Abstract

Introduction: The lumbar and thoracic spine sustain 90% of all spinal fractures. There is ongoing discussion regarding the most effective treatment for thoracolumbar burst fractures. Clinical results of long-segment and short-segment posterior stabilization are contrasted in this research.

Materials and methods: There were thirty patients in total; fifteen underwent short-segment stabilization, and fifteen underwent long-segment stabilization. All patients were assessed and treated according to the advanced trauma life support (ATLS) protocol. Mobilization was started as tolerated. Postoperative X-rays were taken, and follow-up occurred monthly for six months, then every two months thereafter up to one year. Functional outcomes were determined by employing the ASIA impairment scale and VAS scores.

Results: Long-segment group operating times were 136.1±11.31 and 79.4±11.7 minutes, respectively, substantially longer than those of short-segment groups (p<0.05). In comparison to short-segment groups, long-segment groups experienced a statistically significant greater blood loss of 1263.3±151.74 and 876.7±189.8 ml, respectively (p<0.05). The ASIA impairment scale measurement, change in Beck’s index, and kyphotic angle were statistically insignificant among both groups. In the long-segment group, most patients had a Denis pain scale score of P2 (40%) and a Denis work scale score of W3 (46.7%), but in the short-segment group, most patients had scores of P3 (60%) and W4 (46.7%). We encountered no major complications.

Conclusion: While long-segment fixation provides greater stability, short-segment fixation results in less operative time and blood loss without compromising clinical outcomes. Longer follow-up studies with larger sample sizes are recommended.

ASIA impairment scale, Beck’s index, Dennis scale, ODI, VAS

Nine out of ten spinal fractures are thoracic and lumbar spine fractures. Any type of trauma can cause a thoracolumbar fracture; however, falls from a height or road traffic accidents are the most frequent causes. This causes the slightly flexed spine to be compressed vertically; rotational, shear, and extension stresses can also result in different fracture forms. Apart from being a substantial cause of illness and mortality, neurological impairments are linked to 20% of them.^[1] Patients in the active age group constitute the majority of them. There have been improvements in intraoperative monitoring, more stable fixation devices, and diagnostic imaging techniques over the last few decades.^[2] Literature suggests that the extensive usage of steroids to lessen secondary injuries to the central nervous system has also gained enough consideration.^[3–6] Despite these developments, orthopedic surgeons still face challenges in treating these fractures. Several surgical techniques have been developed to treat thoracolumbar burst fractures. These include combination therapies using anterior and posterior spinal approaches and direct anterior decompression via corpectomy, along with posterior segment or long segment pedicle screw fixation.^{[7, 8]}

There is still disagreement over the best surgical strategy for treating thoracolumbar burst fractures. Whether short or lengthy posterior fixation produces better results is a matter of debate.^{[9, 10]} Although it increases stability, long posterior fixation—which uses pedicle screws as well as rods positioned 2 levels above and below the fracture—may put undue strain on the lower discs and lead to over-instrumentation. Short posterior fixation, on the other hand, preserves more mobility segments and lessens the load on the surrounding discs by using screws that are one level above and below the fracture.^{[11, 12]}

In spite of its benefits, some research has linked brief fixation to increased kyphotic collapse and hardware failure rates.^[13] A number of studies have compared and evaluated the clinical and radiological outcomes of short versus long posterior fixation techniques.^{[14, 15]}

In this regard, our study’s objective is to examine the relative merits of short-segment posterior stabilization and long-segment posterior stabilization in the treatment of thoracolumbar burst fractures.

Materials and methods

We conducted a prospective interventional study between October 2021 and October 2023. Institutional Ethics Committee clearance was obtained before conducting this study. All patients gave informed consent. We conducted this study on thirty patients who presented to the emergency room having thoracolumbar burst fractures. They were initially assessed and treated according to the ATLS protocol. A comprehensive neurological and clinical assessment was performed on each subject. The level of the injury sustained by the patient was diagnosed with clinical and radiological correlation according to the AO spine trauma classification system.

