- Email: [email protected]
Unstable upper and middle thoracic fractures. Preliminary experience with a posterior transpedicular correction-fixation technique
Journal of Clinical Neuroscience (2005) 12(5), 529–533 0967-5868/$ - see front matter ª 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2004.11.006
Clinical study
Unstable upper and middle thoracic fractures. Preliminary experience with a posterior transpedicular correction-fixation technique M. Payer MD Department of Neurosurgery, University Hospital of Geneva, Geneva, Switzerland
Summary A number of conservative and operative approaches have been described for the treatment of unstable traumatic upper and middle thoracic fractures. The advantage of surgical correction and fixation/fusion lies in its potential to restore sagittal and coronal alignment, thereby indirectly decompressing the spinal cord. A consecutive series of 8 patients with unstable traumatic upper and middle thoracic fractures is reviewed. In all patients, polyaxial pedicle screws were inserted bilaterally into the two levels above and below the fracture. Rods that were less contoured (‘‘undercontoured”) than the regional hyperkyphosis at the injured level, were anchored to the caudal four screws. The cranial four screws, with the vertebrae to which they were inserted, were then progressively pulled posteriorly onto the undercontoured rods with rod reducers, thus correcting the hyperkyphosis and anterolisthesis. The mean follow-up was 15 months. The mean regional kyphosis was 23 preoperatively, 17 postoperatively and 18 at follow-up. The mean anterolisthesis was 8 mm preoperatively, 1 mm postoperatively and 1 mm at follow-up. No hardware failure occurred. Five patients with complete spinal cord injury at presentation made no neurological recovery, two patients with incomplete spinal cord injury initially (ASIA B), recovered substantially (to ASIA D), and the patients who were neurologically intact at presentation remained so. ª 2005 Elsevier Ltd. All rights reserved. Keywords: spine trauma, thoracic fracture, pedicle screw fixation
INTRODUCTION The treatment of unstable upper and middle thoracic fractures (T1-10) is controversial and rarely addressed exclusively in the literature.1,2 The advantages of surgical stabilization are immediate restoration of the sagittal and coronal alignment in kyphotic fractures and the correction of an anterolisthesis in subluxation fractures, thereby leading to indirect decompression of the spinal canal. Fixation and fusion furthermore can prevent progressive hyperkyphosis, which can result in local pain, neurological deterioration, and the psychological consequences of a ``hunchback” deformity. Posterior thoracic fixation can be performed by means of pedicle screws or laminar hooks. Thoracic pedicle screw fixation can be technically challenging, as anatomical work has shown a smaller pedicle diameter in the upper and especially middle thoracic spine than in the lower thoracic spine.3–6 Additionally, biomechanical studies have found much stronger pull-out forces with thoracic pedicle screws than with laminar hooks.7,8 The current study was performed to evaluate correction and fixation/fusion in 8 consecutive patients with unstable upper and middle thoracic fractures, using polyaxial pedicle screws two levels above and below the fracture and ‘‘undercontoured” rods. MATERIALS AND METHODS Eight consecutive patients (7 men and 1 woman) with a mean age of 40 years (range 20–65) with traumatic unstable upper and middle thoracic fractures were operated by the author over a 2-year Received 2 July 2004 Accepted 30 November 2004 Correspondence to: Dr Michael Payer, Department of Neurosurgery, University Hospital of Geneva, Rue Micheli-du-Crest 24, 1211 Geneva 14, Switzerland. Tel.: +41 22 372 82 24; Fax: +41 22 372 82 20; E-mail: [email protected]
period (Table 1). Patient hospital charts and radiological examinations were reviewed from the hospital admission, and all patients were clinically and radiologically reviewed in the outpatients clinic by the author. The average follow-up was 15 months (10– 19 months). Preoperative neurological injury was assessed according to the ASIA classification9 (Table 2). Initial spinal cord injury was complete (ASIA A) in five patients, incomplete (ASIA B) in two patients, and no spinal cord injury was present in one patient (ASIA E). Plain antero-posterior and lateral radiographs and computed tomography (CT) scan of the thoracic spine with reconstruction revealed a type B injury in four patients and a type C lesion in four patients according to the Magerl classification10 (Table 3).
Surgical technique After hemodynamic and pulmonary stabilization of the thoracic trauma that was always concomitant, and other abdominal or peripheral injuries, surgery was performed at a mean of three days (range 1–6) after the accident, under general anesthesia with the patient in the prone position. Gel cushions were placed under the upper thorax and the iliac crests bilaterally, freeing the abdomen. After midline incision and exposure of the spine, pedicle screws were inserted bilaterally into the two levels below and two levels above the fracture. Anteroposterior and lateral fluoroscopy was used to control the entry point, which was positioned at the junction of the proximal edge of the transverse process with the lamina.11 Using the funnel technique12 with a high speed drill, a starting hole 30–40 mm in depth was created and palpated with a probe to ensure bony surroundings. Under fluoroscopic control, polyaxial top-loading self-tapping titanium screws (CD M8, Medtronics, Memphis, TN), 4.5 mm in diameter and 30–40 mm in length, were inserted. Screw insertion was performed in a slightly converging inclination in the middle thoracic spine and a more converging angle in the upper thoracic spine according to the 529
530 Payer
Table 1 Patient summary Patient
Injury
Deformity Pre-op
Neurological Post-op
Operation
Deformity Post-op
SD 21M
T3/4 Type B
33 kyphosis 9 mm ant-listh
ASIA A
T1-5
19 kyphosis 2 mm ant-listh
EC 44M
T4/5 Type B
26 kyphosis 7 mm ant-listh
ASIA A
T2-7
LZ 65F
T4/5 Type B
29 kyphosis 7 mm ant-listh
ASIA A
SF 47M
T4 Type B
21 kyphosis 0 mm ant-listh
AJ 43M
T4/5 Type C
LK 44M
Follow-up
Deformity Follow-up
Neurological Follow-up
Pain Follow-up
19 m
20 kyphosis 2 mm ant-listh
ASIA A
2/10
15 kyphosis 0 mm ant-listh
14 m
15 kyphosis 0 mm ant-listh
ASIA A
3/10
T2-7
2 kyphosis 2 mm ant-listh
12 m
24 kyphosis 2 mm ant-listh
ASIA A
2/10
ASIA E
T2-6
13 kyphosis 0 mm ant-listh
10 m
14 kyphosis 0 mm ant-listh
ASIA E
2/10
23 kyphosis 32 scoliosis 5 mm ant-listh
ASIA B
T2-7
20 kyphosis 11 scoliosis 0 mm ant-listh
17 m
22 kyphosis 10 scoliosis 0 mm ant-listh
ASIA D
0/10
T5/6 Type C
21 kyphosis 11 mm ant-listh
ASIA B
T2-8
10 kyphosis 0 mm ant-listh
15 m
10 kyphosis 0 mm ant-listh
ASIA D
2/10
SS 36M
T4 Type C
33 kyphosis 9 mm ant-listh
ASIA A
T2-6
20 kyphosis 1 mm ant-listh
15 m
22 kyphosis 1 mm ant-listh
ASIA A
3/10
LK 20M
T2/3 Type C
19 kyphosis 16 mm ant-listh
ASIA A
C7-T5
15 kyphosis 0 mm ant-listh
14 m
15 kyphosis 1 mm ant-listh
ASIA A
2/10
Indicates patient initials, age and sex; injury level and type; preoperative deformity (ant-listh = anterolisthesis) and neurological state; level of surgery; follow-up period in months; deformity, neurological state and thoracic back pain (on a visual analog scale 0–10) at follow-up.
