Cerebral Haemorrhage Overview
Cerebral Haemorrhage Overview
A cerebral haemorrhage is bleeding into the brain tissue itself — an artery that has burst, not one that has blocked. From the outside the two look identical: a dropped face, a weak arm, speech that has gone wrong. That is why the first examination in any suspected stroke is an immediate CT, and why the result of that scan changes the treatment completely rather than slightly. What follows is governed by two questions. The first is how much blood has been released, where it sits, and whether it is still growing — blood continues to escape in the first hours in a substantial minority of patients, and that window is what neurocritical care and neurosurgery are working against. The second question is what the patient takes that stops blood from clotting, because that is something the first hour must actively reverse. This guide is about those two questions and the decisions that follow from them.
-
National
Stroke Center construction unit, and a national GCP base
-
3D DSA
Alongside 3.0T MRI, dual-source CT, PET-CT and 24-hour video EEG
-
48 hours
Written second-opinion report after complete records arrive
-
124+
Countries and regions our international patients come from
What a cerebral haemorrhage is, and why the CT comes first
A cerebral haemorrhage is blood escaping from an artery into the brain tissue itself. The distinction that matters is not academic but mechanical: in an ischaemic stroke an artery is blocked and the tissue beyond it starves; here an artery has burst and the tissue is damaged both by the pressure of the escaping blood and by the loss of supply to everything the vessel used to feed. Haemorrhage accounts for roughly one stroke in ten to one in seven — a higher proportion in East Asian populations than in Europe or North America — yet it takes a disproportionate share of stroke deaths and of the disability that follows.
Two consequences follow immediately. The first is that the treatments are mirror images. Clot-dissolving drugs and clot-retrieval procedures exist to open a blocked artery, and both are dangerous when the artery has bled. Aspirin and anticoagulants, which protect against clots, make a bleed worse. So the very first step in any suspected stroke is not treatment but a non-contrast CT, which shows fresh blood as a bright area and settles in minutes which of the two conditions is being dealt with. Everything else waits on that image.
The second consequence is that the emergency here runs on time in the opposite direction from the one most people have heard about. There is no drug that dissolves the blood and no procedure that reliably restores the tissue it has damaged. What can be done is to stop the bleeding from continuing, to protect the brain around it, and to relieve the pressure it creates. Much of that work happens in the first few hours, because a haematoma that is still expanding is a different problem from one that has stopped — and in imaging series a substantial minority, around a third, do continue to expand after the first scan.
Where the blood sits — and what that usually means
The single most useful thing a CT report can add after the word "haemorrhage" is where it is. Location is not a detail for the diagram; it points at the mechanism, and the mechanism decides what happens next and what has to be looked for.
Deep haemorrhage — in the putamen, the thalamus, the pons or the cerebellum — is the classic pattern of long-standing high blood pressure. Decades of raised pressure damage the small penetrating arteries that supply these structures, and when one of them finally gives way the bleeding begins in the middle of the brain rather than at its surface. This is the most common pattern worldwide, and it is the reason blood pressure is a risk factor that can actually be acted on.
Lobar haemorrhage sits near the cortex, in the outer rim of a hemisphere, and in an older patient this pattern points in a different direction: cerebral amyloid angiopathy, in which a protein fragment deposits in the walls of the small vessels on the brain's surface and makes them brittle. It becomes steadily more common with age, it tends to bleed again, and it carries implications that blood-pressure treatment alone does not address — including how to think about an anticoagulant the patient may still need for their heart. On a CT the two look like the same white blob. MRI is what separates them.
A third group is described by what it is not: bleeding caused by something structural — an arteriovenous malformation, an aneurysm, a cavernous malformation, or a vein that has clotted — or by a disorder of clotting, a low platelet count, liver disease, or a drug such as cocaine. These are more likely in a younger patient and in someone with no history of high blood pressure, and finding them changes the plan entirely: the question stops being only "how do we manage this bleed" and becomes "what is the malformation, and when do we treat it".
Volume, location and the direction of travel
A haemorrhage is not graded once. It is described along several axes at the same time, and the axis that matters most in the first hours is not a static number but a direction: is this bleed still getting bigger?
