University of Nebraska - Lincoln

DigitalCommons@University of Nebraska - Lincoln

Publications, Agencies and Staff of the U.S. U.S. Department of Commerce

Department of Commerce

2011

Quality Assessment of Filtered Smoked Yellowfin Tuna (Thunnus

albacares) Steaks
Lori F. Pivarnik
University of Rhode Island, Pivarnik@uri.edu

Cameron Faustman
University of Connecticut - Storrs

Santiago Rossi
University of Rhode Island

Surendranath P. Suman
University of Kentucky

Catherine Palmer
University of Rhode Island

See next page for additional authors

Follow this and additional works at: https://digitalcommons.unl.edu/usdeptcommercepub

Part of the Environmental Sciences Commons

Pivarnik, Lori F.; Faustman, Cameron; Rossi, Santiago; Suman, Surendranath P.; Palmer, Catherine; Richard,
Nicole L.; Ellis, P. Christopher; and DiLiberti, Michael, "Quality Assessment of Filtered Smoked Yellowfin
Tuna (Thunnus albacares) Steaks" (2011). Publications, Agencies and Staff of the U.S. Department of
Commerce. 307.
https://digitalcommons.unl.edu/usdeptcommercepub/307

This Article is brought to you for free and open access by the U.S. Department of Commerce at
DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Publications, Agencies and
Staff of the U.S. Department of Commerce by an authorized administrator of DigitalCommons@University of
Nebraska - Lincoln.
Authors
Lori F. Pivarnik, Cameron Faustman, Santiago Rossi, Surendranath P. Suman, Catherine Palmer, Nicole L.
Richard, P. Christopher Ellis, and Michael DiLiberti

This article is available at DigitalCommons@University of Nebraska - Lincoln: https://digitalcommons.unl.edu/

usdeptcommercepub/307
Quality Assessment of Filtered Smoked Yellowfin
Tuna (Thunnus albacares) Steaks
Lori F. Pivarnik, Cameron Faustman, Santiago Rossi, Surendranath P. Suman, Catherine Palmer, Nicole L. Richard,
P. Christopher Ellis, and Michael DiLiberti

Abstract: Filtered smoke (FS) has been used to preserve taste, texture, and/or color in tuna and other fish species. This
treatment is particularly important in color preservation during frozen storage. The objective of this study was to compare
changes in the quality profiles of FS-treated and untreated (UT) yellowfin tuna (Thunnus albacares) steaks stored in 3 ways:
room temperature (21 to 22 ◦ C), refrigerated (4 to 5 ◦ C), and iced (0 ◦ C). FS and UT steaks were processed from the
same lot of fish and analyzed for chemical, microbiological, lipid oxidation, color, and sensory profiles. Similar trends
were seen for microbial proliferation and accumulation of apparent ammonia and total volatile base nitrogen (TVB-N)
during the storage temperatures evaluated. Notable exception in quality profile was found in lipid oxidation which was,
as expected, lower for treated samples at all storage temperatures for TBARS (P < 0.05) and lower or significantly (P <
0.05) lower for POV values. FS increased the initial redness value significantly (P < 0.05). Unlike UT product, there was
no loss of color value concomitant with quality changes for FS-treated tuna for all storage temperatures evaluated.
Keywords: carbon monoxide, color, filtered wood smoke, seafood quality, tuna steaks

Practical Application: The overall goal of this project was to evaluate filtered smoked tuna steaks as to the impact on the
overall quality profile. As a color-stabilizing technology, it could mask deteriorating quality.

Introduction products—snapper, mahi mahi, marlin, swordfish, and tilapia—are

Color of meat and seafood has a strong influence on consumer also being treated to help preserve color. During the FS process,
acceptance. A bright red color is an important quality determi- natural wood smoke is filtered to remove all or most of the taste

S: Sensory & Food

nant in seafood, particularly tuna, as the market value is based and odor components, and particulate matter from the vapor phase
on this attribute. After being cut and during standard commercial of the smoke. This leaves CO as a significant component to bind

Quality
frozen storage of tuna, there is a rapid formation of a brown color to the heme molecule and “fix” the red color in dark muscle
(Anderson and Wu 2005). Color changes in tuna, from red/purple (Kristinsson and others 2006b). In addition, improving myoglobin
to red-brown to brownish-red, reflect the chemical oxidation state stability may improve lipid stability. The heme proteins are strong
of the myoglobin within the muscle. The reduced ferrous heme catalysts of lipid oxidation in muscle (Richards and Hultin 2002;
iron ion (Fe2+) in the myoglobin molecule is prone to oxidation Undeland and others 2004). Research on meat and tuna muscle
to the ferric form (Fe3+), which affects the color, taste, and tex- has shown an interaction between oxymyoglobin oxidation and
ture (Faustman and Cassens 1990; Chan and others 1998). The lipid oxidation (Faustman and others 1989; Lee and others 2003).
use of carbon monoxide (CO) either alone or as part of a filtered Oxymyoglobin oxidation by-products are prooxidative towards
wood smoke (FS) process, has been applied to seafood in an ef- unsaturated fatty acids, and conversely, the process of lipid oxida-
fort to maintain the desirable color attributes during storage and tion can generate chemical species that predispose myoglobin to
transportation. When CO complexes with the heme iron of myo- rapid oxidation (Greene 1969; Greene and others 1971; Faustman
globin in meat it forms the stable red pigment, carboxymyoglobin. and others 2010). FS treatment could result in fewer oxidation
Carboxymyoglobin is more resistant to oxidation than oxymyo- products of fatty acids, which in turn, would result in delayed
globin because CO has a stronger binding affinity to the heme iron production of off-flavors and off-odors in meat. However, the
than oxygen does (Livingston and Brown 1981). Although tuna use of color-stabilizing technologies with fresh meat might mask
has been a primary product treated by this process, other seafood microbial spoilage (Faustman and others 1989).
Filtered smoke treatment, when compared to UT controls, has
been shown by previous research to decrease bacterial load, in-
crease oxidative stability, and increase red color stability in stored
MS 20101210 Submitted 10/25/2010, Accepted 5/25/2011. Authors Pivarnik,
Rossi, Palmer, and Richard are with Univ. of Rhode Island, Dept. of Nutrition tuna (Ludlow and others 2004; Kristinsson and others 2008), mahi
and Food Sciences, West Kingston, RI, USA. Author Faustman is with Univ. of mahi (Demir and others 2004; Kristinsson and others 2007), and
Connecticut, Dept. of Animal Science, Storrs, CT, U.S.A. Author Suman is with Univ. Spanish mackerel (Garner and Kristinsson 2004). The researchers
of Kentucky, Dept. of Animal and Food Sciences, Lexington, KY, U.S.A. Author Ellis speculated that CO2 (21% of FS) and CO (18% of FS) worked
is with Rhode Island Dept. of Health, Food Chemistry Lab., Providence, RI, U.S.A.
together to reduce and suppress microbial growth and that a va-
Author DiLiberti is with Natl. Technical Services, USDC/NOAA/NSIP, Gloucester,
MA, U.S.A. Direct inquiries to author Pivarnik (E-mail: Pivarnik@uri.edu). riety of other compounds in FS may impart antimicrobial effects.
Impacts observed when FS was applied were not always seen when


