Clinical-Imaging Nomogram for Estimating Hematoma Expansion After Spontaneous Intracerebral Hemorrhage

When a blood vessel inside the brain bursts, the resulting bleed is a medical emergency known as a spontaneous intracerebral hemorrhage. Unlike a stroke caused by a blocked artery, this event involves active bleeding that fills the space within the brain tissue. The danger lies not just in the initial injury, but in what happens next. In many cases, the clot of blood continues to grow, expanding into the surrounding brain and causing further damage. This growth often occurs within the first few hours after a patient arrives at the hospital, making it difficult for doctors to predict who will suffer this worsening condition and who will remain stable. Because the outcome for a patient depends heavily on whether the bleed stops or spreads, the ability to identify those at risk early on is a critical challenge for medical teams.

A team of researchers at Beijing Jingmei Group General Hospital set out to solve this problem by looking for a pattern in the data available the moment a patient walks through the door. They reviewed the records of 279 adults who had suffered a spontaneous brain bleed between 2019 and 2025. Every patient in this group had undergone a standard head scan, known as a computed tomography, or CT, both when they were first admitted and again a short time later. The researchers defined a significant expansion as a growth in the blood clot of more than 6 milliliters or an increase of at least one-third of its original size. By comparing the scans and the patients' initial health records, they aimed to find a reliable way to forecast which clots would grow.

The study began by gathering a wide range of information for each patient. This included basic details like age and sex, medical history such as diabetes or previous strokes, and habits like smoking or drinking alcohol. The team also recorded vital signs taken upon arrival, including blood pressure and a score measuring the patient's level of consciousness. Laboratory results were pulled from blood tests, checking levels of platelets, glucose, and calcium. Finally, the researchers carefully examined the CT scans, measuring the size of the initial blood clot, noting its location, and looking for specific visual clues. Two of these visual clues were particularly important: whether the blood clot looked uneven or mixed in texture, and whether there were tiny, separate spots of bleeding floating near the main clot, a feature the researchers called the satellite sign.

After sorting through the data, the researchers found that 89 of the 279 patients, or about 32 percent, experienced this significant expansion of their blood clot. When they compared the patients who had expanded clots against those who did not, several clear differences emerged. The group with growing clots tended to have higher blood pressure upon arrival, lower levels of consciousness, and larger initial blood clots. Their scans were more likely to show the uneven texture and the satellite sign. Interestingly, the group with expanding clots also had lower levels of ionized calcium in their blood. While factors like smoking history and the use of blood-thinning medications showed some connection in the initial review, they did not hold up as strong predictors when all the factors were analyzed together.

The researchers then used a statistical method to combine the most important factors into a single, practical tool. They selected six specific variables that worked best together to predict the outcome: the patient's blood pressure, their level of consciousness, the initial size of the clot, the uneven texture of the clot on the scan, the presence of the satellite sign, and the level of ionized calcium. These six elements were woven into a visual chart called a nomogram. This tool allows a doctor to take the values for these six factors, plot them on the chart, and receive a calculated probability of whether the patient's hematoma will expand. The model proved quite effective at distinguishing between patients who would and would not experience growth, correctly identifying the risk in a large majority of cases within the study group.

Despite the promising results, the authors are careful to frame this work as a step forward rather than a final solution. The tool was built and tested using data from a single hospital, and the researchers acknowledge that it needs to be checked against data from other hospitals to ensure it works universally. They also note that the study was retrospective, meaning it looked back at past records rather than following patients forward in time, which can sometimes make a model look better than it truly is. The study does not claim that high blood pressure causes the bleeding to grow, only that the two are linked in a way that helps prediction. Similarly, the role of calcium is viewed as a useful signal rather than a proven cause.

The value of this research lies in its simplicity and its use of information that is already available in almost every emergency room. The model does not require expensive new tests or complex imaging; it relies on the standard blood pressure cuff, the routine blood draw, and the initial CT scan that every patient receives. By combining these ordinary pieces of data into a structured prediction, the nomogram offers a way to spot high-risk patients early. This could help doctors decide who needs closer monitoring or more frequent scans. However, the researchers emphasize that this tool is currently a candidate for clinical use, not a rule that should be followed without further proof. Before it can guide life-or-death decisions, it must be validated in a broader, independent group of patients to confirm that its predictions hold true everywhere.