Abstract
Optimal management of patients on warfarin who require oral surgery has been controversial. Historically the clinician has had to balance the risk of thromboembolism by reducing or stopping anticoagulant therapy, against the risk of triggering excessive postextraction bleeding if anticoagulation is maintained at therapeutic levels during surgery. Patients on anticoagulants have impaired fibrin formation that is more susceptible to normal fibrinolysis and is believed to be the major cause of postextraction bleeding. This report reviews normal haemostasis, oral anticoagulants, thromboembolism, the fibrinolytic system and oral fibrinolysis and the traditional methods of managing patients on oral anticoagulants. A technique, derived from that originally described by Sindet-Pedersen et al. (1989), was used where teeth were extracted from patients in whom therapeutic warfarin levels were maintained. The antifibrinolytic agent, tranexamic acid was used as a mouth rinse four times a day for seven days to reduce fibrinolysis and subsequent bleeding.
The aim of this study was to verify that this technique is a safe, simple, effective and acceptable method of patient management. The second objective was to identify potential risk factors that may increase the likelihood of bleeding.
One hundred consecutive warfarin patients with an International Normalised Ratio (INR) between 1.9 and 4.0 on the day of surgery and who required dental extractions were recruited to the study. Following removal of teeth, patients were instructed to use 5 millilitres of 10% tranexamic acid syrup as a mouthrinse 4 times a day for 7 days, to record bleeding that required pressure to control and their mouthrinse usage. The researcher collected demographic data, details of the state of anticoagulation, details of the surgery and details of bleeding that required additional management. Data entry and analysis were conducted using the statistical computer programme SPSS. Descriptive statistics were produced for the sample demographics, the haemostasis screen, the postextraction bleeding profile, the teeth removed and mouthrinse acceptance and utilisation. Identification of potential risks that might increase the likelihood of bleeding was carried out using the chi-square test and the independent samples t test as appropriate. Multivariate analysis was then performed using logistic regression.
Of the 100 patients treated, 8 reported bleeding at home after day 1 that was controlled by pressure. A further 8 patients reported to the researcher's dental surgery where bleeding was controlled by local measures. No patients required hospital admission for systemic management of bleeding.
Statistically significant risk factors for postextraction bleeding were: A pre-extraction INR equal to or greater than 3.0; periodontal pockets equal to or greater than 5 millimeters in depth; maxillary molar teeth and patients on long term aspirin therapy that was stopped seven days before the extraction.
Factors not statistically significant for an increased risk of bleeding included: The number of teeth removed; raising a mucoperiosteal flap; removing bone; pre-extraction activated partial thromboplastin time; pre-extraction bleeding time; time on anticoagulants and compliance with tranexamic acid mouthrinses.
The post-operative use of tranexamic acid mouthrinses in patients who have teeth removed at therapeutic warfarin levels is a safe, simple, effective and acceptable method of reducing postextraction bleeding.