A comparison of drugs targeting the renin-angiotensin system and global cognition in aging

Abstract

Background

Hypertension (HTN) is a major risk factor for Alzheimer's disease (AD) and related dementias. Renin-angiotensin system (RAS) medications, including angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin II receptor blockers (ARBs), have been proposed to mitigate cognitive decline, but evidence remains inconsistent.

Objective

Compare global cognition over time among ARB, ACEI, and non-RAS drug users.

Methods

Data from the Rush Memory and Aging Project (1997–2006) included 2019 participants. Global cognition was assessed using a composite z-score of 19 cognitive tests. Participants were categorized by medication use: ARB, ACEI, or non-RAS. Analysis of covariance (ANCOVA) was performed to compare cognitive outcomes over 9 years, adjusting for baseline cognition, social determinants, and medical factors. Subgroup analyses evaluated cognitive domains and the impact of HTN.

Results

At baseline, ARB and ACEI users had more vascular risk factors and medical conditions than non-RAS users. At 9 years, ARB users had slightly better global cognitive outcomes compared to ACEI (LS-mean difference: 0.015, 95% CI [−0.175, 0.205]) and non-RAS users (LS-mean difference: 0.06, 95% CI [−0.11, 0.23]), though differences were not statistically significant. ACEI users showed significant improvement in working memory compared to non-RAS users (LS-mean: 0.157, 95% CI [0.002, 0.313]). Among participants without HTN, ACEI use was associated with significantly reduced cognitive decline compared to non-RAS use.

Conclusions

ARBs may benefit global cognitive function, while ACEIs may yield improvements in specific domains like working memory. Further studies are needed to confirm these findings and explore the impact of drug characteristics such as blood-brain-barrier penetration.

Keywords

aging Alzheimer's disease cognitive function renin-angiotensin system

Introduction

Over 55 million people around the world are living with Alzheimer's disease (AD) and related dementias. Hypertension (HTN) is one of the more common risk factors for AD development and progression.¹ HTN is quite prevalent, affecting 74% of individuals 60 years of age and older.^2,3 Medications targeting the renin-angiotensin system (RAS) have been identified as potential treatments for AD. The RAS is a hormonal system that regulates blood pressure, fluid balance and contributes to oxidative stress and inflammation. These medications interact with the RAS by either increasing the level of Renin (dihydropyridone calcium channel blockers, thiazide diuretics), decreasing Beta-1 mediated renin production (beta blockers), inhibiting angiotensin II activity at the AT1 receptor (angiotensin receptor blockers), or inhibiting angiotensin II production (angiotensin converting enzyme inhibitors, ACEI).⁴ Specifically, calcium channel blockers and angiotensin receptor blockers (ARB) have been associated with lower risk of dementia in observational studies^5,6 One hypothesis is that these medications interact with the RAS by increasing Renin and Angiotensin II production which results in neuroprotective effects by promoting AT4 production. Furthermore, angiotensin-converting enzyme (ACE) has been reported to degrade amyloid-β (Aβ), which comprises the neuritic plaques associated with AD suggesting and ACEI could promote Aβ plaque accumulation.⁴

Through mechanisms that promote the angiotensin 2 receptor, Angiotensin 4 receptor and choose of RAS drugs that cross the blood-brain barrier (BBB) and interact with the Brain RAS, there is much to learn about the relationship between RAS medications and cognition as a method of promoting more informed choice when RAS medications are needed. ACE has been reported to degrade Aβ which comprises the neuritic plaques associated with AD suggesting and ACEI could promote Aβ plaque accumulation.⁴ ACEI also increase ang I which can lead to an increase in MasR promoting vasodilation, antiinflammation and antioxidative effect.⁷ In this study we focus on ACEI and ARB, which are often used interchangeably for a number of medical conditions. In addition to HTN, RAS medications are used for diabetes (DM), chronic kidney disease, cardiovascular events, and proteinuria.^7–11 Both ACEI and ARB decrease the activity of Angiotensin II which causes vasoconstriction and elevated blood pressure (BP), cerebral vascular dysfunction through vascular remodeling and inflammation, and Aβ production.^7,11–14 ACEI inhibits ACE, which decreases the conversion from Angiotensin I to Angiotensin II, while ARB bind to the Angiotensin II receptor type 1 (AT1R) to block the activity of Angiotensin II and subsequently causes vasodilation and reduced blood pressure.^11,12

Despite these theoretical benefits, studies comparing the impact of ARBs and ACEIs on cognition in older adults have produced inconsistent findings. Some studies suggest ARBs may confer greater benefits in slowing cognitive decline or preserving hippocampal volume compared to ACEIs.¹⁵ Cognitive outcomes assessed in prior studies include episodic memory, semantic memory, delayed recall, verbal memory, executive function, and visuospatial ability.^8,10,15–28 The variability in findings across previous studies may be attributed to differences in study designs, participant characteristics, and cognitive assessment methods. Importantly, many prior studies have focused on individual cognitive domains rather than a measure overall cognitive performance. This study employs a standardized battery of 19 cognitive tests and calculating a composite global cognitive function score. While this represents a comprehensive test of global cognitive function, other screening tests of global cognitive function are accessible during clinical assessments when antihypertensive medications are being considered. In this study we compare global cognitive function over 9 years among ARB, ACEI and non-RAS drug users.

Methods

Study population

This prospective longitudinal study included 2019 study participants from the Rush Alzheimer's Disease Center Memory and Aging Project cohort. The Memory and Aging Project is a longitudinal, epidemiologic clinical-pathologic cohort study approved by the Institutional Review Board of Rush University Medical Center that started in 1997.²⁹ There were 2247 available participants in the baseline year (1997). For this study, participants were excluded if there was missing information on baseline global cognitive function, and if they had a history of head injury. Participants were followed for up to 9 years from 1997 to 2006 to assess changes in global cognitive function.

