Oral phase dysphagia in facial onset sensory and motor neuronopathy

Abstract Introduction Facial onset motor and sensory neuronopathy (FOSMN) is a rare disease whose cardinal features are initial asymmetrical facial sensory deficits followed by bulbar symptoms and spreading of sensory and motor deficits from face to scalp, neck, upper trunk, and upper extremities in a rostral–caudal direction. Although bulbar involvement is frequently observed in FOSMN, dysphagia in these patients has not been fully described. In this study, we aimed to characterize dysphagia as a prognostic factor in FOSMN by investigating our institutional case series. Methods We retrospectively reviewed the medical records, including swallowing function tests, of six patients with FOSMN (three men and three women) who were thoroughly examined at Kyushu University Hospital between 1 January 2005 and 30 November 2017. Results Average age at onset was 58.5 years; average disease duration was 5.7 years. All patients developed bulbar dysfunction and dysphagia (at an average of 1.8 and 2.6 years from onset, respectively), resulting in choking episodes in three patients, percutaneous endoscopic gastrostomy placement in three, and recurrent aspiration pneumonia in one. Four of five patients evaluated with videofluoroscopic swallowing studies had poor oral retention, leading to bolus flowing into the pharynx before swallowing; the fifth patient showed poor lingual transfer. Fiberoptic endoscopic evaluation of swallowing revealed leakage of blue‐dyed water from the mouth to the pharynx in three patients because of poor oral retention, but only mild pharyngeal phase dysphagia in all four cases evaluated. Conclusions Oral phase dysphagia predominates in the early stage of FOSMN.

In this study, we aimed to clarify clinical and laboratory features in this series of patients with FOSMN. We used videofluoroscopic swallowing studies (VFSS) and fiberoptic endoscopic evaluation of swallowing (FEES) to evaluate oral phase dysphagia as a critical prognostic factor in FOSMN.

| Patients and standard protocol approvals
We conducted a retrospective survey of FOSMN cases evaluated in the Department of Neurology at Kyushu University Hospital, Japan, between 1 January 2005 and 30 November 2017. The diagnosis of FOSMN was largely based on the clinical features reported by Vucic et al. (2006). However, there is one report of a patient with FOSMN who suffered from dysarthria and dysphagia at onset followed by sensory symptoms in the face (Broad & Leigh, 2015). Therefore, in this study, the diagnostic and inclusion criteria of FOSMN were as follows: Patients had either (1) slowly progressive sensorimotor symptoms in the face beginning with sensory abnormality or (2) bulbar symptom at onset followed by sensory and motor symptoms in the face, and (3) (Hyodo's score), which is widely accepted and used by Japanese otolaryngology specialists because it is useful, simple, and clinically validated (Hyodo, Nishikubo, & Hirose, 2010).

| Evaluation of swallowing
In brief, this scoring method was designed to be simple and reliable in daily clinical practice. It consists of the following four parameters: (A) degree of salivary pooling at the vallecular and piriform sinuses, (B) glottal closure reflex induced by touching the epiglottis or arytenoid with the tip of an endoscope, (C) swallowing reflex initiation assessed according to "white-out" timing, and (D) pharyngeal clearance after swallowing of blue-dyed water. Each parameter was scored on a scale from 0 to 3, where 0 was normal, 1 was mild impairment, 2 was moderate impairment, and 3 was severe impairment. The total score (0 to 12) was used as an index of swallowing function. In this scoring system, a score greater than 7 indicates a serious risk for aspiration (Hyodo et al., 2010). VFSS and FEES findings were also scored according to the Penetration-Aspiration Scale (PAS), which is a widely used 8-point clinical scale for rating penetration and aspiration (Colodny, 2002;Rosenbek, Robbins, Roecker, Coyle, & Wood, 1996).

| Case reports
We identified six patients with FOSMN (three women and three men) during the study period. Table 1 summarizes their clinical and demographic features. Average age at onset was 58.5 years (range 44-83 years); average disease duration was 5.7 years (range 2-12 years). All six patients developed sensory impairment in the face, including the oral cavity; sensory impairment spread in a rostral-caudal direction to the scalp, neck, upper trunk, and upper extremities. We briefly describe the clinical course, neurological findings, and results of laboratory tests, imaging, and neurophysiological examinations below.

| Patients 1 and 2
Patients 1 and 2 were previously reported (Hokonohara et al., 2008;Sonoda et al., 2013). It is worth noting that Patient 1 developed progressive dysphagia and died by choking 12 years after the onset of disease.

| Patient 3
History A 60-year-old man noticed numbness in his face and oral cavity, which slowly progressed and spread to the scalp, neck, and shoulders. Two years after onset, he developed difficulty swallowing and slurring of his speech. Three years after onset, he was referred to our hospital.

