Cardiac biomarkers combined with neuroimaging localization predict long-term outcomes in acute stroke patients Authors: Haşim Tüner, Fuat Polat, Yalçın Velibey, Gülsüm Saruhan Durmaz, Mehmet Ali Özbek, Sait Terzi Published: 2026 ABSTRACT AbstractCardiac biomarker elevation is common after stroke, but its relationship with stroke localization and long-term outcomes remains incompletely characterized. This prospective observational cohort study enrolled 695 patients (380 ischemic stroke, 135 TIA, 180 hemorrhagic stroke) from October 2020 to July 2023. High-sensitivity cardiac troponin I (hs-cTnI), NT-proBNP, and CK-MB were measured at admission and at 12–24 h. Ischemic strokes were categorized by localization (insular involvement, anterior non-insular, posterior circulation, subcortical-only). Primary outcomes were 1-year all-cause mortality and major adverse cardiac and cerebrovascular events (MACCE). Biomarker elevations varied significantly by stroke type and localization. Hemorrhagic stroke showed the highest hs-cTnI (median 28.7 ng/L) and troponin elevation rate (54.4%). Among ischemic strokes, insular involvement demonstrated the highest biomarker levels (hs-cTnI 45.8 ng/L, NT-proBNP 892 pg/mL, troponin elevation 73.9%; all p < 0.001 versus other localizations). At 1 year, overall mortality was 20.4% and MACCE 31.4%. Elevated troponin independently predicted mortality (HR 3.24, 95% CI 2.18–4.82), MACCE (HR 2.86), and recurrent stroke (HR 1.92). A synergistic interaction between troponin elevation and insular involvement was identified for mortality (interaction HR 2.15, p = 0.029) and MACCE (interaction HR 1.84, p = 0.016). Hs-cTnI demonstrated good discriminative ability (AUC 0.782 for mortality, 0.745 for MACCE). Cardiac biomarker elevation, particularly in insular strokes, independently predicts adverse outcomes. Routine biomarker assessment combined with stroke localization data may enhance risk stratification and guide targeted interventions.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-57238-2.Keywords: Stroke, Cardiac biomarkers, Mortality, Major adverse cardiac and cerebrovascular events, Insular cortexSubject terms: Biomarkers, Cardiology, Diseases, Medical research, Neurology Introductıon Stroke remains a leading cause of mortality and long-term disability worldwide, with an estimated 12.2 million new cases and 6.6 million deaths annually1. Elevated cardiac biomarkers, particularly high-sensitivity cardiac troponin (hs-cTn) and NT-proBNP, are observed in 18–60% of stroke patients without concurrent acute coronary syndrome, depending on the biomarker and population studied2,3. The pathophysiological mechanisms underlying cardiac biomarker elevation are multifactorial, including neurogenic cardiac injury, pre-existing cardiac disease, hemodynamic stress, and systemic inflammation4,5. Neurogenic cardiac injury, mediated through excessive catecholamine release and autonomic dysregulation, can result in direct myocardial damage manifesting as troponin elevation and electrocardiographic abnormalities in the absence of obstructive coronary artery disease6. This phenomenon has been particularly well-documented following subarachnoid hemorrhage but is increasingly recognized across all stroke subtypes. Recent state-of-the-art reviews have further characterized the neuroanatomic and neurochemical basis of post-stroke cardiac injury, collectively termed “stroke-heart syndrome” emphasizing that cardiac complications significantly contribute to post-stroke morbidity and mortality7,8. The insular cortex is a critical region for autonomic regulation, and its damage, particularly to the right insula, may lead to profound autonomic imbalance with heightened sympathetic outflow9,10. However, comprehensive data examining cardiac biomarker profiles across different stroke localizations, including anterior, posterior, and subcortical territories, remain limited11. The prognostic significance of cardiac biomarker elevation after stroke has been established in multiple cohort studies12,13. A recent systematic review confirmed that troponin elevation is independently associated with increased all-cause mortality after ischemic stroke, with pooled hazard ratios of 1.5 to 4.014. NT-proBNP elevation also predicts cardiovascular complications and recurrent stroke15. The temporal dynamics of serial troponin measurements (delta troponin) may provide additional prognostic information16. Despite these associations, the independent predictive value of cardiac biomarkers for specific long-term outcomes, particularly MACCE, remains incompletely characterized when accounting for stroke subtype and anatomical localization. Current