Abstract
Background: Diet is an important source of microbial exposure, potentially protecting against allergic disease. However, changes in dietary habits may have altered the intake of live food microbes. Aim: We quantitatively assessed the intake of live food microbes in Dutch adults and compared these estimates with those obtained from duplicate portions. Methods: In 34 Dutch adults (20-70 years), we estimated the food-microbial content of their diet for three dominant groups: total contaminating bacteria (TCB), lactic acid bacteria (LAB), and yeasts/moulds (YM). A food-microbial load database was compiled with minimum, best, and maximum estimated levels of these food microbes (in colony forming units (CFU)/g food). To estimate microbial intake, the amounts of food consumed (in grams) based on three 24 h dietary recalls were multiplied by the corresponding microbial content/gram of food. For validation, one 24 h duplicate portion per person was analysed for microbial content by conventional plate counting. We applied a one-way ANOVA to assess interindividual variation in microbial exposure, a two-way ANOVA to assess intraindividual variation in microbial intake, the so-called MOM2 analysis and regression analysis to identify foods contributing most to the level and variation of microbial intake, and Bland-Altman plots to assess the agreement between microbial intake estimated from 24-HDR and microbial counts determined in duplicate food samples. Results: The estimated total microbial exposure varied considerably across individuals (p < 0.05), ranging from 5.7 to 11.6 log (5.4 × 10(5)-4.4 × 10(11)) CFU/day. The exposure to TCB ranged from 2.5 to 11.4 log (3.0 × 10(2)-2.5 × 10(11)) CFU/day, LAB from 3.4 to 11.5 log (2.3 × 10(3)-3.0 × 10(11)), and YM from 2.6 to 9.6 log (3.6 × 10(2)-4.3 × 10(9)) CFU/day. Also, intraindividual exposure differed significantly (p < 0.05). Estimates were confirmed by microbial analysis of 24 h duplicate foods, revealing total microbial levels of 6.4-11.0 log (2.8 × 10(6)-9.4 × 10(10)) CFU/day. On average, the best estimated total microbial exposure was 0.6 log CFU higher than values determined by plate counting of the duplicate foods. Foods responsible for the observed variation were identified. Conclusions: The intake of live food microbes among Dutch adults varied considerably, ranging from nearly a million to more than 100 billion cells per day. Further validation of the food-microbial database is required.