Preoperative and postoperative neurologic status was assessed according to the American Spinal Cord Injury Association modified Frankel Impairment Scale.27 As radiographic determination of fusion can be difficult with anterior thoracolumbar instrumentation, a construct was deemed stable in the absence of motion in flexion-extension films, lack of significant radiolucency at the interbody graft-vertebral body junction, and no evidence of interval change in angulation in a >1-year period.16,28 Average patient radiographic and clinical follow-up was 31.1 months (range 9-50 months).
Analysis of variance and Tukey post-hoc tests tested differences across patient demographics, mechanism of injury, AO classification injury types, reconstruction methods, and type of anterior spinal instrumentation. Matched preoperative and postoperative measurements were analyzed with paired t tests. Differences in arthrodesis probability among injury types and instrumentation/reconstruction methods were analyzed with a likelihood ratio test. Statistical analysis was performed with SAS (for Windows, version 8.00; SAS Institute, Cary, NC).
No patients deteriorated neurologically as a result of operative treatment. Thirty of 33 patients (91%) with incomplete injuries improved at least one modified Frankel grade (range one to three grades). Three of four patients categorized as modified Frankel A demonstrated some improvement, whereas all three modified Frankel E neurologically intact patients (7.5%) remained unchanged (see Table 1).
Preoperative canal compromise, based on preoperative CT scan, averaged 68.5% (range 32-100%). Average vertebral body height loss at the injured level was 44.5% (range 30-60%). Mean preoperative segmental kyphosis measured 22.7° (range 10-42°, SD 8.3°) with significant (P < 0.0001) early postoperative correction to 7.4° (range 0-28°, SD 7.4°). At latest radiographic follow-up, angulation was significantly changed but only by an average 2.1° (range −6-7°) to 9.3° (range 0-28°, SD 8.2°) (P < 0.001) (Figure 4). However, latest postoperative sagittal angulation remained significantly different from preoperative angulation (P < 0.0001). One patient with a type C2.1 injury (type B1 flexion-distraction injury with posterior ligamentous disruption and associated rotational component) experienced early failure due to a technical error with a malpositioned derotational screw, allowing recurrence of translational and rotational deformity. This required an early posterior thoracolumbar instrumentation for salvage, which went on to uneventful healing. Aside from this patient, 37 of 39 (95%) patients demonstrated apparently stable constructs by radiographic evaluation at latest follow-up. Two patients had evidence of pseudarthrosis and subsequently underwent successful posterior thoracolumbar arthrodesis with instrumentation; both had type B2.3 injuries (flexion-distraction injury with type A vertebral body burst fracture and posterior element fractures through pars interarticularis) reconstructed with structural iliac graft and University plate instrumentation. There was no statistically significant difference in the apparent arthrodesis rates at 95% confidence interval likelihood testing among the different AO injury classes (P = 0.21) or anterior thoracolumbar implant types and reconstruction methods (P = 0.18).
Three of 40 patients (8%) required either early or delayed supplemental posterior thoracolumbar arthrodesis with instrumentation. There was no progression of initial neurologic deficit in these or any of the other 37 patients. There were no intraoperative or late vascular injuries. Three patients demonstrated radiographic evidence of minor screw loosening that did not progress; no cases required removal of anterior thoracolumbar implants. Two patients who underwent thoracolumbar approaches (5%) developed low thoracic dermatomal pain from intercostal neuralgia that was improved after a series of intercostal nerve blocks. Perioperative complications included two cases of pneumonia/atelectasis, one urinary tract infection, and one superficial wound infection that was successfully treated with antibiotics.
The anterolateral approach allows direct decompression of ventral osseous and soft tissue pathology, offering superior canal clearance as compared with posterior indirect (ligamentotaxis) and posterolateral decompression techniques.10,11,22,24,25,29-31 Although some have reported that this improved anterior decompression results in better neurologic recovery as compared with the posterior management of thoracolumbar fractures,1,11,12,22 others found no significant difference.31,32 Initial uninstrumented reconstruction methods with simple anterior strut grafting resulted in unacceptably high rates of pseudarthrosis, ranging from 10% to 100%.1,9,10,33 Therefore, two-stage anterior decompression and posterior instrumentation were recommended.1,10,14,17,20,22,34
Early anterior thoracolumbar instrumentation, developed by Dwyer, Hall, and Zielke,1,33 was used in the correction and stabilization of scoliosis. However, these devices were biomechanically insufficient in the setting of unstable thoracolumbar injuries.10,11,35,36 More recently, anterior thoracolumbar instrumentation has significantly evolved, greatly increasing its utility in treating thoracolumbar trauma.17,36-38 Current devices allow both distraction and compression, with greater deformity correction and improved load-sharing ability.1,17,36,38,39
Numerous biomechanical studies have demonstrated the stability provided by these newer, more rigid designs.36,40,41 An et al17 evaluated the biomechanical characteristics of four different types of anterior thoracolumbar instrumentation (three of which were the Kaneda device, University plate, and Z-plate, used in the authors' current study) in a calf spine model with anterior and middle column defects. All showed significant stabilizing effects, and all restored axial rotation stability. In another calf spine model, Gurr et al40 compared the mechanical stiffness of an anterior construct (Kaneda device) with posterior pedicle screws, concluding that posterior systems spanning five levels produced results similar to the Kaneda device spanning three levels.
As a result of these biomechanical improvements in anterior thoracolumbar instrumentation,17 several investigators began using these constructs in the single-stage management of acute or late thoracolumbar injuries.1,2,11,12,15,28,42 Kostuick11 reported on a series of 49 patients with “burst injuries,” 25 of whom underwent early or delayed anterior decompression and stand-alone instrumentation. Clinical results were very good, with an average 1.6 Frankel grade neurologic improvement and, among this subgroup, no pseudarthroses. Kaneda et al15 initially reported on their early results in the anterior decompression and Kaneda device stabilization of 110 patients with thoracolumbar burst fractures with neurologic deficits. Thirteen years later, they included longer-term follow-up (mean 8 years) on 150 such patients, the largest published series to date.2 Reportedly, these were all burst fractures according to the Denis classification, with a preoperative mean canal stenosis of 47% (improved to 2% postoperative mean) and mean kyphotic deformity of 19° (corrected to 7° postoperatively, without significant loss at latest follow-up). Ninety-five percent improved neurologically by at least one Frankel grade. Schnee and Ansell16 reported on 25 patients with thoracolumbar burst fractures, 15 of which they considered three-column injuries. Although not specifically delineated, at least four of these three-column injury patients underwent successful stand-alone anterior decompression and reconstruction. In their series, preoperative mean stenosis was 48.3% and mean kyphosis 16.8°, corrected to 2.9° postoperatively. Sixteen of 17 patients with neurologic deficit improved.
With few exceptions,13 the majority of studies describing stand-alone anterior treatment of thoracolumbar injuries have been in reference to the burst fracture groups of the Denis and McAfee classifications. These represent two-column (anterior and middle) injuries, corresponding to the type A3 injury of the AO classification. Although some authors label some of these injuries “three column” due to a posterior element injury in the form of an associated longitudinal laminar split fracture, this does not result in inherent insufficiency of the posterior ligamentous complex.3,16,19 Thus, these fractures would not be considered type B or C injuries in the AO classification system. However, it is certainly possible that some of the series reporting stand-alone anterior thoracolumbar treatment of two-column Denis burst fractures did include subtle type B1.2 or C1.3 AO classification injuries (true three-column injuries).
