Effect of time of day and fasting duration on measures of glycaemia: analysis from the Whitehall II Study

Abstract

We aimed to study diurnal variation in glucose regulation by examining the effects of time of day and fasting duration on fasting plasma glucose (FPG), 2h post-load plasma glucose (2hPG) and HbA(1c) levels. We analysed data from 5,978 non-diabetic white men and women from the prospective Whitehall II Study. All studied participants fasted for at least 8h before a clinical examination, which included an OGTT and anthropometric measurements. We fitted mixed-effects models for FPG, 2hPG and HbA(1c) as outcome variables, and time of day and/or fasting duration as explanatory variables. Models were adjusted for age, BMI and study phase. Time of day and fasting duration were associated inversely with FPG and positively with 2hPG. The mean difference between measures at 08:00 and 15:00hours in men/women was -0.46 (95% CI -0.50, -0.42) mmol/l/-0.39 (95% CI -0.46, -0.31) mmol/l and 1.39 (95% CI 1.25, 1.52) mmol/l/1.19 (95% CI 0.96, 1.42) mmol/l for FPG and 2hPG, respectively. HbA(1c) levels were independent of either time. Time of day and fasting duration were independently associated with 2hPG. In contrast, the effect of fasting duration on FPG was markedly attenuated with adjustment for time of day. Ageing, but not obesity, was associated with increased diurnal variation in glucose tolerance. Both time of day and fasting duration should be considered in clinical practice and epidemiological studies, since they have clinically relevant effects on FPG and 2hPG levels. As biochemically expected, HbA(1c) levels are independent of time of blood sampling and fasting duration.