Rheumatology 2006;45:1276–1281 Advance Access publication 4 April 2006

doi:10.1093/rheumatology/kel091

Polyarteritis nodosa when applying the Chapel Hill nomenclature—a descriptive study on ten patients D. Selga1, A. Mohammad1, G. Sturfelt2 and M. Segelmark1 Objectives. Polyarteritis nodosa (PAN) is a term that has been used to describe a wide variety of vasculitic conditions. In 1994, the Chapel Hill Consensus Conference on the nomenclature of systemic vasculitides proposed that the name classical PAN should be restricted to diseases where there is arteritis in small and medium-sized arteries without the involvement of smaller vessels. Our aim was to describe the symptoms and course of disease in PAN when the microscopic forms are excluded. Methods. All patients with a diagnosis of PAN treated in our departments during the period 1990–2002 were eligible for this study. The diagnosis had to be confirmed by biopsy, angiography or electromyography. Results. Ten patients were eligible for the study. The median age was 46 yrs. Renal involvement was seen in 70% at diagnosis. After 5 yrs, 57% had experienced a relapse, which is equivalent to the relapse rate seen in microscopic polyangiitis. Organ damage was assessed by the Vasculitis Damage Index (VDI) and after 5 yrs cardiovascular and neuropsychiatric damage dominated, followed by renal damage. During the follow-up, two patients developed end-stage renal disease. The annual incidence of PAN in our local catchment area was estimated to 1.6 per million and year. Conclusions. When applying the Chapel Hill nomenclature, PAN is a rare but severe disease with a high incidence of renal involvement and frequent relapses. KEY WORDS: Polyarteritis nodosa, PAN, Vasculitis, Outcome, Incidence, Relapses.

Introduction In 1866, Kussmaul and Maier [1] described a patient dying from a previously unknown disease for which they gave the descriptive name of periarteritis nodosa. This and the other similar designations such as polyarteritis nodosa (PAN) were subsequently used as diagnostic terms for an increasing number of different vasculitic conditions. The separation into specific subgroups of vasculitides was started by Wegener [2] in 1939. In 1948, a microscopic form of polyarteritis was described by Davson et al. [3] who suggested that segmental necrotizing glomerulonephritis was a feature of this form of the disease. In 1952, Zeek [4] pointed out that the term PAN had gradually encompassed a wide variety of vascular conditions and proposed a classification for necrotizing angiitides in which PAN was separated from other forms. Although many authors adopted this terminology, most continued to include glomerulonephritis as a manifestation of PAN. Moreover, the American College of Rheumatology (ACR) classification criteria since 1990 [5] do not distinguish between classical PAN and microscopic polyangiitis (MPA). In 1970, Tre´po and Thivolet [6] described hepatitis-induced PAN. This secondary form of vasculitis was more common in Western Europe in the 1970s due to hepatitis B virus (HBV) infection caused by contaminated blood transfusions. Today, hepatitis-associated PAN is treated with antiviral drugs and the clinical course after such treatment is usually benign. The relationship between anti-neutrophil cytoplasmic antibodies (ANCA) and small vessel vasculitis was first described by van der Woude [7] in 1985. In 1994, the Chapel Hill Consensus Conference (CHCC) on the nomenclature of systemic vasculitides [8] proposed that the name classical PAN should be restricted to disease where there is arteritis in small and medium-sized arteries without the involvement of smaller vessels. According to the 1

CHCC nomenclature, PAN is a systemic necrotizing vasculitic disease of unknown aetiology characterized by inflammation and necrosis of small and medium-sized arteries. After the publication of the CHCC document many patient series have been presented describing the clinical features of MPA [9–12]. However, with the exception of a small study by Kirkland et al. [13], we have found no PAN study that strictly comply with the CHCC nomenclature; consequently, the clinical features for PAN are not completely known when applying the new definitions. The aim of the present study was to describe the symptoms, signs and clinical course of a cohort of patients who fulfil the CHCC definition for PAN.

