[Experience with laboratory diagnosis of Clostridium difficile].

BACKGROUND: Clostridium difficile is currently a significant cause of nosocomial diarrhea. For several years, the number of infectious cases in the community has also been increasing. Since the beginning of 2010, quite a large increase in the number of Clostridium difficile infections (CDIs) has been noted in Pardubice Regional Hospital (PRH). The objectives of this study were to describe and evaluate the methods used in the laboratory diagnosis of CDIs in PRH, and to describe the laboratory diagnostic algorithm used here.

MATERIAL AND METHODS: Samples of stools were taken from symptomatic patients hospitalized or examined in the outpatient departments of PRH from 1 July 2010 to 31 December 2012. For the detection of glutamate dehydrogenase (GDH) and toxin A/B, the dual test based upon the principle enzyme immunoassays C. Diff Quik Chek Complete, Techlabo (D-EIA) was used. The system GeneXpert PCR Cepheid (PCR) was used for confirmation of laboratory findings. Since the beginning of 2011, all the GDH-positive samples were cultured.

RESULTS: A total of 2,040 samples were examined. The D-EIA test was used for examination of 2,014 samples. Of those, 1,373 (68.2 %) samples were GDH- and toxin A/B-negative. In 359 (17.8 %) samples, both GDH and toxin A/B were detected. The D-EIA sensitivity and specificity for detecting toxigenic strains in stool samples were 21.8% and 97.2%, respectively. The PPV and NPV rates calculated for the populations with prevalence rates of disorders of 5%, 10%, 20% and 50 % were 0.29, 0.46, 0.66, 0.88 and 0.96, 0.92, 0.83, 0.55, respectively. The sensitivity and specificity of GDH for the detection of Clostridium difficile in stools were 100.0% and 96.2%, respectively. PCR examination was carried out in 140 samples. Of those, 82 samples were PCR-positive. The gene for the production of toxin B was detected in 47%, the finding suspected for ribotype 027 (gene for toxin B, binary toxin and deletion of tcdC) in 48%. In 5% of the samples, the gene for toxin B and the gene for the binary toxin were detected.

CONCLUSION: Considering the low sensitivity of the D-EIA test for detecting the toxigenic strain of Clostridium difficile, if used as the only one, a two-step algorithm was introduced for routine laboratory examination of infections with Clostridium difficile in the Clinical Microbiology Department of PRH. In the first step, the D-EIA test diagnosed 86 % of examined samples in 30 minutes as positive (GDH +; toxin A/B +) or negative (GDH -; toxin A/B -). The examination with PCR in the second step increased the number of patients diagnosed with CDI. The test results are available within two hours. This enables quick introduction of isolation measures in the departments of PRH and appropriate antibiotic treatment of the patients.