Radiological evaluation of C1 pedicle screw anatomic feasibility

Journal of Clinical Neuroscience xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Opinion paper

Radiological evaluation of C1 pedicle screw anatomic feasibility Bryden Dawes a,⇑, Yuliya Perchyonok b,c, Augusto Gonzalvo a,c a

Department of Neurosurgery, Austin Health, 145 Studley Road, Heidelberg, Victoria 3070, Australia Department of Radiology, Austin Health, 145 Studley Road, Heidelberg, Victoria 3070, Australia c The University of Melbourne, Parkville, Victoria 3010, Australia b

a r t i c l e

i n f o

Article history: Received 2 May 2017 Accepted 5 January 2018 Available online xxxx Keywords: Atlas C1 vertebra Cervical pedicle screws Cervical posterior fixation Internal fixation

a b s t r a c t C1 pedicle screw instrumentation is a recently documented technique, which may have benefits over other more popularised procedures, however it may not be possible in all patients. This study aims to investigate the applicability of the C1 pedicle screw technique to a cohort of patients through assessment of radiological parameters. A retrospective review of 150 consecutive patients undergoing computer tomography (CT) of the cervical spine was performed. Based on defined parameters images were assessed for feasibility of placement of pedicle screws. C1 pedicle height (PH), pedicle width (PW), screw trajectory length (LML) and width (LMW) were recorded with PH  4 mm defined as the primary outcome measure. A total of 115 patients and 230 C1 pedicles were examined. The mean PH was found to be 5.1 mm. 207 pedicles were deemed suitable for placement of pedicle screw instrumentation. Overall C1 pedicle screw instrumentation was deemed possible in 94.6% of male pedicles and 84% of female pedicles. C1 pedicle screw instrumentation is feasible to be performed in more than 90% of the patients. Crown Copyright Ó 2018 Published by Elsevier Ltd. All rights reserved.

1. Introduction Posterior instrumentation of the first cervical vertebra may be required in the management of numerous spinal pathologies including degenerative spine, trauma and tumour surgery. Multiple fixation techniques have been described including sublaminar wiring, C1-C2 transarticular screws, monocortical and bicortical lateral mass screws and pedicle screws. The most widely used technique, described by Harms et al. [1] in 2001, involves screw placement within the lateral mass of C1. This technique utilises an entry point at the centre of the lateral mass with a convergent and cranial trajectory. Whilst successful in achieving rigid fixation and allowing reduction, the technique requires careful dissection of the paravertebral venous plexus and has risk of C2 nerve root injury. In addition bicortical screw placement, which achieves improved biomechanical results, may place ventral structures such as the hypoglossal nerve and internal carotid artery at risk. Recently a technique of monocortical pedicle screw instrumentation has been described [2–4]. The technique utilises an entry point on the posterior arch, which whilst requiring careful preparation of the posterior arch of C1 and the use of image guidance methods in most cases, negates the need to enter the spinal canal and expose the lateral mass and adjacent venous plexus. Addition⇑ Corresponding author. E-mail address: [email protected] (B. Dawes).

ally the trajectory passes along the posterior arch, enabling placement of longer screws and maximising the screw-cortical bone interface. This improves the screw purchase with recent cadaveric studies demonstrating improved pull out strength comparable to bicortical lateral mass screws [5,6]. The variable dimensions of the C1 pedicle are the major limiting parameters in the technique. Anatomical variation of the vertebral artery or an anomalous posterior arch may also limit the technique’s applicability. Cannulation of the posterior arch relies on a canal of adequate size to avoid splitting of the cortical bone and potential injury to the adjacent vertebral artery. Typically, a cutoff of 4 mm has been utilised when using a 3.5 mm cannulated screw, however smaller pedicles may accept screw placement due to bone flexibility (Tables 1 and 2). We propose that despite its limitations that the C1 pedicle screw technique is widely applicable and should be considered when assessing instrumentation methods. We aim to assess the feasibility of C1 pedicle screws in the Australian population through retrospective assessment of radiological parameters.

