The Development and validation of an RP-HPLC-FD method for the multi-residue analysis of antiparasitic macrocycliclactones in cattle edible tissues Authors: Busra Aslan Akyol, Nurullah Guclu, Kubra Deliklitas, Oguzhan Korkut, Fazilet Sen Metin, Cengiz Gokbulut Published: 2026 ABSTRACT AbstractA method was developed and validated for the simultaneous determination of five macrocyclic lactones (ivermectin, abamectin, doramectin, moxidectin, and eprinomectin) in edible tissues (meat, liver, kidney, and fat) of cattle using a modified QuEChERS extraction procedure. Residue levels were determined by high-performance liquid chromatography with fluorescence detection. Recovery rates for all drugs across all edible tissues ranged from 73.91 to 96.15%. The limits of detection and quantification for macrocyclic lactones in cattle edible tissue samples ranged from 0.85 to 1.25 µg/kg and 2.55 to 3.80 µg/kg, respectively. Application of the method to real samples revealed the presence of macrocyclic lactone residues in 18 of 168 samples evaluated. One cattle meat sample (40 µg/kg, ivermectin) exceeded the maximum residue limit, while all other positive samples were below their respective maximum residue limits. This developed and validated method offers an alternative for the simultaneous analysis of multiple macrocyclic lactone residues in cattle edible tissues intended for human consumption, thanks to fewer processing steps, lower solvent volume, and fewer chemicals. Unlike other methods in the literature, this technique uses a very small sample amount (1 g) and solvent volume (3 ml). While these findings indicate that the developed method can be applied to detect macrocyclic lactone residues in edible tissues of cattle, potential matrix effects should be carefully evaluated, especially in matrices with high lipid content, such as adipose tissue (fat), given the target compounds high lipophilicity.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-57294-8.Keywords: Macrocyclic lactones, Ivermectin, Multi-residue, HPLC, Cattle, Meat, LiverSubject terms: Biochemistry, Biological techniques, Chemical biology, Chemistry, Drug discovery Introduction Avermectins and milbemycins, a class of macrocyclic lactones (MLs), are globally recognized as anthelmintic drugs for managing parasitic infections. Their widespread adoption in various animal species, including those destined for food production, stems from their remarkable efficacy against both internal and external parasites, even at minimal doses. These compounds are highly valued for their broad spectrum of action and low mammalian toxicity1. Abamectin (ABM), doramectin (DRM), eprinomectin (EPM) and ivermectin (IVM) are classified under the avermectin group, while moxidectin (MXD) is a member of the milbemycins. In terms of chemical structure, a key distinction between avermectins and moxidectin lies at the C13 position. Avermectins possess a saccharide group attached at this position, whereas moxidectin, which belongs to the milbemycin class, lacks this saccharide moiety (Fig. 1). Due to their pronounced lipophilic properties, these compounds accumulate within animal tissues, leading to prolonged residue persistence in the edible tissues of both small and large ruminant species2,3. The European Union (EU) has established Maximum Residue Limits (MRLs) for macrocyclic lactones (IVM, ABM, DRM, EPM, MXD) in the edible tissues (meat, fat, liver and kidney) of cattle. These limits are set at 10–50 µg/kg for meat, 10–500 µg/kg for fat, 20–1500 µg/kg for liver and 10–300 µg/kg for kidney4. Chemical structures of antiparasitic macrocyclic lactones. The profile of drug residues in animal products is influenced by numerous factors, including the animal species, dose, route of administration, pharmacokinetic and physicochemical properties of the drugs, and the nature of the products themselves. Consequently, the occurrence of residues exceeding the MRLs in food products is primarily caused by the failure to comply with the established withdrawal periods. These residues pose significant public health risks, potentially leading to hypersensitivity reactions, carcinogenicity, mutagenicity, reproductive issues, bacterial resistance, and disruption of normal gut flora5,6. Consequently, every veterinary drug used requires a defined withdrawal period after treatment. This period ensures that residue concentrations in the food product do not exceed the MRLs, the legally permissible maximum concentrations of veterinary drugs in food products derived from treated animals3. However, if these prescribed withdrawal periods are not adhered to, resulting in residue formation, the development of sensitive and straightforward analytical methods for their detection becomes critically important7. These compounds have generally been analysed using high-performance liquid chromatography (HPLC) with different detectors: ultraviolet (UV), photodiode array (PDA), mass spectrometry (MS), tandem mass spectrometry (MS/MS), and fluorescence (FLD) after derivatisation8–10. Although UV detection combined with HPLC (HPLC-UV) has been used to determine MLs in various matrices11, its sensitivity is often insufficient for analysis. To achieve the required sensitivity, some methods for determining ML in food samples are based on HPLC with FLD (HPLC-FLD) following derivatisation with trifluoroacetic anhydride and methylimidazole12–15. Additionally, thorough cleanup and extraction procedures are essential for analyses of complex matrices. Methods such as solid-phase extraction (SPE)16, liquid-liquid extraction (LLE)17 and immunoaffinity columns (IAC)18 