Patients and methods Population details The study was performed at the Departments of Nephrology and Rheumatology at the Lund University Hospital, which is a tertiary hospital for a region in South Sweden with a population of 1 575 000, and also a local hospital for a district with 282 900 inhabitants. The largest community in this district is the city of Lund with around 100 000 inhabitants. There are several mediumsized communities where a large percentage of the working population commutes to and from Lund and Malmo¨ (the third largest city in Sweden). There is a substantial rural population but 81% live in communities with more than 1000 inhabitants according to Statistics Sweden (www.scb.se). Females make up 50.4% of the population. The age distribution is as follows: 0–14 yrs (18.8%); 15–34 yrs (27.1%); 35–54 yrs (27.5%); 55–74 yrs (19.1%) and 75 yrs and above (7.5%). Twelve percent of the inhabitants are born outside Sweden, most of them being immigrants from other Nordic countries or Southern Europe.

Department of Nephrology and 2Department of Rheumatology, Lund University Hospital, Lund, Sweden. Received 9 September 2005; revised version accepted 17 February 2006. Correspondence to: D. Selga, Department of Nephrology, Lund University Hospital, S-221 85 Lund, Sweden. E-mail: [email protected] 1276

ß The Author 2006. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: [email protected]

Polyarteritis nodosa The distances in this area are short; no community is more than 40 km from Lund, and the distance between Lund and Malmo¨ is only 17 km. Beside the University Hospital in Lund there is one smaller hospital in the city of Landskrona, without any specialists in rheumatology and nephrology.

1277

five factors found to predict excess mortality in that study were renal insufficiency (serum creatinine
140 mol/l), proteinuria (>1 g/day), severe gastrointestinal involvement, cardiomyopathy and CNS involvement. Glomerular filtration rate (GFR) was calculated from the Cockcroft–Gault equation [16] in adult patients and from the Schwartz equations [17, 18] in children and adolescents.

Patients All the patients with a diagnosis of PAN treated in our departments, including those living outside our local catchment area, during the period January 1990–December 2002 were eligible for this study. Every person in Sweden has a unique personal identification number and all the patients are registered in the patient administrative registry with ICD codes at every visit and admission to the hospital. The patients were identified by means of the patient administrative registry, searched for PAN and related diagnoses, using ICD 9 codes 446 and 447 for the period 1990–97, and the ICD 10 codes M30 and M31 from 1998. For the period 1997–2002, the search was extended to include six other departments in Lund and the selected departments at the four neighbouring hospitals (including Malmo¨ University Hospital). In addition, the registries at the Department of Pathology were searched using the term ‘vasculitis’. This extended search was part of an ongoing larger epidemiological study. Patients were included in the study if they had a disease compatible with systemic vasculitis with histological evidence of vasculitis affecting small or medium-sized arteries and/or with angiographic features of multiple micro-aneurysms and multiple stenoses and/or with an electromyography (EMG) typical for mononeuritis multiplex. A documented ANCA-test, determined by indirect immunofluorescence (IIF) assay and/or enzyme-linked immunosorbent assay (ELISA) to proteinase-3 (PR3)-ANCA or myeloperoxidase (MPO)-ANCA, either at diagnosis or relapse was also required. ANCA is strongly associated with small vessel vasculitis and is usually absent in reports of PAN. Although a positive ANCA does not exclude a diagnosis of PAN, according to the CHCC definitions we consider ANCA to be a marker of small vessel disease and, therefore, to exclude a diagnosis of PAN. Patients were excluded from the study if there were histological or clinical signs of small vessel vasculitis such as pulmonary capillaritis, with lung bleeding or glomerulonephritis, as confirmed by histology or indicated by cellular casts in the urinary sediment or positive PR3-ANCA or MPO-ANCA. Patients were not included if the symptoms and signs could be attributed to another disease, e.g. cancer, infection, other forms of vasculitis including hepatitis-associated vasculitis. Fulfillment of the ACR criteria established by ACR [5] was not considered mandatory for this study.

