ARTICLE IN PRESS

FOOD
MICROBIOLOGY
Food Microbiology 25 (2008) 313–323
www.elsevier.com/locate/fm

Microbial ecology of food contact surfaces and products of small-scale

facilities producing traditional sausages
Antonia S. Gounadakia, Panagiotis N. Skandamisa,b,,
Eleftherios H. Drosinosb, George-John E. Nychasa
a
Laboratory of Microbiology and Biotechnology of Foods, Department of Food Science and Technology,
Agricultural University of Athens, Iera Odos 75, Athens 11855, Greece
b
Laboratory of Food Hygiene and Quality Control, Department of Food Science and Technology, Agricultural University of Athens,
Iera Odos 75, Athens 11855, Greece
Received 20 May 2007; received in revised form 29 September 2007; accepted 1 October 2007
Available online 6 October 2007

Abstract

The microbial status in 7 small-scale facilities (SSFs) producing traditional fermented and/or dry sausages was investigated. It was
shown that the hygienic status of the processing environment and equipment plays an essential role in the microbial stability and safety of
the final products. The current study revealed that the majority of the sampling sites (control points) tested were highly (44 log CFU/
cm2) contaminated by spoilage flora (i.e. Pseudomonas, Enterobacteriaceae), with knives, tables and mincing machines being the most
heavily contaminated surfaces. Moreover, Listeria monocytogenes, Salmonella spp. and Staphylococus aureus were detected in 11.7%,
26.4%, and 11.7% of the food contact surfaces, respectively. The presence of these pathogens seemed to be associated with high numbers
of one or more specific groups of the ‘house-flora’ on the sampling sites of the facilities; however, high numbers of ‘house-flora’ do not
always suggest the presence of pathogens. With regard to product samples, batter samples were heavily contaminated with the ‘house-
flora’ present on surfaces and equipment of the processing facilities while by the end of processing (final products) LAB constituted the
predominant microbial flora of all products. The low initial levels of S. aureus and Salmonella found in batter samples as well as the
combination of hurdles (mainly awo0.92, average pH ca. o5.0 and competitive effect of natural flora) in the final products were able to
inhibit and/or eliminate these pathogens; however, the detection of L. monocytogenes in 3 out of the 7 final products examined is
indicative of cross-contamination. Our findings further indicate that inadequate hygiene practices within small-scale-processing facilities
may result in loss of microbial control. Therefore, this study addresses the need for strict control measures within SSFs producing
traditional fermented sausages.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: Microbial ecology; Hygiene practices; Safety control; Traditional sausages

1. Introduction improve food safety management systems (e.g. Hazard

Analysis and Critical Control Point, HACCP), as well as to
Traditional deli meats, such as fermented and/or dry comply with the imposed official food safety regulations in
sausages, are increasingly popular and represent a sig- order to provide safe and quality products to today’s
nificant proportion of the food market. This worldwide health-conscious consumers. Indeed, generally accepted
trend has forced traditional producers to adapt and prerequisite programmes such as Good Manufacturing
Practices (GMPs), Good Hygiene Practices (GHPs), and
Corresponding author. Laboratory of Microbiology and Biotechnol-
Standard Operating Procedures (SOPs) could provide the
foundation for effective HACCP implementation, in terms
ogy of Foods, Department of Food Science and Technology, Agricultural
University of Athens, Iera Odos 75, Athens 11855, Greece. of the basic environmental and operating conditions that
Tel.: +30 210 5294684; fax: +30 210 5294683. are necessary for the production of safe food (Maunsell
E-mail address: pskan@aua.gr (P.N. Skandamis). and Bolton, 2005). However, the management systems

0740-0020/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fm.2007.10.001
 ARTICLE IN PRESS
314 A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323

mentioned above is not an easy task for small-scale Considering the above, the objective of this study was to
producers, who experience difficulties in complying with survey the global microbial ecology (processing environ-
the prerequisite programmes generally adopted from large ment and products) of 7 small-scale facilities (SSFs)
processing plants (Mossel et al., 1998). This lack of control producing traditional fermented and/or dry sausages.
programs within traditional facilities producing fermented
and/or dry sausages is of particular concern for product’s 2. Materials and methods
safety, especially when their production relies on natural
microbial populations originating from raw materials and 2.1. Collection of samples
processing ‘house-flora’ or otherwise stated ‘resident
microflora’ (Leriche et al., 2003; Chevallier et al., 2006). Samples were collected from 7 SSFs, producing tradi-
Moreover, evidence exists that inadequate hygiene tional fermented and/or dry sausages. Sausages were
practices within food processing plants may result in the mainly manufactured with pork or beef lean, fat, NaCl
contamination of product with pathogens (Metaxopoulos and spices, with or without addition of starter cultures. All
et al., 2003) and therefore pose a subsequent risk in the facilities were divided into 3 main areas: reception of raw
product’s safety. On the other hand, complete elimination of materials, product processing, and product’s storage and
pathogens from raw materials (Eisel et al., 1997) and food handling. Surface samples were collected from 6 control
processing environment (Tompkin, 2002) is difficult, parti- points (CPs) of the processing environment and on
cularly when many food pathogenic are known to be able to equipment from each small-scale facility: 3 from machines
attach on food contact surfaces (Fonnesbech-Vogel et al., (mincing, mixing, and stuffing), 1 from the cutting table, 1
2001; Jessen and Lammert, 2003; Deza et al., 2005) and from wall of the storage cold room, and 1 from knives.
remain viable even after cleaning and disinfection (Frank Food-contact surface sampling was performed by swab-
and Koffi, 1990). Therefore, evaluation of the presence and/ bing a delimited area (5  100 cm2) according to the ISO
or the level as well as the control of technological, spoilage 18593 (2004) suggested protocol, after the scheduled
and pathogenic microorganisms is essential in order to cleaning and/or disinfection procedures. More specifically,
produce fermented and/or dry sausages that satisfy the swabbing was carried out with a sterile wet cloth
criteria of hygienic quality, organoleptic characteristics and (40 cm  40 cm; Laboratories Humeau, France), containing
food safety (Chevallier et al., 2006). 15 ml of a neutralizing solution (10% v/v). From each CP 5
Available studies exist for the evaluation of the effec- swabs were taken, 3 (3  100 cm2; n ¼ 3) for the enumera-
tiveness of currently applied hygienic practices within high- tion of technological and spoilage flora and 2 (2  100 cm2)
throughput food producers such as deli-meat producers. for detection of L. monocytogenes (n ¼ 1) and Salmonella
Numerous such studies have demonstrated that product spp. (n ¼ 1). Immediately after swabbing, all samples were
contamination may originate from the processing environ- stored at 4 1C and analyzed within 24 h.
ment (Eisel et al., 1997; Samelis and Metaxopoulos, 1999). Triplicate product samples were taken at different phases
However, the majority of these studies have focused either of the processing: (a) batter samples: just after stuffing, (b)
on monitoring the prevalence of specific pathogens such as 1-week sausages: after fermentation period, and (c) final
Listeria monocytogenes, Escherichia coli O157:H7 and products: at the time of commercialization. More specifi-
Salmonella spp. (Lunden et al. 2003a, b; Peccio et al., cally, at each sampling phase, 3 sausages (100 g) from the
2003; Holan et al., 2004) or on specific spoilage bacteria same batch were taken and stored at 4 1C and analyzed
such as Pseudomonas spp. within the processing environ- within 24 h. From each sample (100 g), sub-samples of
ment and equipment while the overall microbial ecology of 25 g were used for the microbiological analysis of
the food processing environment has not been fully technological and spoilage flora as well as the enrichment
elucidated. Several other studies have dealt with the of L. monocytogenes and Salmonella spp. as described in
microbial ecology of various meat products rather than the next paragraph.
on equipment, as an attempt to indicate possible micro-
biological risks (Samelis and Metaxopoulos, 1999; Drosi- 2.2. Microbiological analysis
nos et al., 2005; Murphy et al., 2005; Trevenot et al., 2005).
Taking into account the fact that the prevalence of For environmental samples, wet cloths were transferred
pathogens is strongly associated with the ‘house-flora’ of to stomacher bags containing 25 ml sterile buffered
the premises (Tompkin, 2002; Carpentier and Chassaing, peptone water solution (BPW, Merck) and macerated for
2004), there is a need for further analytical information on 1 min in a stomacher (Lab Blender 400, Seward Medical,
possible contamination sources within the processing London). For meat products, microbiological testing was
environment as part of the microbiological hazard conducted by the aseptic transfer of 25 g portions to
identification. Such data will also provide small-scale separate stomacher bags containing 225 ml sterile 0.1%
traditional deli-meat producers with information feasible peptone water with salt (NaCl, 0.85%, w/v). Samples were
to their processes, regarding the impact and the ability of then macerated for 1 min in a stomacher (Lab Blender 400,
current GMP/GHP practices applied in enhancing food Seward Medical, London). For environmental and sausage
safety control. samples, serial dilutions were made in 9 ml quarter strength
 ARTICLE IN PRESS
A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323 315

