PROJECT

Impact of Xpert MTB/XDR in early diagnosis and treatment and 6-month mortality in patients with drug-resistant tuberculosis: a pragmatic quasi-experimental study

March 2025 - March 2027

Objectives:

We aim to determine the reduction in time to diagnosis and time to treatment for drug-resistant tuberculosis with the implementation of Xpert MTB/XDR (intervention) compared to current standard diagnostic algorithm in NTEP (control)

Study Overview:

Detecting tuberculosis that does not respond to commonly used medicines early is crucial for helping patients recover and survive. A newer test called Xpert MTB/XDR can identify resistance to some important TB medicines within a few hours. However, this test is not yet routinely used in the national TB programme. This study aims to determine whether using this test helps detect drug-resistant TB earlier, start treatment sooner, and reduce deaths among patients suspected of having drug-resistant TB. The test will be introduced in selected areas of Chennai, and the results will be compared with those from standard TB tests. Patients will be followed for six months to assess whether their outcomes improve.

Study Approach:

We proposed a quasi-experimental study among 756 individuals in five NTEP districts of Chennai. Three districts would be offered Xpert MTB/XDR for all the patients with presumptive DRTB and the remaining two districts will follow the routine NTEP diagnostic algorithm. We will record time to diagnosis and treatment in both the arms and evaluate mortality at 3 and 6 months. Independent sample t-test or Mann-Whitney U test will be used to compare the outcomes between intervention and control districts, depending on the distribution of the data. Sensitivity and specificity will be calculated by comparing Xpert MTB/XDR results with phenotypic DST.

Expected Public Health Impact:

The study will demonstrate the effectiveness of Xpert MTB/XDR in reducing the diagnostic delay in patients with resistance to isoniazid and fluoroquinolone, which is currently detected in LPA. If fluoroquinolone resistance is detected by Xpert MTB/XDR, it will also help guide appropriate modification of the treatment regimen. Subsequently, this will enable quick decision-making and reduce the treatment delay. Overall, reducing the diagnostic delay and treatment initiation will improve patient outcomes, increase the proportion of patients with favorable outcomes, and reduce mortality due to DRTB. The study will also indirectly give diagnostic accuracy of Xpert MTB/XDR and strengthen the case for utilising this molecular test in the routine diagnostic algorithm in NTEP