Data collected at the time of diagnosis The patient case records were reviewed in detail. From the time of diagnosis the following data were retrieved: age, gender, the first clinical signs attributable to vasculitis, clinical signs at the time of diagnosis, the time period to diagnosis (defined as the time from the first signs attributable to vasculitis until the date vasculitis was diagnosed by angiography or biopsy), serum creatinine, weight, length (for children), urine albumin, urinary sediment findings and blood pressure at the time of diagnosis, serology for hepatitis B and C, disease activity at diagnosis as assessed by the Birmingham Vasculitis Activity Score (BVAS) and the five-factor score (FFS). The BVAS was created by a group in Birmingham in 1994 [14] to measure the disease activity in patients with a variety of systemic vasculitides. It scores abnormality ascribable to the presence of active vasculitis. The maximum possible score is 63. The FFS was created by Guillevin et al. [15] in 1996 by evaluating different parameters for predicting the outcome in patients with PAN. The

Data collected during follow-up The patients were followed until December 2003. During follow-up, the following data were recorded: treatment, time and cause of death, time of end-stage renal disease (ESRD), time of and symptoms at relapse and Vasculitis Damage Index (VDI) at 1 and 5 yrs. VDI was created in 1997 [19] to aid in the separation of damage from disease activity in systemic vasculitis. The VDI scores damage due to non-healing scars from any cause that has occurred since the onset of vasculitis. It is a cumulative assessment of organ dysfunction, damage or scarring and either remains stable or increases with time. It includes assessment of ten organ systems: musculoskeletal; skin; ear, nose and throat; pulmonary; cardiovascular; renal; gastrointestinal; peripheral vascular; ocular and neuropsychiatric; and also other damages like premature gonadal failure, marrow failure, diabetes mellitus, chronic chemical cystitis, malignancy and other features considered to be an important scar or consequence that has arisen since the onset of disease. Sixty-four items of damage are scored. Relapse was defined as the occurrence of one of the following: (i) general symptoms, i.e. malaise, fever, weight loss, myalgia, arthralgia, in conjunction with a rise in CRP, when infection and other causes were ruled out; (ii) gastrointestinal or renal infarction or bleeding from aneurysms, as seen at surgery or strongly suspected by angiography; (iii) new mononeuritis multiplex; (iv) a rapid rise in serum creatinine levels and blood pressure; (v) testicular pain; (vi) skin vasculitis; and (vii) uveitis, when (i–vii) were considered to be due to vasculitis. For the comparison of relapses, 62 consecutive patients with MPA with biopsy-proven renal involvement from a previous study [20] were analysed regarding time until the first relapse.

Ethics The study was approved by the local Ethical Committee at the Faculty of Medicine, Lund University.

Statistics The comparisons between BVAS and VDI were carried out with the Spearman rank correlation test using GraphPad InStat version 3.

Results Fifteen patients with a probable diagnosis of PAN were identified in this study. Five patients all diagnosed before 1983, were excluded because possible co-existence of small vessel vasculitis could not be ruled out. Median age at diagnosis for the excluded patients was 53 yrs (range 28–61). Renal involvement was seen in three of these patients and three had skin involvement at diagnosis. Two patients had nerve involvement and lung involvement was seen in two. Median follow-up time was 14 yrs (range 14–45). One patient had a relapse 10 yrs after the start of treatment. Ten patients, five males and five females, fulfilled all the criteria listed above and were included in the study. Four patients were diagnosed by biopsy (testis, muscle, epididymis and skin), four by angiography and one was positive in both procedures. One patient

