LIVER TRANSPLANTATION 12:972-978, 2006 ORIGINAL ARTICLE Successful ABO-Incompatible Pediatric Liver Transplantation Utilizing Standard Immunosuppression With Selective Postoperative Plasmapheresis Thomas Heffron, 1,2 David Welch, 2 Todd Pillen, 2 Massimo Asolati, 6 Gregory Smallwood, 3 Phil Hagedorn, 2 Chang Nam, 2 Alexander Duncan, 4 Mark Guy, 2 Enrique Martinez, 4 James Spivey, 4 Patricia Douglas, 2 Carlos Fasola, 1 Jill De Paolo, 2 John Rodriguez, 1 and Rene Romero 2,5 1 Department of Surgery, Emory University School of Medicine, Atlanta, GA; 2 Children s Healthcare of Atlanta at Egleston, Atlanta, GA; 3 Department of Pharmacy, Emory University Hospital, Atlanta, GA; 4 Department of Medicine, Emory University School of Medicine, Atlanta, GA; 5 Department of Pediatrics, Emory University School of Medicine, Atlanta, GA; and 6 Dallas VA Medical Center, Dallas, TX Transplanting blood group A, B, or O (ABO)-incompatible (ABO-I) liver grafts has resulted in lower patient and graft survival with an increased incidence of vascular and biliary complications and rejection. We report that, without modification of our standard immunosuppression protocol, crossing blood groups is an acceptable option for children requiring liver transplantation. In our study, ABO-I liver grafts regardless of recipient age have comparable long-term survival (mean follow-up of 3.25 yr) with ABO-compatible grafts without any difference in rejection, vascular or biliary complications. From January 1, 1999 to October 1, 2005, we studied 138 liver transplants in 121 children: 16 (13.2%) received an ABO incompatible liver allograft. One-year actuarial patient survival for ABO-matched grafts vs. ABO-I grafts was 93.0% and 100%, respectively, whereas graft survival was 83.4% and 92.3%. Additionally, 6 of 16 (37.5%) ABO-I transplanted children had 8 rejection episodes, whereas 47 patients (44.8%) had 121 rejection episodes in the ABO-compatible group. There were no vascular complications and 2 biliary strictures in the ABO-I group. Plasmapheresis was not used for pretransplantation desensitization and was only required in 1 posttransplantation recipient. No child was splenectomized. Six of the 16 children were older than 13 yr of age, suggesting the possibility of successfully expanding this technique to an older population. In conclusion, our outcomes may support the concept of using ABO-I grafts in a more elective setting associated with split and living donor liver transplants. Liver Transpl 12:972-978, 2006. 2006 AASLD. Received October 27, 2005; accepted February 2, 2006. Liver transplantation involving blood group A, B, and O (ABO)-incompatible (ABO-I) grafts has historically resulted in low patient and graft survival, along with a significantly higher incidence of acute cellular rejection, and increased postoperative hepatic vascular and biliary complications. This has limited the practice of crossing blood groups to emergent cases only. Many centers have developed novel immunosuppressive regimens in an attempt to improve the viability of ABO-I grafts and have included plasmapheresis and splenectomy. Rydberg et al. 1 suggests that crossing the ABObarrier may be successful particularly in children of younger age. However, ABO-I adult emergency liver transplantation still results in inferior patient and graft survival compared to ABO-identical transplants. Egawa et al. 2 concludes that ABO-I living donor transplants may be performed with relative safety in infants under 1 yr of age using standard immunosuppression. However, the same study found that children over 1 yr of age demonstrated poor long-term graft survival. 2 Reports of success with older age groups often detail small sample groups, such as the case reported Abbreviations: ABO, blood groups A, B, and O; ABO-I, ABO incompatible. Address reprint requests to Thomas Heffron, MD, 1405 Clifton Rd., Atlanta, GA 30322. Telephone: 404-785-6743; FAX: 404-785-1831; E-mail: thomas.heffron@choa.org DOI 10.1002/lt.20760 Published online in Wiley InterScience (www.interscience.wiley.com). 2006 American Association for the Study of Liver Diseases.
