Descriptive epidemiology of acute myocardial infarction in Kuwait, 1978

Bulletin ofthe WorldHealth Organization, 61 (3): 509-516 (1983) i World Health Organization 1983 Descriptive epidemiology of acute myocardial infarction in Kuwait, 1978 RASHED A. AL-OWAISH,1 In 1978 there were 428 cases of acute myocardial infarction (AMI) among the 18 000 patients admitted to the medical wards of the two main general hospitals in Kuwait; of these, 360 (84%) were males and 171 (40%) were Kuwaitis. The male crude incidence rate (13.1 per 10 000population peryear) was 3.7 times thefemale rate; male age-group specific rates were greater than correspondingfemale ratesfor all age groups, although the ratio of male to female rates decreased with increasing age. AMI incidence rates increased with increasing age in both sexes for Kuwaitis and non-Kuwaitis. Kuwaitis living in high or medium socioeconomic status areas had significantly higher AMI incidence rates (6.6 and 4.5 per 10 000 population per year, respectively) than Kuwaitis in low status areas (2.5 per 10 000 population per year). Managers, adminis- trators, professionals, and technical workers had the highest AMI rates, though the mean age and socioeconomic status did not differ among those affected in these occupational groups. Twenty per cent of the patients had been admittedpreviouslyfor AMI, 22% had a hypertension history, 30% a diabetes mellitus history, and 71% a smoking history. Female patients were more often hypertensive and diabetic, but less often smokers, than males. Kuwaitis, diabetics, and those with a previous admission for acute myocardial infarction had increased mean serum cholesterol levels. Of the 16% ofpatients who had died in the hospital, halfdied within 48 hours ofadmission. Those who died were older and had lower systolic and diastolic bloodpressures and lower serum cholesterol levels than the survivors. Twenty-seven per cent of those with extensive anterior myocardial infarcts died compared with 12% ofthose with acute infarcts in other locations. Those who died within 48 hours of admission had lower systolic and diastolic blood pressures than those who died later. Coronary heart disease is the major cause of death in most industrial countries. As the toll of infectious diseases declines, it is also emerging as an important source of morbidity and mortality in many nonindus- trial countries. This source of premature disability and death will be reduced more through the control of risk factors than through treatment. Modern treatment (e.g., cor- onary care units and aorto-coronary bypass surgery) is expensive and may reduce mortality only slightly. Besides, treatment can affect only those patients who survive long enough after the onset to reach medical care (1). Certain factors affect the risk of coronary heart disease: age; sex; nationality; family history; marital status; socioeconomic status; personal his- tory of hypertension, diabetes mellitus, or cigarette smoking; and an elevated blood cholesterol level (1). To determine the feasibility of reducing the disability ' Chief of Ports, Frontiers Health Division, Department of Public Health & Planning, Ministry of Public Health, Kuwait. and death due to such heart disease in nonindustrial countries, it is first necessary to describe the extent of the problem there. Acute myocardial infarction (AMI), which is a result of coronary heart disease, has an immediate mortality of between 30% and 40/o (2-4). Forty per cent of AMI attacks are fatal within 28 days and almost 60/o of these deaths occur within one hour of the onset (i.e., before any medical intervention). Thus, even the most effective treatment cannot prevent a large proportion ofAMI deaths. Moreover, people who survive the first year after an infarct continue to have greater mortality than expected for their age and sex (1). Two small studies previously indicated the AM] incidence and mortality in Kuwait (5, 6) . The present retrospective study investigates the data for 1978 iin Kuwait, attempts to identify the relationships with some of the previously mentioned risk factors, and compares the results with those from similar studies in Kuwait and elsewhere. 