Patients aged ≥18 years who sustained unstable TLFs with a Load Sharing Classification (LSC) score of ≥7 and no associated neurological deficit were included in the study [17]. Patients under 18 years of age, those with associated visceral or appendicular injuries, neurological deficits, prior spine surgery, and elderly patients with osteoporotic fractures, tumors, and infections were excluded. Among 70 patients during the study period, four patients were excluded based on the exclusion criteria, and the remaining 66 patients were randomly assigned to group 1 (long intermediate screw [LIS] group) and group 2 (short intermediate screw [SIS] group) using computer-generated numbers. The allocation details of the study are presented using the Consolidated Standards of Reporting Trials guidelines for reporting randomized trials (http://www.consort-statement.org/) (Fig. 1). Patients in group 1 underwent short-segment fixation with LIS whereas those in group 2 underwent short-segment fixation with SIS. In our study, an intermediate screw was classified as an SIS if it occupied <50 % of the anteroposterior (AP) vertebral body diameter, and as an LIS if it occupied >50% of the AP vertebral body diameter. Specifically, the length of the SIS in this study was ≤35 mm, whereas the length of the LIS was ≥40 mm (Fig. 2). All patients were followed up for a minimum period of 2 years.

All patients were positioned prone on radiolucent bolsters with adequate padding for all bony prominences. A standard posterior midline approach was utilized, and SSPF was performed with pedicle screw insertion one vertebra above and one below the fracture level using the freehand technique. Intermediate screws were inserted into both pedicles of the fractured vertebra. The diameter of the pedicle screws used at the nonfractured vertebra was 6.5 mm, whereas 5.5 mm screws were used at the fracture level. The intermediate screw heads were left slightly proud to act as a push point and also achieve a reduction of kyphosis. No additional effort was made to achieve a complete reduction of kyphosis or restore vertebral height through rod over-contouring, compression, or distraction. After confirming the pedicle screw position using fluoroscopic guidance, the wound was closed in layers. Two experienced spine surgeons performed all the surgeries. Postoperatively, patients were encouraged to start mobilization as tolerated with a thoracolumbar brace. All patients were advised to wear the brace for 1 month, after which it was discontinued. All patients underwent periodic follow-up with clinical and radiologic evaluations.

The patient’s demographics and injury characteristics (age, sex, mode of injury) were analyzed. Preoperative radiographs and computed tomography scans were used to classify the fracture morphology according to the AO Spine Thoracolumbar Injury Classification and the McCormack Load Sharing Classification.

Intraoperative parameters, such as operative time and blood loss, were recorded. Clinical outcomes, namely Visual Analog Scale (VAS) score and Oswestry Disability Index (ODI) were evaluated preoperatively, one month postoperatively, and at the final follow-up.

The following radiological parameters were evaluated preoperatively, immediately postoperatively, and at the final follow-up (Fig. 3): (1) Anterior body height (ABH) was measured from the anterosuperior to the anteroinferior corner of the fractured vertebra. (2) Posterior body height (PBH) was measured from the posterosuperior to the posteroinferior corner of the fractured vertebra. (3) Anterior vertebral body compression percentage was calculated using the ABH/PBH ratio, which measures the degree of vertebral height compression. (4) Local kyphosis angle (LKA) was measured as the angle between both endplates of the fractured vertebra. (5) Regional kyphosis angle (RKA) was measured as the angle formed between a line drawn parallel to the superior endplate of the cranial vertebra above the fracture and a line drawn parallel to the inferior endplate of the caudal vertebra below the fracture.

Fracture healing, implant failure, revision surgery, and surgical complications were recorded until the final follow-up. Implant failure was defined as the occurrence of any of the following events: screw breakage or pullout, rod breakage or displacement, screw or rod deformation, screw head dislodgement, implant loosening, or progressive deformity.

Data analyses were performed using IBM SPSS ver. 21.0 (IBM Corp., Armonk, NY, USA). Continuous variables were compared using the unpaired t-test, while categorical variables were analyzed using Fisher’s exact or the chi-square test. For comparing continuous variables between groups, the one-way analysis of variance test was used, followed by the least significant difference test for post hoc analysis. Statistical significance was set at a p-value less than 0.05.

The study included 66 patients, with 33 patients in each group. The mean age of patients in the LIS and SIS groups was 38.4 years and 40.1 years, respectively. The LIS group consisted of 26 males and seven females, while the SIS group included 28 males and five females. The most common mode of injury in both groups was a fall from height (16 patients in the LIS group, 14 patients in the SIS group) followed by road traffic accidents (12 patients in the LIS group, eight patients in the SIS group) (Table 1).

The analysis of radiological parameters is presented in Table 2.

