Nitrate reductase assay using sodium nitrate for rapid detection of multidrug resistant tuberculosis.

We validated the nitrate reductase assay (NRA) for the detection of multidrug-resistant Mycobacterium tuberculosis (MDR-TB) using sodium nitrate (NaNO3) in replacement of potassium nitrate (KNO3) as nitrate source. NaNO3 is cheaper than KNO3 and has no restriction on use which facilitates the implementation of NRA to detect MDR-TB.

Tuberculosis (TB) remains as a serious health problem worldwide. In 2008, 9.4 million new cases of the disease were reported (8).Besides, there were approximately 400,000 new annual cases of TB caused by multidrug resistant (MDR) strains, defined as resistant to at least isoniazid (INH) and rifampicin (RIF) (8). Rapid detection of drug resistance is an urgent priority to identify patients who are not responding to the standard treatment and to avoid the transmission of resistant strains (4).Recently, the WHO endorsed new non-commercial drug susceptibility testing methods for the detection of MDR-TB patients, and among them, the nitrate reductase assay (NRA). NRA is based on the capacity of M. tuberculosis to reduce nitrate to nitrite, which is easily detected in a colored reaction (5).NRA was initially standardized using potassium nitrate (KNO3) (1) as nitrate source, but due to its higher cost and restriction of use in several countries, some authors have used sodium nitrate (NaNO3) as nitrate source (3, 6, 7). However, there is no study comparing the accuracy of NRA using NaNO3in replacement of KNO3. In this study, we evaluated the use of the NaNO3 in the NRA test to detect MDR-TB in comparison with KNO3 and results were compared to those obtained with the conventional proportion method (PM) performed on Löwenstein-Jensen (LJ) medium.

One-hundred and six M. tuberculosis strains belonging to the collection of the Institute of Tropical Medicine of Antwerp, Belgium, were studied. The strains were cultured in LJ medium and incubated at 37°C for three weeks. The PM was performed according to Canetti et al(2), using critical concentrations of 0.2 μg/mL for INH and 40 μg/mL for RIF. The NRA was performed according to the previously reported methodology (1). The NRA was carried out in LJ medium incorporating 1.0 mg/mL of NaNO3 or KNO3. The critical concentration of INH and RIF were the same used in the PM.

Table 1 shows the sensitivity and specificity obtained with the NRA using NaNO3 or KNO3 compared to the PM. Drug susceptibility testing for RIF showed a sensitivity of 95% with KNO3 and 96% with NaNO3. Specificity was 97% for both nitrate sources. For INH the sensitivity was 97% with KNO3 and 99% with NaNO3 while the specificity was 96% and 93%, respectively.

Table 1

Sensitivity and specificity of the NRA using KNO3 and NaNO3 compared to the PM method.

	NRA-KNO3						NRA-NaNO3
Drug	PM	R (n)	S (n)	Sensitivity (%)	Specificity (%)	R (n)	S (n)	Sensitivity (%)	Specificity (%)
RIF	R	72	4	95	97	73	3	96	97
	S	1	29			1	29
INH	R	76	2	97	96	77	1	99	93
	S	1	27			2	26

Out of the 106 strains tested, results were available after 10 days for 101 strains (95.3%) using KNO3 and for 104 strains (98.1%) using NaNO3. All strains were positive after 14 days with both reagents.

This study showed that the NRA gave similar results using KNO3 or NaNO3. as nitrate source. NRA using NaNO3 showed high sensitivity and specificity for RIF (96% and 97%, respectively) and INH (99% and 93%, respectively). These results are in agreement with previous studies presented in a meta-analysis that evaluated the accuracy of the NRA for the detection of MDR-TB. According to that meta-analysis most of the studies that applied NRA to test M. tuberculosis isolates reported a sensitivity and specificity > 94% for RIF and >92% for INH (5). Another important finding in the present study was that 98% of the strains showed results in 10 days with NRA using NaNO3. This percentage was higher than that obtained using KNO3 and also in a previous study where 87 % of the strains gave results in 10 days (1).

KNO3 is considered as a class of “explosive” and consequently difficult to obtain in some countries such as Brazil. Additionally, NaNO3 is cheaper than KNO3 and has no restriction of use.

Taking into account the high sensitivity and specificity obtained using NaNO3 and the rapid availability of results with 98% of the strains in 10 days, our study validates the use of NaNO3 as the source of nitrate for NRA.