Oral mannitol for bowel preparation: a dose-finding phase II study.

Background: Successful bowel preparation (BP) for colonoscopy depends on the instructions, diet, the laxative product, and patient adherence, which all affect colonoscopy quality. Nevertheless, there are no laxatives which combine effectiveness, safety, easy self-administration, good patient acceptance, and low cost. However, mannitol, a sugar alcohol, could be an attractive candidate for use in clinical practice if it is shown to demonstrate adequate efficacy and safety.

Aims: The present phase II dose-finding study compared three doses of mannitol (50, 100, and 150 g) to identify the best dose to be used in a subsequent phase III study.

Methods: The Boston Bowel Preparation Scale, caecal intubation rate, adherence, acceptability, and safety profile, including measurement of potentially dangerous colonic gas concentrations (CH, H, O), were considered in all patients. A weighted algorithm was used to identify the best mannitol dose for use in the subsequent study.

Results: The per-protocol population included 60 patients in the 50 g group, 54 in the 100 g group, and 49 in the 150 g group. The 100 g dose was the best as it afforded optimal colon cleansing efficacy (94.4% of patients had adequate BP), adherence, acceptability, and safety, including negligible gas concentrations.

Conclusions: The present study demonstrated that the colon cleansing efficacy and safety of mannitol were dose dependent. Conversely, gas concentrations were not dose dependent and negligible in all patients. Combined evaluation of efficacy, tolerability, and safety, using a weighted algorithm, determined that mannitol 100 g was the best dose for the phase III study.