The injuries sustained by the patients were classified according to thoracolumbar injury classification and severity scores (TLICS) (Table 1), and the ASIA grading system (Table 2) was used to quantify the neurological impairment for all the patients in this study. Cobb’s angle and Beck’s index were also calculated preoperatively for all patients (Figs 1, 2). Patients willing to enroll in this study for surgical procedure and follow-up, patients between 20 and 80 years of age, patients with TLICS more than or equal to 4, and patients with one-level thoracolumbar vertebral burst fracture were included in this investigation. Patients with pathological fractures, metabolic bone disease, a TLICS score less than 4, pregnant patients, and patients who were not willing to get enrolled were excluded from the investigation.

Figure 1.

Case 1: preoperative X-ray of L2 burst fracture, Cobb’s angle 13, Beck’s index 0.5.

Figure 2.

Case 1: postoperative X-ray of L2 burst fracture – short-segment fixation, Cobb’s angle 4, Beck’s index 0.7.

Table 1.

Download as

CSV

XLSX

TLICS scoring system: a score of less than 3 indicates a stable fracture, so it can be treated conservatively

Fracture mechanism
Compression fracture	1
Burst	2
Translation	3
Distraction	4
Neurological involvement
Intact	0
Nerve root	2
Cord, conus medullaris, incomplete	3
Cord, conus medullaris, complete	2
Cauda equine	3
Posterior ligamentous complex integrity
Intact	0
Injury suspected/indeterminate	2
Injured	3

Table 2.

Download as

CSV

XLSX

American spinal injury association score

A	Complete	No motor or sensory function in the lowest sacral segment
B	Incomplete	Sensory function below neurological level and in S4 S5, no motor function below the neurological level
C	Incomplete	Motor function is preserved below neurological level; key muscle groups below neurological level have grade <3
D	Incomplete	Motor function is preserved below neurological level; key muscle groups below neurological level have grade >3
E	Normal	Normal motor and sensory function

For evaluating retropulsion of fractured vertebrae, canal compromise, cord status, and posterior longitudinal ligament integrity, radiographs were obtained, and then CT and MRI scans were performed. Additionally, an E-FAST was completed, and clearance from the emergency team was obtained. The random lot method was used to determine whether to move on with short- or long-segment posterior stabilization; accordingly, 15 patients underwent short-segment stabilization and 15 patients underwent long-segment stabilization.

Surgical procedure

The anesthesia team administered general anesthesia, and the patient was then positioned prone on the surgical table. The patient was put in a prone position with bolsters at appropriate sites to keep the abdomen free from tension/pressure. Level of injury was identified and confirmed under C-arm guidance. The conventional posterior technique was used to make the incision. The surgical plane between spinous processes and paraspinal muscles was created once the vertebrae’s spinous processes were recognized. A Cobb’s spinal elevator was used to delete the paraspinal muscles, and a self-retaining retractor was then used to reflect them laterally. The location where the longitudinal axis of the superior facet and the center of the transverse process intersected was used to identify pedicles. After the pedicles were probed and their depth was measured, the pedicle screws were inserted under the supervision of an image intensifier. The dorsal and lumbar pedicles were fitted with screws of the proper size. Thorough decompression was done by doing a laminectomy at the fracture site; both cords and roots were confirmed to be free. Connecting rods were placed and fixed with screws. Horizontal crosslinks were added if needed. Intertransverse fusion was performed with bone graft taken from remnants of decompression. Closure was done in layers, and a drain was inserted with negative suction.

Postoperatively, patients were provided with in-bed mobilization in the postop period. On postoperative day 2, the drain was removed. Dressings were done on postoperative days 2 and 5. Suture removal was done on postoperative day 12. Mobilization was started as per the patient’s tolerance. Stage-appropriate neurorehabilitation was provided by the physiotherapists. Post-op X-rays were taken on postoperative days 1 and 2.