D. Selga et al.

1278 TABLE 1. Selected clinical features of 10 patients with PAN Patient Gender Age (yr) Diagnosis delay (months) Serum creatinine (mol/l)/ GFR (ml/min) Proteinuria (0/þ/þþ/þþþ) Blood pressure (mmHg) Organ involvement

I

II

III

IV

V

VI

VII

VIII

IX

X

Male 8 39 45/131

Male 16 4 45/257

Male 35 6 73/134

Male 38 0.5 144/60

Female 45 1 292/18

Male 48 10 66/120

Female 53 4 65/84

Female 70 27 69/58

Female 72 4 57/67

Female 78 6 58/73

0 110/80 Testes

0 100/45 Gastrointestinal

þ 170/120 Peripheral nerve

þ 200/110 Kidney

þþ 260/110 Kidney

0 190/95 Peripheral nerve

0 110/70 Skin

0 160/90 Peripheral nerve

0 140/70 Kidney

þþ 120/75 Peripheral nerve

Gastrointestinal

Central nervous system Pancreas

Eye Skin Kidney Biopsy BVAS FFS

Spleen Testis 11 0

Kidney

Ileum 12 1

Eye Ear Kidney Muscle 28 0

– 16 2

had typical mononeuritis multiplex confirmed by EMG and nerve biopsy was compatible, but not diagnostic for vasculitis in medium-sized vessels. All diagnostic biopsies showed inflammation and necrosis in small or medium-sized arteries, but not in arterioles and other small-sized vessels. No patient presented multiple positive biopsies. All ten patients fulfilled at least three ACR criteria, median 3 (range 3–5). All patients exhibited multiple negative ANCA tests. One patient was diagnosed at the Paediatric Clinic before ANCA testing was available. He was ANCAnegative at relapse and in all subsequent ANCA tests, IIF as well as ELISA. All the other patients were ANCA-tested at the time of diagnosis. Only one patient was found to have had hepatitis (B and C) that could have contributed to the development of PAN. He was tested positive for HBV 13 yrs before he showed any symptoms attributable to vasculitis. At the time of diagnosis he was positive for HBV antibodies, but negative for HBV antigen. We do not consider PAN in this patient to be HBV induced. At the time when HBV was diagnosed, hepatitis C virus (HCV) was not known and hence he was not tested for this. At the time of diagnosis of PAN he also had antibodies to HCV. During the 13 yrs period, six new cases were diagnosed among patients living in our local catchment area of 282 900 inhabitants. This gives an annual incidence of 1.6 per million. One patient was diagnosed at the Paediatric Clinic before the 13 yrs period and three were referred from areas outside the local catchment area. All excluded patients were diagnosed before this 13 yrs period and would not have influenced the incidence estimate if included. The extended search in registries from other departments and neighbouring hospitals did only yield one patient with possible PAN living in the local study area. That patient was diagnosed in 1983 and ended up in the group excluded because of insufficient data (see aforesaid). Due to the low yield of this search, we did not attempt to extend it further back in time.

Clinical manifestations at diagnosis The parameters assessed at the time of diagnosis are detailed in Table 1. The median age was 46 yrs (range 8–78), and the first symptoms attributable to vasculitis were dominated by myalgia, malaise, headache and fever. The time from the first symptoms attributable to vasculitis until diagnosis (diagnosis delay) varied from 2 weeks to 39 months (median five months). All the patients had systemic symptoms. Renal involvement (proteinuria, haematuria and/or renal microaneurysms in angiography) was seen in seven (70%) patients, but only two had elevated serum creatinine levels at the time of diagnosis. Five patients had renal aneurysms

