Angiographic study in patient presenting with third variceal bleed following transjugular intrahepatic portosystemic shunting. Large varices remain despite placement of multiple coils (open arrow). Laminated thrombus is visible within lumen of stent. Distal splenorenal shunting was performed subsequently (solid arrow).
Survival curve after distal splenorenal shunting. Patients were observed for a mean of 6.5 years and censored according to Kaplan-Meier methods. The 1- and 5-year survival rates were 86.0% and 74.2%, respectively.
Jenkins RL, Gedaly R, Pomposelli JJ, Pomfret EA, Gordon F, Lewis WD. Distal Splenorenal ShuntRole, Indications, and Utility in the Era of Liver Transplantation. Arch Surg. 1999;134(4):416-420. doi:10.1001/archsurg.134.4.416
The distal splenorenal shunt (DSRS) continues to play an important role in the management of recurrent variceal bleeding with minimal negative impact on subsequent orthotopic liver transplantation (OLT).
Hepatobiliary surgery and liver transplantation unit in a tertiary referral medical center.
From August 1, 1985, through October 31, 1997, a single team of surgeons performed 81 DSRS procedures for recurrent variceal hemorrhage. Eleven patients undergoing OLT subsequent to DSRS were compared with a group of 274 patients undergoing OLT without any previous shunt during the same period.
Main Outcome Measures
Operative time, use of blood products, length of hospital stay, perioperative complications, and survival rates.
Operative (30-day) mortality for DSRS was 6% (n=5). From follow-up information available for 74 patients, the 1- and 5-year survival rates were 86.4% (n=64) and 74.3% (n=55), respectively. Recurrent variceal bleeding and hepatic encephalopathy occurred in 5 (6.8%) and 11 patients (14.9%), respectively, after DSRS. In 9 patients, DSRS was used as salvage for failed transjugular intrahepatic portosystemic shunt.
Distal splenorenal shunt is a safe, durable, and effective treatment for controlling recurrent variceal hemorrhage in patients with acceptable operative risk and good liver function. It does not compromise future liver transplantation and can considerably delay the time until transplantation is required. Given the early occlusion rate and need for constant surveillance, transjugular intrahepatic portosystemic shunting should be reserved for patients with Child C classification cirrhosis with chronic hemorrhage or intractable ascites or as an emergency procedure for patients with uncontrollable bleeding using endoscopic therapy.
TREATMENT OPTIONS for the patient with variceal bleeding and portal hypertension have changed dramatically during the past decade. Historically, medical therapy, balloon tamponade, and emergency portal decompression procedures were the mainstay of treatment for the patient with refractory variceal bleeding but were associated with significant morbidity and mortality.1- 4 Refinements in endoscopic sclerotherapy and pharmacological intervention led to an increasing number of patients successfully treated without the need for emergency surgical shunt procedures.5,6
More recently, the development of transjugular intrahepatic portosystemic shunting (TIPS) has added substantially to the armamentarium of therapeutic options for patients with complications of chronic liver disease. Transjugular intrahepatic portosystemic shunting has demonstrated efficacy as an emergency modality for the patient with refractory variceal bleeding, or as the primary therapy for the patient in whom operative risk is considered poor. The use of TIPS is limited, however, by the need for constant surveillance, frequent revisions, and a high rate of early thrombosis.7,8
For the patient with relative preservation of hepatic function and minimal ascites, selective portosystemic shunting still should be considered for the management of variceal bleeding when pharmacological therapy and sclerotherapy have failed.9,10 Although orthotopic liver transplantation (OLT) remains the definitive treatment for the patient with liver failure, many patients with portal hypertension and esophageal bleeding may not require OLT for many years.11
Each year the number of patients waiting for liver transplants increases, yet the number of available organs from suitable donors remains relatively unchanged.12 This places a premium on the development of strategies for treating patients with portal hypertension to delay the need for transplantation. The goal should be to maintain the functional reserve of the remaining liver while minimizing recurrent bleeding without compromising the ability to perform a remedial transplantation. The distal splenorenal shunt (DSRS) procedure developed by Warren et al13 in 1967 fulfills these criteria.
