Rapid detection of microbial contamination in frozen vegetables by automated impedance measurements.

Automated impedance measurements can be used to rapidly assess whether a sample of frozen vegetables contains greater or less than 10(5) organisms per g. Microorganisms growing pureed food samples cause a change in the impedance of the medium when the organisms reach a threshold concentration of between 10(6) and 10(7) organisms per ml. Estimates of the concentration of microorganisms initially present in the food sample can be made by recording the time required for the organisms in the sample to replicate to threshold levels. In this study, the detection times for 357 samples of frozen vegetables were compared with standard plate counts for each sample. The agreement between the two methods in distinguishing samples containing more than 10(5) organisms per g was 92.6% for 257 assorted frozen vegetables and somewhat higher (93 to 96%) when separate cutoff times were used for each type of vegetable. The time required for analysis was about 5 h, compared to the 48 to 72 h required for standard plate counts.     

Detection of microbial concentration in ice-cream using the impedance technique

The detection of microbial concentration, essential for safe and high quality food products, is traditionally made with the plate count technique, that is reliable, but also slow and not easily realized in the automatic form, as required for direct use in industrial machines. To this purpose, the method based on impedance measurements represents an attractive alternative since it can produce results in about 10 h, instead of the 24–48 h needed by standard plate counts and can be easily realized in automatic form. In this paper such a method has been experimentally studied in the case of ice-cream products. In particular, all main ice-cream compositions of real interest have been considered and no nutrient media has been used to dilute the samples. A measurement set-up has been realized using benchtop instruments for impedance measurements on samples whose bacteria concentration was independently measured by means of standard plate counts. The obtained results clearly indicate that impedance measurement represents a feasible and reliable technique to detect total microbial concentration in ice-cream, suitable to be implemented as an embedded system for industrial machines.     

Heterotrophic plate counts of surface water samples by using impedance methods.

Membrane filtration, spread plating, and pour plating are conventional methods used to determine the heterotrophic plate counts of water samples. Impedance methods were investigated as an alternative to conventional methods, since sample dilution is not required and the bacterial count can be estimated within 24 h. Comparisons of impedance signals obtained with different water samples revealed that capacitance produced faster detection times than conductance. Moreover, the correlation between heterotrophic plate count and detection time was highest (r = 0.966) when capacitance was used. Linear and quadratic regressions of heterotrophic plate count and impedance detection time were affected by incubation temperatures. Regressions between heterotrophic plate counts based on conventional methods and detection times of water samples incubated at less than or equal to 25 degrees C had R2 values of 0.878 to 0.933. However, regressions using detection times of water samples incubated at greater than or equal to 30 degrees C had lower R2 values, even though water samples produced faster detection times. Comparisons between broth-based versions of R2A medium and plate count agar revealed that the latter correlated highly with heterotrophic plate count, provided that water samples were incubated at 25 degrees C and impedance measurements were conducted with the capacitance signal (r = 0.966). When the linear regression of this relationship was tested with 100 water samples, the correlation between predicted and actual log10 CFU milliliter-1 was 0.869. These results indicate that impedance methods provide a suitable alternative to conventional methods.     

Heterotrophic plate counts of surface water samples by using impedance methods.

Membrane filtration, spread plating, and pour plating are conventional methods used to determine the heterotrophic plate counts of water samples. Impedance methods were investigated as an alternative to conventional methods, since sample dilution is not required and the bacterial count can be estimated within 24 h. Comparisons of impedance signals obtained with different water samples revealed that capacitance produced faster detection times than conductance. Moreover, the correlation between heterotrophic plate count and detection time was highest (r = 0.966) when capacitance was used. Linear and quadratic regressions of heterotrophic plate count and impedance detection time were affected by incubation temperatures. Regressions between heterotrophic plate counts based on conventional methods and detection times of water samples incubated at less than or equal to 25 degrees C had R2 values of 0.878 to 0.933. However, regressions using detection times of water samples incubated at greater than or equal to 30 degrees C had lower R2 values, even though water samples produced faster detection times. Comparisons between broth-based versions of R2A medium and plate count agar revealed that the latter correlated highly with heterotrophic plate count, provided that water samples were incubated at 25 degrees C and impedance measurements were conducted with the capacitance signal (r = 0.966). When the linear regression of this relationship was tested with 100 water samples, the correlation between predicted and actual log10 CFU milliliter-1 was 0.869. These results indicate that impedance methods provide a suitable alternative to conventional methods.     

