554

CONCISE REPORT

Vasculitis restricted to the lower limbs: a clinical and

histopathological study
M Khellaf, M Hamidou, C Pagnoux, M Michel, J M Brisseau, X Chevallier, P Cohen, L Guillevin,
B Godeau
...................................................................................................................................
Ann Rheum Dis 2007;66:554556. doi: 10.1136/ard.2006.065110

Background: Rare cases of vasculitis restricted to the lower

limbs have been reported, but the characteristics, outcome and
response to treatment of this entity are not well known.
Objective: To describe the clinical, complementary examinations and response to treatment of this rare entity in the first
retrospective series, and to compare data with historical pooled
cases.
Methods: Retrospective analysis of all biopsy-proven cases
observed over a 10-year period in four French tertiary medical
units. Diagnosis of vasculitis restricted to the lower limb
required the absence of any clinical symptom and complementary test finding, suggesting major extramuscular visceral
involvement.
Results: 11 patients were included. Vasculitis restricted to the
lower limb was associated with disabling muscle pain of the
calves. Fever was present in 50% of cases; ankle arthralgia in
50% and skin involvement in 40%. MRI was the cornerstone of
the diagnosis, showing hyperintense signal in T2 weight and in
T1 weight after gadolinium injection. MRI findings correlated
well with clinical outcome and were useful in guiding biopsy.
Muscle biopsy was consistent with a polyarteritis nodosa-type
vasculitis in only 40% cases, whereas a leucocytoclastic
vasculitis was seen for all other cases. Treatment with
corticosteroids was effective in all cases, but there were
relapses requiring immunosuppressive agents in 54% of cases.
Conclusion: Vasculitis of the calf muscles must be considered
for patients with calf pain and with a biological inflammatory
syndrome.

ost cases of systemic vasculitis include multiorgan

involvement and systemic manifestations; occasionally,
only one organ is affected.1 Rare cases of polyarteritis
nodosa (PAN) vasculitis mostly affecting the skeletal leg
muscles have been reported. We report the first series of 11
cases of vasculitis restricted to the lower limbs and compare
their characteristics and outcome with the few cases reported
previously.216

PATIENTS AND METHODS

Data for all patients, with a histologically proven vasculitis
restricted to lower leg muscles followed in four French tertiary
referral medical centres over a 10-year period, were retrospectively analysed. Only cases with a histological proof of
muscle vasculitis were included. The diagnostic criteria for
restricted muscle vasculitis were (1) proven histological
vasculitis on muscle biopsy and (2) absence of clinical
symptoms and complementary test findings suggesting major
extramuscular visceral involvement (ie, heart, central nervous
system, mononeuropathy multiplex, kidney, gastrointestinal, or
upper and lower respiratory tract manifestations).
[Link]

We looked for similar cases in the literature using the

PubMedline Database on the web and the following keywords:
localised polyarteritis nodosa, localised vasculitis, calf
muscle vasculitis or limb-restricted vasculitis.

