Evaluation of minimally invasive percutaneous transpedicular screw fixation as an evolving modality for the treatment of thoracolumbar fractures

Background Percutaneous transpedicular screw fixation was recently introduced as a minimally invasive approach for the treatment of thoracolumbar spine fracture. This approach is superior to the traditional open surgical approaches in decreasing intraoperative blood loss, infection rates, and paraspinal muscle damage with less hospital stay. Objective This study aims to report our surgical experience, the feasibility and safety of percutaneous transpedicular screw fixation of thoracolumbar spine fractures. Method we retrospectively reviewed the charts of all adult patients with thoracolumbar spine fractures who underwent percutaneous transpedicular screw fixation during the period between January 2020 to January 2023 at Ain Shams University Hospitals and Prince Mohamed Ibn Abdelaziz Hospital. Collected data included: demographics, pre-and postoperative neurological assessment, radiological data including type of fracture and cobb’s angle pre-and postoperatively, operative time and amount of intraoperative blood loss. Results Forty-three percutaneous pedicle screw fixation procedures were done for 42 patients. Thirty patients were males. Dorsal spine fractures accounted for 18 cases. Preoperative American spinal injury Association score was E in 37 cases. Mean operative time was 133 min (± 46.5), and mean intraoperative blood loss was 107.1 ml (± 84). The mean follow-up period was 8 months, and the average hospital stay was 7.7 days. Radiological complete bony fusion was achieved in 77% of fractures, and 93% had improved kyphotic angle with a mean correction degree of 7.9 (± 9). Conclusion Percutaneous transpedicular screw fixation is a safe and effective treatment option for the management of thoracolumbar spine fractures with or without spine decompression that provides satisfactory functional outcomes.


Introduction
Each year, there are approximately 5 million new vertebral fractures recorded worldwide.In one epidemiologic study by Hu et al. [1], thoracolumbar fractures accounted for two thirds of the total traumatic injuries of the spinal skeleton.These types of fractures significantly influence the patients, social and financial environment more than other injuries and are highly debatable in terms of management [2].
The decision for open surgical versus minimally invasive treatment of thoracolumbar spine fractures is based on numerous factors like fracture type classification, neurological deficit, general condition of the patient, and associated injuries [3].The traditional open surgical approach (posterior, anterior, or both) is usually associated with satisfactory clinical outcomes, with proper fracture reduction and deformity correction [4].However, it still holds the risk of higher intraoperative blood loss, infection rates, and paraspinal muscle damage with prolonged overall operative time and hospital stays [5,6].
Minimally invasive percutaneous transpedicular screw (MIPS) fixation spine surgeries have been gaining popularity for treating thoracolumbar fractures through minimizing approach-related complications with reduced length of hospital stay, blood loss, and requirements for postoperative analgesia and earlier return to work with similar clinical and radiological results compared to open methods [7][8][9].
Case selection is a critical factor in opting for MIPS fixation to avoid the risk of a postoperative deformity or lifelong kyphosis [10].Literature consensuses suggest that conservative treatment is not advisable in type A3 burst fractures [AO-Magerl classification] [3], young active individuals avoiding prolonged work absence, obesity, polytrauma, and previous deep venous thrombosis (DVT) patients.At the same time, posterior open arthrodesis would be considered an overtreatment.In this setting, a minimally invasive approach would limit overall morbidities and hasten functional recovery, rendering these patients' good candidates for MIPS fixation [2,4].
This study aims to report our surgical experience and the feasibility and safety of thoracolumbar spine fracture MIPS fixation.

Patients and methods
This study was approved by the Ethical Committee of Scientific Research, Faculty of Medicine, Ain Shams University.Then, the study protocol and the ethical approval were approved by the council of Neurosurgery department PMAH hospital.The authors of this retrospective hospital-based study reviewed the medical records of 42 adult patients with thoracolumbar fractures.All patients were operated on in Ain Shams University Hospitals (ASUH), Cairo, Egypt, and Prince Mohamed Ibn Abdelaziz Hospital (PMAH), Riyadh, KSA, in the period between January 2020 and January 2023.We included records for all adult patients with thoracolumbar fractures and thoracolumbar injury classification and severity (TLICS) score ≥ 4 that were operated with percutaneous fixation technique during our study period including their age, sex, type of fracture and neurological status.All included patients had at least 6 months of available follow-up data.Pre-and postoperative radiological images for all patients were analyzed by an independent neuroradiologist.

