Oral sensory and cephalic hormonal responses to fat and non-fat liquids in bulimia nervosa

Abstract

Sensory evaluation of food involves endogenous opioid mechanisms. Bulimics typically limit their food choices to low-fat "safe foods" and intermittently lose control and binge on high-fat "risk foods". The aim of this study was to determine whether the oral sensory effects of a fat versus a non-fat milk product (i.e., traditional versus non-fat half-and-half) resulted in different subjective and hormonal responses in bulimic women (n=10) compared with healthy women (n=11). Naltrexone (50mg PO) or placebo was administered 1h before, and blood sampling began 30 min prior to and 29 min after, a 3 min portion controlled modified sham-feeding trial. Following an overnight fast, three morning trials (fat, naltrexone; fat, placebo; and non-fat, placebo) were administered in a random double-blind fashion separated by at least 3 days. Overall, there were no differences between Fat and Non-Fat trials. Hunger ratings (p<0.001) and pancreatic polypeptide levels (p<0.05) were higher for bulimics at baseline. Bulimics also had overall higher ratings for nausea (p<0.05), fatty taste (p<0.01), and fear of swallowing (p<0.005). Bulimics had approximately 40% higher total ghrelin levels at all time points (p<0.001). Hormones and glucose levels were not altered by the modified sham-feeding paradigm. Naltrexone, however, resulted in an overall increase in blood glucose and decrease in ghrelin levels in both groups (p<0.05, for both). These data suggest that bulimic women have different orosensory responses that are not influenced by opioid receptor antagonism, evident in hormonal responses, or dependent on the fat content of a similarly textured liquid.

Figure 1. Subjective ratings (mean ± SEM) of hunger and nausea before and after the modified sham-feeding session at all three trials

A: Bulimic subjects had overall higher hunger ratings and higher baseline values as indicated by * (p< 0.05). B: Bulimics also had overall higher nausea ratings, but there was no trial, time or interaction effects. In both figures, the black box represents the 3 min modified sham-feeding session.

Figure 2. Subjective ratings (mean ± SEM) of fatty taste and fear of swallowing 4 min after the modified sham-feeding session at all three trials

A: Bulimic subjects had overall higher fatty taste ratings, as indicated by * (p< 0.05). B: Bulimics were also more fearful of swallowing the solutions as indicated by * (p< 0.05).

Figure 3. Blood glucose (mean ± SEM) levels before and after the modified sham-feeding session and glucose (mg/dL) area under the curve at all three trials

A: There was an overall effect for glucose levels to be higher at the fat, naltrexone trials. The black box represents the 3 min modified sham-feeding session. B: Blood glucose AUC was significantly higher at the Fat, naltrexone trials compared with the other two placebo trials, as indicated by * (p<0.05).

Figure 4. Plasma insulin (mean ± SEM) levels before and after the modified sham-feeding session and insulin (μU/mL) area under the curve at all three trials

A: There was an overall effect for time effect for insulin, but no values were significantly different from baseline levels. The black box represents the 3 min modified sham-feeding session. B: Insulin AUC was not significantly different between trials.

Figure 5. Plasma ghrelin (total; mean ± SEM) levels before and after the modified sham-feeding session and insulin (pg/mL) area under the curve at all three trials

A: Bulimics had higher ghrelin levels at all time points as indicated by * (p<0.05). The black box represents the 3 min modified sham-feeding session. B: Plasma ghrelin AUC was significantly higher in bulimics compared with controls (*, p<0.05) and ghrelin levels were also significantly reduced in the naltrexone trials (**, p<0.05).