RESULTS
Clinical characteristics at onset
Eleven patients fulfilled the inclusion criteria. In all patients
except for one, pain was confined to calf muscles and was the
presenting symptom (table 1). In one patient, calf-muscle pain
was associated with quadriceps myalgia. Pain was initially
moderate and worsened during physical exercise in nine
patients. Onset of symptoms was sudden in two patients, with
pain causing rapid impotence of the limbs. An induration of the
painful muscles of between 2 cm and involvement of the whole
calf was present in 8 of 11 patients. Four patients had
cutaneous lesions in the affected calf areas, the lesions being
vascular purpura, erythema around a muscular nodule, superficial dermitis and/or pseudo erythema nodosa of the calf.
Seven patients presented with prolonged fever, three having
temperature peaks .39C (102F). Significant weight loss (2
9 kg) was observed in 5 of 11 patients at presentation.
Arthralgia without arthritis limited to the ankles affected half
of the cases. No patient had major organ involvement,
consistent with the inclusion criteria. The median time between
the onset of symptoms and the final diagnosis was 14 months
(range 235).
Complementary examinations
A biological inflammatory pattern was present in all patients
(table 1). Creatine kinase, aldolase, liver function markers and
serum creatinine were in the normal range in all cases. None of
the patients had severe kidney involvement.
Tests for antinuclear antibodies, anti-neutrophil cytoplasmic
antibodies and cryoglobulinaemia were negative in all but one
patient who had perinuclear anti-neutrophil cytoplasmic
antibodies by indirect immunofluorescence test without antimyeloperoxidase or anti-proteinase 3 specificity by ELISA.
Antibody tests for HIV were negative, and no patient had active
hepatitis B virus or hepatitis C virus infections.
An electromyogram of the legs was performed in four
patients and was found to be normal.
Bone x ray of the lower limbs was normal in all patients
except for one, with new periosteal bone formation bordering
the internal part of the peroneal diaphysis. Technetium
scintigraphy was performed in two patients and showed
localised uptake with new periosteal bone formation.
Magnetic resonance imaging
MRI of the calves was performed in all patients, and
showed normal unenhanced T1-weighted sequences; however,
Abbreviation: PAN, polyarteritis nodosa

Vasculitis of the lower limbs

hyperintense T2-weighted signals in the involved areas

indicated increased muscle fluid content. Gadolinium in T1weighted sequences showed an enhanced signal. Six patients
had several muscle groups involved and nine patients had
bilateral lesions. The main muscle affected was the gastrocnemius muscle (n = 8; fig 1). Other affected muscles were
quadriceps (n = 1), tibialis anterior (n = 2) and peroneus
longus (n = 2). MRI features correlated well with clinical
evolution; subsequent MRI showed disappearance of the
abnormal signal from six patients who initially responded well
to treatment. A new MRI was performed in eight patients who
relapsed, and showed in all cases a recurrence of the
hyperintense signal in the painful area.
Muscle biopsy
A calf muscle biopsy was performed in the abnormal area
localised with MRI in all patients. Of the 11 patients, four had a
surgical biopsy because of the depth of the muscle lesions, and
biopsy was non-surgical in seven patients. All biopsy specimens
were analysed by a trained anatomopathologist of the French
Vasculitis Study Group. Histological analysis revealed leucocytoclastic vasculitis in seven patients and a non-granulomatous necrotising vasculitis in four.
Treatment and outcome
One patient did not receive any treatment because of
spontaneous improvement, followed by a transient relapse
after a 30-month follow-up.

Figure 1 (A) MRI of the limb sequence T1 with gadolinium. (B)

Enhancement of the internal part of the gastronecmius.

555

One patient was treated with 2 mg/day colchicine alone with

a rapid improvement. This patient was, however, dependent on
the treatment, which is still continuing at 32 months of followup.
Prednisone at a median dose of 1 mg/kg body weight per day
(range 0.51) was used as a preferred treatment for nine
patients, leading to substantial improvement in all patients.
However, 8/9 (90%) patients suffered a relapse after a median
of 11 months (range 334) when the median prednisone dose
was 8 mg/day (range 411). The severity of the relapse and/or
the dependence on corticosteroids led to administration of one
or several immunosuppressive agents to six patients, associated
in one patient with plasma exchange (methotrexate, n = 4;
azathioprine, n = 2; cyclophosphamide, n = 1). In one patient,
the relapse was successfully managed by a transient increase of
the daily dose of steroids. One patient received 100 mg dapsone
daily after recurrence with no success and then 400 mg
hydroxychloroquine daily with a better outcome.
None of the 11 patients developed systemic extramuscular
vasculitis during follow-up (median 48 months; range 16120).
At the time of writing this report, four patients were receiving
no treatment, one patient was receiving colchicine, four
patients a low dose of prednisone (315 mg/day) and two
patients prednisone associated with immunosuppressive agents
(azathioprine or methotrexate).

DISCUSSION
Our series confirms the clinical characteristics of vasculitis
restricted to the lower limb: disabling muscle pain, frequently
associated with fever, alteration of general status and inflammatory syndrome. The most frequently involved muscles are
those in the calf, but other lower limb areas, particularly the
quadriceps, may be involved. The nosology of this syndrome is
unclear. Some authors postulate that such localised forms of
vasculitis are cases of systemic PAN diagnosed early and
treated.6 However, the absence of extra-musculoskeletal major
organ involvement in our series even after prolonged follow-up
argues against this hypothesis.
Like other authors,3 5 we highlight the importance of MRI for
diagnosis. In all our patients, MRI revealed the affected area,
confirming the inflammatory process in the muscles and
providing useful guidance for the pathologist or the surgeon
for the muscular biopsy. The very long delay before diagnosis in
our series confirms the poor awareness of this kind of vasculitis
among doctors. MRI should be performed earlier in the course
of the disease in patients with persistent and unexplained calf
pain associated with biological markers of inflammation.
Radiological abnormalities correlated well with the clinical
outcome, so serial MRI might be helpful for monitoring the
course of the disease and response to treatment.
The histology of the previously reported cases of vasculitis
confined to calf muscles was always typical of a medium-vessel
vasculitis consistent with a PAN. By contrast, leucocytoclastic
vasculitis was the most frequent finding on muscular biopsy in
our series, and PAN medium-vessel vasculitis was found in only
40% of cases. The clinical and biological manifestations,
outcomes and responses to treatment did not differ between
these two histological presentations.
Eighteen cases216 of lower limb-restricted vasculitis have
been reported in the literature: corticosteroids seem to be very
effective, and the course and prognosis of the disease were
considered benign. Sensitivity to steroids was confirmed in our
series, and all patients initially responded to steroid treatment.
However, 90% of patients relapsed when the treatment was
tapered and in .50% of patients, immunosuppressive agents
were needed for a better control and as a steroid-sparing
strategy. So the reputation of benign disease attributed to the
[Link]

556

Khellaf, Hamidou, Pagnoux, et al

Table 1 Clinical, biological and histological features of the

11 patients compared with the literature
Clinical, biological and
histological features

New cohort

Pooled cases of the

literature

Median (range) age (years)

Sex ratio (F/M)
Median (range) delay to
diagnosis (months)
Calf pain
Bilateral pain
Calf tumefaction
Skin abnormalities associated
Fever
Weight loss
Median (range) ESR (mm/h)

45 (2572)
1/1.2
14 (235)

48 (2390)
1.3/1
2

100%
63%
72%
36%
63%
45%
46 (23134)

100%
80%
72%
ND
62%
nd
74 (24103)

Median (range) C reactive

protein (mg/l)
Muscular biopsy
Leucocytoclastic vasculitis
PAN histology

35 (9125)

ND

64%
36%

0%
100%

ESR, erythrocyte sedimentation rate; ND, no data; PAN, polyarteritis

nodosa.

localised vasculitis of the calf should be reconsidered, and

patients must be informed of the risk of relapse despite initially
successful corticotherapy. In contrast with the high risk of
relapse, inflammatory phenomena in the vessels remained
localised in muscles in our series, and evolution towards severe
multiorgan vasculitis seems unlikely.
To conclude, doctors should consider the diagnosis of
localised vasculitis in middle-aged patients having persistent
and unexplained calf pain associated with fever and/or altered
general status and abnormal markers of inflammation. MRI of
the legs is best for diagnosis and should be rapidly performed
whenever the cause of persistent pain is not obvious. It shows
features of the localised inflammatory process and can guide
muscle biopsy.
.......................

Authors affiliations

M Khellaf, M Michel, B Godeau, Service de Medecine Interne, Ho

pital
Henri Mondor, Universite Paris XII, Assistance Publique Ho
pitaux de Paris,
Creteil Cedex, France
M Hamidou, J M Brisseau, Service de Medecine Interne, Ho
pital HotelDieu, Universite de Nantes, Nantes Cedex 1, France

[Link]

C Pagnoux, P Cohen, L Guillevin, French Vasculitis Study Group, Ho

pital
Cochin, Reference Center for Vasculitis, Assistance Publique Ho
pitaux de
Paris, Universite Paris Descartes, Paris, France
X Chevallier, Service de Rhumatologie, Ho
pital Henri Mondor, Universite
Paris XII, Assistance Publique Ho
pitaux de Paris, Creteil, Cedex, France
Competing interests: None.
Correspondence to: Dr M Khellaf, Service de Medecine Interne, CHU
Henri Mondor, 51 Avenue de Lattre de Tassigny, 94010 Creteil Cedex,
France; [Link]@[Link]
Accepted 17 October 2006
Published Online First 25 October 2006

REFERENCES
1 Kariv R, Sidi Y, Gur H. Systemic vasculitis presenting as a tumorlike lesion. Four
case reports and an analysis of 79 reported cases. Medicine (Baltimore)
2000;79:34959.
2 Carron P, Hoffman IE, De Rycke L, Peene I, Veys E M, De Keyser F, et al. Case
number 34: relapse of polyarteritis nodosa presenting as isolated and localised
lower limb periostitis. Ann Rheum Dis 2005;64:111819.
3 Eckel CG, Sibbitt RR, Sibbitt WL Jr, Newell JD Jr, Narva A Jr. A possible role for
MRI in polyarteritis nodosa: the creeping fat sign. Magn Reson Imaging
1988;6:71315.
4 Ferreiro JE, Saldana MJ, Azevedo SJ. Polyarteritis manifesting as calf myositis
and fever. Am J Med 1986;80:31215.
5 Gallien S, Mahr A, Rety F, Kambachna M, Lhota F, Cohen P, et al. Magnetic
resonance imaging of skeletal muscle involvement in limb restricted vasculitis.
Ann Rheum Dis 2002;61:11079.
6 Garcia F, Pedrol E, Casademont J, Mellado B, Cordoba R, Cid M. Polyarteritis
nodosa confined to calf muscles. J Rheumatol 1992;19:3035.
7 Garcia-Porrua C, Mate A, Duran-Marino JL, Fernandez-Martinez C, GonzalezGay MA. Localized vasculitis in the calf mimicking deep venous thrombosis.
Rheumatology (Oxford) 2002;41:9445.
8 Gardner GC, Lawrence MK. Polyarteritis nodosa confined to calf muscles.
J Rheumatol 1993;20:9089.
9 Golding DN. Polyarteritis presenting with leg pains. Br Med J 1970;1:2778.
10 Hall C, Mongey AB. Unusual presentation of polyarteritis nodosa. J Rheumatol
2001;28:8713.
11 Hofman DM, Lems WF, Witkamp TD, Putte VD, Bijlsma JW. Demonstration of
calf abnormalities by magnetic resonance imaging in polyarteritis nodosa. Clin
Rheumatol 1992;11:4024.
12 Kamimura T, Hatakeyama M, Torigoe K, Nara H, Kaneko N, Satou H. Muscular
polyarteritis nodosa as a cause of fever of undetermined origin: a case report
and review of the literature. Rheumatol Int 2005;25:3947.
13 Laitinen O, Haltia M, Lahdevirta J. Polyarteritis confined to lower extremities.
Scand J Rheumatol 1982;11:714.
14 Nakamura T, Tomoda K, Yamamura Y, Tsukano M, Honda I, Iyama K.
Polyarteritis nodosa limited to calf muscles: a case report and review of the
literature. Clin Rheumatol 2003;22:14953.
15 Nash P, Fryer J, Webb J. Vasculitis presenting as chronic unilateral painful leg
swelling. J Rheumatol 1988;15:10225.
16 Soubrier M, Bangil M, Franc S, Dubost JJ, Ristori JM, Kemeny JL, et al.
Vasculitis confined to the calves. Report of a case. Rev Rhum Engl Ed
1997;64:41416.