Salmonella spp. and Hygiene Indicator Microorganisms in Chicken Carcasses Obtained at Different Processing Stages in Two Slaughterhouses

Abstract

Chicken meat is considered an important vehicle of foodborne pathogens such as Salmonella spp., demanding an effective control of its contamination during industrial processing. This study aimed to investigate the presence of Salmonella spp. and microbiological indicators at different stages of processing in two slaughterhouses (Sh1, high-capacity; Sh2, low-capacity). Surface samples of chicken carcasses were collected in the following sequential stages: (A) immediately before evisceration, (B) after evisceration, (C) after showering, and (D) after chiller. All samples were submitted for detection of Salmonella spp. and enumeration of mesophilic aerobes, total coliforms, thermotolerant coliforms, and Escherichia coli. The obtained means and frequencies were compared by analysis of variance and chi-square tests (p < 0.05), considering different slaughterhouses and stages of processing. No significant differences were observed between the frequencies of Salmonella spp. obtained at different steps in Sh1 and Sh2 (p > 0.05). Sh2 showed higher levels of microbiological contamination when compared with Sh1 for mesophilic aerobes (in stages B and D), total coliforms and thermotolerant coliforms (stage D), and E. coli (all stages) (p < 0.05). The variation in the levels of contamination by microbiological indicators over the processing indicated the significance of different control procedures adopted by slaughterhouses for the microbiological quality of chicken carcasses.

Introduction

T he production of poultry meat in Brazil has increased significantly because of the demand of the internal market and the export increase observed since the 1990s (MAPA, 2009). In the first 6 months of 2009, the export of chicken meat achieved 1.8 million tons (about U.S. $2.7 billion), besides the 3 million tons for the domestic market (ABEF, 2009).

Because of the great demands from importing countries for products with better quality, including hygienic concerns as to the achievement of poultry carcasses, new methodologies for systematic control have been increasingly used by slaughtering houses. An effective way to perform such control is hygiene indicator monitoring, as proposed by the Hazard Analysis and Critical Control Point (HACCP) (ICMSF, 1988). Although they do not pose any harm to consumers, hygiene indicator microorganisms are frequently used as indicators of pathogens, such as Salmonella spp. and Campylobacter spp. (Scott et al., 2001). Mesophilic aerobes (MA) are frequently used as hygiene indicators in the meat production process, the same way that the presence of Escherichia coli (EC) is considered an indication of fecal contamination (Ghafir et al., 2008). Other indicators, such as enterobacteria, may suggest the presence of specific pathogens in the production environment or in the final product (Martins and Germano, 2007).

The control of pathogen contamination is the main strategy to prevent foodborne diseases. Salmonella spp. is considered a major cause of these diseases and is responsible for 10–15% of the cases of acute gastroenteritis in humans, frequently associated with poultry meat and eggs (Jay et al., 2005). The presence of Salmonella spp. in carcasses depends on the conditions to which poultry were exposed during production until the end of the slaughtering process, including the sanitary conditions in rural properties, transport, and processing in slaughterhouses (Fluckey et al., 2003). In Brazil, there are a great variety of poultry slaughterhouses with different capacities, volumes, procedures, equipment, and tools. These variables can directly affect the contamination by Salmonella spp. and hygiene microbiological indicators (Dickel et al., 2005b).

The objective of the present work was to compare two slaughterhouses with different capacities and processing characteristics as to the presence of Salmonella spp. and hygiene microbiological indicators in important slaughter line stages.

Materials and Methods

Sampling and dilutions

Two slaughterhouses with distinct slaughter capacities were selected in the region of Viçosa, Minas Gerais state, Brazil, being identified as Sh1 (high-capacity of slaughtering) and Sh2 (low-capacity). The main characteristics of each slaughterhouse are described on Table 1. In both slaughterhouses, four sequential stages were defined according Rodrigues et al. (2008), considering relevant sources of microbial contamination: (A) immediately before evisceration, (B) after evisceration, (C) after shower, and (D) after cooling tank (chiller). In each stage, superficial samples were collected from chicken carcasses (Sh1: 27 samples per point, and Sh2: 30). Each sample unit was achieved by swabbing a surface area of 50 cm² with sterile polyurethane sponges in two different locations on the carcass: the chest, near the neck, and the dorsal region, adjacent to the cloacae. The sponges were previously soaked in 10 mL of 0.1% peptone water (H₂Op) (ICMSF, 1986).

Table 1.

Main Characteristics of the Two Slaughterhouses (Sh1, High-Capacity Slaughtering; Sh2, Low-Capacity) Where Microbiological Indicator Microorganisms and Salmonella spp. Were Studied in Different Steps of Processing

Characteristic	Sh1	Sh2
Slaughtering capacity	165,000 poultry/day	4000 poultry/day
Slaughter processing
Bleeding	Automatic	Manual
Defeathering	Automatic	Automatic
Evisceration	Automatic	Manual
Chiller tanks	2	1
Carcass packaging	Automatic	Manual
Final products	Chicken carcasses, poultry cuts, and poultry products	Chicken carcasses
Market	Export (South America, Asia, and Middle East) and internal (all Brazilian states)	Internal (nearby cities)

After swabbing, the sponges were placed in sterile bags and homogenized in a “Stomacher” (400 Seward circulator, West Sussex, United Kingdom) for 1 min with 40 mL of 0.1% H₂Op, to achieve a homogenate with a volumetric equivalent of the sampled area (1 mL = 1 cm²). The obtained homogenates were then submitted to 10-fold dilution using 0.1% H₂Op (ICMSF, 1986).

Salmonella spp.

The methodology described by MAPA (2003a) was followed. An aliquot of 25 mL of the homogenate was added to 225 mL of buffered H₂Op 1%, followed by incubation at 37°C for 20 h (preenrichment). Then, 1 mL was inoculated into selenite cystine broth (incubation at 37°C for 24 h) and 0.1 mL in Rappaport Vassiliadis broth (42.5°C for 24 h in water bath) (selective enrichment). The cultures achieved were streaked onto brilliant green phenol red lactose sucrose agar and xylose lysine deoxycholate agar and incubated at 37°C for 24 h. Once the typical colonies were identified (brilliant green phenol red lactose sucrose agar: colorless or pinkish colonies, and xylose lysine deoxycholate agar: opaque colonies, with or without H₂S production), they were submitted to preliminary biochemical identification in triple sugar iron agar and lysine iron agar and incubated at 37°C for 24 h. The final verification was carried out by serologic agglutination tests with somatic (O) and flagellar (H) polyvalent antisera (Probac do Brazil, São Paulo, SP, Brazil). The samples with characteristic biochemical reactions were considered positive, as well as those with positive reactions in both serological studies.

All the cultures that presented positive results in the serological tests with polyvalent antigens were also submitted to confirmation of belonging to the genus Salmonella by polymerase chain reaction (Oliveira et al., 2002) and serotyping at Fundação Oswaldo Cruz (Rio de Janeiro, Brazil). All the analyses were carried out simultaneously with Salmonella Enteritidis ATCC 13076 as positive control, and all the culture media used were from Oxoid (Basingstoke, United Kingdom).

Microbiological hygiene indicators

The collected samples were also submitted for the enumeration of hygiene indicator microorganisms (MAPA, 2003a). MA were enumerated by direct plating in standard count agar (pour plate) and incubation at 35°C for 48 h. The colonies formed were counted and the results were expressed as colony-forming units/cm².

Total coliforms (TC) were quantified by the multiple tube technique, using the presumptive test in lauryl sulfate tryptose broth (35°C for 24–48 h) and confirmed in 2% brilliant green bile broth (35°C for 24–48 h). The obtained results were expressed as most probable number (MPN)/cm². Based on the TC-positive results, the thermotolerant coliforms (TTC) were detected using EC broth (44.5°C for 24 h in water bath), with final results expressed as MPN/cm².

