Spinal Epidural Abscess

table 1.png

Abscesses arise in the spinal epidural space through a hematogenous route (accounting for approximately 50% of cases) or through direct extension from local infections (i.e. local diskitis or osteomyelitis). (1) Patients usually have on of a number of predisposing factors (See Table 1 right).  The most common isolated bacteria is MRSA with S. epidermidis also being common. (1). Injury to the spinal cord and/or spinal nerve roots can result from the direct mass effects of a growing abscess but can also result from vascular occlusion from septic thrombophlebitis. 

The typical progression of symptoms for a spinal epidural abscess is first, pain at the affected spinal level.  Next patient’s can develop radicular nerve pain radiating from the affected area.  This can then develop into motor weakness and sensory deficits with loss of bowel and bladder control. Finally, paralysis can develop.  Unfortunately, the presentation of these patients is often not dissimilar to patient presenting with benign radicular back pain.  Approximately 50% of patient with a spinal epidural abscess have a fever and only a minority of patients have a triad of back pain, fever, and neuro deficit. (1)

Blood and CSF testing may be helpful in raising the suspicion of disease in some cases but will not be able to confirm the diagnosis.  The white count is often elevated but non-specific and, as always, is “the last refuge for the intellectually destitute” (2) Inflammatory markers such as CRP are almost universally elevated but, again, are highly non-specific. (1)

The imaging study of choice for diagnosing a spinal epidural abscess is MRI with contrast, though CT myelogram also has sensitivity >90% for the diagnosis should your patient have a contraindication for MRI. Epidural abscesses can often track through many spinal levels, so providers should carefully consider obtaining spinal imaging above/below the patient’s primary area of pain or neurologic disability. A retrospective review of 233 adults with spinal epidural abscess, performed by Ju and colleagues (3), attempted to determine clinical and laboratory features predictive of multi-level and skip spinal epidural abscess lesions.  They found that the presence of delay in presentation (>7 days), concomitant infection outside spine and paraspinal region, and an ESR > 95 mm/h predicted a 73% probability of having a skip lesion (multi-level or multi-focal abscess) on neuro imaging.

Prompt recognition and treatment of spinal epidural abscess is critical to prevent and treat devastating neurologic complications from the disease process.  Medical management includes the administration of empiric antibiotics targeted to the likely pathogens (gram (+) coverage with consideration for gram (-) coverage if indicated). Surgical drainage of the epidural abscess is usually needed.  Stratton and colleagues (4) conducted a systematic review and meta-analysis of the literature seeking to identify the risk factors for failed medical management. Reviewing 12 studies with 489 medically treated patients they found a rate of failure for medical management of 29.3%.  Significant heterogeneity in the studies highlighted the need for a consensus definition of failure and only 2 studies reported predictors of failed medical management.  Of the factors reported as predictive of failed medical management, presence of neurologic impairment had the greatest OR (15.2) with the next greatest predictive factor, CRP >115 having an OR of 4.7.  Presence of diabetes, older age (>65 yo), WBC count >12.5, positive blood cultures, and MRSA as the causative agent were also reported to be predictive with OR ranging from 2.5 to 3.8.

Delays in presentation and diagnosis mean that patient’s often present with neurologic disability.  For those patient’s urgent surgical treatment is needed to ensure the best possible outcome.  Avanali and colleagues (5) reported a case series of 23 patients who had delayed surgical treatment.  All 23 patients in the case series had debridement of their epidural abscesses within 24 h of MRI diagnosis but 20/23 were operated on >72 hours after the onset of their neurologic symptoms.  Bowel and bladder dysfunction was seen in 91%o of patients, 19 patients were wheelchair bound, and 21 patients were non-ambulatory prior to surgery. At final follow-up (approximately 4 mo after presentation), only 8 patients were wheelchair bound and all patients had some improvement in bowel/bladder dysfunction, with 4 patients having normal function.  


  1. Darouiche R. Spinal Epidural Abscess. N Engl J Med 2006;355(19):2012–20. 
  2. Christopher Hicks (@HumanFactorz). “Mattu: The white blood cell count is the last refuge for the intellectually destitute #EMUeurope #EMconf #MedEd" 28 September, 2013. [Twitter Post] Retrieved from https://twitter.com/humanfact0rz/status/384055678022987776?lang=en.  
  3. Ju K, Kim SD, Melikian R, Bono C, Harris M. Predicting patients with concurrent noncontiguous spinal epidural abscess lesions. The Spine Journal 2015;15(1):95–101. 
  4. Stratton A, Gustafson K, Thomas K, James MT. Incidence and risk factors for failed medical management of spinal epidural abscess: a systematic review and meta-analysis. Journal of Neurosurgery: Spine 2017;26(1):81–9. 
  5. Avanali R, Ranjan M, Ramachandran S, Devi BI, Narayanan V. Primary pyogenic spinal epidural abscess: How late is too late and how bad is too bad? – A study on surgical outcome after delayed presentation. British Journal of Neurosurgery 2016;12:1–6.