A blow to the head can cause bleeding in or around the brain in several different ways. Different types of bleeding can raise different concerns and require specific management strategies. This page walks through the different factors that may come into play and how Neurotrauma and Neurocritical Care teams may respond.
This resource should not be used to make decisions regarding acute care. The information on this page is intended to inform patients and family members about the basic types traumatic intracranial hemorrhage and the management decisions that clinicians may need to make when interacting directly with patients who present with traumatic brain injury.
Bleeding inside the skull after an injury is not one condition. It is several different conditions that happen to share a cause.
A traumatic intracranial hemorrhage is bleeding inside the skull caused by an injury: a fall, a motor vehicle collision, a sports impact, an assault, or any other blow or sudden movement of the head. It is different from a concussion, where imaging typically looks normal even though something has clearly happened to how the brain is working. When a scan does show blood, the single most useful question a clinician asks is not just "how much blood," but "where is the blood." Location determines the likely cause, what the bleed will look like on a CT scan, how quickly it is likely to change, and how urgently it needs treatment.
To understand why location matters so much, it helps to picture the layers between the outside of the head and the brain itself: the skull, then two layers of tough membrane called the dura mater, then a thinner membrane called the arachnoid, a fluid-filled space, and finally the surface of the brain itself, which has an even thinner membrane (the pia) hugging it directly. These layers create "compartments" that define how blood may expand, spread, and cause injury to the brain itself. The next section walks through each of these compartments and the type of bleed that occurs in it.
Select a type below to see where it occurs, why it happens, and what it typically means for treatment and outlook. The same slider also controls the pressure graph on the right, since not every type of bleed pushes intracranial pressure the same way.
The skull is a closed, rigid container. Blood takes up space at the expense of other skull contents, including the brain itself. This is known as the Monro-Kellie doctrine. At some point, the ability to compensate for the added volume of blood (by decreasing other fluids, including cerebrospinal fluid, for example) diminishes, and pressure starts to rise more quickly. Drag the slider above to see what happens to intracranial pressure as the volume of blood, and any associated swelling, increase — and notice how differently each type of bleed moves along this same curve.
Not every symptom after a bump on the head means a dangerous bleed, and not every dangerous bleed causes symptoms right away. Select a symptom below to see how it is generally regarded.
A combination of the neurological exam, brain imaging, and sometimes invasive monitors, allows us to follow a patient’s status.
The Glasgow Coma Scale (GCS) is a tool that was designed to assess the initial severity of brain injury by allowing medics, doctors and nurses to quickly rate a person’s level of awareness and function in the field or at the bedside. In the emergency department, assessment of GCS helps assign patients to the proper TBI protocol, such as whether intracranial pressure monitoring is indicated. The GCS may be used to track changes over time, but a more specific neurological exam is usually preferred once a patient is settled in the hospital.
A CT scan is almost always the first imaging study after a significant head injury: it is fast, widely available, and shows blood clearly. We look at the type of bleed, its size, whether it is pressing on or shifting the brain, and whether the fluid spaces (ventricles) that normally cushion the brain are being compressed. Repeat CT scans at a set interval are often routine, not necessarily a sign that something has gone wrong. Subdural and epidural hematomas can continue to expand over time, and CT scans are often used to track this, and to assess whether surgery may be required. Contusions, meanwhile, can blossom (enlarge) over the first hours to days, and a repeat scan catches that early, before it shows up as a change in how a patient is behaving.
MRI is not typically the first study in the emergency setting; it takes longer and is harder to perform safely in an unstable patient. Once someone is stable, however, MRI — especially sequences sensitive to blood breakdown products — is considerably more sensitive than CT for diffuse axonal injury, which can look deceptively normal on an initial CT scan despite causing real symptoms. In addition, MRI does a better job showing the brain itself and swelling due to nearby blood. Overall, MRI may be a better indicator of the extent of injury, and may help us determine a patient’s prognosis.
For more severe injuries, imaging and the GCS are not enough on their own — we sometimes place a dedicated pressure monitor directly inside the skull so intracranial pressure (ICP) can be measured continuously in real time, rather than inferred indirectly. This connects directly to the idea explored earlier: once pressure can be measured moment to moment, it can be treated before it causes permanent damage rather than after. There are two main ways this is done.
