Higher CD4 Counts -- But Not Statins -- Cut Risk of Heart Attack, Stroke and Mortality

Statin use did not affect risk of myocardial infarction (MI), stroke and all-cause mortality in a retrospective study of 438 people taking combination antiretroviral therapy. But a higher CD4 count did lower the risk of that combined endpoint in the study, which analyzed data from the U.S. Nutrition for Healthy Living (NFHL) cohort.

Systemic inflammation and metabolic abnormalities persist in people responding well to combination antiretroviral therapy and may contribute to cardiovascular disease and other non-AIDS comorbidities. To see if the anti-inflammatory and anti-lipid impact of statins could affect those outcomes, as well as mortality, NFHL investigators conducted this retrospective analysis.

NFHL is a prospective cohort that began enrolling HIV-positive adults in 1995, excluding people with uncontrolled hypertension, MI or stroke within the past six months, but keeping cohort members in whom those conditions developed after they entered NFHL. This analysis focused on people who began combination antiretroviral therapy in September 2000 or later. The investigators compared the incidence of MI, stroke and all-cause mortality (a composite endpoint) in people who took statins versus those who did not. To identify the factors associated with reaching the composite endpoint, the investigators used two Cox proportional hazards models: one focused on statin use as a binary (yes-or-no) variable and one considering statin duration. Other time-varying predictors in the model were CD4 count and low-density lipoprotein (LDL) cholesterol.

The study group consisted of 438 people, 141 of them (32%) women, 53% white, 32% black and 10% Hispanic. The average age was 44.3 years, average CD4 count 426 and average LDL cholesterol 113 mg/dL. Almost half of the cohort (47%) smoked, and 35% had hypertension. Sixty-seven people (15%) used statins during follow-up, which averaged 275 weeks in people who never used statins and 411 weeks in those who did.

Neither multivariate model found an association between statin use and risk of the composite MI-stroke-mortality endpoint (hazard ratio [HR]: 1.26, 95% confidence interval [CI]: 0.57 to 2.79, for statin use as a binary variable; HR: 0.93, 95% CI: 0.65 to 1.32, for statin duration).

Three variables did predict the composite outcome in both models. Every year of age was associated with a 7% higher risk of reaching the composite endpoint (HR: 1.07, 95% CI: 1.03 to 1.1), smoking conferred a 78% higher risk (HR: 1.78, 95% CI: 1.04 to 3.19) and every 50-cell higher CD4 count lowered the risk 12% (HR: 0.88, 95% CI: 0.83 to 0.94) in both models.

In a sensitivity analysis that eliminated all-cause mortality from the composite endpoint and thus reduced the number of endpoints to 20, only age and CD4 count remained independent predictors of MI and stroke. Every year of age conferred an 8% higher risk, and every 50-cell higher CD4 count meant a 12% lower risk.

The NFHL team noted that a retrospective cohort study like this cannot settle the question of whether statins prevent cardiovascular disease and death in people with HIV. A possibly definitive placebo-controlled trial of pitavastatin (brand name: Livalo) in 6,500 HIV-positive adults (REPRIEVE, ACTG A5332) is recruiting participants. The authors proposed that "CD4 count preservation might play a protective role not only for the whole composite [endpoint] but also for pure cardiovascular outcomes after excluding death as an outcome from the analysis."

Mark Mascolini is a freelance writer focused on HIV infection.