Prognostic value of blood urea nitrogen-to-albumin and C-reactive protein-to-albumin ratios in acute stroke compared with established clinical and radiological predictors: a single-center retrospective cohort study
This single-center retrospective study of 736 acute stroke patients found that while the C-reactive protein-to-albumin ratio (CAR) showed a modest independent association with in-hospital mortality, both CAR and the blood urea nitrogen-to-albumin ratio (BAR) demonstrated limited discriminatory performance compared to established clinical and radiological predictors like lesion volume and stroke severity scores.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
When a stroke strikes, it is a sudden, violent interruption of the brain's blood supply, leaving tissue to starve or bleed out. The immediate aftermath is a race against time, where doctors must quickly gauge how severe the injury is and predict whether a patient will survive the hospital stay. For decades, the medical community has relied on specific tools to make these predictions: a neurological exam that measures how well a patient can move and speak, and brain scans that reveal exactly how much damage has been done. These established methods are powerful, but they require specialized training and equipment. In recent years, researchers have wondered if the body's own chemical signals, found in a routine blood test, could offer a simpler, faster glimpse into a patient's fate. Specifically, they have looked at two numbers derived from standard blood work: the ratio of a waste product called urea to a vital protein called albumin, and the ratio of an inflammation marker to that same protein. These ratios act as a window into the body's stress levels, nutritional state, and inflammatory response, all of which surge after a brain injury.
A team of researchers in Turkey set out to test whether these blood-based ratios could stand alongside the established neurological exams and brain scans as reliable predictors of death in the hospital. They looked back at the records of 736 adults who had been admitted to their emergency department with an acute stroke between 2020 and 2025. The group included patients with both types of stroke: the more common kind caused by a blocked artery, and the less common but often more dangerous kind caused by a burst blood vessel. For every patient, the team calculated the two ratios using blood samples taken the moment they arrived at the hospital, before any major treatments were given. They then compared how well these blood numbers predicted who would die during their hospital stay against the predictions made by the standard tools: the neurological exam score, the brain scan measurements, and other known severity scores.
The results painted a clear picture of what these blood tests can and cannot do. The researchers found that the established tools were far superior at distinguishing between patients who would survive and those who would not. The size of the damaged area in the brain, measured directly from the scans, was the strongest predictor of all, followed closely by the neurological exam score. In contrast, the blood ratios offered only a modest ability to predict the outcome on their own. While the inflammation-based ratio showed a slight, independent link to mortality in patients with the blocked-artery type of stroke, it did not provide the same level of certainty as the brain scans or the physical exam. The waste-product ratio, meanwhile, did not remain a significant predictor once the researchers accounted for the severity of the brain injury and other factors.
The study also explored whether these blood tests worked differently depending on the type of stroke or if adding them to the standard models improved the doctors' ability to predict survival. The analysis showed that the blood ratios did not behave differently between the two stroke types, and adding them to the complex models did not significantly improve the accuracy of the predictions. In fact, the models that relied on the standard neurological and radiological measures were already so accurate that the blood tests added very little new information. The researchers concluded that while these blood ratios are not useless, they should not be viewed as replacements for the established methods. Instead, they might serve as a complementary piece of information, offering a quick, additional hint about the patient's overall stress and inflammation levels, but they cannot yet take the place of the detailed neurological assessment and brain imaging that currently guide care. Before these ratios can be used routinely in hospitals to predict outcomes, the findings need to be confirmed in larger studies across different medical centers.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.