Expandable Intravertebral Titanium Implants for Thoracolumbar Burst Fractures Without Neurological Deficits
Article information
Abstract
Objective
A thoracolumbar burst fracture (TLBF) is defined as the failure of the anterior and middle columns of the vertebra due to high-energy trauma, such as motor vehicle collisions and falls from heights. Debate has continued for decades regarding the standard treatment, especially for TLBFs without neurological deficits (TLBF-WONDs). The aim of this study was to understand the role of expandable intravertebral titanium implants (EITIs) in treating TLBF-WONDs.
Methods
We included patients aged 18–65 years who presented at our hospital Emergency Department with severe back pain (visual analogue scale [VAS] score ≥ 8), were neurologically intact, were diagnosed with TLBF-WOND by either computed tomography or magnetic resonance imaging, underwent percutaneous bilateral transpedicular EITI implantation, and were followed-up for ≥12 months. Radiological and clinical outcomes were analyzed.
Results
Thirty patients satisfied the study inclusion criteria, including 9 men and 21 women, with an average age of 48.2 years. Thirteen A3 and 17 A4 burst fractures were included. The mean duration of hospitalization was 3.3 days. The mean follow-up period was 4.4 years. All patients exhibited significant improvements in radiographical (anterior, middle, and posterior vertebral heights); vertebral kyphotic angle (p<0.001); and functional outcomes (VAS and Oswestry Disability Index scores, p<0.001). One case of cement leakage into the paraspinal muscle was observed; however, no major complications occurred.
Conclusion
Percutaneous bilateral transpedicular EITI placement with cement augmentation under local anesthesia may be an effective strategy for the treatment of high-energy traumatic TLBFs with neurological integrity.
INTRODUCTION
Thoracolumbar burst fractures (TLBFs) account for nearly 10%–20% of all spinal fractures due to the failure of the anterior and middle columns caused by nonphysiological axial loading, mainly caused by motor vehicle accidents and high-energy falls [1-3]. Typical radiographical characteristics include loss of vertebral height, kyphosis angle, interpedicular distance widening, and fragment retropulsion into the spinal canal [2]. Indications for conservative and surgical treatments, surgical approaches, instrumentation constructs, the use of cement augmentation and intermediate screws, and the necessity of implant removal are all under debate, especially for TLBFs without neurological deficits (TLBF-WOND), which account for > 30% of TLBF cases [4-7]. Current evidence shows that there is a leading trend toward surgical intervention to prevent pain and loss of activity, enable faster return to work, and improve functional outcomes and quality of life [5]. Furthermore, recent studies have demonstrated that short-segment posterior instrumentation and fixation result in shorter surgical time, less blood loss, and greater preservation of spinal motion compared to long-segment surgeries, which indicates a trend of minimally invasive and motion preservation surgery in the treatment of TLBF-WONDs [4]. Here, we present a case series of percutaneous bilateral transpedicular expandable intravertebral titanium implants (EITIs) used to treat high-energy traumatic TLBF-WONDs under local anesthesia.
MATERIALS AND METHODS
1. Inclusion and Exclusion Criteria
A prospectively designed, retrospective cohort study was performed at a single-level I trauma center. From 2016–2022, patients presented to our Emergency Department with severe and intractable back pain (visual analogue sale [VAS] score ≥ 8), neurologically intact, and diagnosed with an unstable isolated single-level high-energy related traumatic thoracolumbar incomplete or complete burst fracture (Thoracolumbar AOSpine Injury Score [TL-AOSIS], type A3/A4) [8,9] using anteroposterior and lateral radiography and thoracolumbar computed tomography (CT) or magnetic resonance imaging were included. The unstable criteria in this study were defined as: vertebral height loss ≥ 50%, or kyphosis angle ≥ 20°, or spinal canal compromise ≥ 50%. Patients aged <18 years or >65 years with a VAS score for back pain <7 points, vertebral plana, pedicle rupture or pedicle diameter <6 mm, multilevel vertebral injuries, neurological deficits, osteopenia/osteoporosis (dual-energy x-ray absorptiometry T-score <-1.0) [10,11], multiple myeloma, pathological fractures, ongoing cancer or infection, genetic/congenital spinal disorders or developmental malformations, previous history of spine surgery, and follow-up <12 months were excluded from this study. Finally, 30 consecutive patients who met the inclusion criteria were enrolled. Data were collected from our database and analyzed comprehensively. Radiographical parameters included anterior (A-VBH), mid-(M-VBH), and posterior vertebral body height (P-VBH); vertebral kyphosis angle (VKA); local kyphotic angle (LKA); and percentage of posterior fracture fragment retropulsion. The results are shown in Figure 1. Functional outcomes were measured using the pain VAS and Oswestry Disability Index (ODI) scores. Radiographical parameters and functional scores were recorded before and after the surgery. This study was approved by the Institutional Review Board of Taipei Veterans General Hospital (IRB No. 2024-01-011BC).