Exposure and outcome

Exposure variables for this study were participant use of RAS drugs at the baseline year (1997) in 3 categories: ARB, ACEI, and control group (non-RAS group). Participants supplied all prescribed medications, including vitamins, supplements, and over-the-counter agents, which were taken within the last 2 weeks prior to the evaluation. The containers were directly visualized, and the medications were documented. The outcome variable was a validated Z score of global cognitive functions at a 9-years follow-up period. Global cognitive function was measured by a composite score of 19 tests of cognitive function including episodic memory (word list, word list recall, word list recognition, East Boston immediate recall, East Boston delayed recall, Logical memory immediate, and Logical Memory II delayed); semantic memory (15 items Boston naming, category fluency, 10 items reading test); working memory domain (digits forward, digits backward, and digit ordering); visuospatial ability/perceptual orientation domain (line orientation, 16 items progressive matrices); and perceptual speed domain (number comparison, Stroop color naming, Stroop word reading).²⁹

Covariates

Covariates were selected at baseline 1997: social determinants of health (gender, age, years of education, smoking), physical measures (systolic BP, diastolic BP, body mass index) and medical conditions (depression, and medical conditions summary).²⁹ Medical conditions summary is a composite measure of HTN, DM, heart disease, cancer, thyroid disease, head injury with loss of consciousness and stroke. Head injury with loss of consciousness was excluded as head injury is associated with cognitive impairment and having head injury with loss of consciousness could confound the outcome of cognitive function.³⁰

Statistical analysis

Our study examined differences in means of global cognitive function at 9-years follow-up among the RAS drug categories. Univariate analysis, bivariate analysis, and logistic regression analysis were used to analyze means of baseline age, years of education, medical conditions summary, vascular risks, baseline characteristics, and global cognitive functions among RAS groups at baseline. Analysis of covariance (ANCOVA) was used to compare means of global cognitive function at 9-years follow-up which was regressed by adjusting a regression slope of global outcome cognitive functions residuals on its baseline effects on RAS drugs to eliminate baseline cognitive function effect on RAS drugs.^31,32 Our study includes an unadjusted model (Model A), a model with additional adjustments for social determinants of health (baseline age, gender, race, ethnicity, years of education, smoking) (Model B), and a fully adjusted model adjusted for social determinants of health and medical factors (depression, and medical conditions summary) (Model C). Model A: global cognitive function at 9-year = baseline cognitive function + types of RAS drugs (ARB, ACEI, and non RAS). Model B: global cognitive function at 9-year = baseline cognitive function + types of RAS drugs (ARB, ACEI, and non RAS)) + baseline age + gender + race + ethnicity + years of education + smoking. Model C: global cognitive function at 9-year = baseline cognitive function + types of RAS drugs (ARB, ACEI, and non RAS)) + baseline age + gender + race + ethnicity + years of education + smoking + depression, and medical conditions summary. We utilized longitudinal study from 1997 to 2006 for 9 years with adjusted for baseline global cognitive function. Median follow up years was 4 years, mean follow up was 4.2 years, mode was 4 years and range of follow up was 0 to 9 years.

In an addition analysis, repeated 3 yearly analyses of global cognitive function were conducted until 9 years follow-up. Missing information on global cognitive function was handled with hot deck imputation.³³ Final analyses included all available data with the concept of intent to treat analysis principle. Logistic regression analysis was carried out to find an association between participants’ loss to follow-up (dropout) at 9 years with RAS drugs and baseline global cognitive function. We explored the differences in global cognitive functions and RAS drug use among gender groups, The Cox Proportional Hazards Regression model was utilized to determine the incidence of improved cognitive functions at 9 years from base line among RAS drugs and repeated baseline and global cognitive function at 9 years.^34,35 A priori power analysis showed that a sample size of n = 415, is required to achieve a statistical power of 95%, and n = 975 for power of 90%. Power analysis of our data was assessed in SAS for RAS drugs and the outcome of global cognitive functions at 9 years. A sample of 1, 245 was required to achieve a statistical power of 90%, and a sample of 1, 496 for a statistical power of 95%. In addition, by using R squared power analysis, the sample size 2, 109 with R squared 0.42 in a reduced model and R square 0.51 in a full model achieve 100% of statistical power by R square test in STATA 19 version. All statistical analyses were conducted in Statistical Analysis Software (SAS) 9.4.

Results

Baseline characteristics

Among the 2019 participants in this sample at baseline, 21.1% were taking ACEIs, 14.1% were taking ARB, while 64.8% were not taking a RAS drug (non-RAS). The baseline characteristics of the participants are listed in Table 1. Participant ages ranged from 53 to 100 years old. Approximately 75% of participants were aged 85 years or older. The mean age was similar in the ARB (81.1 ± 7.2) and ACEI (81.6 ± 7.3) groups, and slightly lower in the non-RAS group (80 ± 7.7); p = 0.0035. Years of education was similar in the ARB (14.7 ± 3.1) and ACE (14.5 ± 3.4) groups and higher in the non-RAS group (15.1 ± 3.4); p = 0.0005. In our study, a higher proportion of females (81.5%) were in the ARB group compared to all other groups; p < 0.001. The participants in this sample were primarily white (92.8%), Black (5.2%) and then other races (1.4%); p = 0.500.

Table 1.

Characteristics of participants among RAS categories from the Rush Alzheimer's Disease Center Memory and Aging Project cohort (1997–2006).

