Abstract
BACKGROUND: The risk of short-term death for treatment naive patients dually infected with Mycobacterium tuberculosis and HIV may be reduced by early anti-retroviral therapy. Of those dying, mechanisms responsible for fatal outcomes are unclear. We hypothesized that greater malnutrition and/or inflammation when initiating treatment are associated with an increased risk for death. METHODS: We utilized a retrospective case-cohort design among participants of the ACTG A5221 study who had baseline CD4 < 50 cells/mm(3). The case-cohort sample consisted of 51 randomly selected participants, whose stored plasma was tested for C-reactive protein, cytokines, chemokines, and nutritional markers. Cox proportional hazards models were used to assess the association of nutritional, inflammatory, and immunomodulatory markers for survival. RESULTS: The case-cohort sample was similar to the 282 participants within the parent cohort with CD4 <50 cells/mm(3). In the case cohort, 7 (14%) had BMI < 16.5 (kg/m(2)) and 17 (33%) had BMI 16.5-18.5(kg/m(2)). Risk of death was increased per 1 IQR width higher of log(10) transformed level of C-reactive protein (adjusted hazard ratio (aHR) = 3.42 [95% CI = 1.33-8.80], P = 0.011), interferon gamma (aHR = 2.46 [CI = 1.02-5.90], P = 0.044), MCP-3 (3.67 [CI = 1.08-12.42], P = 0.037), and with IL-15 (aHR = 2.75 [CI = 1.08-6.98], P = 0.033) and IL-17 (aHR = 3.99 [CI = -1.06-15.07], P = 0.041). BMI, albumin, hemoglobin, and leptin levels were not associated with risk of death. CONCLUSIONS: Unlike patients only infected with M. tuberculosis for whom malnutrition and low BMI increase the risk of death, this relationship was not evident in our dually infected patients. Risk of death was associated with significant increases in markers of global inflammation along with soluble biomarkers of innate and adaptive immunity.