On the causes and consequences of hypertension

Abstract

Background: Economic development has resulted in remarkable achievements in human longevity. High blood pressure (BP), or hypertension, remains a leading risk factor for premature death affecting one-third of adults globally. Higher systolic BP undoubtedly causes cardiovascular disease (CVD), but the role of BP in other health outcomes, such as longevity and cancer, and the relative contributions of systolic and diastolic BP remain unclear, as do the origins of higher BP. Early childhood conditions, particularly infancy, have been extensively examined but the role of the other period of rapid growth, i.e., puberty, and its drivers, such as growth and sex hormones, have been less examined and are difficult to address by randomized controlled trials. Moreover, current knowledge from observational evidence is open to confounding and selection bias. Methods: Using a large cohort of older Chinese aged 65+ years from Hong Kong, I validated and recalibrated a risk prediction model of 10-year mortality. I then investigated whether associations of harmful exposures, such as BP, with overall and disease specific mortality in this cohort were attenuated due to selection bias stemming from only recruiting survivors to 65+ years. Using publicly available summary genetic data from the UK Biobank and other smaller studies of younger participants, I applied Mendelian randomization (MR) to examine the associations of systolic and diastolic BP with CVD, longevity and cancer. I also examined the associations of puberty, indicated by age of pubertal maturation, with adulthood BP and possible puberty related underlying drivers, i.e., insulin-like growth factor 1, sex hormone binding globulin (SHBG), testosterone and estradiol. Finally, I conducted an MR-Phenome-wide association study (MR-PheWAS) to examine the possible consequences of earlier male pubertal maturation in the men’s phenome in the UK Biobank. Results: The risk prediction model overestimated mortality risk in older Hong Kong Chinese because of overestimation of baseline mortality rates and inconsistent predictor weightings, most notably for self-rated health. In this cohort BP was not consistently associated with all-cause mortality, and associations with later-onset diseases sharing etiology with common earlier onset conditions were open to selection bias due to selective survival. Using MR, systolic and diastolic BP were independently associated with CVD and shorter lifespan, but were not clearly associated with cancer overall or by subtype after accounting for multiple testing. Pubertal timing was inversely associated with adulthood BP in both men and women independent of childhood body mass index (BMI). Pubertal timing was also positively associated with SHBG possibly via adulthood BMI, which after its adjustment was inversely associated with testosterone in men. In the MR-PheWAS, earlier male pubertal maturation was unfavorably associated with phenotypes related to physical health, such as lung and liver function, and aging, such as balding. Conclusions: Despite the difficulty of interpreting observation evidence, this thesis provided evidence clarifying the role of BP on health, and the importance of pubertal timing on future BP and other health attributes. Future studies on the relevance of these findings in other populations, and feasible strategies preventing early puberty or ameliorating its consequences, are recommended.