Risk factors for secondary vertebral compression fracture after percutaneous vertebral augmentation: a single-centre retrospective study

Abstract

Objective

To determine the incidence of secondary vertebral compression fracture (SVCF) after percutaneous vertebral augmentation (PVA) and its correlative risk factors, and to provide theoretical evidence for clinical practice.

Methods

A retrospective analysis of 288 cases of PVA completed in our hospital from June 2020 to June 2023 was performed, and the patients were divided into the non-secondary vertebral compression fracture group (N-SVCF group) and the secondary vertebral compression fracture group (SVCF group) according to whether SVCF occurred during the postoperative follow-up review. Gender, age, body mass index (BMI), T value of bone mineral density (BMD-T), underlying diseases (hypertension, diabetes mellitus, coronary heart disease, chronic obstructive pulmonary disease), intravertebral vacuum cleft (IVC), amount of bone cement injected, classification of cement diffusion, anterior vertebral recovery ratio, local Cobb angle correction rate, leakage of bone cement into the intervertebral space, and fat infiltration rate (FIR) of paraspinal muscles were collected from the patients. The incidence and risk factors of SVCF after PVA were evaluated using univariate and multivariate logistic regression analysis, and the predictive value of the independent risk factors was evaluated using receiver operating characteristic curve (ROC) to determine the cut-off points at which they were meaningful for the development of SVCF.

Results

In our study, the incidence of SVCF was 14.60% (42/288) in 288 patients who underwent PVA. Univariate analysis showed that age, BMI, fat infiltration rate of paraspinal muscles, cement leakage into the intervertebral space, unilateral/bilateral pedicle puncture approach and presence of IVC were statistically different between N-SVCF and SVCF (P < 0.05). Multifactorial regression analysis and ROC regression analysis revealed that the fat infiltration rate of the psoas major and erector spinae muscles, cement leakage into the intervertebral space, and IVC (P < 0.05) were risk factors for the incident of SVCF after PVA (P < 0.05). Psoas major (FIR) more than 5.490% and erector spinae (FIR) more than 52.413% had a high possibility of the occurrence of SVCF after PVA.

Conclusion

In this study, logistic regression combined with ROC curve analysis indicated that FIR of psoas major and erector spinae, cement leakage in the intervertebral space, and IVC were risk factors for the occurrence of SVCF after PVA. Psoas major (FIR) more than 5.490% and erector spinae (FIR) more than 52.413% had a high possibility of the occurrence of SVCF after PVA.

Introduction

Osteoporosis (OP) is a disease characterized by decreased bone quality strength, which is mainly associated with bone tissue loss and destruction of bone microstructure, and patients with OP are highly susceptible to fracture and their quality of life is greatly reduced [1]. In recent years, the health problems caused by global population aging have become increasingly prominent, and osteoporotic vertebral compression fracture (OVCF) has become more and more common among the elderly, with approximately 1.4 million new vertebral fractures occurring each year worldwide [2]. Some studies have shown that as many as a quarter of people over the age of 50 will experience at least one vertebral fracture in their lifetime [3]. OVCF reduces patients’ mobility and their quality of life, causes significant pain, and raises the risk of death. Conservative treatment of OVCF includes bed rest, use of analgesic medications, physiotherapy and immobilization braces [4]. Although most fractures heal within a few months, most patients experience persistent pain and require hospitalization or long-term care [5]. In addition, conservative treatment is not well tolerated in the elderly population. The simultaneous use of analgesic medications and prolonged external fixation has also been reported to increase the risk of adverse effects, such as falls and constipation [6]. Percutaneous vertebral augmentation (PVA) has the characteristics of restoring local biomechanical stability and reducing pain compared with conservative treatments such as prolonged bed rest, which can significantly improve the quality of life of patients and reduce the occurrence of complications [7]. However, the occurrence of postoperative secondary vertebral compression fracture (SVCF) has gradually attracted the attention of doctors, and more and more scholars have investigated the risk factors of SVCF after PVA. The purpose of this study is to evaluate the incidence and risk factors of SVCF after PVA by retrospective analysis the data of patients who received PVA treatment in our hospital.

Information and methods

General information

Two hundred and eighty-eight patients who underwent PVA in our hospital from June 2020 to June 2023 were selected for this study, and were divided into the non-secondary vertebral compression fracture (N-SVCF group) and the secondary vertebral compression fracture group (SVCF group) according to whether SVCF ( It includes all new vertebral fractures of non-operated vertebrae, which in other words it includes adjacent and non-adjacent vertebrae) occurred during the postoperative follow-up review.