2. Methods A retrospective review was performed on a consecutive series of 150 patients undergoing cervical spine computer tomography (CT) in our institution. To ensure standardized imaging protocols all patients presented through our institution’s Emergency Depart-

https://doi.org/10.1016/j.jocn.2018.01.006 0967-5868/Crown Copyright Ó 2018 Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Dawes B et al. Radiological evaluation of C1 pedicle screw anatomic feasibility. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.01.006

2

B. Dawes et al. / Journal of Clinical Neuroscience xxx (2018) xxx–xxx

Table 1 Results. Total

Male

Female

Mean

±SD

Mean

±SD

Mean

±SD

PH (mm)

Right Left Total

5.2 5.0 5.1

0.9 1.0 1.0

5.5 5.3 5.4

1.0 0.9 1.0

4.9 4.6 4.7

0.8 0.8 0.8

PW (mm)

Right Left Total

12.4 13.3 12.8

2.1 2.1 2.1

12.7 13.8 13.2

2.2 2.2 2.2

12.1 12.6 12.3

2.0 1.8 1.9

LMW (mm)

Right Left Total

9.1 9.1 9.1

1.4 1.4 1.4

9.5 9.5 9.5

1.4 1.3 1.4

8.6 8.6 8.6

1.2 1.3 1.3

LML (mm)

Right Left Total

27.6 27.1 27.4

2.3 2.4 2.4

28.1 27.5 27.8

2.6 2.5 2.5

26.9 26.6 26.7

1.8 2.2 2.0

Table 2 Dichotomised results. Total

PH 4 mm

Right Left Total

Male

Female

Count

%

Count

%

Count

%

104 103 207

90.4 89.6 90

62 61 123

95.4 93.8 94.6

42 42 84

84 84 84

ment with trauma as an indication for cervical spine imaging. Institutional ethics approval was obtained. Patients were eligible for inclusion if they were aged between 18 and 85 and had undergone adequate imaging of the craniocervical junction. Patients were excluded if a fracture or discoligamentous injury was identified with initial or subsequent imaging, if there was evidence of previous cervical spine surgery or if there was significant degenerative disease and anatomical variation affecting C1. For each patient, basic demographic characteristics including gender, age and mechanism of injury was obtained. All examinations were performed on a GE LightSpeed Series VCT 64 slice scanner. For each patient included in the study, volumetric thin slice (0.625 mm) bone algorithm CT cervical spine data was reviewed. Three-dimensional reformats were generated, which were used for all subsequent image manipulation and measurements. All measurements were made using bone windows. Imaging manipulation and measurements were performed in consensus with a Neurosurgeon and Neuroradiologist jointly

reviewing the imaging. Prior to measurement, all images were reconstructed in the axis of the C1 pedicle, parallel to the posterior arch as shown in Fig. 1. A planned screw entry point and trajectory was then made to simulate ideal surgical placement. Four measurements were made bilaterally for each patient as demonstrated in Fig. 2. The dimensions of the planned screw trajectory were measured at the vertebral artery groove, in the coronal plane, for pedicle craniocaudal height (PH) and pedicle width (PW). The width of the lateral mass trajectory corridor (LMW) was measured, defined as the narrowest point perpendicular to the trajectory. This dimension was limited by the foramen transversarium laterally and vertebral canal medially. Additionally the length of trajectory from entry point to the ventral border of the lateral mass (LML) was measured. This simulates the maximum screw length required to achieve monocortical fixation. The outer cortex was used for all measurements (Fig. 3). Statistical analysis was performed using PASW 18 software (SPSS, Chicago Illinois). PH was the primary outcome measure

Fig. 1. Illustrative example of generated three – dimensional reformats. (A) Axial (B) Coronal (C) Sagittal CT reformats of the cervical spine reconstructed in the axis of the C1 pedicles used for measurements.

Please cite this article in press as: Dawes B et al. Radiological evaluation of C1 pedicle screw anatomic feasibility. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.01.006

B. Dawes et al. / Journal of Clinical Neuroscience xxx (2018) xxx–xxx

3

Fig. 2. (A) Coronal CT image of C1. Measurement of pedicle height (PH) and pedicle width (PW). (B) Axial CT image of C1. Measurement of lateral mass width (LMW). Dashed line denotes screw trajectory. (C) Axial CT image of C1. Measurement of lateral mass length (LML).

