Introduction
Lateral lumbar interbody fusion (LLIF) is a less invasive surgical technique used to treat various spinal conditions, including degenerative disc disease, spinal stenosis, and spondylolisthesis [1]. The LLIF procedure offers the advantages of a large bone graft area, high fusion rate, reduced bleeding, limited muscle dissection, and a lower complication rate, while also providing necessary indirect decompression for treating spinal canal stenosis and avoiding nerve injuries [2]. Numerous studies have reported promising clinical outcomes following indirect decompression with LLIF for lumbar degenerative diseases, particularly in elderly patients with multiple comorbidities [3,4]. However, concerns remain regarding implant stability and reoperation when adopting a stand-alone LLIF approach [4].
Recently, LLIF combined with lateral plate fixation (LLIF+LP) has gained increasing popularity as a technique designed to enhance stability in the treated spinal segment. Biomechanical analyses have shown that less invasive adjunctive fixation methods, including LP, can provide sufficient biomechanical stability for LLIF [5,6]. This strategy preserves the minimally invasive nature of LLIF while offering enhanced stability through internal fixation. Additionally, this combined fixation method may reduce cage stress and endplate stress in comparison with pedicle screw fixation [7]. Nevertheless, in certain cases, the indirect decompression effect achieved through LLIF may be inadequate, and may subsequently require a posterior revision for further decompression. While previous studies have examined revision rates and risk factors following stand-alone LLIF [8,9], no research has specifically investigated these outcomes in the context of LLIF+LP, leaving a critical gap in current literature.
Degenerative lumbar disease occurs predominantly in elderly individuals and is often accompanied by varying degrees of paraspinal muscle degeneration, which is increasingly recognized as a significant consideration in spinal surgery [10]. A systematic review and meta-analysis indicated that paraspinal muscle morphometry may predict functional status after lumbar surgery, although its relationship with reoperation rates remains controversial [11]. Moreover, it is still unclear whether paraspinal muscle morphometry contributes to early failure following LLIF+LP surgery. Therefore, the present study aims to investigate the revision rate associated with indirect decompression through LLIF+LP, identify preoperative risk factors related to reoperation, and provide important insights to support clinical decision-making.
Materials and Methods
Study population
This was a retrospective cohort study conducted a single academic institution between June 2020 and July 2023. Patients with degenerative lumbar diseases who underwent LLIF+LP procedures were included. All surgeries were performed by senior spine surgeons at our hospital. Patients were classified into a non-revision group and a revision group. This study was approved by the ethics committee of The Second Affiliated Hospital, Zhejiang University School of Medicine (No. 2023–1042) and written informed consent was obtained from all patients.
The inclusion criteria were as follows: (1) patients who underwent LLIF+LP at the L1 to L5 levels; (2) diagnosis of degenerative lumbar disease with symptoms, including lumbar spinal stenosis (LSS) and/or degenerative spondylolisthesis; and (3) failure of conservative treatment for at least 3 months. The exclusion criteria were as follows: (1) patients with severe cardiopulmonary comorbidities that could influence surgical risk or postoperative recovery; (2) a history of prior multilevel lumbar spine surgeries; and (3) inability to obtain adequate preoperative imaging for radiological assessment.
Baseline demographic data were collected, including age, sex, body mass index (BMI), bone mineral density (BMD), diagnosis, lumbar surgery history, number of fusion levels, surgical segment, first operation time, blood loss, hospital stay, and follow-up duration. Preoperative lumbar computed tomography (CT) was used to measure the Hounsfield unit (HU) value of the lumbar 1 (L1) vertebra to assess BMD [12].
Surgical technique
All patient were treated under general anesthesia and positioned laterally. A small incision was made on the lateral aspect of the abdomen, and the muscles and tissues were carefully retracted to expose the spine. The damaged intervertebral disc (target disc) was excised, and a cage filled with bone fragments was inserted into the central portion of the disc space to restore disc height and provide tension to the ligamentum flavum. A specially designed lateral plate was placed over the treated spinal segment, and screws were inserted through the plate into the vertebral bodies above and below the spacer to enhance stability. The incision was then closed using sutures or staples.
All revision surgeries were performed at the index surgical level. Indications for revision after LLIF+LP included persistent postoperative pain, intraoperative endplate injury, and postoperative cage subsidence. Revision procedures consisted of posterior fusion combined with direct decompression, and all revision surgeries were successfully completed.
