The relationship between the expression of serum asprosin and miR-21 in patients with osteoporosis and delayed healing after OVCF surgery

Abstract

Objective: To explore the relationship between serum asprosin, microRNA-21 (miR-21) expression, and delayed healing after surgery in patients with osteoporotic vertebral compression fractures (OVCF), and to construct a nomogram model. Methods: A prospective study inducted 300 OVCF patients treated with percutaneous vertebroplasty (PVP) from June 2022 to June 2024. Serum asprosin and miR-21 were measured preoperatively, and fracture healing was assessed via X-ray 3 months post-surgery. Patients were categorized into delayed healing and normal healing groups based on outcomes. The least absolute shrinkage and selection algorithm (LASSO) regression identified factors influencing delayed healing, followed by binary logistic regression analysis. A nomogram model was constructed to predict delayed healing, and its predictive value was evaluated using the receiver operating curve (ROC) analysis. Results: Results showed higher rates of delayed weight-bearing (58.11% vs 33.19%), diabetes prevalence (52.70% vs 42.48%), bone cement injection volume (4.68 ± 1.14 mL vs 3.81 ± 1.09 mL), and serum asprosin levels (4.09 ± 1.39 ng/mL vs 3.14 ± 1.07 ng/mL) in the delayed healing group compared to the normal healing group. Serum miR-21 levels were lower in the delayed healing group (0.69 ± 0.19) than in the normal group (0.92 ± 0.31) (p < 0.05). Bone density T scores in OVCF patients correlated positively with asprosin (r = 0.281, p < 0.001) and negatively with miR-21 (r = −0.184, p = 0.001). The LASSO regression identified five factors associated with delayed healing: bone cement volume, delayed weight-bearing, diabetes, and serum asprosin (risk factors), while high miR-21 was protective. Logistic regression indicated significant risk factors with an overall C-index of 0.867 and AUC of 0.868 (sensitivity 0.892, specificity 0.686). After 3 months of treatment, serum Asprosin [(3.46 ± 1.22) ng/ml] in OVCF patients was lower than that before treatment [(3.03 ± 1.03) ng/ml], and serum miR-21 (0.85 ± 0.30) was higher than that before treatment [(0.99 ± 0.33)] (t = 4.039, 4.318, p < 0.05). Conclusion: Serum asprosin and miR-21 levels are closely related to delayed healing after surgery in patients with OVCF. Additionally, the bone cement injection volume, delayed weight-bearing, and concomitant diabetes mellitus are also important factors affecting fracture healing in patients. The nomogram model based on these factors can effectively predict the risk of delayed healing in patients with OVCF after surgery.

Keywords

asprosin delayed healing microRNA-21 nomogram osteoporotic vertebral compression fracture percutaneous vertebroplasty

Introduction

Osteoporosis (OP) is a systemic metabolic bone disease characterized by decreased bone mass, decreased bone strength and increased bone fragility. Osteoporotic vertebral compression fracture (OVCF) is a common complication of OP. OVCF can lead to loss of fracture cone height, kyphosis and intractable back pain.¹ At the same time, the expert consensus in 2021 shows that the mortality rate of OVCF patients after 1 year of fracture is significantly higher than that of the general population, and the 4-year survival rate is only 50 %, which seriously threatens the life safety of patients.²

Percutaneous vertebroplasty (PVP) is a classic surgical procedure for treating osteoporotic vertebral compression fractures (OVCF). It involves percutaneous injection of bone cement into the fractured vertebra to enhance its strength and stiffness, thereby preventing further vertebral collapse and deformity.³ However, as most OVCF patients are elderly, their post-fracture callus formation tends to be slow. As shown in the study by Chen Weina et al.,⁴ the incidence of delayed fracture healing after surgery in OVCF patients is as high as 45.57%. Delayed fracture healing can prevent patients from bearing normal loads or resuming regular work, leaving them with persistent low back pain. Some patients may even require secondary surgery, which not only increases their financial burden but also raises the risk of poor prognosis.⁵ Therefore, there is an urgent clinical need to investigate the factors influencing delayed healing in OVCF patients postoperatively and to construct a predictive model.