management lacks evidence-based guidelines for stroke patients with elevated cardiac biomarkers17. This study aimed to comprehensively characterize cardiac biomarker profiles across stroke types and ischemic stroke localizations and to evaluate their prognostic value for one-year clinical outcomes. We hypothesized that insular involvement would be associated with higher biomarker elevations compared to other localizations, and that integration of biomarker elevation with anatomical localization would enhance risk stratification beyond traditional clinical variables. Methods Study design and population This prospective observational cohort study enrolled patients from October 2020 to July 2023, with follow-up completed through September 2024. Adult patients (≥ 18 years) presenting within 24 h of symptom onset with acute ischemic stroke, TIA, or hemorrhagic stroke confirmed by brain CT or MRI were eligible. Patients were excluded if they had: acute myocardial infarction within 30 days, recent cardiac surgery or PCI within 3 months, severe CKD requiring dialysis (eGFR < 15 mL/min/1.73 m²), active malignancy receiving therapy, life expectancy < 3 months, or incomplete cardiac biomarker data at admission. All patients provided written informed consent. Stroke classification and localization Stroke type was classified as ischemic stroke, TIA, or hemorrhagic stroke based on clinical presentation and neuroimaging. Ischemic stroke subtype was determined using TOAST criteria. For ischemic stroke patients, localization was determined by systematic review of brain CT and/or MRI (DWI preferred, performed in 86.3%) by independent neuroradiologists blinded to clinical outcomes. Inter-rater agreement was high (92.6%; Cohen’s κ = 0.89, 95% CI 0.85–0.93); disagreements were resolved by a senior neuroradiologist. Four mutually exclusive categories were defined1: insular involvement, any infarction involving the insular cortex2; anterior non-insular, MCA or ACA territory without insular involvement3; posterior circulation, PCA or vertebrobasilar territory; and4 subcortical-only, isolated subcortical infarctions without cortical involvement. Cardiac biomarker assessment Venous blood samples were collected at hospital admission (within 6 h of presentation) and at 12–24 h (median 18 h, IQR 14–22). Hs-cTnI was measured using the ARCHITECT STAT High Sensitive Troponin-I assay (Abbott Diagnostics); 99th percentile upper reference limits were 26.2 ng/L (men) and 15.6 ng/L (women). NT-proBNP was measured using the Elecsys proBNP II assay (Roche Diagnostics), and CK-MB by immunoinhibition method (reference range < 5.0 ng/mL). Delta troponin was calculated as peak 24-hour minus admission troponin. Troponin elevation was defined as exceeding the sex-specific 99th percentile. Cardiac evaluation included 12-lead ECG at admission and daily during hospitalization (continuous telemetry ≥ 48 h), and transthoracic echocardiography within 72 h of admission per ASE guidelines. Outcomes Primary outcomes at 12 months were all-cause mortality and MACCE (composite of cardiovascular death, non-fatal MI, recurrent stroke, unplanned coronary revascularization, or heart failure hospitalization). Secondary outcomes included individual MACCE components, any rehospitalization, new-onset atrial fibrillation, and poor functional outcome (mRS ≥ 3). All outcomes were adjudicated by an independent committee of two cardiologists and one neurologist blinded to baseline biomarker values and stroke localization. Statistical analysis Sample size was calculated to provide 80% power to detect a hazard ratio of 1.8 for troponin elevation, assuming 20% mortality, 40% troponin elevation prevalence, and 10% loss to follow-up (α = 0.05, two-sided). Continuous variables are presented as mean ± SD (normally distributed) or median (IQR) (non-normally distributed); categorical variables as n (%). Group comparisons used ANOVA with Tukey HSD, Kruskal-Wallis with Dunn’s post-hoc tests, or chi-squared/Fisher’s exact tests as appropriate. Multivariable Cox proportional hazards regression identified independent predictors of outcomes after verification of the proportional hazards assumption (Schoenfeld residuals). Hazard ratios with 95% confidence intervals are reported. Interaction terms between troponin elevation and insular involvement were examined in ischemic stroke patients. ROC curves were generated and AUCs compared by DeLong’s method; optimal cutoffs were determined by Youden’s index. Decision curve analysis assessed net clinical benefit. Multiple comparisons were adjusted using the Benjamini-Hochberg procedure. All tests were two-sided; p < 0.05 was significant. Analyses used Stata 17.0 