目的：预测初发脑梗塞住院患者出院时的步行能力。方法：回顾性分析１９９１—１９９４年住院的初发脑梗塞患者，共３６０例（１４例死亡），２０４例入院时不能行走，按出院时能否独立行走分为行走组（９５例）和非行走组（１０９例）。对发病时情况，入院后检查结果等参数进行单因素（ｔ检验和ｘ２检验）和多元回归分析。结果：单因素分析中年龄、高血压、糖尿病、尿失禁、大片梗塞及住院天数２组间差异有显著性（Ｐ＜０．０５）；多因素分析中年龄、性别、失语、头晕、住院天数、入院时血压、糖尿病、冠心病、病灶所在半球、大片、多发梗塞、血尿素氮、尿白细胞等１４个因素对步行能力有显著性影响（Ｐ＝０．０１—０．０５）。结论：根据初发脑梗塞患者发病及入院时的资料可以预测出院时的步行能力，并为拟定早期康复计划提供客观依据。 PURPOSE: To predict the ability of ambulatory patients to discharge at first discharge of cerebral infarction. Methods: A total of 360 patients (14 deaths) were admitted to our hospital from 1991 to 1994 with initial cerebral infarction. 204 patients were unable to walk on admission and were divided into walking group (95 cases) and non-walking group Walking group (109 cases). The onset of illness, post-admission test results and other parameters of single factor (t test and x2 test) and multiple regression analysis. Results: There were significant differences in age, hypertension, diabetes mellitus, urinary incontinence, major infarction and length of hospital stay between the two groups in the univariate analysis (P <0.05). In the multivariate analysis, age, gender, aphasia, There were 14 factors such as blood pressure, diabetes mellitus, coronary heart disease, the hemisphere at the lesion, large tracts, multiple infarcts, blood urea nitrogen and urine leucocyte on admission, which had a significant effect on walking ability (P = 0.01-0.05). Conclusion: Based on the data of onset and admission of patients with newly diagnosed cerebral infarction, we can predict the walking ability at discharge and provide an objective basis for formulating early rehabilitation plan.