Background: Immunoglobulin A nephropathy (IgAN) is the most common glomerulonephritis globally. Fewstudies have investigated mortality in patients with IgAN compared with the age- and sex-adjusted generalpopulation.
Study Design: Cohort study with record linkage between the Norwegian Kidney Biopsy Registry, NorwegianCause of Death Registry, and Norwegian Renal Registry.
Setting & Participants: 633 patients diagnosed with IgAN in 1988-2004.Predictor: Estimated glomerular filtration rate (eGFR), age, and sex.Outcomes: Deaths and causes of death before and after the onset of end-stage renal disease through 2008.Results: Mean follow-up was 11.8 (range, 0-20.8) years. During the observation period, the observed
number of deaths was 80 and the expected number was 42.1, resulting in a standardized mortality ratio (SMR)of 1.9 (95% CI, 1.5-2.4). Risk stratification based on initial eGFR showed that SMR was 1.0 (95% CI, 0.6-1.6) ifeGFR was �60 mL/min/1.73 m2, 1.9 (95% CI, 1.3-2.8) if eGFR was 30-60 mL/min/1.73 m2, and 3.6 (95% CI,2.6-5.0) in patients with eGFR �30 mL/min/1.73 m2. Renal replacement therapy (RRT) was initiated in 146patients and 35 of the 80 deaths occurred after the start of RRT. The age- and sex-adjusted SMR was notincreased significantly in the pre-RRT period (1.3; 95% CI, 1.0-1.7), but was increased after initiation of RRT(4.9; 95% CI, 3.5-7.0). The most common cause of death was cardiovascular disease, accounting for 45% of alldeaths.
Limitations: Treatment during follow-up is not known.Conclusions: Mortality in patients with IgAN was twice the expected rate, but not significantly increased
INDEX WORDS: Immunoglobulin A nephropathy; prognosis; mortality; end-stage renal disease; risk factor.
Immunoglobulin A nephropathy (IgAN) is the mostcommon glomerulonephritis globally.1-3 Kidney
disease outcome in patients with IgAN is variable;long-term stable kidney function is observed in manypatients, whereas others progress toward end-stagerenal disease (ESRD).4,5 Risk factors for progressivekidney disease in patients with IgAN are studiedfrequently, and several prognostic models to stratifypatients according to risk of ESRD have been devel-oped.4,6-14 However, few studies have reported whetherthese risk factors are associated with an excess mortal-ity rate in patients with IgAN compared with the age-and sex-adjusted general population. Furthermore,neither mortality rate nor causes of death in patientswith IgAN have been compared with those of patientswith chronic kidney disease (CKD) in general.
Although there is a lack of mortality data forpatients with IgAN, the risk and causes of mortality inpatients with CKD have been studied extensively. It iswell known that the mortality rate is increased substan-tially in patients with CKD overall compared with theage- and sex-adjusted general population. Many stud-ies have shown that the risk of death usually greatlyexceeds the risk of ESRD across all CKD stages, andcardiovascular disease is the most common cause of
death in these patients.15-20 However, in young pa-
Am J Kidney Dis. 2013;xx(x):xxx
tients with CKD with diabetic nephropathy or primarykidney diseases, a much higher risk of progression toESRD than pre-ESRD death has been observed.21-23
In patients with CKD established in renal replacementtherapy (RRT), the mortality rate is higher in patientsreceiving dialysis than in transplant recipients,24 butmortality also is several times increased in transplantrecipients.25 Patients with IgAN are younger andoften have a lower comorbid condition burden thanmost other patients with CKD and thus may haveother risk patterns than the average patient with CKD.
From the 1Department of Clinical Medicine, Renal ResearchGroup, University of Bergen; 2Department of Medicine, Hauke-land University Hospital, Bergen; 3Department of Medicine, Haug-esund Hospital, Haugesund; 4Department of Pathology, Hauke-land University Hospital; 5Department of Clinical Medicine, TheGade Laboratory for Pathology, University of Bergen, Bergen;and 6Section of Nephrology, Department of Transplant Medicine,Oslo University Hospital, Rikshospitalet, Oslo, Norway.
