980

Journal of Food Protection, Vol. 82, No. 6, 2019, Pages 980–987

https://doi.org/10.4315/0362-028X.JFP-18-423
Not subject to U.S. Copyright

Research Paper

Thermal Inactivation of Salmonella in Pâté Made from

Chicken Liver
ANNA C. S. PORTO-FETT,* BRADLEY A. SHOYER, LAURA E. SHANE, MANUELA OSORIA, ELIZABETH HENRY,
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

YANGJIN JUNG, AND JOHN B. LUCHANSKY

U.S. Department of Agriculture, Agricultural Research Service, 600 Mermaid Lane, Wyndmoor, Pennsylvania 19038, USA
MS 18-423: Received 14 September 2018/Accepted 28 January 2019/Published Online 21 May 2019

ABSTRACT
The effect of heating times and temperatures on inactivation of Salmonella in pâté made from chicken liver was evaluated.
Raw chicken liver (ca. 1 kg) was blended in a food processor with two hard-boiled eggs (ca. 100 g total) plus a mixture of
sautéed white onions (100 g), salt (5.0 g), black pepper (2.5 g), and butter (112 g). The tempered (ca. 158C) raw pâté batter was
inoculated with a nine-strain cocktail (ca. 6.5 log CFU/g) of Salmonella, and then ca. 25-g portions were aseptically transferred
into sterile 50-mL polypropylene conical tubes. One set of tubes was completely submerged in a thermostatically controlled,
Journal of Food Protection 2019.82:980-987.

circulating water bath set at 74.98C and cooked to target instantaneous internal temperatures ranging from 60 to 73.98C. An
otherwise similar set of tubes was cooked at 60 to 73.98C, with holding times of 3 to 30 min, in a water bath set at 18C above
each target endpoint cooking temperature. Regardless of the cooking process, when pâté was cooked to a target instantaneous
internal temperature of 60 to 73.98C, pathogen numbers decreased by ca. 1.9 to
6.4 log CFU/g; additional reductions of ca. 0.6
to 1.3 log CFU/g were observed when pâté was cooked to a target instantaneous internal temperature of 60 to 688C and then held
for 3 to 30 min. In related experiments, pâté was prepared as described above, but with inoculated chicken livers (500 g; ca. 5.5
log CFU/g) that were cooked in a frying pan maintained at ca. 1408C for 3 to 8 min with a mixture of sautéed onions, salt, black
pepper, and butter and then blended with a hard-boiled egg. Pathogen numbers within liver cooked in a frying pan decreased by
ca. 1.0 to 4.9 log CFU/g. Collectively, these findings may be useful for establishing cooking guidelines for pâté and thus for
lowering the risk of illness if chicken liver is contaminated with Salmonella and the attendant batter is not handled or cooked
properly.

HIGHLIGHTS
 Cooking may reduce the potential risk of salmonellosis associated with liver pâté.
 A 5-log reduction was achieved when inoculated pâté was cooked to an internal temperature of
73.88C.
 A 5-log reduction was achieved when pâté was made with inoculated liver fried for .8 min at 1408C.
 Findings of this study may be useful for establishing cooking guidelines for liver and pâté.

Key words: Chicken liver; Pâté; Ready-to-eat meats; Salmonella; Specialty meats; Thermal inactivation

Pâté, a traditional European charcuterie, is made with a target instantaneous internal temperature of 73.98C
finely or coarsely minced meats and fat, along with a variety (1658F) to reduce the risk of foodborne salmonellosis (28,
of seasonings and spices. This spreadable meat mixture has 41). There is a growing global culinary trend toward
a texture that can vary from smooth and creamy to dense preparing liver pâté by chefs at restaurants and by
and firm or to coarse and chunky (23, 34). Pâté may be consumers at home using an extensive variety of recipes
served warm, refrigerated, or at room temperature, and it available both on social media applications (i.e., Web sites
may be eaten as an hors d’oeuvre, entree, or component of and blogs) and in print (i.e., magazines and cookbooks).
other dishes. The most traditional pâtés are made with liver Regrettably, many of these recipes endorse the use of low
from poultry, such as chicken, duck, or goose, and are temperatures and short cooking times, as well as cooking
prepared by either sautéing the liver before blending with liver or pâté to a rare degree of doneness (i.e., pink) for
other ingredients or by immersing the blended raw liver and preferred quality attributes, thus circumventing cooking
spice mixture in a water bath (i.e., bain-marie) for a pâté to the required internal temperature of 73.98C. A 2015
predetermined cooking time and temperature. As for any survey conducted by Jones et al. (20) in the United
other poultry products, chicken liver pâté must be cooked to Kingdom reported that 47.8% (492 of 1,030) of consumers
surveyed said that cooking shows on TV, recipes in
* Author for correspondence. Tel: 215-836-3762; Fax: 215-233-6406; magazines, or both contributed directly to how they cooked
E-mail: anna.portofett@ars.usda.gov. and served meat, whereas ca. 45% (64 of 143) of chefs
 J. Food Prot., Vol. 82, No. 6 FATE OF SALMONELLA IN CHICKEN LIVER PÂTÉ 981

surveyed reported a trend toward consumption of chicken and chicken liver products may pose an appreciable threat to
livers that were rarer and more pink as prepared on TV consumer health. However, to our knowledge, there is no
shows. Although undercooked pâté prepared from chicken information in the literature related to cooking times and
liver is recognized worldwide as a major vehicle for temperatures needed to eliminate Salmonella from chicken
transmission of foodborne pathogens such as Salmonella liver pâté. Hence, we validated cooking practices to reduce
and Campylobacter, and as such, has been specifically the potential risk of salmonellosis associated with under-
incriminated in several outbreaks and recalls (7, 15, 24), cooked or poorly handled chicken liver and pâté made
consumers still prefer and chefs continue to prepare pâté therefrom.
that is often undercooked (7, 12, 15, 24, 25, 29, 31).
In the United States, the number of outbreaks and MATERIALS AND METHODS
illnesses attributed to chicken liver and chicken liver pâté Bacterial strains. Approximately equal cell numbers of
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

