lable at ScienceDirect

Food Control 43 (2014) 231e237

Contents lists avai

Food Control

journal homepage: www.elsevier .com/locate/ foodcont

General patterns of background microbiota and selected bacterialpathogens during production of fermented sausages in Serbia

Miroslav Ducic a, Bojan Blagojevic a, Sinisa Markov b, Aleksandra Velicanski b,Sava Buncic a,*

aUniversity of Novi Sad, Faculty of Agriculture, Department of Veterinary Medicine, Trg D. Obradovica 8, 21000 Novi Sad, SerbiabUniversity of Novi Sad, Faculty of Technology, Bulevar Cara Lazara 1, 21000 Novi Sad, Serbia

a r t i c l e i n f o

Article history:Received 1 December 2013Received in revised form11 March 2014Accepted 18 March 2014Available online 26 March 2014

Keywords:Fermented sausagesSalmonellaEscherichia coli O157PorkBeef

* Corresponding author. Tel.: þ381 21 4853440; faxE-mail address: buncic_sava@hotmail.com (S. Bun

http://dx.doi.org/10.1016/j.foodcont.2014.03.0260956-7135/� 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

The presence of Salmonella spp. or Escherichia coli O157 and background microbiota, pH and aw weredetermined in raw fermented sausages produced from pork or beef and without lactic acid bacteriastarters. The investigation was conducted at five meat processing plants, and the sampling was done atfive steps of the production process at each plant. In meat trimmings, total viable count (TVC) rangedaround 6 log CFU/g and around 5e6 log CFU/g in the pork and the beef sausages, respectively. Enter-obacteriaceae count (EBC) ranged in the vicinity of 3e4 log CFU/g, whilst E. coli count (ECC) ranges werecomparably lower (by 1e2 logs). During chopping of both the pork and the beef trimmings, the levels ofTVC, EBC and ECC increased by 1e1.5 logs. After the additives and the spices were added, backgroundmicrobiota tended to slightly decrease, generally more noticeably in pork sausages and with ECC. Duringthe fermentation-drying stage, in both pork and beef sausages, initial TVC levels (6e7 log CFU/g)increased by the mid-process (by approximately 1.5e2 logs) and remained at those levels in finishedproducts. During the same period, lactic acid bacteria (LAB) increased from initial levels of 5.5e6 log CFU/g to around 7e8 log CFU/g in pork and around 8e9 log CFU/g in beef sausages, and became the pre-dominant microbial group. Salmonella spp. was found in the first three stages of the production process(trimmings, trimmings chopping, mixing with additives/spices), in two of three meat processing plants,but not at later stages of the production process. E. coli O157 was found only in one sample of choppedtrimmings in one meat processing plant. The background microbiota patterns and levels were, generally,similar to those commonly reported for raw fermented sausages in other published studies. The initialpresence of foodborne pathogens in raw fermented sausage production may be considered as a potentialmeat safety risk, because in the case of high initial pathogen counts, their total elimination cannot beassumed.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Raw, dry, fermented sausages (further: fermented sausages) areproducts with a long shelf life. Their shelf life relies on biochemicaland physicochemical changes in fermentation and drying processeswhich create a hostile environment for spoilage and pathogenicmicroorganisms. However, it is well known that fermented sau-sages sometimes contain bacterial foodborne hazards, i.e. patho-gens originating primarily from incoming raw materials includingmeat and fatty tissue (Nørrung & Buncic, 2008). The combination ofantimicrobial factors acting in fermented sausages (e.g. low pH,

: þ381 21 6350419.cic).

antagonistic lactic acid microbiota, low water activity) can preventgrowth and, at best, reduce the counts of bacterial pathogens. Thetraditional production process of fermented sausages leads to anappreciable reduction in pathogen counts (Adams &Mitchell, 2002;Nightingale, Thippareddi, Phebus, Marsden, & Nutsch, 2006), butthe total elimination of pathogens cannot be ensured in all cases. Inthe last two decades, epidemiological investigations have shownthat fermented sausages, as the main vehicles for foodbornepathogens, have been involved in several foodborne disease out-breaks (Ethelberg et al., 2009; Gieraltowski et al., 2013; Kuhn,Torpdahl, Frank, Sigsgaard, & Ethelberg, 2011; Luzzi et al., 2007;Sekse et al., 2009).

Verotoxigenic Escherichia coli (VTEC), especially O157:H7 sero-type, can cause bloody diarrhoea and life-threatening hemolytic-uraemic syndrome at low infectious doses in humans. Cattle are the

