Management of spinal infection: a review of the literature
- 2.6k Downloads
Spinal infection (SI) is defined as an infectious disease affecting the vertebral body, the intervertebral disc, and/or adjacent paraspinal tissue and represents 2–7% of all musculoskeletal infections. There are numerous factors, which may facilitate the development of SI including not only advanced patient age and comorbidities but also spinal surgery. Due to the low specificity of signs, the delay in diagnosis of SI remains an important issue and poor outcome is frequently seen. Diagnosis should always be supported by clinical, laboratory, and imaging findings, magnetic resonance imaging (MRI) remaining the most reliable method. Management of SI depends on the location of the infection (i.e., intraspinal, intervertebral, paraspinal), on the disease progression, and of course on the patient’s general condition, considering age and comorbidities. Conservative treatment mostly is reasonable in early stages with no or minor neurologic deficits and in case of severe comorbidities, which limit surgical options. Nevertheless, solely medical treatment often fails. Therefore, in case of doubt, surgical treatment should be considered. The final result in conservative as well as in surgical treatment always is bony fusion. Furthermore, both options require a concomitant antimicrobial therapy, initially applied intravenously and administered orally thereafter. The optimal duration of antibiotic therapy remains controversial, but should never undercut 6 weeks. Due to a heterogeneous and often comorbid patient population and the wide variety of treatment options, no generally applicable guidelines for SI exist and management remains a challenge. Thus, future prospective randomized trials are necessary to substantiate treatment strategies.
KeywordsSpinal infection Spondylodiscitis Vertebral osteomyelitis Spinal epidural abscess Intramedullary abscess Subdural empyema
American Society of Anaesthesiologists
Erythrocyte sedimentation rate
Human immunodeficiency virus
Intramedullary spinal cord abscess
Magnetic resonance imaging
Positron emission tomography
Spinal epidural abscess
Single photon emission computed tomography
White blood cell
Spinal infection (SI) is defined as an infectious disease affecting the vertebral body, the intervertebral disc, and/or adjacent paraspinal tissue . It comprises infection due to a contiguous source (trauma, surgery) as well as due to hematogenous dissemination . SI represents 2–7% of all musculoskeletal infections. There is a bimodal allocation in patients suffering from SI, with one peak below 20 years and the other between 50 and 70 years of age [48, 77]. Gender-specific investigations report a male/female ratio ranging between 2:1 and 5:1 [35, 59]. The incidence varies between 1:20,000 and 1:100,000, and mortality rates range between 2 and 20% in developed countries [1, 7, 44, 70]. However, the incidence of SI has been increasing in the last decades . This may be related to improving imaging techniques and clinical diagnostics, to the progressive number of older patients suffering from chronic diseases, to an increased amount of intravenous drug abusers, and to the growing number of spinal surgery and instrumentation . Patients suffering from SI commonly present with unspecific symptoms, back pain being the most frequently reported (85%), followed by fever (48%) and paresis (32%) [23, 25, 59]. Due to the unspecific presentation, the early detection of SI is still challenging and sufficient treatment is complex.
Pathophysiology and classification
There are numerous factors which may facilitate the development of SI, only some of which can be influenced . Non-influenceable risk factors include but are not limited to advanced patient age, American Society of Anaesthesiologists (ASA) score, and comorbidities like obesity, diabetes mellitus, substance abuse, chronic infection (especially HIV), long-term systemic steroid use, a poor nutritional status, and immunologic incompetence [5, 17, 18]. The most important influenceable factor promoting SI is spinal surgery. Long duration of surgery, high blood loss, type of instrumentation (lumbar and posterior more frequently than cervical and anterior ), and the quantity of operations (revisions, multiple interventions) are significant risk factors, leading to a SI rate of 1–4% after spinal surgery . Particular attention should be paid to surgically treated obese patients, as a subcutaneous fat thickness > 50 mm may lead to a significantly higher risk of postoperative infection and colonization by microorganisms .
Spinal infections are caused by three major agents: bacteria—causing pyogenic infection, fungi—causing granulomatosis infection, and parasites, which are rare. The etiology of SI varies depending on the location of the infection, as different pathogens affect different localizations. Staphylococcus aureus is the most common agent when it comes to spinal epidural abscesses (30–80%). Further frequent pathogens are coagulase-negative Staphylococci and several species of Streptococci, which are also involved in spondylitis/spondylodiscitis and paraspinal abscesses. Likewise, Escherichia coli frequently causes spondylitis/spondylodiscitis and paraspinal infections. Fungal causes of SI are rarely found and particularly include Aspergillus spp., Candida spp., and Cryptococcus neoformans. Although analysis has improved and identification of pathogens is highly pursued, in one third of cases, no organism can be identified [33, 68].
Patients suffering from SI present with an insidious onset and unspecific symptoms including neck/back pain, fever (even though not in an early stage), painful dorsal flexion, and, contingently, neurological deficits [23, 25, 59]. Early diagnosis is hindered by the fact that 30–70% of patients with spondylitis/spondylodiscitis do not show any signs of prior infection . Due to the low specifity of signs, the delay of diagnosis remains an important issue. Even though diagnostic tools and procedures have improved, a delay of 2 to 6 months between first symptoms and diagnosis is reported. As a result, poor outcome is frequently seen [12, 28, 82]. Diagnosis should be supported by clinical, laboratory, and imaging findings [12, 28].
Several clinical routine markers are suitable for diagnosis and evaluation of treatment response. The C-reactive protein (CRP) is considered to be the most specific marker for treatment response, as it returns to normal levels rapidly after successful treatment. Furthermore, CRP is elevated in more than 90% of acute spondylodiscitis cases and is a sensitive marker for bacterial infection. The erythrocyte sedimentation rate (ESR) also is a sensitive marker for infection but has low specificity. It can be used as a marker for therapeutic response only to some degree, as ESR is still increased in 50% of patients with good clinical outcome . The white blood cell (WBC) count, however, is less useful than ESR and CRP, as the presence of a normal WBC count does not exclude the diagnosis of spinal infection . Procalcitonin (PCT), a promising marker to distinguish between bacterial and non-bacterial infection, shows lower sensitivity than CRP in patients with SI. With multiple infected sites, its sensitivity increases. Thus, patients with elevated PCT levels should be considered as suffering from combined infection, and sufficient antibiotic treatment is required . As soon as SI is suspected in a patient, blood and urine cultures should be collected before starting an empiric antibiotic therapy . Up to 59% of positive blood cultures identify the etiological microorganism in patients with monomicrobial pyogenic spondylodiscitis . Based on the fact that biopsy may be superior to blood culture, computed tomography (CT)-guided biopsy of infected tissue should be considered .
Magnetic resonance imaging (MRI) remains the most reliable method to diagnose spondylodiscitis, due to its high sensitivity (96%), high specificity (94%), and capability to provide detailed data on paraspinal tissues and the epidural space [51, 53, 57]. Typical findings in patients with spondylodiscitis are hypointense discs and vertebral bodies in T1-weighted images and hyperintense signals of the same structures in T2-weighted images.
Nevertheless, plain and flexion/extension X-rays should be performed in every baseline evaluation. However, the specificity of native X-rays in the diagnosis of spondylodiscitis is low (59%) and can at most detect irregularities of vertebral endplates and/or low intervertebral disc height in advanced cases . A possible instability during follow-up can be detected reliably by flexion/extension films. Bony changes, i.e., early changes of endplates, and bone necrosis can be detected most precisely by CT. However, it may take 3 to 6 weeks after the onset of symptoms for bony destruction to become evident, which entails a delay in diagnosis, and makes the CT a less expressive diagnostic instrument for SI [42, 85]. Nevertheless, CT is commonly used in CT-guided biopsy and prior to instrumentation procedures . Although MRI is the gold standard in diagnosis of SI, there is no pathognomonic finding on MRI that dependably discriminates between SI and possible neoplasm. In case of suspicion of SI, complementary diagnostic methods can be applied . Sequential bone/gallium imaging and 67Ga-SPECT constitute radionuclide options for SI. Since these two methods are characterized by low specificity, interest increased in fluoro-2-desoxy-D-glucose [18F] PET, which may be a promising technique for distinction, as degenerative changes and fractures normally do not demonstrate FDG uptake [30, 31]. At present, however, radionuclide diagnostics should be reserved for cases of uncertain diagnosis or for special follow-up .
