- Email: [email protected]
CT observation of L5 pedicle screw fixation shielding by the iliac wing width and height
Clinical Neurology and Neurosurgery 189 (2020) 105637
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
CT observation of L5 pedicle screw fixation shielding by the iliac wing width and height
T
Shuai Zhanga, JianYuan Ouyanga, XueLian Pengb, Song Wanga, Qing Wanga,* a b
Department of Spine Surgery, The Affiliated Hospital of Southwest Medical University, China Department of Ultrasonography, The Affiliated Hospital of Southwest Medical University, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Iliac wing width Iliac wing height The 5th lumbar vertebra pedicle screw Three-dimensional CT reconstruction
Objective: The height and width of the iliac wing were measured using three-dimensional computed tomography (3D-CT), and the extent to which the iliac wing occluded the entry point of the 5th lumbar vertebra pedicle screw (L5PS) was observed and graded according to severity. The overall goal was to study the influence of the height and width of the iliac wing on L5PS insertion. Patients and methods: This study included 350 cases (179 males and 171 females) where the L1-S2 region was obtained. We reconstructed 3D-CT in our study for anatomic analysis. The width and height of the iliac wing were measured on CT axial and oblique sagittal images, respectively. The degree of the L5PS entry point was determined by the data of iliac wing width and height, and divided into 0, I, II and III levels. The association between width and height of the iliac wing among males and females was also analyzed in this paper. Results: The incidence of shielding of the L5PS entry point by the height and width of the ilium wing were 20.0 % and 19.4 %, respectively. The overall incidence of obstruction of L5PS by the width of the ilium wing was 27.3 % in males and 12.3 % in females. In total, there were 27, 14, and 8 male patients with level I, II, and III shielding, respectively. There were 12, 7, and 2 females with level I, II, and III shielding, respectively. The overall incidence with which the height of the ilium wing obstructed L5PS was 24.0 % in males and 14.6 % in females, of which there were 23 males with level Ⅰ shielding, 16 males with level Ⅱ shielding, 4 males with level Ⅲ shielding, 13 females with level Ⅰ shielding, 8 females with level Ⅱ shielding, and 4 females with level Ⅲ shielding. Conclusions: The height of the ilium caused obstruction of the L5PS more frequently than the width of the ilium, and males had a higher incidence of iliac wing occlusion than females at both the height and width. The degree of L5PS entry point shielding by the iliac wing width along the horizontal axis was not completely consistent with that of L5PS shielding on the oblique sagittal plane.
1. Introduction Accurate placement of pedicle screws is fundamental to the treatment of spinal trauma, degeneration and other diseases. In two segments in particular, L4/L5 and L5/S1, PS implantation is more difficult due to deep anatomical location, reduced anatomical landmark clarity (caused by facet joint degeneration), and abundant exposure of paravertebral muscle and soft tissue [1–3]. To improve the accuracy of PS implantation, many scholars have studied parameters related to PS fixation, screw implantation techniques, intraoperative navigation, percutaneous puncture, and other techniques [4–7]. However, to date,
there has been sparse data on the effect of the positional relationship between the PS approach point and the iliac wing on successful screw placement, especially at L5 and S1. Choi et al. [8] first reported a correlation between the iliac crest and the L5/S1 intervertebral space on the X-P (anteroposterior and lateral) views of the lumbar spine, and graded the height of the iliac wing according to its relative position. Wang et al. [9] also measured the height of the iliac wing on X-ray images to guide L5PS implantation. The quality of X-ray images is greatly affected by the quality of photographic equipment and position. Further, X-ray is unable to inform 3D spatial relations. The overlap of some bone structures makes it difficult
Abbreviations: L5PS, The 5th lumbar vertebra pedicle screw; 3DCT, three-dimensional computed tomography; 3D, three-dimensionally; MRI, magnetic resonance imaging; CBT, cortical bone trajectory ⁎ Corresponding author at: Department of Spine Surgery, The Affiliated Hospital of Southwest Medical University, No. 25 Taiping Street, Sichuan 646000, Luzhou City, China. E-mail addresses: [email protected] (J. Ouyang), [email protected] (X. Peng), [email protected] (S. Wang), [email protected] (Q. Wang). https://doi.org/10.1016/j.clineuro.2019.105637 Received 29 September 2019; Received in revised form 29 November 2019; Accepted 7 December 2019 Available online 09 December 2019 0303-8467/ © 2019 Elsevier B.V. All rights reserved.
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
minimum internal angle, and the OB' line is the axis of screw implantation with the maximum internal angle. The angle formed by the two lines OA' and OB' is the area in which the L5PS can safely be implanted and is bisected by the reverse extension line of OG (Fig. 1b). As long as the axis of the PS and PS-assisted long-tail instrument is within the range of the safety angles and there is no iliac wing obstruction, the PS can be successfully implanted. Accordingly, when the iliac wing inside the edge of this area does not exceed the reverse extension line of OB', the shielding is defined as level 0 (Fig. 1c); if it exceeds the reverse extension line of OB' but does not exceed the reverse extension line of OG, the shielding is defined as level I (Fig. 1d); if it exceeds the reverse extension line of OG but does not exceed the reverse extension line of OA', the shielding is defined as level II (Fig. 1e); if it exceeds the reverse extension line of OA', the iliac wing will completely shield the L5PS entry point, which is designated as level III shielding (Fig. 1f).
to identify anatomical markers. Jin et al. [10] used CT to observe the effect of iliac wing size on L5PS implantation. In this study, the distance between the medial cortex of the iliac spine and the PS insertion point was used to categorize the degree of shielding. In the simulation, the iliac wing shielded the implanted L5PSs at different points along the axis of screw insertion, without considering the use of long-tail auxiliary devices, such as those for cortical removal, screw channel extension, and tapping, in the process of screw implantation. In addition, the distance between the two points can vary widely due to differences in paraspinal soft tissue, extent of local bone structure degeneration and hyperplasia, and posture. In sum, these research methods have significant limitations in clinical application. This research is aimed to study whether the entry point of L5PS is shielded by the width and height of the iliac wing using 3D-CT reconstruction techniques. Further, methods to classify the degree of occlusion are provided in order to provide a preoperative reference for the accurate implantation of L5PS.
2.3. Methods for measuring and grading L5PS entry point shielding by iliac wing height
2. Materials and methods 2.1. General information
The oblique sagittal view of the vertebral pedicle perpendicular to the upper end plate plane of L5 was used to examine the OG line (Fig. 1). Points C and D of the bone cortex in the narrowest part of the upper and lower diameters of the vertebral pedicle were selected. The middle point of the CD line was defined as E, and the OE line was considered to be the best axis for screw insertion in the sagittal vertebral pedicle. When the PS insertion point is constant, the PS can be inserted by adjusting the angle of the tail. A screw 6.5 mm in diameter was selected, and the intact bone cortex at the narrowest part of the vertebral pedicle was taken as the standard. The upper and lower margins of the screw should be at least 3.25 mm from points C and D and intersect with CD at points C' and D', respectively. When OC' and OD' are connected, the OC' line is the axis of screw implantation with the maximum tail-dip angle, and the OD' line is the axis of screw implantation with the minimum tail-dip angle. The angle formed by the two lines OC' and OD' is the area in which the L5PS can be safely implanted and is bisected by the reverse extension line of OE (Fig. 2b). Accordingly, if the upper edge of the ilium does not exceed the reverse extension line of OC', the shielding is defined as level 0 (Fig. 2c); if it exceeds the reverse extension line of OC' but does not exceed the reverse extension line of OE, the shielding is defined as level I (Fig. 2d); if it exceeds the reverse extension line of OE but does not exceed the reverse extension line of OD', the shielding is defined as level II (Fig. 2e); if it exceeds the reverse extension line of OD', the iliac wing will completely shield the L5PS insertion point, which is thus defined as level III shielding (Fig. 2f).
