Risk Factors for Neurologic Deficits in Patients With Spinal Epidural Abscess: An Analysis of One-Hundred-Forty Cases

Abstract

Study Design

retrospective study.

Objectives

In addition to surgical treatment of spinal epidural abscesses (SEA), a conservative, medical treatment for patients without acute neurologic deficits has been proposed. However, the risk factors for neurologic deficits are unclear. This study aims to identify factors predisposing patients with SEA to neurological impairment.

Methods

All patients treated for SEA between 2008 and 2021 were identified from a prospective vertebral-osteomyelitis registry of a tertiary referral centre. Patient demographics, comorbidities, pathogens, degree of osseous destruction, location of SEA and preoperative neurologic status were retrospectively collected. Differences between patients with (Group 1) and without (Group 2) pretreatment neurologic deficits were assessed by univariate and logistic regression analysis.

Results

A total of 140 patients with SEA were included. Forty-three patients (31%) had a neurologic deficit and 97 patients (69%) had no neurologic deficit prior to therapy. The prevalence of diabetes mellitus (35% vs 19%, P = .03), median visual analogue scale leg pain (8 vs 5, P = .01), median American Society of Anesthesiologists (ASA) Score (3 vs 2.6, P = .003) and mean Body-Mass-Index (29 vs 26, P = .02) differed between Group 1 and 2 in univariate analysis. In multivariable analysis, diabetes mellitus (odds ratio = 2.7), female sex (odds ratio = 2.5) and ASA-Score (odds ratio = 2.4) were significant contributors for neurologic deficits.

Conclusions

In patients with a SEA without neurologic deficits, the ASA score and diabetes mellitus should be considered, especially in female patients. These patients may be at a higher risk for developing a neurologic deficit and may benefit from an early surgical treatment.

Keywords

spinal epidural abscess conservative treatment risk factors neurologic deficit spondylodiscitis vertebral osteomyelitis

Introduction

Spinal epidural abscess (SEA) may present as an isolated event or as a complication of vertebral osteomyelitis (VO). Although SEA is a rare condition, its incidence has increased in recent decades^1,2 due to demographic factors (e.g., an aging population and increased morbidity) and diagnostic improvements such as magnetic resonance imaging (MRI). A vast number of risk factors for the development of SEA have been described in the past,³ the most prevalent ones being diabetes mellitus, intravenous drug abuse, skin infections, and previous invasive procedures (e.g., epidural anaesthesia, extraspinal and spinal surgery, and interventions). Due to unspecific initial symptoms such as backpain and/or fever, early diagnosis is challenging,⁴ and a delay in treatment or treatment failure can have devastating outcomes.⁵ Weakness of the voluntary musculature or sphincter dysfunction is initially present in only a minority of patients.³ If SEA manifests with an acute neurologic deficit (existing no longer than 24-72 hours^2,6), early surgical decompression with or without stabilization and fusion is indicated, in combination with anti-infective treatment. In recent years, there has been a shift in the treatment of SEA without neurologic deficits, with some studies advocating a conservative, medical regime.^7-9 However, this approach remains highly controversial.^10-12 In addition, patients assigned to a conservative treatment need a close clinical observation in case the abscess formation progresses and neurologic deficits occur. A number of studies have attempted to find risk factors for failure of non-operative treatment^13-16 in a subset of patients initially treated conservatively, but the results are inconsistent. This may be due to a nonuniform definition of “treatment failure” and different sets of potential risk factors that were examined. Therefore, it is unclear which patients may benefit from surgical therapy even in the absence of initial neurologic impairment.

In this study, we aim (1) to record the demographic and clinical characteristics of patients with SEA and (2) to identify factors that predispose to the development of neurologic deficits prior to treatment. Further insights may facilitate treatment algorithms for the conservative and surgical management of patients with SEA who do not have neurologic deficits at presentation.

