Von Willebrand disease in dogs: the bleeding disorder to find before the spay, not after it

Signs of Von Willebrand disease

The 10 signs to watch for are listed below. Any one of them is a reason to consult your nearby vet; none of them is a diagnosis.

• Nose bleeds with no injury to explain them
• Bleeding from the gums while teething or chewing
• Blood in the urine, turning it pink, red or brown
• Black tarry stools, or fresh blood in the stool
• Bruising under the skin after ordinary handling
• Bleeding that will not stop from a torn nail
• Prolonged oozing after a spay, neuter, dental or biopsy
• Heavy bleeding after whelping, or during a season
• Oozing from the umbilical cord in a newborn puppy
• Pale gums and sudden weakness after a bleed nobody saw

Breeds on this site that document this condition

Each breed page below carries the figure for this condition with its denominator and its source. The 10 breeds listed are the ones on this site that document it, not a ranking of risk.

What von Willebrand disease is, and why it stays hidden until surgery

Von Willebrand disease is a shortage or a malfunction of von Willebrand factor, the protein that glues platelets to the wall of a cut vessel, and it is the commonest inherited coagulopathy dogs have. The platelets are normal and there are enough of them. They simply cannot stick where they are needed, so the first plug never forms properly and the bleeding runs on.

Von Willebrand factor is a large multimeric glycoprotein made by endothelial cells and by megakaryocytes, and stored in the alpha granules of platelets and in organelles called Weibel-Palade bodies inside endothelial cells. Dogs keep only about 3 percent of their von Willebrand factor inside platelets, against about 20 percent in people, so almost all of a dog's supply is circulating or held in the vessel wall. That detail matters later, because the one drug that raises the factor works by emptying the endothelial store.

The protein does two jobs at once. It bridges the platelet receptor GPIb-IX-V to subendothelial collagen in vessels with fast-moving blood, acting as a sticky tether that slows platelets down long enough for firmer receptors to grip. It also carries factor VIII through the circulation and protects it from being broken down, which is why factor VIII activity is often low in a dog with almost no von Willebrand factor.

The reason this disease ambushes owners is that it does not show up on the tests a clinic runs before an anaesthetic. Most dogs with von Willebrand disease have normal platelet counts and a normal coagulation profile, meaning normal prothrombin time, normal activated partial thromboplastin time and normal fibrinogen. In the Buenos Aires genotyping study, the haemogram, platelet count and coagulation test results were within normal ranges in all 41 dogs that had medical records, with no relationship to genotype at all.

Von Willebrand disease has been identified in 54 breeds according to the Argentine review, in over 50 breeds according to Cornell, and in at least 30 according to VCA eClinPath, Cornell University College… VCA Animal Hospitals. Those three counts are not a contradiction so much as three different cut-offs for what counts as a report. It was first described in 1970, in a family of German Shepherd Dogs.

What it is not

Von Willebrand disease is a distinct disorder and it is not haemophilia. Haemophilia A is a factor VIII defect, it is X-linked, and it affects males almost exclusively. Von Willebrand disease is autosomal, so males and females transmit and express it with equal frequency.

It is also not the same thing as a low platelet count, and confusing the two costs time in India in particular. Pinpoint skin haemorrhages, the petechiae, are rarely seen in von Willebrand disease, and a predisposed dog showing them should be worked up for thrombocytopenia before anyone reaches for a von Willebrand assay. Our page on tick fever in dogs in India covers the commonest Indian cause of that picture.

Type 1, type 2 and type 3 are three different diseases

The three types of von Willebrand disease differ in how much factor is present, in whether the factor that is present works, in how severely the dog bleeds, in which breeds carry them, and in which mutation causes them. Reading a single figure for the disease is therefore close to useless, and breed predisposition is the first thing to establish rather than the last. Within any one breed a single form predominates, so the breed tells you which disease you are dealing with.

Type 1 is a quantitative deficiency, with the normal protein present in reduced amount and all multimer sizes still represented eClinPath, Cornell University College…. It accounts for more than 95 percent of all canine cases Segert JH, Seidel JM, Wurzer WJ. Bleeding is variable and depends on how low the factor runs and on the breed.

Type 2 is a qualitative defect, with low concentration and abnormal structure, specifically a relative loss of the large multimers that do most of the work eClinPath, Cornell University College…. It is rare, and in dogs it is reported in the German Shorthaired Pointer and the German Wirehaired Pointer. Bleeding is more severe than in type 1 and it does not track the antigen level, because the antigen assay counts protein and cannot see multimer size.

Type 3 is a near-total absence of the protein and it is the most severe form eClinPath, Cornell University College…. Cornell records it as familial in the Chesapeake Bay Retriever, the Dutch Kooiker, the Scottish Terrier and the Shetland Sheepdog, and sporadic in the Blue Heeler, Border Collie, Bull Terrier, Cocker Spaniel, Labrador Retriever, mixed breeds and Pomeranian. That sporadic list matters in India, because the Labrador Retriever is on it.

Four mutations, four tests, and no interchangeability

Each type has its own DNA test and the tests do not substitute for one another. Type 1 is read from VWF c.7437G>A. Type 2 in the two Pointer breeds is read from c.1657T>G. Type 3 has three distinct mutations, an intron 16 splice-site change in the Dutch Kooiker, c.255Cdel in exon 4 in the Scottish Terrier, and c.735Tdel in the Shetland Sheepdog. UC Davis puts the warning on all three of its pages, that the mutations have different breed origins and distributions so test selection needs to be breed-appropriate.

