Author(s): ARMERO-SOTILLO, A., RAMOS-CAMPO, D.J., NOBARI, H., BENÍTEZ-MUÑOZ, J.A., GUISADO-CUADRADO, I., ROJO-TIRADO, M.A., BENITO, P.J., PEINADO, A.B., Institution: UNIVERSIDAD POLITÉCNICA DE MADRID, Country: SPAIN, Abstract-ID: 1369

INTRODUCTION:
Maximal oxygen consumption (VO₂max) shows a well established age related decline mainly driven by central cardiovascular limitations (1). In contrast, the behaviour of ventilatory thresholds (VT) is less clear, with studies reporting inconsistent age related patterns and sex specific differences. These discrepancies are partly explained by the different methods used to estimate and determine VT (2,3). In addition, the distinct peripheral mechanisms influencing VT, highlight the need for large scale analyses that clarifies their lifespan trajectory. Therefore, the aim of this study was to model age related changes in VT throughout life, using both classical linear regression and newer approaches such as machine learning.
METHODS:
The cross‑sectional dataset consisted of 3,538 individuals (18–83 years; 79.6% males) who performed a ramp incremental test to volitional exhaustion on either a treadmill or a cycle-ergometer. Oxygen consumption relative to body weight at the VT and VO₂max were determined using standard ventilatory and gas‑exchange criteria (4). Age‑related associations were examined using linear and polynomial regression models complemented by XGBoost to detect complex or non‑linear relationships.
RESULTS:
Both VT showed a clear non‑linear decline with advancing age, with steeper reductions observed in males. VT1 declined between approximately −1.1 and −1.7 mL·kg⁻¹·min⁻¹ per decade depending on sex, while VT2 showed an even larger reduction ranging from −1.3 to −2.6 mL·kg⁻¹·min⁻¹ per decade. VO₂max also decreased substantially up to −3 mL·kg⁻¹·min⁻¹ per decade (in males). Conversely, both thresholds expressed as %VO₂max remained remarkably stable across the lifespan, displaying near‑zero slope values regardless of age, sex or ergometer. These patterns were consistent across all modeling approaches, with little to none differences between lineal regression and XGBoost.
CONCLUSION:
Results indicate a preserved proportional metabolic regulation despite substantial reductions in absolute aerobic capacity. Our findings also highlight the strong relationship between the proportional decrease in consumption at VT and the decrease in VO2max. In addition, our large sample size also allows us to provide a reference for how much oxygen consumption is expected to decrease at the VT over the course of a lifetime. This also underscores the importance of not neglecting the training of both VT and using them as an effective methodology for prescribing training throughout the entire life span.
REFERENCES:
1. Letnes J.M. et al, 2023, DOI: 10.1016/j.ijcrp.2023.200171
2. Anselmi F. et al, 2021, DOI: 10.1111/sms.14007
3. Martini A.D. et al, 2022, DOI: 10.3390/ijerph19073962
4. Rabadán M. et al, 2011, DOI: 10.1080/02640414.2011.571271