Contact Us

Use the form on the right to contact us.

You can edit the text in this area, and change where the contact form on the right submits to, by entering edit mode using the modes on the bottom right. 

           

123 Street Avenue, City Town, 99999

(123) 555-6789

email@address.com

 

You can set your address, phone number, email and site description in the settings tab.
Link to read me page with more information.

Impish Iggies

HOW HEALTH TESTING CAN BE MISLEADING

Tatum Clinton Miller

- By Tatum Clinton Miller

This is a response to a question I was asked by that I decided to make its own post for organizational reasons. The short version of the original question is:

Why would a breeder choose to not test for some joint clearances (such as shoulder and patella)? Is it a lack of reliable tests issue, money and costs, age? […]  Apparently in GSMDs clearing shoulders and patellas isn’t a standard practice. Why would this be? 

There’s a couple of things to be addressed here. I’m going to use Italian Greyhounds as an example a lot, since they’re the breed which I’m most familiar with and they have a lot of lessons to teach us about the usage and value of health testing.

1) Though you can test a dog for literally hundreds of diseases and conditions, there’s no point in testing the dog for conditions which are very rarely or never found in the breed. It just doesn’t make sense to waste money on a test which you know is going to come back with good results. However, some breeders will still do this because it’s an easy way of making their dogs looking healthier than they really are.

For example, Italian Greyhounds are at the bottom of the OFA’s rankings for hip dysplasia, meaning they’re the least affected breed. HD has literally never been reported in IGs according to the OFA data (luxating patellas is their main joint issue, though they are still more toward the middle of the stats as far as affectedness goes). Do some IG breeders still test hips? You bet they do! That way, when the results come in as excellent, which they certainly will, the breeder can then use those results as proof of how awesome and healthy their dogs are. Of course, what they don’t mention is that you could probably go out to any puppy mill, grab a handful of random IGs, test them for HD and get the same results as the dogs from the look-at-me-I’m-so-responsible breeder.

It’s the same deal with the MyDogDNA test. It’s a relatively affordable DNA test which screens every dog tested for 135 genetic diseases, regardless of breed. The IG has only one breed-specific test available in their database, a condition called amelogenesis imperfecta or enamel hypoplasia which causes defects in tooth enamel and therefore facilitates periodontal disease (which is a huge problem in IGs regardless of whether they’re affected by AI or not) and its severity is ranked as mild in the MDDNA databse. However, they also get tested for 134 diseases like every other dog, and as expected, have an extremely high likelihood to come back as “clear” for all those diseases. Again, this is an easy way to make a breeder’s dogs look amazingly healthy because now they can say “Look, our dogs were tested clear for a whole 134 genetic disorders!” which really means nothing. Even worse, IGs do have a whole load of inherited diseases; they just don’t have tests for them. Quoting UC Davis’ research on genetic diversity in IGs:

The IG has the greatest variety and incidences of potentially heritable diseases than any breed that we have studied.

Yikes!

This phenomenon is a lot like an anti-vaxxer mom proudly proclaiming, “Look how healthy my child is, she hasn’t had any negative effects from injections from those awful chemicals they call vaccines!” Congratulations, your child which has never been vaccinated has had zero adverse reactions to vaccines, what an accomplishment!

2) Another thing to consider is that phenotypic tests (like joint testing) are NOT predictive tests. This means that two dogs can have excellent hips and still produce dogs with poor hips. Phenotypic testing only tells you about the health of the dog tested, and does not guarantee the health of their offspring. Hip dysplasia in particular is looking to be less and less heritable (passed down via genetics) and more caused by environmental factors according to recent research, with heritabality estimates ranging between 15-40% according to various studies. Additionally, one recent study found that using Estimated Breeding Values (EBVs) was much more effective at predicting the health of a dog’s offspring than phenotypic tests alone. HD, orthopedic disorders in general, and numerous genetic diseases are polygenic disorders, meaning they are caused by the interaction of more than one gene, so accurate predictive testing is not available for them like it is for a number of monogenic disorders like von Willebrand’s disease or MDR1. Phenotypic tests are still valuable for assessing the overallhealth of a dog to formulate an EBV, but should not be taken as a guarantee of healthy offspring in and of themselves.