Methods

Patient Selection

From January 2008 to January 2021, all consecutive adult patients with VO were prospectively included in a registry of a tertiary referral center using the infrastructure of the “European Spine Tango” registry and the “Deutsche Wirbelsäulengesellschaft (DWG)” registry. A positive ruling from the Ethics-Committee of the Medical Faculty of the University of Cologne (Ruling No. 09-182) was obtained. Written informed consent from each patient for inclusion in the registry was obtained. Patients were diagnosed with VO based on the presence of characteristic back and/or leg pain plus characteristic MRI, abscess, or vertebral body destruction detected by computed tomography (CT). All cases were discussed in an interdisciplinary board between an infectious disease specialist and an orthopaedic surgeon to confirm the diagnosis of VO. Of these, all patients with a diagnosis of SEA confirmed by MRI or CT were identified. Exclusion criteria were incomplete medical records for the factors studied.

Data Collection

After inclusion, basic demographic data such as age, sex, body-mass-index (BMI), level of infected segment, and American Society of Anesthesiologists (ASA) Score were collected prospectively. In addition, the number of affected segments, psoas abscess, preoperative neurologic status (Frankel Scale), initial back pain (measured by visual analogue scale, VAS), initial leg pain, C-reactive Protein (CRP), hemoglobin, albumin, prothrombin time, platelet count, white blood cell count, creatinine, glomerular filtration rate (GFR), causative pathogens, bacteremia, previous spine surgery, and comorbidities (Diabetes mellitus, COPD, malignancy rheumatism, congestive heart failure, chronic kidney failure, alcoholism, intravenous drug abuse, endocarditis) were retrospectively collected. The amount of osseous destruction was quantified by applying the classification by Eysel and Peters¹⁷ on lateral radiographs of the affected spine segment.

Based upon their initial neurologic status, patients were either grouped into Group 1 (impaired neurologic function, Frankel Scale A-D) or Group 2 (intact neurologic function, Frankel Scale E).

Statistical Analysis

Patient demographics and clinical factors were analyzed descriptively and are reported as either mean (± standard deviation), median (interquartile range) or number (percentage). The distribution of independent variables between Groups 1 and 2 was analyzed by univariate tests: Student’s t-Test for continuous variables, Mann-Whitney-U-Test for ordinal variables, and Chi-Square-Test for dichotomous variables. Level of significance was set at P ≤ .05. All variables with a P ≤ .1 in univariate testing were included in a binary logistic regression model with stepwise backward elimination. Odds ratios for variables that were significant in logistic regression (P ≤ .05) are reported with 95% confidence intervals.

Results

Figure 1 shows the process of patient selection. A total of 140 patients with SEA were included in the study. The mean age of the study population was 66 (±11.6) years with a range of 20 to 92. There were 99 (71%) male and 41 (29%) female patients. Forty-three of the 140 patients (33%) presented with a neurologic deficit (Frankel Scale A-D) and were defined as Group 1. Two (5%) of them were Frankel A, 13 (30%) were Frankel B, 15 (35%) were Frankel C, and 13 (30%) were Frankel D. The remaining 97 of the 140 patients (68%) had no neurologic deficit at first presentation (Frankel Scale E) and were defined as Group 2. A causative pathogen could be found in 82% of all 140 cases, the most common one being Staphylococcus aureus (46%). The lumbar spine was solely affected in 90 patients (64%), the mean number of affected segments was 2.8 (±.7). Forty-five patients (32%) had a concomitant abscess in the psoas muscle. Twelve patients (8%) had discitis without erosion of bony structures.

Figure 1.

Flowchart of patient selection for data analysis. 1: spinal epidural abscess.

One-hundred-thirty-two patients (94%) had back pain with a median intensity of 9 (8-10) on the VAS. One-hundred-eight patients (76.9%) had leg pain, with a median intensity of 6 (1-9) on the VAS.

In Group 1, 42 of 43 patients (98%) were treated operatively. One patient decided against surgery. In Group 2, 93 of 97 patients (96%) were treated operatively. Of the remaining four, two decided against surgery, one patient was deemed too morbid for surgery and one patient suffered a myocardial infarction before scheduled surgery. The rates (98% vs 96%) did not differ significantly (P = .8).

