One third of the variability in HDL-cholesterol level in a large dyslipidaemic population is predicted by age, sex and triglyceridaemia: The Paris La Pitié Study.

OBJECTIVE: The objective of this study was to identify key determinants of high-density lipoprotein-cholesterol (HDL-C) level, including subclinical inflammation and insulin resistance, and to determine the prevalence of a low HDL-C phenotype in dyslipidaemic patients at high cardiovascular risk.

METHODS: In a cross-sectional study, we assessed the prevalence of low HDL-C phenotypes in 14 667 dyslipidaemic patients attending our specialised lipid clinic and evaluated the potential relationships between HDL-C level and 16 clinical and biological parameters.

RESULTS: In univariate analysis, women exhibited higher plasma concentrations of HDL-C as compared with men. Levels of triglycerides, fasting blood glucose, uric acid, waist circumference, body mass index, high sensitivity C-reactive protein (hs-CRP), insulin resistance (as HOMA-IR index) and smoking were all negatively correlated with HDL-C, whereas age was positively correlated with HDL-C levels. Moderate drinkers (10-30 g/day) displayed higher HDL-C concentrations as compared with abstinent subjects; in contrast, consumption of more than 30 g alcohol/day was associated with a further non-significant elevation of HDL-C levels as compared to moderate drinkers. Multivariate analysis identified eight independent correlates of HDL-C. Age, sex and TG accounted for 37% of variability in HDL-C; modifiable factors including waist circumference, alcohol consumption and smoking, in addition to HOMA-IR and hs-CRP, accounted for an additional 5% of the variability in HDL-C. Using a cut-off of 40 mg/dL (1.03 mmol/L) for men and 50 mg/dL (1.29 mmol/L) for women, 33% and 28% of men and women displayed low levels of HDL-C.

CONCLUSION: Eight independent determinants of HDL-C account for 41% of variability in HDL-C in our dyslipidaemic population. Three of them, i.e. age, sex and degree of triglyceridaemia accounted for more than one third of such variability. The high prevalence of low HDL-C phenotypes in dyslipidaemic patients at elevated cardiovascular risk emphasises the need for both lifestyle and pharmacological strategies of intervention to raise HDL-C.