- Top of page
- PATIENTS AND METHODS
ABO-incompatible (ABO-I) liver transplantation has been performed essentially in patients with acute liver failure awaiting an urgent liver transplantation. Early results with ABO-I liver transplantation were disappointing with a very low graft survival rate (20–50%). The main risk is the occurrence of severe humoral and cellular rejection, vascular thrombosis, and biliary complications. In order to avoid humoral rejection and improve graft survival, total plasma exchange in combination with an intense immunosuppressive regimen has been proposed to decrease hemagglutinin titers in ABO-I liver grafts. In some centers, this regimen was associated with splenectomy, phototherapy, and portal or arterial intrahepatic infusion therapy; however, as these patients are at high risk of sepsis, a selective approach using antigen-specific immunoadsorption with immunoadsorbent columns has been successfully proposed for ABO-I living donor kidney transplantation. Few cases have been reported following liver transplantation. We report our recent experience with three adult patients (two patients with acute liver failure, and one with severe cirrhosis and hepatic encephalopathy) transplanted in an emergency situation with an ABO-I liver graft and managed with the use of GlycoSorb ABO immunoadsorbent columns and a quadruple immunosuppressive regimen with preservation of the spleen. Eight sessions were performed in the three patients. Antigen-specific immunoadsorption greatly lowered the anti-A hemagglutinin titers. None of the three patients developed acute humoral or cellular rejection. Two patients are alive at 1.5 and 3 years follow-up with a normally functioning graft. The third patient died with a functioning graft, one month after the transplantation, from septic complications.
During the last decade, the number of patients awaiting liver transplantation (LTx) has been dramatically increasing. To overcome the critical shortage of deceased organ donation, the use of marginal or extended criteria organ donors, non-heart beating and living donors, split and domino livers have been widely developed. ABO-incompatible (ABO-I) liver transplantation was first performed in patients with acute liver failure awaiting an urgent liver transplantation. In Eastern countries, where living donor livers provide the only source of organs, ABO-I transplantation has been also proposed for critically-ill patients and for those recipients with hepatocellular carcinoma (1). Viewing the reported literature and the current situation of organ donation, the problem of an ABO blood type incompatibility between voluntary donors and recipients in critical situations seems to be unavoidable.
Early results with ABO-I liver transplantation were disappointing because of the enhanced risk of severe humoral and cellular rejection of the graft, vascular thrombosis, and biliary tract complications. Angiocholitis related to ischemic cholangitis was common, often leading to liver abscesses and loss of the graft (2–4). Less frequently, signs of hyperacute rejection with acute periportal oedema have been described. Hyperacute rejection usually occurs during the first week after transplantation as a consequence of severe immunological reactions and leads to massive hepatocyte necrosis and graft loss (5,6).
In a report of our initial experience with 17/234 patients transplanted with an ABO-I graft in the late 1980s, the one-year patient survival rate was 66%, but the two-year graft survival rate was 30% as compared to 76% for compatible ABO transplants (2). Later, the analysis of our major cohort included 43 ABO-I liver transplantations; the five-year patient and graft survival rates were, respectively, 50% and 20%. The most common observed complications were vascular thrombosis (41%), biliary injury (34%), and hyperacute rejection (20%) (3).
Demetris et al. showed, in patients transplanted with an ABO-I graft, that liver allografts are susceptible to antibody-mediated rejection from preformed complement-fixing hemagglutinin leading to a hemorrhagic infiltration of portal tracts and deposition of fibrinogen and immunoglobulin (Ig)M on sinusoidal wall and vascular endothelium (4). The high rate of biliary complications leading to graft loss might be explained by the fact that the bile ducts may also express ABO antigens and serve as a target for antibody-mediated injury (7).
During the last two decades, several therapeutic regimens have been proposed to decrease hemagglutinin titers in ABO-I liver transplantations in order to avoid humoral rejection. The most common regimen included total plasma exchange in combination with immunosuppressive induction therapy (anti-thymocyte globulin, the monoclonal antibody OKT3, or cyclophosphamide). In some centers this regimen was associated with splenectomy, phototherapy, portal or arterial intrahepatic infusion therapy (6–12). The main risk of these intensive regimens is sepsis, commonly enhanced by the severity of the acute illness. More recently, a selective approach using antigen-specific immunoadsorption with immunoadsorbent columns (GlycoSorb ABO; Glycorex Transplantation AB, Lund, Sweden) (13,14), combined or not to anti-CD20 humanized monoclonal antibodies (rituximab [Mabthera; Roche Pharmaceuticals, Neuilly-Sur-Seine, France]), has been successfully proposed for ABO-I living donor kidney and liver transplantations (14–16). We report here our recent experience with three adult patients who received an ABO-I liver graft and managed with the use of GlycoSorb ABO immunoadsorbent columns and a quadruple immunosuppressive regimen with preservation of the spleen.