Ringer’s solution (Merck, Germany). In order to enumer- selective supplemented, Merck). After incubation at 42 1C
ate the flora, 1 or 0.1 ml portions of appropriate dilution for 24 h, the plates were observed for motility (opaque
were poured or spread onto duplicate plates of the halo). Presumptive Salmonella colonies were streaked onto
appropriate culture medium. Brilliant-green Phenol-red Lactose Sucrose Agar plates
Total viable count (TVC) were obtained from the (BPLS, Merck) and incubated at 37 1C for 24–48 h.
following plate count agar (PCA, Merck), incubated at Suspected Salmonella colonies were isolated and following
30 1C for 48–72 h; lactic acid bacteria (LAB) on the de preliminary testing (Gram-staining, oxidase) were further
Man, Rogosa, Sharpe agar (MRS, Merck), incubated at confirmed by a rapid latex agglutination test (Microgen,
30 1C for 48–72 h; Pseudomonas spp. on Cetrimide– Bioproducts, Surrey, UK).
Fucidin–Cephaloridine (CFC; Merck) agar, incubated at
25 1C for 48 h; Coagulase negative Staphylococcus and 2.4. Physicochemical analysis
Kocuria on Mannitol Salt red phenol agar (MSA, Merck),
incubated at 30 1C for 72 h; Enterobacteriaceae on crystal The pH was measured by immersing the pH probe of a
Violet neutral Red Bile Glucose Agar (VRBG, Merck), digital pH meter (pH 691, Metrohm) in a diluted and
incubated at 37 1C for 24 h; Enterococci on Enterococcus homogenized sample containing 10 g of sausage and 90 ml
Selective Agar acc. To Slanetz and Batley Agar (ME, of distilled water. The aw was measured with a calibrated
Merck); yeasts and moulds on Rose Bengal Chloramphe- electric hygrometer (HygroLab, Rotronic, Bassersdorf,
nicol agar (RBC, Oxoid), incubated at 25 1C for 3–5 days; Switzerland) using a sausage sample of 5 g, according to
Staphylococcus aureus on Baird–Parker agar supplemented the manufacturer’s instructions. pH and aw measurements
with Tellurite Yolk Egg (BP, Merck), incubated at 37 1C were performed in triplicates.
for 48 h. Colonies showing lecithinase activity on Baird–
Parker medium were tested for coagulase, using rabbit 2.5. Statistical analysis
plasma with EDTA (Biomerieux, France).
The microbiological testing (enumeration of technologi-
2.3. Enrichment procedures for L. monocytogenes and cal and spoilage flora) of all samples (CPs of the processing
Salmonella spp. environment and equipment and products samples) was
performed using triplicate samples. Bacterial counts were
For L. monocytogenes, a primary enrichment was converted in log CFU/g for sausage samples and in
performed by suspending 25 g of product samples or the log CFU/cm2 for environmental samples and further
whole wet cloth in 225 or 25 ml of 1/2 FRASER Listeria subjected to analysis of variance using the GLM procedure
selective enrichment broth (Merck), respectively. Following of XLSTAT statistical program (version 2006.06; Addin-
incubation at 30 1C for 24 h (primary enrichment), 0.1 ml soft, France). Tukey’s multiple-range test was used to
portion of samples were transferred to 10-ml FRASER discriminate means. Means were considered to be signifi-
Listeria selective enrichment broth (Merck) and incubated cantly different at the 95% significance level.
at 35 1C for 48 h (secondary enrichment). After each Triplicate samples were also used for the determination
enrichment step the culture was streaked onto both of pathogens (presence/absence) in 25 g of samples
PALCAM Listeria agar plates (30 1C for 48 h) (Merck) throughout processing. The results were reported as the
and Listeria Agar acc. to Ottaviani and Agosti plates number of positive (presence; 3/3) or negative (absence;
(37 1C for 24–48 h) (ALOA, Biolife). Presumptive Listeria 0/3) out of the total number of samples from the 7 facilities.
colonies isolated either from ALOA (blue-turquoise
colonies with or without a precipitation halo) agar plates 3. Results
or from PALCAM Listeria agar plates were confirmed
using biochemical tests. Specifically, suspect colonies were 3.1. Characterization of the SSFs
streaked onto Columbia blood agar (Oxoid, UK) and
incubated at 37 1C for 24 h. b-haemolytic colonies were A total of 55 samples (34 from CPs of the processing
further tested for motility at 25 1C, catalase reaction, and surfaces and equipment and 21 product samples) were
carbohydrate acid production from rhamnose (0.5%, collected from 7 (SSF 1–7) producing fermented and/or dry
Sigma), xylose (0.5%, Sigma), mannitol (0.5%, Sigma), sausages. The majority (6/7) of these SSFs produced
and a-methyl-D-mannoside (0.5%, Sigma) in purple carbo- naturally fermented and ripened sausages (Table 1);
hydrate broth base (Difco, Becton Dickinson). however, variation in the process conditions (duration
For Salmonella spp., enrichment was performed by and temperature of fermentation and/or ripening condi-
suspending 25 g of product samples or the whole wet cloth tions) between facilities was observed (Table 1). Sampling
in 225 or 25 ml of BPW solution, respectively. Initial of the processing surfaces and equipment was performed
suspensions of product and environmental samples were after the cleaning and/or disinfection procedures. The
incubated at 37 1C for 24 h. Following incubation, 3 drops majority (5/7 or 71.4%) of the examined SSFs used
were spotted onto the surface of modified Rappaport– domestic disinfectants (such as chlorine) for cleaning
Vasiliadis medium (MRSV, supplemented with MRSV processing surfaces and equipment, whereas only 42.8%
 ARTICLE IN PRESS
316 A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323

Table 1
Characteristics of the small-scale facilities (SSFs) producing traditional fermented sausages

Small-scale facility 1 2 3 4 5 6 7

Small-scale facility characteristics

Facility on the farm N N N Y Y N Y
Collective facility Y N N N N N N
Butcher on site N Y Y N N Y N
Temperature (1C) within the facility 12 10 17 12 13 15 15
HACCP Y N Y N Y N N
Approved facility by competent authorities Y N Y Y Y N Y
The establishment is producing in compliance of some GHPsa Y N Y N Y N Y
Use of disinfectant/sanitizer Y N N N Y N N
Process characteristics
Mix processing
Mincing machine Y Y Y Y Y Y Y
Cutter Y N N N N N N
Addition of starter culture Y N N N N N N
Addition of sugar Y N N N N N N
Addition of nitrite salt Y N N N N N N
Inoculation of the surface with moulds N N N N N N N
Inhibitors of mould growth N N N N N N N
Fermentation period
Fermentation temperature 1C 20 9 17 12 15–25 15 15
Length of fermentation (days) 7 5 5 3 3 3 2
Drying/maturation period
Natural dryer N Y Y Y Y N Y
Conditioned dryer Y N N N N Y N
Drying temperature 1C 14 12 17 12 15–25 15–25 15–25
Smoking Y N N N N Y Y
Length of drying (weeks) 7 3 2 2 2–3 2–3 2–3
Packaging V/A A A A A A A

Y: yes; N: no; V: vacuum; A: air.

a
Good Hygiene Practices.

(3/7) of the plants had a hazard analysis system in place While Staphylococci/Kocuria were detectable in 82.3%
(Table 1). (28 of 34) of the environmental samples, significant
variation (Po0.05) in their counts was observed in
3.2. Microbiological analysis of the processing environment specific sites of different facilities (Table 2). Their counts
reached 43 log CFU/cm2 in the 85.7% (6/7) of mincing
Considering the house-flora, the microbial counts varied machines as well as in 71.43% (5/7) of mixing machines,
(Po0.05) with the sampling site (CP) and facility (Table 2). knives and tables and the 33.3% (1/3) of stuffing
Specifically, LAB were detected in all of the CPs sampled, and mixing machines, respectively (Po0.05; Table 2).
with their counts ranging between o0.30 (detection limit) The highest counts of Staphylococci/Kocuria (7.6 log
and 7.43 log CFU/cm2, depending on the sampling site CFU/cm2) were recovered from the mincing machine
(Po0.05; Table 2). High LAB counts (45 log CFU/cm2) of SSF 6 (facility operating without HACCP; Tables 1
were recovered from 42.8% (3/7) of knives, 57.1% (4/7) of and 2).
cold rooms’ surfaces and from 33.3% (1/3) of mixing Counts of enterococci were below the detection limit
machines whereas counts in all (3) stuffing machines and in (o0.30 log CFU/cm2) in 17.6% (6/34) of samples from the
57.1% (4/7) of mincing machines and tables ranged mincing machine, the knife of SSF 1 as well as the tables of
between 2 and 5 log CFU/cm2 (Table 2). In any case, SSF 1 and 4 and the cold rooms SSF 3 and 4 (Table 2). In
LAB counts within each CP sampled differed significantly the majority (57.8%) of environmental samples, the levels
(Po0.05) between the majority of facilities (Table 2). The of enterococci were uniformly distributed between 1 and
highest LAB counts (7.43 log CFU/cm2) were recovered 4 log CFU/cm2 (Table 2). However, high levels of enter-
from the mixing machine of SSF 7 (facility operating ococci (44 log CFU/cm2) were recovered from 3 mincing
without HACCP), whereas the lowest counts were recov- machines (SSF 2, 3, and 5), 2 knives (SSF 3 and 4; Po0.05)
ered from all the sampling sites of SSF 1 and the cold room as well as from the tables of SSF 2 and 6 (Po0.05) and the
of SSF 3 (both facilities having HACCP; Tables 1 and 2). cold room of SSF 7 (Table 2).
 ARTICLE IN PRESS
A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323 317

Table 2
Counts (mean log CFU/cm2 and standard deviation (SD); n ¼ 3) of the observed microbial association on specific sampling sites within the processing
environment of different small-scale facilities (SSFs)

Control Facility LAB Staphylococcus/ Yeast and moulds Enterococcus Enterobacteriaceae Pseudomonas TVCa
point identification Kocuria

Mincing 1 o0.30 Ec (0.00)* o1.30 Eb (0.00)* 2.45 Ea (0.21) o0.30 Ec (0.00)* o0.30 Fc (0.00)* o1.30 Eb (0.00)* 3.00 Fa (0.00)
machines 2 4.90 Cc ( 0.11) 3.54 Dd (0.02) 2.84 Ee (0.09) 4.84 Bc (0.03) 6.58 Aa (0.04) 3.60 Cd (0.09) 6.18 Eb (0.02)
3 3.51 De (0.03) 4.99 Cb (0.12) 3.32 Df (0.03) 4.69 Bc (0.12) 3.10 Eg (0.01) 3.73 Cd (0.05) 7.25 Da (0.03)
4 3.70 Dd (0.04) 5.75 BCb (0.21) 4.56 Cc (0.04) 1.96 Ce (0.26) 5.38 Cb (0.04) 7.20 Aa (0.07) 7.27 Da (0.02)
5 3.55 Dd (0.01) 5.19 Cc (0.16) 3.32 Dde (0.09) 5.41 Abc (0.10) 5.61 Bb (0.02) 3.13 De (0.18) 7.59 Ca (0.02)
6 6.48 Ac (0.02) 7.60 Ab (0.43) 5.13 Bd (0.00) 1.30 De (0.00) 4.98 Dd (0.02) 7.44 Ab (0.01) 8.88 Ba (0.03)
7 5.53 Bc (0.07) 6.44 Bb (0.09) 6.19 Ab (0.11) 2.30 Ce (0.00) 5.37 Ccd (0.10) 5.11 Bd (0.02) 9.25 Aa (0.01)

Mixing 4 2.93 Bf (0.06) 5.60 Bc (0.00 ) 3.94 Bd (0.06) 3.00 Bf (0.00) 3.39 Ce (0.03) 5.87 Ab (0.02) 7.48 Ca (0.02)
machines 5 o0.30 Cf (0.00)* 5.24 Bd (0.14) 5.39 Acd (0.15) 3.59 Ae (0.08) 6.50 Ab (0.03) 5.76 Ac (0.14) 7.58 Ba (0.03)
7 7.43 Ab (0.11) 6.29 Ac (0.19) 5.47 Ad (0.01) 1.30 Cf (0.00) 3.88 Be (0.11) 3.56 Be (0.03) 8.28 Aa (0.02)

Stuffing 1 2.36 Bb (0.08) o1.30 Cd (0.00)* o1.30 Cd (0.00)* 2.14 Bbc (0.09) 1.84 Bc (0.34) 2.54 Bab (0.34) 2.90 Ca (0.14)
machines 3 2.56 Bbc (0.03) 2.75 Bab (0.21) 2.69 Bb (0.12) 1.84 Bde (0.09) 2.30 Bd (0.00) o1.30 Ce (0.00)* 4.48 Ba (0.07)
5 4.46 Ae (0.08) 5.69 Ac (0.12) 5.46 Ac (0.08) 2.60 Af (0.00) 4.87 Ad (0.08) 8.41 Ab (0.03) 8.70 Aa (0.02)

Knives 1 o0.30 Ge (0.00)* o1.30 Dd (0.00)* 4.34 Ca (0.03) o0.30 Ee (0.00)* 3.85 Dc (0.05) 4.06 Ebc (0.01) 4.18 Eab (0.26)
2 4.89 Dc (0.08) 6.02 Ab (0.34) o1.30 Ef (0.00)* 3.85 Cd (0.10) 5.64 Cb (0.04) 2.80 Fe (0.28) 7.46 Ca (0.03)
3 2.30 Fd (0.00) 2.45 Cd (0.21) 3.05 Dc (0.21) 4.30 Bb (0.00) o0.30 Ef (0.00)* o1.30 Ge (0.00)* 6.75 Da (0.08)
4 6.40 Ac (0.03) 4.60 Be (0.00) 6.94 Ab (0.02) 4.89 Ad (0.16) 6.80 Bb (0.01) 9.00 Aa (0.02) 9.04 Aa (0.05)
5 5.55 Bd (0.01) 5.63 Acd (0.07) 5.80 Bc (0.07) 3.60 Ce (0.00) 5.64 Ccd (0.02) 6.70 Cb (0.08) 8.23 Ba (0.03)
6 3.97 Ed (0.04) 4.15 Bd (0.00) 4.59 Cc (0.39) 1.30 De (0.00) 3.97 Dd (0.04) 5.90 Db (0.03) 6.42 Da (0.02)
7 5.37 Cb (0.01) 4.75 Bbc (0.21) 4.14 Cc (0.08) 1.30 Dd (0.00) 8.28 Aa (0.02) 8.25 Ba (0.00) 8.81 Aa (0.01)

Tables 1 o0.30 Fd (0.00)* o1.30 Dc (0.00)* 3.02 Da (0.15) o0.30 Ed (0.00)* 1.77 Fb (0.33) 3.15 Ca (0.63) 3.45 Da (0.53)
2 5.14 Bd (0.06) o1.30 Df (0.00)* 4.53 BCe (0.03) 4.58 Ae (0.03) 5.53 BCc (0.01) 6.45 Bb (0.01) 7.43 BCa (0.03)
3 3.06 Ed (0.11) 5.75 Aab (0.21) 3.08 Dd (0.25) 1.30 De (0.00) 3.71 Ebc (0.02) 3.36 Ccd (0.08) 7.14 Ca (0.03)
4 4.47 Cf (0.00) 6.04 Ac (0.06) 5.16 Ae (0.05) o0.30 Eg (0.00)* 5.79 ABd (0.02) 8.29 Ab (0.00) 8.85 Aa (0.06)
5 4.21 De (0.06) 5.19 Bd (0.16) 4.17 Ce (0.05) 3.23 Cf (0.04) 6.49 Ac (0.05) 6.78 Bb (0.00) 8.35 ABa (0.03)
6 5.89 Ac (0.02) 4.30 Ce (0.00) 4.94 ABd (0.08) 4.32 Be (0.03) 4.79 Dd (0.10) 8.26 Ab (0.01) 9.15 Aa (0.02)
7 2.99 Ee (0.04) 4.99 Bc (0.12) 3.37 Dd (0.16) 1.30 Df (0.00) 4.96 Dc (0.05) 5.98 Bb (0.00) 8.36 ABa (0.06)