has risen greatly over the past 2 decades. Indeed, between rifampin-resistant (100 μg/mL; TCI America, Portland, OR)
2000 and 2017 consumption of undercooked chicken liver, derivatives of the following nine Salmonella strains were used
pâté, or both made from chicken liver resulted in 28 to inoculate raw liver: (i) Salmonella Heidelberg WA17839
foodborne outbreaks, with 64% (18 of 28) of these (JF6X01.0122, clinical isolate), (ii) Salmonella Heidelberg
outbreaks occurring between 2014 and 2016 (7, 24). Of WA17853 (JF6X01.0122, food isolate), (iii) Salmonella Heidel-
these 28 outbreaks, 3 (10.7%) outbreaks were caused by berg 2011K-1224 (JF6X01.0058, ground turkey isolate), (iv)
Salmonella and 2 (7.1%) outbreaks were caused by both Salmonella Hadar CO-2955, (v) Salmonella Heidelberg FY-14-6
Salmonella and Campylobacter (2, 7, 24). Undercooked (JF6X01.0045, chicken parts isolate), (vi) Salmonella FSIS
liver and liver products such as pâté, mousse, or meat Control 1, (vii) Salmonella FSIS Control 2, (viii) Salmonella
spreads were identified as the likely vehicle for the above- Heidelberg FSIS11705638 (chicken liver isolate), and (ix)
Salmonella Heidelberg FSIS11705639 (chicken liver isolate).
mentioned outbreaks (7, 24, 39). Among these outbreaks,
Journal of Food Protection 2019.82:980-987.

These strains were kindly provided by Drs. John Johnston

the largest outbreak of salmonellosis (190 cases, no deaths)
(USDA-FSIS, Office of Public Health Science, Fort Collins,
was associated with the consumption of ‘‘not fully cooked’’
CO) and Glenn Tillman (USDA-FSIS, Office of Public Health
chicken liver that was labeled as ‘‘broiled’’ and, as such, Science, Eastern Laboratory, Microbiology Characterization
mislead consumers to eat the product without prior cooking Branch, Athens, GA). Strains were confirmed, cultured, and
(7). Of note, besides partial cooking being a causative factor maintained as described previously (32).
in the above-mentioned outbreaks, consumption of contam-
inated and undercooked liver or chicken liver products in Inoculation and cooking of pâté. For each cooking regimen
commercial and institutional establishments, such as tested, one set of pâté was prepared by blending (ca. 2 min) fresh
nursing homes, was also a common characteristic among raw chicken liver (ca. 1 kg; 48C) in a 2.4-L (10-cup) food
these outbreaks (24, 39). processor (model 70723, Hamilton Beach, Glen Allen, VA) with
Salmonella is widespread in the environment and two hard-boiled eggs (ca. 100 g total; boiled in water for 8 to 10
common among domestic animals and livestock, as well min) plus a mixture of sautéed white onions (100 g), salt (5.0 g;
as within food processing plants and associated food SaltWorks Inc., Woodinville, WA), black pepper (2.5 g; The
products. Poultry meat, including giblets such as the liver Sausage Maker, Buffalo, NY), and salted butter (112 g; Land
O’Lakes Inc., Arden Hills, MN). The batter was cooled to ca. 158C
and gizzard, is a major source of this pathogenic bacterium.
within about 30 min and then inoculated with the nine-strain
Along with the intestinal tract, internal organs such as liver,
cocktail (1 mL of inoculum to 100 g of pâté batter; ca. 6.5 log
gallbladder, spleen, and ceca are reservoirs of Salmonella in CFU/g) of Salmonella. The inoculum was distributed manually
chickens (8, 13, 17). In fact, the global incidence of throughout the batter with the aid of a sterile spatula. Next, ca. 25-
Salmonella in raw chicken liver purchased at retail can be as g portions were aseptically transferred into separate sterile 50-mL
low as 4.8% and as high as 80% (1, 4, 6, 8, 9, 33, 38, 46). In screw-cap conical polypropylene centrifuge tubes (Fisherbrand
the United States, the 2012 U.S. Department of Agriculture, Easy Reader, Thermo Fisher Scientific, Hampton, NH). The tubes
Food Safety and Inspection Service (USDA-FSIS) Raw were completely submerged in a thermostatically controlled,
Chicken Parts Baseline Survey reported a prevalence of circulating water bath (model 8202, Thomas Scientific, Swedes-
Salmonella in giblets of 40.4% (23 of 57 samples) (40). To boro, NJ), a cooking ‘‘process’’ known as bain-marie, and cooked
better understand the risk of Salmonella and Campylobacter to (i) target instantaneous internal temperatures of 60, 63, 65, 68,
associated with specific raw chicken parts such as the neck, 71.1, or 73.98C in a water bath that was stabilized at 74.98C or to
liver, heart, and gizzard, the Other Raw Chicken Parts (ii) target instantaneous internal temperatures of 60, 63, 65, 68,
Sampling Project was conducted by the USDA-FSIS (42); 71.1, or 73.98C in a water bath stabilized at 18C above each target
53.9% (172 of 319 samples) of these samples were cooking temperature; for internal temperatures of 60, 63, 65, and
688C, pâté was also held for 30, 15, 8, and 3 min, respectively. The
contaminated with Salmonella (44). Recently, Jung and
come-up times were measured by inserting a J-type thermocouple
colleagues (22) reported that 59.4% (148 of 249 samples) of
(Omega Engineering Inc., Stamford, CT) into the geometric center
raw chicken liver samples obtained over a 9-month period at of an otherwise similar polypropylene conical tube filled with ca.
retail markets across three states in the Mid-Atlantic region 25 g of noninoculated chicken liver pâté and recording the time for
of the United States tested positive for Salmonella; the batter to reach the target endpoint temperature from a starting
pathogen levels in samples testing positive ranged from temperature of ca. 8.58C. The thermocouple was separately
6.4 most probable number per gram to 2.4 log CFU/g. Thus, connected to a six-channel digital panel temperature indicator
given that poultry liver may harbor Salmonella either (model 500T, Doric Instruments, VAS Engineering Inc., San
internally, externally, or both, undercooked chicken liver Diego, CA). When the liver pâté achieved the target internal
 982 PORTO-FETT ET AL. J. Food Prot., Vol. 82, No. 6

