The ELISA test and the enigma of the single tall bar in the chart

Marcin Śmiałek, Joanna Kowalczyk, Bartłomiej Tykałowski, Andrzej Koncicki 

Chair of Avian Diseases, Department of Veterinary Medicine, University of Warmia and Mazury in Olsztyn, Poland

ELISA tests, and serological assays in general, are highly practical tools for rapid and relatively low-cost diagnosis of infectious diseases and epidemiological monitoring of poultry flocks (also having wider application). What is the one tall bar in an ELISA test result chart?

Appraising the data presented above, it is impossible not to form the impression that the result climbs rapidly as we read rightwards and then suddenly “hits a wall” which it cannot break through. This impression could not be falser, because the result actually “hits” the upper limit of sensitivity of the test. Quite understandably, this may seem odd, given that the average titer fluctuates in a 6,700–6,800 range and the histogram clearly shows the bars not rising at all for 8 out of 18 groups. How may this be explained? It transpires from more detailed analysis of the results shown that the OD of the majority of the tested samples does not exceed a maximum value of 3 (this may not be higher because of the intensity of stain of the diluent in the ELISA plate wells). However, the issue of the high antibody titer value of the sample still remains. In the specific case of this test, the OD of the positive control was also very high (1,082–1,149). If the OD of the tested sample reaches the maximum value and the positive control OD rises at the same time, then the S/P ratio (sample OD / positive control OD) falls, and as a consequence the antibody titer also falls (this being calculated on the basis of the S/P parameter). A paradoxical situation then arises, because mathematically, the average titer of 6,700–6,800, as one perhaps undeserving of particular attention in the normal course of things, transpires to be an out-of-range result in this test. In such cases, the interpretation of the test results must be appropriate and the titer must be acknowledged to be a purely mathematically derived numerical value which may not represent the situation in the flock regarding the specific epidemiological risk tested for.

This summarises an article published in Polski Drobiarstwo [Polish Poultry Farming], issue 1, 2021. The full article may be found at