Article

Patient Characteristics and Outcomes by GN Subtype inESRD

Michelle M. O’Shaughnessy,* Maria E. Montez-Rath,* Richard A. Lafayette,* and Wolfgang C. Winkelmayer†

AbstractBackground and objectives Outcomes-based research rarely focuses on patients with ESRD caused by GN. Thehypotheses were that the GN subtype would clinically discriminate patient groups and independently associatewith survival after ESRD therapy initiation.

Design, setting, participants, & measurements Data were extracted from the US Renal Data System for adultpatients with incident (1996–2011) ESRD attributed to six GN subtypes: FSGS, IgA nephropathy (IgAN), mem-branous nephropathy, membranoproliferative glomeruonephritis, lupus nephritis (LN), and vasculitis. ESRDattributed to diabetes and autosomal dominant polycystic kidney disease served as non-GN comparators. Un-adjusted and adjusted mortality hazard ratios (aHRs) with 95% confidence intervals (95% CIs) were estimatedusing Cox regression (reference, IgAN). Models sequentially adjusted for sociodemographic (model 2), comor-bidity/laboratory (model 3), and ESRD treatment modality (model 4) variables.

ResultsAmong 84,301 patientswith ESRDattributed to GN, themedian age ranged from39 (LN) to 66 (vasculitis)years, male sex ranged from 18% (LN) to 68% (IgAN), and black race ranged from 7% (IgAN) to 49% (LN).Patients with IgAN had the fewest comorbidities and lowest use of hemodialysis (70.1%). After a median follow-up of 2.5 (interquartile range, 1.0–4.9) years, crude mortality was lowest in IgAN (3.7 deaths/100 person years).Compared to IgAN, adjusted mortality was highest in LN (model 4 aHR=1.75; 95% CI, 1.68 to 1.83) and indiabetes (aHR=1.73; 95% CI, 1.67 to 1.79), and was also higher in all other GN subtypes (membranous ne-phropathy: aHR=1.23; 95% CI, 1.17 to 1.29; FSGS: aHR=1.37; 95% CI, 1.32 to 1.42; membranoproliferative GN:aHR=1.38; 95% CI, 1.31 to 1.45; vasculitis: aHR=1.51; 95% CI, 1.45 to 1.58) and in autosomal dominant polycystickidney disease (aHR=1.22; 95% CI, 1.18 to 1.27).

Conclusions This study exposes substantial heterogeneity across GN subtypes at ESRD therapy initiation andidentifies independent associations between GN subtype and post-ESRDmortality. These survival discrepancieswarrant further study, and the utility of current research practice to group GN subtypes together whenevaluating ESRD outcomes should be questioned.

Clin J Am Soc Nephrol 10: 1170–1178, 2015. doi: 10.2215/CJN.11261114

IntroductionGN is the thirdmost common cause of ESRD in patientsinitiating dialysis or receiving a kidney transplant in theUnited States. In the most recent Annual Data Report ofthe US Renal Data System (USRDS), GN accounted for6.3% of patients initiating ESRD therapy in 2011, trailingonly diabetes (43.9%) and hypertension (27.8%), andpatients with GN comprised 14.3% of the prevalenttreated ESRD population (1). Treatment of patients withESRD caused by GN incurs an estimated annual cost toMedicare alone of almost $3 billion (1). Mortality in GNincreases dramatically after the onset of ESRD (2–5);however, the relative contributions from genericESRD-related factors and from the underlying causeof GN to this finding have not formally been evaluated.

The USRDS adopts GN as one of four major cause-of-ESRD stratification variables in published reports,along with diabetes, hypertension, and cystic kidneydisease. However, when juxtaposing the characteristics

and outcomes of patients with GN to those with othercauses of ESRD, it is important to consider that GN isnot a single disease entity but rather a broad diseasecategory comprised of histologically and clinicallydistinct GN subtypes. Alongside differing renal man-ifestations, heterogeneity across GN subtypes withrespect to systemic comorbidities and mortality hasbeen identified in non-ESRD populations (6,7).Whether these phenotypic and prognostic distinctionsdiverge or converge after ESRD development remainslargely unknown.We conducted this study to examine differences

among GN subtypes with respect to demographic andclinical attributes at presentation to ESRD and prog-nosis after initiation of ESRD therapy. We posit thatexploring any such differences would serve to eluci-date and quantify long-term disease-specific risks,explain post-ESRD survival discrepancies betweenGN and non-GN patient groups, and facilitate the

*Division ofNephrology, StanfordUniversity School ofMedicine, Palo Alto,California; and†Section ofNephrology, BaylorCollege of Medicine,Houston, Texas

Correspondence:Dr. Michelle M.O’Shaughnessy,Stanford UniversitySchool of Medicine,Division ofNephrology, 1070Arastradero Road,Suite 3C3108, PaloAlto, CA 94304.Email: moshaugh@stanford.edu

www.cjasn.org Vol 10 July, 20151170 Copyright © 2015 by the American Society of Nephrology

development of a more individualized and equitable patientcare approach.

