Injection Therapy of Esophageal and Gastric Varices
Edward Villa, MD
Uzma D. Siddiqui, MD
Gastrointestinal (GI) variceal bleeding is the most common lethal complication of cirrhosis and accounts for 10% to 30% of all upper GI bleeding. Varices develop at a rate of roughly 8% per year in cirrhotic patients and are present in as many as 50% of cirrhotic patients (5% to 33% in the case of gastric varices).1 Variceal hemorrhage (VH) occurs at a yearly rate of 5% to 15% (25% in 2 years for gastric varices). Mortality at 6 weeks from first esophageal VH is 20% at 6 weeks, and late rebleeding occurs in about 60% of untreated patients within 1 to 2 years of index hemorrhage.1
SCLEROSIS
Sclerosants are tissue irritants that cause vascular thrombosis and endothelial damage, which lead to endofibrosis and vascular obliteration. The sclerosants most commonly used that are available in the United States are fatty acid derivatives (such as ethanolamine oleate, 5% [Ethamolin, QOL Medical, Inc, Woodinville, Sash] and sodium morrhuate, 5% [Scleromate, Glenwood LLC, Englewood, NJ]) or synthetic agents (sodium tetradecyl sulfate, 1% and 3% [Sotradecol, Bioniche Life Sciences, LLC, Belleville, Ontario, Canada; Trombovein, Omega Pharmaceuticals Ltd., Montreal, Quebec, Canada; and Fibro-vein, STD Pharmaceutical, Hereford, England]).2
Sclerotherapy has historically been used for bleeding and nonbleeding esophageal varices and for gastric varices extending into the lesser curvature (type 1 gastroesophageal varices [GOV-1] by Sarin Classification) but has largely been supplanted by variceal ligation (VL). Studies have demonstrated that sclerotherapy results in higher mortality compared to sham therapy for primary prophylaxis of esophageal VH.3,4 There is a 6-month survival of 84% with serial sclerotherapy when combined with vasoactive medications in control esophageal VH, but VL still has been shown to be superior to sclerotherapy with regard to outcomes of
rebleeding, variceal eradication, and the rate of complications.5,6,7 Adverse events include fever, infections, retrosternal discomfort/pain, dysphagia, injection-induced bleeding, esophageal ulceration with delayed bleeding, esophageal strictures, perforation, mediastinitis, pleural effusion, and bronchoesophageal fistulae.8
rebleeding, variceal eradication, and the rate of complications.5,6,7 Adverse events include fever, infections, retrosternal discomfort/pain, dysphagia, injection-induced bleeding, esophageal ulceration with delayed bleeding, esophageal strictures, perforation, mediastinitis, pleural effusion, and bronchoesophageal fistulae.8
CYANOACRYLATE INJECTION
Injection with cyanoacrylate variants such as N-butyl-2-cyanoacrylate (cyanoacrylate, enbucrilate, or Histoacryl [B. Braun, Melsungen, Germany]) and 2-octyl cyanoacrylate (Dermabond [Ethicon, Somerville, NJ]) have been utilized in an off-label use in the United States as a successful therapeutic modality for treatment of patients with bleeding type 2 gastroesophageal varices (GOV-2) and isolated fundal varices (IGV-1). Cyanoacrylate (CA) is a monomer in liquid form that upon contact with hydroxyl ions in water or blood rapidly polymerizes into a hard plastic or glue, which results in hemostasis of VH. CA can be injected in combination with a contrast agent (Lipiodol or Ethiodol) for prevention of polymerization within the injection needle and for X-ray visualization.9
Injection of CA has resulted in resolution of gastric VH in 87% to 100%; reduction in rebleeding in IGV-1’s (85.7%); and eradication of varices in 67.8% to 100%.10,11,12 Recurrent bleeding occurred in 23.3% to 35.2% of patients. Among these, 16.7% were successfully retreated with repeat injection, and there was a treatment failure-related mortality rate of 2.2% to 23.7%.10,11,12 CA injection is superior to VL with regard to hemostasis and rebleeding rates in GOV-1 and IGV-1 hemorrhage.11,12
Most studies quote a 0.5% to 5% adverse event rate.12,13,14 Adverse events include fever, rebleeding, ulceration at the injection site, hemothorax, thromboembolic events (pulmonary emboli, stroke, myocardial infarction, other systemic arterial embolizations, portal and mesenteric venous thrombosis, etc.), fistulization, spontaneous bacterial peritonitis, bacteremia, and pneumonia. The use of endoscopic ultrasound (EUS) and audible doppler solely to assess gastric variceal obturation does not appear to significantly affect adverse events or clinical outcomes.15,16 However, the use of EUS-guided coil deployment performed prior to CA injection into the afferent feeding veins of large fundal gastric varices achieved complete obliteration of the perforating veins in fewer sessions than standard CA injection while resulting in fewer adverse events.17,18,19,20 Preprocedural assessment of the portovenous system should be considered to identify patients with spontaneous portosystemic shunt or patent foramen ovale, and if either is discovered, alternate treatment measures should be considered.14