Grain-Free Dog Food and Heart Disease: What the Data Says (Part Two)

March 20, 2026

grain-free dog food and DCM research analysis showing data-driven canine heart health insights

This is Part 2 of a two-part series. Part 1 covered how the diet-DCM narrative was built: the FDA investigation, the data quality problems, the reporting bias, and the Hill’s lawsuit. This article covers what the peer-reviewed science shows, why the “switch back to grains” recommendation has its own problems, and what all of this means for how you feed your dog.

What the Peer-Reviewed Literature Says

The research base on DCM and diet is real, contested, and far more nuanced than either camp acknowledges. Here is a fair accounting of where it stands.

The ‘Golden Retriever’ Studies

Golden Retrievers became a focal point of the grain-free diet and diet-associated cardiomyopathy (DCM) investigation for a specific reason. That is they have a documented predisposition to taurine-deficient DCM, which notably has been recognized in the veterinary literature since at least the mid-1990s (Kramer, Kittleson et al. 1995). In one report from 2005, a reversible taurine-deficient dilated cardiomyopathy was described in five related Goldens, with marked improvement after taurine supplementation. This was not the typical course for hereditary (primary) DCM (Bélanger, Ouellet et al. 2005). Subsequently, this finding deservingly attracted more research attention. Work following initial investigations demonstrated that Goldens are over-represented in taurine-deficient DCM case series and that many affected dogs improve when diet is changed and taurine is supplemented (Kaplan, Stern et al. 2018). This is a notable difference from the primary hereditary DCM seen in breeds like Dobermans, Great Danes, or Irish Wolfhounds, where disease does not improve with taurine supplementation (O’Grady and O’Sullivan 2004).

In 2018, Kaplan and colleagues published a multicenter prospective observational study describing 24 Golden Retrievers with taurine-deficient DCM and 52 healthy Goldens (Kaplan, Stern et al. 2018). Certain diets and diet characteristics were associated with taurine deficiency, and most affected dogs showed echocardiographic and clinical improvement after a combination of interventions: diet change, taurine and L-carnitine supplementation, and standard heart-failure medications. This strengthened the case that Goldens can develop taurine-responsive DCM linked to diet. However, because multiple treatments were introduced simultaneously, the study cannot isolate the specific contribution of diet change from taurine supplementation and cardiac drugs.

In 2020, the same group published a higher-profile study in PLOS ONE evaluating 86 apparently healthy Golden Retrievers fed “traditional” (grain-inclusive, large-manufacturer) versus “non-traditional” (grain-free, legume-heavy, small-manufacturer) diets (Ontiveros, Whelchel et al. 2020). Dogs in the non-traditional group had, on average, lower taurine concentrations and more frequent echocardiographic changes consistent with subclinical DCM.

Later that year, in September 2020, PLOS ONE issued an ‘Expression of Concern’ for this paper (Ontiveros, Whelchel et al. 2020), citing undisclosed conflicts of interest. The conflicts of interest included co-authors’ ties to a commercial amino acid laboratory that captured data for the study, in addition to substantial heterogeneity between diet groups in factors independently associated with DCM risk. After review by an independent statistical expert, the journal issued a formal correction in August 2023. Those editors concluded that between-group heterogeneity was a “substantial concern” and that conclusions linking diet group to echocardiographic abnormalities were “not well-supported” given the study design. Consequently, the authors corrected article, substantially softening its original claims.

Before the ‘Expression of Concern’ and correction, this paper was frequently cited as strong prospective evidence of a grain-free/taurine/DCM link. The more appropriate interpretation of this study that it is hypothesis-generating – at best. This is because it is a notably confounded observational study that raises legitimate questions but is not a reliable standalone basis for definitive clinical recommendations.

The 18-Month Randomized, Double-Blind Trial

On the other end of the evidence spectrum is an 18-month randomized, parallel-group, double-blind study published in the Journal of Animal Science (JAS) evaluating 60 healthy adult dogs being fed four diets with different ingredient profiles (Morris, Stiers et al. 2025). Diets included grain-free with potatoes and peas, grain-inclusive with peas and pea fiber, grain-inclusive without peas or potatoes, and grain-free with potatoes only. Researchers assessed echocardiographic parameters, whole blood and plasma taurine, urinary taurine, and cardiac biomarkers (troponin-I, NT-proBNP) at 6, 12, and 18 months.

The study found no clinically significant changes or between-group differences in cardiac biomarkers, and whole blood and plasma taurine remained within normal ranges and were unaffected by diet assignment (Morris, Stiers et al. 2025). This is as close to a controlled experiment as can be ethically conducted in healthy dogs. It stands as one of the strongest pieces of evidence that properly formulated grain-free diets (even those containing peas/potatoes) do not predictably induce DCM or meaningful cardiac biomarker changes over 18 months. On the other hand, it does not exonerate every commercial grain-free product, nor does it address dogs with pre-existing cardiac or metabolic risk factors, nor does it answer questions beyond an 18 month period. But it directly challenges the blanket clinical narrative that grain-free diets are inherently cardiotoxic.

