Technical papers
Microbiological assessment of Pangasianodon hypophthalmus at fish-processing plants in Vietnam
2022 Volume 28 Issue 2 Pages 169-177
Details
2022 Volume 28 Issue 2 Pages 169-177
In this study, we evaluated the microbial safety of Pangasius fish at two companies along the Vietnamese Mekong Delta. A microbial assessment scheme was used to diagnose the actual microbiological performance of implemented food safety management systems. The results showed that the microbial safety level profiles were the same at levels 1–2, indicating a poor-to-moderate food safety output of these companies. Microbial quality parameters including total mesophilic counts, Escherichia coli and coliforms, Staphylococcus aureus, and pathogens Listeria monocytogenes and Vibrio cholerae were not in accordance with the microbiological reference standards. Even though the quality of the raw Pangasius fish originating from the two companies was dissimilar as influenced by microbial ecology, the similarity in the microbial safety profiles of the two companies revealed the necessity of validating the efficiency of food safety management systems to improve the safety of the fish product.
In recent years, more attention has been paid to Pangasianodon hypophthalmus fish, a freshwater fish found along the Mekong River, because of its low cost with mild flavor and firm texture (Phan et al., 2009). According to the Vietnam Association of Seafood Exporters and Producers i), the farming area of Pangasius fish in the Mekong Delta (Vietnam) was estimated to reach 6 600 ha; further, the output was expected to reach 1.42 million tons in 2019. The farming sector is concentrated along two branches of the Mekong River in Vietnam, where the provinces of Can Tho, An Giang, and Dong Thap are the leading regions producing Pangasius fish, accounting for over 75% of total national Pangasius fish production. Currently, Pangasius fish farming is implemented in compliance with international food safety and international quality management standards such as Global Good Agricultural Practice (GAP), Aquaculture Stewardship Council (ASC), and Best Aquaculture Practices (BAP) to control the quality of the raw fishi). Moreover, nearly 100 companies in the Mekong Delta processing Pangasius fish have implemented a variety of certifications for Food Safety Management Systems (FSMS), e.g. Hazard Analysis Critical Control Point (HACCP), International Organization for Standardization (ISO 9001), British Retail Consortium (BRC), International Featured Standards (IFS), and Safe Quality Food (SQF) to guarantee the food quality and safety of their productsi). Monitoring the performance of implemented FSMS in line with regulatory documentation is essential in reducing the risk of contamination and ensuring product safety. However, fish companies have faced different issues in terms of notifications owing to failures to meet the demands of export markets (such as microbiological criteria, chemical standards, and residues) (Pigłowski, 2018). The Rapid Alert for Food and Feed databaseii) issued several notifications of frozen Pangasius fish originating from Vietnam contaminated with pathogenic bacteria from 2005 to 2018, i.e., E. coli, Salmonella spp., V. cholerae, and L. monocytogenes. In addition, there were also notifications by the Ministry of Health Labour and Welfareiii) based on microbial contamination of total viable counts (> 7 log CFU/g) and E. coli in Pangasius fish products exported from Vietnam to the Japanese market between 2009 and 2019. The fish sector has dealt with several challenges in meeting the requirements of the importing countries.
A microbial assessment scheme (MAS) has been developed to assess the microbiological quality and safety using the current FSMS within a company according to the principle of low numbers of microorganisms and small variations in microbial counts, indicating an effective FSMS (Jacxsens et al., 2011). From the MAS data, contamination profiles, including microbial counts and distribution of microbial contamination, provide insight into the dynamics of microbial contamination (Jacxsens et al., 2009). In this respect, the MAS protocol has been used to assess the actual microbiological performance of FSMS for meat and fresh vegetable production chains (Daelman et al., 2011; Oses et al., 2012; Njage et al., 2016). In the fish industry, studies have shown high contamination levels and the presence of pathogens, i.e., V. cholerae, and L. monocytogenes, throughout the entire production process of Pangasius fish in Vietnam using MAS (Noseda et al., 2013; Tong Thi et al., 2014). However, a comparison of the microbiological profile of Pangasius fish collected from two different companies has not been conducted. Additionally, the performance of FSMS in actual practice is still variable, although FSMS has been certified and implemented in practice. Therefore, in this study we assessed the microbial safety of this freshwater fish as a result of the implemented FSMS. The obtained results will be valuable for designing appropriate measures to improve and control microbial quality and safety during the processing of Pangasius fish.