The study was conducted in a tertiary referral medical center in Southwest China between January 2015 and June 2018. We reviewed all medical records in our prospectively-maintained database. A simple random sampling method was used to select the CT data of patients who received L1∼S2 thin-layer CT (LightSpeed VCT, GE Healthcare, IN, 64-slice plain, 0.625 mm). All original CT image data were transmitted to an ADM 4.4 workstation for reconstruction and measurement (Advantage Workstation, version ADW 4.4, GE Healthcare, IN). The study was approved by the local institutional review board at the authors’ affiliated institution (K2018009-R). Every patient provided informed consent. Inclusion criteria: 1) Patients with non-spinal or pelvic bone structure disease, such as those affecting the abdominal or pelvic cavity, examined by thin-layer CT. 2) Patients with lumbar or sacral pain examined by thin-layer CT combined with magnetic resonance imaging (MRI) and X-rays, which showed no obvious lumbar vertebral body or intervertebral disc lesions. Exclusion criteria: 1) Patients with deformity, trauma, infection, tumor, metabolic bone disease, etc., affecting the lumbar vertebrae or intervertebral disc space. 2) Patients under 18 years old. 3) Patients previously surgically treated in the lumbosacral region. A total of 350 consecutive patients were enrolled in this study. The 350 CT scan images were reviewed by two orthopedic surgeons (Shuai Zhang and Jianyuan Ouyang). Discrepancies in analysis were resolved by debate until consensus was achieved. 2.2. Methods for measuring and grading L5PS entry point shielding by iliac wing width
2.4. Statistical analysis Statistical analysis was performed using the commercial software package SPSS 19.0 (SPSS, Chicago, Illinois, USA). All results for continuous variables are presented as the mean ± SD, and results for categorical variables are expressed as n. Sex-specific data on the grade reflecting the influence of the width and the height of the iliac wing on each trajectory were compared using Wilcoxon’s rank-sum test. P < 0.05 was considered to reflect statistical significance.
In this study, according to the Magerl [2] method, the intersection of the tangent line of the upper articular processes and the bisector line of the transverse processes of L5 was used as the insertion point. First, the insertion point (O point) was located on the standard anteroposterior view of the surface reconstruction, intercepting the plane parallel to the upper end plate as the measuring plane through point O (Fig. 1a). In the horizontal axis view, point G was defined as the midpoint of line AB connecting the inner and outer cortical points A and B of the narrowest part of the vertebral pedicle, and the OG line was considered the optimum axis for PS placement. With a constant insertion point, the PS can be inserted by adjusting the angle of inclination. A screw with a diameter of 6.5 mm was selected for implantation, and the bone cortex at the narrowest part of the vertebral pedicle was taken as the standard. The lateral edge of the screw should be at least 3.25 mm away from A and B and intersect line AB at A' and B', respectively. When OA' and OB' are connected, the OA' line indicates the
3. Results 3.1. Demographic information There were 530 cases with L1-S2 thin-slice CT scans, excluding: 43 cases of L5/S1 intervertebral disc herniation, 31 cases of L5/S1 spinal canal stenosis, 29 cases under 18 years old, 21 cases of lumbar scoliosis deformity, 19 cases of L5 spondylolisthesis, 12 cases of L5 vertebral fracture, 11 cases of L5/S1 suppurative spondylitis, 7 cases of sacral 2
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
Fig. 1. Classification of L5PS entry point shielding by the width of the iliac wing. a The insertion point (O) was located on the standard anteroposterior view of the surface reconstruction. b Point G was defined as the midpoint of line AB connecting the inner and outer cortical points A and B of the narrowest part of the vertebral pedicle, and the OG line was considered the optimum axis for PS placement. The lateral edge of the screw should be at least 3.25 mm away from A and B and intersect line AB at A' and B', respectively. The OA' line indicates the minimum internal angle, and the OB' line is the axis of screw implantation with the maximum internal angle. c Level 0: the inner edge of the iliac wing does not exceed the reverse extension line of OB' and will not affect L5PS implantation. d Level I: the inner edge of the iliac wing exceeds the reverse extension line of OB' but does not exceed the reverse extension line of OG. e Level II, the inner edge of the iliac wing exceeds the reverse extension line of OG but does not exceed the reverse extension line of OA'. f Level III: if the inner edge of the iliac wing exceeds the reverse extension line of OA', the L5PS implantation point is completely shielded.
3.2. Radiological results
fracture, 3 cases of L5/S1 spinal tuberculosis, 2 cases of L5/S1 spinal tumors and 2 cases of ankylosing spondylitis. A total of 350 cases were included in this study, including 179 males and 171 females, with an average age of 49.4 ± 13.1 years (18–75 years), an average height of 164.6 ± 7.0 cm (150–185 cm), and an average weight of 65.3 ± 9.4 kg (43–86 kg).
Among the 350 patients we observed, the incidence of L5PS entry point shielding greater than level 0 were 20.0 % (70/350) and 19.4 % (68/350) by the height and width of the ilium wing respectively. The width of the ilium wing obstructed L5PS in 27.3 % (49/179) and 12.3 % (43/179) of male and female subjects respectively. Among these were 27 males with level Ⅰ shielding, 14 males with level Ⅱ 3
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
Fig. 2. Classification of L5PS entry point shielding by the height of the iliac wing. a The insertion point (O) was located on the standard anteroposterior view of the surface reconstruction. b Point E was defined as the midpoint of line CD connecting the upper and lower cortical points C and D of the narrowest part of the vertebral pedicle, and the OE line was considered the optimum axis for PS placement. The edge of the screw should be at least 3.25 mm away from C and D and intersect line CD at C' and D', respectively. The OC' line is the axis of screw implantation with the maximum tail-dip angle, and the OD' line is the axis of screw implantation with the minimum tail-dip angle. c Level 0: the upper edge of the iliac wing does not exceed the reverse extension line of OC' and will not affect the L5PS implantation. d Level I: the upper edge of the iliac wing exceeds the reverse extension line of OC' but does not exceed the reverse extension line of OE. e Level II, the upper edge of the iliac wing exceeds the reverse extension line of OE but does not exceed the reverse extension line of OD'. f Level III: the upper edge of the iliac wing exceeds the reverse extension line of OD', meaning that the L5PS implantation point is completely blocked.
shielding, 8 males with level Ⅲ shielding, 12 females with level Ⅰ shielding, 7 females with level Ⅱ shielding, and 2 females with level Ⅲ shielding. The height of the ilium wing obstructed L5PS in 24.0 % and 14.6 % in males and females respectively. Among these, there were 23 males with level Ⅰ shielding, 16 males with level Ⅱ shielding, 4 males with level Ⅲ shielding, 13 females with level Ⅰ shielding, 8 females with level Ⅱ shielding, and 4 females with level Ⅲ shielding. The extent to which the width of the iliac wing shielded L5PS was significantly greater in males than in females (Z=−13.195, P = 0.00), as was the
degree of L5PS shielding by the height of the iliac wing (Z=−12.387, P = 0.00, Tables 1 and 2). The degree of L5PS entry point shielding by the width of the iliac wing on the horizontal axis was not exactly the same as that of the height of the iliac wing on the sagittal plane in each patient. In this study, there were 70 patients with wide iliac wings and 68 patients with high iliac wings. Among these, the degree of shielding by the width and height of the iliac wing were the same in 35 patients and different in 44 patients (Table 3). 4
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
intervertebral space subsidence, L5 spondylolisthesis, and scoliosis. Moreover, the results of our study are somewhat at odds with previous studies based on the concept of a high iliac crest. In this study, it was found that shielding of the L5PS insertion point is actually affected by both the width of the iliac wing on the horizontal axis and the height of the iliac wing on the oblique sagittal axis. The horizontal width of the iliac wing affects the selection of the PS insertion point in terms of the axial position, the angle of introversion, and the depth of screw insertion into the vertebral body. However, the oblique sagittal height of the iliac wing affects the selection of the PS insertion point in terms of the oblique sagittal position, the direction of the head and tail, and the placement of the connecting rods. From our observations, while large horizontal and vertical iliac crest dimensions were often associated with each other, the degree of L5PS insertion point shielding found in both directions was not completely consistent. Thus, it is necessary to distinguish these observations. In this study, the incidence of L5PS shielding was significantly higher in males than in females, and the influence of iliac wing size on L5PS implantation was greater in males than in females. The main reasons for these differences may be as follows: First, as a result of human evolution, due to pregnancy and birth, the transverse diameter is greater than the anterior and posterior diameters in the female pelvis, resulting in the bilateral iliac wings in females being closer to the anterolateral side than in males, thus reducing shielding of the L5PS insertion point [15,16]. Second, pregnancy increases the content of the pelvic cavity. Hormones relax the ligaments of muscles around the pelvis, enabling increased pelvic movement and changes in pelvic morphology, further reducing shielding of the L5PS insertion point [17–19]. Finally, due to the effects of osteoporosis and uterine prolapse caused by menopause, adaptation of the pelvic shape may also lead to decreased shielding of the L5PS insertion point [20,21]. In our surgical experience, the universal spine system has the advantages of standardization and simplification of implants and instruments, shorter segment fixation, small overall volume and easy to assemble upper opening operating system [22,23]. Based on this system, we have found that in cases of level I shielding, as determined by preoperative imaging, a simple reduction in the screw tilt angle (or insertion depth) or a shift in the implantation point is able to achieve satisfactory implantation. In cases of level II shielding, on the horizontal axis, better angles can significantly reduce the depth of screw insertion into the vertebral body and obvious deviation from the pedicle axis, significantly reducing the strength of PS fixation. Although screw implantation can be achieved by internal screw insertion, it can significantly reduce the range of screw insertion direction adjustment, reduce the depth of vertebral PS insertion into the vertebral body, damage the facet joint, accelerate facet degeneration, and increase the risk of vertebral pedicle penetration. On the sagittal axis, when shielding of the L5PS entry point by the iliac wing reaches level II, decreasing the shielding by trimming the angle may increase the risk of damaging the pedicle nerve root through the bottom wall, as would reducing the tailtilt angle. In addition, for screws that obviously deviate from the axis of the vertebral pedicle and form an angle with the upper and lower endplates, complications such as screw cutting may occur after surgery. Therefore, for these patients, we can adopt the percutaneous implantation technique proposed by Guo et al. [24] to minimize the interference of the peripheral soft tissue muscle during implantation. In cases of level III shielding, we cannot apply PS internal fixation treatment. If internal fixation with a vertebral PS is necessary, the cortical bone trajectory (CBT) screw fixation technique proposed by Santoni et al. [25] can be adopted, in which the screw is placed with the head tilted at 25° and an extraversion of 10° to avoid iliac wing shielding to the greatest extent. Another option is the use of the transiliac puncture and implantation technique, but the patient must be informed of the possible complications of iliac graft removal; in this way, a more ideal surgical plan can be formulated. In addition, for patients with an inconsistent degree of L5PS insertion point shielding by the iliac wings on
Table 1 Difference in the degree of L5PS entry point shielding by the width of the iliac wing between males and females. Gender case(n)
Male Female Z P
level[case(%)]
179 171 −13.227 0.000
0
Ⅰ
Ⅱ
Ⅲ
130(72.6) 150(87.7)
27(15.1) 12(7.0)
14(7.8) 7(4.1)
8(4.5) 2(1.2)
Table 2 Difference in the degree of L5PS entry point shielding by the height of the iliac wing between males and females. Gender case(n)
Male Female Z P
level[case(%)]
179 171 −12.405 0.000
0
Ⅰ
Ⅱ
Ⅲ
138(77.1) 146(85.4)
22(12.3) 13(7.6)
15(8.4) 8(4.7)
4(2.2) 4(2.3)
Table 3 Distribution of the degree of shielding by the width and height of the iliac wing. Width
0 Ⅰ Ⅱ Ⅲ Total
Height 0
Ⅰ
Ⅱ
Ⅲ
Total
271 9 0 0 280
9 21 8 1 39
2 6 10 3 21
0 0 6 4 10
282 36 24 8 350
4. Discussion Accurate screw implantation is crucial for maintaining physical integrity and reducing complications [11,12]. However, iliac wing occlusion of the L5PS entry point can increase the difficulty of paravertebral muscle stripping and traction, narrow the exposure range of the entry point, and hinder the application of long tail devices necessary for tasks such as cortex removal, tapping, and screw entry. Entry point occlusion is associated with severe local soft tissue injury, prolonged operation time, increased radiation exposure, repeated screw implantation, and even damage to adjacent vascular and nerve structures due to incorrect screw position. In order to improve the accuracy of screw placement and avoid the complications caused by iliac bone resection, many reports have suggested using intraoperative navigation, robot-assisted methods and percutaneous screw placement technologies. However, in clinical application, it has been shown that the accuracy of screw placement by hand is more than 90 %, and is not associated with the dramatically increased costs of these other alternative methods. In developing countries, hand placement is still the mainstream method [13,14]. The approach used in this study is an effective method to improve the accuracy of manual screw placement. According to our observations, the incidence with which iliac wings obstruct L5PS implantation was slightly higher than that found by Choi and Wang [8,9]. One possible explanation for this discrepancy is that we observed the spatial relationship between the iliac wing and L5PS entry point by CT. Compared with X-ray images, CT images have higher imaging quality, avoid potential interference by surrounding soft tissues, and can be processed after image acquisition. In sum, these benefits allow for more accurate results. Our study included patients with normal lumbosacral bone structure, but the incidence of high iliac spine and wide iliac spine may be increased if they are accompanied by L5/S1 5
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
the study for their valuable contributions.
