Radiologic evaluation of portal steal phenomenon in recipients of liver transplantation

Abstract

It is important to maintain adequate portal flow and to prevent the detrimental effect of portosystemic shunt in recipients following liver transplantation. The purpose of this article is to present a comprehensive review of portosystemic shunts and to illustrate such phenomenon demonstrated on radiologic studies such as Doppler ultrasound, computed tomography, and portogram. It is important for radiologists to be aware of such phenomenon not only in preoperative evaluation of the recipients but also in postoperative screening to detect recurrence of the phenomenon.

Keywords

Liver transplantation Doppler ultrasound computed tomography (CT)

Introduction

Liver transplantation is generally accepted as a curative therapy for end-stage liver disease. Particularly after partial graft liver transplantation, maintaining robust portal perfusion is critical in boosting rapid graft regeneration (1). Occasionally, portosystemic shunts persist even after the transplantation, which may reduce the portal flow and thus threaten the patients’ outcome (2). Therefore, pre- and perioperative detection of portosystemic shunts requiring radiological or surgical interruption is essential for the liver transplantation candidates as well as recipients. In this article, we present a comprehensive review on the preoperative evaluation of portosystemic shunts in recipients of liver transplantation, the intraoperative image-guided interventional or surgical management of the shunt, and the radiologic evaluation for persistent or recurrent shunt following liver transplantation.

Portosystemic shunt

Definition and pathophysiology

In portal hypertensive liver cirrhosis, increment of hepatic sinusoidal resistance results in diversion of portal flow into systemic circulation via various collateral vessels. Consequently, normal hepatopetal portal flow is reduced or even reversed to hepatofugal direction depending on the amount of the shunted or “stolen” flow (3). This peculiar situation of declining portal inflow or bypassing via less resistive route is described as portal steal phenomenon or portosystemic shunt (4). Risk factors for growing collateral shunt vessels greater than 10 mm in caliber are portal vein thrombosis, lower portal venous pressure, and Child-Pugh class C, all sharing pathophysiology of portal hypertension (4).

In return for portal hypoperfusion, hepatic arterial flow reciprocally increases owing to physiologic hepatic arterial buffer response (5). Whether hepatic arterial buffer response is still activated in a cirrhotic liver has been investigated in many studies, and considerable evidence indicates that at least some degree of hepatic arterial buffer response is preserved in liver cirrhosis patients (6 –8). Hepatic arterial vasodilation enhances oxygen delivery to cirrhotic liver short of portal inflow and partially ameliorates worsening liver function (9).

Portosystemic shunt in recipients of liver transplantation

Especially after liver transplantation using partial graft, portosystemic shunts can negatively affect the condition of the graft by stealing portal flow vital for rapid liver regeneration. In the post-transplant period, maintaining adequate portal flow ensures efficient nutrients supply the regenerating graft demands and prevents portal vein thrombosis (1,10). Hence, a persistent portosystemic shunt deteriorates hepatic function and impairs graft regeneration which may even cause small-for-size graft failure after adult living donor liver transplantation (1). Clinical signs of portal steal phenomenon after liver transplantation include elevated aspartate aminotransferase and alanine aminotransferase, hyperbilirubinemia, hepatic encephalopathy, progressive splenomegaly, and thrombocytopenia (11 –14), reflecting graft dysfunction.

The incidence of portosystemic shunt/portal steal phenomenon after liver transplantation is in the range of 2–35%, within the limitation of varied patient demographics and the type of graft used (11,14 –16). According to Fujimoto et al., 22 (27%) of 82 pediatric patients showed diminished portal flow (<10 mL/min/kg) and two patients died of graft failure in part from the portal hypoperfusion after living donor liver transplantation (15). Kyoden et al. reported development of hepatofugal flow in 2% of 180 adult patients who underwent living donor liver transplantation (16), and De Carlis et al. observed 12 (35%) of 34 patients with persistent shunts and eventual serious ischemic damage to the graft in three (9%) patients after orthotopic liver transplantation (11). In addition, many case reports illuminated hepatofugal portal flow provoked by small-for-size graft, acute rejection, or congestion (17 –21).

