ABSTRACT
BACKGROUND: Gastrointestinal bleedings (GIBs) are frequent in cirrhotic patients and lead to high morbidity and mortality. Lately, there have been conflicting reports on the role of and bleeding type [variceal bleeding and nonvariceal bleeding (NVB)]. This study investigated the predictors of mortality in patients with variceal bleeding and NVB with relationship to sex differences. MATERIALS AND METHODS: A total of 271 patients with suspected upper GIB who underwent endoscopy were included. Patients were followed up at 1 week, 6 months and 1 year after admission. Univariate and multivariate logistic or Cox regression analyses investigated correlations of predictive factors and clinical outcomes. Propensity score matching was performed to control for severity of disease and compare groups for sex and bleeding type. RESULTS: A total of 42 patients were excluded (cirrhosis or bleeding not confirmed). The remaining patients were classified by bleeding type into patients with variceal bleeding (n = 115) or NVB (n = 156). Males (n = 155) had higher mortality in variceal bleeding than in NVB, while in females (n = 116) mortality was similar in the two bleeding types. This was confirmed after matching in males (n = 116) and females (n = 82). Further independent predictors of mortality in males were model for end-stage liver disease (MELD) at baseline, blood urea nitrogen, alanine aminotransferase, while in females age, leukocytes, MELD, history of ascites and hepatic encephalopathy. CONCLUSION: This study shows that variceal bleeding has higher mortality in males compared to NVB, while in females the type of GIB does not impact the outcome. This highlights that sex-specific clinical management should be based on bleeding type after endoscopy.
