Live microbes (LM) are naturally present in a variety of commonly consumed foods. These include raw fruits and vegetables, fermented foods, which are defined as foods made through desired microbial growth and enzymatic conversions of food components, and also probiotics, defined as live microorganisms providing health benefits to the host when consumed in adequate amounts (1-3). Interest in consumption of foods that harbor LM has increased recently due to evidence linking these foods to improved health outcomes and reduced risk of chronic diseases, including cardiovascular disease (1-3). However, most research on LM intakes has focused on US population samples or intake of specific foods, and not on total diet (1-3). To-date, no study has quantitatively assessed consumption of LM among the Australian population. The aim of this study was to develop a database for estimating and categorising live microbe (LM) content in common foods and beverages listed in the Australian Eating Survey® (AES) food frequency questionnaire (FFQ) (4). A secondary aim was to apply this database to examine the relationship between LM intake, cardiometabolic health indicators and inflammatory markers.
Estimation of LM content (colony-forming unit (CFU)/per gram) was based on previous published methods (1), with foods assigned to three categories based on expected prevalence of viable microorganisms in different food matrices. CFU LM density was defined as low (<104 CFU/g), medium (104–107 CFU/g), or high (>107 CFU/g). Estimated LM quantities were applied to Australian AUSNUT 2011-13 food and beverage items included in the AES® FFQ. Using cross-sectional baseline data from a sample of healthy adults (n = 58), usual LM intakes were estimated, and relationships with demographic data, anthropometry and clinical chemistry were evaluated. Results were analysed using adjusted weighted-least squares regression. Estimated LM quantities were determined for more than 200 AUSNUT food codes. The greatest number of items (n=226) were classified as low, with most being vegetables (22.6%), cereal-based products and dishes (13.3%), and meat, poultry, and game products and dishes (11.1%). Twenty-one food items were classified as medium (57.1% fruit, 28.6% vegetables, 14.3% milk products), and two items as high (yoghurt and cream sour). Consumption of LM was significantly inversely associated with body weight (p = 0.008), BMI (p = 0.005), waist circumference (p < 0.001), and plasma insulin concentration (p = 0.001), and positively associated with HDL-cholesterol (p < 0.001). No significant associations were observed with fasting, plasma glucose, total, LDL and non-HDL cholesterol, triglycerides and TNF-α.
Findings suggest that more frequent consumption of foods containing LM is associated with lower cardiometabolic and inflammatory risk factors in this sample. Further research is needed to evaluate LM intake and health outcomes in larger studies.