Lopez‐Sanroman 2017.

Methods	Study design: RCT, multicentre Setting: Spain, 22 centres; 2012 to 2015
Participants	Inclusion: Adults (18‐70 years) who underwent a resective surgical procedure (radical or non‐radical) for a CD‐specific lesion at 1 of the participating centres; diagnosis of CD established by generally accepted endoscopic, histological, and/or radiological criteria at least 6 months before surgery; evaluation of disease location by a complete investigation of the gastrointestinal tract (gastroscopy, colonoscopy, and small bowel radiography) within a maximum of 1 year before the index surgery; and ability to start oral nutrition (and, thus, oral medication) within the first 10 postoperative days Exclusion: Contraindications for use of mesalamine; pregnancy or intention of pregnancy within the next 18 months; nursing; short bowel syndrome; clinically significant lactase deficiency; any severe additional disease; diagnosis of primary sclerosing cholangitis; presence of an ileocolonic stoma; more than 3 surgeries preceding the index surgery; and failure to obtain informed consent. Age (IG1 / IG2) median [interquartile range]: overall age not reported; 37.00 years [31.00 to 47.00 years] versus 35.00 years [30.0 to 40.0 years] Sex (M:F): 42:42 overall; (23:16) versus (19/26) Type of surgery: not reported Previous surgery (IG1+IG2): 6 (3/3) Start of intervention after surgery: after surgery (consent obtained before surgery) Medication use (IG1+ IG2): Glucocorticoids‐80 (38/42); Immunosuppressants [thiopurines or methotrexate]‐63 (28/35); Anti‐TNF‐α – 49 (21/28) Smoker (IG1 / IG2): 20 (9/11) Number randomised (N = 85): 40/45 Number analysed (N = 84): (39/40) versus (40/40) Post‐randomisation exclusion (n = 3): (1/40) versus (2/45) Consent withdrawal before treatment‐1 (0/1); Loss to follow‐up ‐2 (1/1)
Interventions	Group 1: AZA 2.5 mg/kg/day for one year + Metronidazole 250 mg three times a day by mouth was added for the first 3 months Group 2: ADA 160 mg subcutaneously, then 80 mg at Week 2, or 40 mg at Week 4 and every 2 weeks thereafter for one year + Metronidazole 250 mg three times a day by mouth was added for the first 3 months All participants: Adherence to therapy was assessed by direct questioning and by counting of returned medication
Outcomes	Duration of study: 52 weeks 1. Endoscopic recurrence defined as i ≥ 2b, 3 and 4 based on Rutgeerts score (24 and 52 weeks) 2. Clinical recurrence defined by 1 of the following: increase in CDAI above 200 (24 and 52 weeks) (CDAI ≥ 200: derived from number randomised ‐ remissions) 3.Radiologic recurrence rate 4.Health Related Quality of Life 5. Adverse events 6.Serious adverse events 7.Withdrawal due to adverse events
Notes	Funding source: unrestricted grant from AbbVie [Spanish Working Group on Crohn’s Disease and Ulcerative Colitis] The funding group had no role in the study design, data collection, data analysis, data interpretation, writing of the report, or decisions concerning publication. The authors had unrestricted access to the data; the decision to submit the paper for publication was solely and entirely to theirs. Conflict of interest: All authors have declared conflict of interest (mainly grants, personal fees, collaboration with AbbVie outside the submitted work, research funding from AbbVie etc.) Power calculation: The difference in the proportion of endoscopic recurrence between treatment groups was estimated at 35% (10% for ADA + metronidazole and 45% for AZA + metronidazole), considering a type 1 error of 5%, a two‐tailed contrast with Yates' continuity correction, 90% power (1‐type II error), and an allocation ratio of 1:1. Therefore, 38 participants per treatment group would be needed. Withdrawals were estimated at 10%. The minimal sample was estimat3d at 84 evaluable participants..
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote:“Central randomisation was based on a pre‐generated block randomisation list stratified by centre.” and “Patients were assigned [1:1] to..” Comment:Central randomisation
Allocation concealment (selection bias)	Low risk	Quote:“Central randomisation was based on a pre‐generated block randomisation list stratified by centre...Allocation was concealed by means of a computer‐generated randomisation schedule without stratification or block allocation”
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote:“Neither patients nor investigators were blinded to the administered treatment” Comment: No blinding of personnel and participants performed
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote:“A video recording of the last 15 cm of the neo‐terminal ileum was evaluated by an endoscopist blinded to treatment allocation and experienced in application of the Rutgeerts score [VP]” and “..MRE, which was evaluated centrally by an experienced blinded reader [JR];” Comment:Outcome assessors we blinded to treatment
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote:“We defined the following populations: 1] the intention‐to‐treat [ITT] population, which included all consenting patients who were randomised and received at least one dose of the study medications” Comment: ITT analysis applied, reasons for withdrawal reported and attrition rates were balanced across groups
Selective reporting (reporting bias)	High risk	Trial registration was available (NCT01564823) and all prespecified outcomes were reported in the study except health related quality of life which was only reported as a P‐value in an abstract
Other bias	Low risk	Quote:“The groups were similar regarding baseline characteristics, including smoking status, previous resections, CD phenotype, previous perianal disease, and previous drug exposure” Comment: Groups well balanced at baseline. No other apparent sources of bias detected