Acute Interstitial Nephritis with Glomerular Capillary IgA Deposition Following SARS-CoV-2 mRNA Vaccination

Abstract

We herein report a case of acute kidney injury (AKI) presenting as acute interstitial nephritis (AIN) after the first dose of the BNT162b2 mRNA vaccine against coronavirus disease 2019 (COVID-19). A 69-year-old man with a history of diabetes and hypertension presented with AKI 4 days after receiving the vaccine. Despite the administration of methylprednisolone pulse treatment, his renal function worsened, which prompted us to initiate temporal hemodialysis. His renal function subsequently improved, and a renal biopsy confirmed AIN and glomerular capillary IgA deposition without apparent crescents. The clinical history and histological findings suggest a relationship between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccination and AIN as a rare side effect.

Introduction

Coronavirus disease-19 (COVID-19) is an ongoing global pandemic caused by infection with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), first identified in December 2019. Several vaccines against SARS-CoV-2 have been developed to combat the COVID-19 pandemic, and clinical trials have confirmed their high efficacy and safety profile (1,2).

Several types of glomerulonephritis following SARS-CoV-2 vaccination, including IgA nephropathy (IgAN), anti-glomerular basement membrane (GBM) glomerulonephritis, anti-neutrophil cytoplasmic antibody (ANCA)-associated glomerulonephritis, membranous nephropathy (MN), minimal change disease (MCD), focal segmental glomerulonephritis (FSGS), and lupus nephritis, have been reported to date (3-11). However, acute interstitial nephritis (AIN) following SARS-CoV-2 vaccination has been considered a rare clinical entity, although reports of such cases have been increasing in conjunction with the large-scale implementation of COVID-19 vaccination worldwide (12-26).

We herein report a case of AIN with glomerular capillary IgA deposition in a patient with diabetes shortly after receiving the SARS-CoV-2 mRNA vaccine. Our case presents unique pathological findings and may offer valuable insight into the pathophysiology of SARS-CoV-2 mRNA vaccine-associated acute kidney injury (AKI).

Case Report

A 69-year-old Japanese man with a medical history of type 2 diabetes mellitus, hypertension, glaucoma, and cataract received the first dose of the BNT162b2 (Pfizer-BioNTech) SARS-CoV-2 mRNA vaccine. His renal function and urinalysis findings had been unremarkable for the previous 2 years, with baseline serum creatinine values ranging from 0.8 to 0.9 mg/dL and urinalysis findings being normal. The most recent glycated hemoglobin (HbA1c) level was 8.2%. The patient's outpatient medications included metformin, sitagliptin, ipragliflozin, amlodipine, and rosuvastatin. No other medications had been recently administered, prior to the administration of the SARS-CoV-2 mRNA vaccine.

Four days after vaccination (day -5), the patient visited another hospital complaining of a prolonged fever and loss of appetite. Laboratory assessments revealed an elevated serum creatinine of 3.9 mg/dL, and a urinalysis showed occult blood (2+) and proteinuria (1+). Drug-induced AKI was suspected, and his medications for diabetes (metformin, sitagliptin, and ipragliflozin) were discontinued. However, his fever persisted, and he was administered acetaminophen as required to reduce the fever. The patient denied the use of nonsteroidal anti-inflammatory drugs.

Eight days after vaccination (day -1), laboratory assessments revealed an elevated serum creatinine level of 5.7 mg/dL, urinalysis showed occult blood (1+), 10-19 red blood cells (RBCs) per high-power field (HPF), and proteinuria (1+), so he was referred to our hospital for nephrology consultation 9 days after vaccination (day 0). Upon admission, the patient was alert and oriented. His body temperature was 36.5°C, blood pressure was 123/78 mmHg, and pulse rate was 90 beats per minute. His physical examination findings were unremarkable, except for mild pedal edema. Laboratory tests revealed a serum creatinine level of 6.5 mg/dL. A urinalysis showed occult blood (2+), 5-9 RBCs per HPF, 1-4 white blood cells (WBCs) per HPF, urinary protein-creatinine ratio of 1.0 g/gCr, urinary N-acetyl-beta-D-glucosaminidase ratio of 23.9 U/g creatinine, and urinary β2-microglobulin levels of 56,074.0 μg/L. SARS-CoV-2 was negative according to reverse transcription loop-mediated polymerase chain reaction (RT-LAMP).

