MRI analysis of fecal volume and dryness: Validation study using an experimental oxycodone-induced constipation model

Background: Opioids are known to affect gastrointestinal motility, but their effect on fluid absorption and secretion is poorly understood in humans. Purpose: To investigate the effect of oxycodone on colonic fecal volume and stool dryness by using a novel MRI-based technique. Study Type: Prospective, randomized, double-blinded, crossover study. Subjects: Twenty-five healthy male volunteers (median age: 24 years [range: 21–56]; mean body mass index [BMI]: 23.9 kg/m ² [range: 22.9–25.0]) without known gastrointestinal disease. Field Strength/Sequence: T ₂-weighted and two-point Dixon MRI scans of the abdomen at 1.5 T. Assessment: Subjects were treated for 5 days with prolonged-release oxycodone or a placebo. Imaging was performed on the first and last study day in each period. Images of the colon were analyzed with semiautomatic k-means-based segmentation software. Regional colonic fecal volumes were quantified excluding gas volume and colon wall. Two-point Dixon and T ₂-weighted MRI signal intensity were assessed as a proxy of colonic stool dryness. Data were obtained in a previously reported study. Statistical Tests: Intraclass correlation coefficients were used to test the reliability of measurements between days, while repeated measures mixed models were applied to test treatment effects. Results: After oxycodone treatment, total colonic fecal volume was significantly increased compared with placebo (mean change 100 mL vs. –13 mL; P = 0.001), with the largest increase (24%) observed in the ascending colon/cecum (P = 0.001). Dixon signal increased (less water in colon content) after oxycodone treatment compared with placebo (mean 0.09 vs. –0.02; P < 0.001). T ₂-weighted signal decreased (less water in colon content) after oxycodone treatment compared with placebo (mean –0.03 vs. 0.03; P = 0.002). Data Conclusion: The 5-day oxycodone treatment increased colonic fecal volume and increased stool dryness compared with placebo. This imaging-based method for noninvasive analysis of colon content has the potential to characterize gastrointestinal symptoms in general, such as in constipation. Level of Evidence: 2. Technical Efficacy Stage: 1. J. Magn. Reson. Imaging 2019;50:733–745.