Bone cement distribution patterns influence cage subsidence after OLIF-AF-SEA in elderly patients with osteoporotic lumbar degeneration

Abstract

Purpose

To investigate the correlation between postoperative bone cement distribution patterns and the incidence of cage subsidence in elderly osteoporotic patients undergoing oblique lateral interbody fusion combined with anterolateral fixation and stress endplate augmentation (OLIF-AF-SEA).

Methods

This retrospective study analyzed 86 osteoporotic patients aged ≥ 60 years who underwent single-level L4-5 OLIF-AF-SEA with a minimum postoperative follow-up of 24 months. Postoperative X-rays were used to classify bone cement distribution patterns, including coronal plane distribution (three-zone method), endplate contact (binary classification), and presence of intervertebral leakage. Cage subsidence was defined as a reduction in disc height of ≥ 2 mm. Statistical analyses included chi-square tests and univariate binary logistic regression.

Results

Among the 86 patients, 16 (18.6%) experienced cage subsidence. Subsidence was significantly associated with unilateral or paramedian cement distribution in the coronal plane, lack of endplate contact, and the presence of intervertebral leakage (all p < 0.001). Logistic regression analysis identified non-bilateral distribution (L4: B = − 4.313; L5: B = − 5.273), absence of endplate contact (B = – 4.234), and intervertebral leakage (B = 3.446) as significant predictors of subsidence. No significant differences were found between the subsidence and non-subsidence groups in terms of age, sex, BMI, or bone quality parameters.

Conclusion

Postoperative bone cement distribution patterns may influence cage stability following OLIF-AF-SEA. Bilateral distribution, contact with both superior and inferior endplates, and absence of intervertebral leakage are closely associated with reduced risk of subsidence. These findings highlight the importance of standardized cement injection strategies and provide a theoretical foundation for improving surgical outcomes in elderly osteoporotic patients undergoing OLIF-AF-SEA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-025-06576-4.

Introduction

With changes in lifestyle and the acceleration of population aging, the incidence of lumbar degenerative diseases—such as disc herniation, spinal stenosis, and spondylolisthesis—has been increasing annually, significantly impacting patients’ quality of life. Over the past decade, oblique lateral interbody fusion (OLIF) has become one of the safe and effective minimally invasive surgical techniques for treating such conditions [1].

Global aging and the burden of degenerative diseases are advancing rapidly [2]. In China, the population aged 60 and above is projected to exceed 400 million by around 2035, accounting for more than 30% of the total population and marking the entry into a stage of deep aging [3]。Elderly patients with lumbar degenerative diseases often experience severe decline in organ and physiological function, commonly accompanied by osteoporosis, which increases surgical risks and complication rates, thereby affecting postoperative outcomes [4]。Building upon prior biomechanical studies and clinical validation, our team proposed the concept and technical approach of Stress Endplate Augmentation (SEA) based on OLIF combined with anterior screw-rod fixation (OLIF-AF) [5]. This technique uses polymethylmethacrylate (PMMA) bone cement to reinforce the stress-bearing endplates in contact with the interbody cage, aiming to prevent postoperative cage subsidence and poor fusion outcomes in elderly osteoporotic patients with reduced bony structural strength. Clinical validation has confirmed the safety and efficacy of this method as a novel treatment option for elderly patients with lumbar degeneration and osteoporosis.

OLIF-AF-SEA has demonstrated favorable stability and safety in clinical practice, significantly reducing the risk of cage subsidence in osteoporotic elderly patients [6]; [7]. However, some patients still present with radiographic signs of poor prognosis postoperatively, suggesting room for further optimization of the technique. Currently, there is a lack of systematic research, both domestically and internationally, on the effect of bone cement distribution characteristics on surgical outcomes in OLIF-AF-SEA. Drawing on previous studies on vertebral augmentation that examined the clinical implications of cement distribution, we have observed in practice that different cement distribution patterns may correlate with radiographic outcomes. Therefore, systematically investigating the relationship between bone cement distribution and postoperative outcomes in OLIF-AF-SEA may provide important theoretical and practical guidance for optimizing intraoperative injection strategies, improving postoperative stability, and reducing the risk of complications.

