Validity of the individualized load-velocity profile to predict one-repetition maximum on a pneumatic leg press device in adults aged 55-81 years

INTRODUCTION: Resistance exercise is the primary strategy to counteract age-related declines in muscle function. Training loads are typically prescribed relative to an individual's one-repetition maximum (1-RM), but direct 1-RM testing is time-consuming and may increase injury risk in older adults. Because movement velocity can predict relative load, this cross-sectional study examined the accuracy of 1-RM estimation from individual load-velocity (L-v) profiles. METHODS: Ninety-six participants (60 males, 36 females; 69 ± 5.8 years) performed unilateral explosive repetitions on a pneumatic leg press at five submaximal loads to generate individualized L-v equations, followed by single repetitions to determine 1-RM. The group's sex-specific average measured velocity at 1-RM (V1RM; males: 0.17 ± 0.08 m·s⁻¹, females: 0.18 ± 0.09 m·s⁻¹) was defined. A corrected V1RM, which accounted for non-linearity of the L-v relationship near maximal load, was also calculated (males: 0.31 ± 0.15 m·s⁻¹, females: 0.33 ± 0.12 m·s⁻¹). To estimate 1-RM, measured V1RM and corrected V1RM values were input into the individual L-v equations. RESULTS: Estimates based on measured V1RM overestimated 1-RM (mean difference = -7.7 ± 7.4 kg; d = 1.05) and did not achieve statistical equivalence. Estimates using corrected V1RM more closely matched measured 1-RM (mean difference = -0.4 ± 6.8 kg; d = 0.06) and demonstrated statistical equivalence, with the 90% confidence interval contained within predefined equivalence bounds (±7.2 kg). CONCLUSION: Although 1-RM estimates derived from the corrected V1RM showed minimal average bias, interindividual variability persisted across both estimation methods, limiting the accuracy of 1-RM predictions at the individual level.