Application of shelf life decision system (SLDS) to marine - CiteSeerX

Application of shelf life decision system (SLDS) to marine - CiteSeerX

International Journal of Food Microbiology 73 (2002) 375 – 382 Application of shelf life decision system (SLDS) t...

142KB Sizes 0 Downloads 6 Views

Recommend Documents

Shelf life of metformin -
Nov 21, 2012 - Tablets - Summary of Product Characteristics (SPC) by Boehringer Ingelheim Limited. Metformin Anti Aging

Shelf Life - Esseco USA
Jan 1, 2016 - Assumes storage in original, unopened container. SHELF LIFE FROM. (or suitable bulk storage) in controlled

Shelf Life - Parker Hannifin
Jan 17, 2013 - basis upon which companies can base their own shelf life requirements. At the O-Ring Division, Parker has

Shelf Life - Masterton Library
Andy Miller wrote in The Guardian of the traditional pleasures of reading being more complex than just enjoyment. They i

Shelf Life of Food Bank Products - Harvesters
Shelf Life of Food Bank Products. Greater Pittsburgh Community Food Bank often distributes food items after the date on

A Proposal of Clinical Decision Support system - CiteSeerX
Clinical Decision Support Systems (CDSS); Electronic health record; Health ... system. The proposed model will take an o

Pineapple Vinegar to Enhance Shelf Life of Carrot and - VTechWorks
Jun 4, 2015 - Pineapple Vinegar to Enhance Shelf Life of Carrot and Mango in Tanzania. Aldegunda Sylvester Matunda. ABST

Application of Membrane Technology to Slaughterhouse - CiteSeerX
Application of Membrane Technology to Slaughterhouse Blood to. Produce Edible Powdered Protein Mixture. Maria I. Kokkora

Shelf Life - SUrface - Syracuse University
THE PRESENCE OF GRACE By Daniel R. Surdam '90. Set in Skaneateles, New York, Surdam's first detective ..... Moore, R. Gi

International Journal of Food Microbiology 73 (2002) 375 – 382

Application of shelf life decision system (SLDS) to marine cultured fish quality K. Koutsoumanis a,*, M.C. Giannakourou b, P.S. Taoukis b, G.J.E. Nychas a a

Laboratory of Microbiology and Biotechnology of Foods, Department of Food Science and Technology, Agricultural University of Athens, Iera Odos 75, Athens 11855, Greece b Laboratory of Food Chemistry and Technology, Department of Chemical Engineering, National Technical University of Athens, 5 Iroon Polytechniou, 15780 Zografou, Greece Received 16 May 2001; accepted 9 August 2001

Abstract Growth of natural microflora of marine cultured, air-packed, sea bass (Dichentrachus labrax) was studied at isothermal conditions in the 0 – 15 C range and kinetically modelled using the four-parameter Logistic equation. Sensory shelf life was correlated to pseudomonad population and sensory acceptability was correlated to a pseudomonad level, NS, of 107. The variability of their initial population was quantitatively shown and a conductance-based rapid method specific to sea bass pseudomonad enumeration was established as a practical means of N0 determination, required in shelf life predictions. Kinetic models, shelf life correlations and N0 data were incorporated into the shelf life decision system (SLDS) shown to be an effective tool for marine cultured sea bass chill chain management leading to optimization of quality of the fish at consumer’s end. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Sea bass; Shelf life; Chill chain

1. Introduction Spoilage of chilled fresh and minimally processed fish is attributed mainly to bacterial activity (Gram and Huss, 1996). The fraction of fish microflora association that is responsible for spoilage depends on intrinsic and extrinsic factors. For example, pseudomonads are the specific spoilage organisms (SSO)


Corresponding author. Tel./fax: +30-1-5294693. E-mail addresses: [email protected] (P.S. Taoukis), [email protected] (G.J.E. Nychas).

of Mediterranean boque (Boops boops) and tsipoura (Sparus aurata) fish stored aerobically (Koutsoumanis and Nychas, 1999), while spoilage of cod fish from Northern European seawater, stored under modified atmosphere packaging conditions, is due to the metabolic activity of Photobacterium phosphoreum (Dalgaard et al., 1993). Establishment of the SSO and their spoilage domain, that is, the range of conditions at which they dominate the spoilage phenomenon, is essential for a quantitative approach to fish quality and shelf life estimation. Modelling the growth of SSO as a function of storage conditions, mainly temperature, provides the means for reliable shelf life prediction. Besides validated growth models, accurate

0168-1605/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 0 5 ( 0 1 ) 0 0 6 5 9 - 6


K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382

correlation of the limit of organoleptic acceptability to the microbial spoilage level of SSO, NS, and a reliable estimate of the initial population, N0, (Koutsoumanis and Nychas, 2000) is required. These data and information on the temperature history, that can be provided down to the fish product unit from the response of suitable time temperature integrators (TTI), allow determination of spent and remaining shelf life at any point of the chill chain (Taoukis et al., 1999b). In most processed food products, the working assumption is that ‘‘zero time’’ parameters, including a target range of initial microbial population, can be set by proper design and control of the processing conditions. On the other hand, initial microflora in fresh fish can vary significantly, depending on a number of environmental factors such as sea water temperature, handling and processing after catch (Huss, 1995). Thus, mathematical models have to be combined with rapid methods of microbial enumeration, in order for any predictive procedure to account for that variability. Among the microbiological methods that can be used to detect microbial counts within a relatively short time, impedance technique is one of the most promising. Since a number of impedance instruments are available (Bolton and Gibson, 1995), conductance measurements have been used for the determination of microbial count in different products (Gibson, 1985; Gibson and Hobbs, 1987; Salvat et al., 1997) by correlating the time to detection (TTD) with viable count numbers (Gibson et al., 1984). The development of a conductance-based rapid method specific to fish SSO enumeration allows in the present work the integration of initial SSO load information, kinetic spoilage models and TTI-based temperature monitoring in the application of the intelligent shelf life decision system (SLDS) to the distribution of marine cultured fish such as sea bass. SLDS is an integrated tool developed to allow quality-based decision making in the food chill chain applied to optimize fish quality at consumer’s end (Giannakourou et al., 2001). With the use of SLDS, decisions with regards to the destination, the storage conditions and the stock rotation of the food products can be taken at critical points of the chill chain based on the actual quality status of each product unit according to its initial microbiological quality and its time – temperature history. In this study, growth models for SSO of sea bass were developed, N0 distribution was determined based on a large number of

initial measurements, and TTI of appropriate response characteristics were selected. The above formed the basis of SLDS which was applied to demonstrate the potential for optimization of quality of marine cultured fish at consumer’s end.

2. Materials and methods Marine cultured sea bass (Dichentrachus labrax), a Mediterannean fish of high consumption and commercial interest in Greece, was studied. Two replicated storage experiments were carried out with ungutted fresh sea bass transported to the laboratory in ice within 6 h after catch. The fish were stored, in individual pouches (not sealed), at controlled isothermal conditions (0, 5, 10 and 15 C) in high-precision low-temperature incubators (Sanyo MIR 153, Sanyo Electric, Ora-Gun, Gunma, Japan). Samples were taken at appropriate time intervals to allow for an efficient kinetic analysis of sensory quality and microbial growth. A survey for the initial pseudomonad population in different fish species was also carried out. Samples of six fresh fish species, gilthead sea bream (S. aurata), sea bass (D. labrax), cod (Gadus capelanus), picarel (Smaris alcedo), trout (Salmon truta) and anchovy (Engraulis encrasicholus) were bought from the central market of Athens and transported to the laboratory in ice within 1 h. 2.1. Sample preparation Fish (25 g) was transferred to a stomacher bag (Seward, London, UK), 225 ml of 0.1% peptone water with salt (NaCl, 0.85%, w/v) was added and homogenized for 60 s with a stomacher (Lab Blender 400, Seward Medical, London, UK). 2.2. Microbiological media and enumeration Samples (0.1 ml) of 10-fold serial dilutions of either treated (inoculated with the isolates) or naturally contaminated fish homogenates were spread on the surface of the appropriate media in Petri dishes for enumeration of the following. (i) Total aerobic viable count on a modified Long and Hammer’s agar (mLHA) (Van Spreekens, 1974) and incubated at

