Accuracy and Reliability of an MR‐Compatible Dorsiflexion Ergometer for Dynamic 31P‐MRS: Comparison With a Clinical Dynamometer in Individuals With and Without Obesity

The clinical application of MR-compatible ergometers for muscle contractile assessment is limited by a lack of validation against standard clinical dynamometers. Moreover, the impact of obesity on the reliability of MR ergometer-based muscle contractile assessments and the quality of phosphorus-31 magnetic resonance spectroscopy (³¹P-MRS) data remains unclear. This study aimed to validate an MR-compatible ergometer against a clinical dynamometer and to evaluate the applicability of ³¹P-MRS in individuals with severe obesity. Twenty adults (35-60 years) were recruited and divided into groups of nonobesity (BMI 18.5-30 kg/m², n = 10) and severe obesity (BMI ≥ 35 kg/m², n = 10), matched for age, sex, and height. Ankle dorsiflexion was assessed using both a clinical dynamometer and an MR ergometer, measuring maximal voluntary isometric contraction (MVIC) and a 4-min isotonic fatiguing exercise. ³¹P-MRS was continuously acquired during the in-scanner exercise. Agreement between devices was assessed using Bland-Altman plots and intraclass correlation coefficients (ICCs). ³¹P-MRS data quality was evaluated based on signal-to-noise ratio (SNR), uncertainty of fit (CRLB), and phosphocreatine (PCr) recovery fit (R²). Pearson's correlations examined relationships between muscle fatigue and metabolic parameters. All subjects successfully completed the protocol on both devices. The MR ergometer demonstrated moderate-to-excellent reliability (ICC ≥ 0.50) for most contractile parameters. While maximal torque, power, and work were underestimated on the MR ergometer (16-28%), this bias was consistent across BMI groups. ³¹P-MRS met preset quality thresholds (SNR ≥ 5, CRLB < 20%, R² ≥ 0.70) in both groups. Dorsiflexion fatigue (reduction in power) correlated strongly (r ≥ 0.77) with metabolic changes, including PCr depletion (R² = 0.68), pH drop (R² = 0.59), PCr recovery time constant (R² = 0.62), and inorganic phosphate accumulation (Pi/PCr) (R² = 0.67). The MR ergometer demonstrated feasibility, acceptable reliability, and consistent ³¹P-MRS data quality across BMI groups. These findings support the use of the MR ergometer for in-scanner dorsiflexor assessments, even in individuals with severe obesity.