van Hoogstraten 2000.

Methods	Randomised parallel group trial.
Participants	Country: Netherlands. Publication language: English. Inclusion criteria: 1. Patients with primary sclerosing cholangitis based on the findings on endoscopic retrograde cholangiography and typical histological lesions (pericholangiolar onion‐skin fibrosis) in combination with both a serum alkaline phosphatase level elevated to more than twice the upper limit of normal and the presence of inflammatory bowel disease. 2. All patients had been treated with ursodeoxycholic acid (mean dose, 12 mg/kg body weight/day) for at least five months without achieving biochemical remission. Excluded criteria: Age over 65 years old, use of immunosuppressive drugs such as corticosteroids, azathioprine, cyclosporin, or methotrexate, pregnancy, evidence of primary sclerosing cholangitis associated autoimmune hepatitis, previous cholecystectomy, presence of a biliary stent, and cirrhosis with a Child‐Pugh score over six. Participants: 1. Budesonide (9 mg/day) group (n = 6) Mean age (years +/‐ SD): 46.4 +/‐ 9.7. Male/female: 4/2. Inflammatory bowel disease: 4. 2. Budesonide (3 mg/day) group (n = 6) Mean age (years +/‐ SD): 38.9 +/‐ 12.3. Male/female: 4/2. Inflammatory bowel disease: 4. 3. Prednisone (10 mg/day) group (n = 6) Mean age (years +/‐ SD): 44.5 +/‐ 10.9. Male/female: 6/0. Inflammatory bowel disease: 4.
Interventions	Budesonide (9 mg/day) group: 9 mg/day budesonide (three 3‐mg budesonide capsules and two placebo tablets of 5 mg prednisone) for eight weeks. Budesonide (3 mg/day) group: 3 mg/day budesonide (one 3‐mg budesonide capsule, two placebo capsules of 3‐mg budesonide, and two placebo tablets of 5 mg prednisone) for eight weeks. Prednisone (10 mg/day) group: 10 mg/day prednisone (two 5‐mg tablets of prednisone and three 3‐mg placebo capsules budesonide) for eight weeks.
Outcomes	Serum alkaline phosphatases activity at the end of treatment. Serum bilirubin level at the end of treatment. Adverse events.
Notes	Follow‐up time: eight weeks. We obtained additional information from the authors.
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Low risk	Random table.
Allocation concealment?	Low risk	Coded medication.
Blinding? All outcomes	Low risk	Identical placebos.
Incomplete outcome data addressed? All outcomes	Low risk	No attrition.
Free of selective reporting?	Low risk	No selective reporting.
Baseline imbalance	Low risk	No imbalance.
Early stopping	Low risk	The trial was carried out as planned.