Table 2 Standard neurological classification of spinal cord injury according to the American Spinal Injury Association, ‘‘ASIA impairment scale’’9 A B C D E
Complete spinal cord injury: No motor or sensory function preserved in the sacral segments S4-S5 Incomplete spinal cord injury: Sensory but not motor function preserved below the neurological level and includes the sacral segments S4-S5 Incomplete spinal cord injury: Motor function preserved below the neurological level, and more than half of the key muscles below the neurological level have a muscle grade less than 3 Incomplete spinal cord injury: Motor function preserved below the neurological level, and at least half of the key muscles below the neurological level have a muscle grade of 3 or more Normal: Motor and sensory function is normal
individual pedicle anatomy. The screw trajectory was chosen parallel to the superior endplate. Titanium rods 5.5 mm in diameter were contoured to physiological lordosis, that is ``undercontoured” in respect to the regional fracture hyperkyphosis, and anchored in the caudal four screws. The cranial four screws with the attached vertebrae were then progressively pulled posteriorly unto the undercontoured rods with rod reducers, thus correcting the hyperkyphosis and anterolisthesis (Figs. 1–3). If there was scoliosis, distraction on the concave side was additionally performed. Finally, the laminae and spinous processes of the instrumented and fractured vertebrae were decorticated; the tips of the spinous
Fig. 2 Postoperative sagittal CT reconstruction of the patient in Figure 1, showing complete reduction of the anterolisthesis and correction of the kyphosis.
Fig. 1 Preoperative sagittal CT reconstruction, demonstrating a type B lesion of T 4/5 with 26 of regional kyphosis and 7 mm anterolisthesis.
Journal of Clinical Neuroscience (2005) 12(5), 529–533
processes of the instrumented and fractured vertebrae were resected and used as a local autograft, augmented by allograft bone chips, and applied posteromedially. Thus, no iliac crest autograft was taken, avoiding a second incision and donor site morbidity. No external immobilisation was applied and the patients were immediately mobilised as much as their concomitant lesions and neurological state allowed. Postoperative plain radiographs and CT scan with sagittal and coronal reconstruction were performed to evaluate deformity correction and screw placement. Regional kyphosis was determined by the angle of Cobb13 between the superior endplate of the first ª 2005 Elsevier Ltd. All rights reserved.
Correction-fixation of unstable thoracic features 531
Fig. 3 Plain antero-posterior (a) and lateral (b) radiographs of the patient in Figure 1 and 2, at follow-up 14 months after T2-7 reduction, fixation and fusion.
intact vertebra above the fracture and the inferior endplate of the first intact vertebra below the fracture. Posttraumatic scoliosis was determined as the angle between the superior endplate of the first intact vertebra above the fracture and the inferior endplate of the first intact vertebra below the fracture. Anterolisthesis was measured in millimeters as the anterior translation of the posterior vertebral body wall of a fractured vertebra in respect to the subjacent vertebra. At follow-up, back pain was assessed using a visual analogue scale (VAS, minimum pain = 0, maximum pain = 10), neurological injury was evaluated according to the ASIA score, and sagittal and coronal alignment, as well as implant position and integrity, were evaluated on plain antero-posterior and lateral thoracic spine radiographs (Fig. 3). RESULTS Mean surgical duration was 211 minutes (range 182–245), mean intraoperative blood loss 630 cc (range 420–860 cc). There were no surgical complications. Sixty-two thoracic and two C7 pedicle screws were inserted in total. Postoperative radiological analysis showed 56 thoracic and both C7 pedicle screws to be correctly placed through the pedicles into the vertebral bodies. Five thoracic pedicle screws were found within the pedicles, but were aimed too cranially and crossed into the disc space. One thoracic pedicle screw was positioned lateral to the pedicle, through the transverse process into the vertebral body. No pedicle screw was in the spinal canal. At follow-up no fracture or displacement of screws or rods were found nor any radiolucency around the pedicle screws. No hardware removal or other second surgery in the thoracic spine was performed in any patient. The mean regional kyphosis was 23 preoperatively (range 11– 33), 17 postoperatively (range 10–22), and 18 at follow-up (range 10–24). Mean anterolisthesis was 8 mm preoperatively (range 0–16 mm), 1 mm postoperatively (range 0–2 mm) and 1 mm at follow-up (range 0–2 mm). Preoperative scoliosis was ª 2005 Elsevier Ltd. All rights reserved.
