In-Fracture Pedicular Screw Placement During Ligamentotaxis Following Traumatic Spine Injuries, a Randomized Clinical Trial on Outcomes

Abstract

Objective

To investigate the efficacy and safety of two different techniques for spinal ligamentotaxis. Spine ligamentotaxis reduces the number of retropulsed bone fragments in the fractured vertebrae. Two different ligamentotaxis techniques require clinical evaluation.

Methods

This was a randomized clinical trial. The case group was defined as one pedicular screw insertion into a fractured vertebra, and the control group as a no-pedicular screw in the index vertebra. Spine biomechanical values were defined as primary outcomes and complications as secondary outcomes.

Results

A total of 105 patients were enrolled; 23 were excluded for multiple reasons, and the remaining were randomly allocated into the case (n=40) and control (n=42) groups. The patients were followed up and analyzed (n=56). The postoperative mid-sagittal diameter of the vertebral canal (MSD), kyphotic deformity correction, and restoration of the anterior height of the fractured vertebrae showed equal results in both groups. Postoperative retropulsion percentage and pain were significantly lower in the case group than in the control group (p=0.003 and p=0.004, respectively). There were no group preferences for early or long-term postoperative complications.

Conclusions

Regarding clinical and imaging properties, inserting one extra pedicular screw in a fractured vertebra during ligamentotaxis results in better retropulsion reduction and lower postoperative pain.

INTRODUCTION

Transpedicular screw (PS) fixation is the gold standard of treatment for spinal fractures and instability. The level of injury, number of fractures, bone density, underlying diseases, postoperative activity expectations, extension of ligamentous injury, and need for advanced or simultaneous spine procedures (e.g., osteotomy and laminectomy) are significant determinants of fusion construct extension and configuration.¹⁹⁾

Regarding neural decompression techniques, the rational approach is that if there are no serious neurologic deficits, the bone fragment retropulsion percent is under 40%–50% of canal volume, and the posterior longitudinal ligament (PLL) is intact, neurosurgeons can reduce the retropulsed fragment via the ligamentotaxis maneuver. Otherwise, retropulsion should be corrected using direct laminectomy-corpectomy and intracanal approaches. There is no consensus on the minimum required amount of retropulsion percentage or other radiological findings to decide whether to decompress the neural elements directly or indirectly. Nevertheless, as much as surgical limitations allow the surgeon to reduce the fragments, it is better to reduce in discretion with the patient's safety and technical challenges.

In the routine ligamentotaxis approach, the vertebrae cephalad and caudal to the instability site were fixed with pedicular screws. The upper and lower vertebrae were distracted from each other using distractors, while tensile forces via PLL extension reduced the fragment out of the spinal canal into the index vertebra. This technique indirectly realigns the bone fragment into the fractured vertebra; therefore, vertebral reshaping is suboptimal. Direct techniques (corpectomy or laminectomy) restore the vertebral canal to its best possible configuration. Disruption of the posterior tension band, increased time consumption, higher blood loss, and extensive neural exposure are the major disadvantages of direct techniques.⁷,²⁵⁾ One modification during ligamentotaxis involves the insertion of an extra screw into the fractured vertebra. In a review of the literature, it has been hypothesized that the pedicular screw can push the bone fragments away from the neural canal and elevate the endplates, thus achieving higher degrees of neural decompression.⁷,⁹⁾ However, the safety and efficacy of these approaches are controversial. Solid clinical trials are required to investigate the safety, efficacy, and clinical outcomes of different techniques during ligamentotaxis in treating spinal fractures.³,²⁴⁾ This randomized clinical trial was designed to address these issues.

MATERIALS AND METHODS

Study design and patients

This study was a single-blinded, randomized, clinical trial with balanced allocation (1:1). It was held in Isfahan, Iran, from July 2020 to March 2022 at Al-Zahra Referral University Hospital.

There are many systems for reporting spinal column fractures. The AO spine classification, Thoracolumbar Injury Classification, and Injury Severity Score System (TLICS) are reliable and clinician-friendly systems for spinal column fractures.⁶,¹²⁾ Inclusion criteria for fusion in spinal fractures were determined based on TLICS/AO scores, patients’ medical condition, neurological status, underlying diseases, acceptable bone density (T score ≥−1.5), and the patient-physician consent reviewed by the neuro spine committee for each patient. Patients with TLICS scores ≥5 and/or mechanically unstable AO subtypes (A4, B, and C subclasses) who could tolerate surgery and had proper medicolegal status were generally scheduled for operation.

