Food Protection Trends - July/August 2021 - 426

pepper (0.15%; w/w), ground white pepper (0.15%; w/w),
ground nutmeg (0.13%), and freeze-dried starter culture
(Bactoferm F-LC; Chr. Hansen A/S, Hørsholm, Denmark)
that was reconstituted in approximately 20 mL of distilled
water. Following mixing, the salami batter was hand stuffed
(10 lb. Sausage Stuffer; The Sausage Maker, Inc., Buffalo,
NY) into 55-mm collagen casings (Globe Packaging, Inc.,
Carlstadt, NJ) and hand tied using butcher twine. Each
casing contained four salamis that were each approximately
260 grams in weight. A single salami in a separate casing
was used as a reference to measure weight loss and diameter
changes during the fermentation and drying process. A total
of 48 salamis were produced for each of the three replications
(n = 144). After stuffing, the salamis were hung in the drying
cabinet (AS50, Impianti Condizionamento Salumifici,
Camposanto, Modena, Italy) and fermented for 48 h at 23°C
and 95% relative humidity (±5%). Following fermentation,
the salamis were dried at 14°C and 75% relative humidity
to a target weight loss of 44%. Drying averaged 5 weeks
among three replications. When salamis completed drying,
they were individually vacuum packaged (Ultravac UV 250,
Koch Packaging, Kansas City, MO) in 3-mil food-grade
pouches (oxygen transmission rate of 50 to 70 cc/m2/24 h;
UltraSource LLC, Kansas City, MO) and stored at ~23° C
for an additional 4 weeks. The pH and aw
targets for the final
product were less than 5.3 and less than 0.90, respectively.
Sampling procedures
For each replication, samples were taken on days 0,
1, 2, 3, 5, and 10 and then every 7 days until day 66. On
each sampling day, three random salamis were taken for
enumeration of pathogens and one salami sample was taken
for determination of pH and aw
. On days 0 and 1, excised
surface samples (25 g) were taken aseptically from duck
trim and pork belly surfaces, prior to salami production.
For enumeration of pathogens in the final product, salamis
were cut in half, and the cores of each half were removed
using a sterile plastic spoon. Twenty-five grams of meat was
homogenized (Seward Stomacher 400 Circulator, Fermionx
Ltd., Worthing, West Sussex, UK) with 100 grams of buffered
peptone water (BD) for 1 min at 260 rpm, and then the
stomachate was serially diluted using 9 mL of buffered
peptone water (BD) blanks. Excised surface samples from
duck and pork belly were homogenized with 100 ml of
buffered peptone water, stomached, and serially diluted as
described above. Each sample was plated onto xylose lysine
deoxycholate agar (BD), modified Oxford agar (BD), CCDA
(Oxoid), and tryptic soy agar (TSA; BD) plates in duplicate
at the relevant dilutions and incubated using the methods
stated above. Resulting colonies were counted and averaged,
and the counts converted to log CFU/g. Colonies matching
the morphologies of Salmonella spp., L. monocytogenes,
and Campylobacter spp. were confirmed using latex
agglutination (Microgen Bioproducts) on each sampling
426 Food Protection Trends July/August
day. Enrichment procedures were conducted according to
the methods of McKinney et al. (31) for Salmonella spp.
and L. monocytogenes and were adapted for Campylobacter
spp. by using Brucella broth (Hi Media Laboratories) and
CCDA (Oxoid) and incubating using the parameters stated
above. pH was measured from the surface of either duck or
pork belly pieces on day 0 and day 1 (A57184 pH electrode,
Beckman Coulter, Inc.) or from the center of a salami for the
remaining samples (Testo 206 pH2, Testo North America,
West Chester, PA). Two pH measurements were taken on
each sampling day, one from each half of the fourth sample
salami. aw samples were measured from a slice taken from
the core of the salami (Aqualab 4TE, Meter Group, Inc.,
Pullman, WA). Only one aw measurement was taken on each
sampling day.
Statistical analysis
Resulting populations of Salmonella spp., L. monocytogenes,
and Campylobacter spp. were analyzed independently using
the general linear model with unique comparisons (α = 0.05)
(SAS 9.4, SAS Institute, Inc., Cary, NC). The detection
limit was calculated as 0.38 using a dilution factor of 5 and
0.5 CFU/g. In order to preserve statistical power, statistical
comparisons were made between the average pathogen
concentration from one sampling day and the concentration
of the same pathogen on the following sampling day.
Statistical analysis was not utilized for aw
and pH data.
RESULTS
Fermentation, drying, and vacuum-packaged storage
were able to achieve >5 log CFU/g reductions on average
in Salmonella spp. (n = 9, P < 0.0001), L. monocytogenes
(n = 9, P < 0.0001), and Campylobacter spp. (n = 9, P <
0.0001) (Table 1). Salmonella spp. achieved an average 5.47
log CFU/g reduction on day 38 of processing, followed by
a final average reduction of 7.03 log CFU/g on day 66. L.
monocytogenes achieved an average 5.20 log CFU/g reduction
on day 59 of processing, with a final average reduction of
5.90 log CFU/g. Campylobacter spp. achieved an average 6.85
log CFU/g reduction on day 45 of processing, with a final
average reduction of 7.19 log CFU/g. The average reductions
in Salmonella spp., L. monocytogenes, and Campylobacter spp.
that occurred after Beefxide treatment of 0.26 log CFU/g,
0.29 log CFU/g, and 0.20 log CFU/g, respectively, were
not significant. Pathogen concentrations did not increase by
more than 1 log CFU/g at any point during the experiment,
indicating that pathogens did not grow during the production
of duck salami. The concentration of bacteria on TSA (BD),
which was presumed to be starter culture, remained greater
than 5 log CFU/g on day 66. Enrichments were conducted
when pathogen concentrations reached levels below the
detection limit (0.39 log CFU/g) of the plating assay (n
= 9). By day 59, all samples produced negative Salmonella
spp. enrichment results. L. monocytogenes produced positive
Food Protection Trends - July/August 2021

Table of Contents for the Digital Edition of Food Protection Trends - July/August 2021
