Autonomic adaptations mediate the effect of hydration on brain functioning and mood: Evidence from two randomized controlled trials

Abstract

Dehydration (water loss >2.0% of body weight) has significant negative effects on physical and mental performance. In two studies the effects of minor hypo-hydration (water loss <1.0% of body weight) on CNS function, mood and cardiovascular functioning were measured. Study 1: On two mornings twelve male participants were exposed to a temperature of 30 °C for four hours and either did or did not drink two 150 ml glasses of water during that time. Study 2: Fifty-six (25 M) individuals were exposed to the same 30 °C environment and randomly allocated to either drink (2 × 150 ml) or not drink. When not given water 0.59% (Study 1) and 0.55% (Study 2) bodyweight was lost. Participant's heart rate variability (HRV) was measured, and they rated their thirst and mood. In study 1, participants participated in an fMRI protocol during which they completed a modified version of the Paced Auditory Serial Addition Test (PASAT), at the end of which they rated its difficulty. Decreases in fMRI BOLD activity in the orbito-frontal cortex, ventral cingulate gyrus, dorsal cingulate cortex, hypothalamus, amygdala, right striatum, post-central gyrus and superior parietal cortex were observed when participants were hypo-hydrated. These deactivations were associated with reduced HRV, greater perceived effort, and more anxiety. In study 2 declines in HRV were found to mediate the effect of hypo-hydration on ratings of anxiety. These data are discussed in relation to a model that describes how autonomic regulatory and interoceptive processes may contribute to the affective consequences of minor hypo-hydration.

Figure 1

The experimental procedure.

Figure 2

The effect of drinking water on the average length of the RR interval, SDNN and RMSSD. Data are the changes (end of the morning minus baseline) for each index compared across the two conditions. Top panel: Study 1. When participants were hypo-hydrated they had a lower R-R (i.e a higher HR) (p < 0.03), lower SDRR (i.e a lower HRV) (p < 0.01) and reduced parasympathetic activity (i.e. lower RMSSD) (p < 0.03). Bottom panel: Study 2. If water was consumed participants had an increase in their average R-R interval (p < 0.001), a larger standard SDNN (p < 0.027), and a larger RMSSD (p < 0.031). RR, interbeat interval, SDNN, standard deviation of normal to normal R-R interval, RMSSD, root mean square of the standard deviation.

Figure 3

Peak activity in PoCG (top) and mOFC (bottom) water >hypo-hydration.

Figure 4

(A) Peak activity in vCG water >hypo-hydration. Correlation between neuronal activity in the vCG and (B). RR interval (C). anxiety and (D). difficulty ratings. Data are Pearson’s r coefficient for difference scores (no water – water).

Figure 5

Schematic illustration of the mediation analysis used in study 2. The total effect of water on rating of anxiety was significant (B = 16.321, LLCI 3.729, ULCI 28.913). The direct effect (when the influence of SDNN was considered) was not significant (B = 12.763, LLCI -0.257, ULCI 25.784). When water was not consumed participants were more anxious (F(1,54) = 4.706, p < 0.034). A decrease in HRV (SDRR) was associated with an increase in anxiety levels (r = 0.335, p < 0.012). The indirect effect Water → SDNN → Anxiety was significant (B = 3.557, LLCI 0.212, ULCI 8.321).

Figure 6

Map showing the brain areas associated with autonomic modulation. Left data from present study. Right areas controlling HRV reproduced from Thayer et al. (2012) with permission.

Figure 7

Arithmetic task and Visual analogue scale example stimuli. The crosses are fixation points when otherwise the screen is blank. At the third image the participant indicated whether the right or left number was correct by pushing a button.