each have drawbacks. SPE is generally expensive, requires large volumes of solvent, and uses impractical cartridges, limiting the number of samples that can be processed. Similarly, LLE requires large amounts of solvent and involves time-consuming processing steps, while IAC is costly and has limited high selectivity to avermectin drugs19,20. Furthermore, one study reported that the initial purification step with C18 SPE before IAC was unnecessary21. The QuEChERS (Quick, Easy, Cheap, Effective, Rugged and Safe) method, a widely recognised and utilised approach for pesticide residue detection in various food matrices, has also been successfully applied to the analysis of veterinary drug residues in animal products. Compared with conventional methodologies, QuEChERS is a rapid, simple, and economical approach that removes matrix impurities with fewer processing steps, resulting in high recovery rates22–24. A review of the literature indicates that studies on detecting ML residues in cattle edible tissues are limited in scope. Current methodologies focus on single ML analysis in a single tissue2 or on a limited number of ML combinations in specific tissues3,25–29. The number of methods that simultaneously perform multiple ML analyses (multiresidue) in more than one edible tissue30 is limited. Existing methods require large sample sizes (2.5–10 g) and solvent volumes (5–30 ml), and involve labour-intensive processing steps30–32. The main objective of this study is to develop a novel method for the simultaneous determination of five MLs (IVM, ABM, DRM, EPM, MXD) in various edible cattle tissues (meat, fat, liver, kidney) using HPLC-FLD, based on simplified preparation steps with minimal sample tissue and solvent consumption. Furthermore, the applicability of this method was tested on edible tissue samples intended for human consumption. Materials and methods Chemicals, reagents and Instrumentation The analytical standards of abamectin (95.6%, CAS number: 71751-41-2), ivermectin (99.9%, CAS number: 70288-86-7), eprinomectin (96.7%, CAS number: 123997-26-2), doramectin (99.9%, CAS number: 117704-25-3) and moxidectin (≥ 98%, CAS number: 113507-06-5), as well as HPLC grade methanol (≥ 99.9%, CAS number: 67-56-1), acetonitrile (≥ 99.9%, CAS number: 75-05-8), magnesium sulphate (≥ 99%, CAS number: 10034-99-8), sodium chloride (≥ 99%, CAS number: 7647-14-5), trifluoroacetic anhydride (≥ 99%, CAS number: 407-25-0), methylimidazole (99%, CAS number: 616-47-7) and glacial acetic acid (> 99%, CAS number: 64-19-7) were supplied by Sigma-Aldrich. Chromatographic analyses were performed using an Agilent 1260 HPLC system (Agilent Technologies, Germany), comprising a binary high-pressure isocratic system for the analysis of MLs. An Agilent binary pump (G1312B) delivered the mobile phase to the analytical column. The sample was injected via an Agilent autosampler (G1367E) connected to an injection valve (Rheodyne®, USA) with a 100-µL variable-loop. An analytical column (Zorbax, Eclipse plus C18, 5 μm, 250 mm x 4.6 mm) was used in conjunction with a nucleosil C18 guard column (Phenomenex). A fluorescence detector (G1321B) was used for detection, with data acquisition performed using ChemStation® Software (C.01.08, Agilent, Germany). All evaporations following sample extraction were performed at 60 °C using a vacuum evaporator (Maxi-Dry Plus, Hettich, Germany). A vortex mixer (622, Isolab, Germany), a homogenisation mixer (T 25 Ultra-Turrax, IKA, Germany), a centrifuge (Hettich Rotina® 380R, Germany), and an ultrasonic bath (Elmasonic® S40-H, Elma, Germany) were used for the extraction procedure. Standard solutions Stock analytical standard solutions (100 mg/L) of the five-ML mixture (IVM, ABM, DRM, EPM, MXD) were prepared in acetonitrile and stored in glass bottles at 4 °C. The stock solutions were diluted with acetonitrile to concentrations of 5, 10, 50, 100, 250, 500 and 1,000 µg/L. These solutions were used to spike drug-free tissue samples at varying levels, thereby generating standard curves and determining the recovery of the extraction procedure. Sample preparations and extraction procedures Tissue samples (50–100 g each) from cattle, specifically meat, liver, kidney and fat, were collected from various markets or butcher shops. To minimise lipid interference during extraction, the leanest meat tissue (specifically M. semitendinosus) was selected. Given the highly lipophilic nature of MLs, visceral (inner) adipose tissue, which has a high drug storage capacity and the slowest elimination1, has been identified as a representative matrix for residue detection. International authorities have also defined adipose tissue, along with the liver, as a target tissue for monitoring ML residues (especially IVM)33. Ensuring the reliability and representativeness of analytical results in heterogeneous tissues (especially liver and fat) is of great importance. Therefore, before weighing the 1 g analytical sample, all tissue samples were completely homogenized using a high-speed tissue homogenizer to obtain a completely homogeneous paste-like structure. The samples were stored at − 20 °C until analysis. Before chromatographic analysis, the frozen samples were thawed at room temperature and thoroughly mixed to ensure complete homogenisation, particularly of any fat layer present. Before analysing tissue samples, method validation was performed by spiking drug-free edible tissue samples (blind matrices) at concentrations of 5, 10, 25, 50, and 100 µg/kg. To account for biological variability, the blind tissue matrices used for method validation were obtained from multiple