Intravenous Immunoglobulin and
Plasmapheresis in Acute Humoral
Rejection: Experience in Renal
Allograft Transplantation
Ruediger W. Lehrich, Paulo N. Rocha,
Nancy Reinsmoen, Arthur Greenberg,
David W. Butterly, David N. Howell, and
Stephen R. Smith
ABSTRACT: Acute humoral rejection (AHR) in kidney treated with IVIG and PP. Two-year graft survival was
transplantation is associated with higher rates of allograft numerically worse in patients with AHR versus ACR
loss when compared with acute cellular rejection (ACR). (78% vs. 85%, p 0.5) but the difference was not
Treatment with intravenous immunoglobulin (IVIG) statistically significant. Graft survival after AHR treated
combined with plasmapheresis (PP) has been used re- with IVIG and PP is much better than it has been
cently in many centers. We report the incidence, clinical historically. IVIG in combination with PP is an effective
characteristics, and outcome of patients with AHR treated treatment for AHR. Graft survival in this setting is
with IVIG and PP. All patients (n 519) at our insti- similar to graft survival in patients with ACR. Human
tution who underwent kidney transplantation between Immunology 66, 350 –358 (2005). © American Society for
January 1999 and August 2003 were retrospectively an- Histocompatibility and Immunogenetics, 2005. Pub-
alyzed and classified according to biopsy results into three lished by Elsevier Inc.
groups: AHR, ACR, and no rejection. AHR was diag-
nosed in 23 patients (4.5%) and ACR in 75 patients KEYWORDS: renal allografts; acute humoral rejection;
(15%). Mean follow-up was 844 23 days. Female sex, intravenous immunoglobulin; plasmapheresis; allograft
black race, and high panel-reactive antibody were risk survival
factors for AHR. Most AHR patients (22 of 23) were
ABBREVIATIONS
ACR acute cellular rejection IVIG intravenous immunoglobulin
AHR acute humoral rejection No REJ no rejection
CDC-AHG complement-dependent cytotoxicity PP plasmapheresis
technique enhanced with antihuman PRA panel-reactive antibody
globulin
INTRODUCTION
Acute humoral rejection (AHR) occurs early, usually and intravascular coagulation [1]. Antidonor antibodies
within 24 to 48 hours after reperfusion. Characteristic are involved in its pathogenesis [2, 3]. Saadi et al. [4]
pathologic features include severe injury to endothelial hypothesized that binding of antidonor antibodies to
cells lining small blood vessels, inflammatory infiltrates, vascular endothelial cells in concert with activation of
From the Duke University Medical Center, Department of Medicine, Address reprint requests to: Dr. Ruediger W. Lehrich, Duke University
Division of Nephrology, Durham, NC, USA (R.W.L., D.W.B., A.G., Medical Center, Department of Medicine, Division of Nephrology, PO Box
S.R.S.); Federal University of Bahia, Department of Medicine, Salvador, 3014, Trent Drive, Durham, NC 27710; Tel: (919) 660-6857; Fax:
BA, Brazil (P.N.R.); and Duke University Medical Center, Department of (910) 684-4476; E-mail: Lehri001@mc.duke.edu.
Pathology, Durham, NC, USA (N.R., D.N.H.). R.W.L. and P.N.R. contributed equally to this study.
Received December 15, 2004; accepted January 19, 2005.
Human Immunology 66, 350 –358 (2005)
© American Society for Histocompatibility and Immunogenetics, 2005 0198-8859/05/$–see front matter
Published by Elsevier Inc. doi:10.1016/j.humimm.2005.01.028

IVIG and Plasmapheresis in Humoral Rejection 351
complement leads to a process that he termed the “acti- sity Medical Center between January 1999 and August
vation” of endothelium. The endothelium is rendered 2003 (n 519). Data collection consisted of a review of
permeable, allowing penetration of immunocompetent medical records. We had a preexisting database contain-
cells, leading to the characteristic picture of inflamma- ing demographic and clinical information on all patients
tion, thrombosis and ischemia [5]. AHR is less fre- transplanted between January 1999 and August 2001;
quently encountered than acute cellular rejection (ACR). data for patients transplanted between August 2001 and
The latter responds well to therapy directed against T August 2003 were added to this database. Follow-up for
lymphocytes as a preventive and therapeutic strategy [6]. all patients was extended to August 2004.
