Which subtypes of degenerative lumbar spondylolisthesis are suitable for oblique lumbar interbody fusion? A retrospective study in China based on the clinical and radiographic degenerative spondylolisthesis classification

Article information

Asian Spine J. 2025;19(1):112-120

Publication date (electronic) : 2025 February 28

doi : https://doi.org/10.31616/asj.2024.0310

Xianghe Wang^1,*, Hongwei Wang^1,*, Xiaosheng Ma¹, Xinlei Xia¹, Feizhou Lyu^1,2, Haocheng Xu^1,†, Hongli Wang^1,†

¹Department of Orthopaedics, Huashan Hospital, Fudan University, Shanghai, China

²Department of Orthopaedics, Shanghai Fifth People’s Hospital, Fudan University, Shanghai, China

Corresponding author: Hongli Wang, Department of Orthopaedics, Huashan Hospital, Fudan University, No.12 Middle Urumqi Road, Shanghai, 200040, China, Tel: +86-21-52887126, Fax: +86-21-52887126, E-mail: wanghongli0212@163.com; wanghongli@huashan.org.cn

Co-corresponding author: Haocheng Xu, Department of Orthopaedics, Huashan Hospital, Fudan University, No.12 Middle Urumqi Road, Shanghai, 200040, China, Tel: +86-21-52887126, Fax: +86-21-52887126, E-mail: xuhaocheng1993@126.com

*These authors contributed equally to this work as the first authors.

†These authors contributed equally to this work as the corresponding authors.

Received 2024 July 28; Revised 2024 November 16; Accepted 2024 November 17.

Abstract

Introduction

Degenerative lumbar spondylolisthesis (DS) is a common cause of chronic low-back pain, significantly affects patients’ quality of life, and increases the burden on the healthcare system [1,2]. Current treatment options include conservative and surgical interventions [2], such as lumbar interbody fusion (LIF). To address evolving indications and enhance safety, Silvestre et al. [3] introduced a minimally invasive technique called oblique LIF (OLIF), offering a unique approach between the abdominal vascular sheath and the psoas major muscle. Given its promising outcomes [4,5], OLIF represents as a potential treatment option for patients with DS. However, selecting the appropriate surgical technique remains challenging because of the wide pathological spectra, ranging from stable spondylolisthesis with stenosis to unstable segments that cause dynamic symptoms. Currently, no studies have assessed which patients with DS are suitable for OLIF.

Mechanistically, OLIF can facilitate indirect decompression and alleviate dynamic compression by longitudinal bracing, repositioning the vertebra, and restoring stability. Compared with traditional posterior approaches, OLIF does not disrupt the posterior spinal column and allows wide exposure for the placement of a larger interbody graft, leading to higher fusion rates and improved stability. These advantages address the therapeutic needs of DS. Therefore, OLIF is widely perceived as an effective treatment for DS [6,7]. However, its effectiveness can vary expressively because of patient heterogeneity. Combining all patients with DS in an analysis may obscure differences among subgroups with variations in vertebral translation, disc collapse, sagittal alignment, and vertebral mobility [8].

Therefore, this study aimed to determine the suitability of different DS subgroups for OLIF based on a clinical classification. Kepler et al. [9] developed the clinical and radiographic degenerative spondylolisthesis (CARDS) classification system to assist in selecting the appropriate surgical technique. It incorporates three radiographic variables, namely, disk height, sagittal alignment, and translation, and the presence of leg pain. This classification significantly correlates with the Oswestry Disability Index (ODI) and postoperative outcomes [10,11] and has excellent reliability and intraobserver consistency [12], which can help simplify the diagnostics and operative planning.

In this study, the preoperative radiological characteristics were assessed, and the incidence of clinical contraindications for OLIF was examined in different DS subtypes. It aimed to identify the subtypes and proportions of patients suitable for OLIF. The study also sought to leverage the CARDS classification for preliminary screening and more precise recommendations.

Materials and Methods

CARDS classification and sample size

Neutral lateral and flexion–extension radiographs were utilized to measure vertebral translation, kyphosis, and bony apposition using the method by Kepler et al. [9]. Participants were categorized into groups A, B, C, and D (Fig. 1) based on the CARDS classification [9]. In clinical practice, the distribution of patients across subtypes differs from that of existing reports, with types C and A being the most prevalent and least common, respectively. The number of patients with type B was fewer than that described in the original classification paper. Accordingly, pre-stratification was implemented to ensure adequate representation of less common subtypes and maintain statistical power. Instead of conventional random sampling, predetermined inclusion targets were set for each subtype (40 for subtype C and 20 for each of the others). The total sample size (approximately 104) achieved meaningful comparisons across subtypes, with a power of 80%, effect size of 0.8, and significance level of 0.05. The proportional allocation to each group was based on the prevalence reported in a previous study [9] and our clinical observations. The stratified random sampling technique was employed to reduce selection bias.

