Spinal epidural abscess

OVERVIEW: What every practitioner needs to know

Spinal epidural abscess represents the accumulation of purulent material in the space between the dura mater and osseo-ligamentous confines of the vertebral canal. Spinal epidural abscess presents a diagnostic challenge, particularly in pediatric populations. The diagnosis of spinal epidural abscess requires a high level of clinical suspicion. In patients with increasing back pain or localized spinal tenderness and elevated erythrocyte sedimentation rate (ESR) with or without leukocytosis, early imaging studies must be considered. Gadolinium-enhanced MRI is the preferred diagnostic technique when spinal epidural abscess is suspected. Early diagnosis and optimal management require knowledge of the presenting symptoms and signs and its pathogenesis.

Are you sure your patient has spinal epidural abscess? What are the typical findings for this disease?

Clinical presentation depends on the predisposing condition and pathogenesis. The typical presentation includes fever, localized spinal tenderness or back pain and progressive neurological deficits, but the classic triad of back pain, fever, and neurologic deficit is seen in about 8%-37% of cases. Back pain is the most common symptom (70-90% of cases, described as relentless and severe), followed by fever (60%-90%), spinal tenderness (up to one-third of cases) and paresthesia (35% of cases).

Neurologic dysfunction is often evident at the time of initial examination (71% have an abnormal neurologic examination at presentation). Radicular complaints or signs of nerve root irritation are present in 12%-47% of patients, with 26%-60% displaying some degree of weakness. Neurological progression often follows fever and back pain, with radicular pain or meningismus preceding frank sensory or motor deficits. Loss of bowel or bladder control is a late presentation commonly concomitant with substantial paraparesis.

An established staging system outlines the progression of the above-mentioned symptoms and signs:

Stage 1, back pain at the level of the affected spine

Stage 2, nerve-root pain radiating from the involved spine area

Stage 3, motor weakness, sensory deficit, bladder and bowel dysfunction

Stage 4, paralysis

What other disease/condition shares some of these symptoms?

Patients presenting with discitis or spontaneous vertebral osteomyelitis as a result of hematogenous dissemination of the avascular disc and the adjacent vertebral endplates have combined discitis and osteomyelitis, which can precede the formation of the spinal epidural abscess. Many patients (in adults) have identifiable risk factors or co-morbidities that predispose to or facilitate spinal infection, such as injected-drug use, diabetes mellitus, hypertension, end-stage renal disease, or COPD.

What caused this disease to develop at this time?

The spinal epidural space is a contiguous vertical sleeve filled with fat, arteries and venous complexes. The location and extent of spinal epidural abscesses are associated with anatomical structures. Spinal epidural abscess is usually located in the posterior space, and anteriorly located abscesses are frequently associated with vertebral osteomyelitis. As an abscess increases in size, it extends along the dural sheath and usually involves several segments.

Infections away from the spinal canal, including infections of the skin and skin structure, respiratory tract, genitourinary system, abdominal viscera, and oral cavity, as well as infective endocarditis, may spread to the epidural space. Abscesses formed by this route are generally localized to the posterior epidural space.

Contiguous spread usually originates from vertebral osteomyelitis or, less commonly, from paravertebral, retropharyngeal or psoas abscess, and this mode of spread generally infects the anterior portion of the spinal canal.

Inoculation of the epidural space can occur during invasive therapeutic and diagnostic procedures, which include spine surgery, epidural catheterization, and central nerve block. Invasive spinal surgery exposes patients to intraoperative and postoperative infection with skin microorganisms or multidrug-resistant organisms.

Other risk factors include penetrating injury to the spine, morbid obesity, primary or iatrogenic immunosuppression, skin infections or abscesses, and transient bacteremia.

Local predisposing factors include spine trauma, spinal surgery, extrathecal injection, or catheter placement into the vertebral canal.

Patients presenting with spinal epidural abscess after invasive procedures are often healthy individuals without co-morbidities and develop a spinal epidural abscess days or weeks after the procedure. Patients present with worsening pain at or near the surgical or puncture site, accompanied by purulent discharge. These findings also vary with the type of procedure, as patients who have procedures that affect disc space, such as discectomies or discograms, can develop postprocedure discitis or osteomyelitis that precedes the spinal epidural abscess.

