Delta oscillation enhancement and alpha oscillation attenuation as electroencephalogram biomarkers of CNS demyelination in patients with hematologic malignancies

Abstract Background: Central nervous system demyelinating disease (CNSD) is a frequent and diagnostically challenging neurological complication of hematologic malignancies. Resting-state neural oscillations reflect cortical function and may provide noninvasive biomarkers for early CNSD detection. This study explored clinical risk factors and analyzed resting EEG power spectral density (PSD) changes to identify reliable biomarkers of CNS demyelination in this patient population. Methods: A total of 168 patients with hematologic malignancies treated at the Institute of Hematology, Chinese Academy of Medical Sciences, were enrolled: 68 with confirmed CNS demyelination (CNSD group) and 100 without (Control group). Clinical variables were recorded for all. Resting 19-channel EEGs were obtained in a subset (20 CNSD vs. 20 Controls). PSD of delta (0.5–4 Hz) and alpha (8–13 Hz) bands was computed via Fourier transform. Logistic regression identified independent clinical risk factors for CNSD. Independent-samples t-tests compared regional and global PSD between groups. Results: (1) Multiple myeloma (OR = 10.60, P = 0.007), acute onset (OR = 7.52, P = 0.010), history of preceding infection (OR = 5.923, P = 0.013), cerebrospinal fluid albumin (OR=7.723, P=0.028), intracranial pressure (OR=1.023, P=0.001), and cerebrospinal fluid IgG index (OR=566.814, P<0.001) were independent risk factors for CNS demyelination in hematological diseases. (2) Global EEG changes: In the CNSD group, alpha-band PSD was significantly reduced (P < 0.001), while delta-band PSD was significantly increased (P = 0.004). (3) Regional EEG changes: In alpha band, PSD decreased across all brain regions (all P < 0.001). Paradoxically, the Fz and Pz channels showed significant alpha-PSD enhancement (both P < 0.001), whereas all other channels exhibited significant decreases (P < 0.05). In delta band, PSD in frontal (F), central-parietal (CP), and temporal (T) regions was markedly higher than controls (F: P < 0.001; CP: P < 0.001; T: P < 0.001), and every individual channel showed increased delta-PSD compared with controls (P < 0.01). Conclusion: Multiple myeloma, acute onset, preceding infection, elevated intracranial pressure, increased CSF albumin, and high CSF IgG index independently predict CNS demyelination in hematologic malignancies. Resting-state EEG reveals global alpha attenuation and delta enhancement, with specific regional and channel-level alterations (notably Fz/Pz alpha increases), which can serve as noninvasive biomarkers for early CNSD detection. Incorporating these electrophysiological measures may facilitate timely diagnosis and improve patient management.