am (1987) 110, 1617-1630

CHRONIC INFLAMMATORY
DEMYELINATING
POLYRADICULONEUROPATHY
A CLINICAL AND ELECTROPHYSIOLOGICAL STUDY OF
92 C A S E S

by P . A. McCOMBE, J . D . P O L L A R D and J . G . McLEOD

(From the Department of Medicine, University of Sydney, Australia)

SUMMARY

Ninety-two patients with chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) have

been studied in order to define better the clinical features, course and prognosis of the condition
and to identify possible aetiological factors. Sural nerve biopsy was performed on 87 subjects.
Electrophysiological studies were undertaken on all patients and demonstrated marked slowing of
motor conduction and impairment of sensory conduction. The onset was usually gradual but there
was a rapid rate of onset in 15 (16%) patients. Males were more commonly affected than females.
Weakness and paraesthesiae were the most common symptoms but pain was frequently a feature.
Age of onset ranged from 2 to 72 years. Sixty patients (65%) had a relapsing course and 32 patients
(35%) a progressive or monophasic course; there was a significantly earlier age of onset in patients
with relapsing disease. Thirty-two patients (35%) gave a history of preceding infection or some other
possible antecedent precipitating event and there was a significantly higher titre for cytomegalovirus
antibodies in the serum of patients with CIDP than in controls. The patients were followed up for
an average time of approximately ten years. Most patients (73%) had made a good recovery and
were independent, but 7 patients had either died or were completely immobilized as a result of their
disease. The value of treatment with corticosteroid therapy, immunosuppressive agents and plasma
exchange is discussed.

INTRODUCTION

The review of Austin (1958) focused attention on the condition of recurrent or

relapsing neuropathy. Thomas et al. (1969) later described 5 patients with relapsing
Guillain-Barre syndrome and drew attention to the similarities between their cases
and those of Austin (1958). Many descriptive terms used to describe the illness
reflected a lack of clear definition of the condition. These included relapsing
corticosteroid dependent polyneuritis (Matthews et al., 1970), steroid responsive
recurrent polyneuropathy (DeVivo and Engel, 1970), relapsing hypertrophic neur-

Correspondence to Professor J.G. McLeod, Department of Medicine, University of Sydney, NSW 2006,
Australia.
1618 P A . McCOMBE AND OTHERS

itis (Dolman and Allan, 1973), subacute demyelinating polyneuropathy responding

to corticosteroid treatment (Oh, 1978) and Guillain-Barre syndrome with slow
progressive onset and persistent elevation of spinal fluid protein (Hinman and
Magee, 1967). Dyck et al. (1975) introduced the term chronic inflammatory poly-
radiculoneuropathy (CIP) to include patients with progressive and chronic mono-
phasic as well as relapsing courses and described the clinical features of 53 patients
with this disorder. Later they used the term chronic inflammatory demyelinating
polyradiculoneuropathy (CIDP) to define the condition (Dyck et al., 1982). A series
of 23 patients with relapsing and chronic monophasic courses and pathological
features of a demyelinating neuropathy was reported from our laboratory (Prineas
and McLeod, 1976). Since that time the number of patients we have studied has
increased to 92 and we now report our experience with this larger series in order
to define better the clinical features, course and prognosis of CIDP and to identify
possible aetiological factors.

METHODS
Selection of patients
All the patients were referred for investigation between 1968 and 1983 inclusive and satisfied the
diagnostic criteria previously described (Dyck et al., 1975; Prineas and McLeod, 1976; Dyck and
Arnason, 1984). They had a demyelinating neuropathy with either a relapsing and remitting course
or a subacute onset followed by steady progression or subsequent remission. For the purposes of
this classification, a relapse was defined as a worsening of symptoms or signs, resulting in an increase
in the disability by one or more grades on the disability scale with subsequent improvement (Prineas
and McLeod, 1976; McCombe et al., 1987). The histological criteria for demyelination were the
presence in the nerve biopsy of demyelination affecting 15% or more of teased fibres or evidence of
active demyelination and inflammatory infiltrates on electron microscopy or both; the elec-
trophysiological criteria were motor nerve conduction velocities of less than 40 m s" ' in the median
and ulnar nerves, or less than 30 m s~' in the peroneal nerve, or terminal latency greater than 7ms
in the median or ulnar nerves or 10ms in the peroneal nerve (McLeod el al., 1973) or conduction
block and/or differential dispersion of the compound muscle action potential (Lewis and Sumner,
1982). All patients had extensive investigations to exclude other causes of neuropathy (McLeod et
al., 1984). Patients with underlying carcinoma or associated paraproteinaemia were excluded from
the study group.