What laboratory studies should you request to help confirm the diagnosis? How should you interpret the results?

Once spinal epidural abscess is suspected, image-guided biopsy as well as cultures of the blood and urine should be obtained for determination of the microbial etiology. If the patient remains clinically and neurologically stable, then antibiotic treatment should be withheld until specimens from the suspected abscess are obtained for cultures.

Diagnostic procedures that penetrate the spinal canal generally should not be done until spinal epidural abscess has been ruled out, because needle puncture of the dura carries the risk of seeding the intrathecal compartment with resultant meningitis, and aspiration caudal to the level of a block to CSF flow can lead to neurologic deterioration. In cases where CSF is evaluated, CSF analysis shows a high level of protein and pleocytosis, suggestive of parameningeal inflammation.

Would imaging studies be helpful? If so, which ones?

The diagnosis of spinal epidural abscess requires a high level of clinical suspicion. In patients with increasing back pain or localized spinal tenderness and elevated ESR with or without leukocytosis, early scanning must be considered. Plain radiographs are unreliable and show vertebral erosion in only a few patients. Contrast-enhanced CT is inadequate for early detection of spinal epidural abscess, and the diagnosis of spinal epidural abscess cannot be reliably excluded by CT. CT can only be recommended in an emergency setting in the absence of MRI. Only CT myelography can show the site of cord compression, and is useful when MRI is contraindicated. Small abscesses in the epidural location can be easily overlooked. Nuclear medicine studies such as technetium, gallium and indium isotope scans play little, if any, role in the diagnosis.

Gadolinium-enhanced MRI is the preferred diagnostic technique when spinal epidural abscess is suspected. MRI visualizes epidural abscess clearly as a T2 hyperintense and mixed-signal mass with marked gadolinium enhancement on T1W1. Images in the axial and sagittal plane are useful for precise surgical planning.

Confirming the diagnosis

If the patient develops fever, back pain or localized spinal tenderness with and/or without neurological deficits, particularly following spine surgery or trauma, or deep-seated infection such as psoas or retropharyngeal abscess, then spinal epidural abscess should be included in the differential diagnosis, and imaging studies should be performed.

If you are able to confirm that the patient has spinal epidural abscess, what treatment should be initiated?

Emergent surgical decompression and drainage is necessary in cases with compression of neural structures. Otherwise, diagnostic collection of abscess fluid for culture is essential.

Antimicrobial therapy

The medical treatment of spinal epidural abscesses without neurological compromise includes bed rest and parenteral antimicrobials. However, because of the risk of progressive neurological dysfunction despite antimicrobial therapy and the correlation between improved outcome and preoperative neurologic function, close neurological monitoring must be maintained and surgical intervention should be done at the earliest sign of deterioration.

Medical management has also been used in patients with neurological symptoms who are at high risk of complications during surgery, in patients with complete paresis for more than 48-72 hours, or in patients with extensive infections involving many vertebral levels of the neural axis.

Antimicrobials should be started before an operative procedure if a substantial delay before surgery cannot be avoided, significant neurologic dysfunction exists, or the patient is septic or critically ill. Empiric antimicrobial therapy should be initiated immediately after cultures have been obtained and must provide coverage against both Gram-positive cocci (including methicillin-resistant S. aureus (MRSA), such as vancomycin) and Gram-negative bacilli (third- or fourth-generation cephalosporins, such as cetriaxone, ceftazidime or cefepime), particularly in patients with distant Gram-negative infection or bacteremia.

No evidence-based guidelines are available on the duration of parenteral antimicrobials for spinal epidural abscesses that do not compress epidural regions. A course of 6 to 8 weeks of parenteral antimicrobials is usually recommended on the basis of clinical efficacy. Some specialists provide an additional 1 to 2 months of oral antimicrobials. When medical management is recommended, close monitoring with frequent neurological examination, serial CBC with WBC counts and differential, ESR and C-reactive protein (CRP), and repeat MRI are essential. Normalization of elevated ESR or CRP may represent the response to treatment.

