Abstract
Polyarteritis nodosa (PAN) is a rare vasculitis affecting predominately medium-sized vessels. Our evolving understanding of PAN, including the role of HBV and discovery of monogenic variants such as deficiency of adenosine deaminase 2 (DADA2), has resulted in idiopathic PAN becoming a very rare condition with an annual incidence of 0.6–19.9/million. The investigative approach is to assess organ involvement and confirm the diagnosis either by imaging or biopsy. Conventional catheter angiography is being replaced by CT or MR angiography, which provide increasingly good resolution of the typical fusiform narrowing and aneurysm. ANCA-associated vasculitis (AAV) should be excluded. DADA2 is being more frequently identified as causing a PAN-like illness, especially in children; cases presenting in adulthood are now recognised. In children and most adults, genotyping for DADA2 should be undertaken. There are no high-quality randomised controlled trials of treatment in idiopathic PAN. Current approaches have been adapted from those used to treat other types of vasculitis. The five-factor score may be used to stratify patients to identify those needing intensive therapy with glucocorticoids combined with CYC. In order to reduce the burden of glucocorticoid toxicity, the ACR has advocated the use of immunosuppressive drugs in addition to steroids in those not initially receiving CYC. Unlike AAV, there appears to be little role for rituximab in the treatment of idiopathic PAN, although it is occasionally used as salvage therapy. HBV-associated PAN is treated with immunosuppression and antiviral therapy. There is reasonably good evidence to support the use of anti-TNF to treat DADA2.
Keywords: polyarteritis nodosa, vasculitis, arteritis, hepatitis B, ANCA, deficiency of adenosine deaminase 2 (DADA2), VEXAS, glucocorticoids, cyclophosphamide, five-factor score
Key messages.
Idiopathic polyarteritis nodosa is now a very rare condition.
Genetic investigation for DADA2 should be undertaken in all children and most adults.
Treatment is glucocorticoids and cyclophosphamide for severely affected patients.
Less severely affected patients may be treated with glucocorticoids and other immunosuppressive drugs.
Relapse is common.
Introduction
Polyarteritis nodosa (PAN) is a rare form of medium-sized vessel vasculitis. PAN was first described by Karl Rokitansky in 1852, when he described the presence of aneurysmal lesions with nodes in multiple arteries at the postmortem of a 23-year-old man. The first complete description was provided by Kussmaul and Maier in 1866, who described ‘periarteritis nodosa as a unique arterial disease’ [1]. Over the past 160 years there have been considerable advances in our understanding of PAN and its place in the constellation of the vasculitides; in particular, recognition of the pathogenic role of HBV in many cases. Furthermore, monogenic vasculitides such as deficiency of adenosine deaminase 2 (DADA2) and VEXAS (vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic) syndrome have been described.
Along with this there have been developments in the treatment of both viral-associated PAN and idiopathic PAN. The rarity of idiopathic PAN means that there are few randomised controlled trials to guide therapy. The aim of this article is to review our current state of knowledge and provide some thoughts about future developments.
Nomenclature
Our concept of PAN has changed since 1866, morphing from a term used for almost all types of vasculitis to one describing a much more circumscribed form affecting predominately medium-sized blood vessels. In 1948, Davson et al. [2] described a variant characterised by microscopic aneurysms, initially known as microscopic PAN. This condition is now known as microscopic polyangiitis (MPA) and, being a member of the ANCA-associated vasculitides (AAVs), falls outside the scope of this review. In 1952, Pearl Zeek [3] was the first to classify vasculitis by vessel size and included PAN in her classification system of five different types of vasculitis. In 1990, the ACR published classification criteria for seven types of vasculitis, including PAN [4] (Table 1). These criteria did not include either MPA or ANCA.
Table 1.
ACR criteria for PAN.
|
|
|
|
|
|
|
|
|
|
The classification of PAN requires the presence of 3 of the 10 criteria; this gives a specificity 86.6% and sensitivity 82.2%.
Adapted from Lightfoot RW, Michel BA, Bloch DA et al. The American College of Rheumatology 1990 criteria for the classification of polyarteritis nodosa. Arthritis Rheum. 1990;33:1088–93.
The Diagnostic and Classification in Vasculitis Study (DCVAS) dataset was used to investigate the performance of the ACR 1990 criteria in a dataset recruited between 2011 and 2017. The sensitivity for PAN was 40.6% in the DCVAS data compared with 82.2% in the ACR 1990 data, with a specificity of 87.8% and 86.6%, respectively [5]. In this study of 162 patients with a diagnosis of PAN, 47 could also be classified as granulomatosis with polyangiitis (GPA), 13 as eosinophilic granulomatosis with polyangiitis (EGPA) and 4 as both GPA and EGPA. This suggests that ACR 1990 criteria perform badly, especially in discriminating between PAN and GPA or EGPA.
The Chapel Hill Consensus Conference (CHCC) in 1994 provided definitions of both PAN and MPA but not classification criteria. The CHCC 1994 defined PAN as ‘Necrotizing inflammation of the small and medium-sized arteries without glomerulonephritis or vasculitis of the arterioles, capillaries, or venules’ [6]. Medium-sized arteries are the main visceral arteries and their branches. Inflammatory aneurysms and stenoses are a common feature of vasculitis affecting medium-sized vessels. It should be stressed that PAN affects predominately but not exclusively medium-sized vessels and that any size artery may be involved.
