ABSTRACT
INTRODUCTION: The issue of whether integrase inhibitors (INSTIs) may confer a higher risk of paradoxical tuberculosis-related immune reconstitution inflammatory syndrome (TB-IRIS) compared with other classes of antiretroviral in people with HIV with a profound level of immunosuppression remains insufficiently explored. We aimed to assess whether such a higher risk exists by examining a cohort of patients with TB-HIV initiating antiretroviral therapy (ART) in Hong Kong. METHODS: This was a retrospective review of 133 patients registered in the TB-HIV Registry of the Department of Health during the period 2014-2021. RESULTS: Sixteen of 70 patients (22.9%; 95% confidence interval [CI] 13.0-32.7) and 14 of 63 patients (22.2%; 95% CI 12.0-32.5) from the INSTI and non-INSTI groups experienced TB-IRIS (p = 0.920). The median intervals between ART initiation and IRIS among patients from the two groups were similar (3 weeks [interquartile range IQR 2.0-7.8] vs. 4 weeks [IQR 2.0-5.1], p = 0.620). The proportion of patients requiring steroid therapy were similar, as were the hospitalization rates. There was no IRIS-related death in either group. The risk of TB-IRIS with INSTI versus non-INSTI was also similar in a stratified analysis in a subgroup of patients with a baseline CD4 count of <50 µL (10/33 [30.3%; 95% CI 14.6-46.0] vs. 10/22 [45.5%; 95% CI 24.7-66.3], p = 0.252) and another subgroup of patients with ART initiated within 4 weeks of anti-TB treatment (10/26 [38.5%; 95% CI 19.8-57.2] vs. 10/23 [43.5%; 95% CI 23.2-63.7], p = 0.721). CONCLUSION: Our cohort study did not offer support for an increased risk of TB-IRIS with INSTIs compared with non-INSTIs, even in severely immunocompromised people with HIV.