AHR is relatively unresponsive to therapies that target T Pathology and Immunopathology
lymphocytes. Treatment focuses rather on removal of
preformed alloantibodies against donor-specific human Biopsies were performed to evaluate allograft dysfunc-
leukocyte antigens (HLA). This can be accomplished by tion; protocol biopsies were not performed. Biopsy re-
means of plasmapheresis (PP) in combination with im- sults were used to classify patients into rejection groups,
munosuppressive agents that inhibit B-cell proliferation, namely ACR or AHR. Patients with no biopsy-proven
such as mycophenolate. evidence of rejection or who did not undergo biopsy were
In renal transplantation, AHR has a poor prognosis for assigned to the no rejection group (No REJ). Transplant
immediate graft survival [6]. Grafts that do survive are biopsy samples were routinely processed and stained by
subject to impaired long-term allograft function, and hematoxylin-eosin, periodic acid–Schiff, methenamine
patients experience early allograft loss [3, 7]. Reports silver, and Masson trichrome methods. ACR was diag-
that used conventional therapy reveal that 1-year graft nosed and graded according to the Banff 1997 criteria
survival does not exceed 15%–50% [8, 9]. New treat- [21]. The diagnosis of AHR was suggested by the fol-
ment strategies that use intravenous immunoglobulin lowing histologic criteria: interstitial infiltrate of inflam-
(IVIG) have been found to be more efficacious. Several matory cells (1) involving more than 25% of biopsy
studies, including a recently published report by our tissue, (2) with predominant focus on peritubular capil-
group [10 –13], describe the combined use of PP or other laries, (3) composed at least in part of neutrophils, and
means of immunoadsorption in conjunction with IVIG (4) associated with minimal tubulitis. AHR was con-
and standard maintenance immunosuppression. firmed on the basis of criteria recently published by the
IVIG has immunomodulatory properties and is effec- Banff group [22]. Patients who had biopsy-proven typ-
tive in treating several autoimmune and inflammatory ical histology for AHR and who had either positive
conditions such as immune thrombocytopenic purpura, staining for C4d or presence of donor-specific HLA an-
hemolytic anemias, and autoimmune neutropenias [14]. tibodies were considered to have confirmed AHR. C4d
Various actions of IVIG have specific relevance to alloan- staining was performed on a frozen section by means of
tibody-mediated acute rejection of transplanted allo- a mouse monoclonal anti-C4d antibody (Biogenesis, San-
grafts. These include neutralization of autoantibodies down, NH). Renal transplant biopsy samples after Feb-
[15], inhibition of activation of endothelial cells [16], ruary 2001 were routinely stained with anti-C4d anti-
downregulation of antibody synthesis as a result of inhi- body. Stored frozen tissue from biopsies performed
bition of B- and T-cell proliferation [17–19], and in- between January 1999 and February 2001 were retro-
creased apoptosis of B cells [20]. These properties of spectively stained for C4d.
IVIG may explain its role as a helpful adjunct in the At the time of initial workup, sera from all kidney
treatment of AHR. transplant candidates were evaluated for the presence of
In our experience with AHR in renal allografts, the anti-HLA immunoglobulin (Ig) G antibodies by means
combined use of IVIG and PP is associated with a 1-year of both the complement-dependent cytotoxicity tech-
graft survival of 81% [10]. In the current study we nique enhanced with antihuman globulin (CDC-AHG)
describe the extension of that experience by expanding and flow cytometry techniques (flow panel-reactive an-
the study group and follow-up. We now describe 23 tibody [PRA]) [23]. The CDC-AHG technique for T-
patients with AHR who were treated with PP and IVIG, cell panel analysis and the complement dependent cyto-
and we report outcome data extending more than 2 toxicity (CDC) (NIH modified, three wash) technique for
years. B-cell panel analysis were used to determine whether any
antibody detected had cytotoxic properties or whether
any IgM was present. Dithiothreitol was used to reduce
MATERIALS AND METHODS
any IgM present, thereby allowing for the detection of
Study Group IgG antibodies by the cytotoxic techniques. Because flow
Our study group consisted of all consecutive kidney or cytometry is three times more sensitive than the CDC-
kidney-pancreas transplants performed at Duke Univer- AHG technique, it was used to determine the presence of

352 R.W. Lehrich et al.
any HLA-directed antibody. Also, the flow cytometry dose of 2 g/kg, was administered after the last PP session.
technique can specifically identify HLA class I– versus However, there was a wide dose variation.
class II– directed IgG antibodies. The primary outcome measure was return to renal
The fine specificity of any antibody detected was replacement therapy after kidney transplantation. Sec-
analyzed by flow cytometry by means of specificity ondary outcome measures were last serum creatinine at
beads, latex beads coated with class I or class II HLA end of follow-up and patient survival.
molecules, as well as single antigen beads and latex beads Statistical Analysis
coated with molecules of a single HLA class I or class II The results were summarized as mean SEM or median
allele (One Lambda, Canoga Park, CA). In this manner, and interquartile range. Continuous variables were com-
the exact specificity of the antibody can be determined to pared by the two-tailed unpaired t-test, and dichotomous
identify alleles in the donor population that would be variables were compared using 2 2 contingency tables
unacceptable antigens. and Fischer’s exact test. Survival analysis was performed
The final cross match included a CDC-AHG T-cell with the Kaplan-Meier method, and comparisons be-
and CDC (three wash) B-cell cross match for all donor- tween survival curves were made by the log-rank test.
recipient combinations, and a flow cytometry T- and Statistical significance was defined as a p value of less
B-cell cross match for all recipients with HLA-directed than 0.05. All data analysis was performed by SAS Sys-
antibodies detected by flow cytometry. tem for Windows, version 8, and SAS Enterprise Guide
PRAs before transplantation were obtained on all (SAS Institute, Cary, NC).
