The size and density of muscle and subcutaneous adipose tissue are independently associated with the risk of mortality in older women with different hip fracture types

Abstract

Aims

To investigate whether the size and density of proximal femur muscle and surrounding subcutaneous adipose tissue (SAT) are associated with the risk of mortality in older women with different hip fractures (HF) types, as well as to assess how this association changes over time following the fracture.

Methods

CT images and data from 378 female patients with low-energy acute HF were retrospectively collected between January 2020 and July 2022, with an average follow-up of four years. The cross-sectional area and density of the gluteal muscle around the hip, thigh muscle, and adjacent SAT were measured. Cox proportional hazard models were used to assess mortality risk adjusted for covariates in different HF types.

Results

At the end of the follow-up period, 58 had died, and 320 had survived. The mean age at death for non-surviving patients (84.5 ± 7.9 years) was significantly higher than that of surviving patients (72.1 ± 10.6 years). In the femoral neck fracture (FNF) group, the size and density of the gluteus medius and minimus muscles (G.Med/MinM), as well as thigh muscle density, were negatively associated with mortality (P < 0.05). In the intertrochanteric fracture (ITF) group, the mid-thigh SAT (MSAT) area and density were negatively and positively correlated with mortality, respectively. Gluteal muscle size was inversely associated with mortality at one year and beyond following a HF. The MSAT area was inversely associated with mortality within 2 years after HF, while the MSAT density was positively correlated with mortality during the 1–2 years following the fracture.

Conclusions

In different HF groups, the size and density of proximal femur muscles and surrounding SAT demonstrated varying associations with post-fracture mortality in older women, independent of age, BMI, and clinical risk scores.

Introduction

Hip fractures (HF) are a significant cause of morbidity and mortality in older adults, both men and women, and represent a major public health issue worldwide, substantially increasing the socio-economic burden [1, 2]. For individuals, a hip fracture can be devastating, leading to a loss of independence, and fewer than one third of patients achieve a full recovery [2]. Approximately 32–59% of older patients with HF are considered frail [3], and these patients tend to have multiple comorbidities compared to younger patients [4]. HF are the most common type of fragility fractures that require immediate inpatient surgical treatment, followed by managed care and rehabilitation to minimize complications [5, 6]. Mortality rates in older patients with HF range from 11.5 to 36% within the first year [7], and this risk gradually increases over the following five years [8, 9]. Elderly patients with high Fracture Risk Assessment Tool (FRAX) scores who experience hip fractures are particularly susceptible to refractures following surgery [10]. To reduce mortality, prognostic prediction is essential, as it necessitates the identification of patients at high risk of death and the factors that contribute to this risk, enabling effective treatment decisions.

Both osteoporosis and sarcopenia are associated with aging, which increases the likelihood of falls among older adults and subsequently raises the risk of HF [11–13]. Osteoporosis is a chronic metabolic bone disease characterized by a decrease in bone mineral density (BMD) and the deterioration of bone microarchitecture [14]. Sarcopenia is a progressive systemic disease affecting skeletal muscle, characterized by the accelerated loss of muscle mass and function [15, 16]. The study found that low BMD and sarcopenia were linked to an increased risk of mortality in older patients with HF [12, 17].

Quantitative computed tomography utilizes 3D imaging to assess not only the BMD of the cortical and trabecular bone [18], but also the adjacent anatomical muscle and adipose tissue [19]. CT-based muscle and fat parameters show promise in predicting the risk of hip fractures and mortality [20–22]. However, only a limited number of cross-sectional studies have investigated muscle and adipose size and composition as potential influencing factors for adverse cardiovascular outcomes and mortality following HF [23–25]. Elderly patients with HFs may experience significant loss of muscle mass. This loss is associated with impaired functional recovery one year after hip surgery [26]. HFs are demanding challenges for patients and healthcare systems. Management cannot be limited to the operating theatre. Given the increase in the burden of disease, the true challenge is in prevention and in developing strategies to improve the quality of life for this group of patients [27]. Intertrochanteric fractures (ITF) pose significant challenges, particularly in elderly patients. Preoperative health status is a crucial predictor of postoperative outcomes in these individuals [28]. To address these challenges, some teams have conducted focused research and discussions on the surgical methods for ITF [29–32]. Orthogeriatric co-management provides better outcomes for patients with multimorbidity, in reducing the risk of adverse events after a hip fracture [33]. Elderly patients with higher hemoglobin levels who have undergone HF treatment may be discharged earlier, reducing the risk of anemia and lowering costs associated with potential readmission [34]. In older women after HF, there is limited knowledge regarding the effects of proximal femur muscle and subcutaneous adipose tissue (SAT) on mortality at various time intervals. The aim of this study was to investigate the association between the size and density of muscle and surrounding SAT and the risk of mortality in older women with different types of HF, as well as to assess the time-dependent nature of this association following the fractures.

