Background: The outcome of CMV/HIV co-infection in infants treated early with combined antiretroviral therapy (cART) in resource-limited settings has not been described. We aimed to estimate the prevalence and identify factors associated with early CMV infection in HIV-infected and non-infected infants included in a study in Cameroon, and to compare HIV disease progression and survival after 1 year of early cART, following infants' CMV status.
Methods: HIV-infected infants followed from birth or from HIV diagnosis before 7 months old and HIV-uninfected infants born to HIV-infected or uninfected mothers were tested for CMV at a median age of 4.0 months [Interquartile range (IQR): 3.4-4.9]. Multivariable logistic regression was performed to identify factors associated with CMV infection. Early cART was offered to HIV-infected infants: mortality, immunological and virological outcomes were assessed.
Results: Three hundred and sixty-nine infants were tested. The proportion of infants infected with CMV at baseline was significantly higher in HIV-infected than in HIV-uninfected groups (58.9% (86/146) vs 30.0% (67/223), p < 0.001). At baseline, median CMV viral load was higher in HIV-infected (3.7 log copies/ml [IQR; 3.1-4.3]) than in HIV-uninfected infants (2.8 log copies [IQR; 2.1-3.4], p < 0.001). cART was initiated in 90% of HIV-infected infants (132/146) at a median age of 4.0 months (IQR; 3.2-5.9); in this sub-group CMV infection was independently associated with being followed from the time of HIV diagnosis rather than from birth (aOR = 3.1, 95%CI [1.2-8.0]), born to a non-single mother (aOR = 3.4[1.4-8.1]), and breastfeeding (aOR = 7.3 [2.7-19.4]). HIV-infected infants were retested after a median of 7.1 months [4.8-9.5]: CMV was undetectable in 37 of the 61 (60.7%) initially CMV-infected cases and became detectable in 8 of the 38 (21.1%) initially CMV-negative cases. After 1 year of cART, the probability of death (0.185 vs 0.203; p = 0.75), the proportion of cases with HIV RNA viral load <400 copies/ml (75.5% vs 61.5%; p = 0.17) and the mean CD4 percentage increase (10.97% vs 6.88%; p = 0.15) did not differ between CMV+ and CMV- infants.
Conclusions: We observed a high prevalence of CMV infection among HIV-infected infants. Early initiation of cART may have limited the negative impact of CMV even in the absence of specific anti-CMV treatment.
Keywords: Co-infection; Cytomegalovirus; Early treated HIV-infected infants; Human Immunodeficiency Virus.