glomeruloneritis

Copyright @ Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Acute Postinfectious Glomerulonephritisin the Modern Era

Experience With 86 Adults and Review of the LiteratureSamih H. Nasr, MD, Glen S. Markowitz, MD, Michael B. Stokes, MD, Samar M. Said, MD,

Anthony M. Valeri, MD, and Vivette D. D’Agati, MD

Abstract: Acute postinfectious glomerulonephritis (APIGN) is

uncommon in adults, and its incidence is progressively declining in

developed countries. To our knowledge there are no modern North

American series addressing epidemiology and outcome. Here we

report the clinical and pathologic findings in 86 cases of adult

APIGN diagnosed by renal biopsy in a large New York referral

center between 1995 and 2005. The male:female ratio was 2:1, and

mean age was 56 years, with 33.7% aged older than 64 years. Of the

patients, 38.4% had an immunocompromised background, including

diabetes (29.1%), malignancy (4.7%), alcoholism (2.3%), acquired

immunodeficiency syndrome (AIDS) (2.3%), and intravenous drug

use (1.2%). The most common sites of infection were upper

respiratory tract (23.3%), skin (17.4%), lung (17.4%), and heart/

endocarditis (11.6%). The 2 most frequently identified infectious

agents were streptococcus (27.9%) and staphylococcus (24.4%).

Hypocomplementemia was present in 73.9% of patients. The most

common histologic patterns were diffuse (72.1%), focal (12.8%),

and mesangial (8.1%) proliferative glomerulonephritis.Outcome analysis was performed on the 52 patients with a

follow-up of �3 months (mean, 25 mo). Among the 41 patients

without underlying diabetic glomerulosclerosis, 23 (56.1%)

achieved complete remission, 11 (26.8%) had persistent renal

dysfunction, and 7 (17.1%) progressed to end-stage renal disease

(ESRD). Of the 11 patients with underlying diabetic glomerulo-

sclerosis, 2 (18.2%) had persistent renal dysfunction, and the

remaining 9 (81.8%) progressed to ESRD (p < 0.001). In patients

without underlying diabetic glomerulosclerosis, correlates of

complete remission were younger age, female sex, lower serum

creatinine at biopsy, and absence of immunocompromised state. By

multivariate analysis, age and serum creatinine at biopsy inversely

correlated with complete remission, and serum creatinine at biopsy

was the only correlate with ESRD. Outcome did not correlate with

any pathologic feature (including crescents) or steroid treatment.

Diabetes and aging have emerged as major risk factors for

adult APIGN. Full recovery of renal function can be expected in just

over half of patients, and prognosis is dismal in those with

underlying diabetic glomerulosclerosis.

(Medicine 2008;87:21–32)

INTRODUCTION

A cute postinfectious glomerulonephritis (APIGN) mostfrequently affects children and is relatively uncommon

in adults. The disease incidence has declined sharply inindustrialized countries over the last 50 years. In the past, thevast majority of cases followed streptococcal infections,typically involving the upper respiratory tract (URT) or skin.In recent years, non-streptococcal infections, particularlystaphylococcal and Gram-negative infections, have becomemore frequent precipitants of adult APIGN28. Adults withan immunocompromised background, especially due to alco-holism, are at higher risk for developing the disease21,28. Mostchildren with APIGN recover completely, but there is generalagreement that the prognosis is more guarded in adults. Whileseveral large studies performed before 1980 evaluated thecourse of nonepidemic adult APIGN4,18,22,23,26, there have beenfew reports in the modern era, and, to our knowledge, none hasaddressed the North American experience21,28,29. Furthermore,there are limited published data regarding the clinicopathologicfindings and outcome in the subset of adults with APIGN whohave underlying diabetic glomerulosclerosis (DGS), eventhough APIGN is one of the most common nondiabeticglomerular diseases to occur in diabetics16,25,27,30,50,56.

In the current report we retrospectively examined 86adult cases of APIGN diagnosed on renal biopsy at a largeNew York referral center from 1995 to 2005, including 16with histologic evidence of DGS. The demographic, clinical,pathologic, and outcome data define an evolving spectrum ofadult APIGN in the new millennium.

METHODSAll native renal biopsies received and processed at

Columbia University Medical Center from 1995 to 2005

Abbreviations: APIGN = acute postinfectious glomerulonephritis, C =serum complement, DGS = diabetic glomerulosclerosis, EM = electronmicroscopy, ESRD = end-stage renal disease, GN = glomerulonephritis,H & E = hematoxylin and eosin stain, IF = immunofluorescence,IVDU = intravenous drug use, LM = light microscopy, MPGN =membranoproliferative glomerulonephritis, URT = upper respiratory tract.

Medicine � Volume 87, Number 1, January 2008 21

From Department of Pathology (SHN, GSM, MBS, SMS, VDD) andDepartment of Medicine, Division of Nephrology (AMV), ColumbiaUniversity, College of Physicians & Surgeons, New York, New York.

This study was presented orally as an abstract at the annual meeting of theAmerican Society of Nephrology, October 31–November 5, 2007, SanFrancisco, CA.

Address reprint requests to: Samih H. Nasr, MD, Department of Pathology,Columbia University, College of Physicians & Surgeons, 630 West168th Street, VC14-224, New York, NY 10032. Fax: 212-342-5380;e-mail: [email protected].

Copyright n 2008 by Lippincott Williams & WilkinsISSN: 0025-7974/08/8701-0021DOI: 10.1097/MD.0b013e318161b0fc


(n = 14,138) were reviewed retrospectively for the diag-nosis of APIGN in adults (aged �16 yr of age). Ninety-twocases were identified, indicating a biopsy incidence of0.6%. The diagnosis of APIGN was established by clini-copathologic correlation. For the purpose of this study, atleast 3 of the following 5 criteria were required for studyentry:

1) clinical or laboratory evidence of infection precedingthe onset of glomerulonephritis (GN);

2) depressed serum complement;3) endocapillary proliferative and exudative GN on

light microscopy (LM);4) C3-dominant or codominant glomerular staining on

immunofluorescence (IF); and5) ‘‘hump-shaped’’ subepithelial deposits on electron

microscopy (EM).Six cases of presumed APIGN were excluded because

they fulfilled only 2 of these criteria (including 4 with typicalhumps on EM but no evidence of preceding infection and 1with a membranoproliferative glomerulonephritis (MPGN)pattern secondary to schistosomiasis). The remaining 86cases fulfilled at least 3 criteria and form the basis of thisreport. Among these, 32 cases fulfilled 5/5 criteria (37.2%),35 fulfilled 4/5 criteria (40.7%), and 19 fulfilled 3/5 criteria(22.1%).

Standard processing of renal biopsies included LM, IF,and EM. For LM, all cases were stained with hematoxylinand eosin, periodic acid-Schiff, Masson trichrome, and Jonesmethenamine silver. For IF, 3-mm cryostat sections werestained with polyclonal FITC-conjugated antibodies to IgG,IgM, IgA, C3, C1q, kappa, lambda, fibrinogen, and albumin(Dako Corp., Carpinteria, CA). EM was performed using aJEOL 100s electron microscope.

