Warfarin INR target for safe dental extraction

Results

Dental extraction in anticoagulated patients with dicoumarinics appears predicTable if the INR ranges are controlled, extracting a single tooth and applying local haemostatic measures to control the haemorrhage. The risk of postoperative bleeding of these patients appears similar to the risk of healthy individuals when the INR value does not exceed 2.2. If we take into account other conditions such as INR greater than 3 or several extractions, the evidence available is limited for answering this question, therefore in these cases it is recommended to contact the patient’s doctor to decide a treatment plan (Grade C recommendation). Without doubt, in this review we have lacked more studies on acenocoumarol, as the majority included patients treated with warfarin or did not specify the vitamin K antagonist medication administered. The studies designed to assess the risk of bleeding must very clearly define aspects such as: the monitoring time of the patient, which we recommend being at least 7 days; the main risk variable, which we recommend being dental extraction, as the medical condition of the patient may vary at different times; the INR range that will be established to include anticoagulated patients; the target population (as the polymorphism of isoenzyme CYP2C9 prevalent in the Asian population affects the clearance of warfarin); as well as also registering whether the patient has previously been treated with antibiotics and concomitant treatment with platelet antiaggregants.

Clinical practice guideline of the spanish society of oral surgery for oral surgery in patients with coagulation disorders. Copyright: © 2024 Medicina Oral S.L. Creative Commons Attribution 2.5 International (https://creativecommons.org/licenses/by/2.5/).

5.1 Warfarin Therapy

According to a systematic review and meta-analysis (2009) while performing minor dental procedures, continuing the regular warfarin dose does not cause increased bleeding when compared with adjusting or discontinuing its dose [113]. A 2015 systematic review concluded that for patients undergoing dental extractions, a regular dose of warfarin can be continued if INR is within the therapeutic range [114]. The benefits of warfarin therapy discontinuation must be weighed carefully against the risk of thromboembolism. If deemed essential, warfarin must be discontinued five days before major surgery and restarted 12–24 hours postoperatively [115]. INR must be measured a day before surgery to monitor adequate reversal of anticoagulation [116]. When warfarin therapy is discontinued in patients with an initial INR of 2.0–3.0 for about 4 to 5 days, the INR falls to the normal range (less than 1.5) on the day of surgery which is considered safe for dental procedures and associated with acceptable risk of perioperative bleeding [117]. The INR reaches the therapeutic range (2.0) after around 3 days of restarting the warfarin therapy [116, 118, 119] (Fig. 1).

Dental Management Considerations for Patients with Cardiovascular Disease—A Narrative Review. Copyright: © 2022 The Author(s). Published by IMR Press. Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/).

5. Anticoagulants

The ratio of the patient prothrombin time to the mean prothrombin time raised to the power of international sensitivity index (ISI) value is the INR of the patient. It determines the anticoagulation status of a patient [104]. It is warranted to know the INR value of a patient on anticoagulation therapy before a dental procedure. It is recommended to check the INR 72 hours prior to an invasive dental procedure in a patient taking long-term anticoagulant therapy and stably anticoagulated on warfarin [105]. A physiologically normal patient has an INR value of 1, and a therapeutic range of 2.0 to 3.0 is considered safe for most indications due to the reduced risk of thromboembolic events [106]. In the case of prosthetic heart valves, a higher INR range of 2.5–3.5 is required [107]. Guidelines on the management of patients on oral anticoagulant therapy (2007) recommend not to discontinue the anticoagulants in patients with stable INR in the range of 2.0–4.0 as the risk of significant bleeding is low for most outpatient dental procedures [105]. American Academy of Oral Medicine (AAOM) Clinical Practice Statement 2016 stated that an INR value of 3.5 (up to 4.0 by some experts) is safe for moderately invasive dental surgical procedures like simple tooth extractions [108]. For non-invasive dental procedures, there is no need to maintain this safety margin [109].

Dental Management Considerations for Patients with Cardiovascular Disease—A Narrative Review. Copyright: © 2022 The Author(s). Published by IMR Press. Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/).

Subgroups

In the warfarin group, the INR level was associated with a risk of postoperative bleeding. The risk of any type of postoperative bleeding was significantly higher in patients with an INR of 2.6–3.1 compared with those with an INR of 2.0–2.5 (unadjusted OR 6.00, 95% CI: 1.46–24.73; adjusted OR 8.50, 95% CI: 1.48–48.89). Two patients in the warfarin group required cessation of anticoagulant therapy. The first, a 75-year-old female with a preoperative INR of 3.0 underwent surgical removal of tooth 28 (FDI World Dental Federation notation) and concomitant apical surgery on the buccal roots of teeth 26. Five days postoperatively, she developed bleeding with an INR of 6.8. The bleeding was controlled with compression, warfarin was discontinued, and she received vitamin K. A potential cause of the elevated INR was high paracetamol intake. The second patient, a 73-year-old male, had a preoperative INR of 2.4 and underwent surgical removal of tooth 37. Three days postoperatively, the patient developed local infection and bleeding. Systemic antibiotics were prescribed, and warfarin was discontinued. Additional suturing and debridement were performed seven days postoperatively. Two other patients in the warfarin group had a preoperative INR of 3.1. Both experienced postoperative bleeding with one patient presenting with grade 2 bleeding.