Follow-up examinations
Attempts were made to contact and reexamine patients during 1983-4. Sixty-seven patients were
reviewed and their disability graded according to the criteria previously defined by Prineas and
McLeod (1976): 0 = normal, 1 -signs but no symptoms or vice versa, 2 = mild motor and/or sensory
symptoms with signs, 3 = moderately disabled by motor and sensory symptoms including ataxia,
4 = requiring assistance with eating or dressing, or using a walking aid, 5 = not ambulant, 6 = dead.

Viral serology
Complement fixation litres of antibody to cytomegalovirus (CMV), Epstein-Barr (EB) virus, herpes
simplex virus and varicella zoster were measured in the Department of Microbiology, Royal Prince
Alfred Hospital, Sydney. IgG antibody to cytomegalovirus was measured in a group of patients by
means of an enzyme immunoassay (Cytomegalisa Test Kit, MA Bioproducts). Patient and control
sera were tested simultaneously. Absorbence of test sera was compared with a standard, and a score
 CHRONIC INFLAMMATORY DEMYELINATING NEUROPATHY 1619

derived for each sample. A score equal to or greater than 0.24 was regarded as positive and was
equivalent to a complement fixation titre of 1:8 or greater.

Sural nerve biopsy

Sural nerve biopsies were performed and prepared for light and electron microscopic examination
and morphometric studies using techniques previously described (Prineas and McLeod, 1976;
McLeod ei al., 1984).

Statistical methods
Mean values were compared using the Student's unpaired t test. Values of ^ = 0.05 were regarded
as significant. For comparison of disability scores, the trend / 2 test was used.

RESULTS

Ninety-two patients (57 males, 35 females) fulfilled the criteria for diagnosis of
CID'P. Sural nerve biopsy was performed on 87 subjects. CSF protein levels in 72
patients on whom lumbar punctures were performed ranged from 0.23 to 5.50 g/1
(mean 1.38, SD 1.11). The results of nerve conduction studies are shown in Table
1 and fig. 1.

TABLE 1. CHRONIC INFLAMMATORY DEM YELI N ATI NG POL YR A DICU LONEU ROPATH Y:
NERVE CONDUCTION STUDIES
Motor conduction Sensoryconduction

Median Dinar Common peroneal

Median Ulnar
Latency Velocity Latency Velocity Latency Velocity Amplitude' Amplitude
(m.v) (m *-') (m.v) (m-.v"') (m.s) (m-s'<) 0'V)

CIDP
MeaniSD 9.1 6.0 30.4 14.4 6.84.3 29.313.1 IO.44.6 29.5 10.9 2.3 4.4 1.63.2
Range 2.6-14.0 9-62 2.8-25 8-54 2.5-23 9-50 0-20 0-17
n 73 77 62 67 48 48 75 69
Controls
MeaniSD 3.4 0.5 56.9 5.0 2.1 0.4 55.74.6 4.6I.O 47.3 4.9 17.97.5 I3.76.4
Range 2.5^.9 49-66 2.1-3.7 47-69 3.3-7.7 38-55 9-40 5-36
n 35 35 35 35 35 35 35 35

Significance P< 0.001 P< 0.001 P< 0.001 P< 0.001 P< 0.001 P< 0.001 />< 0.001 />< 0.001
of difference

Clinical features
Symptoms and signs at presentation
Most patients (72%) presented with a sensorimotor neuropathy; 22% of patients
had a predominantly motor neuropathy and 6% of patients had a sensory neur-
opathy. Weakness was present in 94% of patients and ranged from mild weakness of
the intrinsic muscles of the hands and feet to severe weakness requiring ventilatory
support. Paraesthesiae were reported in 64% of patients and were frequently the
1620 P A . McCOMBE AND OTHERS

70-1

_ 60-
I
X T
o
X
o
f 40
30-
o
c
o
20- I
10-

0J
Median Ulnar Peroneal
FIG. 1. Mean motor conduction velocities in median, ulnar, and common peroneal nerves in control subjects
(open bars) and C1DP patients (hatched bars). Mean values are shown with 1 SD.

presenting symptom when weakness was mild. Pain was a symptom in 20% of
patients in whom it was usually the chief complaint and ranged from a sensation
of burning feet to an aching pain in the muscles. Seventy-two per cent of patients
had clinical evidence of sensory impairment that ranged from increased two-point
discrimination to severe loss of all modalities. Cranial nerve involvement was found
in 16% of patients; ophthalmoplegia was present in 4 patients (4%), facial weakness
in 14 (15%) and bulbar weakness in 6 (6%). Papilloedema was present in 1 patient
whose CSF protein level was 2.2 g/1. Tremor was observed in 3 patients. Autonomic
dysfunction was not a feature of the condition; no significant abnormalities were
found in 14 patients in whom autonomic function studies were performed.

Clinical course
The patients were classified into two major subgroups, relapsing and nonre-
lapsing, according to their clinical course. The relapsing group consisted of 60
patients in whom there was a history of definite relapses and remissions. The
nonrelapsing group consisted of 32 patients in whom there were no relapses and
who had either a subacute onset of a monophasic type of illness or a clinical course
that was chronically progressive.
The age of onset for the whole group of 92 patients ranged from 2 to 72 (mean
35.4, SD 20.3) yrs. There was no significant difference between the age of onset for
males and females. Age of onset for patients in the relapsing group ranged from 2
to 70 (mean 26.8, SD 16.0) yrs and for the nonrelapsing group from 2 to 72 (mean
51.4, SD 17.7) yrs. The difference is significant (P< 0.005) (Table 2,fig.2).
In 15 patients (16%) there was a rapid rate of onset of the neuropathy, the
plateau of disability being reached within a period of four weeks. However these
patients were distinguished from those with the Guillain-Barre syndrome (GBS)
 CHRONIC INFLAMMATORY DEMYELINATING NEUROPATHY 1621

TABLE 2. CHRONIC INFLAMMATORY DEM YELINATING

POLYRADICULONEUROPATHY: AGEOFONSET
Mean SD Range Median

Total group 35.4 20.3 92 2-72 33

Males 37.1 19.4 57 2-71 39
Females 32.6 21.6 35 2-72 25
Relapsing 26.8 16.0 60 2-70 26
Nonrelapsing 51.4 17.7 32 2-72 56

by their subsequent relapsing or progressive course. The age of onset in this group
ranged from 3 to 70 yrs (mean 22; SD 15). In the remaining 77 patients (84%)
there was a subacute onset of symptoms.

Total (n=92)

30-, Relapsing (n=60)

0 10 20 30 40 50 60 70 80
Age of onset (by decades)

FIG. 2. Frequency distribution of age of onset of symptoms in patients with CIDP.

Onset in childhood
There were 10 patients (5 males, 5 females) in whom the age of onset ranged
from 2 to 10 yrs. In 7 patients there was a subacute and in 3 an acute onset of the
disease. Eight patients have been followed up, 7 of whom had no disability and 1
had only minor disability. Three of the patients had experienced relapses in adult
life.
1622 P A . McCOMBE AND OTHERS

Associated diseases
Nine patients had associated diseases with possible allergic or autoimmune
aetiology. These were thyrotoxicosis (4), psoriasis (2), urticaria (1), iritis (1) and
eczema (1). The patients with thyrotoxicosis had been diagnosed and treated before
the onset of the neuropathy and all were euthyroid at the time of diagnosis of
CIDP.

Precipitating factors
History of preceding infection. Twenty-nine patients (32%) gave a history of an
illness in the six weeks preceding the onset of neuropathy (Table 3). In 21 patients

TABLE 3. CHRONIC INFLAMMATORY DEM YELI N ATI NG

POLYRADICULONEUROPATHY: PRECIPITATING FACTORS

Illness No. of subjects

Upper respiratory 18
Gastroenteritis 2
Cystitis 1
Smallpox vaccination 1
Salk polio vaccination 2
Measles infection 1
Varicella infection 2
Mumps infection 1
Tetanus vaccination* 1

(* Pollard andSelby. 1978)

there were nonspecific upper respiratory, gastrointestinal or bladder infections and

in 8 patients the onset of neuropathy was preceded by specific infections or
vaccinations. Complement fixation viral titres were performed on fresh serum of
15 patients. Ten patients (67%) were found to have positive CMV titres ( > 1:4)
but there was no significant elevation of titres for EB virus, herpes simplex or
varicella. Because of the high frequency of positive CMV titres, an enzyme im-
munoassay for cytomegalovirus was performed on the sera of 25 control subjects,
39 CIDP and 22 GBS patients (Table 4). Control subjects were laboratory workers
and medical students, and patients were selected because of their geographic
location and availability for study. The results on CIDP patients were identical to
those obtained with the complement fixation test. The mean titre and the percentage
of patients with a positive CMV titre were significantly greater for the CIDP
patients than the control subjects. There was no significant difference in the scores
of CIDP patients in relapse (mean 0.25, SD 0.14) when compared with those in
remission (mean 0.24, SD 0.15) (/ 3 >0.05). The mean score of patients receiving
corticosteroids (mean 0.27, SD 0.17) did not differ significantly from those not
receiving steroids (mean 0.24, SD 0.14) (P>0.05). There was no correlation of the
score on the CMV immunoassay with age.
Pregnancy. There were 16 women of childbearing age, 9 of whom became
 CHRONIC INFLAMMATORY DEMYELINATING NEUROPATHY 1623

TABLE 4. RESULTS OF ENZYME IMMUNOASSAY FOR C YTOM EG ALO VI RUS

IN CIDP. GBS. AND CONTROL SUBJECTS

Controls GBS patients CIDP patients

No. of subjects 25 22 39
Age of subjects (yrs)
Mean age 29 36 44
SD 5 19 20
Range 23-45 14-68 3-73
Cytomegalovirus test scoresf
Mean score 0.10 0.24* 0.25*
SD 0.08 0.13 0.15
Percentage with positive scores 8% 50% 48%

* P< 0.001. f S0.24 is positive.

pregnant. In 4 of these women onset occurred in pregnancy and in the other 5 there
were relapses related to pregnancy. There was a significant increase in the number
of relapses during the pregnancy and three months postpartum and a tendency for
symptoms to worsen during the third trimester or immediate postpartum period.
The analysis of the association with pregnancy has been reported elsewhere
(McCombe et ai, 1987).

Response to treatment
Seventy-six patients were treated with corticosteroids. In 49 patients (65%) there
was improvement ranging from increased functional ability to complete remission.
In some patients continued corticosteroid therapy was required since relapses
occurred when the steroids were withdrawn. Twelve patients received no benefit
from steroids and were treated with immunosuppressive drugs. Seven patients
received azathioprine with improvement in 4 cases, and 4 of 5 patients treated with
cyclophosphamide improved.
Plasma exchange was performed on 13 patients (Pollard el a!., 1983, 1985). Eight
patients (61%) improved and 2 required regular exchanges ranging from weekly
to third weekly to maintain initial benefit.

Disability at follow-up
Seventy-six patients were alive and available for follow-up and 5 patients had
died as a result of their disease. The interval of time from onset of the disease to
final review or death was 1 to 41 yrs (mean 10.6, SD 9.1). Six patients were lost to
follow-up and 5 others had died as a result of intercurrent illnesses, unrelated to
CIDP. Most patients (73%) had made a good recovery and were independent with
a disability of 2 or less but 2 patients (2%) were grade 5 on the disability scale
because of inability to walk and severe muscle wasting and 5 patients (6%) had
died as a result of their disease (fig. 3, Table 5).
The disability of the patients with relapsing disease was compared with that of
patients with the nonrelapsing type. Patients with the relapsing disease were
1624 P A . McCOMBE AND OTHERS

0 1 2 3 4 5
Grade of disability
FIG. 3. Frequency distribution of grade of disability at follow-up in 81 patients with CIDP.

TABLE 5. C H R O N I C INFLAMMATORY DEM YELIN ATING

P O L Y R A D I C U L O N E U R O P A T H Y : DISABILITY AT FOLLOW-UP

Total Relapsing Nonrelapsin

Disability grade (Prineas and McLeod, 1976)

0 24 (30%) 19 (36%) 5(17%)
1 16(20%) II (21%) 5(17%)
2 19 (23%) 10(19%) 9(31%)
3 14(17%) 7(13%) 7 (24%)
4 1 (1%) 0 1 (3%)
5 2 (2%) 2 (2%) 0
6 Died of CIDP 5 (6%) 3 (6%) 2 (7%)

Other patients
Died of other causes 5 2 3
Lost to follow-up 6 6 0

reviewed after 1 to 41 yrs of illness (mean 12.8; SD 9.6, n = 52). Fifty-seven per
cent of patients had grade 0 or 1 disability and the median disability score was 1.
Patients with the nonrelapsing type of disease were reviewed at 1 to 22 yrs (mean,
6.5, SD, 6.4; n = 29) after the onset of their illness; 34% of patients had grade 0 or
1 disability and the median disability score was 2. The differences are not significant
(Trend/ 2 = 3.3, 0.1 > P > 0 . 0 5 ) . When the nonrelapsing group was further analysed,
the median score for patients with chronic/progressive disorder was 3 and that for
the patients with the subacute remitting disorder was 1.

DISCUSSION
All patients fulfilled the criteria for diagnosis of CIDP (Dyck et ai, 1975; Prineas
and McLeod, 1976; Dalakas and Engel, 1981; Sumner, 1984; Dyck and Arnason,
1984). They had a predominantly symmetric sensorimotor neuropathy with either
 CHRONIC INFLAMMATORY DEMYELINATING NEUROPATHY 1625

a relapsing course or a chronic monophasic or steadily progressive course, without

the presence of an associated causal disease. There was electrophysiological or
histopathological evidence, or both, of peripheral nerve demyelination and com-
monly an elevated CSF protein. The histopathological findings in the sural nerve
did not differ from those described in detail in our earlier report (Prineas and
McLeod, 1976). Other laboratory investigations were normal; patients with para-
proteinaemia were excluded from the present analysis.
We have classified our cases into relapsing and nonrelapsing groups using criteria
similar to those described by others (Thomas et al., 1969; Prineas and McLeod,
1976; Dalakas and Engel, 1981; Schaumburg et al., 1983). Patients who had definite
relapses but without complete recovery are included in the relapsing group; some
of these may fall into the stepwise progressive category described by Dyck et al.
(1975). We did not include in this group patients whose relapses were related only
to reduction or cessation of steroids or immunosuppressive therapy. Cases with a
steadily progressive course, and those with a subacute onset and monophasic
course, have been included in the nonrelapsing group. The difficulties in classifying
precisely the course of the illness in some cases has been previously emphasized
(Schaumburg et al., 1983). Cases of typical GBS in whom a single relapse occurred
with a time course similar to the original episode were defined as recurrent GBS
and not included in the present study (Dyck and Arnason, 1984), but cases that
commenced with an acute onset and then developed a chronic course or had
multiple relapses were included.
Sixty patients (65%) had a relapsing course and 32 (35%) nonrelapsing, pro-
portions similar to those in our earlier study (Prineas and McLeod, 1976). In the
series of Dyck and coworkers (Dyck et al., 1975; Dyck and Arnason, 1984) there
were 34% of patients with a relapsing course and a further 34% of patients who
had a stepwise progressive course. If these two categories together are considered
to be the same as our relapsing group and their gradually progressive (15%) and
monophasic (15%) categories similar to our nonrelapsing group, then the two
series are very similar in their composition.
In our present series there was a predominance of male patients (62%) a finding
that is in accordance with that of other studies (Austin, 1958; Dyck et al., 1975;
Prineas and McLeod, 1976; Oh, 1978; Sladky et al., 1986). As in other series all
age groups were affected by the disease (Dyck et al., 1975; Prineas and McLeod,
1976; Dalakas and Engel, 1981), but there was a significantly earlier age of onset
in patients with the relapsing condition. Early age of onset has been also reported
in other studies of patients with relapsing neuropathy; the average age of onset in
Austin's (1958) series was 23 yrs and 26 yrs in that of Thomas et al. (1969).
The clinical features did not differ substantially from those reported previously
(Dyck et al., 1975; Prineas and McLeod, 1976). Weakness of proximal and distal
muscles, and paraesthesiae are the most common symptoms and reflexes are usually
depressed or absent. Pain was a dominant symptom in 20% of patients, a frequency
similar to that of 15% and 17% in hands and feet respectively reported by Dyck
1626 P. A. M c C O M B E AND O T H E R S

et al. (1975). The cranial nerves were involved in 16% of patients, a frequency
similar to that in other series (Dyck et al., 1975; Dalakas and Engel, 1981). One
patient (1%) had papilloedema, a somewhat lower incidence than that of 7% in
the study of Dyck et al. The mechanism of papilloedema is unknown but is
occasionally seen in the GBS also and may be related to raised intracranial pressure
or an elevated CSF protein (Morley and Reynolds, 1966; Sullivan and Reeves,
1977). Three patients (3%) had an action tremor; intention tremor (Thomas et al.,
1969), ataxic tremor (Prineas and McLeod, 1976) and postural and action tremor
(Dalakas and Engel, 1981) have been previously reported, but the mechanism is
uncertain (Dalakas et al., 1984).
Most patients (72%) presented with a sensorimotor neuropathy; 22% of cases
had a predominantly motor neuropathy and 6% of cases had a predominantly
sensory neuropathy. These findings are similar to those of others (Dyck et al., 1975;
Prineas and McLeod, 1976; Dalakas and Engel, 1981).
Twenty-nine patients (32%) gave a history of an illness or vaccination in the six
weeks preceding the onset of neuropathy. This is higher than the proportion of
14/54 (24%) reported in the literature reviewed by Oh (1978). Dyck and Arnason
(1984) found that 10/57 (19%) of their patients experienced antecedent events
within the three months preceding the onset or exacerbation of the illness (7 received
foreign protein, 3 had upper respiratory tract infections). They concluded that it
was uncertain whether the incidence of antecedent infections exceeded that of the
general population but it is considerably less than that in the GBS, in which
approximately two-thirds of patients give a history of an antecedent acute infectious
illness (Arnason, 1984). Our finding that there was serological evidence of previous
CMV infection in a significantly higher proportion of patients with both CIDP
and GBS than in the control population provides some evidence for an increased
incidence of preceding infection in CIDP. Cytomegalovirus infection has previously
been associated with attacks of GBS (Dowling and Cook, 1981).
Ten patients had an acute or subacute onset of neuropathy in childhood, 3 of
whom had subsequent relapses in adult life. The 8 patients whom we were able to
follow up had little or no disability. Our findings are consistent with those of others
that CIDP in childhood may have an acute, subacute or insidious onset with a
relapsing or progressive course; it is commonly steroid responsive (Colan et al.,
1980; Sladky et al., 1986).
At follow-up, most patients (73%) were not seriously incapacitated by their
illness, having a disability of 2 or less. However, 5 patients had died of the disease
and 2 were quadriplegic and totally dependent. Dyck et al. (1975) found that 60%
of patients were ambulatory and working with minimal disability, 28% were
confined to bed or a wheelchair and 16% had died of the disease. The better
outcome in our patients may reflect improved treatment in recent years with plasma
exchange and immunosuppressive drugs. Patients with a relapsing course tended
to have a more favourable prognosis than those with a nonrelapsing course, but
the differences in disability at follow-up did not reach significant levels.
 CHRONIC INFLAMMATORY DEMYELINATING NEUROPATHY 1627

Forty-nine of 76 patients (65%) in the present study treated with corticosteroids

improved. The efficacy of corticosteroid therapy has been reported by many pre-
vious workers (Austin, 1958; Thomas et al., 1969; Matthews et al., 1970; DeVivo
and Engel, 1970; Dalakas and Engel, 1981; Oh, 1978; Prineas and McLeod, 1976;
Dolman and Allan, 1973; Sladky et al., 1986; Gabreels-Feston et al., 1986). Dyck
et al. (1982) demonstrated in a controlled clinical trial that corticosteroid therapy
resulted in a small but significant improvement compared with no treatment in
both relapsing and progressive cases. Small numbers of our patients who were
unresponsive to steroid therapy appeared to benefit from azathioprine and cyclo-
phosphamide. Most reports of the use of these agents have also been on small
numbers of patients and the value of the treatment remains unproven (Palmer,
1966; Yuill et al., 1970; Walker, 1979; Pentland, 1980; Dalakas and Engel, 1981;
Rosen and Vastola, 1976). One controlled trial showed no better results from
treatment with combined prednisone and azathioprine than with prednisone alone
(Dyck et al., 1985). We have also treated patients who did not respond satisfactorily
to corticosteroid therapy with plasma exchange. Eight of 13 patients (61%)
improved and our finding that some, but not all, patients respond to this treatment
is consistent with the observations of others (Levy et ai, 1979; Server et al., 1979;
Cook et al., 1980; Gross and Thomas, 1981; Toyka et al., 1982; Dyck et al., 1986).
The complication rate with plasma exchange using albumin replacement is so low
in our experience that we use this therapy in preference to corticosteroids in some
patients, particularly those requiring only short courses of plasmapheresis or only
infrequent exchange. Patients with relapsing disease may need only a small number
of exchanges, whereas in chronic progressive cases, exchanges at regular intervals
over long periods of time may be required to maintain benefit. In patients in whom
frequent exchanges are necessary, concomitant immunosuppressive therapy may
help to reduce their frequency (Toyka et al., 1982). Although the role of plasma
exchange in CIDP has been established in a controlled study (Dyck et al., 1986),
there are no such studies of combined immunosuppression and plasma exchange.
Until the pathogenesis of CIDP is better understood or unambiguous evidence of
autoantibody provided, it would seem better to reserve immunosuppression for
patients who fail to respond to corticosteroids or to plasma exchange alone, or
who need frequent exchanges.
The similarities between CIDP and multiple sclerosis, chronic demyelinating
diseases of the peripheral and central nervous systems respectively, have been
previously noted (Prineas, 1970; Dyck and Arnason, 1984). In both diseases immu-
nological mechanisms are probably implicated in the pathogenesis and responsible
aetiological factors include a genetically determined predisposition to the disease
(Stewart et al., 1978; Adams et al., 1979; McCombe et al., 1985) and precipitating
viral infections. The present study has also demonstrated that patients with a
relapsing course have an earlier onset than those with a nonrelapsing course, a
finding similar to that in multiple sclerosis (Confavreux et al., 1980).
1628 P A . McCOMBE AND OTHERS

ACKNOWLEDGEMENTS
Dr P. A. McCombe was a National Health and Medical Research Council Postgraduate Medical
Scholar. We are grateful to Drs J. L. Allsop, L.A. Dawson, J. K. Graham, G.M. Halmagyi, J.T.
Holland, R. Joffe, A. K. Lethlean, J. Leicester, R. Mackenzie, R.A. Ouvrier, P. G. Procopis, J.C.
Walsh, P.M. Williamson, G. Selby, G.A. Wise, W. H. Wolfenden and Professor J.W. Lance for
allowing us to examine patients under their care.

REFERENCES
ADAMS D, FESTENSTEIN H, GIBSON JD, HUGHES RAC, JARAQUEMADA J, PAPASTERIADIS C, SACHS J,
THOMAS PK (1979) HLA antigens in chronic relapsing idiopathic inflammatory polyneuropathy.
Journal of Neurology, Neurosurgery and Psychiatry, 42, 184-186.
ARNASON BGW (1984) Acute inflammatory demyelinating polyradiculoneuropathies. In: Peripheral
Neuropathy, Volume 2. Edited by P.J. Dyck, P.K. Thomas, E.H. Lambert and R. Bunge.
Philadelphia and London: W.B. Saunders, pp. 2050-2100.
AUSTIN JH (1958) Recurrent polyneuropathies and their corticosteroid treatment: with five-year
observations of a placebo-controlled case treated with corticotrophin, cortisone and prednisone.
Brain. 81, 157-192.
COLAN RV, SNEAD OC, OH SJ, BENTON JW (1980) Steroid-responsive polyneuropathy with subacute
onset in childhood. Journal of Pediatrics, 97, 374-377.
CONFAVREUX C, AIMARD G, DEVIC M ((1980) Course and prognosis of multiple sclerosis assessed by
the computerized data processing of 349 patients. Brain, 103, 281-300.
COOK JD, TINDALL RAS, WALKER J, KHAN A, ROSENBERG R (1980) Plasma exchange as a treatment
of acute and chronic idiopathic autoimmune polyneuropathy: limited success. Neurology, New
York, 30, 361-362.
DALAKAS MC, ENGEL WK (1981) Chronic relapsing (dysimmune) polyneuropathy: pathogenesis and
treatment. Annals of Neurology, 9, Supplement, 134-145.
DALAKAS MC, TERAVAINEN H, ENGEL WK (1984) Tremor as a feature of chronic relapsing and
dysgammaglobulinemic polyneuropathies: incidence and management. Archives of Neurology,
Chicago, 41,711-714.
DEVIVO DC, ENGEL WK (1970) Remarkable recovery of a steroid-responsive recurrent poly-
neuropathy. Journal of Neurology, Neurosurgery and Psychiatry, 33, 62-69.
DOLMAN CL, ALLAN BM (1973) Relapsing hypertrophic neuritis. Archives of Neurology, Chicago,
28,351-353.
DOWLING PC, COOK SD (1981) Role of infection in Guillain-Barre syndrome: laboratory confirmation
of herpesviruses in 41 cases. Annals of Neurology, 9, Supplement, 44-55.
DYCK PJ, LAIS AC, OHTA M, BASTRON JA, OKAZAKI H, GROOVER RV (1975) Chronic inflammatory
polyradiculoneuropathy. Mayo Clinic Proceedings, 50, 621-637.
DYCK PJ, O'BRIEN PC, OVIATT KF, DINAPOLI RP, DAUBE JR, BARTLESON JD, MOKRI B, SWIFT T,
Low PA, WINDEBANK AJ (1982) Prednisone improves chronic inflammatory demyelinating
polyradiculoneuropathy more than no treatment. Annals of Neurology, 11, 136-141.
DYCK PJ, ARNASON B (1984) Chronic inflammatory demyelinating polyradiculoneuropathy. In:
Peripheral Neuropathy, Volume 2. Edited by P.J. Dyck, P.K. Thomas, E.H. Lambert and R.
Bunge. Philadelphia and London: W.B. Saunders, pp. 2101-2114.
DYCK PJ, O'BRIEN P, SWANSON C, LOW P, DAUBE J (1985) Combined azathioprine and prednisone
in chronic inflammatory demyelinating polyneuropathy. Neurology, Cleveland, 35, 1173-1176.
DYCK PJ, DAUBE J, O'BRIEN P, PINEDA A, Low PA, WINDEBANK AJ, SWANSON C (1986) Plasma
exchange in chronic inflammatory demyelinating polyradiculoneuropathy. New England Journal
of Medicine, 314, 461^165.
 CHRONIC INFLAMMATORY DEMYELINATING NEUROPATHY 1629

GABREELS-FESTEN AAWM, HAGEMAN ATM, GABREELS FJM, JOOSTEN EMG, RENIER WO, WEEMAES
CM R, TER LAAK HJ (1986) Chronic inflammatory demyelinating polyneuropathy in two siblings.
Journal ofNeurology, Neurosurgery and Psychiatry, 49, 152-156.
GROSS MLP, THOMAS PK (1981) The treatment of chronic relapsing and chronic progressive idiopathic
inflammatory polyneuropathy by plasma exchange. Journal of the Neurological Sciences, 52,
69-78.
HINMAN RC, MAGEE KR (1967) Guillain-Barre syndrome with slow progressive onset and persistent
elevation of spinal fluid protein. Annals of Internal Medicine, 67, 1007-1012.
LEVY RL, NEWKIRK R, OCHOA J (1979) Treating chronic relapsing Guillain-Barre syndrome by
plasma exchange. Lancet, ii, 259-260.
LEWIS RA, SUMNER AJ (1982) The electrodiagnostic distinctions between chronic familial and
acquired demyelinative neuropathies. Neurology, New York, 32, 592-596.
MCCOMBE PA, CLARK P, FRITH JA, HAMMOND SR, STEWART GJ, POLLARD JD, MCLEOD JG (1985)
a-\ antitrypsin phenotypes in demyelinating disease: an association between demyelinating
disease and the allele PiM3. Annals of Neurology, 18, 514-516.
MCCOMBE PA, MCMANIS PG, FRITH JA, POLLARD JD, MCLEOD JG (1987) Chronic inflammatory
demyelinating polyradiculoneuropathy associated with pregnancy. Annals of Neurology, 21, 102
104.
MCLEOD JG, PRINEAS JW, WALSH JC (1973) The relationship of conduction velocity in pathology in
peripheral nerves: a study of the sural nerve in 90 patients. In: New Developments in Elec-
tromyography and Clinical Neurophysiology, Volume 2. Edited by J.E. Desmedt. Basel and
London: S. Karger, pp. 248-258.
MCLEOD JG, TUCK RR, POLLARD JD, CAMERON J, WALSH JC (1984) Chronic polyneuropathy of
undetermined cause. Journal of Neurology, Neurosurgery and Psychiatry, 47, 530-535.
MATTHEWS WB, HOWELL DA, HUGHES RC (1970) Relapsing corticosteroid-dependent polyneuritis.
Journal of Neurology, Neurosurgery and Psychiatry, 33, 330-337.
MORLEY JB, REYNOLDS EH (1966) Papilloedema and the Landry-Guillain-Barre syndrome. Brain,
89, 205-222.
OH SJ (1978) Subacute demyelinating polyneuropathy responding to corticosteroid treatment.
Archives of Neurology, Chicago, 35, 509-516.
PALMER KNV (1966) Polyradiculoneuropathy treated with cytotoxic drugs. Lancet, i, 265.
PENTLAND B (1980) Azathioprine in chronic relapsing idiopathic polyneuropathy. Postgraduate
Medical Journal, 56, 734735.
POLLARD JD, SELBY G (1978) Relapsing neuropathy due to tetanus toxoid: report of a case. Journal
of the Neurological Sciences, 37, 113-125.
POLLARD JD, MCLEOD JG, GATENBY P, KRONENBERG H (1983) Prediction of response to plasma
exchange in chronic relapsing polyneuropathy: a clinico-pathological correlation. Journal of the
Neurological Sciences, 58, 269-287.
POLLARD JD, MCLEOD JG, KRONENBERG H (1985) Plasma exchange in chronic inflammatory de-
myelinating neuropathy. Plasma Therapy and Transfusion Technology, 6, 435.
PRINEAS JW (1970) Polyneuropathies of undetermined cause. Ada Neurologica Scandinavica, 46,
Supplement 44, 1-72.
PRINEAS JW, MCLEOD JG (1976) Chronic relapsing polyneuritis. Journal of the Neurological Sciences,
27, 427-458.
ROSEN AD, VASTOLA EF (1976) Clinical effects of cyclophosphamide in Guillain-Barre polyneuritis.
Journal of the Neurological Sciences, 30, 179-187.
SCHAUMBURG HH, SPENCER PS, THOMAS PK (1983) The inflammatory polyneuropathies (Guillain-
Barre syndrome and related disorders). In: Disorders of Peripheral Nerves. Philadelphia: F.A.
Davis, pp. 25-39.
SERVER AC, LEFKOWITH J, BRAINE H, MCKHANN GM (1979) Treatment of chronic relapsing inflam-
matory polyradiculoneuropathy by plasma exchange. Annals of Neurology, 6, 258-261.
1630 P A . McCOMBE AND OTHERS

SLADKY JT, BROWN MJ, BERMAN PH (1986) Chronic inflammatory demyelinating polyneuropathy
of infancy: a corticosteroid-responsive disorder. Annals of Neurology, 20, 76-81.
STEWART GJ, POLLARD JD, MCLEOD JG, WOLNIZER CM (1978) HLA antigens in the Landry-
Guillain-Barre syndrome and chronic relapsing polyneuritis. Annals of Neurology, 4, 285-289.
SULLIVAN RL, REEVES AG (1977) Normal cerebrospinal fluid protein, increased intracranial pressure,
and the Guillain-Barre syndrome. Annals of Neurology, 1, 108-109.
SUMNER AJ (1984) Chronic demyelinating neuropathies. In: Peripheral Nerve Disorders: A Practical
Approach. Edited by A.K.. Asbury and R.W. Gilliatt. London: Butterworths, pp. 46-57.
THOMAS PK, LASCELLES RG, HALLPIKE JF, HEWER RL (1969) Recurrent and chronic relapsing
Guillain-Barre polyneuritis. Brain, 92, 589-606.
TOYKA KV, AUGSPACH R, WIETHOLTER H, BESINGER UA, HANEVELD F, LIEBERT UG, HEININGER, K,
SCHWENDEMANN G, REINERS K, GRABENSEE B (1982) Plasma exchange in chronic inflammatory
polyneuropathy: evidence suggestive of a pathogenic humoral factor. Muscle and Nerve, 5, 479-
484.
WALKER GL (1979) Progressive polyradiculoneuropathy: treatment with azathioprine. Australian
and New Zealand Journal of Medicine, 9, 184-187.
YUILL GM, SWINBURN WR, LIVERSEDGE LA (1970) Treatment of polyneuropathy with azathioprine.
Lancet, ii, 854-856.
(Received December 9, 1986. Revised March 12, 1987. Accepted March 23, 1987)