Eur Spine J DOI 10.1007/s00586-014-3585-3
ORIGINAL ARTICLE
Transforaminal lumbar interbody debridement and fusion for the treatment of infective spondylodiscitis in the lumbar spine Meng-Ling Lu • Chi-Chien Niu • Tsung-Ting Tsai Tsai-Sheng Fu • Lih-Huei Chen • Wen-Jer Chen
•
Received: 9 December 2013 / Revised: 10 September 2014 / Accepted: 12 September 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose To determine the safety and efficacy of using a single posterior approach with transforaminal lumbar interbody debridement and fusion (TLIDF) plus pedicle screws fixation in treating infective spondylodiscitis in the lumbar spine. Methods Between January 2009 and June 2011, 28 patients with infective spondylodiscitis who underwent TLIDF, using autogenous graft and posterior pedicle screws instrumentation, met the indications for surgery, and completed more than 18 months of follow-up, were included. Clinical outcomes were assessed using a visual analog scale (VAS), the Oswestry Disability Index (ODI), and Kirkaldy-Willis functional outcome criteria. Infection status was evaluated using C-reactive protein levels, erythrocyte sedimentation rate, and clinical symptoms. The interbody fusion status and sagittal alignment of the infected segments were assessed using radiographic studies. Results Intra-operative culture rate was 82.1 %. The most common pathogen was methicillin-resistant Staphylococcus aureus. One post-operative deep wound infection with septic implant loosening and one instance of early aseptic implant loosening were noted. Implants in both patients were subsequently removed. Two infections recurred
M.-L. Lu C.-C. Niu T.-T. Tsai T.-S. Fu L.-H. Chen W.-J. Chen (&) Department of Orthopedic Surgery, Chang Gung Memorial Hospital, Chang Gung University, No 5, Fu-Hsing Street 333, Kweishan, Taoyuan, Taiwan e-mail: [email protected] C.-C. Niu T.-T. Tsai T.-S. Fu L.-H. Chen W.-J. Chen College of Medicine, Chang Gung University, Taoyuan, Taiwan
within 3 months post-operatively, with both subsiding within 3 months after extended antibiotic treatment. VAS and ODI values were significantly improved. The interbody fusion rate was 82.1 % and the lordotic angle of the infected segments and the lumbar spine were corrected by 9.7° and 14.3°, respectively at the last follow-up. Conclusions A single posterior approach with TLIDF and pedicle screws instrumentation for lumbar infective spondylodiscitis provided a satisfactory clinical outcome, adequate infection control and good fusion rate, and this strategy avoids the risks of anterior or staging surgeries. Keywords Infective spondylodiscitis Lumbar spine Transforaminal lumbar interbody fusion Transforaminal lumbar interbody debridement and fusion
Introduction There are several types of spinal infections, including spondylodiscitis, discitis, epidural abscess, septic facet arthropathy and ‘‘central body lesions’’ in tuberculosis. Based on the type of pathogens, the infection can be classified as pyogenic or nonpyogenic. Regardless of whether one subscribes to the arteriole [1] or venous theory [2], the vascularity of vertebrae is extremely rich near the end-plate, which allows pathogens entrapped in the endplate to cause destruction of the disc and adjacent vertebrae. Clinically, spondylodiscitis is the most common type spinal infection, and it occurs most commonly in the lumbar region [3]. Infective spondylodiscitis can result in disc destruction, pathologic fracture, and abscess formation. Pain is the most common symptom and neurologic deficit may affect up to 17 % of the patients [4]. Surgical treatment is indicated for neurologic impairment, spinal
123
Eur Spine J
instability, deformity, unknown pathogen, poorly controlled infection and intractable pain. Treatment of spinal infections remains challenging for spinal surgeons. The goals of surgical treatment are debridement of infective foci, collection of pathogens, decompression of neural element, and re-stabilization of the deformed spine. Regardless of whether the infection is pyogenic or nonpyogenic, the infective foci of infective spondylodiscitis are always located around the involved disc, and hence, debridement of the infected disc with reconstruction of the interbody bony defect must be considered first [5]. A posterior approach with laminectomy is contraindicated, as it will result in global instability. An anterior approach for infective spondylodiscitis may achieve more radical debridement and fusion [6, 7]. The addition of posterior instrumentation to anterior procedures can stabilize the spine and prevent kyphotic deformity [8]. The complication rate of anterior spinal surgery is relatively higher than that of posterior spinal surgery [9], especially when performed in patients with chronic lesions with adhesion, in thoracolumbar and lumbosacral junctions, or in patients with a history of previous anterior surgery. Combining anterior and posterior approaches may increase the risks, especially in immunocompromised patients. The purposes of this study were to determine the safety and efficacy of a single posterior approach with transforaminal lumbar interbody debridement and fusion (TLIDF) plus pedicle screws fixation for infective spondylodiscitis in the lumbar spine.
