Chronic inflammatory demyelinating polyneuropathy of infancy A corticosteroid-responsive disorder.

Chronic Idammatory Demyelinating
Polyneuropathy of Infancy:
A Corticosteroid-Responsive Disorder
John T. Sladky, MD, Mark J. Brown, MD, and Peter H. Berman, MD
We present the clinical, electrophysiological, and histopathological findings in 6 children with early-onset chronic
inflammatory demyelinating neuropathy. The clinical features initially suggested a genetically determined disorder in
each patient. S u r d nerve biopsy showed changes of chronic demyelination with multifocal endoneurial edema and
mononuclear cellular infiltrates. All children improved with corticosteroid therapy.
Sladky JT, Brown MJ, Berman PH: Chronic inflammatory demyelinating polyneuropathy of infancy:
a corticosteroid-responsive disorder. Ann Neurol 20:76-8 1, 1986
Demyelinating peripheral neuropathies in infancy and
early childhood are most frequently genetically transmitted disorders for which effective therapy is not yet
available [9, 191. When the history and examination of
relatives do not disclose evidence of neuropathy in the
family, the possibility of an acquired, potentially treatable disorder should be considered. We encountered 6
infants and young children with chronic demyelinating
neuropathy in whom the clinical features suggested an
inherited disease, but subsequent electrophysiological
and nerve biopsy findings pointed to the diagnosis of
chronic inflammatory demyelinating polyneuropathy
(CIDP). Each of these children had a favorable response to corticosteroid therapy.
Methods
We reviewed the records of all patients referred to the Children’s Hospital of Philadelphia during a seven-year period
for evaluation of neuropathy, all of whom underwent nerve
conduction studies and electromyography in the electrodiagnostic laboratory at the Hospital of the ,University of Pennsylvania. Patients in whom neuropathy was part of a more
widespread neurological disorder, such as a leukodystrophy,
Refsurn’s disease, Friedreichs disease, or Cockayne’s syndrome, were excluded. About half (62) of the 112 remaining
patients had a dernyelinating neuropathy by electrophysiological criteria (motor conduction slowed to 70% or less of
normal velocity, with or without the presence of multifocal
conduction block [ I s , 231). Twenty-five had an acute monophasic illness from which they recovered without specific
therapy and were diagnosed as having acute idiopathic polyneuritis or Guillain-Barre syndrome.
Thirty-seven children had a chronic demyelinating neuropFrom the Department of Neurology, University of Pennsylvania
School of Medicine, and the Division of Neurology, Children’s Hos-
pital of Philadelphia, Philadelphia, PA 19104.
Received Mar 8, 1985, and in revised form Oct 14. Accepted for
publication Oct 29, 1985.
76
athy without clinical or laboratory evidence of an associated
systemic disorder. In 24 of them, neuropathy was thought to
be genetically determined because of a positive family history. W e divided the remaining 13 patients with chronic
demyelinating neuropathy into two groups. Seven had an
acute or subacute evolution of neuropathy followed by a
chronic relapsing or progressive course. These children
clearly had an acquired demyelinating disorder and were
therefore excluded from the study. The other 6 patients had
insidiously progressive motor symptoms over many months
or years and were considered by the referring physician to
have a genetic neuromuscular disease, despite the absence of
neurological abnormalities in the family. In addition to electrophysiological evaluation, each of the 6 underwent sural
nerve biopsy and a trial of corticosteroid therapy. These 6
patients are the subject of this report (Table 1).
Sensory and motor nerve conduction studies were carried
out using standard methods. Sural nerve biopsy specimens
were fixed in 3.6% glutaraldehyde and 2% osmium tetroxide, dehydrated, embedded in epoxy resin, sectioned, and
stained with MAB trichrome for histological evaluation El].
After the initial evaluation, all patients were treated for
four weeks with oral prednisone (2 mglkglday). The dose
was reduced over the next four weeks to 2 mgikg given
orally on alternate days. Because formal motor testing
proved an unreliable means of measuring strength in the
younger patients, we assessed the clinical course by functional testing to determine, for example, the ability to walk
independently, climb stairs, run, and perform such tasks as
dressing and manipulating objects.
Case Report
Patient 1 was first evaluated at 22 months of age for delay in
motor milestones. He was born to normal nonconsanguineAddress reprint requests to Dr Sladky, Department of Neurology,
Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104.
Table I . Clinical Features of 6 Paiients with Chronic Infammatory Demyelination Neuropathy in Early Childhood
Age of
Onset
Age at Initial
Evaluation
Presenting
Symptoms
Distribution of
Weakness
Enlarged
Nerves
M
F
Birth
12 m o
2-3 yr
2 Yr
22 m o
24 mo
6 Yr
7 Yr
P=D,L=A
P=D,L=A
D > P, A > L
D>P,L>A
+
+
+
5
M
Birth
10 yr
6
M
12 m o
10 yr
Motor delay
Motor delay
Hand weakness
Generalized
weakness
Generalized
weakness
Generalized
weakness
M = male; F
=
Patient
No.
Sex
1
2
M
M
3
4
-
D>P,L>A
+
ous parents after an uncomplicated pregnancy and delivery.
The infant was hypotonic at birth and throughout infancy. At
22 months, he was able to crawl and pull himself to a standing position but could not maintain that posture without
assistance. The family history was negative for neuromuscular disease.
General examination revealed palpably enlarged nerves
and bilateral pes planus. Neurological examination demonstrated generalized hypotonia with moderate symmetrical
weakness in both proximal and distal muscle groups. No
sensory or cerebellar deficit could be detected. Deep tendon
reflexes were absent.
Routine laboratory tests, serum protein electrophoresis,
serum and urine amino acid screening, and leukocyte microsomal enzyme analysis were normal. The cerebrospinal
fluid (CSF) protein concentration was elevated (118 mddl).
Median motor nerve conduction was markedly slowed (2.3
mlsec). Sural nerve biopsy revealed features of a severe
chronic demyelinating neuropathy with subperineurial and
endoneurial edema and increased numbers of endoneurial
mononuclear cells.
Prednisone therapy (2 mg/kg/day) was initiated. By the
fourth week of treatment, the patient was able to walk two or
three steps independently for the first time. After six weeks
he could walk six to eight steps despite persistent distal
weakness. Prednisone was reduced to 2 mglkg on alternate
days. After three months of treatment, he was able to ‘‘walk’
in an upright position on his knees and use an infant walker,
even though distal muscle weakness persisted. Deep tendon
reflexes did not return.
Results
Clinical Features
Five of the patients were male and 1 was female. Their
ages at the initial neurological examination ranged
from 22 months to 10 years (Table 1).The presenting
symptom reflected weakness in each patient. The 2
youngest had never walked and were referred for
evaluation of developmental delay. The 4 oldest had
generalized weakness resulting in impaired ambulation; one had lost the ability to walk. The time of onset
was difficult to determine because of the very slow
progression of symptoms. However, 2 patients were
+
-
-
+
+
+
-
D>P,L>A
female; D = distal muscle groups; P = proximal muscle groups; A = arms; L = legs;
Pes Cavus
Deformity
=
present;
-
=
absent.
hypotonic as neonates, 2 had delayed motor development before their first birthday, and 2 had a subtle
decline in motor abilities between 2 and 3 years of age
(Table 1).
Neurological abnormalities were restricted to the
peripheral nervous system. Results of cranial nerve
testing were normal in all patients. Weakness was
prominent, but its pattern and distribution were variable. Motor findings were symmetrical in 5 patients,
with legs somewhat more severely affected than arms
in all but one child, whose presenting symptom was
hand weakness. Proximal and distal muscles were
equally affected in 2 of 6 patients. Distal weakness
predominated in 4. Two older patients had a stockingglove pattern of sensory loss involving large fibers
more than small. Sensory abnormalities could not be
detected in 4 of the patients. Deep tendon reflexes
could not be elicited in 4 and were present only in the
arms of 2. Peripheral nerves were enlarged to palpation in 4 of the 6 patients. One 7-year-old and two 10year-old patients had pes cavus deformities.
N o patient had a history of antecedent illness or
known exposure to a neurotoxin. Family histories were
negative for neuromuscular disease. Clinical examination of available family members in 5 of the patients
and nerve conduction studies of relatives in 2 patients
failed to disclose evidence of neuropathy in these kinships.
Laboratory Features
Routine serum biochemical and hematological studies,
including sedimentation rate, were normal. Results of
measurements of serum phytanic acid, urine amino
acid screening, and leukocyte lysosomal enzyme analyses were normal. Serum protein electrophoresis failed
to demonstrate a paraprotein in any patient. The CSF
protein concentration was elevated in the 5 patients in
whom it was measured (Table 2). None had CSF
pleocytosis.
Sensory nerve action potentials could be elicited in
Sladky et al: Demyelinating Polyneuropathy in Infants
77
Table 2. Laboratory Findings of 6 Patients with Chronic
lnflamrnatovy Demyelinating Neuropathy in Earh Childhood
Signs of Focal" CSF Protein
Patient Median Motor
Velocity ( d s e c ) Dernyelination (mgidl)
No.
1
2
3
4
5
6
2.3
15
11
40
35
3
118
147
61
Not measured
120
338
-
+
+
-
-
+
"Motor conduction velocity difference of more than 5 d s e c between comparable nerves, dispersion of proximally evoked compound motor action potentials, focal conduction block, a n d o r distal
latency prolonged out of proportion to the degree of proximal slowing [15, 231.
- =
absent;
+
=
present; CSF
=
cerebrospinal fluid.
only 2 patients. The evoked potentials were of low
amplitude, and their latencies were prolonged. Motor
responses were unobtainable in the legs in 4 patients.
The amplitude of compound motor action potentials in
the arms ranged from normal to markedly reduced.
Motor conduction velocities were slowed in all patients and were slow enough to indicate demyelination
in 5 of the 6 (Table 2). Focal conduction abnormalities,
including disproportionately prolonged distal latencies,
dispersed compound motor action potentials, focal
conduction block, and disparity of velocities among
7 8 Annals of Neurology
Vol 20 No 1 July 1986
Fig 1 . Cross-section of surd newe from Patient 3,a boy with
chronic injammatoy demyelinatinR polyneuroputhy. There is a
mild degree of nerite fiber loss, as u'eN a.r demyelinated axons and
remyelinated axovzs with disproportionately thin niyelin sheaths.
Subperineurial and endoneuvial edema is eziident. i x 360 before
36%; reduction.,
Fig 2.Cross-section of surd nerre from Patient 2, about 12
months after the onset of chronir inflammatory demyelinuting
polyneuropathy. Fiber density is reduced, and there are tkinly
remyelinated and demyelinated large-c.aliber axons (arrowheads).
Most rrrnaining large-diameter axoris ure J urrounded by
rudimrvztay orziori-bulb formation.r. ( x 900 before 10% reduc-tion.)
Fig 3. Longitadinal section ofsural nerve from Patient 4, about
j v e years after the onset of chronic infEarnmatory demyelinating
polyneuropathy. The distance between nodes of Ranvier (arrowheads) is short, and internodal myelin thickness is variable, indicating segmental demyelination and remyelination. ( x 900 before 56% reduction.)
comparable nerves, were present in 3 of the 6 patients.
Electromyography uniformly demonstrated changes of
chronic partial denervation.
In each case, surd nerve biopsy findings indicated an
active demyelinating neuropathy. Myelinated fiber
numbers were diminished, and there were demyelinated and thinly remyelinated axons (Figs 1, 2). Onionbulb formations were present but consisted of only a
few lamellae (Fig 2). Segmentally demyelinated and
irregular, short, thinly remyelinated internodes were
evident (Fig 3). Occasional axons undergoing wallerian
degeneration were seen, as well as clusters of regenerating axons. The amount of subperineurial and
endoneurial edema was variable but sometimes massive (Figs I, 4). Excess numbers of mononuclear cells
were distributed throughout the endoneurium and in
subperineurial edema fluid (Fig 4). We did not find
large perivascular collections of mononuclear cells,
however, in any of the patients. Vessel walls were
thickened and capillary endothelial cells were enlarged
in 3 of the 6 biopsy specimens, but there was no
vasonecrosis.
Response to Therapy and Course
All patients were stronger within four weeks after
initiation of corticosteroid therapy. Two responded
within the first week, and the remainder improved
more gradually during the second to fourth weeks of
daily therapy. The 5 patients available for follow-up
examination continued to improve during the second
four weeks of therapy. These 5 patients have been
followed for periods ranging from two months to six
years. Two are corticosteroid-dependent and have required plasmapheresis and/or azathioprine therapy to
decrease the side effects of corticosteroids without loss
Fig 4. Longitudinal section of sural newe from Patient 3 , showing marked endoneurial edema and increased numbers o j mononuclear cells within an area of nerve fber lo.is. x 900 before
56% reduction.)
of strength. Three are receiving 0.3 to 1 mgikg of
prednisone on alternate days and have remained stable
or improved over 2 to 23 months. Follow-up electrophysiological testing 2 to 13 months after the initial
studies did not detect meaningful changes in motor
conduction velocities. However, an increase in the amplitude of compound motor action potentials and the
detection of previously unelicitable sensory nerve action potentials coincided with the patients’ clinical improvement.
Discussion
Demyelinating neuropathies are unusual in infancy or
early childhood [9, 13, 19}. The onset of neuropathic
symptoms in this age group suggests an inherited disorder, and therefore an acquired demyelinating neuropathy may not be considered. The diagnostic difficulty in recognizing a genetically determined disorder
is compounded by the apparent clinical and genetic
heterogeneity of familial demyelinating neuropathies
[3,6, 10-12, 22f in which autosornal dominant, recessive, and sex-linked patterns of inheritance are reported, as well as sporadic occurrences [8, 121.
