Diastolic Blood Pressure J-Curve Phenomenon in a Tertiary-Care Hypertension Clinic

Abstract

Concerns exist regarding the potential increased cardiovascular risk from lowering diastolic blood pressure (DBP) in hypertensive patients. We analyzed 30-year follow-up data of 10 355 hypertensive patients attending the Glasgow Blood Pressure Clinic. The association between blood pressure during the first 5 years of treatment and cause-specific hospital admissions or mortality was analyzed using multivariable adjusted Cox proportional hazard models. The primary outcome was a composite of cardiovascular admissions and deaths. DBP showed a U-shaped association (nadir, 92 mm Hg) for the primary cardiovascular outcome hazard and a reverse J-shaped association with all-cause mortality (nadir, 86 mm Hg) and noncardiovascular mortality (nadir, 92 mm Hg). The hazard ratio for the primary cardiovascular outcome after adjustment for systolic blood pressure was 1.38 (95% CI, 1.18–1.62) for DBP <80 compared with DBP of 80 to 89.9 mm Hg (referrant), and the subdistribution hazard ratio after accounting for competing risk was 1.33 (1.17–1.51) compared with DBP ≥80 mm Hg. Cause-specific nonfatal outcome analyses showed a reverse J-shaped relationship for myocardial infarction, ischemic heart disease, and heart failure admissions but a U-shaped relationship for stroke admissions. Age-stratified analyses showed DBP had no independent effect on stroke admissions among the older patient subgroup (≥60 years of age), but the younger subgroup showed a clear U-shaped relationship. Intensive blood pressure reduction may lead to unintended consequences of higher healthcare utilization because of increased cardiovascular morbidity, and this merits future prospective studies. Low on-treatment DBP is associated with increased risk of noncardiovascular mortality, the reasons for which are unclear.