Clinical manifestations

Presentation and course

Acute motor axonal neuropathy presents as an acute, flaccid, symmetrical ascending paralysis with increased cerebral spinal fluid protein, suggesting Guillain-Barré syndrome. Sensory impairment is minimal, and autonomic involvement less common. Peak severity is reached within 5 to 9 days (42). Weakness may involve dysphagia, dysarthria, and facial diplegia and may progress to flaccid quadriplegia and respiratory failure. Extraocular muscle involvement rarely occurs (108). Focal motor involvement was suggested by a small study from South India reporting that 12 patients with acute motor axonal neuropathy out of 84 patients with Guillain-Barré syndrome had prominent finger extensor weakness with relative preservation of finger flexors, wrist flexors, and wrist extensors (46). Reflexes are typically absent in advanced stages of the disease, and deep tendon reflex responses correlate with the severity of weakness; however, normal to increased reflexes have been reported in rare cases (190; 29; 149), especially in acute motor axonal neuropathy patients with positive anti-GM1 antibody (91) and in one case showing hyperreflexia and bilateral papillitis (122). Areflexia or hyporeflexia is a diagnostic criteria for Guillain-Barré syndrome. However, it should be known that in rare instances all Guillain-Barré syndrome subtypes can initially present with hyperreflexia (172). Guillain-Barré syndrome patients with hyperreflexia are more frequently associated with antecedent diarrhea than upper respiratory tract infection and with acute motor axonal neuropathy than acute inflammatory demyelinating polyneuropathy. A vast majority of patients have generalized hyperreflexia, and antiganglioside antibodies are frequently present.

Although autonomic dysfunction is less common, bowel, bladder, and erectile dysfunction have been documented in acute motor axonal neuropathy (148). A patient with acute motor-sensory axonal neuropathy was reported to develop autonomic dysfunction manifesting as severe pulmonary hypertension (143).

Patients with acute motor axonal neuropathy can rarely have concomitant involvement of the CNS. Involvement of the white matter or the spinal cord was reported in acute motor axonal neuropathy cases with hyperreflexia (116; 147). Multiple case reports indicate that some patients with acute motor axonal neuropathy develop altered consciousness or a comatose state and loss of brainstem reflexes likely due to associated Bickerstaff brainstem encephalitis (186; 112).

Cerebellar ataxia has also been reported in association with acute motor axonal neuropathy (16; 99). Lee and Han reported another case of acute motor axonal neuropathy in a 3-year-old male child with atypical presentation and CNS involvement (95). This patient presented with complaints of severe pain and burning sensations as well as flaccid paralysis of the whole body that was worse in the lower extremities. His CSF parameters were in the normal range, but entire spine MRI after gadolinium injection showed high-signal intensity in the cervical and lower thoracic spinal cord as well as cauda equina in the T1-weighted image. Nerve conduction studies were performed on day 13. Sensory nerve action potential (SNAP) results were within normal limits, and motor nerve conduction studies showed inexcitable tibial nerves bilaterally. Electrophysiologic findings were typical of acute motor axonal neuropathy, but the sensory symptoms in this patient were not in concordance with the diagnosis of acute motor axonal neuropathy. Thus, acute motor axonal neuropathy can be seen in patients with atypical sensory symptoms, along with central nervous system involvement (95).

Chi and colleagues presented a unique case of asymmetric acute motor axonal neuropathy with unilateral tongue swelling due to acute hypoglossal nerve palsy (31). The patient initially presented with acute onset of left-sided weakness that mimicked stroke. The muscle weakness was asymmetric, with a left upper limb power of MRC grade 3/5 and a lower limb power of grade 1/5, but a right upper limb power of grade 1/5 and a lower limb power of grade 2/5. Deep tendon reflexes of the four limbs were normal initially. The patient had no history of recent infection or vaccination. Subsequently, the patient progressed to areflexia, and a nerve conduction study was performed that showed motor axonal neuropathy. Therefore, acute motor axonal neuropathy was suspected. Antiganglioside antibodies were checked and IgG anti-GD1b was found to be positive, supporting the diagnosis (31).

Prognosis and complications

Extent and site of axonal injury are most important determinants of prognosis in acute motor axonal neuropathy. The patients with injury confined to distal motor axons can recover rapidly, whereas patients with significant disease burden in spinal roots or proximal nerve trunks are more likely to recover slowly and be left with residual muscle weakness. Serial electrodiagnostic testing is most commonly used to assess the site and extent of axonal injury. MRI of the muscle may be a useful technique for assessing the extent of motor axonal injury and resultant clinical course. Good concordance between clinical, electrophysiological, and MRI findings has been reported (20). Likewise, ultrasound of cervical nerve roots and peripheral nerves was shown to help distinguish axonal (AMAN, AMSAN) from demyelinating (AIDP) Guillain-Barré syndrome variants (115; 98). MRI and nerve ultrasound may, thus, be useful adjunctive techniques in situations where extensive EMG examination is not feasible, as in small children.

Patients with acute motor axonal neuropathy have shown a more rapid progression and an early nadir compared to those with acute inflammatory demyelinating polyneuropathy in a series of 131 patients with Guillain-Barré syndrome (63). Tekgul and colleagues found the children with acute motor axonal neuropathy or acute motor-sensory axonal neuropathy initially recovered slowly, but there was no difference at 12 months compared with acute inflammatory demyelinating polyneuropathy (167). Sung and colleagues performed a retrospective study to predict the functional outcome in patients with the axonal Guillain-Barré syndrome that were admitted in their university hospital between 2003 and 2014 (161). They defined a good outcome as being “able to walk independently at one month after onset” and a poor outcome as being “unable to walk independently at one month after onset.” Seventy-five percent of the patients with acute motor axonal neuropathy in the study were unable to walk independently at one month after admission. According to them a lower Guillain-Barré syndrome disability score at admission, relatively high amplitude of median, ulnar, deep peroneal, and posterior tibial CMAPs, and high amplitude of median, ulnar, and superficial peroneal SNAPs were associated with being able to walk at one month in patients with axonal Guillain-Barré syndrome (161). In another study, Zhang and colleagues reviewed their patients with Guillain-Barré syndrome between 2006 and 2013 in southwest China and reported that acute motor axonal neuropathy group had poorer prognosis compared to the acute inflammatory demyelinating polyradiculoneuropathy group at three months and six months after disease onset (203). Furthermore, a Chinese study reported that in patients with acute motor axonal neuropathy, elevated blood and CSF glucose levels were additive predictors of a poor short-term outcome compared to the AIDP patients with normal glucose levels (47). An Indian study also looked at the outcomes after Guillain-Barré syndrome in a pediatric population and found that children with acute motor axonal neuropathy had more residual muscle weakness and incomplete recovery compared to children with acute inflammatory demyelinating polyradiculoneuropathy (78). A Bangladeshi study examining the prognosis in children with Guillain-Barré syndrome found that at three months after onset of weakness, complete recovery or recovery with minor deficits was significantly higher in patients with demyelinating Guillain-Barré syndrome compared to patients with axonal Guillain-Barré syndrome. Further, cranial nerve palsy and severe muscle weakness were the significant risk factors of poor outcome in children with Guillain-Barré syndrome (57).

Glial fibrillary acidic protein is a protein expressed in the cytoskeleton of mature astrocytes and nonmyelin-forming Schwann cells. Notturno and colleagues found higher levels of glial fibrillary acidic protein in all patients with Guillain-Barré syndrome, with higher levels noted in patients with acute motor axonal neuropathy. Higher levels corresponded to worsening functional status at six months suggesting glial fibrillary acidic protein as a possible diagnostic marker for acute motor axonal neuropathy and a predictive tool for recovery (125). The utility of this biomarker to predict prognosis needs to be validated in larger clinical studies.

