COVID-19 and Glomerular Diseases

Abstract

COVID-19 is a systemic disease, and the kidney is one of the target organs of infection. Kidney injury is common and can occur in up to 40% of patients. Several glomerular diseases have been reported in association with COVID-19. Some are likely related to COVID-19 whereas many are likely coincidental. Glomerular diseases that are frequently reported in COVID-19 and have a plausible mechanistic explanation are likely to be related to COVID-19. In contrast, glomerular diseases that are seldom reported and have no known plausible mechanism, are likely to be unrelated. Collapsing glomerulopathy (CG) is by far the most prevalent. Its association with COVID-19, resembling HIV and CG, led to the newly proposed term “COVID-19 associated nephropathy” or “COVAN.” High-risk APOL1 genotypes are the major risk factor in COVAN patients. Podocytopathy, membranous nephropathy (MN), pauci-immune crescentic glomerulonephritis (GN), and thrombotic microangiopathy (TMA) are also reported. In kidney allografts, CG remains the most common glomerular pathology. Patients typically present with acute kidney injury (AKI) or abnormal urinary findings at the time of or shortly after COVID-19 diagnosis. Treatment of glomerular disease in patients with COVID-19 is challenging. Providers should cautiously consider balancing risks and benefit of immunosuppression, particularly in patients with active diseases. Short-term outcomes vary but generally remain poor with high morbidity and mortality. Future study of long-term outcomes is needed to improve our understanding of glomerular disease associated with COVID-19.

Introduction

Since the COVID-19 pandemic began, over 680 million people have been infected, with a mortality of 6.8 million. COVID-19 is a systemic disease and extrapulmonary manifestations are common.¹ Severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) is an enveloped single stranded RNA virus that belongs to the family Coronaviridae.² SARS-CoV-2 imposes tissue injury via several mechanisms including direct viral invasion, endothelial cell damage, immune dysregulation, and renin angiotensin aldosterone system dysfunction.¹ Angiotensin-converting enzyme 2 is the membrane-bound receptor that binds to the receptor binding site located at the spike region of SARS-CoV-2.³ A disintegrin and metalloproteinase domain-containing protein 10, a disintegrin and metalloproteinase domain-containing protein 17, and transmembrane protease serine 2 are required for angiotensin-converting enzyme 2-SARS-CoV-2 complex to enter target cells via proteolytic cleavage.⁴ Angiotensin-converting enzyme 2 is widely expressed in many tissues, including type 2 pneumocytes, nasal mucosa, nasopharynx, enterocytes of small intestines, skin, bone marrow, spleen, liver, brain, and kidney.^3,5 Although the kidney appears to be one of the target organs affected by SARS-CoV-2, supported by the presence of SARS-CoV-2 in the urine and favorable distribution of angiotensin-converting enzyme 2,6, 7, 8 most kidney biopsies from COVID-19 patients do not demonstrate the presence of SARS-CoV-2 in the tissue.9, 10, 11, 12 AKI is prevalent in patients with COVID-19.^13,14 Earlier published studies, which predated mass vaccination enrollment, showed a high rate of AKI ranging from 20% to 40% with the rate of renal replacement therapy as high as 31%.¹³ A more recent study demonstrated a lower rate of AKI with a pooled incidence of 12.3% with 5% requiring renal replacement therapy. SARS-CoV-2 can cause AKI via direct and indirect effects. Directly, SARS-CoV-2 can cause cellular toxicity, complement activation, immune dysregulation, and coagulopathy.^15,16 Immune dysregulation in COVID-19 is one of the key features in many kidney pathologies and it involves various mechanisms, including molecular mimicry, upregulation of inflammatory cytokines, alteration in immune profile, and autoantibody production.¹⁷ Alternatively, SARS-CoV-2 can indirectly induce AKI via hemodynamic instability, organ cross-talk, or exposure to nephrotoxic agents.¹⁵ Although acute tubular injury (ATI) is common in a kidney biopsy of patients with COVID-19, several other kidney pathologies have been reported.^9,18,19

In this review, we searched the peer-reviewed literature through PubMed on COVID-19 associated kidney pathology in kidney biopsy series published since the beginning of the pandemic until February 2023. We emphasize only those series with available clinical outcomes and relevance to clinical practice. We focus on de-novo glomerular disorders. Details of clinical presentation, treatments, clinical courses, and outcomes that are available in the published articles, are included. We also provide examples of kidney pathology images of unpublished cases from our institution.

Kidney Biopsy and Autopsy Series in Patients With COVID-19

Since the onset of the COVID-19 pandemic until February 2023, there have been 76 kidney pathology-related studies. These studies included 511 native kidney cases and 85 transplant cases. Of the 76 studies, 36 (47.4%) studies are from North America, 16 (21.1%) from Europe, 22 (28.9%) from Asia, 1 (1.3%) from Africa, and 1 (1.3%) from South America.

Of 511 biopsy cases, 410 cases (80.2%) have glomerular lesions, 99 cases (19.4%) have tubulointerstitial lesions, and 2 cases (0.4%) have no significant pathology. Of glomerular pathologies, CG is the most common pathology (n = 148, 36.1%) followed by, primary podocytopathy (n = 45, 11.0%), diabetic nephropathy (n = 37, 9.0%), pauci-immune crescentic glomerulonephritis (GN) (n = 31, 7.6%), membranous nephropathy (MN) (n = 23, 5.6%) and immunoglobulin A (IgA) nephropathy (IgAN) (n = 23, 5.6 %). The remaining 103 (25.1 %) cases have other glomerular pathologies. The detail of glomerular pathologies is shown in Table 1 and Figure 1a.