The Irish Wolfhound Study (Coppinger et al. 2024)

In a 2024 retrospective study of 97 apparently healthy Irish Wolfhounds (Coppinger, Freeman et al. 2024). This breed is documented to have a well-recognized hereditary cardiomyopathy. Coppinger and colleagues compared dogs eating high-pulse diets to those eating low-pulse diets. Results showed no significant differences in echocardiographic measurements between groups, but dogs in the high-pulse group had a significantly higher prevalence of ventricular premature complexes (VPCs) on a 1-minute ECG strip (17% versus 2%). VPCs are a potential early marker of cardiac electrical instability and can precede more serious arrhythmias in some breeds.

The finding is noteable as it involves apparently healthy dogs in a genetically predisposed breed, focused on an arrhythmic endpoint that may precede overt DCM. At the same time, the study is retrospective, limited to one predisposed breed, and based on short ECG strips rather than confirmed cardiomyopathy. It is not a straight line from “legumes cause heart failure,” but a signal that justifies prospective, breed-stratified research. It is also worth noting that senior authors include Drs. Freeman and Rush, who are named defendants in the KetoNatural v. Hill’s litigation; readers can weigh that conflict of interest accordingly.

Owens et al. (2023) Cross-Sectional Study

A 2023 prospective cross-sectional study of 46 healthy client-owned dogs compared those eating traditional grain-inclusive diets to those eating nontraditional, legume-heavy diets (Owens, LeBlanc et al. 2023). Dogs in the nontraditional group had lower indices of systolic function and larger left ventricular volumes, although the authors emphasized these values generally remained within established reference intervals. There were no significant differences in cardiac biomarkers (NT-proBNP or cardiac troponin I) or whole blood/plasma taurine between groups.

These results may provide some evidence that legume-heavy diets may be associated with subtle subclinical echocardiographic differences in some dogs. However, the small sample size, cross-sectional design, and absence of biomarker differences limit how far those findings can be taken in clinical decision-making. This means, these results from 46 dogs cannot and should not be extrapolated across the entire canine population.

The 2020 BSM Partners Literature Review

A June 2020 review in the Journal of Animal Science examined more than 150 published studies on canine DCM and its potential causes (McCauley, Clark et al. 2020). The authors concluded that the available literature did not support a definitive causal link between grain-free or legume-rich diets and DCM, arguing that the BEG framework lacked scientific rigor and that FDA case data were too incomplete and biased to support firm conclusions.

This review does not function as proof that the dietary DCM issue was fabricated. It’s important to emphasize that many of the limitations it highlights in case-series data are real, and that the issue is very real for a subset of dogs. Nevertheless, this study should be weighed with the same critical scrutiny applied to any industry-funded work.

Not surprisingly, the review attracted criticism from some veterinary commentators who argue that it selectively emphasizes methodological flaws in studies supporting a diet association and may understate the clinical significance of reported improvement following diet change and supplementation (SkeptVet 2021). A conflict-of-interest concern also arose in this study as well. BSM Partners, which funded and conducted the work, provides formulation and consulting services to pet food companies including one or more brands that appeared on the FDA’s most-reported list. Nonetheless, the journal issued a clarification regarding the authors’ employment and BSM’s industry role.

It is worth noting that BSM Partners is one company in the pet food industry, among others, that provides formulation and technical support that many smaller boutique companies choose not to use. These smaller companies sometimes underestimate or dismiss the need for formal nutrition expertise, robust validation, and true transparency. In that sense, BSM is attempting to solve the very problem that the DCM investigation indirectly highlighted: the lack of rigorous formulation and published nutritional data behind many commercial diets.

Two 2025 Narrative Reviews with Different Conclusions

A broad 2025 review examined grain-free diets across multiple health dimensions including DCM, allergies, glycemic response, and mycotoxin exposure (Zhang, Ji et al. 2025). For DCM specifically, the evidence was characterized as mixed: FDA data and case series suggest a possible association, but several studies including controlled trials found no association. The overall conclusion was that nutritional adequacy, balance, and individualized assessment matter more than the presence or absence of grain. This is the more measured of the two reviews.

A second 2025 narrative review focused narrowly on diet as a predisposing factor for DCM (Mornard, Brasileiro et al. 2025). It characterizes the cumulative findings as suggesting a strong link between nontraditional, legume-rich diets and DCM, It noted larger left ventricular diameters, reduced systolic function, and more frequent ventricular premature complexes in dogs on such diets. Importantly, this review does not prominently address the methodological criticisms of the case-series literature or the null findings from controlled trials.

The fact that two peer-reviewed narrative reviews published in the same year, drawing on largely the same underlying literature, reached different characterizations reflects a larger point. It highlights a genuine divide in how researchers are currently weighing this evidence, and is exactly why a single headline or a single study citation cannot give you the full picture.