Serum neurofilament light chain (NfL) levels are emerging as a biomarker of axonal damage in many diseases of the CNS and PNS. A study examining serum and CSF NfL levels found that patients with Guillain-Barré syndrome and axonal or mixed axonal-demyelinating pathology showed significantly lower CSF/serum NfL ratios indicative of a peripheral nerve origin of NfL (83). Role of serum NfL levels as a prognostic biomarker in Guillain-Barré syndrome has been examined (104). It was found that baseline serum NfL levels are increased in patients with Guillain-Barré syndrome, are associated with disease severity and axonal variants, and have an independent prognostic value. In another study, Altmann and colleagues found that serum NfL levels measured at hospital admission correlated with clinical outcome in patients with Guillain-Barré syndrome (03). The patients with higher baseline serum NfL levels were more likely to have longer hospitalization time or need for intensive care.

NfL levels are not specific to axonal damage in the PNS but are also elevated with neuronal and/or axonal injury to the CNS. Therefore, two studies proposed novel biomarkers for the specific detection of peripheral nerve damage in immune neuropathies, including acute motor axonal neuropathy. Keddie and colleagues reported that elevated levels of peripherin, an intermediate filament almost exclusively expressed in the axonal cytoskeleton of PNS neurons, was more specific than NfL in distinguishing peripheral nerve axonal damage from CNS-associated axonal injury, and serum peripherin levels distinguished Guillain-Barré syndrome from CIDP and CNS diseases with good diagnostic accuracy. In Guillain-Barré syndrome, serum peripherin levels mostly peaked monophasically during the acute phase of Guillain-Barré syndrome and dropped thereafter, but the Guillain-Barré syndrome cohort size was discussed to be too small to yet assess its utility as a biomarker for routine clinical application (81). Compared to NfL levels, peripherin levels were able to distinguish Guillain-Barré syndrome from chronic inflammatory demyelinating polyneuropathy, with even increased diagnostic accuracy when combining both analytes (77% sensitivity, 89% specificity). Furthermore, unlike NfL, peripherin serum levels were age-independent. Although only five acute motor axonal neuropathy patients were included in this study, peripherin also appears to be a promising biomarker to facilitate the differentiation of axonal from demyelinating Guillain-Barré syndrome variants, but future larger studies are warranted to prove this assertion (81).

In another study, CSF levels of sphingomyelin, a sphingolipid expressed in the myelin sheath, distinguished chronic inflammatory demyelinating polyneuropathy and acute inflammatory demyelinating polyneuropathy from other neurologic diseases and predicted outcome (30). A trend towards higher sphingomyelin levels in acute inflammatory demyelinating polyneuropathy than in acute motor axonal neuropathy patients was seen, but due to a small acute motor axonal neuropathy cohort size (n=3), follow-up studies are required to assess the utility of CSF sphingomyelin levels to distinguish between axonal and demyelinating Guillain-Barré syndrome variants.

A study in Colombia involving children ages 1 to 15 years with Guillain-Barré syndrome and significant disease suggests those with cranial nerve involvement, quadriplegia, and need for ventilation support showed significant motor recovery delay. Muscle strength at day 10 was the most helpful prognosticator (130). They also found that patients with acute motor axonal neuropathy had longer recovery time than those with acute inflammatory demyelinating polyneuropathy. Another study of children in Seoul, Korea, found functional status at nadir was the best prognosticator compared to electrophysiologic studies, although both subtypes showed good outcomes (94).

Nishimoto and colleagues studied 51 children with Guillain-Barré in Japan and found that those with antiganglioside autoantibodies were more frequently diagnosed with acute motor axonal neuropathy (64% vs. 11%). Functional status on admission was similar between antibody-positive and antibody-negative children. A slower recovery was found with 78% of the antibody-negative group at follow-up, compared to 29% of antibody-positive patients, bringing the authors to suggest that antibody testing may be a helpful prognostic tool (124).

Long-term follow up on large cohorts of patients with acute motor axonal neuropathy is not available. Residual symptoms were noted up to six years after syndrome onset in a group of patients with Guillain-Barré syndrome questioned from the Dutch Guillain-Barré syndrome trial. Of 122 Dutch patients, 84 (69%) had no symptoms or only minor neurologic symptoms. Twenty-four patients were able to walk more than 10 meters without assistance. Fourteen were bedbound or only able to walk with support or a walker. Psychosocial status changes concerning work and daily living activity were seen in 63% of patients. Thirty-eight percent had a change in work status, with many able to resume working (22). Based on these studies, up to 10% die from complications of Guillain-Barré syndrome, largely from sepsis, respiratory failure, pulmonary embolism, and autonomic dysfunction; another 10% suffer from disabling weakness or balance abnormalities (106; 144).

In addition to long-term poor conditioning and loss of strength, other complications in the more acute and subacute phases include dysautonomia, infection, respiratory dysfunction, deep venous thrombosis, pulmonary embolism, contractures, peroneal nerve compression palsies, hypercalcemia and heterotopic calcification from immobility, decubitus ulcers, anemia, psychological abnormalities, and poor nutrition (144; 110).

In fulminant Guillain-Barré syndrome cases, in which central and peripheral nervous systems are involved, the patients often have good and relatively rapid recovery of central nervous system function, whereas recovery of peripheral nervous system function is relatively delayed and often incomplete (112). A similar case of acute motor axonal neuropathy has been reported in which the patient developed rapidly progressive weakness and respiratory failure. Patient had absent brainstem and spinal cord reflexes resembling brain death. Acute motor axonal neuropathy was diagnosed in this patient by CSF analysis and nerve conduction velocity testing. Twenty-four-hour electroencephalogram (EEG) showed that the patient had normal brain function. Transcranial Doppler ultrasound showed appropriate blood flow to the brain. Guillain-Barré syndrome rarely presents with severe weakness and brainstem encephalitis, which can mimic brain death. Awareness of the potential clinical manifestations and long-term prognosis of fulminant Guillain-Barré syndrome is crucial for both acute management of these patients, as well as education and counseling provided to family members and other health care providers (139).

Clinical vignette

A 67-year-old man with a previous history of diabetes and residual right hemiparesis from stroke presented to the emergency department after a subacute onset of increasing right hemiparesis. Over one and a half days, he developed quadriplegia, facial diplegia, dysphagia, dysarthria with tachycardia, and fluctuations in blood pressure. Reflexes were increased on the right and normal on the left. Four days later, he required mechanical ventilation. An MRI of his brain did not show evidence of a new stroke. Cerebral spinal fluid was normal. Laboratory tests showed negative anti-GM1, GD1b, GQ1b, and acetylcholine receptor antibodies, with negative HIV, West Nile, cytomegalovirus, Lyme, and mycoplasma. Initial nerve conduction studies on day four showed low compound muscle action potential amplitudes with absent F-waves in the lower extremities and partial conduction block of the ulnar nerve across the right elbow. Repetitive nerve stimulation at 3 and 30 Hz did not reveal decremental or incremental responses. Electromyography showed no evidence of increased insertional or spontaneous activity. He was diagnosed with Guillain-Barré syndrome and treated with plasmapheresis with no evidence of improvement over several weeks. Nerve conduction studies at one month showed further reduced compound muscle action potential amplitudes with absent F-waves in the upper extremities and no change in conduction velocities. Repeat electromyography showed acute denervation with fibrillation and positive sharp waves in all tested muscles. Intravenous immunoglobulin was initiated over the next two months.

Biological basis

Etiology and pathogenesis

Clinical and experimental data support that acute motor axonal neuropathy is an antibody-mediated nerve disorder. Like other variants of Guillain-Barré syndrome, antecedent triggering infections are not uncommon. Acute motor axonal neuropathy is strongly associated with preceding Campylobacter jejuni infection, particularly in Chinese, Japanese, and Bangladeshi populations (72). The bulk of clinical and experimental evidence support the idea that target epitopes are cell surface glycans called gangliosides, which are the constituents of the axolemma of the motor fibers. Autoantibodies against gangliosides arise due to molecular mimicry, and this hypothesis is supported by the following observations:

Acute motor axonal neuropathy pathology and immunopathology is distinct from AIDP and provides clues to the pathogenesis and pathophysiology of this disorder.

