Article Text

Download PDFPDF
Delayed diagnosis of odontoid peg osteomyelitis with bilateral X and XII cranial nerve palsies
  1. Faisal Bashir Chaudhry1,
  2. Samavia Raza2 and
  3. Usman Ahmad3
  1. 1 Department of Stroke Medicine, John Radcliffe Hospital, Oxford, UK
  2. 2 Radiology Department, John Radcliffe Hospital, Oxford, UK
  3. 3 Department of Gastroenterology, Hull and East Yorkshire Hospitals NHS Trust, Hull, UK
  1. Correspondence to Dr Faisal Bashir Chaudhry, faisalbchaudhry{at}gmail.com

Abstract

Upper cervical osteomyelitis is rare. Its presenting features are fever and neck pain, but rarely it can involve lower nerves. MRI is the main imaging modality, but it is difficult to interpret due to the unique anatomy of C1 and C2 vertebra and complex intervertebral joint. We describe a case of a 67-year-old woman, who presented with the complaint of loss of voice, neck pain and fever for 5 days. Despite repeated imaging of neck, the diagnosis was not reached. As the patient’s condition continued to deteriorate, clinical signs of bilateral 10th and 12th cranial nerve paralysis appeared and lead to a focused workup for base of skull pathology. Discussion with the radiologist helped guide the imaging protocol, which leads to the correct diagnosis being made. Treatment was tailored by blood cultures and available images. Temporary immobilisation with a cervical collar and a total of 12 weeks of antibiotics lead to complete remission.

  • radiology (diagnostics)
  • infections
  • musculoskeletal and joint disorders
  • bone and joint infections
  • cranial nerves

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 

Among cases of vertebral osteomyelitis, which itself is an uncommon entity, the involvement of cervical spine is rare, accounting for only 3%–6% of all cases of vertebral osteomyelitis.1 2 Osteomyelitis disease of first and second cervical vertebra is a diagnostic challenge due to multiple reasons. First, due to its proximity to the skull base, disease process can produce neurological deficits, attributed to the involvement of lower cranial nerves, causing diagnostic uncertainty in localisation. MRI is the most sensitive imaging modality in vertebral osteomyelitis.3 On T1-weighted images, the decreased signal intensity of the vertebral bodies and discs and loss of endplate definition are seen. However, the signal intensity of vertebral bodies and discs is increased on T2-weighted images. Due to the unique and complex anatomy of the atlantoaxial joint, imaging can be challenging, even with an MRI. Specific imaging protocol focusing on the base of the skull, combined with a high level of suspicion is required to make the correct diagnosis. The MRI technique is critical, as non-contrast T1 images in all planes without fat saturation are required. Gadolinium-enhanced images without adequate fat saturation techniques can obscure the osteomyelitis.4 Hence, making a diagnosis can be challenging, as images can be difficult to interpret, even with the most reliable imaging modality for spinal diseases like MRI. Delay in diagnosis is linked with increased morbidity and mortality.

Case presentation

A 67-year-old woman, with a history of neck pain and loss of voice, was referred to ENT department of a tertiary care hospital from a local district general hospital, with suspicion of retropharyngeal abscess. The patient had an episode of sore throat about 3 weeks back, from which she had recovered. Now associated with a complaint of low-grade fever of ranging 37.6–38°C and lower neck pain at the back of 7/10 intensity for 5 days, she described a sensation of a lump in her throat. She was previously fit and well and only significant medical history was undergoing hysteroscopy and endometrial biopsy almost a month back. On examination, neck movements were restricted in all directions due to pain and no palpable lumps or lymphadenopathy could be appreciated. There was mild pharyngeal inflammation on throat examination, but tonsils were not swollen or discharging. On flexible nasal endoscopy, there was mild fullness in the posterior pharyngeal wall, otherwise airway was secure, raising the possibility of retropharyngeal abscess. Haematology results showed a white cell count (WBC) of 26×109/L with a neutrophilia of 24×109/L. Biochemistry results showed a plasma C reactive protein (CRP) level of 246 mg/L and mildly deranged liver function tests (bilirubin 33 μmol/L, alanine aminotransferase 79 U/L, alkaline phosphatase 135 U/L).

