Comparison of Model-Based Indices of Cerebral Autoregulation and Vasomotor Reactivity Using Transcranial Doppler versus Near-Infrared Spectroscopy in Patients with Amnestic Mild Cognitive Impairment

• Published in: Journal of Alzheimer's disease Vol. 56; no. 1; pp. 89 - 105
• Main Authors: Marmarelis, Vasilis Z., Shin, Dae C., Tarumi, Takashi, Zhang, Rong
• Format: Journal Article
• Language: English
• Published: London, England SAGE Publications 01.01.2017
• Subjects: Aged; Amnesia - complications; Blood Flow Velocity; Cerebrovascular Circulation - physiology; Cognitive Dysfunction - complications; Cognitive Dysfunction - diagnostic imaging; Cognitive Dysfunction - pathology; Female; Homeostasis - physiology; Humans; Male; Mental Recall - physiology; Mental Status Schedule; Middle Aged; Models, Biological; Spectroscopy, Near-Infrared; Time Factors; Ultrasonography, Doppler, Transcranial; cerebral blood flow; mild cognitive impairment; Alzheimer’s disease; cerebral circulation
• ISSN: 1387-2877; 1875-8908
• DOI: 10.3233/JAD-161004

We recently introduced model-based “physiomarkers” of dynamic cerebral autoregulation and CO2 vasomotor reactivity as an aid for diagnosis of early-stage Alzheimer’s disease (AD) [1], where significant impairment of dynamic vasomotor reactivity (DVR) was observed in early-stage AD patients relative to age-matched controls. Milder impairment of DVR was shown in patients with amnestic mild cognitive impairment (MCI) using the same approach in a subsequent study [2]. The advocated approach utilizes subject-specific data-based models of cerebral hemodynamics to quantify the dynamic effects of resting-state changes in arterial blood pressure and end-tidal CO2 (the putative inputs) upon cerebral blood flow velocity (the putative output) measured at the middle cerebral artery via transcranial Doppler (TCD). The obtained input-output models are then used to compute model-based indices of DCA and DVR from model-predicted responses to an input pressure pulse or an input CO2 pulse, respectively. In this paper, we compare these model-based indices of DVR and DCA in 46 amnestic MCI patients, relative to 20 age-matched controls, using TCD measurements with their counterparts using Near-Infrared Spectroscopy (NIRS) measurements of blood oxygenation at the lateral prefrontal cortex in 43 patients and 22 age-matched controls. The goal of the study is to assess whether NIRS measurements can be used instead of TCD measurements to obtain model-based physiomarkers with comparable diagnostic utility. The results corroborate this view in terms of the ability of either output to yield model-based physiomarkers that can differentiate the group of aMCI patients from age-matched healthy controls.