ABO Incompatible Renal Transplantation: A Paradigm Ready for Broad Implementation : Transplantation

Journal Logo

Clinical and Translational Research

ABO Incompatible Renal Transplantation: A Paradigm Ready for Broad Implementation

Montgomery, Robert A.1,6; Locke, Jayme E.1; King, Karen E.2; Segev, Dorry L.1; Warren, Daniel S.1; Kraus, Edward S.3; Cooper, Matthew4; Simpkins, Christopher E.1; Singer, Andrew L.1; Stewart, Zoe A.1; Melancon, J Keith1; Ratner, Lloyd5; Zachary, Andrea A.3; Haas, Mark2

Author Information
Transplantation 87(8):p 1246-1255, April 27, 2009. | DOI: 10.1097/TP.0b013e31819f2024
  • Free


Despite concerted efforts to increase public awareness about the deepening crisis in organ availability, only modest gains in deceased organ donation have been made during the past decade. During the same period live donation has nearly tripled, indicating that significant progress in closing the gap between organ supply and need will likely come from increasing live donation. One of the most significant barriers to expansion of live kidney donation is ABO incompatibility (ABOi). Based on blood group frequencies in the United States, there is a 36% probability that any two individuals will be incompatible, eliminating up to one third of potential live donors.

Protocols for crossing the blood group barrier using heavy immunosuppression and splenectomy were developed in the 1980s in Europe, Japan, and the United States (1, 2). Early graft loss and infectious complications remained high, preventing the expansion of these protocols around the world (3, 4). We now show that using a protocol involving a brief escalation in immunosuppression consisting of plasmapheresis (PP) and low-dose IVIg, kidneys can be transplanted across any blood group barrier without significant risk of hyperacute rejection, antibody-mediated rejection (AMR), or serious infection. Furthermore, graft survival rates are comparable with compatible live donor grafts.

Most hospitals that perform dialysis and kidney transplantation are capable of performing apheresis and measuring isohemagglutinin titers. Any of these programs should be able to implement the protocol described in this study. In 2006, there were 6432 live donor transplants performed in the United States (OPTN data as of June 16, 2008). Based on the estimates from mathematical simulations, if ABOi was eliminated as a barrier to renal transplantation an additional 1500 transplants could be performed annually (5).


Patient Selection and Protocol

We report the results of a series of 60 consecutive ABOi transplants performed between 1999 and 2007. After informed consent, all patients were treated with a standardized protocol approved by the Johns Hopkins Institutional Review Board that included pre- and posttransplant PP/cytomegalovirus intravenous immunoglobulin (CMVIg) and quadruple, sequential immunosuppression. In addition to this standard protocol, we initially performed a splenectomy at the time of transplant, this was later replaced by a single-dose anti-CD20 administered the night before transplantation, and more recently both interventions have been eliminated.

Plasmapheresis and Intravenous Immunoglobulin

Plasmapheresis was performed using a COBE Spectra (COBE 2997, COBE BCT, Lakewood, CO) centrifuge-driven cell separator as previously described (6). Immediately after each PP treatment, patients received 100 mg/kg of CMVIg (IVIg) (CytoGam, Medimmune, Gaithersburg, MD). The number of treatments was estimated based on the starting isohemagglutinin titer from the algorithm presented in Table 1. In all cases, the titer was reduced to less than or equal to 16 before the transplant. At least two posttransplant protocol PP/IVIg sessions were performed for all patients. A greater number of posttransplant treatments were delivered to patients who had high starting titers, antibody rebound, or AMR.

The number of planned pre- and posttransplant PP/IVIg treatments correlate with the starting isohemagglutinin titer


All patients received tacrolimus and mycophenolate mofetil beginning on the first day of their pretransplant PP/IVIg as previously reported (6). All patients were induced with daclizumab (2 mg/kg initial dose and 1 mg/kg every 2 weeks for 5 total doses). Steroids were used perioperatively, including dexamethasone 100 mg intraoperatively and 25 mg every 6 hr for six doses postoperatively, followed by prednisone 30 mg/day. Prednisone was reduced to 20 mg/day when tacrolimus was in the therapeutic range (8–12 ng/dL), and tapered thereafter to 5 or 10 mg/day.

