Harnessing Scalp Gamma Oscillations to Predict Clinical Outcomes in Patients with Periodic Discharges
Kongpop Sutantikorn1,2, Totsapol Surawattanawong1,2, Chusak Limotai3,4,5
1Division of Neurology, Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
Background:
Periodic discharges (PDs) are frequently observed in critically ill patients and are associated with diverse etiologies and variable prognoses. While high-frequency oscillations have been extensively studied in epilepsy, the clinical significance of scalp gamma oscillations (SGOs)-defined as activity > 30 Hz-remains poorly understood in neurocritical care settings.
Objective:
To evaluate the prognostic value of SGOs in predicting short-term functional outcomes among critically ill patients with PDs.
Methods:
This retrospective cohort study included patients aged > 15 years admitted to our center (2014-2022) who underwent continuous electroencephalograph (EEG) monitoring and had electrographic PDs. EEG data were analyzed using standardized protocols. SGOs were defined as discrete oscillations > 30 Hz superimposed on spike/sharp waveforms. Two parameters were extracted: mean frequency and SGO count (per 60 s). Functional outcome was assessed using the modified Rankin Scale (mRS) at discharge and 90 days postdischarge, with outcomes dichotomized as favorable (mRS 0-3) or poor (mRS 4-6). Multivariable logistic regression was used to identify predictors of outcome.
Results:
A total of 72 patients were included (mean age 66.5 ± 17.1 years; 50% male). Poor outcomes were observed in 77.8% at discharge and 76.4% at 90 days. Both higher mean SGO frequency and increased SGO count were significantly associated with poor functional outcomes. A median SGO frequency > 51.5 Hz (adjusted OR 5.08, 95% CI 1.21-21.34; p = 0.027) and a median SGO count > 14 per 60 s (adjusted OR 5.82, 95% CI 1.11-30.47; p = 0.037) were both independent predictors of poor outcomes after controlling for relevant clinical variables.
Conclusions:
SGOs may serve as a novel and practical biomarker for predicting short-term functional outcomes in critically ill patients with PDs. Their detection using standard EEG protocols supports the feasibility of integrating SGO analysis into routine ICU care. Further prospective studies are warranted to validate these findings.
