Abstract
Introduction
Delayed diagnosis and therapy initiation for pyogenic spondylitis can have severe and fatal consequences. Early diagnosis and intervention are crucial in the treatment of pyogenic spondylitis. This multicenter cross-sectional study with prospective case series aimed to identify factors influencing the time from symptom onset to the diagnosis of pyogenic spondylitis.
Methods
Patients hospitalized with pyogenic spondylitis between 2019 and 2023 were included. Patients were classified into 2 groups: the delayed diagnosis group (>30 days from the onset of initial symptoms to the diagnosis of pyogenic spondylitis) and the early diagnosis group (within 29 days). Risk factors for delayed diagnosis were analyzed.
Results
A total of 74 patients (42 men and 32 women; mean age: 70.2 years) from 5 institutions were included. Univariate analysis of risk factors for delayed diagnosis revealed that the significant risk factors included advanced age (p=0.03), low white blood cell count (p<0.01), low C-reactive protein level (p<0.05), and semi-rigid spinal level, based on the spinal instability neoplastic score classification (p=0.05). Multivariate analysis for delayed diagnosis showed that the location at the semi-rigid spinal level was a significant risk factor (p=0.02). The vertebral bone destruction rate and abscess cavity index in the delayed diagnosis group were significantly higher than those in the early diagnosis group (p<0.01 and p<0.01, respectively).
Conclusions
Significant risk factors for delayed diagnosis of pyogenic spondylodiscitis include infection at the semi-rigid thoracic spinal level. Early diagnosis of spondylodiscitis is crucial because delayed diagnosis can lead to progressive bone destruction and the formation of large abscesses. Increased awareness of thoracic spinal infections, which can easily delay diagnosis, could help in the early diagnosis and treatment of pyogenic spondylodiscitis.
Keywords: pyogenic spondylitis, delayed diagnosis, spinal instability neoplastic score, thoracic spine
Introduction
In recent years, the number of patients with spinal infections, including pyogenic spondylitis, spondylodiscitis, and epidural abscess, has increased due to the aging population and the growing number of immunocompromised patients1-3). Pyogenic spondylitis is one of the most difficult infectious diseases to treat because it often becomes intractable and can be fatal in severe cases3-5). The basic treatment for pyogenic spondylitis is antibiotic therapy, which requires long-term antibiotic administration for at least 1.5 months6). Progressive infections can also lead to increased abscess formation and bone destruction, often necessitating surgical treatment and prolonged therapy2,7).
The causes of intractable and severe pyogenic spondylitis include the following: (1) drug intolerance due to poor drug penetration to the infected lesions, owing to the lack of blood supply to the intervertebral discs where the infection originates. Additionally, multidrug-resistant bacterial infections, such as methicillin-resistant Staphylococcus aureus [MRSA], also contribute to drug intolerance8-10); (2) patient factors, such as comorbid medical conditions and immunocompromised states due to the aging population; and (3) delays in diagnosis and treatment11). The diagnosis of pyogenic spondylitis is often delayed because the initial symptoms, such as pain and fever, are nonspecific and often difficult to diagnose, even with imaging tests in the early stages of disease onset. Delayed diagnosis and treatment of pyogenic spondylitis can result in progressive bone destruction and the formation of intervertebral and epidural abscesses, potentially leading to paralysis and sepsis12). Therefore, early diagnosis and intervention are crucial for treating pyogenic spondylitis.
Magnetic resonance imaging (MRI) and blood tests are useful in diagnosing pyogenic spondylitis. It is critical to understand the risk factors for delayed diagnosis and perform these tests as early as possible to ensure early diagnosis. However, it remains unclear which conditions of pyogenic spondylitis contribute to a delayed diagnosis. This multicenter cross-sectional study with a prospective case series aimed to analyze factors influencing the time from symptom onset to diagnosis of pyogenic spondylitis, with the goal of aiding in its early detection. Additionally, the study evaluated the influence of the time from the onset of initial symptoms to diagnosis on vertebral bone destruction (VBD) and the formation of intervertebral abscesses, as observed through imaging.
Materials and Methods
Design and participants
We conducted a multicenter, cross-sectional study with a prospective case series on pyogenic spondylitis across 5 facilities with spine surgery specialists in the same province. This study was approved by the institutional review board of each facility, and informed consent was obtained from all patients.
