Objectives: Antibody-mediated rejection in patients with positive crossmatches can be severe and result in sudden onset of oliguria, leading to graft loss. In an attempt to prevent posttransplant oliguria, we adopted a preoperative desensitization protocol involving the use of high-dose intravenous immunoglobulin/plasmapheresis and the anti-CD20 antibody, rituximab, in 41 transplant recipients with positive crossmatch test results.
Materials and Methods: We retrospectively examined the clinical courses of the 41 kidney transplant recipients, paying special attention to renal graft function, urine volume, and changes in the titers of donor-specific antibodies.
Results: Four grafts were lost during an average of 4.5-year follow-up. Average graft function was excellent, with a serum creatinine level of 1.3 ± 0.4 mg/dL. Sufficient urine output, with no oliguria or anuria, was achieved postoperatively in 40 of the 41 patients. However, among the 34 patients who underwent graft biopsies, the biopsies revealed acute antibody-mediated rejection in 21 patients (62%), and chronic antibody-mediated rejection in 10 patients (30%).
Conclusions: The high-dose intravenous immunoglobulin treatment included in our desensitization protocol was shown to be safe and effective for achieving successful transplant outcomes and allowed the avoidance of more aggressive B-cell-targeted treatments, such as C5 inhibitors and/or proteosome inhibitors, for preventing posttransplant oliguria and anuria.
Key words : Antibody-mediated rejection, Anuria, Graft dysfunction, Human leukocyte antigen, Oliguria, Plasma exchange, Positive crossmatch tests
Sensitization to human leukocyte antigens (HLA antigens) is one of the most important hurdles to overcome for successful kidney transplant.1 Severe antibody-mediated rejection (AMR) caused by donor-specific antibodies (DSAs) in sensitized recipients is characterized by acute onset of insufficient urine output, oliguria or anuria, and renal dysfunction in the posttransplant period. Without suitable treatment for this critical situation, most cases eventually result in graft loss after a rapid rise in the titers of DSAs and formation of fibrin thrombi and cortical necrosis.2-6 To overcome the critical posttransplant situation of oliguria/anuria, a John Hopkins’ team reported splenectomy and/or eculizumab administration as effective salvage therapies in addition to preoperative desensitization induction with low-dose intravenous immunoglobulin (L-IVIg)/plasmapheresis and rituximab.7-9
In practical clinical settings, it is often difficult to reliably assess the underlying critical condition in the presence of insufficient urine output and to provide appropriate treatment to recipients because these patients may have already received various kinds of desensitization treatments (like plasmapheresis) before transplant. Fluid imbalance in recipients after desensitization may make it difficult to precisely assess the water balance before administration of any salvage treatment.10-12 Thus, suitable therapies to prevent the development of oliguria/anuria in the posttransplant period are desirable.
Since 2011, we have used a desensitization protocol consisting of high-dose IVIg (H-IVIg)/plasmapheresis and high-dose rituximab for kidney transplant recipients with positive crossmatch test results.13 In this retrospective study, we examined the clinical courses of 41 highly sensitized recipients who received desensitization according to this latest protocol, with special attention paid to the onset of oliguria/anuria and renal graft function.
Materials and Methods
Data were extracted from the Japan Academic Consortium of Kidney Transplantation (UMIN Clinical Trials Registry number UMIN 000018327). This study was approved by the ethical committee at the Tokyo Women’s Medical University (approval number 3366R) and was consistent with the 2000 Declaration of Helsinki as well as the 2008 Declaration of Istanbul. All patients provided consent for use of their anonymized data before registration in the Japan Academic Consortium. Details that could disclose the identity of the study patients were omitted.
