Quantitative Assessment of Forearm Muscle Architecture Using Dynamic High-Frequency Ultrasound: Determinants of Grip Strength and Predictive Modeling

Objective The study quantitatively evaluated the relationship between individual forearm muscle architecture and handgrip strength (GS) using dynamic high-frequency ultrasound (US) to develop a multivariate functional predictive model informed by latest international muscle health consensus. Methods Forty-one healthy volunteers (21 females, 20 males; median age 22 years) were enrolled. Maximum GS (standardized isometric dynamometry at 110° elbow flexion) and US-derived muscle thickness (MT) and cross-sectional area (CSA) of seven forearm and hand muscles—palmaris longus (PL), flexor digitorum superficialis (FDS), flexor digitorum profundus (FDP), flexor carpi radialis (FCR), flexor carpi ulnaris (FCU), flexor pollicis longus (FPL), and thenar muscles (TM)—were measured at rest and maximum voluntary contraction (MVC). Surface electromyography (sEMG) from five muscles was recorded simultaneously to assess neural drive. Stepwise multiple linear regression was applied for predictive modeling. Results Significant sexual dimorphism was observed in GS and muscle morphometry (P < 0.05). GS significantly correlated with the MT and CSA of the PL and FCU in both resting and contracted states (P < 0.05). Notably, the primary finger flexor (FDS) and sEMG signals showed no significant correlation with GS (P > 0.05). The final multiple regression model (R² = 0.682, P < 0.001) identified weight, BMI, FCU thickness at MVC (FCUT2), and PL thickness at MVC as key quantitative predictors. FCUT2 was the most significant sonographic contributor (β = 0.430). Conclusions Dynamic high-frequency US provides a precise, quantitative method for functional muscle assessment. Contractile thickness of the FCU and PL are superior predictors of GS, highlighting their role in wrist stabilization during power grip. This model offers potential for early sarcopenia screening and rehabilitation monitoring consistent with AWGS 2025.