Determinants of lateral fusion in single-level oblique lateral lumbar interbody fusion: a retrospective analysis of fusion patterns and clinical outcomes
Article information
Abstract
Study Design
Retrospective cohort study.
Purpose
This study aimed to (1) determine the incidence of lateral fusion following single-level oblique lateral interbody fusion (OLIF); (2) identify risk factors associated with the development of lateral fusion; (3) evaluate the effect of different fusion patterns on interbody cage subsidence rates; and (4) assess whether fusion patterns influence postoperative clinical outcomes.
Overview of Literature
Fusion characteristics following OLIF differ from those seen in conventional transforaminal lumbar interbody fusion, most notably due to lateral fusion marked by extra-vertebral bony bridging (EVB). EVB may develop early postoperatively, suggesting a potential mechanism for early interbody fusion.
Methods
This retrospective cohort study included 153 single-level OLIF cases between January 2016 and December 2023. Postoperative computed tomography was used to classify patients into central fusion, lateral fusion, and non-fusion groups. Demographic, surgical, and radiographic parameters—including osteophyte grade, Hounsfield unit (HU) values, and cage positioning—were analyzed to identify factors affecting fusion. Cage subsidence and clinical outcomes (Oswestry Disability Index [ODI], Visual Analog Scale) were compared across groups.
Results
Lateral fusion occurred in 39.9% of cases, central in 56.9%, and non-fusion in 3.2%. Preoperative osteophytes and higher HU values were associated with lateral fusion (p<0.001). OLIF with standalone cages (OLIF-SA) had a significantly higher lateral fusion rate than OLIF with posterior screw fixation (OLIF-PS) (p=0.002). Smoking was a significant risk factor for non-fusion (p=0.005). No significant difference in cage subsidence was observed between central and lateral fusion, but non-fusion showed more severe subsidence. Clinical outcomes improved across fusion groups, though non-fusion cases had worse ODI scores at follow-up.
Conclusions
Lateral fusion is a distinct OLIF feature influenced by osteophytes, bone density, and fixation type. It does not negatively affect cage subsidence or outcomes, but solid fusion remains essential for recovery. These findings enhance understanding of OLIF fusion and may guide surgical planning.
Introduction
The prevalence of lumbar spine disorders has shown a steady increase, with recent studies estimating that approximately 619 million individuals worldwide suffer from low back pain [1]. Lumbar interbody fusion is a commonly employed surgical intervention for these disorders when conservative treatments fail [2]. Based on the surgical approach, interbody fusion techniques can be broadly categorized into posterior approaches, including posterior lumbar interbody fusion and transforaminal lumbar interbody fusion (TLIF), and anterior approaches, such as anterior lumbar interbody fusion and oblique or lateral lumbar interbody fusion (XLIF) [3]. Oblique lateral interbody fusion (OLIF) has gained traction over the past decade. Initially described by Mayer [4] and formally named by Silvestre et al. [5] in 2012, OLIF has since undergone continuous refinement and is increasingly being adopted for the management of lumbar spine disorders [6].
The OLIF technique is characterized by an anterolateral retroperitoneal approach that accesses the lumbar intervertebral space anterior to the psoas muscle, minimizing disruption to intraspinal structures. This approach circumvents direct manipulation of the spinal canal and adjacent neural structures, allowing for the implantation of larger interbody cages that aid in disk height restoration and improve segmental stability [7]. In many cases, the use of a standalone cage (OLIF stand alone, OLIF-SA) has been shown to achieve favorable clinical outcomes [8]. However, in patients with significant spinal instability or a high risk of cage subsidence, OLIF combined with posterior pedicle screw fixation (OLIF-PS) has been shown to confer greater stability and better clinical outcomes [9].
In conventional TLIF procedures, the time required for interbody fusion typically ranges from 6 to 12 months [10]. Fusion primarily occurs within the intervertebral space, with bony bridging across the endplate largely confined to the intervertebral disk space and rarely extending beyond the lateral margins of the vertebral body. This fusion pattern is primarily centralized within the vertebral body and is therefore categorized as central fusion [11].