PEDIATRIC ABO LIVER TRANSPLANTATION 973 TABLE 1. Immunosuppresive Protocol Months S/P transplantation or acute rejection Predisone per day CsA Level (monoclonal) Tacrolimus level 1 3 Child; 20 kg: 0.3 mg/kg qd; 20 kg: 7.5 mg qd 300 400 ng/dl 15 20 ng/dl 3 6 Child; 20 kg: 0.3 mg/kg qd; 20 kg: 5 mg qd 200 300 ng/dl 10 15 ng/dl 6 12 Child; 20 kg: 0.3 mg/kg qd; 20 kg: 2.5 mg qd 200 300 ng/dl 5 10 ng/dl 12 Child; 20 kg: qod; 20 kg: 5 mg qod 100 200 ng/dl 5 ng/dl Abbreviations; qd, quaquedie, everyday; qod, quaque altera die, every other day; S/P, status post, C s A, cyclosporine A. TABLE 2. Patient Demographics ABO incompatible ABO compatible P No. grafts 16 122 Mean age (yr) 6.5 6.2 8.1 6.2 NS Female 8 (50%) 70 (55.1%) NS Caucasian 7 66 NS African American 8 46 NS Asian 1 5 NS Hispanic 0 10 NS Abbreviation: NS, not significant. TABLE 3. Patient Outcomes ABO compatible grafts (n 122) ABO incompatible grafts (n 16) P Patients with HAT 7 0 NS Patients with bile duct 13 2 NS Rejection episodes 121 8 NS Patients with rejections 47 6 NS Time to first rejection (days) 307 325 144 96 0.001 % 1-yr actuarial patient survival 93 100 NS % 1-yr actuarial graft survival 83.4 92.3 NS Mean follow-up SD (days) 1185 733 925 537 NS Abbreviations: HAT, hepatic artery thrombosis; NS, not significant; SD, standard deviation. by Fang et al. 3 of a 59-year-old woman postoperatively treated with tacrolimus, mycophenolate mofetil, OKT3, steroids, and prostaglandin E1, as well as daily plasmapheresis, for 9 days after transplantation. Prolonged plasmapheresis, such as that noted by Fang et al., 3 is commonly employed in adult ABO-mismatch transplants. Farges et al. 4,5 asserts that increasing immunosuppression and using plasmapheresis postoperatively to reduce the titer of anti-a/b antibodies had little influence on the incidence of hyperacute rejection, vascular thrombosis, and biliary complications. Moreover, splenectomy as an immunosuppressive adjunct as described by Egawa et al., 2 Hanto et al., 6 and Monteiro et al. 7 is still used. In addition to perioperative splenectomy, Monteiro et al. relates the successful use of Rituximab and plasmapheresis immediately before and after ABO-I transplantation. 7 Herein, we report our immunosuppressive approach and surgical experience with ABO-I grafts in pediatric liver transplantation. PATIENTS AND METHODS From January 1, 1999 to October 1, 2005, we prospectively followed 138 pediatric liver transplant recipients under the age of 18 yr. Transplants were performed by a single surgeon. Of these, 16 children received an ABO-I graft (Tables 1-3). Recipient demographics, primary liver disease, biliary and vascular complications, retransplantation rate, patient and graft survival, rejection episodes, and follow-up time, were the variables analyzed from our database. All pediatric recipients were listed according to the United Network for Organ Sharing criteria. Mean follow-up was 3.25 yr. Blood type and cross-match with an irregular anti-
974 HEFFRON ET AL. body screen were performed on all children prior to transplantation. Isohemagglutinins titers were drawn pretransplantation only if irregular antibodies were found on initial screening. Isohemagglutinin titers against ABO antigens were monitored posttransplantation. According to our blood bank protocol, if clinically indicated, a positive A titer was considered to be related to the most common A1 antigen and A2 antigen tested. We defined indications for plasmapheresis as increased isohemagglutinin titers associated with allograft dysfunction, histological evidence of rejection, and/or clinical suspicion of antibody mediated response (i.e., hemolysis). Posttransplantation immunosuppression consisted of single-dose