(P16-025-25) No Between-Group Differences in Cardiometabolic Risk Factors in Individuals With Type 2 Diabetes After 6 Months of Personalized Diet Intervention

Objectives: A secondary analysis to examine whether a diet intervention personalized to predicted postprandial glycemic response (PPGR) in people with type 2 diabetes (T2D) improves lipid markers associated with cardiometabolic health as compared to a standardized diet.

Methods: The Diabetes Telemedicine Mediterranean Diet (DiaTeleMed) study is a remote randomized clinical trial to evaluate the impact of a 3-arm, 6-month dietary intervention on glycemic variability in adults with moderately-controlled T2D (hemoglobin A1c (HbA1c) 6.0 to 8.0%). The usual care control (UCC) arm received diabetes education with an emphasis on the Mediterranean diet, the Standardized (S) arm received the same as UCC plus behavioral counseling and was asked to self-monitor dietary intake using a mobile application. The personalized (P, n = 42) arm received all elements of S plus feedback on PPGR, calculated using a gut microbiome-based machine-learning algorithm. Fasting lipid panels (high-density lipoprotein (HDL), low-density lipoprotein (LDL), triglycerides (TG), total cholesterol (TC), TC-HDL ratio, non-HDL cholesterol) were collected at 0, 3, and 6 months (Quest Diagnostics). For the purposes of this sub-analysis, we combined the arms that did not receive personalization (UCC + S) into one group (non-Personalized (NP, n = 76)). A linear mixed-effects model was employed to check whether the change at months 3 and 6 varied between P and NP. Data are reported as N (%) or mean [SD].

Results: 118 participants contributed data for this analysis. Participants were 60 [12] years old, and mostly women (73 (61.9%)), Caucasian (56 (47.5%)), and non-Hispanic (94 (79.7%)). There were no between-arm differences in baseline characteristics, except TC-HDL ratio was higher in P (p=0.03). No significant differences were observed between P and NP at 3 and 6 months, respectively, in ΔTC (p= 0.72, p=0.75), ΔHDL (p=0.37, p= 0.41), ΔLDL (p=0.92, p=0.99), ΔTC-HDL ratio (p=0.17, p=0.21), Δnon-HDL cholesterol (p=0.88, p=0.57), or ΔTG (p=0.92, p=0.09).

Conclusions: We found no conclusive between-group differences in change in lipid markers at 3 and 6 months. The addition of personalized PPGR feedback to a Mediterranean-style diet does not provide improvements in cardiometabolic risk factors.

Funding Sources: The DiaTeleMed Study is supported by a grant from the National Institute of Health (R01NR018916).