A three-year survey of Florida packinghouses to determine microbial loads on pre- and post-processed tomatoes

Prevention of microbial cross-contamination during postharvest handling is an important step to minimize microbial food safety hazards of produce. Dump tanks and flume systems are widely used in many states (e.g., Florida) to transfer/wash tomatoes, and are one of the most critical points where cross-contamination may occur. This study presents data gathered over three years (2013, 2014, and 2015) on tomatoes collected from five growing regions in Florida to evaluate the risk associated with postharvest processing of tomatoes in commercial packinghouses. A total of 840 and 839 composite samples, from pre- and post-processed tomatoes, respectively, were analyzed for aerobic plate count (APC), and total coliforms (TC) and generic E. coli (EC). The least square mean (LSM) value of APC for all samples (both pre- and post-processed) was 6.0 log CFU/tomato (n = 840), whereas the LSM for TC counts was 4.1 log CFU/tomato (n = 839). Ninety-one (10.8%) and 820 (97.7%) out of 839 samples of post-processed samples had TC and generic EC counts below the detection limit of 1.3 log CFU/tomato, respectively. APC and TC counts in post-processed samples were significantly lower (p < 0.0001) than those in the pre-processed samples. There was no significant difference (p = 0.1011) in the occurrence of generic EC pre- and post-process. APC and TC were significantly higher (p < 0.0001) on samples collected in 2014 than 2013 and 2015, while the generic EC levels were not significantly different between 2013 and 2014. All samples collected in 2015 were negative for generic EC. TC counts varied significantly (p < 0.0001) by season, with highest counts in summer and lowest in the winter, over the three-year period. APC were significantly (p < 0.0001) higher in summer and fall seasons as compare to winter and spring. Microbial loads were significantly higher in the northern sites compared to the southern sites. Tomatoes from site 5 (southernmost) had significantly lower APC and TC (p < 0.0001) than recorded from other four sites. Data from this research demonstrated that the postharvest wash treatments used at the packinghouse surveyed in this study effectively reduced the overall microbial load and prevented cross-contamination.