Materials and methods

Participants

This study retrospectively analyzed data from 1166 female patients with suspected HF who were treated in the Orthopedic Trauma Emergency Department of Shanghai Sixth People’s Hospital between January 2020 and July 2022. Patients were followed for an average of 4 years, from 2020 to 2022 to 2024–2025. For admitted patients, the clinical orthopedic surgeon assessed mobility using the Parker Mobility Score (PMS) prior to the first HF, and evaluated the patient’s general condition before surgery with the American Society of Anesthesiologists (ASA) score and the Charlson Comorbidity Index (CCI). In this facility, CT scans are conducted on patients with suspected or confirmed HF. An experienced radiologist analyzes the CT images to determine the type of fracture, specifically whether it is a femoral neck fracture (FNF) or an intertrochanteric fracture (ITF). All patients underwent a CT examination within 48 h of their fracture. The exclusion criteria for HF patients were as follows: (1) Age younger than 50 years; (2) High-energy HF (such as a fall from a non-standing or sitting height); (3) Previous bilateral HF or pelvic fractures; (4) Conditions that cause long-term limited activity, such as hip dysplasia and avascular necrosis of the femoral head; (5) Inappropriate body position or poor image quality; (6) The patient did not receive surgical treatment. Four years later, patients were followed up by telephone to assess the potential incidence of death.

The study was approved by the Ethics Committee of Shanghai Sixth People’s Hospital (2023-KY-053 (K)) on April 24, 2023. Informed consent was waived because of the retrospective design.

Computed tomography acquisition

All study participants underwent spiral CT imaging of the hip using a uCT 780 scanner (United Imaging Healthcare Co., Ltd., Shanghai, China). CT scans were conducted with the patient in a supine position, encompassing the area from the top of the acetabulum to 3 cm below the lesser trochanter, including both legs. The scan parameters for all CT scans were: 120 kVp, 185 mAs, a 50 cm field of view, a 512 × 512 matrix, a 0.5 s tube rotation time, a 1 mm reconstructed slice thickness, and a standard convolution kernel algorithm. Daily air corrections were performed, followed by an air scan to measure air CT values within a standard limit of ± 6 Hounsfield Units (HU). External calibration was conducted weekly, and bed height was determined using laser alignment, referencing three crosses on the quality assurance body mold.

To cross-calibrate muscle attenuation in this CT scanner, HU values of the water-equivalent material from the European Spine Phantom were measured.

Muscle, SAT and bone parameters assessment

The cross-sectional area (CSA) and density of the specified muscle or muscle groups, along with the mid-thigh SAT (MSAT), were measured on one slice each. In subjects with HF, the non-fractured hip was analyzed. We analyzed the gluteus maximus muscle (G.MaxM) at the greater trochanter level, and the gluteus medius and minimus muscles (G.Med/MinM) at the center of S3. In the thigh, we included all mid-thigh muscles (MM) located 3 cm below the lesser trochanter of the femur, along with the surrounding MSAT.

SliceOmatic software (version 5.0; Tomovision, Montreal, Canada) was used for the analysis. The ‘Region Growing’ tool facilitated semi-automatic segmentation of the muscle and MSAT regions, using preset HU intensity thresholds of -29 to 150 HU for muscle and − 190 to -30 HU for MSAT [35]. CSA and density values for the muscle and MSAT in the regions of interest were calculated automatically. All measurements were performed by the same radiologist, who received training in CT imaging under the supervision of an expert radiologist. As part of the training, approximately 25 sample images were analyzed in collaboration with the expert before the measurement study began. Areal BMD (g/cm²) for the femoral neck (FN), trochanter (TR), intertrochanter (IT), and total hip (TH) was calculated from the CT scans using the CTXA technique (Version 4.2.3, Mindways Inc.). The details have been described in previous research [18].

Data collection

Demographic and anthropometric assessments, such as age and body mass index (BMI), were collected. Additionally, health-related data, including information on hypertension, previous fractures, and type 2 diabetes, were retrieved from hospital medical records. Comorbidity was classified using the American Society of Anesthesiologists (ASA) classification and the Charlson Comorbidity Index (CCI). These tools are useful for predicting morbidity and mortality associated with various diseases [4]. The Parker Mobility Score (PMS), which ranges from 0 to 9, is a validated tool used to assess an individual’s mobility levels [36]. PMS was assessed in two different time frames. The initial assessment (PMS1) was conducted by orthopaedists in the emergency room to evaluate the patient’s mobility during the three months preceding hip fracture surgery. The second round of assessment (PMS2) was conducted during the three months leading up to the patient’s death, using recall records obtained from interviews with family members or friends. For patients who survived, the assessment relied on records gathered from the patient or their family members within three months prior to the clinical doctor’s interview.

