Exercise shifts blood markers most in men, under-50s and the metabolically unhealthy

Guidelines everywhere recommend exercise for cardiovascular health, yet the trial evidence connecting exercise to that benefit through intermediate blood markers has been surprisingly inconsistent. If exercise protects the heart, it should show up in the biomarkers that predict cardiovascular risk: lipids, insulin measures, inflammatory and hormonal markers. Establishing which of these actually move with training, and in whom, matters for designing lifestyle advice that targets the people with the most to gain. This meta-analysis set out to quantify those effects across every available randomised trial in adults without existing cardiovascular disease.

Two researchers selected 160 randomised controlled trials covering 7487 participants, drawn from searches of Medline, Embase and Cochrane Central spanning January 1965 to March 2014. All trials tested exercise interventions in adults without cardiovascular disease, and data were extracted using a standardised protocol. The team then ran a random-effects meta-analysis and systematic review evaluating the effects of exercise on cardiorespiratory fitness and on a wide panel of conventional and novel circulating cardiometabolic biomarkers, and examined whether the effects differed by age, sex and health status.

Exercise significantly raised both absolute and relative cardiorespiratory fitness. Lipid profiles improved, with lower triglycerides and higher HDL cholesterol and apolipoprotein A1. On the glucose side, exercise groups showed lower fasting insulin, lower insulin resistance by the homeostatic model, and lower glycosylated haemoglobin. Exercise groups also had lower leptin, fibrinogen and angiotensin II, though interestingly higher levels of interleukin-18. The effects were modified by age, sex and health status: people under 50, men, and those with type 2 diabetes, hypertension, dyslipidaemia or metabolic syndrome appeared to benefit more.

This analysis strengthens the case that exercise works on cardiovascular risk partly through measurable blood markers, addressing the inconsistency across individual trials by sheer pooled scale. The moderation findings complicate the one-size-fits-all message: the same programme may shift markers more in some groups than others, which the authors note has implications for designing targeted lifestyle interventions. Because the trials span decades and used very different programmes, the analysis cannot say which exercise type, intensity or duration is optimal for any given marker, and the raised interleukin-18 finding sits oddly and invites further study.

This is a reason to think regular training improves fitness and a meaningful set of risk markers even in people who do not have heart disease, covering cholesterol, insulin handling and blood sugar control. It is also a reason to think the people who often feel exercise is least for them, those already carrying metabolic problems like type 2 diabetes or high blood pressure, may see some of the largest biomarker improvements. The analysis describes marker changes rather than counted heart attacks, so it maps the pathway rather than the destination.