Electrophysiological observations in critically ill Guillain–Barre syndrome
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.190271
Source of Support: None, Conflict of Interest: None
Keywords: Critically ill; electrophysiology; Guillain–Barre syndrome; inexcitable nerves; mechanical ventilation
Guillain–Barre syndrome (GBS) is the leading cause for acute neuromuscular weakness worldwide. The disease progresses through a phase of progressive weakness lasting for up to four weeks, followed by a variable plateau phase and then a recovery phase. Paralysis of respiratory muscles is a dreaded complication that occurs in one-fourth of the subjects. Respiratory weakness occurs due to the involvement of both inspiratory and expiratory muscles. Diaphragmatic weakness occurs due to phrenic nerve demyelination. Careful and close monitoring is mandatory to identify patients at high risk for respiratory failure, and triage them to the Intensive Care Unit (ICU) and mechanical ventilation. Clinical features that predict progression to respiratory failure include severe limb weakness, rapid progression, bulbar dysfunction, bifacial weakness, dysautonomia, reduced pulmonary vital capacity, and elevated hepatic enzymes. Antecedent gastrointestinal illness, infection with Campylobacter jejuni and cytomegalovirus, and the presence of anti-GM1 (monosialotetrahexosylganglioside) antibodies are the other factors that portend respiratory paralysis.
Electrophysiological studies are used to confirm the diagnosis of GBS and distinguish it from clinically similar diseases. The findings typically show an evolving pattern of multifocal demyelination in the form of reduction in nerve conduction velocity, partial motor conduction block (CB), abnormal temporal dispersion, prolonged distal latencies, and delayed or absent F or H responses. Electrophysiology also prognosticates recovery from GBS.,,,,, Importantly, it has been used in the early identification of patients who may eventually develop respiratory failure.
However, factors that determine “duration” of respiratory muscle paralysis and mechanical ventilation in GBS have not been studied in large cohorts. Knowledge regarding this may promote the development of strategies to hasten recovery from respiratory paralysis. Since electrophysiological studies reflect the nature and severity of the underlying pathology in peripheral nerves, they may provide crucial information about the nature of abnormalities in ventilated patients. We aimed to analyze in retrospect, the electrophysiological parameters in a large cohort of critically ill patients with GBS and correlate these findings with the duration of mechanical ventilation and outcome at discharge.
Patients satisfying the National Institute of Neurological Disorders and Stroke (1990) diagnostic criteria for GBS, and developing severe weakness requiring mechanical ventilation between July 1997 and June 2007 were retrospectively analyzed. The age, gender, antecedent illness, if any, and duration of mechanical ventilation were recorded. All patients received treatment with plasmapheresis or intravenous immunoglobulin. They were closely monitored for signs of respiratory compromise and were mechanically ventilated in the event of hypoxia, vital capacity <15 mL/kg, PaO2 <70 mmHg, and PaCO2 >45 mmHg. Weaning from the ventilator was initiated when oxygenation was satisfactory, vital capacity was >10 mL/kg, and atelectasis or pulmonary infiltrates were absent in chest X-ray images. The study was approved by the Institute Ethics Committee.
All patients underwent electrophysiological tests, which were carried out at the earliest possible instance, but in the event of rapid progression of illness to respiratory paralysis, the studies were staggered for the duration of mechanical ventilation. The day of illness on which the electrophysiological tests were carried out was noted. Only those patients in whom the following electrophysiological data were available were included in the analysis:
A difference of more than two standard deviations from the mean values standardized at our laboratory was considered abnormal. Patients were categorized based on the following electrophysiological criteria:
Conduction block (CB) was defined by the pCMAP to dCMAP ratio of <0.7. A dCMAP amplitude of at least 20% of the lower limit of normal was considered as an imperative prelude for identifying CB. The cohort was classified into Groups I, II, and III when the duration of mechanical ventilation was 15 days or less, 16–30 days, and 31 days or more, respectively. The outcome at hospital discharge was assessed using the Hughes disability scale (HDS). An attempt was made to correlate the electrophysiological observations with the outcome at hospital discharge and the duration of mechanical ventilation. The data were analyzed using Statistical Product and Service Solutions version 11.0 (SPSS Inc, Chicago).