Methods

Data collection

A retrospective analysis was conducted on patients who underwent OLIF-AF-SEA at our institution between March 2021 and May 2023. Inclusion criteria were as follows: (1) patients who received single-level OLIF-AF-SEA at the L4–5 segment due to degenerative lumbar spine disease; (2) age over 60 years with a diagnosis of osteoporosis (T-score ≤ – 2.5); (3) availability of complete medical records and radiographic imaging data; and (4) a minimum clinical and radiological follow-up period of 24 months. Exclusion criteria included: (1) a prior history of lumbar spine surgery; (2) presence of severe chronic systemic illness or active spinal infection; (3) presence of osteoporotic vertebral fracture at the index or adjacent levels; and (4) incomplete records or loss to follow-up during the study period. All procedures and data collection protocols were approved by the Institutional Review Board of our hospital prior to the initiation of the study.

Surgical procedure

Under general anesthesia, the patient was placed in the right lateral decubitus position, and a skin incision approximately 6 cm in length was made in the lateral abdominal region. The abdominal muscles were separated along their fibers, and the retroperitoneal space was accessed by blunt dissection. The fascia over the ventral surface of the psoas major was then stripped, and the sympathetic chain and ureter were mobilized anteriorly. The belly of the psoas muscle was retracted posteriorly without the use of neuromonitoring. A 22-mm diameter tubular retractor was placed at the target intervertebral disc level. Discectomy was subsequently performed, with disruption of the contralateral annulus and preparation of the endplates. A bullet-shaped PEEK cage, filled with calcium phosphate cement containing rhBMP-2, was selected based on sequential trial implant testing and inserted obliquely into the intervertebral space and rotated laterally. To prevent graft migration or loss during insertion or SEA manipulation, two gelatin sponges were applied to the superior and inferior cage openings and secured with absorbable sutures, forming a sandwich-like containment structure (Supplementary Figure S1). Following cage insertion, stress endplate augmentation was performed. The entry point was selected at the midpoint of the cage in the sagittal plane, approximately 5–10 mm from the peripheral rim of the vertebral endplate. In general, the central axis of the screw was planned to be about 5 mm from the lowest point of the central concavity of the vertebral endplate, as determined by preoperative CT or X-ray imaging. A washer was hammered into the left side of the vertebral body, and a pilot hole was created along the axis of the cage and parallel to the endplate, with a depth approximately 5 mm from the contralateral cortex. The pilot hole was divided into two parts along its axis, and 1.5 mL of bone cement was injected into each part under C-arm fluoroscopic guidance. Subsequently, a lateral screw of appropriate length was inserted into each vertebral body, and a connecting rod of proper length was placed and locked. Finally, the abdominal muscles, subcutaneous tissue, and skin were closed sequentially. All patients were required to wear a customized lumbar brace for three months postoperatively. OLIF-AF-SEA was indicated for the degenerative lumbar disease in elderly osteoporotic patients requiring indirect decompression and fusion, with a favorable oblique lateral corridor and without a need for posterior decompression or multilevel instrumentation. SEA was employed to enhance endplate support and mitigate the risk of cage subsidence in osteoporotic endplates. All procedures were performed by the same experienced spine surgery team to ensure procedural consistency and minimize inter-surgeon variability.

Radiological evaluation

The spatial distribution of bone cement postoperatively was assessed using anteroposterior and lateral lumbar X-rays obtained on the day of surgery. Referring to established methods used in vertebroplasty studies evaluating cement distribution, a “three-zone classification” system was adopted [8]; [9]. On the anteroposterior X-ray, the lateral dispersion of bone cement within the vertebral body was categorized into three types: Type 1 (unilateral type): cement localized predominantly on one side of the vertebral body; Type 2 (paramedian type): cement extending beyond the midline but not fully covering the contralateral side; Type 3 (bilateral type): cement fully covering both sides of the vertebral body (Fig. 1).

Fig. 1.

Classification of lateral cement distribution on postoperative anteroposterior radiographs. Type 1: unilateral distribution; Type 2: paramedian distribution; Type 3: bilateral distribution

Additionally, to further evaluate the relationship between bone cement and endplates as well as its potential risks, a supplementary “binary classification” was applied (Fig. 2). On the lateral X-ray: Cement was classified as Type 0 (non-contact) or Type 1 (contact) based on whether it contacted the superior and/or inferior endplates [10]; Cement was defined as Type 1 (leakage) if it breached the endplate and entered the intervertebral space, or Type 0 (non-leakage) otherwise [11]. Given that bone mineral density is an independent factor influencing cement distribution [12], the study also included the mean Hounsfield unit (HU) value and vertebral bone quality (VBQ) score of the surgical segment as bone quality reference indicators, using measurement methods consistent with our previous studies [13]; [14].

Fig. 2.

Classification of cement–endplate relationship and intervertebral leakage on postoperative lateral radiographs. Left: endplate contact classification (Type 0 = no contact, Type 1 = contact); Right: intervertebral leakage classification (Type 0 = no leakage, Type 1 = leakage across the endplate)

Fig. 3.

A 68-year-old male patient who underwent L4–5 OLIF-AF-SEA surgery. The cement distribution pattern at L5 was classified as type 1 (unilateral). Cage subsidence was observed at the final 2-year follow-up. (A–B) Postoperative day 1 and 2-year follow-up lumbar spine anteroposterior radiographs. (C–D) Postoperative day 1 and 2-year follow-up lateral radiographs. (E–F) Postoperative day 1 and 2-year follow-up sagittal reconstructed CT images

Fig. 4.

A 64-year-old female patient underwent L4–5 OLIF-AF-SEA. Postoperative imaging revealed lack of endplate cement contact at the L5 level. Cage subsidence was observed at the final 2-year follow-up. A–B Standing anteroposterior radiographs of the lumbar spine obtained on postoperative day 1 and at 2 years, respectively. C–D Lateral radiographs of the lumbar spine on postoperative day 1 and at 2 years. E–F Sagittal reconstructed CT scans on postoperative day 1 and at 2 years, demonstrating progressive cage subsidence

Fig. 5.

A 62-year-old female patient underwent L4–5 OLIF-AF-SEA. Intervertebral cement leakage was identified on early postoperative imaging. Cage subsidence was observed at the final 2-year follow-up. A–B Standing anteroposterior radiographs of the lumbar spine obtained on postoperative day 1 and at 2 years, respectively. C–D Lateral lumbar radiographs on postoperative day 1 and at 2 years. E–F Sagittal reconstructed CT scans obtained on postoperative day 1 and at 2 years, showing progressive cage migration into the vertebral body

Fig. 6.

A 68-year-old female patient underwent L4–5 OLIF-AF-SEA. Postoperative imaging showed a Type 3 cement distribution with contact to both superior and inferior endplates and no intervertebral leakage. At the 2-year follow-up, no cage subsidence was observed, and interbody fusion was achieved. A–B Standing anteroposterior lumbar radiographs obtained on postoperative day 1 and at 2 years. C–D Lateral lumbar radiographs on postoperative day 1 and at 2 years. E–F Sagittal reconstructed CT scans obtained on postoperative day 1 and at 2 years, demonstrating stable cage position without migration

Cage subsidence was defined as radiographic evidence of the cage sinking into the vertebral body through the endplate. Disc height (DH) at the L4/5 level was measured on sagittal CT images from the immediate postoperative and follow-up scans. A decrease of more than 2 mm in DH at follow-up was considered indicative of subsidence [15]. All measurements were independently performed by two physicians, and the average value was used for final analysis.

Statistical analysis

The primary outcome variable in this study was cage subsidence (CS), which treated as binary categorical outcome. Explanatory variables included categorical indicators of bone cement distribution, such as lateral dispersion type, endplate contact status, and presence of intradiscal cement leakage. To assess associations between explanatory variables and outcome, Chi-square analysis or univariate binary logistic regression analysis was used. Statistical significance was defined as a two-sided p-value less than 0.05. All statistical analyses were conducted using IBM SPSS Statistics software (version 29.0; IBM Corp, Armonk, NY, USA).