M. Ducic et al. / Food Control 43 (2014) 231e237232

principle reservoir for VTEC, and so consequently, this pathogencan contaminate carcass meat during cattle slaughter as well as thefermented sausages produced from such beef (Nastasijevic,Mitrovic, & Buncic, 2009). Hence, VTEC represent an importanthealth hazard associated with beef fermented sausages (Holcket al., 2011). This risk is potentiated by the facts that VTEC, partic-ularly E. coli O157:H7, appear to be relatively acid tolerant and, asthe infectious dose can be low, the pathogen’s growth in contam-inated foods is not necessary to cause associated foodborne illness(Sartz et al., 2008). On the other hand, Salmonella spp. is one of themajor causes of foodborne diseases in humans, and after eggs andpoultry meat, pork is regarded as the most important source offoodborne salmonellosis. As Salmonella is associated with freshpork, it may be present also in raw fermented sausages containingpork; consumption of which is relatively common in a number ofcountries (EFSA, 2006). The Salmonella risk posed by pork fer-mented sausages is potentiated by the pathogen’s ability to grow infoods with awater activity down to 0.94 (or sometimes even lower)or with pH > 4 (Nørrung, Andersen, & Buncic, 2009). Further, Sal-monella and E. coli O157 can originate not only from rawmeat/fattytissue, but also from other ingredients such as spices (Gieraltowskiet al. 2013; Holck et al. 2011). Consequently, raw fermented sau-sages may represent a significant microbial food safety risk forconsumers. Fermented sausages are one of the most popular typesof meat products in Serbia, and are also potentially very importantexport items, but published information on Salmonella and E. coliO157 in these Serbian products is very limited or is completelylacking, respectively.

Therefore, the main aim of the present study was to find if themost relevant bacterial foodborne pathogens occur in pork

Table 1Formulations and ripening conditions used for fermented sausages production.

Pork sausages (“Sremska”)

Industry A Industry B Industry C

FormulationsMeat trimmings 65 kg 65 kg 65 kgFatty tissue 30 kg 30 kg 30 kgSpices Paprika,

chillies, coriander,garlic, pepper(2.5 kg in total)

Paprika,chillies, coriander,garlic, pepper(2.5 kg in total)

Paprika,chillies, corgarlic, pepp(0.7 kg in t

NaCleNaNO2 mixture 2.5 kg 2.6 kg 2.5 kgDextrose 0 0.5 kg 0.35 kg

Glucono-d-lactone (GDL) 1 kg 1 kg 0.6 kgAscorbic acid 0.025 kg 0.05 kg 0.015 kg

Ripening conditionsDay Ta RHb T RH T

0 24 93 22 98 241 24 87 18 92 242 22 84 18 88 223 19 82 18 86 194 15 77 17 84 155 14 75 17 80 146 14 75 16 70 147 12 75 16 70 128 12 75 16 70 129 12 75 16 70 1210 12 75 16 70 1211 12 75 16 70 1212 12 75 16 70 1213 12 75 16 70 1214 12 75 16 70 12

a T ¼ temperature (�C).b RH ¼ Relative air humidity (%).

(Salmonella) and beef (E. coli O157) raw fermented sausagesthroughout their production processes. Furthermore, the sausageswere characterised with respect to their background microbiota,pH and aw, as these factors affect the fate of the pathogens. Thisinitial study was not aimed at comparing processing conditions indifferent commercial meat plants, but to provide a base-linerationale for our subsequent investigation into novel strategiesfor bacterial pathogen control in fermented sausages.

2. Materials and methods

2.1. Sausages

Investigations were conducted at five commercial meat pro-cessing plants located in the northern part of Serbia. Two types ofraw, dry, fermented sausages were investigated: a) pork sausages(“Sremska”; produced in plants A, B and C) and beef sausages(“Sudzuk”; produced at plants A, E and D). Sausages were producedusing standard raw material composition (Table 1) and standardripening procedure (Table 1); no meat plant used starter cultures.

2.2. Sampling

At eachmeat plant and for each of pork or beef sausage type, thesampling was done at following steps: a) chilled meat destined forfermented meat production (meat trimmings); b) chopped meattrimmings; c) complete sausage batter before stuffing in the casings(chopped trimmings þ fatty tissue þ additives); and d) sausages atthree stages of the production process: at the start (day “0”), at mid-process and at the end of the process. For microbiological analysis,

Beef sausages (“Sudzuk”)

Industry A Industry D Industry E

60 kg 60 kg 82 kg35 kg 35 kg 15 kg

iander,erotal)

Paprika,chillies, onion,pepper, mustard,marjoram,rosemary, basil(2.0 kg in total)

Paprika,chillies, coriander,pepper, turmeric(0.5 kg in total)

Garlic,pepper, celery(0.4 kg in total)

2.4 kg 2.6 kg 2.2 kg0 0 0.3 kg

(þSaccharose 0.2 kg)1.2 kg 1 kg 00.03 kg 0.025 kg 0

RH T RH T RH T RH

93 24 93 20 90 20 8887 24 87 18 88 18 8484 22 84 18 86 18 8082 19 82 18 84 18 8077 15 77 18 82 17 7875 14 75 18 82 17 7875 14 75 16 68 17 7875 12 75 16 68 16 7575 12 75 16 68 15 7575 12 75 16 68 15 7575 12 75 16 68 15 7575 12 75 16 68 15 7575 12 75 16 68 15 7575 12 75 16 68 15 7575 12 75 16 68 15 75

M. Ducic et al. / Food Control 43 (2014) 231e237 233

from each type of sausage, at each related meat plant and at eachabove sampling step, 6 samples were taken (216 in total). Forphysicochemical analysis, from each type of sausage at each relatedplant, 6 sausage samples were taken at three steps: start, middleand end of production process (108 in total). Each samplewas placed in a separate stomacher filter bag (Nasco Whirl-pack15e23 cm; USA) and transported in a chill-bin to the laboratorywithin 1 h.