Based on the positive results for TTC, EC was detected according Kornacki and Johnson (2001). Aliquots of the EC tubes with positive results were streaked on plates of eosin methylene blue agar and incubated at 35°C for 24 h. When typical colonies of EC were present, with a metallic aspect, they were streaked in plate count agar with incubation at 35°C for 24 h. Next, they were streaked on prehydrated Petrifilm™ EC plates (3M Microbiology, St. Paul, MN) and incubated at 35°C for 24 h, for the evaluation of the glucuronidase activity (blue colonies) and EC confirmation. Cultures still suspected were biochemically identified using Bactray 1 and 2 (Laborclin Produtos para Laboratório, Pinhais, PR, Brazil). All the analyses were carried out simultaneously with the positive control, which consisted of a culture of EC isolated from mastitic milk. The final results were expressed as MPN/cm².

Temperature and residual chlorine monitoring

During nine sample collections in Sh2, the temperatures of scalding, prechiller, chiller, and the swabbed carcasses were measured using a digital thermometer. The amount of free residual chloride in the prechiller and chiller was also checked with the use of a digital electronic equipment (LaMotte 1200 Colorimeter; LaMotte, Chertertown, MD), with reading magnification from 0.00 to 5.00 mg/L. In Sh1, the temperatures from distinct stages and residual chlorine from chiller tanks were measured by inspectors from the Ministry of Agriculture (Ministério da Agricultura, Pecuária e Abastecimento, Brazil), as routine control.

Analysis of the results

The frequencies of positive samples for Salmonella spp. at each step and at each slaughterhouse were compared by the chi-square test (p < 0.05). The mean counts of MA, TC, TTC, and EC were compared by analysis of variance in each stage of the processing and in each slaughterhouse (p < 0.05). All the analyses were carried out using the Statistica 7.0 (StatSoft, Tulsa, OK) and XLStat 2009.1.02 (Addinsoft, New York, NY). The temperatures at different stages and of carcasses and the concentrations of chlorine were compared with the values established by the MAPA (1998).

Results and Discussion

A total of 9 samples out of the 108 analyzed from Sh1 were contaminated with Salmonella spp. (8.3%), whereas 6 out of the 119 samples from Sh2 were contaminated by this pathogen (5.0%) (Table 2). Although the slaughterhouses presented different frequencies of Salmonella spp. in the studied stages, no significant differences were found between these frequencies, considering both the stages of processing and the slaughterhouses (Table 2). All Salmonella spp. isolates obtained from Sh2 were confirmed by polymerase chain reaction (Fig. 1) and serotyped as Salmonella Enteritidis.

FIG. 1.

Gel electrophoresis with the products of amplification obtained by the polymerase chain reaction from six cultures of Salmonella spp. (1–6) isolated from superficial samples of chicken carcasses obtained from different processing steps in a small-sized slaughterhouse. M, molecular marker of 100 bp (Promega, Madison, WI); Salmonella Enteritidis ATCC 13076 was used as positive control (+) with a segment of 284 pb; Milli-Q water was used as negative control (−).

Table 2.

Frequencies of the Salmonella spp. at Different Processing Stages in Large- (Sh1) and Small-Sized (Sh2) Slaughterhouses

Stage ^a	Sh1	Sh2	Test ^b
A	1/27	0/29	χ ²: 1.09; df: 1; p > 0.05
B	3/27	2/30	χ ²: 0.35; df: 1; p > 0.05
C	4/27	2/30	χ ²: 1.00; df: 1; p > 0.05
D	1/27	2/30	χ ²: 0.25; df: 1; p > 0.05
Test	χ ²: 3.27; df: 3; p > 0.05	χ ²: 2.03; df: 3; p > 0.05

A, immediately before evisceration; B, after evisceration; C, after shower; D, after cooling tank.

