Background
There are concerns that dolutegravir (DTG) may be associated with excess gestational weight gain (GWG) in pregnant women living with HIV (WLH) but there are few data, including from Africa.
Methods
The Obesogenic oRigins of maternal and Child metabolic Health Involving Dolutegravir (ORCHID) study is evaluating metabolic outcomes of pregnant WLH receiving tenofovir+lamivudine+DTG (TLD) and HIV-negative (HIV-) pregnant women seeking antenatal care at a primary healthcare facility in South Africa. For this analysis, we included 949 pregnant women (388 WLH, 572 HIV-) enrolled with gestational age (GA) <13 weeks (w) and age >16 years (y). Body composition was assessed using air displacement plethysmography (ADP) and serial standardised anthropometry including triceps, subscapular, and suprailiac skinfold thicknesses to distinguish visceral versus subcutaneous fat distributions. We used latent profile analysis (LPA) to identify adiposity patterns and examined whether these patterns varied by HIV status or duration of DTG use. Finally, we used linear regression to examine whether GWG (change in weight between first and third trimester in kg/week) varied by HIV/DTG.
Results
At enrolment, median age was 27y [Interquartile range (IQR), 24-32]; GA 10w [8-12]; BMI 30 kg/m2 [25-35]; fat-mass index (FMI) 13kg/m2 [9–17]; centripetal fat ratio (CPFR) 52 [47-56]. In WLH median duration of DTG use was 219 days [10-646] and 21% of women who initiated TLD were DTG naïve. LPA identified five distinct adiposity patterns: (i)normal weight/normal central fat, (ii)overweight/high central fat, (iii)obese/normal central fat, (iv)obese/low central fat, and (v)morbid obese/high central fat. While women in pattern (i) were significantly younger, neither HIV status nor DTG duration were different across adiposity patterns after adjusting for maternal age. Among 798 women (84%) with third trimester GWG assessment, GWG patterns varied by HIV and BMI (figure 1): WLH experienced 0.03 kg/week lower rate of GWG compared to HIV- women after adjustment for age, parity, education, GA and BMI at enrolment (95% CI: -0.06 to -0.01, p = 0.02). Restricted to WLH, GWG did not vary by TLD duration (not shown).
Conclusions
These reassuring data suggest that DTG is not associated with GWG in this setting. While the patterns of adiposity identified are specific to this population, these are not associated with HIV/DTG.