1. Data and patients
A total of 94 patients treated at the First Affiliated Hospital of Dalian Medical University (37 cases) and Tongji Hospital (affiliated to Tongji Medical College of Huazhong University of Science and Technology, 57 cases) from January 2010 to January 2019 were enrolled as research subjects. The patients were provided informed consent, and the relevant institutional review boards granted ethical clearance for the study. The selection criteria were as follows: 1) patients complained of either pharyngeal pain, pharyngeal foreign body sensation, unilateral neck shoulder radiation pain, or radiation ear pain; 2) a bony eminence was palpable in the fossa; 3) either the 3D CT reconstruction of the styloid process or the anteroposterior, lateral X-ray film of the styloid process met the diagnostic criteria, and the length of the styloid process was > 25 mm. Patients with severe cardiovascular diseases, neuropsychiatric diseases, and neuromuscular diseases were excluded. There were 43 men and 51 women with an average age of 45 ± 8.06 years (range, 25–68 years). The disease course was 3–60 months. Of the 94 cases, 63 had unilateral disease and 31 had bilateral diseases; 20 patients underwent styloid process resection through the external cervical approach, 45 underwent tonsillectomy as well as styloid process shortening, and 29 underwent styloid process shortening along with tonsil preservation.
2. Imageological examinations
Imageological examination was performed with a Siemens 64-slice double helix machine (Siemens, Munich, Germany). The patient was put in a supine position with the mandible raised, such that the auditory orbital line was perpendicular to the scanning table. The measurement method used was as follows(7, 9): based on the 3D CT reconstruction image, the length of the styloid process was measured as the distance between its root and the distal end. On the anteroposterior image, a perpendicular line was drawn parallel to the skull base, and the angle between the centre line of the styloid process and the perpendicular measurement line was noted as the inward angle(10). An acoustic orbital line (a perpendicular line connecting the upper edge of the outer ear hole and the lower edge of the orbit) was drawn on the reconstructed image, and the internal deviation angle was measured as the angle between the perpendicular line and the centre line of the styloid process. If the length of the styloid process was greater than 25 mm(11), the styloid process was considered too long. In this case, the angle of inclination (inward or forward) was greater than 25° and was called the abnormal azimuth angle. The anteroposterior, lateral X-ray film of the styloid process was examined as follows: for scanning, the patient (either standing or reclining) was instructed to keep their head in a lateral position and slightly tilt their body. After scanning, the data was transmitted to a picture archiving and communication system workstation to measure the azimuth angle and the length of the styloid process using the same method as before.
3. Surgical indications
At present, patients diagnosed with Eagle’s syndrome are not uncommon. Some patients with a long styloid process undergo conservative treatments such as non-steroidal anti-inflammatory drugs and glucocorticoids(12), although it has a short-term effect, can easily relapse. The inclusion criteria for surgery were the same as the selection criteria for this study. All three criteria (1, 2, and 3) were considered necessary conditions and were strictly observed, and the imaging data were carefully analysed. Although the X-ray of the styloid process could show the long styloid process bone, the 3D reconstruction of the styloid process was used as the final examination method in this study to determine surgical method. This is because the 3D CT scan of the styloid process is more accurate in displaying the details, and accurate judgement of deflection, interruption, and ossification of the hyoid ligament(10).
4. Selection of the operation method
The patients were placed under general anaesthesia and the appropriate surgical method was selected based on their oropharyngeal palpation, 3D CT reconstruction of the styloid process, and specific selection criteria: 1) for patients in whom the styloid process bone could be touched directly during intraoral palpation, or in whom the distal part of the styloid process could not be directly touched but the CT scan showed that the bone inclined toward the oropharynx and its distal part was relatively close to the oropharynx cavity (which does not cause excessive injury in the parapharyngeal space), styloid process resection was selected for transoral surgery. Whether tonsillectomy was performed depended on whether the patient’s tonsil was too large to affect the surgical incision; 2) for patients whose styloid process could not be touched through the mouth or could be touched under the jaw, or when the CT scan showed that the inclination angle was not large, the external cervical approach was selected. The selection of appropriate surgical methods can lower operation time, injury, postoperative pharyngalgia, and the probability of complications.
4.1. Bimanual examination
The index finger of one hand was placed in the affected tonsillar fossa, and the other finger was placed behind or under the ipsilateral ramus of the mandible(13). Of 74 patients who underwent surgery using the intraoral approach, 49 patients could either touch the hard cord in the unilateral or bilateral tonsillar fossa, or felt tenderness, fullness, or other discomforts at the touchpoint; 10 patients could touch the hard cord behind or under the ramus of the mandible. The other 15 patients could not touch any abnormality. Specific surgical methods: 1) the tonsillar fossa approach: if the tonsil was too large and affected the exposure of the whole styloid process, tonsillectomy was performed first. Based on the position and course of the styloid process, a sickle knife was used to cut the mucosa of the palatoglossal or palatopharyngeal arch in order to separate the superior pharyngeal muscle and expose the distal styloid process. Following this, the styloid process periosteum was cut, and soft tissues such as the muscle and styloid hyoid ligament were excised. The styloid process was covered with a small ethmoid curette from the free end and walked along the styloid process (which was clipped with a vascular clamp), then cut and removed with bone biting forceps. The cut-off length was 10–38 mm. If the residual styloid process was long, it was further stripped and shortened. During the operation, it was necessary to clamp the broken end of the styloid process with a vascular clamp as it is difficult to find once it slips. Small incisions were not sutured; however, large incisions were sutured with 3 − 0 absorbable sutures to eliminate large spaces and prevent parapharyngeal space infection. In patients with tonsillectomy, it was not necessary to suture the tonsillar fossa completely. 2) Using the palatoglossal arch- soft palate approach, the mucosa was cut between the medial sides of the opsigenes on the palatoglossal arch, and the incision was extended to the upper soft palate to a length of 2.5 cm. The remaining steps were the same as those described for surgical procedure 1.
4.2. External cervical approach
An arc-shaped incision was made around the mandibular angle, approximately 1–2 cm from the posterior edge of the ramus to the inferior edge of the mandible. The length of the incision was approximately 4 cm. The skin was cut, and the layers of subcutaneous tissue and platysma were separated. The parotid masseteric fascia was exposed to protect its superficial layer and to cross the facial nerve. The mandibular angle was fully exposed, and the styloid mandibular ligament was located. The styloid process was located, and the attached mandibular ligament of the styloid process and styloid hyoid ligament was cut. The attachment site of the styloid process was approached using the ethmoid curette, and a small pair of bone biting forceps was inserted along the styloid process.
5. Statistical analyses
The SPSS software (IBM SPSS Statistics for Windows, Version 19.0; IBM Corp., Armonk, NY: IBM Corp.) was used to conduct rank sum tests, and P values < 0.05 were considered statistically significant.