Comparison of C(18)-carboxypropylbetaine and standard N-acetyl-L-cysteine-NaOH processing of respiratory specimens for increasing tuberculosis smear sensitivity in Brazil.

Techniques to improve the sensitivity of smear microscopy would facilitate early tuberculosis (TB) diagnosis and disease control, especially in low-income countries where the positive predictive value is high. C(18)-carboxypropylbetaine (CB-18) is a zwitterionic detergent that helps to compensate for the innate buoyancy of mycobacteria, potentially enhancing recovery by centrifugation. Previous data suggest that CB-18 may increase the sensitivity of smear, culture, and molecular amplification diagnostic testing. The goal of the present study was to evaluate if the sensitivity of the smear technique using light microscopy could be improved by treating respiratory samples with CB-18. In the first phase, respiratory specimens were collected consecutively from patients with suspected pulmonary tuberculosis in a tertiary-care hospital in Rio de Janeiro, Brazil (236 specimens were analyzed). After protocol modifications, another 120 respiratory specimens were evaluated. The standard technique was N-acetyl-L-cysteine with sodium hydroxide (NALC-NaOH) treatment, smear concentration with centrifugation, and Ziehl-Neelsen staining. Culture on Löwenstein-Jensen slants was performed on all specimens for use as the "gold standard." No specimens from patients undergoing active TB treatment were included. The initial protocol for CB-18 processing resulted in a sensitivity of 59.6% and specificity of 96.8% compared to standard processing with a sensitivity of 66.0% and specificity of 96.8%. Using the modified protocol, the sensitivity of CB-18 increased to 71.4% with a specificity of 97.0% versus standard processing with a sensitivity of 61.9% and a specificity of 99.0%. The diagnostic yield of acid-fast bacillus smear with CB-18 in the absence of fluorescence microscopy and PCR compared to standard processing with NALC-NaOH was not significantly different, although the power to detect a difference by the modified assay was low.