Since positive blood cultures and imaging findings associated with clinical symptoms do not automatically confirm SI, the definite diagnosis can only be attained by microscopic or bacteriological examination of the affected tissue. Both blood cultures and imaging can be inconclusive or false negative, so that a definite confirmation of SI diagnosis should be aimed for to ensure the quality of management [24, 37, 43, 84].
Intramedullary spinal cord abscesses (ISCA) are very rare with about 120 reported cases since the first description in 1830 . Congenital dermal sinus is implicated as the leading cause of ISCA in children, whereas hematogenous spread of urogenital and lung infections or infective endocarditis is the most common cause in adults. Usually some underlying spinal (cord) pathology provides the basis for ISCA . The overall mortality of ISCA has improved significantly from 90% in 1944  to 4% in 2009 , mostly caused by the availability of sufficient diagnostic facilities, the use of antibiotic agents, and improved surgical techniques. Due to the small number of cases, there is no general algorithm for ISCA and treatment has to be chosen diligently according to the patient’s clinical status. Generally, early drainage and the rapid administration of intravenous antibiotics result in a good prognosis . Surgical drainage within 5 days after onset of symptoms may lead to a significantly better neurological outcome than conservative treatment or delayed drainage [40, 75].
Subdural empyema (SDE), known as an infection located between the dura and the arachnoid, is a very rare condition and mostly results from hematogenous infection or spread of infection from osteomyelitis . SDE may develop in the spinal canal where it is possible to cause rapid compression of the spinal cord. The diagnostic procedure of choice is MRI with gadolinium, followed by CT with myelography [50, 63]. Early surgical drainage followed by appropriate antibiotic therapy is the most promising treatment . Depending on the extension of the lesion, (hemi-)laminectomy over one or more levels may be necessary. In cases of wider spread, flavectomy or laminotomy at several different levels may be required in order to evacuate the infectious material [2, 50]. Generally, SDE represents an extreme medical and neurosurgical emergency and treatment requires prompt surgical drainage and antibiotic therapy, as morbidity and mortality in SDE relate directly to treatment delay. As soon as the central nervous system is directly involved, outcome is poor . Thus, the condition should be treated with great urgency .
Spinal epidural abscess (SEA) also is a life-threatening infection. Mortality rates are estimated at 5–16% worldwide, and less than 50% of surviving patients show full recovery. Although SEA is a relatively rare diagnosis (2.5–3.0/100,000), the incidence has nearly doubled in the last 50 years according to circumstances already mentioned above and an increasing incidence of SI in general [6, 71]. Demographically, male patients are affected more frequently than female patients (ratio 2:1) for unknown reasons. Furthermore, a peak is reported in the sixth decade of life . Spontaneous SEA is mostly a secondary complication of a primary SI like spondylodiscitis, spreading hematogenously via septic thrombosis of epidural veins . Alternatively, iatrogenic SEA is mainly associated with surgical spinal procedures and in particular with epidural anesthesia or spinal injections . Abscesses are most commonly located in the lumbar spine (48%), followed by the thoracic (31%) and the cervical spine (24%) . Mostly, SEAs present as a multisegmental (three to four segments) disease as germs are capable of spreading via the epidural space without anatomical resistance. SEA is found in multiple locations in 20% of cases; therefore, imaging of the whole spine is recommended . As soon as an active infection is suspected, treatment should be started immediately. Decompression combined with systemic antibiotics has been established as gold standard in the last decades, especially in patients with progressive infection and late diagnosis . Nevertheless, management should be guided by individual parameters, as recently good outcomes were also demonstrated for conservative treatment. Conservative management can be considered (1) in patients with severe comorbidities which limit surgical options, (2) in extremely lengthy SEA (holospinal), in which surgery may not make sense or may be unachievable, (3) when SEA is detected early and patients