Our aim was to calculate the predictability of different blood pressure measures for cardiovascular mortality in a cohort of both men and women. We also aimed to determine whether clinically applicable cut-off levels for cardiovascular mortality risk of these measures work well.
A healthcare need investigation from the 1970s was used. Participants aged 46–65 were included, n=788 (390 men and 398 women). The following blood pressure measures were studied: systolic, diastolic, mean, mid, and pulse pressure. The participants were followed for 26 years with respect to cardiovascular mortality through the Swedish Cause-of-Death Register. Isolated diastolic hypertension failed to show significant associations with cardiovascular mortality.
Combined systolic and diastolic hypertension showed twice as high cardiovascular mortality in men and women compared with those with normal blood pressure. Mid arterial blood pressure showed increased significant hazard ratios for all three grades of hypertension in men and for grades 2 and 3 in women with good predictability (area under the curve=0.72 and 0.80, respectively).
Mid arterial blood pressure is strongly associated with cardiovascular mortality. Additional studies in larger populations and with a wider age range comparing mid arterial blood pressure with clinically useful cut-offs of other blood pressure measures are required to corroborate our findings.
aDepartment of Neurobiology, Centre for Family Medicine, Care Sciences and Society, Karolinska Institutet, Huddinge
bDepartment of Public Health and Caring Sciences/Section of Geriatrics, Uppsala University, Uppsala, Sweden
Correspondence to Axel C. Carlsson, MSc, PhD, Department of Neurobiology, Centre for Family Medicine, Care Sciences and Society, Karolinska Institutet, Alfred Nobels Allé 12, S-141 83 Huddinge, Sweden Tel: +46 85 248 866 5; fax: +46 85 248 870 6; e-mail: firstname.lastname@example.org
Received August 2, 2012
Accepted December 21, 2012