Setting: Damien Foundation Bangladesh tuberculosis (TB) control projects.
Objectives: To compare 25% sulphuric acid in water (H(2)SO(4)) with hydrochloric acid in water (HCl) to differentiate acid-fast bacilli in sputum smears stained with 1% carbolfuchsin.
Design: For 1 year, all 158 microscopy laboratories used either H(2)SO(4) or 3%/6%/10% HCl for their routine work, alternating monthly between H(2)SO(4) and HCl. Each month a sample of five smears per laboratory was rechecked blind. After recording qualitative staining aspects, all sample smears were restained before rechecking, using H(2)SO(4) for destaining.
Results: A total of 368,059 H(2)SO(4) and 335,436 HCl smears were routinely read, yielding 7.2% positive or scanty results in both groups. Of these, 9492 were rechecked. There was no difference in false-negatives detected (0.66%, 95%CI 0.44-0.95 for H(2)SO(4) vs. 0.68%, 95%CI 0.46-0.98 for HCl), but apparently there were more false-positives with H(2)SO(4) (2.12%, 95%CI 0.92-4.14 vs. 0.28%, 95%CI 0.00-1.54, P = 0.05). Qualitatively, only 3% HCl yielded significantly inferior differentiation results.
Conclusions: HCl 6-10% in water can be recommended for Ziehl-Neelsen destaining above H(2)SO(4). Diluting is easier and safer, and it may cause less confusion with false-positives during rechecking, including a restaining step.