The authors' current study, reviewing the results of single-stage, stand-alone anterior surgical management of some types of three-column (type B and C) thoracolumbar injuries, demonstrates findings not unlike those previously reported for such treatment of theoretically more stable two-column injuries. Ninety-one percent of the patients with an incomplete neurologic injury developed neurologic improvement of at least one modified Frankel grade; none experienced neurologic deterioration. Aside from one early failure due to a technical error, 37 of 39 (94%) patients appeared to have stable constructs at latest follow-up, similar to results from Kaneda et al.2,15 Two patients who developed known pseudarthrosis underwent successful augmentation with instrumented posterior arthrodesis. Although both occurred in type B2.3.2 injuries (posterior osseous flexion-distraction injury through pars interarticularis with type A anterior vertebral body fracture), this injury subgroup was not significantly different from other type B or C groups (P = 0.21). Significant angulation improvement was achieved postoperatively with a mean preoperative 22.7° kyphosis correcting to a mean postoperative 7.4°. Although some kyphosis returned at latest follow-up, this represented only an average 2.1° loss of correction, with overall sagittal angulation still remaining significantly improved from preoperative angulation (P < 0.0001).
Several investigators have reported that anterior management of acute thoracolumbar injuries allows improved kyphosis correction (and maintenance of that correction), as compared with posterior instrumentation.1,4,8,26,43 This is a result of the restoration of anterior column load bearing, which can be achieved with anterior reconstruction methods, placing the interbody graft material in a biomechanically optimal environment of compression. Alanay et al5 described a technique that replicates this anterior column restoration through posterior intracorporeal transpedicular grafting to prevent angulation and failure of short-segment pedicle instrumentation. Twenty consecutive patients were prospectively randomized into short-segment posterior instrumentation with or without intracorporeal transpedicular grafting. There was no significant difference between the two groups, each with a 40-50% failure rate of >10° correction loss and 10% hardware breakage.5 The importance of maintenance of correction is controversial.19 Malcom et al,44 as well as others,11 have concluded that compensatory hyperlordosis below a kyphotic segment resulted in increased posttraumatic back and buttock pain. Others have found no correlation between clinical outcome and residual kyphosis.19
Additionally, stand-alone anterior thoracolumbar treatment of unstable injuries allows short-segment constructs to be used, saving motion segments (particularly in comparison with long-segment posterior constructs, the traditional treatment method for these types of injuries).17,24,39,41 In this study of unstable three-column injuries, a single-level corpectomy reconstruction with two-segment instrumented arthrodesis resulted in relatively high stability. Short-segment posterior constructs can accomplish the same goal; however, they have been associated with reportedly high failure rates, ranging from 10% to 50%.4,5,8,16,43,45 Subsequently, McCormack and co-workers43 have described a “load-sharing” classification to identify which unstable thoracolumbar fractures are likely to have poor anterior load-bearing capabilities (such as some AO type B1.2, B2.3, and C injuries), resulting in loss of kyphosis correction and posterior instrumentation failure. In those instances, they recommend either a long-segment posterior instrumented arthrodesis or two-stage anterior and posterior procedures.4,43
The authors' current study is limited by its retrospective nature and relatively small population, which restricts the conclusions drawn from reconstruction subgroup comparisons (structural allograft versus structural autograft versus titanium-mesh cage packed with local autograft). Similarly, although all three cases requiring supplemental posterior instrumentation occurred with University plate instrumentation, this was not statistically significant (P = 0.15). Whereas most current anterior thoracolumbar implants perform acceptably well in biomechanical testing, some instrumentation (such as Kaneda-like linked dual-rod constructs) provides significantly greater rigidity than some screw and plate designs (such as the University and Z-plates),19,36 particularly in torsion. Despite this experimental biomechanical advantage, all implants yielded equally acceptable clinical results, pointing toward the important role of meticulous attention to surgical technique in reconstruction of an appropriate load-sharing environment (helping to decrease instrumentation failure rates).36,38
Finally, the authors certainly do not advocate single-stage stand-alone anterior management of all three-column thoracolumbar injuries. No attempt was made to treat any type C3 (rotational shear) or most type C2 injuries with this approach, owing to the significant instability associated with such injury patterns. In these circumstances, this stand-alone construct would not provide an adequate degree of stability.