In a recent study, our team reported that extra-vertebral bony bridging (EVB) occurred in approximately 50.3% of OLIF-SA cases and 30.3% of OLIF-PS cases [11]. Notably, EVB formation typically occurs early in the fusion process, preceding interbody fusion. This early osseous response may enhance segmental stability and, in some cases, eliminate the need for a second-stage posterior fixation procedure [11,12]. Based on these findings, we define this distinct fusion pattern as lateral fusion (Fig. 1).
Lateral and central fusion in oblique lateral interbody fusion (OLIF): OLIF stand-alone (A–C) and OLIF with pedicle screw fixation (D–F). Extra-vertebral bony bridge of lateral fusion is indicated by white circles in (A, B, D, E), while central fusion is indicated by white arrows in (C, F).
To date, no clinical studies have investigated the factors associated with lateral fusion following single-level OLIF. To address this gap, we conducted a retrospective analysis of patients who underwent single-level OLIF at Sir Run Run Shaw Hospital, Zhejiang University, School of Medicine, Hangzhou, China. The objective was to evaluate the rate of occurrence of lateral fusion following single-level OLIF, identify factors contributing to its development, assess the relationship between fusion patterns and interbody cage subsidence rates, and determine whether different fusion patterns affect postoperative clinical outcomes.
Materials and Methods
Study population
We retrospectively analyzed patients who underwent single-level lumbar OLIF surgery at our institution between January 2016 and December 2023. All procedures were performed by a single senior spine surgeon with extensive experience in OLIF procedures. The study was conducted in accordance with the principles of the Declaration of Helsinki. The research protocol was approved by the Institutional Review Board (IRB) of Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou, China (IRB no., 2025-1051), and written informed consent was obtained from all patients.
Inclusion criteria
Patients were included in the study if they met the following criteria: (1) aged between 18 and 80 years; (2) diagnosis of lumbar spine pathology involving the L1–L5 segments—such as lumbar disk herniation (contained herniation), lumbar spinal stenosis, lumbar spondylolisthesis, lumbar instability, or discogenic low back pain—that had not responded to at least 3 months of conservative treatment; (3) underwent single-level lumbar OLIF surgery; (4) availability of complete imaging data, including preoperative anteroposterior and lateral X-rays, computed tomography (CT), and magnetic resonance imaging (MRI) scans, as well as immediate postoperative and 1-year follow-up X-rays and CT images.
Exclusion criteria
Patients meeting any of the following criteria were excluded: (1) long-term corticosteroid or bisphosphonate use known to affect bone healing; (2) revision surgery at the same lumbar level; (3) interbody fusion performed using non-cage techniques, such as autologous iliac bone grafting; (4) OLIF procedures combined with supplementary posterior decompression of the spinal canal; (5) patients who underwent non-pedicle screw fixation techniques (such as PIVOX); and (6) presence of systemic skeletal disorders affecting bone metabolism and ossification, including diffuse idiopathic skeletal hyperostosis, ankylosing spondylitis, hyperparathyroidism, and rickets.
Surgical technique
The OLIF-SA and OLIF-PS procedures are described in the Appendix 1. For all patients, interbody cages and bone substitute materials were sourced from the same manufacturer. Postoperatively, all patients were encouraged to ambulate with lumbar brace support starting on the first day after surgery, with brace immobilization maintained for 3 months.
General clinical data collection
Collected demographic and clinical data included sex, age, preoperative diagnosis, smoking status, body mass index (BMI), and type 2 diabetes. Surgical parameters included the operated spinal level, operative time, and intraoperative blood loss. Additionally, the surgical approach (left- or right-sided), cage height, and cage width were documented.
Radiographic assessment
Three experienced spine surgeons independently conducted preoperative and postoperative imaging assessments. The evaluated imaging modalities included preoperative anteroposterior and lateral radiographs, CT, and MRI, as well as immediate postoperative radiographs and follow-up imaging, including X-rays and CT scans obtained at 1-year postoperatively.
Fusion Features and EVB Assessment [11]
Lateral fusion was defined by the presence of an EVB on one or both sides, extending beyond the vertebral body in coronal CT images at the final follow-up, accompanied by the formation of a central bony bridge within the disk space (Fig. 1A, B, D, E). For EVB assessment, each intervertebral space was evaluated bilaterally. A score of 1 was assigned if EVB was present on one side, and a score of 2 if observed on both sides. Central fusion was defined as Bridwell Grade 1 or 2 on CT imaging, indicating solid fusion occurring primarily within the intervertebral disk space [13,14]. No fusion was defined as Bridwell Grade 3 or 4 on CT imaging, reflecting incomplete or absent interbody fusion [13,14].
