Setting: Three medical college hospitals using light-emitting diode fluorescence microscopy (LED-FM) for diagnosing tuberculosis (TB) in Chhattisgarh, India.
Objectives: To assess and compare the proportion of sputum smear-positive TB patients diagnosed through same-day microscopy (spot-spot) strategy or with the conventional (spot-morning) strategy.
Methods: During November 2012 - March 2013, all consecutively enrolled presumptive TB patients (aged ≥ 18 years) were requested to provide three specimens: two spot specimens collected 1 h apart on the first day and one early morning specimen the next day; these were stained using auramine-O and examined using LED-FM.
Results: Of 1716 (93% of total 1845) presumptive TB patients who provided all three specimens, 218 (13%) were smear-positive: 200 (11.7%) by same-day microscopy and 217 (12.7%) by the conventional method (McNemar's χ(2) 13.5, df 1, P = 0.0002). Eighteen (8.3%) cases were missed by the same-day method.
Conclusion: Although LED-FM is more sensitive to paucibacillary samples, 8% of smear-positive cases were missed using the same-day method. These findings indicate the need to revisit the global applicability of the current World Health Organization recommendation of switching to same-day diagnosis from the conventional policy.