٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 16 20.6%31.7% 7%20.8%33.6% 100 135.3133.2 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Dubai is the second largest of the 7 Emir- ates of the United Arab Emirates (UAE) with almost 700 000 inhabitants. Like other Gulf countries, this Emirate is currently un- dergoing rapid socioeconomic development with the concomitant lifestyle changes of increasing use of “fast foods” and increas- ingly sedentary life, leading to health prob- lems such as increasing rates of obesity and type 2 diabetes, which has emerged as an epidemic problem in this region [1]. Type 2 diabetes represents a real challenge to the health planners in UAE due to its high prevalence and increased economic cost to society [2]. The latter includes its effect on morbidity, employment, productivity, pre- mature mortality and the increased use of health services. At the moment, evidence- based interventions and models are avail- able to continuously improve the quality of diabetes programmes at the community level based on principles of chronic disease management [3–8]. Parallel to the economic reforms un- derway in Dubai, the health sector is also undergoing a process of reform. Since the year 2002, a new leadership for the health system in the Department of Health and Medical Services (DOHMS) in Dubai has been applying its vision to develop the sys- tem to international standards and pursue excellence in health care. Dubai is not only moving forward on reform of the health care system but also for international ac- creditation of this system. To this end, the principles, concepts and tools of total quality improvement have been applied as a core business in the organization of health care in this emirate [9]. Great investments have been made in developing the health care system at all levels by applying total quality improvement. Improving the quality of health care provided to diabetes patients has been iden- tified as a priority area for the continuous quality improvement programme endorsed by the DOHMS. The aims of this study were to conduct a baseline assessment of the quality of diabetes care in primary health care (PHC) settings in Dubai before the implementation of the diabetes quality improvement programme and to measure the impact of the programme on key clinical indicators of diabetes care. The present study was carried out in 16 of 18 family practice centres affiliated to the PHC sector of DOHMS in 2004. Two clinics were excluded from the study as they only provide primary medical serv- ices to expatriates at Dubai airport and Port Rashid. Model used The FOCUS PEDSA quality performance improvement model was used as a frame- work for the PHC system development. The steps of the 1st phase of this model (FOCUS) depends on Finding an opportu- nity for improvement, Organizing a quality improvement team, Clarifying the process, Understanding the problem and Selecting an area for improvement. The 2nd phase of the model (PEDSA) stands for Plan, Do, Study and Act [10]. Strategic planning In applying this model a strategic planning workshop was conducted in April 2003 to identify priority areas for improvement in PHC. The participants were representatives from all PHC sections: doctors, nurses, health educators, pharmacists, administra- tors and customer services. At the work- shop, diabetes mellitus fulfilled the criteria ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما for a priority area for implementing a qual- ity improvement disease management pro- gramme due to its high prevalence, high cost, high variability in practice patterns, high risk of clinical outcomes, inefficient delivery system, potential for changes in patients’ lifestyle to improve outcomes, the availability of clinical and other expertise to develop the programme and the consider- able impact of the disease on the burden of illness in this region [1,2]. One doctor from each PHC centre was invited to attend focus group discussions about current problems of diabetes care in PHC settings and barriers to good diabetes care practice. Three focus discussion groups were formed, each of 5–6 participants, led by a facilitator. Each of the 3 groups inde- pendently reached a consensus about the identified problems and barriers. The 3 groups then met to establish a unified list, suggested solutions and a quality agenda to overcome the current problems and barriers concerning diabetes care and to continuous- ly improve the PHC diabetes programme based on principles and an evidence-based care model of chronic disease management [3–8]. A multidisciplinary quality improve- ment team of 12 members was formed as a task group at the central level to set priori- ties for implementing the quality agenda set by the focus groups discussions. Strategic directions for improving quality of care in- cluded developing decision support, clinical information systems, mobilizing teamwork and delivery systems. Goals and specific objectives were then set to achieve each of these strategic directions. Table 1 shows the obstacles to practising good diabetes care as perceived by the doc- tors in the focus group discussion sessions and the solutions implemented during the quality improvement process. Diabetes care guidelines Diabetes care guidelines were formulated based on the most up-to-date clinical evi- dence in order to develop the decision sup- port system [3–8]. Measurements of body weight, body mass index (BMI) and blood pressure were undertaken in accordance with National Health and Nutrition Ex- amination Survey (NHANES) procedures [11]. Glycosylated haemoglobin (HbA1c) levels were measured in accordance with USA standard methods [12] (normal range 4.2%–6.3%). Serum total cholesterol and triglycerides were measured using a colori- metric assay, serum high-density lipopro- tein (HDL) cholesterol was measured using a direct enzymatic method and low-density lipoprotein (LDL) cholesterol was calcu- lated using the Friedewald formula [13]. The goals mentioned in this study were in accordance with those specified by the American Diabetes Association (ADA) guidelines [8]: HbA1c < 7.2%, LDL cho- lesterol < 100 mg/dL, HDL cholesterol > 45 mg/dL, triglycerides < 150 mg/dL, and systolic blood pressure < 130 mmHg and diastolic pressure < 80 mmHg. Processes in the health centres The clinical information system of the dia- betes programme was developed through establishing a computerized diabetes regis- ter in each PHC centre, and developing key clinical indicators of best practice. Medical records were also developed through intro- ducing colour coding of records, problem lists, drug lists, special follow-up cards for diabetes patients and special forms for an- nual checkups and health education. Developing the delivery system was undertaken through establishing diabetes quality improvement teams at the grassroots level in each of the 16 family practice cen- tres. Each health centre team was composed ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما of a doctor, a nurse, a health educator, a dietician and an administrator. The purpose of establishing these teams was to develop a team approach to diabetes care based on the established guidelines; to develop the role of nurse practitioners in diabetes; and to train the team on methods and tools of performance measurements. Nurses were trained as case managers and clinical audi- tors of diabetes care. A list of all the patients with their file num- bers was obtained from the diabetes register ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما in each PHC centre. The 1st practice audit and the establishment of comprehensive baseline data about diabetes patients started in June 2003. A retrospective analysis of all files identified from the diabetes register was performed by the clinical auditors for the period between May 2002 to May 2003. Files included in this study were only ac- tive diabetes files for diabetes patients who attended the PHC centres for consultation about diabetes at least once during the study period (n = 2548) and the analysis was un- dertaken manually by the clinical auditors. This analysis included a comprehensive assessment of variables related to socio- economic status (age, sex, education, oc- cupation, marital status and employment), profile of diabetes (type of diabetes, type of treatment, duration of diabetes, family his- tory of diabetes and family history of coro- nary heart disease, and several variables related to process and outcome of care. The 2nd audit started in January 2005. Due to shortage of time and staff, only certain key clinical performance indicators of process and outcome of diabetes care were audited. The 16 PHC centres were asked to review active files for all diabetes patients attending for diabetes-related visits over a 1-month period. The files were re- viewed retrospectively over 12 months by the nurse clinical auditors and included data collection for the key clinical indicators, which were: HbA1c, blood pressure, LDL cholesterol, BMI, smoking status and refer- ral for funduscopy examination. The data regarding laboratory investigations were extracted from the computerized laboratory electronic system which was operating by the time of the 2nd audit, while analysis of the remaining variables was still undertaken manually through extracting and analysing data available in medical records. Data analysis was undertaken using SPSS, version 12. Appropriate tests of signifi- cance were performed; unpaired t-test was performed to compare independent sam- ple means and the chi-squared test was performed to compare categorical vari- ables. The data for the continuous vari- ables, HbA1c, blood pressure and LDL cholesterol were converted into categorical data to be benchmarked with other practices regarding best practice standards set by the ADA [8]. The population of the catchment areas of the 16 family practice centres affiliated to DOHMS, Dubai is 614 210 people. How- ever, the total number of registered files in these centres for people who are utilizing the service is 319 197, representing a 52% utilization rate. The total number of regis- tered diabetes patients in these PHC centres is 4903 patients giving a point prevalence of 1.6% of the total registered population in the 16 PHC centres. The total number of active files included in the current study was 2548, representing 51% of registered diabetes patients. Table 2 shows the sociodemographic char- acteristics of the patients with diabetes based on available information from the medical records. The mean (standard deviation) age was 55.3 (11.6) years and 90.3% were 40 years of age, with nearly equal sex distribu- tion, and the majority (66.0%) were of UAE nationality. Table 2 also shows that 94.3% of the diabetes patients were married, 2.7% were single or divorced and 3.0% widowed. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Of the patients, 30.2% were illiterate and 47.3% unemployed. It was noted that 79.3% of the records had missing data regarding education level and more than 50% of the records were missing information regarding the employment and marital status of the diabetes patients. Table 3 shows the profile of diabetes in the patients. The table shows that 90.0% of diabetes patients were diagnosed with type 2 diabetes, 74.5% were on oral hypogly- caemic medication only, 30.7% had had ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما diabetes for > 10 years, 61.8% had a family history of diabetes and 26.3% had a positive history of ischaemic heart disease. Three- quarters of the patients with diabetes were overweight or obese (BMI 25 kg/m2). Missing data were mainly in recording the history of coronary heart disease (67.4%) and family history of diabetes (78.5%). Table 4 compares the key performance clinical indicators of process and outcome of diabetes care in the 16 PHC centres be- tween the 1st and 2nd audit cycles. There were significant improvements in the proc- ess of care for the key clinical performance indicators studied: HbA1c, blood pressure ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما and LDL-cholesterol, recording of BMI and smoking status and referral for funduscopy (P < 0.001). Table 4 also demonstrates a significant improvement in outcome variables of dia- betes care. Mean HbA1c was reduced from 8.7% to 8.1% (P < 0.001; 95% CI: 0.4–0.8) and the proportion of patients achieving the audit target level of HbA1c < 7% increased significantly from 20.6% to 31.7% (P < 0.001). The mean systolic blood pressure fell from 135.3 mmHg to 133.2 mmHg (P < 0.05; 95% CI: 0.6–3.6). While the propor- tion of patients achieving the audit target of systolic blood pressure control < 130 mmHg did not significantly improve, there were significant decreases in the other cat- egories (P < 0.001). The mean LDL-cholesterol decreased from 129.2 mg/dL to 115.4 mg/dL (P < 0.001, 95% CI: 10.8–16.8) and the propor- tion of patients with the audit target < 100 mg/dL increased from 20.8% to 33.6% (P < 0.001). Fewer files were reviewed for the differ- ent clinical indicators of the 2nd audit. This was due to manpower shortages affecting the availability of clinical auditors in each of the 16 PHC centres to submit the required data on time. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Table 5 shows the relationship of outcome of diabetes care with some socioeconomic variables in the 1st audit. No significant relationships were observed between the proportion of patients achieving a target of HbA1c < 7% and LDL target < 100 mg/dL and the socioeconomic variables studied, except age 40 years which was signifi- cantly associated with a higher proportion of diabetes subjects achieving the target (P < 0.05). On the other hand there was a significant relationship between the propor- tion of patients achieving a target of systolic blood pressure < 130 mg/dL and age < 40 years, nationality, literacy and employ- ment (P < 0.05) with higher proportions of non-UAE nationals, literate and employed patients achieving the target. This study documents the impact of imple- menting the quality agenda for improve- ment of the diabetes care programme in light of principles and evidence-based mod- els of chronic disease management [4–8]. Several areas for system improvement were identified based on the care model, and quality improvement teams were formed to undertake such improvements. The main areas identified for system improvements were information systems, decision support and systems delivery. The main outcome measures were to monitor and document the extent of improvement in glycaemic, blood pressure and lipid control. The current study showed that the prevalence of diabetes in the registered practice popula- tion in the 16 PHC centres was 1.6%, while community-based studies in the UAE have reported a rate of 10% [2]. This gap could be explained by the multiplicity of health systems in Dubai, as DOHMS is not the only provider of health services in Dubai. In addition to the private sector, the Ministry of Health has its own health premises and some governmental organizations provide health services for their own employees. Some patients could have more than one health card and have access to more than one health care provider. This gap could be also a reflection of the low detection rate of diabetes in the community and the need to establish screening programmes for the early detection of undiagnosed cases of dia- betes in the community. Research evidence has shown that cases of type 2 diabetes can be missed in elderly patients with vascular problems of sufficient severity to warrant amputation. It has been suggested that se- lective screening of high-risk groups is one solution to the problem of reducing the level of undiagnosed diabetes [10]. The present study showed that only 52% of the registered diabetes patients were ac- tively utilizing the PHC services. This could be because patients with type 1 diabetes receive their care mostly from the hospitals and because the health system in the UAE allows citizens with UAE nationality to register in more than one health system, ac- quire more than one health card and choose to receive medical services from any of the available health systems. In addition, the recently introduced fee-for-service scheme for non-UAE citizens may have also con- tributed to the low utilization rate of the diabetes services. The present study showed that most of dia- betes population had type 2 diabetes, were taking oral hypoglycaemic medications, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما were married and were UAE nationals. The illiteracy rate was 44%, compared with 48% in Saudi Arabia, and the unemployment rate was 47%, compared with 44% in Saudi Ara- bia [11]. The study also showed that about three-quarters of the diabetes patients were overweight, with BMI > 25 kg/m2 compared with 56% in a report from Saudi Arabia [12]. This means that nearly half of the dia- betes population are illiterate, unemployed and mostly overweight. This information reflects the need to use appropriate methods of health education for the illiterate group and give more attention to assess activity levels and promote exercise programmes. Information about age, sex and national- ity of the patients could be found easily in almost all the PHC records. On the other hand, in the 1st audit cycle documenta- tion about marital status, education and employment was poor. This is comparable to some other reports from the Gulf area which found low rates of documentation regarding education and employment status [14] and other reports that demonstrated a marked improvement in documentation of these variables in a diabetes care follow- up audit [15]. The degree of improvement in the documentation of socioeconomic variables was not assessed in the 2nd audit cycle of the current study as it needs a major investment in time to undertake such tasks manually. Nevertheless, staff training pro- grammes following the recommendations emerging from the 1st audit emphasized the importance of documenting such variables. As noted by the doctors in the focus group discussions, the structure of the PHC medical records did not facilitate the process of providing adequate diabetes care due to the lack of diabetes follow-up cards. The latter are considered to be one of the essential items of providing good diabetes care [11]. The introduction of a problem-oriented medical record system is an ideal solution to improve the quality of medical records for chronic conditions. Also the introduction of drug lists, problem lists and diabetes follow-up cards can help time wasted during the consultation for diabetes patients, who usually have multiple problems and polypharmacy needs. In ad- dition, establishing simple clinical indica- tors of care will help audit coordinators retrieve relevant information quickly from the records. Considerable time was devoted to conducting the 1st audit cycle manually; nevertheless, in 2004, DOHMS introduced a new computer-based information system with computerized laboratory, radiology and billing systems. This saved time as it was possible in the 2nd audit cycle to conduct the audit of laboratory results elec- tronically. A full electronic medical record system is planned to be in action by late 2006, which will greatly facilitate the audit process for all the studied variables. Glycosylated haemoglobin levels are an objective measure of metabolic control of diabetes. This study showed a significant improvement in the rate of performing this test from 62% to 82% between the 1st and 2nd audit cycle. This can be compared with rates of performing glycosylated haemo- globin tests ranging from 0% to 60% from Saudi Arabia [14,15], from 83.0% to 93.0% from studies in the United Kingdom (UK) [16,17] and 15%, 44% and 81% from the United States of America (USA) [18–20]. The current study was also able to docu- ment a significant improvement in the rate of measuring blood pressure from 84% to 98%. This compares with rates of blood pressure recording ranging from 66% to 100% in Saudi Arabia [14,15,21], 83% in the UK [16] and 86% in the USA [18]. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Lipid disorders are a common source of co-morbidity in diabetes patients and treating such disorders is important as car- diovascular diseases are currently among the main causes of morbidity and mortality in the Eastern Mediterranean Region [2]. The current study showed a significant improvement in the rate of performing lipid profiles from 64% in the 1st audit to 75% in the 2nd audit. This compares with a testing rate of 73.8% in Saudi Arabia [13] and rates ranging from 31%, 45% to 66% in reports from the USA [18–20]. Our study showed a significant improve- ment in referral rates for funduscopy exami- nation between the 2 surveys from 28.9% to 53.0%. Studies from Saudi Arabia reported referral rates of 33% [15] and 61.5% [21], from the UK of 64.4% to 86% [16,17], and from the USA of 22%, 66% and 28% [18–20]. This study in Dubai showed that the pro- portion of patients with good glycaemic control (i.e. HbA1c levels < 7%) improved from 20.6% to 31.7% in the 2nd audit cy- cle. A report from Australia showed an increase from 18% to 25% in the 2nd au- dit [22], while another Australian report demonstrated a rate of 57% [23], reaching ADA targets. By comparison, data from the USA showed rates between 37% and 44% [24,25]. Another study from the USA dem- onstrated an improvement in patients’ mean HbA1c level from 7.8% to 7.4% [26] which is comparable with the improvement in our study from 8.7% to 8.2%. In the present study, the mean systolic blood pressure dropped from 135.3 mmHg to 133.2 mmHg, while the proportion of diabetes patients reaching the ADA target of systolic blood pressure < 130 mmHg remained the same. In a study from the USA [19] the mean blood pressure was found to be 134.1 mmHg and in another study [24] the proportion of patients achieving the ADA target were 41%. In a study from the UK the mean systolic blood pressure fell from 147 mmHg to 140 mmHg between 2 audits [27]. The current audit showed significant improvements in control of LDL-choles- terol, as the proportion of diabetes patients with LDL-cholesterol level < 100 mg/dL increased from 20.8% to 33.6%. By com- parison, 23% of patients in the USA [24] and 52.8% in Australia [23] achieved ADA targets. This study demonstrated the impact of im- proving some aspects of the system and organization of diabetes care on improv- ing key clinical indicators of the diabetes programme in Dubai. The study focused on mobilizing decision support, teamwork, de- veloping role of nurses in diabetes care and improving information systems. The results suggest that many opportuni- ties for cardiovascular disease risk reduction are still missed in spite of efforts to improve the system of care for diabetes in Dubai Emirate. The extent of improvement that has taken place is still not sufficient to meet the challenge, as a significant proportion of individuals were not meeting the targets of the key clinical indicators. Control of weight and glycaemia are complex proc- esses that require efforts beyond health system service development. There is also still a need to monitor and study the impact on outcome of care of socioeconomic vari- ables in Dubai. Further studies are needed to measure the impact on diabetes outcome measures of increasing the interaction of the health care team with diabetes patients, mobilizing self- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما care efforts and community involvement in the diabetes programme. We are most grateful to members of the technical committee of DOHMS for all their support and commitment to continu- ous quality improvement and driving the health system towards excellence. We deep- ly thank Dr Adnan Julfar, Director of PHC for all his efforts to motivate and empower PHC staff. We acknowledge the paramount contribution of the nursing staff in PHC in this work as clinical auditors, case manag- ers and diabetes educators. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 404 2003 59% 62.6%50.3% ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Le diabète de type 2, maladie chronique en pleine expansion mondiale, pose un problè- me majeur de santé publique [1] en Tunisie. Sa prévalence est passée de 2,3 % en 1977 [2] à 10,2 % en 1995 [3]. Cette augmentation est due à la transition épidémiologique [4], caractérisée par une occidentalisation du style de vie, l’augmentation de l’obésité et du stress [5], ainsi qu’à une adoption rapide de nouvelles habitudes alimentaires et d’un mode de vie sédentaire [6]. Du fait de sa prévalence en accroissement continu et de sa morbi-mortalité menaçante [7], le diabète de type 2 pose aussi un problème de coût non négligeable pour le système de soins [8]. Il est démontré par différentes études qu’une approche globale et agressive de cette maladie est efficiente en termes de coût et d’efficacité [9,10]. En effet, l’édu- cation thérapeutique du patient diabétique fait partie intégrante du traitement comme dans toutes les autres maladies chroniques [11]. Pour gérer leur traitement, les pa- tients diabétiques de type 2 doivent réaliser eux-mêmes des actes techniques et pren- dre des décisions d’ordre thérapeutique, l’ensemble étant désigné par le terme de « comportements de soins » [5]. Or l’adhé- sion à ces comportements de soins dépend de nombreux facteurs : psychologiques, affectifs, familiaux et socioculturels, dont les connaissances nécessaires à la gestion du traitement. Un des rôles des soignants est donc d’intégrer l’enseignement des connaissances de la maladie à un processus d’éducation thérapeutique du patient, axée sur la gestion de sa pathologie. Plusieurs études ont montré que les diabétiques de type 2 ayant bénéficié d’une éducation thérapeutique non seulement connaissaient mieux leur maladie, mais aussi savaient mieux se traiter, étaient mieux équilibrés, faisaient moins de complications et étaient moins longtemps et moins souvent hospita- lisés [11]. Cependant, la pratique d’une éducation thérapeutique fondée sur des données pro- bantes et adaptée aussi bien à la culture locale qu’à la personnalité du patient est en- core peu généralisée dans les structures de soins de première ligne en Tunisie assurant la prise en charge des maladies chroniques, dont le diabète de type 2. D’où l’intérêt de ce travail qui se propose de décrire le niveau des connaissances d’un échantillon repré- sentatif des diabétiques de type 2, suivis en première ligne, à propos de leur problème de santé et de relever par conséquent les éventuelles lacunes dans la pratique de cette dimension thérapeutique primordiale : l’éducation pour la santé. Il s’agit d’une étude épidémiologique des- criptive et transversale ayant porté sur un échantillon représentatif des patients dia- bétiques de type 2 suivis dans la région sanitaire de Sousse pendant la période allant de janvier à avril 2003. Cet échantillon a été stratifié selon le niveau de prise en charge : centres de santé de base et consultations de diabétologie. Le calcul de la taille de l’échantillon a été basé sur l’hypothèse que la prévalence estimée de l’éducation diabé- tique a été de 50 % avec une précision de 5 % et un niveau de confiance de 95 %. Cet- te taille a été majorée de 5 % pour remédier à d’éventuelles données manquantes. Les structures de soins participant à l’étude ont été identifiées à travers un échantillonnage stratifié selon les deux échelons de soins de la première ligne afin d’assurer la représentativité de la popula- tion de l’étude et de pouvoir approfondir l’analyse statistique des données. Ainsi, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما l’enquête s’est déroulée dans deux centres spécialisés de diabétologie et trois centres de santé de base. Dans ces structures de santé de première ligne, aucune éducation thérapeutique structurée et systématique n’est organisée pour les patients chroniques dont les diabétiques de type 2, en dehors d’une éducation pour la santé souvent indi- vidualisée, rapide et aléatoire dans le cadre des consultations médicales. L’éducation du diabétique de type 2 a été évaluée à travers les connaissances que celui-ci avait sur sa maladie. Ainsi, 20 questions à réponses fermées axées sur cinq chapitres ont été posées : la définition du diabète, l’origine du diabète, l’évolution du diabète, les interférences avec la vie courante et les instructions thérapeutiques. Chaque dimension étudiée a été composée de quatre items. Nous avons supposé que les différents items ont été égaux dans leur im- portance. Une réponse correcte a été cotée 1, une réponse fausse ou l’aveu d’une non- connaissance « ne sait pas » a été coté(e) 0. Ainsi, chaque dimension prenait un score allant de 0 à 4. Un patient était considéré éduqué à propos d’une dimension de la maladie diabétique s’il avait correctement répondu au moins à 3 items correspondants. Un score global des connaissances a été donc établi, allant de 0 à 20. Pour faciliter l’interprétation des résul- tats, nous avons procédé à la transformation linéaire des scores des cinq dimensions et du score global selon la formule suivante [12] : score final = [(score d’origine – plus petit score d’origine) 100] / étendue possible des scores d’origine. Ainsi, les nouveaux scores obtenus variaient de 0 à 100. Le niveau des connaissances d’un patient a été considéré « satisfaisant » si le patient avait obtenu un score global 75 %, « moyen » si le score global a été compris entre 50 % et 75 %, et enfin « insuffisant » s’il a obtenu un score global < 50 %. Le traitement statis- tique des données a été effectué au Service d’Épidémiologie et de Statistiques médica- les du CHU Farhat Hached de Sousse sur le logiciel Epi Info 6,04 FR. Cette enquête a porté sur un échantillon représentatif de 404 patients diabétiques de type 2 suivis dans la région sanitaire de Sousse. La population étudiée, avec une moyenne d’âge de 60 ans (E.T. 10,9) était constituée de 66,6 % de femmes. Un patient sur cinq bénéficiait de l’assurance médi- cale gratuite. Le pourcentage des patients inactifs sur le plan professionnel était de 64,4 %. Le profil clinique de ces diabétiques de type 2 est caractérisé par une ancienneté moyenne de diabète de 8 ans (E.T. 6,1), une fréquence élevée de l’obésité (37,6 %), une co-morbidité importante, particulièrement l’hypertension artérielle (71,3 %) ainsi que des complications dégénératives fréquentes essentiellement neurologiques (47,5 %) et oculaires (28,7 %). Connaissance de la dénition du diabète Le signe majeur du diabète (polyurie noc- turne) a été reconnu par la quasi-totalité des patients. Cependant, la connaissance de la macrosomie fœtale comme signe de suspicion de diabète chez la femme en- ceinte a été l’item le moins connu par les patients diabétiques de type 2 (Tableau 1). En effet, la moyenne globale de cette di- mension de l’éducation diabétique a été de 66 (E.T. 23,7). Le pourcentage des patients qui avaient une bonne connaissance de la ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما définition du diabète (connaissent au moins trois items) était de 62,6 %. Connaissance de l’origine du diabète Le tableau 2 montre que la connaissance de l’obésité comme facteur de risque pour le diabète a été mentionnée par 80,9 % des patients diabétiques de type 2 et que seulement 35,1 % des patients savaient que l’hypertension artérielle n’entraînait pas le diabète. La connaissance de l’origine de la maladie diabétique (connaissance de 3 items ou plus) a été reconnue par la moitié des patients (50,3 %). En effet, la moyenne globale de cette dimension de l’éducation diabétique était de 54 (E.T. 28,7). Connaissance de l’évolution du diabète Il ressort du tableau 3 que la majorité des patients diabétiques de type 2 connaissaient le caractère chronique de leur maladie ainsi que les complications dégénératives du diabète. Cependant, seulement 64,9 % des patients interrogés savaient répondre correctement aux quatre items concernant l’évolution du diabète. La moyenne globale de cette dimension de l’éducation diabé- tique était de 85 (E.T. 20,0). Les compli- cations oculaires étaient les plus connues (97,3 %). Connaissance des interférences du diabète D’après le tableau 4, la connaissance de l’interférence du diabète avec la grossesse est l’item le moins connu ; par contre, les soins des pieds ainsi que l’importance des activités physiques sont connus par pres- que la totalité des patients (> 90 %). En plus, 89,1 % des patients connaissaient plus de trois items des interférences de la ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما maladie diabétique avec la vie quotidienne. La moyenne globale de cette dimension de l’éducation diabétique était de 87 (E.T. 17,5). Connaissance des instructions du diabète Les signes de l’hypoglycémie étaient connus par les trois quarts des patients interrogés et 96,3 % des diabétiques de type 2 étaient convaincus de la nécessité d’un traitement à vie sous surveillance médicale (Tableau 5). En plus, 91,6 % des patients connaissaient au moins trois items des instructions impo- sées par la maladie diabétique. La moyenne globale de cette dimension de l’éducation diabétique était de 85 (E.T. 16,2). Seulement 58,9 % (IC 95 % : 53,9 - 63,7) de la population étudiée ont été considérés « bien informés » à propos de leur maladie contre 35,9 % (IC 95 % : 31,2 - 40,8) ayant un niveau de connaissances « moyen ». Le taux des patients diabétiques considérés non éduqués (niveau global des connaissances moyen ou mauvais) vis-à-vis de leur mala- die était de 41,1 %. La moyenne du score global des connaissances était de 75 (E.T. 15,5). La figure 1 illustre la répartition des connaissances des patients diabétiques de type 2 en fonction des cinq dimensions ; les insuffisances majeures des connaissances intéressaient les deux dimensions suivan- tes : l’origine du diabète et sa définition. Le concept d’éducation des patients diabé- tiques a été développé depuis de longues années [13]. C’est l’ensemble des activités ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما pédagogiques conçues pour aider le patient à acquérir des compétences concernant sa santé, alliant son information et par consé- quent, l’adoption et le maintien volontaire d’attitudes positives et de comportements favorables. Cette participation active du patient aide à la réussite du contrôle glycé- mique et retarde l’apparition des complica- tions de la maladie. Ce travail n’a pas été à l’abri de quelques insuffisances méthodologiques limitant par- tiellement sa validité interne et externe. D’une part, l’évaluation de l’éducation thé- rapeutique repose classiquement sur l’ex- ploration des trois axes : les connaissances, les attitudes et les pratiques des patients. Or dans notre étude, seule la composante « connaissances » a été décrite étant donné ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما la facilité de son étude [14,15]. Par ailleurs, il existe une corrélation positive entre les connaissances et les attitudes et les compor- tements : mieux un patient connaît sa patho- logie, plus il adoptera de bonnes attitudes et des pratiques correctes dans la gestion de sa maladie. D’autre part, le recueil des don- nées a été fait au moyen d’un questionnaire spécialement conçu pour ce travail dont la validité est en cours d’étude. Dans la littéra- ture, il existe d’autres questionnaires anglo- saxons, largement utilisés et validés par les équipes de recherche, mais spécifiquement conçus pour les diabétiques de type 1. Cette étude a permis de conclure que seulement 59 % de notre population de diabétiques de type 2 avaient un niveau de connaissances « satisfaisant ». Ce résultat a été décevant puisque des notions élémen- taires de la maladie diabétique (ex. : savoir qu’elle est chronique et que le patient dia- bétique peut s’exposer à des complications cardio-vasculaires, oculaires et rénales, savoir que l’apparition des sueurs chez un diabétique signifie une hypoglycémie) étaient mal connues par les diabétiques qui avaient en plus beaucoup d’idées erronées. Ce manque d’information des diabé- tiques a été confirmé par d’autres études tunisiennes. En effet, le pourcentage des pa- tients diabétiques, bien informés concernant leur maladie, étudiés dans des différentes régions de notre pays variait de 14,7 % à 51 % [16-18]. D’autres études étrangères ont rapporté aussi cette insuffisance d’in- formation chez les diabétiques de type 2 [19-22]. Les différences des méthodes de mesure et d’échantillonnage font que les comparaisons entre les études, particuliè- rement l’estimation des taux des patients éduqués, doivent être interprétées avec pré- caution. Pourtant, l’insuffisance de l’éduca- tion diabétique paraît évidente dans toutes ces études et intéressait l’ensemble des dimensions des connaissances : définition de la maladie, son origine, son évolution, ses interférences avec la vie et enfin les instructions pour sa bonne gestion par le patient. Selon notre étude, les items portant sur la définition de la maladie et sur son origine représentaient les deux principales lacunes dans les connaissances des pa- tients diabétiques de type 2 interrogés, avec un pourcentage de réponses correctes de 62,6 % et 50,3 % respectivement. En effet, seulement la moitié des patients avait une notion correcte de la définition du seuil à partir duquel une personne est considérée diabétique et 40 % savaient que la macrosomie foetale pouvait être un signe de diabète. L’insuffisance d’information sur l’association diabète et grossesse peut être expliquée par le fait que la plupart des patients interrogés étaient des personnes âgées, devenues diabétiques après avoir dépassé l’âge de procréation. Le mécanisme ainsi que les facteurs de risque du diabète n’étaient pas clairs à l’esprit des patients. L’obésité était de loin le facteur de risque le plus connu (80,9 %) ; par contre, 64,9 % des diabétiques interrogés croyaient abusi- vement à la responsabilité de l’hypertension artérielle dans la genèse du diabète de type 2. Cette information erronée est due pro- bablement à la fréquence de l’association de ces deux problèmes de santé. Ces insuf- fisances des connaissances concernant la définition et l’origine du diabète pourraient être expliquées par le fait que les médecins croyaient que les notions de définition et d’origine de la maladie font partie de leur savoir sacré et par conséquent, ils ne s’ef- forçaient pas d’expliquer ces notions aux patients. Les causes de la morbidité et de la morta- lité des diabétiques sont dominées par les ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما complications à long terme, représentées essentiellement par les accidents de l’an- giopathie et de la microangiopathie [23,24] ; pour les combattre, une bonne coopération du patient aux mesures préventives et théra- peutiques est nécessaire. Dans notre étude, le caractère chronique du diabète de type 2 a été rapporté par 88 % des patients et les signes de complications aiguës telles que l’hypoglycémie n’étaient rapportés que par 71 % des patients. Cette défaillance dans les connaissances des patients concernant les complications aiguës du diabète a été retrouvée également dans l’enquête de Ben Ammar [16] ; les signes révélateurs d’une hypoglycémie n’ont été cités que par 4,2 % des patients interrogés. Une étude anglaise [25] a montré qu’il existait des lacunes dans les connaissances des diabétiques interro- gés concernant les complications aiguës de la maladie telles que l’hypoglycémie et l’acidocétose. Les résultats de l’étude amé- ricaine du groupe de Médecine préventive [26] et ceux de l’étude espagnole [20] sur l’évaluation du degré des connaissances des diabétiques de type 2 suivis en première ligne rejoignent les nôtres en montrant la méconnaissance des complications cardia- ques. Cette méconnaissance des complica- tions dégénératives de la maladie diabétique pourrait tenir, d’une part, à la longueur de la période asymptomatique qui précède leur apparition et d’autre part, à un déficit d’in- formation de la part du médecin qui ne don- ne des explications aux complications d’une maladie que lorsqu’elles sont installées. Ce contexte favorise vraisemblablement l’am- bivalence du patient et du médecin, même s’ils sont correctement informés des risques de complications graves et invalidantes. L’interférence du diabète avec la grossesse est l’item le moins connu ; par contre, les soins des pieds ainsi que l’importance des activités physiques et d’un régime adé- quat sont connus par presque la totalité de l’échantillon (> 90 %). Les résultats de notre étude concordent avec ceux des autres enquêtes [16,24,2729]. Les patients savent qu’il est recommandé de pratiquer des exercices physiques, qu’il faut respecter un régime limitant l’apport de glucides et de graisses, cependant ils ne mettent pas systématiquement ces re- commandations en pratique. En effet, la modification de l’alimentation par exemple nécessite un changement de comportement, très difficile à obtenir chez le diabétique de type 2, compte tenu de son âge, de l’ancien- neté de ses habitudes, de son surpoids, de son faible niveau socio-économique et de la place importante de la diététique dans la vie de tous les jours [30,31]. Cette étude a permis de quantifier l’ampleur du problème de l’insuffisance des connais- sances des diabétiques de type 2 suivis en première ligne dans la région sanitaire de Sousse à l’égard de leur problème de santé. Par conséquent, il est primordial de renfor- cer la formation des médecins en matière d’éducation thérapeutique en introduisant davantage les sciences humaines lors de la formation initiale (psychologie, pédagogie, communication, etc.), en sensibilisant très tôt les étudiants à la prise en charge des pathologies chroniques et en incluant le thème de l’éducation diabétique dans le cer- tificat de « Pratique médicale ». En plus, il nous paraît important d’institutionnaliser la pratique de l’éducation thérapeutique pour les patients diabétiques dans les structures de soins de première ligne sous forme de séances d’éducation diabétique en groupe dans les Centres de Santé de Base et d’un séjour éducatif hospitalier en « hôpital de ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما jour » dans les hôpitaux de circonscription pour permettre aux diabétiques d’apprendre convenablement la gestion de leur maladie et d’acquérir des attitudes positives et vali- des face à leur problème de santé. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The vascular endothelium is a dynamic endocrine organ that regulates vascular tone, local homeostasis, and the fibro- inflammatory-proliferative process. Many studies have demonstrate that endothelial dysfunction and activation is one of the ear- liest markers in patients with atherogenic risk factors (e.g. male sex, ageing, hyper- tension, diabetes mellitus, smoking, fam- ily history) in the absence of angiographic evidence of atherosclerosis [1]. Oxidative stress to the vascular endothe- lium is a serious causative factor of vascu- lar endothelial dysfunction, and plays an important role in the pathophysiology of several vascular diseases, including athero- sclerosis, diabetes, neuronal disorders, and ischaemia-reperfusion injury [2]. It was noted that some of the patients attending 6th October national insurance clinic for annual routine laboratory and clinical examinations were diagnosed as hypercholesterolaemic without any clinical symptoms. A few years later, vascular injury was noted during the annual medical examination. Thus, we car- ried out this study to confirm the relation- ship between hypercholesterolaemia and vascular injury. Hypercholesterolaemia is frequently associated with enhanced lipid peroxida- tion [3]. We therefore aimed to investigate the correlation between hypercholesterol- aemia and oxidative stress on the one hand and P-selectin and interleukin-6 (IL-6) (pro-inflammatory cytokine) as markers of endothelial functional status on the other. We contacted asymptomatic Egyptian adult employees of the National Research Centre (males and females) who had been exam- ined at the health insurance outpatient clinic between the beginning of October 2004 and the end of December 2004 and who had been identified as having hypercholesterol- aemia (total cholesterol > 200 mg/dL). We excluded any patient with cardiovascular disease. Of those we invited to take part in our study 65% refused to participate and 35% agreed (40 patients). During the same period, 20 healthy age- and sex-matched personnel whose medical records showed total cholesterol and triglycerides were within normal levels and who had no car- diovascular disease were selected from the medical staff of the National Research Centre to participate in the study as con- trols. This group was retested to confirm their blood lipid status before participating in the study. After taking verbal agreement, all par- ticipants (patients and controls) underwent the following investigations. Blood samples (10 mL) were collected from each participant at the health insurance clinic or the clinic in the National Research Centre, as convenient for the participant. Each sample was divided into 2 × 5 mL por- tions: in the first portion, serum was sepa- rated, divided into aliquots and preserved at –20 °C until used; the blood in the second portion was collected over EDTA and as- sayed on the same day for the determination of glutathione peroxidase activity and lipid peroxidation, measured as thiobarbituric acid-reactive substance (TBARS). Lipid peroxidation was determined as TBARS value according to the method of Mitsura and Midori. The TBARS value was measured as the difference in optical density read at 535 nm and 520 nm [4]. Glutathione-peroxidase activity in whole blood was determined using Ransel kit (Randox Laboratories, Crumlin, UK) ac- cording to the manufacturer’s instructions. Total serum cholesterol was determined using a commercial kit (catalogue number ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 07986B-07/98BioMérieux, Marcy l’Etoile, France) according to the manufacturer’s instructions]. Serum high-density lipoprotein (HDL) cholesterol was determined using a Bio- Mérieux kit (catalogue number 00232B- 04/96) according to the manufacturer’s instructions. Serum low-density lipoprotein (LDL) cholesterol was measured directly using a BioMérieux kit (catalogue number 00238 B-04/96) according to the manufacturer’s instructions. It was read at wavelength 500 nm and calculated using the formula: Asample/Astandard × n mg/dL (dilution coef- ficient n = 387). Human P-selectin was measured by enzyme-linked immunosorbent assay kit for quantitative detection of soluble human P-selectin (Bender MedSystems, Vienna, Austria) according to the manufacturer’s instructions. IL-6 was measured using an immu- noenzymometric assay kit (IL-6 EASIA, Biosource, Nivelles, Belgium) for the quan- titative measurement of human IL-6 in serum according to the manufacturer’s in- structions. Results were expressed as mean and standard deviation (SD). Data were ana- lysed using SPSS, version 10. Data were compared using a paired t-test for inde- pendent variables. Values of P < 0.05 were considered significant. Pearson correlations between different variables were done. Values of P < 0.01 were considered significant (2-tailed). Levels of lipid peroxidation (TBARS) (P < 0.001) and glutathione peroxidase activity as markers for oxidative stress were statisti- cally significantly higher in the hypercho- lesterolaemic (patient) group (P < 0.001). P-selectin (marker for endothelial cell func- tion) and IL-6 (pro-inflammatory cytokine) were also markedly higher in the patient group (P < 0.001) (Table 1). Table 2 shows the correlation (Pearson coefficient, r) between total cholesterol ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما level and LDL-cholesterol level and oxida- tive stress, expressed by lipid peroxidation (TBARS) and glutathione peroxidase activ- ity. Also shown is the correlation with en- dothelial function, expressed by P-selectin and the pro-inflammatory cytokine IL-6. Table 3 shows the correlation between oxidative stress, expressed by lipid per- oxidation and glutathione peroxidase activ- ity, and markers of endothelial function, expressed by P-selectin and pro-inflammatory cytokine IL-6 (P < 0.001). Correlation be- tween Il-6 and P-selectin was also statisti- cally significant. There was a negative, but not statisti- cally significant, correlation between HDL cholesterol and the markers of oxidative stress and endothelial function (Table 3). Hypercholesterolaemia has frequently been associated with enhanced lipid peroxidation [3]. In this study we examined the rela- tion between hypercholesterolaemia and increased oxidative stress. We found a very high positive correlation between choles- terol level, LDL cholesterol and raised oxidative stress, expressed as high lipid per- oxidation and glutathione peroxidase activ- ity. This is consistent with the result of Davi et al., who stated that they had obtained evidence of enhanced lipid peroxidation (in vivo) in hypercholesterolaemic patients [3]. Desideri et al. stated that hypercholes- terolaemia was associated with endothelial activation and increased lipid peroxidation [5]. Lewis et al. showed that oxidative stress stimulates the production and release of platelet-activating factor (PAF) in endothe- lial cells [6]. Other reports from Marathe et al. and Tokumura et al. indicated that PAF as well as PAF-like phospholipids are critical factors in the pathophysiology of vascular endothelial dysfunction under oxidative stress conditions [7,8]. According to Lum and Roebuck, oxi- dative stress to the vascular endothelium is a serious causative factor of vascular endothelial dysfunction and plays an impor- tant role in the pathophysiology of several vascular diseases [2]. LDL is the cholesterol ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما component that is more directly involved in the pathogeneses of vascular dysfunction in hypercholesterolaemic patients. In our study there was increased level of LDL cholesterol and increased oxida- tive stress. LDL becomes pathogenic when subjected to oxidation (increase in oxidative stress) and becomes ox-LDL [9]. Ox-LDL is, in fact, no longer recognized by the LDL receptors; instead it is taken up by scavenger receptors. These are not sub- jected to regulation by the intracellular cholesterol level in the subendothelial mac- rophages which transform to foam cells [10]. Egashira showed that LDL, especially oxidized LDL, inhibits endothelial function through inhibition of NO (nitric oxide) activity by down-regulation of endothelial NO synthase expression, decreased receptor- mediated release of NO, and activation of NO via superoxide anion production [1]. Aikawa et al. concluded that lipid lower- ing reduced production of reactive oxygen species, ox-LDL accumulation and plasma level of anti-ox-LDL IgG; VCAM-1 and MCP-1 expression decreased and NO syn- thase expression increased, and endothelial cells exhibited more normal ultrastructure [11]. P-selectin level was much higher in the hypercholesterolaemic group, with a strong positive correlation with cholesterol and LDL-cholesterol levels. This is in agree- ment with Johnson-Tidey et al., who stated that high P-selectin plasma level in symp- tomatic hypercholesterolaemic patients may represent an index of the presence of atherosclerotic vascular lesions P-selectin expression is endothelial cells overlying atherosclerotic plaques [12]. Davi et al. observed that plasma P- selectin concentration was directly cor- related with LDL levels, which suggests that LDL might have an impact on the series of events that lead to P-selectin expression and release in vivo. They also found that hypercholesterolaemia was associated with elevated plasma P-selectin [13]. P-selec- tin level may be proposed as a marker of endothelial dysfunction in hypercholestero- laemic patients. This agreed with the results of our study as we found a strong positive correlation between P-selectin and mark- ers of oxidative stress expressed by lipid peroxidation and glutathione peroxidase activity. Levels of the proinflammatory cytokine IL-6 were significantly higher in the hyperc- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما holesterolaemic group compared to healthy controls. There was a positive correlation with cholesterol, LDL, oxidative stress and endothelial function. This result agreed with that of Nawawi et al., who observed that hypercholesterol-aemia caused endothelial dysfunction, leading to increased produc- tion of adhesion molecules and cytokines (IL-6) [14]. Desai et al. concluded that the inflammatory cytokine IL-6 was an impor- tant mediator of increased endothelial per- meability via alterations in ultra-structural distribution of tight junctions and morpho- logic changes in shape causing endothelial barrier dysfunction [15]. Blood levels of inflammatory markers have been associated with hypercholes- terolaemia [16]. Cytokines (IL-6, Il-1B) and soluble adhesion molecules have been associated with both hypercholesterolaemia and atherosclerotic diseases. Soluble inter- cellular adhesion molecule-1 and IL-6 have been found to reflect endothelial dysfunc- tion in patients with primary hypercholes- terolaemia [17]. IL-6 has been found to induce oxida- tive stress and endothelial dysfunction by over-expression of the angiotensin II type I receptor [18]. Pro-inflammatory cytokines such as tumour necrosis factor alpha and IL-6 are important mediators of immune response, associated with endothelial dys- function [19]. Hypercholesterolaemia causes endothe- lial microinflammation. IL-6 and P-selectin were also identified as risk factors for car- diovascular disorders. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Coronary artery disease (CAD) often oc- curs in the absence of traditional risk fac- tors. This study evaluated 3 promising new markers for cardiovascular risk: bilirubin, total homocysteine (t-Hcy) and high- sensitivity C-reactive protein (hs-CRP) [1,2]. Natural antioxidant defences have evolved to protect humans against deleteri- ous effects of free radicals. The primary enzymatic defences are intracellular, but other antioxidant defences are largely ex- tracellular, including antioxidative sub- strates such as uric acid and unconjugated bilirubin, the predominant bile pigment in the intravascular compartment. For many years, the bile pigment was considered as a toxic waste product formed during haem catabolism. However, more recent evidence suggests that bilirubin is a potent physiolog- ical antioxidant that may provide important protection against atherosclerosis, CAD and inflammation. In 1994, Schwertner et al. were the first to observe a significant inverse correlation between total bilirubin plasma concentrations and the prevalence of CAD [3]. Subsequently, Hopkins et al. noted that patients with early familial CAD had a mean total serum bilirubin of 8.9 (SD 6.1) µmol/L compared with 12.4 (SD 8.1) µmol/L in healthy control subjects [4]. Low serum bilirubin concentrations have been shown to be independently and inversely associated with an increased risk for CAD [3]. The strength of the association between bilirubin and CAD appears to be similar to that of high-density lipoprotein-choles- terol (HDL-C). The antioxidant capacity of bilirubin and its ability to provide po- tent scavenging of peroxyl radicals have led to suggestions that mildly increased circulatory bilirubin may have a physi- ological function to protect against disease processes that involve oxygen and peroxyl radicals [5,6]. Antioxidant activity and car- dioprotective potential may be attributable to any of the bilirubin forms, including free unconjugated bilirubin, protein-bound unconjugated bilirubin, delta bilirubin or mono- or di-conjugated bilirubin. Under physiological conditions, the predominant circulatory form of bilirubin is the unconju- gated, albumin-bound form [7–9]. In recent years, “non-traditional factors” such as hs-CRP, total homocysteine, as well as oxidative stress, have been proposed as risk factors for the development and progression of atherosclerosis and athero- thrombotic cardiovascular disease [10–13]. The purpose of this study was to examine the relationship between traditional and non-traditional biomarkers of CAD in coro- nary angiography patients and in apparently healthy control subjects. All patients referred to the Department of Cardiology, University of Gaziantep be- tween March 2003 and August 2003 for whom clinical data were available were included in our study. Thus 319 subjects were included who were admitted to hospi- tal with chest pain and underwent coronary angiography. These patients were divided into 2 groups: the CAD group which con- sisted of 262 patients (63 females and 199 males) with stenosis of the coronary arter- ies and the non-CAD group (57 patients; 4 females and 53 males) which consisted of patients in whom CAD was excluded by coronary angiography (degree of stenosis < 20% indicating the absence of clinically rel- evant coronary stenosis). A third group was recruited which consisted of 50 apparently healthy control subjects (17 females and 23 males). The controls underwent physical ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما examination and routine laboratory investi- gation to verify their health status and that they were not taking any drugs. Age- and sex-matched individuals without any cle- arly evident chronic disease were recruited as control subjects. Each angiogram was read jointly by at least 2 cardiologists. Coronary artery le- sions were magnified, traced and measured with calipers to determine the percentage of diameter narrowing of the artery. All coronary angiographies were performed in the same centre. Patients with CAD were further divided into groups according to the maximum coronary stenosis at angiogra- phy: 0%–20% (no detectable CAD), 20%– 49% (mild disease), 50%–70% (moderate disease, and 70%–100% (severe disease). Further classification of severity of the disease was done by counting the number of diseased vessels (0 to 3). All participants were weighed and meas- ured, gave blood samples and were ques- tioned about established cardiovascular risk factors, including diabetes, smoking and hypertension. Although the healthy subjects did not undergo coronary angiograms they had a comprehensive physical examination by a physician, completed the World Health Organization standard Rose questionnaire on chest pain, and answered other ques- tions about their past medical history [14]. None of the individuals in the healthy group had angina or a prior history of CAD. All of them had normal electrocardiograms according to the Minnesota Coding Cri- teria [14]. Obesity was defined as a body mass index (BMI) greater than 27.8 kg/m² as proposed by the National Institutes of Health Consensus Statement [15]. Diabetes mellitus was considered present in patients with a known history of diabetes and in patients with a fasting glucose 126 mg/dL (7.0 mmol/L) according to the American Diabetes Association criteria [16]. All 50 control subjects were monitored for somatic illness throughout the investiga- tion period and were excluded if symptoms of infection or systemic illness were present (acute or chronic liver disease, cancer, renal disorder, rheumatic disease, etc.). Patients diagnosed with acute coronary syndrome 6 months prior to the study were excluded. Additional exclusion criteria included the use of aspirin, S-adenosyl-methionine, vita- min supplements, alcohol, anticonvulsants, estrogen, lipid-lowering therapy and other medications that might affect bilirubin, CRP and homocysteine metabolism. Thus 319 individuals were included in our patient group after these exclusions. The study was approved by the Ethics Committee of Gaziantep University, and the individuals participating in the study gave their informed consent. Blood samples of 319 patients and 50 con- trols (in EDTA tubes and tubes without ad- ditives) were taken at the time of admission between 08:00 and 10:00 after an overnight fast. Blood was centrifuged at 3000 g for 10 minutes at 4 ºC. After separation, the aliq- uots were frozen at –70 °C until analysis. We measured serum total bilirubin by a diazo method with a detergent to accelerate azo-coupling and to prevent the precipita- tion of protein. The test was performed by means of an autoanalyser (Hitachi Modular DP Systems, Roche Diagnostics, and Mannheim, Germany). Total bilirubin levels below 1.1 mg/dL are normal for adults. Measurement is linear from 0.1 to 30 mg/dL. The intra-assay imprecision (coeffi- cient of variation) was 1.3% and inter-assay imprecision (coefficient of variation) was 1.9% at a bilirubin concentration of 2.1 mg/dL. Serum hs-CRP and t-Hcy concentrations were determined with the Immulite® one ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما analyser and Immulite® reagent (DPC, Los Angeles, USA) according to the manufac- turer’s instructions. The assays were linear from 2 to 50 mol/L (t-Hcy) and 10 to 160 mg/L (hs-CRP); calibrators and controls were supplied by manufacturers. Specifica- tions of intra-assay and inter-assay coef- ficients of variation of hs-CRP and t-Hcy assays were assessed from quality control data of the laboratory, which were 5.3% and 6.0% (intra-assay for hs-CRP), 9.1% and 9.9% (inter-assay for hs-CRP), 1.5% to 3.0% (intra-assay for t-Hcy) and 1.7% to 3.2% (inter-assay for t-Hcy). Summary statistics were evaluated for all variables. Differences between the 3 groups were tested with a t-test for independent samples or Mann–Whitney test for con- tinuous variables and a chi-squared test for categorical variables. The difference in cur- rent smoking frequency between the groups was tested by the Fisher exact test. Spear- man correlation coefficients were obtained for biomarkers and each study variable for CAD patients. To determine independent predictors of the presence of CAD, multi- variate logistic regression analysis was done using a model including all variables with a P-value < 0.15 on univariate analysis. None of the results changed if the log of bilirubin and t-Hcy was used instead of the untransformed values. Because the distribu- tion of hs-CRP is rightward skewed, values derived from log-transformed means were used as means for this variable throughout the study; these values virtually coincided with median values. Area under the curve (AUC) values in receiving operating cha- racteristics (ROC) curve (as a measure of discriminating efficacy) were used for com- parison of the diagnostic values of different analyses (including only the CAD and non- CAD groups, using angiography as the gold standard). Optimal cut-off levels, sensitivity and specificity of CRP were selected based on the ROC curves. Only 11 women were under age 50 years and there was no differ- ence in bilirubin, hs-CRP and t-Hcy levels between pre- and postmenopausal women. Therefore, these variables were not further considered in the analysis. Two-tailed P < 0.05 values were considered. All statistical analyses and illustrations were obtained with SPSS, version 9.0 and MedCalc statis- tical software. Summary statistics are given for patients with CAD, without CAD (patients with nor- mal angiogram) and for apparently healthy subjects in Table 1. There was no signifi- cant difference between the groups in BMI, waist/hip ratio and age. Mean [standard deviation (SD)] serum bilirubin levels were significantly higher in apparently healthy subjects [0.81 (SD 0.32) mg/dL] than pa- tients without CAD [0.52 (SD 0.25) mg/dL] and the patients with CAD [0.55 (SD 0.39) mg/dL] who underwent coronary angiogra- phy (P < 0.01). However, t-Hcy [10.7 (SD 5.14) mol/L] and hs-CRP [0.43 (SD 0.61) mg/dL] levels were significantly lower in individuals in the apparently healthy group and the groups undergoing coronary an- giography; without CAD [13.0 (SD 8.61) mol/L, 1.27 (SD 2.78) mg/dL, P < 0.01] and CAD [19.4 (SD 8.73) mol/L, 1.54 (SD 0.87) mg/dL, P < 0.01] respectively. Ad- ditionally, serum t-Hcy and hs-CRP were significantly higher in patients with CAD compared to those without CAD (P < 0.05). Mean serum levels of hs-CRP [1.7 (SD 2.1) mg/dL] and t-Hcy [19.8 (SD 9.6) mol/L] were highest in the patients who smoked (not shown in the table). ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Correlation coefficients of biomarkers of CAD in patients in whom CAD was angi- ographically documented are given in Table 2. There was a negative correlation between bilirubin and sex (male) (r = –0.183, P < 0.01 and r = –0.199, P < 0.01 respectively); in contrast, a significant positive correlation between bilirubin and serum t-Hcy concen- trations was found (r = 0.330, P < 0.001). Bilirubin was also significantly correlated with triglycerides (r = –0.183, P < 0.01) and uric acid (r = 0.127, P < 0.05), but was not significantly correlated with any other study variable. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Total Hcy and hs-CRP concentrations showed many correlations with other study parameters (Table 2). There was a positive correlation between serum t-Hcy and sex, age, smoking, hs-CRP and number of sten- otic vessels. In addition there was positive correlation between hs-CRP and sex, age and t-Hcy. The predictor variables obtained by re- gression analysis for the number of stenotic vessels in CAD patients are given in Table 3. Sex (male, P = 0.001), age (P = 0.004) and t-Hcy (P = 0.0001) were strongly cor- related with the number of stenotic ves- sels (severity of disease); hypertension (P = 0.019) was moderately associated, and HDL-C (P = 0.022), glucose (P = 0.028), total cholesterol (P = 0.047) were weakly associated. Bilirubin, hs-CRP and other parameters were not related to the number of diseased vessels and the degree of occlu- sion (P > 0.05). Optimal cut-off levels and the associ- ated diagnostic performances (sensitivity, specificity and diagnostic value) of serum bilirubin, hs-CRP, t-Hcy, based on ROC analysis, are given in Table 4. Optimal cut-off levels for bilirubin, hs-CRP and t-Hcy providing the maximum efficiency found in patients (n = 319) with CAD were 0.59 mg/dL, 1.09 mg/dL and 12.1 mol/L respectively. ROC curve-based sensitivities ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما of bilirubin, hs-CRP and t-Hcy levels were 70.9%, 50.0%, 76.8% respectively. The specificities of bilirubin, hs-CRP and t-Hcy were 40.4%, 80.7% and 70.2% respectively (data of ROC curves are shown in Figures 1–3). To the best of our knowledge, the present study is the first to assess the diagnostic performance and relationship of bilirubin with hs-CRP and t-Hcy for cardiovascular disease in men and women in an angi- ographically documented design. The study demonstrated that patients with angiograph- ically confirmed CAD had significantly higher serum hs-CRP and t-Hcy levels than non-stenotic patients (patients with normal angiogram) and the apparently healthy con- trol group. These data strongly suggest that serum t-Hcy helps to identify individuals at risk of atherosclerosis (AUC value 0.781), especially among those with elevated hs- CRP and decreased bilirubin levels. t-Hcy showed the highest AUC value (0.781) compared to hs-CRP (0.648) and bilirubin (0.507). In agreement with previous reports, we found that the bilirubin levels in serum were significantly lower in the patients with CAD than in age- and sex-matched controls [3,17,18]. We found that a serum bilirubin concentration of 10.0 µmol/L (0.58 mg/ dL) discriminated between high and low cardiovascular risks. This association was ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما independent of the extent of CAD, BMI, diabetes, hypertension and smoking. Indi- viduals in the top quintile of serum bilirubin concentration had an 80% reduction of the CAD risk compared with individuals in the lowest quintile [4]. In 1995, Breimer et al. performed a prospective study of 7685 middle-aged men enrolled in the British Regional Heart Study and found that both low and high bilirubin concentrations were associated with an increased risk of CAD [7]. More recently, Vitek et al. reported on the prevalence of CAD in individuals with Gilbert syndrome who were found to have a CAD prevalence of 2% compared with 12.1% in the general population [18]. A meta-analysis of 11 studies has shown a negative relationship between serum bi- lirubin concentration and severity of athero- sclerosis in men (r = –0.31, P < 0.0001) [19] but we did not find such an association in either men or women. However, we found that the number of stenotic coronary arteries was significantly associated with elevated serum t-Hcy and hs-CRP concentration. Several researchers have investigated the risk of myocardial infection in individu- als with the UGT1A1*28 allele [20,21]. According to the “oxidative modification hypothesis”, which suggests atherogenesis is initiated by oxidization of low-density lipoprotein particles, it has been suggested that increased physiological concentrations of serum bilirubin may reduce atherogenic risk by reducing oxidation. An involvement of bilirubin in immune reactions and inflam- matory processes has also been documented [22–24]. Smoking causes oxidative stress and production of acute phase reactants, such as CRP, temporary ischaemia, repeti- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 0 20 40 60 80 100 100-Specificity 100 80 60 40 20 0 0 20 40 60 80 100 100-Specificity 100 80 60 40 20 0 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما tion of inflammation and reactive oxygen species. Thus, these factors constitute en- dothelial injury, which increases platelet aggregation, abnormalities of fibrinolysis and smooth muscle cell proliferation, and accelerate the development of thrombosis and atherosclerosis [25,26]. Earlier studies have reported differences in the levels of t-Hcy, ranging from 13.9– 20.1 mol/L in persons with CAD [27,28]. We found a mean t-Hcy level of 19.4 (SD 8.73) mol/L in the CAD group, 10.7 (SD 5.14) mol/L in the healthy group and 13.0 (SD 8.61) mol/L in the non-CAD group. Some differences between reported serum t-Hcy levels may be related to analytical methods and ethnic differences. Bortolotto et al. grouped patients as hypertensive and hypertensive plus CAD. When compared, the plasma t-Hcy levels were significantly higher in the hypertensive plus CAD group [29]. Our study agrees with this study in that we obtained a similar association. The regression coefficient of hypertension with arterial blood pressure was 0.141 (SEM 0.142), P = 0.019. Also, homocysteine enhances oxidative stress. A study in 19 centres in Europe reported high homo- cysteine levels and increased risk of CAD in smokers [30]. We found that the t-Hcy levels tended to increase in the presence of more cardiovascular risk factors, i.e. male gender, older age, diabetes mellitus, hyperlipidaemia and certain chronic dis- eases. As expected, traditional coronary risk factors were more prevalent among those participants with elevated levels of t-Hcy and hs-CRP in our study, as in other studies [31–33]. More recently, McConnell et al. [34] and Lear et al. [35] have reported gen- der differences in C-reactive protein. The observed gender differences have important 0 20 40 60 80 100 100-Specificity 100 80 60 40 20 0 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما implications for the establishment of cut-off points for cardiovascular risk stratifica- tion [36]. We found that lower total serum bilirubin was associated with a higher risk of CAD among men, but the pattern was much clearer in women (Table 2). Contrary to the findings of Djoussé et al. [13], our study provides only suggestive evidence for a lower risk for women. However, the relatively small number of CAD cases in women means that our study had less sta- tistical power in women. Women may be more susceptible to low levels of bilirubin. The fact that the median age at baseline was 50 years in women indicates that most of the women in our study were postmenopausal. It is possible that in this older age group, the effects of bilirubin are not off-set by those of estrogen. A plausible biological mechanism is necessary to support a causal association between serum bilirubin and CAD outcome. The levels of bilirubin may be related to an inflammatory condition in patients with CAD [37,38]. Another possibility is that low bilirubin concentration is not per se a major causative factor in the development of CAD, but rather a reflection of the presence of this ailment. According to this view, low bilirubin is a result of increased oxidative activity in CAD-prone individuals, leading to consumption of a natural antioxidant such as bilirubin [39–41]. Our data suggest that serum bilirubin concentration is more closely associated with the oxidative stress marker serum uric acid level [–0.033 (SEM 0.038), P < 0.587] than smoking. These findings conflict with those that have found subjects who smoke and have low serum bilirubin antioxidant concentrations [20]. Problems in risk assessment also arise from overlapping properties (shared pathophysi- ological pathway) of traditional risk factors such as hypertension, obesity, age, gender, smoking and diabetes [42–49]. To conclude, we found little evidence of an association between the serum concen- tration of bilirubin and atherosclerosis. In contrast, the concentration of novel (t-Hcy and hs-CRP) and traditional risk markers may be stronger markers for atherosclerosis in CAD patients. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Regional consultation on establishing guidelines on management and care for acute coronary conditions The World Health Organization Regional Office for the Eastern Medi- terranean organized the above-mentioned regional consultation to establish guidelines on the management and care for acute coronary 27 29 2007conditions, in Cairo, Egypt, from to March . The objectives of the consultation were: to review the progress made in the management and care of acute coronary conditions among countries of the Regions; and to set regional strategies for the management and care of acute coronary conditions. Experts from Egypt, Islamic Republic of Iran, Lebanon, Pakistan, Qatar, Saudi Arabia, Syrian Arab Republic, Tunisia, United Arab Emirates, United Kingdom, as well as WHO concerned staff, participated in the consultation. • • ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Les maladies cardio-vasculaires sont la cause principale de morbidité et de mor- talité dans les pays industrialisés. Elles sont à l’origine de 41 % de décès 1,2 et la majorité des décès d’origine cardio- vasculaire est due à l’insuffisance coronaire : la pathologie coronaire est la première cau- se de mortalité chez les hommes de plus de 45 ans et chez les femmes de plus de 65 ans 3 . Pour permettre d’importants progrès en matière de traitement et de prévention des maladies cardio-vasculaires, il faut axer la recherche sur la compréhension des méca- nismes qui sont à la base du développement de ces maladies et s’appuyer sur des études épidémiologiques de grande ampleur pour identifier les facteurs de risque internes et externes. Avec le développement socioécono- mique et de l’infrastructure sanitaire, la Tunisie est en pleine phase de transition épidémiologique 4 . Cette transition est caractérisée par une baisse des taux de mortalité, une augmentation de l’espérance de vie, une modification des causes de mortalité au profit des maladies chroniques non transmissibles, les maladies cardio- vasculaires en particulier. Un diagnostic de situation s’impose pour quantifier l’ampleur du problème et étudier la meilleure stratégie de prévention. Nous avons entrepris une étude populationnelle en milieu hospitalier pour décrire la fréquence de différentes pathologies cardio-vasculaires et leurs fac- teurs de risque associés. Dans cette partie du travail, qui implique tous les services hospitalo-universitaires de cardiologie situés dans la région du Sahel en Tunisie et qui regroupe les gouverno- rats de Sousse, Monastir et Mahdia, nous avons étudié la fréquence et la distribution des principaux facteurs de risque cardio- vasculaire chez des sujets coronariens hos- pitalisés entre 1994 et 1998. Cette étude s’est adressée à tous les patients ayant fait un infarctus du myocarde (IDM) ou un angor instable et admis en unités de soins intensifs cardiologiques (USIC) des hôpitaux Sahloul et Farhat Hached à Sousse, de l’hôpital Fattouma Bourguiba à Monastir et de l’hôpital Tahar Sfar à Mahdia entre 1994 et 1998. Un patient était considéré comme atteint d’un infarctus du myocarde s’il avait au moins 2 des éléments suivants : douleur angineuse typique au repos pendant au moins 30 min ; élévation du segment ST > 0.1 mV dans 2 dérivations contiguës frontales ou > 0.2 mV dans 2 dérivations précordiales contiguës à l’électrocardio- gramme (ECG) ; apparition d’une onde Q de nécrose au décours de la douleur ; éléva- tion des créatines phosphokinases (CPK) à plus de deux fois la normale. Un patient était considéré comme ayant un angor instable s’il avait un des éléments suivants : angor de repos ; angor d’effort accéléré, brutalement aggravé ; angor d’ef- fort de novo. Les données cliniques, biologiques et familiales des patients (sexe, âge, poids, taille, hypertension artérielle, diabète, tabac, alcool, antécédents familiaux et personnels de cardiopathie ischémique, profession, lieu de naissance, lieu de résidence) ont été analysées de façon rétrospective à partir des dossiers médicaux. L’obésité a été définie par un indice de masse corporelle (IMC) > 25,8 kg/m2 chez la femme et 26,4 kg/m2 chez l’homme, soit un surpoids supérieur ou égal à 20 % d’après les tables de 1959 de la Metropolitan Life Insurance Company 5 , déterminé sur les valeurs du poids et de la taille mesurés pen- dant l’hospitalisation. Les autres facteurs de risque (diabète, hypertension, dyslipidémie et tabagisme) étaient déterminés d’après les informations contenues dans l’observation médicale ainsi que par l’analyse des médi- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما caments prescrits. L’hypertension artérielle (HTA) a été définie par l’existence d’un traitement antihypertenseur en cours ou en fonction des recommandations de l’OMS 6 , comme l’existence d’une pression ar- térielle systolique > 140 mmHg ou d’une pression artérielle diastolique > 90 mmHg notées à plusieurs reprises jusqu’à la fin de l’hospitalisation. L’existence d’un diabète était déterminée par une glycémie à jeun > 1,26 g/L ou un traitement antidiabétique en cours 7 . La dyslipidémie était définie par une valeur du cholestérol total > 2 g/L et/ou des triglycérides > 1,5 g/L et une valeur du LDL-cholestérol > 1,6 g/L ou un traitement hypolipémiant en cours. Les va- leurs lipidiques étaient élevées dans tous les bilans pratiqués régulièrement au cours de l’hospitalisation. Le tabagisme a été défini à partir de l’interrogatoire lors de l’admission ou du dossier médical préexistant. Les données ont été saisies et analysées sur SYSTAT (System for statistics, ver- sion 5). Les données ont été exprimées en moyenne écart réduit ou en nombre de sujets et pourcentage (%). L’analyse des variables qualitatives a été réalisée par le test du 2, alors que l’analyse des données quantitatives a fait appel au test t de Student ou à l’analyse de variance. Dans tous les cas, un seuil de significativité de 5 % a été utilisé. L’échantillon représentait la totalité des patients admis entre 1994 et 1998 pour in- suffisance coronaire aiguë dans les services de cardiologie des quatre centres hospitalo- universitaires de la région du Sahel en Tuni- sie, soit environ le tiers de l’ensemble des patients hospitalisés pour cette pathologie en Tunisie. Les données de 3455 patients (1741 IDM et 1723 angors instables) ont été analysées. Leur description selon le type de pathologie (IDM ou angor instable) et en fonction de l’âge est indiquée au tableau 1. Les hommes qui représentent 72,4 % de la population totale sont âgés en moyenne de 59,9 ans (E.T. 12,3) (extrêmes : 20 et 97 ans) et les femmes de 63,1 ans (E.T. 10,7) (extrêmes : 24 et 100 ans). La figure 1 montre les principaux facteurs de risque cardio-vasculaire dans l’ensemble de l’échantillon et par sexe. Il y a une différence statistiquement signifi- cative de la prévalence de l’HTA entre les femmes (59,2 %) et les hommes (28,5 %), p < 0,001. Le diabète est retrouvé chez 46,4 % et les femmes sont significativement plus souvent atteintes (56,4 %) que les hommes (42,5 %), p < 0,001. L’obésité a été retrouvée dans 27 % des cas. Elle est signifi- cativement plus fréquente chez les femmes (31,9 %) que chez les hommes (25,1 %), p < 0,001. La dyslipidémie a été retrouvée dans 40,6 % des cas, avec une prévalence plus élevée chez les femmes (43,7 %) par rapport aux hommes (39,4 %), p < 0,025. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما L’hypercholestérolémie se retrouve chez 15,8 % des sujets, l’hypertriglycéridémie chez 8,5 % et la dyslipidémie mixte chez 16,3 %. Le tabagisme a été retrouvé dans 56,8 % des cas. Les hommes sont significa- tivement plus souvent fumeurs (77,4 %) que les femmes (2,9 %), p < 0,0001 ; 8,2 % des sujets fument entre 1 et 10 paquets/année, 18,8 % fument entre 11 et 20 paquets/année, 54,8 % fument entre 21 et 50 paquets/année et 18,1 % fument plus de 50 paquets/année. L’association chez la même personne entre les facteurs de risque étudiés a montré que 7,7 % des sujets coronariens n’ont aucun facteur de risque, 26,3 % ont un seul facteur de risque, 30,5 % en ont deux, 23,9 % en ont trois, 9,9 % en ont quatre et 1,5 % ont cinq facteurs de risque. La distri- bution des facteurs de risque chez l’homme et selon l’âge est indiquée au tableau 2. La prévalence du tabagisme est élevée à tous les âges, variant de 81,8 % pour la tranche d’âge inférieure à 35 ans à 62,5 % pour la tranche d’âge supérieure à 85 ans. Le pour- centage de coronariens hypertendus aug- mente avec l’âge, passant de 1,8 % chez les moins de 35 ans à plus de 35 % lorsqu’on dépasse les 75 ans. Il en est de même pour la prévalence du diabète, qui représente moins de 10 % chez les moins de 35 ans, pour atteindre 49,8 % dans la tranche d’âge 55-64 ans et baisse ensuite à 25,0 % chez les plus de 85 ans. Pour la dyslipidémie, les valeurs sont élevées et croissantes jusqu’à 54 ans pour décroître ensuite progressive- ment au delà de cet âge. Il en est de même pour l’obésité qui suit la même évolution que la dyslipidémie. La distribution des facteurs de risque chez la femme selon l’âge est indiquée au tableau 3. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Chez les femmes du groupe d’âge 35- 44 ans, la prévalence de l’hypertension arté- rielle (38,1 % vs 7,0 %), du diabète (38,1 % vs 26,3 %), et de l’obésité (38,1 % vs 24,9 %) est plus élevée que celle observée chez les hommes pour la même tranche d’âge, alors que la prévalence du tabagisme (8,4 % vs 82,1 %) et de la dyslipidémie (33,3 % vs 43,8 %) est plus basse. Tou- tefois, l’évolution de la prévalence des différents facteurs de risque reste croissante avec l’âge chez les deux sexes. Pendant longtemps, l’étude des maladies cardio-vasculaires et leurs facteurs de risque ne concernaient que les sociétés dévelop- pées 8,9 , puisqu’il n’existe pas de données fiables sur la fréquence, en termes de mor- bidité et de mortalité, des maladies cardio- vasculaires dans les pays en développement. L’étude des maladies cardio-vasculaires et des facteurs de risque associés en Tunisie est intéressante à plus d’un titre. D’abord, très peu de données épidémiologiques sont disponibles pour quantifier le problème en Tunisie, à part quelques études en milieu semi-urbain 10 et urbain 4,11 et touchant un nombre réduit de sujets ou s’intéressant à un seul facteur de risque 12 comme le diabète ou l’hypertension artérielle. En- suite, des données de ce type confirment le phénomène de transition épidémiologique, ce qui devrait permettre au pays de mieux organiser son système de santé pour faire face à cette nouvelle pathologie chronique et coûteuse et d’entreprendre des actions de prévention efficaces à l’échelle du pays entier, comme cela a été le cas pour les pays à forte mortalité cardio-vasculaire 13,14 . Par ailleurs, très peu de données publiées existent à ce jour sur la prévalence en mi- lieu hospitalier des différentes pathologies cardio-vasculaires et de leurs facteurs de risque. Notre étude, qui porte sur une co- horte de 3455 patients originaires de la région du Sahel en Tunisie et hospitalisés pour maladie coronaire, montre un profil de risque cardio-vasculaire qui rappelle, et dépasse pour certains facteurs comme le diabète ou le tabagisme, celui des pays développés et traditionnellement exposés au fléau des maladies cardio-vasculaires com- me la France (HTA 49,9 %, diabète 17,6 % chez des patients angineux 15 , ou encore HTA 42,1 %, diabète 19,2 % chez des su- jets coronariens 16 ). Aux États-Unis, la prévalence du diabète était de 26 % chez les hommes et de 21 % chez les femmes 17 . Dans une autre étude décrivant des patients angineux suivis au Royaume-Uni, l’HTA y est rapportée chez 48 % des malades, le dia- bète chez seulement 9 % des patients 18 . Le pourcentage des malades hypercho- lestérolémiques (12,4 %) est nettement inférieur à celui de l’étude américaine où 34 % des patients étaient concernés, sans différence entre les hommes et les femmes. Il l’est encore davantage, comparé aux étu- des françaises ELAN (Étude longitudinale dans l’ANgor) 15 et CORALI 19 (57 %) et à l’étude européenne EUROASPIRE (European Action on Secondary Prevention through Intervention to Reduce Events) 20 où il atteint 71 % des patients européens et 73 % des patients français. Cette différence avec nos propres résultats est partiellement liée aux particularités alimentaires de cha- que population. L’obésité est retrouvée chez 18,7 % des patients, pourcentage supérieur à ce- lui de l’étude ELAN (14,8 %) et l’étude PREVENIR (10,9 %). Quant au tabagisme, il est noté chez 68,3 % des malades ; ce pourcentage est nettement supérieur à celui de l’étude PREVENIR (48,8 %), à celui de l’étude ELAN et de l’étude EUROAS- PIRE (18 %) et à celui de l’étude CORALI. Dans cette dernière étude datant des années ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 1987-89, 59 % des patients fumaient plus de 10 cigarettes par jour. La fréquence particulièrement élevée du tabagisme chez les hommes (77,4 % vs 2,9 %) explique en grande partie leur plus grande susceptibilité à développer des cardiopathies ischémiques surtout par rapport aux femmes post- ménopausées. Le taux de tabagisme chez les hommes (77,4 %) excède en effet celui des États-Unis (43 %) [21] et celui de la plu- part des pays de la communauté européenne (41 %), à l’exception de la Grèce (61 %) [22]. La femme demeure encore préservée, contrairement aux pays développés. Les résultats de notre étude sur la région du Sahel en Tunisie confirment les résultats déjà publiés et portant sur la région du grand Tunis [23], et montrent que la prévalence des facteurs de risque cardio-vasculaire chez des patients hospitalisés pour maladie coronaire est élevée. Ces données descripti- ves mettent en évidence un profil à haut ris- que cardio-vasculaire de ce type de patients, nécessitant une prise en charge et un suivi régulier dans un contexte de prévalence éle- vée de facteurs de risque cardio-vasculaire. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Caesarean section is considered a safe surgi- cal procedure. Better anaesthesia, improved surgical techniques, more efficacious pre- ventive and therapeutic measures against infections and thromboprophylaxis have contributed to the advancement of safety. Despite this, caesarean section remains associated with higher maternal morbidity and mortality than vaginal delivery [1,2]. Although anaesthesia and surgery carry their inherent risks, maternal morbidity and mortality at caesarean section is more a consequence of either a disease predating pregnancy or induced by pregnancy [3]. About one-third of performed caesarean sections are repeat procedures [4]. Repeat caesarean sections are associated with an increased incidence of placenta praevia and placenta praevia accreta [5–7], scar dehis- cence and rupture [3,8,9]. Health workers in developing countries in general, and Middle Eastern countries in particular, may face different problems. Large families are the desired norm and sterilization is not read- ily acceptable so the prevalence of women with multiple previous caesarean sections is high. Most studies compare patients delivered by caesarean section with those delivered vaginally. The aim of our study, however, was to evaluate maternal complications dur- ing the peripartum period in relation to the number of previous caesarean sections and the risks of maternal morbidity and mortal- ity. The ultimate goal being the provision of evidence-based information that can be used in planning antenatal care and admis- sion for elective or emergency deliveries to units that are well equipped to deal with po- tential complications. Such evidence would have added benefit in counselling couples regarding the advisability of performing tubal ligation in cases of multiple previous caesarean sections. Out of 2209 patients delivered by caesar- ean section at Princess Badea Teaching Hospital between January 2000 and June 2001, the complete medical records of 1739 patients (80%) were reviewed. The review of this large number of files to retrieve the necessary data after the patients were dis- charged was done manually and was time consuming. During the period since the data were collected no change in policies or practice occurred in the management of patients delivered by caesarean and so the situation in the hospital is believed to be the same now. The rate of caesarean section was 18% of all deliveries. The surgical procedure involved a “standard” transverse lower seg- ment caesarean section under general an- aesthesia. These operations were performed by fourth-year obstetric residents or special- ists and tubal ligation only performed if the patient and her husband gave their consent. Cephalosporins were routinely prescribed and anticoagulant prophylaxis was adminis- tered to patients with high thromboembolic risk profiles. Patients were divided into those with no previous caesarean section and those with 1, 2, and 3 or more previous caesarean sections (groups 0, 1, 2 and 3 respectively). In addition to demographic data, parity and gestational age, details of any maternal intra-operative, postoperative and puerperal morbidity and mortality were extracted. Placenta praevia accreta was diagnosed intraoperatively. The following information was extracted from the records: peripartum hysterectomy including hysterectomy for placenta praevia accreta, bowel and bladder injury and any intra- or postoperative blood transfusion due to excessive blood loss during the operation. Puerperal complica- tions in each of the designated groups were ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما also recorded, such as fever (> 38 °C on 2 consecutive measurements, 6 hours apart other than in the first 24 hours) and wound infection presenting as induration, erythema and or purulent discharge. Wound seromas were not included. Descriptive statistics were generated and significance of statistical comparison was determined by the chi-squared test. The Fisher exact test was used for smaller groups. Odds ratios (OR) and 95 confidence intervals (CI) were calculated to identify the association between number of caesarean sections and certain complications. P < 0.05 was considered statistically significant. Of the 1739 patient records reviewed, 1060 (61%) had no previous caesarean section and were undergoing the procedure for the first time, 351 (20%) had 1 previous caesar- ean section, 204 (12%) had 2 and 124 (7%) had 3 or more previous (the highest was 6 caesarean sections). As expected, maternal age and parity were higher in patients with more previous caesarean sections, but this was not the case for gestational age. Tubal ligation was performed in 64 (3.7%) patients (Table 1). Of these, 49 (76.6%) were in high parity patients (36 patients had no previous cae- sarean section, 6 with 1 previous caesarean section, 6 with 2 previous caesarean sections and 1 with 3 previous caesarean sections). Only 15 (23.4%) patients had tubal ligation because of high number caesarean sections (3–6 previous caesarean sections). Table 2 shows the maternal complica- tions recorded according to the presence or absence of previous caesarean sections. Twelve (12) patients required caesarean hysterectomy. Of these, 7 were because of placenta praevia accreta, all of which were anterior. One patient had a hysterectomy with a normally sited placenta that was adherent. Two patients had bowel injury; 1 with history of 2 previous caesarean sec- tions, the other with history of 4. Bladder injury was recorded in 3 women, the first in combination with uterine rupture and no previous caesarean section, the second had 1 previous caesarean section and was delivered in the second stage, the third pa- tient had had 6 previous caesarean sections and was delivered at 37 weeks of gestation and found to have incomplete rupture and extensive adhesions. There was 1 death who was a mother with 3 previous caesarean sections and placenta praevia accreta. She died a few hours after undergoing caesarean hysterectomy because of prolonged shock and multiple organ failure. Generally women with no previous caesarean section and those with previous ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما caesarean sections had a similar rate of complications apart from hysterectomy for placenta praevia accreta, which was sig- nificantly higher in patients with previous caesarean section (P = 0.03) (Table 2). The rates of maternal complications were compared between the 4 groups ac- cording to the number of previous caesar- ean sections (Table 3). The risk for blood transfusion as a result of intra-operative blood loss was similar in patients with no previous caesarean section and those with 1 previous caesarean section. Women with 2 previous caesarean sections were at lower risk of transfusion (OR = 0.31, 95% CI: 0.12–0.77, P = 0.01), while women with 3 previous caesarean sections were at a higher risk of blood transfusion (OR = 1.96, 95% CI: 1.12–3.44, P = 0.02). There was an increased risk of placenta praevia accreta in relation to previous cae- sarean scar. In patients with placenta prae- via, 6 out of 20 (30%) women with previous caesarean section needed a hysterectomy compared to 1 out of 33 (3%) of those with no previous caesarean section (OR = 14.14, 95% CI: 1.65–128.56, P = 0.02). When the study groups were analysed separately, the risk of hysterectomy increased with increasing number of previous caesarean sections (OR = 1.50, 33.00, 132.00 for groups1, 2 and 3 respectively) (Table 3). The risk of scar rupture was not signifi- cantly affected by the number of previous caesarean sections. Women with 1 or 2 previous caesarean sections had the same risk of developing post-operative fever as women without previous caesarean sec- tion, but the risk of fever in women with 3 or more previous caesarean sections was higher (OR = 2.31, 95% CI: 1.04–5.14, P = 0.04) (Table 3). No significant correlation was found between having had any number of previous caesarean sections and the risk of developing wound infection. There were no cases of anaesthetic complications or thromboembolic events in the records re- viewed. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما When undergoing an abdominal delivery, patients with previous 1 or more caesarean sections are exposed to specific problems that are directly related to this mode of delivery [2]. Despite the improvement in technical skills and preventive measures of various potential complications, mater- nal morbidity and mortality at caesarean section are still encountered [1,2]. When counselling patients with previous caesar- ean sections regarding future pregnancies, evidence-based advice is advantageous, which is why we conducted the study. Analysis revealed that blood transfusion was required for 7.6% of patients with no previous caesarean section and for 8.2% of those with previous caesarean section. This ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما is higher than the 1.9% reported by Loverro [5]. The group with 2 previous caesarean sections needed fewer blood transfusions, which is mostly due to the fact that these patients had elective deliveries while those with no or 1 previous caesarean section were mostly emergency deliveries. The significantly higher frequency of blood transfusion in patients with 3 or more previ- ous caesarean sections (P = 0.018) cannot be clearly explained but may be related to associated problems in this group, such as placenta praevia, uterine atony due to twin pregnancy and atony due to grand multipar- ity. Placenta praevia accreta is a serious problem that is strongly associated with the presence of previous caesarean section scar [8–11]. Our study revealed a 14-fold increase in the risk of peripartum hyster- ectomy for this indication in patients with previous caesarean section compared to those with placenta praevia and no previous caesarean section. This risk of hysterectomy increased with increasing number of previ- ous caesarean section. This corroborates data from other surveys which showed a linear correlation between the risk of caesarean hysterectomy in cases of placenta praevia accreta and the number of previous caesarean sections [7,12]. Peripartum hysterectomy is an operation that is almost always performed as an emer- gency and is associated with significant blood loss as suggested by Castaneda et al. [13]. This makes it essential that patients with previous caesarean section and pla- centa praevia should be properly counselled and operated on by senior staff members. Enough blood products must be readily available. Prenatal diagnosis of placenta praevia accreta by transabdominal colour Doppler ultrasound may have an impact on the peripartum clinical management [14,15] Regarding postoperative pyrexia, there was a lower overall incidence of fever in patients undergoing their first caesarean section and those with previous caesar- ean sections compared with the report of Chazotte and Cohen (2.9% and 2.1% re- spectively versus 5.1%) [8]. Women with 3 or more previous caesarean sections had a significantly higher risk of postoperative fever than those with no history of previ- ous caesarean section. Variable degrees of adhesions, extensive tissue handling and longer operative time are possible contribu- tors to this discrepancy. These variables should be the subject of further investiga- tion. Analysis of the difference in wound infection between the subgroups did not reveal any significant differences. The number of women with multiple previous cesarean sections who had opted for tubal ligation was very small which perhaps reflects the desire for large families in the community. Given the risk of complications associ- ated with repeat caesarean delivery, women likely to face this situation should be coun- selled about these risks and encouraged to consider avoiding large families. To mini- mize maternal risk, women with placenta praevia and previous caesarean section should be delivered by the most senior members of staff in adequately equipped hospitals. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Every year around the world some 13 mil- lion premature children are born. Most of these children are born in developing countries, and they account for the largest share of prenatal morbidity and mortality [1–7]. The consequences of preterm deliv- ery for children are extremely serious and a preterm neonate might be affected by them throughout his/her life. Infants born too soon die and suffer disability and other morbidity more frequently than infants born at full gestation [6]. They also represent a large economic burden; extremely preterm neonates might have to stay in the neonatal intensive care unit for 90 days and each day can cost approximately 750 EUR (US$ 1 = 0.76 EUR) [4]. In developed countries, 35% of expenses for treating diseases in children result from preterm delivery [8]. A study in the United Kingdom showed that 62% of mortality in neonates < 1 month and 85% of mortality in infants < 1 year occurred in those born prematurely (gestational age between 22 and 36 weeks) [9]. For women, preterm birth may be the tip of the iceberg of other potentially preventable morbidities and may have its own long-term physiologi- cal and psychological consequences for the mother. Data suggest that preterm delivery has not decreased in the last decade in spite of all known risk factors [4]. In the United States, for instance, preterm birth has risen steadily from 9.4% in 1981 to 11.4% in 1997 [6]. The main causes of preterm deliveries are preterm uterine contractions, > 1 cm dilation or more than 70% effacement of cervix and abnormalities in the uterus or cervix [8]. Research in Finland showed that unmarried status, smoking, low education, age above 35 years and first pregnancy were the most important risk factors for preterm deliveries [10]. Although there are many maternal characteristics associated with preterm delivery, the etiology in most cases is not clear. Research to study risks, etiology and prevention of preterm delivery is therefore necessary in order to reduce the rate of preterm delivery and its adverse health and economic outcomes. In the Islamic Republic of Iran, there are no extensive data about preterm delivery. A controlled trial conducted by Makiabadi in 2 hospitals of Shiraz in which 806 pregnant women were divided into 3 groups (low risk, high risk with training and high risk without training) showed that the incidence of preterm delivery in these groups was 3.1%, 8.6% and 30.4% respectively [11]. The objectives of the present study were to investigate and provide data on the risk factors of preterm delivery in Shiraz city and to introduce a preventive strategy to be integrated in the maternal health care programme in order to reduce the rate of preterm delivery. In this study, all 36 health care centres of Shiraz city were involved. In the first phase of the study, a midwife from the family planning division from each centre was in- vited to take part in a one-week workshop (conducted by a gynaecologist) to introduce the known preterm delivery risk factors and common prevention strategies and also the study structure, objectives and data collec- tion instrument (Table 1). In the second phase, the health care providers trained at the workshop started collecting data on the pregnant women at- tending their centres. A questionnaire was used for data collection and as a screening instrument. The questionnaire devised was based on that of Holbrook et al. [12] and ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما contained 14 items for major risk factors, 14 items for minor risk factors and 12 items for background risk factors (Table 2). Each major, minor and background risk factor scored 10, 5 and 2.5 respectively. If the risk score for a pregnant woman exceeded 10, she was categorized in the high-risk group, otherwise in the low-risk group. The reason for taking a risk score of 10 as the cut-point was based on Makiabadi’s study [11] indicating a score of 10 would give > 90% sensitivity. A high sensitivity would result in including a larger number of pregnant women into the high-risk group, which would have no harmful effects on the women. Additionally, a score of 10 has been used as the cut-point in similar studies by other researchers, for instance Main et al. [13]. Thus, all pregnant women attending the 36 health care centres during 3 February to ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 5 March, 2000 with gestational age less and 36 weeks were screened for the presence of preterm delivery risk factors using the ques- tionnaire and were divided into 2 groups; low risk for preterm delivery and high risk. In the third phase of the study, the high- risk pregnant women were divided into sub- groups of 5 to 10 persons. As an intervention activity, to detect and reduce the preterm delivery risk factors, a training programme was devised by the researchers and was given by the trained health care providers to the subgroups. The programme consisted of a 4-hour training session followed by several routine consulting sessions. The contents of the training were nearly the same as those of the workshop conducted for the health care providers (Table 1). In the routine consulting sessions, previous training was briefly repeated and, if needed, clinical interventions, such as hospitaliza- tion or treatment, were prescribed. For ethical reasons we did not to include a control group (high-risk pregnant without training) in the study. Since the protocol of health care services given to pregnant women in governmental health care centres remained the same from 1992 to 2000, and the criteria for categorizing pregnant wom- en remained the same, the control group of Makiabadi’s study (high risk without training) [11] was used as the control group of this study to assess the effectiveness of the prevention strategy. All the women were followed to deliv- ery and their delivery status was recorded (preterm or full term). As preterm delivery has a complex etiol- ogy with several associated variables, we used a multivariate regression model to analyse the data. Statistical analyses were performed using SPSS, version 11 and SY- STAT. The chi-squared test was used as cri- teria for introducing independent variables into the regression models. Logistic regres- sion (binary and polytomous response) and Cox regression model were used to analyse factors associated with preterm delivery. Polytomous response logistic regression with the following 3 levels was applied for categorizing the outcomes of the pregnant women studied: • Level 1: Preterm delivery • Level 2: Full term ( 37 weeks) delivery with intervention • Level 3: Full term delivery without intervention (reference level) (Makiaba- di’s group) In the Cox regression model, the ges- tational age was considered a continuous variable and was not dichotomized to < 37 and 37 weeks. There were 1117 pregnant women included in the study; 519 (46.5%) and 598 (53.5%) of the pregnant women were labeled as high risk and low risk respectively. The total number of preterm deliveries was 94 (8.4%). The number of preterm deliveries in the low-risk group was 18 (3.0%) and ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما in the high-risk group was 76 (14.6%); the difference was statistically significant (P < 0.001). Comparison of the preterm delivery rate in the high-risk group of this study (14.6%) with that of the control group of the previous study (30.4%) [11] showed the significant effect of the training programme (P < 0.001). Table 3 presents the distribution of some risk factors in the 2 groups. The only significant differences found between the high- and low-risk groups were in education level (P = 0.03) and number of pregnancies (P = 0.002). Table 4 shows the distribution of the main risk factors in the high-risk pregnant women. Based on the scoring procedure, women with at least 1 of these main factors were categorized in the high-risk group. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Polytomous logistic regression analysis is presented in Tables 5. The table presents the odds ratios of the outcome of delivery in level 1 in reference to level 3. As indicated in Table 5, presence of > 1 cm dilation, smoking > 10 cigarettes a day, multifetal gestation and other factors were significant- ly associated with preterm delivery. The regression models also showed that when comparing the delivery outcome between level 2 (full-term delivery with interven- tion) and level 3 (full-term delivery without intervention as the reference level), the only factor retained in the model was smoking > 10 cigarettes a day (odds ratio = 4). Table 6 presents odds ratios from the Cox regression models. In this analysis, more factors were significantly associated with preterm delivery (presence of > 1 cm dilation, multifetal gestation, smoking > 10 cigarettes a day, other factors, injury or trauma during pregnancy, preterm uterine contractions and period of training). In this study, based on the risk factors of preterm delivery, the pregnant women ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما were categorized as low- and the high-risk groups. The frequency of preterm delivery in the study group was 8.4%; 3.0% for the low-risk and 14.6% for the high-risk group. As expected, the difference between the frequencies of preterm delivery in the 2 groups was statistically significant (P < 0.0001). The frequency of preterm delivery in our study was higher than that of preterm delivery found in Denmark with the rate of 3.1% [14] and lower than that of preterm delivery in the United States with the rate of 15.2% [15,16]. The preterm delivery rate in the low-risk group of Makiabadi’s study [11] was not significantly different from the frequency in the low-risk group of our study (P > 0.05). The frequency of preterm delivery among the high-risk group of Makiabadi’s study who had not received training, considered as the control group (30.4%), was significantly higher than that of the high-risk group of the present study who had received training (14.6%) (P < 0.0001). This indicates the effectiveness of combining a training pro- gramme with the routine health care given to high-risk pregnant women. The results of our study are in agreement with previous reports [2,3,17] that suggest that participa- tion in an organized preterm delivery pre- vention programme that emphasizes patient education and frequent provider contact can significantly decrease the incidence of preterm birth. Based on this, the training programme for preterm delivery prevention devised in our study was integrated into the routine health care programme for high-risk pregnant women in all health centres in the Islamic Republic of Iran. Our results show that cervical dilation more than 1 cm increases the risk of preterm delivery about 54 fold. Other strong risk factors retained in the regression models were smoking and multifetal gestational pregnancy which increased the risk of pre- term delivery over 6 fold. This concurs with the findings of other studies [3,8,10]. It was also found that training pregnant ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما women for more than 98 days (14 weeks) could decrease the risk of preterm delivery by approximately 50%. Considering this period as a continuous variable in the Cox regression gave a better result in decreasing the risk of preterm delivery. Comparing the results of the 2 regres- sion models showed that although the first 4 risk factors were nearly the same, the Cox regression model also revealed an associa- tion between preterm delivery and injury or trauma during pregnancy, preterm uterine contractions and the period of training. Thus Cox regression seems to provide a more comprehensive statistical analysis. Our study indicates that the following risk factors increased the risk of preterm deliv- ery among our study population: > 1 cm di- lation of the cervix; smoking > 10 cigarettes a day; multifetal gestational pregnancy; injury or trauma during pregnancy; preterm uterine contractions. Training high-risk pregnant women about the risk factors for preterm delivery and preventive strategies could be an effec- tive way to lower the incidence of preterm delivery and consequently prenatal mortal- ity. Research funding for this study was pro- vided by Shiraz University of Medical Sci- ences and the Ministry of Health, Treatment and Medical Education of the Islamic Re- public of Iran. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 1998–2003 109951371998–200328.7 0001024.80001032.800010 40.50001025.230.8 16.311.300010 3550.700010 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Neural tube defects (NTD) result in congen- ital malformations of the nervous system, the most common of which are anenceph- aly, spina bifida and encephalocele. They are the result of multifactorial disturbances in embryonic neurulation [1]. Numerous risk factors have been identified for NTD. Exposure to methotrexate, valproic acid or aminopterin, maternal diabetes, hyperther- mia, low socioeconomic status, and lack of folate have been shown to increase the risk of NTD. Also, genetic factors are believed to be important [2–5]. The prevalence of NTD at birth varies considerably by country, geographic zone, and ethnic and racial group; it ranges from as high as 1 case in 100 births in some regions of China to about 1 case in 2000 or less in some Scandinavian countries. In many countries the prevalence is approxi- mately 1 in 1000 births [5–8]. Previous studies on NTD have been carried out in some parts of the Islamic Re- public of Iran. In a study on 13 037 births in the capital, Tehran (1969–78), 17.6/10 000 newborns had NTD [9]. In another study of 8585 deliveries in Hamadan (1991–97) (a north-west province), the prevalence of NTD was 50.1/10 000 [10] and in Can- nadajh–Kordestan (another north-west province), out of 14 915 births, 55.0/10 000 newborns had NTD [11]. NTD may lead to spontaneous abor- tion, stillbirth, death in early infancy or a lifetime of disability. In view of the sig- nificant health costs of NTD it important to have baseline data of the prevalence of NTD in every region. This study was done in a referral hospital in the north of the Islamic Republic of Iran during 6 years (1998–2003) to determine the prevalence of different types of NTD and the relations of these abnormalities to factors including sex, maternal ethnicity, maternal age and season. This descriptive cross-sectional study was carried out in Dezyani teaching hospital in Gorgan, the capital city of Golestan prov- ince, north Islamic Republic of Iran. The hospital serves as a referral centre for ob- stetric and gynaecologic problems and a prenatal intensive care unit for all other hospitals and clinics in the south-east of the Caspian Sea border (Golestan province). The region has a population of about 1.5 million and covers an area of about 20 460 km2. Dezyani hospital is one of 17 hospitals and there are more than 500 primary health centres in the region. As a result, patients in this catchment area requiring transfer for the purpose of specialized investigations and treatment (obstetrics and gynaecology) must be referred to Dezyani hospital. All live and stillbirth newborns delivered in the hospital from January 1998 through December 2003 from mothers residing in Gorgan province were screened for con- genital malformations. NTD were defined according to the International classification of diseases, version 10. Stillbirths referred to all fetal deaths after 24 weeks or longer gestation. It was not possible in this study to record data about abortions. All NTD rates were calculated per 10 000 births. Demographic characteristics and gen- eral information of all of the neonates and their parents—sex, date of birth, kind of NTD, mother’s age and ethnicity—were recorded in the medical chart. Other vari- ables such as consanguineous marriage and residence of parents were recorded only for neonates with NTD. Three ethnic groups (native Fars, Turkmen and Sistani) were defined. The native Fars groups are ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما the predominant inhabitants of the region. The Turkmen migrated from other parts of central Asia around 250 years ago and have a rate of interracial marriage of nearly 100%. The Sistani are immigrants from the Iran–Pakistan–Afghanistan border from half a century ago. All data were analysed with SPSS soft- ware and were evaluated and compared with the chi-squared test. A P-value of 0.05 or less was considered statistically significant. Between 1998 and 2003 there were 37 951 births in Dezyani teaching hospital, Gor- gan, with 109 newborns and stillbirths re- corded with NTD. The prevalence at birth of NTD during the 6-year period was there- fore 28.7 per 10 000 births. There were 48 males and 61 females; the rate of NTD was 24.8/10 000 and 32.8/10 000 in males and females respectively (no significant differ- ence) (Table 1). Out of the 109 NTD cases, 62 had spina bifida, 43 were anencephalic and 4 had en- cephalocele. The corresponding prevalence for spina bifida was 16.3/10 000 births (13.9 and 18.8/10 000 for males and females respectively), for anencephaly 11.3/10 000 (9.3 and 13.5/10 000 for males and females) and for encephalocele 1.1/10 000 (1.5 and 0.5/10 000 for males and females). Table 1 shows the rate of NTD by mother’s age; the highest rate of NTD was 50.7/10 000 in newborns with mothers aged 35 years. The NTD rates were 40.5/10 000, 25.2/10 000 and 30.8/10 000 for moth- ers of Turkmen, native Fars and Sistani ethnic groups respectively (no significant difference). This study shows 36 (33.0%) of the parents with affected newborns had consanguineous marriages. Also 63% of the parents resided in rural areas and 37% in urban areas. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Figure 1 shows the rate of NTD each year. The highest rate was in the year 2000 (36.0/10 000). Seasonal variations during the 6-year period were observed. The rate of NTD in October to March (33.0/10 000 births) was higher than July to June (24.0/10 000 births) ( 2 = 2.42, df = 1, P > 0.05). The peak prevalence occurred in December. Our study showed a rate of NTD in this referral hospital in the Islamic Republic of Iran of 28.7/10 000 births. In this research we could not study abortions and therefore our results may be underestimated. Our rate is higher than studies in other countries such as Canada where it was 1.41/1000 [8], in South Africa 1.74/1000 [12], in Germany 15.0/10 000 [13], in the north of England 17.9/10 000 [14], in the north of France 10.9/10 000 [15] and in the United States of America (USA) 9.3 to 14.6/10 000 [5]. The rate is lower than that of China which was 6.0/1000 [16], in Turkey 30.1/10 000 [3] and in north-west of the Islamic Republic of Iran (Hamadan) 50.1/10 000 [10]. These variations in different studies could be ex- plained by the influence of racial, ethnic and social factors in various parts of the world, which are commonly explained as genetic disorders. Geographical, nutritional and socioeconomic and biological factors could also be involved. Other reasons for these variations in birth defect prevalence are the type of sample (referral hospitals would be expected to have higher rates) and method of diagnosis). The rates of NTD in different ethnic groups showed that the Turkmen had the highest rate at 40.5/10 000 but among na- tive Fars it was 25.2/10 000 and among Sistani it was 31.0/10 000. Studies by other researchers also show different NTD rates among different races [5,17,18], suggesting that race and ethnicity may be a factor in the rate of NTD [19]. Spina bifida was the most common NTD in our study, which agrees with ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما other studies [20,21], followed by anen- cephaly and encephalocele. The rate of cystic spina bifida in our study population was 16.3/10 000, which is higher than 6.2/10 000 in France [22], 7.1/10 000 in Texas [5] and 1.09/10 000 in Saudi Ara- bia [23]. It should also be mentioned that our rate is higher than another study in Tehran (central Islamic Republic of Iran) with 3.8/10 000 [9] and in Hamadan (north-west province) with 6.98/10 000 [10]. The rate of anencephaly in our study was 11.3/10 000, which is higher than the other studies such as 6.0/10 000 in South America [24], 6.4/10 000 in Texas [5] and 8.0/10 000 in Tehran [9]. But the rate in our study was lower than in Hamedan with 15.6/10 000, China with 87.0/10 000 and Turkey with 16.4/10 000 [10]. The rate of encephalocele (1.1/10 000) was similar to a study in the USA (1.03/10 000) [5]. Regarding sex differences, our results indicate that the rate of NTD was higher in females than males (male to female ratio = 0.76), as reported by other researchers [5,25,26]. The male to female ratio was 0.69 for anencephaly and 0.74 for spina bifida, which is also comparable to other studies [5,9,17,22–24]. For example, in the USA the ratio for all NTD was 0.62, for anencephaly 0.54 and for spina bifida 0.68 [5]. Our research showed that the highest rate of affected newborns was in mothers aged 35 years (50.7/10 000), with 23.3/10 000 in mothers aged 15–19 years and 28.2/10 000 aged 20–34 years. Our ob- servation of a linear relation between the rate of NTD and increasing maternal age is different from other studies which show a higher risk among younger mothers [27] or, more commonly, a U-shaped curve with higher rates in mothers aged under 19/20 years and over 35 years [5,12,22,28]. Thus age is a complex risk factor in NTD and this issue needs more investigation. In this study a seasonal variation was observed and the rate of NTD was high- er in the October to March period with a peak in December. In a study in Ire- land [27] the rate was higher in January– June (28.0/10 000) compared with July– December (23.2/10 000). Also in Ireland the peak was in April [27]. Some research has shown that the rate of consanguineous marriage is high in NTD births [21,23]. In our study 33% of parents with affected newborns had consanguine- ous marriage, although this rate is lower than in Saudi Arabia (89% of the spina bifida parents) [23] and higher than in South Africa [12]. Another study in the north-west of the Islamic Republic of Iran indicated that the rate of consanguinity among parents with healthy infants was 23% [29]. The possibility that consanguinity could be a risk factor for NTD in a population requires further research. In this study 63% and 37% of parents with affected newborns lived in rural and urban areas respectively. A greater preva- lence of NTDs at birth has been shown for rural areas compared with urban areas [30,31]. A report from China (1988–1991) indicated the prevalence of NTD in rural areas (44.3/10 000) was 3 times higher than urban areas (14.4/10 000) [32]. It may be due to factors such as high population growth rates and socioeconomic factors. According to our findings ethnicity and interfamilial marriage may play a role in the NTD rate in this region of the Islamic Re- public of Iran, although there could also be effects of environmental factors such as ex- posure to toxic agricultural substances and nutritional factors such as folate deficiency. So further investigations are needed, and we recommend that a central registry be set up to record NTD occurring in the south-east Caspian Sea region of the country. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما We extend special thank to the Deputy Chancellor and Director of Research Af- fairs of Golestan University of Medical Sciences for the financial assistance and to the personnel of the newborns and delivery sections in Dezyani Hospital. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Congenital hypothyroidism is one of the most preventable causes of mental retarda- tion. Its prevalence is reported to be 1 in 3000–4000 live births. Early diagnosis and treatment of congenital hypothyroidism can prevent its hazards [1]. The prevalence of congenital hypothy- roidism varies in different areas and among different races. Previous studies have shown that congenital hypothyroidism was more prevalent among Asian neonates than others and parental consanguinity was considered as a cause [2,3]. Besides familial factors, other risk factors, such as environmental, genetic and autoimmune factors, affect the prevalence of congenital hypothyroidism [4–7]. Congenital hypothyroidism has been reported to be more prevalent in the Islamic Republic of Iran than other communities. Three studies in our country, in Fars prov- ince, Tehran and Isfahan, have reported a high prevalence of congenital hypothy- roidism [8–10]. Moreover, recently Or- dookhani et al. reported a high prevalence of permanent congenital hypothyroidism and parental consanguinity among patients with congenital hypothyroidism in Tehran [11]. The aim of our study was to determine the prevalence of congenital hypothyroidism in our region by screening for the disorder in Isfahan. According to the preliminary findings of the study of Hashemipour and colleagues [10], the prevalence of con- genital hypothyroidism is high in our re- gion. Therefore, as well as determining the prevalence of congenital hypothyroidism, we compared parental consanguinity of all referred neonates with that of neonates with confirmed congenital hypothyroidism to determine if this factor plays a role in the disorder. This was a descriptive study carried out from May 2002 to April 2005 and all ne- onates referred from the 17 maternity hos- pitals in Isfahan were evaluated. The study was designed in collaboration with paediat- ric endocrinologists, the head of the Isfahan Endocrine and Metabolism Research Centre, the head of the Social Medicine Department of Isfahan University of Medical Sciences and Health Services and carried out with the research assistance of Isfahan University of Medical Sciences & Health Services. The Research Bureau of Isfahan University of Medical Science and Health Services gave ethical approval for the study. The deans of all the 17 hospitals and the heads of their neonatal sections were informed of the study, and the importance of screening for congenital hypothyroidism was explained. With their approval, we approached all newly delivered mothers in the hospitals and explained to them the disorder, its complications and the method of screening. The coverage percentage of this project was derived by calculating the ratio of the number of referred neonates to live births. Trained nurses in the hospitals com- pleted a questionnaire with mothers which recorded gestational age, neonate’s sex, weight, height, head circumference and nationality and parental consanguinity (1st and 2nd cousin). We recommended that the mothers attend the Isfahan Endocrine and Metabolism Research Centre for screening after discharge from hospitals on the 3rd–7th day of delivery. They were requested to give the questionnaire to the executive commit- tee of congenital hypothyroidism screening in the Centre. This committee consisted of a trained general practitioner and nurses who received the referred women. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Venous blood samples of the neonates were obtained by trained nurses in the Centre on the 3rd–7th day of birth and serum thyroxine (T4) and thyroid stimu- lating hormone (TSH) levels were meas- ured in Isfahan Endocrine and Metabolism Research Centre. An endocrinologist and collaborating general practitioner evaluated the laboratory results, the status of parental consanguinity and determined the neonates who needed to be recalled. Recalls were determined based on the levels of T4 and TSH in neonates born at term who weighed over 2500 g. Neonates born at term with T4 < 6.5 g/dL or TSH 20 mIU/L [12,13], and premature neonates with a low level of T4 for their weight or high TSH level for their age were selected for recall [14]. Neonates who were referred after the 7th day of birth were recalled based on a T4 level < 6.5 g/dL or TSH level > 10 mIU/L [15]. If the TSH level was between 20 and 39 mIU/L then a second laboratory test, includ- ing T4, was carried out. If the TSH level was > 40 mIU/L then as well as carrying out a second laboratory tests, treatment was initi- ated [16]. The second measurement from recalled neonates was performed on the 7th–28th day of birth. According to the 2nd laboratory test, if the levels of TSH and T4 were in the normal range, the neonate was considered to have hyperthyrotropinaemia. If the level of T4 was < 6.5 g/dL or TSH > 10 mIU/L [12] then the neonate was consid- ered to have congenital hypothyroidism. The physician performed physical ex- aminations of the neonates and evaluated the laboratory tests and finally, based on the findings, prescribed levothyroxin 10–15 g per kg per day for hypothyroid neonates. In addition to measurement of TSH level, if term or premature neonates had low levels of T4 according to their weight, additional laboratory tests, such as T3 resin uptake (T3RU) and free T4 index (FT4I), were carried out. According to these results, congenital hypothyroidism was diagnosed and patients underwent treatment. Patients with confirmed congenital hypothyroidism were recommended to undergo thyroid scintigraphy before starting treatment. The levels of TSH and T4 were measured using Iran Kavoshyar kits. The level of TSH and T4 were measured using immuno- radiometric assay and radioimmunoassay respectively with the gamma counter of the Endocrine and Metabolism Research Cen- tre (Berthold LB 12-2111). Sensitivity of the kits was 0.05 mIU/L for TSH and 0.38 g/dL for T4. Thyroid scintigraphy was performed using technetium pertechnetate. Data were analysed using SPSS, version 13 and Epi-Info, 2002. Differences in the frequency of parental consanguinity be- tween screened neonates without congenital hypothyroidism and those with congeni- tal hypothyroidism were compared using the chi-squared test. The odds ratios (OR) and confidence intervals (CI) for parental consanguinity in neonates with congenital hypothyroidism and those without were calculated. P < 0.05 was considered sig- nificant. This study included 93 381 neonates from 17 private and public maternity hospitals in Isfahan which represented 82.8% of live births. Of the neonates, 51.7% were female and 48.3% male, 97% were born at term and 3% premature, and 97.3%, 2.6% and 0.1% were respectively Iranian, Afghan and oth- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما er nationalities (Armenian, Iraqi, Libyan, Yemeni); the nationality of 136 neonates was not determined. Of all the studied neonates, 27.7% had parental consanguinity (both 1st and 2nd cousin); the parents of 61.6% of this group had a consanguineous marriage with a 1st degree relative. Of the 93 381 referred ne- onates, 1038 (1.1%) were recalled; 971 (93.5%) of these underwent laboratory tests and 274 were diagnosed with congenital hypothyroidism and received medical treat- ment. The prevalence of congenital hypothy- roidism was 2.9 per 1000 live births or l in 341 live births; 4 neonates had second- ary hypothyroidism. As regards sex, 161 (58.8%) of hypothyroid neonates were fe- male and 113 (41.2%) male, giving a female to male ratio of 1.4:1. Of the hypothyroid neonates, 251 (91.2%) were born at term and 23 (8.4%) were premature. The rate of parental consanguinity (1st and 2nd cousin) is shown in Table 1; 101 (36.9%) neonates with congenital hypothy- roidism had consanguineous parents (both 1st and 2nd degree relation) and of these, 75 (74.3%) had parental consanguinity with a 1st degree relation. Only 121 (44.1%) of the hypothyroid neonates underwent thyroid scintigraphy (most of them were the patients with pri- mary high TSH levels); the remainder did not undergo the procedure because of poor compliance of their parents. According to the results of the thyroid scan, 82 (68.3%) of the patients with congenital hypothyroidism had a normal thyroid scan, 1 had goitre, 7 (5.8%) had ectopia and 31 (25.8%) had agenesia. Therefore 31.6% of our patients with congenital hypothyroidism had dys- genesia. The overall prevalence of parental con- sanguinity among normal, goitrous, ectopic and agenetic congenital hypothyroidism patients was 46.3, 0.0%, 57.1% and 45.1% respectively. Data about 1st and 2nd cousin parental consanguinity among these patients are presented in Table 2. There was significant association be- tween parental consanguinity (both 1st and 2nd degree relation) and the prevalence of congenital hypothyroidism. Consanguin- ity was more prevalent among neonates with hypothyroidism (OR = 1.53, 95% CI: 0.51–0.85, 2 = 11.33, P = 0.0007). In addi- tion, congenital hypothyroidism was more prevalent in neonates with 1st cousin pa- rental consanguinity as compared with 2nd cousin parental consanguinity (OR = 1.8, 95% CI: 1.13–2.88, 2 = 6.83, P = 0.008). The prevalence of congenital hypothy- roidism worldwide is reported to be 1 in 3000–4000 live births [1]. In our study of 93 381 referred neonates the prevalence of congenital hypothyroidism was l in 349 live ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما births, which is high. Previous studies have reported a high prevalence of congenital hypothyroidism in our country (l in 1433, l in 914 and 1 in 370 live birth in Fars prov- ince, Tehran and Isfahan respectively) [8– 10]. Rates for nearby countries differ. The prevalence of congenital hypothyroidism in Pakistan was reported to be l in 1000 live births [17], while in Saudi Arabia urban areas it was 1 in 2759 live births and in the rural areas it was l in 1538 live births [18]. Our study shows that among neonates with congenital hypothyroidism, parental consanguinity was 1.5 times higher than among neonates without congenital hy- pothyroidism. Various studies have report- ed that congenital hypothyroidism is more prevalent among Asian families than non- Asian ones. Rosenthal et al. evaluated the prevalence of congenital hypothyroidism in the north-west of the United Kingdom (UK) among different races, nationalities and minorities, especially Muslims and Asians [2]. They found that the prevalence rates of congenital hypothyroidism among Asians and non-Asian families were l in 918 and l in 3391 live births respectively. This significant difference may be a result of parental consanguinity among the Asian population. Congenital anomalies, mortality and morbidity were also more prevalent in hypothyroid neonates [2]. Another study in the UK, from 1981 to 1991 showed that the prevalence of congenital hypothyroidism among Pakistani families with consanguin- eous parents (l in 781 live births) was sig- nificantly higher than the total prevalence of congenital hypothyroidism in the UK (l in 2154 live births) [3]. In addition, congenital hypothyroidism prevalence among Indian neonates without consanguinity was l in 5540 live births. A study in Israel showed that the inci- dence of congenital hypothyroidism was higher in Arab families with familial mar- riage than those reported from industrialized countries, but was similar to those found in Saudi Arabia [19]. This may be due to the high degree of consanguineous marriages among Arab populations. Also a study in Saudi Arabia found that congenital hy- pothyroidism was 1.8 times more common in rural areas and they believed that con- sanguinity was the cause [18]. Ordookhani and colleagues have shown high prevalence of familial marriage among cases of con- genital hypothyroidism [11]. Only a few studies have shown no significant relation- ship between consanguinity and congenital hypothyroidism [20]. Overall these findings suggest a role of consanguinity in congeni- tal hypothyroidism and other congenital anomalies. Recently Ordookhani and colleagues reported a high prevalence of consanguinity among patients with permanent congenital ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما hypothyroidism [21]. They concluded that parental consanguinity may be considered a causative factor for the high prevalence of thyroid dysgenesia, which was the common- est cause of permanent congenital hypothy- roidism in their study. Overall, 28.6% of all screened neonates and 47.1% of hypothy- roid patients had parental consanguinity. Our findings (27.7% of all studied neonates and 36.9% of congenital hypothyroidism patients with consanguineous parents) are similar to these. In our study, parental consanguinity, especially with 1st cousin relation, was more prevalent among hypothyroid ne- onates. We did not study the prevalence of transient and permanent hypothyroidism because our patients with congenital hy- pothyroidism had not reached 3 years of age which is necessary for determination of transient and permanent forms of con- genital hypothyroidism. Nonetheless, our findings support the role of consanguinity in the high prevalence of overall congenital hypothyroidism. According to our study, overall parental consanguinity and 1st cousin parental con- sanguinity were present in 47.4% and 39.5% of all infants with dysgenesia (agenetic and ectopic), which is in line with the results of a recent study in Tehran [10] which reported overall and 1st cousin parental consanguin- ity of 55.6% and 33% among congenital hypothyroidism patients with dysgenesia. Although the rate of transient and perma- nent congenital hypothyroidism has not yet been determined, patients diagnosed with dysgenesia are considered permanent congenital hypothyroid. The remainder with normal thyroid scan can be considered transient pending additional studies that will be performed at 3 years of age. Considering different etiologies of con- genital hypothyroidism, dyshormonogenesis is inherited through an autosomal recessive pattern but dysgenesia is a sporadic disor- der. The causes of thyroid dysgenesia are unclear, but are believed to be multifacto- rial, including environmental and complex gene interactions [22]. A recent study has reported that familial factors affect 2% of all dysgenesis cases [23]. As with the study of Ordookhani et al., nearly half of our patients with congenital hypothyroidism with dysgenesia had pa- rental consanguinity which has not been reported by other studies. This may reflect the presence of some unknown mutations in the genes involved in thyroid ontogeny in our population. Considering the high prevalence of congenital hypothyroidism in our country, consanguinity, especially with first degree relatives, clearly appears to have a role in this increased prevalence. More studies are needed in this area. At the same time, interventions are needed to try and reduce the rate, such as public awareness activities to increase people’s knowledge about the condition and the risk of familial marriage. This study was funded by the Research Bu- reau, Isfahan University of Medical Scienc- es and Health Services (grant no. 80350). It was carried out partly in Amin Hospital and partly in Isfahan Endocrine and Metabolism Research Centre. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Child and Adolescent Health and Development website This website aims at sharing information on the work on the health and 5 psychosocial development of children under- carried out by the WHO Regional Office for the Eastern Mediterranean and related activities and initiatives promoted and carried out in countries in the Region. It is intended for public health and programme managers, academia, civil organizations and nongovernmental organizations, students and anyone interested in public child health issues, with special focus on the Eastern Mediterranean Region. The CAH website provides detailed information on the coverage of child health activities at different levels in countries in the Region, especially the implementation of the Integrated Management of Child Health (IMCI) strategy. Specific sections describe the evolution of IMCI over the years in the Region, the process and guidelines to develop national child health policies, work carried out in the area of pre- service education related to child health, health systems issues, the community component, the child health-related Millennium Develop- ment Goals, research work, and advocacy initiatives. The website can be accessed at: http://www.emro.who.int/cah/ ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Refractive error (RE) leading to low vision is one of the most common problems of vis- ual impairment, especially among school- children, and frequently remains undiag- nosed for long periods. The World Health Organization (WHO) has grouped uncor- rected RE with cataract, glaucoma, tracho- ma and macular degeneration, infectious disease, and vitamin A deficiency among the leading causes of blindness and vision impairment in the world [1]. In Egypt, a preliminary national survey done in the Helwan area of Cairo reported that 34% of the recorded disabilities were visual dis- ability [2]. Correction of RE and low vision are one of the priorities of global initiatives for Vision 2020 [3]. In Egypt there are few studies showing the prevalence of errors of refraction. The present study is a prelimi- nary survey to determine the prevalence of RE and low vision among schoolchildren in Cairo, which is the capital of Egypt, with the highest population density in the coun- try. It is the first study on a large sample (6000 students) in 4 educational sectors of different socioeconomic level. Studies have been done in different governorates with smaller populations and in smaller samples (not more that 1000–1500) students [4–8]. This is also the first study to train school doctors and school nurses (as all the schools are covered by the Medical Insurance Or- ganization, the governmental organiza- tion that funds the medical services of all school-age children in primary, preparatory and secondary schools). This was a cross-sectional descriptive study of a cluster random sample of 12 govern- ment schools. The schools were chosen using a multistage random sampling tech- nique. The schools of Cairo governorate were divided into clusters based on Ministry of Education geographic divisions. Four ed- ucational geographic divisions were chosen randomly (El-Waiely, Mataria, Heliopolis and Nozha). The schools of each of the selected educational geographic divisions were stratified into primary schools, preparatory schools for males and preparatory schools for females. Subsets of school clusters were randomly selected from the 3 school strata of the chosen educational areas, i.e. 1 pri- mary school, 1 male preparatory school and 1 female preparatory school independ- ent of their numbers. All the students of the selected schools were included in the screening (age range 7–15 years). Private schools were excluded from the study as their health systems are different from that of the government schools and there is a private school doctor for each school. Information on parents’ education, oc- cupation and employment status was also obtained from the parents of each child by questionnaire, to estimate the socioeconom- ic status according to Park and Park [9]. The health insurance school health physi- cian and the school nurse of each school included in the study were trained by the au- thors how to measure the visual acuity (VA) of the students. Assessment of VA of all the students was done with a Landolt broken ring chart at 6 metres in a well-illuminated room. VA was measured with and without glasses. Children with VA 6/12 or less with or without correction were examined by pinhole test to evaluate the improvement of VA. The school doctors were also trained to evaluate ocular balance, strabismus by cov- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ered and uncovered test, and ocular motility in the 6 cardinal positions. The children with VA 6/18 in both eyes with best correction were categorized as low vision. Children with uncorrected VA were referred to the medical insurance clinics. The type of RE was obtained from the ophthalmic prescription. The collected data were analysed using SPSS, version 7.5. The prevalence of RE among the screened students was estimated. Pearson chi-squared test was applied and differences were considered significant at P < 0.05. From the school records, 6000 students aged from 7 to 15 years were selected to be included in the study; 161 students (2.7%) refused to participate. The remaining 5839 students (3113 females and 2726 males) were screened. The screening revealed that 1292 of the 5839 students (22.1%) had RE (VA 6/12), and 728 (12.5%) had low vision (VA 6/18) (Table 1). Strabismus was found in only 42 students (0.7%). Of the children with RE, 55.7% were myopic, 27.3% hypermetropic and 17.0% astigmatic. Table 2 shows the age distribution of the children with RE: 85.4% were aged 12+ years. Figure 1 shows that the prevalence of RE and low vision were significantly higher among female students compared with males (P < 0.05). Among the students with RE, 42.3% wore glasses and 57.7% had no glasses. The proportion of students with RE and without glasses was slightly higher among families of low socioeconomic status (55.0%), than among families of middle socioeconomic status (52.7%), but this was not significantly different (P > 0.05). A preliminary national survey of disabilities in Egypt in Helwan, Cairo, reported that 34% of the recorded disabilities were visual disability [2]. The present study was a preliminary screening to determine the prevalence of refractive error (RE) among schoolchil- dren from 4 different randomly selected educational geographic divisions. We found the prevalence of RE (VA 6/12) was high—22.1% of school students aged 7–14 years—and 12.5% of them had low vision (VA 6/18). The frequency of students with low vision was mostly among the preparatory-school students. A previous study of primary-school children in one area of Cairo (Shubra) diag- nosed RE in 21.8% of all examined children [5], whereas another study detected a higher prevalence of RE (36.8%) among primary- school children in Giza governorate [6]. In Menofiya governorate it was reported that 17.5% of primary-school children had RE [4]. In Tanta governorate, RE was found in about 39% of 511 primary-school students, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما amblyopia in 0.8% and strabismus in 0.6% [7]. In Al-Minya governorate a study on a random sample of 1588 schoolchildren aged 7–15 years found RE in 11.9%, amblyopia in 3.6% and strabismus in 1.6% [8]. RE may be viewed as resulting from a combination of genetic and environmental factors [10]. Myopia has reached epidemic proportions in many countries such as Ja- pan, Hong Kong, Taiwan and Singapore [11]. The present study found that myopia was the most prevalent type of RE among the screened schoolchildren (55.7%), fol- lowed by hypermetropia (27.3%), and astig- matism (17.0%); strabismus was found in only 0.7%. In a previous study, only 9.2% of the children with RE wore glasses [7]. But the present screening revealed that 42.3% had glasses, although family income had no significant effect. Not wearing glasses may lead to a greater deterioration in VA of the affected children. Thus, awareness about the importance of visual correction may play a significant role in proper manage- ment of RE cases. Further study about the public awareness of early detection of RE and proper management is suggested. The main recommendations from this study are: • Development of national programme for early detection of visual impairment, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما involving both preschoolers and school- children. • Establishment of a school screening pro- gramme and follow-up for RE and low vision involving both preschoolers and schoolchildren. • Governmental support for providing low-cost spectacles and school sup- port to encourage children to wear their glasses. This study was supported by the World Health Organization as a part of the Vision 2020 programme. We would like to thank Professor Medhad Mokhtar, Ministry of Health and Population for his helpful sup- port. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 0482916 9162.7% 1.6%5.5% 1.41000 40%Aamblyopia 32.5% ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Five per cent of worldwide blindness in- volves children younger than 15 years of age; in developing countries this age group constitutes 50% of the population. By World Health Organization (WHO) criteria, there are 1.5 million children worldwide who are blind: 1.0 million in Asia, 0.3 million in Africa, 0.1 million in Latin America and 0.1 million in the rest of the world. There are marked differences in the causes of child- hood blindness in different regions due to different socioeconomic factors. In devel- oping countries, 30% to 72% of such blind- ness is avoidable: 9% to 58% is preventable and 14% to 31% is treatable. The leading cause of blindness in children in developing countries is corneal opacification caused by a combination of measles, xerophthalmia and the use of traditional eye medicine [1]. Infectious diseases, malnutrition, vitamin A deficiency and measles are the main causes of avoidable blindness in children of poor countries [2]. In Sudan there is an absence of popula- tion data on the prevalence of blindness in children. However, the prevalence of blindness in all ages in Sudan was found to be 1.78%, the major causes being cataract, corneal opacities, uncorrected refractive errors, trachoma and onchocerciasis [3]. It has been found that the causes of blindness in children are the same as the causes of under-5 mortality (e.g. premature birth, measles, vitamin A deficiency, con- genital rubella syndrome and meningitis) [4]. Therefore the under-5 mortality rate can be taken as a proxy indicator to estimate the prevalence of child blindness. In Sudan, the under-5 child mortality rate was found to be 104/1000 live births. The number of displaced people in Su- dan has been estimated to be about 4 million [5]. The primary reason for leaving their origin was war, followed by food insecurity, economic opportunities and employment. Approximately 340 000 internally displaced persons (IDPs) live in the 5 official camps, which are located at the peripheries of 3 large cities (Khartoum, Khartoum North and Omdurman). They live in cardboard or mud houses in poor hygiene conditions and with a lack of water supply [3]. The re- sults of an interagency household survey in displaced camps in Khartoum indicate that the crude mortality rate for children under 5 years is near the emergency threshold of 1 per 10 000 per day. The main cause of death in all areas is diarrhoea. Other causes included chest infections, pneumonia and measles [6]. Clearly living conditions are poor in the camps and child mortality is high, which suggests visual impairment may also be high among such children. We aimed to determine the prevalence and leading causes of visual impairment and blindness in children in the above- mentioned camps. The outcome of this study will provide baseline information that may assist in measuring the progress of the VISION 2020 programme in fighting causes of childhood blindness in the country. Children under 16 years who were living in the IDP camps around Khartoum for at least a 6-month period before the study were the target population. The 5 IDP camps in Khartoum State are Mayo Farms, Jebel Awlia, El Salaam, Wad el Bashier and Karton Kassala with a total populations of 72 621 during the study period (5–10 October 2003). Each camp was divided into 30 clusters, each having between 34 and 66 households, i.e. average population per cluster ranged between 408 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما and 792 individuals. Every 7th household in each cluster was visited. The eligible children were those un- der 16 years who had been living in these households for a minimum period of 6 months before the survey and contracted eye problems from birth until the time of the study. Health workers who were trained in conducting field surveys to detect eligible children with eye problems or defective vision were involved in the house-to-house surveys. The health worker asked the par- ents whether their children had recent or previous eye problems, or if they had no- ticed any change in the eyesight of their children. Visual acuity was not measured during household interviews, but children with eye problems were registered and the heads of the households were given a card and requested to be present with their children on pre-determined days in the camp clinics to see the ophthalmic medical assistants and optometrists to assess their eye problems and visual acuity. Children found by the field health workers to have eye problems were later examined in the camp clinics by an examination team using portable equipment. The examining team in each camp consisted of 4 ophthal- mic medical assistants, 1 optometrist, 2 medical officers, who were trained in the eye hospital, 2 health workers from the local health units of the camps, 5 medical students, a records clerk and a driver. At the camp clinic, each child was first registered by the health workers and a history was taken from each child or their parents by the health workers or medical students. Visual acuity was measured in each eye separately at 6 m and 3 m using a Snellen E chart. Im- pairment in visual acuity in children under 5 years was determined by showing bright objects. The following WHO categories of visual loss were used for visual acuity in the better eye: 6/18 or better – not impaired; < 6/18–6/60 – visual impairment; < 6/60–3/60 – severe visual impairment; < 3/60 – blind [7]. Trained ophthalmic medical assist- ants using a 4× magnifying loupe and a hand-held flashlight examined the anterior segment of the child’s eye. Posterior seg- ments were not examined at this level. Acute eye problems were treated free of charge. Children with complicated eye dis- eases or visual acuity < 6/18 were referred to Alwaldain Eye Hospital, Omdurman Province. Transport was offered free of charge. At the referral eye hospital, the ophthal- mologist reviewed the field results. In chil- dren aged 5 years or more the external eye and anterior segment were examined using a slit lamp and posterior segments were ex- amined with a direct ophthalmoscope after dilating the pupils, if necessary. A diagnosis of glaucoma was made on the basis of an- terior and posterior segment signs, as visual testing was not possible. Glaucoma cases were assessed using a Schiotz tonometer to assess intraocular pressure. Visual acuity was measured again and refraction was performed for all children. If there was difficulty measuring visual acuity in children under the age of 3 years of age, they were examined and refracted under local anaesthesia. For each child with a visual acuity of < 6/18 in the better eye, the examining oph- thalmologist sought to identify the reasons for visual loss. All the findings were recorded on a data sheet and were entered into a database. The data were analysed using SPSS, version 10. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The target population included 29 048 children under 16 years of age. Of these, 1115 children were registered by the health workers as having a history of recent or previous eye problems during the house- to-house interviews. Of these, 916 came to the camp clinics where they were examined and treated by the field team; 105 of them were found to have visual acuity < 6/18 and were referred to the eye hospital for further refraction and examination. Out of the 105 referred to the eye hos- pital, 15 were found to be severely visually impaired (visual acuity < 6/60 in the better eye) and 25 blind (visual < 3/60). The prev- alence of severe visual impairment among the referred children was 1.6% whereas the prevalence of blindness was 2.7% and the prevalence of blindness and severe visual impairment together was 4.4% (Table 1). The prevalence of blindness was higher in males compared to females although this difference was not statistically significant, but it was significantly higher in the older age group than in the younger age groups in males (P < 0.05) (Table 2). The commonest anatomical sites of visual loss in the 40 blind and severely visually impaired children were the corneas in 40% (corneal scaring), the higher centre of the brain in 32.5% (amblyopia), the lens (cataract) in 12.5%, retina in 7.5%, the whole globe in 2.5%, optic nerve in 2.5% and uvea (glaucoma) in 2.5% (Table 3). In all, 35 (87.5%) of the blind or severely visually impaired children presented with avoidable causes of blindness and vision loss: vitamin A deficiency (22.5%), trauma (10.0%) and measles (7.5%), which are preventable causes, and amblyopia (32.5%) and cataract (12.5%), which are treatable causes (Table 4), although amblyopia is only treatable up to 9 years of age. Of the 40 children with severe visual impairment and blindness, 23 were referred for surgery, 10/23 (43.4%) for corneal scar- ing and 8/23 (34.8%) for cataract surgery (Table 5). The prevalence of blindness in children varies between countries as it is influenced by socioeconomic status, and it is reflected by the childhood mortality rate. While eye lesions are the main causes in high-income countries, corneal scarring due to measles and vitamin A deficiency are the main causes in low-income countries [1]. In our study the prevalence of blindness and se- vere visual impairment in children in dis- placed camps in Sudan was 1.4/1000, which is consistent with the prevalence of child- hood blindness in low income countries with low under-5 mortality rates, where the prevalence may be as high as 1.5 per 1000 children [8] Our rate is similar to some popula- tion-based studies in other countries. For example, the prevalence of blindness was 1.5/1000 in the age group of 5–15 years in Chile but the main causes of blindness were not available [9]. In India it was found to be 1.2/1000 in the age group of 7–15 years, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما where retinal disorders, corneal opacities, congenital anomaly and amblyopia were the main causes of blindness [9]. In Nepal the prevalence of blindness was found to be 1.5/1000 in the age group of 5–15 years, but the main causes of blindness were not available [9]. Examining the main causes showed that corneal scarring (mostly related to vitamin A deficiency disorders) and lens opacity (cataract) were responsible for 40% and 13% of childhood blindness and severe visual impairment respectively. These 2 conditions are entirely avoidable (preven- tion for vitamin A deficiency disorders and early diagnosis and surgical treatment for cataract). Our result is consistent with the recommendations and priorities of WHO VISION 2020 initiative which identifies corneal scarring and cataract as the major causes of childhood blindness in most de- veloping countries [7]. The majority of surgical operations that were done at the eye hospital were for cor- neal opacities and cataract, which suggests ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما the need to initiate a training programme on paediatric eye surgery for the doctors, so as to prevent further complication to blindness. Although 25 blind cases and 15 with severe visual impairment are a small number, not all cases were determined by initial screening and thus these figures are an underestimate of the number of cases. Nonetheless out data provide useful base- line information for programme planners involved in addressing childhood blindness in Sudan. Improving vaccination for measles and distribution of vitamin A supplements will assist in prevention of severe visual im- pairment and blindness in the majority of children in these camps. A public health eye programme based on VISION 2020 should be established to combat childhood blindness problems focusing on the primary health care level. This study was supported by Muslim Aid UK under their programme of Fighting Child Blindness. Special thanks are extended to the Min- ister of Health, Khartoum State and the staff of the Alwaldain Eye Hospital for their great help during the study. We would also like to thank Mrs Sulafa Ahmed for her valuable help as optometrist. Finally, we thank the community leaders in all the camps in Khartoum State for their support and assistance. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Hearing loss in infants and children may be sensorineural, conductive, or mixed uni- lateral or bilateral and symmetric or asym- metric. It can also be syndromic (involving other identifiable features) or nonsyndromic (isolated hearing loss); congenital or post- natal; prelingual, perilingual or postlingual (i.e. onset before, during, or after speech or language acquisition); and genetic or nongenetic [1]. Sensorineural hearing loss is an extremely common disorder, with a spectrum of effect ranging from an almost undetectable degree of disability to a pro- found alteration in the ability to function in the society. Because its onset is frequently insidious and accompanied by subtle com- pensatory strategies, hearing loss is usually overlooked by physicians and patients [2]. The prevalence of hearing loss in several countries has been estimated as 4%, 7.9% 15% and 10.8% in Saudi Arabia, Pakistan, Kenya and the United States of America respectively [3–6]. It was estimated that 3.9 million children in Africa had mild hearing loss and 1.2 million had moderate to severe hearing loss [7]. In Egypt, there is a scarcity of studies estimating the prevalence of hear- ing loss. The risk factors for conductive and sen- sorineural hearing loss are mutually exclu- sive. Risk factors for the former include middle ear infection, trauma to the tym- panic membrane, foreign body impaction in the external canal and consanguineous marriages, while those for sensorineural include viral neuritis, fracture base, ototoxic drugs, noise exposure and tumours of the cerebello–pontine angle [3,8]. In Europe, as in sub-Saharan Africa, the most prevalent causes of hearing impairment are chronic and suppurative otitis media [7,9,10]. There is a paucity of data regarding the risk factors of hearing loss in Egyptian adolescents. Thus, epidemiologic studies are needed to assess hearing loss in this important age category for setting priorities and designing efficient interventions. The main objective of this study was to estimate the prevalence of hearing loss among secondary-school students in an Egyptian city, and to identify the risk fac- tors associated with it, in order to enable specific preventive measures to be targeted at students with these risk factors. There are around 1.5 million inhabitants of Ismailia city. All secondary schools in the city (19 schools) were included in the study: 10 general schools, 5 commercial schools and 4 technical schools. From 2 classes per school year, 10% of the students in each school year were randomly selected. Over the period September 2003 to July 2004, 2750 students were invited to participate in this study, of whom 2633 accepted (re- sponse rate 95.7%). No significant differ- ence was found between participants and non-participants regarding their age or sex. Students were asked to complete a stand- ardized interview questionnaire, whom in- cluded demographic data (age, sex, school type, number of siblings and order of birth), clinical symptoms and the main risk factors for hearing loss. These factors included family history of consanguinity between parents or hearing problems, and history of ear surgery, infections, trauma, admission to fever hospital and intramuscular injection of antibiotics. All participants were subjected to Weber and Rinne tests for assessment of hearing disabilities. According to the findings of these tests, a student was considered to have normal hearing, conductive hearing loss, sensorineural hearing loss or mixed ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما conductive and sensorineural hearing loss. Only 44 (1.7%) students had an ambigu- ous diagnosis and were excluded from our analysis. Comparative analysis between students with coherent and incoherent di- agnoses showed no statistically significant differences regarding their demographic data or medical and family past history. Initial comparisons between students with different hearing loss and those with normal hearing tests were done using the Pearson chi-squared test for categorical variables. Risk factors for conductive hearing loss are completely different from those for sensorineural hearing loss. Thus, conduc- tive and sensorineural hearing loss were considered as 2 distinct pathologies and we conducted a separate multivariate model for each of them. The multivariable models to describe the risk factors for conductive and sensorineural hearing loss were obtained using stepwise logistic regression analysis initially with a conservative significance level of 0.25. Our dependent variable for this analysis was the state of hearing (whether normal or abnormal) according to Weber and Rinne tests findings. Subsequently, the resulting model was reduced using a likelihood ratio test with a significance level of 0.05. The calibration of the final model was assessed using the Pearson chi-squared goodness-of- fit test, and its discrimination was assessed by the area under the receiver-operating curve. All statistical analyses were per- formed using SPSS, version 11.0. A total of 2589 secondary-school students were included in our study; 1713 (66.2%) were from the 1st grade school year, 430 (16.6%) from the 2nd grade and 446 (17.2%) from the 3rd grade. Nearly half of the stu- dents (49.8%) were from general secondary school and the rest were from commercial (24.8%) or technical school (25.4%). The mean (standard deviation) age of the stu- dents was 15.4 (SD 1.2) years (range 13 to 21 years) with a female to male ratio of 1.3:1. The mean number of siblings of the students was 3.2 (SD 1.6) and only 632 students (24.4%) reported consanguinity between parents. A total of 576 students out of 2589 (22.2%) had hearing loss. Of these, 403 (70.0%) had sensorineural hearing loss, 82 (14.2%) bi- lateral conductive hearing loss, 91 (15.8%) mixed sensorineural and conductive hearing loss. Tinnitus, vertigo and hearing difficulty while talking to friends were the 3 com- monest symptoms related to the different diagnoses of hearing loss. The percentage of students diagnosed as sensorineural, conductive, and mixed sensorineural and conductive hearing loss and complaining of tinnitus were 71.7%, 58.5% and 71.5% respectively. Meanwhile, vertigo and hear- ing difficulties while talking to friends were reported in 51.9% and 50.1% of students di- agnosed as sensorineural, 53.6% and 48.7% of conductive and 64.9% and 57.2% of mixed hearing loss respectively. Table 1 shows the association between the sociodemographic characteristics of the studied groups and different types of hearing loss. Mixed sensorineural and con- ductive hearing loss was higher in females (82.4%) than males (17.6%) compared with those with normal hearing and this differ- ence was highly statistically significant (P < 0.001). Otherwise, no sex differences were ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما noticed in both sensorineural and bilateral conductive hearing loss. The percentage of students at general school diagnosed with bi- lateral conductive hearing loss (65.9%) was higher than that of students at commercial (24.4%) and technical schools (9.8%) com- pared to those with normal hearing. Moreo- ver, the percentage of students at technical school diagnosed as mixed sensorineural and conductive hearing loss (46.2%) was higher than that of students at other school types (28.6% at general schools and 25.3% at commercial schools). The mean number of siblings was only significantly higher in mixed sensorineural and conductive hearing loss compared to those with normal hearing. No association between order of birth and any type of hearing loss was found (data not shown). Past history of ear surgery, otitis media, admission to fever hospital, intramuscular injection of antibiotics and ear disease treat- ment were statistically associated with the presence of sensorineural hearing loss (Ta- ble 2). Similarly, history of ear surgery, ear disease treatment, postnasal discharge and trauma were statistically associated with the presence of bilateral conductive hear- ing loss. Mixed sensorineural and bilateral conductive hearing loss was significantly associated with a history of postnasal dis- charge, otitis media, measles and mumps, intramuscular injection of antibiotics and ear disease treatment. No association of consanguinity was found between family history of hearing problems and the pres- ence of hearing loss. The multivariate logistic regression model identified the number of attacks of otitis media, history of ear disease treatment, history of admission to fever hospital and history of ear surgery as the independent risk factors for sensorineural hearing loss (Table 3). The risk factors possibly associated with conductive hearing loss in the final mul- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما tivariate analysis model are presented in Table 4. A history of ear surgery carried the highest risk for conductive hearing loss (odds ratio = 4.06). The importance of early detection of hear- ing loss that may interfere with the process of learning has been repeatedly reported [11–13]. Our study aimed to estimate the preva- lence of hearing loss among secondary- school students, and to identify the risk factors associated with it. Before reaching conclusions based on the present results, it is necessary to consider a number of po- tential objections to our procedures. Simple tools such as questionnaires and tuning fork tests were used (Rinne and Weber test). These tools are characterized by high specificity and low sensitivity. For instance, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما the specificity of a questionnaire to detect hearing loss is 94%, compared with 62.4% for otoscopy and 84% for tympanometry. However, its sensitivity is 5 times less than the 2 previously mentioned methods [14]. Tuning fork tests are the traditional methods of differentiating conductive from sen- sorineural hearing impairments prior to the advent of pure tone audiometry. However, they cannot substitute for a correctly done pure tone audiometry with a full masking. The Rinne tuning fork tests have high spe- cificity and low sensitivity to detect conduc- tive hearing loss. Thus, when a Rinne test becomes negative it should be a reliable indicator [15–17]. Also, in about 25% of cases, the results of the Weber test do not agree with the results of pure tone audiom- etry. Its interpretation is sometimes difficult in the presence of bilaterally affected ears and it should be applied to unilateral hear- ing loss [15,16]. The results arising from these tools must be treated as provisional diagnoses that need confirmation by a more sophisticated battery of tests including pure tone audiometry and tympanometry, otoa- coustic emissions and augmented brainstem response. Of course, from the history and tuning fork examination we could suspect that a student may have conductive or sen- sorineural hearing loss, and they will be referred to audiological evaluation. The present study revealed a prevalence of all types of hearing loss of 22.2%. This prevalence is higher than the prevalences reported in other developing countries such as Kenya (15%) and Pakistan (8%) [3,6]. This could reflect different environmental exposures as there is a greater exposure to portable music devices, cell-phone use and high-powered output music speakers among young people in our environment. Segal and his coworkers [18,19] reported that up to 70% of hearing loss was sensorineural. This high proportion agrees with the results of the present study, where sensorineural hearing loss was the most prevalent loss, constituting 70% of the cases. A previous English study has shown that there is no effect of sex on hearing loss until age 31 to 40 years [20]. Similarly, our study showed that sex has no effect on either sensorineural or conductive hearing loss. However, it did have an effect on mixed sensorineural and conductive deaf- ness especially in females. There is no clear explanation for this and it could be due to a subgroup anomaly. Regarding the risk factors associated with hearing loss, we found that positive ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما consanguinity and presence of family his- tory of hearing problems among parents of the children were not significantly associ- ated with suspected sensorineural hearing loss. Similarly, O’Hara et al. reported that consanguineous marriages were not as- sociated with hearing impairment [21]. However, consanguinity was the major factor in other studies [22,23]. In our study, family history was not associated with the presence of sensorineural or conductive hearing loss. Billings and Kenna found a family history of sensorineural hearing loss or prematurity and/or complicated perinatal course in 28.6% of their patients [24]. Otitis media seems to have an impact on the development of hearing loss in school- children. In the present study a highly sig- nificant association was found between history of otitis media and the presence of sensorineural or mixed sensorineural and conductive hearing loss. These findings agree with the results of a recent survey conducted in schoolchildren in Green- land [25]. Similarly, previous studies in other developing countries, such as Brazil and Nigeria, have shown that a history of chronic otitis media in schoolchildren car- ried a higher risk for hearing loss [26,27]. Sensorineural hearing loss is found to be significantly associated with episodes of otitis media. These findings are in accord- ance with reports from Alaska, Canada, Greenland and Pakistan [3,25]. In our study, as in the study of Elahi et al. [3], no cases of hearing loss were at- tributable to viral infections such as measles and mumps. A retrospective study carried out in Turkey has shown that febrile illness was the major cause of hearing loss [28]. Similarly, the present study has demon- strated that previous admission to fever hospital was significantly associated with sensorineural hearing loss. In the univariate analysis, trauma was significantly associated with conductive hearing loss. In accordance with this find- ing, Cummings considered head trauma as one of the important risk factors for hearing loss [29]. The univariate and the multivariate analyses have shown that a history of otolaryngologic surgery carried a higher risk for sensorineural and conductive hear- ing loss. Previous studies have also showen a high risk of otolaryngologic surgery for different types of hearing loss [26,30]. The multivariate model in the present study showed that commercial school type was highly associated with conductive hear- ing loss. This might be due to the activities related to this school type, which include noise levels that may exceed the permissible values [31]. If students are being exposed, for example, to hazardous noise levels there is a need to obtain measurements of noise levels and supply students with ear protec- tion in the classrooms. The prevalence of the mixed type of hearing loss was almost the same as the con- ductive type. In the univariate analysis, in the group with mixed hearing loss, females represented 82% of the sample, although sex differences were not noticed in either of the other 2 types of hearing loss (conductive or sensorineural). Again, almost half of the students with mixed type of hearing loss were from technical schools. A factor that affects exclusively girls in technical schools should be explored. A history of measles or mumps was associated with the mixed type of hearing loss; an association that was not evident in the other groups with hearing loss. In conclusion, audiometric screening is highly recommended to detect hearing loss among secondary-school students. Not- withstanding the limitations of the tools ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما used in our study, the administration of a well-structured questionnaire at school en- try, complemented by tuning fork tests may be a practical option for an early detection programme in any developing country. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Most childhood tooth decay could be avoid- ed through simple preventive measures such as screening, monitoring, combined use of fluorides and dental sealants and regular professional care. These measures can save children pain, complicated pro- cedures and high dental treatment costs. In order to apply the measures effective- ly, public health care administrators and decision-makers need the tools, the capacity and the information to assess and monitor oral health needs, choose intervention strat- egies, design policy options appropriate to their own circumstances and improve the performance of the oral health system [1]. Dental screening data can be used to iden- tify trends and to assist in allocating scarce resources [2] and are an important indicator of a community’s health status [3]. One successful model of oral health programmes is school-based screening and education. School-based programmes are increasingly being seen as a strategic way to identify children who need dental care the most and thus improve access to health and social support services for vulnerable populations. Studies show that school den- tal screening is capable of stimulating dental attendance among children with a treatment need [4,5] and improving the dental health status of schoolchildren [6]. However, there are inherent problems and challenges with school-based screen- ing programmes. Al-Tannir discussed the importance of outreach activities in promoting awareness of a dental screen- ing programme and highlighted the need for personal contact by programme staff to ensure that children get the necessary follow-up [7]. Cooperative relationships between dental practitioners and the schools are required [8]. Furthermore, school-based dental screening programmes can be very time- and labour-intensive [9,10]. Other challenges include developing a standard- ized screening protocol, including both the establishment of standardized criteria for referral [11] and sufficient training and calibration of screeners [12]. It is important, however, to highlight that training does not fully resolve problems with consistent iden- tification of dental needs between screen- ers [13]. Another question is who should carry out the screening: dentists and dental hygienists [9] or primary care providers or paediatric dentists [14]. This paper outlines the oral screening programme that has been set up in public schools in 9 governorates in the West Bank region of Palestine, and details the evalua- tion of the effectiveness of the programme in minimizing oral health diseases, particu- larly caries. As a global concern, in 1979 The World Health Organization (WHO) announced that by the year 2000, the global average for dental caries was to be no more than 3 decayed, missing or filled teeth (DMFT) at 12 years of age [15] and in 1989 WHO en- dorsed the promotion of oral health as an in- tegral part of health for all by the year 2000 [1]. To achieve this, WHO urged Member States to establish oral health information systems and offered assistance in efforts to develop these systems and provide them with core standardized methods but with flexibility to expand by adding information relevant to the local situation. To meet national needs and interna- tional requests, an oral health screening programme was launched in West Bank public schools of the Palestinian Authority in 1997, aiming to screen schoolchildren ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما in the 3rd, 7th and 10th grades. It is a part of the national school health programme conducted by the Palestinian Ministry of Health (MOH) in collaboration with other international organizations. The programme is carried out and operated by the MOH’s 17 dental clinics that are distributed all over the West Bank and report to the central office in Ramallah. The programme started with oral screen- ing, educational seminars and students’ referrals to oral clinics operated by the MOH. After that, the programme developed into offering some necessary preventive procedures, such as fluoride gel application and fissure sealants for selected schools in school years 2002–03 and 2003–04. The programme uses WHO models and criteria [16] in screening for dental decay using the DMFT index and for gingival health using the CPITN index [community peri- odontal index of treatment needs]. Later, programme operators added fluorosis and malocclusion assessment to the screening programme. Although the dental screening pro- gramme does not have specific objectives, the school health programme aims to im- prove the general and oral health of school students and increase their access to health services. Programme effectiveness was meas- ured in 2 stages: short-term and long-term. Short-term outcomes can be demonstrated by increasing coverage rate, increasing FT component and care experience index (i.e. more services delivered and more follow- ups made), and increasing the percentage of referred students who actually attended oral clinics (i.e. success in raising awareness of oral health care, increasing follow-up and stimulus for seeking treatment). Long-term outcomes can be demonstrated by lower DMFT values, especially the DT compo- nent. This outcome needs more time to be apparent, because caries progression is a relatively slow process; it takes about 2 years from the initial attack of caries to be clinically evident and be counted as D in the DMFT index. Two methods of evaluation were used: quantitative (retrospective analysis of of- ficial records) and qualitative (in-depth interviews). Data for the evaluation were drawn from 7 annual oral health reports organized by the central oral health department in the Pales- tinian Authority MOH (school years 1997– 98 until 2003–04). These data provide a means for tracking changes in quantifiable behaviours. The oral health offices in the 9 Palestinian governorates send their monthly school oral health screening results to the central office in Ramallah, where they are organized and analysed into the final report. Each school in the designated governorate has the chance to be screened once a year. DMFT data (among other oral health indi- cators) are collected for each student in the selected grades. The data presented here were derived from routine dental examination conducted by the MOH dentists using visual methods without radiography or fibro-optic trans- illumination. In school year 2003–04 a grant from the United Nations Children’s Fund (UNICEF) supplied all the governorates with WHO standardized examination sets for oral screening. The methods used for the diagnosis and reporting of caries experience followed those published by the WHO for oral epidemiological studies [16]. The dental screening programme tar- geted public school students in 3rd, 7th and 10th grades in all the 10 governorates. In ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما this evaluation, only data for 7th and 10th grades were used, because for the first 5 years 3rd graders were screened then sub- stituted by 1st graders. The 7th grade (12- year-olds) and 10th grade (15-year-olds) are global monitoring cohorts; this allowed more permanent teeth to be included in the examination, and would facilitate interna- tional comparisons [17]. Only 9 governo- rates were used in this evaluation, because the 10th (Jerusalem) was added only in the last 4 years. Five indicators were used [18]: 1. DMFT index: the mean number of teeth decayed, missing or filled because of decay among schoolchildren. The index may refer to permanent teeth or baby teeth. It is a general indicator of dental health status among children and is felt to be extremely reliable; the lower the index, the better the dental health of the population. The index records past history and is cumulative [19]. To be recorded as decayed, teeth must have evidence of carious cavitations to the level of the dentine. 2. DT index: the number of decayed teeth divided by the number of examined stu- dents. This shows the caries experience by population. 3. Care experience index: the number of filled teeth divided by the number of examined students [(FT/DMFT) ×100/100]. This indicator assesses the delivery of dental services to the popula- tion. 4. Coverage rate: the number of students examined in each governorate divided by the number of students in that gov- ernorate. 5. Percentage of MOH oral clinic visits: the percentage of oral clinic visits by referred students (referred by screening teams). This shows the effect of oral school health programmes in stimulat- ing students and their parents to seek dental care and the amount of follow-up by the programme itself. This is a valu- able indicator to measure programme effectiveness. Analysis of the data extracted from the MOH’s 7 annual reports was carried out using SPSS software package. The second method used in this report was in-depth interviews with programme admin- istrators, dentists who conducted the oral screening and staff involved in interpreting and analysing data. Interviews comprised a series of questions, typically semi- structured or unstructured, conducted in person. A trained dentist with dental public health experience prepared and conducted those interviews. The advantages and disadvantages of the programme and other programmes operat- ing in the region, concerns about the imple- mentation of the programme, suggestions and recommendations to improve it and general oral health policy in the Palestinian Authority areas were discussed. The analysis of the data revealed the follow- ing outcomes for the 5 indicators. Coverage rate: All governorates showed a steady increase in coverage rates in the first 4 years and a boost in the last 3 years of the evaluation. Overall, the rate increased from 22.8% in 7th grade and 14.5% in 10th grade in 1997–98 to 92.4% and 88.2% respectively in 2003–04. The highest rates were obtained in school year 2003–04; coverage rates in 7th grade were consist- ently higher than 10th grade, as shown in ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Figure 1. A major increase in coverage rates after year 2000–01 occurred after the school oral health administration assigned local dental staff for each region to screen school students, saving time and cost of dentists’ travelling from one region to another. DMFT index: Trends in DMFT over time demonstrated a slight decrease in DMFT scores for the West Bank in the last 2 years (Figure 2). The mean West Bank DMFT scores for 7th grade ranged from 1.35 in 1997–98 to 2.17 in 2001 and from 1.6 in 1997–98 to 2.9 in 2000–01 for 10th graders. In school year 2003–04 the mean DMFT was 1.49 for 7th grade and 1.9 for 10th grade. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The lowest recorded mean DMFT for all areas of the West Bank was 1.35 in 7th grade students and 1.6 in 10th grade in school year 1997–98. The highest DMFT was 2.17 in 7th grade and 2.9 in 10th grade in school year 2000–01. In general, 10th graders had higher DMFT scores than 7th graders. DT index: When the components of DMFT were examined separately, however, we can see that DT was the major contribu- tor to the DMFT scores; FT was the small- est contributor. This was true in both 7th and 10th grades (Figure 3). Thus, although DMFT scores slightly improved over the 7-year period, the DT component had in- creased while FT and care index remained low. The measure of DT was higher in 10th grade than 7th grade children. Care experience index: The care index ranged from 11.9% to 15.5% for the 7th grade children over the 7-year period of the programme. Care experience indices were higher in 10th grade than 7th grade children (Figure 4). The 7th grade children had a care experience index of 14.8% in school year 2003/04 and 10th graders scored 20% in the same year. Percentage of referred students visiting oral clinics: The analysis showed that a small percentage of referred students at- tended dental clinics in school years 2002– 03 and 2003–04 (Figure 5). The percentage was lowest in Salfit and highest in Jericho. School year 2003–04 had lower values than the preceding year, 2002–03. The mean percentage of students who attended dental clinics increased from 9% in 2002/03 to 11% in 2003/04 school year. Regional variations A wide range of DMFT values were seen across the 9 individual governorates. When analysed by governorate, the highest DMFT was recorded in Salfit (7.8 in 7th grade in school year 2002–03 and 7.0 in 10th grade in school year 2001–02), and the lowest DMFT was recorded in Jericho (0.7 in 1999–2000) and Nablus (0.62 in 1998–99 in 7th grade). ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Comments on the programme performance, administration and equipment extracted from the in-depth interviews with the stake- holders highlighted a number of concerns. These are summarized on Table 1, together with recommendations to address the is- sues. The 92% coverage rate in the 7th grade cohort (12-years-olds) in 2003/04 could be considered a high figure when compared with an industrialized country such as the United Kingdom (UK). On average, 16% of the total population of children aged 14 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما years were examined in year 2003–04, 32% less than in the 1998–99 UK survey [20]. However, the coverage rate in the Palestin- ian authority MOH’s screening programme was calculated from the ratio of examined students to the total number of students. This ratio did not consider the total population in the same age group, which could include children who attend non-public schools or other forms of independent learning. The mean DMFT scores were 1.49 and 1.9 in 2003–04 in 7th and 10th grade respec- tively. The mean DMFT index compares favourably with some other countries in the region. Jordan scored 3.3 in 1999 [21], Lebanon scored 5.0 in 1996 [22], the Syrian Arab Republic varied between 1.4 and 2.5 in 2004 [23], Egypt scored 1.2 in 1991, and Tunisia scored 1.3 in 1994, Algeria scored 2.3 in 1987, and the Islamic Republic of Iran scored 4.0 in 1992 [24]. However, the West Bank compared less favourably with more developed countries; DMFT ranged from 0.63 to 1.31 in the UK in different regions [17]. Although high DMFT scores indicate high levels of the disease, unchanging DMFT or even a slightly increasing one might indicate a favourable change in care. This is true when all DT components con- vert to FT components; all the decayed teeth become filled teeth, which points to better treatment services. However, examination of the components of the DMFT index in the 9 governorates of the West Bank did not present this favourable change. DMFT scores in all areas over the 7-year period were largely attributable to the DT compo- nents, not the FT. This was also true in the last 2 years when DMFT showed a small reduction. In most of the governorates under study, a slight increase in the DMFT scores after a gradual decrease in the last 3 years was noticed; this could be due to altering the ex- amination strategy in school year 2003–04. Examiners switched from visual inspection to tactile inspection, using mirror and ex- plorer, which is considered more accurate in caries detection. Higher coverage rates demonstrated more representative samples and eventually more accurate DMFT values; this explains the boost in the DMFT index in the 9 gov- ernorates in year 2000–01. However, these report results could be more meaningful, if change indicators were weighted according to different coverage rates achieved. There were regional variations in DMFT values across the 9 governorates. The low caries levels in Jericho are likely to be due to the high natural fluoride content in their water resources. Nablus and Ramallah also demonstrated low DMFT and this could be attributed to 3 factors: • accessibility: dentists are easier to reach due to the wide availability of public transportation in these urban cities; • affordability: these 2 cities are important commercial and political centres in the West Bank, which translates into more job opportunities, and thus higher indi- vidual incomes; • availability: a high number of dentists work in private practice in these areas. Although the DMF scores of these gov- ernorates (Nablus, Ramallah and Jericho) were within normal levels or even lower than WHO recommendations (DMF < 3 for 12-year-olds), the percentage of DT was higher than the FT component in these governorates. This indicates the need to enhance dental care measures among stu- dents by increasing awareness of good oral hygiene practices. Meanwhile, the school oral health programme should focus on teeth filling at an early stage of caries and application of fissure sealants. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما In contrast, Salfit and Qalqiliah present- ed the highest DMFT scores over the whole programme period; however, due to lack of research, no risk factors have so far been identified in these areas. This great differ- ence in the DMFT values among individual governorates has implications for targeting interventions and services at populations that were identified at risk. The care index results (11.9% to 15.5%) could be considered very low when com- pared with UK values. The care index mean in the UK was 12% in 1995/96 then im- proved to 55% in 2002/2003 [17]. Care experience index reflects the restorative care received by those who have suffered disease; it therefore has to be viewed in con- junction with DMFT. These results are of interest in studying the provision of dental services to the age groups under study. Although the mean percentage of stu- dents who attended dental clinics increased (from 9% in 2002/03 to 11% in 2003/04), the results are still considered low, when compared with other regions in the world. In Northern Ireland, the school-based dental screening programme stimulated 45% of the screened students to attend dental clin- ics, in contrast to 27% of the control group [4]. In Davangere, India, the dental screen- ing programme stimulated 31% of screened students to seek dental treatment, compared with only 10% of the control group [5]. In the West Bank, the low percentage of referred students who actually visited dental clinics could be due to the following: • Accessibility problems due to difficul- ties in transportation and checkpoints, curfews and border closures at the Is- raeli West Bank barrier, especially for students living in rural areas travelling to the city for care. • Insufficient oral clinics operated by MOH that offer free treatments (only 17 clinics throughout the West Bank). • Oral clinics are clustered in urban areas. Suburbs and rural areas lack this type of service. • Inadequate services are provided in those clinics; very little dental special- ties are offered. • Oral clinics operated by MOH operate only from 08.00 to 14.00 hours, the same time period for school attendance. • Dentists who carry out curative treat- ment in the clinics are the same people who are responsible for conducting the school screening; thus they are not al- ways present in the clinics to treat the referred students. Given the political situation in Palestine, the programme operating teams face a number of critical challenges. Like other Palestin- ians in the West Bank, they suffered from the direct and indirect consequences of military activities and border closures. It is reported that the dental teams and com- munity health workers are prevented from passing through particular Israeli military checkpoints for days or weeks [25]. In ad- dition, checkpoints and road blocks which divide the West Bank into 300 separate clusters cut off 70% of the population from reaching essential health care services for weeks and months [26,27]. Some limitations were also encountered in the data collection process: • One of the most important issues that influenced the quality of data collection was the change in examination methods in school year 2003–04 from visual ex- amination only to the use of examination sets which contain probes and mirrors. • The percentage of referred students who actually visited oral clinics was avail- able only in the last 2 years. In addition the 10th governorate, Jerusalem, was ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما added to the screening programme in the last 4 years. • The available data from the MOH re- ports were not always complete because of staff changes and resignations. • Finally, continuous change in staff, dif- ferent training, and lack of commitment to WHO examination criteria [16] make comparability of the data difficult. The existence of the school oral health pro- gramme in the Palestinian Authority areas as a database describing school children’s oral health represents the commitment of the MOH to improving the school oral health care and services. The slight im- provement in DMFT scores in students in some governorates in the last 2 years can be considered progress in caries control. However, maintenance efforts are required to ensure that caries experience at the popu- lation level does not rise in disease-stable areas, and an increase in strategic effort is required to address the high caries level in high-risk areas. Many of the problems with delivery of the school oral health programme are symptomatic of the need for substantial improvement in programme management. The programme administrators and operat- ing teams have divergent understandings and expectations of their roles and responsi- bilities. This is particularly so for observing screening sites, control referrals and follow- ups, and for managing and analysing data. In summary, the MOH strategy for pub- lic dental health should use screening data to ensure that priorities for dental health are being properly identified and met, and that responsibilities for policy and operational activities are appropriately assigned and understood between different stakeholders. A national service plan should be developed by the MOH and other collaborators, includ- ing a reassessment of the appropriateness of the service planning principles in place, and whether the location and scale of public dental clinics established are meeting the needs of the eligible population. A national oral health promotion strategy should be launched which covers community educa- tion, development of an environment sup- portive of good oral health, facilitation of adequate access to fluoride, support for oral health research and development of a high quality trained workforce. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The assessment of patients in coma is a medical emergency. The cause should be identified and, where possible, corrected and the brain provided with appropriate protection to reduce further damage. It then becomes important to identify those pa- tients for whom the prognosis is hopeless and in whom the institution or persistence of resuscitative measures is inappropriate [1]. Most epidemiological studies of coma in children have focused on traumatic coma [2–4]. In the past few years, there have been an increasing number of studies that have looked at non-traumatic coma of different etiologies [5,6]. The aim of this study was to determine the incidence, etiology and outcome of coma, both traumatic and non-traumatic, in paediatric patients in Al-Hassa region, eastern province, Saudi Arabia, a highly populated region, to determine the com- monest causes of coma and death in these patients. This was a retrospective study of medical records conducted in King Fahad Hospital, Hofuf in Saudi Arabia. This hospital is the only referral centre for the whole of Al- Hassa region, which represents almost one- quarter of the area of Saudi Arabia and is the largest oasis in the world [7]. All children, aged 28 days to 12 years, admitted to the paediatric intensive care unit (PICU) with acute alteration of consciousness, between April 1999 and March 2002, were the sub- jects of the study. Depression of conscious- ness was assessed by the modified Glasgow Coma Scale for children (GCS) [8,9] as shown in Table 1. Investigations performed for patients of the study were guided by the clinical pres- entation of the patients and were determined by the consultant in charge of every child. Neurological outcomes [6] were determined as: intact (normal or no change from pre- morbid functioning; seizures, if recorded, are 100% controlled), impaired (alteration of tone, power or reflexes; cranial nerve dysfunction; ataxia; seizures; persistent vegetative states) or death. Data were collected, computed and sta- tistically analysed using Epi-Info, version 6. The chi-squared test was used for compari- son between traumatic and non-traumatic coma. The F-test was used to assess the significance of the relation between GCS and outcome. Differences were considered significant at P 0.05. During the study period, a total of 91 (10.5%) children out of 870 admissions to PICU were admitted with a diagnosis of coma, an average rate of 2.6 patients per month. There were 59 males and 32 females giving a male to female ratio of 1.8:1. Popu- lation census of the region showed that the number of children aged 28 days to 12 years was therefore 635 603 during the study pe- riod. The incidence of coma was therefore 4.77 per 100 000 population per year. Table 2 shows that preschool aged chil- dren constituted the greatest proportion of the cases (43.9%) and two-thirds of all the cases had a GCS of 8 or less. Figure 1 in- dicates that the commonest causes of coma were head trauma (42.9%), followed by infections (25.3%) and polytrauma (9.9%). Figure 2 shows the frequency of different types of infections; gastroenteritis was the commonest cause of sepsis-related coma. Pseudomonas aeruginosa, streptococcal pneumonia and Haemophilus influenzae were the most frequently isolated microor- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ganism from blood, cerebrospinal fluid or tracheal aspirate. Tables 3 and 4 show different outcomes (intact, impairment or death) in relation to the cause of coma and GCS score. Most studies on childhood coma have been done in developed countries and there are few comprehensive data from developing countries where 80% of the world’s chil- dren live [5]. Unsupervised activity is a major risk fac- tor for traumatic coma and its age-specific ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما incidence rises throughout early childhood with the highest rates in the preschool age group [10,11]. In our study, 10.4% of the patients with traumatic coma were in the infant age group compared to 27.9% with non-traumatic coma. Other studies report varying rates for traumatic coma. In an Indian study only 3.2% of the patients were below 2 years [12], in a Japanese study 12.5% of patients were infants [13] and in a Malaysian study 50% were infants [5]. In our study, trauma was the commonest cause of coma during the preschool and school ages and was accompanied by poor outcome (death or impairment) in more than half of the cases. These findings agree with earlier reports on traumatic coma [3,4,10]. Acute brain injury has been reported to be the cause of approximately 100 000 pae- diatric hospital admissions per year in the United States of America [14] and it is the leading cause of death in children older than 1 year [9]. In a Spanish 1-year study on chil- dren, 70% of deaths from traumatic brain injuries occurred within the first 48 hours and mortality ranged between 20% and 35% [15]. In another report on head trauma in children, 38% died and the average length of coma in survivors was 15.5 days; 29% of the survivors were unimpaired at follow up, 9% of had motor deficits but normal intellect and 9% had severe intellectual and motor problems [16]. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Infection was the commonest cause of non-traumatic coma in our study and the commonest cause of coma in infants. Our results about infection-induced coma are in agreement with an English study on non-traumatic coma where infections caused 50.5% of coma in infancy, com- pared to 33.7% and 31.5% in preschool and school-aged children respectively [6]. In a Malaysian study, about two-thirds of the total coma cases were due to infection [5]. The same study considered infection as the most important cause of childhood coma throughout the world. In a Japanese study, 74% of cases of coma were also due to infection, mostly viral [13]. In an Indian study the etiology of coma in 60% of cases with non-traumatic coma was central nervous system (CNS) infection including tubercular meningitis, encephalitis, bacte- rial meningitis, and others [17]. This contra- dicts some reports from developed counties which stressed the importance of hypoxic ischaemic encephalopathy and toxic/meta- bolic causes [11,18,19]. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The pattern of infection varies in differ- ent regions. In our study, the commonest infection was septicaemia complicated by septic shock (56.5%) followed by CNS infections (26.1%). The organisms isolated were diverse but the commonest were Pseu- domonas spp. and H. influenzae. Different infective agents predominate in other parts of the world, for example cerebral malaria in Africa [20] and dengue haemorrhagic fever in South-east Asia [21]. In Japan, measles virus, herpes simplex and rubella are important [13]. In Malaysia, H. influen- zae has been commonly implicated in CNS infection [22,23]. In an English study, Neis- seria meningitides was recovered in 47% of cases where a pathogen was identified [6]. Other causes of coma apart from trauma and infection were found in only 21.9% of our patients. They are heterogenous causes but metabolic disorders, poisoning and ac- cidental causes were the commonest. Acci- dental and neurological causes had the worst outcome in comparison with intoxication, which was followed by complete recovery. Although these findings are similar to previ- ous studies [5,6], the small number of cases did not allow for statistical analysis. In our study the overall mortality rate was 47.2%. The mortality rate was not statistically different in patients with GCS 8 (47.7%) versus those with GCS > 8 (42.3%). However, patients with lower GCS who survived had less favourable outcomes. Although GCS was lower if coma was due to a traumatic cause, the mortality rate in the non-traumatic group was nearly double that of the traumatic group (60.5% versus 35.4%). The reverse was true regarding the proportion with impaired outcome (16.3% versus 22.9%). This was supported by the finding of a mildly significant relationship between GCS and outcome in the non- traumatic group compared to a highly sig- nificant relationship in the traumatic group (Table 4). In non-traumatic coma, etiology rather than GCS was more significantly related to the outcome. Nayana et al. stated that in long-term prediction of outcome in acute non-traumatic coma, GCS is not useful [24]. However, verbal response, a component of GCS, correlates well with long-term functional outcome and intel- ligence quotient. In the Malaysian study, the overall outcome was poor and one-third made a full recovery, one-third recovered with neurological deficit and one-third suc- cumbed to the acute illness [5]. In a recent Indian study, 11% achieved full recovery, 54% showed neurological impairment and 35% died [24]. Survival was significantly better in patients with CNS infection. In the English study, the mortality rate was 45.7% [6]. We conclude that head trauma and infec- tions are the commonest causes of coma. The common reasons for poor outcome included septic shock, severe head trauma, accidental causes and metabolic disorders. GCS in traumatic coma and the specific etiology in non-traumatic coma were the most important prognostic factors for the outcome. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 12–14 2003–2004198136 9.6%7.1%13.2% 3217 93 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Depression in childhood and adolescence is among the commonest and more disa- bling disorders. It has been reported that childhood depression ranges from 2%–6% among the school-aged population, and it seriously affects childhood and may have serious repercussion in adult life [1]. Kler- man and Weissman have indicated that the onset of depression is occurring earlier in life today than in the past [2]. By the year 2020, depression is expected to become the single leading cause of disease burden worldwide [3]. The prevalence of depression in pre- pubertal children is estimated at approxi- mately 2% and 5%–8% for adolescents [4,5]. A community survey of Australian children found that 3.7% of boys and 2.1% of girls at prepubertal age had experienced a depressive episode [6]. The gender ratio is 1:1 in prepubertal children and increases to 2:1 female-to-male ratio in adolescents [7– 9]. In a large study of 1% of total students in preparatory schools in Alexandria, 10% of the students demonstrated depressive scores: 56% of these were girls compared with 44% boys [10]. Identifying one specific primary cause for depression among children and adoles- cents is hard since they are more sensitive to their environment [11,12]. Tension and conflicts come together triggering clini- cal depression with physical symptoms [13,14]. In another study in Alexandria, the psychological profile of students rating high on the depression scale indicated that a lack of communication and the presence of parent–child conflicts ranked first (23%), followed by parental conflicts (21%), and 30% had scholastic problems [15]. There has been considerable debate on the efficacy and safety of psychophar- macological drugs such as tricyclic anti- depressants (TCA) and selective serotonin reuptake inhibitors (SSRI). Some studies have shown empirical limitations compared with placebo concerning improvement of depressive symptoms and disorders among children and adolescents [16–19]. Psycho- logical treatments for child and adolescent depression, including cognitive behaviour therapy (CBT), are therefore now often recommended as a first line of treatment [20]. Several randomized trials attest to the efficacy and safety of CBT in the treatment of depression in children and adolescents [21–24]. Meta-analysis of psychotherapies that have been used in the treatment of child and adolescent depression revealed 63% of those receiving some form of CBT showed significant improvement of symp- toms [25,26]. The main aim of this study was to assess the prevalence of depressive symptoms among early adolescents, and to determine the clinical effectiveness of CBT in the management of depressive symptoms in school settings. The study was conducted during the aca- demic year 2003–2004 in 2 preparatory schools [one boys’ (n = 198) and one girls’ school (n = 136)] in a poor socioeconomic district in Alexandria. They were the only such schools in the district. All students enrolled in the first academic year (aged 12–14 years) were eligible for inclusion. A questionnaire and psychological tests for depression and self-esteem were adminis- tered as baseline screening (pre-test) after ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما having the consent for participation. We used the Arabic version of the Children Depression Inventory (CDI-Arabic version) [15] to measure the severity of depressive symptoms (score range: 0–54, cut-off 25) and the Coopersmith Self-Esteem Inventory [27] to measure the self-esteem (score range: 0–25, cut-off 13 = low self-esteem). A sociodemographic questionnaire was completed by the students which included questions on recent stressful life events over the past 6 months, sociodemographic data, and the type of discipline adopted and child interrelations at school and at home. The child’s scholastic performance was assessed through school grades and levels of achievement, which were obtained from school records. Children with the highest scores on the CDI scale (n = 32) underwent further clinical assessment to confirm depressive symptoms according to the Diagnostic and statistical manual of mental disorders (DSM-IV-TR) diagnostic criteria [28]. A more detailed questionnaire was com- pleted by the interviewers for parents and teachers of the 32 children with the highest scores to further assess sociodemographic data, global functioning, academic achieve- ment, peer relationships, communication skills with friends and the reason for fre- quent absences from school (if applicable). • Children who had been and still were taking antidepressant treatment or ther- apy over the past 6 months (1 girl) • Those who refused to join from the chil- dren with high scores (child or parental causes) (n = 15; 7 boys and 8 girls). • Children with suicidal intent and in need of further referral (1 boy). We used the Cognitive behavioral manual for use with child patients with depressive disorders [29] as a basis for therapy. The intervention sample was 7 boys and 10 girls (n = 17). Nine weekly sessions were conducted at their school during the recreation break; 1 group for the 7 boys and 2 groups for girls with 5 girls each. Each session lasted 60 minutes but some sessions for girls extended to 90 minutes. A set of rules was issued by the children before starting the intervention sessions for control, discipline, respect and confidential- ity of the sessions and were approved by their parents. The main theme for each session was introducing a task or a situation with a role-play approach to solve the different situations. Homework assignments related to the tasks discussed and required them to record at home in writing their own feel- ings towards the applied activity. Parents attended with their children the discussion and explanation about the intervention as an introduction to what would be applied. The first session was an ice-breaking session which was followed by 9 therapeu- tic ones focused on: emotional recognition, self-monitoring, self-reinforcement, activity schedule, communication and interpersonal skills, social problem-solving, cognitive reconstruction I, cognitive reconstruction II, and treatment review plan (final session). The skills were designed to help children learn to get along with peers and adults. We also included parents in some of the sessions where they learned how to encour- age healthy behaviour in their children and become better at providing consistent consequences for negative behaviour and praising positive behaviour. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Statistical analyses were performed using SPSS, version 8.0. The chi-squared test was used for testing the difference in the preva- lence between the sexes and the t-test for testing clinical effectiveness of treatment with CBT. The flow of participants through each stage of the study is shown in Figure 1. This shows that those who were the highest scor- ers on the CDI depression scale and eligible ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما were recruited to CBT programme (n = 17) ; 7 boys and 10 girls. The prevalence of depression was 9.6% among all children; 7.1% in boys and 13.2% in girls, giving a male:female ratio of 1:1.86. Table 1 shows the sociodemographic characteristics of both the high-scorer group (n = 32) and the low-scorer group (n = 300). There were no significant differences between the groups except for parental punishment ( 22 = 11.4, P < 0.0001) and stressful life events over the past 6 months ( 23 = 22.6, P < 0.0001). Table 2 shows the different mean CDI scores for the 32 highest scorers among all ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما age groups for both boys and girls. At 13 years, girls had more depressive symptoms than boys (t5 = 2.838, P < 0.0363). Intervention sessions lasted more than 60 minutes particularly for girls where they felt more suppressed; 8 out of 10 girls re- ported that the extra household duties and the limitations of outdoor creativity that are available to their boy siblings were the ma- jor complaints. In contrast, 5 out of 7 boys mentioned that their deprived families (be- ing the most disadvantaged of the families) made them feel isolated with limited social communication with peers. The improvement after the end of the intervention programme at the post-test was significantly greater than expected. The changes that occurred were statistically significant both on the depression scale (t2 = 9.02, P < 0.0001) and the self-esteem scores (t16 = 18.495, P < 0.0001) (Table 3). Table 4 shows that there were marked improvements in feelings of worthlessness, interpersonal relations and communication skills with staff and peers, interest in activi- ties and school achievement among girls (t2 = 10.9, P < 0.0001) and boys (t2 = 12.6, P < 0.0001). As regards overall improvement differences between girls and boys, boys showed greater improvement than girls which was statistically significant ( 23 = 46.7, P < 0.0001). In our study, we found the prevalence of depressive symptoms to be 9.6% which is of concern. Life events over the past 6 months (every-day stressors) and parental punish- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ment were significantly associated with such symptoms. Kalb and Raymond [30] reported family conflicts and poor inter- family communication and peer problems as stressor factors, while Eliot and Reis [31] considered family instability and insecure relationship as stressors. We found sex differences in the fre- quency of depression, and early adolescent girls were more likely to be affected than boys with a ratio of approximately 2:1. Several studies have reported that girls are more vulnerable to life transition stressors, conform to social rules and utilize internal- izing coping styles more than boys, i.e. they become socially withdrawn [8,32–34]. Pomerantz stated that parents using psychological control attempt to regulate their child’s psychological and emotional development [35]. By constraining verbal expression and invalidating feelings they reinforce the child’s depression. With lack of support, the child is less able to com- municate effectively and more conflicts in relationships emerge. Our study illustrated the association between the use of violent means of discipline by parents and high depressive symptoms as has been reported in another study in Egypt [36]. In addition, maltreated children are significantly more likely to have low cognitive functioning, poorer adaptation to school and academic failure [37]. There has been a wave of sentiment against the use of pharmacological inter- vention as a first-line treatment for child and adolescent depression [20,38]. We ob- served a general concern towards treatment, especially pharmacological drugs, among the parents of our students, either because of fear of drug dependency or social and psychological stigma. The start of CBT sessions revealed the conflict between children and their parents and suggested poor treatment outcomes. However, over the course of treatment, parents developed positive views of CBT (praising positive behaviour of their chil- dren) and children showed rapid response to the treatment as assessed by the pre- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما health issues related to children hindered different approaches directed towards the group with depressive symptoms. Depression among 12–14-year-old schoolchildren from deprived areas in Alex- andria appears to a problem that needs to be addressed and a school mental health pro- gramme should be implemented [40]. Fol- low-up studies with larger sample sizes and adequate control groups will be necessary to evaluate the long-term benefit of CBT. In addition, psychoeducation and parenting skills for parents [21] are recommended together with increased training for mental health and school staff on CBT as a simple and effective tool to decrease depressive symptoms among schoolchildren [41]. and post-intervention depression scores, which are the main indicators for effective short treatment outcome. Thus, our results confirm those reported by DeCupper and colleagues that CBT can be applied within school and is effective in treating children and adolescents with depressive symptoms [39]. Regrettably, because of the limited awareness of the parents of their child’s need for the programme, a large propor- tion of the parents of children showing depressive symptoms refused to enrol their child in the intervention programme. This resulted in only a small sample enrolled in the programme, which is was a limiting factor for the study. Furthermore, the lack of awareness of the school staff on mental ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Neuroleptic-induced movement disorders consistent with the term tardive dyskinesia (TD), which was coined in 1964 [1], were reported in the literature as early as the 1950s. It was described as a movement disorder caused by the prolonged use of neuroleptic drugs. The condition is defined as a disorder characterized by involuntary movements, which may involve oro-facial dyskinesia, coarse tics or choreoathetosis, with abnormal oro-facial movements being the commonest [2]. It usually appears while the patient is still on the offending agent or appears for the first time when the drug is discontinued or its dose is reduced. Reported prevalence rates of TD from across the world have varied widely. Rates from as low as 0.5% to as high as 65% have been reported [3–11]. Some of the confounders that have been proposed as contributing to such wide variability in- clude heterogeneity of the populations un- der investigation, lack of agreement upon definition of what constituted TD and, more importantly, the confounding effect of the type, dosage and duration of the offending agents, and the duration of the illness itself, both prior to and after the commencement of the drug [4,5,7,9,11–13]. Tardive dyskinesia remains an enigma without a convincing explanation to its underlying pathophysiology. Two of the most persuasive current hypotheses are the dopamine hypersensitivity hypothesis and the serotonin–dopamine antagonist hypothesis. The dopamine hypersensitiv- ity hypothesis states that neuroleptics may induce a compensatory dopamine hyper- function owing to the prolonged block- ing effect they have on the receptors. The serotonin–dopamine antagonist hypothesis maintains that drugs which have a high affinity for blocking serotonin receptors in the striatum, such as the new atypicals, may lead to increased release of dopamine, which attenuates the blocking effect of neuroleptics to dopamine receptors in the nigrostriatal system due to the inhibitory effect serotonin has on dopamine release in these areas [14–16]. Although there are no concrete biological or pathological find- ings which support these 2 hypotheses, the clinical empirical evidence lend them some good support. Despite the lack of concrete evidence for the underlying patho- physiology of TD, there have been a few consistently reported risk factors which were found to make some patients more vulnerable to developing TD if treated with neuroleptics. These include the prolonged continued use of neuroleptics, especially in large doses; polypharmacy; advancing age of the patient, particularly > 40 years; brain damage; strong negative symptoms or a strong affective component in schizo- phrenic patients; and the indiscriminate use of anticholinergic agents [7,8,12,17–22]. To the best of our knowledge, preva- lence rate of TD has not been recorded for patients from Arab countries diagnosed with psychosis and being treated with con- ventional neuroleptics. The aim of the study was to determine the prevalence rates of TD among Arab patients with psychosis who had been treated with conventional neuroleptics for a prolonged period. We also looked for any risk factors or protective factors among these patients. We carried out a retrospective cross- sectional study to compare patients with neuroleptic-induced TD with those who did not develop it under comparatively similar conditions. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما We selected the 4 hospitals which had psychiatric services situated in the western region of Saudi Arabia for this study. The Jeddah Psychiatric Hospital, which is the only general psychiatric hospital in the region and serves the city of Jeddah and the towns around, is a state hospital. It has 150 beds and 3 clinics a day, with 100 to 125 patients attending daily. There are 4 consultants, 8 psychiatric specialists and 3 senior registrars working in the hospital. King Fahad General Hospital is the largest state general teaching hospital in Jeddah, with about 900 beds. It is the main general hospital serving the inner city of Jeddah and the surrounding areas. Two psychiatrists provide the services, which are mainly out- patient clinics and consultation–liaison du- ties. Al-Noor General Teaching Hospital in Mecca, the second largest city in the area, is a state general hospital. There are 2 consult- ant psychiatrists and 2 associate specialists, whose main duties include outpatient clin- ics and liaison services in the hospital. They see about 60–75 patients in their clinics per day. King Faisal Specialist Hospital and Research Centre is in Jeddah; it has 1 full time consultant, 1 part time consultant and 1 specialist psychiatrist. Their main duties are outpatient clinics, liaison–consultation services and a limited number of inpatient cases as the hospital is mostly a tertiary care service hospital. All patients suffering from a chronic psychosis who were seen between Janu- ary 1997 and December 2000 in 1 of the 4 participating centres and who had been prescribed neuroleptic drugs for longer than 6 months were selected. Their medical records were examined to identify those patients who were eligible to be included in the study. Taking into consideration all the major drawbacks and pitfalls of a retro- spective, multi-centre study, the following inclusion criteria were used: • age 18–65 years; • diagnosed as suffering from 1 of the fol- lowing: schizophrenia, schizoaffective disorder or bipolar affective disorder that had been present for 3 years; • well-defined diagnostic criteria, includ- ing persistent delusions and/or auditory hallucinations, which must have been documented at least 3 times during fol- low-up; • patient had been taking 1 of the con- ventional (typical) neuroleptics continu- ously for 6 months; • patient had been taking only conven- tional (typical) neuroleptics and none of the new atypical antipsychotics had been used at any time prior to the time of inclusion in the study; • patient had been followed up in the same clinic for at least 3 years. A total of 866 patients were diagnosed with 1 of the 3 conditions and had been taking 1 of the conventional neuroleptics; 783 of them were concurrently taking > 1 anti- cholinergic drugs. Only 151 patients met all the inclusion criteria and only 51 of these (5.9%) were found to have some movement disorder that was consistent with TD. A further 59 patients (6.8%) were described as having symptoms of drug-induced Par- kinson disease. When only those with complete medi- cal records, including the more frequently and consistently reported risk factors, were identified, the records of only 151 patients [116 (67.8%) males and 35 (23.2%) fe- males] were suitable for inclusion in the final analysis. Among these, 115 were Ar- abs (27 of whom developed TD) and 22 were Afro-Arabs (10 of whom developed TD) (Table 1). It was not possible to reli- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ably determine the ethnicity of the other 14 patients. Forty-seven patients were > 40 years old, mean 46.4 [standard deviation (SD) 11.4] years and 114 were 40 years of age, mean 36.7 (SD 12.2) years. From the records, we found 51 patients [38 (74.5%) males and 13 (25.5%) females] had TD; all were described as having oro-facial movement disorder (Table 1). A statistically significant positive association with TD was found only with advancing age of the patients [mean age of those with TD was 46.4 (SD 11.4) years compared with mean age 36.7 (SD 12.2) years for those without TD ( 2 = 4.72; P < 0.01)], longer duration of illness (P < 0.001) (Table 1) and longer duration of treatment (P < 0.001) (Table 2). As almost all our patients had been pre- scribed > 1 drug, we calculated the daily intake in terms of approximate equivalents to chlorpromazine in accordance with the British National Formulary [21]. Only 2 of the drugs used, chlorpromazine ( 2 = 13.70, P = 0.001) and haloperidol ( 2 = 8.70; P = 0.012), had a statistically significant posi- tive association with TD. No association was established between trifluoperazine and TD. We considered very carefully the confound- ing effects that a retrospective, multi-centre study such as this may have had on our findings and conclusions. We were also cognizant of the drawbacks of this study which included: • the lack of any inter-rater reliability or validity tests for the diagnostic skills ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما of the psychiatrists who made the diag- noses; • the diagnostic criteria which were used by the different psychiatrists at the time of diagnosis and the value of their di- agnostic significance in reaching that diagnoses; • what exactly was meant by TD by each of the psychiatrists and how accurate their descriptions and documentations were. Some of our findings were also related to a relatively small sample size which may, therefore, have made any generalizations or conclusions more difficult to draw up. The total number included for the final analysis in this study was only 151 patients. Despite this relatively small sample, how- ever, we believe some of our findings merit careful consideration. The lower overall prevalence rate of TD among our patients (5.9%) is not in keeping with most of the reported rates from studies done in Western Europe and North America, where reported rates are 20%–39%, and in the Far East, where reported rates are 29.0%–40.6% [6,7,10,18,20,22]. Perhaps more important is the inter- ethnic difference in the prevalence rate of TD between Arabs from the Middle East and the Afro-Arabs, who are of African origin. While TD rate was 23.5% among ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما the Arabs, it was 45.5% among Afro-Arabs. Although the size of the sample was rather small, this finding, coupled with the rela- tively high rate of TD (39.7%) reported by Van Harten et al. among the mostly Negroid population in a state public hospital in the Netherlands indicate that some interethnic difference may indeed exist [19]. However, the existence of some variability in rates of TD due to ethnocultural factors has not been consistent or agreed upon by all research- ers. While some researchers believe that some biological or genetic difference in susceptibility to TD may be found [16,23], others believe that susceptibility is most likely related to psychopharmacological factors such as duration of exposure and level of daily intake of neuroleptic drugs rather than to any ethnocultural or biologi- cal factors [8,10,11,22]. Our findings are more in support of differences related to either ethnocultural or genetic factors. If an interethnic difference in susceptibility between Arabs and Afro-Arabs does indeed exist, it will have important financial impli- cations, particularly as almost all the Afro- Arabs included in this study came from poor African countries. Such a background may emphasize the importance of either early adverse neurodevelopmental factors such as poor prenatal care, infections or birth com- plications, or a genetic susceptibility which may be operating in these individuals. We also found that many patients had been prescribed relatively low daily doses of neuroleptics to treat their psychosis. For example, among patients on haloperidol, 65 of them (77.4%) were taking < 20 mg/day with only 8 of those (12.3%) developing TD compared to 8 patients out of 19 pa- tients (42.1%) who were on 20 mg/day. The same was true among patients taking chlorpromazine: 106 of 117 (90.1%) were taking < 400 mg/day and only 15 of them (14.2%) developed TD compared to 5 out of 11 (45.5%) receiving 400 mg/day. This finding, coupled with the consistently reported strong positive association of TD with higher daily doses of neuroleptics [5,7,9,12] suggests that the lower rates of both TD and Parkinson disease among our patients may be mostly related to the low daily doses of neuroleptics needed to effectively treat their condition. This could have important economic implications for the continuation of the conventional neu- roleptics in poorer countries as first choice drugs, especially in countries where the af- fordability of drugs is the main determining factor in treatment. One of the risk factors for increased susceptibility to neuroleptic-induced TD is the indiscriminate prescription of anti- cholinergic agents [4,7,10,12,18,24]. In our study, only 51 of the 783 patients taking 1 anticholinergic drugs developed TD and 59 Parkinson disease, which is not in keeping with what has previously been reported in the literature. We therefore believe that this trend in developing countries for prescrib- ing an anticholinergic drug whenever a neu- roleptic drug is given to a psychotic patient, as has been noted among our patients, may prove to have some protective or preventive role in keeping the rate of TD relatively low. This is in contrast to previous findings where anticholinergics have been identified as one of the risk factors. A longitudinal prospective study would be needed to deter- mine if this is so. In conclusion, in this study we found the overall prevalence rate of TD among Arab patients to be lower than the general rates reported in the literature, as well as the existence of some interethnic difference in rates between Arabs and Afro-Arabs. It is also possible that Arabs who suffer from chronic psychosis might need relatively lower daily doses of neuroleptics to be ef- fective for their illness. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Prevalence of TD and drug-induced Parkinson disease were both relatively low among our patients, and in general, rela- tively low daily doses of conventional neu- roleptics were prescribed. Consequently, we believe that the conventional neurolep- tics could still continue to be first choice antipsychotics for the treatment of chronic psychosis rather than the relatively more expensive new atypical drugs which poorer countries may find it difficult to afford. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Mental health in the Eastern Mediterranean Region. Reaching the un- reached Mental health remains a neglected area of public health. People who suf- fer from mental ill health are among the most vulnerable in society, often from the poorest segments in society. They are the “unreached” Mental health in the Eastern Mediterranean Region. Reaching the un- reached charts the progress made in the provision of mental health care in the countries of the Eastern Mediterranean Region of the World Health 3 1 Organization. It is organized into sections. Part covers the philosophy 2 and components of mental health programmes. Part describes the experiences of the countries of the Region; each country has a section 3 on general health and mental health. Part discusses the key issues in provision of mental health care today and tries to identify the areas for future work, at the regional level. Annexes support the other sections and the work as a whole. The full text of this publication is available free on line at: http://www.emro. 702who.int/dsaf/dsa .pdf ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Nutritional deficiencies are a major health problem in tropical and subtropical regions of the world [1]. Malnutrition is still a problem especially in developing countries, where the total number of underweight and stunted children has not changed dramati- cally since 1980 [2]. In Egypt, the incidence of protein–energy malnutrition (PEM) was found to be 16.5% [3]. Reduction in the supply of energy and several essential nutrients during the early stages of life has profound effects on the nervous system structural and functional development. Malnutrition impairs brain development, decreasing the number of cell replication cycles, reducing total brain DNA and restricting dendritic arborization, thus reducing the connection between neurons [4]. Cintra and colleagues [5] stated that prenatal and chronic malnutrition produces important alterations in the homeostatic and circadian process of sleep, thus altering the sleep–wake cycle. Several neuropeptides affect the sleep–wake cycle [6] and a role for gamma- aminobutyric acid (GABA) transmission has been hypothesized [7]. Lechin et al. investigated the changes of circulating neu- rotransmitters (noradrenaline, adrenaline, dopamine, platelet serotonin, plasma serot- onin and tryptophan) during the sleep–wake cycle to correlate the profile of circulating neurotransmitters with the well-known cen- tral neurocircuitry functioning during the sleep–wake cycle [8]. Among many factors associated with PEM, increased levels of the neurotransmitter serotonin were re- ported to be the cause of sleep disturbances in malnourished rats [9]. Since PEM is still a problem in many countries, its effects on children’s develop- ment and cognition is a concern of many investigators. As far as we know, sleep disturbances in PEM patients have rarely been studied before as most of the data on malnutrition from sleep laboratories is from animal models [5,9,10,11] or anorec- tic adults [12,13]. This study was there- fore designed to detect disturbances in the sleep–wake cycle of Egyptian infants with PEM, comparing oedematous and non-oede- matous cases before and after a nutritional rehabilitation programme, and to correlate the detected abnormalities to the level of the neurotransmitter serotonin. The present study was conducted on 26 infants diagnosed with PEM according to the Wellcome classification [14]. They were recruited from the nutritional unit of the Children’s Hospital and assessed at the Institute of Psychiatry, Ain Shams Uni- versity. The patients were further divided into 2 groups according to the 2 clinically distinct disorders of PEM, oedematous and non-oedematous, Heird’s preferred terms for kwashiorkor and marasmus respectively [3]. Group 1 was 12 infants with the non- oedematous form of PEM (8 males and 4 females), with a mean age of 8.5 months [standard deviation (SD) 3.8 months]. Group 2 comprised 14 infants (8 males and 6 females) with the oedematous form of PEM, with a mean age of 13.7 (SD 7.5) months. Patients of both groups were com- pared with group 3, 10 clinically healthy age- and sex-matched controls (5 males and 5 females), with a mean age of 12.6 (SD 6.8) months. The controls were recruited from among those presenting for dietetic advice, vaccination or circumcision (in males) at the outpatient clinic in the Chil- dren’s Hospital. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما All the recruited patients and controls were from underprivileged families (having low socioeconomic standard according to Park and Park [15]) and had been breastfed for at least 6 months and afterwards weaned while receiving artificial milk products. None of the infants was on any medication or vitamin supplementation at the time of enrolment in the study. An informed written consent was signed by the parents or the legal guardians af- ter obtaining the approval of the ethical committee at the Children’s Hospital, Ain Shams University. Each studied infant was then enrolled in the 3-phase study: pre- interventional assessment (phase 1); a nutri- tional rehabilitation intervention programme according to World Health Organization (WHO) guidelines [16] (phase 2); and a post-interventional assessment 2 months (SD 2 weeks) after enrolment (phase 3). The WHO nutritional rehabilitation pro- gramme starts feeding with a diet which is low in protein, fat and sodium and high in carbohydrates (calorie intake 80–100 kcal/kg/day) as almost all severely mal- nourished infants have infections, impaired liver and intestinal functions and problems related to electrolyte imbalance. At the rehabilitation stage as the infant’s appe- tite returns, the calorie intake is increased to 150–200 kcal/kg/day with an increase in amounts and decrease in frequency. A high protein diet is given and vitamins and minerals (potassium, magnesium and zinc) are continued in increased amounts. Iron is given during this stage to treat the anaemia present. The infant remains in the hospital during the early phase of rehabilitation (at least 3 weeks after admission), and is then followed up in the nutritional rehabilitation outpatient clinic. Assessment in phases 1 and 2 included detailed dietetic history and clinical ex- amination, with special emphasis on the anthropometric measurements and signs of malnutrition as well as the laboratory workup and recording the sleep data. For the laboratory workup, samples of blood were collected from all subjects and proc- essed as clotted venous blood and EDTA anticoagulated blood. Serum samples were used for the determination of liver and kidney functions (Synchron CX-5 Delta, Beckman Inst. Inc., Scientific Instruments Division, Fullerton, USA). Serum level of serotonin was estimated as well in all stud- ied cases by enzyme-linked immunosorbent assay (ELISA) according to Chauveau et al. [17]. The EDTA blood was used for com- plete blood count (Coulter T660, Miami, USA). Assessment of sleep was done through a standard sleep questionnaire in simple Ara- bic language that was answered by the par- ents [18] in addition to polysomnographic evaluation of the study infants and con- trols who were left to sleep spontaneously without any intervention. The apparatus used was the Neurofax EEG-2110 digital electroencephalograph (Nihon Koden Cor- poration, Tokyo, Japan), with 3 dedicated respiratory inputs at the headbox and 4 available DC channels. The EEG 2110 can record a variety of biopotential activities necessary for the sleep laboratory. Poly- smith TM software provided the immediate staging, scoring and reporting of the polys- omnographic recordings. Hypnographic sleep EEG provided the following variables: sleep continuity (in- cluding sleep latency and efficiency and ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما number of arousals) as well as sleep ar- chitecture, including percentage of rapid eye movement (REM) sleep, non-REM (NREM) sleep and slow wave sleep (SWS), as well as REM, NREM and SWS latency periods. The analysis of sleep architecture did not include a differential percentage of each sleep stage (I, II, III, IV), but rather NREM or SWS was considered as a whole. Statistical analysis of the results was done using SPSS, version 10 and Statsoft, ver- sion 5. Non-parametric data were detected by the Shapiro–Wilk test. Student t- and paired t- tests were used for parametric quantitative data and Mann–Whitney U and Wilcoxon matched pairs tests for non- parametric quantitative data in addition to the correlation studies. The numerical data were represented in mean and standard deviation (SD) and median (interquartile range). The differences were considered significant at P < 0.05. The present study revealed significantly lower anthropometric measurements— weight, length and mid arm circumfer- ence—in PEM patients compared with those of the controls at the pre-assessment phase (Table 1). These measurements showed significant improvements after nutritional rehabilitation in both oedematous and non- oedematous patients, although the values did not reach control values (Table 2). The same findings were observed regarding haemoglobin and serum albumin levels (Tables 1 and 2). As regards liver and kid- ney functions, their values were within the normal range for age and sex according to Nicholson and Pesce [19] from the start of the study. Serum serotonin levels were signifi- cantly higher in both studied PEM sub- groups compared with those of the controls (Table 1) and these decreased significantly after nutritional rehabilitation (Table 2). The study also revealed significantly lower skull circumference (% of median for age) and serum albumin with significantly higher serum serotonin in the oedematous compared with the non-oedematous patients (Table 1). Analysis of the sleep questionnaire findings revealed that caregivers reported sleep com- plaints in 76.9% of cases, mostly disturbed sleep rhythm, in both subgroups of PEM infants. As regards the sleep data before nutri- tional rehabilitation, there was no signifi- cant difference between both PEM groups before nutritional rehabilitation (Table 3). Comparing both groups of PEM infants with the control infants showed a significantly lower percentage of NREM sleep, REM la- tency time and 2nd REM time. In addition, the number of minutes of sleep latency was lower in both PEM groups compared with the controls but this result was significant only in non-oedematous patients. On the other hand, the percentage of REM sleep was higher in both PEM groups compared to the controls but this result was significant only in oedematous patients (Table 3). In addition, initial SWS latency time and the percentage of SWS were not significantly different in both subgroups compared with the controls. Table 4 shows the sleep parameters meas- ured after nutritional rehabilitation. Sleep latency increased in both groups and this was significant in non-oedematous infants. The percentage of NREM sleep and the amount of 2nd REM sleep time increased in both groups, with a statistically significant ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما difference in the oedematous group only. The percentage of REM sleep decreased significantly in the oedematous group while the non-oedematous one showed a non- significant increase. Moreover, SWS latency time decreased in both subgroups and this was significant only in oedematous patients. As regards percentage of SWS, it decreased significantly in the non-oedematous group, while the oedematous one showed a non- significant increase. However, in spite of the changes occurring after nutritional reha- bilitation, most of the sleep EEG parameters of both groups of PEM infants did not reach the control values. Figure 1 shows the polysomnography of a PEM patient, showing no clear stage dif- ferentiation of NREM sleep, while Figure 2 shows the polysomnography of the same patient after nutritional management with greater differentiation of sleep stages. As regards the correlation studies, the present study revealed significant positive correlation between the rate of change of REM sleep in non-oedematous PEM infants and both of weight and serum albumin (r = ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 0.88 and 0.80 respectively and P < 0.05 for both). In addition the rate of change of REM sleep showed a negative correlation to serum albumin level in oedematous PEM patients but this was not statistically significant (r = –0.45). There was also a significant negative correlation between the rate of change of SWS and weight in non-oedematous PEM patients (r = – 0.72 and P < 0.05). The sleep data of the PEM infants in our study showed that the sleep latency was significantly lower in the non-oedematous malnourished group before nutritional re- habilitation compared with the controls. However, in the oedematous malnourished group sleep latency was lower than in the controls, but not significantly so. After nu- tritional rehabilitation the sleep latency in- creased in both subgroups of PEM infants, both non-oedematous and oedematous. This indicates that after nutritional rehabilitation, patients of both PEM types needed more time to fall asleep than in the acute phase of PEM illness which denotes an improvement in their condition, i.e. they are less sleepy or lethargic. The percentage of NREM sleep was significantly lower in both non-oedematous and oedematous malnourished cases before nutritional rehabilitation compared with the controls. After nutritional rehabilita- tion, NREM sleep improved in the non- oedematous malnourished group, but was ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما still significantly lower than the control values. Moreover, the oedematous mal- nourished group showed improvement after nutritional rehabilitation but the percentage of NREM sleep was still lower than the control values, although not significantly. The failure of complete recovery may be attributed to the short period of nutritional rehabilitation and the immature brain devel- opment at this age. Before nutritional rehabilitation the per- centage of SWS was higher in the non-oede- matous subgroup of PEM compared to the control values but this was not of statistical significance. This, in addition to the sig- nificantly higher serotonin levels detected in the PEM patients, agrees with Datta et al. [9]. Their animal study found that the mean percentage of SWS in malnourished rats (73.30%) was significantly higher than in well-nourished ones (61.23%) and they attributed this to increased brain serotonin levels. The same results were stated by Mokler et al. [20] who found that malnutri- tion in rats leads to increased serotonin production in the brain which leads to an increase in the percentage of SWS. After nutritional rehabilitation, the per- centage of SWS in the non-oedematous subgroup of PEM decreased and nearly reached the control levels. This is further supported by the significant negative cor- relation between the rate of change of SWS and weight in the non-oedematous PEM infants. Similarly Cintra et al. demonstrated that malnourished rats showed a significant reduction in SWS in the nutritional recovery period pre- and postnatally [5]. Before nutritional rehabilitation the percentage of REM sleep in the non- oedematous malnourished infants was higher than in the controls, but with no sta- tistical significance. However, in the oede- matous malnourished group the percentage of REM sleep was significantly higher than the controls. This may be attributed to im- mature brain development. After nutritional rehabilitation the percentage of REM sleep in the non-oedematous malnourished group increased but not significantly. This was further demonstrated by the significant positive correlation between the rate of change of REM sleep and that of weight and albumin in the non-oedematous group of PEM patients. On the other hand, the percentage of REM sleep decreased sig- nificantly in the oedematous malnourished subgroup, which is the normal expected pattern. This finding is further supported by the negative correlation between the rate of change of REM and that of albumin levels in these patients. Our results of percentage REM sleep in non-oedematous patients are not consist- ent with Siegal who suggested that food consumption could produce an increase in REM sleep [21]. More recently, Cintra et al. found that there was increase in REM sleep during nutritional rehabilitation in rats with PEM [5]. The normalization of REM sleep to near control levels in both subgroups is important, as Shaffery et al. proposed that the primary purpose of the REM phase is to act as an inducer of CNS development in the fetus as well as the neonate [22] and we suggest that it might still be of the same importance during infancy. We hypothesize that the increase in REM sleep in non-oedematous PEM might be related to a transient relative increase in REM following a period of decrease (the so called “REM rebound”), which is supposed to be followed by the stabilization period. It is worth mentioning here is that in spite of the changes occurring after nu- tritional rehabilitation, most of the sleep EEG parameters of both groups of PEM infants did not reach the control values. This could be explained by the work of Robinson ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما et al., who detected electrophysiological abnormalities, persisting despite somatic rehabilitation, of 10 severely malnourished children, and added that this must be associ- ated with the chronic rather than the acute aspects of malnutrition, and can be used to detect any deviation of brain function from normality [23]. In conclusion, the ability to maintain nor- mal progression in sleep–wake maturation is an important index of brain development and may serve to assess how environmental factors, including essential nutrient supply, affect central nervous system development. The present study showed that PEM can have an effect on the sleep–wake cycle, which improves after adequate nutritional rehabilitation. The disturbed serotonin lev- els in PEM could be one of the factors responsible for such changes and this needs further study, together with assessment of the role of other neurotransmitters involved in the central nervous system functional development in PEM. The changes that we have demonstrated in the sleep pattern in PEM infants should be considered seriously, as they could be detrimental to the development of social, behaviour and cognitive functions. We thus recommend proper and early nutritional rehabilitation for PEM infants not only to improve the physical growth parameters but also to improve their sleep pattern. Fol- low-up of these patients for longer periods is also recommended to ensure that the re- sidual sleep changes are reversible and that there are no permanent changes. Further studies on PEM patients are advised on a larger scale to support the current results and to measure the specific dietetic ele- ments which could be the causal factors for the sleep disturbances. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 3% 0.6%PER P<0.05 NPUP<0.05 P<0.05 15–20% ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Peas have been a staple diet of man and livestock since prehistoric times. In certain regions, people have always relied on peas and other pulses to provide protein to com- plement the cereals in their diet [1]. They are now grown in temperate regions all over the world, including higher elevations of the tropics. Cultivation is favoured in dry areas where the weather is cool and moisture is abundant during early growth, but where rainfall is minimal during the later stages of development. Peas are an important part of the crop rotation in many countries of Central America and Europe as well as India, Mayanmar and Pakistan. Green peas have become an important green vegetable in many developed countries. They offer a bulk source of seed protein for man and ani- mals from a relatively short growing season compared with other legumes [2]. Pakistan produces about 16 000 tonnes of peas annu- ally and the area under cultivation is about 3.2 thousand hectares [3]. This study is a continuation of earlier in- vestigations through which we established the effect of cooking and supplementation with different kinds of meat on the nutri- tional improvement of mash (Vigna mung), mung (Vigna radiate), masoor (Lens escu- lenta), lobia (Phaseolus vulgaris) and gram (Cicer arietinum) [4–8]. To assess the extent of improvement in the nutritional quality of peas (Pisum sativum L.) by cooking and supplementation with different kinds of meats, biological trials were conducted on albino rats. Peas (Pisum sativum L.) were procured from the local market and dried in a hot air oven at 105 °C for around 4 hours. Flour was obtained by grinding and sieving through a 20 mm mesh sieve. The flour was stored in airtight jars at room temperature until use. Similarly, flour was obtained and stored after cooking the peas by a conven- tional method as described by Bhatty et al. [4]. Briefly, the peas were put in a pot, covered with fresh water to 2.5 cm above the surface and boiled (100 °C) on a natural gas cooker for 40 minutes at high heat, then simmered for 30 minutes. At this stage peas became tender. Maize starch, corn oil and casein (Merck DGaA, Darmstadt) used for the preparation of the standard diet (protein content 84%) were also purchased from the market. The experimental diets were prepared using raw and cooked peas. Diets were also prepared by replacing 10%, 15% and 20% of the protein of cooked peas with the same amount of protein derived from lean meat: poultry (chicken), mutton or beef. The com- position of the diets is shown in Table 1. The casein diet served as a standard and a nitrogen-free diet was used to determine the endogenous nitrogen. The mineral mixture used in the preparation of the experimental diets was prepared according to the formula of Oser [9] and the vitamin mixture accord- ing to Miller and Bender [10]. Biological evaluation was done by measur- ing the protein quality of diets containing peas in raw and cooked form with and with- out supplementation with meat. Albino rats of the Sprague–Dawley strain were used. During gestation and nursing, the mothers were fed a balanced stock diet. Litters born to different mothers within 24 hours were taken to be of same age. Weaning was done at 21 days of age. The rats were then put on stock diet for 7 days prior to the experiment. They were arbitrarily divided into experimental units ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما of 2 rats each in such a way that the initial weight of the rats in each cage was 90 g; 3 experimental units were randomly allotted to each diet. The rats were fed the allotted diet ad libitum for a period of 10 days. Dur- ing this period fresh, clean water was made available at all times and room temperature was maintained at 24–27 °C. The weight of each replicate was recorded daily. The faecal matter from each cage was collected daily, dried to a constant weight and stored in glass bottles for nitrogen determination. At the end of 10 days trial, all the rats were killed with an overdose of chloroform and their cranial and abdominal cavities were opened. The carcasses of each group, in- clusive of intestinal contents, were weighed before and after drying at 105 °C to constant weight. The dried carcasses were run twice through a domestic mincer and stored in airtight bottles for estimation of body nitro- gen. The nitrogen content of the diet, faeces and carcasses of each group was determined by Khjeldahl’s method [11]. Data obtained was used to determine the protein efficiency ratio (PER), true digestibility (TD) and net protein utilization (NPU) [10]. Samples of peas were analysed for proximate composition (moisture, crude protein, ether extract, crude fibre, total ash and nitrogen free extract, according to standard methods [11]. Samples of the ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 3 kinds of meat were analysed for protein content only. The amino acid analysis of peas was car- ried out by the method of Spackman, Steir and Morre using a Beckman Model 120C amino acid analyser (Beckman, Fullerton, California) [12]. The data obtained for PER, TD and NPU were used for analysis of variance using a completely randomized design. The analy- sis was computed using SPSS-400. Multiple comparisons of means were made using Fisher protected least significant difference (PLSD) test [13]. The change in the proximate principals of raw and cooked peas is given in Table 2. Cooking resulted in a slight reduction in crude protein, crude fibre and ash. The lysine content of raw peas was 3.0%, which decreased to 0.6% after cooking. Comparison of experimental diets con- taining peas only with standard casein diet (Table 3) indicated that the PER of cooked peas was very close to that of the standard casein diet whereas the diet containing raw peas had a PER value almost half of the cooked pea diet. Other biological param- eters, TD and NPU, also showed improve- ment when peas were cooked. Biological evaluation of experimental diets is given in Table 4. On average, in- clusion of 15% protein from poultry meat yielded comparatively better results in terms of PER, TD and NPU as it is cheaper than the other 2 meats. Proximate composition of raw peas in our study was 8.0% moisture and 23.7% pro- tein. Augustin and Klein [14] reported simi- lar amounts of moisture and protein in raw peas. Ali-Khan and Youngs [15] showed the protein content to be 22%–23% in field peas. The variation in crude protein content is a reflection of varietal differences and may be attributed to genetic and environ- mental factors. In this study, ether extract was 2.3% in raw peas. Augustin and Klein reported a lower value [14] and other reports of fat content range from 1.0% to 3.1% [2]. These variations could be due to variety differences. Raw peas had 6.8% crude fibre. Augustin and Klein [14] reported a much higher fibre content. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Nitrogen free extract in raw peas was 62.4% [14]. Savage and Deo reported nitro- gen free extract in the range 60.0%–71.7% [2]. Pea seeds had 4.8% ash; other research- ers have reported ash contents of 2.4%–4.1% [16] and 1.0%–3.4% [2]. Raw peas contained 3.0% lysine, provid- ing well above the recommended require- ment (12 mg/kg body weight per day) [17], making peas an ideal supplement to a cereal based diet. Legumes are considered a good source of lysine and as such provide this essential amino acid to enhance the nutritive value of the protein in mixed diets [18]. Savage and Deo reported lysine content at 6.22%–12.3% in peas [2]. El-Refai, Gouda and Ammar showed that in general the amino acid content changed only slightly during storage except for small decreases in lysine, cystine, methionine and tryptophan [19]. Sarwar, Sosulski and Bell concluded that field peas were superior to soybean when blended with wheat flour or supple- mented with additional amino acids [20]. A slight lowering was observed in proxi- mate crude protein content of peas after cooking. Other studies have found similar changes [19,21]. James and Hove reported that improve- ment in nutritive value on cooking was a result of the destruction of anti-nutri- tive factors [22]. Manan et al. observed that cooking peas resulted in considerable reduction in the phytic acid content of Pa- kistani varieties, without any loss of total phosphorus [23]. The nutritive value of peas considerably improved on cooking, suggesting that other water soluble and or heat labile anti-nutritive factors might be more important than phytic acid in affecting the overall nutritive quality of seeds. It was observed that cooking affected the amino acid profile. All amino acids showed losses during cooking of peas. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Protein efciency ratio (PER) Raw peas had PER 1.4, which increased significantly on cooking to 2.0. James and Hove showed a similar increase, 1.87 to 2.21 [22]. The improvement in nutritive val- ue on cooking could be due to destruction of anti-nutritive factors. Shah also reported a significant increase in the body weight gain of rats due to cooking of the whole seed [24]. Supplementation of a diet based on cooked peas with different types of meat also showed significant improvement over a diet containing raw peas, irrespective of the kind of meat. However, supplementation with different types of meat did not improve the PER significantly over that of the diet containing only cooked peas. True digestibility (TD) The TD of protein of peas increased sig- nificantly on cooking from 74.7% to 79.8%. It has been reported that protein TD of autoclaved peas increased from 85% to 88% [20]. Goodlad and Mather, however, claimed that there were only minor effects of cooking on the digestibility of non-starch polysaccharides and their constituent sugars [25]. Fleming and Vose showed that the in vivo digestibility of raw and cooked starch from peas was high in rat experiments [26]. The increase in digestibility on cooking may be due to the elimination of trypsin and chymotrypsin inhibitors. In our study, TD also increased sig- nificantly when the pea-based diet was supplemented with meat, and increased with increasing level of supplementation. The TD of the diet containing peas supple- mented with 20% mutton was significantly higher than the digestibility of other diets. Net protein utilization (NPU) The NPU of the diet containing raw peas was 41.6% and on cooking it significantly increased to 46.3%. Shah showed NPU values of 42.4%–46.8% in raw peas and 49.0%–52.0% in cooked peas [24]. The NPU values were significantly higher when were supplemented with 20% poultry meat. The NPU also increased with the increase in the levels of supplemental mutton and beef beyond the 10% level. Bell and Youngs reported that biological value of pea protein concentrate alone was low but was consid- erably improved by the addition of methio- nine [27]. Shah reported a non-significant increase in biological value on cooking; reduction in biological value was suggested as being due to the destruction or leaching of essential amino acids during the cooking process [24]. Overall, supplementation of peas with 15% poultry meat optimally enhances the protein quality. Cooking alone can also be used for the improvement of protein quality of peas. Although this study was conducted on laboratory rats, it provides a rationale for the supplementation of peas with small quantities of poultry meat for the nutritional rehabilitation of poorly-fed communities. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Technical consultation to review the regional food-based dietary guidelines The World Health Organization Regional Office for the Eastern Mediter- ranean organized the above-mentioned technical consultation in Cairo, 2 4 2007Egypt, from to April . The objectives of the consultation were: to review the draft regional food-based dietary guidelines under preparation; to incorporate additional relevant food and dietary information from Member States; and to finalize the content and format of the regional food-based dietary guidelines. Experts from Egypt, Islamic Republic of Iran, Lebanon, Pakistan, Qatar, Saudi Arabia, Syrian Arab Republic, Tunisia, United Arab Emirates, United Kingdom, as well as WHO concerned staff, participated in this consultation. • • • ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما End-stage renal disease (ESRD) is a growing problem worldwide and renal replacement therapy is increasingly exerting pressure on health systems [1]. The situation is particu- larly serious in developing countries where health resources are limited. The adjusted incidence of renal replace- ment therapy in 9 European countries has increased from 79.4 per million population (pmp) in 1990–91 to 117.1 pmp in 1998–99, i.e. an average annual increase of 4.8% [1]. This increase occurred largely among older age groups. The incidence of ESRD due to diabetes, hypertension and renal vascular disease almost doubled over the same pe- riod. Similar trends were observed in Japan [2], Australia and New Zealand [3] and the United States of America (USA) [4]. The distribution of patients on haemodi- alysis in the world is strongly influenced by economic factors [5]. In Jordan, the number of patients on haemodialysis has doubled over the past 5 years presenting a real chal- lenge for the scarce health resources. The objective of the study was to report on the epidemiology of haemodialysis in Jordan and to provide current estimates of the associated costs. We conducted a survey in all haemodialysis units in Jordan (56 units) between 1 Septem- ber and 31 October 2003. Each patient was personally interviewed in the haemodialysis unit by the attending nurse, using a struc- tured questionnaire specifically designed for the purpose of this study. Additional data such as hepatitis B and C positivity were collected from the patient’s medical record. The cause of ESRD was provided by the attending physician. We also col- lected relevant data on each haemodialysis unit. This included: number of patients on haemodialysis in the unit at the time of the survey (July 2003), number of patients who died in the unit in 2002, and the number of patients who started haemodialysis in the unit in 2002 and their status in July 2003. Epi-Info, version 6 software was used for data entry and analysis. The distribution of the patients by relevant sociodemographic and clinical characteristics was obtained. The prevalence of haemodialysis by certain variables was obtained using population estimates of Jordan from the Department of Statistics as the denominator for these rates. Incidence of haemodialysis in the year 2002 was obtained by dividing the total number of patients who started haemodialysis in all units in 2002 (irrespective of whether they were still living at the time of the survey) by the population of Jordan in 2002. The case fatality rate in the first year after initiating haemodialysis was obtained by dividing the number of deaths that occurred among all patients who started haemodialysis in 2002 followed through June 30, 2003 by the total starting haemodialysis. The cost of haemodialysis in Jordan for the year 2003 was calculated by adding the cost of the haemodialysis session (US$ 105 in the pri- vate sector and US$ 85 in the public sector), the cost of medications and investigations, the cost of admissions, and the cost of the arterial access (arterio-venous fistula in over 95% of patients). A total of 1711 patients were on haemodi- alysis at the time of the survey in Jordan. Table 1 shows the sociodemographic char- acteristics of the patients on haemodialysis. The age of patients ranged from 5 years to ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 88 years with a mean age (standard devia- tion) of 48.9 (16.7) years. About one-third of the patients were 60 years of age or more. Of the 1711 patients, 957 (55.9%) were male. About 28% of the patients were illiterate and 17% current smokers. Only 13.2% were employed and the vast major- ity was poor (92%) (family income < US$ 420/month). Although the average distance to the haemo- dialysis unit was 13.6 km, a good propor- tion of patients (11.5%) had to travel 30 km or more. The major service provider was the private sector (43.4%) followed by the Ministry of Health (27.4%). The duration on haemodialysis ranged from 1 month to 27 years resulting in a skewed distribution (mean = 42.7 months, median = 28 months). Over 54% of patients were maintained on 3 haemodialysis sessions per week, 45% on 2 sessions per week, and only 9 patients on 1 session per week (Table 2). The proportion of patients on 3 sessions per week was much higher in the private sector (89%) than in ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما the Ministry of Health (32.7%) or the Royal Medical Services (20%). These proportions were inversely correlated with the ratio of patients per haemodialysis machine in the different sectors: 3.1, 4.4 and 5.7 respec- tively. Because of the considerable impact on the quality of life of patients on dialysis, we asked about patients’ sexual activity. For cultural reasons, the frequency of sexual intercourse was assessed only in married males. Out of the 744 patients, 689 (93%) responded to the question. Table 3 demon- strates the frequency of intercourse by age. Overall about 48% were not sexually active at all. As expected, lack of sexual activity increased markedly with age from 6.4% in patients < 40 years to 36.4% in patients 40– 59 years to reach 79.2% at age > 60 years. The hepatitis B and C status of the patients before dialysis was unknown in 12% and 24% of the patients respectively. Serocon- version rates were calculated only for pa- tients with known hepatitis B and C status prior to haemodialysis. The overall positivity rate for hepatitis B virus (HBV) among our patients was 4% (Table 4). Of the 68 positive patients, 17 (25%) seroconverted after starting haemo- dialysis. The percentage of HBV-negative patients who seroconverted after initiation of haemodialysis was 1.2%. Given an aver- age duration of haemodialysis of 3.5 years, the annual seroconversion rate was 0.34% With respect to hepatitis C virus (HCV), the overall positivity rate was 21%. About one-third (32.2%) of HCV-positive patients were negative before starting haemodi- alysis. The percentage of HCV-negative patients who became positive after initia- tion of haemodialysis was 9.2% yielding an annual seroconversion rate of 2.6%. The HCV status before starting haemodialysis was unknown for 24% of this population. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The overall prevalence of haemodialysis in Jordan was 312 pmp. Table 5 shows the prevalence by age and sex. The prevalence increased from 30 pmp below age 20 years to 1822 pmp among the elderly ( 60 years). The prevalence was also higher among males (329 pmp) as compared to females (288 pmp). The total number of patients who started haemodialysis in 2002 in all units, irrespec- tive of whether they were still on haemo- dialysis at the time of the survey, was 578 accounting for an incidence of 111 pmp in 2002 (the total population of Jordan in 2002 was 5.2 million). The total number of deaths among patients on haemodialysis in Jordan during the 1- year period preceding the survey was 292. Of the 578 who started haemodialysis in 2002, 116 had died by 1 July 2003 giving a case fatality of 20% during the first year. Kidney transplantation was carried out for only 5% of those patients. Table 6 presents the causes of ESRD among patients on haemodialysis at the time of the survey. The commonest cause was diabetes mellitus (29.2%) followed by hypertension (18.4%) and glomerulonephritis (12.3%). The category of other causes in Table 6 included, for example, systemic lupus ery- thematosus, analgesic nephropathy, Alport syndrome, and familial Mediterranean fe- ver. For over one-fifth of the patients, the cause was classified as unknown as the attending physician was unable to provide the cause which led to ESRD after full in- vestigation, including renal biopsy. In fact, these cases may be considered as having primary kidney disease although they are categorized separately in Table 6. For patients who started haemodialysis during 2002 and 2003, diabetes mellitus was the cause in 33.4% and 44.0% respec- tively. The estimated total annual cost of haemodi- alysis in Jordan was US$ 29 715 553 (Table 7). Haemodialysis sessions accounted for about three-quarters of the total cost while medications and investigations accounted for an additional 20%. Erythropoietin was ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما regularly supplied to 92% of patients on haemodialysis, calcium to 98%, vitamin D to 91% and iron to 88%. Erythropoietin was given twice weekly on average with the objective of raising the haemoglobin level to 10–12 g/dL. Data on haemodialysis in Jordan are very scarce. The present study has shown an incidence of haemodialysis in Jordan of 111 pmp in 2002, a figure slightly lower than reported from Europe in 1998–1999 (117 pmp) [1] but much higher than the incidence in Baltic countries [6]. In the United States, the incidence of ESRD in the year 2000 was 291 pmp [7]. In Egypt the reported incidence of ESRD in the year 1992 was 200 pmp, of whom 80 pmp were accepted on the dialysis programme [8]. Variation in the incidence of haemodialy- sis largely reflects differentials in accept- ance of patients onto the haemodialysis service, as well as the reference year for the reported incidence, since an increasing trend has been observed almost everywhere. The prevalence of haemodialysis in Jordan seems to be increasing at a high pace from 114 pmp in 1992 [9] to 181 pmp in 1998 [10] reaching 312 pmp in the year 2003 (average annual increase of 8.5%). Such an increase may largely reflect the increasing availability of haemodialysis in Jordan. Consistent with data from other coun- tries [1,4,7], diabetes mellitus was the com- monest underlying cause for haemodialysis among our patients (29.2%) followed by hypertension (18.4%) and glomerulonephri- tis (12.3%). However, considering patients who started haemodialysis in 2003 alone, diabetes accounted for 44% of the cases. Several explanations can be offered, such as more elderly persons accepted on haemo- dialysis, shorter survival of diabetics on haemodialysis, or an actual increase in the contribution of diabetes to ESRD. Diabetes mellitus is a highly prevalent disease in Jordan affecting 13.4% of Jordanians 25 years of age [11]. The mean age of patients on haemodialysis in Jordan has increased from 44.6 years in 1998 [10] to 48.9 years in 2003 and the percentage of patients aged > 60 years has increased from 21% in 1998 [10] to 33% in 2003. In agreement with data from other coun- tries [1], more males (55.9%) than females were being maintained on haemodialysis and the prevalence of haemodialysis was 334 pmp in males compared to 288 pmp in females. HBV positivity was relatively low (4%), and annual seroconversion occurred in 0.34% of our patients after starting haemo- dialysis. HCV positivity among our patients was 21% with an annual seroconversion of 2.6%. These figures are assuring even when compared with data from a number of European countries, Japan and the United States [12]. The DOPPS study showed the prevalence and seroconversion of HBV among haemodialysis patients in 5 European countries, Japan and the United States [12]. The overall prevalence and seroconversion was 3.3% and 0.78 per 100 patient-years respectively. Recent data on HBV in the general population are lacking in Jordan but the prevalence is generally low in se- lected groups such as healthy blood donors (1.8%–3.9%) [13] and multi-transfused patients for hereditary haemolytic anaemia (3.5%) [14]. With respect to HCV, the prevalence among haemodialysis patients in Jordan was much lower than in many countries in our region such as Saudi Ara- bia (72.3%) [15], Kuwait (40%) [16] and Pakistan (68%) [17], but higher than that in developed countries such as Germany (7%) [18]. In Jordan, the prevalence of HCV in the general population is not known, but it ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما is available for subgroups of the population such as healthy blood donors (0.65–2.1%) [13] and multi-transfused patients (40.5%) [14]. A previous study in 6 Ministry of Health haemodialysis units in Jordan re- ported an HCV prevalence of 34.6% [19]. In the present study, HCV prevalence in Ministry of Health haemodialysis units was very similar (33%) compared to 7.3% in the private sector. The most likely explanation for the observed lower rate in the private sector is a reluctance to accept HCV posi- tive patients since neither the private nor the public sector have HCV infection control protocols which could account for such a discrepancy. In contrast, a standard HBV prevention protocol exists in all haemodi- alysis units in Jordan. It should be noted that strict infection control measures in haemodialysis units and isolation policy for HCV-positive patients have been shown to limit HCV transmission as it has done in reducing HBV transmission [12,20–22]. Patient-to-patient transmission of HCV also occurs in haemodialysis units highlighting the importance of the strict enforcement of standard infection prevention measures [23]. Nurse understaffing of haemodialysis units was shown to have a negative effect on HCV transmission [24] probably because of difficulty in ensuring adherence to infection control measures in such a situation. Among the interesting issues in our study was the assessment of sexuality. Total sexual inactivity affected the great majority of patients > 60 years of age and over one- third of those aged 40–59 years. Complete lack of sexual intercourse was a problem even at a younger age for over 6% of the pa- tients. These findings may add to our under- standing of the suffering of haemodialysis patients and their quality of life. Attending physicians should be aware of this problem and be ready to assist their patients in this regard. Sexual dysfunction of patients on haemodialysis has been reported by other studies [25,26]. The fatality rate observed in our study (20%) is similar to rates in the United States where the 1-year fatality rate was approxi- mately 22% in 2002 [27]. However, patients accepted for haemodialysis in Jordan tend to be younger than United States patients, and this may explain the favourable out- come in Jordan. Expansion of haemodialysis services in Jordan has a high cost. Jordan is a de- veloping country with an annual per capita income of approximately US$ 1800. The overall expenditure on haemodialysis in 2003 was about US$ 30 million, i.e. about 4% of the total health expenditure in Jordan. In Turkey, the annual cost of haemodialysis was US$ 22 759 per patient, while the cost of transplantation was US$ 23 393 and US$ 10 028 per patient respectively for the first and second year [28]. In industrialized countries, the annual costs of haemodialysis are much higher, being highest in France (US$ 78 947 per patient) [29] compared to Japan [30] and the United States [31] (US$ 46 000 per patient). Expenditure in Jordan is likely to in- crease rapidly in the near future as more and more of the elderly are accepted into the service. The government of Jordan bears almost all the costs of haemodialysis. Al- though the cost of haemodialysis in Jordan is relatively low (US$ 17 385/patient/year), the burden is huge in a country with limited resources like Jordan, and the efficiency of the allocated resources for haemodialysis should be an issue for consideration. The cost of kidney transplantation in Jordan is not accurately known but has been shown in other countries to be a less costly alternative [28,31]. Rough estimates of the cost of transplantation in Jordan, based on ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما personal communication, range from US$ 20 000–22 000 in the first year, dropping to US$ 10 000–12 000 in the second year, to reach US$ 8000–10 000 later on. Unfortu- nately, kidney transplant is used by only a small minority (5%) of patients with ESRD in Jordan. The reasons behind this are not clear but restrictions on transplants from non-relative donors exist by law to prevent the sale of human organs and may contrib- ute significantly to this underuse. While permitted in Jordan, cadaveric transplants are not widely available, probably because of a lack of efforts to raise the awareness of people of the importance of organ donation after death. Appropriate strategies to pro- mote kidney transplantation are needed and may succeed in containing the escalating costs. However, in the long run, prevention of ESRD is particularly important and can be achieved by early detection and prompt treatment of the major diseases leading to ESRD, namely diabetes, hypertension and glomerulonephritis. Restriction of haemo- dialysis to patients who are likely to benefit from the intervention and excluding patients with advanced multi-organ disease have been suggested [28] as a way to contain cost, but may be unacceptable on ethical grounds. In the present study, we did not collect data on peritoneal dialysis which is rarely used in Jordan. In fact, no programme for peritoneal dialysis exists in Jordan. There is no system to reimburse private physicians for caring for patients on peritoneal dialysis and therefore they direct their patients to haemodialysis. Moreover, as peritoneal dialysis is usually carried out at home, it is more likely to be successful in educated patients. In conclusion, our study has provided a profile of patients on haemodialysis in Jor- dan. Diabetes is the leading cause of haemo- dialysis and its role seems to be increasing. Haemodialysis is expanding rapidly and the burden is high given the limited resources in Jordan. HBV and HCV prevalence and seroconversion are relatively low suggest- ing that the current infection control meas- ures are working and, therefore, should be intensified. Promotion of less expensive interventions such as kidney transplantation and prevention of the main causes lead- ing to ESRD may be needed to curb the escalating costs. Research is needed on the quality of life of patients on haemodialysis and their families. The study was supported by a grant from the National Centre for Diabetes, Endocrinol- ogy, and Genetics in Jordan. We would like to thank Dr Hakim S. Alkadhi, Consultant Urologist, Dr Munib S. Ayoub, Consultant Internist, Dr Abdel-Ka- rim Khawaldeh, Consultant Endocrinolo- gist and Dr Yousof S. Hamzeh, Professor of Nephrology and Internal Medicine for their helpful contribution to this work. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Basic epidemiology The second edition of this popular textbook provides an introduction to the principles and methods of epidemiology. Since publication of the 1993first edition in , Basic epidemiology has become a standard refer- ence for education, training and research in the field of public health 25 and has been translated into more than languages. The second edition provides updated examples of how and why the basics of epidemiology are essential to anyone who is required to understand and apply the principles of disease causation and preven- tion. This edition includes comprehensive chapters on the nature and uses of epidemiology; the epidemiological approach to defining and measuring the occurrence of health-related states in populations; the strengths and limitations of epidemiological study designs, causation, the contribution of epidemiology to the prevention of disease, the promotion of health and the development of health policy, and the role of epidemiology in evaluating the effectiveness and efficiency of health care. Further information about this and other WHO publications is available at: http://www.who.int/publications/en/ ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Tuberculosis (TB) is the most common infectious disease worldwide, and causes the death of about 2–3 million people each year [1]. Incidence of TB has increased dramati- cally recently and a high prevalence of respiratory disease has been reported during recent years [2]. These reports most com- monly come from developing countries. Factors involved in the increased number of TB cases include poverty, immigration, poor health facilities and drug abuse [3]. In addition, irregular use of antituberculosis drugs, incorrect dosage, poor knowledge among general physicians, improper at- tention of doctors, and drug resistance of Mycobacterium tuberculosis can be fac- tors in treatment failure [4]. Studies done throughout the world show that by using the directly observed treatment, short-course (DOTS) strategy, the success rate of treat- ment is about 90%–95%, or even greater [5]. According to the same studies, even in industrialized countries and among knowl- edgeable and educated individuals at least 30% of patients, owing to a perception of improvement, do not take their medication properly and discontinue treatment after a while. In the DOTS strategy, in addition to direct observation of drug consumption, having a standard regimen for treatment and follow-up of patients improves the success rate. Conversely, for the reasons outlined above, treatment regimens other than DOTS have a low success rate, and may lead to TB transmission in the community and mycobacterial drug resistance [5]. Prevalence of TB in the Islamic Re- public of Iran is about 39 per 100 000 [6]. Prevalence in Golestan province, where this study was carried out, is quite high, the second highest in the country [7]. The aim of this study was to evalu- ate the efficacy of the DOTS strategy in reducing the failure rate of TB treatment in comparison with the conventional strategy. The clinical and epidemiological features of 260 smear-positive respiratory TB patients in Gorgan were studied for a period of 2 years. We carried out a follow-up cohort study to investigate the epidemiologic and clinical efficacy of DOTS strategy in treatment of TB patients in residential areas of Gorgan: a few areas classified as rural and one area classified as urban. All new patients (hav- ing had no prior treatment for TB) present- ing to the health services during the period 1998–2000 whose respiratory specimens were smear-positive were included in the study. It should be noted that the rural area in this study was immediately adjacent to the urban area, therefore, the DOTS and the non-DOTS groups were comparable. All the new cases of TB in the patients we studied had been diagnosed with sputum smear-positive tests using the Ziehl–Nielsen technique. The 178 patients in the rural ar- eas were kept on the DOTS regimen (i.e. the drugs were dispensed in the health centre by a health worker, who observed the drugs being taken) in accordance with the national protocol (DOTS is obligatory in rural areas but optional in urban areas). The 82 patients in the urban areas (control group) were treated using a non-DOTS strategy, i.e. the same drug regimen was used and the drugs were dispended in the health centres but pa- tients consumed them at home; there was no observation of their taking the medication. All of the new TB cases underwent a 6-month treatment regimen according to ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما the World Health Organization protocol: 2 months with 4 drugs (isoniazid, rifampin, pyrazinamide, ethambutol) and 4 months with 2 drugs (isoniazid, rifampin) [8]. In the DOTS group, these drugs were given to fasting patients by trained health workers in 15 health centres every morning. In the non-DOTS group, the drugs were taken by the patients themselves without any obser- vation of consumption. During the treatment period, sputum samples were taken at the end of the 2nd month and the beginning of the 5th month. Patients who were smear-positive after the 2nd month had an extra month on the 4-drug regimen. Those smear-positive at the begin- ning of the 5th month were considered treat- ment failures. In these cases, drug resistance was suspected and samples were sent to the Tuberculosis and Respiratory Disease Research Centre in Masih Daneshvari Hos- pital, Tehran for further investigation. Results were entered into SPSS, version 11, statistical software and analysed using 2 and Fisher’s exact tests, with P < 0.05 considered statistically significant. Age, sex, clinical manifestation, TB in the family and also the treatment results (sputum smear status after 2 months and at the beginning of the 5th month) were recorded. A total of 260 smear-positive patients were diagnosed during the study period. The clin- ical manifestations of TB are presented in Table 1. There were more females (51.5%) than males. The largest age group was fe- males aged 15–24 years. There were no significant differences between the 2 groups with regard to age, sex, socioeconomic sta- tus or nationality. Of the 178 patients who were under treatment using the DOTS strategy, 3 (1.7%) were still smear-positive at the beginning of the 5th month. Of the 82 patients who were on the non-DOTS regimen, 6 (7.3%) were still smear-positive at the beginning of the 5th month (Table 2). In the DOTS group, 91.0% were smear- negative at the end of the 2nd month and 98.3% at the beginning of the 5th month. In the non-DOTS (control) patients without direct observation, the corresponding values were 81.7% and 92.6% (P < 0.05). Through numerous studies and experience in various situations, it is now widely ac- cepted that the most important cause of fail- ure in TB treatment programmes is irregular drug-taking by patients, which is a direct consequence of poor motivation [5]. A suc- cessful, cost-effective, community-based programme of directly observed therapy, using volunteers, clinic staff and commu- nity health workers or trained personnel can help ensure adherence to therapy [9]. This is the first report about the efficacy of DOTS strategy in Golestan province in the Islamic Republic of Iran. The rate of negative smears at the beginning of the 5th month is an excellent index for evaluating the efficacy of TB treatment. Treatment ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما failure in patients on the DOTS strategy was much lower than in the non-DOTS group. This correlates well with the results of stud- ies done in other Asian countries [10–15]. In a study in China with a large sample population the treatment failure rate in pa- tients under the DOTS strategy was 6.2% [12]; in our study we had better results with patients on the DOTS strategy. In Iraq treatment failure in patients un- der the DOTS strategy was 2.0% and in the control group it was 5.8%, which is agree- ment with our findings, and emphasizes the importance of the DOTS strategy [14]. Mu- rali and Udaya also reported a lower treat- ment failure rate among DOTS patients, 9% compared to 47% in the non-DOTS group, although the failure rate in our study was lower than theirs [15]. Although these differences were not statistically significant, there is a basis for further investigation. In our study, the high- est rate of infection was among the younger age group, 15–24 years; in a study done in Canada, the highest rate of infection was in those > 65 years [16]. Given that in industrialized countries TB is mostly the reactivated form of primary infection [17], the high prevalence in older age groups may be a result of weaker immunity in those patients. All the patients in our study were, however, new TB cases. The clinical syndrome in this study (spu- tum, cough, fever, haemoptysis) had a good correlation with a previous study [18]. The most common clinical manifestation was cough, which was also found in other stud- ies [13,15,16]. In accord with those of other investiga- tions, the findings of this study demonstrate the effectiveness of the DOTS strategy in treatment of TB patients. Considering conversion rates, our study showed a better efficacy in the DOTS group compared to some studies from other countries, such as a study in India which showed a 90% success rate with the DOTS strategy compared to 81% for the control patients [11]; in a study done in Iraq, the treatment success rate with the DOTS strategy was 96.2% but 76.2% in the control group [13]. In 27.7% of cases there was a positive family history of TB. This agrees with the findings of a study from Masih Daneshvari teaching hospital in Tehran (26.9%) [18]. Therefore, family members of infected pa- tients also need to be checked. In a study in Pakistan it was reported that only 8.5% of family members of TB patients had the disease itself. It is nevertheless an important point to be taken into consideration when managing TB patients [19]. The DOTS strategy is an excellent way to reduce treatment failure, therefore, the ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما health system authorities in our country should make every effort to get the coopera- tion of all the relevant organizations (both private and public sector) to implement the DOTS strategy suggested by the World Health Organization. It is also a safe and effective way of controlling and prevent- ing multidrug resistance in Mycobacterium tuberculosis. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Global tuberculosis control: surveillance, planning, financing. WHO 2007Report The eleventh report in this series charts progress towards the Mil- lennium Development Goals (MDGs) as related to tuberculosis (TB). Focusing on five key indicators – case detection, treatment success, incidence, prevalence and deaths – Global tuberculosis control: sur- veillance, planning, financing presents the fullest possible assessment of progress towards MDG targets in the world as a whole, and in each WHO region and country. The report compiles case notifications and treatment outcomes for 200 2005countries up to the end of . It also investigates how ef- fectively national TB control programmes have begun to implement WHO's expanded Stop TB Strategy, and sets out costs, budgets, ex- penditures and sources of funding. The report summarizes progress on initiatives, including the development of public–private partner- ships in TB control, human resources development, the management of drug-resistant TB, and collaborations in TB and HIV/AIDS control. Further information about this and other WHO publications is available at: http://www.who.int/publications/en/ ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Tuberculosis (TB) has traditionally been regarded as a pulmonary disorder. How- ever, nearly 17.9% of TB cases have only extrapulmonary manifestations [1]. Breast and skin are considered to be rare sites of extrapulmonary mycobacterial infection, comprising 0.1% to 0.5% of all TB cases, respectively [2]. Tuberculous mastitis is an uncommon lesion [1] even in countries where the incidence of pulmonary and ex- trapulmonary TB is still very high [3]. The clinician may confuse TB of the breast with either breast carcinoma or abscess [4–6]. Although the usual form of the disease is unilateral, it occasionally presents bilater- ally [7]. We summarize here 4 cases of tubercu- lous mastitis presenting to the National Re- search Institute of Tuberculosis and Lung Disease over a 5-year period at Shaheed Beheshti Medical Science University. A 37-year-old female was referred to this centre for management of left breast pain. There was no history of recent pregnancy or lactation or a history of breast trauma. She first noticed a painful mass in her left breast 1.5 years ago, 20 days later she noticed a vis- ible lesion on the breast skin with purulent discharge. Subsequently, sinus tracts with purulent discharges appeared in a total of 6 locations. She received at least 8 courses of antibiotic therapy with no therapeutic effect before she was referred to our centre. On physical examination, a painful nod- ular mass was detected in the medial half of the breast. A surgical scar, more than 6 healed sinus tracts, and a region containing exudate were detected (Figures 1 and 2). Sonography showed a well-defined 20 12 14 mm mass in the medial portion of the left breast. Based on the sonography and mammog- raphy results, the patient underwent biopsy with a suspicion of malignancy. Diagnosis of granulomatous mastitis was established from biopsy. A second biopsy of the new sinus tract also confirmed a granulomatous lesion and there was no sign of malignancy. The fine-needle aspiration (FNA) smear was negative for acid-fast bacilli (AFB). FNA was negative on polymerase chain reaction testing for mycobacterium TB and culture-negative for actinomycetes and fungi. The angiotension converting enzyme ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما level was in the normal range. Complete blood counts and liver function tests were all normal. The patient’s purified protein derivative (PPD)/Mantoux skin test was 16 mm in diameter. Chest radiography was intact and there were no changes compatible with TB or sarcoidosis. High resolution computerized tomography of the lung was normal. Although we found no laboratory evi- dence for TB mastitis, due to the Mantoux skin test and strong clinical suspicion, the patient was put on standard anti-TB regi- men of isoniazid, rifampin, pyrazinamide and ethambutol. In the subsequent 6-month period no other sinus tracts appeared and the breast mass showed an apparent remis- sion on multiple sonography. A 41-year-old woman presented with pain and swelling of the left breast. She was a known case of TB from 18 years before and had received a complete 6-month period of anti-TB treatment. The patient’s symptoms had started one year after discontinuation of breast feeding with pain and a mass in the superolateral portion of the left breast. The symptoms progressed over a few months and no associated constitutional symptoms were found. After ultrasound examination, a biopsy was taken with the suspicion of breast car- cinoma. The result was reported as granu- lomatous mastitis and after ruling out other diagnoses, the patient underwent anti-TB treatment with an excellent response; how- ever, she discontinued medication after 3 months. Two weeks later, the left breast mass reappeared, associated with multiple lymph nodes in the left axillary pit. Sonog- raphy supported the prior diagnosis of gran- ulomatous mastitis. This time, a 6-month course of anti-TB therapy was started with a standard regimen and no mass or abscess was reported on ultrasound examination during the course of treatment. After 6 months of follow-up, starting from the end of treatment, no mass or lymphadenopathy appeared and the patient complained only of minor pain. A 19-year-old female was referred to this centre with a complaint of breast pain and retraction of the right nipple with yellowish discharge. She was nulliparous with no his- tory of breast trauma. Her symptoms began 1 year before referral and gradually pro- gressed. She took various different kinds of antibiotics during this period. On physical examination, painful masses were palpated in the centre and superolateral quadrant of the right breast. An enlarged lymph node in the right axilla, dimension 20 × 20 mm, was also palpable on physical examination. Full nipple retraction and purulent yellow- ish discharge from the same nipple was de- tected (Figures 3 and 4). At least 4 openings of sinus tracts were visible with no purulent discharge at the time. On sonography, there was thickening of the breast skin in the superolateral and su- percentral portion. This abnormality along with several hypoechoic lesions suggested abscess formation. There was evidence of left pleural effusion on chest radiography. The PPD/Mantoux skin test was 11 mm. Complete blood count and liver function tests were all normal. Sputum smears were negative; however, direct smear of breast discharge was 1+ positive for AFB, com- patible with World Health Organization criteria. Aspiration of pleural fluid was not performed because of the patient’s dissent. Considering AFB secretions and con- comitant pulmonary and pleural involve- ment, a diagnosis of TB was suggested and a standard anti-TB drug regimen of isoniazid, ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما rifampin, pyrazinamide and ethambutol was prescribed. After beginning anti-TB medication, the breast discharge and pain decreased significantly after 2 months. Direct smear of breast discharge became negative after 2 months although discharges continued. After 6 months of chemotherapy, a decision for surgical intervention was made to al- leviate persistent discharges; however, the patient left the country and the treatment remained interrupted. A 49-year-old woman presented with sup- purative secretions from her left breast. She had suffered from pain and a mass in the left breast for 4 years. An ultrasound- based diagnosis of suppurative abscess had been suggested and she underwent antibiotic therapy plus aspiration. Due to recurrence, repeated courses of antibiotic therapy were given. A sinus tract appeared in the left breast, superior to the nipple with suppurative secretions 8 months previously. Bacteriologic studies at different periods from abscess and sinus tract secretions gave negative results. Mycologic studies led to negative results. The PPD/Mantoux skin test was also negative. Liver function tests and other tests were normal and lung com- puterized tomography scan gave no abnor- mal findings. Smear and culture from sinus tract secretions were sent for bacteriological evaluation which showed negative-smear results. However, AFB colony growth in the culture was confirmed and using dif- ferential tests, Mycobacterium tuberculosis was eventually identified. The first case had a granulomatous lesion in her breast. Granulomatous mastitis is a descriptive and non-specific term which encompasses many specific lesions such as TB, fungal infections, sarcoidosis and granulomatous reactions in carcinoma [8]. Several diagnoses must be ruled out to es- tablish TB mastitis. Idiopathic granulomatous mastitis was first described in 1972 [9], for which TB mastitis should always be considered as a differential diagnosis in women with a positive history of exposure to TB patients [10]. This kind of exposure occurs very frequently in endemic countries. A similar granulomatous reaction has also been de- scribed in actinomycosis of the breast which ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما was ruled out by culture of the patient’s discharge obtained by FNA in our patients [11]. The best description of TB mastitis is by Shinde et al. in India [12]. They showed that a lump in the breast with or without ulceration was the commonest presentation in TB of the breast, the other less common forms being diffuse nodularity and multiple sinuses. Concomitant axillary lymph nodes were found in one-third of the patients. Our second case had similar symptoms. This suggests that a young, multiparous, lactat- ing woman with a similar lesion should always raise the suspicion of TB mastitis, although pre-therapeutic pathologic confir- mation of benign conditions is mandatory. Similar results were reported by Al-Marri et al. from 13 multiparous women with TB of the breast from Qatar [13]. All of them pre- sented with a lump, 2 had nipple discharge and 1 had a palpable axillary node on the same side. In that study all diagnoses were confirmed histologically. In both these previous studies the type of breast lesion was similar to what we found in our cases; in our second case, the breast lesion emerged as an abscess, correspond- ing to Daali’s report [14]. According to different studies, the age of patients with breast TB ranges between 20 and 40 years, which corresponds to the age of our patients. However, many other investigators have also reported breast TB in post-menopausal ages [4,6,15–17]. With regard to pathogenesis, the route of entry of mycobacterial TB to the breast var- ies. Spreading of infection from other TB foci via haematogenic path during primary infection is a common route [18]. TB mastitis should also be considered in immunodeficiency states in which patho- genesis is also via the haematogenic route [19,20]. There is also much controversy about the diagnostic criteria for TB mastitis and different surgical biopsies and FNA have been recommended [13]. However, in several studies it has been shown that the overall rate of positivity of AFB in nipple discharge, FNA and tissue samples was 12.0%–22.7%. [12,16,21]. Although we obtained samples for smear and direct visualization from breast discharge samples and FNA for all our patients, only patient 3 was AFB-positive. The differential diagnoses include duct ectasia, foreign-body giant-cell reaction with fat necrosis, foreign material or an abscess, idiopathic granulomatous mastitis, sarcoidosis, syphilitic gumma and Wege- ner’s granulomatosis [16,22], all of which must be ruled out using proper diagnostic tests. Various modes of therapy ranging from chemotherapy alone to mastectomy have been suggested for tuberculous mastitis [12]. In a study by Shinde et al., 14% of patients required simple mastectomy due to either lack of response to chemotherapy (10%) or large painful, ulcerative lesions involving the entire breast (4%). Axillary dissection was performed in only 8% of patients with large ulcerated axillary nodes. Al-Marri recommends that incisional or ex- cisional biopsy together with anti-TB drugs is the most successful treatment [13]. Daali et al. recommend anti-TB chemotherapy for 9 months [14]. The clinical course was favourable at 6 months. They recommend that anti-TB antibiotic therapy might be accompanied by surgery in case of exten- sion, therefore several authors suggest that anti-TB antibiotic therapy in combination with aspiration or surgical drainage are usu- ally associated with an excellent outcome [4,6,7,23,24]. We recommend at least 9 months of anti-TB therapy and surgery in cases with unsatisfactory treatment results. Although TB of the breast is a rare en- tity, it should always be considered in the ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما differential diagnoses of granulomatous mastitis in endemic areas. It seems that empirical therapy might be effective when primary evaluations for other causes lead to no definite conclusion. Diagnosis is based on histopathology and it is less probable that a microbiological diagnosis be made. Therapy includes at least 6 months of anti- TB medication, and surgery when indi- cated. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Diagnostic and treatment delay in tuberculosis This document presents an in-depth analysis of the health-seeking be- 7 haviour of patients and the health system response in countries of the Eastern Mediterranean Region. This study was conducted in order to obtain reliable information about the extent of diagnostic and treat- ment delay and the factors implicated in the Eastern Mediterranean Region. It is a detailed analysis of the health-seeking behaviour of tu- berculosis patients from onset of symptoms until reaching the health system, final diagnosis and treatment. It also provides a thorough analysis of the health system in relation to tuberculosis. It is envisaged that the information provided by this study could as- sist health policy-makers in devising suitable interventions in order to increase case detection and reduce transmission of infection in the community, and hence achieve proper tuberculosis control. The full text of this document is available free on line at: http://www. 710emro.who.int/dsaf/dsa .pdf ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 72 144 P<0.05 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Brucellosis is a zoonotic infection caused by bacteria of the genus Brucella; species considered important agents of human dis- ease are B. melitensis, B. abortus and B. suis [1]. Brucellosis has worldwide distribution; but nowadays the disease is rare in the United States of America and in many other industrialized nations because of routine screening of domestic livestock and animal vaccination programmes [2–4]. This dis- ease is, however, still a leading zoonosis in the countries of the Eastern Mediterranean Region and a disease of economic impor- tance [5]. Additional losses result from human infection (undulant fever) with its prolonged misery, debility and generalized aching, which may last for months or years [5,6]. Sheep and goats and their products are the main sources of infection. Conse- quently, brucellosis has been an occupa- tional risk for farmers, veterinary surgeons and employees in the meatpacking business [1]. Non-occupational sources of infection include consumption of fresh, unpasteur- ized goat cheese and raw fresh (untreated) milk [7]. In Egypt, the infection is mostly caused by B. melitensis and B. abortus [2]. The Epidemiology and Surveillance Unit of the Egyptian Ministry of Health and Population has recorded a substantial increase in the number of patients with brucellosis in the recent past, from 24 cases in 1988 to 1429 in 1998 [8]. An earlier report (1992) described the distribution pattern of human cases of brucellosis during the period 1982–91: the infection rate was generally low except in 1987 and 1991, when there were marked increases in numbers of cases. This was clearly observed in Alexandria and Meno- fiya, and also in Giza and Domiyat [9]. A full description of the epidemiology of the disease is needed for planning of any intervention for its control. Therefore, the objectives of this study were to define the trend of brucellosis in Alexandria governo- rate over a 16-year period and to determine the risk factors. The study was conducted at the Alexandria Fever Hospital. This hospital was selected because it has well-trained staff and well- equipped laboratories and because of its ac- cessibility. Although the hospital is located in an urban area, its catchment area includes not only districts from Alexandria city, but also from nearby rural areas such as El- Amriya and Abis. Although these areas are administratively related to El-Beheira gov- ernorate, they are near enough to Alexandria that, for any case of fever, it is easier to go to Alexandria Fever Hospital than Damanhour Fever Hospital in El-Beheira [8]. The logbooks of the health directorate over the 16 years 1988–2003 were reviewed to detect the annual number of brucellosis cases. The reliability of data from the log- books was checked by comparing with the same data collected by the Department of Communicable Diseases in the Ministry of Health and Population. A case–control study was also conducted in 2001 to determine the epidemiologic risk factors of brucellosis. Cases of brucellosis were identified from the admission log- books of the Alexandria Fever Hospital. Basic demographic data such as name, age, sex and address were collected from the logbooks. For this study, a case of brucellosis was defined as any case hospitalized with a phy- sician’s diagnosis of brucellosis between January 1999 and October 2000. Diagnosis depended on the presence of clinical symp- toms such as recurrent fever, profuse sweat- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ing, headache and generalized bone pain, with rising titre of brucella IgG antibody > 1/160 using a serum agglutination test (tube and slide). Diagnosis was confirmed by performing a blood culture. The total number of confirmed cases was 120, but only 72 were included in the study (all districts). Of the 48 patients not enrolled in the study, 20 lived in another governorate, 10 were abroad, the address was unclear for 12 and 6 refused to participate. For each case, 2 age-matched (± 2 years) controls were selected from households within the same neighbourhood. Informed consent was obtained from all participants following explanation of the aims and importance of the study. Cases and controls were interviewed in their homes by one of the authors us- ing a standardized questionnaire, which covered demographic data (marital status, occupation, educational level and family size), date of onset of symptoms, infection of other members of the family and the main risk factors. These included contact with animals (goats, cattle, sheep, camels, pigs and dogs), mode of contact (clean- ing farms, delivery or handling of abortus and slaughtering animals), consumption of unboiled milk and milk products (cottage cheese and ice cream) and knowledge about brucellosis. Data were analysed using Epi-Info, ver- sion 6. Odds ratio, chi squared, and Student t-test were used in analysis of data. Test re- sults were considered significant if P-value was 0.05. Factors found to be significant in the univariate analysis were included in the multivariate analysis. The number of reported cases is shown in Figure 1. A mean of 61 cases has been reported annually during the period 1988– ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 2003, with an increase in the number of reported cases, particularly in 1991, 2000 and 2003. Sociodemographic characteristics of the 2 groups are described in Table 1. Age range was 8–69 years for cases [mean 37.1 years, standard deviation (SD) 15.4] and 9–67 years for controls (mean 36.7 years, SD 15.1) (P = 0.8). Those with brucellosis (n = 72) were distributed by age as follows: 10 years 2.7%, 11–20 years 15.2%, 21–30 years 19.4%, 31–40 years 19.4%, 41–50 years 26.3%, 51–60 years 8.3% and > 60 years 8.3%. Male to female ratio was 2:1. The most frequent symptoms were re- current attacks of fever (100%), profuse sweating at night with no prior antipyretic (82%) and headache (75%). The overall case fatality rate was 0.0%. Recurrence of symptoms was experienced by 36% of pa- tients (16 males and 10 females; 4 under the age of 15 years) during the first 6 months of illness. Onset of illness was greatest in spring, 44% of cases, followed by summer, 29%. The association between brucellosis and some risk factors is shown in Table 2. Oc- cupation was an important risk factor for ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما brucellosis. Workers in occupations dealing with animals (including farmers, slaugh- terhouse workers, butchers and veterinary surgeons) had a 2.4-fold higher risk of bru- cellosis than those in occupation not dealing with animals (P = 0.009). Breeding animals was a significant risk factor, but when the type of animal was considered, the odds for presenting with brucellosis was sig- nificantly higher only for goats (P = 0.006). Other animals (cattle, sheep, camels, pigs and dogs) were not associated with the risk of disease. Eating ice cream from street ven- dors was also significantly associated with brucellosis: the odds were 1.8 times higher among those who ate ice cream from street vendors than those who did not (P = 0.04). All significant variables from the univar- iate analysis were included in multivariate analysis using a conditional logistic regres- sion model. Most were not significantly associated with disease in the multivariate model, with the exception of male sex, jobs dealing with animals, breeding goats and eating ice cream bought from street vendors (unknown source) (Table 3). Brucellosis has been recognized as one of the most common zoonoses in the East- ern Mediterranean Region, with more than 45 000 cases reported annually. The epide- miological data on the disease is frequently incomplete. This is partly explained by the lack of proper laboratory facilities in some remote areas as well as by poor cooperation and exchange of information between vet- erinary and health services [10]. In this study, a rise in the number of brucellosis cases was noticed in 1991 and this was in agreement with Wassif et al., who described the distribution of brucel- losis cases in a number of Egyptian gover- norates: the highest number of brucellosis cases was in Alexandria governorate in 1991 [9]. A similar rise was reported in a neighbouring country, the Palestinian Ter- ritories, during the late 1980s (W. Tarazi, unpublished report, 1990). The increase in number of brucellosis cases in 1991 may be a true epidemic of brucellosis or a ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما false increase because of the use of better diagnostic measures and more meticulous notification [9]. In fact, the number reported was probably much lower than the actuality as under-diagnosis and under-reporting of cases is a recognized problem in many developing countries [10]. It has been es- timated that for each reported case, there are at least 2 additional cases that are not reported or not diagnosed [11]. Therefore, the actual number of cases may be up to 3 times as many as the reported number. We also noted another increase in number of cases between 2000 and 2003. This could be related to improvement in the regulated surveillance system for notifying brucel- losis cases accomplished by the Ministry of Health and Population in 2001 [8]. The age group 15–50 years was the most commonly affected with brucellosis in this study. Comparable findings have been reported from Kuwait (mean age 34.4, SD 11.5, years), Saudi Arabia (mean age 33.8, SD 13.9, years) and Djibouti (mean age 31.6 years) [12–14]. Brucellosis is pre- dominately an occupational disease [1], so it would be expected that this age group would be the most affected, having been exposed longer to risk factors related to their occupation. More males were affected than females (2:1) in our study and this is in accord with studies conducted in Sharkia governorate (Egypt), Kuwait, Saudi Arabia and India [9,12,13,15]. This sex distribution in the incidence of brucellosis infection may be because males are more concerned in ac- tivities such as slaughter and handling of carcases, and as a consequence they are at greater risk of exposure to infection. In both children (0–14 years) and adults, the disease was more prevalent among males. Abdou reported similar conclusions in a 1995 report [16]. Male children in the areas studied may be exposed to the same risk factors as male adults. Other studies have, however, observed that the incidence of infection was greater in females than males (or roughly equal) [17,18]; in the study areas, those milking the cows, and thus having a higher chance of contact and acquiring infection, were mainly females. Recurrent attacks of fever with profuse sweating at night with no prior antipyretic were the most prevalent symptoms in our study and in other studies carried out in south Jordan, Yemen and Greece [18–20]. The case fatality rate in our study was 0%; the same rate was reported in a 1998 study from Palestine [11]. The disease is insidious ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما in onset with long-standing fever, so there is more likelihood of diagnosis. In addition, the availability of well-known and effec- tive therapy enhances the outcome of the disease. Recurrence of symptoms during the first 6 months of illness among cases we studied was 3 times higher than that reported by Awad and 11 times higher than reported by Shehata [11,12]. The onset of symptoms showed a sea- sonal pattern with high incidence in spring followed by summer. Several other stud- ies have shown a similar seasonal pattern [9,11,12]. The increase is believed to be linked to the delivery (parturition) season of sheep and goats where there is a greater possibility for direct contact with vaginal discharge, foetuses and placentas, which may play a major role in increasing risk of exposure to infection. People in occupations dealing with animals were at greater risk of developing brucellosis. A study conducted in several regions in Lebanon on 597 persons in high- risk occupations found overall prevalence of IgG and IgM antibodies for brucella was around 60%. Exposure to brucellosis was high among persons in high-risk occupa- tions from all surveyed regions in Lebanon [21]. Similarly in a case–control study con- ducted in Yemen, occupation dealing with animals was a significant risk factor for infection; socioeconomic and educational factors were independent risk factors [22]. The majority of the participants in our study had not been educated beyond preparatory school level and were mostly of a low socio- economic standard so we could not confirm these issues as independent risk factors. Contact with goats was an important risk factor in the group we studied. In a Saudi Arabian case–control study, greater risk for brucellosis was associated with products derived from sheep and goats as opposed to camels and cattle [23]. Direct contact with domestic animals and consumption of raw products of animal origin have also been identified as major risk factors [24]. In a study done in Eritrea, the highest preva- lence was among dairy farm workers and owners in randomly selected dairy-cattle farms, followed by veterinary personnel [25]. A higher risk was associated with the presence of sheep on the farm. Eating ice cream from street vendors (i.e. unknown source, possibly made from the milk of infected animals) was an im- portant source of infection in our study. Kolar also noted that there had been reports of transmission of the disease from eating ice-cream [26]. Consumption of unboiled milk appeared to have no association with brucellosis. In a study from Spain, no statistically significant relationship was demonstrated between consumption of dairy products and being seropositive for brucella antibodies [27]. This may be due to the fact that drinking fresh milk without boiling it is an uncom- mon practice in the study areas owing to fear of contracting other infections such as tuberculosis. Other studies have, however, found that milk and dairy products appeared to be associated with brucellosis [19,28]. In a study on 5726 blood specimens from children aged 14 years for serological evidence of brucellosis, > 60% had a history of both consumption of fresh goat’s milk and close contact with animals [15]. Eating soft cheese and butter were not significantly related to brucellosis in our study; our cases were from urban areas, however, and cheese and butter were bought from sources which used pasteurized milk in manufacturing. Raw milk and milk products from infected sheep, goats and cattle have been cited as important sources of infection with brucellosis; soft cheeses made using traditional methods which do not ensure ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما killing of organism have also been impli- cated [11,13]. Occupation dealing with animals, breed- ing goats, male sex and eating ice cream bought from street vondors were the only significant risk factors in the logistic regres- sion model in our study. Similarly, Bikas et al., using multivariate stepwise analysis, found that that occupation dealing with ani- mals, place of residence, absence of stables and trauma during animal delivery were the most important risk factors remaining in the model [29]. Our findings emphasize the importance of contact infections, i.e. contact with infected animals and their products, as method of transmission of brucellosis rather than in- gestion of contaminated animal products. This means that prevention of brucellosis in man ultimately depends upon its control in the principal animal hosts. Therefore, in- formation is needed concerning the present incidence of brucellosis in livestock (sheep, goats and cattle). A control programme for human brucel- losis would depend to a large extent on public health education about the disease and its risk factors, good administrative arrangement and ensuring the maximum cooperation of the community (particu- larly between health and veterinary authori- ties). Active cooperation between health services and veterinary services should be promoted. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Cutaneous leishmaniasis (CL) is still en- demic and an important health problem in many countries of the Eastern Mediterra- nean Region [1–4]. Three species of Leish- mania parasites cause CL in the old world: L. major, L .tropica and L. aethiopica [5]. Zoonotic cutaneous leishmaniasis (ZCL) is endemic in half of the 28 provinces of the Islamic Republic of Iran. The most important hyperendemic focus of ZCL has been located in Isfahan province, in the central part of the country, where the great gerbil, Rhombomys opimus, is the principle reservoir host of the disease [6]. In the other high risk parts of the country (the west and south-west regions), Tatera indica is the primary reservoir host [7]. In both the above-mentioned areas, Meriones libycus plays a secondary role as a source for trans- mitting of L. major by sandfly vectors [6,7]. In the central parts of the Islamic Republic of Iran (Isfahan province) as well as south- east of the country, female sandflies of Phlebotomus papatasi have been reported as the main vector of CL responsible for L. major infection [8,9]. In southern Islamic Republic of Iran, there is concern that the annual incidence of CL has gradually risen over the past dec- ade, increasing from 1560 cases in 1991 to 3861 in 2001 (unpublished data). The most likely reason for this trend is an increase in human–sandfly contact. This is attributed to the development of villages and the spread of the human population into the habitats of the local vectors, P. papatasi, as well as rodents which act as reservoir hosts. It should be emphasized that most Iranian cases of CL are zoonotic. Our recent studies in some foci of ZCL in southern Islamic Republic of Iran showed that Mer. libycus was the main reservoir host, and R. opimus and T. indica were absent [10,11]. Although there are several studies on different aspects of disease in the region, there has been no comprehensive study on vectors of ZCL in the southern part of the Islamic Republic of Iran. The aim of this study was to determine the species composition of reservoirs and vectors from the field in the Marvdasht focus of ZCL, southern Islamic Republic of Iran. In addition, we evaluated a new mo- lecular technique, nested polymerase chain reaction (PCR), for species determination. The study was carried out in 3 villages (Raja-Abad, Ghorbanlak and Soltan- Velayat) in the Mohammad Abad district (52° 56 E, 29° 54 N) at an altitude 1595 m above sea level and 9–11 km from Mar- vdasht city. The weather is very hot in the summer and cold and snowy in the winter. The major activities of the population are agriculture and animal farming. Rodents were captured monthly by Sher- man live traps during October to December 2003 and April to October 2004. Traps were baited with roasted walnuts and placed in the active burrows. The traps were set in the early morning and evening once a month. Around 20–25 traps were used for each col- lection. Collected animals were transferred to the laboratory for preparation of smears from the ears or any suspected skin lesion of the animals. The prepared smears were stained with Giemsa and checked under light microscope for the presence of Leish- mania parasites. Similar smears were pre- pared from the hand of 2 local cases of CL (a 5-year-old girl from Ghorban-Lak village ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما and a 15-year-old boy from Raja-Abad vil- lage). Identification was made using the standard reference for rodents in the Islamic Republic of Iran [12]. Total DNA was extracted from positive smears by digestion in lysis buffer. The dry smear was scraped with a sterile scalpel and dissolved in 200 µL buffer [50 mM Tris- HCL (pH 7.6), 1 mmol/L EDTA, 1% Tween 20] containing 8.5 µL proteinase K solution (19 µg/µL), in a 1.5 mL tube. The tube was incubated for 2 hours at 56 ° before add- ing 200 µL phenol–chloroform: isoamyl alcohol (25:24:1 by volume). After being shaken vigorously, the tube was centrifuged at 6000 × g for 10 minutes and then DNA was precipitated with 400 µL pure ethanol, re-suspended in 100 µL double-distilled water and stored at –20 ° before being used in PCR. The variable segment on minicircles of kine- toplast DNA from any Leishmania parasites present in the smear scraping was amplified using 2 rounds of nested PCR [13]. The primers were CSB1XR (CGA GTA GCAGAA ACT CCC GTT GA) and CS- B2XF (ATT TTT CGC GAT TTT CGC AGA ACG) for the first round and LiR (TCG CAG AAC GCC CCT) and 13Z (ACT GGG GGT TGG TGT AAA ATA G) for the second round. Reference strains of L. major MHOM/IR/XX/LV114) and L. tropica (MHOM/IR/89/ARD22) from Tehran University of Medical Sciences were used as standards. The first round reaction mixture con- tained 250 µmol/L deoxynucleoside- triphosphate (dNTP), 1.5 µmol/L MgCl2, 1.0 U Taq polymerase, 50 mmol/L Tris-HCl (pH 7.6), 1% Tween 20, and 40 ng each of primers CSB1XR and CSB2XF, in a final volume of 25 µL. The DNA was amplified for 25 cycles in a Progen thermocycler (Thecne, Cambridge, UK) set to run at 94 ° for 1 min, 54 for 1 min, and then 72 ° for 1 min in each cycle. The first round product (1.0 µL of a 9:1 dilution in distilled water) was used as tem- plate for the second round, in a total volume of 25 µL and under similar conditions to those for the first round except using LiR and 13Z as the primers. Sandflies were collected bi-weekly by an aspirator from indoor areas (guest rooms, sleeping rooms, latrines) and livestock sta- bles, 2 hours after sunset during August and September 2004 (the main activity season of sandflies). Temperature and humidity were also recorded. The collected sandflies were transferred to the laboratory and dissected in a drop of normal saline. For each female, the head with the 3 last segments of abdomen was cut and mounted in a slide with a drop of Puri medium for species identification. The remaining body parts of each specimen were transferred to a microtube filled with methanol 95% for DNA extraction and PCR examination. DNA was extracted as described by Mo- tazedian et al. [14]. Briefly, individual female, unfed and parous sandfly bodies were homogenized with a sealed pasture pipette in 1.5 mL tubes. Then 200 µL lysis buffer [50 µmol/L Tris-HCl (pH, 7.6); 1 µmol/L EDTA; 1% Tween 20] and 12 µL proteinase K (19 µg/mL) were added and incubated in 37 °C overnight before 300 µL ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما phenol–chloroform–isoamylalcohol (25:24:1 by volume) were added. After being shaken vigorously, the tube was cen- trifuged at 10 000 rpm for 10 min and then the DNA in the supernatant solution was precipitated with 400 µL cold, pure ethanol, re-suspended in 50 µL double-distilled wa- ter and stored at –20 °C before being used in PCR. Nested-PCR was employed for detec- tion of L. major in sandfly specimens [13]. During this study 126 rodents were captured and identified as Mer. libycus (75.4%), Microtus arvalis (10.3 %) and Cricetulus migratorius (14.3%) (Table 1). Although all collected animals were examined for parasite infection under light microscope, amastigotes were only found in smears of 8 (8.4%) from 95 Mer. libycus (Table 2). Each infected Mer. libycus had at least 1 lesion either on an ear or at the base of its tail. Amastigotes from infected Mer. libycus were similar, with a mean diameter of 4 µm (Figure 1). Parasite infection was observed among males and females animals from 2 of 3 study villages. In our study area, Mer. liby- cus had diurnal activity and each Meriones colony was around 2 m deep, with several burrow entrances. All 30 smears from host and reservoir were amastigote-positive using micros- copy. Out of 30 smears tested, 24 were from Mer. libycus and 6 from 2 local cases of ZCL, which were also shown positive for Leishmania DNA by PCR. For each PCR- positive sample, the second-round products of the nested PCR were identical to those of the L. major reference strain with a main band of 650 bp and distinct from those of the L. tropica standard with its main band of 750 bp (Figures 2 and 3). A total of 200 parous, unfed female sand- flies were checked by PCR for the pres- ence of Leishmania parasite. They were P. papatasi (75%), P. segenti (15%) and P. caucasicus (10%). Among the examined sandflies, only 4 out of 150 P. papatasi (2.7%) were found naturally infected with L. major (Figure 3). Control of leishmaniasis in areas of ende- micity requires a thorough knowledge of Leishmania ecology and epidemiology. There is a major problem for epidemiolo- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما gists both in the identification of reservoir hosts and in the detection of vectors. The suggested criteria for incrimination of a vector are anthropophily and common infection with the same Leishmania parasite as that found in man in the same places [15]. Dissection of sandflies for detection of Leishmania parasite is difficult, and accu- rate estimates of infection are not possible [16]. Isoenzymes provide the gold standard for species identification and reference strains of Leishmania [3]. This method was used by Yaghoobi-Ershadi et al. [8,17] who identified single infections of the strain (zymodeme MON26) of L. major in P. pap- atasi and P. caucasicus in Isfahan province, central Islamic Republic of Iran [8,17]. However, this method has the disadvantage of requiring the culture of a large number of parasites, and primary isolates can easily become contaminated, or in mixed infection yield only the strain that grows fastest in laboratory conditions [9]. At the present time the applicability of molecular techniques (PCR) including kine- toplast DNA for detection and identification of Leishmania within sandflies by DNA hy- bridization has been demonstrated [18–20]. The highly sensitive technique of PCR has been used before for detecting Leishmania in sandflies of the new world [21], Islamic Republic of Iran [9] and India [22]. This is the first report of P. papatasi (collected from indoors) as a main and primary vec- tor of L. major in the southern part of the Islamic Republic of Iran. The results of our study showed that nested-PCR is faster and more reliable for directly detecting L. major in wild populations of P. papatasi in com- parison with other methods, i.e. dissection and isoenzyme techniques [9,16]. This spe- cies has been reported as a principle vector of ZCL transmitted to humans in other parts of the Islamic Republic of Iran [8,9]. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Another important finding of this sur- vey was confirmation of Mer. libycus as a principal reservoir of ZCL in rural regions of Marvdasht focus. This rodent has been reported as a main reservoir in the foci of Arsanjan and Neiriz, adjacent to our study areas [10,11]. In the central and west of the Islamic Republic of Iran, Mer. libycus is a secondary reservoir host, where R. opimus and T. indica are present [1,7]. L. major has also been isolated from Mer. libycus in ZCL foci of Saudi Arabia [23] and Uzbekistan [24]. This animal is common and widespread in central and southern Islamic Republic of Iran. It seems that agriculture and rural development in many parts of this region will lead to more human contact with the rodent–sandfly– rodent cycle of Leishmania transmission. The authors would like to thank the Institute of Public Health Research, Tehran Univer- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما sity of Medical Sciences for financial sup- port of this study. Great appreciation goes to the staff of Kazerun Health Research station for their valuable collaboration in the field. This study is dedicated to our colleague and friend M.M. Gassemi who passed away in a car accident during this study. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ينطسلف ،تيزيرب ةعماج ،ةسدنلها ةيلكو ةيعمتجلماو ةماعلا ةحصلا دهعم :نيورـتكللإا ديبرلا :لوبقلا ، :ملاتسلاا يضارلأاب ةيحصلا ةياعرلا زكارم في ةيبطلا تايافنلا ةلجاعم ةلتحلما ةينيطسلفلا بيطلخا دحمأ ماصع ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 10.5% ][ 13.8% ]22[ 77.3% ][ركُذ ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما في ةيبطلا تايافنلا ةرادإ ،دحمأ ماصع ،بيطلخادهعم .مئاقلا عضولا في ةسارد ينطسلف،تيزيرب ةعماج ،ةيعمتجلماو ةماعلا ةحصلاينطسلف. حسلما .نييطسلفلا ءاصحلإل يزكرلما زاهلجاجئاتنلا ةيحصلا ةياعرلا زكارلم يئبيبلاينطسلف – اللها مار .ةيساسلأا. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The khat plant (Catha edulis Forsk) is a tree of the family Celastraceae that is frequently cultivated in certain areas of East Africa and the Arabian Peninsula. The leaves of the khat plant contain alkaloids structurally related to amphetamine. They are chewed daily by a high proportion of the adult population in Yemen for the pleasant mild stimulant effect. Khat appears to have been used first as a drink prepared from dry leaves, but its effect is weak compared with coffee [1]. It was found later that drying the leaves re- sults in loss of some active constituents [2] and therefore the habit of chewing the green leaves was adopted. For many hundreds of years the custom of chewing khat leaves has been practised for the resulting central stimulant effects [3]. In Yemen, the habit is widespread with a deep-rooted sociocultural tradition. The pleasurable central stimulant properties of khat are commonly believed to improve work capacity, and are used on journeys and by students preparing for examinations and to counteract fatigue. In recent years, because of improved air transport, the consumption of fresh khat leaves has expanded considerably, even to communities in Europe. Early clinical observation suggested that khat had amphetamine-like properties [4]. Subsequent chemical analysis confirmed that the fresh leaves contain a number of compounds, including phenylalkylamine compounds (alkaloids) such as nor- pseudoephedrine (cathine) and alpha ami- nopropiophenone (cathinone), the latter being structurally related [5] and pharmaco- logically similar to amphetamine [6,7]. Khat leaves also contain considerable amounts of tannins (7%–14% in dried material), vitamins, minerals and flavonoids [4,8]. Cathinone is currently believed to be the main active ingredient in fresh khat leaves [7]. Supporters of khat chewing claim that it is useful in diabetic patients because it lowers blood glucose, it acts as a remedy for asthma, it eases symptoms of intestinal tract disorders [9] and maintains social con- tact as a socializing herb [10]. Opponents claim that khat damages health and affects many aspects of life with its adverse social, economic and medical consequences. In Yemen, because of its widespread use, it has become a problem of grave national concern. The objective of this review of the lit- erature was to examine studies on khat, particularly human studies, with special reference to its effects on the body systems and its relationship with common diseases. The effect that accounts for the popularity of khat is its central nervous system stimu- lation, believed to be induced by cathinone, an active ingredient of khat leaves [7]. Cathinone has a more rapid and intense ac- tion compared with cathine due to its higher lipid solubility which facilitates access into the central nervous system. Several studies have shown that the psychostimulant effects induced by chew- ing khat leaves include a moderate degree of euphoria and mild excitement resulting in promotion of social interaction and lo- quacity [8,11,12]. While attaining a state of subjective well-being, the chewers feel an increased alertness and energy together with enhanced depth of perception [7,13]. These effects were found to be maximum between 1.5–3.5 hours after starting to chew [14] and they were progressively replaced by mild dysphoria [14], anxiety, reactive ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما depression [15], insomnia [4,16,17] and anorexia [17,18]. These psychic effects of khat chewing recall those of amphetamine [19], but a major role of environmental factors in the expression of khat effects has also been suggested [14]. It has been proposed that the differences in the effect of khat and amphetamine are quantitative rather than qualitative [20,21]. It has been also reported that the khat-induced behavioural syndrome can resemble hypomania, as it may include hyperactivity and logorrhoea [4]. In recent years khat-induced psychosis has become more common in Europe [22]. Khat may cause a functional psychosis following consumption of exceptionally potent material, when taken in excess or in a predisposed individual [4,23,24]. In the literature reviewed, a number of reports on psychiatric disorders secondary to khat chewing with features of mania-like psy- chosis [25], schizophreniform psychosis [26], paranoid psychosis [27] or symptoms of acute schizophrenia-like psychosis have been documented [11,26]. Moreover, a number of chronic khat chewers experi- enced persistent hypnagogic hallucinations [13]. In one case report from the United Kingdom, khat appeared to induce full- blown paranoid psychosis with the added complication of a suicide attempt [24]. In Kenya, grossly excessive chewing led to psychotic states, which were paranoid in type and transitory in nature [28]. However, when khat is chewed in moderate quantities, there was no increase in psychiatric morbid- ity [28]. Preliminary data on 65 psychotic male patients admitted for psychiatric care in Sana’a because of symptoms uncontrolled by treatment were analysed [29]. They indicated that khat chewing in psychotic patients was likely to be associated with dis- turbance of mood and behaviour, aggrava- tion of delusional symptoms and diminished response to antipsychotic therapy. Previous similar studies have shown that failure to abstain from khat use might prolong a psy- chotic episode, even during treatment with psychiatric medication [11,24–28,30]. Recently, Alem and Shibre considered khat as a substance of abuse and noted that chewing had the potential to complicate psychiatric conditions and forensic events [30]. In comparison with amphetamine, khat is much less likely to cause tolerance [31]. In particular, the stimulant central nervous system effects of khat do not seem subject to the development of tolerance [4], but some degree of tolerance to insomnia [3] and anorexia [32] has been observed in most chronic khat chewers. The issue of dependence on khat has been reviewed by a World Health Organization (WHO) Expert Group on drug dependence, which concluded that khat consumption may induce a persistent psychic depend- ence rather than physical dependence [22] although a certain degree of psychological dependence can occur [33]. However, the psychological withdrawal symptoms after prolonged khat use seem to be limited to lethargy, mild depression, slight trembling and recurrent bad dreams [4,31]. The lack of physical symptoms of withdrawal suggest that only rebound phenomena rather than a specific abstinence syndrome occurs [3,34]. However, the WHO Expert Committee on Drug Dependence recently subjected khat to the preliminary review of psychoactive substances. The Committee reported that khat is believed to be dependence-producing and recommended that there was sufficient information on it to justify a critical review (a fully documented review) at a future meeting [35]. Khat use is often accompanied by the use of other substances. Simultaneous ciga- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما rette smoking is a common habit that might influence khat-induced symptoms [4]. Khat- induced insomnia is frequent and khat users try to overcome this with sedatives or alco- hol [12]. A report from Ethiopia confirmed the simultaneous use of cigarettes, alcohol, gasoline inhalation and glue sniffing with khat among university students, a pattern similar to that reported for substance abuse in other countries [36]. Recent work on healthy Yemeni adult vol- unteers provided evidence that khat chew- ing induced a significant rise of arterial systolic and diastolic blood pressure and pulse rate in comparison with the baseline values [37]. The peak effect on the arterial blood pressure and pulse rate was reached 3 hours after starting to chew, followed by a decline 1 hour after spitting the leaves out. These changes run parallel with changes in plasma cathinone levels during and after khat chewing [38]. Similar blood pressure changes have also been observed in smaller numbers of subjects when pure cathinone in gelatine capsules was taken orally [39] or when khat leaves were chewed [40]. These observations support the suggestion that cathinone is the constituent that is mainly responsible for the increase in arterial blood pressure and pulse rate during khat chew- ing. A possible mechanism is the release of catecholamines from presynaptic storage sites. To gain further insight into the phar- macological effects of khat chewing, a randomized controlled clinical trial of 1 and selective 1 adrenoceptor blockade was conducted on adult Yemeni volunteers [41]. The results indicate that selective 1 adrenoceptor blockade with atenolol pre- vented the elevation of systolic blood pres- sure and increase in pulse rate suggesting these effects are mediated by the stimulant effect of cathinone in khat on 1 adrenocep- tors in the heart. By contrast the intake of indoramin, an 1 adrenoceptor blocker, or placebo premedication failed to antagonize the effects of khat on arterial blood pressure or pulse rate. It can be concluded from the previous studies that khat-induced blood pressure elevation is probably mediated at least in part through its cardiac action. Therefore, khat chewing carries a potential cardiovas- cular risk in patients with hypertension and heart disease [41], and may precipitate the occurrence of cerebrovascular accidents and myocardial infarction in susceptible individuals [42]. The effect of khat chewing on blood pressure and cardiac rhythm among Yemeni patients with hypertension and ischaemic heart disease was also explored using a 24-hour ECG Holter monitor and ambula- tory blood pressure monitor [43]. The study showed a progressive increase in blood pressure and heart rate that occurred rapidly in the 1st hour of khat chewing; the peak effect was reached at the 2nd and 3rd hours and the blood pressure and heart rate values returned to their baseline values by the 4th or 5th hours after stopping chewing. The normal circadian rhythm and the nocturnal blood pressure drop were well preserved in khat-chewing hypertensive patients. The 24-hour Holter monitor revealed ischaemic changes (ST-segment depression) in 20% of patients with ischaemic heart disease and it was associated with sinus tachycardia and extrasystoles during khat chewing. 1 blockade with atenolol reduced the effect of khat on blood pressure and heart rate among a substantial number of patients with ischaemic heart disease and hypertension [43]. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The role of khat chewing as a risk factor for acute myocardial infarction in Yemen was investigated by an experimental and clinical study. In the guinea pig, cathinone induced vasoconstriction of the coronary vascular bed which, unlike amphetamine, was not related to a sympathomimetic ef- fect [44]. In humans, the circadian rhythm of the timing of acute myocardial infarc- tion was shown to be influenced by khat chewing [45]. In non-chewers, there was a progressive increase in number of patients presenting with symptoms of acute myocar- dial infarction from 03:00 to 09:00 hours, and after 15:00 hours there was a gradual decline until there were none in the last 3 hours of the day. This confirms the previ- ously noted rhythmicity in the timing of the onset of acute myocardial infarction which peaks in the early hours of the day. By contrast, the peak period of presentation of acute myocardial infarction in khat chewers was during the afternoon, commencing at 15:00 hours, continuing until 21:00 hours, and then declining to a trough at 03:00 hours. This finding illustrates the shift in the circadian rhythm of acute myocardial infarction presentation associated with khat chewing [45]. The effects of habitual khat chewing on the digestive system mentioned in older studies were based on the clinical observation that khat chewers often complained of symp- toms suggestive of stomatitis, oesophagitis and gastritis. These effects were believed to be caused mainly by the strongly astringent tannins in khat [17,46]. Stomatitis is discussed later. Gastric symptoms were attributed to a hypotonic stomach resulting from the sympathomi- metic action of cathine and its precursor [17]. Recent evidence has shown that khat chewing does indeed delay gastric emptying of a semi-solid meal, probably as a result of the sympathomimetic action of cathinone in khat [19]. Delayed gastric emptying may contribute to an increased rate of gastro- oesophageal reflux manifested as heartburn and acid regurgitation, and to an increased risk of Barrett oesophagus, a precancerous condition (see khat and cancer below). Anorexia frequently follows a khat ses- sion and chewers seldom eat a further sig- nificant meal on the same day. A significant reduction of appetite after khat chewing was recently noted in a study on the subjective effects of khat chewing [18]. This anorectic effect is not due to an effect on ghrelin or peptide YY levels [47] but may be attrib- uted to combined direct central and gastric effects of cathinone in fresh leaves of khat [19]. A common complaint of khat chew- ers is constipation, probably caused by a combination of the astringent properties of the khat tannins and the sympathomimetic properties of cathinone [8]. Habitual users try to attenuate this undesirable effect by food adaptation, notably by eating a meal with high fat content prior to the khat ses- sion in order to facilitate intestinal transit [48]. The constipating effect of khat was suggested by the observation that when a ban was imposed on khat in Aden in 1957, the sales of laxatives decreased by 90%, but returned to the original level soon after the ban was lifted [4]. Recent evidence has shown that chewing khat leaves signifi- cantly slows both the orocaecal transit time [49] and the whole gut transit time [50]. These 2 mechanisms may contribute to the constipating effect of khat. Moreover, khat chewing was found to interfere with the absorption of some orally administered antibiotics, particularly ampicillin [51] and tetracycline [52] resulting in low bioavail- ability. Khat chewing has no effect on gall bladder contraction [53]. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما The liver has been suspected to be particularly vulnerable to the harmful ef- fects of khat use [3,4], and a disturbance in liver function and architecture has been described in experimental animals both on short-term [54] and long-term [55] feeding with Catha edulis leaves. Hepatitis B and C viruses are a cause of major health prob- lems in Yemen; however, no significant relationship was found between hepatitis B surface antigen positivity and khat use in one study conducted in a rural community in Taiz province [56]. In a recent study on acute sporadic hepatitis in adults in Yemen, it was found that hepatitis viruses A to E accounted for only 48.7% of the cases, and in 51.3% of cases no viral cause was identified. There may be an unknown virus responsible or some environmental toxins, such as pesticides in khat leaves [57]. Cer- tainly we have seen patients with abnormal liver function which resolved after khat chewing was suspended and a prospective study is underway. One of the obvious side-effects of chewing khat leaves in males is temporary interfer- ence with micturition with hesitancy and poor flow. Overall urine flow rates were re- cently found to be significantly lower in khat users [58]. This effect is probably mediated through stimulation of 1 adrenoceptors in the bladder neck by the sympathomimetic alkaloid cathinone. These effects were abol- ished by the 1 adrenoceptor blocker, indo- ramin. The consumption of khat is also said to induce an increase in libido, spermator- rhoea and erectile dysfunction [7] but this has not been adequately studied. In the domain of reproductive health, epi- demiological data derived from 1181 de- liveries in Yemen showed that at birth the mean weight of full-term single infants from mothers who chewed khat habitually or occasionally was below average [59]. A study on pregnancy outcome and khat showed a significantly increased incidence of low-birth-weight full-term infants among the offspring of women who chewed khat during pregnancy in comparison to those who were non-chewers during pregnancy [60]. Recent evidence indicates that neonates of mothers who chewed khat during preg- nancy had a significant decrease in all neonatal parameters such as birth weight, length, head circumference and Apgar score at 1 and 5 minutes in comparison with those of mothers who did not chew khat dur- ing pregnancy [61]. This effect was found to increase in severity with the increased frequency and duration of khat chewing during pregnancy. Results obtained from the above-mentioned studies indicate that frequent use of khat during pregnancy may impair intrauterine fetal growth. An ex- perimental study in rats has recently proved that khat can affect intrauterine fetal growth by reducing total fetal fat and weight and by inducing some changes in the chemical composition of fetal organs, particularly the liver, heart and kidneys [62]. This effect was attributed to depletion of carbohydrate material and suppression of DNA and pro- tein synthesis in the fetal organs. Nursing mothers in Yemen frequently complain of poor lactation. Some authors believe that this phenomenon may be re- lated to the use of khat as cathine in khat ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما may inhibit prolactin secretion [3]. Interest- ingly, it has been found that the breast milk of khat-chewing mothers contains cathine, and this compound could even be detected in the urine of one breastfed infant [63]. The effect of khat chewing in diabetic pa- tients is unclear. The very few published papers on this issue make it difficult to draw a conclusion. Some authors believe that the overall effect of khat on diabetic patients is deleterious because the user is less likely to follow dietary advice, and the consump- tion of sweetened beverages with khat ag- gravates hyperglycaemia [3]. A clinical study was conducted on diabetic patients in Yemen over 20 years ago [64]. It showed that when khat-extract was mixed with the glucose given for the glucose tolerance test, there was a significant lowering of blood glucose level in comparison to the non-khat (control) arm of the experiment. This effect was attributed to delayed glucose absorp- tion from the intestine by the action of khat tannins and inorganic ions, particularly magnesium, which produces a substan- tial inhibitory action upon gastrointestinal function. It seems that khat-induced delay of gastric emptying may also play a role in reducing postprandial hyperglycaemia in patients with type 2 diabetes mellitus [19]. By contrast, a similar study on nondiabetic subjects in Somalia showed that khat did not influence to any significant extent blood glucose levels in man [65]. Since khat use is widespread and often persists throughout adult life, a number of studies have been made on the toxicologi- cal aspects of habitual khat use. Owing to its mode of consumption, khat frequently affects the oral cavity and digestive tract. Its effect was found to be clearly dependent on the level of khat consumption [7]. Tumours of the oral cavity (lower maxilla, buccal mucosa and lateral surface of the tongue) were reported in 0.13% of patients seeking treatment over a 2-year period in a stoma- tology clinic in Hodeida, Yemen [66]. Most of them had been habitual khat chewers for more than 20 years, and some of them also chewed shamma (tobacco powder). A simi- lar review of oral cancers presenting over a 2-year period in Asir region of Saudi Ara- bia showed strong circumstantial evidence linking the long-term use of khat with an increased rate of oral malignancies [67]. Tannins in khat can thicken the mucosa of the oropharynx and oesophagus [68] and may be carcinogenic [69]. A recent study in Yemen has shown that oesophageal and gastric carcinoma accounted for 6% of all patients who had an upper gastrointestinal endoscopy (183 out of 3064 patients) over a period of 1 year [70]. A preponderance of women with carcinoma of the mid-oesopha- gus was noted, which was previously only recorded in areas of high prevalence of oesophageal carcinoma. A high frequency of khat chewing and water-pipe smoking (mada’a) was found among both men and women and among a group with tumours of the gastro-oesophageal junction or cardia. This apparent association of khat with car- cinoma of the lower oesophagus might be related to the khat-induced delay of gastric emptying with a subsequent increased risk of gastro-oesophageal reflux and Barrett oesophagus [19]. The effect of chewing khat on the mu- cosal histology of the upper gastrointestinal tract was explored in Yemeni patients pre- senting with dyspepsia [71]. Regular daily khat chewing was not associated with any major effect on the oesophagus or stomach ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما but duodenal ulcers were common in chew- ers. This may have been associated with the high prevalence of smoking in this group. Gastric-type mucosa at the lower end of oesophagus is thought to increase the risk of developing adenocarcinoma 30–125-fold. Although its presence was not related to the intake of khat, its overall prevalence in Yemeni patients was comparatively high (18%). To clarify this point, a case–control study on oesophageal carcinoma in Yemen is planned. The association with mouth cancer has been discussed earlier. The adverse effects of khat chewing on oral–dental tissues were first observed by Laurent [72], Halbach [4] and Luqman and Danowski [3]. They reported that long-term khat chewing caused stomatitis followed by secondary infection. These might be related to mechanical strain on the cheek and other oral tissues as well as chemical irritation of the mucosal surfaces. A high rate of periodontal diseases and low rate of dental caries has been observed among male Yem- eni khat chewers [3]. Mouth dryness, the major complaint following khat chewing, might be due to the sympathomimetic effect of cathinone and/or to excess secretion of saliva during chewing [73]. Hill and Gibson observed the effects on oral and dental tissue among Yemeni males with an average age of 35 years who chewed khat for of 20 years [74]. They found a low prevalence of caries, but universal attrition, temporomandibular joint pain and increased periodontal pocket depth on the khat-chewing side compared with the non-khat chewing side. They also reported increased keratosis on the buccal mucosa in 50% of the cases. In Kenya, Macigo and his colleagues showed that khat chewing was not significantly associated with leukoplakia compared with tobacco and alcohol consumption [75]. Recently, a cross-sectional hospital study among Yemeni khat and non-khat chewers showed an increased risk for a number of oral and paraoral lesions [76]. The study revealed that khat chewing caused many lesions to the supporting structures of the teeth, namely gingivitis, periodontal pocket formation, gingival recession, tooth mo- bility and tooth mortality. Khat chewing caused clicking and pain in the temporo- mandibular joints and led to attrition and staining of teeth and cervical caries, par- ticularly among crystallized sugar consum- ers. Due to continuous mechanical friction and/or the chemical content, khat chewing caused white lesions on the buccal and gingival mucosa. Histopathological study revealed changes to the oral mucosa, such as acanthosis, papillomatosis, ortho- and para-keratosis and intercellular oedema, but not leukoplakia. Concerning saliva and salivary glands, khat chewing results in mouth dryness, en- largement of salivary glands, inflammation and folding of the parotid papilla at the site of khat chewing. Khat chewing also causes obvious facial asymmetry of facial tissues. To conclude, khat chewing appears to pose the following potential health risks. • It may induce disturbance of mood (anx- iety and/or depression, insomnia). • In psychotic patients, it may aggravate thought disturbances (hallucination and delusions), induce aggressive behaviour and create difficulties in treating these patients. • It may cause elevation of arterial blood pressure and pulse rate with subsequent ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما increased cardiovascular risk, particu- larly in hypertensive patients. • It seems to be a common cause of stoma- titis and other problems in the mouth as well as gastro-oesophageal reflux. • It may be associated with increased risk of carcinoma of the mouth and oesophagitis. • It may interfere with absorption of some orally administered antibiotics. • It causes anorexia. • It causes constipation. • It may have a toxic effect on the liver, possibly as a result of pesticides used in khat cultivation. • It is associated with an increased risk of low birth weight infants in khat-chewing pregnant women. Since khat chewing is widespread in Yemen, the following actions are recommended. 1. Increase public awareness of the poten- tial health hazards of khat chewing. 2. Support scientific research on khat in different institutions and universities to explore the different effects of khat on public health. 3. Integrate education about khat into the curricula of primary and secondary schools. 4. Legislate on the use of pesticides in the cultivation of khat in view of their potentially harmful effects on human health. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما WHO Expert Committee on Drug Dependence. Thirty-fourth report: 942Technical Report Series, No This report presents the recommendations of a WHO Expert Commit- tee responsible for reviewing information on dependence-producing drugs to assess the need for their international control. The first part of the report contains a summary of the Committee’s 7 evaluations of substances (dronabinol, oripavine, buprenorphine, butorphanol, ketamine, khat and zopiclone). The second part of the report discusses the guidelines for the WHO review of dependence- producing psychoactive substances for international control. It in- cludes sections on amending the current guidelines, interpretation of specific aspects of the guidelines and access to information necessary for the evaluation of substances. The final section considers other matters including activities of the EMCCDA, the use of pharmacovigi- lance data, promotion of education and information on the appropriate use of psychoactive drugs and the impact of international control on medical availability of substances. Further information about this and other WHO publications is available at: http://www.who.int/publications/en/ ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 1994 ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Tobacco use has been identified as one of the major avoidable causes of morbidity and mortality in society [1,2]. Tobacco use is a known cause of cancer, heart disease, stroke and chronic obstructive pulmonary disease [2]. In Bahrain, despite decades of health promotion and tobacco control legislation, tobacco use is still prevalent among adults and teenagers. The national family health survey in Bahrain in 1995 reported that the overall prevalence of smoking among Bahraini male adults was 22.8% and among adult females was 8.8% [3]. The latest global health professional survey in 2004 showed that 23.8% of Bahraini male physicians were cigarette smokers, while only 0.9% of Bahraini female physicians were smokers [4]. This is lower than earlier studies, in which 26.6% of male physicians and 6.9% of female physicians were smokers [5,6]. Moreover, the prevalence of smoking among youth appears to be rising [7]. Smok- ing among high-school males has risen from 14.8% in the early 1980s to 25.8% in the late 1990s, although the reasons for this increase are unclear [8,9]. In a more recent survey, 33.5% of young males and 11.9% of young females reported being smokers [10]. Smoking is prevalent among medical stu- dents, with 27.5% of male medical students and 2.3% of female students reporting being smokers [11]. The recent trend towards waterpipe smoking is also a source of concern. Anec- dotal evidence suggests a rise in the preva- lence of this type of tobacco use in Bahrain, particularly among women and teenagers. The prevalence of waterpipe smoking was 9.7% among Bahraini male physicians and 3.1% among Bahraini female physicians [4]. The aim of this report is to describe the tobacco control approach in Bahrain, focusing on tobacco legislation, youth and waterpipe smoking, tobacco surveillance, and smoking cessation, exploring the main limitations and suggesting action for further improvement. The history of tobacco control in Bahrain dates back to 1978, when a number of control measures were implemented for the first time, including raising custom duties on cigarettes to 70%, regulating the permis- sible level of tar and nicotine per cigarette and restricting cigarette advertising. The following year (1979) a nongovernmental organization (NGO), the Bahrain Anti- smoking Society, was formed [12]. The society works in collaboration with the Bah- rain government to advocate for tobacco control through prevention activities and lobbying for law review and law enforce- ment. In 1994 a tobacco control decree was issued by the Emir of Bahrain to reinforce the antismoking measures [13]. It stated the following: • No tobacco to be cultivated in Bahrain. • Smoking to be banned in closed public transport and aeroplane flights. • Smoking to be banned in closed public places, government institutions, hos- pitals, educational institutions and air- ports. • No sponsorship by tobacco companies to be allowed for any kind of sports or competitions. • No tobacco factories to be built. • No vending machines for cigarettes to be allowed. • No sale of tobacco to under 18-year- olds. The law also called for the formation of a committee to include members represent- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ing different parties involved in the tobacco control plan. The National Tobacco Control Committee was established in 1995. The committee was headed by the Minister of Health and members of various ministries involved in tobacco control plus NGO rep- resentatives. The committee was assigned various responsibilities, such as identifying the upper limit for nicotine, tar and other toxic substances in cigarettes, limiting ad- vertisements and putting warning signs on advertisements. Despite having been inactive from 1997 until 2005, the commit- tee finally resumed its operation with new members. It is worth noting that the tobacco control approach in Bahrain incorporates the Gulf Cooperation Council, World Health Or- ganization and other internationally recom- mended strategies of legislation and health promotion. In general, primary prevention activities dominated the tobacco control plan in Bahrain. Activities such as media advocacy, school education and community awareness are dominant. The implementa- tion plan is multisectoral, involving the Ministry of Health as the leader, plus other members (Ministry of Education, Minis- try of Information, municipal government, NGOs, private companies, etc.). In terms of tobacco research, Bahrain has participated in two global tobacco sur- veys, the Global Youth Tobacco Survey in 2003 and the Global Health Professional Tobacco Survey in 2002 [10,14]. In addi- tion, a few local studies [5,6,8,9,11] and a national survey [3,15] have been carried out to assess the prevalence of smoking in Bahrain. Bahrain participates at regional and in- ternational level conferences and meetings concerned with tobacco control and lately at the negotiations for the WHO Framework Convention on Tobacco Control (FCTC) [16]. However, as Bahrain hosts the For- mula One motor racing sport, which is sponsored by the tobacco industry, ratifica- tion the FCTC has been delayed. The text is currently being submitted to a special ministerial committee for consideration and approval [12]. Smoking uptake in youth is of particular concern for tobacco control programmers [10,14]. Data show that around three-quar- ters of ever-smokers, including current smokers and ex-smokers, reported that they started to smoke regularly between the ages of 15 and 18 years. A further 1 in 5 of these said they started to smoke before age 15 years [14]. Intervention directed at youth smoking in Bahrain is focused on risk awareness and education on the negative consequences of smoking. A number of school-based programmes have been implemented in recent decades, plus mass media messages and other community-focused programmes [17]. The latest survey among youth in Bahrain indicated that the prevalence of smoking is 33.5% among male high-school students and 11.9% among females (Table 1) [10]. The mean age at which smoking began has been reported to be 16.8 (SD 1.1) years [8]. The survey showed that over half (52.4%) of under-18-year-olds purchase their cigarettes from local stores, and 75.3% were not stopped from purchasing because of their age. With regards to mass media campaigns, it appears that 7 in 10 teenagers saw pro-tobacco messages in the media [7]. Based on the above data, it is evident that there has been a failure of tobacco legislation and there is a need to review the current strategies and legislative ac- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما tion focused on youth smoking, and devise with new interventions. Evidence elsewhere shows that legislative strategies and penal- ties are important for reducing youth smok- ing [18]. This can discourage both uptake and continuation of tobacco use. Smoking tobacco in a waterpipe (shisha, nargile, hookah) seems to be a rising trend among both sexes and teenagers and is a major concern to the public health system [10,19,20]. Anecdotal evidence in Bahrain indicates that waterpipe smoking has in- creased in popularity in Bahrain recently, particularly among women and teenagers. In contrast to cigarette smoking, waterpipe smoking has become an accepted social behaviour and has gained more public toler- ance [4]. The majority of smokers believe that waterpipe smoking is not damaging to health and many cigarette smokers switch to the waterpipe while they attempt to quit [21–23]. There are few studies concerned with waterpipe smoking and its patterns and trends in Bahrain. Data from the national morbidity survey showed that waterpipe smoking is more prevalent among females (8.9%) compared with males (2.0%) (Ta- ble 2) [24]. The Global Youth Tobacco Survey showed that among young people the prevalence of tobacco consumption in the category “other tobacco products” was 18.2% [10]; this is likely to comprise mostly waterpipe use. Bahrain does not grow tobacco or manu- facture cigarettes, and thus all cigarettes available in the market are imported, almost exclusively from Europe and North Amer- ica. There are four main types of tobacco available in Bahrain (cigarette, waterpipe, cigar and pipe) and two types of tobacco are used in waterpipes (narjeela and kadu). Females smoke narjeela compared with males who commonly use kadu. The local ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما tobacco used in waterpipe smoking (mua- sal) is flavoured and processed in Bahrain. The popularity of Bahrain muasal is in- creasing in the local and regional market [19,25]. For cigarettes in Bahrain, by law the maximum permissible limits of nicotine, tar, and carbon monoxide are 0.6 mg, 10 mg and 12 mg per cigarette respectively [26]. However, the main challenge here is the absence of local resources to assess the required specification for the new type of imported cigarettes and tobacco, making it difficult to enforce the law in this area. Moreover, random checking on the ciga- rettes available in the local market is not in operation. Reviewing the available data [27] and Ministry of Health reports [28] shows that periodic data on consumption, import and export of tobacco and on smoking locations are missing. Such information is critical for monitoring and evaluating the control plan. Thus, one of the first initiatives is to main- tain an effective tobacco surveillance and reporting system with other ministries, such as the Ministry of Commerce, the Ministry of Education and other licensing bodies, such as the local municipality. Despite the Emiri decree for tobacco con- trol in 1994, law enforcement is weak in general, and special enforcement activities, such as conducting a regular retail compli- ance check or restrictions on advertising for tobacco and frequent checks on tobacco specifications, are not in operation. Al- though the 1994 law stated that Ministry of Health inspectors can issue an injunction for law breakers, since then, apart from three warning letters issued by the Ministry of Health, no injunctions have been issued. Publicity in the media suggests that pub- lic pressure is growing against the harmful effects of second-hand cigarette smoke and the smoking of waterpipes in closed public places [19,24]. On the other hand, personnel shortages at the Ministry of Health have delayed action and the enforcement of the law in most cases. In addition, the current legislative charges imposed on companies who break the law are hardly significant in terms of business profit for these activities. Interventions targeted at individual smokers are only part of the much broader spectrum of strategies to reduce the prevalence of smoking. A wide variety of smoking ces- sation programmes have been developed and studied over recent years. The two main interventions are pharmacotherapy and behavioural therapy. These contribute about equally to success in quitting. Most evidence indicates that a combination of both methods is the most effective interven- tion [29]. In Bahrain, an anti-smoking clinic was set up by the Bahrain Anti-smoking Soci- ety. However, as it operated on a voluntary basis it has been unable to maintain its serv- ices over time. In 2004 a smoking cessation clinic was set up by the Ministry of Health, but the clinic operates on a part-time basis on limited resources. Nicotine replacement therapy and bupropion hydrochloride tablets are avail- able in the local market; however, their use is very limited because they are expensive and the price is not subsidized. Recently an attempt has been made to formulate national guidelines for smoking cessation [30,31]. Building local capacity with regards to smoking cessation interven- tions is still a big challenge. Affordability ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما and acceptability need to be addressed. The new tobacco control plan suggested by the Chronic Diseases Control Section of the Bahrain Ministry of Health recommended widening the smoking cessation services and integrating them into primary care serv- ices [32]. Training of health care staff on measures for dealing with smokers is being given priority. Recently a number of workshops organized by the Ministry of Health and NGOs focused on national capacity- building in tobacco control [Ministry of Health, Bahrain, Quit smoking workshop, unpublished, 2005]. Ongoing surveillance will document any positive future changes. In Bahrain, to- bacco use surveillance has been monitored through population surveys and tobacco- specific surveys; these include: • Global Youth Tobacco Survey, 2003 [14]; • Global health professionals survey [4]; • Census data and health statistics; • National family health survey, 1995 [3]; • National morbidity survey of 4.5% of the households in Bahrain [24]. Individual research has been carried out on different population groups (school- based youth tobacco surveys, surveys of adults, school administrators, teachers, opinion leaders and health care providers, etc.) [5,6,8,10,11,33]. Inconsistent methods and irregular sur- vey intervals are two of the issues that have made it difficult to know exactly how smok- ing patterns are changing in Bahrain. Furthermore, two significant challenges were noted. First, liaison with other minis- tries is weak regarding the tobacco market (import, export) and specific data on the amount of tobacco consumed, the type of tobacco permitted, the number of permitted smoking places and other tobacco-related activities. Thus, maintaining collaboration and proper reporting channels with other parties involved in tobacco control are be- ing addressed. A special reporting format is been created and will be processed soon. This would allow for a proper surveillance system to be put in place. Second, baseline data on patterns and trends in waterpipe smoking and second-hand smoke are insuf- ficient. Giving the importance of the above data for monitoring the tobacco control interven- tions, further research is required. Tobacco use still constitutes a public health problem in Bahrain and imposes a huge burden with its associated morbidity and mortality. Although tobacco legislation plays a major role in tobacco control policy, in Bahrain there exist two main limitations facing tobacco legislation: first, the failure to enforce the law in the majority of cases and a lack of qualified manpower to do this; second, the delay in ratification of the FCTC. Upward trends in youth smoking preva- lence highlight the need to diversify the approaches within the tobacco control plan. New legislation and enforcement of existing legislation take priority, and there is a need to reduce availability of tobacco products to young people and regulate supply. In addition, teenagers need an alternative to occupy their free time, particularly at school breaks; therefore it is highly recommended to provide special programmes focused on youth during the breaktime. The implemen- ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما tation of such programmes would minimize the peer pressure effect on tobacco intake. The smoking cessation plan needs to be strengthened within the context of primary health care. However, implementation of a cessation plan requires adequate staff train- ing and more resources, such as medication to support smokers who want to quit. It is essential as well to equip medical students and other health professional students with the skills to help smokers to quit smoking. Implementation of a tobacco control plan is multisectoral, involving government and NGOs. Hence, it is recommended to maintain collaboration and surveillance through periodic reporting systems, with the Ministry of Health taking the leadership role. Data collated from various imple- mentation agencies should be analysed and disseminated for policy use. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Setting appropriate priorities for healthcare research should be an integral component of planning in developed and developing countries alike. However, other priorities all too often take precedence over this, es- pecially in developing countries. Coupled with this, the generally lower levels of fund- ing for healthcare research in developing countries may necessitate a more creative use of existing resources, including as wide a range of collaborative coalitions as pos- sible. This paper makes recommendations for knowledge generation and knowledge application partnerships between develop- ing countries in the Eastern Mediterranean Region (EMR), research organizations else- where and, in particular, The Cochrane Col- laboration. The development of healthcare research capacity in the EMR is constrained by fac- tors that have been clearly articulated in the past: “inadequate political commitment; an unfavourable research environment; lack of leadership, and weak management and coordination of research; near absence of linkages and networking among scientists; poorly developed research capacity and inadequate resources” [1]. A recent analysis by the World Health Organization (WHO) of 5 developing coun- tries in the EMR confirmed the fragmented and poorly coordinated status of their na- tional health research systems and the wide variation in research priority setting among them [2]. The final report recommended that “the development of national health re- search systems will need to look beyond the Ministry of Health and/or the Department of Medical Research or Medical Research Council in terms of priorities or an agenda for action”. A range of models has been designed for setting priorities for the allocation of health resources and the conduct of research. These have incorporated qualitative and quantitative methods, quantitative formula- tions, and prioritization matrices but not all have found wide acceptance in developing countries [3,4]. Decisions on priorities need to take ac- count of the best available information, including an evidence-based situation analysis or context analysis, to help inform the process. Information needs to be gath- ered on population health status or burden of disease and contextual data regarding the healthcare system and health research system. If countries within the EMR wish to develop research capacity, build sustainable institu- tions and identify solutions to key national health problems, the research conducted must be grounded on robust scientific data on population morbidity and mortality, as well as the crucial evidence of what works, for whom, and why. Moreover, if the devel- opment of indigenous capacity for research in developing countries within the EMR ran in tandem with policies ensuring that research funding is allocated to addressing ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما local priorities (specifically the most impor- tant determinants of health), this might en- sure that the limited expertise within these countries and the new expertise that will be created, is used to maximum benefit lo- cally, In this way, priority setting would not only crystallize the immediate health goals but would also serve to empower, energize, retain and attract researchers. The Ministerial Summit on Health Research in Mexico City in 2004, convened by the WHO, called for action by national gov- ernments to, “promote access to reliable, relevant, and up-to-date evidence on the effects of interventions, based on systematic reviews of the totality of available research findings” [5]. The Cochrane Collaboration is well placed to work with national governments to achieve this goal. The Collaboration is an international organization dedicated to improving health care for the world’s population by preparing, maintaining and promoting the accessibility of systematic reviews of the effects of healthcare inter- ventions. These high quality Cochrane re- views are available online in The Cochrane Library, and are being used increasingly by policy-makers in the developing world not only to make decisions about service provi- sion but also to help set research priorities. Systematic reviews allow one to exam- ine what research has been done, where, and to what effect. They can underpin deci- sion-making about healthcare interventions and research because they summarize large amounts of information, identify beneficial, harmful or unproven interventions, and highlight gaps in research. Recent initiatives within The Cochrane Collaboration are of increasing relevance to developing countries; for example the Developing Country Network and the Co- chrane Health Equity Field & Campbell Equity Methods Group whose mandate is to ensure that reviews relevant to lower- and middle-income countries, including EMR countries, apply an “equity lens” that ensures the results relate to the disadvan- taged. These entities are developing further methods of examining equity in evidence on the effects of interventions, in addition to ways of integrating differing methods, thereby answering the “for whom and why” elements of what works [6]. This builds on the Collaboration’s commitment to engage with the generators and users of research in developing countries, and now is an ideal time to combine this work with the establishment of a priority-driven approach to health research in the EMR. Therefore, given the limited resources currently available to prepare and maintain systematic reviews, there is an urgent need to identify the priorities for this form of research. These reviews will clarify the benefits and harms of interventions studied in randomized trials, and identify the gaps and priorities for further trials in the EMR. We believe that priorities for future health research should be based on the most important gaps in current knowledge in the relevant part of the world. We propose that the following approach be integrated into regional planning. 1. Analysis of the relevant health burden, using the burden of disease approach, in the context of equity and health in- equalities. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما 2. Systematic reviews of the evidence on the effects of interventions, relevant to the health outcomes sought in the region, with special attention to the dis- advantaged. 3. Consultation on a process of identifying the matches and gaps, with the aim of undertaking a priority setting exercise to: • identify priority topics for Cochrane systematic reviews • examine processes within the agency and among agencies for funding rel- evant new research. 4. Undertaking of a process of knowledge translation for dissemination to, and en- gagement with, users and key stakehold- ers, with a strategy to update this at least every 5 years. We contend that this approach has the po- tential to make a meaningful difference to people in the EMR, whilst allowing the countries of this Region to develop and retain the capacity to set their own research priorities. We are grateful to Professor Mike Clarke, Director of the UK Cochrane Centre, for his comments on an earlier version of this manuscript. The views expressed in this paper represent those of the authors and are not necessarily the views or the official policy of The Cochrane Collaboration. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Gram-negative sepsis is a life-threatening disease especially in an already debilitated patient. The presence of Enterobacteriaceae or any Gram-negative bacteria in the blood- stream may lead to endotoxic shock, sys- temic disease and multiple organ failure. The pseudomonads are among the most common organisms causing nosoco- mial infections [1]. Although Pseudomonas aeruginosa is by far the most commonly isolated species, several other species are occasionally isolated from clinical human specimens as opportunistic pathogens, such as P. stutzeri [2]. P. stutzeri was recently isolated in a newborn pre-term baby in our hospital. This is the first time this organism has been isolated in our laboratory over the past few years. The organism was almost discarded as a plate contaminant, but for its unusual features. It is being reported here in order to alert microbiologists to be vigilant concerning such unusual organisms so that they are not missed and that prompt and appropriate antibiotic therapy can be instituted. T.B. was a male twin delivered in our hospi- tal by emergency caesarean section on 16th December, 2004. The mother was 34 weeks pregnant and had had pre-eclampsia and premature rupture of the membranes for 8 hours before she underwent caesarean sec- tion. The babies’ presentation was breach. At delivery, baby T.B. weighed 1.8 kg. On examination, he was pink and cried nor- mally. His Apgar scores were 6/1 and 8/5. He had mild tachypnea. His heart sounds I and II were normal and there were no murmurs. No abnormality was found in his chest, his abdomen was soft, not distended, and there was no organomegaly. His hip joints were normal. The second twin, a female, was normal. Because of the baby’s tachypnea, neonatal sepsis was suspected. The following investigations were car- ried out: complete blood count (CBC), chest X-ray, erythrocyte sedimentation rate, blood and urine culture, electrolytes, C-reactive protein, TORCH (Toxoplasma gondii, other microorganisms, rubella virus, cytomegalovirus and herpes simplex virus) and occult blood. The baby was then started on intravenous ampicillin 45 mg every 12 hours (which was increased to 90 mg every 12 hours after 48 hours) and intravenous amikacin 13.5 mg every 12 hours. Blood culture was performed in the BACTEC 9240 blood culture system. After 48 hours, the machine indicated a growth, which on Gram stain showed Gram- negative bacilli. A subculture was done as well as direct sensitivity testing using the ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما disc-diffusion technique. The organism was sensitive to all routine antibiotics for Enterobacteriaceae. It grew well on blood agar, Mueller–Hinton agar and on Mac- Conkey agar as a non-lactose fermenter. It was motile and oxidase positive. It grew at 42 °C but not at 4 °C. The colonies were dry, wrinkled and brownish in appearance. If not for the good growth on MacConkey agar, it could pass for a contaminant, such as aerobic spore-bearers. When attempts were made to make new smears and suspen- sion for biochemical tests, removing the colonies from the plates proved difficult. This aroused further suspicion. After much difficulty, a few colonies were removed for Gram-stain, oxidase test and biochemical tests with the API system (bioMérieux, France). The next day, API reactions were all negative. It was thought that this might be due to insufficient organisms in the suspension used. A repeat API was done and the result came out the same as before. Meanwhile, a preliminary report was sent to the nursery as Gram-negative bacilli, possibly Pseu- domonas species, sensitive to both amikacin and ampicillin, which the baby was being treated with. After 3 unsuccessful attempts at identification using the API system, the isolate was sent to the Aseer Central Hos- pital for identification using the Microscan system (Dade Behring, West Sacramento, USA). After 48 hours the organism was identified as P. stutzeri, sensitive to a large number of antibiotics. The results of the other investigations carried out were as follows: blood urea ni- trogen 3 mg/dL (low) (normal 6–20); sodi- um 135.3 mmol/L (low) (normal 136–145); total bilirubin 9.38 mg/dL (high) (normal 0.00–1.00); direct bilirubin 0.39 mg/dL (high) (normal 0.00–0.30). The results of the remaining tests were normal. The baby was continued on amikacin plus ampicillin since the organism was sensitive to both agents. He made rapid progress and the results of the repeat tests were all normal within 2 weeks. The an- tibiotics were discontinued after 10 days of therapy. Repeat blood cultures were carried out on days 13 and 15 and were both negative. The baby was discharged home 1 week thereafter. Before discharge, the electrolytes became normal, haemoglobin was 13.4 g/dL, white blood cell count was 9.1 × 103/ L and platelets 411 × 103/ L. P. stutzeri group of pseudomonads consists of 3 organisms: P. stutzeri, CDC group Vb- 3 and P. mendocina [3]. Among this group, P. stutzeri is the most frequently encoun- tered in clinical specimens. A review of the literature shows that P. stutzeri is most frequently isolated from blood, wounds, the respiratory tract and urine [4]. P. stutzeri has distinctive features. On sheep blood agar, strains appear buff to brown in colour with dry, wrinkled, tough and adherent col- onies. The adherence of P. stutzeri makes the removal of colonies from agar medium difficult. Because of the difficulty in mak- ing suspensions of specific turbidity, iden- tification and susceptibility systems do not work well with this organism [3]. However, P. stutzeri may be identified using the fol- lowing criteria: presence of oxidase, ability to oxidize but not ferment dextrose, growth on MacConkey agar and demonstration of motility by polar flagella [5]. P. stutzeri is a ubiquitous organism found in soil, water and the hospital environments [5]. It is a saprophytic microorganism that rarely causes severe infections [6]. Isolation of P. stutzeri indicates contamination or ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما colonization in hospitalized patients. How- ever, patients with P. stutzeri infections have generally been found to have serious underlying diseases or predisposing risk factors such as haemodialysis [1], menin- gomyelocele [6], human immunodeficiency virus infection [7] and chronic obstructive pulmonary disease [8]. All isolates to date have responded well to a wide range of an- tibiotics including the aminoglycosides, the antipseudomonal penicillins, trimethoprim– sulfamethoxazole and the third generation cephalosporins [4,5]. The isolate from our patient was sensitive to all of these drugs. Whenever a laboratory encounters an unusual organism such as this, every effort should be made to identify it, including sending it to the nearest reference centre. Besides, every clinical isolate must be cor- related with the patient’s symptoms before arriving at a decision. Samples were taken from the baby’s intravenous fluid and line, suction machine, cot, surrounding floor and walls, the sink, and attending physicians and nurses but none of these grew P. stutzeri. The mother however was not sampled due to cultural practices in our area. However, the twin sis- ter who was also admitted into the nursery at about the same time for observation was free of the organism. Although we were unable to identify the exact source of the organism, the probability of the organism coming from the mother’s vagina cannot be excluded, with acquisition by the baby dur- ing parturition. This view is supported by an earlier report from this hospital by Asindi et al. [9] on neonatal sepsis in pre-labour rupture of membranes. At the same time, given that P. stutzeri is an environmental organism, attending healthcare workers should be reminded to observe strictly infection control guidelines, especially scrupulous washing of hands. I am grateful to the infection control nurse, Ms Nilda Acuba, and the staff of the nurs- ery unit and the microbiology laboratory at Abha General Hospital for their help and cooperation in the preparation of this report. I am also grateful to Professor A.O. Osoba of the National Guard Hospital, Jeddah for reviewing the manuscript. ٢٠٠٧ ،٣ ددعلا ،رشع ثلاثلا دلجلما ،ةيلماعلا ةحصلا ةمظنم ،طسوتلما قرشل ةيحصلا ةلجلما Infant and meonatal health and development 28 A healthy start in life is important to every newborn baby. The first days, the neonatal period, is critical. It is during this time that funda- mental health and feeding practices are established. It is also during this time that the child is at highest risk for death. One of the aims of the Infant and Neonatal Health (HNI) team within the WHO Department of Child and Adolescent Health and Develop- ment (CAH) is to reduce infant mortality and to ensure that newborns and infants have the opportunity for a healthy start in life. Further information about the work of the team can be found at: http://www. who.int/child-adolescent-health/OVERVIEW/HNI/neonatal.htm
World Health Organization (WHO) · Journal articles
Eastern Mediterranean Health Journal [2007; Vol.13, Issue 3]
View original document
The full text is hosted by the publishing organisation. lawenc.com indexes the metadata and links to the official source.
Full text
Key facts
Organisation
World Health Organization (WHO)
Document type
Journal articles
Source
World Health Organization