Neurological findings
The patient had decreased sensation to pinprick and light touch in all divisions of the trigeminal nerve bilaterally and in the dermatomes of C1-C4. Corneal reflexes were absent. He had weakness and atrophy of the right facial muscles and dysarthria with tongue weakness.
Gag reflex was also absent. Chorea-like involuntary movement was detected in the extremities with right-sided predominance; however, muscle strength and reflexes of the extremities were normal.

Laboratory findings
The patient's blood and cerebrospinal fluid tests were normal.
MRI of the brain and spinal cord were also normal. A blink reflex study showed delayed ipsilateral R2 (iR2) and contralateral R2 (cR2) responses bilaterally (latency of iR2 was 50.7 ms, and cR2 was not evoked on left-sided stimulation; latencies of iR2 and cR2 on right-sided stimulation were 49.9 and 42.4 ms, respectively; the upper normal limit of R2 latency is <41 ms). A nerve conduction study (NCS) showed decreased sensory nerve action potential amplitudes in the upper limbs. Needle electromyography revealed chronic denervation changes in the sternocleidomastoid muscle.
Dopamine transporter SPECT revealed reduced 123 I-ioflupane binding in the bilateral striata, with left-sided predominance (see Figure 1).

| Patient 4
History A 74-year-old woman developed slurred speech; she noticed numbness in her left lip and oral cavity 10 months later. The numbness spread to the right lip over several months. She also noticed difficulty moving her tongue and weakness in her right arm.

Neurological findings
Neurological examination 14 months after onset showed decreased sensation to light touch and pinprick in the bilateral trigeminal distributions. The patient also had dysarthria and mild weakness in her face, neck, and right arm, with tongue fasciculations. Light touch, pinprick, vibration, and proprioception of her limbs were all intact.
Reflexes, gait, and coordination were normal.

Laboratory findings
The patient's laboratory workup was normal except for the presence of monoclonal protein (IgAλ), hepatitis B surface antigen antibody, and hepatitis C virus antibody. A bilateral blink reflex study showed prolonged iR2 and cR2 responses (45.7 and 46.0 ms, respectively, on left-sided stimulation; 41.8 and 41.3 ms, respectively, on right-sided stimulation).

Diagnosis and treatment
With a diagnosis of FOSMN, the patient received three infusions of IVIg, which relieved her facial numbness completely and improved the delayed iR2 and cR2 responses to left-sided stimulation in the blink reflex test (36.6 and 34.5 ms, respectively; see Figure 2).
However, the patient's right arm weakness spread to the right leg and her bulbar symptoms worsened. Because she experienced worsening of dysphagia and frequent choking on food, water, and saliva despite modification of food and fluid consistency, resulting in about 5 kg of weight loss over a 6-month period, a PEG tube was placed 2 years after initial symptoms.

| Patient 5
History An 83-year-old woman first noticed numbness around the right corner of her mouth, which spread to all areas of the three branches of the bilateral trigeminal nerves. Carbamazepine was prescribed for a presumed diagnosis of trigeminal neuralgia; however, the treatment was not effective. The patient then developed progressive dysarthria and salivation, with difficulty propelling food boluses into the pharynx.

Neurological findings
Neurological examination 2 years after onset showed paresthesia in

Diagnosis and treatment
The patient was initially treated with levothyroxine, which normal- ized TSH and f-T4 levels but did not relieve her symptoms. She was then diagnosed with FOSMN. IVIg therapy was initiated and mildly improved her speech and right-sided dysesthesia. However, treatment did not improve the patient's dysphagia.

| Patient 6
History A 44-year-old man was initially told by others that he could not enunciate well; however, he was not concerned about this change.
Two years after onset, he felt numbness around his mouth and had slurring of his speech. Numbness spread to his cheeks, chin, and bilateral fingers within 2 years. Four years after onset, the patient suffered from loss of ability to retain saliva and water within the mouth.
He was referred to our hospital 5 years after onset. The patient also had an episode of choking on noodles and being transported to a hospital by ambulance before he underwent detailed examination at our hospital.