Table 1.

Glomerular pathologies from native kidney biopsy studies since the beginning of COVID-19 pandemic until February 2023

Pathology	Number	Percentage
Collapsing glomerulopathy	148	36.1%
Primary podocytopathy	45	11.0%
Diabetic nephropathy	37	9.0%
Pauci-immune crescentic GN	31	7.6%
Membranous nephropathy	23	5.6%
IgA nephropathy	23	5.6%
Thrombotic microangiopathy	21	5.1%
Secondary FSGS	15	3.7%
Lupus nephritis	10	2.4%
Anti-GBM GN	9	2.2%
Infection-associated GN	8	2.0%
AL amyloidosis	5	1.2%
PGNMID	5	1.2%
Cryoglobulinemic GN	3	0.7%
Immune complex GN	3	0.7%
MGMID	1	0.3%
Fibrillary glomerulonephritis	1	0.3%
Light chain deposition disease	1	0.3%
AA amyloidosis	1	0.3%
Other
-Nephrosclerosis (chronic changes)	17	4.1%
-Thin basement membrane disease	1	0.3%
-Sickle cell nephropathy	1	0.3%
Total	410	100.0%

FSGS, focal segmental glomerulosclerosis; GBM, glomerular basement membrane; GN, glomerulonephritis; MGMID, membranous-like glomerulopathy with masked IgG kappa deposits; PGNMID, proliferative glomerulonephritis with monoclonal immunuoglobulin deposition.

Figure 1.

Glomerular pathology in native kidney and allograft biopsy. ABMR, antibody-mediated rejection; ATN, acute tubular necrosis; CG, collapsing glomerulopathy; DN, diabetic nephropathy; FSGS, focal segmental glomerulosclerosis; GN, glomerulonephritis; IgAN, IgA nephropathy; LN, lupus nephritis; MN, membranous nephropathy; TMA, thrombotic microangiopathy; TCMR, T-cell mediated rejection.

There are several glomerular diseases reported in patients with COVID-19; however, some glomerular diseases are likely simply coincidental. These lesions are mostly being diagnosed while patients with COVID-19 seek medical attentions for other issues. Although establishing a causal relationship between COVID-19 and glomerular diseases remains challenging, some clues may suggest the possibility of the association. Glomerular diseases that are likely to have a causal relationship with COVID-19 are frequently reported in the literature, more prevalent in the COVID-19 era compared to pre-COVID-19 era,⁹ or have plausible mechanistic explanations such as immune dysregulation, autoantibody production, cytokine upregulation, complement activation, or direct viral toxicity.¹⁵ These entities may include CG, primary podocytopathy, pauci-immune crescentic GN, MN, IgAN, etc. In contrast, some glomerular lesions that are likely coincidental with no other plausible mechanistic links include diabetic nephropathy, AL amyloidosis, AA amyloidosis, among others, etc (Figure 2).

Figure 2.

Approach to glomerular pathology in patients with COVID-19. ATN, acute tubular necrosis; CG, collapsing glomerulopathy; CNI, calcineurin inhibitor; FSGS, focal segmental glomerulosclerosis; GBM, glomerular basement membrane; GN, glomerulonephritis; MCD, minimal change disease; MGMID, membranous-like glomerulopathy with masked IgG kappa deposits; PGNMID, proliferative glomerulonephritis with monoclonal immunoglobulin deposit; TBMD, thin basement membrane disease; TMA, thrombotic microangiopathy.

Glomerular Diseases in Native Kidneys

Kidney Pathology Potentially Related to COVID-19

CG

CG is the major pathological finding in the biopsy series of patients with COVID-19.^9,19 It was first reported by Larsen et al.²⁰ in a Black patient who presented with COVID-19. The patient developed stage 3 AKI and nephrotic range proteinuria. A kidney biopsy revealed CG and diffuse ATI. Genetic testing showed homozygous G1 apolipoprotein L1 (APOL1) variant.²⁰ Subsequent studies have reported similar finding of CG in COVID-19 and high-risk variant APOL1.^19,21, 22, 23, 24, 25, 26 May et al.⁹ showed that 91.7% of African American or Hispanic descents whose kidney biopsy showed CG harbored 2 risk alleles APOL1 variant. On the other hand, only 35.6% of patients with non-CG possessed 2 risk alleles of the APOL1 variant.⁹

This association of high-risk APOL1 genotype and kidney disease has been previously reported in several viral infections such as cytomegalovirus, parvovirus B19, Epstein-Barr virus, and particularly HIV.²⁷ HIV-associated nephropathy (HIVAN) is an aggressive form of kidney disease that occurs in HIV patients, particularly in patients with uncontrolled disease. The biopsy findings in HIVAN typically show diffuse CG, tubular injury, and distended tubules containing proteinaceous cast forming “tubular microcysts.” Electron microscopy (EM) shows collapsed capillaries with wrinkled glomerular basement membrane (GBM) and diffuse foot process effacement (FPE). Tubuloreticular inclusions (TRI s), known as “interferon (IFN) footprint,” are often observed in glomerular endothelium.²⁸ The similarity of clinical and pathological findings in HIVAN and CG in COVID-19 led to the newly proposed term of COVAN or COVID-19-associated nephropathy.²⁷ Pathophysiology of COVAN is thought to be related to immune dysregulation in patients with COVID-19. Several cytokines such as interleukin-1β, -6, -10, and IFN-γ are upregulated in the setting of SARS-CoV-2 infection.²⁹ IFN interact with APOL1 risk variant culminating in glomerular epithelial cell autophagy, mitochondrial dysfunction, and cell injury.²⁷ Although COVAN shares several similarities, including pathological findings and upregulation of IFN, with HIVAN, viral detection in kidneys appears to be more common in HIVAN than in COVAN.³⁰ To date, there are 148 cases of CG in COVID-19, of these, 86 cases have detailed clinical data available.

