Authors :
Presenting Author: Alex Bender, MD, PhD – Massachusetts General Hospital
Mia Bothwell, MD, PhD – Massachusetts General Hospital, Harvard Medical School
Maddie Taubkin, MPH – Massachusetts General Hospital
Kyle Pellerin, BS – Massachusetts General Hospital
Sydney Cash, MD, PhD – Massachusetts General Hospital
Alice Lam, MD, PhD – Massachusetts General Hospital
Rationale:
Spindles (11-16 Hz) and slow oscillations (0.5-1.5 Hz; SO) are distinct neurophysiological elements of the NREM sleep microarchitecture. The coupling of spindles and SO is essential for coordinating interregional communication in cortical networks and memory consolidation in sleep. However, spindles and SO occur locally at different times and in different regions, and it is unknown how the coupling of such oscillations is organized across the brain. In this study, we present a novel relationship between spindles and SO, termed phase precession, and then compare the properties of this relationship between participants with and without epilepsy.
Methods:
We studied 111 adult patients with temporal lobe epilepsy (TLE) and 58 without epilepsy (Cntrl) who were admitted to the Epilepsy Monitoring Unit (EMU) at Massachusetts General Hospital. Continuous scalp EEG was sleep-staged, and spindles and SO were detected, using previously-published algorithms. The SO phases corresponding to spindle detections were plotted against the anterior-posterior scalp position. Circular-linear statistics were then used to determine the best fit line, and quantify the slope, strength (R value), and significance. Linear and logistic regression models were used to test the association of these phase precession parameters with demographic variables (age, sex) and to compare between Cntrl and TLE participants.
Results:
Visual inspection of circular phase plots across the brain topography revealed a striking pattern. There was a systematic shift in phase, with spindles occurring at sequentially earlier SO phases relative to their anterior-posterior position (Fig. 1A). In other words, there was precession of the phases moving from the front to the back of the brain. Phase-position plots confirmed this relationship at both an individual and group level (Fig. 2B). Examining phase precession across demographic factors revealed that there was a significant association with age, but not sex. With advancing age, the slope became flatter (p=0.02, R=0.32) and the strength became weaker (p=0.015, R=0.32). Comparing across groups, and adjusting for age, sex, and spindle count, there was a 25.6% reduction in the strength of phase precession in TLE (p=0.0001, T=-3.93, Rmodel=0.37), and the odds of having statistically significant phase precession was 71% lower in TLE (p=0.019, T=-2.34, chi2model=12.9) compared to Cntrl participants (Fig. 2C).
Conclusions:
Phase precession of spindle-SO coupling represents a novel relationship between spindles and SO across the brain’s cortical surface. We speculate that phase precession provides a potential mechanistic explanation for how the coupling of sleep oscillations could support memory consolidation in a coordinated manner across the brain. Underscoring its relevance, we show that in TLE, a brain disorder associated with dysfunctional neurophysiology and impaired memory, there is a loss of phase precession integrity.
Funding:
NINDS R25NS065743 and Mass General Neuroscience Transformative Scholars Award