– 33 3

Joints Testis Muscle Skin Testis 18 0

Skin 5 0

Heart Muscle

Muscle Peripheral nerve

Kidney Muscle

– 13 0

– 23 1

– 22 0

FIG. 1. Organ system involvement at diagnosis according to Birmingham Vasculitis Activity Score (BVAS) for 10 patients with PAN. The y-axis indicates percentage of patients having enough symptoms from an organ system to get a BVAS score. G, general; Cu, cutaneous; M, mucous membranes/eyes; E, ear, nose and throat; Ch, chest; Ca, cardiovascular; A, abdominal; R, renal; N, nervous system. on angiography, and three of these also had microscopic haematuria and/or proteinuria even though renal infarcts were not noted. The other two patients with renal involvement had microscopic haematuria or low grade proteinuria. No renal biopsies were performed at the time of diagnosis since none of the patients exhibited red cell casts, dysmorphic red cells or other cell casts indicative of glomerulonephritis. One patient underwent renal biopsy during a flare with a rise in serum creatinine, but there were no signs of glomerulonephritis. The median serum creatinine at diagnosis was 65 mol/l (range 45–292). Four patients were hypertensive, including three with malignant hypertension. The vasculitic activity at diagnosis was scored according to BVAS as shown in Fig. 1 and Table 1. The median BVAS score was 17 (range 5–33). Symptoms at diagnosis were dominated by renal and nervous system involvement. The FFS for each patient is shown in Table 1. The FFS varied from 0 to 3 (median 0). Patients with the highest FFS tended to have short diagnostic delay.

Follow-up The patients were followed for 0.5–19.8 yrs (median 6.0 yrs). No patient was lost to follow-up. The parameters assessed during follow-up are detailed in Table 2. All patients received induction

Polyarteritis nodosa

1279

TABLE 2. Major events during follow-up for 10 patients (I–X) with PAN Follow-up (yr)

Initial treatment

ESRD (yr after diagnosis)

Death (yr after diagnosis)

Relapses (yr after diagnosis)

I

19.2

CS þ AZA

–

–

6.8

II

7.2

CS þ CYC

–

–

III

9.4

CS þ CYC

6.5

–

2.0 4.8 3.2

IV

13.9

CS þ CYC

8.7

–

V VI

5 4.8

CS þ CYC CS þ CYC

– –

– –

VII VIII

5 7.9

– –

– –

IX X

2.1 0.4

CS þ AZA CS þ CYC þ PE CS þ CYC CS þ CYC

– –

– 0.4

Patient

5.9 8.9 – 0.7 2.1 2.6 3.2 6.9 – –

Symptoms at relapse Fever, headache, arthralgia, myalgia, weight loss, haematuria Malaise, fever, headache, arthralgia, weight loss Gastrointestinal bleeding Mononeuritis, rise in creatinine, haematuria, proteinuria, retinal vasculitis, pericarditis, hypertension Rise in creatinine, dyspnoea, ECG changes Gastrointestinal bleeding Sensory peripheral neuropathy, mononeuritis multiplex Testicular pain, arthralgia, inflammation of colon Skin nodules, arthritis Myalgia, arthritis, headache Iritis

CS, corticosteroids; CYC, cyclophosphamide; AZA, azathioprine; PE, plasma exchange.

therapy with a combination of steroids and cyclophosphamide (8 patients) or azathioprine (2 patients). In addition to steroids and cyclophosphamide, one patient underwent treatment with plasma exchange. An elderly woman died 5 months after diagnosis without achieving remission. She had chronic heart failure, a progressive disease and repeated infections during immunosuppressive treatment. Autopsy was not performed. In this case, BVAS was 22 and FFS 0 at diagnosis. All the other patients achieved remission within 6 months. Seven patients had altogether 11 relapses. Eight of the relapses were seen in the same organs that were affected at the time of diagnosis. In addition to this, one patient had several episodes of pancreatitis that were considered to be due to previous damage and scarring. These episodes were not classified as relapses. The median time from diagnosis to first relapse was 3.2 yrs (range 0.7–6.8). Within 5 yrs 57% of the patients had experienced a relapse. The relapse tendency for these PAN patients does not seem to differ from a cohort of 62 MPA patients previously described [20], as shown in Fig. 2. During follow-up, organ damage attributable to vasculitis was assessed by the VDI at 1 and 5 yrs after diagnosis. The scores are detailed in Table 3. After 1 yr, cardiovascular and neuropsychiatric damage dominated. Only one of the nine patients that survived the first year had enough renal damage to get a renal VDI score. After 5 yrs of follow-up, again neuropsychiatric and cardiovascular damage dominated, followed by renal damage. Two patients progressed to ESRD and started haemodialysis 6.5 and 8 yrs after diagnosis. Both had malignant hypertension at the time of diagnosis and both had a relapse with rise in creatinine and hypertension. Even though the study was small, according to the Spearman rank correlations test there was a significant correlation between high BVAS at diagnosis and high VDI at follow-up, both at 1 yr (P ¼ 0.0029) (n ¼ 9) and 5 yrs (P ¼ 0.0072) (n ¼ 7).