Fig. 1

The schematic illustration of subgroups of the clinical and radiographic degenerative spondylolisthesis classification system. (A) Type A: complete collapse of the disc space with bony apposition of adjacent vertebral end plates. (B) Type B: disc partially preserved with translation of ≤5 mm. (C) Type C: disc partially preserved with translation of >5 mm. (D) Type D: kyphotic alignment.

To simplify the grouping, the leg pain modifier was not utilized because of the following reasons: (1) All patients included had intermittent claudication or lower limb symptoms. (2) This study focused on magnetic resonance imaging (MRI) and computed tomography (CT) analyses, which reduced the dependency on the leg pain modifier compared with the original classification that relied primarily on lateral radiographs.

Radiological measurements and grading criteria (all observed at the slipped segments)

Preoperative digital lumbar X-ray (anteroposterior, neutral/flexion/extension lateral views), MRI, and CT images were analyzed. The clinical contraindications for OLIF include severe spinal stenosis, significant intraspinal occupancy effects, segmental stiffness, and poor bone quality [6,8].

Central canal stenosis

The most severe degree of stenosis was recorded. This was determined from axial T2-weighted magnetic resonance (MR) images based on GY Lee’s grading system (Fig. 2) [13]. Severe central canal stenosis is indicated by the bundled appearance of the entire cauda equina.

Fig. 2

The schematic illustration of GY Lee’s grading system on central canal stenosis. (A) Mild: anterior cerebrospinal fluid space is mildly obliterated, nerves in cauda equina can be clearly separated from each other. (B) Moderate: cauda equina aggregation. (C) Severe: entire cauda equina as a bundle.

Significant spinal canal occupancy effect

Patients were examined for the presence of the following manifestations: (1) massive disc herniation, in which the herniation extends beyond the intrafacet line [14] on any axial T2-weighted MR image; (2) prolapsed disc sequestration, in which the displaced disc nucleus has lost continuity with the parent disc on MRI [14]; (3) spinal epidural lipomatosis (SEL), with axial epidural fat tissue thickness >7 mm on any T1-weighted MR image [15]; (4) significant bony hyperplasia or ligamentous calcification in the spinal canal, including posterior vertebral osteophytes, ossification of the ligamentum flavum/posterior longitudinal ligament, or facet joint arthropathy on CT images. The thickest axial part of the ossification was >50% of the diameter of the spinal canal. Facet joint arthropathy was graded according to the Pathria grading system (Fig. 3) [13].

Fig. 3

The schematic illustration of Pathria M’s grading system on facet joint osteoarthritis. (A) Mild: mild narrowing and joint irregularity. (B) Moderate: moderate narrowing and joint irregularity, sclerosis, and osteophyte formation. (C) Severe: severe narrowing and almost total loss of joint space, sclerosis, and osteophyte formation. (D) The axial thickest part of ossification was >50% diameter of the spinal canal.

Segmental stiffness

The single-segment lateral Cobb angle was measured on dynamic radiographs to determine the presence of mobility loss (i.e., Cobb angle ≤4°) [16].

Bone quality

Because not all patients underwent dual-energy X-ray absorptiometry (DEXA), the Hounsfield unit (HU) values of the vertebral body on CT images were obtained for bone quality assessment. A threshold of 110 HU was adopted because studies have suggested that vertebral HU values below this level are associated with a higher risk of postoperative cage subsidence [17,18]. In this study, CT was conducted using a uniform protocol.

To avoid intraobserver bias, all EMRs were reviewed by two senior spine surgeons, and a third one was involved in the case of disagreement.

Results

Demographics and clinical characteristics

This study included 100 patients (28 males and 72 females), with an average age of 63.76±9.75 years. Of these patients, 20 were classified into groups A, B, and D and 40 into group C. No significant difference was found in the sex distribution (p=0.471) or slipped segment level (p=0.070) among the groups (Table 1). Patients in group A were significantly older (69.30±6.71 years) than those in group D (60.00±10.05 years, p=0.014). Group B had significantly shorter slippage distances (4.34±0.41 mm, p<0.001). The slippage distances for groups A, C, and D were 7.40±2.32 mm, 7.35±1.70 mm, and 7.67±2.09 mm, respectively, and no significant differences were observed. In group D, the maximum slippage distance was observed predominantly during hyperflexion and was significantly longer than that in neutral positions, signifying an unstable slipped segment.

