An epidural abscess is an infection within the epidural space anywhere in the brain or spinal cord.
Dura mater forms the inner lining of the bony skull, and under normal conditions, there is no space between skull and dura. Intracranial pressure (ICP) elevation related to infections, inflammation, or tumors opens up the epidural space and separates bone from the tissue. This newly formed epidural space may contain blood, pus or an abscess. Below the foramen magnum, the epidural space extends the length of the spine. It has 2 compartments: (1) a true space posterior and lateral to the spinal cord containing a cushioning layer of fat embedded with penetrating arteries and an extensive venous plexus, and (2) a potential anterior space where the dura adheres to the posterior surface of the vertebral body.
Epidural abscesses occur as a result of infections involving the spinal or cranial epidural space. Intracranial epidural abscesses (IEA) are complications of cranial surgery or trauma; they may also complicate otorhinolaryngological infections or other neck and thoracic procedures. Spinal epidural abscess (SEA) can have an acute and chronic presentation. This simple categorization correlates with certain clinical and laboratory manifestations, bacteriological and cerebrospinal fluid (CSF) formulae, anatomic details, and pathology.
Acute SEA is usually less than 2 weeks in duration with fever and signs of systemic inflammation from a hematogenous source. This is in contrast with subtle, afebrile and long-standing chronic SEA that has resulted from a direct extension of vertebral osteomyelitis. Both present with back and radicular pain, but leukocytosis (in serum and CSF) is more likely in the acute form and not so much in the chronic form. Acute forms are posterior to the spinal cord, but chronic forms are commonly anterior to the cord. Gross pathology is purulent and exudative in acute, but with granulation tissue in chronic.
IEA usually starts with an exogenous port of entry, either the paranasal sinuses or ears. Streptococci (usually anaerobe forms like Peptostreptococcus) and other anaerobes such as Cutibacterium (formerly Propionibacterium, part of the skin flora) are usual organisms. Gram-negative bacilli or fungi can also be a cause. They usually originate in the sinuses.
IEA can be a complication of neurosurgery, with the most likely organisms being staphylococci, especially Staphylococcus aureus, and gram-negative bacteria. Infection can also spread inward from osteomyelitis of the skull or fetal monitoring probes applied to the skull during birth.
Bacterial SEA make-up the major cause of this entity. Tuberculous, fungal, and parasitic abscesses of the spinal epidural space typically evolve more insidiously than pyogenic bacterial ones. Other than candida infections, these etiologies are more frequently encountered in tropical and subtropical resource-constrained regions of the world.
The incidence of the SEA is 0.2 to 1.2 cases per 10,000 hospital admissions. Risk factors include diabetes, intravenous drug use, chronic renal failure, alcoholism, or immunosuppression. SEA is nine-time more common than IEA and also more likely to be acute.
Although not common, IEA is still the third most common focal pyogenic intracranial infection, after brain abscess and subdural empyema. It was historically a result of head and neck infections such as sinusitis, mastoiditis, and otitis; but nowadays is mostly a complication of neurosurgical procedures.
Infection enters the epidural space by direct extension from a contiguous site or by hematogenous/lymphatic seeding from a remote site.
Contiguous infections include vertebral osteomyelitis, discitis, retropharyngeal/perinephric/paraspinal/psoas abscesses, decubitus ulcers, and persistent dermal sinus tracts. Local invasion from superficial infections can also occur following penetrating injuries, surgery, or spinal procedures. Such procedures might be as simple as a lumbar puncture or paravertebral injection, or as complex as epidural catheterization or CT-guided needle biopsy.
Approximately 25% to 50% of SEA come from hematogenous spread into the epidural space. Of those, the majority (about 15%) are from skin or soft tissues. Endocarditis and infected intravascular catheters, respiratory tract infections, urinary tract infections, abdominal infections, dental abscesses, and complications from gastrointestinal surgery are also possible.
IEA is mostly a localized lesion with a central collection of pus surrounded by a wall of inflammatory reaction. Dura is rigid and tight around the base of the skull and therefore prevents the downward transmission of the infection into spinal epidural space. SEA spreads quickly since spinal epidural space is a connected space and the infection is more of a granulation tissue rather than pus.
Blood and/or intraoperative cultures define the organisms and histopathology. Staph aureus (coagulase positive or negative) accounts for over 60% of the SEA. The polymicrobial SEA is a rarity and blood cultures are positive in most patients. Further, the pathogens identified by blood culture are almost always identical with the infectious agents cultured from abscess content.
This situation differs completely from that of IEA in which blood cultures are positive in only 10% of patients; and even if positive, they do not represent the actual causative organism. IEA and brain abscess pus often grows mixed organisms.
On initial presentation, the patient with an SEA often has vague and subtle symptoms. There must be a high index of suspicion to diagnose an SEA before it causes neurologic deficits. Localized vertebral tenderness to palpation or percussion is almost always present. This pain will inevitably become more severe and harder to treat. During the illness, usually a few days after the onset of spine tenderness, 90% of patients will develop radicular pain. Most patients will also develop fevers above 38 C. Other nonspecific findings may also be present such as generalized malaise, fatigue, headaches, irritability, or vomiting.
Heusner (1948) conceptualized the manifestations of spinal epidural abscess as progressing through four distinguishable but overlapping stages: (1) spinal ache (or back pain); (2) root (or radicular) pain; (3) weakness; and finally, (4) paralysis. The specific time between the onset of back pain and the development of neurologic deficits can be highly variable. The often rapid evolution from backache to neurologic catastrophe (or even death) forces clinicians to consider this entity in the differential diagnosis of all patients with new or changing back pain, particularly when fever and localized spinal tenderness coexist.
While early symptoms of a backache may be indolent and persist for weeks, severe back pain usually progresses to root pain within 3 to 4 days. This pain is followed by advance signs of spinal cord dysfunction within the next 4 to 5 days. The neurologic deficits at this stage are often still reversible; yet, rapid surgical intervention may be needed because progression to complete paralysis may occur within a few hours despite the chronicity of the process. Neurologic signs depend upon the level of spinal cord involvement. These signs are another factor that influences the differential diagnosis at the time of presentation.
In IEA, signs and symptoms can happen because of infection or because of slowly increasing intracranial pressure (ICP). More common symptoms are fever, headache, lethargy, nausea, vomiting, and photophobia. More common signs are papilledema, sinus drainage, cranial nerve palsies, and focal neurologic deficits. Many times, IEA is a complication of surgery or other invasive processes, and that makes the diagnosis extra difficult since the focus is on the primary process and procedure and not the complication. Anyone who has a fever and headache should be considered for this diagnosis especially if there has been a recent intracranial or head/neck surgical procedure. This is even more critical when there are sources of infection present in the head and neck region, for example, sphenoid/ethmoid sinusitis or chronic otitis media.
Lab findings for these patients are nonspecific. Patients may have a mild leukocytosis and elevated C-reactive protein. Blood cultures are positive in SEA but not so in IEA. Conventional radiography of the spine may not be helpful in SEA because osseous destruction can be absent. Bone and gallium scans and even computed tomography are equivocal and can delay definitive diagnostic testing. The gold standard for diagnosis of SEA is myelography which has mostly replaced by MRI. MRI is as sensitive and specific as CT-myelogram.
For IEA, CT or MRI are diagnostic with the latter showing more details. Nowadays, CT-guided needle aspiration is more frequently used in both entities all across the world.
SEA can be catastrophic in no time. The progression of neurologic deterioration to severe spinal cord dysfunction can occur in a matter of just a few hours, making the diagnosis and treatment imperative. The treatment usually comprises decompressive surgery and drainage of the abscess (considered the gold standard under most circumstances), eradication of the primary underlying infectious focus (if detectable), and parenteral antibiotic therapy (to target the likely etiologic agents such as coagulase-negative staphylococci (CoNS), streptococci, and gram-negative rods). Most authors treat posterior epidural abscesses by posterior laminectomy, drainage of infected and granulation tissue, and normal saline irrigation. The surgeon usually places a drainage system and continue to irrigate the abscess for several days post surgery till the infection is completely resolved. A combination of third-generation cephalosporin (ceftriaxone) with another antimicrobial showing antistaphylococcal activity (rifampicin, nafcillin, or fosfomycin) is recommended. Physicians should treat patients with antibiotic therapy for 4 to 6 weeks unless there is associated osteomyelitis, in which case they should continue the treatment for 6 to 8 weeks.
