Background:  Breast-feeding accounts for up to a third of mother-to-child transmission of HIV. Breast inflammation may contribute to HIV shedding and risk of mother-to-child transmission. Current interventions include maternal single-dose nevirapine (NVP) and NVP infant prophylaxis. Our objectives were to determine the relationship between shedding of HIV RNA in breast milk and mastitis among women receiving single-dose NVP in the HTPN023 clinical trial.

Methods:  We studied 34 women from Chitungwiza, Zimbabwe, who received single-dose NVP at labor. HIV-1 RNA in plasma and breast milk at 2 and 8 weeks post-NVP were quantified by the ultra-sensitive Roche1.5 Amplicor Assay. By concentrating virus from 1 to 2 mL of lactoserum, the lower limit of detection was 12.5 to 22.5 copies/mL.  Electrolytes associated with mastitis (Na⁺ >12 mmol/mL and Na⁺/K⁺) and other breast milk electrolytes (Cl^-, Ca⁺⁺, and Mg⁺⁺) were analyzed by ion-selective electrodes (Stanford Clinical Lab). Correlations between electrolytes and HIV levels were performed on all breast milk samples and one randomly selected breast milk sample (R or L) from each woman. Statistical analyses were conducted on the SAS program. 

Results:  Median plasma RNA was 4.3 log copies/mL and 4.5 log copies/mL at 2 and 8 week, respectively. Breast milk HIV-1 RNA > 50 copies/mL increased significantly from 9/32 (27%), median 1.41 log copies/mL, at 2 weeks to 22/34 (66%), median 2.5 log copies/mL, at 8 weeks (paired Rank Sum, p <0.01). There was a significant correlation (Spearman’s r = 0.52, p <0.01) between 2 and 8 weeks breast milk and plasma HIV RNA. Elevated Na⁺ (>12 mmol/L), was detected in 24/57 (42%) of 2 weeks and 24/59 (41%) of 8 weeks breast milk samples, without significant correlations between 2 weeks HIV RNA and breast milk electrolytes (p >0.05).  However at 8 week, there was a significant correlation between breast milk HIV RNA and Na⁺ and Na⁺/K⁺ (Spearman’s r = 0.53 and 0.47, respectively, p <0.01). At 8 weeks, median breast milk HIV RNA was 698 copies/mL (2.84 log copies/mL) among women with elevated breast milk Na⁺ vs 25 copies/mL (1.40 log copies/mL) among women with NA⁺ < 12 mmol/L (p <0.05).   

Conclusions:  Sub-clinical mastitis or inflammation, as indicated by elevated breast milk Na⁺ was found in ~42% of women at 2 and 8 weeks postpartum.  Significantly higher breast milk HIV-1 RNA in samples with elevated Na⁺ at 8 weeks provides evidence that sub-clinical mastitis is associated with increased HIV RNA shedding and infant exposure. Reducing sub-clinical mastitis and viral shedding may increase the safety of breast-feeding among HIV⁺ women post single-dose NVP.

Keywords: Nevirapine; Mastitis; Breast milk HIV RNA