Characterization of the companies This study was conducted in two Pangasius fish-processing plants located in the Vietnamese Mekong Delta (VMD). Company A is upstream of the Hau River while company B is on the opposite side of the Tien River. Both rivers are two main branches of the Mekong River through Vietnam. Company A is one of the largest companies in the VMD, producing 250 tons of raw fish per day with 2 000–2 500 workers. The company has implemented HACCP, BRC, IFS, and Halal standards, and mainly exports fish to South American countries (Brazil and Colombia), Asia (Thailand, Singapore, and China), and the EU (Spain, Poland, France, and Italy). Company B is a medium-sized company with 800–900 workers (100 tons of raw fish per day), has implemented HACCP, BRC, Halal, and ISO 9001:2000 standards, and mainly exports to Middle Eastern countries (UAE and Qatar) and South Africa. A flowchart of the production process of the two companies is shown in Fig. 1.
Flowchart of the production process of Vietnamese Pangasius fish in two companies. Processing steps shown as bold are only in B company. Critical sampling location (CSL) selected along the processing lines were marked as CSL1 to CSL16
Selection of critical sampling locations Critical sampling locations (CSL) are those in which loss of control will lead to unacceptable food safety problems due to contamination, growth, and/or survival of microorganisms (Jacxsens et al., 2009). In this study, 16 CSL along the processing lines were identified as shown in Fig. 1. The samples included fish (raw fish, intermediate, and final products), workers' hands/gloves, food contact surfaces (FCS), and water that was randomly collected and analyzed once per sampling.
Selection of microbiological parameters Table 1 lists the microbiological parameters examined at each sampling location. Total mesophilic counts (TMC) were selected as an indicator of overall quality. Coliforms, E. coli, and S. aureus were selected as hygiene indicators. The presence of pathogens, such as L. monocytogenes, Salmonella spp., and V. cholerae, was also monitored.
| Microbiological parameters | Fresh fish/filletsa (log CFU / g) |
Food contact surfacesb (log CFU / 100 cm2) |
|
|---|---|---|---|
| Goal | Tolerance | ||
| Total mesophilic counts | 5 | 6 | Good, ≤ 3; moderate 3–4.5; poor ≥ 4.5 |
| Enterobacteriaceae/Coliforms* | 2 | 3 | Good, ≤ 3; moderate 3–4.5; poor ≥ 4.5 |
| E. coli | 2 | 3 | Absence in tested area |
| S. aureus | 2 | 3 | Absence in tested area |
| L. monocytogenes | Absence in 25 g | Absence in 25 g | Absence in tested area |
| V. cholerae | Absence in 25 g | Absence in 25 g | Absence in tested area |
| Salmonella spp. | Absence in 25 g | Absence in 25 g | Absence in tested area |
The “goal” values indicate a target limit for the end product
The “tolerance” values indicate the upper acceptable limit
Definition of sampling frequency Each company was visited for three days in three consecutive weeks. Samples were taken three times per sampling day (at the beginning of production day: 7:00–8:00, at an intermediate time: 11:00–12:00, and the end of the production day: 16:00–17:00) and on three independent sampling days. Overall, a total of 288 samples (or 144 samples per company) were taken in September 2018 (company A) and January 2019 (company B).
Sampling Two Pangasius fillets (approximately 300–500 g) were randomly taken and placed into a Stomacher® bag (Interscience, Ile-de-France, France) using sterile tweezers. For the water sample, approximately 250 mL of water was collected into sterile Stomacher® bags. A sampling of food processing surfaces and personnel hands was performed by the swab method, with dimensions of a 100 cm2 area. The pre-wetted swab was placed in 5 mL maximum recovery diluent (MRD, Merck, Darmstadt, Germany) for quantitative analysis of TMC, E. coli, coliforms, and S. aureus; in 5 mL of Demi-Fraser broth (Merck) for detection of L. monocytogenes; in 5 mL alkaline saline peptone (Merck) for V. cholerae; and in 5 mL buffered peptone water (Merck) for Salmonella spp. All samples were aseptically taken, stored on ice, and transported to the Laboratory of Microbiology and Biotechnology (Can Tho University, Can Tho city, Vietnam) for microbial analyses within 24 h of sampling.
Enumeration of TMC, E. coli, coliforms, and S. aureus Twenty-five grams of fish sample was mixed with 225 mL MRD and homogenized for 1 min in a Stomacher® bag. Furthermore, 1 mL of water sample was transferred to 9 mL of MRD. A decimal serial dilution series was then prepared in the MRD. Enumeration of TMC was determined on plate count agar (Merck) and incubated at 37 °C for 2–3 d. E. coli and coliforms were enumerated on Coliforms Agar Enhanced Selectivity (Merck) by incubating for 24 h at 37 °C. S. aureus was enumerated by spread plate on Baird Parker Agar (Merck) with 25 mL/500 mL Egg Yolk Tellurite Emulsion (Merck) after an incubation period of 48 h at 37 °C, and confirmation of S. aureus occurred with a Bactident® Coagulase (Merck).
Detection of L. monocytogenes, V. cholerae, and Salmonella spp. The detection of L. monocytogenes and Salmonella spp. was performed according to the International Organization for Standardization (ISO) ISO11290-1:1996, and ISO 6579-1:2002 (ISO 1996, 2002). The presence of V. cholerae was carried out as previously described by Tong Thi et al. (2014). For detection of L. monocytogenes, samples were pre-enriched in Demi-Fraser broth by incubation for 24 h at 30 °C. Then, a second enrichment was incubated in Fraser broth (Merck) for 48 h at 37 °C. Subsequently, the culture was streaked on Listeria selective agar (base) acc. OTTAVIANI and AGOSTI (Merck) for 48 h at 37 °C. Typical L. monocytogenes colonies have green-blue color surrounded by an opaque halo, and these were confirmed by carbohydrate fermentation. Concerning the detection of Salmonella spp., samples were pre-enriched in buffered peptone water at 37 °C for 18 h. The first pre-enriched culture was then transferred to Rappaport Vassiliadis soya peptone broth (Oxoid, Basingstoke, UK) and incubated at 42 °C for 24 h. Afterward, a loopful culture was streaked onto xylose lysine deoxycholate agar (Oxoid) and incubated at 37 °C for 24 h. Colonies of presumptive Salmonella spp. (red colonies with a black center) were sub-cultured and confirmed through biochemical tests. For detecting V. cholerae, pre-enriched samples were incubated in alkaline saline peptone water (pH = 8.6) at 42 °C for 6 h. After that, 1 mL of the pre-enriched sample cultures were inoculated into 10 mL of alkaline saline peptone water and incubated for 18 h at 42 °C. The second culture was then streaked onto thiosulfate citrate bile salts sucrose agar (Merck) and incubated at 37 °C for 24 h. Typical colonies (yellow and smooth colonies) were streaked onto tryptone soya agar (Oxoid) supplemented with 1.5% sodium chloride (Merck) at 37 °C for 24 h, followed by confirmatory biochemical tests.
Data interpretation and microbiological safety level profile The actual microbiological assessment was interpreted according to the guidelines, and Vietnamese National Standards (TCVN) are summarized in Table 1 (Sampers et al., 2010; TCVN, 2010; Uyttendaele et al., 2010). The quality of water used for washing and glazing fish has to meet the requirements of potable water standards according to European Union (EU) and TCVN (EU, 1998; TCVN, 2009). A microbiological safety level profile (MSLP) was specified by estimating each parameter for each CSL by a score from 1 to 3, indicating from poor to good food safety (Jacxsens et al., 2009). In this study, the MSLP score should be a maximum of 21 (7 microbiological parameters × 3 score maximums). If the sum of the microbial safety levels is 7 to 8, the assigned score will be 1. If the sum is 9 to 12, then the assigned score will be 1–2. The assigned score is 2 when the sum ranges from 13 to 15. An assigned score of 2–3 will be given when the sum is 16 to 19. Finally, an assigned score of 3 is attributed if the sum is 20 or 21 (Jacxsens et al., 2009; Sampers et al., 2010).
MAS results give information on microbiological problems in an implemented FSMS. The range of microbiological counts at each CSL indicating the extent of variation of the microbial parameters are shown in Table 2, while Table 3 provides information on the contamination level of pathogens.
| CSL | Type of samples | Processing step | Hygiene indicators | Overall indicator | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| E. coli | Coliforms | S. aureus | Total mesophilic counts | |||||||
| A | B | A | B | A | B | A | B | |||
| CSL1 | Fisha | Raw | < 2 | < 2–3.6 (1/9) | 1.5–2.7 | < 2–4.9 (4/9) | <1 | <1 | 3.7–5.4 | 3–7.2 (4/9) |
| CSL3 | Filleting | < 2–2.8 | < 2–3.4 (3/9) | < 2–5.8 (7/9) | < 2–4.9 (6/9) | <1 | 1.6 | 3–6 (1/9) | 5.6–8 (6/9) | |
| CSL7 | Trimming | < 2–3.7 (3/9) | 3–4.7 (9/9) | 2.8–6.4 (9/9) | 3.1–5.9 (9/9) | <1 | <1 | 4.9–8.1 (7/9) | 6–8.5 (9/9) | |
| CSL12 | Cooling | < 2–3.9 (2/9) | 3–4.5 (9/9) | 4.1–6.2 (9/9) | 3–5.5 (9/9) | <1 | <1 | 5.8–7.8 (7/9) | 7–8.8 (9/9) | |
| CSL14 | Packaging | < 1–2.3 | < 2–3.8 (2/9) | 3.1–4.2 (9/9) | < 2–5.8 (4/9) | <1 | 1.2 | 5.5–7.5 (6/9) | 5.3–7.6 (8/9) | |
| CSL4 | Hands/glovesb | Filleting | < 2–3.3 (9/9) | < 1–3.1 (6/9) | 3.5–6.2 (6/9) | 2.7–6 (3/9) | <1 | 1.5 (1/9) | 5.5–7.7 (9/9) | 6–8.2 (9/9) |
| CSL8 | Trimming | < 2–3.3 (9/9) | 2.5–4.4 (9/9) | 4–5.9 (7/9) | 3.3–6.6 (4/9) | <1 | <1 | 5.9–8 (9/9) | 6.3–9 (9/9) | |
| CSL15 | Packaging | < 2–3.1(9/9) | < 1–2 (1/9) | 3.1–4.2 | < 1–2.3 | 1.4 (1/9) | <1 | 5.1–6.8 (9/9) | 4.6–6 (9/9) | |
| CSL6 | Food contact surfacesb | Skinning | < 2–3.6(9/9) | < 2–4.7 (9/9) | 2.3–5.9 (5/9) | 4.2–7 (7/9) | NA | NA | 6.2–7.5 (9/9) | 5.7–8.9 (9/9) |
| CSL9 | Trimming | < 2–5 (9/9) | < 2–4.6 (9/9) | 3.5–6 (7/9) | < 2–6.3 (5/9) | NA | NA | 5.7–8.3 (9/9) | 6.0–8 (9/9) | |
| CSL11 | Washing 2 | < 2–3.5 (9/9) | < 2–4.2 (9/9) | < 2–6 (6/9) | 3.3–6 (5/9) | NA | NA | 6–7.7 (9/9) | 5.3–8.6 (9/9) | |
| CSL16 | Packaging | < 2 (9/9) | < 2 (9/9) | 2–5 (5/9) | < 2–3.3 | NA | NA | < 5–7 (8/9) | 5.3–8.1 (9/9) | |
| CSL2 | Waterc | Bleeding | 3.2–5.3 (9/9) | < 4–5.6 (9/9) | 5.8–7.0 (9/9) | 4.8–7 (9/9) | NA | NA | 8.3–9.6 (9/9) | 7.3–9.4 (9/9) |
| CSL5 | Washing 1 | < 3–3.8 (9/9) | < 4–5.7 (9/9) | 5.1–6.4 (9/9) | 4.9–6.3(9/9) | NA | NA | 6.3–8.2 (9/9) | 6–11.3 (9/9) | |
| CSL10 | Washing 2 | 3–5.2 (9/9) | < 4–5.6 (9/9) | 4–6.8 (9/9) | 4–6.3 (9/9) | NA | NA | 8.3–9.5 (9/9) | 7.3–10 (9/9) | |
| CSL13 | Glazing | < 3–4.1 (9/9) | < 3 (9/9) | 3–6.3 (9/9) | < 3–5.3 (9/9) | NA | NA | 6–7.6 (9/9) | 4–6 (7/9) | |
| Microbiological safety level* | 1 | 1 | 1 | 1 | 2 | 2 | 1 | 1 | ||
Numbers in parentheses show the number of samples that exceeded the tolerance values/total samples; NA: Not analyzed
| CSL | Type of sample | Processing step | L. monocytogenes | V. cholerae | Salmonella spp. | |||
|---|---|---|---|---|---|---|---|---|
| A | B | A | B | A | B | |||
| CSL1 | Fisha | Raw | N | P (2/9) | N | P (1/9) | N | N |
| CSL3 | Filleting | N | N | N | P (1/9) | N | N | |
| CSL7 | Trimming | P (2/9) | P (3/9) | P (3/9) | P (1/9) | P (1/9) | N | |
| CSL12 | Cooling | P (7/9) | P (5/9) | P (6/9) | P (2/9) | N | N | |
| CSL14 | Packaging | P (5/9) | P (3/9) | P (2/9) | P (1/9) | N | N | |
| CSL4 | Hands/glovesb | Filleting | N | P (1/9) | N | N | N | N |
| CSL8 | Trimming | P (1/9) | P (2/9) | P (2/9) | P (1/9) | N | N | |
| CSL15 | Packaging | P (2/9) | N | P (1/9) | N | N | N | |
| CSL6 | Food contact surfacesb | Skinning | N | N | NA | NA | N | N |
| CSL9 | Trimming | P (4/9) | N | NA | NA | N | N | |
| CSL11 | Washing 2 | P (4/9) | P (1/9) | NA | NA | N | N | |
| CSL16 | Packaging | P (1/9) | N | NA | NA | N | N | |
| CSL2 | Waterc | Bleeding | N | P (3/9) | NA | NA | N | N |
| CSL5 | Washing 1 | N | P (1/9) | NA | NA | N | N | |
| CSL10 | Washing 2 | P (6/9) | P (2/9) | NA | NA | N | N | |
| CSL13 | Glazing | P (2/9) | N | NA | NA | N | N | |
| Microbiological safety level* | 1 | 1 | 2 | 2 | 3 | 3 | ||
N: negative, P: positive, NA: Not analyzed
Numbers in parentheses show positive samples/ total samples
In Table 2, for hygiene indicators, a score of 1 was assigned for E. coli and coliforms since most of the samples carried high levels of contamination. For E. coli, nearly 100 out of 144 samples exceeded the acceptable levels on the product or environmental samples at each processing company. E. coli was found above the goal limit on the fillet samples at the filleting step (CSL 3) and subsequent intervention steps (CSL 7, 12, and 14). In addition, high coliforms (> 3 log CFU/g) were also present above the goal limit in the final product (CSL 14). Furthermore, high counts of E. coli and coliforms were observed from workers' personal hands/gloves, FCS, and washing water caused by inadequate personal hygiene practices and poor sanitation procedures. The source of contamination is also related to the Pangasius fish gut, and transmission occurs from the filleting step onward (Noseda et al., 2013; Tong Thi et al., 2013). These unacceptable counts of coliforms and E. coli are in line with other reports in fish processing plants as well (Onjong et al., 2014; Kussaga et al., 2017). Although low levels of S. aureus were found (below the goal limit < 2 log CFU/g) in the fish samples (CSL 1, 3, 7, 12, and 14), this bacterium was still detected in the hands of workers (CSL 4 and CSL 15). Additionally, it is reported that S. aureus shows high biofilm-forming ability in fish processing plants (Gutiérrez et al., 2012; Frozi et al., 2017). Thus, the presence of this bacterium should be evaluated on food contact surfaces, because this is a possible transmission route. For the overall indicators, total mesophilic counts obtained a food safety level of 1 since the initial contamination of raw material was high (CSL 1), as farmed Pangasius fish may be contaminated with bacteria from the farming ponds (ICMSF, 2005). As a consequence, the contamination was spread out to the intermediate (CSL 7) and finished products (CSL 14). Surprisingly, high counts of TMC were detected at CSL 14, exceeding the goal limit of 5 log CFU/g for both companies. High TMC counts may indicate unsatisfactory hygiene practices (Svanevik et al., 2015). The observed counts of TMC, E. coli, and coliforms were primarily influenced by improper control activities such as hand hygiene of personnel, sanitation programs, and water contamination (Kulawik et al., 2016). High variation in bacterial contamination suggests that the implementation of FSMS by both companies A and B in actual practice is not sufficient to control these microbial parameters (Jacxsens et al., 2009). Moreover, the microbial counts of all samples in this study were higher than those analyzed from a company with a daily production capacity of 200 tons as investigated by Noseda et al. (2013) and another company with a capacity of 35 tons as observed by Tong Thi et al. (2014).
The results in Table 3 highlight that L. monocytogenes is a crucial concern in both companies. L. monocytogenes was assigned a score of 1 because it was detected in a considerable number of samples covering both fish and environmental samples, 34 out of 144 and 24 out of 144 samples for companies A and B, respectively. In a study at a catfish processing plant, Chen et al. (2010) also found L. monocytogenes contamination not only from fish at different production steps but also from FCS, water, and the hands of food handlers. Moreover, L. monocytogenes is known to survive in the food production environment for long periods in extreme conditions by forming biofilms (Rodríguez-López et al., 2018). The growth of biofilms on FCS could act as a key source of L. monocytogenes cross-contamination in the food industry (Colagiorgi et al., 2017; Mazaheri et al., 2021). In addition, the raw fish is considered a possible transmission route as L. monocytogenes has been detected on the fish surface and in the stomach lining, gills, and intestine (Jami et al., 2014). Hence, poor handling practices and insufficient cleaning and disinfection procedures are proposed as the main reasons for the predominance of L. monocytogenes. In contrast, a maximum score of 3 was obtained for Salmonella spp., indicating that it was not detected or only found in one analyzed sample. Although the gastrointestinal tracts of mammals, birds, and reptiles are believed to be the natural habitat of Salmonella spp., several studies have reported a high prevalence of Salmonella spp. in aquaculture from fresh fish culture ponds to fish products (Winfield and Groisman, 2003; Lotfy et al., 2011; De Souza Sant'Ana, 2012; Elhadi, 2014; Bibi et al., 2015; Nguyen et al., 2016). However, our data indicate that the contamination level of this pathogen was in line with the criteria. Furthermore, Salmonella spp. were not detected within the entire processing chain of Pangasius fish from previous studies by Noseda et al. (2013) and Tong Thi et al. (2014). Thus, Pangasius fish may not be a potential source of Salmonella spp. contamination. In the case of V. cholerae, a similar score of 2 was assigned, since several samples were detected: 15 samples out of 144 and 7 samples out of 144 for companies A and B, respectively. V. cholerae is abundant in marine and freshwater environments; therefore, fish are considered to be a vehicle for the dissemination of V. cholerae (Senderovich et al., 2010; Halpern and Izhaki, 2017). In our study, V. cholerae was also detected on the hands/gloves of workers at the step of trimming and packaging in both companies, showing insufficient personal hygiene. Hence, the transmission path of this pathogen may come from the gastrointestinal parts during filleting and the hands of workers. In both processing processes at companies A and B, high levels of bacterial contamination were observed in FCS, especially at the step of trimming and washing 2 (Tables 2 and 3). The manual handling operation may induce