One hundred and ninety-one patients with GBS were admitted in our ICU during the study period, and 178 required mechanical ventilation. Ninety-three patients had complete electrophysiological data and were included in this study. Of these 93 patients, there were 59 men and 34 women. The mean age was 33.51 ± 21.4 years (range: 1–84 years). The mean duration from the symptom onset to mechanical ventilation was 7.48 ± 5.3 days (range: 1–30 days). The duration of mechanical ventilation was 28.08 ± 27.9 days (range: 1–159 days). The mean duration from the onset of illness to the conduction of the electrophysiological study was 37.07 ± 29.1 days (range: 1–153 days). The clinical and electrophysiological data and duration of mechanical ventilation are summarized in [Table 1] and [Table 2], respectively. Based on the electrophysiological observations, patients could be classified as primary demyelinating (n = 62), equivocal (n = 16), and inexcitable (n = 14). Only one had primary axonal electrophysiology; in this patient, amplitudes of sensory potentials were normal. Thus, in our cohort, none had features of acute motor sensory axonal neuropathy.
Patients with reduced amplitude of median CMAP, as defined by Cornblath et al., and Winer et al., required a longer duration of mechanical ventilation. The individual motor and sensory nerves were analyzed separately. Inexcitable motor nerves were noted more commonly in patients who required mechanical ventilation beyond 31 days. Conversely, inexcitable sensory nerves were associated with shorter duration of mechanical ventilation (<15 days). CBs were noted in motor nerves; no relation between the duration of mechanical ventilation and CB could be discerned. Patients had a worse outcome at hospital discharge as measured by the Hughes Disability Scale (HDS) when the amplitude of median nerve CMAP was low [Table 3].
Fourteen subjects underwent electrophysiological study prior to mechanical ventilation [Table 4]. Subgroup analysis of these patients showed that one each had inexcitable and equivocal electrophysiology based on Hadden criteria, while the rest had primary demyelinating neuropathy. Low-amplitude median nerve CMAP was noted in two and three patients, respectively, when Winer and Cornblath criteria were applied. The duration of mechanical ventilation was 15 days or less in eight, 16–30 days in three, and >30 days in three patients. Motor nerve conduction studies showed an inexcitable median, ulnar and common peroneal nerves in one, one, and three patients, respectively; these patients recovered within 30 days of mechanical ventilation. Sensory nerve conduction studies showed an inexcitable median, ulnar and sural nerves in eight, nine, and nine patients, respectively. Interestingly, here also, inexcitable sensory nerves were noted more commonly in Group I. In Group I, absent median, ulnar, and sural nerve SNAP was noted in six, seven, and six patients, respectively, while in Group II, this corresponded to one, one, and two patients, respectively. In Group III, one patient each had absent median, ulnar, and sural nerve SNAPs. Owing to the small number, we were unable to compare this subgroup with the rest of our cohort using statistical tools.
GBS is an acute immune-mediated neuropathy that causes varying degrees of weakness. Mechanical ventilation improves the outcome in critically ill patients with respiratory paralysis. Plasma exchange and intravenous immunoglobulins are the only therapies with proven efficacy in GBS that shorten the duration to recovery. Despite this, a portion of subjects with respiratory failure do not recover and require prolonged ventilation, which in turn is associated with increased complications such as pneumonia, sepsis, gastrointestinal bleed, and pulmonary embolism among others,,,, and a correspondingly greater financial burden.
Electrophysiological studies define the extent and nature of peripheral neuropathy in GBS and exclude other causes of acute neuromuscular weakness. They also predict “progression” to respiratory paralysis. Patients with primary demyelination have worse disability, lower pulmonary vital capacity, and are at a heightened risk of respiratory failure and mechanical ventilation., Durand et al., reported that, while 46% and 17% of subjects with demyelinating and equivocal electrophysiology required mechanical ventilation, none of the patients with axonal pattern developed respiratory failure. In our study, which included patients with respiratory failure too, majority had a demyelinating pattern of neuropathy. In contrast, Ye et al., showed that 82% of patients with the axonal form of GBS had severe disease, being bed bound or requiring mechanical ventilation. The ratio of pCMAP to dCMAP amplitude of the common peroneal nerve is a useful marker for progression to respiratory failure., Phrenic nerve conduction and electromyogram also serve to indicate which patients require mechanical ventilation., Intercostal nerve conductions have been used to selectively assess the involvement of respiratory muscles. Both these tests failed to gain popularity because of the lack of widespread technical expertise, coupled with inconsistent results from other studies.
