Analysis of antibiotic prophylaxis in the prevention of rebleeding in patients with acute variceal hemorrhage andChild-Pugh classification A liver cirrhosis

• Main Authors: Min-Chin Chiu, 邱敏欽
• Other Authors: Ming-Shiang Wu
• Format: Others
• Language: en_US
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/6ed8y7

碩士 === 國立臺灣大學 === 臨床醫學研究所 === 107 === Introduction Gastroesophageal variceal bleeding is one of the most common and serious complications in cirrhosis. The existing data has shown that infection had negative influence on hemostasis in cirrhotic patients. Due to immunocompromised and bacterial translocation, cirrhotic patients have higher risks of infection when suffering from gastrointestinal bleeding. Antimicrobial agents are of proved efficacy to prevent infection and rebleeding after variceal hemorrhage, especially in decompensated cirrhosis. However, the discussions and reports are still lacking about Child-Pugh (CP) class A. We thus conducted a retrospective study to analyze the relationship between prophylactic antibiotics and rebleeding and infection during acute variceal hemorrhage in CP class A cirrhosis. Aims To analyze whether antibiotic prophylaxis prevents rebleeding and infection in CP class A cirrhotic patients with acute variceal hemorrhage Methods and Materials In this retrospective study, we reviewed the endoscopic reports and medical records from National Taiwan University Main Hospital and Yun-Lin Branch. Individuals with CP class A cirrhosis, age above 20 years and endoscopically confirmed acute variceal bleeding were enrolled into the study. The exclusion criteria were pregnancies, vital organ failures, prior cerebrovascular accidents, non-liver related malignancies, active antiplatelet or anticoagulant agents use, antibiotic exposure within two weeks before bleeding episode and confirmed infection before endoscopic study. CP class A cirrhotic patients with more poor prognostic factors (ascites and frequent hepatic encephalopathy at baseline condition) were also excluded. Results Our study totally recruited 884 CP class A cirrhotic patients with acute variceal hemorrhage. Among these patients, there were 511 cases with antibiotic prophylaxis while 373 without. The overall five-day rebleeding rate was 1%, six-week rebleeding rate was 13% and infection rate was 13%. After logistic regression analysis of relevant variables, antimicrobial prophylaxis had significant association with endoscopic treatments (OR: 3.7264, 95% CI: 2.5079 to 5.5368), PPI use (OR: 1.6893, 95% CI: 1.2378 to 2.3055), hepatocellular carcinoma (OR: 1.3802, 95% CI: 1.0078 to 1.8902), heart rate (OR: 1.0120, 95% CI: 1.0046 to 1.0194), Hematocrit (OR: 0.9638, 95% CI: 0.9418 to 0.9862) and albumin (OR: 0.5291, 95% CI: 0.3720 to 0.7526). After propensity score matching, there were total 466 patients with 233 in each group enrolled comparison. The analysis showed that the rates of five-day rebleeding, six-week rebleeding and infection showed no significant difference between patients with antimicrobial prophylaxis and those without. Discussion For many years, antibiotic prophylaxis is of confirmed efficacy to prevent infection and rebleeding among cirrhotic patients after acute variceal bleeding. However, recent reports have disclosed that prolonged antibiotic exposures lead to serious complications in cirrhosis. Under consideration of benefits and disadvantages of antimicrobial therapy, a stratified-risk guidance of antibiotics use is suggested when treating variceal hemorrhage. Regardless of several evidence demonstrated in decompensated cirrhosis, there is only limited data in discussion about CP class A group. In our study, we focused on CP class A cirrhosis and explored the association between antibiotic agents and clinical factors relevant to variceal bleeding. The results showed that antibiotics use were significantly associated with endoscopic treatment, PPI use, hepatic malignancies, heart rate, hematocrit and serum album. These findings are more illustrated in clinical scenarios that physicians are tend to administer antibiotics when facing endoscopically treated bleeders, malignant comorbidities, lower liver reserve and clinical suspicion of massive hemorrhages. Furthermore, we performed propensity score matching and the analysis showed that there was no difference in five-day, six-week rebleeding or infection showed between two matched groups. The results indicated that antibiotic prophylaxis provided no obvious preventive effects for CP class A cirrhotic patients during variceal bleeding. Conclusion For acute variceal hemorrhage, antibiotic prophylaxis shows no significant effects on prevention of rebleeding or infection in CP class A cirrhosis.