Send to

Choose Destination
Circulation. 2018 Oct 23;138(17):1805-1814. doi: 10.1161/CIRCULATIONAHA.118.033348.

Association of Blood Pressure Measurements With Peripheral Artery Disease Events.

Author information

Division of Vascular Surgery, Department of Surgery (N.K.I., C.K.L., N.J.L.), Stanford University School of Medicine, CA.
Department of Health Research and Policy (D.S.T., S.M.), Stanford University School of Medicine, CA.
Division of Pediatric Critical Care Medicine, Department of Pediatrics (D.S.T.), Stanford University School of Medicine, CA.
Division of Nephrology, Department of Medicine (T.I.C.), Stanford University School of Medicine, CA.



Current guidelines recommend treating hypertension in patients with peripheral artery disease (PAD) to reduce the risk of cardiac events and stroke, but the effect of reducing blood pressure on lower extremity PAD events is largely unknown. We investigated the association of blood pressure with lower extremity PAD events using data from the ALLHAT trial (Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial).


ALLHAT investigated the effect of different antihypertensive medication classes (chlorthalidone, amlodipine, lisinopril, or doxazosin) on cardiovascular events. With the use of these data, the primary outcome in our analysis was time to first lower extremity PAD event, defined as PAD-related hospitalization, procedures, medical treatment, or PAD-related death. Given the availability of longitudinal standardized blood pressure measurements, we analyzed systolic blood pressure (SBP), diastolic blood pressure (DBP), and pulse pressure as time-varying categorical variables (reference categories 120-129 mm Hg for SBP, 70-79 mm Hg for DBP, and 45-54 mm Hg for pulse pressure) in separate models. We used extended Cox regression with death as a competing risk to calculate the association of each blood pressure component with PAD events, and report the results as subdistribution hazard ratios and 95% CIs.


The present analysis included 33 357 patients with an average age of 67.4 years, 53.1% men, 59.7% white race, and 36.2% with diabetes mellitus. The median baseline blood pressure was 146/84 mm Hg. Participants were followed for a median of 4.3 (interquartile range, 3.6-5.3) years, during which time 1489 (4.5%) had a lower extremity PAD event, and 4148 (12.4%) died. In models adjusted for demographic and clinical characteristics, SBP <120 mm Hg was associated with a 26% (CI, 5%-52%; P=0.015) higher hazard and SBP≥160 mm Hg was associated with a 21% (CI, 0%-48%; P=0.050) higher hazard for a PAD event, in comparison with SBP 120 to 129 mm Hg. In contrast, lower, but not higher, DBP was associated with a higher hazard of PAD events: for DBP <60 mm Hg (hazard ratio, 1.72; CI, 1.38-2.16). Pulse pressure had a U-shaped association with PAD events.


In this reanalysis of data from ALLHAT, we found a higher rate of lower extremity PAD events with higher and lower SBP and pulse pressure and with lower DBP. Given the recent revised blood pressure guidelines advocating lower SBP targets for overall cardiovascular risk reduction, further refinement of optimal blood pressure targets specific to PAD is needed.


URL: . Unique identifier: NCT00000542.


blood pressure; hypertension; hypotension; peripheral arterial disease; standards

[Available on 2019-10-23]

Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center