Figure 1 presents a flowchart showing patient exclusion and inclusion criteria. Between 2011 and 2020, we performed a total of 1102 kidney transplant procedures at the Tokyo Women’s Medical University Hospital. Patients with only DSAs (n = 321) detected by Luminex were not enrolled in this study. In addition, pediatric recipients (n = 113), recipients with no HLA antibodies (n = 500), ABO-incompatible kidney transplants (n = 324), deceased donor kidney transplants (n = 61), and recipients with non-HLA antibodies (n = 65) were excluded from this study. Thirty-four recipients with surgical complications of the urinary tract followed by reoperation or insertion of double J stents were also excluded because these complication would preclude accurate assessment of urine output. There were 41 remaining patients with positive crossmatch tests (positive with either the flow cytometric crossmatch [FCXM] test or the complement-dependent crossmatch [CDCXM] test). A 100% positive result in the CDCXM test is considered as a contraindication for kidney transplant at our institution, similar to protocols in other institutions around the world.
Figure 2 shows the latest immunosuppressive regimen used at our center; this regimen was established in 2011 and includes use of H-IVIg/plasmapheresis and rituximab. In brief, triple immunosuppression with tacrolimus, mycophenolate mofetil, and methylprednisolone was started 1 month before transplant. Figure 2 also shows the preoperative doses of tacrolimus, mycophenolate mofetil, and methylprednisone. After transplant, the dose of tacrolimus was adjusted according to target trough levels,14 as shown in Figure 2.
High-dose intravenous immunoglobulin was administered to the highly sensitized recipients in 4 or 5 divided doses, because H-IVIg has too much fluid and is too viscous to administer to patients undergoing maintenance hemodialysis. The total IVIg dose is 2 g/kg/body weight (body wt) for patients with a positive FCXM test and 4 g/kg/body wt for patients with a positive FCXM and positive CDCXM test. Headache, which can be observed as an adverse event during intravenous infusion of IVIg, can be resolved by lowering the speed of infusion. The number of double-filtration plasmapheresis (DFPP) sessions for these patients is normally 3 or 4; however, this number could vary depending on the antibody titers. Plasmapheresis is performed on alternate days rather than every day, from the point of view of the efficacy of antibody removal.10,12 If crossmatch test positivity persists until immediately before transplant, the last plasmapheresis session is changed from DFPP to whole plasma exchange (PEX). The exchange volume at the time of PEX is calculated according to a previously described formula.10 Rituximab is given to all highly sensitized recipients at a total dose of 500 mg/body, provided on 2 separate days at 200 mg and 300 mg.
Postoperative patient management after transplant surgery
The protocol for postoperative management after transplant surgery is as follows: adjust infusion speed of saline every 2 hours according to urine output over the previous 2 hours in the intensive care unit and then administer the same amount of saline as the urine output in the previous 2 hours by intravenous infusion over the next 2 hours. However, maximum infusion speed should be set at 200 mL/hour to prevent pulmonary edema, and the minimum speed should be set at 50 mL/hour. If there are no problems on posttransplant day 1, the patient is moved from the intensive care unit to the general ward. After the move to the general ward, the amount of fluid administered by intravenous infusion is normally 2000 to 2500 mL/day, followed by tapering to 500 mL/day and/or oral intake of water and foods within 4 days after transplant. On posttransplant day 2, the recipient is required to stand up at bedside to check body weight. In sensitized recipients, body weight is often higher due to insufficient urine output compared with dry weight at the time of the previous dialysis. This overweight state is controlled by the administration of diuretics, with dose adjusted so as to lower the body weight to the previously measured dry weight maintained during the period of hemodialysis. Doppler ultrasonography is performed at the bedside daily by a nephrologist to confirm the renal blood flow. A double J stent is not routinely inserted into the site of the bladder anastomosis. On posttransplant day 4, the balloon urethra catheter is removed, except for patients who have small bladder capacity of less than 50 mL.
