Chapter 197 Spine Infection
Postoperative Spine Infection
Infectious complications of spine surgery are not uncommon and occur in 0.5% to 12% of patients.1–9 Infections can range from limited superficial wound infections or isolated discitis to more serious deep subfascial wound infections, osteomyelitis, epidural abscess, or meningitis. Postoperative infection results in an increase in health-care costs and increases the risk of poor outcomes, including persistent pain, permanent neurologic deficit, and death.1–11 Some of the risk factors associated with surgical site infections include patient age,12,13 obesity,9,14–18 diabetes,18–20 urinary incontinence,9 alcoholism, extended steroid use, tobacco use,18 poor nutritional status,12 prior infection, prior surgery, prolonged hospitalization prior to surgery,18 complete neurologic deficit,21,22 trauma,23 tumor resection,9 prior radiation therapy,24 and the presence of more than three comorbid diseases.9
The rate of postoperative infection is in large part determined by the type of operation. Surgeries without bone grafting or instrumentation have a lower rate of infection. The incidence of infection after intervertebral disc procedures is between 0.5% and 1%.25,26 There is uncertainty whether microdiscectomy with the use of an operative microscope increases the infection rate.27 The rate of infection with laminectomy without fusion is estimated to be around 1.5% to 2%.16,26 The incidence of infection is higher when grafting and instrumentation are used,28 which was first documented in a case series involving the use of Harrington instrumentation for fusion in scoliosis surgery.5 The increased risk has many components, including addition of a foreign body; lengthier, more complicated surgeries; increased blood loss; and the use of prolonged retraction. Instrumented fusion of the lumbar spine carries an infection risk of approximately 2.8% to 6%.16,18,26,29 Surgery after spine trauma carries a 10% risk of postoperative infection.23
Ventral operations have a markedly lower incidence of infection than dorsal approaches, likely due in part to the injury caused by the use of prolonged retraction in dorsal approaches. The addition of a combined anterior and dorsal approach does not seem to increase the risk of infection over that for a dorsal approach alone.23,30 Other surgical factors that have been shown to increase the risk of postoperative infection include increased blood loss (>1 L),18 use of blood transfusion,19 prolonged surgical time (>3 hours),31 multilevel surgical fusions extending to the sacrum,32 and spinal fluid leak.18 The use of drains has not been shown to increase the risk of infection.33 Participation of residents or fellows in the surgical team is not associated with an increased risk of infection.34
The optimal method of dealing with surgical site infections is prevention. Prophylactic antibiotics have been shown convincingly to decrease the rate of postoperative infection, and their use is recommended by published clinical guidelines for all spine operations.35,36 A large meta-analysis found a statistically significant decrease in the rate of postoperative spine infection in those patients given preoperative antibiotics versus controls (2.2% vs. 5.9%).37 Prophylactic antibiotics should be given prior to the start of incision, should be redosed for prolonged procedures, and should not be continued for more than 24 hours. First- or second-generation cephalosporins, such as cefazolin or cefuroxime, adequately cover the most common causes of surgical site infection and are recommended for most patients. Patients with cephalosporin allergies can be given vancomycin or clindamycin. Vancomycin should be used for patients colonized with methicillin-resistant Staphylococcus aureus (MRSA).29,35
One large prospective case-controlled study found that the rate of discitis after microdiscectomy was significantly decreased when a gentamicin-soaked collagen sponge was placed in the cleared disc space versus the rate for historical controls (0% vs. 3.7%).38 The benefit of prophylactic local antibiotics for discectomy procedures is in part due to the poor penetration of IV antibiotics into the relatively avascular intervertebral disc.39 The use of irrigation solution with antibiotics such as bacitracin and gentamycin, or dilute iodine, is widespread, but there is not good evidence demonstrating additional benefit over irrigation with just saline.35,40 Use of chlorhexidine-alcohol instead of povidone-iodine for preoperative skin cleansing has been shown to decrease surgical site infections in randomized controlled trials.41 In a Cochrane review, alcohol-based rubs were found to be equivalent to aqueous chlorhexidine-based scrubs for preoperative hand antisepsis, with aqueous povidone-iodine–based scrubs being inferior.42 Operating rooms with vertical laminar airflow have been demonstrated to decrease infections in dorsal spine fusion surgery.43 Double-gloving; frequent release of retractors to prevent ischemia; and copious, frequent irrigation are reasonable, though unproven, strategies for also minimizing postoperative infection.29,44
Postoperative wound infections can be classified as early onset (occurring <1 month after surgery) or late onset (occurring >1 month after surgery). Late-onset infections include isolated discitis and infections of instrumented fusions by indolent organisms. Staphylococcus aureus is responsible for approximately 50% to 75% of infections, followed by S. epidermidis, gram-negative organisms, and multimicrobrial infections.16,26 Late-onset infections associated with instrumentation are more likely to be caused by more fastidious organisms such as S. epidermidis, Priopionibacterium acnes, or Corynebacterium44,45
The most frequent sign of postoperative spine infection is wound drainage, which in one series was present in 93% of cases.26 Often patients with infection are discharged home after an apparently normal recovery and return because of drainage and associated swelling, tenderness, erythema, and wound dehiscence. The average time of presentation for a postoperative spine infection is 2 weeks, though it can be days for aggressive organisms, such as Clostridium perfringens, or years for indolent infections.26
New or worsening neurologic deficits, such as numbness, urinary or bowel dysfunction, weakness, or paralysis, are ominous signs and should raise suspicion for epidural hematoma or abscess. Postoperative spinal epidural abscess is a neurosurgical emergency that can lead to rapid decline if not promptly diagnosed and treated with antibiotics and surgical drainage.46
Laboratory studies are an important adjunct in diagnosing infection, especially when the wound site does not show obvious signs. Patients often demonstrate leukocytosis, with or without associated neutrophilia, but a large portion of cases will have a normal white blood cell count and differential. Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels are much more sensitive, though nonspecific, markers of inflammation and are almost always significantly elevated with infection. Care must be taken in interpreting these levels since they are initially elevated after any surgery. In uninfected postoperative patients, CRP typically peaks on postoperative day 2 and returns to normal levels between days 5 and 14. ESR peaks around day 5 and can remain elevated for weeks after surgery.47,48 ESR and CRP levels are particularly helpful in diagnosing late-onset infections in patients with minimal signs or symptoms besides worsening or persistent pain, such as postoperative discitis.
Blood cultures should be obtained in all patients in whom a surgical site infection is suspected. Cultures of the skin or drainage site are rarely helpful because they culture normal skin flora. Needle aspiration cultures are much more reliable, and intraoperative cultures are the best option. Since most patients will need surgical debridement, the best course of action is to wait for the results of intraoperative cultures before starting antibiotics. For late-onset infections, it is especially important to continue cultures for at least 7 to 15 days to increase the chance of growing indolent organisms such a P. acnes.45 When surgery is not indicated, as in suspected isolated discitis, and blood cultures are negative, then a percutaneous biopsy should usually be obtained to confirm the diagnosis and guide treatment.
Plain radiography, CT, and MRI can be useful in diagnosing postoperative infection, but their utility is often clouded by the similarities in imaging findings between infection and normal postoperative inflammation. Radiographs can demonstrate retained foreign bodies, disc space narrowing that develops with discitis after 7 to 10 days, erosion of vertebral end plates, or vertebral collapse.49 Loosening of implants can also be revealed, which is often a sign of late-onset infection.50 CT reveals many of these same features but with superior anatomic detail.
Contrasted CT and MRI both have excellent sensitivity for identifying fluid collections, but it can sometimes be difficult to differentiate between abscess and postoperative seroma or hematoma. Either modality helps differentiate superficial from subfascial infections. Contrasted MRI is the best study for determining postoperative epidural abscess.51 MRI with contrast is also the study of choice for diagnosing discitis and osteomyelitis. T1-weighted images reveal hypointensity of the disc and vertebral body. On T2-weighted images, the involved bone and disc are hyperintense due to edema and the involved disc demonstrates a loss of the intranuclear cleft.52 Areas of inflammation enhance with contrast administration.49 Unfortunately, many of the same MRI signal findings are seen with normal postoperative changes.53 Radionuclide studies, such as gallium/bone (technetium) scintigraphy, can be useful when MRI is contraindicated or nondiagnostic due to artifact from implants.49,54
Treatment consists of targeted antibiotics and surgery in nearly all cases, though there are a few exceptions. There is a significant difference between the treatment of spine infections in the presence of grafting and instrumentation versus simple decompression procedures. The treatment is also different for early-onset infections versus that for late-onset infections in the setting of instrumentation. Some very superficial wound infections or stitch abscesses can be treated with empirical antibiotics alone, but there must be no evidence of deeper infection or significant systemic signs of infection. Ensuring adequate nutrition is vital for the successful treatment of all postoperative infections.12,50
Isolated postoperative discitis often presents later than other postsurgical infections with worsening back pain, elevated ESR/CRP, and characteristic findings on radiograph and MRI. Discitis can frequently be treated successfully with 4 to 6 weeks of IV antibiotics alone. This requires obtaining bacterial diagnosis via blood cultures or percutaneous biopsy. Spontaneous fusion across the disc space usually occurs after resolution of the infection. Many surgeons recommend bracing to minimize pain. Surgery for debridement of the disc space is indicated in the setting of new or worsening neurologic deficits, significant associated infection (especially epidural abscess), and progression of infection and bony involvement or worsening pain despite antibiotics.
Surgical debridement is warranted for most other surgeries that do not involve bone graft or instrumentation. This is especially true for subfascial infections. Exploring below the fascia is recommended for all but the most superficial of infections. All necrotic, infected, and foreign material, such as sutures, must be completely debrided. Cultures should be obtained and sent for Gram stain and aerobic, anaerobic, and fungal cultures. Once culture specimens have been sent, empirical broad-spectrum antibiotics, such as vancomycin and a third-generation cephalosporin, can be started. The wound should be copiously irrigated with large volumes, with many institutions using 9 L of bacitracin-containing irrigation.26,50 Pulse-lavage irrigation may improve debridement.
Primary closure over a drain can frequently be used for more superficial infections; however, deeper infections should usually be left open to heal via secondary intention or with a delayed closure. Vacuum-assisted closure (VAC) dressings have been increasing in popularity and are purported to decrease nursing requirements and aid in healing, but no good trials have yet proved their benefit over traditional gauze packing.55–57 Repeat debridement at 48 to 72 hours can be beneficial for treating septic or immunocompromised patients or in the setting of infections caused by multiple organisms or those associated with extensive myonecrosis.26,29,50 Tailored IV antibiotics are continued for 4 to 6 weeks.
Infections after instrumented spine fusions require the same aggressive debridement and irrigation. For early infections, almost all published reports advocate leaving the spinal instrumentation and viable bone graft in place in order to maximize the chance of fusion.1–3,24,26,29,30,32,50,58,59 Loose instrumentation and nonviable bone graft should be removed. Surgical wounds should usually be left open to heal by secondary intention or with delayed closure. Repeat irrigation and debridement is occasionally needed for reasons listed earlier. Levi et al. reported success using an irrigation-suction system for postoperative care.30 IV antibiotics selected based on culture and sensitivies are continued for at least 6 weeks. In addition, Kowalski et al. found a significant decrease in the late recurrence of infection in patients treated with oral suppression therapy for at least 6 months after an initial course of IV antibiotics.24
Late infections after instrumented fusion are usually the result of more fastidious organisms that are capable of creating a glycocalyx covering on implanted hardware that is resistant to antibiotics and the normal immune response.60 Surgical treatment involves debridement, irrigation, and removal of hardware.24,60,61 Fortunately, in most cases bony fusion has already occurred, as assessed on CT or intraoperatively, and the hardware is no longer needed for stabilization. Patients with late infections who have implants left in place have a significantly higher risk of treatment failure.24 Postoperatively, patients should be monitored closely for evidence of pseudarthosis or deformity that would warrant repeat instrumented fusion. After removal of hardware, debridement, and irrigation, the wound can typically be closed over a drain, and appropriate antibiotics are continued for 6 weeks,24,61 although one group reported good outcomes with just 2 days of IV antibiotics followed by 7 days of oral antibiotics.60
Spontaneous Spinal Infections
Discitis
Pyogenic discitis is a bacterial infection of the intervertebral disc that is frequently associated with involvement of the adjacent vertebral end plates (spondylodiscitis). Incidence is estimated to be between 0.2 and 2.4/100,000 each year, with two peaks in age distribution: one in early childhood and another between ages 60 and 70 years.62,63 The relatively high proportion of children with pyogenic discitis is likely related to the anatomy of the disc space during development. Children still have a vascular supply into the nucleus pulposus of the disc, which allows septic emboli to lodge with the disc. The vascular supply in adults only reaches the anulus fibrosus. In the adult population, more men are affected than women.64 Risk factors for developing discitis include invasive procedures, diabetes (11–31%), malignancy, IV drug use (IVDU), immunosuppression, alcoholism, renal failure, and cirrhosis.62,63,65
The majority of cases are caused by S. aureus, with gram-negative rods, Streptococcus, and Enterococcus being the next most frequently involved organisms.63,64 Gram-negative organisms are more common when associated with diabetes, immunocompromise, infections of the genitourinary or gastrointestinal tracts, or IVDU.65 Tuberculosis (TB) and brucellosis are atypical bacterial infections that can cause discitis in endemic regions and in at-risk populations. Fungal discitis is rare but should be considered in patients who are critically ill, immunosuppressed, taking multiple antibiotics, or have an indwelling catheter.66 The most frequent location is the lumbar spine (60%), followed by the thoracic (30%) spine and the cervical spine (10%).62,64
In 1936, Milward and Grout were the first to describe the clinical and radiographic characteristics of interspace infection after the inadvertent introduction of microorganisms into a disc space during a lumbar puncture.67 More than 90% of patients complain of back or neck pain.62 This pain often is not relieved by medications or recumbency. Radicular pain symptoms are not uncommon. Guarding against movement and a positive straight-leg raise may be present. Fever is present in approximately 60% to 70%.63 Neurologic deficits, especially weakness, should raise concern for more a extensive infection such as an epidural abscess. In children, discitis often presents as a refusal to bear weight or walk.68 The time from onset to diagnosis is often months. ESR and CRP are elevated in most patients, but leukocytosis is present in less than half of cases.69
The differential diagnosis for discitis includes other infections of the spine and adjacent structures, trauma, osteoporotic fracture, degenerative disc disease or acute herniation, metastatic disease, and inflammatory spondyloarthopathies. Discitis is often also associated with bacterial endocarditis (3.7–15%). Echocardiography is recommended for patients with spontaneous discitis. Back pain in the setting of endocarditis or bacteremia should lead to an evaluation for discitis.70
MRI with contrast is the image modality of choice for diagnosing spontaneous discitis, with a sensitivity and specificity greater than 90%. If an MRI is not possible, the next most sensitive studies are radioisotope scans followed by CT with contrast. Radiographs also show typical changes such as disc space narrowing and end-plate erosion, but these changes usually take several weeks to develop.49 CT is particularly helpful for guiding percutaneous biopsy.
If possible, antibiotics should be held until cultures can be obtained. Blood cultures are positive in approximately half of discitis cases.64 To maximize yield, three culture specimens taken at different times and locations should be obtained, ideally when the patient is febrile. If after 48 hours there is no growth, a CT-guided percutaneous biopsy should be obtained, which increases the yield to between 60% and 70%.64 Samples should be taken from both adjacent vertebral end plates and the disc itself, and the disc space should be rinsed and aspirated. Biopsy samples are sent for histopathologic studies, aerobic, anaerobic, fungal, and mycobacterial cultures, and stains.62 A second percutaneous biopsy has been shown to increase yield when the first biopsy is negative.69 Occasionally open biopsy is required to obtain a diagnosis.
Approximately three-quarters of spontaneous pyogenic discitis cases can be treated nonoperatively with IV antibiotics tailored according to the results of the cultures and sensitivities.62 There are no good studies comparing various antibiotic regimens, duration of antibiotic therapy, or the role of oral antibiotics after administration of IV antibiotics. Traditionally 4 to 6 weeks of IV antibiotics was recommended, but recent reports suggest there is a decreased risk of recurrence if antibiotics are used for 12 weeks, often with a switch to oral antibiotics after 6 weeks.62,65 Spinal immobilization for pain control, using bracing or short-term bed rest, is recommended by most surgeons.
Response to therapy is demonstrated by diminishing pain, resolution of fever, and progressive decrease in CRP levels.48 Radiographs should be obtained at regular intervals after initiation of treatment and should show sclerosis and osteophyte formation by 3 months if healing is occurring.65 Patients generally progress to fusion over 6 to 12 months. Repeat MRI is generally not helpful and often initially appears worse than pretreatment scans.71 MRI should usually be reserved for patients with new or worsening neurologic symptoms to rule out expanding abscess.
Surgery is reserved for patients with neurologic deficits, especially those with associated epidural abscess and spinal cord compression, for clinical failure after conservative treatment, for treatment of spinal instability or correction of deformities, for obtaining a diagnosis by open biopsy after failed percutaneous biopsy, and occasionally for persistent pain.72
Recurrence rates after a course of antibiotics are generally around 10% (0–16%).62,69 Chronic pain is the most common residual complication. Functional impairment and neurologic deficits occur in a minority.64 Mortality is generally low and usually related to associated sepsis, endocarditis, or underlying disease.63,69
Vertebral Osteomyelitis
Infection of the bones of the spinal column—vertebral osteomyelitis—can occur after trauma, as a result of direct manipulation during surgery, via contiguous spread from adjacent structures, or via hematogenous spread from distant sources.73 Vertebral osteomyelitis is relatively rare, with an estimated incidence of 2.4/100,000 people. Older adults are more likely to be affected, with the incidence increasing from 0.3/100,000 for those younger than 20 years of age to 6.5/100,000 for those older than 70.74 There is a male predominance that also increases with age.74,75
Patients with vertebral osteomyelitis usually have predisposing factors, with the most common being diabetes mellitus, end-stage renal disease requiring dialysis, sepsis, endocarditis, cancer, HIV infection, immunosuppression, alcoholism, and IVDU.73,75,76 Urinary tract infections followed by skin infections were the most common sources of infection.75
A vast majority of cases of pyogenic spine osteomyelitis involve the vertebral body, with only 3% to 12% involving the dorsal elements of the spine.77 Infection generally begins at the highly vascular end plates. Most cases involve two or more contiguous vertebral bodies and the intervening disc. Vertebral osteomyelitis occasionally manifests as the collapse of an isolated vertebral body. The lumbar spine is the most commonly involved (58%), followed by the thoracic spine (30%) and cervical spine (11%).75 Associated epidural abscesses (17%), paravertebral abscesses (26%), and disc space abscesses (5%) are frequent.73
The most common causative organism is S. aureus. Escherichia coli is the most commonly reported gram-negative organism and is especially associated in cases when genitourinary or gastrointestinal infections are the source.75 S. aureus is the most frequent organism in IV drug users, but they also have an increased frequency of Pseudomonas aeruginosa76 In many developing regions of the world TB is a frequent cause of chronic vertebral osteomyelitis.
There is often a substantial delay in the diagnosis of vertebral osteomyelitis due to the nonspecific nature of its presenting symptoms. In one large series the mean time to diagnosis was 1.8 months, with only a quarter of cases being diagnosed in less than a month. The same study revealed that on initial presentation only a quarter of the patients had vertebral osteomyelitis considered in the differential diagnosis. The majority of patients (86%) present with several weeks of worsening back or neck pain.73 Fever has been reported in 35% to 60% of patients.73,78 Neurologic deficits are present in approximately a third of cases, and only around one fifth of patients have localized tenderness.78 Rapidly worsening neurologic deficits and paralysis should raise the concern for associated spinal epidural abscess. Acute worsening of pain is often associated with vertebral collapse. Back pain in the setting of bacteremia, such as with endocarditis, should always lead to an evaluation for vertebral osteomyelitis. The differential diagnosis for vertebral osteomyelitis includes other localized spine infections, osteoporotic or traumatic fractures, spondyloarthopathies, degenerative disc disease, herniated disc, metastasis, and infections such as pancreatitis and pyelonephritis.79
Leukocytosis is present in approximately two thirds of cases, and there is an associated neutrophilia in about a third of cases.80 ESR and CRP measurements are much more sensitive to the presence of inflammation and are elevated in almost all cases of vertebral osteomyelitis.48,80 Serial CRP is more accurate than ESR for gauging response to therapy.48 Blood cultures should be obtained in all patients with suspected vertebral osteomyelitis because 58% (range, 30–78%) of cultures will be positive.75 Identifying the organism via blood cultures often obviates the need for more invasive procedures.
MRI is the best imaging modality for evaluating for vertebral osteomyelitis, with a sensitivity of nearly 100% and accuracy over 90%.81 Findings associated with osteomyelitis include a low intensity on T1-weighted images with loss of the usual hyperintense signal of fat in the bone marrow, hyperintensity on T2-weighted images in the bone or adjacent disc and soft tissues indicative of edema, and enhancement with contrast of the end plates and associated abscesses. There is usually a loss of the intranuclear cleft of involved discs. End-plate destruction is a late finding.81
Combined gallium/bone scintigraphy is the most useful radionuclide study for diagnosing vertebral osteoarthritis in patients unable to have MRI due to incompatible implants or in whom MRI imaging is nondiagnostic. 2-[18F] Flouro-2-deoxy-d-glucose positron emission tomography (FDG-PET) is a promising new modality that also has a high reported accuracy. In contrast, labeled leukocyte imaging is not considered useful in the diagnosis of vertebral osteomyelitis.54,81
The overall goal of treatment of vertebral osteomyelitis is to eliminate the infection while maintaining neurologic function and spinal stability. A majority of patients are able to obtain this goal without surgical intervention with antibiotics and spine immobilization via bracing or bed rest. The overall rate of surgery has likely been rising, and a recent systematic review found that 42% of patients had some form of surgical intervention. The reasons for surgery include open biopsy, spine stabilization (23%), drainage of associated abscess (21%), decompression of the spinal cord (13%), and correction of deformity after infection had been cleared (2%).75 The use of instrumentation for stabilization in patients with acute infection does not appear to increase the risk of relapse when the patient has an appropriate course of antibiotics.82
Antibiotics are chosen based on the results of culturing a causative organism and should be withheld prior to obtaining the culture results if possible. If blood cultures are negative, CT-guided percutaneous biopsy or open biopsy should be obtained. Aerobic, anaerobic, fungal, and mycobacterial culture specimens should be sent. Histopathology is useful in identifying granulomas that might be indicative of tuberculosis or brucellosis.79 Antibiotics are usually continued for at least 6 weeks, but recommended courses ranging from 4 weeks to 3 months have been reported.83 Longer courses should be considered in patients with complicated infections or implanted hardware. Antibiotic courses of less than 4 weeks have a higher incidence of relapse.84 In certain instances oral regimens have been used successfully after a course of IV antibiotics. One randomized trial found similar outcomes in patients treated with a combination of an oral fluoroquinolone and rifampin and in those treated with IV antibiotics.85
Patients should be monitored closely for failure of therapy. Failure of symptoms to improve and persistent CRP elevations at 4 weeks indicate likely treatment failure.48 MRI obtained after starting treatment is a poor predictor of treatment outcome and should be reserved for patients in whom a change in symptoms occurs that might suggest new or worsening abscess.79 Relapse occurs in 1% to 22% of cases and is more likely in patients with recurrent bacteremia, a chronically draining sinus, or a paravertebral abscess.73
The systematic review by Mylona et al. found a mortality rate of 6%, with most deaths attributable to associated sepsis. Approximately a quarter of patients had a significant decrease in quality of life. The most common complications reported were chronic pain (28%), weakness (16%), and dysfunction of the bowels or bladder (7%).75 Predictors of worse outcome include motor weakness or paralysis at presentation, delayed diagnosis (>2 months), and acquisition of the infection in the hospital.73
Spinal Epidural Abscess
Spinal epidural abscess (SEA) is a relatively rare but extremely important clinical condition involving supportive infection in the epidural space of the spinal canal. SEA is considered a neurosurgical emergency because severe neurologic decline or death may become unavoidable if diagnosis and treatment are delayed. The mortality rate from SEA has been reported to be between 4.6% and 31%.86
Incidence reports from longer than 2 decades ago estimated that 0.18 to 1.96 cases of SEA occur per 10,000 hospital admissions.87,88 However, evidence suggests that the incidence has increased as the number of susceptible patients with known risk factors, such as IVDU and HIV infection, has increased.46,86,89 This apparent increase in incidence may in part be due to the fact that the diagnosis of SEA is also made easier due to the advances in medical imaging. The male-to-female ratio was previously reported to be approximately 1:1,87,88 but a large meta-analysis in 2000 revealed a ratio of 1:0.56.90 This predominance is likely related to the higher incidence of trauma, alcoholism, IVDU, and other risk factors in men. SEA is more common in adults, with the majority of cases occurring from ages 30 to 60, but it can occur with any age group, with the youngest reported case being a 10-day-old patient.90 The most common location for SEA is the thoracic spine, followed by the lumbar and lumbosacral regions.
