The hospital stay is a fundamental parameter for the management of hospital costs and resources. Endarterectomy is a surgical procedure commonly made for diminishing the risks of long-term stroke. For subjects undergoing endarterectomy it is important to evaluate the length of stay (LOS) because it may be prolonged in case of complications or comorbidity of the patient. In this paper, we deal with the task of predicting LOS value for patients undergoing endarterectomy of the University Hospital “San Giovanni di Dio and Ruggi d’Aragona” of Salerno and “A.O.R.N. Antonio Cardarelli” of Naples, to compare the procedures applied into the two hospitals. Machine learning algorithms were used in the work and for both hospitals the most accurate algorithm was the Random Forest.