Work

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

Year: 2026

Type: article

Abstract: Stroke-heart syndrome (SHS) significantly affects prognosis in patients with acute ischemic stroke (AIS). Risk stratification tools for in-hospital major adverse cardiovascular events (MACEs) remain limited. This study aimed to develop and validate a nomogram for time-point-specific prediction of in-hospital MACEs in AIS patients using routinely available clinical variables. A total of 871 AIS patients admitted to a tertiary hospital in Urumqi (2020–2023) formed the derivation cohort, and 373 patients (2023–2024) the validation cohort. LASSO-Cox and multivariate Cox regression were used to select predictors and construct a nomogram for predicting MACEs at 3 days, 7 days, and during hospitalization. Discrimination (area under the receiver operating characteristic curve [AUC]), calibration, and decision curve analysis were assessed. X-tile-derived total risk scores defined three risk strata, validated by Kaplan-Meier curves. In the derivation cohort, 124 of 871 patients (14.2%) developed in-hospital MACEs, with 52.4% occurring within the first three days. Seven independent predictors were identified: age, admission National Institutes of Health Stroke Scale (NIHSS) score, carotid artery stenosis, D-dimer, platelet-to-lymphocyte ratio (PLR), creatinine, and lactate dehydrogenase (LDH). The nomogram achieved AUCs of 0.839, 0.764, and 0.698 at 3 days, 7 days, and during hospitalization in the derivation cohort and 0.862, 0.760, and 0.754 in the validation cohort; optimism-corrected AUCs were 0.823, 0.750, and 0.689. Calibration was acceptable in the derivation cohort, but 3-day risk was overestimated in temporal validation (calibration-in-the-large 0.85, calibration slope 1.35); decision curve analysis indicated net benefit across a range of thresholds. Total risk scores stratified patients into low- (0-131), intermediate- (131–184), and high-risk (184–292) groups with significantly different MACE incidence (log-rank P < 0.05). This study developed and temporally validated a nomogram for time-point-specific prediction of in-hospital MACEs in AIS patients at 3 days, 7 days, and during hospitalization, based on seven routinely available predictors. The derived risk stratification system may aid tiered monitoring, but recalibration and external validation are required before clinical use. Not applicable.

Source: BMC Cardiovascular Disorders

DOI: 10.1186/s12872-026-06980-6