daclizumab, tacrolimus, mycophenolate mofetil, and methylprednisolone, and prednisone maintenance for 3 months. Intravenous methylprednisolone 20 mg/kg (up to a 1-gm maximum) was given at the time of the biliary anastomosis, followed by prednisone tapered to 0.3 mg/kg or a maximum of 20 mg/ day on postoperative day 7. Daclizumab was administered at 1 mg/kg and given once intravenously 12 to 24 hours posttransplantation. Mycophenolate mofetil was started orally at 30 mg/kg/day on postoperative day 1. Oral tacrolimus began on postoperative day 7. In the absence of rejection, steroids were discontinued within 12 months after transplantation. Exceptions from steroid withdrawal were children transplanted for autoimmune hepatitis or sclerosing cholangitis. If acute rejection occurred during the steroid wean, acute treatment was given with pulsed steroids, reinstitution of the oral prednisone, and increasing the calcineurin inhibitor trough to a therapeutic level. During both early and late phases of rejection, tacrolimus trough levels were maintained according to Table 1 and mycophenolate mofetil was administered at 15 mg/kg per day. Intravenous methylprednisolone pulse (10-20 mg/kg per dose) and increased tacrolimus trough level were used when antibody titers increased on serial isohemagglutinin assay, and/or evidence of increased indirect bilirubin, hematuria, positive direct antiglobulin test, or increased haptoglobin suggested hemolysis or antibody-mediated response. If the titers did not respond to methylprednisolone pulse and increased immunosuppression, we instituted plasmapheresis. Vascular anatomy of the transplanted liver was assessed by abdominal color doppler ultrasounds, performed intraoperatively and postoperatively for a minimum of 3 consecutive days. Liver biopsies were performed only when elevations in serial liver enzyme assay suggested allograft dysfunction. Standardized histologic grading of rejection was used. Clinical rejection is defined as an increase in the liver enzymes with corresponding liver biopsy or a clinical response to increasing tacrolimus and/or 1 bolus of intravenous steroids without liver biopsy. Immunohistochemistry to diagnose humoral rejection was not performed. Statistical analysis of the data was preformed by utilization of Student s t-test (2 tailed) for all continuous data when comparing 2 groups. When continuous data was compared within the same individuals, pair Student s t-test was utilized. Nominal data were compared by chi-squared and Fisher s exact test where appropriate, while survival was calculated by using actuarial data. Significance was considered P 0.05. RESULTS Of 138 liver transplants (in 121 children) 54% were whole grafts and 46% were partial. A total of 16 ABO-I liver allografts (11.6%) were transplanted; 56.3% were in recipients greater than 3 yr of age and 37.5% in recipients older than 13 yr of age at the time of transplantation. Demographic information was similar between both ABO-I and ABO-compatible groups (Table 2). The blood group pairings included blood type B donors to type A liver recipient (B to A, n 4),AtoB (n 2),AtoO(n 7), and AB to A (n 3). The primary diagnoses at the time of transplantation of the 16 children included: fulminant hepatic failure (n 5); postnecrotic cirrhosis (n 3); biliary atresia with acute decompensation (n 3); and Alagille syndrome (n 1); 4 of these 16 transplantations were retransplantations due to primary nonfunction with initial diagnosis of cryptogenic cirrhosis (n 1), biliary atresia (n 2), and hemangioendothelioma (n 1). All ABO-I recipients had acute liver decompensation on chronic disease requiring admission to the pediatric intensive care unit and met the United Network for Organ Sharing criteria for status 1, except for 1 patient with Alagille syndrome (Pediatric End-Stage Liver Disease score 6) who received an AB liver not utilized by other programs. Posttransplantation surgical