Radiographic measurements. (A) Average A-VBH of adjacent vertebra=(HA1 + HA2)/2; Average M-VBH of adjacent vertebrae=(HM1 + HM2)/2; Average P-VBH of adjacent vertebrae=(HP1 + HP2)/2; A-VBH loss={HA/[(HA1 + HA2)/2]} × 100%; M-VBH loss={HM/[(HM1 + HM2)/2]} × 100%; P-VBH loss={HP/[(HP1 + HP2)/2]} × 100%. HA, anterior vertebral height; HM, midvertebral height; HP, posterior vertebral height; VKA, vertebral kyphotic angle; LKA, local kyphotic angle. All the above parameters were measured using lateral radiographs of the thoracolumbar spine. (B) Posterior fracture fragment retropulsion (%)=(RF/SC) × 100% measured on the axial cut of computed tomography scans at the injury level. RF, retropulsed fragment; SC, spinal canal.
2. Surgical Techniques and Postoperative Care
All patients were placed in the prone position with pillows supporting the upper chest wall and pelvic regions. All patients underwent surgical procedures under local anesthesia alone (without heavy sedation). The surgical site was sterilized and draped in a standard manner. The surgical level was confirmed under fluoroscopic guidance, and the center of the lateral border of the bilateral pedicle was marked under fluoroscopy. A local anesthetic agent (a mixture of 10 mL 2% Lidocaine [Xylocaine Injection 2%, Recipharm Monts, Monts, France] + 10 mL 0.9% normal saline) was injected around the facet joint (5 mL) and layer-by-layer into the subcutaneous area (5 mL). After the anesthetic status was confirmed, a 3-mm stab wound was made approximately 3 cm away from the spinal process along the middle of the transpedicular line. A 1.6-mm Steinman pin was then used to reach the lateral border of the pedicle center in the anteroposterior view and gradually advanced to the center of the vertebral body in the lateral view. Then, an 11G × 4” disposable bone marrow biopsy needle (T-Lok Bone Marrow Biopsy Needle, Argon Medical Devices, Athens, TX, USA) was slowly twisted through the pin to the middle of the vertebral body. The SpineJack system (Stryker Corp., Kalamazoo, MI, USA) was prepared, sterilized with betadine solution, and applied as follows: the 1.6-mm Steinman pin was replaced with a guidewire, and slowly advanced to the anterior border of vertebral body. The 11G biopsy trocar was first replaced with a reamer, after which it was replaced with the SpineJack template. A 5-mm SpineJack was inserted into the vertebral body. These procedures were performed on the contralateral side. The 2 spine jacks were then expanded simultaneously to restore the vertebral height as much as possible. After the position of the SpineJacks were confirmed by fluoroscopy in both anteroposterior and lateral views, bone cement powder (BonOs Inject Spine Cement, OSARTIS GmbH, Münster, Germany) was mixed with 1 g of Vancomycin (Vanco powder for injection, Gentle Pharmaceutical Corp., Yunlin, Taiwan) and prepared in a standard manner; bilateral cement augmentation was performed through the cement injectors and pusher simultaneously. Cementplasty was terminated when the cement reached the posterior border of the vertebral body or started to leak into the anterior/intervertebral disc/paraspinal vessels. The wounds were closed using 3-0 nylon sutures (one on each side). Bracing was applied for 6–8 weeks (Taylor brace for fractures above L3; chairback brace for fractures at or below L3) depending on radiographical fracture healing. Oral analgesic agents (celecoxib 200 mg twice a day and acetaminophen 500 mg every 6 hours as needed) were administered for 3 days for pain control, if not contraindicated. Each patient had their own chart with detailed records, including personal data, injury mechanism, associated trauma, neurological deficits, implantation details, and functional recovery process. Regular follow-ups were arranged at 2, 6, 12 weeks, 6 months, and 1 year postoperatively for all patients. Radiographs (anteroposterior and lateral views) and wound conditions were evaluated during each outpatient visit.
3. Statistical Analysis
Statistical analyses were performed using IBM SPSS Statistics ver. 24.0 (IBM Co., Armonk, NY, USA). Repeated-measures (RMs) analysis of variance (ANOVA) tests and paired sample t-tests were used. Statistical significance was set at p<0.05 (*p<0.05, **p<0.01, and ***p<0.001).
RESULTS
Patient demographic data are presented in Table 1. The average age of the patients at the time of the injury was 48.2±10.9 years (range, 20–64 years). Nine males and 21 females were included. Mean body weight index was 25.0±3.2 kg/m2 (range, 17.1–32.8 kg/m2). The majority of patients were injured by falling from a high place (83.3%, n=25), followed by motor vehicle collisions (10.0%, n=3) and horse-riding accidents (6.7%, n=2). One T11 (3.3%), seven T12 (23.3%), nine L1 (30%), four L2 (13.3%), six L3 (20%), and three L4 burst fractures (10%) were noted, with 13 of them classified as A3 fractures (43.3%), and 17 classified as A4 fractures (56.7%) based on the TL-AOSIS classification system. Ten percent of the patients had other injuries: one clavicle, finger, and toe fracture, each. The quantity of bone cement used for augmentation was 5.7±2.2 mL (range, 2.5–10.0 mL). Mean surgical time was 36.8±12.6 minutes (range, 20–60 minutes). The mean duration of hospitalization was 3.3±1.7 days (range, 2–8 days). The mean follow-up period was 4.4±2.0 years (range, 1.1–7.6 years). One patient (3.3%) experienced cement leakage without neurological deficits or further surgical intervention. No adjacent fractures, pulmonary embolism, or cement implantation syndrome was observed during the follow-up period. Preoperative and postoperative radiographical parameters and functional outcomes are summarized in Tables 2–4.
Patient demographics and clinical characteristics (N=30)
Comparison between preoperative and postoperative radiographic parameters (by repeated-measured ANOVA test) (N=30)
Comparison between preoperative and postoperative functional outcomes (by paired samples t-test) (N=30)
Comparison between postoperative 1 month and 1 year of restoration of radiographic parameters (by paired samples t-test) (N=30)
DISCUSSION
TLBFs are defined by Denis 3-column theory as failures of the anterior and middle columns of the vertebral body caused by sudden nonphysiological axial loading.1 This mostly affects the population of middle-aged adults and is mainly caused by high-velocity motor vehicle collisions or nonincidental falls from heights, which contribute to 37% and 32% respectively, and type A3 (incomplete burst fracture) was found to be the most common fracture pattern and comprised nearly 40% of all thoracolumbar fractures in a previously published meta-analysis [1-3]. Typical characteristics in radiographs and CT imaging characteristics include significant loss of vertebral height, widening of the interpedicular distance, and retropulsion of the fractured fragment into the spinal canal [2]. The most commonly used classifications are the thoracolumbar injury classification and severity score system and TL-AOSIS scores for thoracolumbar traumas [8,9]. Conservative and surgical treatments have been controversial for decades. Classical surgical indications are the presence of significant progressive neurological deficits and spinal instability [2]. However, in the subgroup of TLBF-WOND, the standard treatment remains unknown. Conservative or operative treatment, surgical approaches, use of fixation constructs, necessity of spinal fusion, and the need for implant removal are still controversial [4]. Two studies previously published by Wood et al. [12,13] described a randomized controlled cohort comparing nonoperative (bracing + body casting for 16 weeks) and operative treatment (anterior/posterior spinal arthrodesis) and showed no advantages of surgical intervention, even at 16 years and 22 years of follow-up. Surgery was also found to be less cost-effective than conservative treatment in other studies [14]. In a recent systematic review and meta-analysis published by Chou et al. [15], no or little difference at 6 months follow-up were observed between the 2 approaches in terms of radiographic kyphotic angle, VAS pain score, ODI score, and Roland Morris Disability Questionnaire results. Nevertheless, careful interpretation should be made in terms of operative treatment because the choice of treatment and goals of spinal surgery are evolving. First, the operative approach presented by Wood et al. [12,13] was anterior/posterior spinal fixation with fusion; however, the necessity of spinal fusion is now doubtful in randomized controlled trials that showed comparable outcomes between fusion and nonfusion surgeries [16]. Moreover, minimally invasive spinal surgery techniques have evolved and become increasingly advanced in many countries [17,18]. Some studies have defended the surgical treatment of TLBF-WOND patients, which yields better radiographical outcomes, more rapid mobilization after spinal column stabilization, and reduction in the severity of sepsis and respiratory failure that potentially decreases morbidity and mortality rates for patients operated within 72 hours after trauma [19,20]. A recent review published by Tanasansomboon et al. [4], under current evidence, advocates that posterior-only short-segment instrumentation (PSSI) plus intermediate screws is the most appropriate treatment for TLBF-WOND patients, which is a less invasive approach, with shorter surgical time, lower blood loss, and preservation of more spinal motion segments. Despite the trend of minimally invasive and motion preservation of the spine, early recovery, ambulation, and returning to daily life or work for patients might be other goals in the current trends of spinal surgery [21].
Furthermore, the standard evaluation of spinal instability, which is an indication for surgery, remains uncertain. The most used TLICS classification system evaluates the fracture pattern, neurological status, and posterior ligamentous complex injury, suggesting surgical intervention when scoring equal or >5 points [8]. However, Woo et al. [22] demonstrated that the TLICS classification system does not exactly match the concepts of instability because in the TLBF-WOND group, all scored only 2 points according to that classification system. Similar to TL-AOSIS scoring, TLBF-WOND scores only 3 points for incomplete burst (conservative treatment) and 5 points for complete burst fractures (conservative or operative treatment) [9]. From this point of view, surgical indication should not only be based on the TLICS or TL-AOSIS scoring systems; other strong indicators suggesting spinal instability should also be considered at the time of evaluation, such as loss of vertebral height > 40%–50%, loss of kyphotic angle > 18.8°–30°, canal compromised >50%, and widening of the interpedicular distance, all of which might provide hints of an unstable spine [23-28]. Therefore, there is a room for research surrounding better treatments or consensus for TLBF-WOND patients among institutions.
Over recent years, several studies have suggested the treatment of TLBF-WOND patients with EITI. Wei et al. [29] reported a case series of 33 patients using EITI to treat Magrel type A3 TLBF-WOND caused by trauma, osteoporosis, and malignancy, which resulted in significant restoration of vertebral height and reduction of the retropulsed fragment. Noriega et al. [30] reported a similar pilot study of 44 patient with AO type A3 fractures under general anesthesia, which are caused by acute trauma, the study showed significant improvements in terms of vertebral height, kyphosis angle, and VAS and ODI scores. Other studies have extended the indication to A2 and A4 fractures. Lofrese et al. [31] reported a retrospective case series of 57 patients with AO type A2, A3, and A4 TLBF-WOND, with etiology caused by medium-to high-energy trauma and treated with EITI; they found that treatment within 7 days of trauma have better wedging corrections and functional outcomes, and identified A2 fractures as a significant risk factor for complications. A more recent study published by Giordan et al. [32] demonstrated comparable clinical and radiographical outcomes between EITI (41 patients) and posterior spinal fusion (54 patients) for AO types A2, A3, and A4 TLBF-WOND, but a significantly shorter operative time and length of hospital stay in the group using EITI. The above studies encourage the use of EITI for the treatment of TLBF-WONDs.