Clinical Presentation. The median age was 57 years (interquartile range [IQR] 49, 63). The majority of patients were men (75.6 %) and nearly all the patients were of African ancestry.¹⁹ Cardiovascular comorbidities such as diabetes, hypertension, and obesity were common. About 80% of patients had a kidney biopsy within 1 month of COVID-19 diagnosis.¹⁹ Patients often presented with AKI with median serum creatinine of 7.4 mg/dl. Almost 90% of patients had AKI stage 3. Severe proteinuria was a hallmark of CG with median 24-hour urine protein of 10 g/d and 96% of patients have nephrotic syndrome.¹⁹ About half of patients have hematuria (Supplementary Table S1).

Pathological Findings (Figure 3a). Glomeruli demonstrate glomerular tuft collapse with overlying reactive glomerular epithelium in the Bowman space.²⁰ Features of tubular injury include diffuse simplification with denudation of the brush border. The degree of tubulointerstitial scarring seems to be associated with the time of CG and COVID-19 diagnosis. Moderate to severe tubulointerstitial scarring tends to be more common if a kidney biopsy is performed after 1 month of the COVID-19 diagnosis.¹⁹

Figure 3.

Kidney pathology in native kidney biopsy in patients with COVID-19. (a) The glomerulus shows collapsed capillary tufts with florid parietal epithelial cells hyperplasia, typical of COVID-19 associated nephropathy (COVAN) (Jones methenamine silver, ×400); (b) A segmental cellular crescent and mesangial proliferation are noted in the glomerulus of a patient who had COVID-19 and IgA nephropathy (Jones methenamine silver, ×400); (c) IgA dominant staining of the mesangium confirms a diagnosis of IgA nephropathy. (d) A PR3 positive pauci-immune glomerulonephritis characterized by a focal segmental cellular crescent and lack of mesangial or endocapillary proliferation (PAS, ×400) (e) A biopsy of antiglomerular basement membrane nephritis in a patient with COVID-19 shows a circumferential cellular crescent and glomeruli without proliferative histology (PAS, ×400); (f) The IgG stain demonstrates bright linear staining of the glomerular basement membrane, without staining of tubular basement membranes.

Some patients may have interstitial edema with a predominantly lymphocytic infiltrate.²⁰ Immunofluorescence (IF) is typically negative for immune reactants. EM microscopy typically showed diffuse FPE and cytoplasmic vaculolization.^20,31 TRIs are occasionally found in the glomerular endothelial cell cytoplasm.³¹ Overall, these features are not different from CG in non-COVID-19 patients. Thus far, there has been no evidence of viral detection in kidney biopsy of patients with CG.

Treatment and Outcome. About a third of patients received steroids for COVID-19 and a quarter of patients received either monoclonal antibodies or immunomodulators agents for COVID-19. However, only 6% of patients received antiviral therapy. A quarter of the patients received immunosuppression for CG. These treatments included prednisone alone, calcineurin inhibitor (cyclosporine), and cyclophosphamide. Overall, CG portends poor prognosis. Approximately 70% of patients required dialysis, but half of the patients were able to come off dialysis. The rate of end-stage kidney disease was higher than CG in non-COVID-19 in which the rate of end-stage kidney disease was approximately 30%.³² About 7.0% of the patients died during follow-up. Supplementary Table S1.

Noncollapsing Podocytopathy

Podocytopathy refers to a group of glomerular diseases caused by podocyte injury. This often leads to severe proteinuria and nephrotic syndrome as a result of diffuse FPE.^33,34 Podocytopathy can be classified based on their pathogenesis into permeability-factor mediated, toxic, or genetic.³⁴ Noncollapsing primary podocytopathy (minimal change disease and noncollapsing focal segmental glomerulosclerosis [FSGS]) is the second most common reported glomerular pathology in patients with COVID-19 following CG. Pathophysiology of podocytopathy may be related to upregulation of IFN in COVID-19 rather than direct toxicity from viral invasion because most patients did not have viral particles detected in their biopsy.^9,35 Although, the relationship between APOL1 high risk genotype and noncollapsing FSGS is less defined compared to that with CG, about 86% of African American and Hispanic patients with primary podocytopathy had APOL1 high risk genotypes.⁹ There were total of 45 cases, of which 23 cases (51.1%) were minimal change disease and 22 cases (48.9%) were noncollapsing FSGS. Of these 45 cases, only 21 cases have detailed clinical history available.^10,18,19,35, 36, 37, 38

Clinical Presentation. Similar to podocytopathy in non-COVID-19 setting, patients typically presented with significant proteinuria. The median urine protein was 12 g/d. About two-thirds of patients have evidence of hematuria. The majority (75%) of patients were African American. The median age at presentation was 54 years and 57% were male. About two-thirds of patients presented with stage 3 AKI with median serum creatinine of 4.5 mg/dl. A kidney biopsy is often performed within 2 to 4 weeks of COVID-19 diagnosis.^19,35 Most patients had mild to moderate COVID-19.¹⁹ Summary of clinical characteristics is shown in Supplementary Table S2.