In an attempt to provide a workable system for classification of the vasculitides, an algorithm was developed by consensus based on these definitions and the ACR criteria [7]. This algorithm adopted a stepwise approach to classification using the ACR criteria and CHCC definitions with the aim of minimising the number of unclassifiable patients.
The CHCC in 2012 revised the definitions to emphasize that PAN was not associated with the presence of ANCA: ‘Necrotizing arteritis of medium or small arteries without GN or vasculitis in arterioles, capillaries, or venules, and not associated with ANCAs’ [8]. The CHCC 2012 also introduced the concept of single-organ vasculitis and vasculitis associated with a probable aetiology, among which was hepatitis B–associated vasculitis. PAN occurring in association with a recognised cause is often termed secondary PAN, as opposed to idiopathic PAN where no cause is recognised.
HBV-associated PAN is now defined separately from ‘idiopathic’ PAN in the CHCC system. Cutaneous PAN was defined as PAN with vasculitis confined to the skin without any systemic involvement. The CHCC considered cutaneous PAN as a form of single-organ vasculitis [8]. These evolving changes are summarised in Table 2.
Table 2.
Historical overview of development of PAN.
| Kussmaul and Maier [1] | 1866 | First complete description of PAN |
| Davson et al. [2] | 1948 | Concept of microscopic polyarteritis as a separate entity from PAN with a clinically different presentation |
| Zeek [3] | First classification of vasculitis by vessel size | |
| ACR [4] | 1990 | Classification criteria for PAN among seven types of vasculitis. Poor specificity and sensitivity. MPA and ANCA not included. Not intended for diagnosis of individual cases |
| CHCC [6] | 1994 | First working definition of PAN |
| EMA [7] | 2006 | Algorithm incorporating both CHCC and ACR into a usable system for epidemiological and clinical studies |
| CHCC [8] | 2012 | Updated definitions for vasculitis. PAN distinct from AAV. Recognition of secondary (e.g. viral) causes of PAN |
| Cutaneous PAN considered a single organ vasculitis | ||
| ACR/EULAR | 2022 | Data-driven classification criteria for vasculitis, but PAN still not included |
EMA: European Medicines Agency.
Modified from Watts RA. Idiopathic polyarteritis nodosa—does it still exist? Viewpoint 1: as our knowledge makes progress, idiopathic polyarteritis nodosa is fading away. Rheumatology (Oxford) 2025;64:i79–81.
A consequence of the acceptance in the CHCC 2012 nomenclature of vasculitis associated with a specific aetiology has been the recognition of an increasing number of causes of PAN, including infection, especially HBV but also HCV, HIV, human parvovirus B19, EBV and possibly COVID19, and haematological malignancy such as hairy cell leukaemia and myelodysplasia. PAN occurs as a comorbidity of FMF [9].
In addition, the monogenic disease DADA2, first identified in 2014, has been recognised as causing clinical features that are often indistinguishable from PAN [10, 11]. VEXAS syndrome, described in 2020, due to somatic mutations in UBA1, may present with PAN-like clinical features [12].
Occurrence
Our changing understanding of PAN has had a significant impact on the reported incidence [13]. The incidence of PAN has been reported to range between 0.9 and 8.0/million population per year. It has been reported in all populations, mainly occurring in the age range 40–60 years, with a male:female ratio of 1.5:1 (reviewed in Watts et al. [14]). A recent report from Poland using a hospital morbidity register between 2008 and 2013 suggested that the incidence had decreased from 3.3/million in 2008 to 1.9/million in 2013 [15]. Recent data from Norwich, UK between 2011 and 2020 reported a lower incidence at 0.6/million [16]. Earlier studies did not distinguish between idiopathic PAN and that associated with a specific aetiology, such as HBV. Recent series have shown that the proportion of patients with HBV infection has fallen to <5% [15, 17, 18] and none of the patients in the Norwich cohort had HBV infection [16]. These data reflect the decreasing incidence of HBV due to vaccination and education regarding safe needle-sharing practices and safe sex. As a result of our evolving understanding of PAN, idiopathic PAN has become a very rare disease [19].
There are no current data on the prevalence of PAN, reflecting the decreasing incidence of PAN and making these studies hard to conduct. As yet, there are no epidemiological data on the occurrence of DADA2 or VEXAS syndrome.
Prognosis
Early studies included a mixture of idiopathic PAN and secondary PAN, predominately HBV-PAN as other secondary causes had not been identified. The use of the ACR criteria also resulted in some patients with PAN being classified as PAN whereas they would now be considered as having MPA.
An international study reported that the 1-, 5- and 10-year survival rates were 97.1%, 94.0% and 89.0%, respectively. Predictors of mortality included age >65 years, serum creatine at diagnosis >140 μmol/l, gastrointestinal manifestations and CNS involvement [20].
A French retrospective study conducted in the post-HBV era study reported 1-, 5- and 10-year survival of 92.9%, 87.4% and 80.7%, respectively. Independent risk factors for death in multivariate analysis were age >65 years, necrotic purpura, acute kidney injury and secondary PAN [17].
A recent series from India reported an overall mortality of 10.2% with a median follow-up of 53.3 months [18].
PAN was at one time considered to be a one-shot non-relapsing disease. However, several studies have shown that over prolonged follow-up, relapse in patients without HBV infection is common. A review in 2017 noted that relapse or refractory disease occurred in 12–32% of patients with non-HBV PAN [21]. The international study reported that relapse occurred in 48.5% of patients with idiopathic PAN over a follow-up period of 59.6 months [20].
Clinical features
PAN is a multisystem disease potentially involving multiple organs. An international study of 358 patients recruited from observational cohorts across Europe, Japan and North America reported a mean age at diagnosis of 44.3 years; constitutional symptoms were present in 71.5%, cutaneous in 70.5%, musculoskeletal in 69.1% and neurological in 48.0%. This study excluded patients with HBV-PAN, ADA2 and FMF [20].
The majority of patients present with constitutional symptoms, including cutaneous, musculoskeletal, neuropathic or gastrointestinal involvement (Table 3). A French retrospective study reported on 55 patients with secondary PAN out of a total of 196 patients [17]. The most common causes of secondary PAN were myelodysplastic syndrome (17/55) and solid cancer (13/55). Other causes included lymphoproliferative disorder in nine patients, autoinflammatory disorder in eight patients (including DADA2) and hypereosinophilic syndrome in five patients. Viral infection was uncommon, with no cases of HBV, one HCV and two HIV. There was no difference in clinical features between primary and secondary PAN.
Table 3.
Clinical features of idiopathic PAN in recent series.
| Clinical features | Rohmer et al. [17] | Karadag et al. [20] | Jha et al. [18] |
| Date | 2005–2019 | 1990–2020 | 2005–2023 |
| Patients, n | 141 | 282 | 53 |
| Location | France | International | India |
| Classification | FSVG | EMA | ACR |
| Hepatitis B related | Excluded | Excluded | 3.77 (2 cases) |
| Age (years) | Median 53.5 (IQR 17) | Mean 45.1 (s.d. 17.6) | Mean 38.5 (s.d. 15.2) |
| Male:female ratio | 1.8 | 1.8 | 1.9 |
| Fever, n | 53.9 | 45.2 | 49.1 |
| Weight loss, n | 52.5 | 50.6 | 54.7 |
| Myalgias, n | 36.9 | 51.7 | 67.9 |
| Arthralgia, n | 44.7 | 53.2 | 45.3 |
| Cutaneous, n | 59.6 | 62.5 | 77.4 |
| Necrosis/ulceration, n | 40.4 | 13.9 | 32.1 |
| Hypertension, n | 17.7 | 36.3 | 43.4 |
| Renal involvement, n | 22.7 | 48.2 | 47.2 |
| Abdominal pain, n | 22.0 | 21.8 | 26.4 |
| PNS involved, n | 53.3 | 52.7 | 77.4 |
| CNS involved, n | 8.5 | 16.8 | 9.4 |
| Orchitis, n | 15.6 | 12.6 | 16.1 |
FSVG: French Vasculitis Study Group; EMA: European Medicines Agency; PNS: peripheral nervous system.
The reduction in occurrence of HBV-associated PAN over the past 20 years means that there are no contemporary series of patients. The majority of recent reports have been isolated cases. In the era of HBV-PAN, peripheral neuropathy, abdominal pain, cardiomyopathy, orchitis and hypertension were more common in patients with HBV-PAN compared with non-HBV-PAN [22]. A recent study updating the French experience did not include any patients with HBV infection [17].
Systemic features are common, with the majority of patients having constitutional symptoms including fever >38°C, drenching night sweats, poor appetite and weight loss.
Cutaneous manifestations are frequent, a variety of lesions are seen, including palpable purpura, livedo and subcutaneous nodules. Digital ischaemia leading to gangrene can occur.
Gastrointestinal involvement is one of the more common features of PAN. Symptoms can be non-specific, but occurrence in association with other possible features of PAN should heighten suspicion of PAN. Bowel ischaemia is suggested by postprandial worsening of abdominal discomfort. The small bowel and gall bladder are the most common sites of gastrointestinal involvement. Bowel ischaemia can progress to ulceration and bowel necrosis. The diagnosis of vasculitis may only be made at histological examination of surgically removed organs, such as the gall bladder or lengths of ischaemic bowel.
Renal involvement is due to involvement of the intrarenal arteries, presenting with haematuria, proteinuria and renovascular hypertension. Haematuria and proteinuria is due to rupture of microaneurysms in the renal parenchyma and not glomerulonephritis. Active urinary sediment with significant proteinuria is suggestive of glomerulonephritis and a diagnosis of AAV, particularly MPA. Serum creatinine is usually within the reference range. In men orchitis can occur.
Neurological involvement is frequent. Peripheral nerve involvement is due to ischaemia of the vasa nervorum. This can present classically as a mononeuritis multiplex, but a length-dependent predominately sensory neuropathy can also occur. Typical symptoms are pain, paraesthesia and motor loss. CNS involvement is uncommon at presentation, occurring more commonly as a late manifestation.
