In the trend of an aging society, degenerative lumbar spondylosis causes disability. It increases the socioeconomic burden worldwide [1,2]. Lumbar fusion surgery is widely used for treating advanced degeneration of the lumbar spine, requiring extensive decompression and restoring stability. Transforaminal lumbar interbody fusion (TLIF) is one of the standards in lumbar fusion surgeries. TLIF has been proven effective and safe in minimizing injuries to neural structures [3]. After facetectomy, the surgical corridor between traversing and exiting nerve roots allows placing the cage into the intervertebral space after discectomy. Currently, there are minimally invasive techniques of TLIF, including tubular or endoscopic-assisted procedures. These techniques mainly emphasize fewer injuries to paraspinal muscles to decrease postoperative pain and enhance recovery after the operation. However, the goal of the lumbar fusion surgery is solid bony fusion of the involved vertebrae. Factors affecting the fusion rate should be the priority when making decisions. Interbody fusion generally involves inserting bone grafts or interbody spacers to facilitate fusion between adjacent vertebral bodies. Therefore, various designs of interbody cages, bone substitutes, and expanders have been developed. These technologies and biological products are essential parts of lumbar fusion surgeries.

In recent years, expandable cages have been novel devices used in minimally invasive lumbar fusion surgeries. Expandable cages allow for customizable fit within the intervertebral space, enabling surgeons to restore the alignment and disc height. The expandable cages can be inserted into the intervertebral space in a compact form and then expanded within the disc space. The expandability allows the surgeon to pass the cage through a small corridor during minimally invasive procedures. The surgeon also has greater control over the disc space distraction to decompress the neural structures. Furthermore, expandable cages can accommodate additional bone graft material, enhancing the potential for successful fusion. The increased surface area available for bone growth within the cage promotes bony ingrowth, fusion, and the development of a robust structural bridge between adjacent vertebral bodies.

Some researchers have shown promising results in the applications of expandable cages [4,5]. Comparable fusion rate and significant correction of lumbar lordosis (LL) have been reported. Though expandable cages offer many advantages, the potential complications and efficacies of newly developed expandable cages require further study. Evidence regarding the expandable cages was insufficient, and the advancement of expandable devices could improve clinical outcomes with novel technology. Therefore, the current study compared the titanium expandable cage with the conventional static cage in TLIF patients.

A prospective study was conducted to elucidate the radiographic and clinical outcomes of expandable cages used for TLIF. The control group consisted of TLIF patients with static cages. The Institutional Review Board (IRB) approved the study protocol (IRB No. (954)111-33), and informed consent from all patients was obtained before the operation. The participants were not randomly assigned to each group; after the surgical consultation, they were allocated to either group according to their personal preference. A single surgeon performed all surgeries in the study.

The study was conducted from October 2022 to January 2024. The patient allocation criteria were as follows: (1) age was at least 20 years, (2) chronic low back pain with or without radiculopathy due to spondylolisthesis or spinal stenosis, (3) at least 12 weeks conservative treatment, (4) disease level was between L1–2 and L5–S1, (5) without previous lumbar spine surgery, (6) without spinal neoplasm or infection, and (7) without history of metabolic bone disorders. The follow-up period was 6 months.

The patients were divided into static and expandable groups, respectively. Polyetheretherketone (PEEK) static cages were routinely used for all patients before October 2022. Expandable cages (LIF-PISCES, BAUI, New Taipei City, Taiwan) were available and used afterward. The expandable cages were constructed with titanium incorporated into the expanding mechanism in the cranio-caudal direction. Patients in both groups underwent standard minimally invasive TLIF with percutaneous transpedicular screws fixation. The bone grafts for fusion were a mixture of autograft and demineralized bone matrix.

The clinical outcomes were evaluated by patient-reported outcomes, including the visual analogue scale (VAS) score for back and leg pain and the Oswestry Disability Index (ODI). A project assistant recorded the measures preoperatively, 1, 3, and 6 months postoperatively at the outpatient clinic. Radiographic assessments were performed using standing anterior-posterior, lateral plain radiographs and dynamic flexion-extension films. The disc height, segmental lordosis (SL), and LL were measured using lateral radiographs. The intervertebral disc height was the mean of the sum of the height at the anterior and posterior edges. To correct the magnification of the plain films, the measured disc height was normalized with the anteroposterior diameter of the upper vertebral body to get the disc height index (DHI). Fusion was defined by the dynamic radiographs as follows: (1) an angular change of the index level measured by Cobb angle of the endplates was less than 2°, and (2) the change in distance between any 2 spinous processes on the dynamic was less than 2 mm. Cage subsidence was measured using lateral radiographs and defined as more than 2 mm sinking of the cage into the endplate. Two independent observers made all radiological interpretations, and the mean of the values was used for analysis.

All continuous variables were expressed as the mean±standard deviation and tested for normality using the Shapiro-Wilk test. Normally distributed variables were analyzed using an independent sample t-test. A Wilcoxon signed-rank test was used to compare the change in the radiographic parameters. The VAS and ODI scores were analyzed using repeated measures to analyze variance. Categorical variables were compared using the chi-square test or Fisher exact test. All statistical analyses were conducted with IBM SPSS Statistics ver. 22.0 (IBM Co., Armonk, NY, USA). Statistical significance was defined as p<0.05.

Forty patients were enrolled in the study: 20 in the expandable group and 20 in the static group. The mean age of the expandable group was 68.0±7.4 years, similar to the static group, which was 68.9±6.9 years. There were 8 males and 12 females in the expandable group. The static group included 9 males and 11 females. Nine patients underwent 2-level fusion in the expandable cage group. Ten patients in the static cage group underwent 2-level fusion. There were no significant differences in patient characteristics, including body mass index, smoking, bone mineral density (BMD), or surgical levels. All patients had a follow-up at least 6 months after the operation (Table 1).

The VAS of back or leg pain and ODI scores showed significant improvement after the operation in each group. The mean VAS of the back in the expandable cage group decreased from 5.9±3.5 to 1.9±1.8 in the first-month postoperation (p<0.05). The mean VAS of the leg in the expandable cage group also improved from 3.7±4.0 to 1.0±1.6 (p<0.05). Patients in the static cage group also improved in both back and leg pain after the operation (VAS of the back from 6.2±4.4 to 2.3±2.5; VAS of the leg from 4.1±3.9 to 1.5±1.3, p<0.05). The pain improvement remained influential during the 6-month follow-up. After the operation, the patients’ ODI in the expandable cage group significantly decreased from 23.4±7.7 to 15.7±8.0 (p<0.05). Patients in the static cage group also had significant functional improvement (ODI from 26.3±9.4 to 17.4±7.3). There was no significant difference between the groups in VAS and ODI scores at each assessment point (Table 2). There were no complications such as neural injury, dural tear, implant failure or migration, or wound infection in both groups. There was no revision surgery during the study period in both groups.

The fusion rate assessed by dynamic radiography was significantly higher for the expandable cage group than the static cage group at 6-month follow-up. Five and 4 patients had cage subsidence during the 6-month follow-up in the expandable and static cage group, respectively. The DHI significantly increased from 0.33±0.06 to 0.47±0.08 in the expandable cage group (p<0.05). The disc height increase was also significant in the static cage group (from 0.35±0.11 to 0.43±0.13, p<0.05). There was no significant difference between the 2 groups in disc height reduction. As for the LL and SL, both groups had an increased trend after the operation. However, there was no significant difference regarding the lordosis between the groups (Table 3).

The intervertebral cages for lumbar interbody fusion have evolved in different aspects, such as material, shape, structures, and mechanics. Recently, expandable cages have shown promising results and gained much attention. By different mechanics, the expandability of the newly designed cages can be placed into the intervertebral space through a smaller channel and bone window. This feature also helps decrease paraspinal muscle injury and lower the risk of neural injury during cage implantation. The expandable cage is superior to having a better reduction in intervertebral space and correction in SL [6]. An efficient fusion could be possible with an ideal cage-bone interface material and structures [7,8].

Though the present study did not show better correction in the alignment of sagittal parameters by expandable cages, it revealed that a porous titanium expandable cage could enhance bone fusion in TLIF patients. The efficient bone fusion can promote stability and decrease the subsidence rate. Comparable clinical outcomes in the current study were consistent with most previous studies [9-11]. One retrospective study by Woodward et al. [12] showed a more pronounced improvement in pain and ODI in the expandable cage group. The expandable cage used in their study was multidirectional, and the radiological results showed a significant increase in disc and foraminal height. The bone graft was morselized autologous bone and recombinant bone morphogenic protein. The fusion rates were more than 90% and similar between the 2 groups at the 1-year follow-up. Similar fusion rates and better restoration of disc height and SL were reported by Hawasli et al. [13] in a retrospective analysis. These differences in cage profiles require further study to show the benefits of the specific mechanics or materials.

Previous studies have reported controversial results in cage subsidence when expandable cages are applied [14]. A recent meta-analysis showed no significant difference in subsidence rate between the expandable and static cages [15]. Some risk factors have been discussed, such as cage position, BMD, obesity, or stand-alone cage fusion [16-19]. A larger size of the static cage is usually necessary to achieve an adequate reduction of intervertebral disc height and expand the intervertebral foramen dimension. However, using a sizeable static cage may cause iatrogenic endplate injury while inserting the device [20,21]. The endplate injury is related to subsidence, biomechanical instability, and failure of indirect decompression or fusion [22,23]. Theoretically, an expandable cage may decrease the endplate injury while correcting the SL [24]. The cage is compact for effortlessly passing the tract into the intervertebral space before expansion. Therefore, an expandable cage can avoid vigorous abrasion that might cause intraoperative endplate injury. Though the distraction of an expandable cage can increase the disc height and improve sagittal alignment, it is crucial to avoid excessive force to achieve greater disc height or lordosis [25]. The overdistraction of the intervertebral space may contribute to subsidence that compromises the restoration of disc height and alignment [26]. In one retrospective study, there was a trend of lower satisfaction due to subsidence in patients using an expandable cage for TLIF [27]. Therefore, the surgeon should operate the expandable cage carefully to avoid distraction and endplate injury.

The methods for evaluation of fusion status include dynamic radiographs, computed tomography (CT), and surgical exploration. In general, the most common method for postoperative assessment in the outpatient clinic is plain radiography. Plain radiography is convenient to evaluate the patient at each time point. The previous studies also applied dynamic radiographs to assess fusion status. The lack of significant motion between vertebrae is highly suggestive of successful fusion. Though CT imaging can provide accurate 3-dimensional anatomical information to confirm intervertebral bridging bone, serial CT scans during each postoperative follow-up time do not adhere to ethical principles due to increased radiation exposure to the participants. Therefore, dynamic radiography was used in the current study to evaluate the fusion status. However, there were no available criteria with sufficient accuracy and reliability for radiographic assessment of fusion status. The standard descriptive criterion on plain radiography or CT includes osseous bridging, absence of motion, or absence of static signs of nonunion [28]. The variable criteria might affect the results. To improve the reliability of the radiological results, combined assessment with plain radiography and CT scan is necessary in future studies.

The present study has some limitations. A randomized controlled trial minimizes selection bias and confounding factors while evaluating and comparing the different devices. The sample size was small, and the follow-up period was relatively short. Therefore, the current study might have low statistical power to reach a conclusion. It is difficult to conclude the superiority of the current expandable cage in terms of subsidence or durability in correcting lordosis. Limited data may not highlight the differences between static and expandable cages. There was a lack of CT images to evaluate the fusion status.

The present study showed that the titanium expandable cage can provide an efficient bone fusion compared with the static PEEK cage. Though the clinical outcomes and correction of the sagittal alignment were comparable, the newly designed expandable cage may be an alternative in TLIF patients. Further studies are necessary to evaluate the benefits and long-term outcomes of applying expandable cages.