Ascertainment of death causes

The primary outcome of this study was all-cause mortality, which was assessed through telephone interviews with clinicians to determine survival status and cause of death.

Statistical analyzes

All statistical analyses were conducted using MedCalc (version 15.2.2, MedCalc Software bvba) and IBM SPSS (version 22.0, IBM). Differences were considered significant at P < 0.05. Continuous variables are presented as the mean ± standard deviation (SD), while categorical variables are shown as counts and percentages (%). The comparisons among various survival status categories were conducted using the two-sample non-parametric Wilcoxon or Kruskal–Wallis tests for continuous variables and the chi-squared test or Fisher’s exact test for categorical variables. The primary outcome was mortality following surgery. Cox proportional hazard models were used to assess the associations of four outcomes with various muscle, SAT, and bone parameters across different types of HF. The analysis compared patients who survived until the end of the follow-up period with the following mortality outcomes: (1) patients died in the first year after HF; (2) patients died in the second year; (3) patients died over the following 2-year period; (4) the total number of patients who died. Additionally, the association of different muscle, SAT, and bone parameters with the overall risk of death at various times after surgery was evaluated. Age, BMI, ASA, CCI, and Parker Mobility Scores obtained prior to HF surgery were used as covariates. The cumulative mortality was calculated using Kaplan-Meier survival analysis, with median values of muscle and SAT parameters serving as cut-points for explanation.

Results

Study cohort characteristics

Among the 1166 patients with low-energy HF included in the study, 245 were lost during follow-up. After reviewing medical records and CT images, an additional 543 patients were excluded from the analysis: 81 were younger than 50 years, 185 had high-energy fractures, 42 had prior bilateral hip or pelvic fractures, 39 had chronic diseases that limited mobility, 101 had inappropriate body positioning or poor image quality, and 95 had not received surgical treatment. After exclusions, the study cohort comprised 378 participants: 264 with FNF and 114 with ITF. At the end of the follow-up period, 58 patients (15.3%) had died. The mean age of deceased patients was 84.5 ± 7.9 years, which was significantly higher than the mean age of surviving patients at 72.1 ± 10.6 years. The characteristics of the study cohort are detailed in Table 1. Compared to deceased patients, surviving patients had higher parameters for muscle, SAT, and bone, except for MSAT density, which showed no significant difference (P = 0.05). The differences in the other parameters were statistically significant (P < 0.01). Among patients with HF, various surgical methods were used, including femoral head replacement and internal fixation. There was a significant difference in the percentages of surgical methods between surviving and deceased patients (P < 0.01). The number of patients in the age groups 50–59, 60–69, 70–79, 80–89, and 90 + was 41, 93, 116, 97 and 31, respectively.

Table 1.

