As part of my ongoing coverage of the risks and benefits associated with neutering, I wanted to review a recent article on the subject, this one looking specifically at cancer risk and behavioral problems in Vizslas.
Zink, MC. Farhoodly, P. Elser, SE. Ruffini, LD. Gibbons, TA. Riegr, RH. Evaluation of the risk and age of onset of cancer and behavioral disorders in gonadectomized Vizslas. J. Amer Vet Med Assoc. 2014;244(3):309-319.
This study involved an analysis of data collected in 2008 through an online survey of owners of Vizslas in the U.S. and other countries (U.S. owners made up about 87% of the responses, with almost all the others coming from the UK, Australia, and Canada). Information was collected on about 2,500 dogs, and both cancer and behavioral problems were reported in about 25% of these.
The authors looked at the cancers and behavioral problems reported by owners as well as the age when individuals were neutered, if they were, the sex, and the age at which the medical problems examined were reported (though this last figure often had to be guessed at). The authors specifically excluded many conditions from the analysis, including some previously reported to be associated with neuter status. They did not, for example, consider orthopedic diseases because these were uncommon (~9% of the dogs). Oddly, they did not consider skin conditions either, though these were reported in about 20% of the dogs.
The reported results cover several cancers that are especially common in Vizslas, including Mast Cell Tumors (MCT), Hemangiosarcoma (HAS), and Lymphoma (LSA), as well as behavioral problems (noise phobias, separation anxiety, and various forms of aggression). The general results, broken down by age of neutering, are reported in the tables below.
The odds of MCT and LSA were higher for neutered than intact animals. The odds of HSA was higher for neutered females than for intact females, but there was no relationship between neutering and HSA risk for males. The odds of cancers other than these three were also higher for neutered than for intact animals. For all of these cancers, the odds were higher in those neutered after 12 months of age than in those neutered earlier.
For behavioral problems evaluated, the odds of having such a problem were higher in dogs neutered before 6 months of age than in intact dogs. There were no differences in the odds of behavior problems between intact dogs and those neutered after 6 months of age with the exception of storm phobia which was more common in neutered animals overall than in intact animals.
There was no difference in the age at death or the longevity of neutered dogs compared with intact dogs.
Limitations of the Study
The first potential limitation of this study is that the population of dogs included are potentially not representative of the general pet dog population. Only one breed was included, and only 60% were reportedly kept as “primarily a family pet,” with most of the others being used for show or hunting activities. About 23% were reported to have had offspring and almost half (43%) of the dogs were intact, whereas surveys suggest over 80% of the overall owned dog population is neutered. The average age at death was also reported to be 9 years, which seems quite young compared to many other breeds and mixed-breed dogs of similar size. Since there are genetic factors involved in many health conditions, and potentially developmental and environmental factors associated with how dogs are kept, caution must be used in extrapolating results from one population to another.
Another significant limitation of this study is the method of data collection. All data was collected by anonymous online questionnaire, with no attempt to verify the accuracy or validity of these data. Diagnoses of cancer and behavioral problems and assessment of age at neutering were based entirely on reports of owners, sometimes many years after the fact.
This raises a host of concerns. Owners may have reported diagnoses incorrectly, such as misidentifying cancers or reporting benign tumors as “cancer.” Owners may have been more likely to report cancer and/or neutering information if they believed there to be a relationship between the two or if they knew one purpose of the study was to examine such a relationship. Owners also identified cancer and skin conditions as top health concerns, suggesting a population of respondents particularly interested in these conditions, which might have affected their rate of reporting them.
There is also no way to identify what if any differences there were between people who participated in the survey and people who did not, or between the dogs owned by these different groups. It is likely that people who were aware of the survey and motivated to complete it differed in numerous ways from other Vizsla owners and owners of other kinds of dogs, and this again could affect the health conditions reported and the risk factors affecting them.
Another issue is that this study looked at potential risks posed by neutering, but it did not include assessment of most of the potential benefits of this procedure in this population. For females, for ecample, neutering is believed to be protective against mammary cancer (though the evidence is not as strong as commonly supposed), which in some populations is a very common and frequently malignant type of cancer. In this population, mammary cancer was reported in less than 1% of the females in the study, a rate dramatically less than in other populations studied. This suggests either that this population is a much lower risk of mammary cancer than others, in which case the protective effect of neutering might not be meaningful, or that the incidence of this disease was underreported.
Similarly, uterine infections (pyometra) are a common and serious disease in intact females, and these can be completely prevented by neutering. Yet the rate reported in this study was quite low. 22 cases were reported, which would be a rate of about 4% of the ~535 intact females included. Other studies have reported rates of 10-50% depending on age, so either this population has an unusually low rate of this disease, or the incidence was not accurately reported.
Health problems and cost or disruption for owners associated with estrus, reproductive behaviors, or actual reproduction were also not evaluated in this study.