Triggering factors for persistent portosystemic shunt after liver transplantation

Fortunately, after liver transplantation using adequate-sized grafts, normal hepatic resistance usually reduces portal pressure and clears up the portosystemic shunt with subsequent collapse of collateral vessels. However, the major pathways of portosystemic shunts such as splenorenal shunts may persist even after liver transplantation, especially when the diameter of the shunting vessel is greater than that of the recipient portal vein (2,13,14). Also, settings where intrahepatic vascular resistance fails to fall (e.g. small-for-size graft, rejection, and volume overload) can prevent spontaneous regression of the collateral vessels (11,12,22,23) which would further serve as draining channel.

Therefore, identification of possible pathways of portosystemic shunts during the pre-transplantation evaluation, appropriate surgical or interventional interruption of these pathways, and timely detection of persistent or recurrent shunts during the post-transplantation follow-up could potentially improve the outcomes of liver transplant recipients.

Preoperative evaluation of potential portosystemic shunt in recipients of liver transplantation

Various collateral pathways are developed in portal hypertensive liver cirrhosis. Thus, numerous routes of intra- or extrahepatic portosystemic shunts can be found in the liver transplantation candidates. Common pathways of extrahepatic portosystemic shunts include splenorenal shunt, gastrorenal shunt, gastroesophageal shunt, mesocaval shunt, portophrenic shunt, and para-umbilical shunt (Fig. 1) (4,22,24). While intrahepatic shunts and some extrahepatic shunts such as portophrenic or para-umbilical shunts can be eliminated with harvest of the native liver in the recipients, gastrorenal shunts, splenorenal shunts, mesocaval shunts, gastrophrenic shunts, and other minor non-removable shunts may persist and potentially contribute to portal steal following liver transplantation (4,14,25). Splenorenal shunt is one of the major concerns due to their higher prevalence and tendency to grow larger (4,26).

Fig. 1.

Illustration of the common pathways of extrahepatic portosystemic shunts in portal hypertensive liver cirrhosis. GEV, gastroesophageal vein; GRS, gastrorenal shunt; IMV, inferior mesenteric vein; LGV, left gastric vein; LPV, left portal vein; LRV, left renal vein; MCS, mesocaval shunt; PPS, portophrenic shunt; PUV, para-umbilical vein; PV, portal vein; SMV, superior mesenteric vein; SRS, splenorenal shunt; SV, splenic vein; SVC, superior vena cava. *To retroperitoneal-paravertebral vein; ^†To gonadal vein; ^‡To internal iliac vein.

In addition to supreme portability, Doppler ultrasound enables real-time monitoring of flow velocity and direction. In pre-transplantation evaluation of the recipients, ultrasound can reveal various collateral vessels by direct observation of aberrant, engorged vessels draining the portal flow to the systemic circulation. Superficially-located shunts such as a para-umbilical shunt can be readily visualized by ultrasound. Larger deep-seated shunts as well can be visualized; engorged venous collaterals related with gastroesophageal shunt, gastrorenal shunt, splenorenal shunt, and mesocaval shunt can be detected with meticulous inspection (Figs. 2 –4). Doppler studies are useful in distinction between the flow stagnation and the portal steal. Depiction of hepatofugal blood flow in the engorged collateral vessels confirms the presence of portal steal. In these cases, the velocity of portal flow decreases, and even can be bidirectional, reversed, or undetectable due to steal phenomenon (1).

Fig. 2.

Gastroesophageal shunt demonstrated on Doppler ultrasound in a 64-year-old male patient with liver cirrhosis. (a) On transverse scan, Doppler ultrasound shows normal hepatopetal blood flow direction at splenic vein (SV) heading for portal confluence (*). IVC, inferior vena cava; SMA, superior mesenteric artery. (b) Doppler ultrasound of the level just superior to (a) shows hepatofugal blood flow at left gastric vein (LGV), appearing as red color (arrow), instead of blue, in the gastrohepatic space. (c) Sagittal Doppler sonogram demonstrates hepatofugal flow through the dilated coronary vein and subsequent paraesophageal varix (arrows), indicating gastroesophageal shunt.