complications in the ABO-compatible group included hepatic artery thrombosis (n 7; 5.8%) and biliary complications (n 13; 10.7%). There were 2 biliary and no vascular complications in 16 ABO-I recipients (Table 3). The 2 children with biliary complications developed anastomotic strictures within 12 days of transplantation. One child had a left lateral living donor allograft with a small donor bile duct connected to a Roux-en-Y limb. Balloon dilatation with percutaneous biliary stenting for 2 weeks corrected the structural defect in this child. No longer term (more than 3 yr later) bile duct sequelae have occurred. The second adolescent developed an anastomotic stricture on postoperative day 13 that was stented percutaneously. He remains with this stent 11 months later. Neither patient had sustained hypotension, hepatic artery thrombosis, or severe allograft rejection that would predispose to biliary strictures. The 1-yr actuarial patient survival for ABO-compatible vs. ABO-I grafts was 93.0% and 100%, whereas graft survival was 83.4% and 92.3%, respectively. The 16 children in the ABO-I group received a total of 18 liver allografts. Four patients received an emergent ABO-I liver transplant for primary nonfunction of the first graft. All ABO-I children were alive and well 1 yr after transplantation. One child died 471 days after transplantation in a local emergency room from necrotizing pneumonitis with a functioning graft. Six of the 16 (37.5%) ABO-I allograft children had 8
PEDIATRIC ABO LIVER TRANSPLANTATION 975 Figure 1. Comparison of Posttransplant Mean Total Bilirubin between ABO-compatible and ABO-I recipients. Figure 3. Comparison of Posttransplant Mean Alanine Aminotransferase between ABO-compatible and ABO-I recipients. Figure 2. Comparison of Posttransplant Mean Aspartate Aminotransferase between ABO-compatible and ABO-I recipients. rejection episodes, whereas 47 of the 121 (45.1%) children in the in the ABO-compatible group had 121 rejection episodes. The mean time to first rejection in the ABO-I group was 144 96 days posttransplantation. The mean time to first rejection was 307 325 days in the ABO-matched group. Liver enzymes for both groups are similar 2 yr after transplant (Figs. 1-3 Antibody titers were recorded only in the presence of irregular antibodies. Other than 1 patient who received plasmapheresis, there were no significantly increased anti-b titers within the ABO-I group (Fig. 4). The anti-a titers of those in the ABO-I group can be found in Figure 5. The anti-a titers show only minor variations within the ABO-I group and did not require intervention. Patient Synopsis The patients presented here are summarized in Table 4. Patient #1 (patient age: 0.75 yr; patient weight: 5kg; Figure 4. Anti-B titer values in Patient #3 pre- and postplasmapheresis. reason for transplant: primary nonfunction), after a successful liver transplant, presented to a local emergency department 471 days later in full cardiac arrest secondary to necrotizing streptococcal pneumonia with septicemia and died with a functioning graft. Patient #2 (17.5 yr, 87 kg, acute fulminate hepatic failure) had 1 episode of rejection confirmed by liver biopsy on postoperative day 41. Patient #3 (13 yr, 55 kg, primary nonfunction) had anti-b titers of 1:32 or less immediately posttransplantation. The patient s anti-b titer increased to 1:1,024 with hemolysis between day 5 and 6 and peaked at day 8 (1:4,096) with concomitant bilirubin at 8.4 milligrams/deciliter (mg/dl), aspartate aminotransferase of 180 Unit/Liter (U/L), alanine aminotransferase of 377 U/L, and gamma glutamyl transpeptidase of 245 U/L even after 1 gm of solumedrol IV on days 5 and 6. The direct antiglobulin test was positive. Plasmapheresis