Taurine: The Leading Hypothesis and Its Limits

Taurine deficiency has been a nutritionally established cause of DCM in cats since at least the 1980s (Hayes 1982, Pion, Kittleson et al. 1987, Pion, Kittleson et al. 1992). In cats, taurine is an essential amino acid that must be obtained from diet. At the time, this was a breakthrough that led to mandatory taurine supplementation of commercial cat food and largely eliminated the problem in that species (Pion, Kittleson et al. 1992). The hypothesis that a parallel mechanism might operate in some dog breeds is plausible and supported by evidence in specific contexts, particularly Golden Retrievers, who have a documented predisposition to taurine-deficient DCM.

Dogs can synthesize their own taurine from methionine and cysteine, and therefore taurine was not historically considered essential or required at a minimum level in commercial pet foods (Hand, Thatcher et al. 2010). Since the most abundant dietary source of taurine is muscle meat, the assumption was that commercial foods with animal protein would supply enough (Spitze, Wong et al. 2003). However, breed-specific differences, genetic factors, and health status mean that some dogs may require dietary taurine. Inflammatory states are also relevant here: taurine plays a role in immune cell function and modulating oxidative stress, and chronic inflammation can increase taurine demand and utilization. A dog with underlying gut inflammation, systemic inflammatory burden, or dysbiosis may have elevated taurine requirements that dietary intake alone does not meet, independent of diet type. As a result, taurine should be considered conditionally essential in canine nutrition multiple studies demonstrate several factors that may render endogenous synthesis insufficient, leading to taurine‑deficient DCM in some dogs (Li and Wu 2023), regardless of diet type.

What taurine deficiency does not explain is that many dogs reported to have diet-associated DCM have normal blood taurine concentrations (McCauley, Clark et al. 2020). This finding has been replicated across multiple studies and is not contested. If simple systemic taurine depletion were the primary driver, we would expect to see deficiency consistently. We do not. The current mechanistic thinking is considerably more complex: interactions between dietary fiber fractions, bile acid cycling, sulfur amino acid bioavailability, gut microbiota composition, and breed-specific metabolic differences may all contribute to functional taurine insufficiency at the tissue level even when blood taurine appears normal (Ko, Backus et al. 2007, Ko and Fascetti 2016, Donadelli, Pezzali et al. 2020, Quilliam, Ren et al. 2021).

This means taurine status is not a single number to measure once and interpret in isolation. It is a dynamic metabolic state influenced by diet composition, body condition, gut function, and breed genetics. Assessing whether any given dog is “taurine sufficient” is considerably more complicated than a single blood draw suggests. The leap from “some affected dogs have low taurine” to “grain-free diets cause taurine deficiency” happened much faster than the underlying science justified.

Why Taurine Measurement Is More Complicated Than a Blood Draw

The finding that many dogs with diet-associated DCM had normal blood taurine concentrations has been interpreted in two directions: by some as evidence that taurine is not involved, and by others as evidence that the measurement is inadequate. The second interpretation is better supported by what we know about taurine physiology.

Both whole blood and plasma taurine concentrations reflect recent dietary intake and short-term metabolic flux more than they reflect tissue taurine stores (Tôrres, Biourge et al. 2022). The relationship between circulating taurine concentrations and cardiac muscle taurine concentrations in dogs is poorly characterized, and the data are limited. A dog can have circulating taurine within the laboratory reference interval while having suboptimal taurine availability at the tissue level, particularly under conditions of increased biosynthetic demand or impaired recycling. Studies reporting normal blood taurine in affected dogs are not exonerating taurine as a mechanistic contributor, but rather this is exposure of a measurement gap. Reference interval-based blood taurine testing was developed in a different clinical context and was not validated as a tool for ruling out functional taurine insufficiency in the context of a complex dietary exposure question. Using it as though it has shaped the interpretation of a substantial portion of this literature.

“Grain-Free” Is a Marketing Category, Not a Nutritional Mechanism

This is a point that gets lost in almost every public discussion of this topic.

“Grain-free” is a marketing designation, not a nutrient profile. Two products can carry the same label and have nothing else in common. One may be formulated by a board-certified veterinary nutritionist and/or PhD/MS nutritionist, with documented amino acid (protein) bioavailability, third-party AAFCO nutrient profile verification, and published feeding trial data. Another may come from a company with no in-house nutrition expertise, legume fractions dominating the first five ingredients, no feeding trial history, and no published data on sulfur amino acid adequacy. The label tells you what is not in the bag. It also tells you nothing about whether the formulation is sound, whether protein quality has been characterized and verified, or whether the amino acid profile supporting taurine synthesis has ever been measured.

The BEG framework was coined to describe a cluster of characteristics that often appeared together in implicated products. These were small manufacturers, novel proteins, high legume inclusion, and absence of AAFCO feeding trial data. It bundles multiple independent variables under one label. “Grain-free” was the most marketable part of that label, so it became the shorthand. The mechanism, to the extent one exists, almost certainly has nothing to do with the absence of grain and everything to do with specific ingredient composition, processing quality, level of processing, and nutritional completeness.