Patients hospitalized with pyogenic spondylitis between 2019 and 2023 were included in this study. Pyogenic spondylitis was diagnosed by spine surgery specialists at the 5 facilities based on a comprehensive assessment of clinical examinations, blood tests, bacterial culture results, MRI, and computed tomography (CT) findings. Patients with pain in the affected area, a high inflammatory response on blood tests, and signal changes on MRI indicating vertebral osteomyelitis or abscess formation, or bone destruction suggestive of infection on CT were diagnosed with pyogenic spondylitis based on the Infectious Diseases Society of America (IDSA) guidelines2). Two spine surgery specialists at the main research facility confirmed that the registered cases had pyogenic spondylitis based on the clinical records, blood tests, and images. We excluded patients whose consent was difficult to obtain, those whose duration of symptom onset was unknown because they could not be interviewed (e.g., patients with dementia or delirium), and patients with iatrogenic infections such as postoperative infections or infections following epidural catheter placement or block procedures.
At the time of spinal infection diagnosis, patients' background information, including initial symptoms (such as neck pain, back pain, low back pain, fever, and fatigue), date of symptom onset, comorbidities, and physical activity, was obtained through medical history interviews. Performance status (score ranging from 0 to 4, with higher scores indicating greater limitations in daily life) was assessed based on this collected information13,14). Physical examination included an assessment of muscle strength in the extremities, paralysis, and ambulation to evaluate the Frankel classification (A-E)15,16). Blood tests, including white blood cell (WBC) count, C-reactive protein (CRP) level, and serum albumin levels, were performed to assess the inflammatory response and nutritional status. Blood cultures and specimens from the infection site were tested for culture and identification of the causative organism.
Analysis of risk factors for delayed diagnosis
A delayed diagnosis was defined as patients who required more than 30 days from the onset of initial symptoms to the diagnosis of pyogenic spondylitis. Pola et al.17) reported a median time to diagnosis of pyogenic spondylodiscitis of 30 days, which was used as the basis for defining a delayed diagnosis group. Patients were classified into 2 groups based on this criterion: the delayed diagnosis group (>30 days) and the early diagnosis group (≤29 days), and the risk factors for the delayed diagnosis were analyzed.
The risk factors for delayed diagnosis were assessed in relation to the following factors. Patient factors included sex (male or female, with male as the reference), age (in years), presence of comorbid medical problems including diabetes mellitus, malignant neoplasm, hypertension, cardiovascular disease, and kidney disease (yes/no, with no as reference), clinical assessment including Frankel classification (A-C, or D-E, with A-C as the reference), and performance status (0-4). Infection factors included causative organism identification, multidrug-resistant bacterial infection (yes/no, with no as reference), blood test results including WBC (×104 cells/μL), CRP (mg/dL), albumin (g/dL), and infected spine level (classified into the following groups: junctional [occiput-C2, C7-T2, T11-L1, L5-S1], mobile [C3-C6, L2-L4], and semi-rigid [T3-T10] based on the Spinal Instability Neoplastic Score [SINS])7,18).
Imaging evaluation was performed at the start of inpatient treatment and compared between the delayed and early diagnosis groups. The extent of bone destruction and the size of the abscess cavity were assessed using CT and MRI findings as follows: the abscess cavity index (ACI) was evaluated on T2-weighted lumbar sagittal MRI images and measured as the height of the abscess cavity showing a high-signal area as a percentage of the height of the infected disc spinal unit, which includes the height of the abscess cavity and the cephalocaudal vertebral body (Fig. 1a)12). The rate of VBD was measured as the ratio of the height of the bone destruction area to the normal vertebral body height on sagittal lumbar spine CT images (Fig. 1b)12). Three slices of CT and MRI midsagittal images and 1/4 from the lateral side of the vertebra on their left and right sides were used for these measurements, and the mean values were used for evaluation. In the case of multilevel spinal infections, the ACI and VBD of each spinal unit were measured, and the maximum value was evaluated12).
Figure 1.
(a) Sagittal computed tomography/vertebral bone destruction rate: height ratio of bone-destruction area (B) to normal vertebral body height (A). (b) T2-weighted sagittal magnetic resonance imaging/abscess cavity index: height ratio of the abscess cavity (B) to disc-spinal unit (A+B+C).
