Influence Of Age, Sex And An Occurrence Of Cardiovascular Event On Seasonal Variations In Blood Pressures In Hypertensive Patients: Insights From The Ascot Cohort

Objective: In hypertensive patients, the extent to which patient-related characteristics influence seasonal blood pressure (BP) variations is unclear. We used repeated BP-measurements in the ASCOT trial to investigate the influence of age, sex, geographical residence and cardiovascular (CV) event occurrence, on the association between month-of-year and BP-level. Design and method: 18,746 hypertensive patients, with 5-year median follow-up and a median (IQR) of 14 [12–17] visits, contributed to these analyses. We developed mixed effects models to estimate mean systolic-BP (SBP) for each month of the year. Models were adjusted for a priori confounders (age, sex, BMI, total cholesterol, presence of diabetes, smoking status, ethnicity, socio-economic status, randomized treatment-allocation) and in-trial period. Subgroup analyses were conducted by age-groups, sex, geographical residence, and occurrence of CV-event or not. Age was appropriately categorised (see figure) and geographical region was split into two-groups: Scandinavian countries and Scotland and England and Ireland. We estimated mean SBP-difference between the coldest pair of months (January/February) and warmest (July/August): overall; and separately by subgroups; evaluating for presence of interactions. Results: Mean SBP differs significantly across months of year, with higher BPs during winter and lower BPs during summer (mean SBP-difference between coldest and warmest pairs: 3.07 [95% CI 2.87, 3.27] mm Hg, p < 0.001). The trend was similar in each age-group, but with a significantly higher mean SBP-difference amongst older (= >75 years) patients (4.59 [3.87, 5.30] mm Hg) vs younger (<55 years) (1.97 [1.47, 2.46]) (interaction p-value < 0.001). The magnitude of mean SBP-difference differs substantially by geographical region with a mean SBP-difference of 1.99 [1.70, 2.27] mm Hg amongst those in Scandinavia and Scotland vs. 4.10 [3.82, 4.39] amongst those in England and Ireland (interaction p-value < 0.001). Those who had a CV-event had significantly higher seasonal BP-variability vs. those who didn’t, and amongst those with a CV-event: the seasonal BP-variability was higher after the event vs before (p-value = 0.034). No interaction with gender was apparent.Conclusions: The magnitude of seasonal BP-variability is influenced by age, residence, and an occurrence of CV-event in hypertensive patients. These findings may have important clinical implications for BP-control strategies.