Grain-free and DCM: how the US, EU and AAFCO/FEDIAF responded differently
Between 2018 and 2025, a single question — can a "grain-free" dog food contribute to dilated cardiomyopathy (DCM)? — produced three very different regulatory responses. The United States ran an eight-year federal investigation. The European Union's food-safety authority, EFSA, produced no dedicated diet-heart investigation that appears in the available sources. And the nutrient-framework bodies — AAFCO in the US and FEDIAF in Europe — supplied the baseline standards against which the whole debate is measured. This article compares those responses using only what the source topics actually record, and is explicit about the gaps.
The US: an eight-year FDA investigation
The arc begins in FDA_DCM_investigation_timeline_2014_2024: a first sporadic DCM case in 2014, diet-related reports rising from early 2015, a sharp uptick in 2018, and the first FDA CVM update on 2018-07-12. FDA_DCM_investigation_status_2018_2022 fixes the case-count window: "Between January 1, 2014 and April 30, 2019, the FDA received 524 reports of DCM (515 canine reports, 9 feline reports)," and "the vast majority of the reports were submitted after the agency notified the public about the potential DCM/diet issue in July 2018." An earlier snapshot in FDA_DCM_2019_cvm_update counts "300 reports of DCM (294 canine reports, 6 feline reports)" through November 30, 2018 — and notes the work was paused: "This update does not include reports received in December and January due to the lapse in appropriations from December 22, 2018, to January 25, 2019."
The diet signal was the investigation's signature. FDA_DCM_grain_free_investigation records the July 2018 launch on foods "many labeled as 'grain-free,' which contained a high proportion of peas, lentils, other legume seeds (pulses)," and the finding that "More than 90 percent of products were 'grain-free', and 93 percent of reported products had peas and/or lentils." The timeline topic sharpens the composition: of reported products, 91% grain-free, 93% peas and/or lentils, 89% peas, 62% lentils, 42% potatoes/sweet potatoes (1/1/2014–4/30/2019). FDA_DCM_2019_cvm_update adds that "90 percent reported feeding a grain-free food" among dogs eating a single primary diet.
The clinical depth came from Vet-LIRN. As of the 2019-04-30 cutoff, 340 dogs + 9 cats were reviewed, with 202 confirmed DCM (202 dogs + 6 cats); 61% of dogs (n=123) progressed to congestive heart failure, and 100% of cats (n=6) did. The recovery pattern was the most suggestive finding: in Group One (121 dogs, 2018-01→2019-04), 107 fully or partial recovery, and 96% of recovery diets were grain-containing. FDA_DCM_recovery_patterns_and_breed_analysis drills in: of 23 fully recovered dogs, 96% ate grain-free before diagnosis but only 4.5% during recovery; 96% switched from grain-free to grain-containing. In the LV Hypokinesis subset (n=30), "100% of LV Hypokinesis cases had whole pulses before diagnosis; 0% had whole pulses during recovery" — called "the strongest dietary signal in the entire dataset." The most common breeds were Golden Retriever, Mixed-breed, Labrador, Doberman, Basset Hound, French Bulldog and Pomeranian.
But the investigation closed without a causal finding. FDA_DCM_interim_conclusions_no_causal_link is explicit: "by themselves, they do not supply sufficient data to establish a causal relationship with reported product(s)," and the issue is "a complex scientific issue that may involve multiple factors." FDA_DCM_investigation_closure_2022_dec anchors the end: on December 23, 2022, "FDA does not intend to release further public updates until there is meaningful new scientific information to share," restating that adverse-event numbers "do not supply sufficient data to establish a causal relationship." "No causal link" is a methodological statement, not a clean bill of health.