The case group was defined by cephalad and caudal fixation plus an extra PS insertion in the index vertebral body. The control group was defined as those with cephalad and caudal fixation without screws in the fractured vertebra.

All patients who met the criteria for spine fusion using the ligamentotaxis technique were enrolled (FIGURE 1). Exclusion criteria were: refusal to participate, terminal stage medical conditions, neurological deterioration requiring other types of spine surgeries, severe osteoporosis (T-score <−2.5) for any reason, and mortality before the procedure. The remaining patients were randomly allocated to the case (n=40) and control (n=42) groups. The randomization sequence was generated using the block randomization technique using Stata 9^® software (StataCorp LP, College Station, TX, USA).

FIGURE 1. Flow diagram summarizing the study in a step-by-step fashion.

n: number of patients.

Operation

Under general anesthesia and in the prone position, after prep and drape in a sterile fashion, using C-arm navigation, the level of fracture was marked, paravertebral muscles were stripped off, and all gross anatomical and pathological findings (posterior ligamentous complex [PLC] and bony element integrity status, hematoma, and extension of spine injury beyond the expected level) were documented by the circulating nurse or residents. Using C-arm fluoroscopy, pedicular screws were inserted into the vertebral bodies using a freehand technique. Depending on the allocation (case or control group), one extra-pedicular screw was inserted into the fractured vertebral level. The surgeon had to choose an intact pedicle for in-fracture PS insertion, and there was no left- or right-side priority. In cases of bilateral pedicular fractures or poor screw-grip strength during the intraoperative examination, the screw was removed, and the patient was excluded from the case group (n=4).

The final fusion construct properties (overall number of polyaxial pedicular screws, screw diameter, screw length, connecting rod diameter, and rod curvature) were determined based on multiple factors. The fracture level, fracture morphology, PLC integrity, bone density status, and postoperative rehabilitation expectations of patients were the most important determinants. It is important to emphasize that the study was designed for fracture-zone instrumentation properties and not for overall fusion construct features, as discussed above.

After constructing the assembly, the surgeon extensively decorticated the posterior bony structures (the lamina, spinous, and transverse processes). A generous mixture of allograft bone graft–bone marrow aspiration contents combined with 500 mg of vancomycin powder was spread over the bony surfaces and all over the construct. To prevent proximal junctional kyphosis (PJK) and improve osteogenesis, we extended decortication to two levels above the construct as an extra effort. We placed a considerable number of bone grafts cephalad to the decorticated surfaces. Depending on the construct level, 1–2 drains were inserted submuscularly, and then paravertebral muscles, fascia, subcutaneous tissues, and skin were repaired in anatomic layers.

Postoperative period

Patients were transferred to intensive care units (ICU) or neurosurgery wards according to their medical status. Routine postoperative orders were followed, and the patients were discharged after 2–3 days. All the patients underwent the same analgesic protocol. The postoperative visual analog score (VAS) of the patients was documented 48 hours postoperatively. Routine postoperative spine multidetector computed tomography (MDCT) was obtained from all patients, and the values (postoperative retropulsion percentage, screw position, axial anteroposterior [AP] diameter at the fractured level, and correction ratio) were measured and documented by spine fellows.

An independent spine fellow and a qualified neuroradiologist evaluated and assessed the pre- and postoperative spine images. The inter-rater reliability was calculated using the Percent Agreement for Two Raters. Negligible value differences were defined as 1 mm for 2-dimensional data, 2 mL for volumetric values, and 2° for angle calculations.

The neuroradiologist performed volumetric analysis of the intracanalicular lesions (retropulsion percent, hematoma, or space-occupying lesions) using Materlised Mimics® digital software in the Digital Imaging and Communications in Medicine (DICOM) interface.