Materials and methods Thirty patients with infective spondylodiscitis of the lumbar spine underwent TLIDF and pedicle screws instrumentation between January 2009 and June 2011 at the author’s institution. Patients who were followed-up for more than 18 months were enrolled. Of these, 28 were included in this retrospective study. Medical records, imaging studies, laboratory data, neurologic function data and functional outcomes were reviewed and analyzed. The indications for TLIDF included: progressive neurologic deficit, spinal instability, deformity, intractable pain, unknown pathogen, and a failed response to antibiotic treatment. Patients displaying destruction of more than 50 % [based on pre-operative radiography and magnetic resonance imaging (MRI)] of vertebral bodies were excluded. We modified the surgical procedure of transforaminal lumbar interbody fusion (TLIF), and performed debridement, fusion and instrumentation using a single posterior approach, naming the modified procedure TLIDF. Surgical time, the affected level, the instrumented level,
123
intra-operative blood loss, and complications were recorded. All patients were encouraged to ambulate while wearing a Taylor brace for 12 weeks. Patients were followed-up at 1, 3, 6 months and then annually. Imaging studies including pre-operative and post-operative plain radiography of the lumbar spine were performed. All patients underwent MRI of the lumbosacral spine before operation. Using the superior and inferior endplates of the diseased vertebra as reference points, the lordotic angle of the lesion site was recorded pre-operatively, post-operatively, and on a yearly basis thereafter. The lumbar lordotic angle measured from lower end-plate of T12 to upper end-plate of S1 was also recorded simultaneously. The interbody fusion status was analyzed at the final follow-up. Radiographic outcome were assessed by a radiologist who was blinded to the operation. Solid fusion was defined as continuous trabeculae bridging across the interface of the treated segments, whereas the presence of any radiolucent interruption at the treated segments was defined as pseudarthrosis. The fusion was classified as inadequate when the radiographic image was vague which could not be defined as solid or pseudarthrosis. According to the culture results and recommendations of the infectious disease specialist, antibiotic treatment was initiated after surgery. Laboratory data including a white blood count with a differential count, erythrocyte sedimentation rates (ESRs), and C-reactive protein (CRP) levels were checked serially and recorded. According to the clinical symptoms and the ESR and CRP data, pyogenic spondylodiscitis and fungus infection were treated with 4–6 weeks of parenteral antibiotics followed by oral antibiotic for totally 3 months. Tuberculosis was treated with 12 months of oral anti-tuberculosis drug administration. Clinical outcomes were evaluated using the patientassessed quantitative measurement of the visual analog scale (VAS) for back pain, the Oswestry Disability Index (ODI) and Kirkaldy-Willis functional outcome criteria. Neurologic status was evaluated using the American Spinal Injury Association (ASIA) scale. Operative procedure While under general anesthesia, each patient was placed on a four-poster spinal frame in the prone position. Posterior iliac bone was harvested first. One posterior midline skin incision, followed by a bilateral para-median approach (Wiltse approach), was performed. The TLIDF side was determined pre-operatively, according to the severity of bony destruction, epidural abscess, and radiculopathy. Unilateral facetectomy and limited laminotomy were performed to expose the infected disc at the foraminal level. The pus, infected disc, end-plate and sequestrum were removed using the TLIF procedure and the psoas abscess
Eur Spine J
Fig. 1 A 64-year-old woman who experienced tuberculosis infection and underwent anterior debridement twice in another hospital. X-ray revealed L1–2 disc space narrowing with end-plate destruction and kyphosis (a); T1-weighted magnetic resonance imaging with contrast revealed vertebral body destruction and psoas abscess (b, c). The patient underwent TLIDF with two above and two below posterior instrumentation (T11-L4) (d). 1-year post-operative radiograph
revealed solid interbody fusion at L1–2 (e). Spine computed tomography revealed continuous trabeculae bridging across the interface of the treated segments on the sagittal view (f). Adequate debridement and decompression were noted on the axial view (g). Another axial view revealed no screw loosening and good screw position (h)
was drained posterolaterally under blunt dissection. In patients with neurologic compromise, the dural compression from the anterior infected lesion could be well-
decompressed, synchronously. The samples were cultured for aerobic and anaerobic microbes, tuberculous bacilli and fungi and the extracted tissue was sent for
123
Eur Spine J
histopathological examination. After adequate debridement and curettage, the autogenous iliac bone grafts (corticocancellous and cancellous bone chips) were inserted for interbody fusion. Trans-pedicle screws were inserted into the intact bodies one level superior and one level inferior to the diseased disc. However, in the presence of greater bony destruction or osteoporosis, the fixation was extended over a longer distance.
