Conference Coverage

Lipid variability predicts cardiovascular events, diabetes onset


Key clinical point: Variability in fasting lipids over time in statin-treated patients is of prognostic importance.

Major finding: More than 30 mg/dL of visit-to-visit variability in triglycerides was independently associated with a 34% increase in risk of coronary events and a 31% increase in cardiovascular events.

Study details: This was a post hoc analysis of the clinical impact of visit-to-visit variability in fasting lipids in 9,572 participants in the randomized, double-blind TNT trial, all of whom were on statin therapy.

Disclosures: The presenter reported serving as a consultant to Pfizer, which sponsored the TNT trial, as well as to several other companies.



ANAHEIM, CALIF.Variability in fasting lipid levels over time in statin-treated patients is of prognostic importance, David D. Waters, MD, reported at the American Heart Association scientific sessions.

More specifically, above-average visit-to-visit variability in fasting triglycerides, LDL cholesterol, or HDL cholesterol in atorvastatin-treated patients with known coronary artery disease proved to be a strong and independent predictor of coronary and cardiovascular events in a post hoc analysis of the landmark Treating to New Targets (TNT) trial (N Engl J Med 2005;352:1425-35).

Dr. David D. Waters of University of California, San Francisco Bruce Jancin/Frontline Medical News

Dr. David D. Waters

Also, high variability in triglyceride and LDL cholesterol levels – but not HDL cholesterol – independently predicted new-onset diabetes, added Dr. Waters, emeritus professor of medicine at the University of California, San Francisco.

The TNT trial randomized more than 10,000 subjects with known coronary artery disease and a baseline LDL cholesterol level below 130 mg/dL to receive either 10 or 80 mg/day of atorvastatin, with fasting lipids measured in a central laboratory at 3 and 12 months, then annually. The trial demonstrated that high-intensity statin therapy was more effective at preventing cardiovascular events than moderate-intensity therapy, thereby ushering in major changes in clinical practice guidelines.

The TNT investigators had previously reported that higher visit-to-visit variability in LDL cholesterol was independently associated with an increased rate of cardiovascular events during the median 4.9 years of study follow-up. In a multivariate regression analysis, each 1 standard deviation increase in average successive variability – that is, the average absolute difference between successive LDL cholesterol values – was associated with a 16% increase in the risk of any coronary event, an 11% increase in risk of any cardiovascular event, a 10% increase in MI, an 17% increase in stroke, and a 23% higher all-cause mortality independent of assignment treatment, achieved LDL cholesterol, demographics, and baseline cardiovascular risk factors.

At the AHA meeting in Anaheim, Dr. Waters presented an expanded analysis of 9,572 TNT participants that incorporated visit-to-visit variability in HDL cholesterol and triglycerides (J Am Coll Cardiol. 2015 Apr 21;65[15]:1539-48). Patients with 1 standard deviation of average successive variability (ASV) in triglycerides – that is, more than 30 mg/dL of visit-to-visit variability – had a 9% increased risk of coronary events during follow-up in a multivariate analysis. Patients with more than 4 mg/dL of variability in HDL cholesterol had a 16% increased risk compared with those with lesser variability.

“For both coronary and cardiovascular events, most of the increased risk appears to reside in the uppermost quintile,” the cardiologist observed.

Indeed, when the investigators divided patients into quintiles of ASV, the top quintile in terms of triglyceride variability had a 34% greater risk of coronary events, a 31% increase in risk of cardiovascular events, a 63% increase in stroke, a 65% increase in nonfatal MI, and a 92% greater likelihood of new-onset diabetes compared with patients in the lowest quintile of ASV. In contrast, these risks were not significantly elevated in the second, third, and fourth quintiles.

Similarly, patients in the top quintile for HDL cholesterol ASV had a 50% greater rate of coronary events, a 56% increased risk of cardiovascular events, a 70% increase in stroke, and a 61% increase in nonfatal MI, compared with those in the lowest quintile. Again, risks weren’t significantly increased in the second through fourth quintiles. Unlike with triglycerides, greater variability in fasting HDL cholesterol over time wasn’t predictive of new-onset diabetes.

Observers noted that these findings could be clinically relevant for patients who remain at high residual risk for atherosclerotic cardiovascular events even after aggressive LDL cholesterol lowering.

Variability in levels of the three lipids was only weakly correlated.

Dr. Waters made a plea to his audience, “The mechanisms accounting for these associations are unknown. If you can suggest for me any possibility of what the causes are, I’d be very happy to hear it and go back to try to verify it.”

He reported serving as a consultant to Resverlogix, CSL Limited, the Medicines Company, Pfizer, and Sanofi-Aventis.

Next Article: