• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Food

  • Print
  • Share
  • E-mail

Processing Parameters Needed to Control Pathogens in Cold Smoked Fish Chapter VI. Control of Food Safety Hazards during Cold-Smoked Fish Processing

(Table of Contents)

 

Scope

This section describes a general process of cold-smoking finfish (see also Appendix A) from the fishing vessel to the consumer. It was not the intention of the panel for this section to be a Hazard Analysis Critical Control Point (HACCP) plan. This section is limited to the following significant concerns: pathogenic bacteria, production of biogenic amines, and parasites. Other chemical or physical hazards are not included. Each step of this general process is analyzed in terms of identifying potential hazards, control points, and processing parameters when scientifically substantiated. Verification of the processing parameters is included in Appendix C. It is beyond the scope of this document to capture all of the wide variety of cold-smoking methods. Therefore, differences from the cold-smoking process described here may require a different analysis and evaluation. See Appendix D for survey data received from cold-smoked processor respondents. Appendix B describes some of the most common vessel practices. This chapter addresses some recommendations by the Association of Food and Drug Officials (AFDO 1991). It is noteworthy to clarify, however, that although developed by experts and widely used, these recommendations have not necessarily been scientifically evaluated.

1. Receiving

1.1. Background and support information

Hazards posed by fish and fish products vary and depend on the fish environment, both while the fish was alive and after it was caught. For example, a wild-caught, scombrotoxin-susceptible fish that has been kept for some time in the water or on a boat before being iced or frozen can have elevated toxic levels of biogenic amines. Vessel practices would help prevent biogenic amine development in scombrotoxin-susceptible species. Waters where fish are harvested may contain bacteria or spores that may be pathogenic to humans (Clostridium botulinum or Listeria monocytogenes). Fish also may come in contact with pathogenic microorganisms during harvesting, handling on board, and offloading and transportation to a smoking facility. Rinsing wild-caught fish with potable water (not environmental surface water) may aid in reducing L. monocytogenes in certain areas during certain seasons. Additionally, wild-caught fish may carry parasites that could be infectious to humans.

Aquaculture provides a year-round supply of fresh fish and a significant amount of raw material for cold-smoked fish processors. In general, aquaculture has the potential to be a highly controlled and hygienic operation at the initial harvesting and handling step, particularly in the case of salmon. The microbial flora present on aquacultured fish is affected by water quality and feed composition. Water quality can range from that in raceways or tanks fed by microbe-free spring water, well water, ozone-treated water, and ultraviolet light-treated water, to that in mud ponds purposely enriched with untreated human or animal waste. Hence, regardless of whether a fish has been wild-caught or farm-raised, it may carry various types of pathogenic bacteria when it arrives at the smoking facility. On the positive side, farm-raised fish that are fed with a controlled diet, such as processed feed pellets, and reared in closed systems with purified or treated water have been shown to be free of parasites (Angot and Brasseur 1993). Due to the amount of variation in water quality and conditions for farm-raised fish, a processor must consult with the supplier and not assume that the aquaculture product is parasite free.

The quality of the water should be one of the factors that determine the stringency of harvest and post-harvest treatments. Processors of cold-smoked fish should be concerned with the microbiological quality of the primary product they receive from suppliers since product from contaminated sources may not be appropriate for production of cold-smoked products. In general, good sanitation procedures should be applied throughout harvest, transportation, storage, and post-harvest handling.

Transportation of the fish from harvest or processing to the cold-smoking facility is another area of concern. Fish can be transported frozen or refrigerated (typically iced and thus cooled to 0 °C (32 °F). Basic sanitation practices and temperature controls need to be in place in transportation vehicles and containers. Temperature should be monitored at all times during transportation and storage before processing, especially if the fish is being frozen to control parasites or belongs to a scombrotoxin-susceptible fish.

In the United States, direct treatment of finfish to reduce microbial load is permitted after harvest and before processing. Chlorine solution dips (Eklund and others 1993), which require intense management to avoid recontamination, have been replaced by chlorine solution rinses or sprays that are followed by a rinse with potable water. Although it has been suggested that rinsing of the fish is important to reduce numbers of pathogenic microorganisms, no data on the effect of this procedure could be found in the scientific literature.

In summary, efforts directed to reduce or minimize microbial load on fish destined for cold-smoking must be initiated as early in the production cycle as practical. The most critical interventions, however, are those directed to prevent growth of specific bacteria that may eventually lead to health hazards. At all times, from harvest through receiving, Good Manufacturing Practices (GMPs), Good Hygienic Practices (GHPs), appropriate sanitation procedures (that is, Sanitation Standard Operating Procedures, [SSOPs]), and control of the product temperature should be applied. After harvest, the fish should be cooled as soon as practically possible and maintained at a temperature below 40 °F (4.4 °C) until processed. Ideally, fish is stored at 32 °F (0 °C) in ice. The use of temperature-recording devices is highly recommended.

 

1.2. Potential hazards

Incoming fish may harbor parasites. Scombrotoxin-susceptible fish may have been temperature-abused and contain unsafe levels of biogenic amines. Additionally, fish may be contaminated with C. botulinum or L. monocytogenes.

 

1.3. Control point

Receiving is a control point for aquacultured and wild-caught fish. For scombrotoxin-susceptible fish, receiving is a control point to screen for harmful, unsafe levels of biogenic amines. If the product is received frozen and there are no other freezing steps in the process, this is a control point for parasites.

 

1.4. Processing parameters

If scombrotoxin-susceptible fish are received directly from the harvest vessel, all lots should be accompanied by documentation certifying proper time and temperature handling of the fish. Refrigerated scombrotoxin-susceptible fish should be received at an internal temperature of 40 °F (4.4 °C) or less. Table VI-1 shows the estimated shelf life of scrombrotoxin-forming species at various storage temperatures. In addition to the time and temperature parameters, analytical testing for histamines should be done periodically on samples of edible fish flesh (FDA guidance limits the histamine level to 50 ppm). A recent report on Hawaiian fishery (Kaneko 2000) concluded that odors of decomposition were reliable indicators of biogenic amines risk and that sensory evaluation is an effective monitoring measure. Even though an experienced processor may be doing the evaluation, sensory analysis is a very subjective and not a quantitative monitoring method; therefore, a more objective method (analytical method or temperature record) should be in place. Practically, most companies use a sensory evaluation of incoming scombrotoxin-susceptible species with a maximum receiving temperature for refrigerated raw material received. If sensory analysis points to a high biogenic amine level, analytical testing is performed.

If product is received frozen to control for parasites, a number of time and temperature combinations have been recommended in the United States, such as holding the fish at -10 °F (-23 °C) for 60 h, or less than -4 °F (-20 °C) for 7 d, or -31 °F (-35 °C) internally for 15 h. For a more detailed explanation of freezing regimes, see Chapter V.

 

Table VI-1. Approximate Safe Shelf Life of Scrombrotoxin-Forming Species at Various Storage Temperatures (FDA 1998)

Product temperature Safe shelf life (d) Safe shelf life (d) with delayed cooling
0 °F (-17.8 °C) No limit No limit
32 °F (0 °C) 14 8
38 °F (3.3 °C) 10 7
40° F (4.4° C) 7 5
50 °F (10 °C) 3 0
70 °F (21.1 °C) 0 0
90 °F (32.2 °C) 0 0

2. Fresh or frozen storage

2.1. Background and support information

Before smoking, fish may be stored fresh or frozen. Fresh and frozen storage areas should be maintained in a clean and sanitary manner. Refrigerated fish should be stored so that their internal temperature minimizes the production of toxic biogenic amines and the outgrowth of L. monocytogenes. Frozen storage may potentially be a control point for parasites, if there is no other freezing step in the process. In this case, freezing of fish to the proper internal temperature and for the proper length of time is necessary to kill parasites in the incoming product that would otherwise survive the cold-smoking process.

 

2.2. Potential hazards

Contamination of the raw material or outgrowth of pathogenic microorganisms may occur if the fish is not maintained in a sanitary facility with proper refrigeration conditions.

If the frozen storage is intended to be the step to kill parasites, and the product is not frozen at the proper temperature for the proper length of time, parasites may become a hazard in the finished product.

 

2.3. Control point

If fish are stored under refrigeration, this is a control point for biogenic amines in scombrotoxin-susceptible species. If fish are stored frozen, this could be a control point for parasites, provided a previous freezing step has not been included as the control point for parasites or there is no other freezing step later in the process.

2.4. Processing parameters 

Refrigerated fish should be stored so that their internal temperature is less than 40 °F (4.4 °C). For freezing, a number of time and temperature combinations have been recommended, such as holding the fish at -10 °F (-23 °C) for 60 h, or less than -4 °F (-20 °C) for 7 d, or -31 °F (-35 °C) internally for 15 h. For a more detailed explanation of freezing regimes, see Chapter V.

3. Thawing, washing, and rinsing

3.1. Background and support information

Thawing frozen raw fish is necessary to process cold-smoked fish. Many cold-smoked fish processors receive frozen fish, either aquaculture-produced or wild-caught, for processing. Thawed or fresh fish also needs to be thoroughly washed and rinsed after being received into the plant and then again after being butchered and processed. For these activities, it is necessary to use cold, potable, and continuously flowing water.

 

3.2. Potential hazards

Frozen raw fish should be thawed under sanitary conditions. Although this step is short and one would not anticipate biogenic amine production or pathogen growth, AFDO guidelines state that frozen raw material should be thawed under temperature control (AFDO 1991). This step is not considered to be a control point but it should be performed following GMPs and GHPs.

4. Butchering and evisceration

4.1. Background and support information

Butchering and evisceration may occur at the smoking facility or at the supplier's processing plant before the raw material is shipped to the smoker. Many cold-smoked fish processors receive raw material in the form of fillets or other products that have already been completely or partially processed (eviscerated, headed, gutted, and skinned). Temperature records are needed and the product should be checked for cleanliness when it is received into the processing facility.

If fish are eviscerated or butchered at the smoking facility, AFDO guidelines recommend that gutting be performed "with minimal disturbance of the intestinal tract contents." The guidelines further recommend that fish should be butchered in a room or area that is separate from the rest of the smoking and processing facility. The fish, especially the body cavity, should be washed and rinsed thoroughly with potable water after butchering.

 

4.2. Potential hazards

This step does not present a significant or special potential hazard and consequently is not considered to be a control point. Nevertheless, butchering and evisceration should be done promptly and should follow GMPs and GHPs.

5. Washing and rinsing 

See section 3 above.

6. Sorting, sizing, and salting

6.1. Background and support information

Salting of fish can be conducted by two different methods, brining and dry-salting. Brining is the process by which the fish is soaked or steeped (Kassem 1977) in a solution consisting of water, salt, sugar, various spices and flavorings, phosphates, and, depending on the recipe and species of fish (allowed in the United States for sable, salmon, shad, chub, and tuna), sodium nitrite. Brining may also be accomplished by injecting a liquid brine solution, usually with an automated or mechanized system but occasionally by hand, into the fish. In dry salting, fish are placed directly into a salt mixture usually consisting of some variation of the previously mentioned ingredients without the water.

Brining solutions are often referred to as a percent of a totally saturated salt solution and may be measured with a salometer or other methods. A saturated brine solution is one in which no more salt will dissolve into the water (aqueous phase). A salometer measures in degrees (100 °S), which can be thought of as "percent saturation" of the brine solution or directly in % salt. Tables and guides are available for preparing salt brines from the U.S. Sea Grant Extension Service (Hilderbrand 1973).

Salting should be as uniform as possible, with the correct amount of salt or brine solution absorbed into the fish flesh of each piece. To accomplish this, the fish must be sorted by size and thickness and spread out uniformly prior to salting so they will absorb the same amount of salt. Eklund pointed out that "even under the best salting conditions, it is nearly impossible to obtain the predetermined concentration of salt in all samples even within the same lot of fish" (Eklund 1989). For example, when 50 samples from a commercial operation with a target of 3.5% (WPS) were tested, WPS concentrations of the final product ranged from 2.8 % to 6.0 %, with the majority of the samples being above 4.0 %. The inherent variability in this important process must be taken into consideration.

Proper salting is essential to cold-smoked fish processing for three reasons: 1) it usually reduces the moisture content of the fish, affecting both texture and shelf life; 2) it imparts essential flavors; and 3) it is important as a preservative and inhibitor of microbial growth (Kassem 1977). Eklund points out that with cold-smoked products, the addition of sugars would provide an advantage in that the carbohydrate source would "encourage the growth of lactic acid and other spoilage bacteria and lower of the product pH which can be inhibitory to C. botulinum" in the final product (Eklund 1989). Other researchers, however, noted that if the product is vacuum-packaged and chilled, very rapid growth of lactic acid bacteria (Carnobacteria spp.) is observed, independent of addition of carbohydrates (L. Gram 2000; personal communication; unreferenced).

 

6.2. Potential hazards

 

Listeria monocytogenes and C. botulinum spores present on a single fish could contaminate an entire batch within the brine solution. To minimize microbial growth and cross-contamination, temperature control of the brine solution during brining is recommended, particularly if brining is done for more than a few hours. If brine injection is used, needles can get contaminated and spread a pathogen to other fish. Fish may be underbrined if the brine solution is too diluted or if the brine soak time is too short, potentially allowing botulinum toxin formation in the final product.

This step is especially important in that WPS needs to be high enough to inhibit the outgrowth and toxin formation of C. botulinum in the final product. Brining should be done at a salt concentration that will provide the appropriate concentration in the final product (that is, 3.5% WPS in the final product). Acidity (pH), salt, moisture (water activity), or some combination of these act in combination to inhibit Clostridia outgrowth and toxin formation. The salt level and refrigerated storage of the final product along with competing microflora can prevent the growth of C. botulinum type E and nonproteolytic types B and F. The reader is referred to the C. botulinum section for a review of the scientific literature on this subject.

 

6.3. Control point

Salting, including sizing and sorting of the fish, is a control point because the presence of enough salt in the fish is essential to inhibit the outgrowth of Clostridia species and to prevent the formation of toxins, particularly in vacuum-packaged finished products. The sizing and sorting step is an integral part of proper salting in cold-smoked fish processing. Batches of fish should be checked going into or being removed from the brine. Portions that are too thick or too large should be removed and cut to the proper size.

 

6.4. Processing parameters

A number of parameters need to be considered when salting. These include: 1) a minimum salt concentration of the brine solution, usually measured at the beginning of salting; 2) a minimum ratio of brine solution or dry salt mix to fish so that each fish is adequately exposed to the brine; 3) a minimum salting time to allow the salt to adequately and uniformly absorb into the fish; and 4) a maximum temperature of the salting. A further requirement is that the fish or fish portions be of a uniform thickness and size so that in the given amount of time for the batch, all of the portions absorb sufficient salt. As mentioned above, extra thick or large pieces in a batch would not be sufficiently salted, meaning they would not have a high enough salt concentration in the final product to inhibit Clostridia growth. Conversely, small or thin pieces would be oversalted at the end of this step, compromising the sensory characteristics. The AFDO guidelines recommend that the temperature of the brine not exceed 60 °F (15.5 °C) at the beginning of salting. If the salting is longer than 4 h, it is recommended that the salting be done under refrigeration.

The amount of salt, volume of the brine, weight of the fish, and duration and temperature of the process should be calculated empirically by the processor. All these parameters need to be established with the objective of obtaining a final product with at least 3.5% salt concentration in the water phase, if the product is to be vacuum-packaged (see Packaging, section 11). No scientific data specifically address the question of the salt concentration needed to inhibit C. botulinum toxin formation on air-packaged products. Although it is believed that spoilage of aerated products will act as a safeguard against botulism, not enough scientific evidence supports this idea (see Chapter III).

7. Rinsing, draining, and preparation

7.1. Background and support information

The AFDO guidelines recommend rinsing fish with fresh water after salting. This process, sometimes called "freshening," involves rinsing in cold, potable water for a certain amount of time. After the rinse, the fish may be laid on a rack to begin the drying and smoking stage.

 

7.2. Potential hazards

Cross-contamination with pathogens that may be present on the fish is unlikely during freshening due to rinsing with potable water. Consequently, the freshening step is not considered to be a control point.

8. Drying and cold-smoking

8.1. Background and support information

A number of cold-smoking procedures involve a drying stage with no smoke added to the product. This can be considered a separate step or an initial part of the cold-smoking step. The product is held at a specific temperature for a specific amount of time before the smoke is introduced. The duration and temperature of this initial drying step constitute the "art" of smoking fish. The parameters that are considered for this initial drying component include the type or species of fish, its fat content, and humidity. If the humidity is high, the fish will tend to be wetter longer and the drying step must be adjusted accordingly. Smoke should be applied before a pellicle (dried skin-like layer) forms on the outside of the fish fillets. If a pellicle forms first, the effectiveness of the smoke is markedly reduced. The goal is to have the surface of the fish portions a little "tacky" so the smoke can be absorbed (R. Martin 2000; personal communication; unreferenced).

There is a variety of literature on the smoking of fish and seafood products (Gilbert and Knowles 1975; Storey 1982). Variations include temperature, time, humidity, types of control (from a thermometer hanging in the smokehouse to sophisticated microprocessor feedback-controlled systems), types of smoke used (natural, generated, or liquid), and the design and types of smokehouses or kilns. Usually, fish are laid or arranged uniformly (that is, on racks) to prevent contact, ensuring that the portions receive a uniform exposure to the smoke and drying.

Traditional smokehouses or kilns allow natural airflow or convection, and their use requires a lot of expertise. Modern smokehouses or kilns control the airflow mechanically or electrically. Many also have refrigeration units; temperature monitors and controls; and microprocessor control systems that can regulate smoke exposure, humidity, drying, and temperature. This permits strict control of the complex smoking operation. Wood smoke can be generated by burning wood or, more commonly, by heating sawdust or small wood chips. Wood that has been treated with chemicals or preservatives should not be used for smoking. A wide variety of woods has been used for smoking, including oak, hickory, mahogany, pine, whitewood, cherry, apple, alder, mesquite, beech, birch, and maple to impart various flavors and colors. Some of the "original Scottish processes used smoldering peat or moss" (Storey 1982). According to some authorities, wood smoke is generally comprised of two components: the visible "tarry droplets" and the invisible gaseous or vaporous component (Storey 1982). Chemically, wood smoke contains hundreds of compounds, including a number of phenolic substances and volatile acids (Storey 1982; Gilbert and Knowles 1975). Liquid smoke products, which are produced by condensing the wood smoke in some fashion, are generally available. From a culinary perspective, the purpose of the smoking step in cold-smoked fish processing is to add flavor components and modify the taste and texture of the fish.

 

8.2. Potential hazards

Cold-smoked fish is defined as a product in which the fish flesh proteins show incomplete coagulation. Practitioners cite a range of temperatures as the upper-bound limit of cold-smoking, but the true defining limit is to leave the fish proteins partially undenatured or uncoagulated.

Although it is natural to assume that the product being smoked gets hot from exposure to the smoke and is thus cooked, cold-smoking is not a cooking step due to the low temperatures involved. In some instances the addition of smoke has been shown to have some inhibitory effect on microorganisms (Storey 1982). Research indicates that short term cold-smoking (< 24 h, as recommended by the AFDO guidelines [AFDO 1991]) causes a reduction rather than an increase in numbers of L. monocytogenes (see Chapter II for a review of this subject).

Contribution of smoke to the inactivation of pathogens, however, is not of significant importance in the overall process (see Chapters II and III for a review of the scientific literature). Rather, the low temperatures of smoking permit the survival of a significant portion of the natural microbiological flora on the fillets. Several studies have shown that growth of L. monocytogenes in smoked fish is hampered by a high background microflora (Rørvik and others 1991). It is a common belief that the natural flora will spoil the fish prior to botulinum toxin formation in cases of temperature abuse of air-packaged or unpackaged cold-smoked fish. On the contrary, vacuum packaging or modified atmosphere packaging (MAP) inhibits the growth of the natural flora so the rate of spoilage under temperature-abuse conditions is lower and toxin production may occur before spoilage. Consequently, relying on the competition from the naturally occurring background flora or on spoilage organisms to restrict C. botulinum is not a reliable and reproducible method of controlling toxin formation.

 

8.3. Control point

The cold-smoking step is a control point for C. botulinum growth and toxin production and for L. monocytogenes growth. Although cold-smoking will not completely eliminate either the microorganisms or the spores of concern, inadequate temperature and time control could exacerbate the hazards. Consequently, it is important to control the temperature and time during smoking so that the background microflora is not eliminated.

9. Cooling

9.1. Background and support information

Cold-smoked fish products must be rapidly cooled to minimize possible growth of L. monocytogenes and bacteria capable of producing biogenic amines in the products.

 

9.2. Potential hazards

Formation of biogenic amines is possible if scombrotoxin-susceptible species are used in the cold-smoking process. Also, low levels of biogenic amines have been found in non-scombrotoxin-susceptible species such as cold-smoked salmon (Gram and Huss 2000). Formation of amines in cold-smoked fish products during vacuum-packed storage may be caused by lactic acid bacteria, which have the ability to form biogenic amines (Leisner and others 1994). The primary amine-producing bacteria are species of Enterobacteriaceae, which typically are associated with temperature-abuse before packing. Rapid cooling of the product is important to reduce the growth rate of bacteria capable of forming biogenic amines.

Other potential hazards include pathogen growth and cross contamination of the cold- smoked product with pathogens. Listeria monocytogenes can survive the salting and cold-smoking process and will grow at refrigerator temperatures (Ben-Embarak and Huss 1992; Hudson and Mott 1993; Rørvik and others 1991). Thus, although complete assurance that L. monocytogenes is absent from cold-smoked fish products is not possible (Anonymous 1995; FAO 1999; Gram and Huss 2000), its numbers on the finished product may be kept low by rapid cooling of the cold-smoked product, following strict GMPs, and implementing and adhering to appropriate SSOPs.

 

9.3. Control point

Cooling is a control point for scombrotoxin-susceptible species. In addition, proper cooling will prevent or minimize growth of pathogenic bacteria, including L. monocytogenes.

 

9.4. Processing parameters

The AFDO guidelines state that all finished product must be cooled to a temperature of 50 °F (10 °C) or less within 3 h after cooking and further cooled to a temperature of 38 °F (3.3 °C) or below within 12 h after cooking. The New York State guidelines are similar to the AFDO guidelines except that the product must be cooled to a temperature of 50 °F (10 °C) or less within 5 h after cooking (Corby 1991).

It should be noted, however, that these temperature guidelines are under discussion as to their validity and applicability for the cold-smoked fish processing industry. Additional research is needed at this step to determine scientifically based temperature requirements at the cooling step.

10. Slicing and cutting

10.1. Background and support information

High priority must be given to the slicing and cutting step to help control possible recontamination or cross-contamination of the cold-smoked fish product with pathogens. Clean, sanitary food contact surfaces are essential. A processor should have well-designed and comprehensive SSOPs and must follow GMPs to control contamination of the cold-smoked fish product at this step. The development and verification of effective sanitation and cleaning programs will help reduce the prevalence of L. monocytogenes on the slicing and cutting equipment and in the general processing environment. Such programs will help reduce contamination of the cold-smoked fish product. Recent studies have shown that one potential source of L. monocytogenes in cold-smoked salmon is the slicer, where specific DNA types can occupy a niche for extended periods of time (Fonnesbech Vogel, Ojeniyi, Ahrens and others 2001).

As Eklund and others (1993, 1995) suggested, in addition, to inactivating the bacterium on the incoming raw product and inhibiting growth of survivors in the final product, processors should strictly follow GMPs to prevent recontamination of L. monocytogenes during processing. Because the source of L. monocytogenes may be other than the incoming product, following GMPs is currently the most important measure to minimize L. monocytogenes presence in the final product.

 

10.2. Potential hazards

Potential hazards during slicing and cutting include possible cross-contamination of the cold-smoked product with pathogens such as L. monocytogenes.

 

10.3. Control point

The slicing and cutting step is not a control point but is an extremely important processing operation. Strict adherence to SSOPs and GMPs to control cross contamination of product with pathogens is essential; in particular, effective SSOPs can be used to minimize or prevent cross-contamination with L. monocytogenes.

11. Packaging and labeling

11.1. Background and support information

Reduced Oxygen Packaging (ROP) is defined as any packaging procedure that results in a reduced oxygen level in a sealed package (FDA 1999). Furthermore, the Food Code defines the following: 1) Controlled Atmosphere Packaging (CAP) is the packaging of a product in a modified atmosphere followed by maintaining subsequent control of that atmosphere; 2) Modified Atmosphere Packaging (MAP) is the packaging of a product in an atmosphere that had a one-time modification of gaseous composition so it differs from air; and 3) Vacuum Packaging (VAC) removes the air from a package and seals it so a near-perfect vacuum remains inside the package. For fatty fish species, packaging that excludes oxygen is preferred to prevent rancidity caused by oxidation of lipids.

The United States seafood industry uses a variety of films and packages to protect the product and extend the shelf life of cold-smoked fish products. Packaging protects and preserves foods, providing an additional mechanism for marketing products by improving shelf life, convenience, freshness, and quality (Lord 2000). These packaging materials include vacuum bags and flexible pouches, vacuum shrink barrier bags and films, vacuum multi-layer bags, thermoformed and rollstock laminates, oxygen-permeable shrink bags, film overwraps, rigid sheeting, films for semi-rigid packages, and pre-formed trays. The types of packaging and film-permeability characteristics depend on the intended use and nature of the product. Packages can be designed to control film permeability to adjust oxygen, carbon dioxide, and moisture levels. Such information can be used to help extend product distribution shelf life without odor buildup within the package.

Other parameters being equal, gas permeability characteristics of the film will determine the microflora that survives in a packaged product. Therefore, gas permeability is an important parameter and should be taken into account when doing research and making decisions on food safety issues. Specifications for gas permeability, however, are product- and use-specific and are usually established at ambient temperatures under moderate humidity conditions (that is, 23 °C and 50% R.H.) using a variety of testing and verification methods. These various testing conditions make it difficult to compare gas transmission rates between films. In addition, film permeability rates also change when the film is stretched or when it contacts the product. Data are also limited on gas-transmission rates of films at product-chill temperatures (Robertson 1993).

 

11.1.1. Pathogens

Temperatures used in the preparation of cold-smoked fish are inadequate to eliminate C. botulinum spores; thus, other controls such as temperature and NaCl must be included to ensure its safety. Numerous studies address the formation of botulism toxin under vacuum. A few of the studies are done with fresh, hot-smoked fish and cold-smoked fish.

Eklund (1992) conducted studies on growth of C. botulinum and subsequent toxin formation in hot-smoked fish packaged in an oxygen-permeable film (1.5 ml polyethylene; oxygen transmission 7,195 cc/m2/24 h at 760 mm Hg 23 : C and 0% R.H.; CO2 transmission 22,858 cc/m2/24 h) and under vacuum in an oxygen-impermeable film (108 cc/ m2/24 h; CO2 transmission 526 cc/m2/24 h) (Table VI-2). He reported that higher concentrations of NaCl were required to inhibit toxin formation by C. botulinum in O2-impermeable films compared with O2-permeable films.

 

Table VI-2. Toxin production by Clostridium botulinum type E (103 spores/100 g) in hot-process whitefish steaks vacuum-packaged in O2-permeable (7195 cc/ m2/24 h), or O2-impermeable (108 cc/ m2/24 h) films (adapted from Eklund 1992)

Storage time,

d at 25 °C

Water phase salt (%) Quantity of C. botulinum toxin formed (MLD50/g)
    Oxygen-permeable film Oxygen-impermeable film
3 1.8 50 2500
  2.6 0 500
  3.4 0 5
5 1.8 250 50000
  2.8 0 2500
  3.5 0 0
7 1.8 2500 50000
  2.8 0 5000
  3.5 0 50

In addition, under temperature-abuse conditions, toxin formation can occur in VAC- or MAP-products before or at the same time as spoilage (Eklund 1992). Although the studies were conducted at 25 °C (77 °F), Eklund reported that concentrations of NaCl, the combination of NaCl and nitrite, or other preservatives needed to inhibit C. botulinum were similar for growth and toxin formation at 25 °C (77 °F) as at 10 °C (50 °F). Toxin formation, however, was dramaticall compared with 25 °C. We should emphasize, though, that these studies were performed on hot-smoked fish, which are different microbiologically from cold-smoked fish.

Dufresne and others (2000) recently completed studies on the effect of packaging film permeability and storage temperatures on C. botulinum type E growth and toxin formation in cold-smoked and hot-smoked trout products. Summaries of her findings on cold-smoked fish are contained in Tables VI-3-5.

 

 

Table VI-3. Challenge studies with Clostridium botulinum

type E spores (102/g) on cold-smoked trout fillets with 1.7% NaCl (WPS) packaged in films of different oxygen transmission rates and stored at 4 °C (Dufresne and others 2000)

Oxygen transmission rate

cc/m2/d/atm@ 24 °C, 0% R.H.

Sensory shelf life1 based on odor (d) No. of toxic samples at each sampling interval (d) 2
    7 14 21 28
12 ~28 0 0 0 0
2,950 ~28 0 0 0 0
4,920 ~22 0 0 0 0
10,040 ~18 0 0 0 0

1Time (days) to reach a score of 3.5 on a hedonic scale of 1 to 7. (7=Extremely desirable, 1=Extremely undesirable)

2Trypsinized extract (in duplicate)

 

 

Table VI-4. Challenge studies with Clostridium botulinum

type E spores (102/g) on cold-smoked trout fillets with 1.7% NaCl (WPS), packaged in films of different oxygen transmission rates and stored at 8 °C (Dufresne and others 2000)

Oxygen transmission rate

cc/m2 /d/atm @ 24 °C, 0% R.H.

Sensory shelf life1 based on odor (d) No. of toxic samples at each sampling interval (d) 2
    7 14 21 28
12 ~14 0 0 0 0
2,950 ~13 0 0 0 0
4,920 ~16 0 0 0 1
10,040 ~ 6 0 0 0 1

1Time (d) to reach a score of 3.5 on a hedonic scale of 1 to 7. (7=Extremely desirable, 1=Extremely undesirable)

2Trypsinized extract (in duplicate)

 

 

Table VI-5. Challenge studies with Clostridium botulinum

type E spores (102/g) on cold-smoked trout fillets with 1.7% NaCl (WPS), packaged in films of different oxygen transmission rates and stored at 12 °C (Dufresne and others 2000)

Oxygen transmission rate

cc/m2 /d/atm @ 24°C, 0% R.H.

Sensory shelf life1 based on odor (d) No. of toxic samples at each sampling interval (d) 2
    7 14 21 28
12 ~11 0 1 2 1
2,950 ~12 0 1 2 2
4,920 ~11 0 1 2 2
10,040 ~ 6 0 1 2 2

1Time (d) to reach a score of 3.5 on a hedonic scale of 1 to 7. (7=Extremely desirable, 1=Extremely undesirable)

2Trypsinized extract (in duplicate)

In Dufresne's studies, product shelf life for cold-smoked trout ranged from 18-28 d at storage temperatures of 4 °C (39 °F) with no toxin formed in any sample (Table VI-3). At 8 °C (46 °F) (Table VI-4), cold-smoked trout packaged in films with oxygen transmission rate (OTR) of 10,040 cc/m2/24 h spoiled at 6 d of storage, and toxin formation occurred at 28 d of storage. Spoilage occurred at 16 d, and toxigenisis occurred by 28 d in cold-smoked trout packaged in films 4,920cc/m2/24 h (Dufresne and others 2000). Spoilage of products packaged in films with OTRs of 12 and 2,950 cc/m2/24 h occurred at 13-14 d, but there was no toxin formation by 28 d of storage. It should be emphasized that packages with high OTRs were toxic after 28 d, whereas packages with low OTRs were not toxic.

At 12 °C (54 °F) (Table VI-5), spoilage in smoked trout packaged in film with OTRs of 12, 2,950, and 4,920 cc/m2/24 h occurred between 11 and 12 d with toxin formation by 14 d of storage. Cold-smoked trout packaged in films with an OTR of 10,040 cc/m2/24 h spoiled at 6 d of storage, and toxin formation occurred by 14 d of storage (Dufresne and others 2000). From Dufresne's data it would seem that at abuse temperatures (8-12 °C) spoilage may proceed toxin pro particularly when packaged in high O2 permeable films. Spoilage may help prevent the consumption of toxic fish, but it cannot be relied upon solely as a control to prevent foodborne botulism.

These studies illustrate the need for additional research to identify the processing conditions and characteristics that must be present for a product to be considered air-packaged. Results from these studies suggest that film with OTRs of 7,195 cc/m2/24 h or greater could be considered air-packaged. Additional research on O2 and CO2 film permeability characteristics is needed to better define these parameters. Nevertheless, storage of smoked trout at 4 °C 8 °C or less resulted in no toxin formation prior to spoilage in the cold-smoked product, regardless of film OTRs (Dufresne and others 2000; Eklund 1984, 1992).

 

Listeria monocytogenes is another pathogen of concern in cold-smoked fish. Studies indicate that L. monocytogenes grows well on the finished cold-smoked product at refrigerated temperatures (Farber 1991). Peterson and others (1993) reported that vacuum-packaging initially suppressed growth of L. monocytogenes by 10-100 fold in samples with 3% or 5% WPS. Neither 3% nor 5% water-phase salt by itself, however, sufficed to prevent the growth of L. monocytogenes in vacuum-packaged or O2-permeable film-packaged, cold-processed salmon during prolonged storage at 5 °C (41 °F) or 10 °C (50 °F). The authors suggested that in addition to NaCl, other inhibitors such as smoke and sodium nitrite (NaNO2) could be used to inhibit L. monocytogenes in cold-smoked fish. Pelroy and others (1994b) reported that NaNO2 enhanced the inhibitory effect of NaCl on L. monocytogenes at refrigeration tempera the inoculum was low and the storage temperature was 5 °C. This inhibitory effect decreased as the storage temperature increased to 10 °C and the inoculum level increased. These results emphasized the importance of reducing or eliminating L. monocytogenes and adequate refrigeration during all stages of storage or cold-smoked fish products.

 

 

11.1.2. U.S. packaging and labeling requirements

The botulism outbreaks from hot-smoked fish during the 1960s were caused by a combination of inadequate processing of the products, WPS concentration in most cases less than 1%, and gross temperature-abuse during distribution (Eklund 1992). It is interesting to note that although WPS guidelines are part of state regulations in New York, Michigan, and Minnesota, current U.S. HACCP guidelines do not recommend specific WPS concentrations for cold-smoked fish.

Current AFDO guidelines (followed by New York, Michigan, and Minnesota) recommend that cold-smoked products contain a WPS level of at least 2.5% for air-packaged fish; or a minimum WPS level of 3.5% for vacuum- or modified atmosphere-packaged fish; or a combination of at least 3% WPS and a nitrite level of 100-200 ppm. Each container of processed fish must contain identification indicating where the product was packaged, the year and day packed, and the period during which the product was packaged. Packing codes shall be changed with sufficient frequency to allow identification of lots during sale and distribution. The label should also state the need for refrigerated storage (AFDO 1991). As mentioned in the scope section of this chapter, although developed by experts and widely used, these recommendations have not necessarily been scientifically evaluated. Adequate refrigeration is the most important factor for the safe distribution of smoked fish products and has been recommended in multiple occasions (Eklund 1992; Eklund and others 1982; Eklund and others 1988; Pelroy and others 1982).

 

11.1.3. Canadian packaging and labeling requirements

The Canadian document "Food and Drugs Act and Regulations" states under Division 21 (B.21.025.) that no person shall sell marine and fresh water animals, or marine and fresh water animal products, that are packed in a container that has been sealed to exclude air and that are smoked or to which liquid smoke or flavour or liquid smoke flavour concentrate has been added, unless (a) the container has been heat-processed after sealing at a temperature and for a time sufficient to destroy all spores of the species Clostridium botulinum; (b) the contents of the container contain not less than nine percent salt, as determined by official method PO-38, Determination of Salt in Smoked Fish, dated March 15, 1985; (c) the contents of the container are customarily cooked before eating; or (d) the contents of the container are frozen and the principle display panel of the label of the container carries the statement "Keep Frozen Prior to Use" in the same size types used for the common name of the contents of the container (Health Canada 1994).

The Canadian government, however, allows refrigerated storage of smoked products if they are packaged in containers with an oxygen permeability equal to or greater than 2,000-cc/m2/24 h at 24 °C at 1 atm. These products must also be stored at 4 °C or less and should have a label that states shelf life must not exceed 14 d from the date of packaging. Processors and retailers should record the type of film used and its permeability characteristics (Health Canada 1994).

 

11.2. Potential hazards

Hazards that may arise from the time of final product manufacture to the time of consumption within the product packaging include growth of pathogens such as L. monocytogenes or C. botulinum and production of botulin toxin on the finished product. In addition, biogenic amine formation is possible if scombrotoxin-susceptible species are used for the cold-smoking process.

 

 

11.3. Control point

The packaging and labeling step is not considered a control point, since it is not possible to label safety into a product. Nevertheless, label information identifying appropriate storage temperatures and time for safety is critical to control biogenic amine formation in scombrotoxin-susceptible species, as well as C. botulinum growth and toxin formation in cold-smoked products, especially if packaged in a reduced-oxygen environment. This label information is also important in reducing the growth rate of L. monocytogenes, although temperature will not prevent its growth. A warning label for populations at high risk of developing listeriosis may be considered for cold-smoked fish and other ready-to eat foods in the risk category. Such a label could indicate that these products may constitute a health hazard to immunocompromised individuals and pregnant women.

 

11.4. Processing parameters

All finished products should be labeled to advise on refrigeration temperatures and storage time. This panel concluded that, with the appropriate salt concentrations (that is, 3.5% WPS), the product should be refrigerated at 40 °F (4.4 °C) or less for no more than 5 wk (for a detailed discussion on this subject, see the Chapter III). This would mean the final product needs to maintain that temperature during storage, distribution, retail, and home storage and that the recommended time from manufacture to consumption is 5 wk. Frozen product must also be labeled to indicate that the "product shall remain frozen until it is thawed at refrigeration temperatures."

12. Storage and distribution

12.1. Background and support information

Cold-smoked fish products can be stored refrigerated or frozen. Companies may freeze and hold the product in frozen storage for 1-2 weeks prior to distribution and shipment. This is done primarily for inventory control. Other companies simply refrigerate products and distribute and ship them as soon as possible. In addition, cold smoked products are also distributed via overnight carriers and government postal services.

Kalish (1991) conducted audits of temperature readings at more than 50 major warehouses and distribution centers. She reported that most warehouses and distribution centers maintained refrigeration temperatures within the recommended temperature range (0 - 3.3 °C, 32 - 38 °F), although a few (number not specified) were as high as 10 °C (50 °F). The rotation of product in warehouses and distribution centers was good (Kalish 1991).

 

12.2. Potential hazards

Potential hazards during refrigerated storage and distribution include pathogens such as L. monocytogenes or C. botulinum and botulin toxin formation on the finished product. In addition, biogenic amine formation is possible if scombrotoxin forming species are used for the cold-smoking process. Frozen storage is not considered a control point, since pathogens will not grow and biogenic amine will not form during storage.

 

12.3. Control point

Both storage and distribution are control points for biogenic amine formation in both aerobic-packaged and ROP cold-smoked scombrotoxin susceptible species. Storage and distribution are also control point for C. botulinum growth and toxin formation in cold-smoked products. Although chill temperature during storage and distribution will also reduce the growth rate of L. monocytogenes, it will not prevent its growth.

 

12.4. Processing parameters

Cold-smoked fish should be stored and distributed at storage temperatures of less than 40 °F (4.4 °C) or frozen. Studies indicate that storage at 4 °C (39 °F) or less resulted in no botulin toxin formation prior to spoilage in cold-smoked products at salt levels of 1.7% WPS (Dufresne and others 2000) (for a more detailed discussion see Chapter III). Those temperatures will also reduce the growth of scombrotoxin forming species and of L. monocytogenes. Frozen distribution is not considered a control point, as pathogens will not grow and biogenic amines will not form during frozen storage.

13 Retail

13.1. Background and support information

The Food Code provides criteria that must be met by the HACCP plans of operators that handle ROP products (FDA 1999). (Note: the Food Code is a voluntary recommendation from the FDA and is not codified in all states). The Food Code prohibits products from being packaged in ROP at the retail level and requires maintenance of adequate refrigeration during the entire shelf life of the product. Nevertheless, temperature-abuse does occur during retail storage. Studies have shown that at the retail level, product rotation procedures were inadequate, as sales of product dictated product rotation frequency (Kalish 1991).

Approximately 2,000 retail stores, including back-room storage facilities and chill cases, were checked. Kalish reported that only 37% of refrigerated foods were stored within the 32-38 °F (0-3.3 °C) range. Products were also found stacked on the floor without any refrigeration, and the temperature of many refrigerated cases was 44 °F (6.7 °C) with some as high as 56 °F (13.3 °C) (Kalish 1991).

 

13.2. Potential hazards

Potential hazards during all retail operations include pathogens such as L. monocytogenes or C. botulinum on the finished product. In addition, biogenic amine formation is possible if scombrotoxin-susceptible species are used for the cold-smoking process.

 

13.3. Control point

Temperature and time control at all retail operations are control points to control biogenic amine formation in both air-packaged and ROP cold-smoked scombrotoxin-susceptible species and to control C. botulinum growth and toxin formation in cold-smoked products. Low temperature will also reduce the rate of growth of L. monocytogenes, although it will not prevent its growth. If the product is distributed frozen to the retailer, frozen storage is not considered a control point, as pathogens will not grow and biogenic amine will not form during storage. Thawing of the product, however, should be conducted at refrigeration temperatures.

 

13.4. Processing parameters

Cold-smoked fish should be stored, handled, prepared, and displayed at temperatures <40 °F (4.4 °C) or frozen. Studies indicate that storage at 4 °C or less resulted in no botulin toxin formation prior to spoilage in cold-smoked products at salt levels of 1.7% WPS (Dufresne and others 2000) (for a more detail discussion see Chapter III). Those temperatures will also reduce the growth of scombrotoxin-forming species and L. monocytogenes.

If the product is received frozen, it should kept stored and displayed frozen or thawed under refrigerated temperatures and handled as indicated in the previous paragraph.

14. Consumer

14.1. Background and support information

Beard (1991) reported that consumers should be educated about the potential hazards associated with food products and provided information on proper product rotation. Scientists have recommended that better and more effective consumer education programs are needed to reduce the incidence of foodborne disease outbreaks (Garrett 1987; Sachs 1989; NMFS 1991b). Consumers should be educated about potential risks associated with home preparation aspects of food consumption (NMFS 1991a).

For each of the years from 1993 through 1997, the most commonly reported food preparation practice that contributed to foodborne disease was improper holding temperature; the second most commonly reported practice was inadequate cooking of food (MMWR 2000).

Beard (1991) reported that out of 14 home refrigerators and 11 freezers, only 7 refrigerators and 1 freezer had thermometers. Refrigerator temperatures ranged from 0-13 °C (32-55 °F). The panel recommends labeling frozen products with thawing instructions and storing the thawed product below 40 °F (4.4 : C).

 

14.2. Potential hazards

Potential hazards at its consumer stage include pathogens such as L. monocytogenes or C. botulinum, or botulin toxin formation on the finished product. In addition, biogenic amine formation is possible if scombrotoxin susceptible species are used for the cold-smoking process.

 

14.3. Control point

The refrigerator of the consumer is a control point to control biogenic amine formation in both air-packaged and ROP cold-smoked scombrotoxin-susceptible species and to control C. botulinum growth and toxin formation in cold-smoked products. Low temperature will also reduce the rate of growth of L. monocytogenes, although it will not prevent its growth.

 

14.4. Processing parameters

Cold-smoked fish should be stored at storage temperatures of <40 °F (4.4 °C). Studies indicate that storage at 4 °C or less resulted in no botulin toxin formation prior to spoilage in cold-smoked products at salt levels of 1.7% WPS (Dufresne and others 2000) (for a more detailed discussion see section 11 of this chapter and Chapter III). Those temperatures will also reduce the growth of scombrotoxin-forming species and L. monocytogenes.

 

References

[AFDO] Association of Food and Drug Officials. 1991 June. Cured, salted, and smoked fish establishments good manufacturing practices [model code]. York (PA): Association of Food and Drug Officials. 7 p.

Angot V, Brasseur P. 1993. European farmed Atlantic salmon (Salmo salar L.) are safe from Anisakid larvae. Aquaculture 118:339-44.

Anonymous. 1995. International forum supports above-zero tolerance for Listeria in low risk foods. World Food Chem News July 26:15-7.

Beard TD. 1991. HACCP and the home: the need for consumer education. Food Technol 45(4):123-4.

Ben Embarek PK, Huss HH. 1992. Growth of Listeria monocytogenes in lightly preserved fish products. In: Huss HH, Jakobsen M, Liston J, editors. Quality assurance in the fish industry. Amsterdam: Elsevier. p 293-303.

Corby JJ. 1991. Circular 102 Rules and regulations relating to fish processing and smoking establishments pursuant to Article 17 of the Agriculture and Markets Law. Albany, NY: New York State Department of Agriculture and Markets, Division of Food Inspection Services. Part 262 of Title 1 of the Official Compiliation of Codes, Rules, and Regulations of the state of New York.

Dufresne I, Smith JP, Liu JN, Tarte I, Blanchfield B, Austin JW. 2000. Effect of films of different oxygen transmission rate on toxin production by Clostridium botulinum type E in vacuum packaged cold and hot smoked trout fillets. J Food Saf 20:251-68.

Eklund M. 1984. Effect of CO2 modified atmospheres and vacuum packaging on Clostridium botulinum and spoilage organisms of fishery products. Published in: Proceedings of First National Conference on Seafood Packaging and Shipping; 1982 Nov 15-17 [Washington, DC] and 1982 Dec 7-9 [Seattle, WA]. p 298-331.

 

Eklund M. 1989. Comments and research data for the proposed establishment of standards for the manufacture, packaging, and labeling of processed fish including smoked fish [testimony to the New York Department of Agriculture and Markets]. [New York]: Northwest Fisheries Center, Utilization Research Division.

Eklund MW. 1992. Control in fishery products. In: Hauschild AHW, Dodds KL, editors. Clostridium botulinum: Ecology and control in foods. New York: M Dekker. p 209-32.

Eklund MW, Pelroy GA, Paranjpye R, Peterson ME, Teeny FM. 1982. Inhibition of Clostridium botulinum types A and E toxin production by liquid smoke and NaCl in hot-process smoked-flavored fish. J Food Prot 45(10):935-41.

Eklund M, Pelroy G, Poysky F, Paranjpye R, Lashbrook L, Peterson M. 1993 July. Summary of interim guidelines for reduction and control of Listeria monocytogenes in or on smoked fish [internal report]. Seattle: Northwest Fisheries Science Center. July 1993. 14 p.

Eklund MW, Peterson ME, Paranjpye R, Pelroy GA. 1988. Feasibility of a heat-pasteurization process for the inactivation of nonproteolytic Clostridium botulinum types B and E in vacuum-packaged, hot-process (smoked) fish. J Food Prot 51(9):720-6.

Eklund MW, Poysky FT, Paranjpye RN, Lashbrook LC, Peterson ME, Pelroy GA. 1995. Incidence and sources of Listeria monocytogenes in cold-smoked fishery products and processing plants. J Food Prot 58(5):502-8.

[FAO] Food and Agriculture Organization. 1999 May. Report of the FAO expert consultation on the trade impact of Listeria in fish products. Rome: FAO. FAO Fisheries Report nr 604. 34 p.

Farber JM. 1991. Listeria monocytogenes in fish products. J Food Prot 54(12):922-4, 934.

[FDA] Food and Drug Administration. 1998. Fish & Fisheries Products Hazards & Controls Guide. 2nd ed. Washington, D.C.: FDA, Office of Seafood. 276 p.

[FDA] Food and Drug Administration. 1999. Food Code. Washington, DC: U.S. Department of Health and Human Services, Public Health Service, Food and Drug Administration.

Fonnesbech Vogel B, Ojeniyi B, Ahrens P, Due Skov L, Huss HH, Gram L. 2001. Elucidation of Listeria monocytogenes contamination routes in cold-smoked salmon processing plants detected by DNA-based typing methods. Appl Enviro Microbiol. Forthcoming.

Garrett ES. 1987. Testimony before the committee on Agriculture, Nutrition, and Forestry. Washington, DC: United States Senate. June 11, 1987. Report nr S1813.

Gilbert J, Knowles M. 1975. The chemistry of smoked foods: a review. J Food Technol 10:245-61.

Gram L, Huss HH. 2000. Fresh and processed fish and shellfish. In: Lund BM, Baird-Parker TC, Gould GW, editors. The microbiological safety and quality of food. Gaithersburg (MD): Aspen. p 472-506.

Health Canada. 1994 Aug 16 [updated 1997 Mar 19]. Food and drugs act and regulations: division 21 - prepared fish (B.21.025) [online publication - not the official Canada Gazette pages]. Available from: Health Canada's Food Directorate web site at http://www.hc-sc.gc.ca/food-aliment/english/publications/acts_and_regulations/food_and_drugs_acts/d-text-2.pdf

Hilderbrand K. 1973. Preparation of salt brines for the fishing industry. Corvallis: Oregon State Univ., Oregon Sea Grant Program (#OSU SG 22). Report nr NSGL# ORESU - G - 73 - 002. Grant nr NOAA - 73072505. 4 p.

Hudson JA, Mott SJ. 1993. Growth of Listeria monocytogenes, Aeromonas hydrophila and Yersinia enterocolitica on cold-smoked salmon under refrigeration and mild temperature abuse. Food Microbiol 10:61-8.

Kalish F. 1991. Extending the HACCP concept to product distribution. Food Technol 45(4):119-20.

Kaneko J, (PacMar, Inc., Honolulu, Hawaii). 2000. Development of a HACCP-based strategy for the control of histamine for the fresh tuna industry [A report by PacMar, Inc. pursuant to National Oceanographic and Atmospheric Administration]. Honolulu (Hawaii): PacMar; 2000 July 31. NOAA Award No. NA86FD0067. 48 p.

Kassem CL. 1977. Smoking fish at home - a step by step guide. Blacksburg (VA): Virginia Polytechnic Institute and State Univ, Cooperative Extension Service. VPI-SG-300-2.

Leisner JJ, Millan JC, Huss HH, Larson LM. 1994. Production of histamine and tyramine by lactic acid bacteria isolated from vacuum-packed sugar-salted fish. J Appl Bacteriol 76:417-23.

Lord JB. 2000. Entering the new millennium: the food industry in transition. NFPA Feb:9-15.

[MMWR] Morbidity Mortality Weekly Report. 2000 Mar 17. Surveillance for foodborne disease outbreaks - United States, 1993-1997. MMWR 49(SS01):1-51.

[NMFS] National Marine Fisheries Service. 1991a. HACCP prototype model food service/consumer education. Model seafood surveillance project. Pascagoula (MS): NMFS, Office of Trade and Industry Services, NSIL. 1991 Dec.

[NMFS] National Marine Fisheries Service. 1991b. HACCP regulatory model smoked and cured fish. Model seafood surveillance project. Pascagoula (MS): NMFS, Office of Trade and Industry Services, NSIL. 1991 Dec.

Pelroy GA, Eklund MW, Paranjpye RN, Suzuki EM, Peterson ME. 1982. Inhibition of Clostridium botulinum types A and E toxin formation by sodium nitrite and sodium chloride in hot-process (smoked) salmon. J Food Prot 45(9):833-41.

Pelroy GA, Peterson ME, Paranjpye R, Almond J, Eklund M. 1994b. Inhibition of Listeria monocytogenes in cold-process (smoked) salmon by sodium nitrite and packaging method. J Food Prot 57(2):114-9.

Peterson ME, Pelroy GA, Paranjpye RN, Poysky FT, Almond JS, Eklund MW. 1993. Parameters for control of Listeria monocytogenes in smoked fishery products: sodium chloride and packaging method. J Food Prot 56(11):938-43.

Robertson GL. 1993. Food Packaging: Principles and Practice. Hughes H, editor. New York: M Dekker. 676 p.

Rorvik LM, Yndestad M, Skjerve E. 1991. Growth of Listeria monocytogenes in vacuum-packed, smoked salmon during storage at 4° C. Int J Food Microbiol 14:111-8.

Sachs S. 1989. Forward to a margin of safety: the HACCP approach to food safety education. Washington, DC: USDA, FSIS, ILA. 1989 June.

Storey RM. 1982. Smoking. In: Aitken A, Mackie IM, Merritt JH, Windsor ML, editors. Fish handling and processing. 2nd ed. Aberdeen: Ministry of Agriculture, Fisheries and Food, Torry Research Station; Edinburgh: HMSO. p 98-114.