iga nephropathy with crescents in kidney transplant recipients
TRANSCRIPT
T
●
mrlbgCrIwcghwai©
I
Iwmssotvl
ntitrlektwolsadch
A
ransplantation
IgA Nephropathy With Crescents in Kidney Transplant Recipients
Jolanta Kowalewska, MD, Shan Yuan, MD, Neodjema Sustento-Reodica, MD,Roberto F. Nicosia, MD, PhD, Kelly D. Smith, MD, PhD, Connie L. Davis, MD,
and Charles E. Alpers, MD
Background: Crescentic glomerulonephritis is an uncommon finding in renal allografts. Recurrence or de novoesangial deposition of immunoglobulin A (IgA) in renal allografts most often is clinically benign, but some case
eports have shown that IgA nephropathy in renal allografts can present as crescentic glomerulonephritis and mayead to rapid deterioration of graft function and/or graft loss. Methods: We reviewed diagnoses of all allograftiopsies at University of Washington Medical Center (Seattle, WA) from 1989 to 2003 and found 33 cases oflomerulonephritis with crescents. Eight of these cases were the result of recurrent or de novo IgA nephropathy.linical and pathological features of these patients were reviewed. Results: Six of 8 cases with crescents were the
esult of recurrent IgA nephropathy, and 2 cases were presumptive de novo IgA nephropathy. Of the 8 patients withgA nephropathy with crescents, 6 patients presented clinically with increasing serum creatinine levels; 4 patients,ith proteinuria; and 4 patients, with hematuria. In 6 patients, there was 10% to 30% involvement of glomeruli, withrescents partially or completely filling urinary spaces. The other patients showed lesser (�7% of sampledlomeruli) involvement. Four patients with IgA nephropathy with crescents developed renal failure and returned toemodialysis therapy. Three patients had a benign clinical course, with stabilization of renal function. One patientas lost to follow-up. Conclusion: We identified a cohort of patients with glomerulonephritis with crescents in renalllografts with IgA nephropathy as the cause. In half the affected patients, this led to early progressive renalnsufficiency and return to hemodialysis therapy. Am J Kidney Dis 45:167–175.
2004 by the National Kidney Foundation, Inc.
NDEX WORDS: Immunoglobulin A (IgA) nephropathy; crescents; kidney allograft; renal failure.
lanclnosgeragtts
Mt
b
UP9
MMUNOGLOBULIN A (IgA) nephropathyis the most common form of glomerulopathy
orldwide.1-3 It has been estimated that approxi-ately 15% to 40% of patients with long-
tanding IgA nephropathy will progress to end-tage renal disease (ESRD).4,5 Although a varietyf therapeutic interventions show promise inreating patients with IgA nephropathy and pre-enting ESRD, there is no reliable cure or estab-ished therapy for this disease.
For patients who progress to ESRD, the alter-ative to long-term dialysis therapy is kidneyransplantation. In 1975, only 7 years after theirnitial description of the entity of IgA nephropa-hy, Berger et al6 reported the first case of recur-ent IgA in a kidney allograft, which was fol-owed by several similar reports.7-9 It has beenstimated that approximately 20% to 60% ofidney transplant recipients with IgA nephropa-hy as the cause of their original kidney diseaseill develop recurrent disease.7-13 The majorityf patients with recurrent IgA nephropathy fol-ow a benign course, characterized by micro-copic hematuria and intermittent proteinuria. In
small subset of patients (up to 20%), theisease is more aggressive, characterized clini-ally by progressive worsening of renal function,
ematuria, proteinuria, and subsequent graftmerican Journal of Kidney Diseases, Vol 45, No 1 (January), 200
oss.8,14,15 Occasionally, these patients are char-cterized by mesangial proliferative glomerulo-ephritis with additional features of glomerularellular crescent formation (crescentic glomeru-onephritis). Crescentic glomerulonephritis in re-al allografts has been reported previously, butverall, it is considered rare and most oftenecondary to systemic disease (vasculitis, anti–lomerular basement membrane antibody dis-ase, or systemic lupus).16 The de novo occur-ence of crescentic glomerulonephritis in renalllografts also has been reported, including anti–lomerular basement membrane disease in pa-ients with Alport’s syndrome.17 Our review ofhe English-language literature identified onlyporadic cases of IgA nephropathy with crescen-
From the Department of Pathology; and Department ofedicine, Division of Nephrology, University of Washing-
on, Seattle, WA.Received April 23, 2004; accepted in revised form Septem-
er 28, 2004.Address reprint requests to Jolanta Kowalewska, MD,
niversity of Washington Medical Center, Department ofathology, Box 357470, 1959 NE Pacific St, Seattle, WA8195. E-mail: [email protected]© 2004 by the National Kidney Foundation, Inc.0272-6386/04/4501-0019$30.00/0
doi:10.1053/j.ajkd.2004.09.0305: pp 167–175 167
tgcpa
rro2wegittht
C
tc2akaccttutdrmkwrrcIuw
otlalr4iml1c
ar
L
vicssigsmchmcecgtefttpcfict6mnnnpueadp
I
scic
KOWALEWSKA ET AL168
ic glomerulonephritis occurring in renal allo-rafts.18-24 We report here clinical and pathologi-al features in the largest cohort to date ofatients with IgA nephropathy with crescentsrising in their allograft kidneys.