K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382

10 C for 7 days. The medium was composed of (g l  1 distilled water): proteose peptone (Sigma, P 0431), 20; gelatine (Merck, 4070), 40; K2HPO4, 1; NaCl, 10; agar (Oxoid, L11), 15; ammonium ferric citrate, 0.25. (ii) Pseudomonads on cetrimide fusidin cephaloridine agar (CFC, Oxoid code CM 559, supplemented with SR 103) and incubated at 20 C for 2 days (Mead and Adams, 1977). (iii) Yeasts on rose bengal chloramphenicol agar, incubated at 25 C for 7 days and wrapped in aluminium foil to avoid any adverse effects of light. For Enterobacteriaceae and hydrogen sulphide-producing bacteria, 1.0 ml was inoculated into 10 ml of molten (45 C) violet red bile glucose agar (VRBGA, Oxoid code CM 485) and iron agar (IA, Oxoid code CM 867), respectively. After setting, a 10-ml overlay of molten medium was added. For the former, incubation was at 30 C for 24 h. The large colonies with purple haloes were counted (Mossel et al., 1979). Iron agar plates were incubated at 20 C for 4 days (Gram et al., 1987). Black colonies formed by the production of H2S were enumerated after 2– 3 days (Gennari and Campanini, 1991). Three replicates of at least three appropriate dilutions (Anonymous, 1978) were enumerated. All plates were examined visually for typical colony types and morphological characteristics associated with each growth medium. In addition, selectivity of each medium was checked routinely by Gram-staining and microscopic examination of smears prepared from randomly selected colonies from all media. 2.3. Sensory evaluation of fish quality Whole fish was examined by a trained sensory panel of six persons who evaluated the odour of raw fish and the taste and odour of cooked fish. Fish were scaled, gutted and gilled before cooking. Fish were cooked whole, individually wrapped steam tightly in aluminum foil, at 180 C for 30 min. An adaptation of a simple three-point scoring system (Dalgaard et al., 1993; Taoukis et al., 1999a,b) was used. Taste and odor were judged and recorded in appropriate forms with descriptive terms reflecting the organoleptic evaluation of quality deterioration. Rating was assigned on a continuous 0 – 3 hedonic scale (0 being the highest quality score and 2 the limit of acceptance).


2.4. Impedance measurements for the enumeration of pseudomonads The rapid automated bacterial impedance technique (RABIT) from Don Whitley Scientific (Shipley, UK) was used for the impedance measurements. A test lasted 48 h with a measurement every 6 min; the TTD signal appeared when there were three consecutive measurements of 5 mS minimum. Metronidazole, carbenicilline, cetrimide, cycloheximide, diamide (MCCCD) (Salvat et al., 1997) was used as conductance medium. Fifty-four fish samples of different microbiological quality were tested. Fish (25 g) was transferred to a stomacher bag (Seward), 225 ml of 0.1% peptone water with salt (NaCl, 0.85%, w/v) was added and homogenized for 60 s with a stomacher (Lab Blender 400, Seward Medical). A volume (500 ml) of this dilution was used to inoculate the impedance tube and 100 ml was used to inoculate CFC agar (Mead and Adams, 1977). Each colony growing on CFC agar was confirmed as pseudomonads by an oxidase test in which a solution of 1% N,N,NV,NV-tetramethylphenylene-1,4-diamine dichlorohydrate, mixed in equal part with 1% a-naphydratethol alcoholic solution, was poured onto the surface of the agar. The oxidasepositive colonies appeared dark blue. 2.5. Growth model development The growth data from the enumeration of different groups of microbial association were modelled as a function of time to estimate the maximum specific growth rate and the Lag phase. The log-transformed form of the four-parameter Logistic model was used (Dalgaard, 1995):  logN ðtÞ ¼ log Nmin þ

Nmax  Nmin 1 þ exp½lmax ðt  ti Þ


In Eq. (1), t is the time (h), N(t) is the number of microorganisms at time t (cfu/g), Nmin and Nmax are the minimum and maximum asymptotic cell concentration (cfu/g), lmax is the maximum specific growth rate (h  1) and ti is the time (h) when half of the maximum cell concentration is reached. The duration


