Introduction

Novel severe acute respiratory syndrome by coronavirus-2 (SARS-CoV-2) disease (COVID-19) ranges from asymptomatic to severe cases, which are characterized by a severe acute respiratory syndrome. Even mild forms of COVID-19 have been associated with various autoimmune manifestations and accordingly this infection has been proposed as a trigger of several autoimmune diseases [1]. Molecular mimicry and hyperinflammation due to hyperstimulation of the immune system seem to be the potential mechanisms of autoimmunity in COVID-19 [2] and may lead to the appearance of previously non-existent autoantibodies [3]. Furthermore, complement activation in COVID-19 has been shown to activate platelets and neutrophil extracellular traps (NETs) [1, 4] involved in multiple autoimmune diseases [3].

A study conducted in China in patients with critical SARS-CoV-2 pneumonia showed a 50% prevalence of antinuclear antibodies [5] and another study found anti-neutrophil cytoplasmic antibodies (ANCA) in 13% of these patients [6]. Other studies showed an increased incidence of positivity for lupus anticoagulant and antiphospholipid antibodies in COVID-19 patients [7].

Furthermore, there are some reports in the literature describing de novo development of autoimmune diseases associated with COVID-19 [2, 3, 8]. Additionally, patients with preexisting autoimmune diseases may undergo reactivation of their disease after SARS-CoV-2 infection; however, the evidence in the literature about this process is limited. Herein, we report two cases of reactivation of IgA vasculitis after COVID-19 infection.

Case presentation

Case 1

A 27-year-old white male was admitted to our rheumatology department in January 2021 presenting with diffuse arthralgias and cutaneous purpuric lesions in the upper and lower limbs. The patient had been diagnosed with Henoch–Schönlein purpura 3 years earlier, but had no other medical history of interest. At diagnosis, the patient had cutaneous purpura, articular and renal disease (mesangial proliferative glomerulonephritis with IgA deposits). He was successfully treated with oral glucocorticoids for 1 year achieving sustained remission without subsequent relapses.

In our assessment, physical examination revealed generalized palpable purpura distributed over all the extremities, gluteal region and abdomen, without evidence of arthritis, gastrointestinal symptoms or associated fever. A month before, the patient had suffered an asymptomatic SARS-CoV-2 infection diagnosed by positive real-time polymerase chain reaction (RT-PCR) test in nasopharyngeal swab sample, which was indicated due to close contact with a COVID-19-positive subject.

Laboratory results showed normal full blood count, liver and renal function tests, as well as normal coagulation profile, erythrocyte sedimentation rate and C-reactive protein values. Serum IgA levels were increased (357 g/L), while IgG and IgM were normal and the rest of the autoimmune assays, including ANCAs, were negative. Urinalysis was normal.

Given the suspicion of an IgA vasculitis flare, a skin purpuric lesion was biopsied, showing histological results consistent with leukocytoclastic vasculitis. Immunofluorescence microscopy demonstrated predominant IgA deposition thereby confirming IgA vasculitis relapse. The patient received 50 mg/day of prednisone with improvement of purpuric lesions. Three months later, the patient developed microhematuria, with adequate renal function, which was resolved by treatment with azathioprine at a dose of 1.5 mg/kg/day.

Case 2

A 62-year-old Hispanic woman presented to our department with purpuric lesions in the upper limbs with no other associated symptoms. She had been diagnosed with IgA vasculitis four years before, which consisted of anterior scleritis, joint and cutaneous involvement with positive skin biopsy at two different times (immunofluorescence-confirmed leukocytoclastic vasculitis with IgA deposition). ANCA antibodies were negative and C3 decrease was observed, consistent with hypocomplementemic IgA vasculitis. Treatment with high doses of glucocorticoids and methotrexate was enough to achieve remission, which was maintained after glucocorticoid withdrawal under methotrexate monotherapy.

In March 2020, the patient presented symptoms suggestive of SARS-CoV-2 infection, but did not require admission and recovered remaining isolated at home. She then tested positive for IgG COVID-19 antibodies in the next month. Three months later, the patient advanced her scheduled appointment to our clinic. Physical examination revealed a palpable purpura on arms, without signs of arthritis. She was afebrile and her pulse and blood pressure were normal. Laboratory results revealed a reduction in glomerular filtration rate (49 ml/min/1.73 m2 vs previous of 90 ml/min/1.73 m2), mild anemia and lymphopenia with normal acute phase reactants. Urinalysis showed microhematuria and proteinuria up to 1.4 g/24 h. ANCA determination tested positive for proteinase 3 (anti-PR3) antibodies (459 IU/ml, normal upper limit 20 IU/ml) with low complement levels of C3.

