Post-traumatic (and postsurgical) Guillain-Barré Syndrome: a rare, but treatable entity

Abstract

Guillain-Barré syndrome (GBS) is an acute, monophasic, polyradiculoneuropathy usually provoked by a preceding infection. The cardinal features are progressive weakness in the upper and lower limbs accompanied by loss of deep tendon reflexes. The diagnosis is made on the basis of the clinical history and examination findings, supported by typical cerebrospinal fluid and electrophysiology findings. Trauma and surgery are well understood but rare precipitants of GBS, which clinicians should be aware of, in order not to miss an opportunity to use immunomodulatory therapies. Furthermore, the presence of postsurgical or post-traumatic GBS should prompt careful assessment for underlying malignancy or autoimmune disease associated with an acute demyelinating polyradiculoneuropathy. Here, we present a case of post-traumatic GBS and discuss the potential mechanisms that might underlie this, as well as the investigations and treatment that should be considered.

Background

Guillain-Barré syndrome (GBS) is an acute, monophasic, polyradiculoneuropathy usually provoked by a preceding infection, with well-documented associations with Campylobacter jejuni, Mycoplasma spp, Epstein-Barr virus and recently the Zika virus.¹ GBS has also been associated with vaccinations¹ and malignancies.^{2 3} Clinicians should also be aware of the associations of GBS with surgery and trauma^{4 5} so as to ensure that patients are appropriately investigated and receive timely immunomodulatory therapies. We would encourage clinicians to register any cases of post-traumatic or postsurgical GBS with international registers,⁶ in order that the incidence of this association can be better documented, and appropriate awareness raised.

Case presentation

A 46-year-old man with no prior medical history presented with back pain and ascending weakness following a paragliding accident while on holiday in Spain thirteen days earlier. He had fallen four metres, fracturing his pelvis and T9 vertebra. He also fractured his left distal humerus, which required open reduction and internal fixation under general anaesthetic. Aside from ongoing back pain, he remained well and returned to the UK.

There were no prodromal coryzal or gastrointestinal symptoms, or insect bites prior to his accident. Ten days after the trauma and surgery, he developed paraesthesia in his upper and lower limbs. Despite this, he managed to attend his outpatient fracture clinic appointment, where his arm was recasted. His symptoms of limb parasthesiae were investigated with radiographs of the spine which were deemed to be normal.

The parasthesiae progressed, and the following day the patient felt unwell and light-headed. He collapsed while walking to a General Practice appointment and was taken to hospital by ambulance.

In the emergency department, a trauma CT scan revealed a compression fracture of the T9 vertebrae. Blood tests showed anaemia with a thrombocytosis, but otherwise unremarkable inflammatory markers. A neurosurgical opinion was sought and an MRI scan of the head and whole spine was advised. This showed oedema surrounding T1, T4, T5, minor loss of T6 anterior height with no increase in signal, and the T9 fracture (figure 1). Oedema around the lower sacrum was also noted. His trauma CT was subsequently reviewed and a fracture of the sacrum confirmed.

Figure 1.

MRI spine showing oedema surrounding T1, T4, T5, minor loss of T6 anterior height with no increased in signal, and T9 fracture (indicated by yellow arrow).

Despite these relatively stable MRI appearances, his neurology continued to deteriorate. He was still mobilising to the bathroom on day 2 of admission, but by day 4 he could no longer walk. A neurology consultant opinion was sought. On examination, he was noted to have labile blood pressure. He had mild weakness of neck flexion and extension but there was no bulbar dysfunction. The patient weighed 97 kg, and had a forced vital capacity of 4.16 L, equating to 42.8 mL/kg (normal range 60–70 mL/kg). All other cranial nerves were intact. Examination of the limbs revealed a symmetrical loss of power (grade 3+ to 4/5). Sensation was reduced to pinprick below the wrists and ankles bilaterally. Vibration sense was reduced to the costal margin. There with loss of deep tendon reflexes in the upper and lower limbs. Plantar reflexes were flexor.

Investigations

Routine blood tests on admission showed mild anaemia with otherwise unremarkable inflammatory markers. Serum was negative or normal for Lyme, syphilis, HIV, hepatitis B, C, E, antinuclear antibody (ANA), extractable nuclear antigens (ENA), anti smooth muscle antibody, antimitochondrial antibody, gastric parietal cell antibody, vitamin B₁₂, folate, homocysteine, thyroid function tests, serum protein electrophoresis, HbA1c and copper. Ganglioside (GM1, GM2, GD1a, GD1b, GQ1b), neurofascin-155, 140/186, contactin-1 and Caspr antibodies were also negative.

