Use of a Scald Additive to Reduce Levels of Salmonella Typhimurium During Poultry Processing

Transcription

1 Use of a Scald Additive to Reduce Levels of Salmonella Typhimurium During Poultry Processing S. R McKee, 1 J. C. Townsend, and S. F. Bilgili Department of Poultry Science, Auburn University, Auburn, AL ABSTRACT This study was conducted to evaluate the efficacy of a scald additive, RP scald, to reduce Salmonella Typhimurium (ST) levels on inoculated poultry carcasses. The RP scald (contains sodium hydroxide) in a 1% solution has a ph of 11.0, which may reduce bacteria levels on carcasses. In this study, 600 broilers (Ross 708 straight run, 6 wk of age) with 300 broilers in each of 2 experimental trials were divided into 4 scald treatments (inoculated with ST) and 2 noninoculated groups. The treatment groups included 4 scald treatments (n = 50 per experimental group per trial): soft scald (SS; 50 C for 90 s), soft scald with 1.0% added RP scald (SSRP), hard scald (56.6 C for 45 s; HS), and hard scald with 1.0% added RP scald. The Key words: scald additive, Salmonella, hard scald, soft scald, antimicrobial treatment INTRODUCTION Salmonella continues to be one of the most common causes of foodborne illness in the United States, with Salmonella Typhimurium (ST) being one of the most common serotypes causing salmonellosis (CDC, 2006). Under the current HACCP system, all broiler processing plants are required to meet Salmonella performance standards that are based on the national baseline study conducted in 1994 and Under these standards, if 20% or more broiler carcasses test positive for Salmonella, then the processing plant fails to comply. A study in the United States found that 33.9% of retail broiler carcasses tested positive for Salmonella over a 20-wk sampling period (Simmons et al., 2003). Therefore, additional intervention strategies during various stages of poultry processing are needed to help plants reduce Salmonella on poultry. Adding alkaline scald additives to scald tanks may help reduce bacterial loads on carcasses. Humphrey et al. (1981) reported a reduction of fecal matter on feathers and a reduction in total bacterial counts when scald water 2008 Poultry Science Association Inc. Received February 6, Accepted April 3, Corresponding author: mckeesr@auburn.edu noninoculated groups (n = 50 per group per trial) are represented by SS0 and HS0. After defeathering, carcass rinses were collected for ST detection. Results indicated that inoculated broilers from hard scald with 1.0% added RP scald had the lowest Salmonella recovery, whereas carcasses from the SS treatment with no RP additive had the highest ST recovery. In trial 1, the SSRP was more effective in reducing ST than HS alone; however, this trend was not consistent. In trial 2, HS alone was more effective in ST reduction than SSRP. Within each scald temperature, the addition of RP scald increased ST reduction; therefore, RP scald may be effective in reducing ST on broiler carcasses in poultry scalder applications, particularly when hard scald temperatures are used Poultry Science 87: doi: /ps ph was adjusted to 9.0 ± 0.2 by adding sodium hydroxide to scald water. The RP scald (Duchem Industries, Newnan, GA) is a commercial scald additive that has been used in commercial scald tanks to help reduce the appearance of bruising on broilers. The active ingredient in RP scald is sodium hydroxide, which creates a highly alkaline environment when mixed with water. It is thought that the alkaline environment created by RP scald may reduce microbial levels in scalding applications. Although the high ph may reduce bacteria, some potential negative effects of using high ph compounds could include the possibility of inactivating chlorine in the chiller if the chiller ph was raised as a result of residual sodium hydroxide on the carcasses. Monitoring and controlling the ph of the chiller would remedy this issue. In addition to adding a high ph compound to the scalder, temperature of the scald water can also influence the levels of bacteria. Generally, 2 types of scald applications are used in poultry processing. These scalding methods are referred to as soft scald and hard scald. Although scalding times, temperatures, and scalder configurations vary considerably among plants, a soft scald generally refers to scalding at 53.0 C for 120 s, and this scalding method is designed to leave the waxy stratum corneum layer (cuticle) of the skin on the carcasses where pigments may be deposited (Sams, 2001). Alternatively, hard scald applications are higher temperatures for a shorter time, 1672

2 USE OF A SCALD ADDITIVE TO REDUCE SALMONELLA 1673 but this method removes the cuticle, resulting in a pale appearance of the carcass. Hard scald temperatures have a greater effect on reducing levels of bacteria compared with lower temperatures (Notermans and Kampelmacher, 1975). Because temperature alone may impact bacterial survival in the scalder, it is important to determine the isolated effect of scalding temperature as well as the combined effect of temperature with RP scald in terms of their ability to decrease Salmonella on inoculated carcasses. Therefore, the objective of this study was to validate the efficacy of RP scald to reduce Salmonella on inoculated carcasses when used in soft and hard scald applications. MATERIALS AND METHODS Salmonella Inoculum Preparation Test tubes containing 10 ml of tryptic soy broth (Acumedia, Acumedia Manufacturers Inc., Lansing, MI) were inoculated with a frozen culture of a nalidixic acidresistant strain of ST. After incubation at 37 C for 24 h, a 10- L loopful of the ST culture was streaked onto xylose lysine tergitol 4 Agar (XLT4; Acumedia, Acumedia Manufacturers Inc., Lansing, MI) containing Tergitol 4 supplement and 0.1% nalidixic acid (Difco Laboratories, Detroit, MI). All XLT4 plates used in this experiment contained the supplement and nalidixic acid so that the media would be selective for recovering the nalidixic acid-resistant strain of ST. The plates were incubated at 37 C for 48 h. Black, isolated colonies were picked from the XLT4 plates, and fresh tryptic soy broth tubes containing 0.1% nalidixic acid were inoculated with one colony per test tube. Test tubes were incubated 20 to 24 h. A stock culture of 10 8 ST was prepared. Fecal Slurry Preparation To mimic natural conditions, a fecal slurry was used as the vehicle for ST inoculation because fecal matter is generally positive for Salmonella in broiler flocks that are shedding the microorganism. Equal proportions of distilled water and fecal matter collected from the Auburn University Poultry Research Unit were used to make 500 ml of a fecal slurry. The fecal matter was autoclaved (121 C for 30 min at 30 psi, Getinge 522LS Steam Sterilizer, Rochester, NY) to kill any existing microorganisms and then cooled to room temperature. Microbial analyses of the autoclaved fecal matter were carried out to confirm that the autoclave cycle killed microorganisms. Approximately 180 ml of the autoclaved fecal matter was reserved for use as the sterile fecal slurry. A total of 300 ml of inoculated fecal slurry was prepared by adding 120 ml of the stock 10 8 ST inoculum to 180 ml of autoclaved fecal matter, giving a 1:1.5 ratio of ST to fecal matter. In other words, 1 ml of 10 8 ST was present in each 2.5 ml of the stock of inoculated fecal slurry. Broilers Broilers (Ross 708 straight run, 42 d of age, n = 600 total; Aviagen, Huntsville, AL) were used for the experiment. In trial 1, broilers (n = 300) were obtained from a commercial grower and transported to the Auburn University Poultry Research Unit (<1.5 h). Birds were subjected to an 8-h feed withdrawal but were maintained on water (ad libitum) until time of transport. In trial 2 (Ross 708 straight run, n = 300) were raised at the Auburn University Poultry Research Unit. Birds were randomly placed into floor pens (0.08 m 2 / bird). Birds were provided standard cornsoy broiler starter (fed d1to15), grower (fed d 15 to 29), finisher (fed d 29 to 43), and withdrawal (fed d 43 to 49) feeds containing 22.0, 20.0, 17.8, 16.6% CP and 3,085, 3,115, 3,140, and 3,175 kcal of ME/kg, respectively. Rations and water were supplied ad libitum under continuous lighting. At 42 d of age and after an 8-h feed withdrawal (water was continued), birds were cooped (5 birds/coop) and taken to the Auburn University Poultry Research Processing Plant. Processing Parameters In each of 2 experimental trials (different days and different flocks of birds), birds (n = 300 total) were divided into 4 scald treatments (inoculated with ST) or 1 of 2 control groups (noninoculated) with a total of 25 birds per group in each of 2 replications (one morning and one afternoon, same flock of birds). Inoculated and control groups (noninoculated) consisted of birds subjected to soft scald, 50 C for 90 s (SS), soft scald with 1.0% added RP scald (SSRP), hard scald (HS; 56.6 C for 45 s), hard scald with 1.0% added RP scald (HSRP), or control groups, which were soft scald (SS0) and hard scald (HS0) with no inoculation and no RP scald additive. Treatment controls were SS and HS groups without RP scald. Before the start of each experimental group, approximately 100 additional broilers were processed at the Auburn University Poultry Processing Unit to increase the organic load in the scalder. Specifically, the broilers were hung on shackles, electrically stunned (50 V, 20 ma, 400 Hz) via 1% saline stunner bath (custom built, 1.52 m) with a metal plate running along the bottom attached to an electrical stun control box (model IA, Georator Corp., Manassas, VA). Following stunning, birds were killed by exsanguination through a unilateral neck cut followed by a 95-s bleed-out. After bleed-out, the birds were scalded in a 2.44-m-long single pass, steaminjected scalder (custom built, Cantrell, Gainesville, GA). The birds were defeathered in a 1.22-m-long disk-picker (custom built, Meyn, Oostzaan, the Netherlands) and removed from the processing line. During the experiment, soft scald treatments were run before hard scald treatments, and for each scalding treatment, the order of processing consisted of noninoculated controls (SS0 or HS0), inoculated birds subjected to soft and hard scald temperatures (SS or HS), and inoculated birds with RP scald (SSRP or HSRP) added to the scalder.

3 1674 MCKEE ET AL. Within each scald temperature (SS, 90 s at 50 C orfor HS, 45 s at 56.6 C), a 0.1% concentration of RP scald was achieved by adding 1.14 L of concentrated solution to 1, L of scald water. This level was chosen because it was consistent with recommended usage level by the manufacturer for scalding applications. After adding the RP scald, the solution was mixed by hand with plastic paddle mixers. Following addition of the RP scald, the remaining inoculated birds were processed. After the 3 soft scald treatments were processed and before starting the hard scald treatments, the scald tank was emptied, rinsed with water (110 C), cleaned (Turbo Zyme, St. Charles, IL, 30 min of contact time followed by rinse), and sanitized [ZEP Perosan (acidic peroxygen sanitizer), Zep Manufacturing Company, Atlanta, GA]. The sanitizer was applied at 149 ppm with a contact time of 10 min. After sanitizing, the scald tank was swabbed for microbial analysis and then refilled with water for processing of the hard scald treatments, which began noninoculated controls followed by HS and HSRP. The picker fingers were also rinsed (water 110 C), cleaned (Turbo-Zyme), rinsed, and sanitized (Zep Perosan acid) before the hard scald treatments were processed. Bird Inoculation and Experimental Treatment Application Birds were processed by hanging birds on the shackle of the processing line and processing (as described previously) through bleed-out. After bleed-out, birds were removed from the shackle line and placed breast-side-up on a stainless steel table. Each bird was inoculated with a sterile fecal slurry or fecal slurry containing 10 8 ST. Using sterile 2.5-mL scoops, the fecal slurry inoculum was administered in 2.5 ml aliquots (10 8 ST per bird) on the breast side along the feather tract so that the fecal slurry was applied directly on the skin of the broiler. The fecal slurries were allowed 10 min to dry. The birds were rehung on the shackles and passed through the scald tank for 90 s at 50 C or for 45 s at 56.6 C for soft scald and hard scald treatments, respectively. Carcass Sampling After each scalding treatment, the birds were defeathered and aseptically removed from the shackles. New York dressed carcasses were placed into sterile stomacher bags and rinsed with 200 ml of buffered peptone water (BPW, Difco) for 1 min using a rocking motion to assure that the exterior surfaces were rinsed. The rinse volume was altered from 400 ml as per USDA s protocol to 200 ml for this study because pre-experimental trials revealed better recovery of Salmonella from carcasses under the current experimental conditions. In pre-trial evaluation, the rinsate ph was checked to make sure the BPW could buffer the high ph scalder additive used. When carcasses were treated with a 1% solution of RP scald and rinsed in 200 ml, the BPW rinsate ph was approximately 7.5. Therefore, no additional ph adjustment was deemed necessary. The rinsate was then transferred to a sterile bottle and then divided