Semi-quantitative study of magnetic resonance imaging features of disc-condylar complex in patients with anterior disc displacement without reduction of temporomandibular joint

Abstract

Purpose

To investigate the differences in magnetic resonance imaging (MRI) characteristics between anterior disc displacement with reduction(ADDWR) and anterior disc displacement without reduction (ADDWoR) of the temporomandibular joint (TMJ) by semi-quantitative analysis of the disc-condylar complex. Reliable imaging markers were selected to predict ADDWoR.

Methods

The risk factors of ADDWoR were analyzed by univariate and multivariate logistic regression methods, and the receiver operating curves(ROC) were plot to evaluate the diagnostic ability.

Results

There were statistically significant differences in disc shape, displacement degree, condylar shape, bone change, position and joint effusion between the two groups(p<0.05). Univariate logistics showed that fold-curved articular disc, degree of ADD, condylar bone change, beak condyle, central condyle and effusion grades were all related factors in the diagnosis of ADDWoR. Multivariate logistic showed fold-curved articular disc, the degree of ADD and the grade of joint effusion were independent factors for the diagnosis of ADDWoR. The area under ROC curve for ADD-degree and joint effusion analysis for the prediction of ADDWoR was 0.909 and 0.745,respectively.

Conclusion

Based on semi-quantitative analysis of MRI signs of disc-condylar complex, it can effectively evaluate the influencing factors of ADDWoR and provide imaging reference for clinical disease prediction of temporomandibular joint (TMJ) patients.

Background

Temporomandibular joint disorder(TMD) is a disease involving the mandibular muscles and/or the temporomandibular joint (TMJ), and is a general term for pain and dysfunction of the TMJ complex and masticatory muscles [1]. Causes include physical, psychological, genetic and environmental factors [2]. It is common worldwide and occurs mainly in individuals aged 20–45 years. Disc displacement and condylar imbalance are the most common causes. In the normal disk-condylar relationship, the posterior articular disk band is located near the 12 o ‘clock position above the condyle. When the articular disk moves to the anterior condyle or other non-12 o ‘clock position for some reason, it is considered to be a articular disk displacement, in which the anterior disc displacement(ADD) is the most common. Previous studies believed that ADD is a self-limiting disease, and its clinical symptoms such as pain and movement disorders can be relieved naturally, without treatment or active disc reduction [3]. However, with the development of the disease, the clinical symptoms of most patients can not be completely relieved, and there will be changes such as increased degree of ADD, joint disk deformation, condylar bone absorption, joint effusion. Even the disc progressed from anterior disc displacement with reduction(ADDWR) to anterior disc displacement without reduction (ADDWoR), and secondary facial deviation, mandibular retraction and other deformities [4].

Magnetic resonance imaging(MRI), as a non-invasive imaging method, has excellent performance in analyzing soft tissue and fluid in the TMJ complex, and is the gold standard for evaluating the TMJ complex, disc-condylar relationship, and disc displacement [5]. Based on the TMJ complex and disc-condylar relationship, combined with multiple MRI indicators, more complete image data can be provided [6]. Screening out imaging markers with diagnostic efficacy can improve the clinical diagnosis level of ADDWoR. Especially for patients with difficulty in opening, the prediction of ADDWoR on closed mouth images is helpful to promote the treatment and management of TMD, a chronic disease, and has important clinical significance. Most previous studies have only focused on the single relationship between the intervertebral disc and the condyle, neglecting the quantitative and qualitative understanding of its integrated structure. Some studies have quantified the TMJ with higher-level software outside of workstations, which is not suitable for clinical application. Therefore, we used 1.5T MRI to conduct a semi-quantitative study from three aspects: articular disc, condyle, and joint effusion. We integrated more imaging indicators, conducted quantitative or grading, and constructed the ADDWoR diagnostic prediction model, aiming to screen out imaging indicators that are more in line with clinical diagnosis. Data analysis and measurement were all carried out on the original workstation rather than external software, providing more direct and accessible imaging references for daily clinical diagnosis.

