Correlation of Psoas Muscle Index with Fragility Vertebral Fracture: A Retrospective Cross-Sectional Study of Middle-Aged and Elderly Women

Objective To investigate the correlation of psoas muscle index (PMI) with fragility vertebral fracture. Methods A total of 184 middle-aged and elderly women were included in the study. We measured the bilateral psoas muscle area on the picture archiving and communication system (PACS) from computed tomography images and calculated PMI. We observed lateral radiographs of the thoracolumbar spine and assessed vertebral fractures using the Genant semiquantitative method. The T-score, bone mineral density (BMD) of the lumbar (L)1-4, femoral neck, and trochanter were measured by dual-energyX-ray absorptiometry (DXA). The data was collected and then statistically analyzed. Results The PMI of the nonosteoporosis group was higher than that of the osteoporosis group (P value = 0.006). Height in the nonosteoporosis group was higher than that in the osteoporosis group (P value = 0.013). Weight, body mass index (BMI), left psoas muscle area, BMD of the L1-4, femoral neck, femoral trochanter, and T-score in the nonosteoporosis group were higher than those in the osteoporosis group (P value <0.001). The right psoas muscle area in the nonosteoporosis group was higher than that in the osteoporosis group (P value = 0.008). The incidence of combined thoracolumbar fracture was significantly higher in the osteoporosis group than that in the nonosteoporosis group (P value <0.001). For nonosteoporosis subjects, the PMI of the vertebral fracture group was lower than that of the nonvertebral fracture group (P value = 0.034). Conclusions A decrease in height, weight, BMI, bilateral psoas muscle area, and PMI is associated with osteoporosis. Combined thoracolumbar fractures are more common in osteoporosis. Sarcopenia may be an independent risk factor for nonosteoporotic vertebral fractures.


Introduction
Musculoskeletal disorders have become signifcant public health problems as the population ages. Irwin Rosenberg frst used sarcopenia to defne sarcopenia in 1989, and since then, muscle function has become essential in describing sarcopenia [1]. Sarcopenia is a progressive decrease in body muscle mass and/or a decrease in muscle strength or muscle physiological function associated with age. Te relationship between sarcopenia and osteoporosis has increasingly become a hotspot, and signifcant progress has been made in clinical and basic research [2][3][4]. Vertebral fracture is considered a sign of osteoporosis and is the most common fragility fracture. It is of great signifcance to study the relationship between paraspinal muscles and thoracolumbar osteoporotic fractures and to further early intervention treatment for diagnosing and treating thoracolumbar vertebral compression fractures.
Psoas muscle index (PMI) is used to evaluate skeletal muscle mass through computed tomography (CT). PMI is positively correlated with total skeletal muscle volume and can be used to diagnose sarcopenia [5,6]. A recent study showed that PMI measured by CT in patients with spinal degeneration was positively correlated with bone mineral density (BMD). It was considered a useful tool for assessing osteoporosis and fracture risk [7]. Bilinc Dogruoz Karatekin suggested that PMI may be associated with hip fractures in patients with osteoporosis. Selective psoas strengthening exercises may be benefcial for hip fracture prevention and posthip fracture rehabilitation program [8]. Is sarcopenia related to fragility vertebral fracture? Tis retrospective cross-sectional study explores the association between sarcopenia and fragility vertebral fracture using PMI.

Subject. 214 women who visited the Sixth Afliated
Hospital of Xinjiang Medical University from January 2021 to December 2021 were enrolled. Inclusion criteria: women over 50 years of age with clinical syndromes of back pain and limited spinal mobility, with aninterval among CT, spinal radiographs, and BMD examination within 3 months. Based on imaging and medical history, we excluded 12 cases with vertebral fractures caused by a high-energy trauma or degenerative disease based on the qualitative criteria used in clinical practice for diferential diagnosis of vertebral fractures; 2 cases who had taken calcitriol or alendronate sodium tablets for 3 months; 1 case who had hypothyroidism; 3 cases of severe lumbar degeneration; 4 cases of pathological vertebral fracture; 1 case of septic spondylitis; 5 cases of spinal tuberculosis; 2 cases of brucella spondylitis. Finally, 184 women aged 55-90 (70.5 ± 7.4) years were included in the study.
We collected clinical data from the subjects including age (y), menopausal age (y), height (m), weight (kg), and history of fractures. Weight divided by the square of height is the body mass index (BMI) (kg/m 2 ) calculation formula.

