Mitsutoshi Senoh, Haru Kato, Hitoshi Honda, Tadashi Fukuda, Yasuaki Tagashira, Hiroko Horiuchi, Hiroshi Chiba, Daisuke Suzuki, Naoto Hosokawa, Hidetaka Kitazono, Yasuhiro Norisue, Hisashi Kume, Nobuaki Mori, Hideo Morikawa, Saeko Kashiwagura, Akiko Higuchi, Hideaki Kato, Makoto Nakamura, Saori Ishiguro, Sayuri Morita, Hideaki Ishikawa, Takuya Watanabe, Katsuyuki Kojima, Izumi Yokomaku, Tatsuya Bando, Kayoko Toimoto, Kei Moriya, Kei Kasahara, Seigo Kitada, Junko Ogawa, Haruko Saito, Harumi Tominaga, Yousuke Shimizu, Fumi Masumoto, Kayoko Tadera, Junichi Yoshida, Tetsuya Kikuchi, Ichiro Yoshikawa, Tatsuyuki Watanabe, Masahisa Honda, Kuniko Yokote, Takao Toyokawa, Hiroko Miyazato, Mika Nakama, Cedric Mahe, Kimberly Reske, Margaret A Olsen, Erik R Dubberke
Anaerobe 60 102107-102107 2019年12月 査読有り
BACKGROUND: The optimal and practical laboratory diagnostic approach for detection of Clostridioides difficile to aid in the diagnosis of C. difficile infection (CDI) is controversial. A two-step algorithm with initial detection of glutamate dehydrogenase (GDH) or nucleic acid amplification test (NAAT) alone are recommended as a predominant method for C. difficile detection in developed countries. The aim of this study was to compare the performance of enzyme immunoassays (EIA) detecting toxins A and B, NAAT detecting the toxin B gene, and GDH compared to toxigenic culture (TC) for C. difficile as the gold standard, in patients prospectively and actively assessed with clinically significant diarrhea in 12 medical facilities in Japan. METHODS: A total of 650 stool specimens were collected from 566 patients with at least three diarrheal bowel movements (Bristol stool grade 6-7) in the preceding 24 h. EIA and GDH were performed at each hospital, and NAAT and toxigenic C. difficile culture with enriched media were performed at the National Institute of Infectious Diseases. All C. difficile isolates recovered were analyzed by PCR-ribotyping. RESULTS: Compared to TC, the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of EIA were 41%, 96%, 75% and 84%, respectively, and for NAAT were 74%, 98%, 91%, and 92%, respectively. In 439 specimens tested with GDH, the sensitivity, specificity, PPV, and NPV were 73%, 87%, 65%, and 91%, and for an algorithm (GDH plus toxin EIA, arbitrated by NAAT) were 71%, 96%, 85%, and 91%, respectively. Among 157 isolates recovered, 75% of isolates corresponded to one of PCR-ribotypes (RTs) 002, 014, 018/018", and 369; RT027 was not isolated. No clear differences in the sensitivities of any of EIA, NAAT and GDH for four predominant RTs were found. CONCLUSION: The analytical sensitivities of NAAT and GDH-algorithm to detect toxigenic C. difficile in this study were lower than most previous reports. This study also found low PPV of EIAs. The optimal method to detect C. difficile or its toxins to assist in the diagnosis of CDI needs further investigation.