A generalized equation for predicting peak oxygen consumption during treadmill exercise testing: mitigating the bias from total body mass scaling.

Abstract

Background: Indexing peak oxygen uptake (VO₂peak) to total body mass can underestimate cardiorespiratory fitness (CRF) in women, older adults, and individuals with obesity. The primary objective of this multicenter study was to derive and validate a body size-independent scaling metric for VO₂peak. This metric was termed exercise body mass (EBM).

Method: In a cohort of apparently healthy individuals from the Fitness Registry and the Importance of Exercise National Database, we derived EBM using multivariable log-normal regression analysis. Subsequently, we developed a novel workload (WL) equation based on speed (Sp), fractional grade (fGr), and heart rate reserve (HRR). The generalized equation for VO₂peak can be expressed as VO₂peak = Cst × EBM × WL, where Cst is a constant representing the VO₂peak equivalent of one metabolic equivalent of task. This generalized equation was externally validated using the Stanford exercise testing (SET) dataset.

Results: A total of 5,618 apparently healthy individuals with a respiratory exchange ratio >1.0 (57% men, mean age 44 ± 13 years) were included. The EBM was expressed as Mass (kg)^0.63 × Height (m)^0.53 × 1.16 (if a man) × exp (-0.39 × 10^-4 × age²), which was also approximated using simple sex-specific additive equations. Unlike total body mass, EBM provided body size-independent scaling across both sexes and WL categories. The generalized VO₂peak equation was expressed as 11 × EBM × [2 + Sp (in mph) × (1.06 + 5.22 × fGr) + 0.019 × HRR] and had an R ² of 0.83, p < 0.001. This generalized equation mitigated bias in VO₂peak estimations across age, sex, and body mass index subgroups and was validated in the SET registry, achieving an R ² of 0.84 (p < 0.001).

Conclusions: We derived a generalized equation for measuring VO₂peak during treadmill exercise testing using a novel body size-independent scaling metric. This approach significantly reduced biases in CRF estimates across age, sex, and body composition.