Introduction
Transforaminal lumbar interbody fusion (TLIF), introduced by Harms and Rolinger [1] in 1982, has become a widely accepted standard surgery for most lumbar degenerative conditions, including spondylolisthesis, spinal stenosis, and discogenic pain with signs of instability. Over time, modifications to the original technique have aimed to minimize muscle trauma while maintaining stability. These advancements include minimally invasive techniques, endoscopic approaches, bilateral cage insertion, modular interbody cages, and combinations of unilateral screws with either spinous process anchors or contralateral translaminar facet screws [2–4].
Several meta-analyses have compared unilateral pedicle screw fixation (UPSF) with TLIF to bilateral pedicle screw fixation (BPSF), revealing no significant differences in terms of fusion rate, pain relief (Visual Analog Scale [VAS] scores for back and leg pain), functional disability (Oswestry Disability Index [ODI]), complication rates, or length of hospital stay [5–7]. The UPSF group showed the advantages of lesser blood loss and shorter operation time but with a higher incidence of cage migration [6,7].
Foraminal stenosis is a major cause of leg pain (and to some extent axial back pain) in degenerative lumbar disorders due to neural ischemia, venous congestion, and nerve conduction defects [8]. It can be addressed surgically by direct decompression with facetectomies or indirectly by increasing the intervertebral disc height. This concept of indirect decompression has been extensively studied in oblique lumbar interbody fusion and lateral lumbar interbody fusion, and to a lesser extent in TLIF/unilateral-TLIF [9,10]. Hunt et al. [11] reported that 2.5% of patients undergoing TLIF procedures developed new-onset contralateral radiculopathy. Therefore, analysis of foraminal morphological changes is crucial, especially in unilateral-TLIF surgery, where the contralateral side remains uninstrumented.
Furthermore, biomechanical studies have shown that unilateral fixation constructs are less stiff than bilateral fixation constructs [12]. This stable fixation with lesser rigidity can prevent adjacent segment degeneration (ASD), an inevitable consequence of fusion surgery [13].
The indirect decompression effects of UPSF–TLIF or its mid- to long-term effects on adjacent discs have not been investigated in contemporary literature [9]. Our study addresses this knowledge gap by investigating the outcomes of over 300 UPSF patients treated at the Apollo Hospitals, Chennai, India, shedding light on these critical aspects.
Materials and Methods
The study protocol was approved by the Institutional Ethics Committee of Bio Medical Research, Apollo Hospitals, Chennai (IRB approval no., AMH-C-S-032/05-24).
Study design
This was a retrospective matched case-control study adhering to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. Informed consent was obtained from all patients. We retrospectively reviewed 825 consecutive patients who underwent single-level TLIF surgery with either UPSF or BPSF between 2012 and 2020. Patients were included if they were skeletally mature with predominant lower-limb radiculopathy or claudication/back pain secondary to degenerative disc disease, grade 1 degenerative spondylolisthesis, or recurrent disc protrusion, and had failed conservative treatment for at least 6 weeks. The extended indications for USPF-TLIF were patients with severe lumbar canal stenosis, bilateral radiculopathy, or degenerative scoliosis. Patients with grade 2 or higher spondylolisthesis, lytic spondylolisthesis, active infection, multilevel involvement (>2 levels), osteopenia (dual-energy X-ray absorptiometry T score <−1.5), and body mass index (BMI) of >30.0 kg/m2 were excluded.
Out of the 407 UPSF–TLIF and 418 BPSF–TLIF patients reviewed, 319 and 331 patients, respectively, who completed 4-year follow-ups and met matching criteria formed the study population. All surgeries were performed by a single surgeon using a standardized pedicle screw system. Interbody cages were made of either titanium or polyetheretherketone material, with either bullet or banana shapes.
Demographic characteristics of patients, operative time, blood loss, length of hospital stay, implant costs, and complications were evaluated. Functional outcomes were assessed using VAS score, ODI, and 36-item Short Form Health Survey preoperatively and at 6 months, 1 year, 2 years, and 4 years post-surgery. Plain lumbar spine radiographs were obtained at each follow-up to assess fusion and monitor complications. Fusion rates were assessed using Bridwell interbody fusion grading. ASD was assessed in terms of loss of disc height, instability, or facetal arthropathy. Computed tomography (CT) was used to assess patients with a doubtful union in plain radiographs. Postoperative follow-up magnetic resonance imaging (MRI) was performed in select patients with preoperative severe canal/neural foraminal stenosis. Parameters compared were space available for cord/cross-section area of cord (CSA), foraminal height (FH), and foraminal width (FW), as described by Fujiwara et al. [14] with modifications [9,15,16]. CSA was measured on axial T2-weighted MRI at mid-disc level. Similarly, foraminal morphology was assessed on sagittal images at the level of the mid-pedicle (Fig. 1A, B).