found in one patient with 32, corrected to 10 postoperatively. Mean surgical kyphosis correction was 6 immediately postoperatively and 5 at follow-up, average translation correction 7 mm postoperatively and at follow-up. Neurologically, all five patients with complete initial spinal cord injury remained unchanged at follow-up (ASIA A), the two patients with initially incomplete lesions (ASIA B) recovered substantially to ASIA D, and the initially neurologically intact patient remained intact. Thoracic back pain according to the VAS at follow-up was between 0 and 3 in all patients with a mean value of 2. No patient was taking regular pain medication for back pain. No patient was limited in his daily activities because of back pain and all patients except for one patient (LZ, Table 1) were living independently at follow-up. This patient remained in the rehabilitation center for the remainder of her life due to severe depression, obesity, recurrent urinary and pulmonary infections, and finally died after pulmonary embolism. DISCUSSION Traumatic unstable upper and middle thoracic fractures are rare. There is no consensus regarding optimal treatment and various conservative and operative options have been described in the literature with different inclusion criteria, follow-ups and evaluation tools.14–22 There are no prospective randomised or controlled studies. A surgical fixation procedure for an unstable upper or middle thoracic fracture should be safe, correct the frequent kyphotic deformity and anterolisthesis and thereby decompress the spinal cord and maintain the deformity correction long-term.2,21,22 The above described technique meets these criteria. A posterior midline approach is less invasive than the posterolateral or anterior approach to the thoracic spine. Regional hyperkyphosis and anterolisthesis could be safely and reliably corrected by anchoring undercontoured rods in a block of four pedicle screws below the fracture, then pulling the kyphotic or Journal of Clinical Neuroscience (2005) 12(5), 529–533
532 Payer Table 3 Magerl’s comprehensive classification of thoracic and lumbar injuries10 Type A Injuries
Vertebral body compression
A1 A2 A3
Impaction fractures Split fractures Burst fractures
Type B Injuries
Anterior and posterior element injury with distraction
B1 B2 B3
Posterior disruption, predominantly ligamentous (flexion-distraction injury) Posterior disruption, predominantly osseous (flexion-distraction injury) Anterior disruption through the disc (hyperextension-shear injury)
Type C Injuries
Anterior and posterior element injury with rotation
C1 C2 C3
Type A injuries with rotation Type B injuries with rotation Rotational-shear injuries
anteriorly translated fractured spine with the block of four pedicle screws above the fracture posteriorly unto the undercontoured rods by means of rod reducers. Finally, deformity correction was maintained in all cases throughout the observation period. Pedicle screw insertion in the upper and middle thoracic spine is challenging due to the specific osseous morphology, as has been shown in various reports.3–6 However, with exact anatomical knowledge and meticulous technique, thoracic pedicle screws can be used safely.23 In this series, the recently described funnel technique12 was found to be helpful. Lateral fluoroscopy of the upper thoracic spine is often of poor quality, but usually sufficient to position the pedicle screw parallel to the superior endplates. This screw path has recently been biomechanically analysed to have a significantly higher pullout strength than a cephalo-caudad pedicle screw along the anatomical axis of the pedicle.24 Anteroposterior fluoroscopy was reliable in finding the pedicle eyes, thus confirming the entry point of the pedicles. Care must be taken to install the fluoroscopy tube perpendicular to the instrumented vertebra by compensating for the local kyphosis in order to obtain a true anteroposterior image. In the current series, a pedicle screw diameter of 4.5 mm and a length between 30 and 40 mm seemed biomechanically sufficient, while being small enough to be accommodated by the upper and middle thoracic pedicles. According to a recent study, thoracic pedicle screws can even be placed outside very narrow thoracic pedicles, using a trajectory through the transverse process into the vertebral body, a technique recently described as “extrapedicular placement of screws in the thoracic spine”.25 Biomechanical work showed a 75% peak insertional torque for thoracic pedicle screws at the neurocentral junction, which was measured 15 to 20 mm from the dorsal cortex of the pedicle.24 This may explain the sufficient biomechanical strength even in suboptimal screw placements, provided the screw traverses the neurocentral junction. The increased pull-out strength of thoracic pedicle screws as compared to laminar hooks, pedicle hooks or transverse process screws7,8 allowed safe and reliable correction of the frequently found kyphotic deformity or anterolisthesis in the unstable upper and middle thoracic fractures in this series. Thus, a supplementary anterior approach in these three-column injuries was deemed unnecessary. A solid fusion in the thoracic spine is difficult to assess, but the absence of thoracic back pain in daily living in combination with maintained sagittal and coronal deformity correction at the followup makes fusion very probable in the patients in the current series. Substantial neurological recovery was observed in both patients with an initial incomplete spinal cord injury with ASIA B score, improving to ASIA D at follow-up. These two patients with 5 and 11 mm anterolisthesis respectively had bilateral pedicle shear fractures, which protected the spinal cord from severe compression. This finding is in accordance with a recent report, showing bilateral pedicle shear fractures to be present in all 15 published thoracic fracture dislocations without neurological deficit.26 Journal of Clinical Neuroscience (2005) 12(5), 529–533
To the author’s knowledge there are only two other reports on pedicle screw fixation of upper or middle thoracic fractures in the English literature.1,2 Yue et al., in their series, also performed transpedicular fixation two levels above and below the fracture.2 They used varying screw diameters from 4.2 to 7 mm according to the pedicle size. In contrast to the current series, they anchored undercontoured rods in the cranial vertebrae first and pulled the caudal vertebrae onto undercontoured rods in order to correct kyphotic deformity. This series included 16 patients with fractures between T1 and T10. At a median follow-up of 22 months they observed no hardware failure, no loss of correction, no painful hardware requiring removal and no postoperative neurological worsening. Schweighofer et al. briefly described a series of 6 patients with Magerl type B or C lesions in the upper and middle thoracic spine, operated using a posterior approach with Louis or Steffee plates.1 The technique is not described. One patient had hardware dislocation 6 days after surgery and required posterior reoperation and anterior fusion. Follow-up was between one and four years and no neurological change was observed. No systematic description of pain status is given or any radiological follow-up. Shortcomings of the presented study are its retrospective nature, the limited case number and the lack of a control group. However, it is a uniform series of exclusively upper and middle thoracic fractures, all of the unstable type B or C, in which a consistent technique was applied. CONCLUSION In this retrospective series of eight patients with traumatic unstable upper and middle thoracic fractures, kyphosis and anterolisthesis correction and thereby spinal canal decompression could be safely and reliably performed with polyaxial thoracic pedicle screws and undercontoured rods. Solid fixation was achieved, and the deformity correction could be maintained throughout the observation period in all patients. Exact anatomical knowledge of thoracic pedicles and a meticulous technique for thoracic pedicle screw insertion are mandatory.
REFERENCES 1. Schweighofer F, Hofer H, Wildburger R, Stockenhuber N, Bratschitsch G Unstable fractures of the upper thoracic spine Langenbecks Arch Chir 1997; 382: 25–28. 2. Yue J, Sossan A, Selgrath C, et al. The treatment of unstable thoracic spine fractures with transpedicular screw instrumentation: a 3-year consecutive series Spine 2002; 27: 2782–2787. 3. Ebraheim N, Jabaly G, Xu R, et al. Anatomic relations of the thoracic pedicle to the adjacent neural structures Spine 1997; 22: 1553–1556. 4. Panjabi MM, O’Holleran J, Crisco J, et al. Complexity of the thoracic spine pedicle anatomy Eur Spine J 1997; 6: 19–24. 5. Vaccaro A, Rizzolo S, Allardyce T, et al. Placement of pedicle screws in the thoracic spine. Part I: Morphometric analysis of the thoracic vertebrae J Bone Joint Surg Am 1995; 77: 1193–1199.
ª 2005 Elsevier Ltd. All rights reserved.
Correction-fixation of unstable thoracic features 533
6. Vaccaro A, Rizzolo S, Balderston R, et al. Placement of pedicle screws in the thoracic spine. Part II: An anatomical and radiographic assessment J Bone Joint Surg Am 1995; 77: 1200–1206. 7. Heller J, Shuster J, Hutton W. Pedicle and transverse process screws of the upper thoracic spine. Biomechanical comparison of loads to failure Spine 1999; 24: 654–658. 8. Liljenvist U, Hackenberg L, Link T, et al. Pullout strength of pedicle screws versus pedicle and laminar hooks in the thoracic spine Acta Orthop Belg 2001; 67: 157–163. 9. Maynard F, Bracken M, Creasey G, et al. International standards for neurological and functional classification of spinal cord injury Spinal Cord 1997; 35: 266–274. 10. Magerl F, Aebi M, Gertzbein S, Harms J, Nazarian S. A comprehensive classification of thoracic and lumbar injuries Eur Spine J 1994; 3: 184–201. 11. Lenke L, Rinella A, Kim YJ. Freehand thoracic pedicle screw placement Semin Spine Surg 2002; 14: 44–52. 12. Yingsakmonkol W, Karaikovic E, Gaines R. The accuracy of pedicle screw placement in the thoracic spine using the “funnel technique” J Spinal Dis 2002; 15: 445–449. 13. Kuklo T, Polly D, Owens B, Zeidman S, Chang A, Klemme W. Measurement of thoracic and lumbar fracture kyphosis: evaluation of intra-observer, interobserver, and technique variability Spine 2001; 26: 61–65. 14. Abraham D, Herkowitz H, Katz J. Indications for thoracic and lumbar spine fusion and trends in use Orthop Clin North Am 1998; 29: 803–811. 15. Bohlman H Treatment of fractures and dislocations of the thoracic and lumbar spine J Bone Joint Surg Am 1985; 67: 165–169. 16. Capen D, Gordon M, Zigler J, et al. Non-operative management of upper thoracic spine fractures Orthop Rev 1994; 10: 818–821.
ª 2005 Elsevier Ltd. All rights reserved.
17. Edwards C, Levine A. Early rod-sleeve stabilization of the injured thoracic and lumbar spine Orthop Clin North Am 1986; 17: 120–145. 18. Flesch J, Leider L, Erickson D. Harrington instrumentation and spine fusion for unstable fractures and fracture dislocations of the thoracic and lumbar spine J Bone Joint Surg Am 1977; 59: 143–153. 19. Floman Y, Fast A, Pollack D. The simultaneous application of an interspinous compressive wire and Harrington distraction rods in the treatment of fracture-dislocation of the thoracic and lumbar spine Clin Orthop 1986; 205: 207–215. 20. Katonis P, Kontakis G, Loupasis G, et al. Treatment of unstable thoracolumbar and lumbar spine injuries using Cotrel-Dubousset instrumentation Spine 1999; 24: 2352–2357. 21. Korovessis P, Baikousis A, Stamatakis M. Use of the Texas Scottish Rite Hospital instrumentation in the treatment of thoracolumbar injuries Spine 1997; 22: 882–888. 22. McCullen G, Vaccaro A, Garfin S. Thoracic and lumbar trauma Orthop Clin North Am 1998; 29: 813–828. 23. Suk S, Kim W, Lee S, et al. Thoracic pedicle screw fixation in spinal deformities: are they really safe? Spine 2001; 26: 2049–2057. 24. Lehman R, Polly D, Kuklo T, et al. Straight-forward versus anatomic trajectory technique of thoracic pedicle screw fixation: a biomechanical analysis Spine 2003; 28: 2058–2065. 25. Husted D, Yue J, Fairchild B, Haims A. An extrapedicular approach to the placement of screws in the thoracic spine: an anatomic and radiographic assessment Spine 2003; 28: 2324–2330. 26. Shapiro S, Abel T, Rodgers R. Traumatic thoracic spinal fracture dislocation with minimal or no cord injury. Report of four cases and review of the literature J Neurosurg 2002; 96: 333–337.