Materials and MethodsStudy Population and Data SourcesAll adult patients aged $18 years who initiated ESRD

therapy with hemodialysis, peritoneal dialysis, or kidneytransplantation between January 1, 1996, and December 31,2011, were retrospectively identified from the USRDS, a na-tional registry of almost all patients with treated ESRD. Pa-tients with ESRD attributed to one of six common GNsubtypes—FSGS, IgA nephropathy (IgAN), membranousnephropathy (MN), membranoproliferative GN (MPGN), lu-pus nephritis (LN), and vasculitis—were selected as the prin-cipal study cohort. These patients were identified usingcause of ESRD diagnostic codes (Supplemental Appendix)obtained from Medical Evidence Reports (form CMS-2728).These reports are submitted by attending nephrologists tothe Centers for Medicare and Medicaid Services within 45days of a patient commencing ESRD therapy. Patients withESRD attributed to diabetes (DN) and autosomal dominantpolycystic kidney disease (ADPKD) were selected as non-GN comparator groups. Missing or uncertain cause ofESRD or a defined cause other than the eight of interestwere the sole study exclusion criteria.

Patient CharacteristicsBaseline sociodemographic and clinical data were ex-

tracted from USRDS Patient and Medical Evidence files.Age, sex, race (white, black, Asian, or other), Hispanicethnicity (yes/no), Medicaid insurance (yes/no), and geo-graphic region (Northeast, Midwest, South, or West) wereselected as sociodemographic variables. Initial ESRD ther-apy modalities were defined as hemodialysis, peritonealdialysis, or kidney transplantation, using USRDS-definedcodes (Supplemental Appendix). Date of first kidneytransplantation was also obtained for patients initiatingESRD therapy with dialysis. Baseline comorbidities (repor-ted in Medical Evidence Reports to be present currently orwithin the last 10 years) included DN, heart failure, coro-nary heart disease, cerebrovascular disease, hypertension,chronic obstructive pulmonary disease, current smoking,cancer, peripheral vascular disease, and nonambulant sta-tus (unable to ambulate or transfer). Laboratory values(reported in Medical Evidence Reports as measured within45 days before commencing ESRD therapy) included albu-min, hemoglobin, and creatinine (used to calculate Modi-fication of Diet in Renal Disease [MDRD] eGFR).

OutcomesDeath was our primary study outcome. Cause of death

(reported on CMS-2746 Death Notification Forms), usingcollapsed USRDS-defined categories (Supplemental Ap-pendix), was a secondary outcome. Date and cause ofdeath were ascertained from USRDS Patient files. Patientswere censored at the end of study (January 1, 2012).

Statistical AnalysesCross-tabulation and distribution plots were used to

examine unadjusted differences in baseline characteristicsbetween groups. Categorical variables were summarized

as frequencies and proportions, and continuous variableswere summarized as medians and interquartile ranges ormeans6SDs, as appropriate. Differences in mortality wereexamined using time-to-event analysis. Cumulative sur-vival curves were derived by the Kaplan–Meier methodand compared using the log-rank test. Mortality hazardratios (HRs) with 95% confidence intervals (95% CIs) wereestimated using Cox proportional hazards regression, strat-ified by year of ESRD therapy initiation, with IgAN asthe referent group. Model 1 was unadjusted, model 2was adjusted for sociodemographic characteristics (age,sex, race, ethnicity, and Medicaid insurance), and model3 included additional adjustment for baseline comorbidityand laboratory variables. To adjust for ESRD therapy mo-dality at baseline and subsequent access to transplantation,we fitted a fourth model (model 4) that added to model 3baseline modality as a fixed covariate and post-ESRD trans-plantation as a time-dependent covariate. This allowed pa-tients who initiated ESRD therapy with dialysis but laterreceived a transplant to obtain a second ESRD treatmentrecord starting on the transplant date. Squared terms wereincluded for all continuous variables (age and laboratoryvalues). Proportionality was examined using plotted log(-log) survival curves.Approximately 32% of patients had at least one missing

variable. To handle these missing data, we assumed themto be missing at random and used standard multipleimputation techniques to impute up to 32 datasets (8). Inaddition to including all model 4 covariates, the imputa-tion model included the event indicator and the Nelson–Aalen estimator of the cumulative marginal hazard H(T),where T is the time to event or censoring (9). Imputationswere performed separately by year of ESRD therapy initi-ation and assumed a joint modeling approach (10). LogHR from the models applied to each imputation datasetwere then combined, as described by Little and Rubin (11).As a sensitivity analysis, models were repeated using com-plete case analysis.All datawere analyzed using SAS version 9.4 software (SAS

Institute, Cary, NC). This study was approved by an InternalReview Board of Stanford University School of Medicine.