Results Of these 30 patients, two were lost to follow-up and excluded from this study. The remaining 28 patients (13 men and 15 women) were followed-up for more than 18 months. The mean age at surgery was 60.4 years (range 37–86 years). The mean follow–up period was 20 months (range 18–36 months). No patient died due to spinal infection; however, one patient died from cancer 18 months after the operation. Three patients underwent percutaneous endoscopic debridement and drainage, and two underwent anterior debridement prior to this surgical intervention. 15 patients received empiric antibiotics before the operation. The involved levels were T12-L1 in two patients, L1–2 in two patients, L2–3 in two patients, L3–4 in two patients, L4–5 in nine patients, L5–S1 in 10 patients and L1-2-3 in one patient. The mean operative time was 179.6 ± 30.3 min (range 90–240 min), and the mean blood loss was 712.5 ± 299.9 mL (range 200–1, 400 mL). The instrumentation levels were one above one below in 12 patients, two above two below in 14 patients (including two patients who required iliac screw fixation) (Fig. 1), and three above three below in two patients. Infection status Twenty-six patients exhibited hematogenous infection. Two patients exhibited post-operative infections (1 discectomy and 1 instrumented vertebral fracture combined infection). The intra-operative culture rate was 82.1 %. The most common pyogenic pathogen was methicillin-resistant Staphylococcus aureus (7 patients) and the most common non-pyogenic pathogen was tuberculosis (3 patients) (Table 1). All histopathologic examinations revealed acute and chronic inflammation, and two examinations also revealed granulomatous inflammation. For most patients, the CRP levels and ESRs normalized within 3 months after operation. One patient (L5–S1 infection, Klebsiella pneumoniae) experienced a post-operative deep wound infection that extended to the L4–5 level. He underwent wound debridement; transforaminal discectomy at L4–5; and removal of implants. The pathogen was found to be drug-
123
Table 1 Cultures of spinal infection Pathogen
%
No. of patients
MRSA
25
7
Tuberculosis
10.7
3
Klebsiella pneumoniae
10.7
3
MSSA
7.1
2
Pseudomonas aeruginosa
3.6
1
Salmonella
3.6
1
Enterococcus faecalis
3.6
1
Gram-negative bacilli, G group
3.6
1
Aerococcus viridans
3.6
1
CoNS
3.6
1
Fungus (Cladosporium) Multiplea
3.6 3.6
1 1
No organism
17.9
5
Culture rate
82.1
28
MRSA methicillin–resistant Staphylococcus aureus, MSSA methicillin sensitive Staphylococcus aureus, CoNS coagulase negative Staphylococcus a
Pseudomonas aeruginosa, Enterococcus faecalis and Candida
resistant and the infection was eventually controlled. Two patients experienced recurrent back pain with elevated CRP levels during the 3 months after discharge. Both patients received extended intravenous antibiotics for 4–6 weeks, and the infections subsided. Neurologic status Two patients displayed cauda-equina syndrome, with both recovering after the operation. The pre-operative neurologic status was ASIA C in one patient; ASIA D in five patients, and ASIA E in 22 patients. All patients returned to ASIA E within 3 months and no intra-operative nerve injury was noted. Radiography The interbody fusion status was evaluated 1 year after surgery and at the last follow-up. Solid union, inadequate union, and pseudarthrosis occurred in 82.1 (23/28), 10.7 (3/28) and 7.2 % (2/28) of patients, respectively (Fig. 1). The total fusion rate was 82.1 %. The lordotic angle at the diseased level was 1.9 ± 13.1° pre-operatively, -10.4 ± 11.1° postoperatively, and -7.8 ± 10.4° at the last follow-up. The lumbar lordotic angle was -25.4 ± 13.8° pre-operatively, -38.8 ± 10.6° post-operatively, and -39.7 ± 15.4° at the last follow-up. The lordotic angle at the diseased level and lumbar spine were corrected by means of 9.7 and 14.3°, respectively, at the final follow-up. There was one case of postoperative aseptic implant loosening within 6 months.
Eur Spine J
The implant was removed because of persistent irritation from the loosened implant. Collapsed union of the interbody fusion was noted. Outcome The mean VAS score of back pain improved from 8.3 ± 0.8 pre-operatively to 4.8 ± 1.1, 3.2 ± 1.9, 2.6 ± 1.4 and 2.4 ± 1.5 at 1 week, 1, 3 months, and 1 year post-operatively, respectively. The mean ODI value changed from a mean of 80.2 ± 15.7 pre-operatively to 21.6 ± 16.2 at 1 year post-operatively, an overall improvement of 58.6 %. Based on the Kirkaldy–Willis functional outcomes criteria, 10, 15, 1 and 2 patients had excellent, good, fair and poor outcomes, respectively. The satisfactory rate was 89.3 %.
Discussion Antibiotics represent the primary treatment option for infective spondylodiscitis. Unfortunately, some patients require surgical debridement owing to poor infection control, neurologic deficit, back pain, spine instability or for pathogen culture. Surgery can be performed using a combined anterior and posterior approach [10–13], an anterior approach alone [14], or a posterior approach alone with posterior interbody fusion [15]. However, surgery-related morbidity is higher in anterior surgery than in posterior surgery [9, 16], and the combined approach may increase surgical risks. Although the anterior surgical approach is a gold standard for acute spondylodiscitis, it is more challenging when performed in patient with chronic infection with adhesion, in thoracolumbar and lumbosacral junctions, or in patients with a history of previous anterior surgery (Fig. 1). There were no major complications noted in this study, and it may be theorized that complications from anterior spinal surgery could be prevented using this modified posterior approach. The posterior approach for infective spondylodiscitis is still controversial. Laminectomy is contraindicated for infective spondylodiscitis, as it can lead global instability. Some authors described a one-stage posterior approach using a posterior lumbar interbody fusion (PLIF) technique with pedicle screw fixation for infective spondylodiscitis and achieved good infection control [15, 17]. PLIF requires a laminectomy and retraction of the adhesive nerve roots, and it may increase the risks of nerve injury and posterior deep wound infection and result in spinal instability. Consequently, the utility of PLIF for spondylodiscitis remains unclear. TLIF with interbody cages or bone grafts is widely accepted for treating degenerative disc disease of the lumbar
spine. TLIF is simpler and safer than PLIF [18]. Gautam et al. [19] reported good results using TLIF for lumbar tuberculosis. Our series included both non-pyogenic (3 tuberculosis and 2 fungal) and pyogenic lumbar spine infections. TLIDF can achieve adequate debridement and obtain tissue for culture (culture rate 82.1 %). Through inserting massive bone grafts, instead of cages, TLIDF can enhance interbody fusion (fusion rate 82.1 %), and prevent cage subsidence or biofilm formation. TLIDF can also preserve posterior element and prevent post-laminectomy instability. Posterior instrumentation for infective spondylodiscitis may improve the surgical outcome, accelerate healing, and prevent kyphotic deformity formation [8, 20], but the use of instrumentation in spinal infection remains controversial. Few studies reported the safety of posterior spinal instrumentation using a posterior approach only [15, 19]. We performed short fusion via TLIDF combined with short or long instrumentation (the implants could be removed after solid fusion). The sagittal alignment of the treated levels and lumbar spine were corrected by averages of 9.7 and 14.3°, respectively, at the final follow-up. The mean lordotic angle of lumbar spine at the last follow-up was -39.7°. Instrumentation-related complication was rare. TLIDF with pedicle screws fixation for infective spondylodiscitis is safe and it may provide immediate stability and acceptable interbody fusion rates and maintain post-operative alignment. The VAS score for back pain was improved from 8.3 to 4.8 at 1 week post-operatively, and to 2.4 after 1 year. Ambulation and rehabilitation with adequate brace protection could be performed as soon as possible after surgery. There was a 58.6 % improvement of the ODI 1 year after surgery. The Kirkaldy–Willis function outcomes were evaluated, and the satisfactory rate was 89.3 %. All patients with neurologic deficits displayed improvement. TLIDF with posterior pedicle screws instrumentation could improve back pain, neurologic deficits, and functional outcome. In conclusion, a single posterior approach with TLIDF and pedicle screws instrumentation for infective spondylodiscitis of the lumbar spine is safe and effective. This strategy has the additional benefit of avoiding the risks of anterior or staging surgeries. Conflict of interest
None.
References 1. Wiley AM, Trueta J (1959) The vascular anatomy of the spine and its relationship to pyogenic vertebral osteomyelitis. J Bone Jt Surg Br 41:796–809 2. Batson OV (1967) The vertebral systemic of veins as a means for cancer dissemination. Prog Clin Cancer 3:1–18
123
Eur Spine J 3. Hadjipavlou AG, Mader JT, Necessary JT, Muffoletto AJ (2000) Hematogenous pyogenic spinal infections and their surgical management. Spine (Phila Pa 1976) 25:1668–1679 4. Sapico FL, Montgomerie JZ (1979) Pyogenic vertebral osteomyelitis: report of nine cases and review of the literature. Rev Infect Dis 1:754–776 5. Mehta JS, Bhojraj SY (2001) Tuberculosis of the thoracic spine. J Bone Jt Surg Br 83(6):859–863 6. Emery SE, Chan DP, Woodward HR (1989) Treatment of hematogenous pyogenic vertebral osteomyelitis with anterior debridement. Spine (Phila Pa 1976) 14:284–291 7. Hodgson AR, Stock FE, Fang HS, Ong GB (1960) Anterior spinal fusion. The operative approach and pathological findings in 412 patients with Pott’s disease of the spine. Br J Surg 48:172–178 8. Chen WJ, Wu CC, Jung CH, Chen LH, Niu CC, Lai PL (2002) Combined anterior and posterior surgeries in the treatment of spinal tuberculous spondylitis. Clin Orthop Relat Res 398:50–59 9. Sasso RC, Best NM, Mummaneni PV, Reilly TM, Hussain SM (2005) Analysis of operative complications in a series of 471 anterior lumbar interbody fusion procedures. Spine (Phila Pa 1976) 30:670–674 10. Rath SA, Neff U, Schneider O, Richter HP (1996) Neurosurgical management of thoracic and lumbar vertebral osteomyelitis and discitis in adults: a review of 43 consecutive surgically treated patients. Neurosurgery 38:926–933 11. Korovessis P, Petsinis G, Koureas G, Iliopoulos P, Zacharatos S (2006) Anterior surgery with insertion of titanium mesh cage and posterior instrumented fusion performed sequentially on the same day under one anesthesia for septic spondylitis of thoracolumbar spine: is the use of titanium mesh cages safe? Spine (Phila Pa 1976) 31:1014–1019 12. Ruf M, Stoltze D, Merk HR, Ames M, Harms J (2007) Treatment of vertebral osteomyelitis by radical debridement and
123
13.
14.
15.
16.
17.
18.
19.
20.
stabilization using titanium mesh cages. Spine (Phila Pa 1976) 32:E275–E280 Pee YH, Park JD, Choi YG, Lee SH (2008) Anterior debridement and fusion followed by posterior pedicle screw fixation in pyogenic spondylodiscitis: autologous iliac bone strut versus cage. J Neurosurg Spine 8:405–412 D’Aliberti G, Talamonti G, Villa F, Debernardi A (2012) The anterior stand-alone approach (ASAA) during the acute phase of spondylodiscitis: results in 40 consecutively treated patients. Eur Spine J 21(Suppl 1):S75–S82 Lee JS, Suh KT (2006) Posterior lumbar interbody fusion with an autogenous iliac crest bone graft in the treatment of pyogenic spondylodiscitis. J Bone Jt Surg Br 88:765–770 Fritzell P, Hagg O, Wessberg P, Nordwall A (2002) Chronic low back pain and fusion: a comparison of three surgical techniques: A prospective multicenter randomized study from the Swedish Lumbar Spine Study Group. Spine (Phila Pa 1976) 27:1131–1141 Endres S, Wilke A (2012) Posterior interbody grafting and instrumentation for spondylodiscitis. J Orthop Surg (Hong Kong) 20:1–6 Yan DL, Pei FX, Li J, Soo CL (2008) Comparative study of PILF and TLIF treatment in adult degenerative spondylolisthesis. Eur Spine J 17:1311–1316 Zaveri GR, Mehta SS (2009) Surgical treatment of lumbar tuberculous spondylodiscitis by transforaminal lumbar interbody fusion (TLIF) and posterior instrumentation. J Spinal Disord Tech 22(4):257–262 Rayes M, Colen CB, Bahqat DA et al (2010) Safety of instrumentation in patients with spinal infection. J Neurosurg Spine 12:647–659
ORIGINAL ARTICLE
Transforaminal lumbar interbody debridement and fusion for the treatment of infective spondylodiscitis in the lumbar spine Meng-Ling Lu • Chi-Chien Niu • Tsung-Ting Tsai Tsai-Sheng Fu • Lih-Huei Chen • Wen-Jer Chen
•
