Are grapes toxic to dogs? What the tartaric acid evidence actually is, why no safe dose exists, and where kishmish sits in an Indian kitchen

Are grapes toxic to dogs, and what the answer actually rests on

Grapes are toxic to dogs on the evidence available, and the honest version of that sentence is longer than any ranking page prints it. Grapes, raisins, Zante currants, sultanas and tamarind have all been followed by acute kidney injury in dogs, and the substance believed to do it was only proposed in 2022.

Start with what the veterinary reference actually says, because it is more careful than the pages quoting it. Ingestion of grapes, raisins and Zante currants, all members of the Vitis genus, has occasionally been associated with the development of renal failure in dogs, and tamarind ingestions have also resulted in renal failure. True currants of the genus Ribes, a different plant that shares the English name, have not been associated with renal injury. One proposed toxic principle in these fruits is tartaric acid, which dogs excrete poorly because they lack the organic acid transporters other species have, allowing it to accumulate in the proximal renal tubular cells.

Read the word proposed. For roughly twenty years after the first report in 2001 the toxic agent was simply unknown, and pesticides, mycotoxins, herbicides and vitamin D were all screened for and not found. Tartaric acid is the first candidate that has survived scrutiny, and the case for it now rests on three separate strands of work.

The three strands the tartaric acid hypothesis stands on

Each strand answers a different question, and none of them is a dosing trial in living dogs.

The cell work is the piece most people have not caught up with, and it is the strongest. It found that tartaric acid is a species-specific nephrotoxicant in dogs because of an interspecies difference in organic anion transporter-4 expression, and that transfecting the human transporter into canine cells reduced the damage by 57 per cent. That is a mechanism with a named protein behind it, which is a long way from where this subject sat in 2019.

Four things follow from all of that about what actually counts as a grape for this purpose.

• In, at any quantity. Fresh grapes of any colour, raisins, kishmish, munakka, sultanas and Zante currants, which are all Vitis fruit.
• In, at a far higher concentration. Tamarind pulp, at 8 to 18 per cent tartaric acid against 0.35 to 1.1 per cent in grapes, and cream of tartar, which is the potassium salt itself.
• Out, on current evidence. Grape juice, grape jam, grape seed oil, vine leaves and wine, none of which has been associated with toxicosis, which two sources attribute to tartrates being removed in processing.
• Out, because it is a different plant. True currants of the genus Ribes, which share an English name with Zante currants and nothing else.

One practical consequence falls out of the chemistry immediately. Tartrates are often removed, or detartrated, from commercial products such as grape juice, jam and wine, which the authors offer as the explanation for why toxicosis has not been reported after those products. Cornell's canine health centre says the same thing, listing grape juice, jelly, leaves, grape seed oil and wine as not associated with toxicosis, likely because processing removes or reduces tartaric acid. That is not permission to pour juice into a bowl. It is a clue about what the toxin is attached to, and it matters again further down this page.

Why there is no safe dose, and what every published number really is

No safe dose exists because the published case literature shows no reliable relationship between how much a dog ate and how badly it was hurt. Several numbers circulate anyway, and they are four different kinds of number doing four different jobs.

The figure almost every page repeats is the veterinary reference's action line. In general, more than one grape or raisin per 4.5 kg of body weight may contain enough tartaric acid to pose a risk of renal effects in dogs, and the manual prints it immediately after noting that the tartaric acid content varies, resulting in variable effects following ingestion. It is a precautionary threshold above which risk stops being treated as negligible. It is not a prediction, and it is not what has been observed to harm dogs.

What has been observed is a different set of numbers, and they sit far higher.

What those numbers mean for a 10 kg dog

Run the two lowest observed doses against a 10 kg dog and the gap becomes obvious. The lowest raisin dose ever reported to have caused kidney injury works out at 28 grams of kishmish, which is a small handful. The lowest grape dose works out at 196 grams, which is most of a bunch. Both figures are Bonbeno's arithmetic on El Bahri's published g/kg values and nothing else. Against those, the action line for the same dog is a third grape.

The ratio between the two observed figures is worth a second look. Grapes and raisins differ by a factor of seven, 19.6 against 2.8, while ICAR-NRCG measures raisin recovery at about 25 per cent, meaning roughly 4 kg of grapes makes 1 kg of kishmish. Drying concentrates the fruit about fourfold. The published raisin figure is more conservative than drying alone accounts for, which is consistent with raisins being the form dogs are most often reported eating rather than with raisins being chemically worse than the arithmetic says.

Why your dog eating grapes last year tells you nothing

A previous exposure without harm is not evidence of tolerance, and the literature is unusually blunt about it. Eubig and colleagues found no relationship between survival and the dose of grapes or raisins ingested, and noted that acute kidney injury does not consistently develop at dosages similar to or higher than those in their own report. Some dogs have remained free of signs after as much as 1 kg of raisins while others died after a handful. Cornell puts it in one line for owners, that the exact amount of fruit that causes toxicity is unknown and some dogs may become sicker than others.

The idiosyncrasy was visible in the laboratory work too. Experiments in rabbits found significant individual variation in response to tartrates at low dosages, and that there was not a direct correlation between dosage and degree of renal damage, although large dosages invariably caused marked effects. Variable tartaric acid content between cultivars, growing conditions and ripeness explains part of the inconsistency, and individual sensitivity appears to explain the rest Cornell University College of Veterinary….

How often ingestion actually harms a dog, and why two true answers look opposite

Grape ingestion harms a small minority of dogs and kills very few, and grape toxicosis kills about half the dogs who develop it. Both statements are correct, and the difference between them is not biology. It is which dogs each study counted.

Four published series have measured this, and read side by side they look like a contradiction.

The selection rule that explains the whole table

Two of those studies started with dogs that ate grapes and asked what happened next. Two started with dogs that already had kidney failure and asked what they had eaten. A study built the second way cannot produce a low injury rate, because injury was the entry ticket. Quoting its mortality figure as the risk of eating a grape is the single commonest error on this subject, and it runs through the ranking pages.

The numbers that answer an owner's actual question are the first two rows. Across 606 UK dogs that ate grapes, raisins, currants or sultanas, every one of the 338 who had creatinine or urea measured came back inside the reference interval, and all 606 survived to discharge. Across 139 US dogs with confirmed ingestion, kidney injury appeared in 6.7 per cent and one dog died. A fifth series, a Dutch poisons-centre cohort of 95 dogs, points the same way, and its figures are set out on our Diwali sweets page rather than repeated here.

Three limits keep that from being reassurance. The whole-ingestion cohorts are dogs whose owners noticed and acted, so they measure the risk after a clinic visit rather than the risk of a grape. The UK authors state their own finding conditionally, that there was no evidence of significant kidney injury in 32 of 33 cases where supportive measures were introduced and repeated creatinine assessments were performed. In the US series, 122 of 139 dogs, or 88 per cent, underwent gastrointestinal decontamination, and the authors say plainly that the retrospective design means no conclusion about whether that helped can be drawn.

What happens inside a dog, hour by hour, and where the sources disagree

Vomiting comes first, usually within 6 to 12 hours, and kidney failure declares itself over the following one to three days. The published timelines do not quite agree with each other, and the disagreement is small enough to publish and large enough to matter when you are watching a clock.

Most dogs with grape, raisin or tamarind toxicosis develop vomiting or diarrhoea within 6 to 12 hours of ingestion, with lethargy, anorexia, abdominal pain, weakness, dehydration, polydipsia and tremors also reported. Cornell gives the window as 6 to 24 hours and adds drooling and abdominal distension to the early list. On the failure itself the same veterinary reference contradicts its own page: its summary says progression to oliguric or anuric renal failure may occur within 24 to 48 hours, while its clinical findings section says 24 to 72 hours and its pathogenesis section says generally within 72 hours. Take the widest of those and watch for three days.

