Paradoxical tuberculosis immune reconstitution inflammatory syndrome (TB-IRIS) in HIV patients with culture confirmed pulmonary tuberculosis in India and the potential role of IL-6 in prediction

Abstract

Background: The incidence, manifestations, outcome and clinical predictors of paradoxical TB-IRIS in patients with HIV and culture confirmed pulmonary tuberculosis (PTB) in India have not been studied prospectively.

Methods: HIV+ patients with culture confirmed PTB started on anti-tuberculosis therapy (ATT) were followed prospectively after anti-retroviral therapy (ART) initiation. Established criteria for IRIS diagnosis were used including decline in plasma HIV RNA at IRIS event. Pre-ART plasma levels of interleukin (IL)-6 and C-reactive protein (CRP) were measured. Univariate and multivariate logistic regression models were used to evaluate associations between baseline variables and IRIS.

Results: Of 57 patients enrolled, 48 had complete follow up data. Median ATT-ART interval was 28 days (interquartile range, IQR 14-47). IRIS events occurred in 26 patients (54.2%) at a median of 11 days (IQR: 7-16) after ART initiation. Corticosteroids were required for treatment of most IRIS events that resolved within a median of 13 days (IQR: 9-23). Two patients died due to CNS TB-IRIS. Lower CD4(+) T-cell counts, higher plasma HIV RNA levels, lower CD4/CD8 ratio, lower hemoglobin, shorter ATT to ART interval, extra-pulmonary or miliary TB and higher plasma IL-6 and CRP levels at baseline were associated with paradoxical TB-IRIS in the univariate analysis. Shorter ATT to ART interval, lower hemoglobin and higher IL-6 and CRP levels remained significant in the multivariate analysis.

Conclusion: Paradoxical TB-IRIS frequently complicates HIV-TB therapy in India. IL-6 and CRP may assist in predicting IRIS events and serve as potential targets for immune interventions.

Figure 1. Associations between pre-ART clinical and laboratory characteristics with subsequent TB-IRIS events.

Baseline (pre-ART) characteristics significantly different between patients that developed TB-IRIS events in the follow up and those who did not were assessed to test associations with risk for IRIS in univariate and multinomial logistic models. Relative risks (RR) are for values below or above the threshold levels displayed, which were estimated close to median values for the overall study population. Adjustment was performed for all variables presented and also included age and gender. 95% CI, 95% confidence interval.

Figure 2. Pre-ART plasma levels of IL-6 or CRP distiguish individuals at higher risk for TB-IRIS.

(A) Pre-ART plasma concentrations of IL-6 (left panel) and CRP (right panel) were compared between individuals who developed TB-IRIS during study follow up and those who did not using the Mann-Whitney test. (B) Correlation between IL-6 and CRP values was tested in both groups of patients using the Spearman test. In (B), dotted lines represent the median pre-ART values of IL-6 or CRP in the entire study population. Gray areas highlight those patients displaying values for both biomarkers above their respective median values within the study population. (C) The associations between systemic levels of IL-6 and CRP with risk for subsequent TB-IRIS were assessed by univariate and multivariate models. Relative risks (RR) are per standard deviation increase after log₁₀ transformation. RR were adjusted for baseline age, gender, weight, hemoglobin, hematocrit, sputum culture grade, presence of extra pulmonary TB, presence of miliary TB, days to ART initiation, plasma HIV RNA levels and CD4⁺ T-cell count. CI, confidence interval.