ResultsThe final study population comprised 84,301 patients

with ESRD attributed to six major GN subtypes: 34,330(40.7%) with FSGS; 13,012 (15.4%) with IgAN; 7177 (8.5%)with MN; 5193 (6.2%) with MPGN; 16,463 (19.5%) with LN;and 8126 (9.6%) with vasculitis (Figure 1). In addition,36,272 patients with ADPKD and 720,001 patients withDN were studied as external non-GN comparator groups.Sociodemographic characteristics varied across GN sub-

types (Table 1). The median age ranged from 39 (LN) to 66years (vasculitis). There were approximately twice asmany men as women within all primary GN subtypes,ranging from 60% men in MPGN to 68% men in IgAN.Sex was balanced in vasculitis (52% men), and womenpredominated in LN (18% were men). Black race was over-represented in LN (48.6%) and FSGS (36.2%), Asian racewas overrepresented in IgAN (15.4%), and white race wasoverrepresented in vasculitis (88.8%). Medicaid insurancewas most common in LN (31.0%).

Clin J Am Soc Nephrol 10: 1170–1178, July, 2015 GN Subtype and ESRD Outcomes, O’Shaughnessy et al. 1171

Baseline reported comorbidity and laboratory characteristicsdiffered among GN subtypes. Congestive heart failure (8.3%),cerebrovascular disease (2.4%), and peripheral vascular disease(2.9%) were least common in IgAN, whereas DN (8.9%),atherosclerotic heart disease (6.3%), and cancer (1.6%) were leastcommon in LN. Serum albumin was highest in IgAN (3.560.7g/dl) and lowest in MN (2.960.9 g/dl). Hemoglobin was high-est in IgAN (10.161.9 g/dl) and lowest in LN (9.461.8 g/dl).Among non-GN comparator groups, patients with ADPKD

had a relatively favorable comorbidity and laboratory profile,comparable with that in IgAN. Patients with DN had thehighest comorbidity burden of all; however, hemoglobin andalbumin levels were higher than in some GN subtypes.The proportion of patients receiving a preemptive kidney

transplant was substantially higher in IgAN (11.9%) than inLN (3.1%) and vasculitis (1.5%). Use of peritoneal dialysisas an initial ESRD therapy was also higher in IgAN (16.8%)than in LN (10.8%) and vasculitis (6.5%).

On follow-up, 84,301 patients with ESRD attributed toGN contributed 431,657 person years of observation,during which time 33,774 deaths were observed (crudemortality rate, 7.8/100 person years). Unadjusted 5-yearsurvival ranged from 45.5% (95% CI, 44.2% to 46.7%) invasculitis to 81.5% (95% CI, 80.7% to 82.2%) in IgAN, asillustrated in Kaplan–Meier survival plots (Figure 2). Thesesurvival curves corresponded to mortality rates rangingfrom 3.69 (95% CI, 3.56 to 3.83) per 100 person years inIgAN to 15.91 (95% CI, 15.45 to 16.36) per 100 person yearsin vasculitis (Table 2). Survival discrepancies are furtherreflected in unadjusted HRs for mortality (Figure 3, Sup-plemental Table 1), which differed by .4-fold across GNsubtypes (HR=4.15; 95% CI, 3.96 to 4.35 in vasculitis). Ad-justment for sociodemographic characteristics attenuatedHRs in all subgroups except for LN, such that LN becamethe GN subtype associated with the highest sociodemo-graphic-adjusted mortality (model 2). Additional adjustment

Figure 1. | Flow diagram of cohort assembly. ADPKD, autosomal dominant polycystic kidney disease; DN, diabetes-related ESRD; IgAN, IgAnephropathy; LN, lupus nephritis; MN, membranous nephropathy; MPGN, membranoproliferative GN.

1172 Clinical Journal of the American Society of Nephrology

Tab

le1.

Baselinech

arac

teristicsac

cordingto

GN

subtype

Cha

racteristic

Prim

aryGN

Subtyp

esSecond

aryGN

Subtyp

esNon

-GN

Com

paratorGroups

FSGS

IgAN

MN

MPG

NLN

Vascu

litis

DN

ADPK

D

n(%

amon

gGN

type

s)34

,330

(40.7)

13,012

(15.4)

7177

(8.5)

5193

(6.2)

16,463

(19.5)

8126

(9.6)

720,00

1(n/a)

36,272

(n/a)

Age

,med

ian(IQR),y

51(38–

65)

44(33–

57)

59(47–

71)

53(41–

65)

39(29–

50)

66(54–

75)

63(54–

72)

54(47–

64)

