Background
Tuberculosis (TB) causes considerable morbidity and mortality in pregnancy and postpartum, particularly in women living with HIV (WLH), but there are few data from cohorts of women on dolutegravir (DTG). In addition, while risk of active TB generally decreases with increasing body mass, there are few insights into this association during pregnancy and postpartum.
Methods
Consecutive WLH and HIV seronegative women (HIV-) aged >18y and <20w gestational age (GA) attending routine antenatal primary care in Cape Town, South Africa were enrolled in the Obesogenic oRigins of maternal and Child metabolic Health Involving Dolutegravir (ORCHID) study. All WLH were on tenofovir+lamivudine+DTG. Body mass index (BMI) was calculated and categorised as underweight (<18.5 kg/m2), normal (18.5–24.9 kg/m2) or overweight/obese (>25 kg/m2). A correction factor was applied to weight based on GA to estimate pre-pregnancy weight. Follow-up through 24m postpartum included assessment of active TB diagnoses and treatment from routine public sector services based on symptom screening +/- bacteriologic confirmation. GA was by ultrasound; TB events were classified as prevalent (already on treatment at conception) or incident (newly diagnosed during pregnancy or postpartum). TB incidence rates (IRs) were calculated and mixed effects Poisson models were used to assess the association between pre-pregnancy BMI and TB IRs.
Results
Among 1920 women (mean age 28y, median GA at enrolment 13w), 42% were WLH (median duration on ART 58m, IQR 26–98); among WLH 7% had a CD4 <200 cells/uL and 89% had a VL <1000 copies/mL. TB prevalence was 0.4% at conception. TB IRs in WLH and HIV- women were 1,658 (95% CI 942–2920) and 99 (95% CI 14–704) per 100,000 person years, respectively (Table). Adjusting for age, HIV status and CD4 count, education and pregnancy status, TB IRs were higher in WLH with CD4 >200 cells/uL (incidence rate ratio [IRR] 5.0; 95% CI 1.1–22.4) and in WLH with CD4<200 cells/uL (IRR 26.9; 95% CI 2.5–293.9) vs HIV- women. TB IRs were significantly increased if pre-pregnancy BMI was <18.5 kg/m2 vs 18.5–24.9 kg/m2 (IRR 32.6; 95% CI 2.34-455.2). IRs for pre-pregnancy overweight/obese were higher than for normal BMI women but this was not statistically significant (IRR 2.42; 95% CI 0.43–13.53).
Conclusions
TB IRs remained high in WLH despite ART in this setting. Rates appeared higher in underweight vs normal weight women and may also be increased in overweight/obese women however this requires further investigation.
Table or Figure
![]()