Comparative Analysis of Preoperative and Postoperative Muscle Around Hip Joint by Computed Tomography Between Intertrochanteric Fracture and Femur Neck Fracture

Background: This study assessed the changes in hip muscles by comparing the preoperative and postoperative CT scan results between patients with intertrochanteric versus femoral neck fractures. Methods: 48 patients who received surgical treatment for intertrochanteric or femoral neck fractures from February 2013 to February 2019 and underwent pelvic computed tomography(CT) preoperative and postoperatively aged 65 and older with a minimum follow-up of 1 year were included. The subjects were divided into two groups: 26 patients with intertrochanteric fracture and 22 patients with femoral neck fracture. We measured the cross-sectional area(CSA) and attenuation of the gluteus medius(G.med), gluteus minimus(G.min), iliopsoas(IP), and rectus femoris(RF) on the contralateral side. Patient basic data were collected from medial records including sex, age, height, weight, BMI, BMD, Harris hip score (HHS), and length of follow-up until the nal visit.

The number of hip fracture patients continues to rise with an aging population. Surgical intervention is also increasingly applied in the treatment of hip fracture for improvement of the quality of life, pain relief, and prevention of complications associated with prolonged bed con nement. 1,2) Decline in muscle mass, lower muscular strength, and sarcopenia de ned as low physical performance related with aging have emerged as a serious problem. 3,4) Sarcopenia is profoundly related with poor health status and high mortality and elderly patients with sarcopenia are not only at high risk of hip fracture [5][6][7] , but also may have aggravated health conditions, leading to high mortality. 8,9) Intertrochanteric and femoral neck fractures are the most common hip fractures in elderly patients. These fractures are similar in fracture site, but different in location and direction of applied force. Moreover, in regards to the surgical treatment of choice, intertrochanteric fracture is usually treated with reduction and internal xation such as intramedullary nailing, while femoral neck fracture is mainly treated with hip arthroplasty.
Previous studies on sarcopenia primarily used dual-energy X-ray absorptiometry (DEXA) in the analysis. 10,11) Of these studies, no study has been performed comparing intertrochanteric and femoral neck fractures. In the comparison of the fractures, DEXA can provide rough estimation based on the amount of radiation absorbed, but the analysis on precise muscle mass and density of different muscles is impossible. 12) Furthermore, since injury mechanism at the time of fracture, the type and direction of the muscles involved and surgical options are different between intertrochanteric and femoral neck fractures, the type of muscles hypertrophied and atrophied are predicted to be different. However, only a few studies have been conducted to compare the difference. The measurement of muscle mass and attenuation on the axial computed tomography(CT) scan has been reported to a good indicator for sarcopenia in previous literature. 13) Therefore, this study aimed to perform a comparative analysis on pre-and postoperative muscle mass between the intertrochanteric fracture and femoral neck fracture groups by accurately measuring muscle mass around the hip joint using CT scan.

Subjects
This study was performed after gaining institutional review board(IRB) approval (DAUHIRB-20-124) from investigators' institution. This study reviewed 97 patients who had postoperative pelvic CT scan among 700 patients who underwent surgery due to intertrochanteric or femoral neck fractures from February 2013 to February 2019. Of these, 42 patients who had bilateral hip surgery, ambulatory disability, neuromuscular disorder, or hemiparesis due to cerebrovascular disease were excluded from the analysis and a total of 48 patients were retrospectively reviewed among 90 patients aged 65 and older with a minimum follow-up of 1 year. The subjects were divided into two groups: 26 patients in the intertrochanteric fracture group and 22 patients in the femoral neck fracture group. (Fig. 1)

Surgical Methods
All operations were performed by a single surgeon. In 26 patients with intertrochanteric fracture, adequate reduction obtained on a fracture table was preoperatively con rmed with C-arm radiography, and intramedullary nail insertion was performed using the PFNA-II nail system (Proximal femoral nail antirotation-II, Asian version, DePuy Synthes, Oberdorf, Switzerland). In 22 patients with femoral neck fracture, bipolar hip arthroplasty was conducted using a Bencox® hip stem (Corentec, Seoul, Korea) in a lateral position with modi ed Gibson approach.