4307 -509- R. A. AL-OWAISH MATERIALS AND METHODS The subjects in this study were selected from the 18 000 medical admissions to Amiri and Sabah Hos- pitals in 1978. These two are the only general hospitals and receive more than 95% of the AMI cases in Kuwait. In Amiri Hospital, only cases admitted as AMI but not with ischaemic pain due to an old in- farction were included. In Sabah Hospital, AMI cases were selected from those who had been admitted to the coronary care unit and those who had been classi- fied by the medical records department under ICD-8 codes 410 ("acute myocardial infarction"), 411 ("other acute and subacute forms of ischemic heart disease"), and 412 ("chronic ischemic heart disease") (7). A case was included in the study if the hospital consultant and at least one other physician had made the diagnosis of acute myocardial infarction based on the following criteria: (i) typical chest pain of more than 30 minutes duration; (ii) serial changes in the electrocardiogram (ECG) indicating ST segment elevation associated with terminal inversion of T waves and the loss of initial QRS potential (i.e., development of pathogenic Q waves of 0.04 second duration or greater); (iii) rises in serum glutamic oxalacetic trans- aminase (SGOT) above 60 units and in lactic dehydro- genase (LDH) above 500 units (8). The following information was abstracted from the hospital record of each case: age; sex; nationality; occupation; place of residence (an index of socio- economic status); marital status; referral source; duration of stay; date of admission; number of pre- vious admissions for AMI; family history of hyper- tension, diabetes mellitus, or AMI; personal his- tory of diabetes mellitus, hypertension or obesity; smoking history; condition at discharge; location of the infarct; serum cholesterol after admission; and blood pressure at admission. Too few records had information on family history or personal history of obesity to be useful in further analysis. Diabetes mellitus was considered present if the patient had a definite history of and took treatment for diabetes, or if the patient's blood sugar had been recorded on more than two occasions (at least 48 hours after onset of the infarction) as being above 7.22 mmol/l (130 mg/dl). Hypertension was con- sidered present if the patient had a definite history of and took treatment for hypertension. Patients were classified as nonsmokers, mild smokers (1-20 cigar- ettes per day), moderate smokers (2140 cigarettes per day), and heavy smokers (41-60 cigarettes per day). The 1975 Kuwait census was the source for denomi- nators of the age-, sex-, nationality- and occu- pational-specific incidence rates (9). Statistical tests used in the analysis included t-tests for independent samples; chi-square tests with Yates' correction (X2YC); chi-square tests for independence; the Mantel- Haenszel chi-square test; Fisher's exact test for 2-by-2 tables; 95% Poisson confidence intervals; and mul- tiple linear logistic regression (10, 11). Probability values less than 0.05 were considered statistically significant. Crude odds ratios (ORCR) and Mantel- Haenszel odds ratios (ORMH) measured the strength of associations between various variables. RESULTS A total of 428 cases (231 from Amiri Hospital and 197 from Sabah Hospital) met the AMI criteria. Table 1 shows the incidence rates according to sex, age group, and nationality. There was no one under the age of 20, the largest percentage (31 %) of cases occurring in those aged 60 years or older, where the age-specific incidence rate was the greatest (51.2 per 10 000 population per year). Age-specific incidence rates increased with increasing age for both males (84% of the cases) and females, in both Kuwaitis (401o of the cases) and non-Kuwaitis. Male age- specific rates were greater than comparable female rates in Kuwaitis and non-Kuwaitis. These age- specific rates, however, were similar within the male