Kyphotic angulation was defined as a kyphotic deformity at the fracture level, usually caused by a collapse of the anterior column height. It was divided into two categories: non-significant (<5° angulation) and clinically significant (>5° angulation). PJK was defined as more than 10° angulation deformity compared to the mean standard Cobb’s angle values in the operated zone at least 12 weeks after surgery. PJK and sagittal balance variables were measured by using a long-standing cassette. Screw pseudoarthrosis was defined as radiolucencies around screws after three months of follow-up MDCT scans, which could be only radiologically apparent or clinically symptomatic. All the patients received prosthetic thoracic, thoracolumbar, or lumbar rigid orthoses for at least 3–4 months.

Follow-up

After 2 weeks, the patients were admitted to the hospital’s neurospine clinic. After 3 months, a routine low-dose spine MDCT scan and long-standing cassette were obtained from all patients to evaluate the fusion rate or long-term complications of the surgery, such as pseudoarthrosis, proximal junctional kyphosis, adjacent segment deformities, and long-term sagittal balance parameters. To determine the ultimate PJK status, we rechecked the patients 12 months postoperatively. All patients were followed-up until August 2022. The patients’ demographic and medical data, laboratory results, long-standing cassette values, spine MDCT scans, magnetic resonance imaging (MRI) findings, AO/TLICS scores, VAS and all related medical records were documented by the surgical team (nurses, residents, fellows, and attending physicians).

Outcome measures

The degree of retropulsion correction, mid-sagittal AP diameter (MSD), anterior height of the fractured vertebra, postoperative VAS score, postoperative kyphotic deformity correction, and early postoperative neurological status were defined as primary outcomes. The long-term neurological status of patients and surgical complications were determined as secondary outcomes.

Ethics approval

All procedures were performed under the institutional and/or national research committee’s ethical standards, including the 1964 Helsinki Declaration and later amendments or comparable ethical standards. The board members of the Isfahan University neurosurgery department supervised and approved this report on behalf of the Ethical Committee of Isfahan University of Medical Sciences (IR.MUI.MED.REC.1399.1185).

RESULTS

A total of 56 patients were included in the final stage of the study (case: n=18, 32.1%; control: n=38 patients, 67.9%). The patients’ mean age was 43.6±13.42 (range: 19–68 years), 71.4% were males, and 28.6% were females.

Statistical analysis showed that the case and control groups were homogeneous in terms of age, sex, and level of fracture undergoing surgery (p=0167, p=0.928, and p=0.95, respectively) (TABLE 1).

TABLE 1. Demographic preview of the patients.

Variables		All the patients (n=56)	Groups		p-value
			Control (n=38)	Case (n=18)
Age (years)		43.66±13.42	47.27±12.6	41.94±13.62	0.167
Sex					0.928
	Male	40 (71.4)	27 (71.1)	13 (72.2)
	Female	16 (28.6)	11 (28.9)	5 (27.8)
Level of the major injury					0.950
	L1	7 (12.5)	5 (13.2)	2 (11.1)
	L2	8 (14.3)	5 (13.2)	3 (16.7)
	L3	7 (12.5)	5 (13.2)	2 (11.1)
	L4	5 (8.9)	3 (13.2)	2 (11.1)
	L5	2 (3.6)	2 (5.3)	0 (0.0)
	T5	2 (3.6)	1 (2.6)	1 (5.6)
	T7	2 (3.6)	1 (2.6)	1 (5.6)
	T8	3 (5.4)	2 (5.3)	1 (5.6)
	T9	3 (5.4)	3 (7.9)	0 (0.0)
	T10	5 (8.9)	2 (5.3)	3 (16.7)
	T11	5 (8.9)	5 (10.5)	1 (5.6)
	T12	7 (12.5)	5 (13.2)	2 (11.1)

Values are presented as mean ± standard deviation or number (%).

All patients had an intact neurological status pre- and postoperatively (American Spinal Injury Association [ASIA] score E). The mean pre- and postoperative retropulsion percents were 37.76±10.39 and 8.5446±5.20, respectively. Postoperative retropulsion percent was significantly lower in the case group (p=0.003).

The mean values for kyphotic angle were 12°±6.58° and 11.48°±7.19° in preoperative assessment and 1.18°±4.87° and 0.17°±4.69° in postoperative surgeries in the case and control groups, respectively (p=0.602 and p=0.460, respectively). Regarding kyphotic deformity correction, both the case and control groups had effectively redacted kyphotic deformities and restored sagittal balance (p=0.001; TABLES 2 & 3).