Journal of Clinical Neuroscience (2005) 12(5), 529–533
Clinical study
Unstable upper and middle thoracic fractures. Preliminary experience with a posterior transpedicular correction-fixation technique M. Payer MD Department of Neurosurgery, University Hospital of Geneva, Geneva, Switzerland
Summary A number of conservative and operative approaches have been described for the treatment of unstable traumatic upper and middle thoracic fractures. The advantage of surgical correction and fixation/fusion lies in its potential to restore sagittal and coronal alignment, thereby indirectly decompressing the spinal cord. A consecutive series of 8 patients with unstable traumatic upper and middle thoracic fractures is reviewed. In all patients, polyaxial pedicle screws were inserted bilaterally into the two levels above and below the fracture. Rods that were less contoured (‘‘undercontoured”) than the regional hyperkyphosis at the injured level, were anchored to the caudal four screws. The cranial four screws, with the vertebrae to which they were inserted, were then progressively pulled posteriorly onto the undercontoured rods with rod reducers, thus correcting the hyperkyphosis and anterolisthesis. The mean follow-up was 15 months. The mean regional kyphosis was 23 preoperatively, 17 postoperatively and 18 at follow-up. The mean anterolisthesis was 8 mm preoperatively, 1 mm postoperatively and 1 mm at follow-up. No hardware failure occurred. Five patients with complete spinal cord injury at presentation made no neurological recovery, two patients with incomplete spinal cord injury initially (ASIA B), recovered substantially (to ASIA D), and the patients who were neurologically intact at presentation remained so. ª 2005 Elsevier Ltd. All rights reserved. Keywords: spine trauma, thoracic fracture, pedicle screw fixation
INTRODUCTION The treatment of unstable upper and middle thoracic fractures (T1-10) is controversial and rarely addressed exclusively in the literature.1,2 The advantages of surgical stabilization are immediate restoration of the sagittal and coronal alignment in kyphotic fractures and the correction of an anterolisthesis in subluxation fractures, thereby leading to indirect decompression of the spinal canal. Fixation and fusion furthermore can prevent progressive hyperkyphosis, which can result in local pain, neurological deterioration, and the psychological consequences of a ``hunchback” deformity. Posterior thoracic fixation can be performed by means of pedicle screws or laminar hooks. Thoracic pedicle screw fixation can be technically challenging, as anatomical work has shown a smaller pedicle diameter in the upper and especially middle thoracic spine than in the lower thoracic spine.3–6 Additionally, biomechanical studies have found much stronger pull-out forces with thoracic pedicle screws than with laminar hooks.7,8 The current study was performed to evaluate correction and fixation/fusion in 8 consecutive patients with unstable upper and middle thoracic fractures, using polyaxial pedicle screws two levels above and below the fracture and ‘‘undercontoured” rods. MATERIALS AND METHODS Eight consecutive patients (7 men and 1 woman) with a mean age of 40 years (range 20–65) with traumatic unstable upper and middle thoracic fractures were operated by the author over a 2-year Received 2 July 2004 Accepted 30 November 2004 Correspondence to: Dr Michael Payer, Department of Neurosurgery, University Hospital of Geneva, Rue Micheli-du-Crest 24, 1211 Geneva 14, Switzerland. Tel.: +41 22 372 82 24; Fax: +41 22 372 82 20; E-mail: [email protected]
period (Table 1). Patient hospital charts and radiological examinations were reviewed from the hospital admission, and all patients were clinically and radiologically reviewed in the outpatients clinic by the author. The average follow-up was 15 months (10– 19 months). Preoperative neurological injury was assessed according to the ASIA classification9 (Table 2). Initial spinal cord injury was complete (ASIA A) in five patients, incomplete (ASIA B) in two patients, and no spinal cord injury was present in one patient (ASIA E). Plain antero-posterior and lateral radiographs and computed tomography (CT) scan of the thoracic spine with reconstruction revealed a type B injury in four patients and a type C lesion in four patients according to the Magerl classification10 (Table 3).
Surgical technique After hemodynamic and pulmonary stabilization of the thoracic trauma that was always concomitant, and other abdominal or peripheral injuries, surgery was performed at a mean of three days (range 1–6) after the accident, under general anesthesia with the patient in the prone position. Gel cushions were placed under the upper thorax and the iliac crests bilaterally, freeing the abdomen. After midline incision and exposure of the spine, pedicle screws were inserted bilaterally into the two levels below and two levels above the fracture. Anteroposterior and lateral fluoroscopy was used to control the entry point, which was positioned at the junction of the proximal edge of the transverse process with the lamina.11 Using the funnel technique12 with a high speed drill, a starting hole 30–40 mm in depth was created and palpated with a probe to ensure bony surroundings. Under fluoroscopic control, polyaxial top-loading self-tapping titanium screws (CD M8, Medtronics, Memphis, TN), 4.5 mm in diameter and 30–40 mm in length, were inserted. Screw insertion was performed in a slightly converging inclination in the middle thoracic spine and a more converging angle in the upper thoracic spine according to the 529
530 Payer
Table 1 Patient summary Patient
Injury
Deformity Pre-op
Neurological Post-op
Operation
Deformity Post-op
SD 21M
T3/4 Type B
33 kyphosis 9 mm ant-listh
ASIA A
T1-5
19 kyphosis 2 mm ant-listh
EC 44M
T4/5 Type B
26 kyphosis 7 mm ant-listh
ASIA A
T2-7
LZ 65F
T4/5 Type B
29 kyphosis 7 mm ant-listh
ASIA A
SF 47M
T4 Type B
21 kyphosis 0 mm ant-listh
AJ 43M
T4/5 Type C
LK 44M
Follow-up
Deformity Follow-up
Neurological Follow-up
Pain Follow-up
19 m
20 kyphosis 2 mm ant-listh
ASIA A
2/10
15 kyphosis 0 mm ant-listh
14 m
15 kyphosis 0 mm ant-listh
ASIA A
3/10
T2-7
2 kyphosis 2 mm ant-listh
12 m
24 kyphosis 2 mm ant-listh
ASIA A
2/10
ASIA E
T2-6
13 kyphosis 0 mm ant-listh
10 m
14 kyphosis 0 mm ant-listh
ASIA E
2/10
23 kyphosis 32 scoliosis 5 mm ant-listh
ASIA B
T2-7
20 kyphosis 11 scoliosis 0 mm ant-listh
17 m
22 kyphosis 10 scoliosis 0 mm ant-listh
ASIA D
0/10
T5/6 Type C
21 kyphosis 11 mm ant-listh
ASIA B
T2-8
10 kyphosis 0 mm ant-listh
15 m
10 kyphosis 0 mm ant-listh
ASIA D
2/10
SS 36M
T4 Type C
33 kyphosis 9 mm ant-listh
ASIA A
T2-6
20 kyphosis 1 mm ant-listh
15 m
22 kyphosis 1 mm ant-listh
ASIA A
3/10
LK 20M
T2/3 Type C
19 kyphosis 16 mm ant-listh
ASIA A
C7-T5
15 kyphosis 0 mm ant-listh
14 m
15 kyphosis 1 mm ant-listh
ASIA A
2/10
Indicates patient initials, age and sex; injury level and type; preoperative deformity (ant-listh = anterolisthesis) and neurological state; level of surgery; follow-up period in months; deformity, neurological state and thoracic back pain (on a visual analog scale 0–10) at follow-up.