Statistical analysis was performed using IBM SPSS ver. 26.0 (IBM Corp., Armonk, NY, USA). Pre- and postoperative data were compared using the Wilcoxon signed rank and Friedman tests, followed by post-hoc tests. Mann Whitney U test was applied to compare continuous variables between two groups, while the chi-square test was used for categorical variables. Two-tailed p-values <0.05 were considered indicative of statistical significance for all tests.
Surgical technique
UPSF–TLIF
The patient was positioned prone on a Jackson table. A standard midline posterior approach was utilized, and paraspinal muscles were elevated subperiosteally only on the symptomatic side. Pedicle screws were inserted using a free-hand technique (Fig. 1D). On the same side, a partial medial facetectomy, laminotomy, and decompression were performed. The disc space was approached after pedicle screw-based distraction or using interbody distractors. The disc was resected using standard instruments (rongeurs, shavers, and curettes). The disc space was then packed with autologous graft, prepared from facetectomy and laminotomy tissue, combined with bone marrow aspirate (BMA) harvested from the posterior iliac crest using a Jamshidi needle. In cases of inadequate autologous bone graft, 1 g of demineralized bone matrix was used along with BMA.
An adequate-sized single cage filled with graft and BMA was inserted into the prepared disc space. For banana cages, the modular handle was used to orient the cage in the coronal plane. Following cage placement, pedicle screw-based compression was applied, and the final cage position was confirmed using fluoroscopy. Lastly, the wound was closed in layers.
BPSF–TLIF
The BPSF–TLIF technique differed from the UPSF–TLIF technique primarily in its bilateral approach. Paraspinal muscle elevation and pedicle screw insertion were performed on both sides while preserving midline structures. Medial facetectomy, laminotomy, decompression, and cage insertion were done on the symptomatic side. However, in cases of severe canal stenosis, bilateral decompression was performed. No additional posterolateral fusion was done in any patient.
Results
The clinical characteristics of the study population are summarized in Table 1. The two groups were matched in terms of age, sex, diagnosis, duration of symptoms, and risk factors for fusion (e.g., BMI ≤2 kg/m2, diabetes, and smoking). The follow-up period ranged from 4 to 12 years (mean=50 months).
The mean operating time and intraoperative blood loss were significantly lesser in the UPSF group (p<0.0001). The length of hospital stay was not significantly different, since the majority of patients in both groups were discharged either on the same day or on postoperative day 1. There was no significant between-group difference in terms of pre- and postoperative/follow-up ODI and VAS scores for back and leg pain (p>0.05) (Table 1).
At the 4-year follow-up, radiological evidence of fusion was observed in 94.3% of patients in the UPSF group (Fig. 1B) and 96.4% of patients in the BPSF group; the between-group difference in this respect was not statistically significant (p<0.05). While CT scans showed early fusion through the cage in both groups (Fig. 1D), complete radiographic fusion of the entire disc space occurred earlier in the BPSF cohort compared to the UPSF cohort.
The UPSF cohort showed a significant increase in the CSA of the cord from 78.41±33.03 mm2 preoperatively to 177.05±46.49 mm2 postoperatively (p<0.001) (Figs. 2A–D, 3). This improvement was consistent across all indications. The FH and FW increased from 12.65±3.33 mm and 7.95±1.71 mm preoperatively to 15.86±3.14 mm and 9.34±1.54 mm postoperatively, respectively (p<0.05) (Figs. 2E–H, 3 and Table 2). The BPSF cohort also showed a similar increase in these parameters (p<0.05) (Table 2).
Disc height was adequately restored in the UPSF–TLIF cohort, increasing from 9.7 mm preoperatively to 12.2 mm postoperatively, with a minimal loss of correction to 11.58 mm at the last follow-up (p<0.001). Similar maintenance of disc height and significant correction of segmental lordosis was attained in the BPSF group (i.e., 11.76o–15.46o) (Fig. 4A–C, Table 3).
