Ulcerative colitis: rectal ASA is better than rectal steroids for improving active disease.

Clinical bottom line (level 1a)

  1. Patients with active distal ulcerative colitis who take rectal corticosteroids are more likely to have symptomatic and endoscopic improvement (NNT = 4 at 3 weeks) , and are more likely to be in symptomatic remission (NNT = 12 at 3 weeks) .
  2. There is no clear difference between rectal budesonide and other rectal steroids.
  3. Rectal aminosalicylic acids are more effective than corticosteroids at inducing symptomatic remission (NNT = 10 at 4 weeks) , and endoscopic remission (NNT = 9 at 4 weeks) .
  4. Rectal aminosalicylic acids are more effective than budesonide at inducing symptomatic remission (NNT = 20 at 4 weeks) .
Marshall and Irvine: Gut 1997; 40: 775-781
Expires May 2003

The study

Systematic review of all randomised controlled trials of
  • Patients: active distal ulcerative colitis
  • Intervention: rectal corticosteroids for at least two weeks compared with another medication or placebo
  • Outcome: objective measurement of subsequent improvement or remission (using symptomatic, endoscopic or histological criteria chosen by original authors)


    • Articles found in all languages using MEDLINE, EMBASE, 1966 to 1996 (search terms: 'inflammatory bowel disease', 'therapy', and 'topical administration', 'enema' or 'suppository' ) and bibliographies and review article references were also searched.

    Selection criteria: as above
    Appraisal criteria: selected and appraised by two independent reviewers using set criteria (not detailed in text)
    Articles excluded if:

    33 studies found - 11 double-blinded RCTs, 1 cohort study
    • Studies chosen if documented disease margin distal to the splenic flexure on radiographic studies or < 60 cm from anal verge at flexible sigmoidoscopy or colonoscopy.
    • Patient received hydrocortisone, prednisolone, betamethasone.
    No information on heterogeneity.

    The evidence

    Outcome Time to outcome CER OR
    (95% CI)    		 NNT
    (95% CI)
    rectal corticosteroids v. placebo: symptomatic improvement 2-3 weeks /
    (34%)    		 0.21
    (0.07 to 0.71)     		4
    (3 to 14)
    rectal corticosteroids v. placebo: endoscopic improvement 2-3 weeks /
    (38%)    		 0.27
    (0.10 to 0.77)     		4
    (3 to 17)
    rectal corticosteroids v. placebo: symptomatic remission 2-3 weeks /
    (9%)    		 0.07
    (0.02 to 0.29)     		12
    (11 to 16)
    rectal corticosteroids v. placebo: endoscopic remission 2-3 weeks /
    (17%)    		 0.34
    (0.10 to 1.20)     		10
    (NNT = 7 to infinity;
    NNH = 37 to infinity)
    rectal corticosteroids v. rectal 5-ASA: symptomatic improvement 2-7 weeks /
    (34%)    		 1.36
    (0.88 to 2.09)     		-14
    (NNT = 36 to infinity;
    NNH = 6 to infinity)
    rectal corticosteroids v. rectal 5-ASA: endoscopic improvement 2-7 weeks /
    (38%)    		 1.06
    (0.61 to 1.85)     		-72
    (NNT = 9 to infinity;
    NNH = 7 to infinity)
    rectal corticosteroids v. rectal 5-ASA: symptomatic remission 2-7 weeks /
    (9%)    		 2.42
    (1.72 to 3.41)     		-10
    (-18 to -6)
    rectal corticosteroids v. rectal 5-ASA: endoscopic remission 2-7 weeks /
    (17%)    		 1.89
    (1.29 to 2.76)     		-9
    (-37 to -26)
    rectal corticosteroids v. rectal 4-ASA: symptomatic improvement 2-7 weeks /
    (34%)    		 3.88
    (1.29 to 11.6)     		-3
    (-17 to -2)
    rectal corticosteroids v. rectal budesonide: symptomatic improvement 2-7 weeks /
    (34%)    		 2.08
    (0.84 to 5.14)     		-6
    (NNT = 26 to infinity;
    NNH = 3 to infinity)
    rectal corticosteroids v. rectal budesonide: endoscopic improvement 2-7 weeks /
    (38%)    		 1.40
    (0.87 to 2.25)     		-12
    (NNT = 31 to infinity;
    NNH = 5 to infinity)
    rectal corticosteroids v. rectal budesonide: symptomatic remission 2-7 weeks /
    (9%)    		 0.85
    (0.44 to 1.63)     		80
    (NNT = 21 to infinity;
    NNH = 20 to infinity)
    rectal corticosteroids v. rectal budesonide: endoscopic remission 2-7 weeks /
    (17%)    		 1.14
    (0.69 to 1.88)     		-52
    (NNT = 22 to infinity;
    NNH = 9 to infinity)
    rectal 5-ASA v. rectal budesonide: endoscopic improvement 4 weeks /
    (38%)    		 0.58
    (0.27 to 1.22)     		8
    (NNT = 4 to infinity;
    NNH = 21 to infinity)
    rectal 5-ASA v. rectal budesonide: endoscopic remission 4 weeks /
    (17%)    		 0.95
    (0.43 to 2.10)     		140
    (NNT = 11 to infinity;
    NNH = 8 to infinity)
    rectal 5-ASA v. rectal budesonide: symptomatic remission 4 weeks /
    (9%)    		 0.41
    (0.18 to 0.94)     		20
    (14 to 200)

    • Adverse effects were poorly reported. Budesonide caused less adrenocorticoid axis suppression than conventional corticosteroids (weighted mean difference for cortisol concentration 119 nmol/l).

    Comments

    1. There was not enough data to compare rectal and oral corticosteroids, or compare foam enemas and suppositories
    2. Little information given on the trial participants - assumed to have mild-to-moderate ulcerative colitis.

    Citation

    1. Marshall JK, and Irvine EJ: Rectal corticosteroids versus alternative treatments in ulcerative colitis: a meta-analysis. Gut 1997; 40: 775-781
    Search Terms: colit* in Cochrane
    Contributor: David Ford and Chris Ball, October 2000
    Reviewer:

    Clinical Question.
    Patient active ulcerative colitis
    Intervention or Exposure rectal corticosteroids
    Comparison other medication or placebo
    Outcome remission