(n/%)	ARBsn = 297 (14.1%)	ACEIsn = 445 (21.1%)	Non-RASn = 1367 (64.8%)	p
Age, years (mean, SD) **	81.1 ± 7.2	81.6 ± 7.3	80.0 ± 7.7	0.004
Years of education (mean, SD) ** ^§	14.7 ± 3.1	14.5 (3.4)	15.1 (3.4)	<0.001
Gender *** ^§				<0.001
Male (550/26.08)	53 (17.9)	150 (33.8)	347 (25.4)
Female (1548/73.40)	242 (81.5)	292 (65.6)	1, 014 (74.2)
Race ^§				0.500
White (1958/92.84)	272 (91.6)	410 (92.1)	1276 (93.3)
Black (110/5.22)	16 (5.4)	28 (6.3)	66 (4.8)
Other (30/1.42)	7 (2.4)	4 (0.90)	16 (1.4)
Ethnicity				0.500
Hispanic (107/5.07)	20 (6.7)	23 (5.2)	64 (4.7)
Non-Hispanic (1991/94.40)	275 (92.6)	419 (94.2)	1297 (94.9)
Physical Measures (mean, SD)
Systolic Blood Pressure mmHg ** ^§§	137.3 ± 18.1***	136.0 ± 19.3	133.84 ± 17.7	0.001
Diastolic Blood Pressure mmHg ^§§	74.4 ± 12.0	73.7 ± 12.1	74.91 ± 11.0	0.125
Body Mass Index (kg/m²) *** ^§§	29.2 ± 6.2***	28.3 ± 5.8	26.51 ± 4.9	<0.001
Medical conditions (n, /%)
Hypertension ***^§ (1105/52.39)	253 (85.2)	374 (84.0)	478 (35.0)	<0.001
Diabetes ***^§ (280/13.28)	65 (21.9)	104 (23.4)	111 (8.1)	<0.001
Stroke**^§§§				0.003
Highly Probable (28/1.33)	4 (1.4)	5 (1.1)	19 (1.4)
Probable (145/6.88)	22 (7.4)	54 (12.1)	69 (5.1)
Possible (105/4.98)	16 (5.4)	22 (4.9)	67 (4.9)
Not present (1672/79.28)	234 (78.8)	329 (73.9)	1, 109 (81.1)
CHF ** ^§§§ (91,4.31)	20 (6.7)	30 (6.7)	41 (3.0)	0.001
Depression				0.284
Highly probable	3 (1.0)	5 (1.1)	7 (0.5)
Probable	11 (3.7)	10 (2.2)	25 (1.8)
Possible	15 (5.1)	18 (4.1)	57 (4.2)
Not present (1947/ 92.80)	268 (90.2)	411 (92.6)	1268 (93.4)
Smoking (n, %)				0.24
Never (1216/57.66)	178 (59.9)	229 (51.5)	809 (66.53)
Former (822/38.98)	113 (38.1)	198(44.5)	511 (37.38)
Current (52/2.47)	4 (1.4)	14 (3.1)	34 (2.49)
Medical condition summary (mean ± SD) *** ^§	1.8 ± 1.0	1.8 ± 0.9	1.09 ± 1.0	<0.001
Vascular risks *** (mean ± SD) ^§	1.5 ± 0.7	1.6 (0.8)	0.8 (0.7)	<0.001
Medication Utilization (n, %)
Polypharmacy (5 or more) (1,561, 74.02) ***	278 (17.8)	413 (26.46)	870 (55.7)	<0.001
Diuretics *** 690 (32.72)	169 (24.5)	210 (30.4)	311 (45.1)	<0.001
Beta Blockers (587, 27.83)	109 (18.6)	157 (26.8)	321 (54.7)	<0.001
Calcium Channel Blocker (475, 22.52)	103 (21.6)	125 (26.3)	247 (52.0)	<0.001
Other antihypertensive (24,1.14)	6 (25.0)	10 (41.7)	8 (33.3)	0.004
Global Cognitive Functions (mean ± SD)
Baseline	0.05 ± 0.6	−0.09 ± 0.7	0.01 ± 0.7	0.350

* < 0.05, ** < 0.01 *** < 0.0001, ^§: Missing ≤ 20 ^§§: Missing = 20–100 ^§§§: > 100.

Systolic blood pressure at baseline was 137.3 ± 18 in the ARB group, 136.0 ± 19 in the ACEI, and 133.8 ± 18 in the non-RAS group; p = 0.0012. BMI was highest in the ARB group (29.2), followed by ACEI (28.3, then non-RAS group (26.5), p ≤ 0.001. Most medical conditions (HTN, Diabetes, stroke, CHF) were more prevalent in the ARB and ACEI group compared to the non-RAS group. The medical condition summary was similar in the ARB (1.8 ± 1) and ACEI group (1.8 ± 1) and lower in the non-RAS group (1.09 ± 0.9). Vascular risks were more prevalent among ACEI (1.6 ± 0.8) and ARB (1.5 ± 0.7) groups compared to the non-RAS group (0.8 ± 0.7)), p < 0.0001. The prevalence of polypharmacy (5 or more medications) was 55.7% in the non-RAS group, 26.5% in the ACEI group, and 17.8% in the ARB group; p < 0.0001 The use of other antihypertensive agents beyond RAS medications was higher in the non-RAS group, however the use of these agents was more prevalent in the ACEI group compared to the ARB group, p ≤ 0.001. Baseline global cognitive function was not statistically significantly different among the groups: 0.046 ± 0.58 in the ARB group, −0.089 ± 0.67 in the ACEI group and 0.013 ± 0.67 in the non-RAS group; p = 0.35.

