Bleeding time does not predict gastrointestinal bleeding in patients with cirrhosis

Abstract

Background/Aims: Bleeding time, a laboratory test which explores primary hemostasis, may be prolonged in cirrhosis, but whether abnormal bleeding time identifies patients with cirrhosis who are at risk of bleeding has never been investigated. The aim of this study was to analyze the relationship between bleeding time and the risk of gastrointestinal bleeding. Methods: Eighty consecutive patients with liver cirrhosis (47 males, 33 females; age, 60±9 years; range 31 to 83 years) and esophageal varices were enrolled in the study. Results: In the whole series of patients bleeding time was 11±6 min; it increased as the degree of liver deficiency increased, from low to severe ( p=0.007). During 14±9 (median (range): 12 (1–34)) months of follow-up, 28 (35%) patients experienced gastrointestinal bleeding. They had a longer bleeding time, higher incidence of previous bleeding, more severe liver failure and larger variceal size than patients who did not bleed. However, multivariate analysis (Cox's model) showed that only previous bleeding, liver failure and variceal size were independently associated with bleeding. Similar data were obtained in patients with moderate-severe liver insufficiency (B and C degree according to Child-Pugh's classification). In patients who had never bled ( n=54), the severity of liver failure and variceal size were independent predictors of bleeding. Conclusions: This study shows that bleeding time is not a predictor of gastrointestinal bleeding in patients with cirrhosis.