Micro December 2018

Results were compared using a paired statistical analysis (8). Inoculating strains are listed in 1

2

Table 10.

In addition, 12 of the 31 food matrices (fresh raw ground beef, raw frozen chicken wings, 3

raw milk, whole liquid egg, tuna sushi, smoked salmon, bunched spinach, pasteurized carrot 4

juice, ready-made sandwiches, raw vegetable with salad dressing, chicken feed and soybean 5

meal) and both environmental surfaces were compared to ISO 4832:2006 and ISO 16649- 6

2:2001 reference methods as part of a harmonized validation study with MicroVal. As such, the 7

comparison to the ISO methods was conducted per ISO 16140-2:2106 validation guidelines (9). 8

The same inoculated lots of material were used for all comparisons. Five replicate portions for 9

each level of each matrix were prepared and diluted in PSS, as directed in the ISO methods, and 10

then tested by the 3M Petrifilm Rapid E. coli /Coliform Count Plate, ISO 4832:2006 and ISO 11

16649-2:2001. Results were compared using a paired statistical analysis. An uncontaminated 12

level was not tested in this comparison, as it is not required by ISO 16140-2: 2016. 13

To prepare the artificially contaminated foods, an isolated colony from each inoculating 14 strain was transferred from trypticase soy agar with 5% sheep blood (SBA) into ISO-BPW and 15

incubated at 35 ± 1 o C for 16–24 h. Following incubation, the culture was diluted to a desired 16

target level using ISO-BPW as the diluent and a bulk lot of each matrix was inoculated and 17

homogenized by hand. Refrigerated foods were held for 48–72 h at 2–8°C and frozen foods 18

were held for 2 weeks at -20°C or less prior to analysis. For all low moisture, shelf-stable foods, 19

a lyophilized inoculum was used to spike a bulk lot of each matrix and then homogenized. Low 20

moisture foods were held at ambient temperature (20–25°C) for two weeks. 21

16