Table 1

Comparison of demographics and clinical features

Central canal stenosis

Among subtypes, patients exhibited varied severities of central canal stenosis (Table 2, Fig. 4A). Significant differences were observed between groups A and B (p=0.048). Moreover, 50% of the patients in group A and 42.5% in group C had severe central canal stenosis, which were both significantly higher than those in groups B (15%) and D (15%).

Table 2

Degree of central canal stenosis in subtypes

Fig. 4

(A) The stenosis grade of central canal (CC). (B) The level of facet joint arthropathy. (C) The Δlateral Cobb angle in dynamic positions. (D) Significant bony hyperplasia in each subtype. ^*p<0.05. ^**p<0.01. ^***p<0.001.

Significant spinal canal occupancy effect

As shown in Table 1, group A had the highest rates (15%) of massive disc protrusions/herniations and higher rates of posterior osteophytes (10%) and ligament ossification (10%). However, these differences were not significant. Only three SEL cases were observed, and no differences were found among the groups. In the assessment of facet joint hypertrophy (Table 3, Fig. 4B), group C showed the most severe condition. Specifically, 27.5% of the patients with group C had severe hypertrophy, with an additional 17.5% exceeding this severity level. Groups A (15%+15%) and D (10%+10%) showed moderate severity. In contrast, group B exhibited the mildest degree of hypertrophy, with only 10% showing severe hypertrophy. These data indicate a notable variation in facet joint hypertrophy, with group C showing a more severe disease than group B (p=0.033). When posterior osteophytes, ligamentum flavum ossification, and facet joint hypertrophy were considered together as “bony hyperplasia,” the prevalence was significantly higher in group A than in group B (p=0.047) (Fig. 4D).

Table 3

Degree of facet joint arthropathy in subtypes

Discussion

OLIF and other anterior or lateral approaches, including anterior LIF (ALIF) and lateral LIF (LLIF), are all indirect decompression techniques. Although they share similar concerns in patient evaluation, they also have differences. Indirect decompression techniques may not provide adequate decompression in severe spinal stenosis and bony or other forms of spinal canal occupancy. In addition, rigid segments can be challenging to distract during surgery, which may damage the endplate or posterior structures, even leading to surgical failure. Compared with ALIF, the size of the OLIF graft is limited because of the oblique insertion direction, rotational placement, and preservation of the anterior longitudinal ligament. As a result, OLIF provides less effective decompression in severe stenosis than ALIF [19,20]. Specifically, the initial posteromedial trajectory of OLIF carries a small risk of displacing disc or ligamentous material into the central canal or contralateral neuroforamen [19], making it relatively contraindicated in patients with severe central canal stenosis. OLIF provides lower stability when using supplementary fixation because of the smaller graft size but offers higher stand-alone stability because of the preservation of the anterior longitudinal ligament [21]. In patients with facet tropism, OLIF is more suitable than LLIF because its anterolateral approach minimizes disruption to posterior structures and reduces the risk of facet asymmetry and lumbar plexus injury [22]. Overall, although each technique presents distinct strengths and limitations, our findings provide not only insights for OLIF but also references for other indirect decompression procedures in patients with DS.

DS can be categorized into four stages: destabilization, slippage, stenosis, and restabilization [23,24]. Currently, no clear identification criteria have been established for patients with DS suitable for indirect decompression surgery [25]. The results revealed that different DS subtypes have distinct clinical characteristics, and not all patients with DS are suitable for OLIF as a surgical option.

Different age groups may exhibit different DS subtypes. In particular, patients with type A tend to be older than those with type D. Although lower lumbar discs tend to lose height with advancing age [26], disc height alone cannot serve as a universal indicator of painful disc degeneration because it does not contribute to radiating pain. Type A is the near-total loss of the intervertebral space height, indicating advanced degeneration. Furthermore, bone proliferation and contact between endplates improve stability in the spondylolisthesis segment. The improved stability could delay medical attention because patients typically present with intermittent claudication. In contrast, type D is distinguished by segmental kyphosis. With reduced lumbar lordosis (LL), the back muscles receive increased load, resulting in back pain and higher visual analog scale and ODI scores [12,27], which require earlier medical interventions. Furthermore, no significant associations were found between age and other variables.

