Alternative indicator bacteria analyses for evaluating the sanitary condition of beef carcasses

Sponge samples were obtained from 47 (study 1) and 32 (study 2) beef carcasses in a small plant over 6 months. In study 2, slaughter equipment surfaces were also sampled. In study 1, the Petrifilm method was used to count presumptive Escherichia coli and spread plating on kanamycin esculin azide (KEA) agar with and without 40% added bile was used to count presumptive Enterococcus spp. Qualitative testing for presumptive E. coli and Enterococcus spp. in study 1 was done using lauryl sulfate tryptone broth (LST) + 4-methylumbelliferyl-beta-D-glucuronide (MUG) and KEA + 40% bile broth, respectively. In study 2, LST + MUG was used as a most probable number (MPN) method along with the Petrifilm method. In the two studies, 8 (17.0%) and 11 (34.4%) carcasses were contaminated with presumptive E. coli; all but one contaminated carcass contained <1 CFU/cm2. Presumptive Enterococcus spp. were recovered from 15 carcasses (31.9%) in study 1, but the KEA + 40% bile agar method lacked specificity (only 31.3% of isolates confirmed as Enterococcus spp.) The LST + MUG and Petrifilm methods were significantly (P < 0.05) related in terms of detecting presumptive E. coli, but the presence of presumptive Enterococcus spp. was not significantly related to the presence of presumptive E. coli. However, on slaughter plant equipment in Study 2 there was a statistically significant (P < 0.05) relationship between the presence of presumptive E. coli and presumptive Enterococcus spp. In study 2, there was no significant (P < 0.05) difference in numbers of presumptive E. coli (obtained using Petrifilm) on carcasses chilled 1 day (n = 16) and 7 days (n = 16), although more of the 7-day carcasses were contaminated (five and seven carcasses, respectively). For samples testing positive for presumptive E. coli, the 95% confidence intervals obtained using the LST + MUG MPN method included the Petrifilm value for all but one sample.     

Consumer assessment of beef palatability from four beef muscles from USDA Choice and Select graded carcasses

Consumer sensory analysis was conducted to determine differences in beef palatability between two quality grade categories [Upper 2/3 (Top) Choice and Select] and four muscles [longissimus lumborum (LL), gluteus medius (GM), serratus ventralis (SV), and semimembranosus (SM)]. Generally, tenderness, flavor, and overall liking scores were more desirable for Top Choice compared to Select, regardless of muscle. Consumers rated LL as more tender (P < 0.05) than SV and SM, but similar to GM (P = 0.52). Overall and flavor acceptability were similar (P > 0.05) between LL, GM, and SV, regardless of quality grade. Consumer overall liking was most highly correlated with flavor liking (r = 0.85). When tenderness was acceptable, flavor and juiciness played a major role in determining overall acceptability. Overall liking of GM and SV from Top Choice carcasses was superior to LL from Select carcasses and comparable to LL from Top Choice carcasses.     

The electrical stimulation of beef carcasses

Beef carcasses were stimulated shortly after slaughter at varying voltage, pulse frequency and duration, with the aim of accelerating the fall of pH, destruction of ATP and onset of rigor in the muscles, so that the meat could be chilled rapidly after slaughter without danger of cold-shortening. Optimal effects were produced at 700 V, 25 Hz for 2 min, to give a total of 3000 pulses. In undressed carcasses stimulation induced a pH fall to 6.0 within 1 h of slaughter and to 5.7 within 2.5 h in the major muscles of the forelimb, back and thigh, representing a gain of more than 8 h over the time required in unstimulated carcasses hanging at 16°C. Similar results were obtained with dressed carcasses and sides, after allowing for the 50 min or so lost in dressing. The apparently high rate of pH fall from 6.3 to 5.7 after stimulation had ceased, compared with that over the same pH range in unstimulated carcasses, can be attributed mainly to the different muscle temperatures in the two cases (38 and 26°C respectively, in four major muscles). The muscles of stimulated carcasses showed no deleterious effects of stimulation. The observed drip loss from the hindlimb jointed 6 days after slaughter was not significantly greater than that from unstimulated carcasses.     

In-Plant evaluation of a lactic acid treatment for reduction of bacteria on chilled beef carcasses.