K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382

of the lag phase (Lag) was calculated as described by Dalgaard (1995). The obtained estimates of lmax (h  1) and Lag phase (h) were further expressed as a function of temperature using the Arrhenius model:    EAl 1 1  lnðlmax Þ ¼ lnðlref Þ  ð2Þ T Tref R

1 ln Lag

EALag ¼ lnðLagref Þ  R

1 1  T Tref


where T (K) is the absolute temperature; Tref is the reference temperature (273 K); lref (h  1) and Lagref are the maximum specific growth rate and Lag phase, respectively, at reference storage conditions (Tref); EAl and EALag (kJ/mol) are the activation energies referring to lmax and Lag phase, respectively; and R is the universal gas constant. All data were fitted using nonlinear regression with the Fig. P version 2.5 software (Anonymous, 1995).

3. Results and discussion The experimental data for the growth of the different measured components of the natural microflora of

sea bass are shown in Fig. 1 along with the fitted Logistic growth curves for two representative isothermal conditions. Pseudomonads were the dominant bacteria at all temperatures tested. Growth of pseudomonads followed closely the decrease of sensory quality and end of shelf life coincided with an average level of 107 (Table 1). This is in agreement with results reported for other Mediterranean fish species (Taoukis et al., 1999a,b; Koutsoumanis and Nychas, 2000). The pseudomonad kinetic parameters (lmax, Lag phase and Nmax) are presented in Table 1. The results showed that the storage temperature did not affect the maximum concentration (Nmax) of pseudomonads which was found to be constant with an average value of 8.07 F 0.21 log10 cfu/g (average F S.D.). Further, the temperature dependence of pseudomonad kinetic parameters was modelled using the Arrhenius equation (Fig. 2). The estimated values and statistics of the Arrhenius model parameters are shown in Table 2. The activation energies for lmax and Lag phase were very close (74.0 and 72.1 kJ/mol, respectively). Similar EA values for pseudomonads have been reported in studies on other Mediterranean fish species (Taoukis et al., 1999a,b; Koutsoumanis et al., 2000a,b). Having established quantitative mathematical expressions for the growth of SSO and its dependence from temperature, prediction of remaining shelf life of the fish at any point of the chill chain requires reliable estimates of the initial population of SSO. This can

Fig. 1. Development of the microbial flora of Mediterranean sea bass (D. labrax) stored aerobically at 0 (a) and 10 (b) C (n: total viable count, .: pseudomonads, E: H2S-producing bacteria, : Enterobacteriaceae, 1: yeasts, x: lactic acid bacteria).

K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382


Table 1 Kinetic parameters of pseudomonad growth and shelf life of sea bass (D. labrax) stored aerobically at 0, 5, 10 and 15 C T (C)

N0 (log cfu/g)

Nmax (log cfu/g)

lmax (h  1)

Lag time (h)

Shelf life (h)

NS (log cfu/g)

0 5 10 15

3.30 F 0.14a 3.58 F 0.10 4.24 F 0.08 4.16 F 0.19

8.19 F 0.20 8.18 F 0.08 7.76 F 0.05 8.17 F 0.14

0.0523 F 0.001 0.0939 F 0.007 0.202 F 0.008 0.267 F 0.015

38.6 F 2.4 22.5 F 3.3 10.2 F 0.1 8.02 F 1.0

216.5 F 10.6 107.0 F 4.2 48.5 F 5.0 32.5 F 3.5

7.25 F 0.11 6.98 F 0.20 7.27 F 0.30 6.97 F 0.40

N0: initial cell concentration. Nmax: maximum cell concentration. NS: cell concentration at organoleptical rejection. lmax: maximum growth rate. a Value F S.D.

exhibit a significant variation from batch to batch, depending on extrinsic factors such as sea water temperature of catching, handling and processing after catch. To get a quantitative overview of the range of N0 variability, the initial count of pseudomonad population was surveyed mainly on samples of marine cultured sea bream and sea bass but also on four other species of interest. Tested fish was as close as possible to commercial ‘‘zero time,’’ obtained upon arrival in ice to the retail fish shop, within hours from harvest or catch. The results of the survey performed in the present study are summarized in Table 3 and illustrate the interspecies and intraspecies variation of the initial count of pseudomonads.

