SFEBES2018 Poster Presentations Diabetes & cardiovascular (27 abstracts)
1Kings College London, London, UK; 2Diabetes Association of Sri Lanka, Rajagiriya, Sri Lanka.
Background: Existing interventional trials show high dietary protein intake can reduce glycaemia in type 2 diabetic (T2DM) individuals. At present, there is limited data on this relationship in a young, healthy population at risk of T2DM. This study investigates the association of animal and plant-based dietary protein intake on fasting blood glucose (FBG) and 2-hour oral glucose tolerance test (OGTT) concentrations in a young South-Asian population at high-risk of type 2 diabetes.
Methods: A post-hoc analysis of a randomised controlled trial of 1250 South-Asian participants aged <18 assigned to intensive (3-monthly, n=573) or control (12-monthly, n=677) lifestyle modification. Median follow-up was 3 years, with participants annually self-reporting dietary data using a food-frequency questionnaire. Total protein indices and plant-protein ratios were calculated for each participant. Pearsons correlation and multiple linear regression models, including age, gender, change of waist circumference, and change in physical activity assessed relationships between protein intake and i) FBG and ii) 2-hour post-OGTT blood glucose.
Results: Total protein intake was significantly positively correlated to FBG in year 3 (r=0.059, P=0.042), but significantly negatively correlated with OGTT in year 4 (r=−0.119, P=0.001). Between baseline and trial completion, total protein (animal and plant-based) was neither significantly related to changes in FBG (β=−0.025, P=0.56) nor changes in OGTT (β=−0.0015, P=0.72). Changes in plant protein ratio were neither significantly related to changes in FBG (β=−0.026, P=0.54) nor changes in OGTT (β=−0.030, P=0.48).
Conclusion: Total protein intake was weakly positively correlated with FBG, contrasting the negative correlation seen with OGTT in years 3 and 4, but overall regression found non-significant relationships. Furthermore, total protein intake in this population was low and despite multiple regression accounting for key confounders, recommendations from these findings are at present limited.