Follow-up was done monthly once every six months and two months once thereafter for up to one year. Cobb’s angle and Beck’s index were measured at the one-year follow-up (Figs 3, 4). The Denis Pain and Work Evaluation Scale was employed for evaluating patients during the follow-up period. The acquired data was processed using SPSS version 16 [IBM]. While the categorical variables—such as sex, mode of injury, diagnosis, and ASIA impairment scale—were categorized as frequency and percentage, the numerical variables—such as age, Beck’s index, and Cobb’s angle—were summarized as mean and standard deviation.

Figure 3.

Case 2: preoperative X-ray of L2 burst fracture, Cobb’s angle 21, Beck’s index 0.5.

Figure 4.

Case 2: postoperative X-ray of L2 burst fracture, Cobb’s angle 11, Beck’s index 0.8.

Independent t-tests for numerical variables and chi-square tests for categorical data were employed for comparing independent and dependent variables. Repeated measures of analysis of variance (RMNOVA) were for comparing timeframes within each group. Pie charts, bar charts, and box and whisker plots were employed for illustrating variables’ distribution. Every p-value <0.05 was regarded as significant.

Among the 30 participants of the study, exactly half of the patients underwent long-segment stabilization, and the remaining underwent short-segment stabilization. The mean age of participants was 38±17 years in the long-segment group and 35±12.7 years in the short-segment group. There was equal distribution of males and females in both groups, with males being the majority, 73% in each group. The majority (60%) of the patients sustained RTA in the long-segment group, while more than half (53.3%) in short-segment group were from falls from height (Table 3).

The majority of patients sustained L3 fractures in both the long-segment (53%) and short-segment groups (47%). Similarly, most patients (43%) in the long-segment group exhibited levels fused at D11-L3, while in the short-segment group, most of them (46.7%) had fusion from D12-L2 (Table 4).

When the two groups were compared using TLICS, the majority scored 6 in both long (60%) and short-segment (80%) stabilization. The TLICS comparison was statistically insignificant, suggesting that both groups had an equal distribution (Table 5).

The average operating time was 136.1±11.31 minutes for the long-segment group and 79.4±11.7 minutes for the short-segment group. This difference was found to be statistically significant (p❬0.05), suggesting that surgeries in the short-segment group required significantly less time compared to the long-segment group. Regarding neurological status, the ASIA impairment scale was classified as grade D across all time points for both groups. The lack of significant difference between the groups suggests a comparable neurological distribution throughout the study period (Table 6).

The Beck’s indexes in the preoperative period at 3, 6, and 12 months were statistically insignificant at all timelines, inferring equal distribution in both groups (Table 7).

The longitudinal comparison of the Beck index score at different timelines in both the long- and short-segment groups shows a significant increase in score (p❬0.05) from the preoperative period to one year after surgery (Table 8).

The Cobb’s angles in the preoperative period and postoperative 3, 6, and 12 months were 16.1, 4.5, 4.6, and 4.8 in the long-segment group and were 17.2, 5.1, 5.1, and 5.6 in the short-segment group. The difference between the two groups was statistically insignificant at all timelines (Table 9).

The longitudinal comparison of Cobb’s angle at different timelines in the long- as well as short-segment groups shows a significant decline in Cobb’s angle (p❬0.05) from the pre-operative period to one year after surgery (Table 10).

The majority of the patients had P2 (40%) in the Denis pain scale and W3 (46.7%) in the Denis work scale in the long-segment group, while most patients had P3 (60%) in the Denis pain scale and W4 (46.7%) in the Denis work scale in the short-segment group (Table 11).

Nine patients experienced complications, such as bedsores and urinary tract infections. Four individuals in the short-segment group and one in the long-segment group had urinary tract infections. Three patients in the short-segment group and 1 in the long-segment group were found to have bedsores. With the exception of four patients—three from the short-segment group and one from the long-segment group—nearly every patient was mobilized within 12 days following surgery.

Furthermore, 2 patients in the long-segment group had cross-links reinforced into their constructions, whereas 3 patients in the short-segment group exhibited cross-links employed to stabilize their connecting rods (Table 12).

