Small fiber neuropathy and vasculitis.

ARTHRITIS & RHEUMATISM
Vol 40, No. 6, June 1097, pp 1173-1 177
0 1997, Amcrican College of Rheumatology
1173
SMALL FIBER NEUROPATHY AND VASCULITIS
DAVID LACOMIS, MICHAEL J. GIULIANI, VIRGINIA STEEN, and HENRY C. POWELL
Vasculitis-associated neuropathy usually presents as multiple mononeuropathies or sensorimotor
polyneuropathies that affect large nerve fibers; painful
small fiber sensory neuropathy has not previously been
described in association with vasculitis. This report
describes 2 patients with small fiber neuropathy in
whom vasculitis was found to be present. Patients with
small fiber neuropathy for which no other cause has
been identified should be evaluated for the presence of
vasculitis.
The diagnosis of small fiber neuropathy (SFN) is
made based on the presence of distal dysesthesias,
frequently with loss of pain and temperature sensation in
the setting of normal tendon reflexes and strength,
relative preservation of large fiber sensory modalities,
and normal findings on conventional nerve conduction
studies (which evaluate only large myelinated axons) (1).
Recent advances in electrodiagnostic techniques (in
quantitative sensory and sweat testing) facilitate objective confirmation of the clinical diagnosis of SFN (2), but
identification of the root cause is not possible in the
majority of patients (1).A careful search for inherited as
well as toxic and metabolic conditions such as diabetes
mellitus, alcoholism, amyloidosis, and cancer (2) is often
uninformative, and these patients are often thought to
have an idiopathic neuropathy that is treated
symptomatically.
Another cause of SFN may be disordered autoimmunity. Gorson and Ropper (3) described 20 elderly
patients with idiopathic SFN, noting that 3 seemed to
respond to intravenous gamma globulin treatment.
Suarez et a1 (4) postulated an immune-mediated mechDavid Lacomis, MD, Michael J. Giuliani, MD, Virginia Steen,
MD: University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; Henry C. Powell, MD, DSc: Veterans Administration Research and University of California, San Diego.
Address reprint requests to David Lacomis, MD, Neurology
Department, University of Pittsburgh School of Medicine, Room
F878, Presbyterian University Hospital, Pittsburgh, PA 15213.
Submitted for publication September 26, 1996; accepted in
revised form January 7, 1997.
anism as a cause of another small fiber disturbance,
idiopathic autonomic neuropathy, noting an antecedent
viral illness in many patients and perivascular mononuclear cell infiltration in a sural nerve specimen from 1.
A potentially treatable autoimmune pathogenesis
is known to play a role in the vasculitides, but these
disorders have not been associated with SFN. During the
last 2 years, we have identified 2 patients with SFN
associated with vasculitis, and herein we report their
clinical, neurophysiologic, and pathologic features.
DIAGNOSTIC METHODS
Nerve conduction studies were performed on the
sural, superficial peroneal, radial, median, and ulnar
nerves. Motor responses and F-waves were recorded
from the tibia], peroneal, median, and ulnar nerves.
Routine concentric needle electromyography (EMG)
included at least a foot and a distal leg muscle. Quantitative sensory testing was performed utilizing the forcedchoice method (patient 1) and the 4-2-1 stepping algorithm (patient 2) on a microcomputer system (Case IV;
W.R. Electronics, Stillwater, MN) (5). Quantitative
sudomotor axon reflex testing (QSART) was performed
according to the method of Low et a1 (6).
Five micron-thick formalin-fixed sural nerve sections were stained with hematoxylin and eosin,
trichrome, and Congo red, and with leukocyte common
antigen by immunohistochemical methods. Sections
were also processed for plastic embedding, ultrastructural study, and nerve fiber teasing by routine methods.
Morphometric analysis of lp sections (patient 1) was
performed via computerized image analysis (NIH Imaging Program 1.43, part no. PB93-504868; available from
NTIS, Springfield, VA) measuring axon diameters of
myelinated fibers. Morphometric analysis was not performed on the biopsy specimen from patient 2, due to
the obvious loss of both large and small myelinated
fibers and because the interfascicular variability in axon
loss made representative sampling problematic.
LACOMIS ET AL
1174
CASE REPORTS
Patient 1. Clinical and laboratoryfindings. Patient 1, a
71-year-old woman with bipolar disorder treated with phenelzine, developed slowly worsening painful electrical sensations
that began in the soles of both feet and, after 8 months,
ascended to the calves. There was minimal intermittent cramping and numbness in the toes. No weakness, imbalance, lack of
coordination, lightheadedness, dry mouth, dry eyes, or bowel
or bladder dysfunction was noted. There were no antecedent
infections or other illnesses. The patient was concerned that
her symptoms were psychological in origin.
