C. difficile infection is common in patients with hematologic malignancy. There is increasing recognition that molecular (polymerase chain reaction, PCR) based testing lacks specificity for infection, while detecting patients with colonization. The objective of our study was to evaluate characteristics of patients with toxin enzyme immunoassay (EIA) versus PCR positive C. difficile test results.
A retrospective review of inpatients at a tertiary care academic center with hematologic malignancy and a positive C. difficile test from 1/2015 to 1/2016 was performed. Data on demographics, comorbidities, clinical features, and outcomes were collected using medical record review. Characteristics were compared between patients with EIA versus PCR positive test results using chi-squared or Fisher’s exact test for categorical variables and Wilcoxon rank sum test for continuous variables.
A total of 130 patients were included: 51% and 49% had a PCR positive and EIA positive result, respectively. Diagnoses included AML (42%), multiple myeloma (22%), and Non-Hodgkin’s lymphoma (13%). Antibiotic exposure was similar, with a median of 4 days of anti-pseudomonal antibiotics received in the prior 30 days. There was no difference in history of a positive C. difficile test in the prior year (12% in the EIA group, 10% in the PCR group, P=0.71).
Patients with EIA positive results were more likely to have a WBC ≥15/mm3 (18% vs 6%, P=0.02). However, there were no differences in presence of fever, stool frequency, or imaging evidence of colitis at the time of testing. Medications in the prior 72 hours were similar, including use of proton pump inhibitors of ~40% and of laxatives of 28%. Clinical outcomes were also similar between patients with EIA versus PCR positive tests: all-cause death (22% vs 20%), recurrent CDI (9% vs 13%), colectomy (1% vs 4%), and megacolon (0% vs 3%). Most patients received treatment with oral vancomycin for a median duration of 14 days.
In patients with hematologic malignancy, those with EIA versus PCR positive C. difficile test results were clinically similar. These findings suggest that algorithms for testing and treatment of C. difficile in hematologic malignancy patients will need to be specifically targeted towards this immunocompromised population.
D. Pegues, None
K. Alby, None
C. Gilmar, None
K. Bink, None
T. Gorman, None
A. Moore, None
J. Han, None
See more of: Poster Abstract Session