temperature, tubes were either removed from the water bath or Cincinnati, OH). A 10-mL portion of the filtrate from each sample
held for up to 30 min (i.e., postcooking holding times) at the was transferred to a sterile 15-mL screw-cap conical polypropyl-
designated target endpoint temperatures. After cooking and ene centrifuge tube (Thermo Fisher Scientific) with the aid of a
holding, the tubes were removed from the water bath and sterile pipette. Pathogen numbers were determined by serial
immediately cooled in an ice-water bath for up to 30 min. For diluting the filtrate in 0.1% sterile peptone water as needed and
pâté cooked to target instantaneous internal temperatures of 60 to surface plating onto duplicate xylose lysine Tergitol 4 (Difco, BD)
73.98C in a water bath that was stabilized at 74.98C, in each of the agar plates plus rifampin (100 μg/mL; TCI America). Plates were
seven trials, three pâté samples were analyzed at each target incubated for 24 h at 378C. When pathogen levels decreased to
temperature tested (N ¼ 7 trials; n ¼ 3 pâté samples per below the detection limit (0.46 log CFU/g) by direct plating,
temperature tested per trial). For pâté cooked to target instanta- samples were enriched as described previously (32). Colonies
neous internal temperatures of 60 to 73.98C, with holding times in typical of Salmonella were counted, and numbers are expressed as
a water bath that was stabilized at 18C above each target endpoint log CFU per gram.
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

cooking temperature, in each of the four trials, three pâté samples

were analyzed at each target temperature tested (N ¼ 4 trials; n ¼ 3 Proximate composition analyses. Levels of fat, carbohy-
pâté samples per temperature tested per trial). drate, moisture, protein, ash, salt, pH, and water activity were
In related experiments, a similar set of pâté was prepared by determined in 300-g composite samples of pâté made from
sautéing white onions (50 g), salt (2.5 g; SaltWorks Inc.), black chicken liver that was cooked to a target instantaneous
pepper (1.25 g; The Sausage Maker), and salted butter (56 g; Land temperature of 73.98C, in two of the four trials (two samples
O’Lakes Inc.) for about 2 min in a nonstick frying pan (T-Fal total, 300 g each), in 300-g composite samples of artisanal pâté
C51905, Tefal, Rumilly, France) using a flat-surface electric made from chicken liver and in 300-g composite samples of
ceramic heating plate (Isotemp 11-100-49SH, Fisher Scientific, artisanal pâté made from beef liver; artisanal samples of pâté were
Pittsburgh, PA) set at ca. 2008C (3908F). Next, raw chicken livers obtained on two different visits to a farmers market in
(ca. 500 g) that were inoculated internally by separately injecting Pennsylvania. Analyses were determined by methods approved
Journal of Food Protection 2019.82:980-987.

100 μL of the Salmonella cocktail into three different locations and described in the Official Methods of Analysis of AOAC
(ca. 33 μL per location; ca. 5.5 log CFU/g) of each liver with the International (OMA) (3), and the tests were performed by a
aid of a 1-mL 25-gauge syringe (Difco, BD, Franklin Lakes, NJ) private testing laboratory.
were directly added to the sautéed mixture in the frying pan and
then cooked for 3, 4, 5, 6, 7, or 8 min. Livers were ‘‘flipped’’ once Statistical analyses. Means and standard deviations were
using an alcohol-sterilized, stainless steel fork at the approximate calculated from individual sets of data for each of the separate
midpoint between the initial and target endpoint cooking times. trials using the statistical function option that is provided with
Liver temperatures were recorded using a hand-held, Type K Excel 2010 software (Microsoft, Redmond, WA). The analysis of
digital thermocouple (range, 73 to 2608C [100 to 5008F]; variance (ANOVA) for pâté made from chicken liver cooked in a
accuracy, 60.58C [60.98F]; model AquaTuff Waterproof 351 water bath was analyzed as a two-factor mixed linear model using
digital thermocouple, Cooper-Atkins, Middlefield, CT). The Proc Mixed (SAS 14.3, SAS Institute Inc., Cary, NC), with
temperatures of three livers (n ¼ 3) were taken by separately temperature as the fixed factor and trial as the random factor. For
inserting the probe into the approximate geometric center of each pâté made from chicken liver cooked in a frying pan, ANOVA was
liver after flipping, as well as by separately inserting the probe into analyzed as a four-factor mixed model using Proc Mixed (SAS
three additional livers immediately after cooking when livers were Institute Inc.), with temperature and time as fixed factors, day as
removed from the frying pan (n ¼ 3). Also, the temperatures of the repeated factor, and trial as the random factor. Differences in
three additional livers (n ¼ 3) were measured after cooking and lethality observed for each preparation method and combinations
removal from the frying pan by cutting or splitting each liver in thereof were considered as significant using the Sidak method at a
half with the aid of an alcohol-sterilized, stainless-steel fork and P  0.05 significance level.
inserting the probe into the ‘‘cold spot’’ as perceived by visual
inspection (i.e., still pink). Within ca. 1 min after being removed RESULTS AND DISCUSSION
from the frying pan, the sautéed liver, sautéed onions, salt, pepper,
and butter mixture along with hard-boiled egg was transferred into Every year in the United States, ca. 1 million illnesses,
a food processor (model 70723, Hamilton Beach) with the aid of a ca. 19,000 hospitalizations, and ca. 400 deaths are attributed
sterile spatula and blended for ca. 1 min at room temperature. to Salmonella, with an average economic loss of ca. $3.7
Portions of the liver pâté were transferred from the food processor billion (5, 18, 35, 36). Poultry meat and poultry products,
into sterile weigh boats that were then covered with another weigh including liver, are a common reservoir of Salmonella. As
boat. Microbiological analyses were conducted immediately after evidenced by the 361 human illnesses and 46 hospitaliza-
blending and after storage at 4 or 218C for ca. 2 h to mimic tions attributed to consumption of undercooked chicken
consumers behavior of cooling pâté at room or refrigeration liver or pâté made from chicken liver between 2000 and
temperature, respectively, just before consumption. Other samples 2016 (24), there is a public health concern related to
were stored at 48C for 3 days to mimic consumer behavior for consumption of chicken liver pâté prepared (intentionally)
storing leftovers. In each of the four trials, three pâté samples were
to a rare degree of doneness based on consumer’s
analyzed at each cooking time (N ¼ 4 trials, n ¼ 3 pâté samples per
preference and request. Therefore, we validated cooking
sampling interval per trial).
practices to reduce the potential risk of salmonellosis
Microbiological analyses. Cells of Salmonella were recov- associated with the consumption of pâté made from chicken
ered from both uncooked and cooked pâté by adding 25 g of pâté liver.
and 25 mL of 0.1% sterile peptone water (Difco, BD) to a sterile Pâté made from chicken liver was formulated based on
filter bag (BagFilter S, Microbiology International, Frederick, a compilation of 12 recipes randomly selected from the
MD) and macerating for 2 min in a Stomacher 400 (Seward, Internet, using only common ingredients among the selected
 J. Food Prot., Vol. 82, No. 6 FATE OF SALMONELLA IN CHICKEN LIVER PÂTÉ 983