Results

Patients

Between March 2021 and May 2023, a total of 201 consecutive patients who underwent OLIF-AF-SEA for lumbar degenerative disease at our center were retrospectively reviewed. After applying the inclusion and exclusion criteria, 86 patients were ultimately enrolled in the study. The primary reasons for exclusion included history of prior lumbar surgery, incomplete imaging or clinical records, coexisting systemic infections or severe chronic disease, and failure to complete the minimum 24-month follow-up.

Of the 86 patients included, 16 (18.6%) experienced radiographically confirmed CS, while 70 (81.4%) did not. Baseline demographic and clinical characteristics were comparable between the CS and non-CS groups. No statistically significant differences were observed in age (66.50 ± 1.90 vs. 65.54 ± 4.03 years, p = 0.358), sex distribution (female: 81% vs. 86%, p = 0.653), BMI classification (p = 0.690), BMD (p = 0.890), or comorbidities including diabetes, hypertension, hyperlipidemia, and chronic pulmonary or cardiac conditions (all p > 0.2). Notably, bone quality parameters such as HU values and VBQ scores showed no significant intergroup difference (HU: p = 0.628; VBQ: p = 0.139), suggesting comparable baseline bone status (Table 1).

Table 1.

Patient demographic and clinical characteristics

Characteristic	Non-CS (n = 70)	CS (n = 16)	p-Value
Demographic
Age, y	65.54 ± 4.031	66.50 ± 1.897	0.358
Sex (Female)	60(86)	13(81)	0.653
BMI, kg/ Obese Nonobese BMD (T-value) Imaging-derived bone quality HU VBQ	37(53) 33(47) -3.2 ± 0.25 89.12 ± 16.47 4.17 ± 1.01	6(38) 10(63) -3.2 ± 0.33 86.84 ± 18.97 4.60 ± 1.14	0.690 0.890 0.628 0.139
Clinical history
Diabetes Hypertension ACS COPD Hyperlipidemia	10(14) 32(46) 21(30) 8(11) 8(11)	2(13) 5(31) 3(19) 2(13) 3(19)	0.852 0.292 0.365 0.904 0.429
Operative Procedure time, hour	1.68 ± 0.31	1.81 ± 0.24	0.122

Patient demographic and clinical characteristics

CS, Cage Subsidence; HU, Hounsfield Unit; VBQ, Vertebral Bone Quality; ACS, Acute Coronary Syndrome; COPD, Chronic Obstructive Pulmonary Disease; BMI, Body Mass Index; BMD, Bone Mineral Density

Radiological results

Radiological evaluation revealed a significant association between postoperative cage subsidence and cement distribution patterns. As shown in Table 2, the laterality pattern of cement at both L4 and L5 levels, the presence of endplate cement contact, and intervertebral cement leakage were all significantly correlated with cage subsidence (all P < 0.001). Specifically, subsidence rates varied substantially across different lateral distribution categories: for L4 cement laterality, 85.7% in Type 1 (unilateral), 61.5% in Type 2 (paramedian), and only 3.0% in Type 3 (bilateral); for L5, the respective rates were 80.0%, 73.3%, and 1.5%. Subsidence was more frequent in patients without cement-endplate contact (88.9%) compared to those with contact (10.4%), and more common among those with intervertebral cement leakage (83.3%) than those without (13.8%).

Table 2.

Pearson Chi-square analysis of the association between cement distribution parameters and cage subsidence

Variable	Category Levels	Cage Subsidence (%)	χ²	df	p-value
L4 Cement Laterality Pattern	1 / 2 / 3	85.7/ 61.5 / 3.0	47.214	2	< 0.001*
L5 Cement Laterality Pattern	1 / 2 / 3	80.0/ 73.3 / 1.5	54.843	2	< 0.001*
Endplate Cement Contact	0 / 1	91.7 / 6.8	49.160	1	< 0.001*
Intervertebral Cement Leakage	0 / 1	13.8/83.3	17.846	1	< 0.001*

*, statistical significance between the two groups (P < 0.05)

To further clarify the relative impact of each factor, we performed binary logistic regression using dichotomized cement laterality variables (Type 3 [bilateral] vs. Types 1 and 2), as summarized in Table 3. The results demonstrated that non-bilateral cement distribution at both L4 and L5 was associated with a significantly increased risk of subsidence (L4: B = − 4.313, P < 0.001; L5: B = − 5.273, P < 0.001). Similarly, absence of cement-endplate contact (B = -4.234, P < 0.001) and presence of intervertebral cement leakage (B = 3.446, P = 0.003) were both associated with a significantly increased risk of subsidence.