into 3 parts for each analysis. All microbial analyses were based on methods described in the USDA/FSIS Microbiology Laboratory Guidebook (USDA, 2004) unless noted by modification. The rinsate was transferred into sterile bottles and then transported to the laboratory [<3 miles (<4.8 km)] for microbial analyses. Microbial Analyses Direct plating on XLT4 media with no pre-enrichment was used for the microbial analyses of the carcass rinses due to the high level of ST inoculum (10 8 ) in the fecal slurry. In duplicate, rinsate samples were direct plated by dispensing 1 ml of rinsate into a Petri dish, as well as dispensing 1 ml of rinsate into 9 ml of BPW per dilution tube. Serial dilutions were carried out through Dilutions were direct plated into Petri dishes, and approximately 20 ml of XLT4 media ( 50 C) containing Tergitol 4 supplement and 0.1% nalidixic acid was poured into the dishes. The Petri dishes were gently swirled to ensure proper dispersion of the bacteria. The plates were incubated at 37 C for 24 h, and typical Salmonella colonies were identified (Waltman et al., 1998) and counted. To determine the microbial status of the scalder surfaces, 3 distinct areas (front, middle, and back sections) of the interior of the scald tank were swabbed for Salmonella detection during 3 time frames: before processing began, after the tank was emptied and sanitized after each scald group, and after processing. A sterile sponge (Bio Pro Sampling Systems, International Bioproducts, Redmond, VA) was moistened with 10 ml of BPW, and the tank was swabbed. The swab was placed in a sterile sample bag containing 50 ml of BPW, bringing the total volume of BPW to 60 ml. The sponge samples were incubated at 37 C for 20 to 24 h. Next, 0.5 ml of the sample was transferred into 10 ml of Tetrathionate broth (Difco Laboratories) plus iodine supplement (Difco Laboratories), and 0.1 ml of the sample was transferred into 10 ml of Rappaport Vassiliadis broth (Difco Laboratories). The samples were incubated at 37 C for 22 to 24 h for the Tetrathionate broth and 42 C for 24 h for the Rappaport Vassiliadis broth. Following this step, samples were streaked for isolation on XLT4 agar plates using one 10 L loopful of sample for each plate. Presumptive positive Salmonella colonies were then identified (Waltman et al., 1998) and counted. For controls (SS0 and HS0) that were not inoculated and expected to be Salmonella negative, detection was reported as percentage positive. Water Sample Collection Scald water samples were collected for chemical oxygen demand (COD), biological oxygen demand (BOD), total water hardness (Mg and Ca), total soluble solids (TSS), and microbial detection of Salmonella in the scald water at the following stages: before processing began, after each soft scald treatment, after the scald tank was

4 USE OF A SCALD ADDITIVE TO REDUCE SALMONELLA 1675 refilled, before the second processing began, after each hard scald treatment, and after processing. Using sterile containers, three 500-mL samples of scald water were collected for COD, BOD, and TSS measurements. Water was collected from the front, middle, and end sections of the scald tank so that the water samples would be representative of the entire scald tank. In addition, three 100-mL scald water samples were collected in sterile containers from the front, middle, and end sections of the scald tank for water hardness and microbial detection of ST. Water samples were placed on ice, and COD and BOD measurements were carried out within 6 h of sampling. Water Quality Analyses During processing, ph and temperature of the scald water were monitored using a calibrated ph meter and digital thermometer (Thermo Orion ph Meter Model 720A Pittsburgh, PA). Measurements were taken at the following time frames: before processing began, after each soft scald treatment, after the scald tank was refilled, before the second processing began, after each hard scald treatment, and after processing. The measurements at each time interval were taken at both ends of the scald tank. Specifically, 10-mL water samples (5 samples per testing site per period) were collected and tested. For microbial detection of Salmonella in scald water samples, 50 ml of scald water was combined with 50 ml of BPW and enriched for 24 h at 37 C. Samples were tested for Salmonella using the culture methods described previously To measure COD, water samples were prepared in triplicate (500 ml each). Of the 500 ml, 20 ml was used for the heat of dilution technique for measuring COD as described by Boyd (1979). To measure biological oxygen demand, the 5-d BOD test was used. Specifically, scald water samples were collected in triplicate (500 ml each), and 200 ml of water sample plus 200 ml of distilled water was used to dilute the water sample. After the dilution, water samples were allowed time for the water temperature to equilibrate to 20 C. Two milliliters of seed inoculum solution and 1 nutrient buffer pillow were added to each BOD bottle. The diluted water samples were then transferred into BOD bottles, and the dissolved oxygen was measured on a polargraphic dissolved oxygen meter and probe (Model 57, YSI, Yellow Springs, OH). Dissolved oxygen measurements were obtained each day for 5 d. Scald water samples from each treatment were analyzed for total water hardness using a titration drop test kit (Model HA-4P, Hach Company, Loveland, CO), whereas TSS was measured in triplicate for each water sample. For TSS determination, glass fiber filters (Fisher Scientific) were dried in an oven at 103 C for 24 h and then weighed. Following this step, the water samples were shaken thoroughly to disperse the solids within each sample, and then 100 ml of the sample was passed through the glass fiber filters. The filters were removed, Table 1. Recovery of Salmonella from carcasses after treatment with RP scald additive combined with soft and hard scald temperatures in trial 1 Treatment % Positive cfu/sample Soft scald SS0 (noninoculated) 0 ND 1 SS (ST/no RP) 100 D 3.37 ± 0.21 d SSRP (ST with RP) 28 B 1.82 ± 0.34 b Hard scald HS0 (noninoculated) 0 ND HS (ST/no RP) 80 C 2.62 ± 0.43 c HSRP (ST with RP) 12 A 0.95 ± 0.15 a a d, A D Values within the same column with different superscripts are significantly different (P < 0.05). 1 ND = levels of Salmonella were not determined in presumed negative control groups. placed in an oven for 24 h at 103 C, and weighed back, and TSS was calculated. Statistical Analysis Colony-forming units were multiplied by 200 (200-mL carcass rinse solution used indicating each sample size) and then transformed to log 10 and reported as cfu/sample. Because 0 cannot be directly analyzed with the statistical model, we used 0.5 log 10 cfu for statistical analyses. Data were analyzed using the general linear model of SAS software (SAS Institute, 2003), and a probability level of P < 0.05 indicated significance. Differences and interactions between temperature, additive, and inoculum levels were compared using Duncan s multiple range tests. Results are reported means with standard errors. For Salmonella expressed as percent positive, positive samples were indicated by a 1 and negative samples were indicated by a zero. Data was subjected to the ANOVA procedure using the general linear model of SAS. Significance was indicated by a P < RESULTS AND DISCUSSION Differences in recovery of ST existed between trial 1 and trial 2. In trial 1, ST was not detected on the control birds (noninoculated) from the soft scald control or the hard scald control groups. However, in trial 2, the control birds (noninoculated) tested positive for Salmonella in the SS0 and HS0 treatments. As a result, levels of Salmonella recovery for trial 2 were higher, indicating that the control birds were contaminated before the start of the experiment. As a result, the data from trials 1 and 2 will be presented separately. Results from trial 1 are illustrated in Table 1. The average scald water temperature in trial 1 was C ± 1.95 (45 s) and C ± 1.51 (90 s) for hard scald and soft scald treatments, respectively. Results confirmed that the soft scald and hard scald (SS0 and HS0) noninoculated control birds remained negative (Table 1). Among the inoculated groups with no RP scald added, HS treatment had lower levels of ST recovered as well as fewer Salmonella-positive birds when compared with the SS treatment (Table 1). These results suggest that hard scald tempera-