C 2011 Institute of Food Technologists
R

doi: 10.1111/j.1750-3841.2011.02276.x Vol. 76, Nr. 6, 2011 r Journal of Food Science S369
Further reproduction without permission is prohibited
 Assessment of filtered smoked tuna. . .

pure CO was used (Kristinsson and others 2006a). The application thickness of UT steaks were 236.9 ± 86.2 g and 20.3 ± 2.7 mm,
of pure CO gas on tuna quality resulted in a delay of undesirable respectively. The average weight and thickness of FS steaks were
color changes when high gas concentration (100%) was used; low 235 ± 73.2 g and 20.0 ± 3.5 mm, respectively. UT steaks were
levels (4%) did not impact color. In addition, FS has been shown vacuum-packaged and frozen immediately. Remaining steaks were
to have a greater impact on microbial growth than pure CO due treated with filtered smoke.
to possible residual impacts in the muscle tissue (Kristinsson and
others 2006b).
Thus far, research has focused on the effects of CO alone and/or Filtered smoke treatment
FS treatment on fish quality as delineated by microbial, lipid oxida- A total of 4 to 6 randomly sampled steaks were placed in a
tion and color stability assessments (Garner and Kristinsson 2004; vacuum bag, air removed, and treated with filtered smoke using
Ludlow and others 2004; Kristinsson and others 2007, 2008). pilot scale equipment at Clearsmoke Technologies. The smoke
Only mahi-mahi was studied for sensory attributes with semi- contained 16% CO (verified on-site). Bags were held under re-
trained sensory analysts (Kristinsson and others 2008). No study frigeration for about 48 h, flushed, vacuum-packaged, sealed, and
has integrated expert sensory assessments and traditional seafood frozen after treatment. All UT and FS samples were held for 30 d
spoilage indicators (for example, TVB) with microbial, lipid ox- and shipped frozen, overnight, to the Univ. of Rhode Island where
idation, and color determinants for a complete quality profile of they were held frozen until storage trials began.
tuna that has been treated with FS. This additional information
would be necessary to more fully assess the impact of FS on per- Storage and preparation of fish samples
ceived quality of seafood. Concerns have been raised regarding All frozen samples were thawed at refrigerated temperature (4 to
quality and safety of these products which may be disguised by a 5 ◦ C) prior to time/temperature storage trials. Three steak sam-
“fresh-looking” piece of fish (Anderson and Wu 2005). The orig- ples were placed next to each other in unclosed, oxygen permeable
inal draft of the Food Safety Enhancement Act (H.R. 3/26/2009; plastic bags and stored at room temperature (22 ◦ C), refrigerated
H.R. July 2009) reflected this on-going concern by including a (4 to 5 ◦ C), and in ice (0 ◦ C). UT and FS products were packaged
directive for the FDA to conduct a safety assessment on the use of separately. Fish stored at room temperature were evenly spaced out
carbon monoxide on meat, poultry, and seafood products. While in plastic trays under a fume hood. Products held at refrigerated
this order does not appear to be included in the final amended temperatures were stored in plastic trays and spaced evenly. Sam-
bill, it indicated that there were still concerns about this particular ples held in ice were positioned upright in an insulated cooler,
technology on quality and safety of food. Therefore, more infor- surrounded by ice with the opening to the plastic bags oriented
mation is still needed to completely evaluate the impact of filtered to prevent melting ice water from entering the bag. A day “zero”
smoke on the indicators of quality of seafood. represented the first set of samples.
The overall goal of this project was to evaluate commercially Packages for each sampling interval were assigned a ran-
processed, filtered smoked tuna steaks as to the overall quality dom 3-digit code and duplicate bags of fish samples, each bag
profile. Specific objectives included establishing (1) profiles of with 3 steaks, were selected randomly at each time point. One
chemical quality indicators, (2) enumeration of microbial growth, bag containing samples (n = 3) was immediately tested for
S: Sensory & Food

(3) lipid oxidation changes, (4) sensory assessment parameters, and aerobic plate counts and then vacuum-packaged (Super Vac,
(5) color change and stability.
Quality

Smith Equipment, Clifton, N.J., U.S.A.), frozen and stored at

−70 ◦ C for chemical analyses. The other package was vacuum-
Materials and Methods packaged, stored at −70 ◦ C until overnight shipment for sensory
testing at the Natl. Marine Fisheries Service, Gloucester, Mass.,
Raw material U.S.A.
Yellowfin tuna (Thunnus albacares) was obtained through Samples for color and lipid oxidation were transported on ice to
ClearsmokeR Technologies, Atlanta, Georgia. The fish were the Univ. of Connecticut. Fish samples were stored and sampled
caught by long lines in the Gulf of Mexico in December and under the same conditions as at the URI facility. Color deter-
were headed, gutted, and fins removed on-board. The fish were minations were made on each steak followed by lipid oxidation
shipped via ground transportation from Louisiana, packaged in measurements.
ice, and held under refrigeration upon receipt at the Clearsmoke
Technologies facility. The 4 to 5 d old fish were rated as Grade 2.
Grade 2 tuna is used in the lower end sashimi markets or high end Microbiological assessment
restaurants, where tuna is prepared semi cooked (Bartram 1996; Psychrotrophic bacteria counts were obtained by the standard
Ledafish 2011). The main differences between Grade 1 and Grade agar plate count per AOAC (1990) method 966.23 and FDA (1995)
2 tuna is with respect to red color; where Grade 1 tuna has bright BAM. Total aerobic plate counts on fish tissue were done asep-
red muscle and translucent flesh (clarity) compared to Grade 2 tuna tically. Approximately one-half of each steak was blended using
with red muscle and some translucency (Bartram 1996; Ledafish a Handy-ChopperTM (Black-Decker, Shelton, Conn., U.S.A.).
2011). Total of 3 whole tuna, from the same lot, were used for Total of 11 g of the blended tissue were transferred into indi-
this project. One upper and one lower tuna loin from each fish vidual sterile Whirlpak bags containing 99 mL of sterile Butter-
were separated into 2 groups—one that would be treated with FS. fields Phosphate buffer and pulsified, using a PulsifierTM (Filtaflex
and one that would not be treated (UT). Each group contained Ltd., Almonte, Ontario, Canada) for 60 s. This was followed by
equal numbers of pieces from upper and lower loins. A total of standard dilutions and plating using the spread plate technique
48 steaks for FS and UT samples were used for this study and the onto plate count agar and incubated at 25 ◦ C for 48 h follow-
steaks were sampled randomly. The average weight and thickness ing accepted methods of sampling, preparation, and enumeration
measurements were taken on 14 (n = 14) randomly selected tuna (Banwart 1989; FDA 1995; Venkitanarayanan and others 1997; Jay
steaks for untreated and treated samples. The average weight and 2000). Mean values represented the 3 determinations.