Most patients with SEA have an underlying predisposing condition such as diabetes, end-stage renal disease with dialysis,91 cirrhosis, medical immunosuppression for transplant, chronic steroid therapy, HIV, malignancy and related chemotherapy, alcoholism, or previous trauma or spine intervention.86,88,90,92 Approximately half of SEA cases are caused by hematogenous spread from a focus of infection, which can be either arterial or via the paravertebral venous plexus.93 The most common source of infection is skin abscesses.90 Other commonly reported sources include IVDU, indwelling venous or arterial catheters, dental abscesses, bacterial endocarditis, urinary tract infections, and respiratory infections. Iatrogenic introduction of disease via surgery, epidural anesthesia, or corticosteroid injection, among other causes, is another important source for the introduction of bacteria. Finally, contiguous spread to the epidural space can occur and has been reported from such sources as adjacent psoas abscesses, decubitus ulcers, abdominal infections, pyelonephritis, mediastinitis, and pharyngeal abscesses.
The most common microbial agent in SEA is S. aureus, which causes two-thirds to three-quarters of all cases.46,90,94 Of concern is the increasing prevalence of MRSA, which in some reports represents almost 40% of abscesses.95 Overall, aerobic gram-positive organisms, such as S. epidermidis, Streptococcus viridans, Enterococcus, and Propionibacterium, among others, account for nearly 80% of SEA cases.
Coagulase-negative staphylococci, such as S. epidermidis, are more common in patients who have undergone invasive spine procedures or who have implanted foreign bodies. Gram-negative organisms, such as E. coli, Enterobacter, Salmonella, Proteus, Serratia, and Pseudomonas, among others, are more likely to be involved when the source of the infection is gastrointestinal infection or urinary tract infection. Pseudomonas is more likely in patients with IVDU.96 Multiple organisms can be found in up to 10% of abscesses. Anaerobic cultures should always be obtained since anaerobic bacteria, such as Bacteroides and Peptostreptococcus species, also rarely cause SEA.87,97 Other causes of SEA include atypical bacterial infections, such as TB, brucellosis, and actinomycosis, for which acid-fast bacillus staining and extended cultures may be necessary; fungal infections, such as aspergillosis, in patients who are immunocompromised; and even parasitic infections, such as echinococcosis and dracunculiasis.90
Diagnosis begins with recognition of the clinical presentation. Heusner’s classic description of the presentation of SEA in 1948 describes four stages: (1) severe back pain, local tenderness and fever; (2) signs of spinal irritation such as Kernig sign, neck stiffness, and radicular pain; (3) development of neurologic deficits such as weakness, fecal or urinary incontinence, and sensory deficits; and (4) progression of weakness to paralysis.98 Most patients with SEA do not present with such a characteristic course and often initially present with only complaints of isolated back pain. For this reason it is very common for the diagnosis of SEA to be initially missed. The most common signs and symptoms are back pain (71%) and fever (66%),90 and the combination should always raise the possibility of SEA. Atypical signs and symptoms such as localized tenderness to percussion, thoracic radicular pain, and pain with recumbency should also raise red flags. Symptoms of systemic infection such as chills, night sweats, or sepsis may be present.99 New-onset neurologic deficits are more common with cervical and thoracic SEA and need to be rapidly evaluated due to the possibility of progression. Approximately one third of patients present with some degree of paralysis.90
In a patient with back pain, the addition of laboratory tests to identify systemic signs of inflammation can greatly enhance screening for pyogenic spine infection. The leukocyte count, ESR, and CRP are often elevated, though a normal lab value by itself should never be used to rule out the possibility of SEA. The incidence of leukocytosis is approximately 68% to 78%.100,101 Approximately 94% to 100% of patients have an elevated ESR.90,100 Leukocytosis and an elevated ESR are relatively nonspecific symptoms and must be interpreted in the setting of the patient’s condition as a whole. All patients thought to have SEA should also have blood cultures drawn as this can help with the diagnosis of SEA as well as identifying the offending pathogen.
MRI findings in SEA reveal an epidural mass that is hypointense to isointense on T1-weighted images and hyperintense on T2-weighted images. The abscess usually enhances with contrast administration, often as a linear rim enhancement surrounding a nonenhancing core that represents purulent material.102 More heterogeneous enhancement may be present if the abscess has more of a phlegmon consistency as opposed to liquid pus.44 MRI is also excellent for identifying other conditions that may mimic SEA, such as spinal tumors, transverse myelitis, spinal cord infarction, or intervertebral disc herniation.
Radiographs and CT are useful for evaluating for adjacent osteomyelitis. Bony erosion or destruction may be seen on radiographs after 4 to 6 weeks of infection; however, delineation of bone involvement is appreciated much better with CT imaging. CT may show evidence of inflammation, such as stranding in the paravertebral soft tissues. After contrast administration, SEA may be identified as an enhancing epidural mass. However, there are reports of CT imaging alone missing a relatively high proportion of SEAs.103 CT imaging can be useful for surgical planning, especially in deciding whether fusion and instrumentation will be needed after surgical debridement.
If MRI is not possible, CT-myelography can be a very useful diagnostic substitute. Prior to widespread availability of MRI, myelography and then CT-myelography were the gold standards for radiologic diagnosis of SEA. With CT-myelography contrast is directly injected into the thecal sac and an epidural mass can be identified as blockage of flow from above or below, depending on where the puncture was performed. In a direct comparison study, MRI and CT-myelography were found to be equally sensitive (91% vs. 92%).104 CT-myelography is less specific than MRI and provides less information about the characteristics of the epidural mass. Moreover, CT-myelography requires an invasive procedure that can introduce infection or spread an epidural infection to a subdural space, causing a subdural empyema or meningitis.105 Since SEA in the upper cervical spine is relatively rare, a lateral C1-2 puncture is often recommended to limit the chance of traversing the epidural collection during the myelogram. A myelogram also allows for the evaluation of CSF, which reveals associated meningitis in up to 15% of patients. However, most authors recommend against lumbar puncture for the purpose of CSF examination alone in the evaluation of suspected SEA due to the low specificity and risk of seeding infection.51,90,104–106
Many conditions can present with back pain and signs and symptoms of inflammation in a manner similar to SEA. A large percentage of SEA cases are still initially misdiagnosed, leading to a delay in treatment. In a large meta-analysis, the most common initial misdiagnosis was intervertebral disc herniation, followed by meningitis, vertebral osteomyelitis, sepsis, endocarditis, and spinal tumors.90 The differential diagnosis for SEA should also include epidural metastasis, acute transverse myelitis, subdural empyema, intramedullary abscess, epidural hematoma, autoimmune spondylitis, discitis, infections of adjacent structures (pyelonephritis, psoas abscess, etc.), vascular malformations, subarachnoid hemorrhage, and lymphoma. Acute transverse myelitis is more common than SEA and typically presents with rapidly progressing neurologic deficits without significant back pain. A history of a recent viral illness would also make acute transverse myelitis more likely.
The next step in the diagnosis of SEA after obtaining radiologic imaging is to obtain a culture of the organism causing the infection. Obtaining cultures prior to starting antibiotics is imperative to ensure the highest yield; however, this is not always possible when the patient is frankly septic. Antibiotics given prior to culture or biopsy can result in a failure to isolate a bacterial source and lead to the patient being unnecessarily treated with broad-spectrum antibiotics. The easiest method for isolation of the bacterial source involves obtaining blood culture specimens, which should be obtained from multiple sites and at different time points. Ideally, at least three separate blood specimens are obtained while the patient has spiking fevers. The yield from blood samples in the setting of SEA is between 30% and 60%.44
If blood cultures are negative, then percutaneous needle biopsy with either fluoroscopic or CT guidance should be performed.107 If the biopsy is unrevealing, a repeat percutaneous biopsy or open biopsy may be necessary. Antibiotics should be withheld, even in a patient with rapidly progressing neurologic symptoms, until an adequate sample can be obtained at the time of surgery. All biopsy samples should be sent for aerobic, anaerobic, mycobacterial, and fungal stains and cultures. Occasionally extended periods of incubation are necessary for fastidious organisms.
All patients with SEA need antibacterial treatment. After obtaining appropriate cultures or biopsy specimens, the patient is usually started on broad-spectrum antibiotics that are tailored once the species and antibiotic sensitivities are known. The initial antibiotic regimen should have activity against S. aureus, the most common organism, should be tolerable for weeks of therapy, and should have good bone penetration due to the frequency of adjacent osteomyelitis.108 Many regimens of empiric antibiotics have been reported in the literature.90 The combination of a synthetic penicillin with activity against Staphylococcus, such as nafcillin, with a third- or fourth-generation cephalosporin for gram-negative rod coverage is one option. Vancomycin should be substituted for nafcillin in areas with a high prevalence of MRSA or if the patient is allergic to penicillin.109 Antipseudomonal coverage should be considered in patients with a history of IVDU.96 Duration of antibiotic treatment is usually at least 6 weeks and sometimes longer in patients with extensive bony involvement or immunocompromise. Consultation with an infectious disease specialist is highly recommended for determining the best antibiotic regimen and the length of treatment and for monitoring for adverse effects of the antibiotics selected.
Surgical decompression with debridement and drainage of infected material is the other cornerstone of treatment of SEA. Traditionally, the recommendation was that all patients with SEA should undergo urgent surgical decompression and drainage. The guiding principle was “ubi pus, ibi evacua.”110 There are numerous reports of deterioration when surgical decompression was delayed. This principle still holds for most patients; however, there have been a number of reports of good outcomes in carefully selected patients who have received antibiotics or antibiotics plus percutaneous drainage without surgical decompression. Rigomonti et al. described the use of nonoperative treatment in patients in whom the causative organism had been identified and who had minimal or no neurologic symptoms, medical comorbidities precluding surgery, extensive spine involvement, or stable paraplegia. Overall they found that two thirds of these patients had a good outcome without surgical intervention.46 Since then, a number of other nonrandomized comparisons have been conducted of nonoperative versus operative management in similarly selected groups of patients that have found conflicting data, with some groups finding that nonoperative management can provide equivalent results as long as close monitoring is performed,111–113 as opposed to a report by Curry et al. that indicated worse outcomes in patients treated conservatively.114 The most important determinant of the need for surgery is the presence of neurologic deficits, which if present, and not long standing, warrant urgent decompression. Any patient being treated nonoperatively needs to be monitored extremely closely for the development of neurologic deficits, which occur in approximately 20% of patients,115 that would require urgent surgical intervention.
Surgical approach is dictated by the location of the abscess either ventral or dorsal to the spinal cord. The majority of SEAs are located dorsally88 and are managed with laminectomy, drainage of pus, debridement of infected material, and copious irrigation, with the option of postoperative drainage or suction-irrigation.116 Care must be taken not to cause instability by extending the laminectomy too far laterally. The possibility of concomitant subdural abscess should be considered if the dura appears tense and nonpulsatile after evacuation of the epidural pus.117 Laminotomy for the drainage of dorsal abscesses has also been described, which may be especially beneficial in children to minimize the risk of future spinal instability.118,119 Monofilament suture is recommended for closure.
An SEA located ventrally to the thecal sac usually requires an anterior approach, especially since concomitant discitis or osteomyelitis is often present and requires debridement as well. A ventral liquid pus collection, as opposed to a more solid, granulomatous collection, can sometimes be drained via a dorsal approach, especially below the conus medullaris; however, the great majority of cases require an anterior approach to achieve adequate decompression and drainage. Anterior approaches require graft with or without placement of instrumentation. The risk of infection of graft material or instrumentation is minimized with adequate debridement of infected bone and tissue. Numerous studies have shown that infection does not preclude the use of autograft, allograft, or instrumentation when combined with adequate debridement and appropriate courses of antibiotics.120–123
Cervical SEAs tend to become symptomatic more rapidly, have a higher risk of severe neurologic deficits, and an overall higher mortality rate.124 The severity of presentation is due to the smaller epidural space in the cervical region. Cervical SEAs are more likely to be ventrally located and have associated discitis and osteomyelitis for which an anterior approach with resection of disc and bone and reconstruction with graft with or without instrumentation is warranted.
Thoracic epidural abscesses also often present with significant neurologic compromise and present a technical challenge due to the difficulty of approach to a ventral SEA. The region involved dictates the approach. The cervicothoracic junction is particularly difficult to access. Partial sternotomy or manubrial resection44 and transpedicular, lateral extracavitary,125,126 or parascapular extrapleural127 approaches have all been used to access this region. Anterior approaches in this region often necessitate concurrent dorsal instrumentation and fusion to prevent instability.122,128 Anterior approaches to the midthoracic spine include thoracotomy,44 lateral extracavitary,125,126 or the retropleural approach.129 Anterior approaches to the lower thoracic spine can be accomplished via a thoracoabdominal approach.
Extremely large SEAs involving the entire length of the spine have been reported. Medical management is often recommended for multisegmental SEA; however, there are case reports of good outcomes with extended laminectomy and surgical debridement of cervicothoracolumbar SEAs.127
Outcomes for patients with SEA have been improving since Walter Dandy’s 1926 review, which reported an 81% mortality rate.130 A comprehensive meta-analysis found that mortality has decreased from around 34% during the period of 1954 to 1960 to 15% between 1991 and 1997. Moreover, the number of patients experiencing complete recovery has improved from 28% to 41% in the same time periods.90 Mortality from SEA is likely now closer to 5% to 10%, usually caused by sepsis, meningitis, or underlying disease.95 The best predictors of outcome are neurologic status at presentation and duration of neurologic deficits.88,103,104,131 Patients with paralysis of more than 48 to 36 hours’ duration have very little chance of recovery. Permanent paralysis still occurs in approximately 4% to 22% of patients.95 Location of the SEA in the cervical or thoracic region, older age, significant cord compression, and delay of diagnosis also negatively affect the overall prognosis.115 Unfortunately, an estimated one half of cases of SEA are still initially misdiagnosed, emphasizing the importance of clinical suspicion to aid in the early diagnosis of SEA prior to permanent neurologic injury.95
Spinal Subdural Empyema
Spinal subdural abscesses are localized infections inside the dura surrounding the spinal cord with or without associated meningitis. No estimate of incidence is available, and the total number of patients presented in case reports is less than a hundred.115,117,132 The vast majority of reported cases are caused by S. aureus, although a number of other gram-positive cocci, gram-negative rods, anaerobes, and tuberculosis microbes have been reported as causative agents. The lumbar region is the area most likely to be involved, followed by the thoracic and cervical regions.132 Cases of simultaneous spinal and cranial subdural empyemas have been reported.133 The reported cases reveal an approximately equal distribution between men and women, with the most common age range being between 50 and 70, though all ages can be affected.132
Patients with spinal subdural abscesses have many of the same predisposing factors as patients with SEA. The pathogenesis is thought to be most commonly hematogenous spread from a distant source, such as a skin abscess. Other sources include direct extension from infected CSF, introduction of infection during surgery, direct inoculation during lumbar puncture or spinal anesthesia, and occasionally via a dermal sinus tract.134,135 Injury to the spinal cord can be caused by direct compression, associated meningitis, or inflammation of the vessels of the spinal cord, causing thrombosis or hemorrhagic infarction.
Clinical signs and symptoms, as well as progression of neurologic deficits, are similar to SEAs except that subdural abscesses are more likely to present with signs of meningeal irritation and are less likely to demonstrate localized tenderness. The most common symptoms are of back pain and fever, followed by neurologic deficits. Most reported cases presented with symptoms developing in a subacute fashion (1–8 weeks).117 Spinal subdural empyema is often initially misdiagnosed for one of the more common conditions on the differential diagnosis, including transverse myelitis, epidural hematoma or abscess, vertebral osteomyelitis, discitis, or intradural tumor.
The best radiologic tool for diagnosing a subdural empyema is MRI with contrast because of its sensitivity, noninvasiveness, and ability to clearly identify the location and extent of the abscess.136 CT-myelography may be useful when the patient is unable to undergo an MRI, but there are reports of difficulty in differentiating a subdural abscess from an epidural abscess with a CT-myelogram.137,138 There is often minimal or no involvement of bone, making radiographs and noncontrasted CT less useful.
The management of subdural empyemas involves urgent surgical drainage and IV antibiotics. The surgery usually involves laminectomy, durotomy, drainage of pus and debridement of infected tissue, and copious irrigation. The dura is often found to be tense and nonpulsatile. If an extended area is involved, multiple separate laminectomies might be used to minimize the risk of future instability. Aerobic, anaerobic, and mycobacterial cultures should be obtained to guide antibiotic treatment. Though spinal subdural empyemas caused by fungal infection have not been reported, fungal cultures should probably also be obtained. Antibiotics should only be started after obtaining cultures, unless the patient is frankly septic. Intraoperative ultrasound can aid in identifying the true extent of the abscess and ensuring complete drainage.139 Associated epidural abscess or osteomyelitis should also be debrided, and, if such an infection is present, the dermal sinuses should be excised. Most patients are found to have frank pus at surgery, even if they presented with chronic symptoms (>8 weeks of symptoms).117
The importance of surgical drainage is illustrated in the review by Bartels et al. of 44 patients, in which only one of five patients treated conservatively survived, in contrast to the 81% survival rate of patients treated with surgery in addition to antibiotics. Moreover, in the surgical group, approximately 30% of patients made a complete recovery, with the remainder of survivors showing improvement.117 Broad-spectrum IV antibiotics, similar to the regimen used for SEA, are started once cultures are obtained and narrowed appropriately once species and sensitivities are known. Antibiotics are continued for at least 6 weeks. Spinal subdural empyemas have a high mortality rate if not properly managed, but with prompt diagnosis, surgical drainage, and appropriate IV antibiotics, a majority of patients should survive and have a good recovery.117,140
Intramedullary Pyogenic Spinal Cord Abscess
Intramedullary spinal cord abscess (ISCA) is a rare lesion involving infection contained within the spinal cord parenchyma. Slightly more than a hundred cases have been reported in the literature.141 Only one case of ISCA was found in a series of 40,000 autopsies.142 The proportion of pediatric cases is much higher for ISCA than epidural or subdural spinal abscesses, with approximately 40% of all cases presenting prior to age 20 and 25% of cases occurring in children younger than age 10.143 The increased preponderance of pediatric patients is a result of the role played by congenital spine lesions in the pathophysiology of the disease, especially dermal sinuses, which are responsible for approximately a quarter of infections.144 Overall, the thoracolumbar and lumbrosacral regions are the most commonly involved, especially in patients with midline deformities. In ISCA patients without midline deformities, the cervical region was more likely to be involved.141 These patients were also more likely to be adults. Most cases of ISCA are solitary, but some have multiple foci of infection and a number of cases of holocord ISCA have been documented.145–147
The majority of currently diagnosed ISCAs have a cryptogenic source, likely from transient bacteremia from breaks in the mucosa or skin. Other sources include contiguous spread from congenital defects, surgery, or trauma, and hematogenous spread from known sources including IVDU.144 Many of the patients, especially in the adult population, have comorbid conditions that impair the immune response (e.g., HIV), similar to those seen in the SEA population.92
Staphylococcus species are the most common causative organisms, followed by Streptococcus species. However, the proportion of ISCAs caused by S. aureus is much lower than in other types of spinal abscess. ISCAs caused by Listeria, Brucella, Actinomyces, gram-negative rods including Pseudomonas, anaerobes, Mycobacterium tuberculosis, Histoplasma, Toxoplasma, Candida, and parasites have been reported.141,148 Unfortunately, around 30% of cases strongly suggestive of ISCA have negative cultures.148
Clinically, ISCAs have been categorized by duration of symptoms as acute (<1 week), subacute (1–6 weeks), or chronic (>6 weeks).149 Acute cases often present with pain, fever, leukocytosis, elevated ESR, and progressive neurologic defects similar to acute transverse myelitis, though, like SEAs, presentation can be quite varied. Chronic ISCA often presents in a fashion similar to intramedullary spinal cord tumors, with a predominance of neurologic deficits occurring in the absence of fever or systemic signs of inflammation.150 Patients can also have concurrent meningitis or brain abscesses.
MRI with contrast is the most useful diagnostic test for identifying ISCA and delineating its location and extent. Most lesions are hypointense on T1-weighted imaging, hyperintense on T2-weighted imaging, and demonstrate nodular or rim enhancement on postcontrast T1 imaging. The areas of T2 signal abnormality are usually more extensive than that seen with the postcontrast T1 sequences and likely represent adjacent edema. The differential diagnosis for ISCA includes intramedullary spinal cord tumors, acute transverse myelitis, multiple sclerosis, Guillain-Barré syndrome, and spinal cord infarct.151 Increased diffusion restriction on diffusion-weighted imaging at the center of the lesion may help differentiate ISCA from intramedullary tumors. CT-myelography can be used to diagnose an intramedullary enlargement and is useful when MRI is not possible, but its specificity and ability to delineate anatomic details is limited.
Treatment entails surgical drainage and IV antibiotics. The surgical approach involves laminectomy, dorsal myelotomy centered at the site of maximal thickness of the abscess, drainage of the abscess with collection of specimen for cultures, and copious irrigation.152 The specimen should be sent for Gram stain and aerobic, anaerobic, mycobacterial, and fungal cultures. The use of a limited laminectomy and myelotomy to allow a drainage catheter to be passed rostrally has been described for holocord ICSAs.146 Excision of any associated dermal sinus should also occur. The risk of ISCA in patients with dermal sinuses is one of the reasons children should be screened for midline back abnormalities and why dermal sinuses should be prophylactically excised at the earliest possible time.146
Broad-spectrum IV antibiotics are started after obtaining cultures and are narrowed as speciation and specificities become available. The choice of empiric antibiotics is similar to those used for SEA except that high-dose ampicillin should be added because of the relative frequency of Listeria monocytogenes in reported cases of ISCA.141 The duration of IV antibiotics should be at least 4 to 6 weeks, with follow-up neurologic examinations, laboratory studies, and imaging to demonstrate successful response to therapy and resolution of the lesion.144 A few case reports and reviews discuss successful treatment of ISCA with antibiotics alone, but the role of medical management alone has yet to be defined.141
ISCA mortality rates have been progressively improving from the reported 90% mortality rate in cases in the preantibiotic era (1830–1944).153 A recent review of cases from 1977 to 1997 reported a mortality rate of 8%.144 Rapidly progressive deficits increase the likelihood of a poor outcome.149 Most survivors have an improvement in neurologic status after treatment with surgery and antibiotics, but a substantial proportion, approximately 70%, have persistent neurologic deficits.144
Atypical Bacterial Infections
Tuberculosis
Tuberculosis may involve the vertebral column, the epidural space, the dura mater, the arachnoid, or the spinal cord itself. Tuberculous spinal infection most commonly involves the vertebral body. However, in up to 10% of patients the neural arch, transverse processes, or spinous processes may be affected. Tuberculous spondylitis most commonly involves the lower thoracic and upper lumbar vertebrae and most commonly affects the vertebral body. It is usually confined to a single level. Tuberculous spondylitis, with subsequent spinal cord compression, continues to be a major public health problem throughout much of the world.
Spinal tuberculosis can be treated either medically or surgically. Treatment objectives include healing the disease, preventing or minimizing neurologic dysfunction, and preventing any further gibbus deformity. Treatment with pharmacologic agents has been shown to be successful in multiple series. The current recommendations for the first-line antibiotic treatment of spinal tuberculosis involve 6 to 9 months of isoniazid and rifampin with ethambutol and pyrazinamide added for the first 2 months.154 This shorter 6- to 9-month course was equivalent to longer 18-month regimens involving isoniazid and para-aminosalicylic acid or ethambutol in clinical trials.155,156 M. tuberculosis isolates should be tested for drug susceptibilities to guide therapy.
When the spine is stable and neurologic signs are absent or minimal, initial therapy should be pharmacologic rather than surgical. More than 85% of patients with Pott paraplegia make an excellent recovery with pharmacotherapy. Routine focal debridement and abscess evacuation in addition to antibiotics has not been shown to provide major benefit over antibiotics alone.156 In general, surgery should be reserved for diagnostic biopsy, spinal instability, severe deformity, significant abscesses, open draining sinuses, or myelopathy.157 Surgery often requires ventral decompression and stabilization. These ventral procedures in the thoracic spine are extensive and are often dangerous, fraught with the potential for catastrophe. In addition, any deformity correction obtained with surgery may subsequently recede with time.
Brucellosis
The causative agents in brucellosis are small, nonmotile, non–spore-forming, aerobic gram-negative coccobacilli that are commonly found in domestic animals, including Brucella melitensis (goats), Brucella abortus (cattle), B. canis (dogs), and B. suis (swine). The organism is usually transmitted to humans by ingestion of contaminated products, skin wound contamination from infected animal tissues, and inhalation of aerosols. The disease affects approximately 500,000 people per year worldwide.158 Increasing use of milk pasteurization has resulted in a decreasing incidence of brucellosis in the United States. Likewise, brucellosis is also uncommon in other developed countries because of milk pasteurization.
Brucella infections are often asymptomatic. Initial infection leads to immunity in more than 90% of cases. After an incubation period of 10 days to 3 weeks, the patient typically develops a low-grade fever, malaise, lymphadenopathy, hepatosplenomegaly, and diffuse arthralgias. The infection spreads through the lymphatic system, resulting in acute systemic infection and chronic relapsing disease (undulant fever). However, a classic undulant fever rarely occurs in patients.159
Failure to provide adequate treatment at this stage can result in involvement of almost any organ system. After the initial illness, which may last for several days to weeks, relapse occurs in approximately 5% of patients. Relapses seldom occur in appropriately treated patients and often are the result of focal suppurative lesions. Musculoskeletal involvement is the most common complication of brucellosis. The spine is most commonly affected. Brucellar spondylitis typically develops secondary to chronic brucellosis and occurs in 10% to 50% of patients with brucellosis.160 Brucellosis is one of the major causes of spondylitis in the Mediterranean basin. Of those patients with spinal brucellosis, approximately 12% have some degree of spinal cord compromise.
Brucellar spondylitis should be part of the differential diagnosis of any patient with back and radicular pain in a region where brucellosis is endemic. Lumbar involvement is most common. Localized back pain is the most common symptom. It may be present even at rest. In most cases, radiating pelvic and girdle pains are often noted, along with restriction of movement, muscle spasms, tenderness, and signs of nerve root involvement. Neurologic deficits occur in approximately 20% of patients. Formation of a paraspinal or epidural abscess is uncommon but can occur with severe infections.161
Involvement of the spine can be either focal or diffuse, with a predilection to the lumbar region. Hallmarks of focal brucellar spondylitis include vertebral end-plate erosion and sclerosis, inflammatory changes, and intact discs. Features of diffuse brucellar spondylitis include osteomyelitis of neighboring vertebrae, involvement of the intervening disc, and epidural extension.162 Radiographic changes occur relatively late in the course of the disease and are similar to but less severe than those observed with tuberculosis. Plain radiographs demonstrate disc space involvement with erosion of the adjacent cortical bone, preservation of relatively intact vertebral architecture despite the amount of infection present, and absence of gibbus formations. Plain radiographs demonstrate a thinning of the disc space and erosion of the vertebral body adjacent to the involved disc. This epiphysitis usually occurs in the ventral-rostral angle and may be the main sign of bone destruction. Osteophytic bridging occurs across the infected disc interspace. CT demonstrates destruction of both cortical and cancellous bone. MRI and CT findings are similar for tuberculosis and brucellosis, except that tuberculosis produces more kyphosis and paraspinal abscess formation.
The mainstay of treatment for brucellosis is antibiotic therapy. A recent systematic review recommends that the first-line treatment for brucellosis should be combination therapy with doxycycline for 6 weeks and gentamicin for the initial 2 weeks, with the optional addition of rifampin for a total of 6 weeks.163 Spinal brucellosis appears to have a higher incidence of recurrence, which leads some authors to recommend an antibiotic course of 6 months.164
Actinomycosis
Most patients can be treated nonoperatively with antibiotics and spinal immobilization. Before the advent of penicillin, approximately 75% of cases were diagnosed postmortem. The disease involves the vertebrae and ribs in less than 1% of patients who have actinomycosis. Isolated vertebral body infections can be adequately treated nonoperatively with aggressive antibiotic therapy, usually IV penicillin G for 6 weeks followed by 6 to 12 months of oral penicillin or amoxicillin.165 Tetracycline, doxycycline, and clindamycin have all been used successfully in patients with penicillin allergies. Indications for operative intervention include epidural infection with spinal cord compression, large abscesses, and progressive spinal deformity.
Nocardiosis
If there is no spinal cord compression or large abscess, medical therapy alone is often sufficient. Sulfonamides, in conjunction with appropriate surgery, have been the mainstay of treatment since the 1940s. The use of trimethoprim-sulfamethoxazole (TMP-SMX) for 6 or more months has been described for the treatment of systemic nocardiosis.166 Many other antibiotics have been used, either alone or in combination. The optimal duration of therapy is uncertain, but because of the possibility of relapse, treatment is often continued for many months after apparent cure. A poor response to treatment may be related to the presence of a second pathogen.
Syphilis
The syphilitic gumma lesion is composed of micro-organisms and the local tissue reaction to the organisms. Gummas are rare, destructive, and usually symptomatic, causing collapse and neurologic deficits. The clinical features of spinal gumma are often difficult to distinguish from those of coincident neuropathy, which is often present, and biopsy is necessary for the diagnosis of spinal gumma. The indications for surgically treating syphilitic gummas are similar to those for treating tuberculous spondylitis. The treatment of choice for syphilis is penicillin, though the dosing depends on the specific manifestation of the disease.167
Fungal Infections
Fungal infections of the spine are rare and generally occur in debilitated, diabetic, or immunocompromised patients. Patients with acute leukemia, patients with lymphoma, recipients of organ transplants, and those receiving chemotherapy are particularly susceptible.168 Accurate diagnosis is often delayed because other medical conditions mask the diagnosis and because fungal spondylitides are often indolent in nature. Notably, sporadic cases of fungal osteomyelitis have been reported in immunocompetent patients.169
Aspergillosis
Infection typically is acquired by inhaling small spores (conidia). Although uncommon, spine involvement nearly always results from hematogenous spread from the lungs; though postoperative Aspergillus discitis has been reported after lumbar discectomy.170 Vertebral involvement can also result from contiguous spread from the lungs to the vertebral bodies. Vascular invasion is common in immunocompromised patients, and it leads to tissue necrosis with abundant hyphal proliferation.171 In patients with chronic granulomatous disease, vascular invasion is uncommon and hyphae are sparse.
Clinically, sinus tract formation is characteristic, though often not present. The incidence of epidural abscess formation, in association with neurologic deficits, is high. Diagnosis is established by percutaneous or open biopsy. Voriconazole has been demonstrated to be superior to amphotericin B for the treatment of invasive aspergillosis. Liposomal amphotericin, posaconazole, itraconazole, caspofungin, or micofungin are all possibilities for salvage therapy in cases refractory to voriconazole therapy. Treatment duration is usually 6 to 8 weeks minimum for patients with normal immune systems. Immunocompromised patients likely require longer treatment and usually warrant chronic suppression therapy.172
In addition to antibiotics, surgical debridement with resection of devitalized bone and cartilage is important to achieve cure in most cases. With Aspergillus discitis, early surgery with vigorous surgical debridement combined with antifungal treatment, yields a good outcome in most cases.170,173 There are rare reports of successful treatment with antifungal agents alone.174,175 The prognosis of patients with Aspergillus spondylitis overall is guarded, in large part due to the patient’s usually poor condition and numerous comorbities.
Blastomycosis
Blastomycosis is caused by Blastomyces dermatitidis, a dimorphic fungus that is endemic to the midwestern, southeastern, and south central United States. Primary infection in humans occurs by inhalation of conidia, which then convert to the yeast phase in the lung. The incubation period for acute pulmonary infection is 30 to 45 days. The symptoms are nonspecific, and acute pulmonary infection may occasionally be undetected. Extrapulmonary disease involving the skin, bones, genitourinary system, or CNS occurs in approximately 25% to 40% of cases and is much more common in immunocompromised patients. The organism spreads hematogenously from the lungs to the spine.176 Men are affected nine times more commonly than women, particularly those with a history of alcohol abuse.
Before the availability of effective antimicrobial therapy, the mortality rate exceeded 60%.177 CNS blastomycosis should be treated with IV liposomal amphotericin B for 4 to 6 weeks followed by oral azole therapy (fluconazole, itraconazole, or voriconazole) for at least 12 months. Though mild pulmonary blastomycosis often resolves without treatment, antibiotic therapy should be considered to prevent dissemination to extrapulmonary sites.178 Surgery is generally reserved for patients with severe infections, such as epidural abscesses, that are causing neurologic deficits.
Coccidioidomycosis
The primary focus of disease is the lungs, but the disease becomes disseminated in 0.5% of cases, and the rate of extrapulmonary disease is much higher in immunocompromised hosts.179 The organism enters the body when arthroconidia are inhaled into the lungs and can secondarily disseminate via hematogenous spread. Osseous lesions are found in 20% of those with disseminated disease. Vertebral lesions are most common in the thoracic and lumbar spine. Multicentric disease is common. Radiographic studies usually reveal that the intervertebral disc is relatively uninvolved compared with the vertebral body, pedicles, and transverse processes. Contiguous rib involvement is also common. Paraspinal abscesses and skin tracts are common.
Extrapulmonary coccidioidomycosis is usually treated with oral ketoconazole, itraconazole, or fluconazole. IV amphotericin B is indicated for rapidly progressive infections or with infections not responsive to initial therapy. Surgical debridement is indicated for large abscesses, neurologic compromise, spinal instability, or failure of medical therapy.180
Cryptococcosis
Cryptococcosis is a subacute or chronic infection caused by a yeastlike fungus surrounded by a gelatinous capsule, Cryptococcus neoformans. Infection is acquired by inhalation of the aerosolized organism. Because many cases of cryptococcal osteomyelitis occur in normal hosts, it should be considered in the differential diagnosis even in a normal host.181 Osseous involvement occurs in less than 5% of all cases and resembles cold abscesses. Sinus tracts and abscess formation are rare.
The onset of cryptococcosis is usually insidious. The bone infection typically has an indolent course. Both serum and CSF agglutination tests are available. Testing for cryptococcal antigen in the serum should be done prior to invasive diagnostic procedures.181 Radiographic studies show lucent lesions of the vertebral bodies with sharply scalloped margins and little, if any, reactive sclerosis or periosteal new bone formation. The disc spaces are typically unaffected.
Parasite Infections
Cysticercosis
Cysticercosis is suggested by a history of infection with an adult worm, multiple subcutaneous nodules, typical symptoms, previous residence in highly endemic regions where undercooked pork may have been eaten, and eosinophilia. Indirect hemagglutination tests may be positive in the blood or spinal fluid; however, a negative result does not rule out cysticercosis.
Both albendazole and praziquantel have proven effective in the treatment of both cerebral and spinal cysticercosis. Recent analysis favors albendazole over praziquantel.182 Because of the CNS’s inflammatory reaction to dying parasites, concomitant administration of steroids to minimize this inflammatory reaction has been recommended. Excision of the cyst has produced significant improvement in patients with neurologic deficits secondary to neural compression.
Echinococcosis
All patients with echinococcosis should undergo surgical debridement if possible, with complete excision of the cyst and affected bone and stabilization, as needed. Unfortunately the relatively high risk of cyst rupture (up to 44% in extradural cases) can cause anaphylaxis or seed the surgical site, leading to future recurrence.183 A hyperosmolar saline solution washout is recommended intraoperatively to prevent cyst recurrence. Although its value remains to be established, presumably, the hypertonic saline solution disrupts any residual cysts osmotically. Albendazole and mebendazole have both been used to treat echinococcosis, with most evidence favoring albendazole as the primary agent. Treatment with albendazole should be continued for a minimum of 3 to 4 months, with one report suggesting 1 year of treatment after neural decompression.184
Recent studies have observed an improved prognosis for what was once thought to be a uniformly fatal disease. However, for spinal echinococcosis, reports of recurrence rates range from 30% to 100%, and mortality rates range from 5% to 50%.185 Response to therapy should be monitored with serial imaging, given the high incidence of recurrence.
Schistosomiasis
The parasitic flukes (trematodes) of the Schistosoma genus can rarely involve the CNS, including the spinal cord. Three species have been implicated in causing diseases of the spinal cord: Schistosoma mansoni, S. haematobium, and S. japonicum186 The trematode larvae gain access to humans by piercing the skin of people exposed to infected waters. Once the larvae penetrate the skin, they gain access to the vascular system, develop into adults, and produce eggs, which are passed out of the urine (S. haematobium) or feces (S. mansoni and S. japonicum). Their natural hosts are freshwater snails.
The most typical clinical picture is that of transverse myelitis or myeloradiculopathy. The lumbosacral region is the most commonly affected region, though all regions of the spinal cord, including the cervical spine,186 can be affected. In the acute stage of lumbosacral myeloradiculopathy the patient presents with lumbar pain, lower extremity sensory changes, and bladder dysfunction, followed by lower extremity weakness, sexual dysfunction, or bowel dysmotility.
