Potential of Stratified Medicine for High Blood Pressure Management: A Modeling Study Using NHANES Survey Data

International audienceThe 2017 American College of Cardiology/American Heart Association hypertension guidelines lowered the thresholds for defining and treating hypertension. However, the SPRINT trial showed substantial heterogeneity in benefits and harms of intensive antihypertensive treatment depending on patients’ characteristics. We aimed at illustrating the potential gains of personalizing intensive antihypertensive treatment. Using the US National Health and Nutrition Examination Survey 2011 to 2014 (n=2067), and prediction models derived from the SPRINT trial, we computed expected benefits and harms of intensive antihypertensive treatment for individuals aged 50 or more. We compared 2 interventions: (1) intensive antihypertensive treatment for all individuals meeting the 2017 American College of Cardiology/American Heart Association thresholds and (2) a stratified medicine strategy excluding from intensive treatment individuals with predicted unfavorable benefit-risk. Outcome measures were model-predicted 5-year risk of cardiovascular events or death (myocardial infarction, acute coronary, stroke, acute decompensated heart failure, and cardiovascular-related death), and severe adverse events (hypotension, syncope, electrolyte abnormalities, bradycardia, and acute kidney injury). Per 2017 American College of Cardiology/American Heart Association guidelines, 40.1 million (39.2%) US individuals aged 50 or more should initiate or intensify antihypertensive treatment, thereby preventing cardiovascular events for 795 000 individuals and inducing severe adverse events for 848 000 over 5 years. A stratified treatment strategy could decrease the number of individuals treated by 21.2 million (52.9%) and reduce the number of individuals with severe adverse events by 38.3%, with 11.7% fewer individuals with cardiovascular events prevented. Personalizing antihypertensive treatment according to predicted benefits and harms could spare treatment for more than half individuals while reducing harms 3× more than benefits