TABLE 2. Pre- and postoperative mean values in case and control groups.

Quantitative variables		In all patients (n=56)	Groups		p-value (between groups)
			Control (n=38)	Case (n=18)
Kyphotic angle
	Preoperative		11.48±7.19	12.00±6.58	0.602
	Postoperative		0.17±0.469	1.18±0.487	0.460
		p-value (within group)	0.001	0.001
Anterior column height
	Preoperative	13.77±3.17	13.70±4.42	13.8±2.44	0.921
	Postoperative	18.19±4.04	18.23±4.35	18.16±3.95	0.954
		p-value (within group)	0.000	0.019
MSD
	Preoperative	9.14±1.27	9±1.05	9.21±1.37	0.555
	Postoperative	14.09±1.61	14.22±1.37	14.03±1.73	0.690
		p-value (within group)	0.032	0.099
Retropulsion percent
	Preoperative	37.12±11.3	33±10.88	39.07±11.14	0.060
	Postoperative	8.54±5.20	5.66±4.82	9.90±4.85	0.003
		p-value (within group)	0.0001	0.038
VAS score
	Preoperative	6.76±0.99	7±1.08	6.65±0.93	0.231
	Postoperative	3.46±0.57	3.77±0.64	3.31±0.47	0.004
		p-value (within group)	0.288	0.083

Values are presented as mean ± standard deviation.

MSD: mid-sagittal anteroposterior diameter, VAS: visual analog score.

TABLE 3. Detailed values for surgical and imaging results.

Variables		No.	Mean	SD	Minimum	Maximum	ANOVA p-value (between groups)
Preoperative anterior column height (mm)							0.921
	Case	18	13.7083	4.42700	1.80	19.60
	Control	38	13.8000	2.44231	9.60	18.20
	Total	56	13.7705	3.17368	1.80	19.60
Postoperative anterior column height (mm)							0.556
	Case	18	18.8311	4.29134	11.39	25.20
	Control	38	18.1429	3.94909	11.80	30.20
	Total	56	18.3641	4.03592	11.39	30.20
Preoperative MSD (mm)							0.555
	Case	18	9.0000	1.05370	6.80	11.00
	Control	38	9.2184	1.37838	6.50	12.00
	Total	56	9.1482	1.27746	6.50	12.00
Postoperative MSD (mm)							0.690
	Case	18	14.2250	1.37384	11.25	16.20
	Control	38	14.0382	1.73195	7.80	17.10
	Total	56	14.0982	1.61527	7.80	17.10
Retropulsion reduction rate							0.943
	Case	18	29.3333	6.48981	15.00	40.00
	Control	38	29.1711	8.52839	15.00	45.00
	Total	56	29.2232	7.87107	15.00	45.00
Preoperative retropulsion percent (%)							0.173
	Case	18	35.0000	8.22478	20.00	45.00
	Control	38	39.0789	11.14131	20.00	65.00
	Total	56	37.7679	10.39754	20.00	65.00
Postoperative retropulsion percent (%)							0.003
	Case	18	5.6667	4.82640	0.00	20.00
	Control	38	9.9079	4.85444	5.00	25.00
	Total	56	8.5446	5.20077	0.00	25.00
Kyphotic angle reduction rate							0.797
	Case	18	12.0000	6.57983	−5.00	18.00
	Control	38	11.4816	7.18827	−10.00	23.00
	Total	56	11.6482	6.94278	−10.00	23.00
Preoperative VAS score							0.231
	Case	18	7.0000	1.08465	5.00	9.00
	Control	38	6.6579	0.93798	5.00	9.00
	Total	56	6.7679	0.99070	5.00	9.00
Postoperative VAS score							0.004
	Case	18	3.7778	0.64676	3.00	5.00
	Control	38	3.3158	0.47107	3.00	4.00
	Total	56	3.4643	0.57094	3.00	5.00
Follow up period (months)							0.248
	Case	18	19.3333	1.45521	18.00	23.00
	Control	38	19.9211	1.87993	18.00	24.00
	Total	56	19.7321	1.76317	18.00	24.00

SD: standard deviation, ANOVA: analysis of variance, MSD: mid-sagittal anteroposterior diameter, VAS: visual analog score, No.: number of patients.

The mean values for the anterior height of the fractured vertebra in the case and control groups were 13.70±4.42 mm and 13.80±2.44 mm in the preoperative and 18.83±4.29 mm and 18.14±3.95 mm in postoperative assessments. None of them were superior to the others (p=0.921 and p=0.556, respectively). The anterior column was effectively reconstructed in both groups (p=0.000).

The mean values for the MSD of the fractured body in the case and control groups were 9±1.05 mm and 9.22±1.38 mm in the preoperative imaging workup and 14.22±1.37 mm and 14.04±1.73 mm in the postoperative cases, showing equal efficacy (p=0.555 and p=0.690, respectively). Regarding the MSD of the vertebral reduction ratio, both the case and control groups had significantly decompressed intracanalicular space (p=0.001; TABLE 2).

Regarding pre- and postoperative VAS scores, the median VAS scores in both groups were 7 and 3, respectively, which showed statistically significant superiority in the case group compared to the control group (p=0.004, TABLE 3). It is important to mention that the VAS score was defined as general back or dorsal pain and was not localized to a single level.

FIGURES 2, 3, 4 display pre- and postoperative results of posterior-only ligamentotaxis. In some patients, the retropulsed fragments were entirely reduced, while the reduction ratio was less achievable in more compound fractures. The mean follow-up period was 19 months (18–24 months).

FIGURE 2. Pre- and postoperative images of a 37-year-old male presented with L4 burst fracture and operative as the control group. despite of severe canal compromise (A-C, retropulsion percent=90%) the patient was neurological intact (ASIA score E). Postoperative images shows 40% reduction ratio in retropulsion(D, E). The patient was discharged with intact neurological status (ASIA score E).

ASIA: American Spinal Injury Association.

FIGURE 3. The 45-year-old male presented with T11 fracture with intact neurological status which operated in case groups with one extra-pedicular screw in T11. Pre- and postoperative retropulsion per cents were 40% and 2%, respectively. The patient was discharged with intact neurological status.

FIGURE 4. Two cases (A, B) demonstrate a successful retropulsion reduction while preserving posterior elements.

DISCUSSION

Regarding the severity of spinal trauma, bone fragments often intrude into the spinal canal following vertebral fractures, predisposing patients to neurological deficits. There are 2 methods to remodel the intracanal space and reduce the retropulsed bone fragment: direct and indirect. Direct techniques include laminectomy and corpectomy to decompress neural tissues. This technique is associated with significant bleeding, prolonged operation duration, potential intraoperative challenges for the medical team and patient, longer hospital stays, and increased health system costs.²,⁴,¹⁴,²⁰⁾ Ligamentotaxis can indirectly reduce retropulsed bone fragments and resolve the compressive effect on the spinal canal through posterior distraction in the presence of an intact PLL. One PS was inserted in a vertebra above (cephalad) and one PS in the lower vertebral body (caudal) during ligamentotaxis. The neurosurgeon attempted to reduce the retropulsed fragment with distraction forces and indirectly decompress the spinal canal without further laminectomy. Ligamentotaxis appears to be a fast and effective method for patients with spinal fractures. This technique is better to be selected for patients with intact neurological status, canal compromise <50%, and, most importantly, intact PLL; however, none of these criteria are absolute and require multiple randomized clinical trials (RCTs) to establish the evidence.¹⁶,¹⁷⁾

The efficacy of canal decompression is classically assessed by comparing pre- and postoperative MSD or retropulsion values and, most importantly, neurological outcomes. There are no published studies on ligamentotaxis and reduction in retropulsed bone fragments without laminectomy or intracanal space remodeling in RCTs.

MSD is a mono-dimensional spine parameter that only provides data on a single AP diameter of the vertebral canal. Simultaneously, the retropulsion percentage is assessed using a three-dimensional volumetric calculation. Regarding the equal efficacy of both case and control groups in MSD values despite different postoperative retropulsion results, it could imply that retropulsion percentage is a better indicator of intracanal decompression than MSD (TABLE 2).