Asprosin is a peptide hormone and a C-terminal cleavage product of fibrillin. In the review of Ovali MA,⁶ it was mentioned that Asprosin can affect the bone mineral density of patients through various ways such as glucose metabolism and regulation of insulin sensitivity. MicroRNA-21 (miR-21) is located on human chromosome 17q23.2. miR-21 has always been considered as a proto-oncogene. At present, clinical research focuses on angiogenesis, inflammation and immunity.^7,8 However, some studies have shown that miR-21 is also involved in bone diseases. In the study of Li X,⁹ miR-21 can be directly involved in regulating the expression of bone marrow mesenchymal stem cells, osteoblasts and osteoclasts, and regulating bone tissue balance. Therefore, it is speculated that serum Asprosin and miR-21 may be involved in the postoperative fracture healing of OVCF patients. In view of this, this study focuses on the analysis of the relationship between the two and the postoperative delayed healing of patients.

Materials and methods

Study design

A prospective study was conducted, and 300 OVCF patients admitted to the hospital from June 2022 to June 2024 were selected as the study subjects. All patients underwent preoperative serum asprosin and miR-21 testing. Three months post-surgery, X-rays were used to assess fracture healing, and the patients were divided into a delayed healing group and a normal healing group. The baseline data of the two groups were statistically analyzed and compared. The least absolute shrinkage and selection algorithm (LASSO) regression was used to screen the factors that may affect the delayed healing of OVCF patients after surgery, and binary logistics regression analysis was performed. A nomogram model for delayed healing of OVCF patients after surgery was constructed, and the receiver operating characteristic (ROC) curve was drawn to evaluate the value of the nomogram model in predicting delayed healing of OVCF patients after surgery.

Patient population

Inclusion criteria: ① OVCF meets the diagnostic criteria of the 2021 consensus guidelines²; ② PVP treatment is performed in accordance with the “Expert Consensus on Percutaneous Vertebroplasty Surgical Techniques”¹⁰; ③ Surgery is performed by the same group of physicians, and there are no serious complications after the surgery; ④ Single vertebral compression fracture with no symptoms of spinal cord compression; ⑤ Age ≥18 years. Exclusion criteria: ① Secondary osteoporosis; ② Recent use of medications that may affect bone metabolism; ③ Concomitant thyroid disease; ④ Combined with other orthopedic diseases; ⑤ History of thoracolumbar surgery; ⑤ Combined with immune system diseases.

According to the above inclusion and exclusion criteria, a prospective study was conducted and 300 patients with OVCF admitted to the hospital from June 2022 to June 2024 were selected as study subjects. This study was approved by the hospital’s medical ethics committee, and informed consent was obtained from all patients.

Baseline data

Baseline data of the patients were collected through the hospital medical records system, including gender, age, body mass index (BMI), fracture site, bone mineral density T score [measured using a dual-energy X-ray absorptiometry device (Osteosys Co., Ltd, CFDA approval number: 20172061588)], height recovery rate of injured vertebral, improvement of kyphosis Cobb angle, bone cement injection volume, cause of the disease, delayed weight-bearing after surgery (unable to complete weight-bearing exercises more than 12 weeks after surgery), combined hyperlipidemia, hypertension, and diabetes, smoking history and drinking history.

Criteria for delayed healing and grouping

According to the criteria in the Diagnosis and treatment of nonunion of osteoporotic thoracolumbar fractures,¹¹ X-ray examination (Canon, CFDA approval number: 20173061346) was performed 3 months after surgery. Delayed healing was defined as no or minimal callus formation, visible gaps at the fracture site, and sclerotic bone edges. Patients with delayed fracture healing were assigned to the delayed healing group (n = 74), while those with normal healing were assigned to the normal healing group (n = 226).

Serum asprosin and miR-21 testing methods

Before surgery, 5 mL of fasting venous blood was collected from the patients and centrifuged at 3500 rpm for 15 minutes with a 10 cm radius using a centrifuge centrifuge (Fresenius Kabi AG, CFDA approval number: 20183100418) to separate the serum. Serum asprosin levels were measured using enzyme-linked immunosorbent assay (ELISA). Total RNA was extracted from the serum using the TRIzol method, and reverse transcription was performed using a reverse transcription kit to convert total RNA into complementary DNA (cDNA). Reverse transcription conditions were 37°C for 15 minutes and 85°C for 5 seconds. PCR amplification was conducted using cDNA as a template. Primer sequences were as follows: forward primer 5′-GCGGTAGCTTATCAGACTGA-3′, reverse primer: 5′-TGCGTGTCGTGGAGTC-3′. PCR reaction conditions were: 95°C for 10s, 95°C for 5s, 60°C for 30 s, for a total of 40 cycles. Using U6 as an internal control, and the relative expression level of miR-21 was calculated using the 2- ^∆∆CT method. All kits were provided by Roche Molecular Systems, Inc.

Statistical methods

SPSS25.0 software was used for data processing. The measurement data were expressed as “ $\bar{x}$ ± s”. The independent sample t test was used for comparison between the two groups. Paired t test was used for intra-group comparison. The count data were expressed as n% and compared using the χ² test. The LASSO regression algorithm was used with the “glmnet” package in the R4.1.3 to screen the factors related to delayed healing after surgery in OVCF patients, and binary logistic regression analysis was performed. The R (R4.1.0) software package and the “rms” package were used to establish a nomogram model for predicting delayed healing after surgery in OVCF patients. The “caret” package was used for bootstrap method to calculate the consistency index (C-index). The receiver operating characteristic (ROC) curve was drawn, and the area under the curve (AUC) was calculated to test the predictive value of the nomogram model. The significance level was set at α = 0.05.

Results

Comparison of baseline data between the delayed healing group and the normal healing group

After enrolling 74 cases in the delayed healing group and 226 cases in the normal healing group, the patients’ gender, age, body mass index (BMI), fracture site, bone mineral density T score, intraoperative blood loss, vertebral height recovery rate, kyphosis Cobb angle improvement, bone cement injection volume, cause of disease, delayed postoperative weight-bearing, combined hyperlipidemia, hypertension, and diabetes, smoking history and drinking history, serum levels of asprosin, and miR-21 were analyzed. The results showed that the delayed weight-bearing, the proportion of patients with diabetes and the bone cement injection volume, and the serum asprosin level in the delayed healing group [58.11% (43/74), 52.70% (39/74), (4.68 ± 1.14) ml, (4.09 ± 1.39) ng/ml] were higher than those in the normal healing group [33.19% (75/226), 42.48% (96/226), (3.81 ± 1.09) ml, (3.14 ± 1.07) ng/m] (p < 0.05). In addition, the serum miR-21 level in the delayed healing group [(0.69 ± 0.19)] was lower than that in the normal healing group [(0.92 ± 0.31)] (p < 0.05). See Table 1 for details.

Table 1.

Comparison of baseline data between the delayed healing group and normal healing group.