TABLE 1. Proximate composition of pâté made from chicken liver tested herein and of artisanal pâté made from chicken liver or beef
liver purchased at a local farmers market
Analysis Experimental chicken liver pâté Artisanal chicken liver pâté Artisanal beef liver pâté

Fat (%) 10.25a 6 1.30 8.49b 6 0.55 28.07b 6 0.71

Protein (%) 13.53 6 0.40 17.38 6 1.59 16.06 6 0.62
Moisture (%) 74.29 6 1.13 70.59 6 2.01 50.41 6 2.14
Carbohydrate (%) 0.20 6 0.14 1.34 6 0.17 2.27 6 2.14
Ash (%) 1.93 6 0.15 1.75 6 0.34 2.31 6 0.07
Salt (%) 0.58 6 0.02 0.90 6 0.25 1.75 6 0.06
Acidity (%) 0.72 6 0.30 1.51 6 0.27 0.93 6 0.29
Water activity 0.990 6 0.001 0.991 6 0.001 0.969 6 0.005
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

pH 5.94 6 0.40 4.79 6 0.42 5.34 6 0.45

a
Average of two trials. Proximate composition was determined in 300-g samples of pâté made from chicken liver in two of the four trials.
b
Average of two samples. Proximate composition was determined in 300-g samples of artisanal pâté made from chicken liver and of
artisanal pâté made from beef liver.

recipes: chicken liver, butter, onions, salt, black pepper, and moisture content (74.3%), and high fat level (10.3%).
hard-boiled eggs. The generic formulation tested herein Moreover, no appreciable differences in proximate compo-
may represent the worst-case scenario for thermal inactiva- sition were observed between the pâté made from chicken
tion of Salmonella in pâté, because other recipes call for liver tested herein and the artisanal pâté made from chicken
inclusion of ingredients such as alcoholic beverages (i.e., liver that was obtained at a farmers market; however,
Journal of Food Protection 2019.82:980-987.

bourbon), vinegar, herbs, and additional spices. Such significant differences in the levels of water activity (0.968),
additions to the base recipe may result in greater moisture (50.4%), and fat (28.1%) of artisanal pâté made
inactivation of Salmonella during cooking (19), since these from beef liver were observed compared with chicken liver
additional ingredients have intrinsic antimicrobial proper- pâté prepared in this study and with the artisanal pâ té made
ties. However, Hutchison et al. (19) reported no significant from chicken liver (Table 1). It is well established that the
differences on thermal inactivation of Campylobacter in thermal tolerance of Salmonella in foods can be critically
pâté prepared with liver that was not prewashed in impacted by the aforementioned intrinsic factors, although
comparison with liver that was prewashed with vinegar extrinsic factors such as product geometry and the location
(5%, v/v) or essential oils before cooking to a target internal where the cells are attached (e.g., wing joint versus breast or
temperature of 60 or 638C in a water bath. In addition, it is thigh) in the meat product may also affect thermal tolerance
noteworthy that 41.7% (5 of 12 recipes) of the recipes of the pathogen (11, 14, 21, 27, 28, 30). In meat products,
compared herein recommended cooking liver to a rare fat content plays a major role in thermal inactivation of
degree of doneness (i.e., pink, rosy, bloody) as determined Salmonella due to the lower heat conductivity and reduced
by color rather than temperature. None of the recipes water content of the fat (30, 37). For example, studies
provided instructions to cook liver to a degree of doneness reported that thermal resistance of Salmonella in ground
based on achieving a target internal temperature; the recipes turkey and ground chicken was greater in products with
only suggested cooking liver ‘‘thoroughly’’ based on times higher fat content (12 versus 2% fat) at temperatures
that varied from 5 to 10 min. Likewise, Hutchison and ranging from 58 to 748C, and that longer D-values were
colleagues (19) assessed 40 recipes of chicken liver pâté observed for chicken skin (47.4% fat) than those observed
selected from books, Web sites, and blogs and reported that for ground chicken meat (10.3% fat) at temperatures
22.5% (9 of 40 recipes) clearly stated to cook liver to a rare ranging from 55 to 708C (21, 27). Thus, the composition
degree of doneness (i.e., ‘‘still pink inside’’), with only 1 of chicken liver pâté may also provide a favorable
recipe (2.5%) stating to cook liver to a specified internal environment for the survival of Salmonella or may enhance
temperature, with that being 708C. Godwin et al. (16) also the thermal tolerance of Salmonella during cooking since it
reported that among 242 recipes used for preparing egg has a fat content of ca. 10% (Table 1). Because chicken liver
dishes, 92% (222 of 242 recipes) of the recipes provided a is a common reservoir of Salmonella, consumption of pâté
time or a range of times as the indicator of doneness, rather that is improperly prepared or handled may result in illness,
than a required endpoint internal temperature (i.e., 71.18C). especially since relatively few cells may cause illness for
Collectively, these studies demonstrated that consumer some people (10, 26).
education about proper use of a food thermometer for Regardless of the process, higher cooking temperatures
cooking is needed to lower the public health risk associated or longer come-up times or extended cooking and holding
with consumption of undercooked poultry. times resulted in greater inactivation of Salmonella in
The proximate compositions of the pâté made from the chicken liver pâté. For example, cooking pâté to a target
chicken liver tested herein, as well as artisanal pâté made temperature of 60 to 688C in a water bath stabilized at 18C
from chicken liver and artisanal pâté made from beef liver, (Fig. 1) higher than the target cooking temperature was
are presented in Table 1. Proximate compositional analyses significantly (P , 0.05) more effective at inactivating cells
established that the pâté prepared in this study was of a of Salmonella than cooking pâté at 60 to 688C in a water
relatively high pH (pH 5.9), high water activity (0.99), high bath stabilized at 74.98C (Fig. 2). This difference in lethality
 984 PORTO-FETT ET AL. J. Food Prot., Vol. 82, No. 6

FIGURE 1. Thermal inactivation of Sal-

monella in chicken liver pâté cooked in a
water bath set at 18 C above the target
cooking temperature (N ¼ 4, n ¼ 3). Error
bars represent the standard deviation of
the mean. The numbers following the
cooking temperature on the x axis are the
times in minutes that samples were main-
tained at the indicated target endpoint
temperature. Means with different letters in
a column are significantly (P  0.05)
different by the Sidak test. For a given
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

cooking temperature, means with an aster-

isk (*) are significantly (P  0.05) different
by the Sidak test.

(ca. 0.9 to 2.1 log CFU/g) between both procedures may be pâté was cooked to a target internal temperature of 60 to
attributed to come-up times; pâté cooked in a water bath 688C and held for 3 to 30 min.
stabilized at 74.98C reached the target internal temperature For pâté that was prepared with inoculated livers that
Journal of Food Protection 2019.82:980-987.

in less time (5.8 to 19.5 min) than when cooked in a water were cooked with butter, onions, salt, and pepper in a frying
bath stabilized at 18C above each target cooking tempera- pan maintained at 139 6 4.78C for 3 to 8 min before
ture (19.5 to 21.8 min). Of note, differences in lethality were blending, as expected, greater inactivation (P , 0.0001) of
not observed (P . 0.05) for pâté cooked to a target internal Salmonella was observed in pâté made from liver cooked
temperature of 71.1 and 73.98C when cooked in a water for longer times (Fig. 3). Specifically, Salmonella numbers
bath stabilized at 74.98C or for pâté cooked in a water bath were reduced by ca. 1.0, 1.5, 3.2, 3.5, 4.9, and 4.6 log CFU/
stabilized at 18C above a target endpoint cooking temper- g in pâté cooked for 3, 4, 5, 6, 7, and 8 min, respectively
(Fig. 2). Cooking inoculated liver plus other pâté ingredi-
ature of 73.98C. These data were expected because the
ents for 3 to 8 min in a frying pan and then blending with
temperature of the water bath was 74.98C for both heating
hard-boiled egg followed by storage for 2 h at 218C allowed
regimens. The results showed that for pâté cooked to target
for an increase in pathogen levels of ca. 0.1 to 0.8 log CFU/
instantaneous internal temperatures of 60, 63, 65, 68, 71.1, g. Likewise, similarly treated pâté stored for 2 h at 48C
and 73.98C in a water bath stabilized at 74.98C, pathogen allowed for an increase of ca. 0.1 to 0.5 log CFU/g in
numbers decreased by ca. 0.3, 0.6, 1.2, 2.6, 4.9, and
6.2 pathogen numbers. In contrast, after 3 days of storage at
log CFU/g, respectively. Likewise, when pâté was cooked 48C, pathogen numbers remained relatively stable (ca. 0.1
to target instantaneous internal temperatures of 60, 63, 65, log CFU/g change). The average internal temperature of the
68, 71.1, and 73.98C in a water bath stabilized at 18C above sautéed chicken liver at flipping was 21.9 6 11.2, 40.4 6
each target cooking temperature, pathogen numbers were 12.2, 32.7 6 17.6, 43.3 6 11.9, 47.2 6 10.4, and 47.8 6
reduced by ca. 1.9, 2.5, 3.8, 5.2, 6.0, and
6.2 log CFU/g, 7.68C for livers cooked for 3, 4, 5, 6, 7, and 8 min,
respectively. Additional reductions (P , 0.05) in pathogen respectively. The average internal temperature of the
numbers of ca. 0.6 to 1.3 log CFU/g were observed when sautéed chicken liver immediately after cooking for 3, 4,

FIGURE 2. Thermal inactivation of Sal-

monella in chicken liver pâté cooked in a
water bath set at 74.98 C (N ¼ 7, n ¼ 3).
Error bars represent the standard devia-
tion of the mean. Means with different
letters in a column are significantly (P 
0.05) different by the Sidak test.
 J. Food Prot., Vol. 82, No. 6 FATE OF SALMONELLA IN CHICKEN LIVER PÂTÉ 985
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.
Journal of Food Protection 2019.82:980-987.

FIGURE 3. Thermal inactivation of Salmonella in chicken liver pâté prepared with liver cooked for 3 to 8 min in a frying pan using a
single flip. Error bars represent the standard deviation of the mean (N ¼ 4, n ¼ 3). Livers were inoculated, pan-fried along with other
ingredients, and then blended in a food processor with the hard-boiled egg component of the recipe. Portions of the blended pâté were
then stored at 48 C for 2 h or 3 days or stored at 218 C for 2 h.

