Secondary Logo

Journal Logo


A New Surgical Strategy for Infective Spondylodiscitis

Comparison Between the Combined Antero-Posterior and Posterior-Only Approaches

Choi, Sung Hoon MD; Koo, Ja Wook MD; Hur, Jeong Min MD; Kang, Chang-Nam MD, PhD

Author Information
doi: 10.1097/BRS.0000000000003544
  • Free

Spondylodiscitis is the most common infectious disease of the spine; it involves a primary infection of the vertebral bodies, particularly the endplates, initially, followed by progressive involvement of the adjacent intervertebral disc and other vertebral bodies.1,2 Its reported prevalence is 1:100,000, and its mortality rate is reported to be between 2% and 20%.3 Despite advances in the diagnosis and treatment of spinal diseases, the incidence of infective spondylodiscitis has increased along with the increase in older and/or immunosuppressed patients, as well as with the use of local injection therapy.4–6

Because there is no consensus on the use of instrumentation for active infective spondylodiscitis, various surgical treatment modalities have been discussed.7–16 Anterior debridement, using a combined antero-posterior approach, arthrodesis with autologous bone grafting, and posterior instrumentation are the most common surgical treatment modalities. When the infection is accompanied by anterior collapse of the vertebral body, effective removal of the infected tissue via an anterior approach with additional posterior instrumentation has been reported to yield good results, with increased fusion rates.8,9,14,15 However, in immunosuppressed patients or those in poor general condition, the combined antero-posterior approach has been associated with a high rate of postoperative complications, such as recurrence of infection, failure of internal fixation, and pseudarthrosis. Consequently, some reports have described satisfactory results with interbody fusion and posterior instrumentation using posterior-only approaches.16–18

To our knowledge, there have been virtually no studies that have compared the clinical outcomes of stepwise surgical treatment for infective spondylodiscitis via a combined antero-posterior approach and via a posterior-only approach with long-level posterior instrumentation and interbody fusion of the infected segment.19 This study aimed to compare the radiographic and clinical results of surgical treatment using a posterior-only approach or a combined antero-posterior approach in patients with infective spondylodiscitis.


Patient Enrollment

This study protocol was approved by the institutional review board of our hospital. Patients in whom medical treatment failed, despite adequate antibiotics after diagnosis of infective spondylodiscitis in thoracic or lumbar vertebrae, from January 2007 to December 2016, were enrolled. They were divided into an antero-posterior group (AP group) that received stepwise surgical treatment using a combined antero-posterior approach, and a posterior-only group (P group) that received treatment via a posterior-only approach. Infective spondylodiscitis was diagnosed based on clinical symptoms, laboratory data, magnetic resonance imaging findings and were confirmed by identification of the pathogen through blood and tissue cultures.20 The clinical manifestations of these patients varied from progressively aggravated back pain, fever, newly developed radiating pain to the lower extremity, or weakness, to sepsis.

Inclusion criteria for this study were a requirement for surgical treatment because of the failure of medical treatment, despite appropriate antibiotic treatment after initial diagnosis of pyogenic spondylodiscitis. During the medical treatment period, if pathologic fracture, pain aggravation, or failure of C-reactive protein (CRP) normalization occurred, we considered the use of surgical debridement. Patients were excluded if they were immunocompromised, if they had any source of infection anywhere else in the body, other than the spine, had postoperative infective spondylodiscitis, or who underwent debridement alone, had a follow-up period less than 2 years, or had neurological deficits at the time of initial diagnosis, as these patients required more aggressive debridement and reconstruction surgery. Depending on the type of pathogen, the extent of bone destruction and progression of the disease varied; therefore, we excluded patients with fungal and tuberculous spondylodiscitis, to ensure consistency in the study.21

Blood and urine were cultured in all patients prior to administering antimicrobials, and appropriate antibiotics were administered after the strain type had been identified conclusively. When no microbe was isolated, broad-spectrum antibiotics were used continuously. Intravenous antibiotics were administered for 6 to 8 weeks until CRP levels normalized, and thereafter, orally administered antibiotics were used until the erythrocyte sedimentation rate (ESR) normalized.