and R 4.2.0. Results Study population and baseline characteristics A total of 695 patients were enrolled: 380 (54.7%) with ischemic stroke, 135 (19.4%) with TIA, and 180 (25.9%) with hemorrhagic stroke (Fig. 1). Mean age was 67.1 years; ischemic stroke patients were oldest (68.5 ± 12.3 years; p = 0.012). Sex distribution was similar across groups (57% male; p = 0.953). Among ischemic stroke patients, localizations were: anterior non-insular (n = 188, 49.5%), insular involvement (n = 92, 24.2%), posterior circulation (n = 85, 22.4%), and subcortical-only (n = 15, 3.9%). Baseline clinical characteristics are presented in Table 1; localization-specific data are presented in Supplementary Table S1. Study population distribution by stroke type and ischemic stroke localization. Table 1Baseline clinical characteristics and cardiac biomarkers by stroke type.VariableIschemic Stroke (n = 380)TIA (n = 135)Hemorrhagic Stroke (n = 180)p-valueDemographicsAge (years), mean ± SD68.5 ± 12.366.2 ± 11.864.7 ± 13.50.012Male sex, n (%)218 (57.4)76 (56.3)102 (56.7)0.953BMI (kg/m²), mean ± SD27.8 ± 4.628.2 ± 4.326.9 ± 4.80.087 Vascular risk factors Hypertension, n (%)289 (76.1)98 (72.6)152 (84.4)0.021Diabetes mellitus, n (%)142 (37.4)48 (35.6)58 (32.2)0.515Dyslipidemia, n (%)201 (52.9)73 (54.1)67 (37.2)0.002Current smoking, n (%)95 (25.0)32 (23.7)61 (33.9)0.062Atrial fibrillation, n (%)98 (25.8)24 (17.8)22 (12.2)< 0.001Prior stroke/TIA, n (%)78 (20.5)31 (23.0)28 (15.6)0.198Known CAD, n (%)87 (22.9)29 (21.5)31 (17.2)0.327Heart failure, n (%)34 (8.9)8 (5.9)12 (6.7)0.442CKD (eGFR < 60), n (%)112 (29.5)32 (23.7)45 (25.0)0.363 Clinical presentation Admission SBP (mmHg)156 ± 28148 ± 24178 ± 35< 0.001Admission HR (bpm)82 ± 1878 ± 1684 ± 200.019NIHSS at admission, median (IQR)8 (4–15)2 (1–3)10 (6–18)< 0.001mRS at discharge, median (IQR)3 (2–4)1 (0–2)4 (3–5)< 0.001 Cardiac biomarkers at admission hs-cTnI (ng/L), median (IQR)18.5 (8.2–52.3)6.8 (3.5–14.2)28.7 (12.4–78.5)< 0.001Troponin > 99th percentile, n (%)167 (43.9)28 (20.7)98 (54.4)< 0.001NT-proBNP (pg/mL), median (IQR)485 (168–1342)246 (98–678)612 (245–1685)< 0.001CK-MB (ng/mL), median (IQR)3.2 (1.8–5.4)2.1 (1.2–3.5)4.1 (2.3–7.2)< 0.001hs-CRP (mg/L), median (IQR)5.8 (2.3–12.4)3.2 (1.5–7.8)8.9 (3.8–18.2)< 0.001D-dimer (µg/mL), median (IQR)0.82 (0.45–1.65)0.58 (0.32–1.12)1.24 (0.68–2.48)< 0.001 Peak troponin (24 h) hs-cTnI (ng/L), median (IQR)24.8 (10.5–68.4)8.2 (4.1–16.5)38.6 (16.2–102.3)< 0.001Δ Troponin (ng/L), median (IQR)4.5 (1.2–15.8)1.2 (0.3–3.8)8.4 (2.8–24.5)< 0.001 ECG & echocardiography New-onset AF, n (%)34 (8.9)6 (4.4)18 (10.0)0.182QTc prolongation (> 450 ms), n (%)112 (29.5)28 (20.7)67 (37.2)0.008ST-T changes, n (%)145 (38.2)32 (23.7)78 (43.3)0.002LVEF < 50%, n (%)56 (14.7)12 (8.9)28 (15.6)0.195RWMA present, n (%)67 (17.6)14 (10.4)34 (18.9)0.114BMI, body mass index; CAD, coronary artery disease; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; SBP, systolic blood pressure; HR, heart rate; NIHSS, National Institutes of Health Stroke Scale; mRS, modified Rankin Scale; IQR, interquartile range; hs-cTnI, high-sensitivity cardiac troponin I; CK-MB, creatine kinase-MB; hs-CRP, high-sensitivity C-reactive protein; AF, atrial fibrillation; LVEF, left ventricular ejection fraction; RWMA, regional wall motion abnormality. p-values are overall from one-way ANOVA (normally distributed continuous variables), Kruskal-Wallis test (non-normally distributed continuous variables), or chi-squared test (categorical variables). Pairwise post hoc comparisons were performed using Tukey HSD (ANOVA) or Dunn’s test (Kruskal-Wallis); all pairwise differences between stroke types were statistically significant at p < 0.05 for biomarkers with overall p < 0.001, except where noted. Baseline clinical characteristics and cardiac biomarkers by stroke type. BMI, body mass index; CAD, coronary artery disease; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; SBP, systolic blood pressure; HR, heart rate; NIHSS, National Institutes of Health Stroke Scale; mRS, modified Rankin Scale; IQR, interquartile range; hs-cTnI, high-sensitivity cardiac troponin I; CK-MB, creatine kinase-MB; hs-CRP, high-sensitivity C-reactive protein; AF, atrial fibrillation; LVEF, left ventricular ejection fraction; RWMA, regional wall motion abnormality. p-values are overall from one-way ANOVA (normally distributed continuous variables), Kruskal-Wallis test (non-normally distributed continuous variables), or chi-squared test (categorical variables). Pairwise post hoc comparisons were performed using Tukey HSD (ANOVA) or Dunn’s test (Kruskal-Wallis); all pairwise differences between stroke types were statistically significant at p < 0.05 