An evaluation of some microbiological and ATP bioluminescence methods for the recovery and detection of bacterial contamination from food contact and environmental surfaces

Abstract

Plant hygiene and food contact surface cleanliness are key prerequisites to the management of food quality and safety, and may form a critical control point within Hazard Analysis Critical Control Point food safety management systems. Several methods exist with which to monitor food contact surface cleanliness, with a recent survey of the UK food industry indicating that ATP bioluminescence, cotton hygiene swabbing and agar contact methods are the most commonly adopted. Despite their widespread use, little is known about the relative efficiency with which these methods recover contaminating surface bioburden. The purpose of the work reported was to critically evaluate these hygiene methods for assessing food contact and environmental surface cleanliness within the food industry. On surfaces sampled while dry, cotton swabbing was found to be the least efficient of the methods, with bacterial recovery rates ranging from < 0.1% on surfaces sampled while dry, and from 0.25% to l6% on surfaces sampled while wet. Minimum detection limits (MDLs) ranged from 102 to 108 cfu/100 cm2 depending upon surface moisture level, organism type and the nature of the organism release method used. Absolute recovery rates were influenced by organism type and by a number of sampling variables, with surface moisture level having the greatest effect on recovery. Organism recovery rates were not found to vary greatly over swab storage times typical of those found in industry during swab transportation, but the method was found to have poor reproducibility with coefficients of variation of up to l64% being recorded for sampling marginally unclean stainless steel surfaces. Agar contact dip slides were found to be more reproducible than cotton swabbing, with minimum detection limits on inoculated surfaces sampled while wet being consistent at 102 cfu/100 cm2, and from 102 cfu/100cm2 to >107 cfu/100 cm2 on inoculated surfaces sampled while dry. Different ATP detection systems were found to have different minimum detection limits when individual components of total ATP detection limit were evaluated. These ranged from 104 to 106cfu/100 cm2 when used to sample inoculated stainless steel surfaces while dry. On identical inoculated surfaces sampled when either wet or dry, the minimum detection limit was found to be consistent at l04cfu/l00 cm2. A technique for determining microbial ATP levels was developed. Microbial ATP values from a range of food contact and environmental surfaces within different food processing environments correlated well with microbial colony count data, with R2 values ranging from 0.65 to 0.93 before cleaning and from 0.50 to 0.94 after cleaning. Results are discussed within the context of surface cleanliness assessment in the food industry and should help industry develop appropriate strategies for surface hygiene monitoring.