different animals. Before being used in enrichment (spiking) experiments, these samples were rigorously verified to be free of target drug residues. Each candidate sample was analyzed using the developed HPLC-FLD method. Only samples that showed no chromatographic interference peaks (signal-to-noise ratio, S/N < 3) and a flat baseline at the specific retention times of the target macrocyclic lactones were approved as true blind matrices and pooled for validation studies. Sample preparation involved derivatisation following a salt-assisted liquid-liquid extraction, with modifications based on the QuEChERS extraction method34. Specifically, 1 g of edible tissue samples (meat, liver, kidney and fat) was placed into six 15 ml capped plastic tubes. A multi-residue spiked matrix was then prepared by adding 50 µL of an analytical standard mixture containing IVM, ABM, DRM, EPM, and MXD to achieve concentrations of 5, 10, 25, 50, and 100 µg/kg. Subsequently, 1 ml of distilled water was added, and the mixture was thoroughly mixed. To initiate protein precipitation, 3 ml of acetonitrile was added to each tube. The tubes were vortexed for 1 min, then centrifuged at 5,000 rpm for 5 min. The supernatant was transferred to an additional 15-ml tube. Then, 0.4 g of magnesium sulphate (MgSO4) and 0.2 g of sodium chloride (NaCl) were added to each tube. The tubes were vortexed for 1 min, then centrifuged at 5,000 rpm for 5 min. The supernatant was transferred to a 10-ml glass tube and evaporated to dryness using a vacuum evaporator at 60 °C for approximately 2.5 h. For pre-column derivatisation, 0.1 ml of acetonitrile was added to the tubes and vortexed for 15 s. Subsequently, 0.1 ml of 1-methylimidazole (1:1) was added and vortexed for 15 s, followed by the addition of 0.15 ml of trifluoroacetic anhydride (1:2) and 0.05 ml of glacial acetic acid, which were vortexed. The samples were then transferred to 0.4 ml HPLC vials and incubated at 60 °C for 20 min, followed by cooling at 4 °C for 10 min. Finally, 35 µl of the prepared samples were injected into the HPLC system for analysis. Chromatographic conditions Chromatographic analyses were performed using an HPLC-FLD with excitation at 365 nm and emission at 475 nm. An analytical column was used in conjunction with a nucleosil C18 guard column. The mobile phase consisted of acetonitrile/methanol/ultrapure water (60:40:3, v/v/v). The analysis was carried out isocratically at a flow rate of 1.2 ml/min. The injection volume was 35 µl, and the chromatographic run time was 12 min. The mobile phase composition was adjusted as necessary before analysis of study samples to ensure complete chromatographic separation of the five target drug peaks from any co-eluting endogenous compounds in the tissue samples. Method validation The proposed analytical method was validated in accordance with the essential analytical performance criteria of regulation 2021/808 with EU requirements35. The validation was carried out using spiked edible tissue samples, and the following parameters were considered: linearity, response and range, accuracy and precision (repeatability and between-day reproducibility), selectivity, sensitivity, decision limit (CCα), and detection capability (CCβ). The intra- and inter-assay precision of the extraction and chromatographic analysis procedures was assessed by processing five spiked blank tissue samples (at 5, 10, 20, 50, and 100 µg/kg) as four replicate cattle edible tissue samples within a single day and on four separate days, respectively. The linearity of the analytical standards was assessed by analysing the five molecules at multiple levels spanning the entire working range. Additionally, the method’s linearity was evaluated using calibration curves prepared from replicates (n = 6) of blank tissue samples spiked with the five MLs at five concentrations (5, 10, 20, 50 and 100 µg/kg). The slope of the relationship between peak areas and drug concentration was determined by least squares linear regression. The linear concentration range was determined by linear least-squares regression, and linearity was assessed by the R2 value. To calculate signal suppression in the matrix-induced HPLC-FLD response, the matrix effect (ME) (%) was determined by comparing the slope of the calibration curve obtained from standard solutions with that from matrix-matched solutions spiked to the same concentration levels. Application of the method to real samples The developed method was applied to determine the presence of residues in cattle edible (meat, fat, liver, kidney) tissues. Samples of meat (72), fat (36), liver (30) and kidney (30), each weighing at least 50 g, were collected from various markets or butcher shops within the provincial borders of Balikesir in 2024–2025. The samples were promptly transported to the laboratory and stored at − 20 °C until analysis. They were then analysed using the previously described method to determine residue levels, and the results were used to confirm the validity of the developed method. Results and discussion HPLC-FLD chromatograms obtained for the simultaneous determination of MLs in cattle meat, liver, fat and kidney samples (analytical standard mixture, blank tissue and spiked tissues) are presented in Figs. 2, 3, 4 and 5, respectively. During chromatographic analysis, a characteristic peak was observed at the beginning of the chromatogram (approximately 2 min). This signal originates from the methylimidazole and trifluoroacetic anhydride reagents used in the pre-analytical derivatization process. Since the reagent peak left the system early, it did not have any interfering effect on the measurement of the target analytes or on peak separation. No interference was observed at the elution times of the drug peaks when blank tissue samples were analysed. A chromatographic analysis of MLs was performed in 12 min using isocratic elution. The intra- and inter-assay precision of the extraction and chromatographic analysis procedures was assessed by processing five spiked blank tissue samples (Supplements Figs. S1, S2, S3, and S4). The recoveries from the extraction procedure were determined by comparing the results obtained from extracted spiked tissue samples. Method precision was calculated using inter- and intra-day repeatability, expressed as relative standard deviation (RSD), as shown in Table 1. Chromatograms of an analysed standard mixture (A, 10 ng/ml), blank meat sample (B), spiked meat sample (C, 10 ng/g), and residue-positive [doramectin (D, 4.2 ng/g)]) and ivermectin (E, 40 ng/g)] meat samples processed by the suggested method. Chromatograms of an analysed standard mixture (A, 10 ng/ml), blank liver sample (B), spiked liver sample (C, 10 ng/g), and ivermectin residue-positive liver sample (D, 9.5 ng/g)] processed by the suggested method. Chromatograms of an analysed standard mixture (A, 10 ng/ml), blank fat tissue sample (B) and spiked fat tissue sample (C, 10 ng/g), processed by the suggested method. Chromatograms of an analysed standard mixture (A, 10 ng/ml), blank kidney tissue sample (B) and spiked kidney tissue sample (C, 10 ng/g), processed by the suggested method. Table 1Data from the regression analysis, recoveries (%), inter- and intra-day variations, LODs, and LOQs of the developed method for the determination of the anthelmintic macrocyclic lactones in various edible tissues of cattle.Tissue typesDrugsSpiked Conc. (ng/g)Recovery (%) ± SDIntra-day(RSD)(n = 20)Inter-day(RSD)(n = 16)Slope (ng – 1)Intercept r 2 LOD (ng/g)LOQ (ng/g)MuscleEPM5, 10, 25, 50 and 10093.03 ± 3.124.936.861.933− 0.6150.99991.13.50MOX91.86 ± 1.744.458.551.145− 0.5970.99990.852.55ABM91.85 ± 1.631.3612.281.575− 1.2420.99980.902.80DRM91.03 ± 3.000.9511.241.714− 1.6400.99840.952.87IVM91.10 ± 2.851.3410.831.676− 1.8520.99850.972.93FatEPM83.51 ± 5.797.76.933.7230.2490.99981.253.80MOX80.23 ± 6.328.227.951.850− 0.4040.99860.852.55ABM83.87 ± 5.609.056.632.043− 0.4040.99860.902.80DRM85.18 ± 6.577.717.702.043− 0.1160.99900.952.87IVM84.26 ± 7.227.598.262.4410.0120.99930.972.93LiverEPM86.34 ± 3.4310.413.983.8032.0250.99401.13.50MOX89.09 ± 4.556.375.101.7530.6970.99710.852.55ABM87.27 ± 4.839.185.532.0441.9470.99610.902.80DRM86.65 ± 4.519.75.202.4032.1450.99620.952.87IVM87.50 ± 5.5510.836.352.0032.7400.99710.972.93KidneyEPM88.83 ± 6.559.387.364.7341.3580.99901.13.50MOX85.55 ± 6.098.547.073.0870.8360.99890.852.55ABM87.07 ± 6.3610.57.254.1171.0270.99980.902.80DRM88.33 ± 6.138.846.904.2960.2800.99910.952.87IVM88.82 ± 5.3210.836.003.9791.0740.99960.972.93 Data from the regression analysis, recoveries (%), inter- and intra-day variations, LODs, and LOQs of the developed method for the determination of the anthelmintic macrocyclic lactones in various edible tissues of cattle. Various analytical methods (Table 2) for detecting ML residues in edible tissues of cattle have been reported in the literature8. Most of these approaches involve acetonitrile extraction, followed by clean-up on C8, C18 or HLB sorbents. Most contemporary sample preparation procedures are based on the C8 clean-up protocol developed by researchers36. This procedure has been adopted by several research groups, who substitute the C18 sorbent, which offers stronger sorption of non-polar co-extractives. The use of a liquid chromatography (LC) gradient in analytical methods is recommended to mitigate matrix carryover peaks, particularly in liver tissue19. A limited number of multi-analysis methods that allow for the quantitative analysis of ML residues and meet the European Commission validation criteria have been reported in the literature34. Among these methodologies, QuEChERS-based extraction approaches stand out for their advantages, including low volume, environmentally friendly use of organic solvents, high cost-effectiveness, and fast processing time. Furthermore, it has been reported that the QuEChERS approach offers a significant advantage over SPE in analytical workflows, providing higher sample throughput in a single step and optimised recovery rates15,19,27,28,37. Some studies have established IAC-based clean-up methods for the pre-analytical isolation of ML residues from cattle tissues, preceding both HPLC and LC-MS/MS analysis38,39. While IAC offers benefits such as enhanced selectivity, a notable limitation is that these IAC techniques typically encompass only avermectins, as the antibodies used do not cross-react with milbemycins8. An HPLC-FLD study analysing EPM residue in edible tissues from cattle reported recovery rates of 95 ± 11% for fat, 93 ± 12% for liver, 103 ± 13% for meat, and 87 ± 13% for kidney25. These high recovery rates are thought to be due to the study’s large sample size (5 g) and SPE steps using high volumes (10–15 ml) of solvents such as methylene chloride and acetone. In this study, the recovery values for EPM were determined as 86.34 ± 3.43%, 93.03 ± 3.12%, 88.83 ± 6.55%, and 83.51 ± 5.79% for liver, meat, kidney, and fat tissues, respectively. In this developed method, the extraction process was simplified as much as possible. In the analyses, 3 ml of acetonitrile was used per 1 g of tissue sample; extraction was performed without the need for additional solvent systems or complex clean-up steps. In this study, the sample mass was set