Craniocerebral Trauma Diagnosis
Craniocerebral Trauma Diagnosis
The assessment of a head injury is ordered by the clock rather than by convenience: airway, breathing and circulation first, then the response, then the scan, then the decision. Nothing in it is complicated; what makes it work is doing the parts in that order, and repeating the parts that move. This page sets out what the first hour contains, how the Glasgow Coma Scale is taken and read, what the scan establishes and what it cannot, the two delayed presentations that are most often missed, and the judgement of when to repeat rather than when to order once.
The first hour: what happens in what order
The order is the treatment. Every part of the assessment matters, but the parts that decide the most come first, and the commonest failure in this disease is a scan taken before the blood pressure is supported. Low oxygen and low blood pressure are the two secondary injuries most strongly associated with a worse outcome, and both are correctable in minutes by anyone who looks for them.
| Order | What happens | Why it is there | Common error |
|---|---|---|---|
| 1. Airway, breathing, circulation | Oxygen given, the airway protected if the person cannot protect it, blood pressure supported, and the neck protected until the spine is cleared | Low oxygen and low blood pressure cause more avoidable damage than anything else in this disease, and they are treatable in minutes | Going to the scanner before the blood pressure is supported, because the scan feels more urgent than the numbers |
| 2. The mechanism | What force, what height, what speed, what surface, whether the head stopped suddenly or rotated, and who saw it | It sets which patterns are likely, and whether the neck needs imaging too | Accepting "he just fell" without asking from what height, onto what, and whether anyone saw it |
| 3. The response | Eye opening, speech and movement, recorded as three components; pupils recorded with size and reactivity | This is the monitor. Every later decision is a comparison against it | Recording a single number once, and recording it after sedation rather than before |
| 4. Emergency non-contrast CT | Reported for blood, fracture, shift and pressure | It decides whether someone goes to theatre within the hour | Treating a normal scan as the end of the assessment rather than as one finding within it |
| 5. The decision | Theatre, admission to a unit that can escalate, or discharge with a written plan and someone who knows what to watch for | Each of the three is a real decision, and the third one is the one most often done badly | Discharging without a written plan, or admitting somewhere that cannot escalate if the person deteriorates |
| 6. Bloods and the bleeding question | Clotting, platelets, blood sugar, and — in an older adult — exactly which medicines affect bleeding | Anticoagulation turns a small collection into a large one, and it is reversible if it is known about in time | Asking "is he on blood thinners" and accepting a yes without asking which one and when the last dose was |
Two things are worth adding. The neck is part of the assessment, because a head injury mechanism is frequently a neck injury mechanism, and cervical spine imaging is decided by the mechanism and the examination rather than by the headache. And the clock is recorded — the time of the impact is the reference point for every later comparison, and it is the one piece of information that cannot be reconstructed afterwards. What the first signs look like is set out on the symptoms page.
The Glasgow Coma Scale, and how it is read
The scale records three things: whether the eyes open, whether speech makes sense, and whether movement obeys a command. It is written as three component scores rather than as one total, because the total hides the detail that the decision depends on — a person who opens their eyes but does not speak is in a very different position from one who speaks but does not obey.
| Component | What is recorded | What can move it falsely | Why it is repeated |
|---|---|---|---|
| Eye opening | Spontaneous, to speech, to pain, or none | Facial swelling can close the eyes; sedation abolishes it | A change from spontaneous to only-to-pain is a deterioration whoever the patient is |
| Verbal response | Orientated, confused, inappropriate words, sounds, or none | A breathing tube removes it entirely; language and hearing differences move it | It is the component most sensitive to a slowly enlarging collection in the first hours |
| Motor response | Obeys commands, localises pain, withdraws, abnormal posturing, or none | Sedation, paralysis, and a low blood pressure all move it | It is the component that carries the most prognostic weight, and it is read once sedation and perfusion are accounted for |
| Pupils | Size and reactivity, recorded separately from the score | Previous eye surgery, eye drops, and some medicines | A pupil that stops reacting is a surgical sign rather than a finding to be noted and reviewed tomorrow |
| Mild / moderate / severe | Conventionally grouped by the total, with severe at the bottom of the range | The grouping is a shorthand for handover, not a verdict | The group can change in either direction in the first day; the trajectory is the real finding |
The one sentence to hold on to: the score is a monitor, not a verdict. It is read once sedation, a breathing tube, low blood pressure, low oxygen and alcohol have been accounted for, and its value lies almost entirely in the comparison between one recording and the next.
What the scan establishes, and what it cannot
Computed tomography is the right first test and it answers the questions that decide the theatre: is there blood, where is it, how big is it, is there a fracture, and is the brain being pushed. That is a lot, and it is not the whole of the disease. Knowing what it is not looking at is what prevents the normal scan from becoming a false reassurance.
