Assessing awareness in severe Alzheimer’s disease

Jonathan Huntley, Daniel Bor, Feng Deng, Marco Mancuso, Pedro A. M. Mediano, Lorina Naçi, Adrian M. Owen, Lorenzo Rocchi, Avital Sternin, Robert HowardFrontiers in Human Neuroscience 2023 · DOI

Abstract

There is an urgent need to understand the nature of awareness in people with severe Alzheimer's disease (AD) to ensure effective person-centered care. Objective biomarkers of awareness validated in other clinical groups (e.g., anesthesia, minimally conscious states) offer an opportunity to investigate awareness in people with severe AD. In this article we demonstrate the feasibility of using Transcranial magnetic stimulation (TMS) combined with EEG, event related potentials (ERPs) and fMRI to assess awareness in severe AD. TMS-EEG was performed in six healthy older controls and three people with severe AD. The perturbational complexity index ($PCI^{ST}$) was calculated as a measure of capacity for conscious awareness. People with severe AD demonstrated a $PCI^{ST}$ around or below the threshold for consciousness, suggesting reduced capacity for consciousness. ERPs were recorded during a visual perception paradigm. In response to viewing faces, two patients with severe AD provisionally demonstrated similar visual awareness negativity to healthy controls. Using a validated fMRI movie-viewing task, independent component analysis in two healthy controls and one patient with severe AD revealed activation in auditory, visual and fronto-parietal networks. Activation patterns in fronto-parietal networks did not significantly correlate between the patient and controls, suggesting potential differences in conscious awareness and engagement with the movie. Although methodological issues remain, these results demonstrate the feasibility of using objective measures of awareness in severe AD. We raise a number of challenges and research questions that should be addressed using these biomarkers of awareness in future studies to improve understanding and care for people with severe AD.

Introduction

In the advanced stages of Alzheimer’s disease (AD), patients often lose the ability to communicate verbally, creating a significant challenge for clinicians and caregivers. Without reliable behavioral indicators, it is difficult to determine whether a patient retains any level of conscious awareness or experiences "moments of lucidity." Historically, assessing awareness has relied on subjective reports, which are impossible to obtain in severe AD.

This pilot study investigates whether objective neurophysiological biomarkers—originally developed for assessing awareness in anesthesia and disorders of consciousness (DoC) like vegetative states—can be applied to severe AD. By moving beyond behavioral observation, researchers aim to probe the "hidden" awareness of these patients to better inform person-centered care.

Methodology

The researchers employed a multimodal approach, conducting three case-control pilot studies with a cohort of eight healthy older controls (mean age 73.4) and four patients with severe AD (mean age 80.5).

1. TMS-EEG and the Perturbational Complexity Index

Transcranial Magnetic Stimulation (TMS) was used to "ping" the brain with a magnetic pulse, while the resulting electrical activity was recorded via EEG. From these recordings, the Perturbational Complexity Index ($PCI^{ST}$) was calculated. This metric quantifies the integration and differentiation of brain activity; high complexity is a hallmark of consciousness, whereas low complexity is associated with unconscious states like deep sleep or anesthesia.

Participants underwent a visual masking paradigm to identify specific ERP markers of awareness. They were presented with "critical stimuli" (either a face or a mask) for durations of 33ms (typically below the threshold for conscious perception) or 200ms (typically consciously perceived).

Figure 1: Visual perception task paradigm. Critical stimuli (face or mask) were presented at 33 ms or 200 ms durations within a visual masking sequence.

The study focused on two primary ERP components:

  • Visual Awareness Negativity (VAN): Associated with "phenomenal consciousness" or the raw subjective experience of a stimulus.
  • Late Positivity (LP): Associated with "access consciousness," involving higher-level cognitive processing and working memory.

3. fMRI and Naturalistic Viewing

To assess higher-order engagement, a single patient and controls watched an engaging Alfred Hitchcock short film ("Bang! You’re Dead") during fMRI scanning. Independent Component Analysis (ICA) was used to determine if the patient’s brain networks (auditory, visual, and fronto-parietal) synchronized with those of healthy controls.

