New Nomogram Predicts Heart Risks After Stroke

Researchers at a tertiary hospital in Urumqi have developed a statistical tool designed to predict the timing of major adverse cardiovascular events in patients recovering from acute ischemic stroke. The team created a nomogram, a graphical calculation device, to estimate the risk of these events at specific time points: three days, seven days, and throughout the entire hospital stay. This approach aims to provide clinicians with a more precise method for risk stratification using variables that are routinely available in clinical practice.

The research involved a total of 1,244 patients, divided into two distinct cohorts for development and validation. The derivation cohort consisted of 871 patients admitted between 2020 and 2023, while the validation cohort included 373 patients admitted from 2023 to 2024. In the derivation group, 124 patients, representing 14.2% of the total, developed in-hospital major adverse cardiovascular events. A notable finding was the concentration of these events in the early phase of recovery, with 52.4% occurring within the first three days of admission.

To identify the most relevant predictors, the researchers employed Least Absolute Shrinkage and Selection Operator Cox regression, followed by multivariate Cox regression. This statistical process narrowed down a large set of potential variables to seven independent predictors that significantly influenced the risk of adverse events. These predictors included patient age, the admission National Institutes of Health Stroke Scale score, the presence of carotid artery stenosis, D-dimer levels, the platelet-to-lymphocyte ratio, creatinine levels, and lactate dehydrogenase levels.

The performance of the nomogram was evaluated using the area under the receiver operating characteristic curve, a standard metric for assessing the accuracy of predictive models. In the derivation cohort, the model achieved AUC values of 0.839 for the three-day prediction, 0.764 for the seven-day prediction, and 0.698 for the overall hospitalization period. When tested on the independent validation cohort, the AUCs were 0.862, 0.760, and 0.754, respectively. To account for potential overfitting, the researchers calculated optimism-corrected AUCs, which were 0.823, 0.750, and 0.689.

Calibration, which measures how well the predicted probabilities match the observed outcomes, was also assessed. In the derivation cohort, the calibration was acceptable, indicating that the model’s predictions were generally aligned with actual events. However, in the temporal validation cohort, the model overestimated the risk at the three-day mark. The calibration-in-the-large value was 0.85, and the calibration slope was 1.35, suggesting that the predicted risks were higher than the observed risks for this specific time point.

Using the total risk scores generated by the nomogram, the researchers stratified patients into three distinct risk groups: low, intermediate, and high. The low-risk group had scores ranging from 0 to 131, the intermediate-risk group from 131 to 184, and the high-risk group from 184 to 292. Kaplan-Meier survival curves demonstrated that these groups had significantly different incidences of major adverse cardiovascular events, with a log-rank P-value of less than 0.05.

The study concludes that the developed nomogram offers a time-point-specific approach to predicting in-hospital major adverse cardiovascular events in acute ischemic stroke patients. By utilizing seven routinely available predictors, the tool provides a practical framework for identifying patients at elevated risk early in their hospital stay. However, the authors emphasize that the model requires recalibration and external validation in different populations before it can be widely adopted in clinical practice. The overestimation of risk in the validation cohort underscores the need for further refinement to ensure accurate predictions across diverse patient groups.

Subject of Research: Development and validation of a nomogram for predicting in-hospital major adverse cardiovascular events after acute ischemic stroke

Article Title: Predicting stroke-heart syndrome: a nomogram for time-point-specific prediction of in-hospital major adverse cardiovascular events after acute ischemic stroke

Article References: Qin, M., Xu, X., Dong, Y., Xie, Y., Sun, Y., Qiao, M., & Wang, H. (2026). Predicting stroke-heart syndrome: a nomogram for time-point-specific prediction of in-hospital major adverse cardiovascular events after acute ischemic stroke. BMC Cardiovascular Disorders. https://doi.org/10.1186/s12872-026-06980-6

Image Credits: AI Generated

DOI: 10.1186/s12872-026-06980-6

Keywords: acute ischemic stroke, stroke-heart syndrome, nomogram, cardiovascular risk, clinical prediction, risk stratification, Urumqi, BMC Cardiovascular Disorders, Predicting, stroke-heart, syndrome, time-point-specific