- Email: [email protected]
Effect of Delayed Graft Function in Hypersensitized Kidney Transplant Recipients
Effect of Delayed Graft Function in Hypersensitized Kidney Transplant Recipients Marcos López-Hoyos, Gema Fernández-Fresnedo, Emilio Rodrigo, Juan Carlos Ruiz, and Manuel Arias ABSTRACT: There is increased evidence about the deleterious effect of delayed graft function (DGF) in both short- and long-term kidney graft outcome. Among the mechanisms involved in the production of DGF, immune factors play a role, especially in the level of hypersensitization. From the 1389 patients transplanted at our hospital until November 2004, it has been found that the presence of moderate and high levels of sensitization, as measured by panel-reactive antibodies, is a risk factor for suffering from DGF. Further, DGF was associated with poor graft survival, and the risk was even higher when DGF was combined with moderate/high panel-reactive antibodies. Recent data demonstrate the usefulness of ABBREVIATIONS AR acute rejection DGF delayed graft function HLA human leukocyte antigen HS hypersensitized
Now that kidney transplantation has become the preferred choice for treating end-stage renal disease, and now that acute rejection (AR) and short-term graft outcome have been resolved, the challenge we now face is to improve the long-term graft outcome and the patient outcome [1]. These aims have been complicated by the ever-increasing number of patients on transplant waiting lists and the difficulties associated with the shortage of available organs [2]. Further, the criteria for selecting donors are expanding, with marginal kidneys now being transplanted, and an important percentage of patients undergo repeat renal transplantations [3]. These patients have been sensitized by previous human leukocyte antigen (HLA)-mismatched transplants and, probably, by From the Services of Immunology and Nephrology, Hospital Universitario Marqués de Valdecilla, Santander, Spain. Address reprint requests to: Dr. Manuel Arias, Service of Nephrology, Hospital Universitario Marqués de Valdecilla, Santander, Spain; Fax: 34 942 320415; E-mail: [email protected]. Received December 15, 2004; accepted January 19, 2005. Human Immunology 66, 371–377 (2005) © American Society for Histocompatibility and Immunogenetics, 2005 Published by Elsevier Inc.
intravenous immunoglobulins in the management of hypersensitized patients in terms of short-term outcome. It remains to be demonstrated whether this therapy is able to ameliorate the higher ischemic injury that kidneys undergo from these immunologically high-risk patients. Human Immunology 66, 371–377 (2005). © American Society for Histocompatibility and Immunogenetics, 2005. Published by Elsevier Inc. KEYWORDS: delayed graft function; graft outcome; hypersensitization; intravenous immunoglobulins; ischemic injury
IVIG PRA SGF TNF-␣
intravenous immunoglobulin panel-reactive antibody slow graft function tumor necrosis factor alpha
blood transfusions and pregnancy [4]. Such patients are termed “hypersensitized” (HS) when exhibiting 50% or more panel-reactive antibodies (PRAs), although there is controversy about the exact PRA to define them, mostly because any titer of antibodies is associated with poorer graft survival [5]. Hypersensitized recipients with PRA ⬎50% represent 9.2% of the waiting list at our hospital. These patients usually stay on the waiting lists of transplant centers for a long time because the crossmatch tests are positive—and if they are not, the requirements for HLA matching are so strict that is almost impossible to find a suitable donor. The use of marginal kidneys, together with the increased number of HS patients, largely contribute to the development of delayed graft function (DGF), which is one of the major risk factors for poor graft outcome and long-term patient outcome [6]. Important advances have been made by improving the perioperative management, but DGF is still a major impediment to the progression of kidney transplanta0198-8859/05/$–see front matter doi:10.1016/j.humimm.2005.01.026
372
FIGURE 1 Incidence of acute rejection (AR), delayed graft function (DGF), and slow graft function (SGF) in the last 8 years at the University Hospital Marques Valdecilla. DGF was defined as need for dialysis in the first week after transplantation. SGF was defined as a slow recovery of graft function that permits avoidance of dialysis.
tion. With regard to hypersensitization, recent approaches have been published that allow successful transplantation of crossmatch–positive receptors, mainly on the basis of the use of intravenous immunoglobulins (IVIG) and plasmapheresis [7]. There are different definitions of DGF, although the most widely used is that of poor urine output and requirement for dialysis within the first week after renal transplantation [8]. This definition has the limitations of nephrologist readiness to perform the dialysis, delay in diagnosis by up to a week, and inclusion of different levels of graft dysfunction [9]. To overcome these limitations, we and others have proposed the creatinine reduction ratio on posttransplant day 2 as an earlier parameter of renal allograft function [9, 10]. We believe that this is a more predictive factor for kidney graft
M. López-Hoyos et al.
function during the first year than the performance of dialysis because this parameter detects a group of renal transplant recipients with poor graft survival, even though they do not need dialysis therapy [10]. The immediate graft function is in the opposite side, and it represents active diuresis and rapid decreases in serum creatinine. In addition, another concept between DGF and immediate graft function has been introduced: slow graft function (SGF). Slow graft function means slow recovery of graft function characterized by a moderate degree of diuresis and slow decrease in serum creatinine, which permit avoidance of dialysis treatment [6, 11]. At our hospital, the mean incidence of DGF and SGF during the last 8 years was 37.49% and 15.51%, respectively (Figure 1). We observed a relationship between DGF and AR in the last 5 years, with a clear decrease in their incidence until 2002. New immunosuppressive therapies have probably allowed better management of rejection. However, the last 2 years have seen an increase the frequency of AR episodes, perhaps because of the calcineurin inhibitor–sparing protocols [12] and because of the increased use of expanded criteria (to address the donor shortage) in the last few years at our center [6]. A number of risk factors for DGF and SGF have been described: donor tissue quality, brain death, perioperative management, recipient variables, and immune factors [6, 13]. The aim of the present report was to focus on the immune risk factors for developing any type of immediate graft dysfunction and how the use of IVIG is rapidly changing this field. It is well accepted that HS patients have a higher frequency of DGF, probably because preformed HLA antibodies are not detected in the crossmatch and induce a silent AR that is manifested as DGF [14]. Figure 2 illustrates the incidence of AR and immediate graft
FIGURE 2 Frequency of acute rejection (AR) and delayed graft function (DGF) according to the current (A) and peak (B) panel-reactive antibody (PRA). Low level of sensitization was considered for PRA ⬍25%, moderate for PRA 26%–50%, and high for PRA ⬎51%. DGF-grouped patients experienced both delayed and slow graft function.
