Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to page options Skip directly to site content

Guide to the Application of Genotyping to Tuberculosis Prevention and Control

Tuberculosis Genotyping Case Studies: How TB Programs Have Used Genotyping

False-Positive Culture Investigation

In early 2003, a state mycobacteriology laboratory began testing the Mycobacterium Growth Indicator Tube (MGIT) 960 system as a potential replacement for their current BACTEC 460 system for culture of M. tuberculosis in broth media. The MGIT 960 is an automated culture system that automatically checks culture tubes for growth every hour and does not depend on a technician reviewing culture results twice each week, as was required by the former system. During February and March 2003, the BACTEC and the MGIT systems were used in parallel.

In March 2003, several M. tuberculosis isolates cultured in the state laboratory were reported to be resistant to isoniazid (INH) and streptomycin. In April 2003, several isolates also demonstrated resistance to isoniazid (INH) and ethambutol. Since these susceptibility patterns are unusual, the concern for possible false-positive culture results was raised and a formal investigation was initiated.

Spoligotyping and MIRU analysis identified three separate genotyping clusters during this time period. The first cluster consisted of two isolates with isoniazid and streptomycin resistance. One of the isolates came from a patient with three positive cultures and an abnormal chest x-ray; the other came from a patient with only one positive culture and a history of blunt trauma to the chest. This patient’s specimen was processed 1 day after the first patient’s specimen was processed for drug susceptibility testing.

The second cluster involved isolates from six patients, each with only one positive culture. Only one patient had a clinical picture suggestive of tuberculosis, and this patient was in his 11th month of anti-TB treatment administered by directly observed therapy. The spoligotyping and MIRU results from these six isolates matched a quality-control strain used by the laboratory. In each instance, patient specimens were processed for drug susceptibility testing within 1-2 days of the quality-control strain.

The third cluster involved only one patient. This patient had only one positive culture result and had a clinical picture that was not particularly suggestive of tuberculosis disease. This patient’s isolate matched the genotyping pattern of a proficiency strain the laboratory processed 1 day before during drug susceptibility testing.

A review of laboratory procedures revealed that new laboratory protocols were required for use of the MGIT system. These protocols stipulated that a manual micropipettor should be used to inoculate tubes for drug susceptibility testing and for adding sterile supplement to broth tubes for culture of new specimens. Additional micropipettors had been ordered, but they had not arrived yet at the time of the contamination. Therefore, the same manual micropipettor was used to inoculate tubes for DST each afternoon and to add sterile supplement to broth tubes each morning. These broth tubes were subsequently inoculated with incoming specimens/isolates via sterile, disposable pipettes.

This investigation highlights several common findings when false-positive cultures occur. All of the patients with false-positive cultures had only one positive culture result (all the true cases had more than one positive culture result). All of the patients with false-positive culture results did not have clinical pictures that were particularly suggestive of active tuberculosis, and in all cases, the contaminated cultures and the cultures that were the sources of the contamination were processed within 1-2 days of each other. Genotyping helped the laboratory staff and clinicians communicate rapidly and terminate unnecessary treatment.

As you will see when we discuss algorithms for identifying possible false-positive culture results, each of these findings are red flags and should be evaluated each time a genotyping cluster is reported.