χ ², chi-square; df, degree of freedom; p, level of significance.

Lopes et al. (2007) found a low frequency of Salmonella spp. in a poultry slaughterhouse, as the pathogen was detected in only 1.6% of the analyzed chicken carcasses (i.e., one positive sample before entrance into the chiller tank and another after the chiller tank exit). Different results were found by Dickel et al. (2005a), who compared the presence of Salmonella spp. in slaughterhouses with different slaughtering capacities: there was no isolation of Salmonella spp. in the small-sized slaughterhouse, whereas in the larger slaughterhouse, 50% of the samples were positive, with oscillation during the processing and a predominance of contamination after evisceration. Evisceration is a significant factor in contamination by Salmonella spp. because it was carried out in a mechanized way, favoring the rupture of viscera of animals with nonstandardized sizes (Dickel et al., 2005b). Despite the identification of main steps for Salmonella spp. contamination (Table 1), the obtained results in both slaughterhouses met the demands of the Program for the Reduction of Pathogens, with values below 23.5% (MAPA, 2003b).

The mean values and the analysis of the results of the microbial contamination in each stage of the processing and each slaughterhouse are presented in Table 3. A significant reduction was observed for MA, TC, and TTC in Sh1 only between C and D stages (p < 0.05), whereas the mean counts of EC did not present significant differences in the processing stages (p > 0.05). In Sh2, MA presented a significant reduction between B and C stages, whereas TC, TTC, and EC were reduced between C and D stages (p < 0.05). These results indicate the importance of the shower (C) in Sh2 and the chilling tank (D) in both slaughterhouses to promote the reduction of contamination by hygiene indicator microorganisms. Such results are similar to those observed by Gill et al. (2006), who suggested that MA are associated not only with the skin of the poultry carcass, but also with superficial and adhered organic matter, which is easily eliminated by showers, as observed in Sh2 (Table 3). Defects in the adjustment of the water pressure of these showers may affect their efficiency in the reduction of the superficial microbial contamination (Bolder, 1997), as observed in MA in Sh1 (Table 3). Further, a significant reduction in the contamination by coliforms between these stages (TC, TTC, and EC) in both slaughterhouses was not observed, possibly because of the higher adhesion capacity of these microorganisms (Gill et al., 2006). Cason et al. (2004) affirmed that several microorganisms strongly adhere to the carcasses because they are sheltered in the skin folds and are thus resistant to removal during processing.

Table 3.

Mean Counts (± Standard Deviation) of Mesophilic Aerobes, Total Coliforms, Thermotolerant Coliforms, and Escherichia coli at Four Steps of Chicken Processing in Large- (Sh1) and Small-Sized (Sh2) Slaughterhouses

		Slaughterhouse
Indicator microorganism	Stage	Sh1	Sh2
Mesophilic aerobes	A	4.07 ± 0.50^a,A	4.41 ± 0.47^a,A
	B	3.75 ± 0.54^b,A	4.53 ± 0.60^a,A
	C	3.65 ± 0.38^a,A	3.90 ± 0.45^a,B
	D	2.16 ± 0.75^b,B	4.28 ± 0.72^a,A,B
Total coliforms	A	3.10 ± 0.67^a,A	3.40 ± 0.51^a,A
	B	3.28 ± 0.50^a,A	3.36 ± 0.44^a,A
	C	3.12 ± 0.60^a,A	3.08 ± 0.57^a,A
	D	1.52 ± 0.46^b,B	2.49 ± 0.51^a,B
Thermotolerant coliforms	A	2.94 ± 0.65^a,A	3.40 ± 0.50^a,A
	B	3.12 ± 0.68^a,A	3.19 ± 0.61^a,A
	C	2.92 ± 0.60^a,A	3.03 ± 0.59^a,A
	D	1.38 ± 0.44^b,B	2.42 ± 0.49^a,B
Escherichia coli	A	1.71 ± 0.76^b,A	3.35 ± 0.50^a,A
	B	2.10 ± 1.11^b,A	3.19 ± 0.62^a,A
	C	1.57 ± 0.45^b,A	3.01 ± 0.58^a,A
	D	0.85 ± 0.47^b,A	2.41 ± 0.49^a,B