Preoperative assessment
All patients included in the study were subjected to thorough neurological assessment preoperatively and all were clinically classified according to the American Spinal Injury Association (ASIA) Impairment Scale.Polytraumatized patients were all assessed for associated injuries.
All patients had a preoperative computerized tomographic scan (CAT scan) together with magnetic resonance imaging (MRI) of the thoracolumbar spine.Fracture levels were recorded and were all categorized according to both AO-Magerl classification and Thoracolumbar injury classification and severity score (TLICS).

Postoperative assessment
The authors evaluated the available clinical and radiological data immediately postoperatively, at three months and again after six months.
Clinical evaluation • Neurological status: determined by measuring immediate postoperative American Spinal Injury Association (ASIA) score.• Postoperative wound cosmesis through visual analog scales (VAS) evaluated by the patient.A scale from 1 to 5 was considered unsatisfactory, six and seven were considered fair, while a scale from 8 to 10 was considered good cosmetically.• Postoperative infection was recorded if occurred.

Radiographic evaluation
• All patients had plain X-ray radiography preoperatively and immediately postoperative.• CAT scans were done for all patients preoperatively, immediately postoperatively for screw position verification and pedicle violations, and again at three and six months to monitor fusion.• The degree of kyphosis according to the difference in the mean Cobb's angle preoperatively and six months postoperatively and comparing the average correction obtained by the technique.

Surgical protocol
Three types of percutaneous fixation systems were utilized through all of our patients included in this study.Thirty-five fractures were treated with the CD Horizon Longitude TM system (Medtronic-2600 Sofamor Danek Drive, Memphis, TN 38132).VIPER PRIME ® (DePuy Synthes Spine, Raynham, MA, USA) was utilized in six fractures and ES2 system (Stryker, Allendale, New Jersey, USA) in the last two fractures.All procedures were performed under general anesthesia with patients placed prone on a lumbar frame/Jackson table.Entry points were identified under fluoroscopic guidance where the tip of the tap (Jamshidi) needle was placed over the lateral margin of the pedicle and the needle shaft then aligned parallel to the end plate with lateral to medial trajectory 15 degrees.The needle was inserted gently till it tapped the pedicle and then advanced 1 to 1.5 cm deeper through the vertebral body.The trocar was removed from the needle, and a guide wire was inserted through the needle.The same procedure was repeated for all pedicles in the segment planned for fixation.The needle was removed, and a 1.5-cm skin incision was done over the guidewire.The fascia and muscles were then dilated to allow for screw placement using three dilators.The first two dilators were removed, leaving the third dilator to serve as a tissue protection sleeve during the taping step.The pedicle was then tapped by placing the tap over the guide wire and through the third dilator guided by the C-arm image.In VIPER PRIME ® (DePuy Synthes Spine, Raynham, MA, USA), no need to use Jamshidi needle except in case of small pedicles only.
Screw placement After tapping, the dilator was removed, and a cannulated pedicle screw was placed over the guidewire at each level and advanced through the pedicle into the bodies then the guidewires were removed.
Rod placement Rods of suitable lengths and shape were prepared.The rod inserter was attached to the screw extenders and the rod was inserted percutaneously through a small incision created cranially or caudally (in longitude system only, meanwhile in ES2 and VIPER PRIME ® systems the rod was introduced through the upper or lower screw entry point).The rod was then situated and passed through the screw head extenders into the screw crowns, and then the screw caps were inserted into each screw to capture the rod.Distraction or compression of the system was performed when required followed by final tightening of the screw caps and then skin closure (Fig. 1).
The short segment fixation (one level above and one below the fractured vertebra including unilateral or bilateral index screws) was done in 23 fractures as follows: 7 patients with 4 screws, 2 patients with 5 screws (unilateral index screw) and 14 patients with 6 screws (bilateral index screws).Long segment fixation was performed in 20 fractures.Overall, a total number of 282 screws were implanted with a percutaneous technique utilizing multiple paramedian incisions only in 38 fractures, coupled with a small midline incision for laminectomy in five fractures (Fig. 2).The midline skin incision was used for laminectomy and decompression in cases with neurological deficits.

