This website will no longer be updated.

Thank you for all of your support! Please visit me at the new URL, - See you there!

Friday, February 19, 2021

New Diet-Associated DCM Study: Feb 2021

     Another retrospective study (Walker et al. 2021) concerning dilated cardiomyopathy in dogs was made available for pre-proof access, meaning that the article has been peer-reviewed and accepted for publication, but will undergo some further editing/formatting for final display in the journal. This research, published in the Journal of Veterinary Cardiology, can be accessed with a subscription or single purchase, but is otherwise not available to the general public. 

    The study itself is somewhat similar to the paper published in December (Freid et al. 2020). It is a retrospective analysis of dogs diagnosed with DCM and CHF (congestive heart failure) at NC State University College of Veterinary Medicine between 1/1/2015 and 07/10/2019. Dogs were excluded if their diet history was unknown. The analysis divided dogs into two groups based on whether their diet was grain-free (GF) or grain-inclusive (GI) at the time of diagnosis. Dogs eating GF at the time of diagnosis were switched to GI as part of the treatment. They're therefore referred to as pGF (prior grain-free). The authors further divided the pGF group based on the length of time a GF diet had been consumed (6 months - 2 years, 2-3 years, and 3+ years). 

    The findings of this study are consistent with previous research, providing further evidence that grain-free diets are associated with a lower mortality, reversible form of DCM, which in turn is associated with improved outcomes following diagnosis. This study does not provide insight to the exact etiology of this observation, but reinforces the need for continued investigation. The authors acknowledge several limitations, including small sample size, and encourage further research. 

Summary of Data:

  • 67 dogs met the inclusion criteria, 43 eating GF and 24 eating GI. 
    • 23 breeds were represented, 19 in the GF group and 11 in the GI group
      • Great Danes, Dobermans, and Labrador Retrievers were most represented in both groups, 18/43 (42%) in GF and 16/24 (67%) in GI. 
    • Age varied significantly between groups, with GF median age of 6 years and GI median age of 9 years
  • 21/43 (49%) GF dogs died and 19/24 (79%) GI dogs died during the study period. 
    • The effect of diet on survival was not statistically significant when considered overall
  • Survival was also assessed in the context of dogs that lived at least 8 days following initial diagnosis, to consider the impact of treatment therapy on dogs that lived to undergo it.
    • For these dogs, previous diet was significantly associated with survival, with pGF dogs that switched diet having better outcomes than dogs eating GI at time of diagnosis
      • Within this group of dogs that survived at least one week, median survival time was 465 days for pGF and 263 days for GI. 
      • Consuming GF diet for a longer time prior to diagnosis was associated with greater mortality (risk of death).
  • On average, pGF dogs reduced medication doses over time while GI dogs increased or maintained medication doses over time. 
  • There was a significantly greater decrease in LVIDdN (diameter of the left ventricle of the heart) over time in the twenty six pGF dogs re-evaluated when compared to the seven GI dogs re-evaluated. 

Additional Resources

Published Research

DCM Resources / Reading

    Sub-group for Veterinary Professionals

General Pet Nutrition Resources / Reading

WSAVA: FAQ, Guide to Nutrition on the Internet (Dogs) (Cats)

Like this article? Follow me on Social Media: @AllTradesDVM

Facebook     Twitter     Instagram


  1. WTF kind of study is this? No controls, 3 breeds that are prone to DCM. We are switching food if they fed GF, but not if they were fed GI, looks like my dogs have a better chance of reducing meds if fed GF, how do we know this is diet associated DCM, did they get taurine supplements, carnitine supplements, the same meds? What ages were the dogs? Who cares - because I guess veterinary research doesn't have to pass any kind of rigorous scientific process at all.

    1. Thanks for your comment!
      In research, there are levels of evidence. For emerging issues, research often works up the pyramid as information is collected and eventually filtered. What you seem to have in mind as a gold standard for 'rigorous scientific process' is a randomized controlled trial, also referred to as RCTs. However, there are two kinds of research that fall below that 'rung' on the pyramid, so to speak. These are cohort studies and case controlled studies. In this case, what we're looking at is a cohort study, defined as "A study which identifies two groups (cohorts) of patients, one which did receive the exposure of interest, and one which did not, and following these cohorts forward for the outcome of interest."
      In this case, the exposure of interest was grain-free diets, and the outcome of interest is progression or reversal of DCM/CHF.