at 1 g to establish the optimum balance between green analytical chemistry principles and analytical performance (sensitivity and accuracy). In preliminary optimization trials, using lower masses (e.g., 0.5 g) reduced detectability for target analytes, while masses of 2.0 g and above led to a significant increase in matrix effects and co-extracted interferences. The need for higher solvent volumes (> 5 ml) to overcome these drawbacks created by increased sample mass contradicts the environmentally friendly nature of the method. In conclusion, using a combination of 1 g sample and 3 ml of solvent minimized solvent consumption and achieved recovery rates ranging from 73.91% to 96.15% for all drugs in all tissues, as well as intraday and inter-day analyses (RSD < 13%). Most existing HPLC-based methods in the literature focus on the analysis of single or a limited number of MLs across a limited number of tissues. The main disadvantages of these protocols include high solvent and sample consumption, as well as labour-intensive extraction procedures that increase operational costs and analysis time2,3,25,27–32. This can make the methods difficult to implement, especially in cases requiring analysis of multiple tissues and multi-residues. In another study, an LC-MS/MS analytical method was developed to isolate eight ML residues (abamectin B1a, abamectin 8,9-Z isomer B1a, emamectin benzoate B1a, emamectin benzoate 8,9-Z isomer B1a, IVM, eprinomectin B1a, DRM, and MXD) from cattle tissues, including fat, liver, and meat. Samples were extracted with an acetone-aqueous ammonia mixture and partitioned into isooctane. The extracts were concentrated, re-suspended in n-hexane, and then diluted with acetonitrile before LC-MS/MS analysis10. Despite continuous advances in immunochemical methodologies (e.g., ELISA, DELFIA), no single antibody has yet been developed that demonstrates effective cross-reactivity with both avermectins and milbemycins. This suggests the potential to enhance moxidectin-specific antibodies, possibly through genetic engineering or chemical modifications, to broaden their recognition to include avermectins. Nevertheless, a significant constraint of these approaches is their customary reliance on initial sample preparation steps, such as SPE, which consequently prolongs the overall analytical duration. Therefore, for the wider application of immunochemical methods to detect ML residues, it has been emphasised that improvements in sample preparation procedures (e.g., ultrafiltration or immunoaffinity extraction) are necessary19. Table 2Comparison of analytical methods and performance for the quantification of the macrocyclic lactones in various edible tissues of cattle reported in the literature.DrugsTissueAnalytical instrumentSample preparationLinearity rangeLOD-LOQRecovery, % (SD)ReferencesIVMCattle meat, liver, fat, kidneyLC-MS/MSLLE2.5–200 µg/kg< 2.50 µg/kg (LOQ)– 3 DRMCattle meatHPLC-FLDSPE0.1–100 ng/g–– 2 IVM, EPR, EMA, DRM, ABM, MXDCattle meatLC-MS/MS and HPLC-FLDSPE–0.1–1.1 µg/kg (LOD),0.4–3.4 µg/kg (LOQ)88.9–100.7% 32 IVMCattle meat, fat, liver, kidneyHPLC-FLDACN–1 µg/kg (LOQ)76.7–93.9% 29 ABM, DRM, MXD, İVM, Milbemectin A3, A4Cattle meatHPLC-FLDSPE1–300 ppb0.21–0.44 ppb (LOD),0.64–1.34 ppb (LOQ)73.3–110% 31 EPMCattle meat, liver, kidney, fatHPLC-FLDSPE–1 ng/g (LOD),2 ng/g (LOQ)87–100% 25 EPM, ABM, DRM, IVMCattle liver, meatHPLC-FLDSPE0–100 ng/ml0.5–1 ng/g (LOD),1–2 ng/g (LOQ)70.31–93.65% 30 ABM, MXD, DRMCattle meatUHPLC-MS/MSQuEChERS50–200 µg/kg0.007–66 µg/kg (LOD),0.011–1.13 µg/kg (LOQ)92.8–99.3% 27 ABM, IVMCattle meatHPLC-PDAQuEChERS9–600 µg/kg3.2–12.5 µg/kg (LOD),9–30 µg/kg (LOQ)90.24–93.86% 28 EPM, eprinomectin; MXD, moxidectin; ABM, abamectin; DRM, doramectin; IVM, ivermectin; EMA, emamectin. Comparison of analytical methods and performance for the quantification of the macrocyclic lactones in various edible tissues of cattle reported in the literature. 0.1–1.1 µg/kg (LOD), 0.4–3.4 µg/kg (LOQ) 0.21–0.44 ppb (LOD), 0.64–1.34 ppb (LOQ) 1 ng/g (LOD), 2 ng/g (LOQ) 0.5–1 ng/g (LOD), 1–2 ng/g (LOQ) 0.007–66 µg/kg (LOD), 0.011–1.13 µg/kg (LOQ) 3.2–12.5 µg/kg (LOD), 9–30 µg/kg (LOQ) EPM, eprinomectin; MXD, moxidectin; ABM, abamectin; DRM, doramectin; IVM, ivermectin; EMA, emamectin. Linearity and range The linearity of the developed method was successfully demonstrated through both analytical standards and matrix-matched calibration curves in the 5–100 µg/kg range. Least squares regression analysis revealed a consistent relationship between peak areas and drug concentrations, yielding correlation coefficients (R²) of [≥ 0.9961] (Table 1) for the five macrocyclic lactones studied. This validated 5–100 µg/kg study range encompasses the international regulatory limits (MRLs) for the target analytes. However, the study range of the developed method (5–100 µg/kg) is lower than that reported in other studies in the literature [3,27,28]. Accuracy and precision The accuracy of the method was assessed by a recovery study, in which cattle edible tissue samples were spiked with known concentrations of all analytical standards (ABM, IVM, EPM, DRM, and MXD) at three levels (12, 32 and 74 µg/kg for meat; 15, 38, and 57 µg/kg for fat; and 14, 25, and 78 µg/kg for liver and kidney tissue samples), and mean percentage recovery and standard deviation were calculated. The extraction efficiency (recovery) of the five analytes was determined by comparing the peak areas obtained from spiked tissue samples with those from the injection of analytical standards after pre-column derivatisation. The precision and accuracy results are presented in Table 