Thoracic and abdominal computed tomography showed swelling of the bilateral kidneys and left renal calculus but were otherwise unremarkable with no signs of hydronephrosis. The blood culture results were negative on admission. The serological evaluation showed a positive anti-GBM of 5.6 U/mL; antinuclear, anti-double-strand DNA antibodies, complement, anti-neutrophil cytoplasmic autoantibodies, hepatitis B and C virus, and HIV virus were all negative. Serum free light chain levels and protein electrophoresis with immunofixation results were normal. An evaluation of the serum Ig concentration revealed hypergammaglobulinemia of IgG (2,591 mg/dL; normal: 870-1,700 mg/dL), IgG4 (167 mg/dL; normal: 4.8-105 mg/dL), IgM (209 mg/dL; normal: 33-190 mg/dL), and IgE (1,043 mg/dL; normal ＜216 U/mL), whereas IgA 307 mg/dL (normal: 110-410 mg/dL) was in the normal range.

We initially suspected rapidly progressive glomerulonephritis, one of the most common causes of which is crescentic glomerulonephritis, such as anti-GBM glomerulonephritis and ANCA-associated glomerulonephritis with a poor renal prognosis and a risk of life-threatening pulmonary hemorrhaging. We therefore performed steroid treatment before conducting a renal biopsy. Since the patient's clinical history showed a rapid decline in the renal function over a few days, we considered steroid pulse therapy likely to provide a significant benefit in improving the renal function while withdrawing the renal replacement therapy. Therefore, we initiated methylprednisolone pulse therapy on day 2, starting with 3 daily pulses of 0.5 g, followed by 50 mg/day of oral administration. On day 3, his renal function worsened to a serum creatinine level of 8.1 mg/dL and urea nitrogen of 78 mg/dL, and he became complicated with hyperkalemia (5.8 mmol/L) and metabolic acidosis (15.8 mmol/L). Hemodialysis was therefore initiated with percutaneous internal jugular vein catheterization using a double lumen catheter. His renal function eventually improved after the administration of corticosteroids, and hemodialysis was discontinued on day 8. On day 9, he developed a catheter-related bloodstream infection (CRBSI) due to methicillin-sensitive Staphylococcus aureus; consequently, the catheter was removed immediately with the initiation of vancomycin and piperacillin tazobactam.

His bacteremia rapidly improved with prompt treatment, and antibiotics were de-escalated to cefazoline. After adequate antibiotic treatment, a percutaneous renal biopsy was performed on day 24. The renal biopsy specimens contained seven glomeruli, including one with global sclerosis. Glomerular crescent formation was not observed. The expansion of the mesangium was mild without apparent mesangial hypercellularity (Fig. 1A). Some glomeruli exhibited neutrophil infiltration in the lumens of the capillary loops, indicating a recent history of catheter-related bacteremia (Fig. 1B). Inflammatory cells, mainly composed of lymphoid and plasma cells, partially infiltrated the renal interstitium with tubulitis (Fig. 1C, D). Arteriolar hyalinosis and intimal thickening were partially present, reflecting a history of diabetes and hypertension. Immunofluorescence (IF) revealed linear IgA staining along the capillary loops and mesangial IgM deposition (Fig. 2A-C). In addition, IF staining for C3c showed weak granular staining along the capillary loops (Fig. 2D). IF results for IgG, C1q, and C4c were negative. Electron microscopy (EM) revealed global foot process effacement, slight thickening of the glomerular basement membrane, and a small nodule with an increased mesangial matrix (Fig. 3A). An electron-dense exudative lesion was present in the subendothelial space of the glomerular tuft (Fig. 3B) and in the area of segmental sclerosis (Fig. 3C). An exudative lesion was also revealed in the arteriolar wall (Fig. 3D). In addition, a fibrous glomerular crescent was also found in one of the three glomeruli.

Figure 1.

