M:Fo

odMi

crobio

logy

&Sa

fety

Role of Peroxyacetic Acid, Octanoic Acid,Malic Acid, and Potassium Lactate on theMicrobiological and Instrumental ColorCharacteristics of Ground BeefAnand Mohan, F. W. Pohlman, J. A. McDaniel, and M. C. Hunt

Abstract: This study evaluated the effects of peroxyacetic acid (PAA), malic acid (MA), octanoic acid (OA), and potassiumlactate (KL) followed by mixing with trisodium phosphate (TSP) and an ultra-chilled CO2 snow shower on microbialcounts of Escherichia coli (EC), coliform (CF), aerobic plate count (APC), and Salmonella Typhimurium (ST) on inoculatedbeef trimmings and the instrumental color attributes of the resultant ground beef. Beef trimmings inoculated with ECand ST were treated with either 0.02% PAA; 2% MA; 0.04% OA; or 2% KL, followed by mixing with 10% TSP andrapid chilling with CO2 snow shower. Treated trimmings were then ground, packaged, displayed under simulated retailconditions, and sampled on days 1, 2, 3, 5, and 7 for microbial counts and instrumental color characteristics. PAA, MA,OA, and KL reduced (P < 0.05) the microbial counts of EC, CF, APC, and ST during display. Among treatments, OA wasmost effective on EC, CF, ST, and APC during retail display. Chilling beef trimmings with CO2 improved instrumentalcolor characteristics of the produced ground beef but made little difference in reducing microbial counts during display.During retail display, ground beef produced from beef trimmings treated with antimicrobials tended to maintain redness,myoglobin redox form stability (630 nm/580 nm), and overall instrumental color characteristics.

Keywords: antimicrobial agents, beef trim, Escherichia coli, ground beef, malic acid, octanoic acid, potassium lactate,peroxyacetic acid, Salmonella Typhimurium

Practical Application: This research provides a practical and cost-effective decontamination technology for beef processorsthat can be immediately implemented in the ground beef production chain. Using antimicrobial intervention coupledwith rapid chilling could benefit the meat industry by preserving the quality attributes of ground beef during retail displayunder aerobic packaging environment.

IntroductionGround beef has been a concern to producers and consumers

because of its high potential risk of food-borne illness. Groundbeef is produced using trimmings from different cattle and carcasslocations, so grinding and mixing during production create a highpotential for cross-contamination. In fact, ground beef is the lead-ing source of Escherichia coli (EC) and Salmonella infections in theUnited States (Center for Disease Control and Prevention [CDC]2009). In the United States every year, an estimated 76 millioncases of food-borne illnesses occur, resulting in $5 to $17 billionin economic and productivity losses annually (Edwards and Fung2006). In 2002, 19 million pounds of ground beef were recalled,

MS 20110899 Submitted 7/26/2011, Accepted 12/8/2011. Author Mohan is withDept. of Food Science & Technology, Food Science Bldg., Univ. of Georgia, Athens,G.A. 30602, U.S.A. Authors Pohlman and McDaniel are with Dept. of AnimalScience, Univ. of Arkansas, Fayetteville, Ark. 72701, U.S.A. Author Hunt is withDept. of Animal Sciences & Industry, Weber Hall, Kansas State Univ., Manhattan,Kans. 60616, U.S.A. Direct inquiries to author Mohan, Dept. of Food Science &Technology, Univ. of Georgia, 100 Cedar St., 240 Food Science Bldg., Athens, G.A.30602-2610, U.S.A. (E-mail: anandm@uga.edu).

and in 2007, 2 recalls removed 27.4 million pounds of groundbeef from food supplies because of suspected EC contamination(USDA-FSIS recall case archives).

In spite of decades of research, ground beef remains a majorsource of food-borne pathogens and safety recalls. Decontami-nating beef trimming destined for ground beef production maybe the most important step in controlling food-borne pathogenicbacteria during production. Recent studies have suggested that de-contaminating beef trimmings before grinding can effectively re-duce bacterial counts in ground beef (Pohlman and others 2002a,2002b; Stivarius and others 2002).

The meat industry continues to see new pathogens that presentobstacles in meat safety for producers and consumers. Dorsa andothers (1997) found that using hot water or organic acid in process-ing beef carcasses effectively reduced bacteria but had detrimentaleffects on quality characteristics. Pohlman and others (2009) foundthat KL and sodium metasilicate treatments maintained oxymyo-globin content and redness of color through simulated retail display.TSP has been approved for use on beef carcasses and reduces bacte-rial attachment to carcass surfaces (Dickson and others 1994; Kimand Slavik 1994; Jimenez-Villarreal and others 2003a). In addi-tion, ground beef decontamination using TSP has shown promise

C© 2012 Institute of Food Technologists R©doi: 10.1111/j.1750-3841.2011.02600.x Vol. 00, Nr. 0, 2012 � Journal of Food Science M1Further reproduction without permission is prohibited

M:FoodMicrobiology&

Safety

Microbial safety of ground beef . . .

in stabilizing ground beef color as part of a system for increasingthe safety and color shelf-life of beef products (Pohlman and others2002b; Jimenez-Villarreal and others 2003b).