CT neck with contrast on the day of presentation helped rule out the retropharyngeal abscess. It only showed the marked degenerative change of the cervical spine most pronounced at C6-7 with loss of the intervertebral disc height space (figure 1). In light of these findings and clinical features of neck pain and fever, suspicion of discitis was raised and an MRI cervical spine with gadolinium enhancement was arranged, which did not aid in making a positive diagnosis as images were poor quality due to patient anxiety. The only abnormality shown was narrowing of the cervical canal by the posterior spondylotic process at C5/6, C6/7 and C7/T1. The degenerative changes at these levels did not show an excessive enhancement to support a diagnosis of discitis. No pathological enhancement was evident in the para-spinal soft tissues or posterior to the pharynx. Hence, both CT and MRI head supported no evidence of acute disease process (figure 2).

Figure 1

CT neck with contrast.

Figure 2

MRI cervical spine: (A) T1-weighted TSE, (B) T2-weighted TSE. TSE, turbo spin-echo.

The patient was started on benzylpenicillin and metronidazole initially, but switched to intravenous flucloxacillin 2 g four times a day, after blood cultures grew Staphylococcus  aureus on day 2. The patient was transferred to acute general medicine for further workup of an unclear source of sepsis.

On admission to our ward, the assessment showed slurring of words and slowing of speech, difficulty annunciating and dribble with fluid intake. Although both uvula and tongue showed no deviation on examination, there was diminished palatal movement and minimal tongue protrusion, signifying bilateral lower cranial nerve palsies. Repeat flexible nasal endoscopy confirmed reduced vocal cord movements. There were no signs of meningeal irritation. Cardiac examination and subsequent echocardiogram were normal. Inflammatory markers on admission to our unit were still high though partially improved with WBC count of 18×109/L and CRP of 103 mg/L.

Based on these clinical findings, a list differential diagnosis was formulated including (1) Lemeirre’s syndrome due to previous history of sore throat, (2) posterior cranial fossa abscess based on bulbar nerve involvement and (3) C1-2 discitis or osteomyelitis with surrounding oedema/pus, based on neck pain, hypoglossal palsy and methicillin-susceptible S. aureus bacteraemia. Based on this differential, available images were discussed with a radiologist who highlighted that imaging sequence is not ideal for the base of the skull and degraded by motion artefact. A repeat MRI head and neck with special focus on the craniocervical junction was arranged. It subsequently showed extensive enhancing oedema surrounding the craniocervical junction and within the marrow of the odontoid peg. These appearances suggested osteomyelitis of the odontoid peg and C1, with associated facet joint septic arthritis. There were extradural empyema and longus colli pyomyositis (figure 3).

Figure 3

MRI base of skull/cervical spine with gadolinium: (A) T1 Spectral Presaturation with Inversion Recovery (SPIR) axial, (B) T1 SPIR coronal, (C) STIR sagittal.

The patient was later discussed at bone infection Multidisciplinary team (MDT), where it was agreed to continue with antibiotics at a high dose and frequency of 2 g flucloxacillin every 4 hours. No spinal intervention was planned, but a cervical collar was recommended for stabilisation, for the length of treatment.

Outcome and follow-up

The patient continued to improve and was discharged home with a plan to continue intravenous antibiotics for a total of 12 weeks with home IV team.

Discussion

Vertebral osteomyelitis is a rare condition and makes only 1%–7% of all cases of osteomyelitis. Of that, the involvement of cervical spine is even rare, accounting for just 3%–6% of all cases of vertebral osteomyelitis,1 but when it happens, it is a rapidly progressing disease with a potential of high morbidity or mortality and demands quick intervention.2 Osteomyelitis of the odontoid process is even rarer and requires a high degree of suspicion for diagnosis to be made.5 List of differentials for lesions involving the lower cranial nerves can be vast, involving genetic, vascular, traumatic, iatrogenic, infectious, immunological, metabolic, nutritional, degenerative or neoplastic process.6 However, lower cranial nerve palsy, when seen in combination with neck pain and fever, points the clinician in the right direction. A combination of hypoglossal and vagus nerve palsy, causing dysphonia due to vocal cord paralysis and dysphagia due to weakness of muscles of tongue, is known as Tapia syndrome and is linked with nerve trauma and stretching of nerve tissue during orotracheal intubation.7 8 Rare case has been reported of bilateral hypoglossal and recurrent laryngeal nerve palsy with airway instrumentation,9 but to our knowledge, no case of bilateral hypoglossal and vagus nerve palsy has ever been reported in the literature with odontoid peg osteomyelitis.

The common mechanisms by which bacteria can reach the spine are haematogenous spread, contiguous spread or direct inoculation from trauma or surgery.10 The risk factors include advancing age, diabetes mellitus, poor nutritional status, alcoholism, dental work, liver disease, injection drug use, infective endocarditis, degenerative spine disease, spinal surgery, corticosteroid use or any other immunocompromised state.11–13 Genitourinary tract is also a common primary source of haematogenous spread of bacteria, which may have been the case in this patient, given a history of recent instrumentation. Although in many cases, a primary source cannot be identified.14 What aided in the diagnosis of this case was the growth of Staphylococcus aureus in blood cultures, which highlights the importance of taking repeated blood cultures in patients with fever with unknown source of infection. S. aureus is the most common culprit for vertebral osteomyelitis,15 but other pathogens include gram-negative enteric bacilli, pyogenic and non-pyogenic streptococci, Pseudomonas aeruginosa, coagulase-negative staphylococci and rarely candida.16 17

In the case described, as per a logically dictated diagnostic pathway of neck pain and fever, an MRI spine was organised for the case, but what made an already challenging diagnostic process, even more difficult, was the fact that initial imaging showed no signs of an acute disease. This is partly because of the unusually complex anatomy of atlantoaxial joint, which is a combination of three synovial joints, making the interpretation of routine diagnostic images, quite difficult.18 Due to this complex bony and ligamentous anatomy, imaging of the odontoid peg continues to be a challenge.19 Discussing the case with a neuroradiologist with the diagnostic suspicions helps correct protocol to be sequenced, to delineate the base of the skull and upper cervical spine.

Even after reaching a diagnosis, treatment may still be a tricky matter. Although most cases are due to monomicrobial infections, polymicrobial infections can happen,15 and while culture and sensitivity of biopsy sample are awaited, or it is decided not to biopsy, broad-spectrum antibiotics covering gram-positive, gram-negative and anaerobic bacteria should be used.15 Not all cases need a surgical intervention like decompression or fixation and review of the literature shows that 8–12 weeks long course of intravenous antibiotics, while immobilisation of neck with a collar or brace, may suffice in many cases.5 20

Cervical osteomyelitis is a rapidly progressing disease and the threat of quadriparesis remains the biggest fear of treating clinicians.21 22 Antibiotics have remarkably decreased morbidity and mortality in vertebral osteomyelitis. About 25% suffered a fatal outcome in the preantibiotic era.23 This number has been reduced to <5% with the use of appropriate antibiotics. The rate of residual neurological deficit is roughly <7%.10 Delay in diagnosis remains a major factor that has a negative impact on morbidity and mortality.24

Learning points

  • Odontoid peg osteomyelitis is a life-threatening disease with high morbidity and mortality, if not diagnosed and treated in time.

  • Due to the complexity of presentation and rarity of the disease, a high level of suspicion is required to make the correct diagnosis. Delay in diagnosing is a real possibility, as imaging even gold-standard imaging using MRI may not be diagnostic. Serial imaging and blood cultures would be advisable in patients with similar presenting features like fever and neck pain.

  • MRI remains the mainstay of diagnosis and should be used judiciously in cases of neck pain and fever.

  • It is prudent that all images be discussed and requested, with enough emphasis placed on the clinical picture and differential diagnosis. This not only helps in assigning the desired protocol of imaging sequence but also helps in the interpretation of findings.

References

Footnotes

  • Contributors FBC: was the main clinician involved in management of the case and has contributed to writing up the abstract, main stem of case report and discussion. SR: guided in radiological image interpretation, image selection, literature review and contributed to writing up discussion. UA: helped in literature review and contributed to writing up discussion and conclusion.

  • 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.

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

  • Patient consent for publication Obtained.