To INFINITY and Beyond: What Have We Learned and What Is Still Unknown About Blood Pressure Lowering and Cognition?

Rebecca F. Gottesman, MD, PhD


Circulation. 2019;140(20):1636-1638. 

Management of elevated blood pressure (BP) as a means to reduce mild cognitive impairment (MCI) and dementia is a promising area for prevention of these common, but devastating, neurologic conditions. Even before definitive clinical trial data provided evidence that BP reduction was a means for prevention, the National Academies officially recommended BP reduction as an important focus for dementia prevention.[1] When SPRINT (Systolic Blood Pressure Intervention Trial),[2] and its accompanying cognitive-focused substudy SPRINT-MIND (SPRINT Memory and Cognition in Decreased Hypertension), were completed, definitive clinical trial evidence demonstrated that aggressive BP control reduced incident MCI (hazard ratio, 0.81), as well as a combined outcome of MCI and dementia (hazard ratio, 0.85),[3] but not dementia specifically. SPRINT-MIND directly tested this question by randomizing more than 9000 individuals with hypertension to standard control (goal systolic BP <140 mm Hg) versus aggressive control (goal systolic BP <120 mm Hg). When combined with other previously conducted clinical trials, the net benefit of BP lowering on dementia reduction became significant (relative risk, 0.93 [95% CI, 0.86–1.00]).[4]

Although the SPRINT-MIND trial answered major questions relating to prevention of MCI and dementia through BP control, several unanswered questions remain. SPRINT controlled BP when evaluated at clinic visits, but was not able to evaluate more subtle or ambulatory fluctuations in BP, nor did it provide key information about neurologically vulnerable populations (such as those with prior stroke or with cerebral small vessel disease) who might have a greater, or perhaps even a lesser, benefit from BP lowering. In addition, questions remain about the mechanism by which BP reduction might impact cognitive outcomes even in a relatively short follow-up period. In the recently published SPRINT-MIND magnetic resonance imaging substudy, intensive treatment of BP over the period of 3.4 years attenuated the rise in white matter hyperintensities, an indication of cerebral small vessel disease, which are commonly associated with hypertension, as well as the decline in the total brain volumes.[5] Although these results provide some information about a mechanism by which BP management might reduce MCI or dementia, the amount of benefit on these imaging markers was relatively small, and it remains unclear how much of the cognitive effect is explained by measurable changes on brain imaging.

In this issue of Circulation, White et al present the results of the INFINITY trial (Intensive Versus Standard Blood Pressure Lowering to Prevent Functional Decline in Older People), which attempts to answer some of these remaining questions.[6] In INFINITY, 199 adults with systolic hypertension and white matter hyperintensities on magnetic resonance imaging underwent ambulatory BP monitoring, and were randomized to a 24-hour mean systolic BP of <130 mm Hg (intensive therapy) versus standard therapy (goal systolic BP <145 mm Hg). Over 3 years, despite success in achieving the desired split in BP levels, neither cognition nor gait speed trajectories differed in the 2 groups; however, consistent with the SPRINT-MIND effect, the more aggressively treated group had less progression in white matter hyperintensity volumes.

By considering ambulatory BP, INFINITY allowed the investigators to more tightly control mean 24-hour BP. Ambulatory BP monitoring provides other theoretical advantages over clinic-based assessment, including consideration of BP variability and nighttime dipping patterns; neither of these was the focus of the INFINITY randomization structure, but are important and relevant aspects of ambulatory BP patterns. In the Three-City Study, BP variability, but not mean BP, measured using ambulatory devices was associated with greater risk of dementia.[7] Higher nighttime BPs have been associated with more white matter hyperintensity progression,[8] and less nocturnal systolic BP dipping has been associated with worse executive function.[9]

