Embolization of spontaneous splenorenal shunt for after‐TIPS hepatic encephalopathy in a patient with cirrhosis and variceal bleeding

Abstract

Potential conflict of interest: Nothing to report. Transjugular intrahepatic portosystemic shunt (TIPS) is a well‐established treatment for complications arising from portal hypertension.1 However, hepatic encephalopathy (HE) is a common complication after TIPS insertion. The incidence rate of HE after TIPS insertion is 15%‐48%.2 TIPS implantation with or without gastroesophageal variceal embolization remains questionable in general,2 but it is particularly questionable when TIPS is used in patients with splenorenal shunt (SRS). SRS is a spontaneous portosystemic shunt (SPSS) that persists in 14%‐21% of patients with cirrhosis.4 A previous study suggested that SPSS embolization alone is an effective and safe treatment for chronic HE.5 However, combining TIPS implantation with SPSS embolization for patients who present with variceal bleeding (VB) remains an issue. We report on a case of improved post‐TIPS HE after embolizing the coexisting SRS in a patient with cirrhosis with VB. Case Report A 34‐year‐old man who had experienced repeated melena and hematemesis three times (total, 600 mL) for 13 days with post–hepatitis B cirrhosis required hospitalization on March 16, 2012. The Child‐Pugh score was B/8 (total bilirubin level: 0.92 mg/dL; serum albumin: 3.17 g/dL; mild ascites, no encephalopathy), and the hemoglobin level was 9.2 g/dL. An upper endoscopy revealed significant gastroesophageal varices (GEV), and a large SRS (11 mm in diameter) was observed by angiography (Fig. 1A). Hematemesis and melena disappeared after pharmacological treatment. To further decrease the portal pressure and prevent variceal rebleeding, we implanted a TIPS using an 8‐mm Bard polytetrafluoroethylene‐covered stent. The portosystemic pressure gradient decreased from 36 to 15 mmHg. Diagnostic endoscopy showed improved GEV 2 days later, and ascites disappeared a few days after TIPS implantation. Given the lack of evidence in the literature regarding the role of embolotherapy during the TIPS procedure, SRS was not embolized at that time (Fig. 1B).Figure 1: Portal venogram of a patient with a spontaneous splenorenal shunt undergoing a TIPS procedure with subsequent embolization in the SRS. (A) A large spontaneous SRS (black arrow) is shown. (B) The SRS (black arrow) was not embolized after the addition of the TIPS (white arrow). (C) The former TIPS (white arrow) and a coexisting spontaneous SRS (black arrow) are shown before embolization. (D) SRS was embolized using an Amplatzer plugging device (black arrow) through the femoral vein, and the patent TIPS is shown (white arrow).The patient developed lethargy and disorientation 5 days after TIPS insertion, and a diagnosis of grade II HE was made. HE occurred repeatedly in the following days, despite a continuous treatment of branched‐chain amino acids and lactulose. To treat the post‐TIPS encephalopathy effectively, we embolized the SRS using an Amplatzer plugging device (18 mm in diameter) from the femoral vein through the left renal vein to the SRS 37 days after TIPS placement (Fig. 1C,D). After embolization, the HE symptoms gradually disappeared in approximately 6 hours. The Child‐Pugh score changed to A/5 (total bilirubin level: 1.15 mg/dL; serum albumin: 3.95 g/dL), and the hemoglobin level was 9.3 g/dL 1 year later. There was no recurrence of HE or any significant increase in the aggravation of preexisting varices at the 2‐year follow‐up. Discussion Recently, SPSS has received increasing attention because it appears in 46%‐70% of patients with refractory HE.5 In TIPS patients, the large SRS diverts portal flow, resulting in the diversion of portal venous flow into the left renal vein, which decreases both blood flow within the stent and hepatopetal perfusion.3 Consequently, the SRS may increase the incidence of shunt dysfunction and liver dysfunction. Moreover, the existence of two shunts induces excessive portosystemic shunts, thereby increasing the incidence of HE episodes. We have established that TIPS implantation decreases the portal pressure for the prophylaxis of rebleeding and that SRS embolization anatomically forces the splenic venous flow into the main portal vein, which increases the previously reduced blood flow. Therefore, this method may improve liver function and maintain stent patency. Furthermore, SRS embolization shifts the splenic venous flow from the extrahepatic shunt into the TIPS, which allows for only one intrahepatic shunt, thereby effectively lowering the risk of rebleeding and HE episodes. Currently, there is no clear therapeutic strategy regarding the management of SPSS patients in liver cirrhosis and no definite consensus on the treatment of post‐TIPS HE. This case demonstrated that the combined therapy may be superior to TIPS alone owing to the mitigation of post‐TIPS HE. TIPS insertion combined with SPSS embolization may be an alternative treatment in the management of such patients. Prospective, randomized trials are warranted to further confirm these findings. Authors names in bold designate shared co‐first authorship.