2.3. Microbiological analysis

In the laboratory, all sausage casings were removed usingsterilised instruments to produce sausage samples. From eachsample (except meat trimmings) 25 g was taken, added to 225ml ofMaximum Recovery Diluent (MRD; Oxoid, UK), homogenized withstomacher (easyMIX, BioMerieux, France) for 2 min and furtherdecimal dilutions were made in MRD (ISO method 6887-1:1999).Samples of meat trimmings, weighing approximately 500 g, weremixed with MRD (ratio 1:1) in the bag, homogenized manuallyfrom outside so as to rinse the meat surface, and further decimaldilutions were made in MRD.

2.3.1. Determination of total viable count (TVC), Enterobacteriaceaecount (EBC) and generic E. coli count (ECC)

For TVC, 1 ml volumes from appropriate sample dilutions werespread onto Aerobic Count Plate Petrifilms (3M Health Care, St.Paul, USA), incubated at 30 �C for 72 h and the colonies counted(AFNOR validated method 3M 01/1-09/89). For EBC 1 ml volumesfrom sample homogenates or their appropriate dilutions werespread onto Enterobacteriaceae Count Plates Petrifilms (3M HealthCare, St. Paul), incubated at 37 �C for 24 h and typical colonies werecounted (AFNOR validated method 3M 01/06 09/97). For ECC, 1 mlvolumes from sample homogenates or their appropriate dilutionswere spread onto E. coli/Coliform Count Plates petrifilms (3MHealth Care, St. Paul), incubated at 37 �C for 48 h and typical col-onies were counted (NordVal validated method 3M 014-11).

2.3.2. Determination of lactic acid bacteria count (LAB)For LAB, after sample homogenisation and decimal dilutions in

MRD, final dilution steps was made in de Man, Rogosa and Sharpe(MRS) broth (Biokar Diagnostics, France) and 1 ml volumes werespread onto Aerobic Count Plate Petrifilms (3M Health Care) whichwere incubated anaerobically at 37 �C for 48 h.

2.3.3. Detection of E. coli O157 in SudzukE. coli O157 was selected as representative also for other Shiga-

toxin E. coli (STEC)/Enterohaemorrhagic E. coli (EHEC). From eachsample homogenate, 25 ml was transferred to 225 ml of enrich-ment medium (mEC þ Novobiocin selective enrichment broth;Merck, Darmstadt, Germany), incubated at 37��C for 24 h, followedby detection of E. coliO157 using an immunochromatographic rapidtest (Singlepath� E. coli O157; Merck). According to the manufac-turer’s manual, this E. coli O157 rapid test is AOAC validated, with adetection limit of 1 CFU/25 g sample and has both sensitivity andspecificity of >99%.

2.3.4. Detection of Salmonella spp. in SremskaAccording to ISO 6579:2002 method, each sample was incu-

bated in buffered peptone water (BPW, Oxoid, UK) for 18 h at 37 �C(pre-enrichment step). Subsequently, 1 ml was transferred to 10 mlof Muller-Kauffmann tetrathionate-novobiocin broth (MKTTn,Oxoid) and 0.1 ml was transferred to Rappaport-Vassiliadis Brothwith soya (RVS, Oxoid) and incubated at 37 �C and 41.5 �C,respectively, for 24 h (enrichment step). Then, both xylose lysinedeoxycholate agar (XLDA, Oxoid) and brilliant green agar (BGA,

Oxoid) plates were inoculated from RVS and MKTTn broths. All theplates were incubated at 37 �C for 24 h. Suspected colonies (withblack centre and/or pink from XLDA and pink from BGA) werepurified on a nutrient agar and confirmed using Vitek� 2 Compactequipment (BioMerieux, France) and serologically confirmed byusing poly O somatic anti-serum (Pro-Lab Diagnostics, Canada).

2.4. Physicochemical analysis

Water activity (aw) in sausage samples was determined byusing LABSwift-aw set Euro-plug&BAT equipment (Novasina,Switzerland); pH was determined by a hand-held pH meter (Testo205; Germany). Both methods were according to the manufac-turer’s instructions.

2.5. Analysis of results

The results were obtained by testing six samples for eachsampling point, hence six values for each parameter (microbio-logical or physicochemical) were used to calculate the means,standard deviation and differences betweenmeans (p< 0.95) usingMS Excel. Before that, TVC, EBC, ECC and LAB results (initiallycalculated as bacterial colony forming units; CFU per g of sample)were converted to logarithms10 CFU/g.

3. Results

3.1. Pork fermented sausages

Results for general microbial characterisation of pork sausagesare presented in Table 2. The general patterns of changes of back-ground microbiota (TVC, EBC, ECC, LAB) throughout the productionof pork sausages were similar in meat plants A, B and C (Table 2),hence will not be commented on individually for each plant. Withrespect to rawmaterials for pork sausage preparation, TVC, EBC andECC in meat trimmings were relatively high (around 6 log, 3e4 logand 1e3 log, respectively), and these further increased in the nextproduction step (trimmings chopping), and then somewhatdecreased after spices and additives were added to the batter. LABlevels were not investigated in the raw materials.

