Quantifying agitation in sedated ICU patients using heart rate and blood pressure

Agitation is a significant problem in the intensive care unit (ICU), affecting 71% of sedated adult patients during 58% of ICU patient-days (Fraser and Riker 2001 NY Health-Syst. Pharm. 20 17-24). Subjective scale-based assessment methods are currently used to assess the level of patient agitation, but are limited in their accuracy and resolution. This research develops an objective agitation measurement method using heart rate variability (HRV), systolic blood pressure (BP) and blood pressure variability (BPV) data, processed by wavelet transforms and autoregressive signal processing. A fuzzy inference system (FIS) is developed to classify changes in these signals that correlate with observed patient agitation, and combine them into a final agitation level. Proof of concept clinical trials on 13 normal subjects and 5 ICU patients has been performed to verify the validity of this approach in comparison with agitation graded by nursing staff using the Riker sedation-agitation scale (SAS). Results show good correlation with medical staff assessment with no false positive results during calm periods. Clinically, this initial agitation measurement method promises the ability to consistently and objectively quantify patient agitation to enable better management of sedation and agitation through optimized drug delivery leading to reduced length of stay.