EP1248951A2

EP1248951A2 - Diagnostic method involving haemoglobin, haematocrit or red blood cell count

Info

Publication number: EP1248951A2
Application number: EP01901248A
Authority: EP
Inventors: Toby Grahame Knowles
Original assignee: University of Bristol
Current assignee: University of Bristol
Priority date: 2000-01-17
Filing date: 2001-01-17
Publication date: 2002-10-16
Also published as: AU771538B2; AU2690801A; WO2001053830A2; GB0001058D0; WO2001053830A3

Abstract

A method of predicting the ability of an individual animal to resist disease particularly transmissible disease at an early stage of its life, by measuring levels of haematocrit or red blood cell count.

Description

Diagnostic Method

The present invention relates to a method of predicting the growth potential of an animal and/or the ability of an animal to resist disease, particularly transmissible diseases. The methodology is simple to apply, results are almost instantaneous and the procedure is non-traumatic.

The ability to predict disease or growth rate in young stock is of use in tailoring a system of husbandry to best economic advantage. There are numerous examples; First, where an animal is likely to succumb to disease it would be possible to provide routine prophylactic treatment to avoid the damaging growth and economic effects of disease. Second, where a small number of cattle are selected for optimum production from amongst a larger population, the ability to predict growth rate would allow more objective selection. This is the case with heifers selected to replace dairy cows culled from the dairy herd. The lifetime milk yield of a cow has been shown to be strongly influenced by growth rate as a heifer. Third, in rearing cattle for beef, different methods of feeding are utilised, dependent upon the growth potential of the animals. Fourth, being able to predict the disease resistance or growth potential of an animal would allow a better index of monetary value.

The use of blood biochemical, and haematological measures as an aid to diagnosis of disease in animals continues to increase, particularly in animals of commercial value, as the automation and ease with which samples can be analysed improves. Diagnoses can be made, or aided by a comparison of the values or variables measured in a clinical case with those found in the general healthy population. At present the process is relatively crude, i.e. values are considered to be normal if they lie within a range plus or minus two standard deviations from the mean of the healthy population in the case where the values for the variable can be approximated by a Gaussian (normal) distribution or, more generally, a range which contains the central 95% of values found in the healthy population (see Farver, T. B. et al (1997) Academic Press, San Diego), where the values are not well approximated by a Gaussian distribution. Differences in analytical methods and differences between geographically separated populations of animals mean that reference values provided by a local analytical service, which are based upon their own measurements, generally provide more accurate standards, however, reference values published in the literature can also provide a useful guide. Reference ranges for blood biochemical and haematological variables can be found in volumes such as Benjamin, M. M. (1984) Outline of Veterinary Clinical Pathology 3^rd Edition, The Iowa State University Press, Ames, Iowa, Radostits and others (1994) Veterinary Medicine: A Textbook of the Diseases of Cattle, Sheep, Pigs, Goats and Horses. Bailliere Tindall, Kaneko and others (1997) Clinical Biochemistry of Domestic Animals (Fifth Edition). Academic Press, San Diego, Kerr, M. G. (1988) Veterinary Laboratory Medicine: Clinical Biochemistry of Domestic Animals (Fifth Edition). Academic Press, San Diego and ^• Schalm, O.W. (1984) Manual of Bovine Hematology: Anemias/Leukocytes/Testing, Veterinary Practice Publishing Company, Santa Barbara, California. Published reference ranges are usually only given for the adult animal and these can be misleading if applied to young animals because in these animals there are often quite dramatic changes in the baseline values of the variables which are part of the normal process of growth.

Various methods have been described where a blood component is measured to predict the vulnerability to a specific disease in humans (EP-A-0669400; RU 2142639; RU 2138053; RU 2126152; RU 2097763) or immune competency (WO96/14565A1). Shamberer Yu N et al in SU1717039 have described a method of predicting growth potential only in 3 month old cattle by measuring a steroid hormone. In EP-A-0170494 there is described a method of assessing passive immunity in calves by measuring the IgG levels which would have been received from the colostrum (mother's first milk).

The inventors have carried out studies of the reference ranges for blood biochemical and haematological variables in young animals and have defined a surprising correlation between the blood biochemical and haematological measurements obtained and the future growth potential, as relative percentage live weight gain over time, and the ability of the animal to resist disease, in terms of a normal (not raised) white blood cell count. (Knowles T. G. et al (2000), The Veterinary Record, 147, 593-598). Specifically, measurement of red blood cell count (RBC), haematocrit and blood haemoglobin have been shown by the inventors to be useful in such a predictive role at an early stage of an animal's life: The physical measurement of these blood parameters may be performed by any known procedure.

Being able to predict the disease susceptibility of an animal to a transmissible disease from birth is advantageous in that the most healthy offspring can be selected by the farmer. A high resistance to disease can be used to predict better growth potential. The invention is not limited to any particular disease and therefore can be widely used to predict the general future health of an animal offspring. The method is simple to use, robust and more sensitive that those in the prior art.

The invention predicts whether an animal is more susceptible to transmissible diseases, i.e diseases which result in an immune response (as evidenced by a raised white blood cell count and/or neutrophil count). Such diseases are generally a. result of infection by bacteria, mycobacterium, viruses or other biological agents and can take the form of gastrointestinal disease, respiratory disease, infection of any part of the body by a biological agent.

The invention can be used at birth or within the first week of life to predict disease susceptibility in later life.

According to a first aspect of the invention there is provided a method of predicting the ability of an animal to resist disease comprising:

a. taking a blood sample from the animal; and b. measuring the level of one or more variables selected from haematocrit levels, red blood cell count or haemoglobin levels present in the blood sample; and c. using the level of the one or more variables to predict the disease susceptibility of the animal and to categorise the animal as having normal predicted disease susceptibility or higher than normal predicted disease susceptibility. In this way it is possible to use the level of the one or more variables to predict the disease susceptibility of the animal and to categorise the animal as having normal predicted disease susceptibility or higher than normal predicted disease susceptibility.

Preferably, the disease is a transmissible disease for example respiratory disease, enteritis, pneumonias, corona virus, joint-ill, navel-ill, staphlococcal infection, streptococcal infection, or pasteurellosis infection.

Other variables include serum albumin levels, alkaline phosphatase levels, β-hydroxybutyrate levels, plasma cortisol levels, serum creatine kinase levels, creatinine levels, iron levels, plasma fibrihogen levels, serum γ glutamate levels, plasma glucose levels, haptoglobin levels, serum non-esterified fatty acid levels, total protein levels, transferrin levels, triglyceride levels, urea, γ globulin, basophil levels, eosinophil levels, packed cell volume levels, haemoglobin levels, lymphocyte levels, mean cell haemoglobin levels, mean cell haemoglobin concentration, mean cell volume levels, monocyte levels, band neutrophil levels, neutrophil levels, platelet counts, and white blood cell count.

Where one of the variables is red blood cell count (RBC), the measured RBC (in counts of 10¹²/1) can be correlated with the predicted disease susceptibility of an animal using the formula:

score = 17.0367 - (1.6423 x RBC)

wherein a score of less than 0 indicates that the animal has a higher than normal predicted disease susceptibility.

Where one of the variables measured is haematocrit, the measured haematocrit level (as a percentage) can be correlated with the predicted disease susceptibility of an animal using the formula:

score = 12.0008 -(0.2468 x Haematocrit) wherein a score of less than 0 indicates that the animal has a higher than normal predicted disease susceptibility.

Where one of the variables measured is haemoglobin level, the measured haemoglobin levels (in units of g/dl) can be correlated with the predicted disease susceptibility of an animal using the formula:

score = 12.0008 - (0.2468 x Haemoglobin)

In the method according to the invention, the further the score is below 0, the higher the predicted level of disease susceptibility of the animal.

The inventors have found that the higher the predicted level of disease susceptibility, the lower the growth potential of the individual animal.

According to a second aspect of the invention there is provided a diagnostic kit for use in a method of determining the ability of an animal to resist disease comprising means to measure the level of one or more variables present in a blood sample where the variables are selected from haematocrit levels, red blood cell count or haemoglobin levels.

Other variables include serum albumin levels, alkaline phosphatase levels, β-hydroxybutyrate levels, plasma cortisol levels, serum creatine kinase levels, creatinine levels, iron levels, plasma fibrinogen levels, serum γ glutamate levels, plasma glucose levels, haptoglobin levels, serum non-esterified fatty acid levels, total protein levels, transferrin levels, triglyceride levels, urea, γ globulin, basophil levels, eosinophil levels, packed cell volume levels, lymphocyte levels, mean cell haemoglobin levels, mean cell haemoglobin concentration, mean cell volume levels, monocyte levels, band neutrophil levels, neutrophil levels, platelet counts, and white blood cells count. According to a third aspect of the invention there is provided a method of predicting the growth potential of an individual animal, comprising:

a. taking a blood sample from the animal; and b. measuring the level of one or more variables selected from haematocrit levels, red blood cell count or haemoglobin levels; and c. using the level determined in (b) to predict the growth potential of the animal.

The animal may be from a domesticated species. Preferably, the animal is a calf. The animal may be less than one year old. Preferably, the blood sample is taken from birth to within the first three days after birth.

The animal may be a human. The animal is less than one year old. Preferably, the blood sample is taken from birth to within the first three days after birth.

A fourth aspect of the invention provides a diagnostic kit for use in a method of determining the growth potential of an animal comprising means to measure at least one of haematocrit, red blood cell count and/or haemoglobin in a blood sample.

Embodiments of the invention will now be described, by way of example only, and with reference to the accompanying Figures 1 to 5, in which:

Figures 1 (A) to 1 (F) are graphs showing the changes in levels of (A) albumin, (B) ALP,

(C) BHB, (D) CK, (E) cortisol and (F) creatinine in calves from birth to 83 days; and

Figures 2 (A) to 2 (F) are graphs showing the changes in (A) Fe, (B) fibrinogen, (C) GGT,

(D) girth, (E) glucose and (F) haptoglobin in calves from birth to 83 days; and

Figures 3 (A) to 3 (F) are graphs showing the changes in (A) NEFA, (B) total protein, (C) transferin, (D) triglycerides, (E) urea and (F) ZST in calves from birth to 83 days; and Figures 4 (A) to 4 (F) are graphs showing the changes in (A) haematocrit, (B) haemoglobin, (C) lymphocytes, (D) mean cell HB, (E) mean cell HB CN and (F) mean cell volume in calves from birth to 83 days; and

Figures 5 (A) to 5 (E) are graphs showing the changes in (A) monocytes, (B) neutrophils, (C) platelets, (D) RBC and (E) white cells in calves from birth to 83 days.