Data and methods

The retrospective study was approved by the ethics review committee of NO.1 Orthopedics Hospital of Chengdu, and was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2013, and the need to obtain informed consent was waived. This study included patients who visited the TMJ outpatient department of our hospital from January 2022 to June 2023. According to the DC/TMD diagnostic criteria [2], they were accompanied by TMD symptoms (joint pain, clicking or difficulty in opening the mouth) and were evaluated as ADD by MRI. Maxillofacial trauma and tumors, previous surgical history of TMJ, congenital malformations and specific inflammation were excluded. Cases with poor scan image quality were also excluded. The course of the disease ranges from 1 week to 3 years. Some patients have medical experience, but due to insufficient information, the details are not detailed.A total of 154 patients were included, and at least one joint met the criteria. Finally, A total of 279 joint images were included to evaluate the degree of ADD, reduction or not, disk-condylar relationship, and the grades of joint effusion.

Scanning method

Open and closed mouth scanning of the temporomandibular joint was performed with a combined imaging 1.5T magnetic resonance scanner (UNITED IMAGING uMR 588, China) and 16-channel cranial orthogonal coil. Scanning parameters: Field of view 120 mm×120 mm; Matrix 250 mm×192 mm; Layer thickness 2–3 mm; The layer spacing was 0.5 mm. T1-weighted imaging(T1WI), T2-weighted imaging(T2WI), T2-weighted imaging fat-suppressed sequence(T2WI-fs) and Proton density weighted imaging(PDWI) were routinely scanned. Oblique coronal images (T1WI) and oblique sagittal images (PDWI and T2WI-fs) were collected at the closed position. The oblique sagittal phase (PDWI) is collected at the open mouth position.

Image analysis

The images were analyzed jointly by one radiologist with an intermediate professional title and one specialist in TMJ with a senior professional title. All images were observed on the PACS system as the original DICOM images, and all data were measured on the workstation.

Temporomandibular joint disc morphology

Based on the closed position of the oblique sagittal plane PDWI, the disc morphology was divided into 5 types. (1) Biconcave: normal articular disc, the middle band is narrow, and the posterior and anterior bands are completely visible. (2) Average thickness: the front, middle and back of the joint disc with equal thickness. (3) Convex: the middle band of the joint disc increases in thickness and is shaped like a round ball. (4) Fold-curved: joint disc contracture, shortened length, irregular shape. (5) Stretching: the joint disc is elongated and uneven in thickness.

Degree of ADD, ADDWR, ADDWoR

At the closed oblique sagittal position, a line was drawn between the lowest point at the back of the articular fossa and the lowest point of the articular tubercle, marking the midpoint of the condylar portion of this line. In the closed oblique sagittal position, a line is drawn between the lowest point of the posterior fossa and the lowest point of the articular tubercle, marking the midpoint of the line through the condylar portion. The Angle between the line that passes through the midpoint and the midpoint of the rear edge of the joint disc, and the vertical line that passes through the midpoint, is the disk boundary Angle [7], as shown in Fig. 1a, and the normal disk boundary angle is within 10° of the front and back.

Fig. 1.

(a) Schematic diagram for measuring the disk boundary Angle; (b) Mild ADD, the disk boundary Angle = 23.8°; (c) Moderate ADD, the disk boundary Angle = 46.5°; (d) severe ADD, the disk boundary Angle = 68.3°

In the open oblique sagittal image, the disk-condylar position relationship returned to normal and was defined as ADDWR. The disk-condylar position relationship could not return to normal, and the disc remained in front of the condyle, which was defined as ADDWoR.

The group was divided according to the Angle size. When the Angle value was in the range of 11° to 30°, it was classified as the mild group, as shown in Fig. 1b. When it is in the range of 31° to 50°, it is classified as moderate group, as shown in Fig. 1c. When it is above 50, it is classified as severe group, as shown in Fig. 1d. A total of 3 groups.

Condyle morphology and location

According to the standard reported by Luo [8], based on T2WI-fs, the morphology of the condyle was classified as oval, flat or beak-shaped. And according to whether condyle bone injury, divided into normal, joint surface erosion(The surface of bone is blurred and rough) and bone injury(Local bone defect). According to conical beam CT’s judgment method of condyle position [9], joint space was quantitatively measured in MR Image, and condyle position was divided into three groups: anterior position, central position and posterior position, as shown in Table 1; Fig. 2.

Table 1.