and female groups for both nationality groups. Only among those aged 50-59 years were the rates in Kuwaiti males about 400o of those in their non- Kuwaiti counterparts. The ratio of male to female age-specific incidence rates in both nationality groups decreased with increasing age from 29:1 in those aged 20-39 years to 2.3:1 in those aged 60 or older. Age- specific incidence rates did not differ by specific nationality (e.g., Kuwaiti, Jordanian-Palestinian, Indian-Pakistani). Place of residence served to define three levels of socioeconomic status. AMI incidence rates increased with higher status among both Kuwaitis and non- Kuwaitis, although this gradient was statistically significant only between the lower status and the middle and upper status Kuwaitis (Table 2). Only the rate in lower status Kuwaitis was significantly lower than that in lower status non-Kuwaitis; rates in the middle and upper status groups were comparable in both groups of nationalities. The reduced rate in lower status Kuwaitis might be an artifact due to the relatively larger percentage of population aged less than 20 years in this group than in the other groups in Table 2. The age distribution of various places of resi- dence in Kuwait has not been published. If the popu- lation denominators in Table 2 were corrected to 510 ACUTE MYOCARDIAL INFARCTION IN KUWAIT Table 1. Acute myocardial infarction: age-specific incidence rates by nationality and sex, Kuwait, 1 978a Nationality Age group Males Females(years) Cases Population Rateb 95% C.l.c Cases Population Rate 95% C.l.C (No.) (No.) Kuwaiti 20-39 9 60 686 1.5 0.7-2.8 1 63 544 0.2 0.0-0.9 40-44 12 8 611 13.9 7.2-24.3 1 7 720 1.3 0.0-7.2 45-49 16 7 228 22.1 12.7-35.9 3 5 881 5.1 1.1-14.9 50-54 14 5 365 26.1 14.3-43.8 2 5 082 3.9 0.5-14.2 55-59 13 4 094 31.8 16.9-54.3 7 3 687 19.0 7.6-39.1 > 60 68 9 032 75.3 58.5-95.4 23 8 846 26.0 16.5-39.0 All ages 133 95 016 14.0 11.7-16.6 37 94 760 3.9 2.7-5.4 Non-Kuwaiti 20-39 31 125 757 2.5 1.7-3.5 0 73 722 0.0 0.0-0.5 40-44 37 22 478 16.5 11.6-22.7 0 8 949 0.0 0.0-4.1 45-49 43 15 008 28.7 20.7-38.6 0 5 770 0.0 0.0-6.4 50-54 56 8 435 66.4 50.1-86.2 4 3 391 11.8 3.2-30.2 55-59 34 3990 85.2 59.0-119.1 10 1 952 50.7 24.3-93.3 ; 60 26 4 135 62.9 41.4-92.1 16 3 971 40.3 23.0-65.4 All ages 227 179 803 12.6 11.0-14.4 30 97 755 3.1 2.1-4.4 Both 20-39 40 186 443 2.1 1.5-2.9 1 137 266 0.1 0.0-0.4 40-44 49 31 089 15.8 11.7-20.8 1 16 669 0.6 0.0-3.3 45-49 59 22 236 26.5 20.2-34.2 3 11 651 2.6 0.5-7.5 50-54 70 13 800 50.7 39.5-64.1 6 8 473 7.1 2.6-15.4 55-59 47 8 084 58.1 42.7-77.3 17 5 639 30.1 17.6-48.3 >60 94 13 167 71.4 57.7-87.4 39 12 817 30.4 21.6-41.6 Allages 360 274819 13.1 11.8-14.5 67 192 515 3.5 2.7-4.4 One Kuwaiti male had an unknown age greater than 19 years; the sex of another Kuwaiti aged 40-49 years was unknown. b Rate = cases per 10 000 population per year. c 95% Poisson confidence interval. Table 2. Acute myocardial infarction: incidence rates by socioeconomic status and nationality, Kuwait, 1978a Kuwaiti Non-Kuwaiti Socioeconomic status No. of Populationb Rate' 95% C.l.d No. of Populationb Ratec 95% C.l.d cases cases Upper 50 75521 6.6 4.9-8.7 15 21 539 7.0 3.9-11.5 Medium 46 102 600 4.5 3.3-6.0 84 164 979 5.1 4.1-6.3 Lower 72 287 033 2.5 2.0-3.2 148 329 255 4.5 3.8-5.3 Total 168 465 154 3.6 3.1-4.2 247 515 773 4.8 4.2-5.4 e Information on the socioeconomic status of 3 Kuwaitis and 10 Non-Kuwaitis was not available. b Population estimates available only for all ages. c Rate = cases per 10 000 population per year. d 95% Poisson confidence interval. 