the horizontal and sagittal axes, patients with higher degrees of shielding should be considered for this type of implantation. Although the Magerl [2] method was used to select the entry point of pedicle screws in our study, surgeons can alternatively select an entry point according to the method of Roy-Camille and Du [3,4]. Indeed, according to different entry points, shielding results may be different. In addition, this study is a retrospective study of imaging from a single center. Further, eligible patients were all from Southwest China. It is necessary to further verify the validity and repeatability of this study with a significant amount of samples from multiple centers. However, these method of observing the degree to which the iliac wing occludes the L5PS entry point from the perspective of imaging can provide reference for further clinical research. Finally, we did not exclude patients with lumbosacral transitional vertebrae for the following reasons:
References [1] H.H. Boucher, A method of spinal fusion, J. Bone Joint Surg.-Br. Vol. 41 (2) (1959) 248–259. [2] F.P. Magerl, Stabilization of the lower thoracic and lumbar spine with external skeletal fixation, Clin. Orthop. 189 (189) (1984) 125–141. [3] R. Roycamille, G. Saillant, C. Mazel, Internal fixation of the lumbar spine with pedicle screw plating, Clin. Orthop. Relat. Res. 203 (203) (1986) 7–17. [4] D.U. Xin ru, L.X. Zhao, Y.M. Zhang, Anatomical study of the adjacent structures to the top point of the “∧” shape crest and its relevance, Chin. J. Spine Spinal Cord. 11 (2) (2001) 89–92. [5] T. Laine, T. Lund, M. Ylikoski, J. Lohikoski, D. Schlenzka, Accuracy of pedicle screw insertion with and without computer assistance: a randomised controlled clinical study in 100 consecutive patients, Eur. Spine J. 9 (3) (2000) 235–240. [6] M. Law, A.F. Tencer, P.A. Anderson, Caudo-cephalad loading of pedicle screws: mechanisms of loosening and methods of augmentation, Spine 18 (16) (1993) 2438–2443. [7] M. Matjaz, D. Igor, V. Matjaz, B. Tomaz, R. Gregor, A multi-level rapid prototyping drill guide template reduces the perforation risk of pedicle screw placement in the lumbar and sacral spine, Arch. Orthop. Trauma Surg. 133 (7) (2013) 893–899. [8] K.C. Choi, C.K. Park, Percutaneous endoscopic lumbar discectomy for L5-S1 disc herniation: consideration of the relation between the Iliac Crest and L5-S1 disc, Pain Physician 19 (2) (2016) E301–E308. [9] Y.L. Wang, X.Y. Wang, et al., L5–S1 disc degeneration and the anatomic parameters of the iliac crest: imaging study, Eur. Spine J. 24 (11) (2015) 2481–2487. [10] H.M. Jin, X.Q. Bai, X.X. Pan, et al., Does the iliac wing influence L5 pedicle screw fixation? World Neurosurg. 5 (113) (2018) e302–e308. [11] Mac-Thiong J M, H. Labelle, M. Rooze, et al., Evaluation of a transpedicular drill guide for pedicle screw placement in the thoracic spine, Eur. Spine J. 12 (5) (2003) 542–547. [12] T. Ma, Y.Q. Xu, Y.B. Cheng, et al., A novel computer-assisted drill guide template for thoracic pedicle screw placement: a cadaveric study, Arch. Orthop. Trauma Surg. 132 (1) (2012) 65–72. [13] F. Roser, M. Tatagiba, G. Maier, Spinal robotics: current applications and future perspectives, Neurosurgery 72 (1) (2013) A12–A18. [14] Young-Woo Kim, Lawrence G. Lenke, Yongjung J. Kim, et al., Free-hand pedicle screw placement during revision spinal surgery, Spine 33 (10) (2008) 1141–1148. [15] K.R. Chin, A.F. Kuntz, H.H. Bohlman, S.E. Emery, Changes in the iliac crest-lumbar relationship from standing to prone, Spine J. 6 (2) (2006) 185–189. [16] M.M. Abitbol, The shapes of the female pelvis. Contributing factors, J. Reprod. Med. 41 (4) (1996) 242–250. [17] A.E. Anderson, C.L. Peters, B.D. Tuttle, J.A. Weiss, Subject-specific finite element model of the pelvis: development, validation and sensitivity studies, J. Biomech. Eng. 127 (3) (2005) 364–373. [18] G.S. Huerta-Enochian, V.L. Katz, L.K. Fox, J.A. Hamlin, J.P. Kollath, Magnetic resonance–based serial pelvimetry: Do maternal pelvic dimensions change during pregnancy? Am. J. Obstet. Gynecol. 194 (6) (2006) 1689–1694. [19] K.J. Lehmann, A. Wischnik, K. Zahn, M. Georgi, Do the obstetrically relevant bony pelvic measurements change? A retrospective analysis of computed tomographic pelvic x-rays, Rofo. 156 (5) (1992) 425–428. [20] K.J. Lehmann, A. Wischnik, K. Zahn, M. Georgi, Do the obstetrically relevant bony pelvic measurements change? A retrospective analysis of computed tomographic pelvic x-rays, Rofo 156 (5) (1992) 425–428. [21] K. Narumoto, M. Sugimura, K. Saga, Y. Matsunaga, Changes in pelvic shape among Japanese pregnant women over the last 5 decades, J. Obstet. Gynaecol. Res. 41 (11) (2015) 1687–1692. [22] Q.Z. Jia, J.C. Gao, C.M. Zhang, Surgical treatment of kyphosis with universal spine system, Chin. J. Reparative Reconstructive Surg. 17 (4) (2003) 276–278. [23] A. Macchiarola, F.P.D. Carlo, F.P.D. Pietro, et al., USS internal fixator in lumbar and thoracolumbar vertebral fractures, Chir. Organi Mov. 85 (2) (2000) 177–184. [24] J. Guo, L. Guo, J. Gao, Q. Ling, Z. Yin, E. He, Does a deep seated L5 vertebra position with respect to the iliac crests affect the accuracy of percutaneous pedicle screw placement at lumbosacral junction? BMC Musculoskelet. Disord. 18 (180) (2017) 1–6. [25] B.G. Santoni, R.A. Hynes, K.C. Mcgilvray, et al., Cortical bone trajectory for lumbar pedicle screws, Spine J. 9 (5) (2009) 366–373.