3. The overall range of MLs recovery rates in all analyzed cattle edible tissue samples was between 73.91% and 96.15%. The overall range of intraday and between-day variation (RSD) values, reflecting method accuracy, was between 0.95% and 12.28% (Table 1). The obtained recovery and accuracy data fall within the acceptable limits specified in the relevant validation criteria of the European Commission (2021/808) [35]. The lowest recovery rates for all MLs were found in fat tissue (73.91% to 91.75%). This is thought to be due to the challenging matrix nature of fat. Because of the nature of adipose tissue (fat), co-extracts that cannot be removed from the environment along with the analyte during extraction can lead to significant matrix suppression and, consequently, analyte loss in chromatographic analyses [40–42]. Matrix effects in fat tissue also varied between − 22.50% and − 13.08% (Table 4). ML analysis in sample tissues with a high lipid load, such as adipose tissue, increases the risk of signal suppression and analyte loss due to their lipophilic structure. However, with the low-volume solvent strategy, small sample tissue size, and simplified workflow applied in this study, the recovery values indicated in Table 1 were obtained without resorting to any additional purification steps. Table 3Precision and accuracy data (n = 3 for each added concentration) for the determination of the anthelmintic macrocyclic lactones in various edible tissues of cattle.Tissue typesAnalytesAdded (µg)Measured ± SD (µg/kg)Precision (RSD%)Accuracy (%)MeatEPM12.0012.90 ± 0.564.67107.5132.0030.42 ± 1.604.9995.9174.0073.17 ± 1.772.4098.88MXD12.0013.42 ± 0.756.24111.8732.0031.05 ± 2.116.5897.0274.0071.80 ± 1.622.1997.03ABM12.0012.68 ± 0.453.77105.6532.0032.30 ± 0.521.63100.9374.0072.37 ± 2.833.8297.79DRM12.0013.16 ± 0.786.53109.6732.0031.79 ± 0.000.0099.3474.0074.35 ± 1.542.08100.47IVM12.0012.47 ± 0.433.58103.8932.0030.81 ± 1.083.3896.2874.0071.79 ± 1.141.5497.02FatEPM15.0014.40 ± 0.452.9995.0538.0038.75 ± 2.406.31101.9857.0055.51 ± 3.175.5797.38MXD15.0014.12 ± 1.217.9893.1838.0038.52 ± 3.007.89101.3857.0058.32 ± 1.953.43102.31ABM15.0014.31 ± 0.674.4294.4338.0038.98 ± 2.446.42102.5757.0057.66 ± 2.925.12101.16DRM15.0014.94 ± 0.805.2598.6038.0039.60 ± 2.396.28104.2257.0058.71 ± 2.624.60103.00IVM15.0015.01 ± 0.704.6399.0938.0038.76 ± 0.962.52101.9957.0059.44 ± 1.332.33104.27LiverEPM14.0013.73 ± 0.463.2296.7025.0025.59 ± 1.174.67102.3678.0076.09 ± 4.105.2597.56MXD14.0014.55 ± 0.906.34102.5025.0026.25 ± 2.519.03105.0078.0076.28 ± 5.246.7197.80ABM14.0014.25 ± 0.684.79100.3625.0024.05 ± 1.184.7196.1978.0075.86 ± 1.562.0197.25DRM14.0014.28 ± 1.158.10100.5925.0024.91 ± 1.003.9999.6578.0077.56 ± 2.563.2899.44IVM14.0014.37 ± 0.503.51101.1925.0023.85 ± 1.797.1895.4178.0076.45 ± 1.081.3998.01KidneyEPM14.0014.23 ± 1.097.71100.2225.0025.79 ± 0.421.69103.1478.0080.76 ± 6.588.44103.54MXD14.0013.97 ± 0.543.7998.3925.0025.71 ± 0.351.41102.8478.0076.78 ± 4.976.3798.44ABM14.0014.46 ± 0.503.49101.8125.0024.91 ± 1.977.8799.6278.0076.78 ± 4.976.3798.44DRM14.0014.49 ± 0.785.50102.0525.0025.57 ± 1.355.41102.2978.0077.91 ± 6.147.8899.88IVM14.0013.68 ± 0.704.9696.3225.0025.82 ± 0.923.69103.2778.0077.15 ± 6.548.3998.91EPM. eprinomectin; MXD, moxidectin; ABM, abamectin; DRM, doramectin; IVM, ivermectin. Precision and accuracy data (n = 3 for each added concentration) for the determination of the anthelmintic macrocyclic lactones in various edible tissues of cattle. EPM. eprinomectin; MXD, moxidectin; ABM, abamectin; DRM, doramectin; IVM, ivermectin. Table 4The decision limit and detection capability values of the method were calculated at the maximum residue limit (n = 16) of the anthelmintic macrocyclic lactones in various edible tissues of cattle.Tissue typesDrugsMRL (µg/kg)(Codex Alimentarius)MRL (µg/kg)(EU Codex)Added (µg/kg)Measured (µg/kg)SDError α(1.64 × SD)CCα (µg/kg)CCβ (µg/kg)ME (%)MeatEPM100503229.421.602.6222.6225.25− 2.98–(− 10.50)MXD–503231.050.751.2321.2322.46− 14.53–(− 3.67)ABM–107473.671.833.0023.0026.00− 15.40–(− 5.34)DRM10403231.790.200.3320.3320.66− 12.44–(− 7.71)IVM30301212.470.430.7120.7121.41− 13.43–(− 3.04)FatEPM2502505755.911.372.2522.2524.49− 22.50–(− 8.99)MXD–5005758.321.352.2122.2124.43− 20.65–(− 11.17)ABM–101415.010.671.1021.1022.20− 22.15–(− 8.22)DRM1501503839.061.392.2822.2824.56− 19.33–(− 13.08)IVM4001005758.441.031.6921.6923.38− 20.04–(− 10.27)LiverEPM200015007878.001.883.0823.0826.17− 27.92–(− 14.65)MXD–1007876.280.951.5621.5623.12− 20.91–(− 15.09)ABM–201414.560.901.4821.4822.95− 19.60–(− 11.78)DRM1001007877.561.001.6421.6423.28− 22.43–(− 9.94)IVM8001007876.451.792.9422.9425.87− 21.90–(− 10.31)KidneyEPM3003007880.162.784.5624.5630.12− 14.71–(− 6.55)MXD–502525.710.350.5720.5721.15− 22.62–(− 12.68)ABM–107876.611.682.7622.7625.51− 22.19–(− 10.32)DRM30602525.571.362.2322.2324.46− 19.98–(− 13.17)IVM100302525.820.921.5121.5123.02− 20.57–(− 9.28)MRL, maximum residue limit; ME, matrix effect; CCα, the decision limit; CCβ, the detection capability; EPM, eprinomectin; MXD, moxidectin; ABM, abamectin; DRM, doramectin; IVM: ivermectin.CCα = CMRL + 1.64 x SD20 representative MRL—spiked samples.CCβ = CCα + 1.64 x SD20, representative samples spiked at the CCα level. The decision limit and detection capability values of the method were calculated at the maximum residue limit (n = 16) of the anthelmintic macrocyclic lactones in various edible tissues of cattle. MRL, maximum residue limit; ME, matrix effect; CCα, the decision limit; CCβ, the detection capability; EPM, eprinomectin; MXD, moxidectin; ABM, abamectin; DRM, doramectin; IVM: ivermectin. CCα = CMRL + 1.64 x SD20 