Incidence, mortality and lethality of immune reconstitution inflammatory syndrome in hiv- infected patients starting highly active antiretroviral therapy: systematic review and meta-analysis

Abstract

Background The Immune Reconstitution Inflammatory Syndrome (IRIS) is a relatively frequent complication in patients who start ART and it has, over all, not been consistently described before. We made a systematic review and meta-analysis to obtain its incidence and lethality. Methods We included retrospective and prospective cohorts that unspecifically evaluated IRIS in HIV-infected adults initiating HAART, with a minimum follow-up of 6 months. We searched LILACS, PUBMED, Cochrane Library, SCOPUS and Google Scholar databases, and assessed study quality with the Newcastle-Ottawa Scale (NOS). Rates were estimated with a 95% confidence interval using binomial distribution random-effects pooled model. Results We included 8,124 patients from 8 different countries. IRIS incidence ranged from 38‰ to 314‰ patients, with a 170‰ patients as pooled index. It was most common in patients with a baseline T CD4+ ≤100 cells/mm3 and from high-income countries. Pooled mortality and lethality were 10‰ and 4%, respectively. Mortality was more frequent in patients with ≤100 cells/mm3 T CD4+ baseline count and in middle-income countries. Lethality was slightly higher in patients with lower T CD4+ baseline count, regardless of where they came from. Conclusions We found a high overall IRIS incidence in HIV HAART-naïve patients, higher in ≤100 cells/mm3 T CD4+ baseline cell count group. Nonetheless, it varies according to studied population and clinical context. Moreover, lethality was homogenous in all studies. We consider further research on costs on diagnosis and management of IRIS should be done so that cost-effective interventions to avoid this phenomenon can take place. MeSH terms: Immune Reconstitution Syndrome; HIV infections; HAART; Meta-Analysis; Humans.