Testicular involvement with orchitis presenting with pain, swelling or induration occurs in ≈20% of patients. Orchitis is one of the most characteristic features of idiopathic PAN [22] and is more common in polyarteritis complicated by gastrointestinal involvement and vasculitis associated with HBV infection. It is rarely the first manifestation of the disease, is generally unilateral and is caused by ischaemia of the testicular artery.
Respiratory involvement due to vasculitis is uncommon. Asthma is suggestive of EGPA. Interstitial lung disease can be the presenting feature of MPA. Granulomata occur in GPA but not PAN.
ENT involvement is very uncommon and, if present, should suggest alternative diagnoses, in particular GPA.
Ocular involvement is also uncommon and, if present, is suggestive of GPA, Behçet’s syndrome, Erdheim–Chester disease and VEXAS syndrome.
Single-organ PAN is by definition PAN isolated to a single organ. The diagnosis is critically dependent on the intensity of the search for other organ manifestations. Isolated cutaneous PAN is the most common form of single-organ PAN, but any organ can be involved. Isolated polyarteritis of the appendix, gall bladder, uterus or testis is well recognized, though uncommon. Vasculitic peripheral neuropathy can occur without other systemic or constitutional symptoms. The diagnosis of PAN in more uncommon organs such as the gall bladder or uterus is often only made by histological examination of the resected specimen.
A recent study from France compared the clinical features of patients with idiopathic PAN with VEXAS-PAN and chronic myelomonocytic (CMML)-associated PAN (CMML-PAN). Patents with VEXAS-PAN and CMML-PAN were older than those with idiopathic PAN. VEXAS-PAN was associated with more frequent skin involvement (100%), orchitis (67%) and ocular disease (58%) than primary PAN. The clinical features of CMML-PAN were similar to those of idiopathic PAN [23].
DADA2 is the first recognised monogenic vasculitis syndrome. The condition is due to autosomal recessive mutations of the ADA2 gene encoding the enzymatic protein adenosine deaminase, which catalyses the conversion of free adenosine and 2′-deoxyadenosine to inosine and 2′-deoxyinosine, respectively. Decreased activity of the ADA2 enzyme leads to increased adenosine levels, which trigger a pro-inflammatory cascade. The first description of DADA2 was in 2014 when two groups independently described an inherited form of vasculitis that most frequently affected children with features strongly suggestive of PAN [10, 11].
A large number of mutations have been reported affecting many functions of the enzyme, leading to highly variable clinical manifestations. The carrier frequency of pathogenic ADA2 variants has been estimated to be ≈1 in 236 persons, making DADA2 one of the more common monogenic autoinflammatory conditions. The vasculopathy can mimic PAN. DADA2 should be suspected in individuals presenting with early onset fever, rash and stroke, even in the absence of a family history. The initial patients were all children, but adult-onset DADA2 is now well-recognised, with up to 50% of patients in some series being adult onset [24].
VEXAS syndrome is a recently described adult-onset monogenic autoinflammatory condition characterised by systemic inflammation and bone marrow failure [12]. It is caused by acquired mutations in the UBA1 gene restricted to haematopoietic cells, disrupting the functioning of the E1 ubiquitin-activating enzyme, leading to impaired protein degradation pathways and variable clinical manifestations. The mutations drive ubiquitin-dependant protein homeostasis, triggering proteotoxic and inflammatory stress that drives systemic inflammation, cytopaenias and clonal haematopoiesis. Clonal haematopoiesis is the clonal expansion of haematopoietic stem and progenitor cells driven by somatic mutations of genes involved in epigenetic regulation, tumour suppression, signal transduction and RNA splicing. Somatic mutations in UBA1 can result in clonal haematopoiesis. Clonal haematopoiesis of indeterminate potential (CHIP) is associated with an increased risk of atherosclerotic disease. The somatic mutations lead to proliferation of haematopoietic progenitor cells resulting in increased activation of granulocytes and monocytes and inflammation. Mutations in UBA1 often coexist with other CHIP mutations.
A systematic review of published cases reported skin involvement is the most common clinical feature, affecting >81.8% of patients, with constitutional symptoms in 69.4%, respiratory in 61.3%, arthropathy (arthritis or arthralgia) in 47.3% and ocular in 44.3%. Skin involvement includes vasculitis lesions (neutrophil dermatoses, purpura and livedo), nodules, erythema nodosum and Sweet syndrome. Myelodysplasia was present in 35.8% of patients [25]. Neurologic involvement is relatively uncommon. Venous thromboembolism occurs in 42%. Chondritis (39%) can resemble relapsing polychondritis, most frequently affecting the ears.
Diagnostic approach
The diagnosis of PAN is based, like other types of vasculitis, on the clinical suspicion of vasculitis followed by investigations to confirm the presence of vasculitis, its specific subtype and organ involvement. Polyarteritis is a multisystem disease, hence the investigative strategy is, by necessity, broad, to exclude infection (including endocarditis), malignancy, pseudo-vasculitic syndromes such as cholesterol emboli, atrial myxoma and anti-phospholipid syndrome. Other types of vasculitis, including AAV, should be excluded.