While Doppler ultrasound excels in depicting the direction and velocity of blood flow, computed tomography (CT) is superior in displaying anatomic details and provides essential information needed for pre-transplantation planning of surgical or interventional interruption of portosystemic shunts. Triple-phase CT scan consisting of unenhanced scan, arterial phase, and portal venous phase with vascular reformatting techniques such as maximum intensity projection or volume rendering is useful in revealing anatomic details of portosystemic shunts (Figs. 3 and 5) (27 –29). Regarding gastrorenal or splenorenal shunts, it is generally agreed that the size of the shunt may matter; although splenorenal shunts smaller than 10 mm at the level of drainage into left renal vein is likely to collapse spontaneously after implantation of a normal vascular resistance graft, larger shunts may require percutaneous or surgical closure. It should also be noted, although rare, if there is an anatomic variation in the left renal vein such as circumaortic or retroaortic variant for successful interruption of shunt in such cases (Fig. 6). Therefore, the diameters and locations of shunts observed during the portal venous phase should be precisely described in detail in the reports for pre-transplantation CT scans (2,29).

Fig. 3.

Gastrorenal shunt in a 49-year-old liver cirrhosis patient. (a) Doppler ultrasound of the portal vein shows hepatofugal blood flow, suggesting the presence of portal steal. (b) Transverse sonogram shows the dilated coronary vein (arrowheads) in the gastrohepatic space. (c) Maximal intensity projection image of portal venous phase CT demonstrates the hypoplastic portal vein (dashed circle), the dilated coronary vein (arrow) and gastric varix (*), and gastrorenal shunt (arrowhead).

Fig. 4.

Portal steal phenomenon via splenorenal shunt revealed on Doppler ultrasound in a 55-year-old male patient with portal hypertensive liver cirrhosis. (a, b) Doppler ultrasound of the portal vein (a) and splenic vein (b) shows hepatofugal blood flow (arrows), suggesting the presence of portal steal. (c–e) Doppler ultrasound of the left upper quadrant shows the dilated splenic vein (*) around the spleen (SPL) and dilated veins of splenorenal shunt pointing to the left renal vein (arrowheads). 3D arrows indicate the flow direction.

Fig. 5.

Major mesocaval shunt in a 47-year-old female patient with portal hypertensive liver cirrhosis. (a) On the volume rendering image, markedly distended mesenteric varix (arrowheads) as well as splenic varix (*) is intuitively visualized. (b) On axial image, drainage of the mesenteric varix into right posterior aspect of the inferior vena cava is well demonstrated (arrowhead).

Fig. 6.

Portal steal via circumaortic left renal vein detected on pre-transplantation CT in a 54-year-old female with portal hypertensive liver cirrhosis. (a, b) Axial CT scans show massive splenomegaly (SPL), splenic varices (arrowheads), and distension of the left renal veins (arrows) due to large splenorenal shunt. Note that the left renal veins course anterior (a) and posterior (b) to the aorta, so-called circumaortic left renal vein. (c) On the volume rendering image, large splenorenal shunt via both the left renal vein at its usual anteaortic course (*) and more caudally located retroaortic left renal vein (arrowhead) are noticeable.

More recently, magnetic resonance imaging (MRI) has been highlighted as a feasible tool for measuring portal flow. Several studies have quantitatively measured and compared the velocity and volume of portal venous as well as hepatic arterial flow between portal hypertensive liver cirrhosis patients and healthy volunteers, showing significantly reduced portal flow in the liver cirrhosis patients (30 –33). Further studies on use of MRI in direct quantification of shunt flow via collateral vessels would follow.

Intraoperative image-guided interventional or surgical management of portosystemic shunt

In adult living donor liver transplantation, surgeons may try to ligate major portosystemic shunts larger than 10 mm if technically feasible. During the operation, Doppler ultrasound helps determining better candidate for elective shunt ligation by demonstrating portal flow velocity and volume increment upon manual compression of the shunt (2,27,34).

Specifically, after covering the high-frequency transducer with a sterile sleeve, the abdominal cavity is filled with a warmed saline irrigation solution to facilitate ultrasound transmission. The transducer is placed directly upon the portal vein. After adjusting the sampling gate to encompass the whole lumen of the portal vein, real-time, continuous tracing of mean peak velocity and volume of portal flow is performed before and after the tentative compression or temporary ligation of the shunt to assess the effectiveness of permanent ligation. If continuous tracing function is not available or the measurement is imprecise, portal flow volume (Q) can be manually calculated by area of portal vein (A) multiplied by mean velocity of portal flow (v) averaged (Q = v × A).

