Where the analysis done by Kronfeld on stress in dogs goes awry is in its implication that this conversion of protein reserves occurs during a mild or transient period of emotional turmoil. In point of fact, catabolism of proteins only begins after an extended duration of severe stress, as a consequence of an extreme condition like a long sled race or a bad infection. Therefore, a mildly stressed animal probably needs carbohydrates (and perhaps fats) far more than supplemental protein, since the former can be quickly and easily converted into bodily fuel. And in the case of the stress induced by mild illness, it is probably best to let the animal follow his natural instincts, and simply let him fast. The metabolism and consequent nutritional requirements of an animal in training--a racing sled dog differ both qualitatively and quantitatively from those of an animal subjected to sudden and unexpected stress, whether of the transient, mild or the more severe type. Finally, there may be real danger in feeding a dog more protein than his body actually needs. Recent studies have documented a strong causal link between regular supplementation of an animal's diet with heavy doses of protein and progressive kidney disease. And it is virtually certain that a large daily ration of protein will exacerbate any pre-existing kidney conditions, especially in older dogs, since the damaged organ is simply unable to handle the extra work required to excrete the additional levels of toxic byproducts of increased protein catabolism.
Murphy, D.H. (1983). Too much of a good thing: Protein and a dog's diet. International Journal for the Study of Animal Problems, 4(2), 101-107.