Diagnostic Performance of Multiplex Allele-specific Polymerase Chain Reaction for Isoniazid Resistance in Mycobacterium tuberculosis: Comparison with Phenotypic Mycobacteria Growth Indicator Tube Drug-susceptibility Testing

BACKGROUND: Rapid detection of isoniazid (INH) resistance remains important for tuberculosis management, although targeted molecular assays may not capture the full genetic heterogeneity underlying phenotypic resistance. This study evaluated multiplex allele-specific polymerase chain reaction (MAS-PCR) for detecting INH resistance-associated molecular patterns in Mycobacterium tuberculosis . METHODS: This cross-sectional diagnostic accuracy study included 100 sputum specimens, comprising 70 acid-fast bacilli smear-positive and 30 smear-negative specimens. Phenotypic INH susceptibility was determined using Mycobacteria Growth Indicator Tube drug-susceptibility testing (DST-MGIT). MAS-PCR interrogated resistance-associated regions, with particular emphasis on katG315 and ahpC. Diagnostic performance and agreement were assessed in the primary evaluable population of 64 smear-positive specimens with paired MAS-PCR and DST-MGIT results. RESULTS: Among the 64 evaluable specimens, MAS-PCR yielded 5 true-positive, 3 false-negative, 0 false-positive, and 56 true-negative results. Sensitivity was 62.5% (95% confidence interval [CI]: 24.5%-91.5%), specificity of 100.0% (95% CI: 93.6%-100.0%), positive predictive value of 100.0% (95% CI: 47.8%-100.0%), negative predictive value of 94.9% (95% CI: 85.9%-98.9%), and accuracy of 95.3% (95% CI: 86.9%-99.0%). Overall agreement was 95.3%, with substantial agreement beyond chance (Cohen's κ≈0.745). The exact McNemar's test showed no evidence of systematic directional discordance ( P = 0.250). The targeted katG315 mutation-associated pattern occurred exclusively among phenotypically resistant specimens, whereas no targeted ahpC pattern was detected. CONCLUSION: MAS-PCR demonstrated high specificity and substantial agreement with DST-MGIT among evaluable smear-positive specimens but incomplete sensitivity for phenotypic INH resistance. These findings support its use as a complementary targeted molecular approach rather than a replacement for phenotypic DST.