Antibody Titer

Isohemagglutinin titers were determined using standard serological techniques (tube test) and then converted to the anti-human globulin test phase (7). The titer endpoint was considered to be the reciprocal of the highest dilution demonstrating agglutination. Human leukocyte antigen (HLA) crossmatch techniques including antiglobulin-enhanced lymphocytotoxicity (anti-human globulin-complement dependent cytotoxicity [CDC]) with T cells, one-wash CDC (1wCDC) with B cells, and flow cytometry with T and B cells were performed as previously described (8). A crossmatch was considered positive when (1) cell death with the patient's serum was more than or equal to 10% above background in cytotoxicity assays or (2) the ratio of median channel obtained with a patient's serum was more than or equal to 3 times that obtained with negative control serum in a flow crossmatch. When present, anti-HLA class I and II donor specific antibody (DSA) were identified by (1) ELISA using soluble HLA antigens as targets, specifically, GTI Quik-ID and GTI Quick-ID class II (GTI Diagnostics, Brookfield, WI) were used, or (2) using a color-coded multianalyte bead immunoassay on the Luminex platform, which uses beads coated with soluble HLA molecules as targets (Life Screen, Life Match I for Class I and Life Match ID for class II, Tepnel Lifecodes, Stanford, CT, and Single Antigen Beads, One Lambda, Canoga Park, CA). The multianalyte immunoassays were performed according to the manufacturer's instructions as reported elsewhere (8, 9). Isohemagglutinin and HLA DSA titers were monitored before and after each PP/IVIg treatment, 72 hr after the last PP/IVIg treatment, weekly for the first month, and at 2, 3, 6, and 12 months. A rise in titer more than or equal to 32 triggered a biopsy, and if AMR was diagnosed histologically PP/IVIg was initiated.

Histopathologic Analysis of Biopsies and Diagnosis of Rejection

Biopsies were obtained when clinically indicated or by protocol. All biopsies were analyzed by routine light microscopy using hematoxylin-eosin, periodic acid-Schiff, methenamine silver, and Masson's trichrome stains. For all biopsies, a small (3–4 mm) piece of cortical or medullary tissue was also taken for C4d staining, which was performed by indirect immunofluorescence on cryostat sections using a mouse monoclonal anti-human C4d antibody (Quidel, San Diego, CA) at 1:40 dilution, followed by fluorescein isothiocyanate-conjugated goat anti-mouse IgG (Jackson Immunoresearch Laboratories, West Grove, PA). Staining for C4d in peritubular capillaries (PTC) was graded as diffuse (positive in >50% of specimen), focal (>10% of specimen, but <50%), or absent. Clinical and subclinical acute cellular rejections were graded according to Banff '97 criteria (10). AMR (clinical or subclinical) was diagnosed as being present when all of the following three criteria were met (11): (1) positive (diffuse or focal) C4d staining in PTC; (2) at least one of the following: marginated leukocytes (neutrophils or mononuclear cells) in at least 10% of cortical PTC, moderate or severe glomerulitis (Banff g score >2), or transmural necrotizing arteritis; and (3) presence of circulating anti-donor antibodies, including anti-blood group antibodies. In addition, for each biopsy the following chronic indices were semiquantitatively evaluated (0–3 scale indicating increasing severity) according to Banff'97 criteria (10): transplant glomerulopathy (cg), interstitial fibrosis (ci), tubular atrophy (ct), arterial intimal fibrosis (cv), and a chronic SUM score (cg+ci+ct+cv) consisting of the sum of the semiquantitative scores for these indices was determined.

Treatment of Rejection

Cellular rejection episodes (both clinical and subclinical) with Banff'97 grades of 1A or 1B were treated with a 3-day pulse of dexamethasone 100 mg/day followed by a taper. If the Banff score was 2A, 2B, or 3, patients received a 7-day course of antithymocyte globulin. AMR (both clinical and subclinical) was treated with reinitiation of PP/IVIg until clinical improvement was achieved or AMR resolved histologically (12, 13).