Clinical data, including demographic information,presenting clinical and laboratory findings, medical history,treatment, and follow-up, were obtained from referral formssubmitted at the time of biopsy and from follow-uptelephone interviews with the referring nephrologist. Thefollowing clinical definitions were used: nephrotic rangeproteinuria, �3.0 g/d; hypoalbuminemia, serum albumin<3.5 g/dL; renal insufficiency, serum creatinine >1.2 mg/dL;nephrotic syndrome, nephrotic range proteinuria, hypoalbu-minemia, and peripheral edema; and hypertension, systolicblood pressure >140 mmHg, diastolic blood pressure >90mmHg, or ongoing treatment with antihypertensive medi-cations. For outcome analysis, complete remission wasdefined as normalization of serum creatinine to baselinelevels or to a creatinine �1.2 mg/dL (for those patients inwhom baseline creatinines were unavailable); persistentrenal dysfunction was defined by elevation of serumcreatinine 0.2 mg/dL above baseline levels or follow-upcreatinine >1.2 mg/dL (for those in whom baseline levelswere unavailable); and end-stage renal disease (ESRD) wasdefined as requiring renal replacement therapy.

Continuous variables are reported as mean ± SD.Statistical analysis was performed using SPSS for Windows,version 15.0 (SPSS, Chicago, IL). Analysis was performedusing exact statistical methods. Univariate analysis was

performed using the Mann-Whitney U test, the Kruskal-Wallis test, and the Fischer exact test as appropriate forvariable type. Multivariate analysis was performed usingbinary and multinomial logistic regression analysis. Statis-tical significance was assumed at p < 0.05.

The study was approved by the Institutional ReviewBoard of Columbia University Medical Center.

RESULTS

Clinical Features of Adult APIGNThe majority of patients were white (59.3%) and male

(66.3%) (Table 1). The mean age was 56 years, and one-thirdof patients were older than 64 years. Of the patients, 38.4%had an immunocompromised background, and the mostfrequent predisposing factor for infection was diabetes(present in 29.1% of patients). The site of infection wasidentified in 83.7% of patients; the 4 most common siteswere URT (in 23.3% of patients), skin (17.4%), lung(17.4%), and heart/endocarditis (11.6%) (Table 2). Theinfectious agent was identified in 58.1% of patients (bypositive culture in the case of nonstreptococcal infection andby positive culture, elevated anti-streptolysin O antibody, oranti-DNase B antibody in the case of streptococcal infection)(see Table 2). Streptococcus and staphylococcus were by farthe 2 most frequent causative agents, found in 27.9% and24.4% of patients, respectively. URT was the most frequentsite of streptococcal infection (46%), while the skin was themost common site of staphylococcal infection (38%). Two-thirds of patients with staphylococcal infections werediabetic compared with only 8% of patients with strepto-coccal infections. Some patients with clinically evident infec-tion but negative cultures had been treated with antibioticsbefore culture. The mean time from clinical onset of infectionto renal disease was 3 weeks (including 4 wk for URTinfection, 3 wk for skin and lung infections, and 2 wk forendocarditis). In 7 patients (8.1%), comprising 4 with pneu-monia, 2 with endocarditis, and 1 with urinary infection, theonset of GN coincided with the first clinical recognition ofinfection. The time from clinically apparent onset of renaldisease to biopsy was �2 months in 62 of 67 patients (92.5%)with available data.

Clinical Renal Characteristics at Presentation in70 Patients Without Underlying DGS

Twenty patients (28.6%) had new-onset hypertension,and 22 additional patients (31.4%) had longstandinghypertension at the time of biopsy (Table 3). Peripheraledema was present in approximately half of the patients.The mean 24 h urine protein was 3.6 g, and 40.7% ofpatients had nephrotic range proteinuria. Nephrotic syn-drome was present in 29.1% of patients. Hematuria waspresent in 89.6% of patients, including macroscopichematuria in 23.9%. Red blood cell casts were reportedin only 14.9% of patients. Leukocyturia was present in themajority of patients. The mean peak serum creatinine was3.96 mg/dL, while the mean serum creatinine at biopsy was3.89 mg/dL. Seven patients (10%) had a peak serum

22 n 2008 Lippincott Williams & Wilkins

Nasr et al Medicine � Volume 87, Number 1, January 2008


creatinine in the normal range, 6 of whom had resolvingAPIGN on biopsy (3 mesangial proliferative GN and 3focal endocapillary proliferative GN). More than two-thirdsof patients had serum creatinine >2 mg/dL at biopsy.Baseline serum creatinine was elevated in only 6 of the 31patients with available data. Testing for serum complement(C) was performed in 58 patients, 63.8% of whom had adepressed C3, 48.3% had a depressed C4, and 67.2% had adepressed C3 or C4.

Clinical Renal Characteristics at Presentation in16 Patients With Underlying DGS

Compared to APIGN patients without underlyingDGS, the subgroup of APIGN patients who had underlyingDGS on biopsy exhibited a greater degree of proteinuria(mean, 4.1 g/24 h), and a higher serum creatinine at baseline(mean, 2.2 mg/dL) and at the time of renal biopsy (mean,6.44 mg/dL) (see Table 3). All patients had microhematuria,while macrohematuria was seen in only 1 patient (6.3%). C3was depressed in all patients, whereas C4 was low in 53.3%of patients.

Pathologic FindingsThe most frequent histologic pattern of glomerular

injury on LM was diffuse endocapillary proliferative andexudative GN (Figure 1), found in 72.1% of patients, fol-lowed by focal endocapillary proliferative and exudative GN(12.8% of patients), and mesangial proliferative GN (8.1%)(Table 4). Only 2 patients had an MPGN pattern, including1 with an infected ventriculoperitoneal shunt and 1 withstaphylococcal skin infection. Diffuse necrotizing and cres-centic GN (defined by crescents or necrosis involving �50%of glomeruli, according to the World Health Organization9)was seen in 4 patients. Two of these patients had strepto-coccal endocarditis, of which 1 had C-ANCA seropositivity;1 had streptococcal URT infection; and 1 had staphylococcalskin infection.

LM Findings in 70 Patients WithoutUnderlying DGS

The mean glomerular count was 18 (range, 5–46) andthe mean percentage of sclerotic glomeruli was 9.9%(Table 5). Glomerular neutrophil infiltration was seen in90% of cases. Cellular crescents were present in approxi-mately one-third of cases but affected �20% of glomeruli inonly 14.3% of cases. Fibrous crescents, on the other hand,were less frequent, seen in 12.9% of cases, and affected�20% of glomeruli in only 2 cases. Segmental glomerularnecrosis, characterized by intracapillary fibrin deposition,rupture of the glomerular basement membrane, and fibrinextravasation into the Bowman space, was encountered in

TABLE 2. Site of Infection and Infectious Agent (86 Patients)

Site of Infection No. of Patients (%) Infectious Agent No. of Patients (%)

Upper respiratory tract 20 (23.3) Streptococcus 24 (27.9)

Skin 15 (17.4) Staphylococcus 21 (24.4)

Pneumonia 15 (17.4) Pneumococcus 1 (1.2)

Endocarditis 10 (11.6) Pseudomonas 1 (1.2)

Osteomyelitis 4 (4.7) Enterococcus 1 (1.2)

Urinary tract infection 4 (4.7) Propionibacterium acnes 1 (1.2)

Deep-seated abscess 2 (2.3) Candida 1 (1.2)

Ventriculoperitoneal shunt 1 (1.2) Unknown 36 (41.9)

Phlebitis 1 (1.2)

No clinical evidence of infection 14 (16.3)

TABLE 1. Demographics and Predisposing Factors ForInfection (86 Patients)

No. of Patients (%)

Male:female 57:29 (66.3:33.7)

Age, yr (mean ± SD) 56 ± 16

15–24 3 (3.5)

25–34 8 (9.3)

35–44 12 (14.0)

45–54 16 (18.6)