Using antimicrobial interventions to decontaminate meat con-tinues to advance, but most decontamination treatments are lesseffective in the presence of organic matter (Siragusa 1995). More-over, it is difficult to design a chemical decontamination treatmentthat can be applied directly to raw beef products without adverselyaffecting the color, taste, or smell of the product. More effectivestrategies of antimicrobial treatments and application methods thatare safe, easy to use, cost effective, and keep organoleptic prop-erties acceptable are needed (Mohan and others 2009; Quilo andothers 2009). Currently available decontamination treatments andmethods are somewhat effective in reducing microbial populations;however, many of these treatments damage quality attributes.

The beef processing industry has spent more than $750 mil-lion over the last decade to improve beef product safety (Huffman2002; Koohmaraie and others 2005; Sofos 2008), but they con-tinue to seek new and better decontamination technologies onbeef trimmings intended for production of ground beef to elimi-nate incidents of food-borne pathogens and subsequent safety re-calls (USDA-FSIS 2007). Decontamination of meat products withantimicrobials near the end of production would be an effectivestrategy in reducing pathogenic bacteria populations in the finalproduct. However, the challenge of direct antimicrobial applica-tion is that the active substances can be neutralized rapidly uponcontact as they diffuse from the surface into the food mass. Fur-thermore, many antimicrobial agents affect meat color adversely(Jimenez-Villarreal and others 2003b). Consumers demand notonly safe meat products but also high quality products with goodcolor, taste, and appearance (Baublits and others 2006a, 2006b).Maintaining a bright red color of meat involves a delicate interac-tion of applied antimicrobial and myoglobin chemistry. Sequentialapplication of TSP with rapid chilling by CO2 shower after antimi-crobial treatment may have potential for preserving the efficacy ofapplied antimicrobial as well as the quality attributes of groundbeef. Therefore, the objectives of this study were: (1) to evaluatethe effectiveness of antimicrobial interventions on reducing E. coli,Salmonella Typhimurium, coliform, and aerobic bacteria of pre-inoculated beef trimmings, and (2) to assess sequential applicationof 10% TSP with rapid chilling by CO2 shower on preserving theadverse impact of antimicrobial treatment on ground beef instru-mental color characteristics.

Materials and Methods

Bacterial preparation and inoculationInoculums of EC (ATCC #25922) and a nalidixic acid resis-

tant strain of Salmonella Typhimurium (ST, ATCC #333310NR)were prepared from frozen (−80 ◦C) stock cultures. Frozen cul-tures of EC and ST were thawed, and 0.1 mL of EC suspensionwas inoculated into 60 separate 40 mL aliquots of brain heart in-fusion (BHI) (BD, BBLTM, Becton Dickinson and Co., Sparks,Md., U.S.A.), and 0.1 mL of ST suspension was inoculated into60 separate 40 mL aliquots of BHI with nalidixic acid (Fisher Sci-entific, Fair Lawn, N.J., U.S.A.). Following 18 h of incubation at37 ◦C, bacteria were then harvested by centrifugation (3500 ×g for 20 min at 25 ◦C; Beckman GS-6 series, Fullerton, Calif.,U.S.A.) and resuspended with 40 mL of 0.1% buffered peptonewater (BPW) (Difco Laboratories, Becton Dickinson and Co.,Sparks, Md., U.S.A.) and pooled to make a bacterial cocktail. Thebacteria (log 108 CFU/mL EC and log 108 CFU/mL ST) were

cooled to 4 ◦C and then mixed. Beef trimmings were inoculatedby pouring the bacterial cocktail over the trimmings and mixingthoroughly in a tumbler to ensure all trimmings were equally ex-posed to inoculum. The inoculated beef trimmings were placedin a sterile bag and stored at 4 ◦C cooler for 12 to 14 h to allowfurther microbial attachment.

Antimicrobial treatment application and sampleprocessing

The antimicrobial treatments, 0.02% peroxyacetic acid (PAA),0.04% octanoic acid (OA), 2% malic acid (MA), 2% potassiumlactate (KL; UltraLac KL-78, Hawkins, Inc., Minneapolis, Minn.,U.S.A.), and 10% (w/v) trisodium phosphate (TSP; ICL Perfor-mance Products, St. Louis, Mo., U.S.A.), were prepared by mixingantimicrobials with an appropriate amount of water to achieve tar-geted concentration. Control treatments were: inoculated sampleswith (IN + CO2) and without CO2 (IN); uninoculated sampleswith (UN + CO2) and without CO2 (UN) treatment. UN andUN + CO2 were used only for instrumental color measurementsand not for microbial analysis. Inoculated beef trimmings weretreated with 1 of the following 4 antimicrobials: (1) 0.02% PAA;(2) 0.04% OA; (3) 2% MA; and (4) 2% KL. These samples wereeach subsequently mixed with 10% (w/v) TSP and chilled withCO2 snow in a CO2 meat mixer (Model 814, Food ProcessingEquipment Co., Springdale, Ark., U.S.A.). For treatments, 5.4 kgof beef trimmings (80% lean and 20% fat) were placed in a CO2

mixer with 1 L of one of the antimicrobial solutions and mixed for3 min; samples were then chilled with CO2 for 30 s. Afterward, thebeef trimmings were ground twice using a Hobart grinder (Model310, Hobart Inc., Troy, Ohio, U.S.A.) with a 3.2-mm plate. Aftergrinding, 200 g of the ground beef was placed on absorbent pads infoam trays and overwrapped with polyvinyl chloride film (oxygentransmission rate of 14000 cc/mm2/24 h/L atm; Koch Supplies,Inc., Kansas City, Mo., U.S.A.). The ground beef was stored anddisplayed under retail display conditions (4 ◦C; continuous deluxewarm white fluorescent lighting; 1600 lx; Phillips, Inc., Somerset,N.J., U.S.A.) for 7 d in a retail display case.