During pork sausage fermentation-drying, TVC reached 7.5e8.7 log CFU/g and then remained practically stable until the sausageproduction was completed (Table 2). EBC and ECC dramaticallydecreased during sausage the fermentation-drying stages, andwere not found in the finished sausages. The initial levels ofnaturally-occurring LAB (no starter cultures were used) immedi-ately after stuffing the batter into casings were around 5.5 log CFU/g(Table 2). Expectedly, during pork sausage fermentation-dryingstages, LAB significantly increased and this was predominant mi-crobial group (around 7e8 log CFU/g) among the backgroundmicrobiota at the time sausages were finished.

Salmonella spp. was selected to be investigated during theproduction process of pork sausages, as it is the most relevantbacterial foodborne pathogen associated with raw pork. Thepathogen was found in raw materials during sausage batter prep-aration in two meat plants, but not in the third (Table 2). In plant C,Salmonella spp. was found in meat trimmings, chopped trimmingsand in sausage batter with added additives. In plant A, the pathogenwas found only in choppedmeat trimmings. However, in both plantC and A, Salmonella spp. disappeared during the pork sausagefermentation-drying process i.e. was not found in finishedsausages.

The main physicochemical parameters in the sausages, pH andaw, are shown in Fig. 1. The initial pH of the sausages was relativelylow, around 5.4e5.5 in plant A and C and as low as 4.9e5.0 in plant

Table 2Main microbial characteristics of pork sausage during the production process.

Stage of production TVC EBC ECC LAB Salmonella spp.

Log10 CFU/g � SD

Meat plant AMeat trimmings 6.05 � 0.21 3.82 � 0.27 1.00 � 0.32 ND NFChopped trimmings 6.68 � 0.11 4.57 � 0.25 2.64 � 0.24 ND þ(1 sample of 6)Complete batter with additives 6.43 � 0.14 3.84 � 0.27 1.82 � 0.14 ND NFSausages at start (day 0) 6.23 � 0.07 3.39 � 0.34 1.13 � 0.55 5.58 � 0.14 NFSausages in mid-process (day 7) 8.08 � 0.22 1.54 � 0.18 NF 8.53 � 0.17 NFFinished sausages (day 15) 7.77 � 0.11 NF NF 8.23 � 0.11 NFMeat plant BMeat trimmings 5.95 � 0.22 3.91 � 0.54 2.91 � 0.48 ND NFChopped trimmings 7.22 � 0.11 5.59 � 0.09 3.43 � 0.34 ND NFComplete batter with additives 6.25 � 0.1 4.52 � 0.24 2.71 � 0.2 ND NFSausages at start (day 0) 6.45 � 0.32 4.64 � 0.16 3.42 � 0.19 5.49 � 0.1 NFSausages in mid-process (day 7) 7.5 � 0.17 1.3 � 0.24 NF 7.91 � 0.27 NFFinished sausages (day 15) 7.29 � 0.12 NF NF 7.43 � 0.19 NFMeat plant CMeat trimmings 5.74 � 0.3 3.32 � 0.25 2.26 � 0.32 ND þ(4 samples of 6)Chopped trimmings 6.42 � 0.11 3.97 � 0.24 2.78 � 0.19 ND þ(2 samples of 6)Complete batter with additives 5.97 � 0.05 3.57 � 0.21 2.0 � 0.07 ND þ(1 sample of 6)Sausages at start (day 0) 6.25 � 0.15 3.67 � 0.15 1.96 � 0.2 5.75 � 0.15 NFSausages in mid-process (day 7) 8.69 � 0.23 1.44 � 0.24 NF 8.79 � 0.15 NFFinished sausages (day 15) 8.76 � 0.1 NF NF 8.81 � 0.06 NF

TVC e Total viable count; EBC e Enterobacteriaceae count; ECC e Escherichia coli count; LAB e Lactic acid bacteria count; ND e Not determined; SD e Standard deviation; NFeNot found.

M. Ducic et al. / Food Control 43 (2014) 231e237234

B, which can be attributed to incorporation of the acidulant (GDL)in sausage batter. During the fermentation-drying stages, the pHfurther decreased so that in middle of the process it was 4.9e5.0,and remained at that level or slightly increased in finished sau-sages. The initial aw (0.95) decreased gradually during sausagefermentation-drying stages and reached values below 0.9 (0.83e0.76) in the finished sausages.

3.2. Beef fermented sausages

Results for general microbial characterisation of beef sausagesare presented in Table 3. The general patterns of changes of back-groundmicrobiota (TVC, EBC, ECC, LAB) throughout the productionof beef sausages were similar in meat plants A, D and E (Table 3),hence will not be commented on individually for each plant. Withrespect to rawmaterials for beef sausage preparation, TVC, EBC andECC in meat trimmings were relatively high (around 4.7e6 log CFU/g, 1.7e3.5 log CFU/g and 0e2 log CFU/g, respectively), furtherincreased in the next production step (trimmings chopping) andthen somewhat decreased after spices and additives were added tothe batter (Table 3). LAB levels were not investigated in the rawmaterials.

Fig. 1. pH and aw in pork sausages.

During beef sausage fermentation-drying, TVC reached levels of8e8.5 log CFU/g and remained practically the same in finishedsausages (Table 3). EBC and ECC decreased during sausagefermentation-drying stages (most dramatically in plant E), and inthe finished sausages were at level of around 2e3 log CFU/g, exceptin plant E where no ECC was found. The initial levels of naturally-occurring LAB (no starter cultures were used) immediately afterstuffing the batter into casings were around 6 log CFU/g (Table 3).Expectedly, during beef sausage fermentation-drying stages, LABsignificantly increased and was predominant microbial group(around 8e9 log CFU/g) among the background microbiota at thetime the sausages were finished.

