A 3-year follow-up of hypertension in Delhi.

Research / Recherche A 3-year follow-up of hypertension in Delhi* N. Gopinath,' S.L. Chadha,2 S. Shekhawat,3 & R. Tandon4 A follow-up study of hypertension was carried out among adults in Delhi 3 years after an initial community- based epidemiological survey of the same population. The treatment and the severity status of 1115 out of 1749 individuals with hypertension detected in the initial survey were compared with those observed in the follow-up. The proportion of treated cases with controlled blood pressure rose from 10.8% to 60.8%. Among the cohort of 3611 subjects aged 25-64 years who were normotensive in the initial sur- vey, 132 new cases of hypertension, were detected. The annual incidence of hypertension was the same in men and women (12.2 per 1000). Diabetes and regular alcohol consumption were significant risk factors for hypertension, being present in 13 and 7 cases, respectively. Electrocardiograms (ECGs) were recorded for 871 of the 1115 cases of hypertension. Abnormal ECGs were exhibited by 307 cases (35.2%), of which 24 (2.7%) had had myocardial infarction, 133 (15.3%) had ischaemic ST-T changes, 54 (6.2%) had left ventricular hypertrophy, and 96 (11.0%) had conduction defects and arrhythmias. Introduction Over the period 1985-87 a community-based epi- demiological study on coronary heart disease and hypertension was carried out in Delhi among 25-64- year-olds using a randomized house-to-house survey. Electrocardiograms (ECGs) were obtained for clini- cally detected cases of hypertension, and normoten- sive adults were selected as controls from every second household. The methodology and results of this study have been reported elsewhere (1). We carried out a follow-up of the same popula- tion 3 years after the initial investigation. Our find- ings on coronary heart disease have already been published (2). The present article concems the re- evaluation of normotensive adults and patients with hypertension. Materials and methods The same cohort of adults that we studied in 1985-87 was re-examined after an average interval From: Sitaram Bhartia Institute of Science and Research, B-16 Mehrauli Institutional Area, New Delhi - 110 016, India. 1 Director. 2 Consultant in Community Health. Requests for reprints should be sent to this author. 3 Biostatistician. 4Consultant Cardiologist. Reprint No. 5519 of 3 years. This community-based follow-up study had the following objectives: - to evaluate the progress of hypertensive patients detected in the initial epidemiological study; and - to identify new cases of hypertension in adults who had been normotensive in the initial survey, and hence to determine the incidence of hyper- tension in the study population. Mild hypertension was defined as a diastolic blood pressure of 90-104 mmHg (12-13.9 kPa), moderately severe hypertension as 105-114mmHg (13.8-15.2 kPa), and severe hypertension as >1 15 mmHg (.15.3 kPa). Isolated systolic hyperten- sion was defined as systolic blood pressure >160 mmHg (>21.3 kPa) and diastolic blood pres- sure <90 mmHg (<12.0 kPa). The blood pressures of those who exhibited hypertension were measured twice, with 10-minutes' complete rest between the measurements. The average of the two readings was then calculated. General information on the study subjects' age, sex, socioeconomic status, marital status, occupation, physical activity, height, weight, dietary habits includ- ing salt intake, smoking, and alcohol consumption had already been obtained during the initial epidemi- ological investigation, and any changes in these fac- tors were noted during the follow-up study on the same forms that had been used previously. Suitable forms were designed for recording other data. The medical and family histories of hypertension and Bulletin of the World Health Organization, 1994, 72 (5): 715-720 © World Health Organization 1994 715 N. Gopinath et al. findings of detailed clinical examinations were re- corded. Efforts were made to obtain 12-lead ECGs from known hypertension cases, from newly detected cases, and from a subsample of individuals who had been normotensive in the initial survey. All ECGs were interpreted using the Minnesota codes (3). The cluster sampling method and the criteria for diagnos- ing