Crohn's disease: no benefit from anti-tuberculous chemotherapy.

Clinical bottom line (level 1b)

  1. Patients with Crohn's disease who had anti-tuberculous chemotherapy had more adverse effects (NNH = 5 at 2 years) , without fewer courses of corticosteroids.
  2. The effect on need for surgery or new stricture formation was unclear.
Swift et al: Gut 1994; 35: 363-368
Expires May 2003

The study

Double-blinded concealed randomised trial without intention-to-treat
Setting: three acute hospitals, UK and Germany

130 patients (aged range 16 to 70 years; mean 36, 54% female) with active Crohn's disease within the previous four months (diagnosed by clinical, histological and radiological evidence)

Excluded if
  • severe acute disease
  • toxic megacolon
  • hepatic or renal impairment
  • pregnant or planning to be pregnant
  • on anticoagulants, anticonvulsants or sulphonylurea


    • Control Group: (n = 65, 63 analysed): placebo
    Experimental Group: (n = 65, 63 analysed): rifampicin 450 mg po once daily if <50 kg; 600 mg po once daily if >50 kg; ethambutol 15 mg/kg daily; isoniazid 300 mg po once daily
    Patients continued other Crohn's disease medication as necessary.
    97% followed for 2 years
    Outcome notes:
    • new stricture formed : diagnosed radiologically

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    adverse effects 2 years 3
    (4.76%)    		 17
    (27.0%)    		 -467%
    (-1738% to -75%)    		 -22.2%
    (-34.4% to -10.1%)    		 -5
    (-10 to -3)
    surgery required 2 years 14
    (22.2%)    		 15
    (23.8%)    		 -7%
    (-103% to 43%)    		 -1.59%
    (-16.3% to 13.1%)    		 -63
    (NNT = 6 to infinity;
    NNH = 8 to infinity)
    required corticosteroids 2 years 54
    (85.7%)    		 54
    (85.7%)    		 0%
    (-15% to 13%)    		 0.00%
    (-12.2% to 12.2%)    		 inf
    (NNT = 8 to infinity;
    NNH = 8 to infinity)
    new stricture formed 2 years 1
    (1.59%)    		 5
    (7.94%)    		 -400%
    (-4059% to 40%)    		 -6.35%
    (-13.7% to 1.00%)    		 -16
    (NNT = 7 to infinity;
    NNH = 100 to infinity)

  • There were no significant differences in the CDAI scores noted at any time.

    • Citation

    1. Swift GL, Srivastava ED, Stone R, et al: controlled trial of anti-tuberculous chemotherapy for two years in Crohn's disease. Gut 1994; 35: 363-368
    Search Terms: Crohn* in Cochrane
    Contributor: Chris Ball and Clare Wotton, October 2000
    Reviewer:

    Clinical Question.
    Patient Crohn's disease
    Intervention or Exposure anti-tuberculous therapy
    Comparison placebo
    Outcome need for corticosteroids, side effects