S370 Journal of Food Science r Vol. 76, Nr. 6, 2011

Assessment of filtered smoked tuna. . .

Sample preparation for apparent ammonia and TVB-N on the tuna steaks using a Minolta Chromameter CR 200 (Osaka,
analyses Japan) calibrated to a standard white plate. Illuminant used was
Vacuum-sealed frozen samples were thawed using cold running C (6774K) and the measuring area was 8 mm. All fish samples
water in preparation for chemical analyses. Thawed steaks were ho- were analyzed in their vacuum packages. Total of 3 different mea-
mogenized, individually, using a handy-chopper (Black-Decker). surements were taken across each product surface and averaged to
Each homogenate was then sampled for apparent ammonia and obtain the mean value for each experimental unit. In addition, the
TVB determinations resulting in 3 measurements. All homog- color penetration was measured in representative samples. Samples
enized samples were vacuum-packaged (Super Vac) in pouches were sliced in half, perpendicular to the long axis of the muscle,
(Market Sales Co., Newton, Mass., U.S.A.) and frozen at −70 ◦ C and measured (mm) relative to the depth of surface red coloration.
until analysis was performed. This was done over storage to determine if the surface color layer
changed in a manner consistent with what has been observed in
Apparent ammonia by ISE red meat (Faustman and Cassens 1990).
Apparent ammonia was determined on all homogenates using
the ISE procedure (AOAC 2000) method 999.01 as originally Expert sensory assessment
outlined by Pivarnik and others (1998). Briefly, 5 g of comminuted Vacuum-packaged, frozen fish samples were shipped overnight
fish tissue samples were blended with 95 mL of water for 2 min, to the USDC/NOAA sensory laboratory in Gloucester, MA. This
and pH adjusted with 2 mL alkaline ion-strength adjuster (ISA) laboratory fulfilled all of the requirements mandated by the ASTM
solution. The compounds contributing to the ammonia response STP 913 (ISO 1993) guidelines for physical design of sensory eval-
were immediately determined using a precalibrated Orion model uation laboratories. Upon arrival, the samples were checked to
95–12 ammonia gas-sensing, ion-specific electrode, and an Orion ensure that the frozen integrity had been maintained and placed
Model 290A portable pH/ISE meter (Thermo Orion, Beverly, in a −80 ◦ C freezer and stored for no longer than 1 mo before
Mass., U.S.A.). All results were reported as miligram apparent sensory evaluation of raw product. Prior to a scheduled sensory
ammonia per 100 g of fish tissue (Pivarnik and others 1998). session, samples were prepared in an odor-free area that was sepa-
rated from the sensory testing facility. The vacuum-sealed samples
TVB-N analysis were tempered at 4 ◦ C for 12 h, and brought to room tempera-
TVB analysis was conducted on a double trichloroacetic acid ture in running water immediately before sensory evaluation. All
(TCA) extraction. TVB concentrations were determined by distil- packages were identified by random 3-digit codes and placed in
lation and titration as specified by published procedures (Malle and booths.
Tao 1987; Malle and Poumeyrol 1989). Briefly, 50 g comminuted Three expert inspectors, internationally calibrated and trained,
fish were blended in 100 mL 7.5% (w/v) TCA solution at high and who had achieved test scores of greater than 85% during past
speed for 2 min. The homogenate was filtered through Whatman testing sessions, were chosen to conduct the sensory assessments on
nr 1 filter paper. A 25 mL aliquot of the TCA extract was pipetted all samples. Each inspector had at least 10 y of practical experience
into a distilling flask (250 mL Tecator digestion tube) and 10 mL and had been trained by attending a minimum of 3 international
10% NaOH was added. Steam distillation (Tecator Kjeltec Model free trade harmonization workshops. After training, using a 100

S: Sensory & Food

1002 distillation unit) lasted until 75 mL liquid was collected in mm line, each inspector was presented with 120 test samples and
had to achieve a score of 85% or better and “expert” status accord-

Quality
a 125 mL Erlenmeyer flask containing 10 mL Kjeldahl indicator
solution (4 g boric acid in distilled water containing 0.7 mL 0.1% ing to statistical criteria agreed upon by FDA and NMFS in the
alcoholic solution of methyl red and 1.0 mL 0.1% alcoholic solu- U.S. and the Canadian Food Inspection Agency (Reilly and York
tion of bromocresol green diluted to 100 mL in distilled water) per 1993). Each inspector also met International Standards (ISO) for
AOAC (1995) method 980.10. The green alkaline distillate was sensory evaluation criteria (ISO 1993). A minimum of 3 sensory
back-titrated with 0.025 N sulfuric acid to its original red color. experts is required to achieve statistically valid results for qual-
ity and decomposition evaluation and to ensure a high degree of
Lipid oxidation acuity and reproducibility (Poste and others 1991; Sims and others
The thiobarbituric acid (TBA) procedure of Yin and others 1992; Pivarnik and others 2001). The analysts independently eval-
(1993) was used to assess lipid oxidation and reported as TBA- uated each sample for appearance, texture, and odor in the raw
Reactive Substances (TBARS). Analysis of peroxide values (POV) state. Sample evaluation was conducted in a facility illuminated
was conducted by a modified procedure of Shantha and Decker with artificial daylight and red amber lights.
(1994). Whole muscle steak samples were stored in vacuum pack- Analysts evaluated each sample using a standardized ballot de-
ages for the storage portion of this study to prevent oxygen pen- veloped during international exercises involving harmonization
etration into the package. For the TBA test, BHA was added for product sensory standards and criteria that USDC/NOAA
into the acid tissue extract to prevent the production of oxidation seafood analysts had participated in developing. The ballot con-
products during the analytical preparation of the sample. Briefly, sisted of the 3-digit sample number, an unstructured 100-mm line
triplicate 5 g samples were each added to 10 mL distilled water scale, a place to indicate whether the sample passed or failed for de-
and 12.5 mL 20% trichloroacetic acid (TCA). The mixture was composition, and a space to write any descriptive information for
homogenized for 1 min in a Waring blender and filtered through each sample. Overall evaluation, while primarily reflecting odor,
Whatman nr 1 filter paper. A total of 1 mL filtrate was mixed with incorporated some general textural and appearance qualifiers as
1 mL 20 mM aqueous TBA and incubated at 25 ◦ C for 21 h. The verification for the odor assessment. The line scale represented
absorbance was measured at 532 nm and reported as TBARS. the degree of continuous deterioration of the sample, where 0 =
no deterioration and 100 = severe decomposition. Samples ob-
Color assessment taining sensory scores of greater than 50 were considered unac-
L ∗ (lightness), a∗ (redness), and b∗ (yellowness) values (Faustman ceptable, as agreed during an International Free Trade agreement
and Phillips 2001) were recorded from 2 different surface locations workshop and detailed during the harmonization workshops.

Vol. 76, Nr. 6, 2011 r Journal of Food Science S371

 Assessment of filtered smoked tuna. . .

Following independent evaluation, a panel leader collected FS product after 8 d of storage. Application of Clearsmoke (FS) to
pass/fail decisions, numerical data, and any terminology that de- yellowfin tuna was also shown by Kristinsson and others (2008) to
scribed sensory characteristics. depress microbial growth. However, in some of the studies, while
the same FS was applied, it was done in a controlled laboratory
Statistical analysis environment, using a commercial cylinder as the source of the
All assays were determined in triplicate and the results were FS gas at 18% CO (Kristinsson and others 2006b, 2008). The gas
reported as means. The significance of the differences were de- applied in this study reflected commercial application in a pilot
termined between treated and untreated samples, at the different plant facility using 1 d old compressed FS gas with a CO con-
temperature storage conditions by two-way analysis of variance centration of 16%. This variation could account for the difference
(ANOVA) followed by F-tests and student t-tests using the Excel in the results obtained reflecting application parameters and/or
data analysis tools (Microsoft Excel 2000, Microsoft Corp., Red- level of gas saturation in the product (not evaluated in this study).
mond, Wash., U.S.A.). Significance of difference was reported at This disagreement in microbial profiles reported by the different
P ≤ 0.05. studies is noteworthy since they could reflect the importance of
standardized protocol for FS application and its ultimate impact on
Results and Discussion microbial assessment. In this study, filtered smoke did not appear to
Microbiological analyses indicated that at time zero, there was a impact microbial proliferation either initially or during prolonged
significant (P < 0.05) difference between UT and FS samples storage at the temperatures evaluated.
(Figure 1), where initial microbial concentrations were 3.9 ± The other quality indices measured did not show a quality
0.39 log CFU/g fish and 5.0 ± 0.4 log CFU/g fish, respec- profile impact of treatment over untreated samples during storage
tively. However, no differences (P > 0.05) in microbial pro- at the different storage temperatures. Accumulation of apparent
liferation were found between UT and FS samples when the ammonia, with the exception of room temperature storage, and
entire storage period was evaluated. These results were unex- TVB-N in UT and FS tuna samples did not differ (P > 0.05)
pected, since the results of previous research had shown an im- during storage (Figure 2 to 4). FS-treated product did have slightly
pact of FS treatment on lowering the initial microbial levels higher apparent ammonia at time zero, 19 mg/100g compared
in tuna. While similar microbial growth trends to this study with 16.7 mg/100g for untreated product, and significantly greater
were observed in tilapia, where no differences were found over (P < 0.05) accumulation during room temperature storage.
storage time at room temperature and refrigerated storage tem- A notable exception to the lack of difference in quality profile
perature (Leydon and others 2005), other research has shown indicators between FS and UT samples was for lipid oxidation.
an impact of treatment on the initial microbial load in mahi Lipid oxidation indicators were lower (P < 0.05) for treated sam-
mahi (Kristinsson and others 2007) and tuna (Ludlow and others ples at room temperature and iced for POV values (Figure 5) and
2004; Kristinsson and others 2008). Kristinsson and others (2007) at all storage temperatures for TBARS (Figure 6). As expected,
showed that while lower (P < 0.05) microbial concentrations these values were lower at lower storage temperatures (P < 0.05).
were determined for FS-treated mahi mahi during initial storage at The FS treatment depressed the oxidative rancidity development
4 ◦ C, no differences (P > 0.05) were determined between UT and (P < 0.05) during all storage temperatures evaluated. Other
S: Sensory & Food
Quality

10 Figure 1–Microbiological proliferation in FS

and UT tuna steaks stored at room
temperature (22 ◦ C), refrigerated (4 to 5 ◦ C),
9 and in ice (0 ◦ C).

7
CFU / g fish (log 10)

UT Room
5
FS Room
UT Refrigerated
4 FS Refrigerated
UT Ice
3 FS Ice

2
UT and FS samples significantly different (P <0.05) at time zero.

0
0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360

Time (hours)

S372 Journal of Food Science r Vol. 76, Nr. 6, 2011

Assessment of filtered smoked tuna. . .

researchers have also shown that CO and/or FS treatments, when Kristinsson and others 2007; Faustman and others 2010). Stabiliza-
compared with untreated products, effectively depressed lipid ox- tion of heme proteins would be expected to reduce lipid oxidation
idation in mahi mahi (Kristinsson and others 2007), tuna (Ludlow rates (Kristinsson and others 2007) as the relationship between
and others 2004), and Spanish mackerel (Garner and Kristins- myoglobin and lipid oxidation in yellowfin tuna has been docu-
son 2004). The major components involved in lipid oxidation mented (Lee and others 2003; Faustman and others 2010). Other
of fish muscle are the heme proteins, hemoglobin, and myo- components of FS include carbon dioxide (CO2 ), and gaseous
globin (Richards and Hultin 2002; Richards and Dettmann 2003; phenolics (Kowalski 1999; Kristinsson and others 2007). Carbon

45 Figure 2–Apparent ammonia and TVB-N in FS

and UT tuna steaks stored at room
temperature (22 ◦ C).
40 UT and FS apparent ammonia samples
significantly different (P <0.05) over storage
35

30
mg / 100 g fish

25

20

15
UT ammonia
10 FS ammonia
UT TVB-N
FS TVB-N
5

0
0 5 10 15 20 25 30 35 40

Time (hours)

S: Sensory & Food

Quality
40 Figure 3–Apparent ammonia and total volatile
base-nitrogen (TVB-N) in treated (FS) and
untreated (UT) tuna steaks stored refrigerated
35 (4–5 ◦ C).