1 There is still debate about the definition and nomenclature of lumbosacral transitional vertebrae. 2 When lumbar sacralization exists, the reduction or even disappearance of the L5/S1 intervertebral space may increase the occlusion by the ilium wing of the entry point of lower lumbar pedicle screw. 3 Lumbar sacralization increases the incidence of lumbar pain and head adjacent segment degeneration. Sacral lumbarization increases lumbar motion and the incidence of instability in the caudal two segments, whereas the possibility of lumbosacral transitional vertebrae implanting pedicle screw increases. 4 Whether there is a corresponding variation of iliac wing and other bone structures in the lumbosacral transitional vertebrae has not been reported. Further study of the spatial relationship between the lumbosacral transitional vertebrae and the ilium wing is also one of the important directions for our future research. 5. Conclusions The incidence of occlusion was higher in height than in width at L5PS. Additionally, males have a higher incidence of iliac wing occlusion than females at both the height and width. The degree of L5PS entry point shielding by the iliac wing width along the horizontal axis was not completely consistent with that of L5PS shielding on the oblique sagittal plane. CRediT authorship contribution statement Shuai Zhang: Conceptualization, Methodology, Software, Investigation. JianYuan Ouyang: Data curation, Investigation, Writing - original draft. XueLian Peng: Methodology, Visualization, Software. Song Wang: Validation, Visualization. Qing Wang: Supervision, Writing - review & editing. Declaration of Competing Interest The authors declare that they have no conflict interest. Acknowledgements The authors thank the other investigators, staff, and partcipants of
6
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
CT observation of L5 pedicle screw fixation shielding by the iliac wing width and height
T
Shuai Zhanga, JianYuan Ouyanga, XueLian Pengb, Song Wanga, Qing Wanga,* a b
Department of Spine Surgery, The Affiliated Hospital of Southwest Medical University, China Department of Ultrasonography, The Affiliated Hospital of Southwest Medical University, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Iliac wing width Iliac wing height The 5th lumbar vertebra pedicle screw Three-dimensional CT reconstruction
Objective: The height and width of the iliac wing were measured using three-dimensional computed tomography (3D-CT), and the extent to which the iliac wing occluded the entry point of the 5th lumbar vertebra pedicle screw (L5PS) was observed and graded according to severity. The overall goal was to study the influence of the height and width of the iliac wing on L5PS insertion. Patients and methods: This study included 350 cases (179 males and 171 females) where the L1-S2 region was obtained. We reconstructed 3D-CT in our study for anatomic analysis. The width and height of the iliac wing were measured on CT axial and oblique sagittal images, respectively. The degree of the L5PS entry point was determined by the data of iliac wing width and height, and divided into 0, I, II and III levels. The association between width and height of the iliac wing among males and females was also analyzed in this paper. Results: The incidence of shielding of the L5PS entry point by the height and width of the ilium wing were 20.0 % and 19.4 %, respectively. The overall incidence of obstruction of L5PS by the width of the ilium wing was 27.3 % in males and 12.3 % in females. In total, there were 27, 14, and 8 male patients with level I, II, and III shielding, respectively. There were 12, 7, and 2 females with level I, II, and III shielding, respectively. The overall incidence with which the height of the ilium wing obstructed L5PS was 24.0 % in males and 14.6 % in females, of which there were 23 males with level Ⅰ shielding, 16 males with level Ⅱ shielding, 4 males with level Ⅲ shielding, 13 females with level Ⅰ shielding, 8 females with level Ⅱ shielding, and 4 females with level Ⅲ shielding. Conclusions: The height of the ilium caused obstruction of the L5PS more frequently than the width of the ilium, and males had a higher incidence of iliac wing occlusion than females at both the height and width. The degree of L5PS entry point shielding by the iliac wing width along the horizontal axis was not completely consistent with that of L5PS shielding on the oblique sagittal plane.
1. Introduction Accurate placement of pedicle screws is fundamental to the treatment of spinal trauma, degeneration and other diseases. In two segments in particular, L4/L5 and L5/S1, PS implantation is more difficult due to deep anatomical location, reduced anatomical landmark clarity (caused by facet joint degeneration), and abundant exposure of paravertebral muscle and soft tissue [1–3]. To improve the accuracy of PS implantation, many scholars have studied parameters related to PS fixation, screw implantation techniques, intraoperative navigation, percutaneous puncture, and other techniques [4–7]. However, to date,
there has been sparse data on the effect of the positional relationship between the PS approach point and the iliac wing on successful screw placement, especially at L5 and S1. Choi et al. [8] first reported a correlation between the iliac crest and the L5/S1 intervertebral space on the X-P (anteroposterior and lateral) views of the lumbar spine, and graded the height of the iliac wing according to its relative position. Wang et al. [9] also measured the height of the iliac wing on X-ray images to guide L5PS implantation. The quality of X-ray images is greatly affected by the quality of photographic equipment and position. Further, X-ray is unable to inform 3D spatial relations. The overlap of some bone structures makes it difficult
Abbreviations: L5PS, The 5th lumbar vertebra pedicle screw; 3DCT, three-dimensional computed tomography; 3D, three-dimensionally; MRI, magnetic resonance imaging; CBT, cortical bone trajectory ⁎ Corresponding author at: Department of Spine Surgery, The Affiliated Hospital of Southwest Medical University, No. 25 Taiping Street, Sichuan 646000, Luzhou City, China. E-mail addresses: [email protected] (J. Ouyang), [email protected] (X. Peng), [email protected] (S. Wang), [email protected] (Q. Wang). https://doi.org/10.1016/j.clineuro.2019.105637 Received 29 September 2019; Received in revised form 29 November 2019; Accepted 7 December 2019 Available online 09 December 2019 0303-8467/ © 2019 Elsevier B.V. All rights reserved.
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
minimum internal angle, and the OB' line is the axis of screw implantation with the maximum internal angle. The angle formed by the two lines OA' and OB' is the area in which the L5PS can safely be implanted and is bisected by the reverse extension line of OG (Fig. 1b). As long as the axis of the PS and PS-assisted long-tail instrument is within the range of the safety angles and there is no iliac wing obstruction, the PS can be successfully implanted. Accordingly, when the iliac wing inside the edge of this area does not exceed the reverse extension line of OB', the shielding is defined as level 0 (Fig. 1c); if it exceeds the reverse extension line of OB' but does not exceed the reverse extension line of OG, the shielding is defined as level I (Fig. 1d); if it exceeds the reverse extension line of OG but does not exceed the reverse extension line of OA', the shielding is defined as level II (Fig. 1e); if it exceeds the reverse extension line of OA', the iliac wing will completely shield the L5PS entry point, which is designated as level III shielding (Fig. 1f).
to identify anatomical markers. Jin et al. [10] used CT to observe the effect of iliac wing size on L5PS implantation. In this study, the distance between the medial cortex of the iliac spine and the PS insertion point was used to categorize the degree of shielding. In the simulation, the iliac wing shielded the implanted L5PSs at different points along the axis of screw insertion, without considering the use of long-tail auxiliary devices, such as those for cortical removal, screw channel extension, and tapping, in the process of screw implantation. In addition, the distance between the two points can vary widely due to differences in paraspinal soft tissue, extent of local bone structure degeneration and hyperplasia, and posture. In sum, these research methods have significant limitations in clinical application. This research is aimed to study whether the entry point of L5PS is shielded by the width and height of the iliac wing using 3D-CT reconstruction techniques. Further, methods to classify the degree of occlusion are provided in order to provide a preoperative reference for the accurate implantation of L5PS.
2.3. Methods for measuring and grading L5PS entry point shielding by iliac wing height
2. Materials and methods 2.1. General information
The oblique sagittal view of the vertebral pedicle perpendicular to the upper end plate plane of L5 was used to examine the OG line (Fig. 1). Points C and D of the bone cortex in the narrowest part of the upper and lower diameters of the vertebral pedicle were selected. The middle point of the CD line was defined as E, and the OE line was considered to be the best axis for screw insertion in the sagittal vertebral pedicle. When the PS insertion point is constant, the PS can be inserted by adjusting the angle of the tail. A screw 6.5 mm in diameter was selected, and the intact bone cortex at the narrowest part of the vertebral pedicle was taken as the standard. The upper and lower margins of the screw should be at least 3.25 mm from points C and D and intersect with CD at points C' and D', respectively. When OC' and OD' are connected, the OC' line is the axis of screw implantation with the maximum tail-dip angle, and the OD' line is the axis of screw implantation with the minimum tail-dip angle. The angle formed by the two lines OC' and OD' is the area in which the L5PS can be safely implanted and is bisected by the reverse extension line of OE (Fig. 2b). Accordingly, if the upper edge of the ilium does not exceed the reverse extension line of OC', the shielding is defined as level 0 (Fig. 2c); if it exceeds the reverse extension line of OC' but does not exceed the reverse extension line of OE, the shielding is defined as level I (Fig. 2d); if it exceeds the reverse extension line of OE but does not exceed the reverse extension line of OD', the shielding is defined as level II (Fig. 2e); if it exceeds the reverse extension line of OD', the iliac wing will completely shield the L5PS insertion point, which is thus defined as level III shielding (Fig. 2f).