The efficacy of stand-alone single-stage anterior decompression and reconstruction of unstable three-column thoracolumbar injuries, utilizing current-generation anterior spinal instrumentation, was studied. Modern anterior spinal instrumentation and reconstruction techniques can allow some types of unstable three-column thoracolumbar injuries to be treated in an anterior stand-alone fashion similar to those previously reported for stable burst fractures without posterior column disruption. Sagittal alignment was restored and maintained, 91% of patients developed neurologic improvement of at least one modified Frankel grade, and 94% of patients appeared to have stable constructs. The advantages of this anterior stand-alone technique in unstable AO B-type fractures are allowing direct anterior decompression of neural elements, improvement in segmental angulation, and acceptable rates of arthrodesis without the need for supplemental posterior instrumentation. Caution should be used with extremely unstable C-type fractures with rotational, translational, and shearing instabilities. These fractures and fracture-dislocations should first undergo posterior stabilization.
1. Ghanayem AJ, Zdeblick TA. Anterior instrumentation in the management of thoracolumbar burst fractures. Clin Orthop
2. Kaneda K, Taneichi H, Abumi K, et al. Anterior decompression and stabilization with the Kaneda device for thoracolumbar burst fractures associated with neurological deficits. J Bone Joint Surg Am
3. Magerl F, Aebi M, Gertzbein SD, et al. A comprehensive classification of thoracic and lumbar injuries. Eur Spine J
4. Parker JW, Lane JR, Karnikovic EE, et al. Successful short-segment instrumentation and fusion for thoracolumbar spine fractures. Spine
5. Alanay A, Acaroglu E, Yazici M, et al. Short-segment pedicle instrumentation of thoracolumbar burst fractures. Does transpedicular intracorporeal grafting prevent early failure? Spine
6. Criscitiello AA, Fredrickson BE. Thoracolumbar spine injuries. Orthopedics
7. Harrington RM, Budorick T, Hoyt J, et al. Biomechanics of indirect reduction of bone retropulsed into the spinal canal in vertebral fracture. Spine
8. Sasso RC, Cotler HB. Posterior instrumentation and fusion for unstable fractures and fracture dislocations of the thoracic and lumbar spine. Spine
9. Hitchon PW, Al Jurf A, Kernstine K, et al. Management options in thoracolumbar fractures. Contemp Neurosurg
10. Dunn HK. Anterior stabilization of thoracolumbar injuries. Clin Orthop
11. Kostuik JP. Anterior fixation for fractures of the thoracic and lumbar spine with or without neurologic involvement. Clin Orthop
12. McAfee PC, Bohlman HH, Yuan HA. Anterior decompression of traumatic thoracolumbar fractures with incomplete neurological deficit using a retroperitoneal approach. J Bone Joint Surg Am
13. McGuire R. The role of anterior surgery in the treatment of thoracolumbar fractures. Orthopedics
14. Dimar JR, Wilde PH, Glassman SD, et al. Thoracolumbar burst fractures treated with combined anterior and posterior surgery. Am J Orthop
15. Kaneda K, Abumi K, Fujiya M. Burst fractures with neurologic deficits of the thoracolumbar spine. Results of anterior decompression and stabilization with anterior instrumentation. Spine
16. Schnee CL, Ansell LV. Selection criteria and outcome of operative approaches for thoracolumbar burst fractures with and without neurologic deficit. J Neurosurg
17. An HS, Lim TH, You JW, et al. Biomechanical evaluation of anterior thoracolumbar spinal instrumentation
18. Miyakoshi N, Abe E, Shimada Y, et al. Anterior decompression with single segment spinal interbody fusion for lumbar burst fracture. Spine
