The Acute Effect of Adequate Water Intake on Glucose Regulation in Low Drinkers

Background: Arginine vasopressin (AVP), a key hormone in fluid balance, may be a modifiable contributor to hyperglycemia [ 1 ]. Low daily water drinkers often exhibit increased urine concentration and copeptin, a surrogate marker for AVP [ 2 , 3 ]. Objective: The primary purpose was to investigate the acute effect of adequate water intake on daily glucose concentration in low drinkers. Secondarily, the study examined if adequate water intake could improve glucoregulatory hormonal profiles in low drinkers. Methods: Seven healthy (5 males, 2 females; age 43 ± 7 years, BMI 31 ± 3) low drinkers were recruited using a water frequency questionnaire and a 24-h urine sample. Participants were recruited using social media channels and flyers in local community. Classification of a low drinker was defined by a fluid intake (water and other beverages) <1.5 L·day −1 in males or <1.0 L·day −1 in females and a 24-h·UOsm of >800 mmol·kg −1 . In a crossover counterbalanced design, participants remained in the laboratory for 11 h (07:00–18:00) and were provided either the Institute of Medicine’s recommended amount of water excluding fluid from food (males: 3 L·day −1 , females: 2 L·day −1 ; high water intake, HWI) or an amount representing the bottom quartile of water consumption observed in the National Health and Nutrition Examination Survey (males: 0.5 L·day −1 , females: 0.4 L·day −1 ; low iater intake, LWI) (Table 1) [ 4 , 5 ]. Caloric intake was standardized to body weight (100 kJ·kg −1 ) with an identical ratio of macronutrients and time of consumption between trials (Table 1). At 07:00, fasted baseline blood was drawn. Subsequent blood draws performed across the next 11 hours were analyzed for copeptin, glucose, insulin, glucagon, cortisol, and GLP-1 (Table 1). All urine voids during the 11-h protocol were pooled and analyzed for osmolality and glucose ( n = 4). A two-way (water intake × time) repeated-measures ANOVA was used to determine differences in hydration and glucoregulatory measures. Dependent t tests were used to measure differences in urine samples. Statistical significance was determined a priori at an alpha of 0.05. Results: Participants were confirmed as low drinkers according to daily fluid intake, 24-h·UOsm, and copeptin (water frequency questionnaire volume: 823 ± 403 mL·day −1 , 24-h·UOsm: 961 ± 105 mmol·kg −1 , copeptin: 8.17 ± 3.05 pmol·L −1 ). During the experiments, 11-h·UOsm (HWI: 224 ± 48 mmol·kg −1 , LWI: 956 ± 120 mmol·kg −1 ), plasma osmolality, and copeptin were lower in HWI as than in LWI ( p = < 0.05, Fig. 1). There was a borderline significant main effect of water intake on plasma glucose ( p = 0.07, Fig. 2) and total urinary glucose output (HWI: 51.4 ± 6.9 mg, LWI: 40.1 ± 10.4 mg, p = 0.07). Cortisol was significantly higher in LWI as than in HWI ( p = 0.009, Fig. 2); however, no pairwise differences were observed in post hoc analysis. Glucagon, insulin, and GLP-1 were similar between trials ( p > 0.05). Conclusion: Acute increases in water intake may mildly reduce daily plasma glucose concentrations in low drinkers. This may be due to acutely increased urinary glucose output when low drinkers are given adequate amounts of water. Increased water intake also led to decreased cortisol concentration.