55–64 18 (20.9)

Over 64 29 (33.7)

Race

White 51 (59.3)

Hispanic 13 (15.1)

Black 9 (10.5)

Asian 3 (3.5)

Undisclosed 10 (11.6)

Predisposing factors for infection 33 (38.4)

Diabetes mellitus 25 (29.1)

Malignancy 4 (4.7)

Immunosuppressive therapy 3 (3.5) (2 had DM,1 had malignancy)

Alcoholism 2 (2.3) (1 had DM)

Cirrhosis 2 (2.3) (both had DM)

Acquired immunodeficiencysyndrome (AIDS)

2 (2.3)

Intravenous drug use 1 (1.2) (also had DM)

Severe malnutrition 1 (1.2)

n 2008 Lippincott Williams & Wilkins 23

Medicine � Volume 87, Number 1, January 2008 Acute Postinfectious Glomerulonephritis in Adults


21.4% of cases. Interstitial inflammation with a mixedinfiltrate of lymphocytes, monocytes, plasma cells, and rareneutrophils was present in 64 cases (91.4%) and was typicallyfocal. In 3 cases (4.3%) the interstitial neutrophils were nu-merous, and 5 cases (7.1%) had neutrophil casts. Second-ary acute tubular injury with luminal ectasia, epithelialsimplification, loss of brush border, and nuclear enlargementwas a common feature, identified in 60% of cases. The degreeof tubular atrophy and interstitial fibrosis ranged from absentto mild in the majority of cases. Focal septic vasculitis waspresent in 2 cases (2.9%). Arteriosclerosis ranged from absent(24.2% of cases) to mild (40% of cases) to moderate (32.9%of cases) to severe (2.9% of cases).

LM Findings in 16 Patients WithUnderlying DGS

Sampling for LM included a mean of 18 glomeruli(range, 3–81 glomeruli), and a mean of 21.8% of glomeruliwere globally sclerotic (Table 6). Glomerular neutrophilinfiltration was present in all cases and was typically diffuse.Cellular crescents and necrosis were seen in one-quarter ofcases, whereas fibrous crescents were not identified. Allcases showed mesangial sclerosis and thickening of glomer-ular and tubular basement membranes, typical of DGS. Themesangial sclerosis was nodular in 12 cases (75%) anddiffuse in 4 cases (25%) (Figure 2). The degree of tubularatrophy and interstitial fibrosis ranged from mild (37.5% ofcases) to moderate (43.8%) to severe (18.8%). Arterio-sclerosis and arteriolar hyalinosis were present in all cases

and ranged from mild (1 case) to moderate (11 cases) tosevere (4 cases).

IF FindingsStandard IF on frozen tissue was performed in 83

biopsies with available glomeruli (Table 7). In 1 of the 3

TABLE 3. Clinical Characteristics at Presentation (86 Patients)

APIGN Patients Without DGS

No. (%) (n = 70)

APIGN Patients With DGS

No. (%) (n = 16)

New-onset hypertension 20/70 (28.6) 2/16 (12.5)

Long-standing hypertension 22/70 (31.4) 8/16 (50.0)

Peripheral edema 37/70 (52.9) 9/15 (60.0)

Mean 24 h urine protein, g 3.6 4.1

Proteinuria <1 g/24 h 13/59 (22.0) 4/13 (30.8)

Proteinuria 1–3 g/24 h 22/59 (37.1) 4/13 (30.8)

Proteinuria > 3 g/24 h 24/59 (40.7) 5/13 (38.5)

Full nephrotic syndrome 16/55 (29.1) 3/13 (23.1)

Hematuria (microscopic or macroscopic) 60/67 (89.6) 15/15 (100)

Macroscopic hematuria 16/67 (23.9) 1/16 (6.3)

RBC casts 10/67 (14.9) 3/14 (21.4)

Leukocyturia 37/67 (55.2) 10/14 (71.4)

Mean baseline serum creatinine, mg/dL (range) 2.2 (0.9–4.4)

Mean serum creatinine at biopsy, mg/dL 3.89 6.44

Creatinine �1.2 mg/dL 7/70 (10.0) 0/16

Creatinine 1.2–2 mg/dL 15/70 (21.4) 0/16

Creatinine > 2 mg/dL 48/70 (68.6) 16/16 (100)

Low C3 37/58 (63.8) 15/15 (100)

Low C4 28/58 (48.3) 8/15 (53.3)

Low C3 or C4 39/58 (67.2) 15/15 (100)

FIGURE 1. The most common histologic pattern of adultAPIGN is a diffuse and global endocapillary proliferative glo-merulonephritis. Theglomerular capillaries are globally occludedby numerous infiltrating neutrophils, as well as proliferation ofglomerular endothelial and mesangial cells (H & E, � 400).




remaining biopsies lacking glomeruli on frozen tissue, IFwas performed on pronase-digested paraffin-embeddedtissue, a less sensitive technique for detecting C3 tissue de-position31. The most common staining pattern on IF, seenin 82% of cases, was granular mesangial and glomerularcapillary wall, resembling the ‘‘starry sky pattern’’ describedby Sorger et al51 (Figure 3). In 7 cases (8.4%), the stainingwas predominantly glomerular capillary wall ‘‘garlandpattern.’’ In the remaining 8 cases (9.6%), only mesangialpositivity was detected, ‘‘mesangial pattern.’’ Of the 8 caseswith mesangial pattern, 3 had features of resolving APIGNon LM (2 mesangial proliferative GN and 1 focalproliferative GN), 1 showed necrotizing and crescentic GN,and 4 showed diffuse proliferative GN including 3 withunderlying DGS. C3 was positive in all 84 cases (100%) andwas the dominant or codominant immune reactant detectedin glomeruli in all 83 cases in which IF was performed onfrozen tissue, with a mean staining intensity of 2.4+. Themost common immunoglobulin deposited in glomeruli wasIgG (mean intensity when positive, 1.3+) present in 65.5% ofcases. IgM staining was less frequent (44% of cases) and lessintense (mean intensity when positive, 1.0+). In 8 cases(9.6%), IgM staining was stronger than IgG and IgA, 3 ofwhich were associated with URT infection, 3 with endocar-ditis, and 2 with unknown infection sites. Glomerular IgA

staining was detected in 44% of cases with a mean intensitywhen positive of 1.3+. In 8 cases (9.6%), IgA was thedominant immunoglobulin deposited in glomeruli (intensity�1+), 7 of which were from patients with staphylococcalinfection and underlying DGS. The remaining case was froma nondiabetic subject with an undetermined pathogen. C1qdeposition was detected in approximately one-third of casesbut was of low intensity (mean, 0.8+). Staining for kappa andlambda was similar with regard to frequency and intensity.Diabetic patients with or without nephropathy commonlyshowed weak linear staining of glomerular and tubularbasement membranes for IgG and albumin.

EM FindingsEM was performed on all 83 biopsies with available

glomeruli (Table 8). Subepithelial electron dense depositswere present in 94% of cases, and typically exhibited ahump-shaped appearance (Figure 4). Subepithelial humpswere also seen with the trichrome-stained LM sections in 2of the 3 biopsies lacking glomeruli for EM. In patients withunderlying DGS, the subepithelial deposits were small andsparse. In cases of resolving APIGN, the subepithelialdeposits were preferentially located at the glomerular base-ment membrane reflection over the mesangium (that is, themesangial waist). Subendothelial deposits were typicallysmall and infrequent, but were nonetheless detectable inthree-quarters of cases. Mesangial deposits were found in90.4% of cases and were abundant in 55.4% of cases. Bydefinition, patients with underlying DGS displayed mesan-gial sclerosis and thickening of the glomerular and tubularbasement membranes. The mean degree of foot processeffacement was 60% (range, 15%–100%).