Demographic data reviewed included age, level of the lesion, pathogenic organism, underlying disease, bone mineral density (BMD), and body mass index (BMI). For radiological data, preoperative height loss, regional wedge angle using Cobb's angle, and kyphotic angle were recorded. Lordosis was defined as a negative value and kyphosis as a positive value. Visual analogue scale (VAS) scores for evaluating back pain, initial white blood cell count, ESR, and CRP levels were also analyzed. The following surgical data were analyzed: fusion level, fusion rate, fusion time, estimated blood loss (EBL), operation time, and complications after surgery in terms of implant failure, screw pullout, pseudarthrosis, revision surgery, and kyphotic angles immediately postoperatively and at the final follow-up. The Kirkaldy–Willis standard was used as functional scale. These data are shown in Table 1.

The Criteria of Kirkaldy–Willis for Functional Outcome

Surgical Technique

All surgical procedures were performed by a single spine surgeon. A combined antero-posterior approach was performed in patients who had preoperative height loss exceeding 50%, local kyphosis exceeding 25° (in those with thoracolumbar lesions), or loss of lordosis (in those with lumbar lesions). In the AP group, the patient was placed in a lateral position and corpectomy was performed via transthoracic, transpleural retroperitoneal, or retroperitoneal approaches. Anterior interbody fusion was performed using an autogenous iliac tricortical strut bone or autogenous rib bone graft. Afterward, the patient was placed in a prone position and long-level posterior instrumentation performed, without fixation of infected vertebrae, through the posterior midline approach. Pedicle screws were placed from two segments above to two segments below the infected vertebrae (Figure 1A–H).

Figure 1
Figure 1:
(A) Preoperative antero-posterior (AP) radiograph, (B) preoperative lateral radiograph, (C) preoperative T1-weighted magnetic resonance (MR) image, (D) preoperative T2 fat-suppression-weighted MR image, (E) postoperative AP radiograph, (F) postoperative lateral radiograph, (G) last follow-up coronal computed tomography (CT) image, (H) last follow-up sagittal CT image.

For the P group, the patient was placed in the prone position and a posterior approach was used to perform en bloc partial laminectomy of the infected vertebrae (Figure 2A, B). Then, infected tissues, including the intervertebral disc, endplate, and parts of the vertebral body were debrided, and autogenous laminar strut bones were grafted to achieve interbody fusion. Long-level instrumentation was performed from two segments above to two segments below, but excluding the infected vertebrae (Figure 3A–H)

Figure 2
Figure 2:
Intraoperative clinical photograph. (A) En bloc partial laminectomy of infected vertebrae. (B) Laminar strut bone (lateral).
Figure 3
Figure 3:
(A) Preoperative antero-posterior (AP) radiograph. (B) Preoperative lateral radiograph. (C) Preoperative T1-weighted magnetic resonance (MR) image. (D) Preoperative T2 fat-suppression-weighted MR image. (E) Postoperative AP radiograph. (F) Postoperative lateral radiograph. (G) Last follow-up coronal computed tomography (CT) image. (H) Last follow-up sagittal CT image.

All patients were instructed to wear a thoracic lumbar sacral orthosis for 3 months. During outpatient follow-up, antero-posterior and lateral radiographs were taken at 3, 6, 12, 18, and 24 months postoperatively, and once every year thereafter. Additionally, spine computed tomography scans were performed 6, 12, and 24 months postoperatively, and arthrodesis was evaluated based on the Bridwell interbody fusion grading system.

Statistical Analysis

Differences between the two groups were evaluated by independent t tests or the Mann–Whitney U test, and continuous variables were compared by paired t tests and Wilcoxon signed-rank tests. To identify factors that affected the decision about the surgical approach taken, univariate and multivariate stepwise logistic regression analyses were performed. To determine the related cut-off value, receiver operator characteristic (ROC) curve analysis was used. Statistical significance was set at P < .05. Statistical analysis was performed using SPSS software (version 22.0; SPSS Inc., Chicago, IL) and MedCalc Version 18.10.2 (Medcalc Software bvba, Ostend, Belgium).