Shi et al. [23] found that the postoperative vertebral body height in patients with TLFs depends on the preoperative vertebral compression and the degree of fracture comminution, based on their study of 186 patients. The greater the compression of the fractured vertebral body, the more severe the damage to its structure and the greater the degree of expansion required to restore vertebral body height [24]. During internal fixation with pedicle screws, repositioning the fractured cortical bone is challenging, and the collapsed cancellous bone cannot be fully restored, increasing the risk of postoperative vertebral collapse [25,26]. Consequently, the degree of preoperative vertebral compression may be a predictor of postoperative vertebral body height. In the present study, both the LIS and SIS groups showed restoration of ABH and PBH, as well as improvement in the ABH/PBH ratio. This was likely attributable to the fulcrum effect of the intermediate screw. However, our analysis did not reveal any significant difference between the two groups, suggesting that postoperative vertebral body height restoration is independent of the length of the intermediate screw used. Instead, our findings indicate that vertebral body height restoration indirectly depends on the degree of fracture comminution. In contrast, Öztürk et al. [22] found that the use of thick and LISs resulted in significantly better preservation of PBH, likely due to the long anterior vector arm during the indirect reduction of fracture fragments.

The use of the intermediate screw led to improvements in LKA in both groups, with mean corrections of 8.35° in the LIS group and 8.72° in the SIS group in the immediate postoperative period. Notably, the LIS group showed a significant improvement in regional kyphosis (8.8°) compared to the SIS group (4.48°) in the immediate postoperative period. However, the correction was not sustained until the final follow-up. This suggests that long intermediate screw constructs provide better regional kyphosis correction in the short term. The late kyphosis recurrence in our patients may be attributed to vertebral body collapse and the presence of associated cranial (predominantly) and caudal intervertebral disc injuries.

In the present study, all patients had an LSC score of ≥7, indicating severe vertebral body comminution and complete loss of anterior column support. However, the use of a lengthy intermediate screw (>50% of vertebral body diameter) increased the immediate stability and stiffness of the injured spinal segments, effectively supporting the endplate and preventing collapse. This additionally provided anterior column support, facilitating sustained kyphosis correction, which was not achievable in the SIS group (<50% of vertebral body diameter). These results are consistent with a biomechanical study by Liu et al. [27], where the restoration of vertebral body height was maintained with the screw inserted into 60% of the vertebral body. In another similar study, increasing the screw depth from 50% to 80% of the vertebral body enhanced screw stability under flexural and torsional loads by 30% [28].

Studies have shown that TLFs are often accompanied by injuries to the upper and lower intervertebral disc [29]. This can lead to postoperative complications, including progressive kyphotic deformity due to intervertebral disc space collapse. Progressive disc space narrowing may be caused by the adaptation of the disc to morphologic changes in the osseous endplate or the disc creeping into the central depression of the endplate [30]. Similarly, Kim et al. [31], in a retrospective analysis involving 42 patients with TLFs, identified disruption of adjacent intervertebral discs as a major reason for postoperative morphologic change, leading to a gradual reduction in the anterior disc height. This may eventually lead to kyphosis recurrence without additional bony morphological changes in the fractured vertebra. Moreover, the anterior portion of the intervertebral disc adjacent to the fractured vertebra is particularly vulnerable to injury from flexion and compression forces, especially in cases with severe anterior body collapse [31].

Our study’s findings align with those of Mohammed et al. [32], who observed a loss of kyphosis correction during follow-up in 36 patients treated with SPFF using an intermediate screw. They attributed the loss of correction to the fracture configuration but also suggested that other clinical factors may contribute [32]. Formica et al. [33] investigated these factors and found that obese patients (body mass index >30 kg/m²) experienced a significant loss of kyphosis correction. Variables such as sex, smoking habit, and preoperative neurological status showed no significant association with the loss of correction [33].

Research indicates that the pedicle, rather than the vertebral body, contributes approximately 80% of the stiffness and 60% of the pullout strength [19,34]. Zindrick [35] reported no difference in pullout strength between similarly sized screws inserted halfway or completely into the anterior cortex. They noted that the more cortices penetrated, the stronger the fixation [35]. In the present study, longer screws (extending >50% of the vertebral body diameter) provided initial anterior column support to the fractured vertebra, but subsequent loss of kyphosis correction occurred, likely due to severe anterior vertebral body comminution and associated disc injury. Furthermore, in highly comminuted fractures (LSC score ≥7), using longer and larger diameter screws poses a risk of distracting the bone fragments, potentially impeding fracture healing and compromising the canal. Nonetheless, no neurological complications related to the use of longer intermediate screws were encountered in the present study.

Some limitations of this study should be acknowledged. This was a single-center study with a limited sample size. Moreover, the follow-up period was limited to 2 years, which is considered relatively short. A multicenter study with a large sample size and a relatively long follow-up period is required to obtain more robust evidence.