In the UPSF group, adjacent segment disc heights remained relatively stable, measuring 11.392±2.03 mm preoperatively, 10.975±1.93 mm at 4-year follow-up, and 10.03±1.88 mm in some patients at 8-year follow-up (Fig. 4A–C, Table 3). The BPSF group exhibited a higher incidence of ASD affecting 14 patients (4.47%) at 4–8-year follow-up. This was significantly higher than that in the UPSF group, where ASD occurred in only three patients (1.06%) (p<0.04). Nine of these 14 patients in the BPSF group underwent surgery for ASD using UPSF at the adjacent level, either retaining or removing the previously operated level implants (Fig. 4D–F). The remaining five patients underwent BPSF–TLIF. In contrast, among the three UPSF patients with ASD, two were converted to bilateral fixation and extended to a higher level, while one patient was extended with UPSF–TLIF. There was a significant difference between preoperative and last follow-up lower adjacent disc height values in UPSF compared to the BPSF group (Table 4).
Complication rates were comparable between the UPSF and BPSF groups (6.58% versus 6.04%, p>0.05) (Table 1). Superficial surgical site infections in both groups responded well to oral antimicrobial therapy. Postoperative radiculopathy due to pedicle screw malpositioning occurred in one UPSF patient and two BPSF patients but improved after repositioning. Two BPSF patients experienced unexplained foot drop (indirect neural injury). One patient achieved full recovery within 3 weeks, while other patients had persistent weakness. Most cases of cage migration were documented between 6 weeks and 3 months postoperatively. While three cages required repositioning, the remaining cage migrations were asymptomatic, ultimately resulting in solid arthrodesis. The pseudarthrosis rate in the UPSF group was higher than that in the BPSF group (2.5% versus 1.5%), but this difference was not statistically significant (p>0.05).
Discussion
TLIF is the preferred technique for treating lumbar degenerative disc disease compared to postero-lateral fusion and posterior lumbar interbody fusion (PLIF) [17,18]. UPSF for TLIF was developed to minimize muscle trauma, preserve posterior elements, and reduce gross destabilization of the spine.
Kabins et al. [19] in 1992 and Suk et al. [20] in 2000 evaluated the relative clinical effectiveness of UPSF against BPSF, demonstrating similar clinical and radiologic outcomes. Over the last 2 decades, several systemic reviews and meta-analyses, have consistently shown that unilateral surgical technique provides equivalent fusion rates with reduced intraoperative bleeding, shorter hospital stays, and faster rehabilitation [5–7,17,18]. Our pilot study conducted in 2022 corroborated these findings [21]. The present study also showed similar findings. Additionally, at our institution, the total hospital cost was 15% less for UPSF–TLIF than for the BPSF construct.
Few studies have raised concerns regarding the higher rates of pseudoarthrosis, cage migration, and subsidence after UPSF [17–20,22]. The fusion rates in our study were comparable between the two groups (94.3% in the UPSF group and 96.4% in the BPSF group). These results align with previous studies by Said et al. [17], Fujimori et al. [18], Kabins et al. [19], and others, which documented fusion rates of 84.6%–93.8% for unilateral instrumented TLIF and 94.3%–96.3% for bilateral instrumented TLIF. The high fusion rates observed in our study can be attributed to meticulous disc-space preparation, adequate autologous bone grafting/usage of BMA in the fusion bed, and usage of the largest-sized cages. Additionally, retaining the posterior midline structures, especially the interspinous and supraspinous ligaments, likely contributed to enhanced stability [23]. A recent meta-analysis by Xu et al. [7] reported a nonunion rate of 10.29% in the unilateral group and 4.73% in the bilateral group, although the pooled estimates were not significantly different.
The use of modular banana cages (signature cage; Globus Medical Inc., Audubon, PA, USA) facilitated optimal cage placement, boasting a large surface area and high volumetric graft holding capacity. Its insertion handle enabled precise positioning. Biomechanical studies have shown that anterior and coronal cage placement at the anterior 1/3 to middle 1/3 junction significantly enhances stiffness, with the cage sharing more load under axial compression compared to posterior cage placement [24]. Postoperative CT scans revealed a favorable disc occupancy ratio, with the cage occupying approximately 30% of the disc space and the anterior 20% occupied by graft, resulting in nearly 50% support (Fig. 1C). This optimal graft-cage configuration likely contributed to the study’s high fusion rates.
The UPSF–TLIF technique utilized large-sized cages to increase the disc space, with cage lengths selected to extend to the contralateral side of the disc space. This ensured indirect decompression, especially on the contralateral unopened side, by increasing its FH (Fig. 2E–H) [25]. The clinical symptoms were relieved by the reduction of the bulging annulus and stretching of the redundant ligamentum flavum, increasing the FW [25]. Notably, maintained disc heights at the operated level were observed in both groups, with no significant cage subsidence at the 4-year follow-up (p<0.001).
