Herd diagnostics and selecting solution for intestinal problems in clinical practice

An important work function for the practicing veterinarian is to solve various problems in swine herds. Diagnostics and making a diagnosis are an important part of solving a problem. 

The diagnosis is central for describing the problem, identifying the problem and, finally, making a change that will hopefully lead to the problem being solved.

A diagnosis can either be made for individual animals or, more frequently in swine herds, for a herd – the so-called herd diagnosis. In this connection, it is important to always remember that herd diagnostics are based on the individual animal when the work is carried out. 

There are different types of diagnosis. A diagnosis can be clinical, pathological, microbiological and/or based on laboratory tests. There are a number of fundamental principles connected to diagnostics including analytical sensitivity and specificity, diagnostic sensitivity and specificity plus positive and negative predictive values, apparent prevalence and socalled true prevalence. The reader can consult ordinary textbooks on epidemiological subjects in respect of these principles. 

Before starting a diagnostic work-up, there are several considerations to be made. For example, how many pigs should be examined, which pigs should be examined and how should they be examined. The first question is what you actually want to achieve – what is the diagnostic purpose? This could e.g. be which infections are causing a cough, is a diarrhoea caused by bacterial infection, how sick are the pigs and what does this cost in respect of productivity, would vaccination be relevant, when would vaccination be relevant or does Lawsonia intracellularis have an impact on e.g. growth in this herd. 

The choice of the diagnostic method depends entirely on which of these areas one wishes to clarify.

 

With regard to sample size calculation, it is the case that we should look at how many pigs should be examined. There are a number of statistical formulas for this, but some biological considerations are also involved. 

The statistical considerations consist of epidemiological principles which can be found in various textbooks, but in the real world this often fails as the costs of laboratory diagnostics or practical issues like time available exceed the statistically calculated number. In that case, one must make the best assessment based on practice, economy and statistics. 

With regard to the biological considerations in respect of the sample size, it is also important whether it is mixed infections or pure infections. Here it is again a question of what one wishes to examine. 

If we look at diarrhoea problems among weaners, we know that mixed infections are a big problem. Here it is therefore necessary to take samples from a quite significant number of pigs in order to uncover which infections are present. On the other hand, if there are problems with an outbreak of pleuropneumonia, then fewer pigs could actually be examined as each sick animal are more likely to represent the problem. 

With regard to selection of pigs to examine, there are various methods that can be used: random, convenient, targeted judgement and cluster sampling. Again, this depends a lot on what one wishes to examine, but in the real world cluster sampling and judgment sampling are very often used. Likewise, convenient sampling is often used because the pigs that have died anyway can be used for necropsy. 

One should of course always be aware whether the pigs that are examined are in fact representative for the problem one wishes to investigate. 

An important thing is also the variation that occurs over time. One should be aware that in respect of pigs we are making diagnostics for the future because we assume that the pigs we examine today are representative for the pigs that will also get diarrhoea or coughing in the future so our findings here will also apply to the pigs in the next batch. 

However, we know that this is far from the case. There are great variations between week batches both in respect of pneumonia (e.g. in respect of lung scores at slaughterhouses), the occurrence of intestinal infections in diarrhoea among weaners and finisher pigs and there are other examples.

 

Finally, there are a number of issues concerning diagnostic tests of which one should be aware. Different methods have different properties in respect of assessing whether an infection is present or not or whether an infection actually has a significance for the disease or pig from which the sample was taken. 

For example, we can say that serology is good in respect of determining whether an infection is present in a herd while quantitative PCR is better if one wishes to examine whether an animal is actually sick and has reduced productivity due to an infection. Methods for detection of L. intracellularis are good examples of these differences. 

There is probably also no denying that histology with detection of agents in histological tissue is the golden standard available for most diseases. 

In the future, a change will probably occur in the way diagnostics are done. Among other things, this includes more regular monitoring by using methods that test for many agents at the same time. The taking of samples will also be e.g. with ropes in the form of saliva samples, sock samples as we know them from diarrhoea or samples from air. 

In addition, there will probably be progress in pen-site diagnostics which will make it possible for us to make a diagnosis on one animal or a group of animals already on the farm. 

 

Examples in respect of intestinal diseases from practice are e.g. that microbiological tests of post-weaning diarrhoea have shown that Escherichia coli is not always the cause, but on the contrary that rotavirus is the cause or that the diarrhoea is feed-related. It would therefore not be relevant to vaccinate against E. coli but on the other hand rather to make a feed-related adjustment. 

In the same way, another example concerns weaners and finisher pigs where detection of L. intracellularis through normal qualitative PCR or serology does not say anything about whether an infection has an impact on the pigs. Here a quantitative PCR test where the number of bacteria is counted would be able to identify whether there is an impact on the pigs’ productivity. In practice, it has turned out that in herds with low excretion numbers of L. intracellularis treatment of or vaccination against L. intracellularis is not relevant. 

Similarly, there are cases of neonatal diarrhoea in newborn piglets, where no agent is detected. This means that it is not a primary, infectious cause but on the other hand e.g. cold or poor milk production of sows. Again examples of how the clinical, pathological and microbiological examination leads to a decision on e.g. not to start up vaccination which would otherwise have been done. 

Likewise, continuous monitoring of e.g. PCV2 by quantitative PCR has shown that vaccination is not necessary in all cases. Based on this, it is therefore possible to identify whether a vaccination should be stopped or be continued. 

 

All the things described above apply to the all types of diagnoses and all the different kind of examinations shall be combined for interpretation of the results. 

At the end, the most important thing is whether the diagnostics made actually make a difference in respect of handling the problem one wishes to solve.