Introduction

Antibiotics are widely used for therapeutic and prophylactic purposes in human and veterinary medicine and also to promote growth and increase feed efficiencies in food producing animals [1]. However, abused use of antibiotics and their presence in the food of animal origin are of concern due to development of resistance in target pathogens against antibiotics, induced allergic reactions in some hypersensitive individuals, and potential compromise of the human intestinal and immune systems [2, 3]. There is a diverse range of chemical substances with antimicrobial activity. Among them, phenicols, including chloramphenicol (CAP), thiamphenicol (TAP) and florfenicol (FF) are readily available broad-spectrum antibiotics. They are effective against a wide variety of Gram-negative and Gram-positive bacteria, including most anaerobic organisms. Chloramphenicol was first isolated in 1947 from cultures of Streptomyces venezulae, and synthetic production began in 1949 [4]. CAP was used widely for many years in veterinary practice, both therapeutically and prophylactically, due to its excellent pharmacokinetic and antibacterial properties. However, CAP is hematotoxic to humans and is associated with side effects such as bone marrow aplasia (loss of ability to produce blood cells), and therefore, aplastic anaemia. These effects are unrelated to dosage and generally fatal [5,6,7]. TAP and FF have been used as alternatives to CAP, since they do not have the same side effects. However, even though it is prohibited, CAP is still used because of its efficacy and relatively low cost as well as the availability and prevention of some infectious diseases in mammals, birds, bees and aquaculture. In addition to its illegal use, products of animal origin can contain CAP residues because of its occurrence in the environment [8]. According to the literature, chloramphenicol can still be found in several food matrices, suggesting its continued use [6, 9,10,11,12,13,14,15]. Besides, there is little information available regarding the occurrence of its analogues in foods of animal origin and environment. Therefore, sensitive and reliable methods for the analysis of phenicols are needed.

Various analytical methods have been reported for the determination of TAP and FF in food, such as gas chromatography (GC) [16, 17], liquid chromatography (LC) [18,19,20,21], GC–mass spectrometry (MS) [40,41,42,43,44,45,46,47,48,49,50].

In the current study, various combinations of solvents, as described by previous reports, were evaluated. The method used by **e et al. [51] was found to be the most relevant for the extraction of all phenicols. It consists of extraction with ethyl acetate/acetonitrile/ammonium hydroxide mixture, followed by hexane wash and without further clean-up step. However, using this method, CAP gave poor recoveries. A second extraction with mixture of acetonitrile/ethyl acetate/methanol was evaluated separately [52]. A third extraction used only ethyl acetate [2]. Ethyl acetate was better for the recovery of most of the remaining CAP and the rest of the pharmaceuticals as compared to previous two extraction methods. Ethyl acetate, due to its polarity, provides a high selective extraction of nonpolar compounds including fat [41]. After extraction feed samples were centrifuged,12 mL of extract was evaporated at the nitrogen stream. Residues were resuspended in 2 mL of 30% methanol in Milli-Q water, mixed, then added 5 mL of Milli-Q water and then 2 mL of n-hexane was added to defat the sample, then manually mixed and centrifuged.

SPE was used for sample clean-up. Two different SPE disposable cartridges were tested for clean-up and pre-concentration. The method of cartridges preparation was the same for C18 BondElut and Strata-X cartridges. The cartridges were conditioned with methanol and water, than samples were loaded. Cartridges were washed with water and dried under vacuum for 5 min. Next phenicoles were eluted with methanol. The experiment showed that better recoveries and pure extract was obtained with the BondELUT C18 cartridge than Strata-X cartridges. Therefore, for the purification of phenicols from feed matrix we selected BondELUT C18 cartridges.

Method Validation

The developed procedure was designed to obtain a method for detection and quantification of florfenicol, thiamphenicol and chloramphenicol in non-target feed. Due to the lack of a certified reference material for all analytes under investigation, the accuracy and precision of the method in terms of repeatability (RSD %, intra-day precision), reproducibility (RSD %, inter-day precision) and uncertainty were determined using blank feed samples fortified with known amounts of analytes. The accuracy and precision values were calculated at three validation levels for each compound on three different days. The mean extraction recoveries ranged from 80.7% to 93.1% for CAP, 79.3% to 101.4% for TAP, 83.2% to 93.9% for FF. Overall mean recoveries ranged from 86.8% to 92.7% for all the three analytes. The method precision was estimated in terms of RSD. The intra-day and inter-day variations calculated for all the analytes ranged from 4.5% to 10.9% and 8.4% to 13.5%, respectively. The values of CCα (decision limit) and CCβ (detection capability) were calculated from precision experiments for all the compounds and are presented in Table 1. The LOD and the LOQ values were estimated from blank feed extracts. Matrix effects were ± 36%, which is in compliance with SANTE/11945/2017 requirements [53]. The expanded uncertainty was estimated to be in the range of 29.8–34.8%, depending on the analyte. Specificity is the ability of the method to distinguish between the analyte of interest and the other substances (impurities or matrix components) that may be present in a test sample. In the evaluation, the specificity of blank feed samples was analysed by the MS detector. The results obtained with blank samples were compared with CAP, TAP and FF spiked samples and no interfering peaks were observed. The method was tested in in-house comparisons by analysing samples to which known concentrations of antibiotics were added by various analysts and by analysing real feed samples. The method was shown to be appropriate for all the analytes with acceptable accuracy and precision. All validation parameters are presented in Table 1.

Table 1 Validation parameters of optimised LC–MS
Full size table

Conclusions

A quantitative LC–MS–ESI method was optimized for the simultaneous determination of three fenicols (chloramphenicol, thiamphenicol and florfenicol) in feed samples. Good separation of all the compounds in less than 20 min and pure chromatograph image was obtained. The method has been successfully applied to the analysis of feed samples for poultry and swine. The results obtained using the method in the analysis of commercial feed samples showed that the method is able to quantify phenicols at carry-over level in different kinds of feeds.