Abstract WP221: Relation Between Blood Pressure and Stroke Mortality in the United States Veteran Population

Abstract

Introduction: Improved control of hypertension is largely credited for declines in stroke incidence and mortality. Previous studies have reported worse outcomes after ischemic stroke with lower blood pressures prior to the stroke. We investigated the relationship between pre-stroke systolic blood pressure (SBP) and mortality after ischemic stroke among U.S. Veterans. Methods: This longitudinal analysis assessed mortality risk after incident stroke in a national sample of Veterans Health Administration hospitalizations between 2002 and 2007. We included patients who had admission ICD-9 codes for ischemic stroke (433.x1, 434, and 436, excluding 434.x0) and ≥1 outpatient blood pressure measurements 1 to 18 months prior to stroke. We defined 6 categories of average baseline SBP: <120, 120-129, 130-139, 140-149, 150-159, and ≥160 mm Hg. Multivariable Cox analyses were used to relate baseline SBP to all-cause and vascular mortality determined using the National Death Index. Results: There were 29,458 hospitalizations (mean age 67±12 years, 98% male) for incident stroke. During 4.1±3.3 years mean follow-up, there were 15,489 deaths (54%). There were 6,629 vascular deaths (23%). In a fully adjusted model, with high normal pre-stroke SBP (130-139) as the reference, risk of all-cause mortality was increased for patients with very low-to-normal SBP (<120), normal SBP (120-129), and very high SBP (≥160) (Table 1A). The risk of vascular mortality increased with very low (<120) and very high (≥160) SBP levels, but was not increased with normal SBP (120-129) (Table 1B). Conclusions: In U.S. Veterans hospitalized with first-ever stroke, very low and very high pre-stroke systolic blood pressure levels increased risk of all-cause and vascular mortality. Optimal SBP targets after stroke, particularly in patients with low pre-stroke SBP, merit further investigation.