An open access analysis titled, “Investigation of diets associated with dilated cardiomyopathy (DCM) in dogs using foodomics analysis,” garnered national press in early August, 2021. This is fairly uncommon for research studies. We have a few questions for the study’s researchers.
#1. According to the paper, “The study’s primary objective, therefore, was to apply a metabolomics approach to identify biochemical compounds that differ between commercial dog foods that have been associated with canine DCM and in more traditional commercial dog foods.”
Considering the Dunham-Cheatham et al. (2021) research study that found, “Despite pet foods compositions being simplistic and consistent, many of the tested foods were adulterated. Here, the term adulteration is defined as foods that included DNA from animals or plants that were not listed in the package ingredient list as well as foods that did not include DNA from animals or plants that were listed in the package ingredient list. These results are consistent with those from Palumbo et al. (2020), that showed 16 of the 18 tested commercial pet foods were adulterated.”
Before the actual metabolomics analysis, were all of the 18 diets selected for the study first analyzed for ingredient accuracy using DNA metabarcoding?
#2. You mention corn as an oft-used ingredient in traditional commercial grain-containing dog food diets.
Why was it not included as a comparison along with the rice, peas and lentils instead of chicken/turkey?
#3. You mention, “Deficiencies of a variety of nutrients, such as thiamine, magnesium, choline, vitamin E and selenium, have been associated with DCM in humans or animal models.” Then note, “Mineral supplements and probiotics were excluded from ingredients analysis.”
If deficiencies of the minerals selenium and magnesium have been associated with DCM, why were they not included in the analysis?
#4. In the paper, you state, “Seven of the 8 vitamins that were significantly lower in 3P/FDA diets were B vitamins: pyridoxine (vitamin B6), thiamine (vitamin B1), folate (vitamin B9), pantothenate (vitamin B5), and riboflavin (vitamin B2) were lower in 3P/FDA diets (Table 2).”
Upon review of Table 2, we noticed you provided the statistical difference – rather than the level present for each vitamin in each food.
Why were the levels of B vitamins not provided for all of the diets analyzed?
#5. Only two food samples were analyzed from each selected product, and not all were obtained from different bags of food (some samples were from a single bag).
Could this have biased the analyses?
#6. One author selected the diet samples and some were obtained directly from owners of dogs with DCM; most of these foods had been opened and stored for variable periods of time, first by the owner and then by the investigators.
Surely, this introduced variability could bias the data if certain nutrients were degraded during storage conditions. Please clarify.
#7. Dovetailing off Questions 4 through 6, did all of the foods meet Association of American Feed Control Officials (AAFCO) minimum standards for pyridoxine, thiamine, folic acid (the synthetic form of folate), pantothenic acid (pantothenate), and riboflavin?
#8. We focus here on folate as an example. According to the analysis, folate was correlated with the chicken/turkey diet. According to the USDA FoodData Central, chicken breast has 3 micrograms per 100 grams vs. chicken liver which has 588 micrograms in raw form and 578 micrograms when cooked. Turkey breast has 7 micrograms, but the liver has 677 micrograms (raw) and 691 micrograms (cooked).
However, chicken/turkey had a lower P value than – even though it was still positively correlated – than the rice diet according to the study. Unenriched white rice only has 7 micrograms of naturally occurring folate. Rice is often enriched with folic acid. In cooked form, the enriched rice has 58 micrograms of folate.
Peas (or the diets high in peas – we were not sure if the categories in Figure 2 represent the diets or the ingredients) had a negative association with folate. Granted, cooked peas appear to lose some folate during the cooking process with 57 micrograms per 100 grams cooked as compared to raw green peas at 65 micrograms.
Logic would have it that if turkey or chicken liver were included in the diets that they would have a stronger positive correlation with folate compared to the other two categories. What parts of the turkey and/or chicken were included in the diets?
Secondly, was the rice enriched? If so, logically it would be equal to the peas in folate amount unless more rice was used in proportion to the rest of the recipe than the pea recipes?
We quickly looked at two grain-free kibble dog food bags (one with peas and one with lentils) and read that folic acid was added as a supplement. We do not know if one of these diets was included in your analysis.
Was folic acid added to any of the diets analyzed in your study? If so, which ones? Was that supplementation included as a contributor to the level of folate present in the diet?
Maybe these are the right questions to ask, maybe they are not. Maybe we missed something. After reading the paper a few times, we still would like some clarification.
Finally, we are not saying this research is incorrect. However, when we read headlines such as, “New ‘piece of the puzzle’: Why some dog foods may be linked to deadly heart disease,” we do not want companion dog parents making potentially erroneous decisions based on a study funded in part by a large pet food company that has a vested interest.
Yes; we have mentioned studies funded by brands that were implicated by the FDA as producing products that were associated with causing DCM. However, they never received the national press attention of the present study.
In any event, everyone can agree that DCM in dogs is a complicated condition with both known and unknown variables.
Remember, if you’ve stopped feeding grains to your companion dog, think back to the many reasons why you stopped. It could be to prevent leaky gut syndrome, to help curb food sensitivities or intolerances to a particular grain, to maintain optimal weight in your dog, etc.
If you are worried, we understand. We suggest having your veterinarian take a blood sample to measure the methionine, cysteine and taurine levels in both whole blood and plasma, and send it to a diagnostic laboratory (University of California – Davis or University of Wisconsin) experienced with the appropriate reference ranges for circulating taurine, or have an echocardiogram performed. If the levels are lower than normal for dogs, please discuss the appropriate next steps with your veterinarian.
Next week, stay tuned for more commentary on the FDA role in the matter of DCM.