976 HEFFRON ET AL. Figure 5. Pre- and Post-transplant Anti-A titers for ABO-I group. Four patients had immeasurable titers. was started on day 8 at 80% of total blood volume with resultant decrease in bilirubin to 5.7, aspartate aminotransferase to 76, alanine aminotransferase to 82, and gamma glutamyl transpeptidase to 98. A 4-fold increase in anti-b titers was noted, resulting in an increased total volume exchange to 150% starting on day 12. By day 17 all lab values had returned to normal and plasmapheresis was stopped. By day 21 the anti-b titer was 1:32 and bilirubin was 0.9, with complete resolution of the hemolysis. This patient had 2 biopsy-proven episodes of acute cellular rejection at 95 and 169 days posttransplantation, which were responsive to pulsed intravenous steroids (Fig. 4). Patient #4 (16.8 yr, 55 kg, subacute hepatic necrosis) was released from the hospital without complications and has had no additional problems. Patient #5 (14.0 yr, 51 kg, autoimmune) received a total of 3 transplants. This child s first ABO-matched transplant was lost to sepsis; the second allograft was an ABO-I allograft and was lost on postoperative day 19 due to complications following a percutaneous transhepatic cholangiogram. This patient was retransplanted with an ABO-compatible liver. Patient #6 (4.3 yr, 23 kg, acute fulminate hepatic failure) received steroid boluses for increased ABO-B antibody titers of 1:512 on postoperative days 14-16 and increased total bilirubin from 2.4 to 6.8 mg/dl. This patient responded well to intravenous steroids and did not require plasmapheresis. Patient #7 (1.4 yr, 10 kg, hemangioendothelioma) received an ABO-I allograft for primary nonfunction. This child had 1 rejection on postoperative day 311 and responded to intravenous steroids. Patient #8 (2 yr, 12.5 kg, acute fulminate hepatic failure) received a left-lateral segment from a living donor. This patient required a percutaneous transhepatic cholangiogram with biliary stent for stricture. Patient #9 (13.8 yr, 42.4 kg, acute fulminate hepatic failure) was released from the hospital without hepatic complications. Patient #10 (5.0 yr, 15.2 kg, Alagille syndrome) electively received an AB liver that would not have been used as there was no match nationally and it was turned down by all other centers. On postoperative day 98, the patient was diagnosed with clinical rejection, which was treated by increasing immunosuppression and 1 dose of intravenous steroids. Biopsy revealed mild rejection on postoperative day 197, which was successfully treated with 2 doses of intravenous steroids; the patient was discharged from the hospital. Patient #11 (0.5 yr, 6.3 kg, decompensated biliary atresia) was released from the hospital without complications and has had no interval hepatic complications. Patient #12 (0.35 yr, 6.3 kg, decompensated biliary atresia) was released from the hospital without complications and has had no additional problems. Patient #13 (0.25 yr, 5.66 kg, congenital cirrhosis) was released from the hospital without complications and has had no additional problems. Patient #14 (0.88 yr, 8.0 kg, biliary atresia) received her second allograft after her first liver decompensated on postoperative day 10. She was released from the hospital and has had no interval hepatic complications. Patient #15 (1.92 yr, 11.9 kg, biliary atresia) was released from the hospital without complications and has had no additional problems. Patient #16 (17.58 yr, 47.0 kg, acute fulminate hepatic failure due to hepatitis B) developed a bile duct stricture on postoperative day 15. The patient required a biliary stent, but otherwise has been without acute cellular rejection to date. DISCUSSION Many surgical and medical advances over the last 3 decades have dramatically improved patient and graft survival for children with end-stage liver disease requiring liver transplantation. However, recipient need for organs has outpaced donor availability and split livers have been underutilized in the United States. Historically, ABO-I grafts have been limited to emergency situations and have resulted in inferior