Bleeding after a tooth extraction: how to stop it

Most people reading this have gauze in their mouth and a sink they have already looked into more than once. The urgent material is therefore first, because the situations that need help are not always the ones that look most dramatic. If you were given written post-operative instructions after your surgery, those instructions govern. This page is general, and where the two differ, follow the sheet you were handed.

Everything else on this page concerns the ordinary version of this problem, which is oozing rather than bleeding, and which settles with pressure applied properly for long enough. The distinction is not how red it looks. It is whether fresh blood is welling back into the socket while you watch.

The mouth produces saliva continuously, and blood colours it out of all proportion to its volume. A quantity of blood too small to measure will turn a whole mouthful pink, and the same quantity swirled into a basin of water produces a red cloud that looks like a serious event. Nothing you can see in a sink tells you how much blood has actually been lost.

The same effect operates overnight. A pillowcase with a palm-sized red-brown patch on it in the morning is the ordinary result of several hours of oozing mixed with saliva from a slightly open mouth, and it is one of the more common reasons patients make contact in the first twenty-four hours genuinely frightened. Putting a dark towel over the pillow before the first night removes the question entirely.

The test that actually distinguishes the two takes a minute. Sit upright under a good light, wipe the area gently with a clean, damp piece of gauze — wipe, do not scrape into the socket — and then watch it for sixty seconds. If bright red blood wells up and refills the socket while you are looking, that is active bleeding and pressure is the answer. If what you see is a dark, still clot with pink saliva around it, that is oozing, and it needs nothing done to it.

Some blood is swallowed, and that accounts for two things patients are rarely warned about. Nausea on the evening of surgery is common, and swallowed blood is a frequent cause of it rather than the analgesia, which usually takes the blame. Stools darker than usual for a day or two afterwards often have the same explanation. That explanation should not be stretched indefinitely: black, tarry stools, stools that stay dark beyond a couple of days, or dark stools in anyone taking an anticoagulant need a medical assessment rather than being written off as swallowed blood. Vomiting blood, or material resembling coffee grounds, is a different matter again and belongs in the warning above.

Expect saliva to stay pink-tinged for the rest of the day of surgery, and sometimes into the following morning. That is not bleeding. Treating it as bleeding — with repeated gauze, repeated inspection and repeated rinsing — is one of the reliable ways to turn a settled socket back into a bleeding one.

Gauze does not stop bleeding by absorbing blood. It stops bleeding by holding steady mechanical pressure against the wound for long enough that a clot forms and stays where it formed. Everything that matters about the technique follows from that, and there are only two ways to get it wrong: pressure in the wrong place, and pressure that keeps being interrupted.

Where the pad goes

Fold or roll the gauze into a firm pad about the thickness of a finger — thick enough that biting compresses it rather than closing your teeth straight through it. The pad has to sit directly on the gum over the socket, not lie flat across the chewing surfaces of the teeth in front of it. A flat pad wedged between the molars props the jaw open and delivers almost no pressure to the wound, which is why a patient can soak through a dozen pieces of gauze to no effect.

Dampen the gauze slightly with cool water before placing it. Dry gauze bonds to the forming clot and pulls it away on removal, which restarts the bleeding at the exact moment you thought it had finished.

How to hold it

1. Sit upright, or prop yourself up on pillows. Do not lie down while applying pressure.
2. Place the pad over the socket and close your teeth firmly and continuously. Firm, not gentle, and not a rhythmic chew.
3. Set a timer for thirty minutes. An estimated thirty minutes is almost always short of thirty minutes.
4. Do not talk more than you have to, and do not open your mouth to answer the door.
5. Do not lift the pad to check. Not at five minutes, and not at twenty.
6. When the timer goes, wet the outside of the pad with a little cool water, then take it out slowly and sideways rather than pulling it straight down.
7. Look, using the sixty-second test above, before deciding whether a second pad is needed.

How many rounds is reasonable

Most sockets settle within one properly held round, and where they do not, a second round is the reasonable next step. If frank red bleeding is still running after two full rounds of unbroken, correctly placed pressure, the sequence available at home has been worked through, and the next step is contact rather than a third pad. That is the threshold used here, and it does not move according to the time of night.

The most common reason gauze appears not to work is checking. A clot forming in a socket is a fragile structure for its first several minutes; lifting the pad to look at it disrupts what has formed and returns the clock to zero. Thirty minutes of pressure interrupted four times is not thirty minutes of pressure. It is four short attempts, none of which lasted long enough to do anything.

