Bulk staining of smears: no demonstrated risk of bacilli transfer from a positive to a negative smear.

Despite a theoretical risk of transfer of bacilli from a positive to a negative smear, bulk staining is routinely performed in many laboratories. To assess this risk in our laboratory, two smears were made from each sputum specimen and stained with auramine: one smear was stained on a rack and the second using the bulk method. Smears were read blind using a fluorescence microscope. A total of 811 sputum specimens were analysed. No acid-fast bacilli transfer was observed even when staining solution jars had not been renewed for 3 days. Bulk staining is rapid and cheap, and could be used in laboratories with a high workload in low-resource settings.

Smear-negative, culture-positive pulmonary tuberculosis among patients with chronic cough in Cotonou, Benin.

SETTING: The main tuberculosis (TB) centre in Benin, West Africa, where only 2% of adult pulmonary TB cases are sputum smear-negative, all other pulmonary cases being smear-positive. OBJECTIVES: To assess the burden of smear-negative, culture-positive pulmonary TB among TB suspects in Cotonou, and to estimate the total number of non-smear-positive TB cases at country level. DESIGN: For 1 year, one morning sputum culture was performed for every TB suspect (cough lasting >3 weeks, as defined in Benin's national guidelines) with three negative sputum smears (fluorescence technique). RESULTS: Of 214 TB suspects for whom culture was performed, only 22 smear-negative, culture-positive cases were identified. During the same period, 831 sputum smear-positive cases were diagnosed. Culture therefore contributed only 2.6% of the total number of bacteriologically proven cases. CONCLUSION: These results show the relatively low input of culture in TB diagnosis among chronic coughers in Cotonou, Benin, and demonstrates that the expected number of non-smear-positive TB cases in Benin is probably much lower than the World Health Organization's current annual estimates.

Comparison of two LED fluorescence microscopy build-on modules for acid-fast smear microscopy.

SETTING: National Reference Laboratory, Benin. OBJECTIVES: To compare the performance of Fraen FluoLED and LW Lumin light-emitting diode (LED) fluorescence microscopy modules. DESIGN: Acid-fast bacilli (AFB) smears, routinely examined with a classical fluorescence microscope, were blindly re-read with both LED systems at 200x magnification. Smears with discordant results were rechecked on all systems at 200x, and 100 randomly chosen smears were read again at 400x. Confirmed presence of AFB with any system was accepted as a true positive. RESULTS: A total of 1937 smears were examined by all systems. The Fraen and LW detected 895 (46.2%) and 817 (42.2%) positive and scanty positive smears. After rechecking 201 smears, 15 false-positive and 61 false-negative results were declared for Fraen, against 11 and 135 for LW. The systems had similar false-positive rates (1.7% for Fraen and 1.4% for LW), but differed significantly regarding detection of confirmed microscopy positives (93.5% and 85.6% respectively, P < 0.00001). A high correlation between both LED systems was found at 400x magnification. CONCLUSIONS: The Fraen LED fluorescence microscopy module performed significantly better than the LW LED at the most efficient 200x magnification. It was also more appreciated by all users. The LW module may perform equally well at higher magnification.