Treatment involves the use of schistosomicidal drugs, such as praziquantel, to kill the adult worms and stop the production of eggs. Treatment with corticosteroids has been recommended as an adjuvant to praziquantel to stop the inflammatory response to the eggs that have already been deposited, though length of treatment has not been defined, with some groups recommending up to 6 months of steroid treatment.187 Surgical intervention is reserved for acute paraplegia or evidence of medullary compression, medically refractory cases, or biopsy when the diagnosis is unclear. Surgery usually involves laminectomy for decompression and possible release of spinal nerve roots with biopsy if diagnosis is uncertain.188 Resection of focal lesions has also been described.186
Cottle L., Riordan T. Infectious spondylodiscitis. J Infect. 2008;56(6):401-412.
Khoo L.T., Mikawa K., Fessler R.G. A surgical revisitation of Pott distemper of the spine. Spine J. 2003;3(2):130-145.
Kowalski T.J., Berbari E.F., Huddleston P.M., et al. The management and outcome of spinal implant infections: contemporary retrospective cohort study. Clin Infect Dis. 2007;44(7):913-920.
McHenry M.C., Easley K.A., Locker G.A. Vertebral osteomyelitis: long-term outcome for 253 patients from 7 Cleveland-area hospitals. Clin Infect Dis. 2002;34(10):1342-1350.
Reihsaus E., Waldbaur H., Seeling W. Spinal epidural abscess: a meta-analysis of 915 patients. Neurosurg Rev. 2000;23(4):175-204. discussion 205
Watters W.C., Baisden J., Bono C.M., et al. Antibiotic prophylaxis in spine surgery: an evidence-based clinical guideline for the use of prophylactic antibiotics in spine surgery. Spine J. 2009;9(2):142-146.
Weinstein M.A., McCabe J.P., Cammisa F.P.Jr. Postoperative spinal wound infection: a review of 2,391 consecutive index procedures. J Spinal Disord. 2000;13(5):422-426.
1. Abbey D.M., Turner D.M., Warson J.S., et al. Treatment of postoperative wound infections following spinal fusion with instrumentation. J Spinal Disord. 1995;8(4):278-283.
2. Glassman S.D., Dimar J.R., Puno R.M., Johnson J.R. Salvage of instrumental lumbar fusions complicated by surgical wound infection. Spine (Phila Pa 1976). 1996;21(18):2163-2169.
3. Keller R.B., Pappas A.M. Infection after spinal fusion using internal fixation instrumentation. Orthop Clin North Am. 1972;3(1):99-111.
4. Kostuik J.P., Israel J., Hall J.E. Scoliosis surgery in adults. Clin Orthop Relat Res. 1973;93:225-234.
5. Lonstein J., Winter R., Moe J., Gaines D. Wound infection with Harrington instrumentation and spine fusion for scoliosis. Clin Orthop Relat Res. 1973;96:222-233.
6. Roberts F.J., Walsh A., Wing P., et al. The influence of surveillance methods on surgical wound infection rates in a tertiary care spinal surgery service. Spine (Phila Pa 1976). 1998;23(3):366-370.
7. West J.L.3rd, Ogilvie J.W., Bradford D.S. Complications of the variable screw plate pedicle screw fixation. Spine (Phila Pa 1976). 1991;16(5):576-579.
8. Balderston R.A., Blumberg K.D. Infection in spine surgery. In: Balderston R.A., An H.S., editors. Complications in spinal surgery. Philadelphia: Saunders; 1991:157-168.
9. Olsen M.A., Mayfield J., Lauryssen C., et al. Risk factors for surgical site infection in spinal surgery. J Neurosurg. 2003;98(Suppl 2):149-155.
10. Calderone R.R., Garland D.E., Capen D.A., Oster H. Cost of medical care for postoperative spinal infections. Orthop Clin North Am. 1996;27(1):171-182.
11. Kirkland K.B., Briggs J.P., Trivette S.L., et al. The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol. 1999;20(11):725-730.
12. Klein J.D., Hey L.A., Yu C.S., et al. Perioperative nutrition and postoperative complications in patients undergoing spinal surgery. Spine (Phila Pa 1976). 1996;21(22):2676-2682.
13. Tenney J.H., Vlahov D., Salcman M., Ducker T.B. Wide variation in risk of wound infection following clean neurosurgery. Implications for perioperative antibiotic prophylaxis. J Neurosurg. 1985;62(2):243-247.
14. Andreshak T.G., An H.S., Hall J., Stein B. Lumbar spine surgery in the obese patient. J Spinal Disord. 1997;10(5):376-379.
15. Capen D.A., Calderone R.R., Green A. Perioperative risk factors for wound infections after lower back fusions. Orthop Clin North Am. 1996;27(1):83-86.
16. Massie J.B., Heller J.G., Abitbol J.J., et al. Postoperative posterior spinal wound infections. Clin Orthop Relat Res. 1992;284:99-108.
17. Patel N., Bagan B., Vadera S., et al. Obesity and spine surgery: relation to perioperative complications. J Neurosurg Spine. 2007;6(4):291-297.
18. Wimmer C., Gluch H., Franzreb M., Ogon M. Predisposing factors for infection in spine surgery: a survey of 850 spinal procedures. J Spinal Disord. 1998;11(2):124-128.
19. Ho C., Sucato D.J., Richards B.S. Risk factors for the development of delayed infections following posterior spinal fusion and instrumentation in adolescent idiopathic scoliosis patients. Spine (Phila Pa 1976). 2007;32(20):2272-2277.
20. Simpson J.M., Silveri C.P., Balderston R.A., et al. The results of operations on the lumbar spine in patients who have diabetes mellitus. J Bone Joint Surg [Am]. 1993;75(12):1823-1829.
21. Rechtine G.R., Bono P.L., Cahill D., et al. Postoperative wound infection after instrumentation of thoracic and lumbar fractures. J Orthop Trauma. 2001;15(8):566-569.
22. Sponseller P.D., LaPorte D.M., Hungerford M.W., et al. Deep wound infections after neuromuscular scoliosis surgery: a multicenter study of risk factors and treatment outcomes. Spine (Phila Pa 1976). 2000;25(19):2461-2466.
23. Blam O.G., Vaccaro A.R., Vanichkachorn J.S., et al. Risk factors for surgical site infection in the patient with spinal injury. Spine (Phila Pa 1976). 2003;28(13):1475-1480.
24. Kowalski T.J., Berbari E.F., Huddleston P.M., et al. The management and outcome of spinal implant infections: contemporary retrospective cohort study. Clin Infect Dis. 2007;44(7):913-920.
25. Bongartz E.B., Ulrich P., Fidler M., Bernucci C. Reoperation in the management of post-operative disc space infection. Zentralbl Neurochir. 1994;55(2):120-124.
26. Weinstein M.A., McCabe J.P., Cammisa F.P.Jr. Postoperative spinal wound infection: a review of 2,391 consecutive index procedures. J Spinal Disord. 2000;13(5):422-426.
27. Dauch W.A. Infection of the intervertebral space following conventional and microsurgical operation on the herniated lumbar intervertebral disc. A controlled clinical trial. Acta Neurochir (Wien). 1986;82(1-2):43-49.
28. Kanafani Z.A., Dakdouki G.K., El-Dbouni O., et al. Surgical site infections following spinal surgery at a tertiary care center in Lebanon: incidence, microbiology, and risk factors. Scand J Infect Dis. 2006;38(8):589-592.
29. Sasso R.C., Garrido B.J. Postoperative spinal wound infections. J Am Acad Orthop Surg. 2008;16(6):330-337.
30. Levi A.D., Dickman C.A., Sonntag V.K. Management of postoperative infections after spinal instrumentation. J Neurosurg. 1997;86(6):975-980.
31. Wimmer C., Gluch H. Management of postoperative wound infection in posterior spinal fusion with instrumentation. J Spinal Disord. 1996;9(6):505-508.
32. Picada R., Winter R.B., Lonstein J.E., et al. Postoperative deep wound infection in adults after posterior lumbosacral spine fusion with instrumentation: incidence and management. J Spinal Disord. 2000;13(1):42-45.
33. Payne D.H., Fischgrund J.S., Herkowitz H.N., et al. Efficacy of closed wound suction drainage after single-level lumbar laminectomy. J Spinal Disord. 1996;9(5):401-403.
34. Banco S.P., Vaccaro A.R., Blam O., et al. Spine infections: variations in incidence during the academic year. Spine (Phila Pa 1976). 2002;27(9):962-965.
35. Brown E.M., Pople I.K., de Louvois J., et al. Spine update: prevention of postoperative infection in patients undergoing spinal surgery. Spine (Phila Pa 1976). 2004;29(8):938-945.
36. Watters W.C., Baisden J., Bono C.M., et al. Antibiotic prophylaxis in spine surgery: an evidence-based clinical guideline for the use of prophylactic antibiotics in spine surgery. Spine J. 2009;9(2):142-146.
37. Barker F.G. Efficacy of prophylactic antibiotic therapy in spinal surgery: a meta-analysis. Neurosurgery. 2002;51(2):391-400. discussion 400–401
38. Rohde V., Meyer B., Schaller C., Hassler W.E. Spondylodiscitis after lumbar discectomy. Incidence and a proposal for prophylaxis. Spine (Phila Pa 1976). 1998;23(5):615-620.
39. Rhoten R.L., Murphy M.A., Kalfas I.H., et al. Antibiotic penetration into cervical discs. Neurosurgery. 1995;37(3):418-421.
40. Haines S.J. Topical antibiotic prophylaxis in neurosurgery. Neurosurgery. 1982;11(2):250-253.
41. Darouiche R.O., Wall M.J., Itani K.M.F., et al. Chlorhexidine-alcohol versus povidone-iodine for surgical-site antisepsis. N Engl J Med. 2010;362(1):18-26.
42. Tanner J., Swarbrook S., Stuart J. Surgical hand antisepsis to reduce surgical site infection. Cochrane Database Syst Rev. 2008;1:CD004288.
43. Gruenberg M.F., Campaner G.L., Sola C.A., Ortolan E.G. Ultraclean air for prevention of postoperative infection after posterior spinal fusion with instrumentation: a comparison between surgeries performed with and without a vertical exponential filtered air-flow system. Spine (Phila Pa 1976). 2004;29(20):2330-2334.
44. Tandon N., Vollmer D.G. Infections of the spine and spinal cord. In: Winn R.H., editor. Youmans neurological surgery. ed 5. Philadelphia: Saunders; 2004:4363-4394.
45. Haidar R., Najjar M., Boghossian A.D., Tabbarah Z. Propionibacterium acnes causing delayed postoperative spine infection: review. Scand J Infect Dis. 2010;42:405-411.
46. Rigamonti D., Liem L., Sampath P., et al. Spinal epidural abscess: contemporary trends in etiology, evaluation, and management. Surg Neurol. 1999;52(2):189-196. discussion 197
47. Thelander U., Larsson S. Quantitation of C-reactive protein levels and erythrocyte sedimentation rate after spinal surgery. Spine (Phila Pa 1976). 1992;17(4):400-404.
48. Khan M.H., Smith P.N., Rao N., Donaldson W.F. Serum C-reactive protein levels correlate with clinical response in patients treated with antibiotics for wound infections after spinal surgery. Spine J. 2006;6(3):311-315.
49. Kothari N.A., Pelchovitz D.J., Meyer J.S. Imaging of musculoskeletal infections. Radiol Clin North Am. 2001;39(4):653-671.
50. Beiner J.M., Grauer J., Kwon B.K., Vaccaro A.R. Postoperative wound infections of the spine. Neurosurg Focus. 2003;15(3):E14.
51. Martin R.J., Yuan H.A. Neurosurgical care of spinal epidural, subdural, and intramedullary abscesses and arachnoiditis. Orthop Clin North Am. 1996;27(1):125-136.
52. Boden S.D., Davis D.O., Dina T.S., et al. Postoperative diskitis: distinguishing early MR imaging findings from normal postoperative disk space changes. Radiology. 1992;184(3):765-771.
53. Ross J.S., Zepp R., Modic M.T. The postoperative lumbar spine: enhanced MR evaluation of the intervertebral disk. AJNR Am J Neuroradiol. 1996;17(2):323-331.
54. Love C., Patel M., Lonner B.S., et al. Diagnosing spinal osteomyelitis: a comparison of bone and Ga-67 scintigraphy and magnetic resonance imaging. Clin Nucl Med. 2000;25(12):963-977.