In this study, we review a single center's experience in performing DSRS during a 12-year period in a single institution with an active liver transplantation program. We explore the utility of DSRS in the management of recurrent variceal bleeding, outcome analysis, and impact on subsequent OLT.
From August 1, 1985, through October 31, 1997, DSRS was performed in 81 patients in the liver transplantation unit at the Beth Israel Deaconess Medical Center, Boston, Mass. Patient medical records were reviewed for demographics, cause of the liver disease, Child classification, indication for shunting, intraoperative blood loss, transfusion requirement, perioperative surgical complications, length of hospitalization, postoperative encephalopathy, and rebleeding rates. A group of 11 patients undergoing OLT subsequent to DSRS (DSRS was performed at the study center in 6 patients and at referring hospitals in 5) was compared with a group of 274 patients who underwent OLT only during the same interval. Thirty-day operative mortality and long-term survival rates were evaluated.
The primary indication for DSRS was recurrent variceal bleeding. In 9 patients, however, DSRS was used for salvage after TIPS and subsequent development of acute occlusion and rebleeding (Figure 1).
Seven of 81 patients were unavailable for follow-up and were excluded from statistical analysis. Of the 74 patients in whom complete data were available, Child classification was A in 57 (77%) and B in 17 (23%) at the time of surgery. Endoscopy showed esophageal varices alone in 52 patients (70%) and esophageal and gastric varices in 22 (30%). Primary sclerotherapy was performed in all patients to control initial bleeding episodes. In cases of medical failure with recurrent hemorrhage refractory to sclerotherapy, TIPS or DSRS was considered. Distal splenorenal shunting was performed only in patients with Child A or B classification. Patients were selected for transplantation based on the development of end-stage liver failure in accordance to guidelines promulgated by the United Network for Organ Sharing.14
Unless otherwise indicated, all data are presented as mean±SEM. Continuous data were compared using Student t test. Thirty-day survival data were compared using χ2 test. Statistical significance was set at P<.05 in all cases. Survival data were calculated using the Kaplan-Meier product-limit method.
Fifty-two men (64%) and 29 women (36%) had a mean age of 56.9±1.2 years (range, 35-88 years) at the time of surgery. The primary causes of portal hypertension and variceal bleeding are summarized in the following tabulation:
Hepatitis C disease assumed increasing importance in later years and undoubtedly contributed to cirrhosis in patients with cryptogenic or alcoholic liver disease before serologic markers became routinely available.
Hospital course is summarized in the following tabulation:
Perioperative and long-term complications are summarized in Table 1. Most complications developed during the perioperative phase and required observation or short-term treatment. Operative 30-day mortality was 6% (n=5). Child classification in 3 of these patients was A; in 2, B. Causes of death included adult respiratory distress syndrome and multiple organ failure (n=2), coagulopathy and uncontrolled bleeding (n=2), and cardiac arrhythmia (n=1). Of particular note is the finding that thrombosis of the portal vein developed within the first week of surgery in 5 patients, contributing to delay in resolution of postoperative ascites in all. Three of these 5 patients have subsequently required OLT. One of these patients required an iliac vein graft to reconstruct the portal vein, whereas the other 2 underwent thrombectomy and standard anastomosis.
Long-term complications included recurrent variceal bleeding and hepatic encephalopathy, which occurred in 6.8% (n=5) and 14.9% (n=11) of patients, respectively. These data reflect occurrences throughout a mean survival time of 6.5 years. Figure 2 shows the Kaplan-Meier survival curve for all patients who underwent successful DSRS. The 1- and 5-year survival rates were 86.4% (n=64) and 74.3% (n=55), respectively.
During the study, 11 patients underwent OLT a median of 5.1 years following DSRS (range, 0.4-10.1 years). We compared these patients with 274 patients who underwent OLT during the same period without any preexisting shunt. The relative impact of previous DSRS on operative outcome of OLT is shown in Table 2. There was no significant statistical impact on duration of surgery, use of blood products, or 30-day operative mortality. The DSRS was ligated after hepatic revascularization in 5 patients, and was left intact in the remaining 6.