The practical consequence is that a clear result on the wrong test is worthless. A Scottish Terrier returned clear on the type 1 test has been tested for a variant its breed does not carry, and its actual risk, the type 3 deletion, was never looked at. Read the report for which variant was assayed, not just for the word clear.

Where the Shetland Sheepdog sits in two lists at once

Cornell's classification table names the Shetland Sheepdog under type 1 and again under familial type 3, and UC Davis sells a type 3 test with a Sheltie-specific mutation, c.735Tdel. Both are correct. A breed can carry more than one variant, and the antigen assay cannot tell you which, because it measures the amount of protein and not the reason there is less of it.

The severity gap between the types is not theoretical, and one study measured it directly across three breeds. In 260 clinically affected dogs, mean plasma von Willebrand factor antigen was 15 percent of normal in affected Doberman Pinschers, 8 percent in affected Shetland Sheepdogs and 0 percent in affected Scottish Terriers. The affected Dobermans were also diagnosed later, at a mean age of 4.6 years against 1.9 years for Shelties and 1.7 years for Scotties, because a dog with no factor at all declares itself early.

Even the site of bleeding differs by breed. The commonest site of mucosal bleeding was the oral or nasal cavity in Scottish Terriers and Shetland Sheepdogs, and the urogenital tract in Doberman Pinschers.

One disagreement is worth flagging because it changes how a result is read. VOSD, the page that ranks first in India for this query, states that type 2 may show normal or near-normal factor levels that nonetheless do not function correctly. Cornell classifies canine type 2 as low concentration with abnormal structure, and eClinPath describes it as low antigen with a relative decrease in high molecular weight multimers. The description VOSD gives fits a recognised human subtype. For a dog, the safer reading is Cornell's, and either way the point stands that in type 2 the antigen number underestimates the risk.

Why a DNA result does not tell you whether your dog will bleed

A type 1 DNA result predicts the phenotype badly, and the studies that genotyped whole cohorts say so in their own numbers. Most dogs carrying two copies of the Doberman variant never bleed, and some dogs carrying one copy do. That is the single most misread thing about this disease, and it is the same penetrance trap the site already documents on MDR1 drug sensitivity in dogs.

Start with the cleanest cohort. Researchers in La Plata genotyped 89 Doberman Pinschers from the province of Buenos Aires for c.7437G>A and found 32 dogs homozygous normal, 41 heterozygous and 16 homozygous mutant, giving a mutant allele frequency of 0.41. Forty-one of those 89 dogs had medical records covering surgeries, parturition and other bleeding episodes.

Of the five homozygous mutant dogs with records, three had no evidence of clinical signs at all, and only 6 of 27 heterozygous dogs showed any. Every homozygous normal dog was normal, which is what makes the authors read the inheritance as dominant with incomplete penetrance rather than recessive.

The bleeding time told the same story. Mean buccal mucosal bleeding time was 2.36 minutes in the five homozygous mutant dogs, 2.15 minutes in the 27 heterozygotes and 1.66 minutes in the nine homozygous normal dogs, and single-factor analysis of variance put the difference at p equals 0.403, which is not significant. A genotyped, clinically documented Doberman cohort could not separate its own genotypes by bleeding time.

The Kromfohrlaender work is the other half of the picture, because it shows penetrance running the other way. Eighty-seven dogs were genotyped for the same c.7437G>A variant, 80.5 percent carried it, 65.5 percent as heterozygotes and 15.0 percent as homozygotes, and 21 percent of the whole cohort had bleeding symptoms. Serum von Willebrand factor ran from 28 to 137 percent in wild-type dogs, 3 to 77 percent in heterozygotes and 1 to 23 percent in homozygotes. Because heterozygotes both dropped their factor and bled, the authors concluded that in this breed inheritance is autosomal dominant with incomplete penetrance rather than recessive.

Three respected sources, three different inheritance patterns

The Royal Kennel Club describes type 1 as autosomal-recessive, and states that a dog inheriting only one copy will have no signs of the disease but will be a carrier. UC Davis lists the mode of inheritance for exactly the same variant as autosomal dominant with incomplete penetrance, and says a dog with one copy may be affected. Cornell's Riney Center calls the type 1 variant recessively inherited while adding that carriers can have reduced factor levels, and Cornell's own diagnostic laboratory describes the pattern in many breeds as incomplete dominance, because one abnormal gene appears sufficient to cause abnormal and variable bleeding.

All four are current pages from serious institutions and they do not agree. This page prints the spread rather than picking one, because the practical instruction is the same under every version. Treat a heterozygous dog as a dog whose bleeding risk is unknown rather than as a dog that is safe, and confirm it with a factor measurement before an operation.

The self-selection problem in every published frequency

Breed frequencies for this disease come from dogs whose owners chose to test them, which is a self-selected pool and not a census. The largest unselected estimate comes from a panel screen of over 100,000 mixed breed and purebred dogs, where the type 1 variant ranked eighth by disease allele frequency among mixed breed dogs at 0.768 percent, and fourth in the combined purebred sample at 1.460 percent. Those are allele frequencies across a whole tested population, and they sit an order of magnitude below any breed figure on this page.

The same disagreement shows up inside one breed. That panel screen found the type 1 variant in 13 of 160 Kromfohrlaender, which is 8.1 percent, against 80.5 percent of 87 dogs in the breed-specific study. Both are real measurements of the same variant in the same breed, and the gap is what sampling does. Ask which dogs were tested before you believe any percentage.

The type 2 marker that was probably never the cause

There is a harder version of this problem, and it affects results already sitting in owners' files. The panel screen reported type 2 carriers in more than ten breeds on the originally published variant VWF c.4937A>G, including 6 of 17 Canaan Dogs and 3 of 8 Boykin Spaniels. Its authors then sent plasma from dogs homozygous for that variant for antigen testing, and every result came back borderline or within the laboratory's reference range, which they read as further support that the originally published variant is in fact not the causal variant.

The Utrecht group had reached the same place from the genetics. Sequencing a German Wirehaired Pointer with type 2 disease found four amino-acid-altering variants, healthy dogs of the breed were homozygous for two of them, Chinese Crested dogs homozygous for c.4937G had no signs of a bleeding disorder, and only c.1657G was found homozygous exclusively in affected dogs. A type 2 result on the old marker is not evidence that a dog has type 2 disease. Check which variant the laboratory reported.

What the bleeding looks like, and when to go in today

Bleeding in von Willebrand disease is mucosal and it is disproportionate to what caused it. A dog with the trait can live years without a single episode and then bleed for hours from a torn nail, a lost tooth or a routine spay incision. The pattern to recognise is bleeding from a lining rather than into a joint, and bleeding that does not stop when it should.

In puppies the first sign is often a procedure nobody thought twice about. Excessive bleeding has been described after tail docking, dewclaw removal, tattooing and simple teething, and the Royal Kennel Club adds prolonged bleeding from the umbilical cord at birth. Our guide to puppy teething covers what normal looks like at that age.

In adults the trigger is usually surgery. VCA puts it plainly, that bruising or bleeding after a surgical sterilisation procedure may be the first time any abnormality is noticed. Cornell's own teaching image for severe type 3 disease is a Shetland Sheepdog with severe bruising along the incision line after an ovariohysterectomy, and the caption records that the bruising occurred in spite of a pre-operative transfusion.

Bitches carry a second window. The Royal Kennel Club lists prolonged bleeding during a season and after whelping among the recognised signs, and factor levels themselves swing more during pregnancy and heat than at any other time.

One case, and why cosmetic surgery is the worst place to find out

A five-year-old Doberman Pinscher was presented with sudden left-sided bleeding around the eye after surgery to replace a prolapsed third eyelid gland. The eye was pushed forward, the whole conjunctiva was haemorrhagic, and the dog was blind on that side with no direct or consensual pupillary light reflexes. Computed tomography showed a large retrobulbar haematoma around the optic nerve, with no orbital fracture and no history of trauma.

The work-up is the lesson. Buccal mucosal bleeding time was prolonged beyond four minutes with a normal platelet count, von Willebrand factor antigen came back at 24 percent against a reference range of 50 to 150 percent, and a type 1 mutation was confirmed by PCR. A single blood test before the operation would have changed the plan. Our page on cherry eye in dogs covers the surgery that triggered it.

Crespi's cohort makes the point about elective cosmetic surgery specifically. The medical records of the genotyped Dobermans list ear cropping and tail docking among the surgeries performed on them. Cropping the ears of a Doberman puppy whose von Willebrand status nobody has checked is an avoidable bleeding risk taken for appearance, and Cornell's Riney Center lists avoiding cosmetic surgeries first among its management strategies.

What makes an affected dog bleed worse on a given day

Bleeding tendency in an affected dog is not constant. Infections, endocrine disorders and certain medications can all exacerbate it, and eClinPath adds hormonal fluctuations around heat cycles and pregnancy to the list. Plasma factor levels themselves fluctuate day to day even in healthy dogs, and that fluctuation is exaggerated during pregnancy, during a season, and in any dog with systemic illness, especially liver disease or an inflammatory disorder.

Three tests, and the different question each one answers

To work out what your dog's result means, first work out which of three tests produced it, because the buccal mucosal bleeding time, the von Willebrand factor antigen assay and the DNA test answer three different questions and none of them substitutes for the others. One measures function on the day, one measures how much protein is circulating, and one reads a single letter of the genome.

The bleeding time is a screen, not a diagnosis

Cornell's clinical pathology reference says exactly where the buccal mucosal bleeding time belongs. It can be used as an in-clinic test for von Willebrand disease, and the worked example it gives is a Doberman with unknown factor results before surgery, but the test is not sensitive or specific for the disease and therefore does not replace more specific assays. VCA reaches the same practical position from the other direction, saying that at minimum a buccal mucosal bleeding time may be considered before a planned spay or neuter.

Both statements are true at once, and the way to hold them together is this. A prolonged bleeding time in a predisposed breed is a reason to stop and investigate. A normal bleeding time is not clearance.

Reading an antigen result

The Cornell Comparative Coagulation Laboratory reports each dog as a percentage against a 100 percent standard and publishes three intervals. Normal is 70 to 180 percent, borderline is 50 to 69 percent, and abnormal is 0 to 49 percent. The most severely affected dogs sit below 25 percent, and dogs with the type 3 form typically run at 1 percent or less. eClinPath adds a fourth line that is the one a surgeon cares about, that most dogs which bleed because of von Willebrand disease have antigen values below 35 percent.

• Normal, 70 to 180 percent. The dog is considered clear of the trait and at low risk of expressing or transmitting it.
• Borderline, 50 to 69 percent. The dog cannot be classified as carrier or clear from this measurement alone; a second assay, a test mating or a DNA test resolves it.
• Abnormal, 0 to 49 percent. The dog is considered a carrier, is at risk of transmitting an abnormal gene, and some individuals will express a bleeding tendency.
• Below 35 percent. Most dogs that actually bleed from this disease sit here.

