Loopamp?百日咳菌検出試薬キットDの 臨床的評価

摘要

A clinical performance study of aloop-mediated isothermal amplification (LAMP)-based test reagent (Loopamp? Bordetella pertussis detection kit D) was conducted in 6 medical institutions. A total of 103 samples from a prospective study and 88 preserved positive samples (nucleic acid extract) were analyzed. Pertussis detection by four assays in18 samples from patients with a definitive diagnosis, of pertussis in the prospective study was compared to clinical diagnosis. Pertussis was detected with a sensitivity, specificity, and overall agreement rate of 27.8%, 100.0%, and 87.4% using the LAMP-based test; 38.9%, 100.0%, and 89.3% using reaレtime PCR analysis, which is also a genetic test; 11.1%， 100.0%, and 84.5% using the cultural method; and 93.8%， 86.8%, and 88.1% using the anti-PT antibody assay, although some samples were indeterminate. The sensitivity was highest for the anti-PT antibody assay and lowest for the cultural method. The specificity was as high as 100.0% for the cultural method, real-time PCR analysis, and LAMP-based test. It was shown that the genetic tests had higher sensitivity than the cultural method and higher speciiicity than the anti-PT antibody assay. According to the diagnostic flowchart for pertussis described in the Japanese Respiratory Society Guidelines for Management of Cough (2nd edition),11samples were supposed to be positive when analyzed using the LAMP-based test reagent; in actuality, 5 of the11samples were positive (5/11， 45.5%). The LAMP-based test had an approximately 2.5 times higher detection rate than the cultural method. Since the remaining 6 samples were negative when analyzed by both the LAMP-based test and real-time PCR analysis, the amount of Bordetella pertussis in the samples may have been very low. In the prospective study, the two reagents for genetic testing yielded inconsistent results in only 2 samples (negative with the LAMP-based test reagent and positive in the real-time PCR analysis); however, sequence analysis revealed that the bacteria in these 2 samples were Bordetella species other than Bordetella pertussis. When the 88 preserved positive samples were analyzed using the two reagents for genetic testing, the LAMP-based test reagent had a sensitivity of 96.5% and an overall agreement rate of 96.5%. The amounts of Bordetella pertussis in the 3 positive samples that were negative in the LAMP-based test may have been below the detectable limit. The LAMP-based test results were highly correlated with the real-time PCR results. In addition, the LAMP-based test reagent is very reliable and requires no specialized instrumentation, indicating that the test can easily be performed in ordinary hospital settings and contribute to early diagnosis of pertussis, which is stilldi伍cult to diagnose in the early stage.