Abstract
Background
Multidrug-resistant tuberculosis (MDR-TB) in HIV endemic settings is a major threat to public health. MDR-TB is a substantial and underreported problem in Sub-Saharan Africa (SSA), with recognised cases projected to increase with advancement in diagnostic technology. There is paucity of review evidence on treatment outcomes and antiretroviral (ART) uptake among MDR-TB patients with HIV in SSA. To address this gap a review of treatment outcomes in HIV patients co-infected with MDR-TB in the SSA region was undertaken.
Methods
Three databases (Medline, Web of Science, CINHAL), Union on Lung Heath conference proceedings and grey literature were searched for publications between January 2004 and May 2018. Records were assessed for eligibility and data extracted. Random effect meta-analysis was conducted using STATA and Cochrane’s review manager.
Results
A total of 271 publications were identified of which nine fulfilled the inclusion criteria. Data was collected from 3,368 MDR-TB and HIV co-infected patients from four SSA countries; South Africa (6), Lesotho (1), Botswana (1) and Ethiopia (1). The most common outcome was cure (34.9% cured in the pooled analysis), this was followed by death (18.1% in pooled analysis). ART uptake was high, at 83% in the pooled analysis. Cure ranged from 22.2% to 57.7% among patients on ART and from 28.6% to 54.7% among those not on ART medication. MDR-TB and HIV coinfected patients were less likely to be successfully treated than HIV negative MDR-TB patients (Risk Ratio = 0.87, 95% CI 0.97, 0.96).
Conclusion
Treatment outcomes for MDR-TB and HIV coinfected patients do not vary widely from those reported globally. However, treatment success was lower among HIV positive MDR-TB patients compared to HIV negative MDR-TB patients. Prompt antiretroviral initiation and interventions to improve treatment adherence are necessary.
Treatment outcomes and antiretroviral uptake in multidrug-resistant tuberculosis and HIV co-infected patients in Sub Saharan Africa: a systematic review and meta-analysis