55. Ploumis A., Mehbod A.A., Dressel T.D., et al. Therapy of spinal wound infections using vacuum-assisted wound closure: risk factors leading to resistance to treatment. J Spinal Disord Tech. 2008;21(5):320-323.
56. Yuan-Innes M.J., Temple C.L., Lacey M.S. Vacuum-assisted wound closure: a new approach to spinal wounds with exposed hardware. Spine (Phila Pa 1976). 2001;26(3):E30-E33.
57. Ubbink D.T., Westerbos S.J., Evans D., et al. Topical negative pressure for treating chronic wounds. Cochrane Database Syst Rev. 2008;3:CD001898.
58. Mok J.M., Guillaume T.J., Talu U., et al. Clinical outcome of deep wound infection after instrumented posterior spinal fusion: a matched cohort analysis. Spine (Phila Pa 1976). 2009;34(6):578-583.
59. Stambough J.L., Beringer D. Postoperative wound infections complicating adult spine surgery. J Spinal Disord. 1992;5(3):277-285.
60. Clark C.E., Shufflebarger H.L. Late-developing infection in instrumented idiopathic scoliosis. Spine (Phila Pa 1976). 1999;24(18):1909-1912.
61. Viola R.W., King H.A., Adler S.M., Wilson C.B. Delayed infection after elective spinal instrumentation and fusion. A retrospective analysis of eight cases. Spine (Phila Pa 1976). 1997;22(20):2444-2450. discussion 2450–2451
62. Cottle L., Riordan T. Infectious spondylodiscitis. J Infect. 2008;56(6):401-412.
63. Hopkinson N., Stevenson J., Benjamin S. A case ascertainment study of septic discitis: clinical, microbiological and radiological features. QJM. 2001;94(9):465-470.
64. Legrand E., Flipo R.M., Guggenbuhl P., et al. Management of nontuberculous infectious discitis: treatments used in 110 patients admitted to 12 teaching hospitals in France. Joint Bone Spine. 2001;68(6):504-509.
65. Grados F., Lescure F.X., Senneville E., et al. Suggestions for managing pyogenic (non-tuberculous) discitis in adults. Joint Bone Spine. 2007;74(2):133-139.
66. Chia S.L., Tan B.H., Tan C.T., Tan S.B. Candida spondylodiscitis and epidural abscess: management with shorter courses of anti-fungal therapy in combination with surgical debridement. J Infect. 2005;51(1):17-23.
67. Milward F., Grout J. Changes in intervertebral discs following lumbar puncture. Lancet. 1936;2:183.
68. Fernandez M., Carrol C.L., Baker C.J. Discitis and vertebral osteomyelitis in children: an 18-year review. Pediatrics. 2000;105(6):1299-1304.
69. Friedman J.A., Maher C.O., Quast L.M., et al. Spontaneous disc space infections in adults. Surg Neurol. 2002;57(2):81-86.
70. Le Moal G., Roblot F., Paccalin M., et al. Clinical and laboratory characteristics of infective endocarditis when associated with spondylodiscitis. Eur J Clin Microbiol Infect Dis. 2002;21(9):671-675.
71. An H.S., Seldomridge J.A. Spinal infections: diagnostic tests and imaging studies. Clin Orthop Relat Res. 2006;444:27-33.
72. Butler J.S., Shelly M.J., Timlin M., et al. Nontuberculous pyogenic spinal infection in adults: a 12-year experience from a tertiary referral center. Spine (Phila Pa 1976). 2006;31(23):2695-2700.
73. McHenry M.C., Easley K.A., Locker G.A. Vertebral osteomyelitis: long-term outcome for 253 patients from 7 Cleveland-area hospitals. Clin Infect Dis. 2002;34(10):1342-1350.
74. Grammatico L., Baron S., Rusch E., et al. Epidemiology of vertebral osteomyelitis (VO) in France: analysis of hospital-discharge data 2002–2003. Epidemiol Infect. 2008;136(5):653-660.
75. Mylona E., Samarkos M., Kakalou E., et al. Pyogenic vertebral osteomyelitis: a systematic review of clinical characteristics. Semin Arthritis Rheum. 2009;39(1):10-17.
76. Patzakis M.J., Rao S., Wilkins J., et al. Analysis of 61 cases of vertebral osteomyelitis. Clin Orthop Relat Res. 1991;264:178-183.
77. Babinchak T.J., Riley D.K., Rotheram E.B.Jr. Pyogenic vertebral osteomyelitis of the posterior elements. Clin Infect Dis. 1997;25(2):221-224.
78. Priest D.H., Peacock J.E. Hematogenous vertebral osteomyelitis due to Staphylococcus aureus in the adult: clinical features and therapeutic outcomes. South Med J. 2005;98(9):854-862.
79. Zimmerli W. Clinical practice. Vertebral osteomyelitis. N Engl J Med. 2010;362(11):1022-1029.
80. Jensen A.G., Espersen F., Skinhøj P., Frimodt-Møller N. Bacteremic Staphylococcus aureus spondylitis. Arch Intern Med. 1998;158(5):509-517.
81. Palestro C.J., Love C., Miller T.T. Infection and musculoskeletal conditions: imaging of musculoskeletal infections. Best Pract Res Clin Rheumatol. 2006;20(6):1197-1218.
82. Carragee E., Iezza A. Does acute placement of instrumentation in the treatment of vertebral osteomyelitis predispose to recurrent infection: long-term follow-up in immune-suppressed patients. Spine (Phila Pa 1976). 2008;33(19):2089-2093.
83. Roblot F., Besnier J.M., Juhel L., et al. Optimal duration of antibiotic therapy in vertebral osteomyelitis. Semin Arthritis Rheum. 2007;36(5):269-277.
84. Sapico F.L., Montgomerie J.Z. Pyogenic vertebral osteomyelitis: report of nine cases and review of the literature. Rev Infect Dis. 1979;1(5):754-776.
85. Schrenzel J., Harbarth S., Schockmel G., et al. A randomized clinical trial to compare fleroxacin-rifampicin with flucloxacillin or vancomycin for the treatment of staphylococcal infection. Clin Infect Dis. 2004;39(9):1285-1292.
86. Pereira C.E., Lynch J.C. Spinal epidural abscess: an analysis of 24 cases. Surg Neurol. 2005;63(Suppl 1):S26-S29.
87. Baker A.S., Ojemann R.G., Swartz M.N., Richardson E.P.Jr. Spinal epidural abscess. N Engl J Med. 1975;293(10):463-468.
88. Darouiche R.O., Hamill R.J., Greenberg S.B., et al. Bacterial spinal epidural abscess. Review of 43 cases and literature survey. Medicine (Baltimore). 1992;71(6):369-385.
89. Akalan N., Ozgen T. Infection as a cause of spinal cord compression: a review of 36 spinal epidural abscess cases. Acta Neurochir (Wien). 2000;142(1):17-23.
90. Reihsaus E., Waldbaur H., Seeling W. Spinal epidural abscess: a meta-analysis of 915 patients. Neurosurg Rev. 2000;23(4):175-204. discussion 205
91. Obrador G.T., Levenson D.J. Spinal epidural abscess in hemodialysis patients: report of three cases and review of the literature. Am J Kidney Dis. 1996;27(1):75-83.
92. Nussbaum E.S., Rigamonti D., Standiford H., et al. Spinal epidural abscess: a report of 40 cases and review. Surg Neurol. 1992;38(3):225-231.
93. Batson O.V. The function of the vertebral veins and their role in the spread of metastases. Ann Surg. 1940;112(1):138-149.
94. Khan S.H., Hussain M.S., Griebel R.W., Hattingh S. Title comparison of primary and secondary spinal epidural abscesses: a retrospective analysis of 29 cases. Surg Neurol. 2003;59(1):28-33. discussion 33
95. Darouiche R.O. Spinal epidural abscess. N Engl J Med. 2006;355(19):2012-2020.
96. Koppel B.S., Tuchman A.J., Mangiardi J.R., et al. Epidural spinal infection in intravenous drug abusers. Arch Neurol. 1988;45(12):1331-1337.
97. Lechiche C., Le Moing V., Marchandin H., et al. Spondylodiscitis due to Bacteroides fragilis: two cases and review. Scand J Infect Dis. 2006;38(3):229-231.
98. Heusner A.P. Nontuberculous spinal epidural infections. N Engl J Med. 1948;239(23):845-854.
99. Lifeso R.M. Pyogenic spinal sepsis in adults. Spine (Phila Pa 1976). 1990;15(12):1265-1271.
100. Maslen D.R., Jones S.R., Crislip M.A., et al. Spinal epidural abscess. Optimizing patient care. Arch Intern Med. 1993;153(14):1713-1721.
101. Del Curling O.Jr., Gower D.J., McWhorter J.M. Changing concepts in spinal epidural abscess: a report of 29 cases. Neurosurgery. 1990;27(2):185-192.
102. Rothman S.L. The diagnosis of infections of the spine by modern imaging techniques. Orthop Clin North Am. 1996;27(1):15-31.
103. Danner R.L., Hartman B.J. Update on spinal epidural abscess: 35 cases and review of the literature. Rev Infect Dis. 1987;9(2):265-274.
104. Hlavin M.L., Kaminski H.J., Ross J.S., Ganz E. Spinal epidural abscess: a ten-year perspective. Neurosurgery. 1990;27(2):177-184.
105. Redekop G.J., Del Maestro R.F. Diagnosis and management of spinal epidural abscess. Can J Neurol Sci. 1992;19(2):180-187.
106. Krauss W.E., McCormick P.C. Infections of the dural spaces. Neurosurg Clin North Am. 1992;3(2):421-433.
107. Chew F.S., Kline M.J. Diagnostic yield of CT-guided percutaneous aspiration procedures in suspected spontaneous infectious diskitis. Radiology. 2001;218(1):211-214.
108. Leys D., Lesoin F., Viaud C., et al. Decreased morbidity from acute bacterial spinal epidural abscesses using computed tomography and nonsurgical treatment in selected patients. Ann Neurol. 1985;17(4):350-355.
109. Mampalam T.J., Rosegay H., Andrews B.T., et al. Nonoperative treatment of spinal epidural infections. J Neurosurg. 1989;71(2):208-210.
110. Strohecker J., Grobovschek M. [Spinal epidural abscess: an interdisciplinary emergency. Zentralbl Neurochir. 1986;47(2):120-124.
111. Siddiq F., Chowfin A., Tight R., et al. Medical vs surgical management of spinal epidural abscess. Arch Intern Med. 2004;164(22):2409-2412.
112. Savage K., Holtom P.D., Zalavras C.G. Spinal epidural abscess: early clinical outcome in patients treated medically. Clin Orthop Relat Res. 2005;439:56-60.
113. Karikari I.O., Powers C.J., Reynolds R.M., et al. Management of a spontaneous spinal epidural abscess: a single-center 10-year experience. Neurosurgery. 2009;65(5):919-923. discussion 923–924
114. Curry W.T.Jr., Hoh B.L., Amin-Hanjani S., Eskandar E.N. Spinal epidural abscess: clinical presentation, management, and outcome. Surg Neurol. 2005;63(4):364-371. discussion 371
115. Khanna R.K., Malik G.M., Rock J.P., Rosenblum M.L. Spinal epidural abscess: evaluation of factors influencing outcome. Neurosurgery. 1996;39(5):958-964.
116. Garrido E., Rosenwasser R.H. Experience with the suction-irrigation technique in the management of spinal epidural infection. Neurosurgery. 1983;12(6):678-679.
117. Bartels R.H., de Jong T.R., Grotenhuis J.A. Spinal subdural abscess. Case report. J Neurosurg. 1992;76(2):307-311.
118. Fischer E.G., Greene C.S.Jr., Winston K.R. Spinal epidural abscess in children. Neurosurgery. 1981;9(3):257-260.
119. de Villiers J.C., Cluver P.F. Spinal epidural abscess in children. S Afr J Surg. 1978;16(2):149-155.
120. Przybylski G.J., Sharan A.D. Single-stage autogenous bone grafting and internal fixation in the surgical management of pyogenic discitis and vertebral osteomyelitis. J Neurosurg. 2001;94(Suppl 1):1-7.
121. Schuster J.M., Avellino A.M., Mann F.A., et al. Use of structural allografts in spinal osteomyelitis: a review of 47 cases. J Neurosurg. 2000;93(Suppl 1):8-14.