Grain-inclusive diets are not automatically safe or well-formulated. The ironic part to this entire story is that the real transparency gap is not limited to small or boutique manufacturers. Some veterinary brands, including Hill’s, Purina, and Royal Canin, do not have (or at least won’t provide) AAFCO feeding trial substantiation for every formula in their lineup. More concerning, none of these brands publicly provide a full analyzed AAFCO nutrient profile for their products when asked, even on the veterinary level. They publish guaranteed minimums and maximums which are the floor and ceiling required by law. These are not the real analyzed values that would allow a nutritionist, veterinarian, or informed consumer to evaluate what a dog is actually consuming. A guaranteed minimum for crude protein tells you nothing about amino acid profile, sulfur amino acid (taurine) bioavailability, or whether the formulation has ever been assessed for taurine precursor adequacy. That information usually does exist internally. It is simply not shared under the guise of ‘proprietary or intellectual’ reasoning.

This matters because the entire premise of the “reputable company” qualifier embedded in the Tufts reporting protocol, and implied throughout the BEG framework, rested on the assumption that major brands represent a nutritionally characterized, validated, and verifiable baseline. That assumption is false. As of this writing, the only companies willing to provide that level of data transparency would themselves be classified as BEG, which may just be the most ironic part about this whole story. Regardless, the DCM investigation, whatever its methodological limitations, was never a legitimate basis for treating grain inclusion or brand size as proxies for nutritional rigor. Those are marketing variables. The relevant variable is data transparency. It is not brand recognition, not ingredient composition, and not market share.

What the “Switch Back to Grains” Narrative Gets Wrong

The diet-associated DCM narrative that grain-inclusive diets are the safer default, rests on a framing assumption that has gone largely unexamined. That is conventional grain-inclusive dry dog food represents a well-characterized, safe baseline against which grain-free diets should be evaluated. That assumption was never tested, and therefore has not been proven.

Agricultural Contaminants: The Risk That Doesn’t Get Discussed

Grains are agricultural commodities, and they carry the contamination profile of agricultural commodities. Mycotoxins which include aflatoxins, deoxynivalenol, and fumonisins are produced by mold species that colonize corn, wheat, and other grain crops both in the field and in storage (Macías-Montes, Rial-Berriel et al. 2020). Aflatoxin contamination in corn-based pet foods has caused documented mass fatality events in dogs, with multiple FDA-reported outbreaks over the past two decades (PFI 2011, FDA 2021). Pesticide and herbicide residues represent an additional consideration, particularly for crops with well-established high agricultural chemical exposure profiles (Macías-Montes, Luzardo et al. 2025). These are regulatory-documented, recurring risks associated with specific grain ingredients. Risks that were never meaningfully incorporated into the diet-DCM narrative as a plausible contributing factor or alternative health concern. That omission is not a minor oversight. It reflects a framework that treated grain inclusion as inherently protective while leaving its actual contamination burden unexamined.

Digestibility, Bile Acid Cycling, and What We Don’t Yet Know

Not all grains are equivalent in how dogs process them, and this matters for the taurine hypothesis specifically. Certain grain fractions, particularly insoluble fiber components, are poorly fermented or digested in the canine gastrointestinal tract (Sunvold, Fahey et al. 1995, de Godoy, Kerr et al. 2013). Impaired digestibility has implications for bile acid cycling: when bile acids are excreted in feces rather than recirculated, the body’s demand for taurine synthesis increases (Donadelli, Pezzali et al. 2020). If specific grain fiber fractions are contributing to bile acid loss, then a grain-inclusive diet is not categorically protective in the way the clinical messaging implied.

This is not a settled question. The literature characterizing the bile acid and taurine metabolic effects of specific grain fractions in dogs is thin. Recommending a return to grain-inclusive diets as the evidence-based safer choice requires evidence that those diets are, in fact, safer along this specific metabolic pathway. That evidence does not robustly exist.

After Extrusion, the Label Tells You Less Than You Think

When starch-based ingredients are processed through high-temperature, high-pressure extrusion, which is how most dry pet food is manufactured, starch granules gelatinize (Perry, Valach et al. 2022). The degree of gelatinization, and consequently the glycemic and digestive behavior of the finished product, is driven substantially by processing conditions rather than the identity of the starch source. A well-formulated kibble built on rice or oatmeal and a well-formulated kibble built on potato or peas are not nutritionally equivalent in every respect, but the structural differences between them at the point of canine digestion are considerably smaller than ingredient list comparisons suggest. The name on the label identifies the source; it does not fully describe what the dog’s gastrointestinal tract encounters.

Adding Grains to a Complete Diet Is Not a Solution

Some owners have reported their veterinarian advising them to add cooked grains such as rice or pasta to a grain‑free diet as a way to reduce perceived diet‑associated DCM risk, rather than changing the underlying formulation (Annonymous 2025). This deserves direct challenge. A complete and balanced commercial diet is formulated to deliver specific macronutrient ratios, amino acid profiles, and micronutrient targets per calorie consumed. Adding calorie-dense, low-protein carbohydrate sources dilutes protein density, distorts the amino acid balance, which includes the sulfur-containing amino acid precursors relevant to taurine synthesis, and contributes to caloric excess without proportional nutritional return. If impaired taurine recycling is the mechanism of concern, reducing the dietary protein fraction and increasing carbohydrate load is not consistent with addressing that mechanism. It is the kind of reductive clinical translation that occurs when a hypothesis moves into practice faster than the evidence justifies.