Characteristics		Delayed healing group (n = 74)	Normal healing group (n = 226)	χ²/t	p
Gender	Male	39 (52.70)	129 (57.08)	0.433	0.510
Gender	Female	35 (47.30)	97 (42.92)	0.433	0.510
Age	≥60 years	61 (82.43)	177 (78.32)	0.575	0.448
Age	<60 years old	13 (17.57)	49 (21.68)	0.575	0.448
BMI (kg/m²)		24.62 ± 1.17	24.78 ± 1.36	0.908	0.365
Fracture site	Lumbar vertebra	48 (64.86)	152 (67.26)	0.144	0.705
Fracture site	Thoracic vertebra	26 (35.14)	74 (32.74)	0.144	0.705
BMD T score		−3.06 ± 0.64	−2.95 ± 0.72	1.171	0.243
Intraoperative blood loss (ml)		82.54 ± 9.69	81.04 ± 9.04	1.217	0.225
Vertebra height recovery rate (%)		79.62 ± 20.42	76.14 ± 18.34	1.561	0.120
Kyphosis Cobb angle improvement (°)		5.67 ± 1.55	5.36 ± 1.39	1.618	0.107
Bone cement injection volume (ml)		4.68 ± 1.14	3.81 ± 1.09	5.905	<0.001
Causes of injury	Slip	40 (54.05)	124 (54.87)	0.041	0.998
	Fall	21 (28.39)	64 (28.32)
	Heavy lifting injuries	9 (12.16)	27 (11.95)
	Other	4 (5.41)	11 (4.87)
Delayed postoperative weight-bearing	Yes	43 (58.11)	75 (33.19)	14.511	<0.001
Delayed postoperative weight-bearing	No	31 (41.89)	151 (66.81)	14.511	<0.001
Combined hyperlipidemia	Yes	22 (29.73)	58 (25.66)	0.471	0.492
Combined hyperlipidemia	No	52 (70.27)	168 (74.34)	0.471	0.492
Combined hypertension	Yes	39 (52.70)	96 (42.48)	2.355	0.125
Combined hypertension	No	35 (47.30)	130 (57.52)	2.355	0.125
Combined diabetes	Yes	20 (27.03)	32 (14.16)	6.442	0.011
Combined diabetes	No	54 (72.97)	194 (85.84)	6.442	0.011
Smoking	Yes	28 (37.84)	71 (31.42)	1.040	0.308
Smoking	No	46 (62.16)	155 (68.58)	1.040	0.308
Drinking	Yes	32 (43.24)	88 (38.94)	0.431	0.512
Drinking	No	42 (56.76)	138 (61.06)	0.431	0.512
Asprosin (ng/ml)		4.09 ± 1.39	3.14 ± 1.07	6.160	<0.001
miR-21		0.69 ± 0.19	0.92 ± 0.31	5.697	<0.001

Correlation between serum asprosin, miR-21 levels, and bone mineral density T score in OVCF patients

The bone density T score of OVCF patients was positively correlated with asprosin (r = 0.281, p < 0.001) and negatively correlated with miR-21 (r = −0.184, p = 0.001). The scatter plots are shown in Figures 1 and 2.

Figure 1.

Scatter plot of bone density T score and serum Asprosin levels in OVCF patients.

Figure 2.

Scatter plot of bone density T score and serum miR-21 levels in OVCF patients.

LASSO regression screening results for delayed healing in OVCF patients

The LASSO regression model identified five potential factors associated with postoperative delayed healing in patients with OVCF. These factors are: bone cement injection volume, postoperative delayed weight-bearing, combined diabetes, serum asprosin levels, and serum miR-21 levels. See details in Figures 3 and 4.

Figure 3.

LASSO regression coefficient path.

Figure 4.

LASSO regression cross validation results.

Regression analysis of influencing factors of delayed healing in OVCF patients

The occurrence of delayed healing after surgery in OVCF patients was used as the dependent variable (“1” = delayed healing group, “0” = normal healing group), and the bone cement injection volume, postoperative delayed weight-bearing, combined diabetes, serum asprosin levels, and serum miR-21 levels were used as independent variables (postoperative delayed weight-bearing and combined diabetes, both 1 = “yes”, 0 = “no”, and the rest were continuous variables). The results of binary Logistic regression analysis showed that delayed weight-bearing, combined diabetes, bone cement injection volume, and high serum asprosin levels were risk factors for delayed healing after surgery in OVCF patients (OR >1, p < 0.05), and high serum miR-21 levels were protective factors (OR <1, p < 0.05). See Table 2 for details.

Table 2.

