HIV/HCV-coinfected patients in a French cohort had 60% lower odds of insulin resistance, according to a large longitudinal analysis, with almost half of them having insulin resistance and almost half reporting marijuana use.
The study, which was published in the journal Clinical Infectious Diseases, analyzed data from the HEPAVIH cohort, an ongoing prospective study in France following HIV/HCV-coinfected adults.
In general, people infected with HCV (hepatitis C) run a high risk of insulin resistance and diabetes. Coinfection with HIV compounds that risk. A high proportion of HCV- and HIV-infected people in France use cannabis, especially those infected through needle sharing. General-population research in the U.S. links marijuana use to a lower risk of obesity and diabetes. To see whether such associations hold true in people coinfected with HIV/HCV, HEPAVIH researchers conducted this longitudinal study.
HEPAVIH cohort members complete a questionnaire every 12 months about sociodemographics and behavior, including marijuana use. This analysis focused on cohort members who had one or more study visits in which cannabis use and fasting insulin were both recorded. The researchers used a HOMA-IR (homeostasis model assessment of insulin resistance) cutoff of > 2.77 to indicate insulin resistance, and the analysis excluded people who already had diabetes.
Among 1,324 HEPAVIH members, 703 (53%) met the entry criteria. Two thirds of the study group were men, median enrollment age stood at 44 years, 72% had undetectable HIV viral loads and 72% had a normal weight. At their first study visit, 21% of the participants reported occasional marijuana use in the past four weeks, 12% regular use and 13% daily use. Almost half of the group (46%) had HOMA-IR > 2.77, indicating insulin resistance, while 30% had HOMA-IR > 3.8.
The investigators used two multivariate models to identify independent predictors of HOMA-IR > 2.77. One model included cirrhosis and one excluded cirrhosis. In both models, any marijuana use lowered the odds of insulin resistance 60% (odds ratio [OR]: 0.4, 95% confidence interval [CI]: 0.2 to 0.5 in model with cirrhosis). The association between marijuana use and insulin resistance held true in three sensitivity analyses, one of which included people with diabetes.
The model without cirrhosis identified three other variables independently associated with lower odds of insulin resistance: female gender (OR: 0.4, 95% CI: 0.3 to 0.7), detectable HIV viral load (OR: 0.6, 95% CI: 0.4 to 1.0), and drinking three or more cups of coffee daily (OR: 0.6, 95% CI: 0.4 to 1.0). Two variables independently predicted higher odds of insulin resistance: previous stavudine (d4T, Zerit) use (OR: 1.8, 95% CI: 1.1 to 2.9) and being overweight/obese versus normal weight (OR: 3.2, 95% CI: 1.8 to 5.5).
Why marijuana may lower insulin resistance risk remains undetermined. Cannabis users in this and other studies tend to weigh less than nonusers, and higher weight boosts the risk of insulin resistance, but the insulin resistance association in this study was independent of weight. Some research suggests marijuana may affect insulin resistance risk by modifying adiponectin levels, but the evidence supporting that hypothesis is inconsistent. The researchers suggested evaluating cannabis-based agents in clinical research and practice for possible benefit in preventing insulin resistance.
Mark Mascolini is a freelance writer focused on HIV infection.