Measurement of joint space and judgment of condyle position

Parameter	Definition
Anterior joint space (AJS)	Draw the highest point of the articular fossa, the line tangent to the condylar front, and the distance from the tangent point to the corresponding articular fossa on the vertical line of the tangent point, as shown in Fig. 2a.
Posterior joint space (PJS)	Draw the highest point of the articular fossa, the line tangent to the posterior margin of the condyle, and the distance from the tangent point to the corresponding articular fossa on the vertical line of the tangent point, as shown in Fig. 2a.
Anterior position	(PJS-AJS)/(PJS + AJS) = R, if R>12%, defined as anterior position.
Central position	(PJS-AJS)/(PJS + AJS) = R, if -12%< R<12%, defined as central position.
Posterior position	(PJS-AJS)/(PJS + AJS) = R, if R<-12%, defined as posterior position.

Fig. 2.

(a) Schematic diagram of joint space measurement; (b) Anterior position, AJS = 3.18, PJS = 4.77, R = 20%; c. Central position, AJS = 3.4, PJS = 3.62, R = 3.1%; d. posterior position, AJS = 4.57, PJS = 2.79, R=-24.2%

Joint effusion grading

Referring to the standards reported by Koh [10] and Larheim [11], etc., joint effusion was classified into 4 levels based on the closed-mouth oblique sagittal T2WI-fs: (1) No effusion: no high-signal effusion shadow was observed in joint space; (2) Mild effusion: point-like or linear high signal shadow; (3) Moderate effusion: between mild and marked, presenting as small triangular or strip-shaped high-signal aggregations; (4) Marked effusion: Liquid pools and high signal levels. As shown in Fig. 3.

Fig. 3.

(a) No effusion. (b) The arrow indicates mild effusion. (c) The arrow indicates moderate effusion. d. The arrow indicates marked effusion

Data analysis

SPSS Statistics X9.1 software was used for statistical analysis of the eligible data. Mean age was described by mean ± standard deviation and independent sample t test was used. The sex distribution, disc shape, degree of ADD, condyle bone shape, bone change, condyle position and joint effusion were examined by Chi-square test. Single factor binary logistic regression analysis was used to preliminarily screen out the variables that might be meaningful. The preliminary screening boundary value was P<0.1. If the difference in factors was statistically significant (P<0.1), the indicator was included in multivariate binary logistic regression analysis to further validate the effect and estimate the effect size. Receiver operating curve (ROC) was used to evaluate the prediction accuracy of the model. P<0.05 was considered statistically significant.

Results

Patient characteristics

In this study, 154 patients with ADD were included, and unilateral or bilateral TMJ images were collected, totaling 279 cases. They were divided into ADDWR(n = 146) and ADDWoR(n = 133) groups according to whether the joint disc was reset. There were no significant differences in gender distribution and mean age between the two groups. As is shown in Table 2.

Table 2.

Comparison of MR image features of ADDWR and addwor

	ADDWR n = 146	ADDWoR n = 133	t/χ2	p
Sex
Male	19(13%)	12(9%)	1.123	0.289
Female	127(87%)	121(91%)
Age	27.7 ± 9.7	26.6 ± 8.8	1.479	0.225
Articular disc morphology
Biconcave	53(36.3%)	50(37.6%)	49.557	<0.0001*
Average thickness	49(33.6%)	7(5.3%)
Convex	2(1.4%)	9(6.8%)
Fold-curved	30(20.5%)	62(46.6%)
Stretching	12(8.2%)	5(3.7%)
Degree of ADD
Mild	92(63%)	5 (3.7%)	174.282	<0.0001*
Moderate	38(26%)	9(6.8%)
Severe	16(11%)	119(89.5%)
Condylar morphology
Oval	92(63%)	48(36.1%)	22.424	<0.0001*
Flat	7(4.8%)	20(15%)
Beak	47(32.2%)	65(48.9%)
Condylar bone
Normal	51(34.9%)	25(18.8%)	9.148	0.01*
Surface erosion	69(47.3%)	79(59.4%)
Bone defect	26(17.8%)	29(21.8%)
Condylar position
Anterior position	21(14.4%)	8(6%)	14.854	0.001*
Central position	87(59.6%)	63(47.4%)
Posterior position	38(26%)	62(46.6%)
Joint effusion
None	31(21.2%)	8(6%)	61.733	<0.0001*
Mild	91(62.3%)	44(33.1%)
Moderate	21(14.4%)	60(45.1%)
Marked	3(2.1%)	21(15.8%)

