Evaluating the Role of Sleep in the Development of Intensive Care Unit Delirium

PROJECT SUMMARY More than one third of patients in the intensive care unit (ICU) develop ICU delirium, an acute, fluctuating form of cognitive dysfunction that is associated with increased mortality, and longer ICU lengths of stay and days of mechanical ventilation; and disproportionately increases the risk for long-term cognitive impairment.

Disrupted sleep in the ICU is thought to contribute to development of ICU delirium, however the sleep-delirium relationship has not been rigorously investigated. Although ICU patients maintain adequate sleep duration per 24 hours, sleep architecture is severely disrupted. Namely, there is a marked decrease in rapid eye movement (REM) sleep compared to healthy adults.

Insufficient REM sleep has been associated with neurodegenerative disease such as dementia, which has similar characteristics to ICU delirium.

Less REM sleep has been observed in patients with, compared to without ICU delirium, although the temporality is unclear.

To date, few studies have investigated sleep prior to development of ICU delirium, and no studies account for objective baseline sleep.

We propose a novel approach to address these critical gaps by leveraging a population of thoracic surgery candidates with a planned post-operative ICU admission to enable baseline sleep evaluation.

Research: The K99 phase study will evaluate clinical and demographic predictors of percentage of days with ICU delirium or coma (DoDC) in the acute ICU phase (days 1 up to 14) in a retrospective electronic health record study of thoracic surgery patients (N=4,849) at the University of Pennsylvania Health System (UPHS) from 2015-present.

As sleep is not routinely assessed in the ICU, it is not possible to evaluate sleep from medical records.

Therefore, we will first identify non-sleep predictors of ICU delirium to then be able to evaluate if sleep indices measured in future studies explains a substantial percentage of the variance in ICU delirium not accounted for by non-sleep predictors.

The R00 phase study will evaluate the difference in mean minutes of REM sleep per night for patients with and without ICU delirium in a prospective observational cohort study of thoracic surgery candidates from UPHS (N=148).

Sleep will be assessed using wireless electroencephalography at baseline (pre-ICU), and post-operatively in the ICU for 3 days each. In a subset of the R00 cohort (N=20), frequency of awakenings from sleep due to the ICU environment will be assessed. Results from these studies will be used to inform development of future sleep-promoting interventions.

Training: To achieve overall career goals, the training plan will build upon the candidate?s background in critical care and sleep in healthy individuals by affording her in-depth training in learning about sleep in clinical and ICU populations, and training in ICU delirium mechanisms and development of predictive models.

This training will also include advancing her scientific dissemination skills, as well as development of the skills needed to become a leader in the scientific community.

A variety of approaches will be used to achieve these goals including formal coursework, hands-on training, and structured one-on-one mentorship.