Inclusion criteria: (1) T value of bone mineral density (BMD-T) <-2.5; (2) the patient’s first vertebral fracture was a single-segment thoracic/lumbar vertebral fracture; (3) the patient’s first surgical treatment with PVA; (4) the follow-up time was more than 12 months; (5) informed consent and signing of the informed consent form; (6) complete imaging and medical records; (7) the surgical instruments and bone cement used in the operation were provided by the same manufacturer and brand.

Exclusion Criteria: (1) Vertebral compression fractures resulting from high energy, such as car accidents, falls from height, etc.; (2) Patients with mental or neurological disorders or malignant tumours and tuberculosis; (3) Patients with severe coagulation disorders or metabolic disorders; (4) Patients with preoperative known allergy to bone cement or contrast mediums; and (5) Patients with pulmonary and cardiac disorders, or with hepatic & renal disorders, who are unable to tolerate the surgery.

Dropout criteria: (1) cases of systemic disease or death during follow-up; (2) those who have missed a visit or have incomplete imaging and medical records.

Evaluation metrics

Gender, age, body mass index (BMI), BMD-T, underlying diseases (hypertension, diabetes mellitus, coronary heart disease, chronic obstructive pulmonary disease), intravertebral vacuum cleft(IVC), amount of bone cement injected, classification of cement diffusion, anterior vertebral recovery ratio(AVRR), Cobb angle correction rate, leakage of bone cement into the intervertebral space, and fat infiltration rate(FIR) of paraspinal muscles(MF, multifidus; ES, erector spinae; PM, psoas major) were collected from the patients.

Lumbar spine BMD -T

BMD-T values of the lumbar spine (L1-L4) were measured by a professional radiologist via DXA (OsteoSys-DEXXUMT, Korea).

Extent of fat infiltration

Images were obtained using a 1.5T MRI scanner (SIEMENS-MAGNETOM ESSENZA Galaxy, Germany) and a standard spinal array coil. The imaging protocol consisted of acquiring axial and sagittal T2-weighted images. The parameters of the T2-weighted images were as follows: slice thickness of 4 mm, slice interval of 1 mm. The axial position was located in the middle of the sagittal plane, and the sagittal T2-weighted image of the fracture site was acquired in the sagittal plane. Axial T2-weighted images of the lumbar spine were obtained at the L4/5 disc level [8]. Image J (Image J Version 1.5E, National Institutes of Health, Bethesda, Maryland, USA) software was used to manipulate the MRI images and to designate the region to be measured as the region of interest (ROI). Paraspinal muscle degeneration is characterized by muscle atrophy and fat infiltration, the extent of which can be quantified by cross-sectional area (CSA) and FIR [9]. The total CSA of the MF, ES and PM was measured on the left and right sides separately by outlining the fascial boundaries of each muscle. The intramuscular fat CSA was then separated from the original segmentation along the fascial boundaries using a thresholding method to obtain a specific range of signal intensities for the adipose and muscular tissues (Fig. 1). Where paravertebral muscle FIR = ( paraspinal intramuscular fat CSA/paraspinal muscle CSA) × 100% [10].

Fig. 1.

(A) Cross-sectional area of paraspinal muscles and intermuscular fat area measurements (MF multifidus; ES, erector spinae; PM, psoas major); (B) ImageJ’s automatic threshold setting distinguishes fat from muscle. The range of the yellow curve is the right MF range, and the red part within the yellow curve is the fat area of the right MF

Cement diffusion classification

According to the diffusion mode of the bone cement in the postoperative frontal and lateral X-ray films, it was classified into four types (Fig. 2): (1) Diffuse type: the bone cement was uniformly distributed in more than 70% of the vertebral body length (BL), vertebral body width (BW), and vertebral body height (BH) in the anterior/posterior and lateral X-rays; (2) Block type: The cement was uniformly distributed over 70% of BL and BH and between 40 and 70% of BW; (3) Double-band type: the cement was divided into two regions on anterior/posterior and lateral radiographs, with each strip being less than 40% of the BH; (4) Single band type: the cement was concentrated in a single region in both anterior/posterior and lateral radiographs, which was less than 40% of the BH [11].

Fig. 2.