hypertension were the same as those employed in the initial survey (1). Obesity was expressed as body mass index (weight in kg/(height)2 in m2), values >25 being taken as the cut-off point. Diabetes was assessed on the basis of clinical history supported by documentary evidence of drug and dietary therapy. Results The age and sex distributions of the resurveyed sub- jects are shown in Table 1. Of the 1749 cases of hypertension detected in the initial epidemiological survey, 1115 were available for re-examination. A total of 309 cases had left Delhi and were not trace- able, while 74 had died. The remaining 251 cases were not available, despite repeated efforts to contact them. Severity status The severity of hypertension in the 1115 patients re- examined is shown in Table 2. The proportion of cases with controlled blood pressure rose from 10.8% in the initial survey to 60.8% in the follow- up. Of the 732 subjects previously categorized as having mild hypertension, 460 were found to be nor- motensive upon follow-up, while 231, 34, and 7 res- pectively, had mild, moderate, and severe hyperten- sion. Similarly, of the 142 subjects with moderate hypertension in the initial survey, 68 were control- led, whereas 7, 17 and 50, respectively, had mild, moderate, and severe hypertension at follow-up. Of the 58 individuals who had had severe hypertension, 21 were controlled, and 20, 8, and 9, respectively, had mild, moderate, and severe hypertension in the follow-up examination. Of the 62 cases of isolated systolic hypertension recorded in the initial study, 11 had developed diastolic hypertension (mild in 10 cases and moderate in 1 case); 37 had become nor- motensive; and for the remaining 14 there was no change. Of the 678 cases with controlled hyperten- sion in the follow-up study, 116 (108 with mild hypertension and 8 with isolated systolic hyperten- sion) had not received any antihypertensive drug therapy but had followed nonpharmacological meas- ures for the control of hypertension. Treatment The treatment and control status of the 11 15 subjects with hypertension who were re-examined is shown in Table 3. The control status improved in each cate- gory of severity of hypertension. The proportion who were treated but had uncontrolled hypertension fell from 35% to 24.7%, whereas the proportion who were treated and controlled rose from 10.8% to 60.8%. The proportion of subjects with untreated and uncontrolled hypertension fell from 54.2% to 14.5%. Complications There was a history of angina pectoris in 71 of the subjects with hypertension who were re-examined, myocardial infarction had occurred in 16, and stroke had affected 13 individuals. The severity of hyper- tension in the subjects who developed complications is shown in Table 4. ECG findings for the re-examined subjects with hypertension In the initial epidemiological survey, 1417 of the 1749 detected cases of hypertension had ECGs recorded, and of these 1011 gave normal ECG traces. Of the 406 cases with abnormal ECGs, left ventricular hypertrophy was present in 80; ST-T changes (Minnesota codes 4.1.2, 5.1, and 5.2) occur- red in 117; left ventricular hypertrophy (Minnesota codes 3.1 and 3.3) and ischaemic ST-T changes were present in 40; changes associated with myocar- dial infarction in 59; and conduction defects (various Table 1: Distribution of the hypertensive and normotensive study subjects, by age group and sex No. with hypertension No. with normotensiona TotalAge group (years) Males Females Total Males Females Total Males Females Total 25-34 56 55 111 449 952 1 401 505 1 007 1 512 35-44 94 152 246 338 667 1 005 432 819 1 251 45-54 162 220 382 279 428 707 441 648 1 089 55-64 146 230 376 242 256 498 388 486 874 Total 458 657 1 115 1 308 2 303 3 611 1 766 2 960 4 726 a No. who were normotensive at time of initial epidemiological survey. WHO Bulletin OMS. Vol 72 1994716 A 3-year follow-up of hypertension in Delhi Table 2: Comparison of the severity status of the cases of hypertension among the study subjects In the initial sur- vey and upon re-evaluation Initial severity: Mild Moderate Severe Controlled ISHa Total Severityy on re-evaluation No. 732 (65.7)b 142 (12.7) 58 (5.2) 121 (10.8) 62 (5.6) 1 115 Mild 231 (31.6) 7 (4.9) 20 (34.5) 25 (20.7) 10 (16.1) 293 (26.3) Moderate 34 (4.6) 17 (12.0) 8 (13.8) 4 (3.3) 1 (1.6) 64 (5.7) Severe 7 (1.0) 50 (35.2) 9 (15.5) 66 (5.9) - - Controlled 460 (62.8) 68 (47.9) 21 (36.2) 92 (76.0) 37 (59.7 678c (60.8) ISHa _ - _ - 14 (22.6) 14 (1.3) a Isolated systolic hypertension. b Figures in parentheses are percentages. c Includes 116 cases not taking antihypertensive drug therapy. types of bundle branch block) and arrhythmias (tachycardia, bradycardia, ventricular premature beats, and atrial fibrillation) in 110 cases. Of the 117 cases with ST-T changes in the initial survey, 72 were re-examined: myocardial infarction had occur- red in 8 cases and angina pectoris in 41. Of the 80 cases with left ventricular hypertrophy, 49 were re- examined: two had had a myocardial infarction and 16 had developed angina pectoris. Of the 40 with left ventricular hypertrophy and ischaemic changes, 28 were re-examined: six had had a myocardial infarc- tion and 14 had developed angina pectoris. Thus of 237 subjects with abnormal ECGs in the initial sur- vey, 149 were re-evaluated and 16 and 71, respec- tively, had developed the clinical manifestations of myocardial infarction and angina pectoris during the 3-year interval between the studies. ECGs were obtained for 871 of the cases of hypertension who were re-examined. The ECGs were abnormal in 307 cases (35.2%), with evidence of myocardial infarction in 24 (2.7%), ST-T changes in 133 (15.3%), left ventricular hypertrophy in 54 (6.2%), and conduction defects and arrhythmias in 96 cases (1 1.0%). Causes of death The responses of relatives to questions about the causes of death among subjects were not satisfactory. On occasions, information given by different family members was conflicting and unreliable. Nor were death certificates always helpful in this regard. However, for 35 of the 74 deaths, the documentary evidence and the statements of relatives revealed that 24 had died of a heart attack, 7 had had a stroke, and 4 had died of renal failure. The exact causes of death for the remaining 39 subjects could not be establi- shed. Data collected during the initial survey showed that, of the 74 persons who died, 35, 17, 12, and 3 respectively, had had mild, moderate, severe, and isolated systolic hypertension. The hypertension of the remaining seven individuals who died had been controlled. ECGs had been obtained for 63 of these 74 patients during the initial survey. A total of 40 had exhibited abnormal findings: 8 had typical myo- cardial infarction patterns, 12 had ST-T changes; 4 had left ventricular hypertrophy; 2 had left ventricu- lar hypertrophy with strain; and 14 had conduction defects and arrhythmias. Table 3: Comparison of the treatment and control status of 1115 cases of hypertension examined in the initial and follow-up studies Initial survey Follow-up survey Males Females Total Males Females Total Treated and 144 (31.4)a 246 (37.4) 390 (35.0) 115 (25.1) 160 (24.3) 275 (24.7) uncontrolled Treated and 55 (12.0) 66 (10.0) 121 (10.8) 276 (60.3) 402 (61.2) 678b(60.8) controlled Untreated and 259 (56.6) 345 (52.6) 604 (54.2) 67 (14.6) 95 (14.5) 162 (14.5) uncontrolled Total 458 657 1 115 458 657 1 115 a Figures in parentheses are percentages. b Includes 116 who did not use antihypertensive drug therapy. WHO Bulletin OMS. Vol 72 1994 717 N. Gopinath et al. Table 4: Distribution of complications at follow-up among patients found to have hypertension In the initial survey At initial survey: No. with complications at follow-up: Severity status No. of cases Angina pectoris Myocardial infarction Stroke Mild 732 41 (5.6)a 5 (0.7) 2 (0.3) Moderate 142 9 (6.3) 4 (2.8) 5 (3.5) Severe 58 3 (5.2) 3 (5.2) 6 (10.3) ISHb 62 11 (17.7) 4 (6.4) - Controlled blood pressure 121 7 (5.8) - - Total 1 115 71 (6.4) 16 (1.4) 13 (1.2) a Figures in parentheses are percentages. b Isolated systolic hypertension. Incidence of hypertension On re-examination of the cohort of 3611 adults (1308 males, 2303 females) who were normotensive in the initial survey, 132 new cases (48 males, 84 females) of hypertension were detected. The overall incidence was thus 12.2 per 1000 adults per annum, and this value held for both sexes. The incidences by age and sex are shown in Table 5. Of these new cases, 90 were mild, 10 were moderate, 10 were of isolated