Clinical and radiological outcome parameters
The Visual Analog Scale (VAS) for back and leg pain, the Oswestry Disability Index (ODI), and the Japanese Orthopaedic Association (JOA) score were evaluated preoperatively and at the last follow-up. For patients who underwent revision surgery, patient-reported outcomes at the final follow-up were collected after the revision procedure. All patients underwent preoperative lumbar magnetic resonance imaging (MRI) and CT. The grading systems were selected based on previous validation in lumbar degenerative disease populations. Lumbar central canal stenosis (CSS) was classified into four grades (A–D) based on the space of the anterior cerebrospinal fluid [13]. Lateral recess stenosis (LRS) was evaluated on axial T2-weighted image and graded as 0 (no stenosis), 1 (partial lateral stenosis), or 2 (complete lateral stenosis) [14]. Lumbar foraminal stenosis (FS) was assessed on sagittal MRI and divided into four grades: grade 0 (no FS), grade 1 (mild FS), grade 2 (moderate FS), and grade 3 (severe FS) [15]. The Pfirrmann grading system was used to classify lumbar disc degeneration on T2-weighted MRI, ranging from grade I to V [16]. Facet joint degeneration (FJD) was graded on CT as 0 (normal), 1 (mild), 2 (moderate), and 3 (severe) [17]. Cartilage endplates of the target segment were evaluated for Modic changes. In this study, ligamentum flavum thickness greater than 4 mm was defined as ligamentum flavum hypertrophy (Fig. 1) [18]. Fat infiltration (FI) of the lumbar multifidus muscles was visually classified using standard criteria: grade 0 (<10% FI), grade 1 (10%–50% FI), and grade 2 (>50% FI) (Fig. 2) [19].
Statistical analysis
All analyses were performed using IBM SPSS ver. 27.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as means±standard deviation, and categorical variables were presented as counts and percentages (%). The analysis of variance test was used to compare the continuous variables between groups, while the chi-square test or Fisher’s exact test was applied for categorical variables. Associations between radiographic parameters and revision rate were analyzed using univariable and multivariable logistic regression models. Variables with p<0.05 in univariable analysis were included in the multivariable model to reduce degrees of freedom and minimize overfitting. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) and the Hosmer-Lemeshow goodness-of-fit test. Internal validation was performed with 1,000 bootstrap resamples. For patients undergoing multilevel surgery, radiological assessments were performed at the level with the most severe preoperative LSS and/or degenerative spondylolisthesis. Only the side with the greatest severity was considered for analysis. Statistical significance was set at a two-sided p-value <0.05.
Inter-rater reliability for all imaging parameters was assessed by two independent senior spine surgeons. Categorical grading variables (such as FI, CSS, LRS, FS, Pfirrmann grade, FJD, and Modic changes) were evaluated using Cohen’s kappa coefficients, while continuous measurements (such as ligamentum flavum thickness) were assessed using the intraclass correlation coefficient (ICC). Agreement was generally excellent, with representative values including a kappa of 0.82 for multifidus fatty infiltration grading and an ICC of 0.88 for ligamentum flavum thickness.
Results
Demographical and surgical information
A total of 163 patients (195 lumbar fusion segments) were included in this study. All patients were classified into a non-revision group (n=144) and a revision group (n=19), resulting in an overall revision rate of 11.7% (19/163). The primary indications for revision surgery included residual stenosis in eight patients (42.1%), leg pain in five patients (26.3%), cage subsidence in three patients (15.8%), and endplate injury in three patients (15.8%). Patients who underwent posterior decompression revision surgery had a mean time to revision of 8 days (range, 7–89 days). The mean age of the study population was 63.6±8.6 years (range, 35–85 years). Ninety-four patients (57.7%) were female. The mean BMI and BMD were 24.9±3.0 kg/m2 and 124.0±38.6 HU, respectively. The most common diagnosis leading to surgery was LSS/spondylolisthesis (69.9%). Most patients (92.4%) had no previous lumbar surgery history. One hundred and thirty-two patients (81%) underwent a single-level fusion, whereas 31 patients underwent two-level fusions. The most frequently treated surgical segment was L4–5 (82.2%), followed by L3–4 (33.7%), L2–3 (2.5%), and L1–2 (1.2%). The mean operative time for the initial surgery was 74.7±28.0 minutes and the mean blood loss was 15.8±12.7 mL. The mean hospital stay and mean clinical follow-up duration were 7.4±3.5 days (range, 2–21 days) and 628.9±285.5 days (range, 170–1,299 days), respectively (Table 1).
