Setting: Seventeen health facilities in a Tuberculosis Unit, and the Tuberculosis Research Centre (TRC), Chennai, India.
Aim: Evaluation of restaining by the auramine-phenol method for detection of acid-fast bacilli (AFB) in direct smears of samples transported in cetylpyridinium chloride (CPC) solution.
Methods: Among patients attending the above health facilities, 730 samples were collected in CPC and transported to the TRC. Two direct smears were prepared from each sample, one stained by the usual auraminephenol method (primary staining) and the other stained again by the same method (restaining) for examination by fluorescence microscopy. All the samples were processed for culture of Mycobacterium tuberculosis.
Results: A significantly higher proportion (49.6%) of samples were positive by restaining compared to primary staining (32.5%, P < 0.0001). Of 362 samples positive by restaining, 38.7% were negative by primary staining. The yield of different grades of smears was significantly higher with restaining than with primary staining (P < 0.0001). More smear-negative culture-positive smears were observed with the primary than with the restaining method (178/400 [44.5%] vs. 78/ 400 [19.5%], P < 0.001).
Conclusion: The rate of detection of AFB in direct smears made from sputum samples transported in CPC was higher on restaining than on primary staining.