Validation of a liquid chromatography–electrospray ionization tandem mass spectrometric method to determine six polyether ionophores in raw, UHT, pasteurized and powdered milk
Mararlene Ulberg Pereira ⇑, Bernardete Ferraz Spisso, Silvana do Couto Jacob, Mychelle Alves Monteiro, Rosana Gomes Ferreira, Betânia de Souza Carlos 1, Armi Wanderley da Nóbrega
A B S T R A C T
This study aimed to validate a method developed for the determination of six antibiotics from the poly- ether ionophore class (lasalocid, maduramicin, monensin, narasin, salinomycin and semduramicin) at residue levels in raw, UHT, pasteurized and powdered milk using QuEChERS extraction and high perfor- mance liquid chromatography coupled to tandem mass spectrometry (HPLC–MS/MS). The validation was conducted under an in-house laboratory protocol that is primarily based on 2002/657/EC Decision, buttakes in account the variability of matrix sources. Overall recoveries between 93% and 113% with relative standard deviations up to 16% were obtained under intermediate precision conditions. CCa calculated values did not exceed 20% the Maximum Residue Limit for monensin and 25% the Maximum Levels for all other substances. The method showed to be simple, fast and suitable for verifying the compliance of raw and processed milk samples regarding the limits recommended by Codex Alimentarius and those adopted in European Community for polyether ionophores.
Keywords:
Coccidiostats Anticoccidials Ionophores
Veterinary drug residues QuEChERS
LC–MS/MS
Milk
1. Introduction
Polyether ionophore antibiotics are widely used in animal feed- ing to prevent and treat coccidiosis, a disease caused by protozoans parasites of the Eimeria genus (mainly Eimeria bovis and Eimeria suernii) residing in the intestinal mucosa. Besides being anticoccidial agents, they act as growth promoters in cattle and pigs, improving feed efficiency and rate of weight gain (The Merck Veterinary Manual, 2012). In ruminants are still employed to increase milk pro- duction in lactating cows (Lindsay & Blagburn, 1995). Six polyether ionophores are approved in Brazil (lasalocid, maduramicin, monensin, narasin, salinomycin and semduramicin) but only monensin and lasalocid are authorized for cattle and lac- tating cows. Salinomycin is permitted for beef cattle to increase weight gain (Brasil, 2008).
These antibiotics influence the contractility of muscle tissue, producing acute pharmacological effects on the cardiovascular sys- tem, with increased coronary blood flow and coronary dilation. Victims of coronary artery disease may be more susceptible to adverse effects (Elliott, Kennedy, & McCaughey, 1998). To protect human health different regulatory authorities have established Maximum Residue Limits (MRLs) or Maximum Levels (MLs) for some polyether ionophores (124/2009/EC, 2009; 37/2010/EU, 2010; Friedlander & Sanders, 2009). In 2008 the Joint FAO WHO Expert Committee on Food Additives and Contaminants (JECFA) evaluated some polyether ionophores in respect of their toxicity and safety and the Codex Alimentarius Commission recommended a Maximum Residue Limit of 2.0 lg kg—1 for monensin in milk, later adopted by European Commission (37/2010/EU, 2010; Friedlander & Sanders, 2009). In 2009, the European Commission, by the opinion of the European Food Safety Authority (EFSA) estab- lished MLs (tolerances) for the presence of some polyether iono- phore residues in foodstuffs, including milk. MLs of 1 lg kg—1 were adopted for lasalocid and narasin and 2 lg kg—1 for salinomycin, maduramicin and semduramicin (124/2009/EC, 2009).