The study was conducted in a tertiary referral medical center in Southwest China between January 2015 and June 2018. We reviewed all medical records in our prospectively-maintained database. A simple random sampling method was used to select the CT data of patients who received L1∼S2 thin-layer CT (LightSpeed VCT, GE Healthcare, IN, 64-slice plain, 0.625 mm). All original CT image data were transmitted to an ADM 4.4 workstation for reconstruction and measurement (Advantage Workstation, version ADW 4.4, GE Healthcare, IN). The study was approved by the local institutional review board at the authors’ affiliated institution (K2018009-R). Every patient provided informed consent. Inclusion criteria: 1) Patients with non-spinal or pelvic bone structure disease, such as those affecting the abdominal or pelvic cavity, examined by thin-layer CT. 2) Patients with lumbar or sacral pain examined by thin-layer CT combined with magnetic resonance imaging (MRI) and X-rays, which showed no obvious lumbar vertebral body or intervertebral disc lesions. Exclusion criteria: 1) Patients with deformity, trauma, infection, tumor, metabolic bone disease, etc., affecting the lumbar vertebrae or intervertebral disc space. 2) Patients under 18 years old. 3) Patients previously surgically treated in the lumbosacral region. A total of 350 consecutive patients were enrolled in this study. The 350 CT scan images were reviewed by two orthopedic surgeons (Shuai Zhang and Jianyuan Ouyang). Discrepancies in analysis were resolved by debate until consensus was achieved. 2.2. Methods for measuring and grading L5PS entry point shielding by iliac wing width
2.4. Statistical analysis Statistical analysis was performed using the commercial software package SPSS 19.0 (SPSS, Chicago, Illinois, USA). All results for continuous variables are presented as the mean ± SD, and results for categorical variables are expressed as n. Sex-specific data on the grade reflecting the influence of the width and the height of the iliac wing on each trajectory were compared using Wilcoxon’s rank-sum test. P < 0.05 was considered to reflect statistical significance.
In this study, according to the Magerl [2] method, the intersection of the tangent line of the upper articular processes and the bisector line of the transverse processes of L5 was used as the insertion point. First, the insertion point (O point) was located on the standard anteroposterior view of the surface reconstruction, intercepting the plane parallel to the upper end plate as the measuring plane through point O (Fig. 1a). In the horizontal axis view, point G was defined as the midpoint of line AB connecting the inner and outer cortical points A and B of the narrowest part of the vertebral pedicle, and the OG line was considered the optimum axis for PS placement. With a constant insertion point, the PS can be inserted by adjusting the angle of inclination. A screw with a diameter of 6.5 mm was selected for implantation, and the bone cortex at the narrowest part of the vertebral pedicle was taken as the standard. The lateral edge of the screw should be at least 3.25 mm away from A and B and intersect line AB at A' and B', respectively. When OA' and OB' are connected, the OA' line indicates the
3. Results 3.1. Demographic information There were 530 cases with L1-S2 thin-slice CT scans, excluding: 43 cases of L5/S1 intervertebral disc herniation, 31 cases of L5/S1 spinal canal stenosis, 29 cases under 18 years old, 21 cases of lumbar scoliosis deformity, 19 cases of L5 spondylolisthesis, 12 cases of L5 vertebral fracture, 11 cases of L5/S1 suppurative spondylitis, 7 cases of sacral 2
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
Fig. 1. Classification of L5PS entry point shielding by the width of the iliac wing. a The insertion point (O) was located on the standard anteroposterior view of the surface reconstruction. b Point G was defined as the midpoint of line AB connecting the inner and outer cortical points A and B of the narrowest part of the vertebral pedicle, and the OG line was considered the optimum axis for PS placement. The lateral edge of the screw should be at least 3.25 mm away from A and B and intersect line AB at A' and B', respectively. The OA' line indicates the minimum internal angle, and the OB' line is the axis of screw implantation with the maximum internal angle. c Level 0: the inner edge of the iliac wing does not exceed the reverse extension line of OB' and will not affect L5PS implantation. d Level I: the inner edge of the iliac wing exceeds the reverse extension line of OB' but does not exceed the reverse extension line of OG. e Level II, the inner edge of the iliac wing exceeds the reverse extension line of OG but does not exceed the reverse extension line of OA'. f Level III: if the inner edge of the iliac wing exceeds the reverse extension line of OA', the L5PS implantation point is completely shielded.
3.2. Radiological results
fracture, 3 cases of L5/S1 spinal tuberculosis, 2 cases of L5/S1 spinal tumors and 2 cases of ankylosing spondylitis. A total of 350 cases were included in this study, including 179 males and 171 females, with an average age of 49.4 ± 13.1 years (18–75 years), an average height of 164.6 ± 7.0 cm (150–185 cm), and an average weight of 65.3 ± 9.4 kg (43–86 kg).
Among the 350 patients we observed, the incidence of L5PS entry point shielding greater than level 0 were 20.0 % (70/350) and 19.4 % (68/350) by the height and width of the ilium wing respectively. The width of the ilium wing obstructed L5PS in 27.3 % (49/179) and 12.3 % (43/179) of male and female subjects respectively. Among these were 27 males with level Ⅰ shielding, 14 males with level Ⅱ 3
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
Fig. 2. Classification of L5PS entry point shielding by the height of the iliac wing. a The insertion point (O) was located on the standard anteroposterior view of the surface reconstruction. b Point E was defined as the midpoint of line CD connecting the upper and lower cortical points C and D of the narrowest part of the vertebral pedicle, and the OE line was considered the optimum axis for PS placement. The edge of the screw should be at least 3.25 mm away from C and D and intersect line CD at C' and D', respectively. The OC' line is the axis of screw implantation with the maximum tail-dip angle, and the OD' line is the axis of screw implantation with the minimum tail-dip angle. c Level 0: the upper edge of the iliac wing does not exceed the reverse extension line of OC' and will not affect the L5PS implantation. d Level I: the upper edge of the iliac wing exceeds the reverse extension line of OC' but does not exceed the reverse extension line of OE. e Level II, the upper edge of the iliac wing exceeds the reverse extension line of OE but does not exceed the reverse extension line of OD'. f Level III: the upper edge of the iliac wing exceeds the reverse extension line of OD', meaning that the L5PS implantation point is completely blocked.