In the column Stage, A indicates immediately before evisceration; B, after evisceration; C, after shower; D, after cooling tank. In the column Slaughterhouse, distinct superscript capital letters indicate significant differences between the mean values presented in the same column, for each microbiological group and each slaughterhouse. Distinct superscript lowercase letters indicate significant differences between the mean values in the same line, in both slaughterhouses. Analysis of variance, p < 0.05.

The passage through the chiller was important to promote a significant reduction in TC and TTC in carcasses processed in both slaughterhouses, besides MA in Sh1 and EC in Sh2. This reducing effect promoted by this stage was described by Allen et al. (2000), who performed a study on the efficiency of immersion cooling tanks and observed a significant reduction in the contamination by coliforms in chicken carcasses. In a similar study, Huezo et al. (2007) observed that after passing through the chiller, chicken carcasses had a 90% reduction of coliform populations, especially EC.

By comparing the results achieved in both slaughterhouses in the different processing stages, higher counts were found in the samples collected in Sh2, for all the indicator microorganisms studied in stage D, after passing through the chiller (p < 0.05) (Table 3). Comparing the results obtained from temperature measurement (Fig. 2), it can be observed that Sh2 presented high variation, all of them out of the limits established by MAPA (1998), whereas Sh1 presented these values in conformity with MAPA parameters (data not shown). Considering the residual chlorine (Fig. 3), despite the high variation in both chiller tanks, all measurements were between 1 and 5 mg/L, as established by MAPA (1998) and observed in Sh1 (data not shown). Variations in Sh2 may explain the lower reduction of the indicator microorganisms in the last stage of processing, because the water from the cooling tanks may contribute to the microbial contamination to chicken carcasses (Gill et al., 2006).

FIG. 2.

Variation of the measured temperatures from different processing steps (A, scalding; B, precooling; C, cooling) and inner part of chicken carcasses (D) during 9 sampling days in a small-sized slaughterhouse. In each figure, black points indicate the measured temperatures and the continuous lines indicate minimum (in A) and maximum reference values (in B–D) (MAPA, 1998).

FIG. 3.

Variation of the free residual chloride concentration in precooling (▴) and cooling (▲) tanks, obtained from a small-sized slaughterhouse.

It was observed that MA means were significantly higher in the B stage (evisceration) and EC in all the stages (p < 0.05) (Table 3). A plausible reason for the higher microbial contamination in Sh2 after evisceration is the slaughtering structure in that facility. At this stage, the operation is carried out manually in Sh2, which may lead to a higher risk of contamination by operator handling or from improperly cleaned tools. Russell and Walker (1997) stated that the increase in the contamination by MA after the evisceration of the chicken carcasses is inevitable. Göksoy et al. (2004), however, did not find any effect of evisceration on the contamination by MA in the same product.

The relevance of the research on EC was different for each slaughterhouse. In Sh1, EC was characterized in only 61% of the TTC (Table 3). According to Voidarou et al. (2007), EC cannot be used as a classic indicator in chiller tanks, because EC was found in only 30% of the TTC. However, EC was present in 99.5% of the TTC in Sh2, demonstrating that, in this case, this group can be considered as a good hygiene indicator. These results suggest that even when the original microbiota is the same, there are, over the slaughter process, changes in its behavior because of the different technologies employed by each slaughterhouse.

