Methods

Randomized, parallel group, double‐blind, placebo‐controlled trial. Study medication packaged centrally and labeled with code number. Placebo tablets looked like corresponding active tablet.

Participants

All patients with Crohn's disease seen by the collaborating European centres were considered for admission. Patients who had quiescent disease (Crohn's diesease activity index [CDAI] < 150) were eligible. Patients less than 18 years of age or who had a diagnosis of Crohn's disease more than 2 years prior to the start of the study and those who were pregnant or anticipating pregnancy were excluded. A total of 237 patients with quiescent disease were randomized.

Interventions

Patients received either: 1) placebo; 2) sulphasalazine 3 g/d; 3) 6‐methylprednisolone 8 mg/d; or 4) combination sulphasalazine 3 g/d and 6‐methylprednisolone 8 mg/d for up to 2 years.

Outcomes

"Relapse" was deemed to be a worsening of the CDAI to a level of greater than 150. "Failure" had several definitions including: death due to Crohn's disease, pending surgery for complications of Crohn's disease, development of new fistulas or abscesses, persistence of fever for > 14 consecutive days, rise of CDAI during an acute phase treatment cycle, no change or insignificant decrease of CDAI (< 60 points) during 3 cycles of acute phase treatment, failure to achieve CDAI < 150 or results of interim barium radiograph or endoscopic examination which documented a worsening of the patient's condition compared to the baseline examination.

Notes

1) Details of the study patient characteristics were not available for the group of patients with quiescent disease alone but were given for the overall group of both active and quiescent disease.
2) Although it appears that "treatment failure" was defined as a CDAI > 150 and a need to institute an acute phase treatment, worsening of interim barium radiograph or endoscopic examination was also listed as a possible reason for treatment failure. Such worsening was the reason for discontinuation in one patient in the 6‐methylprednisolone group, five patients in the sulphasalazine group, one patient in the combination group and no patients in the placebo group.
3) Patients who developed a "relapse" as defined by a CDAI score > 150 had the opportunity to receive acute phase treatment cycles without discontinuation from the trial if they responded to treatment. However, it appears that the life table from which the outcome data was derived used "failure and relapse" as the endpoint so that such patients would have been counted as an "outcome" once the CDAI rose above 150.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Randomized, parallel group, placebo‐controlled trial. Study drugs and/or identical placebo tablets were supplied to study patients. Patients all received at least two varieties of tablets.

Participants

Patients with established diagnosis of Crohn's disease who were actively being followed at one of 14 NCCDS study centres and who had a CDAI score of less than or equal to 150 were eligible. Patients with proctitis only, children under age 15, women who were pregnant or who planned to become pregnant during the study, patients with active tuberculosis or systemic fungal infection, patients with diabetes, significant liver disease or hypertension requiring more than thiazide therapy were excluded.
A total of 274 patients (133 females, 141 males) were randomized. Of these, 127 (46%) had prior abdominal surgery for Crohn's disease, 25 (9.3%) had colonic disease only, 120 (48.8%) had small bowel disease only and 149 (41.9%) had both small bowel and colonic diseae. Mean age was 31.8 years.

Interventions

Patients were randomized to one of four treatment arms: 1) placebo; 2) prednisone 0.25 mg/kg/d; 3) azathioprine 1 mg/kg/d; or 4) sulphasalazine 0.5 g/15 kg/d for a period of up to 24 months. Patients who were reactive to intermediate strength PPD (purified protein derivative) were given isoniazid 300 mg/d as tuberculosis prophylaxis.

Outcomes

Primary outcome was "relapse" defined as withdrawal from the study for any of the following reasons: CDAI > 150 and over 100 points greater than initial CDAI for two consecutive weeks, need for surgery, development of new fistula other than simple fistula‐in‐ano, persistence of daily fever for over 14 days in a row, and interim barium x‐rays judged at the study centre to be worse than the initial x‐rays.

Notes

1) Included 48 patients with surgical resection within 1 year of randomization and no evidence of recurrent disease.
2) "Relapse" definition included radiological worsening. This outcome was reported in 1, 2 , 1 and 3 patients in placebo, sulphasalazine, prednisone and azathioprine groups respectively.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Randomized, parallel group, double‐blind, placebo‐controlled trial. Patients were randomly allocated to receive either prednisone or identical placebo tablets.

Participants

Patients with Crohn's disease recruited from one centre in Wales. Eligible patients included those who were free from symptoms and in whom there was no indication for steroid treatment. Patients were divided into 3 subgroups: I) bowel resection within 1 year with no residual disease; II) bowel resection within 1 year but with residual disease; III) active Crohn's disease within previous 12 months but no surgery during that time. Patients were excluded if they were more than 60 years of age, had evidence of liver disease, active peptic ulcer, severe hypertension, diabetes, previous pulmonary tuberculosis or were pregnant.

59 patients were available for these analyses.

Interventions

Prednisone 7.5 mg/d or placebo for up to 3 years. Children under 15 years of age received prednisone 7.5 mg on alternate days. Eight patients continued pre‐existing sulphasalazine therapy (dose range 1 ‐ 5 g/d). Lomotil, kaolin, vitamins and folic acid were permitted co‐interventions.

Outcomes

Clinical relapse as defined by a requirement for additional or open‐label prednisone to control recurrent or persistent abdominal symptoms. At least two clinicians were involved in the decision to initiate open label prednisone.

Notes

1) Eight of 33 patients in the prednisone group and 6 of 26 patients in the control group withdrew from the study. This was presumed to be due to treatment failure or relapse. Additional unreported patients may have been withdrawn during the first 6 months following randomization for reasons other than those due to Crohn's disease. These patients were not included in the report.
2) A single relapse managed by a course of open label corticosteroids no longer than 6 weeks was permitted without considering it to be a treatment failure or relapse. This occurred in 5 patients (4 prednisone and 1 control).

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear