ICU and You · The Dental Files · Number 1
Teeth in the school-age child, for intensivists who were never taught any
We put a metal blade into a child's mouth, lever it against their front teeth, and then spend the next several days looking directly at those teeth every hour without ever once thinking about them. Dentistry is the one region of the body we routinely instrument and never examine.
The Dental Files is a small series about the bits of the mouth that turn up in intensive care whether we invite them or not. Not dentistry — there are dentists for that — but the overlap: the tooth you knocked out, the abscess that closed the airway, the prophylaxis you were supposed to give, and the ten-year-old whose incisor has one hour to live.
A nine-year-old is intubated for status asthmaticus. Afterwards the registrar notices a gap in the upper arch that they cannot account for, and reasons that a child of nine is losing teeth anyway.
That reasoning is correct about a third of the time and catastrophic the rest, and telling the difference takes about fifteen seconds if you know what you are looking at. So let us start with the part nobody teaches.
Twenty primary teeth. Thirty-two permanent teeth, or twenty-eight if you leave out the wisdom teeth. The primary set is complete by about two and a half.
Then, from about six to twelve years, the child is in mixed dentition — primary and permanent teeth in the mouth at the same time, some firm, some frankly mobile, some half-erupted. That window maps almost exactly onto the school-age band, which is why this piece belongs in this fortnight.
At about six, the first permanent molar erupts behind the last primary molar. It does not replace anything. Nothing falls out to announce it, no gap precedes it, and it arrives looking new and small at the back of a mouth full of baby teeth.
Everybody — parents, nurses, and most doctors — assumes it is a baby tooth. It is not. It is a permanent molar that has to last sixty or seventy years, and it is the most commonly neglected and most commonly lost tooth in the mouth.
Working rule: in a child of six to twelve, count back from the midline. The sixth tooth is permanent, always, however new it looks.
The first permanent teeth to arrive are the lower first molars and the lower central incisors, both at six to seven. Upper central incisors follow at seven to eight. Canines and the primary molars hang on much longer — canines are shed at nine to twelve, second primary molars at ten to twelve — so in an eight-year-old, a wobbly tooth at the front is almost certainly permanent and about to be a problem, while a wobbly tooth further back is almost certainly primary and about to be nothing.
Two digits. First digit the quadrant, counting clockwise as you look at the patient, starting upper right. 1, 2, 3, 4 for the permanent quadrants; 5, 6, 7, 8 for the primary ones. Second digit the position from the midline — one to eight for permanent teeth, one to five for primary. Said aloud digit by digit: thirty-six is "three-six".
| Code | Tooth |
|---|---|
| 11 / 21 | Permanent upper central incisors — the ones the laryngoscope meets |
| 16 26 36 46 | The four first permanent molars. The six-year molars |
| 51 / 61 | Primary upper central incisors |
| 85 | Primary lower right second molar |
A perfectly good ICU entry: “61 and 71 mobile pre-intubation; 11, 21 erupted and firm; 16/26/36/46 present and firm.” It takes fifteen seconds and it is the difference between a documented finding and an argument.
Dental damage is the most commonly claimed anaesthetic complication in Australia — something like eighteen per cent of anaesthetic claims — which is wildly out of proportion to its clinical severity and tells you something about how people feel about their teeth.
Damage requiring actual repair or extraction runs at roughly one in four and a half thousand general anaesthetics. Small prospective studies in which a dentist formally examined children after intubation report much higher rates of minor injury, around ten to sixteen per cent, but those are counting bruised teeth that no anaesthetist would ever record — do not put those two numbers side by side. What both agree on is where the risk sits: difficult laryngoscopy and repeated attempts.
This is the dental emergency most likely to arrive in your resus bay attached to a trauma call, and the one where the window closes while everybody is busy with something more obviously important.
What determines whether a replanted permanent tooth survives is the periodontal ligament cells on the root surface. They die when they dry out, and they are destroyed by scrubbing. Everything below follows from that one fact.
Dry time is the number that matters, not time since the accident. Under 15 minutes dry, the cells are probably alive. Under 60 minutes, compromised but worth replanting. Over 60 minutes, they are gone and the goal changes to keeping the tooth as a space maintainer. A tooth that sat in milk for three hours but was picked up in the first minute has a dry time of one minute.
Water is hypotonic and lyses the very cells you are trying to save. Milk works because its osmolality and pH are close enough to physiological. That said, the international guideline is explicit that water is still better than letting the tooth dry out — so the rule is milk or saline first, water only if there is nothing else, never in preference.
The root apex of a primary incisor sits directly against the developing permanent tooth bud. Pushing it back risks driving it into the follicle and damaging the permanent successor — enamel defects, a bent root, or a tooth that never erupts properly. It can also ankylose and block the successor, and it can be aspirated. Clean, give analgesia, arrange dental review. Do not replant.
Which leaves the question the registrar is actually standing there with, holding a tooth, at two in the morning.
Put it in milk and call someone. The cost of storing a primary tooth in milk is nothing. The cost of binning a permanent incisor is permanent.
This is the section to read even if you skip the rest, because the guidance moved recently and a lot of us are still carrying the old version.
First, the scale of it. In 2024 there were 506 diagnoses of acute rheumatic fever across the reporting jurisdictions, and 93 per cent were in First Nations people. The rate in First Nations Australians was about 52 per 100,000 against 2.7 per 100,000 in non-Indigenous Australians — a nineteen-fold difference. And it is a disease of exactly this age group: 45 per cent of all cases were in children aged five to fourteen. In the Northern Territory the First Nations rate is 386 per 100,000.