Protocol Biopsies

Prereperfusion and 1 hr postreperfusion biopsies were obtained at the time of surgery. Surveillance biopsies were performed at 1, 3, 6, and 12 months after transplantation. Subclinical rejection was treated and followed up by an additional surveillance biopsy 1 month later.


Summary statistics including mean, median, and range were obtained. Kaplan-Meier methodology was used to estimate patient and death-censored graft survival at 1, 3, and 5 years. All analyses were performed using STATA 10.0 (Stata Corp., College Station, TX).



Between 1999 and 2007, 60 patients received an ABOi kidney transplant through the Johns Hopkins Incompatible Kidney Transplant Program. All patients who were cleared for transplantation were entered into the cohort and none were excluded due to the magnitude of their ABO titers. All patients were informed about the option of kidney paired donation (KPD) and several patients received an ABOi graft as part of a KPD transplant. Desensitization was initiated in 61 patients during the study period. One patient had a starting titer 4096 and as we approached the transplant using daily PP/IVIg and AB type fresh frozen plasma (FFP) he had paradoxical rebound in his isohemagglutinins during each treatment. Recipients ranged in age from 20 to 73 years, 52% were women, and 73% were white. Seventeen recipients had a history of previous transplant, and 10 patients were considered sensitized with a peak panel-reactive antibody (PRA) greater than 80%. Eleven patients were transplanted across both a blood-type barrier and a positive crossmatch (+XM), and six patients received an ABOi kidney transplant from an HLA identical sibling (Table 2). Thirty-two transplants were performed using kidneys from donors with blood-type A1 (Fig. 1A). The median starting isohemagglutinin titer was 128 (Fig. 1B), and the median number of pre- and posttransplant PP treatments was 6 and 5, respectively (Table 3).

Demographic characteristics of blood-type incompatible (ABOi) recipients transplanted through the Johns Hopkins Incompatible Kidney Transplant Program between 1999 and 2007
Distribution of donor blood types and starting isohemagglutinin titers among ABO incompatible recipients transplanted through the Johns Hopkins Incompatible Kidney Transplant Program between 1999 and 2007. The donor and recipient blood types are listed in Panel A. Panel B shows the number of patients with each starting titer. The median starting isohemagglutinin titer was 128. All titers represent anti-human globulin (AHG) assays.
Summary of isohemagglutinin titers and desensitization treatments required before and after ABOi kidney transplantation

Patient and Allograft Survival

There were three patient deaths in the series. All three patients died with functioning grafts, and the causes of death included West Nile encephalitis (serum creatinine [SCr] 1.2 mg/dL), sudden cardiac death (SCr 1.2 mg/dL), and metastatic liver cancer (SCr 1.6 mg/dL). The West Nile encephalitis was likely acquired from an FFP transfusion and represents the only major infection in the cohort (14). Patient survival at 1, 3, and 5 years was 96.3%, 96.3%, and 89.4%, respectively (Table 4).

Patient and graft survival among 60 ABOi kidney transplant recipients transplanted at the Johns Hopkins Hospital between 1999 and 2007

Death-censored graft survival at 1, 3, and 5 years was 98.3%, 92.9%, and 88.7%, respectively (Table 4). There were a total of four graft losses. Only one graft was lost in the first year. This graft loss was secondary to severe thrombotic microangiopathy, not related to an increase in DSA or isohemagglutinin. The other three graft losses were secondary to recurrent disease (n=2) or patient noncompliance (n=1). In addition, four patients had BK viremia including one of the patients who lost their graft secondary to recurrent disease. The other three have cleared their viremia with a reduction in immunosuppression. The mean current creatinine clearance (CrCl) for the cohort is 61.2 mL/min at a mean follow-up of 29.1 months (Table 5).

