UPETSG 1970.

Study characteristics
Methods	Study design: multicentre, randomised controlled trial Method of randomisation: telephone to Drug Assignment Center Blinding: double‐blind Duration: October 1968 to August 1970 Exclusions post‐randomisation: 1 Losses to follow‐up: 1
Participants	Country: USA Setting: 14 centres No. of participants: 160: 82 in urokinase group, 78 in heparin group Age (years): not clearly stated Sex: 47 men, 35 women in urokinase group; 45 men, 33 women in heparin group Inclusion criteria: well‐documented clinical episode suggesting PE had occurred within 5 days of start of therapy Exclusion criteria: recent operation, contraindication to use of anticoagulant or thrombolytic therapy
Interventions	Treatment group: 2 brands of urokinase ‐ Urokinase (Abbott) and Winkinase (Sterling‐Winthrop) ‐ were given randomly by 12‐hour infusion. Loading dose of 2000 CTA IU/lb in 10 minutes, then 2000 CTA IU/lb for 12 hours via infusion pump Control group: heparin; loading dose of 75 IU/lb, then 10 IU/lb/h for 12‐hour infusion Both groups then received heparin for a minimum of 5 days, followed by heparin or warfarin therapy for a total of 14 days. Dosage not clear Follow‐up: 6 to 18 hours after completion of test drug infusion, then at 2 weeks and 3, 6, and 12 months
Outcomes	Morbidity during 2‐week study period Complications Pulmonary angiography (24 hours) Lung scanning Haemodynamic measurements
Funding	No details were provided
Declaration of interests	No details were provided
Notes	—
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Drug Assignment Center
Allocation concealment (selection bias)	Low risk	Telephone to Drug Assignment Center
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Double‐blind
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blind
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing data
Selective reporting (reporting bias)	Unclear risk	No information provided
Other bias	Unclear risk	Small sample size may be a source of other bias