IDEXX Results Interpretation of Serological Tests
IDEXX Results Interpretation of Serological Tests
+ + + + + + + + + + + CREATING CLARITY
1 VOL.1 RESULTS INTERPRETATION OF SEROLOGICAL TESTS
General avian serology
Contents
Serology – definition.................................................................................................................................................................. 4
As mentioned, several methods can be used to detect antibodies and antigens in avian serology (Fig. 1.), but in practice
ELISA and the hemagglutination inhibition (HI) test are most often used, which allow not only to answer what antibodies
are present in the serum, but also how many antibodies are contained in it (ELISA) or what is their hemagglutination
activity. Currently, the plate agglutination test and the agar gel precipitation test (qualitative tests) are less frequently
used, although they can still be very useful in some diseases. This applies in particular to the plate agglutination test
used in the diagnosis of mycoplasmas or the agar gel plate test (AGPT) test used to diagnose Marek’s disease
(Szeleszczuk, 1995).
The scientific basis for the introduction of ELISA for the diagnosis of poultry diseases was developed by a team of
researchers from the US University of Maryland (Snyder et al., 1983; Snyder et al., 1983; Snyder et al., 1984; Snyder
et al., 1986).
The use of the ELISA test for serological diagnosis of poultry diseases enabled the emergence of a new strategy in
veterinary care over large-scale poultry farming by introducing serological monitoring to the broad practice (Snyder
et al. 1985; Szeleszczuk, 1997).
Functional
Agglutination
- Plate Agglutination Test (RPA)
Hemagglutination
- Hemagglutination inhibition (HI) test
Neutralization
- Seroneutralization test (SN)
Quantitative
Chemical and physical methods ELISA
Antigens
Add sample
Wash + conjugate
Wash + substrate
OD
Positive result
Borderline titer
Negative result
Titer
Fig. 3. Relationship between antibody concentration (titer) and optical density (OD) in an indirect test
Antigens
Add sample
Wash + conjugate
Wash + substrate
OD
Negative result
Borderline titer
Positive result
Titer
Fig. 5. Relationship between antibody concentration (titer) and optical density (OD) in the blocking ELISA
Chicken, turkey,
falcon, flamingo,
IDEXX ELISA test for the detection quail, Japanese
Avian
INFAvian* Influenza A of antibody to influenza A virus quail, laughing S/P > 0,5
influenza virus
Ab Test in animal serum gull, house
finch, starling,
and herring gull
Box 1. Statistics
In addition to the central tendency described by the mean titer (AMT, GMT), scatter is a fundamental
sample characteristic in serology testing in poultry. A poultry flock consists of many individuals whose
reaction to contact with the antigen leads to different levels of serological response. This diversity is
a natural phenomenon resulting from the genetic structure of a given individual or the current efficiency
of its immune system. Therefore, the obtained, different results of serological tests must be processed
using indicators describing the degree of their differentiation. In the case of elaborating the results of
serological tests from a given population (flock), it is assumed that they have the form of the so-called
normal distribution, otherwise known as the Gaussian distribution, illustrated by the Gaussian curve,
which distribution is the most important theoretical probability distribution in statistics (Moczko et al.,
1998). The normal distribution is also the most intuitive statistical distribution. In short, it describes
a situation in which most cases are close to the average result, and the more a given result deviates from
the average, the less numerous it is. The simplest way to say is that in populations with a normal distribu
tion, most serum titers are close to the mean value, and the further we move away from the mean result,
the fewer sera there are (Fig. 6).
The most commonly used measures of dispersion in medical research are: variance, standard deviation
and coefficient of variation (Andrasiak et al., 2018). Dispersion (scattering, scattering, dispersion) is the
variation in the observed values of a variable; it is the greater, the more these values deviate from the cen
tral tendency, i.e., the mean titer. The desired probability function of the random variable has the shape of
a normal curve (Gaussian Function). Gaussian curve (Fig. 6.) describes a normal distribution in which the
curve is symmetrical about the axis determined by the mean value and reaches its maximum here (high
and low titer values occur less frequently, the more they differ from the sample mean).
Box 2. Statistics
Incidence
-3 -2 -1 0 +1 +2 +3
68%
95%
99,7%
Standard deviation
As you can see, about 68% of the observations are close to the mean, within one standard deviation of the mean
(a measure of distance in the language of statistics). As you move away from the mean, the Gaussian curve descends.