Renal Function among ABOi kidney transplant recipients transplanted at Johns Hopkins Hospital between 1999 and 2007

There were no cases of hyperacute rejection or graft losses from AMR. All kidneys functioned promptly and none of the patients required dialysis during the first week posttransplant. There was one case of severe AMR treated successfully with urgent rescue splenectomy and PP/IVIg. Eleven episodes of AMR occurred among 10 patients and four of these rejection episodes also had histologic evidence of cellular rejection (Table 6). Among patients who developed AMR, there was no significant difference in starting (pretreatment) isohemagglutinin titers. ABOi-only patients who experienced AMR showed a concomitant rise in isohemagglutinin titer (32–256). Two patients who received an ABOi and +XM transplant experienced a significant AMR despite low isohemagglutinin titers (≤16) but showed a rise in HLA DSA. There were seven episodes of subclinical AMR documented on protocol biopsies. There were 11 clinically apparent cellular rejections and 10 subclinical cellular rejection episodes (Table 6).

Incidences of acute rejection episodes among ABO incompatible kidney transplant recipients transplanted at Johns Hopkins Hospital between 1999 and 2007

Table 7 shows mean chronic SUM scores (cg+ci+ ct+cv) for the 53 patients who had protocol biopsies performed at 6 or 12 months; each patient also had a protocol biopsy at 1 month posttransplantation. Notably, the mean chronic SUM score on 1-month protocol biopsies was 1.3 and was mainly caused by non-zero cv scores in 38 biopsies, most likely reflecting arteriosclerosis present in the donor kidney. On both 6- and 12-month biopsies, mean chronic SUM scores were significantly higher in patients who developed AMR (clinical or subclinical) than in those who did not. Only 9 of 36 patients who had a 12-month protocol biopsy showed evidence of transplant glomerulopathy (cg≥1), and five of nine experienced AMR and two others were combined ABOi and +XM grafts. Mean cg scores at 12 months were significantly higher in patients with AMR than without AMR (0.8±0.7 vs. 0.1±0.4, P=0.005 by Wilcoxon rank sum test), in first-era patients than in second-era patients (0.5±0.6 vs. 0.1±0.2, P=0.008), and in ABOi-only grafts compared with ABOi and +XM grafts (0.8±0.9 vs. 0.2±0.4, P=0.007). However, 6- and 12-month SUM scores and increases in these scores from 1 month to 6 and 12 months were not significantly different in first and second era patients or in those receiving ABOi-only versus ABOi and +XM grafts (P>0.10 in all cases).

A comparison of chronic SUM scores among patients who did or did not experience AMR

Incompatible Kidney Transplant Program ABOi Protocol Evolution

With the recognition from our early clinical experience that accommodation was rapid after ABOi transplantation and AMR was relatively uncommon, tended to be mild, and occurred in the first 2 weeks, we challenged the historic paradigm that splenectomy was a necessary prerequisite for successful engraftment. First, we eliminated the splenectomy and substituted anti-CD20 (15, 16). We hypothesized that tight control of isohemagglutinin titers with PP/IVIg during this early window of graft vulnerability would allow us to eliminate anti-CD20 as well.

First Era—Splenectomy or Anti-CD20 or Both

A total of 14 patients underwent splenectomy before transplantation. Three patients received anti-CD20 therapy in addition to splenectomy before transplant. Fifteen patients were treated with anti-CD20 instead of splenectomy. The mean follow-up on this cohort is 42.0 months (range, 0.3–101 months). The donor blood types were distributed as follows: 19 A1, 4 A2, 3 B, and 6 AB. The median starting titer for this group was 128 (range, 4–1024) and the median number of pre- and posttransplant treatments were 6 (range, 2–31) and 4 (range 2–12), respectively. The median current SCr in this group is 1.3 mg/dL (range, 0.8–1.8 mg/dL), which corresponds to a median CrCl of 63.2 mL/min (range, 39.3–96.9 mL/min). There were six clinical and four subclinical cases of AMR in this group. Eleven patients experienced previous transplants. The mean 1-year delta SUM score was 2.4.

