TO THE EDITOR—We read with interest the review titled “Effective utilization of evolving methods for the laboratory diagnosis of Clostridium difficile infection” [1]. The article mentioned that the Infectious Diseases Society of America/Society for Healthcare Epidemiologoy of America guidelines stipulate that once diarrhea is present, histopathologic presence of pseudomembranes is one test that can be used to define C. difficile colitis [2]. We performed a correlation of pathology with C. difficile toxin enzyme immunoassay (EIA) and molecular-based assay (polymerase chain reaction [PCR]) in stool while validating the latter. Of the 56 patients used in the laboratory validation, only 7 had a colonoscopy with biopsies performed. In these 7 patients, 2–4 stool samples had been obtained, and at least 1 sample had been positive for the toxin in each of the patients. The C. difficile gene was detected by PCR in only 3 of the 7 patients. The biopsies had been obtained within a range of 1–60 days of the positive stool sample. Only 1 patient showed pseudomembranous colitis in the colonoscopy and by histopathology 16 days after the toxin and gene were detected. The other 2 patients in which the toxin and gene were detected showed focal active colitis and an enterocutaneous fistula. The pathology in the 4 patients with presence of toxin (positive EIA) but no evidence of the gene (negative PCR) varied from minimal inflammation to active ulcerative colitis. Our colonoscopic and histopathologic results underscore the great variability found when C. difficile is detected by EIA and PCR in stool. There have been few studies that correlate the presence of C. difficile and endoscopic and histopathologic findings [3–6]. In adults with C. difficile–associated diarrhea, pseudomembranes were detected in 51%–89% of patients by endoscopy and in 63% by histology; the remainder showed normal endoscopies or erythema [2, 3]. In a study by Mogg et al [6], only 31% of patients with diarrhea had pseudomembranous colitis histologically and presence of C. difficile toxin in stool, 27% had positive histology and negative C. difficile toxin, and 42% had suspicious histology with positive toxin in stool. The presence of pseudomembranes may not occur in patients with evidence of C. difficile–associated diarrhea in patients who are neutropenic or immunosuppressed, have received hematopoietic stem cell transplantation, or have ulcerative colitis [4]. The majority of cases of pseudomembranous colitis are due to C. difficile; however, some are due to other pathogens [7–9]. El-Gammal et al [10] looked at the use of C. difficile toxin results by clinicians and found that results had minimal impact on treatment decisions. At our institution, there was skepticism of the C. difficile PCR results even though the validation showed better sensitivity and specificity than the toxin (EIA) assays and need for testing fewer samples. Due to the increased sensitivity of the PCR test, clinicians have raised concerns about potential colonization with a toxin-producing strain. Studies that address the clinical usage of the different test results as well as studies that correlate signs, symptoms, colonoscopy, and histopathology with the newer tests are needed.