5, 6, 7, and 8 min was 43.5 6 10.5, 57.1 6 12.5, 67.9 6 batter was more uniform and reproducible than that
12.2, 75.6 6 14.1, 70.2 6 2.5, and 81.6 6 8.48C, observed for otherwise similar livers that were pan fried
respectively. Regardless of the cooking time, cold or pink before blending (19, 45).
spots were observed for all cooking times tested. The To our knowledge, there is a lack of published
average temperature of cold or pink spots after cooking information on thermal inactivation of Salmonella in pâté
chicken liver for 3, 4, 5, 6, 7, and 8 min was 37.9 6 5.3, made from chicken liver. However, there are a few
42.1 6 6.3, 50.1 6 6.2, 52.3 6 12.1, 55.5 6 5.0, and 63.5 publications on viability of Campylobacter in chicken liver
6 8.28C, respectively. Similarly, Jones et al. (20) reported and related products in response to thermal treatments (19,
that the average internal temperature of pan-fried chicken 20, 45). The results obtained herein for Salmonella are in
liver cooked via pan frying at moderate to high heat on an general agreement with results reported in related studies
electric stove top for 3, 4, 5, 6, and 7 min was 54.8, 64.0, published on thermal inactivation of Campylobacter in
65.5, 72.0, and 718C, respectively. Regardless of the chicken liver or in chicken liver pâté. For example,
cooking time in the present study, the temperature of the Hutchison et al. (19) demonstrated that cooking chicken
pâté after blending ranged from 53.5 to 61.68C, whereas liver pâté via a water bath set at 1008C to a target internal
after 2 h of storage at 4 or 218C, the temperature of pâté temperature of 63 or 688C was sufficient to eliminate
decreased to 4.6 to 7.88C and 22 to 23.98C, respectively. Campylobacter naturally present (ca. 3.0 to 3.5 log CFU/g)
These results established that after preparing pâté, the in the liver, whereas when pâté was cooked to a target
product should be chilled promptly to 48C to prevent internal temperature of 608C, pathogen numbers decreased
multiplication of any cells of Salmonella that may survive by ca. 2.3 log CFU/g. In another study, Whyte and
heat treatment. These results are in general agreement with colleagues (45) reported that chicken liver should be pan
other studies reporting that cooking chicken liver in a frying fried with butter to an internal temperature
708C and then
pan for pâté preparation resulted in products that were not held for
2 min at
708C to eliminate naturally occurring
uniformly cooked and that displayed an internal temperature Campylobacter present in chicken liver (ca. 0.6 to 4.9 log
that was lower than the surface temperature of the organ CFU per liver reduction). The authors reported that an
meat. Although cooking pâté made from chicken liver in a internal temperature of
708C was achieved after cooking
water bath required longer times than when cooking in a liver for
2.5 min with multiple flipping, and that the core
frying pan to achieve the target internal temperature, of the liver remained pink after cooking for 5 min. Lastly,
previous studies demonstrated that when pâté was cooked Jones et al. (20) demonstrated that when inoculated chicken
in a water bath, the distribution of heat throughout the pâté livers were cooked in a frying pan with 10 g of butter for 1
 986 PORTO-FETT ET AL. J. Food Prot., Vol. 82, No. 6

to 7 min at moderate heat, the internal temperature ranged 2. Anonymous. 2013. Multistate outbreak of Campylobacter jejuni
from 36 to 728C and the survival rate of Campylobacter infections associated with undercooked chicken livers – northeastern
United States, 2012. Morb. Mortal. Wkly. Rep. 62:874–876.
ranged from 100 to 8.3%. According to the authors, a
3. AOAC International. 2000. Official methods of analysis of AOAC
logistic model predicted a 98% survival rate of Campylo- International, 17th ed. AOAC, Gaithersburg, MD.
bacter in chicken liver cooked in a frying pan to an internal 4. Arroyo, G., and J. A. Arroyo. 1995. Detection of Salmonella
temperature of 528C, whereas if cooked to an internal serotypes in edible organ meats from markets in Madrid, Spain. Food
temperature of either 708C without holding, or 768C, or Microbiol. 12:13–20.

888C, the predicted survival rate decreased to 22, 5, and 5. Batz, M. B., S. Hoffmann, and J. G. Morris. 2012. Ranking the
disease burden of 14 pathogens in food sources in the United States
,0.001%, respectively. Variations in thermal inactivation of
using attribution data from outbreak investigations and expert
foodborne pathogens in chicken liver pâté among studies elicitation. J. Food Prot. 75:1278–1291.
may be attributed to factors such as the number of times 6. Carramiñana, J. J., J. Yangüela, D. B. Parmo, C. Rota, A. I. Agustin,
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

livers were flipped during cooking, variability in tempera- A. Ariño, and A. Herrera. 1997. Salmonella incidence and
ture of cooking surfaces, types of appliances used for distribution of serotypes throughout processing in a Spanish poultry
cooking, cooking methods (e.g., water bath versus frying slaughterhouse. J. Food Prot. 60:1312–1317.
pan), holding times after cooking, and differences in 7. Centers for Disease Control and Prevention. 2012. Multistate
outbreak of human Salmonella Heidelberg infections linked to
formulation (e.g., moisture, salt, spices, and fat content), ‘‘Kosher broiled chicken livers’’ from Schreiber Processing Corpo-
as well as strain-to-strain variation and microbiological ration (final update). Available at: https://www.cdc.gov/salmonella/
methods used to recover the pathogen. 2011/chicken-liver-1-11-2012.html. Accessed 27 March 2018.
Collectively, our findings may be useful for establish- 8. Cox, N. A., L. J. Richardson, R. J. Buhr, J. K. Northcutt, J. S. Bailey,
ing cooking guidelines for chicken liver pâté and, thus, for P. F. Cray, and K. L. Hiett. 2007. Recovery of Campylobacter and
lowering the risk of foodborne illness if naturally contam- Salmonella serovars from spleen, liver and gallbladder, and ceca of
six- and eight-week-old commercial broilers. J. Appl. Poult. Res.
inated chicken liver is not cooked properly. Despite
Journal of Food Protection 2019.82:980-987.