Table 3.

Binary logistic regression analysis of the relationship between cement distribution parameters and cage subsidence

Variable	Non-CS (n = 70)	CS (n = 16)	B	p-Value
L4 cement Laterality Pattern Non-Bilateral Bilateral L5 cement Laterality Pattern Non-Bilateral Bilateral Endplate Cement Contact Non-Contact Contact Intervertebral Cement Leakage Non-Leakage Leakage	6(7) 64(91) 5(7) 65(93) 1(1) 69(99) 69(99) 1(1)	14(88) 2(13) 15(94) 1(6) 8(50) 8(50) 11(69) 5(31)	-4.313 – 5.273 – 5.023 3.446	< 0.001* < 0.001* < 0.001* 0.003*

B, Beta coefficients

*, statistical significance between the two groups (P < 0.05)

These results underscore the clinical relevance of postoperative cement distribution patterns in relation to cage subsidence. Representative imaging cases with corresponding radiographic outcomes are provided below to illustrate these associations.

Disccusion

In this retrospective study of patients undergoing OLIF-AF-SEA, we investigated the relationship between cement distribution patterns and postoperative cage subsidence. Our findings revealed that certain radiographic features—including unilateral or near-midline cement spread in the coronal plane, absence of cement–endplate contact, and intervertebral leakage—were significantly associated with higher rates of subsidence at the two-year follow-up. Notably, bilateral cement spread corresponded with the lowest incidence of subsidence and demonstrated a statistically significant association with reduced risk in subsequent binary logistic regression analysis. To objectively evaluate cement distribution, we adapted a trichotomous classification method widely used in percutaneous vertebral augmentation literature and applied it to both anteroposterior and lateral lumbar radiographs. Additional binary classification systems were used to assess cement–endplate interface and evidence of cement leakage across the disc space. These combined frameworks enabled reproducible characterization of cement patterns and identification of distribution types potentially detrimental to implant stability.

To the best of our knowledge, this is the first study to systematically investigate the relationship between bone cement distribution and clinical outcomes in patients undergoing OLIF-AF-SEA. Although previous studies have explored the use of bone cement augmentation in vertebral compression fractures, particularly in percutaneous vertebroplasty and balloon kyphoplasty, its role in lumbar interbody fusion surgery has been rarely reported. The bone cement distribution classification framework used in this study was based on the “three-zone” method proposed in prior PVP/PKP literature, categorizing cement dispersion patterns into three types in the axial and coronal planes. Clinical studies have shown that the lateral distribution type of bone cement within the vertebral body can affect vertebral strength, stiffness, and the risk of recompression [16, 17]. In terms of basic research, most scholars have found through finite element analysis that bilateral distribution of bone cement results in better biomechanical performance of the vertebrae [18, 19]. Lin et al. [20] in a study of 93 PVP patients, found that the contact between bone cement and the superior and inferior endplates was closely related to postoperative vertebral height loss, with poor endplate contact being associated with a higher likelihood of recompression. Zhang et al. [21] in a review of 172 elderly patients with vertebral fractures, noted that when the cement contacted both endplates, the incidence of recompression was significantly lower than in patterns where cement was confined to the central vertebral body; logistic regression analysis also confirmed this distribution pattern as a protective factor against recompression. Specifically, radiographic features such as bilateral cement dispersion and a larger cement–endplate contact area are considered beneficial for improving biomechanical stability. Gu et al. [22] achieved effective control of cement leakage and optimized its intravertebral dispersion through technical modifications, which reduced the risk of cage subsidence in postoperative patients. The findings of this study further support this view, revealing that lack of endplate contact and intervertebral cement leakage were both significantly associated with poorer radiographic outcomes. This also provides a theoretical foundation for future biomechanical validation and surgical technique optimization for this procedure.