for biomarkers with overall p < 0.001, except where noted. Hypertension was most prevalent in hemorrhagic stroke (84.4% vs. 76.1% ischemic vs. 72.6% TIA; p = 0.021). Atrial fibrillation was most frequent in ischemic stroke (25.8% vs. 17.8% TIA vs. 12.2% hemorrhagic; p < 0.001). Stroke severity (NIHSS) was greatest in hemorrhagic stroke (median 10, IQR 6–18) and ischemic stroke (median 8, IQR 4–15), with minimal deficits in TIA (median 2, IQR 1–3; p < 0.001). Cardiac biomarkers by stroke type Biomarker elevations differed significantly across stroke types. Admission hs-cTnI was highest in hemorrhagic stroke (median 28.7 ng/L, IQR 12.4–78.5), intermediate in ischemic stroke (median 18.5 ng/L, IQR 8.2–52.3), and lowest in TIA (median 6.8 ng/L, IQR 3.5–14.2; p < 0.001). Troponin elevation rates were 54.4%, 43.9%, and 20.7%, respectively (p < 0.001). NT-proBNP and CK-MB followed similar patterns (Table 1; Fig. 2). Distribution of cardiac biomarker levels across stroke types and ischemic stroke localizations. Cardiac biomarkers by ıschemic stroke localization Among ischemic stroke patients, insular involvement demonstrated the highest biomarker elevations. Median admission hs-cTnI was 45.8 ng/L (IQR 22.4–112.5) in insular strokes versus 28.6 ng/L in posterior circulation, 14.2 ng/L in anterior non-insular, and 9.8 ng/L in subcortical-only (p < 0.001). Troponin elevation rates were 73.9%, 52.9%, 36.2%, and 20.0%, respectively (p < 0.001). NT-proBNP was also highest in insular strokes (median 892 pg/mL, IQR 358–2145; p < 0.001). Peak hs-cTnI and delta troponin followed the same pattern (Supplementary Table S1, Figs. 2 and 3). Cardiac biomarker levels and one-year outcomes stratified by stroke subgroup. Right insular injury (n = 58) was associated with higher median hs-cTnI (52.4 ng/L) than left insular injury (n = 27, 36.5 ng/L; p = 0.042). Insular laterality did not independently predict mortality (HR 1.24, 95% CI 0.68–2.25; p = 0.483) or MACCE (HR 1.18, 95% CI 0.71–1.96; p = 0.524) on multivariable analysis. ECG and echocardiographic findings QTc prolongation was most common in hemorrhagic stroke (37.2%) and insular ischemic strokes (45.7%; p = 0.003 for localization comparison). ST-T wave changes were most frequent in insular strokes (58.7%) and hemorrhagic stroke (43.3%; p < 0.001). New-onset atrial fibrillation during hospitalization was highest in insular strokes (19.6%; p = 0.004). Regional wall motion abnormalities were most frequent in insular strokes (30.4%; p = 0.007) (Table 1 and Supplementary Table S1). As echocardiography was performed post-stroke in all patients, these findings may reflect neurogenic injury, unmasked pre-existing cardiomyopathy, or both; causal attribution to acute neurogenic injury alone cannot be established. Primary outcomes at 12 months Overall mortality was 20.4% (142/695), highest in hemorrhagic stroke (31.1%), followed by ischemic stroke (20.5%) and TIA (5.9%; p < 0.001). Cardiovascular mortality occurred in 12.5% overall. MACCE occurred in 31.4% (218/695), with rates of 40.0%, 33.7%, and 13.3% in hemorrhagic stroke, ischemic stroke, and TIA, respectively (p < 0.001) (Table 2; Fig. 3). Table 2Twelve-month clinical outcomes stratified by stroke type and troponin elevation status.OutcomeOverall (n = 695)Ischemic Stroke (n = 380)TIA (n = 135)Hemorrhagic Stroke (n = 180)p-valuePrimary outcomesAll-cause mortality, n (%)142 (20.4)78 (20.5)8 (5.9)56 (31.1)< 0.001Cardiovascular mortality, n (%)87 (12.5)52 (13.7)4 (3.0)31 (17.2)< 0.001MACCE (composite), n (%)218 (31.4)128 (33.7)18 (13.3)72 (40.0)< 0.001 Secondary outcomes Individual MACCE parameters CV death, n (%)87 (12.5)52 (13.7)4 (3.0)31 (17.2)< 0.001Non-fatal MI, n (%)34 (4.9)24 (6.3)3 (2.2)7 (3.9)0.146Recurrent stroke, n (%)98 (14.1)58 (15.3)8 (5.9)32 (17.8)0.006Unplanned revascularization, n (%)28 (4.0)19 (5.0)4 (3.0)5 (2.8)0.439HF hospitalization, n (%)56 (8.1)38 (10.0)3 (2.2)15 (8.3)0.018Any rehospitalization, n (%)312 (44.9)178 (46.8)42 (31.1)92 (51.1)< 0.001New-onset AF, n (%)67 (9.6)42 (11.1)8 (5.9)17 (9.4)0.236Poor functional outcome (mRS ≥ 3), n (%)285 (41.0)162 (42.6)18 (13.3)105 (58.3)< 0.001 Outcomes by troponin status Troponin positive (> 99th percentile, n = 293)All-cause mortality108 (36.9)62 (37.1)6 (21.4)40 (40.8)0.124MACCE156 (53.2)96 (57.5)8 (28.6)52 (53.1)0.006Recurrent stroke58 (19.8)36 (21.6)2 (7.1)20 (20.4)0.185 Troponin negative (n = 402) All-cause mortality34 (8.5)16 (7.5)2 (1.9)16 (19.5)< 0.001MACCE62 (15.4)32 (15.0)10 (9.3)20 (24.4)0.012Recurrent stroke40 (10.0)22 (10.3)6 (5.6)12 (14.6)0.203MACCE, major adverse cardiac and cerebrovascular events; CV, cardiovascular; MI, myocardial infarction; HF, heart failure; AF, atrial