patient and graft survival compared to ABO-compatible grafts. In 1986, Gordon et al. 8 reported decreased patient (45%) and graft (26%) survival for ABO-I liver transplants at 1 yr. In 1990, Gugenheim et al. 9 published a series of 234 liver transplants in which the 2-yr survival was 76% for ABO-compatible grafts and 30% for ABO-I grafts. He noted an increased incidence of biliary complications and vascular thrombosis as well as histological findings compatible with antibody mediated rejection. In 1995, Farges et al. 4,5 reported greatly decreased patient and graft survival with increased incidence of rejection and biliary and vascular complications in the ABO-I group. In a large series Bjoro et al. 10 found that patients who received an ABO-I liver had a much poorer graft survival rate (40 vs. 64% for ABO-compatible grafts) at 1 yr. Most recently, various groups have attempted to use
PEDIATRIC ABO LIVER TRANSPLANTATION 977 TABLE 4. ABO-I Patient Outcomes Age (yr) Gender Primary diagnosis UNOS status Transplant type Donor ABO Recipient ABO Complication Status 1 0.75 F Biliary atresia; 1 LLS B A Died-sepsis PNF #2 2 17.5 M FHF 1 Whole A B Rejection Alive 3 13 F PNC; PNF #2 1 Whole B A Rejection Alive 4 16.8 M PNC 1 Whole A O Alive 5 14 M PNC 1 Whole A B Hema Alive 6 4.3 M FHF 1 LLS A O Alive 7 1.4 M Tumor; PNF#2 1 LLS A O Rejection Alive 8 2 M FHF 1 LR-LLS B A Biliary stricture Alive 9 13.8 F FHF 1 Whole B A Alive 10 5 F Alagille syndrome PELD score 6 Whole AB A Rejection Alive 11 0.5 F Biliary atresia 1 Whole A O Alive 12 0.35 F Biliary atresia 1 LLS A O Alive 13 0.25 F Congenital 1 Whole AB A Alive cirrhosis 14 0.88 F Biliary atresia; 1 LLS AB A Alive PNF #2 15 1.92 F Biliary atresia 1 LL A O Alive 16 17.58 M FHF; Hepatitis B 1 Whole A O Biliary stricture Alive Abbreviations: LLS, left lateral segment; LL, left lobe; Hema, graft loss secondary to hematoma from invasive procedure; PNF, primary nonfunction; FHF, fulminant hepatic failure; PNC, postnecrotic cirrhosis. plasmapheresis, splenectomy, and novel immunosuppression regimens in an attempt to improve graft survival when crossing blood groups. Hanto et al. (2003) reported one and 5 yr patient and graft survival rates of 71.4% and 61.2%, respectively, in 14 adult patients, using a regimen of total plasma exchange, splenectomy and quadruple immunosuppression. Stegall 11 comments on the importance of this approach, with the lack of humoral rejection despite high anti-blood group titers and suggests extending this option to recipients with cancer or long wait times. In addition, he states that ABO-I liver transplantation leads to long-term graft loss secondary to vascular thrombosis and biliary complications. 11 In pediatric recipients, results in ABO-I liver transplantation have been mixed. Egawa et al. 2 notes that children less than 1 yr of age survive significantly longer with significantly less biliary complications than those greater than 1 yr of age. Morbidity and mortality appear to increase with age. As early as 1999, Varela- Fascinetto et al. 12 reported no significant difference in 10-yr patient or graft survival in children receiving 28 ABO-I grafts compared with those receiving 72 ABOcompatible or identical grafts. In addition, rejection, vascular thrombosis, and biliary complications were not significantly different. While 1 of 3 of those receiving an ABO-I graft had transient increases in antibody titers, only 3 of the 28 grafts required treatment. As recently as 2003, Szymczak et al. 13 reported decreased pediatric 1-yr patient and graft survival of 57% associated with increased late acute and chronic rejection in ABO-I grafts. In 1994, Tanaka et al. 14 reported 13 ABO-I pediatric liver transplants from living-related donors (mother or father) with preoperative plasmapheresis. Posttransplantation plasmapheresis was reserved for isoagglutinin titers greater than 64. A 77% patient and graft survival resulted