Petruson 1974.

Methods	Allocation: double‐blind, parallel‐group, placebo‐controlled randomised trial with 10 days of treatment and 10 days of follow‐up
Participants	Setting: inpatients, Sweden Sample size: 68 Number randomised: 31 in intervention group, 37 in placebo group Number completed: not reported Participant (baseline) characteristics: Age: intervention group mean 56 years; control group mean 56 years Gender: not recorded Risk factors: 69% of participants were taking acetylsalicylic acid and 26% had hypertension Inclusion criteria: all hospitalised patients with epistaxis Exclusion criteria: none mentioned
Interventions	Intervention group: Cyklokapron (tranexamic acid) 1 g, 3 times daily, started 1 hour after hospitalisation, for 10 days Comparator group: placebo 1 tablet, 3 times daily, started 1 hour after hospitalisation, for 10 days Use of additional interventions: all but 3 of the participants were treated with anterior and/or posterior nasal packing at presentation
Outcomes	Outcomes of interest in the review Primary outcomes Control of epistaxis: re‐bleeding ‐ proportion of patients with re‐bleeding in a period of up to 10 days post‐intervention. This was calculable from the data presented. "Adverse effects" (severity not defined) Secondary outcomes Length of hospital stay (days) Other outcomes reported by the study Number of episodes of re‐bleeding (not patients) requiring further intervention Frequency and severity of re‐bleeding episodes using a point scale (0 points = no bleeding to 6 points = large amount of bleeding requiring packing) measured in 12‐hour periods over 10 days. Minimum number = 0 (no bleeding at any point); maximum number (theoretical) = 6 x 10 x 2 = 120). The authors felt that measurement of blood loss was not possible due to the patients swallowing a considerable volume. At the end of the 10‐day study period the participants were brought back to the hospital and questioned about whether any further bleeding had occurred after discharge.
Declarations of interest	None stated
Funding	None declared
Notes	Participants lost to follow‐up: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Assignment ... by random numbers" Comment: this was probably done
Allocation concealment (selection bias)	Unclear risk	Comment: no information was provided about allocation concealment
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Double blind", "The labels of the bottles bore only the patients' serial number" Comment: this was probably done. The bottles did not reveal whether they contained the treatment or placebo
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Double blind" Comment: this was probably done
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: "One to three days after the drug treatment was finished the patients visited the hospital again" Comment: it is not stated whether all patients returned or not
Selective reporting (reporting bias)	Unclear risk	A standard bleeding score was used for all patients Hospitalisation time was recorded from patient records Adherence to these protocols would prevent selective reporting Number of units required for blood transfusion was not recorded
Other bias	Unclear risk	Quote: "The tampons must be regarded as errors in the evaluation of the therapy effect. When the tampons were taken away or moved small bleeding sometimes started. These small bleedings were not recorded." Comment: this will have had an unknown effect on the overall outcome