REVIEWS

Epidemiology, pathogenesis,
treatment and outcomes of infection-​
associated glomerulonephritis
Anjali A. Satoskar   1*, Samir V. Parikh2 and Tibor Nadasdy   1
Abstract | For over a century , acute ‘post-​streptococcal glomerulonephritis’ (APSGN) was the
prototypical form of bacterial infection-​associated glomerulonephritis, typically occurring after
resolution of infection and a distinct infection-​free latent period. Other less common forms of
infection-​associated glomerulonephritides resulted from persistent bacteraemia in association
with subacute bacterial endocarditis and shunt nephritis. However, a major paradigm shift in
the epidemiology and bacteriology of infection-​associated glomerulonephritides has occurred
over the past few decades. The incidence of APSGN has sharply declined in the Western world,
whereas the number of Staphylococcus infection-​associated glomerulonephritis (SAGN) cases
increased owing to a surge in drug-​resistant Staphylococcus aureus infections, both in the
hospital and community settings. These Staphylococcus infections range from superficial skin
infections to deep-​seated invasive infections such as endocarditis, which is on the rise among
young adults owing to the ongoing intravenous drug use epidemic. SAGN is markedly different
from APSGN in terms of its demographic profile, temporal association with active infection and
disease outcomes. The diagnosis and management of SAGN is challenging because of the lack
of unique histological features, the frequently occult nature of the underlying infection and the
older age and co-​morbidities in the affected patients. The emergence of multi-​drug-resistant
bacterial strains further complicates patient treatment.

Glomerulonephritides constitute a major cause of kid-
ney injury and frequently require a kidney biopsy for
aetiological diagnosis1. Among glomerulonephritides
associated with immune complexes, lupus nephritis,
IgA nephropathy (IgAN), infection-​associated glomeru­
lonephritis (GN) and membranous GN are the most
commonly encountered forms; cryoglobulinaemic GN
is relatively less common1. The immune system, and in
particular the complement system, has an important
role in all forms of immune complex-​associated GN2.
A complex interplay of environmental triggers, genetic
1
Department of Pathology,
Wexner Medical Center,
predisposition and dysregulated immunity lead to the
The Ohio State University, formation and accumulation of immune complexes
Columbus, OH, USA. in the glomeruli2. The presence of immune complexes
2
Department of Internal triggers the influx of effector immune cells, cytokine
Medicine, Wexner Medical release and secretion of enzymes, such as matrix metal-
Center, The Ohio State loproteinases (MMPs), that collectively damage the glo-
University, Columbus,
OH, USA.
merular capillary tuft; this phenotype is characteristic of
proliferative immune complex GN1–3.
*e-​mail: anjali.satoskar@
osumc.edu Infection is one of the most important triggers for the
https://doi.org/10.1038/ development of acute GN — bacterial, viral and para-
s41581-019-0178-8 sitic infections have all been implicated4. However, the
term ‘infection-​associated GN’ is typically reserved for
GN triggered by bacterial infections and the glomeru-
lar pheno­type is temporally associated with the infec-
tion (Fig. 1). Infection-​associated glomerulonephritides
include post-​infectious GN, which develops after com-
plete resolution of the infection and an infection-​free
latent period, as well as GN associated with an ongo-
ing infection. In both cases, infection might be acute or
chronic. Endocarditis-​associated GN is classified as an
infection-​associated GN, although it can also be caused
by non-​bacterial microorganisms, such as fungi5.
Classic acute post-​infectious GN secondary to
group A β-​haemolytic Streptococci (GAS) infections,
also termed acute post-​streptococcal GN (APSGN),
was the prototypical bacterial infection-​associated GN
throughout most of the last century6–8. However, from
the mid-1900s, the incidence of APSGN started to
gradually decline in most developed countries9–12. This
decline was largely attributed to improved living stand-
ards and socioeconomic conditions, which reduced
transmission of infection and prevented disease epidem-
ics13. A surge in the incidence of methicillin-​resistant

Nature Reviews | Nephrology

Reviews
Key points
• In the last century, acute post-​streptococcal glomerulonephritis (APSGN) was the
prototypical infection-​associated GN.
• Over the past few decades, the incidence of APSGN declined sharply in the Western
world, whereas the number of Staphylococcus infection-​associated glomerulonephritis
(SAGN) cases increased. The surge in staphylococcal infections, mainly methicillin-​
resistant Staphylococcus aureus (MRSA), is a probable contributing factor.
• APSGN still occurs with high frequency among highly populated and economically
disadvantaged communities around the world, where group A β-​haemolytic
streptococcal infections are common.
• APSGN remains the most common cause of acute GN among children, whereas SAGN
has very different demographic features and mainly affects the elderly population.
• SAGN is associated with ongoing infection when the patient presents with acute GN
and antibiotic treatment is the mainstay of management; aggressive
immunosuppressive treatment is clearly contraindicated and detrimental to the
patient when the infection is still active.
• Infective endocarditis is one of the most frequent infections implicated in SAGN;
epidemic intravenous drug use is a major contributing factor, as well as the increased
use of cardiac devices in elderly patients.
Staphylococcus aureus (MRSA) infections over the past
two decades further changed the epidemiological land-
scape of infection-​associated GN14–20. Staphylococcus
infection-​associated GN (SAGN) is now far more pre­
valent than APSGN, at least in developed countries19–25.
Factors implicated in the upsurge of SAGN cases include
lifestyle changes and iatrogenic risk factors, which are
increasingly pertinent owing to the growing number of
elderly patients with comorbidities such as diabetes mel-
litus and obesity who need long hospital stays, as well
as the increase in prolonged use of indwelling catheters
and central lines, implants and cardiac devices5,26. The
rise in intravenous drug use (IVDU), which can lead
to direct bloodstream infections and endocarditis, is
another major contributing factor27.
In this Review, we focus primarily on the chang-
ing epidemiology of infection-​associated GN, charac-
terized by the emergence of SAGN and the decline in
APSGN. We discuss the diagnostic pitfalls, histological
findings and approaches to the management of patients
with SAGN. We also describe classic APSGN and
endocarditis-​associated GN, and briefly discuss other,
less common forms of infection-​associated GN.
Acute post-​streptococcal GN
Epidemiology
The association between bacterial infections with acute
GN was observed as early as the 1800s in patients with
scarlet fever caused by infection with Streptococcus pyo-
genes, the prototypical GAS6–8. GAS are known to cause
a wide spectrum of illnesses, including superficial infec-
tions (such as pharyngitis and skin infections), invasive
infections (such as cellulitis, necrotizing fasciitis and
pneumonia) and toxin-​induced diseases; collectively,
Scabetic lesions
Contagious skin condition these diseases constitute a large disease burden glob-
caused by mites, tiny insect-​ ally9–12. Sequelae of a GAS infection (that is, APSGN
like parasites that infect the and acute rheumatic fever) are usually triggered by
top layer of the skin. upper respiratory tract infections (such as pharyngitis
Pyoderma
and tonsillitis) in colder climates and by skin infections
Bacterial skin infection with or superimposed bacterial infection of scabetic lesions in
formation of skin pustules. warmer climates4. Nephritogenic S. pyogenes strains 12,
4 and 1 are associated with APSGN triggered by throat
infections, whereas APSGN secondary to skin infec-
tions is linked to S. pyogenes M types 49, 42, 2, 57 and
60 (refs28–30). Not every infection is followed by acute
GN; therefore, the attack rate of APSGN is variable31.
In fact, of all children infected with the various nephrito-
genic strains of S. pyogenes, less than 2% show clinically
obvious signs of acute GN2,32. Host immune responses
are likely to determine whether GN will develop
following infection.
APSGN mainly affects patients in the paediatric age
group, predominantly those aged 3–15 years4. Epidemics
of APSGN were reported from the 1950s to as recently
as 2001 — ~12 epidemics with more than 100 patients
and ~16 epidemics with less than 100 cases — in both
rural and urban settings, as well as among military
recruits32–37. The risk of nephritis in epidemics ranged
from 5% when associated with throat infections to as
high as 25% when secondary to pyoderma caused by
strains of type 49 streptococci30,36,38.
However, the incidence of classic APSGN has slowly
declined since the 1980s, particularly in middle- to high-​
income countries. Collectively, improved living stand-
ards, socioeconomic growth, better hygiene practices,
less crowding in households, prevention and treatment
of scabies among children, and better access to health
care have contributed to the reduction in APSGN
cases9–13. However, these primary prophylactic measures
have not always been successfully implemented in less
developed countries and among economically deprived
indigenous populations worldwide. GAS infections and
their sequelae thus remain a ‘disease of disparities’9–13 as
demonstrated by the worldwide incidence distribution
of APSGN (Fig. 2)
According to the World Health Organization (WHO),
of the estimated 470,000 new annual cases of APSGN
worldwide, 97% occur in regions with poor socioeco-
nomic status12. Globally, APSGN continues to be the
most common cause of acute nephritis in children but
primarily occurs in developing countries13,37,39. One
analysis that evaluated data from 11 large population-​
based studies on the global burden of APSGN reported
a median incidence of 24.3 and 6.2 APSGN cases per
100,000 person-​years in children from less devel-
oped countries and from more affluent communities,
respectively10. The incidence of APSGN remains high
in heavily populated tropical regions of the world and in
indigenous communities; among Aboriginal Australians,
for example, the incidence is disproportionately high
— 239 per 100,000 individuals per year, compared
with an incidence of 7.33 cases per 100,000 individu-
als per year among non-​indigenous Australians10–14,40–42
(Fig. 2). A study from Auckland, New Zealand, reported
that, between 2007 and 2015, the annual incidence of
APSGN among children aged 1 to 14 years was 17 times
higher among Pacific Islanders and 7 times higher
among members of the Maori community than among
the individuals of European descent in the country.
Ethnic background, socioeconomic conditions and
access to health care were significantly associated with
disease incidence39,43. Of note, this study also showed that
the incidence of APSGN decreased over a 9-year period
www.nature.com/nrneph
 Reviews

Bacterial infection-associated glomerulonephritis

Acute post-streptococcal (or post- Staphylococcus infection-associated Glomerulonephritis

infectious) glomerulonephritis (APSGN) glomerulonephritis (SAGN) of other infections

• Occurs following resolution of • Associated with S. aureus and • Subacute bacterial

infection and an infection-free
latent period of 1–8 weeks
• Abrupt onset of
glomerulonephritis
• Usually involves paediatric
patients
S. epidermidis infections endocarditis
• Infection is usually active when • Shunt nephritis
glomerulonephritis develops • Indwelling central venous
• Aggressive immunosuppressive therapy catheter infections
is contraindicated, especially when • Deep-seated visceral
active infection is evident abscesses

Fig. 1 | Schema for classification of bacterial infection-​associated glomerulonephritis. Bacterial infection-​associated

glomerulonephritis (GN) mainly refers to acute post-​streptococcal GN (APSGN; also termed post-​infectious GN),
Staphylococcus infection-​associated GN (SAGN) and GN associated with other bacteria that cause endocarditis, shunt
infections or central venous line infections. Viral infection-​triggered glomerular diseases such as membranous GN,
cryoglobulinaemic GN and collapsing glomerulopathy are not included in this classification.

Acute nephritic syndrome
Syndrome characterized by
abrupt onset oedema and
dark ‘cola-​coloured’ or
‘tea-​coloured’ urine, with or
without renal impairment and
sub-nephrotic proteinuria.
Hypertensive urgency
Marked elevation of blood
pressure without target organ
damage, such as pulmonary
oedema, cardiac ischaemia,
neurological deficits or acute
renal failure.
Anti-​streptolysin O
Antibody targeted against the
toxic Streptolysin O enzyme
produced by Group A
Streptococcus bacteria.
following the implementation of community-​based
programs to proactively identify and treat pharyngitis
(commonly termed sore throat) and sepsis associated
with skin infections, improve vaccination provision and
provide early and more accessible community-​based
treatment of streptococcal infections with antibiotics to
reduce the transmission of GAS infections41.
It should also be emphasized that in the vast majority
of cases, kidney disease is subclinical, which is thought
to be 4 to 19 times more common than symptomatic
disease28. As pharyngitis is a common occurrence in the
community and can be multi-​factorial, detailed practice
guidelines for the diagnosis and management of GAS-​
induced pharyngitis were established in 2002 and fur-
ther updated in 2012 (refs44–46). These guidelines have
certainly helped physicians improve their management
of patients, but large prospective or outcome studies
have not been performed to evaluate the effectiveness
of these guidelines.
APSGN in elderly patients. APSGN is predominantly a
childhood disease but can also affect individuals aged
over 15 years. In the adult population, it is more fre-
quently reported among elderly individuals (aged over
60 years)47,48. As the proportion of elderly individuals
continues to increase in Western populations, more cases
of adult APSGN might be encountered and even biop-
sied because of the atypical clinical features and addi-
tional co-​morbidities among these patients. Collectively,
changes in the immune system secondary to the ageing
process and concurrent co-​morbidities, such as dia-
betes mellitus, hypertension and atherosclerotic heart
disease, as well as increased exposure to nephrotoxins
(including antibiotics and cancer chemotherapeutic
drugs), greatly contribute to an immunocompromised
state in elderly individuals and are thought to predispose
them to APSGN49. One study reported that 59% of their
109 elderly patients with infection-​associated GN had
diabetes mellitus, 14% had an underlying malignancy,
4% were alcoholic and 1% had malnutrition, liver cirrho-
sis or myelofibrosis47. Another study48 analysed 393 renal
biopsy samples from elderly patients with type 2 diabetes
and found that 9.4% of patients had post-​infectious GN
(the majority of cases were APSGN); 6.6% of biopsy
samples also showed evidence of concomitant diabetic
kidney disease.
Diagnosis of infection-​associated GN in elderly
patients can be difficult because the concomitant
co-​morbidities might mask the underlying infection.
Patients might not show specific signs and symptoms of
infection such as a high fever and leucocytosis. Instead,
clinical features might be non-​specific and related to
pre-​existing co-​morbidities such as worsening control
of diabetes, worsening cardiac failure or generali­zed
fatigue4,26. Compared with children, renal impair­ment
in the elderly with APSGN tends to be severe and pro­
longed; complete recovery might not occur and the
overall outcome in the elderly is also less favourable47,48.
Clinical and laboratory findings
As mentioned before, the majority of APSGN cases are
subclinical, but patients with clinical disease typically
present with acute nephritic syndrome32 (Table 1). Salt and
water retention leading to facial oedema are prominent
features and may be disproportionate to the degree of
renal dysfunction4. One paediatric study from Southern
India reported hypertension and hypertensive urgency
in 50% and 23% of admitted patients, respectively50.
Hypertensive encephalopathy or left ventricular dys-
function were reported in the same study, but these are
usually reversible with resolution of the acute illness;
21.3% of patients developed acute kidney injury (AKI)50.
Nephrotic syndrome and rapidly progressive GN (that
is, the crescentic form of the disease) are rare4. Severe
AKI that requires dialysis is uncommon in APSGN
(incidence of 3–5% in developing countries), but this
treatment is used when severe metabolic derangements
are present or in cases of refractory volume overload11.
Serum levels of complement proteins (mainly comple-
ment component C3) are almost always low during the
acute phase of APSGN, probably owing to activation of
the complement cascade and increased consumption
of complement factors due to formation of immune com-
plexes. Anti-​streptolysin O (ASO) serum titres increase as