Basic blood investigations include full blood count, inflammatory markers and renal and liver function. A full blood count may show a normochromic normocytic anaemia with leucocytosis and thrombocytosis. There may be evidence of an underlying cause such as a myelodysplastic syndrome or lymphoproliferative disorder. Eosinophilia points to a diagnosis of EGPA.
Inflammatory markers such as CRP and ESR are usually elevated. Serum creatinine is often normal. Abnormal liver function suggests the possibility of a viral hepatitis. There is often a polyclonal increase in serum gamma globulins.
Serum ANCA status should be established; patients with PAN are ANCA negative. Positive ANCA should direct suspicion towards one of the AAVs, including MPA.
Although increasingly uncommon, viral serology should be checked, including HBV, HCV and HIV, especially in patients with abnormal liver function.
PAN was traditionally diagnosed on catheter angiography of the renal or mesenteric arteries, the typical findings being fusiform narrowing and aneurysm in small and medium-sized arteries. Catheter angiography is now being superseded by CT or MR angiography. CT may be better at visualising the distal mesenteric arteries than MR. There remains a possible role for catheter angiography in patients in whom there is a high index of suspicion [26].
18F-FDG-PET/CT has been useful in the diagnosis of PAN, especially muscular involvement, as increased tracer uptake may also be seen in bone marrow and the spleen. It is particularly useful in patients with non-specific features and for excluding other differential diagnoses such as malignancy and infection [27].
The gold standard for diagnosis is a biopsy showing fibrinoid necrosis in medium-sized arteries.
There may be pointers towards an underlying cause, such as an underlying malignancy. A bone marrow examination may be required to confirm myelodysplasia or lymphoproliferative disorder. Suggestion of a solid malignancy should be investigated with a PET/CT scan.
The possibility of DADA2 should be considered in children and there are increasing reports of the diagnosis being made in adults. Genetic screening for DADA2 should be undertaken.
Patients presenting with myelodysplasia, including those with unclassifiable myelodysplastic syndrome as part of a vasculitic syndrome should be screened for UBA1 mutations as well as other CHIP mutations. Testing is especially necessary when haematological abnormalities coexist with inflammatory features such as recurrent fever, relapsing polychondritis, cutaneous vasculitis or Sweet syndrome.
Differential diagnosis
The differential diagnosis of PAN is very broad, including other types of vasculitis, infections and non-inflammatory conditions. The differential diagnosis includes EGPA, MPA, IgA vasculitis, cryoglobulinaemic vasculitis and drug-induced vasculitis. Infections include infectious endocarditis and mycotic aneurysm with embolization. A number of non-inflammatory conditions may mimic PAN, including atrial myxoma, fibromuscular dysplasia and segmental mediolysis.
Pathology
The pathological lesions of PAN are focal necrotizing inflammatory lesions that extend through the wall of small and medium-sized arteries. The inflammation is characterized by fibrinoid necrosis and pleomorphic cellular infiltration by predominantly macrophages and lymphocytes and variable numbers of neutrophils and eosinophils. A significant eosinophilic infiltrate favours the diagnosis of EGPA. The normal architecture of the vessel wall, including the elastic lamina, is disrupted. Thrombosis or aneurysmal dilation may be present at the site of the lesion. Healed areas of arteritis show proliferation of fibrous tissue and endothelial cells, which may lead to occlusion of the vessel. Lesions at all stages of progression and healing may be seen. The focal nature of the inflammation increases the risk for a false-negative biopsy result when the tissue sample is small. The finding of perivascular inflammation or intimal proliferation without fibrinoid necrosis of the arterial wall suggests but does not confirm the diagnosis of polyarteritis.
Assessment of specific organ involvement
Renal function should be routinely assessed including urine protein:creatinine ratio. PAN causes renovascular hypertension due to involvement of the renal arteries rather than a glomerulonephritis. Significant nephrotic-range proteinuria should prompt consideration of an alternative diagnosis, including a glomerulonephritis. Imaging of renal arteries may diagnostic, with demonstration of microaneurysms and/or stenotic lesions.
Imaging is advocated to establish a diagnosis and the extent of abdominal involvement. The gold standard test traditionally was catheter contrast angiography, however, this is being replaced by CT or MR angiography as the resolution of these techniques provides information on the vessel wall that conventional angiography cannot.
If PAN is suspected of involving the skin, a biopsy is the recommended method of confirming the diagnosis. A deep skin biopsy should be obtained, reaching down to the medium-sized vessels of the dermis [26]. A punch biopsy may not reach sufficiently deep into the dermis.
Neurophysiology typically shows a mononeuritis multiplex. It is important in cases of suspected neurological involvement to undertake neurophysiology of all four limbs, not just clinically affected regions, as this may uncover evidence of involvement of clinically unsuspected nerves.
Where possible, the diagnosis should be confirmed with a biopsy. A combined nerve and muscle biopsy should be obtained, as this provides a better diagnostic yield [26]. Biopsies should be taken from a clinically involved nerve and should be from a sensory nerve, not a motor nerve, to avoid causing a motor deficit.
