Predicting the course of Guillain-Barre syndrome

Guillain-Barre syndrome is a post-infectious immunemediated polyneuropathy characterized by rapidly progressive weakness, mostly accompanied by sensory disturbances. By definition, maximum weakness is reached within 4 weeks from onset of disease; however, most patients reach their nadir within 2 weeks. The disease is treatable, but a substantial proportion of patients need mechanical ventilation or remains disabled for a long time. Some patients have a highly progressive disease course that requires mechanical ventilation within hours or a couple of days. Respiratory dysfunction must be anticipated and promptly recognized. Pulmonary decompensation frequently occurs during sleep in the supine position when breathing is almost entirely dependent on the diaphragm. At least three factors contribute to the development of respiratory insufficiency: bulbar dysfunction, respiratory muscle weakness, and secondary insufficiency due to atelectasis or pneumonia. In most centres, the majority of patients with Guillain-Barre syndrome will, at least initially, be admitted to a general ward. Respiratory failure, which occurs in at least 17–30% of patients, is the most severe complication. Ideally, patients at high risk of developing respiratory insufficiency would be identified at an early stage of disease.

In this issue of The Lancet Neurology, Durand and colleagues6 provide simple parameters that can be used to predict respiratory failure in Guillain-Barre syndrome. They aimed to identify clinical and electrophysiological predictors using a prognostic model based on classification trees. The authors prospectively assessed a set of related clinical factors such as the speed of progression, as expressed by the duration between onset of the disease and admission, inability to lift the head, and vital capacity. Antecedent diarrhoea, liver dysfunction, and serum antibodies against various gangliosides were also included in the model. Additionally, they undertook electrophysiological assessment (electromyography) on average 2 days after admission and classified the patients as demyelinating, axonal, equivocal, unexcitable, or normal. All clinical and electrophysiological data were used in the predictive model. A random selection of two-thirds of the patients was used as a fitting set; validation was done in the remaining third. Patients with demyelinating Guillain-Barre syndrome were at highest risk of requiring mechanical ventilation. A proximal/distal compound action potential (p/dCMAP) ratio of the common peroneal nerve of above 56% was a good dichotomising parameter, which, along with vital capacity of above 80%, allowed the identification of a group of patients with Guillain-Barre syndrome at very low risk for developing respiratory failure.

The data from this carefully executed study can be useful in clinical practice, especially when considering the need for admission to an intensive care unit and when monitoring pulmonary function very closely. The study also gives additional information about the prognosis of the disease after 6 months, although this was not the primary objective. A very recent prognostic study that used only readily obtainable clinical factors in the early course of Guillain-Barre syndrome is more adequately designed for predicting outcome after 6 months.

Apart from possible selection bias, as only 154 of 302 patients were included because they had a complete electromyograph by the same neurophysiologist before mechanical ventilation, and the suggested favourable effect of plasma exchange compared with intravenous immunoglobulin, which might be explained by a less frequent use of plasma exchange in severe cases, several other issues need further discussion.

From a practical point of view, Guillain-Barre syndrome in general is a clinical diagnosis that can be made without use of electromyography. Furthermore, it is not always possible to get a proper electromyographic examination 24 h a day. Clinical factors, such as age, difficulties with swallowing, progression of weakness, and a reduction of vital capacity (or other simple pulmonary parameters), are more easy to obtain and should preferentially be used. The authors, however, included a large number of these clinical factors in their analysis and vital capacity seemed to be a remaining factor predicting the need for mechanical ventilation. From a pathophysiological point of view it is diffi cult to understand why examination of a nerve in the leg and not the median nerve in the arm or even the phrenic nerve would be the best indicator of developing respiratory dysfunction.

Despite a defned set of criteria, in clinical practice it may be difficult to make a clear and definite distinction between axonal and demyelinating findings.7 In the cohort of Durand and colleagues,6 a substantial proportion of non-ventilated patients had equivocal electrophysiology, which could have changed into overt demyelination if these patients were examined a few days later. Would that have affected the prognostic model? Some patients with Guillain-Barre syndrome are only mildly affected, as may be the case in patients with acute motor axonal neuropathy (AMAN). Patients with AMAN, however, can also have a very rapidly progressive neuropathy needing mechanical ventilation. Although the study by Durand and co-workers provides a useful prognostic algorithm for respiratory progression in Guillain-Barre syndrome in general, individual patients can have an unpredictable rapid progression and all patients with the disease should be closely monitored at regular intervals, generally at least every 4–6 h, including those with mild weakness or axonal features on electromyographic examination. This monitoring includes the observation of possible autonomic dysfunction as well as the risk of developing respiratory failure. Very rapid disease progression, bilateral facial palsy, inability to cough, tachypnoea, sweating, tachycardia, and use of accessory respiratory muscles are key clinical values that can be assessed easily many times a day, even when it is difficult to measure reliable vital capacities. If a patient is at high risk of needing mechanical ventilation, this should be anticipated and discussed in a relatively quiet period and not during emergency intubation. Knowledge of prognostic factors can substantially improve patient care. New studies should investigate the relation between common peroneal nerve blockade, demyelination in general, and a reduced vital capacity with respect to the occurrence of mechanical ventilation. These studies are needed to replicate the findings of Durand and colleagues in an independent group of patients with Guillain-Barre syndrome and should preferentially also investigate other parameters related to disease course in the acute and chronic phases of the disease.

Pieter A van Doorn, Bart C Jacobs
Department of Neurology (PAvD, BCJ) and Department of
Immunology (BCJ), Rotterdam, Erasmus MC, Netherlands
p.a.vandoorn@erasmusmc.nl
We have no conflicts of interest.

1 Hughes RAC, Cornblath DR. Guillain-Barré syndrome. Lancet 2005;
366: 1653–66.

2 Asbury A, Cornblath DR. Assessment of current diagnostic criteria for
Guillain-Barré syndrome. Ann Neurol 1990; 27: S21–24.

3 Hughes RAC, Wijdicks EFM, Barohn R, et al. Practice parameter:
immunotherapy for Guillain-Barré syndrome: report of the Quality
Standards Subcommittee of the American Academy of Neurology.
Neurology 2003; 61: 736–40.

4 Van Doorn PA. Treatment of Guillain-Barré syndrome. J Peripheral Nerv Syst
2005; 10: 113–27.

5 Hughes RAC, Wijdicks EFM, Benson E, et al. Supportive care for patients
with Guillain-Barré syndrome. Arch Neurol 2005; 62: 1194–98.

6 Durand MC, Porcher R, Orlikowski D, et al. Clinical and electrophysiological
predictors of respiratory failure in Guillain-Barré syndrome. Lancet Neurol
2006; 5: 1021–28.

7 Hadden RD, Cornblath DR, Hughes RA, et al. Electrophysiological
classifi cation of Guillain-Barré syndrome: clinical associations and
outcome. Ann Neurol 1998; 44: 780–88.

8 Van Koningsveld R, Steyerberg EW, van Doorn PA, Jacobs BC. A bedside
prognostic scoring system for Guillain-Barré syndrome. J Neuroimmunol
2006; 178: 223.

9 McKhann GM, Cornblath DR, Griffi n GW, et al. Acute motor axonal
neuropathy: a frequent cause of acute fl accid paralysis in China. Ann Neurol
1993; 33: 333–42.

10 Lawn ND, Fletcher DD, Henderson RD, et al. Anticipating mechanical
ventilation in Guillain-Barré syndrome. Arch Neurol 2001; 58: 893–98.