Nature Reviews | Nephrology

Reviews

France Denmark Czech Republic Mainland China

(1981–1985) (1985–1997) (1994–2000) (1979–2002)
0.6 7.5 4.43 13.2

Italy
(1996–2000) North and South Korea
United States 0.29 (1973–1995)
(1999–2006) 6
0.78 Tunisia
(1975–2005)
44 Hong Kong
French (1993–1997)
Caribbean 2.6
Nigeria
(1979–1980) (1986–1991)
20 24.3
India
(1986–2002) Australia
Brazil Jordan (1993–1995)
(1999–2005) (1970–1980) 39.24
239 (Aboriginal
22.8 25 Australians)
Singapore
(1985) 7.33 (non-Aboriginal
Chile Saudi Arabia 10.8 Australians)
(1980–1989) (1979–1983)
18.1 25
New Zealand
Kuwait (1981–1984)
(1980–1984) 5.9
19.5 46.5 (Pacific Islanders)
50.5 (Maori tribes)

Fig. 2 | global ApSgN incidence. World map showing the incidence of acute post-​streptococcal glomerulonephritis
(APSGN) per 100,000 person years. Data are based on biopsy studies and large population-​based calculations performed
during the specified study periods9–12. The incidence of APSGN has sharply declined since the 1980s but remains high
in some regions of the world. APSGN is still the most common cause of acute GN among children worldwide. Incidence
of APSGN can vary greatly among different communities even within a nation. This is very evident in Australia and
New Zealand, where the incidence of APSGN is much higher among Aboriginal Australians and Pacific Islanders,
respectively , who live in some of the most impoverished regions of these countries. Disease prevalence is linked to the
level of socioeconomic development of the community. Incidence is also high in tropical regions of the world affected
by overcrowding.

patients convalesce and rising ASO levels provide the Treatment

best evidence of recent infection with GAS46,51.
Pathology
Histologically, APSGN is characterized by diffuse glo-
merular endocapillary hypercellularity, frequently with
numerous polymorphonuclear leukocytes, also known
as “exudative glomerulonephritis”4 (Fig. 3a,b; Table 2).
Crescents are uncommon, but might be seen in severe
cases. Direct immunofluorescence detects complement
C3 (with or without IgG) and produces distinct stain-
ing patterns. A ‘garland pattern’ refers to coarse granular
‘lumpy-​bumpy’ staining along the glomerular capillary
walls (Fig. 3c), whereas a ‘starry sky pattern’ describes
a randomly distributed, finely granular pattern with
few interspersed large deposits. The third pattern is
predominantly mesangial staining4. Examination by
electron microscopy reveals large, often numerous,
dome-​shaped subepithelial electron-​dense deposits
or ‘humps’ (Fig. 3a,d). Although humps can be seen in
several other glomerulonephritides, they are the most
characteristic finding in APSGN4. These subepithelial
deposits are especially numerous during the first few
weeks of acute GN, after which their number declines.
Sparse mesangial and subendothelial deposits are also
typically present.
The treatment of APSGN focuses on the clinical symp-
toms of acute nephritis and is usually temporary, as clin-
ical manifestations typically begin to improve 1–2 weeks
after disease onset and renal function usually returns to
baseline levels within 4 weeks, even in cases of cres-
centic GN47. Restriction of daily sodium consumption
to 2,000–2,500 mg and fluid intake to 2,000 ml with
diuretic therapy is recommended to manage hyperten-
sion and volume overload52. Anti-​hypertensive agents,
including vasodilators such as calcium channel blockers,
are required to control blood pressure, but angiotensin-​
converting enzyme (ACE) inhibitors are avoided in
APSGN as the condition is usually self-​limited and dur-
ing acute nephritis, ACE inhibitors may further worsen
renal function and cause hyperkalaemia52. The short
duration of treatment for APSGN also means that any
long-​term benefit of ACE inhibitors might be negligi-
ble. Hypertensive encephalopathy and AKI with urae-
mia can occur in severe cases of APSGN. In such cases,
both parenteral therapy to manage blood pressure and
dialysis should be pursued to prevent life-​threatening
neurological and cardiovascular complications until the
acute nephritis resolves53.
Antibiotics are not warranted in APSGN, as the acute
infection has subsided by the time nephritis develops.

www.nature.com/nrneph
 Reviews

Table 1 | Comparison of clinical and laboratory features in ApSgN and SAgN

Feature ApSgN SAgN
Age group Mainly paediatric age group (peak incidence in the • Mainly adults, between 50 and 80 years of age
first decade) • Younger patients with IVDU
Bacterial strain Nephritogenic strains of Streptococcus pyogenes (GAS) MRSA , MSSA , MRSE, MSSE
Type of Streptococcal pharyngitis, tonsillitis, mastoiditis, Endocarditis, skin infections (usually leg ulcers),
infection peritonsillar abscess, otitis media, pyoderma, cellulitis, skin abscesses, osteomyelitis, septic
streptococcal superinfection of scabies arthritis, pneumonia, bacteraemia of unclear
source, post-​surgical site infections
Infection-​free • 1–2 weeks for streptococcal sore throat • No infection-​free latent period
latent period • 3–6 weeks for pyogenic streptococcal skin • Infections can be occult and deep-​seated
infections • Co-​morbid conditions may mask signs of
• In many patients, GN may remain a subclinical infection. Infection may come to attention
disease after the patient presents with acute nephritis
Clinical • Renal function normal or mildly elevated serum • AKI, microscopic haematuria, heavy
presentation creatinine levels. Dark-​coloured urine (‘cola-’ or proteinuria (nephrotic range)
‘tea-​coloured’ urine). Proteinuria sub-​nephrotic • Worsening of underlying co-​morbid
range conditions such as diabetes mellitus, HTN,
• Abrupt onset oedema (prominent facial oedema) heart failure
• HTN in half of the affected children • Subset of patients present with LCV rash
Laboratory • Low C3, normal C4 levels • Low C3 in 30 to 50% of patients, C4 usually
findings • Rising ASO titres normal
• Positive ANCA serology in a subset of patients.
Often dual positivity for myeloperoxidase and
proteinase 3. Usually , titres are low positive
Outcome • Usually complete recovery in paediatric patients • Prognosis is unpredictable
• For adult patients, prognosis is less favourable. • Poorly controlled diabetes, advanced age,
Rare cases can progress to chronic kidney disease endocarditis, ANCA positivity with crescents
when the kidney shows numerous crescents in the biopsy portend worse prognosis
AKI, acute kidney injury ; ANCA , anti-​neutrophil cytoplasmic antibody ; APSGN, acute post-​streptococcal glomerulonephritis;
ASO, anti-​streptolysin O; C3, complement C3 protein; GAS, group A β-​haemolytic Streptococci; HTN, hypertension; IVDU,
intravenous drug use; LCV, leukocytoclastic vasculitis; MRSA , methicillin-​resistant Staphylococcus aureus; MRSE, methicillin-​
resistant Staphylococcus epidermidis; MSSA , methicillin-​sensitive Staphylococcus aureus; MSSE, methicillin-​sensitive
Staphylococcus epidermidis; SAGN, Staphylococcus infection-​associated glomerulonephritis.

However, during the acute infection, antibiotics are
recommended to reduce the probability of developing
APSGN. Prophylactic antibiotics should also be con-
sidered for individuals in close contact with patients
affected by group A streptococcal infection in loca-
tions where the prevalence of disease is high to mitigate
the spread of disease and prevent disease outbreaks11.
Prophylactic antibiotics are not recommended for close
contacts in industrialized nations where prevalence
is low as the risk of infection transmission does not
outweigh the risk associated with antibiotic use44.
Renal biopsy is typically not warranted when APSGN
is suspected, as the clinical history and presentation are
specific and the disease is usually transient. However, if
the clinical presentation is atypical, or if clinical manifes-
tations persist or worsen over time, then a kidney biopsy
should be considered.
Definitive recommendations regarding the use of
immunosuppression for the treatment of APSGN can-
not be made based on the currently available evidence,
which was mostly obtained from small retrospective
studies. Only one prospective study exists, in which
children with biopsy-​proven crescentic APSGN received
either supportive care or immunosuppression54. After
3 months of treatment, kidney biopsy samples were col-
lected from 8 of 10 patients. The clinical and histologi-
cal outcomes of the two treatment groups, including the
level of parenchymal scarring, were similar. However,
the rate of improvement in renal function was signifi-
cantly faster in patients who received immunosuppres-
sive therapy than in those who received supportive care
alone54. In severe cases of APSGN, high-​dose corticos-
teroids, starting with pulse methylprednisolone may be
warranted to suppress acute inflammation, limit chronic
renal damage and facilitate rapid renal recovery55,56.
Outcome
The prognosis of APSGN, even in patients with acute
crescentic GN and renal failure, is generally excellent56–60.
Renal function is rarely diminished and the development
of end-​stage renal disease (ESRD) due to APSGN is
extraordinarily rare. However, abnormal urine findings
with microscopic haematuria and low-​level proteinuria
persist in up to 20% of patients with APSGN as long as
5 years following the episode of acute GN58.
As discussed above, APSGN typically occurs after an
infection with GAS or, occasionally, group C or group
G streptococci10,11. Additionally, nephritogenic antigens
such as nephritis-​associated plasmin receptor (NAPlr)
and streptococcal pyrogenic exotoxin B (SPEB) have
been implicated in the pathogenesis of APSGN11 (dis-
cussed later). However, outcomes have not been differ-
entiated according to bacterial strain or antigens and this
remains an area of active research.
Renal outcome and overall prognosis are generally
worse in adults with APSGN than in children, as elderly

Nature Reviews | Nephrology

Reviews

a Mesangial cell NAPIr Bowman's space Podocyte

SPEB
Fibrinogen

C3
Fibrinogen
Antibody Plasmin
MMPs
Immune Monocyte
Cytokine complex
Plasmin C3 Neutrophil
MMPs

Endothelial
cell
ECM GBM

b Mesangial and endocapillary c C3 immune

hypercellularity complex deposits d Subepithelial humps

Fig. 3 | glomerular pathology and histological features in ApSgN. a | Streptococcal pyrogenic protease exotoxin B (SPEB)
and nephritis plasmin-​binding protein (NAPlr) are thought to be two leading streptococcal antigens involved in the
pathogenesis of acute post-​streptococcal glomerulonephritis (APSGN). Both antigens interact with plasmin, which is
thought to have an important role in the activation of complement and the release of matrix metalloproteinases (MMPs)
and collagenases, immune cell infiltration and activation (including cytokine release), as well as changes in the glomerular
basement membrane (GBM) and mesangial extracellular matrix (ECM). These changes probably allow immune complexes
to breach the filtration barrier and accumulate as subepithelial humps. Fibrinogen is another acute phase reactant matrix
protein that binds to and is deposited with antigen–antibody complexes. b | Light microscopy image of endocapillary
proliferative GN, stained with haematoxylin & eosin (H&E), 400 × magnification. c | Direct immunofluorescence staining
for complement protein C3 shows lumpy-​bumpy , coarsely granular staining in the mesangium and glomerular capillary
loops, 400 × magnification. d | Electron microscopy image of subepithelial ‘humps’; tissue stained with uranyl acetate,
30,000 × magnification.

Superantigens
Class of antigens that binds
directly to HLA molecules
without internal processing of
the antigenic peptide, resulting
in non-​specific activation of
T lymphocytes and polyclonal
T cell activation with massive
cytokine release.
patients are mostly affected49,58–62. Elderly individuals
have less renal reserve because of ageing and are more
likely to develop glomerulosclerosis and permanent kid-
ney damage than children, who typically have healthy
kidneys before developing APSGN47,52. Moreover, adults
are more likely to develop azotaemia, congestive heart
failure and nephrotic-​range proteinuria during the
acute phase of the illness. During follow-​up, 30–50% of
elderly patients have residual hypertension, persistently
abnormal urinalysis or persistently elevated serum crea­
tinine levels, and focal global glomerulosclerosis on
kidney biopsy58,59.
Staphylococcus infection-​associated GN
Epidemiology
Historically, staphylococci were implicated in the devel-
opment of GN in the context of chronic infections such
as in some cases of endocarditis or ventriculo-​atrial
shunt infections caused by coagulase-​negative S. epi-
dermidis4. Only a few isolated case reports published in
the 1970s and 1980s described diffuse proliferative GN
in patients with acute staphylococcal infections, which
included lung abscess, skin infections, cellulitis and
pneumonia63–67. Most cases were secondary to MRSA,
but some involved coagulase-​negative S. epidermidis
and methicillin-​sensitive S. aureus (MSSA). In the mid-
1990s, reports from Japan described S. aureus infections
with concomitant acute GN and implicated staphylococ-
cal superantigens in the disease68–73. The next few case
series, mainly published in the USA, highlighted the
IgA-​dominant nature of the immune complex deposits
in SAGN21–25 and the concomitant presence of leukocy-
toclastic vasculitis in a subset of patients with SAGN,
which mimicked IgA vasculitis (formerly known as
Henoch–Schönlein purpura)74–76.
SAGN was shown to mainly affect older adult males
and its incidence peaked between the fifth and sev-
enth decades of life26,75,76 (Table 3). SAGN can affect the
younger population as well, especially in the context of
IVDU-​associated endocarditis26,77–80.

