Single molar extraction in a patient on dual antithrombotic therapy
A 71-year-old man is referred to the oral surgery clinic for removal of the mandibular right first molar (46). The tooth has gross caries, a necrotic pulp, grade II mobility and a 7 mm buccal probing depth with suppuration; mouth opening is 32 mm and there is no trismus, floor-of-mouth swelling or submandibular lymphadenopathy. Medical history: atrial fibrillation on apixaban 5 mg twice daily, percutaneous coronary intervention eight months ago with a drug-eluting stent (clopidogrel 75 mg daily since), type 2 diabetes with HbA1c 8.4%, and hypertension controlled on amlodipine. eGFR is 62 mL/min. He is anxious and asks whether he must stop his "blood thinners" before surgery. The clinic has no access to platelet function testing, and a routine cardiology opinion takes two weeks. His last apixaban dose was this morning. He reports occasional gum bleeding when brushing and one episode of epistaxis lasting ten minutes last month.
1. How should his antithrombotic therapy be managed around the extraction, and what is your approach to local anaesthesia and analgesia? Justify each decision.
Read the worked answer
Two risks must be balanced, and they are not symmetrical.
Thrombotic risk. A drug-eluting stent placed eight months ago means he is still inside the 12-month window in which uninterrupted dual antiplatelet therapy is advised. Stopping clopidogrel now risks stent thrombosis, an event with a mortality of roughly 20–40% and far more dangerous than bleeding from one socket. Clopidogrel therefore continues.
Bleeding risk. A single, atraumatic extraction is a low-bleeding-risk procedure. Apixaban (half-life about 12 hours, no routine monitoring) does not need to be stopped; residual anticoagulation can be managed locally. Omitting one morning dose is an accepted option for higher-risk procedures, but here it buys little, because he remains on clopidogrel and platelet inhibition cannot be reversed by waiting, while leaving an infected, suppurating tooth untreated is a real harm. Do not bridge with low-molecular-weight heparin, and never stop both drugs.
Local anaesthesia. Buccal infiltration with 4% articaine and 1:100,000 adrenaline is preferred. A conventional inferior alveolar block is not absolutely contraindicated but carries a small risk of pterygomandibular haematoma, so use it only if infiltration fails, and aspirate meticulously in two planes.
Analgesia. Paracetamol, not NSAIDs, because of the additive antiplatelet effect and renal and gastrointestinal risk in a diabetic patient. Document the discussion and seek non-urgent cardiology advice for any future surgery.
2. Set out your local haemostatic protocol, your take-home instructions, and your response to a telephone call at 11 pm saying he is bleeding.
Read the worked answer
Before operating: confirm blood pressure is reasonably controlled (ideally below 140/90) and that he has eaten and taken his medication as usual.
During surgery: use a periotome and controlled luxation to expand the socket without fracturing the buccal plate; curette granulation tissue and irrigate. Pack the socket to the crest with oxidised regenerated cellulose or a collagen plug, place a figure-of-eight 3-0 silk suture under tension, then apply a moist gauze pack for a full 20 minutes before inspecting. Accept the socket only when there is no oozing from the socket or the soft tissue margins after a further five minutes of pressure.
Prescription: tranexamic acid 4.8% mouthwash, 10 mL rinsed for two minutes four times daily for five days, not swallowed, with nothing eaten or drunk for 30 minutes afterwards. Add paracetamol and a written advice sheet: no rinsing, spitting, sucking or straws for 24 hours, no smoking, cool soft diet, sleep with the head elevated, avoid strenuous exercise and hot food and drink, plus a contact number.
The 11 pm call: most late bleeding is soft-tissue oozing or a lost clot, not arterial. Ask how much blood, whether it is oozing or pumping, and whether he is dizzy, pale or tachycardic. If he is stable, advise firm pressure on rolled gauze for 20–30 minutes sitting upright, then a tranexamic acid rinse. If bleeding continues, is pulsatile, or there are signs of haemodynamic compromise, he must attend the emergency department; never try to manage significant haemorrhage by telephone. Review him the next morning, resuture or apply topical thrombin, and check haemoglobin if blood loss has been substantial.
3. Does the HbA1c of 8.4% change your plan? State your decision and the follow-up you would arrange.
Read the worked answer
Poor glycaemic control impairs neutrophil function, collagen synthesis and microcirculation, and is associated with increased post-extraction infection, delayed healing and dry socket. An HbA1c of 8.4% is above the ideal but is not an absolute bar to surgery, and the decision must be weighed against the harm of leaving a suppurating source of infection.
My decision: proceed, because this is source control rather than purely elective surgery, but optimise what can be optimised. Schedule an early-morning appointment, confirm he has taken his medication and eaten, avoid systemic steroids, prescribe chlorhexidine 0.12% rinses for one week, and give the extraction with minimal trauma. Defer any additional elective surgical or implant treatment until control improves.
Follow-up: telephone review at 48 hours and clinical review at five to seven days to detect infection, delayed healing or dry socket early; reinforce that smoking, if present, and poor control both worsen outcomes. Write to his physician asking for optimisation of diabetes management, since an HbA1c above 8% is a marker of risk for any future oral surgery. Explain to the patient that hyperglycaemia mainly raises infection and healing risk rather than bleeding risk, so it does not change the antithrombotic plan, and that good hygiene and early reporting of pain or swelling are his main protections.