Prevalence of extensively drug-resistant tuberculosis and drug resistance patterns in Asia and Africa (2000-2025): systematic review and meta-analysis.

backgroundExtensively drug-resistant tuberculosis (XDR-TB) remains a significant global health issue, especially in areas with high incidence like Asia and Africa. XDR-TB is characterized by resistance to rifampicin, fluoroquinolones, and at least one Group A drug such as bedaquiline or linezolid, making treatment difficult and raising the risk of death. Although concern about XDR-TB is increasing, there is still limited data on its prevalence and resistance trends in Asia and Africa.

methodsA systematic review and meta-analysis were conducted following PRISMA guidelines, with the protocol registered in PROSPERO. Four electronic databases, which are PubMed, Scopus, ProQuest and Science Direct, were systematically searched for studies published between 2000 and 2025. Twenty-two studies met the inclusion criteria and were assessed for quality using the Joanna Briggs Institute tool. Meta-analysis using a random-effects model was performed in RStudio to estimate pooled prevalence rates of XDR-TB, as well as resistance to key anti-tuberculosis drugs. Given the evolution of the XDR-TB definition in 2021, studies were evaluated for definitional consistency. Where necessary, differences between pre-2021 and post-2021 definitions were acknowledged and addressed through subgroup interpretation.

resultsThe pooled prevalence of XDR-TB was estimated at 16.5% (95% CI: 5.8-27.1), with substantial regional variation. Higher prevalence was observed in Asia (12.3%) compared to Africa (0.4%), although the low estimate in Africa likely reflects limited diagnostic capacity and incomplete drug susceptibility testing. Resistance to XDR-defining drugs was uniformly high, with isoniazid and rifampicin resistance observed in 100% of cases, consistent with the definition of XDR-TB. Among fluoroquinolones, high resistance rates were observed, particularly for ofloxacin (88.4%) and levofloxacin (90.4%), while ciprofloxacin showed complete resistance in the limited studies available. Resistance to last-resort Group A drugs was variable, with bedaquiline (31.7%) and linezolid (60.0%) resistance reported based on single-study estimates. High resistance was also observed among second-line and legacy drugs, including amikacin (88.8%), capreomycin (83.7%), kanamycin (76.7%), and streptomycin (97.4%), reflecting widespread resistance across both historical and current treatment regimens. However, interpretation of these estimates should consider variability in diagnostic capacity and differences in XDR-TB definitions across study periods.

conclusionThe findings highlight the substantial burden and underreported nature of XDR-TB, particularly in Africa, where diagnostic limitations may obscure true prevalence. While newer anti-TB drugs remain largely effective, increasing resistance trends threaten their future utility. The high resistance rates to both first-line and Group C drugs underscore the urgent need for strengthened diagnostic infrastructure, robust antimicrobial stewardship, and accelerated development of new anti-tuberculosis therapies. Effective policy implementation and international collaboration are essential to curb the escalation of drug resistance and improve treatment outcomes. These findings underscore the urgency of evidence-based drug policy reforms that strengthen antimicrobial stewardship, ensure equitable access to effective XDR-TB diagnostics and medicines, and integrate surveillance of resistance trends into national and global TB control strategies.