Microbial samplingOn days 1, 2, 3, 5, and 7 of retail display, microbial enu-

meration was carried out by aseptically removing 25 g samplesfrom the ground beef surface using a sterile scalpel; the sam-ples were placed in sterile whirl pack bags (Nasco, Ft. Atkin-son, Wis., U.S.A.) with 225 mL of 0.1% buffered peptone water.Samples were homogenized for 2 min in a stomacher (Model400 Lab Stomacher; Seward, London, UK), and serial 10-folddilutions were made. Spread plating was done in duplicate foraerobic plate count (APC) and EC/coliform (CF) counts usingPetrifilm R© plates (3M Corp., St. Paul, Minn., U.S.A.). ST countswere performed on Salmonella Shigella agar (Difco Laboratories,Becton Dickinson and Co., Sparks, Md. 21152, U.S.A.) contain-ing nalidixic acid. Plates were then incubated at 37 ◦C in an aerobicincubation chamber (VWR Model 5015 and Model 3015 incuba-tors, VWR Scientific, Cornelius, Oreg., U.S.A.). All counts wererecorded as colony forming units per gram (CFU/g).

Instrumental colorOn days 1, 2, 3, 5, and 7 of simulated retail display, instrumental

color of the ground beef samples was measured using a Hunter-LabMiniscan XE Spectrophotometer (Model 45/0-L, Hunter Asso-ciates Laboratory, Inc., Reston, W.Va., U.S.A.). The samples wereevaluated using illuminant A, 10◦ observer, 2.54 cm dia aperture

M2 Journal of Food Science � Vol. 00, Nr. 0, 2012

M:Fo

odMi

crobio

logy

&Sa

fety

Microbial safety of ground beef . . .

for the Commission Internationale de l’Eclairage (CIE; L∗, a∗, and b∗)color values. In the visible spectrum 630 nm/580 nm reflectanceratio were used to estimate the proportion of oxymyoglobin con-tent (Hunt and others 1991). In addition, hue angle [tan−1(b∗/a∗)]and the saturation index [(a2 + b2)0.5] were also calculated accord-ing American Meat Science Association Meat Color Guidelines.Three measurements were taken for each sample and averaged forstatistical analysis.

Statistical analysisThe experimental design was a randomized complete block with

3 replications (representing different batches of beef trimmings)with a 6 × 5 factorial design consisting of 6 different treatmentapplications and 5 display days (1, 2, 3, 5, and 7). The modelincluded the main effects of antimicrobial treatment, day of display,and treatment × day interactions. The Proc Mixed procedure ofSAS (2003) was used to perform type-3 tests of fixed effects. Leastsquares means for protected F-tests (P < 0.05) were separated byusing least significant differences (LSD; P < 0.05).

Results

Effects of antimicrobial treatment on EC populations inground beef

Treating beef trimmings with PAA, MA, OA, and KL beforegrinding reduced EC counts by approximately 1-log (P < 0.05)over IN and IN + CO2 controls on display day 1 (Table 1). Onday 2 of display, PAA and OA treatments showed reductions of ECcounts by 2.0-log (P < 0.05) over IN and 1.1-log (P < 0.05) overIN + CO2 control. Similarly, MA and KL treatments reducedEC counts by 1.1 to 1.4 logs over IN and by 0.2 to 0.5 logs(P < 0.05) over IN + CO2 on day 2 of the retail display. Treatingbeef trimmings with OA, PAA, and MA reduced EC counts by2.1, 1.1, and 1.0 logs (P < 0.05) during display day 3 over IN.When compared to IN + CO2, treating beef trimmings withOA, PAA, MA, and KL reduced EC counts by 2.9, 1.9, 1.8, and0.8 logs (P < 0.05) during the display day 3. On retail displayday 5, beef trimmings treated with OA and PAA reduced ECcount by approximately 2.0 and 1.0 logs (P < 0.05) over IN andIN + CO2. Beef trimmings treated with OA reduced EC countsby approximately 3.0-log (P < 0.05) while MA reduced EC countsby 1.0-log (P < 0.05) over IN and IN + CO2.

Effects of antimicrobial treatment on CF populations ofground beef

Treating beef trimmings with PAA, MA, OA, and KL beforegrinding reduced CF counts by 0.4 to 1.2 logs (P < 0.05) over

Table 1–Least squares means† for log CFU£/g counts of Escherichiacoli (EC) evaluated for antimicrobial treatment∗ on bulk groundbeef during 7 d of display time.