The effectiveness of a lactic acid treatment consisting of spraying a 4% L-lactic acid solution (55 degrees C at source) on chilled beef carcasses to reduce bacterial populations was tested in a commercial slaughter environment. All carcasses had been treated with a proprietary decontamination treatment composed of a hot water spray followed by a lactic acid spray prior to chilling. Bacterial groups used to indicate reductions included aerobic plate count (APC), total coliform count, and Escherichia coli count, and samples were examined from the brisket, the clod, and the neck regions of 40 untreated and 40 treated carcass sides. Depending on the carcass surface region, APCs were reduced by 3.0 to 3.3 log cycles. Log coliform and E. coli counts were consistently reduced to undetectable levels. The small reductions observed for coliforms are attributable to counts on untreated carcasses already being near the lower detection limit of the counting method. The percentage of samples with detectable numbers of coliforms (positive samples) on untreated carcasses ranged from 52.5 to 92.5%, while 0.0% of the samples collected from treated carcasses contained detectable coliforms. Percent E. coli-positive samples ranged from 7.5 to 30.0% on untreated carcasses and 0.0% after treatment of carcass sides. These results indicate that a hot lactic acid spray with increased concentration and time of application may be effectively implemented for an additional decontamination treatment of chilled beef carcasses prior to fabrication.     

Glycolysis and associated changes in beef carcasses

Post-mortem glycolysis was investigated at several locations in intact sides of beef. The time required for the pH to fall to 6.0 in six major hindquarter muscles ranged from 2.2 to 13.6 h, varying with muscle and depth in the carcass. A good correlation (r=0.97) was observed between the rate of ATP turnover and muscle temperature. In the muscles in the round the rate of glycolysis increased with depth, as did the extent of denaturation of myofibrils, creatine phosphokinase and also drip loss.     

Prevalence and antibiotic susceptibility of Salmonella isolated from beef animal hides and carcasses

This study determined the prevalence of Salmonella on beef animal hides and carcasses and antimicrobial susceptibility profiles against a panel of 13 antibiotics. In each of the eight commercial packing facilities, of which five processed primarily heifers and steers and the remaining three processed primarily cows and bulls, hide and carcass sponge swab samples were obtained immediately before hide removal and before carcass chilling, respectively. Overall, prevalence of Salmonella on external surfaces (hides) of cattle was 15.4% (49 of 319), whereas prevalence after dehiding and other slaughtering/dressing processes, including the application of decontamination treatments, was, as expected, reduced (P < 0.05) to 1.3% (4 of 320) on carcass surfaces. From 53 total Salmonella-positive hide and carcass samples, 526 biochemically confirmed isolates were obtained to determine antimicrobial susceptibility profiles. Of 53 Salmonella-positive samples, individually, 24 (45.3%), 17 (32.1%), 17 (32.1%), 11 (20.8%), 8 (15.1%), 8 (15.1%), 8 (15.1%), 4 (7.5%), and 2 (3.8%) samples yielded at least one isolate resistant to amoxicillin/clavulanic acid, tetracycline, streptomycin, sulfonamides, ampicillin, ampicillin/sulbactam, chloramphenicol, gentamicin, and trimethoprim/sulfamethoxazole, respectively. None of the Salmonella-positive samples yielded an isolate resistant to ceftriaxone, ciprofloxacin, enrofloxacin, or levofloxacin. Although none of the samples yielded an isolate simultaneously resistant to three or four antimicrobials, a total of eight samples yielded at least one isolate resistant to five or more antimicrobials tested. Included among the 18 group B-positive samples were three samples that, individually, yielded at least one Salmonella Typhimurium var. Copenhagen DT104 isolate resistant to at least six antimicrobials tested. Results from this study support current prudent therapeutic and subtherapeutic antimicrobial use recommendations.     

