In previous studies, the investigators used retrospective analysis of cases of acute upper gastrointestinal bleeding in patients with liver cirrhosis from the Fifth Medical Center of the General Hospital of Beijing PLA, China from January 2018 to May 2019. The investigators performed univariate and multivariate analyses of rebleeding risk and death risk based on all data. Then, based on the analysis of 85% of the sampled data, the investigators randomly selected 85% of the patient data to build a model, and then used the remaining 15% of the patient data for model validation. Re-bleeding risk scores and death risk scores were established, respectively.

This study intends to prospectively verify the two risk scoring systems described above. After statistical calculations, about 500 patients with liver cirrhosis who plan to undergo emergency gastroscopy for acute upper gastrointestinal bleeding within the next 5 months at the Fifth Medical Center of Beijing General Hospital of China performed in adult patients. The investigators will exclude patients with incomplete or lost follow-up records. Perform patient self-control,using the existing upper gastrointestinal bleeding risk scores (AIMS65, Rockall, and Blatchford) and the previous scoring system model separately, compared with the actual rebleeding rate and mortality for comparison. To verify and revise the rebleeding risk score and death risk score that the investigators constructed earlier.The data were statistical processed by a professional statistician.

The establishment of an acute upper gastrointestinal bleeding rebleeding and death risk scoring system for patients with liver cirrhosis can help distinguish patients with high or low risk of rebleeding or death to determine the patient's treatment needs.

In previous studies, the investigators used retrospective analysis of cases of acute upper gastrointestinal bleeding in patients with liver cirrhosis from the Fifth Medical Center of the General Hospital of Beijing PLA, China from January 2018 to May 2019. The survey and return visits were completed by the patient's clinician, and the subject researcher was responsible for reviewing and including a complete questionnaire for eligible patients. The database includes demographic information, vital signs and other key clinical manifestations, endoscopic findings, laboratory test results, imaging studies and complications, and rebleeding and deaths within 42 days. Covers known risk factors for rebleeding and death.

First, the investigators performed univariate and multivariate analyses of rebleeding risk and death risk based on all data. Second, based on the analysis of 85% of the sampled data, the investigators randomly selected 85% of the patient data to build a model, and then used the remaining 15% of the patient data for model validation. Re-bleeding risk scores and death risk scores were established respectively.

This study intends to prospectively verify the two risk scoring systems described above. After statistical calculations, about 500 patients with liver cirrhosis who plan to undergo emergency gastroscopy for acute upper gastrointestinal bleeding within the next 5 months at the Fifth Medical Center of Beijing General Hospital of China Performed in adult patients. the investigators will exclude patients with incomplete or lost follow-up records. One researcher was responsible for patient identification and data collection during the initial review, and one researcher performed subsequent data entry, which was statistical processed by a professional statistician.

Perform patient self-control,using the existing upper gastrointestinal bleeding risk scores (AIMS65, Rockall, and Blatchford) and the previous scoring system model separately, compared with the actual rebleeding rate and mortality for comparison. To verify and revise the rebleeding risk score and death risk score that the investigators constructed earlier.

The establishment of an acute upper gastrointestinal bleeding rebleeding and death risk scoring system for patients with liver cirrhosis can help distinguish patients with high or low risk of rebleeding or death to determine the patient's treatment needs.

Inclusion Criteria:

• Patients with histologically proven cirrhosis or clinical and ultrasonographic data compatible with the diagnosis of cirrhosis
• Above 18 years old patients
• Clinical evidence of bleeding (hematemesis and/or melena) during the previous 24 hours
• Undergo the emergency endoscopy within 2 hours at admittance to emergency room or in the general ward from the initial evaluation
• Who agree to participate in the study

Exclusion Criteria:

• Unfit for resuscitation
• Pregnancy and lactation

Patients with histologically proven cirrhosis or clinical and ultrasonographic data compatible with the diagnosis of cirrhosis were included.