Article Text

Download PDFPDF
CASE REPORT
Rare case of anterior cervical discectomy and fusion complication in a patient with Zenker’s diverticulum
  1. Mauro Dobran,
  2. Maurizio Gladi,
  3. Fabrizio Mancini and
  4. Davide Nasi
  1. Neurosurgery, Ospedali Riuniti di Ancona, Università Politecnica delle Marche, Ancona, Italy
  1. Correspondence to Dr Mauro Dobran, dobran{at}libero.it

Abstract

We present a case of Zenker’s diverticulum in a 45-year-old woman, occurred as complication after anterior cervical discectomy and fusion for a cervical spine injury. The oesophageal complication occurred 12 months after vertebral cervical surgery and presenting symptoms were fever, dysphagia and neck pain with evidence of retropharyngeal infection. We performed a posterior cervical stabilisation C3-D1 by screws and rods and a second anterior left cervical approach with anterior plate removing and oesophageal wall break repairing with a sternohyoid muscle patch. Despite pharyngo-oesophageal diverticulum may be a complication of anterior cervical surgery (traction diverticulum), in case of an already present true Zenker’s diverticulum, delayed complication may occur without cervical hardware pull-out.

  • neurology (drugs and medicines)
  • bone and joint infections
  • neuroimaging
  • trauma CNS /PNS
  • neurosurgery

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Background  

Anterior cervical spine surgery is the procedure of choice to treat cervical trauma, disc herniation and degenerative disease; in these patients oesophageal and pharyngeal injury may occur during surgery. The perforation of pharyngo-oesophageal diverticulum of Zenker is rare and literature reports only few cases. Zenker’s diverticulum is an outpouch of the posterior oesophageal wall located proximally to the upper oesophageal sphincter; it may be a direct consequence of surgical procedure (traction diverticulum) or present at the surgical time (pulsion diverticulum). Surgical repair is possible and prompt recognition and treatment avoid serious complications such as abscess, mediastinitis, pneumonia and sepsis.

We present a case of Zenker’s diverticulum which occurred as complication in a patient who underwent anterior cervical discectomy and fusion (ACDF) for a cervical spine injury. Due to the presence of abscess with mediastinal extension, hardware has been removed and oesophageal repair was performed by wall suturing and muscle flap reinforcement.

Case presentation

We present the case of a 45-year-old woman admitted to Neurosurgical Clinic of Ancona for cervical spine injury. Neurological examination revealed flaccid tetraparesis with hypoesthesia at T1 level. Cervical spine CT scan and MR image showed acute traumatic lesion of the C6-C7 intersomatic disc, C6-C7 somatic body lussation and anterior longitudinal ligament rupture; intramedullary signal intensity on the T2-weight images at C6 level was also documented (figure 1).

Figure 1

Cervical spine trauma with C6-C7 dislocation and anterior longitudinal ligament (ALL) rupture.

The patient underwent surgical procedure of anterior C6-C7 cervical discectomy and fusion with screws and plate (figure 2). The postoperative period was uneventful and 9 days after surgery she moved to the rehabilitation centre where she gradually improved with onset of some movement of the shoulders.

Figure 2

Anterior cervical discectomy and fusion at C6-C7 level.

After 3 months, she suffered from acute onset of bilateral cervical and brachial pain, fever and neurological worsening.

Investigations

The MR and CT scan images of cervical spine (with contrast media) showed retropharyngeal abscess extended into mediastinum (figure 3). Laboratory analysis showed neutrophilic leucocytosis (white cells=15.700/mmc, neutrophil=14.300/mmc) and high inflammatory markers (PCR=23.62 mg/dL, fibrinogen=873 mg/dL). Cerebrospinal fluid microbiological examination was negative. Antibiotic therapy immediately started with meropenem (1 g×3/die, rifampicin 600 mg/die and daptomycin 500 mg/die). The endoscopic oesophageal examination performed for dysphagia revealed sacciform recess (3.5 cm) like Zenker’s diverticulum with screws and plate visible inside the fundus through a break of the posterior wall (figure 4). Oesophageal transit and lumen were normal.

Figure 3

Retropharingeal infection with extension into mediastinum.

Figure 4

Oesophagogram and endoscopy showing diverticulum with clear distinction from true oesophageal lumen. Evidence of plate and screw inside diverticulum.

