Simple, Rapid TB Diagnosis Feasible in Low-Resource, High-Burden Settings

Session: C15 TB Game Changers: Diagnostics, HIV, Drug-Resistance, and TreatmentTuesday, May 17, 2016, 10:15‒10:30 a.m.Location: Room 3016/2018 (West Building, Level 3), MOSCONE CENTER

Newswise — ATS 2016, SAN FRANCISCO – A streamlined approach to tuberculosis (TB) diagnosis requiring a single sputum sample and providing rapid, accurate results to patients proved feasible in rural Uganda, according to research presented at the ATS 2016 International Conference.

At four community health centers, patients learned the same day as their visit if their sputum was positive for TB when analyzed using fluorescence microscopy. If negative, the sputum was sent immediately to a lab, where it was reanalyzed using GeneXpert® MTB/RIF (Xpert), a much more sensitive test. Xpert results were reported back to the health center via automated text messaging.

The goal of the pilot was two-fold (1) to test the feasibility of this approach in a country where TB is endemic and (2) to assess its ability to increase the numbers of patients tested and initiated on treatment for TB, said lead author Priya B. Shete, MD, clinical instructor and research fellow at the University of California, San Francisco. Before the study, patients were typically required to provide two sputum samples, often on separate days, and microscopic analysis was rarely completed the day of the patient’s visit. Access to Xpert testing relied on sputum being transported on average once per week to a testing facility with results brought back the following week.  Of the 822 patients referred for TB testing, researchers found :• 12 percent had TB, of whom 75 percent were diagnosed using fluorescence microscopy and 25 percent were diagnosed using Xpert.• 67 percent testing positive for TB using fluorescence microscopy began treatment within 1 day.• 67 percent testing positive for TB using Xpert began treatment, on average, within 6 days.• 20 percent testing positive for TB did not start treatment.

The researchers are now expanding their study to 20 Ugandan health centers, where they will measure cost effectiveness of the approach and test different tools for further increasing treatment initiation rates, including sending text messages to the patient and offering financial incentives to patients. Although the cost of TB diagnosis and treatment is free in Uganda, as it is in most countries where the burden of TB is high, Dr. Shete said direct and indirect costs are often “catastrophic for patients in Uganda with chronic cough.” Those costs include lost wages, childcare and transportation.

Dr. Shete added that performing rapid molecular testing on-site, which may be possible with newer technology, could be needed to maximize diagnosis and treatment initiation rates.

The researchers were encouraged that pilot clinics increased the percentage of TB testing for patients whose symptoms were consistent with TB. In previous studies, researchers found that only 21 percent of such patients were worked up for TB.“Clinicians in Uganda can see 50 to 100 patients a day, so they often can’t focus on one disease,” Dr. Shete said. “We identified people at each clinic, however, who could focus on TB and worked with the clinic to improve the training of all clinicians who might see TB patients. The results are promising, but there is more work to be done.”Contact for study: Priya B. Shete, MD, [email protected]

Abstract 11955Feasibility of a Streamlined SIngle-saMPLE (SIMPLE) TB Diagnosis and Treatment Initiation Strategy in UgandaP.B. Shete1, T. Nalugwa2, K. Farr1, C. Ojok2, M. Nantale2, L. Chaisson3, D. Dowdy4, A. Katamba2, D. Moore5, A. Cattamanchi11San Francisco General Hospital- University of California San Francisco - San Francisco, CA/US, 2Makerere College of Health Sciences -Kampala/UG, 3Johns Hopkins University - Baltimore, MD/US, 4Johns Hopkins - Baltimore, MD/US, 5London School of Hygiene and Tropical Medicine - London/UK

BackgroundIn high burden countries, many patients with tuberculosis (TB) whopresent to community health centers are lost to follow-up before TB canbe diagnosed or treated, leading to ongoing transmission. A primaryreason is that the standard approach of collecting sputum specimensover multiple days for microscopic examination is not only insensitive butalso inconvenient and costly for patients. We report on the feasibility of apatient-centered, SIngle-saMPLE (SIMPLE) TB diagnosis strategy.

MethodsThe SIMPLE TB diagnosis strategy includes: 1) Single-sample LEDfluorescence microscopy (analysis and reporting of two smear resultsfrom the initial specimen within two hours) and 2) Daily transport ofsmear-negative sputum samples to GeneXpert® MTB/RIF (Xpert) testingsites. In a single-arm interventional pilot study, we evaluated thefeasibility of these components of the SIMPLE TB diagnosis strategy atfour community health centers in Uganda. Using data from TB laboratoryand treatment registers, we evaluated process measures that reflectimplementation of each intervention component and TB diagnosisoutcomes.

ResultsOf 822 consecutive patients referred for TB testing, 465 (57%) werefemale and their median age was 37 years (IQR 26-48). Overall 100/822(12%) patients were diagnosed with TB. Two smears from the initial spotspecimen were analyzed and reported on the same-day for 779/822(95%) patients. Overall, 73 (9%) were smear-positive (68 onsputum1/smear1, 3 on sputum1/smear2 and 2 on sputum2). Of the 71smear-positive patients identified from the first sample, 31 (44%)started treatment on the same-day, 16 (23%) on the next day, and 12(17%) did not start treatment. Sputum was transported to an Xperttesting site within one business day of the initial visit for 670/706 (95%)patients with negative smear results from the first sputum sample. Xpertresults were positive for TB in 27 smear-negative patients (incrementalTB yield 27%), of whom 18 (67%) initiated therapy (mean time totreatment initiation 6 days, 95% CI 1.7-8.3 days).