Well-designed studies have shown that statins reduce the risk of cardiovascular events in elderly patients. It is unclear whether this beneficial effect is caused by the lowering of cholesterol levels, given that there appears to be no association between high cholesterol levels and all-cause mortality risk in this population. In an effort to provide further evidence that statins benefit the elderly independently of cholesterol level, Karlamangla and colleagues examined the relationship between changes in cholesterol level and health effects in a cohort of healthy older adults.

Participants included 1,189 men and women 70 to 79 years of age who were followed for a mean of 28 months to determine changes in cholesterol level and then for an additional 57 months to determine health outcomes. Baseline data and blood test results were collected, and a complete battery of laboratory tests was repeated at the end of the study in a random subsample of 267 participants. The primary outcome was mortality, and secondary outcomes were myocardial infarction and stroke. Physical and cognitive functioning also were assessed.

Baseline levels of total and high-density lipoprotein (HDL) cholesterol in the sub-sample were 218 mg per dL (5.60 mmol per L) and 47 mg per dL (1.20 mmol per L), respectively. There was a wide range of change in non-HDL cholesterol levels. In analyses that stratified the study population by quartiles, mortality decreased when non-HDL cholesterol increased. In the continuous analysis, increasing non-HDL cholesterol levels decreased mortality risk but was not associated with cardiovascular event risk or cognitive decline.

Participants with increasing non-HDL cholesterol levels also had better function with regard to activities of daily living. Some modifications of these findings were found in yet another analysis. While most of the effects held, participants with the highest cholesterol levels (245 mg per dL or higher [6.30 mmol per L]) did have a higher odds ratio for mortality than the two other cholesterol ranges analyzed. Patients with levels of 191 to 244 mg per dL (4.90 to 6.30 mmol per L) had the lowest odds ratio for mortality. In addition, cardiovascular disease status at baseline made a difference in functional and cognitive decline. There was a lower odds ratio in functional and cognitive decline with increasing non-HDL cholesterol levels, but only in participants without baseline cardiovascular disease.

After recording changes in cholesterol levels over 2.5 years, a cohort was followed for 4.5 additional years to correlate cholesterol levels with mortality, cardiovascular events, and physical and cognitive function. Overall, those with increasing non-HDL cholesterol levels had a lower mortality risk and a lower risk of cognitive and physical decline. The authors suggest several reasons why cholesterol may have a protective effect, including the possibility that aging cells need higher cholesterol levels to maintain themselves. They also suggest that high cholesterol levels may not adversely affect certain persons, and these are the patients who survive into older age. Whatever the reason, the implication of these findings is that the criteria for hypercholesterolemia in older persons might not be the same as those in younger persons, and the elderly might not require treatment at the same thresholds. It is too soon to make clinical recommendations without further trials.