From Pages to Practice


Published August 7, 2019


Jack is a 55-year-old man diagnosed with essential hypertension, with a systolic blood pressure of 145 mm Hg and a diastolic blood pressure of 85 mm Hg. His primary care physician started him on an antihypertensive medication 6 months ago. Since then, Jack has been monitoring his blood pressure at home using an automatic cuff monitor. Recently, he noticed that although his systolic blood pressure remained below 140 mm Hg, his diastolic blood pressure was consistently above 90 mm Hg. When Jack visits his primary care physician for a review of his blood pressure, he asks about the significance of the diastolic reading.

The focus on hypertension, its treatment, and its effect on cardiovascular health has primarily been on the role of systolic blood pressure. The 2017 American College of Cardiology hypertension guidelines set the blood pressure thresholds for treatment at 140/90 mm Hg, and 130/80 mm Hg for high-risk patients (adults with clinical cardiovascular disease or an estimated 10-year atherosclerotic cardiovascular disease risk of  ≥10%). Diastolic blood pressure, although still recorded and mentioned in target values, has received relatively little attention.

In a retrospective cohort study recently published in the NEJM, Flint et al. examined the effects of systolic and diastolic blood pressures on cardiovascular outcomes in the general outpatient population. They found that systolic and diastolic hypertension were each independently associated with adverse cardiovascular outcomes (myocardial infarction, ischemic stroke, or hemorrhagic stroke), although systolic hypertension had a greater effect.

For Jack, his current blood pressure increases his risk of adverse cardiovascular events. Going forward, he will require treatment and monitoring for both diastolic and systolic hypertension.

The following NEJM Journal Watch summary explains the study and results in more detail:


Associations Between Systolic and Diastolic BP and Cardiovascular Outcomes

Allan S. Brett, MD and Harlan M. Krumholz, MD, SM reviewing Flint AC et al. N Engl J Med 2019 Jul 18

Fifty years ago, diastolic blood pressure (BP) was thought to be more predictive of adverse cardiovascular (CV) events than systolic BP, but epidemiologic studies eventually overturned that idea. More recently, systolic BP has been deemed more important, although both systolic and diastolic BP targets are recommended in guidelines (NEJM JW Gen Med Dec 15 2017 and J Am Coll Cardiol 2018; 71:e127). Now, researchers have explored relations between BP and 8-year CV outcomes (myocardial infarction or stroke) in more than 1 million adults (median age, 53) from northern California's Kaiser Permanente health system. The analysis used weighted averages of each person's BP measurements (median, 22 measurements per person) during the 8-year observation period.

In multivariable analyses, systolic and diastolic BP were each associated independently with increased risk for CV events, but the effect was greater for systolic than for diastolic BP. For systolic BPs of 136 and 160 mm Hg, the predicted 8-year risks for a CV event were 1.9% and 4.8%, respectively. For diastolic BPs of 81 and 96 mm Hg, CV event rates were 1.9% and 3.6%, respectively. Relative risks for CV events (but not necessarily absolute risks) were similar regardless of use of antihypertensive medication. CV risk increased when diastolic BP was in the 60s or lower (i.e., a “J-curve” relation), but this relation disappeared with adjustment for age and other covariates.

Comment — General Medicine

Allan S. Brett, MD

This report confirms that in a contemporary population-based cohort, both systolic and diastolic blood pressure are independent predictors of adverse cardiovascular outcomes.

Comment — Cardiology

Harlan M. Krumholz, MD, SM

This interesting study is more than an assessment of a single blood pressure value, but also an evaluation of the weighted average BP over time, indicating the BP burden. I wonder if variations in the patterns of BP carry additional prognostic importance and whether certain treatments were more effective in mitigating risk. I would have liked to see how the prognostic importance of the weighted average compared with a single measurement at the beginning of the study period, which is how many studies are conducted. To stratify risks, it may be that we should be using weighted averages, which could be calculated from electronic medical records.

Browse more From Pages to Practice »

Ken is a 2019-2020 NEJM Editorial Fellow and a paediatrician working in the National Health Service of the United Kingdom. He graduated from Imperial College School of Medicine and is training in general paediatrics in London.