Two laboratory details change what a first blood test means. Serum creatinine tends to rise before urea nitrogen and rises disproportionately higher than it. Hypercalcaemia and a high calcium times phosphorus product are characteristic: they were present in 62 per cent and 95 per cent of the dogs in whom they were measured in the ASPCA series, and extremely high values were negative prognostic indicators. In the Bern series the raised total calcium was the single most distinctive feature separating grape cases from other causes of kidney injury.

Three windows are worth watching at home, alongside whatever the clinic has arranged.

• The first 6 to 24 hours. Vomiting, diarrhoea, drooling, loss of appetite and lethargy, with fruit sometimes visible in the vomit for up to 12 hours. Vomiting was the single most consistent sign in every published series and occurred in all 43 dogs in the ASPCA cohort.
• Day one to day three. Increased thirst and urination, weakness or incoordination, swelling of the limbs, trembling or tremors. Decreased urine output, ataxia and weakness were each associated with a worse outcome in the ASPCA series.
• Any point, and treat it as an emergency. A dog that stops producing urine. Anuric renal failure is the point at which most dogs in the published series died or were euthanased.

The neurological signs most pages never mention

A Swiss referral series found that severe neurological signs are a major feature of this poisoning, which almost no owner-facing page carries. Of 15 dogs with grape or raisin toxicosis, 11, or 73 per cent, had marked forebrain, cerebellar or vestibular signs at presentation, against only 2 of 74 control dogs with kidney injury from other causes. The findings included impaired consciousness, generalised ataxia in nine dogs, tremors, head tilt, drifting and hypermetric gait.

What makes it stranger is what was not found. The signs were not associated with the severity of the azotaemia or with systemic hypertension, neuropathological examination of the brain and spinal cord was unremarkable in all four dogs examined, and every affected dog that survived recovered neurologically in full, with the signs regressing over 2 to 20 days. The authors describe a reversible functional brain injury specific to grape toxicosis and unrelated to uraemia.

Here the sources part company and we publish the split rather than pick. The Vet Times clinical review attributes the neurotoxicity to hypercalcaemia blocking sodium movement through voltage-gated channels. The Bern authors looked for exactly that kind of association and found none between laboratory values and neurological signs. One is a proposed mechanism, the other is a negative finding in the only series that measured it.

Where this page stops and the clinical pages start

Acute kidney injury in a dog has several causes and grapes are a minority of them, which is why a vet will not simply accept your account and stop thinking. In the Bern hospital, 54 of the 74 control dogs with kidney injury, 73 per cent, had leptospirosis. That is the differential an Indian vet is weighing too, and our page on leptospirosis in dogs in India covers it properly. Ethylene glycol and cholecalciferol are the other classic differentials the reference names.

The first hour belongs to a different page and its conclusion is short. Do not make your dog vomit. That is a veterinary decision, taken on the individual animal, and our guide to foods toxic to dogs sets out the published emesis evidence, the contraindications and what to take with you. One physiological fact is worth carrying there: gastric emptying of tartaric acid is delayed in dogs, which is the likely reason grapes and raisins have been recovered in vomit as late as 8 and 13 hours after ingestion. Late is not too late to be worth the drive.

Where kishmish actually sits in an Indian kitchen

Kishmish sits in far more Indian homes than fresh grapes do, because India dries about 27 per cent of its entire grape crop into raisins. The dried fruit is the Indian form of this hazard, and it is a year-round resident of the kitchen rather than a seasonal visitor.

ICAR's National Research Centre for Grapes at Manjri Farm, Pune, publishes the structure of the trade. Of roughly 2,980 thousand tonnes of grapes produced from 138 thousand hectares, about 71 per cent is eaten fresh, nearly 27 per cent is dried for raisin production, 1.5 per cent goes to winemaking and 0.5 per cent to juice. About 95 per cent of the grapes come from Maharashtra and Karnataka, with raisins made mainly in the Sangli, Solapur and Nashik districts of Maharashtra and the Vijayapura and Bagalkot districts of Karnataka. Tamil Nadu and Andhra Pradesh make up most of the rest of the growing area.

The variety matters for a reason that is not obvious. Growers dry Thompson Seedless and its clonal selections such as Tas-A-Ganesh, Sonaka and Manik Chaman, plus Sharad Seedless and its clones, and they harvest at more than 22 degrees Brix because more sugar means more raisin recovery. Drying runs from mid-February to mid-April in open sheds at 40 to 45 degrees Celsius and 20 to 30 per cent relative humidity, taking 10 to 14 days, and it has to finish well before the monsoon because above 70 per cent humidity the berries rot on the racks. The finished kishmish holds 14 to 16 per cent moisture.

That last figure is the one to hold. A raisin is a grape with about four fifths of its water gone, which is why the published harmful dose for raisins is so much lower than for grapes, and why a quantity that looks trivial in a bowl is not trivial at all.

An independent handle on how much acid that represents comes from human nutrition research. A crossover trial in 13 healthy adults treated 5 grams a day of cream of tartar as the tartaric acid equivalent of 120 grams of raisins. Set that beside the two cream of tartar dogs described further down this page, one of which ate 6.76 grams and was in kidney failure within 30 hours. A portion of raisins a human study hands a volunteer for breakfast carries roughly the mass of the same compound that put a 9.6 kg dachshund in a clinic. The dog's problem is not the quantity of fruit. It is the transporter it does not have.

The storage advice no Indian kitchen can follow

ICAR's own recommendation is to store raisins at 10 degrees Celsius and 50 to 60 per cent relative humidity, because raisins are hygroscopic and contact with moisture causes mould, rot and fermentation, with candying and syrup formation above 25 degrees. No home kitchen in Mumbai, Delhi, Chennai or Kolkata is at 10 degrees in any month of the year. The same document explains why the drying window closes in mid-April: once the humid pre-monsoon air arrives, relative humidity above 70 per cent rots the fruit on the racks. So households do the sensible thing and decant kishmish into an airtight jar on a shelf, which is precisely the container a dog can reach, knock over and empty.

Where the dried fruit then turns up is the practical question, and the sweet-shop half of it is already mapped. Our Diwali sweets page lists the mithai and the cooked dishes that carry kishmish, and our guide to Eid and your pet covers sheer khurma, where chopped dates and soaked raisins look identical in opaque milk. The routes below are the ones those pages do not cover.

The Ayurvedic cabinet nobody checks

Chyawanprash contains grapes, which is a sentence we did not find on any page about dogs. The classical formulation, reviewed in the Indian Journal of Traditional Knowledge, lists Vitis vinifera under the common names draksha and munnaka among roughly 50 plants, and munnaka appears again in the standard preparation method. ICAR-NRCG confirms the trade from the other end, recording that dried coloured grapes are called monukka and that demand is higher for making Ayurvedic medicines.

Now the honest part, because the finding is interesting and the risk is probably not large. In the classical method the plants are boiled in about 16 litres of water, the mixture is reduced to a quarter, the decoction is filtered and the marc is discarded. The grape solids do not reach the jar, the material is heated for hours, and nobody has measured tartaric acid in a finished chyawanprash. We cannot tell you the dose is negligible and we will not tell you it is dangerous.

What we can tell you is simpler. Chyawanprash is a sugar, ghee and honey jam given by the spoon to humans, it is a household remedy in northern India rather than a veterinary product, and no dog needs it. If your dog has licked a spoon, say so when you telephone, and say the word draksha, because your vet will not think of it.

Does cooking kishmish change anything? The question India most needs answered

Cooking may reduce the tartaric acid in grapes and raisins, nobody has measured by how much, and three sources give three different answers. This is the biggest unanswered question on the subject for Indian owners, because most of the kishmish an Indian dog meets has been cooked.

Set the three statements beside each other, because reading them in one place is the only way to see the size of the gap.

Three sources, three positions, and not one measurement between them. The 2022 authors are the only ones who state their own uncertainty, and they say the temperature and duration are not in the literature at all.

This is where the subject stops being a translation of an American page. In an Indian home the raisin is rarely eaten raw. It is boiled in milk, fried in ghee, steamed in rice or baked into a cake, and the quantity that reaches a dog usually arrives already cooked. So the exposure route that matters most in India runs straight through the one variable nobody has studied. That is not a reason to relax. It is a reason to say what is known and what is not, which no page serving Indian readers currently does.