Neurological findings
Neurological examination showed decreased sensation to pinprick and light touch in all divisions of the trigeminal nerves bilaterally and dysesthesia in the bilateral maxillary and mandibular nerve areas and fingers. Corneal reflexes were absent. The patient had mild weakness and atrophy of the right masseter, facial, and arm muscles. He also had dysarthria, dysphagia, and weakness and atrophy of the tongue with fasciculations. Gag reflex was absent. Reflexes of the extremities were diminished. Gait and coordination were normal.

Laboratory findings
The patient's general laboratory workup was normal. An NCS revealed reduced sensory nerve action potential amplitudes in the bilateral median and ulnar nerves. Needle electromyography showed chronic denervation and reinnervation findings in the trapezius muscles. Left blink reflex studies showed delayed cR2 response (latency 42.9 msec), with normal responses to right-sided stimulation.

Diagnosis and treatment
Based on his clinical history and findings, the patient was diagnosed with FOSMN. He was followed up but chose to receive no treatment.  retention was observed in three (Patients 2, 3, and 6); this leakage was not counted in Hyodo's scoring (Hyodo et al., 2010). In PAS scoring of FEES, neither penetration nor aspiration was detected in any patient (Patients 3-6). These findings suggest that patients with FOSMN had dysphagia that was more predominant in the oral phase than the pharyngeal phase, especially in the early course of the disease. The findings also indicate that the risk of aspiration in patients with FOSMN can be underestimated if only the easy-touse scoring systems of FEES are used.

| D ISCUSS I ON
The Consistent with a previous report (Vucic et al., 2006), all our patients developed sensory impairment in the face, spreading in a rostral-caudal direction. Bulbar involvement preceded sensory disturbance in Patients 4 and 6, which is consistent with the report by Broad and Leigh (Broad & Leigh, 2015). We thus consider all our cases to fit the category of FOSMN. All patients developed bulbar dysfunction with dysphagia early in the course of the disease, which resulted in choking, PEG insertion, recurrent aspiration pneumonia, and even death. Therefore, we consider bulbar involvement one of the most life-threatening factors in FOSMN, along with respiratory failure, as seen in Patient 2. According to a survey of the literature that does not include the present cases, nine patients died an average of 3.4 years after symptom onset; the identified causes of death were aspiration pneumonia and/or respiratory insufficiency (Barca et al., 2013;Broad & Leigh, 2015;Fluchere et al., 2011;Truini et al., 2015;Vucic et al., 2006;Ziso et al., 2015).
We judged bulbar involvement in our patients to be severe, according to their complaints and clinical signs such as weight loss, prolongation of mealtime, and decreased food consumption. However, the swallowing findings according to Hyodo's FEES scoring in our patients were unexpectedly mild. This discrepancy can be explained by the fact that Hyodo's score focuses on the pharyngeal phase of swallowing, whereas patients with FOSMN predominantly have difficulty in the oral phase. Hyodo's total scores are useful for evaluating the ability of oral food intake: subjects with total scores below 4 or 5 have satisfactory oral feeding ability, whereas those with total scores above 9 or 10 have difficulty with oral feeding (Hyodo et al., 2010). Despite these previous findings, the Hyodo's scores of patients with FOSMN who needed PEG insertion or had episodes of choking were between 2 and 4, which should have been adequate for oral feeding. We did not TA B L E 2 Evaluation of swallowing in six patients with FOSMN Note. L, latest evaluation; NA, not applicable; NE, not examined; NM, not mentioned; P, evaluated just before insertion of percutaneous endoscopic gastrostomy; PAS, Penetration-Aspiration Scale.
a Total scores and subscores (in brackets) are based on Hyodo's score (Hyodo et al., 2010), described in the Section 2. b Patients 1 and 2 were evaluated before the use of Hyodo's endoscopic scoring system. use the Langmore score (Langmore, 2001) to evaluate oral phase dysphagia in the present study because the enormous quantity of data it nucleus (Vucic et al., 2006(Vucic et al., , 2012Ziso et al., 2015). These pathological changes of the pontine and medullary cranial nerve nuclei may well explain impairment of oral retention and dysphagia in patients with FOSMN. Therefore, poor oral retention and premature material flow resulting in oral phase dysphagia appear to be characteristic findings in FOSMN.
By contrast, patients with ALS usually develop not only oral but also pharyngeal phase dysphagia when they require modification of foods or tube feeding (Higo, Tayama, & Nito, 2004). Patients with spinal and bulbar muscular atrophy (SBMA), another type of motor neuron disease, also develop dysphagia in both oral and pharyngeal phases. In patients with SBMA, pharyngeal phase abnormalities are more frequently observed than oral phase abnormalities (Banno et al., 2017). Patients with Parkinson's disease, one of the most common neurodegenerative diseases, also frequently develop dysphagia. The most characteristic finding of dysphagia in Parkinson's disease is residues after swallowing; leakage from the oral cavity before swallowing is not frequently observed in these patients (Pflug et al., 2018). Therefore, the findings of dysphagia in patients with FOSMN seem to be relatively unique to this condition. According to our literature survey, bulbar involvement was present in at least 43 of 47 FOSMN cases (91%) (Barca et al., 2013;Broad & Leigh, 2015;Cruccu et al., 2014;Dalla Bella et al., 2014Dobrev et al., 2012;Fluchere et al., 2011;Hokonohara et al., 2008;Isoardo & Troni, 2008;Karakis et al., 2014;Knopp et al., 2013;Sonoda et al., 2013;Truini et al., 2015;Vucic et al., 2006Vucic et al., , 2012Ziso et al., 2015).
Based on the present findings, we propose that bulbar involvement manifesting with dysphagia could be a critical prognostic factor in FOSMN and that oral phase dysphagia should be carefully evaluated in patients with FOSMN.
We found discrepancy in PAS scores between VFSS and FEES  (Dobrev et al., 2012;Knopp et al., 2013;Sonoda et al., 2013;Vucic et al., 2006Vucic et al., , 2012 and partially respond to various immunotherapies (Fluchere et al., 2011;Hokonohara et al., 2008;Knopp et al., 2013;Sonoda et al., 2013). Together with the pathological observation of lymphocytic infiltration in peripheral nerves, cervical root ganglion, and ventral nerve root in one autopsy case (Sonoda et al., 2013), these findings suggest an inflammatory component to this condition. However, neuronal loss and gliosis in the nuclei of affected cranial nerves, including the trigeminal nucleus and tract, dorsal root ganglia, anterior horn cells, and anterior spinal roots with or without TDP-43-positive inclusions (Sonoda et al., 2013;Vucic et al., 2006Vucic et al., , 2012Ziso et al., 2015), indicate the essential neurodegenerative nature of this disease. Therefore, it should be emphasized that early intervention for dysphagia is important in managing patients with

