Molecular Characterization and Diagnosis of Nosocomial Clostridium difficile Infection in Hospitalized Patients

Background: Toxigenic Clostridium difficile is one of the prevalent diarrheagenic pathogens in hospitalized patients. Objectives: The study assessed the ability of three diagnostic methods in identifying C. difficile strains. The genotyping of the isolates was done, as well. Methods: Stool samples were subjected to three different diagnostic methods including direct stool culture, glutamate dehydrogenase enzyme immunoassay (GDH-EIA), and direct stool PCR for the detection of the tcdA and tcdB genes. The sensitivity and specificity of the tests were evaluated. The genotyping was done by the PFGE method. Results: Of 120 samples, 20 (16%) were positive for C. difficile based on PCR, while 15 (12.5%) and 12 (10%) were positive according to GDH-EIA and direct stool culture. Among patients with C. difficile-associated diseases (CDAD), 11 (61%) were more than 65-years-old. The specificity of PCR, GDH-EIA, and direct culture was almost similar and equal to 100%, but their sensitivity was 90%, 70%, and 60%, respectively. The positive predictive value (PPV) was lower for GDH-EIA than for the other two methods, and the highest negative predictive value (NPV) was related to the PCR method. The results showed a high similarity between the isolates, and only were three pulsotypes differentiated among the isolates. Conclusions: The specificity and sensitivity of the direct stool PCR method were higher than those of the other two methods. Although PCR inhibitors may reduce its ability for the correct diagnosis of negative samples, it seems to be a reliable method for the detection of C. difficile infection. The weakness of the GDH-EIA method was its lower PPV, which can cause false-positive results. Toxin patterns and pulsotypes of C. difficile isolates revealed a high similarity between the strains isolated from the same units.