Y. Du, J. Wang, S. Li, C. Meireles, A. Saliba, A. Castillo, M. Goros, J. Gelfond, BY. Choi, L. Qi, C. Li, et al. (2025). Digitally enhanced ketogenic diet versus low-fat diet for cardio-renal-metabolic health in a predominantly Hispanic adult population with overweight or obesity: Pilot randomised clinical trial. Diabetes, obesity & metabolism, 27(11), 6154-6166. https://doi.org/10.1111/dom.70000

PubMed 40770907

[Abstract]
AIMS: Individuals respond differently to diets for weight loss and cardio-renal-metabolic health. Advances in digital health and metabolomics hold promise for personalised lifestyle interventions. The study evaluated the feasibility and the preliminary efficacy of a digitally enhanced ketogenic diet compared with a low-fat diet, and explored the potential to use metabolites to guide personalised lifestyle interventions via digital health tools.

MATERIALS AND METHODS: The study randomised 60 adults living with overweight or obesity to a ketogenic diet (n = 30) or a low-fat diet (n = 30). Both groups received digital lifestyle interventions, including digitally delivered education sessions, mobile and wearable devices to monitor lifestyle behaviour and health indicators, and weekly individualised feedback through digital communication platforms. Clinical outcomes (e.g., weight, blood pressure (BP), HbA1c) and targeted metabolites were collected at baseline, 3 and 6 months. Intention-to-treat analysis and linear mixed models were used to assess group and time differences.

RESULTS: The study successfully enrolled the planned sample size (N = 60) within 1 year despite COVID-19-related challenges. Retention rates were 73.3% at 3 months and 71.7% at 6 months. Participants were 58.3% Hispanic. No significant differences in clinical or metabolic outcomes were observed between the groups, except that the ketogenic group had significantly higher ketone levels than the low-fat group at 3 months (b = 0.2, 95% CI = [0.05, 0.35], p = 0.015). Both groups showed significant weight and Body Mass Index reductions at 3 and 6 months. HbA1c improved only in the ketogenic group (baseline: 6.0 ± 0.9%; 3-month: 5.5 ± 0.7%, p < 0.001; 6-month: 5.7 ± 0.8%, p = 0.004), and systolic blood pressure (BP) improved in the ketogenic group at 3 months (131.2 ± 12.9 to 123.1 ± 11.2, p = 0.006). Most targeted metabolites (e.g., isoleucine, leucine) decreased significantly from baseline to 3- (~21/29 metabolites) and 6-month (~22/29 metabolites) within both groups.

CONCLUSIONS: Digitally enhanced lifestyle intervention was feasible. Both diets, enhanced by digital tools, were efficacious in weight reduction. The ketogenic diet showed potential benefits for glycaemic and BP control. Metabolite changes reflect sensitive responses to lifestyle interventions. Further research is warranted to explore the integration of digital tools and metabolic data for developing personalised lifestyle interventions.