Regression analysis of influencing factors for delayed healing in OVCF patients.

Influencing factors	B	SE	Wald	p	OR	95% confidence interval
Bone cement injection volume	0.712	0.156	20.684	<0.001	2.037	1.1499–2.768
Postoperative delayed weight bearing	1.083	0.339	10.228	0.001	2.954	1.521–5.736
Combined diabetes	1.065	0.407	6.859	0.009	2.901	1.307–6.436
Serum asprosin	0.791	0.152	27.227	<0.001	2.205	1.638–2.967
Serum miR-21	−2.922	0.663	19.448	<0.001	0.054	0.015–0.197
Constant	−5.334	1.068	24.942	<0.001	-	-

Construction of a nomogram prediction model for delayed healing in OVCF patients

Based on the factors with differences in Table 2 (bone cement injection volume, delayed weight bearing, combined diabetes, serum asprosin, serum miR-21), a nomogram prediction model for delayed healing after surgery in OVCF patients was constructed (see Figure 5). The nomogram model was validated using the Bootstrap internal validation method (see Figure 6), and the results showed that the C-index value was 0.867.

Figure 5.

Nomogram for delayed healing after OVCF surgery.

Figure 6.

Calibration curve of nomogram for delayed healing after surgery in OVCF patients.

The value of the nomogram model in predicting delayed healing after OVCF surgery

The ROC curve was drawn to perform internal validation of the nomogram model. The results showed that the AUC of the nomogram for predicting delayed healing after surgery in OVCF patients was 0.868 (95% confidence interval 0.824-0.911, p < 0.001), the specificity was 0.686, the sensitivity was 0.892, and the Youden index was 0.578. See Figure 7 for details.

Figure 7.

ROC curve for the nomogram model predicting delayed healing after surgery in OVCF patients.

Serum levels of asprosin and miR-21 in OVCF patients before treatment and 3 months after treatment

After 3 months of treatment, serum Asprosin [(3.46 ± 1.22) ng/ml] in OVCF patients was lower than that before treatment [(3.03 ± 1.03) ng/ml], and serum miR-21 (0.85 ± 0.30) was higher than that before treatment [ (0.99 ± 0.33)] (t = 4.039, 4.318, p < 0.05).

Discussion

As a mature minimally invasive spinal surgery, PVP is widely used in clinical practice due to its advantages in improving local spinal symptoms and restoring limb mechanical strength. Most OVCF patients can achieve good recovery of the fractured vertebra after PVP surgery, however, a small number of patients experience delayed healing or even non-healing of fractures after surgery. The research data of Jia Jinlong et al.¹² showed that the incidence of delayed healing of fractures after PVP surgery in OVCF patients was 26.53%, which is similar to the 24.67% in this study.

Fracture healing is a trauma repair process in which osteoblasts and osteoclasts interact with each other, and the proliferation and differentiation of both play a decisive role in this process.¹³ Recent studies have shown that some adipose factors can regulate the proliferation and differentiation of osteoclasts and osteoblasts, inhibit the expression of intracellular osteogenic signals, and thus affect bone metabolism and participate in bone healing.¹⁴ Asprosin is a secretory adipose factor secreted by white adipose tissue. Some studies have shown that changes in asprosin levels are closely related to bone mineral density in diabetic patients¹⁵ miR-21 is an important member of the miRNA family. Existing studies have shown that miR-21 plays an important role in embryonic development and cell differentiation.¹⁶ Furthermore, Meng Yubin’s¹⁷ doctoral thesis found that high expression of miR-21 can increase the expression of osteogenic-related genes in human umbilical cord mesenchymal stem cells, thereby participating in the osteogenic differentiation process. Therefore, serum asprosin and miR-21 levels may be related to postoperative fracture healing in OVCF patients.

