Patients' description of rectal effluent and quality of bowel preparation at colonoscopy

Background: There are few data evaluating how accurately patients can predict the quality of their colonoscopy preparation. Objective: The aim of this study was to assess whether patients' description of rectal effluent predicts preparation quality as assessed per endoscopist. Design: Prospective, cross-sectional. Setting: Three outpatient endoscopy units at Indiana University Medical Center. Patients: Patients undergoing colonoscopy were enrolled. Interventions: Patients were given a questionnaire assessing their preparation based on the description of their last rectal effluent. This was compared with endoscopists assessment of preparation. Main Outcome Measurements: Correlation between the patient's description of the last effluent and endoscopist's assessment of preparation. Results: Of the total 429 patients, 59% were male and 75% were white. There was only slight agreement between the patients' description of effluent and the endoscopists' description of preparation (Cohen kappa statistic, 0.067). However, patients reporting brown liquid or solid had a 54% chance of having fair or poor preparation. Ingestion of <90% of the preparation, male gender, use of medications associated with constipation, and comorbid conditions were independent predictors of fair or poor preparation. Limitations: No validated system to assess the quality of the bowel preparation or for patients to assess their preparation. Conclusion: Patients' description of last rectal effluent is not a reliable predictor of quality of preparation per the endoscopist, but patients reporting their last effluent as brown liquid or solid have a substantial likelihood of inadequate preparation. These patients may benefit from additional preparation, which may be particularly useful if it can be administered in the endoscopy unit followed by colonoscopy on the same clay. (Gastrointest Endosc 2010;71:1244-52.)