Cold 1 o0.30 Ff (0.00)* 3.75 Ca (0.04) 2.89 Eab (0.16) 2.23 Cde (0.04) 2.54 Fcd (0.05) 2.69 Dbc (0.12) 2.80 Ebc (0.28)
rooms 2 5.26 Da (0.07) 3.04 Dc (0.06) 5.15 Cab (0.02) 2.37 Cd (0.10) 2.96 Ec (0.02) 5.32 Ba (0.03) 4.98 Db (0.02)
3 o0.30 Fc (0.00)* 6.17 ABa (0.12) o1.30 Fb (0.00)* o0.30 Ec (0.00)* o0.30 Gc (0.00)* o1.30 Eb (0.00)* 7.18 BCa (0.02)
4 3.15 Ed (0.05) o1.30 Ee (0.00)* o1.30 Fe (0.00)* o0.30 Ef (0.00)* 4.07 Cc (0.04) 4.68 Cb (0.01) 6.87 Ca (0.02)
5 5.55 Cc (0.07) 6.25 ABb (0.24) 4.55 Dd (0.03) 2.63 Bf (0.04) 3.76 De (0.04) 5.74 Ac (0.01) 7.38 Ba (0.01)
6 6.28 Bc (0.04) 6.29 Ac (0.00) 7.35 Bb (0.02) 1.30 De (0.00) 5.66 Bd (0.05) 5.80 Ad (0.08) 7.92 Aa (0.02)
7 6.60 Ab (0.03) 5.83 Bd (0.04) 7.83 Aa (0.12) 4.68 Af (0.05) 6.28 Ac (0.02) 5.23 Be (0.05) 7.92 Aa (0.01)

*Counts below the detection limit of the method; however for the means of statistical analysis the value of the detection limit (either 0.30 or 1.30) was used
for each microbial group. Within each row, underlined values are indicative of the dominant microorganism confirmed at significance level of 0.05.
abcdefg: means within a row sharing at least a common letter are not significantly different (Po0.05).
ABCDEFG: for each CCP, means within a column sharing at least a common letter are not significantly different (Po0.05).
a
Total viable count.

Counts of Enterobacteriaceae ranged between 4 and Considering the hygienic status of the SSFs, TVC were
9 log CFU/cm2, in 71.4% (5/7) of mincing machines and used to compare the overall contamination levels among
tables as well as the stuffing and mixing machines of SSF 5 the seven facilities. In general, the majority (27/34) of the
(Po0.05; Table 2). Such high levels were also recovered environmental samples were characterized by high con-
from 57.1% (4/7) to 42.8% (3/7) of knives and cold rooms, tamination levels (TVC levels of 46 log CFU/cm2) while
respectively (Po0.05; Table 2). The levels of Pseudomonas the statistical analysis revealed that in the 51.8% (14/27) of
as well as of yeasts and moulds also ranged between 4 and these samples Pseudomonas were the predominant micro-
9 log CFU/cm2, in 21 and 19 of all (34) environmental organisms followed by Enterobacteriaceae, yeasts and
samples tested, respectively (Po0.05; Table 2). Specifically, moulds and Staphylococci/Kocuria (Table 2). With regard
the levels of Pseudomonas exceeded 7 log CFU/cm2 in the to specific sampling sites, the knives of SSF 4–7, the tables
stuffing machine of SSF 5, the mincing machines and the and of SSF 2, 4–6 as well as the mincing machines of SSF 4
tables of SSF 4 and 6 (P40.05) as well as the knives of SSF and 6 were among the most heavily contaminated surfaces
4 and 7 (Po0.05; Table 2), whereas similar levels of yeasts (Table 2). Regardless of sampling site, SSF 1 was always
and moulds were recovered from the cold rooms of SSF 6 characterized by the lowest TVC levels (facility operating
and 7 (Po0.05; Table 2). with HACCP), while the highest TVC levels were recovered
 ARTICLE IN PRESS
318 A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323

Table 3
Overall prevalence and distribution of pathogenic bacteria on specific sampling sites within the processing environment

Control points Samples containing unsatisfactory levels (%)

L. monocytogenes Salmonella spp. S. aureus

a b c a
Positive samples SSFID % Samples Positive samples SSFIDb % Samples c
Positive samplesa SSFIDb % Samplesc

Mincing machines 1/7 (14.29) 4 2.94 2/7 (28.57) 2; 5 5.88 1/7 (14.29) 4 2.94
Mixing machines 0/3 (0.00) – 0.00 0/3 (0.00) – 0.00 1/3 (33.33) 4 2.94
Stuffing machines 1/3 (33.33) 5 2.94 0/3 (0.00) – 0.00 1/3 (33.33) 5 2.94
Knives 1/7 (14.29) 2 2.94 3/7 (42.86) 2; 5; 7 8.82 0/7 (0.00) – 0.00
Tables 0/7 (0.00) – 0.00 1/7 (14.29) 3 2.94 1/7 (14.29) 2 2.94
Cold rooms 1/7 (14.29) 6 2.94 3/7 (42.86) 2; 5; 6 8.82 0/7 (0.00) – 0.00
Total 4/34 11.76 9/34 26.46 4/34 11.76
a
% Distribution per control point.
b
Small-scale facility identification.
c
Overall samples.

Table 4
Changes (mean and standard deviation (SD); n ¼ 3) in product characteristics (pH, aw) at different manufacturing stages of the examined products

Small-scale facility 1 2 3 4 5 6 7

pH
Batter 5.68 Ab (0.06) 6.08 Ba (0.00) 4.48 Ce (0.01) 4.91 Ad (0.02) 4.92 Ad (0.02) 5.34 Bc (0.01) 5.28 Ac (0.01)
1-week sausage 4.66 Bc (0.16) 6.11 Aa (0.00) 4.76 Bc (0.06) 4.78 Ac (0.11) 4.71 Bc (0.01) 5.25 Cb (0.01) 5.26 Ab (0.02)
Final product 4.06 Cd (0.06) 6.04 Ca (0.00) 4.92 Ac (0.04) 4.73 Ac (0.01) 4.73 Bc (0.01) 5.72 Ab (0.01) 4.72 Bc (0.12)
aw
Batter nd nd nd nd nd nd nd
1-week sausage nd nd nd nd nd nd nd
Final product 0.865 d (0.01) 0.925 a (0.02) 0.915 ab (0.01) 0.885 cd (0.02) 0.905 abc (0.01) 0.920 ab (0.02) 0.895 bc (0.02)

nd: not determined.

abcdefg: means within a row sharing at least a common letter are not significantly different (Po0.05).
ABCDEFG: for pH readings, means within a column sharing at least a common letter are not significantly different (Po0.05).

from the processing environment of SSF 5 (facility levels of water activity (aw) was observed among final
operating with HACCP) and SSF 4, 6, and 7 (facilities products ranging between 0.92 and 0.87 (Table 4).
operating without HACCP) (Po0.05; Table 2). Regarding the microbial flora of the products, LAB
With regards to pathogenic flora, Salmonella was levels ranged between 4 and 8 log CFU/g in all batter
isolated from 4 sampling sites at a rate of 26.46% of the samples (Table 5). By the end of fermentation (1-week
samples, mainly from 5 facilities (Table 3). Similarly, sausage), LAB levels exceeded 8 log CFU/g (P40.05) in
L. monocytogenes was present at 4 sampling sites of 4 71.4% (5/7) of products (Table 5; Fig. 1). Conversely their
facilities resulting in an overall frequency of 11.7% over the counts remained stable or slightly increased during
total samples (Table 3). S. aureus was also found to persist ripening (end product) (Table 5) constituting the predomi-
in the environment of 3 facilities and it was isolated nant microbial flora of all products (Table 5; Fig. 1).
from 4 sampling sites at the same frequency as that of Significant variation (Po0.05) in the counts of Staphy-
L. monocytogenes (Table 3). lococci/Kocuria was observed (Table 5). Their population
ranged between 3 and 5 log CFU/g in 6 out of 7 batter
3.3. Microbiological analysis and physicochemical samples; however, their levels declined below 2 log CFU/g
characteristics of the sausages in 2 and in 5 of the products by the end of fermentation
and ripening, respectively (Table 5; Fig. 1).
With regards to the physicochemical characteristics of Enterococci were uniformly distributed between 2 and
products, significant variation (Po0.05) in the pH of 6.5 log CFU/g in 4 of the batter samples (Po0.05; Fig. 1),
the batter samples was observed between different SSFs whereas their counts declined below the detection limit
(Table 4). Specifically, in the majority (4/7) of final (o1 log CFU/g) by the end of ripening in 5 out of the 7
products, the pH varied from 4.72 to 4.92 (P40.05), products. Pseudomonas and Enterobacteriaceae were re-
whereas in 2 of the final products the pH was above of 5.3 covered from all (21) samples and their levels varied
(Table 4). In addition, significant variation (Po0.05) in the significantly between 2 and 8 log CFU/g(Po0.05; Table 5;
 ARTICLE IN PRESS
A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323 319

Table 5
Changes in the counts (mean log CFU/g and standard deviation (SD); n ¼ 3) of the observed microbial association at different stages during the
production of the examined products

Productsa Facility LAB Staphylococcus/ Yeast and Enterococcus Enterobacteriaceae Pseudomonas TVCb
identification Kocuria moulds

Batter 1 5.82 Bf (0.01) 3.00 Ed (0.00) 3.49 De (0.02) 1.81 Fd (0.05) 3.77 Cd (0.01) 3.79 Cf (0.01) 7.50 Ac (0.14)
2 7.69 Ab (0.00) 4.30 BCbc (0.03) 3.78 Cd (0.00) 2.18 Dd (0.20) 2.35 De (0.49) 4.87 Be (0.12) 8.18 Aa (0.04)
3 7.42 Ac (0.00) 4.12 Fc (0.16) 4.71 Ec (0.05) 6.35 Ca (0.04) 2.93 Ge (0.04) 5.22 Dd (0.11) 6.65 Be (0.00)
4 6.38 Be (0.05) 4.12 Fc (0.00) 4.72 Dc (0.02) o1.00 Ge (0.00)* 4.30 Ecd (0.03) 6.13 Cc (0.00) 7.00 Ad (0.00)
5 8.01 Ba (0.03) 7.13 Ea (0.07) 7.42 CDa (0.01) 5.65 Fb (0.07) 7.38 DEa (0.18) 8.34 Aa (0.04) 7.69 Cbc (0.00)
6 6.98 Ad (0.02) 4.60 Bbc (0.43) 4.95 Bb (0.07) o1.00 Ce (0.00)* 4.83 Bc (0.02) 7.00 Ab (0.03) 6.73 Ae (0.00)
7 4.02 Eg (0.01) 4.94 Db (0.00) 4.79 Dc (0.04) 3.54 Ec (0.34) 5.68 Cb (0.16) 7.00 Bb (0.04) 7.72 Ab (0.00)

1-week 1 8.82 Aa (0.00) 4.39 Dd (0.12) 2.87 Ee (0.12) 2.77 Ec (0.10) 4.91 Ce (0.02) 5.52 Be (0.06) 8.90 Ab (0.09)
sausage 2 8.67 Aa (0.01) o2.00 Ff (0.00)* 2.95 De (0.00) o1.00 Gd (0.00)* 2.30 Ef (0.00) 4.13 Cf (0.07) 7.11 Be (0.05)
3 8.84 Aa (0.11) 4.00 Ee (0.00) 5.30 Db (0.06) 6.48 Ca (0.04) 7.32 Ba (0.16) 7.11 Bb (0.03) 8.87 Ab (0.06)
4 7.80 Bb (0.01) o2.00 Ff (0.00)* 5.04 Ec (0.00) o1.00 Gd (0.00)* 6.79 Cb (0.02) 6.25 Dd (0.00) 8.56 Ac (0.00)
5 8.81 Ba (0.01) 7.90 Ca (0.01) 7.96 Ca (0.07) 6.00 Ea (0.06) 6.53 Dc (0.04) 7.94 Ca (0.06) 9.21 Aa (0.08)
6 7.76 Bb (0.09) 7.17 Cb (0.08) 5.02 Fc (0.02) o1.00 Gd (0.00)* 5.68 Ed (0.04) 6.94 Dbc (0.06) 8.80 Ad (0.12)
7 8.81 Aa (0.09) 6.72 Bc (0.01) 3.98 Ed (0.03) 5.24 Db (0.34) 5.84 Cd (0.02) 6.80 Bc (0.05) 8.04 Abc (0.00)

Final 1 9.18 Ab (0.00) o2.00 Db (0.00)* 5.62 Ec (0.01) 6.84 Ea (0.09) 4.91 Ce (0.02) 8.17 Ba (0.01) 8.73 Ab (0.03)
sausage 2 6.48 Bg (0.01) o2.00 Fb (0.00)* 3.60 Cf (0.00) o1.00 Gc (0.00)* 2.19 Ef (0.02) 3.00 Dd (0.00) 7.90 Ae (0.05)
3 8.73 Ad (0.12) 6.93 Da (0.04) 7.71 Bb (0.04) o1.00 Ec (0.00)* 7.32 Cb (0.16) 7.93 Ba (0.02) 8.25 Ad (0.02)
4 7.80 Bf (0.01) o2.00 Fb (0.00)* 5.11 Ed (0.05) o1.00 Gc (0.00)* 6.83 Cc (0.01) 6.30 Dc (0.00) 8.45 Ac (0.02)
5 8.52 Be (0.02) o2.00 Eb (0.00)* 7.89 Ca (0.08) 6.45 Db (0.21) 6.30 Dd (0.00) 8.09 BCa (0.20) 9.09 Aa (0.04)
6 8.98 Ac (0.02) 6.60 Ca (0.43) 5.08 Dd (0.02) o1.00 Ec (0.00)* 7.83 Ba (0.02) 7.11 Cb (0.04) 8.65 Ab (0.01)
7 10.03 Aa (0.03) o2.00 Eb (0.00)* 4.44 De (0.01) o1.00 Fc (0.00)* 2.00 Ef (0.00) 6.54 Cc (0.04) 8.45 Bc (0.02)

*Counts were below the detection limit of the method; however for the means of statistical analysis the value of the detection limit (either 1 or 2) was used
for each microbial group. Within each row, underlined values are indicative of the dominant microorganism confirmed at significance level of 0.05.
ABCDEFG: Means within a row sharing at least a common letter are not significantly different (Po0.05).
abcdefg: For each product period, means within a column sharing at least a common letter are not significantly different (Po0.05).
a
Batter sample: just after stuffing; 1-week sausage: after fermentation period; and final sausage: product at the time of commercialization.
b
Total viable count.

Fig. 1). More specifically, high levels (X7 log CFU/g) of small-scale businesses may lack the in-house knowledge
Pseudomonas were recovered from the batter samples of and resources to correctly implement HACCP. Indeed, the
SSF 5–7 while by the end of ripening such high levels were majority of the SSFs studied (Table 1) operated without an
recovered from the products of SSF 1, 3, 5, and 5 (Table 5). HACCP system in place, using insufficient sanitation
Moreover, high levels (X4 log CFU/g) of Enterobacteria- procedures (i.e. using solely rinsing with water or domestic
ceae were recovered from the final products of SSF 1, and cleaners). An insufficient cleaning and disinfection may not
3–6. Furthermore, yeasts and moulds were detected in all remove all meat or fat residues, and these residues may act
samples (Table 5) and their counts varied significantly as vehicles for spoilage or pathogenic bacteria from
between approx. 3 and 8 log CFU/g (Po0.05; Table 5; surfaces to meat (Gill and McGinnis, 2004) and thereafter
Fig. 1). to sausages.
Sausage samples were also examined for the presence It has been demonstrated that the ‘house-flora’ of
and distribution of pathogenic bacteria throughout proces- facilities producing traditional fermented and/or dry
sing (Table 6). S. aureus was not detected at any sample sausages includes useful microorganisms for the fermenta-
(Table 6). Salmonella and L. monocytogenes were recovered tion (e.g. LAB) and the flavour of sausage (i.e. Staphylo-
from batter samples of SSF 4 and 5 at a frequency of cocci/Kocuria), as well as spoilage (i.e. Pseudomonas) and
14.2% (Table 6). However, Salmonella was not detected at pathogenic flora (i.e.. L. monocytogenes, S. aureus)
any final product, whereas L. monocytogenes was detected (Chevallier et al., 2006). Indeed, the current study revealed
in 3 of the final products (Table 6). that the majority of the tested food contact surfaces were
highly contaminated by spoilage flora (Pseudomonas,
4. Discussion Enterobacteriaceae, and yeasts/moulds) (Table 2). Our
results clearly showed that the knives, the tables and the
It is widely recognized that GHPs form the basis or a mincing machines were among the most heavily contami-
vital part of HACCP (FAO/WHO, 2004). Under the nated surfaces within the SSFs (Table 2). In agreement with
current food hygiene regulations, all food businesses within other studies, our results also demonstrated the high
EU except for primary producers are required to operate variability of the microbial loads on surfaces and equip-
food safety management procedures based on HACCP ment in different facilities (Talon et al., 2007). Several
principles (EU Directive EC No. 852/2004). However, workers have attributed this variability to the microbial
 ARTICLE IN PRESS
320 A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323

