468. Diagnosis of Clinical Clostridium difficile Infection: An Unmet Challenge

BACKGROUND: Diagnosis of Clostridium difficile infection (CDI) is challenging. The reason is two-fold: (a) lack of unique symptoms and (b) lack of a gold standard test for CDI. We studied variation in CDI rates when different diagnostic algorithms were utilized. In addition, we compared patients who met the clinical definition of CDI with different diagnostic assays. METHODS: This is a retrospective study at an academic medical center (401-bed) conducted over 12 months (January 2017–December 2017). A stool sample that tested positive by polymerase chain reaction (PCR) for C. difficile (n = 81) was then tested for glutamate dehydrogenase (GDH) and toxin enzyme immunoassay (EIA). Additionally, all PCR-positive cases were also tested for toxin production by cytotoxic neutralization assay (CCNA). Clinical C. difficile was defined as three or more loose stools within 24-hour time period. Clinical data were obtained from review of charts. This definition was applied to all community-onset and hospital-onset cases. RESULTS: C. difficile was detected in 81 symptomatic patients by PCR test. Of these, 41.9% met the clinical definition of diarrhea. Of the 81 patients, toxin EIA and GDH were positive in 29.6% (24/81) and 4% met the clinical definition. CCNA was positive in 66.67% (54/81) and only 9% met the clinical definition. The CDI rate (per 10,000 patient days) was 10.2 in the PCR positive group; 3.02 in toxin EIA and GDH group and 6.81 in CCNA group. Duration of diarrhea was longer when functional assays (toxin EIA and/or CCNA) were positive, i.e., 48 hours after diagnosis, 22.7% (18/79) of patients with a positive CCNA and EIA had diarrhea while only 6% (3/49) of the patients with GDH and PCR positive tests (nonfunctional assays) had diarrhea (P = 0.013). The difference was statistically significant. All 81 patients were started on CDI treatment within 24 hours of diagnosis. Of note, there was no laxative use contributing to symptoms in these cases. CONCLUSION: CDI rates differ with various diagnostic algorithms. Duration of diarrhea was significantly longer when functional assays (CCNA or toxin EIA) were positive. Inclusion of both a functional assay (EIA and/or CCNA) and a clinical definition of CDI can improve the diagnostic accuracy of CDI. A combination of clinical judgment and functional assays is required for an accurate diagnosis of CDI. DISCLOSURES: All authors: No reported disclosures.