16:477–480.
consumer preferences for partially cooked liver for use in 9. D’Aoust, J.-Y., U. T. Purvis, R. M. Coulter, and K. Weiss. 1985.
pâté, our results demonstrated that ca. a 5-log reduction was Salmonella in fresh pork liver and chicken liver in Canada: a 1979–
only achieved when pâté was cooked to a target internal 80 survey. Can. Inst. Food Sci. Technol. J. 18:323–325.
temperature of
73.98C in a water bath or was made with 10. D’Aoust, J. Y., D. W. Warburton, and A. M. Sewell. 1985.
chicken liver cooked for .8 min (internal temperature ca. Salmonella typhimurium phage-type 10 from cheddar cheese
implicated in a major Canadian foodborne outbreak. J. Food Prot.
828C) in a frying pan maintained at ca. 1408C. Regardless of
48:1062–1066.
the cooking method used to prepare pâté, the endpoint 11. Doyle, E., and A. S. Mazzota. 2000. Review of studies on the thermal
temperature of the product must be measured with a food resistance of salmonellae. J. Food Prot. 63:779–795.
thermometer to ensure that the required minimum target 12. Forbes, K. J., F. J. Gormley, J. F. Dallas, O. Labovitiadi, M. MacRae,
endpoint temperature (i.e., 73.98C) is achieved (28, 41, 43). R. J. Owen, J. Richardson, N. J. C. Strachan, J. M. Cowden, I. D.
These findings demonstrated that to lower any public health Ogden, and C. C. McGuigan. 2009. Campylobacter immunity and
coinfection following a large outbreak in a farming community. J.
risk potentially associated with consumption of pâté made
Clin. Microbiol. 47:111–116.
from chicken liver that may be naturally contaminated with 13. Gast, R. K., R. Guraya, D. R. Jones, and K. E. Anderson. 2013.
Salmonella, food handlers and consumers must cook, Colonization of internal organs by Salmonella Enteritidis in
handle, and store product at appropriate temperatures and experimentally infected laying hens housed in conventional or
avoid cross-contamination with raw food before consump- enriched cages. Poult. Sci. 92:468–473.
tion. Studies are ongoing to educate consumers and chefs 14. Giaccone, V., P. Catellani, and L. Alberghini. 2012. Food as cause of
and to evaluate the effectiveness of high pressure processing human salmonellosis. Available at: https://www.researchgate.net/
publication/221922701_Food_as_Cause_of_Human_Salmonellosis.
and the use of food-grade antimicrobial agents to lessen the Accessed 27 July 2018.
levels of naturally occurring Salmonella that may be present 15. Glashower, D., J. Snyder, D. Welch, and S. McCarthy. 2017.
in or on raw chicken liver. Outbreak of Campylobacter jejuni associated with consuming
undercooked chicken liver mousse – Clark County, Washington,
ACKNOWLEDGMENTS 2016. Morb. Mortal. Wkly. Rep. 66:1027.
16. Godwin, S., C. Maughan, and E. Chambers IV. 2016. Food safety:
We are extremely grateful to the following colleagues who in large
recommendations for determining doneness in consumer egg dish
measure contributed to the successful completion of this study by sharing
recipes and measurement of endpoint temperature when recipes are
their time, talents, resources, and opinions: John J. Johnston, William
followed. Foods 5:1–10.
Lanier, Karen Becker, Kis R. Hale, Willian Shaw, Udit Minocha, Christine
17. He, G. Z., W. Y. Tian, N. Qian, C. Cheng, and S. X. Deng. 2010.
Alvarado, Meryl Silverman, Mark Carter, Cesar Morales, and Glenn
Quantitative studies of the distribution pattern for Salmonella
Tillman (USDA-FSIS), and James A. Lindsay, Mark E. Berrang, and Mary
Enteritidis in the internal organs of chicken after oral challenge by
Camp (USDA, Agricultural Research Service). Mention of trade names or
a real-time PCR. Vet. Res. Commun. 34:669–676.
commercial products in this publication is solely for the purpose of
18. Hoffman, S., M. B. Batz, and J. G. Morris. 2012. Annual cost of
providing specific information and does not imply recommendation or
illness and quality-adjusted life year losses in the United States due
endorsement by the U.S. Department of Agriculture (USDA).
to 14 foodborne pathogens. J. Food Prot. 75:1292–1302.
19. Hutchison, M., D. Harrison, I. Richardson, and M. Tchórzewska.
REFERENCES 2015. A method for the preparation of chicken liver pâté that reliably
1. Abdellah, C., R. F. Fouzia, C. Abdelkader, S. B. Rachida, and Z. destroys campylobacters. Int. J. Environ. Public Health 12:4652–
Mouloud. 2009. Prevalence and anti-microbial susceptibility of 4669.
Salmonella isolates from chicken carcasses and giblets in Menknès, 20. Jones, A. K., D. Rigby, M. Burton, C. Millman, N. J. Williams, T. R
Morocco. Afr. J. Microbiol. Res. 3:215–219. Jones, P. Wigley, S. J. O’Brien, and P. Cross. 2016. Restaurant
 J. Food Prot., Vol. 82, No. 6 FATE OF SALMONELLA IN CHICKEN LIVER PÂTÉ 987

cooking trends and increased risk for Campylobacter infection. 34. Ruhlman, M., and B. Polcyn. 2013. Pâté and terrines: the Cinderella
Emerg. Infect. Dis. 22:1208–1215. meat loaf. In Charcuterie: the craft of salting, smoking and curing, p.
21. Juneja, V. K., B. S. Eblen, and G. M Ransom. 2001. Thermal 201–251. W. W. Norton and Company Inc., New York.
inactivation of Salmonella spp. in chicken broth, beef, pork, turkey, 35. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M.-A.
and chicken: determination of D- and z-values. J. Food Sci. 66:146– Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011.
152. Foodborne illness acquired in the United States – major pathogens.
22. Jung, Y., A. C. S. Porto-Fett, B. A. Shoyer, E. Henry, L. E. Shane, M. Emerg. Infect. Dis. 17:7–15.
Osoria, and J. B. Luchansky. 2019. Prevalence, levels, and viability 36. Scharff, R. L. 2012. Economic burden from health losses due to
of Salmonella in and on raw chicken livers. J. Food Prot. 82:834– foodborne illness in the United States. J. Food Prot. 75:123–131.
843. 37. Senhaji, A. F., and M. J. Loncin. 1977. The protective effect of fat on
23. Kowalski, J., and The Culinary Institute of America. 2011. the heat resistance of bacteria (I). J. Food Technol. 12:203–216.
Forcemeats, p. 148–201. In The art of charcuterie. John Wiley and 38. Sodagari, H. R., Z. Mashak, and A. Ghadimianazar. 2015. Prevalence
Sons, Hoboken, NJ. and antimicrobial resistance of Salmonella serotypes isolated from
Downloaded from jfoodprotection.org by University of Otago - Library on 09/03/19. For personal use only.

24. Lanier, W. A., K. R. Hale, A. L. Geissler, and D. Dewey-Mattia. retail chicken meat and giblets in Iran. J. Infect. Dev. Ctries. 9:463–
2018. Chicken liver–associated outbreaks of campylobacteriosis and 469.
salmonellosis, United States, 2000–2016: identifying opportunities 39. Stapp-Kamotani, E., and K. Barlow. 2018. Amend Food Code FSIS
for prevention. Foodborne Pathog. Dis. 15:726–733. doi:10.1089/ chicken liver compliance guide in Annex. Available at: http://www.
fpd.2018.2489. foodprotect.org/issues/packets/2018ScribePacket/issues/I_029.html.
25. Little, C. L., F. J. Gormley, N. Rawal, and J. F. Richardson. 2010. A Accessed 4 May 2018.
recipe for disaster: outbreaks of campylobacteriosis associated with 40. U.S. Department of Agriculture, Food Safety and Inspection Service.
poultry liver in England and Wales. Epidemiol. Infect. 138:1691– 2012. The nationwide microbiological baseline data collection
1694. program: Raw chicken parts survey. Available at: http://www.fsis.
26. Matthews, K. R., K. E. Kniel, and T. J. Montville. 2017. Food usda.gov/shared/PDF/Baseline_Data_Raw_Chicken_Parts.pdf. Ac-
microbiology: an introduction, 4th ed. ASM Press, Washington, DC. cessed 24 May 2018.
27. Murphy, R. Y., T. Osaili, L. K. Duncan, and J. A. Marcy. 2004. Effect 41. U.S. Department of Agriculture, Food Safety and Inspection Service.
of sodium lactate on thermal inactivation of Listeria monocytogenes 2013. Giblets and food safety. Available at: https://www.fsis.usda.
Journal of Food Protection 2019.82:980-987.

and Salmonella in ground chicken thigh and leg meat. J. Food Prot. gov/wps/portal/fsis/topics/food-safety-education/get-answers/food-
67:1403–1407. safety-fact-sheets/poultry-preparation/giblets-and-food-safety/ct_
28. National Advisory Committee for Microbiological Criteria for index. Accessed 24 August 2018.
Foods. 2007. Response to the questions posed by the Food Safety 42. U.S. Department of Agriculture, Food Safety and Inspection Service.
and Inspection Service regarding consumer guidelines for the safe 2016. Other raw chicken parts sampling project. Available at: https://
cooking of poultry products. J. Food Prot. 70:251–260. www.fsis.usda.gov/wps/wcm/connect/5ef0f371-0ce7-4eaf-9cdc-
29. Norton, S., E. Huhtinen, S. Conaty, K. Hope, B. Campbell, M. Tegel, f e 6 1 8 3 5 6 d 5 6 / 7 2 - 6 . p d f ? M O D¼A J P E R E S & C O N V E RT _
R. Boyd, and B. Cullen. 2012. A large point-source outbreak of TO¼url&CACHEID¼5ef0f371-0ce7-4eaf-9cdc-3fe618356d56. Ac-
Salmonella Typhimurium linked to chicken, pork, and salad rolls cessed 24 May 2018.
from a Vietnamese bakery in Sydney. Western Pac. Surveill. 43. U.S. Department of Agriculture, Food Safety and Inspection Service.
Response J. 3:1–8. 2018. Minimizing the risk of Campylobacter and Salmonella
30. Osaili, T., C. L. Griffis, E. M. Martin, B. L. Beard, A. Keener, and J. illnesses associated with chicken liver. Available at: https://www.
A. Marcy. 2006. Thermal inactivation studies of Escherichia coli fsis.usda.gov/wps/wcm/connect/b3f4efe7-27d4-4c39-bce7-
O157:H7, Salmonella, and Listeria monocytogenes in ready-to-eat 0 11 b 7 b b d 1 e 7 d / C h i c k e n - L i v e r- G u i d a n c e - J u l y - 2 0 1 8 . p d f ?
chicken-fried beef patties. J. Food Prot. 69:1080–1086. MOD¼AJPERES. Accessed 6 August 2018.
31. Parry, A., E. Fearnley, and E. Denehy. 2012. ‘‘Surprise’’: outbreak of 44. U.S. Department of Agriculture, Food Safety and Inspection Service.
Campylobacter infections associated with chicken liver pâté at a 2018. Sampling results for FSIS regulated products: calculation (Jul
surprise birthday party Adelaide, Australia, 2012. Western Pac. 1, 2017–Jun 30, 2018). Available at: https://www.fsis.usda.gov/wps/
Surveill. Response J. 3:16–19. wcm/connect/68f5f6f2-9863-41a5-a5c4-25cc6470c09f/Sampling-
32. Porto-Fett, A. C. S., J. E. Call, and J. B. Luchansky. 2008. Validation Project-Results-Data.pdf?MOD¼AJPERES. Accessed 16 August
of a commercial process for inactivation of Escherichia coli 2018.
O157:H7, Salmonella Typhimurium, and Listeria monocytogenes 45. Whyte, R., J. A. Hudson, and C. Graham. 2006. Campylobacter in
on the surface of whole muscle beef jerky. J. Food Prot. 71:918–926. chicken livers and their destruction by pan frying. Lett. Appl.
33. Procura, F., D. J. Bueno, S. B. Bruno, and A. D. Rogé. 2017. Microbiol. 43:591–595.
Prevalence, antimicrobial resistance profile and comparison of 46. Zdragas, A., K. Mazaraki, G. Vafeas, V. Giantzi, T. Papadopoulos,
methods for the isolation of Salmonella in chicken liver from and L. Ekateriniadou. 2012. Prevalence, seasonal occurrence and
Argentina. Food Res. Int. Available at: https://doi.org/10.1016/j. antimicrobial resistance of Salmonella in poultry retail products in
foodres.2017.08.008. Accessed 13 August 2018. Greece. Lett. Appl. Microbiol. 55:308–313.