Posterior lumbar interbody fusion with an autogenous iliac crest bone graft in the treatment of pyogenic spondylodiscitis

Transcription

1 Spine Posterior lumbar interbody fusion with an autogenous iliac crest bone graft in the treatment of pyogenic spondylodiscitis J. S. Lee, K. T. Suh From the Pusan National University Hospital, Pusan, Korea There are few reports on the treatment of pyogenic lumbar spondylodiscitis through the posterior approach using a single incision. Between October 1999 and March 2003 we operated on 18 patients with pyogenic lumbar spondylodiscitis. All underwent posterior lumbar interbody fusion using an autogenous bone graft from the iliac crest and pedicle screws via a posterior approach. The clinical outcome was assessed using the Frankel neurological classification and the criteria of Kirkaldy-Willis. Under the Frankel classification, two patients improved by two grades (C to E), 11 by one grade, and five showed no change. The Kirkaldy-Willis functional outcome was excellent in five patients, good in ten and fair in three. Bony union was confirmed six months after surgery in 17 patients, but in one patient this was not achieved until two years after operation. The mean lordotic angle before operation was 20 (-2 to 42 ) and the mean lordotic angle at the final follow-up was 32.5 (17 to 44 ). Two patients had a superficial wound infection and two a transient root injury. Posterior lumbar interbody fusion with an autogenous iliac crest bone graft and pedicle screw fixation via a posterior approach can provide satisfactory results in pyogenic spondylodiscitis. J. S. Lee, MD, Assistant Professor K. T. Suh, MD, PhD, Professor Department of Orthopaedic Surgery Pusan National University, College of Medicine, 1-10 Ami- Dong, Seo-Gu, Pusan , Korea. Correspondence should be sent to Professor K. T. Suh; kuentak@pusan.ac.kr 2006 British Editorial Society of Bone and Joint Surgery doi: / x.88b $2.00 J Bone Joint Surg [Br] 2006;88-B: Received 6 October 2005; Accepted after revision 5 January 2006 Surgical treatment for pyogenic lumbar spondylodiscitis may be required for those who do not respond to antibiotic treatment or in the presence of a collapsed lumbar body, an epidural abscess or a neurological deficit. 1-7 Although various approaches to the surgical management of pyogenic lumbar spondylodiscitis have been described, 1,3,6,8-12 many patients in whom medical management has failed have been treated by debridement, followed several weeks later by internal fixation. 1,2,13-15 Some have found percutaneous transpedicular discectomy to be an excellent treatment in the early stages of uncomplicated spondylodiscitis. 16,17 Others have suggested that internal fixation can be left in place if debridement is performed and a prolonged course of intravenous antibiotics given. 8,11,18 Others have used antibioticimpregnated methylmethacrylate 8 and continuous suction/irrigation systems. 10 However, there are few reports on the treatment of pyogenic lumbar spondylodiscitis using posterior lumbar interbody fusion and pedicle screw fixation. 19,20 The aims of this procedure are to remove the infected disc, promote natural wound healing by firm fixation and to facilitate early mobilisation and rehabilitation. Patients and Methods This study was a retrospective review of 18 patients treated by one surgeon (JSL) for pyogenic lumbar spondylodiscitis between October 1999 and March Their details are shown in Table I. The medical records were reviewed and the patients recalled for followup. There were ten women and eight men. Their mean age at the time of surgery was 57.7 years (44 to 71), and the mean period of follow-up was 32 months (24 to 48). Most complained of severe lower back pain and disturbance of gait over a period of one to eight months. Of these, 16 had neurological deficits (Frankel 21 B to D) which included motor weakness, sensory changes, pain radiating to the lower limbs, and urinary symptoms. There were five patients with diabetes mellitus, three were hypertensive, two had a gastric ulcer and one was in chronic renal failure. The L2-L3 level was affected in one patient, L3-L4 in six, L4-L5 in eight and L5-S1 in three. Between one and three levels were fused. The presumed cause of the infection was an epidural injection in nine patients, a post-laminectomy infection in four, and in five the cause was unknown. Operation was indicated when the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels did not return to normal VOL. 88-B, No. 6, JUNE

