BACKGROUND AND OBJECTIVES To determine clinical predictors of in-hospital mortality in patients with middle cerebral artery infarcts (MCAI). PATIENTS AND METHODS Data from 1.355 patients with MCAI were obtained from consecutive strokes included in the "Sagrat Cor Hospital of Barcelona Stroke Registry". Demographic, clinical, neuroimaging and outcome variables in the subgroup of patients who died were compared with those in the surviving subgroup. The independent predictive value of each variable on the development of death was assessed with a logistic regression analysis. Two predictive models were constructed. A first model was based on demographic, risk factors and clinical variables (total 14 variables). A second model was based on demographic, risk factors, clinical and outcome variables (total 20). RESULTS In-hospital death was observed in 235 patients (17.3%). Early seizures (OR=4.49; CI 95%: 1.77-11.40), 85 years old or more (OR=2.61; CI 95%: 1.88-2.60), atrial fibrillation (OR=2.57; CI 95%: 1.89-3.49), limb weakness (OR=2.55; CI 95%: 1.40-4.66), cardiac heart disease (OR=2.33; CI 95%: 1.43-3.80) and sensory symptoms (OR=2.29; CI 95%: 1.68-3.12) appeared to be independent prognostic factors of in-hospital mortality in the first predictive model. In addition to these variables, cardiac complications (OR=5.50: CI 95%: 3.21-9.40), peripheral vascular complications (OR=3.74; CI 95%: 1.58-8.85), previous cerebral infarct (OR=1.89: CI 95%: 1.27-2.80), infections (OR=1.82; CI 95%; 1.27-2.61), and lacunar infarcts (OR=0.02; CI 95%: 0.01-0.17), appeared to be independent prognostic factors of in-hospital mortality in the second model. CONCLUSIONS Clinical features easily obtained at the patient's bedside help clinicians to predict in-hospital mortality in patients with MCAI. Early seizures and age 85 years old or more, were the main clinical predictors of in-hospital mortality.