The tenderising effect of electrical stimulation of beef carcasses

Electrical stimulation (ES) of beef carcasses soon after death has an accelerated tenderising effect on the musculature, under conditions of slow cooling (8 h at 16°C and then storage in still air at 1°C). The effect is large in the LD muscles, reducing the shear force on day 1 of storage from 11 to 6 kg; on day 14, the difference is still 3·3 kg. These differences would be detected by taste panels. The St muscles show a similar, but less pronounced, effect which would probably not be detected by taste panels. The accelerated tenderisation due to ES can be accounted for by the higher temperatures obtaining in stimulated muscles at the onset of rigor. Rapid cooling soon after death reduces the effect almost to zero. Hence, the extra tenderisation cannot be due to muscle damage during ES. Histological examination shows that stimulated muscles have longer sarcomeres than the controls; they do not exhibit damage. However, with slow cooling, irregular bands of denatured sarcoplasmic protein are deposited within the fibres of stimulated muscles, similar to those found in PSE pig muscles. There is also some shortening of sarcomeres in the region of the bands. The protein is deposited on the myofibrillar surfaces. In spite of the PSE-like appearance, there is no significant increase in drip from the stimulated muscles at 48 h after death.     

Frequencies of injection-site lesions in muscles from rounds of dairy and beef cow carcasses

The frequency of injection-site lesions in muscles from top sirloins and rounds in fed cattle carcasses is well documented; this study characterizes the frequency and severity of lesions in muscles from rounds of beef and dairy cow carcasses. Audits were conducted in 1998, 1999, and 2000 on 3190 rounds from cow carcasses. Outside round muscles were cut into 1.25-cm slices to characterize lesions. In 1998, 31% of beef rounds and 60% of dairy rounds had an injection-site lesion. Frequency of lesions in beef rounds significantly declined 5 percentage points between 1998 and 1999 and 6 percentage points between 1999 and 2000. The frequency of lesions in dairy rounds significantly declined 9 percentage points between 1998 and 1999 and 16 percentage points between 1999 and 2000. Frequencies of injection-site lesions in muscles of beef rounds were significantly lower than those in muscles of dairy rounds in all 3 yr. Injection-site lesions were most common between the hooks and pins of the hindquarter of beef cattle and between the pins and hocks of the hindquarter of dairy cattle. Clear lesions and woody calluses exceeded 89% and occurred more frequently than did other kinds of lesions in muscles of beef and dairy rounds in 1998, 1999, and 2000 audits. Of all injection-site lesions, between 3 and 5% were cystic in muscles of beef rounds, similar to the 2 to 4% of cystic lesions found in muscles of dairy rounds. Although yearly data indicate trends in declining frequencies of injection-site lesions, the need remains for educational programs and continued improvements in beef quality assurance practices among both beef and dairy producers.     

Antibiotic resistance among coliform bacteria isolated from carcasses of commercially slaughtered chickens

A total of 322 coliform bacteria Escherichia coli, Enterobacter spp., Citrobacter spp., Klebsiella spp. and Serratia spp., were isolated from 50 carcasses of commercially slaughtered chickens. Their resistance to ampicillin, tetracycline, gentamicin, chloramphenicol, cephalotine, cotrimoxazole, nalidixic acid and nitrofurantoin, were determined. The most commonly found resistance was to tetracycline followed by cephalotine, cotrimoxazole and nalidixic acid. A large percentage of E. coli (41%) and Klebsiella spp. (38%) showed multiple antibiotic resistance.     

Hot boning of intact carcasses: a procedure to avoid central nervous system self-contamination in beef and beef products

Compared with traditional boning of split refrigerated carcasses, hot boning of intact carcasses (the removal of meat from the skeleton prerigor) provides several commercially important cuts, may improve quality and reduce refrigeration costs, and may reduce the contamination of carcasses with central nervous system (CNS) tissue. In a comparative study of hot boning of intact and split carcasses, the CNS tissue contamination of intact carcasses was negligible (as measured with the CNS-related proteins glial fibrillary acidic protein and S-100 protein), but split carcasses were highly contaminated. The same trends were observed for dissection worktables used during the boning process. Most current boning plants have processing lines that are organized for boning carcass quarters, where the carcasses, in addition to transversal division, also are split horizontally. This part of the boning process was incorporated in the design of our study. Nine of the 18 intact carcasses were split horizontally between thoracic vertebrae 10 and 11 before they were hot boned. CNS tissue contamination was not detected on the carcass site related to this procedure. The amount of CNS tissue contamination was similar in boned cuts and minced meat from split and intact carcasses, except in the forerib. Boning of split carcasses appears to reduce CNS tissue contamination significantly to a level comparable to that of intact hot-boned carcasses.     