DGF in Hypersensitized Kidney Recipients
FIGURE 3 Mean number of dialysis sessions received to recover from the delayed graft function according to the current (gray bars) and peak (black bars) panel-reactive antibocy (PRA). Levels of PRA were considered as indicated in Figure 2.
dysfunction (both DGF and SGF) in 427 cadaver kidney recipients transplanted from 1997 to date (from our total cohort of 1398 recipients) according to the PRA at the moment of transplantation and according to the historical peak reactivity, measured by the complement-dependent crossmatch test. Data demonstrate that the higher current PRA of the recipients, the higher the incidence of graft dysfunction. This effect was slightly more marked for moderate current PRA, although it can be attributed to the better HLA matching in the donor selection and to the higher level of immunosuppression received by recipients with high current PRA. As expected, the grade of sensitization was accompanied by a higher incidence of AR. When considering the historical PRA, higher-peak PRA was followed by higher incidence of AR and DGF (Figure 2). Differences were significant between groups with moderate- to high-peak PRA and the group with low-peak PRA (p ⬍ 0.05 for AR and p ⬍ 0.01 for DGF). There was no significant difference between the two groups of patients with moderate- or high-peak PRA in the frequency of DGF, but the percentage of patients with AR was higher (p ⬍ 001) in the group with high-peak PRA than in the one with moderate-peak PRA (Figure 2). Taken together, these data demonstrate that the presence of a moderate or high grade of sensitization is a risk factor for immediate graft dysfunction, and subsequently, for AR. As a parameter of severity of the DGF, we quantified the number of dialysis sessions undergone by those patients with DGF, according to both current and peak PRA (Figure 3). There were no differences when considering the current PRA, but patients with moderate-peak PRA underwent a higher number of dialysis sessions to recover from DGF, although differences did not reach statistical significance. At our center, as in others [5], standard practice is to
373
consider only the current negative crossmatch in spite of an historical positive crossmatch. In view of the data presented, the existence of high-peak PRA is a risk factor for suffering from DGF. The role of peak PRA in the immediate graft function must probably be revised. It is possible that special attention must be paid to factors affecting DGF, such as cold ischemia time or immunosuppression with calcineurin inhibitors. Additionally, new crossmatch methods to detect even very low levels of donor-specific alloantibodies have been developed (i.e., flow cytometry), and consequently, this source of immune injury will probably change [15]. The effect of donor-specific alloantibodies detected by these new methods on DGF and AR should be revisited. Moreover, their role in chronic allograft nephropathy and transplant glomerulopathy must be more deeply investigated. Prior HLA-mismatched kidney transplantation, together with previous transfusions or pregnancies, is a classical risk factor for allosensitization [4] and can be responsible for DGF. We sought to determine whether there was any association between any of those sensitizing factors with the production of DGF. Previous kidney transplantation seemed to confer a higher risk of DGF, although this effect was not related to the number of transplants received. Data were not so clear with regard to the number of transfusions, but the lack of blood transfusion seemed to protect patients from DGF (Figure 4). Previous pregnancies did not seem to have any effect on
FIGURE 4 Frequency of recipients that experienced delayed graft function (DGF) (gray bars) or did not (black bars) according to number of previous renal transplants received (A) or blood transfusion (B).
374
the DGF frequency in our population of recipients (data not shown). The deleterious effect of DGF on the graft survival has been previously described by our group [16]. Such effect is especially significant in the first year after transplantation but not in the long term. There is some controversy in the literature with regard to this issue [13, 16 –20]. It is clear that the combination of DGF with AR episodes contributes to graft loss [16, 21], and it is probable that the worst outcome associated with DGF alone observed in some centers might be explained by the development of silent AR that is not detected unless protocol biopsies are performed [14]. The lack of a standard, objective definition of DGF makes the consensus of the effect of DGF in long-term outcomes difficult. At our hospital, the development of AR episodes was identified as a major risk factor for graft survival, although the association of AR with DGF (defined as the need for dialysis sessions in the first week after transplantation) resulted in the worst combination for graft survival (Figure 5). In the absence of AR, graft survival was significant higher in those recipients with immediate graft function than in those with DGF (p ⬍ 0.05). It was not possible to analyze the effect on graft survival of DGF in combination with high PRA because the presence of high-peak or current PRA (defined as ⬎50%) was the worst risk factor for graft survival (Figure 5), and it interferes with the Cox analysis. Although the aim of the present work is to focus on HS patients and the effect of alloimmunization of DGF, it must be kept in mind that many of the nonimmune factors occurring during kidney transplantation may contribute to both DGF and hypersensitization. One of the most important factors associated to increased risk of DGF is a long cold ischemia time [6, 13]. At the same time, it has been described that longer cold ischemia time induces a more immunogenic kidney, with higher production of HLA class I antibodies and higher risk of AR, independent of the existence of HLA mismatches [22]. Thus, preservation management may cause ischemia-reperfusion injury with accumulation of leukocytes and endothelial activation that finally result in elevation of cytokines that induce T- and B-cell activation that may lead to cell- and/or antibody-mediated rejection and poor graft survival [23]. Efforts to improve the preservation variables must be made, especially in HS kidneys recipients in whom the risk for alloimmune responses is extremely high. It is quite difficult to isolate the effect of allosensitization from early graft dysfunction to clearly define their role in graft and patient survival. Nonetheless, it is evident that the number of HS patients in the waiting list is exponentially increasing and that efforts must be made to give these patients a chance to receive a new
