A simple model based on routine clinical data can provide individualized cerebral small-vessel disease risk estimates and may help prioritize magnetic resonance imaging evaluation and intensify the management of modifiable risk factors.
Cerebral small-vessel disease was present in nearly half of the hospitalized patients who underwent brain MRI.
The final multivariable logistic regression model included six variables: hypertension, glycated hemoglobin, homocysteine, C-reactive protein, triglycerides, and total cholesterol.
The prediction model demonstrated good discriminative ability for distinguishing patients with MRI-defined CSVD from those without.
The model showed acceptable calibration after bootstrap internal validation.
Decision curve analysis demonstrated a net clinical benefit of the model across a wide range of threshold probabilities.
An online dynamic nomogram was developed to enable individualized, point-of-care cerebral small-vessel disease risk estimation.
This research suggests that a small set of routine clinical measurements — blood pressure history (hypertension), blood sugar control (glycated hemoglobin), homocysteine levels, inflammation marker (C-reactive protein), and cholesterol-related values (triglycerides and total cholesterol) — can meaningfully predict whether a hospitalized patient has cerebral small-vessel disease (CSVD) as detected by MRI. CSVD refers to damage to the small blood vessels in the brain and is associated with stroke, dementia, and other neurological problems. In this study of 164 hospitalized adults, nearly 45% were found to have CSVD on MRI, highlighting how common this condition is in clinical settings. The researchers built a statistical model using these six variables and found it performed well, correctly distinguishing patients with and without CSVD about 82% of the time. They also created an online interactive tool (a dynamic nomogram) that clinicians can use to quickly calculate an individual patient's estimated risk of having CSVD based on their specific test results. The model showed reliable calibration (meaning its predictions matched real-world outcomes well) and demonstrated clinical usefulness across a broad range of risk thresholds. This research suggests that doctors may be able to use simple, already-available blood tests and clinical information to identify which patients are most likely to have CSVD before — or instead of — ordering an MRI, potentially helping prioritize imaging resources and target treatments for modifiable risk factors like high blood pressure, elevated cholesterol, and inflammation. However, the study was conducted at a single center with a relatively small sample and only used internal validation, so the model's performance in other populations and settings would need to be confirmed in future studies.
Liu X, Wang X, Xie R, Li M, Liu Z, Pan Z. (2026). A web-based dynamic nomogram for individualized risk prediction of magnetic resonance imaging-defined cerebral small-vessel disease: Model development and internal validation.. The Journal of international medical research. https://doi.org/10.1177/03000605261466575