Changes in Inflammatory Marker Concentrations during Randomized Feeding in Women Infected with HIV-1 Subtype C
Elizabeth Russell1,2, T Mohammed2, B Jorowe2, L Smeaton1, R Hoffman3, J Currier3, S Moyo2, M Essex1, and S Lockman1,2,4
1Harvard Sch of Publ Hlth, Boston, MA, US; 2Botswana-Harvard AIDS Inst Partnership, Gaborone; 3Univ of California, Los Angeles, US; and 4Brigham and Women`s Hosp, Boston, MA, US
Background: Limited data exist on the association between pregnancy/breastfeeding and inflammation in HIV+ women.
Methods: To estimate changes in inflammatory markers postpartum and assess the association of these markers with feeding strategy, we measured C-reactive protein (CRP), D-dimer, IFN-ϒ, IL-6, IL-10, and TNF-α at 34 weeks gestation and again at 6 months postpartum in HIV+ women randomized to breastfeeding (n = 86) vs formula feeding (n = 75), randomly selected from the Botswana Mashi PMTCT trial: >30% of IL-10, TNF-α and IFN-ϒ results were below the lower limit of detection of the assay. Outcomes were analyzed as detected/undetected. CRP, D-dimer, and IL-6 were natural log transformed. Groups were compared with either c2 or Wilcoxon Rank-Sum tests. Changes between formula feeding and breast feeding subcohorts over time were compared using regression models including the following covariates: baseline marker level, baseline plasma HIV-1 viral load, age, clinic site, and education level. ART was initiated between time points for 9% breast-feeding and 12% formula-feeding women (c2 p = 0.56).
Results: For all participants, CRP and D-dimer significantly decreased between the thirrd trimester and 6 months postpartum, while IL-6, IFN-ϒ, and IL-10 significantly increased. TNF-α concentrations remained unchanged. No significant longitudinal changes in concentration of IL-6 or IL-10 were seen in breast-feeding women, yet there were significant increases in formula-feeding women. When testing the association of these changes with feeding strategy, IL-10 changes were not significantly different between breast feeding and formula feeding (p = 0.09), yet were significantly different for IL-6 (p = 0.003).
Conclusions: Breastfeeding may provide some level of immune tolerance in HIV+ pregnant women by damping IL-6 responses. Whether this is maintained over time, or influences disease progression, warrants further investigation.
a BD=below limit of detection