Research Article: Evaluation of Xpert MTB-RIF guided diagnosis and treatment of rifampicin-resistant tuberculosis in Indonesia: A retrospective cohort study

Date Published: February 28, 2019

Publisher: Public Library of Science

Author(s): Arto Yuwono Soeroto, Bony Wiem Lestari, Prayudi Santoso, Lidya Chaidir, Basti Andriyoko, Bachti Alisjahbana, Reinout van Crevel, Philip C. Hill, Helen Cox.


Rifampicin-resistant tuberculosis (RR-TB) is largely underdetected in Indonesia. Xpert MTB/RIF (Xpert) has recently been introduced, prioritizing patients at risk of RR-TB, followed by phenotypic drug-susceptibility (DST) if rifampicin resistance is detected.

This study investigated Xpert-based management of presumptive RR-TB cases under routine practice in West Java, Indonesia.

We examined all records of patients tested with Xpert in the referral hospital for West Java in 2015–2016. We measured loss across a limited cascade of care, time to Xpert diagnosis and the commencement of initial second-line treatment, and identified factors associated with diagnostic and treatment delay. Additionally, we analyzed the appropriateness of treatment according to DST results.

Of 3415 patients with presumptive RR-TB, 3215 (94%) were tested by Xpert, of whom 339 (10.5%) were diagnosed as RR-TB. 288 (85%) of 339 RR-TB patients started initial second-line TB treatment, with 48 (14%) patients being lost between diagnosis and pre-treatment assessment. Second-line treatment was commenced at a median of 41 days (IQR 29–70) after RR-TB diagnosis. Delays in both diagnosis and treatment initiation were observed in 104 (52%) of 201 RR-TB patients with identifiable referral date. Rural residence was associated with delay to diagnosis (adjusted OR 2.7; 95%CI 1.5–5.2) and treatment initiation (adjusted OR 2.0; 1.2–3.4). Of 162 patients with available DST result, 107 (66%) had multidrug-resistant tuberculosis (MDR-TB) and 32 (20%) had either pre-extensively drug resistant (pre-XDR) or extensively drug resistant tuberculosis (XDR-TB). We estimated that with the current algorithm 41% of pre-XDR or XDR-TB patients are diagnosed, and 33% of them started on an appropriate treatment regimen.

Many patients with Xpert-diagnosed RR-TB either do not start MDR-TB treatment or encountered diagnostic and treatment delays under programmatic conditions in Indonesia, and most pre-XDR and XDR-TB cases remain undiagnosed. Further expansion and ongoing quality improvement of RR-TB services are urgently needed.

Partial Text

The emergence of rifampicin-resistant tuberculosis (RR-TB) has become a threat to global TB control. Suboptimal TB treatment, poor treatment outcomes, low adherence to TB treatment, and poorly regulated private sectors have contributed to the increase of MDR-TB [1, 2]. Most incident of RR-TB cases are thought to result from direct transmission rather than treatment-related resistance acquisition [3], like in South Africa where 52% of RR-TB is caused by direct transmission [4]. Thus, improving RR-TB diagnosis should be a high priority. Xpert MTB/RIF assay (Xpert) allows for universal limited drug-susceptibility testing (DST) with rapid detection of RR-TB [5]. The World Health Organization (WHO) recommends Xpert as an initial diagnostic tool for presumptive RR-TB patients [6]. Since 2011, the global roll-out of Xpert has increased detection of RR-TB diagnosis three- to eight-fold compared to conventional testing [7].

This study has identified specific losses and delays with respect to diagnosis and treatment initiation of RR-TB in Indonesia. Of 3415 patients referred with presumptive RR-TB over a 2-year period, almost 10% (n = 339) were diagnosed with Xpert RR-TB. For this group it took a median of two weeks to obtain an Xpert result and an additional 25 days until start of second-line treatment, with 14% of patients lost between diagnosis and MDR-TB treatment initiation. Patients who came from outside the city were more likely to have diagnostic and treatment delay. Just over half of those with Xpert-diagnosed RR-TB who started initial second-line treatment had a phenotypic DST result. Of those with DST results, 66% had MDR-TB and 20% had either pre-XDR or XDR-TB. If the phenotypic DST data are extrapolated to the whole study population of registered patients, then the system diagnosed only 41% of pre-XDR or XDR-TB patients and started only 33% on an appropriate treatment regimen. Together these data provide important insights into the challenges of diagnosing and treating DR-TB in this setting.




Leave a Reply

Your email address will not be published.