Rapid enumeration of microorganisms in foods by the direct epifluorescent filter technique.

Filtration of "stomachered" food suspensions through nylon filters (pore size, 5 microns) removed most of the food debris without affecting the recovery of microorganisms. Two to ten milliliters of these prefiltered suspensions could be filtered in the direct epifluorescent filter technique (DEFT). The technique takes less than 30 min to complete and has a lower sensitivity of less than 60,000 microorganisms per g for all products examined. Vegetative bacterial cells, spores, fungal hyphae, and yeasts could be distinguished with the technique. For fresh meat and fish, the DEFT count of prefiltered suspensions agreed well with the plate count of unfiltered suspensions over the range of 10(4) to 10(10)/g (correlation coefficient of 0.91). For frozen meat and fish and frozen vegetables, the two counting methods had correlation coefficients of 0.87 and 0.66, respectively. The poor correlation for frozen vegetables was due to the inclusion in the DEFT count of nonviable bacteria killed by the blanching process used to inactivate enzymes. Good agreement was obtained between the prefiltered DEFT count and unfiltered plate count for cooked meats, cream doughnut, and whole peppers. Possible reasons for the poor agreement between the DEFT count and plate count for certain products are discussed.     

Examination of Market Foods for Coliform Organisms

Food specimens (490) in nine categories were examined for total aerobic plate count and numbers and types of coliform organisms, including the enteropathogenic Escherichia coli (EEC). The total counts were compared with various suggested standards, and a limit of 100,000/g appeared to be a realistic goal, except for certain food types with a high level of natural flora. Plate counts in VRB were compared to counts obtained by isolation by enrichment in LST Broth, and the latter method produced a higher percentage of isolations. The presence of E. coli was determined by use of EC Medium incubated at 44.5 +/- 0.1 C. Only 40.4% of the positive EC tubes, however, contained E. coli. It appeared that a limit of 10 coliform organisms per g as a suggested standard could be met with several types of foods. Isolation of EEC was obtained only three times, or in 0.6% of the specimens.     

Microbiological quality of frozen cauliflower, corn, and peas obtained at retail markets

The microbiological quality of blanched frozen cauliflower, cut corn, and peas at the retail level was determined. At 35 degrees C, mean aerobic plate count (APC) values for cauliflower, corn, and peas, respectively, were 30,000, 6,100, and 4,700 per g; at 30 degrees C, the mean APC values were 45,000, 8,500, and 6,800 per g, respectively. Geometric means for coliform, Escherichia coli, and Staphylococcus aureus counts for all three vegetables were less than 10 per g.     

Microbiological quality of frozen cauliflower, corn, and peas obtained at retail markets.

The microbiological quality of blanched frozen cauliflower, cut corn, and peas at the retail level was determined. At 35 degrees C, mean aerobic plate count (APC) values for cauliflower, corn, and peas, respectively, were 30,000, 6,100, and 4,700 per g; at 30 degrees C, the mean APC values were 45,000, 8,500, and 6,800 per g, respectively. Geometric means for coliform, Escherichia coli, and Staphylococcus aureus counts for all three vegetables were less than 10 per g.     

Estimation of microbial contamination of food from prevalence and concentration data: application to Listeria monocytogenes in fresh vegetables

A normal distribution and a mixture model of two normal distributions in a Bayesian approach using prevalence and concentration data were used to establish the distribution of contamination of the food-borne pathogenic bacteria Listeria monocytogenes in unprocessed and minimally processed fresh vegetables. A total of 165 prevalence studies, including 15 studies with concentration data, were taken from the scientific literature and from technical reports and used for statistical analysis. The predicted mean of the normal distribution of the logarithms of viable L. monocytogenes per gram of fresh vegetables was -2.63 log viable L. monocytogenes organisms/g, and its standard deviation was 1.48 log viable L. monocytogenes organisms/g. These values were determined by considering one contaminated sample in prevalence studies in which samples are in fact negative. This deliberate overestimation is necessary to complete calculations. With the mixture model, the predicted mean of the distribution of the logarithm of viable L. monocytogenes per gram of fresh vegetables was -3.38 log viable L. monocytogenes organisms/g and its standard deviation was 1.46 log viable L. monocytogenes organisms/g. The probabilities of fresh unprocessed and minimally processed vegetables being contaminated with concentrations higher than 1, 2, and 3 log viable L. monocytogenes organisms/g were 1.44, 0.63, and 0.17%, respectively. Introducing a sensitivity rate of 80 or 95% in the mixture model had a small effect on the estimation of the contamination. In contrast, introducing a low sensitivity rate (40%) resulted in marked differences, especially for high percentiles. There was a significantly lower estimation of contamination in the papers and reports of 2000 to 2005 than in those of 1988 to 1999 and a lower estimation of contamination of leafy salads than that of sprouts and other vegetables. The interest of the mixture model for the estimation of microbial contamination is discussed.     

