Sindet‐Pedersen 1989.

Methods	Trial design: double‐blind RCT Follow up: 7 days
Participants	Countries: Denmark and Sweden Setting: 2 hospitals: Aarhus University Hospital in Aarhus, Denmark, and Karolinska Hospital in Stockholm, Sweden Inclusion criteria: Individuals with a cardiac‐valve stenosis, a cardiac‐valve prosthesis or a vascular prosthesis on oral anticoagulant treatment at a therapeutic level ‐ who were to undergo oral surgery ‐ PP between 10% ‐ 20% or TT result between 5% ‐ 12% (USA: PT ratio of 1.4‐2.0 or an INR level between 2.5 ‐ 4.8) Exclusion criteria: ‐ No informed consent ‐ Use of aspirin and other antiinflammatory drugs in the 2 weeks before surgery Sample size: TXA group: ‐ Total number of participants: N = 19 ‐ Mean number of surgically treated teeth per participant: N = 3.3 Placebo group: ‐ Total number of participants: N = 20 ‐ Mean number of surgically treated teeth per participant: 2.8 Numbers analysed/randomised: ‐ Total: N = 39/39 Participant characteristics: ‐ Age (mean, range): TXA group: Aarhus 56.3 years (20 ‐ 72), Stockholm mean 59.7 (47 ‐ 69); placebo group: Aarhus 48.7 years (19 ‐ 65), Stockholm 62.6 years (38 ‐ 77) ‐ Sex N (M/F): N = 22/17 ‐ INR PP between 10% ‐ 20% or TT result between 5% ‐ 12% (USA: PT ratio of 1.4 ‐ 2.0 or an INR of 2.5 ‐ 4.8)
Interventions	Intervention group: Before suturing: irrigation with 10 mL 4.8% TXA solution Postoperatively: rinsing with 10 mL TXA mouthwash for 2 min 4x daily for 7 days Control group: Before suturing: irrigation with 10 mL placebo solution Postoperatively: placebo mouthwash for 2 min 4x daily for 7 days
Outcomes	1. Number of postoperative bleeding episodes requiring intervention 2. Side effects or other adverse events 3. Number of minor postoperative bleeding episodes (defined as self‐limiting, usually with local pressure, that does not require medical attention) 4. Major bleeding, requiring transfusion of packed red blood cells Bleeding was defined as postoperative bleeding that could not be controlled by compression with gauze for 20 min
Notes	All intervention and control groups were eligible for the review
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Random numbers were used to assign consecutive patients to treatment groups." (Methods, page 841)
Allocation concealment (selection bias)	Low risk	"Random numbers were used to assign consecutive patients to treatment groups. Each number corresponded to a medication package with 28 bottles, each containing 10 mL of one of the solutions. Randomization was carried out independently for each of the clinics." (Methods, page 841)
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"The study was carried out as a double‐blind investigation with parallel groups, comparing the hemostatic effect of a sterile aqueous solution of 4.8 percent tranexamic acid used as a mouthwash with that of a placebo solution " (Methods, page 841) Participants followed the same treatment schedule. "Before sutures were applied, the operative field was irrigated with 10 ml of the test solution. After surgery, the patients were instructed to rinse their mouths with 10 ml of the assigned solution for two minutes and then to expectorate the solution (four times a day for seven days)." (Methods, page 841)
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	This was not clearly discussed in the article
Incomplete outcome data (attrition bias) All outcomes	Low risk	"Forty‐five patients were recruited…" "...Six patients did not meet the requirements for inclusion. Thus, 39 patients were eligible for further analysis." (Methods, page 841)
Selective reporting (reporting bias)	Low risk	All pre‐specified outcome measures are reported for all trial participants
Other bias	Unclear risk	It is unclear whether participants could undergo multiple randomisations