It is not a dental disease. Acute rheumatic fever follows Group A streptococcal infection. Dental caries is caused by entirely different organisms — Streptococcus mutans and friends — and there is no causal path from bad teeth to rheumatic fever. Poor oral health and ARF cluster in the same communities because they share the same social determinants, not because one causes the other. Do not teach otherwise.
It is not only throat infection. The Australian guideline is now explicit that skin infection triggers ARF as well as pharyngitis, and that streptococcal impetigo is very common in First Nations children. That is a real divergence from the international teaching most of us grew up with, and it changes what counts as a treatable precursor.
Where teeth genuinely enter the story is the other end: a child with established rheumatic heart disease has a damaged valve, and oral viridans streptococci are a classical cause of infective endocarditis. Which brings us to prophylaxis.
Australian guidance used to recommend endocarditis prophylaxis for rheumatic heart disease only in Aboriginal and Torres Strait Islander people. That restriction has been removed. Prophylaxis is now recommended for all patients with rheumatic heart disease, whoever they are.
The genuinely Australian part of the story survives, but it is a different point from the one most people remember: Australian guidance lists rheumatic heart disease as an indication for prophylaxis at all, which the American, British and European guidelines do not. That divergence is driven by the epidemiology above.
The cardiac conditions warranting prophylaxis before dental work are: a prosthetic valve or prosthetic material used in valve repair; previous infective endocarditis; a ventricular assist device; congenital heart disease only where there is an unrepaired cyanotic defect or residual defect adjacent to prosthetic material; and rheumatic heart disease.
It is needed for extractions, implants, biopsy or removal of soft tissue or bone, scaling above or below the gum line, and — note this, given the section above — replanting an avulsed tooth. It is not needed for local anaesthetic through healthy tissue, radiographs, or adjusting appliances.
| Situation | Child dose, orally, 60 minutes before |
|---|---|
| No penicillin allergy | Amoxicillin 50 mg/kg up to 2 g |
| Non-severe penicillin allergy | Cefalexin 50 mg/kg up to 2 g |
| Severe penicillin allergy | Azithromycin 10 mg/kg up to 500 mg, or doxycycline |
Clindamycin is no longer recommended for this indication in Australia, and neither is vancomycin. A good many of us would still have reached for clindamycin without thinking. Oral amoxicillin given an hour beforehand holds adequate levels for around four hours, which is worth knowing when a list runs late.
Fluoride works topically and after the tooth has erupted, not systemically beforehand, which is why the advice is to spit rather than rinse. It reduces demineralisation, promotes remineralisation with a more acid-resistant mineral, and inhibits bacterial metabolism in plaque.
Fluorosis — the mottling — comes from swallowing too much fluoride while the crown is mineralising inside the jaw, which is why the amounts are restricted in the very young. A useful thing to be able to tell a worried parent: a ten-year-old cannot develop fluorosis of their permanent front teeth now. That window has closed.
Probably toxic dose is around 5 mg/kg of elemental fluoride; around 15 mg/kg is potentially lethal in a small child.
Ordinary toothpaste is 1,000 ppm, which is 1 mg of fluoride per gram. For a 20 kg child that puts the probably-toxic dose at 100 g of toothpaste — an entire family tube. Ordinary toothpaste ingestion is almost always benign.
So the question to ask is which product. High-fluoride prescription toothpaste is 5,000 ppm — the same child reaches a toxic dose at 20 g, which is an entirely plausible accident. Fluoride tablets and professionally applied gels likewise.
Significant toxicity is GI first — fluoride forms hydrofluoric acid in the stomach and is directly corrosive — then hypocalcaemia, hypomagnesaemia and hyperkalaemia, then QT prolongation, ventricular dysrhythmia and arrest, which can come without warning up to six hours later. Oral calcium or milk binds fluoride in the gut. Monitor, and give IV calcium and magnesium early and repeatedly rather than cautiously — these arrhythmias are notoriously refractory to everything else. Ring the Poisons Information Centre on 13 11 26.
Ludwig's angina is a rapidly spreading cellulitis of the submandibular, sublingual and submental spaces — bilateral, brawny, and in its early phase without a drainable collection. The name comes from the Latin for to strangle, which is fair.
In adults it is overwhelmingly odontogenic, from lower second and third molars whose roots sit below the mylohyoid line. In children it often is not, partly because a school-age child has no third molars and unerupted second molars, so the classic adult route is not available. Paediatric cases follow oral trauma, sialadenitis, gingivostomatitis, immunosuppression — and frequently no identifiable source at all. Where it is dental in this age group, suspect a grossly carious first permanent molar, which brings us neatly back to the beginning.
For antibiotics after replanting an avulsed tooth, the international dental guideline advises against doxycycline under twelve years, while Australian paediatric hospital guidance uses an eight-year cut-off and prescribes it above that. So for a ten-year-old with a replanted incisor the two documents point opposite ways. Follow your local Australian guidance and Therapeutic Guidelines — but know the disagreement exists rather than being surprised by it at the bedside.
Have you ever knocked a tooth out, or watched one go?
Most of us have, or have been standing next to it. What happened to the tooth, what did the unit do next, and — the part I am most interested in — was the right thing done in the first ten minutes, or did everybody find out afterwards what should have happened?
I would also like to hear from anyone who works where rheumatic heart disease is common, on what dental access actually looks like for those children in practice, as opposed to what the guideline assumes. Named or anonymous, entirely as you prefer.
The button opens a reply straight to me. If your device blocks it, write to icuandyou@icloud.com instead.