Second Era—No Splenectomy and No Anti-CD20

Twenty-eight of the 60 patients have been desensitized in the absence of both splenectomy and anti-CD20 and this is our current standard practice. The median starting titer for this cohort was 128 (range, 16–512), which is the same as the first era. There were five clinical and three subclinical cases of AMR. Three patients had clinical cell-mediated rejection (CMR). The median SCr and CrCl are 1.3 mg/dL (range, 0.7–2.7 mg/dL) and 60.5 mL/min (range, 32.6–167.5 mL/min), respectively. The mean 1-year delta SUM score was 1.7. The only statistically significant difference between the two eras was the length of follow-up, which was 21 months (range, 0.4–43.5 months) in era 2.

HLA Identical and ABOi

Among patients with broad HLA sensitization crossing the blood group barrier with a kidney from an HLA identical sibling can be a good solution. Five sensitized (PRA>80) patients presented to us with ABOi, HLA identical siblings (Table 2). In addition, one patient in this group had a low degree of sensitization (PRA 6). The median peak PRA for this cohort was 92%, making their prospects for transplantation poor. All six patients had A1 donors. Two patients underwent splenectomy, one underwent both splenectomy and received anti-CD20, and the other three had neither splenectomy nor anti-CD20. The median initial titer was 128 (range, 128–512). There were two episodes of clinical AMR and one episode of clinical CMR in this cohort. The one early death in our series occurred in this group and was due to West Nile Encephalitis (14). The other five grafts are functioning with a mean current CrCl of 58.1 mL/min (range, 51.5–86.4 mL/min). The mean delta SUM score of this group at 1 year was 1.8.

Combined ABOi and Positive Crossmatch

Eleven patients had combined HLA and ABOi. These patients were felt to be at significant risk of AMR and received splenectomy alone (n=3) or in combination with anti-CD20 (n=1). Six of the remaining patients received anti-CD20 alone, and one patient was transplanted without splenectomy or anti-CD20. The median age of this group was 36 years and 9 of 11 were women (Table 2). These patients had multiple risk factors for sensitization including previous transplants (n=6), pregnancies (n=5), and blood transfusions (n=6). The median follow-up for this group is 15.9 months (range, 0.3–64.4 months). Despite this being a high-risk group, there were only four episodes of clinical AMR, two occurring in the same patient. Two of the episodes also had features of CMR. Ten grafts are currently functioning and the median current CrCl is 65.1 mL/min (range, 39.3–96.9 mL/min). The mean delta SUM score of this group at 1 year was 3.3.


In this study, we showed that blood-type incompatibility, once believed to be a formidable barrier to renal transplantation, can be crossed with excellent results and only a brief escalation in immunosuppression. These findings challenge the paradigm of the last several decades, namely, that ABOi transplants require intensive, long-lasting preconditioning followed by heavy maintenance immunosuppression to produce results that in the short term are inferior to blood-type compatible transplants. Further, our findings suggest that ABOi transplants can be accomplished with a low risk of AMR and graft loss without the need for splenectomy or B-cell ablative therapy that might place patients at increased risk of infection or malignancies. Using a short course of PP and low-dose IVIg with standard maintenance immunosuppression, the death-censored graft survival of 60 consecutive ABOi kidney transplants at 1, 3, and 5 years was 98.3%, 92.9%, and 88.7%, respectively. These results compare favorably with data from the United Network of Organ Sharing national database that reports a 1-, 3-, and 5-year graft survival for live donor compatible kidney transplants of 95.1%, 87.8%, and 79.7%, respectively (based on OPTN data as of June 16, 2008). At median follow-up of 29 months, the current median CrCl of the cohort is 61.2 mL/min, portending good long-term allograft function.

The study details the evolution of our desensitization therapy from the traditional heavy immunosuppression regimens used when we began our series to a protocol that would be predicted to gain wider acceptance by the transplant community. We have demonstrated that eliminating splenectomy and anti-CD20 does not change the risk of AMR or graft loss, which is a significant finding as historically the need for these interventions have raised financial and safety concerns and has served as a potent disincentive for other centers to adopt ABOi protocols. In addition, patients who receive a combined positive crossmatch and ABOi kidney transplant as well as those sensitized patients with ABOi, HLA identical siblings have excellent outcomes after desensitization. Because both isohemagglutinin monitoring and PP services are available at centers that perform kidney transplantation, we believe that our protocol could be implemented at most transplant centers throughout the United States. We have previously estimated that this would translate to a 23% annual increase in live donor transplants (5).