Timing the sample matters more than owners expect. Plasma levels fluctuate day to day in normal healthy dogs, and the fluctuation is exaggerated during pregnancy, during a season, and in any dog with systemic illness, so the values are most useful as genetic predictors when taken from a healthy dog at a quiet time. Puppies can be sampled from six to eight weeks of age.

Why the antigen assay cannot find carriers reliably

The Dutch Kooiker work put a number on this. Among 39 obligatory carriers of the type 3 mutation, dogs whose pedigree position made carrier status certain, antigen ranged from 30 to 114 percent with a median of 64 percent, and was subnormal in only nine animals. Thirty of 39 known carriers would have tested in the normal or borderline band. That is the gap the DNA test was built to close.

The Scottish Terrier paper says the same thing about its own breed. Standard protein-based tests are accurate for identifying affected Scottish Terriers, but are not reliable for identifying carriers unless multiple replicate assays are performed. In a voluntary survey of 87 dogs submitted by Scottish Terrier owners, the new PCR assay found 15 carriers and four affected dogs, two of which had previously been shown to have undetectable factor.

What a general-population study found when it compared all four

The closest thing to an unselected prevalence study anywhere in a middle-income country comes from Brazil. Researchers at Sao Paulo State University tested 350 dogs of various ages and breeds around Botucatu, all of them with no historical or clinical evidence of abnormal bleeding, and evaluated von Willebrand factor antigen, buccal mucosal bleeding time, activated partial thromboplastin time and factor VIII activity against one another. Prevalence in that population was 1.43 percent, and the antigen assay was the best laboratory test for diagnosing the disease.

That 1.43 percent is worth holding beside the breed numbers further down this page. In an ordinary mixed clinic population, this is an uncommon disease. In a Doberman line, it is not.

The pre-surgical plan, in general terms and with no numbers on a syringe

To plan a surgery safely in a dog with von Willebrand disease, the veterinary team needs the diagnosis in hand before the date is set, a factor level to work from, and a decision about whether a blood product will be on the premises. This page describes what those plans look like in the published series. It does not give amounts, because the amount is the veterinarian's judgement for the animal in front of them.

The best evidence that an elective spay can be done safely in an affected dog is a multi-institutional case series of 20 dogs. Sixteen had von Willebrand disease and all sixteen were Doberman Pinschers, with a median von Willebrand factor expression of 19 percent and an interquartile range of 18 to 30 percent. All sixteen were pretreated with desmopressin and four also received cryoprecipitate.

No postoperative complications, including signs of haemorrhage, were reported for any dog in that series. One dog had an iatrogenic splenic laceration during trocar insertion that required conversion to an open laparotomy, and even that dog did not bleed abnormally afterwards. The authors concluded that minimally invasive ovariohysterectomy or ovariectomy may be considered in female dogs affected with these coagulopathies when they are pretreated.

That is the argument of this whole page in one study. The same operation, in the same disease, is a routine day surgery when the diagnosis is known in advance and a crisis when it is not.

What desmopressin does, and what it does not do

Desmopressin acetate works by emptying a store rather than by making new protein. It stimulates release of von Willebrand factor from the Weibel-Palade bodies of endothelial cells, raising antigen values and shortening the buccal mucosal bleeding time for up to four hours in Dobermans with type 1 disease. Repeated administration has diminishing effect because the store is finite. It does not work at all in type 3 disease, where there is no endothelial store to release, and Cornell's Riney Center states the same exclusion.

The response is also incomplete and hard to predict. Sixteen non-anaemic Dobermans with type 1 disease were tested before and one hour after desmopressin, and baseline buccal mucosal bleeding time was prolonged in 12 of 14 dogs measured. Bleeding time shortened significantly after the drug in 6 of 7 dogs, and the point-of-care closure time shortened significantly across the group, but only one dog raised its plasma factor concentration by more than 35 units per decilitre, and there was no correlation between the rise in factor and the improvement in closure time.

In healthy dogs the effect is measurable and dose-dependent. A 2026 Japanese trial in 27 healthy Beagles under isoflurane anaesthesia found that desmopressin shortened buccal mucosal bleeding time in a dose-dependent way and raised factor VIII antigen within an hour, with no other significant changes in the haematology, coagulation or blood pressure measurements.

Two cautions belong beside all of that. eClinPath states that not all dogs with type 1 disease respond, so the drug should not be relied upon to achieve haemostasis in surgery, although it may be useful as an adjunct to transfusion. VCA adds the practical objection, that it is not recommended for regular use because it is expensive and not all dogs respond. Ask your veterinarian to decide, with your dog's own factor result in front of them.

Blood products, and why plasma beats whole blood here

Severe haemorrhage in this disease is controlled with fresh plasma, fresh frozen plasma or cryoprecipitate, and plasma products can be transfused before an operation to prevent surgical haemorrhage Cornell University College of Veterinary…. Whole blood should be used only in a dog that is hypoxic from anaemia, and even then component therapy is preferred, because whole blood does not carry enough von Willebrand factor to stop the bleeding. Plasma products are also the safer long-run choice, since these dogs often need repeated transfusions across a lifetime and every exposure to donor red cell antigens raises the risk of a transfusion reaction.

There is one more trick in the literature that matters where donors are scarce. VCA notes that the donor dog may itself be treated with desmopressin before blood collection, which raises the factor level in the donated blood.