Sex,

male

61.6

67.6

66.0

60.4

18.2

52.4

52.2

54.0

Race

White

59.0

75.1

69.5

74.5

43.4

88.8

64.9

82.0

Black

36.2

6.6

26.0

18.1

48.6

7.5

27.8

13.1

Asian

3.5

15.4

3.0

4.9

5.8

1.9

4.4

3.8

Other

1.4

2.9

1.5

2.4

2.1

1.8

2.8

1.0

Hispa

nicethn

icity

8.6

12.4

10.0

10.6

16.2

8.9

17.1

8.6

Med

icaidinsu

red

18.4

13.4

16.5

20.2

31.0

11.9

28.3

12.5

ESRD

therap

yHem

odialysis

78.4

70.1

82.0

80.9

85.5

91.6

91.4

70.5

Peritone

aldialysis

14.5

16.8

12.8

11.6

10.8

6.5

7.1

15.6

Trans

plan

t6.1

11.9

4.2

6.3

3.1

1.5

0.9

12.7

Missing

1.0

1.3

1.0

1.2

0.7

0.4

0.6

1.2

Geo

grap

hic

region

Northeast

20.0

19.6

20.4

20.9

16.1

19.3

16.8

19.7

Midwest

23.3

23.0

23.6

24.7

18.4

27.2

20.5

22.9

South

39.6

30.3

37.9

32.9

44.1

33.2

39.8

36.1

West

16.6

26.6

17.0

20.5

20.3

20.0

20.9

20.2

Missing

0.6

0.5

1.2

1.0

1.1

0.3

2.0

1.1

Com

orbidities

Diabe

tes

13.3

9.3

15.0

14.9

8.9

15.9

88.0

7.5

Heartfailu

re14

.68.3

18.8

17.9

15.0

18.2

38.5

8.4

Coron

aryhe

artd

isease

12.2

6.5

15.2

10.6

6.3

14.5

28.1

9.6

Cereb

rova

scular

even

t4.1

2.4

5.7

4.2

5.2

5.3

10.5

4.3

Hyp

ertens

ion

78.5

79.1

78.3

77.7

75.1

67.3

81.1

79.8

COPD

5.7

2.7

7.2

6.8

2.4

10.2

7.6

3.5

Current

smok

er7.4

4.8

7.2

8.5

4.2

4.4

4.7

5.8

Can

cer

4.8

2.8

6.2

5.7

1.6

6.0

3.9

3.6

PVD

5.1

2.9

6.2

5.0

3.5

7.6

19.8

3.4

Non

ambu

lant

1.6

0.9

2.6

2.2

2.4

3.5

6.6

1.0

Lab

oratorymea

suremen

tsAlbumin,g

/dl

3.36

0.8

3.56

0.7

2.96

0.9

3.06

0.8

2.96

0.8

3.06

0.7

3.16

0.7

3.86

0.6

Albumin

missing

23.1

22.5

22.3

22.3

23.6

23.2

25.5

23.4

Hem

oglobin,

g/dl

10.161.8

10.161.9

10.061.8

9.96

1.8

9.46

1.8

9.66

1.7

9.96

1.6

10.561.8

Hem

oglobinmissing

11.4

11.5

11.4

11.7

9.9

10.6

10.5

11.7

Creatinine,

mg/

d8.46

4.2

8.46

4.0

7.76

3.8

7.46

3.7

7.46

3.5

7.56

3.6

6.36

2.8

7.66

3.3

Creatininemissing

2.0

1.5

1.6

1.8

1.4

1.1

1.2

1.4

eGFR

,ml/min

per1.73

m2

8.76

4.1

8.46

4.0

9.26

4.5

9.56

4.6

9.46

4.6

8.76

4.3

10.664.7

8.56

3.8

eGFR

missing

2.6

2.1

2.6

2.9

2.5

1.7

2.2

2.0

Value

sarepe

rcen

tage

s,mean6

SDs,or

asothe

rwiseindicated

.IQR,interqu

artilerang

e;COPD,chron

icob

structivepu

lmon

arydisease;P

VD,p

eriphe

ralv

ascu

lard

isease;eGFR

,eGRFus

ingthe

4-va

riab

leMod

ification

ofDietin

Ren

alDisease

form

ula;

IgAN,IgA

nephrop

athy

;MN,m

embran

ousne

phrop

athy

;MPG

N,m

embran

oproliferativeGN;L

N,lup

usne

phritis;D

N,d

iabe

tes-

relatedESR

D;n

/a,

notap

plicab

le;A

DPK

D,a

utosom

aldom

inan

tpo

lycystickidne

ydisease.

Clin J Am Soc Nephrol 10: 1170–1178, July, 2015 GN Subtype and ESRD Outcomes, O’Shaughnessy et al. 1173

for baseline comorbidity and laboratory characteristics(model 3) and for ESRD therapy modality, including time-varying adjustment for kidney transplantation (model 4),attenuated HRs further within all GN subtypes, as comparedwith IgAN. Even after accounting for these differences in thecase mix, however, pronounced mortality differences acrossGN subtypes persisted. Compared with patients with ESRDcaused by IgAN, adjusted (model 4) mortality hazards were23% (95% CI, 17% to 29%), 37% (95% CI, 32% to 42%), 38%(95% CI, 31% to 45%), 51% (95% CI, 45% to 58%), and 75%(95% CI, 68% to 83%) higher in MN, FSGS, MPGN, vascu-litis, and LN, respectively. Adjusted mortality in IgAN was

also lower than in ADPKD (adjusted mortality hazard ratio[aHR]=1.22; 95% CI, 1.18 to 1.27), a disease with a generallyfavorable prognosis in ESRD (12,13), whereas mortality inLN (aHR=1.75; 95% CI, 1.68 to 1.83) was similar to in DN(aHR=1.73; 95% CI, 1.67 to 1.79), a disease with a particu-larly poor prognosis in ESRD (14).We observed some differences across GN subtypes with

respect to primary cause of death (Figure 4, SupplementalTable 2). Cardiovascular disease accounted for the highestproportion of deaths within all GN subtypes, ranging from34.2% in vasculitis to 44.6% in FSGS. The highest propor-tion of infection-related deaths was observed in LN (14.0%),

Figure 2. | Kaplan–Meier survival curves. ADPKD, autosomal dominant polycystic kidney disease; DN, diabetes-related ESRD; IgAN, IgAnephropathy; LN, lupus nephritis; MN, membranous nephropathy; MPGN, membranoproliferative GN.