General characteristics of study participants

Characteristics (mean ± standard deviance)	Died						P valuea
	(1) < 1 year	(2) 1–2 years	(3) > 2 years	(4) Subtotal	(5) Alive	Total	(1) vs. (2)	(1) vs. (3)	(2) vs. (3)	(1) vs.(2) vs.(3)	(4) vs. (5)
Sample size	20	16	22	58	320	378	—	—	—	—	—
Age (years)	86.55 ± 7.98	83.56 ± 7.63	83.18 ± 8.06	84.45 ± 7.93	72.13 ± 10.64	74.02 ± 11.18	0.14	0.12	0.89	0.21	<0.01
Surgery, % (n)							<0.01	<0.01	<0.01	<0.01	<0.01
Femoral head Replacement	15.00(3)	25.00(4)	54.55(12)	32.76(19)	47.81(153)	45.50(172)
Internal fixation	85.00(17)	75.00(12)	45.45(10)	67.24(39)	52.19(167)	54.50(206)
BMI (kg/m2)	21.00 ± 2.22	21.24 ± 3.14	23.23 ± 3.70	21.91 ± 3.22	22.32 ± 3.24	22.25 ± 3.23	0.52	0.02	0.20	0.06	0.51
Parker Mobility Score 1	7.05 ± 3.49	7.81 ± 2.23	8.45 ± 1.18	7.79 ± 2.50	8.41 ± 1.69	8.31 ± 1.84	0.68	0.38	0.45	0.61	0.03
Parker Mobility Score 2	2.05 ± 3.41	4.25 ± 3.34	4.59 ± 3.26	3.62 ± 3.47	7.05 ± 3.06	6.53 ± 3.36	0.02	0.01	0.80	0.02	<0.01
ASA score	2.30 ± 0.66	2.00 ± 0.37	1.91 ± 0.29	2.07 ± 0.49	1.81 ± 0.51	1.85 ± 0.52	0.13	0.01	0.41	0.04	0.001
CCI	5.20 ± 0.89	4.94 ± 0.44	4.77 ± 0.53	4.97 ± 0.67	4.19 ± 1.03	4.31 ± 1.02	0.31	0.03	0.38	0.12	<0.01
Fracture types, % (n)							<0.01	<0.01	<0.01	<0.01	<0.01
FN	15.00(3)	43.75(7)	50.00(11)	36.21(21)	75.94(243)	69.84(264)
IT	85.00(17)	56.25(9)	50.00(11)	63.79(37)	24.06(77)	30.16(114)
Prevalent diabetes, % (n)	45.00(9)	43.75(7)	40.91(9)	43.10(25)	27.81(89)	30.16(114)	<0.01	<0.01	<0.01	<0.01	<0.01
Prevalent hypertension, % (n)	40.00(8)	50.00(8)	59.09(13)	50.00(29)	47.19(151)	47.62(180)	<0.01	<0.01	<0.01	<0.01	<0.01
G.MaxM area (cm2)	28.11 ± 4.68	26.39 ± 5.48	29.04 ± 5.54	27.99 ± 5.26	33.19 ± 6.60	32.39 ± 6.68	0.26	0.48	0.15	0.28	<0.01
G.MaxM density (HU)	26.31 ± 8.82	27.98 ± 6.11	26.80 ± 7.44	26.96 ± 7.52	32.76 ± 7.81	31.87 ± 8.04	0.52	0.88	0.64	0.81	<0.01
G.Med/MinM area (cm2)	25.67 ± 5.02	23.90 ± 5.25	25.68 ± 6.47	25.19 ± 5.63	29.41 ± 5.79	28.76 ± 5.96	0.34	0.71	0.70	0.68	<0.01
G.Med/MinM density (HU)	31.96 ± 6.39	32.31 ± 5.46	33.48 ± 6.85	32.63 ± 6.26	38.10 ± 6.64	37.27 ± 6.87	0.94	0.61	0.73	0.88	<0.01
MM area (cm2)	73.04 ± 14.29	64.70 ± 9.34	78.80 ± 15.59	72.92 ± 14.59	83.98 ± 15.29	82.28 ± 15.68	0.06	0.31	0.01	0.02	<0.01
MM density (HU)	36.66 ± 6.78	36.26 ± 3.32	37.48 ± 5.93	36.86 ± 5.60	40.97 ± 5.29	40.34 ± 5.53	0.43	0.90	0.36	0.62	<0.01
MSAT area (cm2)	67.55 ± 18.21	68.84 ± 28.06	87.68 ± 26.07	75.54 ± 25.69	89.71 ± 26.16	87.54 ± 26.55	0.80	0.02	0.04	0.03	<0.01
MSAT density (HU)	-97.03 ± 9.16	-96.12 ± 8.47	-100.58 ± 5.10	-98.13 ± 7.77	-100.53 ± 5.16	-100.16 ± 5.69	0.67	0.17	0.17	0.26	0.05
TH aBMD (mg/cm2)	503.35 ± 88.85	550.63 ± 118.68	590.50 ± 118.63	549.45 ± 113.66	652.99 ± 115.20	637.10 ± 120.74	0.20	0.01	0.64	0.05	<0.01
FN aBMD (mg/cm2)	453.20 ± 85.57	485.81 ± 92.12	510.14 ± 89.66	483.79 ± 90.74	541.36 ± 94.24	532.52 ± 95.87	0.43	0.06	0.44	0.19	<0.01
TR aBMD (mg/cm2)	324.95 ± 67.11	398.00 ± 118.15	394.59 ± 102.71	371.52 ± 101.11	452.72 ± 100.06	440.26 ± 104.29	0.03	0.02	0.86	0.03	<0.01
IT aBMD (mg/cm2)	618.25 ± 107.32	679.56 ± 151.86	720.36 ± 130.13	673.90 ± 134.45	790.54 ± 136.35	772.65 ± 142.26	0.23	0.01	0.53	0.03	<0.01

* BMI, body mass index; ASA, American Society of Anesthesiologists; CCI, Charlson Comorbidity Index; G.MaxM, gluteus maximus muscle; G.Med/MinM, gluteus medius and minimus muscle; MM, midthigh muscle; MSAT, mid-thigh subcutaneous adipose tissue; TH, total hip; FN, femoral neck; TR, trochanter; IT, intertrochanter; aBMD, areal bone mineral density;

Note: Parker Mobility Score 1: assessment obtained prior to hip fracture surgery; Parker Mobility Score 2: assessment within 3 months prior to death; Group without death: assessment of mobility prior to follow-up visit

^aP value were obtained using chi-square tests or Fisher’s exact test for categorical variables, while two-sample Wilcoxon tests were used for continuous variables

Correlation of muscle, SAT and bone parameters with mortality after hip fracture

After adjusting for age, BMI, and any scores from the ASA, CCI or Parker mobility scores obtained before surgery, the results were confirmed using traditional Cox proportional hazards models, as shown in Tables 2 and 3. In the FNF group, the risk of death decreased with increasing values of the G.Med/MinM area, G.Med/MinM density, and MM density (P < 0.05). In the ITF group, the MSAT area was negatively correlated with mortality, while MSAT density showed a positive correlation with mortality. Although TH aBMD was associated with a patient’s risk of death, the effect was relatively small.

Table 2.

HRs of muscle, adipose and bone parameters per SD decrease for FNF death risk

	Unadjusted		Adjusteda		Adjustedb		Adjustedc
	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value
G.MaxM area (cm2)	0.91(0.85,0.97)	< 0.01	0.93(0.86,1.01)	0.07	0.93(0.87,1.01)	0.08	0.93(0.86,1.00)	0.06
G.MaxM density (HU)	0.92(0.87,0.97)	< 0.01	0.97(0.90,1.04)	0.34	0.96(0.90,1.03)	0.27	0.96(0.90,1.03)	0.27
G.Med/MinM area (cm2)	0.86(0.79,0.93)	< 0.01	0.88(0.80,0.96)	< 0.01	0.88(0.81,0.96)	< 0.01	0.88(0.81,0.96)	< 0.01
G.Med/MinM density (HU)	0.86(0.80,0.92)	< 0.01	0.90(0.82,0.98)	0.01	0.89(0.82,0.97)	0.01	0.89(0.82,0.97)	0.01
MM area (cm2)	0.97(0.94,1.00)	0.02	0.99(0.95,1.03)	0.52	0.99(0.95,1.02)	0.42	0.99(0.95,1.02)	0.47
MM density (HU)	0.87(0.81,0.93)	< 0.01	0.91(0.84,1.00)	0.04	0.91(0.84,0.99)	0.03	0.91(0.84,0.99)	0.03
MSAT area (cm2)	0.99(0.97,1.01)	0.15	0.99(0.97,1.01)	0.36	0.99(0.97,1.01)	0.30	0.99(0.97,1.01)	0.29
MSAT density (HU)	0.97(0.89,1.06)	0.50	0.96(0.88,1.04)	0.32	0.96(0.89,1.04)	0.34	0.96(0.88,1.05)	0.35
TH aBMD (mg/cm2)	1.00(0.99,1.000)	0.08	1.00(1.00,1.00)	0.85	1.00(1.00,1.00)	0.81	1.00(1.00,1.00)	0.73
FN aBMD (mg/cm2)	1.00(0.99,1000)	0.06	1.00(0.99,1.00)	0.64	1.00(0.99,1.00)	0.68	1.00(0.99,1.00)	0.52
TR aBMD (mg/cm2)	1.00(0.99,1.00)	0.14	1.00(1.00,1.01)	0.88	1.00(1.00,1.01)	0.98	1.000(1.00,1.01)	0.97
IT aBMD (mg/cm2)	1.00(0.99,1.00)	0.07	1.00(1.00,1.00)	0.77	1.00(1.00,1.00)	0.75	1.00(1.00, 1.00)	0.66
Parker Mobility Score 1	0.83(0.69,0.99)	0.04	NA	NA	NA	NA	NA	NA
Parker Mobility Score 2	0.82(0.73,0.92)	< 0.01	NA	NA	NA	NA	NA	NA
ASA score	2.50(1.20,5.20)	0.02	NA	NA	NA	NA	NA	NA
CCI	2.90(1.84,4.57)	< 0.01	NA	NA	NA	NA	NA	NA

* FNF, femoral neck fracture; aBMD, areal bone mineral density; ASA, American Society of Anesthesiologists; CCI, Charlson Comorbidity Index; HR, hazard ratio; SD, standard deviance; CI, confidence interval; G.MaxM, gluteus maximus muscle; G.Med/MinM, gluteus medius and minimus muscle; MM, midthigh muscle; MSAT, mid-thigh subcutaneous adipose tissue; TH, total hip; FN, femoral neck; TR, trochanter; IT, intertrochanter

Note: Parker Mobility Score 1: assessment obtained prior to hip fracture surgery; Parker Mobility Score 2: assessment within 3 months prior to death; Group without death: assessment of mobility prior to follow-up visit

^aAdjusted for age, BMI, and ASA score

^bAdjusted for age, BMI, and CCI

^cAdjusted for age, BMI, and Parker Mobility Score prior to hip fracture surgery

Table 3.