Log (CFU/g)/display time (d)

Treatment∗ 1 2 3 5 7

IN 6.4b 7.4d 6.5c 6.7c 7.3d

IN + CO2 6.4b 6.5c 7.3d 6.4c 7.2d

PAA 5.4a 5.4a 5.4b 5.5b 7.4d

MA 5.6a 6.0b 5.5b 6.2c 6.1b

OA 5.3a 5.4a 4.4a 4.4a 4.4a

KL 5.4a 6.3bc 6.5c 6.7c 6.9c

†abcd = within a column, numbers with different superscripts are statistically different(P < 0.05).£CFU = colony forming units. Standard error for treatment × display time (d) = 0.14to 0.19.∗IN = untreated inoculated control; IN + CO2 = untreated inoculated control treatedwith chilled with CO2; PAA = 0.02% peroxyacetic acid; MA = 2% malic acid;OA = 0.04% octanoic acid; KL = 2% potassium lactate.

IN + CO2 on display day 1 (Table 2). Beef trimmings treated withPAA, MA, OA, and KL reduced CF counts by 1.0, 0.9, 1.1, and0.9 logs (P < 0.05) over IN on day 2 of the display. Also, ultra-chilling beef trimmings with CO2 resulted in lower counts of CFafter treatment with PAA (1.1-log), MA (1.0-log), OA (1.1-log),and KL (0.9-log) (P < 0.05) over IN + CO2 on day 2 of thedisplay.

On day 3 of the retail display, reduced CF counts were found intreatments with PAA (1.1-log) and OA (1.2-log) over IN. Usingthe ultra-chilled CO2 shower was more effective in lowering theCF counts, reducing them by 1.4 and 1.5 logs (P < 0.05) overIN + CO2. On day 5 of the retail display, beef trimmings treatedwith PAA and OA reduced CF counts by 1.4 and 2.5 logs of (P <

0.05) over IN control; CF counts were reduced in treatments withPAA (2.2-log), MA (1.4-log), OA (3.3-log), and KL (1.1-log)(P < 0.05) over IN + CO2 control. Treating beef trimmings withMA and OA reduced CF counts by 0.9 and 2.8 logs (P < 0.05)over IN and 1.0 and 3.0 logs over IN + CO2 during the displayday 7.

Effects of antimicrobial treatment on APC populations ofthe ground beef

Antimicrobial treatment of beef trimmings with PAA, MA, OA,and KL reduced APC counts by approximately 1.7, 0.8, 1.5, and1.5 logs (P < 0.05) over IN during day 1 of the retail display(Table 3). Rapid chilling of beef trimmings with a CO2 snowshower after antimicrobial treatments with PAA, MA, OA, and KLlowered APC counts by 1.5, 0.6, 1.3, and 1.2 logs (P < 0.05) on

Table 2–Least squares means† for log CFU£/g counts of coliform(CF) evaluated for antimicrobial treatment∗ on bulk ground beefduring 7 d of display time.

Log (CFU/g)/display time (d)

Treatment∗ 1 2 3 5 7

IN 6.6c 6.8b 6.7cd 6.9d 7.4c

IN + CO2 6.8c 6.8b 6.9d 7.8e 7.6c

PAA 5.8a 5.7a 5.6a 5.5b 7.6c

MA 6.2b 5.8a 6.4bc 6.4c 6.6b

OA 5.6a 5.7a 5.4a 4.5a 4.7a

KL 5.6a 5.9a 6.3bc 6.7cd 7.4c

†abcde = within a column, numbers with different superscripts are statistically different(P < 0.05).£CFU = colony forming units. Standard error for treatment × display time (d) = 0.17to 0.21.∗IN = untreated inoculated control: IN + CO2 = untreated inoculated control treatedwith chilled with CO2; PAA = 0.02% peroxyacetic acid; MA = 2% malic acid; OA =0.04% octanoic acid; KL = 2% potassium lactate.

Table 3– Least squares means† for log CFU£/g counts of aerobicplate counts (APC) evaluated for antimicrobial treatment∗ on bulkground beef during 7 d of display time.

Log (CFU/g)/display time (d)

Treatment∗ 1 2 3 5 7

IN 7.2c 7.7b 7.6bc 7.0b 7.8b

IN + CO2 7.0c 7.8b 7.7c 8.1c 8.2b

PAA 5.5a 7.9b 7.2b 7.9c 7.9b

MA 6.4b 7.6b 7.8c 7.9c 7.9b

OA 5.7a 5.6a 4.9a 4.8a 7.3a

KL 5.8a 7.9b 7.5b 8.1c 8.1b

†abc = within a column, numbers with different superscripts are statistically different(P < 0.05).£CFU = colony forming units. Standard error for treatment × display time (d) = 0.17to 0.20.∗IN = untreated inoculated control; IN + CO2 = untreated inoculated control treatedwith chilled with CO2; PAA = 0.02% peroxyacetic acid; MA = 2% malic acid;OA = 0.04% octanoic acid; KL = 2% potassium lactate.

Vol. 00, Nr. 0, 2012 � Journal of Food Science M3

M:FoodMicrobiology&

Safety

Microbial safety of ground beef . . .

display day 1 as compared to IN + CO2 (Table 3). During retaildisplay of ground beef on days 2 to 7, antimicrobial treatmentsusing PAA, MA, and KL did not significantly (P < 0.05) reduceAPC counts over either IN or IN + CO2. The OA treatmentremained effective, lowering (P < 0.05) APC counts by 2.0, 2.7,and 3.3 logs during days 2, 3, and 5 of the retail display over IN(Table 3).

