Effects of fructose vs glucose on regional cerebral blood flow in brain regions involved with appetite and reward pathways

Abstract

Importance: Increases in fructose consumption have paralleled the increasing prevalence of obesity, and high-fructose diets are thought to promote weight gain and insulin resistance. Fructose ingestion produces smaller increases in circulating satiety hormones compared with glucose ingestion, and central administration of fructose provokes feeding in rodents, whereas centrally administered glucose promotes satiety.

Objective: To study neurophysiological factors that might underlie associations between fructose consumption and weight gain.

Design, setting, and participants: Twenty healthy adult volunteers underwent 2 magnetic resonance imaging sessions at Yale University in conjunction with fructose or glucose drink ingestion in a blinded, random-order, crossover design.

Main outcome measures: Relative changes in hypothalamic regional cerebral blood flow (CBF) after glucose or fructose ingestion. Secondary outcomes included whole-brain analyses to explore regional CBF changes, functional connectivity analysis to investigate correlations between the hypothalamus and other brain region responses, and hormone responses to fructose and glucose ingestion.

Results: There was a significantly greater reduction in hypothalamic CBF after glucose vs fructose ingestion (-5.45 vs 2.84 mL/g per minute, respectively; mean difference, 8.3 mL/g per minute [95% CI of mean difference, 1.87-14.70]; P = .01). Glucose ingestion (compared with baseline) increased functional connectivity between the hypothalamus and the thalamus and striatum. Fructose increased connectivity between the hypothalamus and thalamus but not the striatum. Regional CBF within the hypothalamus, thalamus, insula, anterior cingulate, and striatum (appetite and reward regions) was reduced after glucose ingestion compared with baseline (P < .05 significance threshold, family-wise error [FWE] whole-brain corrected). In contrast, fructose reduced regional CBF in the thalamus, hippocampus, posterior cingulate cortex, fusiform, and visual cortex (P < .05 significance threshold, FWE whole-brain corrected). In whole-brain voxel-level analyses, there were no significant differences between direct comparisons of fructose vs glucose sessions following correction for multiple comparisons. Fructose vs glucose ingestion resulted in lower peak levels of serum glucose (mean difference, 41.0 mg/dL [95% CI, 27.7-54.5]; P < .001), insulin (mean difference, 49.6 μU/mL [95% CI, 38.2-61.1]; P < .001), and glucagon-like polypeptide 1 (mean difference, 2.1 pmol/L [95% CI, 0.9-3.2]; P = .01).

Conclusion and relevance: In a series of exploratory analyses, consumption of fructose compared with glucose resulted in a distinct pattern of regional CBF and a smaller increase in systemic glucose, insulin, and glucagon-like polypeptide 1 levels.

Figure 1. Mean Change in Hypothalamic Cerebral Blood Flow, Mean Plasma Glucose Response, and Mean Plasma Insulin Response

X-axis represents time points when hypothalamic cerebral blood flow (CBF) was measured (A) or when blood was sampled (B,C). Error bars indicate 95% CIs. To convert glucose values to mmol/L, multiply by 0.0555; insulin values to pmol/L, multiply by 6.945.

Figure 2. Regional Cerebral Blood Flow Response to Ingestion of Glucose or Fructose

A, Regional cerebral blood flow (CBF) response to glucose ingestion. B, Regional CBF response to fructose ingestion. The images represent paired t tests for postdrink vs baseline for 20 participants. The blue regions identify the areas in the brain with significantly decreased regional CBF after glucose or fructose ingestion compared with baseline. There were no brain regions with increased regional CBF after either fructose or glucose ingestion. Significance threshold set at P<.05, 2-tailed, family-wise error whole-brain corrected. Z is defined from top to bottom on the Montreal Neurological Institute (MNI) atlas with the origin passing through the anterior commissure- posterior commissure (AC-PC) line. MNI coordinates were used to define brain regions.

Figure 3. Functional Connectivity Analysis

A, Functional connectivity analysis for glucose ingestion at baseline, with bilateral hypothalamus as the seed region. Hypothalamus response to glucose ingestion was functionally connected to the caudate, putamen, and thalamus response. B, Functional connectivity analysis for fructose ingestion at baseline, with bilateral hypothalamus as the seed region. Hypothalamus response to fructose ingestion was functionally connected to the thalamus response. The images represent paired t tests for postdrink vs baseline for 20 participants. Yellow and red regions identify areas in the brain with magnetic resonance imaging signal responses correlated with the hypothalamic response. Significance threshold set P<.05, 2-sided, family-wise error whole-brain corrected. Montreal Neurological Institute (MNI) coordinates were used to define brain regions.

Figure 4. Region-of-Interest Correlation Analysis

A, Axial brain slice representing averaged data for 19 participants. Pink regions illustrate brain areas with a change in regional cerebral blood flow (CBF) that correlated with the change in plasma insulin levels after glucose ingestion. Montreal Neurological Institute (MNI) coordinates were used to define brain regions. B, Corresponding scatterplot showing the correlation between change in plasma insulin levels and the change in regional CBF to the left caudate and putamen following glucose ingestion for the 19 participants. The solid line in the scatterplot corresponds to the regression line (line of best fit). Difficulties with blood sampling in 1 participant limited analysis to 19 participants. To convert insulin values to pmol/L, multiply by 6.945.