Date of Completion


Embargo Period



Blair T. Johnson and Tania B. Huedo-Medina

Field of Study



Master of Science

Open Access

Open Access



Purpose: Aerobic exercise (AE) lowers blood pressure (BP) 5-7 mmHg among those with hypertension, but there is considerable variability in the BP response to AE. Genetic predispositions account for 40-65% of this variability; however, identifying genetic variants that associate with the BP response to AE is a challenge. We performed a meta-analysis to integrate the small number of studies that examined the influence of candidate genes on the BP response to AE. Methods: Studies retrieved included an acute or chronic AE intervention; BP before and after AE by genotype; and candidate gene polymorphisms. Effect sizes were the standardized mean difference of BP post-vs. pre-exercise for AE training interventions, and post-vs. pre-exercise BP vs. control for acute AE interventions. Effect sizes were disaggregated for genotype and adjusted for baseline sample features. Analyses followed fixed‑effects assumptions. Results: 11 AE training (N=2646) and 4 acute AE (N=50) studies qualified. AE training interventions were performed at 62.3±7.5% (Mean+SD) maximum oxygen consumption for 43.8±11.6 min×session-1,3.6±1.2 d×wk-1 for 15.3±7.6 wk. Participants were mostly non-Hispanic white (N=1,736) men (N=1,278) and women (N=1,360), 44.2±10.7 yr with a BP of 134.4±11.9/78.6±9.3 mmHg and body mass index of 26.9±2.6 kg.m-2. The effect of exercise on the BP response to AE training was small but statistically significant for systolic BP (SBP) (d+ = -0.21 [95% CI = -0.247, -0.168], -3.1 mmHg, I2=77.8%) and diastolic BP (DBP) (d+ = ‑0.20 [95% CI = -0.235, -0.158], -1.9 mmHg, I2=62.2%). Sample features explained 59.1-71.5% of the variability in the BP response to AE training (P< 0.001), and reductions were greater among samples that had a higher resting BP (SBP: β = -0.68, P< 0.001; DBP: β=-0.56, P=0.01), that were younger (SBP: β=0.34, P<0.01; DBP: NS, P>0.05), and that included more women than men (SBP: β = 0.41, P<0.001; DBP: β=0.52, PAGT) M235T (rs699) polymorphism showed a significant association with the DBP response to AE training (Multiple R=0.058, P=0.02), explaining 0.3% of the variability in the DBP response. Pairwise comparisons of AGT M235T genotypes showed those with the AGT MM genotype reduced DBP 2.9 mmHg more in response to AE training compared to those with the AGT TT genotype (Multiple R=0.076, P=0.02). Acute interventions were performed at 50.1±10.1% maximum oxygen consumption for 40 min·session-1. Participants were men, 44.1±1.0 yr with a BP of 145.7±1.7 / 85.8±0.9 mmHg and body mass index of 29.9±0.3 kg.m-2. BP responses to acute AE were large and heterogeneous for SBP (d+ = -0.62 [95% CI = ‑0.75, ‑0.50], ‑5.5 mmHg, I2 =48%), and small and homogeneous for DBP (d+ =-0.28 [95% CI = ‑0.40, ‑0.16], ‑1.7 mmHg, I2 =0%). Sample features explained 55.2-82.3% of the variability in the BP response to acute AE (P< 0.001), while candidate gene polymorphisms explained a marginally significant 4.6-6.0% of the variability (P=0.08). Analyses of individual polymorphisms were not feasible due to the low numbers of interventions and observations. Conclusions: Despite our attempt to increase the sample size to detect polymorphism associations with the BP response to AE, sample features explained most of the variability across trials, although the AGT M235T polymorphism is promising. These findings reinforce the notion that most genetic variants explain only a small amount of variability in the response of health/fitness phenotypes to exercise, if any. Future research efforts seeking to explain the variability of health/fitness phenotypes to exercise such as BP should explore sample features known to influence the phenotype of interest, as well as the multiple levels of gene regulation using high throughput screening in larger, more ethnically diverse samples of men and women with HTN.

Major Advisor

Linda S. Pescatello