Preoperative osteophyte assessment and grading
Preoperative osteophyte formation was evaluated using the classification by Nathan [15] (Appendix 2). Osteophytes graded ≥2 were considered indicative of significant osteophyte formation [16]. The assessment was performed bilaterally, with the left and right sides evaluated independently.
Hounsfield unit measurement
Hounsfield unit (HU) values were measured using the method described by Schreiber et al. [17]. Mid-sagittal CT reconstruction images of the lumbar spine were used to guide the analysis. Corresponding axial CT slices displaying the largest visible cortical bone region adjacent to the upper and lower endplates of the adjacent vertebrae were selected for HU measurement (Appendix 3).
Cage-pedicle lateral index
Postoperative lumbar CT scans were used to evaluate interbody cage positioning. Coronal reconstruction images were first examined to identify the lateral marker line of the cage (Appendix 4A). The corresponding sagittal plane was then used to localize this marker line (Appendix 4B). Subsequently, axial CT images at the pedicle cross-sectional levels of the superior (Appendix 4C) and inferior (Appendix 4D) vertebrae were analyzed.
The distance between the lateral border of the pedicle and the designated cage marker line was measured for both the superior and inferior vertebrae. The cage-pedicle lateral index (CPLI) was defined as the mean distance between the cage marker line and the lateral border of the pedicle on superior and inferior vertebrae. A positive CPLI value indicated that the cage marker line was positioned lateral to the pedicle’s lateral border, while a negative value indicated that it was positioned medially within the pedicle’s lateral border.
Cage position index
On lateral X-ray images, the distance from the center of the interbody cage to the anterior vertebral margin was measured (Appendix 5, red line “a”) along with the anteroposterior diameter of the vertebral body (Appendix 5, blue line “b”). The cage position index (CPI) was calculated using the formula a/b.
Cage settling and subsidence
At the final follow-up, sagittal CT images were used to assess the region with the greatest degree of endplate collapse. In cases of bilateral endplate collapse, the measurement was taken from the side exhibiting the most severe collapse. Endplate collapse of ≤2 mm was defined as cage settling, whereas collapse exceeding 2 mm was classified as cage subsidence. Cage subsidence was further graded based on the extent of endplate collapse: Grade 1: >2 mm to ≤4 mm; Grade 2: >4 mm to ≤6 mm; Grade 3: >6 mm [11] (Appendix 6).
Clinical functional assessment
Clinical functional outcomes were assessed using the Oswestry Disability Index (ODI, 0–50 points) and the Visual Analog Scale (VAS, 0–10 points). These scores were recorded preoperatively, immediately after surgery, and at the 12-month follow-up. Functional outcomes were then compared between different fusion patterns.
Statistical analysis
Normally distributed continuous variables were presented as mean±standard deviation, and between-group differences were assessed using the independent t-test. The Mann-Whitney U test was used for non-normally distributed continuous variables. Categorical variables were expressed as frequencies (percentages) and compared between two or more groups using the chi-square test or Fisher’s exact test. When intergroup differences were statistically significant, pairwise comparisons were conducted using the chi-square test or Fisher’s exact test with Bonferroni correction for multiple comparisons. Multinomial logistic regression was used for variables with three categories. All statistical analyses were conducted using IBM SPSS Statistics ver. 29.0 (IBM Corp., Armonk, NY, USA).
Results
General characteristics
A total of 225 cases of single-level lumbar OLIF were performed during the study reference period. Of these, 153 cases (64 males and 89 females; mean age, 61.9±10.8 years) met the inclusion criteria, including 86 cases in the OLIF-SA group and 67 cases in the OLIF-PS group. Seven cases originally planned for OLIF-SA underwent secondary posterior fixation due to cage migration or subsidence during follow-up. These cases were subsequently reclassified into the OLIF-PS group. As a result, 79 cases were included in the OLIF-SA group, and 74 cases in the OLIF-PS group (Fig. 2).
Study flowchart. OLIF, oblique lateral interbody fusion; OLIF-SA, OLIF stand-alone; OLIF-PS, OLIF with pedicle screw fixation; DISH, diffuse idiopathic skeletal hyperostosis.