Pathological Findings. Kidney pathology may vary depending on the degree of injuries. Light microscopy may show no significant abnormalities as in minimal change disease. Subtle changes such as mild mesangial hypercellularity may be reported. Tip or not otherwise specified lesions are reported in patients with FSGS. ATI is very common and seen in almost every case. The degree of interstitial fibrosis and tubular atrophy is usually mild but some cases may show severe changes,¹⁹ which raises suspicion of possible undiagnosed podocytopathy preceding COVID-19. Microcysts may present in a small number of patients.¹⁹ IF is typically negative for immunoreactants. EM findings demonstrate diffuse FPE with occasional TRIs.³⁵ Phagolysosomal activity may be observed by EM.¹⁰

Treatment and Outcome. Twenty-seven percent of patients received steroids therapy for COVID-19 and 18% received antiviral therapy. The majority of patients did not receive immunosuppression. Approximately one-third of patients received immunosuppression, all of which were prednisone. Unlike CG, only 37.5 % of patients required dialysis. Patients had a better prognosis compared to CG. Twenty-five percent of patients had complete renal recovery and only 6% required long-term dialysis (Supplementary Table S2).

MN

MN is the most common adult nephrotic syndrome in the Western world.³⁹ Approximately 80% of cases are associated with phospholipase A2 receptor antibody.³⁹ Other antigens associated with MN include thrombospondin 7A, neural epidermal growth factor-like 1 protein, Exostosin 1 or 2, or protocadherin FAT1,^40,41 which have expanded our understanding of MN. However, some viral infections, primarily hepatitis B virus, can be associated with MN.⁴² There are reports of MN cases that occurred in temporal relationship with COVID-19 (23 cases) but only 7 cases have clinical information available.^18,43, 44, 45, 46, 47 Pathophysiology of MN in COVID-19 remains unclear but may be related to immune dysregulation resulting in autoantibodies production.⁴⁶

Clinical Presentation. The median age on presentation was 72 years (IQR 63, 81 years). Almost all patients (85.7%) were male. There were 33% Asian, 33% Hispanic, 17% Black and 17% White. Median time of onset was within 1 week of COVID-19 diagnosis. Median serum creatinine upon biopsy was 1.8 mg/dl (IQR 1.0, 7.0). The median proteinuria was 6.8 (IQR 4.6, 8.7) g/d and nearly all patients (83.3%) had hematuria. Phospholipase A2 receptor antigen was tested in all 7 patients and detected in 4 (57.1%) patients. No other antigen has been reported in the literature thus far (Supplementary Table S3).

Pathological Findings. Glomeruli typically showed segmental mild to moderate mesangial hypercellularity and matrix expansion.^44,46 Diffuse thickening of basement membrane is common. There was no report of endocapillary hypercellularity, crescents, or thrombi in these cases. Glomerulosclerosis and interstitial fibrosis and tubular atrophy were typically mild to moderate.¹⁸ Microcysts were observed in some cases.¹⁸ ATI was reported in some cases.^46,47 IF showed typical granular peripheral capillary wall staining of IgG similar to non-COVID-19 MN cases. EM generally demonstrated diffuse FPE with primarily subepithelial immune deposits.⁴⁶ Mesangial and intramembranous immune deposits were sometimes reported.⁴⁴ Abundant TRIs were identified in the endothelial cell cytoplasm of 1 case.⁴⁶

Treatment and Outcome. The treatment and outcome are limited due to scarcity of reports and short-term follow-up. Two out of 7 patients received steroids for COVID-19 and only 1 patient received antiviral therapy. Four patients (66.7%) received immunosuppression as follows: 1 with tacrolimus, 1 with cyclophosphamide, 1 with methylprednisolone, and 1 with rituximab.^18,44, 45, 46 Two patients received conservative management. Two patients (40%) required dialysis and eventually died. Four patients (66.6%) had persistent kidney dysfunction. Of these, 3 patients received cyclophosphamide, tacrolimus or rituximab (Supplementary Table S3).^18,44,45

IgAN

IgAN is the most common GN globally and is more common in the East and Pacific Asian countries.⁴⁸ Patients typically present with microscopic or gross hematuria. Proteinuria is common and is associated with progression of kidney disease. The risk of progression to end-stage kidney disease varies from 5% to 15% at 5 years and 10% to 50% at 20 years.⁴⁹ Some patients may present with rapidly progressive GN with high burden of glomerular crescents.⁵⁰ Pathophysiology of IgAN in COVID-19 remains to be elucidated but it is possible that upregulated IgA may result in development of IgAN. In fact, IgA is the dominant immunoglobulin in the early course of COVID-19, and is associated with clinical severity.^51,52 Similar to a report of IgAN post COVID-19 vaccine, COVID-19 itself may unmask subclinical IgAN in asymptomatic patients.⁵³ There were total of 23 patients with IgAN in association with COVID-19. Of these, 13 patients had clinical data available.^47,54, 55, 56, 57, 58, 59, 60, 61, 62, 63

Clinical Presentation. COVID-19 patients with IgAN tended to be older with median age of 64.5 years (IQR 41, 79.5 years) compared to 34 years for IgAN in the general population.⁶⁴ The median onset of IgAN was 3 weeks (IQR 0, 7) post COVID-19. One patient had hematuria and proteinuria 14 months before COVID-19. However, this patient had worsened proteinuria and hematuria after COVID-19 and a kidney biopsy showed IgAN.⁵⁶ Majority (69.2 %) of patients were male, and 63.6 % were White. The median serum creatinine upon biopsy was 3.1 mg/dl (IQR 1.9, 4.0). Of the 13 patients, 2 patients did not have AKI, 1 patient had AKI stage 1, 5 patients had AKI stage 2, and 5 patients had AKI stage 3. Nearly all patients (91.7%) had hematuria and the median proteinuria was 3.7 g/d (IQR 1.3, 4.9) (Supplementary Table S4). Interestingly, 7 of 13 patients (53.8%) also had concomitant cutaneous vasculitis.^{47,57,59,60,62}