FOSMN.
One of our patients had chorea-like involuntary movement from the onset. Although this involuntary movement could have been incidental, a patient from another institute in Japan was also reported to have chorea-like hyperkinesis (Taguchi et al., 2016). In our patient, haloperidol effectively controlled involuntary movements and 123 I-ioflupane binding was decreased in the striata, suggesting that presynaptic dopaminergic abnormality was the underlying mechanism of this chorea (Booth et al., 2015a,b). It is possible that chorea is a rare manifestation of FOSMN, in which multiple systems may undergo degeneration. Careful observation of larger numbers of patients with FOSMN is required to clarify this issue.
FOSMN is a developing disease without clearly established diagnostic criteria. We chose the inclusion criteria in this study on the basis of previously reported clinical features of patients with FOSMN ( Barca et al., 2013;Broad & Leigh, 2015;Cruccu et al., 2014;Dalla Bella et al., 2014Dobrev et al., 2012;Fluchere et al., 2011;Hokonohara et al., 2008;Isoardo & Troni, 2008;Karakis et al., 2014;Knopp et al., 2013;Sonoda et al., 2013;Truini et al., 2015;Vucic et al., 2006Vucic et al., , 2012Ziso et al., 2015). Moreover, we consecutively enrolled all patients who fulfilled the criteria in the present study. Therefore, we believe that there was no serious selection bias that could have distorted our conclusions. However, selection bias cannot be completely ruled out because this study was retrospective and patient enrollment was based on our own inclusion criteria.
In conclusion, bulbar involvement with oral phase dysphagia can be a life-threatening prognostic factor in patients with FOSMN; this symptom is manageable with early PEG intervention. Neurologists should be aware of this condition and pay particular attention to dysphagia.

ACK N OWLED G M ENTS
We thank Rebecca Tollefson, DVM, from Edanz Group (www. edanzediting.com/ac) for editing a draft of this manuscript.