Helping patients resume their regular daily activities is the main goal of surgical surgery for thoracolumbar burst fractures. Restoring spinal column stability as well as decompressing the spinal canal to allow for early patient mobilization are the primary objectives, regardless of the therapeutic strategy selected. However, there is ongoing discussion over the best approach for treating thoracolumbar burst fractures.^[16]

In this investigation, we found that the average age of patients was 38±17 years in the long-segment group and 35±12.7 years in the short-segment group. The majority of the overall population was male. The most prevalent mode of injury in this study population was RTA, followed by FFH. Khurjekar et al.^[17] found that the average age of 92 patients who sustained thoracolumbar fractures was 32 years. The most common mechanisms of injury were falls from height (50%) and RTAs (46.7%), predominantly affecting young adult males.

The T12 and L2 vertebrae are injured in around 60% of thoracolumbar trauma patients. This area acts as a transitional zone between the more flexible, lordotic lumbar spine and the comparatively inflexible, kyphotic thoracic spine, which is supported by the rib cage. This region of the spine is especially vulnerable to fractures because of the differential mobility and biomechanics of these contiguous spinal segments, which result in substantial mechanical stress upon trauma.^[18] In this study, the majority of the patients in both groups sustained L1 fractures; subsequently, the majority of patients in the short-segment group underwent fixation from the D12-L2 level, and the majority of patients in the long-segment group underwent fixation from the D11-L3 level.

A dorsolumbar spinal injury with neurological dysfunction is an orthopedic emergency that increases morbidity and death significantly. We noticed that the TLICS score was 6 in the majority of patients in both groups. The ASIA impairment scale in both groups was class D at all timelines. Differences between the two groups were statistically insignificant at all timelines, inferring equal distribution in both groups. In a comparative study done by Bhatia et al., they observed that functional along with neurological outcomes based on ASIA scoring were similar in both groups.^[19]

The Beck’s index in the preoperative period and postoperative 3, 6, and 12 months were 0.6, 0.77, 0.8, and 0.77, respectively, in the long-segment group and were 0.6, 0.77, 0.8, and 0.76, respectively, in the short-segment group. In research conducted by Sapkas et al.^[16], they identified that a statistically insignificant difference was present in the increase of Beck’s index in both groups.

We noticed that the increase in Beck’s index in the long-segment group from the pre-operative value of 0.6±0.7 to the post-operative value of 0.77±0.047 and the increase in Beck’s index in the short-segment group from the pre-operative value of 0.6±0.5 to the post-operative value of 0.76±0.049 were both statistically significant. These are similar to research conducted by Tiwari et al.^[20], who demonstrated that long-segment fixation results in reduction of the Beck’s index from a mean preoperative ratio of 0.543 to a mean postoperative ratio of 0.888. And another study^[21] on long segment stabilization for thoracolumbar fractures showed that the Beck’s index significantly increased from a preoperative mean of 0.78±2.65 to 0.86±0.40 at the final follow-up. These findings prove that both long- and short-segment stabilization techniques are effective in achieving a satisfactory increase in postoperative Beck’s index.

The long segment group’s Cobb’s angle was 16.1 before surgery and 4.5, 4.6, and 4.8 three, six, and twelve months later, respectively. The comparable values for the short-segment group were 17.2 prior to surgery and 5.1, 5.1, and 5.6 at the same postoperative intervals. At no point in time were the two groups’ variations in Cobb’s angle statistically significant. Aggarwal et al.^[22] and Guven et al.^[23] demonstrated similar results, showing no discernible difference between long-segment and short-segment posterior fixation procedures concerning correcting the initial kyphotic angle.

