Associated factors of extended treatment duration for drug susceptible tuberculosis

Study Population and Design
From 2018 to 2020, patients with active tuberculosis (TB) at National Taiwan University Hospital (NTUH) were enrolled in this study. Inclusion criteria required microbiologically confirmed TB that was susceptible to all first-line anti-TB medications. Exclusion criteria included known HIV co-infection, pregnancy, age under 20 years, loss to follow-up during treatment, unclear medical records, and treatment durations between 226–269 days.

Treatment Course Definitions
Patients were categorized into two groups based on treatment duration:

• Standard treatment: 180–225 days

• Extended treatment: >270 days

A logistic regression model was used to analyze factors associated with extended treatment. Multivariable analysis included age, gender, and variables with p-values < 0.1 in univariate analysis. Confounding variables—such as poor treatment response, disseminated or bony TB, and incomplete PZA therapy—were excluded. Variables causing complete separation in the model (e.g., organ transplantation, cytopenia) were also removed.

Study Cohort
Out of 279 eligible patients, 221 were included in the final analysis: 141 received standard treatment, and 80 received extended treatment.

Demographic comparisons showed no significant difference in age (65.2 vs. 65.4 years, p = 0.919) or male gender distribution (64.5% vs. 75.0%, p = 0.108).

Significant Clinical Differences
The extended treatment group had significantly higher proportions of:

• Hepatitis B virus (HBV) infection (12.5% vs. 5%, p = 0.043)

• Recent cancer treatment (18.8% vs. 8.5%, p = 0.025)

• Extrapulmonary TB (15% vs. 3%, p = 0.002), including pleural, spinal, soft tissue, urinary/prostatic, gastrointestinal, CNS, lymphatic, nasal, and ear TB

• Organ transplantation (2.5% vs. 0%, p = 0.06, trend only)

There were no significant differences between groups in smoking status, BMI, AFS grade, or autoimmune diseases.

Microbiological Outcomes and Treatment Response
After two months of treatment, the extended group showed non-significant trends toward:

• Higher persistent sputum AFS positivity (16.4% vs. 11.3%, p = 0.58)

• Higher culture positivity for Mycobacterium tuberculosis (10% vs. 5%, p = 0.274)

Adverse Drug Events (ADEs)
Common ADEs in both groups included hyperuricemia (76.6% vs. 68.8%, p = 0.208), skin rash (21.3% vs. 31.3%, p = 0.108), and gastrointestinal upset (22% vs. 22.5%, p = 0.999). However, the extended treatment group had significantly higher rates of:

• Elevated liver aminotransferases (30% vs. 17.7%, p = 0.043)

• Hyperbilirubinemia (12.5% vs. 5%, p = 0.042)

• Cytopenia (8.8% vs. 0%, p = 0.001)

• Neuropathy (13.8% vs. 4.3%, p = 0.011)

Skin itching was more common in the standard group (15.6% vs. 1.3%, p < 0.001). Pyrazinamide (PZA) was the most frequently implicated drug in ADEs, especially in the extended group (68.8% vs. 47.5%, p = 0.003). Treatment interruption was significantly more common in the extended group (46.3% vs. 18.4%, p < 0.001). No significant differences were observed in lab parameters including hemogram, liver function, or uric acid levels.

Reasons for Extended Treatment
The most common reasons for extended TB treatment were:

• Disseminated or extrapulmonary TB (17.5%)

• Persistent positive sputum AFS after 2 months (16.3%)

• Cavitary lesions (6.3%)

• Residual lesions after standard treatment (11.3%)

Regression Analysis
In models adjusting for age, gender, and significant variables:

• Pre-treatment factors:

  • Recent cancer treatment (AOR 2.40, 95% CI: 1.04–5.55, p = 0.041)

  • HBV infection (AOR 2.63, 95% CI: 0.94–7.38, p = 0.067, borderline)

• Combined pre- and post-treatment model:

  • HBV infection (OR 3.11, 95% CI: 1.06–9.11, p = 0.039)

  • Recent cancer treatment (OR 3.09, 95% CI: 1.26–7.56, p = 0.013)

  • Elevated aminotransferase levels (OR 2.40, 95% CI: 1.19–4.86, p = 0.014)

  • Skin rash and neuropathy showed trends toward association (p = 0.057 and p = 0.051, respectively)

Conclusion

This real-world study found that extended anti-TB treatment occurred in approximately 28.7% of patients. Beyond classic indicators like disease severity and poor microbiologic response, host factors—including HBV infection, organ transplantation, and recent cancer therapy—were strongly associated with prolonged treatment. Adverse drug events, especially those leading to discontinuation of PZA, played a critical role in extending treatment duration.

Source: Liu, C.Y., Chen, R.T., Shu, C.C. and Lin, S.Y., 2025. Prevalence, clinical reasons and associated factors of extended treatment duration for drug susceptible tuberculosis–a real-world experience. Scientific Reports, 15(1), pp.1-8.