Global cognition results at 9 years follow-up

Table 2 describes with differences in the means of global cognitive function at 9 years follow-up among the RAS drug categories. Our results comparing ARB use to ACEI use did not result in statistically significant differences in global cognitive function at 9 years follow-up. Participants on ARB had reduced scores of global cognitive functions at 9 years compared with those in ACEI in the unadjusted model (Model A). After adjusting for Social determinants (Model B) and Medical factors (Model C), participants in the ARB groups had increased scores of global cognitive functions at 9 years compared with those in the ACEI group (LS-means: 0.015, 95% CI [−0.175, 0.205]). Similarly, participants taking any RAS drug (ARB versus non-RAS (LS-means: 0.060, 95% CI [−0.109, 0.229]) and ACE versus non-RAS [LS-means: 0.045, 95% CI [−0.094, 0.185]) had better global cognitive functions than those of non-RAS drugs (Table 2).

Table 2.

Differences in global cognitive functions at 9 years by RAS drug use.

Baseline n = 2, 109	Model A	Model B	Model C
Baseline n = 2, 109	LS-means (95% CI)	LS-means (95% CI)	LS-means (95% CI)
ARB	−0.38 (−0.54, −0.23) ***	−0.35 (−0.50, −0.20) ***	−0.34 (−0.50, −0.19) ***
ACEI	−0.36 (−0.48, −0.24) ***	−0.36 (−0.48, −0.25) ***	−0.36 (−0.48, −0.24) ***
non-RAS	−0.39 (−0.46, −0.33) ***	−0.40 (−0.4632, −0.34) ***	−0.40 (−0.47, −0.34) ***
ARB versus ACEI	−0.02 (−0.22, 0.17)	0.02 (−0.17, 0.20)	0.01 (−0.18, 0.20)
ARB versus non-RAS	0.01 (−0.16, 0.18)	0.06 (−0.11, 0.22)	0.06 (−0.11, 0.23)
ACEI versus non-RAS	0.03 (−0.1024, 0.17)	0.04 (−0.09, 0.17)	0.05 (−0.09, 0.19)
ARB	−0.38 (−0.54, −0.23) ***	−0.34 (−0.49, −0.19) ***	−0.35 (−0.50, −0.19) ***
ARB versus non-RAS	0.003 (−0.16, 0.17)	0.05 (−0.1138, 0.21)	0.05 (−0.12, 0.21)
ACEI	−0.36 (−0.50, −0.24) ***	−0.36(−0.48, −0.25) ***	−0.36 (−0.48, −0.24) ***
ACE versus non-RAS	0.03 (−0.10, 0.16)	0.03 (−0.10, 0.16)	0.03 (−0.10, 0.17)

* < 0.05, ** < 0.01, *** < 0.0001, Model A: ARBs + base line cognitive function; Model B: A + Social determinants; Model C: B + medical factors.

Global cognition results in 3-year intervals

In the repeated 3-year analysis with an observed period at 3-year follow up, 6-year follow up, and 9-year follow-up (Table 3), there were no statistically significant difference when ARB use was compared to ACEI use at 3 years (LS-means: 0.105, 95% CI [−0.014, 0.223]), 6 years (LS-means: 0.123, 95% CI [ −0.030, 0.2767]), and 9 years (LS-means: 0.043, 95% CI [−0.201, 0.288) in the fully adjusted model (Model C). In the fully adjusted model (Model C), ARB use compared to non-RAS use was associated with higher global cognitive function at 3 years (LS-means: 0.035, 95% CI [−0.070, 0.139]), 6 years (LS-means: 0.098, 95% CI [−0.036, 0.231]) and 9 years (LS-means: 0.112, 95% CI [−0.099, 0.323]), while ACEI versus non-RAS was associated initially with lower global cognition at 3 years (LS-means: −0.082, 95% CI [−0.169, 0.005]) and 6 years (LS-means:−0.045, 95% CI [−0.156, 0.065), but with higher cognition at 9 years 0.037, 95% CI [−0.139, 0.213]). Again, these results were not statistically significantly different (Table 3).

Table 3.

Repeated measurements of global cognitive functions by RAS drug at 3-year intervals.

Baseline n = 2, 109	At 3 years (n = 1, 498)	At 6 years (n = 984)	At 9 years (n = 561)
Baseline n = 2, 109	LS-means (95% CI)	LS-means (95% CI)	LS-means (95% CI)
Model A
ARB	0.01 (−0.07, 0.09)	−0.08 (−0.19, 0.02)	0.32 (−0.48, −0.16) **
ACEI	−0.10 (−0.16, −0.04) **	−0.21 (−0.28, −0.13) ***	−0.36 (−0.48, −0.24) ***
non-RAS	−0.01 (−0.04, 0.02)	−0.16 (−0.21, −0.12) ***	−0.39 (−0.46, −0.33) ***
ARB versus ACEI	0.11 (−0.01, 0.22)	0.12 (−0.03, 0.28)	0.04 (−0.20, 0.29)
ARB versus non-RAS	0.02 (−0.08, 0.12)	0.08 (−0.05, 0.21)	0.08 (−0.14, 0.29)
ACEI versus non-RAS	−0.09 (−0.17, −0.00) *	−0.04 (−0.15, 0.06)	0.03 (−0.14, 0.20)
Reference baseline global cognitive function (mean square)	0.059	2.75 ***	33.44 ***
Model B
ARB	0.02 (−0.06, 0.10)	−0.07 (−0.17, 0.03)	−0.29 (−0.45, −0.13) **
ACEI	−0.10 (−0.15, −0.04) **	−0.21 (−0.29, −0.14) ***	−0.36 (−0.4841, −0.25) ***
None	−0.01 (−0.04, 0.02)	−0.17 (−0.21, −0.12) ***	−0.40 (−0.46, −0.33) ***
ARB versus ACEI	0.12 (−0.001, 0.23)	0.14 (−0.01, 0.29)	0.08 (−0.16, 0.31)
ARB versus non-RAS	0.03 (−0.07, 0.14)	0.10 (−0.03, 0.23)	0.11 (−0.10, 0.31)
ACEI versus non-RAS	−0.08 (−0.16, −0.001) *	−0.05 (−0.15, 0.06)	0.03 (−0.13, 0.20)
cognitive function (mean square)	1.00 **	2.39 ***	6.28 ***
Model C
ARB	0.02 (−0.06, 0.10)	−0.07 (−0.17, 0.03)	−0.29 (−0.45, −0.13) **
ACEI	−0.09 (−0.16, −0.03) *	−0.21 (−0.29, −0.13) ***	−0.36 (−0.49, −0.24) ***
non-RAS	−0.01 (−0.05, 0.02)	−0.17 (−0.21, −0.12) ***	−0.40 (−0.47, −0.33) ***
ARB versus ACEI	0.12 (−0.001, 0.23)	0.14 (−0.01, 0.29)	0.08 (−0.16, 0.31)
ARB versus non-RAS	0.03 (−0.07, 0.14)	0.10 (−0.04, 0.23)	0.1121 (−0.10, 0.32)
ACEI versus non-RAS	−0.08 (−0.17, 0.005)	−0.05 (−0.16, 0.07)	0.04 (−0.14, 0.21)
	3.95 **	0.68 *	0.32