A related detail supports treating processing with caution rather than relief. In grapes, during ripening an increased proportion of tartaric acid is precipitated into salts such as potassium bitartrate, and both the free acid and the salt are present in the fruit. Cornell names tartaric acid and its salt potassium bitartrate together as the likely cause, and notes that the concentration varies with ripeness. Heat that decomposes one need not do the same to the other, and the toxicity of potassium bitartrate in dogs has never been formally evaluated.

Two cream of tartar cases show what the salt alone can do, and they are the cleanest dose data in the field. A 28 kg dog that ate four cups of homemade play dough containing 13.52 grams of cream of tartar, about 483 mg/kg, was vomiting within two hours, ataxic by 18 hours, azotaemic at presentation and euthanased at about 96 hours after seizures. A 9.6 kg miniature dachshund that ate 6.76 grams, about 704 mg/kg, began vomiting within two hours and presented 30 hours later with a creatinine of 5.9 mg/dL. Two teaspoons of a baking-aisle powder did that.

Imli, and what six dogs can and cannot tell you

Tamarind has caused fatal kidney injury in dogs, and the entire canine evidence for it is six animals in one case series. Our Diwali page carries the concentration comparison and the general argument. What follows is the dose data in those six cases, which has not been published for an Indian audience anywhere.

Four dogs that ate tamarind and two that ate cream of tartar were identified from the ASPCA poison centre database, decontamination was not performed and treatment was delayed in all of them, two became anuric and one oliguric, four were euthanased and the outcome is unknown for two. The reason to set out the individual cases is that they are the only record of how much tamarind it took.

Fourteen grams is the smallest quantity in the series, and it was the low end of an estimate rather than a measurement. It was followed by anuric kidney failure in a 9.2 kg dog. Read it as a warning rather than a threshold, because four animals cannot establish one and the paper does not claim they can.

Two findings from our own SERP work belong here, because they are the measure of how badly this is covered. A PubMed search for canine tamarind toxicosis run on 22 October 2026 returned exactly one indexed paper, the series above. And Google India's People Also Ask box for the query can dogs eat tamarind returned four questions on 22 October 2026, of which not one was about dogs: what kind of candy can dogs eat, how to eat sweet tamarind, what is the English name for imli, and are tamarind seeds edible. There is no question cluster for imli and dogs in India, in the country where imli is in the chutney, the rasam and the sambar.

Our Diwali sweets page sets out the concentration comparison, what is established and what is not. The short version for this page is that a licked plate is a reason to watch and a candy, a block or a bowl of concentrated chutney is a reason to telephone.

What this page cannot prove, and what India is missing

Everything above rests on case reports, poison-centre retrospectives and one cell-culture study, and none of it was done in India. Naming the gaps is more useful than hiding them, because each one changes how much weight a number deserves.

Seven limits apply to the evidence on this page.

1. No controlled dosing study exists in living dogs, and none is likely, so every dose figure here is either a precautionary line or the lowest quantity someone happened to record in a dog that was already ill.
2. The two cohorts with reassuring numbers counted dogs whose owners noticed and reached a clinic Reich CF, Salcedo MC, Koenigshof AM, and the two with frightening numbers counted dogs that were already in kidney failure Schweighauser A, Henke D, Oevermann. Neither population is the population of dogs that eat grapes.
3. Tamarind rests on four dogs, and cream of tartar on two. The toxicity of potassium bitartrate in dogs has never been evaluated.
4. The mechanism is supported by canine cells in a dish rather than by dogs, and the authors of the 2022 series state that the mechanism by which tartaric acid causes renal tubular necrosis was not described in the literature.
5. Whether cooking changes a raisin is unmeasured, and the only source that raises it says the temperature and duration are not in the literature.
6. Cats are unresolved. Anecdotal reports exist of renal failure in cats and ferrets after they ingest grapes or raisins, and there are no published case reports in either species. Absence of published cases is not evidence of safety.
7. A PubMed search for Indian canine grape or raisin toxicosis run on 22 October 2026 returned no items found. There is no Indian denominator for any figure on this page, and we are not going to invent one.

