Investing time in microscopy: an opportunity to optimise smear-based case detection of tuberculosis.

SETTING: St. Elizabeth Hospital, rural North-west Cameroon, an area with high tuberculosis (TB) and human immunodeficiency virus (HIV) incidence.
OBJECTIVE: To measure the time spent during routine sputum smear microscopy and assess whether reexamining slides for 10 min translates into higher case detection of smear-positive cases.
DESIGN: A prospective observational study over a 6-month period with three components: 1) timing of routine sputum smear examination; 2) blinded reexamination of all slides for 10 min and results compared with initial readings; and 3) blinded re-examination, by the original microscopists, of a portion of false-negative slides mixed with true negatives for 10 min.
RESULTS: A total of 204 patients submitted 612 sputum specimens for screening. The median routine examination time was 2 min 6 seconds (interquartile range 1:30-2:30). A 10 min examination significantly increased the number of positive smears from 82 to 116 (P = 0.0083), and overall case detection from 28 to 48 patients (P = 0.011). On review by the original readers, more than half of the false-negative slides were reported as positive after 10 min.
CONCLUSION: Sputum smear microscopy has low sensitivity if performed too quickly, and 10 min re-examination significantly increases case detection. Ensuring that smears are examined for the recommended duration may be a simple and low-cost way to improve case detection.