fibrillation; mRS, modified Rankin Scale. Percentages represent proportions within each subgroup. p-values reflect chi-squared or Fisher’s exact tests for overall group differences; pairwise post hoc comparisons between stroke types confirmed significant differences for all primary outcomes except where p > 0.05 is explicitly noted. Twelve-month clinical outcomes stratified by stroke type and troponin elevation status. MACCE, major adverse cardiac and cerebrovascular events; CV, cardiovascular; MI, myocardial infarction; HF, heart failure; AF, atrial fibrillation; mRS, modified Rankin Scale. Percentages represent proportions within each subgroup. p-values reflect chi-squared or Fisher’s exact tests for overall group differences; pairwise post hoc comparisons between stroke types confirmed significant differences for all primary outcomes except where p > 0.05 is explicitly noted. Secondary outcomes Individual MACCE components included cardiovascular death (12.5%), non-fatal MI (4.9%), recurrent stroke (14.1%), unplanned coronary revascularization (4.0%), and heart failure hospitalization (8.1%). All-cause rehospitalization occurred in 44.9%. Poor functional outcome (mRS ≥ 3) occurred in 41.0%, highest in hemorrhagic stroke (58.3%) (Table 2). Outcomes by troponin elevation status Among 293 patients with elevated troponin, mortality was 36.9% versus 8.5% in 402 non-elevated patients (4.3-fold difference). MACCE occurred in 53.2% versus 15.4% (3.5-fold difference), and recurrent stroke in 19.8% versus 10.0% (Table 2). Kaplan-Meier curves stratified by troponin status and stroke type confirmed significantly worse cumulative event-free survival in the elevated troponin group (log-rank p < 0.001 for all primary outcomes; Supplementary Figure S1). Multivariable predictors of adverse outcomes In Cox regression for all-cause mortality, troponin elevation was the strongest predictor (HR 3.24, 95% CI 2.18–4.82; p < 0.001). Other independent predictors included age (HR 1.68 per 10 years), NIHSS (HR 1.08 per point), atrial fibrillation (HR 1.82), reduced eGFR (HR 1.54), elevated NT-proBNP (HR 1.42 per log unit), and insular involvement in ischemic stroke (HR 1.88; all p < 0.05) (Table 3; Fig. 4). For MACCE, troponin elevation (HR 2.86) and delta troponin (HR 1.28 per log unit) independently predicted outcomes alongside age, NIHSS, atrial fibrillation, reduced eGFR, and diabetes. For recurrent stroke, atrial fibrillation was the strongest predictor (HR 2.14), followed by prior stroke/TIA (HR 2.08), troponin elevation (HR 1.92), and hypertension (HR 1.68; all p < 0.05) (Table 3; Fig. 4). Table 3Multivariable cox proportional hazards regression analysis for one-year outcomes.VariableHR (95% CI)p-valueModel 1: Predictors of All-Cause Mortality (n = 695)Troponin > 99th percentile3.24 (2.18–4.82)< 0.001Age (per 10 years)1.68 (1.42–1.98)< 0.001Male sex1.12 (0.78–1.61)0.532NIHSS at admission (per point)1.08 (1.05–1.11)< 0.001Atrial fibrillation1.82 (1.24–2.68)0.002eGFR < 60 mL/min/1.73 m²1.54 (1.08–2.21)0.018Stroke type (ref: Ischemic)- TIA0.28 (0.13–0.58)0.001- Hemorrhagic1.76 (1.21–2.57)0.003NT-proBNP (per log unit)1.42 (1.18–1.71)< 0.001Insular involvement (ischemic only)1.88 (1.15–3.07)0.012 Model 2: predictors of MACCE (n = 695) Troponin > 99th percentile2.86 (2.12–3.86)< 0.001Age (per 10 years)1.38 (1.21–1.57)< 0.001Male sex1.28 (0.96–1.69)0.088NIHSS at admission (per point)1.06 (1.04–1.08)< 0.001Atrial fibrillation1.64 (1.21–2.23)0.001eGFR < 60 mL/min/1.73 m²1.36 (1.02–1.81)0.035Diabetes mellitus1.42 (1.08–1.87)0.013Stroke type (ref: Ischemic)- TIA0.35 (0.21–0.57)< 0.001- Hemorrhagic1.45 (1.08–1.95)0.014Δ Troponin (per log unit)1.28 (1.12–1.46)< 0.001 Model 3: predictors of recurrent stroke (n = 695) Troponin > 99th percentile1.92 (1.28–2.88)0.002Age (per 10 years)1.24 (1.06–1.46)0.008Male sex1.18 (0.79–1.76)0.412NIHSS at admission (per point)1.04 (1.01–1.07)0.006Atrial fibrillation2.14 (1.42–3.23)< 0.001Prior stroke/TIA2.08 (1.36–3.18)0.001Hypertension1.68 (1.02–2.78)0.042Stroke type (ref: Ischemic)- TIA0.38 (0.18–0.79)0.010- Hemorrhagic1.28 (0.83–1.98)0.265Model 4: interaction analysis - troponin × insular involvement (ischemic stroke only, n = 380) For all-cause mortality Troponin > 99th percentile2.82 (1.52–5.24)0.001Insular involvement1.94 (1.08–3.48)0.026Troponin × Insular interaction2.15 (1.08–4.28)0.029Age (per 10 years)1.72 (1.38–2.14)< 0.001NIHSS (per point)1.09 (1.05–1.13)< 0.001Atrial fibrillation1.86 (1.15–3.01)0.011 For MACCE Troponin > 99th percentile2.58 (1.64–4.06)< 0.001Insular involvement1.76 (1.12–2.76)0.014Troponin × Insular interaction1.84 (1.12–3.02)0.016Age (per 10 years)1.42 (1.19–1.69)< 0.001NIHSS (per