153 Failure of fecal toxin A/B to reliably detect Clostridium difficile in immunocompromised patients: Toxigenic cultures as new gold standard

Friday, March 19, 2010
Grand Hall (Hyatt Regency Atlanta)
Andreas F. Widmer, MD, MS , University Hospital Basel, Basel, Switzerland
Anne M. Strandén, PhD , University Hospital Basel, Basel, Switzerland
Gregory Mansella, MD , University Hospital Basel, Basel, Switzerland
Marc Dangel, MPH , University Hospital Basel, Basel, Switzerland
Lukas Fenner, MD, MIH , University of Bern, Bern, Switzerland
Reno Frei, MD , University Hospital Basel, Basel, Switzerland
Background: Clostridium difficile (CD) is the most common pathogen of nosocomial diarrhea. Rapid treatment with metronidazole or vancomycin is crucial to avoid severe CD-associated disease, especially with the emerging strain NAP1/027 ribotype. Most laboratories rely on direct EIA for fecal Toxin A/B, but current European guidelines emphasize toxigenic cultures. This study evaluated the additional value of toxigenic cultures compared to fecal toxin A/B only.
Objective: Between 2004 and 2008, all consecutive stool specimens submitted for CD were analyzed by two methods: 1. toxin A/B detection directly from stool (CDIFF TOX A/B II; TechLab/Wampole, Blacksburg, VA), and 2. if negative, repeated from the isolate, when culture became positive (toxigenic culture). First isolates from all consecutive inpatients were included. Representative isolates were molecularly characterized including PCR-ribotyping. Clinical data were collected using a standardized case report form by full chart review.
Methods: Between 2004 and 2008, all consecutive stool specimens submitted for CD were analyzed by two methods: 1. toxin A/B detection directly from stool (CDIFF TOX A/B II; TechLab/Wampole, Blacksburg, VA), and 2. if negative, repeated from the isolate, when culture became positive (toxigenic culture). First isolates from all consecutive inpatients were included. Representative isolates were molecularly characterized including PCR-ribotyping. Clinical data were collected using a standardized case report form by full chart review.
Results: 225 toxin A/B positive cases were analyzed. Fecal toxin A/B was positive directly from stool in 129 (57.3%) cases. Toxigenic culture increase the sensitivity by 74% (p<0.0001): An additional 96 (42.7%) cases were identified by toxigenic culture only. Surprisingly, 6/16 (27.5%) hypervirulent PCR-ribotype 027 cases from an outbreak were detected by toxigenic culture only.
The main risk factor for CD being negative by direct toxin testing, but positive by toxigenic culture was a specimen from patients with hematologic malignancies (OR 9.7, CI95 2.5-29).
Conclusions: Toxigenic culture significantly increases the sensitivity by 74% (p<0.0001). Cultures for CD should not be abandoned, in particular in patients hospitalized in hematology units and likely patients with immunosuppression.