2 766 J. S. LEE, K. T. SUH Table I. Details of the patients Case Age (yrs) Gender Medical problem * Spinal stenosis Epidural abscess Level Area for instrumentation Duration of symptoms (mths) Presumed cause Organism Post-operative antibiotics (wks) 1 68 M L3-L4 L3-L4 3 Epidural injection S. aureus 6 IV/12 PO 2 54 F GU + - L4-L5 L4-L5 2 Epidural injection No growth 6 IV/8 PO 3 52 F DM - + L5-S1 L4-S1 3 Post-laminectomy S. aureus 6 IV/8 PO 4 49 M L4-L5 L4-L5 5 Unknown C. luteola 6 IV/8 PO 5 71 M L3-L4 L3-L4 6 Post-laminectomy S. aureus 6 IV/6 PO 6 62 F DM + + L3-L4 L3-L4 2 Epidural injection No growth 7 IV/6 PO 7 53 F HT - + L4-L5 L4-L5 3 Epidural injection MRSA 8 IV/12 PO 8 51 F HT - + L4-L5 L4-S1 1 Unknown No growth 6 IV/12 PO 9 66 M L2-L3 L2-L3 2 Post-laminectomy No growth 6 IV/6 PO F HT - + L3-L4 L2-L5 5 Epidural injection S. aureus 6 IV/4 PO M L4-L5 L4-L5 4 Unknown No growth 6 IV/8 PO F DM - - L4-L5 L4-S1 1 Epidural injection MRSA 12 IV/24 PO M CRF + + L3-L4 L3-L5 5 Epidural injection B. cepacia 4 IV/6 PO F L5-S1 L4-S1 2 Post laminectomy No growth 6 IV/6 PO M DM - + L4-L5 L4-L5 1 Epidural injection S. aureus 6 IV/8 PO F GU + - L3-L4 L3-L5 4 Unknown No growth 6 IV/8 PO M L4-L5 L4-S1 8 Unknown No growth 6 IV/6 PO F DM + + L5-S1 L5-S1 2 Epidural injection No growth 6 IV/6 PO * GU, gastric ulcer; DM, diabetes mellitus; HT, hypertension; CRF, chronic renal failure +, present; -, absent IV, intravenously; PO, orally and pain persisted after the administration of antibiotics for more than three weeks, or when neurological changes were noted as a result of an abscess compressing the dura. Posterior lumbar interbody fusion with posterior fixation was indicated in patients with a small abscess in the anterior portion of the vertebral body connected to a disc, so that drainage could be accomplished through the posterior approach. It was also indicated in those with spinal stenosis and those in whom the anterior approach could not be used because the abscess was in the posterior portion of the dura. The white blood cell, ESR and CRP levels were determined and plain radiographs, radionucleide studies and magnetic resonance imaging (MRI) were undertaken before surgery. A computed tomography (CT)-guided fine-needle biopsy of the infected disc space was performed in all patients. The disc material and infected end-plate removed during surgery were sent for culture and histology. Plain and dynamic radiographs were obtained in all patients 3, 6 and 12 months after surgery. At the final follow-up evaluation, the fusion and measurements of the lordotic angle were carried out by two surgeons (JSL, KTS). CT was also used to assess the fusion and the formation of a bone bridge. The radiographs were graded according to the criteria of Lee, Vessa and Lee 22 (Table II). In our study we classified both definitive and probable fusion as a fusion. The lumbar lordotic angle was measured from the upper end-plate of T12 to the upper end-plate of S1 preoperatively, post-operatively, and at the final follow-up. The functional outcome was classified according to the Frankel classification 21 and the criteria of Kirkaldy-Willis (Table III). 23 Operative technique. A posterior laminectomy and pedicle screw fixation was performed under general anaesthesia. When possible, a pedicle screw was not applied directly to the infected vertebral body. When the infection and destruction of the vertebral body was severe, we fixed one level above or below the segment. After pedicle screw fixation, the disc space was distracted and the infected endplate, disc and soft tissue removed. The space created after Table II. The criteria of Lee et al 22 for results of radiographic fusion Grade Description Definitive fusion Definitive bony trabecular bridging across the graft-host interface, no motion (< 3 ) on flexionextension radiograph, and no gap at interface Probable fusion No definitive bony trabecular crossing, but no detectable motion and no identifiable gap at the interface Possible pseudarthrosis No bony trabecular crossing, no movement but identifiable gap at the interface Definite pseudarthrosis No traversing trabecular bone, definitive gap, and movement > 3 Table III. The criteria of Kirkaldy-Willis et al 23 for functional outcome Grade Description Excellent The patient has returned to his normal work and other activities with little or no complaint Good The patient has returned to his normal work but may have some restriction in other activities, and may on occasion after heavy work have recurrent back pain requiring a few days rest Fair The patient has to reduce his working capacity, taking a lighter job or working part-time, and may occasionally have recurrence of pain requiring absence from work for one to two weeks, once or twice a year Poor The patient does not return to work THE JOURNAL OF BONE AND JOINT SURGERY

3 POSTERIOR LUMBAR INTERBODY FUSION WITH AN AUTOGENOUS ILIAC CREST BONE GRAFT 767 Fig. 1a Fig. 1b Case 3 A 52-year-old woman with diabetes mellitus developed an L5-S1 staphylococcal spondylodiscitis. She underwent posterior lumbar interbody fusion with an autogenous strut iliac graft. a) pre-operative anteroposterior and lateral radiographs show L5-S1 disc interspace narrowing and irregular end-plates (arrow); b) T2-weighted sagittal and contrastenhanced T1-weighted axial images show an epidural abscess (arrow) and destruction of the L5-S1 disc interspace; c) lateral radiograph taken two years after surgery shows radiolucency around the pedicle screw. The functional outcome was excellent. The autogenous strut iliac graft has united (arrow). Fig. 1c meticulous debridement was filled with an autogenous iliac bone graft. A tricortical iliac graft was used in all patients. After this graft had been firmly fixed by compression of the space, a posterolateral graft was performed in order to obtain a 360 fusion. The patients were mobilised three days after surgery and wore a lumbosacral orthosis for three months. The appropriate antibiotics were administered intravenously for a mean of 6.4 weeks (4 to 12) after operation, followed by oral administration for a mean of 8.6 weeks (4 to 24). Statistical analysis. Serial changes of white blood cell, ESR, and CRP levels and lordotic angle were assessed using the Friedman test with Dunn s post-multiple comparison test. The differences in these measurements between the stenotic and the non-stenotic groups and the patients with and without an epidural abscess were analysed using the Mann- Whitney U test. The final results at follow-up between these groups of patients were compared using Fisher s exact test as appropriate. A value of p < 0.05 was defined as statistically significant. Results Bony union and change in lordotic angle. Bony union was obtained by six months after operation in 17 patients (Fig. 1). In one patient union was not achieved until two years after surgery (Fig. 2). The mean lordotic angle before operation was 20 (-2 to 42 ), improving after the procedure to 34.3 (19 to 47 ). At the final follow-up it was 32.5 (17 to 44 ) (Table IV). These changes were statistically significant (p < 0.05). Clinical. Although a pre-operative needle biopsy was performed in all patients, bacterial growth was obtained in only three cases. Bacteria were cultured from material removed from the disc and the infected end-plate in nine patients, including Staphylococcus aureus in five, methicillin-resistant Staphylococcus aureus in two, Cheryscomonas luteola in one, and Burkholderia cepacia in one. Histological examination of the material removed from the disc and the infected end-plate confirmed the presence of pyogenic lumbar spondylodiscitis in all patients. VOL. 88-B, No. 6, JUNE 2006

