# The Cognitive Disintegration Model of Delirium: A Clinical Cohort Study Investigating Neural Network Connectivity and Local Slow Wave Activity

> **NIH NIH K23** · UNIVERSITY OF WISCONSIN-MADISON · 2020 · $74,098

## Abstract

Project Summary/Abstract
Delirium is a sudden state of confusion that is associated with increased morbidity and mortality, impaired long-
term cognition and loss of independence. Although there are substantial costs – financial, societal and individual
– delirium is unfortunately bereft of therapies, largely due to the limited understanding of its pathogenesis. The
long-term goal of the proposed research program is to develop preventative and therapeutic approaches to
delirium. In this application, I seek to identify the neural correlates of delirium, and subsequent cognitive decline
as the first step in a translational research program to address this therapeutic void. The strategy is based on
the Cognitive Disintegration model: we hypothesize that delirium results from an acute breakdown in neural
network connectivity. In essence, delirium results when “information integration” falls below a critical threshold
in vulnerable networks. We use network connectivity as a surrogate of the capacity to integrate information. We
will test the hypothesis that impaired preoperative cingulate functional connectivity is associated with increased
risk of postoperative delirium following adjustment for confounding variables. This will inform patient risk
stratification and the development of preclinical models of delirium. An important observation that motivates this
approach is the finding that cognition is not globally impaired in delirium; rather, specific cognitive functions, such
as attention and executive function, are particularly affected. Furthermore, while the electroencephalogram
(EEG) hallmark of delirium is increased slow wave activity, the patients are not asleep. We reconcile this
difference by hypothesizing that there is “local” slow wave activity in delirium. If proven, this would provide a
mechanism for the breakdown in network connectivity in delirium and provide rationale for the study of the
vulnerability of specific brain regions in both future clinical and translational studies. In experimental extensions,
we plan to study the change in connectivity of different brain regions associated with delirium through source
reconstruction of the EEG. Finally we will address the role of pre-delirium connectivity in the associations
between delirium and long-term cognitive decline, hypothesizing that preoperative connectivity modulates the
impact of delirium on long-term cognition. This will provide greater insight into the long-term impact of delirium.
The funding will be used to conduct a perioperative cohort study, collecting clinical, cognitive, imaging and EEG
data. This application will provide comprehensive insights into the neural network changes predisposing
to, and associated with, postoperative delirium and its long-term cognitive sequelae. While postoperative
delirium is a major public health issue warranting significant scientific focus and investment, we also expect our
findings to have wider ramifications for the neuroscience of sl...

## Key facts

- **NIH application ID:** 9821168
- **Project number:** 5K23AG055700-03
- **Recipient organization:** UNIVERSITY OF WISCONSIN-MADISON
- **Principal Investigator:** Robert D Sanders
- **Activity code:** K23 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $74,098
- **Award type:** 5
- **Project period:** 2017-12-01 → 2020-05-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9821168

## Citation

> US National Institutes of Health, RePORTER application 9821168, The Cognitive Disintegration Model of Delirium: A Clinical Cohort Study Investigating Neural Network Connectivity and Local Slow Wave Activity (5K23AG055700-03). Retrieved via AI Analytics 2026-05-22 from https://api.ai-analytics.org/grant/nih/9821168. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*