Several analytical methods based on liquid chromatography– electrospray ionization tandem mass spectrometry (LC–MS/MS) have been used to determine one or more residues of polyether ionophores in different matrices, such as eggs, tissues and feed (Dubois, Pierret, & Delahaut, 2004; Matabudul, Conway, Lumley, & Sumar, 2001; Mortier, Daeseleire, & Peteghem, 2005; Rokka & Peltonen, 2006; Shao et al., 2009; Spisso et al., 2010; Tkáciková, Kozˇárová, & Máté, 2010; Vincent, Chedin, Yasar, & von Holst, 2008). However, few surveys have investigated, using LC–MS/MS, residues of polyether ionophores in milk, compared to other matri- ces. So far, only ten manuscripts discuss about LC–MS/MS methods for milk (Bo, Luo, Cao, Xing, & Pang, 2009; Clarke, Moloney, O’Mahony, O’Kennedy, & Danaher, 2013; Dai & Herrman, 2010; Kim, Bahn, Kang, & Kim, 2012; Nebot, Iglesias, Regal, Miranda, Cepeda et al., 2012; Nebot, Iglesias, Regal, Miranda, Fente et al., 2012; Nász, Debreczeni, Rikker, & Eke, 2012; Thompson, Noot, & Kendall, 2011; Zhan et al., 2012; Cordroc’h apud Friedlander & Sanders, 2009).
Among these ten studies, only two authors mentioned the use of commercial processed milk during validation. Nász et al. (2012) described the use of processed milk with different fat con- tents and Bo et al. (2009) cited the use of powdered milk. However, both methods require a purification step in sample preparation using solid phase extraction (SPE) technique. Considering the need to assess the presence of polyether ionophore residues in milk in Brazil and the lack of standardized methods, this paper presents a study that aimed to determine six antibiotics from the polyether ionophore class (lasalocid, maduramicin, monensin, narasin, salinomycin and semduramicin) in raw and processed milk (UHT, pasteurized and powder) employing QuEChERS approach (Quick, Easy, Cheap, Effective, Rugged, Safe) and high performance liquid chromatography coupled to tandem mass spectrometry (HPLC–MS/MS).
2. Materials and methods
2.1. Chemicals
Methanol (MeOH) and acetonitrile (ACN) of high-performance liquid chromatography (HPLC) grade were purchased from J.T. Baker (Phillipsburg, NJ, USA) and Merck (Darmstadt, Germany), respectively. The chemicals sodium acetate (NaOAc) and formic acid (FOA) were Merck Suprapur® reagents. Anhydrous sodium acetate (NaOAc) and anhydrous magnesium sulfate for analysis employed in QuEChERS were also obtained from Merck. Ultra pure water was provided by a Milli-Q system (Millipore, Bedford, MA, USA). Standards of narasin A (NAR), salinomycin A (SAL) and the nigericin sodium salt (NIG) were purchased from Sigma–Aldrich (St. Louis, MO, USA). Lasalocid A sodium salt (LAS), maduramicin alpha ammonium salt (MAD) and monensin A sodium salt hydrate (MON) were obtained from Dr. Ehrenstorfer (Augsburg, Germany). Sodium semduramicin (SEM) from Phibro Animal Health was kindly donated by the Community Reference Laboratory (CRL) Bundesamt fur Verbraucherschutz und Lebensmittelsicherheit (Berlin, Germany).
2.2. Preparation of standard solutions
Stock standard solutions of 1 mg mL—1 were prepared by dis- solving each standard in methanol and stored at 70 °C, except for LAS that was prepared at 10 lg mL—1 given that the standard is provided as an acetonitrile solution at 0.1 mg mL—1. These solutions were stored in cryotubes at or below 70 °C for at least 3 years for SEM and 2 years for MON, SAL, MAD, NAR, NIG and LAS. Intermediate and working standard solutions were freshly pre- pared at several concentrations by appropriate dilution of stock standard solutions in methanol.
2.3. Milk samples
For method validation, samples of raw and processed whole milk (powder, UHT and pasteurized) were employed. These samples were previously tested to confirm they were free from the target analytes.