with 3 of 13 patients dying from portal vein thrombosis, sepsis, and graft dysfunction. Takayma et al. 15 reported 3 cases of pediatric ABO-I liver transplantation with patient and graft survival of 100% at 1 yr. In our series, we used our standard immunosuppressive protocol without preoperative plasmapheresis or splenectomy. Despite a high percentage of higher-risk older age pediatric group (6/16 were older than 13 yr of age while 3/16 were older than 16 yr of age), a 1-yr actuarial patient survival of 100% and graft survival of 83% was achieved. The sole graft loss was secondary to intrahepatic hemorrhage after a percutaneous transhepatic cholangiogram that did not reveal a biliary stricture or leak. The sole patient loss was due to necrotizing pneumonitis 471 days after liver transplant. The patient died with a functioning graft. Hemolysis, responsive to plasmapheresis has been reported after solid organ transplantation. 16,17 In our series, the 1 patient with hemolysis that required plasmapheresis had an increased anti-b titer in the absence of measurable anti-a titer (Fig. 4), a positive direct antiglobulin test with clinical evidence of rejection, and high bilirubin. This was the only patient with high preformed antibodies prior to transplantation. Few mechanisms might be responsible for this immunemediated hemolysis: preformed antibodies, antibodies derived from donor passenger lymphocytes, or an associated minor mismatch type of immune reaction. 18 The complexity of posttransplantation hemolytic reactions is beyond the scope of this article. Of note, other than 2 bouts of rejection, this patient is alive and well with
978 HEFFRON ET AL. normal liver function tests more than 5 yr after transplant. We believe that interleukin-2 blockade with monoclonal antibodies (through CD25) may provide modulation of activated T and B cells early after transplantation and allow starting calcineurin inhibitors after increasingly longer intervals. However, we do not feel that this is as potent an immunosuppressive regimen compared to that reported that by other groups with less favorable results. Our study is at a later time frame than previous studies in these small, technically challenging patients. Many of the poor results previously attributed to crossing blood groups may have been secondary to the poor condition of the recipient and the increased incidence of vascular and biliary complications in this earlier time frame. Biliary and vascular complications have decreased in centers with experience. 14,19,20 Recently Kim et al. 21 have reported improved survival in children. Of importance, 100% survival at 1 yr, even in small patient groups of 16 children is better than the 93% survival reported in our larger patient group of ABO-compatible recipients. Limitations of this study include the relatively small number of patients and the difficulty of analyzing liver transplantation based on perioperative isohemagglutinin titer. Liver allografts are thought to be more resistant to immunological attacks and may actually provide a protective effect against antibody-mediated response. This may make the sole anti-ab titers more difficult to interpret. Moreover, presence of humoral rejection was not evaluated. In our experience, pediatric recipient patients with ABO-I grafts have long-term outcomes comparable or superior (100%) with ABO-compatible grafts (93%) without any difference in rejection or in vascular or biliary complications. Unlike other reports, recipient age did not appear to impact patient or graft survival. We have achieved optimal outcomes without modifying our standard immunosuppressive protocol and without preoperative plasmapheresis or perioperative splenectomy. Crossing blood groups has played a significant role in avoiding pediatric waitlist mortality in emergent patients at our center. This supports the concept of using ABO-I grafts in an elective setting associated with split and living donor liver transplants. Further studies are necessary to confirm our conclusions. REFERENCES 1. Rydberg L. ABO-incompatibility in solid organ transplantation. Transfus Med 2001;11:325-342. 