Calpain-calpastatin and toughness in M. longissimus from electrically stimulated lamb and beef carcasses

Shear strength, pH, temperature, μ-calpain, m-calpain and calpastatin levels were measured over a two-week post-slaughter period in Longissimus lumborum et thoracis (LD) from six lamb and six beef carcasses. All carcasses were subjected to high voltage electrical stimulation. The toughness of the beef LD determined by a MIRINZ tenderometer at 24 h post-slaughter showed a strong correlation (r=0.91) with pH of the LD at 3 h. Beef LD toughness at 14 days was correlated (r=0.84) with initial m-calpain levels. In both lamb and beef, LD toughness at 4 and 14 days respectively was also correlated with initial levels of calpastatin (r=0.85, 0.83, respectively). The strong correlation between calpastatin and the rate of tenderisation indicates that the calpain system is closely linked to the proteolytic breakdown of myofibrillar proteins. There is also evidence of an interaction between pH and μ-calpain activity. The μ-calpain, m-calpain, calpastatin, pH and temperature kinetic changes which occurred during the post-mortem ageing of beef and lamb LD were applied to a computer program which predicted rate of meat tenderisation by calculating in situ calpain activity. The closeness of fit between the predicted rate of meat tenderisation and the observed tenderness values of beef and lamb LD indicates that the post-mortem activity of μ-calpain is the major determinant of variations in tenderness. However, application of the meat tenderisation predictive program to LD from individual animals revealed that the program was not sufficiently robust for this use.     

Number and distribution of bacteria on some beef carcasses at selected abattoirs in some member states of the European communities

In seven member countries of the European Communities, three abattoirs were visited on three occasions in each of two surveys and at each visit ten beef carcasses were sampled, before chilling, at defined sites on the neck, brisket, forerib and medially on the round. In Survey I, samples were plated for total viable count (TVC) at 30° (ISO 2293) and Enterobacteriaceae at 37° (ISO 5552); in Survey II only TVCs were made. This paper is confined to analyses of the TVCs in the two surveys. Data from each country were analysed separately as sampling methodology may not have been sufficiently reproducible by different workers to allow between-countries comparison. Variations among visits to particular abattoirs and abattoir × site interactions made comparisons among abattoirs invalid within five of the seven countries. To effectively monitor differences between abattoirs within most countries it would be necessary to make more than three visits to each abattoir. Despite abattoir × site interactions in three countries in Survey I and four countries in Survey II, comparisons between sites were generally valid because of the consistent high contamination of the brisket. In the remainder of countries the abattoir × site interaction was too large to allow valid comparisons between sites. It is recommended that at least three or four sites are sampled in future surveys as only one site per carcass would underestimate the number of more heavily contaminated carcasses.     

Attachment of bacteria to meat surfaces: A review

The mechanism of attachment of bacteria to meat surfaces involves two consecutive stages: reversible and irreversible attachment. The dependence of attachment on various factors is discussed: hydrophobicity and cell surface charge, bacterial structures, type of meat surface, pH, temperature, presence of chemical substances, etc. The consequences of these findings for slaughter hygiene, kitchen hygiene, etc., are clear; avoid bacterial contamination.     

The detection of central nervous system tissue on beef carcasses and in comminuted beef.

We report the development and validation of a fluorescent enzyme-linked immunosorbent assay (ELISA) for glial fibrillary acidic protein (GFAP), which can be used as a rapid and sensitive method to detect CNS tissue in meat products. The fluorometric assay is sensitive to 0.2 ng GFAP and has an intra-assay coefficient of variation (CV) of 2.0% and an interassay CV of 14.1%. Bovine spinal cord and brain demonstrate dose-response curves that are parallel to GFAP standards, whereas peripheral sciatic nerve and cervical ganglia also cross-react at high tissue levels. The use of another central nervous system marker, syntaxin 1-B, was not effective for neural tissue detection. Less than 1.0 ng GFAP per mg tissue was found on most beef subprimals and advanced meat recovery (AMR) product. Occasional samples contained higher levels of GFAP, probably because of contamination by the carcass-splitting saw, incomplete removal of the spinal cord, or a chance sampling of a major nerve. Further reduction of CNS content was facilitated by removal of the cervical vertebrae and the spinal canal prior to processing beef chuck bones through AMR equipment. The presence of GFAP was very low (0.037 ng/mg) in beef patties collected from major processors throughout the USA. The presence of normal sausage ingredients or heating the product to 80 degrees C for 60 min did not affect the detection of GFAP. Heating the product to 115 degrees C for 100 min eliminated the detectability of GFAP.     