* < 0.05, ** < 0.01, *** < 0.0001, Model A: ARBs + base line cognitive function; Model B: A + Social determinants; Model C: B + medical factors.

Cox proportional hazard model

In the Cox proportional hazard regression analysis (Table 4) results were not statistically significant. Participants in the ARB group were 1.2 times more likely to have improved global cognitive function (HR: 1.2, 95% CI [0.65, 2.07]) at 9 years follow-up compared to those in the ACEI group (Model A). In the full adjusted model (Model C), participants in the ARBs group were 1.2 times more likely to have improved global cognitive function (HR: 1.2, 95% CI [0.69 2.25]) at follow-up 9 years compared to those of ACEIs. In the fully adjusted model participants on ARB compared to non-RAS were 1.0 times more likely to have improved global cognition (HR: 1.0, 95% CI [0.63, 1.72]), while participants in ACEIs groups compared to the non-RAS group were 83% less likely to have improved global cognitive function at 9-year follow-up.

Table 4.

Proportional Hazard Models for the incident of improved global cognitive function at 9 years follow-up.

	Model A(HR, 95% CI)	Model B(HR, 95% CI)	Model C(HR, 95% CI)
ARBs versus ACEIs	1.2 (0.7, 2.1)	1.3 (0.7, 2.3)	1.2 (0.7, 2.2)
ARBs versus non-RAS	0.8 (0.5, 1.3)	1.0 (0.6, 1.6)	1.0 (0.6, 1.7)
ACEIs versus non-RAS	0.7 (0.5, 1.1)	0.8 (0.5, 1.0)	0.8 (0.5, 1.3)

* < 0.05, ** < 0.01, *** < 0.0001, Model A: ARBs + base line cognitive function; Model B: A + Social determinants; Model C: B + medical factors; HR: hazard ratio.

Exploratory analysis

Further exploration among the cognitive domains of episodic memory, semantic memory, working memory domain, visuospatial ability/perceptual orientation domain, and perceptual speed domain resulted in statistically insignificant differences when ARB was compared to ACEI at 9 years follow-up (not shown). When the ACEI group was compared to the non-RAS group, the ACEI group had significant improvements in working memory compared to the non-RAS group at 9 years when adjusted for social factors, physical measures, and medical factors (LS-mean 0.157, 95% CI 0.002, 0.313). On the other hand, the ARB group had non-significant improvement in working memory (LS-means: 0.082497, 95% CI [−0.102269, 0.267263]) compared to non-RAS groups. There were no significant differences among genders.

Discussion

In this study, participants in the ARB group had better global cognitive function at 9 years compared with those in the ACEI group, although the results were not statistically significant. Global cognitive function is a comprehensive measurement of an individuals’ cognitive function that can be examined using cognitive tests that can be administered in during clinical care among practitioners who would prescribe antihypertensive agents. Such tests are helpful at the population screening level to detect individuals with cognitive impairment and to track their progress with interventions.³⁶ Our findings on global cognition adds to the results from other studies by Ouk, Edward, Olschewski, Gouveia, and Ahmed et al.^{8,10,15–20,22,23,27,28} that found that ARBs improve cognitive function. However, in these studies the measures of cognitive functions and comparisons were diverse. In a prospective study with 3.2 year follow up, Moran et al. found a slow rate of brain atrophy on MRI in participants on ARB compared to those on ACEI, though there was no significant different in neuropsychological assessments in participants on either ARB or ACEI group.¹⁸ A retrospective study by Deng et al. showed that ARBs were associated with lower risk of progression to dementia compared to ACE and other classes of antihypertensive agents.³⁷ In a study by Li and colleagues, a reduced incidence of dementia was noted in a predominantly male cohort (97%) who were taking ARBs compared to the ACEI lisinopril (HR: 0.81, 95% CI [0.68 to 0.96]). However, Li et al. used the outcome of incident dementia, nursing home admission, or death instead of cognitive measurements such as global cognition. In our study, there were no significant differences among genders and our cohort was predominantly female (73%).

