Tenofovir use is associated with reduced bone mineral density (BMD) in HIV subjects. Trabecular bone score (TBS), an index of bone microarchitecture, determined from grey scale variation in lumbar spine dual energy X-ray images, is also reduced in treated HIV subjects with fracture.  It is unclear what proportion of reduced BMD is due to HIV infection and what proportion is related to treatment. There are limited data on the effects of tenofovir on BMD or TBS in a non HIV population.  The recent introduction of PrEP (pre-exposure prophylaxis) to prevent HIV transmission provides the opportunity to study the effects of tenofovir in a non HIV population.  We examined longitudinal BMD and TBS in a group of HIV-negative men initiating tenofovir-based PrEP.

Thirty one men were enrolled: mean age 37.9 years (SD 9.8), mean lumbar BMD was 1.18 g/cm² (SD 0.16) with mean T score -0.29 (SD 1.3).  Mean TBS was 1.30 (SD 0.11). At 12 months, lumbar BMD decreased by 2% (-0.02 g/cm²) from baseline (95% CI -0.04 - -0.01; P=0.008).  There was no significant change in TBS at 12 months (-0.02, 95% CI -0.06-0.02; P=0.35).

Conclusion: The decrease in BMD in this group of HIV-negative men confirms the negative effect of tenofovir on BMD independent of the effects of HIV. The lack of change in TBS is of uncertain significance. Possible causes include a slower rate of change in TBS compared to BMD, reducing sensitivity to detect a decrease, or predominant cortical bone loss on tenofovir therapy.  Further studies are warranted.