M. López-Hoyos et al.
renal transplant. In recent years, successful kidney transplantation protocols for HS patients have been developed, mostly on the basis of the use of IVIG [7]. This therapy is able to abrogate the donor-specific crossmatch test in vitro and to predict which patients are most likely to respond in vivo, allowing successful transplantation [24 –26]. Results with high-dose IVIG before transplantation demonstrated a 2-year patient and graft survival of 97.5% and 89.1%, respectively, and the incidence of AR was 29% [24]. Because of the short experience with this type of protocol, special care must be taken to avoid unexpected adverse effects. A case report recently described a kidney transplant recipient in whom the combination of antithymocyte globulin and IVIG resulted in acute renal transplant injury [27]. A possible explanation for this phenomenon was the interference of IVIG with lymphodepletion by antithymocyte globulin because CD3⫹ cell counts remained elevated. Finally, preconditioning regimens consisting of plasmapheresis and lowdose cytomegalovirus hyperimmunoglobulin, along with standard immunosuppression, have even allowed successful transplantation of simultaneous ABO-incompatible and crossmatch–positive kidneys [28]. Although the mechanisms seem to be very complex and not fully understood, IVIG is thought to act mainly by neutralizing and/or removing donor specific antibodies by antiidiotypic antibodies present in such preparations [29]. This action depends on the half-life of infused IVIG. Besides this passive action, IVIG may actively inhibit B-cell production of donor-specific alloantibodies by different mechanisms that last longer than the halflife of IVIG. First, they can control the B-cell repertoire that migrates from the bone marrow to the peripheral lymphoid organs, although this has been only described in murine models [30]. Second, IVIG may interact with CD32 or Fc␥ receptoer IIB (Fc␥RIIB) that transmits inhibitory signals on B cells, and it may even induce apoptosis [31]. Finally, it has been demonstrated that IVIG inhibits the expression of the costimulatory molecule CD19 on the surface of activated human B cells [24]. On the other hand, immunoglobulins present in IVIG preparations may block complement-mediated endothelial injury through the high avidity of their Fc fragments for C3b and C4b. Through this mechanism, the complement cascade is blocked and the generation of the membrane-attack complex is abolished. As noted for CD32 on B cells, IVIG can block the binding of alloantibodies to the activating Fc receptors on inflammatory leukocytes, such as macrophages and neutrophils, responsible in part for the DGF [29]. Polymorphisms of tumor necrosis factor (TNF)-␣ associated with a higher TNF-␣ expression have been recently associated with a slight risk for DGF [32] and IVIG might also interfere with it by inducing the expression of antiinflammatory cyto-
DGF in Hypersensitized Kidney Recipients
375
FIGURE 5 Kaplan-Meier analysis of graft survival comparing patients according to current and peak panel-reactive antibody (PRA) (A) and to existence of delayed graft function (DGA) and acute rejection (AR) episodes (B). (A) Patients with current (thick solid line) and peak (dashed line) PRA over 50% (squares) exhibited a significantly worse graft survival than those patients with PRA ⬍50% (circles, p ⬍ 0.001 in both cases). (B) Development of AR (dashed line) was associated with a poor graft survival, especially when it was combined with the development of DGF (circles).
kines [7, 29]. At the same time, IVIG contains antibodies directed to multiple cytokines (i.e., interleukin [IL]1␣, IL-6, TNF-␣) [7]. Taken together, although the main function attributed to the therapy with IVIG in HS kidney recipients is related to the reduction in the titers of HLA antibodies and inhibition of humoral immune responses, they also inhibit the function of most of the
inflammatory cells (from endothelial cells to T cells) that participate in the ischemia-reperfusion injury during kidney transplantation. Efforts to decrease the number of HS patients on the waiting list for a renal transplant are being made at our center in collaboration with other centers from northern Spain. The proposed protocol is based on that of Jordan et
376
al. [24]. Patients must have a PRA ⬎50% and a time on the waiting list of more than 6 months. The pretransplant protocol is based on the therapy with IVIG at dosages of 2 g/kg once a month, for three consecutive months. One week after the last infusion, the measurement of HLA antibodies should be performed and must reflect a decrease of 50% in PRA. In such a case, transplantation must be performed immediately with a blood group– compatible, negative crossmatch organ (HLA matching is not necessary). From a clinical point of view, the possibility of using a previous in vitro test to demonstrate a PRA reduction provides a rational basis for the selection of HS patients to be transplanted [24]. Postransplant therapy includes thymoglobulin (in the first 10 days after transplantation), steroids, mycophenolate mofetil, and tacrolimus, together with IVIG (at 48 hours, 20 days and 40 days after transplantation). Obviously, in this type of protocol, active exchange programs are especially necessary. There is no doubt that desensitization with IVIG has opened doors to HS patients. Nonetheless, this approach is novel, and results, although promising, are still scarce. Data indicate that allosensitization and the response to desensitization are heterogeneous. Further studies will add important knowledge to this field. In summary, the association of DGF and HLA antibodies is deleterious. Several mechanisms that lead to an ischemia injury with inflammatory phenomena into the kidney have been suggested. Such injury finally results in an impairment in the glomerular filtration rate and a poor graft survival. New approaches, including IVIG therapy, will permit better management of DGF and improved graft survival in these immunologically high-risk kidney recipients. ACKNOWLEDGMENT
This work was supported by the Fondo de Investigación Sanitaria (RITC C03/03).
REFERENCES 1. Pascual M, Theruvath T, Kawai T, Tolkoff-Rubin N, Cosimi AB: Strategies to improve long term outcomes after renal transplantation. N Engl J Med 346:580, 2002. 2. Cecka JM: The OPTN/UNOS renal transplant registry 2003. Clin Transpl 1, 2003. 3. Ojo AO, Hanson JA, Meier-Kriesche H, Okechukwu CN, Wolfe RA, Leichtman AB, Agodoa LY, Kaplan B, Port FK: Survival in recipients of marginal cadaveric donor kidneys compared with other recipients and wait-listed transplant candidates. J Am Soc Nephrol 12:589, 2001. 4. Sanfilippo F, Vaughn WK, Bollinger RR, Spees EK: Comparative effects of pregnancy, transfusion, and prior graft rejection on sensitization and renal transplant results. Transplantation 34:360, 1982.