Measurement of Bone Mineral
Density. BMD of the lumbar spine and one side of the hip joint was measured by dual-energyX-ray absorptiometry (DXA) (GE, Lunar Prodigy). Measurement process: the subjects lay supine in the middle of the examination table with their hands fat on both sides of their bodies. Te posterior-anterior program scanned the spine and hip joints. After scanning, the software automatically generates data and measurement reports. Te collected data included BMD of lumber (L)1-4, femoral neck, femoral trochanter, and T-scores. Te L1-4 vertebrae, femoral neck, and femoral trochanter were used as regions of interest (ROI), and the lowest T-score of the 6 ROIs was used to diagnose osteoporosis. According to the DXA osteoporosis diagnostic criteria recommended by the academic organization of osteoporosis in China [9], participants were defned as osteoporosis (T-score ≤−2.5 standard deviations (SD)), osteopenia (−2.5 SD < T-score ≤ −1 SD), and normal bone mass (T-score > −1 SD).

Evaluation of Vertebral Fracture and Psoas Muscle Index.
All subjects underwent lateral radiographs of the thoracic (T) 4-L5. One radiologist read radiographs and determined vertebral fractures according to the Genant semiquantitative method [10]. Te criterion for vertebral fracture was a reduction of more than 20% in the anterior, middle, or posterior vertebral height. Te site of the fracture was recorded.
T12 to sacral (S)1 vertebral body of subjects were scanned by a 128-slicedual-source CT scanner (Siemens, Germany). CT scanning parameters are as follows: tube voltage 110 kV, tube current 100 mA, and layer thickness 1 mm. One radiologist who was blinded to BMDof the subjects selected CT axial images of the L3 vertebrae by picture archiving and communication system (PACS) (Heart Shadow International), and then manually delineated the border of the bilateral psoas muscle. Te software automatically calculated the bilateral psoas muscle area, as shown in Figures 1(a) and 1(b). Total psoas muscle area divided by the square of height is the PMI (mm 2 /m 2 ) calculation formula.

Statistical Analysis.
We used the Statistical Product and Service Solutions 19.0 software package for statistical analysis. Measurement data were expressed as the mean-± standard deviation. Enumeration data were expressed as numbers (percentages). A Kolmogorov-Smirnov test was used to test whether the data conformed to a normal distribution. Te homogeneity of variance for the measurement data was assessed using a Levene test. Measurement data of the two groups were compared using an independent sample t-test. A comparison of sample rates between the two groups was performed using a chi-square test or Fisher's exact test. P value < 0.05 was considered statistically signifcant.
All subjects were postmenopausal women. Height in the nonosteoporosis group was higher than that in the osteoporosis group (P value � 0.013). Weight, BMI, left psoas muscle area, BMD of the L1-4 vertebral body, femoral neck, femoral trochanter, and T-score in the nonosteoporosis group were higher than those in the osteoporosis group (P value <0.001). Te right psoas muscle area in the nonosteoporosis group was higher than that in the osteoporosis group (P value � 0.008). Te PMI in the nonosteoporosis group was higher than in the osteoporosis group (P value � 0.006). Tere was no signifcant diference in age, age of menopause, or the incidence of vertebral and other fractures (humerus fracture, forearm fracture, femur fracture, and patella fracture) between the two groups (P value � 0.355, P value � 0.185, P value � 0.117, P value � 0.484, P value � 1, P value � 0.674, and P value � 0.233, respectively), as shown in Table 1.