As much as 95% of observations are in the range of -2 to +2 standard deviations from the mean value. Extreme values
(at the ends of the Gaussian curve) are represented by a negligible percentage of observations (Fig. 8.). This means that
when the distribution of antibody levels is close to the normal distribution, we can say that there are no anomalies in the
tested sample, the obtained data are normal, there are few extreme observations, and a significant part of the observa
tions is concentrated around the mean. The occurrence of such anomalous distributions means that the properties of the
normal distribution cannot be applied to them, and thus many statistical tests cannot be used, because their results may
be distorted by the occurrence of a non-standard distribution of results (Fig. 9.). In practice, the normal distribution of
results is most often disturbed by combining chicks from different parent flocks with different immunological status in the
hatchery. This practice, sometimes referred to as chicks mixing, makes it very difficult to determine the proper timing of
vaccinations, e.g., against Gumboro disease (Fig. 10.).
Fig. 7. Distribution of IBD virus antibody titers in 39 flocks of day-old broiler chicks
In the presented example, it can be observed that the more measurements we make, the more the resulting curve will
resemble a symmetrical bell. If (which is of course impossible) we analyzed a group with an infinite number of cases, we
would get a normal distribution. In practice, the sample we study always has a limited number of elements, but it is worth
remembering that the more elements we include in the study, the better the actual distribution of the examined variable
in the analyzed population will be reproduced. At this point, it should be strongly emphasized that too few samples tested
may cause the results of the serological test to be unreliable and misleading. In practice, it is recommended to collect
23-30 blood samples, regardless of herd size.
Fig. 8. Example of distribution of ELISA results in a flock of day-old broiler chicks from a very well IBD vaccinated parent flock
In Fig. 8. the effect of examining a flock of one-day-old broiler chicks coming from a very well-protected parent flock was
illustrated. Noteworthy is the very low value of the coefficient of variation (CV - 10.4%) and the difference between GMT
and AMT of only 50 and the very low value of the standard deviation (954). In such a herd, scheduling vaccination against
Gumboro disease will not be a problem.
Fig. 10. Example of the variability of serological test results in the case of chicks from different parents’ flocks.
To describe the data obtained in quantitative serological tests, it is necessary to use descriptive statistics to better
understand the level of specific antibodies in the population, and its distribution. The values obtained because of the use
of descriptive statistics are also the basis for further, more advanced methods of statistical analysis, such as testing
differences between groups or examining relationships between selected indicators (e.g., the level of specific antibodies
against a given pathogen and the performance of the flock).
The variance of the denominators x1, x2, ..., xn is the arithmetic mean of the squared deviations from their arithmetic mean:
The standard deviation is the square root of the variance. The standard deviation determines approximately how much
all statistical units of a given population differ on average from the arithmetic mean of the variable under study. It is assu
med that the value of the standard deviation for a normally distributed population should not be greater than 50% of the
mean value. The standard deviation provides us with the necessary knowledge about whether the results in a particular
group of results are like each other - whether the group of titers is similar or different, and if so, to what extent (Moczko
et al., 1998).
We start the analysis of the test result by reading the description of the tested material. It is located above the histogram
and can have a very different form, but it always contains the research direction in the form of an acronym (Tab. 1.).
A description identifying the flock is printed here. This is extremely important information for the correct interpretation of
the result. Although the recording solutions used by specific laboratories differ in form, they should allow full identification
of the ordering party and the tested material.
The system of describing/marking samples in serological monitoring tests of poultry flocks in the country is not unified,
which is often an obstacle to a full analysis of the serological test result. Very often, in practice, the correct interpretation
of the serological monitoring result is hindered by the lack of additional information relevant for this assessment, such as
the age of the birds or the vaccination program.
The recommended structure of material and stock description is given in Tab. 2. and the description scheme most often
used in practice is contained in Tab. 3.
1. 2. 3. 4. 5. 6.
Tab. 3. The most used description in practice of serologically tested samples taken from a poultry flock.
Fig. 11. Report of IBD test results performed using xChekPlus* software
10/12/AP/AB01/NOWAK/EŁK/S3/K3/6-0/CB/F15/120301 -IBD
Count: 23
10
Mean: 8527
9
GMean: 7738
8
SD: 3278
7
%CV: 38,4
6
Count
Min: 2899
5
Max: 13219
4
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Titer Groups
Fig. 12. A printout of the result of the test for antibodies to Gumboro disease virus
Case Comment
10/12/22/AP/AB01/Jan Nowak/Elk/S3/K3/6-0/CB/F15/120301/P/ ///M1HI//N1AS//I1MS//I10VW//G18HW/G24HW
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
S3/ Sector
6.
K3/ House
12. /// Three slashes separate flock data from the vaccination program
16. I10VW// Vaccination against IB at day 10 with a variant vaccine by drinking water
17. G18HW/ Gumboro disease vaccination at day 18 with an intermediate plus vaccine by drinking water
18. G24HW Vaccination against Gumboro disease at day 24 with an intermediate plus vaccine by drinking water
H = HVT
N Marek’s disease MD I = injection
R = Rispens
A = asymptomatic S = spray
N Newcastle disease ND
L = lentogenic W = in drinking water
M = Mass S = spray
I Infectious bronchitis IB
V = variant W = in drinking water
P = intermediate
G Gumboro disease IBD W = in drinking water
H = intermediate plus
S = spray
A Metapneumovirus infections APV
W = in drinking water
S = spray
C Coccidiosis Coccidiosis
W = in drinking water
The most important advantage of the proposed sample coding system is the ability to quickly search for results of inte
rest, stored in the databases of programs used to generate test results. With proper coding, it is possible to prepare va
rious reports, e.g., collecting the results of blood tests from the age of 6 weeks in one chart., taken by a specific submitter,
vaccinated against Gumboro disease.
In statistics, 2D histograms are a graphical representation of the frequency distribution of a selected variable(s), in which
the columns (bars) are plotted above the group ranges (denominator groups 0-18), and the height of the columns is pro
portional to the number of sera in each titer group. All sera with titers below the cut-off titer (negative) are in group 0.
In the given example, there are 23 tested sera in this group.
Titer group number. To facilitate the recording and analysis of serological monitoring data, the obtained titer was grouped
into titer groups according to the rule given in Table 6.
Next to the histogram, there are basic statistical data, most often used when analyzing the test result.
Quantity
The geometric mean of the titers is the nth root of their product. The geometric mean in statistics is most often used to
calculate the average severity of changes. In serology, the geometric mean of titers is most often used, as it better reflects
the distribution of titers in the studied population of birds.
The geometric mean of the denominator x1, x2, x3, ..., xn is expressed by the formula:
x1 – serum titer 1
x2 – serum titer 2
x3 – serum titer 3
xn – serum titer n
n – quantity of tested sera
Arithmetic mean
Arithmetic mean titer (AMT), sometimes referred to as mean titer, by default - mean titer.
The arithmetic mean of the titers is the sum of these titers divided by their number. The arithmetic mean belongs to the
classic average measures and expresses the average level of the observed feature.
The arithmetic mean of x1 , x2 , x3 , ..., xn is given by:
x1 – serum titer 1
x2 – serum titer 2
x3 – serum titer 3
xn – serum titer n
Standard deviation – standard deviation is a classic measure of variability, next to the arithmetic mean, the most frequen
tly used concept used in the language of statistics. Standard deviation (SD), defined as the square root of the quotient
of the sum of squares of deviations of the variable (X) from the arithmetic mean of the set (x¯). Intuitively, the standard
deviation tells you how widely the values of a quantity are spread around its mean, the greater the standard deviation, the
greater the variance in the studied population. It is a measure of the distance of individual results from the average. The
further a given result is from the mean in units of standard deviation, the more unusual it is, i.e., it is inconsistent with the
normal distribution.
n
i=1
% CV
Coefficient of variation - is a classic and extremely useful measure of titer distribution variability in the analysis of serolo
gical test results. The coefficient of variation is a relative measure, i.e., it depends on the size of the arithmetic mean, as it
expresses the variability of a feature in relation to the value of this mean. The coefficient of variation is usually expressed
as a percentage.
The coefficient of variation characterizes the variability of the result. If the CV is less than 40%, it is assumed that the
dispersion of the results, i.e., the differentiation of changes between individual individuals, is satisfactory, which may
mean that all birds in the flock reacted similarly to the vaccine administration. A high coefficient of variation means that
the variation of titers between individuals in the flock is too large, which is not a favorable phenomenon. The valuation of
the coefficient of variation value is particularly important when determining the timing of vaccination against Gumboro
disease. Reliable determination of the optimal date of vaccination is possible only when the coefficient reaches the value
40%. In samples taken from birds vaccinated with live vaccines, the value of this ratio should not exceed 60%.
It is short for “minimal” - antibody titer. The assessment of the minimum titer value is important, for example, when evalu
ating the results of day-old chicks, as it allows to assess whether all tested individuals were sufficiently protected against
early infection.
Max
Itis short for “maximum” antibody titer. The assessment of the maximum titer value is important, for example, when evalu
ating the results of tests in vaccinated birds suspected of infection with field virus. Very high maximal titers indicate that
such a high serological response cannot be expected from the applied vaccination program, therefore they may confirm
clinical symptoms characteristic of a given disease.
Operator
Data
Next to the histogram, the result of the ELISA test is presented as a table data (Tab. 7). The TEST MATERIAL field contains
information about the test, which enables its identification, the reason for testing and the code of the technician perfor
ming the test.
The first column contains data describing the results of testing control sera - positive and negative in duplicate, followed
by the numbers of the tested sera, and in the second column their location in a specific well on the microtiter plate.
Optical density
It is worth emphasizing that the only variable determined empirically in the ELISA test is the optical density value in the
third column of the data table. Optical density (OD), sometimes called extinction or absorbance, although not quite cor
rectly, is a physical quantity described by the tenth logarithm of the quotient of the intensity of a monochromatic beam
entering the absorbing medium and the intensity of the beam passing through this medium (according to the Lambert-
Beer law). In simpler terms, optical density for transparent materials is measured by the ratio of the intensity of light
that has passed through a particular layer of a substance to the light incident on it. Theoretically, it is assumed that the
S/P ratio
Since the rate of the enzyme reaction is dependent on the conditions under which the assay is performed, especially on
the temperature in the laboratory, the relative amounts of antibodies in the test samples are calculated by reference to the
positive control, corrected for the OD values of the negative control sera. In a laboratory where the test is performed at
a higher temperature, the OD values are higher than those found in a laboratory performing the test at a lower tempera
ture. A comparison of OD values may therefore be unreliable. Regardless of the test conditions, the result expressed as
the value of the S/P ratio in indirect tests is comparable, because the proportions are maintained. For this purpose, the
OD value of the test serum, i.e., the sample taken for testing, is divided (S, sample) by the OD value of the positive control
serum of the kit (P, positive). The value of the coefficient is calculated from the formula:
In the blocking ELISA format, the S/N ratio is calculated. Regardless of the test conditions, the result expressed as the
value of the S/N ratio in the blocking format tests is comparable, because the proportions are maintained. The value of
the S/N ratio is calculated from the formula:
Cut-off titer
The so-called the cut-off titer is the value of the titer of S/P or S/N ratio at which the result is considered positive. This
value is set by the kit manufacturer. It is usually determined empirically by testing a large number of positive and negative
samples. Values are selected that maximize the sensitivity and specificity of the test. Depending on the direction of the
examination and the type of test, it ranges from 0.2 to 0.6. For example, samples with an S/P greater than 0.2 (titer greater
than 396) when tested with the IDEXX IBD Ab assay are considered positive.
396
OD value 0,0 0,2 >6,0
Antibody concentrations
The titer is one of the basic parameters assessed during the interpretation of the serological result. The titer is calculated
according to the formula provided by the test manufacturer, and it varies depending on the formula used by the test manu
facturer to calculate it. All formulas are based on the OD value and the S/P ratio calculated on this basis.
The sample algorithm has the following formula:
*Refers to the S/P ratio at a 1:500 dilution to the end point, defined as the reciprocal of the highest dilution that gives
a reading above the cut off value.
In indirect tests by IDEXX, the cutoff titer is usually 396 (dilution 1: 500; S/P > 0.2) [Tab. 6.].
As shown in Figure 14, depending on the ELISA manufacturer’s titration formula, the results of the same sera from diffe
rent manufacturers may vary significantly, which should be considered when interpreting test results. If comparisons are
necessary, use S/P (or S/N) values that are independent of the titer calculation formula.
The next column in the table of results contains information to which group of titers the test result of a given serum belongs.
25000
20000
15000
GMT
10000
5000
0
A B C A B C A B C
Age of birds
Fig. 14. The level of antibodies against IBDV in serological monitoring performed using tests from 3 manufacturers (A, B and C) in 1. day
on 3. and 6. week in a herd vaccinated against Gumboro disease (Szeleszczuk et al., 2014)
In poultry practice, the hemagglutination inhibition test is also widely used for the serodiagnosis of infectious diseases
caused by hemagglutinating pathogens. It is a serological test used to detect antibodies formed after spontaneous infec
tion, as well as post-vaccination antibodies. It uses the fact that hemagglutinin of surface viral receptors has a clumping
effect on red blood cells (Hirst phenomenon). Regarding poultry diseases, this feature has e.g., Infectious Bronchitis virus,
Newcastle disease, Egg Drop Syndrome 1976, and Avian Influenza virus. The HI test is also used in the serodiagnosis of
mycoplasma infections. Specific antibodies, which are a humoral immune response to infection or vaccination, can inhibit
agglutination by binding to hemagglutinins of pathogens, i.e., they have the ability to inhibit pathogen-induced erythrocyte
agglutination. If the test serum contains specific antibodies against haemagglutinin, the addition of the test pathogen
causes an antigen/antibody reaction. This is visible after the addition of erythrocytes. They are no longer agglutinated as
the reaction is inhibited (Fig. 15.).
The technique of performing the HI test consists in introducing the test serum into the wells of the microplate and making
its successive dilutions, and then adding the pathogen of a certain concentration (e.g., 4 hemagglutinating units) and
a suspension of red blood cells (Fig. 16.). If the tested serum contains specific antibodies directed against this patho
gen, its hemagglutinating properties are blocked, and as a consequence the blood cells settle in a compact group at the
bottom of the well, forming a nub. If there are no antibodies specific for a specific subtype in the tested serum, then the
pathogen, due to its ability to agglutinate blood cells, causes them to settle at the bottom of the well in the form of a cha
racteristic mesh. The result is read visually. The antibody titer is the reciprocal of the last serum dilution that completely
inhibits hemagglutination. The strengths of the test are: the possibility of selecting a specific antigen for testing depen
ding on the changing epizootic situation, the possibility of simultaneous detection of antibodies to different subtypes of
Dilution
1/1024
1/256
1/128
1/512
Neg.
Titer log2
Pos.
1/64
1/32
1/16
1/4
1/8
1/2
Titer
1 16 4
2 128 7
3 2 1
4 1024 10
Serum
5 32 5
6 8 3
7 32 5
8 256 8
1 2 3 4 5 6 7 8 9 10
Dilution log2
Fig. 15. Rules for reading the result of the hemagglutination inhibition test
µl 25 µl 25 µl 25 µl 25 µl 25 µl 25 µl 25
25
25
25
1 8 3 Neg.
2 8 3 Neg.
3 16 4 Pos.
4 8 3 Neg.
5 16 4 Pos.
6 16 4 Pos.
7 8 3 Neg.
8 16 4 Pos.
9 8 3 Neg.
10 8 3 Neg.
11 8 3 Neg.
12 8 3 Neg.
13 32 5 Pos.
14 8 3 Neg.
15 8 3 Neg.
16 8 3 Neg.
17 8 3 Neg.
18 8 3 Neg.
19 8 3 Neg.
20 8 3 Neg.
21 8 3 Neg.
22 8 3 Neg.
23 8 3 Neg.
Statistical summary:
Mean: log23.3 Min.: log23
Pos/Neg: 5/18 Interpretation of the test result:
Max.: log25 Positive sample: log2>3
4
log2 titer
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Serum number
20
18
16
14
Number of samples
12
10
0
Pos. Neg.
2 1 No immunity.
Lack of protection
Negative test result
- in case of infection
4 2 possible clinical signs
8 3
Poor immunity.
Fisrt vaccination
16 4 Partial protection
with live vaccines
- mild symptoms
32 5
64 6
128 7
Protection against
Revaccination with
mortality.
live vaccines
No clinical symptoms
256 8
512 9
1024 10
Live and inactivated vaccines
– intensive program
2048 11
Field virus
infection - disease
4096 12
8192 13
Tab. 8. General indications for interpreting the results obtained in the HI test.
+ + + + + + + + + + + CREATING CLARITY