Received: 9 December 2013 / Revised: 10 September 2014 / Accepted: 12 September 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose To determine the safety and efficacy of using a single posterior approach with transforaminal lumbar interbody debridement and fusion (TLIDF) plus pedicle screws fixation in treating infective spondylodiscitis in the lumbar spine. Methods Between January 2009 and June 2011, 28 patients with infective spondylodiscitis who underwent TLIDF, using autogenous graft and posterior pedicle screws instrumentation, met the indications for surgery, and completed more than 18 months of follow-up, were included. Clinical outcomes were assessed using a visual analog scale (VAS), the Oswestry Disability Index (ODI), and Kirkaldy-Willis functional outcome criteria. Infection status was evaluated using C-reactive protein levels, erythrocyte sedimentation rate, and clinical symptoms. The interbody fusion status and sagittal alignment of the infected segments were assessed using radiographic studies. Results Intra-operative culture rate was 82.1 %. The most common pathogen was methicillin-resistant Staphylococcus aureus. One post-operative deep wound infection with septic implant loosening and one instance of early aseptic implant loosening were noted. Implants in both patients were subsequently removed. Two infections recurred
M.-L. Lu C.-C. Niu T.-T. Tsai T.-S. Fu L.-H. Chen W.-J. Chen (&) Department of Orthopedic Surgery, Chang Gung Memorial Hospital, Chang Gung University, No 5, Fu-Hsing Street 333, Kweishan, Taoyuan, Taiwan e-mail: [email protected] C.-C. Niu T.-T. Tsai T.-S. Fu L.-H. Chen W.-J. Chen College of Medicine, Chang Gung University, Taoyuan, Taiwan
within 3 months post-operatively, with both subsiding within 3 months after extended antibiotic treatment. VAS and ODI values were significantly improved. The interbody fusion rate was 82.1 % and the lordotic angle of the infected segments and the lumbar spine were corrected by 9.7° and 14.3°, respectively at the last follow-up. Conclusions A single posterior approach with TLIDF and pedicle screws instrumentation for lumbar infective spondylodiscitis provided a satisfactory clinical outcome, adequate infection control and good fusion rate, and this strategy avoids the risks of anterior or staging surgeries. Keywords Infective spondylodiscitis Lumbar spine Transforaminal lumbar interbody fusion Transforaminal lumbar interbody debridement and fusion
Introduction There are several types of spinal infections, including spondylodiscitis, discitis, epidural abscess, septic facet arthropathy and ‘‘central body lesions’’ in tuberculosis. Based on the type of pathogens, the infection can be classified as pyogenic or nonpyogenic. Regardless of whether one subscribes to the arteriole [1] or venous theory [2], the vascularity of vertebrae is extremely rich near the end-plate, which allows pathogens entrapped in the endplate to cause destruction of the disc and adjacent vertebrae. Clinically, spondylodiscitis is the most common type spinal infection, and it occurs most commonly in the lumbar region [3]. Infective spondylodiscitis can result in disc destruction, pathologic fracture, and abscess formation. Pain is the most common symptom and neurologic deficit may affect up to 17 % of the patients [4]. Surgical treatment is indicated for neurologic impairment, spinal
123
Eur Spine J
instability, deformity, unknown pathogen, poorly controlled infection and intractable pain. Treatment of spinal infections remains challenging for spinal surgeons. The goals of surgical treatment are debridement of infective foci, collection of pathogens, decompression of neural element, and re-stabilization of the deformed spine. Regardless of whether the infection is pyogenic or nonpyogenic, the infective foci of infective spondylodiscitis are always located around the involved disc, and hence, debridement of the infected disc with reconstruction of the interbody bony defect must be considered first [5]. A posterior approach with laminectomy is contraindicated, as it will result in global instability. An anterior approach for infective spondylodiscitis may achieve more radical debridement and fusion [6, 7]. The addition of posterior instrumentation to anterior procedures can stabilize the spine and prevent kyphotic deformity [8]. The complication rate of anterior spinal surgery is relatively higher than that of posterior spinal surgery [9], especially when performed in patients with chronic lesions with adhesion, in thoracolumbar and lumbosacral junctions, or in patients with a history of previous anterior surgery. Combining anterior and posterior approaches may increase the risks, especially in immunocompromised patients. The purposes of this study were to determine the safety and efficacy of a single posterior approach with transforaminal lumbar interbody debridement and fusion (TLIDF) plus pedicle screws fixation for infective spondylodiscitis in the lumbar spine.