systolic hypertension, and 22 were control- led using pharmacological measures. Risk factors for hypertension The prevalences of risk factors for the newly identi- fied cases of hypertension and the controls are given in Table 6. A total of 190 cases of coronary heart disease were excluded from the control groups because of the existence of common risk factors. There were statistically significant differences be- tween those with hypertension and the controls for the following major factors: diabetes mellitus (9.8% of those with hypertension, 1.0% of controls); and regular alcohol consumption (5.3% of those with hypertension, 1.2% of controls). Regular alcohol consumption was defined as at least two drinks per day, 5-6 days per week; casual drinkers were not included. Of individuals who had hypertension, 24.2% were obese, compared with 22.1% of the control group. Smoking was not identified as a risk factor. Discussion To the best of our knowledge the present study is the first community-based follow-up investigation of hypertension carried out in India. Our findings pro- vide information on the incidence of hypertension, the pattem of severity, and the treatment status of patients 3 years after an initial survey. The sharp rise in the proportion of treated patients whose blood pressure was controlled at the time of the follow-up (60.8%) compared with the sit- uation at the time of the initial survey (10.8%) is due to the health education and strong motivation of patients by field survey staff, who emphasized the importance of regular antihypertensive therapy, non- pharmacological measures, and monitoring of blood pressure. Since hypertension by itself is asymptomat- ic, the importance of its long-term effects that can lead to serious complications is generally not real- ized by patients. Many patients do not seek treatment because of ignorance, carelessness, inadequate coun- selling by health care professionals, and the lack of a strong patient-doctor relationship. The link between elevated blood pressure and the incidence of coro- nary heart disease and stroke is well recognized. In the present study, 41 of the 71 cases of angina pec- toris and 5 of the 16 cases of myocardial infarction who developed these complications between the two studies had had mild hypertension in the initial sur- vey. The duration of hypertension among these cases with complications is not known precisely. The Table 5: Annual incidence of hypertension among study adults, by age and sex Males Females Total Age group No. of Incidence No. of Incidence No. of Incidence (years) n new cases (per 1000) n new cases (per 1000) n new cases (per 1000) 25-34 449 10 7.4 952 19 6.7 1 401 29 6.9 35-44 338 16 15.8 667 24 12.0 1 005 40 13.3 45-54 279 11 13.1 428 28 21.8 707 39 18.4 55-64 242 11 15.2 256 13 16.9 498 24 16.1 Total 1 308 48 12.2 2 303 84 12.2 3 611 132 12.2 WHO Bulletin OMS. Vol 72 1994718 A 3-year follow-up of hypertension in Delhi Table 6: Prevalence of risk factors In newly detected cases of hypertension and the control group New hypertension Risk cases Control groupa factor (n = 132) (n = 3 289) Smoking 10 (7.6)b 408 (12.4) Family history 16 (12.1) 478 (14.5) Obesity 32 (24.2) 727 (22.1) Diabetes mellitus 13C (9.8) 32 (1.0) Alcohol consumption 7C (5.3) 39 (1.2) a Free of hypertension and coronary heart disease at the time of initial survey and follow-up survey. b Figures in parentheses are percentages. c P<0.001. initial survey records of 13 of the cases of hyperten- sion who had a stroke showed that 2, 5, and 6 cases, respectively, had mild, moderate, and severe hyper- tension. The association between severity and the risk of vascular complication was more significant for stroke than for coronary heart disease (see Table 4). There were 11 cases of angina pectoris and 4 cases of myocardial infarction among those with iso- lated systolic hypertension. This indicates that the severity of blood pressure cannot be completely characterized by the levels of diastolic blood pres- sure. Raised systolic blood pressure is an indepen- dent risk factor for coronary heart disease (4). In the present study, the blood pressure of 7 of the 87 cases of coronary heart disease was controlled by adequate antihypertensive therapy. Our findings indicate that cases