Differential diagnosis

Hardware pull-out with subsequent posterior oesophageal wall injury should be differentiated from primary Zenker’s diverticulum on which pressure necrosis due to contact with a correct positioned hardware may develop.

Treatment

We performed a posterior cervical stabilisation C3-D1 by screws and rods and a second anterior left cervical approach with the help of thoracic surgeon. In this second approach we removed anterior plate and oesophageal wall break was sutured with a patch of sternohyoid muscle. Six days after surgery, barium swallow showed normal oesophageal transit without contrast leakage through fistula but the diverticulum was still present (figure 5).

Figure 5

Postoperative oesophagogram with no contrast leakage but diverticulum is still present.

Outcome and follow-up

At the moment the patient suffers from stable clinical spastic tetraparesis without signs of prevertebral infection, and nutrition is allowed by percutaneous endoscopic gastrostomy. For the residual diverticulum further surgical procedure (diverticulectomy or diverticulopexy) is under evaluation by thoracic surgeons.

Discussion

ACDF is one of the most commonly performed neurosurgical procedures for patients with spondylotic myelopathy, radiculopathy or cervical trauma with generally good outcome. In a series of 1015 ACDF cases, the mortality rate was 0.1% and overall morbidity rate was 19.3%.1 The most common complications were dysphagia (9.5%), haematoma (5.6%), recurrent laryngeal nerve palsy (3.1%), dural tear (0.5%), oesophageal perforation (0.3%), worsening of myelopathy (0.2%), Horner’s syndrome (0.1%), instrumentation failure (0.1%) and wound infection (0.1%). Other rare described complications are thoracic duct injury, vertebral artery lesions, carotid artery injury, spondylodiscitis, meningitis, jugular vein thrombosis, abscess and cervical spine deformity.2–9

Oesophageal injuries are well-known complications of ACDF, but their precise incidence is still debated.10 When an intraoperative complication occurs prompt management must be performed: identification of lesion can be accomplished by oesophagoscopy or oesophagus filling with indigo carmine by nasogastric tube. To reduce the risk of oesophageal injuries during ACDF surgery, gentle retraction and careful dissection should be performed.

Oesophagus wall perforations may occur even a long time after surgery and are difficult to identify. Subcutaneous emphysema may indicate oesophageal perforation and symptoms like dysphagia or odynophagia should be further investigated with plain cervical films, CT of thorax, neck and oesophagram. Delayed recognition of oesophageal perforations may lead to mediastinitis, neck abscess and sepsis, with a mortality rate of 50%. Among pharyngo-oesophageal injuries related to ACDF, pharyngo-oesophageal diverticulum (Zenker’s diverticulum) perforation is rare with only few cases reported in literature.

The pharyngo-oesophageal diverticulum (Zenker’s diverticulum) is the most common oesophageal diverticulum originally described by Zenker and Von Ziemssen in 1887.11 Anatomically, it consists of mucosal outpouching herniated into the Killian’s triangle that is a zone of weakness located between the oblique fibres of the thyropharyngeus muscle and horizontal fibres of the cricopharyngeus muscle.12 Zenker’s diverticulum is usually present in older patients due to the progressive loss of tissue elasticity and muscle tone with age. It has been supposed that, because of tissue weakness, swallowing incoordination may lead to abnormal bolus pressure in pharynx with progressive herniation of submucosa through Killian’s triangle. Gradual enlargement of the diverticulum causes its migration downward into prevertebral space, preferentially on the left side, and even into the mediastinum. Initially, patients are asymptomatic but, as the sac size increases, halitosis, voice changes, retrosternal pain and regurgitation of foul-smelling undigested material are common presenting symptoms.13 Diagnosis is made by barium oesophagography in lateral view. Surgical fixation (diverticulopexy) and resection (diverticulectomy) are the two open procedures performed to treat Zenker’s diverticulum, usually associated to myotomy of thyropharyngeus and cricopharyngeus muscles. For diverticula larger than 3 cm, same outcome can be obtained with endoscopic procedure that consists of division of the common wall between oesophagus and diverticulum using laser or stapler.14

Some authors suggest an anatomical and pathophysiological distinction between true Zenker’s diverticulum and ACDF-related diverticulum.15 Zenker’s diverticulum is a false diverticulum that involves herniation of submucosal layer alone caused by pulsion forces, while ACDF-related diverticulum is a traction diverticulum derived from traction forces due to fibrous scar tissue between oesophagus, prevertebral tissue and cervical plate. In this case, the outpouching hernia consists of all oesophagus layers, including muscle layer. Despite we got no histological tissue exam because no diverticulectomy was performed, intraoperative diverticula in present case strongly resemble true Zenker’s diverticulum because of the absence of clear muscle layer.

