Do all placebos fit the definition of a “placebo”? The variation in glycemic response of different placebos in healthy individuals

BACKGROUND Placebo is defined as an inert substance not possessing any biological effect. But this definition may not apply to all placebos. We previously noticed that a cornstarch‐placebo decreased the glycemic response to a 75g‐oral glucose tolerance test (75g‐OGTT) compared with a water‐control. Our objective was to investigate the effect of five common placebos on glycemia. METHODS Using a double‐blind, randomized, multiple‐crossover design, 10 healthy subjects (gender:6M:4F, age:33.1±4y, BMI:27.1±1.7kg/m2) received 7 treatments: 9g glucose‐placebo, lactose‐placebo, lactulose‐placebo, wheat‐bran‐placebo, and cornstarch‐placebo and two water controls. Each treatment was given 40‐min before a 75g‐OGTT with blood drawn at −40, 0, 15, 30, 45, 60, 90, and 120‐min. RESULTS Two‐way ANOVA showed a significant effect of treatment(p<0.001) and time (p<0.001) on incremental glycemia, with no interaction. Glucose‐placebo significantly reduced AUC by 41±15.4% (p<0.001) and peak glycemia (p<0.005) compared with the mean of the two water‐controls. It also reduced AUC (p<0.001) and peak glycemia (p<0.005) significantly compared with the cornstarch, lactulose and wheat‐bran placebos. CONCLUSIONS Glucose may not fit the definition of a true placebo. The implication is that the use of glucose as a placebo may lead to underestimation of efficacy in glycemic testing protocols. Travel grant: Inovobiologic, Calgary, AB.