At the last follow-up, the FH and FW showed significant increases of 3.21 mm and 1.39 mm, respectively (p<0.05). Kim et al. [9] analyzed 66 MRI scans and found that unilateral-approach minimally invasive TLIF significantly improved both quantitative and qualitative parameters of the central canal and contralateral neural foramen. They noted significant postoperative increases in FH and FW from 11.8±2.0 to 14.7±2.5 and 4.9±1.5 to 6.5±1.8, respectively (all p<0.001).
Our study revealed a 2.3-fold increase in CSA in the UPSF–TLIF cohort, comparable to the BPSF–TLIF cohort. Liu et al. [25] reported a similar increase in CSA from 164.36±37.13 mm2 to 211.59±48.12 mm2 at the last follow-up after MIS TLIF with unilateral side cage insertion.
Sethi et al. [3], Goel et al. [12], and Lee et al. [13] used cadaver specimens to demonstrate that UPSF–TLIF provides less rotational and lateral bending stability, resulting in off-axis movement, and less stiffness compared to BPSF. However, the necessity of more rigid fixation remains debated [3,9,12]. A finite element study by Ambati et al. [26] revealed that unilateral posterior fixation preserved over 50% of intact motion in lateral bending compared to <10% after bilateral fixation. This has its merits as well as demerits.
Theoretically, UPSF may reduce ASD by decreasing pressure over the adjacent discs. Our study found that the UPSF group better maintained disc heights at the upper and lower levels compared to the BPSF–TLIF group, with statistically significant differences in lower adjacent disc height. This aligns with the experimental study by Goel et al. [12] in 1991 demonstrating that unilateral instrumentation reduces stress shielding on the fixed vertebrae and diminishes peak stress on the upper and lower adjacent levels. However, Fernandez-Fairen et al. [27] suggested that in theory, once fusion is complete, the number of pedicle screws (unilateral or bilateral fixation) may not significantly impact ASD risk. Notably, our study found a lower rate of ASD requiring revision surgery in the USPF group (1.06%) compared to the BPSF group (4.47%) at the 4-year follow-up. Long-term studies are needed to confirm this finding. A 10-year follow-up study by Kim et al. [9] showed a lower rate of radiological ASD and clinical outcomes in the UPSF group (55.9% versus 72.7% in the BPSF group).
The UPSF–TILF may be particularly useful in multilevel degenerative conditions, allowing for targeted intervention (based on the exact source of pain) and significant symptom relief without extensive deformity correction surgery (Fig. 4G). For ASD after PLIF/TLIF, the extension of the construct with UPSF–TLIF at the adjacent level may prevent the cascade of ASD from progressing (Fig. 4D–F).
The adequacy of UPSF for stabilizing two-level constructs has been debated. Zhang et al. [28] demonstrated the safety and effectiveness of two-level UPSF–TLIF with no increased incidence of pseudoarthrosis. The fusion rates were 87.5% (7/8) and 91.2% (11/12) in the unilateral and bilateral groups, respectively. Despite encouraging findings, long-term outcomes of multilevel UPSF–TLIF require further investigation.
Cage migration is one of the main reasons for reoperation, which can cause neurological deterioration and nonunion. Aoki et al. [22] followed-up 125 individuals and identified bullet-shaped cage, pear-shaped disc space, undersized cage, higher pelvic diameter to height ratio, and presence of scoliotic curvature as potential risk factors for cage migration. In our study, 10 incidents of cage migration occurred, especially with bullet-shaped cages at higher levels (L3–L4 and L2–L3). However, meticulous endplate preparation, insertion of maximum-sized cages, anterior–horizontal positioning, and fixing the screws posteriorly under compression minimized cage migration and pseudoarthrosis rates.
This study boasts several strengths, including the largest patient cohort in each group, a single-surgeon series ensuring standardized surgical technique, consistent single-pedicle screw and intervertebral cage systems, and follow-up assessments using validated patient-reported outcome measures. However, some limitations of the study should be acknowledged. This was a retrospective single-center study, which may have introduced an element of bias. Other limitations include the absence of subgroup analysis based on cage morphology or material and the exclusion of multilevel disease. Long-term studies are required to assess the effects of UPSF on ASD.
Conclusions
UPSF in TLIF demonstrates comparable results to BPSF in terms of patient-reported clinical outcomes, fusion rates, and complication rates. However, UPSF offers several advantages over BPSF in terms of shorter operative time, decreased blood loss, shorter hospitalization, lower incidence of ASD, and lower costs in select patients. Furthermore, UPSF facilitates significant indirect neural foraminal and canal decompression on the contralateral side, both radiologically and clinically.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