METHODS
After obtaining institutional review board approval foreview of patient records (03-8118-E01), we studied allenal allograft biopsy specimens received at the Universityf Washington Medical Center (Seattle, WA) from 1989 to003 (n � 2,959). We found 33 cases of glomerulonephritisith crescents (1.1%), distinct from the epithelial cell prolif-
ration of collapsing glomerulopathy or focal and segmentallomerulosclerosis. Cellular crescents were defined accord-ng to the widely accepted World Health Organization defini-ion as 2 or more layers of proliferating cells and inflamma-ory cells within Bowman’s space.25 Eight of these patientsad an underlying diagnosis of IgA nephropathy (24%), andhese form the substance of this report.
linical DataClinical data are listed in Table 1. Mean patient age at the
ime of the index biopsy diagnosis of IgA nephropathy withrescents was 42.6 years (range, 22 to 59 years). There werewomen and 6 men. Six patients had recurrent IgA nephrop-
thy. The presumptive causes of renal failure of the nativeidneys in the 2 remaining patients were reflux nephropathynd postinfectious glomerulonephritis, both diagnosed clini-ally without biopsy confirmation or biopsy exclusion ofoncurrent IgA nephropathy. Four patients received kidneyransplants from living-related donors (LRDs), and 3 pa-ients, from deceased donors (cadaveric). For 1 patient whonderwent transplantation at another institution, this informa-ion was unavailable. To rule out the possibility of donor-erived deposition of IgA in the index biopsy specimens, weeviewed donor or early posttransplantation biopsy speci-ens of our patients. We had available only 2 early donor
idney biopsy specimens for patients 7 and 8. These biopsiesere performed at 1 and 3 months after transplantation,
espectively. They did not show histological or immunofluo-escence features of IgA nephropathy. Therefore, in theseases, we can exclude the possibility of preexisting donorgA nephropathy. For the remaining 6 patients, we arenable to confirm that finding based on earlier biopsies,hich were either not performed or not available.Mean time from kidney transplantation to the index bi-
psy was 8.2 years (range, 1.75 to 20 years). Indications forhe allograft renal biopsy were increasing serum creatinineevels in 6 patients (75%), proteinuria in 4 patients (50%),nd hematuria in 4 patients (50%). Mean serum creatinineevel at the time of the biopsy was 3.4 mg/dL (300 �mol/L;ange, 1.9 to 6.5 mg/dL [168 to 575 �mol/L]). Proteinuria inpatients ranged from 1 to 3� by dipstick analysis and was
n the nephrotic range in the 2 patients for whom it waseasured during a 24-hour period. Four patients had border-
ine hypertension (diastolic pressure � 90 mm Hg), but onlypatient (no. 5) had overt hypertension. The presence of
irculating serum antineutrophil cytoplasmic antibodies and S
ntinuclear antibodies was examined in 1 patient (no. 8), andesults were negative.
RESULTS
ight Microscopy
Biopsy specimens for 7 of 8 patients showedarying degrees of focal global glomerulosclerosisnvolving 4.8% to 40% of glomeruli in individualases. Glomeruli without global glomerulosclero-is showed varying degrees of mesangial expan-ion caused by both increased matrix and cellular-ty. In 3 patients, approximately 3.4% to 20% oflomeruli showed additional features of focal andegmental glomerulosclerosis characterized by seg-ental expansion of the matrix, obliteration of the
apillary lumina by matrix, and accumulations ofyaline. All patients showed accumulations ofononuclear cells in the urinary space forming
ellular to fibrocellular crescents (Fig 1Ato D). Thextent of glomerular involvement by these cres-ents ranged from 6.9% to 30% of nonscleroticlomeruli. The extent of interstitial fibrosis andubular atrophy was minimal to moderate and gen-rally was proportional to the degree of global andocal segmental glomerulosclerosis. In all patients,here was mononuclear inflammatory cell infiltra-ion of the interstitium, which, in the majority ofatients, was localized predominantly to areas ofhronic fibrosing tubulointerstitial injury. These in-ltrates generally lacked features characteristic ofellular rejection in better preserved tubulointersti-ial parenchyma. However, in 2 patients (nos. 5 and), interstitial inflammation was associated withild interstitial edema and interpreted as interstitial
ephritis, probably secondary to severe active IgAephropathy. Although tubulitis was not promi-ent, the possibility of an ongoing active rejectionrocess in these patients could not be excludednequivocally.Arterial vessels showed mild to mod-rate degrees of intimal fibrosis, but no evidence ofctive vascular rejection was identified in any in-ex biopsy specimen, including the patient withrevious biopsy-proven vascular rejection.
