NDT Advance Access originally published online on July 10, 2007
Nephrology Dialysis Transplantation 2007 22(10):3048-3051; doi:10.1093/ndt/gfm460
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
On-demand strategy as an alternative to conventionally scheduled post-transplant immunoadsorptions after ABO-incompatible kidney transplantation
1University Hospital Freiburg, Renal Division, Germany, 2University Hospital Freiburg, Surgery Division, Transplant Unit, Germany, 3University Hospital Freiburg, Department of Transfusion Medicine, Germany and 4University Hospital Freiburg, Department of Pathology, Germany
Correspondence and offprint requests to: J. Wilpert, Renal Division, Hugstetter Strasse 55, 79106 Freiburg, Germany. Email: jochen.wilpert{at}uniklinik-freiburg.de
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Background. Since 2001, approximately 100 ABO-incompatible kidney transplantations have been performed in Europe. The standard protocol, employed by most transplant centres, uses rituximab and scheduled pre-emptive antigen-specific immunoadsorption on post-operative days 3, 6 and 9.
Methods. Our centre has performed 22 ABO-incompatible kidney transplantations since 2004, using a different approach; like in Sweden, all patients received immunoadsorptions preoperatively, but instead of scheduling pre-emptive post-transplant immunoadsorptions, we submitted patients to immunoadsorptions post-operatively only, if their isoagglutinine titers (IgG-Anti-A or -B) exceeded certain thresholds. These thresholds were greater than 1 : 8 in the first post-operative week and greater than 1 : 16 in the second post-operative week, respectively.
Results. A shorter pre-operative length on dialysis, a blood-type constellation of donor A1/recipient 0 and 9a high initial starting-titer were identified as predictors for post-operative immunoadsorptions.
Conclusion. Using this on-demand strategy, our data reveal that a titer-dependent protocol reduces costs at no additional risk for the patient.
Keywords: ABO-incompatible; IgG titer; immunoadsorption; indirect antiglobulin test; kidney transplantation
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Since the advent of ABO-incompatible kidney transplantation in Europe, approximately 100 ABO-incompatible kidney transplantations have been performed across European transplant centres. The protocol most widely employed in Europe, is the one published by Tydén et al. [1,2], using rituximab and antigen-specific immunoadsorptions, whereas the Japanese protocol is based on splenectomy and plasmapheresis or double-filtration plasmapheresis, respectively [3]. The risk for early humoral allograft rejection after an ABO-incompatible transplantation is highest within the first 2 weeks after transplantation [3], and correlates with the post-operative isoagglutinine titers [4]. This has prompted many centres to employ a protocol with fixed scheduled post-operative immunoadsorptions during the first 2 weeks after transplantation [2,5].
Our group has chosen a different approach. Since isoagglutinine titers are very sensitive to predict the risk of humoral rejections, we hypothesized that post-operative immunoadsorptions to prevent rejections, are only necessary in the presence of measurable isoagglutinine titers. We therefore closely monitored the post-operative isoagglutinine titers, and performed immunoadsorptions only, if titers were above certain threshold values (i.e. ‘on-demand strategy’). Our data suggest that costly ‘overtreatment’ and potentially critical ‘undertreatment’ can be prevented by this approach. Here we report on our experience with this modified protocol in a retrospective analysis of 22 ABO-incompatible transplantations.
|
Subjects and methods |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Twenty-two ABO-incompatible kidney transplants have been performed according to a protocol that shares most of its features with the widely used Tydén [2] method. Average donor age was 53 (±10) years. The average recipient was 45 (±11) years old and had been on dialysis for 44 (±32) months. All ABO incompatibilities were accepted with the exception of 2-antigen mismatches, i.e. donor AB, recipient 0. There were 15 A1, 2 A2 and 5 B major mismatches. The indirect antiglobulin test (IAT) was used to determine the initial IgG anti-A or anti-B titer against donor erythrocytes, measured before the administration of rituximab. The median initial IgG titer was 1 : 128 (range 1 : 8 up to 1 : 1024).