Table 2. Follow-up details of the study population, including unadjusted mortality rates, by GN subtype

GN Subtype No. ofPatients (%)

Person Timeat Risk, y

No. ofDeaths (%)

Mortality Rate per 100 PersonYears (95% Confidence Interval)

IgAN 13,012 (15.4) 76,901 2839 (21.8) 3.69 (3.56 to 3.83)FSGS 34,330 (40.7) 179,364 13,721 (40.0) 7.65 (7.52 to 7.78)LN 16,463 (19.5) 84,585 6741 (41.0) 7.97 (7.78 to 8.16)MPGN 5193 (6.2) 26,869 2405 (46.3) 8.95 (8.59 to 9.31)MN 7177 (8.5) 34,482 3382 (47.1) 9.81 (9.48 to 10.14)Vasculitis 8126 (9.6) 29,456 4686 (57.7) 15.91 (15.45 to 16.36)

IgAN, IgA nephropathy; LN, lupus nephritis; MPGN, membranoproliferative GN; MN, membranous nephropathy.

1174 Clinical Journal of the American Society of Nephrology

comparing with rates of #10.6% among primary GN sub-types. Malignancy-related deaths were comparatively rare,ranging from 2.3% in LN to 6.5% in MN.

Sensitivity Analyses: Imputed versus Complete DataAnalysesBecause of computational limitations, we were unable to

generate 32 imputed datasets to reflect our 32% missingdata frequency, as previously suggested (8). Instead, wegenerated results using eight and 15 imputed datasets. Assummarized in Supplemental Figure 1 and SupplementalTable 1, complete case analyses tended to yield somewhatlarger associations compared with multiply imputed anal-yses. Results did not materially change when increasingfrom eight to 15 imputed datasets, suggesting that addi-tional efficiency was unlikely to be gained by extendingimputation beyond 15 datasets.

DiscussionIn this large, national, study of patients with ESRD at-

tributed to GN who initiated dialysis or received a preemp-tive kidney transplant between 1996 and 2011, we identified

considerable sociodemographic and clinical differencesacross six important GN subtypes. We also observed markedsurvival discrepancies that persisted even after adjustmentfor sociodemographic and clinical factors. SomeGN subtypes(e.g., IgAN) conferred a particularly favorable prognosis, su-perior to that in ADPKD, whereas others (e.g., LN) displayedshortened survival, similar to in patients with ESRD causedby DN. The most obvious conclusion to draw from thesefindings is that combining GN subtypes into a single dis-ease category, as is current practice in research and publichealth reporting, is of questionable utility; this approach failsto recognize the heterogeneity and complexity pervadingthis patient group.Our study is not the first to report differential survival

outcomes across GN subtypes. A single-center, retrospectivestudy of 580 Taiwanese patients with biopsy-proven GN, notyet requiring dialysis (mean eGFR, 70.4633.8 ml/min per1.73 m2), reported a lower baseline comorbidity and cyto-toxic treatment burden in patients with IgAN than withMN or FSGS (6). Unadjusted mortality, after a medianfollow-up of 5.9 years, was significantly lower in IgAN (4.6%)compared with MN (17.2%) and FSGS (14.4%); however,adjustment for between-group demographic and clinical

Figure 3. | Unadjusted and adjusted mortality hazard ratios. Model 1 is unadjusted; model 2 is demographic adjusted; model 3 is de-mographic and comorbidity adjusted; and model 4 is demographic, comorbidity, and ESRD therapy modality adjusted. ADPKD, autosomaldominant polycystic kidney disease; DN, diabetes-related ESRD; IgAN, IgA nephropathy; LN, lupus nephritis; MN,membranous nephropathy;MPGN, membranoproliferative GN.

