Does the Nose Know? The Odiferous Diagnosis of Clostridium difficile-Associated Diarrhea

TOTHE EDITOR — Clostridium difficile—associated diarrhea (CDAD) is a common and emerging problem in hospitals; patients present with clinical symptoms of leukocytosis and abdominal pain, and the disease is associated with significant morbidity and mortality [1]. Diagnosis relies on various modalities, including laboratory tests (e.g., tissue culture or ELISA), invasive procedures (e.g., colonoscopic examination) and, at times, clinical suspicion and response to empirical treatment [2]. An “urban legend” exists among nursing specialties that CDAD has a unique odor that would allow for olfactory diagnosis; however, this has never been scientifically tested. A majority of nurses interviewed in preparation for this study were quite confident in their ability to make a positive diagnosis of CDAD on the basis of olfactory examination. Physician requests for stool analysis for C. difficile toxin were reviewed at 2 separate teaching hospitals in Dayton, Ohio, over a 6-month period. Nursing staff were interviewed, and a survey (including questions regarding nursing demographic data and specific stool-related data) was completed before stool test results were available. Nurses were included if they had the “perceived” ability to diagnose CDAD by smell and were directly involved in the processing or examination of the patient's stool. Nurses were excluded from the study if their patients were receiving empirical therapy with either metronidazole or oral vancomycin for>24 h. CDAD was diagnosed by standard testing in the hospital laboratory (i.e., by ELISA or tissue culture). A total of 138 nursing staff surveys were completed. Sensitivity and specificity for the odiferous diagnosis of CDAD were 0.55 (95% CI, 0.33–0.77) and 0.83 (95% CI, 0.76–0.90), respectively. The positive predictive value and negative predictive value were 0.35 (95% CI, 0.19–0.52) and 0.92 (95% CI, 0.86–0.97), respectively. The overall accuracy of the nurses in diagnosing or excluding CDAD was 79%. Length of nursing experience did not impact the nurses' ability to diagnose; however, stool quality (i.e., whether the stool was partially formed or formed) did impact the accuracy of the diagnosis. This study suggests that nursing staff were able to exclude a diagnosis of CDAD with a high degree of confidence and accuracy. However, confirmation of the diagnosis on the basis of odor and stool quality was insensitive. Results of this preliminary study, if confirmed, may have a broad impact on the approach to treatment of hospitalized patients who develop diarrhea. The odiferous characteristics associated with the stool of patients with CDAD may impact decisions regarding the initiation of empirical therapy (perhaps limiting the overuse of antibiotics, such as metronidazole), infection-control practices (limiting unnecessary patient isolation), and the necessity for stool testing for CDAD; moreover, these findings may possibly lead to new diagnostic modalities, such as an “electronic sniffer” [3]. Potential conflicts of interest. S.D.B. and J.M.B.: no conflicts.