Bacteriology of Dehydrated Space Foods

The initial bacteriological requirement established in 1964 for space foods by the U.S. Army Natick Laboratories are: a total aerobic plate count ( 300,000), chocolate ice cream cubes (20,000), and each of four samples of chocolate candy (12,000 to 61,000); (ii) coliforms: two out of three vanilla milk drinks (16 and 127) and one beef hash bar (14); (iii) fecal coliforms: one sample of chicken soup and gravy base positive; (iv) fecal streptococci: two samples of peanut cubes (40 and 108), coconut cubes (75), chicken soup and gravy base (2,650), beef soup and gravy base (33), and five out of six flavored milk drinks (23 to 300); (v) salmonellae: one each of chicken and beef soup and gravy base were positive.     

Estimation of Microbial Contamination of Food from Prevalence and Concentration Data: Application to Listeria monocytogenes in Fresh Vegetables

A normal distribution and a mixture model of two normal distributions in a Bayesian approach using prevalence and concentration data were used to establish the distribution of contamination of the food-borne pathogenic bacteria Listeria monocytogenes in unprocessed and minimally processed fresh vegetables. A total of 165 prevalence studies, including 15 studies with concentration data, were taken from the scientific literature and from technical reports and used for statistical analysis. The predicted mean of the normal distribution of the logarithms of viable L. monocytogenes per gram of fresh vegetables was −2.63 log viable L. monocytogenes organisms/g, and its standard deviation was 1.48 log viable L. monocytogenes organisms/g. These values were determined by considering one contaminated sample in prevalence studies in which samples are in fact negative. This deliberate overestimation is necessary to complete calculations. With the mixture model, the predicted mean of the distribution of the logarithm of viable L. monocytogenes per gram of fresh vegetables was −3.38 log viable L. monocytogenes organisms/g and its standard deviation was 1.46 log viable L. monocytogenes organisms/g. The probabilities of fresh unprocessed and minimally processed vegetables being contaminated with concentrations higher than 1, 2, and 3 log viable L. monocytogenes organisms/g were 1.44, 0.63, and 0.17%, respectively. Introducing a sensitivity rate of 80 or 95% in the mixture model had a small effect on the estimation of the contamination. In contrast, introducing a low sensitivity rate (40%) resulted in marked differences, especially for high percentiles. There was a significantly lower estimation of contamination in the papers and reports of 2000 to 2005 than in those of 1988 to 1999 and a lower estimation of contamination of leafy salads than that of sprouts and other vegetables. The interest of the mixture model for the estimation of microbial contamination is discussed.     

Evaluation of the MicroFoss system for enumeration of total viable count, Escherichia coli and coliforms in ground beef

This study was conducted to evaluate the performance of the MicroFoss system (Biosys, Ann Arbor, MI) for enumeration of total viable organisms, Escherichia coli and coliforms in ground beef. The system performance was compared to that of the USDA Bacteriological Analytical Method (BAM) reference culture methods. The correlation coefficients for the regression lines comparing the MicroFoss system detection times to the results of plate count methods for the total viable counts, coliform counts and the most probable number (MPN) method for E. coli were -0.95, -0.96 and -0.97, respectively. Tests comparing the reproducibility of data generated independently by two technicians on the same batch of samples showed no significant differences (P>0.05) in the MicroFoss detection times and culture results. The plate count methods for the total viable counts and coliform counts, and the MPN method for E. coli required 10, 11 and 22 times, respectively, the amount of time to complete tests compared to the length of time required to perform these tests using the MicroFoss system. The MicroFoss system produced reproducible data and provided a rapid and cost-efficient alternative method for enumeration of TVC, coliforms and E. coli in ground beef.     