Correlation between Psoas Muscle Index and Fragility
Vertebral Fracture. Of all participants, 107 cases had vertebral fractures including 57 osteoporotic and 50 nonosteoporotic. For the nonosteoporotic vertebral fracture group, 37 had osteopenia and 13 had normal bone mass. Tere was no signifcant diference between the PMI of the vertebral fracture group and that of the nonvertebral fracture group (t value � −1.384 and PP value � 0.168), as shown in Figure 2(a), and there was no signifcant difference between the PMI of the osteoporotic vertebral fracture group and that of the nonosteoporotic vertebral fracture group (t value �−0.684 and P value � 0.495), as shown in Figure 2 For subjects with osteoporosis, there was no signifcant diference between the PMI of the vertebral fracture group and that of the nonvertebral fracture group (t value � 0.846 and P value � 0.4), as shown in Figure 2(c). For subjects with nonosteoporosis, the PMI of the vertebral fracture group was lower than that of the nonvertebral fracture group, and the diference was statistically signifcant (t value � −2.155 and P value � 0.034), as shown in Figure 2(d).
We divided the subjects into three categories according to the fracture site. Te result showed that the incidence of combined thoracolumbar fracture was signifcantly higher in the osteoporosis group than that in the nonosteoporosis group (chi-square value � 12.331 and P value < 0.001), as shown in Table 2.