M. López-Hoyos et al.
5. Gebel HM, Bray RA, Nickerson P: Pre-transplant assessment of donor-reactive, HLA-specific antibodies in renal transplantation: contraindication vs risk. Am J Transplant 3:1488, 2003. 6. Peeters P, Terryn W, Vanholder R, Lameire N: Delayed graft function in renal transplantation. Curr Opin Crit Care 10:489, 2004. 7. Jordan S, Cunningham-Rundles C, McEwan R: Utility of intravenous immune globulin in kidney transplantation: efficacy, safety, and cost implications. Am J Transplant 3:653, 2003. 8. Halloran PF, Hunsicker LG: Delayed graft function: state of the art. Summit meeting, Scottsdale, AZ, November 10 –11, 2000. Am J Transplant 1:115, 2001. 9. Govani MV, Kwon O, Batiuk TD, Milgrom ML, Filo RS: Creatinine reduction ratio and 24-hour creatinine excretion on posttransplant day two: simple and objective tools to define graft function. J Am Soc Nephrol 13:1645, 2002. 10. Rodrigo E, Ruiz JC, Piñera C, Fernandez-Fresnedo G, Escallada R, Palomar R, Cotorruelo JG, Zubimendi JA, Martin de Francisco AL, Arias M: Creatinine reduction ratio on post-transplant day two as criterion in defining delayed graft function. Am J Transplant 4:1163, 2004. 11. Park JH, Yang CW, Kim YS, Lee SH, Choi YJ, Kim YS, Moon IS, Koh YB, Bang BK: Comparisons of clinicopathological correlations between immediate and slow graft function in renal transplant recipients. Clin Transplant 16:18, 2002. 12. McTaggart RA, Tomlanovich S, Bostrom A, Roberts JP, Feng S: Comparison of outcomes after delayed graft function: sirolimus-based versus other calcineurin-inhibitor sparing induction immunosuppression regimens. Transplantation 78:475, 2004. 13. Cecka JM, Shoskes DA, Gjertson DW: Clinical impact of delayed graft function for kidney transplantation. Transplant Rev 15:57, 2001. 14. Qureshi F, Rabb H, Kasiske BL: Silent acute rejection during prolonged delayed graft function reduces kidney allograft survival. Transplantation 74:1400, 2002. 15. Lopez-Hoyos M, Fernandez-Fresnedo G, Pastor JM, Arias M: Anticuerpos anti-HLA postrasplante: un nuevo método de monitorización. Nefrologia 24:62, 2004. 16. Arias M: Impact of the delayed graft function in hypersensitized kidney transplant recipients. Transplant Proc 35:1655, 2003. 17. Marcen R, Orofino L, Pascual J, De la Cal MA, Teruel JL, Villafruela JJ, Rivera ME, Mampaso F, Burgos FJ, Ortuño J: Delayed graft function does not reduce the survival of renal transplant allografts. Transplantation 66:461, 1998. 18. Fernandez-Juarez G, Marcen R, Pascual R, Teruel JL, Rivera ME, Villafruela JJ, Mampaso F, Burgos FJ, Ortuño J: Prolonged delayed graft function decreases graft sur-
DGF in Hypersensitized Kidney Recipients
19.
20.
21.
22.
23. 24.
25.
vival in transplant patients taking cyclosporine. Transplant Proc 34:338, 2002. Boom H, Mallat MJK, De Fijter JW, Zwinderman AH, Paul LC: Delayed graft function influences renal function, but not survival. Kidney Int 58:859, 2000. Gentil MA, Alcaide MP, Algarra GR, Pereira P, Toro J, Gonzalez-Roncero F, Lopez M, Bernal G, Mateos J: Impact of delayed graft function on cadaveric kidney transplant outcome. Transplant Proc 35:689, 2003. McLaren AJ, Jassem W, Gray DWR, Fuggle SV, Welsh KI, Morris PJ: Delayed graft function: risk factors and the relative effects of early function and acute rejection on long-term survival in cadaveric renal transplantation. Clin Transplant 13:266, 1999. Bryan CF, Luger AM, Martinez J, Muruve N, Nelson PW, Pierce GE, Ross G, Shield CF, Warady BA, Aeder MI, Helling TS: Cold ischemia time: an independent predictor of increased HLA class I antibody production after rejection of a primary cadaveric renal allograft. Transplantation 71: 875, 2001. Friedewald JJ, Rabb H: Inflammatory cells in ischemic acute renal failure. Kidney Int 66:486, 2004. Jordan SC, Vo A, Bunnapradist S, Toyoda M, Peng A, Puliyanda D, Kamil E, Tyan D: intravenous immune globulin treatment inhibits crossmatch positivity and allows for successful transplantation of incompatible organs in living-donor and cadaver recipients. Transplantation 76:631, 2003. Akalin E, Ames S, Sehgal V, Fotino M, Daly L, Murphy B, Bromberg JS: intravenous immunoglobulin and thymoglobulin facilitate kidney transplantation in complementdependent cytotoxicity B-cell and flow cytometry T- or
377
26.
27.
28.
29.
30.
31.
32.
B-cell crossmatch-positive patients. Transplantation 76:1444, 2003. Gloor JM, DeGoey S, Ploeger N, Gebel H, Bray R, Moore B, Dean PG, Stegall MD: Persistence of low levels of alloantibody after desensitization in crossmatch-positive living-donor kidney transplantation. Transplantation 78: 221, 2004. Gallay BJ, Perez RV, Ramsamooj R: Acute renal transplant injury and interaction between antithymocyte globulin and pooled human immunoglobulin. Clin Transplant 18:327, 2004. Warren DS, Zachary AA, Sonnenday CJ, King KE, Cooper M, Ratner LE, Shirey RS, Haas M, Leffell MS, Montgomery RA: Successful renal transplantation across simultaneous ABO incompatible and positive crossmatch barriers. Am J Transplant 4:561, 2004. Kazatchkine MD, Kaveri SV: Immunomodulation of autoimmune and inflammatory diseases with intravenous immune globulin. N Engl J Med 345:747, 2001. Sundblad A, Marcos M, Malanchere E, Castro A, Haury M, Huetz F, Nobrega A, Freitas A, Coutinho A: Observations on the mode of action of normal immunoglobulin at high doses. Immunol Rev 139:125, 1994. Toyoda M, Pao A, Petrosian A, Jordan SC: Pooled human gammaglobulin modulates surface molecule expression and induces apoptosis in human B cells. Am J Transplant 3:156, 2003. Alakulppi NS, Kyllönen LE, Jäntti VT, Matinlauri IH, Partanen J, Salmela KT, Laine JT: Cytokine gene polymorphisms and risks of acute rejection and delayed graft function after kidney transplantation. Transplantation 78: 1422, 2004.