Application of ultra-chilled CO2 shower enhanced the antimi-crobial efficacy of OA resulting in reduction of APC counts by2.1 and 2.8, and 3.3 logs (P < 0.05) over IN + CO2 on days 3, 5,and 7 of display. Treating beef trimmings with PAA, MA, or KLdid not effectively reduce (P < 0.05) APC counts during displaydays 2 through 5 over IN and IN + CO2. On day 7 of the display,all treatments were ineffective.

Effects of antimicrobial treatment on SM populations ofground beef

Treating beef trimmings with PAA, MA, OA, and KL be-fore grinding reduced SM count by 2.2, 1.1, 2.1, and 1.8 logs(P < 0.05) over IN and IN + CO2 on display day 1 (Table 4).The reductions continued on day 2 of the display with SM countsof 2.3, 0.9, 1.8, and 2.2 logs (P < 0.05) over IN and IN +CO2 controls. On day 3 of the retail display, PAA, MA, OA, andKL treatments reduced SM counts by 2.2, 1.1, 2.2, and 1.9 logs(P < 0.05) over IN and IN + CO2 controls.

On day 5 of the retail display, among all decontamination treat-ments, PAA and OA reduced SM counts (1.4 and 2.3 logs) morethan MA and KL (0.9 and 1.0 logs) (P < 0.05) compared with IN.On day 7, PAA, MA, OA, and KL treatments reduced SM countsby 1.4, 0.9, 2.3, and 1.0 logs (P < 0.05) over IN.

Table 4–Least squares means† for log CFU£/g counts of Salmonella(SM) evaluated for antimicrobial treatment∗ on bulk ground beefduring 7 d of display time.

Log (CFU/g)/display time (d)

Treatment∗ 1 2 3 5 7

IN 5.5d 5.6d 5.6d 5.6d 5.8d

IN + CO2 5.5d 5.6d 5.5d 5.6d 5.8d

PAA 3.4a 3.4a 3.4a 4.1b 4.4b

MA 4.4c 4.7c 4.5c 4.6c 4.9c

OA 3.4a 3.9b 3.4a 3.4a 3.5a

KL 3.7b 3.5a 3.7b 4.4c 4.7c

†abcd = within a column, numbers with different superscripts are statistically different(P < 0.05).£CFU = colony forming units. Standard error for treatment × display time (d) = 0.17to 0.20.∗IN = untreated inoculated control; IN + CO2 = untreated inoculated control treatedwith chilled with CO2; PAA = 0.02% peroxyacetic acid; MA = 2% malic acid;OA = 0.04% octanoic acid; KL = 2% potassium lactate.

Effects of antimicrobial treatment on instrumental colorcharacteristics

The effects of treatments on instrumental color characteristicsof the ground beef samples are provided in Table 5. UN andUN + CO2 exhibited higher L∗-values than IN, IN + CO2,and inoculated samples treated with PAA, MA, OA, and KL.Among inoculated samples, beef trimmings treated with onlyCO2 and KL were relatively darker than those treated with MAand OA.

The a∗-value was highest (P < 0.05) for IN as compared withUN and UN + CO2, IN + CO2, and inoculated samples treatedwith PAA, MA, OA, and KL. Redness (a∗-value) of ground beefsamples treated with PAA and KL was higher (P < 0.05) than MAand OA. The oxymyoglobin content (630 nm/580 nm reflectanceratio) of inoculated ground beef samples treated with PAA, MA,and KL was relatively higher (P < 0.05) than samples treated withOA, UN, and UN + CO2, and IN and IN + CO2.

The saturation index value was highest (P < 0.05) for UNand UN + CO2 IN controls. Ground beef samples treated withPAA, MA, OA, and KL exhibited lower saturation index values,indicating less red color intensity in the displayed ground beefsamples. The hue angle value was higher (P < 0.05) indicatingmore discoloration for UN and UN + CO2 and samples treatedwith OA than IN and IN + CO2 and samples treated with PAA,MA, and KL.

Effect of duration of display on the main effects ofinstrumental color characteristics

The effects of duration of sample display on instrumental colorcharacteristics are summarized in Table 6. Individual day effects

Table 6–Least squares means† for the effects of duration of dis-play time on instrumental color characteristics of bulk groundbeef treated with antimicrobials during 7 d of display time.

Log (CFU/g)/display time (d)ColorParameters 1 2 3 5 7

L∗ 39.6b 38.0ab 37.3a 37.1a 37.4a

a∗ 23.4d 19.7c 18.1bc 17.1b 10.1a

b∗ 18.2c 15.7b 15.6b 15.1b 12.9a

Hue angle 37.9a 38.7ab 41.2bc 42.3c 52.0d

SI� 29.6d 25.2c 24.0c 22.9b 16.4a

630/580 ratio 2.1b 2.2b 2.0b 2.9b 1.4a

†abcd = within a row, numbers with different superscripts are statistically different(P < 0.05).∗UN = untreated uninoculated control not chilled with CO2; UN + CO2 = untreateduninoculated control chilled with CO2; IN = untreated inoculated control; IN + CO2 =untreated inoculated control treated with chilled with CO2; PAA = 0.02% peroxyaceticacid; MA = 2% malic acid; OA = 0.04% octanoic acid; KL = 2% potassium lactate.Standard error = 0.17 to 0.93.�SI = saturation index: intensity or the degree of red color saturation.630/580 ratio: 1 = brown; 5 = bright purplish red.

