Use of Automated Infrared Pupillometry to Predict Delirium in the Intensive Care Unit: A Prospective Observational Study

Abstract

Introduction Delirium is an acute state of brain dysfunction prevalent among critically ill patients. Disturbances in the sympathetic neurons, including cholinergic neurons, have been reported to cause delirium by upsetting the balance of neurotransmitter synthesis, release, and inactivation. The cholinergic system mediates pupillary constriction as a response to light stimulation, and this reflex can be measured using automated infrared pupillometry (AIP). The relationship between delirium and AIP parameters has been examined. The Confusion Assessment Method for the Intensive Care Unit (CAM ICU) and the Intensive Care Unit Delirium Screening Checklist (ICDSC) are used for assessing delirium. However, that between the ICDSC score and AIP parameters remains unclear. Objective To examine the relationship between AIP parameters and the various categories of delirium as defined by the ICDSC score (delirium, subsyndromal delirium, no delirium). Methods This prospective observational study included patients aged ≥18 years admitted to the intensive care unit (ICU) from May 2018 to September 2018. ICU patients were classified into delirium, subsyndromal delirium, and no delirium groups according to the ICDSC score during ICU stay. The pupillary light reflex was assessed in both eyes immediately after admission using AIP with a portable infrared pupillometer. Logistic regression analyses were used to estimate the odds ratio to examine the relationship between the severity of delirium as assessed by the ICDSC score and the AIP parameters. Results In total 133 patients were included in the study. Based on the ICDSC scores, 41.4% of patients had no delirium, 40.6% had subsyndromal delirium, and 18% had delirium. Dilation velocity (DV) measured by AIP was significantly different among the delirium, subsyndromal delirium, and no delirium groups. Post-hoc comparisons showed that DV was significantly slower in the delirium group than in the no delirium group but was not significantly different between the subsyndromal delirium and no delirium groups. After adjusting for patients’ sex and age at enrollment, DV was shown to be independently associated with delirium. Conclusion This study suggests that the use of AIP at ICU admission may improve the identification of patients at a high risk of developing delirium.

Association Between Pupil Light Reflex and Delirium in Adults With Traumatic Brain Injury: Preliminary Findings

“…However, inconsistent findings on the use of pupillometry for ICU delirium screening purposes have also been reported. In non-TBI ICU patients, a recent study 18 found no significant association between these 2 PLR variables (pupil constriction percentage and velocity) and delirium but still showed a decrease in values in patients with delirium compared with those without. Notably, other studies conducted during cardiac surgery and postanesthesia care have emphasized the importance of dilation velocity (defined as the recovery phase of the PLR measured in millimeters per second) and the Neurological Pupil index (an algorithm that includes size, latency, constriction velocity, and dilation velocity) as valid predictors of delirium.…”

“…16 However, its application in assessing delirium has only recently been explored, particularly among non-TBI patients. [17][18][19] This technology quantitatively measures the resting pupil diameter and its changes in response to external light stimulation, known as the pupillary light reflex (PLR). 20 The procedure uses an infrared illumination system device with a digital video signal processor, commonly called a video pupillometer.…”