Actigraphy and cEOG for Delirium Detection in the ICU.

Smook, S. (Simone) (2014) Actigraphy and cEOG for Delirium Detection in the ICU. thesis, Medicine.

Text SmookSP.pdf Restricted to Registered users only Download (1MB)

Abstract

Objective To analyze whether changes in actigraphy and cEOG are indicative for delirium and whether these techniques can be used in future to detect delirium in the ICU. Methods This is an ongoing case control observational study including delirious and control patients in the ICU. Patients were categorized into the delirious or control group based on CAM-ICU scores which were performed three times a day. Delirious patients were categorized as hyperactive, hypoactive or mixed delirious based on RASS scores. We registered 24 hours a day for the length of the patient’s stay in the ICU or for a maximum of 5 days. Delirious patients were compared with control patients as well as delirious patients with themselves if they showed delirious and non-delirious episodes. Results In total 11 patients were included, five delirious patients and six control patients. Two delirious patients were mixed delirious, two were hyperactive delirious and one hypoactive delirious. As we studied a small number of patients only tentative indicative trends could be observed. We found that delirious patients were significantly more active during nighttime (p<0.05) and had a significant more disturbed day/night rhythm in the percentage of time being active (p<0.05) compared to control patients. When observing patients individually, there seemed to be a difference in activity between subtypes of delirium. One hyperactive delirious patient showed evidently more expressive activity and a hypoactive delirious patient clearly showed lower and less activity compared to control patients and to their own non-delirious episodes. Due to technical problems no registration of blinks was possible in some patients, resulting in five control patients and three delirious patients. The algorithm to detect blinks was a newly developed algorithm which needs to be optimized in future. For now, we found in delirious patients a significant lower velocity of the upward side of a blink compared to control patients (p=0.04). Next to that, two of the three delirious patients showed an increase of blink duration during their delirious episodes compared to the non-delirious episodes. Conclusion Both activity and blinks were affected in delirious patients. The continuation of this study should establish whether actigraphy and cEOG can discriminate delirious from non-delirious episodes and whether it can recognize delirium at an early stage.

Item Type:	Thesis (Thesis)
Supervisor name:	Beishuizen, Dr. A. and Putten, Prof. dr. ir. M.J.A.M. van and Heide, Dr. ir. E.M. van der and Medical Spectrum Twente, Enschede
Faculty:	Medical Sciences
Date Deposited:	25 Jun 2020 10:46
Last Modified:	25 Jun 2020 10:46
URI:	https://umcg.studenttheses.ub.rug.nl/id/eprint/765

Actions (login required)

View Item