As shown in Table 2, the differences in MR Image features between the two groups were compared. The articular disc morphology was divided into 5 types, of which more were double-concave in ADDWR and more were fold-curved in ADDWoR, and there was a statistical difference between the two groups(p<0.0001). The degree of ADD was divided into mild, moderate and severe. The proportion of mild(63%) in ADDWR was the highest, while the proportion of severe(89.5%) in ADDWoR was the highest, and there was a statistical difference between the two(p<0.0001). Condylar morphology was divided into 3 types, oval shape(63%) was more common in ADDWR, and beak shape(48.9%) was more common in ADDWoR, and there was a statistical difference between the two(p<0.0001). For the changes of condylar bone, surface absorption accounted for the highest proportion in both groups, but there was a statistical difference between them(p<0.01). The condylar position was more common in the central position between the two groups, but the difference between the two groups was statistically significant(p<0.001). Joint effusion was divided into 4 grades according to the amount of effusion, and the difference between the two groups was statistically significant(p<0.0001).

Univariate and multivariate logistic regression analysis of ADDWoR image features was shown in Table 3. The results showed that fold-curved articular disc, degree of ADD, condylar bone change, beak condyle, central condyle and effusion grades were all related factors in the diagnosis of ADDWoR. In addition, fold-curved articular disc, the degree of ADD and the grade of joint effusion were independent factors for the diagnosis of ADDWoR. ROC analysis for the diagnosis of ADDWoR was presented in Fig. 4. The area under ROC curve (AUC) for ADD-degree and joint effusion analysis for the prediction of ADDWoR was 0.909 (95% confidence interval (CI) of AUC = 0.8871–0.947) and 0.745 (95% confidence interval (CI) of AUC = 0.687–0.804), respectively.

Table 3.

Univariate and multivariate logistic regression analysis for addwor

	Univariate analysis OR(95%CI)	p	Multivariate analysis OR(95%CI)	p
Fold-curved joint disc	3.377(1.994–5.717)	<0.0001*	2.485(1.037–5.954)	0.041*
Degree of ADD	15.198(8.644–26.72)	<0.0001*	15.103(7.891–28.906)	<0.0001*
Condylar bone	1.554(1.092–2.212)	0.014*	1.516(0.796–2.889)	0.206
Beak condyle	2.013(1.239–3.273)	0.005*	0.727(0.319–1.656)	0.448
Central position	0.61(0.38–0.981)	0.041*	1.713(0.738–3.975)	0.21
Joint effusion	3.832(2.596–5.657)	<0.0001*	3.033(1.771–5.194)	<0.0001*

Fig. 4.

ROC prediction curve of ADD-degree and joint effusion for ADDWoR

Disscusion

The main findings of this study are as follows: (1)The degree of ADD, fold-curved disc, condylar bone change, beak condyle, condylar central position, and the grades of joint effusion, are all factors affecting ADDWoR. (2) The degree of ADD, fold-curved disc and the degree of joint effusion, are independent risk factors affecting ADDWoR. (3) The ROC for predicting the ADDWoR showed that the degree of ADD had a high predictive value, while the joint effusion had a average predictive value.

Whether the TMJ disc is displaced and the subtype of displacement are of great significance for clinical treatment decisions, and ADD is the most common type [12]. The joint disc is fibrous tissue and the normal disc is biconcave. With the opening and closing movements of the mouth, the thickness of the disc changes adaptively to balance the pressure in the joint cavity [13]. Abnormal joint movements, such as bad habits like teeth grinding and unilateral chewing, as well as asynchronous movements of the discor-condyle structure, can lead to degeneration of the collagen fibers in the intervertebral discs, disrupt the pressure balance within the joint cavity, and cause displacement of the intervertebral discs [14, 15]. Past studies have categorised disc morphology as diverse [16, 17]. In our study, disc morphology was classified as biconcave, average thickness, convex, fold-curved, and stretching. Articular disc deformities accounted for a high percentage of 63.1%. The morphology of the articular disc has significance on whether it is reset or not. During open-closed movements, the articular disc is subjected to compressive stress, tensile stress and shear stress [18]. In normal joints, the stress distribution is uniform. The sliding of the condyle along the articular disc may generate abnormal stress, causing damage to the articular disc and leading to the aggravation of TMD [19]. Continuous stress on the articular disc may cause it to deform, increase the friction of the temporomandibular joint, and prevent the articular disc from repositioning [20]. The middle band of the articular disc is the thinnest and bears the least resistance, while the anterior and posterior bands are thicker and bear more force [21]. Due to the constant opening and closing motion, the articular disc is subjected to condylar extrusion, impact or other factors, with the middle band as the folding point, the anterior and posterior bands are reflexed and bent, or continue to be shortened, folded or even change into a rounded-convex shape, and gradually develop into an irreducible state [22]. In this study, biconcave type was the most common in the ADDWR group (36.3%), and fold-curved type was the most common in the ADDWoR group (46.6%). Moreover, we used binary logistic regression to screen independent risk factors from MRI features as predictors of whether the articular disc was retrievable. The result shows that the fold-curved articular disc is an independent risk factor for ADDWoR(OR = 2.485, p = 0.041).