OutcomeClinical follow-up was available in 66 patients (76.7%)

(Table 9). The duration of follow-up was <1 month in8 patients (12.1%), 1 to <3 months in 10 patients (15.1%),and �3 months in 48 patients (72.7%). The mean duration of

TABLE 4. Histologic Pattern on LM (86 Patients)

No. of Patients (%)

Mesangial proliferative GN 7 (8.1)

Focal proliferative GN 11 (12.8)

Diffuse proliferative GN 62 (72.1)

MPGN 2 (2.3)

Crescentic and necrotizing GN* 4 (4.6)

*Defined by the presence of crescents and/or necrosis affecting �50% ofglomeruli.

TABLE 5. LM Findings of 70 APIGN Patients Without Underlying DGS

Pathologic Findings No. of Cases %

No. of glomeruli, mean (SE) 18 (1.3)

No. of sclerotic glomeruli, mean (SE) 2 (0.5)

Sclerotic glomeruli, % (SE) 9.9 (1.6)

No. of cases with glomerular neutrophil infiltration (focal/diffuse) 63 (21/42) 90.0 (33.3/66.7)

No. of cases with cellular crescents (<20% of glomeruli, �20% of glomeruli) 26 (16, 10) 37.1 (61.5, 38.5)

No. of cases with fibrous crescents (<20% of glomeruli, �20% of glomeruli) 9 (7, 2) 12.9 (77.8, 22.2)

No. of cases with necrosis (<20% of glomeruli, �20% of glomeruli) 15 (8, 7) 21.4 (53.3, 46.7)

Interstitial inflammation: none/focal/diffuse* 6/58/6 8.6/82.9/8.6

Acute tubular injury: none/focal/diffuse* 28/27/15 40.0/38.6/21.4

Tubular atrophy and interstitial fibrosis: none/mild/moderate/severey 30/33/5/2 42.9/47.1/7.1/2.9

Arteriosclerosis and arteriolar hyalinosis: none/mild/moderate/severe 17/28/23/2 24.2/40.0/32.9/2.9

Abbreviation: SE = standard error.*Definitions: focal, �50% of cortical surface area; diffuse, >50%.yDefinitions: mild, 0%–25% of cortical surface area; moderate, 26%–50%; severe, >50%.




follow-up for all 66 patients was 19 months. Of the 18patients with <3 months of follow-up, 4 achieved completeremission, 8 had persistent renal dysfunction (1 of whomdied of septicemia), and 6 remained dialysis dependent,including 5 patients who died (3 due to septicemia andmultiorgan failure and 2 of undetermined causes).

For the purpose of outcome analysis, only patients witha follow-up of �3 months (48 patients) and patients with <3months of follow-up and complete remission (4 patients)were included. Eleven of the 52 patients had underlyingDGS. The mean duration of follow-up for the 48 patientswith �3 months of follow-up was 25 months (range, 3–120mo). Of the 41 APIGN patients without underlying DGS,23 (56.1%) achieved complete remission, 11 (26.8%) hadpersistent renal dysfunction, and 7 (17.1%) progressed toESRD including 2 patients who died, 1 of septicemia and 1of multiple myeloma. In this group without underlying DGS,on univariate analysis of 3-way outcome (complete remis-sion vs. persistent renal dysfunction vs. ESRD), correlates ofcomplete remission were younger age, female sex, lowerserum creatinine at time of biopsy, and the absence ofimmunocompromised state. There were no significantcorrelates between renal outcome and site of infection (p =0.07) or infectious organism (p = 0.31) or any otherdemographic, clinical, or serologic factor. Similarly, therewere no significant pathologic correlates of renal outcome,although patients with complete remission showed lessinterstitial inflammation and less acute tubular injury thanthe other 2 groups (both p = 0.051).

Correlates of serum creatinine >2 mg/dL at biopsy onunivariate analysis included older age, male sex, history ofpre-existing hypertension, lower serum albumin, a low C4,and greater endocapillary proliferation, interstitial inflam-mation, acute tubular injury, and more cellular crescents. Onmultivariate analysis, only a lower serum albumin (p =0.015), low C4 (p = 0.002), interstitial inflammation (p =0.012), endocapillary proliferation (p = 0.039) and acutetubular injury (p = 0.015) remained significant. The presenceof cellular crescents on biopsy did not correlate with risk of

ESRD (p = 0.071). A diffuse endocapillary proliferativepattern correlated with presence of nephrotic syndrome (p =0.021).

By multivariate analysis (binary logistic regressionanalysis), age (hazard ratio [HR], 0.938; 95% confidenceinterval [CI], 0.889–0.989; p = 0.017) and serum creatinineat biopsy (HR, 0.746; 95% CI, 0.557–0.999; p = 0.049) weresignificant and independent inverse correlates of completeremission, and higher serum creatinine at biopsy (HR, 1.486;95% CI, 1.033–2.137; p = 0.033) correlated with ESRD. Bymultinomial logistic regression analysis, age was the onlyclinical characteristic that distinguished complete remissionfrom persistent renal dysfunction (HR, 1.068; 95% CI,1.006–1.132; p = 0.030) and ESRD (HR, 1.087; 95% CI,1.002–1.180; p = 0.046). Serum creatinine at biopsy waslower in patients with complete remission versus ESRD (HR,1.659; 95% CI, 1.046–2.629; p = 0.03), as was percentage

TABLE 6. LM Findings of 16 APIGN Patients With Underlying DGS

Pathologic Finding No. of Cases %

No. of glomeruli, mean (SE) 18 (4.5)

No. of sclerotic glomeruli, mean (SE) 6 (3)

Sclerotic glomeruli, % (SE) 21.8 (4)

Pattern of DGS: nodular/diffuse 12/4 75/25

No. of cases with glomerular neutrophil infiltration (focal/diffuse) 16 (3/13) 100 (18.8/81.2)

No. of cases with cellular crescents (<20% of glomeruli, �20% of glomeruli) 4 (3, 1) 25.0 (75.0, 25.0)

No. of cases with necrosis (<20% of glomeruli, �20% of glomeruli) 4 (4, 0) 25.0 (100, 0)

Interstitial inflammation: none/focal/diffuse 1/13/2 6.3/81.3/12.5

Acute tubular injury: none/focal/diffuse 5/7/4 31.2/43.8/25.0

Tubular atrophy and interstitial fibrosis: mild/moderate/severe 6/7/3 37.5/43.8/18.8

Arteriosclerosis and arteriolar hyalinosis: mild/moderate/severe 1/11/4 6.3/68.8/25.0

Abbreviations and definitions: See Table 5.

FIGURE 2. An example of APIGN superimposed on diabeticglomerulosclerosis illustrates the combination of severe glo-merular neutrophil infiltration and distinct mesangial nodulesof sclerosis (H & E, � 400).




of global sclerosis (HR, 1.072; 95% CI, 1.001–1.148; p =0.048).

Thirty-three percent of the 52 patients with outcomedata were treated with steroids. The indication for steroidtherapy was renal insufficiency with or without crescents. Ofthis group, 70% had complete remission, 18% had persistentrenal dysfunction, and 12% progressed to ESRD. Nocorrelation was found between steroid therapy and renaloutcome: the 17 steroid-treated patients included 16 of 35nondiabetics and 1 of 16 diabetics (p = 0.009). When patientswith underlying DGS were excluded from the analysis,complete remission occurred in 12 of 17 patients treated withsteroids versus 10 of 23 non-steroid-treated patients (p =0.116). Thirty-five patients were known to have been treatedwith antibiotics. There was not sufficient information aboutthe adequacy of antibiotic therapy to draw conclusionsabout outcome.