Table 2 Standard neurological classification of spinal cord injury according to the American Spinal Injury Association, ‘‘ASIA impairment scale’’9 A B C D E
Complete spinal cord injury: No motor or sensory function preserved in the sacral segments S4-S5 Incomplete spinal cord injury: Sensory but not motor function preserved below the neurological level and includes the sacral segments S4-S5 Incomplete spinal cord injury: Motor function preserved below the neurological level, and more than half of the key muscles below the neurological level have a muscle grade less than 3 Incomplete spinal cord injury: Motor function preserved below the neurological level, and at least half of the key muscles below the neurological level have a muscle grade of 3 or more Normal: Motor and sensory function is normal
individual pedicle anatomy. The screw trajectory was chosen parallel to the superior endplate. Titanium rods 5.5 mm in diameter were contoured to physiological lordosis, that is ``undercontoured” in respect to the regional fracture hyperkyphosis, and anchored in the caudal four screws. The cranial four screws with the attached vertebrae were then progressively pulled posteriorly unto the undercontoured rods with rod reducers, thus correcting the hyperkyphosis and anterolisthesis (Figs. 1–3). If there was scoliosis, distraction on the concave side was additionally performed. Finally, the laminae and spinous processes of the instrumented and fractured vertebrae were decorticated; the tips of the spinous
Fig. 2 Postoperative sagittal CT reconstruction of the patient in Figure 1, showing complete reduction of the anterolisthesis and correction of the kyphosis.
Fig. 1 Preoperative sagittal CT reconstruction, demonstrating a type B lesion of T 4/5 with 26 of regional kyphosis and 7 mm anterolisthesis.
Journal of Clinical Neuroscience (2005) 12(5), 529–533
processes of the instrumented and fractured vertebrae were resected and used as a local autograft, augmented by allograft bone chips, and applied posteromedially. Thus, no iliac crest autograft was taken, avoiding a second incision and donor site morbidity. No external immobilisation was applied and the patients were immediately mobilised as much as their concomitant lesions and neurological state allowed. Postoperative plain radiographs and CT scan with sagittal and coronal reconstruction were performed to evaluate deformity correction and screw placement. Regional kyphosis was determined by the angle of Cobb13 between the superior endplate of the first ª 2005 Elsevier Ltd. All rights reserved.
Correction-fixation of unstable thoracic features 531
Fig. 3 Plain antero-posterior (a) and lateral (b) radiographs of the patient in Figure 1 and 2, at follow-up 14 months after T2-7 reduction, fixation and fusion.
intact vertebra above the fracture and the inferior endplate of the first intact vertebra below the fracture. Posttraumatic scoliosis was determined as the angle between the superior endplate of the first intact vertebra above the fracture and the inferior endplate of the first intact vertebra below the fracture. Anterolisthesis was measured in millimeters as the anterior translation of the posterior vertebral body wall of a fractured vertebra in respect to the subjacent vertebra. At follow-up, back pain was assessed using a visual analogue scale (VAS, minimum pain = 0, maximum pain = 10), neurological injury was evaluated according to the ASIA score, and sagittal and coronal alignment, as well as implant position and integrity, were evaluated on plain antero-posterior and lateral thoracic spine radiographs (Fig. 3). RESULTS Mean surgical duration was 211 minutes (range 182–245), mean intraoperative blood loss 630 cc (range 420–860 cc). There were no surgical complications. Sixty-two thoracic and two C7 pedicle screws were inserted in total. Postoperative radiological analysis showed 56 thoracic and both C7 pedicle screws to be correctly placed through the pedicles into the vertebral bodies. Five thoracic pedicle screws were found within the pedicles, but were aimed too cranially and crossed into the disc space. One thoracic pedicle screw was positioned lateral to the pedicle, through the transverse process into the vertebral body. No pedicle screw was in the spinal canal. At follow-up no fracture or displacement of screws or rods were found nor any radiolucency around the pedicle screws. No hardware removal or other second surgery in the thoracic spine was performed in any patient. The mean regional kyphosis was 23 preoperatively (range 11– 33), 17 postoperatively (range 10–22), and 18 at follow-up (range 10–24). Mean anterolisthesis was 8 mm preoperatively (range 0–16 mm), 1 mm postoperatively (range 0–2 mm) and 1 mm at follow-up (range 0–2 mm). Preoperative scoliosis was ª 2005 Elsevier Ltd. All rights reserved.