HRs of muscle, adipose and bone parameters per SD decrease for ITF death risk

	Unadjusted		Adjusteda		Adjustedb		Adjustedc
	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value
G.MaxM area (cm2)	0.91(0.86,0.97)	< 0.01	0.96(0.90,1.03)	0.28	0.96(0.90,1.03)	0.29	0.96(0.89,1.03)	0.22
G.MaxM density (HU)	0.97(0.93,1.01)	0.09	1.00(0.96,1.05)	0.98	1.00(0.96,1.05)	0.95	1.00(0.96,1.05)	1.00
G.Med/MinM area (cm2)	0.93(0.88,0.99)	0.03	0.99(0.93,1.06)	0.77	0.99(0.93,1.06)	0.77	0.99(0.93,1.06)	0.76
G.Med/MinM density (HU)	0.97(0.93,1.01)	0.15	1.03(0.97,1.09)	0.36	1.03(0.97,1.09)	0.36	1.02(0.97,1.08)	0.41
MM area (cm2)	0.96(0.94,0.99)	< 0.01	0.99(0.96,1.01)	0.31	0.99(0.96,1.01)	0.31	0.98(0.96,1.01)	0.25
MM density (HU)	0.96(0.91,1.01)	0.11	1.01(0.95,1.08)	0.69	1.02(0.95,1.08)	0.63	1.01(0.95,1.08)	0.68
MSAT area (cm2)	0.98(0.96,0.99)	< 0.01	0.98(0.96,1.00)	0.01	0.98(0.96,1.00)	0.01	0.98(0.96,1.00)	0.01
MSAT density (HU)	1.06(1.02,1.11)	< 0.01	1.05(1.01,1.09)	0.02	1.05(1.01,1.09)	0.02	1.05(1.01,1.09)	0.02
TH aBMD (mg/cm2)	0.99(0.99,1.00)	< 0.01	1.00(0.99,1000)	0.04	1.00(0.99,1.00)	0.04	1.00(0.99,1.00)	0.03
FN aBMD (mg/cm2)	0.99(0.99,1.00)	< 0.01	1.00(0.99,1.00)	0.42	1.00(0.99,1.00)	0.37	1.00(0.99,1.00)	0.30
TR aBMD (mg/cm2)	0.99(0.99,1.00)	< 0.01	1.00(0.99,1.000)	0.05	1.00(0.99,1.00)	0.04	1.00(0.99,1.00)	0.03
IT aBMD (mg/cm2)	1.00(0.99,1.00)	< 0.01	1.00(0.99,1.000)	0.06	1.00(0.99,1.00)	0.05	1.00(0.99,1.00)	0.04
Parker Mobility Score 1	0.96(0.83,1.11)	0.59	NA	NA	NA	NA	NA	NA
Parker Mobility Score 2	0.81(0.72,0.90)	< 0.01	NA	NA	NA	NA	NA	NA
ASA score	1.94(0.99,3.78)	0.05	NA	NA	NA	NA	NA	NA
CCI	2.59(1.22,5.47)	0.01	NA	NA	NA	NA	NA	NA

* ITF, intertrochanteric fracture; aBMD, areal bone mineral density; ASA, American Society of Anesthesiologists; CCI, Charlson Comorbidity Index; HR, hazard ratio; SD, standard deviance; CI, confidence interval; G.MaxM, gluteus maximus muscle; G.Med/MinM, gluteus medius and minimus muscle; MM, midthigh muscle; MSAT, mid-thigh subcutaneous adipose tissue; TH, total hip; FN, femoral neck; TR, trochanter; IT, intertrochanter

Note: Parker Mobility Score 1: assessment obtained prior to hip fracture surgery; Parker Mobility Score 2: assessment within 3 months prior to death; Group without death: assessment of mobility prior to follow-up visit

^aAdjusted for age, BMI, and ASA score

^bAdjusted for age, BMI, and CCI

^cAdjusted for age, BMI, and Parker Mobility Score prior to hip fracture surgery

When we divided patients into ‘low’ and ‘high’ groups using median cutoff values for muscle and SAT parameters, the calculations from Kaplan-Meier survival curves revealed the following findings: In the FNF group, patients with low values in the following parameters had significantly lower cumulative survival (P < 0.05) (Fig. 1): G.MaxM area, G.MaxM density, G.Med/MinM area, G.Med/MinM density, MM area, MM density. In the ITF group, patients with low MM area and MSAT area had significantly lower cumulative survival (P < 0.01), while those with high MSAT density also had significantly lower cumulative survival (P < 0.01) (Fig. 2).

Fig. 1.