of hypertension with ECG abnormalities such as ST-T changes and left ventricular hypertrophy are potentially at risk of developing complications. Of the 149 persons with abnormal ECGs in the initial survey who were re-examined, 58.4% had developed coronary heart disease. The findings are consistent with the obser- vation that the presence of abnormal ST-T segments or left ventricular hypertrophy in ECG traces is an indicator of future coronary heart disease. Previous- ly, it has been reported that subjects whose ECGs indicated left ventricular hypertrophy had a higher incidence of clinically manifest coronary heart dis- ease than those without this condition (5). Five per- sons whose ECGs in the initial survey were consis- tent with typical left ventricular hypertrophy had normal ECGs upon follow-up evaluation. Their records showed that their blood pressure was con- trolled by regular treatment. Regular antihyperten- sive therapy can therefore reverse ventricular hyper- trophy and result in a normal ECG trace (6). The incidence of hypertension was the same in male and female study subjects (see Table 5). Among men the maximum incidence (15.8 per 1000) occurred for the age group 35-44 years, whereas among women the highest incidence (21.8 per 1000) was for the age group 45-54 years. The incidence was low for both sexes in the age group 25-34 years. In the USA an epidemiological follow-up study based on data obtained during a national survey showed that after average follow-up times of 9.5 years the incidence of hypertension for white men and women was 5.3% and 4.4%, respectively, for the age group 25-34 years, and 18% and 23%, respec- tively, for the age group 55-64 years (7). In other age groups the rates for men and women were almost equal. The interpretation of studies of the incidence of hypertension depends on the case definition used. The cut-off points in the present study were 160 mmHg (21.3 kPa) and 90 mmHg (12.0 kPa) for systolic and diastolic blood pressure, respectively. Apart from treated individuals, the study included people with isolated systolic hypertension. The present study had the following limitations: a high proportion of patients with hypertension were not available for re-evaluation; poor responses to questions about causes of death were often given by family members; and many death certificates were inadequate. However, the major limitation of the initial epidemiological study was that the diagnosis of hypertension was based on the measurement of blood pressure during a single visit. Some patients placed in the mild category might have exhibited normal blood pressure on repeated measurements (8). However, such patients could still be placed in the potentially high-risk category for developing hypertension. Kannel & Dawber have proposed that in adults of any age and either sex, even modest elevation of blood pressure - systolic or diastolic, casual or basal - are associated with a substantial increase in the risk of coronary heart disease (4). Conclusions The present study shows that there is a need to or- ganize a community-based programme for identify- ing individuals with hypertension, bringing them into medical facilities for further evaluation, and maintain- ing a high proportion of them in a long-term control programme. Family physicians have a vital role to play in this respect since the majority of cases of hypertension are treated by them. Health education through the mass media should be designed to wam people about this common public health problem, and to persuade them to have their blood pressure regularly monitored, to seek treatment if required, and to ensure the continuity of such treatment. Hypertension is a chronic condition with serious complications, among which are coronary heart disease, stroke, and renal failure. Systematic man- WHO Bulletin OMS. Vol 72 1994 719 N. Gopinath et al. agement of hypertension therefore has great potential for reducing morbidity and mortality among large numbers of people with high blood pressure, in- cluding the large proportion with mild hypertension. Acknowledgements We thank Dr P.K. Malhotra and Dr O.P. Tiwari; Mr N. Singh, Mr A. Basra, Mrs L. Nagpal, and Mrs N. Masson; and Mrs S. Chawla and Mrs N. Kalra for their valuable assistance in the