What Does a Low Cholesterol Mean?
Abstract & Commentary
Synopsis: Serum albumin and HDL cholesterol levels predict prognosis in elderly people with low serum cholesterol values.
Source: Volpato S, et al. J Am Geriatr Soc. 2001;49: 1142-1147.
Volpato and colleagues from the National Institute on Aging had previously published results from the Established Populations for Epidemilogic Studies in the Elderly (EPESE) which showed that the association of low cholesterol levels and adverse coronary events was spurious and based on failure to adjust for comorbidity and frailty.1 Their study suggested that serum albumin was a marker for poor health. The same group studying the same population had earlier shown that low high-density lipoprotein cholesterol (HDL-C) levels were associated with mortality from coronary heart disease.2 This study’s goal was to determine whether combining albumin and HDL-C levels could stratify the mortality risk among elderly people with low serum cholesterol. During the sixth annual follow-up of the EPESE, blood was drawn from 4128 subjects (mean age, 78.7 years, 64% women). These people were followed for up to 6 years (median, 4.9 years). The outcome of interest was time until death. During the follow-up period, 1117 participants died (27%). Nonfasting blood samples were assayed for total cholesterol, HDL-C, and albumin at baseline. The subjects were classified into 3 groups, low cholesterol (< 160 mg/dL), normal-to-borderline high cholesterol (161-239 mg/dL), and high cholesterol (> 240 mg/dL). Those in the low cholesterol group numbered 393 (9.5%). (One of these subjects was taking a lipid-lowering agent.) Not surprisingly, low-cholesterol subjects had higher all-cause mortality (170 deaths or 43%) than the other 2 groups (29% and 17%, respectively).
To examine the interplay of HDL-C and albumin, the participants were assigned to 1 of 4 categories, low HDL-C/low albumin, high HDL-C/low albumin, low HDL-C/high albumin, and high HDL-C/high albumin. The definitions were as follows: high HDL-C > 47 mg/dL and high albumin > 38 g/L. After adjusting for age, sex, race, body mass index, smoking, alcohol use, and blood pressure, those participants with high albumin had a lower relative risk (RR) of death than those participants with low albumin (RR = 0.57, 95% CI, 0.41-0.79). When the high albumin group was subdivided by HDL-C level, the RR of death for low HDL-C/high albumin subjects was 0.68 (95% CI, 0.47-0.99). For high HDL-C/high albumin the RR was even lower, 0.38 (95% CI, 0.20-0.68). Surprisingly, a high HDL-C level did not confer any benefit to individuals with low albumin.
Comment by Allan Wilke, MD
This article addresses one of the paradoxes of medicine: in some studies when mortality is plotted against serum cholesterol levels, a U-shaped curve is produced. In 1990 the National Heart, Lung, and Blood Institute convened a conference to investigate why this happens.3 The conference concluded that there is a direct relationship between death from coronary artery disease and cholesterol levels, and there is an inverse relationship for a host of other diseases (eg, lung cancers, respiratory disease, digestive disease, and trauma). A study from the Framingham project showed that while low cholesterol levels in people younger than 50 years were associated with longevity, in people older than 50, the association was less clear.4 What confounded the association were those people with spontaneously falling cholesterol values, who were in poor health.
As physicians, we seem to handle this paradox well. We work to lower our patients’ cholesterol levels, all the while knowing that low cholesterol levels are associated with poor outcome in certain patients. Who are these patients? This study makes it clear. They are those folks who also have low albumin levels.
References
1. Corti MC, et al. Ann Intern Med. 1997;126:753-760.
2. Corti MC, et al. JAMA. 1995;274:539-544.
3. Jacobs D, et al. Circulation. 1992;86:1046-1060.
4. Anderson KM, et al. JAMA. 1987;257:2176-2180.
Dr. Wilke is an Assistant Professor of Family Medicine, Medical College of Ohio, Toledo.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content