Poor Outcomes Seen With First-Line Treatment of Isoniazid-Resistant TB

Use of the first-line regimen for tuberculosis (TB) currently recommended by the World Health Organization (WHO) results in poor treatment outcomes in patients with isoniazid-resistant TB, according a systematic literature review and meta-analysis published in Lancet Infectious Diseases.¹

The study was an update of a 2008 literature review of randomized trials from 1965 to June 2008 of retreatment regimens in previously treated patients with TB and in patients with TB infection that was resistant to isoniazid.² The recent iteration included studies selected for that review, plus additional cohort studies and clinical trials from January 1, 2008, to March 31, 2015, in which results were reported on the standardized treatment of patients with isoniazid-resistant TB, but not those patients also resistant to rifampicin. Data on drug-sensitive patients included in the studies were likewise analyzed. Ultimately, 19 cohort studies and 33 trials involving 3744 patients with isoniazid-resistant TB and 19,012 patients with drug-sensitive TB were eligible for inclusion.

In an email interview with Infectious Disease Advisor, coauthor Richard Menzies, MD, from McGill University’s Respiratory Epidemiology and Clinical Research Unit, Montreal, Quebec, Canada, explained the impetus for the meta-analysis: “Many authoritative statements simply stated that ‘isoniazid resistance doesn’t matter,’ and that treatment can be given without regard to isoniazid resistance. This is based on very limited data and in large part comes from the fact that isoniazid-resistant patients were treated in many randomized trials and they did not seem to have significantly worse outcomes. But this is because in many randomized trials, the isoniazid resistance subgroup [was] only 10 or 20 patients, so of course only 2 or 3 of them might have failed or relapsed, and perhaps only 1 would have acquired multidrug resistance, which would never have been detected as statistically significant.”

Results from the meta-analysis showed that of patients with isoniazid-resistant TB who received the WHO-recommended regimen for those without prior treatment, treatment failure was seen in 11%, relapse in 10%, and acquired multidrug resistance in 8%. In contrast, those who received the WHO regimen for previously treated patients experienced treatment failure, relapse, and acquired multidrug resistance at rates of 6%, 5% and 3%, respectively. “The big risk is acquired multidrug resistance; that is, going from isoniazid-resistant TB, which is, after all, not that bad, to multidrug-resistant TB, which has very high failure and death rates,” commented Dr Menzies.

Dr Menzies also noted that in most countries, patients without a history of prior TB treatment do not typically receive drug sensitivity testing before initiation of the standard first-line TB regimen. “In a large country like India, this can mean thousands of patients develop multidrug-resistant TB every year simply because they have unrecognized isoniazid resistance. I mention India because the prevalence of isoniazid resistance is over 10% in new patients (people who never been treated before). In a country like Vietnam, where 18% of new patients have isoniazid resistance, this would be an even more important phenomenon. Our results indicate that presumptive treatment of all new cases with the WHO category 1 treatment without knowing whether they have isoniazid resistance is an important contributor to the global multidrug resistance epidemic.” 

(Some quotations were edited for clarity.)

References

1. Gegia M, Winters N, Benedetti A, van Soolingen D, Menzies D. Treatment of isoniazid-resistant tuberculosis with first-line drugs: a systematic review and meta-analysis [published online November 16, 2016]. Lancet Infect Dis. doi:10.1016/S1473-3099(16)30407-8
2. Menzies D, Benedetti A, Paydar A, et al. Standardized treatment of active tuberculosis in patients with previous treatment and/or with mono-resistance to isoniazid: a systematic review and meta-analysis. PLoS Med. 2009;6(9):e1000150. doi: 10.1371/journal.pmed.1000150