P230 Isoniazid mono-resistant pulmonary tuberculosis and its clinical outcomes: a prospective multicenter cohort study in Korea

<h3>Background</h3> Resistance to isoniazid is the most common type of anti-tuberculosis drug resistance. The Republic of Korea is an intermediate-tuberculosis-burden and high-income country with increasing burden of elderly people with tuberculosis. We evaluated effects of isoniazid resistance on treatment outcomes in people with pulmonary tuberculosis. <h3>Methods</h3> Among 1,126 adults ≥19 years with pulmonary tuberculosis enrolled in a multicentre prospective cohort study of pulmonary tuberculosis (COSMOTB) between 2019 and 2021, those with isoniazid mono-resistant tuberculosis and with pan-susceptible tuberculosis were identified. Isoniazid-resistance was defined based on results of molecular or phenotypic drug susceptibility tests. Those with rifampicin resistance, other types of drug resistance, and unknown drug resistance profiles were excluded. All patients were regularly followed up until the end of anti-tuberculosis treatment. We evaluated factors associate with treatment outcome among pulmonary TB cases. Multivariable logistic regression models were employed to evaluate whether isoniazid-resistance was associated with an unfavourable outcome; death, failure, loss-to-follow-up, still-on-treatment, and transfer-out. For the sensitivity analysis, we identified cases of death, failure, and still-on-treatment to redefine the secondary unfavourable outcome and conducted additional multivariable logistic regression analyses to assess the effect of isoniazid-resistance on the redefined unfavourable outcome. <h3>Results</h3> Among 758 pulmonary TB cases, 54 (7.1%) had isoniazid mono-resistance. Frequency of unfavourable outcomes was significantly higher in those with isoniazid resistance, compared to those with pan-susceptible tuberculosis (31.5% vs. 17.9%; p=0.014). Old age, low body mass index, cancer, prior tuberculosis history, presence of initial symptoms, and initial severe disease (either positive smear result or presence of cavity on chest x-ray) were significantly associated with unfavourable outcome. Using multivariable logistic regression model, isoniazid resistance was independently associated with unfavourable outcomes (adjusted odds ratio [aOR], 2.00; 95% confidence interval [CI], 1.03–3.89]). In the sensitivity analysis, isoniazid resistance was also significantly associated with redefined unfavourable outcome (aOR, 2.18; 95% CI, 1.07–4.44). <h3>Conclusions</h3> Isoniazid mono-resistant pulmonary tuberculosis was predictive of unfavourable outcome. Effective treatment regimens for isoniazid-resistant tuberculosis are needed to improve outcomes.