2.4. Preparation of samples and matrix-matched calibration standards
Samples of UHT processed whole milk, tested and proved to be blank samples, were spiked with the target analytes at 0.5– 7 lg kg—1 concentration levels for MON, MAD, SAL, SEM and 0.25–3.5 lg kg—1 for LAS and NAR. NIG was used as internal standard at 15 lg kg—1 concentration. A blank sample, besides one con- taining only the internal standard NIG, was also prepared. For spiking, 2.0 mL of milk were pipetted into a 50 mL polypropylene centrifuge tube, added of 50 lL NIG solution at 0.6 lg mL—1, and 50 lL polyether ionophores standard solution in each concentra-Acetonitrile was added to the samples in two portions of 4.0 mL, stirring in vortex for 1 min after each addition. After, 0.8 g of anhy- drous magnesium sulfate and 0.2 g of anhydrous sodium acetate were added and stirred in vortex for 1 min. After 5 min resting, samples were centrifuged at 10,000 rpm (12,857 rcf) for 5 min at 4 °C. An aliquot of 250 lL of the supernatant was transferred to a 15 mL polypropylene centrifuge tube and evaporated to dryness under a gentle nitrogen stream 46 ± 1 °C. The dried extracts were redissolved with 1 mL of dilution solvent (5 mmol L—1 NaOAc: MeOH, 70:30, v/v), vortexed for 15 s and filtered through a 0.22 lm PVDF filter into amber vials prior to LC–MS/MS analysis.
2.5. Validation of the analytical method
Method validation was performed under a laboratory protocol primarily based on 2002/657/EC Decision. The evaluated validation parameters were linearity, absolute matrix effect, selectivity (including relative matrix effect), limit of detection, limit of quan- tification, decision limit (CCa), detection capability (CCb), trueness (bias), precision (repeatability and intermediate precision) and ruggedness.
Linearity was assessed for each analyte by statistical analysis of calibration curves obtained by spiking matrix at the beginning of the procedure and the concentrations of the target analytes were equivalent to 0.25, 0.5, 1, 1.5, 2, 2.5, 3 and 3.5-fold MRL/ML. Injections were performed in triplicate randomly.
Method absolute matrix effect, regarding suppression or enhancement of ionization signal, considering the LC–MS/MS tech- nique, was studied by comparing the calibration curve for each analyte in standard solution (solvent curve) to that obtained in the matrix with addition of the analyte after the extraction steps, that means, spiked milk samples at the end of the procedure (final curve in matrix).
Selectivity and relative matrix effect were tested by using eight blank samples of four whole milk types (raw, powder, pasteurized and UHT), and for each type of milk two different samples were chosen. Calibration curves were constructed for each type of milk by spiking at the end of the procedure, at equivalent concentra- tions of 0.25, 0.5, 1, 1.5, 2, 2.5-fold the MRL/ML, including zero level, with addition of NIG as internal standard. A blank milk sam- ple from each type was also processed.
Repeatability and intermediate precision were studied by using 6 samples (two raw, two UHT, one pasteurized and one powdered milk). Samples were spiked at the beginning of the procedure at concentration levels corresponding to 0.5, 1 and 1.5-fold the MRL/ML. Injections were done in random triplicate. A pool of blank sample was prepared with the six selected samples and also ana- lyzed. To quantify these eighteen samples, calibration curves were done using UHT whole milk as matrix, spiked at the beginning of the procedure, at six concentration levels plus zero level, all with NIG addition as internal standard. Concentrations were equivalent to 0.5, 1, 1.5, 2, 2.5 and 3-fold the MRL/ML (1, 2, 3, 4, 5 and Ruggedness was evaluated by central composite design at two levels, with triplicates in central point, 23. Time for analyte–matrix interaction, extraction time with vortex and evaporation tempera- ture in the nitrogen evaporator, which could potentially affect results, were evaluated. With this approach the effects of these fac- tors as well as the effects of their interactions could be detected. These factors were slightly changed and eleven tests were per- formed to evaluate the three factors. Each factor was set at two dif- ferent levels: the interaction time at 5 and 15 min, extraction time at 30 and 90 s and the evaporation temperature at 44 ± 1 °C and 48 ± 1 °C. The central point was set at the midpoints of the three factors: 10 min, 60 s and 46 ± 1 °C and was performed in triplicate. Samples were analyzed in random order. The evaluation was per- formed from peak areas comparison.
