Skip to main content
ARS Home » Pacific West Area » Davis, California » Western Human Nutrition Research Center » Obesity and Metabolism Research » Research » Publications at this Location » Publication #340926

Title: Structural equation modeling of food craving across the menstrual cycle using behavioral, neuroendocrine, and metabolic factors

item KRISHNAN, SRIDEVI - University Of California
item TRYON, REBECCA - State Of California
item AGRAWAL, KARAN - University Of California
item WELCH, LUCAS - Volunteer
item Horn, William
item Newman, John
item Keim, Nancy

Submitted to: Physiology & Behavior
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/10/2018
Publication Date: 7/11/2018
Citation: Krishnan, S., Tryon, R.R., Agrawal, K., Welch, L.C., Horn, W.F., Newman, J.W., Keim, N.L. 2018. Structural equation modeling of food craving across the menstrual cycle using behavioral, neuroendocrine, and metabolic factors. Physiology and Behavior. 195:28-36.

Interpretive Summary: Many women experience food cravings during the latter part of their menstrual cycle, but little is known about the underlying causes of this craving. In this study healthy young women with regular menstrual cycles were studied to explore how different biological and behavioral factors might interact to influence craving of sweet foods or craving of rich (fatty) foods. Structural equation modeling was used to delineate the important factors, and we found that certain derivatives of long chain fatty acids called endocannabinoids were associated with craving sweets and other related derivatives were associated with craving fats. These findings suggest that an alteration of the type of dietary fats consumed may be a key to changing the endocannabinoid profiles in the circulation, which in turn, might lessen craving symptoms.

Technical Abstract: The luteal phase craving behaviors often lead to increased food intake, and weight gain over time in women. The regulation of this craving is not well understood, and its manifestation also appears vastly different. Methods: Seventeen healthy premenopausal normal weight women were enrolled in this study. The volunteers were studied twice during their menstrual cycle - during the follicular and the luteal phase. We measured fasting estradiol, progesterone, leptin, and endocannabinoids (AEA and 2-AG – ECs) as well as several endocannabinoid like chemicals (ECLs), specifically several N-acylethoanolamides (NAEs) and monoacyglycerols (MAGs). We used a Food Craving Inventory (FCI) to record cravings for fat-rich and sweet-rich foods during the luteal phase. We also used three factor eating questionnaire (TFEQ) to measure restrained eating behaviors in these women. We used Spearman’s correlation tests and structural equation modeling (SEM) to understand the association between ECs, ECLs, ovarian and satiety hormones and cravings. Results: As expected, leptin and estradiol were inversely associated with ECs and ECLs (p<0.05). ECs and ECLs were positively associated with craving fats, and showed a trend towards being inversely associated with sweet-food craving in these women. SEM indicated that the follicular and luteal phase milieu were associated with cravings, and that MAGs were positively associated with craving sweets (RMSEA: 0.08, '2 p: 0.27, indicating no difference between hypothesized and actual model), while NAE’s were positively associated with craving fats (RMSEA = 0.00, '2 p: 0.49). Further, MUFA derived NAE’s (oleyoylethanolamide) was positively associated with craving fats, but ratio of PUFA derived NAE (eg: LEA – linoleyolethanolamide) to MUFA derived NAE, as well as ratio of SFA derived NAE (eg: SEA – stearoylethanolamide) were inversely associated with craving fats. Conclusions: Dietary fatty acid intake can alter circulating ECLs profiles. Cravings for sweet-rich or fat-rich foods during the luteal phase may be altered by altering dietary fatty acid intake.