You hear the noise before you see it. Then comes the silence, that very long second in which they do not cry, and then the scream. And while you pick them up you are already asking yourself the only question that matters: do we go or not?
Here you will find the answer in order: the signs that do not allow waiting, what the doctor will look at and why, and what to do in the twenty-four hours afterwards. At the end there is the part that concerns us, and it is small: the neck that stays painful. This article does not replace a medical assessment: when in doubt, go, and nobody will look at you sideways.
When to go, and how quickly
Two lists, because they are not the same thing.
The first collects the heaviest signs, largely those that the clinical rules classify as high risk, plus seizures. Here you call 112: do not wait, do not "see if it passes", do not search the internet.
- does not wake up or struggles to wake up, is confused, does not recognise you, responds slowly, keeps asking the same question;
- has had a seizure;
- has weakness, tingling, numbness, pupils of different sizes, double vision, difficulty walking or talking;
- is bleeding or leaking blood or clear fluid from the nose or ear, or has a bruise behind the ear or around the eyes that appeared after the injury;
- you feel a soft part, a step or a dip on the skull.
The second list is emergency department today: without rushing, but without putting it off until tomorrow. These are largely the signs that the rules place at intermediate risk, plus two things the guidelines treat separately: age under 3 months and a painful neck. Taken one at a time the risk stays low, but it is the doctor who has to put them together.
- has vomited, especially more than once;
- lost consciousness even for just a few seconds, and now is awake and normal;
- has a severe headache;
- has neck pain or cannot turn their head;
- is under 3 months old;
- is under 2 years old and the bump is not on the forehead;
- the fall was significant: more than 90 cm under the age of 2, more than 1.5 metres from the age of 2 up, or a road accident with a rollover or ejection, or hit while on foot or on a bike without a helmet, or struck on the head by an object thrown with force;
- simply seems different from usual to you. In the criterion doctors use under the age of two there is literally an item reading "behaves normally according to the parents": your perception is inside the algorithm, not outside it.
Outside these two lists, in the great majority of cases, the child is observed at home.
How rarely it is serious (and why it needs saying)
The Italian guidelines of the Italian Society of Paediatric Emergency and Urgent Medicine point out that in the United States between 600,000 and 700,000 children a year arrive at the emergency department with head trauma, and that over 95% are mild. Among mild injuries, Glasgow Coma Scale 14 to 15, that is a child who is awake and responding, a lesion visible on CT appears in about 5% of the children who are given a CT scan.
The largest study, PECARN, enrolled 42,412 children under 18 with a Glasgow of 14 to 15 in 25 North American emergency departments:
- 35.3% received a CT scan;
- 0.9% (376 children) had a clinically important brain injury, defined as death, need for neurosurgery, intubation for more than 24 hours or admission for at least 2 nights;
- 0.1% (60 children) ended up in the operating theatre.
Two opposite readings to hold together: the probability that tonight ends badly is very low, and precisely because it is so low you cannot rely on a feeling. You need a list of criteria, which is what the doctor will apply.
The questions the doctor will ask (and what they are looking for)
Two decision rules came out of that study, one for children under 2 years and one for those from 2 years up: the Italian guidelines have adopted them for every child with a Glasgow of 14 or more. Some signs are high risk and lead straight to a CT scan; others are intermediate risk and open a fork between CT and observation; if neither is present, the child is at very low risk and a CT scan can be avoided.
| Under 2 years | From 2 years up | |
|---|---|---|
| High risk (CT) | altered state of consciousness (Glasgow of 14 or less, agitation, drowsiness, repeated questions); palpable skull fracture | altered state of consciousness; signs of a base of skull fracture |
| Intermediate risk (CT or observation) | haematoma of the scalp in the occipital, parietal or temporal area; loss of consciousness of 5 seconds or more; abnormal behaviour according to the parents; serious mechanism of injury | vomiting; loss of consciousness of any duration; severe headache; serious mechanism of injury |
These rules are built to miss nothing, not to be precise. In an independent validation on 20,137 children in ten hospitals in Australia and New Zealand, sensitivity was 100% under 2 years and 99.0% above, with a negative predictive value above 99%: if the child comes through these criteria clean, the probability that something important is present is remote.
A note so that you do not feel under suspicion: the doctor will ask you exactly how it happened, and may ask you a second time. It is the protocol for all children, not a suspicion about you.
The three signs that keep parents awake most
The bump
The most frequent and the most misunderstood. In an analysis of children under 24 months, 2,998 had a bump as their only sign: a clinically important injury appeared in 12 of them (0.4%) and none needed neurosurgery. Among those who were given a CT scan, however, 8.8% showed an abnormality.
What shifts the risk is not how ugly it looks, but four things: the younger age, a non-frontal location (to the side or behind), the size and a significant mechanism of injury. A bump on the forehead at two and a half is not the same thing as a large bump at the back of the head of a four-month-old.
Vomiting
Among 42,112 children with mild head trauma, 5,557 (13.2%) had vomited. Among those who vomited with no other sign, a clinically important injury appeared in 2 cases out of 815 (0.2%), against 2.5% of those who vomited together with something else. And one detail that dismantles a widespread anxiety: neither the time at which the vomiting started nor the time since the last episode was significantly associated with risk. There is no magic hour to count.
"They lost consciousness for a moment"
An analysis from the same group found loss of consciousness in 15.4% of cases: with any loss of consciousness the risk of a significant injury was 2.5% against 0.5% without, but when it was isolated, the only sign present, it fell to 0.5% (13 children out of 2,780). The same holds for mechanism: among children whose only factor was a significant fall or an accident, a clinically important injury appeared in 0.3% under 2 years and 0.6% above.
The thread that connects these three paragraphs is always the same: one sign on its own weighs little, two or three together weigh a lot.
Why waiting a few hours is a choice, not a delay
If the doctor says "let's keep an eye on the child for a while", they are not parking you: they are applying the strategy with the most evidence behind it.
In the PECARN study, 5,433 children (14%) were observed before a decision on CT. They received fewer CT scans (31.1% against 35.0%, adjusted odds ratio 0.53) with a rate of significant injuries that was the same (0.75% against 0.87%). The Italian guidelines are explicit: for children at intermediate risk, initial observation should be preferred, for a minimum of 4 to 6 hours from the injury over 3 months; under 3 months it can go up to 24 hours.
The reason for avoiding it when you can is concrete. A study across seven American health systems estimated that the risk of radiation-induced leukaemia is higher precisely after head CT scans under the age of 5, at 1.9 cases per 10,000 scans. And the European EPI-CT cohort, covering 948,174 people who had a CT scan before the age of 22 in nine countries, estimated that for every 10,000 children examined today, 1 to 2 will develop a blood cancer attributable to radiation in the following 12 years.
Small numbers, which multiplied by hundreds of thousands of children a year become the reason the algorithms exist: not to deny you the CT scan, but so that it is not done for those who do not need it.
The 24 hours after: what to really watch
If you go home, the task is one: a responsible adult keeps an eye on the child during the first 24 hours. That is how it is written in the Italian guidelines, together with the recommendation that you be given the signs in writing for which to come back. Return to the emergency department if even one of these appears:
- a headache that gets worse instead of easing;
- vomiting that continues or comes back later;
- difficulty waking, unusual drowsiness, confusion;
- seizures, strange movements, weakness in an arm or a leg;
- inconsolable crying or irritability that does not settle, in a small child;
- loss of balance, double vision, difficulty speaking;
- neck pain that appears or increases.
In a small child the most reliable signal remains the simplest: are they eating, comforting themselves, playing, sleeping and waking up as usual?
What is not needed (and you will be told so anyway)
- Skull X-rays. The Italian guidelines advise against them, with a strong recommendation, as a screening test for significant brain injuries: a normal X-ray rules nothing out of what matters.