A thin pressure-sensing probe is passed through a small hole (“bolt”) in the skull and a few centimeters into the brain tissue itself. It is quick to place, does not require finding a specific fluid space, and gives a continuous, reliable pressure number at the bedside. Its limitation is that it only measures pressure — it cannot be used to drain fluid the way an EVD can.
A thin catheter is guided into one of the brain’s fluid-filled ventricles and connected to an external monitor. Beyond measuring pressure, an EVD can also actively drain cerebrospinal fluid (or blood, in the case of IVH) to directly relieve pressure — making it both a monitoring tool and a treatment, which is why it is often preferred when the ventricles are enlarged or when IVH is present.
An ICP monitor reports the average pressure in the brain, and also provides a continuous pressure waveform, one pulse per heartbeat. The shape of that waveform carries information about the brain’s compensatory reserve (its “compliance” or ability to compensate for increased mass effect from bleeding or swelling), often before the average pressure itself looks concerning on its own. Learn about the ICP waveform below.
Some bleeds are so large that they press on the brain to compromise blood-flow and function. When that hematoma is large enough, or the patient is declining, removing it is the most direct way to relieve pressure.
Because these are often arterial and can expand quickly, all but the smallest are typically evacuated through a craniotomy promptly, and even small ones are watched closely in the ICU rather than sent home.
Often evacuated when it is causing significant "mass effect" or neurological decline; a small one in a patient who is neurologically stable may instead be watched closely with repeat imaging. A slower-forming, older version of this same condition is covered separately on our Chronic Subdural Hematoma page.
Contusions that have "blossomed" enough to cause severe mass effect or swelling may require a decompressive craniectomy to relieve pressure on the brain.
Epidural hematomas classically feature a lucid interval: because the blood has not yet touched the brain directly, a patient can seem entirely normal for a stretch of time before rapidly declining as the bleed expands and the pressure begins to affect neurological function. This is why even a small, asymptomatic epidural hematoma is watched closely rather than dismissed.
When there is no single mass to remove, or after one has been removed, treatment shifts to directly controlling pressure inside the skull. Select each tier to see what it involves and why it comes where it does.
Water moves toward whichever side has proportionally less free water because random molecular motion, over time, evens out the water fraction on both sides of a membrane that only water can cross. Giving a hypertonic IV solution (like mannitol or 3% saline) raises the solute concentration in the blood vessel; water then moves from the (relatively more dilute) brain tissue into the vessel until the two sides re-equilibrate. That water, no longer in the swollen brain, ultimately leaves the body in urine, visibly shrinking brain tissue as it leaves, counteracting swelling.
Outcomes vary enormously depending on the type of bleed, its severity, and a patient’s age and overall health. Even knowing these things, precisely predicting long-term outcomes for any particular individual with traumatic brain injury is still fraught with difficulty and uncertainty.
Seizures can occur after a significant brain injury, and the risk is higher with more severe injury. Cognitive and behavioral changes are common after moderate to severe injury; bilateral frontal contusions in particular are associated with longer-term behavioral change. Diffuse axonal injury is the hardest of these injury types to predict early on, because visible injury on imaging can look mild even when symptoms are significant.
For patients who need a decompressive craniectomy, possible complications include CSF leak, infection, a fluid collection under the scalp (pseudomeningocele), hydrocephalus requiring a shunt, and, less commonly, worsening of brain shift immediately after the skull is opened (sometimes called paradoxical herniation). We discuss these tradeoffs directly rather than only after the fact.
Select each stage below to see what typically happens and why.
Care for a traumatic intracranial hemorrhage is delivered by our Neurotrauma division, working directly with our Neurocritical Care colleagues.
Rhode Island Hospital is a high-volume Level I Trauma Center serving Rhode Island and the surrounding region, and our Neurotrauma division manages the full range of these injuries, from small bleeds that only need close observation to the most severe injuries requiring emergency surgery. Our 18-bed Neurocritical Care Unit (NCCU) is staffed by neurosurgeons, neurointensivists, and critical care nurses who specialize specifically in diseases of the brain and spine, with advanced imaging immediately available. While the team below leads this care, all of our neurosurgery faculty contribute to the care of patients with traumatic brain injury.