Seventy-nine patients (43 men and 36 women) with infective spondylodiscitis, in whom conservative treatment (mean: 3.1 ± 1.2 wks) failed, were analyzed. Eight patients were lost to follow-up within 2 years postoperatively. The follow-up rate at the last visit was 90.8% (79/87). The mean age at the time of operation was 65.2 years (21–88 yrs), and the mean follow-up period was 70 months (24–151 mos). There were 59 patients in the AP group and 20 in the P group, and there were 12 thoracic, 6 thoracolumbar, and 61 lumbar vertebrae cases.

Table 2 compares preoperative variables in the AP and P group. Initial height loss, regional wedge angle, and preoperative kyphotic angle were significantly higher in the AP group. There were no differences between groups in terms of comorbidity, BMD, and BMI. Table 3 compares the postoperative variables in the two groups. Seventy-three patients (92.4%) had posterior instrumentation removed at 9 to 12 months postoperatively. Six patients (4 [6.8%] in the AP group, 2 [10%] in the P group, P = 0.64) required adjustment of instrumentation at 12 months postoperatively. There were no significant differences between the two groups for fusion rate or fusion time. However, EBL, operation time, functional outcome, and last VAS score were significantly lower in the P than in the AP group, and the corrected kyphotic angle was significantly higher in the AP (12.8°) than in the P group (5.3°).

Preoperative Demographic Data of 76 Patients Underwent Surgical Treatment Through Antero-Posterior or Posterior-only Approach
Comparison of the Postoperative Outcomes in Patients With Infective Spondylodiscitis Through Antero-Posterior or Posterior-only Approach

In terms of functional results, the AP group exhibited 69.5% excellent/good results: 21 were excellent, 20 good, 15 fair, and 3 poor. The P group exhibited 95% excellent/good results: 12 were excellent, 7 were good, and 1 was fair. There was no wound infection or transient neurological deficits in the postoperative period, and there was no recurrence of infection in any of the patients after surgery. With regard to complications, pseudarthrosis was observed in three patients in the AP group and none in the P group; however, this difference was not statistically significant. None of the patients required revision surgery because there were no relevant symptoms.

Kyphotic angle and VAS score improved significantly in both groups after the operation (Table 4). According to univariate and stepwise multivariate logistic regression, the regional wedge angle (P < 0.001, odds ratio 0.84, 95% confidence interval 0.75–0.93) was the only factor that affected the decision for using the combined antero-posterior or posterior-only approach (Table 5). With an optimal cut-off value for the regional wedge angle of 8.2° in the ROC curve, sensitivity was 60% and specificity 89% (Figure 4).

Changes in Kyphotic Angle and Visual Analogue Scale Score of Each group
Factors Affecting the Decision Regarding the Surgical Approach, Based on Stepwise Multivariate Logistic Regression
Figure 4
Figure 4:
Receiver operating characteristic curve for the regional wedge angle of the collapsed vertebra. A cutoff value of 8.2° demonstrated a sensitivity of 60% and specificity of 89.8%.


The anterior spinal column is the most commonly destroyed structure in infective spondylodiscitis.22 Traditionally, anterior corpectomy along with autogenous iliac tricortical strut grafts have been performed via an anterior approach. However, this has several disadvantages, including technical difficulty, need for long-term immobilization, and prolonged use of an orthosis, which can lead to general deterioration.11,15,23,24 Additionally, as issues with kyphosis progression and with the graft or vertebral body arose in cases treated via an anterior approach, several authors have reported the usefulness of additional posterior instrumentation through a stepwise operation. Additional posterior instrumentation had many advantages, such as early ambulation, shorter orthosis use, and reduced graft destruction and kyphosis deformity.11,15,23,24 However, as a 2-stage operation, it increases surgery and anesthesia times and blood loss, as well as morbidity and hospitalization. Therefore, the appropriate approach should be chosen carefully in patients with infective spondylodiscitis, particularly those in poor general condition.

In this study, EBL, operation time, functional impairment, and last VAS scores were all lower in the P group, indicating less morbidity is associated with the posterior than with the antero-posterior approach. No wound infection, neurological deficits, recurrence of infection, or failure of internal fixation occurred in patients in either group during the follow-up period. Pseudarthrosis was the only complication and occurred only in the AP group. However, none of these patients displayed symptoms and thus none required revision surgery. The overall mean preoperative kyphotic angle was (+) 9.5° and (–) 1.3° at final follow-up. The mean corrected lordosis angle was 10.8°, which was statistically significant (P < 0.001). The corrected kyphotic angle was significantly higher in the AP (12.8°) than in the P group (5.3°; P = 0.003). In terms of clinical outcomes, 75% of our patients overall had excellent/good results; the mean preoperative VAS score was 7.9 points, which decreased to 2.1 points at final follow-up, indicating a significant lower back pain reduction (P < 0.001).

