Cardiovascular

Cerebral microbleeds on susceptibility-weighted imaging, cognitive domains, and serum neurofilament light chain in adults with hypertension.

TL;DR

In adults with primary hypertension, greater CMB burden was independently associated with MCI and selected cognitive domains, while sNFL had a numerically higher AUC than CMB count for identifying MCI and adding CMB count did not improve discrimination beyond sNFL alone.

Key Findings

CMBs were detected in nearly half of hospitalized hypertensive adults, and MCI was similarly prevalent in this population.

  • Study included 173 hospitalized adults aged 31–95 years with a pre-existing diagnosis of primary hypertension.
  • CMBs were detected in 82 patients (47.4%).
  • MCI was identified in 83 patients (48.0%).
  • CMB burden was categorized as 0, 1–4, or ≥5 lesions.
  • Brain MRI with susceptibility-weighted imaging was used to detect CMBs.

Greater CMB burden was independently associated with significantly higher odds of MCI after multivariable adjustment.

  • Adjusted odds ratio for MCI with greater CMB burden: 4.53 (95% CI 1.95–10.55).
  • Models were adjusted for age, sex, education, systolic blood pressure, diabetes, and prior stroke.
  • Sensitivity analyses additionally adjusted for other cerebral small vessel disease markers, modeled CMB count as a continuous variable, and excluded participants with prior stroke.
  • Cognitive impairment was assessed using the Montreal Cognitive Assessment (MoCA).

Greater CMB burden was associated with lower scores in executive function, orientation, attention, and visuospatial function, but not with memory or language.

  • Six MoCA-derived cognitive domain z-scores were calculated and analyzed.
  • Associations survived multivariable adjustment and false-discovery-rate correction.
  • Memory and language domain scores were not significantly associated with CMB burden after correction.
  • This suggests a domain-specific rather than global pattern of cognitive vulnerability linked to CMB burden.

Cognitive domain scores did not differ according to CMB location.

  • CMB location was examined as a potential modifier of cognitive domain associations.
  • No significant differences in any cognitive domain score were found based on CMB location.
  • This finding was consistent across the multivariable-adjusted analyses.

Both CMB count and serum neurofilament light chain (sNFL) were inversely correlated with cognitive performance.

  • sNFL was measured by enzyme-linked immunosorbent assay (ELISA).
  • Higher CMB count and higher sNFL levels were each associated with worse cognitive performance.
  • The inverse correlations were observed across the study sample of 173 hypertensive adults.

sNFL alone discriminated MCI better than CMB count alone, and adding CMB count to sNFL did not improve discrimination.

  • AUC for CMB count alone for identifying MCI: 0.683.
  • AUC for sNFL alone: 0.794.
  • AUC for the combined model (CMB count + sNFL): 0.792.
  • The combined model had a significantly larger AUC than CMB count alone (p < 0.001).
  • The combined model did not differ significantly from sNFL alone (p = 0.768).

The study design was cross-sectional and single-center, and the authors identified prospective validation as a prerequisite before clinical application.

  • The study was conducted at a single center with 173 hospitalized adults.
  • Cross-sectional design precludes causal inference.
  • Authors stated: 'prospective validation is required before clinical application.'
  • Findings are described as potentially helping 'characterize cognitive vulnerability in hypertension.'

What This Means

This research suggests that in people with high blood pressure (hypertension), the presence of tiny brain bleeds called cerebral microbleeds (CMBs)—detected on a specialized type of MRI scan—is strongly linked to a higher risk of mild cognitive impairment (MCI). Nearly half of the 173 hospitalized hypertensive adults studied had CMBs, and nearly half had MCI. Patients with a higher number of CMBs were more than four times as likely to have MCI compared to those with none, even after accounting for age, education, blood pressure levels, diabetes, and prior stroke. Importantly, higher CMB burden was specifically associated with worse performance in thinking skills like executive function, orientation, attention, and visuospatial ability, but not with memory or language—suggesting that CMBs affect some cognitive abilities more than others in this population. The study also examined a blood protein called serum neurofilament light chain (sNFL), which is released when brain nerve fibers are damaged. Both higher CMB counts and higher sNFL levels were each linked to worse overall cognition. When the researchers tested how well each measure could identify who had MCI, sNFL performed better on its own (AUC 0.794) than CMB count alone (AUC 0.683), and combining the two did not meaningfully improve accuracy beyond sNFL by itself. This research suggests that measuring sNFL through a blood test may be a more practical and informative tool than counting CMBs on MRI for identifying cognitive impairment in hypertensive patients, though the two measures capture related aspects of brain injury. Because this was a single-center study examining patients at one point in time, the authors caution that these findings need to be confirmed in larger, longer-term studies before they could be used to guide clinical decisions.

Have a question about this study?

Citation

Li Z, Zhao X, Zhou D, Zhao C. (2026). Cerebral microbleeds on susceptibility-weighted imaging, cognitive domains, and serum neurofilament light chain in adults with hypertension.. Frontiers in neurology. https://doi.org/10.3389/fneur.2026.1909512