A number of studies have examined ARB use in terms of specific cognitive domains or specific neuroimaging findings. Edward et al. found that participants on ARBs had larger hippocampal volumes (R2 = 0.83, p = 0.05) and brain parenchymal fraction (R2 = 0.83, p = 0.01), which was associated with significantly better performance on tests of episodic and verbal memory, language, and executive function than those with ACEI. Ouk et al. found that participants with ARBs had better delayed recall scores (1.094, 95%CI [1.007, 1.188]) compared to those with ACEIs. Hajjar and Nation et al. found that less amyloid deposition in brain autopsy in participants with ARBs and accumulation in CSF in those with ARBs compared to other antihypertensive including ACEI and no antihypertensives.^15,17 Cosarderelioglu et al. found that ARBs lowered Aβ burden in participants with normal cognitive functions but not in those with AD (cognitive performance measured by 19 tests (five cognitive domains, including episodic memory, semantic memory, working memory, perceptual speed, and visuospatial ability)).²²Oliveira et al found improvement in cognitive functions among late-onset AD participants who was on ACEI treatment (APOE ε4 non-carriers) or ARBs (mostly APOE ε4 carriers) at 1 year follow-up with adjustment of APOE ε4 carrier status and genotypes of the ACE insertion/deletion polymorphism.²¹ However, Oliveira et al. assessed cognitive functions through caregivers’ impressions rather than the battery of cognitive tests.²¹

In our study, participants taking any RAS drug (ARB versus non-RAS (LS-means: 0.0601, 95% CI [−0.1086, 0.2287]) and ACE versus non-RAS [LS-means: 0.0454, 95% CI [−0.0943, 0.1851]) had better global cognitive functions than those of non-RAS drugs. This is consistent with other studies comparing RAS drugs to placebo or non-RAS antihypertensives. A longitudinal study by Wharton et al. showed better cognitive function based upon digit Symbol (significant-add p value or CI) and non-significantly (logical Memory delayed, Boston naming category fluency (animals), CDR-SOB) in participants taking RAS compared to those of non-RAS over 5 years.²⁶

Anderson et al. found that participants taking ARBs were more likely to have reduced cognitive impairment [OR: 0.97, 95% CI (0.81–1.17), p = 0.76] compared with placebo during 54 months follow-up.²⁴ It is important to note that there are a number of studies that show no relationship between the RAS class of antihypertensive medication and incident cognitive decline or dementia.^38,39

A number of factors could have contributed to the findings in this cohort. There were notable differences between the ARB and ACEI groups. While the baseline age and medical condition summary were similar among the two groups, the ARB group had more females, less polypharmacy, slightly higher education and slightly lower vascular risk than the ACEI group. The finding that ARB users had less polypharmacy was an interesting finding that requires further investigation to determine if use of ARB is protective in lessening the prevalence of polypharmacy. Additional adjustments to Model C for polypharmacy and vascular risk yielded similar results when ARB was compared to ACEI [LS-Means 0.015445 95% CI −0.175013 0.205904]. In our study, there were no significant differences among genders.

There were notable differences between the RAS groups (ARB and ACEI groups) when they were compared to the non-RAS groups. There was a higher prevalence of medical conditions such as HTN, DM and CHF in the RAS groups compared to the non-RAS group.

To our knowledge, our study is one of the first studies to investigate global cognitive function over time with respect to the effect of base line RAS drugs. Our study measured global cognitive functions as a composite of 19 tests of cognitive functions rather than separated variables. The findings of non-significant reduced in global cognitive function in the ARBs group compared with those in ACEIs could be explained by more recent observations that RAS drugs can be categorized by their ability to cross the brain barrier. Centrally acting RAS drugs that cross the BBB inhibit angiotensin II at the AT1R (ARB) and reduce the angiotensin II production (ACEI) which limits vasoconstriction, oxidative stress, neuroinflammation and vascular remodeling, all of which are associated with neurodegeneration.^7,19,40,41 In this study we did not look at central versus peripherally acting RAS drugs. A recent meta-analysis of 14 cohorts from 6 countries (n = 12,849), found that centrally acting RAS (c-RAS) drugs that cross the BBB were associated with less memory decline over 3 years compared to peripherally acting RAS (p-RAS) drugs that do not cross the BBB.⁴²

There were no significant associations between the study participant who dropped out by 9 years follow-up those who remained in the study with respect to use of RAS drugs. However, those who dropped out had lower global cognitive function than those who remained in the study there was a significant association between those who dropped out and base line global cognition functions (Supplemental Tables).

Strengths and limitations of study

Among several strengths of this study, our study measured global cognitive function as a composite z scores of 19 tests of cognition and was the first study to apply this composite global cognitive function score to compare the effect of RAS drugs. We added to the evidence from previous studies of the use of RAS by examining the effects of ARB use to ACEI use and comparing both ARB or ACEI use to controls (non-RAS drugs) over a 9-year follow-up period in the same cohort. We applied ANCOVA where base line cognitive function was adjusted at 9 years follow-up. This increased the precision of the comparison by including covariates as a source of variation, and hence reduced the error of variance, increased the power and narrowed the CI. Furthermore, we limited information bias by excluding participants who did not have baseline global cognitive function. All of this strengthened the internal validity of the findings of the study.

There were a number of limitations in our study. Given the limited number of participants who were not white, our results are less generalizable to other groups such as Asians, Blacks, and Hispanics. Additionally, the survey design of this cohort relied upon participants to provide information about medication use and medical conditions, which could have resulted in information bias and recall bias regarding RAS drug use and medical conditions. We did not examine dose and duration of medication use.

