Visit-to-visit variability of lipid measurements and the risk of myocardial infarction and all-cause mortality: A prospective cohort study

Background and aims Previous studies suggested that increased visit-to-visit variability in lipid measurements is associated with cardiovascular disease in specific or high-risk populations. Because it is unknown whether this notion applies to the general population, we investigated whether lipid variability has additive effects on the risk of all-cause mortality and myocardial infarction (MI) in the general population. Methods We identified 51,620 subjects from the Kailuan cohort who had no history of MI, stroke and cancer and who underwent ≥3 health examinations from 2006 to 2010. Variability in total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) measurements was measured using the coefficient of variation (CV), standard deviation (SD), variability independent of the mean (VIM), and average real variability (ARV). Cox proportional hazards models were used to calculate the hazard ratios (HRs). Results During a median of 7.03 years of follow-up, 426 (1.21) incidents of MI and 2243 (6.45) incidents of all-cause mortality occurred. In the multivariable-adjusted model, the HRs comparing participants in the highest versus lowest quartile of the CV for HDL-C were 1.56 (95% CI, 1.17–2.08; p for trend<0.01) for MI and 1.22 (95% CI, 1.09–1.37; p for trend<0.01) for all-cause mortality. For the highest quartile compared with the lowest quartile of LDL-C variability, the risk of MI and all-cause mortality increased by 34% (HR, 1.34; 95% CI, 1.03–1.82; p for trend<0.05) and 19% (HR, 1.19; 95% CI, 1.04–1.36; p for trend<0.05), respectively. We did not observe any significant association between TG variability and MI or all-cause mortality. Conclusions These findings suggest that high visit-to-visit HDL-C and LDL-C variability is associated with an increased incidence of MI and all-cause mortality in a Chinese community population.