www.nature.com/nrneph
 Reviews

Table 2 | histological features in ApSgN and SAgN on kidney biopsy

Technique ApSgN
Light • Fairly uniform
microscopy • Diffuse or focal endocapillary
hypercellularity
• Exudative lesions (intracapillary
polymorphonuclear leukocytes (admixed
mononuclear cells) may be present
• Acute tubular necrosis may be present
• Few scattered RBC casts
• Variable degree of interstitial
inflammation and oedema
Immuno­ • Bright coarse granular ‘lumpy-​bumpy’
fluorescence C3 staining along the glomerular capillary
loops and mesangium with or without IgG
• Garland pattern — densely packed large
confluent deposits along the capillary loops
resembling membranoproliferative GN
• Starry sky pattern — irregular granular
distribution of smaller deposits
• Late in the disease course, only mesangial
staining seen
Electron • Large, scattered and usually abundant
microscopy discrete dome-​shaped electron-​dense
subepithelial deposits described as ‘humps’
• Few intramembranous, subendothelial
and mesangial electron-​dense deposits
APSGN, acute post-​streptococcal glomerulonephritis; C3, complement protein C3; EM, electron microscopy ; FSGS, focal segmental glomerulosclerosis;
RBC, red blood cells; SAGN, Staphylococcus infection-​associated glomerulonephritis.
SAgN
• Wide spectrum
• Mesangial hypercellularity
• Endocapillary hypercellularity (segmental or global) with exudative lesions in some cases
• Segmental necrotizing lesions
• Nodular diabetic glomerulosclerosis with superimposed mesangial hypercellularity
• FSGS lesions usually absent. If present, primary IgA nephropathy suspected
• Large capillary wall deposits resembling ‘wire-​loop lesions’
can be seen in a small subset of patients with concomitant liver cirrhosis
• Acute tubular necrosis frequently seen and is usually severe, with marked flattening of
tubular epithelial cells
• Scattered RBC casts are common
• Interstitial inflammation is mild to moderate and patchy
• Most common pattern consists of mild to moderate IgA and moderate to strong C3.
Weak or no IgA in 25% of the cases; not always IgA-​dominant
• Exclusive C3 staining with negative IgA and IgG in 15% of biopsy samples
• Pauci-​immune pattern — weak to negative IgA , C3, IgG in 13% of samples
• In some cases, staining may be weak despite the presence of prominent subepithelial
humps on EM
• Co-​dominant IgG and IgA in 40% of the biopsies. λ-​light chain staining may be
stronger than κ-​light chain staining, similar to that seen in primary IgA nephropathy ;
however, the immune complexes are polyclonal in nature
• Small to medium-​sized mostly mesangial immune-​type deposits
• Few subendothelial and intramembranous deposits
• Subepithelial ‘humps’ seen in 31% of biopsy samples
• In a small subset of patients with concomitant liver cirrhosis, large confluent
mesangial, subendothelial and subepithelial deposits may be present

The exact incidence of SAGN is difficult to determine
and no large studies exist that describe the incidence and
prevalence of this disease in the general population.
In our practice and in other reports, 0.8% to 0.9% of
total native kidney biopsy samples showed SAGN21–26.
One study reported five cases of IgA-​d ominant
SAGN out of 4,600 biopsy samples (0.1%) collected
from 2000 to 2002 and, in a subsequent study, the same
researchers identified 93 cases of infection-​associated
GN out of 10,080 biopsy samples (0.9%) from elderly
patients collected between 2000 and 2010 (refs21,47). The
overall frequency of culture-​proven SAGN in overall
renal biopsy material seems low, but the percentage
would be much higher if calculated among biopsy sam-
ples with acute GN. Over the past 13 years, we have
seen at least 40 cases in which the biopsy findings
are highly suggestive of SAGN, but this is difficult to
prove because of the lack of positive culture results at
the time of biopsy (A.A.S., unpublished observations).
Therefore, the true incidence of SAGN might be higher
than reported.
Both S. aureus and S. epidermidis infections can lead
to SAGN4. MRSA is the leading cause of nosocomial
infections caused by a single bacterial pathogen in the
USA and is estimated to cause approximately 44% of all
hospital-​acquired infections14. The incidence of MRSA
infections acquired in the community, also known as
community-​acquired MRSA, has also been increas-
ing since the mid-1990s and has reached epidemic
proportions14–20. Community-​acquired and hospital-​
acquired cases of MRSA can overlap and are thus no
longer differentiated. Among the general population,
the prevalence of permanent and transient colonization
with S. aureus is approximately 20% and 60%, respec-
tively81–84. S. epidermidis is a skin commensal organism,
but has emerged as a potential pathogen owing to its
ability to survive in biofilms, which enables it to suc-
cessfully survive on implantable and indwelling medical
devices (such as central venous catheters) increasingly
used in medical care84.
Clinical and laboratory findings
Patients with SAGN usually present with AKI, micro-
scopic haematuria and nephrotic-​range proteinuria
(Box 1); infection is typically still ongoing when patients
develop SAGN4,21–26 (Table 1). Superficial infections, such
as skin infections, cellulitis or superficial abscesses, may
be clinically apparent. However, in the case of deep-​
seated infection such as osteomyelitis, endocarditis, deep
visceral abscesses or dental abscesses, the presence of
an infection may not be so obvious. Fever is not always
present and disease symptoms may be non-​specific, such
as fatigue, weight loss, exacerbation of other chronic
conditions such as diabetes mellitus or congestive heart
failure26,74. Blood cultures might also be negative (Table 3)
and local wound cultures as well as imaging studies
might be required to identify the source of infection26.
A subset of patients (20%) develop a purpuric skin rash
(leukocytoclastic vasculitis) similar to that observed in
IgA vasculitis74–76. Positive anti-​neutrophil cytoplasmic
antibody (ANCA) serology has been reported in cases
of systemic infection, especially in patients with bac-
terial endocarditis, further complicating the diagnosis
of SAGN5,26,85. The presence of normal serum C3 levels

Nature Reviews | Nephrology

Reviews

Table 3 | Clinical and demographic features of patients with SAgN in the three largest published case series
Clinical and demographic Satoskar et al.a (2016)26 Bu et al.b (2015)79 Nasr et al.c (2011)47
features
Absolute Frequency Absolute Frequency Absolute Frequency
number (%) number (%) number (%)
Caucasian 74 95 31 39.7 82 75
African American 3 3.8 0 – 5 5.0
Asian 1 1.2 42 53.8 2 2.0
Other ethnicity 0 – 5 6.4 10d 9.0
Males 61 78 63 81 80 73
Females 17 22 15 19 29 27
Underlying diabetes mellitus 32 41 18 23.1 53 49
ANCA positive 9 of 41 22 0 – 5 5
Hepatitis C positive 22 28 NA – NA –
Heart diseases NA – 13 16.7 1 1
Malignancy NA – 12 15.4 15 14
Staph strain 78 100 53 68 50 46
MRSA 42 59 38 48.7 23 21
MSSA 17 27 14 17.9 16 15
MRSE 3 1.20 NA NA
MSSE 2 1.20 3 3.8 6 5
Staph strain unknown 7 11 0 – 8 7
Gram-​negative bacteria 0 – 3 3.8 11 11
Mixed Staph and/or Gram-​ 7 9 NA – NA –
negative bacterial infection
Streptococcus 0 – 3 2.6 17 16
Rickettsia spp., Chlamydia spp., 0 – 4 5.0 1 1
Acinetobacter spp.
No culture growth 0 – 10 12.8 37 34
AKI at presentation 78 100 75 83 48 46
Blood culture positive 39 50 NA – NA –
Local wound culture positive 43 55 68 87 72 66
Both cultures positive 4 5 NA – NA –
Low C3 19 of 64 30 45 54.2 57 of 83 69
Low C4 9 of 64 14 NA – 29 of 83 35
Both C3 and C4 low 9 of 64 14 NA – 26 of 83 31
Purpuric lower extremity rash 16 20.5 NA – NA –
Nephrotic range proteinuria 35 of 73 48 78 21 54 of 109 50
Endocarditis 18 21 NAe – 7 6
Osteomyelitis or joint infection 17 22 13 16 8 7
Respiratory tract infections 6 8 13 16 28 26
Bacteraemia 10 14 3 8.4 3 3
Skin infections 17 22 19 19.2 31 28
Post-​surgical site infection 2 1 NA – NAf –
Visceral abscess 6 8 15 19.2 5 5
Urinary tract infection 2 3 2 3.6 14 13
Other infections 10 13 NA – NA –
Unknown infection 0 – 9 2.4 19 17
AKI, acute kidney injury ; ANCA , anti-​neutrophil cytoplasmic antibody ; C3, complement protein C3; C4, complement protein C4;
MRSA  , methicillin-​resistant Staphylococcus aureus; MRSE, methicillin-​resistant Staphylococcus epidermidis; MSSA , methicillin-​
sensitive Staphylococcus aureus; MSSE, methicillin-​resistant Staphylococcus epidermidis; NA , data not available; SAGN, Staphylococcus
infection-​associated glomerulonephritis; Staph, staphylococcal. aSingle-​centre study of staphylococcal culture-​positive biopsy
samples (n = 78, mean age = 55 ± 12.1 years, age range = 21–91 years). bPooled analysis of biopsy samples from previously published
studies (n = “78”, mean age = 58 ± 17 years, age range = 21–89 years). cSingle-​centre study of infection-​associated glomerulonephritis
(GN) in elderly individuals (n = 109, age range = 65–90 years). dPatient ethnicity undisclosed. eIn this study , endocarditis cases were
classed as visceral abscess. fIn this study , post-​surgical site infections were classed as skin infections.

www.nature.com/nrneph

Box 1 | Criteria for the diagnosis of SAgN
No single clinical or pathological feature is pathognomonic for Staphylococcus
infection-​associated glomerulonephritis (SAGN) or other bacterial infection-​associated
GN. Clinical features, biopsy findings, culture results and urinalysis findings should all
be taken into consideration at the time of diagnosis. Also, not every patient with SAGN
might fulfil all diagnostic criteria74.
Definitive diagnostic criteria
• Culture-​proven staphylococcal infection (ongoing or in the recent past).
• Acute onset proliferative GN with IgA and complement protein C3 containing
glomerular immune complex deposits, acute kidney injury, nephrotic-​
range proteinuria and (usually microscopic) haematuria. The degree of
glomerular hypercellularity and the extent of glomerular crescents can be
highly variable.
Additional criteria
Although not definitive, these features might suggest the possibility of infection-​
associated GN. In the presence of these features, an underlying staphylococcal
infection should be carefully investigated. This is especially true in cases of GN and develop in persistent, smouldering or relapsing glo-
AKI, with recent onset severe proteinuria and microscopic haematuria in combination
with any of the following:
• Presence of potential risk factors for infection — diabetes mellitus, intravenous drug
use, recent surgical or invasive procedure, prosthetic devices such as pacemakers,
heart valves or orthopaedic devices, poor dentition and/or tooth abscesses,
multiple trauma with open wounds, non-​healing ulcers or post-​amputation wounds
in diabetic patients, indwelling central or peripheral intravenous catheters or a direct consequence of SAGN. Of note, the presence
ventriculo-peritoneal shunt.
• Low serum C3 levels
• Leukocytoclastic vasculitic rash (LCV) — IgA staining often seen in the biopsy of
affected skin.
• Positive anti-​neutrophil cytoplasmic antibody (ANCA) serology (titre might only be
mildly positive), atypical ANCA or dual specificity for ANCA antibodies — proteinase 3
and myeloperoxidase. Lupus serologies, rheumatoid factor and cryoglobulin test are
usually negative, but patients with endocarditis may show one or more of these
serologies as well.
• Predominantly C3 immunofluorescence detected in the mesangial and capillary walls
of the glomerulus (with or without IgA staining) along with electron-​dense immune
deposits on ultrastructural examination.
• Subepithelial humps identified by electron microscopy.
also does not exclude the possibility of SAGN. In our
SAGN cohort, 30% of patients had low serum C3 levels26,
whereas in the cohort of elderly patients, reported by
Nasr et al., 69% of patients had low C3 serum levels47.
This discrepancy might be due to the inclusion of elderly
patients with APSGN (16%) in the cohort analysed by
Nasr and colleagues.
Pathology
The histological features of SAGN are those of an active
proliferative immune complex GN with or without
focal crescents (Table  2) . IgA and C3 staining by
immuno­fluorescence is also characteristic. Owing to
the lack of unique histological features, presence of
a wide spectrum of histological changes and typically
adult presenta­tion, SAGN can be confused with several
Pauci-​immune staining
pattern other forms of glomerulonephritides on kidney biopsy
Lack of or very few immune (Table 4).
complex deposits.
Light microscopy. Glomerular lesions in SAGN are char-
Hyaline thrombi
Intracapillary proteinaceous
acteristic of proliferative immune complex-​mediated
globules resembling hyaline GN (Table 2). The degree of proliferation is hetero­geneous
material. and ranges from purely mesangioproliferative lesions
Nature Reviews | Nephrology
Reviews
to focal or diffuse endocapillary proliferative lesions
(Fig. 4a,b), with or without segmental necrotizing lesions
and cellular crescents. Crescents were present in approx-
imately one-​third of the SAGN biopsies in our study26
and 37% of all biopsies from the Nasr et al. study47,
which included SAGN and APSGN cases; these cres-
cents ranged from small subtle segmental necrotizing
lesions to large cellular crescents. In our cohort26, 60%
of patient samples had endocapillary hypercellularity
and in 55% of those biopsies with endocapillary hyper-
cellularity, we observed exudative lesions consisting of
endocapillary polymorphonuclear leukocytes (Fig. 4c). As
many patients with SAGN are also affected by diabetes,
underlying nodular diabetic glomerulosclerosis might
also be present (Fig. 4d).
Focal segmental glomerular sclerosis (FSGS), or
segmental glomerular scarring lesions, which usually
merular diseases, are rare in SAGN because of the acute
nature of the disease. When present, such lesions are
more likely to be related to hypertensive disease or other
types of chronic kidney disease, prevalent in the older
age group in which SAGN typically occurs, rather than
of these glomerular lesions might also be indicative of
primary IgAN (Table 4).
Immunofluorescence microscopy. SAGN is character-
ized by the presence of IgA and C3 in immune complex
deposits, typically in the mesangium, and some along
the glomerular capillary loop segments21–26. SAGN is
often referred to as IgA-​dominant infection-​associ-
ated GN21,24,25, but it should be emphasized that strong
IgA staining is not present in every case of SAGN.
Immunofluorescence staining intensity is measured on
a scale of 0 to 3+. In our patient cohort, most biopsy
samples had mild to moderate (1 to 2+) IgA stain-
ing (Fig. 4e) and 25% of biopsy samples were negative
or had only trace IgA staining26. These observations
indicate that trace or even absent IgA staining does not
exclude the possibility of SAGN — the disease might
not always be IgA-​dominant. C3 staining is typically
concurrent with and often stronger than the stain-
ing for IgA; C3 staining was predominantly moderate
to strong (2 to 3+) in our biopsy samples26 (Fig. 4f). In
some cases of SAGN (14% in our study), C3 staining
might be mild or absent and the biopsy sample might
thus have a pauci-​immune staining pattern. Cryoglobulin-​
like globular glomerular capillary hyaline thrombi are
rarely seen (detected in three biopsy samples in our
cohort)23,26. Of note, we and others reported a subset
of patients with bacterial infection and concomitant
liver cirrhosis, with strong glomerular immunofluores-
cence staining for IgA, C3 and IgG, and large confluent
immune-​type deposits in the glomerular capillary loops
and the mesangium on electron microscopic examina-
tion86,87. It is possible that chronic liver dysfunction com-
promises the normal mechanism of immune complex
removal from the circulation, leading to the deposition
of large immune complexes in the kidney. Underlying
infection as the cause of GN should be ruled out in
such cases.
Reviews