Statistical analyses
The results are expressed as the mean±standard deviation. Statistical analyses were performed using Fisher's exact probability test and binomial logistic regression analysis, as appropriate. Multivariate analyses of risk factors for delayed diagnosis were performed using the following procedure. First, binomial logistic regression analysis was performed as a univariate analysis to quantify the association between candidate predictors and delayed diagnosis. Multinomial logistic regression was performed on the variables identified as significant in the univariate analysis. For all analyses, a p-value <0.05 was considered statistically significant.
Results
Patients
Overall, 74 patients (42 men and 32 women; mean age: 70.2 years) from 5 institutions were included in this study. The diseases consisted of 68 cases of pyogenic spondylodiscitis and 6 cases of epidural abscess without spondylodiscitis. The infection levels were as follows: cervical spine in 4 patients, thoracic spine in 14 patients, and lumbar spine in 61 patients (including duplicate cases). According to the SINS criteria, 20 patients (27.0%) were classified as junctional, 45 patients (60.8%) as mobile, and 9 patients (12.2%) as semi-rigid at the spine level of infection.
Sixty-six patients (89.9%) had comorbid medical conditions, including 24 with diabetes mellitus, 22 with malignant neoplasms, 28 with hypertension, 15 with cardiovascular disease, and 10 with kidney disease (all included as duplicate cases). Clinical assessment showed that 5 cases (6.8%) were non-ambulatory (Frankel classification A-C), and the average performance status was 2.6±1.3. Causative organisms were identified in 49 (66.2%) patients. The identified organisms included methicillin-sensitive Staphylococcus (24 cases), MRSA (9 cases), Streptococcus (7 cases), Enterococcus (2 cases), Klebsiella (2 cases), Escherichia coli (1 case), Enterobacter (1 case), Pseudomonas (1 case), and Citrobacter (1 case). Of the 49 cases in which the causative organisms were identified, 9 (18.4%) were infected by multidrug-resistant bacteria.
The average period from the onset of initial symptoms to diagnosis for all 74 cases was 23.9 days (range: 1-150 days). Forty-eight patients (64.9%) were diagnosed within 29 days of the onset of initial symptoms and were classified into the early diagnosis group, whereas 26 patients (35.1%) required more than 30 days from the onset of initial symptoms for diagnosis and were classified into the delayed diagnosis group (Table 1).
Table 1.
Univariate Analysis of Risk Factors for Delayed Diagnosis (Patient Factors).
| Early diagnosis n=48 (%) | Delayed diagnosis n=26 (%) | P-value | Odds ratio | 95% CI | ||
|---|---|---|---|---|---|---|
| Sex | Male | 25 (59.5) | 17 (40.5) | 0.27 | 0.58 | 0.21 – 1.54 |
| Female | 23 (71.9) | 9 (28.1) | 1.0 (ref.) | |||
| Age, (yr) | 68.1±8.9 | 74.1±13.7 | <0.05 | 1.04 | 1.00 – 1.09 | |
| Referral from another hospital | 29 (64.5) | 16 (35.5) | 0.56 | 1.05 | 0.39 – 2.79 | |
| Comorbidities | ||||||
| Diabetes mellitus | + | 15 (62.5) | 9 (37.5) | 0.77 | 1.16 | 0.42 – 3.21 |
| - | 33 (66.0) | 17 (34.0) | 1.0 (ref.) | |||
| Malignant neoplasm | + | 12 (54.5) | 10 (45.5) | 0.23 | 1.88 | 0.67 – 5.23 |
| - | 36 (69.2) | 16 (30.8) | 1.0 (ref.) | |||
| Hypertension | + | 17 (60.7) | 11 (39.3) | 0.56 | 1.34 | 0.50 – 3.55 |
| - | 31 (67.4) | 15 (32.6) | 1.0 (ref.) | |||
| Cardiovascular disease | + | 10 (66.7) | 5 (33.3) | 0.87 | 0.90 | 0.27 – 3.00 |
| - | 38 (64.4) | 21 (35.6) | 1.0 (ref.) | |||
| Kidney disease | + | 8 (80.0) | 2 (20.0) | 0.29 | 0.42 | 0.08 – 2.13 |
| - | 40 (62.5) | 24 (37.5) | 1.0 (ref.) | |||
| Clinical examination | ||||||
| Frankel classification | A-C | 2 (40.0) | 3 (60.0) | 0.81 | 1.25 | 0.19 – 8.00 |
| D, E | 46 (66.7) | 23 (33.3) | 1.0 (ref.) | |||
| Performance status | 2.5±1.4 | 2.7±1.1 | 0.62 | 1.10 | 0.75 – 1.61 | |
Data are presented as n (%) or means±standard deviations.
