Setting: Despite the high prevalence of tuberculosis in Korea, false-positive cultures have not yet been reported. At Pusan National University Hospital, in which positive mycobacterial culture specimens were 2.8 daily on average, 12 specimens from 10 patients requested on the same day were positive for Mycobacterium tuberculosis.
Objective: To identify an episode of laboratory cross-contamination in a tertiary care hospital.
Design: All isolates were analyzed by restriction fragment length polymorphism, and patients' medical records were reviewed.
Results: All isolates from 10 patients with supposed cross-contamination were identical. Anti-tuberculosis drugs were administered to two patients unnecessarily, resulting in adverse drug reactions in one patient. In one patient who had known tuberculous empyema, the medication course was probably lengthened unnecessarily.
Conclusions: An episode of laboratory cross-contamination in mycobacterial cultures occurred, possibly due to droplets splashed from a sample in a centrifuge tube. Consequently, the aerosol soiled the tip of the dispenser and contaminated the following specimens sequentially. This episode of laboratory cross-contamination resulted in some modifications in our methods of specimen processing and interpretation of the results.