Safety and efficacy of very low LDL-cholesterol intensive lowering: a meta-analysis and meta-regression of randomized trials

Abstract

Abstract Aims We performed a study-level meta-analysis to provide more robust evidence on safety of very low LDL-cholesterol (LDL-C) levels. Background Concerns on the safety of LDL-C values achieved with potent lipid-lowering therapies have been raised. Methods and results We searched randomized trials reporting clinical outcomes with intensive lipid-lowering treatments leading to very low (<40 mg/dL) LDL-C levels vs. a control group with higher LDL-C levels. Only studies with follow-up duration ≥ 3 months were considered. Primary endpoint was the incidence of various safety measures. A total of 10 randomized trials were overall included, with 38 427 patients being in the very low LDL-C group vs. 70 668 in the control group. Median follow-up duration was 28.8 months. The incidence of all safety outcomes was similar in the two groups: non-cardiovascular death: OR 1.13, 95% CI 0.87–1.45; P = 0.36; any adverse events: OR 1.00, 0.90–1.11, P = 0.94; adverse events leading to drug discontinuation: OR 1.00, 0.87–1.15, P = 0.99; cancer: OR 1.02, 0.95–1.10, P = 0.57; haemorrhagic stroke OR 0.89, 0.66–1.20, P = 0.44; new-onset diabetes: OR 1.16, 0.91–1.47, P = 0.23; neurocognitive disorders: OR 0.97, 0.91–1.04, P = 0.41; haepatobiliary disorders: OR 0.99, 0.83–1.18, P = 0.93; muscle disorders: OR 0.94, 0.77–1.13, P = 0.49; cataract: OR 1.28, 0.78–2.10, P = 0.34. The rates of major adverse cardiovascular events were significantly lower in the very low LDL-C group: OR 0.82, 0.72–0.94, P = 0.005. Conclusion This meta-analysis indicates that very low LDL-C levels on intensive lipid-lowering treatments are not associated with any adverse event and maintain a persistent reduction of cardiovascular events.