Experimental Feeding Trial Introduces New Questions to Diet/DCM Observations
In this study, privately owned Huskies were fed in four treatments groups over 20 weeks and had serial evaluations via echocardiogram and blood parameters. To maintain control over the diet and extraneous variables, the Huskies were housed in a kennel facility for the duration of the study.
All four diets fed for the study were experiment-specific diets formulated by Champion Petfoods. Whole pulse ingredients were used at inclusion rates of 0%, 15%, 30%, and 45%. The whole pulse content was divided evenly (5%, 10%, and 15% inclusion) between peas, pinto beans, and a 50/50 mixture of chickpeas and lentils. All diets were chicken-based for protein. Pea starch was used at varying concentrations to balance the macronutrient content of the different diets. The diets were formulated to exceed AAFCO recommendations Methionine and Cysteine by > 60%. These amino acids are precursors to taurine, an important nutrient for the heart. Over the 20 week course of the feeding period, dogs received serial echocardiographic examinations, body composition assessments, amino acid measurements, cardiac biomarker measurements, and basic blood panels. No diet-dependent statistically significant relationships were observed within or between the groups over the 20 week period.
Pea starch, a pulse legume derived ingredients, was used in different amounts in the diets. The use of pea starch is justified in the study design portion, "Pea starch was chosen as the starch ingredient for all diets for balancing macronutrients to prevent potential confounding effects related to introducing additional ingredients that differed between the Control and Pulse diets. " The pea starch contributed primarily starch (~70%), but also protein (7-10%), and fiber (1-2%), to the diets. The highest inclusion of pea starch was in the Pulse15 diet, at 24.2%.
With no changes observed over the 5 month feeding period, does this suggest pulses are a red-herring in DCM cases? Not necessarily. The following points should be considered:
This study only included Siberian Huskies. As the authors acknowledge in the discussion, Huskies are a performance breed. As a genetically unique population of dogs, they may have differences in their metabolism that make them more resistant to the development of nutrient-mediated DCM than other breeds. It is worth noting that in the 2019 "data dump" from the FDA, no Huskies were noted among reported cases.
Any given pulse ingredient was only included at up to 15% in the diet. Previous studies have identified peas as the potential strongest candidate for a causative factor in these cases. Lentils, which are also frequently identified as a common factor in implicated diets, were only included at a maximum of 7.5%. It is possible that a diet containing a higher proportion of peas or lentils would lead to different results. Pinto beans and chickpeas may not be as much of a risk factor in the development of atypical DCM as peas and lentils.
The experimental diets in this study were formulated very intentionally. They also exceed several AAFCO minimum nutrient requirements. Commercially available diets using high amounts of pulse legumes may be very different from these experimental diets- they may not be as methodically formulated, and they may not exceed nutrient minimums. While it may be possible to formulate a diet with high amounts of pulse legumes safely, it doesn't mean that diets with high amounts of pulse legumes are currently formulated safely.
While 5 months is a long period to conduct a controlled feeding trial, it may be insufficient to fully capture long-term outcomes.
The authors state, "Previous clinical studies were unable to investigate whole pulse ingredients as the single variable causative factor; however, current data suggest that including pulse ingredients in dog food is not a causative factor of DCM in dogs who are not genetically predisposed." Given the considerations discussed above, this construction of the conclusion seems overreaching. However, later in the study discussion, more specific statements highlighting the limitations of the research, including the points raised above, are made.
There are several relevant conflicts of interest that must be acknowledged. As has been frequently stressed on this blog, conflicts of interest alone do not detract from study findings, but should be examined critically particularly in the context of the discussion and conclusions of a paper, or the way the paper is characterized in press releases.
These disclosures were made in full transparency in the paper as-published, in line with research ethics and expectations. "Christopher PF. Marinangeli is an employee of Protein Industries Canada and a former employee of Pulse Canada and Kellogg Canada." The mission of PIC is to "accelerate innovation and the competitiveness of the Canadian plant protein sector." Jennifer Saunders-Blades is Research and Nutrition Manager at Champion Petfoods. Darcia Kostiuk is the senior veterinarian at Champion Petfoods. Janelle Kelly is a Nutrition Research Scientist for Champion Petfoods. Anna Kate Shoveller serves on the Champion Petfoods Scientific Board. Champion Petfoods was one of the top named brands in the FDA's 2019 report on non-hereditary DCM.
Other disclosures related to nutrition, but not specifically DCM or the use of pulses in pet food, included "Adronie Verbrugghe is the Royal Canin Veterinary Diets endowed chair in Canine and Feline Clinical Nutrition and declares that they serve on Health and Nutrition Advisory Board for Vetdiets." and " Shoshana Verton-Shaw was previously employed by Hill’s Pet Nutrition, and her current position as a Registered Veterinary Technician in Clinical Nutrition is supported by Purina Pro Plan Veterinary Diets."
Further research is needed to characterize the long-term safety of pulse legume products in pet foods, specifically in the context of atypical dilated cardiomyopathy. Controlled, experimental studies such as this provide valuable additions to the body of knowledge surrounding this disease. Future studies could include additional breeds of dog, focus on individual pulse legumes at escalating inclusions, and use non-pulse starches in the control diets and for equalizing the starch content across diets. There may also be benefit to comparing experimental diets such as these to available commercial diets to assess for differences in total nutrient content and bioavailability, various biochemical compounds, and digestibility.
Ultimately, the findings of this study do not provide me as a clinician sufficient encouragement that the pulse-containing diets that are currently commercially available are safe for long-term feeding. I will continue to discourage pet owners from selecting diets with pulse legumes, especially peas and lentils, listed in the first 5-10 ingredients of a pet food. I hope that the pet food industry will continue to conduct further research on the use of various pulse legumes in dog food. I wish that such studies had happened before these ingredients were saturated into the market at high inclusion rates, but there is no putting the cart back in front of the horse. This study does demonstrate that a carefully formulated high-pulse diet that exceeds AAFCO minimum requirements for nutrients can be fed for 5 months without cardiac changes in Huskies.
It's worth noting that less than 1 month after this paper was published, another paper was published demonstrating that dogs diagnosed with DCM while eating non-traditional diets have a better prognosis if their diet is changed, highlighting the role of diet in this disease process.