Pathological Findings. Kidney pathology of IgAN in COVID-19 was similar to IgAN in non-COVID-19 cases, including segmental sclerosis, mesangial hypercellularity, and endocapillary hypercellularity. However, COVID-19 IgAN cases may exhibit more active features such as diffuse endocapillary hypercellularity with neutrophilic exudation, and crescents.^47,56,59,62 Of 13 patients, 7 (53.8%) had crescentic lesions.^{47,55,56,59,60,62,63} IF showed granular mesangial staining for IgA, C3, kappa and lambda light chain.⁶¹ EM displayed mesangial, para-mesangial and rare subendothelial electron deposits.⁶¹ Significant podocyte FPE was observed in some cases.^47,62 Perez et al.also reported the presence of coronavirus particles in the cytoplasm of tubular cells via immunohistochemistry. However, PCR for SARS-CoV-2 was negative.⁶¹ (Figure 3b and c).

Treatment and Outcome. Thirty percent and 15% of patients received steroids and antivirals for COVID-19, respectively. Only 1 patient required dialysis. This patient had serum creatinine of 7.8 upon presentation. A kidney biopsy showed concomitant acute interstitial nephritis and crescentic lesions. This patient had epidural abscess, which precluded the patient from receiving aggressive immunosuppression but received high dose steroids with improvement in serum creatinine but remained on dialysis at the time of the report.⁵⁵ Nine patients required immunosuppression and all of them received steroids. Three patients received additional treatment with rituximab, cyclophosphamide, or tonsillectomy.^60,62,63 Of patients with immunosuppression, 2 patients had complete renal recovery, 6 patients had partial renal recovery (1 with negative proteinuria but with partial creatinine improvement; 5 with partial improvement in proteinuria ranging between 40% and 90%),^{57,59,60,62,63} and 1 patient remained on dialysis. Three patients were treated conservatively with renin angiotensin aldosterone system inhibitor.^54,56,58 Of these, 1 had complete recovery of proteinuria and the other 2 had improvements in serum creatinine, but the amount of proteinuria was not reported.^54,56,58 (Supplementary Table S4).

Pauci-Immune Crescentic GN

Pauci-immune crescentic GN is a group of vasculitis that involves kidneys and extrarenal organs. Majority of patients (90%) are associated with antineutrophil cytoplasmic antibody (ANCA), including proteinase-3 and myeloperoxidase.⁶⁵ ANCA-negative pauci-immune GN accounts for 10% of cases.⁶⁵ In patients with genetic predisposition for ANCA-associated GN, exposure to certain triggers (e.g., infections, drugs, or environmental factors) may lead to neutrophilic activation and ANCA production.⁶⁶ There is growing evidence to support the role of infections in both ANCA-associated and ANCA-negative GN.^67,68 Cytomegalovirus, Ebstein-Barr virus, and dengue virus are reported to be associated with ANCA-associated GN.⁶⁷ The underlying mechanism of developing pauci-immune crescentic GN may be related to molecular mimicry, ANCA autoantibody production, and formation of neutrophil extracellular traps.^69,70 Thus far, there are a total of 31 reports of COVID-19 associated pauci-immune crescentic GN; of these, 16 patients had clinical data available.^38,70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80

Clinical Presentation. The median age was 62.5 years (IQR 38.5, 76.4). Majority of patients were female (56.5 %) and majority were White (37.5 %). The median serum creatinine was 4.3 mg/dl (IQR 2.5, 6.1). Patients typically presented with severe AKI with 73.3 % of patients developing AKI stage 3. All patients had proteinuria and the median proteinuria was 3.3 g/d (IQR 1.8, 5.0). This was quite high compared to non-COVID-19 ANCA patients.⁸¹ There was heterogeneity in reporting ANCA antibody. Myeloperoxidase accounted for 43.7% of cases followed by proteinase-3 (25.0 %), C-ANCA (12.5%), C and P-ANCA (6.3%), myeloperoxidase + proteinase-3 (6.3%) and ANCA-negative 6.3%). Onset was typically within 1 week of COVID-19 diagnosis (IQR 0, 2.5). Of 13 patients that reported onset information, 9 patients were diagnosed within 1 week of COVID-19. Four patients were diagnosed at least 2 weeks after COVID-19 diagnosis; of these, 3 patients had myeloperoxidase antibody (Supplementary Table S5).^71,72,74,78

Pathological Findings. Light microscopy shows moderate to diffuse cellular to fibrocellular crescents associated with ruptured Bowman’s capsule.48, 49, 50, 51, 52, 53, 54 The degrees of glomerulosclerosis ranges from mild to severe.⁷² ATI is common with inflammatory infiltrates in some cases, and tubulitis may be observed.⁷⁵ Inflammatory infiltrates are nonspecific and include T-cells, B-Cells, plasma cells, and macrophages.^74,75 Red blood cell cast may be observed in severe hematuria.⁷¹ Necrotizing arteritis with fibrinoid necrosis and peritubular capillaritis are noted.⁷⁴ IF shows minimal to no staining of immunoreactants except fibrinogen in areas of crescents and necrosis. EM shows no or few electron-dense deposits and shows no viral particles.^75,79 Uppal et al.⁷⁹ were the only investigators who tried to detect the presence of SARS-CoV-2 in the kidney, which was negative by immunohistochemistry. As expected, patients who had a kidney biopsy done more than 1 month after COVID-19 appeared to have a greater degree of interstitial fibrosis and tubular atrophy.^71,72,74 (Figure 3d).