Kaplan-Meier curves of FNF group for probability of death by low (blue line) vs. high (green line) parameter values using median as cut-points. (A) G.MaxM area, (B) G.MaxM density, (C) G.Med/MinM area, (D) G.Med/MinM density, (E) MM area, (F) MM density, (G) MSAT area, (H) MSAT density. CI, confidence interval; HR, hazard ratio

Fig. 2.

Kaplan-Meier curves of ITF group for probability of death by low (blue line) vs. high (green line) parameter values using median as cut-points. (A) G.MaxM area, (B) G.MaxM density, (C) G.Med/MinM area, (D) G.Med/MinM density, (E) MM area, (F) MM density, (G) MSAT area, (H) MSAT density. CI, confidence interval; HR, hazard ratio

Relationship between muscle and SAT parameters and mortality at annual intervals

After adjusting for age, BMI, ASA, CCI, and Parker mobility scores obtained before surgery, Fig. 3 compares the adjusted hazard ratios for all patients who died in the first, second, or later years after their HF. The analysis revealed that: In muscle parameters, G.MaxM area (HR 0.89, 95% CI 0.80–0.98, P = 0.02; HR 0.92, 95% CI 0.85–0.99, P = 0.03) and G.Med/MinM area (HR 0.90, 95% CI 0.81–0.99, P = 0.03; HR 0.92, 95% CI 0.85–0.99, P = 0.03) were inversely associated with mortality at 1 year and beyond following HF. MM area (HR 0.91, 95% CI 0.87–0.96, P < 0.01) was inversely associated with mortality during the 1–2 years after HF. There was no significant correlation between hip muscle density and mortality.

Fig. 3.

Adjusted* hazard ratios of death risk per one SD decrease of G.MaxM area (A), G.MaxM density (B), G.Med/MinM area (C), G.Med/MinM density (D), MM area (E), MM density (F), MSAT area (G), and MSAT density (H) with time-frame. *Adjusted for age, BMI, ASA score, CCI and Parker Mobility Score prior to hip fracture surgery

Among SAT parameters, the MSAT area (HR 0.97, 95% CI 0.95-1.00, P = 0.03; HR 0.97, 95% CI 0.94–0.99, P = 0.02) was inversely associated with mortality within 2 years after HF, while the MSAT density (HR 1.09, 95% CI 1.02–1.17, P = 0.01) was positively correlated with mortality during the 1–2 years following the fracture.

Discussion

The new finding of this study is that, among older women with HF, the size and density of the G.Med/MinM, and thigh muscle density are associated with mortality in the FNF group, whereas the MSAT area and density are linked to mortality in the ITF group. This indicates that muscle parameters in the FNF group may be more critical for mortality than adipose parameters, whereas in the ITF group, adipose parameters appear to be more significant than muscle parameters. Additionally, we found that gluteal muscle size was inversely associated with mortality one year and beyond after hip fracture. The MSAT area was negatively associated with mortality within two years after the fracture, while MSAT density showed a positive association with mortality during the 1–2 year period post-fracture. These findings suggest that increases in the size of muscle and SAT are more important for reducing mortality. Therefore, opportunistic screening using hip CT scans may be an effective strategy for identifying patients at risk of death at an early stage.

Increases in adiposity within skeletal muscle and other tissues are reported as age-related processes that can lead to mitochondrial dysfunction and impaired oxidative metabolism [37–39]. Aging has been associated with an increasing tendency of precursor cells, such as bone marrow mesenchymal cells and muscle satellite cells, to express an adipocytic phenotype rather than an osteoblastic or myocytic phenotype [19]. In skeletal muscle, age-related decreases in the ability of muscle fibers to process triglycerides lead to increased storage of lipids in the form of droplets that form along the cell membrane [40]. Sarcopenia is considered an independent factor in evaluating the prognosis of patients with medical conditions such as endovascular aortic aneurysm and pulmonary embolism [41, 42]. In addition, decreased muscle mass and increased fat mass are associated with an elevated risk of developing cardiovascular disease [43, 44]. Low muscle mass is correlated with an increased incidence of left ventricular diastolic dysfunction, which is one of the predictors of heart failure development [45]. The replacement of skeletal muscle by non-contractile components, such as lipids, appears to contribute to age-related losses in skeletal muscle function, alongside the loss of muscle mass. This results in a decrease in muscle strength and reduced performance, as well as diminished mobility, increasing the risk of trauma, such as falls [4]. In general, lower muscle mass is associated with reduced physical activity, and impaired muscle strength and function may lead to a loss of bone strength.