511 R. A. AL-OWAISH reflect only those aged 20 years or older, then the rate in lower status Kuwaitis might approach the rates in these other groups. Of the 268 persons with known occupations, male administrators and managers had the highest AMI rates (46.2 per 10 000 population aged 15 years old and older), followed by professional and technical workers (16.6); sales workers (4.6) and construction workers and labourers (4.3) had the lowest rates (Table 3). Neither age nor socioeconomic status accounted for this difference in rates among the various occupations. Retired and other workers were 7 years older on the average than administrators and managers, although AMI rates in the former group were about 10% of the latter's. Mean socioeconomic status did not differ among the occupational groups. Over 9807 of the 259 male cases with known marital status were married; 73% of males aged 20 years or older in the general population were married. The AMI rate among married men (12.6 per 10 000 men aged 20 or older per year) was 23 times that among single, divorced, or widowed (i.e., unmarried) men (0.55). Age confounds this relationship between AMI and being currently married: splitting the data into two age groups (20-49 years and > 50 years) yields a summary ORMH = 13.1, about half the crude ratio. If male cases with unknown marital status were mostly unmarried, then the percentage of male cases who were married would more closely resemble that of men aged 20 years or older in the general population. The AMI rate in married men would also be very much like that in unmarried men. Seventy-nine per cent of the female cases were married, which is close to the 77%o of women aged 20 years or older in the general population who were married. The AMI rate among married women (3.0 per 10 000 women aged 20 years or older per year) was comparable to that among unmarried women (2.7). Neither nationality nor socioeconomic status affected the AMI rates in married and unmarried persons of either sex. Twenty-two per cent of patients (N = 374) had a hypertension history, which was more com- mon among females (44%o) than males (18%o) (ORCR = 3.71, X2yc = 19.1, P < 0.0001). Although female patients were usually older than males (Table 1), increased age did not account for the association of sex with hypertension history: the percentage of patients with a hypertension history did not increase with age (2= 4.45, P = 0.108; ORMH= 3.70 for a sex- and hypertension-history association adjusted for age, P < 0.0001). Thirty per cent of patients (N = 374) had a diabetes mellitus history, the incidence being higher among females (42%o) than males (27/o) (ORCR = 1.89, xyc = 4.26, P = 0.039). The increased age of female patients accounted for the association of sex with diabetes history: the percentage of patients with a diabetes history increased with age (X2 = 7.83, P = 0.020; ORMH = 1.56 for a sex- and diabetes- history association adjusted for age, P = 0.14). Forty per cent of the Kuwaiti patients, but only 31 7o of the non-Kuwaitis, had a diabetes history (ORCR= 2.17, x'c = 10.74, P < 0.001). Seventy-one per cent of the patients (N = 295) were smokers, with fewer females (2507o) than males (75%o) (ORCR = 0.11, XYc = 29.08, P<0.0001). Even after adjusting for age, females smoked less often than males (x2= 5.78, P<0.0001, ORMH= Table 3. Acute myocardial infarction: incidence rates by occupation and sex, Kuwait, 1978 Males Females Occupation No. of (%) Rate' 95% C.l.b No. of (%) Ratea 95% C.l.b cases cases Professional, technical 46 (21) 16.6 12.2-22.2 0 (0) 0.0 0.0-2.6 Administrative, managerial 13 (6) 46.2 24.6-79.1 1 (2) 232.6 5.9-1296 Clerical and related 27 (13) 8.0 5.3-11.6 0 (0) 0.0 0.0-8.7 Sales 11 (5) 4.6 2.3-8.3 0 (0) 0.0 0.0-117.9 Housewives, housekeepers 40 (19) 6.4 4.6-8.7 2 (4) 1.3 0.2-4.6 Agricultural, fishing, etc. 4 (2) 5.2 1.4-13.3 0 (0) 0.0 0.0-2838 Production labourers 45 (21) 4.3 3.1-5.7 0 (0) 0.0 0.0-139.2 Retired, others 30 (14) 5.4 3.6-7.6 49 (94) 2.5 1.8-3.2 Total 216 6.8 5.9-7.7 52 2.2 1.7-2.9 a Rate = cases per 10 000 population aged 1 5 years or older. b 95% Poisson confidence interval. 