Table 1 shows the demographic characteristics as well as clinical factors for each group and the results of univariate analysis. Table 2 shows the distribution of osseous destruction according to the classification by Eysel and Peters.¹⁷ Univariate analysis revealed that the prevalence of diabetes mellitus, leg pain, ASA score, and BMI were significantly differently distributed between the two groups. These factors as well as sex (P ≤ .1) were included in the binary logistic regression model. In the regression model, the presence of diabetes mellitus (P = .02), ASA score (P = .01), and female sex (P = .03) were significant. Figure 2 and Table 3 depict the odd ratios with 95%-confidence-intervals of the significant variables.

Table 1.

Patient Demographics and Clinical Factors for Group 1 (Patients with Neurologic Deficit) and Group 2 (Patients without Neurologic Deficit), as Well as P-Values of Univariate Testing.

	Group 1 (n = 43)	Group 2 (n = 97)	P-value
Age, (years)	65 (±10.4)	66 (±12.3)	.69
Sex
Male, n (%)	26 (61)	73 (75)	.08**
Female, n (%)	17 (39)	24 (25)	.08**
BMI¹ (kg/m²)	29 (±7.1)	26 (±5.2)	.02*
ASA² grade	3 (3-3)	3 (2-3)	.003*
Leg pain (VAS³)	8 (5-10)	5 (0-8)	.01*
Back pain (VAS)	8 (7-10)	9 (8-10)	.36
Number of affected segments	1.5 (±1)	1.3 (±.5)	.82
Localization of affected segment
cervical, n (%)	3 (7)	4 (4)	.88
upper thoracic (Th1–Th6), n (%)	2 (5)	7 (7)	.57
lower thoracic (Th7–Th12), n (%)	9 (21)	13 (13)	.26
lumbar, n (%)	25 (58)	65 (67)	.31
multilocal, n (%)	4 (9)	8 (8)	.84
Psoas abscess, n (%)	13 (30)	32 (33)	.75
Pathogen detected, n (%)	38 (88)	77 (80)	.20
Bacteremia, n (%)	16 (37)	40 (40)	.74
Previous spine surgery, n (%)	13 (30)	18 (19)	.13
Previous back infiltration, n (%)	5 (12)	17 (18)	.17
MRSA⁴, n (%)	2 (5)	4 (4)	.89
Diabetes mellitus, n (%)	17 (40)	18 (19)	.008*
Endocarditis, n (%)	3 (7)	3 (3)	.30
COPD⁵, n (%)	5 (12)	11 (11)	.96
Congestive heart failure, n (%)	7 (16)	13 (13)	.65
Chronic kidney failure, n (%)	4 (9)	12 (12)	.60
Rheumatism, n (%)	2 (5)	8 (8)	.45
Malignancy, n (%)	6 (12)	17 (18)	.60
Alcoholism, n (%)	5 (12)	8 (8)	.53
IV drug use, n (%)	3 (7)	6 (6)	.86
CRP⁶ (mg/L)	119 (±94)	105 (±100)	.42
Hemoglobin (g/dL)	10.7 (±1.9)	11.3 (±2.1)	.15
White Blood cell count (10⁹/L)	12.2 (±5.1)	10.8 (±8.1)	.29
Platelet count (10⁹/L)	336 (±122)	304 (±149)	.16
Serum-creatinine (mg/dL)	.9 (±.5)	1.1 (±.7)	.30
GFR⁷ (ml/min)	81 (±35)	79 (±29)	.74
Serum-albumin (g/L)	30 (±8)	33 (±7)	.13
Prothrombin time ratio (%)	87 (±19)	93 (±20)	.11

Data is reported as either mean (± standard deviation), median (interquartile range) or absolute number (percentage). *P ≤ .05; **P ≤ .1

¹Body-Mass-Index.

²American Society of Anesthesiologists.

³Visual Analogue Scale.

⁴Methicillin-resistant Staphylococcus Aureus.

⁵Chronic, Obstructive, Pulmonary Disease.

⁶C-reactive Protein.

⁷Glomerular Filtration Rate.

Table 2.

Distribution of Osseous Destruction in Group 1 (Neurologic Impairment) and Group 2 (no Neurologic Impairment) According to the Classification by Eysel and Peters.¹⁷

Stage	Description	Group 1 (n = 43)	Group 2 (n = 97)	P-value
I	decrease of the intervertebral space	3 (7%)	9 (9%)	0.5
II	Erosion of end plates	22 (51%)	59 (61%)	0.1
III	Spinal deformity with kyphosis	18 (42%)	29 (30%)	0.9
IV	Reactive bone formation, ankylosis with kyphotic malalignment	0	0	—

Data is presented as absolute number (percentage).

Figure 2.

Forest plot depicting the odds ratios (●) and 95% confidence intervals (├─┤) of significant factors in the multivariable analysis. 1: American Society of Anesthesiologists.

Discussion

In this retrospective study of a tertiary referral center of 140 VO patients with spinal epidural abscess, we aimed to characterize the patient population suffering from this disease and find risk factors for a neurologic deficit. According to our findings, SEA patients with diabetes mellitus, female patients and patients with a higher ASA score are at a greater risk for a neurologic deficit.

With a broad age distribution, a male predominance and S. aureus as the most common pathogen, our cohort is typical of populations in other studies on SEA.³

The incidence of SEA has nearly doubled in recent decades. Factors such as an increasing life expectancy, widespread availability of MRI (the diagnostic gold standard) and the prevalence of spinal procedures have been attributed to this increase.² In everyday clinical practice a minority of patients presents with the classic clinical triad of back pain, fever, and neurologic impairment.¹⁸ Only 26% of SEA patients present with beginning neurologic deficits, such as weakness of the voluntary musculature, and only 30% of patients present with a sphincter dysfunction.³ Once a neurologic deficit has occurred, the outcome is significantly worse, and even surgical treatment may still result in sequalae.³ Yao et al¹⁹ report that 80% of patients showed an improvement of neurologic deficits after surgery in their retrospective study on 53 cases. Although no patient had a deterioration of neurologic deficits after surgery, 40% had residual neurologic deficits 3.5 years after surgery. Patients with conservative treatment and no initial neurologic deficit show a deterioration of neurologic function in 25%-38% of cases.^15,16 Seventy-three percent of patients with complete neurologic impairment after SEA achieved a relevant improvement after receiving inpatient rehabilitation, but none had a complete recovery, according to Koo et al.²⁰ Early surgical intervention for acute neurologic deficits leads to significantly better outcomes in terms of neurologic recovery and long-term quality of life.²¹ The treatment of choice of SEA has traditionally consisted of decompressive surgery and administration of anti-infective therapy. However, in recent years, a trend toward conservative treatment of SEA in patients without neurologic deficits can be observed.^7-9 Although a number of studies have found that this subset of patients does not perform worse with successful medical treatment,^2,21-27 the rate of failure and conversion to surgery is quite high, up to 49%.^13,14,26-28 The rate of surgically treated patients with SEA was quite high in our study and did not differ significantly between patients with and without neurologic deficit (98% vs 96%). The high rate of surgery in patients without neurologic deficit was due to the high amount of osseous destruction in this group (91%, Table 2). Although radiographs have a sensitivity of only 82% and specificity of only 57%,²⁹ we used the classification by Eysel and Peters¹⁷ to quantify the amount of vertebral destruction, because it is the only published radiological classification for this purpose.³⁰

The evidence on conservative treatment of SEA is based solely on retrospective case-series. To our knowledge, no prospective observational studies exist, probably due to ethical reasons. There are four retrospective studies^13-16 that investigated the risk factors for failure of conservative treatment of SEA. They report inconsistent results. In a study of 128 conservatively treated patients with SEA, Patel et al¹³ identified diabetes, CRP greater than 115 mg/L, white blood cell count greater than 12.5 × 10⁹/L, and bacteremia as risk factors for treatment failure, defined as worsening neurologic function or increased/intolerable pain. Kim et al¹⁴ in a study of 142 conservatively treated patients, found diabetes mellitus, age older than 65 years, incomplete or complete spinal cord deficits, and methicillin-resistant Staphylococcus aureus (MRSA) as risk factors for failure of treatment. Kim et al. defined failure of treatment as persistent sepsis, the development of increasing back pain and/or neurologic deficits, progression of SEA on serial radiographs, or death despite antibiotic treatment for more than a week. Diabetes mellitus, sensory changes, active malignancy, an existing motor deficit, pathologic or compression fracture in the affected levels, and dorsal location of the abscess were the risk factors for treatment failure found by Shah et al¹⁶ in a retrospective analysis of 367 cases with initial conservative treatment of SEA. More recently, Hunter et al¹⁵ in a retrospective study of 58 conservatively treated SEA patients, found that Maori ethnicity (their study was conducted in New Zealand) and elevated white blood cell count were associated with treatment failure (defined as worsening neurology, lack of clinical improvement, or failure of laboratory markers to respond). The risk factor models by Patel and Kim were externally validated by Stratton et al³¹ on their retrospective cohort of 75 patients with SEA initially treated conservatively. The ability to predict treatment failure was found to be poor for both models, with Patel’s model having a slightly better ability to discriminate between treatment success and failure.

Table 3.

Results of the Multivariable Analysis.

Factor	P-Value	Odds Ratio [95%-CI]
BMI¹	.14	1.05 [.9 - 1.1]
ASA²	.01	2.4 [1.2 - 4.6]
Diabetes	.02	2.7 [1.1 - 6.7]
Leg pain	.30	1.9 [.5 - 6]
Female sex	.03	2.5 [1.1 - 5.8]

¹Body-Mass-Index.

²American Society of Anesthesiologists.

Our study focused on the presence of neurologic deficits at the time of diagnosis, as this is a clear and undisputed indication for surgical treatment. The equally high rate of surgery in both groups should not be relevant for the further analysis.

However, it is a somewhat different outcome parameter than the occurrence or worsening of neurologic deficits during conservative treatment in the above four studies. This limits comparability of their results. Furthermore, while the occurrence or worsening of neurologic deficits was included in all definitions of treatment failure in these studies, other factors were also considered (e.g., death, ongoing pain, elevated inflammation markers). In addition, some clinical factors considered by us were not included in any of the four studies by Hunter, Kim, Patel and Shah, like ASA score, endocarditis, COPD, rheumatism, congestive heart failure, GFR, creatinine and vice versa (sepsis, fever, intraspinal localization of abscess, hemodialysis, ethnicity, smoking, HIV status).

However, we believe that from a pathophysiologic perspective, the factors associated with the onset or worsening of neurologic deficits during treatment of SEA may be similar to the factors associated with a neurologic deficit at diagnosis of SEA. The significant factors we found (female sex, ASA score, diabetes mellitus) are all factors that are relatively stable over time.

Kim, Patel and Shah, but not Hunter, found diabetes mellitus to be a risk factor for treatment failure. We were able to confirm this finding. In our study, presence of diabetes mellitus had an odds ratio of 2.6. High Body-Mass-Index had a significantly different distribution in patients with and without neurologic deficit in univariable analysis, but not in multivariable analysis. This may be due to the high correlation of obesity and diabetes mellitus.⁹

The ASA classification system is a common tool for assessing a patient’s health status prior to surgery and is routinely used by anesthesiologists worldwide.³² It is a six-point scale ranging from Grade 0 (healthy patient) to Grade 6 (brain dead patient) based on the evaluation of a patient’s systemic medical conditions. It is related to postoperative outcomes in numerous fields of surgery.³³ While the grading is simple to apply, it has been described as partially inconsistent, with low to moderate interrater reliability.^34,35 This is probably due to its subjective nature. Furthermore, Daabiss³³ pointed out that the grading’s relying on “systemic” diseases may be confusing to the user (e.g., a heart attack is strictly spoken not a systemic disease, but has severe effects on overall health). Surprisingly, in our study, a higher ASA score was also associated with neurologic deficits, with an odds ratio of 2.6. This may indicate that more morbid patients are at a higher risk for developing a neurologic deficit. A reason for this may be an association of high morbidity with immunosuppression. On the other hand, this finding may be due to confounding, as a neurologic deficit – although not a “systemic” condition – may have resulted in a higher ASA score. In our cohort, 70% of patients in Group 1 had a Frankel Scale of C or lower, indicating the inability to walk. ASA Score was not considered in the studies by Hunter,¹⁵ Kim,¹⁴ Patel,¹³ and Shah¹⁶ and further investigation of this in future studies has to prove the validity of our finding.

