Practical Utility of Lipid Predictive Equations in a Diabetic Population

Abstract

In the 1960s, Ancel Keys created an equation to predict how total blood cholesterol levels are affected by dietary modification. This equation has since been modified and applied to triglycerides, LDL, and HDL cholesterol. We assessed the practical utility of 11 published equations in Type 2 diabetics using self‐reported dietary data. In a larger clinical trial, 21 participants completed three daily food logs followed by a two‐week controlled feeding period. Fasting lipid profiles were assessed prior to study entry and after completion of the study diet. The difference between the participants’ personal diets and the study diet was used to predict lipid changes between the eating periods. The predicted lipid changes were compared to the observed lipid changes using a correlation analysis. For total cholesterol, the Original Keys Equation yielded the only predicted change that was significantly correlated with the observed change (r=0.45, p=0.04). For HDL cholesterol, the Yu equation yielded the only predicted change that was significantly correlated with the observed change (r=0.47, p=0.03). The predicted changes in LDL cholesterol and triglycerides were not correlated with the observed changes. Our findings demonstrate that for self‐reported dietary data, the Original Keys Equation provides the best estimate of total cholesterol changes in a diabetic population.Funding: American Pistachio Growers