Clin J Am Soc Nephrol 10: 1170–1178, July, 2015 GN Subtype and ESRD Outcomes, O’Shaughnessy et al. 1175

differences was not performed, and outcomes after ESRDdevelopment were not examined. A second study of 1943Korean patients with primary GN again demonstrated asurvival advantage in IgAN compared with MN, FSGS,and MPGN (7). In the subset of patients who developedESRD (n=257), 10-year survival risks of 85%, 80% (approx-imately), 61%, and 26%, respectively, were reported. Mor-tality comparisons accounting for case-mix differenceswere, again, not performed. With the exception of this sub-group analysis, most other studies examining mortalityoutcomes after ERSD development in GN were restrictedto single GN subtypes (3,15–17) or combined subtypesinto a single GN category (18,19), precluding direct com-parisons across GN subtypes.This study translates these previous findings, derived

from non-ESRD, non-United States patient populations to anationally representative cohort of United States patientswith treated ESRD. It expands on prior studies by com-paring outcomes not only across primary GN subtypes butalso across secondary GN subtypes and non-GN relatedcauses of ESRD. Importantly, this study also addresses thequestion of whether GN subtype is independently asso-ciated with post-ESRD survival or whether mortalitydifferences are explained by differences in case mix alone.We confirm that post-ESRD survival in IgAN not onlyexceeds survival in other primary GN subtypes but also

exceeds survival in secondary GN subtypes and ADPKD.This latter finding conflicts with prior reports suggestinginferior survival in GN compared with ADPKD. For ex-ample, a study of 44,240 Brazilian patients with treatedESRD reported a demographic-adjusted relative risk formortality, with reference to hypertension-associated ESRD,of 0.93 (95% CI, 0.89 to 0.98) in GN compared with 0.69(95% CI, 0.61 to 0.78) in ADPKD (18). Within the UnitedStates, 1-year mortality is reportedly 2-fold higher in primaryGN than in cystic renal disease (10% versus 5%, respectively)(1). Our data explain these findings by demonstrating aspectrum of risk within GN: patients with IgAN have asurvival advantage over patients with ADPKD that iscounterbalanced by a survival disadvantage in other GNsubtypes. This finding escapes detection when individualGN subtypes are combined together in ESRD outcomesresearch.The substantially higher mortality observed in patients

with LN in this study warrants further mention. High ratesof infection (20–22), cardiovascular disease (23), and hos-pitalization (21,23) were previously described in patientswith ESRD caused by LN; however, direct comparisonswith other GN subtypes are largely lacking, and studiesinvestigating mortality report conflicting findings (21,23–27).We determined that unadjusted crude mortality in LNexceeded mortality identified in many other GN subtypes,

Figure 4. | Cause of death categories by GN subtype. ADPKD, autosomal dominant polycystic kidney disease; DN, diabetes-related ESRD;IgAN, IgA nephropathy; LN, lupus nephritis; MN, membranous nephropathy; MPGN, membranoproliferative GN.

1176 Clinical Journal of the American Society of Nephrology

despite this being the youngest patient group with thehighest proportion of patients who were black and patientswho were women, factors which should have portended afavorable prognosis (28). Indeed, adjustment for sociode-mographic factors uniquely increased the relative risk formortality in LN with respect to IgAN, in contrast with therisk attenuation that was observed in all other GN subtypes(Figure 3). Further adjustment for differences in clinicalcharacteristics, including access to kidney transplantation,reduced the HR for mortality in LN somewhat, but it re-mained almost 2-fold higher than in the referent group,IgAN.Our study has several limitations. First, we cannot con-

firm the validity of GN subtype designations obtained fromthe USRDS. A previous study measured agreement be-tween biopsy-based diagnoses and USRDS-derived diag-noses among 227 patients with biopsy-proven GN (29).Poor overall agreement was largely explained by missing(57%) and GN not histologically examined (9%) diagnosessubmitted to the USRDS; positive predictive values exceeded90% once a specific GN subtype was selected. Agreementalso improved after 1995, following revisions to the Med-ical Evidence Report diagnostic coding system in thatyear. Nevertheless, we could not always distinguish pri-mary from secondary forms of GN (e.g., primary from sec-ondary MN), and our study findings are not necessarilyapplicable to nonbiopsied or misclassified patients withGN (i.e., false negatives) who may differ fundamentallyfrom correctly classified patients. Second, as a retrospectiveobservational study, associations between GN subtype andmortality cannot be assumed to represent causation. Wecould not distinguish the influence of GN-related factors(e.g., nephrotic syndrome, systemic inflammation, immu-nosuppressive therapy) from unmeasured or residualnon-GN related factors. Some misclassification of comor-bidities is likely (e.g., DN was reported in only 88% ofpatients with ESRD caused by DN), and detailed socio-economic data were not available; however, we proposethat misclassification of these confounding variables islikely to be nondifferential and to bias findings towardthe null. At the same time, adjustment for laboratory var-iables and initial ESRD treatment modality may overadjustfor disease mediators, particularly in those GN subtypesthat typically display a more rapid and unpredictablecourse to ESRD (e.g., LN, vasculitis). Albumin was lowestin the GN subtypes most typically associated with nephroticsyndrome (MN and MPGN) and systemic inflammation(LN and vasculitis), whereas hemoglobin was lowest inthose GN subtypes most likely to be treated with immu-nosuppressive therapy (LN and vasculitis), suggestingdirect disease- and treatment-mediated effects. Finally,our findings apply to patients with progressive GN whosurvive to ESRD and should not be generalized to patientswith mild, treated, or remitted GN without ESRD or topatients who die before developing ESRD.Despite these limitations, our study has a number of

