Should women have lower blood pressure targets than men? Sex differences in blood pressure and cardiovascular disease in the UK Biobank

BackgroundRecent studies show that the risk of cardiovascular disease (CVD) increases from a lower nadir of systolic blood pressure (SBP) in women than men, and increases thereafter at a greater rate. This has led to a suggestion that sex-based SBP thresholds are required. We aimed to investigate sex differences in the associations of SBP and incident atherosclerotic CVD in a large prospective cohort. Methods420,649 UK Biobank participants with no prior history of CVD were included. Age-adjusted sex-specific risks, relative risks (RRs) and risk differences (RDs) relating SBP to CVD were estimated using Poisson and Cox regression. ResultsOver 13.2 years of follow-up there were 28,628 CVD events. CVD risks across BP levels showed a "J-shape", and were higher in men than women at all BP levels. The lowest risks were at SBP 100-<105 mmHg (events per 10,000 person-years [95%CI]: 15.6 [11.8-23.1]) and 110-<115 (47.2 [41.8-53.0]) among women and men, respectively. Compared with SBP 100-<110, sex-specific RRs at above 120 were higher in women than men, but RDs were higher in men than women at all levels of SBP. Furthermore, compared to men at 100-<110 (i.e. the men with least risk), risks in women were lower at all levels of SBP below 170. ConclusionsCVD risk is lowest for women at a slightly lower SBP than men and RRs for CVD increase with SBP at a slightly steeper rate in women. However, both risks and RDs in women are never greater than in men. This evidence suggests that women should not have lower BP thresholds. Clinical Perspective What is New?O_LIWhile risks are higher at lower BP levels in women than men when both sexes are considered independently, consideration of the risks and sex-combined analyses do not support treating women at lower BP levels than men. C_LI What are the Clinical Implications?O_LIIn line with 2017 ACC/AHA guideline, antihypertensive treatment thresholds should be based on absolute risk calculated from sex-specific algorithms, rather than BP level in isolation. C_LIO_LIBoth risks and relative risks must be taken into account to understand sex-differences in the relationship of BP-CVD and whether these differences are clinically meaningful. C_LI