An eighth gap is worth stating separately because it is the practical one. The tartaric acid content of Indian kishmish has not, as far as we could find, been measured for this purpose. India dries about 27 per cent of its grape crop and has a national research centre for the crop, so the measurement is within reach of somebody. Until it is made, the figures a vet will use here are Spanish, American and British grape numbers applied to Thompson Seedless dried in a shed near Sangli.

What India does have, and where it is

Veterinary kidney support exists in India and most owners do not know it. The Madras Veterinary College nephrology unit in Chennai states that animals with acute and chronic renal failure are managed with intermittent haemodialysis and continuous renal replacement therapy on par with human standards, naming drug toxicities, snake poisoning and leptospirosis-induced renal failure among the causes it treats. Its critical care unit runs from 1 pm to 8 am.

In the north, Guru Angad Dev Veterinary and Animal Sciences University at Ludhiana expanded its dialysis unit in April 2022 and reports an exponential increase in renal failure cases over the preceding four to five years, with 4 to 5 dogs a day presenting to the medicine outpatient department with renal failure, of which 2 to 3 meet its criteria for dialysis. That is a real Indian caseload figure for kidney failure in dogs, from a government university, and it is the only one we found.

Outside those centres it is thin. When a Chennai Labrador with acute renal failure was dialysed in 2020, the veterinarians said that while the Veterinary College in Vepery has dialysis facilities, there are not many private clinics offering the service because it is costly and requires technical skills, and the dog's owner said he had never known dialysis was available for dogs at all. Find out tonight what your own vet refers to, because the Bern survival figures were earned in a hospital with a dialysis unit.

What to do in an Indian home, year round

Keep every form of grape behind a door rather than out of sight, and treat any ingestion as a telephone call on the same day. Prevention here is about the packet and the cooking pot rather than about one festival week.

Six habits cover the realistic routes, and none of them costs anything.

1. Store kishmish, munakka and dry-fruit mixes in a closed cupboard rather than a jar on an open shelf, since ICAR's own storage advice is unachievable at home and the jar is what a dog can reach.
2. Check the kitchen counter while anything with kishmish is cooking, because the open packet beside the pot is a different exposure from the finished dish.
3. Ask what is in a dish before letting anyone hand your dog a taste, naming kishmish and imli specifically, since both hide inside food whose name gives no warning.
4. Weigh your dog, or keep a current weight on your phone, because every published figure on this page is per kilogram and a guess costs accuracy at the one moment it matters.
5. Keep fresh grapes off low tables through the January to April season, when Indian table grapes are cheapest and most households have a bowl out.
6. Find out now, not later, which clinic near you handles kidney cases and whether it refers for dialysis.

If your dog has already eaten something, four things decide what happens next and none of them is a home remedy. Telephone a veterinary surgeon the same day even if the dog is bright. Say the dog's weight in kilograms, what it ate, roughly how much, and the time to the nearest fifteen minutes. Take the packet or a photograph of the ingredient list. Expect the clinic to want repeat bloods over two to three days rather than one panel.

Do not try to make the dog vomit at home, and never use salt water. Emesis is a veterinary decision taken on the individual animal, and our guide to foods toxic to dogs sets out why, including the exposures where it is contraindicated outright.

Community dogs deserve a line, because the routes are the same and the outcomes are worse. Leftover sweets and bakery containing kishmish are a common and well-meant gift, and a dog with no owner has nobody to notice vomiting at 2 am. Our guide to feeding community dogs in India covers what to give instead.

The rest of the Indian plate is covered elsewhere, and sending you there is more useful than summarising it. Our guide to Indian food safe for dogs sorts any dish with four questions, and our page on what fruits dogs can eat runs the Indian fruit bowl against India's own food tables. Grapes are the one genuinely toxic fruit on that list, and everything else there is arithmetic.