Continuous use of direct oral anticoagulants during and after simple and surgical tooth extractions: a prospective clinical cohort study. © The Author(s) 2025. Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/).

Discussion

It is now well-established that uncomplicated dental extractions can be safely carried out without interruption of VKAs therapy provided INR is maintained <3.5 and local hemostatic measures are performed to control bleeding (Nematullah et al., 2009). However, no clear guidelines exist on the perioperative management of patients receiving DOAC (Brennan et al., 2019). Recommendations range from the continuation of DOAC during dental extractions to omitting one or two doses of the drug before the procedure (Brennan et al., 2019). A recent survey by Precht et al. (2019) has demonstrated 94% of dental practitioners continue with VKAs during single tooth extractions but 62% of them interrupt DOAC therapy. Given such ambiguity, the results of our review present some clarity on the risk of hemorrhage with uninterrupted DOAC as compared to uninterrupted VKAs after dental extractions. We found a statistically significant reduced risk of hemorrhage with DOAC as compared to VKAs. Individually, none of the included studies noted a significant difference and all of them concluded that patients on DOACs have a similar bleeding tendency as compared to VKAs.

Bleeding Outcomes After Dental Extraction in Patients Under Direct-Acting Oral Anticoagulants vs. Vitamin K Antagonists: A Systematic Review and Meta-Analysis. Copyright © 2021 Hua, Huang and Huang. Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/).

Results

In anticoagulated patients undergoing simple extractions it is not recommended to stop the anticoagulant treatment, recommending the application of local haemostatic measures (Grade C recommendation). No local haemostatic measure has been demonstrated as superior to the others, therefore it is recommended to use those which have more evidence, such as sutures, collagen or cellulose sponges, as well as compression with gauze soaked in tranexamic acid (Grade C recommendation). The appearance of postoperative haemorrhages is associated with the age of the patient, therefore in older patients undergoing simple extractions the use of local haemostatic measures is recommended (Grade C recommendation). Future lines of research should study the effect of different local haemostatic measures through randomised clinical trials evaluating the effect of these measures on anticoagulated patients undergoing extractions to allow valid conclusions to be obtained. These trials must be carried out on large samples of patients where the anticoagulation is not stopped and with INR in the range recommended in the main guidelines and protocols. With regard to the above, patients with an INR of between 2 and 4 should be included, as it is common to find patients with INR levels over 3, and to thereby be able to evaluate whether these haemostatic measures continue to be effective for such INR values. Additionally, the effectiveness of these haemostatic measures should be studied in patients who combine anticoagulant and antiaggregant treatment, to thereby be able to create protocols which guide clinical practice in this type of patient.

Clinical practice guideline of the spanish society of oral surgery for oral surgery in patients with coagulation disorders. Copyright: © 2024 Medicina Oral S.L. Creative Commons Attribution 2.5 International (https://creativecommons.org/licenses/by/2.5/).

Dental procedures: extraction and implant-related factors

Previous studies reported that extracting up to three teeth was safe if the PT-INR was <3.5 in the absence of other bleeding risks.31,34 This study revealed no significant difference between the number of teeth treated and postoperative bleeding. However, because scaling with a low risk of bleeding involves the treatment of at least one-third of the jaw or all teeth, dental treatment variables require adjustment. Multivariate logistic regression analysis showed that the risk of postoperative bleeding increased if ≥4 teeth were treated.

Dental Complications in Antithrombotic Patients: Evidence From a Nationwide Cohort and a Single-Institution Dataset. © 2026 The Authors. Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/).

Hemostatic measures in patients with anticoagulant or antiplatelet therapy

For most patients receiving oral anticoagulants and/or antiplatelet agents, local compression is sufficient to active hemostasis [109, 110], although, in particular, an increased INR, difficulty in achieving hemostasis intraoperatively, and higher serum creatinine concentrations may be additional risk factors for postoperative bleeding [111]. If further measures are necessary, the use of sutures, collagen, oxidized cellulose, absorbable sponges, chitosan, bone wax, and fibrin glue, for example, have proven to be effective [112–114] and can be used to achieve successful hemostasis even with INR values outside the therapeutic range [115]. However, evidence as to which hemostatic regimen should be preferred cannot be extracted from the literature [86, 109, 112]. In a systematic Cochrane literature review, Engelen et al. evaluated the efficacy of antifibrinolytics in preventing bleeding complications in patients receiving oral anticoagulation (vitamin K antagonists or DOACs) undergoing dental surgery. Overall, a positive effect of locally applied tranexamic acid (mostly 5% concentration, 3–4 × /days) to prevent oral bleeding events could be demonstrated [116].

Bleeding disorders in implant dentistry: a narrative review and a treatment guide. © The Author(s) 2022. Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/).