E. coli O157 was selected to be investigated during the produc-tion process of beef sausages, as it is the most relevant bacterialfoodborne pathogen associated with raw beef. The pathogen wasfound in raw materials during sausage batter preparation in onemeat plant only, but not in the other two (Table 3). Specifically,E. coli O157 was found in chopped trimmings in plant D, but itdisappeared during the beef sausage fermentation-drying processi.e. was not found in the finished sausages.

The main physicochemical parameters in the beef sausages, thepH and aw, are shown in Fig. 2. The initial pH of the sausages inmeat plants D and E was 5.6e5.7 and decreased during thefermentation-drying process to around 5.2e5.3. In plant A, both theinitial and the final pH (5.4 and 4.8, respectively) were comparablylower. The relatively low initial pH in sausages (particularly in plantA) can be attributed to incorporation of the acidulant (GDL) insausage batter. The initial aw in sausages (around 0.95) decreasedgradually during the fermentation-drying stages and reachedvalues around 0.85 in the finished sausages.

4. Discussion

4.1. Raw material preparation stage of fermented sausagesproduction

The levels of TVC in meat trimmings destined for the pork or thebeef sausages were rather high; around 5e6 log CFU/g (Tables 2and 3). However, previously reported TVC levels in raw material

Table 3Main microbial characteristics of beef sausage during the production process.

Stage of production TVC EBC ECC LAB E. coli O157

Log10 CFU/g � SD

Meat plant AMeat trimmings 5.74 � 0.25 3.1 � 0.46 0.97 � 0.86 ND NFChopped trimmings 7.36 � 0.13 5.1 � 0.3 4.1 � 0.17 ND NFComplete batter with additives 7.23 � 0.12 4.94 � 0.26 3.89 � 0.24 ND NFSausages at start (day 0) 7.03 � 0.23 4.92 � 0.17 3.84 � 0.21 6.02 � 0.1 NFSausages in mid-process (day 10) 8.43 � 0.15 3.12 � 0.18 1.97 � 0.37 8.84 � 0.17 NFFinished sausages (day 20) 8.24 � 0.19 2.99 � 0.25 1.33 � 0.73 8.69 � 0.06 NFMeat plant DMeat trimmings 4.78 � 0.35 3.44 � 0.72 1.88 � 1.58 ND NFChopped trimmings 6.48 � 0.58 5.58 � 0.26 5.18 � 0.3 ND þ(1 sample of 6)Complete batter with additives 6.41 � 0.57 5.31 � 0.36 4.67 � 0.39 ND NFSausages at start (day 0) 6.63 � 0.75 5.5 � 0.86 4.55 � 0.61 6.09 � 0.51 NFSausages in mid-process (day 6) 8.07 � 0.44 3.19 � 0.29 2.09 � 0.64 8.23 � 0.13 NFFinished sausages (day 13) 8.29 � 0.31 3.05 � 0.29 1.96 � 0.8 8.41 � 0.22 NFMeat plant EMeat trimmings 6.08 � 0.12 1.68 � 1.52 NF ND NFChopped trimmings 7.68 � 0.15 6.38 � 0.23 1.96 � 1.03 ND NFComplete batter with additives 7.59 � 0.12 6.14 � 0.21 1.25 � 1.41 ND NFSausages at start (day 0) 7.63 � 0.03 6.18 � 0.17 1.51 � 1.21 6.45 � 0.25 NFSausages in mid-process (day 8) 8.49 � 0.04 2.73 � 0.39 NF 8.66 � 0.11 NFFinished sausages (day 15) 8.34 � 0.08 1,82 � 0.19 NF 8.66 � 0.04 NF

TVC e Total viable count; EBC e Enterobacteriaceae count; ECC e Escherichia coli count; LAB e Lactic acid bacteria count; ND e Not determined; SD e Standard deviation; NF eNot found.

M. Ducic et al. / Food Control 43 (2014) 231e237 235

destined for pork fermented sausages were roughly one log loweri.e. 4e5 log CFU/g (Comi et al., 2005; Kozacinski et al., 2008). Theenteric indicator organisms (EBC and ECC) were also present inboth the pork and the beef trimmings (Tables 2 and 3). Altogether, itcould be considered that the higher levels of background micro-biota in the pork and the beef trimmings indicate that there is aneed for improvement of their hygienic status, this being an indi-rect measure to reduce the risk of enteric pathogens in the product.

During subsequent chopping of both the pork and the beeftrimmings, the levels of all the groups of background microbiota(TVC, EBC, EEC) clearly increased (Tables 2 and 3). This is due to thefact that meat chopping exacerbatesmicrobiological contaminationbecause the exposed surface is increased and juices liberated fromthe meat act as a substrate for microbial growth (Castano, GarciaFontan, Fresno, Tornadijo, & Carballo, 2002), and also because themicrobial contamination of the chopping equipment itself can betransferred into the meat. The results obviously confirm the criticalimportance of the hygiene and related process control at this stepfor the microbial properties of the fermented sausages. In practice,this can be achieved through implementation of: a) GHP (Good

Fig. 2. pH and aw in beef sausages.

