Food Safety

Can Dogs Eat Grapes or Raisins? How Many Is Dangerous

One grape per 10 lb of dog is the only threshold anyone publishes - and it is a threshold, not an allowance.

Grapes are the food that broke the usual rules. There is no dose that vets can call safe, the toxic ingredient is still only a proposal rather than a proven fact, and two dogs of the same size can eat the same handful with completely different outcomes. What is not in doubt is the consequence when it goes wrong: acute kidney failure, and once a dog stops producing urine, a prognosis that the MSD Veterinary Manual describes bluntly.

The quick answer
  • No. Grapes, raisins, sultanas, currants and tamarind are all on the same MSD topic, and all are implicated.
  • MSD's one published threshold: more than one grape or raisin per 4.5 kg (10 lb) of bodyweight may carry enough tartaric acid to pose a risk.
  • That is a threshold, not an allowance. It is the line above which risk is recognised, not a quantity anyone has shown to be safe below.
  • Colour, seeds and organic status make no difference, and neither does cooking.
  • Cats: anecdotal reports only, no published cases. That is not the same as "safe" - see below.
  • Call your vet the same day. Kidney failure develops over 24-72 hours, and treatment works best before it starts.

Why grapes are dangerous, and why nobody could say why for twenty years

For most of the time that grape toxicosis has been recognised, the mechanism was a genuine mystery. Pesticides, moulds, heavy metals and mycotoxins were all investigated and none of them explained it. That mattered practically, not just academically: without knowing the toxic principle, nobody could work out a dose.

The current best candidate is tartaric acid, and MSD is careful about how it says so: "One proposed toxic principle in these fruits is tartaric acid." Proposed. It fits - grapes and tamarind are both unusually high in tartaric acid, and dogs appear to be poor at clearing it - but it has not been settled, and you should treat anyone who tells you the question is closed with some caution.

What the tartaric acid hypothesis did give the profession is the first usable rule of thumb, and it is the only number MSD publishes:

The only published threshold

"More than one grape or raisin per 4.5 kg (10 pounds) of body weight may contain enough tartaric acid to pose a risk."

Read that sentence carefully, because it is easy to read backwards. It says that above this amount there is a recognised risk. It does not say that below it a dog is fine. Tartaric acid content varies enormously between grape varieties, growing conditions and ripeness, so the same number of grapes is not the same dose twice.

What that works out to, dog by dog

Applying MSD's single sentence to some common weights. This table is our arithmetic on that one published threshold - MSD does not print it, and it exists only to show how startlingly small the numbers are:

Dog Rough weight MSD's threshold is more than
Chihuahua, Yorkshire terrier 3 kg Under one grape
Miniature dachshund, Pomeranian 5 kg About one grape
French bulldog, cocker spaniel 12 kg About two or three grapes
Border collie, springer spaniel 20 kg About four grapes
Labrador, German shepherd 32 kg About seven grapes
Great dane, mastiff 60 kg About thirteen grapes

A Labrador is a large, robust animal, and seven grapes is a rounding error in its day. That is the entire problem with this food: the quantity that matters is far below the quantity anyone would think to worry about, and well below the quantity that gets left on a coffee table.

Raisins deserve their own paragraph

MSD's threshold is written per fruit - "one grape or raisin" - not per gram. A raisin is a grape with most of the water taken out, so gram for gram it carries far more of everything the grape contained. If the threshold really is per fruit, a small box of raisins is a much larger exposure than the same weight of grapes, and a small box of raisins is a very ordinary thing for a dog to find in a lunchbox, a cereal bowl, a hot cross bun, a scone or a Christmas cake.

This is the everyday version of grape poisoning. It is rarely a dog eating a bunch of grapes off the vine.

What happens, and when

MSD sets out a two-stage timeline, and the gap between the stages is what catches owners out.

  • 6-12 hours: "Most dogs with grape, raisin, or tamarind toxicosis develop vomiting or diarrhea within 6-12 hours after ingestion." This is the stage that looks like an ordinary upset stomach, and it is why so many of these cases arrive at the vet late.
  • 24-72 hours: renal failure develops. Alongside it, MSD lists lethargy, anorexia, polydipsia (drinking much more than usual), dehydration, and rising serum creatinine and BUN on bloodwork.

The important point is that the second stage is not announced by the first being dramatic. A dog who vomits once, then seems brighter, is not a dog who has got away with it.

Why vets treat this one so aggressively

MSD on prognosis: "once anuric renal failure develops, most dogs die or are euthanized", and "Anuric dogs are unlikely to survive unless peritoneal dialysis or hemodialysis is initiated; even then, the prognosis is guarded."

Anuric means the kidneys have stopped producing urine. Veterinary dialysis exists in a small number of referral centres and is expensive and not always available. This is why the recommended treatment is early and hard rather than watchful: MSD calls for prompt GI decontamination followed by aggressive IV fluid therapy for a minimum of 48 hours. Both of those are far easier to do well on day one than on day three.

What to do if they've eaten some

  1. Ring your vet now, even if it was one grape and even if your dog is bouncing around. The treatment that changes the outcome is time-limited, and "wait and see" spends exactly the window in which it works.
  2. Say what, how much, and when. Grapes, raisins, sultanas, currants and the dried fruit inside a baked product are all different exposures, and so is "about four" versus "possibly the whole box". If you're not sure how many, say that rather than guessing low.
  3. Take the packaging. For a baked item it is often the only way to tell how much dried fruit was in it.
  4. Don't try to make your dog sick at home. MSD lists ropinirole, apomorphine and hydrogen peroxide among the emetics used, but inducing vomiting is a clinical decision with real risks and the drugs used are veterinary ones. Let the practice make that call.
  5. Go even if it has been a day or two. Fluid therapy still helps once kidney values are rising, and the history is what points your vet at the right diagnosis quickly.

In the UK, Animal PoisonLine (01202 509 000) is a 24-hour service for owners, run by the Veterinary Poisons Information Service; there's a per-call fee. In the US, the ASPCA Animal Poison Control Center is on (888) 426-4435, 24 hours, also for a fee.

What about cats?

Here we have to be honest about the limits of the evidence, because the honest answer is more useful than a confident one. MSD's topic is titled Toxicosis in Dogs, and on other species it says: "anecdotal reports exist of renal failure in cats and ferrets after ingesting grapes or raisins, although there are no published case reports in these species."

So: reports exist, published evidence does not. That is not a clean bill of health, and it is not a reason to relax. It does mean that anyone quoting you a feline toxic dose for grapes is quoting something that has not been established. In practice most cats show no interest in grapes at all, which is probably why the case reports are thin. If your cat has eaten grapes or raisins, ring your vet and describe it - the same as you would for anything unknown.

The things that make no difference

  • Colour. Red, green and black have all been involved.
  • Seedless or seeded. Neither the seeds nor the skin have been shown to be the problem.
  • Organic, home-grown or shop-bought. The pesticide theory was investigated and did not hold.
  • Cooking or baking. A raisin in a scone is still a raisin.
  • Having eaten them before without a problem. This is the most dangerous false reassurance of the lot. Susceptibility appears to vary between individual dogs, and a dog that tolerated grapes once has not been proven immune.

Not sure about something else your pet got into?

Check any food instantly in the Supernormal app's food safety checker - or try the Pet Food Safety Checker for the foods we've covered so far.

Get the app

Related guides: Pet Food Safety Checker · How Much Chocolate Can a Dog Eat? · Can Dogs Eat Onions or Garlic? · Can Dogs Eat Eggs?

Frequently asked questions

How many grapes can a dog eat?
None that anyone can call safe. The only threshold the MSD Veterinary Manual publishes is that more than one grape or raisin per 4.5 kg (10 pounds) of bodyweight may contain enough tartaric acid to pose a risk. For a 10 kg dog that is roughly two grapes. It is the point above which risk is recognised, not a quantity shown to be safe below it - tartaric acid content varies between varieties and growing conditions, and susceptibility appears to vary between individual dogs.
Are raisins worse than grapes for dogs?
Almost certainly, in practice. MSD states its threshold per fruit - "one grape or raisin" - rather than per gram, and a raisin is a grape with most of the water removed. So the same weight of raisins is a far larger number of fruit. It also matters that raisins turn up in things dogs actually get hold of: cereal, lunchboxes, hot cross buns, scones, flapjacks and Christmas cake.
What happens if a dog eats grapes?
It arrives in two stages. MSD reports that most affected dogs develop vomiting or diarrhoea within 6-12 hours, which looks like an ordinary stomach upset. Renal failure then develops 24-72 hours after exposure, with lethargy, loss of appetite, increased thirst, dehydration and rising creatinine and BUN on bloodwork. A dog who seems brighter after the first stage has not necessarily escaped the second.
My dog ate one grape - should I worry?
Ring your vet and describe it, whatever the number. For a small dog, one grape is already at or above MSD's published threshold. For a large dog it is below it, but "below the threshold" is not the same as "shown to be safe", because that threshold was derived from the tartaric acid hypothesis rather than from a dose at which dogs have been proven unharmed. The call costs nothing and the treatment window is short.
Are grapes toxic to cats?
The honest answer is that nobody knows. MSD's topic covers dogs, and on other species says anecdotal reports exist of renal failure in cats and ferrets after eating grapes or raisins, although there are no published case reports in these species. Reports without published evidence is not a clean bill of health. Most cats show no interest in grapes, which is probably why the evidence is thin. If your cat has eaten some, ring your vet.
Does it matter whether the grapes were seedless, organic or cooked?
No. Colour, seeds, skin, organic status and cooking have all been looked at and none of them explains the toxicity. Pesticides, moulds, heavy metals and mycotoxins were all investigated and ruled out, which is why the mechanism was unknown for so long. A raisin baked into a scone is still a raisin.
My dog has eaten grapes before and was fine. Is he immune?
No, and this is the most dangerous piece of false reassurance around this food. Susceptibility appears to vary between individual dogs, and a dog who tolerated grapes once has not been shown to be protected next time. The amount, the variety and the individual animal all vary.
What will the vet do?
MSD's recommended treatment is prompt GI decontamination, followed by aggressive IV fluid therapy for a minimum of 48 hours, with kidney values monitored. That is deliberately front-loaded, because the prognosis once anuric renal failure has developed is poor: MSD states that most such dogs die or are euthanised, and that anuric dogs are unlikely to survive without dialysis, which is available at few referral centres and carries a guarded prognosis even then.