Antihypertensive combinations linked with divergent cognitive impairment risks

In a longitudinal study of 28,057 adults aged 65 or older who received blood pressure-lowering medication in Valencia, Spain, different antihypertensive regimens were associated with markedly different probabilities of being diagnosed with cognitive impairment. Loop-diuretic monotherapy was associated with a lower hazard of impairment, while ACE-inhibitor monotherapy and selective beta-blocker monotherapy were associated with higher hazards after adjustment for age, sex and several concomitant treatments. Among people receiving combination therapy, the combination of an angiotensin II receptor blocker and a selective beta-blocker was associated with roughly a 45% lower hazard than other antihypertensive combinations.

The retrospective cohort study analyzed electronic health records from Arnau de Villanova Hospital and Hospital La Fe for 2002–2021. Participants had to have received antihypertensive treatment for at least six months, and those who developed cognitive impairment during follow-up had to have started treatment at least one year before diagnosis. The researchers identified 1,016 cases of cognitive impairment using diagnostic codes for Alzheimer disease and related disorders; the remaining 27,041 participants were classified as cognitively unimpaired. Mean follow-up was 1,438 days, or about 3.9 years.

Cox proportional-hazards models were used to compare the time to impairment diagnosis for individual drug classes used alone and in fixed combinations. Model 1 adjusted for age and sex, while Model 2 additionally adjusted for use of antidiabetics, antithrombotic agents, cardiac therapy and lipid-lowering drugs. For each monotherapy comparison, the reference group consisted of users of another antihypertensive class alone; for combination comparisons, it consisted of users of other two-drug regimens. The researchers also produced unadjusted Kaplan–Meier survival curves and tested the proportional-hazards assumption.

Loop diuretics were the only monotherapy class with a statistically significant lower adjusted hazard of cognitive impairment in both models. Their hazard ratios were 0.75 with a 95% confidence interval of 0.62–0.90 in Model 1 and 0.76 with a 95% confidence interval of 0.63–0.92 in Model 2. ACE inhibitors showed the opposite pattern: the hazard ratio was 1.35, with a 95% confidence interval of 1.09–1.66, in Model 1 and 1.33, with a 95% confidence interval of 1.08–1.63, in Model 2.

Selective beta-blockers used as monotherapy were associated with a higher hazard only after the broader adjustment. Their hazard ratio was 1.18, with a 95% confidence interval of 1.00–1.39, in the age- and sex-adjusted model, and 1.21, with a 95% confidence interval of 1.02–1.43, in the fully adjusted model. ARB monotherapy did not have a statistically significant lower hazard in either Cox model, with adjusted hazard ratios of 0.89 and a 95% confidence interval of 0.76–1.03 in Model 2. In the unadjusted survival analysis, however, ARB users had higher probability of remaining free of a cognitive-impairment diagnosis during follow-up.

The strongest combination result involved ARBs with selective beta-blockers. Relative to other antihypertensive combinations, this regimen was associated with hazard ratios of 0.52, with a 95% confidence interval of 0.28–0.96, in Model 1 and 0.54, with a 95% confidence interval of 0.29–0.99, in Model 2. Selective beta-blockers were also used more frequently by participants who remained cognitively unimpaired: 12.91% of unimpaired combination-therapy users received this combination, compared with 6.55% of impaired users. No other evaluated antihypertensive combination had a statistically significant association in the Cox models.

The results describe associations, not proof that one medication prevents cognitive decline. Exposure was inferred from prescriptions rather than confirmed adherence, and the records lacked detailed baseline blood pressure, lifestyle and socioeconomic information. The cohort also included both incident and prevalent medication users because complete pre-entry prescription histories were unavailable, potentially introducing prevalent-user and survivor biases. No mortality data were included, and participants receiving frequent medical monitoring may have been more likely to undergo opportunistic cognitive screening. These limitations make confounding by indication especially important: ACE inhibitors and beta-blockers are often prescribed for heart failure, coronary disease, atrial fibrillation or other conditions that may themselves affect cognitive outcomes.

The authors describe the findings as exploratory and hypothesis-generating, and no correction was made for multiple comparisons. Consequently, the nominal p-values should not be treated as confirmatory evidence: testing many drug classes and regimens increases the chance that at least one association will appear statistically significant. The six-month exposure requirement and one-year diagnostic lag were intended to reduce reverse causality, but they could also select particular patient groups and contribute to immortal-time bias. Independent cohorts and prospective studies that measure blood pressure, adherence, comorbidity, socioeconomic status and mortality are therefore needed before the associations can be interpreted as effects of drug selection.

Several biological mechanisms could plausibly connect antihypertensive treatment with cognition, although the study could not determine which mechanism produced its observations. Lower blood pressure may reduce cerebral vascular injury, while individual drug classes may influence neuroinflammation, oxidative stress, cerebral perfusion or neuronal survival. Loop diuretics have been investigated in relation to neuroinflammatory processes and amyloid accumulation, and ARBs may affect angiotensin-receptor signaling and cerebral blood flow. A combined ARB–beta-blocker regimen could also provide better blood pressure control by acting on the renin–angiotensin and sympathetic systems. More intensive clinical monitoring of people on combination therapy could also improve management of other illnesses, meaning the observed association may reflect healthcare utilization rather than a unique pharmacological effect. The study supports further investigation of medication choice and cognitive trajectories in older adults, but it does not establish that changing an antihypertensive regimen will prevent dementia.

Subject of Research: Associations between antihypertensive drug classes and cognitive impairment in older adults

Article Title: Impact of ARBs plus beta-blockers and other antihypertensives on cognitive impairment: a longitudinal study

Article References: Garcia-Zamora, M., García-Lluch, G., Moreno, L., Pardo, J., & Cháfer-Pericas, C. (2026). Impact of ARBs plus beta-blockers and other antihypertensives on cognitive impairment: a longitudinal study. Frontiers in Pharmacology, 17, Article 1954416. https://doi.org/10.3389/fphar.2026.1954416

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1954416

Keywords: antihypertensives, cognitive impairment, angiotensin receptor blockers, ACE inhibitors, beta-blockers, loop diuretics, hypertension, dementia prevention, observational cohort, electronic health records, Impact, ARBs