Event Abstract Back to Event Migraine patients exhibit an imbalance between feedback and feedforward visual signalling. Marco Lisicki1, 2*, Kevin D'Ostilio1, 2, Gianluca Coppola3, Alain Maertens De Noordhout1, 2, Vincenzo Parisi3, Jean Schoenen1, 2 and Delphine Magis1 1 Headache Research Unit. CHU de Liège, Belgium 2 University of Liège, Belgium 3 Fondazione G.B. Bietti (IRCCS), Italy Introduction The aim of this study was to investigate visually induced alpha/beta (feedback) and gamma (feedforward) power in episodic migraine without aura patients during or outside of an attack, and to compare the results to those of chronic migraine patients and healthy volunteers. Materials and Methods The study involved 20 healthy volunteers (mean age: 36.1 ± 11.4 years), 30 episodic migraine without aura patients recorded during a headache free interval (±72 hours) (mean age: 33.3 ± 11.9 years), 20 episodic migraine without aura patients recorded during a migraine attack (mean age: 32.7 ± 9.1 years), and 20 chronic migraine patients (mean age: 40.3 ± 12.7 years). Pattern reversal (3.1 reversals/second, 68’ squares) Visual Evoked Potentials (VEP) recordings were performed in the electrophysiology laboratory of the Headache Research Unit (Neurology Department, Centre Hospitalier Régional de la Citadelle, Liège, Belgium). Six hundred trials, each lasting 250ms, were continuously registered. After DC subtraction, epochs were band-pass filtered between 1 and 100 Hz. Artiefacted epochs were rejected (<6%). The Fast Fourier Transform was applied on each epoch to compute spectral decomposition. Log-transform of single-trial spectral power was performed before averaging. To estimate power, the area under the curve of activity at the peak and nearby surrounding frequencies (4 to 16 Hz for alpha/beta - feedback and 40 to 60 Hz for gamma - feedforward) was calculated for each individual. Fully automated electrophysiological analyses were performed using EEGLAB running in MATLAB (The MathWorks Inc.). Statistical analyses and graphs were performed in Prism version 6.00 for Windows (GraphPad Software, La Jolla, California, USA). The significance level for all tests was set at p<0.05. Results Mean number of monthly migraine days and total headache days were 5.5 ± 3.5 / 7.3 ± 4.1 for headache free episodic migraine patients, 5.9 ± 3.6 / 8.6 ± 6.6 for episodic migraine patients in the ictal period, and 15.8 ± 6.4 / 23.9 ± 5.7 for chronic migraineurs. Mean alpha/beta power was significantly higher in headache free episodic migraine patients compared to healthy volunteers. Conversely, gamma power was higher in ictal episodic migraine patients as well as in chronic migrainerus. A significant decrease in the feedback/feedforward (alpha/beta-to-gamma) ratio was found in both the ictal and chronic migraine groups with respect to the headache free group. This ratio was negatively correlated with the number of total monthly headache days (r= -0,33; p= 0,02). However, when considering only migraine-specific days, the same correlation did not reach statistical significance (r=-0,2496; p=0,08). Discussion Prior evidence has shown that feedforward volleys upcoming from visual afferents induce oscillations in the primary visual cortex in the gamma range. Conversely, feedback activity arriving from higher order visual areas (i.e. V2-V4) induces oscillatory activity within the alpha/beta frequency bands.(1,2) Our results show that, while an increased feedback activity (alpha/beta power) characterizes the visual responses of episodic migraine patients when they are headache free, elevated feedforward activity (gamma power) is the main feature found in patients during the ictal period and in chronic migraineurs. Given their physiological properties, a possible interpretation of our finding of increased visually induced alpha/beta-to-gamma ratio in headache free migraine patients is that feedback activity limits feedforward volleys during the inter-ictal period; a restraining feature that is lost during migraine attacks. Figure 1