A microbial evaluation of the effects of preparation method and storage of blended feeds

Background Patient-led feeding using blended (pureed) food via gastrostomy tube is increasing although it is not currently recommended (1). Concerns include the associated risk of micro-organism contamination in non-sterile blended feeds (2) which may cause gastrointestinal or systemic infection. Although the microbial contamination of blended feeds has been assessed, (3) there is a lack of data to estimate the effect of preparation method and storage on contamination. Therefore, the aim of this study was to undertake a microbial evaluation of blended feed prepared using different methods and stored for three different periods.
Method A blended feed was designed for a healthy, hypothetical 70-year old man and prepared using three methods: A) professional blender + extra fine sieve, cleaned using sterilising tablets; B) jug blender + standard sieve, cleaned with cold water; C) stick blender no sieve, cleaned with hot soapy water. Each feed was sampled immediately after preparation and after 24 and 48 hours in a domestic fridge. Samples were diluted, spread on seven types of agar and incubated aerobically (except Columbia blood agar: anaerobically) at 30 or 37C. Total colony forming units (CFU) were counted in triplicate. Bacterial colonies of unique morphologies were randomly selected for identification by Gram staining, oxidase test, catalase test and API 20NE strips. ANOVA was used to compare CFU across groups for method and storage. Ethical permission was not required.
Results There was no significant difference between total bacterial CFU of blended feeds prepared using different methods, P=0.771 (Table 1). The impact of storage time on bacterial CFU in blended feed A varied with increase in colonies on some agars but, overall, was not significantly different P=0.097 (Table 2). The genus of bacteria identified included Enterococcus, Bacillus, lactose-fermenting Enterobacteriaceae, Pseudomonas and Staphylococcus. Pathogens, such as Clostridium spp., Salmonella spp. and Vibrio spp., were not identified by phenotypic tests used.
Table 1 Comparison of mean total CFU/g of blended feeds prepared using three methods [DATA NOT DISPLAYED]
Table 2 Log CFU/g blended feed A grown on seven agar types after incubation [DATA NOT DISPLAYED]
Discussion The bacteria genera identified were expected due to the use of non-sterile food ingredients. It is not possible to predict whether these are potentially harmful without identifying the bacteria to species level using alternative approaches such as 16S rRNA PCR and sequencing. This study is limited to one hypothetical blended feed recipe, its lab-based design and by the absence of Listeria spp. testing. Future studies are required.
Conclusion There is potential concern about bacterial contamination of blended feeds but this does not appear to be influenced by the method of preparation or storage used in this study.
References
1Breaks, A., Smith, C., Bloch, S et al. (2018) Blended diets for gastrostomy fed children and young people: a scoping review. J. Hum. Nutr. Diet. doi: 10.1111/jhn.12563.
2British Dietetic Association (2013) Use of liquidised food with enteral feeding tubes, policy statement. Retrieved 7 July from https://www.bda.uk.com/improvinghealth/healthprofessionals/policystatement_liquidisedfood
3Sullivan, M.M., Sorreda-Wequerra, P., Santos, E.E. et al., (2001) Bacterial contamination of blenderized whole food and commercial enteral tube feedings in the Philippines. J. Hosp. Infect. 49:268-273.