The most common presenting symptoms of CMⅠ include cervical neck pain, occipital headache, movement dysfunction, paresthesia, muscle atrophy, or unsteady gait. Comparatively, symptoms of cranial neuropathies, such as hoarseness, choking, and swallowing difficulty can often be ignored by patients and even neurosurgeons. In previous study, we reviewed the medical records of 140 CMⅠ patients from January 2002 to September 2006 in our hospital. There were only 15 cases (10.7%) which reported having “ difficulty in swallowing” , or “ choking while drinking” , while 72 cases (51.4%) reported “ no difficulty in swallowing” . The remaining 53 cases (37.6%) did not clearly indicate presence or absence of dysphagia. In order to study changes of swallowing function in CMⅠ patients, we surveyed swallowing function of 126 CMⅠ patients in our hospital from January 2007 to July 2010^[21]. In the study, 34 cases (26.9%) clearly reported dysphagia of varying severity. Most patients admitted difficulty in swallowing or choking while drin-king after careful history taking. The average interval between the onset of symptoms and the diagnosis was (7.3± 1.8) years, while the onset of dysphagia and choking was only (2.5± 0.9) years. Dysphagia occurred more often in advanced CMⅠ , but was frequently overlooked by clinical practitioners.

Dysphagia may give rise to clinically relevant complications. Aspiration occurs in 50% of patients with oropharyngeal dysphagia, and is associated with a mortality of up to 50%^[22]. Additionally, new studies show that dysphagia risk factor for low respiratory tract infections (LRTIs) and community-acquired pneumonia (CAP) in the elderly.

We analyzed other associated symptoms’ relationship to dysphagia in CMⅠ patients. A higher percentage of CMⅠ patients with dysphagia (15/20)ex-perienced symptoms caused by posterior cranial nerve damage, such as decreased sweating or hypoesthesia on the affected side of the face, as compared with the control group (5/22, P=0.001). The ratios of those with ataxia and positive pyramidal signs in the dysphagia group (9/20, 9/20, respectively) were also higher than those in the non-dysphagia group (3/22, 3/22, respectively), and there were significant differences(P=0.025). These associations suggest a patho-physiological mechanism of dysphagia in CMⅠ .

Manometry is a standard method of characterizing oropharyngeal and esophageal motor function by providing measurements of pressures, peristalsis, and coordination. There was a significant difference in UES relax ratio measurement (75.3% vs. 63.1%, P=0.023). The UES normally prevents food reflux into the pharynx. Surprisingly, the UES relax ratio was higher in the dysphagia group, suggesting that excessive relaxation of UES might cause food refluxing and coughing. On the other hand, UES rest pressures in the dysphagia group [64.3(18.8-151.2)mmHg] were higher than those in the non-dysphagia group[55.4(11.6-166.2)mmHg], though not significantly dif-ferent(P=0.180). These findings suggest that dysphagia in CMⅠ patients may be a dysfunction in the neurological pathway of pharyngeal muscle movement. Unfortunately, other parameters, including UES relaxation time, cannot be measured using water-perfusion manometry, which is an important parameter in estimating UES relaxation function. Solid-state manometry (SSM)would better measure UES pressure in future studies.

From a neurological point of view, swallowing control is governed by three main elements: the afferent system, the brain stem swallowing center and the efferent system. Oropharyngeal afferents project to supramedular structures and to the brain stem swallowing center, in the medulla oblongata, to allow invo-luntary onset of the pharyngeal swallow response and modulate volitional swallowing. The afferent neurons of the oropharynx and larynx involved in swallowing are the maxillary branch of the trigeminal nerve (vcranial nerve), the pharyngeal branch of the glossopharyngeal nerve (GPNph, Ⅸ cranial nerve) and two branches of the vagus nerve (Ⅹ cranial nerve)— the pharyngeal branch and the superior laryngeal nerve (SLN). These afferents are formed in part by non-myelinated (type C) or thinly myelinated (type A) fibers, and are sensitive to mechanical stimuli (pressure, touch), temperature changes, and chemical stimuli (hydrogen ions, taste stimuli)^[23]. As dysphagia associated with CMⅠ is found to frequently be associated with symptoms caused by posterior cranial nerve damage and ataxia and pyramidal signs, this suggests that the underlying pathophysiology is comprised of a set of sympathetic and parasympathetic nerve fiber injuries.

In our study, we found that the location of syrin-xon MRI was highly associated with the presence of dysphagia symptoms, as the percentage of syringomyelia affecting the bulbar or upper cervical region was significantly higher in the dysphagia group 15/20 vs. 7/22, P=0.001). To our surprise, diameter and involved segments of syringomyelia did not show the same correlation. This may mean that injury of nerve fibers passing specifically through the bulbar and upper cervical region is important in the pathogenesis of dysphagia. Postoperative improvement of syringomyelia was found for all the 36 patients with syrinx which was consistent with mate analysis results made by Arnautovic et al^[24]. In that study, the authors reviewed 145 English-language reports of pediatric, adult, and combined (adult and pediatric) surgical series of patients with CMⅠ published from 1965 through August 31, 2013. They concluded that 78% of the CMⅠ patients who had syringomyelia improved or resolved postoperatively.

Consistent with our previous study^[24], dysphagia was improved (according to the patients’ self-evaluation) in half of the cases shortly after posterior fossa decompression with duraplasty. We believe that improvement of dysphagia may be a meaningful index to evaluate the effect of surgery in the early postoperative stage. According to our study, headache/neck pain, symptoms caused by posterior cranial nerve da-mage, and pyramidal signs may be associated with dysphagia. However, there were no differences in the outcomes of these symptoms between the two groups. In the study of Arnautovic et al^[24], most (80%) reported postoperative neurological outcomes as follows: 75% improved, 17% showed no change, and 9% experienced worsening. Postoperative headaches improved or resolved in 81% of the patients^[23]. We need further study to identify whether the dysphagia and other associated symptoms or signs outcomes correlate in these patients.

A limitation in this study is the small number of patients in each group. Evaluation of the total incidence, risk factors, assessment, intervention and post-operative change requires a systematic analysis of a larger sample size.

Neurogenic dysphagia is thought to be a relatively rare symptom in adult CMⅠ patients without atlantoa-xial dislocations. However, it may cause fatal complications. Dysphagia associated with CMⅠ was usually accompanied by symptoms caused by posterior cranial nerve damage, ataxia, and positive pyramidal signs. Location of the syringomyelia affecting specifically the bulbar or upper cervical region was associated with dysphagia in CMⅠ patients. The difference of UES rest pressure and relax ratio in dysphagia patients versus non-dysphagia patients suggests that the pathophysiology of dysphagia may be a dysfunction in the neurological pathway of pharyngeal muscle movement. Dysphagia etiology work-up should include CMⅠ in the dif-ferential diagnosis. We hope that this study highlights the importance of dysphagia in the CMⅠ adult population, and shows how it can be a meaningful index for improvement in symptoms after posterior fossa decompression. In the future, we hope to further characterize mechanisms of dysphagia in CMⅠ adult patients using SSM in a larger patient population.

(Edited by LIU Shu-ping)