A male dominance in SEA has been described in the meta-analysis by Reihsaus et al,³ and this is also true for our ungrouped study population which consisted of 71% male patients. Despite this overrepresentation of male sex, female sex was found to be associated with a neurologic deficit in our logistic regression model, with an odds ratio of 2.8. Female sex as a risk factor has not yet been described. A reason for a higher risk of neurologic deficits in female patients may be sex differences in spinal canal anatomy. Women have a narrower lumbar spinal canal and a smaller epidural space compared to men.³⁶ Hence, a smaller epidural mass may be needed to cause a symptomatic stenosis of the spinal canal in women.

While back pain is the typical and most common SEA symptom described, leg pain is usually not surveyed on or reported in studies.³ A large majority of patients of our cohort had strong leg pain, which was also a significant risk factor in univariate analysis. Radicular pain may be an expression of a beginning stenosis of the spinal canal in the early stadium of SEA, where the abscess affects the nerve root but is not yet large enough to cause more severe symptoms. It may therefore be seen as a warning sign or precursor of neurologic deficits. Shah et al¹⁶ reported an incidence of leg pain of 73% in their cohort, which was similar to the incidence in our cohort (77%). However, in Shah’s study, it was not a significant factor in univariable analysis. Since leg pain would be an easily queried, potential risk factor, it should be considered in future studies for additional research on this.

Limitations

Our study has several limitations. Although data on patient demographics were extracted from a prospective registry, most clinical data were collected retrospectively. Incomplete medical records were excluded and may contribute to selection bias.

Our cohort is from a tertiary referral center specializing in spine surgery. Therefore, the study population may not be representative of a general SEA population. Transfer of our results to conservatively treated SEA patients should be done with caution.

Although comparable to other studies on risk factors for SEA, the size of our study population may be too low to find statistical significance for some factors that had borderline P-values.

Conclusion

In patients with a SEA without neurologic deficits, the ASA score and the presence of diabetes mellitus should be considered, especially in female patients. This subset of patients may be at a higher risk for developing a neurologic deficit and may benefit from an early surgical treatment. In general, research on risk factors for developing neurologic deficits in SEA yields inconsistent results and is solely based on retrospective studies. While our study confirmed diabetes mellitus as a known risk factor, other previously reported risk factors were not found to be significant in our study. On the other hand, female sex and ASA Score have not been previously found to be risk factors. These findings should be validated in future prospective observational studies.

Footnotes

Author Contributions

AY and KS conceived the study and were involved in conception and planning of the study design. NJ, KZ and AY supervised the collection of the data. KS and NK extracted the data from medical records, KS did the statistical analysis. The collected data and results were discussed and interpreted with KS, NK, KZ, NJ, CH, PE and AY. All authors wrote and revised the manuscript. All authors have read and approved of the final submitted manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

positive ruling from the Ethics-Committee of the Medical Faculty of the University of Cologne (No. 09-182) was obtained

ORCID iD

Krishnan Sircar

References

Rigamonti

Liem

Sampath

, et al. Spinal epidural abscess: Contemporary trends in etiology, evaluation, and management. Surg Neurol. 1999;52(2):189-197. doi:10.1016/S0090-3019(99)00055-5.

Darouiche

. Spinal epidural abscess. N Engl J Med. 2006;355(19):2012-2020. doi:10.1056/NEJMRA055111.