mmunofluorescence
In all cases, a portion of the biopsy tissue wastudied by means of immunofluorescence micros-opy. The number of glomeruli available formmunofluorescence examination in individualases ranged from 2 to 11 per level section.
ections were stained for IgG, IgA, IgM, � and �
Table 1. Clinical Features of Kidney Transplant Recipients With IgA Nephropathy With Crescents
Case
PatientAge
(y)/Sex
Disease inNativeKidney
Type ofAllograft
MismatchedHLA
Antigens
Time FromTransplantion
(y) Treatment ProtocolClinical
Presentation Kidney Biopsy
BloodPressure(mm Hg)
SerumCr
(mg/dL) Proteinuria UrinalysisTreatment AfterKidney Biopsy
Outcome/Follow-upCr (mg/dL)
1 43/M IgA LRD NA 20 Oral steroid,cyclophosphamide
Chronic risein serumcreatinine,proteinuria
7% GS, 3.4 SS,7% crescents
140/90 4.3 3� NA NA Return to HD/12mo; attransplant 2,5 y
2 48/M IgA LRD 0 (identicaltwin)
17 Oral steroid,azathioprine
Recent rise inserumcreatinine,proteinuria
30%-40% GS, 0SS, 12.5%crescents
130/95 1.9 10.77 g/24 h 3� Blood azathioprine, oralsteroid,propranolol, ASA
Return to HD/6mo; died ofpulmonaryembolism 8 y
3 53/F IgA Unknown NA 6 Azathioprine,calcineurin inhibitor
Proteinuria 14% GS, 0 SS,20% crescents
NA 2.3 3.5 g/24 h NA NA Unknown
4 41/M IgA CAD NA 1.75 Oral steroid,calcineurin inhibitor
Hematuria 8% GS, 20% SS,30% crescents
140/84 5 NA NA IV steroid,plasmapheresis
Return to HD/1 mo
5 59/M PIGN LRD DR13 3 Oral steroid,calcineurin inhibitor
Recent rise inserumcreatinine
8% GS, 0 SS, 17%crescents
175/92 2.7 NA NA IV steroid Return to HD/2 mo
6 38/M IgA CAD A11,29,B35,50
9 Oral steroid,calcineurininhibitor,azathioprine
Recent rise inserumcreatinine,proteinuria
10.5% GS, 5.2%SS, 7%crescents
140/88 2 NA NA ACE inhibitor Functioning/4 y Cr 1.6
7 22/F IgA LRD A24, B70 3 Oral steroid,calcineurininhibitor,azathioprine
Recent rise inserumcreatinine,13 weekspregnant
0 GS, 0 SS, 13%crescents
136/90 6.5 3� 2� Blood IV steroid, calcineurininhibitor, ACEinhibitor
Functioning/5 y Cr 1.3
8 37/M Reflux LRD 0 (identicaltwin)
5 Calcineurin inhibitor,MMF
Hematuria 4.8% GS, 0 SS,14.3% crescents
NA 2.7 1� 3� Blood Oral steroid Functioning/6 y Cr 1.7
Previously reported cases54/M18 IgA CAD NA 0.75 Oral steroid Rapid rise in
serumcreatinine,grosshematuria
60% Crescents NA 9 NA Blood Plasmapheresis,cyclophosphamide,oral steroid
Return to HD/2 mo
18/M19 MPGN CAD A1, DR4 0.5 Oral steroid,calcineurin inhibitor
Rise in serumcreatinine,hematuria
80% Crescents NA 3.8 0.7 g/24 h Blood Plasmapheresis,cyclophosphamide,IV steroid
Return to HD/4 mo
43/M20 IgA CAD NA 4 Azathioprine, oralsteroid, calcineurininhibitor
Abdominalpain
40% Crescents NA 7.1 NA NA plasmapheresis, IVsteroid
Functioning/30d Cr 1.2
51/M21 NA LRD NA 4 Oral steroid,calcineurin inhibitor
Rise in serumcreatinine,hematuria
50% Crescents NA 11.6 10.6 g/24 h Blood IV steroid Functioning/1 y Cr 2.5
47/M22 IgA CAD B7, B38,DR13
0.75 OKT3, Oral steroid,calcineurininhibitor, tacrolimus
Rise in serumcreatinine,hematuria,proteinuria
33% Crescents 180/100 2.7 4.48 g/24 h Blood MMF, oral steroid,ACE inhibitor
Deteriorating/6mo Cr 3.7
NOTE. To convert serum creatinine in mg/dL to �mol/L, multiply by 88.4.