Articular disc deformation varies with the degree of ADD [23]. As the degree of displacement becomes more severe, the length of the disc becomes smaller, the thickness increases, and the length/thickness ratio decreases, and as the disc is displaced further forwards, both anterior displacement and deformation become severe at the same time [24]. In the process of ADDWoR, the articular disc will adaptively change in the early stage, and as the disease progresses the degree of anterior displacement of the articular disc increases, the internal structure is disorganized, the deformation of the articular disc is aggravated, and the degree of displacement is subsequently aggravated by interactions [25], which in the long term affects the effect of late clinical treatment. In our study, there was a statistically significant difference in the degree of displacement between the ADDWR group and the ADDWoR group(p<0.0001), with severe ADD accounting for the most ADDWoR (89.5%) and mild ADD accounting for the most ADDWR (63%). The degree of ADD is an independent risk factor for ADDWoR (OR = 15.103,p<0.0001), and it has a high predictive value for ADDWoR(Fig. 4). This indicates that the more severe the degree of ADD is, the greater the possibility that ADD is irreversible.

The temporomandibular joint(TMJ) is a unique double linkage joint in the craniofacial region, with the condyle constituting a pivotal anatomical structure that exhibits significant variability [26]. It is intricately linked and interacts with the articular disc. During movement, the condyle is subject to vertical forces that can be mitigated by the articular disc. However, when the articular disc is deformed, the cushioning effect is compromised, resulting in an imbalance of forces that alters the condyle’s position and morphology [27]. The present study revealed statistically significant disparities in condylar position, morphology, and bone alterations between the ADDWR and ADDWoR groups. In our study, oval condyle accounted for a higher proportion (63%) in the ADDWR group, while beak condyle accounted for a higher proportion (48.9%) in the ADDWoR group, which was consistent with previous studies [28]. It is noteworthy that within an individual, the condyles on both sides of the joint may exhibit divergent morphologies. This variation can be attributed to disparate occlusal styles, which give rise to disparate condylar forces. Prolonged overloading has been demonstrated to disrupt the dynamic equilibrium between condylar osteoblasts and osteoclasts, culminating in bone changes or osteomyelitis [25]. The univariate analysis of this study indicated that the beak condyle, central condyle, or bone changes are all influencing factors for ADDWoR(Table 3). The change of condyle position after ADD is the focus of scholars’ research, and the measurement of joint space is the most intuitive manifestation to understand the condyle position [29]. Cone-beam computed tomography(CBCT) provides high-resolution three-dimensional images, allowing for the assessment and quantification of facial bone tissue in the real dimension without significant magnification or deformation. It offers high anatomical accuracy and is the preferred imaging method for evaluating the bone condition of the TMJ [30]. Most of our cases only have MRI images. Our research referred to the CBCT standard and measured the joint space on MRI [9]. Previous scholars have believed that condyle position is significantly related to ADD. If ADD is mild, the condyle will shift backward, and the condyle will return to the central position as the dislocation of the articular disc becomes more severe [31]. But our study found that condylar central position had an effect on ADDWoR(OR = 0.61, p = 0.041), but the degree of ADD was an independent factor for ADDWoR(OR = 15.103, p<0.0001), which seemed to be contradictory to previous studies. However, the relationship between condyle position change and ADDWoR remains unclear. Measurement of the joint space on MRI instead of CBCT, and at a two-dimensional level instead of a three-dimensional level, may result in accuracy biases [4]. Anyway, this suggests that the measurement and evaluation of condylar bone changes will help to prevent the disease from progressing to a more severe stage.

