New onset type 2 diabetes mellitus risks with integrase strand transfer inhibitors-based regimens: A systematic review and meta-analysis

OBJECTIVES: The development of diabetes mellitus (DM) in patients taking integrase strand transfer inhibitors (INSTIs) has raised concerns. It's critical because, in most guidelines, INSTIs are the preferred third agent at first-line regimens. This study investigates the excess risk of developing DM among people living with HIV (PWH) on INSTIs-based regimens compared to those with other combination antiretroviral therapies (cART). METHODS: A search from PubMed, clinicaltrials.gov, Latin America and Caribbean health sciences literature, Cochrane, and google scholar to retrieve case-control and cohort studies were done. The literature search was performed for studies from January 2007 to January 2021. Data were extracted from studies and pooled as risk ratios (RR) with a 95% confidence interval (CI) using Stata 14 software. The protocol was registered in PROSPERO, ID: CRD42021230282. RESULTS: This review included ten studies, resulting in 62 400 participants. There was no significant difference in the incidence of DM between participants receiving INSTIs-based regimens versus other cARTs (RR 0.97, 95% CI: 0.92–1.03; participants = 50 958; studies = 4; I(2) = 86.8%, chi-square = 22.67). There is no statistically significant difference in DM among people treated with INSTIs-based regimens compared to those treated with boosted protease inhibitors (PIs)-based regimens (RR 0.97, 95% CI 0.92–1.03; participants = 49 840; studies = 3; I(2) = 89.3%, chi-square = 18.65). DM incidence was lower in INSTIs-based regimens than in those using non-nucleoside reverse transcriptase inhibitors (NNRTIs)-based regimens (RR 0.80, 95% CI 0.69–0.91; participants = 42 346; studies = 2; I(2) = 0%, chi-square = 0.18). CONCLUSION: The present review shows a nonsignificant difference in the incidence of DM in patients receiving INSTIs-based regimens compared to other regimens. However, there was a lower incidence of DM in the INSTIs group compared to the NNRTIs-based and PIs compared to the NNRTIs-based. When the INSTIs drugs dolutegravir, raltegravir, and elvitegravir were compared, there was a lower incidence of DM in raltegravir compared with elvitegravir.