INFINITY specifically recruited a population with preexisting white matter hyperintensities (volumes of white matter hyperintensities of 20.2 and 21.1 cm3 for the 2 groups); although SPRINT-MIND, as part of SPRINT, did not require white matter hyperintensities for recruitment, most individuals had at least some white matter lesion volume (mean baseline, 3 cm3; interquartile range, 1.5–6.2 cm3), but clearly had less baseline disease than INFINITY participants did. Although white matter hyperintensities are very strongly associated with hypertension, and thus it is reasonable to hypothesize that individuals with more white matter hyperintensities might have the greatest potential benefit from aggressive BP lowering, there is also a theoretical concern that some white matter lesions may be related to hypoperfusion and lower cerebral blood flow,[10] which might indicate that a group with preexisting small vessel disease might actually do worse if exposed to lower BP. This is a plausible concern given the known rightward shift in the cerebral autoregulatory curve that occurs in individuals who are chronically hypertensive; exposing a chronically hypertensive individual to a lower mean BP could theoretically lead to problems with maintaining adequate cerebral perfusion. Importantly, INFINITY showed that this population did tolerate more aggressive BP control (although not at as tight a level as was tested in SPRINT-MIND), and even found slowing of progression of white matter hyperintensities in the aggressively treated group. It still remains unknown what the optimal range of BP, or way to measure it, might be for a population with preexisting small vessel disease, but the reassuring safety data from INFINITY further support a similar approach in this population as in the rest of the population, with lower BP being more ideal.

Do these studies, taken together, provide adequate insight to answer all remaining questions regarding BP control and risk of dementia? Unfortunately, no: some questions still remain. SPRINT-MIND suggested that BP goals of <120 were optimal, but failed to find a reduction in dementia rates specifically. INFINITY similarly did not find evidence for a cognitive benefit over their 3-year follow-up period, but found a benefit with a slightly higher BP target on white matter hyperintensity progression, which itself is strongly associated with cognition. The most likely reason for the lack of a clear cognitive benefit is the relatively short follow-up period for these studies. The most compelling evidence from observational studies emphasizes that BP in midlife is more strongly associated with cognitive change, MCI, and dementia than is later-life BP.[11–13] Thus, it is possible that ideal BP management might need to start in midlife to have a maximal benefit, and most likely needs to continue for decades. This design isn't feasible to test in a clinical trial, but emphasizes the importance of evaluating both longitudinal observational data and clinical trial data to best guide prevention strategies for hypertensive adults. Most likely, treatment of hypertension to a goal systolic BP of <120 mm Hg (as outlined in SPRINT and SPRINT-MIND), starting as early (and at as young an age) as possible, is the best way to prevent late-life cognitive decline and dementia from a BP standpoint. Although midlife is typically cited as a key time period, it is highly likely that even earlier exposure to hypertension is at least as harmful,[14] if not more so. Importantly, "from a blood pressure standpoint" is one of the only current promising areas in the prevention of cognitive change, MCI, and dementia.

An additional unanswered question in the field relates to mechanism of a BP lowering effect on cognitive outcomes. Although both showed significant results, neither INFINITY nor SPRINT-MIND showed a huge benefit on white matter hyperintensity progression, despite the fact that, for INFINITY, preexisting white matter hyperintensity volumes were higher and thus risk of progression might be even greater. Hypertension impacts the brain through several mechanisms, including ones not currently easily measured. Evidence points to alterations in neurovascular function induced by hypertension, as well as changes in amyloid clearance and alterations in blood-brain barrier function, as well as reductions in cerebral blood flow as mentioned previously;[15] these may not all result in visible small vessel lesions or even visible brain atrophy. Further understanding of the mechanism by which hypertension, and therefore its management, impacts dementia and cognitive performance will allow not only for advances in prevention and treatment, but also for identification of surrogate markers that will allow for more efficient study of any potential therapeutic advances.

INFINITY adds further important insight into the importance of BP as a target in the prevention of white matter hyperintensity progression and dementia, and demonstrates the safety of more intensive BP control in individuals with preexisting small vessel disease. Several large clinical trials are being conducted to investigate the benefit of intensive BP lowering on the brain structure and function in middle-aged and older adults based on either office or ambulatory BP monitoring (NCT02913664 and NCT03354143). The additional data are likely to provide information regarding BP targets and the role of BP variability and nocturnal patterns, which may further guide best practices regarding BP screening in the context of prevention of dementia. Although at one point the questions regarding BP, its management, and its relation to dementia were infinite, tremendous progress has been made.