patients. CDC-AHG, enzyme-linked immunosorbent as-
say, and flow cytometry techniques were used, and peak
historic PRAs were recorded for the purpose of this RESULTS
study. If flow cytometry PRAs were available (they were Between January 1999 and August 2003, a total of
routinely used starting in January 2000), they were 519 patients underwent a kidney or combined
reported. If flow cytometry PRAs were not available, kidney-pancreas transplantation at our institution. Mean
enzyme-linked immunosorbent assay or CDC-AHG follow-up was 884 23 days. Seventy-five patients had
PRAs were reported. Flow cytometry PRAs were avail- at least one episode of ACR. On the basis of light
able in the majority of patients. For simplicity of pre- microscopic findings, 29 patients had AHR. The re-
sentation, the results are presented as T- and B-cell cently developed consensus criteria for the diagnosis of
PRAs, regardless of the method used. AHR were then used to confirm AHR [22]. This en-
Patient Characteristics, Treatment, and Outcome tailed C4d staining of frozen sections of transplant biopsy
Charts were screened for demographic patient character- samples and screening for donor-specific HLA antibod-
istics, namely sex, race, age, and type of transplant (liv- ies. C4d staining was performed on all 29 biopsy sam-
ing donor or deceased donor). Furthermore, the following ples, and donor-specific HLA antibody screening was
performed on 23 (79%) of 29 patients. In 13 of 29
transplant-specific data were extracted: cold ischemia
patients, donor-specific HLA antibody screening was
time, presence of delayed graft function (defined as the
performed at the time of rejection. In 10 of 29 patients,
need for renal replacement therapy in the first week after
donor-specific HLA antibody screening was performed at
transplantation), type of induction therapy (daclizumab the time of transplantation. We defined AHR as pres-
or antithymocyte globulin), historic peak T- and B-cell ence of typical findings on light microscopy and presence
PRA, and modality of treatment (PP and IVIG, PP of either C4d staining or presence of donor-specific an-
alone, IVIG alone, pulse methylprednisolone, antithy- tibodies. With this stringent approach, we were able to
mocyte globulin, and muromonab-CD3). confirm AHR in 23 patients and excluded the remaining
Maintenance immunosuppression in all patients con- 6 patients from the analysis (Figure 1).
sisted of a calcineurin inhibitor (tacrolimus or cyclospor- The baseline characteristics of patients who developed
ine), mycophenolate, and prednisone. Patients who were ACR, AHR, or No REJ are summarized in Table 1.
identified as having AHR received a combined regimen Although most demographic values did not differ be-
of PP and IVIG (Gamimune, Bayer Biological Products, tween the groups, patients who experienced AHR were
Research Triangle Park, NC; or Venoglobulin, Alpha significantly more likely to be black and female (AHR
Therapeutic, Los Angeles, CA). A typical PP regimen vs. No REJ: p 0.0161 and p 0.0003, respectively).
consisted of four daily sessions (range 3– 6 days) with 5% Age and donor source were similar among groups.
human albumin replacement. The number of PP sessions Table 2 lists the clinical characteristics of our cohort.
was based on the clinical response to therapy as measured Cold ischemia time was similar in all three groups.
by urine output and serum creatinine. IVIG, usually at a Patients who developed AHR were significantly more

IVIG and Plasmapheresis in Humoral Rejection 353
Patients in the AHR group had a significantly higher
mean T- and B-cell PRA compared with patients in the
ACR or No REJ group (mean B-cell PRA: AHR vs. No
REJ and ACR: p 0.0001 and p 0.0001, respectively;
mean T-cell PRA: AHR vs. No REJ and ACR: p
0.0001 and p 0.0002, respectively). When historic peak
PRAs were categorized as negative ( 10%), moderate
( 10%– 50%), or high ( 50%), we observed a bimodal
distribution of peak PRAs in the AHR group. Negative B-
and T-cell PRAs were found in 47.8% and 47.9% of
patients with AHR, respectively. High B- and T-cell
PRAs were identified in 43.5% and 47.9%, respectively.
The majority of patients in the ACR and No REJ group
were found to have a negative PRA (Figure 2).
The treatment of AHR consisted of PP and IVIG in
FIGURE 1 Retrospective analysis of all consecutive kidney almost all patients (22 of 23). One patient received PP
and kidney-pancreas transplants performed at Duke University alone. Eleven patients additionally received pulse meth-
Medical Center between January 1999 and August 2003. ylprednisolone therapy, and seven patients received ei-
Follow-up was extended until August 2004. ther Thymoglobulin or OKT3 (Table 2). Most (20 of 23)
patients responded to therapy with improved renal func-
likely to have had delayed graft function (AHR vs. No tion. Of the nonresponders, one patient required hemo-
REJ: p 0.0005). Use of induction therapy was similar dialysis on postoperative day 1 and underwent transplant
in all three groups. The time to rejection was defined as biopsy on day 6, the findings of which revealed AHR.