Overall, the study found no difference in the longevity or overall mortality of neutered versus intact dogs. This is in contrast to other studies which suggest neutered animals may live longer on average than intact animals. More importantly, it calls into question the significance of the reported increase in cancer risk in neutered dogs. If neutered dogs are truly at significantly higher risk of often fatal diseases like HAS and LSA, one might expect intact animals to live longer as a results of being less likely to experience these diseases. And if the two groups have roughly the same life expectancy, perhaps there are benefits to neutering not reported here that counterbalance the risks discussed?
This study contributes useful new information to the ongoing process of evaluating the risks and benefits of neutering. It supports information from other studies, in Rottweilers and Golden Retrievers, that suggest neutering may increase the risk of some cancers, such as hemangiosarcoma and lymphosarcoma, in breeds predisposed to develop these diseases.
The study also has a number of significant limitations. The dogs in the study were all of one breed, and they differed in a number of ways from the general pet dog population, so findings in this group may not be applicable to other populations. The data was collected though anonymous questionnaires completed by owners, often years after the events being asked about, and there was no way to confirm the accuracy or validity of these reports. There is also a high risk that the people who chose to participate in the survey, and the dogs they own, are quite different from the general pet owning population and their pets, in their concerns, knowledge, and pet care practices.
The study did not examine many of the risks posed by being intact, which have to be considered in weighing the overall risks and benefits of neutering. Rates of pyometra and mammary cancer, common and serious medical problems prevented by neutering females, were far lower in this study than generally reported elsewhere, suggesting either that the study population was quite different from other dog populations or that the rates of these diseases were not accurately reported.
And it is unclear how significant the reported increase in the risk of cancers in neutered animals really is since there was no overall difference in the longevity of neutered and intact animals. If neutered animals are much more likely to get cancer, it is surprising that they tended on average to live just as long as intact dogs.
Overall, this study supports the current trend towards questioning the dogma of routine neutering for all dogs. The risks and benefits are likely to vary according to breed, age, and many other variables, and a one-size-fits-all approach is not ideal. Unfortunately, a great deal of additional research will need to be done for dog owners and veterinarians to have confidence in specific recommendations for individual dogs.
Pingback: Benefits and Risks of Neutering–An Evidence-Based Approach | The SkeptVet Blog
http://guardianlv.com/2014/02/single-injection-dog-sterilization-zeuterin-available-starting-monday/ how’s this looking?
The Zn injections have been around in variou forms for a while, but they haven’t caught on. I don’t know much about them, but I have heard concerns about consistency of sterilization, pain on injection, and occasional necrosis. I can see a roole for such a thing.
I have used it on my own pets when it was sold under another brand name. One advantage or disadvantage is that the testicle still produces a reduced amount of hormone. I will be interesting what the company uses to promote a possible benificial effect of leaving 50% hormone production.
Let me start here. First the study was designed and partially funded by Viszla owners to be a study about their breed. Which for those that participated in the survey would I would think tend to make them more attuned to give correct answers. If their memory is faulty they have their medical records to rely upon. Most dog owners do not forget the date of death of their dog, nor if indeed they were told if that beloved dog had cancer, and when they were told this. And indeed if they were “incentivized” it would be to give accurate information. After all who is going to be funding future studies????
An on line survey can contain with-in certain cross checks to help validate the information.
The anonymous part as described might just well be, that the contributors were told that their contributions were anonymous.
To bring up the fact that this study can’t be necessarily attributed to the larger dog population because it was designed and directed to only one breed, well hey maybe yes maybe not, but the study was designed to look at Viszlas.
cancer is a big issue in dogs, and not just for the Viszla breed but for many breeds.
Now this study showed some surprising results on pyrometra and mammary tumours, but is also surprising to me, in not only this study, but in other recent studies , why is there not any reference to the “hallmark” studies, that of “course” must have been done, before the veterinary medical community decided that 6mths old across all breeds is the optimal time to spay or neuter for the “HEALTH”
I don’t know why the study designers and funders decided to only look at a few things that were negative, and not to what is the benefits.. but gosh hey here is a thought mayhap the rates of cancer;skin,and other problems in Viszlas was what they wanted to look at.
So it remains in this and many other studies conclusions including your analysis, that one as a dog owner need to truly discuss the risks and benefits for their dog of this breed and the timing of same for s/n. And it behooves the vet community to come up with and train for alternatives to OVE and Castration, to truly give pet owners a “choice” in this decision
And one other comment where is the learned discourse, on what studies can be attributed to the larger dog population as a whole? What is the criteria, how was it developed, based upon what assumptions?
