Colonic stool burden on computed tomography does not correlate with bowel habit: a cross-sectional study.

PURPOSE: Stool burden on abdominal radiographs is not reflective of constipation, perhaps because of variable fecal shadowing or limited inter-rater agreement. These limitations are hypothetically mitigated by computed tomography (CT). This research aimed (i) to evaluate whether bowel movement frequency, stool form, or a diagnosis of functional constipation correlate or associate with stool burden identified on CT, and (ii) to investigate whether physicians use CT in their assessment of a patients' bowel function. METHODS: Patients attending for non-emergent abdominal CT imaging were invited to participate by completing a bowel questionnaire. Stool burden identified on imaging was determined and inter-rater reliability was evaluated in a subgroup of patients (n = 20). Descriptive and comparative analyses were performed. An online questionnaire evaluated the use of abdominal imaging in assessing patients' bowel function in a cross-section of ordering physicians. RESULTS: The patient response rate was 19% (96/516). No clinically useful associations between fecal burden and stool form, bowel motion frequency, gastrointestinal symptoms or a diagnosis of constipation was identified. The physician response rate was 35% (33/94). Opportunistic use of data was more common than deliberate use. However, an abdominal radiograph or CT scan had been requested by 42% and 18% of physicians respectively to assess patient's chronic bowel habit. Approximately 30% of physicians believed evidence supported this practice. CONCLUSIONS: Physicians may use abdominal CT in their assessment of patients' chronic bowel habits. However, objective assessment does not support this practice. Colonic stool burden does not correlate with patient-reported symptoms or a criteria-based diagnosis of constipation.

The diagnostic yield of urgent colonoscopy in acute lower gastrointestinal bleeding.

BACKGROUND AND AIMS: Lower Gastrointestinal Bleeding (LGIB) is one of the leading causes of ER visits. Colonoscopy is indicated in all patients with LGIB, yet the time frame for performing colonoscopy remains unclear. Whether or not urgent endoscopic evaluation improves outcomes of LGIB has been questioned. We therefore aimed to examine the success of urgent colonoscopy in identifying the source of LGIB. PATIENTS AND METHODS: A retrospective study was conducted in which timing of colonoscopy was divided into urgent (performed within the first 24 hours of presentation) and delayed (performed following 24 hours of hospitalization). Data on clinical presentation, investigations and endoscopic findings was collected. Risk ratios were calculated and regression analysis was used to examine associations and identify predictors of endoscopic success. RESULT: A total of 183 patients underwent colonoscopies. 55.4% of colonoscopies were performed within 24 hours of presentation. A source of LGIB was identified in 55.7% of first attempt colonoscopies. Endoscopic intervention was required in 10.9% of cases and rebleeding occurred in 24.6%, of which 6.5% required hospital re-admission. 2.7% required emergency colectomy and the calculated mortality rate was 1%. Risk ratios comparing urgent to delayed colonoscopy for source of LGIB identification, colectomy and mortality were 1.01, 4.8 and 1.2, respectively. Age and timing of colonoscopy appeared to be predictive of colectomy on regression analysis. CONCLUSIONS: Urgent colonoscopy for LGIB did not improve the rate of identification of the source of bleeding, colectomy rate or mortality rate and was predictive of the need for emergency colectomy.