There were no recent toxic exposures or human immunodeficiency virus risk factors, nor was there a family history of
neuropathy. The patient drank -325 ml of wine daily for a
duration of 2 years until 2 years prior to the onset of
paresthesias. She had had no dysesthesias prior to the development of pain in her feet.
The general examination findings at the time of presentation to us, 8 months after the onset of symptoms in the
patient's feet, were unremarkable. Cranial nerves, muscle
strength and tone, tendon reflexes including ankle jerks,
proprioception, vibration, coordination, and gait were normal.
There was a mild decrease in pinprick and cold sensation
distally in both feet.
The following laboratory tests yielded negative or
normal results: Westergren erythrocyte sedimentation rate
(ESR), complete blood cell count, serum protein studies and
immunoelectrophoresis, Lyme serology, rapid plasma reagin
(RPR) testing, thyroid function studies, hepatitis B and C
antibody screens, measurement of glucose hemoglobin A,C
(HbA,C) electrolytes, alanine aminotransferase, aspartate
aminotransferase, alkaline phosphatase, lipids, gamma glutamyl transpeptidase, blood urea nitrogen, creatinine, calcium,
phosphorus, anti-Hu antibody, sulfatide antibody, anti-myelinassociated glycoprotein, vitamin B,,, antinuclear antibody
(ANA), rheumatoid factor, Ro/SS-A and La/SS-B antibodies,
antineutrophilic cytoplasmic autoantibodies, and cryoglobulins, and chest radiography, and mammography. Phenelzine was stopped, and trials of topical capsaicin and lidocaine,
oral amitriptyline, phenytoin, baclofen, mexiletine, and carbamazepine were undertaken, but without alleviation of the
pain.
Results of electrodiagnostic testing. Nerve conduction
and EMG results were normal, and included sural sensory
amplitudes of 11 pV (right) and 9 pV (left) (normal >3).
F-waves and needle examination results were also normal.
Quantitative sensory testing (QST) revealed an inability to
sense temperature changes (from 10" to 48°C) in the foot.
Vibratory thresholds were normal in the foot (eighty-eighth
percentile; normal below ninety-fifth percentile) and only
mildly elevated in the hand (ninety-seventh percentile; normal
below ninety-fifth percentile). Quantitative sudomotor axon
reflex testing (QSART) revealed an absent sweat response in
the foot (normal 0.16-3.03 pl) (7) and normal proximal leg and
forearm responses.
Results of pathologic studies. Incisional biopsies of the
left sural nerve and medial gastrocnemius muscles were performed. Paraffin-embedded sections of the sural nerve specimen revealed multiple small to moderately large foci of
perivascular epineurial lymphocytic infiltrates, mainly involving venules and small arterioles. Small cuffs of lymphocytes
Figure 1. Pathologic appearance of a sural nerve specimen from
patient 1. A, A small epineurial blood vessel is surrounded by
mononuclear inflammatory cells (paraffin embedded; hematoxylin and
eosin stained). B, A deeper level of the same vessel reveals nuclear
debris (arrows) associated with degeneration of the wall (paraffin
embedded; hematoxylin and eosin stained). C, Darkly staining lymphocytes (arrowheads) are present within the wall of a small arteriole.
A pale endothelial cell nucleus is identified for comparison (arrow)
(leukocyte common antigen stained). I), A plastic-embedded sural
nerve section reveals normal density of large myelinated axons, with no
acute axonal degeneration (toluidine blue stained). Bars = 25 pm.
surrounded a few endoneurial blood vessels. One epineurial
vessel exhibited degeneration of the vascular wall with infiltration by lymphocytes and rare polymorphonuclear leukocytes
(Figures 1A and B). Several other vessels, including small
arterioles (30-50 pm in diameter), exhibited transmural invasion by lymphocytes (Figure 1C). Larger arteries were normal,
and there was no vascular occlusion. Plastic-embedded sections revealed a grossly normal population of myelinated fibers
(Figure 1D). Morphometric analysis revealed a mild increase
in small myelinated axons and no significant dropout in large
axons (compared with a published age-matched normal control) (8). Only 1 of 100 teased myelinated nerve fibers contained signs of axonal degeneration. Electron microscopy
(EM) revealed many empty stacks of Schwann cell processes
and collagen pockets associated with nonmyelinated axons
(Figure 2). The distinctive profiles of regenerating small and
medium-sized myelinated axon clusters were observed. The
muscle specimen was unremarkable except for mild perivascular mononuclear cell inflammation without vasculitis.
Clinical course. Prednisone (40-50 mgiday) was administered orally for 12 weeks for nonsystemic vasculitis and then
was slowly tapered, without subjective improvement. Dysesthesias slowly worsened and ascended to the thighs over 2% years.
Nerve conductions remained normal. Due to debilitating pain,
intravenous cyclophosphamide treatment was started. Unfortunately, shortly after starting cyclophosphamide, the patient
committed suicide. An autopsy was not performed.
Patient 2. Clinical and laboratory findings. Patient 2, a
48-year-old woman, reported distal burning paresthesias that
had begun in the feet several years prior to our evaluation and
then progressed to involve the hands. She denied weakness or
gait imbalance, but she felt that her sweating was diminished.
There were no disturbances of bowel or bladder. She noted
infrequent lightheadedness.
SFN AND VASCULITIS
1175
Figure 2. Electron microscopy of a sural nerve specimen from patient 1. There are stacks of Schwann cell
processes (several are identified by arrows) lacking unmyelinated axons, and many collagen pockets are present
(see arrowheads for examples), consistent with a loss of unmyelinated axons. Bar = 1 pm.
The patient reported a 3-year history of polyarthralgias, Raynaud’s phenomenon, dry eyes, fatigue, thinning of
hair, and intermittent oral but no genital ulcerations. There
were several remote episodes of possible pleuritis. There was
no exposure to alcohol or toxins and no family history of
peripheral neuropathy.
One year prior to presentation, ANA had been found
to be positive at a titer of 1540 (normal <1:40), in a nucleolar
pattern. At the time of presentation, the ESR was 25 mmihour
(normal <20). Anti-double-stranded DNA, anti-Sm, antiRNP, anti-RoiSS-A, anti-LaiSS-B, and antimitochondrial antibodies were absent. Results of the following studies were
negative or normal: hepatitis B and C profiles, serum protein
electrophoresis, thyroid function studies, RPR, prothrombin
time, partial thromboplastin time, and measurement of complement (including C3 and C4), liver enzymes, HbA,C, cryoglobulins, cholesterol, triglycerides, and antisulfatide and antineutrophilic cytoplasmic antibodies.
The general examination results were normal except
for obesity. Schirmer’s test revealed normal tear production.
Neurologic examination revealed normal cranial nerves,
strength, tendon reflexes, and gait. Pinprick and temperature
sensations were reduced in the feet, and vibration was mildly
reduced at the great toes only.
Results of electrodiagnostic testing. Unilateral nerve
conduction studies and EMG findings were normal and included a right s u r d sensory amplitude of 12 pV. Quantitative
sensory testing revealed abnormally elevated thresholds to
thermal and vibratory stimuli in the hand and foot. Quantita-
tive sudomotor axon reflex testing revealed persistent (“hungup”) responses diffusely, with excessive sweat volumes in the
forearm and proximal leg.
Results of pathologic studies. A left sural nerve biopsy
was performed. Many small and medium-sized epineurial
blood vessels (50-80 p m in diameter) from the sural nerve
specimen were surrounded by mononuclear inflammatory
cells. Several exhibited transmural invasion by inflammatory
cells, along with destruction of the vessel walls (Figures 3A and
B). There was variation in dropout in myelinated fibers from
fascicle to fascicle. In the plastic-embedded sections, occasional degenerating axon profiles were noted. Five percent of
teased fibers contained myelin ovoids (consistent with axonal
degeneration). EM revealed occasional stacks of Schwann cell
processes and collagen pockets adjacent to 1 o r 2 nonmyelinated axons; these changes were milder than those noted in
patient 1. Occasional regenerating clusters of small axons were
observed.
Clinical course. Due to the suspicion of possible systemic vasculitis, patient 2 received 6 monthly intravenous
pulses of cyclophosphamide and daily oral prednisone. The
dysesthesias initially improved, but worsened when the cyclophosphamide was temporarily withdrawn because of hematuria. Symptoms stabilized with resumption of this regimen. The
patient still has distal dysesthesias but without new symptoms
while on a regimen of moderate oral doses of prednisone
alone, and the neurologic features are unchanged 12 months
since the diagnosis of vasculitis.
LACOMIS ET AL
1176
A
B
Figure 3. Pathologic appearance of a sural nerve specimen from patient 2. A, A toluidine blue-stained plastic-embedded section from the sural
nerve reveals portions of fascicles exhibiting mild to moderate dropout in small and large myelinated fibers and an epineurial arteriole (lower left)
that is invaded by mononuclear cells. The vessel wall is damaged, and the internal elastic lamina is obliterated. There is mild perivascular
inflammation involving the small vessels in the center. B, A paraffin-embedded, hematoxylin and eosin-stained section shows an arteriole
surrounded and invaded by mononuclear inflammatory cells. There is disruption of the normal vessel wall architecture. Bar = 30 pm.
DISCUSSION
Both of our patients had the hallmark clinical
features of small fiber neuropathy without EMG or
clinical features of large fiber neuropathy (1). The
results of QST (temperature sensation more affected
than vibratory sensation) were consistent with a SFN in
patient 1 and identified mostly subclinical involvement
of large sensory fibers in addition to small fiber involvement in patient 2. Decreased distal sweating (patient 1)
and persistent (“hung-up”) sweat responses (patient 2)
seen on QSART denoted postganglionic sympathetic
(small fiber) dysfunction (9). These objective data were
important in convincing us to pursue the evaluation in
patient 1, who had uncertain subjective symptoms in the
setting of a significant psychiatric disorder.
Patient 1 had no recent risk factors for polyneuropathy, and sural nerve biopsy was performed mainly to
evaluate for amyloid infiltration. Although the past
exposure to alcohol was significant, and alcohol toxicity
with or without malnutrition causes SFN, the time of
maximum exposure was remote and the patient was
asymptomatic during the exposure (there was no history
of malnutrition). The identification of vasculitis was
unexpected. Patient 2 had clinical features that were
suggestive of, but not diagnostic of, systemic lupus
erythematosus (SLE), i.e., polyarthralgias, oral ulcers,
elevated ANA titer, and possible pleuritis (10). She also
had fatigue and dry eyes. Because of the known (although relatively uncommon) association of vasculitic
neuropathy with SLE (9), it was initially suspected that
the cause of the dysesthesias in this patient was peripheral nerve vasculitis. Sjogren’s syndrome is another
consideration, especially in light of the report of dry
eyes, but SS-A and SS-B antibodies were negative on
repeated testing, and tearing was actually normal when
measured objectively. In addition, the known sensory
neuropathy of Sjogren’s syndrome is a large fiber disturbance (ganglionopathy) (11).
In patient 1, EM revealed many flat stacks of
Schwann cell processes consistent with a loss of unmyelinated axons. In addition, the small axonal sprouts
noted on EM probably account for the apparent increase
(due to regeneration) in small myelinated axons that was
noted by morphometric analysis. Our light microscopy
findings excluded significant acute degeneration or regeneration of large myelinated axons, but they did reveal
a small vessel vasculitis. Although active necrosis of
vascular walls was not noted, there was transmural invasion
by inflammatory cells, as well as chronic degenerative wall
changes (“lymphocytic vasculitis”) (Figure 1).
The histopathologic findings in patient 2 revealed
changes more typical of vasculitic neuropathy, including
a spatial pattern of dropout of small and large myelinated fibers (interfascicular variability) that is thought to
be due to ischemia from vasculitis of the vasa nervorum.
Some dropout in large as well as small fibers is not
unusual in patients with clinical and electrophysiologic
features of predominantly distal SFN, and some patients
who present with small fiber symptoms later develop
symptoms and signs of large fiber neuropathy (2). In our
SFN AND VASCULITIS
patient, apparently enough large fibers were spared so
that sensory nerve conductions were normal, and there
was no significant loss of proprioception, vibration, or
tendon reflexes detectable clinically. However, some
large fibers were lost, and this loss was reflected in the
abnormal vibratory thresholds found on QST.
The presumed mechanism of axon damage in our
2 patients is ischemia; however, we were not able to
prove this. There is some support, from an animal model
using partial arterial infarction, for the notion that small
nerve fibers are more vulnerable to ischemia (12), but
further study in humans is required, especially since the
typical spectrum of vasculitic neuropathies includes primarily large fiber sensorimotor neuropathies such as
multiple mononeuropathies (mononeuritis multiplex)
and symmetric sensorimotor polyneuropathies. Multiple
mononeuropathies are usually caused by necrotizing
arteritis involving large epineurial arterioles (up to 250
pm) (13). Distal symmetric polyneuropathies, which
account for 19-37% of all cases of vasculitic neuropathy,
may occur in the setting of systemic as well as nonsystemic (isolated peripheral nervous system involvement)
vasculitis (9). In the series reported by Dyck et al, only 1
of 21 patients with nonsystemic vasculitic polyneuropathy had a sensory disturbance alone, and this patient
had large fiber involvement (sensory ataxia) clinically
(13). Nonsystemic vasculitic polyneuropathy, as in our
patient 1,may have a more benign course than its systemic
counterparts. Interestingly, smaller arterioles tend to be
involved (13). Patient 2 had an undifferentiated connective
tissue disease, but a small vessel vasculitis was identified
only in peripheral nerve in this patient also.
The small vessel involvement is somewhat reminiscent of an apparently immunosuppression-responsive
“lymphocytic vasculitis” that has also been identified in
some diabetic patients with sensorimotor neuropathies
and proximal neuropathies. The vasculitis is usually not
associated with fibrinoid necrosis ( 1 4 3 ) . Since fibrinoid
necrosis may not be easily identified in small-caliber
vessels, the findings of transmural inflammation with
disorganization of vascular walls and the presence of
nuclear debris seem to warrant a trial of typical immunosuppressive therapy for vasculitis.
It is important to note that laboratory findings,
including ESR, are usually normal in patients with
nonsystemic vasculitic polyneuropathy (13), as they were
in patient 1. Even patients with connective tissue diseases and vasculitic neuropathy may have only mild
elevations in the ESR, as in patient 2 (9).
Our observations add predominant small fiber
1177
neuropathy to the spectrum of vasculitic neuropathies
and suggest the need for further study of the role of
vasculitis in some cases of “idiopathic” SFN or SFN
associated with connective tissue diseases. We postulate
that, in some patients, small vessel vasculitis might cause
relatively indolent ischemia, which first affects the function of small axons.
ACKNOWLEDGMENTS
We are grateful to our institution’s electromyography
and histopathology technicians for their excellent assistance
and to Dr. David Chad for his thoughtful review of the
manuscript. The Department of Veterans Affairs provided
equipment for the histometric studies. Dr. Kent Berkey kindly
referred the first patient to us, and Dr. Chester Oddis assisted
in management of that patient.
REFERENCES
1. Stewart JD, Low PA, Fealey RD: Distal small fiber neuropathy:
results of tests of sweating and autonomic cardiovascular reflexes.
Muscle Nerve 15:661-665, 1992
2. Stewart JD, Low P A Small fiber neuropathy. In, Clinical Autonomic Disorders. Edited by PA Low. Boston, Little, Brown, & Co.,
1993
3. Gorson KC, Ropper AH: Idiopathic small fiber neuropathy. Acta
Neurol Scand 92:376-382, 1995
4. Suarez GA, Fealey RD, Camilleri M, Low PA: Idiopathic autonomic neuropathy: clinical, neurophysiologic, and follow-up studies on 27 patients. Neurology 44:1675-1682, 1994
5. Gruener G, Dyck PJ: Quantitative sensory testing: methodology,
applications, and future directions. J Clin Neurophysiol 11:568583, 1994
6. Low PA, Caskey PE, Tuck RR, Fealey RD, Dyck PJ: Quantitative
sudomotor axon reflex test in normal and neuropathic subjects.
Ann Neurol 14:573-580, 1983
7. Low PA: Clinical Autonomic Disorders. Boston, Little, Brown, &
Co., 1993
8. Jacobs JM, Love S: Qualitative and quantitative morphology of
human sural nerve at different ages. Brain 1085397-924, 1985
9. Kissel JT, Slivka AP, Warmolts JR, Mendell JR: The clinical
spectrum of necrotizing angiopathy of the peripheral nervous
system. Ann Neurol 18:251-257, 1985
10. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield
NF, Schaller JG, Tala1 N, Winchester RJ: The 1982 revised criteria
for the classification of systemic lupus erythematosus. Arthritis
Rheum 25:1271-1277, 1982
11. Griffin JW, Cornblath DR, Alexander E, Campbell J, Low PA,
Bird S, Feldman EL: Ataxic sensory neuronopathy and dorsal root
ganglionitis associated with Sjogrens syndrome. Ann Neurol 27:
304-315, 1990
12. Parry GJ, Brown MJ: Selective fiber vulnerability in acute ischemic neuropathy. Ann Neurol 11:147-154, 1982
13. Dyck PJ, Benstead TJ, Conn DL, Stevens JC, Windebank AJ, Low
PA: Nonsystemic vasculitic neuropathy. Brain 110343-854, 1987
14. Said G, Goulon-Goeau C, Lacroix C, Moulonguet A: Nerve biopsy
findings in different patterns of proximal diabetic neuropathy. Ann
Neurol 35:559-569, 1994
15. Krendel DA, Costigan DA, Hopkins LC: Successful treatment of
neuropathies in patients with diabetes mellitus. Arch Neurol
52:1053-1061, 1995