Clinical and Radiological Analysis
The basic data of the patients were retrospectively reviewed based on medical records including sex, age, height, weight at the time of surgery and at the nal follow-up, body mass index (BMI), bone mineral density (BMD), Harris hip score (HHS) related with function, and length of follow-up until the last visit.
Axial CT images of the pelvis just below the sacroiliac (SI) joint and at the level of the lesser trochanter (LT) were used for radiographic analysis of the gluteus medius(G.med), gluteus minimus(G.min), rectus femoris(RF), and iliopsoas(IP) muscles. The Cross-sectional area(CSA) and density of the muscles were measured by drawing a line connecting the outer margin of the muscles on axial CT images using the PACS system (INFINITT PACS, version 3.0; INFINITT Healthcare Co., Seoul, Korea), expressed in mm 2 and Houns eld units (HU). The CSAs of the gluteus medius, gluteus minimus, and iliopsoas were measured just below level of the SI joint and the CSA of the rectus femoris was measured at the level of the LT. The largest cross-sectional areas around those levels were measured (Fig. 2). Assuming that measurement errors could occur due to hematoma and swelling on the affected side of the pelvis, muscle CSA and density were preoperatively measured on the contralateral hip. The muscles were measured twice at a two-week interval by the same experienced surgeon, and the mean value was used.

Statistical Analysis
The CSA, cross-sectional area per weight (CSA/Wt, mm 2 /kg), and attenuation of each muscle were measured pre-and postoperatively, and the measured values were compared between the intertrochanteric fracture and femoral neck fracture groups. All statistical analyses were performed using SPSS software (version 25.0; IBM Co., Armonk, NY, USA). Chi-square test was used comparing the two groups on sex. Independent t-test was conducted to test the difference between the two groups on age, BMI, BMD, HHS related with function, and length of follow-up until the last visit; and on CSA, crosssectional area per weight (CSA/Wt, mm 2 /kg) and attenuation of each muscle. Paired t-test was done to test pre-and postoperative difference within a group. P-values of less than 0.05 were considered statistically signi cant. Correlations of the CSA and cross-sectional area per weight (CSA/Wt, mm 2 /kg) with HHS related with function were analyzed using Pearson's correlation tests. Statistical signi cance was considered at p < 0.05.