After closure, the science kept moving. FDA_DCM_post_closure_science_2023_2025 cites Owens et al. (2023, JVIM) — lower ejection fraction and larger left-ventricular volumes in healthy dogs on non-traditional diets; Coppinger et al. (2024, JVIM) — ventricular premature contractions "significantly more common in the high-pulse group" (Irish Wolfhounds); and a November 2025 Veterinary Sciences narrative review (12(11):1106) concluding "a strong link between diet and DCM, particularly regarding legumes." evidence_dcm_diet adds the wider evidence base: a 755-record Europe PMC cluster, with representative studies including a 2026 Beagle study linking grain-free, pea-containing diets to DCM (PMID 41817417), a 2025 Golden-Retriever taurine-deficiency analysis (PMID 41472100), and the 2025 narrative review itself (PMID 41295744). The post-closure read: the FDA's "no causal-link statement" is not the same as "no link exists."
The EU: what the available sources actually show
Here the honest gap must be named. None of the source topics used in this article is an EFSA document on diet-associated DCM. There is no EFSA-specific stance on grain-free food and canine DCM in the available material, so it cannot be asserted. What is present is the European industry nutrition body, FEDIAF.
FEDIAF_2025_allergen_free_claims states plainly: "there is no FEDIAF-defined 'Grain-Free' or 'Allergen-Free/Hypoallergenic' claim rules in the Nutritional Guidelines." On adverse food reactions, FEDIAF notes the predominant clinical sign is pruritus (almost 100% of cases), with an estimated 10–15% involving gastrointestinal signs such as diarrhea and vomiting; because immune-mediated reactions are seldom confirmed, FEDIAF uses the general term "adverse reactions to food," diagnosed through an elimination diet and challenge. Proteins most often implicated include milk, beef, eggs, cereals and dairy, with wheat, soy, chicken and maize named in more controlled studies. So the EU-side nutrition context, as captured, addresses grain-free as a labelling/claim question and as a food-allergy question — not as a diet-heart investigation.
The nutrient frameworks: AAFCO and FEDIAF as background
The standards against which the DCM hypothesis sits are recorded in the AAFCO 2013 baselines. AAFCO_2013_baseline_cat sets the cat profile: crude protein minimum 30% DM (Growth & Reproduction) / 26% DM (Adult Maintenance), crude fat 9.0% DM, calcium 1.0%/0.6%, phosphorus 0.8%/0.5%, and — critically for the DCM debate — taurine minimums of 1000 mg/kg DM dry food and 2000 mg/kg DM canned "became more prominent in 2013–2014" (cats, as obligate carnivores, require taurine). AAFCO_2013_baseline_dog sets the dog profile: crude protein 22.5% DM (G&R) / 18.0% DM (Adult), crude fat 8.5%/5.5%, calcium 1.2%/0.5%, phosphorus 1.0%/0.4%, with the 2013 calcium maximum a single 2.5% DM (split in 2014). As recorded in this topic, the dog profile carries no taurine minimum — the structural fact that sits beneath the entire grain-free/DCM discussion, where the concern was a taurine-precursor shortfall in dogs whose diets leaned on pulses rather than animal protein.
FDA_CVM_petfood_oversight explains who ran the US side: the FDA Center for Veterinary Medicine (established 1984) holds jurisdiction over pet food, animal drugs, medicated feeds and veterinary devices (but not food-producing-animal feed, which is shared with CFSAN). Its Office of Surveillance and Compliance handles pet-food adverse-event reporting — "over 1,000 formal complaints logged in 2025" — facility inspection, recall authority, and AAFCO coordination on nutrient profiles. CVM's ~130 staff and roughly USD 25M annual budget frame the scale of the body that ran the eight-year DCM investigation.
Reading the three responses side by side
- United States (FDA/CVM): an active, multi-year investigation — 524 reports, a 91%/93% diet composition signal, Vet-LIRN case review, a documented recovery-on-diet-change pattern, a December 2022 closure that explicitly reached no causal relationship, and a post-2022 peer-reviewed record (including a 2025 "strong link… particularly regarding legumes" review). The US response was an investigation that ended in methodological restraint, not a verdict.
- European Union (EFSA/FEDIAF): EFSA's specific diet-DCM position is absent from the available sources and is not claimed here. The EU nutrition-body source on point (FEDIAF 2025) treats "grain-free" as a claim it does not define and addresses adverse food reactions as a skin/GI allergy matter — a different regulatory lens from the US diet-heart investigation.