The contribution of electrophysiological parameters in predicting progression to respiratory failure has been the subject of many studies. None have systematically correlated electrophysiological parameters and the “duration” of mechanical ventilation or recovery from respiratory muscle paralysis in GBS. A couple of studies have analysed the clinical parameters to identify patients who require a longer duration of mechanical ventilation. Serial pulmonary function tests were assessed in GBS requiring mechanical ventilation for < and >3 weeks. An integrated pulmonary function score based on daily vital capacity and maximal inspiratory and expiratory pressures was found to correlate with duration of mechanical ventilation. In another study of forty mechanically ventilated GBS patients, the lack of foot flexion and the presence of sciatic nerve CB were found to be associated with prolonged mechanical ventilation of >15 days; the caveat is that neither the clinical nor the electrophysiological parameter was adequately defined in the methodology. We could not compare and contrast our observation with other studies, as none had systematically compared the duration of mechanical ventilation with electrophysiological abnormalities in a large cohort of critically ill GBS.
In the current cohort, two-thirds of the patients could be weaned off the ventilator within one month; and, one-third (31/93) required prolonged mechanical ventilation beyond one month. Since our cohort consisted of patients who required mechanical ventilation, demyelinating electrophysiology was over-represented in our cohort. The reason why GBS with demyelinating electrophysiology has a higher likelihood of developing respiratory dysfunction is an open question. Demyelination is multifocal, affecting multiple peripheral nerves, as evidenced by the presence of CBs in the upper and lower limbs; it is possible that the same produces CB of the phrenic nerve. Besides, the pathogenic immune mechanisms underlying demyelinating and axonal GBS are different, with more widespread and progressive damage occurring in the demyelinating form., It can be extrapolated that the duration of respiratory muscle paralysis is likely to be a function of the severity of inflammation. Greater damage to peripheral nerves is reflected as low amplitude or absent evoked motor response.
Analysis of individual nerves showed that inexcitable motor nerves were significantly associated with a longer duration of mechanical ventilation beyond 31 days. Likewise, low amplitude of median nerve CMAP correlated with requirement of mechanical ventilation >31 days. Low CMAP amplitudes of the motor nerves and inexcitable nerves are known to be powerful predictors of an adverse outcome in GBS, including death and inability to walk independently at 48 weeks., 11, ,,,, In a large cohort of 114 patients with GBS, admitted into the ICU, sixty required mechanical ventilation. Inexcitable nerves were associated with poor maximal recovery. In this study, we show that inexcitable motor nerves and low amplitude CMAP of median nerve are associated with delayed recovery from respiratory paralysis in GBS. Electromyography was not performed in our study. Inexcitable nerves may develop during the course of illness on serial electrophysiological examination. This may be a reflection of the ongoing immune-mediated damage to peripheral nerves.
Inexcitable peripheral nerves are seen in only a small proportion of patients, i.e. in 1.4–19% of patients with GBS., 31, ,,, Inexcitable nerves occur due to severe axonal damage or demyelination with CB in the terminal segments of motor axons.,,,,, Infrequent autopsy studies have provided an insight into the pathological basis of nerve inexcitability in GBS. Differential changes are noted in spinal and peripheral nerves in fulminant GBS with inexcitable nerves. Extensive demyelination of spinal nerves occurs with macrophage dominant infiltrate. Peripheral nerves, on the other hand, demonstrate axonal degeneration with only a few residual denuded axons. A combination of widespread demyelination of the proximal spinal roots and secondary Wallerian degeneration More Details of the peripheral nerves due to severe inflammation results in nerve inexcitability.