512 ACUTE MYOCARDIAL INFARCTION IN KUWAIT 0.17). Among the 65% of smokers whose amount of smoking was recorded, females smoked less than males (Fisher's exact test P = 0.045). Twenty per cent of the patients (N = 428) had had at least one previous admission for AMI (13.3% for one and 6.807o for two or more AMIs). Those with a previous AMI did not differ from those with their first AMI with respect to sex, age, hypertension his- tory, smoking history, amount smoked, systolic blood pressure, or diastolic blood pressure. The largest percentages of cases were admitted in January (11.77o) and September (11.4%70), and the smallest in December (3.3 0o). The average number of admissions per week was 8, with a range from 2 to 17. Forty-eight per cent came from the clinic or poly- clinic, and 43%o from the casualty department. On admission, 24% of the patients (N = 348) had serum cholesterol levels less than 4.65 mmol/l (180 mg/dl) and 16%7o had levels greater than or equal to 6.72 mmol/l (260 mg/dl). Neither sex, age, history of smoking, nor socioeconomic status affected the serum cholesterol levels. Kuwaitis had higher mean serum cholesterol levels, 5.85 mmol/l (226 mg/ dl), than non-Kuwaitis, 5.4 mmol/l (209 mg/dl) (P = 0.001, t-test for independent samples). Patients with a history of diabetes mellitus had higher mean serum cholesterol levels, 5.95 mmol/l (230 mg/dl), than those without such a history, 5.45 mmol/l (211 mg/dl) (P = 0.0008, t-test). Patients with a prior admission with an infarct also had higher serum cholesterol levels, 5.9 mmol/l (228 mg/dl), than those admitted with their first AMI (213 mg/dl) (P = 0.03, t-test). Each of these three factors Kuwaiti nationality, history of diabetes mellitus, and previous admission with AMI -independently favoured an increased serum cholesterol level: those with all three had a mean cholesterol level of 6.86 mmol/l (265 mg/dl); those with none had a mean level of 5.3 mmol/l (205 mg/dl). Admission systolic blood pressures fell below 100 mmHg (13.3 kPa) in 7% of the patients (7 had 0 mmHg systolic pressures) and exceeded 160 mmHg (21.3 kPa) in 160o. Nine patients had 0 mmHg admission diastolic blood pressures and 12% had diastolic pressures above 100 mmHg (13.3 kPa). Ad- mission blood pressures were unrelated to demo- graphic or other potential risk factors besides a past history of hypertension. Fifty-eight per cent of the patients had an anterior myocardial infarct (271o anteroseptal, 24% extensive anterior, 7% antero- lateral), 41%0o had an inferior and 1%o a posterior myocardial infarct. Demographic or other potential risk factors did not affect the location of the infarct. The location did, however, affect the in-hospital mortality rate. Eighty-four per cent of the patients survived till discharge, although 4% were discharged against medical advice. Of the 16% who died in the hospital, half died within 48 hours of admission. Those who died were older and had lower admission systolic and diastolic blood pressures and lower serum cholesterol levels (Table 4). In each quartile of duration of stay, the mean serum cholesterol level of those who died was lower than that of the survivors. Those who died within 48 hours of admission had lower systolic and diastolic blood pressures than those who died later in the hospital; 6 out of the 7 persons with 0 mmHg ad- mission systolic blood pressures and 6 out of the 9 with 0 mmHg diastolic blood pressures died within 48 hours of admission. Twenty-seven per cent of those with extensive anterior myocardial infarcts died, com- pared with 12%o of those with infarcts elsewhere (ORCR = 2.63, X2c = 11.19, P= 0.0008); the former also had the lowest mean admission systolic and dias- tolic blood pressures. Smokers had a lower in- hospital mortality (8%) than nonsmokers (19/o), although mortality among heavy smokers (8%o) did not differ significantly from that for moderate (6%o) or light smokers (8%7o). A multiple linear logistic regression with the five variables- age, serum choles- terol, presence of an extensive anterior myocardial infarct, smoking history, and systolic blood pressure - selected only the first three as statistically signifi- cant (P < 0.05) prognostic indicators of in-hospital mortality. DISCUSSION Kuwait occupies the northwestern corner of the Gulf, bounded on the east by the Gulf itself, on the south by Saudi Arabia, and on the north and west by Iraq. Kuwait covers about 17 800 kM2, and in 1975 the population was about one