The authors hypothesized that one extra pedicular screw in a fractured vertebra (case groups) could reduce the retropulsed fragment into the index vertebra and elevate the superior end plate, thus reconstructing the anterior column and restoring a more biomechanically stable vertebral column alignment. While all the patients operated on were considered to have ligamentotaxis, the study was divided into case and control groups. The statistical analysis confirmed the superiority of the in-fracture screw (case) in comparison to the no-screw (control) in the fractured vertebral body in reducing the postoperative retropulsion percentage (p=0.003). This is of great value for clinical situations requiring further correction of retropulsed fragments while preserving posterior elements for better bony fusion to prevent unfavorable long-term outcomes.

Acetaminophen, nonsteroidal anti-inflammatory drugs, and opioids are the most commonly used postoperative pain medications following spine surgery. In terms of postoperative pain and spine surgery, the literature recommends a combination of oral and parenteral drugs to minimize pain, increase mobilization, and improve quality of life. However, drug side effects, especially in older adults, can interfere with the optimal benefits of analgesics.¹,²¹⁾ To date, no study has compared postoperative pain related to different surgical techniques for ligamentotaxis following spinal injury. Our study showed that the postoperative VAS score in the case group was significantly lower than that in the control group (p=0.004). To our knowledge, there is no well-delineated evidence explaining this finding, but the authors hypothesize that stronger solid fixation via 1-extra PS, further reduction, neural decompression, and nociceptive-ablative procedures during PS placement at the fracture site could be explanations. The authors would like to encourage further research from this perspective in future studies.

Lee et al. reported that the prevalence of PJK in posterior fixation surgeries ranges from 6%–41%.¹⁵⁾ In the literature review, the associated risk factors were spinopelvic fixation, osteoporosis, female sex, combined anterior-posterior approaches, bisphosphonate consumption, higher deformity correction, and residual sagittal imbalance. In the literature review, most studies reported degenerative spinal deformities. Trauma, which is the primary study population, has been less investigated. The use of hooks, wires, cement augmentation, and other types of instrumentation have been reviewed. However, none of these are absolute measurements to reduce the incidence of PJK.⁸,¹⁰,¹⁸⁾ Laminectomy, facet damage, and surgical resection of PLC compromise the posterior stabilizing elements of the spine, resulting in a higher risk of PJK development. Yin et al.²⁴⁾ conducted an RCT on injured vertebral pedicle instrumentation and cross-segment pedicle instrumentation and found that in-fracture PS insertion can restore vertebral height but has no preventive effect on long-term ASD.

In this RCT, the authors tried to preserve the natural stabilizing spine elements (lamina, spinous processes, and PLC) and restore the anterior column as much as possible to prevent possible future deformity, thereby lowering the chances of PJK occurrence. Conversely, preservation of posterior bony elements, extensive decortication, and the application of generous amounts of bone grafts on fusion constructs are considered to help promote osteoblastic and osteoinductive activities over the fusion construct, thus achieving a more solid bone-metal integration over the injured segment. These advantages can be achieved merely by posterior-only ligamentotaxis without further laminectomy or osteotomy. There was one case of screw pull-out (1.7% and after four months follow in the control group) in the lowest level of construct in follow-up images, which was revised to 5.4% radiologic screw pseudoarthrosis, and three cases of new in-construct kyphosis at the level of fracture (5.4%, kyphotic angulation, range=1°–3°). None of the patients were clinically significant and were observed conservatively during the follow-up period (p=0.855).

It is worth mentioning that the prevalence of nonsignificant kyphosis in both groups was approximately 5%. The prevalence of radiologic pseudoarthrosis and screw loosening was approximately 5.3% in both groups. Although this difference was not statistically significant (p=0.855), the clinical perspective provides excellent information that the presence of pseudoarthrosis can predict or accompany the emergence of a new kyphosis deformity (TABLE 4).

TABLE 4. Surgery associated outcomes in short and long-term follow up (mean 19 months).