At present, mechanistic studies on OLIF combined with bone cement augmentation remain limited. Bereczki et al. pointed out through finite element analysis that in osteoporotic models, an appropriate volume of bone cement injection (approximately 3–4 cm³ per screw) can effectively enhance the initial stability of the fusion segment, with the augmentation effect approaching that of the traditional bilateral pedicle screw fixation system to some extent [23]. Clinical observations indicate that cage subsidence often occurs during the early postoperative stage and is closely related to insufficient endplate load-bearing capacity and inadequate bony structural support [24, 25]. In OLIF-AF-SEA, the injection of bone cement is intended to enhance the local load-bearing capacity of the endplate region; however, if the cement distribution is asymmetric or lacks sufficient contact with the endplate, it becomes difficult to form an effective supporting structure at the cage–vertebral interface. When the endplate is subjected to prolonged focal high-stress states, stress concentration is likely to occur, accelerating bony fatigue and microfractures, ultimately triggering subsidence [26]. On the other hand, if bone cement penetrates the endplate and enters the intervertebral space during surgery, it not only compromises the integrity of the supporting endplate but may also disrupt the intended stability of the intervertebral environment [27]. Clinical experience suggests that such leakage typically occurs in the context of osteoporosis, abnormal endplate morphology, or poor injection control [28, 29]. Additionally, in some cases, subsidence may not occur immediately after surgery but may develop gradually during the increasing load-bearing phase, as repetitive stress at the cage–cement–endplate contact area may lead to progressive interface collapse or endplate failure. Therefore, the intraoperative distribution pattern of bone cement not only affects immediate stability but also plays a decisive role in the long-term load transfer pathway and endplate stress distribution pattern.

In this study, a PEEK cage was chosen due to its elastic modulus being more similar to that of bone [30], which may reduce stress shielding and allow for a more physiological load transfer across osteoporotic endplates. Its radiolucency also makes it easier to see how the graft is incorporated and how the bones are healing on postoperative imaging [31]. However, titanium cages have their own benefits, such as better immediate mechanical strength, more resistance to deformation, and better osseointegration potential because of their rough surface and material properties [32]. These biomechanical characteristics may confer increased stability in specific surgical scenarios. Even so, titanium cages can also cause higher stress concentrations at the cage–endplate interface, especially in osteoporotic bone. This may predispose to subsidence in elderly patients. Considering that SEA introduces bone cement into the vertebral body, understanding the interplay among cage material, cement distribution, and endplate support is essential. In our technique, bone cement augmentation is primarily used to reinforce the stress-bearing region of the endplate directly beneath and above the cage, thereby creating a stronger supporting platform for the interbody construct. Furthermore, while PEEK provides favorable load-sharing characteristics, cement augmentation compensates for its lower intrinsic stiffness by improving the integrity of the cage–endplate interface and facilitating progressive osseous integration.