Now that kidney transplantation has become the preferred choice for treating end-stage renal disease, and now that acute rejection (AR) and short-term graft outcome have been resolved, the challenge we now face is to improve the long-term graft outcome and the patient outcome [1]. These aims have been complicated by the ever-increasing number of patients on transplant waiting lists and the difficulties associated with the shortage of available organs [2]. Further, the criteria for selecting donors are expanding, with marginal kidneys now being transplanted, and an important percentage of patients undergo repeat renal transplantations [3]. These patients have been sensitized by previous human leukocyte antigen (HLA)-mismatched transplants and, probably, by From the Services of Immunology and Nephrology, Hospital Universitario Marqués de Valdecilla, Santander, Spain. Address reprint requests to: Dr. Manuel Arias, Service of Nephrology, Hospital Universitario Marqués de Valdecilla, Santander, Spain; Fax: 34 942 320415; E-mail: [email protected]. Received December 15, 2004; accepted January 19, 2005. Human Immunology 66, 371–377 (2005) © American Society for Histocompatibility and Immunogenetics, 2005 Published by Elsevier Inc.
intravenous immunoglobulins in the management of hypersensitized patients in terms of short-term outcome. It remains to be demonstrated whether this therapy is able to ameliorate the higher ischemic injury that kidneys undergo from these immunologically high-risk patients. Human Immunology 66, 371–377 (2005). © American Society for Histocompatibility and Immunogenetics, 2005. Published by Elsevier Inc. KEYWORDS: delayed graft function; graft outcome; hypersensitization; intravenous immunoglobulins; ischemic injury
IVIG PRA SGF TNF-␣
intravenous immunoglobulin panel-reactive antibody slow graft function tumor necrosis factor alpha
blood transfusions and pregnancy [4]. Such patients are termed “hypersensitized” (HS) when exhibiting 50% or more panel-reactive antibodies (PRAs), although there is controversy about the exact PRA to define them, mostly because any titer of antibodies is associated with poorer graft survival [5]. Hypersensitized recipients with PRA ⬎50% represent 9.2% of the waiting list at our hospital. These patients usually stay on the waiting lists of transplant centers for a long time because the crossmatch tests are positive—and if they are not, the requirements for HLA matching are so strict that is almost impossible to find a suitable donor. The use of marginal kidneys, together with the increased number of HS patients, largely contribute to the development of delayed graft function (DGF), which is one of the major risk factors for poor graft outcome and long-term patient outcome [6]. Important advances have been made by improving the perioperative management, but DGF is still a major impediment to the progression of kidney transplanta0198-8859/05/$–see front matter doi:10.1016/j.humimm.2005.01.026
372
FIGURE 1 Incidence of acute rejection (AR), delayed graft function (DGF), and slow graft function (SGF) in the last 8 years at the University Hospital Marques Valdecilla. DGF was defined as need for dialysis in the first week after transplantation. SGF was defined as a slow recovery of graft function that permits avoidance of dialysis.
tion. With regard to hypersensitization, recent approaches have been published that allow successful transplantation of crossmatch–positive receptors, mainly on the basis of the use of intravenous immunoglobulins (IVIG) and plasmapheresis [7]. There are different definitions of DGF, although the most widely used is that of poor urine output and requirement for dialysis within the first week after renal transplantation [8]. This definition has the limitations of nephrologist readiness to perform the dialysis, delay in diagnosis by up to a week, and inclusion of different levels of graft dysfunction [9]. To overcome these limitations, we and others have proposed the creatinine reduction ratio on posttransplant day 2 as an earlier parameter of renal allograft function [9, 10]. We believe that this is a more predictive factor for kidney graft
M. López-Hoyos et al.
function during the first year than the performance of dialysis because this parameter detects a group of renal transplant recipients with poor graft survival, even though they do not need dialysis therapy [10]. The immediate graft function is in the opposite side, and it represents active diuresis and rapid decreases in serum creatinine. In addition, another concept between DGF and immediate graft function has been introduced: slow graft function (SGF). Slow graft function means slow recovery of graft function characterized by a moderate degree of diuresis and slow decrease in serum creatinine, which permit avoidance of dialysis treatment [6, 11]. At our hospital, the mean incidence of DGF and SGF during the last 8 years was 37.49% and 15.51%, respectively (Figure 1). We observed a relationship between DGF and AR in the last 5 years, with a clear decrease in their incidence until 2002. New immunosuppressive therapies have probably allowed better management of rejection. However, the last 2 years have seen an increase the frequency of AR episodes, perhaps because of the calcineurin inhibitor–sparing protocols [12] and because of the increased use of expanded criteria (to address the donor shortage) in the last few years at our center [6]. A number of risk factors for DGF and SGF have been described: donor tissue quality, brain death, perioperative management, recipient variables, and immune factors [6, 13]. The aim of the present report was to focus on the immune risk factors for developing any type of immediate graft dysfunction and how the use of IVIG is rapidly changing this field. It is well accepted that HS patients have a higher frequency of DGF, probably because preformed HLA antibodies are not detected in the crossmatch and induce a silent AR that is manifested as DGF [14]. Figure 2 illustrates the incidence of AR and immediate graft
FIGURE 2 Frequency of acute rejection (AR) and delayed graft function (DGF) according to the current (A) and peak (B) panel-reactive antibody (PRA). Low level of sensitization was considered for PRA ⬍25%, moderate for PRA 26%–50%, and high for PRA ⬎51%. DGF-grouped patients experienced both delayed and slow graft function.