Cement diffusion classification. A: Diffuse type; B: Block type; C: Double band; D: Single band; CFR: Cement filling rate; BW: Body width; BH: body height; BL: body length; CW: cement width; CH: cement height; CL: cement length. A-P: anterior/posterior position

AVRR and postoperative local Cobb angle correction rate

The AVRR was calculated by dividing the difference between the anterior heights of the injured vertebral bodies in the postoperative and preoperative periods (d-a) by the average value of the anterior heights of the adjacent upper and lower vertebral bodies ((b + c)/2) (Fig. 3) [12].

Fig. 3.

Diagram of the anterior vertebral recovery ratio of the injured vertebrae. a: Preoperative height of the anterior margin of the injured vertebrae. b: Preoperative height of the anterior margin of the upper vertebrae of the injured vertebrae. c: Preoperative height of the anterior margin of the lower vertebrae of the injured vertebrae. d: Postoperative height of the anterior margin of the injured vertebrae

AVRR = (postoperative vertebral anterior height - preoperative vertebral anterior height) / average of the anterior heights of the adjacent upper and lower vertebrae × 100% (Fig. 3).

The local Cobb angle was measured on lateral radiographs. Vertical lines were made by drawing horizontal lines at the superior margin of the adjacent upper vertebra of the injured vertebra and the inferior margin of the adjacent lower vertebra of the injured vertebra. The intersection angle of these two vertical lines is the Cobb angle (Fig. 4) [13].

Fig. 4.

Schematic diagram of local Cobb angle changes from preoperative to postoperative. a: Preoperative local Cobb angle. b: Postoperative local Cobb angle

Local Cobb angle recovery rate = (preoperative local Cobb angle - postoperative local Cobb angle)/preoperative local Cobb angle × 100% (Fig. 4).

Statistical processing

R studio (version 4.2.2) software was applied for data analysis. Comparison of count data was performed using the X² test, measurement data were performed using x ± s, and comparisons between groups were performed using the independent samples t-test. The independent risk factors for SVCF in postoperative PVA patients were analysed using univariate and multivariate logistic regression, and the predictive value of the independent risk factors was analysed using the receiver operating characteristic curve (ROC), the area under curve (AUC), and the predictive value of the independent risk factors were analysed using the ROC curve. The predictive value of the independent risk factors was analysed, and the sensitivity, specificity and accuracy were analysed by using the Youden’s index to determine the cut-off point of FIR of the paravertebral muscles. P < 0.05 is considered statistically significant.

Results

General information

According to the inclusion and exclusion criteria, a total of 288 patients were included in the study, including 56 males and 232 females; age 51–92 years (71.25 ± 8.25); BMI 13.31–35.84 kg/m² (22.94 ± 3.24); 42 of them presented with SVCF, and the incidence of SVCF was 14.60% (42/288). The 42 SVCFs included T6, T8, T10, L3, L4, L5, T8&T11, T8&L4, T10&L3, T11&T12, L1&L3, L2&L3 fracture were all 1 case, T9 and L1&L2 were both 2 cases, there were 4 cases in L2, 5 cases in each of T11 and L1, and 12 cases in T12 (Fig. 5).

Fig. 5.

Secondary vertebral compression fracture locations distribution

Univariate analysis

There were no statistically significant differences in gender, BMD-T, cement injection volume, cement diffusion classification, AVRR, local Cobb angle correction rate, diabetes mellitus, coronary heart disease, chronic obstructive pulmonary disease, and hypertension in the SVCF group compared with the N-SVCF group (P > 0.05). Age, BMI, FIR of paraspinal muscles (PM, ES, MF), cement leakage, unilateral/bilateral pedicle puncture access and IVC were statistically different (P < 0.05), as shown in Table 1.

Table 1.

Univariate Analysis of Risk factors for SVCF

Variable		SVCF	N-SVCF	x2/t	P value
Age		75.86 ± 7.842	70.47 ± 8.071	-4.016	< 0.001**
Gender	Male	8	48	0.005	0.944
	Female	34	198
BMI		21.994 ± 3.284	23.108 ± 3.217	2.923	0.004*
BMD-T		-3.020 ± 1.083	-3.173 ± 0.813	-1.072	0.285
PM(FIR)		10.319 ± 6.440	4.786 ± 4.452	-5.354	< 0.001**
ES(FIR)		61.176 ± 17.479	40.102 ± 17.457	-7.118	< 0.001**
MF(FIR)		67.293 ± 19.044	46.988 ± 17.253	-6.941	< 0.001**
Cement injection volume		4.893 ± 1.031	5.294 ± 1.271	1.940	0.053
Anterior Vertebral Restoration Rate		4.079 ± 2.048	3.664 ± 2.578	-0.986	0.325
Local Cobb angle correction rate		33.143 ± 18.402	28.522 ± 19.229	-1.448	0.149
Classification of bone cement diffusion	Diffuse type	30	198	3.145	0.370
	Block type	11	38
	Double-band type	0	1
	Single band type	1	9
DM	No	37	196	1.646	0.199
	Yes	5	50
CHD	No	38	224	0.015	0.903
	Yes	4	22
COPD	No	40	243	2.639	0.104
	Yes	2	3
Hypertension	No	23	138	0.026	0.872
	Yes	19	108
Puncture approach	Unilateral	6	13	4.717	0.030*
	Bilateral	36	233
Bone cement leakage	No	17	219	60.315	< 0.001**
	Yes	25	25
IVC	No	33	241	36.369	< 0.001**
	Yes	9	3

SVCF, secondary vertebral compression fracture; N-SVCF, non-secondary vertebral compression fracture; BMI, body mass index; BMD-T, T value of bone mineral density; PM, psoas major; ES, erector spinae; MF, multifidus; FIR, fat infiltration rate; DM, diabetes mellitus; CHD, coronary heart disease; COPD, chronic obstructive pulmonary disease; IVC, intravertebral vacuum cleft; Compared with the other groups, **P < 0.001, *P < 0.05

Multivariate analysis

Factors that were statistically differenced in the univariate analysis: age, BMI, FIR of paravertebral muscles (PM, ES, MF), cement leakage, unilateral and bilateral pedicle puncture access and IVC were included in the multivariate logistic regression analysis (see Table 2 for details). The multivariate logistic regression analysis showed that PM (FIR), ES (FIR), cement leakage and IVC were risk factors for the incidence of SVCF.

Table 2.

Multivariate logistic regression analysis of risk factors for SVCF

Variable	B	SE	Z	P	OR	95%CI
Age	0.024	0.033	0.536	0.464	1.024	0.960–1.093
BMI	-0.053	0.070	0.574	0.449	0.948	0.826–1.088
PM(FIR)	0.102	0.041	6.307	0.012*	1.108	1.023–1.199
ES(FIR)	0.035	0.012	8.433	0.004*	1.035	1.011–1.060
MF(FIR)	0.022	0.014	2.304	0.129	1.022	0.994–1.052
Puncture approach	1.010	0.793	1.625	0.202	0.364	0.077–1.721
Bone cement leakage	-2.202	0.490	20.168	< 0.001**	9.041	3.459–23.635
IVC	-3.159	0.915	11.917	< 0.001**	23.542	3.917–141.490

SVCF, secondary vertebral compression fracture; OR, odds ratio; CI, confidence interval; BMI, body mass index; PM, psoas major; ES, erector spinae; MF, multifidus; FIR, fat infiltration rate; IVC, intravertebral vacuum cleft; Compared with the other groups, **P < 0.001, *P < 0.05

ROC curve

Multivariate logistic regression analysis demonstrated that PM(FIR), ES(FIR), cement leakage and IVC were all risk factors for the development of SVCF after PVA. To clarify the cut-off point at which the above risk factors were meaningful for SVCF, a ROC curve was plotted for 288 patients (Fig. 6). The AUC for PM(FIR) was 0.788, the Youden’s index was 0.482, corresponding to a cut-off point of 5.490%, sensitivity 76.2% and specificity 72.0%. ES(FIR) had an AUC of 0.817, a Youden’s index of 0.511, corresponding to a cut-off point of 52.413%, a sensitivity of 71.4%, and a specificity of 79.7%. Bone Cement Leakage had an AUC of 0.746, a Youden’s index of 0.493, a sensitivity of 59.2%, and a specificity of 98.8%. IVC had an AUC of 0.601, with a Youden’s index of 0.202, a sensitivity of 21.4%, and a specificity of 89.8%. The above influences combined together predicted an AUC of 0.895 for SVCF (Table 3).

Fig. 6.

ROC curve for predicting the occurrence of SVCF after PVA. PM, psoas major; ES, erector spinae; IVC, intravertebral vacuum cleft

Table 3.

Cutoff point、Youden’s index and AUC

Variable	Cutoff point	Sensitivity(%)	Specificity(%)	Youden’s index	AUC	95%CI	P
PM(FIR)	5.490	76.2	72.0	0.482	0.788	0.712–0.863	< 0.001**
ES(FIR)	52.413	71.4	79.7	0.511	0.817	0.749–0.884	< 0.001**
Bone cement leakage	/	59.2	89.8	0.493	0.746	0.653–0.839	< 0.001**
IVC	/	21.4	98.8	0.202	0.601	0.498–0.704	0.037*
Comprehensive prediction	0.130	83.3	84.8	0.681	0.895	0.836–0.954	< 0.001**

AUC, Area Under Curve; PM, psoas major; ES, erector spinae; FIR, fat infiltration rate; IVC, intravertebral vacuum cleft; Compared with the other groups, **P < 0.001, *P < 0.05

Discussion

The purpose of this study was to determine the risk factors associated with the incidence of SVCF after PVA. Multifactorial logistic regression analysis showed that PM(FIR), ES(FIR), cement leakage and IVC were all risk factors for the occurrence of SVCF. Moreover, we tested the predictive efficacy of independent risk factors by ROC curves. We noticed that PM(FIR), ES(FIR), and cement leakage had good predictive ability for predicting the occurrence of SVCF after PVA, while IVC had weak predictive ability. PM(FIR) more than 5.490% and ES(FIR) more than 52.413% had a high possibility of the occurrence of SVCF after PVA. Meanwhile, when we combined all the above risk factors, the AUC was 0.895, which was higher than the AUC of any other single risk factor, indicated that the combination of the above independent risk factors had a great ability to predict the occurrence of SVCF in postoperative patients after PVA. It may be due to the factors that the occurrence of SVCF is not a result of each risk factor alone, but a result of the combined effect of multiple risk factors.

The mechanism of the development of SVCF after PVA is complex, and in the light of previous literature, we broadly classified the risk factors for the development of SVCF after PVA into two aspects. This includes patient biological characteristics (advanced age, low BMI, paraspinal muscle degeneration, IVC, etc.) and surgical manipulation-induced biomechanical changes (amount of cement injected, classification of bone cement diffusion, cement leakage, etc.).

Biological characteristics

Age, BMI, BMD-T

One study suggested that the risk of SVCF increased by 3.0% for each additional year of age [14], but others have suggested that age is not a risk factor for the development of SVCF [15], while our results showed that age is not a risk factor.

In studies of OP, low BMI is a risk factor for vertebral fractures, and slimmer individuals are more likely to re-fracture [16]. All of the above are consistent with the results of our univariate regression analysis, but they were not statistically difference in the multivariate regression analysis owing to the interaction of the variables.

Consistent with the previous results of some scholars [17–19], the present study showed that lumbar BMD-T was not a risk factor for SVCF, whereas other scholars suggested that low BMD was significantly associated with the occurrence of SVCF [20, 21]. It could be caused by the fact that dual-energy X-ray absorptiometry measures BMD in only one plane, and the measurement is affected by many factors (concomitant compression fracture, spinal degeneration, spinal bone mass, etc.). Moreover, bone strength is a combination of BMD and bone architecture, therefore, basing on BMD-T alone without considering bone mass is not sufficient for predicting the risk of SVCF [19].

Paraspinal muscle degeneration

Paraspinal muscles are considered to be the main stabilisers of the spine, and as the adipose tissue is not contractile, fat infiltration impairs the quality of the paraspinal muscles [22]. It has been reported that paraspinal muscle degeneration is a risk factor for SVCF after PVA, and fatty infiltration of the paraspinal muscles is significantly associated with delayed healing of OVCF [23]. The FIR of the ES is higher in patients with OP [24].The ES contributes significantly to the maintenance of spinal stability among all paravertebral muscles, and also plays a key role in reducing the risk of insufficient healing in patients with OVCF [25]. When ES mass is impaired through fatty infiltration, it may have a reduced supportive effect on the lumbar dorsal muscles, which may in consequence increase the risk of the development of SVCF after PVA [26]. When the posterior lumbar paraspinal muscles (ES and MF) are degraded, the PM may increase muscular activity and reduce fat infiltration to stabilise the lumbar spine and maintaining lumbar lordosis [27]. When PM fatty infiltration is abundant, spinal stability is altered, which in turn increases the risk of vertebral collapse [28]. MF atrophy or fatty degeneration is not a risk factor for SVCF after PVA, and our results are consistent with those of a recent study [29]. Therefore assessing the FIR of the paravertebral muscles could help to identify individuals with a high risk of SVCF. It has been demonstrated that vitamin D and calcium supplementation has a positive effect on slowing down the process of muscle atrophy [30, 31]. Therefore, vitamin D and calcium supplementation may be able to slow down the process of paraspinal muscle atrophy and thus reduce the incidence of SVCF. Patients who underwent low back muscle exercises (five-point brace, three-point brace, gluteal bridge and one-point brace) not only had a significantly lower incidence of OVCF, but also contributed to the maintenance of BMD [32]. Therefore, it is recommended that patients with OVCF should actively exercise their low back muscles after undergoing PVA.