A percutaneous renal biopsy was performed showing findings of rapidly progressive glomerulonephritis with fibroepithelial crescents compatible with a diagnosis of ANCA-associated vasculitis. The patient was treated with 3 methylprednisolone boluses of 500 mg/day and intravenous rituximab (4 weekly doses of 375 mg/m2) as induction therapy, followed by oral prednisone 1 mg/kg/day. After 3 months, renal function was partially recovered, proteinuria and ANCA levels were notably reduced and cutaneous lesions were improved. Currently, the patient remains with 10 mg of oral prednisone and rituximab.

Search strategy and case selection

A literature review was performed to identify studies focused on the development of ANCA vasculitis or IgA vasculitis after COVID-19. Accordingly, MEDLINE database was accessed through PubMed and searched for articles published in English or Spanish between March 2020 and October 2021. The search strategy used the following key terms related to vasculitis and COVID-19: (“IgA vasculitis” OR “Henoch–Schönlein purpura” OR “ANCA vasculitis” OR “granulomatosis with polyangiitis” OR “microscopic polyangiitis”) AND (“COVID-19” OR “SARS-CoV-2” OR “coronavirus”). Search strategy is represented in Fig. 1. We selected only patients with confirmed positive test for COVID-19 by serology or RT-PCR in nasopharyngeal swab. Cases with insufficient information to confirm the diagnosis of IgA vasculitis or ANCA vasculitis (positive biopsy, positive ANCA or increase of IgA levels) were not considered. To capture all the available literature, articles were selected with no filter in terms of design, included case reports and case series, and with no limits in the age of the patients. Abstracts or not published results were not included. This search strategy was applied at three different times (May 2021, August 2021 and October 2021) covering until October 25th 2021. Finally, a total of 29 articles were selected for evaluation.

Fig. 1
figure 1

Flowchart of the bibliographic search strategy and selection criteria. ANCA anti-neutrophil cytoplasmic antibodies, COVID-19 coronavirus disease 2019, RT-PCR reverse transcription polymerase chain reaction, SARS-CoV-2 severe acute respiratory syndrome coronavirus-2

Full size image

Results

We identified 16 reports describing 18 cases (12 women/6 men) of new-onset ANCA-associated vasculitis in COVID-19 patients and presumably related to this disease [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23]. The main characteristics of these cases are presented in Table 1. In eight cases the onset of ANCA vasculitis coincided with the infection, six of them had pneumonia. Of the 18 cases, 16 presented organ-threatening disease: 13 patients presented renal involvement (pauci-immune glomerulonephritis) and 11 patients had diffuse alveolar hemorrhage, in 9 of these patients both diseases coexisted. Three patients also presented pulmonary involvement, pulmonary nodules with cavitary lesions in two patients and hyper eosinophilic bronchiolitis in one patient. Only seven patients had leukocytoclastic vasculitis and two patients had arthritis.

Table 1 Reported cases of ANCA-associated vasculitis related to COVID-19
Full size table

Regarding the type of ANCA, anti-MPO antibodies were more frequently detected than anti-PR3 antibodies (nine and seven patients, respectively). All patients received steroids and most cases were treated with immunosuppressive combined therapy: nine with rituximab, nine with plasmapheresis, six with cyclophosphamide (CYC), two with intravenous immunoglobulins (IVIG), one with mycophenolate mofetil (MMF) and the other one with azathioprine. In the follow-up, most patients responded well to glucocorticoids and immunosuppressive agents but two patients died as a result of the disease (both with diffuse alveolar hemorrhage) [10, 20]. Another patient died as a consequence of multiple infections [23]. Notably, three of these cases had a preexistent autoimmune disease [12, 13, 23]. Another case simultaneously developed an antiphospholipid syndrome [18].

Concerning IgA vasculitis related to COVID-19, 15 cases (12 men; 3 women) in 13 reports have been published to date [24,25,26,27,28,29,30,31,32,33,34,35,36], which are summarized in Table 2. Half of the cases were diagnosed in childhood. These cases presented new-onset IgA vasculitis, most of them with palpable purpura (13 patients) and 8 patients developed renal disease (IgA nephropathy). Additionally, eight patients presented with gastrointestinal involvement and three patients presented with arthritis. In eight patients, the onset of vasculitis coincided with the infection. All of these patients were treated with glucocorticoids and four patients received immunosuppressants for renal involvement (1 rituximab, 2 MMF, 1 CYC) with favorable renal response in all cases [25, 30, 36]. No deaths were identified.