In accordance with the Swedish protocol, rituximab (375 mg/m2) was given 4 weeks prior to scheduled transplantation date. On the seventh pre-operative day a triple immunosuppressive regimen was initiated (mycophenolic acid, tacrolimus, prednisone) and antigen-specific immunoadsorptions (Glycosorb®, AB Transplantation, Lund, Sweden) were started on alternating days (2–3 plasma volumes per session). The Glycosorb® ABO column is a low-molecular carbohydrate column with A or B blood group antigen linked to a sepharose matrix. The column effectively and specifically depletes anti-A or anti-B antibodies without any apparent side effects [2]. Intravenous immunoglobulins (ivIG 0.5 g/kg) were administered 1–5 days ahead of the scheduled operation. Pre-operative immunoadsorptions were performed until IgG-anti-A/B titers equaled 1 : 4 or less on the morning before transplantation.
In contrast to the Swedish protocol, where post-operative immunoadsorptions are performed according to a definite schedule on post-operative days 3, 6 and 9, we carried out post-operative immunoadsorptions only if titers mandated us to do so [6].
We determined titers daily after successful grafting during the first post-operative week. In the second week after the transplant, we collected titers on days 8, 10, 12 and 14, respectively.
If post-operative IgG anti-A or anti-B titers exceeded 1 : 8 in post-operative week one and/or 1 : 16 in post-operative week 2, we carried out immunoadsorptions. If titers remained below these pre-defined thresholds, no immunoadsorptions were performed.
Results are expressed as mean ± SD unless otherwise stated. Student's t-test was used for inter-group comparisons and a P-value 
0.05 considered to be statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Twenty-two ABO-incompatible kidney transplantations were performed at our centre since April 2004 (Table 1).
|
Patient and graft survival
The mean follow-up until march 2007 was 18 months (range 1–35 months). Mean serum creatinine at a mean follow-up of 17 months was 1.63 ± 0.48 mg/dl, corresponding to an estimated creatinine-clearance of 47 ml/min/1.73 m2. One patient died with a functioning graft 4 months after transplantation due to sepsis caused by Clostridium difficile. A total of 41 protocol- or indication biopsies were performed; no humoral and only one late cellular rejection were observed in these biopsies. Calcineurin-inhibitor toxicity was diagnosed in three cases. BK-virus nephropathy was documented in one case. Two patients experienced asymptomatic infections with cytomegalovirus. There were no significant differences between these patients and a control collective that received a blood-group compatible kidney transplant from a living donor during the same period of time (63 patients) (data not shown).
Immunoadsorptions and isoagglutinine titers
Post-operative immunoadsorptions were performed in 32% of all patients (7/22), because their titers exceeded the thresholds of greater than 1 : 8 for post-operative week 1 (POW1), and greater than 1 : 16 for post-operative week 2 (POW2), respectively. These seven patients required a total of 29 immunoadsorptions (mean 4.1/patient) during the post-operative period. Increment and titer-peak varied extensively between patients. Four of the seven patients required immunoadsorptions during POW1 as well as during POW2. Two of the seven patients only needed immunoadsorptions during POW1 (patient #9: days 1, 2 and 3; patient #22: days 1, 2, 3 and 4). One patient exclusively needed immunoadsorptions during POW2 (patient #10: days 8, 9, 10, 11 and 12) (Figure 1). The seven patients who required additional extracorporeal treatments did not differ in their baseline characteristics with regards to donor age, recipient age, sex, cardiovascular risk factors, number of HLA mismatches or percentage of living-related donations. Protocol biopsies did not show any group-related differences on light microscopy or immunofluorescence.
|
Of note, the pre-operative dialysis duration appeared to correlate inversely with the requirement for further immunoadsorptions. Patients with additional post-operative immunoadsorptions had been dialyzed for an average of 27 (±26) months, whereas patients who did not require further treatments had been on dialysis for 51 (±33) months (P = 0.12). All three patients who underwent pre-emptive kidney transplantations (n = 2, patient #1 and #22), or who had only been on dialysis for <3 months before the operation (n = 1, patient #14) needed post-transplant treatments.
In addition, we identified differences in blood-type constellations as a risk factor for additional immunoadsorptions. Only patients with the donor blood-type A1 and recipient blood-type 0 had to be submitted to additional treatment cycles. None of the patients with other blood type combinations had to be treated post-operatively.
Finally, we observed that higher initial IgG-anti-A/B-titers are more likely associated with the need for additional immunoadsorptions in the post-operative period. Our seven index patients presented with a median starting titer of 1 : 256 (range 1 : 128–1 : 1024), while patients without post-operative treatments exhibited a median titer of 1 : 64 (range 1 : 8–1 : 1024).
Outcome of patients receiving post-operative immunoadsorption vs patients without post-operative immunoadsorptions
When comparing the outcome of the seven patients who underwent post-operative immunoadsorptions with the remaining 15 patients, no significant differences were detectable (Table 2). Mean follow-up in both groups was 17 months (range 1–35). Renal function determined by creatinine-clearance in the on-demand group was 50 ml/min/1.73 m2 (MDRD formula) (range 28–75) while follow-up in the no-treatment group of patients showed 46 ml/min/1.73 m2 (range 31–66). The incidence of surgical or medical complications did not differ between groups.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
In this report, we present data of 22 patients that underwent ABO-incompatible kidney transplantations between April 2004 and February 2007. As opposed to other groups, we adopted an on-demand strategy for post-operative immunoadsorptions that uses threshold isoagglutinine titers to implement post-operative immunoadsorptions instead of a scheduled regimen of pre-emptive treatments. Following this strategy, in our cohort of 22 patients, 15 patients (68%) did not undergo any post-operative immunoadsorptions. The decision of not submitting these patients to additional post-operative treatments was based on the fact that their donor-specific isoagglutinine titers remained under a defined threshold, and no clinical signs and symptoms of early humoral rejection were observed in the untreated patients.
The Swedish protocol [1] employs scheduled post-operative immunoadsorptions on days 3, 6 and 9, following successful grafting, irrespective of titer levels and clinical data. Our data suggest that immunoadsorptions are only necessary when a relevant amount of blood-type-specific antibodies is detectable in the recipient's circulation. Our threshold values are similar to protocols published by Gloor et al. [7], who used plasmapheresis when isoagglutinine-titers were above 1 : 8, or 1 : 16 in post-operative weeks 1 and 2, respectively. Despite the arbitrary fashion of these thresholds, the good renal outcome (no episodes of rejection) of the 15 untreated patients provides sufficient support for our algorithm.
This on-demand strategy saved a total of 45 immunoadsorptions (equivalent to approximately 180.000 Euro). Despite the good tolerance and low side effects of immunoadsorptions, our on-demand protocol also avoided the procedure-associated risk for the patient. Our protocol requires daily titer measurements in post-operative week 1, and titer measurements on every other day in post-operative week 2. Experienced laboratory staff is necessary to complete the assays quickly. However, despite these extra measures and assay costs of approximately 3000 Euro, our approach documents that standard protocols may be associated with costly ‘overtreatment’.
Seven of the 22 patients in our collective displayed post-operative isoagglutinine titers above our threshold values. These patients were submitted to an extra cycle of treatments, despite the lack of any clinical evidence for rejection. On average 4.1 ± 1.5 post-operative treatments were needed to lower titers into the desired target range. Thus, patients, who exhibit rising titers post-transplant should be submitted to more immunoadsorptions than the three pre-emptive treatments suggested by the Swedish protocol. We also observed that the titers peaked at completely different times in this group of patients (Figure 1). It is noteworthy that peaks can fall in between days of fixed scheduled treatments, potentially facilitating the advent of unrecognized humoral rejections. We therefore believe that pre-emptive scheduling in these seven patients could have easily resulted in ‘undertreatment’, facilitating the occurrence of humoral rejections.
