What Is EIC in Dogs? Exercise-Induced Collapse Explained

Exercise-induced collapse, or EIC, is a genetic neuromuscular condition in which a dog loses control of its hind limbs and sometimes its entire body during intense physical activity. It stems from a specific mutation in a gene called DNM1, and the condition follows an autosomal recessive inheritance pattern, meaning a dog needs two copies of the mutant gene to be affected. Labrador Retrievers are by far the most commonly diagnosed breed, though the mutation has been found in several others. Episodes typically resolve on their own within minutes, but they can be alarming and, in rare cases, life-threatening.

What Happens During an Episode

A typical EIC episode begins after about five to twenty minutes of intense exercise. The dog may first appear slightly uncoordinated in the hind legs, swaying or taking wide, wobbly steps. Within seconds this progresses to a dragging gait and then full collapse of the rear limbs, sometimes spreading to the front legs as well. During the episode the dog is usually conscious and alert, still looking around and sometimes trying to move forward, which distinguishes EIC from fainting or seizures. Body temperature often rises because of the preceding exertion, but the collapse itself is not caused by overheating.

Most dogs recover completely within five to thirty minutes of rest, regaining coordination gradually from front to back. In a small number of reported cases, dogs have died during or shortly after severe episodes, though this outcome is uncommon. What makes EIC distinctive is its repeatability: if the same dog is exposed to the same intensity of exercise under similar conditions, the collapse tends to happen again. Owners often notice a pattern, and that pattern is what sends them to the veterinarian.

The DNM1 Mutation Behind EIC

The genetic culprit was identified in 2008 by researchers who mapped the EIC locus to canine chromosome 9 using genome-wide analysis. They found a mutation in the dynamin 1 gene (DNM1) that causes a single amino acid substitution in a highly conserved region of the protein.¹ Dynamin 1 plays a critical role in how nerve cells recycle the tiny packets of chemical signals they use to communicate with muscles. During low-intensity activity, other forms of dynamin can compensate, so affected dogs seem perfectly normal on a casual walk. But during sustained, high-intensity exertion, the nervous system relies heavily on dynamin 1 to keep up with the rapid-fire signaling muscles demand. When that protein is defective, the signaling system essentially runs out of capacity and the connection between nerves and muscles starts to fail.

The condition is autosomal recessive, which means a dog must inherit one copy of the mutated gene from each parent to be at risk of collapse.² Dogs with just one copy are carriers: they can pass the mutation to their offspring but typically show no symptoms themselves. This carrier status is part of the reason EIC has persisted in popular breeds for so long. Carriers look and perform normally, so without genetic testing there is no obvious reason to exclude them from breeding programs.

Which Breeds Are Affected

Labrador Retrievers carry the DNM1 mutation at surprisingly high rates. Depending on whether you look at conformation show lines, field trial and hunt test lines, or pet and service dog lines, the frequency of carriers ranges from roughly 18% to 38%. The frequency of dogs homozygous for the mutation (those with two copies) ranges from about 2% to nearly 14%.³ Field trial lines tend to sit at the higher end of that range, likely because intense drive and athleticism have been selected for alongside the hidden mutation.

Beyond Labradors, the same DNM1 mutation has been confirmed in Chesapeake Bay Retrievers, Curly-Coated Retrievers, Boykin Spaniels, Pembroke Welsh Corgis, and mixed-breed dogs believed to have Labrador heritage.³ A large genetic screening study across Belgium, the Netherlands, and Germany also identified the EIC-associated mutant allele in additional breeds beyond those originally reported, including in mixed-breed dogs whose parentage included known carrier breeds.⁴ Interestingly, Golden Retrievers, Flat-Coated Retrievers, Nova Scotia Duck Tolling Retrievers, and 15 other non-retrieving breeds tested in one major study showed no evidence of the DNM1 mutation at all.³

The fact that Golden Retrievers are clear is worth highlighting because many people assume all retrievers share the same risk. They do not, at least not for this particular mutation.

Triggers and What Makes Episodes More Likely

Not every bout of exercise causes a collapse in an affected dog. The key ingredient seems to be sustained high-intensity effort combined with excitement. Activities most commonly linked to episodes include retrieving drills (especially repetitive marked retrieves), hard running during games of fetch, upland bird hunting, and agility training at high speed. A leisurely jog or a casual swim rarely triggers a collapse.

Ambient temperature and the dog’s emotional state both appear to lower the threshold. Hot, humid weather makes episodes more likely, probably because the dog’s body temperature climbs faster and the nervous system is already under greater metabolic stress. High excitement levels, the kind of electric enthusiasm many Labradors bring to field work, also seem to play a role. Some owners report that their dog can exercise at the same intensity without collapsing on days when the dog is calmer, but will collapse quickly during a scenario that generates intense eagerness, like the presence of live birds.

This combination of physical intensity and emotional arousal is important for management, because it means that simply reducing exercise duration may not be enough if the dog remains in a state of extreme excitement during whatever shortened activity you allow.

How Many Affected Dogs Actually Collapse

Having two copies of the DNM1 mutation does not guarantee that a dog will show symptoms. In a large survey-based study, about 84% of homozygous Labrador Retrievers had experienced at least one episode of collapse by four years of age.³ That leaves roughly one in six genetically affected dogs apparently unaffected, at least by the time of the survey. Why some homozygous dogs never collapse is not fully understood. Possible explanations include a naturally calmer temperament that keeps exercise intensity below the threshold, differences in other genes that partially compensate for defective dynamin 1, or simply a lifestyle that never demands the kind of sustained effort that triggers an episode.

It is also worth noting that collapse was reported in about 9% of dogs with no copies of the mutation and about 8% of carriers in the same study.³ Those numbers remind us that exercise-related collapse in dogs is not always EIC. Heat stroke, cardiac arrhythmias, metabolic disorders, and other conditions can cause a dog to go down during exercise, which is why a genetic test alone does not replace a full veterinary workup.

Getting a Diagnosis

EIC is diagnosed through DNA testing rather than observation alone. A simple blood draw or cheek swab is sent to a veterinary genetics laboratory, which looks for the specific G767T point mutation in exon 6 of the DNM1 gene. The test returns one of three results: clear (no copies of the mutation), carrier (one copy), or affected (two copies).⁵ Several commercial laboratories now offer the test, and many breeders routinely screen their breeding stock.

Because the mutation is so common in Labrador populations, testing is worthwhile even for dogs that have never shown symptoms. In one Japanese sample of 133 Labrador Retrievers, about 4.5% were homozygous for the mutation and nearly 38% were carriers.⁵ Those numbers are broadly consistent with findings in North American and European populations, confirming that the mutation is widespread globally.

A veterinarian may also run bloodwork, perform a cardiac evaluation, and possibly check thyroid function to rule out other causes of collapse before attributing episodes to EIC. The genetic test confirms the potential, but the clinical picture still matters.

Conditions That Look Like EIC but Are Not

Several other conditions in dogs can mimic the appearance of EIC, and telling them apart matters because their management differs.

Heat stroke: Dogs with heat stroke collapse because their core body temperature has risen to dangerous levels, damaging organs. Unlike EIC, heat stroke usually involves heavy, continuous panting, brick-red gums, vomiting, or diarrhea, and the dog does not bounce back within minutes of rest. Heat stroke is a medical emergency requiring aggressive cooling and veterinary care.
Cardiac syncope: A dog that faints during exercise due to a heart arrhythmia or structural heart disease typically loses consciousness briefly, which most EIC dogs do not. The collapse tends to be sudden and total rather than the progressive hind-limb weakness seen in EIC, and recovery is almost instantaneous once blood flow normalizes.
Myasthenia gravis: This autoimmune condition affects the neuromuscular junction in a way that superficially resembles EIC, with exercise-related weakness that improves with rest. However, myasthenia gravis often involves facial muscle weakness, difficulty swallowing, and megaesophagus, none of which are features of EIC.
Border Collie collapse: An episodic nervous system disorder triggered by strenuous exercise that occurs in Border Collies and related herding breeds. Researchers have found that it has a heritable component, with estimates suggesting genetics account for roughly half of the variation in susceptibility, but the condition involves different genomic loci than EIC and is not caused by the DNM1 mutation.⁶ If you have a collapsing Border Collie, testing for the DNM1 mutation will almost certainly come back clear, and a different diagnostic path is needed.

Because these conditions overlap in their outward appearance, owners who witness a collapse should try to note the details: how quickly it came on, whether the dog remained conscious, which limbs were affected first, how long recovery took, and what the gums looked like. A short video on your phone is even better. That information helps a veterinarian narrow the differential diagnosis considerably.

Living With an EIC-Affected Dog

There is currently no drug or procedure that fixes the underlying genetic defect. Management revolves around avoiding the conditions that trigger episodes. In practice, this means learning your individual dog’s threshold and staying below it. Some affected dogs can swim at moderate intensity, go for hikes, and play in the yard without incident. Others are exquisitely sensitive and will collapse during a short burst of excited retrieving on a warm day.

