Derivation and validation of BOREAS, a risk score identifying candidates to develop cold-induced hypertension

Abstract

Introduction: Blood pressure increases in cold periods, but its implications on prevalence of hypertension and on individual progression to hypertension remain unclear. Our aim was to develop a pre-screening test for identifying candidates to suffer hypertension only in cold months among non-hypertensive subjects.

Methods: We included 95,277 subjects registered on a primary care database from Girona (Catalonia, Spain), with ≥ 3 blood pressure measures recorded between 2003 and 2009 and distributed in both cold (October-March) and warm (April-September) periods. We defined four blood pressure patterns depending on the presence of hypertension through these periods. A Cox model determined the risk to develop vascular events associated with blood pressure patterns. A logistic regression distinguished those nonhypertensive individuals who are more prone to suffer cold-induced hypertension. Validity was assessed on the basis of calibration (using Brier score) and discrimination (using the area under the receiver operating characteristics).

Results: In cold months, the mean systolic blood pressure increased by 3.3 ± 0.1 mmHg and prevalence of hypertension increased by 8.2%. Cold-induced hypertension patients were at higher vascular events risk (Hazard ratio=1.44 [95% Confidence interval 1.15-1.81]), than nonhypertensive individuals. We identified age, diabetes, body mass index and prehypertension as the major contributing factors to cold-induced hypertension onset.

Discussion: Hypertension follows a seasonal pattern in some individuals. We recommend screening for hypertension during the cold months, at least in those nonhypertensive individuals identified as cold-induced hypertensive by this assessment tool.

Keywords: Blood pressure; Season; Temperature; Vascular disease; Winter.