Use gauze rather than whatever is to hand. Cotton wool, tissue paper and kitchen roll shed fibres into the socket and lift the clot when they come out. Gauze is supplied for the first day and is sold in any pharmacy; buying a second packet on the way home is worth the detour.

A plain black tea bag, damp and cool, is the usual home substitute once the supplied gauze has run out. Black tea contains tannins, which are astringent, and the conventional explanation is that they act on the surface of the wound and help a forming clot hold. That explanation is widely taught rather than firmly settled, and the more dependable half of the answer is duller: a tea bag is a well-shaped, absorbent pad that sits neatly over a molar socket and delivers pressure exactly where pressure is wanted.

How to use one

1. Take a plain black tea bag. Ordinary breakfast tea is what is wanted — herbal and fruit infusions are not tea and contain little or no tannin, and green tea contains less than black.
2. Wet it with cool water and squeeze it out so it is damp rather than dripping. Never hot, and never brewed and then used warm.
3. Place it directly over the socket, exactly as you would place gauze, and bite firmly.
4. Hold it for twenty to thirty minutes by a timer, without checking.
5. Wrap it in a single layer of damp gauze first if you would rather it did not come apart in your mouth.

It stains. The gum around the socket and sometimes the neighbouring teeth will look brown for a day or two, and dark debris around a tea-treated socket looks alarming to anyone who does not know what caused it. The staining is superficial and comes off. No sugar, no lemon, no milk, and do not use loose leaves, which are difficult to clear out of a socket afterwards.

Where a tea bag earns its place is the socket that keeps oozing into the evening after the supplied gauze has run out, or the one that starts again at eleven at night. What it does not do is change the threshold. A tea bag held in place for a full round with fresh blood still running underneath it is saying the same thing a second gauze pad would say: this needs to be looked at rather than managed at the kitchen sink.

A clot in a fresh socket has roughly the mechanical strength of a set jelly, and it is holding onto a wound that has a blood supply on both sides of it. Almost everything on the list below works through one of three mechanisms: suction, raised pressure in the vessels supplying the site, or direct mechanical disturbance. Knowing which mechanism applies makes the restrictions easier to keep, because most of them stop mattering as soon as the socket has closed over.

Two of those deserve a note. The hot drink is the one patients walk into, because tea is what people reach for when they feel unwell, and a socket that had settled will often start oozing within a few minutes of the first hot mug. Lukewarm is genuinely different from hot here, and a numb lip judges temperature poorly in any case.

The other is the night. Bleeding that restarts in the small hours, and the pillow that goes with it, is usually a posture problem rather than a wound problem. Lying flat raises the pressure in the veins of the head, and a clot that was adequate while sitting up is not necessarily adequate while horizontal. Two pillows for two nights removes most of it.

Exercise is worth being specific about, because "take it easy" tends to be heard as "do slightly less than usual". What matters is anything that raises the pulse and the blood pressure: running, weights, contact sport, moving furniture, a long hot shower. Walking at an ordinary pace is fine from the day after surgery.

The instinct of nearly every patient on a blood thinner is to skip a dose or two around surgery. It is a reasonable-sounding conclusion and it is the wrong one, for a reason worth understanding rather than simply accepting. Bleeding from a socket is a local problem with local solutions: it is visible, it is reachable, and it responds to pressure, to material packed into the socket, and to a suture. A clot forming on a coronary stent, on a mechanical heart valve, or in an artery supplying the brain is none of those things.

What changes if you take one of these drugs is not the method but the timing of the call. Work through one full thirty-minute round of properly placed pressure, and if the socket is still bleeding frankly at the end of it, make contact then rather than at the end of a second round. Say which medication you take: a vitamin K antagonist, one of the direct oral anticoagulants, a single antiplatelet drug, or two antiplatelet drugs taken together are four different conversations, and what is advised differs between them. If the office cannot be reached, an emergency department is the right destination rather than a longer wait at home.

The reason to make contact early is that a socket can be seen, cleared and closed in a surgical office, and none of that is available at a kitchen sink. Persisting past the threshold does not add anything, and it costs time. What is actually done for bleeding of this kind is set out in the last section of this page. None of it involves altering what you take.

Two smaller points belong at the consultation rather than on the day of surgery. Fish oil, high-dose vitamin E, ginkgo and several other supplements affect platelet function, and they are worth writing on the medication list, because patients rarely think of them as medication at all. And where an anticoagulant is running, surgery is generally planned for earlier in the day and earlier in the week, so that a problem on the evening of surgery is a problem that arises while the practice is open.