Variables	Events	All the patients (n=56)	Control (n=38)	Case (n=18)	p-value
Direct instrument related events	Acceptable screw	53 (94.6)	36 (94.7)	17 (94.4)	0.964
	Pedicle medial breach	3 (5.4)	2 (5.3)	1 (5.6)
Long term complications	Screw pseudoarthrosis	6 (10.7)	2 (11.1)	4 (10.5)	0.855
	PJK	2 (3.6)	1 (5.6)	1 (2.6)
	Non	48 (85.7)	33 (86.8)	15 (83.3)

Values are presented as number (%).

PJK: proximal junctional kyphosis.

Despite the study by Yin et al.,²⁴⁾ our late-onset postoperative complications, including PJK, adjacent segment deformity (ASD), osteodiscitis, and wound infections, follow-up evaluations showed no considerable complications.²⁴⁾ These results are significantly more promising than the high incidence of PJK reported in previous studies. These results carry an important clinical message: if the surgeon saves the bony bed and other posterior elements of the spinal column, bony fusion will be much more solid; thus, long-term complications such as PJK and ASD are less frequently seen.

In a study of 42 patients, Xue et al.²³⁾ found no cases of treatment failure during the 28-month follow-up period. In this study, the kyphosis correction percentage was 74.3%, and there was a significant improvement in the mean vertebral height at the end of the follow-up period compared to pre-treatment values.²³⁾

The present study showed that ligamentotaxis could significantly improve anterior column reconstruction, so the percentage of anterior column height correction reached approximately 38% compared to preoperative values (p=0.000). The results of the present study showed that ligamentotaxis led to an improvement in the MSD, so that mean values were reduced from 9.1482±1.27 mm in the preoperative phase to 14.09±1.61 mm in the postoperative evaluation, reflecting an effective expansion of the AP diameter of the vertebral canal following surgery (p=0.001).

The correlation between residual postoperative retropulsion and postoperative neurological status is controversial; however, most studies have reported no association between postoperative bone fragment residues and neurological status. Reportedly, the remaining bony fragments will be resolved and remodeled over time.⁵,¹³,²²⁾

Mueller et al.¹⁷⁾ reported a series of 36 patients who underwent ligamentotaxis for spinal fractures. They reported a 10% correction in retropulsion percentage after the operation and improved MSD by approximately 7%.¹⁷⁾ In the current study, the maximum value of canal compromise was 65%, which is relatively high compared to the literature. The correction ratios of retropulsion and MSD in postoperative MDCT were 29% and 55% in both groups, demonstrating considerable effectiveness of the operation in both groups compared to the preoperative status of the mentioned parameters (p=0.000).

Notably, Mueller et al.¹⁷⁾ performed anterior corpectomy in some cases, which ultimately resulted in the best odds of clinical outcomes and statistical results for pure ligamentotaxis without further procedures. Lower operative duration, lower blood loss, and a lower overall risk of surgery are some of the essential benefits of posterior-only ligamentotaxis. Laminectomy or corpectomy, although indicated and necessary, completely confounds the results of a pure posterior-only surgical approach.¹⁶⁾ In this study, posterior-only ligamentotaxis were performed without laminectomy or osteotomies. Our results are even more promising than those reported by Mueller et al.¹⁷⁾

It is worth mentioning that Hu et.al are currently conducting an RCT on modified pedicle screw placement at the fracture level for the treatment of thoracolumbar burst fractures.¹¹⁾ The upcoming results could potentially enlighten current research obscurities.

This study was conducted during the coronavirus disease 2019 pandemic in Iran, which limits the total number of cases. The drop rate after randomization in the case group caused an unbalanced distribution, limiting the generalizability of the study. We used a 1.5 Tesla MRI machine with a low resolution of PLL injuries, limiting PLL tension. This can lead to a lower reduction ratio after distraction. A higher-resolution MRI (3 T<) is desirable. The authors could not completely match the TLICS/AO classification grades for each case-control pair; however, these scores had no proven effect on the surgical planning.

CONCLUSION

In terms of post-traumatic kyphotic deformity correction, retropulsion reduction ratio, vertebral canal decompression, anterior column reconstruction, and posterior-only ligamentotaxis in patients with intact PLL and normal neurological status (ASIA score E), regardless of whether in-fracture pedicular screw insertion is safe and effective.

The postoperative retropulsion percentage and postoperative VAS scores were lower in the case group. Long-term complications of spinal surgery, such as PJK, ASD, and infectious complications, have not been reported.