An MRI spine showed a stable T9 fracture as well as a fracture of the sacrum, but no evidence of any cord compression (figure 1).

Cerebrospinal fluid (CSF) profile showed cytoalbuminological dissociation with protein of 3.2 g/L (normal range 0.18–0.45) and white cell count of <1 x10ˆ9/L(normal range <3); oligoclonal bands were negative. CSF culture showed no growth. In view of the normal routine blood results, absence of fever or systemic symptoms suggestive of infection, further CSF studies were not performed. An ECG showed sinus rhythm.

A CT chest abdomen and pelvis performed in trauma workup did not reveal organ pathology. Nerve conduction studies (table 1) demonstrated reduced amplitude and speed of sensory and motor responses in the upper and lower limbs, with some features of acquired motor conduction block. Electromyography (EMG) examination of the lower limbs (tibialis anterior, extensor digotorum brevis and gastrocnemius muscles) demonstrated electrical silence at rest, together with lack of volitional activity on attempted voluntary contraction. No fibrillations were seen.

Table 1.

Nerve conduction study data consistent with the demyelianting (AIDP) variant of GBS, as indicated by reduced conduction velocities in motor nerves, prolonged distal motor latencies and compound muscle action potential dispersion to proximal stimulation

MCNV data	Lat ms	Amp mv	Dur ms	Dist mm	CV m/s
R median
Wrist—APB	6.9 (≤4.4)	1.0 (≥4)	13.1	NR	NR
Elbow—Wrist	14.5 (<3.0)	0.6 (≥4)	11	305	40.1 (≥49)
R Peroneal
Ankle—EDB	6.2 (<6.5)	0.5 (≥2)	8	NR	NR
Pop fos—Ankle	24.2 (<6.7)	0.2 (≥2)	6.5	460	25.6 (≥44)
L Peroneal
Ankle—EDB	9.1 (<6.5)	0.4 (≥2)	4.1	NR	NR
Pop fos—Ankle	22.7 (<6.7)	0.3 (≥2)	4.4	460	33.8 (≥44)
F-Responses	Mean F ms	Max F ms	Min F ms	M ms	F-M ms
R median
Wrist—APB	39.6	43.5	37.4 (<31)	7.8	29.7
R Peroneal	NR	NR	NR	NR	NR
Ankle—EDB	NR	NR	NR	NR	NR
SNCV data	Lat ms	Amp mv	Dur ms	Dist mm	CV m/s
R hand sensory
Med F2—wrist	NR	NR	NR	NR	NR
Uln F5—wrist	NR	NR	NR	NR	NR
R lower limb sensory
Sural—ankle	1.8	1.0 (≥7)	6	90	50 (≥40)
L lower limb sensory
Sural—ankle	1.6	0.9 (≥7)	8	90	56.3 (≥40)
R radial
Forearm—hand	1.4	0.8 (≥10)	4	75	52.8 (≥50)

Normal values are provided in brackets.

Evidence of sural sparing; as shown by preserved sural nerve amplitude and conduction velocities are also a hallmark of AIDP.

APB, Abductor Pollicis Brevis; CV, Conduction Velocity; EDB, Extensor Digitorum Brevis; GBS, Guillain-Barré syndrome; MNCV, Motor Nerve Conduction Velocity; NR, No Response; SNCV, Sensory Nerve Conduction Velocity.

Differential diagnosis

New limb weakness in the context of a history of trauma is concerning for the presence of cord pathology (including compression, inflammation or haemorrhage) or intracranial haemorrhage or stroke. These diagnoses were quickly excluded with MRI of the head and whole spine.

While both demyelinating and axonal forms of GBS are well described in the literature, the distinction between the two is not always clear, and some cases are equivocal.^{7 8} Caution needs to be applied to interpretation of electrodiagnostic data, which should be carefully considered together with clinical features and CSF findings.

In our study, the reduced conduction velocity in motor nerves, prolonged distal motor latencies, compound muscle action potential (CMAP) dispersion to proximal stimulation supported the diagnosis of a primary demyelinating process, in keeping with the diagnosis of the Acute Inflammatory Demyelinating Polyneuropathy (AIDP) variant of GBS, according to internationally recognised criteria.^{8 9} There were some features of acquired conduction block, as indicated by prolonged ‘F’ wave latencies, although motor amplitude responses were low. The presence of preserved sural nerve amplitude and conduction velocities in our case are supportive of sural sparing, another recognised hallmark of AIDP.¹⁰ By contrast, to make a diagnosis of the axonal variants of GBS, none of the above features of demyelination should be seen in any nerve, although one demyelinating feature in one nerve is permitted if the distal CMAP is <10% of the lower limit of normal.