Complement-dependent cytotoxicity test, flow cytometric crossmatch test, and solid-phase assay using Luminex single antigen beads
The crossmatch test using FCXM is superior to conventional CDCXM in terms of its sensitivity for detecting DSAs. Since 1983, when this finding was first reported, we have conducted crossmatch examinations using both methods.15 For the CDCXM test, the recipient’s serum potentially containing anti-HLA DSAs is added to the donor lymphocytes, along with complement. The proportion of lysed cells is assessed, and the crossmatch is graded as being mildly, moderately, or strongly positive.
For the FCXM test, donor lymphocytes are added to the recipient’s serum, followed by incubation for 30 minutes at room temperature. After 2 washes, phycoerythrin-labeled CD19 (Pharmingen) and cytochrome-labeled CD3 (Pharmingen) are added, and the reaction is allowed to occur for 30 minutes at 4 °C. After 3 washes, fluorescein isothiocyanate-labeled anti-human immunoglobulin G antibody (Pharmingen) is added as the secondary antibody, and the cells are fixed in 2% formalin-phosphate buffered saline. The FCXM test is performed using a fluorescein-activated cell sorter (Becton Dickinson), and a positive FCXM is defined by a channel shift of >10.
For the solid-phase assay (Luminex single antigen bead assay), manufacturer’s instructions are followed. The positive range at our institution is defined as a mean fluorescence intensity of over 800.
Patients undergo 2 or 3 biopsies with a 16-gauge needle. All recipients undergo posttransplant protocol biopsies (at 0 hour, at <6 months, and at >6 months), as well as episode biopsies. All biopsies are evaluated by light microscopy and immunofluorescence staining for C4d. In brief, the specimens are fixed with 10% phosphate-buffered formalin (pH 7.2), embedded in paraffin, and cut into 2-μm sections. The sections are stained with hematoxylin and eosin, periodic acid-Schiff, Masson’s trichrome, and periodic acid methenamine silver stains for light microscopy.
For immunohistochemistry, paraffin sections on glass slides coated with saline are stained with a peroxidase-labeled streptavidin-biotin staining kit (DAKO). The primary antibodies used are rabbit polyclonal antibodies against immunoglobulin G, immunoglobulin A, C3, and immunoglobulin M (Hoechst, Behringwerke). Pathological findings are then classified according to the Banff 2007 working classification and the Banff 2005 updated edition and comparatively evaluated among the recipients classified according to sensitization status.16,17 Briefly, AMR was defined as (1) C4d and/or (rarely) immunoglobulin deposition in the peritubular capillaries, (2) serologic evidence of circulating antibodies to donor HLA antigens, and (3) morphologic evidence of acute tissue injuries. Chronic AMR was also defined using the Banff criteria. At our institution, a single pathologist made the diagnoses based on examination of the graft specimens.
Data are expressed as means ± SD. Statistical analysis was performed with SAS version 9.4 TS1M5 (SAS Institute). One-way analysis of variance was used to compare normally distributed continuous variables, and the Kruskal-Wallis H test was used to evaluate skewed or discrete ordinal variables. The chi-square test was used to compare nominal scale variables. A 2-tailed P value < .05 was considered statistically significant by the biostatistics datacenter (STATZ Institute, Tokyo Japan).13
Patient and donor background and clinical course after transplant
As shown in Table 1, the total number of recipients enrolled in this study was 41, consisting of 19 men and 22 women. The immunological test status was CDCXM positive and FCXM positive in 7 recipients and CDCXM negative and FCXM positive in the remaining 34 recipients. The cause of sensitization was a history of kidney transplant or transplant of other organs (ie, heart and liver) in 21 recipients and sensitization by pregnancy, such as in a spousal pair (husband to wife), in 15 recipients. The remaining 5 recipients were sensitized by blood transfusion or the reason was indeterminate. The average dose of IVIg in these recipients was 2.5 ± 2.0 g/kg/body wt. The average number of DFPP in the posttransplant period was 4.5 ± 1.2. Table 2 shows the background of donors in this study.