A national survey of the microbiological quality of beef carcasses and frozen boneless beef in Australia

The third national baseline microbiological survey of Australian beef carcasses and frozen boneless beef was conducted in 2004. Carcasses (n=1155) sampled at 27 slaughter establishments had a mean aerobic plate count (at 25 degrees C) of 1.3 log CFU/cm2. Escherichia coli was isolated from 8.0% of the cacasses, with a mean count of -0.8 log CFU/cm2 for positive samples. On samples from 24 boning (fabrication) plants (n=1082), the mean aerobic plate count for frozen boneless beef was 1.3 log CFU/g, and the mean count for the 1.8% of samples with detectable E. coli was 1.5 log CFU/g. E. coli O157: H7 was isolated from 1 of 1,143 carcasses and from 0 of 1082 boneless samples. Salmonella was isolated from 0 of 1155 carcasses and from 1 of 1082 samples of boneless product. No Campylobacter spp. were isolated from carcasses or boneless beef. Coagulase-positive staphylococci were isolated from 28.7% of beef carcasses and 20.3% of boneless beef samples, and positive samples had a mean count of 0.3 log CFU/cm2 and 0.8 log CFU/g, respectively.     

Attachment of shiga toxigenic Escherichia coli to beef muscle and adipose tissue

Shiga toxigenic Escherichia coli (STEC) serotypes are important foodborne pathogens that cause gastrointestinal disease worldwide. An understanding of how STEC strains attach to surfaces may provide insight into the potential persistence of and contamination with STEC in food environments. The initial attachment of a selection of STEC serotypes to beef muscle and adipose tissue was evaluated for isolates grown in planktonic and sessile culture. Initial experiments were performed to determine whether attachment differed among STEC strains and between the two modes of growth. Viable counts were obtained for loosely and strongly attached cells, and the strength of attachment (Sr) was calculated. All bacterial isolates grown in sessile culture attached in higher numbers to muscle and adipose tissue than did bacteria in planktonic cultures. For all attachment assays performed, mean concentrations for loosely attached cells were consistently higher than concentrations for strongly attached cells. The mean concentrations for strongly attached bacteria for planktonic and sessile cultures were significantly higher (P < 0.05) on adipose than on muscle tissue. However, some strains of STEC, particularly those from sessile culture, did not differ in their attachment to muscle or adipose tissue. Sr values were not significantly different (P > 0.05) among STEC isolates for all assays. No correlation was found between bacterial hydrophobicity and surface charge values (previously determined) and production of surface structures, viable counts, and Sr values. STEC grown in planktonic and sessile culture seems to behave differently with respect to attachment to muscle and adipose tissue. Cells in sessile culture may have a greater potential to strongly attach to meat surfaces.     

Attachment of Salmonella choleraesuis choleraesuis to beef muscle and adipose tissues.

The attachment of Salmonella choleraesuis subsp. choleraesuis ATCC 15790 to beef muscle and adipose tissues was investigated. S. choleraesuis was found to adhere in higher numbers to muscle than to fat. The charge and the hydrophobicity of the surface of S. choleraesuis were evaluated by measurement of electrophoretic mobility, the contact angle with water, adhesion to hexadecane, and hydrophobic interaction chromatography. The overall negative charge of S. choleraesuis was masked by the high electrolyte concentration in the attachment medium (0.15 M phosphate-buffered saline). This bacterium was shown to possess a hydrophilic surface. Electrostatic interactions do not affect the attachment of S. choleraesuis to both lean and fat tissue, and there was no evidence for a role of hydrophobic interactions. However, the attachment of S. choleraesuis was reduced by 90% after mechanical removal of the flagella or after treatment of the bacteria with specific antiflagella serum. This reduction was attributed to a loss of bacterial mobility leading to a reduction in the number of cells reaching the tissue during the period of contact. Treatment of the tissue with a concentrated suspension of flagella or treatment of the bacteria with antisomatic serum (OMD) did not reduce the attachment of S. choleraesuis to tissues, indicating an absence of specific attachment sites for flagella or antigen O on the beef tissue surface.     