2. Egawa H, Oike F, Buhler L, Shapiro AM, Minamiguchi S, Haga H, et al. Impact of recipient age on outcome of ABOincompatible living-donor liver transplantation. Transplantation 2004;77:403-411. 3. Fang WC, Saltzman J, Rososhansky S, Szabo G, Heard SO, Banner B, et al. Acceptance on an ABO-incompatible mismatched (AB to 0) liver allograft with the use of Dacluzamib and mycophenolate mofetil. Liver Transpl 2000; 6:497-500. 4. Farges O, Kalil AN, Samuel D, Saliba F, Arulnaden JL, Debat P, et al. The use of ABO-incompatible grafts in liver transplantation: A life-saving procedure in highly selected patients. Transplantation 1995;59:1124-1133. 5. Farges O, Nocci Kalil A, Samuel D, Arulnaden JL, Bismuth A, Castaing D, et al. Long-term results of ABO-incompatible liver transplantation. Transplant Proc 1995;27:1701-1702. 6. Hanto D, Fecteasu A, Alonso M, Valente JF, Whiting JF. ABO-Incompatible liver transplantation with no immunological graft losses using total plasma exchange, splenectomy, and quadruple immunosuppression: evidence for accommodation. Liver Transpl 2003;9:22-30. 7. Monteiro I, McLaughlin L, Fisher A, de la Torre AN, Koneru B. Rituximab with plasmapheresis and splenectomy in ABOincompatible liver transplantation. Transplantation 2003; 76:1648-1649. 8. Gordon RD, Iwatsuki S, Esquivel CO, Tzakis A, Todo S, Starzl TE. Liver transplantation across ABO blood groups. Surgery 1986;100:342-348. 9. Gugenheim J, Samuel D, Reynes M, Bismuth H. Liver transplantation across ABO blood group barriers. Lancet 1990;336:519-523. 10. Bjoro K, Ericzon B, Kirkegaard P. highly urgent liver transplantation: possible impact of donor-recipient ABO matching on the outcome after transplantation. Transplantation 2003;75:347-353. 11. Stegall M. ABO-incompatible liver transplant: is it justifiable? Liver Transpl 2003;9:31. 12. Varela-Fascinetto G, Treacy SJ, Lillehei CW, Jonas MM, Lund DP, Kevy SV, et al. Long-term results in pediatric ABOincompatible liver, Transplant Proc 1999;31:467-468. 13. Szymczak M, Kalicinski P, Kaminski A, Ismail H, Drewniak T, Nachulewicz P, et al. Liver transplantation across ABO blood groups in children. Transplant Proc 2003;35:2273-2274. 14. Tanaka A, Tanaka K, Kitai T, Yanabu N, Tokuka A, Sato B, et al. Living related liver transplantation across ABO blood groups. Transplantation 1994;58:548-553. 15. Takayama J, Ohkohchi N, Oikawa K, Asakura T, Kawagishi N, Kikuchi H, et al. Living related liver transplantation in patients with ABO incompatibility. Transplant Proc 1998;30:3504-3506. 16. Gregoire JR. Immune hemolytic anemia after renal transplantation secondary to ABO-minor mismatch between the donor and recipient. J Am Soc Nephrol 1993;4:1122-1126. 17. Hareuveni M, Merchav H, Austerlitz N, Rahimi-Levene N, Ben-Tal O. Donor anti-jk causing hemolysis in a liver transplant recipient. Transfusion 2002;42:363-367. 18. Panaro F, DeChristopher PJ, Rondelli D, Testa G, Sankary H, Popescu M, et al. Severe hemolytic anemia due to passenger lymphocytes after living-related small bowel transplant (Case Report). Clin Transplant 2004;18:332-335. 19. Heffron TG, Welch D, Pillen T, Fasola C, Redd D, Smallwood GA, et al. Low incidence of hepatic artery thrombosis after pediatric liver transplantation without the use of intraoperative microscope or parenteral anticoagulation. Pediatr Transplant 2005;9:486-490. 20. Guarrera JV, Sinha P, Lobritto SJ, Brown RS Jr, Kinkhabwala M, Emond JC. Microvascular hepatic artery anastomosis in pediatric segmental liver transplantation: microscope vs loupe. Transpl Int 2004;17:585-588. 21. Kim JS, Groteluschen R, Mueller T, Ganschow R, Bicak T, Wilms C. Pediatric transplantation: the Hamburg experience. Transplantation 2005;79:1206-1209.