Oral Presentation 50th International Society for the Study of the Lumbar Spine Annual Meeting 2024

REFINING GENDER-BASED MUSCULOSKELETAL MODELS: THE IMPACT OF AGE ADJUSTMENT (#MP-10b)

Nima Ashjaee 1 , Sidney Fels 2 , John Street 3 , Thomas Oxland 4
  1. Biomedical Engineering, University of British Columbia, Vancouver, BC, Canada
  2. Electrical and Computer Engineering, University of British Columbia, Vancouver, BC, Canada
  3. University of British Columbia, Vancouver, BC, Canada
  4. Mechanical Engineering, University of British Columbia, Vancouver, BC, Canada

INTRODUCTION

Gender-specific musculoskeletal models have played a valuable role in enhancing the prognosis and surgical planning for spinal deformities [1–3]. Another important demographic factor in adult spinal deformity is subject age, but age-specific generic models do not exist. Recognizing the age-dependent decrease in muscular max-isometric-force—a key indicator of muscular strength closely linked with muscle cross-sectional area—this study integrates age considerations into the prevailing gender-based models. By doing so, we aim to provide a more sophisticated tool that improves conventional gender-based paradigms. Our objective is to evaluate whether age-modified muscle properties can refine load predictions in gender-specific musculoskeletal models.

METHODS

We utilized a musculoskeletal modeling framework (OpenSim 3.3; Figure 1), building on established studies [4,5]. Featuring 250 subjects from the Framingham Heart Study [6], the dataset was partitioned into training (80%, N = 200) and testing sets (20%, N = 50). We simulated 11 different static postures, including neutral standing and various trunk flexion and extension positions, each chosen based on recommendations of Mokhtarzadeh et al. [7]  to elicit maximum Load-to-Strength ratios. Using regression analysis on the training set, for each primary muscle group, we identified a relationship between subjects’ age and max-isometric-force. In the testing phase, the derived equations were employed to recalibrate the generic max-isometric-force values based on age. The fully subject-specific models, wherein both spine curvature and muscle properties were tailored to the specifics of the subject, were treated as the gold standard. The difference in axial compression load prediction from L5 to T1 between the modified models were then compared against the fully subject-specific models to evaluate the efficacy of the age-adjusted generic models.

RESULTS

The established relationships between age and the max-isometric-force of major muscle groups (example: Trapezius; Figure 2) revealed a notable enhancement in load prediction when age-adjusted models were employed. There was a marked reduction in the relative difference from 7.4% (traditional gender-based model) to 4.7% when employing the age-adjusted models—a 37% decrease in error (Figure 3). This percentage decrease in error was relatively consistent across the 11 simulated postures, with the relative difference percentage falling below 6% for 10 postures and under 5% for 8 of them.

DISCUSSION

The integration of age adjustments into muscle properties improved load prediction of our biomechanical simulations, bringing them closer to what we achieve with fully subject-specific models. This improvement was consistent across different postures, highlighting the effectiveness of age-adjusted method. In clinical settings, where custom muscle data is scarce, our findings suggest that using age-adjusted models is a better alternative to the traditional, less personalized models. They offer a practical solution for patient assessment and treatment planning. Therefore, researchers should consider adopting age-adjusted modeling as a standard practice to enhance the relevance and accuracy of their work.

 

6554364fa657b-Generic+musculoskeletal+models.png

Figure 1. Musculoskeletal models in various postures.

 

6554364fa657b-regression+analysis.png

Figure 2. Regression analysis of age vs max-isometric-force. The blue dashed lines indicate the average (generic) max-isometric-force value.

 

6554364fa657b-error+reduction+age+adjusted.png

Figure 3. Error reduction after age adjustment (37% on average) with errors below 6% for 10 of 11 postures.

  1. [1] Bruno AG, Bouxsein ML, Anderson DE. Development and validation of a musculoskeletal model of the fully articulated thoracolumbar spine and rib cage. J Biomech Eng 2015;137.
  2. [2] Bassani T, Stucovitz E, Qian Z, Briguglio M, Galbusera F. Validation of the AnyBody full body musculoskeletal model in computing lumbar spine loads at L4L5 level. J Biomech 2017;58:89–96. https://doi.org/10.1016/j.jbiomech.2017.04.025.
  3. [3] Labelle H, Aubin C, Jackson R, Lenke L, Newton P, Parent S. Seeing the Spine in 3D : How Will It Change What We Do ? 2011;31:37–45.
  4. [4] Bruno AG, Mokhtarzadeh H, Allaire BT, Velie KR, De Paolis Kaluza MC, Anderson DE, et al. Incorporation of CT-based measurements of trunk anatomy into subject-specific musculoskeletal models of the spine influences vertebral loading predictions. J Orthop Res 2017;35:2164–73. https://doi.org/10.1002/jor.23524.
  5. [5] Schmid S, Burkhart KA, Allaire BT, Grindle D, Bassani T, Galbusera F, et al. Spinal compressive forces in adolescent idiopathic scoliosis with and without carrying loads: a musculoskeletal modeling study n.d.:1–15.
  6. [6] Anderson D, Mokhtarzadeh H, Allaire B, Burkhart K, Bouxsein M. Subject-specific spine models for 250 individuals from the Framingham Heart Study 2020. https://doi.org/10.7910/DVN/SJ5MVM.
  7. [7] Mokhtarzadeh H, Anderson DE, Allaire BT, Bouxsein ML. Patterns of Load-to-Strength Ratios Along the Spine in a Population-Based Cohort to Evaluate the Contribution of Spinal Loading to Vertebral Fractures. J Bone Miner Res 2021;36:704–11. https://doi.org/10.1002/jbmr.4222.