Table 5–Least squares means† for main effects of applying antimicrobial treatment on instrumental color characteristics of bulkground beef during 7 d of display time.

Treatment∗ColorParameters UN UN + CO2 IN IN + CO2 PA MA OA KL

L∗ 52.0c 52.5c 49.2b 46.2a 47.7b 48.7b 48.3b 46.0a

a∗ 17.5b 17.4b 19.1c 16.8b 16.6b 16.7a 14.8a 16.3b

b∗ 17.4c 16.8bc 16.1b 15.1b 14.7ab 14.4ab 14.3ab 14.0a

Hue angle 45.5b 44.5b 41.2a 42.0ab 41.6a 40.8a 44.9b 40.4a

SI� 24.9bc 24.4b 25.2c 22.8a 22.2a 22.1a 20.7a 21.6a

630/580 ratio 2.1a 2.1ab 2.3ab 2.3b 2.5c 2.5d 2.4bc 2.5c

†abcd = within a row, numbers with different superscripts are statistically different (P < 0.05).∗UN = untreated uninoculated control not chilled with CO2; UN + CO2 = untreated uninoculated control chilled with CO2; IN = untreated inoculated control; IN + CO2 =untreated inoculated control treated with chilled with CO2; PAA = 0.02% peroxyacetic acid; MA = 2% malic acid; OA = 0.04% octanoic acid; KL = 2% potassium lactate.Standarderror = 0.01 to 1.4.�SI = saturation index: intensity or the degree of red color saturation. 630/580 ratio: 1 = brown to 5 = bright purplish red.

M4 Journal of Food Science � Vol. 00, Nr. 0, 2012

M:Fo

odMi

crobio

logy

&Sa

fety

Microbial safety of ground beef . . .

(P < 0.05) were observed for L∗, a∗, b∗, and hue angle, saturationindex values, and 630 nm/580 nm reflectance ratio. Duration ofdisplay did not affect ground beef lightness (L∗) after day 1 of retaildisplay. The ground beef redness (a∗-values), and yellowness (b∗-values) decreased gradually from day 1 through day 7 of display.The hue angle increased while saturation index decreased fromday 1 to 7 of display.

DiscussionMicrobial contamination of the raw ground beef is a serious

public concern in part because it is produced from trimmingsfrom different cattle and carcass locations. Subsequent grind-ing and mixing during production creates a high potential forcross-contamination. As a preventive safety measure, a numberof decontamination techniques have been used to keep groundbeef free from food-borne pathogens. Peroxyacetic acid, malicacid, octanoic acid, and potassium lactate have been demon-strated to be effective against foodborne pathogens under a widevariety of meat processing conditions. An obvious prerequisitein the meat industry is to develop and implement an alterna-tive decontamination strategy as an effort to improve the safetyof meat products with no or minimal impact on overall qualityattributes.

The main objective of this study was to investigate the effective-ness of peroxyacetic acid, malic acid, octanoic acid, and potassiumlactate combined with sequential mixing in TSP and ultra-chilledCO2 as an effective intervention strategy to improve the safety ofground beef with minimal impact on quality characteristics.

The results were as expected for these potent antimicrobialagents: OA was the most effective on reducing EC, CO, ST,and APC during retail display. These results confirm the results inanother study, where (Pohlman and others 2002a, 2002b) foundthat 10% TSP reduced EC, CO, ST, and APC counts by 0.8, 0.7,0.7 and 0.6 logs CFU/g, respectively, compared with a control ofinoculated ground beef.

The PAA treatment reduced CO, EC, and ST counts for displaydays 1, 2, 3, and 5. Gill and Badoni (2004) reported about 1-logreduction for EC and CO using a conventional spray method with0.02% PAA and/or 0.16% acidified sodium chlorite. However,King and others (2005) demonstrated that spraying chilled carcasssurfaces with 0.02% PAA at 43 ◦C for 15 s had no effect onrifampicin-resistant E. coli O157:H7 and Salmonella Typhimuriumcounts. Farrell and others (1998) evaluated PAA as a sanitizerfor meat contact surfaces and reported that PAA was effective inreducing the bacterial load, but PAA did not completely eliminateE. coli O157:H7.

In the present study, MA reduced bacterial populations less thandid PAA and OA for EC, CO, and APC (Table 1 to 3). PAA wasslightly more effective in reducing the bacterial load (EC, CO,and ST) on the inoculated beef trimmings as shown by a >1-log CFU/g decrease in bacterial populations. Dorsa and others(1998) observed that treating beef trimmings with 2% lactic acidwas not effective in reducing mesophilic aerobic bacteria through7 d of storage. The antimicrobial effects of organic acids such asmalic acid on decontamination of beef trimmings depend on manyfactors, including processing temperature, pH, metabolic activity,and microbial interaction with the product (Mohan and others2010a, 2010b). Meat by itself has a very good buffering capacityand can tolerate a large range of changes in pH.

In this study, we focused on developing a novel strategy toinhibit the growth of EC and SM using organic acids and othercommonly used antimicrobials on beef trimmings from which

ground beef was produced. In addition, our strategy of using ultra-chilled CO2 shower for rapid chilling of beef trimmings beforegrinding takes advantage of maintaining raw ground beef qualityattributes during simulated retail display. Application of 10% TSPand ultra-chilled CO2 shower following antimicrobial treatmentpreserved the efficacy of applied antimicrobials and extended theinstrumental color attributes of ground beef during retail display.Chilling the beef trimmings with CO2 had little effect on reducingpathogens.

Results from this study suggest that using antimicrobial treat-ments on beef trimmings before grinding did not affect the overallinstrumental color characteristics of ground beef during simulatedretail display. These findings suggest that using PAA, MA, OA,and KL as antimicrobial interventions on beef trimmings withCO2 and TSP before grinding can effectively reduce bacterialloads without adversely affecting the overall quality attributes ofthe ground beef.

Quilo and others (2009) treated beef trimmings with 3% KLfollowed by 4% sodium metasilicate sampled for 7 d under sim-ulated retail display and reported only numerical differences forEC and ST counts. A similar study by Pohlman and others (2009)demonstrated that beef trimmings inoculated with EC and STand treated with 3% KL reduced bacteria only slightly less at thebeginning of display but increased in effectiveness for EC, CO,APC, and ST over 7 d of display.

Ground beef produced from beef trimmings treated with PAA,MA, OA, and KL were less red; however, antimicrobial treatmentsexhibited slightly higher numerical values for 630 nm/580 nmratio. Pohlman and others (2002a) found that ground beef treatedwith 10% TSP and 0.5% cetylpyridinium chloride treated groundbeef were significantly more (P < 0.05) red (a∗). Jimenez-Villarrealand others (2003b) found that ground beef patties treated with10% TSP and a control did not differ (P > 0.05) in L∗ value, andthe 10% TSP treatment was more (P < 0.05) red (a∗) in colorthan a control. Consumers associate redness with freshness anddiscriminate against discolored meat, making redness of meat animportant factor (Hood and Riordan 1973; Morrissey and others1994; Mohan and others 2010c).

ConclusionThis report examined the effects of PAA, MA, OA, and KL

with subsequent rapid chilling of beef trimmings before grindingas an intervention strategy to reduce bacterial populations whilemaintaining overall quality during retail display. The results of thisstudy indicate that OA, which is commonly used in microbiolog-ical intervention strategies, was the most effective antimicrobialagent currently approved for use in beef carcass tissue and beeftrimmings. Using antimicrobials such as PAA, OA, and MA inground beef production system can effectively reduce total bac-terial populations with minimal impact on color characteristics.Therefore, treating beef trimmings before grinding with PAA,OA, MA, and KL should improve ground beef safety and en-hance shelf life. These results have practical application in the meatindustry for developing intervention strategies to decontaminatebeef trimmings destined for ground beef. Applying sequential in-terventions coupled with rapid chilling will preserve the qualityattributes of the ground beef during retail display under aerobicpackaging environment.

ReferencesBaublits RT, Pohlman FW, Brown Jr AH, Johnson ZB. 2006a. Effects of enhancement with

differing phosphate types, concentrations, and pump rates, without sodium chloride, on beefbiceps femoris instrumental color characteristics. Meat Sci 72:503–12.

Vol. 00, Nr. 0, 2012 � Journal of Food Science M5

M:FoodMicrobiology&

Safety

Microbial safety of ground beef . . .

Baublits RT, Pohlman FW, Brown Jr AH, Johnson ZB. 2006b. Enhancement with varyingphosphate types, concentrations, and pump rates, without sodium chloride on beef bicepsfemoris quality and sensory characteristics. Meat Sci 72:404–14.

Center for Disease Control and Prevention (CDC). 2009. Multistate outbreak of E. coliO157:H7 infections associated with beef from JBS Swift Beef Company. Available from:http://www.cdc.gov/ecoli/2009/0701.html. Accessed Mar 16, 2011.

Dickson JS, Cutter CG, Siragusa GR. 1994. Antimicrobial effects of trisodium phosphate againstbacteria attached to beef tissue. J Food Prot 57:952–5.

Dorsa WJ, Cutter CN, Siragusa GR. 1997. Effects of acetic acid, lactic acid and trisodium phos-phate on the micro flora of refrigerated beef carcass surface tissue inoculated with Escherichiacoli 0157:H7, Listeria innocua, and Clostridium sporogenes. J Food Prot 60:619–24.

Edwards JR, Fung DYC. 2006. Prevention and decontamination of Escherichia coli O157:H7on raw beef carcasses in commercial beef abattoirs. J Rapid Methods Automat Microbiol14:1–95.

Farrell BL, Ronner AB, Wong ACL. 1998. Attachment of Escherichia coli O157:H7 in groundbeef to meat grinders and survival after sanitation with chlorine and peroxyacetic acid. J FoodProt 61:817–22.