the interval between renal transplantation and the diag- The patient was treated with methylprednisolone at that
nostic biopsy. Not surprisingly, AHR was diagnosed point. PP and IVIG were initiated after a second biopsy
earlier than ACR (median at day 6 vs. day 70, p was performed on day 12, which revealed persistent
0.0071). However, two patients were found to have AHR. AHR could not be reversed, and the patient
AHR late in their transplant course, at days 147 and continued to need renal replacement therapy. Transplant
843. Precipitating factors were unclear in the first pa- nephrectomy was performed 7 months after the trans-
tient but medication noncompliance was found to be the plant. The second patient was diagnosed on postopera-
cause in the second patient. All other patients in the tive day 2 with AHR and was treated with PP and IVIG
AHR group (21 of 23) experienced rejection between starting on day 3. The patient was discharged requiring
days 3 and 14. hemodialysis and died on postoperative day 30. The
TABLE 1 Demographic characteristics of patients who underwent renal
transplantation with and without acute rejection
All AHR ACR No REJ
Characteristic (n 513) (n 23) (n 75) (n 415)
Age (years),
mean SD 46 0.6 45 2.6 42 1.5 47 0.6
Sex, n (%)
Male 302 (59%) 5 (22%) 45 (60%) 252 (61%)
Female 211 (41%) 18 (78%)a 30 (40%) 163 (39%)
Race, n (%)
White 292 (57%) 8 (35%) 31 (41%) 253 (61%)
Black 214 (42%) 15 (65%)b 44 (59%) 155 (37%)
Other 7 (1%) 0 0 7 (2%)
Type of transplant,
n (%)
Living donor 197 (38%) 8 (35%) 24 (32%) 165 (40%)
Cadaveric
transplant 316 (62%) 15 (65%) 51 (68%) 250 (60%)
a
p 0.0003 vs. No REJ.
b
p 0.0161 vs. No REJ.

354 R.W. Lehrich et al.
TABLE 2 Clinical characteristics of patients who underwent renal
transplantation
Characteristic AHR (n 23) ACR (n 75) No REJ (n 415)
Transplant characteristics
Cold ischemia time,
mean SD 11.7 2.5 20.2 1.1 18.8 0.7
Delayed graft function,
n (%) 13 (56%)a 19 (25%) 90 (22%)
Induction therapy, n (%) 17 (74%) 43 (57%) 241 (58%)
Peak B-cell PRA,
mean SD 39%bc 8.7% 7% 2.5% 6% 1.0%
Peak T-cell PRA 43%de 8.9% 9% 2.9% 8% 1.1%
Time to rejection,
median (IQR) 6f (5–8) 70 (7–356) NA
Therapy for rejection, n (%)
PP IVIG 22 (96%) 0 NA
PP alone 1 (4%) 0 NA
IVIG alone 0 0 NA
Pulse methylprednisolone 13 (57%) 37 (49%) NA
Thymoglobulin or OKT3 7 (30%) 38 (51%)g NA
a
p 0.0005 AHR vs. No REJ.
b
p 0.0001 ACR vs. No REJ.
c
p 0.0001 AHR vs. ACR.
d
p 0.0001 AHR vs. No REJ.
e
p 0.0001 AHR vs. ACR.
f
p 0.0071 AHR vs. ACR.
g
In combination with pulse methylprednisolone in patients with ACR.
Cold ischemia time (hours); time to rejection (days).
third patient was found to have AHR on postoperative DISCUSSION
day 7; therapy with PP and IVIG was immediately
initiated. Subsequently, this patient developed systemic We report the results of a single-center retrospective
inflammatory response with acute respiratory distress analysis of incidence and outcome of AHR in renal
syndrome, which was thought to be related to the epi- transplantation. The central findings of this study are as
sode of acute rejection. A transplant nephrectomy was follows: (1) AHR occurs with an incidence of 4.4%,
performed on postoperative day 13. affects predominantly highly sensitized patients, and is
In the No REJ group, cumulative 2-year graft survival observed early in the transplant course; (2) the combina-
was 94%, which was significantly higher than in both tion of IVIG and PP is an effective strategy for the
rejection groups (ACR vs. No REJ: p 0.0001; AHR treatment of AHR; and (3) 2-year graft survival of AHR
vs. No REJ: p 0.0002). Patients with ACR and AHR
with this regimen is better than in historic controls and
had 2-year graft survival of 85% and 78%, respectively
comparable to graft survival in ACR.