Among all baseline variables, only the mean length of hospital stay showed a statistically significant difference between the two groups. Patients in the revision group had a significantly longer hospital stay (13.7±4.0 days) (Table 1). No hardware complications, vertebral body fractures, nerve damage, infections, or adjacent segment disease were observed at the final follow-up.
Clinical and radiological outcomes
Univariate logistic regression analysis demonstrated that ligamentum flavum hypertrophy and the FI grade of the lumbar multifidus muscle were significantly correlated with the revision rate. No significant differences were observed for other variables, including the CSS grade, LRS grade, FS grade, Pfirrmann grade, FJD grade, or endplate Modic change (Table 2). Furthermore, multivariate logistic regression analysis identified ligamentum flavum hypertrophy (OR, 8.038; 95% CI, 1.757–36.766; p=0.007) and severe multifidus FI (OR, 14.394; 95% CI, 1.650–125.603; p=0.016) as independent predictors of revision (Table 3). The logistic regression model demonstrated good discrimination (AUC=0.786; 95% CI, 0.68–0.893; p<0.001) (Fig. 3) and adequate calibration (Hosmer-Lemeshow test, p=0.703). Internal validation using 1,000 bootstrap resamples confirmed consistent OR estimates and model stability, indicating a low risk of overfitting. Fig. 4 shows a case of a patient with ligamentum flavum hypertrophy who required posterior revision surgery following LLIF+LP. An additional illustrative case of a patient with grade 2 FI of the lumbar multifidus muscle who underwent posterior revision surgery after LLIF+LP is presented in Fig. 5.
Compared with preoperative values, the VAS, ODI, and JOA scores of patients in both groups improved significantly after surgery (p<0.001). However, there were no significant differences between the revision and no-revision groups in preoperative or postoperative patient-reported outcomes for VAS back pain, VAS leg pain, ODI, and JOA scores (p>0.05) (Table 4).
Discussion
LLIF circumvents several challenges and morbidity risks associated with anterior or posterior lumbar interbody fusion surgeries and has gained popularity as a minimally invasive procedure for managing various spinal degenerative diseases [1]. Combining LLIF with lateral plate fixation represents an advanced approach that provides additional stability and may improve surgical outcomes in patients with lumbar degenerative disease. In the present study, 19 patients (11.7%) required a posterior revision procedure for further decompression among those who underwent LLIF+LP. The significant parameters associated with revision risk included ligamentum flavum hypertrophy (>4 mm) and the FI grade of the lumbar multifidus muscle. Clinical outcomes in both groups demonstrated marked improvement in VAS, ODI, and JOA scores, and revision surgery did not adversely affect postoperative results. Overall, LLIF+LP appears to be an effective minimally invasive surgical treatment for appropriately selected patients with degenerative lumbar disease.
Previous research has reported that stand-alone LLIF is associated with a relatively high reoperation rate, reaching up to 26% [8], which may limit its widespread application. Aichmair et al. [20] reported that a 21.2% (11/52) reoperation rate following single-level LLIF for adjacent segment disease. Additionally, revision lumbar fusion procedures have been associated with an increased risk of reoperation and poorer clinical outcomes compared with primary fusions [21]. The revision rate observed in the present study 11.7% is consistent with findings from prior studies on LLIF [22,23]. Interestingly, Nguyen et al. [24] reported an even lower reoperation rate of 5.7% after anterior lumbar interbody fusion or LLIF, reflecting the growing preference for less invasive fusion techniques in appropriately selected patients. Notably, the mean time to revision in our cohort was only 8 days (range, 7–89 days), markedly shorter than that reported in studies on stand-alone LLIF [22,24]. This finding suggests a tendency toward early surgical failure, likely due to inadequate indirect decompression in certain high-risk cases.