The possibility of DADA2 should be considered in children, and there are increasing reports of the diagnosis being made in adults. Genetic screening for DADA2 should be undertaken.
The systemic nature of VEXAS syndrome means that it should be considered in patients presenting with an autoinflammatory, possibly vasculitic condition, especially with haematological abnormalities. The diagnosis is made on genetic testing for UBA1 mutations.
Assessment of disease severity
The five-factor score (FFS) was initially described in 1996 as a prognostic tool utilising serum creatinine, proteinuria, cardiomyopathy and gastrointestinal and CNS involvement. The patients used to develop the FFS were derived from a mixed cohort including non-HBV-PAN, HBV-PAN, MPA and EGPA [28]. The French vasculitis study group showed that age >65 years, hypertension and gastrointestinal manifestations were independent risk factors for death [22]. The global study showed an increased FFS was associated with increased mortality. Independent predictors of mortality were age at diagnosis ≥65 years, serum creatinine >140 µmol/l, gastrointestinal involvement and CNS involvement [20].
Therapeutic approaches
There are no randomised controlled trials to guide treatment of idiopathic PAN. The treatment approach has been adapted from other forms of vasculitis, mainly AAV, where there are now a number of high-quality trials to guide therapy. Glucocorticoids combined with CYC remain the lynchpin of treatment, but direct evidence of efficacy is lacking as early trials of this combination did not adequately distinguish between idiopathic PAN and HBV-PAN or other types of vasculitis. The ACR guidelines conditionally recommend pulse intravenous steroids over high-dose oral steroids in patients with life-threatening manifestations such as visceral infarction [26] (Fig. 1).
Figure 1.

Flow chart showing the approach to treatment for the management of PAN. Source: Chung et al. 2021 American College of Rheumatology/Vasculitis Foundation guideline for the management of polyarteritis nodosa. Arthritis Care Res (Hoboken). 2021;73:1061–1070
For patients with less severe disease, oral glucocorticoids are preferred, being associated with less glucocorticoid toxicity. As with other forms of vasculitis, lower glucocorticoid dose regimens are being explored.
CYC has been studied in PAN. In a small randomised study, 24 patients were treated with glucocorticoids and then randomised to long- (12 pulses) or short-duration (6 pulses) CYC treatment. Remission was achieved in 88% of patients with a 32% relapse rate at 10 years [21]. The ACR recommends conditionally initiating treatment with CYC and high-dose steroids [26]. Patients unable to tolerate CYC may be treated with other immunosuppressive drugs, including AZA, MTX and MMF. There is little direct trial data supporting the use of these drugs.
The combination of high-dose glucocorticoids and CYC is associated with significant morbidity. As a result, regimens using lower-dose glucocorticoids for some patients or biologic treatments are being explored. For many of these agents the data are derived from case series and not prospective randomised trials. A single-centre observational study from India suggested that MMF was similar to CYC in maintaining relapse-free survival [18].
The long-standing standard of care is based on the use of the FFS to identify patients who may need more aggressive treatment. Less severe disease is treated with glucocorticoids alone, while more severe disease is treated with glucocorticoids in combination with immunosuppressives, traditionally CYC. This has recently been challenged by the ACR guidelines group, who conditionally advocate the use immunosuppressive agents combined with glucocorticoids in preference to glucocorticoids alone, the rationale being that the overall steroids dose may be reduced with use of the combination [26].
The ACR guidelines group recommends that plasmapheresis should not be used in patients with PAN, as there is no evidence to support its use [26].
Biologic agents that target pro-inflammatory cytokines have been used successfully in other types of vasculitis. Rituximab has become the standard of care in AAV, replacing CYC in most situations, but there is little evidence to support its use in idiopathic PAN. It is occasionally used as a salvage therapy before anticytokine agents.
Relapsing or refractory disease occurs frequently in patients with severe non-HBV-PAN. An open-label multicentre study of refractory cases suggested that tocilizumab may have a better response rate in these patients than either TNF blockers or rituximab [29]. Cottu et al. [30] recently reported on a series of refractory patients treated with tocilizumab showing rapid cutaneous benefit and glucocorticoid sparing.
The optimum duration of treatment has not been established. The duration of treatment therefore depends on the patient’s clinical condition and development of drug toxicities. In other types of potentially relapsing vasculitis, shorter less intense treatment has been associated with an increased rate of relapse. Thus the duration of treatment should be individualised.
The successful use of Janus kinase (JAK) inhibitors has been reported in isolated case reports [31, 32].
Treatment for virus-associated PAN, particularly HBV, is a combination antiviral therapy with immunosuppression to clear the viral load and reduce vascular inflammation. Immunosuppression is typically high-dose glucocorticoids combined with, in severe cases, CYC or AZA, or MTX for less severe cases. Antiviral agents include lamivudine given for at least 6 months. The antiviral component of treatment should be in line with current guidelines for treatment of HBV.
The treatment of DADA2 depends on the clinical manifestations. The genotype does not predict disease severity. Glucocorticoids and other immunosuppressive agents have been tried. Anti-TNF is the mainstay of treatment, being beneficial for vasculitis and inflammatory lesions. The mortality is ≈20% at 10 years [24]. The use of anti-TNF in DADA2 is strongly recommended by the ACR guidelines group [26].