Hygienic Practice) pre-requisite program including regular cleaningand sanitation (disinfection) of meat chopping equipment; and b)HACCP (Hazard Analysis and Critical Control Points) planwithmeatchopping/sausage batter preparation step as one of the CCPs(Critical Control Points) that should include microbiological testingto verify the equipment cleaning/sanitation effectiveness and tomonitor microbiological changes in meat pre- and post-chopping.

In contrast, after additives and spices were added to the porkand the beef chopped trimmings, the background microbiota in theresultant batters tended to slightly decrease (Tables 2 and 3), butthe differences were not statistically significant because the rela-tively and commonly large variability of individual results. Theadditive-caused microbial decrease could be attributed to thenegative, stress-mediated effects of saltenitrite combination(Adams & Moss, 2008, chap. 4, chap. 9) and potentially someinhibitory compounds from spices as well (Drosinos, Skandamis, &Mataragas, 2009) on the viability and/or culturability of themicroorganisms.

With respect to bacterial pathogens, Salmonella spp. was foundin the pork trimmings, chopped pork trimmings and combinedpork sausage batter in one meat plant, as well as in chopped porktrimmings in another plant, but was not found in the third plant(Table 2). Furthermore, E. coli O157 was found in beef choppedtrimmings in one meat plant, although not in the other two plants(Table 3). These differences are reflection of at least two factors.Firstly, the occurrence of pathogens depends on their carriage byrelated meat-producing animals slaughtered in the abattoir as wellas the meat plant process hygiene; and both differ between plants.Secondly, normally, the pathogens are present in/on meat at verylow cell counts and with extremely uneven distribution; hencetheir recoveries by sampling are associated with high variability.Nevertheless, considering the well-known limited pathogen-reduction potential of the subsequent fermentation-drying stagesin raw sausages production (Getty, Phebus, Marsden, Fung, &Kastner, 2000; Lücke, 2000, chap. 19) the results of the presentstudy confirm the critical relevance of the rawmaterial preparationstage for microbial status and ultimate safety of the resultant fer-mented sausages.

M. Ducic et al. / Food Control 43 (2014) 231e237236

4.2. Fermentation-drying stage of fermented sausages production

In the pork and the beef sausages, the initial levels of TVC (6e7 log CFU/g; Tables 2 and 3) were very similar to initial TVC levels incommercial fermented sausages in some other countries (Samelis,Metaxopoulos, Vlassi, & Papa, 1998). However, roughly 1e1.5 loglower TVC levels were found by Kozacinski et al. (2008), Drosinoset al. (2005) and Comi et al. (2005). Not surprisingly, lower initialTVC was also reported for experimentally-produced fermentedsausages (Zivkovic et al., 2012). TVC increased (by 1.5e2 logs) bythe mid-process of the pork and the beef sausages and remained atthose levels in finished sausages. Similar TVC levels in finishedfermented sausages were reported also in other studies (Capita,LLorente-Marigomez, Prieto, & Alonso-Calleja, 2006; Drosinoset al., 2005; Samelis et al., 1998; Zivkovic et al., 2012), althoughhigher TVC levels were found by Garcia Fontan, Lorenzo, Parada,Immaculada, and Carballo (2007).

Although in the pork and the beef sausages (produced withoutstarter cultures in the present study) initial levels of naturally-occurring LAB were relatively low (roughly 5 log CFU/g), theyintensively developed so levels of around 8e9 log CFU/g werefound in mid-process and in finished sausages. In other studies(Comi et al., 2005; Samelis et al., 1998), LAB in fermented sausagesalso grew rapidly, so reached maximum values and became thepredominant microbial group within the first three days of thefermentation. Universally for all fermented sausages, rapid LABgrowth is associated with simultaneous exposure to increasedenvironmental temperatures, drop of partial oxygen pressure (i.e.redox potential) and drop of pH in the sausages. These conditionssuit LAB, being mesotrophs, microaerophilic, halotolerant (hencemore low aw tolerant) and acid-resistant.

At the same time, the initially predominant microbial groupsbrought in the sausage batter by chilled, raw meat (trimmings) ewhich are psychrotrophs, aerobic, comparably more acid sensitiveand not halotolerant (i.e. mainly Enterobacteriaceae) e becomesuppressed and markedly reduced in counts. As observable inFigs. 1 and 2, significant drop of both pH (helped particularly byGDL; an acidulant) and aw took place in the sausages, hencecreating favourable conditions for LAB and suppressive conditionsfor Enterobacteriaceae. In addition, it is also known that naturalspices, due to their manganese content, facilitate the kinetics ofsome essential enzymes of LAB (Adams &Moss, 2008, chap. 4, chap.9). Accordingly, both EBC and ECC decreased during the pork andthe beef sausage production process (Tables 2 and 3).

The net result of all these mechanisms, which occurred simul-taneously during the sausage fermentation, is a so-called major“microbiota shift” in fermented sausages. The “microbiota shift”can be speeded-up by the use of lactic acid bacteria-based startercultures, commercial or autochthonous (Cenci-Goga et al., 2012),which also helps to achieve desired fermentation and productquality parameters and potentially contributes to the pathogeninhibition in the product. The final LAB counts in sausages in thepresent study (produced without starter cultures) were verysimilar to those found in Greek, Italian and Spanish traditionalsausages (Capita et al., 2006; Comi et al., 2005; Drosinos et al.,2005; Fernandez-Lopez, Sendra, Sayas-Barbera, Navaro, & Perez-Alvarez, 2008) and roughly one log lower than in the Spanish“androlla” (Garcia Fontan et al., 2007).

