Abstract:
Sub-Saharan Africa accounts for high tuberculosis cases that result from
widespread HIV infections, which is exacerbated by injection substance use.
Immunologically, HIV critically impairs cell-mediated host responses to
Mycobacterium tuberculosis. IFN-γ, IL-10 and Acrp30 are key mediators of
systemic inflammation. Although circulating IFN- and IL-10 levels are
increased, Acrp30 levels are lowered and associated with disease severity among
HIV and TB co-infected non-susbstance users. In contrast, circulating IFN- and
Acrp30 levels are decreased while IL-10 levels are upregulated among injecting
heroin addicts. However, no studies to date have reported on these cytokine
profiles among Kenyan HIV-1 and TB co-infected injection drug users. This
study, therefore, investigated plasma IFN-γ, IL-10 and Acrp30 levels among
IDUs, and their association with CD4+ T cell counts, HIV-1 viral load and BMI.
A cross-sectional study was conducted from August, 2012-November, 2013 using
138 participants recruited at Bomu hospital; a major centre for rehabilitation of
drug and substance users in Mombasa County. Following informed consent, IDUs
were enrolled through respondent driven sampling, snowball and makeshift
methods while convenience and purposive sampling were used for recruiting the
control group. IDUs and controls were screened for HIV and TB respectively
through Determine™ and Bioline™ rapid tests, and Ziehl Neelsen stained sputum
smears. Subsequently, the study participants were categorised into: HIV-1/TB coinfected ART-naive (n=9) and -experienced (n=27); HIV-1 mono-infected ARTnaive (n=26) and -experienced (n=13); TB mono-infected (n=21), HIV-1 negative
and TB uninfected (n=25) IDUs and controls (n=17). Demographic, drug use
information and physical measurements were recorded using assisted interviews.
EDTA venous blood samples were collected and used for preparing plasma and
enumerating CD4+ T cell counts. Frozen plasma samples were used for
determining cytokine concentrations, and HIV-1 viral load. CD4+ T cell counts
were enumerated using flow cytometry; cytokine levels were measured using a
sandwich ELISA technique, while HIV-1 viral load was determined by RT-PCR,
respectively. Across-group comparisons in continuous data were performed using
Kruskal Wallis followed by post-hoc Dunn’s tests. Plasma IFN-γ (P<0.0001), IL10 (P<0.0001) and Acrp30 (P=0.006) levels differed significantly across groups.
IFN-γ levels were high in co-infected ART-naive (P<0.001) and -experienced
(P<0.001), and HIV-1 mono-infected ART-experienced (P<0.001) IDUs relative
to healthy controls. IL-10 levels were elevated in uninfected IDUs (P<0.001)
compared to healthy controls. Acrp30 levels were lower in TB mono-infected
(P<0.01) relative to controls. IFN-γ/IL-10 ratio varied across-groups (P<0.0001)
and higher in co-infected ART-naive (P<0.001) and -experienced (P<0.001), and
HIV-1 mono-infected ART-experienced (P<0.001) compared to uninfected IDUs.
The IFN-γ/Acrp30 ratio also differed across groups (P<0.0001) with HIV-1
mono-infected ART-experienced (P<0.001), and co-infected ART-naive (P<0.001) and -experienced (P<0.001) IDUs exhibiting higher ratio relative to
uninfected IDUs. CD4+ T cells correlated inversely with Acrp30 (=-0.717,
P=0.030) levels in TB mono-infected IDUs whereas BMI correlated positively
with Acrp30 (=0.523, P=0.022) among co-infected ART-naive IDUs,
respectively. Altogether, circulating IFN-, IL-10 and Acrp30 production is
altered in ART-naive and -experienced HIV-1 and TB co-infected IDUs,
suggesting a role as disease markers in HIV and TB co-infection among IDUs.