suffer from no or only minor neurologic deficits, or (4) when the patient presents with complete paralysis for more than 72 h . However, 6–49% of initially conservatively treated patients still show progression of neurologic deficits, so that treatment has to be reconsidered carefully after meticulous re-evaluation [6, 19]. Special attention should be paid to patients suffering from high-risk factors for failure of medical treatment. Those include diabetes mellitus, MRSA infection, neurological impairment involving the spinal cord, acute or progressive deficit, CRP > 115 mg/L, WBC count > 12.5 × 109 cells/L, ring-like enhancement on MRI, and bacteraemia. Non-operative treatment should only be offered with close monitoring as the initial line of treatment and does not guarantee the prevention of operative therapy . Surgical options for the treatment of SEA include minimally invasive or endoscopic procedures, multisegmental decompression, instrumentation, and ventral debridement of the disc. The choice of the surgical approach depends on the consistency of the space-occupying lesion: solid (granulation tissue) or liquid (abscess). Dorsally accessible abscesses call for a minimally invasive drainage via laminotomy and irrigation, whereas dorsally located granulation tissue should be treated with multisegmental decompression and/or resection. Apart from that, ventrally located granulation tissue, which is usually associated with spondylodiscitis, is most effectively treated by corpectomy, reconstruction, and instrumentation . Overall, patients being surgically treated show a good outcome in more than 60%. Those convincing results cause some authors to recommend surgical therapy of SEA even without associated neurological deficits and especially when lesions are located in the cervical or thoracic spine. As mortality of SEA is still reported to be 5–20%, therapeutic decisions have to be made promptly and thoughtfully and influencing outcome predictors have to be considered. It has to be realized that SEA treated with medical management alone has a high risk for failure, if the patient is older than 50 years and suffers from severe comorbidities like diabetes or MRSA infection and neurologic compromise. In the absence of these risk factors, non-operative management may be considered as the initial line of treatment, but only with very close monitoring [45, 64]. Nevertheless, surgery should stay the treatment of choice in any case of doubt.
Established surgical treatment options for spondylodiscitis
Minimally invasive/endoscopic debridement
Percutaneous instrumentation without debridement
Decompression and debridement plus instrumentation
Discectomy/corpectomy plus instrumentation
Complex anteroposterior reconstruction plus instrumentation
Additional antibiotic treatment (modified to intraoperative culture results) has to be administered for at least 6 to 12 weeks in all cases . The final result in conservative as well as in surgical treatment always is bony fusion, so that conservative management does not preserve segmental motion.
Extension of infection into the paraspinal tissues is commonly addressed by treating the underlying spinal infection. If surgical debridement of, e.g., spondylodiscitis is performed, adjacent anterior and lateral abscesses are usually drained as well. Antibiotic treatment reaches the paraspinal tissues generally easier than the non-vascularized disc space. Only in the case of larger abscess formation percutaneous CT- or ultrasound-guided drainage may be indicated .
The most common paraspinal site of infection, the iliopsoas abscess (IPA) can be subdivided in primary and secondary acquired disease. Primarily arisen IPA is a rare condition and mostly occurs due to hematogenous or lymphatic spread of a pathogen from a distant infectious site. On the contrast, the majority of cases seen represent secondary IPA, most commonly caused by inflammatory spinal or skeletal pathology . The recommended first-line therapy is broad spectrum antibiotic therapy, covering Staph. aureus as the most frequent pathogen accountable for IPA . Especially small abscesses (< 3 cm) may be treated efficiently with antibiotic therapy alone . In the case of progress or insufficient medical therapy, image-guided percutaneous drainage is a very effective and safe alternative to the open surgical drainage . Nevertheless, in secondarily acquired, multiloculated IPA accompanied by florid spinal infection, open surgical drainage plus treatment of the underlying cause is the treatment of choice .
Due to a heterogeneous and often comorbid patient population and the wide variety of treatment options, no generally applicable guidelines for SI exist and management remains a challenge. In synopsis of the reviewed articles and personal experience, medical and surgical options should always be considered, but the authors favor surgical treatment in case of doubt or progression. With the increasing incidence of spinal infection particularly in elderly and frail patients, our personal opinion supports early surgery, as many patients are admitted in the preseptic phase and often deteriorate in their general condition thereafter. What is more, in the early stage of infection, less invasive procedures may still be possible, while extensive resections may be required at later stages, which are commonly characterized by massive bony destruction. Even in severely affected patients with extensive infection and exceedingly high CRP levels, the authors nowadays tend towards surgery, if medically feasible. Nevertheless, future prospective randomized trials are necessary to substantiate treatment strategies.
Open access funding provided by University of Innsbruck and Medical University of Innsbruck.
No funding was received for this research.
Compliance with ethical standards
Conflict of interest
CT receives speaker honoraria and grant research support from medical implant companies. All other authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.
For this type of study, ethical approval is not required.
- 7.Artenstein AW, Friderici J, Holers A, Lewis D, Fitzgerald J, Visintainer P (2016) Spinal epidural abscess in adults : a 10-year clinical experience at a tertiary care Academic Medical Center. Open forum. Infect Dis 3(4):ofw191Google Scholar
- 24.Euba G, Narvaez JA, Nolla JM, Murillo O, Narvaez J, Gomez-Vaquero C et al (2008) Long-term clinical and radiological magnetic resonance imaging outcome of abscess-associated spontaneous pyogenic vertebral osteomyelitis under conservative management. Semin Arthritis Rheum 38(1):28–40CrossRefPubMedGoogle Scholar
- 44.Kehrer M, Pedersen C, Jensen TG, Lassen AT (2014) Increasing incidence of pyogenic spondylodiscitis: a 14-year population-based study. J Inf Secur 68(4):313–320Google Scholar
- 46.Klemencsics I, Lazary A, Szoverfi Z. (2016) Risk factors for surgical site infection in elective routine degenerative lumbar surgeries. Spine J [Internet]. Elsevier Inc.;1–7Google Scholar
- 55.Lin Y, Li F, Chen W, Zeng H, Chen A, Xiong W (2015) Single-level lumbar pyogenic spondylodiscitis treated with mini-open anterior debridement and fusion in combination with posterior percutaneous fixation via a modified anterior lumbar interbody fusion approach. J Neurosurg Spine 23(6):747–753CrossRefPubMedGoogle Scholar
- 58.Mückley T, Schütz T, Kirschner M, Potulski M, Hofmann G, Bühren V (2003) Psoas abscess: the spine as a primary source of infection. Spine (Phila Pa 1976) 28(6):E106–E113Google Scholar
- 64.Pradilla G, Ardila GP, Hsu W, Rigamonti D, Epidural abscesses of the CNS (2009) Lancet Neurol [Internet]. Elsevier Ltd 8(3):292–300Google Scholar
- 69.Schafer F, Mattle HP (1994) Neurologic manifestations of Staphylococcus Aureus infections. Analysis of 43 patients. Arch Neurol Psychiatr 145(1):25–29Google Scholar
- 70.Schoenfeld AJ, Wahlquist TC (2015) Mortality, complication risk, and total charges after the treatment of epidural abscess. Spine J [Internet] Elsevier Inc 15(2):249–255Google Scholar
- 81.Tschugg A, Lener S, Hartmann S, Rietzler A, Neururer S, Thomé C. (2017) Primary acquired spondylodiscitis shows a more severe course than spondylodiscitis following spine surgery: a single-center retrospective study of 159 cases. Neurosurg Rev 41(1):141–147Google Scholar
- 84.Turunc T, Demiroglu YZ, Uncu H, Colakoglu S, Arslan H (2007) A comparative analysis of tuberculous, brucellar and pyogenic spontaneous spondylodiscitis patients. J Inf Secur 55(2):158–163Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.