found in one patient with 32, corrected to 10 postoperatively. Mean surgical kyphosis correction was 6 immediately postoperatively and 5 at follow-up, average translation correction 7 mm postoperatively and at follow-up. Neurologically, all five patients with complete initial spinal cord injury remained unchanged at follow-up (ASIA A), the two patients with initially incomplete lesions (ASIA B) recovered substantially to ASIA D, and the initially neurologically intact patient remained intact. Thoracic back pain according to the VAS at follow-up was between 0 and 3 in all patients with a mean value of 2. No patient was taking regular pain medication for back pain. No patient was limited in his daily activities because of back pain and all patients except for one patient (LZ, Table 1) were living independently at follow-up. This patient remained in the rehabilitation center for the remainder of her life due to severe depression, obesity, recurrent urinary and pulmonary infections, and finally died after pulmonary embolism. DISCUSSION Traumatic unstable upper and middle thoracic fractures are rare. There is no consensus regarding optimal treatment and various conservative and operative options have been described in the literature with different inclusion criteria, follow-ups and evaluation tools.14–22 There are no prospective randomised or controlled studies. A surgical fixation procedure for an unstable upper or middle thoracic fracture should be safe, correct the frequent kyphotic deformity and anterolisthesis and thereby decompress the spinal cord and maintain the deformity correction long-term.2,21,22 The above described technique meets these criteria. A posterior midline approach is less invasive than the posterolateral or anterior approach to the thoracic spine. Regional hyperkyphosis and anterolisthesis could be safely and reliably corrected by anchoring undercontoured rods in a block of four pedicle screws below the fracture, then pulling the kyphotic or Journal of Clinical Neuroscience (2005) 12(5), 529–533
532 Payer Table 3 Magerl’s comprehensive classification of thoracic and lumbar injuries10 Type A Injuries
Vertebral body compression
A1 A2 A3
Impaction fractures Split fractures Burst fractures
Type B Injuries
Anterior and posterior element injury with distraction
B1 B2 B3
Posterior disruption, predominantly ligamentous (flexion-distraction injury) Posterior disruption, predominantly osseous (flexion-distraction injury) Anterior disruption through the disc (hyperextension-shear injury)
Type C Injuries
Anterior and posterior element injury with rotation
C1 C2 C3
Type A injuries with rotation Type B injuries with rotation Rotational-shear injuries
anteriorly translated fractured spine with the block of four pedicle screws above the fracture posteriorly unto the undercontoured rods by means of rod reducers. Finally, deformity correction was maintained in all cases throughout the observation period. Pedicle screw insertion in the upper and middle thoracic spine is challenging due to the specific osseous morphology, as has been shown in various reports.3–6 However, with exact anatomical knowledge and meticulous technique, thoracic pedicle screws can be used safely.23 In this series, the recently described funnel technique12 was found to be helpful. Lateral fluoroscopy of the upper thoracic spine is often of poor quality, but usually sufficient to position the pedicle screw parallel to the superior endplates. This screw path has recently been biomechanically analysed to have a significantly higher pullout strength than a cephalo-caudad pedicle screw along the anatomical axis of the pedicle.24 Anteroposterior fluoroscopy was reliable in finding the pedicle eyes, thus confirming the entry point of the pedicles. Care must be taken to install the fluoroscopy tube perpendicular to the instrumented vertebra by compensating for the local kyphosis in order to obtain a true anteroposterior image. In the current series, a pedicle screw diameter of 4.5 mm and a length between 30 and 40 mm seemed biomechanically sufficient, while being small enough to be accommodated by the upper and middle thoracic pedicles. According to a recent study, thoracic pedicle screws can even be placed outside very narrow thoracic pedicles, using a trajectory through the transverse process into the vertebral body, a technique recently described as “extrapedicular placement of screws in the thoracic spine”.25 Biomechanical work showed a 75% peak insertional torque for thoracic pedicle screws at the neurocentral junction, which was measured 15 to 20 mm from the dorsal cortex of the pedicle.24 This may explain the sufficient biomechanical strength even in suboptimal screw placements, provided the screw traverses the neurocentral junction. The increased pull-out strength of thoracic pedicle screws as compared to laminar hooks, pedicle hooks or transverse process screws7,8 allowed safe and reliable correction of the frequently found kyphotic deformity or anterolisthesis in the unstable upper and middle thoracic fractures in this series. Thus, a supplementary anterior approach in these three-column injuries was deemed unnecessary. A solid fusion in the thoracic spine is difficult to assess, but the absence of thoracic back pain in daily living in combination with maintained sagittal and coronal deformity correction at the followup makes fusion very probable in the patients in the current series. Substantial neurological recovery was observed in both patients with an initial incomplete spinal cord injury with ASIA B score, improving to ASIA D at follow-up. These two patients with 5 and 11 mm anterolisthesis respectively had bilateral pedicle shear fractures, which protected the spinal cord from severe compression. This finding is in accordance with a recent report, showing bilateral pedicle shear fractures to be present in all 15 published thoracic fracture dislocations without neurological deficit.26 Journal of Clinical Neuroscience (2005) 12(5), 529–533
To the author’s knowledge there are only two other reports on pedicle screw fixation of upper or middle thoracic fractures in the English literature.1,2 Yue et al., in their series, also performed transpedicular fixation two levels above and below the fracture.2 They used varying screw diameters from 4.2 to 7 mm according to the pedicle size. In contrast to the current series, they anchored undercontoured rods in the cranial vertebrae first and pulled the caudal vertebrae onto undercontoured rods in order to correct kyphotic deformity. This series included 16 patients with fractures between T1 and T10. At a median follow-up of 22 months they observed no hardware failure, no loss of correction, no painful hardware requiring removal and no postoperative neurological worsening. Schweighofer et al. briefly described a series of 6 patients with Magerl type B or C lesions in the upper and middle thoracic spine, operated using a posterior approach with Louis or Steffee plates.1 The technique is not described. One patient had hardware dislocation 6 days after surgery and required posterior reoperation and anterior fusion. Follow-up was between one and four years and no neurological change was observed. No systematic description of pain status is given or any radiological follow-up. Shortcomings of the presented study are its retrospective nature, the limited case number and the lack of a control group. However, it is a uniform series of exclusively upper and middle thoracic fractures, all of the unstable type B or C, in which a consistent technique was applied. CONCLUSION In this retrospective series of eight patients with traumatic unstable upper and middle thoracic fractures, kyphosis and anterolisthesis correction and thereby spinal canal decompression could be safely and reliably performed with polyaxial thoracic pedicle screws and undercontoured rods. Solid fixation was achieved, and the deformity correction could be maintained throughout the observation period in all patients. Exact anatomical knowledge of thoracic pedicles and a meticulous technique for thoracic pedicle screw insertion are mandatory.