Of the 11 patients with underlying DGS, 2 (18.2%) hadpersistent renal dysfunction and the remaining 9 (81.8%)progressed to ESRD, including 2 patients who died due tosepticemia. This was a much poorer renal outcome than

that of patients without underlying DGS (p < 0.001). Noneof these patients was treated with steroids.

DISCUSSIONIn the current study we analyze the clinical, pathologic,

and outcome data in a series of 86 adults with nonepidemicAPIGN. To our knowledge, this is the largest biopsy-basedseries and the only North American series in the modern era.Previous studies have suggested that the incidence of APIGNin Western Europe is declining, likely due to an improvedstandard of living and earlier treatment of pharyngealinfections49. The biopsy incidence of adult APIGN in thecurrent study was lower than that reported in 3 relativelyrecent European studies (0.6% vs. 1.7%–4.6%)21,28,29

TABLE 7. Glomerular Findings on IF (84 Cases)

No. of Positive Cases

(%)

Mean Intensity

When Positive*

IgG 55 (65.5) 1.3+

IgM 37 (44.0) 1.0+

IgA 37 (44.0) 1.3+

C3 84 (100) 2.4+

C1 30 (35.7) 0.8+

Kappa 57 (67.9) 1.3+

Lambda 59 (70.2) 1.3+

*Scale: trace (0.5%), 1–3+.

FIGURE 3. A representative immunofluorescence imageshows coarsely granular positivity (starry sky pattern) in aglobal glomerular capillary wall and mesangial distribution.(Anti-IgG, � 400).

TABLE 8. EM Findings (83 Cases)

No. of Cases %

Mesangial deposits (segmental/global) 75 (29/46) 90.4

Subendothelial deposits (segmental/global) 62 (41/21) 74.7

Subepithelial deposits (segmental/global) 78 (45/33) 94

Subepithelial humps 75 90.4

FIGURE 4. The most characteristic ultrastructural finding ishump-shaped subepithelial deposits involving the peripheralglomerular capillary walls andmesangial waist. The deposits arelarge, without intervening spikes. There is localized effacementof foot processes overlying the humps. (Electron micrograph,� 2000).




(Table 10). In our patient population, there was a malepredominance (2:1), similar to other studies12,21,28,29,40. Thepercentage of adult patients older than 64 years in the currentstudy was 33.7%, compared with 5.6% in the classic 1974study by Baldwin et al4 from New York City, 3.8% in the1979 report by Lien et al23 from South Australia, and 0% inthe 1942 study by Ellis14 and the 1936 study by Richter40.A longer life expectancy coupled with an increasingprevalence of comorbid conditions such as diabetes likelyaccounts for the predominance of elderly patients in thecurrent cohort.

Most cases of APIGN reported in the literature from1960 to 1980 occurred in patients without significant pastmedical history4,23,26. In the 1980s, an important associationbetween intravenous drug use (IVDU) and endocarditis wasappreciated in the United States34. Other factors predisposingto infection have been documented in a substantial fractionof patients described in more recent studies. In WesternEurope, alcoholism has become the most important riskfactor for APIGN, found in 12%–57% of patients21,28,29 (seeTable 10). In the current study, over one-third of patients had1 or 2 predisposing factors for infection; diabetes was by farthe most frequent factor, found in 29.1% of patients. Only2 patients were alcoholics (2.3%), 1 of whom was alsodiabetic. Only 1 patient had a history of IVDU, and thispatient was also diabetic.

Classically, APIGN occurs after streptococcal pharyn-gitis or skin infection4,26. More recently, other sites of

infection and more diverse organisms have been linked toadult APIGN. Considering all 242 patients compiled fromthe current study and the 3 modern European studies, thesites of infection included URT in 59 patients (24%), skin in39 (16%), lung in 33 (14%), endocarditis in 21 (9%), andteeth in 17 (7%). The responsible bacterium was streptococ-cus in 68 patients (28%), staphylococcus in 44 patients(17%), and Gram-negative bacteria in 25 patients (10%).Importantly, 7%–16% of patients had no clinical evidence ofinfection preceding the renal disease, and in 24%–59% ofpatients the offending microorganism was not identified (seeTable 10). These latter data indicate that APIGN should beincluded in the clinical differential diagnosis of nephriticsyndrome in adults, especially in patients with diabetes oralcoholism, even in the absence of a history of infection. Thelatent period between the streptococcal infection and theonset of renal disease was reported to be shorter after URTinfection (1–3 wk) than after skin infection (3–6 wk)11,30,35.In contrast, in our cohort the mean latent period after URTinfection was 4 weeks compared with 3 weeks after skin andlung infections and 2 weeks after endocarditis. We note thatin many cases infection was only discovered at the timeAPIGN was diagnosed, indicating that the infection may gounrecognized for some time. In 20% of patients withendocarditis and 27% of patients with pneumonia in thecurrent study and in 19% of patients with endocarditisreported by Majumdar et al24, the renal disease was diag-nosed at the same time as the infection. Srivastava et al53

reported 11 children in whom pneumonia and APIGNoccurred within 72 hours of each other.

The classic glomerular pattern of injury associatedwith infection, particularly after transient acute streptococcalpharyngitis or pyoderma, is diffuse endocapillary prolifera-tive and exudative GN. This pattern was present inapproximately three-quarters of patients in the current study.Focal endocapillary and exudative GN and mesangioproli-ferative GN were less common, seen in 12.8% and 8.1% ofpatients, respectively. The latter 2 patterns likely representthe resolving phases of the disease11,42, although some caseswith these milder degrees of injury never develop moresevere lesions11. Cellular crescents were present in aboutone-third of cases but affected >25% of glomeruli in only 9%of cases. In contrast, one-third of cases reported by Montsenyet al28 showed crescents in >25% of glomeruli.

By IF, the glomerular deposits stained dominantly forC3 (mean intensity, 2.4+) with weaker and more focalstaining for IgG (mean intensity, 1.3+). These findings aresimilar to those reported by others11. An explanation of thisstaining pattern is that the presumed nephritogenic antigen ofAPIGN, the streptococcal pyogenic exotoxin B (SPE B), is acationic cysteine proteinase that can bind directly to theglomerular basement membrane through charge interactionswithout formation of immune complexes5. The nephritogenicantigen or antigens appear to activate complement by thealternative or lectin pathways rather than through classicalpathway activation on IgG41.

Crescentic and necrotizing GN was seen in 4 patients(4.6%), 2 of whom had endocarditis, including 1 with C-ANCA

TABLE 9. Outcome and Prognostic Indicators* in 41 PatientsWithout Underlying DGS

CR PRD ESRDy p Value

No. of patients 23 11 7

Percentage of patients 56.1% 26.8% 17.1%

Age (yr) 46.5 62.9 63.6 0.005

Male:Female 11:12 8:3 7:0 0.029

Immunocompromised state, % 9 55 29 0.011

Serum creatinine, mg/dL

Baseline 1.1 0.9 1.4 0.076

At biopsy 3.0 4.8 6.6 0.006

Follow-up 1.1 2.0 10.0 <0.001

UVp, g/24 h

At biopsy 4.25 2.77 4.43 NS

Follow-up 0.64 1.02 5.13 NS

Glomerulosclerosis, % 9.7 10.5 22.4 NS

Interstitial inflammation 1.1 1.6 1.9 0.051

Acute tubular injury score (0–2+) 0.7 1.2 0.9 0.051

TA-IF average score (0–3+) 0.6 0.7 1.3 0.106

Abbreviations: CR = complete renal recovery, PRD = persistent renaldysfunction (renal insufficiency), ESRD = end-stage renal disease, UVp =urinary protein excretion, NS = not significant, TA-IF = tubular atrophy andinterstitial fibrosis.

