Objective：To investigate the predictive value of N - terminal pro - brain natriuretic peptide（NT - proBNP）and high - sensitivity cardiac troponin T（hs-cTnT）for long-term mortality risk in patients with acute ischemic stroke（AIS），and to develop and validate corresponding prediction models. Methods：This single - center retrospective study consecutively enrolled AIS patients who underwent thrombectomy at the First Affiliated Hospital of Nanjing Medical University between January and December 2022，with a 2- year follow up. Cox regression and LASSO regression were used to identify factors associated with all-cause mortality. Three predictive models were constructed：a basic model，Model 1（basic model + NT-proBNP），and Model 2（basic model+hs-cTnT），the predictive performance of these models was compared. Results：A total of 230 AIS patients were included in the final analysis and were randomly assigned to the training set（n=146）and testing set（n=84）at a 3∶2 ratio. During follow-up，83 all-cause mortality events occurred，with a mortality rate of 37.2%. Multivariate Cox regression showed that for every 1 000 pg/mL increase in NT-proBNP，the 2-year all-cause mortality increased by 27%（HR=1.27，95% CI：1.15-1.40，P ＜0.001），while ln（hs-cTnT）elevation showed no significant association with mortality risk（HR =1.11，95% CI：0.89-1.38，P=0.372）. Cox regression and LASSO regression identified the following mortality- related variables：history of atrial fibrillation，postoperative National Institutes of Health Stroke scale（NIHSS）score，baseline hemoglobin，white blood cell count，and random blood glucose，which formed the basic model. The area under the curve（AUC）values for the basic model were 0.816（training set）and 0.778（testing set）. Model 1 achieved higher AUC values（0.866 and 0.799， respectively），demonstrating improved predictive ability. Model 2 showed limited improvement（AUC=0.811 and 0.788）. Conclusion： NT - proBNP is an independent predictor of all - cause mortality in AIS patients and enhances performance of traditional clinical indicator models，supporting individualized management of AIS.