contamination from hands or FCS into the trimmed fillets and then spread out in subsequent processing steps. FCS, including cutting boards, plastic trays/baskets, or conveyor belts are at high risk of contamination in fish processing plants (Consuelo et al., 2014; Sheng and Wang, 2020). Moreover, we observed that the duration required at the cooling step was commonly extended by 10–12 h during actual processing at these companies. In addition, the containers were not adequately cleaned owing to a lack of processing equipment, and the temperature was not regularly checked to guarantee it was below 4 °C. These conditions reflect insufficient sanitation measures in both companies. Thus, the implementation of good hygiene practices needs to be improved in actual situations.
From the food safety level of each microbiological parameter shown in Tables 2 and 3, a similar food safety level (11 of a possible 21) was observed among the microbiological parameters for companies A and B, respectively. The MSLP score was lower than the maximum value, hence improvement of the current FSMS is needed in both fish-processing plants. As a result of the MSLP, an overall score of 1–2 was assigned, indicating that the microbial risk throughout the processing lines was considerably high. Herein, both companies A and B seem to have the same problem with the control of L. monocytogenes, V. cholerae, too high viable counts of TMC, and coliforms that were present in the final product. In contrast to company A, company B had problems with the quality of raw fish (the presence of pathogens, excessively high counts of TMC, and coliforms). Moreover, there were differences in production time (approximately 4 h compared to 5 h for companies A and B, respectively) and the scale of production between these companies. From the aspect of different locations, there were substantial dissimilarities between the Tien and Hau rivers in terms of riverbed elevation, including water-level fluctuation and turbidity (Binh et al., 2018). This difference may lead to a variance in water quality and bacterial pollution in the two main rivers of the VMD. Moreover, Nguyen et al. (2008) reported that the bacterial ecology of Pangasius fish differs depending on the habitat. Supposedly, the situation of the two companies causes the difference in their safety output. Despite this, the microbiological assessment results showed that food safety output was similar at both companies regardless of their current status. It can thus be suggested that cross-contamination inside the fish-processing environment may contribute to the similarity of microbial safety profiles at both companies. This, therefore, validates the efficiency of FSMS and how it plays a vital part in guaranteeing food safety at Pangasius fish-processing plants.
In conclusion, in the present study, we found that the microbiological assessment of two Pangasius fish-processing companies was influenced by different features. Even though the quality of raw Pangasius fish was different at each plant, the microbiological problem was indistinguishable at the company level. This emphasizes that the environment inside fish-processing plants has a significant impact on the microbial safety of the products made. Hence, cleaning and disinfection programs (processing areas and equipment), proper hand hygiene, and improving sanitation must be sufficiently accounted for at fish-processing companies to guarantee food safety output.
Acknowledgements This work was supported by the Can Tho University Improvement Project VN14-P6 and supported by a Japanese ODA loan.
Conflict of interest There are no conflicts of interest to declare.