Limits of detection (LODs) and limits of quantification (LOQs) were determined using spiked samples at the lower validation level in intermediate precision experiments. LODs and LOQs were calculated considering a signal-to-noise ratio equal to 3 and 10, respectively, for the first confirmation transition (the second most intense transition) by S-to-N using peak-to-peak script available in Analyst® software. The highest values obtained in the 3 intermedi- ate precision experiments were considered. Decision limits (CCa) and detection capabilities (CCb) were cal- culated by overall calibration curve procedure from the same inter- mediate precision data, applying weighted regression analysis.
2.6. LC–MS/MS instrumentation and conditions
LC–MS/MS conditions were almost the same as reported previ- ously (Spisso et al., 2010), but optimization of the parameters injection volume and source temperature was required. A Shimadzu Prominence HPLC instrument (Kyoto, Japan) equipped with a quaternary pump (LC-20AD), a membrane degasser (DGU-20A5), an auto-sampler (SIL-20AC), a column oven (CTO- 20AC) and a controller system (CBM-20A) was used. It was inter- faced to an API5000 triple quadrupole mass spectrometer (Applied Biosystems/MDS Sciex, Foster City, CA, USA) fitted with the Tur- boIonSpray® source. Analyst® V1.4.2 LC/MS control software was used. Sample aliquots (stored at 4 °C in the auto-sampler) of 25 lL were injected on an ACE C18 analytical column (50 mm × 2.1 mm i.d., 3 lm particle size, 100 Å), with a guard column of the same material (Advanced Chromatography Technologies, Aberdeen, Scot- land). Gradient elution was performed with water, acetonitrile and methanol all containing 0.1% formic acid (mobile phases A, B and C, respectively) at 35 °C and at a flow rate of 0.3 mL min—1. The run started at 7% B, followed by a 4-min linear gradient to 80% B, imme- diately changed to 95% B (4.10 min), followed by a linear gradient to 100% B at 6 min. This eluent was maintained up to 8 min, when a 0.5-min linear gradient to 100% C was performed. The column was washed for 3 min in 100% C, the initial condition was reestablished in 0.5 min, while 6 min was required to re-equilibration. The total run time was then 18 min. The MS/MS instrument was tuned by infusion at 10 lL min—1 of 50 ng mL—1 polyether ionophore solu- tions prepared in 0.1% FOA with 2 mmol L—1 NaOAc:ACN (20:80, v/v). Positive electrospray ionization technique (ESI+) in multiple reaction monitoring (MRM) acquisition mode was used to monitor three ions for each substance. Nitrogen was employed as nebulizer and dryer gas (gas 1 and gas 2 = 55 psi), collision activated dissoci- ation (CAD) gas (10, arbitrary unit) and CurtainTM gas (gas 3 = 10 psi).
Other parameters selected during automatic tuning were: ionspray potential = 4500 V, source temperature = 450 °C, entrance poten- tial = 10 V, resolution Q1 and Q3 = unit. The parameters decluster- ing potential (DP), collision energy (CE) and collision exit potential (CXP) were optimized for each MRM transition with dwell times equal to 20 or 25 ms (Table 1).
2.7. Identification and quantification
Polyether ionophores were detected as sodium adducts, [M +Na]+. The first MRM transition as shown in Table 1 was related to the product ion with the greatest intensity (base peak) and used for quantitative purposes. Two qualifying ions were used for con- firmation. Identification was performed according to Commission Decision 2002/657, considering ion ratios, relative retention times and minimum signal-to-noise ratios. The chromatographic peaks were integrated with the IntelliQuan algorithm in the Analyst® software. A signal-to-noise ratio of the peaks equal or greater than 3:1 was required for detection. Validation samples were quantified (using peak areas) in matrix-matched calibration curves fitted by weighted regression analysis using a factor of 1/y. Solvent calibra- tion curves were also performed in order to compare to matrix- matched calibration curves. Concentrations in the samples were calculated directly from pre-extracted spiked calibration curves, so recovery corrections were not applied.