The earliest pathological changes in acute motor axonal neuropathy are subtle and affect the nodes of Ranvier of motor fibers in the ventral roots (52). These changes consist of lengthening of the nodal gap. Recruitment of macrophages to the nodes occurs early on. At the immunopathological level, it is noted that early in the pathogenetic process IgG binds at the nodes of Ranvier, leading to activation of complement suggested by C3d deposition, which causes macrophage recruitment (55). These macrophages then insert their processes into the nodal gap and open the periaxonal space, which is normally impermeable, to endoneurial constituents including antibody and complement. Immunopathological analysis at this stage shows deposition of IgG and complement activation marker C3d in periaxonal space and on the internodal axolemma in late cases (55). Macrophages are then recruited to the periaxonal internodal space after the disruption of paranodal sites of the Schwann cell myelin sheath attachment to the axon. This leads to axonal shrinkage and separation away from Schwann cell plasmalemma. The axon survives for some time before undergoing Wallerian-like degeneration (52). It is important to emphasize that not all patients develop axonal degeneration, and patients with rapid recovery likely develop pathological changes restricted to the nodes of Ranvier. The motor nerve terminal is another site of injury in acute motor axonal neuropathy, and this part of the motor axon is susceptible to antibody-mediated injury because it lies outside of the blood-nerve barrier (64). Nerve injury restricted to this site provides another potential rationale to explain rapid recovery as the degenerated axon must grow a short distance to reconnect with target muscle fibers and restore function (64). Overall, these pathological features indicate antibody and complement-dependent injury that is restricted to motor nodes or motor nerve terminals in patients with rapid recovery. A variable extent of axonal degeneration is seen in patients with prolonged disease course.

Antiganglioside antibodies are found in the serum of a proportion of patients with Guillain-Barré syndrome. These antibodies are polyclonal, predominantly IgG, and generally complement-fixing IgG1 and IgG3 (128; 66). Clinico-serological correlations between Guillain-Barré syndrome variants and anti-ganglioside antibodies indicate that anti-GM1 or -GD1a can be detected in 50% to 60% of patients with acute motor axonal neuropathy in the Far East (200; 140; 73; 66; 127). Antibodies to related minor gangliosides GalNAc-GD1a and GM1b are found in motor-predominant Guillain-Barré syndrome in about 10% to 15% of cases (191; 189). Identification of antibodies to complex gangliosides has led to a resurgence of interest in new antibody specificities (85), and studies on a large set of patients with acute motor axonal neuropathy are awaited. It has been proposed that antibody binding to a ganglioside can be enhanced or entirely abolished by the close association of a second ganglioside. Antibodies to ganglioside complexes may explain the lack of antibodies found in sera when only single ganglioside activities were present. Overall, clinical studies implicate GM1 and GD1a as the major target antigens in acute motor axonal neuropathy.

Gangliosides, the target antigens of anti-ganglioside antibodies, are sialic-containing glycosphingolipids that are widely distributed in the mammalian tissues but are particularly enriched in the nervous system. The ceramide portion of gangliosides anchors them into the plasma membrane, whereas oligosaccharide moieties extend into the extracellular space from the cell surface making them accessible to antibodies in the environment. Gangliosides are classified based on the number and linkage of the sugar backbone and attached sialic acids. There are many ganglioside species but GM1, GD1b, GD1a, and GT1b are most abundant in peripheral nerves. Acute motor axonal neuropathy associated gangliosides GM1 and GD1a are localized at the nodes of Ranvier and in motor nerve terminals (154; 48). Furthermore, preferential staining of motor nerve fibers has been demonstrated with monoclonal anti-GD1a antibodies in rodents (32) and also with human anti-GD1a antibodies from a patient with acute motor axonal neuropathy (38). A large number of biochemical studies indicate that there are no consistent differences in ganglioside content of motor and sensory fibers to explain preferential motor nerve fiber injury in acute motor axonal neuropathy. Although GD1a is present in similar amounts in motor and sensory fibers, a quantitative difference in staining of these fibers was demonstrated by Gong and colleagues with preferential motor axonal staining. By increasing antibody concentration, however, some staining was noted in sensory fibers. Similar differences in staining were not detected with GM1. Gong and colleagues suggest antibody accessibility to motor and sensory nerves, and variation in susceptibility to injury may contribute to distinctions between acute motor axonal neuropathy and acute motor-sensory axonal neuropathy (48). Lopez and colleagues developed a specific GD1a monoclonal antibody preferentially staining motor versus sensory nerves, suggesting structurally different conformation of GD1a in motor and sensory fibers (101). Acute motor axonal neuropathy and acute motor-sensory axonal neuropathy share a common immunological profile, and likely both are part of a spectrum of immune-mediated forms of attack on nerve axons (52; 192).

Antecedent events like infections are common in acute motor axonal neuropathy with Campylobacter jejuni gastroenteritis most frequently associated with acute motor axonal neuropathy. Haemophilus influenzae, mycoplasma infection (59; 163; 137), H1N1 influenza infection (87), dengue infection (159), Cikungunya infection (93), Japanese encephalitis virus (138), Salmonella infection (61), Zika infection (34), and Staphylococcus aureus endocarditis (33) have also been reported to precede acute motor axonal neuropathy.

The coronavirus (SARS-CoV-2) 2019 epidemic declared in China soon spread to the rest of the world (80). SARS-CoV-2 is another viral infection that can associate with acute motor axonal neuropathy (44) and other atypical Guillain-Barré syndrome variants (165).

The autoimmune inflammatory pathogenic mechanisms in COVID-19-related acute motor axonal neuropathy are not clear (151). The management of this disease in patients with COVID-19 does not differ from those testing negative for the SARS-CoV-2 virus, although respiratory distress and intensive care admissions are more frequent (133). Mortality has not been shown to be increased compared to control groups (71). Optimal treatment for COVID-19-related Guillain-Barré syndrome remains to be determined (01). Interestingly, an epidemiological study did not find phenotypic clues of SARS-CoV-2 being causative of Guillain-Barré syndrome (80).

Wagner and Bromberg reported a case of a 46-year-old male with previously unrecognized HIV infection who presented with acute motor axonal neuropathy (179). Dardis presented a case of 22-year-old woman with congenitally acquired HIV who had prolonged CD4 depletion due to noncompliance with highly active antiretroviral therapy (HAART) (36). She developed features of acute motor axonal neuropathy following an episode of pyelonephritis. The patient’s CD4 count was less than 10/µl, which is the lowest CD4 count reported for a patient with HIV and an axonal neuropathy. Different studies have reported cases of HIV patients presenting with AIDP features or developing AIDP as a complication of HIV infection, but only four cases of acute motor axonal neuropathy have been reported in HIV patients to date. This study provides evidence of acute motor axonal neuropathy in a patient with congenital HIV infection and prolonged noncompliance with antiretroviral treatment. The precipitating event in this case does not seem to be solely the change in CD4 count because varying levels of CD4 count have been noticed at the time of diagnosis. Macrophage activation is thought to play a role in the pathogenesis in this situation. Macrophages serve as reservoir for HIV and are auto-activated by HIV, which causes secretion of interferon gamma. Widespread systemic mobilization of macrophages is known as “macrophage activation syndrome” (MAS). The author hypothesized that acute motor axonal neuropathy and AIDP could be organ specific variants of macrophage activation syndrome (36).

