Studies were eligible for inclusion if participants (adults and children) had Crohn's disease in medically or surgically induced remission. Studies needed to report study sizes and patient characteristics. The intervention needed to be enteral nutrition with either an elemental, semi-elemental or polymeric formula. The outcome had to be clinical or endoscopic relapse. Eligible studies could not compare different types of enteral nutrition. It appeared that trials with no comparison group were eligible for inclusion. Case series were excluded.
Within included studies and where reported: most studies were conducted in Japan (one was in Canada and one in UK); follow-up duration ranged from one to three years for most studies; enteral nutrition delivery included nasogastric tube infusion and oral intake; comparators included low fat diets and low residue diets; prophylactic medications included sulfasalazine, steroids and 5-aminosalicylic acid; relapse or recurrence definitions varied, and used the Crohn's disease activity index, International Organisation for the Study of Inflammatory Bowel Disease scores, Dutch activity indices and Rutgeerts scores; and follow-up duration ranged from one month to over 19 years.
Two reviewers independently conducted the study selection; disagreements were resolved by consensus.