Surgical treatment

The primary goals of surgical treatment of spinal epidural abscess are rapid decompression, abscess drainage, and microbiological diagnosis. Vertebral osteomyelitis associated with spinal epidural abscess can result in structural instability that warrants surgical correction.

Supportive therapy

Immobilization is indicated for patients with substantial axial or potential spinal instability and should be maintained until pain resolution or spinal stability.

What are the adverse effects associated with each treatment option?

N/A

What are the possible outcomes of spinal epidural abscess?

Neurological deficits, including spinal cord injury, are the major complication of spinal epidural abscess. Epidural abscess can injure the spinal cord either (1) directly by compression of the spinal cord or cauda equina produced by abscess expansion within the closed space of the vertebral canal, or (2) indirectly as a result of vascular occlusion of spinal cord arterial flow, thrombosis of venous outflow, or abscess-induced vasculitis. The first direct injury hypothesis is supported by the demonstration of neurological improvement after surgical decompression, while the second indirect injury hypothesis is supported by the rapidity of onset, irreversibility of the functional deficit in some cases, and demonstration of thrombosed vessels in a few postmortem examinations. The basis for the neurologic sequelae caused by spinal epidural abscess, however, remains incompletely understood.

The severity of neural deficit at presentation has a major influence on neurologic outcome. Factors known to affect neurological recovery include diabetes mellitus, history of rheumatoid arthritis, involvement of cervical spine, and surgical delay greater 72 hours after neurologic dysfunction. Irreversible paralysis, the most serious complication of spinal epidural abscess, continues to affect 4% -22% of patients, and about half of such cases have been attributed to delayed diagnosis and suboptimal management.

Early diagnosis, microbiological identification, and prompt antimicrobial therapy can improve prognosis.

Recurrence rates are 0%-4% with optimal therapy. Clinical and/or radiological follow-up is needed to indicate that recurrence has not developed. Relapsing infection after spinal instrumentation is not rare, and clinicians should be increasingly vigilant when following up these patients.

Spinal epidural abscess location may also affect natural history and response to therapy. Patients with lumbosacral abscess have better outcomes than patients with cervical or thoracic abscess. A dorsal spinal epidural abscess results from the endogenous seeding of the dorsal epidural fat; as the abscess enlarges, neural compression may occur early. Conversely, a ventral spinal epidural abscess may result from either endogenous seeding of the ventral epidural fat or seeding of the disc space with secondary extension into the ventral epidural space. A ventral epidural abscess is, therefore, more likely to present with systemic symptoms (e.g., fevers) before neurologic deficits, and patients with a ventral epidural abscess may present earlier in the disease process. As expected, patients presenting with sepsis have poor prognosis. Those who have a delay in diagnosis or treatment or present with persistent bowel or bladder dysfunction have the worst outcome.

Because of co-morbidities and poor premorbid status, mortality and morbidity associated with spinal epidural abscess are substantial, particularly in developing countries. About 5% of patients die from uncontrolled sepsis, or other underlying illnesses.

What causes this disease and how frequent is it?

Approximately 0.2 to 1.2 of every 10,000 people admitted to the hospital have spinal epidural abscess. The current annual incidence of epidural spinal abscess has apparently increased, probably as a result of the aging of the general population, the growing numbers of injected-drug users and patients with HIV infection, a rise in the use of diagnostic and therapeutic spinal interventions, and the improved sensitivity of neuroradiologic imaging techniques. Spinal epidural abscesses are more prevalent in patients who are 30 years of age and older, but they can affect young people, even young children. The male to female ratio is about 2 to 1.

Gram-positive cocci, including staphylococcus and streptococcus, are the most common causes of spinal epidural abscess, with S. aureus accounting for 50%-70% of cases.

Gram-negative organisms such as E. coli, Haemophilus influenza, Proteus spp, as well as enterococci and some fungi can also cause spinal epidural abscess.

Injected-drug users constitute a separate population, in which Pseudomonas spp are commonly isolated (Table I).