The problem with recall bias and with misclassification of different kinds in studies that use similar methods is a well-recognized source of error. It is not something unique to this study, nor does it automatically invalidate the results. However, it is an uncontrolled source of bias that potentially reduces the internal and external validity of the trial. Similarly, the limitation to one breed clearly does reduce the extent to which we can generalize the results to other breeds or mixed breeds. Again, that doesn’t mean the results aren’t worth considering, but it does mean we cannot reasonably use the results of this one trial to dramatically alter our understanding of the risks and benefits of neutering or our specific practices. It is one bit of evidence out of hundreds of studies with its own strengths and weaknesses, and it must be evaluated in that context.
Automatic neutering of all dogs of both sexes at 6 months of age is absolutely an arbitrary practice with little in the way of evidentiary support. The trick is that when we reject such an arbitrary, non-evidence-based practice we be careful not to replace it with another.
I have this very kind of complex, nuanced discussion with owners every day, and I think more and more vets are doing so. Unfortunately, both vets and owners tend to want simply, broadly applicable general rules, and biology is just too complex for that to be a reliable strategy.
I’m afraid I don’t understand your question.
My question was simple. In any given study/research design and or later on the review of said study/research; what are the criteria needed to be met, to insure this study could be attributed across all breeds of dogs? With 80 million or so dogs in the USA, what would be the percentage of the population you would need to sample/test to come to say a probability factor of 95, for any given one disease?
In a very simplistic view of the world, a canine is a canine is a canine. We can (I thought) identify the canine species through DNA anaylsis. ANd yes I do know that there “appears” to be certain specific cancers associated to certain particular breeds, but again cause and effect has not been determined. Not to mention the environmental impact has not yet been studied on cancer in dogs.
Ah, so you’re asking about the generalizability of study results. There is no simple answer, and it is not, I’m afraid, a case of a canine is a canine is a canine. Intensive artificial selection by humans, with severe genetic bottlenecks and founder effects, has led to a fair bit of genetic diversity among domestic dogs with regard to clinically meaningful characteristics. Some breeds are unequivocally at higher risk for specific diseases than others purely due to genetic differences. So caution is always required when extrapolating study results to a population different from the one studied.
As an example, herding breeds have a higher prevalence of a gene (MDR1) which allows the drug ivermectin to enter the brain and cause neurologic symptoms. Most dogs can take quite high doses of this drug with no problems, but selected breeds, such as Australian Shepherds, are very likely to have serious adverse effects. If this were a new drug and clinical trials studying whether it was a safe and effective parasite treatment were done only on Australian Shepherds, the results would clearly show the drug was too dangerous to use and it would not be approved. That, unfortunately, would result in the rejection of a drug which is, in fact, very safe and useful for most other breeds. So we have to recognize that even very small genetic differences make it hard to generalize between different study populations.
Now, it’s true that it would require a pretty large number of dogs to fairly sample all breeds in one study, but that isn’t really necessary. Different studies with different populations conducted over time will converge on a result that is more generalizable than the results of any single study. The important thing here, is that we recognize the limitations of single studies and not race off to make dramatic changes in our practices on the basis of single studies that might or might not reflect the reality for dogs not in the particular study population tested.
And yes, environmental risk factors have been studied in regards to canine cancer, though of course there is a great deal we don’t know. Certain lawn chemicals, for example, have been associated with an increased risk of bladder cancer in Scottish Terriers, who are genetically at much greater risk already for this disease than many other breeds. However, it is not clear that these chemicals present the same increased risk in other breeds, because both environment and genes can interact in setting the risk of a specific disease. So all of these factors have to be considered in a detailed, nuanced way, which is why I caution against over-interpreting the Viszla study or any other single study when evaluating the risks and benefits of neutering.
Thank you for reporting these scientific, impartial studies. I believe we, as a dog owner population have been lied to and manipulated. I also believe that most people, faced with real scientific fact will make the good and right decision for themselves, the community and their dogs..
As a Vizsla owner I found this very helpful. My puppy is now 5 1/2 months old. The hunting trainer and the vet are giving me two different opinions on whether I should, or should not neuter my male. I also am finding different opinions on when to do it, if I so choose. While earning my doctorate I enrolled in many courses on study analysis and potential limitations/strengths. You did a great job breaking down many of them in this study. For me the discussed study is directly applicable. I am concerned that public opinion often takes precedence over actual evidence when it comes to many health decisions. A good example is the use of homeopathic medicine in the USA. You don’t find much evidence supporting it, yet its popularity is growing at exponential rates. I’m not saying it doesn’t have its purpose, or that it doesn’t work. That is a discussion for another time. I am simply using it as a prime example of lacking in valid double blind placebo controlled research to support the claims that are made, yet wildly popular in public opinion. This study did have its limitations, but should not be thrown out as useless either. It is another piece of knowledge to add to my bank for making a decision.
Thanks for the great work SkeptVet. Keep it up!
Yes this is a complex subject (unlike, I will say, homeopathy, which is pretty conclusively a placebo :-). I’m glad my article was useful to you.