[THE MINIMUM NUMBER OF SPUTUM SMEAR SAMPLES NEEDED FOR THE DIAGNOSIS OF TUBERCULOSIS].

[Objective] To determine whether three sputum examinations with fluorescent staining are necessary to diag- nose tuberculosis (TB) in our hospital. [Patients] From April 2005 to December 2012, 379 TB patients were admitted and received anti-TB therapy in our hospital. [Methods] A retrospective study was conducted to assess the positivity rates of sputum smears based on three exami- nations. The positivity rate of first sputum smear and the cumulative smear-positive rates in the second and third were determined. Then, we also determined difference of positivity rates in sputum properties, sampling procedures and cavity formation. [Results] Of the 379 patients who met the screening criteria, 300 tested positive based on the first sputum smear (79.2%). The positivity rate of the first sputum smears was higher in the purulent sputum group than in the mucous sputum group (91.2% vs. 72.3%). Cavity formation, and sputum extraction procedures were not related to the positivity rate of the first sputum smears. In the mucous sputum group, the cumulative smear-positive rate in the second test significantly rose, but did not rise in the third test. [Conclusions] Three sputum smear examinations were necessary in patients who submitted mucous sputum samples. It is important to get purulent sputum.