      If you'd like to read the entire study, which includes the rationale for diet change, the medical therapies received by the dogs in question, the ages of the dogs, and more specific information, I recommend purchasing the full article, which I cannot provide here.

    2. This info about their hearts and meds also conflicts with this study - sounds like another case of "today eggs are great for you, and tomorrow they will kill you."

      With the number of veterinary nutritionists that don't believe grain-free to be a problem, the wider study of 67,000 dogs, the FDA being more careful than ever with how it words things, the fact that most of the dogs are still golden retrievers, boxers, dobermans, danes etc. this whole GF=DCM is really losing its footing. But then again- grain free simply means absence of grains. To believe that GF=DCM, you have to believe there is some magical property to grains (that wild dogs/wolves do NOT consume) prevents DCM, despite having a good number of anti-nutrients, no taurine, no carnitine or anything else remarkable of them.

      Additionally, grain-inclusive foods have killed a good number of dogs lately with aflatoxins, a topic that the veterinary community is oddly silent on...

    3. Could you be more specific about what findings you believe are in conflict between these two studies, so that I can more appropriately discuss it without us talking past one another?

      What number of veterinary nutritionists "don't believe grain-free to be a problem"? Most that I've connected with do believe there is an issue. I'd also argue that the opinion of cardiologists is somewhat more important here, since they're the ones diagnosing and treating these patients, and are invariably more familiar with DCM as a disease process. That isn't to undercut the importance of nutritionists assisting with researching this, since they can provide valuable insight into what unique aspects of certain diets may be influencing the development of disease.

      What wider study of 67,000 dogs are you referencing? I'd be happy to review it if you provide a link.

      It's no surprise that the FDA is being "more careful than ever" with how they word things-- they've met substantial political pushback from pulse growers and certain segments of the pet food industry. If you aren't familiar with that pushback, I've written about it before and can refer you to the post.

      I think you have a fundamental misunderstanding of where many clinicians stand on this issue. No one is alleging that grains have some magical anti-DCM property, and very few people are asserting that the absence of grains is the issue (some are, and I would say that they don't have a base to stand on in that). Rather, most are acknowledging that some key factor influencing the atypical development of DCM is significantly more common among grain-free diets than grain-inclusive diets. That feature could be anything, from a greater propensity for poor formulation practices to a higher likelihood of including novel ingredients, and it could be entirely unrelated to the actual absence of grains itself.

      Aflatoxin contamination can be prevented through appropriate quality control, and the veterinary community has not been silent on the recalls or the issue of aflatoxins. The aflatoxin issue is not broadly associated with 'grain-inclusive foods' but rather isolated quality control failures of specific, identifiable manufacturers. If dogs were dying sporadically of an unknown toxin exposure and there were dozens of manufacturers represented and 90% of the diets were grain-inclusive, then it might be more appropriate to broadly associate the issue with grain-inclusive foods.

    4. So you are unaware of the multiple veterinary nutritionists that have spoken out on the issue but say that I have a misunderstanding of the clincical opinions here? You are so condescending I'm done lol. Google the study. It's not hard.

    5. I can't find it. Unless you can provide more details about it (like the title, the journal, or the authors), I have to assume it doesn't actually exist. Please link it, its not hard!


    7. Pretty rude to assume your readers are lying. Maybe you'll learn some humility and tact after you graduate.

    8. I didn't assume you were lying- I said that in the absence of being able to locate it myself or have it cited by someone referencing it, I have to assume it doesn't exist. Not because I think you're a liar, but because I cannot confirm the paper's existence to myself. I asked you (or, potentially, other anonymous posters that also referenced it vaguely) several times to provide it. I'm sorry that you found my response rude or untactful. I appreciate the feedback and will work to avoid that perception moving forward. Personally, I found your demand to "google it," off-putting. I'm glad we can share these feelings openly and improve our communication!

      Thank you for linking what you referenced. It's immediately apparent why a google search didn't readily yield it. This paper is a pre-print that has not been accepted for publication by a journal yet. That doesn't mean it isn't worth talking about, though!