Abbreviations: PIGN, postinfectious glomerulonephritis; MPGN, membranoproliferative glomerulonephritis; CAD, cadaveric donor; GS, global glomerulosclerosis; SS, segmental glomerulosclerosis; Cr, creatinine; HD,
hemodialysis; NA, not available; IV, intravenous; ASA, acetylsalicylic acid; ACE, angiotensin-converting enzyme; MMF, mycophenolate mofetil; OKT3, monoclonal anti-CD3 antibodies.
IGA
NE
PH
RO
PA
TH
YW
ITH
CR
ES
CE
NT
S169
rnci(r
KOWALEWSKA ET AL170
Fig 1. Spectrum of morphological findings in patients with IgA nephropathy with crescents. (A) Low-power view ofenal cortex with some glomeruli involved in segmental necrosis and/or crescents. (B) Glomerulus with a small focus ofecrosis (arrow). (C) Mesangial proliferative changes with segmental necrosis of glomerular tuft and adjacent cellularrescent. (D) A full cellular crescent with only a few residual capillary loops in the plane of section. (E) Patient 1,mmunofluorescence staining for IgA in the glomerular mesangial region, typical of all patients included in this series.F) Patient 7, electron microscopy confirmed the presence of immune-type electron-dense deposits in mesangial
egions. ([A-D] Methenamine silver–periodic acid–Schiff stain; original magnification [A] �100; [B-D] �400.)
lafsIepcttsmTCcii
E
ftiutttasmd(lpssmtgs(ite
T
topb
wmctbrlvsfsticf(darOtps(ttPda
C
(ot6Trrfapnroahqc[
IGA NEPHROPATHY WITH CRESCENTS 171
ight chains, and complement components C1qnd C3 and scored using a semiquantitative scalerom 0 to 4�, as previously described.26 Biopsypecimens showed 2 to 4� granular staining forgA within mesangial areas (Fig 1E) and focallyxtending to glomerular capillary walls. The sameattern of staining was observed with C3, � lighthains, and � light chains, but with lesser intensi-ies than with IgA. In 5 cases, there was addi-ional mesangial staining of trace to 1� in inten-ity for IgM, and in 1 case, there was 1�
esangial staining for IgG in some glomeruli.here was no significant glomerular staining for1q, fibrinogen, or albumin. There was no signifi-ant staining of tubular basement membranes,nterstitium, or arteriolar vessels for any of themmune reactants.
lectron Microscopy
Electron microscopic evaluation was per-ormed in 5 patients. In the remaining patients,issue submitted for electron microscopy stud-es did not contain intact glomeruli suitable forltrastructural examination (patient 5) or elec-ron microscopic evaluation was deferred (pa-ients 2 and 7). Glomeruli examined ultrastruc-urally showed expansion of the mesangialreas, predominantly by increased matrix andometimes by an increase in cellularity. Someesangial areas were impregnated with well-
efined immune-type electron-dense depositsFig 1F). In patient 1, some glomerular capil-ary walls contained similar electron-dense de-osits in subendothelial locations. In patient 3,ome capillary walls were reduplicated andhowed cellular interposition between splitembranes. Preservation of foot processes of
he overlying glomerular epithelial cells variedreatly among patients, ranging from well pre-erved (patients 4 and 6) to extensively effacedpatients 1, 3, and 8). There was no evidence ofmmune-type electron-dense deposits withinubular basement membranes, interstitium, orxtraglomerular vessels.