The severity of central canal stenosis is crucial when evaluating the suitability for indirect decompression surgery. A small dural sac cross-sectional area (<44 mm²) may contribute to indirect decompression failure because approximately 20% of patients with severe lumbar stenosis require direct decompression after an indirect procedure [28,29]. The incidence of severe central canal stenosis is high in groups A and C. Both groups exhibited comparatively extensive slippage distances, which can cause misalignment in vertebral canals and accumulation of contracted ligaments. Group A demonstrated significantly decreased intervertebral height and a more severe degree of osteophyte formation, which exacerbate spinal canal stenosis. This condition leads to a higher incidence of soft tissue or bony compression [23]. However, group D had less severe spinal canal stenosis, despite having a similar maximum slippage distance. This can be attributed to more pronounced compression under dynamic conditions compared with the supine position during MRI. Therefore, restoring LL and intervertebral stability is more critical and more feasible for group D.

Facet joint degeneration is associated with DS development and segmental stability. Sagittal changes in facet joints contribute to the initiation of spinal slippage [23]. Subsequent segmental instability increases joint stress, leading to accelerated hypertrophy and partial stability restoration. Compared with other groups, group C exhibited significantly higher rates of severe facet joint hypertrophy. Conversely, group D showed low rates of severe facet joint hypertrophy, and the presence of kyphosis indicates insufficient support of the anterior spinal column, further exacerbating instability and significantly enlarging the dynamic range of motion (ROM). Similarly, Kanayama et al. [30] found that segmental kyphosis predicted an increased dynamic ROM and easier reduction.

Bone quality is an essential factor in surgical options. In this study, the HU value was used as a surrogate measure of bone quality. Compared with DEXA, the current gold standard, HU values enabled the evaluation of cortical and trabecular bones separately in the degenerative spine [31]. Although CT imaging protocols and patient anatomy can influence the measurements, the consistently strong correlation between HU values and bone strength highlights its utility in clinical settings as a reliable osteoporosis indicator [18,32]. Our results showed no significant difference in bone quality among subtypes.

In summary, group B had a shorter slippage distance, partial loss of intervertebral height, and mild bony hyperplasia. Conversely, group D is characterized by instability and segmental kyphosis. These two subtypes have fewer clinical contraindications and align closely with the benefits of indirect decompression. Therefore, they are particularly suitable for OLIF, and group D would particularly benefit from improved LL correction [33]. For group A, the significant intervertebral height reduction aligns well with the restoration of the intervertebral height through indirect decompression. However, group A exhibited significant bony hyperplasia, segmental stiffness, and severe central canal stenosis simultaneously. These characteristics indicate potential spinal fusion, difficulty in intervertebral distraction, and possible indirect decompression failure. Therefore, although some patients in group A achieved favorable surgical outcomes, OLIF should not be the optimal surgical option for most of them.

Among patients in group C, with a 42.5% prevalence of severe central canal stenosis, many exhibited no additional contraindications. Despite unsatisfactory immediate postoperative relief in OLIF, it can achieve comparable short-term outcomes to transforaminal LIF/posterior LIF in severe lumbar stenosis and show positive long-term prognosis [5,34]. These results are likely attributed to the spinal stabilization. These findings indicate the potential of OLIF for treating more patients with type C. Nevertheless, more robust clinical evidence is needed, particularly for cases involving severe lumbar stenosis. Currently, we recommend comprehensive assessments for group C and personalized treatment planning, with particular attention to facet joint hypertrophy.

This study is one of the few that assessed the suitability of OLIF for different DS subtypes based on the CARDS classification. Findings of combined MRI and CT data analyses offer a more comprehensive and refined understanding of the radiological characteristics of heterogeneous subgroups. Moreover, it allows for faster preliminary identification of patient characteristics and factors that require further consideration.

This study has some limitations. First, it is a small-sample, single-center study. Second, only preoperative radiological data of patients were evaluated, and clinical outcomes were not analyzed, which limits the clinical relevance of the results. Lastly, all participants were inpatients who required surgery and presented with severe clinical symptoms, which may limit the representativeness of our findings for the general population of patients with DS.

Conclusions

The results of this study evaluating 100 patients with DS based on the CARDS classification revealed that different subtypes exhibited distinct radiological characteristics. Nearly half of the patients presented with clinical contraindications. Groups B and D were more suitable candidates for OLIF. In contrast, OLIF is not recommended for group A because of the high incidence of bony hyperplasia, severe central canal stenosis, and segmental stiffness. A comprehensive evaluation and individualized assessment are recommended for group C before considering OLIF as a surgical option.

Article information Continued