Received November 5, 2012. Accepted in revised form April 25,2013.
Address correspondence to Rune Bjørneklett, MD, PhD, Depart-ment of Medicine, Haukeland University Hospital, 5021 Bergen,Norway. E-mail: [email protected]
The present study has analyzed mortality data for633 Norwegian patients with IgAN with up to 20years of follow-up and compared these data with theoverall age- and sex-adjusted population. We hypoth-esized that mortality risk would be associated stronglywith risk of ESRD.
METHODS
RegistriesUsed in thePresent Study
The Norwegian Kidney Biopsy Registry was established in1988 and has registered clinical, biochemical, immunologic, andmorphologic data from patients who have had a kidney biopsyperformed in Norway (current population of 5 million) since 1988.All kidney biopsies have been evaluated by an experienced nephro-pathologist at Haukeland University Hospital in Bergen. TheNorwegian Renal Registry is located at Oslo University Hospital,Rikshospitalet, and has registered all patients in Norway withESRD (starting maintenance dialysis treatment or receiving akidney transplant) since 1980. The Norwegian Cause of DeathRegistry is part of Statistics Norway and information in thisregistry is based on the compulsory Norwegian death certificate.The Norwegian Population Registry is a part of the tax office inNorway; all Norwegian citizens are registered with a unique11-digit personal number, and deaths are registered on a weeklybasis.
Identificationof PatientsWith IgAN
For this study, which was approved by the Regional EthicsCommittee, we included all patients with a diagnosis of IgAN inthe Norwegian Kidney Biopsy Registry in 1988-2004. Diagnosisof IgAN was based on criteria in the World Health Organizationmonograph of kidney disease, and the most common morphologicfindings were mesangioproliferative changes with mesangial depos-its of IgA.26
Methodof Follow-up
Using the 11-digit personal identification number, reliable re-cord linkage was performed with the Population Registryof Norway, and we identified all patients who had died by theend of 2008. Cause of death was obtained from the NorwegianCause of Death Registry. Causes of death were classified asvascular disease, malignant disease, infectious disease, or otherdisease/accident. Using the same personal identification number,record linkage also was done with the Norwegian Renal Registry,and all patients who had developed ESRD by the end of 2008 wereidentified.
DefinitionofObservationPeriod
The observation period was defined as the period between thedate of the diagnostic kidney biopsy and death or end of 2008,whichever came first.
Definitionof IgANRiskGroups
To risk stratify the patients in our cohort, we used 3 methods.First, we divided the cohort into 3 groups according to the initialestimated glomerular filtration rate (eGFR) at time of kidneybiopsy. Low, moderate, and high risk were defined as eGFR �60,30-60, and �30 mL/min/1.73 m2, respectively. Second, we di-vided the cohort into 3 groups according to proteinuria (proteinexcretion �1, 1-2.9, and �3 g/24 h) at time of kidney biopsy.Third, we used a recently published Japanese scoring system13 that
we have validated on the same cohort used in the present study27;
2
the scoring system allowed prediction of 10-year risk of ESRD inpatients with IgAN and stratified patients into low-, moderate-, andhigh-risk groups. We defined low risk as �10%, moderate risk as10%-50%, and high risk as �50% expected risk of ESRD.
VariablesUsed inRisk Stratification
EstimatedGFR
We used the CKD-EPI (CKD Epidemiology Collaboration)creatinine equation (whites) to calculate eGFR. We stratifiedpatients based on eGFR data into 3 groups: �60, 30-59.9, and �30mL/min/1.73 m2. Serum creatinine level was not standardized toisotope-dilution mass spectrometry and therefore was reduced by5%.28
Proteinuria
Proteinuria was measured quantitatively and reported as gramsper 24 hours or grams per liter in �80% of cases. Grams per literwas converted to grams per 24 hours by multiplication with 1.5. Inthe other cases, urine dipstick test was used. Results were con-verted as follows: 1�, 0.3 g/24 h; 2�, 1.0 g/24 h; and 3�, 3 g/24 h.When a urine dipstick result was missing, notification of indicationfor biopsy on the Kidney Biopsy Registry report form was used.When the answer to the question “proteinuria yes/no” was no,proteinuria was set to protein excretion �1 g/24 h; if it was yes,proteinuria was set to protein excretion of 1-2.9 g/24 h. Finally,when the answer to the question “nephrotic syndrome yes/no” wasyes, proteinuria was set to protein excretion �3 g/24 h. Westratified patients into groups with proteinuria with protein excre-tion �1, 1-2.9, and �3 g/24 h.