Discussion
Sarcopenia and osteoporosis are age-related declines in the quantity and quality of muscles and bones. Yeung et al. demonstrated that sarcopenia had positive correlation with fall and fractures in the elderly through a meta-analysis [11].  Zanchetta et al. found that sarcopenia was associated with increased fall risk, osteoporosis, and vertebral fractures in postmenopausal women [12]. Te participants in this study were middle-aged and elderlywomen. Unlike in men, BMD in women is susceptible to estrogen levels. Compared with men of the same age, postmenopausal women are more likely to develop osteoporosis and have a higher risk of fragility fractures due to a sharp drop in estrogen levels [13,14]. CT is the preferred method for quantitative assessment of sarcopenia. Skeletal muscle index (SMI) and PMI are the more commonly used indicators in research. SMI was calculated as the total skeletal muscle area at the level of the L3 vertebrae divided by the square of the height. Te calculation method of PMI is similar to that of SMI. Te calculation of PMI uses the psoas muscle area rather than the total skeletal muscle area. Our study used PMI as an assessment tool for sarcopenia, not only because of its ease of calculation but also to the anatomical function of the psoas muscle. As we all know, the psoas muscle adjacent to both sides of the lumbar spine is diferent from other trunk muscles, and it plays a crucial role in maintaining the upright posture of the human body. Te psoas muscle connects the trunk and lower extremities to maintain standing and walking functions. Impairment of its function can seriously reduce the stability of body posture while afecting the human body's ability to fex the hip or maintain a standing posture. PMI is often used in research on Asian countries, especially Japan and South Korea [15][16][17]. A new PMI-based criterion for skeletal muscle mass has been established using data from healthy young Asian adults, which defnes a cut-of value for sarcopenia in Asian populations [18]. In addition to evaluating sarcopenia, PMI can predict long-term mortality in young men with chronic or acute liver failure [19]. Several other studies have shown that PMI can be applied to Marfan syndrome and the prognosis of diferent tumor surgeries [20,21].
Both sarcopenia and osteoporosis share the same pathophysiological basis and have a similar adverse efect on the health of older adults. Our results showed that the bilateral psoas area and PMI of postmenopausal women without osteoporosis were higher than those in the  International Journal of Endocrinology osteoporosis group. It indicates that reduction in skeletal muscle volume and mass as the main pathological feature of sarcopenia is associated with osteoporosis. Te mechanism is relatively complex, mainly including the infuence of the mechanical load of muscle contraction on the mechanical force of bone and the complex and precise endocrine regulation mechanism between muscle and bone [22,23]. Te mechano-regulatory system hypothesis states that muscle contraction directly provides mechanical bone stimulation, promoting osteogenesis [24]. Both muscles and bones have endocrine functions and are regulated by a variety of factors [1]. Several studies have reported the association between sarcopenia and fragility vertebral fracture [25][26][27]. A survey by Wang et al. showed that sarcopenia was an independent risk predictor of refracture in patients with an osteoporotic vertebral fracture [25]. Tetsuro et al. showed that sarcopenia and decreased calf muscle mass were more common in patients with acute osteoporotic vertebral fractures than in patients without osteoporotic vertebral fractures [27]. Our study found that the PMI of the vertebral fracture group was lower than that of the nonvertebral fracture group in the subjects with nonosteoporosis (P value � 0.034). In contrast, the diference in PMI between them in the subjects with osteoporosis was not signifcant. Tis study suggests that sarcopenia may be a risk factor for nonosteoporotic vertebral fractures. Terefore, delaying sarcopenia progression and improving skeletal muscle function can reduce the risk of fragility fractures in people without osteoporosis, and these measures may not be benefcial in reducing the risk of fragility fractures in people with osteoporosis.
Our study showed that the height, weight, and BMI of the nonosteoporosis group were higher than those of the osteoporosis group. It suggests that the protective efect of high body weight is probably due to the more signifcant loading on bones. Other studies suggest that body weight and BMI may be protective factors for BMD [28,29]. Te mechanism may be that decreased sex hormone-binding globulin and increased free sex hormones positively afect BMD in obese patients [30]. In addition, we found that the incidence of osteoporotic and nonosteoporotic fractures was 64% and 52.6%, respectively, and there was no statistical diference (P value � 0.117). Several studies have shown that the incidence of fragility fractures with osteopenia and normal bone mass is not low [31][32][33]. A Korean study showed that more than half of the fragility fractures occurred in women with osteopenia or normal bone mass during the 10-yearfollow-up period [31]. Pasco followed up with 616 postmenopausal women aged 60-94 years. Te results showed that 26.9% of fragility fractures were in the osteoporotic group, while 73.1% of fragilities occurred in women without osteoporosis (56.5% in women with osteopenia, 16.6% in women with normal bone mass) [32].
Our study found that combined thoracolumbar fractures were more common in the osteoporosis group (P value < 0.001). Te progression of osteoporosis is associated with a preferential loss of horizontal trabeculae, the efect of which is reduced ability of the vertebral body to withstand axial compressive forces. Te inclination of this bony structure results in vertebral compression fractures, most of which occur in the thoracic and thoracolumbar spine [34].
Te study also has some limitations. Te study was aimed at middle-aged and elderly women, and the sample was small. Te proportion of subjects with osteoporosis, osteopenia, and normal bone mass was not balanced. We combined osteopenia and normal bone mass into the nonosteoporosis group. Terefore, this study cannot more strictly interpret the association of PMI alone with normal bone mass or osteopenia vertebral fracture. Tis study did not make statistics on the number, location, severity, and time of vertebral fractures. Some subjects had mild lumbar degeneration or spondylolisthesis, which may afect the accuracy of BMD.

Conclusions
A decrease in height, weight, BMI, bilateral psoas muscle area, and PMI is associated with osteoporosis. Combined thoracolumbar fractures are more common in osteoporosis. Sarcopenia may be an independent risk factor for nonosteoporotic vertebral fractures.

Abbreviations
PMI: Psoas muscle index SMI: Skeletal muscle index CT: Computed tomography DXA: Dual-energy X-ray absorptiometry SD: Standard deviation BMI: Body mass index BMD: Bone mineral density T: Toracic L: Lumbar S: Sacral ROI: Region of interest.

Data Availability
Te datasets cannot be made publicly available, and restrictions are applied to the availability of these data. Yihui Zhang should be contacted if someone wants to request the data from this study.

Disclosure
Tis manuscript was submitted as a preprint in the link "https://www.researchsquare.com/article/rs-1528619/v1." Tis manuscript is not diferent from the preprint, and it was not revised before submission to the journal [35].

Conflicts of Interest
Te authors declare that they have no conficts of interest.
International Journal of Endocrinology 5