DGF in Hypersensitized Kidney Recipients
FIGURE 3 Mean number of dialysis sessions received to recover from the delayed graft function according to the current (gray bars) and peak (black bars) panel-reactive antibocy (PRA). Levels of PRA were considered as indicated in Figure 2.
dysfunction (both DGF and SGF) in 427 cadaver kidney recipients transplanted from 1997 to date (from our total cohort of 1398 recipients) according to the PRA at the moment of transplantation and according to the historical peak reactivity, measured by the complement-dependent crossmatch test. Data demonstrate that the higher current PRA of the recipients, the higher the incidence of graft dysfunction. This effect was slightly more marked for moderate current PRA, although it can be attributed to the better HLA matching in the donor selection and to the higher level of immunosuppression received by recipients with high current PRA. As expected, the grade of sensitization was accompanied by a higher incidence of AR. When considering the historical PRA, higher-peak PRA was followed by higher incidence of AR and DGF (Figure 2). Differences were significant between groups with moderate- to high-peak PRA and the group with low-peak PRA (p ⬍ 0.05 for AR and p ⬍ 0.01 for DGF). There was no significant difference between the two groups of patients with moderate- or high-peak PRA in the frequency of DGF, but the percentage of patients with AR was higher (p ⬍ 001) in the group with high-peak PRA than in the one with moderate-peak PRA (Figure 2). Taken together, these data demonstrate that the presence of a moderate or high grade of sensitization is a risk factor for immediate graft dysfunction, and subsequently, for AR. As a parameter of severity of the DGF, we quantified the number of dialysis sessions undergone by those patients with DGF, according to both current and peak PRA (Figure 3). There were no differences when considering the current PRA, but patients with moderate-peak PRA underwent a higher number of dialysis sessions to recover from DGF, although differences did not reach statistical significance. At our center, as in others [5], standard practice is to
373
consider only the current negative crossmatch in spite of an historical positive crossmatch. In view of the data presented, the existence of high-peak PRA is a risk factor for suffering from DGF. The role of peak PRA in the immediate graft function must probably be revised. It is possible that special attention must be paid to factors affecting DGF, such as cold ischemia time or immunosuppression with calcineurin inhibitors. Additionally, new crossmatch methods to detect even very low levels of donor-specific alloantibodies have been developed (i.e., flow cytometry), and consequently, this source of immune injury will probably change [15]. The effect of donor-specific alloantibodies detected by these new methods on DGF and AR should be revisited. Moreover, their role in chronic allograft nephropathy and transplant glomerulopathy must be more deeply investigated. Prior HLA-mismatched kidney transplantation, together with previous transfusions or pregnancies, is a classical risk factor for allosensitization [4] and can be responsible for DGF. We sought to determine whether there was any association between any of those sensitizing factors with the production of DGF. Previous kidney transplantation seemed to confer a higher risk of DGF, although this effect was not related to the number of transplants received. Data were not so clear with regard to the number of transfusions, but the lack of blood transfusion seemed to protect patients from DGF (Figure 4). Previous pregnancies did not seem to have any effect on
FIGURE 4 Frequency of recipients that experienced delayed graft function (DGF) (gray bars) or did not (black bars) according to number of previous renal transplants received (A) or blood transfusion (B).
374
the DGF frequency in our population of recipients (data not shown). The deleterious effect of DGF on the graft survival has been previously described by our group [16]. Such effect is especially significant in the first year after transplantation but not in the long term. There is some controversy in the literature with regard to this issue [13, 16 –20]. It is clear that the combination of DGF with AR episodes contributes to graft loss [16, 21], and it is probable that the worst outcome associated with DGF alone observed in some centers might be explained by the development of silent AR that is not detected unless protocol biopsies are performed [14]. The lack of a standard, objective definition of DGF makes the consensus of the effect of DGF in long-term outcomes difficult. At our hospital, the development of AR episodes was identified as a major risk factor for graft survival, although the association of AR with DGF (defined as the need for dialysis sessions in the first week after transplantation) resulted in the worst combination for graft survival (Figure 5). In the absence of AR, graft survival was significant higher in those recipients with immediate graft function than in those with DGF (p ⬍ 0.05). It was not possible to analyze the effect on graft survival of DGF in combination with high PRA because the presence of high-peak or current PRA (defined as ⬎50%) was the worst risk factor for graft survival (Figure 5), and it interferes with the Cox analysis. Although the aim of the present work is to focus on HS patients and the effect of alloimmunization of DGF, it must be kept in mind that many of the nonimmune factors occurring during kidney transplantation may contribute to both DGF and hypersensitization. One of the most important factors associated to increased risk of DGF is a long cold ischemia time [6, 13]. At the same time, it has been described that longer cold ischemia time induces a more immunogenic kidney, with higher production of HLA class I antibodies and higher risk of AR, independent of the existence of HLA mismatches [22]. Thus, preservation management may cause ischemia-reperfusion injury with accumulation of leukocytes and endothelial activation that finally result in elevation of cytokines that induce T- and B-cell activation that may lead to cell- and/or antibody-mediated rejection and poor graft survival [23]. Efforts to improve the preservation variables must be made, especially in HS kidneys recipients in whom the risk for alloimmune responses is extremely high. It is quite difficult to isolate the effect of allosensitization from early graft dysfunction to clearly define their role in graft and patient survival. Nonetheless, it is evident that the number of HS patients in the waiting list is exponentially increasing and that efforts must be made to give these patients a chance to receive a new
M. López-Hoyos et al.
renal transplant. In recent years, successful kidney transplantation protocols for HS patients have been developed, mostly on the basis of the use of IVIG [7]. This therapy is able to abrogate the donor-specific crossmatch test in vitro and to predict which patients are most likely to respond in vivo, allowing successful transplantation [24 –26]. Results with high-dose IVIG before transplantation demonstrated a 2-year patient and graft survival of 97.5% and 89.1%, respectively, and the incidence of AR was 29% [24]. Because of the short experience with this type of protocol, special care must be taken to avoid unexpected adverse effects. A case report recently described a kidney transplant recipient in whom the combination of antithymocyte globulin and IVIG resulted in acute renal transplant injury [27]. A possible explanation for this phenomenon was the interference of IVIG with lymphodepletion by antithymocyte globulin because CD3⫹ cell counts remained elevated. Finally, preconditioning regimens consisting of plasmapheresis and lowdose cytomegalovirus hyperimmunoglobulin, along with standard immunosuppression, have even allowed successful transplantation of simultaneous ABO-incompatible and crossmatch–positive kidneys [28]. Although the mechanisms seem to be very complex and not fully understood, IVIG is thought to act mainly by neutralizing and/or removing donor specific antibodies by antiidiotypic antibodies present in such preparations [29]. This action depends on the half-life of infused IVIG. Besides this passive action, IVIG may actively inhibit B-cell production of donor-specific alloantibodies by different mechanisms that last longer than the halflife of IVIG. First, they can control the B-cell repertoire that migrates from the bone marrow to the peripheral lymphoid organs, although this has been only described in murine models [30]. Second, IVIG may interact with CD32 or Fc␥ receptoer IIB (Fc␥RIIB) that transmits inhibitory signals on B cells, and it may even induce apoptosis [31]. Finally, it has been demonstrated that IVIG inhibits the expression of the costimulatory molecule CD19 on the surface of activated human B cells [24]. On the other hand, immunoglobulins present in IVIG preparations may block complement-mediated endothelial injury through the high avidity of their Fc fragments for C3b and C4b. Through this mechanism, the complement cascade is blocked and the generation of the membrane-attack complex is abolished. As noted for CD32 on B cells, IVIG can block the binding of alloantibodies to the activating Fc receptors on inflammatory leukocytes, such as macrophages and neutrophils, responsible in part for the DGF [29]. Polymorphisms of tumor necrosis factor (TNF)-␣ associated with a higher TNF-␣ expression have been recently associated with a slight risk for DGF [32] and IVIG might also interfere with it by inducing the expression of antiinflammatory cyto-
DGF in Hypersensitized Kidney Recipients
375
FIGURE 5 Kaplan-Meier analysis of graft survival comparing patients according to current and peak panel-reactive antibody (PRA) (A) and to existence of delayed graft function (DGA) and acute rejection (AR) episodes (B). (A) Patients with current (thick solid line) and peak (dashed line) PRA over 50% (squares) exhibited a significantly worse graft survival than those patients with PRA ⬍50% (circles, p ⬍ 0.001 in both cases). (B) Development of AR (dashed line) was associated with a poor graft survival, especially when it was combined with the development of DGF (circles).
kines [7, 29]. At the same time, IVIG contains antibodies directed to multiple cytokines (i.e., interleukin [IL]1␣, IL-6, TNF-␣) [7]. Taken together, although the main function attributed to the therapy with IVIG in HS kidney recipients is related to the reduction in the titers of HLA antibodies and inhibition of humoral immune responses, they also inhibit the function of most of the
inflammatory cells (from endothelial cells to T cells) that participate in the ischemia-reperfusion injury during kidney transplantation. Efforts to decrease the number of HS patients on the waiting list for a renal transplant are being made at our center in collaboration with other centers from northern Spain. The proposed protocol is based on that of Jordan et
376
al. [24]. Patients must have a PRA ⬎50% and a time on the waiting list of more than 6 months. The pretransplant protocol is based on the therapy with IVIG at dosages of 2 g/kg once a month, for three consecutive months. One week after the last infusion, the measurement of HLA antibodies should be performed and must reflect a decrease of 50% in PRA. In such a case, transplantation must be performed immediately with a blood group– compatible, negative crossmatch organ (HLA matching is not necessary). From a clinical point of view, the possibility of using a previous in vitro test to demonstrate a PRA reduction provides a rational basis for the selection of HS patients to be transplanted [24]. Postransplant therapy includes thymoglobulin (in the first 10 days after transplantation), steroids, mycophenolate mofetil, and tacrolimus, together with IVIG (at 48 hours, 20 days and 40 days after transplantation). Obviously, in this type of protocol, active exchange programs are especially necessary. There is no doubt that desensitization with IVIG has opened doors to HS patients. Nonetheless, this approach is novel, and results, although promising, are still scarce. Data indicate that allosensitization and the response to desensitization are heterogeneous. Further studies will add important knowledge to this field. In summary, the association of DGF and HLA antibodies is deleterious. Several mechanisms that lead to an ischemia injury with inflammatory phenomena into the kidney have been suggested. Such injury finally results in an impairment in the glomerular filtration rate and a poor graft survival. New approaches, including IVIG therapy, will permit better management of DGF and improved graft survival in these immunologically high-risk kidney recipients. ACKNOWLEDGMENT
This work was supported by the Fondo de Investigación Sanitaria (RITC C03/03).
REFERENCES 1. Pascual M, Theruvath T, Kawai T, Tolkoff-Rubin N, Cosimi AB: Strategies to improve long term outcomes after renal transplantation. N Engl J Med 346:580, 2002. 2. Cecka JM: The OPTN/UNOS renal transplant registry 2003. Clin Transpl 1, 2003. 3. Ojo AO, Hanson JA, Meier-Kriesche H, Okechukwu CN, Wolfe RA, Leichtman AB, Agodoa LY, Kaplan B, Port FK: Survival in recipients of marginal cadaveric donor kidneys compared with other recipients and wait-listed transplant candidates. J Am Soc Nephrol 12:589, 2001. 4. Sanfilippo F, Vaughn WK, Bollinger RR, Spees EK: Comparative effects of pregnancy, transfusion, and prior graft rejection on sensitization and renal transplant results. Transplantation 34:360, 1982.