*On univariate analysis. Patients had a follow-up of �3 months or <3months with CR.

yTwo patients with ESRD died.




seropositivity. These 2 cases showed extracapillary prolifer-ation (without significant endocapillary hypercellularity)and prominent tuft necrosis on LM, glomerular depositionof C3 with or without IgG on IF, and exclusively mesangialdeposits on EM. The remaining 8 patients with endocarditis(80%) had diffuse endocapillary proliferative and exudative

GN. In a 2000 study from the United Kingdom, Majumdaret al24 analyzed the renal pathologic findings in 62 patientswith infectious endocarditis. Sixteen patients had GN,diagnosed on biopsy in 9 and on autopsy on 7. In contrastto our findings, pauci-immune crescentic and necrotizing‘‘vasculitic’’ GN was the most frequent pattern, present in

TABLE 10. Current Spectrum of Nonepidemic APIGN in Adults, Present and Recent Reports

First Author (ref )

Keller (21) Montseny (28) Moroni (29) Nasr (Present Report)

Period of study 1984–1993 1976–1993 1979–1999 1995–2005

Country Germany France Italy United States

No. of patients 30 76 50 86

Biopsy incidence, % 4.5 4.6 1.7 0.6

Median age (yr) 49 NA 54 58

M:F 3:1 2.4:1 1.5:1 2:1

Patients with alcoholism, % 57 30 12 2

Patients with diabetes, % NA 8 10 29

Most common site ofinfection (%)

Teeth (23) Skin (13) URT (28) Skin (25)Lung (18)Teeth (13)Endocarditis (13)

URT (44) Lung (16)Urinary tract (12)

URT (23) Skin (17.5)Lung (17.5)Endocarditis (12)

Most common bacteria (%) Streptococcus* (40)Staphylococcus (13)

Staphylococcus (17)Gram-negativebacteria (14)Streptococcus* (14)

Streptococcus* (47)Gram-negativebacteria (22)Staphylococcus (12)

Streptococcus* (28)Staphylococcus (24)

Patients with no clinicalevidence of infection, %

10 7 12 16

Patients with no identifiableoffending microorganism, %

43 59 24 42

Follow-up, mo (mean) 1–76 (12.5) 1–108 20–138 (90) 3–120 (25)y

CR, % 64 28 43 56

PRD, % 28 53 47 27

ESRD, % 4 8 10 17

Death, % 4 11 10 5

Correlates of renal outcome

CR Age <40 yr, URTinfection, endocapillaryhypercellularity,Starry sky pattern

Absence of underlyingdiseasez, absence ofinterstitial inflammationz,absence of crescents,absence of subendothelialdeposits

Younger agez, femalesex, absence ofimmunocompromisedstate, lower serumcreatinine at biopsyz,absence of interstitialinflammation

PRD Alcoholismx,crescenticglomerulonephritisx

Nephrotic syndrome,crescenticglomerulonephritis,interstitial fibrosis

ESRD Higher serumcreatinineat biopsyz,underlying DGS

Abbreviations: See previous tables. NA = not available.*Determined by elevated ASLO titers and/or culture.yOnly the 41 patients without underlying DGS who had a follow-up�3 mo (37 patients) or <3 mo with CR (4 patients) were included for outcome analysis.zBy multivariate analysis.xChronic renal dysfunction was more common in patients with alcoholism and those with crescentic glomerulonephritis, although statistical analysis was

not performed.




11 patients (69%). Of the remaining 5 patients, 3 (19%) haddiffuse endocapillary proliferative GN and 2 (12%) hadMPGN24. In the older literature, focal GN was the mostcommon pattern associated with endocarditis, althoughdiffuse proliferative GN was not infrequent3,6,34. Only2.3% of patients in the current study and 8% of patientsreported by Montseny et al28 showed MPGN, attesting to therarity of this morphologic pattern in patients with infection.MPGN is encountered in the majority of patients with shuntnephritis and rarely in association with pneumonia, deep-seated infections, and osteomyelitis11,17,28,55.

Multiple studies with long follow-up have shown thatthe majority of children and patients with the epidemic formof APIGN have an excellent prognosis13,19,38,45. In a 2001Japanese study, Kasahara et al19 followed 138 children withnonepidemic APIGN diagnosed based on the presence ofhematuria, transient hypocomplementemia, and evidence ofgroup A beta-hemolytic streptococcal infection. None of thepatients had renal insufficiency. All patients had normaliza-tion of serum complement within 12 weeks, disappearance ofhematuria within 4 years, and disappearance of proteinuriawithin 3 years19. In a study of 534 children and adults withAPIGN in Trinidad (361 epidemic and 173 sporadic), 96.5%had complete recovery after 12–17 years of follow-up38. Notall APIGN epidemics are associated with good outcome. A2005 prospective study from Brazil of 56 patients, mostlyadults, who had acute GN following an outbreak of group CStreptococcus zooepidemicus found a less favorable long-term outcome47. After a mean follow-up of 5.4 years, 30% ofpatients were hypertensive, 49% had reduced creatinineclearance, and 22% had microalbuminuria47. Poor outcomesalso have been reported in adults with nonepidemic APIGN.According to the older literature, the rate of completeremission ranges from 53% to 76%22,30,40. Considering the 3recently reported studies from Europe and the current study,28%–64% of patients recovered completely, 27%–53% hadpersistent renal dysfunction, 4%–17% progressed to ESRD,and 4%–11% died21,28,29 (see Table 10). The poor outcomesin the modern era likely relate to older patient age and morefrequent comorbidities. The statistically significant histolog-ic and clinical predictors of renal outcome are listed inTable 10. On multivariate analysis, performed only in thecurrent study and the study by Moroni et al29, correlates ofcomplete remission were younger age, lower serum creati-nine at biopsy, absence of underlying disease (specificallyDGS in our study), and the absence of interstitial in-flammation (which was of borderline statistical significancein our univariate analysis but did not hold up in a mul-tivariate analysis that incorporated age and serum creatinineinto the model). The only significant predictor on multivar-iate modeling of ESRD in our study was a higher serumcreatinine at biopsy.

Previous studies in which statistical analysis was notperformed reported that the presence of abundant crescentswas associated with poor outcome10,52,57, whereas in thecurrent study crescents did not correlate with outcome. Thelack of detectable impact of crescents on clinical outcomemay have been the result of the small percentage of

glomeruli with crescents (only 9% of cases showed crescentsaffecting >25% of glomeruli, and 4.6% of cases showedcrescents affecting >50% of glomeruli).

The efficacy of immunosuppressive therapy forAPIGN is controversial. Roy et al43 compared the outcomein 5 children with crescentic poststreptococcal GN who weretreated with prednisone, azathioprine, cyclophosphamide,dipyridamole, and heparin followed by warfarin to that in 5children who were managed with supportive care only. Theyfound that quintuple therapy offered no advantage oversupportive care43. Evidence supporting the use of steroidtherapy for crescentic APIGN is largely anecdotal39,54,57.Although a significant subgroup (33% of patients) in thecurrent study received steroid therapy for variable periods oftime, no correlation could be found between this form oftherapy and outcome.