Renal biopsy findings. (A) Light microscopy shows glomeruli with mild expansion of the mesangium (PAS staining, original magnification ×400). (B) Some glomeruli exhibit endocapillary hypercellularity with neutrophils in capillary tufts (PAS staining, original magnification ×400) (C-D) Interstitial inflammatory infiltration in a localized area (C: PAS staining, original magnification ×200) and tubulitis. (D: PAS staining original magnification ×400, yellow arrowheads). PAS: Periodic acid-Schiff

Figure 2.

Immunofluorescence of the glomerulus (original magnification ×200). (A) IgG was negative. (B) IgA shows linear staining along the capillary loops. (C) IgM shows mesangial granular deposits. (D) C3c presents weak staining along the capillary loops.

Figure 3.

Electron microscopy (EM) observations. EM reveals extensive foot process effacement, slight thickening of the glomerular basement membrane, and a small nodule with an increased mesangial matrix (A: original magnification ×1,500). Electron-dense deposition is shown in the subendothelial space of the glomerular tuft (B: original magnification ×3,000) and in the area of segmental sclerosis (C: original magnification ×4,000). Electron-dense deposition is also revealed in the arteriolar wall (D: original magnification ×1,500).

Oral prednisolone was gradually tapered to 5 mg/week during hospitalization, and his serum creatinine level stabilized at around 3.0 mg/dL. The patient was discharged on day 33 of hospitalization. Oral prednisolone was tapered continuously during the follow-up visits, although persistent kidney injury was detected, with a serum creatinine level of approximately 3.0 mg/dL (Fig. 4).

Figure 4.

Clinical course before and after vaccination. Changes in the serum level of Cr (mg/dL), CRP (mg/dL), UPCR (g/gCr), and urinary NAG (U/g creatinine). Cr: creatinine, CRP: C-reactive protein, UPCR: urine protein-to-creatinine ratio, NAG: N-acetyl-β-d-glucosaminide ratio

Discussion

We encountered a case of biopsy-proven AIN in concurrence with glomerular capillary IgA deposition and diabetic changes following administration of the SARS-CoV-2 mRNA vaccine. Although AIN following SARS-CoV-2 vaccine is considered to be a rare entity, approximately 23 such cases have been reported (12-26). Relapse of underlying IgG4-related interstitial nephritis following SARS-CoV-2 mRNA vaccination has also been described (27). However, it is difficult to prove the causality between vaccination and AIN because tubulointerstitial nephritis has a wide range of etiologies, including medication (e.g. non-steroidal anti-inflammatory drugs, proton pump inhibitors, and antibiotics), autoimmune diseases (such as Sjögren's syndrome, IgG4-related disease, sarcoidosis, and rarely ANCA-associated vasculitis), infection, and tubulointerstitial nephritis and uveitis syndrome (28-30). In our case, we considered there to be a strong relationship between the SARS-CoV-2 mRNA vaccination and AIN for the following reasons: a persistent fever and AKI within a short duration following vaccination, no other additional medicines prescribed before hospitalization, and few clinical signs suggestive of infection or autoimmune disease at the initial visit. Furthermore, tubulointerstitial injury was more remarkable than glomerular injury, which may correlate with the acute decline of the renal function post-vaccination. Findings also suggested that staphylococcal infection and antibiotics after steroid therapy may have contributed to the tubulointerstitial damage.

Both immune activation and hypersensitivity to the vaccine and polyethylene glycol (PEG) excipients have been proposed to be involved in the pathophysiology of SARS-CoV-2 vaccine-associated AIN (13). In our case, we did not observe any physical signs of hypersensitivity reactions, such as cutaneous rash, itching, or asthma. Furthermore, we did not perform a drug-induced lymphocyte stimulation test, which detects the presence of drug-sensitized T cells; however, clinical findings of a persistent fever and elevated IgE levels suggested a type I allergic reaction to the vaccine. SARS-CoV-2 can directly infect tubular cells via angiotensin-converting enzyme 2 receptor and induce cell injury with subsequent fibrosis, as confirmed by autopsy kidney samples from patients with COVID-19 as well as human-induced pluripotent stem cells (31). In addition, cytokine storms triggered by viral infection may be responsible for tubular injury in patients with COVID-19 (32,33). These observations suggest that SARS-CoV-2 mRNA vaccines can trigger tubular damage and fibrosis via similar mechanisms, although this may require further investigation.

