Mantzaris 2009.
| Methods | Prospective, randomized, controlled, 1‐year trial Only the endoscopist and pathologist were blinded Analyzed by the intention to treat principle |
|
| Participants | Patients (N = 77, AZA n = 37 and budesonide n = 38) aged 18‐67 years, with steroid dependant Crohn's ileocolitis or proximal colitis in remission (CDAI < 150)
Steroid‐dependency was described as individuals having at least 1 flare in the previous 6‐12 months which was successfully treated with steroids, but had relapsed again during steroid tapering or just after steroid withdrawal Patients included in the study must have recently flared (within 1 month of study inclusion) and were receiving a starting dose of oral prednisolone of 0.75 mg/kg a day that was tapered gradually to the lowest dose that effectively maintained remission for at least 1 month prior to randomization |
|
| Interventions | Azathioprine 2‐2.5 mg/kg/day or budesonide 6‐9 mg/day controlled‐release tablets (dose was dictated by prednisolone dependence, such that patients being maintained with 15‐30 mg/day of prednisolone would receive 9 mg/day BUD and those maintained on < 15 mg/day prednisolone would receive 6 mg/day BUD) for 1 year with extension out to 18 months in those still in remission at the 1 year mark | |
| Outcomes | Primary outcome: mucosal healing and histologic remission Secondary outcomes: annual relapse rate, time in remission, discontinuation of study medications, changes in CDAI or health‐related quality of life, safety Treatment failure was defined as "withdrawal of a patient from the study irrespective of cause" Relapse classification: Mild (CDAI 150–219), moderate (CDAI 220–450) and severe (CDAI > 450) |
|
| Notes | Compliance and use of other medications were also monitored | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Computer‐based random number generator |
| Allocation concealment (selection bias) | Unclear risk | Not adequately described |
| Blinding (performance bias and detection bias) All outcomes | High risk | Blinding was limited to the endoscopist and pathologist Patients were aware of the intervention they were receiving, but were unaware of the second study arm |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Missing data was accounted for with explanation 8 patients (6 due to relapse and 2 from adverse events) from the AZA arm and 14 patients from the budesonide arm (all due to relapse) failed Analysis was performed on an intention‐to‐treat basis |
| Selective reporting (reporting bias) | Low risk | Primary outcome and some secondary outcomes described in the methods section were reported as results |
| Other bias | Low risk | The study appears to be free of other sources of bias |