Aging
Initiation of angiotensin II receptor blocker- versus angiotensin-converting enzyme inhibitor-based antihypertensive medication regimens and cognitive decline: An analysis of SPRINT Ryan M. Andrews* Ryan M. Andrews Daniel K. Addo Tom H. Greene Jordana B. Cohen Catherine G. Derington Ransmond O. Berchie Adam P. Bress
Delaying and preventing Alzheimer’s disease and related dementias (ADRD) are urgent public health tasks. Both randomized controlled trials and observational studies show that lowering blood pressure with antihypertensive medications protects against ADRD; however, less is known about whether all antihypertensive medications confer the same cognitive benefits. In this study, we analyzed data from the Systolic Blood Pressure Intervention Trial (SPRINT), which randomized participants to intensive versus standard systolic blood pressure control. Our analytic sample consisted of N=1,977 participants who were followed up to 7 years and who were new users of two major classes of antihypertensive medications: angiotensin II receptor blockers (ARB) and angiotensin-converting enzyme inhibitors (ACEI). To reduce bias, we conducted our study using target trial emulation procedures, with time zero defined as the initiation of ARB or ACEI therapy and eligibility restricted to those who were not taking ARB or ACEI at the SPRINT baseline visit but initiated ARB or ACEI therapy within 12 months thereafter. Our outcome of interest was the difference in the mean rate of change in cognitive performance over time on the Montreal Cognitive Assessment for ARB versus ACEI initiators, which we estimated based on a modified intent-to-treat analysis where we fit a generalized estimating equation model with an exchangeable correlation structure and stabilized inverse probability of treatment weights to address potential confounding. Overall, we found some evidence that ARB versus ACEI therapy causes a differential effect on rates of cognitive change in Montreal Cognitive Assessment scores (B=0.018 per year, 95% CI: 0.002, 0.034). Sensitivity analyses changing our model specifications and/or using alternative measures of cognitive performance yielded similar results.