Food Protection Trends - April 2011 - (Page 213)

the 99th percentile of cattle treated with 0.3 mg per kg of body weight will have tissue residues below these MRLs (15). Veterinary drug residue monitoring, which is statistically based, unbiased, and scheduled, involves random sampling, analysis, and reporting of results. It is used to provide information, usually annually, about the occurrence and concentrations of chemical residues in specific populations of food animals or food items. Monitoring data may be helpful for estimating chemical residue trends and identifying high risk circumstances for chemical non-conformities (i.e., chemical residues above an established tolerance or MRL), and to protect consumer health and international trade (6, 7, 17, 29, 30). Residue monitoring programs should ensure, with 95% confidence, that a residue problem over a maximum acceptable prevalence (commonly 1%) will be detected (6, 29). In Mexico, the Toxic Residues Program is based on several standards and recommendations, such as those published by Codex Alimentarius, the Food and Agriculture Organization, World Health Organization, World Organization for Animal Health, Compound Evaluation System of the United States Department of Agriculture (USDA) and the European Directive 96/23/CE (22, 23, 24). As part of the Toxic Residues Program, liver samples from cattle in Mexico were collected between 2006 and 2008 (262, 289 and 110 animals in 2006, 2007 and 2008, respectively) and analyzed for ivermectin residues. None of the samples contained residues above the established MRL (20, 22, 23, 24). The objective of this study was to determine the prevalence of non-conforming ivermectin residues in cattle slaughtered in federally inspected abattoirs in Nuevo Leon, a northern Mexican state with a reputation for high quality beef and the potential for expanded production (10, 25, 27). MATERIALS AND METHODS a binomial statistical model was used for the calculation of the desired sample size (1–3, 9, 11). The following formula was used: n = log α /[ log (1−p)] where p = 0.10 and α = 10%. The number of samples collected from each abattoir was in proportion to its annual slaughter capacity (8). Samples were collected using a circular systematic random sampling plan, based on the number of slaughtering hours per day and week for each abattoir (14). Each abattoir was provided with a sampling schedule that specified the date and time for sample collection. At the designated times, veterinary inspectors assigned to the abattoir by the Agriculture Department of Mexico (SAGARPA) identified an animal on the line for sampling and collected liver samples (at least 250 g; location not specified) that were placed in sterile plastic bags, individually labelled and frozen. Within one week of collection, the frozen samples were shipped to a central laboratory in a cooler with ice packs or dry ice, accompanied by information regarding the sex, age, and origin of the sampled animals. Samples were analyzed at the “Laboratorio Central Regional de Monterrey” located in Guadalupe, Nuevo Leon, which is accredited by the Mexican Accreditation Entity, based on the quality standard ISO/IEC 17025:2005. Analysis for ivermectin residues was performed by High Performance Liquid Chromatography (HPLC) as specified in the Mexican standard NOM-020-ZOO-1995 (21). Samples that produced positive results were re-tested for confirmation, by use of the same method. All data were recorded in an electronic database (Microsoft Excel 2007). The prevalence estimates and the binomial exact confidence intervals were calculated by a simple binomial method in Stata 10 MP (Stata Corporation, College Station, TX) and R 2.9.2 software (Statistics Department of the University of Auckland, New Zealand). Descriptive statistics and graphics were calculated and developed using Epi Info™ v.3.5.1 (Centers for Disease Control and Prevention, Atlanta Georgia). RESULTS December, 2009. Most of the cattle originated from Nuevo Leon (n = 178; 76%), but others were from Tabasco, Coahuila, Zacatecas, Michoacan, Aguascalientes, Durango, San Luis Potosí, Tamaulipas and Yucatan. The mean age of the animals was 26.8 months with a standard deviation of 9.5 months. Fifty-four percent of the animals were female and 46% were castrated males. Most samples (n = 227; 97%) contained no detectable levels of ivermectin residue. There were four different levels (concentrations) among the samples that contained at least some residue (Table 1). Only one sample contained a nonconforming level (149 µg/kg). Samples containing at least some ivermectin residue (i.e., both conforming and non-conforming) originated from seven batches of cattle slaughtered within three of the six abattoirs. The average size of these batches was 30 animals (range 4–90); they originated from Nuevo Leon (3), Tabasco (2) and Tamaulipas (1); the origin of one of the positive samples was unknown. Four of the seven animals with non-zero concentrations were male; the average age of affected animals was 43 months (range 23–80 months). The non-conforming sample came from a 72 month-old male from Nuevo Leon that was from a batch of 4 animals. DISCUSSION The source population was cattle slaughtered in all six federally inspected abattoirs located within Nuevo Leon between mid-November and midDecember, 2009. Cattle of all ages, breeds and origins (including other Mexican states) were included in the study. A total of 230 liver samples was taken. Because of economic constraints, a 90% level of confidence was selected and In total, 234 cattle liver samples were randomly collected during the study period of mid-November to mid- The observed prevalence of nonconforming (>100 µg/kg) ivermectin residues in cattle liver in this study was small (0.43%) and the vast majority of samples (97%) contained no detectable residue. The 95% confidence intervals of the prevalence of non-conforming residue obtained during this study was consistent with the confidence intervals for ivermectin residue from data gathered during the 2006–2008 residue monitoring program in Mexico (95% exact binomial confidence interval: 0, 0.006). We focused on this region of Mexico (Nuevo Leon) and tested animals at this time of year (mid-November to mid-December) because of suspicions that the frequency of ivermectin use was high. However, we did not attempt to measure the extent of ivermectin use in the study population. The consistency of our findings with those from the general slaughter population of Mexico (23–25) suggests that if the frequency of use was indeed high in this region, appropriate drug withdrawal APRIL 2011 | FOOD PROTECTION TRENDS 213

Table of Contents for the Digital Edition of Food Protection Trends - April 2011

Food Protection Trends - April 2011
Contents
Sustaining Members
Reflections of Your President
Commentary from the Executive Director
Prevalence of Ivermectin Residues in Cattle Slaughtered in Federally Inspected Abattoirs in Nuevo Leon, Mexico
FDA’s Domestic and Imported Cheese Compliance Program Results: January 1, 2004 —December 31, 2006
New Members
What’s Happening in Food Safety
Industry Products
Special Contributors and Sponsors
Preliminary Program
Ivan Parkin Lecture
John H. Silliker Lecture
Activities
General Information
Registration Rates
Coming Events
Advertising Index
Journal of Food Protection Table of Contents
Booklet Order Form
Membership Application
Celebrating 100 Years of IAFP: 1960s

Food Protection Trends - April 2011