As shown in Table 3, the follow-up period was 4.5 ± 3.5 years. Four of the grafts were lost (4/41, 10%) during this period. Cause of graft loss was suicide and chronic AMR in 1 recipient each and cardiovascular events in the remaining 2 recipients. Average serum creatinine level and eGFR were 1.3 ± 0.4 mg/dL and 49 ± 15 mL/min, respectively, except in the 4 patients with graft loss.
Urine output and peripheral blood platelet count in the 41 desensitized recipients in the early phase posttransplant
Figure 3 and Figure 4 show the average daily urine output and the average daily peripheral blood platelet count, respectively, in the 41 recipients of this study during the first 15 days posttransplant. All recipients showed a gradual decrease of urine output accompanied by a decrease in platelet count immediately posttransplant.
As shown in Figure 3, the maximum urine output was 2600 ± 1100 mL/day on posttransplant day 1, whereas the minimum urine output was 1320 ± 620 mL/day on day 7.0 ± 4.5 after transplant surgery. The urine outputs on posttransplant days 3, 5, 7, 9, and 11 were significantly less compared with the urine output on posttransplant day 1 (P < .05). As shown in Figure 4, the minimum platelet count of 9.5 × 104 ± 3.2 × 104/mm3, on average, was observed 3.0 ± 2.1 days posttransplant. Platelet counts on posttransplant days 1, 3, 5, 7, and 9 were significantly lower than the count before transplant (P < .05). The changes in the peripheral blood platelet count and urine output ran almost parallel to each other; that is, the daily urine output decreased and increased in accordance with decrease and increase of the peripheral blood platelet count.
Of the 41 recipients, only 1 recipient showed oliguria with a urine output of less than 400 mL/day posttransplant. This patient was the only patient in this study who temporarily needed 2 sessions each of hemodialysis and extracorporeal ultrafiltration method to maintain body weight and to remove uremic toxins. Other than this patient, none of the patients needed hemodialysis or any other rescue treatments, including additional rituximab, plasmapheresis, or IVIg. None of the 41 recipients who received desensitization according to the latest protocol showed anuria with a urine output of less than 100 mL/day.
Titer changes of donor-specific antibodies after the desensitization protocol
Figure 5 shows the titer change of DSAs after the desensitization protocol and at last follow-up, compared with baseline titers, in 15 highly sensitized recipients. The monitoring period from before to after transplant was almost 4 weeks. Last follow-up monitoring refers to monitoring for DSAs at the latest outpatient visit. Among the 41 recipients, 15 recipients whose serum samples had been refrigerated and stored were investigated by Luminex assay. As shown in Figure 5, in general, class I antibodies (patients 1, 4, 5, 7, 9, 10) as well as DQ antibodies (patients 2, 3, 9, 11, 15), regardless of the titers, were responsive to this latest desensitization protocol. DR antibodies were not very responsive to the desensitization protocol in the early phase; however, low titers of DR antibodies (patients 9, 13, 14) responded to the desensitization regimen in the early phase, but high titers of DR antibodies (patient 8) proved quite resistant. Patient 3 had high titers of DR4 and its cross-reactive group DR53. The titers of these antibodies decreased transiently after desensitization, only to increase again, eventually resulting in chronic AMR. Patient 6 showed markedly high titers of DSAs (DR12), which were responsive to the sensitization regimen because this DR12 alone may be produced without the related cross-reactive group. Patient 6 has remained stable with very mild AMR; in patient 3, the graft was lost with the development of severe AMR.
As shown in Table 3, among the 41 recipients, 34 recipients underwent graft biopsies, which revealed acute AMR in 21 patients (62%), acute T-cell-mediated rejection in 14 patients (41%), chronic AMR in 10 patients (30%), and interstitial fibrosis-tubular atrophy in 6 patients (18%). Graft biopsies among the 34 recipients showed no evidence of rejection.
Despite the numerous efforts made to increase the availability of organs from deceased donors by education and by improvements in public awareness, organ donation still remains modest globally. In Japan, more than 90% of kidney transplant procedures during the past 2 decades have been from living donors.18 No donor exchange program has been developed yet in Japan, most likely because of ethical and religious barriers. In contrast, the number of repeat transplant candidates, who constitute an immunologically high-risk population, on deceased- and living-donor transplant wait lists have been increasing rapidly year by year because of the high level of panel reactive antibody assays.18
Immunologically high-risk recipients sometimes experience unusual clinical manifestations, such as oliguria/anuria,2 low-grade fever, spiky systolic blood flow with low diastolic blood flow, indicative of high resistance of the peripheral renal blood flow, on Doppler ultrasonography, and graft dysfunction in the posttransplant period. Graft biopsies can reveal evidence of typical AMR, including peritubular capillaritis, glomerulitis, and deposition of C4d according to Banff criteria.16,17 Although most of these immunologically high-risk recipients are likely to respond to the usual treatment of a short course of L-IVIg/plasmapheresis and low-dose rituximab (200 mg/body; anti-CD20 antibody) with recovery of sufficient urine volume, some recipients can have severe AMR accompanied by oliguria/anuria, which can lead to graft dysfunction or graft loss. Insufficient urine volume at the onset of oliguria/anuria immediately after kidney transplant makes it difficult to treat these patients because sensitized recipients often receive aggressive desensitization treatments prior to transplant, including high-dose globulin administration and many sessions of plasmapheresis. These courses of treatment often result in fluid imbalance inside versus outside the blood vessels in the recipient’s body, thus making it difficult to administer appropriate treatment in recipients with insufficient urine volume in the immediate posttransplant period.
For candidates of HLA-incompatible kidney transplant procedures, we adopted a desensitization protocol in 2005 consisting of plasmapheresis and low-dose rituximab (200 mg/body); this protocol is almost similar to the ABO-incompatible kidney transplant protocol that was already in place since 2000.19-22 Unfortunately, a protocol that includes only plasmapheresis and low-dose rituximab can be ineffective in recipients with high titers of anti-HLA antibodies and strongly positive crossmatch tests, regardless of whether the FCXM or CDCXM test was used. Most of these recipients can show insufficient urine volume and graft dysfunction. Although rescue therapy of additional use of B-cell-targeted treatments could be used postoperatively, not all patients respond, resulting in graft loss.
To prevent oliguria/anuria and inhibit humoral immunity in the posttransplant period, in 2011, we adopted a protocol consisting of H-IVIg (2-4 g/kg), in addition to plasmapheresis and high-dose rituximab (500 mg/body), with increase of rituximab from 200 to 500 mg/body in line with other studies.23-26
In our retrospective study of 41 immunologically high-risk recipients who underwent this new desensitization protocol between 2011 and 2020 at a single center study, only 1 recipient (patient 3; see Table 3) developed oliguria posttransplant. He had three DSAs (DR4, DR53, DQ8), including a cross-reactive group antibody. The titers of DR4 and DR53 decreased after a series of desensitization treatments, only to increase again at the last follow-up, with eventual graft loss by 2 years posttransplant. The titer of DQ8 continued to be suppressed after the desensitization. Graft biopsy showed chronic severe active AMR, as well as mixed-type rejection. In contrast, the single appearance of class II DR12 in patient 6, of DR15 in patient 8, and of DR7 in patient 12 continued to be suppressed, even though the titers were quite high before the desensitization treatment. These class II antibodies may be produced alone without related production of a cross-reactive group antibody. These changes in DSAs according to the DSA titers and DSA class after desensitization are almost similar to those described in previous reports.23-26
Of note, we found that changes in urine output were almost parallel to those of the peripheral blood platelet count in the posttransplant period. The decreases in urine output and peripheral blood platelet count may be caused by immunological responses of AMR and/or a preoperative aggressive desensitization protocol for DSAs (plasmapheresis and PEX), although these changes were not observed in any ABO-incompatible transplants under the almost similar desensitization protocol. The difference in the clinical courses between HLA-incompatible and ABO-incompatible settings may be because of difference in physiological characteristics of each antibody, anti-HLA, and anti-blood type antibodies.14 No rescue therapy for acute AMR was adopted for any of the patients in our study, except for patient 3, because of excellent and stable graft function accompanied by good blood flow to the graft shown on Doppler ultrasonography. Of 41 patients in our study, all but patient 3 showed stable urine output of at least 1000 mL/day without any additional treatment. Postoperative splenectomy and/or administration of the C5 inhibitor eculizumab has been reported as an effective treatment for posttransplant oliguria with a urine output of 400 mL/day or less to counter severe AMR.1-3 However, although we have used these treatments as rescue for patients with severe AMR at our center, we believe that emergency operations, such as splenectomy, immediately after transplant27 and also several administrations of eculizumab28 and/or bortezomib29 in severely desensitized recipients who have already received desensitization therapy may increase not only the surgical risk but also the infectious risk.27-29 Thus, we believe the combination of H-IVIg/plasmapheresis and high-level rituximab may be an effective noninvasive treatment to prevent posttransplant oliguria/anuria.
Both H-IVIg and L-IVIg are used in the transplant setting to reduce sensitization and to treat steroid-resistant rejections. The effectiveness of IVIg has been reported mainly for high-dose H-IVIg given monthly.30,31 In contrast, the effectiveness of lower doses of IVIg, such as 100 to 500 mg/kg, remains difficult to assess. The uncertainty about the numerous mechanisms involved in the control of humoral activities in the body definitely contributes to the uncertainty of how best IVIg should be used in the field of transplantation. In other fields, for example, in the neurological field, H-IVIg has become established over the past few decades as an important component, exerting dose-dependent effects in the management of autoimmune disease.32,33
Among the limitations of this study were its retrospective nature, lack of a comparator group, and the small sample size. Prospective multicenter comparative studies of various desensitized protocols used to avoid posttransplant oliguria/anuria are needed. Also, studies that include a longer follow-up duration for determining the graft survival rate are also needed because of the high incidence of AMR after transplant previously reported by our team in patients treated with H-IVIg.10 Aubert and colleagues reported that AMR with preexisting DSA was associated with an acceptable and better allograft survival compared with AMR with de novo DSAs during a similar 4-year follow-up, supporting the transplant of highly sensitized patients and the early detection of AMR.34 Although since December 2019 H-IVIg has been covered by the National Health Insurance in Japan, another limitation is its high cost. With the consideration of medical economics and the scarcity of medical resources like H-IVIg, H-IVIg should be limited to patients with a strongly positive crossmatch test, and not patients with only DSA but a negative crossmatch test, prior to transplant.
Volume : 19
Issue : 10
Pages : 1032 - 1040
DOI : 10.6002/ect.2021.0234
From the Department of Urology, Tokyo Women’s Medical University, Tokyo, Japan
Acknowledgements: We acknowledge the support of Katsunori Shimada, PhD (STATZ Institute, Tokyo, Japan), who provided expert assistance with the statistical analysis. The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest.
Corresponding author: Hideki Ishida, Department of Urology, Tokyo Women’s Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo, Japan 162-8666
Figure 1. Patient Flowchart
Figure 2. Immunosuppressive Regimens for Patients With Strong Sensitization at Tokyo Women’s Medical University Hospital Between 2011 and 2020
Table 1. Background of Recipients Before Transplant
Table 2. Background of Donors Before Transplant
Table 3. Results of Recipients After Transplant
Figure 3. Urine Output in the 41 Highly Sensitized Recipients Treated With the Latest Protocol
Figure 4. Peripheral Blood Platelet Count in the 41 Highly Sensitized Recipients Treated With the Latest Protocol Before and After Transplantl
Figure 5. Titer Changes of Donor-Specific Antibodies Before and After Desensitization Protocol