With posteriorly located abscesses which are extensive and spread out, one can use CT-guided needle aspiration instead of surgery in select cases with no major spinal cord compression signs or neurologic deficit.  Highly selected patients with no neurologic deficit or patients who have a contraindication to surgery or absolutely do not want it, might be treated medically with high dose antibiotics. One should be cautious when applying such none-surgical methods.
IEA treatment usually requires a combination of a drainage procedure and antibiotic therapy, like most abscesses. Someone can use craniotomy or simple burr holes for drainage. If it involves the dura, graft or other occlusive approaches might be helpful. After clinicians get samples and cultures, they should start antibiotics as soon as possible. Since this is a relatively rare disease, information available for antibiotic choice is mostly anecdotal. They should choose antibiotics for Streptococcus, Hemophilus, and anaerobes to cover infections that have spread from ear, sinuses or other areas of head and neck. Third-generation cephalosporins plus vancomycin plus metronidazole is one regimen. They can then change treatment based on cultures and sensitivities.  Vancomycin does not cross the blood-brain barrier; therefore, some authors advocate using Rifampin in its place while others consider a short course of steroids. Linezolid and daptomycin have also been used, but data of their efficacy is very limited. Antibiotic therapy is usually 6 to 8 weeks of intravenous agents since oral agents will probably not achieve a therapeutic level in CSF that is needed for cure. Clinicians might manage sinus-related IEA in children without cranial drainage procedures if they can do an effective sinus drainage procedure and the abscess is not showing signs of increased ICP. 
Clinicians must evaluate all patients who experience acute or progressive back pain, fever, and local spine tenderness for the possibility of the SEA. Despite its sometimes subtle and variable presentation, the diagnosis needs to be considered promptly. Not all symptoms and signs classically attributed to an epidural abscess are present in every patient; especially children may exhibit only atypical features. In patients with a chronic course, fever and systemic complaints may be minimal. Research series have noted that half of the patients with SEA are diagnosed as something else during early evaluations.
IEA can mimic any intracranial mass lesion such as primary or secondary brain tumors, abscesses, hematomas or subdural empyema. Also in differential diagnosis is meningitis, vascular inflammations, and giant cell arteritis.
With prompt and accurate diagnosis and time-sensitive management, the prognosis for both IEA and SEA is very good.
Although uncommon, an epidural abscess is a potentially devastating infection, and its neurologic complications are life-changing. The positive is that prompt diagnosis and treatment can often prevent or reverse all this.
Once diagnosed with an SEA, clinicians should treat patients with the same urgency accorded to the most critical patients. Occasionally, even with recognition and intervention, the affected patient can still suffer devastating neurological complications. The feared neurological complications of the SEA could arise from either pressure causing compression of the spinal cord or septic thrombophlebitis causing ischemic necrosis. The former mechanism is probably more common, but vascular compromise is likely responsible for the sudden deterioration of otherwise stable patients.
Physicians rarely use antibiotics alone because some patients will suffer neurologic deficit progression despite antibiotics. Complications can ensue abruptly and unpredictably, without warning. The resultant paresis or paralysis may not be reversible even if surgery is performed urgently.
Studies related to these entities are rare since these are emergency cases which require a quick course of action. However, from the information available, early recognition and treatment are critical. This requires suspicion from all healthcare providers, especially those who encounter patients first (primary care physicians, nurse practitioners and physician assistants). All should be familiar with these entities and at least consider them in differential diagnosis; therefore, clinicians may recognize the patients quickly and be able to manage them promptly. Whenever there is doubt about the diagnosis, consult a neurosurgeon as soon as possible.