Reihsaus

Waldbaur

Seeling

. Spinal epidural abscess: A meta-analysis of 915 patients. Neurosurg Rev. 2000;23(4):175-204. doi:10.1007/PL00011954.

Davis

Wold

Patel

, et al. The clinical presentation and impact of diagnostic delays on emergency department patients with spinal epidural abscess. J Emerg Med. 2004;26(3):285-291. doi:10.1016/J.JEMERMED.2003.11.013.

Tuchman

Pham

Hsieh

. The indications and timing for operative management of spinal epidural abscess: Literature review and treatment algorithm. Neurosurg Focus. 2014;37(2). doi:10.3171/2014.6.FOCUS14261.

Leys

Lesoin

Viaud

, 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. doi:10.1002/ANA.410170408.

Ziu

Dengler

Cordell

Bartanusz

. Diagnosis and management of primary pyogenic spinal infections in intravenous recreational drug users. Neurosurg Focus. 2014;37(2):E3. doi:10.3171/2014.6.FOCUS14148.

Smith

Kochar

Manjila

, et al. Holospinal epidural abscess of the spinal axis: Two illustrative cases with review of treatment strategies and surgical techniques. Neurosurg Focus. 2014;37(2). doi:10.3171/2014.5.FOCUS14136.

Adogwa

Karikari

Carr

, et al. Spontaneous spinal epidural abscess in patients 50 years of age and older: A 15-year institutional perspective and review of the literature: clinical article. J Neurosurg Spine. 2014;20(3):344-349. doi:10.3171/2013.11.SPINE13527.

10.

Connor

Chittiboina

Caldito

Nanda

. Comparison of operative and nonoperative management of spinal epidural abscess: A retrospective review of clinical and laboratory predictors of neurological outcome. J Neurosurg Spine. 2013;19(1):119-127. doi:10.3171/2013.3.SPINE12762.

11.

Epstein

. What are we waiting for? An argument for early surgery for spinal epidural abscesses. Surg Neurol Int. 2015;6(Suppl 19):S504-S507. doi:10.4103/2152-7806.166894.

12.

Epstein

. Timing and prognosis of surgery for spinal epidural abscess: A review. Surg Neurol Int. 2015;6(Suppl 19):S475-S486. doi:10.4103/2152-7806.166887.

13.

Patel

Alton

Bransford

Lee

Bellabarba

Chapman

. Spinal epidural abscesses: Risk factors, medical versus surgical management, a retrospective review of 128 cases. Spine J. 2014;14(2):326-330. doi:10.1016/j.spinee.2013.10.046.

14.

Kim

Melikian

, et al. Independent predictors of failure of nonoperative management of spinal epidural abscesses. Spine J. 2014;14(8):1673-1679. doi:10.1016/J.SPINEE.2013.10.011.

15.

Hunter

Cussen

Baker

. Predictors of failure for nonoperative management of spinal epidural abscess. Glob spine J. 2021;11(1):6-12. doi:10.1177/2192568219887915.

16.

Shah

Ogink

Nelson

Harris

Schwab

. Nonoperative management of spinal epidural abscess: Development of a predictive algorithm for failure. J Bone Joint Surg Am. 2018;100(7):546-555. doi:10.2106/JBJS.17.00629.

17.

Eysel

Peters

. Spondylodiscitis. In: Peters

Klosterhalfen

, eds. Bakterielle Infektionen Der Knochen Und Gelenke. New York, NY: Thieme; 1997:52-68.

18.

Krishnamohan

Berger

. Spinal epidural abscess. Curr Infect Dis Rep. 2014;16(11):1-6. doi:10.1007/S11908-014-0436-7.

19.

Yao

Lin

Chou

Wang

Liu

Chang

. Risk factors for residual neurologic deficits after surgical treatment for epidural abscess in the thoracic or lumbar spine. Spine J. 2020;20(10):1638-1645. doi:10.1016/J.SPINEE.2020.05.001.

20.

