Exploring the association between muscle mass and strength measures and hip geometry in males and postmenopausal females aged 50 and older with low bone mass.

Abstract

Background: Dual-energy X-ray absorptiometry (DXA) scans and hip structural analysis (HSA) software allow for the analysis of bone geometry in the proximal femur. Bone architecture is influenced by muscle, as described by the Utah paradigm, yet the specific muscle-bone geometry relationship remains poorly understood, thereby revealing a knowledge gap. Therefore, the investigation aimed to examine the relationship between lower limb muscle or strength measures and hip geometry, as measured by DXA scans, in a population of males and post-menopausal females aged 50 and above with low bone mass.

Methods: We led a cross-sectional, exploratory secondary analysis of the baseline data from the Finding the Optimal Resistance Training Intensity For Your (FORTIFY) Bones trial. Statistical analysis was completed using R Studio (Version 4.4.2), wherein 24 multivariable linear regressions were developed using section modulus and buckling ratio of the femoral neck, intertrochanteric region and femoral shaft as dependent variables. Lower left limb lean soft tissue (LLLLST), muscle quality index (MQI), 30-second chair stand test and normalized knee extension strength were examined as independent variables in separate models for each dependent variable. All models were controlled for age, while only models using normalized knee extension strength or 30-second chair stand test were controlled for BMI.

Results: In a sample of 324 participants (93% female, 83% Caucasian) descriptive statistics as mean (SD) were as follows: BMI 24.59 (4.33) kg/m2, normalized knee extension strength 85.91 (31.51) Nm, LLLLST 6.183 (1.2325) kg, MQI 13.68 (3.95) Nm/kg, 30-second chair stand test score 15.80 (4.33), femoral neck section modulus 1.187 (0.280) cm3, femoral neck buckling ratio 13.434 (2.948), intertrochanteric region section modulus 3.734 (0.936) cm3 , intertrochanteric region buckling ratio 9.612 (1.834), femoral shaft section modulus 2.091 (0.444) cm3, femoral shaft buckling ratio 3.149 (0.846). Models using LLLLST explained the greatest amount of variance of section modulus at the femoral shaft (Unstandardized ß=0.272, 95%CI [0.247 to 0.298]; R2=0.5840, p<0.001), intertrochanteric region (Unstandardized ß=0.520, 95%CI [0.460 to 0.581]; R2=0.4686, p<0.001) and femoral neck (Unstandardized ß=0.126, 95%CI [0.105 to 0.146]; R2=0.3147, p<0.001). Models using knee extension strength explained the second greatest amount of variance of section modulus at their respective sites. Specifically, models using normalized knee extensor strength explained 38.26% of section modulus variance at the femoral shaft (Unstandardized ß=0.00560, 95%CI [0.00434 to 0.00687]; R2=0.3826, p<0.001), 35.11% of section modulus variance at the intertrochanteric region (Unstandardized ß=0.0111, 95%CI [0.00832 to 0.01391]; R2=0.3511, p<0.001) and 18.22% of section modulus variance at the femoral neck (Unstandardized ß=0.00366, 95%CI [0.00273 to 0.00460]; R2=0.1822, p<0.001).

Conclusion: Our findings indicate that the models using LLLLST and normalized knee extension strength moderately or strongly explain section modulus depending on the region of interest. The relationships illustrated here add to our understanding of the muscle-bone relationship, specifically by highlighting the muscle-bone geometry relationship. Future investigations should determine whether changes in lean soft tissue or strength result in changes in bone geometry within participants over time.