Studies have suggested that portal flow should be at least 1000–1200 mL/min to prevent portal hypoperfusion after liver transplantation (10,27). However, several case reports warranted cautions on using Doppler ultrasound as an only guidance to perform shunt ligation. In those cases, portal flow immediately after reperfusion was sufficient on Doppler ultrasound, only to result in devastating portal steal within a few days (1,2,13).

Portography also can be utilized during the operation. After directly puncturing the isolated inferior mesenteric vein, guide wire and catheter are advanced to the superior mesenteric vein or splenic vein under the guidance of fluoroscopy. Next, venograms are obtained to judge the portal venous flow and portosystemic shunts (35). Intraoperative portogram is advantageous since it can localize culprit collateral vessels that persistently cause portal steal even after the ligation of larger shunts. It is also useful in evaluating the completeness of the shunt occlusion by demonstrating portal inflow change before and after the ligation or other intervention (Fig. 7). Portogram can reveal ancillary abnormality of the portal vein, e.g. stenosis or thrombosis (35). Moreover, it can guide intraoperative transvenous embolization of portosystemic shunt.

Fig. 7.

Real-time intraoperative monitoring of portal flow using Doppler ultrasound in a 59-year-old male patient. The patient had robust portal flow volume on Doppler ultrasound.

Transvenous embolization of portosystemic shunt under the guidance of portography in conjunction with surgical ligation has shown good technical and clinical success (35). Intraoperative transvenous embolization can be performed when surgical ligation is incomplete or difficult to approach due to anatomic location, e.g. superior mesenteric varix or varix at the splenic hilum. Transvenous embolization of splenorenal shunts either during or after the liver transplantation has been more frequently reported (34 –36), though it can be technically challenging in managing larger varixes for fear of migration of embolic material (Fig. 8) (26,35).

Fig. 8.

Transvenous embolization of splenorenal shunt in a 64-year-old female patient with splenorenal shunt. (a) On the volume rendering image, large splenorenal shunt (arrowhead) is evident. LRV, left renal vein; SV, splenic vein. (b) Intraoperative splenic venogram shows large splenorenal shunt (arrowhead). LRV, left renal vein; SV, splenic vein. (c) The final splenic venogram after embolization using coils and histoacryl shows complete occlusion of the shunt.

Radiologic evaluation for persistent or recurrent portosystemic shunt following liver transplantation

Doppler ultrasound is useful in monitoring portosystemic shunts during the follow-up evaluation of the recipients after liver transplantation. After liver transplantation, portal flow peaks on the very early postoperative period due to elevated cardiac output of liver cirrhosis patients (1,27,37), followed by gradual decrease in line with the graft regeneration (1). While this trend of portal flow change is within the expected course of variation, if there is persistent or recurrent portosystemic shunts early after liver transplantation, portal flow velocity can be markedly decreased (Fig. 9), sometimes with bidirectional, hepatofugal, or no detectable flow (2,13,38). Collapse of portal vein and/or hepatofugal flow in the shunting vessels can also be observed (2,13). In one study, the authors anecdotally reported reversal of left portal vein flow to be an early indicator of persistent or recurrent portal steals in patients with splenorenal shunt (26).

Fig. 9.

Recurrent portal steal through the mesocaval shunt in a 58-year-old female patient who underwent liver transplantation for portal hypertensive liver cirrhosis. (a, b) Immediately after living donor liver transplantation, mesenteric varix (dashed oval) is relatively collapsed on CT (a) and adequate portal flow velocity is noted on Doppler ultrasound (b). Stent has been intraoperatively placed over the anastomosis and recipient portal vein due to extensive bland portal vein thrombosis. (c) Follow-up study at 6 months after reveals markedly engorged mesenteric varix (arrowheads) and progressive splenomegaly (SPL), indicating the recurrent or persistent portal hypertension. (d) On Doppler ultrasound, portal flow velocity has reduced to approximately <20 cm/s. Under the suspicion of recurrent portal steal through the mesocaval shunt based on CT and Doppler ultrasound findings, urgent surgical exploration was performed for interruption of the portosystemic shunt. (e) Intraoperative portogram via splenic vein shows non-opacification of the left part of the mesenteric varix seen on (c), after multiple surgical ligation of the varix. Patent portal vein stent and good flow to the liver is also noted. (f) On follow-up Doppler ultrasound after the shunt ligation, portal flow velocity has restored.