IVC

Similar to our findings, other scholars have suggested that IVC is a risk factor for the development of SVCF after PVA [33]. This may be attributed to the fact that the fibrocartilage membrane surrounding the IVC prevents cement diffusion, while the bonding between the cement and the surrounding cancellous bone is not satisfactory, resulting in poor cement diffusion after PVA [34], thereby increasing the risk of the development of SVCF after PVA. Li et al. found that IVC was associated with a higher incidence of cement leakage [35], and a comparably higher rate of cement leakage increased the probability of SVCF [36]. In order to reduce the incidence of cement leakage caused by IVC, one scholar pumped the fluid in the injured vertebrae when the puncture needle arrives at the vertebral clef and injected bone cement-gelatin sponge composite during the surgical procedure, and verified the method significantly reduced the incidence of bone cement leakage [12]. A fracture of the vertebrae results in a bony discontinuity, and in turn the central portion of the callus undergoes cystic degeneration, allowing the synovial membrane, which secretes synovial fluid, to encapsulate the entire cavity [37]. And the fluid it secretes is transudate [38].

Surgical manipulation-induced biomechanical changes

Previous research has demonstrated that the risk of SVCF in patients with cement leakage is 4.60 times higher than that in patients without cement leakage [36], whereas in our study it was 9.04 times higher, thus we suggest that cement leakage during surgery should be avoided to the extent possible. This is probably because leakage of bone cement into the intervertebral space mechanically irritates the vertebral endplates and accelerates disc degeneration, which increases the stress on the adjacent vertebrae and thus increases the risk of SVCF [21]. Consistent with our studies, other scholars have suggested that the amount of bone cement injected, the classification of bone cement diffusion and the occurrence of SVCF are not significantly associated [39, 40]. However, the greater the amount of bone cement, the more likely it is to cause cement leakage [41]. Therefore, the amount of bone cement injected during PVA should be carefully monitored as much as possible, and injected while fluoroscopy is performed to avoid cement leakage.

Shortcomings and prospects

Above all, as a retrospective study with a limited sample size, there is the possibility of a certain degree of systematic bias, and some of the data may have been omitted due to the occurrence of asymptomatic SVCF. In addition, an exercise prescription for postoperative PVA patients could be developed, so as to analyse whether the exercise prescription could reduce the FIR of the paraspinal muscles of the patients, and thus decrease the incidence of SVCF. We will conduct a large-sample, multicentre prospective study to confirm the role of these factors and determine the best preventive strategies and management measures in conjunction with clinical practice.

Conclusion

In this study, logistic regression combined with ROC curve analysis indicated that FIR of PM and ES, cement leakage in the intervertebral space, and IVC were risk factors for the occurrence of SVCF after PVA. PM(FIR) more than 5.490% and ES(FIR) more than 52.413% had a high possibility of the occurrence of SVCF after PVA.

Abbreviations

SVCF

Secondary vertebral compression fracture

PVA

Percutaneous vertebral augmentation

BMD-T

T value of bone mineral density

BMD-T

Bone mineral density

IVC

Intravertebral vacuum cleft

FIR

Fat infiltration rate

ROC

Receiver operating characteristic curve

OP

Osteoporosis

OVCF

Osteoporotic vertebral compression fracture

BMI

Body mass index

MF

Multifidus

ES

Erector spinae

PM

Psoas major

ROI

Region of interest

CSA

Cross-sectional area

BL

Body length

BW

Body width

BH

Body height

AVRR

Anterior vertebral recovery ratio

AUC

The area under curve

Author contributions

JT, SYW contributed to the conception of the study. SYW prepares, creates, or expresses content for publication, especially writing first drafts, including substantive translations. JT, TL, JNW, XKW and LLC prepares or presents content for publication, especially comments, annotations, or revisions, including such work that occurs before and after publication. JT and WX verifies the replication and reproduction of research results, experiments or other research outputs, both in whole and in portions. SYW creates tables and pictures.

Funding

No funding.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The Affiliated Hospital of Wuhan Sports University Ethics Committee approved this retrospective observational study (approval no. 672HRBC20240912-L32).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.