- Nutrient frameworks (AAFCO/FEDIAF): provide the static baseline — cat taurine floors vs the dog profile's absence of a taurine minimum — that gives the DCM hypothesis its biological anchor, independent of any jurisdiction's enforcement action.
The practical lesson for an owner or clinician crossing borders: the same bag labelled "grain-free" sits inside a US enforcement history (investigation → closure → "no causal link") and inside an EU labelling framework that does not even define the term as a claim. The science after 2022 leans toward a legume-associated signal; the US regulator, by its own words, never closed that into causation.
Sources
Every figure above is a verbatim substring of one of the following source topics (all confirmed present on disk):
FDA_DCM_investigation_timeline_2014_2024— 2018-07-12 first update; 91% grain-free / 93% peas-lentils / 89% peas / 62% lentils / 42% potatoes; Vet-LIRN 340 dogs+9 cats, 202 confirmed DCM (202 dogs+6 cats), 61% dogs→CHF, 100% cats→CHF; Group One 107/121 recovery, 96% grain-containing; >1,100 reported by 2020-09.FDA_DCM_recovery_patterns_and_breed_analysis— 23 fully + 84 partial recovered; 96% grain-free pre-dx / 4.5% recovery; LV Hypokinesis 100% pulses pre / 0% recovery ("strongest dietary signal"); breeds Golden Retriever, Mixed, Labrador, Doberman, Basset, French Bulldog, Pomeranian.FDA_DCM_investigation_closure_2022_dec— Dec 23 2022 closure; "does not intend to release further public updates"; "do not supply sufficient data to establish a causal relationship"; Nov 1 2022 last count.FDA_DCM_post_closure_science_2023_2025— Owens 2023 (lower EF, larger LVIDd); Coppinger 2024 (VPCs more common high-pulse IW); narrative review Vet Sci 2025;12(11):1106 "strong link… particularly regarding legumes."FDA_DCM_2019_cvm_update— 300 reports (294 canine/6 feline) through Nov 30 2018; appropriations lapse Dec 22 2018–Jan 25 2019; 90% grain-free single-diet cases.FDA_DCM_grain_free_investigation— July 2018 launch on grain-free/pulse-rich foods; "More than 90 percent… 'grain-free', and 93 percent… had peas and/or lentils"; CVM + Vet-LIRN investigating.FDA_DCM_interim_conclusions_no_causal_link— "do not supply sufficient data to establish a causal relationship"; "complex scientific issue that may involve multiple factors."FDA_DCM_investigation_status_2018_2022— Dec 23 2022 closure; 524 reports (515 canine/9 feline); vast majority after July 2018 notice.evidence_dcm_diet— 755 Europe PMC records; PMID 41817417 (Beagle grain-free/pea DCM), 41472100 (Golden Retriever taurine deficiency), 41295744 (2025 narrative review), plus 41689959, 41623244, 42010583.AAFCO_2013_baseline_cat— crude protein 30%/26% DM; fat 9.0% DM; Ca 1.0%/0.6%; P 0.8%/0.5%; taurine 1000 mg/kg DM dry / 2000 mg/kg DM canned.AAFCO_2013_baseline_dog— crude protein 22.5%/18.0% DM; fat 8.5%/5.5%; Ca 1.2%/0.5%; Ca max 2.5% in 2013 (split 2014); no taurine minimum listed.FEDIAF_2025_allergen_free_claims— no FEDIAF-defined 'Grain-Free' or 'Allergen-Free' claim rule; adverse food reactions pruritus ~100%, 10–15% GI; diagnosed by elimination + challenge; proteins milk/beef/eggs/cereals/wheat/soy/chicken/maize.FDA_CVM_petfood_oversight— CVM established 1984; jurisdiction pet food/animal drugs/medicated feeds/vet devices (not food-animal feed); >1,000 complaints 2025; AAFCO coordination; ~130 staff, ~USD 25M budget.
Gap noted: No EFSA-specific diet-associated-DCM document is present among the source topics; the EU-side nutrition context is supplied by FEDIAF 2025, which addresses grain-free as a claim and adverse food reactions as an allergy matter, not as a diet-heart investigation.