 |
|
||||
| Home | Login | |||||
| About | Editorial board | Articles
|
NSI Publications
|
Search | Instructions | Online Submission | Subscribe | Videos | Etcetera | Contact |
|
||||||||||||||||||||
|
|
|
Accuracy of Patient-Specific, 3D-Printed Laminofacetal Based Trajectory-Guide for Pedicle Screw Placement in Subaxial Cervical and Thoracic Spine
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.375394
Keywords: 3D printing, pedicle screw, subaxial cervical spine, thoracic spine, trajectory guide
Pedicle screws are an established modality for spinal instrumentation. Fluoroscopic assisted placement, which remains the most common method of insertion, has shown higher breach rates ranging from 5–54%, especially in difficult areas such as the cervical and thoracic spine.[1],[2],[3],[4],[5],[6] Compared to pedicles in the lumbar spine, the inter-individual and inter-segmental variability in pedicle dimension and orientation of the subaxial cervical and thoracic spine is higher.[6],[7],[8] Additionally, intra-operative fluoroscopic visualization of cervicothoracic junction may be difficult to interpret. This poses a significant safety concern regarding the injury to the vertebral artery, spinal cord, nerve roots, viscera, and dural tear.[9],[10] This problem is much more pronounced in patients with pre-existing complex spinal deformities. Pedicle screw placement techniques have evolved over decades to decrease its breach rate. Traditionally, surgeons have been using anatomical landmark-based techniques and fluoroscopic-assisted methods to place the pedicle screws. The recently developed navigation systems and robotic-assisted screw placement have been shown to have better accuracy; however, these are very expensive and require high-end training.[11],[12],[13],[14],[15],1[6],[17],[18] Majority of the hospitals do not have these facilities, especially in developing nations. Patient-specific, three-dimensional (3D)-printed trajectory guide for screw placement is a new technique and undergoing evolution. Different techniques have been used to design such trajectory guides with varying success as the breach rates ranged from 1.5–19%.[18],[19],[20],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30],[31],[32],[33],[34] These guides may prove to be an asset in complex spinal deformity cases involving the cervical and thoracic spine. The higher breach rates in some studies may be attributed to the inept design of the template. In this study, we prospectively sought to evaluate a patient-specific, 3D-printed trajectory guide based on the lamina, facet, and base of the spinous process for placement of pedicle screws in subaxial cervical and thoracic spine in the clinical setting.
The study was conducted in a tertiary-care teaching referral hospital attached to a medical college from September 2018 to August 2019. Approval from the institutional ethical committee was granted. Written informed consent was taken from each patient authorizing treatment, radiographic examination, and photographic documentation. The study involved 23 consecutive cases with subaxial-cervical or thoracic spine pathologies necessitating pedicle screw fixation. They were divided into two groups: group A had the cases without spinal deformity (for example; spinal trauma, cervical spondylotic myelopathy), and group B with pre-existing spinal deformity (for example; healed spinal tuberculosis, congenital scoliosis, and adolescent idiopathic scoliosis). All surgeries were performed by a single, fellowship-trained senior spine surgeon. The pre-operative planning was done using a CT scan with 1 mm cuts to produce Digital Imaging and Communications in Medicine (DICOM) images. DICOM images were imported to a segmentation software MIMICS version 21.0 for Windows (Materialise, Belgium) and a 3D reconstruction of desired surgical area was created [Figure 1]a. This model was exported as an “stl” (stereolithography) file and transferred into the computer-aided design (CAD) program 3-MATIC version 9.0 for windows (Materialise, Belgium). “Trim tool” was used to truncate the final reconstructed model (to trim the ribs) and to separate the individual vertebra selected for screw placement [Figure 1]b. The cylinder-shaped trajectory was virtually placed at the desired level within the 3D model manually in consultation with the operating surgeon. The size of the cylinder was chosen in accordance with the standard pedicle screws. The model was evaluated by a “slicing and transparency tool” for assessing any breach in the pedicle. We also assessed the width of the pedicle to predetermine the size of the pedicle screw. A surface is created over the lamina “wave brush mark tool” which is continuous over the proximal part/base of the spinous process extending up to the facet on either side. “Uniform offset tool” is used to create a template of 4 mm thickness from the previously created surface [Figure 2]a. The right and left cylinder guide, the template surface created, and connecting bridge were combined using the “boolean union tool” to create the trajectory guide [Figure 2]b. Additional shapes (sphere and box) were added using the “create primitive object tool” at the junction of the bridge, cylinders, and the laminal surface medially to improve the stability of the trajectory guide [Figure 2]c. The video file detailing each step is enclosed [Video 1]. The trajectory guide and spine model were 3D printed as fused deposition modeling (FDM) in da Vinci Jr. 1.0 Pro printer (XYZ Printing, Thailand) using polylactic acid (PLA) spool material. Such trajectory guide was printed individually for each level. These 3D-printed pedicle screw trajectory guides were placed over the similarly printed spine model and K-wires were passed through them to assess any breach in the pedicles [Figure 3].