Microbiological quality of frozen shrimp and lobster tail in the retail market.

The microbiological quality of three frozen shrimp products and frozen lobster tail at the retail level was determined. The number of retail units of the four products examined and the geometric means for aerobic plate counts at 30 and 35 degrees C, respectively, were: 1,464 units of cooked, peeled shrimp--13,000 and 7,200 per g; 1,468 units of raw, peeled shrimp--860,000 and 300,000 per g; 1,300 units of raw, in-shell shrimp--800,000 and 300,000 per g; 1,315 units of lobster tail--140,000 and 42,000 per g. Geometric means for coliform, Escherichia coli, and Staphylococcus aureus counts for all products were < 10 per g.     

A microassay-based procedure for measuring low levels of toxic organophosphorus compounds through acetylcholinesterase inhibition

Using a microtiter plate spectrophotometric system, an assay procedure was developed for the following toxic organophosphorus compounds: 1,2,2-trimethylpropyl ester of methylphosphonofluoridic acid (1, soman); ethyl N,N-dimethylphosphoramidocyanidate (3, tabun); O-ethyl S-[2-[bis(1-methylethyl)amino]ethyl]- methylphosphonothiolate (4, VX); the diethyl 4-nitrophenyl ester of phosphoric acid (5, paraoxon); and bis(1-methylethyl) phosphorofluoridate (6, DFP). The procedure, based on the Ellman assay method, uses inhibition of eel acetylcholinesterase (0.01 unit per well) to carry out the determination of inhibitor concentrations for both a standard curve and the unknown samples on a single 96-well microtiter plate. On a typical plate, samples of both unknowns and standards (a minimum of six concentrations were used per standard curve) were assayed five times per sample, with three control (uninhibited) enzyme activity points included for each sample. The time required for carrying out a single plate was approx 30 min. Sensitivity for the most potent acetylcholinesterase inhibitor tested was 0.4 nM under the conditions used for a typical assay. It should be noted, however, that no attempt was made to optimize the assay procedure for sensitivity.     

Culture medium for selective isolation and enumeration of Gram-negative bacteria from ground meats.

We developed a new medium, designated peptone bile amphotericin cycloheximide (PBAC) agar, which contains (per liter) 10 g of peptone, 300 mg of bile salts, 1 mg of amphotericin B, 1 g of cycloheximide, and 15 g of agar. When 21 samples of fresh ground beef were studied and plate count agar counts were used as references, we obtained a mean recovery of 28% of total counts with violet red bile agar overlay, whereas we obtained 48% recovery with PBAC agar. With 12 samples of frozen ground beef, recovery on violet red bile agar overlay was 29% of the recovery on plate count agar, whereas the corresponding value on PBAC agar was 45%. PBAC agar allowed the enumeration of 1.4 times as many gram-negative bacteria as violet red bile agar overlay. None of eight strains of gram-positive bacteria and none of eight strains of yeasts grew on PBAC agar. Of 158 colonies randomly selected from pour plates of eight fresh ground meat samples, 95% stained gram negative. In comparison, only 70% of 151 colonies selected from corresponding plate count agar plates were gram negative. The lack of background color, turbidity, and ease of use make PBAC agar easier to handle than other media used for gram-negative bacteria, such as violet red bile agar, violet red bile agar overlay, and crystal violet tetrazolium agar. In the preparation PBAC agar, all ingredients are autoclaved together except amphotericin B, which is filter sterilized and added before the plates are poured.     

Freeze-dried mixed cultures as reference samples in quantitative and qualitative microbiological examinations of food

Two simulated food samples in the form of mixtures of bacteria, freeze-dried in small glass vials, were tested for their suitability as reference materials in food microbiology. Stability and homogeneity of the samples were assessed during a 1 year period when stored at 2–6°C. Samples were examined for aerobic plate counts, coliform counts, Staphylococcus aureus, Clostridium perfringens, Bacillus cereus, enterococci, yeasts and salmonellas. The samples were homogeneous but the number of colony forming units of some organisms declined during the time period studied. The maximum rate of decline (0·8 log units per year) was observed for the coliforms. Precision estimates (reproducibility and repeatability) for the different testing methods used are given. When compared with these data, the rate of decline was considered to be of less importance.     

Microbiological quality of some spices and herbs in retail markets.

