Swallowing is an incredibly complex sensorimotor task involving much of the voluntary and autonomically controlled musculature of the head and neck, several cranial nerves, as well as a central nervous system pathway that has yet to be fully described. Surgical approaches to the cervical spine can interrupt or lead to dysfunction at several points in this chain and lead to the development of dysphagia. Additionally, instrumentation or intrinsic factors such as osteophytes can contribute to speech or swallowing disorders in these patients.

The action of swallowing has typically been divided into three functional phases: oral, pharyngeal, and esophageal (1). The oral phase begins when solid or liquid enters the mouth. The muscles of mastication (temporalis, masseter, medial, and lateral pterygoid) under the control of the mandibular division of the trigeminal nerve power the chewing motion to decrease particle size, while the intrinsic muscles of the tongue, innervated by the hypoglossal nerve, manipulate the bolus and move it toward the oropharynx (1). The voluntary phase of swallowing is initiated as the tongue presses the bolus against the hard palate, activating mechanoreceptors that initiate peristaltic movement of the tongue. As the bolus enters the oropharynx, passing under the faucial arch, involuntary swallowing is triggered by free nerve endings of the glossopharyngeal and vagal nerves (1).

The pharyngeal phase begins as the leading edge of the bolus passes dorsal to the faucial arch. The velopharyngeal muscles elevate the soft palate to seal the nasopharynx as the bolus enters the muscular pharynx, and respiration ceases momentarily. As the bolus enters the dorsal pharynx, contractions of the pharyngeal constrictor muscles strip the bolus from the oropharynx and hypopharynx. The superior and recurrent laryngeal nerves (RLNs) contribute to the pharyngeal plexus, which provides neuronal control to these muscles. Passage of the bolus through the upper esophageal sphincter (UES) represents the completion of the pharyngeal phase (1, 2 and 3).

The final phase of swallowing, the esophageal phase, begins after passage of the bolus through the UES. Under involuntary control, the bolus is propelled through the length of the esophagus by the inner (circumferential) and outer (longitudinal) layers of the muscularis externa, finally passing through the lower esophageal sphincter near the gastroesophageal junction (1, 2 and 3).

While complex, understanding of the normal swallowing mechanism allows the clinician to critically evaluate the patient with postoperative dysphagia in an attempt to determine where the interruption in the normal pathway exists as well as preoperatively identifying at-risk structures that may be encountered in the surgical field. Evaluating the pattern of dysphagia in a patient with postoperative dysphagia may allow the surgeon to identify the etiology of dysphagia (4). Table 115.1 lists at-risk structures for different surgical levels and potential consequences to the swallowing pathway if their function is impaired. In addition, some data suggest a central spinal cord mechanism of dysphagia exists, as up to 66% of patients presenting with myelopathy exhibit preoperative swallowing dysfunction on modified barium swallow (5). The authors of this study postulate that cervical spinal cord compression from cervical spondylosis may lead to dysphagia by disrupting the normal function of preganglionic, sympathetic outflow or spinal afferents, which in turn interrupts local reflex mechanisms.

The true incidence of dysphagia after cervical spine surgery has been difficult to determine, and published reports vary widely. Three notable prospective studies have attempted to accurately describe the incidence of postoperative dysphagia.

Frempong-Boadu et al. (5) evaluated 23 consecutive patients with barium swallow studies and videolaryngoendoscopy both preoperatively and at 1 week and 1 month postoperatively. It is notable that 11 (48%) of these patients had radiographic swallowing abnormalities preoperatively, but none reported clinical symptoms. At the 1-week followup, 66% of the patients with normal preoperative studies had radiographic evidence of a swallowing abnormality, but again, none reported clinical symptoms. No correlation with age, diabetes, hypertension, or smoking with changes on radiographic studies was found in this small series. There was also no correlation between rates of postoperative dysphagia and duration of endotracheal tube (ET) cuff inflation. There was a nonsignificant trend toward increased rates of radiographic abnormalities with multilevel surgery.

We reported on the incidence of dysphagia in 224 patients undergoing anterior cervical spine surgery (6). The incidence and severity of dysphagia was assessed prospectively at 1, 2, 6, and 12 months postoperatively using a subjective scoring system The incidence of dysphagia of any severity was 50.2%, 32.2%, 17.8%, and 12.5% at 1, 2, 6, and 12 months, respectively. The majority of these patients reported mild dysphagia, and at 6 months, only 4.8% of patients reported moderate or severe symptoms. Multiple-level surgery was a risk factor for dysphagia at 1 and 2 months but not thereafter. No correlation was found between the development of dysphagia symptoms and age, instrumentation, location of levels, and type of procedure (corpectomy vs. discectomy or primary vs. revision). A follow-up study designed to evaluate potential risk factors for the development of dysphagia found very similar incidence at these time points as well as an incidence of 13.6% at 24 months (7). Statistically significant risk factors for dysphagia at final follow-up were female gender (18.3% vs. 9.9%), revision surgery (27.7% vs. 11.3%), and greater than 2-level surgery (19.3% vs. 9.7% for two or less levels).

Smith-Hammond et al. (8) similarly performed a prospective study of patients undergoing anterior cervical surgery over a 3-year period evaluating for the incidence of dysphagia. A group of patients undergoing dorsal cervical and dorsal lumbar procedures over the same time period were included as a control. In this series, the incidence of dysphagia in the anterior cervical group was 47% compared to 21% in the dorsal cervical group and 0% in the dorsal lumbar group at 2.0 ±1.5 days postoperatively as assessed by videofluoroscopy. In the anterior cervical group, the vast majority of patients (71% at 2 months and 94% overall) returned to a normal diet without permanent swallowing dysfunction. The only statistically significant risk factor for the development of dysphagia was age greater than 60. Other comorbidities including hypertension, diabetes mellitus, smoking, gastroesophageal reflux disease, chronic obstructive pulmonary disease, alcohol abuse, substance abuse, coronary artery disease, cancer, and myelopathy were not found to be significant risk factors. Gender was not assessed as a risk factor. This may be because most of their patients were males.

Yue et al. reported on a long-term series of 74 patients treated with anterior cervical discectomy and fusion (ACDF) through a standard Smith-Robinson approach. The average follow-up was 7.2 years (9). Seventy-seven percent of the patients had 1- or 2-level ACDF. Patients were retrospectively asked about any symptoms of dysphagia at a single follow-up visit, which varied from 5.4 to 11.1 years postoperatively. Of these patients, 35.1% reported some dysphagia symptoms at final follow-up. Of 13 patients who reported moderate or severe dysphagia, 10 reported that the benefit of the surgery outweighed the inconvenience of the dysphagia symptoms.