point)1.07 (1.04–1.10)< 0.001HR, hazard ratio; CI, confidence interval; NIHSS, National Institutes of Health Stroke Scale; eGFR, estimated glomerular filtration rate; TIA, transient ischemic attack; MACCE, major adverse cardiac and cerebrovascular events.Models 1–3 adjusted for age, sex, NIHSS, atrial fibrillation, eGFR, and stroke type. Model 4 restricted to ischemic stroke patients and adjusted for posterior circulation involvement. Additional sensitivity analyses adjusting for hypertension, smoking status, diabetes mellitus, known coronary artery disease, and heart failure yielded consistent results (see Sensitivity Analyses in Results). Hazard ratios reflect independent associations after multivariable adjustment; these analyses are associative and do not establish causality. Multivariable cox proportional hazards regression analysis for one-year outcomes. HR, hazard ratio; CI, confidence interval; NIHSS, National Institutes of Health Stroke Scale; eGFR, estimated glomerular filtration rate; TIA, transient ischemic attack; MACCE, major adverse cardiac and cerebrovascular events. Models 1–3 adjusted for age, sex, NIHSS, atrial fibrillation, eGFR, and stroke type. Model 4 restricted to ischemic stroke patients and adjusted for posterior circulation involvement. Additional sensitivity analyses adjusting for hypertension, smoking status, diabetes mellitus, known coronary artery disease, and heart failure yielded consistent results (see Sensitivity Analyses in Results). Hazard ratios reflect independent associations after multivariable adjustment; these analyses are associative and do not establish causality. Multivariable predictors of one-year outcomes: forest plots of adjusted hazard ratios. Interaction analysis In ischemic stroke patients (n = 380), significant synergistic interactions were identified between troponin elevation and insular involvement for mortality (interaction HR 2.15, 95% CI 1.08–4.28; p = 0.029) and MACCE (interaction HR 1.84, 95% CI 1.12–3.02; p = 0.016) (Table 3). Patients with both elevated troponin and insular involvement (n = 68) had the highest risk: mortality 50.0% (adjusted HR 15.42, 95% CI 6.89–34.52) and MACCE 69.1% (adjusted HR 11.24, 95% CI 6.42–19.68). The relative excess risk due to interaction (RERI) for mortality was 5.46 (95% CI 0.84–10.08; p = 0.021), confirming a significant synergistic effect beyond additive contributions. Predictive performance of cardiac biomarkers For 1-year mortality, hs-cTnI showed optimal discriminative ability (AUC 0.782, 95% CI 0.741–0.823; cutoff 24.5 ng/L, sensitivity 79.8%, specificity 72.5%). For MACCE, hs-cTnI achieved AUC 0.745 (cutoff 19.4 ng/L, sensitivity 74.3%, specificity 68.9%). For recurrent stroke, AUC was 0.672 (cutoff 18.2 ng/L). Hs-cTnI consistently outperformed NT-proBNP (AUC 0.724 for mortality) and CK-MB (AUC 0.658) across all outcomes (Fig. 5). The addition of hs-cTnI to a baseline model incorporating age and sex significantly improved discrimination for both mortality and MACCE (DeLong p < 0.001 for both). Decision curve analysis demonstrated net benefit of troponin-based stratification over treat-all and treat-none strategies across threshold probabilities of 10–40% (Supplementary Figure S2). Receiver operating characteristic curves for cardiac biomarkers predicting one-year outcomes. Sensitivity analyses Troponin elevation remained independently predictive after adjusting for known coronary artery disease and heart failure (HR 2.98; p < 0.001) and after excluding patients with these conditions (HR 3.12; p < 0.001). Across CKD strata, troponin elevation remained predictive (HRs 2.84–3.52). Using admission troponin alone, the HR for mortality was 2.94 (p < 0.001). Dıscussıon This prospective cohort study of 695 stroke patients demonstrates that cardiac biomarker elevations vary significantly across stroke types and ischemic stroke localizations, and independently predict adverse 1-year outcomes. The study extends current knowledge in three principal respects: by systematically characterizing biomarker profiles across four distinct ischemic stroke localizations; by demonstrating a statistically significant synergistic interaction between insular involvement and troponin elevation; and by providing multivariable-adjusted hazard estimates for the full spectrum of MACCE components with concurrent assessment of hs-cTnI, NT-proBNP, and CK-MB. Our overall troponin elevation rate of 42.2% falls within the previously reported range of 18–60%18,19. The gradient from TIA to ischemic stroke to hemorrhagic stroke aligns with the severity of neurological injury and degree of autonomic dysregulation. The median admission hs-cTnI of 28.7 ng/L in