Objective: To assess the need for testing a third sputum sample for the diagnosis of active pulmonary tuberculosis (PTB).
Method: Review of the records of all sputum culture proven cases of active PTB in the state of Qatar diagnosed from 2002 to 2008.
Results: During the 7-year study period, 687 cases of culture proven PTB were diagnosed in Qatar. Five hundred one cases had at least one positive sputum smear for acid-fast bacillus (73%). Among these, 363 cases had 3 consecutive sputum samples tested. There was no statistically significant difference in sensitivity between the first, second, and third sputum samples in detecting the acid-fast bacilli on direct fluorescent microscopy (81.3%, 83.5%, and 82.3%, respectively). The sensitivity of the first sample in establishing diagnosis was 81.3%, increasing to 96.2% if 2 sputum specimens were tested (14.9% incremental benefit). Testing a third specimen increased the sensitivity to 100% (3.8% incremental benefit).
Conclusion: Based on our results and to deal with the increasing number of cases of suspected PTB seen in the State of Qatar in the last few years as well as to minimize the increasing demand for isolation rooms and the burden on our laboratory, we recommend that testing only 2 sputum specimens for acid-fast bacilli instead of 3 is probably safe and is adequate for the diagnosis of PTB.