shielding, 8 males with level Ⅲ shielding, 12 females with level Ⅰ shielding, 7 females with level Ⅱ shielding, and 2 females with level Ⅲ shielding. The height of the ilium wing obstructed L5PS in 24.0 % and 14.6 % in males and females respectively. Among these, there were 23 males with level Ⅰ shielding, 16 males with level Ⅱ shielding, 4 males with level Ⅲ shielding, 13 females with level Ⅰ shielding, 8 females with level Ⅱ shielding, and 4 females with level Ⅲ shielding. The extent to which the width of the iliac wing shielded L5PS was significantly greater in males than in females (Z=−13.195, P = 0.00), as was the
degree of L5PS shielding by the height of the iliac wing (Z=−12.387, P = 0.00, Tables 1 and 2). The degree of L5PS entry point shielding by the width of the iliac wing on the horizontal axis was not exactly the same as that of the height of the iliac wing on the sagittal plane in each patient. In this study, there were 70 patients with wide iliac wings and 68 patients with high iliac wings. Among these, the degree of shielding by the width and height of the iliac wing were the same in 35 patients and different in 44 patients (Table 3). 4
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
intervertebral space subsidence, L5 spondylolisthesis, and scoliosis. Moreover, the results of our study are somewhat at odds with previous studies based on the concept of a high iliac crest. In this study, it was found that shielding of the L5PS insertion point is actually affected by both the width of the iliac wing on the horizontal axis and the height of the iliac wing on the oblique sagittal axis. The horizontal width of the iliac wing affects the selection of the PS insertion point in terms of the axial position, the angle of introversion, and the depth of screw insertion into the vertebral body. However, the oblique sagittal height of the iliac wing affects the selection of the PS insertion point in terms of the oblique sagittal position, the direction of the head and tail, and the placement of the connecting rods. From our observations, while large horizontal and vertical iliac crest dimensions were often associated with each other, the degree of L5PS insertion point shielding found in both directions was not completely consistent. Thus, it is necessary to distinguish these observations. In this study, the incidence of L5PS shielding was significantly higher in males than in females, and the influence of iliac wing size on L5PS implantation was greater in males than in females. The main reasons for these differences may be as follows: First, as a result of human evolution, due to pregnancy and birth, the transverse diameter is greater than the anterior and posterior diameters in the female pelvis, resulting in the bilateral iliac wings in females being closer to the anterolateral side than in males, thus reducing shielding of the L5PS insertion point [15,16]. Second, pregnancy increases the content of the pelvic cavity. Hormones relax the ligaments of muscles around the pelvis, enabling increased pelvic movement and changes in pelvic morphology, further reducing shielding of the L5PS insertion point [17–19]. Finally, due to the effects of osteoporosis and uterine prolapse caused by menopause, adaptation of the pelvic shape may also lead to decreased shielding of the L5PS insertion point [20,21]. In our surgical experience, the universal spine system has the advantages of standardization and simplification of implants and instruments, shorter segment fixation, small overall volume and easy to assemble upper opening operating system [22,23]. Based on this system, we have found that in cases of level I shielding, as determined by preoperative imaging, a simple reduction in the screw tilt angle (or insertion depth) or a shift in the implantation point is able to achieve satisfactory implantation. In cases of level II shielding, on the horizontal axis, better angles can significantly reduce the depth of screw insertion into the vertebral body and obvious deviation from the pedicle axis, significantly reducing the strength of PS fixation. Although screw implantation can be achieved by internal screw insertion, it can significantly reduce the range of screw insertion direction adjustment, reduce the depth of vertebral PS insertion into the vertebral body, damage the facet joint, accelerate facet degeneration, and increase the risk of vertebral pedicle penetration. On the sagittal axis, when shielding of the L5PS entry point by the iliac wing reaches level II, decreasing the shielding by trimming the angle may increase the risk of damaging the pedicle nerve root through the bottom wall, as would reducing the tailtilt angle. In addition, for screws that obviously deviate from the axis of the vertebral pedicle and form an angle with the upper and lower endplates, complications such as screw cutting may occur after surgery. Therefore, for these patients, we can adopt the percutaneous implantation technique proposed by Guo et al. [24] to minimize the interference of the peripheral soft tissue muscle during implantation. In cases of level III shielding, we cannot apply PS internal fixation treatment. If internal fixation with a vertebral PS is necessary, the cortical bone trajectory (CBT) screw fixation technique proposed by Santoni et al. [25] can be adopted, in which the screw is placed with the head tilted at 25° and an extraversion of 10° to avoid iliac wing shielding to the greatest extent. Another option is the use of the transiliac puncture and implantation technique, but the patient must be informed of the possible complications of iliac graft removal; in this way, a more ideal surgical plan can be formulated. In addition, for patients with an inconsistent degree of L5PS insertion point shielding by the iliac wings on
Table 1 Difference in the degree of L5PS entry point shielding by the width of the iliac wing between males and females. Gender case(n)
Male Female Z P
level[case(%)]
179 171 −13.227 0.000
0
Ⅰ
Ⅱ
Ⅲ
130(72.6) 150(87.7)
27(15.1) 12(7.0)
14(7.8) 7(4.1)
8(4.5) 2(1.2)
Table 2 Difference in the degree of L5PS entry point shielding by the height of the iliac wing between males and females. Gender case(n)
Male Female Z P
level[case(%)]
179 171 −12.405 0.000
0
Ⅰ
Ⅱ
Ⅲ
138(77.1) 146(85.4)
22(12.3) 13(7.6)
15(8.4) 8(4.7)
4(2.2) 4(2.3)
Table 3 Distribution of the degree of shielding by the width and height of the iliac wing. Width
0 Ⅰ Ⅱ Ⅲ Total
Height 0
Ⅰ
Ⅱ
Ⅲ
Total
271 9 0 0 280
9 21 8 1 39
2 6 10 3 21
0 0 6 4 10
282 36 24 8 350
4. Discussion Accurate screw implantation is crucial for maintaining physical integrity and reducing complications [11,12]. However, iliac wing occlusion of the L5PS entry point can increase the difficulty of paravertebral muscle stripping and traction, narrow the exposure range of the entry point, and hinder the application of long tail devices necessary for tasks such as cortex removal, tapping, and screw entry. Entry point occlusion is associated with severe local soft tissue injury, prolonged operation time, increased radiation exposure, repeated screw implantation, and even damage to adjacent vascular and nerve structures due to incorrect screw position. In order to improve the accuracy of screw placement and avoid the complications caused by iliac bone resection, many reports have suggested using intraoperative navigation, robot-assisted methods and percutaneous screw placement technologies. However, in clinical application, it has been shown that the accuracy of screw placement by hand is more than 90 %, and is not associated with the dramatically increased costs of these other alternative methods. In developing countries, hand placement is still the mainstream method [13,14]. The approach used in this study is an effective method to improve the accuracy of manual screw placement. According to our observations, the incidence with which iliac wings obstruct L5PS implantation was slightly higher than that found by Choi and Wang [8,9]. One possible explanation for this discrepancy is that we observed the spatial relationship between the iliac wing and L5PS entry point by CT. Compared with X-ray images, CT images have higher imaging quality, avoid potential interference by surrounding soft tissues, and can be processed after image acquisition. In sum, these benefits allow for more accurate results. Our study included patients with normal lumbosacral bone structure, but the incidence of high iliac spine and wide iliac spine may be increased if they are accompanied by L5/S1 5
Clinical Neurology and Neurosurgery 189 (2020) 105637
S. Zhang, et al.
the study for their valuable contributions.