3) This is related more to health testing in general than the specific tests you mentioned, but I’m including it anyway since it’s relevant to your search for a good breeder. There are some health tests on the market which do not actually definitively mean anything for the health of the dog. One example is the dilated cardiomyopathy (DCM) test for Dobermans, which only identifies one of the many genes responsible for DCM, meaning it’s possible for a dog to develop DCM regardless of their test results. It’s more or less a risk factor for developing the disease than anything.

You’re probably wondering how it’s even possible for breeders to actually want to buy these pointless tests, so I’ll explain using Italian Greyhounds as an example.

One thing in particular that IG breeders will never tell you is that the biggest problem for the breed, immune disease, has NO genetic test. Quoting the UC Davis research on IG genetic diversity:

IGs have one of the highest incidences of autoimmune disease and it manifests in virtually every clinical form known to humans or dogs. Clinical forms of autoimmunity in the breed include Addison’s disease (hypoadrenocorticoism), immune mediated thyroiditis (hypothyroidism),  immune mediated polyarthritis and meningitis, systemic lupus erythematous, systemic vasculitis, panniculitis, masticatory myositis,  autoimmune hemolytic anemia, autoimmune thrombocytopenia, lupoid onychodystrophy, pemphigus, rheumatoid arthritis, and immune mediated orchitis (male sterility).   

There’s a study that was published in 2012 titled Multiple autoimmune diseases syndrome in Italian Greyhounds: preliminary studies of genome-wide diversity and possible associations within the dog leukocyte antigen (DLA) complex which looked at autoimmune disease in IGs and tried to determine its pathology and epidemiology in the breed. The study concluded that AI disease in IGs is likely not the result of inheriting many different AI diseases separately but rather the inheritance of an AI disease complex, meaning dogs inherit genes which predispose them to AI diseases in general, not just a specific one or two, which is why IGs that suffer from AI disease will often suffer from 2-3 AI diseases within their lifetime instead of just one. The study also found that about 9% of IGs are affected by this AI disease complex, that is about one in ten. This should be truly alarming, but you rarely hear IG breeders discuss it, though they will proudly display the excellent results of all other health tests which they’ve conducted on their dogs, whether they are actually meaningful or not.

So what are IG breeders doing about this looming problem of AI disease? Well, it’s known that a predisposition to AI disease is often caused by a lack of diversity (heterozygosity) in the MHC or DLA complex in dogs, so the obvious thing for IG breeders to do is invest in developing a genetic diversity test.

And lo, thus the UC Davis Italian Greyhound Genetic Diversity Test was born. Unsurprisingly, the diversity and distribution of DLA haplotypes in IGs is pretty bad (I actually JUST wrote about this last night). The research done in creating this test found that there was a genetic distinction between American and European IG populations, with slightly different DLA distributions. This announcement made IG breeders jump for joy because it seemingly offered hope for the IG’s problem of low genetic diversity without crossbreeding, which is highly taboo and considered a last resort by these breeders. All they had to do was do this new diversity testing, outcross with dogs from other countries whenever possible, and IG immune health will improve. Right? Well…

The following paragraph is an exact quote from the conclusion of UC Davis’ IG research:

Will better selection of mates increase genetic diversity and lower IR scores and significantly reduce the incidence of these major health problems? We believe that it will, but the proof will come when and if it is widely applied. The number and complexity of heritable disorders in the breed will not make it easy to find causes, develop tests, and eliminate bad genes. Therefore, the best immediate solution is to dilute out these undesirable traits making it less likely to mate dogs that carry them. There is still considerable genetic diversity in both US and European IGs to allow for a significant reapportionment of existing diversity. The genetic difference between European and US dogs also offers the additional option of “outcrossing.” However, the reluctance to mix these two lineages of dogs has to be overcome.

I bolded that one sentence because its wording is extremely important. The UC Davis researchers subtly say theydo not know whether their own genetic diversity testing will increase IG genetic diversity enough to “significantly reduce the incidence of these major health problems,” just that they “believe that it will.” Believing something is true is not good enough to make it fact, which is why we do scientific research to examine data from an objective perspective to determine what is verifiably true or false.

The Institute of Canine Biology posted an article on this topic of “belief” versus emipirical fact a couple days ago. Here’s a good quote:

hen I posed the question at the top [of this article] in the ICB Breeding for the Future Facebook group, I was surprised to see that many people thought the statement was true and just as many thought it was false. Also, many of the explanations used words or phrases like “believe”, “think”, “this makes sense”, “it’s logical that…”, and so on, and there were also examples from personal experience to support an answer. […]  The whole point of science is to figure out how the world works through observation and experimentation, and one thing we all know (or should) is that there are many things we might think we “know” from limited personal experience that turn out not to be true. […]  A trained scientist should be keenly aware of the difference between what they think might be true about something based on experience, a hunch, or a few observations (and they might call it a hypothesis), and what is true (or accepted to be true) based on rigorous science. 

Even coming from trained geneticists, saying “we believe that” is NOT good enough. Though UC Davis does not explicitly claim that their IG diversity test will actually facilitate the improvement of IG health, pretty much every IG breeder out there does.

Okay, so maybe you can accept that genetic diversity testing may not definitively improve IG immune health, but it’s worth a try, right? Hold on a minute! There’s a big problem with this: overlooking existing evidence that contradicts this idea by way of confirmation bias. There is in fact empirical evidence which contradicts the idea that improved heteorzygosity within IGs (whether by genetic diversity testing or other means) truly reduces the risk for immune disease in particular. Let’s look back at the first study I referenced about the AI disease complex within IGs:

A panel of 24 highly polymorphic simple tandem repeat (STR) markers across 20 autosomes demonstrated that affected and unaffected dogs were not distinguishable from the population as a whole by heterozygosity, F-statistics, and principal component analysis (PCA). However, analysis of allele frequencies at each STR loci identified regions of increased or decreased disease risk on four chromosomes. A similar genetic analysis using 109 single nucleotide polymorphisms (SNPs) across the DLA region showed differences between affected and unaffected dogs. PCA and zygosity mapping of DLA SNPs from unrelated dogs demonstrated two distinct subpopulations among the affected individuals. One population was very homozygous and the other closely resembled unaffected dogs in its heterozygosity, suggesting the evolution of a disease prone bloodline as a result of non-random selection.

If you’re not familiar with the terminology, that paragraph probably sounds like gibberish, so here’s my simplified paraphrasing: the study found no difference in both overall genetic diversity (heterozygosity) AND diversity specifically in the DLA region (the part of the genome which dictates the immune system) between sick and healthy IGs; furthermore, though some of the sick IGs had low genetic diversity (homozygosity) in the MHC (DLA region), some sick IGs had the same level of genetic diversity in the MHC as healthy IGs!

This could mean two things:

1) There is no association between heterozygosity, especially in the MHC, and immune health. We know this isn’t true through other scientific studies. It’s very basic knowledge in immunogenetics that heterozygosity in the MHC is protective against AI disease. In the chapter on canine immunogenetics in The Genetics of the Dog ed. Ostrander & Ruvinsky, this information is in the very beginning of the chapter as foundational knowledge for the field.

