Part 1: Your dog has hip dysplasia because your breeder watches too much ER and not enough House.

I am a veterinary radiologist. I sit in a dark room looking at x-rays of dogs and cats all day. Every day. Unfortunately, my job is becoming increasingly frustrating.

Frequently, I am called on to referee between veterinarians, pet owners, and breeders when someone is trying to place blame for a genetic disorder in a pet; or when breeders, pet owners, or police and military units are trying to predict which puppies are going to suffer from heritable diseases in the future.

A veterinary radiologist is called to referee because we are trying to use inappropriate imaging diagnostics, such as x-rays, to predict the future. It seems that breeders and pet owners watched too many episodes of ER. In that TV show, the doctors look at an x-ray and had all the answers from that one test. The truth of the matter is that they should have spent more time watching House. If they did, they would know that imaging tests do not usually have all of the answers and one test leads to more tests as we try to figure out what the first test meant.

In this two part series I will explain why relying on poorly designed imaging tests, and pretending the world is like the TV show ER, has, in part, failed to stomp out genetic disorders such as hip dysplasia.  In order to do that, we need to get our geek on and learn something about imaging diagnostics such as x-rays.

 

Sensitivity and Specificity:

The next 5 minutes of your life are going to be a little painful. Deal with it but don’t give into your nerd rage.  Keep reading. This little bit of background is essential to understanding why your breeder said your dog would not have hip dysplasia and now it does.

Scientists describe the ability of an imaging test, such as x-rays, to predict disease with two terms. These are:

 

  • Sensitivity: The sensitivity of a test refers to how many cases of a disease a particular test can find.
  • Specificity: The specificity of a test refers to how accurately a test diagnoses a particular disease without giving false-positive results.

If a test has a high sensitivity, it will identify all patients with the disease we are testing for.  Detecting disease is great but the problem is that many highly sensitive tests are prone to having false positives. Stated another way, highly sensitive tests can identify every single patient with the disease but a lot of these patients with a positive test result do not really have disease.

What we really need is a test that when it is positive it means the dog has disease. Tests that also satisfy this requirement are also said to be highly specific or have a high specificity for the disease we are testing for.

In the real world, however, tests that have a high specificity for are, in many cases, not that sensitive. Stated another way, if the test is positive you can pretty much guarantee that the patient has the disease we are testing for. The trade off, however, is that there will be patients that do not test positive when they really have the disease (i.e. a false negative.)

As you can see, there is a trade off between sensitivity and specificity with many tests including imaging diagnostics such as x-rays.  Furthermore, in many cases, tests that are both highly sensitive and highly specific (i.e. no false positives and no false negatives) either do not exist or are inconvenient or expensive to run.

Veterinarians, nonetheless, have a job to do and we must use our less than ideal tests to do our job. In order to overcome the limitations of the tests we have, veterinarians have two broad options. These are:

  • Option 1: Run tests that are highly sensitive but lack specificity as screening tests. Screening tests tend to identify all pets with a disease but they often include some accidental positive results. In order to overcome this limitation, we then run a second test with a high specificity to confirm the results the screening tests1.
  • Option 2: Do our best and try to draw meaningful conclusions from tests that are prone to false positives and false negatives.

Take home message #1: Attempting to perform genetic screenings and predict the future health of animals using imaging diagnostics such as x-rays, in the manner that  most pet owners and breeders try to do, falls under option #2.

Stated another way, these tests are not capable (e.g. they are not sensitive enough or specific enough) to allow us to draw the conclusions we try to draw from them and this is one of the reasons why your dog has hip dysplasia and the guarantee your breeder gave you about the future health of your dog might not be as solid as you would hope.

In the second article in this series, we look at some of the tests that breeders and owners use to predict hip dysplasia, why hip dysplasia screening often falls short, and how we can do a better job.

 

NOTES:

1. The problem with this situation is that pet owners get upset that they have to run test after test. They often think the vets does not know what they are doing when they get contradictory results or get upset that the vet is spending their money

About the author: Dr. Matt Wright DVM MS DACVR is a veterinary radiologist who sits in a dark room looking at x-rays all day long. Although he feels that his essays are well thought out, his opinions often fall outside the mainstream. The ideas and concepts in his essays are his and only his and do not necessarily reflect those of his clients, colleagues, or employer. He is the author of several books including Breedfreak – The Dog Breed Guide for Normal People with Real Lives, Families, Houses, and Budgets.