2092-P: Impact of Insulin Resistance on Response to a Meal by Salience, Executive Control, and Default Mode Resting State Networks in Humans

Functional connectivity (fc) in the Salience Network (SN) is reduced by a meal. We hypothesized that disengagement of the SN may be mirrored by increased fc in Executive Control (ECN) or Default Mode (DMN) networks (representing a switch away from interoceptive neural processing after feeding) and that insulin resistance might impact shifts in postprandial neural processing. Subjects (N=111) oversampled for obesity had body composition, fasting glucose and insulin evaluated and underwent resting-state fMRI before and after a standardized meal. Masks for ECN (Seeley2007) and DMN (Grecius2003) were applied to obtain the fc pre-, post-meal and change (%). Subgroups with low (N=37) and high (N=36) HOMA-IR were created from tertiles. Among all subjects (28±9 y; BMI 30.5±6 kg/m2), ECN fc was reduced by a meal (z=-2.9, P=0.004), but DMN fc did not change (z=-0.02, P=0.98). The low and high HOMA-IR groups had mean BMIs of 26.8 and 34.7±5. Fc response to a meal, adjusted for age, sex and fat mass, did not differ between low and high HOMA groups for ECN (group*time chi2(1)=2.39, PInt=0.12) or SN (chi2(1)=2.52, PInt=0.11); both groups reduced fc. However, for DMN a group*time was found (chi2(1)=9.12, PInt=0.003, adjusted), such that the low HOMA group increased post-meal fc (Fisher z-score 0.36 vs. 0.41, P=0.04), but high HOMA group decreased fc (0.37 vs. 0.31, P=0.03). Change (%) in connectivity was positively correlated among all 3 networks (P<0.002). A significant interaction for DMN vs. ECN (P=0.02) by HOMA group was present. Participants with high HOMA had a blunted correlation between DMN and ECN (coef=0.12, P=0.28) compared to those with low HOMA (coef=0.61, P=0.001). Fc was reduced by a meal in the ECN and not modified in the DMN on average. However, DMN fc increased with feeding in non-insulin resistant individuals as hypothesized, but decreased in those with insulin resistance, suggesting that insulin resistance impacts shifts in neural processing in response to a meal. Disclosure L.E. Sewaybricker: None. S.J. Melhorn: None. M.K. Askren: None. M. Webb: None. V. Tyagi: None. M. De Leon: None. T.J. Grabowski: None. E. Schur: None. Funding American Diabetes Association (1-17-ICTS-085 to E.S.); Sao Paulo Research Foundation (2017/00657-0); National Institutes of Health (DK089036, DK098466); University of Washington Nutrition Obesity Research Center (P30 DK035816); Diabetes Research Center (P30DK017047); Institute of Translational Health Sciences (UL1TR000423)