Koo

Townson

Dvorak

Fisher

. Spinal epidural abscess: A 5-year case-controlled review of neurologic outcomes after rehabilitation. Arch Phys Med Rehabil. 2009;90(3):512-516. doi:10.1016/J.APMR.2008.09.567.

21.

Behmanesh

Gessler

Quick-Weller

, et al. Early versus delayed surgery for spinal epidural abscess: Clinical outcome and health-related quality of life. J Korean Neurosurg Soc. 2020;63(6):757-766. doi:10.3340/JKNS.2019.0230.

22.

Arko

Quach

Nguyen

Chang

Sukul

Kim

. Medical and surgical management of spinal epidural abscess: A systematic review. Neurosurg Focus. 2014;37(2). doi:10.3171/2014.6.FOCUS14127.

23.

Eltorai

AEM

Naqvi

Seetharam

Brea

Simon

. Recent developments in the treatment of spinal epidural abscesses. Orthop Rev. 2017;9(2). doi:10.4081/OR.2017.7010.

24.

Karikari

Powers

Reynolds

Mehta

Isaacs

. Management of a spontaneous spinal epidural abscess: A single-center 10-year experience. Neurosurgery. 2009;65(5):919-923. doi:10.1227/01.NEU.0000356972.97356.C5.

25.

Tang

Lin

Liu

. Spinal epidural abscess—experience with 46 patients and evaluation of prognostic factors. J Infect. 2002;45(2):76-81. doi:10.1053/JINF.2002.1013.

26.

Savage

Holtom

Zalavras

. Spinal epidural abscess: Early clinical outcome in patients treated medically. Clin Orthop Relat Res. 2005;439:56-60. doi:10.1097/01.BLO.0000183089.37768.2D.

27.

Siddiq

Chowfin

Tight

Sahmoun

Smego

. Medical vs surgical management of spinal epidural abscess. Arch Intern Med. 2004;164(22):2409-2412. doi:10.1001/ARCHINTE.164.22.2409.

28.

Curry

Hoh

Amin-Hanjani

Eskandar

. Spinal epidural abscess: Clinical presentation, management, and outcome. Surg Neurol. 2005;63(4):364-371. doi:10.1016/J.SURNEU.2004.08.081.

29.

Modic

Feiglin

Piraino

, et al. Vertebral osteomyelitis: Assessment using MR. Radiology. 1985;157(1):157-166. doi:10.1148/RADIOLOGY.157.1.3875878.

30.

(DWG) DG für O und OC (DGOOC) und DW . Diagnostik und Therapie der Spondylodiszitis – S2 k-Leitlinie. Version 1.0. https://www.awmf.org/uploads/tx_szleitlinien/151-001l_S2k_Diagnostik-Therapie-Spondylodiszitis_2020-10.pdf

31.

Stratton

Faris

Thomas

. The prognostic accuracy of suggested predictors of failure of medical management in patients with nontuberculous spinal epidural abscess. Glob Spine J. 2018;8(3):279. doi:10.1177/2192568217719437.

32.

Irlbeck

Zwißler

Bauer

. ASA-klassifikation. Anaesthesist. 2017;66(1):5-10. doi:10.1007/s00101-016-0246-4.

33.

Daabiss

. American society of anaesthesiologists physical status classification. Indian J Anaesth. 2011;55(2):111-115. doi:10.4103/0019-5049.79879.

34.

Haynes

Lawler

PGP

. An assessment of the consistency of ASA physical status classification allocation. Anaesthesia. 1995;50(3):195-199. doi:10.1111/J.1365-2044.1995.TB04554.X.

35.

Sankar

Johnson

Beattie

Tait

Wijeysundera

. Reliability of the American society of anesthesiologists physical status scale in clinical practice. Br J Anaesth. 2014;113(3):424-432. doi:10.1093/BJA/AEU100.

36.

Pierro

Cilla

Maselli

Cucci

Ciuffreda

Sallustio

. Sagittal normal limits of lumbosacral spine in a large adult population: A quantitative magnetic resonance imaging analysis. J Clin Imaging Sci. 2017;7:35. doi:10.4103/JCIS.JCIS_24_17.