Pet Food Processing Chemistry Questions

The diet-associated DCM investigation focused narrowly on ingredient category as the variable of concern. It did not ask what happens to food during manufacturing, or more specifically the thermal processing of pet food. The thermal processing conditions required to produce dry pet food at commercial scale generate a predictable suite of bioactive compounds with documented physiological consequences (van Rooijen, Bosch et al. 2014, Oba, Hwisa et al. 2022, Bridglalsingh, Archer-Hartmann et al. 2024, Xiong, Li et al. 2024). The only question is not what goes into the bag. Other questions about byproducts of the manufacturing process, and what lifetime exposure to those outputs does to the animal consuming them are also important.

One of these byproducts, advanced glycation end products (AGEs) are formed when reducing sugars react non-enzymatically with proteins, lipids, or nucleic acids under heat. The reaction pathway is the Maillard reaction, and it is not incidental to pet food manufacturing, rather it is intrinsic to it. High-temperature extrusion generates well-characterized AGE species including Nε-carboxymethyllysine (CML) and Nε-carboxyethyllysine (CEL) at concentrations that substantially exceed what is found in minimally processed food. This is not a grain-free problem or a grain-inclusive problem. It is a processed food problem.

AGEs are not inert. They cross-link collagen and elastin, reducing tissue compliance in load-bearing structures including the myocardium (Wasim, Mahmood et al. 2022). They activate the receptor RAGE, triggering inflammatory signaling, oxidative stress, and endothelial dysfunction. Chronic RAGE activation has been associated with fibrosis, impaired cardiac contractility, and vascular stiffening in human and animal models (Hooshiar, Esmaili et al. 2022). The cardiac implications of lifetime AGE exposure in dogs have not been studied. That is the point.

AGEs are not the only concern. Acrylamide is a probable human carcinogen whose metabolite glycidamide is genotoxic (Perera, Hewavitharana et al. 2021). It forms predictably from asparagine and reducing sugars above 120°C, meaning starchy kibble of any formulation is a reliable source. High-temperature protein processing generates heterocyclic amines. Lipid oxidation produces malondialdehyde and 4-hydroxynonenal, both of which modify proteins and DNA and compound the AGE burden. Furans and oxidized lipid polymers accumulate in rendered fat fractions and persist in the finished product. None of these are manufacturing defects. They are predictable outputs of the thermal conditions required to produce pet food at commercial scale.

None of these compounds appeared in the FDA’s diet-associated DCM investigation. None were weighed against the speculative legume association when veterinary organizations recommended returning to conventional grain-inclusive kibble. The processing chemistry of conventional dry pet food escaped scrutiny not because it was demonstrated safe, but because it was the historical norm. Historical norm is not the same as demonstrated safety. That asymmetry is a scientific problem this field has not yet addressed.

What Adequate Evidence Would Actually Look Like

The diet-associated DCM literature has generated substantial noise. It has not generated a study that could settle the question it raised. It is worth being honest about why. The answer is because the study that would actually answer this question is not merely difficult. It may be practically and ethically impossible to execute.

A properly designed interventional study would require, at minimum:

Randomized diet assignment with standardized formulations and fully documented ingredient sourcing.
Cardiologist-confirmed echocardiographic endpoints, evaluated blinded to diet group, across follow-up periods long enough to capture a disease that typically develops over years, not weeks or months.
Whole blood taurine measured alongside tissue-level assessment, because circulating taurine and cardiac muscle taurine do not reliably correlate.
Pharmaceutical and supplementation interventions withheld or rigorously controlled, which is where the ethical problem becomes unavoidable. If a dog develops echocardiographic changes consistent with DCM during the study period, withholding pimobendan or other medications known to improve outcomes is not ethically defensible.

The study design that would generate clean causal data cannot be run in dogs with any reasonable ethical framework.

Sample size requirements compound the problem. DCM occurs in under 0.5% of non-predisposed dogs. Detecting a dietary effect on a condition with that background prevalence requires enrollment numbers that no single institution can feasibly support. And the breed question adds another layer:

A study powered to detect an effect in the general dog population would almost certainly miss breed-specific metabolic vulnerabilities.
A breed-specific study cannot be generalized to the broader population.

These are not the same question, and they cannot be answered by the same study. What this means in practice is that the evidentiary standard used to reshape clinical guidelines and consumer behavior in 2018, a voluntary case series filtered through a biased reporting protocol, was not a temporary gap waiting to be filled by better research. It reflects a genuine structural constraint on what this field can produce. The question was raised with epidemiological tools that cannot establish causation. The study design that could establish causation faces ethical and logistical barriers that have not been resolved. That is the honest state of the science, and the public deserves to have it described plainly.