Few studies have examined risk factors for revision surgery following LLIF. Wang et al. [25] identified bony LRS as the sole independent risk factor for inadequate indirect decompression after extreme lateral interbody fusion (XLIF), recommending direct decompression in such cases. Nguyen et al. [24] reported that smaller preoperative canal diameters were linked to an increased need for revision decompression following stand-alone LLIF. Another study by Wang et al. [26] indicated that postoperative FS increased the likelihood of revision surgery after LLIF; however, their analysis primarily focused on postoperative rather than preoperative factors. In contrast, our study emphasizes the association between preoperative imaging features and the likelihood of early revision following LLIF+LP. We identified ligamentum flavum hypertrophy (>4 mm) and grade 2 FI of the lumbar multifidus muscle as significant preoperative predictors of early revision. Differences in findings compared with previous studies mat stem from the fact that earlier analyses predominantly involved stand-alone LLIF or XLIF procedures, which lack additional fixation and differ biomechanically from LLIF+LP.
Ligamentum flavum hypertrophy is a well-recognized contributor to LSS [27] and correlates with the severity of neurogenic intermittent claudication [28]. Recent work by Yabe et al. [29] suggests that apparent thickening of the ligamentum flavum on imaging may result not only true hypertrophy but also buckling caused by disc space narrowing. In our study, ligamentum flavum hypertrophy was identified as a significant preoperative risk factor for revision following LLIF+LP, likely due to its role in indirect decompression failure. Excessively thickened ligamentous tissue may not adequately retract even after disc height restoration, leading to persistent neural compression. Therefore, for patients presenting with pronounced ligamentum flavum hypertrophy, direct posterior decompression, including ligament resection, should be considered during preoperative planning to reduce the likelihood of revision surgery.
Recently, growing attention has been directed toward the role of paraspinal muscles as key elements in spinal degenerative pathology. A meta-analysis revealed that multifidus FI may serve as a useful parameter for stratifying patients undergoing lumbar surgery according to their risk of severe functional disability and low back pain [11]. Although evidence regarding that predictive value of FI in multifidus and erector spinae muscles for revision surgery remains inconsistent [11], another systematic review reported an association between preoperative paraspinal muscle degeneration and multiple postoperative complications after lumbar surgery [30]. Our study demonstrated that severe FI of the lumbar multifidus muscle is significantly associated with a higher revision rate following LLIF+LP. This finding suggests that advanced FI may compromise postoperative spinal stability and increase the risk of indirect decompression failure. For patients with severe multifidus FI, posterior spinal fusion and internal fixation may be warranted to compensate for muscle degeneration and enhance overall structural support.
To our knowledge, this is the first study to examine the relationship between ligamentum flavum hypertrophy and multifidus fatty infiltration with revision rates following LLIF+LP, offering novel insights for preoperative planning. Nevertheless, several limitations must be acknowledged. First, this was a retrospective review conducted at a single institution with a relatively small sample size and limited follow-up period, potentially reducing the statistical power of the findings. Multicenter prospective studies with larger cohorts and longer follow-up durations are required to validate these results. Second, the analysis primarily focused on patient-related factors, while important surgical parameters such as cage height, width, and lordosis angle were not evaluated. These factors may influence the effectiveness of indirect decompression, particularly in patients with severe ligamentum flavum hypertrophy. Third, fatty infiltration of the multifidus muscle was assessed using a qualitative grading system (grades 0–2). Although validated, quantitative measures such as cross-sectional area on MRI or fat mass estimation using CT HUs would provide more objective and reproducible assessments. Fourth, sagittal alignment parameters, including lumbar lordosis and segmental angles before and after surgery, were not analyzed. Inadequate correction of segmental lordosis may contribute to reoperation risk. Fifth, certain potential confounders such as smoking status, diabetes, osteoporosis therapy, and long-term corticosteroid use, were not consistently available and could not be incorporated into the analysis. Finally, both single-level and two-level fusion cases were included. These groups may differ in biomechanical characteristics and risk profiles, and subgroup analysis was not feasible due to the limited sample size.
Conclusions
For patients with lumbar degenerative disease, LLIF+LP represents a valuable surgical technique that achieves a low revision rate and favorable clinical outcomes. Preoperative evaluation of ligamentum flavum thickness and multifidus muscle fatty infiltration is strongly recommended. In patients with ligamentum flavum thickness greater than 4 mm or grade 2 fatty infiltration, early consideration of posterior decompression may help reduce the risk of revision surgery.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