30

25
mg / 100 g fish

20

15

UT ammonia
10
FS ammonia
UT TVB-N
FS TVB-N
5

0
0 1 2 3 4

Time (days)

Vol. 76, Nr. 6, 2011 r Journal of Food Science S373

 Assessment of filtered smoked tuna. . .

dioxide and N2 might impact bacterial growth, but these gases phenols remain; however, their antioxidative function would be
have not shown a strong correlation with color and/or lipid oxi- minor compared to CO (Kristinsson and others 2006a).
dation (Kristinsson and others 2006a). Phenolics may contribute Significantly (P < 0.05) different expert sensory evaluations
to the antioxidant impact of the FS. Although many of the partic- are shown for UT and FS tuna steaks stored at all trial tempera-
ulate phenols are removed during filtration, some minor gaseous tures in Figure 7 to 9. The chemical and microbial quality profile

35 Figure 4–Apparent ammonia and TVB-N in FS

and UT tuna steaks stored in ice (0 ◦ C).

30

25
mg / 100 g fish

20

15

10
UT ammonia
FS ammonia
UT TVB-N
5 FS TVB-N

0
0 3 6 9 12 15
S: Sensory & Food

Time (days)
Quality

165 Figure 5–Peroxide values for FS and UT tuna

UT and FS room temperature and iced samples significantly steaks stored at room temperature (22 ◦ C),
150 different (P <0.05) over storage refrigerated (4 to 5 ◦ C), and in ice (0 ◦ C).

135

120
Peroxide value meq/kg

105

90

75

60

UT Room
45
FS Room
UT Refrigerated
30 FS Refrigerated
UT Ice
15 FS Ice

0
0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360

Time (hours)

S374 Journal of Food Science r Vol. 76, Nr. 6, 2011

Assessment of filtered smoked tuna. . .

measurements for refrigerated (Figure 7) and iced (Figure 8) tuna and FS samples were at borderline quality after 4 d at refrigeration
steaks were supported by the sensory evaluation, where the time temperatures and were considered unacceptable after 15 d on ice,
of rejection or borderline acceptability (≥ 50 mm) of the tuna in agreement with the microbial proliferation profile observed.
steaks were not different between UT and FS products. Similar However, when fish was stored at abusive or room temperature
results were obtained for mahi-mahi where FS and UT samples, (Figure 9), sensory assessment indicated significant differences in
stored at 4 ◦ C, both crossed above the limit of sensory accept- acceptability between UT and FS, with UT samples having ac-
ability at the same time (Kristinsson and others 2007). Both UT celerated unacceptable sensory scores. During the storage trials,

1.0 Figure 6–TBAR values for FS and UT tuna

steaks stored at room temperature (22 ◦ C),
UT and FS samples significantly different (P <0.05) refrigerated (4 to 5 ◦ C), and in ice (0 ◦ C).
for all storage conditions

0.8
UT Room
FS Room
UT Refrigerated
TBARS A 532 nm

0.6 FS Refrigerated
UT Ice
FS Ice

0.4

0.2

0.0
0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360

S: Sensory & Food

Time (hours)

Quality
100 Figure 7–Expert sensory evaluation of FS and
UT tuna steaks stored refrigerated (4 to 5 ◦ C).
UT
90 FS

80

70
Sensory score

60

50

40

30

20

10

0
0 1 2 3 4

Time (days)

Vol. 76, Nr. 6, 2011 r Journal of Food Science S375

 Assessment of filtered smoked tuna. . .

in addition to the documented added benefit of FS treatment, developed quickly and dramatically to levels that may have been
lower temperatures would also suppress the development of ox- more easily identified by sensory analysts in the UT product. This
idative rancidity. Concentrations of oxidative compounds formed could account for the sensory differences (P < 0.05) obtained
(Figure 5 and 6) at lower temperatures, though lower in FS-treated between UT and FS tuna stored at room temperature.
products than UT, may not have been sufficiently high enough in Finally, color is an important quality attribute of meat and
either product to have impacted expert sensory assessment of UT seafood, especially for tuna where the market value is based on
or FS tuna. However, at room temperature oxidative rancidity muscle appearance and color. Lightness (L ∗ ), yellowness (b∗ ), and

100 Figure 8–Expert sensory evaluation of FS and

UT tuna steaks stored in ice (0 ◦ C).
UT
90 FS

80

70
Sensory score

60

50

40

30

20

10

0
0 3 6 9 12 15
S: Sensory & Food

Time (days)
Quality

100 Figure 9–Expert sensory evaluation of FS and

UT tuna steaks stored at room temperature (22
◦
UT UT and FS samples significantly C).
90 FS different (P <0.05) over storage
80

70
Sensory Score

60

50

40

30

20

10

0
0 5 10 15 20 25 30 35 40

Time (hours)

S376 Journal of Food Science r Vol. 76, Nr. 6, 2011

Assessment of filtered smoked tuna. . .

redness (a∗ ) are presented in Table 1 to 3, respectively. No dif- Faustman and others 2010). The redox state of the heme iron
ferences were observed for initial L ∗ and b∗ color values between dictates the color of fish muscle (Lee and others 2003). The re-
UT and FS tuna (Table 1 and 2, respectively). Overall, there were duced ferrous heme iron ion (Fe2+) in the myoglobin molecule
no clear trends that showed that treatment affected L ∗ or b∗ color is prone to oxidation to the ferric form (Fe3+), which affects the
values consistently. There were no changes in L ∗ for any of the color, taste, and texture (Faustman and Cassens 1990; Chan and
storage conditions for FS fish (Table 1). While b∗ showed a slight others 1998). The CO component of FS binds with the heme
significant (P < 0.05) increase in FS tuna during storage in ice, iron of myoglobin forming carboxymyoglobin, which is more re-
this was not observed for other samples evaluated (Table 2). No sistant to oxidation than oxymyoglobin because CO has a stronger
effect of treatment on L ∗ and b∗ was reported by other researchers binding affinity to the heme iron than oxygen does (Livingston
(Kristinsson and others 2006b). However, redness, a∗ value, which and Brown 1981; Kristinsson and others 2007). While forma-
is considered an important color parameter since red color is highly tion of Maillard products could contribute to surface browning
valued for quality, was impacted by the filtered smoke treatment through aldehyde groups of lipid oxidation products and amines,
for all storage temperatures. Treatment significantly (P < 0.05) this has been more typically associated with cooked or freeze-dried
increased a∗ at time zero, with values of 5.1 ± 1.2 and 9.7 ± meat products. Conversion of red ferrous oxy-heme proteins (that
1.4 for UT and FS tuna, respectively (Table 3). These results agree is, myglobin, hemoglobin) to brownish ferric met-heme proteins
with Kristinsson and others (2006c), who reported an approximate has been attributed to the color changes typically observed in
2-fold increase in a∗ for tuna steaks, grades A and B, treated with fresh/frozen tuna (Chow and others 2004; Sohn and others 2005;
either 100% CO or FS (20% CO). When grade C brown tuna Thiansilakul and others 2011). While it has been well documented
steaks were treated with 100% CO, the red color was restored. that FS and/or CO treatment stabilizes red color in tuna, mahi
However, color restoration for grade C steaks did not occur with mahi, and Spanish mackerel during frozen or refrigerated stor-
FS (20% CO; Kristinsson and others 2006c). A significant (P < age, research has indicated that redness faded upon aerobic storage
0.05) decrease in a∗ was observed for the UT product during all albeit at a much slower rate that what occurs in untreated fish
storage temperatures evaluated in this study. This initial discol- (Kristinsson and others 2006b). The color change with storage
oration difference may be due to the impact of oxidation of heme time did not occur in this study. As stated previously, the differ-
proteins due to cutting and freezing, with the color stabilized by ence seen in color stability between the current study and that
the treatment of tissue with FS (Anderson and Wu 2005). How- reported by other researchers could be attributed to treatment
ever, there were no significant (P > 0.05) changes in a∗ for treated methodology. Therefore, if applications of FS are not uniformly
tuna steaks during any of the storage temperatures evaluated. applied and different procedures are used, then assurances of the
Discoloration in stored meat is generally attributed to the quality profiles of fish, as mirrored by color changes, may not be
change in the myoglobin redox status (Richards and Hultin 2002; the same.

