After an acute variceal bleeding, early decision for aggressive management of patients with worse prognosis may improve outcomes. The effectiveness of currently recommended standard therapy (drugs plus endoscopic ligation) for different risk subgroups and the validity of available risk criteria in clinical practice are unknown. We analyzed data of 301 consecutive cirrhotic patients admitted with esophageal variceal bleeding. All patients received antibiotics, somatostatin, and in 263 early endoscopic therapy. A stratified 6-week mortality assessment according to risk (low-risk: Child-Pugh B without active bleeding or Child-Pugh A; high-risk: Child-Pugh B with active bleeding or Child-Pugh C) was performed. A multivariate analysis was conducted to elaborate a new risk classification rule. Among the 162 patients receiving emergency ligation, 14% rebled and 16% died. Standard therapy was very effective in all risk strata, even in high-risk patients, specially if eligible for therapeutic trials (child <14, age ≤75 years, creatinine ≤3.0 mg/dl, no hepatocellular carcinoma, or portal thrombosis), showing this stratum a 10% mortality. In patients receiving ligation, Child-Pugh C patients with baseline creatinine <1.0 mg/dl showed similar mortality to Child-Pugh A or B patients (8% vs. 7%, respectively). Only Child-Pugh C patients with creatinine ≥1.0 were at a significant higher risk (Child-Pugh C: 46% mortality if creatinine ≥1.0 vs. 8% if creatinine <1.0, P=0.006). The combination of somatostatin, antibiotics, and endoscopic ligation after an acute variceal bleeding in a real-life situation is associated with very low mortality. Child-Pugh C patients with baseline creatinine ≥1.0 mg/dl should be considered high-risk patients in this setting.