research; and Mrs S. Muralidharan for secretarial assistance. Resume Hypertension: 6tude de suivi au bout de 3 ans A Delhi Une 6tude de suivi a 6t6 men6e a Delhi 3 ans apres une etude 6pid6miologique sur l'hyperten- sion, basee dans la communaut6, effectu6e selon une m6thode d'enquete de porte-en-porte rando- misee. Les sujets atteints d'hypertension et les adultes normotendus identifies lors de la premiere 6tude ont et6 re6valu6s. Sur les 1749 hyperten- dus identifies lors de la premiere enquete, 1115 ont pu etre r6examines et 74 6taient d6c6d6s. On a analys6 la s6verite de l'hypertension et le traite- ment de ces 1115 patients. La proportion de cas dont la tension art6rielle 6tait bien contr6l6e est passee de 10,8% dans la premiere enquete a 60,8% dans l'etude de suivi. Sur les 732 cas d'hypertension l6gere relev6s dans la premiere 6tude, 460 se sont aver6s normotendus lors du suivi, 231 ont montr6 une hypertension l6gbre, 34 une hypertension moderee et 7 une hypertension s6vere. Au total, 58 sujets pr6sentaient une hypertension severe lors de la premiere 6tude; lors de l'examen de suivi, 21 d'entre eux ont mon- tr6 une tension art6rielle contr6l6e, 20 une hyper- tension l6gere, 8 une hypertension moder6e et 9 une hypertension severe, alors que la proportion des hypertendus non trait6s et non contr6l6s est passee de 54,2% a 14,5%. Sur les 74 d6ces enregistres, 24 ont ete dus a des crises car- diaques, 7 a des accidents c6r6brovasculaires et 4 a une insuffisance r6nale; les causes de d6ces des cas restants sont mal connues. Lors du r6examen des 1115 hypertendus de 1'6tude initiale, 71 (6,4%) ont montr6 des ant6ce- dents d'angine de poitrine, 16 (1,4%) des ant6ce- dents d'infarctus du myocarde et 13 (1,2%) des ant6c6dents d'accident c6rebrovasculaire. Sur les 237 sujets hypertendus et qui presentaient des anomalies de l'electrocardiogramme (ECG) (modi- fications du segment ST-T, hypertrophie ventricu- laire gauche et hypertrophie ventriculaire gauche accompagn6e de modifications ischemiques) lors de 1'enquete initiale, 149 ont ete re6valu6s; 16 d'entre eux pr6sentaient les manifestations cli- niques d'un infarctus du myocarde et 71 celles d'une angine de poitrine. Sur les 3611 sujets normotendus de l'enquete initiale, 132 se sont averes hypertendus au bout de 3 ans. L'incidence annuelle de l'hypertension a ete la meme chez I'homme que chez la femme (12,2 pour 1000). Parmi les facteurs de risque de l'hypertension, le diabete et la consommation reguliere d'alcool tiennent une place importante et sont pr6sents dans 9,8% des nouveaux cas pour le premier et 5,3% des nouveaux cas pour la seconde. La proportion de sujets obeses dans le groupe des hypertendus et le groupe t6moin 6tait de 24,2% et 22%, respectivement. On a fait des 6lectrocardiogrammes chez 871 des 1115 hypertendus de l'etude initiale. Les tra- c6s ont 6t6 anormaux chez 307 d'entre eux (35,2%): 24 (2,7%) avaient eu un infarctus du myocarde, 133 (15,3%) avaient present6 des modifications isch6miques du segment ST-T, 54 (6,2%) une hypertrophie ventriculaire gauche et 96 (11,0%) des troubles de la conduction et des arythmies. References 1. Chadha SL et al. Prevalence, awareness and treat- ment status of hypertension in urban population in Delhi. Indian journal of medical resarch (B), 1990, 92: 233-240. 2. Chadha SL et al. A 3-year follow-up study of coro- nary heart disease in Delhi. Bulletin of the World Health Organization, 1993, 71(1): 67-72. 3. Rose CA et al. Cardiovascular survey methods. Geneva, World Health Organization, 1982 (WHO Monograph Series, No. 56). 4. Kannel WB, Dawber TR. Atherosclerosis as a paediatric problem. Journal of paediatrics, 1972, 80: 544-554. 5. Kannel WB. CHD risk factors: a Framingham study update. Hospital practice, 1990, 25(17): 119-130. 6. Fouad FM, Nakashima Y, Tarazirc Salcedo EE. Left ventricular hypertrophy in hypertensive patients treated with methyldopa. American journal of cardi- ology, 1982, 49: 795-801. 7. Huntley JC, La Crocx AZ, Havlik RJ. Race and sex differentials in the impact of hypertension in United States. Archives of internal medicine, 1989, 149: 780-788. 8. Kilcoyne MM, Richter RW, Alrup MA. Adolescent hypertension. I: detection and prevalence. Circula- tion, 1974, 50: 758-764. 720 WHO Bulletin OMS. Vol 72 1994