The microbiological quality of 10 spices or herbs was determined by a national survey at the retail level. Aerobic plate count values for the 10 products ranged from less than 100 to 3.1 X 10(8) per g; mean values of the individual spices or herbs ranged from 1,400 to 820,000 per g. Coliform counts ranged from less than 3 to 1.1 X 10(6) per g; however, mean values were less than 20 per g for all products. Escherichia coli counts ranged from less than 3 to 2,300 per g. Except for celery seed, which had a mean value of 7 per g, all mean values were less than 3 per g. Yeast and mold counts were made for 5 of the 10 products. Mean values were generally low; the highest mean (290 per g) was obtained for cinnamon.     

Microbiological quality and safety of zoo food.

Two types of commercial products for feeding zoo animals (a frozen meat product, referred to as zoo food, and a dry product, referred to as dry food) were microbiologically examined for spoilage organisms (aerobic, psychrotrophic, coliform, Escherichia coli, mold, and yeasts) and pathogens (Salmonella spp., Listeria monocytogenes, and Campylobacter jejuni). Levels of microorganisms in frozen ground zoo food were compared with those in frozen ground beef and frozen ground turkey meat. The level of microbial contaminants in frozen ground zoo meat was found to be similar to that in frozen ground beef and higher than that in frozen ground turkey meat. Sixty percent of the frozen zoo meat samples were Salmonella positive, and all of the samples were L. monocytogenes positive. Dry zoo food was documented to have microbial levels lower than those in frozen zoo meat; the pathogen levels were less than 1/25 g of food. Defrosting zoo meat at 10, 25, and 37 degrees C for 24 h showed that 10 degrees C is the best temperature for defrosting frozen ground zoo meat loaves (length, 9 in. [22.8 cm]; radius, 2 in. [5.1 cm]) without affecting the microbiological quality or safety of the product.     

Microbiological quality and safety of zoo food

Two types of commercial products for feeding zoo animals (a frozen meat product, referred to as zoo food, and a dry product, referred to as dry food) were microbiologically examined for spoilage organisms (aerobic, psychrotrophic, coliform, Escherichia coli, mold, and yeasts) and pathogens (Salmonella spp., Listeria monocytogenes, and Campylobacter jejuni). Levels of microorganisms in frozen ground zoo food were compared with those in frozen ground beef and frozen ground turkey meat. The level of microbial contaminants in frozen ground zoo meat was found to be similar to that in frozen ground beef and higher than that in frozen ground turkey meat. Sixty percent of the frozen zoo meat samples were Salmonella positive, and all of the samples were L. monocytogenes positive. Dry zoo food was documented to have microbial levels lower than those in frozen zoo meat; the pathogen levels were less than 1/25 g of food. Defrosting zoo meat at 10, 25, and 37 degrees C for 24 h showed that 10 degrees C is the best temperature for defrosting frozen ground zoo meat loaves (length, 9 in. [22.8 cm]; radius, 2 in. [5.1 cm]) without affecting the microbiological quality or safety of the product.     

Assessment of the bacteriological activity associated with granular activated carbon treatment of drinking water.

Bacteriological analyses were performed on the effluent from a conventional water treatment pilot plant in which granular activated carbon (GAC) had been used as the final process to assess the impact of GAC on the microbial quality of the water produced. Samples were collected twice weekly for 160 days from the effluents of six GAC columns, each of which used one of four different empty-bed contact times (7.5, 15, 30, and 60 min). The samples were analyzed for heterotrophic plate counts and total coliforms. Effluent samples were also exposed to chloramines and free chlorine for 60 min (pH 8.2, 23 degrees C). Bacterial identifications were performed on the disinfected and nondisinfected effluents. Additional studies were conducted to assess the bacteriological activity associated with released GAC particles. The results indicated that heterotrophic plate counts in the effluents from all columns increased to 10(5) CFU/ml within 5 days and subsequently stabilized at 10(4) CFU/ml. The heterotrophic plate counts did not differ at different empty-bed contact times. Coliforms (identified as Enterobacter spp.) were recovered from the nondisinfected effluent on only two occasions. The disinfection results indicated that 1.5 mg of chloramines per liter inactivated approximately 50% more bacteria than did 1.0 mg of free chlorine per liter after 1 h of contact time. Chloramines and chlorine selected for the development of different bacterial species--Pseudomonas spp. and Flavobacterium spp., respectively.(ABSTRACT TRUNCATED AT 250 WORDS)