Subjects
Of all 26 patients in the intertrochanteric fracture group, 6 (23.1%) were males and 20 (76.9%) were females. Of all 22 patients in the femoral neck fracture group, 9 (40.9%) were males and 13 (59.1%) were females. In chi-square analyses, there was no signi cant difference between the two groups. The mean age was 79. 8    in the femoral neck fracture group. An independent t-test was performed to compare the means for two groups, but no statistically signi cant difference was exhibited (p > 0.05). (Table 1) 2. Cross-sectional area of muscles In the correlation analysis between postoperative cross-sectional area per weight (CSA/Wt, mm 2 /kg) and HHS related with function, Pearson's correlation coe cient was 0.18, -0.33, -0.21, and 0.27 in the gluteus medius, gluteus minimus, iliopsoas, and rectus femoris, respectively. All p-values were larger than 0.05, showing no signi cant correlation. (Table 4) 3. Attenuation of muscles The attenuation (Houns eld unit, HU) of the gluteus medius and gluteus minimus was 32.19 (6-56) and 23.65 (4-38) in the intertrochanteric fracture group and 30.59 (11-61) and 17.36  in the femoral neck fracture group preoperatively; and 23.54 (5-45) and 10.92  in the intertrochanteric fracture group and 21.18 (4-36) and 11.09    (Table 5) The pre-and postoperative attenuation of all 48 patients enrolled to the study was 31.46 (6-61) and 22.46 (4-45) in the gluteus medius; 20.77 (4-38) and 11.00  in the gluteus minimus; 46.77 (12-71) and 36.96  in the iliopsoas; and 45.08  and 35.85  in the rectus femoris, respectively. A signi cant reduction was shown in all muscles (p < 0.001). (Table 6) Discussion Skeletal muscle mass and strength are inversely proportional to increasing age. The decline of skeletal muscle mass and strength with age are accelerated after 65 years of age and results in various social problems such as physical impairment, reduced quality of life, and an increase in mortality. 14) The socioeconomic burden of these problems has been raised as one of the serious concerns, and as a result, sarcopenia has recently gained much attention. 15,16) Sarcopenia can increase the risk of hip fracture, and cause dysphagia or voiding dysfunction due to muscle weakness. 17,18) These conditions are considered as indicators of frailty and the loss of independence. 19) Tatara 20) and Ellman et al. 21) have suggested muscle force as a crucial factor for proper growth and preservation of the bony skeleton, and Ford et al. 22) have reported the in uence of imbalance of muscles around hip joint on the hip joint. However, few previous studies have been done to evaluate muscles around hip joint by dividing into two groups of intertrochanteric and femoral neck fractures.
The conventional methods for measuring body composition include DEXA, bioelectrical impedance, CT, MRI, etc. Although DEXA and bioelectrical impedance analysis can be used in estimating the overall conditions of the skeleton, but those are not suitable for muscle-speci c analysis. MRI allows a precise analysis of muscle condition and mass, attenuation, fatty in ltration around skeletal muscle, but cannot be used in patients with metallic prostheses. On the contrary, Mitsiopoulos et al. 23) addressed that measurements from CT and the volumes of skeletal muscle and adipose tissue in cadavers were nearly the same. According to a report by Rasch et al. 24) , the use of CT scan allows simple circumscribing of large muscle bellies and minimizes measurement errors with easy identi cation of the bony landmarks of the pelvis.
Of all muscles around the hip joint, the gluteus medius serves a key role in abduction at hip joint, provides stability of the pelvis during a single-leg stance and exhibits Tredelenberg's sign in case of insu ciency. 25,26) Since the gluteus minimus aids the abduction similar to the gluteus medius in function, this muscle was included in the analysis. The iliopsoas muscle was also included because it serves as the prime muscle of hip exion and has been considered a single indicator of sarcopenia, 27,28) contributing to stability of the lumbar spine that may cause problems such as HNP in cases of tightness or spasm. 29) The rectus femoris was included in this study because this muscle crosses over both the hip and knee joint and maintains stability of the femur during walking and in case of weakness of rectus femoris, excessive knee extension may occur due to limitation of active knee extension.
In a recent study, Kim et al. 30) have reported that the weakness of the iliopsoas and spine extensors is the major risk factor of hip fracture. Paganini et al. 31) and Chilibeck et al. 32) have suggested that the strengthening of skeletal muscles through active exercise is effective in preventing hip fracture, increasing BMD and reducing the risk of falling. Gschwind et al. 33) have addressed that hip fracture can be prevented by exercise that improves muscle balance and strength in the older population. However, no studies have been performed to determine which speci c muscles need to be strengthened in rehabilitation for patients with intertrochanteric and femoral neck fractures. According to the results of this study, the CSA of hip abductors (gluteus medius and gluteus minimus) was signi cantly higher in the femoral neck fracture group, while the CSA of hip exors (iliopsoas and rectus femoris) was signi cantly greater in the intertrochanteric fracture group. These ndings are expected to be helpful in determining rehabilitation exercises appropriate for patients with different fracture types. No signi cant difference was shown in muscle attenuation between the two groups. However, a signi cant decrease in attenuation after surgery appears to be attributable to age-associated increase in muscle fat in ltration.
This study was limited in certain aspects. First, this comparative study was limited by the relatively small sample size. Second, the difference in the actual muscle function and power may present because assessment was done based on muscle CSA and attenuation, instead of accurately measuring the power of muscles at the hip joint in each patient. Finally, since preoperative ambulation ability and the overall function of the musculoskeletal system between the two groups were not fully controlled, bias may arise in the comparison of pre-and postoperative changes.

Conclusions
This study detected a signi cant difference in hip abductors and exors on axial CT images between the intertrochanteric fracture and femoral neck fracture groups. The difference seems to be attributed from the type and direction of external forces applied to fracture occurrence. These ndings demonstrate that the rehabilitation program for hip abductors and exors respectively will be helpful for rehabilitation of patients with these fractures. nancially supported by the Dong-A University research fund. Availability of data and materials The data generated and/or analysed during the current study are not publicly available but are available from the corresponding author on reasonable request. Ethics approval and consent to participate The current study was performed in accordance with institutional review board approval from Dong-A university hospital (DAUHIRB-   Figure 1