The printed trajectory guides were labeled for corresponding vertebral levels and sterilized using ethylene oxide. Intraoperatively, these patients- and spinal-level specific trajectory guides were placed over the corresponding vertebra after required surgical exposure, which was based on the lamina, facet, and base of the spinous process and held manually. The pedicle screw trajectory was then drilled to a depth corresponding with the pre-operative plan with a guidewire of 1 mm diameter [Figure 4]. The trajectory guide was removed and cannulated drill bit and tap was used over the guidewire. The pedicle track was felt with a ball-tipped pedicle sound before and after tapping. The pedicle screw was delicately inserted along the same trajectory.
Outcome measures: The primary outcome measure was the accuracy of the screw placement, which was assessed using the Gertzbein and Robbins classification[35] system on postoperative CT scan [Figure 5]. It was classified as grade A when the screw is completely within the pedicle, and grade B, C, D, and E when the screw breaches the pedicle's cortex by up to 2.0; 2.1–4.0; 4.1–6.0; and more than 6.0 mm. Grade A screw placement was considered “perfect”, whereas, grades A and B were considered “clinically acceptable”.
Grade C–E screw placement was considered to have a significant deviation from the desired trajectory and, thus, was considered “clinically unacceptable”. The secondary outcome measures were surgical time, blood loss, and radiation exposure. Any event of complication (for example; injury to the vertebral artery, spinal cord, nerve roots, viscera, and dural tear) was also recorded.
Patient-specific characteristics Out of a total of 23, 13 patients were females and the rest were males. Thirteen cases of group A involved 7 cases of spinal trauma (5 in the cervical spine, 2 in the thoracic spine), and 6 cases of cervical spondylotic myelopathy, whereas, 10 cases of group B involved 4 cases of healed tuberculosis (two each in cervical spine and cervicothoracic junction), 4 cases of congenital scoliosis, and 2 cases of adolescent idiopathic scoliosis. The mean age of the patients was 28.39 years (range; 4 years to 70 years). The patients of group B were younger as the mean age was 14.3 years (range; 4 years to 23 years) as compared to 39.23 years (range; 21 years to 70 years) in group A. This difference in the mean age was statistically significant (P-value <0.001). Treatment-related characteristics A total of 194 pedicle screws were placed (114 screws in the cervical and the rest 80 in the thoracic spine). A total of 92 pedicle screws were placed in 13 patients categorized under group A, whereas 102 were placed in other 10 patients categorized under group B. The mean surgical time was 174.69 min (range; 135 min to 197 min) for group A and 205.7 min (range; 165 min to 265 min) for group B (P = 0.012). The mean intraoperative blood loss was 453.84 ml (range; 250 ml to 650 ml) for group A and 535.0 ml for group B (range; 325 ml to 800 ml) (P = 0.24). The mean number of intra-operative C-arm exposure shots was 2.61 (range; 1 to 6) per case or 0.4 per screw for group A and 4.6 (range; 2 to 8) per case or 0.51 per screw for group B (P = 0.008) [Table 1].