2.8. Data analysis
Besides Analyst®, Statistica® 8.0 software (StatSoft, USA) was used for regression analysis and for planning and analysis of the 2-level factorial design employed to assess the ruggedness of the analytical method.
3. Results and discussion
3.1. Optimization of the extraction and LC–MS/MS method
The full results of extraction method development are described elsewhere (Pereira, 2011). Due to lower MRL values established for polyether ionophores in milk, compared to those set for eggs, the instrument response was not enough when exactly the same LC–MS/MS conditions employed for eggs were used. So, modifications in the parameters injection volume and source temperature were required, in order to improve method sensitivity for the analytes maduramicin and lasalocid. Mass spectrometry optimization was performed by modifying source temperature (600 °C, 550 °C and 450 °C), with triplicate injections of a spiked sample at 0.5 MRL/ML, No significant differences were observed for maduramicin at 600 °C and 550 °C, but an improvement of 44% in peak area was observed from 550 °C to 450 °C. Different injection volumes (15 and 25 lL) were tested using a spiked sample at 0.5 MRL/ML. Responses were pro- portional to the volumes for all analytes, without prejudice to the chromatographic profile, and thus 25 lL was chosen as the injec- tion volume providing the required sensitivity for all analytes, including lasalocid.
3.2. Selectivity, relative matrix effect and absolute matrix effect
Seventeen milk samples from different origins and brands were evaluated and no analytes or matrix components from raw, pow- dered, pasteurized and UHT milk samples were observed, in the elution region of the target substances. Fig. 1 shows MRM chro- matograms of three transitions for MON in a raw blank milk sam- ple and in a raw milk sample spiked at 2.4 lg kg—1 (CCa).
Relative matrix effect was evaluated from calibration curves constructed with different sources of matrix. This approach was used by Matuszewski (2006) to evaluate matrix effect of bioanalyt- ical methods and the use of different matrices for analytical method validation has been recommended by Codex Alimentarius (2011) for multi residue methods. This work followed the same strategy adopted by Matuszewski (2006), considering the sample variability and its possible effect in different concentra- tions, aiming to achieve conditions most similar to those in routine analysis.
The slopes of the calibration curves were evaluated and no sig- nificant differences were observed between them (adopting the criteria of a relative standard deviation value of the slopes lower than 15%), indicating that undetectable endogenous substances present in the different sources of matrices did not influence the detector response. Since the evaluated matrices behaved similarly (the relative standard deviation values of the slopes for the six tar- get analytes ranged from 3% to 8%), a relative matrix effect was not featured. Fig. 2 shows MON calibration curves obtained from eight different matrices. Then, the method proved to be selective with respect to matrix components and applicable to different types of milk.
Evaluation of absolute matrix effects was performed by comparing the slopes of the calibration curves constructed with ana- lytes in solvent and in matrix (spiked at the end of the procedure), using 2-tailed Student t-test. Linearity of both calibration curves was proven using the same statistical tools described in Section 3.3. The comparison of the slopes by the t-test indicated that cali- bration curves were statistically different at a = 0.05, for all ana- lytes, which showed that absolute matrix effects were present. So, analyte quantification must be done in matrix-match calibra- tion curves.
3.3. Linearity
MRM chromatograms obtained for the lower concentration level (0.25 MRL/ML) were evaluated in order to check the signal- to-noise ratio for the first confirmation transition (the second most intense transition) and for the quantification transition (first most intense transition). For MON, NAR, SAL and SEM the lowest studied concentration, 0.25 MRL/ML, can be considered the beginning of the interval linear range because the signal-to-noise ratio was P10 for all peaks in both transitions (Instituto Nacional de Controle de Qualidade em Saúde, 2013). For LAS and MAD, the low- est studied level showed signal to noise ratio 610 for the two tran- sitions. The concentration immediately above that provided this reason, and 0.5 MRL/ML was considered the lower limit of the lin- ear range.
Levene’s test indicated heterogeneity of variances for all ana- lytes (p 6 0,05). Considering data heteroscedasticity, the method of weighted least squares regression (WLSR), was employed for the construction of all calibration curves (weight 1/y, where y is the area). Weighted linear coefficients of determination (R2) for the cali- bration functions constructed with spiked matrix were P0.98. The fit of the mathematical model of WLSR was checked and confirmed comparing the first order linear regression model with the second order polynomial nonlinear regression model. Thus, method linear range was 0.25–3.5 MRL/ML for MON, NAR, SAL and SEM and 0.5–3.5 ML for LAS and MAD. Internal standard correction did not improve calibration func- tions. So, NIG was used just for qualitative purposes, to calculate relative retention times.
3.4. Repeatability, intermediate precision and recovery
LAS showed the highest repeatability relative standard devia- tion (RSDr), 15%, for the level of 1 lg kg—1 and NAR the highest intermediate precision relative standard deviation (RSDRw), 16%, for the level of 0.5 lg kg—1 (Table 2). Recoveries ranged from 94% to 110% under repeatability conditions and from 93% to 113% under intermediate precision conditions (Table 2). For all analytes, for the three studied concentrations, recoveries, RSDr and global RSDRw were acceptable, considering the specifica- tions laid down by European Commission and by Codex Alimenta- rius Commission (Codex Alimentarius Commission, 2009; Codex Alimentarius Commission, 2011; Commission Decision, 2002). Clarke et al. (2013) obtained similar results for recoveries and RSDr for some analytes, but it is important to point out that the method had been validated only for raw milk, unlike the present study which employed six different sources of matrix, including raw and processed milk during the validation study.
3.5. Limits of detection (LODs), limits of quantification (LOQs), decision limits (CCa) and detection capabilities (CCb)
The estimated limits of detection ranged from 0.03 lg kg—1 to tally could lead to lower LOD values than those estimated by extrapolation, it was decided to adopt the highest values as the analytes’ LODs, because they were calculated using the six differ- ent types of milk matrix and thus can be considered most representative. CCa and CCb values are presented in Table 3 and were calculated from the global curves with spiked samples obtained by the weighted least squares method. CCa and CCb values in this study were at most 25% and 50%, respectively, above the MRL/LM.
3.6. Ruggedness (minor changes)
The method showed to be robust in the studied range, for minor changes of the factors analyte/matrix interaction time and vortex extraction time. Only evaporation temperature presented signifi- cant effects (a = 0.05) for the analyte MON (p = 0.0266) and NIG (p = 0.0275) according to analysis of variance (ANOVA). No interac- tion between the three studied factors was observed. The evaporation temperature in nitrogen evaporator is a critical point of the method and should be controlled at 46 °C (with maximum variation of ±1 °C).
3.7. Analysis of real samples
The validated method was applied to samples acquired in the local market. 14 of 102 samples were contaminated with MON (Pereira et al., 2015). Fig. 3 shows a MRM chromatogram of a milk sample containing MON traces (concentration higher than LOD, 0.06 lg kg—1, and lower than LOQ, 0.1 lg kg—1), showing the high sensitivity achieved by the analytical method.
4. Conclusions
Sample preparation using QuEChERS approach was suitable for the extraction of six polyether ionophores in milk. As main advan- tages, the extraction method is fast, easy to perform and present a relatively low consumption of reagents and solvents, which makes it suitable for routine analysis. One of the features of the validated method is its applicability to different types of processed milk (pasteurized, UHT and powder), in addition to raw milk. The method showed to be suitable for the intended use, since validation parameters results fulfilled the criteria advocated in the field of veterinary drug residues in food and is therefore able for assessing the conformity of milk samples regarding the MLs and MRL recommended by Codex Alimentarius and European Community.