million, half of whom were native Kuwaiti. In this developing country, the main causes of mortality are changing from infec- tious diseases to the more chronic degenerative and neoplastic diseases common in the developed coun- tries (9). One of these changes is the recent increase in mortality from coronary heart disease. The epidemi- ology of AMI in Kuwait mimics some of the patterns seen in the USA and Europe, particularly in the 1950s and 1960s (12, 13). The 1978 AMI incidence rates were lower in Kuwait than in the latter areas at that time, although age-specific rates in Kuwait resemble comparable rates in some European communities studied more recently (14). But male AMI rates were also higher than female rates. AMI incidence rates increased with age in both sexes, with a 29:1 male to female ratio in those aged 20-39 years, narrowing to 2.3:1 in those aged 60 years and older. In the Framingham study, the male to female incidence ratio 513 R. A. AL-OWAISH Table 4. Acute myocardial infarction: characteristics of survivors, early hospital deaths, and late hospital deaths, Kuwait, 1978 Deaths in hospital in: Survivors < 48 hours > 48 hours Characteristic Mean S.D." No.' Mean S.D.' No.' Mean S.D.' No.' Age (years)b 52 12 360 58 13 33 61 14 34 Systolic blood pressure, on admission, in mmHg (kPa)c 140 (18.7) 29 (3.9) 360 101 (13.5) 56 (7.5) 34 131 (17.5) 29 (3.9) 34 Diastolic blood pressure, on admission, in mmHg (kPa)d 89 (11.9) 17 (2.3) 359 64 (8.5) 38 (5.1) 34 80 (10.7) 26 (3.5) 34 Serum cholesterol in mmol/l (mg/dl)' 5.7 (219) 1.2 (46) 318 4.8 (186) 0.64 (25) 5 4.7 (183) 0.88 (34) 25 a S.D. = standard deviation; No. = number of patients. b Survivors were significantly younger than either those who died within 48 hours of admission (P = 0.0183) or those who died later (P = 0.0001); the latter 2 groups did not differ by age (P = 0.33). ' Survivors had significantly higher admission systolic blood pressures than those who died within 48 hours (P = 0.0003) but not those who died later (P = 0.09); those who died later also had significantly higher admission systolic blood pressures than those who died earlier (P = 0.008). d Survivors had significantly higher admission diastolic blood pressures than those who died within 48 hours (P = 0.0008) but not those who died later (P = 0.073); those who died later also had significantly higher admission diastolic blood pressures than those who died earlier (P = 0.047). ' Survivors had significantly higher serum cholesterol levels than those who died later (P < 0.0001) but not those who died earlier (P = 0.12); the latter 2 groups did not differ by serum cholesterol level (P = 0.85). was 13:1 in those aged 30-44 years and 2:1 in older patients (15). The increased rates in upper and middle status Kuwaitis and in managers, administrators, professionals, and technical workers parallel the trends in the USA and Europe 20 to 30 years ago (13), but are unlike current trends (16, 17). The percentage of AMI patients with various risk factors both resembles and differs from that of AMI patients in previous studies in Kuwait and elsewhere. Twenty-two per cent of patients in our study had a hypertension history, like that obtained by Al Yousef in Kuwait earlier (5) but lower than the 740%o of AMI cases with hypertension in Framingham over an 18- year follow-up (8). Thirty per cent of our patients had a diabetes mellitus history, which is higher than the 19% observed in Kuwait previously (5) and the 8% in Framingham (8). Seventy-one per cent of the patients in our study were smokers, which is also higher than in Framingham (57%). Twenty per cent of our patients had had at least one previous admission for AMI, like that reported from elsewhere (18). Unfor- tunately, not enough information has been published from these other studies to adjust for possible differ- ences in age and sex distributions among the studies. The observed 16% hospital mortality is 40% less than that seen before in Kuwait (5, 6). Diagnostic differences or improved care may account for some of this decrease. The prognostic