- Routine transfontanelle ultrasound and infrared devices to look for haematomas: advised against as routine by the same guidelines, which also ask to avoid repeated routine CT scans when the first one showed a clinically non-significant lesion: it is repeated only if the neurological picture changes.
- Waking them every hour. The 2025 position statement of the Canadian Paediatric Society says that waking children at night to assess their neurological status is not necessary as a routine. And in the systematic review that fed the international consensus on concussion, disturbed sleep was associated with slower recovery: an observed association, not proof that waking them does harm.
- Stopping them sleeping for half an hour after the bump. None of the guidelines cited here ask for this. If they are sleepy at nap time and wake up normally, they are sleeping, not getting worse.
- Ice is for the bump, not for what lies underneath: it gives relief and reduces swelling, and that is where it ends. Wrap it in a cloth.
If it is a concussion: the days after
In the Canadian 5P study, of 3,063 children and teenagers aged 5 to 17 who arrived at the emergency department within 48 hours of a concussion, symptoms persisted at 28 days in 31.0% of those who came to the final check. One child in three: it is not rare, and it is useful to expect it.
In the same study, what weighed more was female sex, age 13 and up, an already diagnosed migraine, a previous concussion with symptoms lasting more than a week, headache, sensitivity to noise and tiredness, but with modest discriminative ability (area under the curve 0.71 and 0.68), so much so that the authors themselves write that something else is needed before it can be used.
On what to do, by contrast, the direction has reversed compared with ten years ago. Among 2,413 children and adolescents followed for 28 days, those who had resumed some physical activity within 7 days had fewer persistent symptoms: 24.6% against 43.5% in the crude analysis, and the difference held after adjustment (11.4 percentage points with propensity score; relative risk 0.74). It needs saying in full: in the subgroup who were still symptomatic at 7 days, the difference disappeared as soon as the groups were matched by propensity score (47.2% against 51.5%, not significant).
The 2023 systematic review that fed the international Amsterdam consensus, across 46 studies, found in a meta-analysis of 7 studies an average earlier recovery of 4.64 days (95% CI: 2.59 to 6.69) with physical activity or prescribed exercise, and concludes that complete rest until symptoms disappear is not effective. The current model: a couple of days of light activity and little screen time, then a gradual return to activity below the symptom threshold. Not darkness and silence for a week.
The neck that stays: the part that concerns us
First the neck has to be checked by a doctor. In the PECARN study on the cervical spine, of 22,430 children enrolled after an injury, and already selected because they arrived by ambulance, were assessed by the trauma team or were already referred for an X-ray of the neck, 433 (1.9%) had a confirmed cervical injury. The factors that indicate a non-negligible risk are neck pain, pain on pressure over the midline, an altered state of consciousness and a significant head or trunk injury; in the validation cohort the rule had a sensitivity of 94.3% and a negative predictive value of 99.9%. So: a painful neck after an identical bump needs medical assessment, not compresses and not exercises.
After that, once injuries have been ruled out, if neck pain, headache and dizziness that do not go away remain, there is room for physiotherapy, and it has to be described with its limits. The reference trial is small: 31 participants aged 12 to 30 with persistent symptoms after a sports concussion. In the group that received, in addition to the standard pathway, combined vestibular and cervical rehabilitation, 73% (11 of 15) had been cleared to return to sport within 8 weeks, against 7% (1 of 14) of the controls. The 2023 systematic review included 13 studies, only one of high quality, five at high risk of bias, so much so that a meta-analysis was not possible, and it recommends cervicovestibular rehabilitation for adolescents and adults with dizziness, neck pain or headache lasting more than 10 days.
The limits are large: small numbers, an effect of a combined intervention (not of neck work alone), evidence in adolescents and adults. For young children there is no evidence of this kind, and anyone who promises you a protocol for a four-year-old citing these studies is citing them wrongly.