FSIS Notice, 65-07. 2007. Washington, DC: United States Department of Agriculture, FoodSafety and Inspection Service.

Gill CO, Badoni M. 2004. Effects of peroxyacetic acid, acidified sodium chlorite or lactic acidsolutions on the microflora of chilled beef carcasses. Int J Food Microbiol 91:43–50.

Hood DE, Riodran EB. 1973. Discoloration in pre-packed beef. J Food Technol 8:333–48.Huffman RD. 2002. Current and future technologies for the decontamination of carcasses and

fresh meat. Meat Sci 62:285–94.Hunt MC, Action JC, Benedict RC, Calkins CR, Cornforth, DP, Jeramiah LE, Olson DC, Salm

CP, Savell JW, Shivas SD. 1991. Proceedings of 44th Annual Reciprocal Meat Conference,9–12 July 1991. Manhattan, Kansas: Kansas State University. p 3–170.

Jimenez-Villarreal, JR, Pohlman FW, Johnson ZB, Brown Jr. AH. 2003a. Effects of chlorinedioxide, cetylpyridinium chloride, lactic acid and trisodium phosphate on physical, chemicaland sensory properties of ground beef. Meat Sci 65:1055–62.

Jimenez-Villarreal, JR, Pohlman FW, Johnson ZB. Brown Jr. AH. 2003b. Lipid, instrumen-tal color and sensory characteristics of ground beef produced using trisodium phosphate,cetylpyridinium chloride, chlorine dioxide or lactic acid as multiple antimicrobial interven-tions. Meat Sci 65:885–91.

Kim JW, Slavik MF. 1994. Trisodium phosphate treatment of beef surfaces to reduce Escherichiacoli O157:H7 and Salmonella Typhimurium. J Food Sci 59:20–2.

Koohmaraie M, Arthur TM, Bosilevac JM, Guerini M, Shackelford SD, Wheeler TL. 2005.Post-harvest interventions to reduce/ eliminate pathogens in beef. Meat Sci 71:79–91.

King DA, Lucia LM, Castillo A, Acuff GR, Harris KB, Savell JW. 2005. Evaluation of peroxy-acetic acid as a post-chilling intervention for control of Escherichia coli O157:H7 and Salmonella Typhimurium on beef carcass surfaces. Meat Sci 69:40–407.

Mohan A, Dunn J, Hunt MC, and Sizer CE. 2009. Inactivation of Bacillus atrophaeus sporeswith surface-active peracids and characterization of formed free radicals using electron spinresonance spectroscopy. J Food Sci 74:M411–7.

Mohan A, Muthukrishnan S, Hunt MC, Barstow TJ, Houser TA. 2010a. Kinetics of myoglobinredox form stabilization by malate dehydrogenase. J Agric Food Chem 58:6994–7000.

Mohan A, Hunt MC, Muthukrishnan S, Barstow TJ, Houser TA. 2010b. Myoglobin redox formstabilization by compartmentalized lactate and malate dehydrogenases. J Agric Food Chem58:7021–9.

Mohan A, Hunt MC, Barstow TJ, Houser TA, Muthukrishnan S. 2010c. Effects of malate,lactate, and pyruvate on myoglobin redox stability in homogenates of three bovine muscles.Meat Sci 86:304–10.

Morrissey PA, Buckley DJ, Sheehy PJA, Monahan FJ. 1994. Vitamin E and meat quality. ProcNutr Soc 53:289–95.

Pohlman FW, Stivarius MR, McElyea KS, Waldroup AL. 2002a. Reduction of E. coli SalmonellaTyphimurium, coliforms, aerobic bacteria, and improvement of ground beef color usingtrisodium phosphate or cetylpyridinium. Meat Sci 60:349–56.

Pohlman FW, Stivarius MR, McElyea KS, Waldroup AL. 2002b. Reduction of microorganismin ground beef using multiple intervention technology. Meat Sci 61:315–22.

Pohlman FW, Morse DPN, Quilo SA, Brown Jr. AH, Crandall PG, Baublits RT, Story RP,Bokina C, Rajaratnam G. 2009. Microbial, instrumental color and sensory characteristics ofground beef processed from beef trimmings treated with potassium lactate, sodium metasilicate,peroxyacetic acid or acidified sodium chlorite as single antimicrobial interventions. J MuscleFoods 20:54–69.

Quilo SA, Pohlman FW, Brown Jr AH, Crandall PG, Morse DPN, Baublits RT, AparicioJL. 2009. Effects of potassium lactate, sodium metasilicate, peroxyacetic acid, and acidifiedsodium chlorite on physical, chemical and sensory properties of ground beef patties. Meat Sci82:44–52.

Siragusa GR. 1995. The effectiveness of carcass decontamination systems for controllingthe presence of pathogens on the surfaces of meat animal carcasses. J Food Safety 15:229–38.

Sofos JN. 2008. Challenges to meat safety in the 21st century. Meat Sci 78:3–13.Stivarius MR, Pohlman FW, McElyea KS, Apple JK. 2002. Microbial, instrumental color and

sensory color and odor characteristics of ground beef produced from beef trimmings treatedwith ozone or chlorine dioxide. Meat Sci 60:299–305.

M6 Journal of Food Science � Vol. 00, Nr. 0, 2012