Use of a cutoff range in identifying mycobacteria by the Gen-Probe Rapid Diagnostic System

Commercial DNA probes (Gen-Probe Corp., San Diego, Calif.) for Mycobacterium tuberculosis complex Mycobacterium avium, and Mycobacterium intracellulare were compared with conventional methods for accuracy, applicability, and speed for the identification of putative isolates of the M. tuberculosis and M. avium complexes. Results are expressed as percent hybridization. Values of greater than 15% were considered positive, and values of less than 5% were negative. Cultures having hybridization values within an indeterminate range of 5 to 15% were repeated. Mycobacterial isolates resembling M. tuberculosis and M. avium complex from cultures of 589 specimens, representing 432 patients, were entered into this study; 294 cultures were tested with the M. tuberculosis complex probe, and 326 cultures were tested with the M. avium probe. In all cases, probe results agreed with our biochemical identification of the isolates. The M. intracellulare probe was used with 117 isolates morphologically resembling M. avium complex, and one false-negative result was observed. Seventy-two cultures gave initial hybridization results that fell within the indeterminate range and were repeated. If the manufacturer's recommended 10% cutoff value had been used, the original hybridization values would have resulted in 27 misidentified cultures, 16 false-negatives and 11 false-positives.