In the light of superior anatomic depiction, CT readily demonstrates the findings related with recurrent portal steal in the long term follow-up. A meticulous inspection of signs of succeeding portal hypertension such as progression of splenomegaly, ascites, development of new portosystemic collaterals, narrowing of portal vein, and portal hypertensive colopathy, in conjunction with Doppler ultrasound monitoring of portal flow change and clinical inspection for hepatic dysfunction may provide a clue to diagnose recurrent portal steal (Fig. 9).

Although rare, portal steal may recur with recanalization of ligated shunt vessel, which can be readily demonstrated by CT. Doppler ultrasound may also provide a clue if there is a marked decrease in the portal flow velocity compared with the prior examination.

Conclusion

Persistent portal steal via portosystemic collaterals can be disastrous after liver transplantation by restraining liver regeneration. Preoperative detection of potential pathways of shunts benefits patients by optimizing the surgical and interventional strategies. After liver transplantation, meticulous monitoring is necessary for timely detection of recurrent portal steal.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Lee

Moon

Ahn

. Ligation of left renal vein for large spontaneous splenorenal shunt to prevent portal flow steal in adult living donor liver transplantation. Transpl Int 2007; 20: 45–50.

Horrow

Phares

Viswanadhan

. Vascular steal of the portal vein after orthotopic liver transplant: intraoperative sonographic diagnosis. J Ultrasound Med 2010; 29: 125–128.

Wachsberg

Bahramipour

Sofocleous

. Hepatofugal flow in the portal venous system: pathophysiology, imaging findings, and diagnostic pitfalls. Radiographics 2002; 22: 123–140.

Ikegami

Shirabe

Nakagawara

. Obstructing spontaneous major shunt vessels is mandatory to keep adequate portal inflow in living-donor liver transplantation. Transplantation 2013; 95: 1270–1277.

Eipel

Abshagen

Vollmar

. Regulation of hepatic blood flow: The hepatic arterial buffer response revisited. World J Gastroenterol 2010; 16: 6046–6057.

Iwao

Toyonaga

Shigemori

. Hepatic artery hemodynamic responsiveness to altered portal blood flow in normal and cirrhotic livers. Radiology 1996; 200: 793–798.

Aoki

Imamura

Kaneko

. Intraoperative direct measurement of hepatic arterial buffer response in patients with or without cirrhosis. Liver Transpl 2005; 11: 684–691.

Gülberg

Haag

Rössle

. Hepatic arterial buffer response in patients with advanced cirrhosis. Hepatology 2002; 35: 630–634.

Zipprich

Steudel

Behrmann

. Functional significance of hepatic arterial flow reserve in patients with cirrhosis. Hepatology 2003; 37: 385–392.

10.

Castillo-Suescun

Oniscu

Hidalgo

. Hemodynamic consequences of spontaneous splenorenal shunts in deceased donor liver transplantation. Liver Transpl 2011; 17: 891–895.

11.

De Carlis

Del Favero

Rondinara

. The role of spontaneous portosystemic shunts in the course of orthotopic liver transplantation. Transpl Int 1992; 5: 9–14.

12.

Braun

Bar-Nathan

Shaharabani

. Postshunt hepatic encephalopathy in liver transplant recipients. Transplantation 2009; 87: 734–739.

13.

Cho

Kim

Lee

. Ligation of left renal vein as a salvage procedure for splenorenal shunt after living donor liver transplantation: a case report. Transplant Proc 2009; 41: 4243–4245.

14.

Sanada

Mizuta

Urahashi

. Impact of posttransplant portosystemic shunts on liver transplantation. World J Surg 2012; 36: 2449–2454.

15.

Fujimoto

Moriyasu

Nada

. Influence of spontaneous portosystemic collateral pathways on portal hemodynamics in living-related liver transplantation in children. Doppler ultrasonographic study. Transplantation 1995; 60: 41–45.

16.

Kyoden

Sugawara

Kaneko

. Hepatofugal portal flow after living donor liver transplantation. Hepatogastroenterology 2007; 54: 1164–1166.

17.

Sugimoto

Kaneko

Marui

. Reversal of portal flow after acute rejection in living-donor liver transplantation. J Hepatobiliary Pancreat Surg 2001; 8: 573–576.

18.

