Field assessment of the direct nitrate reductase assay for rapid detection of multidrug-resistant tuberculosis in Honduras.

SETTING: The national TB reference laboratory and four health care units connected to the national laboratory network in Honduras, Central America. OBJECTIVE: To evaluate the performance of the direct nitrate reductase assay (NRA) for rapid, low-cost detection of multidrug-resistant tuberculosis (MDR-TB) in a resource-limited setting. DESIGN: Consecutive smear-positive samples (n = 185) were prospectively analysed with NRA and compared to the proportion method on Löwenstein Jensen medium (PM-LJ) to detect resistance to isoniazid (INH) and rifampicin (RMP). RESULTS: The NRA sensitivity, specificity, positive and negative predictive values for INH and RMP were respectively 100%, 99%, 91%, 100% and 80%, 100%, 100%, 99%. Good agreement was observed between NRA and PM-LJ (κ > 0.8). CONCLUSION: The direct NRA is a reliable alternative for rapid and low-cost identification of MDR-TB cases in resource-limited settings.

Evaluation of the nitrate reductase assay for rapid detection of extensively drug-resistant tuberculosis.

BACKGROUND: New diagnostic tools are needed to support tuberculosis (TB) control strategies, particularly in low- and middle-income countries with a high prevalence of TB. OBJECTIVE: To evaluate the nitrate reductase assay (NRA) for the rapid detection of resistance to isoniazid (INH) and rifampicin (RMP), as well as to second-line drugs such as ofloxacin (OFX) and kanamycin (KM). DESIGN: To determine diagnostic accuracy, 192 selected clinical isolates of Mycobacterium tuberculosis were used to compare NRA with BACTEC 460TB for rapid detection of multidrug-resistant (MDR) and extensively drug-resistant (XDR) TB strains. RESULTS: A good agreement between NRA and the BACTEC 460TB reference method was observed, with good sensitivity and excellent specificity for INH, RMP and OFX. Results for KM were also promising, although the sensitivity for the detection of KM resistance should be improved. CONCLUSION: NRA is a diagnostic tool of promise for the timely detection of M. tuberculosis resistance to first- and second-line drugs. Our study showed a clear potential for the prompt detection of both MDR- and XDR-TB cases. Further studies are needed to optimise the testing of second-line drugs.

Improved sputum microscopy for a more sensitive diagnosis of pulmonary tuberculosis.

Diagnosis of tuberculosis in low-income countries is hindered by the low sensitivity of direct sputum smear microscopy. We compared an improved method based on liquefaction of sputum with NaOCl followed by centrifugation with standard direct smear in a central hospital and at peripheral health centres in Honduras. Specificity was high and sensitivity significantly better with the NaOCl method. Moreover, this technique is safe, inexpensive and easy to perform. We recommend its implementation to enable rapid, sensitive laboratory diagnosis of pulmonary tuberculosis, especially in resource-poor settings where culture is not possible.

DNA fingerprinting of Mycobacterium tuberculosis strains from patients with pulmonary tuberculosis in Honduras.

Mycobacterium tuberculosis isolates from 84 patients with pulmonary tuberculosis in Honduras were characterized by restriction fragment length polymorphism analysis. Seventy-three different IS6110 patterns were found; 63 of these were unique and 10 were shared by two to three strains each. Thus, no ongoing spread of any specific clone of bacteria could be demonstrated.

PIP: The relationship among clinical isolates of Mycobacterium tuberculosis from patients with pulmonary tuberculosis in Honduras was investigated through use of DNA fingerprinting. Restriction fragment length polymorphism analysis of isolates from 84 patients revealed 73 different IS6110 patterns. 63 of these patterns were unique and 10 were shared by 2-3 strains each. Smear-positive samples were found for 64 (76%) of all patients, and an isolate derived from at least 1 smear-positive patient was represented in each cluster of strains with identical patterns. Identical or highly related strains--suggestive of ongoing transmission--were more common in men than women, in patients 15-42 years of age, and in those from the Tegucigalpa area. 11 of 76 tested patients were HIV-infected, and only 5 of 11 strains isolated from tuberculosis patients with AIDS had unique IS6110 banding patterns. The fact that no clonal spread of tuberculosis among AIDS patients or those with drug-resistant tuberculosis could be demonstrated indicates that drug-resistant tuberculosis in Honduras emerged in unrelated patients rather than from an ongoing spread of initially resistant M. tuberculosis strains.

Drug-resistant Mycobacterium tuberculosis and atypical mycobacteria isolated from patients with suspected pulmonary tuberculosis in Honduras.

BACKGROUND: Tuberculosis is a major health problem in Central America. In Honduras, with an incidence rate of 81/100,000, it is an increasingly common cause of morbidity and hospitalization. This study was conducted to examine drug-resistant tuberculosis and prevalence of infection with atypical mycobacteria in Honduran patients with suspected pulmonary tuberculosis. METHODS: Pulmonary specimens from 235 Honduran patients with suspected tuberculosis were examined by acid-fast smears and culture. The 95 mycobacterial strains isolated were identified to species level and drug susceptibility tests were carried out. Resistant Mycobacterium tuberculosis strains were tested for susceptibility to six additional drugs. Their possible relationship was studied by DNA restriction fragment length polymorphism. RESULTS: Drug-resistant strains were found in 13 of 85 culture-verified tuberculosis patients, including 10 with isolates of multidrug-resistant bacteria. Seven of the patients with multidrug-resistant tuberculosis had smear-positive disease. Nine of them had a history of specific therapy. Two patients with drug-resistant disease were shown to be infected by identical strains. Only one of 11 HIV-positive patients had drug-resistant tuberculosis. Most resistant strains were susceptible to ciprofloxacin, amikacin, kanamycin, and pyrazinamide. Atypical mycobacteria were isolated from 10 patients with suspected tuberculosis. Seven of them were receiving antituberculosis chemotherapy and five had smear-positive samples. CONCLUSIONS: These results illustrate the importance of mycobacterial culture and subsequent species identification and in vitro susceptibility testing for identification of patients with drug-susceptible or drug-resistant tuberculosis and those infected or colonized with other mycobacteria.