LDL cholesterol after starting statins and cardiovascular events in primary prevention

Background

The cardiovascular and mortality benefits of achieving guideline-recommended concentrations of low-density lipoprotein (LDL) and non-high-density lipoprotein (non-HDL) cholesterol with statins in primary prevention remain uncertain.

Methods and findings

In this new-user cohort study, we used electronic healthcare databases from Hong Kong (HK), New Zealand (NZ) and the United Kingdom (UK) to identify 185,237 patients initiating statins between 1 January 2007 and 31 December 2012 (with follow-up to 31 December 2017) without atherosclerotic cardiovascular disease, heart failure, or chronic kidney disease. Associations between LDL and non-HDL cholesterol measured within 2 years post-initiation with the outcomes of major adverse cardiovascular events (MACE) or all-cause mortality were assessed in landmark analyses using multivariable Cox regression adjusted for age, sex, hypertension, atrial fibrillation, diabetes mellitus, dyslipidaemia, obesity, smoking, alcohol misuse, rheumatoid arthritis, systemic lupus erythematosus and connective tissue disorders, severe mental illness, antihypertensive drugs, antidiabetic drugs, antiplatelet drugs, and anticoagulant drugs. Estimates were pooled using random-effects meta-analysis. At 2 years, LDL cholesterol was <1.4 mmol/L in 7,084 (3.8%), 1.4–1.7 mmol/L in 21,893 (11.8%), 1.8–2.5 mmol/L in 81,941 (44.2%) and ≥2.6 mmol/L in 74,294 (40.1%) patients. Mean duration of follow-up for the primary outcome of MACE was 6.5 years in HK, 5.6 years in NZ and 5.4 years in UK. versus an LDL cholesterol ≥2.6 mmol/L, adjusted hazard ratios (HR) for MACE were 0.80 (95% CI [0.77,0.83]) for 1.8–2.5 mmol/L, 0.79 (0.66,0.94) for 1.4–1.7 mmol/L and 0.82 (0.60,1.12) for <1.4 mmol/L. Non-HDL cholesterol showed similar estimates for MACE (e.g., HR 0.75, [0.72,0.79] for 2.6–3.3 mmol/L versus ≥3.4 mmol/L). In continuous analysis, the association with MACE was approximately log-linear across achieved LDL and non-HDL cholesterol in HK and the UK, whereas in NZ estimates were less consistent and imprecise at the lowest concentrations. After adjustment for confounding factors, very low LDL (HR 1.06, [0.96,1.16]) and non-HDL cholesterol (HR 0.98, [0.84,1.14]) were not associated with all-cause mortality. Limitations include heterogeneity across cohorts, which may reflect confounding by indication where treatment intensity was guided by cardiovascular risk, and residual confounding from comorbidities, lifestyle interventions and disease severity. Estimates are also conditional on patients having initiated statins, survived to the landmark, and undergone follow-up lipid testing.

Conclusions

Lower achieved LDL and non-HDL cholesterol after statin initiation was associated with a reduction in MACE, with limited evidence of an increase in all-cause mortality at very low concentrations after adjustment for confounding. Additional reductions in cardiovascular events could be obtained by achieving guideline-recommended non-HDL cholesterol levels of at least <3.4 mmol/L after statin initiation for primary prevention.