Dear Sir. Concerning your issue about the comparison of quantitation of antibiotic residues in food by ELISA and LC-MS. Antibiotic residue in meat is a serious public health concern due to its harmful effects on consumer health. This study aimed at estimating the residue levels of four commonly used antibiotics in meat samples using three analytical methods (ELISA, TLC and HPLC). A total of 150 samples of raw meat from sales points were analysed for ciprofloxacin, streptomycin, tetracycline, and sulphanilamide residues. Overall, ELISA analysis showed that 56, 34, 18, and 25.3% of the samples tested positive for ciprofloxacin, streptomycin, sulphanilamide and tetracycline residues respectively while TLC and HPLC detected 21.4, 29.4, 92.5, and 14.6%, and 8.3, 41.1, 88.8, and 14.6% of the samples as containing the residues, with ciprofloxacin and sulphanilamide having the lowest and

highest occurrence, respectively. Furthermore, the concentrations of antibiotic residues were in the ranges of 19.8–92.8, 26.6–489.1, 14.2–1280.8, and 42.6–355.6 _g/kg with ELISA, while HPLC detected concentration ranges of 20.7–82.1, 41.8–320.8, 65.2–952.2 and 32.8–95.6 _g/kg for sulphanilamide, tetracycline, streptomycin, and ciprofloxacin, respectively. Mean ciprofloxacin and streptomycin residue levels were above the Codex/SA MRL recommended limit, while 3% of the samples contained multidrug residues. Although some of the mean residues levels were below the permissible limits, the co-occurrence of multidrug residues in some of the samples calls for concern. I think the following below links may help you in your analysis:

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.1021.1290&rep=rep1&type=pdf http://www.ipcbee.com/vol5/34-X00061.pdf

Thanks

Isam Eldin Hussein Elgailani is correct concerning ELISA, TLC and HPLC but you asked about ELISA and LC-MS.

Let's look at the difference between the two techniques. ELISA is based upon the shape of part of the analyte. That means that an ELISA test can be prone to non-specificity. That can be good or bad depending on your need. Maybe you want to see "penicillins" in general. In that case, specificity might not be a good idea. Maybe you want an antibody that is sensitive to the beta-lactam structure.and identifies penicillins and cephalosporins. Or maybe you want to see the original drug and its metabolites--at least some of the metabolites may be cross-reactive. On the other hand, if you want absolute identification, it is hard to beat LC-MS. And, by spiking samples with a deuterated form, you can get absolute sensitivity despite complex matrices.

Of course, ELISA tests are relatively inexpensive and LC-MS has a significant capital expense. But if you already have an LC-MS available, to you the capital expenditure is zero.

So it comes down to what you really want to do. What are the most important goals of your study? Will you have simple or complex matrices? How much sensitivity is required? Is cross-reactivity a feature or a problem? Is an LC-MS available for use? Once you answer these questions, you can decide which technique is appropriate.

ELISA have been promoted in the 1980ies as being able to outperform or at least compete with HPLC-MS/MS.

ELISAs can be very sensitive and thus indeed compete with HPLC-MS/MS. However they are chromic in cross reactivity not only to other compounds of the same class but also to completely non target compounds. - It is quite an effort to get these cross reactions under control. ELISAs are designed to be high throughput (hundreds of samples per weeks) single compound tests. As the antibody is targeting one compound only, but the test can easily be conducted in 96 well plates. The prerequisite for ELISAs are that antibodies for the chosen compounds are available.

On the other hand currently the sensitivity of HPLC-MS/MS is unsurpassed. It is generally only a matter of a few hours to include a new compound into an existing multimethod. HPLC-MS/MS is in principle a multimethod allowing for analysing 20-100 compounds per chromatographic run. procedures on verification have been established and are unproblematic. However, HPLC-MS/MS is still taking time - most labs do 1-6 analysises per hour i.e 24-150 samples per day.

Considering costs: main costs in all analysises are personell costs ;-) in my world about 90% even though I work with relative expensive HPLC-MS/MS equipment.

this just as addendum to the quite well written other comments.