EMG data are not required to make a diagnosis of GBS, and is often not undertaken as part of diagnostic workup.⁸ In this case, limited needle EMG testing was undertaken. It is difficult to correlate the electrical silence observed at rest with the clinical examination of power at Medical Research Council (MRC) grade 3 to 4+/5 on admission; explanations could be that electrodiagnostic studies, undertaken at day 14 postsymptom onset, correlated with the nadir of the patient’s illness, or alternatively, that cooperation was limited. No fibrillations were seen at rest, as can be observed with the axonal form of GBS after 7–14 days.⁷

It is notable that most case reports of post-traumatic GBS are reported as being axonal,¹¹ but the electrodiagnostic features in our case, on balance, are more supportive of a demyelinating process. Furthermore, demyelinating variants of GBS are more commonly associated with a higher CSF protein and absence of circulating antibodies associated with gangliosides of peripheral nerve myelin antigens,¹² both seen in our case.

Taken together, the acute onset of ascending weakness, sensory disturbance, areflexia and dysautonomia, with the profile of albumino-cytological dissociation on CSF, and electrodiagnostic evidence of a demyelinating polyneuropathy, made it most likely that the diagnosis in this case was one of the AIDP variant of GBS.

The underlying reason for the development of GBS was initially unclear. The patient had been well during his holiday in Spain, with no prodromal coryzal or gastrointestinal symptoms, insect bites or other illness. His inflammatory markers were normal, and there was no serological evidence of infection.

Where no clear infectious precipitant for GBS is apparent, other differential diagnoses accounting for rapidly progressive weakness and sensory loss should be considered, and appropriate investigations undertaken. Where CSF lymphocytosis is present, haematological malignancies, HIV seroconversion and Lyme myelitis and radiculitis should all be excluded. The presence of toxic, metabolic and nutritional causes (including B₁₂ deficiency, malnutrition, porphyria) should be considered, in addition to autoimmune processes, acute muscle and neuromuscular junction disorders.

In this case, myelopathy had been excluded with appropriate imaging and the CSF profile and serological tests excluded haematological malignancy, HIV and Lyme disease. The clinical presentation, in particular the sensory changes and electrophysiological studies (table 1), made the diagnosis of a neuromuscular disorder or myopathy clinically unlikely. Dysautonomia and limb weakness can also occur in the context of Lambert-Eaton Myasthenic syndrome,⁹ although the case presentation is acute rather than subacute or chronic.

Malignancy^2–4 and autoimmune disease⁴ should also be considered as potential precipitants for GBS. In this case, the patient had a CT Chest, abdomen and pelvis as part of trauma workup, which did not demonstrate any solid organ pathology. Likewise, systemic examination and serological testing (including ANA, ENA, antismooth muscle antibody, antimitochondrial antibody) did not reveal any underlying autoimmune disease.

The existences of post-traumatic and postsurgical GBS have been reported in the literature previously^{4 5} although the true incidence is likely to be under-reported. The trauma experienced by this patient was of greater magnitude than the surgery, so it is likely that this is the primary precipitant for the subsequent development of GBS. Furthermore, the chronological course, in particular—the 13-day interval between the paragliding accident, and development of weakness and sensory loss supports the diagnosis of trauma-induced GBS.

Treatment

The patient was treated with intravenous immunoglobulin at 2 g/kg over 5 days with Deep Venous Thrombosis (DVT) prophylaxis and cardiac monitoring.

Outcome and follow-up

Following a 5-day course of intravenous immunoglobulin, his limb strength was much improved; power of intrinsic hand muscles, hip and knee flexors had improved to grade 4/5, with all other muscles at grade 5/5. He continued to have some autonomic disturbance with episodes of labile blood pressure, which were managed conservatively. He was constipated and prescribed laxatives to manage this. There was no arrhythmia on cardiac monitoring. Gabapentin was prescribed for neuropathic pain in his hands and feet.

His pelvic fracture was managed conservatively with analgesia and careful weight bearing.

Two weeks following admission, the patient was able to stand independently, but significant sensory ataxia made mobilising difficult. However, with ongoing physiotherapy, he made good progress and 3 weeks after admission, was mobilising with the assistance of two people. His left arm cast was removed after fracture clinic review, and was healing well.

Four weeks following admission, the patient was mobilising independently, and was discharged home with community therapies in place for ongoing rehabilitation. He completed a marathon 6 months following discharge, although with a slower time than previously.

Discussion

GBS is an acute monophasic polyneuropathy typified by the presence of areflexia and CSF cytoalbuminological dissociation. Demyelinating, axonal and nodal/paranodal subtypes are now recognised. At least half of GBS cases are associated with antecedent infection. The putative mechanism underlying this is molecular mimicry, where antibodies raised against the infective agent cross-react with endogenous peripheral nerve antigens as demonstrated in Acute Motor Axonal Neuropathy associated with GM1-ganglioside antibodies and C. jejuni infection. The AIDP subtype, by contrast, is related to CD4 +T cell mediated inflammation and macrophage-induced demyelination. GBS has also been described after vaccination where microbial antigens are administered with an immunological adjuvant.^{1 13}

GBS has been noted to occur following surgery.^{4 5} A recent population-based study found 3.1% of GBS patients had undergone surgery in the 60 days prior to disease onset, compared with 2.1% who underwent surgery over the same interval 1 year previously (OR 1.48, 95% CI 1.2 to 1.8). Longer operations and surgery on digestive organs and bone were more likely to be associated with GBS.⁴ A retrospective study from the Mayo Clinic spanning 18.5 years, gave a maximum, uncorrected, absolute GBS risk of 1 case for every 1600 such surgeries.⁴ Although small, this risk is in the same order of magnitude as for C.jejuni (1 in 1000) and Zika virus (~1 in 4000) infections.

Post-traumatic GBS occurring independently of surgery has also been described.¹¹ However, many patients with trauma also have surgery, so patients with initial trauma appear in case series of postsurgical GBS and vice-versa.^{4–6 13} The trauma seemed of greater magnitude than the surgery in our case and post-traumatic GBS seemed the more intuitive description of his condition.

The mechanisms by which either trauma or surgery increase the risk of GBS is uncertain but are likely to be similar; current theories include transient immunosuppression, subclinical postoperative/trauma infection, or immune reaction to antigens released during surgery/trauma.⁵

Given the frequency of both trauma and surgery, an interesting question is why GBS is not reported more often in these contexts. While under-reporting of mild cases may be one explanation, a retrospective study⁴ indicated that independent risk factors such as age, malignancy and prior autoimmune disease need to align, predisposing to the development of postsurgical GBS. Postsurgical GBS should, therefore, prompt careful assessment for undiagnosed malignancy or autoimmune disease.⁴

To summarise, post-traumatic and postsurgical GBS is rare but treatable and as such clinicians should have a heightened suspicion for GBS in these situations. Identified cases should be registered with international GBS registers (eg, the International Guillain-Barré Outcome Study) to help further understand the association.⁶

Patient’s perspective.

I felt very battered after my accident, with large bruises in the middle and at the bottom of my back, and my arm plated and in plaster. After 12 days of slow improvement, I woke up with pins and needles in my fingers and toes, and began to feel weak, both worsening over the next 24 hours. I stumbled on the pavement on the way to my GP and had to be helped up—he sent me straight to A&E. Over the next few days I continued to deteriorate, lost the strength to walk, or even lift my head from the pillow, and eventually found difficulty swallowing. Only after CT scans was it found I had fractured a vertebra and sacrum; an MRI scan, lumbar puncture, and eventually nerve conduction tests followed. It was reassuring to be given the diagnosis of GBS, but frustrating not to know how long recovery would take—each day I asked the doctors how much longer before I could stand, walk, run again. Difficulty moving and pain in my fingers and toes made life uncomfortable, and even after I could sit then stand again, a return to normal life seemed a long way off. I walked out of hospital under my own steam (just) and within a week of returning home was able to walk with a stick to the local park. Even though I was able to run a marathon 6 months after leaving hospital, not until more than a year later did I finally feel fully recovered.

Learning points.

• Both surgery and trauma are associated with Guillain-Barré Syndrome (GBS).

• Clinicians should have a low index of suspicion for the diagnosis of GBS in postsurgical patients with new weakness in order that immunomodulatory therapies can be started in a timely manner.

• Postsurgical and post-traumatic GBS should prompt evaluation for the presence of underlying malignancy or systemic autoimmune disease.

• Postsurgical and post-traumatic GBS cases should be registered with prospective trial groups (eg, International Guillain-Barré Outcome Study) to understand true incidence, specific clinical features and underlying mechanisms of disease.