| Test | What it is for | What it does well | What it does not do |
|---|---|---|---|
| Emergency non-contrast CT | The first decision: theatre, admission, or discharge with a plan | Blood, fracture, shift and pressure, within minutes, in a person who cannot hold still | It does not show a concussion, it understates diffuse axonal injury, and it shows a collection at the moment it was taken rather than the moment it will matter |
| Repeat CT | Answering the question raised by a change in the examination | Whether a collection has enlarged, which is the commonest cause of a deterioration that was not present on arrival | A scan repeated on a clock rather than on a change finds less and exposes more |
| CT of the cervical spine | Clearing the neck where the mechanism or the examination warrants it | Fracture and alignment, at the same sitting as the head | It is not done after every head injury; it is decided by mechanism, neck pain and examination |
| MRI at 3.0T | Where symptoms persist with a normal CT, or where the extent of injury needs characterising | Diffuse axonal injury, small contusions, and the patterns computed tomography cannot resolve | It is not a screening test after every injury, and it rarely changes acute management |
| CT angiography, where indicated | Where the mechanism or the pattern raises a question about a blood vessel | A traumatic aneurysm, a dissection, or a vascular injury along a penetrating track | Ordered for a specific question rather than as part of a routine panel |
| EEG, where indicated | Where a seizure, or an unexplained fluctuation in consciousness, is in question | Whether an event was a seizure, and whether there are seizures that are not visible from the bedside | A normal recording excludes neither; subclinical seizures after severe injury are worth looking for specifically |
The principle in one line: the examination is the monitor, and the scan is what you do when the monitor changes. Why the mechanism produces this pattern is set out on the causes page.
The two delayed presentations that are missed
Both of these arrive by getting better first, and both are the reason this page exists rather than being a footnote.
| Extradural haematoma | Chronic subdural collection | |
|---|---|---|
| Who | Any age; classically a younger person after a blow to the side of the head with a fracture | An older adult, frequently taking an anticoagulant or antiplatelet medicine, and frequently after a fall nobody mentioned |
| What the source is | Usually an artery, torn at a fracture — which is why it enlarges fast and under pressure | Usually a bridging vein, torn by the movement of the brain — which is why it enlarges slowly and quietly |
| The timing | Minutes to hours. The lucid interval is brief, and the deterioration is rapid | Weeks. The person is well, then gradually not, and the connection to the fall is never made |
| How it presents | A period of apparent clarity, then drowsiness, a worsening headache, vomiting, and a pupil that stops reacting | Confusion, unsteadiness, a change in personality, a gradual memory failure, or simply not being oneself — rarely a headache |
| What it is mistaken for | A good sign: he is talking and wants to go home | Dementia, a stroke, or age. It is the most consequential misattribution in the whole disease |
| What catches it | A period of observation in a setting that can escalate, with a repeat scan triggered by the examination rather than by the clock | Anyone asking about a fall. Any older adult who has changed in the weeks after a fall should be scanned, whatever the fall looked like |
| What happens if found | Surgical evacuation, urgently. This is the pattern where an hour changes everything | Often drained through a small opening, with a large and rapid improvement. The most treatable pattern in this disease, and the most often missed |
There is a third delayed presentation worth knowing, less dramatic and more common: a contusion that enlarges over the first day or two and begins to behave like a collection. It is the reason a repeat scan is triggered by the examination rather than by a schedule, and it is more likely in anyone whose clotting is impaired.
What else is looked for, and what is often not
A head injury is rarely the only thing that happened, and the parts of the assessment that get skipped are usually the ones that were nobody's job in particular.
- The medicines that affect bleeding. Which anticoagulant, which antiplatelet, what dose, and when the last one was taken. This is the single question that most changes the risk of the next twenty-four hours, and it is often answered with a yes and no more.
- Why the person fell. A fall is a symptom as often as it is an accident. A heart rhythm, a blood pressure that drops on standing, a medicine that was changed, an infection, a hypoglycaemia, or a seizure all deserve a look, and none of them is found by scanning the head.
- The neck. Cervical spine injury travels with head injury often enough that the neck is part of the assessment rather than a separate question raised later if there is pain.
- Other injuries. A person who has fallen from height or been in a collision has usually injured something else, and the distracting pain of a limb is the commonest reason a deterioration in consciousness is noticed late.
- The skull base. Fluid from the nose or ear, bruising behind the ear or around the eye, and a fracture through the base of the skull carry a risk of infection that changes management.
- The person's baseline. What they were like before — in memory, in mobility, in mood — is the reference every later comparison is made against, and it has to come from someone who knew them. It is the part of the assessment no hospital can reconstruct retrospectively.
- A written plan for the next forty-eight hours. Who is watching, what they are watching for, and what they should do about it. Discharge without one is the most common failure in the whole pathway, and it is the one that costs the least to fix.
Watching for a change rather than waiting for a time
Observation in this disease is an intervention, not a pause. Its purpose is to catch the deterioration that secondary injury announces before it has finished happening, and that purpose is served by watchfulness rather than by the passage of hours.
What triggers a repeat scan. A fall in the level of consciousness rather than a rise. A new pupil abnormality. New weakness or altered speech. A seizure. A headache that is worsening rather than settling, with vomiting that recurs. Each of these is an examination finding, and each of them is the trigger rather than the clock.
What legitimately waits. Magnetic resonance imaging for persistent symptoms. Neuropsychological testing, which is more informative once the acute phase has passed than during it. Decisions about long-term rehabilitation intensity. And any judgement about the final outcome, which in this disease is made over months rather than in the first week.
What is recorded while watching. The three component scores, the pupils, and — for a person at home — the specific things that matter: whether they can be woken, whether they know where they are, whether the vomiting has recurred, and whether the headache is better or worse than it was six hours ago. A written note of these, taken by whoever is watching, is worth more to the next clinician than any recollection.
The closing principle: a head injury declares itself over time rather than at a moment. The first assessment is rarely the last word, and a revision is the disease becoming clearer rather than the first clinician having failed. What happens once the decision is made follows on the treatment page.