strengths. We report findings derived from population-based data that are broadly applicable to all patients withESRD attributed to GN receiving dialysis or with a func-tioning kidney transplant in the United States. Studyinvestigators did not collect primary data or adjudicateoutcomes, virtually eliminating investigator bias. We used

sophisticated statistical techniques to overcome shortfallsinherent to observational study design, including multipleimputation methods to handle missing data and use ofsequentially adjusted models to minimize confounding.In summary, we have identified in a large, nationally

representative, ESRD cohort that patients classified intoindividual histologic GN subtypes differ considerably fromone another with respect to sociodemographic, clinical,laboratory, and ESRD therapy modality characteristics. Wefurthermore determined that GN subtype independentlyassociates with survival after initiation of ESRD treatment,even after accounting for differences in case mix. We pro-pose that GN subtype be addressed in all future studies ofpatients with ESRD caused by GN, to elucidate explana-tions for observed survival differences and to identify modifi-able factors amenable to targeting in interventional trials andpublic health strategies.

AcknowledgmentsDr. O’Shaughnessy was supported by a Ben J. Lipps Research

Fellowship of the ASN Foundation for Kidney Research. TheStanford Nephrology Fellowship Program was supported by grantT32-DK007357. Dr. Winkelmayer receives salary and research sup-port through the endowed Gordon A. Cain Chair in Nephrology atBaylor College of Medicine.The abstract from this article was presented to the American

Society of Nephrology Kidney Week meeting, November 2014,Philadelphia, Pennsylvania.Themanuscriptwas reviewedandapproved forpublicationbyan

officerof theNational InstituteofDiabetesandDigestiveandKidneyDiseases. Data reported herein were supplied by the US Renal DataSystem. Interpretation and reporting of these data are the re-sponsibility of the authors and in no way should be seen as officialpolicy or interpretation of the United States government.

DisclosuresDr. Winkelmayer reports having served as an advisor or consul-

tant, unrelated to the topic of this article, to ACUMEN, Amgen,Astra-Zeneca, Bayer, Keryx, Medtronic, Mitshubishi-Tanabe, andRockwell Pharma.Dr. Lafayette reports having served as an advisoror consultant, unrelated to the topic of this article, to Genentech,Fibrogen, and Questcor. Drs. O’Shaughnessy and Montez-Rathhave no financial disclosures to report.

References1. USRD System: USRDS 2013 Annual Data Report: Atlas of

Chronic Kidney Disease and End-Stage Renal Disease in theUnited States, Bethesda, MD, National Institutes of Health, Na-tional Institute of Diabetes and Digestive and Kidney Diseases,2013

2. Knoop T, Vikse BE, Svarstad E, Leh S, Reisæter AV, Bjørneklett R:Mortality in patients with IgA nephropathy. Am J Kidney Dis 62:883–890, 2013

3. Lee H, Kim DK, Oh KH, Joo KW, Kim YS, Chae DW, Kim S, ChinHJ: Mortality of IgA nephropathy patients: A single center expe-rience over 30 years. PLoS One 7: e51225, 2012

4. Mok CC, Kwok RC, Yip PS: Effect of renal disease on the stan-dardized mortality ratio and life expectancy of patients withsystemic lupus erythematosus. Arthritis Rheum 65: 2154–2160,2013

5. Yap DY, Tang CS, Ma MK, Lam MF, Chan TM: Survival analysisand causes of mortality in patients with lupus nephritis. NephrolDial Transplant 27: 3248–3254, 2012

6. Chou YH, Lien YC, Hu FC, LinWC, Kao CC, Lai CF, ChiangWC,Lin SL, Tsai TJ, Wu KD, Chen YM: Clinical outcomes and

Clin J Am Soc Nephrol 10: 1170–1178, July, 2015 GN Subtype and ESRD Outcomes, O’Shaughnessy et al. 1177

predictors for ESRD and mortality in primary GN. Clin J Am SocNephrol 7: 1401–1408, 2012

7. Lee H, Kim DK, Oh KH, Joo KW, Kim YS, Chae DW, Kim S, ChinHJ: Mortality and renal outcome of primary glomerulonephritisin Korea: Observation in 1,943 biopsied cases. Am J Nephrol 37:74–83, 2013

8. Montez-Rath ME, Winkelmayer WC, Desai M: Addressingmissing data in clinical studies of kidney diseases. Clin J Am SocNephrol 9: 1328–1335, 2014

9. White IR, Royston P: Imputing missing covariate values for theCox model. Stat Med 28: 1982–1998, 2009

10. van Buuren S: Flexible Imputation of Missing Data, Boca Raton,FL, Chapman and Hall/CRC Press, 2012

11. Little RJA, Rubin DB: Statistical Analysis with Missing Data, 2ndEd., Hoboken NJ, John Wiley & Sons, 2002

12. Perrone RD, Ruthazer R, Terrin NC: Survival after end-stage renaldisease in autosomal dominant polycystic kidney disease: con-tribution of extrarenal complications to mortality. Am J KidneyDis 38: 777–784, 2001