What the Evidence Supports and Doesn’t

What Appears to Be Real

Some dogs diagnosed with DCM have been eating certain grain-free, legume-rich diets at the time of diagnosis. A subset of those dogs have shown echocardiographic and clinical improvement after a combination of interventions that usually includes diet change, taurine and L-carnitine supplementation, and standard cardiac medications. The clinical cases and improvements are real; what is not clear is how much of that improvement is specifically attributable to diet.
In breeds with an existing genetic predisposition to cardiomyopathy, including Irish Wolfhounds and possibly Golden Retrievers, high-pulse or nontraditional dietary patterns are associated with measurable differences in cardiac risk markers such as ventricular premature complexes and subtle changes in systolic function in some cross-sectional studies, although most values often remain within reference intervals and causality has not been established.
Taurine status is a meaningful variable in at least some dogs, particularly Golden Retrievers and certain spaniels, and can interact with diet composition in ways that affect risk of taurine-responsive DCM. However, many suspected diet-associated DCM cases have normal circulating taurine, so taurine deficiency cannot explain all of them.
Reversal or improvement of cardiac changes following intervention is documented in a subset of dogs with suspected diet-associated or taurine-deficient DCM. That pattern of improvement is more consistent with a nutritional or secondary process than with classic hereditary DCM. However, because diet, supplementation, and medications are typically changed together, the specific contribution of diet alone remains uncertain.

What Is Not Established

A definitive causal mechanism. Taurine deficiency explains some cases; it does not explain all of them. The alternative hypotheses which is that fiber-mediated bile acid interference with taurine recycling, antinutritional factors in legumes, compromised amino acid bioavailability, microbiome-mediated effects remain speculative, with some having more plausibility than others.
That “grain-free” as a category increases DCM risk compared to all other properly formulated, tested diets. The 18-month RCT did not find this. The FDA did not conclude this. The best reading of the literature is that specific formulation profiles in some grain-free products, but not grain-free itself, may be risk-modifying in susceptible dogs.
That the initial FDA case database was representative of the true population of DCM cases in grain-free-fed dogs, given the documented reporting protocol at Tufts and the surge in reports following public announcement.

How NorthPoint Pets Approaches This

NPP sells raw diets, fresh diets, freeze-dried diets, and grain-free kibble alongside grain-inclusive options. We also work closely with veterinarians across a range of diet and health conversations. We have some financial stake in how this question is answered, but truthfully, we carry products on both sides of this debate and our business doesn’t hinge on the outcome. What we do have a stake in is giving clients an accurate read on the evidence, because that’s the foundation the rest of what we do is built on. So rather than a position statement, what follows is an honest accounting of what the science shows and where it genuinely falls short.

We do not tell clients that grain-free diets cause DCM. The evidence does not justify that level of certainty.
We do not tell clients that grain-free diets are automatically safe. Formulation quality, ingredient sourcing, amino acid bioavailability, and batch consistency vary enormously across grain-free products, as they do across all commercial pet foods.
We evaluate products on evidence of nutritional competence: AAFCO data, manufacturer transparency about analyzed nutrient levels, amino acid profile data, company nutrition expertise, and track record. These criteria apply regardless of whether a food contains grain.
For high-risk breeds like Dobermans, Great Danes, Irish Wolfhounds, Golden Retrievers we recommend discussing targeted monitoring with your veterinarian: taurine assessment (ideally with breed-specific reference intervals), baseline echocardiography, and awareness of early DCM symptoms. The diet question is part of that conversation, not a substitute for clinical assessment.
We don’t manage diet choices based on marketing labels. “Grain-free,” “ancient grains,” “ancestral,” “wild-inspired” because none of these terms carry nutritional meaning that can be evaluated without looking at the actual analyzed nutrient data behind them.

Bottom line for your dog: DCM is real. Diet-related DCM is real in a subset of dogs. The evidence for “grain-free as a category causes DCM” is not strong, and the data collection that drove the public narrative had documented methodological problems. The honest answer, which no one in this conversation has commercial incentive to give you, is that we don’t fully understand the mechanism yet, the risk is real but narrower than headlines suggested, and diet quality is a more meaningful variable than the presence or absence of grain.

If you want to work through your dog’s specific diet, risk factors, and what monitoring might be appropriate, that’s exactly the kind of conversation we’re built for at NorthPoint Pets.

The information in this article is for educational and informational purposes only and is not a substitute for individualized veterinary medical advice, diagnosis, or treatment. Always consult your veterinarian or a board- certified specialist regarding your pet’s specific health needs, diagnostic testing, and dietary recommendations. Never disregard professional veterinary advice or delay seeking it because of something you have read here.

REFERENCES:

Annonymous. (2025). “Vet said add rice to our dog’s grain-free food to reduce DCM risk. Need a 2nd opinion.” Retrieved March 11, 2026, from https://www.reddit.com/r/AskVet/comments/1iqvb2a/vet_said_add_rice_to_our_dogs_grainfree_food_to/.

Bélanger, M. C., M. Ouellet, G. Queney and M. Moreau (2005). “Taurine-deficient dilated cardiomyopathy in a family of golden retrievers.” J Am Anim Hosp Assoc 41(5): 284-291.