Treatment and Outcome. Although patients typically presented at the time of COVID-19 diagnosis, only 12.5% received steroids and 18.8% received antiviral agents for COVID-19. Of 16 patients, 15 required immunosuppression for the vasculitis. Thirteen patients received steroids plus other agents, whereas 2 received steroids alone. Three patients received plasmapheresis; of these, 2 received cyclophosphamide and 1 received rituximab.^70,71,76 Five patients (31.3%) required dialysis and 2 were able to discontinue dialysis after treatment. Of 16 patients, only 2 achieved complete recovery, 10 were with persistent renal dysfunction, 3 were dialysis dependent, and 1 patient died (Supplementary Table S5).

TMA

TMA is a systemic disease characterized by endothelial injury. Clinical manifestations of TMA are diverse and vary greatly from organ-limited symptoms to systemic features, including thrombocytopenia and microangiopathic hemolytic anemia.⁸² Kidney pathology shows endothelial swelling with or without thrombi. Intraluminal schistocytes may be present in the vessels. EM typically shows endothelial swelling and subendothelial lucency.⁸³ There are several factors associated with TMA, including drug-induced TMA, atypical hemolytic uremic syndrome, autoimmune diseases, glomerular diseases, bone marrow transplant, malignancies, genetic disorders, and infections.⁸² Many infections have been reported to induce TMA such as E.coli (Shiga toxin),⁸⁴ HIV,⁸⁵ influenza,⁸⁶ cytomegalovisus, Epstein-Barr virus, and others.⁸⁷ Pathogenesis of infection related TMA may be related to a second hit from infections in patients with predisposing complement gene mutations.^86,87 In COVID-19, direct viral toxicity to vascular endothelium, immune dysregulation, hyperinflammatory state, and uncontrolled complement pathway activation may play a role in pathogenesis of TMA in COVID-19.^88,89 There were a total of 21 patient reports of renal TMA; of these, 16 patients had clinical data available. However, there were additional 5 cases with evidence of TMA along with other glomerular pathologies; of these, 4 had CG and 1 had FSGS.³⁵

Clinical Presentation. The median age was 43 years (IQR 35.8, 67.5). The majority were male (56.3%) and White (44.4%). All patients suffered severe AKI, all stage 3 with median serum creatinine of 4.8 mg/dl (IQR 3.0, 7.4). Patients typically have subnephrotic range proteinuria, median urine protein of 1.4 g/d (0.7, 3.9). Seventy percent of patients have hematuria. Nearly 80% of patients have systemic features of TMA. Three patients have evidence of alternative complement dysregulation. One patient had low complement factor H and elevated CBb, 1 had CFHR1 and 3 gene mutations and 1 had variance of unknown significance in complement factor I and risk polymorphism in complement factor H gene.90, 91, 92 (Supplementary Table S6).

Pathological Findings. Light microscopy shows endothelial swelling with glomeruli and arterial thrombi. GBM remodeling with double contouring and cross-linked fibrin aggregated in glomerular capillaries can be observed in EM.²⁵ ATI and interstitial inflammation can be seen. In a case with severe TMA, diffuse coagulative cortical necrosis ensued.⁹⁰

Treatment and Outcome. Patients with TMA tended to receive specific therapy for COVID-19 more than other types of glomerular diseases; the therapies include steroids (46.2%), monoclonal antibodies or immunomodulators (33.3%), or antiviral agents (15.4%). The outcome and prognosis of patients with TMA in COVID-19 are grim. Two-thirds of patients required dialysis. Three patients received immunosuppression (2 with steroids and 1 with rituximab).^35,77,93 Eight patients received eculizumab. However, there were no differences in the rate of renal recovery or dialysis between patients with or without eculizumab. Twenty-three percent of patients died. (Supplementary Table S6).

Other Immune-Complex GN

There have been 8 reports of infection-associated GN as part of a large pathology cohort in patients with COVID-19. However, there was no detailed clinical data available.⁹ There were 3 reports of immune complex GN and 2 cases had clinical data available.^94,95 Sethi et al.⁹⁵ reported on a 25-year-old White female with history of genetic FSGS because of SMARCAL1 gene mutation. The patient was admitted due to diarrhea and was found to have Escherichia coli O157:H7-associated hemolytic uremic syndrome. She was also positive for SARS-CoV-2. She was treated with eculizumab with improvement in serum creatinine, but her proteinuria increased to 13 g/d from baseline of 3.7 g/d leading to a kidney biopsy. A kidney biopsy revealed a membranoproliferative GN pattern of injury. There were subendothelial wire-loop lesions. No crescents or fibrinoid necrosis was detected. IF showed bright staining for full house immunoreactions and also C4d. EM showed numerous large subendothelial and mesangial immune deposits. TRIs were absent. Interestingly, there was no evidence of endothelial injury.⁹⁵ Given “full house” appearance, an extensive autoimmune panel was checked, and was negative. The patient was given high dose methylprednisone followed by prednisone taper and mycophenolate mofetil. Three weeks after treatment, the patient still had significant proteinuria with slightly worsening kidney function.⁹⁵ Danis et al.⁹⁴ reported on a 59-year-old male who developed nephrotic syndrome and AKI 2 months post COVID-19. Serum creatinine upon presentation was 5.2 mg/dl and 4.2 g/d of proteinuria. Serological testing was negative. A kidney biopsy showed immune complex GN with positive IgG and C3 immunoreactants. The patient was treated with cyclophosphamide and high dose steroids and kidney function returned to normal at 2 weeks.⁹⁴ Etiology of immune complex GN in COVID-19 remains elusive but may be related to B-cell activation and hypergammaglobulinemia similar to HIV-associated immune-complex GN.⁹⁵ Pathophysiology of immune complex GN in COVID-19 may be similar to immune complex GN in HIV patients whereby polyclonal B-cell activation and hypergammaglobulinemia contribute to immune complex.⁹⁵ In fact, COVID-19 has been associated with IL-6 upregulation which is an important growth factor for B-cell differentiation and maturation.⁹⁶