hemorrhagic stroke reflects the more profound catecholaminergic surge typically observed in intracranial hemorrhage20,21. Neurogenic cardiac injury, mediated through excessive catecholamine release and autonomic imbalance, represents a direct brain-heart interaction that can produce myocardial damage without obstructive coronary artery disease4,5. Additional contributing factors include pre-existing cardiac disease unmasked by the acute stress of stroke, hemodynamic fluctuations, systemic inflammation, and demand ischemia22,23. The significant rise in peak troponin at 24 h, particularly in insular strokes (delta troponin 12.8 ng/L), suggests ongoing myocardial injury rather than merely pre-existing disease, consistent with the independent prognostic value of delta troponin observed in the TRELAS cohort13,24. The marked elevation of cardiac biomarkers in insular strokes, with troponin elevation rates of 73.9%, corroborates the critical role of the insular cortex in cardiovascular autonomic regulation9,25. The right insula modulates sympathetic and parasympathetic outflow to the heart; damage leads to unopposed sympathetic activation and catecholamine-mediated myocardial injury6. Previous experimental and clinical studies have documented cardiovascular effects of insular stimulation and insular stroke-related ECG abnormalities, troponin elevations, and wall motion abnormalities in the absence of coronary artery disease26–28. Our study extends these observations to include higher rates of QTc prolongation, new-onset atrial fibrillation, and regional wall motion abnormalities in insular strokes compared to all other localizations. Troponin elevation was the strongest independent predictor of all-cause mortality (HR 3.24), MACCE (HR 2.86), and recurrent stroke (HR 1.92), consistent with prior studies reporting hazard ratios of 1.5 to 4.012,14,32. The absolute risk differences were striking: 36.9% vs. 8.5% mortality and 53.2% vs. 15.4% MACCE in elevated versus non-elevated troponin groups. Within elevated troponin patients, TIA showed notably lower mortality than ischemic or hemorrhagic stroke (21.4% vs. 37.1% vs. 40.8%; p = 0.124), suggesting that stroke type continues to contribute to prognosis alongside biomarker status. Cardiovascular mortality accounted for 61.3% of all deaths, emphasizing that stroke should be viewed as a marker of systemic vascular disease with sustained cardiovascular risk. The high rate of recurrent stroke (14.1%) and identification of atrial fibrillation as the strongest predictor (HR 2.14) underscores the importance of rhythm monitoring and anticoagulation in stroke survivors35. The synergistic interaction between troponin elevation and insular involvement is, to our knowledge, novel. Patients with both factors (n = 68) had 1-year mortality of 50.0% and MACCE of 69.1%, substantially higher than either factor alone. This synergistic effect likely reflects sustained autonomic dysregulation from insular damage, perpetuating sympathetic hyperactivity, impaired parasympathetic tone, and increased susceptibility to arrhythmias and cardiac dysfunction throughout recovery29,30. Hs-cTnI demonstrated good discriminative performance (AUC 0.782 for mortality, 0.745 for MACCE) and incrementally improved discrimination beyond a model with age and sex alone (DeLong p< 0.001). NT-proBNP provided additional independent prognostic information (AUC 0.724 for mortality), consistent with prior data31–35. These values support the clinical utility of biomarker assessment for risk stratification in stroke patients. Several limitations warrant acknowledgment. This was a single-center study, limiting generalizability. Patients with recent MI, PCI, cardiac surgery, or severe CKD requiring dialysis were excluded, limiting applicability to these high-risk groups. Absolute infarct volume was not systematically measured; NIHSS is an imperfect proxy for infarct burden, and future studies should incorporate volumetric lesion analysis to disentangle the effects of insular location from overall infarct size. CT-based localization in 13.7% of ischemic stroke patients may introduce some misclassification bias. Distinguishing neurogenic wall motion abnormalities from pre-existing ischemic cardiomyopathy was not always possible without pre-stroke echocardiography; causal inferences regarding neurogenic cardiac injury should be interpreted cautiously. Cardiac MRI and coronary angiography were not performed systematically. Follow-up was limited to one year. Conclusıons In this prospective cohort study of 695 stroke patients, cardiac biomarker elevations were common and varied significantly by stroke type and anatomical localization, with