Results

The preliminary findings suggest a complex picture of awareness in severe AD.

Capacity for Consciousness

In the TMS-EEG study, healthy controls exhibited $PCI^{ST}$ scores between 34.4 and 44.6. In contrast, the three AD patients scored significantly lower (16.5, 20.8, and 22.7). These values hover near or just below the established threshold for consciousness (~21) used in other clinical groups, suggesting a reduced and potentially fluctuating capacity for awareness.

Phenomenal vs. Access Consciousness

The ERP results provided evidence for a dissociation between different types of awareness. Two AD patients demonstrated VAN responses similar to controls when viewing faces, suggesting that the raw, phenomenal experience of seeing might remain intact. However, their LP responses were notably diminished, indicating a breakdown in the higher-level cognitive processing required to bring that perception into working memory.

Sensory vs. Cognitive Engagement

fMRI data showed that while the AD patient had robust activation in sensory networks (auditory and visual), their higher-order fronto-parietal networks did not correlate with those of the controls. This suggests that while the patient was processing the sensory input of the movie, they may not have been following the narrative or experiencing the same level of conscious engagement as the controls.

Discussion and Future Directions

This study demonstrates the feasibility of using objective biomarkers to assess awareness in severe AD, though several challenges remain. Significant cortical atrophy in AD patients complicates TMS targeting and fMRI normalization. Furthermore, patient tolerance for long, noisy scanning sessions is a limiting factor.

Despite these hurdles, the results suggest that patients with severe AD may retain "phenomenal" awareness even when "access" consciousness is impaired. Future research should involve larger cohorts and longitudinal designs to track how these biomarkers change over the course of the disease. Investigating "moments of lucidity" through these objective lenses could provide vital insights into the internal lives of patients, ultimately improving the quality of person-centered care.

Summary updated

Abstract

There is an urgent need to understand the nature of awareness in people with severe Alzheimer's disease (AD) to ensure effective person-centered care. Objective biomarkers of awareness validated in other clinical groups (e.g., anesthesia, minimally conscious states) offer an opportunity to investigate awareness in people with severe AD. In this article we demonstrate the feasibility of using Transcranial magnetic stimulation (TMS) combined with EEG, event related potentials (ERPs) and fMRI to assess awareness in severe AD. TMS-EEG was performed in six healthy older controls and three people with severe AD. The perturbational complexity index (PCI<sup>ST</sup>) was calculated as a measure of capacity for conscious awareness. People with severe AD demonstrated a PCI<sup>ST</sup> around or below the threshold for consciousness, suggesting reduced capacity for consciousness. ERPs were recorded during a visual perception paradigm. In response to viewing faces, two patients with severe AD provisionally demonstrated similar visual awareness negativity to healthy controls. Using a validated fMRI movie-viewing task, independent component analysis in two healthy controls and one patient with severe AD revealed activation in auditory, visual and fronto-parietal networks. Activation patterns in fronto-parietal networks did not significantly correlate between the patient and controls, suggesting potential differences in conscious awareness and engagement with the movie. Although methodological issues remain, these results demonstrate the feasibility of using objective measures of awareness in severe AD. We raise a number of challenges and research questions that should be addressed using these biomarkers of awareness in future studies to improve understanding and care for people with severe AD.
BibTeX
@inproceedings{huntley2023assessing,
  author = {Jonathan Huntley and Daniel Bor and Feng Deng and Marco Mancuso and Pedro A. M. Mediano and Lorina Naçi and Adrian M. Owen and Lorenzo Rocchi and Avital Sternin and Robert Howard},
  title = {{Assessing awareness in severe Alzheimer’s disease}},
  booktitle = {Frontiers in Human Neuroscience},
  year = {2023},
  doi = {10.3389/fnhum.2022.1035195},
}
Figure