Role of plasmapheresis in early allograft dysfunction following deceased donor liver transplantation

Abstract

The role of plasmapheresis in liver failure and hepatic encephalopathy is undefined and its use as a strategy to salvage patients with severe allograft dysfunction after liver transplantation remains investigational. We present a case of early allograft dysfunction following deceased donor liver transplantation (DDLT) where plasmapheresis was effective as a bridge to recovery and possibly avoiding a retransplantation. A 16 years old boy, known to have decompensated Wilson’s disease underwent DDLT at our Public Sector Hospital. He received a healthy liver from a brain-dead donor, whose liver was considered too large for the boy. The graft was reduced in situ to a left lobe graft. Surgery was uneventful and the recipient was well for the initial 96 h. On Doppler and further computed tomography scan, a partial portal vein thrombus was noted. He was reexplored and a Fogarty endothombecteomy was performed. Following the second surgery, he developed severe allograft dysfunction with a peak bilirubin of 40 mg/dL. He underwent imaging to rule out technical causes for the dysfunction, followed by a liver biopsy, which revealed acute cellular rejection. Multiple cycles of plasmapheresis were initiated. Over the next two weeks, the graft demonstrated a gradual recovery. He was discharged on the 30^th postoperative day, with a serum bilirubin of 5.5 mg/dL. He remains well on follow-up, with the liver function tests improving further. Our report demonstrates the beneficial effect of plasmapheresis, which appears to be an effective treatment option for early allograft dysfunction following liver transplantation and may obviate the need for retransplantation.

Key Words: Liver transplantation; Allograft dysfunction; Plasmapheresis

Core tip: We demonstrate the beneficial effects of plasmapheresis, which appears to be an effective treatment option for early allograft dysfunction following liver transplantation and may obviate the need for retransplantation.

INTRODUCTION

With expanding indications and increasing demand for liver transplantation (LT) donor organ shortage is a major limitation. Early allograft dysfunction (EAD) is not an uncommon entity, especially in transplantation with organs from marginal donors[1]. The incidence of EAD varies between 1.4%-23%, with a median range of 5%-6%[1-3]. This wide range of incidences is attributable to the myriad of definitions which exist for EAD although most definitions are a combination of elevated bilirubin, international normalized ratio (INR), transaminases, and hepatic encephalopathy.

EAD leads to increased morbidity and may result in mortality and liver support therapies need to be instituted[2-5]. In severe forms, retransplantation may be the only treatment modality. If the duration of early graft dysfunction passes uneventfully, the patient often recovers spontaneously[1-6]. Measures like liver support devices lessen the hepatic metabolic burden and may help in the recovery of graft function[2-4]. Plasmapheresis has been used in acute liver failure, but its role in supporting dysfunctional liver allografts remains unclear[7,8].

In this brief report, we present a case of allograft dysfunction following deceased donor liver transplantation (DDLT) where the graft was salvaged using multiple cycles of plasmapheresis.

CASE REPORT

A 16-year-old boy weighing 33 kg, with decompensated Wilson’s disease underwent DDLT at our Public Sector Hospital. He received a healthy liver from a 34-year-old brain-dead donor. The donor had an initial Sodium value of 194 meq/L, which was controlled and brought down to 164 meq/L at the time of organ retrieval. The donor had one episode of significant hypotension. As the donor liver was considered too large for the boy, it was reduced in situ into a left lobe graft.

The recipient operation was uneventful with a total blood loss of 1000 mL. The end lactate was 4.2 mmol/L from a peak of 10.2 mmol/L in the anhepatic phase. The total cold ischemia time was 210 min. The graft had an accessory artery from the left gastric artery, taken as a cuff from the celiac axis and anastomosed to the common hepatic artery of the recipient. The surgery was uneventful and the recipient was well initially; being extubated on the 1^st postoperative day (POD). Immunosuppression was initiated with steroids (Methyl-Prednisolone 0.25 mg/kg per day) and calcineurin inhibitors (Tacrolimus 0.03 mg/kg per day) from POD 1. On the 5^th POD, his drain output increased from 600 mL in 24 h to 1700 mL and his serum bilirubin which had dropped to 3.8 mg/dL, went upto 9.3 mg/dL. Doppler showed poor flow in the portal vein. On further imaging, he was noted to have a partial portal vein thrombus. He underwent emergency re-exploration when a Fogarty endothrombectomy was done and the graft was revascularised with an iliac vein interposition graft for the portal venous anastomosis. During the second surgery, the graft was noted to be very stiff. Following this, he developed severe allograft dysfunction (rising serum bilirubin > 10 mg/dL over 3 consecutive days in the absence of biliary complications). Over the next 5 d his bilirubin increased up to 23.5 mg/dL, while his transaminases remained normal. He underwent repeat imaging which ruled out technical causes for the dysfunction including a patent portal vein. Liver biopsy was performed which was suggestive of moderate acute cellular rejection, there was no evidence of antibody mediated rejection. He received pulsed steroid therapy (Methyl-Prednisolone 20 mg/kg per day on consecutive three days). Despite the steroid pulse, the graft dysfunction did not abate and the hyperbilirubinemia persisted on an upward trend, peaking at 40.8 mg/dL on the 15^th POD. In an effort to salvage the graft, plasmapheresis was initiated on the 15^th POD.

Plasmapheresis was done on 5 consecutive days using continuous flow centrifugal technology based Spectra Optia Apheresis system (Terumo BCT, Denver, CO, United States). Acid citrate dextrose-A anticoagulation and dual vascular access were used. Patient's total blood volume (BV) was calculated as per Nadler's formula and Plasma volume (PV) was calculated according to the formula PV = BV × (1 - Hematocrit)[8]. 1.0 PV was processed in each session with 100% replacement using 5% albumin solution and blood group specific fresh frozen plasma. The inlet: Anticoagulant ratio was kept 1:12 to 1:15 and blood flow rate kept between 45-50 mL/min. Baseline calcium was monitored before each procedure and 20 mL 10% calcium gluconate was given prophylactically during the procedure to prevent citrate toxicity. Continuous monitoring of pulse and blood pressure was carried out during the procedure to prevent any adverse events related to the procedure. No serious adverse effects were observed during the procedure. Complete blood count, INR, liver function tests, renal function tests, arterial ammonia, arterial blood gas analysis, were performed every 12 hourly irrespective of the timings of the plasmapheresis.

His bilirubin showed a steady fall and by the 5^th cycle of plasmapheresis it had dropped to 15 mg/dL (Figure 1). In the interim, he had an episode of fever with chills, and grew K. pneumonia in his blood culture. This was successfully treated with appropriate antibiotics (Piperacillin-Tazobactum). No obvious source for the infection could be discerned. He was discharged on the 30^th POD being asymptomatic, tolerating oral diet well, with stable vital signs and with a serum bilirubin of 5.5 mg/dL. He remains well on follow-up, with the latest liver function tests showing a total bilirubin of 1 mg/dL, 4 mo after transplantation.

Figure 1 Liver function tests and its response to plasmapheresis. Red arrow: Plasmapheresis; Bilirubin in mg/dL; AST: Aspartate transaminase in IU/L; INR: International normalized ratio.

DISCUSSION

Although the pathophysiological basis for early allograft dysfunction has not been wholly elucidated; it appears to be a critical interplay between donor factors, recipient characteristics, and intra-operative events[1,2,4,5].

Despite a few studies including one by Park et al[4] having shown plasmapheresis to be beneficial in severe graft dysfunction; the role of plasmapheresis remains undefined in graft dysfunction[1-4]. The mechanisms by which plasmapheresis is beneficial hasn’t been completely elucidated, but it does remove the plasma containing free and protein-bound toxic substrates and infuse fresh plasma, as well as clotting factors and albumin, thus functioning as a liver support; creating a milieu conducive to liver regeneration[1-6]. Plasmapheresis is an important adjunct in the treatment of hepatic encephalopathy as it improves blood-clotting, hyperbilirubinemia, and hyperammonemia; acting as a bridge to LT[7,8].

In a series from Japan, all 46 patients with liver failure following LT improved with plasmapheresis[5]. In another recent study by Choe et al[3] consisting of 143 patients with EAD of whom 107 underwent Plasmapheresis. There was a significant improvement in the 1-mo and 1-year survival of this subgroup of patients as compared to those who did not undergo plasmapheresis. A report from Johns Hopkin also suggested that plasmapheresis may aid in the recovery of primary allograft nonfunction following liver transplantation[9,10].

As demonstrated in our patient, a single plasmapheresis session cannot be expected to provide a definite beneficial effect in patients with a failing liver graft[2,3]. Repeated sessions appear necessary to achieve cumulative effects. The timing and interval of plasmapheresis must be adjusted on a case-by-case basis, by daily determination of patient’s general condition and liver graft function[2-4]. In a dysfunctional liver, the liver enzymes often fluctuate, depending on the condition of the liver graft, and hence cannot be used to assess the effectiveness of plasmapheresis. Prothrombin time is readily affected by the plasma infusion and is also not a predictable marker of the effectiveness of plasmapheresis[3,11]. Serum bilirubin appears to be the most reliable parameter to base decision regarding the initiation, continuation and termination of plasmapheresis[2,4,11].

In countries, where retransplantation may not be a feasible option due to the lack of availability of donor grafts and/or the huge financial burden involved, plasmapheresis appears to be a readily available artificial liver support system with the added advantage of being economical, simple and easy to use.

Apart from benefiting the patient, effective management of early allograft dysfunction helps a unit improve its overall efficiency and sets a benchmark for excellence in care by showing an enhancement in healthcare delivery in general.

In conclusion, our report demonstrates the beneficial effect of plasmapheresis, which appears to be an effective treatment option for early allograft dysfunction following DDLT and may obviate the need for retransplantation.

COMMENTS

Clinical diagnosis

Early liver allograft dysfunction.

Differential diagnosis

Arterial complications, venous complications, biliary complications, rejection, infection.

Laboratory diagnosis

Early liver allograft dysfunction.

Imaging diagnosis

Early liver allograft dysfunction.

Pathological diagnosis

Early liver allograft dysfunction.

Treatment

Plasmapheresis.

Term explanation

Plasmapheresis - total plasma exchange.

Experiences and lessons

Useful not to disregard a simple but very effective procedure such as plasmapheresis in treating early allograft dysfunction.

Peer-review

The paper is good, although it doesn’t seem to get different conclusion from the larger case series already published.