With respect to the foodborne pathogens tested, Salmonella spp.and E. coli O157 were found in the pork and the beef raw materials,respectively, during the sausage batter preparation stage. In anearlier study, the presence of E. coli O157 in chopped beef trim-mings in Serbia was also reported (Nastasijevic et al., 2009).However, in the present study, the pathogens were not found in anysample of sausages either in the middle or at the end of the

production process, in any meat plant (Tables 2 and 3). This wasprobably due to low initial counts of the pathogens, their sup-pression by antagonistic effects of the LAB and adverse effects of theenvironmental conditions (nitrite-salt, low pH, low aw) i.e. either totheir dying off or development of “viable but not culturable” status.

Nevertheless, it is highly questionable whether the pathogenswould have been eliminated from the finished product if theirinitial counts had been higher. Namely, it is generally consideredthat the pathogen-reduction potentials of naturally fermented rawsausages production processes are most commonly around one logand up to 2e3 logs (Holck et al., 2011; Lücke, 2000). However, itcould be presumed that initial contamination of raw pork and beefwith Salmonella spp. and E. coli O157, respectively, can be some-times at levels higher than those (Sauer, Majkowski, Green, & Eckel,1997). Therefore, to ascertain total elimination of the pathogensfrom these products under all scenarios, achieving their several-logreduction during the production process would be required(Anonymous, 2001). Because several-log reduction efficacy cannotbe consistently and reliably achieved through the “natural” rawfermented sausage production process alone, further studies intodevelopment of additional pathogen-elimination strategies arenecessary and planned.

5. Conclusion

In the present study, typical Serbian types of pork and beef rawfermented sausages were characterised with respect to the mainbackground microbiota patterns and presence of the main food-borne pathogens throughout the commercial production process.The backgroundmicrobiota patterns and levels in the present studywere, generally, as those commonly reported for the fermentedsausages production in other countries. Salmonella spp. and E. coliO157 were found in the raw materials during the pork and the beefsausage, respectively, batter preparation stage, but not during thefermentation-drying stages and nor in the finished products.Nevertheless, the initial presence of these pathogens may beconsidered as a meat safety risk, as their total elimination in thecase of high initial counts cannot be assumed. Hence, furtherstudies will be focused on novel pathogen elimination strategies forraw, fermented sausages.

Acknowledgements

The study is a part of Serbian Ministry of Education, Science andTechnological Development project (TR 31034).

References

Adams, M., & Mitchell, R. (2002). Fermentation and pathogen control: a riskassessment approach. International Journal of Food Microbiology, 79, 75e83.

Adams, M., & Moss, M. (2008). Food microbiology (3rd ed.). Cambridge: The RoyalSociety of Chemistry.

Anonymous. (2001). Performance standards for the production of processed andpoultry products; proposed rule U. S. Department of Agriculture. Food Safety andInspection Service.

Capita, R., LLorente-Marigomez, S., Prieto, M., & Alonso-Calleja, C. (2006). Micro-biological profiles, pH and titratable acidity of Chorizo and Salchichon (twoSpanish dry fermented sausages) manufactured with ostrich, deer, or porkmeat. Journal of Food Protection, 69, 1183e1189.

Castano, A., Garcia Fontan, M., Fresno, J. M., Tornadijo, M. E., & Carballo, J. (2002).Survival of Enterobacteriaceae during processing of Chorizo de cebolla, aSpanish fermented sausage. Food Control, 13, 107e115.

Cenci-Goga, B. T., Rossitto, P. V., Sechi, P., Parmegiani, S., Cambiotti, V., & Cullor, J. S.(2012). Effect of selected dairy starter cultures on microbiological, chemical andsensory characteristics of swine and venison (Dama dama) nitrite-free dry-cured sausages. Meat Science, 90, 599e606.

Comi, G., Urso, R., Iacumin, L., Rantisou, K., Cattaneo, P., Cantoni, C., et al. (2005).Characterisation of naturally fermented sausages produced in the north east ofItaly. Meat Science, 69, 381e392.

M. Ducic et al. / Food Control 43 (2014) 231e237 237

Drosinos, E. H., Mataragas, M., Xiraphi, N., Moschonas, G., Gaitis, F., &Metaxopulos, J. (2005). Characterisation of the microbial flora from a traditionalGreek fermented sausage. Meat Science, 69, 307e317.

Drosinos, E. H., Skandamis, N. P., & Mataragas, M. (2009). Antimicrobials treatment.In F. Toldra (Ed.), Safety of meat and processed meat (pp. 255e296). New York:Springer.

EFSA. (2006). Risk assessment and mitigation options of Salmonella in pig pro-duction (2006). The EFSA Journal, 341, 1e131.

Ethelberg, S., Smith, B., Torpdahl, M., Boel, J., Jensen, T., Nielsen, E. M., et al. (2009).Outbreak of non-O157 Shiga toxin-producing Escherichia coli infection fromconsumption of beef sausage. Clinical Infection Diseases, 48, 78e81.