100 Batter samples load of the processed food (Lunden et al., 2003b), the
environmental characteristics of the processing facility
80 (temperature, moisture of surfaces) (Chevallier et al.,
2006) and the efficiency of cleaning programs (Reij and
60 Den Aantrekker, 2004); however, the recovery of sub-
stantial numbers (e.g. 45 log CFU/cm2) of Pseudomonas,
40 Enterobacteriaceae, and TVC from the food contact
surfaces of most of the processing facilities, as found in
20 this study, was unexpected (Table 2). This is of particular
concern for the hygienic status of the facilities, since TVC
0 and levels of Enterobacteriaceae are widely used to provide
0-<2 2->4 4->6 6->8 >8 an indication of hygiene and the likelihood of post-
100
1-week sausages processing contamination (Eisel et al., 1997; FSAI, 2001)
as well as the presence of pathogens (FSAI, 2001).
Relative Frequency (%)

80
In this study, S. aureus was recovered from the food
60
contact surfaces of 3 processing facilities at a frequency of
11.8%. It is well known that S. aureus is a principal
40 component of human and animal skin (Borch et al., 1996)
and is frequently isolated from meat-processing facilities,
20 because the microorganism migrates from the carcass
surface to meat during the cutting operation (Borch
0 et al., 1996; Nel et al., 2004; Chevallier et al., 2006).
0-<2 2->4 4->6 6->8 >8 Furthermore, due to the warm and moist environment of
100 meat-processing facilities, proliferation of S. aureus may
Final products occur during processing, especially if cleaning and disin-
80 fection procedures are insufficient (Borch et al., 1996).
Therefore, the distribution of S. aureus within the house-
60 flora of the facilities as found in this study may be the
outcome of cross-contamination of the environment and
40 equipment from raw meat and/or from the food handlers.
However, its distribution may lead to post-process
20 contamination and therefore pose an increased risk in
products that support staphylococcal growth (i.e. products
0 of pH 44.3 or aw40.85; ICMSF, 1996).
0-<2 2->4 4->6 6->8 >8
Salmonella species were recovered from 26.5% of the
Lof CFU/g (Range)
environment and equipment of processing facilities. It has
TVC Pseudomonas Enterococcus been reported that once a production line (e.g. slaughter-
Staphylococci / Kocuria Yeasts & Moulds ing) is contaminated with Salmonella spp. the microorgan-
ism will establish itself on the machinery, equipment and
Enterobacteriaceae Lactic acid bacteria hands of workers (Berends et al., 1997). Indeed, a study in
Fig. 1. Overall distribution (%) of the observed microbial association at
8 slaughtering plants indicated that the prevalence of
different stages during the production of the examined products. Batter presumptive Salmonella spp. was 1.3% (Bacon et al., 2002).
sample: just after stuffing; 1-week sausages: after fermentation; final Moreover, heavily soiled surfaces would be expected to
products: at the time of commercialization. improve the survival of some Salmonella serotypes on such

Table 6
Occurrence of pathogenic bacteria at different stages during the production of the examined products in different facilities

Productsa L. monocytogenes Salmonella spp. S. aureus

Positive samplesb SSFIDc Positive samplesb SSFIDc Positive samplesb SSFIDc

Batter 3/21 (1) 5 (3/3) 1 (3/21) 4 (3/3) 0 (0/21) –

1-week sausage 0/21 (0) – 0 (0/21) – 0 (0/21) –
Final product 9/21 (3) 2 (3/3); 3 (3/3); 5(3/3) 0 (0/21) – 0 (0/21) –
a
Batter: batter sample; 1-week sausage: after fermentation; final product: ready to consumption.
b
Positive samples out of the overall samples tested; value in parenthesis indicate the number of facilities gave positive samples.
c
Small-scale facility identification; number in parenthesis indicate the positive replicates of the facility.
 ARTICLE IN PRESS
A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323 321

surfaces (De Cesare et al., 2003). Therefore, the markedly (Table 5). Their levels in the final products were similar
higher results obtained in this study may be explained by to those found in other naturally fermented sausages
the heavily soiled nature of the food contact surfaces (Rebecchi et al., 1998; Samelis et al., 1998; Cocolin et al.,
analyzed. 2001; Lebert et al., 2007). Accordingly, the levels of
Consistent with other studies, our results showed that Staphylococci/Kocuria in the majority (6/7) of batter
11.8% of the environmental samples were positive for samples were similar to those found in other naturally
L. monocytogenes (Salvat et al., 1995; Tompkin, 2002; fermented sausages (o5 log CFU/g) (Samelis et al., 1998;
Chevallier et al., 2006). In a previous study, it was shown Drosinos et al., 2005); however, by the end of ripening
that on locations where high aerobic counts were harbored (final products) their population reduced to below the
Enterobacteriaceae were accompanied, and in such loca- detection limit (2 log CFU/g) in 5 out of the 7 products.
tions L. monocytogenes was also likely to exist (Jessen and This elimination reflects their poor competitiveness in the
Lammert, 2003). Other studies have related the ability of presence of actively growing aciduric bacteria (Samelis
this organism to colonize food contact surfaces, in the et al., 1998; Drosinos et al., 2005).
presence of Pseudomonas spp. (Sasahara and Zottola, High levels of enterococci were enumerated in 2 of the
1993) or Listeria innocua (Bourion and Cerf, 1996); end-products tested (Table 5). Several studies have shown
however, in our study, the presence of L. monocytogenes that levels of enterococci increase at early fermentation
could not be correlated only with counts of Enterobacter- stages (4104 CFU/g), while their levels in the end-products
iaceae and Pseudomonas spp., since all sampling sites were may vary between 102 and 105 CFU/g (Rebecchi et al.,
also highly contaminated by spoilage (yeasts-moulds) 1998; Samelis et al., 1998; Montel, 2000; Franz et al., 2003).
or technological (LAB, Staphylococci/Kocuria) flora Moreover, high levels of spoilage flora (Enterobacteriaceae
(Table 2). Meat processing lines are continuously con- and Pseudomonas) were enumerated in the majority of the
taminated with L. monocytogenes through incoming raw final products studied (Table 5). In general, Enterobacter-
products, yet lines that utilize the fermentation process do iaceae and Pseudomonas are usually counted in the first
not remain persistently contaminating (Lunden et al., days of fermentation of both animal and plant origin
2003b). This suggests that competing flora may prevent foods, while their levels are usually eliminated during
L. monocytogenes from becoming established on processing maturation due to the strong competitive effect of LAB on
surfaces (Lunden et al., 2003b). Our results indicate that the rest endogenous flora (Samelis et al., 1998; Spyropou-
the competitive flora (LAB) had no destructive effect on lou et al., 2001; Drosinos et al., 2005). However, high levels
the recovery of pathogens since the presence of of these microorganisms have been reported to survive
L. monocytogenes was associated with the strong contam- during the maturation of fermented sausages (Lebert et al.,
ination of the processing environment by other flora (i.e. 2007), and such levels (4104 CFU/g) indicate product of
Pseudomonas, Enterobacteriaceae). low microbiological quality (FSAI, 2001).
Overall, spoilage (44 log CFU/cm2; Pseudomonas, Considering the safety of fermented sausages, a rapid pH
Enterobacteriaceae) and pathogenic flora were recovered drop below 5.3 during fermentation has proved to be
from the environment and equipment of almost all SSFs important for the inhibition of Salmonella and S. aureus if
tested (Table 2; SSF 2, 4–7), with SSFs 4–7 being the products are fermented at temperatures above 18 1C
characterized as the most heavily contaminated facilities. (Lucke, 2000). In the majority of products (6 out of 7) of
Of the 3 facilities (SSF 1, 3, and 5) that operated under the present study, the pH was below 5.3 during fermenta-
HACCP, 1 (SSF 5) had substantial microbial loads (Table tion and the temperature was below 18 1C. Nevertheless,
2) similarly or higher than those that did not have HACCP the low levels of S. aureus and Salmonella found in batter
in place (SSF 2, 4, 6, and 7). This suggests that in this facility samples as well as the combination of hurdles during the
the HACCP was not applied properly. Moreover according fermentation and ripening (aw, pH; Table 6) may have been
to food safety and inspection systems, L. monocytogenes is able to eliminate both pathogens. Regarding the contam-
considered ‘reasonably likely to occur’ especially due to post- ination rate of Listeria, our results (11.8% in batter) were
processing contamination and its detection must be addressed similar to those of other studies showing the incidence of
in the HACCP plan (Tompkin, 2002). Since control of L. monocytogenes in almost all sausage preparations
L. monocytogenes is rather achieved through a variety of (Samelis et al., 1998; Encinas et al., 1999; Trevenot et al.,
prerequisite programs (Tompkin, 2002) that form a solid base 2005). These results are indicative of the incidence of this
for HACCP (Maunsell and Bolton, 2005) if such programs pathogen in raw meat and in other materials used for
are inadequately applied, the presence of L. monocytogenes is sausage production (Samelis et al., 1998). Although the
very likely. Therefore, the occurrence of pathogenic and above-mentioned studies have shown that sausage ripening
spoilage flora within the processing surfaces and equipment tends to yield products with or without low levels of
as found in this study is more likely indicative of the poor L. monocytogenes, the detection of the microorganism in 3
cleaning and disinfection procedures applied and this may of the 7 final products raise concerns about possible
directly affect the quality and safety of the products. product cross-contamination. However, further investiga-
With regard to product contamination, LAB constituted tion is required to analyze higher number of products from
the major microflora of the fermented sausages studied various processing facilities.
 ARTICLE IN PRESS
322 A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323

