Background: Stroke is the most frequent cause of neurogenic oropharyngeal dysphagia. Its frequency is greater than 50% in the acute phase. The early clinical evaluation of swallowing disorders can help define approaches and avoid oral feeding, which may be detrimental to the patient. The aim of this study was to identify predictive clinical factors associated with enteral tube feeding (ETF) in acute stroke patients to develop an ETF predictive score. Methods: The medical records of 1104 acute ischemic stroke patients from our prospective stroke database were reviewed. Clinical factors as age, sex, blood pressure, glycemia, NIHSS score, Glasgow coma scale, previous Rankin, localization, and classification of acute stroke, and comorbidity index were analyzed. Logistic multivariate regression was used to identify perfect predictors of early ETF placement. The sample was randomly divided into two samples (30 and 70% of the sample) to proceed the internal score validation. Results: 1104 patients were enrolled. Mean age was 65.9 years old (SD 13), male patients were 51.7%, mean Glasgow score was 13.9 (SD 1.9), NIHSS at admission was 7.2 (SD 5.7), mean ASPECT score was 9.3 (1.6). Multivariate logistic regression disclosed age (odds ratio [OR] 1.02; CI 95% 1.00-1.03, p=0.005), initial NIHSS (OR 1.15; CI 95% 1.11-1.19, p<0.001) and NIHSS dysarthria subitem (OR 1.76; CI 95% 1.36-2.37, p<0.001) as independent predictive variables. The EFT propensity score was constructed based on these variable. A score equal or above four correctly classified 75% of patients (sensitivity = 67%, specificity = 79%). A ROC curve (AUC = 0.78, CI 0.75 - 0.81; p = 0.015) was constructed taking the formal phonoaudiological evaluation as gold-standard. Discussion: The ETF score allows us to quickly identify and indicate the use of ETF in acute stroke patients avoiding delay in starting enteral feeding and reducing the risk of bronchoaspiration, morbidity, and mortality due to pneumonia. Besides, the ETF score might optimize the patient care in hospitals where the speech therapist is not always available. Conclusion: Combining data from age, NIHSS at admission and the dysarthria subitem could be a strong and useful predictor to evaluate and decide about ETF in acute stroke patients.