Discussion In the present study, we report on ten patients diagnosed with PAN, who all lack direct evidence of small vessel disease such as glomerulonephritis or capillaritis and indirect evidence such as ANCA. Since the publication of the CHCC nomenclature, many studies have been published describing the microscopic form of polyarteritis [9–12]. To the best of our knowledge there is only one study before [13] the present study that describes classical PAN

FIG. 2. Cumulative sustained remission (censored for death) during 5 yrs of follow-up for 10 patients with PAN (solid line) and 62 patients with microscopic polyangiitis (MPA) (dashed line).

strictly adhering to this nomenclature. As in other studies using less restrictive inclusion criteria PAN typically developed subacutely, with the onset of constitutional symptoms over weeks to months. Only few had an acute onset, and in those cases the symptoms were severe with malignant hypertension, renal insufficiency and gastrointestinal involvement. Even though we had excluded patients with glomerulonephritis, organ involvement at diagnosis was dominated by renal involvement. In our study, seven patients (70%) had signs of renal involvement at the time of diagnosis, which is more than what the other authors have found. Fortin et al. [21] found 44% renal involvement and Agard et al. [22] found 22% in their studies. Two patients (20%) in our study developed ESRD during follow-up and both had relapses with renal involvement. This shows that patients with PAN must be followed closely to avoid deterioration of renal function. The ACR document has positivity for HBV as a classification criterion for PAN. Recent reports show that HBV-associated arteritis responds to anti-viral therapy and that relapse after such treatment seems to be rare. If the clinical course, outcome and

D. Selga et al.

1280

TABLE 3. Vasculitis Damage Index (VDI) at 1 and 5 yrs after diagnosis of PAN I 1 yr Musculoskeletal Skin/mucous membranes Ocular ENT Pulmonary Cardiovascular Peripheral vascular disease Gastrointestinal Renal Neuropsychiatric Other damage/drug reaction Total VDI score at 1 yr Total VDI score at 5 yrs

II 5 yrs

1 yr

III 5 yrs

IV

1 yr

5 yrs

1

1

1 yr

V 5 yrs

1 yr

VI 5 yrs

1 yr

VII 1 yr

VIII

5 yrs

1 yr

IX

5 yrs

1

1 1

1 1

1 yr

1 1

1

1

1

1

1

1

1 1 2

1 1 2

1 2

1 2

1

1 1

1

1 1

2 1

3 1

1 5

5 2

1 3

5

1 0

2

1 1

5 4

VDI is registered at 1 yr in 9 patients (I–IX) and for 7 patients also at 5 yrs. The score for each organ system and the total score for each patient is registered.

recent reports from other countries have also shown a relatively low incidence [26]. In comparison, MPA, which is reported to have an annual incidence of 8.4–11.6 per million [26], seems to be around five times more frequent. In conclusion, our small series find PAN to be a rare but severe disease. Renal involvement with malignant hypertension and infarctions is common, and may lead to ESRD. Our data contradict the earlier opinion that PAN is a self-limiting disease with few relapses. We encourage others to publish case series to help revealing the true clinical picture of this old disease when applying the new definition.

Key messages

Rheumatology

response to treatment differ between PAN with and without HBV association, we consider it unwise to lump these two conditions into one disease entity. For one patient in our series, we had difficulty in interpreting the role of viral infection. In this case, HBV infection could be tracked 13 yrs back and at the time of diagnosis he was positive for HBV antibodies but negative for HBV antigen. Guillevin et al. [23] have described patients with HBV-induced PAN as an acute disease, occurring shortly after infection. In their study, the median time from HBV infection to development of PAN was 2 months. We do not consider PAN in our patient to be HBV-induced, but we cannot completely rule out this possibility. Peripheral nerve involvement is a feature that may be more prevalent in HBV-associated PAN. We found this complication in five of our patients (50%). This is less than what the others have reported [22, 24]. In one report on patients with HBV-related PAN, the incidence of peripheral neuropathy was 83% [23]. Relapse is another feature that most probably is influenced by the percentage of HBV-associated cases, as such cases seem to have a lower relapse rate. In the present series, relapses were frequent; 57% of the patients had experienced a relapse within 5 yrs. We compared this with a group of MPA patients whose results we have reported previously [20]. Interestingly, the curves were very similar (Fig. 2). The low number of PAN cases does, of course, introduce a great risk of a type one error. Other groups have shown lower relapse rates [25], but in a small report from Australia with HBV-negative PAN patients, frequent relapses were also seen [13]. Different prognostic factors have been used for systemic vasculitis. BVAS at diagnosis was higher in those patients who had the highest VDI at 1 and 5 yrs and in the patient who died. BVAS is a marker of vasculitic activity, and although the acute damage can be reversible, our findings indicate that it has a prognostic potential. Guillevin et al. [15] have shown FFS to be a prognostic factor for mortality in patients with PAN and Churg–Strauss Syndrome. The small size of our study prevents firm conclusion, but it is notable that the only patient who died had a FFS of zero. It is also difficult to draw prognostic conclusions on single factors, but we note that two out of three patients with malignant hypertension at diagnosis developed ESRD. The incidence and prevalence figures vary between studies depending on the classification criteria used. From our data, we estimated an annual incidence of 1.6 per million in our part of Sweden. This is less than what many other authors have found, but it is probably due to our strict classification criteria. More

 Polyarteritis nodosa is a rare but severe disease with a high incidence of renal involvement and frequent relapses.  These patients must be followed carefully to prevent relapses and further organ damage.

Acknowledgements This study was supported by grants from the Swedish Medical Research Council (Project K-2005-73X-09487-15A) and the Swedish Renal Foundation.

References 1. Kussmaul A, Maier R. Uber eine bisher nicht beschreibene eigenthu¨mliche Arterienerkankung (Periarteriitis nodosa), die mit Morbus Brightii und rapid fortschreitender allgemeiner Muskella¨hmung einhergeht. Deutsches Arch f klin Med 1866;1:84–517. 2. Wegener F. Uber eine eigenartige rhinogene Granulomatose mit besonderer Beteiligung des Arteriensystem und der Nieren. Beitr Pathol Anat 1939:36–38. 3. Davson J, Ball J, Platt R. The kidney in periarteritis nodosa. Q J Med 1948;17:175–205. 4. Zeek PM. Periarteritis nodosa: a critical review. Am J Clin Pathol 1952;22:777–90.