Studies of extra low voltage electrical stimulation of mature beef carcasses

The quality characteristics of biceps femoris (BF) and semimembranosus (SM) roasts obtained from mature cow carcasses treated with a commercial extra low voltage (30 V) electrical stimulation (LVES) system were determined. LVES was applied for either 2 (ESII) or 4 min (ESIII). Evaluations were conducted on meat obtained from control sides (no ES) aged for either 48 h (Ia) or 7 days (Ib) and from ES sides aged 48 h. ES caused a reduction (P<0·001) in pH values at 2 and 6 h post mortem. At 24 h, the pH of muscles from all carcasses was about 5·5. ES duration did not influence muscle pH. Rib-eye muscle colour for ESII and ESIII carcasses was lighter and brighter (P<0·05) than that of control carcasses. Generally stimulated BF roasts had greater cooking losses than control Group Ib roasts; SM roasts from ES carcasses had lower losses than comparable to Group Ib roasts. ES duration had no effect on per cent cooking losses. Trained panelists generally detected few significant effects in BF roasts due to ES. Warner-Bratzler data indicated that ESII and ESIII BF roasts were similar and significantly more tender than comparable control Group Ib samples: OTMS data indicated that all BF roasts were similar in tenderness. However, SM roasts from ES carcasses were judged more soft (Groups II and III) and tender (ESII) than comparable control roasts. Instrumental measurements of tenderness for SM roasts tended to support the taste panel results. Generally, duration of LVES had no effect on the eating quality of either BF or SM roasts. Since LVES effects on the palatability of SM roasts were evident but the effects of stimulation of BF roasts were few, further studies of this LVES system are needed before its use can be recommended. Generally, increasing post-mortem ageing time for mature control carcasses did not influence either BF or SM roast quality.     

Importance of airborne contamination during dressing of beef and lamb carcasses

Carcasses along slaughter lines were exposed to normal slaughterhouse air or ultraclean air provided from a unit fitted with a HEPA filter. In cattle slaughterhouses, aerobic viable counts were measured by sponging the brisket at the end of the line to determine whether the slaughterhouse air had led to contamination of the carcasses. Furthermore, a replica cattle carcass with settle plates attached was exposed to similar conditions. The greatest contamination of the plates occurred at the hide puller (P < 0.01). The use of ultraclean air reduced the deposition of organisms onto settle plates (P < 0.01). The airborne route contributed to contamination in cattle slaughterhouses, but other vectors were more important. Further study of contamination of the brisket, at the time that it was first exposed, showed that knives transfer contamination from the hide. The use of ultraclean air at this position showed that the airborne route was a contributor to contamination (P < 0.1), but it was not the greatest vector. In lamb slaughterhouses, the highest counts on settle plates were found at the fleece puller (P < 0.05). The highest counts on the lamb carcasses were found on the brisket exposed from the start of the line to just after the fleece puller (P < 0.05). There was no clear relationship between the measured counts and the concentration of organisms in the air, indicating that the airborne route in lamb slaughterhouses contributes less to carcass contamination than do the surface contacts.     

Prevalence on beef carcasses of Mycobacterium avium subsp. paratuberculosis DNA

Fifty samples were collected from each of skinned and dressed carcasses, from each of culled beef breeding cows and fed beef cattle <18 months old at two beef packing plants A and B, and from culled dairy cows at a packing plant C. The 450 samples were collected by swabbing an area of about 1000 cm2 in the anal region of each carcass. DNA extracted from each swab was tested for the IS900 and F57 sequences of the Mycobacterium avium subsp. paratuberculosis (MAP) genome by two stage, nested polymerase chain reaction (PCR) procedures. An internal amplification control (IAC) was detected in 45 or more of each group of 50 DNA preparations. IS900 and F57 were detected in some IAC-positive preparations from all and all but one of the groups of carcasses, respectively. Of the IAC-positive preparations in each group, between 6 and 54% were positive for IS900, and between 4 and 20% were positive for F57. When preparations were tested by single stage, quantitative PCR procedures, IS900 was detected in two samples but F57 was detected in none. The MAP DNA on carcasses was probably derived from small numbers of MAP from the environment that contaminated the animals' hides.