Our findings were consistent with previous studies that showed that ARBs were associated with improved cognitive function.^{19,23,25,43–45} The findings did not establish causality as we did not investigate the incidence of change in cognitive function or onset of dementia, we examined differences in means of global cognitive function over time. Our study achieved specificity in agreeing with the findings of previous studies of RAS improvement on cognition.⁴⁶

Conclusion

The findings of our study add to current knowledge of the relationship between RAS drugs and cognitive function in aging although the findings did not reach statistical significance. Further longitudinal studies with larger samples and in more diverse racial and ethnic groups would be more generalizable. Additionally further studies are needed to determine if there are benefits when RAS medications are classified by their ability to cross the blood brain barrier.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251365541 - Supplemental material for A comparison of drugs targeting the renin-angiotensin system and global cognition in aging

Supplemental material, sj-docx-1-alz-10.1177_13872877251365541 for A comparison of drugs targeting the renin-angiotensin system and global cognition in aging by Claudene J George, Kay Thwe Kyaw, Charles B Hall, Erica F Weiss, Joe Verghese and Peter Abadir in Journal of Alzheimer's Disease

Footnotes

Acknowledgments

The authors have no acknowledgments to report.

Ethical considerations

The Memory and Aging Project is a longitudinal, epidemiologic clinical-pathologic cohort study approved by the Institutional Review Board of Rush University Medical Center that started in 1997.

Author contributions

Claudene George: Conceptualization; Data curation; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision; Writing – original draft; Writing – review & editing.

Kay Thwe Kyaw: Formal analysis; Methodology; Software; Writing – original draft; Writing – review & editing.

Charles Hall: Methodology; Writing – review & editing.

Erica Weiss: Methodology; Writing – review & editing.

Joe Verghese: Methodology; Writing – review & editing.

Peter Abadir: Conceptualization; Methodology; Writing – original draft; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a National Institute of Health/National Institute of Aging (grant number K23 AG062807-02) and the Johns Hopkins Older Americans Independence Center. The funding source has no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

Data can be made available with permission from the Rush Center for Memory Aging.

Supplemental material

Supplemental material for this article is available online.

References

Alzheimer's Disease Facts and Figures. https://www.alz.org/media/Documents/alzheimers-facts-and-figures.pdf (2024).

National Health and Nutrition Examination Survey 2017–March 2020 Prepandemic Data Files – Development of Files and Prevalence Estimates for Selected Health Outcomes. Hyattsville, MD: https://doi.org/10.15620/cdc:106273, 2021.

National Center for Health Statistics Hypertension. https://www.cdc.gov/nchs/fastats/hypertension.htm (2024).

van Dalen

Marcum

Gray

, et al. Association of angiotensin II-stimulating antihypertensive use and dementia risk: post hoc analysis of the PreDIVA trial. Neurology 2021; 96: e67–e80. 20201105.

den Brok

MGHE

van Dalen

Abdulrahman

, et al. Antihypertensive medication classes and the risk of dementia: a systematic review and network meta-analysis. J Am Med Dir Assoc 2021; 22: 1386–1395.e1315.

Schroevers

Hoevenaar-Blom

Busschers

, et al. Corrigendum to “Antihypertensive medication classes and risk of incident dementia in primary care patients: a longitudinal cohort study in the Netherlands” [The Lancet Regional Health-Europe, Volume 42, July 2024, 100927]. Lancet Reg Health Eur 2024; 47: 101132.

Cosarderelioglu

Nidadavolu

George

, et al. Brain renin-angiotensin system at the intersect of physical and cognitive frailty. Front Neurosci 2020; 14: 586314.

Ouk

Rabin

, et al. The use of angiotensin-converting enzyme inhibitors vs. angiotensin receptor blockers and cognitive decline in Alzheimer's disease: the importance of blood-brain barrier penetration and APOE ε4 carrier status. Alzheimers Res Ther 2021; 13: 43.

Hao

Wang

Guo

, et al. Effects of ACEI/ARB in hypertensive patients with type 2 diabetes mellitus: a meta-analysis of randomized controlled studies. BMC Cardiovasc Disord 2014; 14: 148.

10.

N-C

Lee

Whitmer

, et al. Use of angiotensin receptor blockers and risk of dementia in a predominantly male population: prospective cohort analysis. Br Med J 2010; 340: b5465.

11.

Hill

RD VP

. Angiotensin II Receptor Blockers (ARB). Treasure Island, FL: StatPearls Publishing, 2024.

12.

Wiesmann

Roelofs

van der Lugt

, et al. Angiotensin II, hypertension and angiotensin II receptor antagonism: roles in the behavioural and brain pathology of a mouse model of Alzheimer's disease. J Cereb Blood Flow Metab 2017; 37: 2396–2413.

13.

Chen

Yan

, et al. Amyloid beta: structure, biology and structure-based therapeutic development. Acta Pharmacol Sin 2017; 38: 1205–1235.

14.

Jackson

Eldahshan

Fagan

, et al. Within the brain: the renin angiotensin system. Int J Mol Sci 2018; 19: 876.

15.

Hajjar

Brown

Mack

, et al. Impact of Angiotensin receptor blockers on Alzheimer disease neuropathology in a large brain autopsy series. Arch Neurol 2012; 69: 1632–1638.

16.

Edwards

Ramirez

Callahan

, et al. Antihypertensive treatment is associated with mri-derived markers of neurodegeneration and impaired cognition: a propensity-weighted cohort study. J Alzheimers Dis 2017; 59: 1113–1122.

17.

Nation

Yew

. Older adults taking AT 1-receptor blockers exhibit reduced cerebral amyloid retention. J Alzheimers Dis 2016; 50: 779–789.

18.

Moran

Xie

Poh

, et al. Observational study of brain atrophy and cognitive decline comparing a sample of community-dwelling people taking angiotensin converting enzyme inhibitors and angiotensin receptor blockers over time. J Alzheimers Dis 2019; 68: 1479–1488.

19.