Hypertension
We defined hypertension as blood pressure �140/90 mm Hg. Incases with missing data for blood pressure (�10%), we used theanswer to the question regarding indication for biopsy (hyperten-sion yes/no) to group patients.
HistologicPicture
We used a slightly modified version of the Japanese histologicgrading system for IgAN (grades 1-4)13 that we previously haveused in this study cohort.27 We stratified the cohort into patientswith IgAN with histologic grades 1-2 versus grades 3-4.
Statistical Analyses
To compare the mortality rate in patients with IgAN, we calcu-lated standardized mortality ratio (SMR). SMR is defined as theratio between the observed and expected numbers of deaths. Theexpected risk of death in the total Norwegian population is given insex-stratified tables from Statistics Norway cross-tabulated for theperiod of interest (1986-1990, 1991-1995, and so forth) and age(15-19, 20-24 years, and so forth). The expected number of deathsin our study was calculated by multiplying the number of person-years in the relevant 5-year period and 5-year age group with theexpected death risk given in the table from Statistics Norway. APoisson distribution for death incidence in the general populationwas assumed when calculating the 95% confidence interval (CI).
First, SMRs were calculated for all patients with IgAN. Second,SMRs were calculated and compared in low-, moderate-, andhigh-risk groups using all methods for risk stratification described.It was evident that mortality risk was associated closely with riskof ESRD, and in further analyses, we calculated SMRs for theperiods before and after reaching ESRD and separately in patientsreceiving dialysis and transplant recipients. In transplant recipi-ents, we calculated SMR separately in patients with and without
transplant loss.
Am J Kidney Dis. 2013;xx(x):xxx
Mortality in IgAN Patients
Kaplan-Meier statistics were used to calculate cumulative riskof ESRD in different prognostic groups of patients with IgAN.Kaplan Meier curves were plotted for risk of death in patients withIgAN compared to the expected number in the age-, period-, andsex-matched general Norwegian population. Log-rank test wasused to test statistical significance. SMRs are given as estimatewith 95% CI, and mean values are given as mean � standarddeviation.
RESULTS
A cohort of 633 patients with IgAN was identified.Mean duration of follow-up was 11.8 (range, 0.0-20.8) years, with a total of 7,464 person-years. Meanage at the start of observation was 39 � 16 years and74% of patients were men. Further cohort characteris-tics are listed in Table 1.
Table 1. Cohort Character
Variable All (N � 633)
Age (y) 39.0 � 16
Male sex 467 (74)
Proteinuria category�1 g/24 h 300 (47)1-2.9 g/24 h 194 (31)�3 g/24 h 139 (22)
Hypertension 283 (45)
Aggressive morphologic pictureb 362 (57)
Diabetes mellitus 10 (1.6)
Note: Values for categorical variables are given as numberdeviation.
Abbreviations: eGFR, estimated glomerular filtration rate; IgANaeGFR calculated by the Chronic Kidney Disease Epidemiol
creatinine, so creatinine level reduced by 5%28).bAggressive morphologic picture: grade 3 or 4 according to a s
IgAN (grades 1-4).13,27
IgAN pa�ents = 633
Not ESRD = 487
Alive = 442 Dead = 45
-CVD = 17
-Cancer = 9
-Infec�on = 6
-Other = 13
ESRD, dialysis
Alive = 11
Figure 1. Flow chart shows outcomes of the 633 patients with
As shown in Fig 1, 80 patients died during follow-up. Forty-five patients died before and 35 died afterreaching ESRD. During follow-up, 146 patients pro-gressed to ESRD; of these, 119 patients received akidney transplant and death-censored transplant lossoccurred in 14 patients.
Causes of deaths were vascular disease in 36 (45%)patients, malignant disease in 13 (16%), infectiousdisease in 15 (19%), and other causes in 16 (20%).Causes of death are given separately for patients withIgAN dying pre-ESRD, while being treated with dialy-sis, or after transplantation in Fig 1.
As also shown in Fig 2 and Table 2, 20-yearcumulative risks of ESRD in low-risk groups were
entage); values for continuous variables, as mean � standard
unoglobulin A nephropathy.ollaboration creatinine equation (with nonstandardized serum
y modified version of the Japanese histologic grading system for
ESRD = 146
= 27
ad = 18
D = 12
cer = 2
c�on = 4
ESRD, transplanted = 119
Alive = 102 Dead = 17
-CVD = 7
-Cancer = 2
-Infec�on = 5
-Other = 3
unoglobulin A nephropathy (IgAN) in study cohort. Abbreviations:
istics
>
(perc
, immogy C
lightl
only
De
-CV
-Can
-Infe
imm
3
Knoop et al
14.7% (eGFR �60 mL/min/1.73 m2), 6.6% (protein-uria with protein excretion �1 g/24 h), and 6.5%(Japanese prognostic model). In moderate-risk groups,20-year cumulative risks of ESRD were 68.4% (eGFR
Figure 2. Kaplan-Meier plots of cumulative renal survival in633 patients with immunoglobulin A nephropathy (IgAN), withrisk stratification based on (A) estimated glomerular filtration rate(eGFR), (B) degree of proteinuria, and (C) Japanese prognosticmodel to predict 10-year risk of end-stage renal disease (ESRD)in patients with IgAN (low risk, �10%; moderate, 10%-50%; andhigh, �50%).13,27 Death before ESRD is treated as a censoringevent in these analyses.
of 30-60 mL/min/1.73 m2), 52.9% (proteinuria with
4
protein excretion of 1-2.9 g/24 h), and 48.7% (Japa-nese prognostic model), and in high-risk groups, riskswere 76.5% (eGFR �30 mL/min/1.73 m2; few pa-tients with �10 years of observation), 57.6% (protein-uria with protein excretion �3 g/24 h), and 80.2%(Japanese prognostic model).
Results from SMR analyses of the total IgANcohort and different risk groups are shown in Fig 3(Japanese prognostic model) and Table 3 (all prognos-tic models). The SMR in the total cohort during theobservation period was 1.9 (95% CI, 1.5-2.4). Interest-ingly, low-risk patients did not show an increasedSMR. In moderate-risk patients, the SMR was in-creased significantly to 1.5, and in high-risk patients,the SMR was 3.1.
The SMR in the pre-ESRD observation period wasnot significantly increased: 1.3 (95% CI, 1.0-1.7) forthe entire cohort. Stratification showed that pre-ESRDSMRs in the low-risk group (all methods) were closeto 1.0; in the moderate-risk group, SMR was signifi-cantly increased (1.8) in only the eGFR risk model. Inthe high-risk group, pre-ESRD SMRs were increasedsignificantly to approximately 2 in all prognosticmodels.
SMRs were 4.9 (3.5-7) in patients during RRT, 10.3in dialysis patients, and 3.2 in transplant recipients.
The SMR in the post-ESRD period in the 14transplant recipients with transplant loss was 8.1, andSMR in the 105 patients without transplant loss was2.5.
DISCUSSION
The major new finding in the present study is thatthe age- and sex-adjusted mortality rate in Norwegianpatients with IgAN is approximately twice that of thegeneral Norwegian population. Of note, an increasedmortality rate is observed in only moderate- andhigh-risk patients with IgAN. Furthermore, SMR isincreased significantly in the pre-ESRD period in onlyhigh-risk patients with IgAN and patients with initialeGFR �60 mL/min/1.73 m2. A strongly increasedmortality rate (SMR, 4.9) is observed in patients withIgAN during RRT. As expected, in patients receivingRRT, a much higher mortality rate was observed indialysis patients compared with transplant recipients(SMRs, 10.3 vs 3.2). In transplant recipients, trans-plant loss was associated with a substantially in-creased death risk (SMRs, 8.3 vs 2.5). Another impor-tant finding in our study is that the mortality rate inlow-risk patients with IgAN, representing �50% ofthe cohort, equals that of the general population.