Materials and methods Thirty patients with infective spondylodiscitis of the lumbar spine underwent TLIDF and pedicle screws instrumentation between January 2009 and June 2011 at the author’s institution. Patients who were followed-up for more than 18 months were enrolled. Of these, 28 were included in this retrospective study. Medical records, imaging studies, laboratory data, neurologic function data and functional outcomes were reviewed and analyzed. The indications for TLIDF included: progressive neurologic deficit, spinal instability, deformity, intractable pain, unknown pathogen, and a failed response to antibiotic treatment. Patients displaying destruction of more than 50 % [based on pre-operative radiography and magnetic resonance imaging (MRI)] of vertebral bodies were excluded. We modified the surgical procedure of transforaminal lumbar interbody fusion (TLIF), and performed debridement, fusion and instrumentation using a single posterior approach, naming the modified procedure TLIDF. Surgical time, the affected level, the instrumented level,
123
intra-operative blood loss, and complications were recorded. All patients were encouraged to ambulate while wearing a Taylor brace for 12 weeks. Patients were followed-up at 1, 3, 6 months and then annually. Imaging studies including pre-operative and post-operative plain radiography of the lumbar spine were performed. All patients underwent MRI of the lumbosacral spine before operation. Using the superior and inferior endplates of the diseased vertebra as reference points, the lordotic angle of the lesion site was recorded pre-operatively, post-operatively, and on a yearly basis thereafter. The lumbar lordotic angle measured from lower end-plate of T12 to upper end-plate of S1 was also recorded simultaneously. The interbody fusion status was analyzed at the final follow-up. Radiographic outcome were assessed by a radiologist who was blinded to the operation. Solid fusion was defined as continuous trabeculae bridging across the interface of the treated segments, whereas the presence of any radiolucent interruption at the treated segments was defined as pseudarthrosis. The fusion was classified as inadequate when the radiographic image was vague which could not be defined as solid or pseudarthrosis. According to the culture results and recommendations of the infectious disease specialist, antibiotic treatment was initiated after surgery. Laboratory data including a white blood count with a differential count, erythrocyte sedimentation rates (ESRs), and C-reactive protein (CRP) levels were checked serially and recorded. According to the clinical symptoms and the ESR and CRP data, pyogenic spondylodiscitis and fungus infection were treated with 4–6 weeks of parenteral antibiotics followed by oral antibiotic for totally 3 months. Tuberculosis was treated with 12 months of oral anti-tuberculosis drug administration. Clinical outcomes were evaluated using the patientassessed quantitative measurement of the visual analog scale (VAS) for back pain, the Oswestry Disability Index (ODI) and Kirkaldy-Willis functional outcome criteria. Neurologic status was evaluated using the American Spinal Injury Association (ASIA) scale. Operative procedure While under general anesthesia, each patient was placed on a four-poster spinal frame in the prone position. Posterior iliac bone was harvested first. One posterior midline skin incision, followed by a bilateral para-median approach (Wiltse approach), was performed. The TLIDF side was determined pre-operatively, according to the severity of bony destruction, epidural abscess, and radiculopathy. Unilateral facetectomy and limited laminotomy were performed to expose the infected disc at the foraminal level. The pus, infected disc, end-plate and sequestrum were removed using the TLIF procedure and the psoas abscess
Eur Spine J
Fig. 1 A 64-year-old woman who experienced tuberculosis infection and underwent anterior debridement twice in another hospital. X-ray revealed L1–2 disc space narrowing with end-plate destruction and kyphosis (a); T1-weighted magnetic resonance imaging with contrast revealed vertebral body destruction and psoas abscess (b, c). The patient underwent TLIDF with two above and two below posterior instrumentation (T11-L4) (d). 1-year post-operative radiograph
revealed solid interbody fusion at L1–2 (e). Spine computed tomography revealed continuous trabeculae bridging across the interface of the treated segments on the sagittal view (f). Adequate debridement and decompression were noted on the axial view (g). Another axial view revealed no screw loosening and good screw position (h)
was drained posterolaterally under blunt dissection. In patients with neurologic compromise, the dural compression from the anterior infected lesion could be well-
decompressed, synchronously. The samples were cultured for aerobic and anaerobic microbes, tuberculous bacilli and fungi and the extracted tissue was sent for
123
Eur Spine J
histopathological examination. After adequate debridement and curettage, the autogenous iliac bone grafts (corticocancellous and cancellous bone chips) were inserted for interbody fusion. Trans-pedicle screws were inserted into the intact bodies one level superior and one level inferior to the diseased disc. However, in the presence of greater bony destruction or osteoporosis, the fixation was extended over a longer distance.
Results Of these 30 patients, two were lost to follow-up and excluded from this study. The remaining 28 patients (13 men and 15 women) were followed-up for more than 18 months. The mean age at surgery was 60.4 years (range 37–86 years). The mean follow–up period was 20 months (range 18–36 months). No patient died due to spinal infection; however, one patient died from cancer 18 months after the operation. Three patients underwent percutaneous endoscopic debridement and drainage, and two underwent anterior debridement prior to this surgical intervention. 15 patients received empiric antibiotics before the operation. The involved levels were T12-L1 in two patients, L1–2 in two patients, L2–3 in two patients, L3–4 in two patients, L4–5 in nine patients, L5–S1 in 10 patients and L1-2-3 in one patient. The mean operative time was 179.6 ± 30.3 min (range 90–240 min), and the mean blood loss was 712.5 ± 299.9 mL (range 200–1, 400 mL). The instrumentation levels were one above one below in 12 patients, two above two below in 14 patients (including two patients who required iliac screw fixation) (Fig. 1), and three above three below in two patients. Infection status Twenty-six patients exhibited hematogenous infection. Two patients exhibited post-operative infections (1 discectomy and 1 instrumented vertebral fracture combined infection). The intra-operative culture rate was 82.1 %. The most common pyogenic pathogen was methicillin-resistant Staphylococcus aureus (7 patients) and the most common non-pyogenic pathogen was tuberculosis (3 patients) (Table 1). All histopathologic examinations revealed acute and chronic inflammation, and two examinations also revealed granulomatous inflammation. For most patients, the CRP levels and ESRs normalized within 3 months after operation. One patient (L5–S1 infection, Klebsiella pneumoniae) experienced a post-operative deep wound infection that extended to the L4–5 level. He underwent wound debridement; transforaminal discectomy at L4–5; and removal of implants. The pathogen was found to be drug-
