Quantifying altered oxygen kinetics and reducing metabolic test times for children with cerebral palsy: a dual-exponential Bayesian modeling approach.

Prior research using indirect calorimetry has shown that children with cerebral palsy (CP) exhibit significantly increased energetic costs during walking. However, metabolic testing to obtain oxygen cost is challenging. As a result, differences in oxygen uptake kinetics (V̇o₂) in CP compared with their typically developing peers remain unexplored. Step changes in work rate have been shown to result in an exponential V̇o₂ response with three distinct phases 1) cardiodynamic, 2) primary, and 3) steady-state. Here, we applied a dual-exponential Bayesian model to assess the time constant of the primary phase V̇o₂ response from resting to walking in children with CP. In addition, we evaluated the model's ability to estimate steady-state V̇o₂ using shorter test durations. From a sample of 263 children with CP, the median V̇o₂ time constant was 33.1 s (5th-95th percentile range: 14.5-69.8 s), significantly longer than reported values for typically developing children (range of means: 10.2-31.6 s). Furthermore, the model accurately estimated steady-state V̇o₂ using only the first 3 min of metabolic data compared with the typical 6 min used in current clinical practice. The 3-min estimate explained >95% of the 6-min estimate variance, with <5% mean absolute error. Slower oxygen kinetics in children with CP suggest impairments in metabolic control, potentially contributing to their higher energy demands. Although the exact mechanisms remain unclear, this study provides valuable insights into the walking energetics of children with CP and presents a more efficient method for analyzing V̇o₂ for this population.NEW & NOTEWORTHY We used a Bayesian model to study oxygen kinetics in children with CP. First, we quantified the oxygen kinetics time constant, finding no significant differences across GMFCS, sex, or age despite slower kinetics compared with typically developing peers. Next, we validated the model's ability to estimate steady-state V̇o₂ with only 3 min of data, halving the usual testing time. This method has important implications for faster, accurate metabolic analyses in clinical and research settings.