Having established the importance of an accurate estimate of N0 of the shelf life determining SSO, in time short enough to allow for use as input to any model-based quantitative shelf life estimates, a specific impedance method enumerating pseudomonads associated with sea bass was explored. It was developed based on a selective conductance medium previously used for the detection of pseudomonads in poultry meat (Salvat et al., 1997). A large number of sea bass samples (54) of different microbiological quality were tested and a satisfactory correlation between TTD and pseudomonad count (log cfu/g) was established (Fig. 3). The parameters and statistics of the regression line equation are shown in Table 4.

Fig. 2. Arrhenius plot for the effect of temperature on the maximum growth rate (a) and Lag time (b) of pseudomonad growth on sea bass (D. labrax) stored under aerobic conditions.


K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382

Table 2 Parameters and statistics of the Arrhenius plot for lmax, Lag time of pseudomonad growth and shelf life of aerobically stored sea bass (D. labrax) Parameter Value

F 95.0% CI



ln(SL) = ln(SLref)+[(  EASL/R)(1/T  1/Tref)]  84.5 11.8 0.981 208.1 35.7

EAl lref

ln(lmax) = ln(lref)+[(  EAl/R)(1/T  1/Tref)] 74.0 12.5 0.972 0.0540 0.009

EALag Lagref

ln(1/Lag) = ln(1/Lagref)+[(  EALag/R)(1/T  1/Tref)] 72.1 15.3 0.956 37.6 8.32

Fig. 3. Regression line between time to detection (TTD) and pseudomonad counts on sea bass.

Using the above technique, samples with 103 cfu/g, a common initial pseudomonad level, were detected in less than 20 h. Having demonstrated the ability to obtain reliable, on time, N0 estimates for fresh sea bass, the next step would be to explore the potential to apply it in improving the management of chill chain distribution of this product. This could be achieved through the application of the integrated SLDS. SLDS is based on growth models of SSO, the determination of the initial SSO population, N0, continuous temperature monitoring in the chill chain with TTI and the correlation of sensory acceptability to an SSO population level, NS. It allows the calculation of the actual remaining shelf life of individual product units (boxes or even single packs) at critical points of the chill chain. Further, based on the distribution of remaining shelf life, product management decisions are made at each point, resulting in the narrowest possible distribution of actual quality at the point of consumption. These

principles were applied in the case of the management of fresh, chilled sea bass, harvested at the marine culture site of a big fish company, packed in 5-kg boxes and distributed to local and export markets. Impendance sampling and application of a TTI unit is assumed for each box. Suitable TTIs should have a temperature sensitivity of response in the 70 kJ/mol range. Such TTIs, for chill fish monitoring, were kinetically tested and reported by Taoukis et al. (1999b). Management points can be shipment points, bulk sale centers and retail store stock rotation. In the studied case, one such point is the central distribution center where product from the same initial shipment is split in half and is forwarded to two different retail markets, a close and a distant one. The split instead of being random according to conventional first in first out (FIFO) practice is based on the actual individual spoilage condition of the product units, that is, prod-

Table 3 The minimum value, maximum value, mean number and standard deviation (log cfu/g) of the initial pseudomonad count in six Mediterranean fish species

Count Mean Min Max S.D.

Sparus aurata (gilthead sea bream)

Dichentrachus labrax (sea bass)

Gadus capelanus (cod)

Smaris alcedo (picarel)

Salmon truta (trout)

Engraulis encrasicholus (anchovy)