Among the 153 cases, fusion assessment at the final follow-up showed that 61 segments (39.9%) achieved lateral fusion, 87 segments (56.9%) exhibited central fusion (Appendices 7, 8), and five segments (3.2%) demonstrated no evidence of fusion. This resulted in an overall fusion rate of 96.8%. Detailed data are presented in Table 1.
Demographic data of included patients (n=153)
Influence of different factors on fusion features
The study cohort was categorized into three groups based on fusion pattern: central fusion (n=87), lateral fusion (n=61), and non-fusion (n=5).
Baseline characteristics
There were no significant differences among the three fusion groups in terms of age (p=0.127), sex (p=0.629), BMI (p=0.583), presence of type 2 diabetes (p=0.171), preoperative diagnosis (p=0.161), surgical segment (p=0.407), cage height (p=0.663), cage length (p=0.091), CPLI (left: p=0.357; right: p=0.281), or CPI (p=0.884) (Table 2).
Influence of different factors on fusion feature
Smoking status
Smoking status was significantly associated with fusion outcomes (p=0.005). The non-fusion group had the highest prevalence of smokers (80.0%), compared to 21.3% in the lateral fusion group and 14.9% in the central fusion group. However, no significant difference in smoking prevalence was observed between the lateral and central fusion groups (Table 2).
Bone mineral density (HU values)
HU values, used as a surrogate for bone mineral density, differed significantly across the fusion groups (p<0.001). The lateral fusion group exhibited the highest HU values (superior endplate: 467.14±78.11; inferior endplate: 435.11±66.63), while the non-fusion group showed the lowest (325.88±36.74 and 336.12±36.06, respectively (Table 2).
Preoperative osteophytes
Preoperative osteophytes were significantly associated with fusion patterns (p<0.001). Among the 71 patients with osteophytes, 43 showed lateral fusion, 25 showed central fusion, and three showed no fusion. In contrast, among the 82 patients without osteophytes, 62 achieved central fusion, 18 had lateral fusion, and two showed no fusion (Table 2).
Fixation technique
Fusion outcomes differed significantly between fixation types (p=0.002). Among the 79 patients who underwent OLIF-SA, 40 (50.6%) developed lateral fusion and 39 (49.4%) achieved central fusion. In comparison, among the 74 patients treated with OLIF-PS, 21 (28.4%) exhibited lateral fusion, 48 (64.9%) showed central fusion, and 5 (6.8%) had no fusion (Table 2).
Multinomial logistic regression
Multinomial logistic regression analysis was performed to identify factors associated with different fusion patterns, using central fusion as the reference category. Several variables were associated with an increased likelihood of lateral fusion compared to central fusion. Notably, higher HU values in both the upper vertebral inferior endplate (UVIE) and lower vertebral superior endplate (LVSE) regions were positively associated with lateral fusion. Specifically, for each unit increase in UVIE HU, the odds of lateral fusion increased by 0.7% (odds ratio [OR], 1.007; 95% confidence interval [CI], 1.000–1.013; p=0.042), while each unit increase in LVSE HU was associated with a 0.9% increase in odds (OR, 1.009; 95% CI, 1.001–1.017; p=0.022). The presence of preoperative osteophytes significantly increased the likelihood of lateral fusion (OR, 0.373; 95% CI, 0.151–0.924; p=0.033). In contrast, the use of a pedicle screw fixation was associated with a significantly lower rate of lateral fusion (OR, 2.66; 95% CI, 1.081–6.543; p=0.033) (Table 3).
Multinomial logistic regression analysis for fusion feature
Influence of preoperative osteophyte grading and surgical approach on EVB
Preoperative osteophyte grading
A significant association was observed between preoperative osteophyte severity and EVB formation (p<0.001), suggesting that osteophyte development may play a critical role in EVB formation following OLIF. Among patients with Grade 0 osteophytes, only seven exhibited EVB, while 44 had no EVB, and two experienced non-fusion. Grade I osteophytes were present in 25 EVB cases, compared to 139 cases in the non-EVB group. In contrast, higher-grade osteophytes (Grade II–III) were associated with an increased incidence of EVB. Specifically, Grade II osteophytes were observed in 20 EVB cases and 30 non-EVB cases. Grade III osteophytes demonstrated the strongest association with EVB, present in 25 EVB cases compared to only six in the non-EVB group (Table 4).