the horizontal and sagittal axes, patients with higher degrees of shielding should be considered for this type of implantation. Although the Magerl [2] method was used to select the entry point of pedicle screws in our study, surgeons can alternatively select an entry point according to the method of Roy-Camille and Du [3,4]. Indeed, according to different entry points, shielding results may be different. In addition, this study is a retrospective study of imaging from a single center. Further, eligible patients were all from Southwest China. It is necessary to further verify the validity and repeatability of this study with a significant amount of samples from multiple centers. However, these method of observing the degree to which the iliac wing occludes the L5PS entry point from the perspective of imaging can provide reference for further clinical research. Finally, we did not exclude patients with lumbosacral transitional vertebrae for the following reasons:
References [1] H.H. Boucher, A method of spinal fusion, J. Bone Joint Surg.-Br. Vol. 41 (2) (1959) 248–259. [2] F.P. Magerl, Stabilization of the lower thoracic and lumbar spine with external skeletal fixation, Clin. Orthop. 189 (189) (1984) 125–141. [3] R. Roycamille, G. Saillant, C. Mazel, Internal fixation of the lumbar spine with pedicle screw plating, Clin. Orthop. Relat. Res. 203 (203) (1986) 7–17. [4] D.U. Xin ru, L.X. Zhao, Y.M. Zhang, Anatomical study of the adjacent structures to the top point of the “∧” shape crest and its relevance, Chin. J. Spine Spinal Cord. 11 (2) (2001) 89–92. [5] T. Laine, T. Lund, M. Ylikoski, J. Lohikoski, D. Schlenzka, Accuracy of pedicle screw insertion with and without computer assistance: a randomised controlled clinical study in 100 consecutive patients, Eur. Spine J. 9 (3) (2000) 235–240. [6] M. Law, A.F. Tencer, P.A. Anderson, Caudo-cephalad loading of pedicle screws: mechanisms of loosening and methods of augmentation, Spine 18 (16) (1993) 2438–2443. [7] M. Matjaz, D. Igor, V. Matjaz, B. Tomaz, R. Gregor, A multi-level rapid prototyping drill guide template reduces the perforation risk of pedicle screw placement in the lumbar and sacral spine, Arch. Orthop. Trauma Surg. 133 (7) (2013) 893–899. [8] K.C. Choi, C.K. Park, Percutaneous endoscopic lumbar discectomy for L5-S1 disc herniation: consideration of the relation between the Iliac Crest and L5-S1 disc, Pain Physician 19 (2) (2016) E301–E308. [9] Y.L. Wang, X.Y. Wang, et al., L5–S1 disc degeneration and the anatomic parameters of the iliac crest: imaging study, Eur. Spine J. 24 (11) (2015) 2481–2487. [10] H.M. Jin, X.Q. Bai, X.X. Pan, et al., Does the iliac wing influence L5 pedicle screw fixation? World Neurosurg. 5 (113) (2018) e302–e308. [11] Mac-Thiong J M, H. Labelle, M. Rooze, et al., Evaluation of a transpedicular drill guide for pedicle screw placement in the thoracic spine, Eur. Spine J. 12 (5) (2003) 542–547. [12] T. Ma, Y.Q. Xu, Y.B. Cheng, et al., A novel computer-assisted drill guide template for thoracic pedicle screw placement: a cadaveric study, Arch. Orthop. Trauma Surg. 132 (1) (2012) 65–72. [13] F. Roser, M. Tatagiba, G. Maier, Spinal robotics: current applications and future perspectives, Neurosurgery 72 (1) (2013) A12–A18. [14] Young-Woo Kim, Lawrence G. Lenke, Yongjung J. Kim, et al., Free-hand pedicle screw placement during revision spinal surgery, Spine 33 (10) (2008) 1141–1148. [15] K.R. Chin, A.F. Kuntz, H.H. Bohlman, S.E. Emery, Changes in the iliac crest-lumbar relationship from standing to prone, Spine J. 6 (2) (2006) 185–189. [16] M.M. Abitbol, The shapes of the female pelvis. Contributing factors, J. Reprod. Med. 41 (4) (1996) 242–250. [17] A.E. Anderson, C.L. Peters, B.D. Tuttle, J.A. Weiss, Subject-specific finite element model of the pelvis: development, validation and sensitivity studies, J. Biomech. Eng. 127 (3) (2005) 364–373. [18] G.S. Huerta-Enochian, V.L. Katz, L.K. Fox, J.A. Hamlin, J.P. Kollath, Magnetic resonance–based serial pelvimetry: Do maternal pelvic dimensions change during pregnancy? Am. J. Obstet. Gynecol. 194 (6) (2006) 1689–1694. [19] K.J. Lehmann, A. Wischnik, K. Zahn, M. Georgi, Do the obstetrically relevant bony pelvic measurements change? A retrospective analysis of computed tomographic pelvic x-rays, Rofo. 156 (5) (1992) 425–428. [20] K.J. Lehmann, A. Wischnik, K. Zahn, M. Georgi, Do the obstetrically relevant bony pelvic measurements change? A retrospective analysis of computed tomographic pelvic x-rays, Rofo 156 (5) (1992) 425–428. [21] K. Narumoto, M. Sugimura, K. Saga, Y. Matsunaga, Changes in pelvic shape among Japanese pregnant women over the last 5 decades, J. Obstet. Gynaecol. Res. 41 (11) (2015) 1687–1692. [22] Q.Z. Jia, J.C. Gao, C.M. Zhang, Surgical treatment of kyphosis with universal spine system, Chin. J. Reparative Reconstructive Surg. 17 (4) (2003) 276–278. [23] A. Macchiarola, F.P.D. Carlo, F.P.D. Pietro, et al., USS internal fixator in lumbar and thoracolumbar vertebral fractures, Chir. Organi Mov. 85 (2) (2000) 177–184. [24] J. Guo, L. Guo, J. Gao, Q. Ling, Z. Yin, E. He, Does a deep seated L5 vertebra position with respect to the iliac crests affect the accuracy of percutaneous pedicle screw placement at lumbosacral junction? BMC Musculoskelet. Disord. 18 (180) (2017) 1–6. [25] B.G. Santoni, R.A. Hynes, K.C. Mcgilvray, et al., Cortical bone trajectory for lumbar pedicle screws, Spine J. 9 (5) (2009) 366–373.
1 There is still debate about the definition and nomenclature of lumbosacral transitional vertebrae. 2 When lumbar sacralization exists, the reduction or even disappearance of the L5/S1 intervertebral space may increase the occlusion by the ilium wing of the entry point of lower lumbar pedicle screw. 3 Lumbar sacralization increases the incidence of lumbar pain and head adjacent segment degeneration. Sacral lumbarization increases lumbar motion and the incidence of instability in the caudal two segments, whereas the possibility of lumbosacral transitional vertebrae implanting pedicle screw increases. 4 Whether there is a corresponding variation of iliac wing and other bone structures in the lumbosacral transitional vertebrae has not been reported. Further study of the spatial relationship between the lumbosacral transitional vertebrae and the ilium wing is also one of the important directions for our future research. 5. Conclusions The incidence of occlusion was higher in height than in width at L5PS. Additionally, males have a higher incidence of iliac wing occlusion than females at both the height and width. The degree of L5PS entry point shielding by the iliac wing width along the horizontal axis was not completely consistent with that of L5PS shielding on the oblique sagittal plane. CRediT authorship contribution statement Shuai Zhang: Conceptualization, Methodology, Software, Investigation. JianYuan Ouyang: Data curation, Investigation, Writing - original draft. XueLian Peng: Methodology, Visualization, Software. Song Wang: Validation, Visualization. Qing Wang: Supervision, Writing - review & editing. Declaration of Competing Interest The authors declare that they have no conflict interest. Acknowledgements The authors thank the other investigators, staff, and partcipants of
6