In this investigation, Cobb’s angle for long segment fixation significantly decreased from a preoperative value of 16.1°±3.09° to 4.8°±2.46° at 1-year follow-up, and in the short-segment group, it significantly decreased from a preoperative value of 17.2°±2.7° to 5.6°±2.72° at the 1-year post-op period. In a study done to evaluate the effectiveness of short-segment fixation^[24], they showed that Cobb’s angle significantly declined from 11.6° preoperatively to 3.5° postoperatively. Another study for long-segment fixation^[21] showed that Cobb’s angle significantly decreased from preoperative 21.83±4.5 degrees to postoperative 12.5±3 degrees. These findings suggest that both long-segment and short-segment stabilization are effective in improving Cobb’s angle of fractured segments.

The mean operating time for the long-segment stabilization was 136 minutes and the mean blood loss was 1263 ml, both of which were significantly more as compared to short-segment group in our study. Sapkas et al.^[16] in their study similarly identified that blood loss was significantly more in long-segment fixation recording as 1200 ml. Long segment fixation necessitated a longer surgical time, averaging 130 minutes, according to Osmon et al. This corroborates the finding that long-segment stabilization often entails a longer recovery period and more blood loss.^[16] A meta-analysis also showed that the two methods differed significantly in the amount of time needed for surgery, with long segment fixation requiring more time. Short-segment instrumentation, on the other hand, could reduce the dangers connected with extended operation. However, after combining data from several trials, there was statistically insignificant difference in the amount of blood lost during surgery among two groups. This is probably due to several factors encompassing expertise of the surgeon, the size of the screw, and individual anatomical variances.^[25]

The majority of the patients had P2 (40%) in the Denis pain scale and W3 (46.7%) in the Denis work scale in the long-segment group, while most patients had P3 (60%) in the Denis pain scale and W4 (46.7%) in the Denis work scale in the short-segment group. A study conducted by Kim et al.^[26] showed higher improvement in Dennis’s work and pain scale in the long stabilization group. Several technical changes have been suggested to enhance the construct’s stiffness within the limitations of short-segment fixation, including the addition of extra hooks at the screw levels and cross-links.^[27–29] Similarly in the current investigation, in the short-segment group, 3 patients had their connecting rods fixed with cross-links, while 2 patients in the long-segment group had their constructs augmented by cross-links. The rest of the patients in both groups did not have cross-link augmentation.

Despite the fact that short-segment pedicle instrumentation is frequently regarded as the best method, several studies have found that it has a significant failure rate. Four different types of hardware failure were found in their analysis of early failures for thoracolumbar burst fractures using this method: progressive kyphosis due to screw bending, segmental kyphosis caused by the caudal screw in the lumbar construct breaking, and kyphosis caused by either vertebral collapse or translation without any hardware deformation.^{[30, 31]} However, in our investigation, none of these issues were noted.

Although several methods assert that they provide the finest treatment, there are not any widely recognized, scientifically supported recommendations. When there is substantial biomechanical instability or the possibility of neurological damage, surgery is often recommended for thoracic as well as lumbar spine fractures. Radiographic indicators including Beck’s index and Cobb’s angle indicate statistically insignificant differences among groups in this investigation. Reduced blood loss and shorter operating time were linked to short-segment fixation. Additionally, insignificant difference in clinical outcomes between short-segment and long-segment instrumentation groups was observed. With no discernible differences in long-term results, these findings show that LS, as well as SS fixation, are equally successful in treating segmental kyphosis along with vertebral body deformation.

The present study is acknowledged to have some limitations. The total number of patients included in the study was 30, with 15 patients in each group. To elucidate the observed variations, a more substantial sample size would have been advantageous. The subsequent period of observation lasted one year. However, a more extensive follow-up would be required to ascertain any statistically significant differences. Consequently, it is anticipated that this study will contribute to the promotion of enhanced quality management and evidence-based practices.

Sai Deiv Ramkumar M: conceptualization, methodology; Benjamin Vinodh: software, validation, formal analysis, R Tarunprasad: investigation, resources, data curation, writing–original draft; Arun Kumar Chandhuru: writing–review and editing, visualization, supervision.

The authors have no funding to report.

The authors have declared that no competing interests exist.

The authors have no support to report.