herapy
Six of 8 patients were on immunosuppressionreatment protocols consisting of a combinationf steroids and either cyclophosphamide, azathio-rine, or calcineurin inhibitor (2- or 3-drug com-
ination therapy). Two patients (nos. 3 and 8) tere on a steroid-free immunosuppression regi-en (azathioprine or mycophenolate mofetil and
alcineurin inhibitor; Table 1). The posttransplan-ation course of 2 patients had been complicatedy single episodes of moderate to severe cellularejection (patient 4) and mild to moderate cellu-ar rejection with an additional component ofascular rejection (patient 7). The kidney biopsypecimen from patient 4, obtained 1 month be-ore the index biopsy containing crescents, alsohowed recurrent IgA nephropathy (based on 1o 2� mesangial IgA deposition detected bymmunofluorescence), but no evidence of cres-ent formation. The kidney biopsy specimenrom patient 7 at the time of the rejection episode3 years before the index biopsy with crescents)id not show evidence of recurrent IgA nephrop-thy. Rejection episodes were treated with ste-oids (patient 4) or a combination of steroids andKT3 (patient 7). After the renal biopsy diagnos-
ic of IgA nephropathy with crescents, mostatients were treated with a bolus of intravenousteroids. One patient underwent plasmapheresispatient 4) in addition to intravenous steroidherapy, and 1 patient was administered an angio-ensin-converting enzyme inhibitor (patient 6).atient 7, who was pregnant at the time of theiagnostic biopsy, did not receive any specificdditional treatment.
linical Outcome
Despite intensified therapy, 4 of 7 patients57%) with crescentic IgA nephropathy devel-ped renal failure and returned to hemodialysisherapy within 1 year (patient 1, 1 year; patient 2,months; patient 4, 1 month; patient 5, 2 months;able 1). One of these patients received a secondenal transplant, with no clinical evidence ofecurrence in the second allograft after 7 years ofollow-up. The biopsy specimen of the secondllograft kidney obtained 4 months after retrans-lantation showed no evidence of recurrent IgAephropathy at that time. The 3 other patientsemained on hemodialysis therapy; 1 patient diedf pulmonary embolism approximately 4 yearsfter the index biopsy. The 3 remaining patientsad a more favorable clinical course, with subse-uent stabilization of kidney function (serumreatinine levels ranging from 1.3 to 2.8 mg/dL115 to 248 �mol/L] in individual patients after 4
o 8 years of follow-up). These 3 patients, who
rugdut
ntImiamcvacgoort(c
ta1tpttptatgmptgmpo22ra
fop
arpharptsap2rk28rnlops
hanwnIipbabppak
w(cttiw
KOWALEWSKA ET AL172
etained their original allografts and did notndergo retransplantation, also generally had lesslomerular involvement by crescents in the in-ex biopsy specimen. None of these 3 patientsnderwent a subsequent renal biopsy. One pa-ient was lost to follow-up.
DISCUSSION
We report the first series of recurrent and deovo IgA nephropathy in renal allografts withhe additional features of crescent formation.gA nephropathy in native kidneys occasionallyay present with glomerular crescents. This find-
ng is uncommon and is estimated to account forpproximately 3% to 5% (when �50% of glo-eruli are involved by crescents) and 32% of
ases (if any number of glomeruli are in-olved).27 Similar findings in allograft kidneysre even less common, and to date, only a fewase reports are available in the English-lan-uage literature.18-22 Between 1989 and 2003 atur institution, we encountered 21 cases (0.7%f total number of kidney allograft biopsies) ofecurrent or de novo IgA nephropathy, and 8 ofhese cases (38%) showed crescent formationrange, 7% to 30% of glomerular involvement byrescents).
All previously reported patients (n � 5) withhese findings have been young men with a meange of 44.2 years (range, 18 to 54 years; Table).18-22 Four patients underwent kidney transplan-ation after renal failure secondary to IgA ne-hropathy, and 1 patient had membranoprolifera-ive glomerulonephritis as a cause of failure ofhe native kidney. The clinical presentation ofreviously reported patients was similar, charac-erized by microscopic hematuria, proteinuria,nd increasing serum creatinine levels. Averageime from transplantation to the diagnostic allo-raft biopsy was 23 months (range, 5 to 48onths). Allograft biopsy specimens in these
atients showed IgA nephropathy with the forma-ion of crescents involving an average 52% oflomeruli (range, 30% to 80%). Reported treat-ents usually were a combination therapy of
lasmapheresis and/or intravenous steroids. Threef these patients had a poor response to therapy,patients returned to hemodialysis therapy withinand 4 months, and 1 patient had deteriorating
enal function at the time of the report (6 months
fter the biopsy). Two remaining patients had dunctioning grafts 1 and 4 months after the bi-psy. None of these patients underwent retrans-lantation.Our findings, both clinical and morphological,
re similar to those observed in individuallyeported cases. The clinical presentation of allatients included in the present series includedematuria, proteinuria, and increasing serum cre-tinine levels. In some patients, the rapid loss ofenal function fit clinical criteria for rapidlyrogressive glomerulonephritis. In our group,here was a predominance of men (75%), corre-ponding to previously reported cases in whichll patients to date (100%) were men. Meanatient age in our group was 42.6 years (range,2 to 59 years), similar to that of the othereported cases. The source of the transplantedidney was either an LRD (71% in our study and0% in other cases) or deceased donor (28% and0%, respectively). LRD transplants have beeneported to pose a greater risk for recurrent IgAephropathy compared with deceased donors oriving unrelated transplants,8,28 and findings inur study group are consistent with such a predis-osition. We lack data to establish this relation-hip.
