Diagnosis of Pharyngeal Fistula by Video Laryngoscopy and Its Nonsurgical Treatment: a Case Report.
Dong Wang,Jun Dong,Meng Lu,Bao-hui Yang,Hao-peng Li,Xi-jing He
DOI: https://doi.org/10.1111/os.12097
2014-01-01
Orthopaedic Surgery
Abstract:Anterior cervical discectomy and fusion (ACDF) is a widely used and effective procedure for diseases of the cervical vertebrae1, 2. However, several postoperative complications can readily occur, for example, damage to adjacent organs such as the esophagus and trachea. Pharyngeal fistula, a rare and serious complication which can occur after cervical surgery can lead to wound, vertebral and/or mediastinal infection, resulting in a high mortality. The reported morbidity of this disease following anterior cervical surgery is 0.2% to 1.49%3, 4. Either operative or conservative intervention is reported effective in the management of pharyngeal fistula. Surgical treatment such as debridement and exploratory surgery, primary suturing of the perforation, sternocleidomastoid myoplasty have been reported3. However, as spontaneous recovery of this disease is possible, reoperation is not always necessary. The conservative treatment generally includes clinical observation, extra oral feeding, and systemic antibiotics. The diagnosis of pharyngeal fistula can be achieved by imaging or endoscopic studies5. Here, we report a case of pharyngeal fistula after ACDF and discuss the possible pathogenesis and treatment. A 36-year-old man with a history of encephalitis resulting in intermittent head spasm presented to us complaining of neck pain and numbness on the right side of his body. Magnetic resonance imaging (MRI) examination revealed disc herniation at the C3–4 level (Fig. 1). Because there were clear indications for surgery, he underwent anterior cervical decompression at the C3–4 level and insertion of an appropriately sized titanium cage packed with autogenous bone grafts into the anterior defect. In view of his head spasm, the senior author decided to perform fusion with internal fixation and cervical pedicle screw fixation via a combined anterior and posterior approach. The plate and screws appeared satisfactory on a postoperative radiograph (Fig. 2). Three days postoperatively, he developed a sore throat, dysphagia and odynophagia. However, no abnormalities were found on physical examination of the lungs and chest. Various investigations, including chest X-ray, white blood cell count, erythrocyte sedimentation rate, and C-reactive protein concentration, were negative. Eventually, a pharyngeal fistula was detected by video laryngoscopy (Fig. 3). Preoperative T2 sagittal image from cervical spinal MRI showing severe cervical spondylitis with swan-neck deformity and severe compression of the spinal cord at the C3–4 level (white arrow). Postoperative cervical lateral view radiograph showing that the instrumentation is satisfactorily placed. First electronic laryngoscopy: a pharyngeal fistula almost 3 mm × 5 mm with a pseudomembrane covering its surface (in the red circle) is visible. A nasogastric tube was immediately inserted and intravenous rehydration and cephalosporin were administered. A sinus emerged through the apparently-healed wound 15 days after surgery, accompanied by oozing of mucus and partial anterior subcutaneous emphysema. After removal of the pseudomembrane inside the sinus, the wound was thoroughly flushed with gentamicin and metronidazole and packed with povidone-iodine gauze. The patient also received a daily metronidazole mouthwash and intravenous nutrition and his condition improved considerably. Although the exudate from the wound had dramatically decreased by the 20th day of daily replacement of the wound dressing, the diameter of sinus had increased. Forty-five days postoperatively, during which time daily dressing replacement and medication had been continued; iodoform was substituted for povidone-iodine gauze in the sinus to decrease the size of the wound. Subsequently, laryngoscopic reexamination revealed that the diameter of the fistula had decreased remarkably (Fig. 4). By 60 days after surgery, the external wound had completely healed over. On laryngoscopy, the original fistula was found to have closed and only pseudomembrane was visible on the surface (Fig. 5). The patient was eventually discharged without any clinical sequelae. Second electronic laryngoscopy: the pharyngeal fistula is 2 mm × 3 mm (in the red circle). Third electronic laryngoscopy: the pharyngeal fistula has almost healed (in the red circle). Pharyngeal fistulas, a rare and potentially lethal complication of cervical surgery6, tends to occur in the postoperative period. However, they can develop weeks to months or even longer after such procedures7. For example, Gazzeri et al. reported a patient in whom esophageal perforation developed 11 years after ACDF8. This complication is so rare that, up to now, it has only been documented by anecdotal case reports7-11. The reported mortality of pharyngoesophageal fistula is 20% to 65%8. Typically, esophageal perforation occurs after screw fixation of the lower cervical spine anteriorly, whereas pharyngeal fistulas may form after surgery on higher cervical vertebral segments. There are three main causes of pharyngeal fistula formation9: errors in surgical techniques/judgments, biological factors and postoperative trauma. They are easily caused by injuring the triangular space, which is a weak area immediately above the cricopharyngeal threshold. In our case, we performed decompression at the C3–4 level, corpectomy of the C3 vertebra and insertion anteriorly of an appropriately sized titanium cage packed with autogenous bone grafts. Thus, over-resection of post-pharyngeal flap led to a pharyngeal muscle defect. Various therapies have been used to manage pharyngeal fistulas. Surgical revision is necessary if the fistula has been caused by complications of surgery such as migration of loose screws or plate fracture. During revision operation, it suggested that internal cervical screws or other fixation be left in place if they are all stable to minimize the risk of developing a new fistula8. The surgeon should focus on suturing the incision in layers, from the external to the internal mucosa6. However, because spontaneous resolution of fistulas can occur, reoperation is not always necessary6, 10, 11. In some cases, the fistula is too small to require a second surgery2. Vogel and Strong described a case in which a gallbladder T-tube was used to close the defect12. In our case, the pharyngeal fistula was caused by the over-resection of the post-pharyngeal flap during the procedure. Thus, a revision procedure was unnecessary. Use of a nasogastric tube and/or intravenous line to provide nourishment obviates the need for swallowing, thus contributing to closing of the fistula. In our case, infection was prevented by packing the fistula with iodoform gauze; this also contracts the interior walls of the sinus. Thus, our patient was managed non-surgically by insertion of a nasogastric tube, iodoform gauze packing of the fistula and intravenous nutrition. Early identification and immediate treatment of pharyngeal fistula reduces adverse effects. However, in the early stages some patients are asymptomatic6. In our experience, the possibility of fistula should be considered when symptoms such as throat pain, dysphagia, odynophagia or hoarseness, develop after cervical surgery. Various means of diagnosing pharyngeal fistulas and esophageal perforation have been reported; these include barium meal, direct visualization, plain radiography and CT (with esophageal contrast)1, 8, 13. However, none of these is sufficiently sensitive. Taylor and colleagues compared three techniques for detecting esophageal perforation using intra-esophageal dye injection and concluded that full reliance should not be placed on intra-esophageal dye injection via nasogastric tube alone when attempting to diagnose esophageal perforation14. In the patient described in this article, video laryngoscopy revealed a fistula from the posterior wall of the pharyngoesophageal junction to the trachea. Eventually, video laryngoscopic examination demonstrated that the pharyngeal fistula had completely healed. To our knowledge, this is the first reported case of pharyngeal fistula after ACDF diagnosed definitively by video laryngoscopy and managed non-surgically without any significant clinical complications.