Fernandez-Lopez, J., Sendra, E., Sayas-Barbera, E., Navaro, C., & Perez-Alvarez, J. A.(2008). Physico-chemical and microbiological profiles of “salchichon” (Span-ish dry-fermented sausage) enriched with orange fiber. Meat Science, 80,410e417.

Garcia Fontan, M., Lorenzo, J., Parada, A., Immaculada, F., & Carballo, J. (2007).Microbiological characteristics of “androlla”, a Spanish traditional pork sausage.Food Microbiology, 24, 52e58.

Getty, K. J. K., Phebus, R. K., Marsden, J. J. L., Fung, D. Y. C., & Kastner, C. L. (2000).Escherichia coli O157:H7 and fermented sausages: a review. Journal of RapidMethods and Automation in Microbiology, 8, 141e170.

Gieraltowsky, L., Julian, E., Pringle, J., Macdonald, K., Quilliam, D., Marsden-Hang, N.,et al. (2013). Nationwide outbreak of Salmonella Montevideo associated withcontaminated imported black and red pepper: warehouse membership cardsprovide critical clues to identify the source. Epidemiology and Infection, 141,1244e1252.

Holck, L. A., Axelsson, L., Mari Rode, T., Hoy, M., Mage, I., Alvseike, O., et al. (2011).Reduction of verotoxigenic Escherichia coli in production of fermented sausages:a review. Meat Science, 89, 286e295.

ISO. (1999). Microbiology of food and animal feeding stuffs e Preparation of testsamples, initial suspension and decimal dilutions for the microbiological exami-nation e Part 1: General rules for the preparation of the initial suspension anddecimal dilutions (EN ISO 6887e6891:1999). International Organization forStandardization.

ISO. (2002). Microbiology of food and animal feeding stuffs e Horizontal method fordetection of Salmonella spp. (EN ISO 6579:2002). International Organization forStandardization.

Kozacinski, L., Drosinos, E., Caklovica, F., Cocolin, L., Gasparik-Reichardt, J., &Veskovic, S. (2008). Investigation of microbial association of traditionally fer-mented sausages. Food Technology and Biotechnology, 46, 93e106.

Kuhn, K., Torpdahl, M., Frank, C., Sigsgaard, K., & Ethelberg, S. (2011). An outbreak ofSalmonella Typhimurium traced back to salami, Denmark, April to June 2010.Eurosurveillance, 16.

Lücke, F. K. (2000). Fermented meats. In B. M. Lund, T. C. Baird-Parker, & G. W. Gould(Eds.), The microbiological safety and quality of food. Gaithersburg: Aspen Pub-lishers Inc.

Luzzi, I., Galetta, P., Massari, M., Rizzo, C., Dionisi, A. M., Filetici, E., et al. (2007). AnEaster outbreak of Salmonella Typhimurium DT 104A associated with tradi-tional pork salami in Italy. Eurosurveillance, 12, 149e152.

Nightingale, K. K., Thippareddi, H., Phebus, R. K., Marsden, J. L., & Nutsch, A. L.(2006). Validation of a traditional Italian-style salami manufacturing process forcontrol of Salmonella and Listeria monocytogenes. Journal of Food Protection, 69,794e800.

Nastasijevic, I., Mitrovic, R., & Buncic, S. (2009). The occurrence of Escherichia coliO157 in/on feces, carcasses and fresh meats from cattle. Meat Science, 82, 101e105.

Nørrung, B., & Buncic, S. (2008). Microbial safety of meat in the European Union.Meat Science, 78, 14e24.

Nørrung, B., Andersen, J. K., & Buncic, S. (2009). Main concerns of pathogenic mi-croorganisms in meat. In F. Toldra (Ed.), Safety of meat and processed meat (pp.3e29). New York: Springer.

Samelis, J., Metaxopoulos, J., Vlassi, M., & Pappa, A. (1998). Stability and safety oftraditional Greek salami-a microbiological ecology study. International Journalof Food Microbiology, 44, 69e82.

Sartz, L., De Jong, B., Hjertqvist, M., Plym-Forshell, L., Alsterlund, R., Lofdahl, S., et al.(2008). An outbreak of Escherichia coli O157:H7 infection in southern Swedenassociated with consumption of fermented sausage; aspects of sausage pro-duction that increase the risk of contamination. Epidemiology and Infection, 136,370e380.

Sauer, J. C., Majkowski, J., Green, S., & Eckel, R. (1997). Foodborne illness outbreakassociated with a semi-dry fermented sausage product: a review. Journal ofFood Protection, 60, 1612e1617.

Sekse, C., O’Sullivan, K., Granum, P. E., Rorvik, L. M., Wasteson, Y., & Jorgensen, H. J.(2009). An outbreak of Escherichia coli O103:H25-bacteriological investigationsand genotyping of isolates from food. International Journal of Food Microbiology,133, 259e264.

Zivkovic, D., Radulovic, Z., Aleksic, S., Perunovic, M., Stanisic, N., & Radovic, C.(2012). Chemical, sensory and microbiological characteristics of Sremskasausage (traditional dry-fermented Serbian sausage) as affected by pig breed.African Journal of Biotechnology, 11, 3858e3867.