Markers of Microbial Translocation in Hiv-infected and Hiv-negative Adults in a Kenyan Urban Centre: a Cross-sectional Comparative Study

Abstract

The incidence of non-communicable diseases in persons infected with HIV has been on the increase, due to persistent chronic immune activation and inflammation despite effective HAART. Passage of microbes from the gastrointestinal system across a damaged epithelium, termed microbial translocation (MT), is believed to be a key factor in the continued inflammation. The effect of MT on the immune system is influenced by environmental and genetic factors, and, if present, is a potentially modifiable driver of the chronic immune activation. This study sought to determine the presence of MT using indirect markers lipopolysaccharide (LPS) and soluble CD14 (sCD14) in HIV-infected adults, with HIV-uninfected age-matched participants as a comparative population. METHODS This was a 1:1 age-matched cross-sectional comparative survey conducted at the Mbagathi District Hospital Comprehensive Care (CCC) and Voluntary Testing and Counselling (VCT) centres, and the Kenyatta National Hospital Voluntary Counselling Centre. HIV-infected persons were drawn from adults attending the CCC, while HIV-uninfected persons were matched in age to within five years of the test population, and drawn from attending the VCT centres. Presence of MT was assessed using LPS and sCD14 levels in plasma. Focused medical history that included CD4 and plasma viral load levels, and drug history for the HIV-infected population was obtained from the patients’ medical charts. Statistical analysis was done on IBM® SPSS® Statistics v23. Mean levels of LPS and sCD14 from the two groups were compared using the Wilcox signed-rank test. A p value of ≤ 0.05 was considered significant. RESULTS Ninety-four participants were included in the primary analysis, 47 HIV-infected and 47 HIV-uninfected. The age ranged from 21–61 years for all participants, with a mean age of 47.5±9.0 years for the HIV-infected group and 47.7±9.8 years in the HIV-uninfected group. Female participants were 51.1% in the HIV-infected group and 66% in the HIV-uninfected group. All the HIV-infected participants were on HAART, with a mean 12.9±3.6 year duration of use, with 91% (n=46) of them having an undetectable HIV viral load.

Mean plasma LPS for the HIV-infected group was 0.95±0.43 EU/mL compared to 0.98±0.42 EU/mL for the HIV-uninfected. There was no statistically significant difference between the two groups (p=0.38). Mean plasma sCD14 values were 447.74±360.22 pg/mL (median 316 pg/mL; IQR 250.40–441.83) and 797.19±364.27 pg/mL (median 742 pg/mL; IQR 541.53–981.65) for the HIV-infected and HIV-uninfected, respectively. The HIV-uninfected group had significantly higher sCD14 levels than the HIV-infected group (p=0.00). The HIV-uninfected group showed greater variability than the HIV-infected group for both LPS and sCD14 levels. There was no correlation between LPS and sCD14 levels in either group. CONCLUSION While this study demonstrated detectable indirect markers of microbial translocation in both HIV-infected and HIV-uninfected groups, HIV infection was not associated with higher levels of either LPS or sCD14.