From the AGA Journals

AGA Clinical Practice Update: Management of bleeding gastric varices



When classifying gastric varices during endoscopy, experts suggest not only describing their location but also their size and whether any high-risk stigmata, such as discolorations and platelet plugs, are present.

In a clinical practice update from the American Gastroenterological Association, Zachary Henry, MD, of the University of Virginia, Charlottesville, and associates also proposed an alternative nomenclature for locating gastric varices (GV). “In practice, most gastroenterologists use the Sarin classification with the main distinction being cardiofundal versus lesser curvature GV. However, the vascular supply and corresponding therapy for GV and esophageal varices are often different, so a merged classification, such as Sarin’s, can be problematic for therapeutic planning purposes,” they wrote in Clinical Gastroenterology and Hepatology, referring to the classification system published by Shiv K. Sarin, MD, DM, and colleagues. They suggested that a merged classification, such as Sarin’s, can be “problematic for therapeutic and planning purposes” because “the vascular supply and corresponding therapy for GV and [esophageal varices] are often different.” Instead, they advised that an “alternative nomenclature based on location within the stomach is clearer and facilitates correlation with vascular imaging.” Another approach is to add risk factors for bleeding, such as an estimate of variceal size and high-risk stigmata (discolored marks, platelet plugs), to Sarin classification.

Diagnosis and treatment of bleeding GV are complex, and multidisciplinary management by hepatologists, interventional radiologists, and interventional endoscopists is optimal, the experts wrote. Data and clinical guidelines do not support primary prophylaxis to prevent bleeding of GV. The authors offered an algorithm for initial management of suspected portal hypertensive GV bleeding based on both endoscopic and vascular anatomy; it includes assessment of circulatory and respiratory status, vasoactive drug administration, antibiotic prophylaxis, and more.

An early goal is confirming bleeding source and attempting to classify the bleeding site; this can be complicated by presence of intragastric blood that obscures the cardia and fundus and underlying GV. Further steps may include temporizing: “Temporizing measures to halt active bleeding are often not the definitive treatment of choice to prevent rebleeding from GV, whereas definitive measures such as endoscopic cyanoacrylate injection (ECI) or endovascular treatments are often not feasible in the acute, diagnostic setting.”

When definitive endoscopic treatment is preferred, ECI of bleeding GV is the therapy of choice because other approaches may be complicated by location and bleeding risk of GV, although band ligation may be appropriate in lesser curve GV. Specific ECI techniques have not been compared directly in studies, according to the update authors; however, “the specific cyanoacrylate agent should favor the fastest polymerization time to avoid embolization and inducing GV bleeding.” This has meant 4-carbon (butyl) preparations are preferred to 8-carbon (octyl) preparations, they noted.

After treatment, endoscopy is performed every 2-4 weeks so that the ECI can be repeated as needed until obliteration is complete. The experts suggested that, after eradication of GV, an endoscopic reevaluation within 3-6 months should be scheduled, then annually thereafter. Any de novo or recurrent GV during the long-term follow-up may require additional imaging and multidisciplinary exploration to determine potential mechanisms and need for alternative treatments, the authors advised.

According to the practice update, transjugular intrahepatic portosystemic shunt can be used when the GV is receiving significant inflow from the coronary vein or the patient has significant complications from portal hypertension. When TIPS is used, the experts suggest also performing endovascular sclerosis or direct embolization of GV, if possible. For patients with a gastrorenal shunt, balloon-occluded retrograde transvenous obliteration (BRTO) of bleeding GV is considered optimal if local expertise is available and the patient lacks severe complications from portal hypertension. Endoscopy should be performed within 48 hours after BRTO to confirm obliteration of the vascular flow. If residual flow is detected, “cyanoacrylate injection should be performed,” the experts wrote. To confirm that GV are obliterated and check for any vascular complications, they suggest performing CT or MR within 4-6 weeks after BRTO and then as clinically indicated. In addition, surveillance endoscopy is important to identify and treat any esophageal varices that could have been worsened by increased portal pressures.

No funding sources were reported. The experts reported having no conflicts of interest.

Next Article: