Background: Gastric varices are probably associated with thrombocytopenia. However, the prevalence and severity of thrombocytopenia are unknown in this clinical setting.
Objective: To assess platelet count and doppler ultrasound on hepatic and splenic venous system as non-invasive predictors of gastric varices in patients with liver cirrhosis with no history of previous endoscopic or surgical intervention for portal hypertension.
Patients and Methods: An observational case control study was accomplished at Al-Azhar University Hospitals, Hepatogastroenterology and infectious diseases Departments, at Cairo from September 2018 to December 2019. One hundred and twenty patients with liver cirrhosis without past history of previous endoscopic or surgical intervention for portal hypertension were selected, and divided into three equal main groups:Group A: Cirrhotic patients without varices, Group B: Cirrhotic patients with esophageal varices only, and Group C: Cirrhotic patients with gastric varices, which divided into two subgroups: Group Ca:Six patients with isolated gastric varices and Group Cb:Thirty four patients with gastroesophageal varices. All patients were subjected to upper gastrointestinal endoscopy,complete    blood   picture (CBC) and abdominal Doppler ultrasonography study on hepatic and splenic venous systems.
Results: Group Cb recorded the highest values of splenic size (P<0.001), ascites (p=0.006), portal vein diameter (P<0.001) and collaterals (P<0.001) compared to other groups. Group Ca recorded the lowest values of the mean portal vein flow velocity (MPVV)(p < 0.001), the mean splenic vein flow velocity(M.SVV) (P=0.026), and recorded the highest values of portal vein congestion index (P.CI) (P<0.001), portal vein thrombosis (PVT)(P<0.001), the mean values of the splenic vein diameter (P<0.001), cross sectional area of splenic vein (p < 0.001), splenic vein congestion index(S.CI) (p < 0.001), as well as abnormal blood direction in splenic vein(P=0.019) compared to other groups. Also, portal vein cross sectional area and the Child-Turcotte-Pugh score class “C" showed statistically significant difference (p < 0.001) between group A which recorded the lowest value compared to group B which recorded the highest value. Also, platelet count recorded no statistically significant difference between the four groups.
Conclusion: Doppler ultrasound can be an easy, cheap and safe predictor of gastric varices, while platelet count has no any significance in predicting of gastric varices.