Olschewski

Nazarzadeh

Lange

, et al. The angiotensin II receptors type 1 and 2 modulate astrocytes and their crosstalk with microglia and neurons in an in vitro model of ischemic stroke. BMC Neurosci 2024; 25: 29.

20.

Gouveia

Fonseca

Silva

, et al. Intranasal irbesartan reverts cognitive decline and activates the PI3K/AKT pathway in an LPS-induced neuroinflammation mice model. Int Immunopharmacol 2024; 128: 111471.

21.

Oliveira

FFD

Almeida

SSD

Chen

, et al. Pharmacogenetics of angiotensin modulators according to APOE- μ4 alleles and the ACE insertion/deletion polymorphism in Alzheimer's disease. Acta Neuropsychiatr 2023; 35: 346–361.

22.

Cosarderelioglu

Nidadavolu

George

, et al. Angiotensin receptor blocker use is associated with upregulation of the memory-protective angiotensin type 4 receptor (AT4R) in the postmortem brains of individuals without cognitive impairment. Geroscience 2023; 45: 371–384.

23.

Ahmed

Ishrat

Pillai

, et al. Role of angiotensin system modulation on progression of cognitive impairment and brain MRI changes in aged hypertensive animals – A randomized double- blind pre-clinical study. Behav Brain Res 2018; 346: 29–40.

24.

Anderson

Teo

Gao

, et al. Renin-angiotensin system blockade and cognitive function in patients at high risk of cardiovascular disease: analysis of data from the ONTARGET and TRANSCEND studies. Lancet Neurol 2011; 10: 43–53.

25.

Takeda

Sato

Takeuchi

, et al. Angiotensin receptor blocker prevented β-amyloid-induced cognitive impairment associated with recovery of neurovascular coupling. Hypertension 2009; 54: 1345–1352.

26.

Wharton

Goldstein

Zhao

, et al. Modulation of renin-angiotensin system may slow conversion from mild cognitive impairment to Alzheimer's disease. J Am Geriatr Soc 2015; 63: 1749–1756.

27.

Barthold

Joyce

Wharton

, et al. The association of multiple anti-hypertensive medication classes with Alzheimer's disease incidence across sex, race, and ethnicity. PLoS One 2018; 13: e0206705.

28.

Schroevers

Eggink

Hoevenaar-Blom

, et al. Antihypertensive medication classes and the risk of dementia over a decade of follow-up. J Hypertens 2023; 41: 262–270.

29.

Bennett

Schneider

Buchman

, et al. Overview and findings from the rush Memory and Aging Project. Curr Alzheimer Res 2012; 9: 646–663.

30.

Shively

Scher

Perl

, et al.

Dementia resulting from traumatic brain injury: what is the pathology?

Arch Neurol 2012; 69: 1245–1251.

31.

Silknitter

Wisnowski

Montgomery

. The analysis of covariance: a useful technique for analysing quality improvement experiments. Qual Reliab Eng Int 1999; 15: 303–316.

32.

Owen

Froman

. Uses and abuses of the analysis of covariance. Res Nurs Health 1998; 21: 557–562.

33.

Andridge

Little

. A review of hot deck imputation for survey non-response. Int Stat Rev 2010; 78: 40–64.

34.

Hosmer Jr

Lemeshow

May

. Applied Survival Analysis, Regression Modelling of Time-to-Event Data. 2nd Edition. Hoboken, NJ: John Wiley & Sons, Inc., 2008.

35.

Allision

. Survival Analysis Using SAS: A Particle Guide. 2nd Edition. Cary, NC: SAS Institute Inc., 2010.

36.

Riello

Rusconi

Treccani

. The role of brief global cognitive tests and neuropsychological expertise in the detection and differential diagnosis of dementia. Front Aging Neurosci 2021; 13: 648310.

37.

Deng

Jiang

Wang

, et al. Angiotensin receptor blockers are associated with a lower risk of progression from mild cognitive impairment to dementia. Hypertension 2022; 79: 2159–2169.

38.

Peters

Yasar

Anderson

, et al. Investigation of antihypertensive class, dementia, and cognitive decline: a meta-analysis. Neurology 2020; 94: e267–e281.

39.

Walker

Davies

Martin

, et al. Comparison of antihypertensive drug classes for dementia prevention. Epidemiology 2020; 31: 852–859.

40.

Kobiec

Otero-Losada

Chevalier

, et al. The renin-angiotensin system modulates dopaminergic neurotransmission: a new player on the scene. Front Synaptic Neurosci 2021; 13: 638519.

41.

Jeneson

Squire

. Working memory, long-term memory, and medial temporal lobe function. Learn Mem 2012; 19: 15–25.

42.

Moriarty

Manly

, et al. Blood-brain barrier crossing renin-angiotensin drugs and cognition in the elderly: a meta-analysis. Hypertension 2021; 78: 629–643.

43.

Zhou

Pavel

Macova

, et al. AT1 Receptor blockade regulates the local angiotensin II system in cerebral microvessels from spontaneously hypertensive rats. Stroke 2006; 37: 1271–1276.

44.

Brown

Malik

, et al. Two opposing functions of angiotensin-converting enzyme (ACE) that links hypertension, dementia, and aging. Int J Mol Sci 2021; 22: 13178.

45.

AbdAlla

Langer

, et al. ACE Inhibition with captopril retards the development of signs of neurodegeneration in an animal model of Alzheimer's disease. Int J Mol Sci 2013; 14: 16917–16942.

46.

Fedak

Bernal

Capshaw

, et al. Applying the Bradford Hill criteria in the 21st century: how data integration has changed causal inference in molecular epidemiology. Emerg Themes Epidemiol 2015; 12: 14.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB