Questioning the utility of oxidative stress measurements as biomarkers of physiological condition and fitness

Markers of oxidative stress are widely used as biomarkers of health, ageing, and physiological stress. However, their reliability as biomarkers remains uncertain due to high intraindividual variation, obscuring associations with environmental conditions, lifestyle, frailty, and physiological indicators of health such as telomere length. Whilst numerous longitudinal studies exist, individual repeatability of oxidative stress measurements is rarely reported. This study presents the first meta-analysis assessing individual repeatability of oxidative stress markers, comprising 123 repeatability estimates from 22 studies. We found that oxidative stress exhibits low individual repeatability (Intraclass correlation = 0.164), regardless of marker type, taxa, sex, study design, or environment. We also found that different markers of oxidative stress were often poorly correlated, which could be due to their low repeatability. This flags serious limitations regarding the utility of oxidative stress measurements as health biomarkers. To illustrate this point, we simulated causal effects of oxidative stress on telomere length to reveal statistical power limitations on the detection of a relationship between oxidative stress and telomere length when individual repeatability is low. Our simulations reveal that substantially larger samples sizes are required than those typically used in this field. On a more positive note, we also show that increasing the number of repeated samples can improve statistical power. Such longitudinal studies would create valuable opportunities for untangling the causes of intraindividual variation in oxidative stress, elucidating the functional consequences of oxidative stress for health and ageing.