Lupus Nephritis

There were 10 cases of lupus nephritis (LN) in patients with COVID-19 but only 1 case had available clinical data. Kudose et al.¹⁸ reported a 27-year-old Asian female with a history of class 2 LN who developed AKI with a serum creatinine of 2.5 mg/dl from a baseline of 0.9 mg/dl. The patient also developed nephrotic syndrome with 9.2 g/d of proteinuria, albumin 2.0 g/dl, and microscopic hematuria. A kidney biopsy was performed which showed evidence of LN class 4 and 5. There was endocapillary and mesangial hypercellularity associated with crescents. There was diffuse interstitial inflammation. EM showed diffuse FPE without evidence of TRIs or viral particles.¹⁸ The patient received steroids but subsequently expired due to multiorgan failure from COVID-19. The pathogenesis of lupus and other autoimmune diseases and COVID-19 may be related to molecular mimicry between SARs-CoV-2 and self-antigens and immune dysregulations.⁹⁷

Anti-GBM Disease

There were a total of 9 cases of anti-GBM disease but only 7 cases with reported clinical data.^18,98, 99, 100 Of 7 patients, 6 were Asian, and 1 was African American. The median age was 48 years (IQR 32, 57). Four patients (57.1%) were males. A kidney biopsy typically showed crescentic GN with evidence of anti-GBM GN with all glomeruli affected¹⁸ (Figure 3e and f). Patients typically present with very high serum creatinine, median 9.5 mg/dl (4.5, 17.0).^18,98, 99, 100 All patients received immunosuppression with cyclophosphamide and steroids. Five patients received plasmapheresis and 2 declined the therapy. Six patients required dialysis upon presentation but only 2 patients were able to discontinue dialysis.

Proliferative GN With Monoclonal Immunoglobulin Deposit

May et al.⁹ demonstrated that proliferative GN With Monoclonal Immunoglobulin Deposit (PGNMID) was more enriched in COVID-19 native kidney biopsies compared to the pre-COVID-19 era.⁹ There were a total of 5 cases of PGNMID in the literature. Four cases were from May et al. ⁹ and 1 case was reported by Shieh et al.¹⁰¹ This patient was the only patient with detail clinical information. The patient was a 71-year-old woman with normal baseline serum creatinine, 0.7 to 0.9 mg/dl. She was diagnosed with COVID-19 and was treated with remdesivir and dexamethasone. One month post COVID-19, her respiratory status improved but she developed generalized swelling. She had AKI with increased serum creatinine at 3.3 mg/dl but progressed to peak at 8.5 mg/dl before dialysis initiation. She had nephrotic syndrome with 24-hour urine protein of 5.2 g/d and serum albumin 1.9 g/dl. Kidney biopsy showed PGNMID with monoclonal IgG3 kappa deposits and only 10% to 20% interstitial fibrosis. Serum and urine monoclonal protein were negative. She was treated with cylocphosphamide, bortezomib, and dexamethasone for 2 cycles with an improvement in kidney function (serum creatinine 2.1 mg/dl) and subsequent discontinuation of dialysis. However, she developed recurrent AKI with serum creatinine up to 6.7 mg/dl shortly after stopping antiplasma cell therapy, requiring reinitiation of dialysis. A repeat kidney biopsy showed similar finding of PGNMID. Cylocphosphamide, bortezomib, and dexamethasone was reinitiated with good response, and she was able to come off dialysis. At the last follow-up, her serum creatinine were stabilized at less than 2.0 mg/dl with over 80% reduction in proteinuria.^9,101 There was a report that COVID-19 may cause an acute monoclonal gammopathy in patients without previous evidence of monoclonal gammopathy similar to other viral infections.⁹⁶ Upregulation of IL-6 in COVID-19 may promote B-cell differentiation and maturation into plasma cell resulting in monoclonal gammopathy.⁹⁶ It is also possible that other immune response to viral infection may explain the association with PGNMID similarly to previous reports.^102,103

Kidney Pathology Potentially Unrelated to COVID-19

There were reports of monoclonal protein-related lesions in patients with COVID-19 including cryoglobulinemic GN, AL amyloidosis, light chain deposition disease, and membranous-like glomerulopathy with masked IgG kappa deposit.^9,19 Other lesions include AA amyloidosis, thin GBM disease, and sickle cell nephropathy.^9,19,104 It is less likely that these lesions are associated with COVID-19 but rather a coincidental finding due to rarity of these lesions even in the COVID-19 era or lack of plausible pathophysiological association.