123
Table 1 Cultures of spinal infection Pathogen
%
No. of patients
MRSA
25
7
Tuberculosis
10.7
3
Klebsiella pneumoniae
10.7
3
MSSA
7.1
2
Pseudomonas aeruginosa
3.6
1
Salmonella
3.6
1
Enterococcus faecalis
3.6
1
Gram-negative bacilli, G group
3.6
1
Aerococcus viridans
3.6
1
CoNS
3.6
1
Fungus (Cladosporium) Multiplea
3.6 3.6
1 1
No organism
17.9
5
Culture rate
82.1
28
MRSA methicillin–resistant Staphylococcus aureus, MSSA methicillin sensitive Staphylococcus aureus, CoNS coagulase negative Staphylococcus a
Pseudomonas aeruginosa, Enterococcus faecalis and Candida
resistant and the infection was eventually controlled. Two patients experienced recurrent back pain with elevated CRP levels during the 3 months after discharge. Both patients received extended intravenous antibiotics for 4–6 weeks, and the infections subsided. Neurologic status Two patients displayed cauda-equina syndrome, with both recovering after the operation. The pre-operative neurologic status was ASIA C in one patient; ASIA D in five patients, and ASIA E in 22 patients. All patients returned to ASIA E within 3 months and no intra-operative nerve injury was noted. Radiography The interbody fusion status was evaluated 1 year after surgery and at the last follow-up. Solid union, inadequate union, and pseudarthrosis occurred in 82.1 (23/28), 10.7 (3/28) and 7.2 % (2/28) of patients, respectively (Fig. 1). The total fusion rate was 82.1 %. The lordotic angle at the diseased level was 1.9 ± 13.1° pre-operatively, -10.4 ± 11.1° postoperatively, and -7.8 ± 10.4° at the last follow-up. The lumbar lordotic angle was -25.4 ± 13.8° pre-operatively, -38.8 ± 10.6° post-operatively, and -39.7 ± 15.4° at the last follow-up. The lordotic angle at the diseased level and lumbar spine were corrected by means of 9.7 and 14.3°, respectively, at the final follow-up. There was one case of postoperative aseptic implant loosening within 6 months.
Eur Spine J
The implant was removed because of persistent irritation from the loosened implant. Collapsed union of the interbody fusion was noted. Outcome The mean VAS score of back pain improved from 8.3 ± 0.8 pre-operatively to 4.8 ± 1.1, 3.2 ± 1.9, 2.6 ± 1.4 and 2.4 ± 1.5 at 1 week, 1, 3 months, and 1 year post-operatively, respectively. The mean ODI value changed from a mean of 80.2 ± 15.7 pre-operatively to 21.6 ± 16.2 at 1 year post-operatively, an overall improvement of 58.6 %. Based on the Kirkaldy–Willis functional outcomes criteria, 10, 15, 1 and 2 patients had excellent, good, fair and poor outcomes, respectively. The satisfactory rate was 89.3 %.
Discussion Antibiotics represent the primary treatment option for infective spondylodiscitis. Unfortunately, some patients require surgical debridement owing to poor infection control, neurologic deficit, back pain, spine instability or for pathogen culture. Surgery can be performed using a combined anterior and posterior approach [10–13], an anterior approach alone [14], or a posterior approach alone with posterior interbody fusion [15]. However, surgery-related morbidity is higher in anterior surgery than in posterior surgery [9, 16], and the combined approach may increase surgical risks. Although the anterior surgical approach is a gold standard for acute spondylodiscitis, it is more challenging when performed in patient with chronic infection with adhesion, in thoracolumbar and lumbosacral junctions, or in patients with a history of previous anterior surgery (Fig. 1). There were no major complications noted in this study, and it may be theorized that complications from anterior spinal surgery could be prevented using this modified posterior approach. The posterior approach for infective spondylodiscitis is still controversial. Laminectomy is contraindicated for infective spondylodiscitis, as it can lead global instability. Some authors described a one-stage posterior approach using a posterior lumbar interbody fusion (PLIF) technique with pedicle screw fixation for infective spondylodiscitis and achieved good infection control [15, 17]. PLIF requires a laminectomy and retraction of the adhesive nerve roots, and it may increase the risks of nerve injury and posterior deep wound infection and result in spinal instability. Consequently, the utility of PLIF for spondylodiscitis remains unclear. TLIF with interbody cages or bone grafts is widely accepted for treating degenerative disc disease of the lumbar
spine. TLIF is simpler and safer than PLIF [18]. Gautam et al. [19] reported good results using TLIF for lumbar tuberculosis. Our series included both non-pyogenic (3 tuberculosis and 2 fungal) and pyogenic lumbar spine infections. TLIDF can achieve adequate debridement and obtain tissue for culture (culture rate 82.1 %). Through inserting massive bone grafts, instead of cages, TLIDF can enhance interbody fusion (fusion rate 82.1 %), and prevent cage subsidence or biofilm formation. TLIDF can also preserve posterior element and prevent post-laminectomy instability. Posterior instrumentation for infective spondylodiscitis may improve the surgical outcome, accelerate healing, and prevent kyphotic deformity formation [8, 20], but the use of instrumentation in spinal infection remains controversial. Few studies reported the safety of posterior spinal instrumentation using a posterior approach only [15, 19]. We performed short fusion via TLIDF combined with short or long instrumentation (the implants could be removed after solid fusion). The sagittal alignment of the treated levels and lumbar spine were corrected by averages of 9.7 and 14.3°, respectively, at the final follow-up. The mean lordotic angle of lumbar spine at the last follow-up was -39.7°. Instrumentation-related complication was rare. TLIDF with pedicle screws fixation for infective spondylodiscitis is safe and it may provide immediate stability and acceptable interbody fusion rates and maintain post-operative alignment. The VAS score for back pain was improved from 8.3 to 4.8 at 1 week post-operatively, and to 2.4 after 1 year. Ambulation and rehabilitation with adequate brace protection could be performed as soon as possible after surgery. There was a 58.6 % improvement of the ODI 1 year after surgery. The Kirkaldy–Willis function outcomes were evaluated, and the satisfactory rate was 89.3 %. All patients with neurologic deficits displayed improvement. TLIDF with posterior pedicle screws instrumentation could improve back pain, neurologic deficits, and functional outcome. In conclusion, a single posterior approach with TLIDF and pedicle screws instrumentation for infective spondylodiscitis of the lumbar spine is safe and effective. This strategy has the additional benefit of avoiding the risks of anterior or staging surgeries. Conflict of interest
None.
