Non-endoscopic Prediction of the Presence of Esophageal Varices in Cirrhotic Patients at Gabriel Touré University Hospital

Abstract

Introduction: Digestive hemorrhage due to ruptured esophageal varices (EV) is a serious progressive complication. The diagnosis of these EVs is identified through upper digestive endoscopy, which is an invasive procedure. Objective: The aim of this study was to determine non-endoscopic factors predictive of the presence of EVs in patients with chronic liver disease in a hospital setting. Methods: This was a cross-sectional study conducted in the Hepato-Gastroenterology Department of Gabriel Touré University Hospital, from July 2017 to July 2019. The study focused on patients with chronic liver disease, who were compared according to the presence or absence of VO. All patients underwent upper digestive endoscopy and also benefited from non-invasive examinations. The Chi-square test was used to compare our results, which were considered significant for a probability of p<0.05. Results: We collected 54 cases of chronic liver disease with the presence of varices, including 35 men and 19 women, with a mean age of 48.7 ± 14.4 years, which were compared to 54 other cases of liver disease without varices, with a mean age of 42.6 years. Hematemesis was the reason for consultation in 33.3% of patients with varices (p = 3×10^-5); liver disease was incidentally discovered in 22.3% of cases without varices (p = 2×10^-3). Clinically, the presentation in the presence of varices was dominated by low blood pressure (p = 0.0021), ascites (p = 0.018), hematemesis (p = 10^-7), melena (p = 10^-7), and anemia (p = 0.0002); in the absence of varices, hepatomegaly (p = 0.6830), ascites (p = 0.0018), and jaundice (p = 0.6674) were most frequently reported. Biologically, 83.3% of patients with varices had anemia (p = 0.00005). Cytolysis (51%) and anemia (46.3%) were present in the absence of variceal bleeding (VO) with respective p-values of 0.0275 and 0.00005. The severity of liver disease, classified as Child-Pugh B and C, was predictive of the presence of VO at 37% and 24.1%, respectively. On abdominal ultrasound, ascites (77.8%), a heterogeneous liver (66.7%), and a dilated portal trunk (33.3%) were significantly associated with the presence of VO. Anemia associated with ascites (PPV=72.5%) or with a central venous catheter (CVC) (PPV=100%) was predictive of VO. An atrophic liver and a dilated portal trunk associated with a CVC strongly predicted the presence of VO (PPV=100%). Any hepatic abnormality identified on abdominal ultrasound, when associated with anemia and a CVC, had a positive predictive value (PPV) of 100% for the presence of VO. Conclusion: Since upper gastrointestinal endoscopy is an invasive procedure, it would be valuable to use non-invasive methods to predict the presence of esophageal varices. Our results could serve as a useful tool to limit the indication for gastroscopy to only those patients who are truly likely to have esophageal varices.