(Figure 3). There was no significant difference in 2-year
graft survival between rejection groups (ACR vs. AHR: When AHR was defined as allograft dysfunction with
p 0.50). Regarding patient survival, there was a sig- typical light microscopic findings, as well as the presence
nificant difference between patients in the ACR group of positive C4d staining or of donor-specific antibodies,
and No REJ (2-year patient survival: ACR 95% vs. No the incidence of AHR in our study is comparable to
REJ 98%, p 0.013). There were two deaths in the previous assessments of incidence and falls well into the
AHR group (2-year patient survival: AHR 95%), but described range of 3%–10% [7, 24, 25]. The use of
mortality difference between the AHR group and No evaluating allograft dysfunction for AHR with a com-
REJ did not reach statistical significance (AHR vs. No bined approach consisting of light microscopic evalua-
REJ: p 0.09) (Figure 4).. Last follow-up mean serum tion, immunofluorescence staining for C4d, and screen-
creatinine of patients with functioning allografts for the ing for donor-specific antibodies is now well established
AHR, ACR, and No REJ groups were 1.8 mg/dl, 1.5 and has been validated in several retrospective cohort
mg/dl, and 1.6 mg/dl, respectively (Table 3). studies [7, 22, 24, 25]. We think that this approach

IVIG and Plasmapheresis in Humoral Rejection 355
FIGURE 3 Kaplan-Meier allograft survival curves with
groups as follows: No REJ group (solid line), ACR group
(dashed line), and AHR group (dotted line). Numbers above
x-axis at months 0, 6, 12, 18, 24, 30, and 36 represent number
of patients after censoring event. Cumulative graft survival was
significantly better in the No REJ group compared with ACR
and AHR (ACR vs. No REJ: p 0.0001; AHR vs. No REJ: p
0.0002). Graft survival between the AHR and ACR groups
was not significantly different (p 0.50).
that AHR occurs early in the transplant course, with a
median onset after transplantation measuring days rather
than weeks [7]. The median onset of AHR in our study
was 6 days, with 50% of patients developing AHR
between days 5 and 8 after transplantation. However, we
were able to identify one patient who developed AHR
FIGURE 2 PRA frequencies according to PRA intensity.
Solid bars B-cell PRAs; open bars T-cell PRAs. PRAs
10% were considered negative; PRAs 10%–50% were con-
sidered moderately elevated; and PRAs 50% were consid-
ered high. There were significantly more patients with high
PRAs in the AHR group compared with the ACR or the No
REJ group. *B-cell PRA: AHR vs. ACR and No REJ: p
0.0001 and p 0.0001, respectively. **T-cell PRA: AHR vs.
No REJ and ACR: p 0.0001 and p 0.0001, respectively.
FIGURE 4 Kaplan-Meier patient survival curves with
groups as follows: No REJ group (solid line), ACR group
allowed us to reliably identify all patients with AHR to (dashed line), and AHR group (dotted line). Numbers above
retrospectively study the effectiveness of therapy with x-axis at months 0, 6, 12, 18, 24, 30, and 36 represent number
IVIG and PP in AHR. of patients after censoring event. Cumulative patient survival
was significantly better in the No REJ group compared with
Patients in the AHR group had clinical features sim- ACR (ACR vs. No REJ: p 0.013). AHR patient survival was
ilar to those previously described in patients with acute not statistically different when compared with the ACR or the
alloantibody-mediated rejection. It is well established No REJ group.

356 R.W. Lehrich et al.
TABLE 3 Outcome characteristics
Characteristic AHR (n 23) ACR (n 75) No REJ (n 415)
Follow-up (days) 764 109 944 58 880 26
2-year graft survival 78%a 85%b 94%
2-year patient survival 95% 95%c 98%
Last creatinine, median (IQR) 1.8 (1.4–2.6) 1.5 (1.2–1.9) 1.6 (1.3–1.8)
a
p 0.0002 AHR vs. No REJ.
b
p 0.0001 ACR vs. No REJ.
c
p 0.013 ACR vs. No REJ.
2-year graft survival (%); 2-year patient survival (%); last creatinine, median (IQR) (mg/dl).
precipitated by medication noncompliance more than 2 models of inflammatory activation of endothelial cells,
years after transplantation. It is conceivable that sup- IVIG has been demonstrated to inhibit tumor necrosis
pressed memory B cells were reactivated when immuno- factor – and interleukin 1 –induced gene transcription
suppression was suboptimal, leading to late acute of adhesion molecules and cytokines [16, 29]. IVIG
alloantibody-mediated rejection. Highly sensitized pa- likely behaves as normal IgG and IgM regarding the
tients are more likely to develop AHR [3, 7, 26]. control of autoreactivity of antibodies in human plasma
Women in our cohort were significantly more likely to [30]. Normal IgG and IgM function includes the sup-
develop AHR. This may relate to the higher rate of pression of migration of B-cell populations from the bone
sensitization observed in women as a result of previous marrow to secondary lymphoid organs, as found in mice
pregnancies. Last, elevated PRAs are markers of sensiti- [31]. Furthermore, IVIG has been demonstrated to
zation. It is therefore not surprising that patients in the downregulate specific autoreactive B-cell populations in
AHR group had significantly elevated historic peak animal models of inherited immunodeficiency [32]. It
PRAs. However, PRA distribution was bimodal, and can therefore be speculated that IVIG might have prop-
roughly half of patients in the AHR group had a negative erties that are helpful in decreasing donor-specific anti-
PRA. Thus, a detectable PRA does not identify all body load and reducing harmful B-cell populations in
patients at risk for AHR. This emphasizes the notion patients with AHR.