Fractures associated with falls in the elderly are a significant morbidity and mortality source [46–48]. There is a high probability that the person will not recover functionality to pre-fracture levels [49], frequently, shifting from a situation of independence to dependence on others. This ultimately leads to difficulties in self-care, declining functionality, loss of quality of life, increased comorbidities associated with immobility, fear of falling again, and decreased average life expectancy. As mortality and morbidity are high, these fractures directly impact public health. Gregori et al. [50] conducted five physical function tests—hand grip strength, one-leg standing, timed up and go, walking speed, and the 30-second chair stand—on 3,028 women aged 75 to 80. The results indicated that one-leg standing was the most reliable test for predicting fracture risk among older women. Retrospective studies [4, 20] have indicated that the muscle density of the paravertebral and psoas muscles, as measured by CT, is associated with mortality, but not with muscle size. A study by Boutin RD et al. [24] found that CT findings of decreased size and attenuation of the thoracic paravertebral muscles are associated with reduced overall survival in elderly patients with HF. However, Byun SE et al. [23] found that the area of the psoas muscle is associated with 1-year mortality. The similarity between our study and previous research is that hip peripheral muscle size and density were associated with mortality in patients. However, this association was observed only in the FNF group, while muscle size and density did not correlate with mortality in the ITF group. The gluteus muscles are widely regarded as some of the strongest muscles in the human body, with the G.MaxM being the largest and partially representative of the overall muscle mass [51]. The G.Med/MinM, often referred to as the “rotator cuff of the hip” [52], attach to the greater trochanter of the femur. They serve as the primary abductors and rotators of the hip, playing a crucial role in maintaining balance during normal gait [53, 54]. In our study, after adjusting for age, BMI, ASA, CCI, and Parker mobility scores, the areas of the G.MaxM and G.Med/MinM were associated with mortality one year and beyond after HF, while hip muscle density was not significantly associated with mortality. This finding is consistent with that of Wang L et al. [55], which showed that the G.MaxM area is associated with mortality two years and beyond after HF.

There is growing evidence that fat atrophy is an adverse prognostic factor linked to increased cancer-related mortality and all-cause mortality [56, 57]. Recently, Zhu W et al. [58] reported that reduced subcutaneous adiposity is linked to increased mortality in elderly women undergoing HF surgery. The study by Ahn TR et al. [59] found that high attenuation of thigh subcutaneous adipose tissue before surgery was correlated with increased mortality in elderly patients undergoing surgical treatment for proximal femoral fractures. Our findings are consistent with previous research indicating that patients with low MSAT area and high MSAT density had significantly lower cumulative survival in the ITF group, while no such association was observed in the FNF group. Although the prognostic implications of adipose tissue remain unclear, potential explanations may include fibrosis or inflammation of adipose tissue, as well as decreased energy reserves associated with reduced adipose tissue mass [60]. These factors could lead to increased attenuation and decreased area on CT. BMD is recognized as a typical risk factor for fractures and is also associated with mortality [61]. In our cohort, although TH aBMD in the ITF group was linked to a patient’s risk of death, the actual effect was minimal.

This study has several limitations. First, it was a retrospective analysis, and a significant number of patients were lost to follow-up during the follow-up period. Second, while muscle and SAT were evaluated at the individual CT level, a comprehensive 3D analysis of the entire muscle or muscle groups, along with corresponding levels of SAT, was not conducted. Third, despite performing multivariate analysis, there may still be residual or unmeasured confounding factors. Fourth, our study focused exclusively on women over the age of 50, which limits the applicability of the findings to older men. Finally, although our exclusion criterion was set at < 50 years to define a postmenopausal cohort, the number of participants in the 50–70 year age range was limited. Consequently, our findings are most representative of and generalizable to the very old women who constitute the typical hip fracture population, and we were unable to perform meaningful stratified analyses by narrower age groups.

Conclusions

Our study found that among older women with HF, the size and density of the G.Med/MinM and thigh muscle density are associated with mortality in the FNF group. In contrast, the area and density of MSAT are linked to mortality in the ITF group. Specifically, gluteal muscle size was inversely associated with mortality at one year and beyond following a HF. The MSAT area was negatively associated with mortality within two years post-fracture, while MSAT density showed a positive association with mortality between one and two years after the fracture. Therefore, for patients with different types of HF, effectively improving the size or density of hip muscles and SAT may be crucial for enhancing the survival rate of hip fracture patients—a significant issue that needs to be addressed in future research.

Author contributions

Y.L. is the primary author and contributed to the manuscript in the following ways: conception and design, patient selection, analysis and interpretation of data, manuscript drafing and revision. The following authors contributed as described: R.J. (analyzed data, interpreted results), H.Y. and Y. H. (participated in the design of the study, analyzed data, interpreted results, and revised the manuscript), J.L. (participated in the design of the study, analyzed data, interpreted results, and revised the manuscript), and Y.L. (study supervisor, conception and design, data analysis and interpretation, manuscript revision). All authors reviewed the manuscript.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to the policies and confidentiality agreements adhered to in our laboratory but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.