Fasting may not be necessary before lipids testing, study finds

In this post hoc follow-up analysis of a randomized clinical trial, triglyceride levels were slightly higher in nonfasting samples than fasting samples, and cholesterol levels in both types of samples were similar.


Measurement of lipids appears to yield the same results regardless of whether patients fast beforehand, according to a recent study.

In a post hoc prospective follow-up of a randomized clinical trial in which fasting and nonfasting lipid levels were measured in the same patients before randomization, researchers compared the association of each type of measurement with coronary and vascular outcomes. They also evaluated whether using nonfasting lipid measurements would misclassify risk in patients who were considered eligible for statin therapy.

The primary endpoint of the original trial, the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid Lowering Arm (ASCOT-LLA), was major coronary events, defined as nonfatal myocardial infarction and fatal coronary heart disease, and participants were randomly assigned to receive atorvastatin calcium, 10 mg/d, or placebo. Secondary analyses looked at an expanded endpoint of atherosclerotic cardiovascular disease (ASCVD) events, defined as fatal or nonfatal myocardial infarction, fatal or nonfatal stroke, or death from ASCVD. The current study was supported by an investigator-initiated institutional research grant from Pfizer, as well as grants from the National Heart, Lung, and Blood Institute and awards from the National Institute for Health Research and the Healthcare National Health Service Trust. Pfizer also provided financial support for ASCOT. The results of the current study were published May 28 by JAMA Internal Medicine.

Overall, 8,270 patients from the ASCOT-LLA were included in the current study, 6,855 of whom had no previous vascular disease. Most of the patients (82.1%) were men; the mean age was 63.4 years. Nonfasting and fasting lipid levels were measured four weeks apart. ASCOT-LLA was terminated early after a median 3.3 years of follow-up because the risk reduction with atorvastatin calcium versus placebo was highly significant.

Triglyceride levels were slightly higher in nonfasting samples than fasting samples, and cholesterol levels in both types of samples were similar. Associations between coronary events and nonfasting lipid levels were similar to those seen with fasting lipid levels. Among all patients, adjusted hazard ratios were 1.32 (95% CI, 1.08 to 1.61) per 40-mg/dL of LDL cholesterol with nonfasting levels and 1.28 (95% CI, 1.07 to 1.55) for fasting levels (P=0.007 and P=0.008, respectively).

When patients with preexisting vascular disease were excluded, adjusted hazard ratios were 1.42 (95% CI, 1.13 to 1.78) and 1.37 (95% CI, 1.11 to 1.69), respectively (P=0.003 for both comparisons). There was a very high concordance between fasting and nonfasting lipid levels and classification into ASCVD categories, 94.8% when the 2013 American College of Cardiology/American Heart Association pooled risk cohort equations were used and 98.6% when the QRISK2 algorithm was used.

The authors noted that most of the patients in their study were white and that the inclusion criteria might limit the findings' generalizability. However, they concluded that the association with cardiovascular risk was similar with nonfasting and fasting lipid levels and that measurements of lipid levels were highly correlated with category of cardiovascular risk regardless of patients' fasting status. Recent changes in U.S. guidelines have recommended broader adoption of nonfasting lipids for routine assessment of cardiovascular risk, the authors noted.

“The results of this study suggest that such a strategy would be highly effective and offer many advantages for cardiovascular risk screening and treatment decisions, including for initiating statin or antihypertensive therapy,” they wrote.