References
Bo, H.-B., Luo, L.-L., Cao, Y.-Z., Xing, Y.-X., & Pang, G.-F. (2009). Determination of six polyether antibiotics residues in milk and milk powder by ultra performance liquid chromatography–tandem mass spectrometry. Chinese Journal of Analytical Chemistry, 37(8), 1161–1166.
Brasil, Ministério da Agricultura, Pecuária e Abastecimento (2008). Divisão de Aditivos. Tabela de aditivos antimicrobianos, anticoccidianos e agonistas com uso autorizado na alimentação animal [Internet]. Available from:
Clarke, L., Moloney, M., O’Mahony, J., O’Kennedy, R., & Danaher, M. (2013). Determination of 20 coccidiostats in milk, duck muscle and non-avian muscle tissue using UHPLC–MS/MS. Food Additives & Contaminants: Part A, 30(6), 958–969.
Codex Alimentarius Commission (2009). Guidelines for the design and implementation of national regulatory food safety assurance programme associated with the use of veterinary drugs in food producing animals. CAC/ GL 71-2009. 38p.
Codex Alimentarius Commission (2011). Proposed draft guidelines on performance characteristics for multi-residues methods (Appendix to CAC/GL 71-2009). CX/ RVDF 12/20/10. December 2011. 19p.
Commission Decision 2002/657/EC (mmission Decision 2002/657). Commission Decision of 12 August 2002 implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results. Official Journal of the European Communities, L221, 8–36.
Commission Regulation (EU) 37/2010 (2010). COMMISSION REGULATION (EU) No 37/2010 of 22 December 2009 on pharmacologically active substances and their classification regarding maximum residue limits in foodstuffs of animal origin. Official Journal of the European Union, L15, 1–72.
Commission Regulation 124/2009/EC (2009). COMMISSION REGULATION (EC) No 124/2009 of 10 February 2009 setting maximum levels for the presence of coccidiostats or histomonostats in food resulting from the unavoidable carry- over of these substances in non-target feed. Official Journal of the European Union, L40, 7–11.
Dai, S. Y., & Herrman, T. J. (2010). Evaluation of two liquid chromatography/tandem mass spectrometry platforms for quantification of monensin in animal feed and milk. Rapid Communications in Mass Spectrometry, 24, 1431–1438.
Dubois, M., Pierret, G., & Delahaut, Ph. (2004). Efficient and sensitive detection of residues of nine coccidiostats in egg and muscle by liquid chromatography– electrospray tandem mass spectrometry. Journal of Chromatography B, 813, 181–189.
Elliott, C. T., Kennedy, D. G., & McCaughey, W. J. (1998). Methods for the detection of polyether ionophore residues in poultry. Critical Review. Analyst, 123, 45R–56R.
Friedlander, L. G., & Sanders, P. (2009). Monensin. In Residue evaluation of certain veterinary drugs, Joint FAO/WHO Expert Committee on Food Additives, 70th Meeting 2008, FAO JECFA Monographs (pp. 109–135, Vol. 6) [Internet]. Available from:
Instituto Nacional de Controle de Qualidade em Saúde (2013). Métodos de Análise para Resíduos de Medicamentos Veterinários em Alimentos: Protocolo de Validação. POP 65.3120.136. Rio de Janeiro, 59p.
Kim, E., Bahn, K., Kang, E., & Kim, M. (2012). Quantitative analysis of lincomycin and narasin in poultry, milk and eggs by liquid chromatography–tandem mass spectrometry. Food Chemistry, 132, 1063–1070.
Lindsay, D. S., & Blagburn, B. L. (1995). Antiprotozoan drugs. In Adams, H. R. (ED.), Veterinary pharmacology and therapeutics. Iowa. 1181p. Chap. 47, 955–983.
Matabudul, D. K., Conway, B., Lumley, I., & Sumar, S. (2001). The simultaneous determination of the ionophore antibiotics in animal tissues and eggs by tandem electrospray LC–MS–MS. Food Chemistry, 75, 345–354.