that donor-specific antibodies other than HLA antibod- In our study, treatment of AHR with PP and IVIG is
ies, and nonclassical HLA antibodies that are not de- associated with 2-year graft survival of 78%. Therapy
tected by the PRA method might play a role in AHR. In with PP alone is associated with inferior results [24, 33],
cardiac transplantation, antiendothelial antibodies have likely because of early rebound of alloantibodies. Historic
been demonstrated to be associated with acute humoral controls indicate that the graft loss without specific
but not ACR [27]. In renal transplantation, antibodies to therapy is 15%–50% [8, 9]. In our experience IVIG and
MHC class I–related A antigen were found to be corre- PP are helpful modalities to treat AHR, but our study
lated with rejection and early graft loss [28]. The pres- and similar studies by other groups must be interpreted
ence of activating antibodies to angiotensin II type 1 with caution. All conducted studies are retrospective.
receptors was associated with steroid resistant rejec- Despite our encouraging results, graft loss in AHR re-
tion in a cohort of kidney transplant recipients who mains higher than in transplant recipients without re-
also had malignant hypertension [29]. jection. New strategies involving alternative treatment
IVIG in combination with PP has been used by us and modalities are being investigated. Thymoglobulin in
others to treat AHR [10 –13]. The commercial prepara- combination with PP was recently used to treat renal
tions of IVIG used in clinical practice contain intact IgG transplant patients with AHR. In a small, uncontrolled
molecules with a distribution of subclasses closely resem- study, no difference in graft survival between the AHR
bling that in the human serum. IVIG represents pooled group and the no rejection group was observed, making
plasma from approximately 3000 –10,000 healthy do- this a promising treatment modality [34]. Rituximab, a
nors [14]. A body of experimental and clinical evidence genetically engineered chimeric human-murine anti–
suggests various potential actions of IVIG that might CD-20 monoclonal antibody, has been used to treat
explain its usefulness in treating AHR. In patients with AHR. This approach appears reasonable because CD-20
autoimmune hemophilia, IVIG has been demonstrated is involved in the regulation of B-cell development and
to neutralize autoantibodies [15]. This is likely because differentiation. In two case reports (one heart transplant
of a high concentration of antiidiotypic antibodies in recipient and one lung transplant recipient), rituximab
IVIG directed against autoantibodies. In experimental was a helpful adjunct in the treatment of AHR [35, 36].

IVIG and Plasmapheresis in Humoral Rejection 357
Patients who are cross-match positive before transplan- demonstrating poor kidney graft survival when acute
tation are at high risk of developing AHR. IVIG alone, rejections are associated with IgG donor-specific
the combination of IVIG and PP, and the combination of lymphocytotoxin. Transplantation 59:357, 1995.
IVIG, PP, and rituximab have been used recently as part 10. Rocha PN, Butterly DW, Greenberg A, Reddan DN,
of desensitization protocols in this patient population Tuttle-Newhall J, Collins BH, Kuo PC, Reinsmoen N,
[37– 41]. These data support additional roles for IVIG Fields T, Howell DN, Smith SR: Beneficial effect of
and PP, namely the prevention of AHR in high-risk plasmapheresis and intravenous immunoglobulin on renal
patients and overcoming contraindications for renal allograft survival of patients with acute humoral rejection.
transplantation. Transplantation 75:1490, 2003.
In summary, we have demonstrated that the combi- 11. White NB, Greenstein SM, Cantafio AW, Schechner R,
nation of IVIG and PP in addition to standard immu- Glicklich D, McDonough P, Pullman J, Mohandas K,
nosuppression containing prednisone, mycophenolate, Boctor F, Uehlinger J, Tellis V: Successful rescue therapy
and calcineurin inhibition effectively salvages renal func- with plasmapheresis and intravenous immunoglobulin for
acute humoral renal transplant rejection. Transplantation
tion in AHR. However, a higher rate of long-term graft
78:772, 2004.
loss warrants more investigation into preventive and
therapeutic measures. 12. Grandtnerova B, Javorsky P, Kolacny J, Hovoricova B,
Dedic P, Laca L: Treatment of acute humoral rejection in
kidney transplantation with plasmapheresis. Transplant
ACKNOWLEDGMENT Proc 27:934, 1995.
R.W.L. is funded by a grant from the James R. Clapp Fellow- 13. Bohmig GA, Regele H, Exner M, Derhartunian V, Kletz-
ship in Nephrology. mayr J, Saemann MD, Horl WH, Druml W, Watschinger
B: C4d-positive acute humoral renal allograft rejection:
effective treatment by immunoadsorption. J Am Soc
REFERENCES Nephrol 12:2482, 2001.