There are currently three options available to a patient whose only willing donor is ABOi: waiting on the deceased donor list, KPD, or desensitization. The average waiting time in the United States for an O recipient is 5.1 years (based on OPTN data, accessed June 16, 2008). Patients who receive a transplant rather than remaining on dialysis on average double their life expectancy (17). Kidneys from live donors have a predicted half-life that is twice that of a deceased donor organ (18). Waiting on the deceased donor list when the possibility of a live donor exists is a poor choice in most instances.

In KPD, incompatible pairs are matched and organs are exchanged between the pairs to produce compatible transplants. Although avoiding the incompatibility through KPD is perhaps the best alternative, only 31% of ABOi pairs have been predicted to find an exchange pair under optimal conditions in a national KPD match run (19). The most common ABOi is an A donor and an O recipient. The O recipient will need an O donor to receive a compatible kidney. O donors are relatively rare in KPD pools. This results in a less than 15% match rate for the most common ABOi combination. In contrast, the A/B or B/A donor/recipient combinations have approximately a 75% match rate. However, efforts by consortia and regional matching programs have had a limited impact during the past 7 years (20). If this changes and KPD reaches its potential nationally, it should be offered preferentially for certain blood-type combinations with the highest match rate. The two modalities can be used together to find the best transplant solution for each donor/recipient phenotype.

In the initial published Japanese series, there was approximately a 10% immediate graft loss due to AMR and 20% during the first year posttransplant (21, 22). Clearly, something has changed in recent years as we and others in Japan, Sweden, and the United States are showing dramatically improved results (23–26). In these series, different methods of antibody depletion have been used including standard PP, double filtration PP, and immunoadsorption. The advantage of immunoadsorption is that it is not believed to deplete total IgG or coagulation factors. However, it does not remove complement either and there may be some advantage to PP in this regard. The Japanese groups do not use protocol posttransplant PP and instead monitor isohemagglutinin levels and treat patients who show an early rise in titers. They also anticoagulate their patients posttransplant. The Swedish and US groups have not found this to be necessary. Possible explanations for improved results in recent times include routine or more careful maintenance in anti-blood group antibodies at or below a titer of 16 for the first few weeks until accommodation has occurred, the inclusion of IVIg in the protocols (although not consistently used in Japan), or the use of mycophenolate mofetil as part of the preconditioning regimen that has anti-B-cell properties. Other groups believe that the inclusion of anti-CD20 in protocols has resulted in improved results although our data suggest otherwise. It is also possible that less immunosuppression has in some way enabled accommodation or reduced morbidity and mortality.

Contrary to previous reports, we have not observed any increase in AMR among patients with high starting isohemagglutinin titers, A1 donors, or AB donors into O recipients. It is also interesting that in all cases isohemagglutinins continue to be present in the recipient's blood (mean titer, 8), albeit at a much lower level than their predesensitization value (mean titer, 128). Rarely, we have observed early antibody rebound that is not associated with AMR (1 patient's titer rose to 256 in the first few weeks after transplant). The phenotype and mechanism of accommodation in ABOi grafts remains poorly understood.

We have performed protocol biopsies on our ABOi patients posttransplant. We have previously described the presence of diffuse C4d deposition in ABOi grafts and shown that in the absence of histologic findings of AMR, C4d is not associated with the development of scarring (27, 28), in contrast to other settings where complement activation can be a marker for antibody-mediated injury. Thus, we evaluated 6- and 12-month protocol biopsies for the presence of chronic changes that might predict a truncated graft half-life. Similar to the findings of Gloor et al. (29), we found a low incidence of transplant glomerulopathy (TG) on 1-year protocol biopsies of ABOi-only grafts (5/30, 17%). Three of the five ABOi-only grafts that developed TG had prior clinical or subclinical AMR. The fraction of 1-year biopsies showing TG was higher among combined ABOi and +XM grafts (4/6, 67%). Two of the four kidneys showing TG had a history of AMR. Indeed the mean cg score at 1 year was significantly higher in patients with AMR than without AMR, as was the mean chronic SUM scores at 12 (and 6) months posttransplantation.

