S152 Investigating the role of fluoroquinolones in the treatment of isoniazid resistant tuberculosis: implications for the 2018 WHO guidance

Abstract

2018 World Health Organization (WHO) guidelines for isoniazid resistant (INH-R) tuberculosis recommend a treatment regimen of four drugs continuously for six months- rifampicin, pyrazinamide, ethambutol, and levofloxacin; with or without isoniazid ([H]RZE-Lfx)- once drug sensitivity testing results are known.1 In the context of a high-income setting, we aimed to assess the regimens used for INH-R tuberculosis and the impact of fluoroquinolones on treatment effectiveness, accounting for INH-R mutations and phenotypic resistance levels. In London 2009–2013, 626 tuberculosis patients with INH-R disease in any anatomical site (3.8% meningeal tuberculosis or other central nervous system involvement) were notified. Treatment regimens and regimen-specific outcomes (table 1) were ascertained systematically from clinical notes. Genotyping of resistance mutations was available for 171 patients. Descriptive analyses of the regimens used were undertaken, followed by univariable and multivariable logistic regression of the association between regimen and regimen-specific outcomes. An isoniazid, rifampicin, pyrazinamide and ethambutol (HRZE) regimen with or without a fluoroquinolone (usually moxifloxacin) was used for 91.1% of the 594 patients with treatment information. Among individuals who completed treatment, median drug durations were 1.8 months, 11.9 months, 2.1 months, 11.8 months and 4.8 months, respectively, and the median overall duration 11.9 months. 15% of patients had a negative regimen-specific outcome. In a univariable logistic regression model the addition of a fluoroquinolone to the HRZE regimen made no difference to the odds of a negative regimen-specific outcome (baseline HRZE, cluster-specific odds ratio 1.05 [95% confidence interval 0.60–1.82], p-value 0.87). Results were similar in a multivariable model adjusting for phenotypic INH-R levels, among other factors (0.99 [0.53–1.85], 0.97), and in a model additionally adjusting for INH-R mutation. In a high-income nation with comprehensive patient management, we find that a HRZE regimen for INH-R tuberculosis that utilises a shorter duration of pyrazinamide than the WHO recommendations has a lower likelihood of negative regimen-specific outcomes. The inclusion of fluoroquinolones- specifically moxifloxacin- in this regimen may be unnecessary. As such, this regimen may result in fewer adverse events than the current WHO recommended therapy. Reference WHO. WHO treatment guidelines for isoniazid-resistant TB2018.