If the crooked neck was there before the fall, that is a different topic, and you will find it in Congenital torticollis in newborns: why timing matters most; if the pain is recurring and unconnected to injuries, in Cervical pain in children: causes, symptoms and what to do; if the mechanism was a whiplash in a car, in Whiplash: symptoms, what to do, recovery.
What we can do (and what we cannot)
Without beating about the bush: urgency is not our job. If you have just heard the noise and are reading with the child in your arms, close this page and call the paediatrician or go to the emergency department.
We come in afterwards, for one thing only: when the emergency department has ruled out the serious stuff, days or weeks have passed, and what remains is a neck that pulls, headache on exertion, a sense of instability. There we look at how far the head turns from one side to the other, what sets off the symptom, and we build a gradual return instead of the sofa.
With Ri-Hub Junior you start with a free twenty-minute introductory consultation by video call with a physiotherapist on the Italian professional register who works with children and teenagers: you tell us what happened and what has remained, we look together at how they move, and we tell you whether it makes sense to work on it or whether the right place is somewhere else. Up to age 10 sessions last 50 minutes and cost 59,90 €; from age 11 to 18 it is the standard session at 29,90 €.
And if there is a complex neurological condition behind it, the right place is not us: those pathways require in-person assessments and specialist centres, and we say so straight away rather than waste your time.
Another domestic accident that frightens and resolves in a minute is the elbow that locks: Nursemaid's elbow.
On the garden trampoline the numbers are surprising, and the risk depends on how many jump on it together: The garden trampoline.
The push scooter and the electric one have completely different injury profiles: Scooters and children.
Sources
- Italian guidelines on the assessment and management of pediatric head injury in the emergency department, Italian Journal of Pediatrics, 2018
- Identification of children at very low risk of clinically-important brain injuries after head trauma: a prospective cohort study, The Lancet, 2009
- Accuracy of PECARN, CATCH, and CHALICE head injury decision rules in children: a prospective cohort study, The Lancet, 2017
- Management of the paediatric patient with acute head trauma, Paediatrics & Child Health (Canadian Paediatric Society), 2025
- The effect of observation on cranial computed tomography utilization for children after blunt head trauma, Pediatrics, 2011
- Risk of traumatic brain injuries in children younger than 24 months with isolated scalp hematomas, Annals of Emergency Medicine, 2014
- Association of traumatic brain injuries with vomiting in children with blunt head trauma, Annals of Emergency Medicine, 2014
- Isolated loss of consciousness in children with minor blunt head trauma, JAMA Pediatrics, 2014
- Prevalence of clinically important traumatic brain injuries in children with minor blunt head trauma and isolated severe injury mechanisms, Archives of Pediatrics & Adolescent Medicine, 2012
- The use of computed tomography in pediatrics and the associated radiation exposure and estimated cancer risk, JAMA Pediatrics, 2013
- Risk of hematological malignancies from CT radiation exposure in children, adolescents and young adults, Nature Medicine, 2023
- PECARN prediction rule for cervical spine imaging of children presenting to the emergency department with blunt trauma, The Lancet Child & Adolescent Health, 2024
- Clinical Risk Score for Persistent Postconcussion Symptoms Among Children With Acute Concussion in the ED, JAMA, 2016
- Association Between Early Participation in Physical Activity Following Acute Concussion and Persistent Postconcussive Symptoms in Children and Adolescents, JAMA, 2016
- Rest and exercise early after sport-related concussion: a systematic review and meta-analysis, British Journal of Sports Medicine, 2023
- Targeted interventions and their effect on recovery in children, adolescents and adults who have sustained a sport-related concussion: a systematic review, British Journal of Sports Medicine, 2023
- Cervicovestibular rehabilitation in sport-related concussion: a randomised controlled trial, British Journal of Sports Medicine, 2014
- Consensus statement on concussion in sport: the 6th International Conference on Concussion in Sport, Amsterdam, October 2022, British Journal of Sports Medicine, 2023
Last source check: September 2026.