13. Reule S, Sexton DJ, Solid CA, Chen SC, Collins AJ, Foley RN:ESRD from autosomal dominant polycystic kidney disease in theUnited States, 2001-2010. Am J Kidney Dis 64: 592–599, 2014

14. Schroijen MA, van de Luijtgaarden MW, Noordzij M, Ravani P,Jarraya F, Collart F, Prutz KG, Fogarty DG, Leivestad T, Prischl FC,Wanner C, Dekker FW, Jager KJ, Dekkers OM: Survival in dialysispatients is different between patients with diabetes as primaryrenal disease and patients with diabetes as a co-morbid condi-tion. Diabetologia 56: 1949–1957, 2013

15. Komatsu H, Kikuchi M, Nakagawa H, Fukuda A, Iwakiri T, ToidaT, Sato Y, Kitamura K, Fujimoto S: Long-term survival of patientswith IgA nephropathy after dialysis therapy. Kidney Blood PressRes 37: 649–656, 2013

16. Romeu M, Couchoud C, Delaroziere JC, Burtey S, Chiche L,Harle JR, Gondouin B, Brunet P, Berland Y, Jourde-Chiche N:Survival of patients with ANCA-associated vasculitis on chronicdialysis: data from the French REIN registry from 2002 to 2011.QJM 107: 545–555, 2014

17. TangW, Bose B, McDonald SP, Hawley CM, Badve SV, Boudville N,Brown FG, Clayton PA, Campbell SB, Peh CA, Johnson DW: Theoutcomes of patients with ESRD and ANCA-associated vasculitis inAustralia andNewZealand.Clin J Am SocNephrol 8: 773–780, 2013

18. Batista PB, Lopes AA, Costa FA: Association between attributedcause of end-stage renal disease and risk of death in Brazilianpatients receiving renal replacement therapy. Ren Fail 27: 651–656, 2005

19. Wolfe RA, Ashby VB, Milford EL, Ojo AO, Ettenger RE, AgodoaLY, Held PJ, Port FK: Comparison of mortality in all patients ondialysis, patients on dialysis awaiting transplantation, and re-cipients of a first cadaveric transplant. N Engl J Med 341: 1725–1730, 1999

20. Chang YS, Liu CJ, Wu TH, Chaou CH, Lin KC, Ou SM, Chen TJ,Chen WS, Chou CT, Tsai CY: Survival analysis in systemic lupus

erythematosus patients on maintenance dialysis: A nationwidepopulation-based study in Taiwan. Rheumatology (Oxford) 52:166–172, 2013

21. Siu YP, Leung KT, Tong MK, Kwan TH, Mok CC: Clinical out-comes of systemic lupus erythematosus patients undergoingcontinuous ambulatory peritoneal dialysis. Nephrol Dial Trans-plant 20: 2797–2802, 2005

22. Huang JW, Hung KY, Yen CJ, Wu KD, Tsai TJ: Systemic lupuserythematosus and peritoneal dialysis: outcomes and infectiouscomplications. Perit Dial Int 21: 143–147, 2001

23. Sule S, Fivush B, Neu A, Furth S: Increased hospitalizations anddeath in patients with ESRD secondary to lupus. Lupus 21: 1208–1213, 2012

24. ChenHA,Wang JJ, ChouCT, Chien CC, ChuCC, SheuMJ, Lin YJ,Chen PC, Chen CH: Predictors of longterm mortality in patientswith andwithout systemic lupus erythematosus onmaintenancedialysis: A comparative study. J Rheumatol 38: 2390–2394,2011

25. Lee PT, Fang HC, Chen CL, Chiou YH, Chou KJ, Chung HM: Poorprognosis of end-stage renal disease in systemic lupus eryth-ematosus: A cohort of Chinese patients. Lupus 12: 827–832,2003

26. Mallett A, Tang W, Clayton PA, Stevenson S, McDonald SP,Hawley CM, Badve SV, Boudville N, Brown FG, Campbell SB,Johnson DW: End-stage kidney disease due to Alport syndrome:outcomes in 296 consecutive Australia and New Zealand Di-alysis and Transplant Registry cases.Nephrol Dial Transplant 29:2277–2286, 2014

27. Ward MM: Cardiovascular and cerebrovascular morbidityand mortality among women with end-stage renal disease at-tributable to lupus nephritis. Am J Kidney Dis 36: 516–525,2000

28. Bloembergen WE, Port FK, Mauger EA, Wolfe RA: Causes ofdeath in dialysis patients: Racial and gender differences. J AmSocNephrol 5: 1231–1242, 1994

29. Layton JB, Hogan SL, Jennette CE, Kenderes B, Krisher J, JennetteJC, McClellan WM: Discrepancy between Medical EvidenceForm 2728 and renal biopsy for glomerular diseases. Clin J AmSoc Nephrol 5: 2046–2052, 2010

Received: November 12, 2014 Accepted: February 26, 2015

Published online ahead of print. Publication date available at www.cjasn.org.

This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.11261114/-/DCSupplemental.

See related editorial, “ESRD Outcomes and GN Subtypes,” onpages 1117–1118.

1178 Clinical Journal of the American Society of Nephrology