Abstract
Purpose: :
Measurements of the width of the retinal arterioles have been shown to be associated with blood pressure and cardiovascular disease morbidity and mortality. Our objective is to describe the impact of systolic blood pressure (SBP) and blood pressure medication use (BPMed) on changes in the central retinal arteriole equivalent (CRAE) over three 5-year intervals in the Beaver Dam Eye Study.
Methods: :
There were 2871 persons 43-86 years of age at the time of the baseline examination who had data available from at least two consecutive examinations performed in 1988-90, 1993-95, 1998-2000, and 2003-05. Retinal vessel measurements were performed using a semi-automated computer program. First, we modeled the relationship of SBP and BPMed at the beginning of the interval to change in CRAE over each 5-year interval. Second, a cross-sectional model was used to estimate change in CRAE over a 5-year interval using different scenarios for changes in SBP and BPMed over the same interval.
Results: :
Mean change in CRAE over each 5-year interval ranged from -0.5 µm to -1.6 µm. In multivariate analyses, both higher SBP (estimated percent change in CRAE per 10 mmHg SBP 0.26; 95% confidence interval [CI] 0.16-0.36) and BPMed use (estimated percent change in CRAE 1.76; 95% CI 0.58-2.93) at the start were associated with an increase in CRAE over the 5-year interval. Using a cross-sectional model, we estimated the effect of changes in both SBP and BPMed on change in CRAE. For example, based on applying this model to an individual who was not taking BP medication and had a SBP of 165 mmHg at a given examination, and then took a BP medication at the next exam and had a SBP of 155 mmHg, an increase in CRAE of 1.79 µm (95% CI 1.14-2.44 µm) over that 5-year interval would result.
Conclusions: :
SBP and BPMed are both related to changes in CRAE over a 5-year interval. From cross-sectional analyses, we show how changes in BPMed over the interval may impact SBP and in turn CRAE such that simply looking at SBP and BPMed at the start of the interval with change in CRAE over the interval may misrepresent the complex relationship between SBP, BPMed and CRAE that occurs over time.
Keywords: retina • clinical (human) or epidemiologic studies: outcomes/complications • clinical (human) or epidemiologic studies: risk factor assessment