Univariate and stepwise multivariate logistic regression analysis showed that the regional wedge angle (P < 0.001, odds ratio 0.835, 95% confidence interval 0.751–0.928) was the only factor affecting the decision regarding an anterior or a posterior approach. The mean preoperative wedge angle was 16.0° for the AP and 8.3° for the P group, and the corrected postoperative wedge angle was 12.8° for the AP and 5.3° for the P group. The optimal cut-off value of the regional wedge angle that predicted choice of a posterior-only approach was 8.2°, with a sensitivity of 60% and specificity of 89%.

Few studies have investigated surgical treatment of pyogenic spondylodiscitis using a posterior-only approach. Lee and Suh16 reported the clinical and radiologic results of a posterior-only approach in patients with pyogenic spondylodiscitis; the patients underwent debridement from the posterior laminectomy as well as removal of infected endplates, followed by lumbar interbody and posterolateral fusion of the affected segments. They reported that a posterior-only approach improved neurological status without recurrence of infection or increase of the kyphotic angle. Additionally, Lin et al17 reported good clinical results without recurrence of infection and worsening of the kyphotic angle in 48 pyogenic spondylodiscitis patients treated with a posterior-only approach. Unlike in Lee and Suh's study,16 infected intervertebral disc and endplate tissues were not removed. In terms of functional outcomes at final follow-up, based on the Kirkaldy–Willis criteria, Lee and Suh16 reported 83% excellent/good results and that of Lin et al17 reported 67%, while our study yielded 95% excellent/good result in the P group. The major differences between these previous studies and our own were as follows. First, we removed infected tissues and performed interbody fusion using the posterior-only approach, while we also performed long-level posterior instrumentation to prevent iatrogenic instability. Second, we directly compared the clinical and radiographic outcomes of a posterior-only approach and a combined antero-posterior approach. Lastly, unlike the study of Lee and Suh,16 we did not include postoperative spondylodiscitis, and we performed lumbar interbody fusion, unlike in the study of Lin et al.17

In infective spondylodiscitis, decompression and debridement can be performed using an anterior approach to remove infected tissues and correct the kyphotic angle effectively; however, the higher morbidity related to this operation remains a concern. In contrast, the posterior-only approach has the drawbacks of insufficient removal of infected tissue and correction of the kyphotic angle. In this study, we found that infected intervertebral discs, vertebral endplates, and vertebral body tissues could be adequately debrided through a posterior-only approach, thus assuring the excellence of the radiographic and clinical results. Moreover, the posterior-only approach could generate solid stability and correct the kyphotic angle using a strut laminar bone along with posterior long instrumentation. Furthermore, the posterior-only approach made it possible to reduce EBL and operation time, diminishing postoperative complications, as well as complications associated with prolonged immobilization. As compared with the anterior approach, which makes it difficult to perform corpectomy, it permitted removal of sufficient infected tissue using an approach familiar to most spine surgeons, and it is therefore judged to be more useful than the AP approach. Therefore, we recommend a posterior-only approach when the regional wedge angle of the collapsed vertebra is less than 8.2°; if this angle is greater, the antero-posterior approach is recommended.

In this study, to preserve motion segments of instrumented vertebra, all patients underwent a second operation to remove instrumentation from 9 months to 12 months postoperatively, depending on the strength of the fused segment. If the fused segment was not sufficiently solid, pedicle screws were reused and adjusted to the healed vertebral body.25 If the fusion was sufficiently solid, all pedicle screws were removed (Figures 5A, B and 6A, B). All patients acquired normalized CRP, reduced pain, and improved ambulation within postoperative 8 weeks. Therefore, our study suggests that, in patients who have prolonged, severe, sustained pain, elevated CRP, or pathologic fracture despite medical treatment, surgical debridement, and strut bone grafting, performed via the appropriate approach, would be helpful to eradicate the infection focus.

Figure 5
Figure 5:
Patient who underwent removal of the implant in the AP-P group. This case did not have enough solidification in the fused segment; the pedicle screws were reused and adjusted to the healed vertebral body. (A) Postoperative antero-posterior radiograph. (B) Postoperative lateral radiograph.
Figure 6
Figure 6:
Patient who underwent removal of the implant in the P group. This case demonstrated sufficient solidification in the fused segment, and all pedicle screws were removed. (A) Postoperative antero-posterior radiograph. (B) Postoperative lateral radiograph.

This study had the following limitations. First, it was a retrospective study of operations conducted by a single spine surgeon, and therefore there may have been selection bias in determining the surgical approach. Second, the study had a relatively small sample size, and therefore, more extensive follow-up research on the factors that determine the combined antero-posterior versus posterior-only approaches in infective spondylodiscitis is needed. However, this study is valuable in that no previous study had directly compared the combined antero-posterior stepwise approach and the posterior-only approach for surgical treatment of infective spondylodiscitis in patients with at least 2 years of follow-up.

In conclusion, in infective spondylodiscitis, interbody fusion with laminar strut bone grafts and long-level internal fixation, avoiding the infected vertebrae, through a posterior-only approach was shown to be as effective as a combined antero-posterior approach in terms of infection control, kyphosis correction, pain control, and functional outcomes. A posterior-only approach is recommended when the regional wedge angle of the collapsed vertebra is less than 8.2°.

Key Points

  • The approach method, invasiveness, fixation levels, and types of bone graft used for surgical treatment of infective spondylodiscitis are controversial.
  • Initial height loss, regional wedge angle, and preoperative kyphotic angle were significantly higher in the group treated with a combined antero-posterior approach than in those treated with the posterior-only approach, while EBL, operation time, functional outcome, and last VAS score were significantly lower in the posterior-only group.
  • In infective spondylodiscitis, interbody fusion with laminar strut bone graft and long-level internal fixation, other than the infected vertebrae, via a posterior-only approach is as effective in terms of infection control, kyphosis correction, pain control, and functional outcomes as that achieved via a combined antero-posterior approach.
  • Regional wedge angle was the only factor that affected the decision between a combined antero-posterior and a posterior-only approach, and a posterior-only approach is recommended when the regional wedge angle of the collapsed vertebra is less than 8.2°.


The statistical analysis used in this study was assisted by the Medical Research Collaborating Center, Hanyang University, Seoul, Republic of Korea.


1. Pandita N, Paul S, Yadav G, et al. Evaluation of challenges in diagnosis of spontaneous subacute pyogenic spondylodiscitis in immunocompetent patients: experiences from a tertiary care center. Asian Spine J 2019; 13:621–629.
2. Sobottke R, Seifert H, Fatkenheuer G, et al. Current diagnosis and treatment of spondylodiscitis. Dtsch Arztebl Int 2008; 105:181–187.
3. Aagaard T, Roed C, Dahl B, et al. Long-term prognosis and causes of death after spondylodiscitis: a Danish nationwide cohort study. Infect Dis (Lond) 2016; 48:201–208.
4. Kehrer M, Pedersen C, Jensen TG, et al. Increasing incidence of pyogenic spondylodiscitis: a 14-year population-based study. J Infect 2014; 68:313–320.
5. Rezai AR, Woo HH, Errico TJ, et al. Contemporary management of spinal osteomyelitis. Neurosurgery 1999; 44:1018–1025. discussion 1025–1026.
6. An HS, Seldomridge JA. Spinal infections: diagnostic tests and imaging studies. Clin Orthop Relat Res 2006; 444:27–33.
7. Lener S, Hartmann S, Barbagallo GMV, et al. Management of spinal infection: a review of the literature. Acta Neurochir 2018; 160:487–496.
8. Fukuta S, Miyamoto K, Masuda T, et al. Two-stage (posterior and anterior) surgical treatment using posterior spinal instrumentation for pyogenic and tuberculotic spondylitis. Spine (Phila Pa 1976) 2003; 28:E302–E308.
9. Dimar JR, Carreon LY, Glassman SD, et al. Treatment of pyogenic vertebral osteomyelitis with anterior debridement and fusion followed by delayed posterior spinal fusion. Spine (Phila Pa 1976) 2004; 29:326–332. discussion 32.
10. Deininger MH, Unfried MI, Vougioukas VI, et al. Minimally invasive dorsal percutaneous spondylodesis for the treatment of adult pyogenic spondylodiscitis. Acta Neurochir (Wien) 2009; 151:1451–1457.
11. Sundararaj GD, Babu N, Amritanand R, et al. Treatment of haematogenous pyogenic vertebral osteomyelitis by single-stage anterior debridement, grafting of the defect and posterior instrumentation. J Bone Joint Surg Br 2007; 89:1201–1205.
12. Ha KY, Shin JH, Kim KW, et al. The fate of anterior autogenous bone graft after anterior radical surgery with or without posterior instrumentation in the treatment of pyogenic lumbar spondylodiscitis. Spine (Phila Pa 1976) 2007; 32:1856–1864.
13. Ito M, Abumi K, Kotani Y, et al. Clinical outcome of posterolateral endoscopic surgery for pyogenic spondylodiscitis: results of 15 patients with serious comorbid conditions. Spine (Phila Pa 1976) 2007; 32:200–206.
14. Korovessis P, Repantis T, Iliopoulos P, et al. Beneficial influence of titanium mesh cage on infection healing and spinal reconstruction in hematogenous septic spondylitis: a retrospective analysis of surgical outcome of twenty-five consecutive cases and review of literature. Spine (Phila Pa 1976) 2008; 33:E759–E767.
15. Pee YH, Park JD, Choi YG, et al. Anterior debridement and fusion followed by posterior pedicle screw fixation in pyogenic spondylodiscitis: autologous iliac bone strut versus cage. J Neurosurg Spine 2008; 8:405–412.
16. Lee JS, Suh KT. Posterior lumbar interbody fusion with an autogenous iliac crest bone graft in the treatment of pyogenic spondylodiscitis. J Bone Joint Surg Br 2006; 88:765–770.
17. Lin CP, Ma HL, Wang ST, et al. Surgical results of long posterior fixation with short fusion in the treatment of pyogenic spondylodiscitis of the thoracic and lumbar spine: a retrospective study. Spine (Phila Pa 1976) 2012; 37:E1572–E1579.
18. Jain A, Jain R, Kiyawat V. Evaluation of outcome of posterior decompression and instrumented fusion in lumbar and lumbosacral tuberculosis. Clin Orthop Surg 2016; 8:268–273.
19. von der Hoeh NH, Voelker A, Hofmann A, et al. Pyogenic spondylodiscitis of the thoracic spine: outcome of 1-stage posterior versus 2-stage posterior and anterior spinal reconstruction in adults. World Neurosurg 2018; 120:e297–e303.
20. Lee BH, Lee HM, Kim TH, et al. Transpedicular curettage and drainage of infective lumbar spondylodiscitis: technique and clinical results. Clin Orthop Surg 2012; 4:200–208.
21. Kim C-J, Song K-H, Jeon J-H, et al. A comparative study of pyogenic and tuberculous spondylodiscitis. Spine (Phila Pa 1976) 2010; 35:E1096–E1100.
22. Rutges JP, Kempen DH, van Dijk M, et al. Outcome of conservative and surgical treatment of pyogenic spondylodiscitis: a systematic literature review. Eur Spine J 2016; 25:983–999.
23. Klockner C, Valencia R. Sagittal alignment after anterior debridement and fusion with or without additional posterior instrumentation in the treatment of pyogenic and tuberculous spondylodiscitis. Spine (Phila Pa 1976) 2003; 28:1036–1042.
24. Linhardt O, Matussek J, Refior HJ, et al. Long-term results of ventro-dorsal versus ventral instrumentation fusion in the treatment of spondylitis. Int Orthop 2007; 31:113–119.
25. Lee JK, Jo YH, Kang CN. Cost-effectiveness analysis of existing pedicle screws reusing technique in extension revision operation for adjacent segmental stenosis after lumbar posterolateral fusion. Spine (Phila Pa 1976) 2016; 41:E785–E790.

corpectomy; infection; initial height loss; kyphosis; posterior instrumentation; spondylodiscitis; surgical approach; wedge angle

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.