The majority of patients (75%) in our groupad an established diagnosis of IgA nephropathys a cause of end-stage renal failure of theirative kidneys, similar to other reported cases, ofhich 80% had an established diagnosis of IgAephropathy as the original cause of ESRD.mmunofluorescence studies were not performedn the native kidneys of 2 of our patients and in 1atient previously reported by others. The possi-ility that these patients had IgA nephropathy inddition to their other clinical diagnoses cannote excluded. However, it is clear that IgA ne-hropathy with crescents most often occurs inatients with preexisting IgA nephropathy and isn uncommon de novo disease in transplantedidneys.In contrast to previously published patients
ith a relatively short time posttransplantationrange, 0.5 to 4 years) at the time of biopsy, ourases are characterized by a longer posttransplan-ation course (range, 1.75 to 20 years). Also, inhe previously reported patients, there was greaternvolvement of glomeruli by crescent formationith a corresponding greater degree of renal
ysfunction in comparison to the lesser involve-
mfenato
intmyobcotgfis1utlrToiFft2(ird[yim��
aw(oIs
rt
ost71gy
ciup7gnlptltyatg(snntItobaid
tptssieppc
IGA NEPHROPATHY WITH CRESCENTS 173
ent by crescents and lesser degree of renalailure in our patients. It appears that the pres-nce of crescents in a kidney allograft with IgAephropathy at any time after transplantation isssociated with graft failure significantly earlierhan might be predicted based on our knowledgef the natural course of IgA nephropathy.For comparison, we also analyzed the remain-
ng 13 cases from our files of recurrent or deovo IgA nephropathy without crescent forma-ion. In this group, there were 5 women and 8en with a mean age of 39 years (range, 24 to 56
ears). Six patients had an established diagnosisf IgA nephropathy based on a native kidneyiopsy. One patient lost his native kidney be-ause of obstructive uropathy; 1 patient, becausef focal and segmental glomerulosclerosis; andhe remaining 5 patients had histories of chroniclomerulonephritis that was not otherwise speci-ed. Twelve patients presented with clinicalymptoms after an average of 8.25 years (range,to 17 years) posttransplantation, and 1 patient
nderwent a 1-year protocol biopsy. All 12 symp-omatic patients had elevated serum creatinineevels, with a mean of 2.2 mg/dL (195 �mol/L;ange, 0.9 to 3.8 mg/dL [80 to 336 �mol/L]).hree of 12 patients (25%) had hematuria, and 5f 12 patients (41%) had proteinuria, all except 1n the subnephrotic range (protein � 3.0 g/24 h).ollow-up was available for 10 patients. Meanollow-up was 4 years (range, 1 to 9 years) afterhe diagnostic biopsy and 9.5 years (range, 4 to6 years) posttransplantation. Two of 10 patients20%) lost their grafts 9 and 1.5 years after thendex biopsy (24 and 2 years posttransplantation,espectively), and 1 patient is experiencing aecline in graft function (creatinine, 3.5 mg/dL309 �mol/L]) 8 years after the index biopsy (14ears after kidney transplantation). The remain-ng 7 grafts (70%) are still functioning, with aean serum creatinine level of 1.4 mg/dL (130mol/L; range, 0.9 to 3.0 mg/dL [80 to 265mol/L]).Patients with recurrent or de novo IgA nephrop-
thy with crescent formation were associatedith a significantly greater chance of graft loss
57%; one unknown outcome) compared withur group of patients with recurrent or de novogA nephropathy without crescents (20%). The
ame conclusion may be drawn from the other deported patients with crescentic IgA in whichhe overall rate of graft loss was 60%.