APIGN is one of the most common diseases to occursuperimposed onDGS, although only limited data are availableconcerning the clinical and pathologic features and outcome ofthis dual glomerulopathy25. Thirty cases have been reportedthat we know of; most include only 1–3 patients, and all but 146

were reported before 19901,2,7,8,15,20,27,36,37,44,46,48,56. Of the 21patients with available data, 15 had preceding streptococcalinfection and 6 had staphylococcal infection. In contrast,among the 16 adults with APIGN andDGS in the current study(which, to our knowledge, represents the largest reportedcohort to date), 10 were secondary to staphylococcus and only3were due to streptococcus. Patients with diabetes have a highincidence of staphylococcal infection, especially of lowerextremity skin and bone, due to higher colonization and morefrequent skin ulcerations. As we previously reported32,33 andre-encountered in this study, the pathologic findings ofAPIGN in patients with underlying DGS differ from thosein patients with pure APIGN in the predominance ofmesangial and subendothelial deposits with few smallsubepithelial deposits. Moreover, IgA is the dominantimmunoglobulin detected in glomeruli in the majority ofcases, especially those with staphylococcal infection. Wehypothesize that the formation of large, hump-shaped,subepithelial deposits is hindered by the thickened glomerularbasement membranes of underlying DGS. IgA dominance canmake it difficult to distinguish this histologic variant ofAPIGN from IgA nephropathy. Features that favor IgA-dominant APIGN over IgA nephropathy include depressedserum complement levels, diffuse endocapillary proliferativeand exudative GN on LM, and subepithelial hump-shapeddeposits on EM32,33,46. The outcome of APIGN in patientswith underlying DGS is dismal, with 9 of 11 patients in thecurrent study progressing to ESRD. Similar poor outcomeshave been reported by others7,27,46. Follow-up was availablein 15 of the 30 cases reported by other investigators2,7,27,36,46,56.Of these, 8 patients became dialysis dependent, 3 had per-sistent renal dysfunction, 3 had complete remission, and 1 diedshortly after diagnosis.

In conclusion, the epidemiology of adult APIGN isshifting. Diabetes and aging have emerged as major riskfactors. Among infectious agents, staphylococcus has becomenearly as common as streptococcus, and is the most frequent




agent in diabetics. Our data indicate that the overall prognosisof adult APIGN is guarded. Over a fourth of patients will havepersistent renal dysfunction and close to a fifth will progressto ESRD. Outcome is particularly poor in patients withunderlying DSG, of whom >80% can be expected to developESRD. As the population ages and diabetes becomes epidemicthroughout the industrialized world, the incidence of adultAPIGN is likely to increase.

ACKNOWLEDGMENT

We thank our many clinical colleagues who kindlyprovided follow-up data on their patients.

REFERENCES

1. Amoah E, Glickman JL, Malchoff CD, Sturgill BC, Kaiser DL, BoltonWK. Clinical identification of nondiabetic renal disease in diabeticpatients with type I and type II disease presenting with renal dys-function. Am J Nephrol. 1988;8:204–211.

2. Aziz S, Cohen AH, Winer RL, Llach F, Massry SG. Diabetes mellituswith immune complex glomerulonephritis. Nephron. 1979;23:32–37.

3. Baehr G. Glomerular lesions of subacute bacterial endocarditis. J ExpMed. 1912;15:330–347.

4. Baldwin DS, Gluck MC, Schacht RG, Gallo G. The long-term course ofpoststreptococcal glomerulonephritis. Ann Int Med. 1974;80:342–358.

5. Batsford SR, Mezzano S, Mihatsch M, Schiltz E, Rodriguez-IturbeB. Is the nephritogenic antigen in post-streptococcal glomerulone-phritis pyrogenic exotoxin B (SPE B) or GAPDH? Kidney Int. 2005;68:1120–1129.

6. Bell ET. Glomerular lesions associated with endocarditis. Am J Pathol.1932;8:639–663.

7. Bertani T, Mecca G, Sacchi G, Remuzzi G. Superimposed nephritis: aseparate entity among glomerular diseases? Am J Kidney Dis. 1986;7:205–212.

8. Cavallo T, Pinto JA, Rajaraman S. Immune complex disease com-plicating diabetic glomerulosclerosis. Am J Nephrol. 1984;4:347–354.

9. Churg J, Bernstein J, Glassock RJ. Renal Disease: Classification andAtlas of Glomerular Diseases. New York:Igaku-Shoin; 1995

10. Cunningham RJ III, Gilfoil M, Cavallo T, Brouhard BH, Travis LB,Berger M, Petrusick T. Rapidly progressive glomerulonephritis inchildren: a report of thirteen cases and a review of the literature. PediatrRes. 1980;14:128–132.

11. D’Agati VD, Jennette JC, Silva FG. Atlas of Nontumor Pathology: Non-Neoplastic Kidney Diseases. Washington, DC: American Registry ofPathology-Armed Forces Institute of Pathology; 2005:269–296.

12. Dodge WF, Spargo BH, Bass JA, Travis LB. The relationship betweenthe clinical and pathologic features of poststreptococcal glomerulone-phritis. A study of the early natural history. Medicine (Baltimore). 1968;47:227–267.

13. Drachman R, Aladjem M, Vardy PA. Natural history of an acuteglomerulonephritis epidemic in children. An 11- to 12-year follow-up.Isr J Med Sci. 1982;18:603–607.

14. Ellis A. Natural history of Bright’s disease. Lancet. 1942;1:1–7.15. Epstein SE, Becker EL. Acute glomerulonephritis in an adult with

longstanding diabetes mellitus. Ann Intern Med. 1961;54:97–103.16. Haas M. Postinfectious glomerulonephritis complicating diabetic ne-

phropathy: a frequent association, but how clinically important? HumPathol. 2003;34:1225–1227.

17. Haffner D, Schindera F, Aschoff A, Matthias S, Waldherr R, Scharer K.The clinical spectrum of shunt nephritis. Nephrol Dial Transplant.1997;12:1143–1148.

18. Jennings RB, Earle DP. Post-streptococcal glomerulo-nephritis: histo-pathologic and clinical studies of the acute, subsiding acute and earlychronic latent phases. J Clin Invest. 1961;40:1525–1595.

19. Kasahara T, Hayakawa H, Okubo S, Okugawa T, Kabuki N, TomizawaS, Uchiyama M. Prognosis of acute poststreptococcal glomerulonephri-tis (APSGN) is excellent in children, when adequately diagnosed.Pediatr Int. 2001;43:364–367.

20. Kasinath BS, Mujais SK, Spargo BH, Katz AI. Nondiabetic re-nal disease in patients with diabetes mellitus. Am J Med. 1983;75:613–617.

21. Keller CK, Andrassy K, Waldherr R, Ritz E. Postinfectious glomeru-lonephritis—is there a link to alcoholism? Q J Med. 1994;87:97–102.

22. Kushner DS, Armstrong SH, Dubin A, Szanto PB, Markowitz A,Maduros BP, Levine JM, River GL, Gynn TN, Pendras JP. Acuteglomerulonephritis in the adult. Longitudinal, clinical, functional andmorphologic studies of rates of healing and progression to chronicity.Medicine (Baltimore). 1961;40:203–240.

23. Lien JWK, Mathew TH, Meadows R. Acute poststreptococcal glomer-ulonephritis in adults: a long-term study. Q J Med. 1979;48:99–111.

24. Majumdar A, Chowdhary S, Ferreira MA, Hammond LA, Howie AJ,Lipkin GW, Littler WA. Renal pathological findings in infectiveendocarditis. Nephrol Dial Transplant. 2000;15:1782–1787.

25. Mazzucco G, Bertani T, Fortunato M, Bernardi M, Leutner M, BoldoriniR, Monga G. Different patterns of renal damage in type 2 diabetesmellitus: a multicentric study on 393 biopsies. Am J Kidney Dis. 2002;39:713–720.

26. McCluskey RT, Baldwin DS. Natural history of acute glomerulone-phritis. Am J Med. 1963;35:213–230.

27. Monga G, Mazzucco G, di Belgiojoso GB, Confalonieri R, Sacchi G,Bertani T. Pattern of double glomerulopathies: a clinicopathologic studyof superimposed glomerulonephritis on diabetic glomerulosclerosis.Mod Pathol. 1989;2:407–414.

28. Montseny JJ, Meyrier A, Kleinknecht D, Callard P. The currentspectrum of infectious glomerulonephritis. Experience with 76 patientsand review of the literature. Medicine (Baltimore). 1995;74:63–73.

29. Moroni G, Pozzi C, Quaglini S, Segagni S, Banfi G, Baroli A, Picardi L,Colzani S, Simonini P, Mihatsch MJ, Ponticelli C. Long-term prognosisof diffuse proliferative glomerulonephritis associated with infection inadults. Nephrol Dial Transplant. 2002;17:1204–1211.

30. Nadasdy T, Silva FG. Acute postinfectious glomerulonephritis andglomerulonephritis caused by persistent bacterial infection. In: JennetteJC, Olson JL, Schwartz MM, Silva FG, eds. Heptinstall’s Pathology of theKidney. 6th ed. Philadelphia:Lippincott Williams &Wilkins; 2007:321–396.

31. Nasr SH, Galgano SJ, Markowitz GS, Stokes MB, D’Agati VD.Immunofluorescence on pronase-digested paraffin sections: a valuablesalvage technique for renal biopsies. Kidney Int. 2006;70:2148–2151.

32. Nasr SH, Markowitz GS, Whelan JD, Albanese JJ, Rosen RM, Fein DA,Kim SS, D’Agati VD. IgA-dominant acute poststaphylococcal glomer-ulonephritis complicating diabetic nephropathy. Hum Pathol. 2003;34:1235–1241.

33. Nasr SH, Share DS, Vargas MT, D’Agati VD, Markowitz GS. Acutepoststaphylococcal glomerulonephritis superimposed on diabetic glo-merulosclerosis. Kidney Int. 2007;71:1317–1321.

34. Neugarten J, Baldwin DS. Glomerulonephritis in bacterial endocarditis.Am J Med. 1984;77:297–304.

35. Nordstrand A, Norgren M, Holm SE. Pathogenic mechanism of acutepost-streptococcal glomerulonephritis. Scand J Infect Dis. 1999;31:523–537.

36. Olivero J, Suki WN. Acute glomerulonephritis complicating diabeticnephropathy. Arch Intern Med. 1977;137:732–734.

37. O’Neill WM, Wallin JD, Walker PD. Hematuria and red cell casts intypical diabetic nephropathy. Am J Med. 1983;74:389–395.

38. Potter EV, Lipschultz SA, Abidh S, Poon-King T, Earle DP. Twelve toseventeen-year follow-up of patients with poststreptococcal acuteglomerulonephritis in Trinidad. N Engl J Med. 1982;307:725–729.

39. Raff A, Hebert T, Pullman J, Coco M. Crescentic post-streptococcalglomerulonephritis with nephrotic syndrome in the adult: is aggressivetherapy warranted? Clin Nephrol. 2005;63:375–380.

40. Richter AB. Prognosis in acute glomerular nephritis. Ann Intern Med.1936;9:1057–1069.

41. Rodriguez-Iturbe B, Batsford S. Pathogenesis of poststreptococcalglomerulonephritis a century after Clemens von Pirquet. Kidney Int.2007;71:1094–1104.

42. Rosenberg HG, Vial SU, Pomeroy J, Figueroa S, Donoso PL, Carranza

C. Acute glomerulonephritis in children. An evolutive morphologic andimmunologic study of the glomerular inflammation. Pathol Res Pract.1985;180:633–643.

43. Roy S III, Murphy WM, Arant BS Jr. Poststreptococcal crescenteric




glomerulonephritis in children: comparison of quintuple therapy versus

supportive care. J Pediatr. 1981;98:403–410.44. Sabnis SG, Antonovych TT. Diabetic nephropathy with superimposed

immune complex glomerulonephritis. South Med J. 1983;76:456–461.45. Sanjad S, Tolaymat A, Whitworth J, Levin S. Acute glomerulonephritis

in children: a review of 153 cases. South Med J. 1977;70:1202–1206.46. Satoskar AA, Nadasdy G, Plaza JA, Sedmak D, Shidham G, Hebert L,

Nadasdy T. Staphylococcus infection-associated glomerulonephritismimicking IgA nephropathy. Clin. J Am Soc Nephrol. 2006;1:1179–1186.

47. Sesso R, Pinto SW. Five-year follow-up of patients with epidemicglomerulonephritis due to Streptococcus zooepidemicus. Nephrol DialTransplant. 2005;20:1808–1812.

48. Sharma HM, Yum MN, Kleit S. Acute glomerulonephritis with diabetesmellitus: report of a case. Arch Pathol. 1974;97:152–154.

49. Simon P, Ramee MP, Autuly V, Laruelle E, Charasse C, Cam G, AngKS. Epidemiology of primary glomerular diseases in a French region.Variations according to period and age. Kidney Int. 1994;46:1192–1198.

50. Soni SS, Gowrishankar S, Kishan AG, Raman A. Non diabetic renaldisease in type 2 diabetes mellitus. Nephrology (Carlton). 2006;11:533–537.

51. Sorger K, Gessler U, Hubner FK, Kohler H, Schulz W, Stuhlinger W,Thoenes GH, Thoenes W. Subtypes of acute postinfectious glomerulo-nephritis. Synopsis of clinical and pathological features. Clin Nephrol.1982;17:114–128.

52. Srivastava RN, Moudgil A, Bagga A, Vasudev AS, Bhuyan UN,Sundraem KR. Crescentic glomerulonephritis in children: a review of 43cases. Am J Nephrol. 1992;12:155–161.

53. Srivastava T, Warady, Alon US. Pneumonia-associated acute glomer-ulonephritis. Clin Nephrol. 2002;57:175–182.

54. Usmani SZ, Shahid Z, Wheeler D, Nasser K. A rare case ofpostinfectious glomerulonephritis caused by pneumococcus in an adultpatient. J Nephrol. 2007;20:99–102.

55. Vella J, Carmody M, Campbell E, Browne O, Doyle G, Donohoe J.Glomerulonephritis after ventriculo-atrial shunt. QJM. 1995;88:911–918.

56. Yum M, Maxwell DR, Hamburger R, Kleit SA. Primary glomerulone-phritis complicating diabetic nephropathy: report of seven cases andreview of the literature. Hum Pathol. 1984;15:921–927.

57. Zent R, Van Zyl Smit R, Duffield M, Cassidy MJ. Crescentic nephritisat Groote Schuur Hospital, South Africa—not a benign disease. ClinNephrol. 1994;42:22–29.



glomeruloneritis

Documents