Bridglalsingh, S., S. Archer-Hartmann, P. Azadi, C. Barbier de La Serre, R. L. Remillard, G. D. Sunvold and J. W. Bartges (2024). “Association of four differently processed diets with plasma and urine advanced glycation end products and serum soluble receptor for advanced glycation end products concentration in healthy dogs.” J Anim Physiol Anim Nutr (Berl) 108(3): 735-751.

Coppinger, L. M., L. M. Freeman, W. D. Tyrrell Jr, S. L. Rosenthal, M. E. Dentino, F. L. Abrams and J. E. Rush (2024). “Echocardiographic and electrocardiographic findings in Irish Wolfhounds eating high-pulse or low-pulse diets.” Journal of Veterinary Internal Medicine 38(3): 1300-1304.

de Godoy, M. R., K. R. Kerr and G. C. Fahey, Jr. (2013). “Alternative dietary fiber sources in companion animal nutrition.” Nutrients 5(8): 3099-3117.

Donadelli, R. A., J. G. Pezzali, P. M. Oba, K. S. Swanson, C. Coon, J. Varney, C. Pendlebury and A. K. Shoveller (2020). “A commercial grain-free diet does not decrease plasma amino acids and taurine status but increases bile acid excretion when fed to Labrador Retrievers.” Transl Anim Sci 4(3): txaa141.

FDA. (2021, January 11, 2021). “U.S. Food & Drug Administration (FDA) Midwestern Pet Foods Voluntarily Expands Recall of Pet Food for Aflatoxin Health Risk | FDA.” Retrieved January 11, 2021, 2021, from https://www.fda.gov/safety/recalls-market-withdrawals-safety-alerts/midwestern-pet-foods-voluntarily-expands-recall-pet-food-aflatoxin-health-risk.

Hand, M. S., C. D. Thatcher, R. L. Remillard, P. Roudebush, B. J. Novtony and L. D. Lewis (2010). Small animal clinical nutrition. Topeka, Kan, Mark Morris Institute.

Hayes, K. C. (1982). “Nutritional problems in cats: taurine deficiency and vitamin A excess.” Can Vet J 23(1): 2-5.

Hooshiar, S. H., H. Esmaili, A. Taherian and S. Jafarnejad (2022). “Exercise, Advanced Glycation End Products, and Their Effects on Cardiovascular Disorders: A Narrative Review.” Heart and Mind 6(3): 139-150.

Kaplan, J. L., J. A. Stern, A. J. Fascetti, J. A. Larsen, H. Skolnik, G. D. Peddle, R. D. Kienle, A. Waxman, M. Cocchiaro, C. T. Gunther-Harrington, T. Klose, K. LaFauci, B. Lefbom, M. Machen Lamy, R. Malakoff, S. Nishimura, M. Oldach, S. Rosenthal, C. Stauthammer, L. O’Sullivan, L. C. Visser, R. Williams and E. Ontiveros (2018). “Taurine deficiency and dilated cardiomyopathy in golden retrievers fed commercial diets.” PLoS One 13(12): e0209112.

Ko, K. S., R. C. Backus, J. R. Berg, M. W. Lame and Q. R. Rogers (2007). “Differences in taurine synthesis rate among dogs relate to differences in their maintenance energy requirement.” J Nutr 137(5): 1171-1175.

Ko, K. S. and A. J. Fascetti (2016). “Dietary beet pulp decreases taurine status in dogs fed low protein diet.” J Anim Sci Technol 58: 29.

Kramer, G. A., M. D. Kittleson, P. R. Fox, J. Lewis and P. D. Pion (1995). “Plasma taurine concentrations in normal dogs and in dogs with heart disease.” Journal of veterinary internal medicine 9(4): 253-258.

Li, P. and G. Wu (2023). “Amino acid nutrition and metabolism in domestic cats and dogs.” J Anim Sci Biotechnol 14(1): 19.

Macías-Montes, A., O. P. Luzardo, A. Zaccaroni and A. Acosta-Dacal (2025). “Pesticides residues in pet food: A market-based study on prevalence and toxicological implications.” Environmental Pollution 376: 126399.

Macías-Montes, A., C. Rial-Berriel, A. Acosta-Dacal, L. A. Henríquez-Hernández, M. Almeida-González, Á. Rodríguez-Hernández, M. Zumbado, L. D. Boada, A. Zaccaroni and O. P. Luzardo (2020). “Risk assessment of the exposure to mycotoxins in dogs and cats through the consumption of commercial dry food.” Science of The Total Environment 708: 134592.

McCauley, S. R., S. D. Clark, B. W. Quest, R. M. Streeter and E. M. Oxford (2020). “Review of canine dilated cardiomyopathy in the wake of diet-associated concerns.” J Anim Sci 98(6): skaa155.

Mornard, L., A. C. M. Brasileiro and M. Marcondes-Santos (2025). “Role of Diet as a Predisposing Factor for Dilated Cardiomyopathy in Dogs: A Narrative Review.” Vet Sci 12(11).

Morris, E. M., C. A. Stiers, L. B. Hancock and K. L. Gross (2025). “Different carbohydrate sources in dog foods supported overall health and cardiac function: an 18-mo prospective study in healthy adult dogs.” J Anim Sci 103.

O’Grady, M. R. and M. L. O’Sullivan (2004). “Dilated cardiomyopathy: an update.” Veterinary Clinics: Small Animal Practice 34(5): 1187-1207.

Oba, P. M., N. Hwisa, X. Huang, K. R. Cadwallader and K. S. Swanson (2022). “Nutrient and Maillard reaction product concentrations of commercially available pet foods and treats.” J Anim Sci 100(11): skac305.

Ontiveros, E. S., B. D. Whelchel, J. Yu, J. L. Kaplan, A. N. Sharpe, S. L. Fousse, A. E. Crofton, A. J. Fascetti and J. A. Stern (2020). “Development of plasma and whole blood taurine reference ranges and identification of dietary features associated with taurine deficiency and dilated cardiomyopathy in golden retrievers: A prospective, observational study.” PLOS ONE 15(5): e0233206.

Owens, E. J., N. L. LeBlanc, L. M. Freeman and K. F. Scollan (2023). “Comparison of echocardiographic measurements and cardiac biomarkers in healthy dogs eating nontraditional or traditional diets.” J Vet Intern Med 37(1): 37-46.

Perera, D. N., G. G. Hewavitharana and S. B. Navaratne (2021). “Comprehensive Study on the Acrylamide Content of High Thermally Processed Foods.” Biomed Res Int 2021(1): 6258508.

Perry, E. B., A. A. Valach, J. M. Fenton and G. E. Moore (2022). “An Assessment of Starch Content and Gelatinization in Traditional and Non-Traditional Dog Food Formulations.” Animals (Basel) 12(23).

PFI. (2011, December 29, 2011). “US FDA issues two more aflatoxin-related dog food recalls.” Pet Food Industry (PFI) Retrieved March 9, 2026, from https://www.petfoodindustry.com/news-newsletters/pet-food-news/article/15456024/us-fda-issues-two-more-aflatoxin-related-dog-food-recalls.

Pion, P. D., M. D. Kittleson, Q. R. Rogers and J. G. Morris (1987). “Myocardial failure in cats associated with low plasma taurine: a reversible cardiomyopathy.” Science 237(4816): 764-768.

Pion, P. D., M. D. Kittleson, W. P. Thomas, M. L. Skiles and Q. R. Rogers (1992). “Clinical findings in cats with dilated cardiomyopathy and relationship of findings to taurine deficiency.” Journal of the American Veterinary Medical Association 201(2): 267-274.

Quilliam, C., Y. Ren, T. Morris, Y. Ai and L. P. Weber (2021). “The Effects of 7 Days of Feeding Pulse-Based Diets on Digestibility, Glycemic Response and Taurine Levels in Domestic Dogs.” Frontiers in Veterinary Science Volume 8 – 2021.

SkeptVet (2021). https://skeptvet.com/2021/01/diet-associated-dilated-cardiomyopathy-dcm-in-dogs-what-do-we-know-so-far/ Accessed March 8, 2026 2026.

Spitze, A., D. Wong, Q. Rogers and A. Fascetti (2003). “Taurine concentrations in animal feed ingredients; cooking influences taurine content.” Journal of Animal Physiology and Animal Nutrition 87(7‐8): 251-262.

Sunvold, G. D., G. C. Fahey, Jr., N. R. Merchen, E. C. Titgemeyer, L. D. Bourquin, L. L. Bauer and G. A. Reinhart (1995). “Dietary fiber for dogs: IV. In vitro fermentation of selected fiber sources by dog fecal inoculum and in vivo digestion and metabolism of fiber-supplemented diets.” J Anim Sci 73(4): 1099-1109.

Tôrres, C. L., V. C. Biourge and R. C. Backus (2022). “Plasma and Whole Blood Taurine Concentrations in Dogs May Not Be Sensitive Indicators of Taurine Deficiency When Dietary Sulfur Amino Acid Content Is Reduced.” Front Vet Sci 9: 873460.

van Rooijen, C., G. Bosch, A. F. van der Poel, P. A. Wierenga, L. Alexander and W. H. Hendriks (2014). “Quantitation of Maillard reaction products in commercially available pet foods.” J Agric Food Chem 62(35): 8883-8891.

Wasim, R., T. Mahmood, M. H. Siddiqui, F. Ahsan, A. Shamim, A. Singh, M. Shariq and S. Parveen (2022). “Aftermath of AGE-RAGE Cascade in the pathophysiology of cardiovascular ailments.” Life Sci 307: 120860.

Xiong, K., M. M. Li, Y. Q. Chen, Y. M. Hu and W. Jin (2024). “Formation and Reduction of Toxic Compounds Derived from the Maillard Reaction During the Thermal Processing of Different Food Matrices.” J Food Prot 87(9): 100338.

Zhang, J., Y. Ji, Y. Yang and Z. Wu (2025). “Grain-Free Diets for Dogs and Cats: An Updated Review Focusing on Nutritional Effects and Health Considerations.” Animals (Basel) 15(14).