The Women’s Diabetes Study was a clinical research study designed to evaluate the efficacy of fatty acid supplementation to the diet of post-menopausal women with diabetes. Conjugated linoleic acid and safflower oil were the treatment oils administered in this randomized, crossover design study. The purpose of this thesis research is to examine the baseline data taken from week 0 of the Women’s Diabetes Study in order to assess the relationships between dietary fatty acid consumption, plasma fatty acids, and markers of health. This is therefore, an observational study, a secondary analysis of the Women’s Diabetes Study.
The first hypothesis states that dietary fatty acid consumption would predict for plasma fatty acid composition. The second hypothesis states that plasma fatty acids would predict for health markers including fasting glucose, HbA1c, insulin, QUICKI, HOMA IR, total, LDL, and HDL cholesterol, CRP and IL6, two markers of inflammation, waist circumference, total body and trunk fat as measured by DEXA, adiponectin, and leptin. Two resulting aims from these hypotheses were to determine if underreporting of energy consumption could have confounded the dietary results and whether the use of oral hypoglycemic and statin medications could impact the markers of health.
We had anticipated correlations for linoleic acid, C18:2n6, an essential fatty acid, and for elaidic acid, C18:1n9, the most common industrially produced trans fat in the diet since neither are synthesized in the body; however, dietary and plasma levels of linoleic acid and elaidic acid were not correlated. Instead, dietary myristic acid, C14:0, was positively correlated with its plasma fatty acid. These data were unable to show a consistent relationship between diet and plasma fatty acid composition.
Upon examining the resting metabolic rate (RMR) as computed from the Mifflin St. Jeor equation, 32% of the subjects had less energy intake than their RMR. This suggests that energy intake was underreported in this study, possibly confounding the dietary to plasma fatty acid correlations.
Long chain omega 6 and omega 3 fatty acids including C:20:4n6, C20:5n3, and C22:6n3 were correlated with positive health markers. Arachidonic acid, required in the pro-inflammatory pathway of eicosanoid production, had no unhealthy correlations. Levels of arachidonic acid were inversely related to the ratio of total cholesterol to HDL. Levels of plasma NEFA positively predicted for both trunk and total body fat. The saturated fat, myristic acid, was inversely related to both trunk and total body fat in the regression models. Plasma fatty acids were indeed related to health markers.
Subjects taking a thiazolidinedione (TZ) or a biguanide had a significantly increased BMI and waist circumference. Additionally, subjects on biguanide had significantly increased weight and trunk fat while TZ users had a significantly increased HbA1c level, all of which were unexpected findings. Subjects taking a statin showed reductions in lipid and CRP levels, as expected. Drugs used for managing glycemia and lipids appeared to have effects that were significant in our models.
Future research examining the plasma levels of arachidonic acid and myristic acid is merited based on the outcomes of this study.