Screw-specific accuracy Out of a total of 194 pedicle screws, 187 screws (96.39%) were placed accurately (grade A), whereas 6 (3.09%) had a grade B breach and another (0.51%) had a grade C breach. Thus, 193 out of a total of 194 pedicle screws (99.48%) were clinically acceptable. In the cervical spine, 110 out of a total of 114 pedicle screws (96.49%) had accurate placement whereas only 4 (2 each in groups A and B) had grade B breach (3.5%). In the thoracic spine, 77 out of a total of 80 pedicle screws (96.25%) were placed accurately, whereas, grade B and grade C breach was observed in 2 (2.5%) and 1 screw (1.25%), respectively (all in group B). Out of a total of 92 pedicle screws in patients with no spinal deformity (group A), 90 screws (97.83%) had grade A placement and 2 screws had grade B breach (2.17%), whereas, 97 (95.09%) out of total 102 pedicle screws were placed accurately in patients with spinal deformity (group B) and 4 had grade B breach (3.92%) and another had grade C breach (0.98%) [Table 2].
There was no clinically relevant complication such as vertebral artery injury, spinal cord, or nerve injury observed in any of the cases in our study. None of the pedicle screws needed revision.
Patient-specific, 3D-printed trajectory guides have improved the application and accuracy of pedicle screw placement in various spinal pathologies. It is especially useful in difficult pedicle instrumentation and complex spinal deformities such as scoliosis, which has intravertebral asymmetry and significant vertebral rotations. With the advancement in technology, these trajectory guides are gaining popularity in assisting surgeons in these difficult intra-operative situations. In this study, we evaluated a patient-specific, 3D-printed laminofacetal-based trajectory guide for the placement of pedicle screws in the subaxial-cervical and thoracic spine in a clinical setting. The newer methods of pedicle screw insertion used are navigation[11],[12],[13],[14],[18] (CT or 3D fluoroscopy based), O-Arm[14],[15],[16] (intraop navigation), and robotic technique.[16],[17],[18] Various studies and meta-analyses have found significantly fewer pedicle violations with the use of these recent technologies compared to conventional methods. The computer-assisted navigation, though proven to be very accurate, has a failure rate of 6–13.9%.[11],[12],[13],[14],[18] The major disadvantage is its inability to obtain the 3D image data and real-time computer reconstruction data. Intra-operative 3D navigation system using O-arm has also been developed recently, which has better accuracy as real-time registration minimizes the error.[14],[15],[16] However, the rates of perforation remain 3.2% to 4.8%, which may result from changes in spinal alignments, such as torsion during drilling and screw placement. As intra-operative registration of bone structures takes additional time, it increases operating time; and the complex navigation equipment often requires additional personnel during surgery with a considerable learning curve. The higher radiation exposure with the navigation methods obviously remains a matter of concern. Robotic-assisted pedicle screw insertion is the latest technique, which has shown high accuracy; however, it requires training and incurs considerable infrastructural costs.[16],[17],[18] In a study by Fan et al.,[18] robotic technique had significantly higher accuracy (91.3%) compared to trajectory guide (81.3%) and CT navigation (84.1%). The availability of these sophisticated technologies remains limited in almost all the institutes in the developing world and the majority of the institutions in developed nations. However, we admit the following shortcomings in our study: (1) There was no control group for the comparison of our results with the conventional technique of pedicle screw placement, (2) there was a limited number of patients having significant pre-existing spinal deformity in our study, (3) though this technique appears economically cheaper than other modern sophisticated methods, we have not done cost-analysis in our study, (4) the technique of planning, printing, and sterilization of trajectory guides is time-consuming. The usual time spent on these steps (after pre-operative CT scan) was approximately 24–36 h. Various studies have used different principles in designing these trajectory guides for the cervical and thoracic spine with varying success [Table 3] and [Table 4]. Studies, which designed it separately for either pedicle have less stability and a higher chance of misplacement.[21],[29] The trajectory guide, which is based on the spinous process, transverse process, and lamina, enables a good lock-and-key type of fit, and provides good stability; however, it requires extensive soft-tissue dissection.[19],[20],[25],[26],[30],[32],[34] Some of these designs have large contact areas that may extend up to the adjacent vertebra, which can hinder the placement of trajectory guide, once the pedicle screws are inserted in the adjacent vertebra.[30],[31] Sugawara et al. designed their template where they did not use the whole spinous process but only used the base of the spinous process for anchoring.[24],[27] This, therefore, avoided extensive soft-tissue dissection over the spinous process. They used three separate guides (location, drill, and screw guide template) at each level and reported an accuracy of 98.5%. In our opinion, such a design also has a large contact area that may extend up to the adjacent vertebra and can hinder the placement of the trajectory guide. Additionally, developing and using three guides for every vertebral level may be a cumbersome and tedious process. Because it will take a long time in printing these trajectory guides, their use in emergency trauma settings may be limited. Therefore, we designed trajectory guide anchoring at the lamina, facet, and proximal part/base of the spinous process, which required limited soft-tissue dissection and avoided the hindrance from adjacent level pedicle screws. Its stability and rigidness were increased because of the bridge part, which also helped in its handling intra-operatively. Owing to its simple design, it took reasonably lesser time to design (approximately 2–5 h) and print (approximately 2–7 h/case depending on etiology and number of levels requiring fixation).
These trajectory guides also improve the accuracy of pedicle screw placement and are fairly inexpensive. Though sufficient studies are not available to prove its efficacy over the other newer methods (such as navigation or robotic technique), there are enough studies to demonstrate its superiority over the conventional free-hand technique.[25],[28],[29],[30],[31],[32],[34] Two meta-analyses concluded that the trajectory guide significantly improves the accuracy of the pedicle screw and reduces the operative time and intra-operative blood loss compared with those of the free-hand technique.[36],[37] Various cadaveric and clinical studies have shown high accuracy of pedicle screws at all levels and pathologies such as fracture-dislocation, myelopathy, infections, tumor, and complex deformities.[19],[20],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30],[31],[32],[33],[34] We used the trajectory guides in the subaxial-cervical and thoracic spine and found overall accuracy of 96.39% (which includes 95.09% accuracy in spinal-deformity cases). A systematic review found costs of these guides anywhere between $300 and >$1,000.[38] Sugawara et al.[24] found the cost of a template and vertebra model $4 to $8 and $8 to $20, respectively, not taking into account the cost of the printer and printing material. The cost of the printer in our setup is approximately $340 and the spool material was $30, which can print about 15–20 guides and 2–3 vertebral models. Thus, the material cost of printing a single trajectory guide and a vertebral model was $1.5 to $2 and $10 to $15, respectively. Because we used the MIMICS and 3-MATIC software from the Department of Mechanical Engineering, IIT Delhi, their cost has not been considered in the above-mentioned cost calculation. Many of the comparative studies and meta-analyses have concluded that patient-specific, 3D-printed trajectory guides help reduce the surgical time,[17],[32],[36],[37],[39],[40] radiation exposure,[17],[31],[32],[40] and intra-operative blood loss.[17],[28],[32],[36],[37],[40] A comparative study found that the surgical time and intra-operative blood loss were the least in the 3D template group when compared to robotic and CT navigation techniques.[17] Garg et al.[32] reported significantly less radiation exposure in the 3D-printed group (5.7 fluoroscopic shots/case) when compared with the free-hand technique (11.9 fluoroscopic shots/case). In our study, the overall surgical time (188 ± 28.53 min), radiation exposure (3.47 fluoroscopic shots/case), and intra-operative blood loss (489 ± 155.36 mL), with no significant difference between groups. These findings were comparable to the previous studies; further substantiating the added benefits of 3D-printed trajectory guides. In addition to these benefits, the technique has few inherent limitations. It requires soft-tissue dissection for accurate placement of these trajectory guides. It may take a long time to prepare these guides, which may go up to 12–15 h for patients with pre-existing complex spinal deformities.
Patient-specific, 3D-printed laminofacetal trajectory guide enables a surgeon to place pedicle screws accurately in the subaxial-cervical and thoracic spine. It may additionally help reduce surgical time, radiation exposure, and blood loss. Declaration of patient consent The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed. Acknowledgments The authors would like to thank Dr (Prof) Sunil Jha, IIT, Delhi for providing the access to the MIMICS and 3-MATIC software for the creation of trajectory guides. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3], [Table 4]
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||
