Unusually low levels of high density lipoprotein cholesterol (HDL-C) were identified as a unique characteristic of the Turkish population several years ago. Turkish men have average HDL-C levels of about 36 mg/dl, and Turkish women have average HDL-C levels of about 42 mg/dl. These values are 10–15 mg/dl lower than those of men and women of western European descent. As a result of the low HDL-C levels, the risk of CHD may be increased as much as 30–40% in Turks compared with other populations with higher HDL-C levels.

In the past year, we have compiled data for an additional 406 Turks who were not included in the database of the original Turkish Heart Study (THS) cohort. The additional subjects were mostly residents of Istanbul and were studied in 1998–1999, whereas the original cohort of ~9000 subjects were studied between 1990–1993. The purpose of this update was to document the current risk factor status of Turks and to consider the appropriateness of recent treatment guidelines for dyslipidemia in a population characterized by “normal” total and low density lipoprotein cholesterol (LDL-C) levels and low HDL-C levels. This is an important issue not only for Turks, but also for CHD patients in the United States who are at risk because of low HDL-C levels. About 20% of CHD patients have low HDL-C levels and normal total cholesterol and LDL-C levels.

Table 1 provides information about the average age, lipid parameters, and body mass index (BMI) of the additional 406 subjects. The average total cholesterol and LDL-C levels were similar to the values of the earlier cohort, but triglyceride levels were about 30 mg/dl greater for both men and women. The ratio of total cholesterol/HDL-C, ~4.6, was similar in men studied in 1990–1993 and 1998–1999, but the total cholesterol/HDL-C values for women increased from 4.2 to 4.8. The average HDL-C levels were 36 mg/dl for men and 42 mg/dl for women and were unchanged from 1990–1993 values. The average BMI of both men and women increased 1–2 kg/m2 from 1990–1993 values, and fasting triglyceride values increased ~30 mg/dl. About two-thirds of the men and one-half of the women had BMI values >25 kg/m2. In the United States about half of the adult population is similarly overweight. The prevalence of smoking, always high in Turkey, has not changed since 1990–1993.

Table1. Plasma lipid and lipoprotein levels in Turkish men and women
living in Istanbul*

The prevalence of obesity in Turks and the well-known inverse relationship of obesity with triglyceride levels and HDL-C levels led us to evaluate the relationship of the plasma lipid parameters with BMI in the 406 follow-up subjects. This is also of interest with regard to whether or not the high prevalence of low HDL-C levels in Turks might be accounted for solely on the basis of obesity. Table 2 shows the average plasma lipid values in Turks according to ranges of BMI. Turkish men and women who are not overweight (BMI <25 kg/m2) have substantially lower average triglyceride levels and higher HDL-C levels than individuals who are overweight (BMI >25 kg/m2). However, the average HDL-C levels of Turkish men and women with BMI <25 kg/m2 were 37 mg/dl and 44 mg/dl, respectively, values that are substantially below the mean HDL-C values of 47 mg/dl in men and 58 mg/dl in women in the United States. These data suggest that obesity affects triglyceride and HDL-C levels in the Turkish population just as in the population of the United States. However, low HDL-C levels are still highly prevalent among the nonobese in Turkey, suggesting that low HDL-C levels are not simply a consequence of obesity or obesity-associated insulin resistance.

Table 2. Changes in mean plasma lipid and lipoprotein levels (mg/dl ± SD) associated with body mass index (BMI).*

The Turkish Heart Study database was also used to study the ability of recently formulated dyslipidemia treatment guidelines to identify high-risk lipid profiles in a population characterized by low HDL-C levels. Existing guidelines in the United States failed to identify for treatment many patients who are at risk because they have low total cholesterol and low HDL-C levels. Consequently, we have developed guidelines for patients primarily at risk because of low HDL-C levels. These guidelines are also widely applicable to high-risk patients in the United States, where about one-quarter of CHD patients have low HDL-C levels.

We have continued our family study of insulin resistance in the Turkish kindreds. Anthropometric and demographic data, as well as blood samples, have been collected from many families in several regions throughout Turkey. Analyses to identify genetic polymorphisms associated with hypertriglyceridemia and low HDL-C levels are currently under way.