Our patient's medical history included diabetes mellitus with poorly controlled glycemia by oral glucose-lowering agents; however, his renal dysfunction was mild with normal urinalysis findings at least two years before admission. Consistent with his medical history, the pathological changes in the early phase of diabetes nephropathies, such as thickening of the GBM, small nodules with increased mesangial matrix, and exudative lesions in the glomeruli and arteriolar wall, were observed in the present case. Furthermore, a recent study demonstrated that immunoglobulin deposition, including IgM and C3, was frequently observed in the pathological findings of diabetic nephropathy (23). Therefore, the present case of mesangial IgM and C3 deposition may have been attributable to diabetic changes.

Reported cases of glomerulonephritis following SARS-CoV-2 mRNA vaccination have also been increasing as the number of vaccinated people is increasing worldwide. IgAN is the most commonly reported type of glomerulonephritis, and both new-onset and relapse cases of the disease have been described (3). Recently, Yokote et al. reported three cases of glomerular capillary IgA deposition following SARS-CoV-2 vaccination (34). We did not consider the present case as one of IgAN for the following reasons: IF showed linear but not granular staining of IgA along the capillary loops, light microscopy did not demonstrate proliferative glomerulonephritis, and EM did not show electron-dense deposits in the mesangial and paramesangial areas.

The linear pattern of IgA along the capillary loops raises the possibility of IgA-type anti-GBM glomerulonephritis, a rare type of anti-GBM disease with only 13 cases reported thus far to our knowledge (35-46). However, commercial-based serological tests for GBM detect IgG-type anti-GBM antibodies but not IgA-type anti-GBM antibodies, so making a diagnosis of IgA-type anti-GBM glomerulonephritis is challenging. Nevertheless, the renal biopsy findings in the present case did not show apparent glomerular crescentic formation, and a urinalysis showed mild microscopic hematuria and non-nephrotic proteinuria, suggesting that glomerular changes had negligible impact on acute renal exacerbation.

It should be noted that CRBSI and antibiotics should be considered contributing factors to renal injury. IgA-dominant acute postinfectious glomerulonephritis (APIGN) has been increasingly recognized as a histological variant of APIGN that typically occurs in association with staphylococcal infection (47). Features more likely to be seen in IgA-dominant APIGN than IgA nephropathy include an initial presentation in the elderly, a history of diabetes mellitus, acute kidney failure, culture-documented staphylococcal infection, hypocomplementemia, diffuse glomerular endocapillary hypercellularity with prominent neutrophil infiltration in light microscopy, stronger IF staining for C3 than IgA, and subepithelial humps on EM (47). Our patient had characteristic findings of APIGN, including an elderly age, history of diabetes, staphylococcal bloodstream infection, neutrophil infiltration in the glomeruli, and subendothelial electron-dense deposits. However, other characteristics of APIGN, such as hypocomplementemia, endocapillary hypercellularity, and stronger IF staining for C3 than IgA, were not present. In addition, mild hematuria and proteinuria suggested weak inflammation of the glomeruli. We therefore considered APIGN unlikely to be the cause of renal injury in our case, although the impact of staphylococcal infection and antibiotics on the tubulointerstitial injury remains possible.

In conclusion, our case represents a distinctive and variable renal pathology of SARS-CoV-2 mRNA vaccination-associated AIN. The immunological response following vaccination may vary among patients, depending on the race, age, sex, and underlying disease. The further accumulation of similar cases will elucidate the pathophysiology of vaccine-associated nephritis, since the number of SARS-CoV-2 mRNA-vaccinated people has been continuously increasing worldwide.

Written and verbal informed consent for publication was obtained from the patient.

Author's disclosure of potential Conflicts of Interest (COI).

Tetsu Akimoto: Endowed department, Terumo, Chugai Pharmaceutical. and Kyowa Kirin.

Financial Support

This case report was supported in part by the Japan Agency for Medical Research and Development (AMED) under Grant Number JP18ek0310010 (to DN).