2) The overall level of homozygosity of IGs as a population is already so high that it’s impossible to genetically distinguish IGs which suffer from AI disease and those that don’t. Since we know that homozygosity in the MHC especially is associated with AI disease, why then does it appear that there’s no statistically significant difference in homozygosity between IGs with AI disease and those without it? There are a few possibilities:

a) There are one or more currently unidentified genes that also affect the probability of developing AI disease in IGs, which may not be found in the MHC itself (this study specifically looked for associations between DLA haplotypes and AI disease and found no definite linkages).

b) AI disease is an epigenetic disorder, meaning there are genes that are causative but environmental factors determine how those genes are expressed. An animal could carry all the “wrong” genes and still not develop AI disease if not exposed to an environmental factor that “triggers” the expression of those genes causing AI disease.

c) A combination of A and B. This seems most likely to me, but of course scientific data is lacking so there is no way to definitively prove it works one way or another.

In any case, the conclusion should be clear: there is NO evidence that increased heterozygosity within IGs is associated with a decrease in the prevalence of AI disease. Even more damning, what data we do have have indicates that the opposite is true: there IS evidence that higher heterozygosity has no impact on the prevalence of AI disease in IGs. This essentially means that the UC Davis IG genetic diversity test is, as far as we know empirically, pointless as far as improving IG AI health.

How could the UC Davis researchers have missed the data in this study? It’s doubtful that they overlooked it, as there are only a tiny number of research studies done on IGs specifically, and you’d think any trained scientist would read through all the research available to guide their own study. What seems more likely is that this information was intentionally not referenced because it might damage the validity of their claim that improved genetic diversity within IGs (assuming the breed is a closed gene pool) will decrease disease, specifically AI disease which is the breed’s major problem. If this claim is invalid, then the idea that the test has any real health benefit for the breed is also invalid. If the test has no real health benefit, then fewer breeders will buy it. You can see where this is going.

The end result is that the researchers behind the test have to use wishy-washy language like “believe” when addressing the idea that the test will improve IG health. It’s also important to keep in mind that in order to improve the health of the IG as a breed assuming the test even provides an effective means of doing this in the first place, it is necessary that most or all IG breeders test their dogs, actually understand what the results mean, and plan all of their breedings by appropriately applying the information gained from testing. Realistically, the chances of all of these things actually happening are pretty damn slim.

The truth is that IG breeders, like most breeders, want a test to fix everything. If there’s a test for everything, then theoretically every genetic defect could be either managed or “bred out” while maintaining the status quo (this turns out to be an unfounded belief, even if it were possible to have a genetic test for everything, which it’s not). If there’s a problem there has to be a test for it, because how are supposed to fix it otherwise? Hmmm. Maybe doing a little more research into population genetics and thinking outside the box of the responsible breeder checklist of required health tests would help. I’m in the process of searching for IG breeders wanting to do just that, with minor luck unfortunately.

You might read everything I’ve written above and conclude that I think health-testing is stupid or pointless, but that’s NOT the case. Knowledge is always preferable to ignorance, and health-testing is a good tool for obtaining a little extra knowledge on any particular dog. I merely think that health tests are overrated and overemphasized, and can be quite misleading about a dog’s health depending on how they’re used. A breeder could test their dogs for everything under the sun and still breed unhealthy dogs. Likewise, a breeder could not test for anything at all and yet still produce dogs that are largely free from health problems (this is more common with working breeders). When health tests are performed simply to tick all the right boxes on the “responsible breeder checklist” but the breeder in question does not utilize smart breeding practices, health testing is of little practical value.

Instead of just looking at health testing when breeder shopping, I would also ask a breeder what kind of problems they’ve encountered in their lines and how common they are. If their answer is “I’ve never had any health problems in my dogs or the dogs I’ve bred,” then run away fast because they are lying. Ask what criteria they used to choose to breed two particular dogs for their planned litter(s). Ask if they use EBVs; more scientifically literate breeders are more likely to do this, and scientific literacy is definitely a desirable feature in dog breeders. If they don’t do any particular health tests which are recommended for their breed, ask them why not. Overall, it’s more helpful to try to see the bigger picture of who the breeder is and what they breed for, rather than getting caught up in minutiae.