The overall mean chronic SUM score at 1 year was 3.3. Other groups have reported lower chronic SUM scores (CSS) from protocol biopsies of ABOi (29) and conventional transplants (30–32). However, their baseline donor biopsies had mean chronic SUM scores in the range of 0.2 to 0.6, whereas the mean chronic SUM score on our 1-month protocol biopsies was 1.3 and was primarily due to the cv component that likely represents donor disease (33). The mean increase in SUM score between 1 and 12 months in our study was only 2.1, comparable with results of Gloor et al. (29) in ABOi-only grafts. Furthermore, when grafts without AMR are considered, the mean increase in SUM score between 1 and 12 months in our patients was only 1.5, similar to the findings in conventional grafts and suggesting that the increase in SUM scores in members of our cohort not developing AMR is unrelated to the blood-type incompatibility or chronic antibody-mediated damage.

This study suggests that with proper pretransplant preparation severe AMR is rare (< 2%). If it does occur these kidneys can often be salvaged with splenectomy (34, 35). The use of routine splenectomy or anti-CD20 does not seem to be essential for engraftment. The results of this study and the work of others support a reevaluation of the paradigm that blood-type incompatibility is a significant barrier to renal transplantation and signal a renaissance in ABOi renal transplantation.


The authors thank Vanessa Collins for her contribution to data collection and management.