A surprising finding in the present IgAN cohort wasthat ESRD occurred 3.2 times more frequently thanpre-ESRD deaths; this is strikingly different from
findings in many other CKD studies.15-20 However,
Am J Kidney Dis. 2013;xx(x):xxx
bNo patients with eGFR �30 mL/min/1.73 m2 reached 20 years of follow up.
Mortality in IgAN Patients
Figure 3. Kaplan-Meier plots of expected versus observed mortality in the immunoglobulin A nephropathy (IgAN) study cohort. (A)All patients, (B) low-risk (�10% risk of end-stage renal disease in 10 years) patients, (C) moderate-risk (10%-50%) patients, and (D)
Table 2. Cumulative Risk of ESRD in All IgAN Patients and After Risk Stratification
No. of Patients Cumulative ESRD Risk
All With ESRD With Death Pre-ESRD 10 ya 15 ya 20 ya
Note: Risk stratification by initial eGFR, proteinuria, and Japanese prognostic model to predict 10-year risk of ESRD in patients withIgAN.13,27
Abbreviations and definitions; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; IgAN, immunoglobulin Anephropathy; y, years after kidney biopsy.
aDeath before ESRD is treated as a censoring event in these analyses.
high-risk (�50%) patients. Risk stratification based on Japanese prognostic model for patients with IgAN.13,27
Am J Kidney Dis. 2013;xx(x):xxx 5
Table 3. Standardized Mortality Ratios in Total, Pre-ESRD, and ESRD Observation Periods
No. ofPatients Total Period Pre-ESRD Period RRT Period
Note: Data for all patients with IgAN and after risk stratification by initial eGFR, proteinuria, and Japanese prognostic model to predict 10-year risk of ESRD in patients with IgAN.13,27
Abbreviations and definitions: CI, confidence interval; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; IgAN, immunoglobulin A nephropathy; pre-ESRD deaths, deaths in patients notestablished in RRT; RRT, renal replacement therapy; RRT deaths, deaths in patients established in RRT; SMR, standardized mortality ratio.A
mJ
Kidney
Dis.2013;xx(x):xxx
6
Knoop
etal
Mortality in IgAN Patients
this observation agrees with a previous study of pa-tients with CKD with primary kidney disease.22 Inprevious CKD studies, a particularly high rate ofdeaths due to cardiovascular disease has been re-ported.16 Also in our study, cardiovascular diseasewas the most common cause of death. However, onlyin the group of patients with IgAN treated withdialysis did cardiovascular disease cause �50% ofdeaths. Preexisting cardiovascular disease in somepatients might explain why kidney transplantationwas not performed.
In our IgAN cohort, a high relative mortality ratewas observed in dialysis patients (SMR, 10.3) andafter transplant loss (SMR, 8.1). This is in line withobservations in other cohorts.24,25,29 We also foundthat SMR in high-risk pre-ESRD patients was moder-ately increased, but lower than reported in most otherCKD studies.15-20 In patients with IgAN with function-ing kidney transplants, SMR is moderately increasedand seemingly rather favorable, but we lack data tocompare with patients undergoing transplantation dueto other diseases. Together with the findings by Menonet al,22 our observations clearly demonstrate that theoverall mortality of patients with IgAN differs substan-tially from that of most other groups of patients withCKD. Thus, data from such studies should not beextrapolated to patients with IgAN.
Our findings provide new information of potentialclinical and prognostic relevance for patients withIgAN. First, our results have implications for theselection of end points when investigating prognosticmodels of IgAN. Most prognostic models have re-ported only renal end points. One recent high-qualitystudy used ESRD or pre-ESRD death as a combinedend point.7,8 Our data, particularly the observationthat SMR is not increased in low-risk patients, indi-cate that pre-ESRD deaths and ESRD perhaps shouldbe studied as separate end points in prognostic modelsof IgAN. Second, it is a clinical experience thatlow-risk patients with IgAN have been denied accessto life insurance, probably due to their general statusas CKD patients. Our results demonstrate that this isnot appropriate and that patients with IgAN should beassessed on an individual basis from an insuranceperspective.
