 |
|
|
| |
High Triglycerides Independently Raise MI Risk in DAD Study
|
| |
| |
17th Conference on Retroviruses and Opportunistic Infections, February 16-19, 2010, San Francisco
Mark Mascolini
High triglycerides in DAD study participants raised the risk of myocardial infarction (MI) 11%, independently of other cardiovascular risk factors, total cholesterol, or high-density lipoprotein (HDL) cholesterol [1]. Because DAD investigators rated this 11% increase "very small," they suggested that fibrates and nicotinic acid "are unlikely to have major impact on the incidence of MI." But not all HIV-heart experts were ready to embrace that conclusion.
Triglycerides can be elevated by antiretroviral therapy, HIV itself, insulin resistance, obesity, and fatty liver disease. But until this DAD analysis, it was unclear whether lofty triglycerides added to MI prognosis after considering total and HDL cholesterol. Furthermore, the DAD team observed, no evidence indicates that trimming triglycerides will ease MI risk in people with HIV.
To address these questions, Signe Worm and DAD colleagues tracked cohort members from enrollment in the study until their first MI, February 1, 2008, or 6 months after the latest clinic visit, whichever came first. The investigators calculated the incidence of first MIs during follow-up according to the latest (time-updated) triglyceride level. To do so, they broke triglyceride levels into six brackets--under 80 mg/dL, 80 to 110, 110 to 150, 150 to 200, 200 to 300, or over 300. (In mmol/L those brackets are under 0.90, 0.90 to 1.25, 1.25 to 1.70, 1.70 to 2.30, 2.30 to 3.45, and over 3.45).
The analysis included 30,703 people with at least one triglyceride reading, 74% of them men and 54% white (but 34% without reported race). Median age stood at 39 years, 37% smoked, 7% had a family history of cardiovascular disease, 2% had a personal history of heart disease, and 3% had diabetes. Three quarters of study participants had taken nucleosides, 60% protease inhibitors, and 34% nonnucleosides.
Through 178,835 person-years of follow-up, the DAD team recorded 580 MIs and 405,756 triglyceride values. Triglycerides were higher in men than women, higher in people currently taking antiretrovirals, higher with a viral load under 500 copies, and lower with a CD4 count under 300 versus over 600.
MI incidence rose linearly from about 0.1 per 1000 person years with triglycerides under 80 mg/dL to about 0.5 per 1000 person years with more than 300 mg/dL. In an unadjusted analysis, every triglyceride doubling raised the MI risk 67% (RR 1.67, 95% confidence interval [CI] 1.54 to 1.80). After adjustment for an array of cardiovascular risk factors, twice-higher triglycerides still inflated MI risk 32% (RR 1.32, 95% CI 1.21 to 1.45). The correlation remained significant at 11% after further adjustment for total and HDL cholesterol (RR 1.11, 95% CI 1.01 to 1.23).
Sensitivity analyses using fasting triglycerides, nonfasting triglycerides, and triglycerides read with an unknown fasting state did not affect the results. The investigators noted that antiretroviral-related triglyceride elevations could not be compared to triglyceride jumps for other reasons. They concluded that, after statistical adjustment, "the residual effect of triglyceride levels . . . of 11% is very small compared with the original unadjusted effect of 67%." Therefore, they proposed that triglyceride-targeting drugs like fibrates and nicotinic acid seem unlikely to affect MI risk.
US researcher Michael Dube, who has studied HIV-related metabolic and cardiovascular abnormalities for years, disagreed with the final conclusion, noting that these drugs may have effects other than direct antitriglyceride activity--for example, on apolipoprotein A. As a result, he urged caution on dismissing use of these agents in people with HIV.
Reference
1. Worm S, Kamara A, El-Sadr W, et al. Triglycerides and the risk of myocardial infarction in the D:A:D study.17th Conference on Retroviruses and Opportunistic Infections. February 16-19, 2010. San Francisco. Abstract 127.
|
| |
|
|
|
|
|