The clinical features of chronic inflammatory polyneuropathy were first delineated by Dyck and colleagues in 1975 161.These authors described a symmetrical motor and sensory polyneuropathy that
occurred in the absence of an associated illness. Weakness affected both proximal and distal muscles, and
generalized hyporeflexia was the rule. Motor nerve
conduction velocities were slowed, and the CSF protein concentration was increased. Characteristic histopathological findings in sural nerve biopsy specimens
included segmental axonal demyelination and remyelination. In addition, modest perivascular and
endoneurial mononuclear cellular infiltrates, subperineurial and endoneurial edema, and onion-bulb formations were present in some nerves. Since that
Sladky et al: Demyelinating Polyneuropathy in Infants 79
description, inflammatory polyneuropathy has been
widely recognized as a potentially treatable form of
acquired neuropathy in older children and adults f2, 5 ,
6, 10, 18, 211. Patient age at the onset of reported
cases of chronic inflammatory polyneuropathy has
ranged from 3 to 68 years. In most cases demyelinating
polyradiculopathy affects previously well adults and
pursues a progressive or chronic relapsing course.
When CIDP develops more slowly and its progression
is more indolent, it may be difficult to distinguish from
a genetically determined disease.
This report presents the clinical, electrophysiological, and histopathological findings in 6 children with
CIDP. Their neurological disorder consisted of (1) the
insidious onset of a chronic progressive neuromuscular
syndrome at an early age; (2) absence of a family history of neurological disease; (3) nerve conduction velocities slowed to a level usually indicative of a demyelinating neuropathy; (4) neuropathological evidence
of a chronic demyelinating neuropathy with inflammatory features; and ( 5 ) increased strength within four
weeks after initiation of a trial of corticosteroid therapy. Previous studies of patients with chronic progressive or relapsing demyelinating polyneuropathy have
included older children 15, 6, 16, 181, and acquired
demyelinating neuropathy has been reported in children as young as 3 years C6, 16, 181. The studies of
Tasker and Chutorian [201 and Byers and Taft 141,
published before the delineation of CIDP, probably
included children with this disorder who developed
weakness in infancy, some of whom also improved
during treatment with corticosteroids. Two of the children in the present study were described as hypotonic
and weak as newborns. CIDP has not, to our knowledge, been reported previously in infants who were
symptomatic as neonates.
Although there is considerable overlap in the spectrum of characteristics encountered in the two groups
of disorders, several clinical and laboratory features
may be helpful in distinguishing CIDP from inherited
motor-sensory neuropathies in childhood. Unlike most
patients with familial demyelinating neuropathies f8,
111, some of our patients with CIDP had proximal
muscles that were as severely involved as distal muscles. Nerve conduction studies revealed multifocal abnormalities in 3 of the 6 patients with CIDP, whereas
conduction usually is uniformly slowed in inherited
demyelinative neuropathies f15, 231. Sural nerve biopsy specimens showed subperineurial and endoneurial edema with increased numbers of endoneurial
mononuclear cells. Onion-bulb formations typically
were small with few lamellae, even in one patient who
had had symptoms for 10 years, whereas onion-bulb
formations may be larger in inherited neuropathies { 1,
81. Inflammatory changes were consistent findings in
80
Annals of Neurology
Vol 20
No 1 July 1986
this group of children. Edema and inflammatory
changes may occur in genetically determined neuropathies {l], but when present, they usually are not
prominent.
Each of the children in this study improved after
daily corticosteroid therapy. Previous studies have
documented clinical improvement of adults with CIDP
in response to immunosuppressant therapy f2, 8, 10,
14, 16, 18,21). Corticosteroids have been found to be
ineffective in the treatment of patients with one form
of familial demyelinating neuropathy (hereditary
motor and sensory neuropathy, type I) (171. A few
patients with a family history of neuropathy have improved after treatment with corticosteroids C71. Such
patients may have an unusual form of genetically determined neuropathy that responds to steroids, or they
may be individuals from kinships with neuropathy who
also have an acquired demyelinating neuropathy.
Examination of the histopathological features of
surd nerve biopsy specimens has been particularly useful in suggesting which children may respond to corticosteroid treatment. These features tend to be distributed in a multifocal fashion along the course of
peripheral nerves, because inflammatory changes may
not be present in a particular biopsy specimen. On the
basis of our experience with this group of children, we
believe that if a careful examination of relatives fails to
disclose evidence of a familial disorder, a child with
progressive demyelinating neuropathy even in the absence of inflammatory changes in the nerve biopsy
specimen should be given a four-week trial of corticosteroids.
The natural history of CIDP is not well characterized, especially in very young children. Some patients
have a self-limited course, whereas others have a relentlessly progressive disease. For this reason the role
of immunosuppression in the long-term treatment of
children with CIDP will prove difficult to define. Until
more precise guidelines are available, we will continue
to strive for symptomatic improvement using the lowest possible prednisone dose in an alternate-day regimen. When the clinical course has been stable for at
least four months, we attempt to withdraw corticosteroids slowly, increasing the dose if a relapse occurs.
For patients with CIDP who require potentially toxic
high doses of corticosteroids daily for many months to
maintain strength or who have a relapse despite this
therapy, we have used azathioprine and plasmapheresis
as adjunctive immunosuppressants, with successful results.
This research was supported by the Muscular Dystrophy Association
and Grant NS08075 from the National lnstitutes of Health.
We thank Kathleen McDevitt for typing the manuscript.
References
1. Asbury AK, Johnson PJ: Pathology of Peripheral Nerve. Philadelphia, Saunders, 1978, pp 136-147
2. Austin JH: Recurrent polyneuropathies and their corticosteroid
treatment. Brain 81:157-193, 1958
3. Bird TD, On J, Giblett ER, et al: Genetic linkage evidence for
heterogeneity in Charcot-Marie-Tooth neuropathy (HMSN
Type I). Ann Neurol 14:679-684, 1983
4. Byers RK, Taf~LT: Chronic multiple peripheral neuropathy in
childhood. Pediatrics 20:517-537, 1957
5. Dalakas MC, Engel W K Chronic relapsing (dysimmune)
polyneuropathy: pathogenesis and treatment. Ann Neurol
~(suPP~
134) : 145, 1981
6. Dyck PJ, Lais AC, Ohta M, et al: Chronic inflammarory polyradiculoneuropathy. Mayo Clin Proc 50:621-637, 1975
7. Dyck PJ, Swanson CJ, Low PA, et al: Prednisone-responsive
hereditary motor and sensory neuropathy. Mayo Clin Proc
57~238-246, 1982
8. Dyck PJ, Thomas PK, Lambert EH, Bunge R: Peripheral Neuropathy. Philadelphia, Saunders, 1984
9. Evans OB: Polyneuropathy in childhood. Pediatrics 6496-105,
1979
10. Feasby TE, Hahn AF, Brown WF: Long-term plasmapheresis
in chronic progressive demyelinating polyneuropathy. Ann
Neurol 14:122, 1983
11. Harding AE, Thomas PK: The clinical features of hereditary
motor and sensory neuropathy types I and 11. Brain 103:259280, 1980
12. Harding AE, Thomas P K Genetic aspects of hereditary motor
and sensory neuropathy (types I and 11). J Med Genet 17:329336, 1980
13. Kasman ML, Bernstein L, Schulman S: Chronic polyradiculoneuropathy of infancy. Neurology (Minneap) 26:565-573,
1976
14. Levy RL, Newkirk R, Ochoa J: Treating chronic relapsing Guillain-Bard syndrome by plasma exchange. Lancet 2259-260,
1979
15. Lewis RA, Sumner AJ: The electrophysiological distinctions between chronic familial and acquired demyelinative neuropathies.
Neurology (NY) 32:592-596, 1982
16. Low N, Schneider J, Carter S: Polyneuritis in children. Pediatrics 22972-990, 1958
17. Prensky AL, Dodson WE: The steroid treatment of hereditary
motor and sensory neuropathy. Neuropediatrics 15203-207,
1984
18. Prineas JW, McLeod JG: Chronic relapsing polyneuritis. J
New01 Sci 27:423-458, 1976
19. Swaiman KF, Wright FS (eds): The Practice of Pediatric NeurolOW. St. Louis, Mosby, 1982, pp 1169-1203
20. Tasker W, Chutorian AM: Chronic polyneuritis of childhood. J
Pediatr 74699-708, 1969
21. Toyka KV, Augspach R, Paulus W, et al: Plasma exchange in
polyradiculoneuropathy. Ann Neurol 8:205-206, 1980
22. Van Weerden TW,Houthoff HJ, Sie 0, et al: Variability in
nerve biopsy findings in a kinship with dominantly inherited
Charcot-Marie-Tooth disease. Muscle Nerve 5 : 185-196, 1982
23. Wilbourn AJ: Differentiating acquired from familial neuropathies by EMG. Electroenceph Clin Neurophysiol43:616, 1977
Sladky e t al: Demyelinating Polyneuropathy in Infants
81