Patients who develop diverticulum after anterior cervical spine surgery are generally young, with a mean age of 40 years, and this may be due to the fact that trauma is the most common cause of surgery indication, followed by cervical spondylosis.16 The level more frequently involved is C5-C6 and it seems that there is no correlation between number of operated levels and risk of diverticulum. Latency between surgery and symptom appearance is about 3 years; there is a single described case of a 45-year-old male patient who developed dysphagia 13 years after an ACDF for cervical trauma.17

Late oesophageal injuries after anterior cervical surgery are generally attributed to plate or screw pull-out that erodes oesophageal wall, but this is not described in all cases.18–20 Several patients with late oesophageal perforation or diverticula without hardware dislodgement are described, probably resulting from pressure necrosis from plate to oesophagus. This is a slow process that may require years to become symptomatic and when it occurs, the patient generally suffers from worsening dysphagia or multiple bouts of pneumonia. In the present case, despite any evidence of plate and screw with endoscopy, no hardware dislodgement has occurred as documented by CT scan and intraoperative view, so a progressive scar tissue formation and erosion of already present Zenker’s diverticulum can be supposed.

The treatment of choice for diverticulum related to anterior cervical surgery is open diverticulectomy, generally with a left approach and oesophageal repair. As in other instrumented spinal surgical procedures, infection is a well-described complication21 22 and generally hardware must be removed to achieve a complete infection resolution; anyway in some cases alternative procedure such as VAC therapy with hardware preservation has been described.23 In this case, prevertebral infection extended to mediastinum was due to oesophageal perforation so we decided to remove anterior cervical hardware after performing a posterior stabilisation with screws and rods avoiding occipital cervical inclusion.24 25 Closure of oesophagus should be performed by a direct suture and a subsequent reinforcement with sternocleidomastoid,26 infrahyoid27 or pectoralis muscle flap. An omental flap has also been described, with good outcome and reduction of infection rate.28 Despite its efficacy in the treatment of Zenker’s diverticulum, endoscopic procedures are not advocated for diverticulum repair because hardware and scar tissue make it impossible.

In the present case, oesophageal complication occurred 12 months after vertebral cervical surgery, was earlier than other reported cases. This early complication was not related to slow scar tissue formation as previously described, but somehow to pressure necrosis over an already present diverticulum consisting of only mucous layer. Probably, subsequent wall perforation led to symptomatic prevertebral infection. During endoscopy, a clear posterior sac was identified with evidence of screws and plate inside it; anyway, both CT scan and intraoperative view showed no hardware dislocation and we supposed a progressive erosion of posterior pouch by hardware. Given the evidence of infection, anterior hardware was removed but we performed a posterior fixation before the anterior revision to maintain cervical spine stability. Oesophageal repair was made by left cervical approach, posterior wall suture and sternohyoid muscle reinforcement.

Some authors suggest an anatomical and pathophysiological distinction between true Zenker’s diverticulum and ACDF-related diverticulum.15 Zenker’s diverticulum is a false diverticulum that involves herniation of submucosal layer alone caused by pulsion forces, while ACDF-related diverticulum is a traction diverticulum derived from traction forces due to fibrous scar tissue between oesophagus, prevertebral tissue and cervical plate. In this case, the outpouching hernia consists of all oesophagus layers, including muscle layer. Despite we got no histological tissue exam because no diverticulectomy was performed, intraoperative diverticula in present case strongly resemble true Zenker’s diverticulum because of the absence of clear muscle layer.