19. Okuyama K, Abe E, Chiba M, et al. Outcome of anterior decompression and stabilization for thoracolumbar unstable burst fractures in the absence of neurologic deficits. Spine
20. Denis F. Spinal instability
as defined by the three-column spine concept in acute spine trauma. Clin Orthop
21. McAfee PC, Yuan HA, Fredrickson BE, et al. The value of computed tomography in thoracolumbar fractures: an analysis of one hundred consecutive cases and a new classification. J Bone Joint Surg Am
22. Bradford DS, McBride GG. Surgical management of thoracolumbar spine fractures with incomplete neurologic deficits. Clin Orthop
23. Chapman JR, Anderson PA. Thoracolumbar spine fractures with neurologic deficit. Orthop Clin North Am
24. Shono Y, McAfee PC, Cunningham BW. Experimental study of thoracolumbar burst fractures. A radiographic and biomechanical analysis of anterior and posterior instrumentation systems. Spine
25. Harris MB. The role of anterior stabilization with instrumentation in the treatment of thoracolumbar burst fractures. Orthopedics
26. Hashimoto T, Kaneda K, Abumi K. Relationship between traumatic spinal canal stenosis and neurologic deficit in thoracolumbar burst fractures. Spine
27. American Spinal Injury Association. Standards for neurologic and functional classification of spinal cord injury. Chicago: ASIA; 1992.
28. Aydin E, Solak AS, Tuzuner MM, et al. Z-Plate instrumentation in thoracolumbar spinal fractures. Bull Hosp Joint Dis
29. Hodgson AR, Stock FE. Anterior spinal fusion: a preliminary communication on the radical treatment of Pott's paraplegia. Br J Surg
30. Brightman RP, Miller CA, Ren GL, et al. Magnetic resonance imaging of trauma to the thoracic and lumbar spine. The importance of the posterior longitudinal ligament. Spine
31. Esses SI, Botsford DJ, Kostuik JP. Evaluation of surgical treatment for burst fractures. Spine
32. Gertzbein SD. Scoliosis Research Society: multicenter spine fracture study. Spine
33. Zdeblick TA, Shirado O, McAfee PC, et al. Anterior spinal fixation after lumbar corpectomy. J Bone Joint Surg Am
34. Mann KA, McGowan DP, Fredrickson BE, et al. A biomechanical investigation of short spinal fixation for burst fractures with varying degrees of posterior disruption. Spine
35. Been HD. Anterior decompression and stabilization of thoracolumbar burst fractures by the use of the Slot-Zielke device. Spine
36. Zdeblick TA, Warden KE, Zou D, et al. Anterior spinal fixators: a biomechanical in vitro study. Spine
37. Jenis LG, An HS. Anterior thoracolumbar instrumentation for tumor or trauma. Semin Spine Surg
38. Kotani Y, Cunningham BW, Parker LM, et al. Static and fatigue biomechanical properties of anterior thoracolumbar instrumentation systems. Spine
39. Hitchon PW, Goel VK, Rogge T, et al. Biomechanical studies on two anterior thoracolumbar implants in cadaveric spines. Spine
40. Gurr KR, McAfee PC, Shih C. Biomedical analysis of anterior and posterior instrumentation systems after corpectomy. J Bone Joint Surg Am
41. Lim TH, An HS, Hong JH, et al. Biomechanical evaluation of anterior and posterior fixations in an unstable calf spine model. Spine
42. Kostuik JP. Anterior spinal cord decompression for lesions of the thoracic and lumbar spine. Spine
43. McCormack T, Karaikovic E, Gaines RW. The load sharing classification of spine fractures. Spine
44. Malcolm BW, Bradford DS, Winter RB, et al. Post-traumatic kyphosis: a review of 48 surgically treated patients. J Bone Joint Surg Am
45. McLain RF, Sparling E, Benson DR. Early failure or short-segment pedicle instrumentation for thoracolumbar fractures. J Bone Joint Surg Am
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
thoracolumbar fracture; instability; spinal instrumentation; anterior plates