Nine RCTs (n=556) were included.
Two trials mentioned allocation concealment; three trials blinded patients and health care staff; most trials reported drop-outs in detail; one trial reported an intention-to-treat analysis; and two trials measured patient compliance with therapy. The number of trial participants ranged from 20 to 271.
Crohn's disease (1 trial): there was no significant difference in disease remission rate after 6 months between a 7-week and 15-week glucocorticoid withdrawal programme.
Bone marrow transplant GVHD (1 trial): there was a quicker (P<0.05) but not more frequent resolution of GVHD in a 21-week versus a 12-week withdrawal programme. The other outcomes did not differ significantly.
Acute asthma attacks (5 trials): three of the trials used abrupt withdrawal (two of which used placebo tapering). There were no significant differences between different glucocorticoid withdrawal programmes for the following: lung function, relapses, hospital admissions, severe exacerbations, median length of stay in hospital, and symptom scores. In all of the trials, patients were receiving medium to high doses of inhaled glucocorticoids during the withdrawal programmes.
COPD (2 trials): in one trial, a 3-day stabilisation course of glucocorticoids followed by a 7-day withdrawal programme was related to statistically significant improved outcomes in lung function, arterial oxygenation and dyspnea on exertion, compared with a 7-day placebo withdrawal. There was no significant difference in acute exacerbation within 6 months. In another study there was no statistically significant difference in treatment failure between an 8-week glucocorticoid, 2-week glucocorticoid and a placebo withdrawal programme.