Fig. 3. Illustrative case of C1 pedicle instrumentation. (A) Sagittal CT (B) Axial CT.

and results dichotomised into greater or equal to, or less than 4 mm. PW, LMW and LML were secondary outcome measures. 3. Results One hundred and fifty consecutive patients were assessed for inclusion in the study. 115 patients were deemed eligible for inclusion. 35 patients were excluded based on defined criteria. 31 patients were aged greater then 85 years, three patients had anatomical variations (two incomplete C1 posterior arches, one anomalous vertebral artery) and one patient had degenerative C1/C2 subluxation. In total, 230 screw trajectories were assessed. The mean age of patients of was 54.0 (range 18–85). There were 65 males and 50 females with an average age of 53.4 (range 18– 85) and 54.7 (range 18–85) respectively. Pedicle height was the smallest dimension and used to determine instrumentation suitability. The mean PH was 5.1 mm (±1.8 mm) with a mean left PH of 5.0 mm (±1.8 mm) and right PH of 5.2 mm (±1.8 mm) respectively. The mean PH was larger in males at 5.4 mm compared with females at 4.7 mm. 23 pedicles had a height less then 4 mm and were deemed inappropriate for instrumentation, 16 in the female group and 7 in the male group. Overall C1 pedicle screw instrumentation was deemed possible in 94.6% of male pedicles and 84% of female pedicles, this difference was statistically significant (P < 0.05). The average PW was 12.3 mm (±3.8 mm) and was larger in the male group. The screw trajectory corridor had a mean width (LMW) and length (LML) of 8.5 mm (±2.4 mm) and 26.6 mm (±4.0 mm) respectively. 4. Discussion The optimal technique for instrumentation of the first cervical vertebra has yet to be determined. The C1 pedicle screw technique

is comparable to other techniques, but may not be applicable in all cases. We were able to demonstrate within a population of patients presenting to a tertiary centre that based on radiological parameters, the technique is widely applicable (94.6% and 84% of male and female pedicles respectively had a height of at least 4 mm). C1 instrumentation via the pedicle has been shown to be a safe technique in clinical practice [3,7,8]. Cadaveric studies have identified the lateral third of the posterior arch under the vertebral artery groove as the ideal entry point with a medial inclination of approximately 10 degrees [2]. The benefit when compared with the Harms technique is the entry point on the posterior arch negating the need for exposure of the lateral mass in the spinal canal avoiding the bleeding from the venous plexus. The C1 pedicle screw technique has also been shown in cadaveric studies to have superior pull out strengths and biomechanical stability to the lateral mass technique [5,6]. Ma et al. found comparable pullout strengths of monocortical pedicle screws to bicortical lateral mass screws [5]. Entry point on the lateral mass allows for placement of longer screws and increases the area of screw/cortical bone interface. The greatest limitation to application of pedicle screw technique is the posterior arch height at the vertebral artery groove. It is generally accepted that a 4 mm screw trajectory dimension is required for a 3.5 mm screw. We have demonstrated that the technique is widely applicable, consistent with earlier research; however smaller mean pedicle sizes have been previously noted. Qian et al. recently looked at C1 pedicle morphology utilizing CT [9]. They found within their cohort, at a Tertiary Chinese Hospital 23.3% of pedicles were less then 4 mm. Cadaveric studies have reported up to 53% of pedicles having a height less the 4 mm [10]. We found the pedicle height of females to be significantly less than in males, 5.4 mm compared with 4.7 mm. This is consistent with previous work [9,10]. This should be considered when assessing patient suitability for the technique. Pedicle height whilst utilized in this study as a parameter for instrumentation suitability may not be the limiting factor. Huang et al. proposed that the presence of an intramedullary canal might allow for placement of screws in pedicles smaller then 4 mm [11]. They achieved pedicle instrumentation in a series of 14 patients with pedicle height less than 4 mm. This is likely due to canal expansion due to bone flexibility, well documented in thoracolumbar pedicle screw instrumentation. The complications related to C1 pedicle screw instrumentation are similar to the Harms technique including vertebral artery injury, hypoglossal nerve injury or occipital neuralgia. Due to the close relationship to the vertebral artery, there is an increased risk of injury with pedicle breach or fracture when compared with the Harms technique. In the clinical series by Thomas et al. [3] and Yeom et al. [8] no vertebral artery injuries were documented. In

Please cite this article in press as: Dawes B et al. Radiological evaluation of C1 pedicle screw anatomic feasibility. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.01.006

4

B. Dawes et al. / Journal of Clinical Neuroscience xxx (2018) xxx–xxx

Menger et al. [7] series of nine patients there was one breach when instrumenting a pedicle with a height of 2.8 mm with no vertebral artery injury. This risk may be reduced by the use of image guidance, minimizing screw misplacement [12]. Image guidance with 3D reconstruction allows surgeons to minimize risk to adjacent structures but also accurately identify landing zones and trajectories rather than opening the canal. This study has several limitations. It represents a retrospective review of radiographic parameters. There was not scope within the research hypothesis for significant clinical information. The major limitation of this study was the inherent error of measurement in the nominated indices. All measurements were performed with agreement of a Neurosurgeon and Neuroradiologist but the small size of the indices are subject to standard error. In addition, the outcomes and conclusions rely on the premise that a pedicle height of 4 mm is the arbitrary cutoff for instrumentation feasibility, which may not be the case. Despite this, based on the current best evidence we were able to show that based on radiographic indices the technique is widely applicable. C1 pedicle screw instrumentation is feasible to be performed in more than 90% of patients within an Australian population.

References [1] Harms J, Melcher RP. Posterior C1-C2 fusion with polyaxial screw and rod fixation. Spine (Phila Pa 1976) 2001;26(22):2467–71. [2] Ma XY et al. Anatomic considerations for the pedicle screw placement in the first cervical vertebra. Spine (Phila Pa 1976) 2005;30(13):1519–23. [3] Thomas JA et al. An alternate method for placement of C-1 screws. J Neurosurg Spine 2010;12(4):337–41. [4] Resnick DK, Lapsiwala S, Trost GR. Anatomic suitability of the C1-C2 complex for pedicle screw fixation. Spine (Phila Pa 1976) 2002;27(14):1494–8. [5] Ma XY et al. C1 pedicle screws versus C1 lateral mass screws: comparisons of pullout strengths and biomechanical stabilities. Spine (Phila Pa 1976) 2009;34 (4):371–7. [6] Fensky F et al. Biomechanical advantage of C1 pedicle screws over C1 lateral mass screws: a cadaveric study. Eur Spine J 2014;23(4):724–31. [7] Menger RP et al. Placement of C1 pedicle screws using minimal exposure: radiographic, clinical, and literature validation. Int J Spine Surg 2015;9:43. [8] Yeom JS et al. Routine insertion of the lateral mass screw via the posterior arch for C1 fixation: feasibility and related complications. Spine J 2012;12 (6):476–83. [9] Qian LX et al. Morphology of the atlas pedicle revisited: a morphometric CTbased study on 120 patients. Eur Spine J 2013;22(5):1142–6. [10] Lee MJ, Cassinelli E, Riew KD. The feasibility of inserting atlas lateral mass screws via the posterior arch. Spine (Phila Pa 1976) 2006;31(24):2798–801. [11] Huang DG et al. Is the 4 mm height of the vertebral artery groove really a limitation of C1 pedicle screw insertion? Eur Spine J 2014;23(5):1109–14. [12] Mason A et al. The accuracy of pedicle screw placement using intraoperative image guidance systems. J Neurosurg Spine 2014;20(2):196–203.

Please cite this article in press as: Dawes B et al. Radiological evaluation of C1 pedicle screw anatomic feasibility. J Clin Neurosci (2018), https://doi.org/ 10.1016/j.jocn.2018.01.006