Drugs to flag on the file

Affected dogs should not be given drugs that interfere with normal clotting mechanisms, and Cornell names aspirin, sulfa-type antibiotics and heparin. The Riney Center adds non-steroidal anti-inflammatory drugs generally, because they interfere with platelet function and add bleeding risk. VCA publishes the longest list, while stating openly that much of it is extrapolated from human research because the specific work has not been done in dogs.

• Non-steroidal anti-inflammatory drugs, including aspirin, meloxicam, carprofen and deracoxib.
• Sulfa-based antibiotics, penicillin, and ampicillin or amoxicillin Cornell University College of Veterinary….
• Heparin VCA Animal Hospitals.
• Phenothiazine tranquillisers and certain antihistamines.
• Oestrogens, certain antacid medications and theophylline.

None of that is a ban. It is a list to hand to whoever is writing the anaesthetic plan, so the risk and the benefit are weighed with the diagnosis known.

The conversation to have before a date is fixed

Five things need to be settled before an elective operation in a predisposed dog, and all five are easier a fortnight early than on the morning.

• Confirm which variant your breed carries, and order that DNA test rather than a generic von Willebrand panel.
• Measure von Willebrand factor antigen from a healthy dog at a quiet time, not during a season and not while the dog is unwell.
• Ask whether the practice can obtain fresh frozen plasma or cryoprecipitate, and how long it would take on the day.
• Ask whether desmopressin is being planned, and accept that it is an adjunct rather than a guarantee.
• Put the drug list and the factor result in writing in the dog's file, and give a copy to any clinic that has not seen the dog before.

The hypothyroidism claim, and what the studies that tested it found

Hypothyroidism is widely repeated as a cause of acquired von Willebrand disease in dogs, and the claim traces back to one small 1990 study. Three later studies set out to reproduce it and none did. The honest position today is that thyroid disease may worsen bleeding in a dog that already has the inherited trait, and that treating the thyroid is not a treatment for the bleeding disorder.

The original finding is real and it is easy to see why it stuck. Fifteen dogs with suspected hypothyroidism had a mean von Willebrand factor antigen of 47.1 percent with a standard error of 12.6, significantly lower than 94.7 percent with a standard error of 5.6 in 14 euthyroid dogs, at P less than 0.0005. Four of the hypothyroid dogs were given thyroxine for a month and all four rose, and a single hyperthyroid dog measured 200 percent. The authors concluded that reduced antigen can be found in association with congenital disease or with disease acquired through hypothyroidism.

Four years later the same question was asked with the thyroid status confirmed rather than suspected. Ten dogs with hypothyroidism had antigen measured before and after roughly two months of replacement thyroxine, and their values were within reference limits throughout, at 126 percent plus or minus 41 before treatment. After treatment the figure fell significantly, to 94 percent plus or minus 39, at P less than 0.01, which is the opposite direction to the hypothesis. The authors wrote that, contrary to findings in some other studies, a deficiency of plasma von Willebrand factor did not appear to result from hypothyroidism.

The strongest test came two years after that, because it made the hypothyroidism rather than finding it. Hypothyroidism was induced in eight adult dogs with radioactive iodine, with four healthy dogs as controls, and antigen and buccal mucosal bleeding time were followed for 23 weeks and then through five weeks of replacement therapy. No significant change in bleeding time was seen at any point, and the authors concluded that hypothyroidism does not induce acquired von Willebrand disease or significant defects in primary haemostasis.

The last study asked the question owners actually care about, which is whether thyroid hormone treats the bleeding. Eight euthyroid Doberman Pinschers with antigen below 15 percent were given levothyroxine and placebo for 30 days each in a two-period, double-blinded crossover design with a washout between. Their mean antigen was 8.9 percent against a reference range of 70 to 180 percent, and their mean collagen-binding activity 11.1 percent against a reference range above 70 percent. Treatment produced laboratory evidence of hyperthyroidism and changed nothing else, with no difference in bleeding time, antigen, collagen binding or factor VIII activity.

So why do the reference pages still name the thyroid

Because the guideline text is making a smaller claim than the one owners hear. Cornell states that infections, endocrine disorders and certain medications may exacerbate signs of bleeding in affected dogs, and eClinPath is more specific, listing endocrine disorders causing deficiencies of steroid or thyroid hormones among the concurrent stress conditions that can worsen haemorrhage in a dog already affected. Worsening an existing bleeding tendency is a different claim from causing the disease, and the controlled studies argue only against the second.

Our page on hypothyroidism in dogs does not list bleeding among the signs of that disease, and on this evidence that is the correct call. If your dog is hypothyroid and also bleeding, treat the two as separate problems until a factor assay says otherwise.

What the screening registers show, and what they cannot show

The Orthopedic Foundation for Animals publishes a von Willebrands DNA register, and reading it in full corrects three things most pages get wrong about this disease. It was pulled live on 27 September 2026 as a 2,209,461-character document, with the in-document control string LABRADOR appearing 109 times, and every row below sums to 99.9 or 100.0 percent once rounding is allowed for.

The first correction is that the Doberman is not merely the worst breed on that register, it is the only breed on it with a substantial affected share. At 6.5 percent abnormal across 1,963 evaluations it sits twice as high as the Boykin Spaniel and more than three times the Kerry Blue Terrier, and 17 of the 22 breeds record no affected dogs at all.

The second correction is that the Doberman figure has moved, and no ranking page says so. The breed's own CHIC page prints the clear percentage by birth-year cohort, and it reads 40.0 percent for dogs born up to 1990, 46.0 percent for 1991 to 1995, 45.2 percent for 1996 to 2000, 48.1 percent for 2001 to 2005, 48.6 percent for 2006 to 2010, 49.7 percent for 2011 to 2015, then 69.5 percent for 2016 to 2020 and 76.4 percent for dogs born in 2021 or later. Two decades of flat results, then a jump. That is what a cheap, definitive DNA test does to a breeding population once enough breeders use it.

The third correction is the biggest and it concerns the numbers everyone quotes. In 1988, 30 percent of 1,363 Scottish Terriers and 28 percent of 4,279 Shetland Sheepdogs tested had abnormal von Willebrand factor antigen, and those two figures are still reproduced on ranking pages today. The DNA register now records the Scottish Terrier at 98.9 percent clear across 1,294 evaluations and the Shetland Sheepdog at 99.3 percent clear across 2,732, with zero affected dogs in either breed.

Those are not contradictory numbers. They are different measurements, taken 38 years apart, in differently selected populations. An antigen below 50 percent is not the same event as carrying a type 3 deletion, and the dogs submitted to a DNA register in 2026 are the descendants of thirty years of selection against exactly that deletion. Quoting the 1988 phenotype figure as though it described a puppy born today is the commonest error on this subject.

The Dutch eradication, and what it proves

The Kooikerhondje is the cleanest demonstration that this works. Type 3 disease was diagnosed in 38 Dutch Kooiker dogs, 19 of them probands referred independently for a moderate to severe bleeding tendency and 19 more found by screening 717 dogs including 356 puppies, with diagnosis resting on undetectable antigen below 1.6 percent. The OFA register now holds 297 evaluations in the breed at 100.0 percent normal, with no carriers and no affected dogs. The Dutch breed club made the DNA test compulsory for parent animals and the register is what that decision looks like measured a generation later.

What the register cannot tell you

Every figure in that table comes from voluntary submissions by owners who chose to test, which is the same self-selection problem that inflates every breed percentage in this field. Two rows show it plainly. The German Shepherd Dog appears with 54 evaluations at 100.0 percent normal, and this is both a breed Cornell lists under type 1 and the breed in which canine von Willebrand disease was first described, in 1970. Fifty-four dogs is not evidence that the breed is clear. It is evidence that almost nobody tests it.

The Royal Kennel Club adds a second limit that applies to anyone buying from an overseas line. It records and publishes results only from laboratories that meet its criteria, and results from laboratories not on that list are not recorded at all. A certificate from an unlisted laboratory may be perfectly accurate and still leave the dog's record blank on the club's Health Test Results Finder.

The breeding arithmetic, and the mistake breeders make in both directions

Cornell's breeding advice is deliberately moderate. Matings between two test-clear parents are predicted to produce only clear puppies, carriers can be bred to test-clear mates with some puppies testing normal, and the proportion of carriers in a line falls gradually without losing desirable traits. Carrier-to-carrier matings are undesirable because they are the ones likely to produce the most severe form, and no dog that expresses abnormal or excessive haemorrhage should be bred at all.

For the recessive types the arithmetic is exact. UC Davis states that matings between two carriers of the type 3 mutation are predicted to produce 25 percent affected puppies, and the same holds for type 2 University of California Davis. One tested parent removes that risk entirely.

• Ask for both parents' results on the variant your breed actually carries, not a generic panel.
• Ask which laboratory ran them, and check it against the kennel club's recorded list if the papers matter to you.
• Ask for a factor antigen figure as well as a genotype where the breed is a type 1 breed, because the genotype predicts bleeding poorly.
• Ask what the breeder does with carriers, and treat a plan to breed carrier to clear as a reasonable answer rather than a red flag.
• Record every answer in writing, because a verbal assurance is not a test result.

Our guide to checking a dog breeder in India covers the paperwork side of that conversation.

Von Willebrand disease in India: the test, the blood bank and the tick problem

The DNA test is obtainable in India, at a rupee price, from an Indian laboratory that ships a swab kit to your door. The blood products are the harder half. India published its first national guidance on animal blood banking in August 2025, that document is explicitly advisory, and it states that most animal transfusions in the country are still done in emergency settings on whichever donor happens to be available.

The DNA test, and a defect in the only Indian list that names all three types

Urban Animal, a company based in Bengaluru and registered at an R.T. Nagar address in Bangalore, sells a cheek-swab Trace Test at Rs 4,500, reduced from Rs 7,500, covering over 130 genetic health insights plus 20 physical traits, with free pan-India shipping and sample pickup and a stated turnaround of four to six weeks. Its published condition list names Von Willebrand disease I, Von Willebrand disease II and Von Willebrand disease III in its haematological block. Its laboratory partner GeneTech, an NABL-accredited Hyderabad laboratory, separately lists a canine traits and disorders test at Rs 7,000 with an 18-day turnaround and over 130 genetic insights, though the string willebrand does not appear anywhere on GeneTech's own page.

So the answer to the question every Indian owner asks is yes, and at a price comparable with sending a swab abroad. There is one defect in that list worth knowing before you read a report from it. The definition printed under all three headings on Urban Animal's condition page is the type III definition, word for word, describing an inherited bleeding disorder characterised by total or near-total absence of von Willebrand factor in the plasma. The words quantitative and qualitative, the two that separate type 1 from type 2, do not appear anywhere on that page at all.

