In 12 years of follow-up for 2748 Framingham Heart Study participants ages 50 to 79, low levels of high density lipoprotein cholesterol (HDL-C) were associated with increased mortality. For men, the relative risk of death in the first HDL-C quintiie (<35 mg/dl) as compared to the top quintiie (>54 mg/dl) was 1.9 for all causes, and 3.6 and 4.1 for death due to cardiovascular and coronary heart disease (CHD), respectively, after adjustment for standard cardiovascular risk factors. In women, corresponding relative risks, comparing the bottom HDL-C quintiie (<45 mg/dl) to the top quintiie (>69 mg/dl), were 1.5, 1.6, and 3.1. With HDL-C considered as a continuous variable, and after adjustment for standard cardiovascular risk factors, highly significant associations were seen with HDL-C and CHD death in both men and women. In addition, a significant HDL-C effect on total mortality and death due to cardiovascular disease was seen in men. In none of the continuous variable analyses was HDL-C associated with cancer death. We conclude that HDL-C is a potent predictor of CHD death in both sexes and has less consistent associations with other types of death. (Arteriosclerosis 8:737-741, November/December 1988) W hile morbidity studies for high density lipoprotein cholesterol (HDL-C) and cardiovascular disease (CVD) are common, investigators have infrequently studied the relationship between HDL-C and mortality.1 -4 There is even less evidence available for potential relationships between HDL-C and noncardiovascular mortality. While some studies have looked into the role of HDL-C in determining mortality for myocardial infarction survivors, 567 the results are largely confined to men, and the impact of HDL-C on mortality for individuals free of coronary heart disease (CHD) has been investigated in less detail. Recent follow-up from the Framingham Study now allows a closer look at the HDL-C to mortality relationship for healthy individuals at baseline, and enables us to determine whether this risk factor, which is so important for CVD morbidity, is equally important for mortality.
MethodsThe Framingham Heart Study has followed 5209 men and women every 2 years since 1948 for the development of cardiovascular and other diseases. 89 At any given examination approximately 80% of the survivors attended, and less than 1 % of the population was lost to follow-up by 1970. During the 1970-1972 examination, fasting levels of HDL-C were determined among the study participants after heparin-manganese chloride precipitation of fresh plasma following a modification of the Lipid Research Clinics Program protocol.
10For this report, all subjects were followed 12 years from the time of HDL-C measurement until death due to all causes, death from CHD, death from CVD (including coronary heart disease, stroke, cardiac failure, and pulmonary circulatory disorders), and death from cancer. None were lost to follow-up. The cause of death was determined by a panel of Framingham Heart Study physicians who reviewed all available evidence relating to ...