Triangular Fixation to Prevent Proximal Screw Pullout in Adult Spinal Deformity Surgery: A Technical Note

Abstract

time was 3 h 33 min, and estimated blood loss was 320 mL. No postoperative complications or neurological compromise was reported. In terms of clinical outcomes, Oswestry Disability Index improved from 56-24%, and visual analog scale score for lower back pain improved from 62 mm to 24 mm at the 1-year follow-up.

Conclusions
Minimally invasive circumferential surgery with triangular xation is effective for preventing proximal screw back-out and surgical invasiveness. With this new technique, surgeons and operating room staff can avoid the risk of adverse events due to intraoperative radiation exposure.

Background
Adult spinal deformity (ASD) affects a large number of patients (8.9% of individuals >40 years old and 68% of those >60 years old and is caused by spinal malalignment, resulting in severe low back pain, neurological dysfunction, and severe deformity of the body [1,2]. Surgical treatment offers improved patient-reported outcomes compared to conservative treatment [3]. However, the complication rate of ASD surgery is reportedly very high (10-45%) [4,5]. Circumferential minimally invasive surgery (cMIS) to reduce the morbidities and perioperative complications of ASD has been receiving increasing attention [6].
Proximal junctional kyphosis (PJK) occurs at the rostral junction between the xed instrumented spinal segment and overlying mobile vertebral levels after ASD surgery. PJK is de ned by a proximal junctional sagittal Cobb angle ≥10°, and also ≥10° more than the preoperative measurement. If this condition requires revision surgery, it is referred to as proximal junctional failure (PJF) [7]. PJF is one of the most common complications of ASD surgery [8]. To prevent PJK and pullout of proximal screws, options such as taping [9], bone cement [10], administration of parathyroid hormone [11], and semi-rigid junctional xation [12] have been proposed. However, these options are di cult to adapt to cMIS techniques. We present herein a novel technique to prevent screw pullout in adult spinal deformity surgery.

Patient history
A 71-year-old woman was referred to our orthopedic department with severe low back pain and gait di culty. Her daily life had been affected by low back pain for more than 6 months.

Physical examination
The patient could not walk or stand unaided. On examination, the patient could walk for only 100 m due to left leg pain and low back pain ( Figure 1A The patient underwent two-stage C-arm-free cMIS. The rst surgery comprised L1-S1 C-arm-free oblique lumbar interbody fusion (OLIF), with an operation time of 3 h 57 min and an estimated blood loss of 240 mL. After 1 week, the second PPS xation was performed and proximal screws (T10, T11) were inserted with a transdiscal approach (T11) and with lower angulation trajectory (T10) to enhance pullout strength ( Figure 4). For this second stage, operative time was 3 h 33 min and estimated blood loss was 320 mL. No postoperative complications or neurological compromise was reported.

Follow-up results and imaging
Follow-up radiography showed no PJK or screw loosening ( Figure 5). In terms of clinical outcomes, Oswestry Disability Index improved from 56% to 24% and visual analog scale score for neck pain improved from 62 mm to 24 mm by the 1-year follow-up.

Operation procedure
First surgery (L1-S1 OLIF) The patient was placed in the right lateral decubitus position on an adjustable hinged carbon operating table (OSI Axis Jackson table; Mizuho, Union City, CA, USA) to perform CT using an O-arm. An axillary roll was placed to protect the neurovascular structures in the axilla. The patient was secured to the Axis Jackson table with tape, and the table was adjusted to approximately 15° convex. The percutaneous reference frame was attached through the sacroiliac joint. The O-arm was then positioned, and 3dimensional (3D) reconstructed images were obtained and transmitted to the Stealth station navigation system Spine 7 ® (Medtronic Sofamor Danek; Minneapolis, MN, USA). After verifying every navigated spinal instrument, the best entry point for each disc was marked by the navigated pinpoint probe.
Typically, three oblique skin incisions of approximately 4 cm each are necessary for this technique.
The subcutaneous fat layers were dissected until the abdominal musculature was reached. The external, internal, and transverse abdominal muscles were divided parallel with the alignment of the muscle bers to avoid cutting the muscle bers. Both index ngers were inserted inside the retroperitoneal space, and were used to follow the internal abdominal wall posteriorly down to the psoas muscle, which can be visualized. The navigated rst direct lateral dilator rested on the anterior border of the psoas muscle at the L5-S1 disc level. Use of a hand-held retractor with illumination placed between the peritoneal contents and the probe was used to minimize the risk of injury to the ureter and vascular structures anteriorly. The retractor assembly was attached to the exible arm and stability pins were inserted to x the retractor. Discectomy was performed using a bayoneted knife, Kerrison rongeurs, pituitary forceps, a navigated Cobb elevatorium, a navigated shaver ( Figure 5) and navigated curved curettes ( Figure 6A The patient was placed in the prone position on the Axis Jackson table. The reference frame was attached around the T11 spinous process and 3D images were obtained from T10 to L3. After every navigated instrument was veri ed, PPS was inserted by navigation ( Figure 9). The length and diameter of pedicle screw were also measured by navigation. After T10-L3 screws were inserted, the reference frame was reattached to L3, and another 3D image was obtained from L4 to the pelvis. Sacral-alar-iliac (SAI) screws are recommended to enhance the pelvic anchors. The proximal screws (T10, T11) are inserted with a transdiscal approach (T11) ( Figure 10) and with lower angulation trajectory (T10) (Figure 11) to enhance pullout strength. This technique results in triangular xation of the upper instrumented vertebra (UIV) to prevent screw pull-out ( Figure 12).
Anteroposterior and lateral radiograms should be obtained to ensure correct placement of pedicle screws and SAI screws. Rods were bent in an appropriate contour and inserted percutaneously. The Axis Jackson table was bent by more than 20° to create good lumbar lordosis. The set screws were gradually tightened to create a greater amount of lumbar lordosis.

Discussion
For ASD, surgery shows better results than conservative treatment, but is associated with many complications [3]. PJK is one such complication. Risk factors for PJK include older age, greater preoperative sagittal imbalance, large correction of sagittal deformity, anteroposterior spinal fusion, fusion to the sacrum, low bone density, and a high body mass index [14]. PJFs involve more serious clinical problems after long segment fusions in patients with ASD and require surgical treatment. PJF is a progressive form of the PJK spectrum, including vertebral fracture of UIV or UIV+1, subluxation between UIV and UIV+1, failure of xation, and neurological de cits [15]. The prevalence of PJF ranges between 1.4% and 25%, with a mean time of 11.4 weeks to 6 months after primary surgery [15,16]. The structural failure that occurs with PJF can present as vertebral body fracture, implant pull-out or breakage, and/or disruption of the posterior osteo-ligamentous complex [16,17].
Several multifactor-based strategies are available to reduce the incidences of PJK and PJF, such as preserving the interspinous and supraspinous ligaments and facet joint capsules at the UIV. With the advent of percutaneous pedicle screw xation by applying minimally invasive surgery (MIS) principles, the soft tissue injuries and perioperative complications associated with open surgery can be minimized [18]. Osteoporosis is another factor responsible for increased hardware failure, such as screw cutout, screw pull out and higher incidences of PJK/PJF, thus increasing the risk of ASD patients needing revision surgeries [19]. Various techniques have been advised to reduce the chances of screw pull out. Expandable screws offer 60% higher pullout strength than traditional pedicle screws, but their availability and cost remain as important constraints. Moreover, these screws are mainly available for open xations [20]. Bicortical screws are longer and 26% stronger than traditional pedicle screws, but this technique carries a major risk of injury to the great vessels lying anteriorly [21]. Screws with cement augmentation show an 86% improvement in pullout strength. This technique is very commonly used, but at the same time carries risks of cement leakage like pulmonary embolism, radicular symptoms and neurological injury [22].
Techniques that can be used in conjunction with MIS without increasing complication rates or adding to the nancial burdens remain scant.
We have suggested an innovative surgical technique to reduce the chances of screw pull-out and PJK/PJF, in addition to maintaining the advantages of MIS procedures. We suggest inserting a transdiscal screw at the UIV-1 level. Abdu et al. in 1994 offered the rst description of using transdiscal screw xation for the management of spondylolisthesis. A trans-discal screw offers better pullout strength than a straight forward screw, at 1.6-to 1.8-times stronger than a traditional pedicle screw [23]. Inserting a trans-discal screw is avoided during UIV to prevent injury to the proximal junctional disc. Conversely, the UIV screw is placed directed cranio-caudally to get make maximum use of the length of the screw and obtain a triangular construct at the proximal end of the construct.
A triangular construct ( Figure 12) offers superior xation strength, since it depends on the mass of bone between the hardware rather than screw purchase alone [24]. Also, in osteoporotic bone, the insertion angle minimally affects the pull-out force. Insertion angle also has no signi cant effect on insertion torque and is based on the material properties of the bone and the length of screw engagement [25]. Longer screws thus offer better purchase. Another advantage of placing the UIV screw cranio-caudally is that signi cantly more bone is available cranial to the screw surface to resist the screw cut-out. Considering all these advantages to this technique with no added complications or nancial burden and its applicability to MIS, we propose that adoption of similar techniques may signi cantly reduce the rates of complications related to bone-implant interface and thus the incidence of PJF.

Conclusion
Proximal triangular xation is a useful technique to prevent screw pullout for adult spinal deformity surgery. This technique is available for cMIS such as percutaneous pedicle screw xation. The institutional ethics committee provided approval for this study (No.306). Necessary consents were obtained from the patient.

Consent for publication
We, the undersigned, give our consent for the publication of identi able details, which can include gures, tables, case history and details within the text to be published in the Journal of Orthopaedic Surgery and Research.

Availability of data and materials
This study does not contain any third material.

Competing interests
The authors declare that they have no competing interests Funding This research did not receive any speci c grant from funding agencies in the public, commercial, or notfor-pro t sectors.

Authors' contributions
MT performed most of the practical work, analyzed the data and prepared the manuscript. RM participated in the planning of the preparation of the manuscript. SA and YF participated in the data gathering. KS and KY supervised the study planning, data analysis and preparation of the manuscript. All authors read and approved the nal manuscript.      The lower angulation trajectory screw Screws of T11 and T12 makes the triangular xation Figure 10 Percutaneous pedicle screw insertion All screws were inserted under navigation guidance Page 18/18 Figure 11 Triangular xation T10 anatomical screw and T11 transdiscal screw make triangular xation.