Our observations are in agreement with thosef others for whom the chance of graft lossecondary to recurrent or de novo IgA nephropa-hy without crescents with a mean follow-up of6 months (6.3 years) posttransplantation was0%.29,30 In our group, there was an overall 20%raft loss, with a longer mean follow-up of 9.5ears.We analyzed additional factors that potentially
ould contribute to allograft loss in our patients,ncluding global and focal and segmental glomer-losclerosis. Biopsy specimens in 7 of 8 of ouratients (87.5%) showed mean involvement of.5% (range, 0% to 14%) of glomeruli by globallomerulosclerosis, which appears to be withinormal limits for the patients’ ages and is un-ikely to be of clinical significance. In only 1atient (no. 2), the biopsy specimen showed 30%o 40% of glomeruli involved by global glomeru-osclerosis, which, considering the long life ofhe kidney allograft in this particular patient (17ears), is likely to represent a feature of chronicllograft injury, possibly contributing in this pa-ient to loss of the graft. Focal and segmentallomerulosclerosis was present in 3 patients37%), 2 of whom lost their grafts. Focal andegmental glomerulosclerosis in allograft kid-eys can be either secondary to recurrent or deovo glomerulopathy or chronic cyclosporineoxicity or represent a chronic allograft injury.31
t usually is impossible to determine what causedhe formation of segmental lesions. The presencef segmental sclerosis in renal allografts haseen associated with greater proteinuria and hasgreater degree of global glomerulosclerosis and
nterstitial fibrosis, all contributing factors toecreased graft survival.31,32
Factors that potentially could contribute tohe recurrence of IgA nephropathy after trans-lantation have been proposed. Among them ishe association with certain HLA antigens,uch as HLA-B12, HLA-B35, and HLA-DR4;hort duration of disease before ESRD; ornfection with cytomegalovirus.9,23,33 How-ver, none of these factors has been highlyredictive of crescent development. Posttrans-lantation immunosuppression protocols andlinical course have no identifiable impact on
isease recurrence8,9,33 in the small number of
phrpootcIot
afucdaNrdrd
w
n
J
aC7
nd7
rt
n3
mN
ns1
gN
a
k
tp
ba
dA1
OiN
DA
gs3
nt
it
gn
up2
cf1
rr(
CY
iPA
l
Rt
KOWALEWSKA ET AL174
atients studied to date. LRD transplantationas been associated with an increased risk forecurrence and graft loss in some series ofatients with IgA nephropathy,33,34 but notthers.35 Our observations suggest that thenly factor identifiable as a significant risk forhe development of IgA nephropathy with cres-ents in an allograft kidney is the presence ofgA nephropathy in the native kidneys. Nother associations can be elicited clearly fromhe presented case series.
We conclude that IgA nephropathy in kidneyllografts characterized by epithelial crescentormation (crescentic IgA nephropathy) is anncommon and aggressive form of disease thatan occur at any time during the course of theisease, often follows a rapid clinical course,nd leads to allograft loss within a short time.o single risk factor can be recognized to be
esponsible for its development. Both the un-erlying pathophysiological characteristics andeliable and effective treatment for this disor-er remain enigmas that require further study.
REFERENCES
1. D’Amico G: The commonest glomerulonephritis in theorld: IgA nephropathy. Q J Med 64:709-727, 19872. Levy M, Berger J: Worldwide perspective of IgA
ephropathy. Am J Kidney Dis 12:340-347, 19883. Donadio JV, Grande JP: IgA nephropathy. N Engl
Med 347:738-748, 20024. D’Amico G, Colasanti G, Barbiano di Belgioioso G, et
l: Long-term follow-up of IgA mesangial nephropathy:linico-histological study in 374 patients. Semin Nephrol:355-358, 19875. Ibels LS, Gyory AZ: IgA nephropathy: Analysis of the
atural history, important factors in the progression of renalisease, and a review of the literature. Medicine (Baltimore)3:79-102, 19946. Berger J, Yaneva H, Nabarra B, Barbanel C: Recur-
ence of mesangial deposition of IgA after renal transplanta-ion. Kidney Int 7:232-241, 1975
7. Wang AY, Lai FM, Yu AW, et al: Recurrent IgAephropathy in renal transplant allografts. Am J Kidney Dis8:588-596, 20018. Odum J, Peh CA, Clarkson AR, et al: Recurrentesangial IgA nephritis following renal transplantation.ephrol Dial Transplant 9:309-312, 19949. Ohmacht C, Kliem V, Burg M, et al: Recurrent immu-
