LDL cholesterol as a predictor of mortality, and beyond: to fast or not to fast, that is the question?

T he necessity of a fasting lipid panel for assessment of atherosclerotic cardiovascular disease (ASCVD) risk has been assumed now for decades. The rationale for this position has been the estimation of levels of low-density lipoprotein cholesterol (LDL-C), a major risk factor for ASCVD, by correcting for nonfasting changes in plasma triglycerides using the so-called Friedewald equation developed during the 1960s at the then National Heart Institute. 1 Because LDL-C determination by ultracentrifugation is time consuming and expensive and direct LDL-C measurements have proven no more accurate than the Friedewald equation, the calculation of LDL-C has remained more than adequate, and was recently endorsed by the 2013 American College of Cardiology (ACC)/American Heart Association (AHA) Risk Assessment Work Group 2 and used in the 2013 ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Disease Risk in Adults. 3 The equation estimates LDL-C after measuring total cholesterol, high-density cholesterol (HDL-C) and triglycerides (TG), and then subtracting from the total cholesterol concentration the HDL-C plus very-low-density lipoprotein cholesterol (VLDL-C), which is estimated by TG/5. This formula is not used when plasma TG >400 mg/dL because in this metabolic setting VLDL particles contain relatively less cholesterol and VLDL-C contribution is therefore overestimated and the calculated LDL-C is falsely low. The scientific rationale for addressing the importance of fasting versus nonfasting is that food consumption results in variable increases in plasma TG. Here the effect is not only that of dietary fat but also carbohydrate. Carbohydrate intake increases VLDL TG secretion within 30 to 90 minutes after ingestion, whereas fat intake in normotriglyceridemic people increases plasma TG via chylomicron assembly and secretion from the small intestine over an interval up to 8 hours. Moreover, even without eating, fasting triglycerides are subject to substantial biological variability (eg, up to 30%). 4 This necessity of fasting to estimate LDL-C risk for all cause mortality and CVD mortality has now been challenged by Doran et al 5 in this issue of Circulation. Their report provides convincing evidence over a long interval of 14 years in middle-aged U.S. men and women that nonfasting versus fasting LDL-C values provide equal predictability for all-cause mortality and CVD mortality (ASCVD + congestive heart failure). Of course congestive heart failure has other causes than coronary heart disease, and it would have been instructive if analyses excluding subjects with congestive heart failure would have been included in the article. …