      This study has a number of limitations, and while the concept is one I find worth pursuing, no meaningful conclusions can be drawn from this data alone.
      For this study, authors contacted 36 practices, of which only 14 provided data. Such a sample, even if all practices had responded, is hardly representative and certainly not adequate for describing DCM caseload over time. There are ~300 cardiologists in the US at over 100 unique practices. According to the FDA's presentation at the Kansas State Forum in September of 2020, the geographic distribution of reports seems to correspond with US population centers. Based on the color scheme of the included graph, it appears that states with the most reports include California, Michigan, Ohio, and Massachusetts. Only one of these four states is represented in the present study.

    9. California has over 20 veterinary practices that employ cardiologists, including the U.C. Davis Veterinary Teaching Hospital where Dr. Joshua Stern is leading research on diet-associated DCM. The study in question does not have data from California. Massachusetts has at least 9 veterinary practices with cardiologists, including Cummings School of Veterinary Medicine at Tufts, where Dr. Lisa Freeman is leading research on diet-associated DCM. BSM Partners only collected data from a single practice in Massachusetts (not Tufts), and that practice provided only two years of data. There are at least 5 clinics with cardiologists in Ohio, including the Ohio State University Veterinary Medical Center, where recruitment is ongoing for clinical study on grain-free diets and DCM. No data in the present study comes from Ohio. Finally, Michigan, where there are at least 5 clinics with cardiologists as well, one of which is Michigan State University Veterinary Medical Center. No data was presented from Michigan.
      It's also worth noting that no information is provided on what relative share of cardiology caseload the responding clinics receive in their region. This is important for a study like this one. There may be a bias in which referral centers local veterinarians recommend, and there may be differences in which clinics within a region are seeing atypical DCM cases. A survey could easily miss an increase over time if the increase was occurring disproportionately at a location not included in the survey.
      Authors write, "Of the 14 cardiology services that provided data, one provided 19 years of data, two provided 15 years of data, four provided between 8-10 years of data, and the remaining seven services provided five years or less of data (Table 1)" When the provided table is assessed, those seven with "five years or less," break down into one providing 1 year of data, one providing 2 years of data, two providing 3 years of data, and three providing 4 years of data. To put this into context, the FDA's first reports began in 2015, with the majority of reports received in 2018, after the public was alerted to the investigation. Cardiologists have reported seeing the rise in atypical cases as early as 2013. Data assessing trends over only the last four years or less is simply insufficient to characterize whether recent years have seen an increase in cases from a previous norm.

    10. Soooo....67,000 dogs in 9 states is less statistically significant than 500 California. Got it. That's what I have to believe if I want to believe grain free is a problem, even though the DCM population of dogs hasn't even gone up one .1% since grain-free sales took off.

    11. The data from the pre-print and the data from the various studies specifically examining the outcomes associated with patients aren't comparable, so calling one more statistically significant than the other isn't a reasonable judgement.

      From the pre-print you shared, we know that 67,243 dogs were seen at 14 cardiology practices over a variable span of time at each clinic, and some of those dogs were diagnosed with DCM. We don't know what diets these dogs were eating and we don't know what the outcomes were for these dogs. We don't know how their parameters varied in relation to factors like age, breed, and diet-history. Half of the practices that returned the survey didn't provide data that extended prior to onset of these atypical DCM cases, so we don't know if a change occurred around the onset of the cases developing.

      We don't know if the composition of grain-free diet formulas changed over time, which would be just an important factor as how much sales of these diets have changed over time.

      We also don't know how incidence of DCM has changed over time. We don't have nearly enough data to assert that it "hasn't even gone up 0.1%" As I discussed above, these practices may not even be the top referral centers for their region, meaning that even if DCM cases were increasing in frequency, their hospital wouldn't necessarily see it. For example, if most general practices in Ohio refer DCM cases to Ohio State University, then survey data from other specialty clinics may not change over time.

      Ideally, we would have at least 10 years of data and we would have it from practices that are the top referral centers in their region. It would also be useful to have a large screening of a random sample of the dog population to see what the underlying rate of occult DCM is, then compare that to other factors like breed, age, co-morbidities, tick exposure, and diet-history.

      To "believe" that certain diets may be a problem, you only have to believe a few things:
      1. Primary DCM is not reversible, but secondary DCM can be.
      2. A significant portion of DCM cases diagnosed in atypical patients are reversing.
      3. These patients are often eating grain-free or pulse-legume rich formulas and are switched off of these diets at the time of diagnosis.

      If that isn't compelling for you, that's fine. No one can force you to believe that there's an issue. But for many clinicians, that's enough to warrant caution and avoidance of the diets until there is more information either exonerating them or implicating them.

