The purpose of this study is to compare propofol associated with fentanyl versus midazolam plus fentanyl for sedation during diagnosis or therapeutic upper gastrointestinal endoscopy (UGE) in cirrhotic patients.

UGE is often performed in cirrhotic patients for the diagnosis and treatment of portal hypertension complications. Current data suggests that propofol sedation may have advantages over benzodiazepines. However, there are few reports comparing propofol versus midazolam in patients with liver cirrhosis. The study's objective is to compare propofol associated with fentanyl versus midazolam plus fentanyl for sedation during diagnosis or therapeutic UGE in cirrhotic patients. A prospective randomized study will include cirrhotic patients (Child A, B or C and ASA 2 or 3), referred for diagnostic or therapeutic UGE, randomized for group I: propofol (0,5 mg/kg up to 400 mg) and fentanyl (0,05 mg); or group II: midazolam (0,1 mg/kg) and fentanyl. Sedation was performed by an exclusively dedicated gastroenterologist. Efficacy (completion of procedures), complications (hypoxemia, hypotension, arrhythmias) and recovery time (elapsed from the end of the procedure and discharge) will be studied.

Inclusion Criteria:

• Liver cirrhosis, any etiology
• ASA II or III
• Child A, B or C
• Age between 18 years and 75 years
• Patients that agree in participate of study and signed the contentment term

Exclusion Criteria:

• Schistosomiasis
• Recuse
• Hepatocellular carcinoma
• Contraindications to drugs
• ASA IV or V
• Hepatic encephalopathy, neurologic diseases
• Opioids, narcotics, MAO inhibitors or benzodiazepines use
• Alcohol abuse