I think your/the hypothesis fails...take certain species of jellyfish that are biologically immortal, they don’t have blood at all.
Coincidentally they are considered top of the food chain, so another knock to the hypothesis.
Then that would be an evolutionary in-spec configuration, since Cnidaria-phylum species have evolved under selection pressure for hundreds of millions of years in that "never losing blood, because it doesn't have any" configuration.
For a similar reason, you wouldn't expect most plants to be troubled/benefited by the loss of solutes (= sap); a maple tree should probably live about the same lifetime whether it's tapped or not, because maple trees evolved mostly without predation. But for plants that have evolved over eons entirely under conditions of predation (e.g. where they're a primary ecological water source, like cacti in the desert—observable by the fact that they possess defenses to predation), you would expect a difference in longevity between those that lose transported solutes to predation, vs. those who don't.
In either case, the hypothesis suggests animals should live longer in the wild than in captivity. So let’s take the only mammal that isn’t more likely to die from age/don’t die from old age, the naked mole rat. We don’t understand why they die and any given day there is a 1 in 10,000 chance they die at any age. In captivity they should have the same lifespan as lab rats (6 years); however, in the wild the have an average lifespan of 17 years and in captivity that expands to 30 years (again in contrast to the hypothesis where wild animals of the same species should live longer than those in captivity).
But what if you also change the oil?
If I remember correctly, it is beneficial for your health to donate blood if you have some predisposition for Fe accumulation (hemochromatosis in the severe form.
TLDR: There are correlations that it reduces circulating iron, circulating cancer cells, arterial hardening, decrease insulin sensitivity, and it increased metabolism. Not a lot of super firm evidence though
A testable hypothesis for this: if you count only the animals that survive long-enough to die by degenerative disease, then animals that get regularly bloodied by their predators or prey species should have longer average lifespans; while animals who don't (e.g. because they have no natural predators; because they're herbivores; because they prey only upon species with no violent defenses) should have shorter average lifespans. This shouldn't be purely a per-species effect, but should also be seen in a lifespan difference between e.g. wild and zoo animals, if the wild animals engage in bloody conflict that the zoo animals don't.
Does anyone know if that hypothesis holds up at all? (Maybe/especially if the question is phrased in terms of heartbeats† rather than years?)
† http://robdunnlab.com/projects/beats-per-life/