REFERENCES 1. Schweighofer F, Hofer H, Wildburger R, Stockenhuber N, Bratschitsch G Unstable fractures of the upper thoracic spine Langenbecks Arch Chir 1997; 382: 25–28. 2. Yue J, Sossan A, Selgrath C, et al. The treatment of unstable thoracic spine fractures with transpedicular screw instrumentation: a 3-year consecutive series Spine 2002; 27: 2782–2787. 3. Ebraheim N, Jabaly G, Xu R, et al. Anatomic relations of the thoracic pedicle to the adjacent neural structures Spine 1997; 22: 1553–1556. 4. Panjabi MM, O’Holleran J, Crisco J, et al. Complexity of the thoracic spine pedicle anatomy Eur Spine J 1997; 6: 19–24. 5. Vaccaro A, Rizzolo S, Allardyce T, et al. Placement of pedicle screws in the thoracic spine. Part I: Morphometric analysis of the thoracic vertebrae J Bone Joint Surg Am 1995; 77: 1193–1199.
ª 2005 Elsevier Ltd. All rights reserved.
Correction-fixation of unstable thoracic features 533
6. Vaccaro A, Rizzolo S, Balderston R, et al. Placement of pedicle screws in the thoracic spine. Part II: An anatomical and radiographic assessment J Bone Joint Surg Am 1995; 77: 1200–1206. 7. Heller J, Shuster J, Hutton W. Pedicle and transverse process screws of the upper thoracic spine. Biomechanical comparison of loads to failure Spine 1999; 24: 654–658. 8. Liljenvist U, Hackenberg L, Link T, et al. Pullout strength of pedicle screws versus pedicle and laminar hooks in the thoracic spine Acta Orthop Belg 2001; 67: 157–163. 9. Maynard F, Bracken M, Creasey G, et al. International standards for neurological and functional classification of spinal cord injury Spinal Cord 1997; 35: 266–274. 10. Magerl F, Aebi M, Gertzbein S, Harms J, Nazarian S. A comprehensive classification of thoracic and lumbar injuries Eur Spine J 1994; 3: 184–201. 11. Lenke L, Rinella A, Kim YJ. Freehand thoracic pedicle screw placement Semin Spine Surg 2002; 14: 44–52. 12. Yingsakmonkol W, Karaikovic E, Gaines R. The accuracy of pedicle screw placement in the thoracic spine using the “funnel technique” J Spinal Dis 2002; 15: 445–449. 13. Kuklo T, Polly D, Owens B, Zeidman S, Chang A, Klemme W. Measurement of thoracic and lumbar fracture kyphosis: evaluation of intra-observer, interobserver, and technique variability Spine 2001; 26: 61–65. 14. Abraham D, Herkowitz H, Katz J. Indications for thoracic and lumbar spine fusion and trends in use Orthop Clin North Am 1998; 29: 803–811. 15. Bohlman H Treatment of fractures and dislocations of the thoracic and lumbar spine J Bone Joint Surg Am 1985; 67: 165–169. 16. Capen D, Gordon M, Zigler J, et al. Non-operative management of upper thoracic spine fractures Orthop Rev 1994; 10: 818–821.
ª 2005 Elsevier Ltd. All rights reserved.
17. Edwards C, Levine A. Early rod-sleeve stabilization of the injured thoracic and lumbar spine Orthop Clin North Am 1986; 17: 120–145. 18. Flesch J, Leider L, Erickson D. Harrington instrumentation and spine fusion for unstable fractures and fracture dislocations of the thoracic and lumbar spine J Bone Joint Surg Am 1977; 59: 143–153. 19. Floman Y, Fast A, Pollack D. The simultaneous application of an interspinous compressive wire and Harrington distraction rods in the treatment of fracture-dislocation of the thoracic and lumbar spine Clin Orthop 1986; 205: 207–215. 20. Katonis P, Kontakis G, Loupasis G, et al. Treatment of unstable thoracolumbar and lumbar spine injuries using Cotrel-Dubousset instrumentation Spine 1999; 24: 2352–2357. 21. Korovessis P, Baikousis A, Stamatakis M. Use of the Texas Scottish Rite Hospital instrumentation in the treatment of thoracolumbar injuries Spine 1997; 22: 882–888. 22. McCullen G, Vaccaro A, Garfin S. Thoracic and lumbar trauma Orthop Clin North Am 1998; 29: 813–828. 23. Suk S, Kim W, Lee S, et al. Thoracic pedicle screw fixation in spinal deformities: are they really safe? Spine 2001; 26: 2049–2057. 24. Lehman R, Polly D, Kuklo T, et al. Straight-forward versus anatomic trajectory technique of thoracic pedicle screw fixation: a biomechanical analysis Spine 2003; 28: 2058–2065. 25. Husted D, Yue J, Fairchild B, Haims A. An extrapedicular approach to the placement of screws in the thoracic spine: an anatomic and radiographic assessment Spine 2003; 28: 2324–2330. 26. Shapiro S, Abel T, Rodgers R. Traumatic thoracic spinal fracture dislocation with minimal or no cord injury. Report of four cases and review of the literature J Neurosurg 2002; 96: 333–337.
Journal of Clinical Neuroscience (2005) 12(5), 529–533