Extrahepatic portal venous obstruction: Is the knife irrelevant?

Abstract

A study on long term outcomes following variceal eradication in patients with extrahepatic portal venous obstruction (EHPVO) was clearly needed. In this issue of the Journal, Thomas et al. present the results of long-term follow up of a cohort of patients treated with endoscopic variceal sclerotherapy (EST). Of the 223 patients treated between 1985 and 1994, they had follow up data available for 198 patients. On a retrospective analysis of these data, 34 (17.2%) patients had rebleeding after initial variceal eradication, 39 (19.7%) had reappearance of esophageal varices and 19 (9.5%) had de novo appearance of fundal varices. The mean duration from the time of eradication of varices to rebleeding was 5.4 (range 1 to 16) years. Based on these results, the authors suggest that endoscopic variceal eradication is an effective treatment modality for bleeding esophageal varices in patients with EHPVO. This study, though retrospective, is quite thorough. 1 It clarifies several questions, while throwing up some new ones. First, the authors base their premise on a cohort of patients in whom they had achieved successful initial hemostasis and long term variceal eradication. A large majority (87%) of their patients presented with active variceal bleeding. However, we remain unsure of the outcome in patients in whom initial EST failed to achieve hemostasis. In 10%–15% of patients, endoscopic treatment may not achieve initial control of bleeding. 2,3 This subset of patients is important since their exclusion may mislead us into believing that EST is successful in all patients with variceal bleeding due to EHPVO. Secondly, 12 patients were lost to follow up from the initial cohort of 223 patients; did these patients die or undergo some other form of treatment elsewhere? No doubt the long term results are important, but without these initial data, establishing the overall efficacy of EST in this group of patients would be incomplete. In the broader clinical perspective, the treatment options available for patients with EHPVO must include surgery. Evidence in the Western literature indicates that if shunt surgery has a role in treatment of portal hypertension, it is in patients with non-cirrhotic portal hypertension, and especially those with EHPVO. 4,5 From India too, excellent long term results of proximal lieno-renal shunt (PLRS) have been reported in such patients. Of 160 children who underwent PLRS with splenectomy at our institution, the overall mortality was 1.9% (elective: 1/140 [0.7%]; emergency 2/20 [10%]), rebleeding rate was 11% and actuarial survival at 15 years by life-table analysis was 95%. 6,7 None of these patients developed encephalopathy. The procedure cured hypersplenism in all the patients who had this. Only one patient developed meningococcal meningitis, which recovered following treatment. Similarly, a study from Chandigarh that included 104 patients with non-cirrhotic portal hypertension who had undergone side-to-side lieno-renal shunt showed a shunt patency rate of 87%, a rebleeding rate of 10%, and no occurrence of encephalopathy after a mean follow up of 54 months. 8 Thus, surgery is a one-time treatment procedure with durable, long term efficacy in preventing variceal rebleeding. Long term studies (15–40 years) from the West have repeatedly shown that shunt surgery with or without splenectomy (proximal lieno-renal, side-to-side splenorenal, distal splenorenal or mesocaval) for EHPVO in children and adults, is associated with shunt patency rates between 90% and 95%, rebleeding rates of 5%–10%, no encephalopathy and 15-year actual survival rates of 95%. 5,9–11 These studies have found portosystemic shunts to be consistently effective for bleeding esophagogastric varices due to EHPVO.