122. Rath S.A., Neff U., Schneider O., Richter H.P. Neurosurgical management of thoracic and lumbar vertebral osteomyelitis and discitis in adults: a review of 43 consecutive surgically treated patients. Neurosurgery. 1996;38(5):926-933.
123. Stone J.L., Cybulski G.R., Rodriguez J., et al. Anterior cervical debridementand strut-grafting for osteomyelitis of the cervical spine. J Neurosurg. 1989;70(6):879-883.
124. Zeidman S.M., Ducker T.B., Raycroft J. Trends and complications in cervical spine surgery: 1989–1993. J Spinal Disord. 1997;10(6):523-526.
125. Benzel E.C. The lateral extracavitary approach to the spine using the three-quarter prone position. J Neurosurg. 1989;71(6):837-841.
126. Larson S.J., Holst R.A., Hemmy D.C., Sances A.Jr. Lateral extracavitary approach to traumatic lesions of the thoracic and lumbar spine. J Neurosurg. 1976;45(6):628-637.
127. Fessler R.G., Dietze D.D.Jr., Millan M.M., Peace D. Lateral parascapular extrapleural approach to the upper thoracic spine. J Neurosurg. 1991;75(3):349-355.
128. Arnold P.M., Baek P.N., Bernardi R.J., et al. Surgical management of nontuberculous thoracic and lumbar vertebral osteomyelitis: report of 33 cases. Surg Neurol. 1997;47(6):551-561.
129. McCormick P.C. Retropleural approach to the thoracic and thoracolumbar spine. Neurosurgery. 1995;37(5):908-914.
130. Dandy W.E. Abscesses and inflammatory tumors in the spinal epidural space (so-called pachymeningitis externa). Arch Surg. 1926;13:477-494.
131. Lu C.H., Chang W.N., Lui C.C., et al. Adult spinal epidural abscess: clinical features and prognostic factors. Clin Neurol Neurosurg. 2002;104(4):306-310.
132. Velissaris D., Aretha D., Fligou F., Filos K.S. Spinal subdural staphylococcus aureus abscess: case report and review of the literature. World J Emerg Surg. 2009;4:31.
133. Baker R.P., Brown E.M., Coakham H.B. Overwhelming cranial and spinal subdural empyema secondary infected sacral decubitus ulcers. Br J Neurosurg. 2003;17(6):572-573.
134. Walker A.E., Bucy P.C. Congenital dermal sinuses: a source of spinal meningeal infection and subdural abscesses. Brain. 1934;57:401-421.
135. Chen C.Y., Lin K.L., Wang H.S., Lui T.N. Dermoid cyst with dermal sinus tract complicated with spinal subdural abscess. Pediatr Neurol. 1999;20(2):157-160.
136. Levy M.L., Wieder B.H., Schneider J., et al. Subdural empyema of the cervical spine: clinicopathological correlates and magnetic resonance imaging. Report of three cases. J Neurosurg. 1993;79(6):929-935.
137. Knudsen L.L., Voldby B., Stagaard M. Computed tomographic myelography in spinal subdural empyema. Neuroradiology. 1987;29(1):99.
138. Theodotou B., Woosley R.E., Whaley R.A. Spinal subdural empyema: diagnosis by spinal computed tomography. Surg Neurol. 1984;21(6):610-612.
139. Price D.B., Gottesman M.H., Adelman R., Schneider S.J. Spinal subdural empyema: appearance on intraoperative sonography. J Ultrasound Med. 1993;12(8):493-495.
140. Lownie S.P., Ferguson G.G. Spinal subdural empyema complicating cervical discography. Spine (Phila Pa 1976). 1989;14(12):1415-1417.
141. Kurita N., Sakurai Y., Taniguchi M., et al. Intramedullary spinal cord abscess treated with antibiotic therapy—case report and review. Neurol Med Chir (Tokyo). 2009;49(6):262-268.
142. Courville C.B. Pathology of the central nervous system, ed 3, Mountainville, CA: Pacific Press, 1950.
143. DiTullio M.V.Jr. Intramedullary spinal abscess: a case report with a review of 53 previously described cases. Surg Neurol. 1977;7(6):351-354.
144. Chan C.T., Gold W.L. Intramedullary abscess of the spinal cord in the antibiotic era: clinical features, microbial etiologies, trends in pathogenesis, and outcomes. Clin Infect Dis. 1998;27(3):619-626.
145. Desai K.I., Muzumdar D.P., Goel A. Holocord intramedullary abscess: an unusual case with review of literature. Spinal Cord. 1999;37(12):866-870.
146. Bunyaratavej K., Desudchit T., Pongpunlert W. Holocord intramedullary abscess due to dermal sinus in a 2-month-old child successfully treated with limited myelotomy and aspiration. Case report. J Neurosurg. 2006;104(Suppl 4):269-274.
147. Kalia V., Aggarwal T. Holocord intramedullary abscess. Indian J Pediatr. 2007;74(6):589-591.
148. Tacconi L., Arulampalam T., Johnston F.G., Thomas D.G. Intramedullary spinal cord abscess: case report. Neurosurgery. 1995;37(4):817-819.
149. Menezes A.H., Graf C.J., Perret G.E. Spinal cord abscess: a review. Surg Neurol. 1977;8(6):461-467.
150. Babu R., Jafar J.J., Huang P.P., et al. Intramedullary abscess associated with a spinal cord ependymoma: case report. Neurosurgery. 1992;30(1):121-124.
151. Murphy K.J., Brunberg J.A., Quint D.J., Kazanjian P.H. Spinal cord infection: myelitis and abscess formation. AJNR Am J Neuroradiol. 1998;19(2):341-348.
152. Morandi X., Mercier P., Fournier H.D., Brassier G. Dermal sinus and intramedullary spinal cord abscess. Report of two cases and review of the literature. Childs Nerv Syst. 1999;15(4):202-206. discussion 207–208
153. Artz P.K. Abscess within the spinal cord. Review of the literature and report of three cases. Arch Neurol Psychiatry. 1944;51:533-543.
154. Treatment of tuberculosis. MMWR Recomm Rep. 2003;52(RR-11):1-77.
155. Controlled trial of short-course regimens of chemotherapy in the ambulatory treatment of spinal tuberculosis. Results at three years of a study in Korea. Twelfth report of the Medical Research Council Working Party on Tuberculosis of the Spine. J Bone Joint Surg [Br]. 1993;75(2):240-248.
156. Five-year assessment of controlled trials of short-course chemotherapy regimens of 6, 9 or 18 months’ duration for spinal tuberculosis in patients ambulatory from the start or undergoing radical surgery. Fourteenth report of the Medical Research Council Working Party on Tuberculosis of the Spine. Int Orthop. 1999;23(2):73-81.
157. Khoo L.T., Mikawa K., Fessler R.G. A surgical revisitation of Pott distemper of the spine. Spine J. 2003;3(2):130-145.
158. Faria F., Viegas F. Spinal brucellosis: a personal experience of nine patients and a review of the literature. Paraplegia. 1995;33(5):294-295.
159. Rajapakse C.N. Bacterial infections: osteoarticular brucellosis. Baillieres Clin Rheumatol. 1995;9(1):161-177.
160. Zaks N., Sukenik S., Alkan M., et al. Musculoskeletal manifestations of brucellosis: a study of 90 cases in Israel. Semin Arthritis Rheum. 1995;25(2):97-102.
161. Paz J.F., Alvarez F.J., Roda J.M., et al. Spinal epidural abscess caused by Brucella: case report. J Neurosurg Sci. 1994;38(4):245-249.
162. al-Shahed M.S., Sharif H.S., Haddad M.C., et al. Imaging features of musculoskeletal brucellosis. Radiographics. 1994;14(2):333-348.
163. Skalsky K., Yahav D., Bishara J., et al. Treatment of human brucellosis: systematic review and meta-analysis of randomised controlled trials. BMJ. 2008;336(7646):701-704.
164. Katonis P., Tzermiadianos M., Gikas A., et al. Surgical treatment of spinal brucellosis. Clin Orthop Relat Res. 2006;444:66-72.
165. Voisin L., Vittecoq O., Mejjad O., et al. Spinal abscess and spondylitis due to actinomycosis. Spine (Phila Pa 1976). 1998;23(4):487-490.
166. Corti M.E., Villafane-Fioti M.F. Nocardiosis: a review. Int J Infect Dis. 2003;7(4):243-250.
167. Workowski K.A., Berman S.M. Sexually transmitted diseases treatment guidelines, 2006. MMWR Recomm Rep. 2006;55(RR-11):1-94.
168. Porter S.D., Noble M.A., Rennie R. A single strain of Candida albicans associated with separate episodes of fungemia and meningitis. J Clin Microbiol. 1996;34(7):1813-1814.
169. Ferra C., Doebbeling B.N., Hollis R.J., et al. Candida tropicalis vertebral osteomyelitis: a late sequela of fungemia. Clin Infect Dis. 1994;19(4):697-703.
170. Assaad W., Nuchikat P.S., Cohen L., Esguerra J.V., Whittier F.C. Aspergillus discitis with acute disc abscess. Spine (Phila Pa 1976). 1994;19(19):2226-2229.
171. Pfausler B., Kampfl A., Berek K., et al. Syndrome of the anterior spinal artery as the primary manifestation of aspergillosis. Infection. 1995;23(4):240-242.
172. Walsh T.J., Anaissie E.J., Denning D.W., et al. Treatment of aspergillosis: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis. 2008;46(3):327-360.
173. Lang E.W., Pitts L.H. Intervertebral disc space infection caused by Aspergillus fumigatus. Eur Spine J. 1996;5(3):207-209.
174. Tang T.J., Janssen H.L., van der Vlies C.H., et al. Aspergillus osteomyelitis after liver transplantation: conservative or surgical treatment? Eur J Gastroenterol Hepatol. 2000;12(1):123-126.
175. Kline M.W., Bocobo F.C., Paul M.E., et al. Successful medical therapy of Aspergillus osteomyelitis of the spine in an 11-year-old boy with chronic granulomatous disease. Pediatrics. 1994;93(5):830-835.
176. Ford L.T., Key J.A. Postoperative infection of intervertebral disc space. South Med J. 1955;48(12):1295-1303.
177. Lagging L.M., Breland C.M., Kennedy D.J., Milligan, et al. Delayed treatment of pulmonary blastomycosis causing vertebral osteomyelitis, paraspinal abscess, and spinal cord compression. Scand J Infect Dis. 1994;26(1):111-115.
178. Chapman S.W., Dismukes W.E., Proia L.A., et al. Clinical practice guidelines for the management of blastomycosis: 2008 update by the Infectious Diseases Society of America. Clin Infect Dis. 2008;46(12):1801-1812.
179. Banuelos A.F., Williams P.L., Johnson R.H., et al. Central nervous system abscesses due to Coccidioides species. Clin Infect Dis. 1996;22(2):240-250.
180. Galgiani J.N., Ampel N.M., Blair J.E., et al. Coccidioidomycosis. Clin Infect Dis. 2005;41(9):1217-1223.
181. Gurevitz O., Goldschmied-Reuven A., Block C., et al. Cryptococcus neoformans vertebral osteomyelitis. J Med Vet Mycol. 1994;32(4):315-318.
182. Abba K., Ramaratnam S., Ranganathan L.N. Anthelmintics for people with neurocysticercosis. Cochrane Database Syst Rev. 2010;1:CD000215.
183. Altinors N., Bavbek M., Caner H.H., Erdogan B. Central nervous system hydatidosis in Turkey: a cooperative study and literature survey analysis of 458 cases. J Neurosurg. 2000;93(1):1-8.
184. Baykaner M.K., Dogulu F., Ozturk G., et al. A viable residual spinal hydatid cyst cured with albendazole. Case report. J Neurosurg. 2000;93(Suppl 1):142-144.
185. Schnepper G.D., Johnson W.D. Recurrent spinal hydatidosis in North America. Case report and review of the literature. Neurosurg Focus. 2004;17(6):E8.
186. Jiang Y.G., Zhang M.M., Xiang J. Spinal cord Schistosomiasis japonica: a report of 4 cases. Surg Neurol. 2008;69(4):392-397.
187. Silva L.C., Maciel P.E., Ribas J.G., et al. Treatment of schistosomal myeloradiculopathy with praziquantel and corticosteroids and evaluation by magnetic resonance imaging: a longitudinal study. Clin Infect Dis. 2004;39(11):1618-1624.
188. Lambertucci J.R., Silva L.C., do Amaral R.S. Guidelines for the diagnosis and treatment of schistosomal myeloradiculopathy. Rev Soc Bras Med Trop. 2007;40(5):574-581.