1. Alexandre GP, Squifflet JP, De Bruyere M, et al. Present experiences in a series of 26 ABO-incompatible living donor renal allografts. Transplant Proc 1987; 19: 4538.
2. Tanabe K, Takahashi K, Sonda K, et al. Long-term results of ABO-incompatible living kidney transplantation: A single-center experience. Transplantation 1998; 65: 224.
3. Hume DM, Merrill JP, Miller BF, et al. Experiences with renal homotransplantation in the human: Report of nine cases. J Clin Invest 1955; 34: 327.
4. Starzl TE, Marchioro TL, Holmes JH, et al. Renal homografts in patients with major donor-recipient blood group incompatibilities. Surgery 1964; 55: 195.
5. Segev DL, Gentry SE, Warren DS, et al. Kidney paired donation and optimizing the use of live donor organs. JAMA 2005; 293: 1883.
6. Segev DL, Simpkins CE, Warren DS, et al. ABO incompatible high-titer renal transplantation without splenectomy or anti-CD20 treatment. Am J Transplant 2005; 5: 2570.
7. Technical Manual—American Association of Blood Banks, [ed. 14]. Bethesda, AABB 2002.
8. Lucas DP, Paparounis ML, Myers L, et al. Detection of HLA class I-specific antibodies by the QuikScreen enzyme-linked immunosorbent assay. Clin Diagn Lab Immunol 1997; 4: 252.
9. Zachary AA, Ratner LE, Graziani JA, et al. Characterization of HLA class I specific antibodies by ELISA using solubilized antigen targets. II. Clinical relevance. Hum Immunol 2001; 62: 236.
10. Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: 713.
11. Racusen LC, Colvin RB, Solez K, et al. Antibody-mediated rejection criteria—An addition to the Banff 97 classification of renal allograft rejection. Am J Transplant 2003; 3: 708.
12. Montgomery RA, Cooper M, Kraus E, et al. Renal transplantation at the Johns Hopkins Comprehensive Transplant Center. Clin Transpl 2003:199.
13. Montgomery RA, Zachary AA. Transplanting patients with a positive donor-specific crossmatch: A single center's perspective. Pediatr Transplant 2004; 8: 535.
14. Shepherd JC, Subramanian A, Montgomery RA, et al. West Nile virus encephalitis in a kidney transplant recipient. Am J Transplant 2004; 4: 830.
15. Sonnenday CJ, Warren DS, Cooper M, et al. Plasmapheresis, CMV hyperimmune globulin, and anti-CD20 allow ABO-incompatible renal transplantation without splenectomy. Am J Transplant 2004; 4: 1315.
16. Tyden G, Kumlien G, Fehrman I. Successful ABO-incompatible kidney transplantations without splenectomy using antigen-specific immunoadsorption and rituximab. Transplantation 2003; 76: 730.
17. Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999; 341: 1725.
18. Hariharan S, Johnson CP, Bresnahan BA, et al. Improved graft survival after renal transplantation in the United States, 1988 to 1996. N Engl J Med 2000; 342: 605.
19. Gentry SE, Segev DL, Montgomery RA. A comparison of populations served by kidney paired donation and list paired donation. Am J Transplant 2005; 5: 1914.
20. Segev DL, Kucirka LM, Gentry SE, et al. Utilization and outcomes of kidney paired donation in the United States. Transplantation 2008; 86: 502.
21. Takahashi K, Tanabe K, Ooba S, et al. Prophylactic use of a new immunosuppressive agent, deoxyspergualin, in patients with kidney transplantation from ABO-incompatible or preformed antibody-positive donors. Transplant Proc 1991; 23(1 Pt 2): 1078.
22. Takahasi K, Ota K, Ooba S, et al. Use of a novel immunosuppressive agent, deoxyspergualin, in a case of ABO-incompatible kidney transplantation. Clin Transpl 1989:306.
23. Genberg H, Kumlien G, Wennberg L, et al. ABO-incompatible kidney transplantation using antigen-specific immunoadsorption and rituximab: A 3-year follow-up. Transplantation 2008; 85: 1745.
24. Ishida H, Miyamoto N, Shirakawa H, et al. Evaluation of immunosuppressive regimens in ABO-incompatible living kidney transplantation—Single center analysis. Am J Transplant 2007; 7: 825.
25. Takahashi K, Saito K, Takahara S, et al. Excellent long-term outcome of ABO-incompatible living donor kidney transplantation in Japan. Am J Transplant 2004; 4: 1089.
26. Gloor J, Stegall M. ABO-incompatible kidney transplantation with and without splenectomy. Transplantation 2006; 82: 720.
27. Haas M, Rahman MH, Racusen LC, et al. C4d and C3d staining in biopsies of ABO- and HLA-incompatible renal allografts: correlation with histologic findings. Am J Transplant 2006; 6: 1829.
28. Haas M, Segev DL, Racusen LC, et al. C4d deposition without rejection correlates with reduced early scarring in ABO-incompatible renal allografts. J Am Soc Nephrol 2009; 20: 197.
29. Gloor JM, Cosio FG, Rea DJ, et al. Histologic findings one year after positive crossmatch or ABO blood group incompatible living donor kidney transplantation. Am J Transplant 2006; 6: 1841.
30. Kee TY, Chapman JR, O'Connell PJ, et al. Treatment of subclinical rejection diagnosed by protocol biopsy of kidney transplants. Transplantation 2006; 82: 36.
31. Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronic allograft nephropathy. N Engl J Med 2003; 349: 2326.
32. Mengel M, Chapman JR, Cosio FG, et al. Protocol biopsies in renal transplantation: Insights into patient management and pathogenesis. Am J Transplant 2007; 7: 512.
33. Haas M, Segev DL, Racusen LC, et al. Arteriosclerosis in kidneys from healthy live donors: Comparison of wedge and needle core perioperative biopsies. Arch Pathol Lab Med 2008; 132: 37.
34. Locke JE, Zachary AA, Haas M, et al. The utility of splenectomy as rescue treatment for severe acute antibody mediated rejection. Am J Transplant 2007; 7: 842.
35. Kaplan B, Gangemi A, Thielke J, et al. Successful rescue of refractory, severe antibody mediated rejection with splenectomy. Transplantation 2007; 83: 99.

ABO incompatible; Kidney transplantation; Antibody-mediated rejection

© 2009 Lippincott Williams & Wilkins, Inc.