References 1. Wiley AM, Trueta J (1959) The vascular anatomy of the spine and its relationship to pyogenic vertebral osteomyelitis. J Bone Jt Surg Br 41:796–809 2. Batson OV (1967) The vertebral systemic of veins as a means for cancer dissemination. Prog Clin Cancer 3:1–18
123
Eur Spine J 3. Hadjipavlou AG, Mader JT, Necessary JT, Muffoletto AJ (2000) Hematogenous pyogenic spinal infections and their surgical management. Spine (Phila Pa 1976) 25:1668–1679 4. Sapico FL, Montgomerie JZ (1979) Pyogenic vertebral osteomyelitis: report of nine cases and review of the literature. Rev Infect Dis 1:754–776 5. Mehta JS, Bhojraj SY (2001) Tuberculosis of the thoracic spine. J Bone Jt Surg Br 83(6):859–863 6. Emery SE, Chan DP, Woodward HR (1989) Treatment of hematogenous pyogenic vertebral osteomyelitis with anterior debridement. Spine (Phila Pa 1976) 14:284–291 7. Hodgson AR, Stock FE, Fang HS, Ong GB (1960) Anterior spinal fusion. The operative approach and pathological findings in 412 patients with Pott’s disease of the spine. Br J Surg 48:172–178 8. Chen WJ, Wu CC, Jung CH, Chen LH, Niu CC, Lai PL (2002) Combined anterior and posterior surgeries in the treatment of spinal tuberculous spondylitis. Clin Orthop Relat Res 398:50–59 9. Sasso RC, Best NM, Mummaneni PV, Reilly TM, Hussain SM (2005) Analysis of operative complications in a series of 471 anterior lumbar interbody fusion procedures. Spine (Phila Pa 1976) 30:670–674 10. Rath SA, Neff U, Schneider O, Richter HP (1996) Neurosurgical management of thoracic and lumbar vertebral osteomyelitis and discitis in adults: a review of 43 consecutive surgically treated patients. Neurosurgery 38:926–933 11. Korovessis P, Petsinis G, Koureas G, Iliopoulos P, Zacharatos S (2006) Anterior surgery with insertion of titanium mesh cage and posterior instrumented fusion performed sequentially on the same day under one anesthesia for septic spondylitis of thoracolumbar spine: is the use of titanium mesh cages safe? Spine (Phila Pa 1976) 31:1014–1019 12. Ruf M, Stoltze D, Merk HR, Ames M, Harms J (2007) Treatment of vertebral osteomyelitis by radical debridement and
123
13.
14.
15.
16.
17.
18.
19.
20.
stabilization using titanium mesh cages. Spine (Phila Pa 1976) 32:E275–E280 Pee YH, Park JD, Choi YG, Lee SH (2008) Anterior debridement and fusion followed by posterior pedicle screw fixation in pyogenic spondylodiscitis: autologous iliac bone strut versus cage. J Neurosurg Spine 8:405–412 D’Aliberti G, Talamonti G, Villa F, Debernardi A (2012) The anterior stand-alone approach (ASAA) during the acute phase of spondylodiscitis: results in 40 consecutively treated patients. Eur Spine J 21(Suppl 1):S75–S82 Lee JS, Suh KT (2006) Posterior lumbar interbody fusion with an autogenous iliac crest bone graft in the treatment of pyogenic spondylodiscitis. J Bone Jt Surg Br 88:765–770 Fritzell P, Hagg O, Wessberg P, Nordwall A (2002) Chronic low back pain and fusion: a comparison of three surgical techniques: A prospective multicenter randomized study from the Swedish Lumbar Spine Study Group. Spine (Phila Pa 1976) 27:1131–1141 Endres S, Wilke A (2012) Posterior interbody grafting and instrumentation for spondylodiscitis. J Orthop Surg (Hong Kong) 20:1–6 Yan DL, Pei FX, Li J, Soo CL (2008) Comparative study of PILF and TLIF treatment in adult degenerative spondylolisthesis. Eur Spine J 17:1311–1316 Zaveri GR, Mehta SS (2009) Surgical treatment of lumbar tuberculous spondylodiscitis by transforaminal lumbar interbody fusion (TLIF) and posterior instrumentation. J Spinal Disord Tech 22(4):257–262 Rayes M, Colen CB, Bahqat DA et al (2010) Safety of instrumentation in patients with spinal infection. J Neurosurg Spine 12:647–659
