In the present study, eye movement abnormalities such as square-wave jerks, abnormal cogwheeling during smooth pursuit, and saccade hypometria were observed in ALS patients. Square-wave jerks and abnormal cogwheeling during smooth pursuit were more frequently observed in ALS patients, especially in ALS patients with bulbar involvement. There were no significant differences in the investigated eye movement parameters between bulbar-onset and spinal-onset patients.
The square-wave jerk rate was markedly higher in ALS patients than in controls. Previous studies have reached a similar conclusion[6, 14], but we subsequently conducted subgroup analyses and found that square-wave jerks were more common in ALS patients with bulbar involvement than in ALS patients without bulbar involvement. The genesis of square-wave jerks remains uncertain, and our present findings are supported by the “brainstem hypothesis” of square-wave jerk generation[33]. This theory, based on the model of Otero-Millan et al.[34], assumes the presence of a disturbance in the brainstem neural circuits. These circuits are mainly formed by excitatory burst neurons (EBNs), inhibitory burst neurons (IBNs), omnipause neurons (OPNs) and their connections with the superior colliculus (SC). Furthermore, electrophysiological and pharmacological inactivation studies in monkeys have suggested that the rostral SC is important in sustaining steady fixation[35, 36]. When ocular fixation is destabilized—that is, the normal suppression of saccades by the rostral SC is disrupted—OPNs allow EBNs and IBNs to fire and may cause square-wave jerks. ALS patients with bulbar involvement are likely to have more extensive brainstem pathology and are therefore more likely to have damage to the brainstem ocular motor network. With respect to the present findings, we speculate that the increased occurrence rate of square-wave jerks may be due to impairment of the brainstem.
There was no such decline in smooth pursuit velocity gain among the ALS patients and those with bulbar involvement. These findings are similar to those reported by Gizzi et al.[10] and Shaunak et al.[14]; however, several studies have found deceased smooth pursuit velocity gain in ALS patients compared to controls [7, 8, 12, 13, 21]. Additionally, we found that abnormal cogwheeling during smooth pursuit was evident more often in ALS patients, especially in ALS patients with bulbar involvement. These abnormal pursuits were also identified by Jacobs et al.[6] and Kang et al.[20]. Smooth pursuit is controlled by cerebro-ponto-cerebellar pathways[26]. The middle temporal (MT)/medial superior temporal (MST) region is important for processing moving stimuli and projects to the frontal eye field (FEF) and supplementary eye field (SEF). These, in turn, project to the nucleus reticularis tegmenti pontis (NRTP), cerebellar dorsal vermis, and fastigial nucleus and are important for smooth pursuit initiation. The MT/MST projects to the dorsolateral pontine nucleus (DLPN), flocculus/paraflocculus, nucleus prepositus hypoglossi and vestibular nucleus and is important for sustaining smooth pursuit. The present findings provided no evidence of cerebellar abnormalities; Jacobs et al. [6] describe one ALS patient with severe bulbar disability who had normal smooth pursuits 1 week before death, and postmortem histopathological examination revealed a loss of neurons and reactive gliosis in the hypoglossal nucleus[6]. This would allow impairments of upper brainstem or supratentorial cerebral lesions, explaining the demonstrated abnormal cogwheeling during smooth pursuit.
Abnormalities of reflexive saccades have been reported, including prolonged saccadic latency[12], reduced saccadic velocity[7, 13, 37], and saccadic dysmetria[8, 20]. However, some studies have also shown no abnormalities of reflexive saccades[10, 14]. Our results do not entirely coincide with those of previous studies; the only abnormality recorded in reflexive saccades was saccade hypometria, but no statistically significant differences were found between the ALS patients and controls, as well as in subgroup analyses. The neural substrate of reflexive saccades comprises both cortical and subcortical components, with the SC playing a vital role[26]. Neurons in the rostral SC are related to visual fixation, and those in the caudal SC are associated with saccades. This functional differentiation suggests that different neural pathways should exist between the ocular fixation and reflexive saccade systems. Given the heterogeneity of ALS, it is necessary to increase the number of patients to finally determine the pattern of eye movement abnormalities in ALS.
Donaghy et al.[17] and Kang et al.[20] reported that some kinds of eye movement abnormalities were more common in ALS patients with bulbar onset than spinal onset. They attributed these findings to extensive brainstem pathology in bulbar-onset disease. However, no significant differences were observed between the bulbar-onset and spinal-onset groups in the investigated eye movement parameters among our ALS patients. A possible reason for the discrepancy between our results and those observed in the two previous studies could be the fact that when spinal-onset ALS patients undergo oculomotor evaluation, they may also show bulbar disabilities, while the analyses of the previous studies did not describe bulbar impairment among spinal-onset ALS patients. The pathological process underlying ALS is histologically characterised by aggregates of the pTDP-43, therefore, the results in the present study suggest that the region of involvement could better indicate the pathophysiological essence of eye movement abnormalities than the type of onset pattern in ALS and provide a more rational explanation for the neural pathways involved and intrinsic pathophysiological mechanism of the disease. With increased understanding of sensitive molecular pathological markers, future neuropathological studies have the potential to identify the neural substrate of these eye movement findings in ALS.
Furthermore, some diseases with clinical manifestations similar to ALS, such as cervical spondylotic amyotrophy (CSA) and multifocal motor neuropathy (MMN), have not been reported to cause eye movement abnormalities. Therefore, oculomotor evaluations might have particular value in helping identify patients who were misdiagnosed.
In addition, this study has some limitations that deserve mention. Our videonystagmography equipment cannot provide the frequency or amplitude of the square-wave jerks. The lack of quantitative characterization likely introduced variability into the data, thus decreasing their statistical power. The associations between these eye movement abnormalities and the severities of functional limitation in ALS patients are not clear and need to be verified through further longitudinal studies.