Requena G, DiBello J, Ma L, Sturmer T, Layton JB. Short term HDL cholesterol (HDL-C) decrease and incidence of cardiovascular (CV) risk. Poster presented at the 34th Annual ICPE Conference; August 26, 2018. Prague, Czech Republic. [abstract] Pharmacoepidemiol Drug Saf. 2018 Aug; 27(S2):468.


BACKGROUND: Several studies have demonstrated an inverse association between prolonged increases in HDL‐C and CV risk. However, the literature describing the effect of short term decreases in HDL‐C on CV risk is limited.

OBJECTIVES: To estimate the risk of CV events in subjects experiencing a decrease in HDL‐C after initiating statin therapy for primary and secondary CV prevention while on short term treatment, compared with those with constant HDL‐C levels.

METHODS: As some statins users experience HDL‐C decreases upon initiation, we conducted a retrospective cohort study comprised of subjects who initiated a treatment with statins and discontinued it within 9 months, using CPRD linked to hospital (HES) and mortality data. HDL‐C measurements were assessed within 9 months before and after statin initiation and subjects were followed for up to 5 years after statin discontinuation for CV events, comparing those with a decrease in HDL‐C (>8%) to those with constant HDL‐C (±8%). The cut‐point was derived from the distribution of HDL‐C change after statin initiation. Major adverse cardiac events (MACE) primary composite endpoint was defined as CV death, myocardial infarction, revascularization, and hospitalized ischemic stroke. We estimated crude, multivariable adjusted, and propensity score adjusted 5‐year risk differences and hazard ratios, comparing Decrease and Constant HDL‐C groups.

RESULTS: From 17 543 subjects, 6454 were in the Decrease group and 11 089 in the Constant group. After adjustment for competing risks and confounding factors the 5‐year cumulative risk difference for MACE events was 1.23% (95% CI: 0.28%, 2.18%). Differences in risk were small in magnitude and were not statistically significant until the 5th year of the follow‐up. In line with this, the hazard ratio across groups for the composite MACE endpoint was higher in the Decrease compared with the Constant group, 1.20 (95% CI: 1.04, 1.39). The elevated risk in this group was mainly driven by an increased risk in ischemic stroke (1.44 95% CI: 1.08, 1.90) and CV death (1.23 95% CI: 0.93, 1.63).

CONCLUSIONS: In our study, we found that patients with a short‐term decrease in HDL‐C levels after statin initiation had a moderately increased risk of developing CV events compared with those with constant HDL‐C levels; however, the magnitude of the difference observed between groups was small and not statistically significant until the 5th year of the follow‐up. This study was sponsored by GSK.

Share on: