Stopping a bleed

Most bleeding after an extraction is local and stops with local measures; the cases that go wrong, on my reading of the reports, tend to be the ones where a systemic cause sat behind a local wound and nobody had asked the question that would have found it. The commonest hidden contributors are drugs the patient does not think of as medication — over-the-counter aspirin, an anti-inflammatory taken for the pain that evening, and herbal supplements, though for the most studied of them, ginkgo, randomised trials found no effect on standard tests of haemostasis — and inherited bleeding disorders, of which von Willebrand disease is the most common and, on a genetic estimate, probably under-diagnosed. A short structured bleeding history, built on the principle of the ISTH bleeding assessment tool, asks about previous extractions, surgery, nosebleeds, bruising and heavy menstrual bleeding, adds family history, and is the cheapest test there is — though bleeding scores discriminated poorly between diagnosed and undiagnosed bleeders in one referred population, and their yield in unselected dental patients has not been measured. Tranexamic acid works by protecting a clot from being dissolved rather than by making one, and its evidence in dentistry is small: benefit against placebo in old trials in patients on warfarin and in haemophilia, and in one modern trial in patients on direct oral anticoagulants no reduction in overall bleeding but fewer delayed bleeds and fewer after multiple extractions. A bleed stops being a dental problem when there are signs of volume loss, when swallowed blood is being vomited, when swelling spreads into the floor of the mouth or the neck, or when competent local measures fail — and hypotension is a late sign, not an early one.