Background:
Dolutegravir (DTG) use has been associated with increased risk for weight gain. We have previously demonstrated that ART-naïve patients starting DTG in Kenya gain significantly more weight compared to those starting a Non-Nucleoside Reverse Transcriptase Inhibitor (NNRTI). In this study we determined the impact of switching from NNRTI, efavirenz (EFV) or nevirapine (NVP), to DTG-containing ART on weight gain in a low-income east African country with high prevalence of HIV and recent large-scale roll-out of DTG.
Methods:
Participants enrolled in the Kenyan Academic Model Providing Access to Healthcare program who had been on NNRTI for at least 24-months prior to switching to DTG were included in the analysis. We excluded participants who switched due to virologic failure, women who were pregnant within 2 years of switching, and participants with missing BMI data at time of switch. Weights within 18-month of switch were included in the longitudinal models of weights over time and were log transformed to identify linear trends pre- and post- switch. Weights over the follow-up period were modeled using piecewise linear mixed effect models.
Results:
23,131 participants met our inclusion criteria with 52% females, 28% with BMI >25 kg/m2, 71% switching from EFV, and 29% switching from NVP. At the time of switch, the mean age was 51 ± 10 years, the mean CD4 count was 201 ± 165 cell/mm3, and the mean BMI was 23 ± 4 kg/m2. Compared to males, females were older (52 vs. 49) years and had higher BMI (24 vs. 22) kg/m2 at the time of switch. Participants gained, on average, 1.36 ± 5.7 kgs during the entire study period at an average rate of 0.59 kg/year. The rate weight increase was significantly higher post-switch compared to pre-switch (0.79 vs. 0.44 kg/year, p< 0.0001) (1A). The rate of weight increase post-switch was higher for females compared to males (0.96 kg/year vs 0.62 kg/year, p< 0.0001) (1B), and for participants switching from EFV compared to NVP (1.12 kg/year vs. 0.002 kg/year, p< 0.0001) (1C).
Conclusions:
In a large HIV cohort from east Africa, on stable NNRTI treatment, switching to DTG-based regimens was associated with a greater rate of weight gain compared to pre-switch. Despite having greater BMI at time of switch, females had greater weight gain post-switch compared to males. Weight gain was predominantly found in those switching from EFV, thus suggesting EFV is more weight suppressive than NVP.
Figure 1. Changes in weight over time among all HIV patients switching from NNRTI to DTG (A). Weight changes by sex and by baseline NNRTI are shown in (B) and (C) respectively. 