Cholesterol, statins, and longevity from age 70 to 90 years

Abstract

Background: The importance of cholesterol as a risk factor among older people, particularly among the very old, is controversial. Whether or not hypercholesterolemia warrants medical concern, and whether statins are beneficial among very old people, remain unresolved common clinical dilemmas. This study examines whether increased total cholesterol (TC) was associated with higher mortality from age 70 to 90, and if statins had a protective effect.

Methods: A representative sample (born 1920-1921) from the Jerusalem Longitudinal Cohort Study (1990-2010) was assessed at ages 70, 78, and 85 for fasting serum TC, low-density (LDL), and high-density lipoprotein (LDL); triglycerides; statin usage; social, functional, and medical domains; and all-cause mortality data (1990-2010). TC was analyzed as either continuous (10 mg/dL increments) or dichotomous variable (high TC >200 mg/dL). Cox proportional hazards models determined mortality hazard ratios (HRs), adjusting for TC, statin treatment, gender, self-rated health, smoking, hypertension, diabetes, ischemic heart disease, neoplasm, body mass index, albumin, and triglycerides.

Results: Prevalence of high TC at ages 70, 78, and 85 was 75% (n = 344), 65% (n = 332), and 34% (n = 237), and statin use was 0%, 17.9%, and 45.4%, respectively. Survival was increased (not significantly) among subjects with high TC >200 mg/dL versus ≤200 mg/dL from ages 70 to 78, 78 to 85, and 85 to 90: 79.1% versus 73.3% (log rank P = .16), 68.7% versus 61.5% (P = .10), and 73.4% versus 70.3% (P = .45), respectively. Survival was significantly increased among subjects treated with statins versus no statins at ages 78 to 85 (74.7% vs 64.3%, log rank P = .07) and 85 to 90 (76.2% vs 67.4%, P = .01). After adjustment, TC (continuous or dichotomous) was not associated with mortality from 70 to 78, 78 to 85, or 85 to 90. In contrast, statins at age 85 were associated with decreased mortality from age 85 to 90 (adjusted HR 0.61, 95% confidence interval 0.42-0.89).

Conclusions: Among older people, cholesterol levels were unrelated to mortality between the ages of 70 and 90. The protective effect of statins observed among the very old appears to be independent of TC.

Keywords: Cholesterol; longevity; longitudinal cohort study; oldest old; statins.