      It was decades between researchers thinking there was a concerningly strong connection between cigarette use and lung cancer and wide public acceptance of that connection. How many lives might have been prolonged or saved had those warnings been heeded earlier?

      If there IS an issue with these diets, it's eventually uncovered and because people were advised to avoid them, more dogs avoid illness.

      If there is NOT an issue with these diets, it's eventually uncovered and people go back to feeding them and maybe pet food companies learn a lesson about marketing claims that aren't evidence-based.

    12. Well for me - mycotoxins are a known issue with grains, and they have caused deaths and recalls for certain this past year. There is no "risk-free" diet period. And how are those not comparable??

      I'm sorry but I'd rather listen to Joe Bartges, George Collings, Ryan Yamka, George Flaherty, Renee Streeter and others.
      They put out a much more convincing argument that grain-free is safe. admit that we don't have the data - and we know cardiology clinics are terrible at keeping records - so we are just going to believe there is a problem without the data? Really? But the data we do have from 14 vet clinics shows no change in the DCM population. Those dogs are only reversing because of meds and taurine/carnitine supplements - that's not new. Legume-rich? You mean like Orijen at 9%?

      Are you aware of the meat content of Orijen/Taste of the Wild versus Purina grain-free? I'll tell you, the Purina is much more "legume" rich than those brands. That's why this whole thing is fishy.

      78-81% of the protein in TOTW High Prairie comes from meat. Versus 34-38% for Purina. So yeah switch to a food that is higher in legumes - yeah that makes sense. By your own logic - Orijen and TOTW are safer choices - and yet you tout Purina.

    13. Mycotoxins are absolutely a known issue, and that's why reputable manufacturers have screening processes in place to avoid contamination with mycotoxins, including Aflatoxin. Nothing in life is risk-free, but that doesn't mean we can't make informed decisions to mitigate risks.

      Experts don't always agree, and that's okay. I have no problem admitting or acknowledging that. In fact, I think it's important to listen to disagreements among experts and consider where discrepancies rest that cause them to disagree.

      Dr. Joe Bartges has a colleague who is also a nutritionist and internist, Dr. Sherry Sanderson, who researched diet-associated DCM in the 90's. She is concerned about these new cases. Why is his opinion worth more to you than hers? (And in turn, why might her opinion be worth more to me than his? To be honest, while he raises criticisms worth considering in regards to the FDA's investigation, he hasn't offered any dialogue that quells the crux of my concern personally, which is the observation of disease reversal in these patients.)

      Drs Collings and Yamka are not veterinarians. While I absolutely believe that PhD nutritionists are an important part of the industry and pet care field, that have valuable and valid expertise to bring to the table, I worry that they may not always have the clinical insight to understand why cardiologists have these concerns for their patients even in the face of limited, emerging evidence. I would love to see more cooperation between veterinarians and non-veterinarians, so that everyone can offer their unique insight and perspective, and less strife and defamation in getting to the bottom of these observations.

      I would also consider that you've listed only nutritionists. Don't you think that the opinions of cardiologists are important here too, especially since they're the ones seeing and treating these cases?

      I admit we lack the data to say whether or not there has been a change over time, absolutely. I don't think we "know cardiology clinics are terrible at keeping records." There are a variety of reasons clinics may have declined participating in the research, including simply not having the time to look up and provide the data to the researchers.

      Absence of evidence is not evidence of absence. In the face of insufficient data, I choose to believe the clinical professionals that are treating these cases and reporting a concerning trend. I hope that one day their concerns can be confirmed or laid to rest by data-- but until that data is available, patient care must be prioritized.

    14. As discussed previously, medications do not reverse DCM, and if taurine/carnitine cause reversal, then there was likely a deficiency secondary to diet.

      By 'legume-rich,' I mean 'higher inclusion rate than has been researched for use in dog diets.' That doesn't need to mean that the majority of protein in the diet is coming from those pulse legumes. Just that there are more of them present than historically used.

      For example, we have research on soybean meal approaching 20% on an as-fed basis. For ingredients like peas and lentils, I've been unable to find research above 2-3% inclusion rates. Mansilla et al. 2019 discussed that some foods on the market appear to have a pulse legume inclusion rate approaching 40%.

      Please don't put words in my mouth. I haven't recommended that anyone feed grain-free Purina diets. I'm personally choosing to avoid any diets with peas, lentils, or pulse legumes unless they're at the end of an ingredients list (least included by weight) until we have more information on the safety of these ingredients at various inclusion rates.

      That said, it is interesting to compare the different approaches that some companies have in grain-free formulation.

      For example, consider the ingredients lists (first 10 listed) below and let me know if any differences stand out to you, specifically in regards to the manner which pulses are included:

      - "Chicken, pea starch, cassava root flour, chicken meal, soybean germ meal, soybean meal, canola meal, beef fat preserved with mixed-tocopherols, dried beet pulp, natural flavor,"
      - "Organic chicken, organic chicken meal, organic cassava root flour, organic pea starch, organic canola meal, organic dried egg, organic coconut oil, organic carrots, organic pea hulls, organic pea protein"
      - "Organic chicken, organic chicken meal, organic cassava root flour, organic pea starch, organic canola meal, organic dried egg, organic coconut oil, organic sweet potatoes, organic pea hulls, organic pea protein"
      -"Deboned chicken, deboned turkey, chicken meal, whole green peas, whole red lentils, whole pinto beans, chicken liver, chicken fat, catfish meal, chickpeas"
      -"Deboned lamb, deboned duck, eggs, lamb meal, goat meal, catfish meal, whole red lentils, whole pinto beans, whole green peas, duck fat, whole green lentils"
      -"Duck, Duck Meal, Chickpeas, Peas, Chickpea Flour, Sunflower Oil (Preserved with Citric Acid), Flaxseed, Dehydrated Alfalfa Meal, Natural Flavors, Salt"
      -"Beef, Beef Broth, Dried Egg Product, Lentils, Peas, Pork Meat and Bone Meal, Potatoes, Sweet Potatoes, Dried Tomato Pomace, Pork Fat"
      -"Chicken, chicken meal, garbanzo beans, lentils, peas, potatoes, chicken fat preserved with mixed tocopherols), pea flour, egg product, tomato pomace"
      -"Fresh deboned chicken, deboned salmon, chicken meal, green peas, red lentils, chickpeas, green lentils, yellow peas, salmon oil (preserved with mixed tocopherols), sun-cured alfalfa meal"

  2. How does this in any way qualify as good science? "Grain free" dog food and "grain inclusive" are in no way commonalities. The variation from one food to another can be staggering. Grain free Purina is nothing like Orijen, the same way Ol' Roy is nothing like Royal Canin. Hell, even dogs themselves are completely varied based on breeds (thanks breeders for creating inbred monstrosities with endless lists of genetic problems). Cohort studies still need to control for commonalities, and this just doesn't. And why on earth are any Dobermanns used for a study involving dietary DCM? Nothing like having a breed that 70% likely to get DCM by just existing to screw with results.

    1. The authors discuss their rationale for the diet groups in their study, if you're interested in reading the full paper. While 'grain free' and 'grain inclusive' both represent a broad range of diets with wide variation, multiple studies have demonstrated reproducible, statistically significant differences between dogs eating one type or the other. All of these studies, this one included, acknowledge the need for further research to hone in on what, specifically, is 1) associated with the broader diet category of grain-free more than it is grain-inclusive and 2) causing atypical, reversible dilated cardiomyopathy. For example, it appears that high inclusion rates of pulse legumes are more common in grain-free diets. If, hypothetically, these ingredients had a dose-dependent cardiotoxic effect, it would explain why these cases are associated with grain-free diets, despite the broad array of diets that fall within that categorization.

      Dobermans, like any other breed, are susceptible to developing DCM from non-hereditary causes. Their predisposition for the disease is acknowledged by the study authors. In some ways, they provide unique insight to primary vs secondary DCM, as a typical Doberman with DCM would never show complete reversal of disease process, and yet some of these Dobermans (broadly, case reported Dobermans, not this study specifically) are.

      I appreciate you sharing your thoughts, and in no way am I trying to say this study is perfect-- but if we wait to conduct any science until we can conduct it in an airtight manner, with no limitations, we'll be waiting forever. The authors acknowledge their limitations and discuss where they advise taking future studies.

    2. Citation needed on those dobermans lol

    3. RE the Dobermans: Vet-LIRN 2019 update, multiple case reports (with verified veterinary records) within the FB citizen science effort cited at the end of the post.


Share what you think: