Practice Parameter: Immunotherapy

for Guillain-Barr syndrome

A report of the Quality Standards Subcommittee (QSS)

of the American Academy of Neurology

RAC Hughes, MD; EFM Wijdicks, MD; R Barohn, MD; E Benson,

DR Cornblath, MD; AF Hahn, MD; JM Meythaler, MD;
RG Miller, MD; JT Sladky; JC Stevens, MD

Published in Neurology 2003;61:736-740.

Objective of the guideline:

To provide an evidence-based statement to guide

physicians in the management of Guillain-Barr
syndrome (GBS).
 The AAN develops these presentation
slides as educational tools for
neurologists and other health care
practitioners. You may download and
retain a single copy for your personal use.
Please contact guidelines@aan.com to
learn about options for sharing this
content beyond your personal use.
Methods of evidence review:
MEDLINE search from 1966 and the Cochrane library
(March 2002).
Polyradiculoneuritis limited by human and cross
referenced with therapy.
Search results were reviewed by at least two
members of the GBS practice parameter group.
Recommendations were graded according to the
levels established by the AANs Quality Standards
Subcommittee (QSS).
AANs
Class of evidence for therapy
Class I. High quality randomized controlled trials
(RCTs)
Class II. Prospective matched group cohort studies or
RCTs lacking adequate randomization
concealment or blinding, or potentially liable to
attrition or outcome ascertainment bias
Class Other studies such as natural history studies
III.
Class Uncontrolled studies, case series, or expert
IV. opinion
AANs
Recommendation Levels
Level Established as effective, ineffective or harmful,
A or as useful/predictive or not useful/predictive
Level Probably effective, ineffective or harmful, or as
B useful/predictive or not useful/predictive
Level Possibly effective, ineffective or harmful, or as
C useful/predictive or not useful/predictive

Level Data inadequate or conflicting; Treatment, test,

U or predictor unproven
Introduction:

Prevalence:
GBS affects between one and four per 100,000 of
the worlds population annually.

Economic Impact:
The costs in the US have been estimated as $110,000
for direct health care and $360,000 in lost productivity
per patient.
Introduction:

Health Outcomes:
Respiratory failure requiring ventilation in
about 25% of patients with GBS
Death in 4% to 15% of GBS patients
Persistent disability in about 20% patients with GBS
Persistent fatigue in 67% of patients with
GBS
Question #1:
Does initial immunotherapy hasten
recovery from GBS symptoms?
Diagnostic criteria
In most studies, the primary outcome measure
used disability scale, where:
0 = normal
1 = symptoms but able to run
2 = unable to run
3 = unable to walk unaided
4 = bed-bound
5 = needing ventilation
6 = dead
Most studies included patients with severe disease,
at least grade 3 on that scale.
Analysis of the evidence
Plasma Exchange
Cochrane review obtained data from six Class II trials
comparing plasma exchange (PE) alone to
supportive care
The PE regimens involved exchanging about one
plasma volume on five separate occasions spaced
out over one to two weeks
One trial which used two plasma volume exchanges
on alternate days for a total of four exchanges
Analysis of the evidence
Author / Year Class Results
Greenwood, II Improved by one or more disability
1984 RCT grades after four weeks
Compare PE with PE group 50%;
supportive treatment
Control group 40%
Osterman, II Improved by one or more disability
1984 RCT grades after four weeks (p<0.025).
Compare PE with PE group 77.8%;
supportive treatment Control group 30%

Complete muscle strength recovery

after one year.
PE group 94.4%;
Control group 80%
Analysis of the evidence
Author/Year Class Results
The Guillain- II Improvement by one or more grades
Barr RCT at one month (p<0.01)
syndrome PE group 59%;
Study Group, Control group 39%
1985
Compare PE with Failed to recover walking unaided
supportive after 6 months. (p<0.05)
treatment
PE group 18%;
Control group 29%

Ventilated patients improvement by

one or more disability grades at one
month (p<0.01)
PE group 50%;
Control group 35%
Analysis of the evidence
Author / Class Results
Year
Farkkila, II RCT Handgrip strength was significantly
1987 greater in the PE group (p<0.001)
Compare
PE with
supportive The mean ( SD) time on ventilator
treatment was slightly shortened
PE group (n=4) 11.7 12.2 days;
Control group (n=3) 15.3 6.1 days

The mean recovery time in days was

almost identical between the two
groups
PE 76.6 88.4 vs.
Supportive Treatment 79.1 55.8
Analysis of the evidence
Author/Year Class Results

French II PE patients recovered walking with

Co-op Group RCT assistance faster than the control
on plasma patients (p<0.01)
exchange in
GBS, Recovered 1 or more disability grades
1987 after 4 weeks
Compare PE with PE group 67/109;
supportive Control group 41/111
treatment

For ventilated patients, time to onset of

recover walking assistance was shorter
in the PE than the control group (p<0.05)
Analysis of the evidence
Author / Class Results
Year
French II RCT In the PE group, time to onset of
Co-op Group motor recovery was significantly
on plasma shortened compared to the control
exchange in group (p=0.0002).
GBS,
1997 The number of patients with one or
Compare PE with more grades of improvement at one
supportive treatment
month was significantly more
PE group 56.5%;
Control group 28.3%
Conclusions
Plasma exchange hastens recovery in non-ambulant
patients with GBS who present within four weeks from the
onset of neuropathic symptoms (Class II evidence).

Plasma exchange also hastens recovery in ambulant

patients who present within two weeks but the evidence is
limited to one trial (Class II evidence).

The effects of plasma exchange and IVIg are equivalent in

patients requiring aid to walk(Class I evidence).

Treatment with CSF filtration has not been adequately

tested (Limited Class II evidence).
Recommendations
PE is recommended in non-ambulant patients
within four weeks from onset (Level A, Class II
evidence).

PE is recommended for ambulant patients within two

weeks from onset (Level B, limited Class II
evidence).
Analysis of the evidence
IV Immunoglobulin
Three trials compared IVIg with PE. The mean
improvement in disability grade four weeks after
randomization was available.
In one Class III trial comparing IVIg with supportive
treatment, seven of nine children who received IVIg
recovered completely by four weeks compared with
two of nine untreated.
Cochrane systematic review found no trials
comparing IV immunoglobulin (IVIg) with placebo.
Analysis of the evidence
Author/Year Class Results

van der II Non- Patients improved by one or

Mech, blinded more grades (p=0.024) after
et al., 1992 RCT four weeks
Compare IVIg with IVIg group 53%;
PE
PE group 34%

Median time to recovery of

unaided walking (p=0.07)
IVIg group 55 days;
PE group 69days
Analysis of the evidence
Author/Year Class Results
Grses, III Alternate Recovered full strength after
1995 allocation four weeks (p=0.06)
Compare IVIG Controlled IVIg group 77.8%;
with supportive
trial Control group 22.2%
treatment
(children)
Median time to recover
unaided walking
IVIg group 15 days (r=11-
20);
Control group 24.5 days
(r=21-28)

After one year all the IVIg

patients had recovered
Analysis of the evidence
Author/Year Class Results

Bril, et al., 1996 II RCT Median time to recover ability

Compare IVIG with to do manual work
supportive treatment
IVIg group 65 days;
PE group 90 days

Mean disability grade

improvement
IVIg group 1.2;
PE group 1.0
Conclusions
Intravenous immunoglobulin has not been
adequately compared with placebo (limited Class II
evidence).
Such comparison is not now needed because, when
started within two weeks from the onset, IVIg has
equivalent efficacy to PE in hastening recovery from
patients with GBS who require aid to walk (Class I
evidence).
Multiple complications were significantly less
frequent with IVIg than with PE (Class I evidence).
There is no evidence concerning the relative efficacy
of PE and IVIg in patients with axonal forms of GBS.
Recommendations
IVIg is recommended for patients with GBS who
require aid to walk within two (Level A
recommendation) or four weeks from the onset of
neuropathic symptoms (Level B recommendation
derived from Class II evidence concerning PE
started within the first four weeks).

The effects of IVIg and plasma exchange are

equivalent. (Level B recommendation Class I
evidence concerning the comparisons between PE
and IVIg started within the first two weeks).
Analysis of the evidence

Combination treatments
One Class I trial showed that PE followed by IVIg
showed no significant benefit compared with PE
alone in any measured outcome.
Analysis of evidence
Author/Year Class Results
PSGBS II No significant difference in
Group, 1997 Single any outcome measure
To compare IVIg blind between any of the three
with PE and with
PE followed by IVIg RCT regimens
The difference between the
change in disability grade
between PE and IVIg was so
small as to fulfill previously
declared criteria for
equivalence
Analysis of evidence
Author / Class Results
Year

Nomura II RCT No significant difference in any

et al., outcome measure
2001
To compare
IVIg with PE
and with PE
followed by
IVIg
Conclusions
Sequential treatment with PE followed by IVIg does
not have a superior effect to either treatment given
alone (Class I evidence).

Sequential treatment with immunoabsorption

followed by IVIg has not been adequately tested
(Limited Class IV evidence).
Recommendations

Sequential treatment with PE followed by IVIg is not

recommended (Level A recommendation, Class I
evidence).

Immunoabsorption followed by IVIg is not

recommended (Level U recommendation, Class IV
evidence).
Analysis of the evidence

Immunoabsorption
An alternative technique to PE, which removes
immunoglobulins.
Has the advantage of not requiring the use of a
human blood product as a replacement fluid.
In a prospective trial there were no differences in
outcome between 11 patients treated with PE and
13 treated with immunoabsorption
Conclusion
There is only limited Class IV evidence from a
single small non-randomized, unblinded study.

Recommendation
The evidence is insufficient to recommend the use of
immunoabsorption (Level U recommendation, Class
IV evidence).
Analysis of the evidence
Steroids
Cochrane systematic review sought all trials of any
form of corticosteroid or adrenocorticotrophic
hormone treatment for GBS. Six randomized trials
were identified.
The corticosteroid regimens included intramuscular
ACTH, intravenous methylprednisolone,oral
prednisolone, or prednisone.
The primary outcome measure in the systematic
review was the improvement in disability grade four
weeks after randomization.
Analysis of evidence
Author/Year Class Results
Swick and II Average disease duration, excluding
McQuillen, RCT one ACTH patient who died
1976 ACTH group 4.4 months;
Effect of ACTH Placebo patients 9.0 months.

Hughes et al., II Less improvement in disability grade

1978 RCT after one, three and 12 months in
Effect of the prednisolone than the untreated
prednisolone patients, which was significant
(p<0.05) for those randomized
within seven days from onset
Analysis of evidence
Author/Year Class Results
Mendell et II Alternate No significant difference in any
al., 1985 allocation outcome
Effect of plasma controlled
exchange and
prednisone
trial
Shukla et al., I RCT No significant difference in any
1988 outcome
Effect of
prednisolone
Analysis of evidence
Author/Year Class Results
GBS Steroid I RCT The mean difference in
Trial Group, disability grade after four
1993 weeks was 0.06 (-0.23 0.36)
Effect of iv grade more improvement in
methyl-
prednisolone
the steroid than the placebo
group
Neither this nor any other
outcome variable showed a
significant difference
Singh et al., II Alternate No significant difference in any
1996 allocation outcome
Effect of CT
prednisolone
Analysis of evidence
Author/Year Class Results
The Dutch III 76% improved one grade;
GBS Group, observational Control group 53% (p=0.04)
1994 series with
Effect of iv historical
methyl-
Prednisolone
controls
added to IVIg
Conclusion
The combined evidence from all trials shows no
benefit from corticosteroids (Class I evidence).
The results of a trial of the combination of
intravenous methylprednisolone and IVIg are
awaited.

Recommendation
Corticosteroids are not recommended in the
treatment of GBS (Level A, Class I evidence).
Question #2:
Are there special issues in the
management of children with GBS?
Analysis of the evidence
GBS in Children
The clinical features of GBS in children are similar to
those in adults except that severe conditions are less
common and axonal forms of the disease are more
frequent in some populations.
In younger children, in particular, pain is frequently the
only symptom they are able to articulate and evidence
of subtle weakness and loss of reflexes may be
overlooked.
There is a lack of adequate randomized controlled
treatment trials in children to define the role of either
PE or IVIg.
Conclusion
There are no adequate randomized controlled
trials of treatment in children.

Recommendation
Plasma exchange or IVIg are treatment options for
treating children with severe GBS (Level B
recommendation derived from class II evidence in
adults).
Future research
More research is needed to evaluate immunotherapy
in GBS, particularly the use of combination
treatments and further treatment after the initial
course.
There is a need to identify patients who are at greater
risk of an adverse outcome and to discover whether
subgroups have differential responses to treatment
(including children, people with axonal forms of GBS,
and Fishers syndrome).
Research should also investigate the best methods of
supportive care for monitoring autonomic and
pulmonary function, weaning from ventilation, treating
pain, managing fatigue, and rehabilitation.
Summary of AAN
recommendations for
immunotherapy for GBS
1. Plasma exchange is recommended in
non-ambulant adult patients with GBS
who present within four weeks from the
onset of neuropathic symptoms. Plasma
exchange should also be considered in
ambulant patients who present within
two weeks from the onset of neuropathic
symptoms.
Summary of AAN
recommendations for
immunotherapy for GBS
2. Intravenous immunoglobulin (IVIg) is
recommended in non-ambulant adult patients
with GBS within two or possibly four weeks
from the onset of neuropathic symptoms. The
effects of plasma exchange and IVIg are
equivalent.

3. Corticosteroids are not recommended in the

treatment of GBS.
Summary of AAN
recommendations for
immunotherapy for GBS
4. Sequential treatment with PE followed by IVIg or
immunoabsorption followed by IVIg is not
recommended for GBS.

5. Plasma exchange or IVIg are treatment options for

treating children with severe GBS.
To access the full guideline please visit:
AAN.com/Guidelines