Validation and interlaboratory comparison of anticoagulant rodenticide analysis in animal livers using ultra-performance liquid chromatography-mass spectrometry.

Anticoagulant rodenticides (ARs) are used to control rodent populations. Poisoning of non-target species can occur by accidental consumption of commercial formulations used for rodent control. A robust method for determining ARs in animal tissues is important for animal postmortem diagnostic and forensic purposes. We evaluated an ultra-performance liquid chromatography coupled with mass spectrometry (UPLC-MS) method to quantify 8 ARs (brodifacoum, bromadiolone, chlorophacinone, coumachlor, dicoumarol, difethialone, diphacinone, warfarin) in a wide range of animal (bovine, canine, chicken, equine, porcine) liver samples, including incurred samples. We further evaluated UPLC-MS in 2 interlaboratory comparison (ILC) studies; one an ILC exercise (ICE), the other a proficiency test (PT). The limits of detection of UPLC-MS were 0.3-3.1 ng/g, and the limits of quantification were 0.8-9.4 ng/g. The recoveries obtained using UPLC-MS were 90-115%, and relative SDs were 1.2-13% for each of the 8 ARs for the 50, 500, and 2,000 ng/g spiked liver samples. The overall accuracy from the laboratories participating in the 2 ILC studies (4 and 11 laboratories for ICE and PT studies, respectively) were 86-118%, with relative repeatability SDs of 3.7-11%, relative reproducibility SDs of 7.8-31.2%, and Horwitz ratio values of 0.5-1.5. Via the ILC studies, we verified the accuracy of UPLC-MS for AR analysis in liver matrices and demonstrated that ILC can be utilized to evaluate performance characteristics of analytical methods.

Determination of Multiresidue Pesticides in Botanical Dietary Supplements Using Gas Chromatography-Triple-Quadrupole Mass Spectrometry (GC-MS/MS).

A simplified sample preparation method in combination with gas chromatography-triple-quadrupole mass spectrometry (GC-MS/MS) analysis was developed and validated for the simultaneous determination of 227 pesticides in green tea, ginseng, gingko leaves, saw palmetto, spearmint, and black pepper samples. The botanical samples were hydrated with water and extracted with acetonitrile, magnesium sulfate, and sodium chloride. The acetonitrile extract was cleaned up using solid phase extraction with carbon-coated alumina/primary-secondary amine with or without C18. Recovery studies using matrix blanks fortified with pesticides at concentrations of 10, 25, 100, and 500 µg/kg resulted in average recoveries of 70-99% and relative standard deviation of 5-13% for all tested botanicals except for black pepper, for which lower recoveries of fortified pesticides were observed. Matrix-matched standard calibration curves revealed good linearity (r(2) > 0.99) across a wide concentration range (1-1000 µg/L). Nine commercially available tea and 23 ginseng samples were analyzed using this method. Results revealed 36 pesticides were detected in the 9 tea samples at concentrations of 2-3500 µg/kg and 61 pesticides were detected in the 23 ginseng samples at concentrations of 1-12500 µg/kg.

Development and validation of a high performance liquid chromatographic method for simultaneous determination of vitamins A and D3 in fluid milk products.

An HPLC method for simultaneous determination of vitamins A and D3 in fluid milk was developed and validated. Saponification and extraction conditions were studied for optimum recovery and simplicity. An RP HPLC system equipped with a C18 column and diode array detector was used for quantitation. The method was subjected to a single-laboratory validation using skim, 2% fat, and whole milk samples at concentrations of 50, 100, and 200% of the recommended fortification levels for vitamins A and D3 for Grade "A" fluid milk. The method quantitation limits for vitamins A and D3 were 0.0072 and 0.0026 µg/mL, respectively. Average recoveries between 94 and 110% and SD values ranging from 2.7 to 6.9% were obtained for both vitamins A and D3. The accuracy of the method was evaluated using a National Institute of Standards and Technology standard reference material (1849a) and proficiency test samples.

Method validation and proficiency testing for determination of total arsenic in apple juice by inductively coupled plasma/ mass spectrometry.

A proficiency test (PT) program for determination of total As in apple juice samples was conducted by the Food Emergency Response Network (FERN) laboratories. An analytical method using inductively coupled plasma (ICP)-MS was validated for this project. The LOD and LOQ were determined to be 0.315 and 2.32 ng/g, respectively. A total of eight apple juice samples were sent to 38 FERN laboratories, and results were statistically evaluated according to ISO 13528:2005. The total As concentrations in the PT samples reported by the participating laboratories were very close to those obtained in the homogeneity and stability tests. The reproducibility, repeatability, interlaboratory, and intralaboratory variability results led to 69% of participating laboratories being rated as satisfactory using the widely accepted Izl score

Detection and enumeration of four foodborne pathogens in raw commingled silo milk in the United States.

A nationwide survey was conducted to obtain qualitative and quantitative data on bacterial contamination of raw commingled silo milk intended for pasteurization. The levels of total aerobic bacteria, total coliforms, Enterobacteriaceae, Escherichia coli, and Staphylococcus aureus were determined using the TEMPO system. The prevalence rates and levels of presumptive Bacillus cereus, E. coli O157:H7, Listeria monocytogenes, and Salmonella spp. were determined in 214 samples. B. cereus was detected in 8.91% of samples, at 3.0 to 93 CFU/ml. E. coli O157:H7 was detected in 3.79 to 9.05% of samples, at <0.0055 to 1.1 CFU/ml, depending on the assay utilized. Salmonella spp. were recovered from 21.96 to 57.94% of samples, at <0.0055 to 60 CFU/ml. L. monocytogenes was detected in 50.00% of samples, at <0.0055 to 30 CFU/ml. The average log-transformed counts of total viable bacteria were slightly lower in samples containing no pathogens. No correlation was observed between the levels of organisms detected with the TEMPO system and the presence or levels of any pathogen except E. coli O157:H7. A higher average log-transformed count of total viable bacteria was observed in samples positive for this organism. The high prevalence rates of target pathogens may be attributed to a variety of factors, including detection methods, sample size, and commingling of the milk in the silo. The effects of commingling likely contributed to the high prevalence rates and low levels of target pathogens because of the inclusion of milk from multiple bulk tanks. The high prevalence rates also may be the result of analysis of larger sample volumes using more sensitive detection methods. These quantitative data could be utilized to perform more accurate risk assessments and to better estimate the appropriate level of protection for dairy products and processing technologies.