Five potential factors related to delayed healing after surgery in OVCF patients were finally screened out through the LASSO regression model: the bone cement injection volume, postoperative delayed weight bearing, combined diabetes, serum asprosin level, and serum miR-21 level. High expression of serum asprosin: Xue Guohui et al.¹⁸ found that asprosin can promote the release of inflammatory factors such as tumor necrosis factor-α (TNF-α) in pancreatic β cells and skeletal muscle cells. TNF-α can directly or indirectly regulate bone metabolism through inflammatory and oxidative stress response pathways, regulate the proliferation and differentiation of osteoclasts and osteoblasts, and accelerate bone loss in OVCF patients.¹⁹ Additionally, TNF-α can enhance the expression of V-ATPase on osteoclasts, thereby enhancing the bone resorption activity of osteoclasts, destroying the balance between osteoblasts and osteoclasts, and thus affecting the postoperative bone healing of patients.²⁰ Asprosin can also promote the expression of Toll-like receptor 4 (TLR4), activate the phosphorylation of JNK, and mediate β-cell inflammatory response, cell dysfunction and apoptosis through the TLR4/JNK signaling pathway, thereby affecting the postoperative recovery of OVCF patients. TLR4 can also disrupt the normal function of osteoblasts and osteocytes, disrupt the balance of bone metabolism, and thus affect bone healing. In a mouse experiment by Cao,²¹ it was found that knocking out TLR4 in diabetic osteoporosis mice could effectively delay the progression of the disease in mice. This shows that asprosin can indirectly affect the bone metabolism level of OVCF patients by promoting TLR4 expression and inflammatory response, thereby increasing the risk of delayed healing after surgery. Low expression of miR-21: With the deepening of clinical research, more and more studies have found that miR-21 is closely related to osteoblast differentiation. Studies have shown that miR-21 can regulate osteogenesis by targeting specific target genes. Spry2 is one of the target genes and is a key regulatory factor for osteoblast differentiation.²² Yang Nan²³ found that miR-21 can upregulate the osteoblast-related gene core binding factor subunit α1 by regulating the signal transduction of Spry2, promote the differentiation of mesenchymal stem cells into osteoblasts, and promote bone healing. In addition to regulating osteoblasts, a 2020 study found that miR-21 can negatively regulate osteoclastogenesis and bone resorption by intervening in the target gene PTEN and activating the PI3K/Akt signaling pathway.²² Therefore, miR-21 can effectively regulate the balance of osteoblasts and osteoclasts and promote bone healing.²⁴

The results of binary logistic regression analysis showed that, in addition to serum asprosin and miR-21 levels, delayed weight-bearing after surgery, combined diabetes, and bone cement injection volume were also influencing factors for delayed healing after surgery in OVCF patients. Delayed postoperative weight-bearing: Fracture healing is a complex biological repair process, which is affected by mechanical stress stimulation and blood supply. Mechanical stress stimulation is one of the important factors, and weight-bearing is one of the common ways of mechanical stimulation of fracture callus.²⁵ Schwarz C^’s²⁶ rat study showed that the newly formed bone tissue at the fracture site of rats in the mechanical loaded group was significantly higher than that in the unloaded group, suggesting that mechanical stress stimulation can stimulate osteoblast proliferation, differentiation, and secretion of extracellular matrix. Additionally, mechanical stress stimulation can also improve the bone microenvironment, promote the release of growth factors at the fracture site, promote local angiogenesis, and thus promote fracture healing. Combined diabetes: Due to changes in glucose homeostasis, diabetic patients can inhibit osteocalcin and insulin-like growth factor-1 and increase reactive oxygen species, thereby causing bone loss. Long-term hyperglycemia can directly affect the function and differentiation of osteoblasts, and has a toxic effect on the differentiation of bone marrow interstitial cells such as adipocytes. It can also increase the production of bone marrow adipocytes by upregulating oxidase proliferator-activated receptor γ, thereby affecting the function of osteoblasts and osteoclasts, leading to delayed fracture healing.²⁷ Bone cement injection volume: A higher injection volume of bone cement indicates more severe disease, making postoperative fracture healing more difficult. Lai Xianliang ‘s²⁸ study showed that a large amount of bone cement injected does not significantly improve the patient’s clinical efficacy, but can cause the bone cement penetration rate to be as high as 30% to 41%, thereby affecting the patient’s postoperative fracture healing. The ROC results showed that the AUC of the nomogram for predicting delayed healing after surgery in OVCF patients was 0.868 (95% confidence interval 0.824-0.911, p < 0.001), the specificity was 0.686 , the sensitivity was 0.892, and the Youden index was 0.578, indicating that the nomogram model constructed based on serum asprosin and miR-21 can effectively predict the risk of delayed healing after surgery in OVCF patients.

However, this study has certain limitations, such as a small sample size and single-center design, which may limit the generalizability of the results. Additionally, only preoperative serum Asprosin and miR-21 levels were evaluated, and continuous monitoring of these levels in OVCF patients was not performed, which could have helped in analyzing the relationship between their changes and fracture healing. Future studies should consider a multi-center design and a larger sample size to validate our findings. Moreover, further mechanistic studies should be included to explore the specific role of serum Asprosin and miR-21 in the physiological process of fracture healing in OVCF patients.

In conclusion, serum Asprosin and miR-21 levels are closely associated with delayed fracture healing in OVCF patients. Additionally, bone cement injection volume, delayed postoperative weight-bearing, and diabetes comorbidity are important factors affecting fracture healing. A nomogram model based on these factors can effectively predict the risk of delayed fracture healing in OVCF patients.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by Anhui Medical University Fund Support Project (2023xkj097) and Key Project of Natural Science Research in Anhui Provincial Universities by the Education Department of Anhui Province (2022AH050749).

ORCID iD

Hong Pan

Data Availability Statement

Data available on request from the authors.*

References

Wang

Zheng

, et al. Analysis of the most influential publications on vertebral augmentation for treating osteoporotic vertebral compression fracture: a review. Medicine (Baltim) 2022; 101(31): e30023.

Chinese Society of Rehabilitation Medicine Osteoporosis Prevention and Rehabilitation Professional Committee . Expert consensus on the diagnosis and treatment of osteoporotic vertebral compression fractures (2021 edition). Chin Med J 2021; 101(41): 3371–3379.

Dai

Liang

Zhang

, et al. Risk factors of vertebral re-fracture after PVP or PKP for osteoporotic vertebral compression fractures, especially in Eastern Asia: a systematic review and meta-analysis. J Orthop Surg Res 2022; 17(1): 161.

Weina

Liang

Liye

. The relationship between serum miR-125a-3p, BMP-2 levels and delayed healing after osteoporotic vertebral compression fractures. Shandong Med J 2023; 63(25): 56–59.

Mick

Fischer

. Delayed fracture healing. Semin Muscoskel Radiol 2022; 26(3): 329–337.

Ovali

Bozgeyik

. Asprosin, a C-terminal cleavage product of fibrillin 1 encoded by the FBN1 gene, in health and disease. Mol Syndromol 2022; 13(3): 175–183.

Ciuffi

Marini

Fossi

, et al. Circulating MicroRNAs as biomarkers of osteoporosis and fragility fractures. J Clin Endocrinol Metab 2022; 107(8): 2267–2285.

Chen

Huang

, et al. Morinda officinalis polysaccharide regulates rat bone mesenchymal stem cell osteogenic-adipogenic differentiation in osteoporosis by upregulating miR-21 and activating the PI3K/AKT pathway. Kaohsiung J Med Sci 2022; 38(7): 675–685.

Guo

Liu

, et al. MicroRNA-21 promotes osteogenesis of bone marrow mesenchymal stem cells via the Smad7-Smad1/5/8-Runx2 pathway. Biochem Biophys Res Commun 2017; 493(2): 928–933.

10.

