Knox 1994.

Methods	Randomised clinical trial.
Participants	Country: USA. Number randomised: 24. Post‐randomisation drop‐outs: 3 (12.5%). Revised sample size: 21. Mean age: 37 years. Females: 7 (33.3%). Separate data for the subgroup with ulcerative colitis: no. Inclusion criteria: Primary sclerosing cholangitis documented by characteristic findings on endoscopic retrograde cholangiopancreatography and liver biopsy. Exclusion criteria: Cytopenia (white blood cell count < 4000 cells/mm3, platelets < 100,000 cells/mm3, or haemoglobin < 10.0 g/dL). Significant cardiac or renal disease (serum creatinine > 2 mg/dL). Pregnancy, lactation, or lack of effective contraceptive methods. Alcoholism. Signs of liver failure manifested by ascites, encephalopathy, variceal bleeding, or muscle wasting. Dominant common bile duct strictures. Follow‐up: 48 months.
Interventions	Participants were randomly assigned to 1 of 2 groups. Group 1: methotrexate 5 mg every 12 hours (15 mg/wk) for 24 months (n = 11). Group 2: identical placebo for 24 months (n = 10).
Outcomes	No outcomes of interest were reported.
Notes	Reasons for post‐randomisation drop‐out: Colectomy for ulcerative colitis (1 participant in the treatment group). Non‐compliance (1 participant in the placebo group). Finding of an unusual bile duct mass of unknown nature seen on protocol endoscopic retrograde cholangiopancreatography (then diagnosed as cholangiocarcinoma) (1 participant in the placebo group).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Comment: This information was not available.
Allocation concealment (selection bias)	Unclear risk	Quote: "The code was broken on patients who were judged to be treatment failures". Comment: Additional details were not available.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "A double‐blind controlled trial of oral‐pulse methotrexate therapy in the treatment of primary sclerosing cholangitis.…Methotrexate (or placebo) was administered orally each week in three divided doses of 5 mg every 12 hours (15 mg/wk) for 2 years in a double‐blind manner. Identical methotrexate and placebo tablets were kindly provided by Lederle laboratories".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "A double‐blind controlled trial of oral‐pulse methotrexate therapy in the treatment of primary sclerosing cholangitis.Methotrexate (or placebo) was administered orally each week in three divided doses of 5 mg every 12 hours (15 mg/wk) for 2 years in a double‐blind manner. Identical methotrexate and placebo tablets were kindly provided by Lederle laboratories".
Incomplete outcome data (attrition bias) All outcomes	High risk	Comment: Post‐randomisation drop‐outs may be related to the treatment that participants received.
Selective reporting (reporting bias)	High risk	Comment: No published protocol was available; no outcomes of interest were reported.
For‐profit bias	High risk	Quote: "Identical methotrexate and placebo tablets were kindly provided by Lederle laboratories..Supported by General Research Center grant MOlRR00054 from the National Institutes of Health and Lederle Laboratories, Pearl River, New York". Comment: The trial was funded by a party with vested interest in the results.
Other bias	Low risk	Comment: no other bias.