| Axis | How it is described |
|---|---|
| Volume of the haematoma | Calculated from the first CT, in millilitres. Supratentorial bleeds are conventionally grouped as small (under about 10 mL), moderate (10–30 mL) and large (over 30 mL), and published series report steadily worse outcomes across those groups. Volume is the number people quote — but it never stands alone, because the same volume means different things in different places. |
| Location | Deep (putamen, thalamus, pons), lobar (the outer rim of a hemisphere), cerebellar, or brainstem. A cerebellar bleed of 3 cm or more that compresses the brainstem or blocks the flow of cerebrospinal fluid is an emergency that may need an operation, while a much larger bleed in a silent part of a frontal lobe may not. Location is what turns a volume into a decision. |
| Intraventricular blood and hydrocephalus | Whether blood has entered the ventricles, and whether it has obstructed the circulation of cerebrospinal fluid. This is often what decides that a drain is needed, and it is a strong predictor of outcome in its own right — sometimes more so than the size of the bleed in the tissue. |
| Level of consciousness | Recorded as the Glasgow Coma Scale and repeated. In reported series it is the strongest single predictor of outcome, and it is also the measurement that changes fastest — which is why it is reassessed rather than recorded once, and why a fall in it is treated as an emergency in itself. |
| Direction of travel in the first hours | Whether the haematoma is expanding or has stopped. A substantial minority enlarge during the first hours after onset, and that expansion is associated with worse outcomes. A spot of contrast leaking inside the haematoma on CT angiography is one of the signs used to judge whether bleeding is ongoing — imperfectly, but it is the closest thing to a live answer that imaging offers. |
| What the patient is taking | An anticoagulant (and which one), an antiplatelet, or neither. This is not a footnote to severity: an anticoagulant changes both the likelihood that the bleed continues and what must be done in the first hour, and it changes the conversation weeks later about whether it can safely be restarted. |
| Function, and time since onset | The functional axis: what the patient can no longer do, how much of that has already begun to return, and how long it has been. This is the axis patients experience, and the one that governs the rehabilitation plan — most of the eventual functional outcome is built over weeks and months after the acute event, not in the first day. |
Read together, these axes answer a different question from the one patients usually arrive with. The question is not "how big was the bleed", but which of three situations this is: a bleed large enough, in a place dangerous enough, to need an operation now; a bleed that must be protected while the brain around it is kept stable; or a bleed that has already stopped, in a patient whose recovery from here is a rehabilitation problem rather than a surgical one.
How a suspected bleed is investigated
The order of these investigations is deliberate, and the first of them is not a formality — it is the decision. Until a CT has been done, no treatment for stroke can safely be given, because the two kinds of stroke require opposite actions.
| Investigation | What it settles |
|---|---|
| Non-contrast CT of the head | Whether this is a haemorrhage at all: fresh blood appears as a bright, well-defined area where an ischaemic stroke looks normal or subtly darker in the first hours. The same scan gives the volume and location, shows whether blood has reached the ventricles, and shows hydrocephalus. It takes minutes, and it is available in every hospital. |
| CT angiography | Whether there is an underlying malformation or aneurysm to treat, and whether contrast is leaking within the haematoma — a sign that the bleeding may still be active. In a younger patient with no history of high blood pressure, this is where a treatable structural cause is usually found. |
| MRI with susceptibility-weighted sequences | Whether the cause looks like amyloid angiopathy rather than blood pressure. Small bleeds scattered in lobar distribution and haemosiderin staining along the cortical surface are the findings that point to it, and they are invisible on CT. MRI also identifies a cavernous malformation, and diffusion-weighted imaging helps where there is doubt about an ischaemic component. |
| Digital subtraction angiography | Reserved for when a vascular malformation is suspected and its treatment is being planned. It gives the detail that decides whether something can be embolised, removed or treated with radiation, and it is performed by the same team that would carry that treatment out. |
| Blood tests and clotting studies | Clotting screen, platelet count, renal function and glucose — together with the specific assay for whichever anticoagulant the patient is taking. This last point is the one most often missed: the way to reverse dabigatran and the way to reverse a factor Xa inhibitor are different, and the way to reverse warfarin is different again. |
A complete report therefore answers five things rather than one: is it blood, how much, where, has it reached the ventricles, and is it still growing.
Why patients travel to China for neurovascular care
A cerebral haemorrhage is an emergency, and no one should travel during one. The treatment that matters most is the one given in the nearest hospital able to provide it, and that is true whether the hospital is in Frankfurt, Lagos or Guangzhou. Travel begins to make sense afterwards, in three situations: when the cause has not been established and the mechanism matters, when a structural cause has been found and its treatment has to be planned, and when the work ahead is rehabilitation.
Our neurology and neurosurgery services are organised as a standardised stroke centre built to national stroke-centre standards, and are a National Stroke Center construction unit and national GCP base. Cerebrovascular cases are planned jointly by neurology, neurosurgery and rehabilitation rather than handed from one department to the next, and interventional work — the treatment of aneurysms, arteriovenous malformations, Moyamoya disease and arterial stenosis — is carried out on 3D DSA, alongside 3.0T MRI, dual-source CT, PET-CT and 24-hour video EEG.
The practical questions are answered in the same way as for any other condition we treat. Records reviewed by a stroke or neurovascular specialist receive a written opinion within about 48 hours of complete records arriving, and the opinion states what the images show, what remains uncertain, and which investigations would settle it. Complete treatment in China typically runs at 30–50% of comparable care in Europe or the United States. We work with more than 30 commercial insurers for direct billing and have received patients from 124 countries and regions. Where the question is whether an old bleed was caused by blood pressure or by amyloid angiopathy — and whether an anticoagulant can be restarted — that is often answerable from images and records alone, and a video consultation is a reasonable first step rather than a journey.