Patients who develop diverticulum after anterior cervical spine surgery are generally young, with a mean age of 40 years, and this may be due to the fact that trauma is the most common cause of surgery indication, followed by cervical spondylosis.16 The level more frequently involved is C5-C6 and it seems that there is no correlation between number of operated levels and risk of diverticulum. Latency between surgery and symptom appearance is about 3 years; there is a single described case of a 45-year-old male patient who developed dysphagia 13 years after an ACDF for cervical trauma.17

Late oesophageal injuries after anterior cervical surgery are generally attributed to plate or screw pull-out that erodes oesophageal wall, but this is not described in all cases.18–20 Several patients with late oesophageal perforation or diverticula without hardware dislodgement are described, probably resulting from pressure necrosis from plate to oesophagus. This is a slow process that may require years to become symptomatic and when it occurs, the patient generally suffers from worsening dysphagia or multiple bouts of pneumonia. In the present case, despite any evidence of plate and screw with endoscopy, no hardware dislodgement has occurred as documented by CT scan and intraoperative view, so a progressive scar tissue formation and erosion of already present Zenker’s diverticulum can be supposed.

The treatment of choice for diverticulum related to anterior cervical surgery is open diverticulectomy, generally with a left approach and oesophageal repair. As in other instrumented spinal surgical procedures, infection is a well-described complication21 22 and generally hardware must be removed to achieve a complete infection resolution; anyway in some cases alternative procedure such as VAC therapy with hardware preservation has been described.23 In this case, prevertebral infection extended to mediastinum was due to oesophageal perforation so we decided to remove anterior cervical hardware after performing a posterior stabilisation with screws and rods avoiding occipital cervical inclusion.24 25 Closure of oesophagus should be performed by a direct suture and a subsequent reinforcement with sternocleidomastoid,26 infrahyoid27 or pectoralis muscle flap. An omental flap has also been described, with good outcome and reduction of infection rate.28 Despite its efficacy in the treatment of Zenker’s diverticulum, endoscopic procedures are not advocated for diverticulum repair because hardware and scar tissue make it impossible.

In the present case, oesophageal complication occurred 12 months after vertebral cervical surgery, was earlier than other reported cases. This early complication was not related to slow scar tissue formation as previously described, but somehow to pressure necrosis over an already present diverticulum consisting of only mucous layer. Probably, subsequent wall perforation led to symptomatic prevertebral infection. During endoscopy, a clear posterior sac was identified with evidence of screws and plate inside it; anyway, both CT scan and intraoperative view showed no hardware dislocation and we supposed a progressive erosion of posterior pouch by hardware. Given the evidence of infection, anterior hardware was removed but we performed a posterior fixation before the anterior revision to maintain cervical spine stability. Oesophageal repair was made by left cervical approach, posterior wall suture and sternohyoid muscle reinforcement.

According to the paper of Duransoy et al 29 in delayed perforations of oesophageal wall related to the loosening of the hardware, the anterior cervical instruments must be removed and the oesophageal perforation repaired by sutures or flap with sternocleidomastoid muscle. In case of instability posterior stabilisation should be performed.

In the present case, oesophageal complication occurred 12 months after vertebral cervical surgery, was earlier than other reported cases. This early complication was not related to slow scar tissue formation as previously described, but somehow to pressure necrosis over an already present diverticulum consisting of only mucous layer. Probably, subsequent wall perforation led to symptomatic prevertebral infection. During endoscopy, a clear posterior sac was identified with evidence of screws and plate inside it; anyway, both CT scan and intraoperative view showed no hardware dislocation and we supposed a progressive erosion of posterior pouch by hardware. Given the evidence of infection, anterior hardware was removed but we performed a posterior fixation before the anterior revision to maintain cervical spine stability. Oesophageal repair was made by left cervical approach, posterior wall suture and sternohyoid muscle reinforcement.

Learning points

  • Diverticulum related to anterior cervical surgery may be the direct consequence of the procedure (traction diverticulum) or it may be already present at the time of surgery as a true Zenker’s diverticulum and it may represent a risk factor for oesophageal injury.

  • Surgeons should be aware of the possibility of intraoperative diverticulum identification and postoperative symptoms such as dysphagia and odynophagia, and signs of infection should always be investigated.

  • Treatment should be performed with open surgery, with or without hardware removal, and oesophageal tears should be sutured and reinforced with a muscle flap.

  • Further studies are necessary to evaluate if preoperative exclusion of diverticulum may be advising to select patients who undergo anterior cervical surgery.

References

Footnotes

  • Contributors MD: Study design. MG: Elaboration data and literature research. DN: Reviewer.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.