The major strengths of the present study are that thecohort is population based and the study group islarge, with long follow-up and many end points,including ESRD and deaths. Furthermore, both identi-fication of patients with IgAN and their end points arebased on high-quality national registries. Age, sex,and time periods are important confounders in allmortality analyses and must be adjusted for appropri-ately. The method used in the present report, SMR,
adequately adjusts for all these confounders.
Am J Kidney Dis. 2013;xx(x):xxx
An important weakness of our study is that thepre-ESRD treatment of our patients with IgAN is notknown because such information has not been regis-tered in the Kidney Biopsy Registry. However, wepreviously have compared and found that renal prog-noses of our patients, after adjustment for risk factors,are similar to those of Japanese patients with IgAN.27
All residents of Norway are covered by the nationalhealth insurance and have good access to health care.Although nephrology practice has changed over thepast decades, it is a fair assumption to state that ourpatients in general have received adequate medicaltreatment in line with recommendations at the time.Another weakness of the present study is lack of dataregarding important comorbid conditions other thanhypertension, such as manifestations of cardiovascu-lar disease and malignancies. Such data have not beenregistered in the Kidney Biopsy Registry and couldhave contributed as explanatory factors to our find-ings regarding mortality in patients with IgAN.
In summary, we have shown that overall SMR ismoderately increased (1.9) in patients with IgAN.Low-risk patients with IgAN have an SMR similar tothat of the general population, whereas a high mortal-ity rate is observed in patients receiving dialysis andafter loss of kidney transplant function. A moderatelyincreased mortality rate is observed in high-risk pre-ESRD and transplant-recipient patients with IgAN.Compared with patients with CKD in general, patientswith IgAN have a favorable prognosis and the prob-ability of reaching ESRD is substantially higher thanthe risk of pre-ESRD death.
ACKNOWLEDGEMENTSSupport: None.Financial Disclosure: The authors declare that they have no
relevant financial interests.
REFERENCES1. D’Amico G. The commonest glomerulonephritis in the world:
IgA nephropathy. Q J Med. 1987;64(245):709-727.2. Floege J, Feehally J. IgA nephropathy: recent developments.
J Am Soc Nephrol. 2000;11(12):2395-2403.3. Levy M, Berger J. Worldwide perspective of IgA nephropa-
thy. Am J Kidney Dis. 1988;12(5):340-347.4. D’Amico G. Natural history of idiopathic IgA nephropathy:
role of clinical and histological prognostic factors. Am J KidneyDis. 2000;36(2):227-237.
5. Geddes CC, Rauta V, Gronhagen-Riska C, et al. A tricontinen-tal view of IgA nephropathy. Nephrol Dial Transplant. 2003;18(8):1541-1548.
6. Alamartine E, Sabatier JC, Guerin C, Berliet JM, Berthoux F.Prognostic factors in mesangial IgA glomerulonephritis: an exten-sive study with univariate and multivariate analyses. Am J KidneyDis. 1991;18(1):12-19.
7. Barbour SJ, Reich HN. Risk stratification of patients with
IgA nephropathy. Am J Kidney Dis. 2012;59(6):865-873.
7
Knoop et al
8. Berthoux F, Mohey H, Laurent B, Mariat C, Afiani A,Thibaudin L. Predicting the risk for dialysis or death in IgAnephropathy. J Am Soc Nephrol. 2011;22(4):752-761.
9. Cattran DC, Coppo R, Cook HT, et al. The Oxford classifica-tion of IgA nephropathy: rationale, clinicopathological correla-tions, and classification. Kidney Int. 2009;76(5):534-545.
10. Chacko B, John GT, Neelakantan N, et al. Presentation,prognosis and outcome of IgA nephropathy in Indian adults.Nephrology (Carlton). 2005;10(5):496-503.
11. D’Amico G. Influence of clinical and histological featureson actuarial renal survival in adult patients with idiopathic IgAnephropathy, membranous nephropathy, and membranoprolifera-tive glomerulonephritis: survey of the recent literature. Am JKidney Dis. 1992;20(4):315-323.