Joint effusion is a dialysate of plasma that provides nutrients to the joints, helps remove metabolites, lubricates the TMJ, and influences pressure distribution [32]. Quantifying joint fluid accumulation on MRI is difficult, and there are no standardized criteria. Some are based on the number of successive layers of liquid signals involved, and some are based on the length, width or volume of the liquid image [33, 34]. We broadly categorize fluid signals into no, mild, moderate, and marked fluid accumulations based on previous experience. A small amount of fluid in the joint cavity may be necessary for physiologic lubrication of the joint. Increased effusion has been observed to be associated with TMJ dysfunction, particularly in cases where dysfunction affects the disc and condylar structures [35]. Many studies in the past have focused on the relationship between joint effusion and pain, but no definitive conclusions have been reached. Most studies have quantified effusion on a two-dimensional level, and the existence of an association between effusion and pain remains controversial. Pain is usually associated with the onset of ADDWoR, and temporomandibular joint effusions are most common in fold-curved discs [36, 37]. Deformed articular discs can affect the metabolic circulation of joint fluid, leading to increased fluid accumulation and joint pain [36]. TMJ and condylar changes due to osteoarthritis are also often associated with disc deformity, which is associated with joint effusion [38]. Joint disorders have been shown to induce joint damage, which in turn prompts the release of pro-inflammatory cytokines, including tumor necrosis factor-alpha, nitric oxide, bradykinin, interleukins and leukotrienes [39]. It has been found that these inflammatory cytokines promote joint destruction and produce inflammatory exudates around the joints. In turn, the exudate leads to increased inflammation, which in turn affects further deterioration of the structural relationships of the joint [40]. No or mild effusion accounted for the majority of ADDWoR, while moderate or marked effusion were more common in ADDWoR (Table 2). Increased joint effusion, an independent risk factor for ADDWoR, had an area under the curve of 0.745 for the predictive model for ADDWoR, with an average predictive effect. This means that the higher the grade of joint effusion, the greater the possibility that ADD will be irreversible.

It is worth noting that our research work still has some limitations. There are only two hospitals in the area where our hospital is located that have a specialist TMJ clinic, and we are one of them. Most patients with TMJ problems can only go to oral, maxillofacial surgery or rehabilitation departments and most of the patients we see are referred from other hospitals where they may have more serious joint problems. This carries the risk of bias in case selection. It is also possible that some patients have been previously treated with poor outcomes, which is likely to affect the changes in joint fluid and complicate the relationship between joint fluid and ADDWoR, thereby affecting diagnostic efficacy. Patients who have received clinical treatments such as manipulative repositioning and acupuncture in Chinese medicine will have a reduction in joint pain and popping, and a reduction in effusion can be seen on MRI at review, but most patients with reduced symptoms refuse to be reviewed again, resulting in a lack of imaging data in our study, which is one of the limitations of our study and one that needs to be explored in further studies in the future. As in previous studies, we used two-dimensional levels to semi-quantify the effusion. The grades of effusion was found to have some diagnostic efficacy in predicting ADDWoR. Further refinement of the quantification method will be undertaken in the future.

In conclusion, MRI has an excellent assessment ability for TMJ. The influencing factors of ADDWoR can be effectively evaluated through semi-quantitative analysis of the MRI signs of the disc-condylar complex. For the diagnosis of ADDWoR, the degree of ADD and joint effusion grades have certain predictive value, which is helpful for evaluating the condition and providing a basis for clinical diagnosis and treatment.

Abbreviations

MRI

Magnetic resonance imaging

ADD

Anterior disc displacement

ADDWR

Anterior disc displacement with reduction

ADDWoR

Anterior disc displacement without reduction

ROC

Receiver operating curves

TMD

Temporomandibular joint disorder

TMJ

Temporomandibular joint

Author contributions

ML and XHC designed and carried out the study, collected the data, wrote and approved the manuscript. KL and HLH designed the study, collected and analyzed data, wrote and approved the manuscript. YT and DZ designed and carried out the study, analyzed the data, cowrote and approved the manuscript. TH and BBD supervised, designed and performed study, analyzed data, wrote and approved the paper. All authors read and approved the final manuscript.

Funding

None.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The retrospective study was approved by the ethics review committee of NO.1 Orthopedics Hospital of Chengdu, and was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2013, and the need to obtain informed consent was waived. All data were anonymized.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.