It is not surprising that patients with high initial isoagglutinine titers need a higher number of immunoadsorptions until they reach the target titer value of less than 1 : 8 prior to surgery. It is still unknown, however, why these patients regenerate isoagglutinines more rapidly. One can hypothesize that higher initial titers reflect a more active state of the immune system, comparable to superior protection from infections by higher antibody titers after vaccination (e.g. in hepatitis B). This could also explain, why we observed a significantly inverse correlation between the time on dialysis and the average isoagglutinine titer in our cohort. Patients who had been on dialysis for more than 2 years exhibited lower isoagglutinine titers (median 1 : 64) than patients who dialyzed less than 2 years (median 1 : 256, P = 0.02). It is striking that all three patients who had been on dialysis for <3 months needed post-operative extracorporeal treatment—a possible sign for a more intact immune system at the beginning of a chronic renal replacement therapy. We caution, however, to conclude that patients who have been on dialysis for a long period of time, can undergo ABO-incompatible kidney transplantation without post-operative immunoadsorption. Close monitoring of titers after surgery will remain mandatory for all patients.
Based on these data we conclude that pre-emptive post-operative immunoadsorptions in ABO-incompatible kidney transplantation can be abandoned. In a majority of cases, they are dispensable and represent an unnecessary ‘overtreatment’. On the other hand, in patients with very high rebound titers after successful grafting, a rigid schedule of set immunoadsorptions might imply the risk of ‘undertreatment’. We therefore suggest using an on-demand strategy that uses immunoadsorptions adapted to individual isoagglutinine titers.
Conflict of interest statement. None declared.
|
Notes |
|---|
*These authors contributed equally to the work.
|
References |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
- Tydén G, Kumlien G, Fehrman I. Successful ABO-incompatible kidney transplantations without splenectomy using antigen-specific immunoadsorption and Rituximab. Transplantation (2003) 76:730–731.[CrossRef][Web of Science][Medline]
- Tyden G, Kumlien G, Genberg H, Sandberg J, Lundgren T, Fehrman I. ABO incompatible kidney transplantations without splenectomy, using antigen-specific immunoadsorption and Rituximab. Am J Transplant (2005) 5:145–148.[CrossRef][Web of Science][Medline]
- 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) 4:1089–1096.[CrossRef][Web of Science][Medline]
- Gloor JM, Lager DJ, Moore SB, et al. ABO-incompatible kidney transplantation using both A2 and non-A2 living donors. Transplantation (2003) 75:971–977.[CrossRef][Web of Science][Medline]
- Ahlenstiel T, Offner G, Strehlau J, et al. ABO-incompatible kidney transplantation of an 8-yr-old girl with donor/recipient-constellation A1B/B. Xenotransplantation (2006) 13:141–147.[CrossRef][Web of Science][Medline]
- Donauer J, Wilpert J, Geyer M, et al. ABO-incompatible kidney transplantation using antigen-specific immunoadsorption and rituximab: a single center experience. Xenotransplantation (2006) 13:108–110.[CrossRef][Web of Science][Medline]
- Gloor JM, Lager DJ, Fidler ME, et al. A Comparison of splenectomy versus intensive posttransplant antidonor blood group antibody monitoring without splenectomy in ABO-incompatible kidney transplantation. Transplantation (2005) 80:1572–1577.[CrossRef][Web of Science][Medline]
Accepted in revised form: 13. 6.07
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Keven and S. Sengul ABO-incompatible kidney transplantation: on-demand strategy Nephrol. Dial. Transplant., May 1, 2008; 23(5): 1773 - 1774. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||