Table 4 | Differential diagnosis of SAgN on kidney biopsy

Disease light microscopy Direct immunofluorescence electron microscopy
SAGN ∙ Endocapillary hypercellularity common, ∙ C3 stronger than IgA ∙ Mesangial deposits most common.
with exudative pattern in some cases ∙ Sometimes weak to negative IgA , but C3 Few small subendothelial deposits
∙ Focal crescents are frequent, but usually present ∙ 31% of the cases show
necrotizing vasculitis is not seen ∙ Some cases are pauci-​immune subepithelial humps
∙ FSGS pattern not seen ∙ Humps not required for diagnosis
Primary IgA ∙ Mesangial hypercellularity most common IgA usually stronger than C3 ∙ Mesangial deposits most
nephropathy ∙ Endocapillary hypercellularity less common
frequent ∙ Absence of subepithelial humps
∙ Crescents are uncommon, and FSGS ∙ Capillary wall deposits
lesions are frequently seen uncommon
IgA vasculitis ∙ Similar to IgA nephropathy ∙ Moderate to strong IgA ∙ Mesangial deposits most
∙ Segmental endocapillary ∙ Mild or moderate C3 common
hypercellularity more frequent ∙ Similar to IgA nephropathy ∙ Few small subendothelial deposits
∙ Occasional crescents
ANCA vasculitis ∙ Crescents are defining lesions ∙ Negative or weak scant granular IgG Few to absent immune complex
∙ Co-​existence of fibrous, fibrocellular and/or C3. Pauci-​immune pattern deposits
and active crescents ∙ Co-​incidental IgA in rare cases
∙ No endocapillary hypercellularity
∙ Necrotizing arterial lesions may be
present (not seen in SAGN)
Incidental IgA deposits Unremarkable glomeruli Mild IgA , usually no accompanying C3 Absent or few mesangial electron-​
dense immune-​type deposits
APSGN ∙ Diffuse endocapillary hypercellularity ∙ Strong C3 with lumpy-​bumpy coarse ∙ Subepithelial humps almost
common granular pattern always present; usually large
∙ Exudative pattern often seen ∙ IgG may be present in early stages and numerous
∙ Crescents are uncommon ∙ IgA absent ∙ Few mesangial deposits are seen
C3 GN (excluding ∙ Mesangial and endocapillary ∙ Strong C3 and weak to absent IgG. ∙ Mesangial, and capillary wall
dense-​deposit disease) hypercellularity IgA absent deposits
∙ Membranoproliferative pattern frequent ∙ Staining can be global or segmental ∙ Humps may be seen, but are not
∙ Crescents are uncommon but may be mesangial and capillary wall required for diagnosis
focal and segmental ∙ Lumpy-​bumpy C3 deposits may be seen
Cryoglobulinaemic GN ∙ MPGN common ∙ Wide spectrum depending on the type of ∙ Microtubular substructure
∙ Intracapillary inflammatory cells are cryoglobulins, usually mixed IgG and IgM frequently seen
monocytes, not PMNs ∙ IgA staining extremely rare ∙ Deposits can be few and small
∙ Segmental hyaline thrombi often seen, ∙ Scant punctate to abundant glomerular
but not required for diagnosis staining
ANCA, anti-neutrophil cytoplasmic antibody; APSGN, acute post-streptococcal glomerulonephritis; C3, complement protein C3; FSGS, focal segmental
glomerulosclerosis; MP GN, membranoproliferative glomerulonephritis; PMNs, polymorphonuclear cells; SAGN, Staphylococcus infection-associated
glomerulonephritis.

Electron microscopy. Electron microscopic examination
of SAGN biopsy samples frequently shows scattered
electron-​dense deposits in the mesangium (Fig.  4g);
however, occasional subendothelial and small subepithe-
lial deposits can also occur (Fig. 4h). Large subepithelial
‘humps’ might be present but are uncommon and fewer
in number in SAGN than in APSGN. In our patient
cohort, only 31% of samples showed subepithelial
humps26.
Treatment
Treatment of SAGN is aimed at eliminating the under-
lying infection and controlling the symptoms secondary
to acute nephritis23. Cultures of microbial isolates are
essential to determine antimicrobial susceptibility and
identify effective therapies to control the infection. In
the case of S. aureus infections, resistance to methicillin
(that is, identification of MRSA versus MSSA strains)
is particularly relevant. However, antibiotics must be
empirically selected in the absence of a positive micro-
bial culture. In the case of occult deep-​seated infections,
a thorough evaluation is usually required to identify
the source of infection; such deep infections might
also require a prolonged antibiotics course of up to
6 weeks.
In addition to clinical examinations and blood cul-
tures, common diagnostic studies include transtho-
racic echocardiogram or, if required, trans-​oesophageal
echocardiogram, abdominal and/or pelvic computed
tomography (CT) scan, magnetic resonance imaging
(MRI) and X-​ray (including dental panoramic X-​ray,
also known as panorex) (Fig. 5).
As with APSGN, supportive care of patients with
SAGN includes the management of hypertension and
fluid overload with diuretic therapy and salt restriction.
Calcium channel blockers are also preferred as anti-
hypertensive drugs, but, in patients with stable renal
function, ACE inhibitors and angiotensin receptor
blockers might be alternative options. Dialysis might
also be necessary to correct electrolyte and metabolic
disturbances. Contrary to APSGN, immunosuppression
should be avoided during active infection in SAGN as
it might hamper clearance of the underlying infection
and increase the risk of death47,68,77,88–92. Moreover, the
risk of developing persistent renal disease and ESRD in
patients who received immunosuppression and in those

www.nature.com/nrneph

who were treated with supportive care alone was sim-
ilar, suggesting that immunosuppression might not be
an effective treatment for SAGN. In the Nasr et al. study,
in which approximately half of the patients had SAGN,
renal injury only resolved in 14% of patients treated
a b
c d
e f
g h
Fig. 4 | histological features in SAgN. a–d | Light microscopy images of Staphylococcus
infection-​associated glomerulonephritis (SAGN) biopsy samples stained with haematoxylin
& eosin (H&E), 400 × magnification: mesangioproliferative lesion with focal endocapillary
hypercellularity (part a); global endocapillary hypercellularity (part b); underlying nodular
diabetic mesangial expansion with superimposed mesangial hypercellularity (part c);
exudative pattern with global endocapillary neutrophilic infiltration (part d). e | Direct
immunofluorescence staining for IgA shows mild (1+) granular mesangial staining, 400 ×
magnification. f | Direct immunofluorescence staining for C3 shows strong (3+) mesangial
and capillary wall granular staining, 400 × magnification. g, h | Electron microscopy images
stained with uranyl acetate: mesangial electron-​dense immune-​type deposits, 25,000 ×
magnification (part g); subepithelial electron-​dense immune-​type deposits, 30,000 ×
magnification (part h).
Nature Reviews | Nephrology
Reviews
with corticosteroids. Additionally, 18% of patients
died because of recurrent sepsis47. In another study
of patients with endocarditis-​related SAGN, immu-
nosuppression did not improve renal recovery and
mortality was higher among patients treated with corti-
costeroids (23.5%) than in those treated with antibiotics
alone (10%)5.
These observations raise the important question of
what to do after the infection has been eradicated but
renal dysfunction persists. One possibility is to biopsy
the patient after they have completed an appropriate
course of antibiotics to determine if nephritis is still
active. Renal dysfunction might persist because of per-
manent kidney damage sustained while the infection
was still active. In cases in which inflammation persists
after the bacterial insult has been eliminated, a course
of corticosteroids could be considered to limit further
damage to the kidneys but should be used with caution
and close monitoring for infection relapse.
Outcome
Most available studies combined post-​infectious GN
with SAGN and reported cumulative outcomes, rather
than evaluating renal prognosis in patients with SAGN
alone. Extrapolating from these studies, approximately
40–77% of patients with SAGN sustain persistent
kidney damage and up to 43% of patients progress to
ESRD47,68,74,77,88,89. In theory, eradication of the infection
should resolve the GN, but available data indicate that
this is not often the case. Risk factors that might con-
tribute to poor renal outcomes in these patients include
a history of diabetes mellitus, presence of diabetic glo-
merulosclerosis, presence of elevated serum creatinine
levels before presentation, marked interstitial fibrosis
on kidney biopsy and advanced age47,77,93. In cases of
SAGN due to subacute or chronic deep-​seated infec-
tions, diagnosis and treatment with appropriate anti-
biotics is often delayed, allowing chronic damage to
accrue. Moreover, antibiotics might take a long time
to effectively eradicate these deep-​seated infections,
especially if they are associated with prosthetic or
implanted devices.
Endocarditis-​associated GN
Epidemiology
Endocarditis has historically been divided into subacute
bacterial endocarditis and acute bacterial endocarditis4.
These terminologies are no longer recommended
because the natural history of infective endocarditis has
been dramatically altered by the use of powerful antibi-
otic therapy, but they have been retained for this Review
to illustrate the changing disease trends. Of note, bacte-
rial endocarditis is now also preferably termed infective
endocarditis to reflect the fact that, although bacterial
infections are the most common cause of endocarditis,
other infectious organisms, including viruses and fungi,
are also implicated5,85,94–97.
Subacute bacterial endocarditis and renal disease.
Subacute bacterial endocarditis usually affects previ-
ously damaged heart valves and the bacterial organ-
isms involved are often oral cavity commensals with
Reviews

Patient presents with signs Risk factors for Staphylococcus infection

of acute glomerulonephritis
Active staphylococcal
Yes infection confirmed?
• Older age
• History of diabetes mellitus
• Intravenous drug use
• Post-surgery patient
• Presence of implantable or indwelling
No medical devices

SAGN Active infection

likely Yes suspected? No

• Treat with Diagnostic tests to consider • Serological work

appropriate • Blood, urine and/or wound culture up as indicated
antimicrobial • Chest X-ray • Perform kidney
therapy based on • Transthoracic echocardiogram biopsy
source of infection (TEE if appropriate)
• Avoid • Abdomen and/or pelvic CT scan
immunosuppression • Panorex

Source of infection
Yes identified? No

Is the biopsy
Yes consistent with SAGN? No

• Identify source of infection SAGN

• Initiate antimicrobial therapy unlikely

Fig. 5 | Algorithmic approach to the management of SAgN. Appropriate antibiotic therapy is the mainstay of treatment
for Staphylococcus infection-​associated glomerulonephritis (SAGN), as the infection is usually active and ongoing.
Staphylococcus aureus infections range from superficial skin infections such as cellulitis and boils, to deep-​seated occult
infections such as endocarditis, osteomyelitis, visceral abscesses and pneumonia. The duration of antibiotics treatment
varies depending upon the nature of the infection, but prolonged courses of up to 6 weeks are required to treat deep-​
seated infections. Aggressive immunosuppressive therapy is contraindicated at least in the early stages of disease, owing
to the possibility of an active ongoing infection. Imaging studies are frequently needed to find the source of infection, as
blood cultures may be negative in cases of occult infection. CT computed tomography ; Panorex, dental panoramic X-​ray ;
TEE, trans-​oesophageal echocardiogram.

Osler nodes
Painful red, raised lesions
found on hands and feet,
associated with a number of
conditions including infective
endocarditis, caused by
immune complex deposition.
Janeway lesions
Non-​tender, small
erythematous or haemorrhagic
macular or nodular lesions on
the palms or soles only a few
millimetres in diameter that
are indicative of infective
endocarditis (in contrast
to Osler’s nodes, which
are painful).
Splinter haemorrhages
Tiny red blood clots running
vertically under the nails. These
can be associated with multiple
autoimmune conditions
including infective endocarditis,
scleroderma, systemic lupus
erythematosus, rheumatoid
arthritis and anti-​phospholipid
syndrome.
low virulence, such as viridans group streptococci,
Streptococcus mutans and the HACEK group of fastid-
ious Gram-​negative bacilli (that is, Haemophilus spp.,
Aggregatibacter spp., Cardiobacterium hominis, Eikenella
corrodens and Kingella spp.)4,5,85. In the pre-​antibiotic
era, valvular deformities secondary to congenital heart
disease or rheumatic fever were common causes of
subacute endocarditis and mainly affected paediatric
patients.
Subacute bacterial endocarditis tends to have a
less destructive but protracted clinical course than
acute bacterial endocarditis and usually involves the
left-​sided heart valves (commonly the mitral valve).
Clinical stigmata such as Osler nodes, Janeway lesions
and splinter haemorrhages are commonly associated with
the condition. Renal lesions due to subacute endocar-
ditis were originally thought to be secondary to renal
infarction and abscess formation attributed to emboli-
zation of valvular vegetations85,97. Diagnostic and technical
advances later demonstrated the immune nature of the
glomerular injury98–104, which is characterized histolo­
gically by diffuse exudative endocapillary proliferative
lesions or membranoproliferative GN (MPGN) lesions and
crescents with few immune complex deposits. True renal
abscesses and/or infarcts may also occur but are more
common in the terminal stages of the disease and are
more likely to be observed in autopsy pathology than in
renal biopsy practice5,85.
Acute bacterial endocarditis and renal disease. Acute
bacterial endocarditis usually involves previously healthy
heart valves and the bacterial organism is highly virulent,
frequently MRSA5,26,85. The infection typically involves
the right-​sided heart valves (mainly the tricuspid valve)
and can lead to the formation of destructive valvular
abscesses that often require surgical replacement of the
valve85. Most renal biopsy samples with endocarditis-​
associated GN currently seen in renal pathology practice
are secondary to acute bacterial endocarditis105–108.
Various host factors contribute to the predominance of
acute bacterial endocarditis. For example, the rise in elderly
patients with co-​morbidities such as diabetes mellitus
and atherosclerotic heart disease who frequently require
artificial heart valves, other intra-​cardiac devices, and
long-​term indwelling catheters. The current opioid epi-
demic, often associated with direct bloodstream infections,
is another important contributing factor that is associated
with a rise in hospitalizations for endocarditis, mostly
related to MRSA104–107. According to data on IVDU patients
admitted to one hospital in the USA, 61% of cases of endo-
carditis and 57% of bacteraemia cases were related to
S. aureus, which was also the leading cause of endocarditis
episodes in a Dutch cohort of IVDU patients108.
However, the exact burden of kidney disease among
patients with infective endocarditis is not known. In a
biopsy-​based study of endocarditis-​associated GN, Boils
et al. reported that 14% of patients were below 30 years