Glomerular Diseases in Kidney Allografts

There were a total of 85 kidney transplant patients. Of these, 58 cases (68.2%) had glomerular pathologies, 23 cases (27.1%) had tubulointerstitial lesions and 4 cases (4.7%) had no significant pathologies. Of 58 cases, 37 (63.8%) had acute or chronic rejections, whereas the remaining 21 cases (36.2%) had other glomerular lesions. The detail of glomerular lesions in kidney allografts is shown in Figure 1b. The increase in percentage of rejection in allograft biopsy was noted in a large series from May et al.⁹ and Daniel et al.¹¹ Majority of cases did not detect any evidence of viral invasion in the kidneys.¹¹ Pathophysiology of allograft rejection in COVID-19 is possibly related to immune dysregulation and reduction in antirejection medication in the setting of active COVID-19.³⁵ It is also possible that COVID-19 acts as a second hit in patients with preexisting donor specific antibodies.¹¹ In this section, we focus on only glomerular lesions excluding allograft rejections. Of 21 cases, CG was the most common reported lesions (n = 8, 38%).

CG

There were 8 cases of CG in allografts with available clinical data.^10,11,22,24 Of 5 cases, 3 received living donor kidneys whereas 4 received deceased donor kidneys. There was no report about the donor in 1 patient.¹⁰⁵ Median age was 49 years (IQR 45.8, 60). There were 4 (50%) African American, 3 (37.5%) White and 1 (12.5%) Hispanics. The median serum creatinine was 6.2 mg/dl (IQR 3.1, 11.3) with proteinuria 7.6 g/d (IQR 4.4, 14.2). There were 3 cases that had APOL1 risk genotype analyzed, 1 was G1/G2,¹⁰⁶ 1 was G0/G2²², and another one was G0/G1.¹⁰ This is different from native kidney CG where high risk APOL1 were more common than allograft CG.⁹ Similar to native kidneys, light microscopy typically revealed collapsing glomerular capillary loops accompanied with reactive or hypertrophic glomerular epithelial and mesangial hypercellularity. Patients typically had severe ATI with microcystic dilation containing proteinaceous content.^10,22,24 IF was typically negative for immunoreactants. EM showed diffuse FPE.¹¹ Six patients had a reduction in immunosuppression.^10,11,22,106 One patient required dialysis and remained dialysis-dependent at the time of the report.²⁴ Three patients had graft failure and 3 patients had persistent kidney dysfunction. Only 1 patient had recovered kidney function. The patient with kidney function recovery also received bamlanivimab therapy.¹¹

Other Lesions

There were several other lesions reported in the literature, including calcineurin toxicity, TMA with cortical necrosis, IgA nephropathy, diabetic nephropathy, LN, and PGNMID.^11,18,43 The frequency of each of these lesions is low, however, and some are possible coincidental.

Future Directions

Although there are several case series and reports of glomerular diseases in patients with COVID-19 since the pandemic began, there is still a need for large population based studies exploring incidence and prevalence of glomerular diseases post COVID-19 similar to studies of glomerular diseases post COVID-19 vaccination.¹⁰⁷ Many glomerular diseases are likely coincidental to COVID-19 rather than having true association. We need studies with control groups to compare the difference in clinical, pathological and prognostication between non-COVID-19 and COVID-19 patients. Importantly, clinicians lack the data regarding treatment particularly how to navigate immunosuppression therapy in the setting of active COVID-19 infection. It also remains unknown whether the outcomes of glomerular diseases associated with COVID-19 would improve with simply COVID-19 therapy or would require additional immunosuppression. Lastly, the type of therapeutic agents, dosage, duration of immunosuppression, and long-term outcomes are the pivotal areas that need to be further explored.

Conclusion

There are various forms of glomerular diseases reported in native and allograft kidneys of patients who recently had COVID-19 (Figure 2). The underlying mechanisms for each disease process remain unclear but likely are varied and multifactorial. A hyperinflammatory state and direct viral cytotoxicity may be responsible for CG and podocytopathy particularly in high risk APOL1 genotypes. Immune dysregulation may incite or unmask some autoimmune process such as ANCA-associated GN, anti-GBM disease, IgAN, or LN. Endothelial injury in conjunction with complement activation may precipitate TMA. In contrast, there are several renal pathologies that are likely unrelated to COVID-19 but are discovered when the patients sought care due to COVID-19. These lesions include certain monoclonal protein related kidney diseases, sickle cell nephropathy, AA amyloidosis, etc. Clinicians should take into consideration the onset of kidney manifestations, types of lesions, chronicity of lesions, and pathophysiological plausibility in deciding which lesions are likely causally related to COVID-19. Treatments of glomerular diseases in the setting of active or recent COVID-19 is challenging. Many patients were managed conservatively. However, immunosuppression may be beneficial in selected patients whose glomerular diseases that, outside of the setting of COVID-19 infection, typically require immunosuppression such as pauci-immune crescentic GN. Although long-term outcomes remain elusive, short-term outcomes generally are guarded. Larger and prospective studies are needed to better understand the epidemiology and outcomes of glomerular diseases following COVID-19, particularly with emerging variants.

Disclosure

All the authors declared no competing interests.

Supplementary Material

Supplementary References.

Table S1. Demographics, clinical presentations, treatment and outcomes of patients with collapsing glomerulopathy.S1-S25

Table S2. Demographics, clinical presentations, treatment and outcomes of patients with noncollapsing podocytopathy.S1^,S6^,S7^,S10^,S24-S26

Table S3. Demographics, clinical presentations, treatment and outcomes of patients with membranous nephropathy.S6^,S8^,S23^,S27-S29

Table S4. Demographics, clinical presentations, treatment and outcomes of patients with IgA nephropathy.S23^,S30-S39

Table S5. Demographics, clinical presentations, treatment and outcomes of patients with pauci-immune crescentic glomerulonephritis.S1^,S5^,S40-S49

Table S6. Demographics, clinical presentations, treatment and outcomes of patients with thrombotic microangiopathy.S1^,S5^,S7^,S8^,S24^,S25^,S50-S56