Background

Low aerobic fitness and obesity are associated with atherosclerosis, and thereforegreatly increase the risk of cardiovascular disease (CVD) and early death. It has long been known that atherosclerosis my begin early in life. Despite this fact, it remains unknown how obesity and aerobic fitness early in life influence the risks of atherosclerosis, CVD and death. Furthermore, it is unknown whether high aerobic fitness can compensate for the risks associated with obesity, and how genetic confounding affects the relationshipsof aerobic fitness with CVD and all-cause mortality. Thus, the main aims of this thesis were to investigate the associations of aerobic fitness in late adolescence with myocardial infarction (Study I), stroke (Study II) and all-cause mortality (Study III), and how genetic confounding influences the relationshipsof aerobic fitness with CVD, diabetes and death (Study IV).

Methods

The study population comprised up to1.3 million men who participated in mandatory Swedish military conscription. During conscription, all conscripts underwent highly standardized tests to assess aerobic fitness, body mass index, blood pressure and cognitive function. A physician also examined all conscripts. Data on subjects’ diagnoses, death and socioeconomic status during follow-up were retrieved using record linkage. Subjects were subsequently followed until the study endpoint, date of death or date of any outcome of interest. Associations between baseline variables and the risks of adverse outcomes were assessed using Cox’s proportional hazard models. Genetic confounding of the relationships between aerobic fitness and diabetes, CVD and death was assessed using a twin population and a paired logistic regression model.

Results

In Study I, low aerobic fitness at conscription was associated with an increased risk of myocardial infarction (MI) during follow-up (hazard ratio [HR] 0.82 per standard deviation increase). Similarly, in Study II, high aerobic fitness reduced the risk of stroke (HR 0.84 for ischemic stroke, HR 0.82 for hemorrhagic stroke; P < 0.001 for all), and obesity was associated with an increased risk of stroke (HR 1.15 for ischemic stroke, HR 1.18 for hemorrhagic stroke; P < 0.001 for all). In Study III, high aerobic fitness was also associated with reduced all-cause mortality later in life (HR 0.49, P < 0.001). High aerobic fitness exerted the strongest protection against death from substance and alcohol abuse, suicide and trauma (HRs 0.20, 0.41 and 0.52, respectively; P < 0.001 for all). Obese individuals with aerobic fitness were at higher risk of MI and all-cause mortality than were normal-weight individuals with low fitness (Studies I and III). In Study IV, fit twins had no reduced risk of CVD or death during follow-up compared with their unfit twin siblings (odds ratio 1.11, 95% confidence interval 0.88–1.40), regardless of how large the difference in fitness was. However, the fitter twins were protected against diabetes during follow-up.

Conclusions

Already early in life, aerobic fitness is a strong predictor of CVD and all-cause mortality later in life. In contrast to the “fat but fit” hypothesis, it seems that high aerobic fitness cannot fully compensate for the risks associated with obesity. The associationsof aerobic fitness with CVD and all-cause mortality appear to be mediated by genetic factors. Together, these findings have implications for the view of aerobic fitness as a causal risk factor for CVD and early death.