oglobulin A nephropathy after renal transplantation: Aignificant contributor to graft loss. Transplantation 64:1493-496, 199710. Berger J, Noel LH, Nabarra B: Recurrence of mesan-
ial IgA nephropathy after renal transplantation. Contrib
ephrol 40:195-197, 1984 K11. Berger J: Recurrence of IgA nephropathy in renalllografts. Am J Kidney Dis 12:371-372, 1988
12. Ramos EL, Tisher CC: Recurrent diseases in theidney transplant. Am J Kidney Dis 24:142-154, 199413. Ponticelli C, Traversi L, Feliciani A, et al: Kidney
ransplantation in patients with IgA mesangial glomerulone-hritis. Kidney Int 60:1948-1954, 200114. Kimata N, Tanabe K, Ishikawa N, et al: Correlation
etween proteinuria and prognosis of transplant IgA nephrop-thy. Transplant Proc 28:1537-1539, 1996
15. Hartung R, Livingston B, Excell L, Disney A, Woo-roffe AJ: Recurrence of IgA deposits/disease in grafts. Anustralian Registry Survey 1980-1990. Contrib Nephrol
11:13-16; discussion, 16-17, 199516. Deegens JK, Artz MA, Hoitsma AJ, Wetzels JF:
utcome of renal transplantation in patients with pauci-mmune small vessel vasculitis or anti-GBM disease. Clinephrol 59:1-9, 200317. Byrne MC, Budisavljevic MN, Fan Z, Self SE, Ploth
W: Renal transplant in patients with Alport’s syndrome.m J Kidney Dis 39:769-775, 200218. Diaz-Tejeiro R, Maduell F, Diez J, et al: Loss of renal
raft due to recurrent IgA nephropathy with rapidly progres-ive course: An unusual clinical evolution. Nephron 54:341-43, 199019. Robles NR, Gomez Campdera FJ, Anaya F, et al: IgA
ephropathy with rapidly progressive course after kidneyransplantation. Nephron 58:487-488, 1991
20. Streather CP, Scoble JE: Recurrent IgA nephropathyn a renal allograft presenting as crescentic glomerulonephri-is. Nephron 66:113-114, 1994
21. Park SB, Joo I, Park KA, et al: Rapidly progressivelomerulonephritis in a renal transplant with recurrent IgAephropathy. Transplant Proc 28:1529-1530, 199622. Benabdallah L, Rerolle JP, Peraldi MN, et al: An
nusual recurrence of crescentic nephritis after renal trans-lantation for IgA nephropathy. Am J Kidney Dis 40:E20,00223. Bachman U, Biava C, Amend W, et al: The clinical
ourse of IgA-nephropathy and Henoch-Schönlein purpuraollowing renal transplantation. Transplantation 42:511-515,98624. Shimizu T, Tanabe K, Tokumoto T, et al: A case of
apid progressive glomerulonephritis with IgA deposits afterenal transplantation. Clin Transplant 15:S11-S15, 2001suppl 5)
25. Churg J, Bernstein J, Glassock R: Renal Disease:lassification and Atlas of Glomerular Diseases (ed 2). Nework, NY, Igaku-Shoin, 1995, pp 134-13526. Alpers CE, Marchioro TL, Johnson RJ: Monoclonal
mmunoglobulin deposition disease in a renal allograft:robable recurrent disease in a patient without myeloma.m J Kidney Dis 13:418-423, 198927. Jennette JC: Rapidly progressive crescentic glomeru-
onephritis. Kidney Int 63:1164-1177, 200328. Brensilver JM, Mallat S, Scholes J, McCabe R:
ecurrent IgA nephropathy in living-related donor transplan-ation: Recurrence or transmission of familial disease? Am J
idney Dis 12:147-151, 1988
tN
n1
snD
g
pr
rp
Ft1
S
IGA NEPHROPATHY WITH CRESCENTS 175
29. Floege J, Burg M, Kliem V: Recurrent IgA nephropa-hy after kidney transplantation: Not a benign condition.ephrol Dial Transplant 13:1933-1935, 199830. Floege J: Recurrent glomerulonephritis following re-
al transplantation: An update. Nephrol Dial Transplant8:1260-1265, 200331. Cosio FG, Frankel WL, Pelletier RP, et al: Focal
egmental glomerulosclerosis in renal allografts with chronicephropathy: Implications for graft survival. Am J Kidneyis 34:731-738, 199932. Jeong HJ, Kim YS, Kwon KW, et al: Segmental
lomerulosclerosis in IgA nephropathy after renal trans- n
lantation: Relationship with proteinuria and therapeuticesponse to enalapril. Clin Transplant 17:108-113, 2003
33. Freese P, Svalander C, Norden G, Nyberg G: Clinicalisk factors for recurrence of IgA nephropathy. Clin Trans-lant 13:313-317, 199934. Bumgardner GL, Amend WC, Ascher NL, Vincenti
G: Single-center long-term results of renal transplanta-ion for IgA nephropathy. Transplantation 65:1053-1060,99835. Frohnert PP, Donadio JV Jr, Velosa JA, Holley KE,
terioff S: The fate of renal transplants in patients with IgA
ephropathy. Clin Transplant 11:127-133, 1997