In conclusion, the current investigation revealed that the References

hygienic status of the processing environment and equip-
ment plays an essential role in the microbial quality and Bacon, R.T., Sofos, J.N., Belk, K.E., Hyatt, D.R., Smith, G.C., 2002.
safety of traditional fermented sausages. Indeed, our Prevalence and antibiotic susceptibility of Salmonella isolated from
beef animal hides and carcasses. J. Food Prot. 65, 284–290.
results suggest that the incidence of certain pathogens, Berends, B.R., Van Knapen, F., Snijders, J.M.A., Mossel, D.A.A., 1997.
such as L. monocytogenes, Salmonella spp. and S. aureus in Identification and quantification of risk factors regarding Salmonella
food processing plants, may be associated with high spp. on pork carcasses. Int. J. Food Microbiol. 36, 199–206.
numbers of one or more specific members of the ‘house- Borch, E., Nesbakken, T., Christensen, H., 1996. Hazard identification in
flora’ (Tompkin, 2002; Carpentier and Chassaing, 2004); swine slaughter with respect to foodborne bacteria. Int. J. Food
Microbiol. 30, 9–25.
however, high levels of ‘house-flora’ are not essentially
Bourion, F., Cerf, O., 1996. Disinfection efficancy against pure-culture
indicative of the presence of a certain pathogen. The and mixed-population biofilms of Listeria innocua and Pseudmonas
insufficient cleaning and disinfection conditions applied in aeroginosa on steel, Teflon and rubber. Sci. Aliments 16, 151–166.
the majority of the facilities likely contribute to reconta- Carpentier, B., Chassaing, D., 2004. Interactions in biofilms between
mination since high levels of the ‘house-flora’ may Listeria monocytogenes and resident microorganisms from food
contaminate traditional sausages at any stage of the industry premises. Int. J. Food Microbiol. 97, 111–122.
Chevallier, I., Ammor, S., Laguet, A., Labayle, S., Castanet, V., Dufour,
processing. According to the microbiological criteria E., Talon, R., 2006. Microbial ecology of a small-scale facility
for foods illustrated by the EU Regulation 2073/2005, producing traditional dry sausage. Food Cont. 17, 446–453.
the majority of the products studied were characterized Cocolin, L., Manzano, M., Cantoni, C., Comi, G., 2001. Denaturing
as unable to support growth of L. monocytogenes (products gradient gel electrophoresis analysis of the 16S rRNA gene V1 region
with pHp4.4 or awp0.92, products with pHp5.0 to monitor dynamic changes in the bacterial population during
fermentation of Italian sausages. Appl. Environ. Microbiol. 67,
and awp0.94) and therefore are of low risk to consumers.
5113–5121.
Two products (SSF 2 and 6) marginally supported growth Commission of the European Communities, 2004. Regulation (EC) No
of L. monocytogenes and might be of higher risk to 852/2004 of the European Parliament and of the Council of 29 April
consumers; however, in a post-process-contamination 2004 on the hygiene of foodstuffs. Off. J. Eur. Union L 139. Available
event, the risk associated with the survival of pathogens, at: /http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:
such as L. monocytogenes, is likely to differ among different 2004:226:0003:0021:EN:PDFS. Accessed 20 February 2006.
Commission of the European Communities, 2005. Commission regulation
fermented and or dry sausages, mainly due to the (EC) No 2073/2005 of 15 November 2005 on microbiological criteria
complexity and the variety of manufacturing processing for foodstuffs. Off. J. Eur. Union L338, 1–26. Available at: /http://
used (Drosinos et al., 2006; Gounadaki et al., 2007; eur-lex.europa.eu/LexUriServ/site/en/oj/2005/l_338/
Skandamis and Nychas, 2007). Therefore, these highly l_33820051222en00010026.pdfS. Accessed 20 February 2006.
popular ready-to-eat products may be a source of exposure De Cesare, A., Sheldon, B.W., Smith, K.S., Jaykus, L.A., 2003. Survival
and persistence of Campylobacter and Salmonella species under
of consumers to L. monocytogenes, depending on
various organic loads on food contact surfaces. J. Food Prot. 66,
their physicochemical characteristics (pH and aw) and 1587–1594.
storage conditions (Gounadaki et al., 2007). However, Deza, M.A., Araujo, M., Garrido, M.J., 2005. Inactivation of Escherichia
the level of exposure remains to be evaluated, for until coli, Listeria monocytogenes, Pseudomonas aeruginosa and Staphylo-
now contamination levels lower than 103 CFU/g have coccus aureus on stainless steel and glass surfaces by neutral
never been documented for foods implicated in cases of electrolysed water. Lett. Appl. Microbiol. 40, 341–346.
Drosinos, E.H., Mataragas, M., Xiraphi, N., Moschonas, G., Gaitis, F.,
invasive listeriosis (FDA/CFSAN, 2003). Apart from Metaxopoulos, J., 2005. Characterization of the microbial flora
revealing the need for strict control measures within SSFs, from a traditional Greek fermented sausage. Meat Sci. 69,
in order to improve the hygiene of the processing 307–317.
conditions and thereby obtaining finished products of Drosinos, E.H., Mataragas, M., Veskovic-Moracanin, S., Gasparik-
higher microbiological standards, this study also suggests Reichardt, J., Hadziosmanovic, M., Alagic, D., 2006. Quantifying
nonthermal inactivation of Listeria monocytogenes in European
motives for further investigation on developing new
fermented sausages using bacteriocinogenic lactic acid bacteria or
disinfection procedures adapted to fermented meat their bacteriocins: a case study for risk assessment. J. Food Prot. 69,
products. 2648–2663.
Eisel, W.G., Linton, R.H., Muriana, P.M., 1997. A survey of microbial
levels for incoming raw beef, environmental sources, and ground beef
Acknowledgments in a red meat processing plant. Food Microbiol. 14, 273–282.
Encinas, J.-P., Sanz, J.J., Garcıa-Lopez, M.L., Otero, A., 1999. Behaviour
of Listeria spp. in naturally contaminated chorizo (Spanish fermented
This study has been partly carried out with the financial sausage). Int. J. Food Microbiol. 46, 167–171.
support of the Commission of the European Communities, Fonnesbech-Vogel, B., Jørgensen, L.V., Ojeniyi, B., Huss, H.H., Gram,
specific RTD program: Quality of Life and Management of L., 2001. Diversity of Listeria monocytogenes isolates from cold-
Living Resources, Key Action 1-Health Food, and smoked salmon produced in different smokehouses as assessed by
random amplified polymorphic DNA analyses. Int. J. Food Microbiol.
Environment, Project QLK1-CT-2002-02240 (TRADI-
65, 83–92.
SAUSAGE). It does not necessarily reflect the Commis- Food and Agriculture Organization of the United Nations (FAO) and the
sions views and in no way anticipates its future policy in World Health Organization (WHO), 2004. FAO/WHO guidance to
this area. governments on the application of HACCP in small and/or
 ARTICLE IN PRESS
A.S. Gounadaki et al. / Food Microbiology 25 (2008) 313–323 323