particularly marked elevations in hemorrhagic stroke and insular ischemic stroke. Elevated hs-cTnI independently predicted all-cause mortality, MACCE, and recurrent stroke at one year after adjustment for traditional risk factors and stroke severity. Insular involvement and troponin elevation demonstrated synergistic effects on adverse outcomes, identifying a particularly high-risk subgroup. These findings support routine cardiac biomarker assessment and integration of neuroimaging localization data for risk stratification in stroke patients. Further research is needed to determine whether targeted interventions in patients identified by this combined approach can reduce post-stroke cardiovascular complications. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Acknowledgements The authors used AI for language editing and minor grammatical corrections during the preparation of this manuscript. The AI tool was not used for study design, data collection, statistical analysis, interpretation of results, or generation of core scientific content. After using this tool for language refinement, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. All scientific content, data analysis, and conclusions represent the independent work and judgment of the authors. The research presented in this manuscript does not involve the development or utilization of any custom software applications or specific code. As such, there is no code availability associated with this study. Author contributions F.P. and H.T. conceived and designed the study. F.P., H.T., Y.V., G.S.D. and M.A.Ö. performed data collection and extraction from electronic health records. F.P. conducted the statistical analyses and drafted the initial manuscript. F.P. and Y.V. critically revised the manuscript for important intellectual content. F.P., H.T., M.A.Ö, and S.T. supervised the study and provided senior oversight. All authors contributed to data interpretation, reviewed and approved the final manuscript, and agree to be accountable for all aspects of the work. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Data availability All relevant data supporting the findings of this study are available upon request and will be provided by the corresponding author. Code availability The research presented in this manuscript does not involve the development or utilization of any custom software applications or specific code. As such, there is no code availability associated with this study. The analysis and findings rely on standard statistical methods and commercially available software tools. For any inquiries related to the methodology or data analysis, please contact the corresponding author. Declarations Competing interests The authors declare no competing interests. Ethics committee approval This study was approved by the Ethics Committee for Non-Interventional Health Sciences Research at Istanbul Yeni Yüzyıl University (Decision No: 2020/02-1036, Date: 13.02.2020). Consent for publication Written informed consent for participation and publication of de-identified data was obtained from all participants. All data were anonymized to protect patient confidentiality, and no personally identifiable information is disclosed in this publication. Informed consent Written informed consent was obtained from all participants who participated in this study. This prospective single-center study included patients who were enrolled after providing written informed consent. All participants were informed about the study procedures and voluntarily agreed to participate. Confidentiality was ensured throughout the research, and all data were handled in accordance with institutional privacy policies and relevant data protection regulations. Consent to participate All participants provided written informed consent prior to inclusion in the study. The study was conducted in accordance with the principles of the Declaration of Helsinki. All procedures adhered to ethical standards to ensure participant safety, privacy, and data security. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Supplementary Material 4 Data Availability Statement All relevant data supporting the findings of this study are available upon request and will be provided by the corresponding author. The research presented in this manuscript does not involve the development or utilization of any custom software applications or specific code. As such, there is no code availability associated with this study. The analysis and findings rely on standard statistical methods and commercially available software tools. For any inquiries related to the methodology or data analysis, please contact the corresponding author.