Interventional Group of Radiology Branch of Chinese Medical Association . Expert consensus on percutaneous vertebroplasty operation technology. Chin J Radiol 2014; 48(1): 6–9.

11.

Huilin

Genlin

Guoqi

, et al. Diagnosis and treatment of nonunion of osteoporotic thoracolumbar fractures. Chin J Orthop 2007; 27(9): 682–686.

12.

Jinlong

Chaoyang

Junjian

, et al. The correlation between miR-24-3p, Smad5 and fracture healing after PVP in patients with osteoporotic spinal fractures. J Media Res 2023; 52(9): 160–164.

13.

Chen

Xiong

, et al. Osteoblast/osteoclast and immune cocktail therapy of an exosome/drug delivery multifunctional hydrogel accelerates fracture repair. ACS Nano 2022; 16(1): 771–782.

14.

Chen

Shi

Xie

, et al. Apelin-13 induces mitophagy in bone marrow mesenchymal stem cells to suppress intracellular oxidative stress and ameliorate osteoporosis by activation of AMPK signaling pathway. Free Radic Biol Med 2021; 163(163): 356–368.

15.

Haiyan

Yamei

, et al. The correlation between serum IL-33, Asprosin levels and type 2 diabetes mellitus complicated with osteoporosis. Shandong Medicine 2023; 63(22): 17–20.

16.

Guan

Tang

, et al. Exosomes derived from microRNA-21 overexpressed adipose tissue-derived mesenchymal stem cells alleviate spine osteoporosis in ankylosing spondylitis mice. J Tissue Eng Regen Med 2022; 16(7): 634–642.

17.

Yubin

. Functionalization of microRNA nanocapsules to promote bone healing on titanium alloy surface[D]. Tianjin, China: Tianjin University, 2016.

18.

Guohui

Lin

Liang

, et al. The expression and significance of adipokines asprosin and metrnl in patients with osteoarthritis. Chin J Clin Lab Sci 2023; 41(12): 912–915.

19.

Wang

. TNF-Alpha and IL-6: the link between immune and bone system. Curr Drug Targets 2020; 21(3): 213–227.

20.

Chen

Liu

, et al. Interaction between ferroptosis and TNF-α: impact in obesity-related osteoporosis. FASEB J 2023; 37(6): e22947.

21.

Cao

Han

Wang

, et al. TLR4 knoekout amelioratesstreptozotocin-indueed osteoporosis in a mouse model of diabetes. Biochem Biophys Res Commun 2021; 26(546): 185–191.

22.

Qin

Wang

, et al. Bone mesenchymal stem cell-derived sEV-encapsulated thermosensitive hydrogels accelerate osteogenes is and angiogenesis by release of exosomal miR-21. Front Bioeng Biotechnol 2022; 9(9): 829136.

23.

Nan

. The mechanism of miR-21 and its target gene Sprouty1 regulating osteogenic differentiation of human bone marrow mesenchymal stem cells[D]. Xi'an, China: Fourth Military Medical University of Chinese Pla, 2012.

24.

Wang

Liu

Wang

, et al. miR-21 promotes osteoclastogenesis through activation of PI3K/Akt signaling by targeting Pten in RAW264.7 cells. Mol Med Rep 2020; 21(3): 1125–1132.

25.

Miramini

Ganadhiepan

Lin

, et al. Influence of muscle loading on early-stage bone fracture healing. J Mech Behav Biomed Mater 2023; 138(138): 105621.

26.

Schwarz

Wulsten

Ellinghaus

, et al. Mechanical load modulates the stimulatory effect of BMP2 in a rat nonunion model. Tissue Eng 2013; 19(1-2): 247–254.

27.

Murray

Coleman

. Impact of diabetes mellitus on bone health. Int J Mol Sci 2019; 20(19): 4873.

28.

Xianliang

Jia

, et al. Effect of bone cement injection volume on compressive cervical spine fracture in lateral percutaneous vertebroplasty. China J Mod Med 2018; 28(5): 68–72.