General guidelines that many owners and veterinary behaviorists recommend include keeping fetch sessions short and low-key, avoiding repetitive high-speed retrieving drills, exercising during cooler parts of the day, and providing frequent rest breaks during any vigorous activity. Training the dog to have a calmer emotional baseline around exciting stimuli, like birds or other dogs, can also help by reducing the arousal component of the trigger equation.

If an episode does occur, the best response is to stop all activity immediately, move the dog to a shaded or cool area, and let it rest. Pouring cool (not ice-cold) water over the dog can help if body temperature is elevated. Most dogs will begin to recover within five to fifteen minutes. If recovery is slow, the dog loses consciousness, or you notice abnormal gum color, seek emergency veterinary care, because the episode may not be EIC after all, or may be one of the rare severe cases.

Many affected dogs live full, happy lives with these adjustments. They can still be active companions; the key is choosing activities that fall below the intensity and excitement level that pushes them over the edge. Nose work, moderate swimming, leash walks, and even low-key agility at a relaxed pace are all options that many EIC-affected dogs handle well.

Breeding Decisions and the Carrier Problem

The high carrier rate in Labrador Retrievers creates a real dilemma for breeders. With carriers making up a quarter to over a third of some breeding populations, removing all carriers from the gene pool would drastically narrow the breed’s genetic diversity, potentially increasing the prevalence of other inherited diseases. The more practical approach, recommended by most veterinary geneticists, is to test all breeding stock and make informed pairings. A carrier bred to a clear dog will produce no affected puppies; roughly half of the offspring will be carriers themselves, and the other half will be clear. Over generations, this strategy can reduce the frequency of the mutation without collapsing genetic diversity.

Breeding two carriers together is where the risk lies: on average, a quarter of those puppies will be homozygous for the mutation and at risk of collapse. Breeding two affected dogs would produce an entire litter of affected puppies. Responsible breeders treat the genetic test as a tool for pairing, not for blanket exclusion, and most breed clubs now encourage or require DNM1 testing before breeding.

EIC in Working and Service Dogs

The practical consequences of EIC are felt most acutely in working dog programs. Labrador Retrievers dominate the world of guide dogs, detection dogs, and search-and-rescue teams, all roles that occasionally demand bursts of sustained, high-intensity effort. A guide dog sprinting to keep pace with a running handler, a detection dog working a large warehouse at speed, or a search dog covering rough terrain in hot weather can all hit the intensity threshold that triggers collapse.

The data show that field trial and hunt test lines carry the mutation at higher rates than pet or conformation lines, likely because the traits selected for in working dogs (high drive, intense focus, tireless retrieving) have been inadvertently co-selected alongside the DNM1 mutation.³ For programs that invest heavily in raising and training each dog, an EIC diagnosis can mean the loss of thousands of dollars and months of effort. This is one of the strongest practical arguments for universal genetic screening early in a working dog candidate’s life, ideally before training even begins.

Some service dog organizations have integrated DNM1 testing into their puppy evaluation protocols, allowing them to place affected dogs in lower-intensity companion roles rather than high-demand working roles. Carriers, who are clinically normal, typically continue through working dog programs without restriction.

Ongoing Research and Unanswered Questions

The discovery of the DNM1 mutation was a major breakthrough, but it did not close the book on exercise-induced collapse in dogs. Roughly one in six homozygous dogs never shows symptoms, and the reasons remain unclear. Some researchers suspect modifier genes, others point to epigenetic factors or even differences in the gut microbiome that might influence how the nervous system handles metabolic stress during exercise. None of these hypotheses has produced definitive answers yet.

There is also the question of whether the DNM1 mutation fully explains all cases of EIC-like collapse even in Labrador Retrievers. The survey data showing that about 9% of genetically clear Labradors still experienced collapse episodes suggests that other genetic or acquired conditions are contributing to the overall picture.³ For Border Collies, the collapse phenotype appears to involve multiple genomic regions and a polygenic architecture, making a single-gene test far less straightforward.⁶

Therapeutic research has explored whether drugs that stabilize synaptic vesicle recycling could treat EIC, but nothing has reached clinical use. The rarity of life-threatening outcomes and the effectiveness of simple management adjustments make it a tough sell for pharmaceutical development. For now, the combination of genetic testing, informed breeding, and exercise modification remains the standard approach, and for most dogs and their owners, it works well.