representative MRL—spiked samples. CCβ = CCα + 1.64 x SD20, representative samples spiked at the CCα level. Selectivity and sensitivity The method’s selectivity was assessed by analysing five spiked tissue samples at concentrations both below and above the established MRL. Selectivity was evaluated by the absence of any interference at the retention time of each ML within the same analysis. Consequently, the methods were determined to be specific to the reference analytical standard of each drug. When blank tissue samples were analysed, no interfering peaks were observed in the chromatograms at the elution times of any analyte, confirming that the compounds were analysed without interference. Nevertheless, a prominent peak was observed within the first 2–3 min post-injection, attributed to the derivatisation agents (methylimidazole and trifluoroacetic anhydride), but it did not adversely affect the analytical results. The Limits of Detection (LOD) and Quantification (LOQ) for each drug were determined by analysing blank tissue samples spiked with the analytical standard mixture, with calculations based on the baseline noise measured at the peak’s retention time. LODs and LOQs were computed using the following equations, in accordance with guidelines [43]: LOD = 3 × standard deviation and LOQ = 10 × standard deviation. The LOQs for the MLs examined in tissue samples ranged from 2.55 to 3.80 µg/kg (Table 1). These results indicate that the method remained below the lowest MRL (10–1500 µg/kg) set by EU regulations for cattle tissues. In this regard, it is believed that the developed method can be used to monitor low levels of ML residues in edible tissues of cattle. The LOQ of the developed method is comparable with literature data for the analysis of single IVM in cattle tissues (liver, fat, meat, and kidney) using advanced techniques such as LC-MS/MS [3]. The results suggest that this HPLC-based approach could be an option for residue analysis in laboratory settings with high operational costs and limited instrumentation. The decision limit (CCα) and detection capability (CCβ) were evaluated in accordance with the EU requirements set out in Commission Decision 2021/808 [35]. CCα is defined as the concentration at which the MRL of the specified compounds can be detected to be exceeded with 95% confidence (α = 5%). In this study, to determine CCα values, 20 blank cattle edible tissue samples were enriched to the legal MRL levels of the examined MLs. Calculations were performed by adding 1.64 times the standard deviation of the relevant concentration to the MRL value. CCβ represents the lowest level at which the analyte concentration in the sample can be distinguished as above the CCα value with 95% confidence (β = 5% false negative probability). This value was obtained by adding 1.64 times the standard deviation of the CCα-enriched samples to the CCα value. The parameters related to CCα and CCβ are presented in Table 4. These parameters of the developed method allow accurate and reliable quantification of residues within MRLs. Application of the method to real samples The fundamental validation parameters were evaluated based on the core principles of EU Regulation 2021/808. Furthermore, the suitability of the method for its intended purpose has been assessed by its ability to accurately and reliably measure macrocyclic lactone residues under the MRLs defined by EU legislation. The method’s applicability was demonstrated by its successful use in determining potential ML residues in meat (72 samples), fat (36 samples), liver (30 samples) and kidney (30 samples), collected from markets and butcher’s shops in Balikesir, Türkiye. Analysis revealed positive ML residues in 4 meat samples (3 containing DRM and 1 containing IVM) and 14 liver samples (all containing IVM). Representative chromatograms of the positive samples are provided in Fig. 2. No residues above the limits of detection were found in any of the analysed kidney and fat tissue samples. Furthermore, ABM and EPM residues were not detected in any samples. In 1.38% of meat samples (95% confidence interval: 0.25%–7.46%), IVM residue (40 µg/kg) was found to be above the MRLs for meat (30 µg/kg), while DRM residues (2.9–4.2 µg/kg) were found in 4.16% (95% confidence interval: 1.43%–11.60%), below the MRLs for meat (40 µg/kg). In 46.67% of liver samples (95% confidence interval: 28.80%–64.50%), IVM residues (3.3–9.5 mg/kg) were found to be below the MRLs (100 µg/kg) for liver. The highly lipophilic character of IVM results in a large volume of distribution in the body and a tendency to concentrate particularly in lipid-rich tissues, such as adipose tissue (fat) [44]. The slow elimination kinetics in the tissue prolong the drug’s tissue half-life, leading to long-term preservation of residues [2,45]. To assess the potential risks of detected IVM residues to consumer health, the Estimated Daily Intake (EDI) was calculated as: (detected concentration (µg/kg) x average intake (kg/day) / average body weight (kg)) (9.5 µg/kg x 100 g/60 kg = 0.0158 µg/kg/day). Based on the Joint FAO/WHO Expert Committee on Food Additives (JECFA) recommended daily intake of 100 g for liver tissue and an average body weight of 60 kg [46], the highest detected residue level (9.5 µg/kg) contributed to a daily intake of 0.0158 µg/kg. In addition, based on the JECFA recommended daily intake of 300 g for meat tissue and an average body weight of 60 kg [46], the highest detected residue level (40 µg/kg) contributed to a daily intake of 0.2 µg/kg. These values are calculated based on the Acceptable Daily Intake (ADI) of 10 µg/kg bw/day (0–10 µg/kg bw/day) for IVM; for a 60 kg person, the daily intake is 600 µg/person/day. The values remain below µg/person/day. Although the detected EDI values are below the ADI limits, the cumulative and adverse effects on non-target systems that may occur as a result of continuous intake of these drugs through edible animal foods may constitute a potential risk factor for public health. In this context, developing analytical approaches with fewer processing steps that are operationally practical and technically accessible is crucial to ensure the continuity of control processes. It is considered that the developed methodology, with its simplified workflow, flexible structure that allows for multi-residue analysis across different matrices, and minimal sample/solvent consumption, can provide an alternative to meet existing analytical requirements. Limitations Seasonal variation in parasite load in cattle directly influences antiparasitic treatment strategies and dosage regimens used in the field. Intensive therapeutic interventions, particularly during periods of increased parasite prevalence (such as humid and mild seasons), are likely to be reflected in residue levels and detection rates in edible tissues. Future studies employing a systematic sampling strategy across different seasons of the year and with a larger sample size will contribute to a clearer understanding of the role of seasonal effects in residue accumulation. One limitation of this study is that the measurement uncertainty profile of the method has not been calculated. Although the developed method meets the requirements of international guidelines in terms of accuracy, precision, and sensitivity, a comprehensive measurement uncertainty budget based on long-term quality control data is planned to be created in future studies for full integration into accredited routine analyses. While the method met the core performance criteria of Regulation (EU) 2021/808, specific long-term analyte stability studies (such as matrix-matched stability and freeze-thaw cycles) were not conducted in the present work. The justification for this omission is based on the well-documented high chemical stability of macrocyclic lactones in both organic solvents and biological tissues reported in extensive prior literature [47–50]. Specifically, extensive within-laboratory reproducibility tests involving different operators, distinct reagent batches, varying environmental conditions, and multiple instruments as well as formal ruggedness and long-term analyte stability evaluations, were beyond the scope of this initial optimization study. To mitigate any potential degradation during our study, strict laboratory protocols were followed: all standard working solutions were freshly prepared, and sample extractions were immediately followed by HPLC-FLD analysis on the same day. Conclusion Given the high sensitivity and selectivity required for MLs analysis in edible food products, the combination of HPLC-FLD and the modified QuEChERS method offers significant advantages. The QuEChERS approach mitigates the limitations of traditional techniques such as SPE or LLE, including the need for expensive cartridges, the use of large volumes of toxic solvents, and lengthy processing steps. Furthermore, it provides a more flexible extraction framework than methods like IAC, which exhibit limited selectivity. The modified QuEChERS protocol used in this study, combined with a pre-column derivatisation step, enabled the simultaneous determination of five MLs in multiple edible tissue samples. This integrated approach, while maintaining the analysis’s operational simplicity, enabled measurements within MRLs (µg/kg) using FLD. However, for routine laboratories with limited access to advanced equipment such as LC-MS/MS, it provides an economical, reliable, and accessible alternative method. The method requires fewer samples and less sample and solvent than the intensive solvent use and complex extraction steps required by systems such as LC-MS/MS. In conclusion, these advantages of the modified QuEChERS method have significant potential for integration into different high-sensitivity detector systems, such as LC-MS/MS, in future research. Additionally, analyses of animal products intended for human consumption have revealed a widespread distribution of IVM contamination, even though residue levels in a large proportion of samples were below the MRL. Therefore, it is recommended that legal withdrawal periods be strictly observed before animals treated with these drugs are sent to slaughter, or that the use of these drugs be avoided in animals with an approaching slaughter schedule. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements During the preparation of this work the author(s) used (Grammarly) in order to language editing assistance. Author contributions C.G. project administration. C.G., B.A.A., N.G., K.D., O.K. and F.S.M. performed material preparation, data collection and analysis. C.G. and B.A.A. wrote the main manuscript text. All authors reviewed the manuscript. Funding This study was supported by the Scientific Research Committee of Balikesir University (Balikesir, Türkiye) as a research project (project no: 2023/155). Data availability The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Should any raw data files be needed in another format they are available from the corresponding author upon reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information files. Should any raw data files be needed in another format they are available from the corresponding author upon reasonable request.