Thus inexcitable nerves reflect the severe and extensive damage to the nerves. Clinically, this is reflected as prolonged-time-to-recovery from respiratory muscle paralysis. Inexcitable nerves at the initial electrophysiology, though rare, are associated with poor recovery. It is intriguing to note that patients with inexcitable sensory nerves required a shorter duration of mechanical ventilation. The clinical relevance and implications for future research of this finding remain to be drawn. This may indicate a less severe immune-mediated damage of the ventral roots and motor nerves relative to the dorsal roots and sensory nerves. As a theoretical corollary, the less severely affected motor nerves may show a more rapid recovery, reflected as faster improvement in the respiratory function and a shorter duration of mechanical ventilation. The immunobiology of GBS implicates the role of antibodies and T cells that bind peripheral nerve proteins. The differential involvement of motor and sensory nerves may be determined by the underlying antigenic target. Antibodies to gangliosides have a role in specific axonal subtypes of GBS. Our cohort comprised almost exclusively of demyelinating neuropathy, and antibodies to gangliosides and ganglioside complexes were not tested. The contribution of critical illness polyneuropathy in causing a delay in the recovery from the ventilator cannot be excluded. Electrophysiological studies of critical illness polyneuropathy show axonal sensorimotor polyneuropathy with normal or mildly reduced nerve conduction velocities. In our cohort, majority of patients underwent electrophysiology soon after being weaned from the mechanical ventilator and they exhibited demyelinating neuropathy.
Plasma exchange and intravenous immunoglobulins have a proven efficacy in shortening the clinical course and in hastening recovery. This was provided to all our patients. There was no significant difference in the age, gender, and the duration between the onset of illness to the initiation of treatment. Thus, we contemplate that severe multifocal pathological changes as reflected by electrophysiological observations were determinants of the duration of mechanical ventilation in our cohort of GBS. As shown in previous studies, patients with inexcitable or low-amplitude CMAP, as defined by Cornblath et al., and Winer et al., fared poorly with respect to motor disability at hospital discharge.,,,
In the current study, there was a significant heterogeneity with respect to the day of illness on which the electrophysiological tests were performed. Electrophysiological changes evolve sequentially with serial examination. Thus, a single electrophysiological test, particularly when performed early in the course of the illness, may be insufficient to classify GBS into axonal or demyelinating forms., Electrophysiological studies in the 4th–6th week of illness reveal the underlying pathology more precisely and are more predictive of outcome., Our patients underwent electrophysiological study after a median of 29.5 days after symptom onset. Thus, we can hypothesize that these findings quite accurately reflect the pathology of underlying neuropathy. The study has certain drawbacks. This is a retrospective study and we could not control the timing of nerve conduction study; only 14 patients underwent electrophysiological tests prior to mechanical ventilation. Serial tests could not be carried out to note the evolution of changes. Besides, electrophysiological tests were not carried out by a single observer, although standard protocols were followed during nerve conduction studies. We did not have data of “control” patients who did not require mechanical ventilation during the same period for comparison. The power of the study rests in the number and the objective of looking at the duration of mechanical ventilation. Thus, it provides a large volume of electrophysiological data in the critically/severely ill GBS requiring mechanical ventilation.
What is already known is that in patients with GBS who develop respiratory paralysis: (i) Demyelinating electrophysiology is common; and, (ii) longer duration of mechanical ventilation and reduced amplitude of motor CMAP increases morbidity and poor outcome. But this study adds certain associations, namely, that abnormalities in motor nerves are more frequent in patients requiring prolonged mechanical ventilation, while sensory nerve abnormalities are commoner in patients who recover within 2 weeks. These associations may add to the objective prognostication. Electrophysiological tests are noninvasive and provide a reasonable idea of the pathology in peripheral nerves. This may be a reflection of the specific underlying pathophysiological processes that govern the patterns of progression and recovery in the critically ill GBS. What remains to be known is the pattern of development of pathological changes underlying persisting respiratory failure beyond the 4th week of illness and the reparative processes that govern the recovery of peripheral nerves, as well as the role of sensory conduction studies in the prognostic algorithm. This will pave way for the development of better and more potent therapies that hasten recovery in the critically ill GBS, in addition to plasmapheresis and intravenous immunoglobulin.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest
[Table 1], [Table 2], [Table 3], [Table 4]