Late Onset Hepatic Encephalopathy after Living Donor Liver Transplant: Successful Reversal by Plug Embolization of Shunt—A Report of Two Cases

Abstract

In liver transplant (LT) recipients, development of portosystemic (PS) shunts can lead to hepatic encephalopathy. Here we report the cases of two patients who presented with hepatic encephalopathy 11 and 15 years after LT. Plug-assisted embolization of PS shunt led to significant improvement in their clinical and neurological symptoms, highlighting the importance of monitoring for PS shunts in long-term LT follow-up.

Keywords

liver transplant - hepatic encephalopathy - portosystemic shunts - vascular plugs - PARTO

Introduction

Liver transplant (LT) is a lifesaving procedure for patients with end stage liver disease (ESLD). Improvement in technique and better postoperative intensive care have improved both short- and long-term survival following LT.

Although spontaneous portosystemic (PS) shunts are a well-known cause of hepatic encephalopathy (HE) in patients with ESLD, their occurrence in the posttransplant period has not been highlighted.

Large PS shunts are often ligated during LT to direct the blood flow to the graft liver and prevent portal steal, which may otherwise lead to inadequate perfusion and graft failure.[1] The shunts that are not ligated or that develop after the LT may in some instances contribute to HE. The initial treatment is usually medical management.

In cases of refractory HE, radiological shunt embolization may help. Herein we describe the clinical presentation and management of two cases of posttransplant refractory HE who presented to us 11 and 15 years after the transplant.

Cases

Patient A

A 65-year-old man underwent living donor liver transplant (LDLT) 11 years back for cryptogenic decompensated cirrhosis with HE. Preoperative triple-phase computed tomography (CT) scan revealed the presence of a large mesenterico-renal shunt that was ligated intraoperatively. Follow-up imaging at 1 month showed complete thrombosis of the shunt.

One year back, the patient had atrial fibrillation (AF) for which he was started on apixaban and amiodarone. One and half month back, the patient presented with waxing and waning sensorium and deteriorated mental functions. On examination, he was drowsy and disoriented, was unable to perform reverse counting and subtraction, and had flapping tremors. CT and magnetic resonance imaging (MRI) of the brain showed no clinically significant findings.

On evaluation, arterial ammonia was found to be elevated (127.9 µmol/L; cutoff value of 60–90 µmol/L) and Doppler examination of the splenoportal axis showed reduced flow velocity in portal vein (PV; 6.6 cm/s; normal: 20–40 cm/s). Based on these findings, the possibility of HE was considered.

Triple-phase contrast CT of the abdomen revealed a large mesenterico-renal shunt arising from the terminal part of the superior mesenteric vein and draining into the left renal vein ([Fig. 1]).

Fig. 1 Maximum intensity projection image from contrast-enhanced computed tomography of patient A shows a large shunt arising from the terminal part of the superior mesenteric vein and draining into the left renal vein. The size of the shunt is approximately 14 mm (white arrow).

For the next 3 days, the patient was managed medically, but with little benefit. Given the large diameter of the shunt and drainage into the left renal vein, plug-assisted retrograde transvenous obliteration (PARTO) was considered for shunt embolization. The benefits of the procedure and the risks involved, including worsening of portal hypertension, were explained to the family.[2]

The procedure was done uneventfully in the angiography suite and the shunt was obliterated with combination of a vascular plug (Amplatzer, Abbott, United States) and Gelfoam injection ([Fig. 2]).

Fig. 2 (A) Left renal venogram of patient A taken via 5-Fr MPA catheter showing the shunt ostia arising from the superior mesenteric vein (black arrow). (B) Venogram of the shunt after deployment of a plug (plug deployed via a long sheath with a diameter of 8 Fr) showing occlusion. (C) Deployed vascular plug in place.

The next day, the patient showed neurological improvement with increased alertness, absence of flapping tremors, and reduction in serum ammonia levels (103 µmol/L). Doppler evaluation revealed hepatopetal flow with increased portal flow velocity (25 cm/s). The patient was discharged in a hemodynamically stable condition 2 days later.

Follow-up after 2 months and 1 year, the patient was active with improved neurological status and his serum ammonia level was 92 and 54 µmol/L, respectively.

Patient B

A 74-year-old man had undergone deceased donor liver transplant (DDLT) 15 years back for hepatitis B virus (HBV) related decompensated chronic liver disease (CLD). He was doing well till 1 month back when he developed generalized body weakness, decreased sleep, and bilateral upper limb tremors. CT and MRI of the brain showed no clinically significant changes.

The serum ammonia level was 110 µmol/L and liver function tests were within normal limits. Doppler examination of the splenoportal axis showed decreased flow velocity in PV (8 cm/s). Contrast-enhanced CT (CECT) abdomen showed normal graft size with cirrhotic changes and multiple collaterals with a prominent lienorenal shunt (22 mm) and gonadal venous shunt (12 mm; [Fig. 3A]).

Fig. 3 (A) Contrast-enhanced computed tomography of patient B shows multiple portosystemic collaterals with a prominent lienorenal shunt. The shunt size is approximately 22 mm (white arrow). (B) Left renal venogram using a 5-Fr MPA catheter of patient B showing the opacification of the gonado-renal shunt (black arrow). (C) Venogram after the deployment of a plug via a long sheath with a diameter of 8 Fr shows decreased flow through the shunt.

Medical management was done for 3 days but was of little benefit and as such embolization of the lienorenal and gonadal vein shunt was considered. The procedure was done in the angiography suite. Initial venograms revealed a gonadal venous shunt with low flow volume. Subsequently after embolization of the lienorenal shunt, repeat interrogation of the gonadal venous shunt revealed increase flow volume of shunt, which was then embolized using a vascular plug (Amplatzer, Abbott, United States; [Fig. 3B, C]).

The next day, the patient showed improved neurological status and there was a decrease in tremors, and the serum ammonia level went down to 105 µmol/L. He was discharged after 2 days in a hemodynamically stable condition.

On follow-up at 1 and 2 months after discharge, there was sustained improvement in mental status. The serum ammonia level went further down to 60 µmol/L.

Discussion

In cirrhotic patients, portal hypertension is known to cause splanchnic vasodilation and hyperdynamic circulation. These factors cause the formation of PS collaterals in approximately 40% cirrhotic patients.[3]

Post-LT, these collaterals may regress spontaneously.[4] If PS shunts persist, they jeopardize the portal inflow, which leads to inadequate perfusion of the liver graft. Large shunt can lead to portal steal syndrome with graft dysfunction and HE.[5] The estimated percentage of the patients with portal steal after LDLT is approximately 0.2%.[1]

There are different algorithms for management of spontaneous PS shunts before, during, or after transplantation. A study from our center concluded that PS shunt ligation may not be necessary if adequate portal flow, good vascular reconstruction, and good graft quality have been ensured.[6]

Endovascular techniques described for PS shunt obliteration include antegrade embolization, balloon-assisted retrograde transvenous obliteration (BRTO), coil-assisted retrograde transvenous obliteration (CARTO)/PARTO, and use of vascular stent grafts.[1]

In our patients, intraoperative shunt ligation was done for patient A as it was large and adequate portal flow was not seen intraoperatively. He presented with refractory HE after 11 years post-LDLT. The development of late encephalopathy in our patient can probably be attributed to various factors. First, the patient was on amiodarone therapy (200 mg/d) for AF for the past 1 year. A cumulative dose of amiodarone as low as 55 g has been shown to cause liver fibrosis or even cirrhosis.[7] Following the development of cirrhosis or fibrosis, small collaterals previously present might have enlarged and precipitated encephalopathy. Second, following the onset of AF, he was also started on apixaban, which might have recanalized any thrombosed vein.[8]

Patient B presented 15 years after his LT with altered mental status and raised ammonia levels.

The clue to HE in these patients was a raised ammonia level. Hyperammonemia is a relatively sensitive marker for detection of PS shunt–related encephalopathy even in patients without cirrhosis.[9] The presence of PS collaterals on CECT strengthened the diagnostic possibility of HE.

In view of the presence of multiple comorbidities in the patients, a minimally invasive endovascular procedure (PARTO) was chosen.

Shunt embolization can result in raised portal pressure, thus causing worsening of gastroesophageal varices and ascites. This is more likely in patients with PV stenosis (PVS) or hepatic venous outflow tract obstruction (HVOTO). In both our cases, there was no imaging evidence of PVS or HVOTO.

To conclude, patent PS shunt should be kept in mind in the follow-up of transplanted recipients with neurological symptoms even if there is no apparent evidence of graft dysfunction. Elevated serum ammonia levels will give a clue and cross-sectional imaging will delineate the shunt, and its obliteration may help in improvement of symptoms.

Conflict of Interest

None declared.

Ethics Approval and Consent to Participate

Ethical committee approval was not needed for this case report. Informed written consent was taken from the patient.

• References

• 1 Jajodia S, Khandelwal AH, Khandelwal R, Kapoor AK, Baijal SS. Endovascular management of portal steal syndrome due to portosystemic shunts after living donor liver transplantation. JGH Open 2021; 5 (05) 599-606
• 2 Álvarez-López P, Campos-Varela I, Quiroga S. et al. Spontaneous portosystemic shunt embolization in liver transplant recipients with recurrent hepatic encephalopathy. Ann Hepatol 2022; 27 (03) 100687
• 3 Zhou HY, Chen TW, Zhang XM, Jing ZL, Zeng NL, Zhai ZH. Patterns of portosystemic collaterals and diameters of portal venous system in cirrhotic patients with hepatitis B on magnetic resonance imaging: association with Child-Pugh classifications. Clin Res Hepatol Gastroenterol 2015; 39 (03) 351-358
• 4 Kim SH, Lee JM, Choi JY. et al. Changes of portosystemic collaterals and splenic volume on CT after liver transplantation and factors influencing those changes. AJR Am J Roentgenol 2008; 191 (01) W8-W16
• 5 Kim B, Kim KW, Song GW, Lee SG. Portal flow steal after liver transplantation. Clin Mol Hepatol 2015; 21 (03) 314-317
• 6 Vijayashanker A, Chikkala BR, Ghimire R. et al. Do natural portosystemic shunts need to be compulsorily ligated in living donor liver transplantation?. J Clin Exp Hepatol 2022; 12 (01) 29-36
• 7 Hussain N, Bhattacharyya A, Prueksaritanond S. Amiodarone-induced cirrhosis of liver: what predicts mortality?. ISRN Cardiol 2013; 2013: 617943
• 8 Wang L, Guo X, Xu X. et al. Anticoagulation favors thrombus recanalization and survival in patients with liver cirrhosis and portal vein thrombosis: results of a meta-analysis. Adv Ther 2021; 38 (01) 495-520
• 9 Gerritzen-Bruning MJ, van den Ingh TS, Rothuizen J. Diagnostic value of fasting plasma ammonia and bile acid concentrations in the identification of portosystemic shunting in dogs. J Vet Intern Med 2006; 20 (01) 13-19

Publikationsverlauf

Artikel online veröffentlicht:
10. Februar 2025