Influence of preoperative osteophytes and surgical approach on EVB formation
Surgical approach
No significant differences were observed among the groups regarding the laterality of the surgical approach (p=0.920) (Table 4).
Effect of fusion feature on cage subsidence
Significant differences in cage subsidence rates were observed across the three fusion groups (p=0.002) (Table 5). The non-fusion group exhibited the highest subsidence rate (100%), whereas both the central (58.6%) and lateral (60.7%) fusion groups had significantly higher proportions of cases without subsidence (p<0.05). The overall rate of cage settling was 22.9%, with no significant difference between the central (25.3%) and lateral (21.3%) fusion groups.
Effect of fusion feature on cage subsidence
Regarding subsidence severity, Grade 1 subsidence occurred at similar rates in the central and lateral fusion groups (11.5% each), but was slightly more frequent in the non-fusion group (20.0%). Grade 2 subsidence was most frequently observed in the non-fusion group (60.0%), significantly higher than in the central (3.4%) and lateral (4.9%) fusion groups (p<0.05). Grade 3 subsidence was rare, occurring in only three cases, with a higher incidence in the non-fusion group (20.0%).
Intra-group and intergroup patient-reported outcomes
Pre- and postoperative ODI and VAS scores
Significant improvements were observed in both VAS and ODI scores following OLIF surgery (p<0.001), indicating substantial pain relief and functional recovery over time. VAS scores decreased from 7.4±1.1 preoperatively to 4.0±0.8 immediately postoperatively, and further improved to 2.1±0.8 at the 1-year follow-up (p<0.001). Similarly, ODI scores declined significantly from 39.7±2.8 preoperatively to 28.3±2.1 postoperatively, and further reduced to 18.2±4.0 at the 1-year follow-up (p<0.001).
Comparison of ODI and VAS scores among different fusion features
No significant differences were observed in preoperative VAS (p=0.937) and ODI (p=0.705) scores among the three fusion groups (Table 6). However, postoperative outcomes varied depending on the fusion pattern. At the 1-year follow-up, VAS scores did not significantly differ among the groups (p=0.143). However, ODI scores showed a significant difference across the three groups (p<0.001), with the non-fusion group demonstrating the highest mean ODI score (25.6±4.5), indicating worse functional outcomes compared to the central (17.6±4.0) and lateral fusion (18.5±3.2) groups.
Comparison of VAS and ODI scores between different fusion features
Intra- and inter-observer reliability
The intra- and inter-observer reliability assessments are presented in the Appendix 9.
Discussion
Lumbar interbody fusion is a widely adopted surgical technique with well-established efficacy for managing lumbar spine disorders that are refractory to conservative treatment [3]. In conventional TLIF, interbody fusion typically occurs over 6–12 months [10], with bony bridging predominantly confined to the intervertebral space and seldom extending beyond the lateral margins of the vertebral body [18]. This characteristic fusion pattern is therefore classified as central fusion. Our previous research revealed a higher incidence of EVB formation in OLIF patients, occurring in 49.0% of OLIF-SA cases and 30.3% of OLIF-PS cases [11]. Notably, early EVB formation was visible on anteroposterior X-rays in some patients as early as 3–6 months after surgery (Fig. 3). These findings suggest that, in OLIF-SA, fusion may initially occur along the vertebral periphery—manifesting as lateral fusion—before progressing centrally. In contrast to TLIF, where bony fusion is largely confined to the intervertebral space, EVB formation represents a distinct fusion phenotype, which we refer to as lateral fusion [11].
A 78-year-old female patient with chronic low back pain and neurogenic claudication underwent stand-alone oblique lumbar interbody fusion at the L3–4 level. (A) Preoperative anteroposterior (AP) radiograph; (B) immediate postoperative (IPO) AP radiograph; (C) 3-month postoperative (POM3) AP radiograph; (D) 6-month postoperative (POM6) AP radiograph; and (E) 12-month postoperative (POM12) computed tomography reconstruction. (A) The preoperative radiograph reveals a prominent left-sided osteophyte at the L3/4 level (arrowhead). (C) At 3 months postoperatively, mild leftward retropulsion of the cage is observed (arrow). Nevertheless, the presence of bridging bone on the left side (arrowhead) indicates early lateral fusion, which contributes to segmental stability despite minor cage migration. (D) At 6 months, the cage position remains stable with no progression of retropulsion, and the continuity of the lateral bony bridge further increases. (E) By 12 months postoperatively, confirms the establishment of lateral bony fusion (arrowhead), eliminating the need for additional posterior instrumentation.
The clinical impact of lateral vertebral body fusion: reducing the need for second surgery
During clinical follow-up, we observed that some OLIF patients developed lateral fusion as early as 3 months postoperatively. Early lateral fusion plays a crucial role in providing early stability to the cage, thereby reducing the risk of cage migration and potentially eliminating the need for secondary posterior fixation in SA-OLIF cases [11]. In instances of cage retropulsion following OLIF-SA, the formation of early lateral fusion effectively stabilizes the cage, preventing further retropulsion and thereby avoiding the need for a revision surgery (Fig. 3). Takami also highlighted that EVB formation, functioning as an initial fusion mechanism, significantly contributes to early cage stability [19].
Effect of fixation type on lateral fusion
Our study highlights the significant influence of fixation type on lateral fusion patterns. Lateral fusion was more frequently observed in the OLIF-SA group, with a notable proportion of early-onset cases. Conversely, a small subset of patients (3.2%) exhibited interbody non-fusion, most of whom were in the OLIF-PS group. A retrospective review revealed that some patients initially planned for OLIF-SA underwent secondary posterior fixation due to cage migration or subsidence identified during follow-up. Consequently, these cases were reclassified into the OLIF-PS group, contributing to the higher incidence of non-fusion observed in this cohort.
The increased prevalence of lateral fusion in OLIF-SA is likely attributable to preserved intervertebral micromotion [11]. This biomechanical phenomenon is analogous to principles observed in long bone fracture healing, where two stabilization strategies exist: absolute stability (e.g., compression plating), which eliminates micromotion and facilitates direct bone healing without callus formation, and relative stability (e.g., bridge plating), which allows controlled micromotion, promoting callus formation and indirect bone healing [20]. OLIF-SA, by preserving micromotion, creates a biomechanical environment akin to relative stability in fracture healing, thereby encouraging peripheral bone formation such as extra-vertebral bridging. Furthermore, the oblique approach used in OLIF minimizes neural disruption and allows for the implantation of larger and taller interbody cages, which enhance disk space support and provide optimal initial stability [7,21]. However, without supplemental pedicle screw fixation, localized micromotion persists—mimicking the relative stability required for bone remodeling.
Within the lumbar intervertebral motion unit, the periosteum-Sharpey fiber-endosteum complex functions analogously to the periosteum in long bones, contributing to bone remodeling and fusion. Notably, in the present study, the HU values of the superior and inferior endplates in lateral fusion cases were significantly higher than those in central fusion cases. This suggests that higher-quality endplates provide more robust support, confining endplate strain within a physiological range conducive to EVB formation. Under repeated physiological loading in a controlled micromechanical environment, the periosteum-Sharpey fiber-endosteum interface promotes EVB development around the intervertebral space. In contrast, OLIF-PS, which incorporates posterior pedicle screw fixation, substantially limits spinal mobility, thereby reducing micromotion and decreasing the likelihood of lateral fusion.
Effect of preoperative osteophytes on lateral fusion
Our study identified a significant association between preoperative lateral osteophytes and increased lateral fusion rates following OLIF. Notably, Grade III osteophytes exhibited the strongest association with EVB formation, with 32.5% of cases demonstrating EVB. Two potential mechanisms may explain this relationship:
Reduction in fusion distance by lateral osteophytes
Pre-existing lateral osteophytes, particularly high-grade (II–III) formations, effectively shorten the required fusion distance. The intervertebral space typically features a concave central region with convex lateral margins, resulting in a longer fusion pathway through the center [22]. In contrast, lateral osteophytes originating from the apophyseal ring often develop horizontal or converging projections, effectively narrowing the lateral intervertebral gap and facilitating earlier fusion. A similar phenomenon has been observed in the cervical spine. Sheng et al. [23] reported that uncinate process fusion occurs more rapidly than endplate fusion due to the shorter bridging distance. Our findings suggest a comparable phenomenon in the lumbar spine, where higher-grade osteophytes, by reducing lateral intervertebral spacing, are associated with an increased incidence of lateral fusion postoperatively.
