Comparison of Clostridium Difficile Associated Diarrhea Detection by Enzyme Linked Immunosorbent Assay (EIA) and Polymerase Chain Reaction (PCR) in Autologous Hematopoietic Stem Cell Transplant (AHSCT) Recipients

Diarrhea is a frequent complication of autologous hematopoietic stem cell transplantation and is a common cause of morbidity. Clostridium difficile associated diarrhea (CDAD) is the most common cause of nosocomial infections. Detection of CDAD is widely performed using enzyme linked immunosorbent assay (EIA) to C. difficile toxins A and B because of its technical ease of use and rapid results. The sensitivity of EIA is 60%, but the sensitivity increases with repeated testing. Recently, detection of CDAD by Polymerase Chain Reaction (PCR) methodology has resulted in rapid detection of amplified C. difficile toxin (CDT) B chromosomal material and has been shown to be more sensitive than EIA in patients suspected of CDAD; however, no comparison of the two methodologies has been performed in recipients of an autologous hematopoietic AHSCT.

This retrospective analysis of AHSCT recipients at our institution evaluates the incidence of CDAD when using either EIA or PCR. CDAD was assessed at the onset of diarrhea in both cohorts. The time period of interest was from 7 days prior to transplantation to 30 days posttransplanation. In the EIA cohort, three consecutive liquid stools were tested for the presence of CDT A and B by EIA. In the PCR cohort, only one liquid stool was sent for sampling. Our institution adopted PCR methodology on 1 February 2010. Demographic and clinical information were collected to assess CDAD risk. Successful treatment of CDAD was defined as resolution of diarrhea within seven days. Previously published data from our institution have shown the incidence of CDAD to be 15% when using EIA methodology. Nominal data were analyzed by Fisher's Exact Test and continuous data were analyzed by Student's T-Test.

A total of 159 patients were screened; and 16 patients were excluded because they did not have a CDAD test performed. Seventy-three were included in the EIA cohort and 69 were included in the PCR cohort. Clinical characteristics were similar between the groups. Known risk factors for CDAD including days in the hospital, days of neutropenia, and days of prophylactic and intravenous antibiotics used were similar between the cohorts. The incidence of a positive CDAD test was higher in the EIA cohort compared to the PCR cohort (18% vs. 7% [p = 0.049]). The duration of diarrhea was longer in the EIA cohort as compared to the PCR cohort (7.3 days vs. 5.5 days [0.016]). Of the three consecutive EIA tests sent, the first was positive only 30% (4/13) of the time. Metronidazole was the first-line agent used in all cases. Response to therapy was poor in both the EIA and PCR cohorts with rates of 30% and 20%, respectively. No fatalities occurred in either group as a result of CDAD.

The incidence of diarrhea was significantly shorter in the PCR cohort as compared to the EIA cohort. The duration of diarrhea was significantly shorter in the PCR group as compared to the EIA cohort. It is possible that repeated testing for CDAD by EIA decreases the specificity of the test or that the EIA is detecting a non-toxigenic C. difficile enterotoxin. Because our patient population has additional causes of diarrhea including mucosal irritation caused by chemotherapy, additional studies are needed to discern the discrepancies between both methodologies.

No relevant conflicts of interest to declare.