Nonfasting and Fasting LDL Equally Predict Mortality in NHANES Analysis

July 11, 2014

NEW YORK, NY — Do patients need to fast before having their cholesterol levels measured? A new analysis suggests maybe not, with researchers reporting that nonfasting LDL-cholesterol levels provide equivalent prognostic value to fasting LDL-cholesterol levels.

Using data from the National Health and Nutrition Survey III (NHANES III), lead investigator Dr Bethany Doran (New York University School of Medicine, NY) and colleagues found that higher nonfasting and fasting LDL-cholesterol levels were associated with an increased risk of all-cause and cardiovascular mortality and that both yielded similar results for the prediction of mortality.

"Based on the clinical guidelines, we've been advising patients to fast before a lipid panel," senior investigator Dr Sripal Bangalore (New York University School of Medicine) told heartwire . "If we see a patient today who had their breakfast or lunch, we ask them to schedule another visit after fasting for eight to 12 hours for the blood test to check the lipid panel. It's a lot of inconvenience for the patient and also for us. It's problematic for patients with diabetes, where we risk hypoglycemia, and in the elderly. In addition, asking patients to come back at a later date may delay starting lipid-lowering therapies, and we may lose patients who simply drop out because they don't have the time to come back again."

Publishing their results online July 11, 2014 in Circulation, the researchers identified 4299 propensity-matched pairs of fasting and nonfasting patients who had their cholesterol levels measured in NHANES III and were followed for a mean of 14 years. LDL-cholesterol levels were stratified by tertiles, with tertile 1, the reference group, including patients with LDL cholesterol <100 mg/dL. For those in tertiles 2 and 3, the LDL-cholesterol levels were >100 to 130 mg/dL and >130 mg/dL, respectively.

Regarding the primary outcome measure, increasing LDL-cholesterol levels were associated with increasing risk of all-cause mortality among those who fasted prior to their lipid panel and those who did not. A test for interaction between fasting status and all-cause mortality was nonsignificant, indicating a lack of association between fasting status and LDL cholesterol with death, say the researchers The C statistic for the prediction of all-cause mortality was 0.59 among those who fasted and 0.58 for the nonfasting group.

Similar results were observed for the secondary outcomes measure of cardiovascular mortality.

"All the guidelines recommendations are based more on expert opinion rather than evidence," said Bangalore. "But we're really looking at the patient when he or she is on their best behavior. We tell them, 'Don't eat your French fries for eight hours and then we'll take your lipid panel.' But that's not what the body is exposed to the majority of the time. What the body is exposed to the majority of the day is the nonfasting state, so this actually makes more sense. I always say fasting is a bit like cramming for an exam. You're staying up late the night before trying to cram and pass the test that way."

In moving to the nonfasting state for measuring lipids, the study suggests that physicians won't lose any of the prognostic value of the tests and it will actually make it easier for patients and physicians alike, added Bangalore. The investigators also found similar prognostic value of nonfasting triglycerides and total cholesterol when compared with fasting values. "This, together with the LDL-cholesterol data, really questions the age-old practice of advising patients to fast before a lipid panel," he said.

Bangalore is on the advisory board for Pfizer. Disclosures for the coauthors are listed in the paper.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.