Am Fam Physician. 2004;69(9):2225-2226
Cholesterol levels have been shown to be a predictor of all-cause mortality in middle-aged patients. However, in elderly patients, cholesterol levels do not appear to follow this pattern. Some studies of elderly patients have shown that higher levels of cholesterol may decrease mortality risks compared with patients who have lower cholesterol levels. In other studies, these findings have not been replicated. In elderly patients, total cholesterol levels tend to decline over time, which may be secondary to the effects of poor health on these levels. Other studies that assessed nutritional parameters have found that low levels of albumin, lymphocytopenia, and low body mass index increase the risk of in-hospital mortality. In addition, markers for inflammation predicted mortality rates. Onder and associates conducted an observational trial that examined whether the total cholesterol level would predict in-hospital mortality and if it could be used as the only marker of health status among elderly patients.
The authors reviewed the hospital admission records of patients admitted to 81 different centers during four specific periods. Each patient responded to a questionnaire at admission, and the information was updated daily by a study physician. Data collected during the study were demographic information, functional and cognitive status, medication use before admission and during hospital stay, and diagnoses at admission and discharge. In-hospital mortality also was recorded. Total cholesterol levels were included in routine laboratory evaluation for all of the patients admitted. Only patients 65 years and older were enrolled in the study. The patients were divided into four groups, based on their total cholesterol level—group 1, less than 160 mg per dL (4.14 mmol per L); group 2, 160 to 199 mg per dL (4.15 to 5.15 mmol per L); group 3, 200 to 239 mg per dL (5.17 to 6.18 mmol per L); and group 4, 240 mg per dL (6.21 mmol per L) and higher.
Of the 6,984 study participants, 2,115 patients were enrolled in group 1; 2,210 in group 2; 1,719 in group 3; and 940 in group 4. The mean age of participants was 78 years, and the average length of stay was 15 days. The total number of in-hospital deaths was 202 during the study period. Analysis of the relationship of total cholesterol level to mortality revealed an inverse relationship; patients in group 1 had a mortality rate of 5.2 percent; in group 2, 2.2 percent; in group 3, 1.6 percent; and in group 4, 1.7 percent. This was a significant trend. When all other variables were controlled, patients with lower cholesterol levels continued to have higher mortality rates than patients with higher cholesterol levels. This trend continued when the data were adjusted for inflammatory markers.
The authors conclude that low total serum cholesterol levels appear to be an independent predictor of short-term mortality rates in hospitalized elderly patients, regardless of the presence or absence of malnutrition, frailty, inflammation, and comorbidities.