Table 2 Reported cases of IgA vasculitis related to COVID-19
Full size table

Discussion

COVID-19 is bringing back many aspects of autoimmune diseases that seemed forgotten. Indeed, SARS-CoV-2 infection could break self-tolerance and trigger systemic autoimmunity. Several reports have suggested that COVID-19 may be followed by immune activation and the development of several autoimmune manifestations [2, 3, 37]; however, there is currently a lack of robust evidence supporting SARS-CoV-2 as the causal trigger of these phenomena [38].

Although the association between COVID-19 and the relapse of vasculitis found in our cases cannot be fully demonstrated and could be incidental, the chronology of events suggests a potential role of the viral infection on the onset of both vasculitis flares. Indeed, other viruses have been proposed as a trigger of vasculitis based on molecular mimicry, tropism for vascular endothelium, immune complex deposition within the vessel walls and autoantibody production [39,40,41]. Causal relationships have been well established between hepatitis C virus and cryoglobulinemic vasculitis and between hepatitis B virus and polyarteritis nodosa [39, 40].

Current evidence supports that the mechanisms involved in SARS-CoV-2 regulation of autoantibody generation, endothelial inflammation and dysfunction, complement activation and NET production may lead to vasculitis [1,2,3,4]. Vasculitis-like phenomena during COVID-19 infection have been described in the literature and numerous reports have described cutaneous vascular lesions in COVID-19 patients [37, 42]. Therefore, some authors suggest that virus–host interactions may lead to both direct and indirect microvasculature damage through endothelial cell inflammation [42]. Additionally, thrombosis, lymphocytic endothelitis, and apoptotic bodies have been found in COVID-19 autopsies [37, 42]. Furthermore, several studies have reported the presence of leukocytoclastic vasculitis in cutaneous biopsies from COVID-19 patients obtained during the active or convalescent phases of this infection [43,44,45].

We present two cases of patients with prior history of IgA vasculitis with ocular, renal, skin and articular involvement, both of them in sustained remission, who developed a new flare of vasculitis shortly after SARS-CoV-2 infection. One case suffered mild COVID-19, while the other remained asymptomatic.

Case 1 had a relapse of the disease with hematuria, arthralgia and cutaneous flare and the second case presented with cutaneous and renal disease consistent with ANCA-associated vasculitis. The absence of ANCA in the patient´s previous history, along with C3 consumption in a well-documented IgA vasculitis prompted us to consider a newly induced ANCA-associated vasculitis; however, recurrent low C3 levels pointed to a relapse of a hypocomplementemic IgA vasculitis, likely in the context of an overlapping vasculitis. We cannot definitely rule out that both entities were present prior COVID19 infection, as ANCA levels were not systematically analyzed during remission at subsequent follow-up.

Our two patients suffered severe manifestations of vasculitis (kidney involvement in both patients). This finding was also described in the reported cases of vasculitis related to COVID-19: renal disease was reported in 15/18 patients with ANCA vasculitis and in 8/15 patients with IgA vasculitis. In line with our findings, the reviewed cases of post-COVID-19 ANCA vasculitis were more severe and with a worse prognosis than those with IgA vasculitis, showing organ-threatening disease in 88% of the cases (16/18) and three deaths. Both renal and pulmonary involvement were very common in ANCA-associated vasculitis. In addition, the reported cases showed a predominance of IgA vasculitis in males and ANCA-associated vasculitis in females.

Current evidence on flares of preexisting autoimmune diseases in COVID-19 patients is very limited. Flares of SLE in patients with COVID-19 have been described [46]. A study in a cohort of Hispanic COVID-19 patients from United States with rheumatic diseases identified COVID-19 positivity as a risk factor for disease flares [47]. To date, we have not found publications reporting cases of relapses of previous controlled IgA vasculitis associated with COVID-19.

Furthermore, to the best of our knowledge, there are no descriptions in the literature of flares of vasculitic diseases associated with COVID-19. Therefore, our cases provide valuable information supporting the novel hypothesis that COVID-19 could act as an immune trigger for vasculitis contributing to flares of prior disease, even with new autoantibody-induced manifestations.

Conclusion

COVID-19 infection could be associated with vasculitis triggering and could induce flares of previous autoimmune diseases. To our knowledge, this is the first description of vasculitis reactivation following COVID-19 infection in patients with preexisting IgA vasculitis. These findings suggest a causal relationship between COVID-19 and vasculitis, although further research is needed to establish solid evidence about this subject.