Blood Pressure Variability Linked to Cognitive Decline in High-Risk Adults

METHODOLOGY

  • Researchers conducted a pooled individual-participant analysis including 11,104 participants from two randomized controlled trials: the SPRINT-MIND (n = 9361; hypertensive adults without diabetes) and ACCORD-MIND (n = 2809; adults with type 2 diabetes).
  • Participants had baseline and follow-up cognitive assessments using the Digit Symbol Coding Test (SPRINT-MIND) or Digit Symbol Substitution Test (ACCORD-MIND). Visit-to-visit systolic BP variability was calculated using SBP readings from the 3-month visit through the final cognitive assessment.
  • Systolic BP variability was quantified using variation independent of mean (VIM) as the primary metric, with SD, coefficient of variation, and average real variability as alternative metrics.
  • The primary outcome was the longitudinal change in global cognitive performance over time, quantified as the annual rate of change in standardized DSST/DSCT Z-score.
  • Median follow-up was 3.94 years; participants contributed a mean of 12.0 BP measurements over the study period.

TAKEAWAY

  • Each 10% increment in systolic BP variation independent of mean was associated with an additional annual decline of 0.008 Z-score units (P = .003) after full adjustment including mean systolic BP.
  • Participants in the highest tertile of systolic BP variability exhibited a faster rate of cognitive decline compared with those in the lowest tertile (regression coefficient, β = −0.005 per year; P = .049).
  • Exploratory stratified analyses suggested the adverse association between BP variability and cognitive decline appeared more pronounced under intensive BP control (ACCORD-MIND β = −0.035 per year; 95% CI, −0.063 to −0.007; SPRINT-MIND β = −0.013 per year; 95% CI, −0.021 to −0.005), although the pooled three-way interaction did not reach statistical significance.
  • The spline curves demonstrated approximately linear dose-response relationships between all BP variability metrics and cognitive decline, with no evidence of nonlinearity (P for nonlinearity = .925 for VIM).

 

https://www.medscape.com/viewarticle/blood-pressure-variability-linked-cognitive-decline-high-2026a1000qjd