All figures show measurements taken at birth and at days 1, 3, 6, 9, 13, 20, 27, 41, 55 and 83 of life. Where two lines appear on a graph they show group means for groups classified by high and low white cell counts. The error bars show the standard error of the mean at that point on the graph.

Method

Fourteen calves were blood sampled from birth to 83 days of age. A blood sample was taken within three hours of birth and at 1, 3, 6, 9, 13, 20, 27, 41, 55 and 83 days of age. On each occasion a measurement of girth, from just behind the forelimbs, was also taken. A colostrum (first milk) sample was obtained from the dam as soon as possible post partum, for determination of specific gravity. The specific gravity of the colostrum is related to the amount of immunoglobulin present in the milk. The measurement of ZST in the blood was used to ensure that transfer of the immunogloblins from the milk to the calf had indeed taken place. The calves were kept and sampled on the farms on which they were born, with no alteration of the usual husbandry practices of the farm.

On each occasion a 10ml blood sample was obtained by jugular venipuncture using an uncoated 10ml monovette. For the different analyses, the sample was then split into five parts; approximately 2ml was placed in each of two 2ml EDTA vacutainers (purple top), a 2ml Lith/Hep vacutainer (green top), a 2ml OxF vacutainer (grey top) and a 2ml uncoated vacutainer (red top). Different vacutainer coatings were required as the different assays required that the blood be preserved in different ways. The aim was to mimic, as closely as possible, the treatment that samples would be subjected to when obtained by veterinarians in the field. After collection, the tubes for biochemical analysis (EDTA, OxF and uncoated) were taken to the Langford Veterinary Investigation Centre analysis by the Veterinary Laboratory Agency (VLA). Where samples were obtained at the weekend they were stored at 4°C before delivery to the VLA on the first working day. The tube for haematological analysis (EDTA) was analysed by the Langford haematology laboratory. Where samples were obtained at the weekend, two whole blood slides were prepared immediately and the remainder of the sample was stored at 4°C before analysis on the first working day. The remaining Lith/Hep tube was centrifuged within 60 minutes of collection and the plasma recovered and stored at -20°C for later analysis of cortisol levels.

The samples delivered to the VLA were analysed for levels of albumin, alkaline phosphatase (ALP), β-hydroxybutyrate (BHB), creatine kinase (CK), creatinine, iron (Fe), fibrinogen, γ-glutamyltransferase (GGT), glucose, haptoglobin, non-esterified fatty acid (NEFA), total protein, transferrin, triglycerides, urea and γ globulin (as measured by the zinc sulphate turbidity test, ZST). The blood fraction analyses and the biochemical analyses carried out by the VLA, and for cortisol, are described in Table 1.

Table 1. Methods of collection and analyses of the biochemical variables. All Sigma kits (Sigma-Aldrich Company Ltd, Poole) were used in conjunction with a Technicon RA-2000 random access, automated analyser.

*Catalogue No. 994-75409 Haematological analyses included eight variables: haematocrit, haemoglobin level, mean cell haemoglobin (mean cell HB), mean cell haemoglobin concentration (mean cell HB CN), mean cell volume, numbers of platelets, red blood cells (RBC), and white cells. Samples were run through an automated impedance cell counter (Baker 9000) to obtain the 'eight parameter, complete blood count'. Blood smear slides were stained with Leishmans' stain (1ml stain for 3 minutes, 2ml 6.8 pH buffered, distilled water for 12 minutes, rinsed in buffer and then allowed to air dry). The manual leukocyte differential count was performed by counting and identifying 100 cells using the 'battlement counting technique' (Jain 1986) to produce counts of band neutrophils (bands), basophils, eosinophils, lymphocytes, monocytes and neutrophils. Bands, basophils and eosinophils are mentioned further later as the majority of counts for these variables were zero.

A repeated measures analysis of variance was used to test for changes in the variables over time and applied separately to test for differences between subsets of the calves. The results are generally presented as graphs to give a better appreciation of the changes over time, where they are present. The error bars on graphs show plus and minus one standard error of the mean. Statistical tests were not applied to the counts of bands, basophils and eosinophils.

Results

In all, fourteen calves were sampled from five different dairy farms. At least two calves were sampled per farm. One female calf (calf M) died in an on-farm accident at 5 days of age, thus only 3 samples were obtained. Three calves in total were sampled on this farm. On one farm a total of four calves were sampled, two of which were twins. In the following results the twins are treated as independent sampling units. All calves were dairy/beef crosses: Friesian/Holstein crossed with Aberdeen Angus, Limousin, Charolais, Belgian Blue or Simmental. Five of the calves were male and nine female.

The farmers reported that they separated the calves from their dams at approximately 48 hours after birth and they all reported that they were weaned by 42 days. For five of the dams it was their first parity, for four, their seventh parity, for two, their second parity and for one each, their fourth, sixth and ninth parity. The details of each calf are summarised in Table 2. The specific gravity of the colostrum from first parity dams was lower than that of the other dams (t = 6.16, p<0.001). It is already well documented that first parity heifers produce colostrum with less immunoglobulin content.

Table 2. General details of each calf in survey.

Calf Sex Sire Farm SG of Parity of Group

Colostrum Dam

A F Belgian Blue 1 1.06320 2 HWC

B F Belgian Blue 1 1.07432 7 HWC

C F Limousin 1 1.07174 2 LWC

D M Belgian Blue 2 1.07394 6 HWC

E F Belgian Blue 2 1.06902 7 LWC

F M Simmental 3 1.07222 9 LWC

G* F Charolais 2 1.07028 7 HWC

H* M Charolais 2 1.07028 7 LWC

I M Simmental 3 4 LWC

J F Aberdeen Angus 4 1.05588 LWC

K F Aberdeen Angus 4 1.06056 LWC

L F Aberdeen Angus 5 1.04876 HWC

M F Aberdeen Angus 5 1.05936 LWC

N M Aberdeen Angus 5 1.04702 LWC

During the preliminary visual inspection of the raw data two different groups of calves were apparent from the white cell counts, those with a raised count between days 3 to 27 and those without. All the variables were tested for differences between these two groups and, where these were significant, the data for each group are shown separately on the graphs. The two groups of calves are identified in Table 2 as high white cell count (HWC) and low white cell count (LWC) and there were significant differences in measurements of girth, haptoglobin, haematocrit, haemoglobin, neutrophils, RBC and white cells between the two groups over time. The changes in levels of albumin, ALP, BHB, CK, cortisol and creatinine in the calves over 83 days are shown in Figure 1; changes in Fe, fibrinogen, GGT, girth, glucose and haptoglobin are shown in Figure 2; changes in NEFA, total protein, transferin, triglycerides, urea and ZST are shown in Figure 3; changes in haematocrit, haemoglobin, lymphoctyes, mean cell HB, mean cell HB CN and mean cell volume are shown in Figure 4; and changes in monocytes, neutrophils, platelets, RBC and white cells are shown in Figure 5. There were significant changes over time in all of the variables measured except serum Fe and the monocyte numbers. The changes in counts of bands, basophils and eosinophils are summarised in Table 3.

Table 3. Counts of bands, basophils and eosinophils for each measurement occasion

The mean counts (x 10^"9) are given and the number of animals upon which the mean is based is given below. The figures in brackets show the number of animals which had at least one positive count throughout the survey.

The majority of measurements for these types of cells showed zero counts, although most animals had a non-zero count on at least one occasion. Five of the calves had consistent zero measurements for bands, one calf had consistent zero counts for eosinophils and one calf had consistent zero counts for basophils. Table 3 shows the number of animals for which there was a non-zero count and the means of the non-zero counts for each measurement occasion. Visual inspection of the data for bands, basophils and eosinophils showed a trend for raised counts of bands in the HWC group for days 3 to 13 and raised eosinophil counts for days 1 and 3. The raised levels tended to be due to one-off high counts for individual calves within the group rather than consistently raised levels within the group.

Discussion

The physiological reasons for the changes in the variables during calf growth are discussed in general terms in texts such as Kaneko, J. et al (1997) Clinical Biochemistry of Domestic Animals (Fifth Edition) Academic Press, San Diego, Jain, N. C. (1986) Schalm's Veterinary Hematology (4^th Edition) Lea and Febiger, Philadelphia and Schalm supra (1984). Essentially, these changes take place as the neonate starts to function as an autonomous creature, no longer dependent upon its dam for oxygen and other nutrients, and as its own independent metabolism begins to function. Red blood cells change as the animal becomes adapted to collect oxygen from its lungs rather than across the placenta. The inventors have provided a more detailed and complete map of the changes in blood variables than was currently available.

Disregarding band neutrophil, basophil and eosinophil numbers, there were marked and significant changes over time in twenty-seven of the twenty-nine variables measured in the survey. It is likely that the trends seen in serum Fe and monocyte numbers over time were real but that the changes were small compared with the natural variability of these measures, thus the changes were not statistically significant. The differences seen between calves in the levels of white cells was not unexpected and the raised levels in the HWC group are normally attributed to some degree of disease challenge.

Of interest are the differences between the HWC and LWC groups in terms of other variables; especially so because the differences between the groups in some of the variables are apparent at birth and these predict the immune response to disease for example by the raised WBC count. Of the seven variables which were different between the HWC and LWC groups, haematocrit, haemoglobin and RBC were different at birth (t = 2.55, p = 0.025; t = 2.60, p = 0.023, t = 3.28, p = 0.007, respectively) and indicated intrinsic differences between the two groups. These differences are large and not obviously correlated with farm, breed and parity. The levels of WBC and neutrophils were not different at birth.