factors for in-hospital mortality among AMI patients in Kuwait resemble those observed in other studies (5, 6, 18-21). Those who died in the hospital were older than the survivors. Those who died within 48 hours of admission had sig- nificantly lower systolic and diastolic blood press- ures, a possible indicator of cardiogenic shock. Those who died had extensive anterior myocardial infarcts, an indicator of massive heart damage. Others have also observed reduced serum cholesterol levels among those AMI patients who died in the hospital (21). Although, for comparable durations of stay, the mean serum cholesterol level of survivors always exceeded that of those who died, the lack of multiple serum cholesterol measurements in individuals prevents an assessment of the relationship between changes in serum cholesterol and subsequent mor- tality. Hypertension or diabetes mellitus histories did not affect hospital mortality in this study as it has in others (5, 19, 20). This report represents only a preliminary study of AMI epidemiology in Kuwait. A companion report 514 ACUTE MYOCARDIAL INFARCTION IN KUWAIT 515 describes a matched case-control study of the risk factors associated with AMI in some of these patients (22). Because the present study is a retrospective review of hospital records, certain information was lacking in some of the case records. This reduces the statistical power of the study as well as permits differential misclassification to bias the conclusions. It was not possible, for example, to analyse the infor- mation about family history or obesity as potential risk factors for AMI. Future studies in Kuwait should involve prospective follow-up of patients as they are diagnosed, with interviews of patients or next-of-kin to obtain more of the required information. ACKNOWLEDGEMENTS This study would not have been possible without the technical assistance of Mr Kamel Al-Saleh, chief of the Vital and Health Statistics Division, Ministry of Public Health, Kuwait. I thank Dr Abdul Razak Al-Yousef, director of the medical department at Sabah Hospital, and Dr Basel Al-Naqeeb, head of the cardiovascular intensive care unit at Amiri Hospital, for their painstaking efforts in planning this study and for helpful suggestions. I am indebted to staff at the United States Centers for Disease Control (in the epidemiology program office and the center for professional development and training, mainly to Dr Matthew Zack) and the Preventive Medicine Division in the Kuwait Ministry of Public Health for their valuable guidance and advice. RtSUMt EPIDEMIOLOGIE DESCRIPTIVE DE L'INFARCTUS AIGU DU MYOCARDE AU KOWEYT, EN 1978 En 1978, il y a eu 428 cas d'infarctus aigu du myocarde (IAM) parmi les 18 000 malades admis dans les services de medecine des deux principaux hopitaux generaux du Koweit; sur ce nombre, 360 (8401.) etaient de sexe masculin et 171 (40%) etaient des Koweitiens. Le taux d'incidence brut chez les hommes (13,1 pour 10 000 par an), etait 3,7 fois plus eleve que celui des femmes; les taux par age chez les hommes etaient superieurs aux taux correspondants chez les femmes, et cela pour tous les groupes d'age, mais le rapport taux hommes/taux femmes diminuait avec l'augmentation de I'age. Les taux d'incidence de l'IAM s'e1evaient avec I'age dans les deux sexes, aussi bien chez les Koweitiens que les non Koweitiens. Parmi les Koweitiens habitant dans des secteurs oui le niveau socio-conomique etait eleve ou moyen, les taux d'incidence de 1'IAM etaient plus hauts (6,6 et 4,5 pour 10 000 par an, respectivement) que chez les Koweitiens des secteurs de bas niveau socio-economique (2,5 pour 10 000 par an). Les directeurs, administrateurs, membres des pro- fessions liberales et techniciens avaient les taux les plus hauts d'IAM, mais l'age et le niveau socio-economique moyens ne differaient pas parmi ceux qui etaient atteints dans ces groupes professionnels. Vingt pour cent des malades avaient e admis anterieurement pour un IAM, 22% avaient des antecedents d'hypertension, 30% de diabete et 71% etaient des fumeurs. Les