12. Donadio JV, Grande JP. IgA nephropathy. N Engl J Med.2002;347(10):738-748.
13. Goto M, Wakai K, Kawamura T, Ando M, Endoh M,Tomino Y. A scoring system to predict renal outcome in IgAnephropathy: a nationwide 10-year prospective cohort study. Neph-rol Dial Transplant. 2009;24(10):3068-3074.
14. Roberts IS, Cook HT, Troyanov S, et al. The Oxfordclassification of IgA nephropathy: pathology definitions, correla-tions, and reproducibility. Kidney Int. 2009;76(5):546-556.
15. Foley RN, Murray AM, Li S, et al. Chronic kidney diseaseand the risk for cardiovascular disease, renal replacement, anddeath in the United States Medicare population, 1998 to 1999.J Am Soc Nephrol. 2005;16(2):489-495.
16. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY.Chronic kidney disease and the risks of death, cardiovascularevents, and hospitalization. N Engl J Med. 2004;351(13):1296-1305.
17. Johnson ES, Thorp ML, Yang X, Charansonney OL, SmithDH. Predicting renal replacement therapy and mortality in CKD.Am J Kidney Dis. 2007;50(4):559-565.
18. Keith DS, Nichols GA, Gullion CM, Brown JB, Smith DH.Longitudinal follow-up and outcomes among a population withchronic kidney disease in a large managed care organization. Arch
Intern Med. 2004;164(6):659-663.
8
19. Patel UD, Young EW, Ojo AO, Hayward RA. CKD progres-sion and mortality among older patients with diabetes. Am JKidney Dis. 2005;46(3):406-414.
20. Rahman M, Pressel S, Davis BR, et al. Cardiovascularoutcomes in high-risk hypertensive patients stratified by baselineglomerular filtration rate. Ann Intern Med. 2006;144(3):172-180.
21. Forsblom C, Harjutsalo V, Thorn LM, et al. Competing-riskanalysis of ESRD and death among patients with type 1 diabetesand macroalbuminuria. J Am Soc Nephrol. 2011;22(3):537-544.
22. Menon V, Wang X, Sarnak MJ, et al. Long-term outcomesin nondiabetic chronic kidney disease. Kidney Int. 2008;73(11):1310-1315.
23. Rosolowsky ET, Skupien J, Smiles AM, et al. Risk forESRD in type 1 diabetes remains high despite renoprotection. J AmSoc Nephrol. 2011;22(3):545-553.
24. Wolfe RA, Ashby VB, Milford EL, et al. Comparison ofmortality in all patients on dialysis, patients on dialysis awaitingtransplantation, and recipients of a first cadaveric transplant. N EnglJ Med. 1999;341(23):1725-1730.
25. Arend SM, Mallat MJ, Westendorp RJ, van der Woude FJ,van Es LA. Patient survival after renal transplantation; more than25 years follow-up. Nephrol Dial Transplant. 1997;12(8):1672-1679.
26. Churg J. IgA nephropathy. In: Glassock R, ed. Renal Dis-ease—Classification and Atlas of Glomerular Diseases. 2nd ed.New York, NY: Igaku-Shoin; 1995:181-202.
27. Bjorneklett R, Vikse BE, Bostad L, Leivestad T, IversenBM. Long-term risk of ESRD in IgAN; validation of Japaneseprognostic model in a Norwegian cohort. Nephrol Dial Transplant.2012;27(4):1485-1491.
28. Skali H, Uno H, Levey AS, Inker LA, Pfeffer MA, SolomonSD. Prognostic assessment of estimated glomerular filtration rateby the new Chronic Kidney Disease Epidemiology Collaborationequation in comparison with the Modification of Diet in RenalDisease Study equation. Am Heart J. 2011;162(3):548-554.
29. Ojo A, Wolfe RA, Agodoa LY, et al. Prognosis after primaryrenal transplant failure and the beneficial effects of repeat transplan-tation: multivariate analyses from the United States Renal Data
![Crescentic IgA Nephropathy[1][1]](https://static.documents.pub/doc/80x56/55cf9903550346d0339b030a/crescentic-iga-nephropathy11.jpg)