www.nature.com/nrneph

Valvular vegetations
Mass of fibrin, platelets,
inflammatory cells and
microcolonies of
microorganisms collecting on
heart valvular surfaces in
infective endocarditis. Sterile
vegetations may also occur
(absence of microorganisms
in them) in systemic lupus
erythematosus.
Membranoproliferative GN
Glomerular inflammatory
disease with hypercellularity
in the expanded mesangium
as well as in the intracapillary
lumina, imparting a nodular–
lobular appearance to the
glomerulus. This can have
many different aetiologies.
Nature Reviews | Nephrology
of age and that S. aureus was the underlying infec­tious
organism in 53% of cases5,85. In our single-​centre cohort
of patients with SAGN, 19% of patients were below
40 years of age, and 73% of those patients had a history
of right-​sided endocarditis associated with IVDU26.
These findings suggest that among young adults with
infection-​associated GN, infective endocarditis might be
the underlying cause in a substantial proportion of cases.
Clinical and laboratory findings
The most common clinical presentation of infective
endo­carditis-​associated GN is AKI, reported in 82% of the
patients in the aforementioned study by Boils et al.5,85.
The same study reported a lower prevalence (8%) of
acute nephritic syndrome (that is, haematuria, hyper­
tension and renal failure). This is a unique finding,
because, in patients with active GN, haematuria and
proteinuria are usually the presenting features. AKI
in patients with endocarditis-​associated GN could be
attributed to the profound compromise of cardiac func-
tion in endo­carditis patients. Up to 60% of patients had
low serum levels of complement proteins (mainly C3).
Some patients with endocarditis also have positive auto­
immune serologies, mainly ANCA (reportedly present
in 18–33% of patients)109–112 and sometimes antinuclear
anti­bodies (ANA), mixed cryoglobulins, rheumatoid
factor and anti-​phospholipid antibodies85,102. In the series
by Boils et al.5,85, among the patients who underwent
serological testing for ANCA (32/62), 25% were ANCA
positive, with either a cytoplasmic ANCA or perinu-
clear staining pattern. ELISA testing frequently detected
ANCA specific to proteinase 3 (PR3) and, in some cases,
both PR3-specific and myeloperoxidase (MPO)-specific
antibodies were detected. The presence of these antibod-
ies can cause serious diagnostic pitfalls (see discussion
on the differential diagnosis of SAGN).
GN due to typical subacute bacterial endocarditis is
becoming rare, but we occasionally encounter biopsy
samples from patients with left-​sided endocarditis
caused by low virulence Streptococci. These patients
might not have the classical clinical stigmata and fevers
may be absent, but anaemia, splenomegaly and purpuric
rash might be important clues suggesting an underlying
infection. Mortality among patients with endocarditis
ranges from 40 to 50% and outcomes have not improved
much over the past two decades85,113,114.
Patients with endocarditis-​associated GN might
have negative blood cultures (prevalence ranges from
2.5 to 31%) 115,116. Moreover, in one report, 19% of
culture-​negative endocarditis patients were afebrile115.
Therefore, trans-​oesophageal echocardiograms, which
are considered to be more sensitive than transthoracic
echocardiograms, are strongly recommended along with
repeated blood cultures. However, according to autopsy
studies, the introduction of echocardiography has not
reduced the undetected diagnosis rate85.
Pathology
The glomerular pattern of injury in endocarditis-​
associated GN is variable and similar to that of SAGN
owing to infections other than endocarditis (Table 2).
Importantly, focal crescents and necrotizing lesions are
Reviews
more frequently described in endocarditis-​associated
GN4,5,85,92. One study reported crescents in 47% of their
biopsy samples, which were either focal (19%) or diffuse
(28%)5, and another92 reported a pauci-​immune crescen-
tic pattern in biopsy samples from two-​thirds of their
patients. In our series of patients with SAGN, among the
18 patients with endocarditis, 9 had crescents (50%)26.
Diffuse or focal endocapillary hypercellularity is also
common in endocarditis-​associated GN and some cases
have an exudative pattern26. A membranoproliferative
pattern is common in subacute endocarditis owing to
persistent long-​standing glomerular injury4.
Immunofluorescence staining shows a variable
degree of deposits, predominantly comprising C3, with
or without IgG. Patients with staphylococcal endo­
carditis might also have mild to moderate IgA deposits,
although IgA staining was absent in 30% (6/18) of our
staphylococcal endocarditis biopsy samples26; however,
a pauci-immune pattern might also be seen. Boils et al.5
reported IgA staining in only 29% of their cases of
endocarditis-associated GN; the remaining cases were
pauci-immune.
Electron microscopy might detect some small mesan-
gial and capillary wall immune-​type deposits, but sub­
epithelial humps are uncommon (reported in 18% of
cases by Boils et al.).
Treatment and outcome
Treatment of endocarditis associated GN is the same as
that described for SAGN. However, it is crucial to differ-
entiate endocarditis-​associated GN with positive ANCA
serology from true ANCA-​associated GN, because the
latter requires aggressive immunosuppressive therapy117.
By contrast, immunosuppression is contraindicated in
endocarditis-​associated GN, even in cases of necrotiz-
ing and crescentic GN, as it does not improve renal out-
comes and is associated with higher mortality compared
with supportive care alone5. Instead, patients with endo-
carditis-​associated GN require prompt treatment with
long-​term intravenous antibiotics to eliminate the infec-
tion and for supportive treatment of the acute nephri-
tis. Valve replacement may also be necessary in severe
cases118. Similar to patients with SAGN, renal recovery
following endocarditis-​associated GN is often incom-
plete and persistent kidney damage or ESRD develops in
up to 50% of cases113,114. The risk factors associated with
chronic kidney damage in endocarditis-​associated GN
are similar to those associated with SAGN. Additionally,
patients with endocarditis-​associated GN may develop
heart failure with reduced ejection fraction, which
might increase their risk of permanent kidney damage.
In younger patients with IVDU-​related endocarditis,
recovery of renal function is possible if diagnosis and
start of appropriate antibiotics are not delayed; cessation
of illicit drug abuse and the degree of valvular damage
are also important factors.
Differential diagnosis of infection-​associated GN
It is very important to consider the differential diagnostic
entities because treatment choices (for example, whether
or not to use immunosuppression) depend on whether the
disease might be driven by infection or autoimmunity.
Reviews
Serum cryoglobulins
Immune complexes that
precipitate at temperatures
below normal body
temperature.
IgA nephropathy and/or IgA vasculitis
Distinguishing SAGN from IgAN and/or IgA vasculitis
is the most frequently encountered diagnostic dilemma,
primarily because of the characteristic IgA immune
deposits seen in SAGN and because IgAN is the most
common cause of GN worldwide119. Importantly, active
IgAN is treated with immunosuppression, whereas
immunosuppression is best avoided in SAGN because
it can aggravate the infection and lead to septic com-
plications88–91. Although the peak incidence of primary
IgAN presentation is in the second and third decades of
life, a smaller subset of patients present later in life and
therefore age at biopsy is not a distinguishing feature4,26.
Recent onset of AKI with nephrotic-​range proteinuria
and low serum C3 levels favour a SAGN diagnosis over
primary IgAN. However, serum C3 levels might be nor-
mal in SAGN; thus, normal serum C3 does not exclude
its diagnosis. Up to 20% of SAGN cases present with
purpuric skin rash due to leukocytoclastic vasculitis.
Because IgA vasculitis is rare in adults, the possibility of
SAGN should be considered when an adult patient pre-
sents with IgA vasculitis-​like symptoms and before com-
mencing immunosuppressive therapies74–76. Histology
can provide useful clues for a differential diagnosis but
overlap can occur (Table 4).
APSGN versus SAGN
Although they are both infection-​associated GNs, it is
important not to label SAGN as ‘post’-staphylococcal
GN, as reported in some of the early literature, as it might
suggest to the nephrologist that the infection has already
resolved120–123. Such an assumption might deprive the
patient of the appropriate antibiotic treatment or even
worse, it might prompt initiation of immunosuppressive
therapy. The presence of IgA staining in the immune
deposits favours a diagnosis of SAGN.
ANCA GN
SAGN biopsy samples might show focal crescents and/
or segmental necrotizing glomerular lesions, as typically
seen in ANCA vasculitis4,26. In a subset of SAGN cases,
immunofluorescence and electron microscopy might
show a pauci-​immune pattern5,26. Additionally, 22% of
our SAGN biopsy samples and 28% of the endocarditis-​
associated GN samples analysed by Boils et al. tested
positive for ANCA5,26,85. In such a scenario, SAGN might
be mistakenly diagnosed as ANCA-​associated GN.
Conversely, nasal carriage of S. aureus or overt infec-
tion are suspected to have a role in the development
of ANCA-​associated granulomatosis with polyangiitis,
which creates a complex pathogenetic dilemma81–84,124–126.
As previously discussed, distinguishing between
SAGN and true ANCA-​associated GN is crucial because
immunosuppressive therapy might be dangerous in
SAGN. The major diagnostic criteria for SAGN and risk
factors for infection (Box 1) should be carefully consid-
ered, and infection should be reasonably excluded before
immunosuppressive therapy is implemented. Subtle his-
tological clues that favour SAGN over ANCA-​associated
crescentic GN include the presence of segmental endo-
capillary hypercellularity in the non-​crescentic glomer-
uli, C3 staining with or without IgA, and the presence of
at least a few immune-​type deposits by electron micro­
scopy4,26. Conversely, presence of fibrinoid necrosis in
vascular walls favours a diagnosis of ANCA-​associated
crescentic GN4,26 (Table 4).
C3 GN
Active ongoing infection is usually absent in true C3
GN, although rare cases can be triggered by infection,
especially streptococcal infection, and therefore C3 GN
might masquerade as APSGN127,128. In the context of
underlying abnormalities in regulatory proteins of the
alternative complement pathway, infection might pro-
vide a ‘second hit’ that triggers uncontrolled activation
of the alternative complement pathway and persistent
complement-​mediated glomerular injury. In contrast to
typical APSGN, C3 GN is characterized by long-​lasting
low serum C3 levels, renal dysfunction, haematuria and
proteinuria, as the symptoms persist long after the epi-
sode of infection has resolved. In such cases, genetic test-
ing for alternative complement pathway abnormalities is
recommended.
A subset of SAGN biopsy samples show dominant
C3 staining with weak to absent IgA and IgG, and
some might also show an MPGN pattern of glomeru-
lar injury26,47. In such cases, SAGN might be mistakenly
diagnosed as C3 GN. Distinguishing features include
the protracted course and intermittent flares seen in C3
GN127,128, which contrast with SAGN. The latter usually
presents as a single episode of AKI with nephrotic-​range
proteinuria4,26,47.
Cryoglobulinaemic GN
Cryoglobulinaemic GN is a form of proliferative GN
associated with the presence of serum cryoglobulins ,
with or without deposition in the glomerular capillar-
ies. Cryoglobulins are classified into three types: type I
(monoclonal), type II (polyclonal) and type III (poly­
clonal)129; type III cryoglobulins can be detected in
chronic bacterial infections 129. We identified three
biopsy samples in our series of 78 cases of SAGN that
contained large, IgA-​containing, cryoglobulin-​like
immune complex deposits4,23,26. However, these depos-
its did not have the microtubular substructure typi­
cally seen in type II cryoglobulinaemic GN. In most
instances, SAGN can be differentiated from cryoglobu­
linaemic GN using a combination of kidney biopsy,
serological and clinical findings. True cryoglobulinae-
mic GN is characterized by distinct glomerular hyaline
thrombi staining for IgG and IgM, normal serum C3
but low C4, and a positive cryoglobulin test and elevated
rheumatoid factor.
Incidental glomerular IgA deposits
In up to 3% of the Western population, light microscopy
might detect mild to moderate incidental IgA staining
in glomeruli with otherwise normal appearance, with
or without discrete deposits detectable by electron
microscopy130. The prevalence of this phenotype is even
higher in Asia (for example, 24% in China and 16% in
Japan)131,132. These deposits might not be clinically rele-
vant. Considering their high prevalence in the general
population, such incidental glomerular IgA deposits
www.nature.com/nrneph
 Reviews

