Neuroimaging of acute myocardial injury in stroke: insights into brain lesion locations and network disconnections

Introduction: Acute myocardial injury (AMI) is frequently observed in patients with acute stroke and may be linked to lesion location within the central autonomic network (CAN). However, neuroimaging findings remain inconsistent, and the neuroanatomical basis remains unclear. We investigated associations between lesion location, structural disconnection, and AMI in acute stroke. Methods: In this single-center cohort, patients with ischemic or hemorrhagic stroke were screened for AMI using serial high-sensitivity cardiac troponin I (hs-cTnI) measurements within 48 h of admission. AMI was defined as hs-cTnI levels exceeding the 99th percentile of the reference limit with >20% dynamic change. Stroke lesions were segmented using Brainomix software and overlaid on a diffusion tensor imaging template to generate indirect structural disconnection maps. A voxel-based multivariate method (Sparse Canonical Correlation Analysis) identified associations between AMI and (a) directly damaged areas and (b) disconnected regions. Results: Among 281 patients (mean age 71.4 ± 12.3 years), 59 (21%) developed AMI. AMI was associated with lesions in the left insula, basal ganglia, and adjacent white matter, as well as with widespread structural disconnection involving CAN regions. It was also significantly associated with older age, female sex, and higher NIHSS scores, but not with atrial fibrillation, diabetes, or hypertension. Conclusion: In this cohort, AMI affected 21% of stroke patients and was significantly associated with specific lesion locations and widespread structural disconnection, including regions implicated in autonomic regulation, but not with traditional cardiovascular risk factors. These findings support the hypothesis that stroke-induced autonomic dysregulation may contribute to AMI development.