The treatment of VEXAS syndrome with glucocorticoids, IL-6 inhibitors and JAK inhibitors provided only partial disease control. Targeted therapies aimed at the mutant clones are being investigated. Allogenic haematopoietic stem cell transplantation is the only curative strategy [33, 34].
Future directions
The history of PAN over the past 160 years shows that it is becoming a progressively rarer disease and it is likely that as other causes, both environmental and genetic, are discovered, truly idiopathic PAN will become vanishingly uncommon. Haematologically driven vasculitides will become a focus of active research expanding beyond VEXAS, exploring the relationship between acquired bone marrow mutations and systemic inflammation.
Current treatment approaches are based on those used for other types of vasculitis, notably AAV. A better understanding of the aetiology will result in more specific and targeted therapies. DADA2 is a good example of this, as anti-TNF is the most effective treatment but appears to have a limited role in idiopathic non-viral PAN. Genetic therapies will be developed for DADA2 and stem cell transplantation for VEXAS.
Conclusions
Our evolving understanding of PAN, including the role of HBV and the discovery of monogenic variants, has resulted in idiopathic PAN becoming a very rare condition. Successful strategies for the prevention of HBV infection have also reduced the occurrence of HBV-PAN. The investigative approach is to assess organ involvement and confirm the diagnosis either by imaging or biopsy. In children and most adults, genotyping for DADA2 should be undertaken. There are no trials of treatment for idiopathic PAN and current approaches have been adapted from those used to treat other types of vasculitis, such as AAV. Glucocorticoids combined with, in severe cases, CYC and in less severe cases with AZA, MTX or MMF is the most widely adopted approach. HBV-PAN is treated with immunosuppression and antiviral therapy. There is reasonably good evidence to support the use of anti-TNF to treat DADA2.
Data availability
No new data were generated or analysed in support of this article.
Author’s contributions
The author was invited to contribute this review, he was responsible for literature review, writing and editing the manuscript.
Funding
No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.
Disclosure statement: The author has declared no conflicts of interest.
References
- 1. Matteson EL. A history of early investigation in polyarteritis nodosa. Arthritis Care Res 1999;12:294–302. [PubMed] [Google Scholar]
- 2. Davson J, Ball J, Platt R. The kidney in periarteritis nodosa. Q J Med 1948;17:175–202. [PubMed] [Google Scholar]
- 3. Zeek PM. Periarteritis nodosa; a critical review. Am J Clin Pathol 1952;22:777–90. [DOI] [PubMed] [Google Scholar]
- 4. Lightfoot RW, Michel BA, Bloch DA et al. The American College of Rheumatology 1990 criteria for the classification of polyarteritis nodosa. Arthritis Rheum 1990;33:1088–93. [DOI] [PubMed] [Google Scholar]
- 5. Seeliger B, Sznajd J, Robson JC et al. Are the 1990 American College of Rheumatology vasculitis classification criteria still valid? Rheumatology (Oxford) 2017;56:1154–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Jennette JC, Falk RJ, Andrassy K et al. Nomenclature of systemic vasculitides. Proposal of an international consensus conference. Arthritis Rheum 1994;37:187–92. [DOI] [PubMed] [Google Scholar]
- 7. Watts R, Lane S, Hanslik T et al. Development and validation of a consensus methodology for the classification of the ANCA-associated vasculitides and polyarteritis nodosa for epidemiological studies. Ann Rheum Dis 2007;66:222–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jennette J, Falk RJ, Bacon PA et al. 2012 Revised International Chapel Hill Consensus Conference nomenclature of vasculitides. Arthritis Rheum 2013;65:1–11. [DOI] [PubMed] [Google Scholar]
- 9. Ozen S, Ben-Chetrit E, Bakkaloglu A et al. Polyarteritis nodosa in patients with familial mediterranean fever (FMF): a concomitant disease or a feature of FMF? Semin Arthritis Rheum 2001;30:281–7. [DOI] [PubMed] [Google Scholar]
- 10. Zhou Q, Yang D, Ombrello AK et al. Early-onset stroke and vasculopathy associated with mutations in ADA2. N Engl J Med 2014;370:911–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Navon Elkan P, Pierce SB, Segel R et al. Mutant adenosine deaminase 2 in a polyarteritis nodosa vasculopathy. N Engl J Med 2014;370:921–31. [DOI] [PubMed] [Google Scholar]
- 12. Beck DB, Ferrada MA, Sikora KA et al. Somatic mutations in UBA1 and severe adult-onset autoinflammatory disease. N Engl J Med 2020;383:2628–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Watts RA, Jolliffe VA, Carruthers DM, Lockwood M, Scott DGI. Effect of classification on the incidence of polyarteritis nodosa and microscopic polyangiitis. Arthritis Rheum 1996;39:1208–12. [DOI] [PubMed] [Google Scholar]
- 14. Watts RA, Hatemi G, Burns JC, Mohammad AJ. Global epidemiology of vasculitis. Nat Rev Rheumatol 2022;18:22–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kanecki K, Nitsch-Osuch A, Gorynski P et al. Polyarteritis nodosa: decreasing incidence in Poland. Arch Med Sci 2019;15:1308–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Fordham S, Ashurst K, Bartoletti A et al. Incidence of ANCA-associated vasculitis and polyarteritis nodosa in Norfolk, UK, from 2011 to 2020. Rheumatology (Oxford) 2025;64:3718–23. [DOI] [PubMed] [Google Scholar]
- 17. Rohmer J, Nguyen Y, Trefond L et al. Clinical features and long-term outcomes of patients with systemic polyarteritis nodosa diagnosed since 2005: data from 196 patients. J Autoimmun 2023;139:103093. [DOI] [PubMed] [Google Scholar]
- 18. Jha A, R C, Yadav B et al. The clinical characteristics and treatment outcomes of patients with systemic polyarteritis nodosa: a single-centre study from India. Rheumatology (Oxford) 2025;64:3710–7. [DOI] [PubMed] [Google Scholar]
- 19. Watts RA. Idiopathic polyarteritis nodosa – does it still exist? Viewpoint 1: as our knowledge makes progress, idiopathic polyarteritis nodosa is fading away. Rheumatology (Oxford) 2025;64:i79–81. [DOI] [PubMed] [Google Scholar]
- 20. Karadag O, Bolek EC, Ayan G et al. Clinical characteristics and outcomes of polyarteritis nodosa: an international study. Arthritis Rheumatol 2024;76:1120–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Samson M, Puéchal X, Mouthon L et al. Microscopic polyangiitis and non-HBV polyarteritis nodosa with poor-prognosis factors: 10-year results of the prospective CHUSPAN trial. Clin Exp Rheumatol 2017;35(Suppl 103):176–84. [PubMed] [Google Scholar]
- 22. Pagnoux C, Seror R, Henegar C et al. Clinical features and outcomes in 348 patients with polyarteritis nodosa: a systematic retrospective study of patients diagnosed between 1963 and 2005 and entered into the French Vasculitis Study Group database. Arthritis Rheum 2010;62:616–26. [DOI] [PubMed] [Google Scholar]
- 23. Cioni P, Nguyen Y, Rohmer J et al. Polyarteritis nodosa associated with VEXAS syndrome and chronic myelomonocytic leukemia: a case–control study. Rheumatology (Oxford). 2026;65:keaf685. doi: 10.1093/RHEUMATOLOGY/KEAF685. [DOI] [PubMed] [Google Scholar]
- 24. Ayan G, Basaran O, Firlatan B et al. Long-term follow-up of anti-TNF treatment in adult and pediatric DADA2 patients: Insights from real-world data. Int J Rheum Dis 2024;27:e15377. [DOI] [PubMed] [Google Scholar]
- 25. Al-Hakim A, Goldberg S, Gaillard S et al. Clinical features in VEXAS syndrome: a systematic review. Rheumatology (Oxford) 2025;64:5217–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Chung SA, Gorelik M, Langford CA et al. 2021 American College of Rheumatology/Vasculitis Foundation guideline for the management of polyarteritis nodosa. Arthritis Rheumatol 2021;73:1384–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Fagart A, Machet T, Collet G et al. Fluorodeoxyglucose positron emission tomography-computed tomography findings in a first series of 10 patients with polyarteritis nodosa. Rheumatology (Oxford) 2022;61:1663–8. [DOI] [PubMed] [Google Scholar]
- 28. Guillevin L, Lhote F, Gayraud M et al. Prognostic factors in polyarteritis nodosa and Churg-Strauss syndrome. A prospective study in 342 patients. Medicine 1996;75:17–28. [DOI] [PubMed] [Google Scholar]
- 29. Hadjadj J, Canzian A, Karadag O et al. Use of biologics to treat relapsing and/or refractory polyarteritis nodosa: data from a European collaborative study. Rheumatology (Oxford) 2022;62:341–6. [DOI] [PubMed] [Google Scholar]
- 30. Cottu A, Grolleau C, Gandon C et al. Tocilizumab for refractory and/or relapsing polyarteritis nodosa. J Am Acad Dermatol. 2026;94:968–70.doi: 10.1016/J.JAAD.2025.11.014. [DOI] [PubMed] [Google Scholar]
- 31. Nakabayashi A, Iguchi E, Kim DS et al. Efficacy of upadacitinib in refractory polyarteritis nodosa: a case report. Oxf Med Case Rep 2025;2:omea199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Zhou S, Zhang Z, Yao Z, Guo Y. Successful treatment of cutaneous polyarteritis nodosa with baricitinib. J Dermatolog Treat 2024;35:2417965. [DOI] [PubMed] [Google Scholar]
- 33. Holden R, Jeelal Y, McLean-Tooke A, Pathmanathan K, Nolan D. VEXAS: a review of current understandings and emerging treatment strategies. Front Immunol 2025;16:1644404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Groarke EM, Turturice B, Patel BA et al. VEXAS syndrome: a comprehensive review of pathogenesis, clinical spectrum, and therapeutic strategies. Lancet 2026;407:637–48.doi: 10.1016/S0140-6736(25)02164-6. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
No new data were generated or analysed in support of this article.
