Routine Molecular Surveillance of Drug-Resistant Tuberculosis: Translating Mutation Proxies into Clinical Governance Intelligence in Rural Eastern Cape

Background: Drug-resistant tuberculosis (TB) remains a major challenge in high-burden settings, where timely identification of emerging resistance and effective governance responses are critical. While routine molecular diagnostics generate large volumes of resistance-associated mutation data, these outputs are typically used for individual patient management and remain underutilized for population-level surveillance and for the application of clinical governance approaches for improved TB care. Methods: We conducted a retrospective cross-sectional analysis of 1386 molecular diagnostic records for Mycobacterium tuberculosis, collected between March 2021 and December 2024, from 30 health facilities in the King Sabata Dalindyebo (K.S.D.) Local Municipality of Oliver Reginald (O.R.) Tambo District. Resistance-associated mutation proxies were identified for loci associated with isoniazid (katG, inhA), fluoroquinolone (gyrA), and second-line injectable agents (amikacin, kanamycin, and capreomycin) through mutations in the rrs locus. Mutation proxy prevalence was examined overall, by age group, over time, and across facilities. Persistence of resistance detection was assessed using consecutive-month analyses to characterize temporal continuity at the facility level. Results: At least one resistance-associated mutation proxy was detected in 72.7% of the analyzed records. Isoniazid resistance predominated, with katG mutation proxies identified in 52.2% and inhA in 20.2% of cases. Mutation proxies associated with fluoroquinolone and second-line injectable resistance were less frequently observed. Temporal analysis demonstrated variability over the study period, with a general decline in overall mutation proxy prevalence alongside a relative increase in inhA-associated mutations. Substantial heterogeneity in resistance patterns was observed across health facilities, with high-volume sites contributing the greatest absolute burden and selected facilities demonstrating sustained persistence of mutation detection over consecutive months. These findings highlight the magnitude, distribution, and persistence of resistance-associated mutation proxies within routine programmatic data. Conclusions: Routine molecular diagnostic data revealed a substantial and heterogeneous burden of drug-resistant Mycobacterium tuberculosis in K.S.D. Local Municipality, characterized by age-specific patterns, temporal shifts, and sustained facility-level persistence. Beyond descriptive epidemiology, routinely generated mutation proxy data can serve as early-warning indicators of clinical governance stress, signaling emerging pressures on TB care systems when resistance patterns persist or worsen. Interpreting these trends can support more anticipatory clinical governance, strengthen resistance surveillance, and guide prioritized interventions in high-burden, resource-constrained settings.