For a Doberman owner that is exactly the wrong way round. Type 1 in a Doberman is the mild quantitative form that often never bleeds Crespi JA, Barrientos LS, and reading the type 3 description against a type 1 result would frighten an owner into cancelling a needed operation. Read the type definitions on this page or on Cornell's, and use the Indian report for the genotype.

Blood products, and the government document that finally names this disease

In August 2025 the Department of Animal Husbandry and Dairying published Guidelines and a Standard Operating Procedure for Blood Transfusion and Blood Bank for Animals in India, a 40-page document prepared after the Veterinary Council of India made a formal request and forwarded guidelines drafted by TANUVAS. A consultative meeting was held at Veterinary Council of India headquarters on 26 June 2025, and a high-level meeting under the Secretary of the department followed on 4 July 2025 at Krishi Bhawan in New Delhi. TANUVAS is the Tamil Nadu Veterinary and Animal Sciences University, and both it and ICAR-Indian Veterinary Research Institute at Izatnagar are named in the document's contributor list.

Two things in it are directly about this disease. Its blood component table lists cryoprecipitate against von Willebrand disease and haemophilia A, describing it as highly concentrated factor VIII and von Willebrand factor and useful in dogs, with fresh frozen plasma listed for coagulopathy and hypoalbuminaemia. Its donor screening section states that canine donors must at a minimum test negative for haemoprotozoan infections such as Babesia and Ehrlichia, and lists von Willebrand factor levels if indicated alongside them.

Read those cold-chain lines together and the Indian reality becomes clear. Cryoprecipitate needs a freezer below minus 30 degrees C and has to be used within four hours of thawing. That is not something a general practice keeps on a shelf for an emergency, which is why the plan has to be made before the surgery rather than during it.

The guidance is honest about where the country starts from. It states that at present India lacks a national regulatory framework and standardised protocols for veterinary transfusion medicine, and that most animal blood transfusions in India are still performed in emergency settings, relying on hospital-available or client-owned donors, without consistent screening, blood typing, or standard operating protocols. It also carries a disclaimer stating that its contents are illustrative and recommendatory and should not be construed as having any statutory or regulatory force.

Where the blood actually is

One named canine blood bank has been operating in India for years, at the Madras Veterinary College campus in Vepery, Chennai, run as a wing of the Department of Clinics as the TANUVAS Animal Blood Bank. Reporting from 2019 recorded it as the only canine blood bank in the country at that time, handling around 200 to 250 canine blood transfusions a year, storing collected canine blood for a maximum of three weeks and frozen plasma for up to a year, and therefore keeping blood on an as-needed basis rather than in stock. Treat the only-one claim as a 2019 statement rather than a current census.

The 2025 guidance proposes a National Veterinary Blood Bank Network and recommends the establishment of state-regulated blood banks, and its contributor list includes both TANUVAS in Chennai and ICAR-Indian Veterinary Research Institute at Izatnagar. Transfusion in Indian small-animal practice is documented and it does happen: an adult female Labrador Retriever presented to the referral polyclinic at ICAR-IVRI Izatnagar with Ehrlichia canis, marked anaemia and thrombocytopenia was transfused on the second day of treatment and recovered.

Why a bleeding dog in India has two plausible diagnoses, not one

Tick-borne infection is the reason a bleeding Indian dog cannot be assumed to have an inherited disorder. A four-city Indian survey using conventional and molecular diagnostics found 261 of 525 dogs, 49.7 percent with a confidence interval of 45.4 to 54.1, infected with one or more tick-borne pathogen. Ehrlichia canis ran at 20.6 percent and Anaplasma platys at 6.5 percent across the whole sample, and both destroy platelets. Concurrent infection with more than one pathogen occurred in 39 percent of cases.

Where you live changes that risk by an order of magnitude. The four sampling sites were chosen to reflect different climatic zones, and they were Sikkim in northern West Bengal, Ladakh in Jammu Kashmir, Delhi as a monsoon-influenced humid subtropical region and Mumbai as a tropical one, sampled from June to September 2008. Ticks were found on 80 percent of dogs in Mumbai and 75.3 percent in Delhi, against 17 percent in Sikkim and 11 percent in Ladakh.

Read the top row against the map. Nearly two in five Delhi dogs and more than one in four Mumbai dogs carried Ehrlichia canis, while not one dog in Sikkim or Ladakh did. A Dobermann kept in a humid coastal city therefore faces a bleeding risk this page cannot separate from its genetics without a blood film, and a Dobermann in a dry montane town faces very little of it.

The clinical discriminator is the kind of bleeding. Pinpoint petechial haemorrhages are rarely seen in von Willebrand disease, so a predisposed dog showing them should be worked up for thrombocytopenia first. Our pages on ehrlichiosis in dogs and tick fever in dogs in India cover that side, and our guide to tick prevention for dogs in India covers keeping it out of the house.

A dog can of course have both, and that is the worst case. A dog with reduced von Willebrand factor that also acquires a platelet-destroying infection does not have two problems so much as one compounded one.

The Indian evidence, and a controlled absence

No published Indian canine von Willebrand series exists, and this is the search behind that statement. Europe PMC was used because NCBI E-utilities returned HTTP 500 on everything on 27 September 2026, and both a title-and-abstract search and an author-affiliation search were run.

The title-and-abstract search for von Willebrand with dog, dogs or canine and India returned exactly one record, a 2023 Research Square preprint from Madras Veterinary College Teaching Hospital on papaya leaf extract in 90 thrombocytopenic dogs with ehrlichiosis, babesiosis or leptospirosis. It matched because von Willebrand factor assays were among the tests performed in the work-up, and it reports no von Willebrand disease result. It is nonetheless the only evidence found that an Indian institution runs the assay at all.

The affiliation search returned one record too, and it is a pure artefact. A human root-canal case report from Government Dental College, Calicut matched because the patient's complaint concerned a maxillary canine tooth and because Bernard-Soulier syndrome involves the von Willebrand factor receptor on the platelet surface. Neither instrument was broken: the control query for hip dysplasia and Labrador returned 110 on-topic canine papers, the same query written in PubMed field syntax returned three European haematology congress abstract books, and the affiliation control for canine parvovirus in India returned 83 genuinely Indian canine papers.

What it costs in India, and what the numbers actually mean

A breed-appropriate DNA test costs between Rs 4,500 and about Rs 4,900 depending on whether you buy it in India or send a swab to California, and the Indian option is a whole panel while the American one is a single variant. Every rupee figure below has been divided by 84.38 and by 74 to check whether it is an Indian price or a converted dollar price, and each row says which it is.

Read that table for what you are buying rather than for the smallest number. The Indian panel is cheaper per condition and far more expensive per unit of certainty about this particular disease, because it reports over 130 conditions on a list whose own descriptions of the three types are identical. The UC Davis test reports one named variant with the mode of inheritance, the three possible genotypes and what each one means printed beside it, and the result can be submitted to the OFA register.

The blood test has a printed price too, and it is the cheapest useful number on this page. Cornell's Comparative Coagulation Laboratory charges USD 37.50 for a von Willebrand factor antigen assay and USD 112.50 for a whole litter, which converts to roughly Rs 3,160 and Rs 9,490 at 84.38 to the dollar, and its collagen binding assay is USD 54.50. Those are American laboratory fees rather than Indian ones, but they set the order of magnitude an Indian clinic's send-out should be compared against.

If your dog is a Doberman, the panel that matches the breed is the better value of the two American options, because the Doberman Pinscher Health Panel at USD 130 covers the breed's other inherited risks in the same submission. Our Dobermann breed profile sets out what those are.

The cost nobody prices, which is the blood

No published Indian rate card for canine cryoprecipitate or fresh frozen plasma was found for this page, and that absence is itself the finding. The government guidance sets storage standards for both products and names cryoprecipitate as the indicated product for this disease, while stating that the country has no national regulatory framework for veterinary transfusion and that most transfusions still run on client-owned donors found on the day. Where a product has no reliable supply chain, it has no reliable price.

The practical planning figure for an Indian owner is therefore not a rupee amount at all, it is a phone call. Ask the practice, before the operation is booked, whether it can source plasma and how many hours that would take.

Living with a dog that has it

Most dogs diagnosed with von Willebrand disease live an ordinary length of life, and the management is a short list of avoidances plus one piece of paperwork. Cornell's Riney Center states that with proper care dogs can have a normal lifespan, and that the strategies are avoiding cosmetic surgeries, giving the right medication before necessary surgeries, and avoiding drugs that further affect platelet or clotting function.

There is no cure for von Willebrand disease, because the defect is written into the gene that makes the protein. What can be changed is every occasion on which the dog is asked to clot, and there are fewer of those than an owner fears.

The single most useful thing you can do costs nothing. Put the diagnosis, the variant tested, the antigen figure with its date and the drug list on one printed page, keep a copy in the dog's file and a photograph of it on your phone, and hand it over at any clinic that has not treated the dog before. A result nobody can find at midnight has bought nothing.

• Decline elective cosmetic surgery outright, because ear cropping and tail docking appear in the surgical records of genotyped Doberman cohorts and both cut vascular tissue for appearance alone Cornell Richard P. Riney Canine Health….
• Flag every planned procedure in advance, including a dental scale, a lump removal, a biopsy and a routine neuter.
• Ask before any new medicine is dispensed, because several common drug classes interfere with platelet function Cornell University College of Veterinary….
• Keep ticks off the dog, because a platelet-destroying infection on top of a factor deficiency is the worst combination available in India eClinPath, Cornell University College….
• Watch the mucous membranes rather than the skin, since mucosal bleeding is the pattern and petechiae point elsewhere.

Bitches need one extra decision made early. A season and a whelping are both bleeding events in an affected dog, and antigen levels swing most during exactly those periods, so the choice between spaying with a plan in place and leaving a bitch entire is better made before the first season than after a difficult one. Our guide to dog sterilisation in India covers the procedure and its Indian costs, and the case series above shows that an elective spay in an affected, pretreated dog carried no bleeding complications in 16 of 16 Doberman Pinschers.

What to tell a new vet in one minute

Lead with the breed and the word. Say that the dog is a breed that carries von Willebrand disease, give the type if you know it, give the antigen percentage and the date it was measured, and say whether the parents were DNA tested. Then ask the two questions that change the plan, which are whether the practice can obtain plasma and how long it would take.

If your dog is a Scottish Terrier, a Kooikerhondje, a Chesapeake Bay Retriever or a Shetland Sheepdog, say type 3 specifically, because desmopressin does not work in those dogs and the plan has to start from a blood product Cornell Richard P. Riney Canine Health…. Our Scottish Terrier profile records the breed's OFA figures alongside its other inherited risks.

Everything on this page is general information. Your own veterinarian, with your dog in front of them, decides what is tested, what is given and when, and nothing here replaces that judgement.