Management of portal hypertension and variceal bleeding is complicated by the variable degree of hepatic synthetic disruption by a diverse group of chronic liver diseases and the number of treatment options available. For the patient with recurrent variceal hemorrhage but adequate hepatic function, controversy exists as to the best method of prophylaxis against future bleeding.1,15,16 In patients with unsuccessful pharmacological and endoscopic control of variceal bleeding, TIPS has evolved recently as a relatively convenient treatment modality, and in many institutions its use is not restricted to patients with poor hepatic reserve. Emerging data suggest the frequent need for TIPS revision for stenosis or thrombosis within the first 1 to 2 years (20%-50%) and the potential for life-threatening complications.7,17- 22 These observations should temper enthusiasm for the use of TIPS in good-risk patients who have the potential for long-term survival once portal hypertension is controlled. Properly used, TIPS should play its pivotal role in the treatment of patients with recurrent variceal hemorrhage who are not candidates for transplantation, in bleeding patients desperately awaiting liver replacement, and in selected patients with intractable ascites.23
Whereas liver transplantation offers the best long-term solution to portal hypertension, end-stage liver failure may not develop for many years in many patients after their first bleeding episode. Therefore, any treatment option to control variceal hemorrhage should reduce the incidence of bleeding, maintain hepatic reserve, and remain durable for years.
There are several advantages to a selective DSRS over other selective and nonselective shunts in controlling variceal bleeding. First, selective decompression of gastroesophageal varices via the splenic circulation maintains prograde portal flow into the liver while effectively decompressing esophageal and gastric varices.24 Maintenance of portal flow may better preserve the delivery of hepatotrophic factors that, in turn, are responsible for preservation of hepatic architecture and function.25 Second, DSRS avoids extensive hepatic hilar mobilization, which might significantly complicate any future transplantation effort. Some caution in the use of DSRS is suggested by the low incidence of postoperative portal vein thrombosis, which occurred in 5 of our patients. Although recanalization of the portal vein may occur during several months, complete or partial thrombosis of the portal vein adds complexity to a subsequent liver transplantation.26,27
Our study confirms that DSRS can be performed in select patients with low operative mortality and a 6.8% rate of rebleeding during a mean follow-up of 6.5 years. Encephalopathy, which occurs frequently after nonselective shunts such as portocaval shunts and TIPS, occurred in only 14.9% of our patients during follow-up and was readily managed by dietary restriction and lactulose.28,29 In patients with DSRS undergoing later transplantation, adhesions may increase the technical difficulties of recipient hepatectomy, but operative times, use of blood products, and survival rates are equivalent to those of the population undergoing liver transplantation as the primary treatment.
Our present algorithm for the management of variceal hemorrhage uses most of the available treatment options, with an emphasis on the immediate identification of patients who are anticipated to need urgent or late transplantation. Octreotide infusion and endoscopic sclerotherapy or banding is used initially in all patients presenting with variceal bleeding. Patients with Child A classification and in whom endoscopic variceal control fails are offered DSRS surgery in anticipation that they will not need transplantation for a prolonged period. Attempts are made to optimize patients with Child B classification who may then also be candidates for DSRS. Patients with Child C classification who are poor candidates for transplantation and patients actively waiting on the transplant list with uncontrollable bleeding are offered TIPS. Surgical portocaval shunts are restricted to rare patients in whom TIPS or DSRS fails or patients with salvageable Budd-Chiari syndrome.30
Given the long waiting times on many transplant lists, evaluation for potential liver transplantation should be instituted early in the patient's clinical course. Patients with chronic liver failure who are referred near the end of their disease process often will not survive the waiting period before a donor organ becomes available.
Distal splenorenal shunting continues to be a safe and effective procedure for controlling recurrent variceal hemorrhage and should be the treatment of choice for patients with acceptable operative risk and good liver function. It also serves as an excellent salvage procedure for patients with favorable Child classification in whom injudiciously placed TIPS fail. Furthermore, DSRS is a durable bridge to transplantation, with a greater than 5-year interval between surgical decompression and OLT in our series.
Presented at the 79th Annual Meeting of the New England Surgical Society, Toronto, Ontario, September 25, 1998.
Corresponding author: Roger L. Jenkins, MD, Division of Hepatobiliary and Transplantation Surgery, Beth Israel Deaconess Medical Center, 110 Francis St, Suite 8C, Boston, MA 02215 (e-mail: firstname.lastname@example.org).