Statistical analysis
Data analysis was performed using Social Science Statistics (SSS).Categorical descriptive data are presented as means and standard deviations (SDs), while median and interquartile ranges for non-parametric continuous data.

Patient demographics and fractures analysis
Our study included a total of 42 patients (30 males and 12 females), sixteen patients underwent fixation in the thoracic spine (from D3 to D12) and 25 in the lumbar spine (from L1 to L4).One of the patients had two distant thoracolumbar burst fractures, one in the L1 vertebra and the other in the D6 vertebra, i.e., we analyzed a total of 43 fractures in our 42 patients.
The most affected location was the thoracolumbar junction (T11-L2) (no = 33; 76.7%).Based on the AO-Magerl classification, most cases were type A fractures (A1 and A3), with only a few cases being type B or type C. All cases were grade 4 and grade 5 according to TLICS classification except for 4 patients that were grade 7.According to the American Spinal Injury Association (ASIA) Impairment Scale [11], all patients were neurologically free (ASIA grade E) preoperatively except for five cases; two were (ASIA grade A) one with sphincteric affection (conus medullaris contusion) and one paraplegic, respectively.Two were (ASIA grade C), and one had (ASIA grade D) with root affection (from L1 to L4). (Table 1).

Perioperative results
None of our cases showed any neurological deterioration postoperatively.All our patients who were neurologically intact preoperatively ASIA grade E (37/ 42 patients) maintained their normal neurological status postoperatively.One out of the 5 patients who had neurological deficits preoperatively had mild improvement of his ASIA score from grade D (4 root values) to grade D (3 root values) in the immediate postoperative follow-up and maintained his improvement through the follow-up period.The other 4 patients with neurological deficits had a stationary deficit throughout the follow-up periods.None of our patients had any new temporary or permanent deficits after the procedure.
The mean follow-up was 8 months, with a minimum of 6 months and a maximum of 36 months.The mean operative time was 133 min (range 72-270) for each fracture.
Operative time was calculated as the time from skin incision to last stitch in skin closure.
The mean operative blood loss was 107.1 ml (range 20-500) for each fracture.We used visual method for blood loss calculation.
Except for the polytrauma patient who stayed for 64 days and the paraplegic patient who stayed for 35 days, the average hospital stay was 7.7 days (range 1-29) (hospital stay was recorded from time of admission till discharge).The relatively long hospital stay was made by some patients with neurological deficits who started rehabilitation during their hospital stay before being discharged.
All our patients reported good postoperative wound cosmesis according to the VAS in the 1-week, 1-month, and 3-month follow-ups, except for one patient who developed a mild superficial wound infection after one month postsurgery.However, the patient responded to conservative treatment and showed adequate healing at three months postoperatively.Radiological complete bony fusion of the fracture line was evident in 42/43 (97.6%) of fractures, with partial fusion in one fracture at 6-month follow-up.Only 7/42 (16.6%) patient who had complete radiological bony fusion of their fractures were apparent on the 3-month follow-up, for the rest of the fractures the radiological bony fusion was completed on the 6-month follow-up.(Fig. 3).

Surgery-related complications
Intraoperatively, there was only one case of a medialized left T5 screw violation that upon removal resulted in a cerebrospinal fluid (CSF) leak from the screw entry track which was controlled by local measures of bone wax and gel foam with no further sequela.Two incidents of dissemblance were noted where a peripheral screw lost connection from the rod, however, this was of no neurological consequence nor radiological change within the Cobb's angle postoperatively and no surgical revision was needed.

Discussion
Starting from the 1970s, reports of the use of pedicle screws and plates to treat thoracolumbar fractures were published [12,13].Since that time, there has been significant paravertebral muscle injury and a high morbidity rate with this type of fixation using the open posterior technique.Conversely, less-invasive methods developed later like percutaneous pedicle screw fixation significantly reduced the degree of iatrogenic muscle damage [14,15].
In 1984, Magerl [16] used an external fixator attached to the spinal vertebrae with screws through the pedicles for the treatment of spinal instability secondary to acute spinal trauma and osteomyelitis.This can be considered the first report of percutaneous spinal fixation in the literature.Kim et al. [17] showed that compared to conventional open fixation methods, percutaneous pedicle screw fixation results in reduced injury to the paravertebral muscle.It has been also reported that percutaneous pedicle screw fixation has the advantage of lower operative time, blood loss, and infection rate compared to the conventional open fixation technique.Also, the percutaneous fixation technique provides better results in terms of functional and pain outcomes for patients with thoracolumbar fractures [15,16].
The difficulty of bone fusion using bone grafts or bone substitutes is considered the main disadvantages of percutaneous fixation.A prospective, randomized trial evaluated pedicle instrumentation with and without fusion, the group without fusion showed superior outcomes.Pedicle screw fixation with fusion, however, appears to be effective in fractures with a high degree of anterior vertebral wedging and significant  levels of vertebral body comminution, according to recent research [18,19].However, there are no specific tools for percutaneous fixation that make fusion conceivable even with the lack of scientific proof of their absolute necessity.However, studies showed that percutaneous fixation provides superior radiological bony fusion and correction of associated spinal deformity [9,[20][21][22][23].In our study, radiological complete bony fusion was achieved in 42/43 (97.6%) of our fractures.
Our results demonstrated a mean Cobb's angle correction similar to that reported in the literature for percutaneous fixation, with an average kyphotic angle correction of 7.9 degrees.A recent meta-analysis showed that the degree of Cobb's angle recovery after percutaneous pedicle screw fixation was superior to that of open pedicle screw fixation, but the difference was not statistically significant [21].Several studies and recently published meta-analyses reported an average Cobb's angle correction of 5.5 degrees.These studies also reported that with using percutaneous screw fixation, a significant improvement in the height of anterior body fracture and the canal compression usually occurred [15,[21][22][23][24].
Percutaneous fixation was often considered unsuitable for spine fractures complicated by neurological deficits since decompression is not possible with this approach.To provide the necessary decompression,  however, a limited posterior midline approach can be used.In this instance, a midline incision is made across the compressed spinal segment after the percutaneous fixation is completed, allowing for limited muscle detachment in the lateral and cranial-caudal planes.There are not plenty of indications for this kind of decompression, though a paramedian mini skin incision for decompression and corpectomies with expandable cage graft was reported in the literature with the use of a percutaneous fixation technique [15,25].In our study, we used this midline technique for decompression in 5 cases with neurological deficits, two with ASIA score grade C and two with ASIA score grade A, and one case with ASIA grade D. We were able to achieve adequate decompression of the fracture in these cases with a limited midline skin incision.Also, we reported no new neurological deficit after the percutaneous fixation in our series.Also, one of our patients who was ASIA grade D preoperatively, reported mild improvement postoperatively.Previous studies have demonstrated that minimally invasive percutaneous fixation offers a less morbid treatment option for patients with thoracolumbar fractures, particularly in terms of reduced blood loss and hospital stay [18,20,26].In our study, we observed a mean blood loss of 107 ml and a mean hospital stay of 7.7 days.These findings align with previously published reports [27].Our mean operative time was 133 min.Also, all our patients reported good postoperative wound cosmesis in the follow-up period, which is also, within the reported range reported in the literature [20,28].
One of the potential complications of percutaneous fixation is screw malposition.It is reported in the literature that percutaneous fixation has a lower screw malalignment than the open fixation technique with an incidence rate of less than 1% for the percutaneous fixation technique [21,24].In our study, we had only a single screw 1/282 (0.35%) which was noticed to be in a medial position and was removed intraoperatively with no associated morbidity or mortality.Two screws in our series 2/282 (0.7%) were found to have lost connection with rods in the postoperative follow-up with no significant associated morbidity that required no revision through the postoperative follow-up period.
While percutaneous fixation systems are generally more expensive than conventional open surgery fixation systems, patients who undergo percutaneous spine fixation typically experience shorter hospital stay and early return to work.Although we did not conduct a cost-effectiveness analysis for our series, it has been reported that percutaneous fixation for thoracolumbar fractures results in lower total hospital charges compared to open surgery, but this difference was not statistically significant [26].Further cost-effectiveness studies are recommended to conclude the cost-benefit of percutaneous fixation versus conventional open technique putting into consideration the length of hospital stay cost, the duration of functional recovery of patients and return to work.

Limitations of our study
Other than the inherited defects of a retrospective study, this study has a limited number of patients with fewer number having associated neurological morbidity that required midline minimal posterior decompression which hindered further feasibility studying of this approach.Also, there were missing data regarding the cost-effectiveness of the percutaneous technique.

Conclusion
Percutaneous fixation is considered a safe and effective technique for patients with thoracolumbar fractures which provides satisfactory functional patients' outcomes [29].It also can be used in the presence of neurological deficits by doing limited midline incision for decompression of neural tissue.The incidence of screw malalignment with percutaneous spine fixation is less than open technique.Percutaneous fixation provides acceptable improvement of radiological parameters in patients with thoracolumbar fractures.Further studies are recommended to evaluate the cost-effectiveness of a potentially expensive technique and the suitable indication in patients associated with a significant lumbar canal compromise.

Fig. 1 A
Fig. 1 A & B A-P and Lateral views intraoperative fluoroscopy showing tap needle (both sides in the middle vertebra) and guide wires in upper and lower levels.C & D A-P and Lateral views intraoperative fluoroscopy showing screws in the pedicles with extenders.E screw extenders projecting from multiple paramedian incisions.F Multiple paramedian skin incisions for screws and rods insertion

Fig. 2
Fig. 2 Pre-and postoperative imaging of 36 years old male patient with L1 burst fracture presented by cauda equina syndrome who underwent MIPS and minimal open decompression through small midline incision.Preoperative MRI sagittal T2, STIR and axial cuts are shown in figures A, B and C, respectively.Preoperative CT scan, axial cut on L1 vertebra showing retropulsed fragment causing the cuada equina compression is shown in figure D. Postoperative CT scan, axial cuts showing D12, D11, L1 vertebra after laminectomy, L2, L3 perfect screw placement with L2 laminectomy and midline fusion at L2 are shown in figures E, F, G, H and I, respectively.Postoperative CAT scan, sagittal cuts showing the whole construct from D11 down to L3 on Right side and Left side are shown in figures J and K, respectively

Fig. 3
Fig. 3 Pre-and postoperative CAT scan imaging of L1 burst fracture.Figure A and B represent preoperative sagittal and axial CAT scan of the fractured L1 body with irregular fracture line margins, respectively.Figure C and D show CAT scan sagittal and axial cuts, respectively, of the same fracture at 8 months with fusion of the fractured fragments.The blue arrow represents the healed body in sagittal film (C) and represents the fused fragment in axial film (D)

Fig. 4
Fig. 4 Pre-and postoperative CAT scan imaging of L1 burst fracture.Figure A represents preoperative sagittal CAT scan showing kyphotic cobb's angle of 22 degrees that was corrected after long segment percutaneous fixation of 2 levels above and 2 levels below to a cobb's angle of 6 degrees as shown in figure (B).The cobb's angle was automatically measured by GE packs software

Table 1
patients' demographics and fracture characteristic data ASIA American spinal injury association, D dorsal, L lumbar, NO number, TLICS Thoracolumbar injury classification and severity score