femmes etaient plus souvent hypertendues et diabetiques mais moins souvent des fumeuses. Chez les Koweitiens diabetiques ou ceux qui avaient deja et hospita- lises pour un infarctus aigu du myocarde, les taux seriques moyens de cholesterol etaient augmentes. Sur les 16% de malades decedes l'hopital, la moitie sont morts dans les 48 heures qui ont suivi I'admission. Les malades decedes etaient plus ages et avaient une pression arterielle systolique et diastolique plus basse et des taux de cholest6rol seriques plus faibles que les survivants. Vingt-sept pour cent des sujets presentant un infarctus anterieur extensif du myo- carde sont decedes contre 12% de ceux qui etaient atteints d'infarctus aigu d'une autre localisation. Chez les malades decedes dans les 48 heures apres l'admission, la pression systolique et diastolique etait plus basse que chez ceux qui sont morts plus tard. REFERENCES 1. MARMOT, M. G. Epidemiological basis for the preven- tion of coronary heart disease. Bull. Wid Hith Org., 57: 331-47 (1979). 2. HONEY, G. E. & TRUELOVE, S. C. Prognostic factors in myocardial infarction. Lancet, 1: 1155-61 (1957). 3. WAHLBERG, F. A study of acute myocardial infarction at the Seraphimer Hospital during 1950-1959. Am. Heart J., 65: 749-57 (1963). 4. BROWN, K. W. G. ET AL. Coronary unit: an intensive care unit for acute myocardial infarction. Lancet, 2: 349-52 (1963). 5. YOUSEF, A. M. & QARAMAN, A. Acute myocardial infarction in Kuwait. J. Kuwait Med. Assn, 3: 69ff (1969). 516 R. A. AL-OWAISH 6. EMARA, M. K. & AL YOUSEF, A. R. Acute myocardial infarction in females in Kuwait. J. Kuwait Med. Assn, 11: 77ff (1977). 7. WORLD HEALTH ORGANIZATION. Manual of the inter- national statistical classification of diseases, injuries, and causes of death. Eighth revision, 1965. Volume 1. Geneva, 1967. 8. SHURTLEFF, D. Some characteristics related to the inci- dence of cardiovascular disease and death: Framingham Study, 18-year follow-up. In: Kannel, W. B. & Gordon, T. (ed.) The Framingham Study, an epidemiological investigation of cardiovascular disease. Washington, 1974. (DHEW Publication No. (NIH) 74-599), pp. 18-20, Tables 3-1 to 3-18. 9. Kuwait annual statistical abstract, 1978. Edition XV. Kuwait City, Ministry of Planning, 1978. 10. DIXON, W. J. & MASSEY, F. J. JR. Introduction to statistical analysis. 3rd ed. New York, McGraw-Hill, 1969. 11. BRESLOW, N. E. & DAY, N. E. Statistical methods in cancer research. Volume 1. The analysis ofcase-control studies. Lyon, International Agency for Research on Cancer, 1980. 12. MORIYAMA, I. M. ET AL. Cardiovascular diseases in the United States. Cambridge, Harvard University Press, 1971. 13. The Registrar-General's decennial supplement. England & Wales, 1951. Occupational mortality, Part 2, Volume 1, Commentary. London, Her Majesty's Stationery Office, 1958. 14. Myocardial infarction community registers. Copen- hagen, WHO Regional Office for Europe, 1977. (Public Health in Europe series No. 5) 15. KANNEL, W. B. Some lessons in cardiovascular disease epidemiology from Framingham. Am. j. cardiol., 37: 269-82 (1976). 16. MARMOT, M. G. ET AL. Employment grade and coronary heart disease in British civil servants. J. epidemiol. commun. health, 32: 244-249 (1978). 17. The Registrar-General's decennial supplement for England and Wales, 1970-72. Occupational mortality Series DS No. 1. London, Her Majesty's Stationery Office, 1978. 18. HELMERS, C. Short- and long-term prognostic indices in acute myocardial infarction. Acta medica Scandi- navica, Suppl. 555: 1-86 (1974). 19. BILLINGS, J. R. ET AL. Prognosis of acute myocardial infarction. Am. j. med., 7: 356-370 (1949). 20. PELL, S. & D'ALONZO, C. A. Immediate mortality and five-year survival of employed men with a first myocardial infarction. New Engl. j. med., 270: 915-922 (1964). 21. FYFE, T. ET AL. Plasma-lipid changes after myocardial infarction. Lancet, 1: 997-1001 (1971). 22. AL-OWAISH, R. A. & ZACK, M. M. Risk factors associated with acute myocardial infarction in Kuwait in 1978. Internat. j. epidemiol., 11 (4): 368-371 (1982).