20th International AIDS Conference - Melbourne, Australia


WEAB0103 - Oral Abstract

Impact of short-term change in body mass index (BMI) after antiretroviral therapy (ART) initiation on subsequent risk of cardiovascular disease (CVD) and diabetes in HIV-positive individuals: the D:A:D study

Presented by Amit C Achhra (Australia).

A.C. Achhra1, A. Mocroft2, P. Reiss3, C. Sabin2, L. Ryom4, S. de Wit5, C. Smith2, A.D. Monforte6, A. Phillips2, R. Weber7, J. Lundgren8, M.G. Law1, The D:A:D Study Group

1Kirby Institute for Infection and Immunity in Society, Sydney, Australia, 2University College London, Infection & Population Health, London, United Kingdom, 3University of Amsterdam, Division of Infectious Diseases and Department of Global Health, Amsterdam, Netherlands, 4University of Copenhagen, Copenhagen, Denmark, 5Centre Hospitalier Universitaire Saint-Pierre, Brussels, Belgium, 6University of Milan Clinic of Infectious Diseases, Milan, Italy, 7University Hospital in Zurich, Zurich, Switzerland, 8Rigshospitalet & University of Copenhagen, Copenhagen, Denmark

Background: ART initiation often results in weight gain. Increased BMI could be detrimental to future health. We modelled the relationship between short term change in BMI after ART initiation and subsequent risk of CVD and diabetes.
Methods: We analysed data from the D:A:D study, a large collaborative multi-national cohort study with well-validated outcomes. The outcomes were new CVD (myocardial infarction/ invasive cardiovascular procedures/ stroke) and diabetes events. We included treatment naive patients initiating ART who had BMI measurements available prior to (within 12 months) and after initiation of ART (1year±6 months post-ART). Patients with CVD or diabetes at study entry were excluded. BMI (weight(kg)/height(meter)2 was categorised as under-weight (< 18.5), normal (18.5-25), over-weight (25-30) and obese(>30). The main covariable of interest was change in BMI (1year post-ART BMI minus pre-ART BMI). Time at risk commenced at 1 year post-ART. Poisson regression models were used to assess the risk of outcomes due to one-year change in BMI, adjusted for and stratified by pre-ART BMI category (to give a category-specific slope), cohort, established risk factors for each outcome (time-updated where possible), and calendar year.
Results: There were 97 CVD events in 43982 person-years (N=9321) and 125 diabetes events in 43278 person-years (N=9193). A majority were male (75%) reporting homosexual transmission (50%) and with a normal pre-ART BMI (64%). The proportion of overweight/obese patients increased from 29.6% pre-ART to 35.5% post-ART. People with baseline BMI in the overweight/obese category tended to have a higher risk of each outcome compared to those in the normal category. The risk of outcome per unit change in BMI statistically significantly varied by baseline BMI category (P< 0.05). For CVD, the relative-risk (RR) (95% confidence-interval, P-value) per unit increase in BMI by baseline BMI category were: underweight: 0.92(0.54-1.56, 0.792), normal: 1.18(1.04-1.32, 0.010), overweight: 0.80(0.62-1.02, 0.070) and obese: 1.08(0.71-1.64, 0.714). Similarly, for diabetes, the estimates were: underweight: 1.16 (0.88-1.53, 0.296), normal: 1.19 (1.06-1.33, 0.003), overweight: 1.22 (1.06-1.40, 0.006) and obese: 1.01(0.85-1.2, 0.941).
Conclusions: Short-term gain in BMI following ART initiation appeared to be associated with a differing risk of CVD and diabetes outcomes depending on the pre-ART BMI.

Back to the Programme-at-a-Glance