might also be coincidently detected in patients with proposed antigens are NAPlr and SPEB25–28,142,124–129. SPEB
other glomerular diseases, including ANCA-​associated has been detected in the subepithelial humps121 and anti-
GN. In such instances, it is also important to carefully bodies against both proteins were also detected in the
consider all the relevant clinical data and risk factors toconvalescent sera of patients with streptococcal infection
avoid erroneously attributing the incidental IgA deposits and APSGN4.
to infection-​associated GN. NAPlr was isolated from the cytoplasmic fraction
of GAS proteins through affinity chromatography,
Pathogenesis of infection-​associated GN using IgGs present in convalescent sera of patients with
All forms of infection-​associated GN are thought to APSGN138,139. NAPlr binds to plasmin with high affin-
result from immune-​mediated glomerular injury trig- ity and retains it in its active form by protecting it from
gered by systemic infection (usually extra-​renal), but its physiological inhibitors. Active plasmin can cause
the precise pathogenetic mechanisms are still unclear. glomerular capillary damage by degrading extracel-
Nephritogenic fractions including intracellular, plasma lular matrix proteins such as fibronectin, laminin and
membrane and extracellular secretory bacterial pro- fibrinolysis — either by itself or through the activation
teins are implicated in the pathogenesis of infection-​ of MMPs. Plasmin can also mediate inflammation by
associated GN 133–140. Antibodies to these bacterial attracting and activating monocytes and neutrophils. Of
antigens are thought to form immune complexes in note, urinary plasmin activity is higher in patients with
the circulation, followed by immune complex deposi- APSGN than in healthy individuals138,139.
tion in the glomerular capillary tuft, which results in One study reported that, in a cohort of 50 patients with
glomerular inflammation, complement activation and APSGN, glomerular deposition of NAPlr was observed
capillary injury4. Deposition of immune complexes can in 100% of patients within 2 weeks of disease onset, and in
result in activation of complement pathways, release of 84% of patients within 30 days of disease onset, indi-
chemotaxins C3a and C5a, recruitment of inflammatory cating that antigen deposition decreases over time138.
cells, cytokine production and release of reactive oxy- Healthy kidney tissue did not stain for NAPlr and only
gen species, plasmin and MMP enzymes — collectively, 4% of samples from patients with other glomeruloneph-
these immune mediators damage the capillary tuft4,139,140. ritides were positive for NAPlr. NAPlr was detected on the
Other postulated mechanisms underlying the formation subendothelial aspect of the glomerular basement mem-
of immune complexes in the kidney include in situ for- brane (GBM) and in endocapillary neutrophils in patients
mation of immune complexes. Antibodies might bind with APSGN, but not in neutrophils from patients with
to bacterial antigens planted in the glomerular capillary rapidly progressive crescentic GN, thus exhibiting
tuft, or bacteria-​specific antibodies might cross-​react some disease specificity. NAPlr+ neutrophils might
with intrinsic glomerular matrix proteins such as colla- have a role in the induction of proteolytic glomerular
gen, laminin and heparan sulphate, as a result of molec- damage138,139.
ular mimicry4,133,136. The theory of molecular mimicry SPEB, the other potential pathogenic antigen described
between bacterial antigens and endogenous glomerular for APSGN, is a cationic cysteine protease secreted as a
matrix proteins was proposed based on the identifica- zymogen140. The pathological role of SPEB is postulated to
tion of antibodies to these matrix proteins in the sera of result from its ability to convert plasminogen to plasmin,
patients with APSGN137. Cationic antigens are especially suggesting that promoting the activity of plasmin might
implicated because they can cross the polyanionic charge be a general pathway in the pathogenesis of APSGN. The
barrier of the glomerular capillary filter and then bind interaction of SPEB with plasminogen is also thought to
to components of the outer basement membrane, form- cleave C3 and activate the alternative complement path-
ing subepithelial humps133,134. This process explains why way, initiating a glomerular immune response. SPEB can
glomerular hypercellularity occurs within the capillary also directly bind to major histocompatibility complex
loops in close proximity to the endothelial surface, even (MHC) class II molecules on antigen-​presenting cells and
though the immune complex deposits (termed humps) to specific Vβ chain of T cell receptors, leading to massive
are subepithelial in location134. These mechanisms lead- activation of T cells.
ing to immune complex formation and deposition in Despite all this experimental evidence, there are seve­
the kidney may explain the latency between the onset of ral pitfalls in our understanding of the pathogenesis of
infection and the development of GN and in some ways APSGN. For instance, circulating immune complexes
resemble the animal models of acute serum sickness141. are not only detected in patients with APSGN, but also
Postulated streptococcal target antigens involved in patients with infection but without GN and they do
in the pathogenesis of APSGN include streptococcal not correlate with disease activity. Several of the bacterial
M protein (derived from the bacterial cell membrane); proteins postulated to have a role in APSGN are also
endostreptosin, also termed preabsorbing antigen present in non-​nephritogenic Streptococci that do not
(derived from the bacterial cytoplasm)135; SPEB, also lead to APSGN. It is possible that nephritogenic antigens
known as nephritis strain-​associated protein (NSAP) or vary in different populations. Importantly, there is still
nephritis plasmin-​binding protein (NPBP)(identified no proof that the antigens present within the immune
as an extracellular protein in bacterial cultures)140; and complexes directly cause APSGN4. Interestingly, NAPlr
NAPlr, which is homologous to streptococcal glycer- in the glomeruli was detected on the mesangial and
aldehyde-3-phosphate dehydrogenase (GAPDH)138,139. subendothelial surface of the GBM138,139; therefore, it is
Many of these proteins have been detected in the glo- not clear how it contributes to the formation of the sub-
meruli during the early phases of GN. Two of the main epithelial humps. Host–pathogen interactions are also

Nature Reviews | Nephrology

Reviews
Cytokine storm
Secretion of large quantities
of activating compounds
(cytokines) due to a surge in
activated immune cells, usually
T lymphocytes.
likely to have an important role and inter-​individual var-
iation is influenced not only by the general health of the
patient, but also by other host-​specific factors, including
genetic variation. Of note, glomerular NAPlr deposition
and plasmin activity were also detected in other glomer-
ular diseases in which recent streptococcal infection was
suspected based on serological testing, including dense
deposit disease, IgA vasculitis and MPGN. These find-
ings are difficult to reconcile, but hint at the notion that
despite many differences, there may also be similarities
in the multi-​step pathogenesis of immune-​mediated
glomerulonephritides as a group1,139.
Compared with APSGN, fewer studies have focused
on the disease pathogenesis of the more recently
described SAGN. Staphylococcal enterotoxins, typi­cally
enterotoxin C, A and toxic shock syndrome toxin 1 were
postulated to act as superantigens and to have an impor-
tant role in disease65,68–73,143. Superantigens activate T cells
by directly binding to MHC class II molecules on anti-
gen-​presenting cells without intracellular processing and
to the Vβ region of the T cell receptor (TCR), irrespective
of TCR antigen specificity144. This non-​antigen specific
binding leads to the activation of large subsets of poly-
clonal T cells, causing a cytokine storm. Activated T cells
then stimulate B cell activation, proliferation and anti-
body production, promoting damage in the glomerular
tuft144. One study of patients with SAGN that presented
as IgA vasculitis-​like clinical syndrome demonstrated
that TCR Vβ chain usage was skewed in patients with
SAGN compared with healthy controls, or with patients
in whom the staphylococcal infection resolved without
triggering GN71. Serum levels of IL-1β, IL-2, IL-6, IL-8
and TNF were also higher in patients with SAGN com-
pared with control groups, suggesting polyclonal T cell
expansion and enhanced immune activation.
Another hypothesis is that SAGN might involve over-​
activation of neutrophils and widespread formation of
neutrophil extracellular traps (NETs) to immobilize
invading microorganisms in response to disseminated
infection145,146. NET release might also expose intracel-
lular neutrophil antigens and promote the generation
of autoantibodies that can further propagate damage to
the glomerular tuft145,146. However, host–pathogen inter-
actions and variability in the host immune response to
bacterial antigens are probably the most crucial factors,
as GN does not develop in all patients with infection.
Compromised host immunity because of advanced age
and other co-​morbidities, the inability of the immune
system to control the infection and persistent or over-
whelming antigenaemia (for example, in the context
of repeated IVDU with contaminated needles) might
be important host-​related susceptibility factors. The
affinity of staphylococcal superantigens to certain
subsets of HLA molecules might also increase patient
susceptibility to staphylococcal infection and SAGN.
In a genome-​wide association study on biological spec-
imens from 4,701 culture-​confirmed cases and 45,344
matched controls, several single-​nucleotide polymor-
phisms on the HLA class II region of chromosome 6
were enriched in the infection group147. These polymor-
phisms were located near the genes encoding HLA-​DRA
and HLA-​DRB1.
GN driven by other infections
Although Streptococcus and Staphylococcus are the
most common pathogens associated with GN, several
other bacteria and non-​bacterial microorganisms can
trigger immune-​mediated GN148. These include Gram-​
positive anaerobes such as Propionibacterium acnes,
Gram-​negative bacteria such as Neisseria gonorrhoeae,
Pseudomonas spp., Brucella spp., Coxiella burnetii,
Yersinia spp., Legionella spp. and Bartonella spp., as well
as Mycoplasma spp. and Treponema spp. Parasitic infec-
tions, such as schistosomiasis, malaria, leishmaniasis and
strongyloidiasis, are also implicated in the development
of proliferative GN (frequently with a membranopro-
liferative pattern or MPGN). More than one-​third of
patients might lack evidence of infection148. Symptoms
of infection might be subtle and, although serum C3
levels are low in 35 to 80% of adults, normal serum
complement levels do not preclude the diagnosis4,26.
Autoimmune serologies such as ANA, rheumatoid factor
and mixed cryoglobulins might also be positive, espe-
cially in chronic infections. Although viral infections
such as hepatitis B, hepatitis C and varicella are known to
cause GN1,4, they are not conventionally included under
the heading of infection-​related glomerulonephritides
and are beyond the scope of this Review.
Shunt nephritis
Shunt nephritis became a rare GN after ventriculo-atrial
shunts were replaced by ventriculo-​peritoneal shunts for
the treatment of hydrocephalus4. However, GN related to
central venous catheter infections has now emerged, espe-
cially in patients who require long-​term hyperalimenta-
tion149. Glomerular morphology can vary in these cases,
but an MPGN pattern is frequently seen, with rare cres-
cents. Immunofluorescence usually shows C3-dominant
coarsely granular staining. Electron microscopy can show
scattered mesangial and few subendo­thelial and sub­
epithelial electron-dense immune-type deposits with or
without subepithelial humps4.
Conclusions
Global studies and statistical analyses indicate that the
incidence of APSGN is declining worldwide, although
it remains high in tropical, densely populated coun-
tries and is disproportionately high among people
with lower incomes. APSGN is still the most common
cause of acute GN in children worldwide. In adults,
SAGN is 3–4 times more common than APSGN and
is increasingly encountered in the elderly population.
Among these patients, SAGN is a source of substan-
tial co-​morbidity due to progression to chronic kidney
disease. Diagnosis of SAGN can be difficult, as clinical
presentation and kidney biopsy findings are not unique
and can resemble other autoimmune glomeruloneph-
ritides. Unfortunately, expensive imaging studies and
prolonged antibiotic courses are needed in the manage-
ment of SAGN. Early aggressive treatment with appro-
priate antibiotics, younger age and the absence of other
co-​morbidities such as diabetes portend a favourable
renal outcome for patients. However, antibiotics such
as vancomycin might themselves contribute to nephro-
toxicity if doses are not carefully titrated and trough
www.nature.com/nrneph
 Reviews

levels are high150. Overall, patient outcomes in SAGN
remain poor.
Future research should investigate non-​invasive
imaging techniques for the detection and localization
of S. aureus infections, especially considering the often
deep-​s eated nature of these infections. A nuclease-​
activated probe for the rapid detection of infection has
been reported in animal experiments, but is probably
not yet ready for application in humans151. Drugs that
reduce bacterial biofilm formation might also be an
important step towards reducing the virulence of these
bacteria and the use of potentially nephrotoxic antibiot-
ics152. More studies on the staphylococcal agr quorum-​
sensing system, involved in the regulation of bacterial
virulence factors, might also help in the development of
anti-​virulence therapeutics153. Identification of bacterial
antigens implicated in infection-​triggered GN has been
an ongoing effort, but it should be emphasized that these
antigens might be different in different regions and
populations of the world.
Reinforcing efforts for effective primary prevention
of streptococcal infections among under-​privileged
and socially disadvantaged communities through
school and community-​based surveillance programmes
remains a top priority. Similarly, control of staphylococ-
cal infections both in the community and in a hospital-
setting, among patients and health care workers, must be
meticulously pursued, along with antibiotics stewardship
programmes, to control the emergence of antibiotics
resistance.
Published online xx xx xxxx

1. Jennette, J. C., Silva, F. G., Olson, J. L., D’Agati, V. D.
(eds) Heptinstall’s Pathology of the Kidney 7th edn
Vol. 1 Ch. 3 91–118 (Wolters Kluwer, 2014).
2. Cook, H. T. & Pickering, M. C. in Heptinstall’s
Pathology of the Kidney 7th edn Vol. 1 Ch. 9 (eds
Jennette, J. C. et al.) 341–366 (Wolters Kluwer, 2014).
3. Satoskar, A. A. & Nadasdy, T. in Silva’s Diagnostic
Renal Pathology 2nd edn Ch. 7 (eds Zhou, X. J. et al.)
243–264 (Cambridge Univ. Press, 2017).
4. Satoskar, A. A., Nadasdy, T. & Silva, F. in Heptinstall’s
Pathology of the Kidney 7th edn Vol. 1 Ch. 10 (eds
Jennette, J. C. et al.) (Wolters Kluwer, 2014).
5. Boils, C. L. in Bacterial Infections and the Kidney
Ch. 4 (eds Satoskar, A. A. & Nadasdy, T.) 87–116
(Springer, 2017).
6. Longcope, W. T. et al. Relationship of acute infections
to glomerular nephritis. J. Clin. Invest. 5, 7 (1928).
7. Dillon, H. C., Reeves, M. S. & Maxted, W. R. Acute
glomerulonephritis following skin infection due to
streptococci of M-​type 2. Lancet 291, 543–545
(1968).
8. Wilmers, M. J., Cunliffe, A. C. & Williams, R. E. Type-12
streptococci associated with acute haemorrhagic
nephritis. Lancet 267, 17–18 (1954).
9. World Health Organization. The current evidence for
the burden of group A streptococcal diseases (WHO,
2005).
10. Carapetis, J. R., Steer, A. C., Mulholland, E. K. &
Weber, M. The global burden of group A streptococcal
diseases. Lancet Infect. Dis. 5, 685–694 (2005).
11. Rodriguez-​Iturbe, B. & Musser, J. M. The current state
of post-​streptococcal glomerulonephritis. J. Am. Soc.
Nephrol. 19, 1855–1864 (2008).
12. Kanjanabuch, T., Kittikowit, W. & Eiam-​Ong, S.
An update on acute postinfectious glomerulonephritis
worldwide. Nat. Rev. Nephrol. 5, 259–269 (2009).
13. World Health Organization. A review of the technical
basis for the control of conditions associated with GAS
infections (WHO, 2005).
14. DeLeo, F. R., Otto, M., Kreiswirth, B. N. &
Chambers, H. F. Community-​associated methicillin-​
resistant Staphylococcus aureus. Lancet 375,
1557–1568 (2010).
15. Tattevin, P., Diep, B. A., Jula, M. & Perdreau-
Remington, F. Long-​term follow-up of methicillin-resistant
Staphylococcus aureus molecular epidemiology after
emergence of clone USA300 in San Francisco jail
populations. J. Clin. Microbiol. 46, 4056–4057 (2008).
16. Tattevin, P., Diep, B. A., Jula, M. & Perdreau-Remington, F.
Methicillin-​resistant Staphylococcus aureus USA300
clone in long-​term care facility. Emerg. Infect. Dis. 15,
953–955 (2009).
17. Nimmo, G. R. USA300 abroad: global spread of a
virulent strain of community-​associated methicillin-​
resistant Staphylococcus aureus. Clin. Microbiol.
Infect. 18, 725–734 (2012).
18. Pan, E. S. et al. Increasing prevalence of methicillin-​
resistant Staphylococcus aureus infection in California
jails. Clin. Infect. Dis. 37, 1384–1388 (2003).
19. King, M. D. et al. Emergence of community-​acquired
methicillin-​resistant Staphylococcus aureus USA 300
clone as the predominant cause of skin and soft-​tissue
infections. Ann. Intern. Med. 144, 309–317 (2006).
20. Chatterjee, S. S. & Otto, M. Improved understanding
of factors driving methicillin-​resistant Staphylococcus
aureus epidemic waves. Clin. Epidemiol. 5, 205–217
(2013).
21. Nasr, S. et al. IgA dominant acute poststaphylococcal
glomerulonephritis complicating diabetic nephropathy.
Hum. Pathol. 34, 1235 (2003).
22. Pola, E. et al. Onset of Berger disease after
Staphylococcus aureus infection: septic arthritis after
anterior cruciate ligament reconstruction. Arthroscopy
19, E29 (2003).
23. Satoskar, A. A. et al. Staphylococcus infection-​
associated glomerulonephritis mimicking IgA
nephropathy. Clin. J. Am. Soc. Nephrol. 1, 1179–1186
(2006).
24. Haas, M., Racusen, L. C. & Bagnasco, S. M.
IgA-dominant postinfectious glomerulonephritis:
a report of 13 cases with common ultrastructural
features. Hum. Pathol. 39, 1309–1316 (2008).
25. Worawichawong, S. et al. Immunoglobulin A-​dominant
postinfectious glomerulonephritis: frequent occurrence
in nondiabetic patients with Staphylococcus aureus
infection. Hum. Pathol. 42, 279–284 (2011).
26. Satoskar, A. A. et al. Staphylococcus infection-​
associated GN — spectrum of IgA staining and
prevalence of ANCA in a single-​center cohort. Clin. J.
Am. Soc. Nephrol. 12, 39–49 (2017).
27. Jackson, K. A. et al. Invasive methicillin-​resistant
staphylococcus aureus infections among persons who
inject drugs — six sites, 2005–2016. MMWR Morb.
Mortal. Wkly Rep. 67, 625–628 (2018).
28. Vosti, K. L., Johnson, R. H. & Dillon, M. F. Further
characterization of purified fractions of M protein from
a strain of group A, type 12 streptococcus. J. Immunol.
107, 104–114 (1971).
29. Nordstrand, A., McShan, W. M., Ferretti, J. J.,
Holm, S. E. & Norgren, M. Allele substitution of the
streptokinase gene reduces the nephritogenic capacity
of group A streptococcal strain NZ131. Infect. Immun.
68, 1019–1025 (2000).
30. Ohkuni, H. et al. Immunological studies of post-​
streptococcal sequelae: serological studies with an
extracellular protein associated with nephritogenic
streptococci. Clin. Exp. Immunol. 54, 185–193 (1983).
31. Anthony, B. F., Kaplan, E. L., Wannamaker, L. W.,
Briese, F. W. & Chapman, S. S. Attack rates of acute
nephritis after type 49 streptococcal infection of the
skin and of the respiratory tract. J. Clin. Invest. 48,
1697–1704 (1969).
32. Rodriguez-​Iturbe, B., Castillo, L., Valbuena, R. &
Cuenca, L. Attack rate of poststreptococcal nephritis
in families: a prospective study. Lancet 317, 401–403
(1981).
33. Poon-​King, T. et al. Recurrent epidemic nephritis in
South Trinidad. N. Engl. J. Med. 277, 728 (1967).
34. Fish, A. J., Herdman, R. C., Michael, A. F., Pickering, R. J.
& Good, R. A. Epidemic acute glomerulonephritis
associated with type 49 streptococcal pyoderma. II.
Correlative study of light, immunofluorescent and
electron microscopic findings. Am. J. Med. 48, 28–39
(1970).
35. Rodriguez-​Iturbe, B. Epidemic post-​streptococcal
glomerulonephritis. Kidney Int. 25, 129–136 (1984).
36. Rodriguez-​Iturbe, B. et al. Epidemic
glomerulonephritis in Maracaibo: evidence for
progression to chronicity. Clin. Nephrol. 5, 197
(1976).
37. Drachman, R., Aladjem, M. & Vardy, P. A. Natural-​
history of an acute glomerulonephritis epidemic in
children — an 11-year to 12-year follow-​up. Isr. J.
Med. Sci. 18, 603–607 (1982).
38. Mori, K., Ito, Y., Kamikawaji, N. & Sasazuki, T.
Elevated IgG titer against the C region of streptococcal
M protein and its immunodeterminants in patients
with poststreptococcal acute glomerulonephritis.
J. Pediatr. 131, 293–299 (1997).
39. Perlman, L. V., Herdman, R. C., Kleinman, H. &
Vernier, R. L. Poststreptococcal glomerulonephritis.
A ten-​year follow-​up of an epidemic. JAMA 194,
63–70 (1965).
40. VanDeVoorde, R. G. 3rd Acute poststreptococcal
glomerulonephritis: the most common acute
glomerulonephritis. Pediatr. Rev. 36, 3–12 (2015).
41. Vogel, A. M. et al. Post-​streptococcal glomerulonephritis:
some reduction in a disease of disparities. J. Paediatr.
Child Health 55, 652–658 (2019).
42. Ramanathan, G. et al. Analysis of clinical presentation,
pathological spectra, treatment and outcomes of
biopsy-​proven acute postinfectious glomerulonephritis
in adult indigenous people of the Northern Territory
of Australia. Nephrology 22, 403–411 (2017).
43. Coppo, R., Gianoglio, B., Porcellini, M. G. &
Maringhini, S. Frequency of renal diseases and clinical
indications for renal biopsy in children (report
of the Italian National Registry of Renal Biopsies in
Children). Group of Renal Immunopathology of the
Italian Society of Pediatric Nephrology and Group
of Renal Immunopathology of the Italian Society of
Nephrology. Nephrol. Dial. Transplant. 13, 293–297
(1998).
44. Shulman, S. T. et al. Clinical practice guideline for the
diagnosis and management of group A streptococcal
pharyngitis: 2012 update by the Infectious Diseases
Society of America. Clin. Infect. Dis. 55, 1279–1282
(2012).
45. Bisno, A. L. et al. Practice guidelines for the diagnosis
and management of group A streptococcal
pharyngitis. Clin. Infect. Dis. 35, 113–125 (2002).
46. Parks, T., Smeesters, P. R., Curtis, N. & Steer, A. C.
ASO titer or not? When to use streptococcal serology:
a guide for clinicians. Eur. J. Clin. Microbiol. Infect. Dis.
34, 845–849 (2015).
47. Nasr, S. H. et al. Postinfectious glomerulonephritis in
the elderly. J. Am. Soc. Nephrol. 22, 187–195 (2011).
48. Mazzucco, G. et al. Different patterns of renal damage
in type 2 diabetes mellitus: a multicentric study on 393
biopsies. Am. J. Kidney Dis. 39, 713–720 (2002).
49. Lien, J. W., Mathew, T. H. & Meadows, R. Acute post-​
streptococcal glomerulonephritis in adults: a long-term
study. Q. J. Med. 48, 99–111 (1979).
50. Gunasekaran, K., Krishnamurthy, S., Mahadevan, S.,
Harish, B. N. & Kumar, A. P. Clinical characteristics
and outcome of post-​infectious glomerulonephritis
in children in Southern India: a prospective study.
Indian J. Pediatr. 82, 896–903 (2015).
51. Rodriguez-​Iturbe, B. et al. Characterization of the
glomerular antibody in acute poststreptococcal
glomerulonephritis. Ann. Intern. Med. 92, 478 (1980).
52. Parikh, S. V., Haddad, N. & Hebert, L. A. in
Comprehensive Clinical Nephrology 5th edn
Ch. 80 (eds Johnson, R., Feehally, J. & Floege, J.)
(Saunders, 2014).

Nature Reviews | Nephrology

Reviews
53. Pavelsky, P. M. Renal replacement therapy in AKI.
Adv. Chronic Kidney Dis. 20, 76–84 (2013).
54. Roy, S. 3rd, Murphy, W. M. & Arant, B. S. Jr
Poststreptococcal crescenteric glomerulonephritis
in children: comparison of quintuple therapy versus
supportive care. J. Pediatr. 98, 403–410 (1981).
55. Melby, P. C. et al. Poststreptococcal
glomerulonephritis in the elderly. Report of a case
and review of the literature. Am. J. Nephrol. 7,
235–240 (1987).
56. Wong, W., Morris, M. C. & Zwi, J. Outcome of severe
acute post-​streptococcal glomerulonephritis in New
Zealand children. Pediatr. Nephrol. 24, 1021–1026
(2009).
57. Zent, R. et al. Crescentic nephritis at Groote Schuur
Hospital, South Africa—not a benign disease.
Clin. Nephrol. 42, 22–29 (1994).
58. Clark, G. et al. Poststreptococcal glomerulonephritis in
children: clinicopathological correlations and long-​term
prognosis. Pediatr. Nephrol. 2, 381–388 (1998).
59. Potter, E. V. et al. Twelve to seventeen-​year
follow-​up of patients with poststreptococcal acute
glomerulonephritis in Trinidad. N. Engl. J. Med. 307,
725–729 (1982).
60. Baldwin, D. S. Poststreptococcal glomerulonephritis.
A progressive disease? Am. J. Med. 62, 1–11 (1977).
61. Buzio, C. et al. Significance of albuminuria in the
follow-up of acute poststreptococcal glomerulonephritis.
Clin. Nephrol. 41, 259–264 (1994).
62. Vogl, W. et al. Long-​term prognosis for endocapillary
glomerulonephritis of poststreptococcal type in
children and adults. Nephron 44, 58–65 (1986).
63. Spector, D. A., Millan, J., Zauber, N. & Burton, J.
Glomerulonephritis and Staphylococcus aureus
infections. Clin. Nephrol. 14, 256 (1980).
64. Danovitch, G. M., Nord, E. P., Barki, Y. &
Krugliak, L. Staphylococcal lung abscess and acute
glomerulonephritis. Isr. J. Med. Sci. 15, 840–843
(1979).
65. Sato, M., Nakazoro, H. & Ofuji, T. The pathogenetic
role of Staphylococcus aureus in primary human
glomerulonephritis. Clin. Nephrol. 11, 190–195 (1979).
66. Maher, E. R., Hamilton, D. V., Thiru, S. & Wheatley, T.
Acute renal failure due to glomerulonephritis
associated with staphylococcal infection. Postgrad.
Med. J. 60, 433–434 (1984).
67. Salyer, W. R. & Salyer, D. C. Unilateral
glomerulonephritis. J. Pathol. 113, 247–251
(1974).
68. Koyama, A. et al. Glomerulonephritis associated
with MRSA infection: a possible role of bacterial
superantigen. Kidney Int. 47, 207–216 (1995).
69. Yoh, K. et al. A case of superantigen-​related
glomerulonephritis after methicillin-​resistant
Staphylococcus aureus (MRSA) infection.
Clin. Nephrol. 48, 311–316 (1997).
70. Kobayashi, M. & Koyama, A. Methicillin-​resistant
Staphylococcus aureus (MRSA) infection in
glomerulonephritis—a novel hazard emerging on the
horizon. Nephrol. Dial Transplant. 13, 2999–3001
(1998).
71. Hirayama, K. et al. Henoch-​Schonlein purpura
nephritis associated with methicillin-​resistant
Staphylococcus aureus infection. Nephrol. Dial
Transplant. 13, 2703–2704 (1998).
72. Yoh, K. et al. Cytokines and T cell responses in
superantigen-​related glomerulonephritis following
methicillin-​resistant Staphylococcus aureus infection.
Nephrol. Dial Transplant. 15, 1170–1174 (2000).
73. Yamashita, Y. et al. Glomerulonephritis after
methicillin-​resistant Staphylococcus aureus infection
resulting in end-​stage renal failure. Intern. Med. 40,
424–427 (2001).
74. Satoskar, A. A. et al. Henoch-​Schönlein purpura-​like
presentation in IgA-​dominant Staphylococcus
infection-​associated glomerulonephritis-​a diagnostic
pitfall. Clin. Nephrol. 79, 302–312 (2013).
75. Montoliu, J. et al. Henoch-​Schonlein purpura
complicating staphylococcal endocarditis in a heroin
addict. Am. J. Nephrol. 7, 137–139 (1987).
76. Kitamura, T., Nakase, H. & Iizuka, H. Henoch-Schonlein
purpura after postoperative Staphylococcus aureus
infection with hepatic IgA nephropathy. J. Nephrol.
19, 687–690 (2006).
77. Hemminger, J. A. & Satoskar, A. A. in Bacterial
Infections and the Kidney Ch. 2 (eds Satoskar, A. A.
& Nadasdy, T.) 37–61 (Springer, 2017).
78. Nasr, S. H., Radhakrishnan, J. & D’Agati, V. D.
Bacterial infection-​related glomerulonephritis in
adults. Kidney Int. 83, 792–803 (2013).
79. Bu, R. et al. Clinicopathologic features of IgA-​
dominant infection-​associated glomerulonephritis:
a pooled analysis of 78 cases. Am. J. Nephrol. 41,
98–106 (2015).
80. Kimata, T., Tsuji, S., Yoshimura, K., Tsukaguchi, H.
& Kaneko, K. Methicillin-​resistant Staphylococcus
aureus-​related glomerulonephritis in a child. Pediatr.
Nephrol. 27, 2149–2152 (2012).
81. Peacock, S. J., de Silva, I. & Lowy, F. D. What determines
nasal carriage of Staphylococcus aureus? Trends
Microbiol. 9, 605–610 (2001).
82. Von Eiff, C., Becker, K., Machka, K., Stammer, H. &
Peters, G. Nasal carriage as a source of Staphylococcus
aureus bacteremia. Study group. N. Engl. J. Med.
344, 11–16 (2001).
83. Wertheim, H. F. et al. Risk and outcome of nosocomial
Staphylococcus aureus bacteremia in nasal carriers
versus non-​carriers. Lancet 364, 703–705 (2004).
84. Foster, T. J. Colonization and infection of the human
host by staphylococci: adhesion, survival and immune
evasion. Vet. Dermatol. 20, 456–470 (2009).
85. Boils, C. L., Nasr, S. H., Walker, P. D., Couser, W. G.
& Larsen, C. P. Update on endocarditis-​associated
glomerulonephritis. Kidney Int. 87, 1241–1249
(2015).
86. Hemminger, J. et al. Acute glomerulonephritis with
large confluent IgA-​dominant deposits associated
with liver cirrhosis. PLOS ONE 13, e0193274
(2018).
87. Takada, D. et al. IgA nephropathy featuring massive
wire loop-​like deposits in two patients with alcoholic
cirrhosis. BMC Nephrol. 18, 362 (2017).
88. Nagaba, Y. et al. Effective antibiotic treatment of
methicillin-​resistant Staphylococcus aureus-​
associated glomerulonephritis. Nephron 92,
297–303 (2002).
89. Riley, A. M., Wall, B. M. & Cooke, C. R. Favorable
outcome after aggressive treatment of infection in a
diabetic patient with MRSA-​related IgA nephropathy.
Am. J. Med. Sci. 337, 221–223 (2009).
90. Kapadia, A. S., Panda, M. & Fogo, A. B. Postinfectious
glomerulonephritis: is there a role for steroids? Indian
J. Nephrol. 21, 116–119 (2011).
91. Zeledon, J. I. et al. Glomerulonephritis causing acute
renal failure during the course of bacterial infections.
Histological varieties, potential pathogenetic pathways
and treatment. Int. Urol. Nephrol. 40, 461–470
(2008).
92. Montseny, J. J. et al. The current spectrum of
infectious glomerulonephritis. Experience with 76
patients and review of the literature. Medicine 74,
63–73 (1995).
93. Wang, S. et al. Clinical, pathological, and prognostic
characteristics of glomerulonephritis related to
staphylococcal infection. Medicine 95, e3386
(2016).
94. Ebright, J. R. & Komorowski, R. Gonococcal
endocarditis associated with immune complex
glomerulonephritis. Am. J. Med. 68, 793–796
(1980).
95. Bookman, I., Scholey, J. W., Jassal, S. V., Lajoie, G.
& Herzenberg, A. M. Necrotizing glomerulonephritis
caused by Bartonella henselae endocarditis. Am. J.
Kidney Dis. 43, e25–e30 (2004).
96. Elzouki, A. Y., Akthar, M. & Mirza, K. Brucella
endocarditis associated with glomerulonephritis
and renal vasculitis. Pediatr. Nephrol. 10, 748–751
(1996).
97. Bell, E. T. Glomerular lesions associated with
endocarditis. Am. J. Pathol. 8, 639–664 (1932).
98. Cordeiro, A., Costa, H. & Laginha, F. Immunologic
phase of subacute bacterial endocarditis. A new
concept and general considerations. Am. J. Cardiol.
16, 477–481 (1965).
99. Bayer, A. S. & Theofilopoulos, A. N. Immunopathogenetic
aspects of infective endocarditis. Chest 97, 204–212
(1990).
100. Gutman, R. A., Striker, G. E., Gilliland, B. C. &
Cutler, R. E. The immune complex glomerulonephritis
of bacterial endocarditis. Medicine 51, 1–25 (1972).
101. Morel-​Maroger, L., Sraer, J. D., Herreman, G.
& Godeau, P. Kidney in subacute endocarditis.
Pathological and immunofluorescence findings.
Arch. Pathol. 94, 205–213 (1972).
102. Agarwal, A. et al. Subacute bacterial endocarditis
masquerading as type III essential mixed
cryoglobulinemia. J. Am. Soc. Nephrol. 8, 1971–1976
(1997).
103. Majumdar, A. et al. Renal pathological findings in
infective endocarditis. Nephrol. Dial. Transplant. 15,
1782–1787 (2000).
104. Eknoyan, G., Lister, B. J., Kim, H. S. & Greenberg, S. D.
Renal complications of bacterial endocarditis.
Am. J. Nephrol. 5, 457–469 (1985).
105. Fatima, S. et al. Epidemiology of infective endocarditis
in rural upstate New York, 2011–2016. J. Clin. Med.
Res. 9, 754–758 (2017).
106. Fleischauer, A. T., Ruhl, L., Rhea, S. & Barnes, E.
Hospitalizations for endocarditis and associated
health care costs among persons with diagnosed drug
dependence — North Carolina, 2010–2015. MMWR
Morb. Mortal. Wkly Rep. 66, 569–573 (2017).
107. Shiffman, M. L. The next wave of hepatitis C virus:
the epidemic of intravenous drug use. Liver. Int. 38,
(Suppl. 1), 34–39 (2018).
108. Bassetti, S. & Battegay, M. Staphylococcus aureus
infections in injection drug users: risk factors and
prevention strategies. Infection 3, 163–169 (2003).
109. Chirinos, J. A. et al. Endocarditis associated with
antineutrophil cytoplasmic antibodies: a case report
and review of the literature. Clin. Rheumatol. 26,
590–595 (2007).
110. Tiliakos, A. M. & Tiliakos, N. A. Dual ANCA positivity
in subacute bacterial endocarditis. J. Clin. Rheumatol.
14, 38–40 (2008).
111. Uh, M., McCormick, I. A. & Kelsall, J. T. Positive
cytoplasmic antineutrophil cytoplasmic antigen with
PR3 specificity glomerulonephritis in a patient with
subacute bacterial endocarditis. J. Rheumatol. 38,
1527–1528 (2011).
112. Langlois, V. et al. Antineutrophil cytoplasmic
antibodies associated with infective endocarditis.
Medicine 95, e2564 (2016).
113. Bor, D. H., Woolhandler, S., Nardin, R., Brusch, J.
& Himmelstein, D. U. Infective endocarditis in the
U.S., 1998–2009: a nationwide study. PLOS ONE 8,
e60033 (2013).
114. Johnson, J. A., Boyce, T. G., Cetta, F., Steckelberg, J. M.
& Johnson, J. N. Infective endocarditis in the pediatric
patient: a 60-year single-​institution review. Mayo Clinic
Proc. 87, 629–635 (2012).
115. Fournier, P. E. et al. Comprehensive diagnostic strategy
for blood culture-​negative endocarditis: a prospective
study of 819 new cases. Clin. Infect. Dis. 51, 131–140
(2010).
116. Hoen, B. et al. Infective endocarditis in patients with
negative blood cultures: analysis of 88 cases from a
one-​year nationwide survey in France. Clin. Infect. Dis.
20, 501–506 (1995).
117. Yates, M. et al. EULAR/ERA-​EDTA recommendations
for the management of ANCA-​associated vasculitis.
Ann. Rheum. Dis. 75, 1583–1594 (2016).
118. Baddour, L. et al. Infective endocarditis in adults:
diagnosis, antimicrobial therapy, and management
of complications. Circulation 132, 1435–1486
(2015).
119. Haas, M. in Heptinstall’s Pathology of the Kidney 7th
edn Vol. 1 Ch. 12 (eds Jennette, J. C. et al.) 463–524
(Wolters Kluwer, 2014).
120. Glassock, R. J. et al. Staphylococcus-​related
glomerulonephritis and poststreptococcal
glomerulonephritis: why defining “post” is important
in understanding and treating infection-​related
glomerulonephritis. Am. J. Kidney Dis. 65, 826–832
(2015).
121. Nadasdy, T. & Hebert, L. A. Infection-​related
glomerulonephritis: understanding mechanisms.
Semin. Nephrol. 31, 369–375 (2011).
122. Nasr, S. H. & D’Agati, V. D. IgA-​dominant postinfectious
glomerulonephritis: a new twist on an old disease.
Nephron Clin. Pract. 119, c18–c26 (2011).
123. Nast, C. C. Infection-​related glomerulonephritis:
changing demographics and outcomes. Adv. Chronic
Kidney Dis. 19, 68–75 (2012).
124. Tadema, H., Heeringa, P. & Kallenberg, C. G.
Bacterial infections in Wegener’s granulomatosis:
mechanisms potentially involved in autoimmune
pathogenesis. Curr. Opin. Rheumatol. 23, 366–371
(2011).
125. Salmela, A. et al. Chronic nasal Staphylococcus aureus
carriage identifies a subset of newly diagnosed
granulomatosis with polyangiitis with high relapse
rate. Rheumatology 56, 965–972 (2017).
126. Oliveira, D. B. G. Linked help from bacterial proteins
drives autoantibody production in small vessel
vasculitis. Med. Hypotheses 112, 24–26 (2018).
127. Sethi, S. et al. Atypical postinfectious
glomerulonephritis is associated with abnormalities
in the alternative pathway of complement. Kidney Int.
83, 293–299 (2012).
128. Smith, R. J. H. et al. C3 glomerulopathy —
understanding a rare complement-​driven renal
disease. Nat. Rev. Nephrol. 15, 129–143 (2019).
129. Ozlok, A. & Yildiz, A. Hepatitis C virus associated
glomerulopathies. World J. Gastroenterol. 20,
7544–7554 (2014).
www.nature.com/nrneph

130. Sinniah, R. Occurrence of mesangial IgA and IgM
deposits in a control necropsy population. J. Clin.
Pathol. 36, 276–279 (1983).
131. Ji, S. et al. The fate of glomerular mesangial igA
deposition in the donated kidney after allograft
transplantation. Clin. Transplant. 18, 536–540
(2004).
132. Suzuki, K. et al. Incidence of latent mesangial IgA
deposition in renal allograft donors in Japan. Kidney
Int. 63, 2286–2294 (2003).
133. Lange, K., Seligson, G. & Cronin, W. Evidence for the
in situ origin of poststreptococcal glomerulonephritis:
glomerular localization of endostreptosin and the
clinical significance of the subsequent antibody
response. Clin. Nephrol. 19, 3–10 (1983).
134. Vogt, A., Batsford, S., Rodriguez-​Iturbe, B. &
Garcia, R. Cationic antigens in poststreptococcal
glomerulonephritis. Clin. Nephrol. 20, 271–279
(1983).
135. Yoshizawa, N. et al. Role of streptococcal antigen
in the pathogenesis of acute post-​streptococcal
glomerulonephritis, characterization of the antigen
and a proposed mechanism for the disease.
J. Immunol. 148, 3110 (1992).
136. Froude, J., Zabriskie, J. B., Buchen, D., Rzucidlo, E. &
Kakani, R. Immunochemical studies of nephritis strain
associated protein (Nsap) and streptokinase (Ska).
Zentralbl. Bakteriol. 22, S203–S205 (1992).
137. Kefalides, N. A. et al. Antibodies to basement
membrane collagen and to laminin are present in sera
from patients with poststreptococcal glomerulonephritis.
J. Exp. Med. 163, 588–602 (1986).
138. Komaru, Y., Ishioka, K., Oda, T., Ohtake, T. &
Kobayashi, S. Nephritis-​associated plasmin receptor
(NAPlr) positive glomerulonephritis caused by
Aggregatibacter actinomycetemcomitans bacteremia:
a case report. Clin. Nephrol. 90, 155–160 (2018).
139. Oda, T. et al. The role of nephritis-​associated plasmin
receptor (NAPlr) in glomerulonephritis associated with
Nature Reviews | Nephrology
streptococcal infection. J. Biomed. Biotechnol. 2012,
417675 (2012).
140. Batsford, S. R., Mezzano, S., Mihatsch, M., Schiltz, E.
& Rodriguez-​Iturbe, B. Is the nephritogenic antigen in
post-​streptococcal glomerulonephritis pyrogenic
exotoxin B (SPE B) or GAPDH? Kidney Int. 68,
1120–1129 (2005).
141. Schick, B. in Scharlach (eds Escherich, T., Shick, B.)
151 (A. Hölder, 1912).
142. Okamoto, S. & Nagase, S. Pathogenic mechanisms of
invasive group A Streptococcus infections by influenza
virus-​group A Streptococcus superinfection. Microbiol.
Immunol. 62, 141–149 (2018).
143. Pla, E. et al. Onset of Berger’s disease after
Staphylococcus aureus infection: septic arthritis after
anterior cruciate ligament reconstruction. Arthroscopy
19, 4 (2003).
144. Delves, P. J., Martin, S. J., Burton, D. R. & Roitt, I. M.
Roitt’s Essential Immunology 11th edn (John Wiley
& Sons, 2011).
145. Safi, R. et al. Neutrophils contribute to vasculitis by
increased release of neutrophil extracellular traps in
Behçet’s disease. J. Dermatol. Sci. 92, 143–150
(2018).
146. Skopelja-​Gardner, S., Jones, J. D. & Rigby, W. F. C.
“NETtling” the host: breaking of tolerance in chronic
inflammation and chronic infection. J. Autoimmun. 88,
1–10 (2018).
147. DeLorenze, G. N. et al. Polymorphisms in HLA
class II genes are associated with susceptibility to
Staphylococcus aureus infection in a white population.
J. Infect. Dis. 213, 816–823 (2016).
148. Kambham, N. & Troxell, M. in Bacterial Infections and
the Kidney Ch. 3 (eds Satoskar, A. A. & Nadasdy, T.)
63–85 (Springer, 2017).
149. Okada, M. et al. Central venous catheter infection-​
related glomerulonephritis under long-​term parenteral
nutrition: a report of two cases. BMC Res. Notes 9,
196 (2016).
Reviews
150. Jeffres, M. N., Isakow, W., Doherty, J. A., Micek, S. T.
& Kollef, M. H. A retrospective analysis of possible
renal toxicity associated with vancomycin in patients
with health-​care associated methicillin-​resistant
Staphylococcus aureus pneumonia. Clin. Ther. 29,
1107–1115 (2007).
151. Hernandez, F. J. et al. Non-​invasive imaging of
Staphylococcus infections with a nuclease-​activated
probe. Nat. Med. 20, 301–306 (2014).
152. Olson, M. E., Slater, S. R., Rupp, M. E. & Fey, P. D.
Rifampicin enhances activity of daptomycin and
vancomycin against both a polysaccharide
intercellular adhesin (PIA)-dependent and
-independent Staphylococcus epidermidis
biofilm. J. Antimicrob. Chemother. 65, 2164–2171
(2010).
153. Olson, M. E. et al. Staphylococcus epidermidis agr
quorum-​sensing system: signal identification, cross
talk, and importance in colonization. J. Bacteriol. 196,
3482–3493 (2014).
Author contributions
A.A.S. researched data for the article, made substantial contri­
butions to discussions of the content and wrote, reviewed and
edited the manuscript before submission. S.V.P. researched
data for the article and wrote part of the manuscript. T.N.
reviewed and edited the manuscript before submission.
Competing interests
The authors declare no competing interests.
Peer review information
Nature Reviews Nephrology thanks B. Najafian, T. Oda and
the other, anonymous, reviewer(s) for their contribution to the
peer review of this work.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.