In 2016, our institution started to treat isolated single-level high-energy traumatic TLBF-WOND (TL-AOSIS type A3/A4) by inserting 2 EITI percutaneously through a bilateral transpedicular approach under local anesthesia; all patients had immediate ambulation 3 hours after surgery under the protection of either a chairback or Taylor Knight brace, and were discharged the day after, if no other associated fractures needed to be treated. A successful clinical case diagnosed with A4 burst fracture is shown in Figures 2–5. In the presented study, after operation, statistically significant improvements were found in terms of the anterior, mid, and posterior vertebral heights (A-VBH, M-VBH, P-VBH, A-VBH loss, M-VBH loss, and P-VBH loss, respectively) and VKA, both in postoperative 1 month and postoperative 1 year after the surgery (p<0.001, RM ANOVA test). However, no significant difference was observed in improvement of the LKA (p=0.086) (Figure 6). Moreover, the VAS pain score and ODI score showed statistically significant improvements (p<0.001, paired samples t-test) 1 month after surgery. However, we did find an interesting result when comparing the restoration of radiographic parameters between the group postoperative 1 month and postoperative 1 year: the percentage of restoration of A-VBH and M-VBH was less significant in the group of postoperative postoperative 1 year when compared to postoperative 1 month, indicating that the height of the anterior and middle vertebrae will become somewhat truncated during 1-year follow-up compared to 1-month follow-up, but with no significant changes in terms of restoration of P-VBH, VKA, and LKA. This might be related to the size of the EITI, quantity of bone cement, and fracture level. More studies are needed for further clarification.
Case presentation. A 62-year-old man fell from a high place, presented with severe back pain (visual analogue scale score: 10), was neurologically intact, and was diagnosed with Thoracolumbar AOSpine Injury Score type A4 complete burst fracture at L4 and a left fourth toe fracture. (A, B) Radiographic anteroposterior and lateral views, showing typical radiographic characteristics of collapse of the anterior and middle columns, and widening of the interpedicular distance. (C–E) Sagittal and axial sections of thoracolumbar computed tomography imaging, showing 3-column collapse, with significant loss of the vertebral height, kyphosis angle, and retropulsed fractured fragment into the spinal canal.
(A–E) Intraoperative radiographic images. Restoration of the maximum vertebral height through the expansion of expandable intravertebral titanium implants with bilateral transpedicular cement augmentation.
Clinical images. (A, B) Two expandable intravertebral titanium implants were expanded simultaneously by 2 spinal surgeons with small skin incisions. (C) Immediate ambulation 3 hours after surgery in the ward.
Follow-up radiographs, anteroposterior and lateral views. (A, B) One month after surgery with good implant and cement position. (C, D) Six Months after surgery with the fracture in union.
(A, B) Graphical presentation of preoperative (preop) and postoperative (postop) radiographic improvements after surgery. VBH, vertebral body height; A-VBH, anterior VBH; M-VBH, mid VBH; P-VBH, posterior VBH; VKA, vertebral kyphotic angle; LKA, local kyphotic angle.
Nevertheless, in employing such an approach, the patient avoids posterior spinal muscle damage, sacrifice of adjacent spinal segment motion, risk of degeneration of the adjacent facet joint, facet joint violation by pedicle screw insertion, risk of general anesthesia, and the necessity of subsequent surgery for the removal of implants (pedicle screws and rods), with a short surgical time, minimal blood loss, and reduced hospitalization costs. It also prevents long-term spinal immobilization caused by bed rest, body casting, and wearing orthoses. However, this is an even more minimally invasive approach than PSSI. Patients begin ambulation immediately after surgery and return to their daily activities or work. This might serve as a compromise between traditional and nonoperative treatments.
This study had certain limitations. First, the sample size was small. Second, this was a single-center, retrospective cohort study. Third, this study lacked a control group to compare clinical outcomes between patients who underwent other treatments. Thus, the findings of this study require further verification using larger prospective studies or randomized controlled trials with sufficient statistical power.
CONCLUSION
In conclusion, percutaneous bilateral transpedicular EITI under local anesthesia may represent an effective strategy for the treatment of high-energy traumatic TLBF-WOND patients.
Notes
Conflict of Interest
The authors have nothing to disclose.
Funding/Support
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Acknowledgments
We thank Ms. Yen-Lin Hou of the Graduate Institute of Public Health Section of Public Health Administration of National Defense Medical Center of Taiwan for the data analysis.