PURPOSE: To review case detection strategy for diagnosing pulmonary tuberculosis by smear microscopy of one versus two versus three sputum specimens.
METHODS: All chest symptomatics (total 1156) in the TU were screened with standard 3 sputum smear examinations (spot-morning-spot, S1MS2) for six months (1 January to 30 June, 2007). For the next six months (1 July to 31 December, 2007), screening of chest symptomatics (total 945) was with a variant (spot-morning-morning, SM1M2 ). Respective data were collected from the TB register and Laboratory Register and the results compared.
RESULTS: Overall smear positive TB in the TU was 330 of 2101 suspects (15.71%). With S1MS2 strategy, 166 (14.36%) of 1156 suspects were smear positive, which was statistically similar to SM1M2 strategy, where 164 (17.35%) of 945 suspects were smear positive (Z = 1.86; p>0.05). In both groups with second sputum smear examination, all patients were diagnosed with no additional yield from the third sputum examination. On further comparison, it was seen that second sputum smear examination had the best yield, significantly better ( p < 0.05) than the first and the third with no difference between first and third (p > 0.05) specimens. It was also seen that the second sample, i.e. M1 detected all smear positive cases (100%).
CONCLUSION: Screening TB suspects with two sputum smears (spot and morning) is as good as the present strategy of examining three specimens. Further, it may be worthwhile to consider only one morning sample to screen TB suspects. This however, should be tested with a larger sample size.
CLINICAL IMPLICATIONS: In a resource-poor country like India, a single or two sputum strategy may be cost-effective, associated with less laboratory work and improved quality of results.
DISCLOSURE: Uttam Dasgupta, None.