32 3.80 2.00 5.66 0.93

24 3.4 2.00 4.3 0.63

10 4.47 3.79 5.47 0.54

10 4.87 3.40 5.78 0.66

10 4.16 2.40 5.47 0.85

10 4.02 2.00 4.87 0.83

K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382 Table 4 Parameters and statistics of the regression line between time to detection (TTD) and pseudomonad count on sea bass (D. labrax) Parameter


log cfu/g = aTTD + b a  0.392 b 10.63

F 95.0% CI


0.032 0.49


ucts with higher load, Nt, are shipped to the close market (B) and consumed sooner and lower Nt products to the distant market (A). This is based on an identity scan of the product and the TTI response. To quantitatively demonstrate the results that can be achieved from the application of the SLDS, for the management of the actual marine cultured sea bass chill chain, a Monte Carlo simulation was applied. Storage temperature conditions used actual data and TTI responses collected during transportation phases of marine cultured fresh fish products (Project FAIRCT96-1090, unpublished data) and at the retail level. Kinetic models and parameters of the SSO pseudomonads developed through this work and representative N0 distribution measured with the described impedance technique were used. All these were included in a FORTRAN 77 programme code (Giannakourou et al., 2001) and 2000 runs of chill chain conditions were generated, based on the distributions of temperature and N0. The results, condensed in Fig. 4, show the probability for chilled sea bass to be consumed at a specific quality level, expressed as remaining shelf life, and demonstrate the optimization of final quality achieved with SLDS compared to the conventional FIFO-based random practice. For the


local market, Fig. 4(a), with the FIFO system, 7% of products were beyond acceptable quality at consumption, whereas with SLDS, unacceptable products were eliminated. For the export market, Fig. 4(b), from 30% with FIFO, unacceptable products were reduced to 13% with SLDS.

4. Conclusion In the present study, kinetic growth models of natural microflora of marine cultured sea bass were developed. Sensory shelf life was correlated to pseudomonads that were recognized as effective SSO and sensory acceptability was correlated to SSO level, NS. The variability of the initial SSO population, reliable estimation of which is required in shelf life determination, was quantitatively shown and a conductancebased rapid method specific to sea bass SSO enumeration was established as a practical means of N0 determination. All these were integrated into the SLDS which was demonstrated to be an effective tool for marine cultured sea bass chill chain management leading to optimization of quality of the fish at consumer’s end.

Acknowledgements The present study was partly supported by the Commission of the European Communities, Agriculture and Fisheries specific RTD programme, FAIRCT95-1090 ‘‘Development, modelling and application

Fig. 4. Distribution of quality (remaining shelf life in hours) at consumption time of sea bass (D. labrax) products when the first in first out (FIFO) or the shelf life decision system (SLDS) chill chain management approaches are used for (a) the local and (b) the export market.


K. Koutsoumanis et al. / International Journal of Food Microbiology 73 (2002) 375–382

of Time Temperature Integrators to monitor chilled fish quality.’’

References Anonymous, 1978. Microorganisms in foods: 1. Their significance and methods of enumeration. International Commission on Microbiological Specifications for Foods (ICMSF), 2nd edn. University of Toronto Press, Toronto, London, p. 45. Anonymous, 1995. Fig. P for windows, user’s Manual version 2.5. Fig. P Biosoft Software, Cambridge, CB2 1LR, UK. Bolton, F.J., Gibson, D.M., 1995. Automated electrical techniques in microbiological analysis. In: Patel, P. (Ed.), Rapid Analysis Techniques in Food Microbiology. Blackie Academic & Professional, London, pp. 131 – 169. Dalgaard, P., 1995. Modelling of microbial activity and prediction of shelf life of packed fresh fish. Int. J. Food Microbiol. 19, 305 – 318. Dalgaard, P., Gram, L., Huss, H.H., 1993. Spoilage and shelf life of cod fillets packed in vacuum or modified atmospheres. Int. J. Food Microbiol. 19, 283 – 294. Gennari, M., Campanini, R., 1991. Isolamento e caratterizzazione di Shewanella putrefaciens da pesce fresco e alterato, carni fresche e alterate, prodotti lattiero-caseari, acqua e suolo. Ind. Aliment. 30, 965 – 976, 988. Giannakourou, M., Koutsoumanis, K., Nychas, G.J.E., Taoukis, P.S., 2001. Development and assessment of an intelligent Shelf life Decision System (SLDS) for quality optimization of the food chill chain. J. Food Prot. 64, 1051 – 1057. Gibson, D.M., Ogden, I.D., Hobbs, G., 1984. Estimation of the bacteriological quality of fish by automated conductance measurements. Int. J. Food. Microbiol. 1, 127 – 134. Gibson, D.M., 1985. Predicting the shelf life of packaged fish from conductance measurements. J. Appl. Bacteriol. 58, 465 – 470. Gibson, D.M., Hobbs, G., 1987. Some recent developments in microbiological methods for assessing seafood quality. In: Kramer, D.E., Liston, J. (Eds.), Seafood Quality Determination. Proceedings of the International Symposium on In Seafood Quality Determination, 10 – 14 November 1986, Anchorage, AK, USA, Elsevier, Amsterdam, pp. 283 – 298. Gram, L., Trolle, G., Huss, H.H., 1987. Detection of specific spoilage bacteria from fish stored at low (0 C) and high (20 C) temperatures. Int. J. Food Microbiol. 4, 65 – 72. Gram, L., Huss, H.H., 1996. Microbiological spoilage of fish and fish products. Int. J. Food. Microbiol. 33, 121 – 137.

Huss, H.H., 1995. Quality and quality changes in fresh fish. FAO Fisheries Technical Paper no. 348. FAO, Rome. Koutsoumanis, K., Nychas, G.-J.E., 1999. Chemical and sensory changes associated with microbial f lora of boque (Boops boops) stored aerobically at 0, 3, 7 and 10 C. Appl. Environ. Microbiol. 65, 698 – 706. Koutsoumanis, K., Nychas, G.-J.E., 2000. Application of a systematic experimental procedure to develop a microbial model for rapid fish shelf life procedure. Int. J. Food Microbiol. 60, 171 – 184. Koutsoumanis, K., Giannakourou, M.C., Taoukis, P.S., Nychas, G.J.E., 2000a. Application of shelf life decision system (SLDS) to marine cultured fish quality. In: Van Impe, J.F.M., Bernaerts, K. (Eds.), Predictive Modelling in Foods — Conference Proceedings. KULeuven/BioTeC, Belgium, pp. 281 – 291 (ISBN 90-804818-3-1). Koutsoumanis, K., Taoukis, P., Drosinos, E., Nychas, G.-J.E., 2000b. Applicability of an Arrhenius model for the combined effect of temperature and CO2 packaging on the spoilage microflora of fish. Appl. Environ. Microbiol. 66, 3528 – 3534. Mead, G.C., Adams, B.W., 1977. A selective medium for the rapid isolation of Pseudomonas associated with poultry meat spoilage. Br. Poult. Sci. 18, 661 – 670. Mossel, D.A.A., Eelderink, L., Koopmans, M., Rossem, F.V., 1979. Inf luence of carbon source, bile salts and incubation temperature on recovery of Enterobacteriaceae from foods using MacConkey-type agars. J. Food Prot. 42, 470 – 475. Salvat, G., Rudelle, S., Humbert, F., Colin, P., Lahellec, C., 1997. A selective medium for the rapid detection by an impedance technique of Pseudomonas spp. associated with poultry meat. J. Appl. Microbiol. 83, 456 – 463. Taoukis, P.S., Koutsoumanis, K., Nychas, G.J.E., 1999a. Modelling of spoilage microflora of boque (Boops boops) as a basis for chilled distribution monitoring with time – temperature indicators. In: Bourgeois, C.M., Roberts, T.A. (Eds.), Predictive Microbiology Applied to Chilled Food Preservation (Proceedings of the International Symposium, Quimper, France, June 16 – 18, 1997). Refrigeration Science and Technology Proceedings Series, International Institute of Refrigeration (IIR), Paris, France, pp. 316 – 325. Taoukis, P.S., Koutsoumanis, K., Nychas, G.-J.E., 1999b. Use of time temperature integrators and predictive modelling for shelf life control of chilled fish under dynamic storage conditions. Int. J. Food Microbiol. 53, 21 – 31. Van Spreekens, K.J.A., 1974. The suitability of a modification of Long and Hammer’s medium for the enumeration of more fastidious bacteria from fresh fishery products. Arch. Lebensmittelhyg. 25, 213 – 219.