5 1676 MCKEE ET AL. tures alone reduce ST regardless of the addition of RP scald. Other researchers have also reported that high scald temperatures reduced Salmonella on poultry carcasses. Specifically, Yang et al. (2001) reported that when the scalding temperature increased from 50 to 60 C, a 2.0 log cfu/cm 2 reduction occurred on chicken skin samples. In the same study by Yang et al. (2001), destruction of S. Typhimurium was observed at a scalding temperature of 55 C. In other studies, scalding at temperatures of 58 to 60 C resulted in greater reductions of Salmonella and Campylobacter, as compared with scalding at 52 C (Notermans and Kampelmacher, 1975; Notermans et al., 1977; Mulder et al., 1978; Oosterom et al., 1983; Wempe et al., 1983). When RP scald was added to the scald water, the ph of the scald water was increased from 7.80 ± 0.18 to ± Results indicated that SSRP and HSRP treatment groups containing the scald additive were better at reducing ST on broiler carcasses than temperature alone (HS; Table 1). Additionally, when compared with SS and HS treatments, both the SSRP and HSRP treatments had lower percentage ST-positive carcasses with the HSRP treatment resulting in the lowest level of ST recovery and fewest percentage-positive carcasses overall (Table 1). These data suggest that raising the ph of the scalder water may be an effective means of reducing Salmonella during the scalding step. Furthermore, a synergistic effect was noted between increased scalding temperatures and a higher scald water ph ( 11.0) in reducing ST levels and number of ST-positive birds. Other researchers have also shown that when ph is greater than 9.0 conditions become bactericidal to Salmonella spp. (Jay, 2000). The mechanism by which an alkaline environment is thought to destroy bacteria is by altering the functioning of enzymes of microbial cells and the transport of nutrients into the cells (Jay, 2000). In a study by Humphrey et al. (1981), scald water ph was stabilized at ph 9.0 ± 0.2 by adding sodium hydroxide, and total bacterial counts in scald water were reduced. However, in the current study, ph measurements were higher than those reported by Humphrey et al. (1981), and higher scalding temperatures (HSRP) were utilized resulting in a significant reduction of Salmonella (Table 1). As stated previously, broilers from trial 2 indicated flock contamination due to the detection of Salmonella on control birds, SS0, and HS0 (Table 2). Trial 2 data are reported separately because a trial by treatment interaction was detected, and the levels of Salmonella were higher overall in trial 2. Specifically, all controls (noninoculated) regardless of scalding temperature were found to be positive for Salmonella (Table 2). Because the media used to detect the Salmonella contained nalidixic acid even for the controls, the Salmonella detected on the controls was nalidixic acid-resistant; however, it was not serotyped in the current study. Other than higher levels of Salmonella being recovered overall in trial 2, there was also a trend difference in treatment efficacy detected between the trials. Specifically, in trial 1 the SSRP (with additive) was more effective in reducing ST than HS temperature alone Table 2. Recovery of Salmonella from carcasses after treatment with RP scald additive combined with soft and hard scald temperatures in trial 2 Treatment % Positive cfu/sample Soft scald SS0 (noninoculated) 100 C ND 1 SS (ST/no RP) 100 C 6.21 ± 0.41 d SSRP (ST /RP) 98 C 4.41± 0.49 c Hard scald HS0 (noninoculated) 48 B ND HS (ST/no RP) 68 B 2.83 ± 0.34 b HSRP (ST/RP) 15 A 0.98 ± 0.21 a a d, A C Values within the same column with different superscripts are significantly different (P < 0.05). 1 ND = levels of Salmonella were not determined in presumed negative control groups. (Table 1). In contrast, in trial 2, the HS without RP was more effective than the SSRP (Table 2). The inability of RP scald to cause a greater Salmonella reduction on broilers in the soft scald treatment may be due to the higher level of background contamination on the broiler flock in trial 2. In addition to higher levels of Salmonella recovery, the number Salmonella-positive carcasses were higher in trial 2 within all the soft scald treatments regardless of the RP scald additive (Table 2). Furthermore, in trial 2 all of the hard scald treatments had lower levels of Salmonella recovery and fewer Salmonella-positive carcasses compared with all of the soft scald treatments including the SSRP treatment containing the additive (Table 2). Similar to trial 1, the HSRP treatment in trial 2 had the lowest recovery of Salmonella and fewer Salmonella-positive carcasses (Table 2). These results support the findings that an alkaline ph environment combined with higher scald temperatures is more effective for reducing Salmonella compared to using lower scalder temperatures that would be consistent with a soft scald