Matuszewski, B. K. (2006). Standard line slopes as a measure of relative matrix effect in quantitative HPLC–MS bioanalysis. Journal of Chromatography B, 830, 293–300.
Mortier, L., Daeseleire, E., & Peteghem, C. V. (2005). Determination of the ionophoric coccidiostats narasin, monensin, lasalocid and salinomycin in eggs by liquid chromatography/tandem mass spectrometry. Rapid Communications in Mass Spectrometry, 19, 533–539.
Nász, S., Debreczeni, L., Rikker, T., & Eke, Z. (2012). Development and validation of a liquid chromatographic–tandem mass spectrometric method for determination of eleven coccidiostats in Milk. Food Chemistry, 133, 536–543.
Nebot, C., Iglesias, A., Regal, P., Miranda, J. M., Cepeda, A., & Fente, C. (2012). Development of a multi-class method for the identification and quantification of residues of antibiotics, coccidiostats and corticosteroids in milk by liquid chromatography–tandem mass spectrometry. International Dairy Journal, 22, 78–85.
Nebot, C., Iglesias, A., Regal, P., Miranda, J. M., Fente, C., & Cepeda, A. (2012). A sensitive and validated HPLC–MS/MS method for simultaneous determination of seven coccidiostats in bovine whole milk. Food Control, 27, 29–36.
Pereira, M. U. (2011) Desenvolvimento de um método para determinação de resíduos de ionóforos poliéteres em leite por CLAE-EM/EM/Mararlene Ulberg Pereira – Rio de Janeiro: INCQS/FIOCRUZ [Internet]. Available from:
Pereira, M. U., Spisso, B. F., Jacob, S. C., Ferreira, R. G., Monteiro, M. A., Costa, R. P., et al. (2015). Ocorrência de resíduos de ionóforos poliéteres em leite UHT comercializado na região metropolitana do Rio de Janeiro. Vigilância Sanitária em Debate: Sociedade, Ciência & Tecnologia. Article in press [Internet]. Available from:
Rokka, M., & Peltonen, K. (2006). Simultaneous determination Nigericin sodium of four coccidiostats in eggs and broiler meat: Validation of an LC–MS/MS method. Food Additives and Contaminants, 23(5), 470–478.
Shao, B., Wu, X., Zhang, J., Duan, H., Chu, X., & Wu, Y. (2009). Development of a rapid LC–MS–MS method for multi-class determination of 14 coccidiostat residues in eggs and chicken. Chromatographia, 69(9/10), 1083–1088.
Spisso, B. F., Ferreira, R. G., Pereira, M. U., Monteiro, M. A., Cruz, T. A., Costa, R. P., et al. (2010). Simultaneous determination of polyether ionophores, macrolides and lincosamides in hen eggs by liquid chromatography–electrospray ionization tandem mass spectrometry using a simple solvent extraction. Analytica Chimica Acta, 682, 82–92.
The Merck Veterinary Manual (2012). [Internet]. Available from:
Thompson, T. S., Noot, D. K., & Kendall, J. D. (2011). Determination of ionophores in raw bovine milk using LC–MS/MS: Application to residue surveillance. Food Chemistry, 127, 321–326.
Tkáciková, S., Kozˇárová, I., & Máté, D. (2010). Liquid chromatography tandem mass spectrometry determination of maduramicin residues in the tissues of broiler chickens. Food Additives and Contaminants, 27(9), 1226–1232.
Vincent, U., Chedin, M., Yasar, S., & von Holst, C. (2008). Determination of ionophore coccidiostats in feedingstuffs by liquid chromatography–tandem mass spectrometry. Part I. Application to targeted feed. Journal of Pharmaceutical and Biomedical Analysis, 47, 750–757.
Zhan, J., Yu, X. J., Zhong, Y. Y., Zhang, Z. T., Cui, X. M., Peng, J. F., et al. (2012). Generic and rapid determination of veterinary drug residues and other contaminants in raw milk by ultra performance liquid chromatography–tandem mass spectrometry. Journal of Chromatography B, 906, 48–57.