Enhancement of fusion potential through osteophyte resection
During OLIF, partial resection of lateral osteophytes is often required to achieve adequate exposure of the intervertebral space. This surgical intervention exposes fresh bone surfaces, creating an optimal biological interface for bone fusion. The exposed bone promotes faster EVB formation, thereby increasing the likelihood of lateral fusion.
Effect of surgical approach on lateral fusion
During OLIF intervertebral space preparation, the cage is inserted through a lateral annular approach, with the contralateral annulus typically perforated prior to cage placement. As the endplates are prepared, bone debris is compressed by the dilator and displaced toward the perforated contralateral annulus. Additionally, surgical disruption of the microvasculature within the vertebral endplates, annulus fibrosus, and adjacent structures—such as the psoas muscle—induces localized bleeding and hematoma formation [24]. These hematomas, rich in organic components, play a critical role in osteogenesis [25]. Although this mechanism would theoretically favor bone bridging on the surgical side, our study found no significant difference in lateral fusion rates between the surgical and contralateral sides. Two potential mechanisms may explain this finding.
Displacement of osteogenic substances to the contralateral side
Bone debris and hematoma are not necessarily confined to the surgical side. Impaction of trial implants may displace these osteogenic materials toward the contralateral side, promoting bone formation beyond the immediate surgical field. Furthermore, disruption of the contralateral annulus during cage placement can also induce hematoma formation, which may enhance EVB formation on the non-surgical side.
Contralateral osteophyte fracture during annular perforation
To optimize apophyseal ring support, contralateral annular perforation is routinely performed during OLIF cage placement. When osteophytes are present on the contralateral side, this maneuver may induce osteophyte fractures, creating a localized bone injury interface that facilitates osteogenic activity and enhances contralateral EVB formation.
Impact of patient factors on fusion pattern: the detrimental effect of smoking
In our cohort, age, sex, obesity, type 2 diabetes, and surgical level showed no significant influence on fusion patterns following single-level OLIF. However, smoking was a significant risk factor for non-fusion. This finding is consistent with previous literature identifying smoking as a key contributor to impaired spinal fusion [26]. The primary mechanism underlying smoking-induced non-fusion involves the direct effect of tobacco-derived toxic substances, such as nicotine, which impair osteoblast proliferation, metabolism, and collagen synthesis. These effects lead to reduced bone density, impaired vascular supply, and compromised local circulation, ultimately hindering intervertebral fusion [27]. Given its systemic impact on bone metabolism, smoking theoretically affects both central and lateral fusion. Our results support this notion: while smoking did not significantly alter the central-to-lateral fusion ratio, it notably increased the risk of non-union.
Additionally, previous studies have suggested that different lumbar spine pathologies create distinct biomechanical environments that may influence clinical outcomes [28]. However, our study found no significant association between specific lumbar spine conditions and the occurrence of lateral versus central fusion patterns. This suggests that the underlying lumbar pathology may not be a decisive factor in determining fusion patterns. Nonetheless, due to the limited sample size of our study, further validation using larger clinical datasets is warranted to confirm these findings.
Cage settling and subsidence rates in OLIF
Recent clinical studies have reported substantial variability in OLIF cage subsidence rates, primarily due to inconsistent definitions of subsidence across studies [29], underscoring the need for a standardized classification.
Several studies have shown that the thickness of the lumbar bony endplate ranges from 0.8 to 1.0 mm. Given this anatomical constraint, minor cage subsidence (≤2 mm) is unlikely to compromise endplate integrity or induce clinical symptoms [30]. Rather, this subtle settling may represent a natural adaptation between the cage and the vertebral structure, contributing to mechanical stabilization.
From a biomechanical perspective, minor cage subsidence (≤2 mm) typically preserves endplate integrity and remains clinically asymptomatic. This self-stabilization phenomenon, in which the cage conforms to the vertebral architecture without compromising structural integrity, is best described as cage settling. Conversely, cage intrusion exceeding 2 mm into the endplate should be classified as cage subsidence, as it compromises structural stability and may contribute to clinical complications.