- Top of page
- PATIENTS AND METHODS
Early experience, in the late 1980s, in ABO incompatible liver transplantation showed serious complications of the graft in more than half of the patients with a major impact on morbidity and mortality. Despite that the liver appears to be less sensitive to antibody-mediated rejection, as compared to other organs, humoral rejection of the graft, ischemic cholangitis and vascular complications were common (2–4). Demetris et al. reported a 46% graft failure rate during the first 30 days for primary ABO-I grafts compared with an 11% graft failure rate for primary ABO compatible, cross-match negative, age, sex and priority-matched control patients (P < 0.02) (4). Recently, the two-year adult recipient survival rate after ABO-I living donor liver transplantation (LDLTx) in Japan was reported to be approximately 70%, while that for ABO-I LDLTx before 2002 was about 40% (17). The risk of graft dysfunction is mainly present during the first month, and tends to vanish later; this is probably related to at least a partial replacement of graft vasculature by endothelial cells that no longer carry ABO antigens (4).
Currently, ABO-I liver transplantation is performed in very rare situations and remains a subject of major debate. However, ABO incompatible liver transplantation could be a life saving procedure in particular situations such as acute liver failure, hepatocellular carcinoma that are at risk of dropout from the waiting list, living related donor transplantation, and in some other critical situations. Liver transplant surgery, performed in an emergency situation, has been considered as a salvage transplantation. Our three patients presented with a severe critical illness, as manifested by the neurological status of coma, hemodynamic instability, and rapid progression of the disease to a fatal outcome in the absence of transplantation. The short- and long-term outcomes of these patients transplanted with an ABO-I liver graft has been very encouraging with the recent improvement in immunosuppressive regimens and apheresis techniques.
It has been reported that the main complications of ABO-I LTx, such as hepatic necrosis and intrahepatic biliary complications, were closely related to the titer of perioperative anti-donor ABO antibodies (18). As a result, most of the efforts to improve the outcome of ABO-I LTx have been directed toward the lowering of anti-blood type antibodies. The impact of preformed anti-donor ABO antibodies and the strategy to reduce their titers play key roles in the success of ABO-I transplantation (8,9,16,19). In elective cases, plasmapheresis has been even initiated before transplantation in order to reduce the titer of IgM or IgG; this is more difficult to realize in emergency cases, such as the three patients described here.
Selective antigen-specific hemagglutinin immunoadsorption has been widely used in ABO-I kidney transplantation, but only in a few reported liver transplant recipients. It has been proposed in order to minimize side-effects and the morbidity related to total plasma exchange (20). Troisi et al. had previously reported in five recipients transplanted with an ABO-I graft, that antigen specific immunoadsorption using GlycoSorb ABO and quadruple immunosuppression, including daclizumab and mycophenolate mofetil, provide high efficiency to lower hemagglutinin titers over time, avoiding the need for splenectomy (13). Daclizumab and mycophenolate mofetil have been successfully used in ABO-I transplantation, therefore also minimizing the risk of infection (21). In the eight sessions performed in our three patients using GlycoSorb A, antigen-specific immunoadsorption showed to be extremely safe and provided efficient capacities to lower significant hemagglutinin titers to very low levels. We did not observe any side-effects related to the specific GlycoSorb immunoadsorption, nor any rituximab-related adverse effects in any of the three recipients.
In addition to the monitoring of hemagglutinin titers and liver enzymes, careful histological surveillance is also mandatory. Acute cellular rejection rates reported in the literature ranged from 20 to 90%, but have been less frequent in recent times (2–4). The three patients each had two liver biopsies performed during the first post-operative month that did not show signs of humoral or cellular rejection, highlighting the efficacy of antigen-specific immunoadsorption and the immunosuppressive regimen.
Toso et al. reported a comparable patient cohort of ABO-I liver transplants (N = 14) with acute liver failure or severely decompensate end-stage disease, intubated, and/or in the ICU. Using a quadruple immunosuppressive protocol without splenectomy, the authors did not observe a significant difference in regard to patient and graft survival when comparing one- and five-year data of ABO-incompatible, -compatible, and -identical liver graft recipients: 64 and 56% of ABO incompatible liver transplants remained functioning after one and five years, respectively. Three ABO-I grafts were lost (one from hyperacute rejection and two from hepatic artery thrombosis) (22). Plasmapheresis was used as adjuvant treatment for rejections in the first 12 patients, and as prophylaxis for high or rising isoagglutinin titers in the subsequent two.