Jones

McCall

Thomas

. Reverse portal flow after liver transplantation–ominous or acceptable? Pediatr Transplant 2010; 14: E34–E37.

19.

Kyoden

Sugawara

Matsui

. Hepatofugal portal flow due to acute cellular rejection. Abdom Imaging 2005; 30: 303–305.

20.

Kita

Harihara

Sano

. Reversible hepatofugal portal flow after liver transplantation using a small-for-size graft from a living donor. Transpl Int 2001; 14: 217–222.

21.

Margarit

Lazaro

Charco

. Liver transplantation in patients with splenorenal shunts: intraoperative flow measurements to indicate shunt occlusion. Liver Transpl Surg 1999; 5: 35–39.

22.

Sadamori

Yagi

Matsukawa

. The outcome of living donor liver transplantation with prior spontaneous large portasystemic shunts. Transpl Int 2008; 21: 156–162.

23.

Duffy

Hong

Farmer

. Vascular complications of orthotopic liver transplantation: experience in more than 4,200 patients. J Am Coll Surg 2009; 208: 896–903. discussion 903–895.

24.

Moon

Lee

Ahn

. Application of intraoperative cine-portogram to detect spontaneous portosystemic collaterals missed by intraoperative doppler exam in adult living donor liver transplantation. Liver Transpl 2007; 13: 1279–1284.

25.

Chmurowicz

Zasada-Cedro

Wojcicki

. Cavoportal hemitransposition for unrecognized spontaneous mesocaval shunt after liver transplantation: a case report. Transpl Int 2013; 26: E46–E49.

26.

Awad

Horrow

Parsikia

. Perioperative management of spontaneous splenorenal shunts in orthotopic liver transplant patients. Exp Clin Transplant 2012; 10: 475–481.

27.

Aucejo

Hashimoto

Quintini

. Triple-phase computed tomography and intraoperative flow measurements improve the management of portosystemic shunts during liver transplantation. Liver Transpl 2008; 14: 96–99.

28.

Kang

Jeong

Choi

. Three-dimensional multi-detector row CT portal venography in the evaluation of portosystemic collateral vessels in liver cirrhosis. Radiographics 2002; 22: 1053–1061.

29.

Pannu

Maley

Fishman

. Liver transplantation: preoperative CT evaluation. Radiographics 2001; 21: S133–S146.

30.

Stankovic

Csatari

Deibert

. A feasibility study to evaluate splanchnic arterial and venous hemodynamics by flow-sensitive 4D MRI compared with Doppler ultrasound in patients with cirrhosis and controls. Eur J Gastroenterol Hepatol 2013; 25: 669–675.

31.

Dyvorne

Knight-Greenfield

Besa

. Quantification of hepatic blood flow using a high-resolution phase-contrast MRI sequence with compressed sensing acceleration. Am J Roentgenol 2015; 204: 510–518.

32.

Dyvorne

Knight-Greenfield

Jajamovich

. Abdominal 4D flow MR imaging in a breath hold: combination of spiral sampling and dynamic compressed sensing for highly accelerated acquisition. Radiology 2015; 275: 245–254.

33.

Roldan-Alzate

Frydrychowicz

Niespodzany

. In vivo validation of 4D flow MRI for assessing the hemodynamics of portal hypertension. J Magn Reson Imaging 2013; 37: 1100–1108.

34.

Al hajjaj

Bonatti

Krishna

. Percutaneous transfemoral embolization of a spontaneous splenorenal shunt presenting with ischemic graft dysfunction 18 months post-transplant. Transpl Int 2008; 21: 816–819.

35.

Kim

Sung

. Transvenous variceal embolization during or after living-donor liver transplantation to improve portal venous flow. J Vasc Interv Radiol 2009; 20: 1454–1459.

36.

Litvin

Atar

Knizhnik

. Stent graft closure of a high flow splenorenal shunt after liver transplantation. Diagn Interv Radiol 2010; 16: 312–314.

37.

Tarantino

Citro

Conca

. What are the implications of the spontaneous spleno-renal shunts in liver cirrhosis? BMC Gastroenterol 2009; 9: 89–89.

38.

Kousoulas

Ringe

Winkler

. Ligation of left renal vein for spontaneous splenorenal shunt to prevent portal hypoperfusion after orthotopic liver transplantation. Case Rep Transplant 2013; 2013: 842538–842538.