An unusual case of acute motor axonal neuropathy has been reported following septic shock due to Acinetobacter baumannii (168). The patient was a 25-year-old man who was admitted to the ICU because of rapidly developing weakness of all four limbs following an episode of fever, sore throat, and dry cough for 10 days before hospitalization. Bronchoalveolar lavage of patient revealed multidrug-resistant A baumannii. Patient had marked flaccid tetraplegia, facial diplegia, absence of tendon reflexes, and no sensory disturbances. NCS examination showed absence of CMAP from all four limbs with the exception of low-amplitude response from the right peroneal nerve, suggesting motor nerve axonal impairment. Because the lipopolysaccharide of A baumannii has structure that is similar to that of C jejuni, it was hypothesized that the infection by A baumannii in this patient may have had a pathogenic role in the development of the acute motor axonal neuropathy via a mechanism of molecular mimicry (168).

Cao-Lormeau and colleagues published the first study on the Zika virus epidemic in Latin America between October 2013 and April 2014, which is the largest epidemic ever recorded (28). On February 1, 2016, WHO declared the suspected link between Zika virus and neurologic disorders. In their case-control study, 42 patients were diagnosed with Guillain-Barré syndrome at the Center Hospitalier de Polynésie Franҫaise (Tahiti, French Polynesia) during the outbreak period. Watrin and colleagues reported clinical and electrophysiological characteristics of this series (180). According to them males predominated with a sex ratio of 2.82 and a mean age of 46 years. All patients except two were native Polynesian. Fifty-five percent of the patients were able to walk unaided at admission (38% at nadir), 24% of patients had swallowing troubles (45% at nadir), 74% had motor weakness of the limbs, and deep tendon reflexes were diminished or not found in the vast majority of patients. Mean duration of progressive and of the plateau phase was 7 and 9 days, respectively. Thirty-eight percent of the patients were admitted to intensive care units and 10 patients underwent tracheotomy. Nerve electrophysiological studies at admission showed marked distal motor conduction alterations that had almost completely disappeared at the fourth month, which was suggestive of an acute motor axonal neuropathy subtype of Guillain-Barré syndrome. Lumbar puncture of 90% of these patients showed elevated proteins, with cell count always less than 50/µl (180). Forty-one patients with Guillain-Barré syndrome had anti-Zika virus IgM or IgG, and all (100%) of the patients had neutralizing antibodies against Zika virus compared with 56% of control group patients. Patients’ Zika and dengue virus status was confirmed with virological investigation including RT-PCR for Zika virus, and both microsphere immunofluorescent and seroneutralization assays for Zika and dengue virus. Antiglycolipid reactivity was assessed in patients with Guillain-Barré syndrome using both ELISA and combinatorial microarrays. Antiglycolipid antibody activity was found in 31% of patients and predominantly against GA1 in 19% of patients by ELISA and 46% of the patients by glycoarray at admission. The typical acute motor axonal neuropathy-associated antiganglioside antibodies were rarely present. History of dengue did not differ significantly between patients with Guillain-Barré syndrome and those in the control group (28).

Another case series of 19 patients with Guillain-Barré syndrome with recent history of Zika virus in Cύcuta, Colombia has been published (11). These patients developed neurologic symptoms at a median of 10 days after the onset of the viral symptoms. Albuminocytological dissociation was found in eight cases. Electrodiagnostic studies confirmed acute motor axonal neuropathy in all patients. Five patients met level 1, eight patients met level 2, and six patients met level 3 diagnostic certainty for Guillain-Barré syndrome in the Brighton classification. Fifteen required respiratory assistance, 16 received IVIG, and three had plasmapheresis. Seventy-nine percent of patients had a Hughes disability scale score of 4 to 5. This study further enforces the association between Guillain-Barré syndrome and Zika virus infection (11).

Hepatitis E infection is increasingly recognized as an antecedent event for Guillain-Barré syndrome (79; 100; 160). Acute hepatitis E infection is also reported as a triggering event in acute motor axonal neuropathy. It was reported that a young woman developed bilateral symmetric weakness involving lower extremities more than upper extremities with no sensory involvement after hepatitis E infection (02). She was found to have acute axonal injury by neurophysiological evaluation and recovered completely 35 days after onset.

Amongst these infectious agents Campylobacter jejuni and H influenzae carry ganglioside-like moieties (114). Campylobacter is discussed further because of its frequent association with acute motor axonal neuropathy and relevance to molecular mimicry hypothesis.

Campylobacter jejuni is a gram-negative rod, which is one of the most common causes of bacterial gastroenteritis world-wide (69; 45; 126). Infection with C jejuni is found in 13% to 72% of patients with acute motor axonal neuropathy or Guillain-Barré syndrome (69; 54), with an overall prevalence estimated around 30% (113). The lipopolysaccharide (LPS) of C jejuni carries ganglioside-like moieties. Several studies have characterized these in Guillain-Barré syndrome- and diarrhea-associated C jejuni strains. GM1-, GD1a-, GalNAc-GD1a-, GM1b-, GT1a-, GD2-, GD3-, and GM2-like structures have all been identified (15; 13; 14; 197; 198; 196; 156; 119; 118). It is these structures that are likely to provide the initial stimulus to autoimmune activation and induction of anti-ganglioside antibodies in patients with post-Campylobacter acute motor axonal neuropathy.

Campylobacter jejuni has been estimated to affect more than 1% of the population per year worldwide, but Guillain-Barré syndrome occurs in approximately 1.5 per 100,000. This suggests that fewer than 0.01% of Campylobacter jejuni cases are associated with Guillain-Barré syndrome (156; 184), raising the hypothesis of host susceptibility. The host properties that confer this susceptibility to Guillain-Barré syndrome after Campylobacter infection remain unknown. Some studies indicate that post-Campylobacter Guillain-Barré syndrome cases preferentially associate with specific HLA alleles (194; 142), but the significance of these findings remains unclear due to lack of confirmatory studies. In the context of host susceptibility, Zhang and colleagues investigated tumor necrosis factor-alpha promoter polymorphism in patients with Guillain-Barré syndrome. Alleles associated with higher levels of TNF alpha were more common in patients with acute motor axonal neuropathy but not with acute inflammatory demyelinating polyneuropathy (206). Another study demonstrated that the presence of specific polymorphisms in the promotor region of the TNF-alpha gene may predict susceptibility to axonal types of Guillain-Barré syndrome (135). Specifically, TNF-alpha 308G/A polymorphism has significant association with acute motor axonal neuropathy and acute motor sensory axonal neuropathy susceptibility, but not with acute inflammatory demyelinating polyradiculoneuropathy (206; 135; 77; 74). Moreover, TNF-alpha 857C/T polymorphism is significantly associated with acute motor axonal neuropathy (135). Similarly, toll-like receptor-4 299Gly allele is present at a higher frequency in patients with the acute motor axonal neuropathy subtype of Guillain-Barré syndrome than in the acute inflammatory demyelinating polyneuropathy subtype (75). A study on 300 Guillain-Barré syndrome patients (60% AMAN/AMSAN) reported that mannose-binding lectin (MBL) 2 gene polymorphisms are not associated with Guillain-Barré syndrome susceptibility, but correlate with Guillain-Barré syndrome severity if the polymorphisms reduce serum MBL levels (75).