Bleeding on the day of surgery is expected. Bleeding that restarts on day two, three or four is a different event, and it has a shorter list of causes.

The usual reason is mechanical. Something disturbed the clot: the first hot meal, a rinse that turned into a swill, a workout taken a day early, a crust of food pressing into the socket, or a suture that came away sooner than intended. Dissolving sutures are meant to last one to two weeks, and one that goes at day three can leave a gum edge that oozes for a few hours before settling.

The less common reason is that the socket is inflamed or infected, and inflamed tissue bleeds readily on the slightest contact. What identifies it is not the bleeding but what accompanies it: swelling that is growing rather than settling after day three, a temperature at or above 100.4°F (38°C), a bad taste that a salt water rinse does not shift, or pain that had been improving and then sharply worsened. Any of those is a same-day contact whether or not the bleeding stops on its own.

The handling of the bleeding itself is unchanged. Sit upright, thirty minutes of firm pressure on a properly placed pad, a damp black tea bag as a second measure, nothing hot, and no rinsing for the rest of that day. If it settles inside two rounds, mention it at the post-operative visit rather than letting it go unrecorded. If it does not settle, make contact.

One thing not to do is keep inspecting. A socket that is examined in the mirror every twenty minutes, with the cheek pulled back and the tongue pushed into it, will keep bleeding for as long as the inspections continue. At that point the examination is the cause.

There is a defined threshold, and it is worth reading before it is needed rather than deciding at two in the morning whether the situation is bad enough to justify a call.

• Fresh red bleeding still running after two separate thirty-minute rounds of firm, unbroken, correctly placed pressure.
• A mouth that fills with blood faster than you can swallow it.
• A large, soft, dark clot that keeps re-forming and lifting away, with bleeding continuing underneath it.
• Bleeding accompanied by faintness, dizziness, clamminess, pallor or a racing pulse.
• Vomiting blood, or material resembling coffee grounds.
• Bleeding in a patient taking an anticoagulant or antiplatelet drug that has not settled after the first thirty-minute round.
• Bleeding that restarts briskly several days after surgery and does not settle with two rounds of pressure.
• Bleeding with difficulty breathing or swallowing, a change in the sound of the voice, or swelling lifting the floor of the mouth or closing the eye — that one is 911 rather than a call to the office.

One further finding belongs here, because a patient dealing with a bleeding socket is usually not thinking about anything else. Numbness, tingling, burning or any altered feeling in the lower lip, chin or tongue that is still present once the local anaesthetic should have worn off is reported to the office on the day you notice it. It is not saved for the post-operative review, and it is not given a few days to see whether it settles on its own. The assessment, the record of what was found and when, and any decision about onward referral are all time-dependent, and reporting late closes options that were open early.

"Same day" means the same day and not the next morning. The practice keeps evening and Saturday hours, which covers part of the window in which this problem usually announces itself. If the office is closed, or if you telephone and do not reach anyone, do not spend the night working through further pads: an urgent care centre or an emergency department is the right destination, and post-extraction bleeding is a problem either can address. Waiting until morning tends to make this particular problem larger rather than leave it the same size, because a socket that has been bleeding for eight hours has usually accumulated a large disorganised clot that holds the wound edges apart and has to be cleared before anything else will work.

What is done for post-extraction bleeding in a surgical office is undramatic, and knowing it makes the call easier to place. Local anaesthetic is used where it is needed. The loose clot is cleared so the source can be seen, because a clot sitting on the wound conceals the bleeding point and holds the wound edges apart. The source is then identified as coming either from the soft tissue at the gum margin or from within the bone of the socket, because the two are managed differently: soft tissue is compressed or closed, while a socket bleeding from bone is packed with a material that supports clotting locally and stays in place while healing proceeds. A suture placed over the packing is common. Which of these applies, and whether anything further is used, is decided at the chair against what is found. Expect to bite on gauze for a period before leaving. The aim of the visit is a socket that holds its own clot without a pad in the mouth.

Things to leave alone: styptic pencils, alum, hydrogen peroxide, ice held inside the mouth against the socket, cotton wool, and any attempt to pack a socket with something not intended for the purpose. Ice against the outside of the cheek helps swelling and does nothing for bleeding. Nothing on that list improves clotting, and several items on it make the next examination harder than it needed to be.

And do not drive yourself. If you have felt faint, or if you have lost enough blood to be considering an emergency department, someone else drives. That is worth arranging on the day of surgery rather than at the point it becomes necessary — having another adult in the house for the first evening is the single practical measure that makes the rest of this page easy to follow.