Table 1–Changes of lightness (L∗ ) color value in untreated and filtered smoked tuna steaks stored and sampled at 3 different time/temperature
conditions: room temperature (22 ◦ C) sampled hourly, refrigerated (4 to 5 ◦ C) sampled daily, and refrigerated on ice (0 ◦ C) sampled daily.

S: Sensory & Food

Untreated Tuna Filtered Smoked Tuna
Sampling time/storage temperature Sampling time/storage temperature

Quality
Hour/22 ◦ C Day/4 to 5 ◦ C Day/0 ◦ C Hour/22 ◦ C Day/4 to 5 ◦ C Day/0 ◦ C
0a 32.3 ± 2.7 0 32.3 ± 2.7 0 32.3 ± 2.7 0a 32.2 ± 1.9 0 32.2 ± 1.9 0 32.2 ± 1.9
6 39.6 ± 1.4 1 33.3 ± 2.0 3 35.4 ± 4.2 6 36.8 ± 4.7 1 31.3 ± 1.1 3 32.2 ± 1.5
12 33.3 ± 2.1 2 31.9 ± 1.6 6 37.4 ± 2.8 12 33.9 ± 1.9 2 31.8 ± 0.7 6 31.8 ± 0.7
18 31.5 ± 1.3 3 35.5 ± 3.1 9 36.7 ± 3.8 18 34.0 ± 1.7 3 34.3 ± 1.4 9 34.5 ± 0.7
26 36.3 ± 2.8 4 33.3 ± 3.1 12 33.8 ± 2.0 26 34.0 ± 1.5 4 31.9 ± 0.5 12 32.0 ± 1.0
32 36.1 ± 0.8 15 34.4 ± 2.8 32 35.5 ± 3.1 15 34.4 ± 2.9
40 35.3 ± 3.1 40 36.2 ± 1.4
Significant difference (P < 0.05) in lightness during storage
YES NO NO NO NO NO
a
Time zero lightness (L ∗ ) color values between untreated and filtered smoke tuna samples were not different (P > 0.05).

Table 2– Changes of yellowness (b∗ ) color value in untreated and filtered smoked tuna steaks stored and sampled at 3 different
time/temperature conditions: room temperature (22 ◦ C) sampled hourly, refrigerated (4 to 5 ◦ C) sampled daily, and refrigerated on
ice (0 ◦ C) sampled daily.
Untreated tuna Filtered smoked tuna
Sampling time/storage temperature Sampling time/storage temperature
Hour/22 ◦ C Day/4 to 5 ◦ C Day/0 ◦ C Hour/22 ◦ C Day/4 to 5 ◦ C Day/0 ◦ C
0a 2.1 ± 1.5 0 2.1 ± 1.5 0 2.1 ± 1.5 0a 1.0 ± 0.3 0 1.0 ± 0.3 0 1.0 ± 0.3
6 5.1 ± 2.7 1 1.7 ± 0.4 3 2.0 ± 0.4 6 1.9 ± 0.7 1 2.6 ± 1.0 3 1.6 ± 0.3
12 3.0 ± 1.5 2 1.7 ± 0.7 6 2.4 ± 0.3 12 1.1 ± 0.6 2 2.0 ± 0.8 6 2.1 ± 0.7
18 2.2 ± 1.1 3 3.2 ± 1.3 9 4.7 ± 2.1 18 1.6 ± 0.5 3 3.1 ± 1.8 9 2.8 ± 0.4
26 2.5 ± 1.1 4 1.7 ± 0.3 12 4.0 ± 0.7 26 1.1 ± 0.6 4 1.9 ± 1.5 12 2.5 ± 0.1
32 2.9 ± 0.8 15 2.6 ± 0.4 32 1.5 ± 0.9 15 2.4 ± 1.1
40 2.2 ± 0.8 40 1.0 ± 0.5
Significant difference (P < 0.05) in yellowness during storage
NO NO NO NO NO YES
a
Time zero yellowness (b∗ ) color values between untreated and filtered smoke tuna samples were not different (P > 0.05).

Vol. 76, Nr. 6, 2011 r Journal of Food Science S377

 Assessment of filtered smoked tuna. . .

Table 3–Changes of redness (a∗ ) color value in untreated and filtered smoked tuna steaks stored and sampled at 3 different time/temperature
conditions: room temperature (22 ◦ C) sampled hourly, refrigerated (4 to 5 ◦ C) sampled daily, and refrigerated on ice (0 ◦ C) sampled daily.
Untreated tuna Filtered smoked tuna
Sampling time/storage temperature Sampling time/storage temperature
Hour/22 ◦ C Day/4 to 5 ◦ C Day/0 ◦ C Hour/22 ◦ C Day/4 to 5 ◦ C Day/0 ◦ C
0 a
5.1 ± 1.2 0 5.1 ± 1.2 0 5.1 ± 1.2 0 a
9.7 ± 1.4 0 9.7 ± 1.4 0 9.7 ± 1.4
6 5.5 ± 1.5 1 4.3 ± 0.8 3 3.3 ± 0.7 6 11.3 ± 0.2 1 11.2 ± 3.7 3 9.3 ± 0.5
12 4.0 ± 0.1 2 2.7 ± 0.7 6 3.0 ± 0.5 12 9.9 ± 0.5 2 10.6 ± 1.3 6 10.3 ± 2.3
18 3.6 ± 0.8 3 3.9 ± 0.4 9 3.6 ± 0.9 18 10.6 ± 1.1 3 11.1 ± 2.9 9 9.8 ± 1.5
26 3.1 ± 1.2 4 3.1 ± 0.2 12 3.4 ± 0.3 26 9.2 ± 0.7 4 9.3 ± 2.6 12 10.4 ± 0.5
32 2.7 ± 0.7 15 2.5 ± 0.5 32 10.5 ± 0.2 15 10.0 ± 0.2
40 3.4 ± 0.3 40 10.0 ± 1.1
Significant difference (P < 0.05) in redness during storage
YES YES YES NO NO NO
a
Time zero redness (a∗ ) color values between untreated and filtered smoke tuna samples were significantly different at P < 0.05.

Conclusions smoke as assessed by color machine vision and lipid oxidation. IFT annual meeting, July
12–16, 2004, Las Vegas, Nev. Abstract 63–8. Chicago, Ill.: Institute of Food Technologist.
Microbial, chemical and expert sensory analyses all showed sim- Faustman C, Cassens RG. 1990. The biochemical basis for discoloration in fresh meat: a review.
ilar patterns of quality and spoilage changes during storage for J Muscle Foods 1:217–43.
Faustman C, Phillips A. 2001. Measurement of discoloration in fresh meat. In: Current protocols
UT- and FS-treated tuna. The impact on enhanced lipid stability in food analytical chemistry. N.Y.: Wiley and Sons Inc.
through decreased oxidation rates in FS-treated samples would be Faustman C, Cassens RG, Schaefer DM, Buege DR, Williams SN, Sheller KK. 1989. Im-
provement of pigment and lipid stability in Holstein steer beef by dietary supplementation of
important in the protection of the omega-3 fatty acids, important vitamin E. J Food Sci 54:858–62.
to health. However, while color, as determined by redness, faded Faustman C, Sun Q, Mancini R, Surendranath PS. 2010. Myoglobin and lipid oxidation inter-
actions: mechanistic bases and control. Meat Sci 86:86–94.
during storage of UT tuna and paralleled the laboratory and sen- FDA. 1995. Bacteriological analytical manual. Gaithersburg, Md.: AOAC Int.
sory assessments, there were no significant changes in red color Garner K, Kristinsson HG. 2004. Quality of Spanish mackerel (Scomberomorous maculates) muscle
in the FS tuna over the storage periods, regardless of tempera- as affected by carbon monoxide and filtered smoke gas treatment. IFT annual meeting, July
12–16, 2004, Las Vegas, Nev. Abstract 49B-12. Chicago, Ill.: Institute of Food Technologist.
ture, even if all other quality profile indicators revealed a declining Greene BE. 1969. Lipid oxidation and pigment changes in raw beef. J Food Sci 34:110–3.
product. Furthermore, in this study, FS treatment did not in- Greene BE, Hsin I, Zipser MW. 1971. Retardation of oxidative color change in raw ground
beef. J Food Sci 36:940–4.
crease the shelf life of the tuna when stored at refrigerated or [H.R. 3/26/2009] U.S. House of Representatives Committee on Energy and Commerce:
iced temperatures. Therefore, a general statement indicating that May 26, 2009 Summary of discussion draft of the Food Safety Enhancement Act of
2009 [Internet]. Washington, D.C. Available from: http://energycommerce.house.gov/
all FS treatments would extend shelf life or depress initial micro- Press_111/20090526/fsea_summary.pdf. Accessed 2010 August 3.
bial growth would not be appropriate. Finally, the lack of color [H.R. July 2009] U.S. House of Representatives Committee on Energy and Com-
merce: July 2009 Summary of H.R. 2749, The Food Safety Enhancement Act of
change, a primary sensory assessment used by consumers when 2009 [Internet]. Washington, D.C. Available from: http://energycommerce.house.gov/
purchasing this product, confirms that FS application could mask Press_111/20090729/hr2749_floorsummary.pdf. Accessed 2010 September 21.
S: Sensory & Food

ISO. 1993. Sensory analysis: general guidance for the selection, training and monitoring of
a lower quality product in the marketplace. assessors. Part 2: experts. ISO 8586–2.
Jay JM. 2000. Low-temperature food preservation and characteristics of psychrotrophic mi-
Quality

crooganisms. In: Jay JM, editor. Modern food microbiology. 6th ed. Md.: Aspen Publishers
Acknowledgments Inc. p 323–39.
This study was funded by a grant from National Oceanic and Kowalski WR. 1999 Oct 26. Process for manufacturing tasteless super-purified smoke for treating
seafood to be frozen and thawed. U.S. patent 5,972,401.
Atmospheric Administration National Marine Fisheries Service Kristinsson HG, Balaban MO, Otwell WS. 2006a. Microbial and quality consequences of aquatic
(NOAA/NMFS) under grant no. NA16FD2301. The project was foods treated with carbon monoxide and filtered wood smoke. In: Otwell WS, Kristinsson
HG, Balaban MO, editors. Modified atmospheric processing and packaging of fish: filtered
assigned contribution no. 5254 by the U.S. Dept. of Agricul- smokes, carbon monoxide and reduced oxygen packaging. Ames, Iowa: Blackwell Publishing.
ture at the Univ. of Rhode Island. This research was supported p 65–86.
by the Univ. of Rhode Island/Agricultural Experiment Station Kristinsson HG, Balaban MO, Otwell WS. 2006b. The influence of carbon monoxide and
filtered wood smoke on fish muscle color. In: Otwell WS, Kristinsson HG, Balaban MO,
(URI/AES) and RI Sea Grant. editors. Modified atmospheric processing and packaging of fish: filtered smokes, carbon
monoxide and reduced oxygen packaging. Ames, Iowa: Blackwell Publishing. p 29–52.
Kristinsson HG, Balaban MO, Otwell WS. 2006c. Color enhancement and potential fraud in
using CO. In: Otwell WS, Kristinsson HG, Balaban MO, editors. Modified atmospheric
References processing and packaging of fish: filtered smokes, carbon monoxide and reduced oxygen
Anderson CR, Wu W-H. 2005. Analysis of carbon monoxide in commerically treated tuna packaging. Ames, Iowa: Blackwell Publishing. p 127–40.
(Thunnus spp.) and mahi-mahi (Coryphaena hippurus) by gas chromatography/mass spectrom- Kristinsson HG, Danyali N, Ua-Angkoon S. 2007. Effect of filtered wood smoke treatment on
etry. J Agric Food Chem 53:7019–23. chemical and microbial changes in mahi mahi fillets. J Food Sci 72:C16–24.
[AOAC] Assoc. of Official Analytical Chemists. 1990. AOAC official method 966.23. Chapter Kristinsson HG, Crynen S, Yagiz Y. 2008. Effect of a filtered wood smoke treatment compared
17. In: Helrich K, editor. Official methods of analysis of AOAC Intl., 15th ed. Arlington, Va.: to various gas treatments on aerobic bacteria in yellowfin tuna steaks. LWT 41:746–50.
AOAC Intl. Lee S, Joo ST, Alderton AL, Hill DW, Faustman C. 2003. Oxymyoglobin and lipid oxidation
[AOAC]. Assoc. of Official Analytical Chemists. 1995. AOAC official method 980.10. In: in yellowfin tuna (Thunnus albacares) loins. J Food Sci 68:1664–8.
Cunniff P, editor. Official methods of analysis of AOAC Intl., 16th ed. Gaithersburg, Md.: Ledafish. Info tuna quality. Available from: http://ledafish.com. Accessed 2011 May 18.
AOAC Intl. Leydon NL, Suman SP, Ellis PC, Palmer C, Faustman LC, Pivarnik LF. 2005. Quality and safety
[AOAC]. Assoc. of Official Analytical Chemists. 2000. AOAC official method 999.01. Chapter assessment of time/temperature storage conditions for filtered smoked tilapia fillets produced
35. In: Horwitz W, editor. Official methods of analysis of AOAC Intl., 17th ed. Gaithersburg, for retail distribution. In: IFT National Meeting, July 16–20, New Orleans, La: Institute of
Md.: AOAC Intl. Food Technologists, Chicago, Ill. Abstract. p 89A–16.
Banwart GJ. 1989. Basic food microbiology. New York: Van Nostrand Reinhold. 771 p. Livingston DJ, Brown WD. 1981. The chemistry of myoglobin and its reactions. Food Technol
Bartram PK. 1996. Proper processing and handling of raw tuna. In: Martin RE, Collette RL, 5:244–52.
Slavin JW, editors. Fish inspection quality control and HACCP: a global focus. New York: Ludlow N, Kristinsson HG, Balaban MO, Otwell WS, Welt B. 2004. Effect of different carbon
Taylor & Francis Group. p 233–7. monoxide and filtered smoke treatments on the quality and safety of yellowfin tuna (Thunnus
Chan WKM, Joo S-T, Faustman C, Sun Q, Vieth R. 1998. Effect of Pseudomonas fluorescens on albacares) muscle. IFT annual meeting, July 12–16, 2004, Las Vegas, Nev. Abstract 49B-26.
beef discoloration and oxymyoglobin oxidation in vitro. J Food Prot 61:1341–6. Chicago, Ill.: Institute of Food Technologists.
Chow CJ, Ochiai Y, Watabe S. 2004. Effect of frozen temperature on autoxidation and aggre- Malle P, Tao SH. 1987. Rapid quantitative determination of trimethylamine using steam distil-
gation of bluefin tuna myoglobin in solution, J Food Biochem 28:123–34. lation. J Food Protect 50:756–60.
Demir N, Kristinsson HG, Balaban MO, Otwell WS. 2004. Quality changes in mahi mahi Malle P, Poumeyrol M. 1989. A new chemical criterion for the quality control of fish: trimethy-
(Coryphaena hippurus) fillets treated by different carbon monoxide concentrations and filtered lamine/total volatile base nitrogen (%). J Food Protect 52:419–23.

S378 Journal of Food Science r Vol. 76, Nr. 6, 2011

Assessment of filtered smoked tuna. . .

Pivarnik LF, Thiam M, Ellis PE. 1998. Rapid determination of volatile bases in fish by using an Sims GG, Farn G, York RK. 1992. Quality indices for canned skipjack tuna: correlation of
ammonia ion-selective electrode. JAOAC Int 81:1011–22. sensory attributes with chemical indices. J Food Sci 57:1112–5.
Pivarnik LF, Ellis PC, Wang X, Reilly T. 2001. Standardization of ammonia electrode method Sohn JH, Taki Y, Ushio H, Kohata T, Shioya I, Ohshima, T. 2005. Lipid oxidations in ordinary
for evaluation seafood quality by correlation to sensory analysis. J Food Sci 66:945–52. and dark muscles of fish: influences on rancid off-odor development and color darkening of
Poste LF, Mackie DA, Butler G, Larmond E. 1991. Laboratory methods for sensory analysis of yellowtail flesh during ice storage. J Food Sci 70:490–6.
food. In: Research Branch, Agriculture Canada, Publication 1864/E. Thiansilakul Y, Richards M, Benjakul S. 2011. Isolation, characterisation and stability of
Reilly TI, York RK. 1993. Sensory analysis application to harmonization of expert assessors myoglobin from Eastern little tuna (Euthynnus affinis) dark muscle. Food Chem 124:254–
of fish products. Abstract. Presented at Quality Control and Quality Assurance for Seafood. 61.
Newport, Oreg., May 16–18. Undeland I, Kristinsson HG, Hultin HO. 2004. Hemoglobin-mediated oxidation of waswhed
Richards MP, Dettmann MA. 2003. Comparative analysis of different hemoglobins: au- minced cod muscle phospholipids: effect of pH and hemoglobin source. J Agric Food Chem
toxidation, reaction with peroxide, and lipid oxidation. J Agric Food Chem 51: 52:4444–51.
3886–91. Venkitanarayanan KS, Faustman C, Hoagland T, Berry BW. 1997. Estimation of spoilage bacte-
Richards MP, Hultin HO. 2002. Contributions of blood and blood components to lipid oxida- rial load in aerobically stored meat by fluorescein diacetate hydrolysis and resazurin reduction.
tion in fish muscle. J Agric Food Chem 50:555–64. J Food Sci 62:601–4.
Shantha, NC, Decker EA. 1994. Rapid, sensitive, iron-based spectrophotometric methods for Yin MC, Faustman C, Reisen JW, Williams SN. 1993. α-Tocopherol and ascorbate delay
determination of peroxide values of food lipids. JAOAC Int 77:421–4. oxymyoglobin and phospholipid oxidation in vitro. J Food Sci 58:1273–81.

S: Sensory & Food

Quality

Vol. 76, Nr. 6, 2011 r Journal of Food Science S379