A large body of experimental work over past 20 years has shed light on the pathogenic role of antiganglioside antibodies. Studies have focused on pathogenic effects of these antibodies on nodes of Ranvier and motor nerve terminals, sites that are affected in patients with acute motor axonal neuropathy. Antibodies bind to nerves at nodes of Ranvier where gangliosides and sodium channels are concentrated. Early studies with patient sera or purified immunoglobulins showed that anti-GM1 antibodies blocked or altered channel function at the nodes (166; 181) though results were not always reproducible (131). Studies with experimental antibodies demonstrate that antibodies bind to nodal and paranodal structures and activate complement, which apparently disrupts cell adhesion molecules necessary for the formation and maintenance of nodal structure and function (164). This disruption of nodal structure correlates with failure of conduction across affected nodes. Willison and colleagues in a series of studies show that antiganglioside (including anti-GD1a) antibodies bind to nerve terminals and cause complement-dependent decreased quantal acetylcholine (ACh) release, resulting in neuromuscular blockade and a calcium-dependent degeneration of the motor nerve terminal (183; 50; 134; 129; 49). The clinical, pathological, and electrophysiologic effects in this model are almost entirely abrogated by the application of an inhibitor of the C5 component of complement (eculizumab), which prevents formation of the membrane attack complex (56). In related series of patch-clamp studies IgG, GQ1b, GD1a, GD1b, and GM1 antibodies have been shown to induce reversible complement independent pre- and post-synaptic blockade depending on the antibody specificity (27; 26; 25). Overall, these studies show that anti-ganglioside antibodies disrupt the nodal or motor nerve terminal function or structure, which could underlie the clinical motor dysfunction seen in patients with acute motor axonal neuropathy. Complement-fixing anti-GM1 ganglioside autoantibodies that target axonal nerve sheaths and Schwann cell membranes are found in both axonal and demyelinating variants. A study by McGonigal and colleagues indicates that different pathomechanisms operate in antiganglioside antibody-mediated primary/direct axonal injury compared to the secondary axonal degeneration following antiganglioside antibody-mediated Schwann cell injury (107). Antiganglioside antibody-mediated primary axonal injury was severe and acute, whereas anti-ganglioside antibody-mediated secondary axonal injury developed subacutely following Schwann cell injury. This bystander axon degeneration early on induced axonal disruption at the nodes of Ranvier and later progressed to involve internodal axon segments.

Development of animal models has allowed fulfillment of Koch-Witebsky postulates supporting that acute motor axonal neuropathy is an autoimmune disorder mediated by autoantibodies (145; 155). Earlier efforts to generate either IgG anti-ganglioside antibodies or neuropathy in animals by active immunization with C jejuni were not successful (120; 86; 97; 136). Active immunization in rodents with C. jejuni lipopolysaccharides generates mainly low titer IgM antibodies (185; 50) due to high tolerance to self-gangliosides (24). Tolerance to self-gangliosides can be circumvented by immunization with gangliosides or C jejuni lipopolysaccharides in immune naive transgenic animals lacking complex gangliosides (102; 24). This model suggests that production of pathogenic antibodies in acute motor axonal neuropathy reflect breakdown of tolerance. Monoclonal anti-ganglioside antibodies generated in these transgenic mice when implanted as an antibody-secreting hybridoma in mice led to the development of axonal neuropathy. The development of a neuropathy was contingent on breakdown of the blood nerve barrier by cytokines generated by the antibody-secreting implanted tumor. This suggests anon-T-Cell-dependent mechanism of breakdown of the blood-nerve barrier (158).

In seminal studies, Yuki and colleagues induced acute motor axonal neuropathy in rabbits by sensitization to GM1 or GM1-containing lipopolysaccharides from Campylobacter jejuni (199; 195; 162). Although all immunized animals produced anti-GM1 antibodies, only a proportion of animals developed clinical disease. Further studies showed that activation of complement and leukocytes only occurred from sera obtained from diseased rabbits, raising the possibility that high affinity of the antibodies is necessary to result in disease (176). In this rabbit model, sodium channel clusters at the nodes of Ranvier were disrupted. IgG was deposited on GM1 at perinodal axolemma first and then extended towards the internodes, with activation of complement and MAC formation. Lengthened nodes showed disruption and loss of sodium channels, with potassium channel cluster alteration only at advanced stages. Schwann cell microvilli, thought to stabilize sodium channels, were also disrupted. Macrophage invasion occurred during the recovery phase, suggesting possible clearance of damaged nerve fibers (164). These animal studies recapitulate vital features of human acute motor axonal neuropathy pathology.

Besides complement, other innate immune effectors like activating FcyRs are involved in inflammatory nerve injury in axonal Guillain-Barré syndrome (153). Studies implicate the role of Fc gamma receptors and macrophages in the pathogenesis of antiganglioside antibody-mediated nerve injury. Zhang and colleagues examined the role of Fc gamma receptors and macrophages in an animal model of Guillain-Barré syndrome. Well-characterized antibody passive transfer sciatic nerve crush and transplant models were used to study the antiganglioside antibody-mediated inhibition of axon regeneration in wild-type, mutant, and transgenic mice with altered expression of specific Fc gamma receptors and macrophage/microglia populations. They performed behavioral testing, electrophysiological testing, morphometry, immunocytochemistry, quantitative real-time PCR, and Western blotting to assess outcome. They concluded that the presence of autoantibodies, directed against neuronal/axonal cell surface gangliosides, in the injured mammalian peripheral nerves switch the proregenerative inflammatory environment to growth inhibitory setting by engaging specific activating Fc gamma receptors on recruited monocyte-derived macrophages to inhibit axon regeneration (202). In a complimentary study He and colleagues examined the role of Fc gamma receptors and macrophages on nodal and axonal injury in a new animal model (58). In this series of studies they found that multiple factors affect antiganglioside antibody-mediated injury to intact nerve fibers including antibody fine specificity, antigen-binding affinity, antigen density in the target nerve fibers, antibody affinity to Fc gamma receptors, breakdown of blood nerve barrier, and extent of inflammation (58).

A study by Zhang and colleagues showed a possible role of soluble receptor for advanced glycation end products (sRAGE) and high mobility group box 1 (HMGB1) in the pathogenesis of Guillain-Barré syndrome (201). They measured serum sRAGE, HMGB1, IL-6, and TNF-α levels in 86 patients with Guillain-Barré syndrome and analyzed associations between sRAGE or HMGB1 and clinical variables in these patients. Similarly, they determined CSF, sRAGE, and HMGB1 levels in a cross-sectional study of 50 patients with Guillain-Barré syndrome who had matched serum samples. The patients with acute motor axonal neuropathy subtype of Guillain-Barré syndrome had significantly lower levels of serum sRAGE compared to healthy controls, whereas serum HMGB1, IL-6, and TNF-α levels in all subtypes of Guillain-Barré syndrome were significantly higher than those in healthy controls. They also observed increased sRAGE levels and decreased HMGB1 levels after treatment. Lower sRAGE and higher HMGB1 levels may be related to the robust autoimmune response that underlies Guillain-Barré syndrome, perhaps by increasing the release of inflammatory cytokines. Serum sRAGE can be used clinically as a biomarker for disease severity in acute motor axonal neuropathy, whereas recombinant sRAGE and anti-HMGB1 interventions seem to be promising emerging therapies for treating Guillain-Barré syndrome (201). A study examined the roles of IL-36 cytokines in Guillain-Barré syndrome (207). Serum IL-36α, β, γ, and interleukin-36 receptor antagonist (IL-36Ra) levels of patients with Guillain-Barré syndrome in the acute and remission phases were measured and compared to healthy volunteers. Significantly higher serum IL-36α and IL-36γ levels were found in the acute phase than in the remission phase. Correlation analyses showed that serum IL-36α and IL-36γ levels in patients with Guillain-Barré syndrome were positively correlated with other proinflammatory factors such as serum IL-17 and TNF-α levels, whereas serum IL-36Ra levels were negatively correlated with IL-17 and TNF-α. Serum and CSF levels of these proinflammatory markers were concordant. Further, the serum and CSF levels of IL-36α and IL-36γ in the axonal variants of Guillain-Barré syndrome were higher than in the demyelinating subtypes.

Epidemiology

The incidence of acute motor axonal neuropathy is essentially unknown given the rarity of the disease and lack of consistent electrodiagnostic studies or strict diagnostic criteria. In studies attempting to differentiate acute inflammatory demyelinating polyneuropathy and acute motor axonal neuropathy, the frequency varies among countries. Acute motor axonal neuropathy was found in 65% to 76% of patients with Guillain-Barré syndrome in Northern China (65; 64), 38% in Japan (127), 38% in a large study in Mexico (117), about 15% in Brazil (40), 3% to 20% in Europe and North America (141; 178; 53), and 22% in a 6-year study in Israel that found an additional 15% to have acute motor sensory axonal neuropathy (84).