The counts of bands, basophils and eosinophils in each calf over time were sporadic and rather variable (see Table 3). This was the case within both the HWC and LWC groups, however, there was some indication of raised levels of bands and eosinophils in the HWC group. Bands and basophils are often completely absent in measurements of the peripheral blood of healthly cattle (Jain, N. C. supra 1986, pl95).

The specific gravity of the colostrum from first parity dams was lower than that of the other dams. This has already been widely reported in the literature e.g. in Pritchett, L. et al (1991) Journal of Dairy Science 74: 2336-2341.

Table 4 shows the normal ranges of the biochemical variables, for cattle, issued by the VLA and the normal ranges of the haematological variables given in Jain, N. C. supra (1984) and Radostits, O. M., et al (1994) Veterinary Medicine: A Textbook of the Diseases of Cattle, Sheep, Pigs, Goats and Horses. Bailliere Tindall. ISBN: 070201592X. These can be compared with the results obtained from this study as an indicator of where values in healthy calves are likely to deviate from the published normal values. Where possible the values for the normal range are indicated in Figures 1 to 5 by a horizontal line. Where the normal value is/are off scale an arrow head indicates the direction of the value(s). There are no normal ranges shown on the graphs of cortisol, girth or transferrin.

Table 4. Published normal blood values in cattle

* No estimates available

Haptoglobin

Mean haptoglobin levels in the calves were within the reference values after 9 days of age. Before 9 days some of the initial values were above the upper reference value but between-calf variation was quite large.

NEFA

Immediately at birth, levels of NEFA were above the upper reference value but rapidly fell to remain fairly centrally positioned within the reference range for cattle. Total Protein

At birth, levels of total protein were approximately 10 g/1 below the lower reference value. Levels rose quickly after birth, probably as a result of colostrum intake, to within the reference range but. fell below during days 13 to 55.

Transferrin

No published reference values were available for transferrin. Overall, there was a fall in percentage saturation from approximately 25 to 15 per cent from birth to day 3, after which there was a roughly linear increase in saturation to approximately 30 per cent by 83 days of age.

Triglycerides

Triglyceride levels in the calves were always within the reference values for cattle. Levels peaked at birth at 0.4 to 0.5 mmol/1 but after day 3 remained roughly in the range 0.2 to 0.3 mmol/1.

Urea

Urea levels in the calves were always within the reference values for cattle but showed a distinctive pattern of change. There was a rapid fall in levels from birth to 6 days. After weaning there was a trend for levels to increase linearly up to 83 days, the end of the sampling period.

ZST

Levels of gamma globulin increased immediately after birth as the calves fed on colostrum. Levels then remained relatively stable at approximately 20 OD units, the generally accepted minimum level for a healthy calf.

Haematocrit and haemoglobin

Values for haematocrit and haemoglobin were high at birth compared with the upper reference values, but declined rapidly over the first 3 days to within the reference ranges.

Lymphocytes, Mean cell HB, Mean cell HB CN and monocytes The values of all these variables were within the published reference ranges for cattle but all showed marked patterns of change from birth to 83 days.

Mean cell volume

Mean cell volume declined steadily from birth to below the lower reference value after 13 days. From 55 days to 83 days of age mean cell volume remained at approximately 34 fl.

Neutrophils, RBC and White Cells

Counts of these cells roughly paralleled each other and tended to be above the upper reference values for the first day of life. Between days 6 to 20 the counts in the HWC group were above the upper reference value.

Platelets

From birth to 3 days of age the platelet count was within the reference range but it had increased rapidly by day 6 to the upper reference value of 800. The count remained above the reference range from day 6 onward.

Overall, it can be seen that the use of normal reference values for adult cattle could be misleading when applied to neonatal calves. When interpreting blood biochemistry and haematology results for young cattle it is important to consider the levels and patterns of change of the variables that are shown in Figures 1 to 5.

The error bars shown on the Figures can be used to give the 'normal reference range' at a particular time, as the standard error of the mean is equal to the standard deviation divided by the square root of the number animals in a sample. As mentioned earlier, normal reference ranges can be based on plus or minus two standard deviations about the mean.

Predicting Disease Susceptability or Reduced Growth

Using the blood variables measured at birth, a number of commonly employed methods are available for classifying the animals into those which are likely to be susceptible to disease or to reduced growth rate. A discriminant analysis using RBC as the predictor variable and using prior probabilities of group membership, based on the number of animals found in each group, leads to the correct classification of 13 of the 14 animals. Logistic regression is another method for producing a classification rule which can be further enhanced by applying the prior probability of group membership. Based on the data available at present, the logistic regression equations using the variables RBC, haematocrit and haemoglobin, individually to classify animals are given in Table 5 below. The prior probability of disease susceptibility, based on the data from the survey, was 0.357 (i.e. 5 out of 14 calves sampled) and of non-susceptibility, 0.643 (i.e. 9 out of the 14 calves sampled).

Therefore, by measuring the RBC of a calf and using the data obtained in the equation shown in Table 5, a resultant score of less than 0 will indicate that the calf falls into the disease susceptible group. In addition, the lower the score is below 0, the greater the susceptibility of the calf to disease.

Table 5 Estimated logistic regression equations for assigning group membership using the blood variables RBC, haematocrit and haemoglobin, individually. The cut-off point is 0. Thus a score of less than 0 indicates membership of the disease susceptible group.

Tables 9 to 14 show the raw data of measurements of RBC, haematocrit, haemoglobin, girth (a measurement of growth), white cell count (WCC) and neutrophil count (NC) (measurements of immune reaction), respectively, from calves {Bos torus) A to N, at birth, and on days 1, 3, 6, 9, 13, 20, 27, 41, 55 and 83 of life.

The data was analysed using the statistics package SPSS (Release 10.0.5) (SPSS Inc.) and the relevant parts of the output are reproduced herein. The animals were initially separated into two groups based on the changes in white blood cell count (WBC) of individual animals (Table 2). The overall differences between the two groups in terms of WBC counts are shown in Figure 5E where one group of calves had elevated WBC counts between 4 to 27 days of age (High white cell count) and the other group did not (Low white cell count). Figures 5D, 4 A, 4B, 2D, 5E and 5B graphically display the data from Tables 9 to 14 broken down by the high and low white cell count classification.

Logistic regression was used to investigate whether the classification could be recreated using variables measured at birth. Table 6 shows the results of three logistic regressions using either RBC, haematocrit or haemoglobin as the predictor of group classification.

The results shown in Table 6 indicate that RBC, haematocrit and haemoglobin can be used to predict group membership. RBC produced the most accurate prediction with 92.9 per cent of cases correctly classified.

There were intrinsic differences between the two groups which make this predictive capacity^' of value. Animals with the high WBC count showed a trend towards reduced growth rate (Table 7 and Figure 2D). Raised levels of white blood cells and raised levels of neutrophils are indicative of a response to a disease challenge. The differences between the two groups in terms of white cell counts and neutrophil counts were significant (Table 8 and Figures 5E and 5B), indicative of greater disease susceptibility in the high white cell count group.

A measurement of the red blood cell count (RBC) of an animal within the first few days of birth can be used to predict growth potential and disease resistance. Animals with counts below 10.38 x 10^/litre are classified as having better growth potential and better disease resistance. The method may be adapted for more accurate use on specific animal populations by means of sampling the population of interest and adjusting the cut-off point for RBC accordingly. Other measurements of blood such as, haematocrit (also known as packed cell volume (PCV)) and blood haemoglobin which are highly correlated with red blood cell count may be used in a similar predictive role. The cut-off points for haematocrit and haemoglobin are 48.6 per cent and 14.5 g/dl, respectively. Measurements below these values predict better growth potential and better disease resistance.

The invention may be implemented using any of the commercially available methods for measuring these parameters in blood.

Table 6 Results of the logistic regression using either RBC, haematocrit or haemoglobin as predictors of group membership. The statistic showing the overall goodness of fit is shown together with the parameter estimates of the model and the percentage of the animals correctly classified into the two groups.

Table 7 The multivariate results of a repeated measures analysis of variance showing a trend for differences in growth rate between the high and low white cell count groups over time (see Figure 2D).

a. Exact statistic

Table 8 Tests of between subject effects from a repeated measures analysis of variance of white cell count and of neutrophil count showing a significant difference in overall levels of count between high and low white cell count groups (see Figures 5E and 5B). White Cell Count

Neutrophil Count

Measure: MEASUREJ Transformed Variable: Avera e

Table 9. Measurements of red blood cell count (RBC) (x 10¹²/litre)

Day

Calf 0 1 3 6 9 13 20 27 41 55 83

A 12.48 10.93 11.65 11.51 12.95 12.71 13.73 12.86 12.59 12.34 11.6

B 11.36 10.88 9.98 9.82 10.99 10.75 10.38 11.41 11.37 11.11 10.9

C 9.25 8.31 7.64 8.77 9.34 9.17 9.15 9.75 9.58 10.13 10.36

D 12.86 10.18 9.32 9.75 10.2 10.19 10.44 10,16 9.95 9.41 9.94

E 10.28 8.46 8.91 9.05 9.45 9.2 8.81 9.09 8.74 10.04 8.29

F 7.96 5.97 5.31 5.71 6.05 6.04 6.19 6.45 6.2 7.22 9.88

G 9.2 7.8 7.92 8.12 8.12 7.78 8.71 7.69 7.99 8.52 9.65

H 8.17 8.23 8.18 8.85 8.27 9.15 9.13 8.52 7.92 9.93 10.4

I 10.25 10.63 9.9 9.88 10.79 11.08 10.96 11.43 11.11 9.3 10.61

J 7.03 8.33 6.84 6.97 5.82 6.58 5.76 5.93 6.5 7.06 9.29

K 6.51 6.74 6.67 6.51 6.54 6.5 6.25 6.22 8.05 7.87 11.5

L 10.57 9.85 9.71 10 9.9 9.39 8.29 8.55 8.06 9.46 10.77

M 9.84 9.33 9.58 N 9.12 8.65 8.53 9.33 9.5 8.18 7.58 7.55 6.63 6.83 6.91

Table 10. Measurements of haematocrit (%).

Day

Calf 0 1 3 6 9 13 20 27 41 55 83

A 56.4 48.2 50.2 48 53.1 50.7 52.9 47.8 43.7 41.5 37.7

B 50.2 47.7 41.6 40 44.5 42.7 39.4 41.6 39.5 37.2 35.4

C 47.7 42 37.2 42.6 43.7 41-4 40 41 37.9 39.4 39.9

D 59.7 45 38.8 40.6 41.8 40.3 39.3 36.8 33.5 30.7 31.1

E 51.1 41.4 41.3 41.1 42.2 40.5 37.3 37.8 34 32.6 30.4

F 32.8 23.9 20.3 21.6 22.4 21.6 21.1 21.1 20.3 25.8 35.6

G 44.1 36.5 36.4 35.9 34.9 32.3 34.7 29.4 28.4 28 31.9

H 39.6 38.4 37.9 39.5 36.3 38.2 36.1 33 29.1 37 38.2

I 45.9 46.7 41.3 40.4 43.8 43.8 41.8 42.2 38.8 31.2 34.5

J 31.2 37.7 28.9 29.3 23.7 25.7 20.8 20.6 20.9 22.2 30.6

K 26 25.9 29.3 27.7 27.1 25.5 23.7 21.5 29.2 28.1 41.4

L 46 40.6 39.7 40.8 39.2 37 33.2 33.3 30.1 34.7 38.1

M 46.4 43.3 43.2

N 38.6 35.5 33.9 36 36.1 30.3 27.6 25.8 21.1 20.8 21

Table 11. Measurements of haemoblobin (g/dl).

Day

Calf 0 1 3 6 9 13 20 27 41 55 83

A 16.9 14.9 16 15.3 16.9 15.9 16.5 15.2 14.2 13.2 12.7

B 15.3 14.1 13 12.1 13.8 13.6 12.5 13.2 12.6 11.9 11.6

C 14.3 13.1 12.1 13.4 13.6 13.3 12.5 12.2 11.4 12.6 12.3

D 17.3 13.6 12 12.7 13 12.6 12.2 11.4 10.6 9.9 10.2

E 15.2 12.5 12.4 12.7 12.8 12.7 11.3 11.6 10.6 10.8 9.7

F 9.8 7 6.3 6.6 7 6.9 6.9 7.1 6.9 8.3 11.2 G 12.7 11.2 11 10.8 10.6 10.5 11 10 9.1 9 10.1

H 11.8 11.9 11.6 11.9 10.7 11.5 10.8 10.2 9.1 11.7 12

I 13.2 13.8 12.5 12.3 13.7 13.8 12.8 13 12.1 9.6 10.8

J 9.3 10.7 8.8 8.9 7.5 8.2 7.1 6.9 7.1 7.6 9.8

K 7.6 7.7 8.8 8.7 8.5 8.2 7.3 7.3 9.4 9.2 13.5

L 14.1 12.6 12.4 12.6 12.4 11.2 10.3 10.1 9.2 10.4 12

M 13.8 12.6 13.2

N 11.7 10.7 10.9 11.3 11.1 9.7 8.8 8.4 7.2 7.6 8.2

Table 12. Measurements of girth (cm).

Day

Calf 0 1 3 6 9 13 20 27 41 55 83

A 80 80 79 82 82.5 83 88 88 92 97.5 112

B 73 72 69.5 74 79 72 80 79.5 84 87 97.5

C 79 78 80 82 83 84 84 86 88 97 103

D 78 77.5 81 77 81.5 80 84 86 90 97.5 97

E 75 75 76 77 77 76.5 77 78 82 89 107

F 80 78 79 79 80 82.5 84 83 89 97 115

G 79 80.5 82 81 82.5 84 86 88 96 101 115

H 84 84 88 84.5 86.5 88 85 89 93 96 104

I 78 78 77 80 84 84.5 87 92 95 110 116

J 77 78.5 80 84 84 87.5 87 90 100 110 117

K 76 75 77 76 78 80 86 91 101 109 110

L 77 80 81.5 83 84 85 87 88 91 100 102

M 82 80 83

N 77 78 79 79 78 83 87 86 94 96 103 Table 13. Measurements of white cell count (x 10⁹/litre).

Day

Calf 0 1 3 6 9 13 20 27 41 55 83

A 19.9 28.2 10.7 12.5 27 16.1 9.9 0.9 7.5 7.5 .8.9

B 16.1 13 4.7 11.6 14.6 19.6 11.8 1 11..22 8.9 10 9.4

C 15.9 17.1 12.9 7.4 13.5 8.1 11 9.9 7.2 8.1 8.6

D 19.2 14.3 8.9 16.6 14.2 11 9.2 8.2 7.3 10.5 8.9

E 13.9 12.6 12 10.1 13.4 14.7 10.6 9.3 16.2 10.3 9.3

F 15.3 11.1 8.5 7.1 10.8 10.7 11.3 12.7 15.7 13.2 15.4

G 13.7 16.3 9.2 7.9 9.2 24.9 17.6 11.2 10.5 7.6 7.8

H 10.4 9.3 9.5 7.9 9.5 8.9 9.2 12.5 9 9.2 7.5

I 15.8 11.9 10.4 9.8 11 10.3 9 0.7 11 10.2 13.8

J 19.5 17.2 11.2 8.6 10.7 11.1 10 8.8 7.2 7.4 10.6

K 12.5 9.7 7.5 7.5 6.6 7.2 7.5 8 7.7 9 12

L 28.6 13.5 10.6 13.1 11.9 16.8 14.3 9.8 9.1 8.8 10.6

M 17.1 9.7 13.5

N 19.2 10 9.2 9.9 10.5 11.9 15.2 8.2 8.7 7.7 8.8

Table 14. Measurements of neutrophil count (x 10⁹/litre)

Day

Calf 0 1 3 6 9 13 20 27 41 55 83

A 16.1 20.9 5.24 7 21.6 8.05 2.48 5.23 1.65 1.58 2.49

B 11.8 8.45 1.69 3.36 5.55 7.45 2.48 2.02 1.6 2.1 2.91

C 11.6 13.9 9.03 2.52 6.75 1.62 4.62 4.46 2.02 2.03 2.06

D 13.2 9.44 4.36 10.6 9.51 5.72 3.96 1.56 0.95 4.41 2.31

E 9.31 6.17 7.68 3.84 4.02 5.59 2.33 2.23 8.59 3.09 2.6

F 11.9 6.88 4.34 1.07 3.24 2.57 3.16 3.68 5.5 3.43 4.47

G 9.32 8.8 4.6 3.24 2.85 16.7 10.2 3.47 3.89 1.52 2.18

H 7.59 6.23 6.08 3.16 5.7 4.72 4.42 5.75 4.95 2.58 1.58

I 11.7 7.14 4.99 3.14 3.63 2.27 1.98 4.39 4.18 1.02 3.31

J 13.7 13.9 5.49 3.18 4.49 5.22 3.6 3.52 1.94 2.89 4.88 K 10.6 6.89 5.7 4.65 4.29 3.17 4.05 2.48 3.16 3.06 3.36

L 22.3 8.24 6.47 5.37 4.64 11.4 5.43 4.8 2.55 2.29 3.29

M 13.9 7.76 9.18

N 11.3 5.3 3.77 5.15 4.2 5.12 6.54 3.12 2.61 2.08 2.73

Claims

1. A method of predicting the ability of an individual animal to resist disease comprising:

a. taking a blood sample from the animal; and b. measuring the level of one or more variables selected from haematocrit levels, red blood cell count or haemoglobin levels present in the blood sample; and c. using the level of the one or more variables to predict the disease susceptibility of the animal and to categorise the animal as having normal predicted disease susceptibility or higher than normal predicted disease susceptibility.

2. A method according to claim 1 wherein the disease is a transmissible disease.

3. A method according to claim 1 or 2 wherein the disease is respiratory disease, enteritis, pneumonias, corona virus, joint-ill, navel-ill, staphlococcal infection, streptococcal infection or pasteurellosis infection.

4. A method according to any preceding claim wherein the measured RBC (in counts of 10^I2/1) is correlated with the predicted disease susceptibility of an animal using the formula:

score = 17.0367 - (1.6423 x RBC)

and wherein a score of less than 0 indicates that the animal has a higher than normal predicted disease susceptibility.

5. A method according to any one of claims 1 to 3 wherein the measured haematocrit level (as a percentage) is correlated with the predicted disease susceptibility of an animal using the formula:

score = 12.0008- (0.2468 x Haematocrit) and wherein a score of less than 0 indicates that the animal has a higher than normal predicted disease susceptibility.

6. A method according to any one of claims 1 to 3 wherein the measured haemoglobin levels (in units of g/dl) is correlated with the predicted disease susceptibility of an animal using the formula:

score = 12.0008 - (0.2468 x Haemoglobin)

7. A method according to any one of claims 4 to 6 wherein the further the score is below 0.5, the higher the predicted level of disease susceptibility of the animal.

8. A method according to any preceding claim wherein the higher the predicted level of disease susceptibility, the lower the growth potential of the individual animal.

9. A method according to any preceding claim wherein the animal is from a domesticated species.

10. A method according to claim 9 wherein the animal is a calf.

11. A method according to any one of claims 1 to 8 wherein the animal is a human.

12. A method according to any one of claims 9 to 11 wherein the animal is less than one year old.

13. A method according to claim 12 wherein the blood sample is taken from birth to within the first three days after birth.

14. A diagnostic kit for use in a method according to any one of claims 1 to 13 comprising means to measure the level of one or more variables present in a blood sample, wherein the variables are selected from haematocrit levels, red blood cell count or haemoglobin levels.

15. A method of predicting the growth potential of an individual animal comprising: a) taking a blood sample from the animal: b) measuring the level of one or more variables selected from haematocrit levels, red blood cell count or haemoglobin levels; and c) using the level determined in (b) to predict the growth potential of the animal.

16. A diagnostic kit for use in a method of determining the growth potential of an animal comprising means to measure the haematocrit level in a blood sample.