4 768 J. S. LEE, K. T. SUH Fig. 2a Case 11 A 65-year-old man with spinal stenosis developed L4-L5 spondylodiscitis and underwent posterior lumbar interbody fusion with an autogenous strut iliac graft. a) contrast-enhanced T1-weighted sagittal and axial images show destruction of the L4-L5 disc interspace and extension along the anterior and posterior longitudinal ligaments; b) pre-operative (left) and postoperative lateral radiographs at one (centre) and two years (right), showing that bony union has been achieved (arrows). Fig. 2b On admission, the mean white blood cell count was 9041 ml (5120 to ), the mean ESR was 71 mm/h (12 to 113), and the mean CRP level was 6.18 mg/dl (1.02 to 14.40). The ESR was reduced after surgery, became normal in seven patients (39%) by six weeks and was normal in 12 patients (67%) at the final follow-up. The CRP became normal within five months after surgery in 17 patients (94%). The mean white blood cell count decreased by six weeks after surgery to 6361 ml (3700 to ; p < 0.01) and at the last follow-up to 6277 ml (3960 to 8370; p < 0.01). With regard to the presence or absence of stenosis and epidural abscess, there was no statistically significant difference in white blood cell count, ESR, CRP level or lordotic angle between patients with stenosis and those with an abscess (p > 0.05). There were post-operative complications in four patients, comprising a superficial wound infection in two and a transient nerve root palsy in two. As assessed by the Frankel neurological classification, 21 two patients improved by two grades (C to E), 11 improved by one grade, and five showed no change. The functional outcome as assessed using the Kirkaldy-Willis criteria 23 was excellent in five patients, good in ten and fair in three. Table V shows the differences in outcome at final follow-up in the presence or absence of stenosis and epidural abscess. They were not statistically significant. THE JOURNAL OF BONE AND JOINT SURGERY

5 POSTERIOR LUMBAR INTERBODY FUSION WITH AN AUTOGENOUS ILIAC CREST BONE GRAFT 769 Table IV. Radiological and clinical results in the 18 patients Case Lordotic angle ( ) Frankel scale 21 Pre-operative Immediate post-operative Final follow-up Pre-operative Final follow-up Post-operative complication Follow-up period (mths) Outcome (Kirkaldy-Willis 23 ) B C 25 Fair D D 48 Good D E 26 Excellent D E 36 Good E E 27 Excellent C D Superficial 36 Good infection D E Transient root 36 Good injury D E 24 Excellent E E 25 Good D E 36 Good C D Transient root 28 Fair injury C C Superficial 29 Fair infection D E 24 Good C D 36 Good C E 24 Excellent D D 31 Good D E 37 Good C E 48 Excellent Table V. Final clinical outcomes according to the presence or absence of the stenosis and epidural abscess Result Stenosis Present (n = 10) Absent (n = 8) p value Epidural abscess Present (n = 10) Absent (n = 8) p value Excellent Good Fair Discussion The incidence of pyogenic lumbar spondylodiscitis after discectomy has been reported to be 0.7% to 0.8% even when prophylactic antibiotic treatment is given. 24,25 The number of spinal procedures for either diagnostic or therapeutic purposes has increased significantly in recent years, which has increased the incidence of pyogenic lumbar spondylodiscitis. The major cause in our patients was an epidural root block carried out in those who were immunocompromised or who had comorbidities such as diabetes mellitus. A total of nine of our 18 patients had received an epidural root block before the infection developed. In the natural course of pyogenic lumbar spondylodiscitis, granulation tissue invades the disc from the subchondral tissue and absorbs disc tissue to suppress the infection. 26 This process would finally result in fusion. Pseudarthrosis would be induced when no interbody fusion occurred. Kyphotic deformity results when the disc becomes malaligned. Although it has been stated that fibrous or osseous union occurs by between six and 24 months after conservative treatment for discitis, Fredrickson et al 4 reported an incidence of osseous union of only 35% of patients. The objectives of surgery are to debride the necrotic material, decompress the neural structures, stabilise the unstable segments by a bone graft and increase the chances of faster wound healing by restoring spinal alignment. 3,9 The anterior approach is used because the infection begins in the anterior part of the spine. An abscess usually involves the anterior spine and an epidural abscess or sequestrated bone compresses the anterior dura. Following an anterior approach an anterior strut graft is used after debridement. However, an anterior interbody fusion performed without internal fixation requires long-term bed rest, a body cast or additional posterior surgery. 27,28 Although Eysel et al 29 have reported good clinical outcomes, including bony union and normal haematological findings following anterior interbody fusion with firm internal fixation, there is a risk of recurrent infection because the metal is at the site of infection. There is also an increased risk of post-operative complications, such as neurological damage, hernia, and injury to the gastrointestinal and genitourinary systems as a result of adhesions around the infected spine. It is difficult to apply this procedure to the lower lumbar spine. There are few reports on the treatment of pyogenic lumbar spondylodiscitis with posterior lumbar interbody fusion and pedicle screw fixation. 19,20 The benefits are early mobilisation because of sound internal fixation, earlier rehabilitation because of reduced morbidity and adequate removal of the abscess in the posterior and anterior dura, sequestrated bone, and any small abscesses in the anterior spine. Posterior instrumentation can be used to correct the spinal deformity, while spinal stenosis, which is usually observed in elderly patients, can be treated simultaneously. The mean age of the 18 patients we studied was 57.7 years VOL. 88-B, No. 6, JUNE 2006

6 770 J. S. LEE, K. T. SUH (44 to 71) and ten had spinal stenosis. There was an epidural abscess in ten patients; it was found in the anterior dura in seven and in the posterior dura in three. If the anterior approach is used in the presence of spinal stenosis, posterior decompression is needed as a secondary procedure. Firm fixation helps to control infection in long bones. We felt that firm internal fixation could help control the pyogenic lumbar spondylodiscitis. However, the presence of an artificial implant where inflammatory tissue is present can induce bacterial attachment and the formation of a biofilm. 30,31 By placing an implant in the cancellous bone of the vertebral body where the blood flow is good and by performing meticulous debridement of necrotic tissue before inserting the implant, the autoimmune system can treat inflammation by preventing bacterial attachment as well as the formation of a biofilm. This can be influenced by preoperative treatment with antibiotics. 29,31 Internal fixation for tuberculous spondylitis is safe 31 and the risk of a recurrence in pyogenic lumbar spondylodiscitis is not high. 29 If possible, in our patients the pedicle screw was not applied directly to the vertebral body causing the infection. When the infection and destruction of the vertebral body was severe, the chance of contact with the infection could be reduced by fixation one level above or below the segment. Compared with anterior debridement and the use of a strut bone graft, the disadvantage of posterior lumbar interbody fusion and pedicle screw fixation is that fixation may be needed for one, two, or more segments when vertebral body destruction is severe. However, we have found that this did not influence the return of our patients to an active life. This study has limitations. The number of patients included was relatively small and studies with larger numbers of patients are required. However, posterior lumbar interbody fusion with an accompanying bone graft can be considered for patients with an adjacent small abscess in the anterior spine and slight vertebral body destruction, those with an epidural abscess in the posterior dura, patients with spinal stenosis, and those with an abscess in the lower lumbar spine. Supplementary Material A further opinion by Mr Brad Williamson is available with the electronic version of this article on our website at This study was supported by Medical Research Institute Grant (2006-9), Pusan National University. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Abramovitz JN, Batson RA, Yablon JS. Vertebral osteomyelitis: the surgical management of neurologic complications. Spine 1986;11: Arnold PM, Baek PN, Bernardi RJ, Luck EA, Larson SJ. Surgical management of nontuberculous thoracic and lumbar vertebral osteomyelitis: report of 33 cases. Surg Neurol 1997;47: Eismont FJ, Bohlman HH, Soni PL, Goldberg VM, Freehafer AA. Pyogenic and fungal vertebral osteomyelitis with paralysis. J Bone Joint Surg [Am] 1983;65-A: Fredrickson B, Yuan H, Olans R. Management and outcome of pyogenic vertebral osteomyelitis. Clin Orthop 1978;131: Gepstein R, Folman Y, Lidor C, et al. Management of pyogenic vertebral osteomyelitis with spinal cord compression in the elderly. Paraplegia 1992;30: Kemp HB, Jackson JW, Jeremiah JD, Hall AJ. Pyogenic infections occurring primarily in intervertebral disc. J Bone Joint Surg [Br] 1973;55-B: Osenbach RK, Hitchon PW, Menezes AH. Diagnosis and management of pyogenic vertebral osteomyelitis in adults. Surg Neurol 1990;33: Dietze DD, Haid RW Jr. Antibiotic-impregnated methylmethacrylate in the treatment of infections with spinal instrumentation: case report and technical note. Spine 1992;17: Emery SE, Chan DP, Woodward HR. Treatment of hematogenous pyogenic vertebral osteomyelitis with anterior debridement and primary bone grafting. Spine 1989; 14: Garrido E, Rosenwasser RH. Experience with suction-irrigation technique in the management of spinal epidural infection. Neurosurgery 1983;12: Levi AD, Dickman CA, Sonntag VK. Management of postoperative infections after spinal instrumentation. J Neurosurg 1997;86: Mann S, Schutze M, Sola S, Piek J. Nonspecific pyogenic spondylodiscitis: clinical manifestations, surgical treatment, and outcome in 24 patients. Neurosurg Focus 2004;17: Graziano GP, Sidhu KS. Salvage reconstruction in acute and late sequelae from pyogenic thoracolumbar infection. J Spinal Disord 1993;6: Liebergall M, Chaimsky G, Lowe J, Robin GC, Floman Y. Pyogenic vertebral osteomyelitis with paralysis: prognosis and treatment. Clin Orthop 1991;269: Rea GL, McGregor JM, Miller CA, Miner ME. Surgical treatment of the spontaneous spinal epidural abscess. Surg Neurol 1992;37: Hadjipavlou AG, Katomis PK, Gaitanis IN, et al. Percutaneous transpedicular discectomy and drainage in pyogenic spondylodiscitis. Eur Spine J 2004;13: Hadjipavlou AG, Mader JT, Necessary JT, Muffoletto AJ. Hematogenous pyogenic spinal infections and their surgical management. Spine 2000;25: Thalgott JS, Cotler HB, Sasso RC, LaRocca H, Gardner V. Postoperative infections in spinal implants: classification and analysis. Spine 1991;16: Przybylski GJ, Sharan AD. Single-stage autogenous bone grafting and internal fixation in the surgical management of pyogenic discitis and vertebral osteomyelitis. J Neurosurg 2001;94: Rath SA, Neff U, Schneider O, Richter HP. Neurosurgical management of thoracic and lumbar vertebral osteomyelitis and discitis in adults: a review of 43 consecutive surgically treated patients. Neurosurgery 1996;38: Frankel HL, Hancock DO, Hyslop G, et al. The value of postural reduction in the initial management of the closed injuries of the spine with paraplegia and tetraplegia. Paraplegia 1969;7: Lee CK, Vessa P, Lee JK. Chronic disabling low back pain syndrome caused by internal disc derangement: the results of disc excision and posterior lumbar interbody fusion. Spine 1995;20: Kirkaldy-Willis WH, Paine KW, Cauchoix J, McIvor G. Lumbar spinal stenosis. Clin Orthop 1974;99: Leung PC. Complication in the first 40 cases of microdiscectomy. J Spinal Disord 1988;1: Lindholm TS, Pylkkanen P. Discitis following removal of intervertebral disc. Spine 1982;7: Fraser RD, Osti OL, Vernon-Roberts B. Iatrogenic discitis: the role of intravenous antibiotics in prevention and treatment. Spine 1989;14: 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 2003;28: Ha KY, Kim YH. Postoperative spondylitis after posterior lumbar interbody fusion using cages. Eur Spine J 2004;13: Eysel P, Hopf C, Vogel I, Rompe JD. Primary stable anterior instrumentation or dorsoventral spondylodesis in spondylodiscitis?: results of a comparative study. Eur Spine J 1997;6: Costerton JW, Irvin RT, Cheng KJ. The bacterial glycocalyx in nature and disease. Ann Rev Microbiol 1981;35: Oga M, Arizono T, Takasita M, Sugioka Y. Evaluation of the risk of instrumentation as a foreign body in spinal tuberculosis: clinical and biological study. Spine 1993; 18: THE JOURNAL OF BONE AND JOINT SURGERY