False-positive results from cultures of Mycobacterium tuberculosis due to laboratory cross-contamination confirmed by restriction fragment length polymorphism.

During 1994, a cross-contamination problem leading to false-positive cultures of Mycobacterium tuberculosis was revealed in a mycobacteriology laboratory processing 30,000 to 35,000 samples per year. Molecular strain typing based on restriction fragment length polymorphism confirmed the contaminations. Out of 1,439 positive cultures of Mycobacterium tuberculosis, 49 samples from 48 patients were suspected to be cross-contaminated. In 37 cases, growth was observed both in BACTEC vials and on solid media, indicating that the contamination took place during the processing of the samples. The majority of the contaminated samples had been handled by the same technician.