less-developed food businesses. Available at: /ftp://ftp.fao.org/docrep/ and HACCP system in Greek meat industry in the production of
fao/009/a0799e/a0799e00.pdfS. Accessed 3 May 2006. cooked sausages and cooked cured meat products. Food Cont. 14,
Food and Drug Administration, Center for Food Safety and Applied 323–332.
Nutrition (FDA/CFSAN), 2003.Quantitative assessment of the Montel, M.C., 2000. Fermented meat products. In: Robinson, R.K., Batt,
relative risk to public health from foodborne Listeria monocytogenes C.A., Patel, P.D. (Eds.), Encyclopedia of Food Microbiology.
among selected food categories of ready-to-eat foods. Available at: Academic Press, London, pp. 744–753.
/http://www.fsis.usda.gov/Science/Risk_Assessments/index.aspS. Mossel, D.A.A., Weenk, G.H., Morris, G.P., Struijk, C.B., 1998.
Accessed 20 August 2004. Identification, assessment and management of food-related micro-
Food Safety Authority of Ireland (FSAI), 2001.Guidelines for the biological hazards: historical, fundamental and psycho-social essen-
interpretation of results of microbiological analysis of some ready- tials. Int. J. Food Microbiol. 39, 19–51.
to-eat foods sampled at point of sale. Guidance Note No. 3. Murphy, R.Y., Hanson, R.E., Duncan, L.K., Feze, N., Lyon, B.G., 2005.
Available at: /http://www.fsai.ie/publications/guidance_notes/gn3.pdfS. Considerations for post-lethality treatments to reduce Listeria mono-
Accessed 15 April 2004. cytogenes from fully cooked bologna using ambient and pressurized
Frank, J.F., Koffi, R.A., 1990. Surface-adherent growth of Listeria steam. Food Microbiol. 22, 359–365.
monocytogenes is associated with increased resistance to surface Nel, S., Lues, J.F.R., Buys, E.M., Venter, P., 2004. The personal and
sanitizers and heat. J. Food Prot. 53, 550–554. general hygiene practices in the deboning room of a high throughtput
Franz, C.M.A.P., Stiles, M.E., Schleifer, K.H., Holzapfel, W.H., 2003. red meat abattoir. Food Cont. 15, 571–578.
Enterococci in foods—a conundrum for food safety. Int. J. Food Peccio, A., Autio, T., Korkeala, H., Rosmini, R., Trevisani, M., 2003.
Microbiol. 88, 105–122. Listeria monocytogenes occurrence and characterization in meat-
Gill, C.O., McGinnis, J.C., 2004. Microbial conditions of air knives before producing plants. Lett. Appl. Microbiol. 37, 234–238.
and after maintenance at a beef packing plant. Meat Sci. 68, 333–337. Rebecchi, A., Crivori, S., Sarra, P.G., Cocconcelli, P.S., 1998. Physiolo-
Gounadaki, A.S., Skandamis, P.N., Drosinos, E.H., Nychas, G.-J.E., gical and molecular techniques for the study of bacterial community
2007. Effect of packaging and storage temperature on the survival of development in sausage fermentation. J. Appl. Microbiol. 84,
Listeria monocytogenes inoculated post-processing on sliced salami. 1043–1049.
J. Food Prot. 70, 2313–2320. Reij, M.W., Den Aantrekker, E.D., 2004. Recontamination as a source of
Holan, J.T., Bird, J., Hall, K.E., 2004. The microbial ecology of high-risk, pathogens in processed foods. Int. J. Food Microbiol. 91, 1–11.
chilled food factories; evidence for persistent Listeria spp. and Salvat, G., Toquin, M.T., Michel, Y., Colin, P., 1995. Control of Listeria
Escherichia coli strains. J. Appl. Microbiol. 97, 68–77. monocytogenes in the delicatessen industries: the lessons of a listeriosis
ICMSF, 1996. Characteristics of microbial pathogens. In: Roberts, T.A., outbreak in France. Int. J. Food Microbiol. 25, 75–81.
Baird-Parker, A.C., Tompkin, R.B. (Eds.), Microorganisms in Foods, Samelis, J., Metaxopoulos, J., 1999. Incidence and principal sources of
Vol. 5. Blackie Academic and Professional, London, pp. 302–304. Listeria spp. and Listeria monocytogenes contamination in
International Organization for Standardization, ISO 18593, 2004. Micro- processed meats and a meat processing plant. Food Microbiol. 16,
biology of food and animal feeding stuffs—horizontal methods for 465–477.
sampling techniques from surfaces using contact plates and swabs, Samelis, J., Metaxopoulos, J., Vlassi, M., Pappa, A., 1998. Stability and
Switzerland. safety of traditional Greek salami—a microbiological ecology study.
Jessen, B., Lammert, L., 2003. Biofilm and disinfection in meat processing Int. J. Food Microbiol. 44, 69–82.
plants. Int. Biodet. Biodeg. 51, 265–269. Sasahara, K.C., Zottola, E.A., 1993. Biofilm formation by Listeria
Lebert, I., Leroy, S., Giammarinaro, P., Lebert, A., Chacornac, J.P., monocytogenes utilizes a primary colonizing microorganism in flowing
Bover-Cid, S., Vidal-carou, M.C., Talon, R., 2007. Diversity of systems. J. Food Prot. 56, 1022–1028.
microorganisms in the environment and dry fermented sausages of Skandamis, P.N., Nychas, G.J.E., 2007. Pathogens: risks and control. In:
small traditional French processing units. Meat Sci. 76, 112–122. Toldra, F. (Ed.), Handbook of Fermented Meat and Poultry.
Leriche, V., Briandet, R., Carpentier, B., 2003. Ecology of mixed biofilms Blackwell Publishing, London, pp. 427–454.
subjected daily to a chlorinated alkaline solution: spatial distribution Spyropoulou, K.E., Chorianopoulos, N.G., Skandamis, P.N., Nychas,
of bacterial species suggests a protective effect of one species to G.J.E., 2001. Survival of Escherichia coli O157:H7 during the
another. Environ. Microbiol. 5, 64–71. fermentation of Sanish-style green table olives (conservolea variety)
Lucke, F.K., 2000. Utilization of microbes to process and preserve meat. supplemented with different carbon sources. Int. J. Food Microbiol.
Meat Sci. 56, 105–115. 66, 3–11.
Lunden, J., Autio, T., Markkula, A., Hellstrom, S., Korkeala, H., 2003a. Talon, R., Lebert, I., Lebert, A., Leroy, S., Garriga, M., Aymerich, T.,
Adaptive and cross-adaptive responses to persistent and non-persistent Drosinos, E.H., Zanardi, E., Ianieri, A., Fraqueza, M.J., Patarata, L.,
Listeria monocytogenes strains to disinfectants. Int. J. Food Microbiol. Laukova, A., 2007. Traditional fry fermented sausages produced in
82, 65–272. small-scale processing units in Mediterranean countries and Slovakia.
Lunden, J., Autio, T., Sjoberg, A.M., Korkeala, H., 2003b. Persistent and 1: Microbial ecosystems of processing environments. Meat Sci., in
non-persistent Listeria monocytogenes contamination in meat and press, doi:10.1016/j.meatsci.2007.05.006.
poultry processing plants. J. Food Prot. 66, 2062–2069. Tompkin, R.B., 2002. Control of Listeria monocytogenes in the food
Maunsell, B., Bolton, D.J., 2005. Guidelines for Food Safety Management processing environment. J. Food Prot. 65, 709–725.
on Farms. Teagasc—The National Food Centre, EU-RAIN. Trevenot, D., Delignette-Muller, M.L., Christieans, S., Vernozy-Rozand,
Metaxopoulos, J., Kritikos, D., Drosinos, E.H., 2003. Examination of C., 2005. Fate of Listeria monocytogenes in experimentally contami-
microbiological parameters relevant to the implementation of GHP nated French sausages. Int. J. Food Microbiol. 101, 189–200.