The findings of this study offer practical guidance for intraoperative procedures in OLIF-AF-SEA. As a critical component in ensuring postoperative stability for elderly patients with osteoporosis, bone cement augmentation during surgery should follow a standardized injection protocol and distribution pattern. Our team currently adopts a three-phase cement injection strategy, which emphasizes sequential delivery to achieve bilateral coverage of the vertebral body. This approach facilitates the formation of a broad and evenly distributed support platform beneath the cage contact area, thereby enhancing load-bearing capacity and improving postoperative stability. In addition to patient-related factors such as bone quality and endplate morphology, subtle variations in intraoperative injection techniques (e.g., injection resistance, fluoroscopic feedback, and endplate conformity) may also influence cement distribution. Our findings suggest that bilateral distribution and firm contact with the endplates can help surgeons refine their injection strategy. Futhermore, precise control of injection depth during the procedure—ensuring that the cement fully conforms to the surfaces of the superior and inferior endplates rather than accumulating in the central region—is a key step in optimizing the quality of the cage–endplate interface. Preventing cement leakage into the intervertebral space also warrants close attention; during the procedure, fluoroscopic monitoring of cement flow should be performed to avoid breaching the endplate and disrupting the native intervertebral structural integrity. Previous studies have also identified a strong correlation between the risk of cement leakage and bone mineral density [33], To minimize this potential confounding factor, our study design matched vertebral bone quality parameters (including HU values and VBQ scores) between the subsidence and non-subsidence groups, thereby ensuring that the analysis more accurately reflects the impact of cement distribution itself on postoperative subsidence. As OLIF techniques become more widely adopted and the proportion of osteoporotic patients undergoing surgery continues to rise, the standardization of bone cement augmentation strategies will play a crucial role in improving surgical success rates and long-term radiographic outcomes. Based on the findings of this study, future advancements may include the development of intraoperative image-guided systems or specialized cement injection devices to enhance the controllability and consistency of cement distribution, ultimately offering safer and more predictable treatment options for elderly patients.

This study systematically evaluated the relationship between bone cement distribution and cage subsidence following OLIF-AF-SEA surgery and proposed a practically applicable classification method with potential clinical value. However, several limitations remain. As a single-center retrospective study, case inclusion was limited by the completeness of previous medical records, which may introduce selection bias. Although all imaging assessments were independently conducted by two spine surgeons to enhance consistency, the retrospective nature of the study makes it difficult to fully control for confounding factors. Future multicenter prospective cohort studies are needed to validate these findings and improve their generalizability. Additionally, the sample size was relatively limited, with imbalanced subgroup sizes in certain bone cement distribution types, suggesting that larger cohorts are necessary to improve statistical precision. Furthermore, while the proposed classification system was informed by prior vertebroplasty literature, it lacks biomechanical or histological evidence, preventing direct conclusions about its effects on load transmission or interface stress. Moreover, the conclusions of this study may not be directly applicable to deformity correction procedures or long-level instrumentation, where biomechanical environments and endplate loading conditions differ substantially from single-level OLIF-AF-SEA. Future studies incorporating finite element modeling, cadaveric experiments, or animal models are warranted to further explore the impact of different cement distribution patterns on cage subsidence. In summary, this study provides a novel radiographic classification framework for assessing subsidence risk after OLIF-AF-SEA, but higher-quality and multidimensional research is required to support and standardize its clinical application.

Conclusion

This study systematically evaluated the relationship between bone cement distribution patterns and cage subsidence in elderly patients with osteoporotic lumbar degeneration undergoing OLIF-AF-SEA. The results demonstrated that bilateral cement dispersion, adequate endplate contact, and absence of intervertebral leakage may be associated with a reduced incidence of postoperative cage subsidence. These findings underscore the clinical importance of standardizing intraoperative bone cement injection strategies. Optimizing cement distribution may enhance postoperative stability and improve surgical outcomes for the growing population of elderly osteoporotic patients with lumbar degenerative diseases.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

X.Hu: Study design, data collection, data analysis, manuscript drafting, final approval. T. Xie: Surgical data interpretation, manuscript revision, final approval. X.Pu: Imaging data extraction, final approval. C.Luo: Clinical data management, final approval. X.Peng: Surgical record review, final approval. X.Wang: Methodology supervision, manuscript review, final approval. J. Zeng: Study conception, overall supervision, final approval.*Xiao Hu and Tianhang Xie contributed equally and should be considered co-first authors of this study.*Corresponding author E-mail: tomzeng5@126.com (J. Zeng) and wxdfrank@wchscu.cn (X. Wang).

Funding

This work was supported by the National Natural Science Foundation of China (82372439); National Key Research and Development Program of China (2023YFC3604404, 2023YFC3604405); National Key Research and Development Program of China (2024YFC3607003).

Data availability

No datasets were generated or analysed during the current study.

Declerations

Ethics approval and consent to participate

Approval for the current study protocol was obtained from the ethics committees of West China Hospital (2023–2285). All data collection and analysis conducted in this study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.