Objectives: To assess the effects of antiretroviral therapy (ART) on bone mineral density (BMD) DESIGN: Randomized comparison of continuous ART (viral suppression group; VS) with intermittent ART (drug conservation group; DC) SETTING: Outpatient clinics in the United States, Australia, and Spain.
Participants: Participants in the Strategies for Management of Antiretroviral Therapy (SMART) Body Composition substudy.
Main outcome measures: Annual hip and spine BMD by dual-energy radiographic absorptiometry (DXA) and spine BMD by quantitative computed tomography (qCT).
Methods: Comparisons were by intention-to-treat analysis, using longitudinal models for change in BMD. Risk factors for BMD loss were evaluated.
Results: The 214 participants (median 44 years, 19% female participants, 73% on ART; median T-scores -0.5 total hip, -0.7 spine DXA, -0.9 spine qCT; 98 randomized to VS and 116 to DC) were followed for a mean 2.4 years. With continuous ART, BMD declined per year by 0.8% (hip), 0.4% (spine DXA), and 2.4% (spine qCT). BMD declined significantly less with intermittent ART. Estimated DC minus VS group differences in mean BMD change through follow-up were 1.4% [hip; 95% confidence interval (CI) 0.6-2.3; P = 0.002], 1.3% (spine DXA; 95% CI 0.1-2.4, P = 0.03), and 3.0% (spine qCT; 95% CI 0.8-5.2, P = 0.007). No consistent drug-specific association with BMD decline was found. In the parent study, 10 of 2753 participants in the VS group and two of 2720 in the DC group reported serious fractures (hazard ratio 4.9; 95% CI 1.1-22.5; P = 0.04).
Conclusion: Continuous ART is associated with decline in BMD and possibly more fractures relative to intermittent, CD4 cell count-guided ART.
Trial registration: ClinicalTrials.gov NCT00027352.