1. Platt JL: Acute vascular rejection. Transplant Proc 32: 14. Kazatchkine MD, Kaveri SV: Immunomodulation of au-
839, 2000. toimmune and inflammatory diseases with intravenous
2. Jeannet M, Pinn VW, Flax MH, Winn HJ, Russell PS: immune globulin. N Engl J Med 345:747, 2001.
Humoral antibodies in renal allotransplantation in man. 15. Sultan Y, Kazatchkine MD, Maisonneuve P, Nydegger
N Engl J Med 282:111, 1970. UE: Anti-idiotypic suppression of autoantibodies to factor
3. Halloran PF, Wadgymar A, Ritchie S, Falk J, Solez K, VIII (antihaemophilic factor) by high-dose intravenous
Srinivasa NS: The significance of the anti– class I antibody gammaglobulin. Lancet 324:755, 1984.
response. I. Clinical and pathologic features of anti– class 16. Xu C, Poirier B, Van Huyen JP, Lucchiari N, Michel O,
I–mediated rejection. Transplantation 49:85, 1990. Chevalier J, Kaveri S: Modulation of endothelial cell func-
4. Saadi S, Holzknecht RA, Patte CP, Stern DM, Platt JL: tion by normal polyspecific human intravenous immuno-
Complement-mediated regulation of tissue factor activity globulins: a possible mechanism of action in vascular
in endothelium. J Exp Med 182:1807, 1995. diseases. Am J Pathol 153:1257, 1998.
5. Platt JL, Vercellotti GM, Dalmasso AP, Matas AJ, 17. Stohl W: Cellular mechanisms in the in vitro inhibition of
Bolman RM, Najarian JS, Bach FH: Transplantation of pokeweed mitogen-induced B cell differentiation by im-
discordant xenografts: a review of progress. Immunol To- munoglobulin for intravenous use. J Immunol 136:4407,
day 11:450, 1990. 1986.
6. Mauiyyedi S, Colvin RB: Humoral rejection in kidney 18. Toyoda M, Zhang X, Petrosian A, Galera OA, Wang SJ,
transplantation: new concepts in diagnosis and treatment. Jordan SC: Modulation of immunoglobulin production
Curr Opin Nephrol Hypertens 11:609, 2002. and cytokine mRNA expression in peripheral blood
7. Crespo M, Pascual M, Tolkoff-Rubin N, Mauiyyedi S, mononuclear cells by intravenous immunoglobulin. J Clin
Collins AB, Fitzpatrick D, Farrell ML, Williams WW, Immunol 14:178, 1994.
Delmonico FL, Cosimi AB, Colvin RB, Saidman SL: 19. van Schaik IN, Lundkvist I, Vermeulen M, Brand A:
Acute humoral rejection in renal allograft recipients: I. Polyvalent immunoglobulin for intravenous use interferes
Incidence, serology and clinical characteristics. Transplan- with cell proliferation in vitro. J Clin Immunol 12:325,
tation 71:652, 2001. 1992.
8. Trpkov K, Campbell P, Pazderka F, Cockfield S, Solez K, 20. Toyoda M, Pao A, Petrosian A, Jordan SC: Pooled human
Halloran PF: Pathologic features of acute renal allograft gammaglobulin modulates surface molecule expression
rejection associated with donor-specific antibody: analysis and induces apoptosis in human B cells. Am J Transplant
using the Banff grading schema. Transplantation 3:156, 2003.
61:1586, 1996. 21. Racusen LC, Solez K, Colvin RB, Bonsib SM, Castro MC,
9. Lobo PI, Spencer CE, Stevenson WC, Pruett TL: Evidence Cavallo T, Croker BP, Demetris AJ, Drachenberg CB,

358 R.W. Lehrich et al.
Fogo AB, Furness P, Gaber LW, Gibson IW, Glotz D, 31. Sundblad A, Marcos MA, Malanchere E, Castro A, Haury
Goldberg JC, Grande J, Halloran PF, Hansen HE, M, Huetz F, Nobrega A, Freitas A, Coutinho A: Obser-
Hartley B, Hayry PJ, Hill CM, Hoffman EO, Hunsicker vations on the mode of action of normal immunoglobulin
LG, Lindblad AS, Marcussen N, Mihatsch MJ, Nadasdy at high doses. Immunol Rev 139:125, 1994.
T, Nickerson P, Olsen TS, Papadimitriou JC, Randhawa 32. Vassilev T, Yamamoto M, Aissaoui A, Bonnin E, Berrih-
PS, Rayner DC, Roberts I, Rose S, Rush D, Aknin S, Kazatchkine MD, Kaveri SV: Normal human
Salinas-Madrigal L, Salomon DR, Sund S, Taskinen E, immunoglobulin suppresses experimental myasthenia
Trpkov K, Yamaguchi Y: The Banff 97 working gravis in SCID mice. Eur J Immunol 29:2436, 1999.
classification of renal allograft pathology. Kidney Int
55:713, 1999. 33. Adams MB, Kauffman HM Jr, Hebert LA, Hussey CV,
Duquesnoy RJ, Tomasulo PA: Plasmapheresis in the
22. Racusen LC, Colvin RB, Solez K, Mihatsch MJ, Halloran treatment of renal allograft rejection. Proc Clin Dial
PF, Campbell PM, Cecka MJ, Cosyns JP, Demetris AJ, Transplant Forum 9:252, 1979.