Differential diagnosis

Associated or underlying disorders

The differential diagnosis of acute ascending paralysis includes acute myelopathy such as from spinal cord compression, infarction, infection, or transverse myelitis. The presence of bowel or bladder dysfunction or a sensory level favors a spinal cord abnormality over a peripheral neuropathy.

A neuromuscular junction abnormality such as myasthenia gravis or botulism should be considered. Although some forms of Guillain-Barré syndrome such as Miller-Fisher syndrome present with ophthalmoparesis, acute motor axonal neuropathy rarely has ocular findings. Botulism generally presents with dilated, unreactive pupils and constipation. Electrodiagnostic studies with repetitive stimulation can be helpful.

West Nile virus infection may produce weakness including an axonal polyneuropathy or a poliomyelitis-type illness on electromyography and nerve conduction studies (121). These patients should present with a febrile illness along with mental status changes at time of weakness, although encephalopathy is not always evident. Poliomyelitis is a consideration in geographical locales where polio is still not eradicated (96), but most patients have a gastroenteritis and fever with acute onset of weakness that tends to be asymmetrical. Lyme disease should have a history of tick bite and erythema migrans. Other infections such as diphtheria may present with acute paralysis, and tick paralysis should improve with removal of the parasite. Paralytic rabies can also mimic axonal Guillain-Barré syndrome (157).

Other considerations should include toxin exposure such as heavy metals and organophosphates that can have a similar presentation. Acute porphyria should have a predominately motor axonopathy, but other symptoms such as abdominal complaints, psychiatric disturbances, and seizures are common (146). Interestingly, a case of undiagnosed acute intermittent porphyria was reported, which was misdiagnosed as recurrent abdominal tuberculosis and was prescribed antituberculous therapy (17). This treatment led to the development of progressive extremity weakness with evidence of severe motor axonal neuropathy.

Studies indicate that patients with limbic encephalitis and positive anti-NMDA and AMPA receptor antibodies can develop acute neuropathic injury. In this context, acute motor axonal neuropathy has been described in a patient with anti-AMPA receptor encephalitis (182).

Diagnostic workup

Electrodiagnostic examination is required to distinguish acute motor axonal neuropathy from other forms of Guillain-Barré syndrome. Ho and colleagues proposed the electrodiagnostic criteria (Table 2) (65).

Diagnosis of acute inflammatory demyelinating polyneuropathy must have one of the following in two or more nerves during the first two weeks of illness:

Diagnosis of acute motor axonal neuropathy:

These criteria have been widely accepted. Hadden and colleagues further modified the classification to include a ratio of the proximal compound muscle action potential amplitude to the distal compound muscle action potential amplitude of less than 0.5 to be further evidence of demyelination (53). However, acute motor axonal neuropathy may show early partial motor conduction block with no further evidence of demyelination or later remyelination (29). The principal abnormalities of acute motor axonal neuropathy are reduced distal compound muscle action potential amplitude and absent F-wave responses (108). When there is electrodiagnostic evidence of acute motor axonal neuropathy and amplitudes of sensory nerve action potentials are below the lower limit of normal, the diagnosis of acute motor-sensory axonal neuropathy should be considered (193).

Table 2. Electrophysiological Separation of Guillain-Barré syndrome Variants

If initial studies are normal and the diagnosis is in question, the nerve conduction studies should be repeated in a few days. Some advocate mandatory serial electrophysiological studies to eliminate diagnostic uncertainty, even if initial studies appear conclusive (170). Uncini and colleagues studied 55 patients with Guillain-Barré syndrome subtypes who had serial electrodiagnostic recordings. The diagnosis was changed in 24% of patients after the follow-up electrodiagnostic study, and the main shift was from equivocal or demyelinating electrophysiology to an axonal subtype of Guillain-Barré syndrome. This shift was related to the reversible conduction failure and the length-dependent compound muscle action potential amplitude reduction (171). In another case report a patient had motor conduction blocks in all peripheral nerves in electrophysiological studies and was diagnosed as having acute inflammatory demyelinating polyradiculoneuropathy. Later, reduction of CMAP amplitudes in posterior tibial nerve, absence of CMAPs in median, ulnar, and peroneal nerves, and loss of motor conduction blocks were found on repeat electrophysiological studies. According to these findings, patient’s diagnosis was changed to acute motor axonal neuropathy. The authors suggest that motor conduction blocks may appear in the early stage of acute motor axonal neuropathy and they disappear on serial studies, hence it is better to repeat electrodiagnostic studies in patients with Guillain-Barré syndrome (187). The difficulty in distinguishing axonal from demyelinating Guillain-Barré syndrome variants within the first four days after disease onset was also pointed out by another study (21). This study stressed the need for serial electrophysiological assessments to delineate acute motor axonal neuropathy from AIDP.

Derksen and colleagues performed a study to evaluate electrodiagnostic abnormalities in patients with Guillain-Barré syndrome and disorders that mimic Guillain-Barré syndrome (39). This study included 38 patients that mimicked Guillain-Barré syndrome at the time of presentation but in fact were confirmed to have other neuropathic illnesses on follow up. The electrodiagnostic testing of these 38 patients was compared with 73 patients with confirmed Guillain-Barré syndrome. This comparison showed that the presence of “spared” sural SNAP with abnormal ulnar SNAP on nerve conduction studies was the most specific finding for demyelinating Guillain-Barré syndrome (39).

Another study by Umapathi and colleagues examined the presence of sural sensory nerve action potential (SNAP) in various subtypes of Guillain-Barré syndrome. They defined sural-sparing as a greater decrease in the median and/or ulnar SNAP compared to decrease in sural SNAP. According to their study sural-sparing pattern was present in both axonal as well as demyelinating subtypes of Guillain-Barré syndrome. Calculating the percentage change in median, ulnar SNAP and comparing it with the change in sural SNAP may enable the electrodiagnostician to detect subclinical sural-sparing and increase the yield of nerve conduction studies in the early stages of Guillain-Barré syndrome. They concluded their study with the hypothesis that if abnormal sural SNAP with normal upper limb SNAP is observed in the initial nerve conduction studies of a patient with suspected Guillain-Barré syndrome then the electrodiagnostician should question the diagnosis, regardless of Guillain-Barré syndrome subtype (169). One study used CMAP scan to differentiate between axonal and demyelinating forms of Guillain-Barré syndrome (41). Plotting the CMAP amplitudes against a range of stimulus intensities results in a dose-response curve that defines the CMAP scan. Patients with acute motor axonal neuropathy and AIDP have pronounced differences in CMAP scan stimulus intensity parameters at the onset of disease, which allows for differentiation between these Guillain-Barré syndrome subtypes.

Cauda Equina Conduction Time (CECT) is another technique investigated to differentiate demyelinating and axonal types of Guillain-Barré syndrome. Matsumoto and colleagues performed a study to compare CECT in nine demyelinating and seven axonal Guillain-Barré syndrome patients. They used Magnetic Augmented Translumbosacral Stimulation (MATS) to activate nerves at both the proximal and distal sites of the cauda equina for the measurement of CECT. According to them CECT was prolonged in all patients with demyelinating Guillain-Barré syndrome who had leg symptoms, but in all patients, the motor conduction velocity at the peripheral nerve trunk was normal. In all the patients with axonal Guillain-Barré syndrome with leg symptoms, motor conduction velocity and CECT were in normal range. So, they concluded that cauda equina is more frequently involved than the peripheral nerve trunk in demyelinating Guillain-Barré syndrome (105).

If a neuromuscular junction abnormality such as botulism or myasthenia gravis is suspected to be the cause of acute flaccid paralysis then repetitive stimulation should be performed.