M. López-Hoyos et al.
5. Gebel HM, Bray RA, Nickerson P: Pre-transplant assessment of donor-reactive, HLA-specific antibodies in renal transplantation: contraindication vs risk. Am J Transplant 3:1488, 2003. 6. Peeters P, Terryn W, Vanholder R, Lameire N: Delayed graft function in renal transplantation. Curr Opin Crit Care 10:489, 2004. 7. Jordan S, Cunningham-Rundles C, McEwan R: Utility of intravenous immune globulin in kidney transplantation: efficacy, safety, and cost implications. Am J Transplant 3:653, 2003. 8. Halloran PF, Hunsicker LG: Delayed graft function: state of the art. Summit meeting, Scottsdale, AZ, November 10 –11, 2000. Am J Transplant 1:115, 2001. 9. Govani MV, Kwon O, Batiuk TD, Milgrom ML, Filo RS: Creatinine reduction ratio and 24-hour creatinine excretion on posttransplant day two: simple and objective tools to define graft function. J Am Soc Nephrol 13:1645, 2002. 10. Rodrigo E, Ruiz JC, Piñera C, Fernandez-Fresnedo G, Escallada R, Palomar R, Cotorruelo JG, Zubimendi JA, Martin de Francisco AL, Arias M: Creatinine reduction ratio on post-transplant day two as criterion in defining delayed graft function. Am J Transplant 4:1163, 2004. 11. Park JH, Yang CW, Kim YS, Lee SH, Choi YJ, Kim YS, Moon IS, Koh YB, Bang BK: Comparisons of clinicopathological correlations between immediate and slow graft function in renal transplant recipients. Clin Transplant 16:18, 2002. 12. McTaggart RA, Tomlanovich S, Bostrom A, Roberts JP, Feng S: Comparison of outcomes after delayed graft function: sirolimus-based versus other calcineurin-inhibitor sparing induction immunosuppression regimens. Transplantation 78:475, 2004. 13. Cecka JM, Shoskes DA, Gjertson DW: Clinical impact of delayed graft function for kidney transplantation. Transplant Rev 15:57, 2001. 14. Qureshi F, Rabb H, Kasiske BL: Silent acute rejection during prolonged delayed graft function reduces kidney allograft survival. Transplantation 74:1400, 2002. 15. Lopez-Hoyos M, Fernandez-Fresnedo G, Pastor JM, Arias M: Anticuerpos anti-HLA postrasplante: un nuevo método de monitorización. Nefrologia 24:62, 2004. 16. Arias M: Impact of the delayed graft function in hypersensitized kidney transplant recipients. Transplant Proc 35:1655, 2003. 17. Marcen R, Orofino L, Pascual J, De la Cal MA, Teruel JL, Villafruela JJ, Rivera ME, Mampaso F, Burgos FJ, Ortuño J: Delayed graft function does not reduce the survival of renal transplant allografts. Transplantation 66:461, 1998. 18. Fernandez-Juarez G, Marcen R, Pascual R, Teruel JL, Rivera ME, Villafruela JJ, Mampaso F, Burgos FJ, Ortuño J: Prolonged delayed graft function decreases graft sur-
DGF in Hypersensitized Kidney Recipients
19.
20.
21.
22.
23. 24.
25.
vival in transplant patients taking cyclosporine. Transplant Proc 34:338, 2002. Boom H, Mallat MJK, De Fijter JW, Zwinderman AH, Paul LC: Delayed graft function influences renal function, but not survival. Kidney Int 58:859, 2000. Gentil MA, Alcaide MP, Algarra GR, Pereira P, Toro J, Gonzalez-Roncero F, Lopez M, Bernal G, Mateos J: Impact of delayed graft function on cadaveric kidney transplant outcome. Transplant Proc 35:689, 2003. McLaren AJ, Jassem W, Gray DWR, Fuggle SV, Welsh KI, Morris PJ: Delayed graft function: risk factors and the relative effects of early function and acute rejection on long-term survival in cadaveric renal transplantation. Clin Transplant 13:266, 1999. Bryan CF, Luger AM, Martinez J, Muruve N, Nelson PW, Pierce GE, Ross G, Shield CF, Warady BA, Aeder MI, Helling TS: Cold ischemia time: an independent predictor of increased HLA class I antibody production after rejection of a primary cadaveric renal allograft. Transplantation 71: 875, 2001. Friedewald JJ, Rabb H: Inflammatory cells in ischemic acute renal failure. Kidney Int 66:486, 2004. Jordan SC, Vo A, Bunnapradist S, Toyoda M, Peng A, Puliyanda D, Kamil E, Tyan D: intravenous immune globulin treatment inhibits crossmatch positivity and allows for successful transplantation of incompatible organs in living-donor and cadaver recipients. Transplantation 76:631, 2003. Akalin E, Ames S, Sehgal V, Fotino M, Daly L, Murphy B, Bromberg JS: intravenous immunoglobulin and thymoglobulin facilitate kidney transplantation in complementdependent cytotoxicity B-cell and flow cytometry T- or
377
26.
27.
28.
29.
30.
31.
32.
B-cell crossmatch-positive patients. Transplantation 76:1444, 2003. Gloor JM, DeGoey S, Ploeger N, Gebel H, Bray R, Moore B, Dean PG, Stegall MD: Persistence of low levels of alloantibody after desensitization in crossmatch-positive living-donor kidney transplantation. Transplantation 78: 221, 2004. Gallay BJ, Perez RV, Ramsamooj R: Acute renal transplant injury and interaction between antithymocyte globulin and pooled human immunoglobulin. Clin Transplant 18:327, 2004. Warren DS, Zachary AA, Sonnenday CJ, King KE, Cooper M, Ratner LE, Shirey RS, Haas M, Leffell MS, Montgomery RA: Successful renal transplantation across simultaneous ABO incompatible and positive crossmatch barriers. Am J Transplant 4:561, 2004. Kazatchkine MD, Kaveri SV: Immunomodulation of autoimmune and inflammatory diseases with intravenous immune globulin. N Engl J Med 345:747, 2001. Sundblad A, Marcos M, Malanchere E, Castro A, Haury M, Huetz F, Nobrega A, Freitas A, Coutinho A: Observations on the mode of action of normal immunoglobulin at high doses. Immunol Rev 139:125, 1994. Toyoda M, Pao A, Petrosian A, Jordan SC: Pooled human gammaglobulin modulates surface molecule expression and induces apoptosis in human B cells. Am J Transplant 3:156, 2003. Alakulppi NS, Kyllönen LE, Jäntti VT, Matinlauri IH, Partanen J, Salmela KT, Laine JT: Cytokine gene polymorphisms and risks of acute rejection and delayed graft function after kidney transplantation. Transplantation 78: 1422, 2004.