Polyarteritis nodosa 5. Lightfoot RW Jr, 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. 6. Trepo C, Thivolet J. Hepatitis associated antigen and periarteritis nodosa (PAN). Vox Sang 1970;19:410–1. 7. van der Woude FJ. Anticytoplasmic antibodies in Wegener’s granulomatosis. Lancet 1985;2:48. 8. Jennette JC, Falk RJ, Andrassy K et al. Nomenclature of systemic vasculitides: the proposal of an international consensus conference. Arthritis Rheum 1994;37:187–92. 9. Guillevin L, Durand-Gasselin B, Cevallos R et al. Microscopic polyangiitis: clinical and laboratory findings in eighty-five patients. Arthritis Rheum 1999;42:421–30. 10. Lane SE, Watts RA, Shepstone L, Scott DG. Primary systemic vasculitis: clinical features and mortality. Q J Med 2005; 98:97–111. 11. Westman KW, Bygren PG, Olsson H, Ranstam J, Wieslander J. Relapse rate, renal survival, and cancer morbidity in patients with Wegener’s granulomatosis or microscopic polyangiitis with renal involvement. J Am Soc Nephrol 1998;9:842–52. 12. Hogan SL, Nachman PH, Wilkman AS, Jennette JC, Falk RJ. Prognostic markers in patients with antineutrophil cytoplasmic autoantibody-associated microscopic polyangiitis and glomerulonephritis. J Am Soc Nephrol 1996;7:23–32. 13. Kirkland GS, Savige J, Wilson D, Heale W, Sinclair RA, Hope RN. Classical polyarteritis nodosa and microscopic polyarteritis with medium vessel involvement – a comparison of the clinical and laboratory features. Clin Nephrol 1997;47:176–80. 14. Luqmani RA, Bacon PA, Moots RJ et al. Birmingham Vasculitis Activity Score (BVAS) in systemic necrotizing vasculitis. Qjm 1994;87:671–8. 15. 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. 16. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976;16:31–41.

1281

17. Schwartz GJ, Haycock GB, Edelmann CM, Spitzer A. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Pediatrics 1976;58:259–63. 18. Schwartz GJ, Gauthier B. A simple estimate of glomerular filtration rate in adolescent boys. J Pediatr 1985;106:522–6. 19. Exley AR, Bacon PA, Luqmani RA et al. Development and initial validation of the Vasculitis Damage Index for the standardized clinical assessment of damage in the systemic vasculitides. Arthritis Rheum 1997;40:371–80. 20. Westman KW, Selga D, Isberg PE, Bladstrom A, Olsson H. High proteinase 3-anti-neutrophil cytoplasmic antibody (ANCA) level measured by the capture enzyme-linked immunosorbent assay method is associated with decreased patient survival in ANCAassociated vasculitis with renal involvement. J Am Soc Nephrol 2003;14:2926–33. 21. Fortin PR, Larson MG, Watters AK, Yeadon CA, Choquette D, Esdaile JM. Prognostic factors in systemic necrotizing vasculitis of the polyarteritis nodosa group – a review of 45 cases. J Rheumatol 1995;22:78–84. 22. Agard C, Mouthon L, Mahr A, Guillevin L. Microscopic polyangiitis and polyarteritis nodosa: how and when do they start? Arthritis Rheum 2003;49:709–15. 23. Guillevin L, Lhote F, Cohen P et al. Polyarteritis nodosa related to hepatitis B virus. A prospective study with long-term observation of 41 patients. Medicine 1995;74:238–53. 24. Guillevin L, Le Thi Huong D, Godeau P, Jais P, Wechsler B. Clinical findings and prognosis of polyarteritis nodosa and Churg–Strauss angiitis: a study in 165 patients. Br J Rheumatol 1988;27:258–64. 25. Gayraud M, Guillevin L, le Toumelin P et al. Long-term follow-up of polyarteritis nodosa, microscopic polyangiitis, and Churg–Strauss syndrome: analysis of four prospective trials including 278 patients. Arthritis Rheum 2001;44:666–75. 26. Watts RA, Gonzalez-Gay MA, Lane SE, Garcia-Porrua C, Bentham G, Scott DG. Geoepidemiology of systemic vasculitis: comparison of the incidence in two regions of Europe. Ann Rheum Dis 2001;60:170–2.