Ultrasound of cervical nerve roots and peripheral nerves is another diagnostic tool to diagnose Guillain-Barré syndrome and distinguish between AMAN/AMSAN and AIDP: AMAN/AMSAN patients revealed milder cervical root enlargement than AIDP patients, and vagal nerve enlargement correlated with autonomic dysfunction in a Chinese study (98). The 2023 EAN/PNS guidelines on diagnosis and treatment of Guillain-Barré syndrome recommend nerve ultrasound (and/or nerve/plexus MRI) as additional diagnostic tools in atypical Guillain-Barré syndrome cases (174).

Cerebral spinal fluid can be evaluated to help with the diagnosis of Guillain-Barré syndrome. After the first week of symptoms, cerebral spinal fluid protein has been seen to be elevated in most patients. Typically cerebral spinal fluid white cells are less than 10, but counts up to 50 cells per mm3 have been recorded in Guillain-Barré syndrome (12). CSF profile in acute motor axonal neuropathy is expected to conform to the preceding observations for Guillain-Barré syndrome in general. A study correlated the CSF total protein levels with different subtypes of Guillain-Barré syndrome (23). This study found significantly higher CSF total protein levels in Guillain-Barré syndrome cases classified as demyelinating compared to axonal subtypes.

The presence of serum antibodies to GM1, GM1b, GD1a, GD1b, GQ1b, and GalNAc-GD1a can help with the diagnosis, although their absence does not exclude acute motor axonal neuropathy; However, their presence does not necessarily confirm acute motor axonal neuropathy. Stool cultures or serological studies for Campylobacter jejuni should be considered in those with a recent gastroenteritis, especially because the HS2, HS4c, HS19, HS23/26c, HS41, and HS1/44c Campylobacter jejuni capsular subtypes are associated with acute motor axonal neuropathy (60).

Koga and colleagues compared serum IgM and IgG antibodies against isolated gangliosides and ganglioside complexes in patients with C jejuni enteritis with and without subsequent neurologic complications. This study indicates that antiganglioside IgM antibodies can be detected in C jejuni enteritis without complication of Guillain-Barré syndrome, and that the detection of antiganglioside IgM antibodies does not always support a diagnosis of Guillain-Barré syndrome (82).

If clinically indicated, urine porphyrins, heavy metals, Lyme titers, West Nile, autoimmune serologies, botulism serologies, and imaging of the spinal cord should be considered to help exclude other causes of acute paralysis.

Management

Despite the availability of two specific immunotherapies, namely, IVIG and plasma exchange, the mainstay of management in Guillain-Barré syndrome or acute motor axonal neuropathy remains the provision of supportive care (including admission to ICU) during the acute phase to prevent complications and facilitate recovery. There are no controlled studies of immunomodulatory therapy in the primary axonal variants of Guillain-Barré syndrome, but anecdotal experience indicates that both plasma exchange and IVIG are beneficial. To date, there is no difference in immunomodulatory treatments in acute motor axonal neuropathy and other subtypes of Guillain-Barré syndrome (09; 64; 53), which was recently affirmed by the 2023 EAN/PNS guidelines on diagnosis and treatment of Gullain-Barré syndrome (174). Due to the ease of its administration and broader patient acceptability, IVIG is used more frequently for the treatment of Guillain-Barré syndrome. The 2023 EAN/PNS guidelines strongly recommend that patients with Guillain-Barré syndrome or acute motor axonal neuropathy be referred to specialized centers with an ICU experienced in managing patients with acute flaccid paralysis. Patients with a rapidly progressing disease, bulbar involvement, and low muscle strength, especially in proximal muscles, are at risk for mechanical ventilation and should be early admitted to an ICU for surveillance (174). This is important as without optimal supportive and/or intensive care this group of disorders still carries significant risk of mortality, and simplicity of IVIG treatment occasionally prevents appropriate referral to centers experienced in managing Guillain-Barré syndrome.

Plasmapheresis improves disability and speeds recovery in Guillain-Barré syndrome compared to supportive treatment alone (04; 05). When using plasmapheresis, the French Cooperative Group suggests treating patients with mild Guillain-Barré syndrome with two exchanges and moderate to severe forms with four exchanges. There was no additional benefit in receiving six versus four exchanges in severe cases (08). The EAN/PNS guidelines recommend two plasma exchange cycles early after disease onset (less than 2 weeks) for mildly affected patients that are able to walk unaided and four to five plasma exchange cycles in severely affected patients (174). In some centers, immunoadsorption (IA) is performed instead of plasma exchange due to purported better tolerability, but randomized controlled trials on immunoadsorption in Guillain-Barré syndrome are currently lacking (103).

The dose of intravenous immunoglobulin is generally set at 2 grams per kilogram and is generally divided into five doses of 400 milligrams per kilogram per dose. The infusion rate should not exceed 200 milliliters per hour or 0.08 milliliters per kilogram per minute. Dalakas suggested giving 1 gram per kilogram for two consecutive days and stated he had not found more adverse side effects with a 2-day infusion (35), but the EAN/PNS guidelines favored the five-days standard therapy over the two-days course, due to a lower probability of treatment-related fluctuations (174). In a small experiment using the rabbit model of acute motor axonal neuropathy, six rabbits were given five days of rabbit gamma-globulin at 400 milligrams per kilogram per day at 3-week intervals at onset of inoculation with bovine gangliosides. All six rabbits developed acute motor axonal neuropathy, although the gamma-globulin treated group developed symptoms at a much slower rate than controls, and there was less axonal degeneration at the anterior roots (123).

No statistical significance was noted when plasmapheresis was compared to intravenous immunoglobulin (173) or when plasmapheresis was followed by intravenous immunoglobulin (09); therefore, the 2023 EAN/PNS guidelines stated equality of IVIG and plasma exchange to treat Guillain-Barré syndrome (174). Most large Guillain-Barré syndrome trials include patients with moderate to severe disease. Given the expense of plasmapheresis and intravenous immunoglobulin treatment, electing not to treat those with mild disease is certainly understandable and in many centers is the protocol. Because the French Cooperative Group found a faster recovery rate in treating those with mild disease, we and the 2023 EAN/PNS guidelines recommend treating all cases of acute motor axonal neuropathy with either intravenous immunoglobulin or plasmapheresis within two weeks of onset or earlier.

Kuwabara and colleagues suggest intravenous immunoglobulin may be superior to plasmapheresis in cases of acute motor axonal neuropathy with positive anti-GM1 autoantibodies. The patients treated with IVIG had significantly lower Hughes grade scores 1, 3, and 6 months after onset and a higher probability to regain independent locomotion at six months. Rapid recovery was more frequent, and delayed recovery was less frequent in the IVIG subgroup (89). The authors postulate that if autoantibodies are pathogenic, intravenous immunoglobulin can displace antibodies bound to motor nerves, possibly preventing complement activation (205). Plasmapheresis would only remove free-circulating antibodies (188).

Hou and colleagues investigated the changes in lymphocyte subsets in patients with Guillain-Barré syndrome treated with IVIG. According to their study, the percentage of CD4+ CD45RO+ T cells was significantly higher, and the percentage of CD4+ CD45RA+ T cells was significantly lower in the AIDP patients than in the healthy control group. After treatment with IVIG the ratio of CD4+/CD8 T cells and the percentage of CD4+CD45RA+ T cells increased and the percentages of CD8+ T cells and CD4+ CD45RO+ T cells decreased significantly, along with the number of CD19+ B cells in the AIDP group of patients. These changes were not obviously notable in the acute motor axonal neuropathy group, although IVIG suppressed immune reactions to a certain degree and prevented aggravation of the clinical condition. The Hughes disability score was significantly lower both in AIDP and acute motor axonal neuropathy groups after therapy with IVIG, and the change in score was not significantly different between the AIDP and acute motor axonal neuropathy groups. These findings hint that there might be other changes in immune function in acute motor axonal neuropathy, and further studies are needed to understand the mechanism of IVIG efficacy in patients with acute motor axonal neuropathy (67).

There is variation in reported relapse rate after immunomodulatory therapy, but overall treatment-related fluctuations occur in about 5% to 10% of patients, irrespective of treatment modality. Retreatment with the original modality is the recommended approach in patients who have relapsed after initial improvement. Whether a second course of immunomodulatory therapy is indicated for those who show no initial improvement remains an unresolved issue. A trial examining the second course of IVIG in such patients is underway at this time. The observational International Second IVIg Dose (ISID) study did not find better outcomes after a second IVIg course in Guillain-Barré syndrome patients who failed initial IVIg treatment (177), which is why the 2023 EAN//PNS guidelines strongly recommend against a second course of IVIg in treatment-refractory patients because it is associated with minimal benefit and increased serious adverse events (174).

Steroids, whether administered orally or intravenously, have been found to be ineffective in treating Guillain-Barré syndrome either orally (68; 06) and are not recommended in the acute stages of disease. However, 500 milligrams of intravenous methylprednisolone given for five days along with five days of intravenous immunoglobulin was found to be more effective than intravenous immunoglobulin alone in an open study of 25 patients (07). A randomized, double-blind, placebo-controlled study with 233 individuals with Guillain-Barré syndrome compared therapy with intravenous immunoglobulin to intravenous immunoglobulin with 500 milligrams of intravenous methylprednisolone. There was no statistical difference in improvement between the two groups. However, a trend towards decreased number of days until independent walking was noted in the steroid-treated group (average 28 days vs. 56 days). The authors stated further investigation using intravenous immunoglobulin with methylprednisolone would be warranted (175). Nevertheless, the 2023 EAN/PNS guidelines recommended against a combined use of IVIG and methylprednisolone due to lack of efficacy of this combinational therapy (174).

In addition to intravenous immunoglobulin or plasmapheresis, supportive therapy should be instituted immediately. Patients with oropharyngeal weakness with an inability to protect their airway or a vital capacity of less than 15 milliliters per kilogram should be considered for elective endotracheal intubation. These patients as well as those with autonomic instability should also be monitored in the intensive care unit. Prevention of infection such as pneumonia or urinary tract infection should be utilized. Deep venous thrombosis and pulmonary embolism prophylaxis, in addition to adequate nutrition possibly by a nasogastric tube or gastrostomy should be addressed (144). Prevention of contractures and decubitus ulcers should be instituted early in immobilized patients with passive range-of-motion exercises and frequent repositioning. Rehabilitative measures such as physiotherapy should be started during the acute phase of the disease and continued afterwards because late recovery can occur (174). Strenuous exercise may cause paradoxical weakness, so therapy should be aimed at improving overall function and introducing strengthening exercises slowly (19). Orthotics should be used for optimal positioning and strength. In addition, complications from long-term illness, such as hypercalcemia of immobilization and anemia, should be monitored (110).

Emerging treatments for patients with Guillain-Barré syndrome include eculizumab, which inhibits complement activation and prevents the formation of membrane attack complex. This treatment has shown to completely inhibit clinical disease in mice (56). Further studies in humans are being done to answer whether eculizumab given together with intravenous immunoglobulin benefits patients with Guillain-Barré syndrome. Eculizumab has been used as an adjunctive therapy to intravenous immunoglobulin in a small randomized, double-blind, placebo-controlled trial of eight patients with Guillain-Barré syndrome (37). This treatment was found to be safe to use in patients with Guillain-Barré syndrome. This small study led to a phase 2 trial in Japan as an add-on therapy with intravenous immunoglobulin (NCT02493725). This treatment was found to be relatively safe, and significantly more patients were able to run at six months in the eculizumab-treated group compared to the placebo group, which was a secondary outcome measure (111), but as primary outcome measures were not met, the 2023 EAN/PNS guideline currently recommends against the use of eculizumab in Guillain-Barré syndrome unless larger trials show more promising results (174). Another complement inhibitor, ie, humanized antibody against the C1q component of complement (92), is also being developed as an add-on therapy with intravenous immunoglobulin for the treatment of Guillain-Barré syndrome (NCT04035135). An additional treatment approach with relevance to Guillain-Barré syndrome pathogenesis is antagonism of neonatal Fc receptor (FcRn). FcRn is central to IgG homeostasis and catabolism. FcRn antagonism shortens the half-life of IgG including circulating pathogenic IgG antineural autoantibodies. FcRn antagonism can reduce antibody-mediated inflammatory nerve injury in experimental models of Guillain-Barré syndrome (204). A number of FcRn inhibitors are in clinical development for different indications, including chronic inflammatory demyelinating polyradiculoneuropathy (NCT04281472), and although the study’s results are not yet published, the latest study update reported that a significant proportion of chronic inflammatory demyelinating polyradiculoneuropathy patients responded to therapeutic FcRn inhibition (10). Given the ample clinicopathologic evidence of autoantibody-mediated nerve injury in Guillain-Barré syndrome, particularly its axonal variants, the use of FcRn inhibitors may extend to this group of disorders.

Outcomes

There are two patterns of recovery in patients with acute motor axonal neuropathy. A rapid recovery after plasmapheresis or intravenous immunoglobulin has been observed in some patients. This may be related to reversible immune-mediated conduction failure at the nodes of Ranvier in motor fibers. A poor recovery suggests Wallerian-like degeneration in motor axons (64; 88). Even so, Hiraga and colleagues reported a patient not walking six months after onset, yet walking independently at 57 months (62). Kuwabara and colleagues found in their study that preservation of deep tendon reflexes and Haemophilus influenzae infection are indicators of a good prognosis (90).

Special considerations

Pregnancy

There is no known published information of acute motor axonal neuropathy in pregnancy. According to national registries in Sweden, Guillain-Barré syndrome risk is lower during pregnancy and increases postpartum (76). There were a few case reports of intravenous immunoglobulin treatment for Guillain-Barré syndrome during 23 to 33 weeks of gestation. Both women responded to treatment and delivered healthy infants (150). In a study Sharma and colleagues performed a retrospective observational analysis to find correlation between pregnancy and Guillain-Barré syndrome. They estimated the incidence of Guillain-Barré syndrome in pregnancy in their cohort between 1.2 and 1.9 cases per 100,000. Their study indicated that the risk of Guillain-Barré syndrome increases in the third trimester and in the first two weeks after delivery. They analyzed 47 patients with pregnancy and Guillain-Barré syndrome and suggested that early diagnosis and prompt intensive supportive care in these cases can improve the prognosis for both the mother and fetus (152). Improvement after plasmapheresis has also been reported (70), but no known clinical trials comparing intravenous immunoglobulin and plasmapheresis exist in pregnancy.

Anesthesia

Specific case reports concerning anesthesia and acute motor axonal neuropathy are unknown to these authors. Perel and colleagues give several recommendations if anesthesia is needed for patients with Guillain-Barré syndrome. Because of decreased sympathetic activity from autonomic instability, they suggest continuous monitoring of the electrocardiogram, blood pressure, and central venous pressure even for minor procedures. Rapid changes in upright position should be avoided, and even minor changes in position should be performed carefully. Medications such as barbiturates and phenothiazines should not be administered, given previous reports of circulatory collapse. Caution should be used with positive pressure ventilation because significant peripheral pooling without reflex venous constriction may occur. They also recommend using sympathomimetic agents if anesthetic agents are needed. Even when low-spinal or epidural analgesia is needed, volume expansion with intravenous fluids should be used and hypotension treated promptly (132). Feldman reported a specific case of cardiac arrest after succinylcholine administration for a Cesarean section in a patient one month after recovery from Guillain-Barré syndrome. Arterial blood showed severe hyperkalemia. Because cholinergic receptors proliferate at extraneuromuscular junction sites after neurologic injury, neuromuscular blockade can increase serum potassium when there is a large amount of muscle involvement (43).