No Effect of Sugar Sweetened Milk on Performance of a Battery of Cognitive Assessment Tests

Abstract

IntroductionDietary sugar has been alleged to play a role in the development of cognitive dysfunction. Experimental procedures in which pure fructose is consumed, often at doses exceeding normal human consumption, may produce changes in cerebral activity and associated cognitive impairment. The practical implication of this is unclear, however, because of the rarity of pure fructose consumption in humans in isolation from other sugars and macronutrients. Therefore the purpose of this study was to test the effects of typical fructose consumption on cognitive function and compare the effects of pure fructose to fructose consumed from more common sources such as high fructose corn syrup (HFCS) or sucrose.MethodsAll participants were apparently healthy and weight stable prior to enrollment (no change in weight >3% over the past three months; n=48, mean age 34.4 ± 11.1 years). Participants consumed sweetened or unsweetened low fat milk in amounts such that the added sugar contributed a target percentage of energy required for weight maintenance: fructose 9% (50th percentile of fructose consumption in the US), glucose 9%, high fructose corn syrup 18%, sucrose 18%; and an unsweetened milk control consumed such that milk contributed 18% of the weight‐maintenance calories over a 10 week period. The energy intake required for weight maintenance was estimated for each participant using the Mifflin St Jeor equation and an appropriate activity factor determined by responses to a physical activity questionnaire. Before and after the intervention participants were assessed on a battery of tests measuring a wide variety of cognitive functions (CANTAB). The tests used were: Spatial Working Memory (SWM), Intra‐Extra Dimensional Set Shift (IED), Stop Signal Task (SST), Delayed Matching to Sample (DMS), Paired Associates Learning (PAL), and Rapid Visual Information Processing (RVP). In addition, participants also performed the Iowa Gambling Test (IGT). Data presented are means ± SD.ResultsThere were small changes in weight (162.0 ± 28.5 vs 163.8 ±29.3) and BMI (26.0 ± 3.5 vs 26.3 ± 3.7) in the entire study population (p<0.001), but no effect of group assignment (interaction p>0.05). In the entire pooled study population, there were decreases in SST Reaction Time on Go (513.8 ± 179.2 vs 477.6 ± 146.2 ms, p<0.05) and DMS Latency (2982.7 ± 742.4 vs 2762.7 ± 569.6 ms, p<0.01), but neither measure was affected by group assignment (interaction p>0.05). However, Reaction Time on SST saw an increase in only the glucose group (146.9 ± 34.9 vs 186.1 ± 52.6 ms, p<0.05), which was significantly different to the change in SSTRT for sucrose and control (p<0.05). No other measures changed from pre to post intervention or were affected by group assignment (p>0.05).ConclusionThese data suggest that when fructose is consumed in the amounts and manner in which is typical in the human diet, previously reported changes in cognitive function are not observed. As such, these previous observations may be due to experimental conditions that do not reflect real life setting.Support or Funding InformationSupported by a grant from The Corn Refiners Association