Postoperative Delirium Severity and Recovery Correlate With Electroencephalogram Spectral Features

Abstract

Background: Delirium is an acute syndrome characterized by inattention, disorganized thinking, and an altered level of consciousness. A reliable biomarker for tracking delirium does not exist, but oscillations in the electroencephalogram (EEG) could address this need. We evaluated whether the frequencies of EEG oscillations are associated with delirium onset, severity, and recovery in the postoperative period.

Methods: Twenty-six adults enrolled in the Electroencephalography Guidance of Anesthesia to Alleviate Geriatric Syndromes (ENGAGES; ClinicalTrials.gov NCT02241655) study underwent major surgery requiring general anesthesia, and provided longitudinal postoperative EEG recordings for this prespecified substudy. The presence and severity of delirium were evaluated with the confusion assessment method (CAM) or the CAM-intensive care unit. EEG data obtained during awake eyes-open and eyes-closed states yielded relative power in the delta (1-4 Hz), theta (4-8 Hz), and alpha (8-13 Hz) bands. Discriminability for delirium presence was evaluated with c-statistics. To account for correlation among repeated measures within patients, mixed-effects models were generated to assess relationships between: (1) delirium severity and EEG relative power (ordinal), and (2) EEG relative power and time (linear). Slopes of ordinal and linear mixed-effects models are reported as the change in delirium severity score/change in EEG relative power, and the change in EEG relative power/time (days), respectively. Bonferroni correction was applied to confidence intervals (CIs) to account for multiple comparisons.

Results: Occipital alpha relative power during eyes-closed states offered moderate discriminability (c-statistic, 0.75; 98% CI, 0.58-0.87), varying inversely with delirium severity (slope, -0.67; 98% CI, -1.36 to -0.01; P = .01) and with severity of inattention (slope, -1.44; 98% CI, -2.30 to -0.58; P = .002). Occipital theta relative power during eyes-open states correlated directly with severity of delirium (slope, 1.28; 98% CI, 0.12-2.44; P = .007), inattention (slope, 2.00; 98% CI, 0.48-3.54; P = .01), and disorganized thinking (slope, 3.15; 98% CI, 0.66-5.65; P = .01). Corresponding frontal EEG measures recapitulated these relationships to varying degrees. Severity of altered level of consciousness correlated with frontal theta relative power during eyes-open states (slope, 11.52; 98% CI, 6.33-16.71; P < .001). Frontal theta relative power during eyes-open states correlated inversely with time (slope, -0.05; 98% CI, -0.12 to -0.04; P = .002).

Conclusions: Presence, severity, and core features of postoperative delirium covary with spectral features of the EEG. The cost and accessibility of EEG facilitate the translation of these findings to future mechanistic and interventional trials.

Figure 1.. Delirium outcomes and EEG recordings.

Participant timelines are provided at an individual patient level by postoperative day (POD), with delirium outcomes based on Confusion Assessment Method (CAM), CAM-ICU, and delirium chart review. Days with delirium absence (blue) and presence (red) are noted. Days with EEG recordings that contributed to analyses are highlighted with an asterisk.

Figure 2.. Occipital EEG alpha and delta relative power discriminate delirium outcomes.

A. Based on EEG recordings acquired during wakeful eye closure, patients with delirium (red) demonstrate a concentration of relative EEG power in the delta (pink, 1–4 Hz) and low theta (blue, 4–8 Hz) frequency bands. Patients classified without delirium (blue) show spectral peaks primarily in the high theta and low alpha (green, 8–13 Hz) frequency bands. B. Power spectra acquired during eyes-open states recapitulate a shift of EEG power from the low alpha and high theta ranges toward low theta and delta. C-D. Receiver-operating characteristic (ROC) curves demonstrating discriminability of delirium outcome based on occipital EEG delta (pink), theta (blue) and alpha (green) relative power acquired during eyes-closed (C) and eyes-open (D) recordings. * p < 0.05, ** p < 0.01 E. Concordance (c-) statistics are provided based on area under receiver operating characteristic (ROC) curves. Bootstrapping analyses yielded 98% confidence intervals for c-statistic estimates. Significant discriminability above chance was noted for the following measures (c-statistic, CI): occipital relative delta power during eyes-closed states (c-statistic: 0.68, 98% CI: [0.50 0.82]), occipital relative alpha power measures during both eyes-closed (c-statistic: 0.75, 98% CI: [0.58 0.87]) and eyes-open states (c-statistic: 0.73, 98% CI: [0.54 0.85]). Other c-statistic measures, including occipital theta relative power (eyes-closed, c-statistic: 0.64, 98% CI: [0.43 0.84]; eyes-open, c-statistic: 0.60, 98% CI: [0.37 0.78]), were not significant as confidence intervals overlapped 0.50. Delirium outcomes were based on Confusion Assessment Method (CAM), CAM-ICU evaluations, and chart review. Comparison results for frontal EEG measures are provided in Supplemental Figure 4.

Figure 3.. Delirium severity correlates with occipital EEG relative power measures.

Generalized mixed-effects models were constructed to correlate delirium severity and EEG metrics. Slopes of generalized ordinal mixed-effects models are reported as the change in severity score/change in EEG relative power. Severity scores did not significantly correlate with EEG delta (A, generalized mixed-effects models slope = 0.38, 98% CI: [−0.51 1.09], p = 0.32) or theta (B, generalized mixed-effects models slope = 0.37, 98% CI: [−0.87 1.60], p = 0.44) relative power measures during eye closure. C. In contrast, occipital alpha relative power (8–13 Hz) negatively correlated with severity (generalized mixed-effects models slope = −0.67, 98% CI: [−1.36 −0.01], p = 0.01). For eyes-open states, delirium severity scores correlated with relative theta power (E, generalized mixed-effects models slope = 1.28, 98% CI: [0.12 2.44], p = 0.007), but did not with either delta (D, generalized mixed-effects models slope = −0.62, 98% CI: [−1.56 0.34], p = 0.97) or alpha (F, generalized mixed-effects models slope = −0.27, 98% CI: [−1.10 0.55], p = 0.39) counterparts. Delirium outcomes are indicated by red (present) or blue (absent). EEG relative power measures are plotted on the logarithmic scale for display purposes. Companion analyses for frontal EEG power are provided in Supplemental Figure 5.

Figure 4.. Delirium core features correlate in severity with occipital EEG relative power metrics.

Grade of inattention, disorganized thinking, and altered level of consciousness (LOC) were evaluated against occipital EEG power measures that correlated with delirium severity in Figure 2. A-B. Severity of inattention correlated negatively with EEG alpha relative power during eyes-closed states (A, generalized mixed-effects models slope = −1.44, 98% CI: [−2.30 −0.58], p = 0.002) and positively with EEG theta relative power derived from eye-open periods (B, generalized mixed-effects models slope = 2.00, 98% CI: [0.48 3.54], p = 0.01). C-D. While severity of disorganized thinking did not correlate with alpha power (C, generalized mixed-effects models slope = −5.99, 98% CI: [−12.53 0.54], p = 0.07), this feature of delirium correlated positively with EEG theta relative power (D, generalized mixed-effects models slope = 3.15, 98% CI: [0.66 5.65], p = 0.01). E-F. Altered LOC was not significantly correlated with either relative alpha power (E, generalized mixed-effects models slope = −1.46, 98% CI: [−3.41 0.48], p = 0.14) or theta power (F, generalized mixed-effects models slope = −5.83, 98% CI: [−102.72 91.05], p = 0.90) measures. Delirium severity scores were based on the long-form Confusion Assessment Method (CAM-S), 0 for disturbance absent, 1 for mild severity, and 2 for marked severity. Overall outcomes for delirium absence (blue) or presence (red) are based on CAM, CAM-ICU, and chart review. Refer to Supplemental Figure 6 for relationships to frontal EEG relative power.

Figure 5.. Correlations among frontal EEG relative power metrics and time during the progression of delirium.

A-B. Frontal (F8-Cz) relative EEG power in the delta (1–4 Hz) frequency band during eyes-closed (A) and open (B) states. Delta relative EEG power during eyes-open correlated positively with time (B, linear mixed-effects models slope = 0.07, 98% CI: [0.02 0.13], p = 0.004) but not during eyes-closed (A, linear mixed-effects models slope = 0.01, 98% CI: [−0.07 0.07], p = 0.78). C-D. Frontal relative EEG theta power during eyes-open states (D), but not eyes-closed (C, linear mixed-effects models slope = −0.02, 98% CI: [−0.07 0.02], p = 0.19), inversely correlated with time (D, linear mixed-effects models slope = −0.05, 98% CI: [−0.12 −0.04], p = 0.002). E-F. Frontal relative alpha power did not correlate with time during both eyes-closed (E, linear mixed-effects models slope = 0.07, 98% CI: [−0.07 0.21], p = 0.20) and eyes-open (F, linear mixed-effects models slope = −0.05, 98% CI: [−0.14 0.03], p = 0.08). Slopes of linear mixed-effects models are reported as the change in the log₁₀ (EEG relative power) over time. Recordings on days of delirium present (red) or absent (blue) were based on Confusion Assessment Method (CAM) evaluations, CAM-ICU, and chart review. Relationships of occipital relative EEG power over time are provided in Supplemental Figure 9.