Background: and hypothesis: We sought to develop and validate the long-term non-hemorrhagic stroke outcome prediction model based on demographic, clinical, and echocardiographic parameters. Material and Methods: Univariate and multivariate logistic regression modeling of long-term survival based on demographic, clinical, and echocardiographic data was performed in 325 consecutive non-hemorrhagic stroke subjects (original cohort). ANOVA, chi-square, Kaplan-Meier, and logistic regression tests were employed. Prediction rules were developed and applied to 1305 patients (validation cohort). The study was approved by the institutional IRB. Results: Only age (hazard ratio 1.6 per decade over 40 years old, 95%CI 1.3-2.2, p<0.001), renal failure (hazard ratio 6.2 95%CI 24.3, p<0.001), and aortic root sclerosis (hazard ratio 2.9 95%CI 1.4-5, p=0.004) were associated with increased all-cause mortality in univariate and multivariate logistic regression analyses. Based on the hazard ratios, risk score was developed with 2 points given for each decade of age over 40 years old, 6 for renal failure, and 3 for aortic root sclerosis. Long-term mortality was compared between 3 groups: low risk (0-5 points), moderate risk (6-10 points), and high risk (more than 11 points). In the original cohort, Kaplan-Meier mean survival estimates were 60+/-1.4 months in low risk, 45+/-2.2 in moderate risk, and 37.4+/-3.3 months high-risk groups (p<0.001). In the validation cohort, Kaplan-Meier mean survival estimates were 53+/-0.8 months in low risk, 47+/-1.3 in moderate risk, and 33+/-2.1 months high-risk groups (p <0.001). When renal function was removed from the analysis, age and aortic root sclerosis still successfully identified patients with poor long term outcomes: 53+/-0.8 months in low risk, 47+/-1.3 in moderate risk, and 34+/-2.4 months high-risk groups (p <0.001). Conclusions: In non-hemorrhagic stroke survivors, long-term prognosis can be predicted with prospectively validated model based on age, presence of aortic root sclerosis, and renal dysfunction. Using such simple, readily available model may help the clinician to identify this high-risk patient group with adverse outcomes, enabling appropriate management and resource utilization.