Our study revealed a significantly higher incidence of severe cage subsidence in non-union cases, emphasizing the clinical relevance of this classification. Specifically, cage settling is typically asymptomatic and does not require intervention, whereas more pronounced cage subsidence is associated with increased clinical symptoms and lower fusion rates, necessitating closer monitoring and potential treatment. Recognizing this distinction may help refine postoperative management strategies and improve patient outcomes following OLIF-SA.
Clinical efficacy of OLIF: the essential role of successful fusion
Our findings confirm the clinical efficacy of OLIF, demonstrating significant postoperative improvements in functional outcomes. Notably, there was no significant difference in clinical function between lateral and central fusion patterns. However, patients who failed to achieve fusion exhibited significantly poorer postoperative functional scores. These results highlight the importance of successful fusion—regardless of pattern—for optimizing long-term recovery following OLIF.
Limitations
This study has several limitations. First, the retrospective study design and relatively small sample size may limit the generalizability of the findings, warranting validation through large-scale prospective studies. Second, the inclusion of patients with diverse lumbar spine pathologies introduces heterogeneity, which may influence interbody fusion patterns and affect the interpretation of statistical outcomes. Third, the follow-up duration was limited to 1 year, by which time fusion rates were already high, potentially limiting further evaluation of fusion progression. Fourth, due to the retrospective nature of the study, CT scans at critical interim time points (e.g., 3 and 6 months postoperatively) were not available, limiting the ability to assess early-stage fusion dynamics. Fifth, the study did not include an analysis of complications such as infection, neurological injury, or implant-related issues, which are important for a comprehensive evaluation of surgical outcomes. Sixth, the absence of a comparative analysis of different surgical approaches limits the generalizability of the findings to specific fusion approaches. A comparative evaluation of various techniques would provide deeper insights into how surgical variations influence fusion patterns and outcomes.
Conclusions
OLIF achieved a high overall fusion rate, with lateral fusion—defined by unilateral or bilateral EVB formation—observed in 39.9% of cases. This fusion pattern is distinct from that seen in traditional TLIF.
Several factors were associated with increased rates of lateral fusion, including the use of OLIF-SA, higher preoperative HU values of adjacent endplates, and the presence of preoperative osteophytes. Smoking was significantly associated with non-fusion. Other variables, such as age, sex, obesity, type 2 diabetes, surgical approach (left vs. right), cage parameters (height, length, anterior-posterior and medial-lateral positioning), surgical level, and preoperative diagnosis, showed no significant association with lateral fusion patterns.
Cage settling and subsidence rates did not significantly differ between the lateral and central fusion groups. However, non-fusion cases demonstrated a significantly higher incidence of severe subsidence. Clinically, OLIF resulted in significant improvements in functional outcomes both immediately after surgery and at the 1-year follow-up. However, patients with non-fusion had poorer functional outcomes.
Key Points
Lateral fusion is a distinct feature of oblique lateral interbody fusion (OLIF), marked by early extra-vertebral bony bridging.
OLIF stand alone, high endplate Hounsfield unit values, and preoperative osteophytes increase the likelihood of lateral fusion, while smoking increases the risk of non-fusion.
Cage subsidence is most severe in non-fusion cases, but rates are similar between lateral and central fusion groups.
All fusion types improve clinical outcomes, but non-fusion patients show significantly worse postoperative Oswestry Disability Index scores.
Solid fusion—lateral or central—is critical for optimal recovery after OLIF.
Notes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Funding
This study was supported by the National Natural Science Foundation of China (Grant No., 82272521, 82302732), Zhejiang Provincial Natural Science Foundation of China (Grant No., LZ23H060002, LQ23H060006).
Author Contributions
All authors contributed to the management and analysis of the case. Conceptualization: TY, SH, FC, ZF. Methodology: TY, LJ, HB. Data collection: TY, SH, FC. Data curation: TY, SH, FC. Formal analysis: TY, SH. Visualization: TY, FC, LJ, ZF. Funding acquisition: LJ, FS, ZF. Project administration: FS, ZF. Supervision: LJ, ZF. Writing–original draft: TY, SH, FC, LJ, HB, FS, ZF. Writing–review & editing: TY, SH, FC, LJ, HB, FS, ZF. Final approval of the manuscript: all authors.