CI, confidence interval
Analysis of risk factors for delayed diagnosis
Risk factors for delayed diagnosis were evaluated using univariate analysis by comparing patient backgrounds between the early and delayed diagnosis groups (Table 1). The period from the onset of initial symptoms to diagnosis was 9.6±6.5 days for the early diagnosis group and 50.3±25.8 days for the delayed diagnosis group. Regarding patient factors, sex ratios were comparable between the 2 groups (p=0.58). The average age of the delayed diagnosis group was 74.1±13.7 years, which was significantly higher than that of the early diagnosis group (68.1±8.9, p=0.03). Patients referred from other hospitals accounted for 45/74 (60.8%): of the 45 patients referred from other hospitals, 29 were in the early diagnosis group and 16 were in the delayed diagnosis group, with no significant difference between the groups (p=0.56) (Table 1). The presence of comorbid medical conditions, including diabetes mellitus, malignant neoplasms, hypertension, cardiovascular diseases, and kidney diseases, was not significantly associated with delayed diagnosis. Regarding clinical examination, the proportion of Frankel classification A-C (non-ambulatory cases) and performance status were equivalent between the 2 groups (p=0.81, p=0.62, respectively) (Table 1).
Univariate analysis of infection factors showed that there was no significant association between delayed diagnosis and identification of the causative organism and multidrug-resistant infections (p=0.69, p=0.90, respectively) (Table 2). Blood test results demonstrated that WBC in the delayed diagnosis group (8.3±3.8×104 cells/μL) was significantly lower than that in the early diagnosis group (11.7±5.5×104 cells/μL, p<0.01). Similarly, the CRP level was significantly lower in the delayed diagnosis group (7.1±5.6 mg/dL) compared to the early diagnosis group (13.5±10.6 mg/dL, p<0.05). Albumin levels were equivalent between the 2 groups. Regarding the infected spine level based on the SINS classification, the delayed diagnosis group in the semi-rigid region (77.8%) was significantly higher than that in the early diagnosis group (22.2%, p=0.05). In summary, the univariate analysis of risk factors for delayed diagnosis revealed that significant risk factors included advanced age, low WBC count, low CRP level, and semi-rigid location of the SINS classification (Table 1, 2). The 4 variables (age, WBC, CRP, and semi-rigidity) identified as significant factors in the univariate analysis were chosen as predictors for multivariate analysis (Table 3). The area under the curve was 0.82 (95% confidence interval: 0.73-0.92), and R2 was 0.23. Multivariate analysis for delayed diagnosis showed that a semi-rigid location was a significant risk factor (p=0.02).
Table 2.
Univariate Analysis of Risk Factors for Delayed Diagnosis (Infection Factors).
| Early diagnosis n=48 (%) | Delayed diagnosis n=26 (%) | P-value | Odds ratio | 95% CI | ||
|---|---|---|---|---|---|---|
| Causative organism identification | + | 31 (64.3) | 18 (36.7) | 0.69 | 1.23 | 0.44 – 3.43 |
| - | 16 (66.7) | 8 (33.3) | 1.0 (ref.) | |||
| Multidrug-resistant bacterial infection | + | 6 (66.7) | 3 (33.3) | 0.90 | 0.91 | 0.21 – 4.00 |
| - | 42 (64.6) | 23 (35.4) | 1.0 (ref.) | |||
| Blood test | ||||||
| White blood cells (×104 cells/μL) | 11.7±5.5 | 8.3±3.8 | <0.01 | 0.81 | 0.70 – 0.95 | |
| C-reactive protein (mg/dL) | 13.5±10.6 | 7.1±5.6 | <0.05 | 0.91 | 0.84 – 0.98 | |
| Albumin (g/dL) | 2.9±0.7 | 2.8±0.7 | 0.82 | 0.93 | 0.48 – 1.80 | |
| Infected spine level classified based on SINS | ||||||
| Junctional (O-C2; C7-T2; T11-L1; L5-S1) | 16 (80.0) | 4 (20.0) | 0.10 | 0.36 | 0.11 – 1.24 | |
| Mobile (C3-6; L2-4) | 30 (66.7) | 15 (33.3) | 0.69 | 0.82 | 0.31 – 2.16 | |
| Semi-rigid (T3-10) | 2 (22.2) | 7 (77.8) | <0.05 | 8.47 | 1.61 – 44.56 |
Data are presented as n (%) or means±standard deviations.
CI, confidence interval; SINS, spinal instability neoplastic score
Table 3.
Multivariate Analysis of Risk Factors for Delayed Diagnosis.
| Partial regression coefficient | Standard error | P-value | Odds ratio | 95% CI | |
|---|---|---|---|---|---|
| Age (yr) | -0.03 | 0.03 | 0.22 | 0.97 | 0.92 – 1.02 |
| WBC | 0.10 | 0.09 | 0.24 | 1.11 | 0.93 – 1.31 |
| CRP | 0.08 | 0.04 | 0.05 | 1.09 | 1.00 – 1.19 |
| Semi-rigid | -2.66 | 1.13 | <0.05 | 0.07 | 0.0076 – 0.64 |
| Intercept | 1.37 | 0.067 – 0.44 |
CI, confidence interval; WBC, white blood cell; CRP, C-reactive protein
Imaging evaluation at the start of treatment
Imaging analysis was performed on 65 patients with spondylodiscitis who underwent CT and MRI on the day of admission (87.8%; 83.3% [40/48 cases] in the early diagnosis group and 96.2% [25/26 cases] in the delayed diagnosis group). In the delayed diagnosis group, most cases were expected to have been urgently admitted to the hospital after the imaging examination. The VBD rate (%) was significantly higher in the delayed diagnosis group than in the early diagnosis group (27.1±13.4 vs 13.3±10.5, respectively, p<0.01) (Fig. 2). The ACI was 0.15±0.08 in the delayed diagnosis group and 0.09±0.046 in the early diagnosis group; it was significantly higher in the delayed diagnosis group (p<0.01) (Fig. 2).
Figure 2.
Vertebral bone destruction rate (%) and abscess cavity index in the early diagnosis group and delayed diagnosis group. Values are in average±standard deviation.
Discussion
In this study, among 74 cases of pyogenic spondylitis, 26 (35.1%) required more than 30 days from the onset of initial symptoms to diagnosis, which we defined as a delayed diagnosis. We clarified the risk factors for delayed diagnosis of pyogenic spondylitis. Focusing on the infected spine level, the semi-rigid group, based on the SINS classification, was identified as a significant risk factor for delayed diagnosis in the multivariate analysis. Additionally, patients in the delayed diagnosis group were older than those in the early diagnosis group and had significantly lower WBC counts and CRP levels in blood tests. Furthermore, the delayed diagnosis group showed a greater degree of VBD and a larger abscess cavity at the start of inpatient treatment on imaging studies compared to the early diagnosis group.
Risk factors for delayed diagnosis of pyogenic spondylitis have been reported, including advanced age and comorbid medical problems11). We found that infection at the semi-rigid spine level, based on the SINS classification, was a significant risk factor for delayed diagnosis. Pluemer et al.19) reported that the SINS classification is used as part of the scoring to determine the treatment strategy for pyogenic spondylodiscitis, and that the severity of infection is assessed based on the level of spinal mobility (junctional, mobile, semi-rigid, and rigid). In diagnosing pyogenic spondylitis, symptoms such as severe pain, fever, and paralysis are critical to suspecting infection and proceeding with the examination as soon as possible. Usually, pyogenic spondylodiscitis creates an abscess cavity and causes VBD, resulting in spinal instability, severe pain, and sometimes paralysis. The semi-rigid spinal level (T3-T10), enclosed by the rib cage and less mobile than the other spinal levels, may be less likely to cause instability when a spinal infection occurs. Pyogenic spondylitis at the semi-rigid spinal level is less likely to cause symptoms in the early stages of infection, which could result in delayed diagnosis. Furthermore, a lower inflammatory response (CRP and WBC), indicating the relatively mild virulence of the infection, may have contributed to the delayed diagnosis by making it difficult to recognize initial symptoms such as pain and fever. For early diagnosis of pyogenic spondylitis, spinal infection should be suspected, and blood tests and MRI scans should be considered when compromised patients, such as those with cancer, diabetes, or those on steroids, present with back pain.
This study also clarified that large abscesses had formed and bone destruction had already progressed at the start of inpatient treatment in cases of delayed diagnosis of pyogenic spondylodiscitis. A large intervertebral abscess space has been reported to cause instability in the infected spinal column, and infection is difficult to control12,13). Furthermore, thoracic spine infections reportedly spread easily from the intervertebral disc space to the thoracic cavity under negative pressure, resulting in the expansion of the abscess into the thorax and intrathoracic infections such as pyothorax, mediastinal abscess, and pleurisy20-22). When the infection extends into the thoracic cavity, drainage of the spinal lesion alone is insufficient to control the infection, and abscess drainage and debridement of the thoracic cavity are necessary, which can be fatal. It is important to keep in mind that early diagnosis of thoracic spinal infections is challenging, as demonstrated in this study. Missing such infections could lead to severe respiratory complications.
This study has some limitations. First, we investigated the time from symptom onset to diagnosis, an evaluation based on the patient's subjective symptoms. Additionally, patients with an unclear onset of symptoms due to dementia or other conditions were excluded from the study, meaning that older patients with comorbidities, who are at risk of infection, may have been excluded. Another limitation was that the detection rate for pyogenic bacteria was not high (66.2%). However, pyogenic spondylitis diagnosis was comprehensively determined and confirmed by several spine surgeon specialists based on the IDSA guidelines2), so these diagnoses can be considered valid. As this was a multicenter observational study, the varying years of experience and specialties of the doctors at the initial consultation may have affected the time to diagnosis. Also, the 5 facilities in this study were all core hospitals in their respective regions, and many patients were referred from other hospitals. However, in actual clinical practice, initial consultations are often conducted by non-specialist spinal surgeons, and the results of this study may reflect the spinal infection diagnosis process in actual clinical practice.
In conclusion, significant risk factors for delayed diagnosis of pyogenic spondylitis include infection at the semi-rigid thoracic spinal level. Early diagnosis of spondylitis is crucial because delayed diagnosis can lead to progressive bone destruction and the formation of large abscesses. Increased awareness of thoracic spinal infections, as revealed in this study, which are easily overlooked due to mild symptoms and low inflammatory responses, is expected to lead to the early diagnosis and treatment of pyogenic spondylodiscitis.
Conflicts of Interest: The authors declare that there are no relevant conflicts of interest.
Author Contributions: T.S., K.Y., and M.T. designed the study; T.S., K.Y., K.Y., J.O., M.K., R.H., H.H., Y.H., H.N., T.E., D.U., H.T., and T.C. performed the experiments and analyzed the data; K.Y., J.O., M.K., H.T., H.S., and M.T. supervised the experiments; T.S., K.Y., and M.T. wrote the manuscript.
Ethical Approval: Approval was obtained by the institutional review board of Hokkaido University Hospital, Obihiro Kosei Hospital, Sapporo City General Hospital, Hakodate Central General Hospital and Hokkaido Medical Center as a multicenter study (Approval code: 018-0393).
Informed Consent: Informed consent was obtained from all patients.
References
- 1.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):653-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Berbari EF, Kanj SS, Kowalski TJ, et al. 2015 Infectious Diseases Society of America (IDSA) clinical practice guidelines for the diagnosis and treatment of native vertebral osteomyelitis in adults. Clin Infect Dis. 2015;61(6):e26-46. [DOI] [PubMed] [Google Scholar]
- 3.Akiyama T, Chikuda H, Yasunaga H, et al. Incidence and risk factors for mortality of vertebral osteomyelitis: a retrospective analysis using the Japanese diagnosis procedure combination database. BMJ. 2013;3(3):e002412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kokabu T, Takahata M, Ishiguro N, et al. Long-term prognosis of hematogenous vertebral osteomyelitis: mortality, quality of life, and pain. J Orthop Sci. 2017;22(5):822-7. [DOI] [PubMed] [Google Scholar]
- 5.Kehrer M, Pedersen C, Jensen TG, et al. Increasing incidence of pyogenic spondylodiscitis: a 14-year population-based study. J Infect. 2014;68(4):313-20. [DOI] [PubMed] [Google Scholar]
- 6.Bernard L, Dinh A, Ghout I, et al. Antibiotic treatment for 6 weeks versus 12 weeks in patients with pyogenic vertebral osteomyelitis: an open-label, non-inferiority, randomised, controlled trial. Lancet. 2015;385(9971):875-82. [DOI] [PubMed] [Google Scholar]
- 7.Schömig F, Li Z, Perka L, et al. Georg schmorl prize of the German spine society (DWG) 2021: Spinal Instability Spondylodiscitis Score (SISS)-a novel classification system for spinal instability in spontaneous spondylodiscitis. Eur Spine J. 2022;31(5):1099-106. [DOI] [PubMed] [Google Scholar]
- 8.Al-Nammari SS, Lucas JD, Lam KS. Hematogenous methicillin-resistant Staphylococcus aureus spondylodiscitis. Spine. 2007;32(22):2480-6. [DOI] [PubMed] [Google Scholar]
- 9.Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the infectious diseases society of America for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children. Clin Infect Dis. 2011;52(3):e18-55. [DOI] [PubMed] [Google Scholar]
- 10.Basak S, Singh P, Rajurkar M. Multidrug resistant and extensively drug resistant bacteria: a study. J Pathog. 2016;2016:4065603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Swanson AN, Pappou IP, Cammisa FP, et al. Chronic infections of the spine: surgical indications and treatments. Clin Orthop Relat Res. 2006;444:100-6. [DOI] [PubMed] [Google Scholar]
- 12.Yamada K, Takahata M, Nagahama K, et al. Posterolateral full-endoscopic debridement and irrigation is effective in treating thoraco-lumbar pyogenic spondylodiscitis, except in cases with large abscess cavities. Eur Spine J. 2023;32(3):859-66. [DOI] [PubMed] [Google Scholar]
- 13.Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5(6):649-55. [PubMed] [Google Scholar]
- 14.Taylor AE, Olver IN, Sivanthan T, et al. Observer error in grading performance status in cancer patients. Support Care Cancer. 1999;7(5):332-5. [DOI] [PubMed] [Google Scholar]
- 15.Frankel HL, Hancock DO, Hyslop G, et al. The value of postural reduction in the initial management of closed injuries of the spine with paraplegia and tetraplegia. I. Paraplegia. 1969;7(3):179-92. [DOI] [PubMed] [Google Scholar]
- 16.Bradford DS, McBride GG. Surgical management of thoracolumbar spine fractures with incomplete neurologic deficits. Clin Orthop Relat Res. 1987;(218):201-16. [PubMed] [Google Scholar]
- 17.Pola E, Taccari F, Autore G, et al. Multidisciplinary management of pyogenic spondylodiscitis: epidemiological and clinical features, prognostic factors and long-term outcomes in 207 patients. Eur Spine J. 2018;27(suppl 2):229-36. [DOI] [PubMed] [Google Scholar]
- 18.Fisher CG, DiPaola CP, Ryken TC, et al. A novel classification system for spinal instability in neoplastic disease: an evidence-based approach and expert consensus from the Spine Oncology Study Group. Spine. 2010;35(22):E1221-9. [DOI] [PubMed] [Google Scholar]
- 19.Pluemer J, Freyvert Y, Pratt N, et al. A novel scoring system concept for de novo spinal infection treatment, the Spinal Infection Treatment Evaluation Score (SITE Score): a proof-of-concept study. J Neurosurg Spine. 2023;38(3):396-404. [DOI] [PubMed] [Google Scholar]
- 20.Lescan M, Lepski G, Steger V, et al. Rapidly progressive paraplegia and pleural empyema: how does that correlate? Gen Thorac Cardiovasc Surg. 2013;61(11):640-2. [DOI] [PubMed] [Google Scholar]
- 21.Bass SN, Ailani RK, Shekar R, et al. Pyogenic vertebral osteomyelitis presenting as exudative pleural effusion: a series of five cases. Chest. 1998;114(2):642-7. [DOI] [PubMed] [Google Scholar]
- 22.Wong M, Lanka L, Hussain Z, et al. Epidural extension of infected chest wall haematoma and empyema causing spinal cord compression. Heart Lung Circ. 2014;23(1):e20-3. [DOI] [PubMed] [Google Scholar]