application. During the 2 trials, water quality measurements such as BOD, COD, water hardness, and TSS were also evaluated to determine if conditions of the scald water affected the efficacy of RP scald as an antimicrobial. The average BOD measurements from scald water samples collected during processing were and mg/l for the hard scald and soft scald water samples, respectively. The average COD of scald water samples collected during processing were and mg/l for hard scald and soft scald water samples, respectively. In trial 1, average water hardness was mg/l in scald water samples. Average water hardness in trial 2 was mg/l. Average TSS measurements were and mg/l in the hard scald and soft scald water samples, respectively. These results suggest that organic matter was present in the scald water during this experiment. Furthermore, RP scald was effective in reducing Salmonella regardless of the organic matter present in the scald water in the current study. Scald water samples were also collected for microbial detection of Salmonella to evaluate the effects of RP scald on the survival of Salmonella in scald water. In trial 1, 33% of the water samples from the soft scald treatments tested positive for Salmonella, whereas only 20% of sam-

6 USE OF A SCALD ADDITIVE TO REDUCE SALMONELLA 1677 ples from the hard scald treatments were positive (data not shown in table). Similarly, in trial 2, 40 and 20% of scald water samples tested positive for Salmonella in the soft scald and hard scald treatments, respectively. In trials 1 and 2, the water samples the treatments containing RP scald were all Salmonella-negative, which indicates that the high ph environment may be an antimicrobial in the scald water. Results from this study indicate that RP scald could be implemented as an effective intervention strategy in poultry processing plants to reduce the incidence of Salmonella during scalding. However, RP scald was shown to be more effective when used in combination with hard scalding temperatures. REFERENCES Boyd, C. E Determination of total ammonia nitrogen and chemical oxygen demand in fish culture systems. Trans. Am. Fish. Soc. 108: Centers for Disease Control and Prevention (CDC) Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food 10 states, United States, MMRW 55: Humphrey, T. J., D. G. Lanning, and D. Beresford The effect of ph adjustment on the microbiology of chicken scaldtank water with particular reference to the death rate of salmonellas. J. Appl. Bacteriol. 51: Jay, J. M Modern Food Microbiology. 6th ed. Aspen Publishers Inc., Gaithersburg, MD. Mulder, R. W. A. W., L. W. J. Dorresteijn, and J. Van Der Broek Cross-contamination during the scalding and plucking of broilers. Br. Poult. Sci. 19: Notermans, S., and E. H. Kampelmacher Heat destruction of some bacterial strains attached to broiler skin. Br. Poult. Sci. 16: Notermans, S., F. M. Van Leusden, and M. Van Schothorst Suitability of different bacterial groups for determining faecal contamination during post-scalding stages in the processing of broiler carcasses. J. Appl. Bacteriol. 43: Oosterom, J., S. Notermans, H. Karman, and G. B. Engels Origin and prevalence of Campylobacter jejuni in poultry processing. J. Food Prot. 46: Sams, A. R First processing: Slaughter through chilling. Pages in Poultry Meat Processing. A. R. Sams, ed. CRC Press, Boca Raton, FL. SAS Institute SAS/STAT Guide for Personal Computers, Version 9.1 Edition. SAS Institute Inc., Cary, NC. Simmons, M., D. L. Fletcher, J. A. Cason, and M. E. Berrang Recovery of Salmonella from retail broilers by a wholecarcass enrichment procedure. J. Food Prot. 66: USDA, Food Safety and Inspection Service Isolation and Identification of Salmonella from Meat, Poultry, and Egg Products. Microbiological Laboratory Guidebook. MLG Waltman, W. D., R. K. Gast, and E. T. Mallinson Salmonellosis. Pages 4 13 in A Laboratory Manual for the Isolation and Identification of Avian Pathogens. 4th ed. D. E. Swayne, J. R. Glisson, M. W. Jackwood, J. E. Pearson, and W. M. Reeds, ed. Am. Assoc. Avian Pathol., Kennett Square, PA. Wempe, J. M., C. A. Genigeorgis, T. B. Farver, and H. I. Yusufu Prevalence of Campylobacter jejuni in two California chicken processing plants. J. Appl. Environ. Microbiol. 45: Yang, H., Y. Li, and M. G. Johnson Survival and death of Salmonella Typhimurium and Campylobacter jejuni in processing water and on chicken skin during poultry scalding and chilling. J. Food Prot. 64: