2.12. Microbiological analysisIn or

2.12. Microbiological analysis
In order to control the microbiological safety of the product and also
to determine the shelf life during refrigerated storage microbial counts
were performed at different times during refrigerated storage at 4 °C
on: i) samples dipped in the preservative solutions and then submitted
to 300 or 600 MPa; ii) beef samples without any treatment.
Approximately 10 g meat samples were taken from each package,
diluted with 90 ml of 0.1 g/100 ml peptone solution, and homogenized
in a Stomacher 400 (Seward, USA) for 120 s at the maximum speed. Serial
dilutions were performed using tubes with peptone 0.1 g/100 ml
and samples were plated on suitable culture media. After incubation,
the number of bacterial colonies grown on each plate were counted
and used to calculate colony forming units (CFU) per gramme of beef
for each sample.
The samples were tested for total aerobic mesophilic microorganisms
(Plate Count Agar PCA, 30 °C, 2 days), total psychrotrophic aerobic
counts (PCA, 4 °C, 7 days), Enterobacteriaceae (AVRB, 37 °C, 24 h),
lactic acid bacteria (MRS, 30 °C, 2 days), yeasts and moulds (YGC,
5 days, 30 °C). These determinations were made weekly in triplicates.
2.13. Statistical analyses
Statistical analyses were done using SYSTAT version 10.0 (SYSTAT Inc.,
USA, 1996). Analysis of variance (ANOVA) was carried out to analyse significant
effects. Least significant difference tests (LSD) were used to perform
pair-wise comparisons between means. Differences in means and
F-tests were considered statistically significant when P b 0.05. Average
values ± standard error of the mean are given throughout the text.
Surface response methodology was applied to model the effect of
solution composition on colour and texture variables. The desirability
function approach was used for optimising multiple responses, to find
the adequate combination of sodium nitrite and ascorbic acid that
produced the best colour in the products expressed in terms of L* and
a*. These procedures were carried out using the Design Expert v.7 software
(State-Ease, Inc., MN, USA).