Fishbein MC, Fogo A, Furness P, Gibson IW, Glotz D,
Hayry P, Hunsickern L, Kashgarian M, Kerman R, Magil 34. Akalin E, Ames S, Sehgal V, Fotino M, Daly L, Murphy
AJ, Montgomery R, Morozumi K, Nickeleit V, B, Bromberg JS: Intravenous immunoglobulin and Thy-
Randhawa P, Regele H, Seron D, Seshan S, Sund S, moglobulin facilitate kidney transplantation in comple-
Trpkov K: Antibody-mediated rejection criteria—an ad- ment-dependent cytotoxicity B-cell and flow cytometry
dition to the Banff 97 classification of renal allograft T- or B-cell crossmatch-positive patients. Transplantation
rejection. Am J Transplant 3:708, 2003. 76:1444, 2003.
23. Pei R, Wang G, Tarsitani C, Rojo S, Chen T, Takemura 35. Aranda JM Jr, Scornik JC, Normann SJ, Lottenberg R,
S, Liu A, Lee J: Simultaneous HLA class I and class II Schofield RS, Pauly DF, Miles M, Hill JA, Sleasman JW,
antibodies screening with flow cytometry. Hum Immunol Skoda-Smith S: Anti-CD20 monoclonal antibody (ritux-
59:313, 1998. imab) therapy for acute cardiac humoral rejection: a case
24. Mauiyyedi S, Crespo M, Collins AB, Schneeberger EE, report. Transplantation 73:907, 2002.
Pascual MA, Saidman SL, Tolkoff-Rubin NE, Williams 36. Garrett HE Jr, Groshart K, Duvall-Seaman D, Combs D,
WW, Delmonico FL, Cosimi AB, Colvin RB: Acute hu- Suggs R: Treatment of humoral rejection with rituximab.
moral rejection in kidney transplantation: II. Morphology, Ann Thorac Surg 74:1240, 2002.
immunopathology, and pathologic classification. J Am 37. Jordan SC, Vo A, Bunnapradist S, Toyoda M, Peng A,
Soc Nephrol 13:779, 2002. Puliyanda D, Kamil E, Tyan D: Intravenous immune
25. Watschinger B, Pascual M: Capillary C4d deposition as a globulin treatment inhibits crossmatch positivity and al-
marker of humoral immunity in renal allograft rejection. lows for successful transplantation of incompatible organs
J Am Soc Nephrol 13:2420, 2002. in living-donor and cadaver recipients. Transplantation
26. Scornik JC, Salomon DR, Lim PB, Howard RJ, Pfaff 76:631, 2003.
WW: Posttransplant antidonor antibodies and graft re- 38. Montgomery RA, Zachary AA, Racusen LC, Leffell MS,
jection: evaluation by two-color flow cytometry. Trans- King KE, Burdick J, Maley WR, Ratner LE: Plasma-
plantation 47:287, 1989. pheresis and intravenous immune globulin provides effec-
27. Fredrich R, Toyoda M, Czer LS, Galfayan K, Galera O, tive rescue therapy for refractory humoral rejection and
Trento A, Freimark D, Young S, Jordan SC: The clinical allows kidneys to be successfully transplanted into cross-
significance of antibodies to human vascular endothelial match–positive recipients. Transplantation 70:887, 2000.
cells after cardiac transplantation. Transplantation 67: 39. Sonnenday CJ, Ratner LE, Zachary AA, Burdick JF, Sa-
385, 1999. maniego MD, Kraus E, Warren DS, Montgomery RA:
28. Sumitran-Holgersson S, Wilczek HE, Holgersson J, Sod- Preemptive therapy with plasmapheresis/intravenous im-
erstrom K: Identification of the nonclassical HLA mole- munoglobulin allows successful live donor renal trans-
cules, mica, as targets for humoral immunity associated plantation in patients with a positive cross-match. Trans-
with irreversible rejection of kidney allografts. Transplan- plant Proc 34:1614, 2002.
tation 74:268, 2002. 40. Gloor JM, DeGoey SR, Pineda AA, Moore SB, Prieto M,
29. Dragun D, Müller DN, Bräsen JH, Fritsche L, Nieminen- Nyberg SL, Larson TS, Griffin MD, Textor SC, Velosa JA,
Kelhä M, Dechend R, Kintscher U, Rudolph B, Hoebeke Schwab TR, Fix LA, Stegall MD: Overcoming a positive
J, Eckert D, Mazak I, Plehm R, Schönemann C, Unger T, crossmatch in living-donor kidney transplantation. Am J
Budde K, Neumayer HH, Luft FC, Wallukat G: Angio- Transplant 3:1017, 2003.
tensin II Type 1–Receptor Activating Antibodies in Re- 41. Gloor JM, DeGoey S, Ploeger N, Gebel H, Bray R, Moore
nal-Allograft Rejection. N Engl J Med 352:558, 2005. SB, Dean PG, Stegall MD: Persistence of low levels of
30. Hurez V, Kaveri SV, Kazatchkine MD: Expression and alloantibody after desensitization in crossmatch-positive
control of the natural autoreactive IgG repertoire in nor- living-donor kidney transplantation. Transplantation 78:
mal human serum. Eur J Immunol 23:783, 1993. 221, 2004.