There are few studies on the differences of contamination between large- and small-sized slaughterhouses. Although Sh2 is a small-sized facility, where a validated control system does not yet exist, in general, comparing both facilities, contamination over the slaughter stages presented punctual differences. Sh2 is a new facility, with low slaughter speed, which may contribute to the control of the contamination in the process. Sh1, on the other hand, has a consolidated HACCP program, but the great volume of slaughter may affect the adequate performance of the quality control procedures. Similar results were found by Tsola et al. (2008), who investigated the impact of the modernization of the main critical control points on an automated slaughterhouse, using microbiological indicators. It was observed that, after modernization, there was a significant reduction in the microbial contamination over the slaughter line, although there were no changes in the main critical control point stressed by the HACCP.

Conclusions

The results of this study showed that the use of indicators in the identification and monitoring of the main stages of microbial contamination in poultry may be valuable, also allowing an evaluation of the efficiency of the quality control established in each slaughterhouse. As to the pathogen studied, it was concluded that the hygiene management in the sampled slaughterhouses did not affect the frequencies of Salmonella spp. Variations in the levels of contamination in different slaughter systems strengthen the HACCP concepts as specific plans of implantation of the system for products and different processes, including poultry slaughters.

Footnotes

Acknowledgments

The authors thank Dr. Ernesto Hofer and Elaine Moura Falavina dos Reis from the Laboratory of Enterobacteria of the Fundação Oswaldo Cruz, for performing the serotyping of the isolated cultures of Salmonella spp. Luís Augusto Nero received financial support from FAPEMIG (Fundação de Amparo a Pesquisa do Estado de Minas Gerais—Programa Pesquisador Mineiro; PPM-00093-09).

Disclosure Statement

No competing financial interests exist.

References

[ABEF] Associação Brasileira dos Produtores e Exportadores de Frango. Estatísticas. www.abef.com.br. 2009 August 19(Online.) (In Portuguese.)

Allen

, Corry

JEL

, Burton

et al. Hygiene aspects of modern poultry chilling. Int J Food Microbiol, 2000; 58:39–48.

Bolder

. Decontamination of meat and poultry carcasses. Trends Food Sci Tech, 1997; 8:221–227.

Cason

, Berrang

, Buhr

et al. Effect of prechill fecal contamination on numbers of bacteria recovered from broiler chicken carcasses before and after immersion chilling. J Food Prot, 2004; 67:1829–1833.

Dickel

, Rodrigues

, Santos

et al. Microbiologia convencional, ELISA e PCR para detecção de Salmonella em abatedouro de frango totalmente automatizado, semi-automatizado de grande porte e semi-automatizado de pequeno porte. Hig Alim, 2005a. 19:79–85(In Portuguese.)

Dickel

, Santos

, Rodrigues

et al.

Ocorrência de Salmonella em abatedouros de aves com tecnologia totalmente automatizada (grande porte), semi-automatizada (médio porte) e semi-automatizada (pequeno porte)

Hig Alim, 2005b. 19:62–67(In Portuguese.)

Fluckey

, Sanchez

, Mckee

et al. Establishment of a microbiological profile for an air-chilling poultry operation in the United States. J Food Prot, 2003; 66:272–279.

Ghafir

, Chin

, Dierick

et al. Hygiene indicator microorganisms for selected pathogens on beef, pork, and poultry meats in Belgium. J Food Prot, 2008; 71:35–45.

Gill

, Moza

, Badoni

et al. The effects on the microbiological condition of product of carcass dressing, cooling, and portioning processes at a poultry packing plant. Int J Food Microbiol, 2006; 110:187–193.

10.

Göksoy

, Kirkan

, Kök

. Microbiological quality of broiler carcasses during processing in two slaughterhouses in Turkey. Poultry Sci, 2004; 83:1427–1432.

11.

Huezo

, Northcut

, Smith

et al. Effect of dry air or immersion chilling on recovery of bacteria from broiler carcasses. J Food Prot, 2007; 70:1829–1834.

12.

[ICMSF] International Commission on Microbiological Specifications for Foods. Microorganisms in Foods 2: Sampling for Microbiological Analysis: Principles and Specific Applications, 2 ^nd. Toronto: University of Toronto Press, 1986.

13.

[ICMSF] International Commission on Microbiological Specifications for Foods. Microorganisms in Foods 4: Application of the Hazard Analysis Critical Control Point (HACCP) System to Ensure Microbiological Safety and Quality. Oxford: Blackwell Scientific Publications, 1988.

14.

Jay

, Loessner

, Golden

. Modern Food Microbiology, 7 ^th. New York: Springer, 2005.

15.

Kornacki

, Johnson

. Enterobacteriaceae, Coliforms, and Escherichia coli as quality and safety indicators. Compendium of Methods for the Microbiological Examination of Foods, 4 ^th. Downes

, Ito

. Washington, DC: American Public Health Association, 2001; 69–82.

16.

Lopes

, Galhardo

, Oliveira

et al. Pesquisa de Salmonella spp. e microrganismos indicadores em carcaças de frango e água de tanques de pré-resfriamento em abatedouro de aves. Semina, 2007; 28:465–476(In Portuguese.)

17.

[MAPA] Ministério da Agricultura, Pecuária e Abastecimento, Brasil. Portaria n° 210, de 10/11/1998. Regulamento Técnico de inspeção tecnológica e higiênico sanitária de carne de aves. Diário Oficial da União26/11/1998Seção 1 1998; 226(In Portuguese.)

18.

[MAPA] Ministério da Agricultura, Pecuária e Abastecimento, Brasil. Instrução Normativa n° 62 de 26/08/2003. Métodos analíticos oficiais para análises microbiológicas para controle de produtos de origem animal e água. Diário Oficial da União18/09/2003Seção 1 2003a. 14(In Portuguese.)

19.

[MAPA] Ministério da Agricultura, Pecuária e Abastecimento, Brasil. Instrução Normativa n° 70 de 06/10/2003. Programa de redução de patógenos: monitoramento microbiológico e controle de Salmonella spp. em carcaças de frangos e perus. Diário Oficial da União10/10/2003Seção 1 2003b. 9(In Portuguese.)

20.

[MAPA] Ministério da Agricultura, Pecuária e Abastecimento, Brasil. Estatísticas. www.agricultura.gov.br. 2009 June 25(Online.) (In Portuguese.)

21.

Martins

, Germano

PML

. Microbiological indicators for the assessment of performance in the hazard analysis and critical control points (HACCP) system in meat lasagna production. Food Control, 2007; 19:764–771(In Portuguese.)

22.

Oliveira

, Santos

, Schuch

DMT

et al. Detection and identification of salmonellas from poultry-related samples by PCR. Vet Microbiol, 2002; 87:25–35.

23.

Rodrigues

ACA

, Pinto

PSA

, Vanetti

MCD

et al. Análise e monitoramento de pontos críticos no abate de frangos utilizando indicadores microbiológicos. Ciência Rural, 2008; 38:1948–1953(In Portuguese.)

24.

Russell

, Walker

. The effect of evisceration on visible contamination and the microbiological profile of fresh broiler chicken carcasses using the Nu-Tech evisceration system and the conventional streamlined inspection system. Poultry Sci, 1997; 76:780–784.

25.

Scott

, Anderson

, Wang

. Mesophilic anaerobic sporeformers. Compendium of Methods for the Microbiological Examination of Foods, 4 ^th edition Downes

, Ito

. Washington, DC: American Public Health Association, 2001; 229–238.

26.

Tsola

, Drosino

, Zoiopoulos

. Impact of poultry slaughter house modernisation and updating of food safety management systems on the microbiological quality and safety of products. Food Control, 2008; 19:423–431.

27.

Voidarou

, Vassos

, Kegos

et al. Aerobic and anaerobic microbiology of the immersion chilling procedure during poultry processing. Poultry Sci, 2007; 86:1218–1222.