Dog Limping: What to Check First, and When It Is Urgent

If your dog is weak or wobbly in the back legs, has changed its bark, is gagging or retching, or has gone downhill over hours, ring a vet now. Along the east and south-eastern coasts of Queensland, New South Wales and Victoria, that picture is a paralysis tick or a snake bite until a veterinarian says otherwise, and both are measured in hours. Do not give your dog paracetamol, ibuprofen, Nurofen, aspirin or any other human painkiller while you wait. Take a short video of the dog walking before you leave the house. Everything below is about a dog that is limping and otherwise itself.

A limp is the commonest reason an Australian dog owner starts searching, and it is the symptom where the American pages that fill the results are least useful.

An American article on limping will run through sprains, cruciate ligaments, arthritis, cut pads and broken nails. All of those happen here. The list it cannot give you is the Australian one: that in tick country in spring and early summer, weakness in the back legs is a paralysis tick until someone has clipped the dog down and looked; that a snake bite can present as weakness and nothing else; that a grass seed drives into a paw in the exact months an American page calls autumn; and that a tick will attach between a dog’s toes as readily as behind its ear.

This page is about sorting one limp from another well enough to know how fast to move, and about what to bring your vet so the appointment is worth what it costs. It cannot tell you what is wrong with your dog, and neither can any other page on the internet.

The three things to do first

Decide whether you are looking at a limp or at weakness. A limp is a dog protecting one leg while the rest of it works normally. Weakness is a dog whose legs are giving way, who is unsteady, who sits down halfway through a walk, or who is wobbly in both back legs at once. Weakness in this country is an emergency until proven otherwise. The rest of this section assumes a limp.
Film it. Thirty seconds of your dog walking away from the camera and back, on a flat hard surface, on a loose lead. Then a second clip of the dog getting up from lying down. This is the single most useful thing you can hand a vet, and the reasons why are further down this page.
Confine the dog and check the foot. Lead only for toileting, no stairs, no jumping on furniture, no off-lead running, until it has been examined. While you are there, look between every toe and under every pad, on all four feet, and run your hands over the whole dog for ticks.

Weakness and lameness are two different problems

This distinction is the most valuable thing on the page, so it is worth being slow about.

Tick paralysis in Australia is caused by Ixodes holocyclus and its relatives, and the 2018 Australian review of lower motor neuron disease by Herndon and colleagues describes the presentation plainly: the clinical signs “commonly start as an ascending limb paralysis first noted in the pelvic limbs and eventually involving the thoracic limbs.” The common pattern starts in the back legs and climbs. The same review notes that asymmetry turns up occasionally, with the front legs worse than the back, so a dog that does not fit the pattern has still not been ruled out.

The clinical scoring system Australian vets use for tick paralysis has four motor grades, and the first one is the reason this page exists. In the version reproduced in that review it reads “Normal or mild weakness and incoordination”; Animal Emergency Australia publishes the same grade 1 as “Mild ataxia or paresis.” Either way, a grade 1 tick dog is a dog whose owner thinks it is limping, or is a bit stiff, or has pulled something in the yard.

The scale has a second axis, and it is the one that kills dogs

Alongside the four motor grades sits a separate respiratory scale running A to D. Animal Emergency Australia gives B as “Mild respiratory compromise – tachypnoeic (RR > 30)”, C as “Moderate respiratory compromise (increased effort, gagging/retching)” and D as “Severe respiratory compromise (dyspnoea and cyanosis).”

A dog can sit high on one axis and low on the other. The same page warns that “the early signs are often subtle and may occur singly and may go unnoticed by both owner and veterinary professional.” A dog that is gagging, retching, breathing hard or barking differently while still walking around the kitchen is a dog in trouble, whatever its legs are doing. That is why the changed bark and the retching sit at the very top of this page.

Three more things about the Australian review are worth carrying around.

The timing is against you right now. Pooled across Australia, nearly three-quarters of tick paralysis cases occur between September and December, with a further 14% across summer. That is a national figure, and the shape of the season shifts along a coastline three thousand kilometres long, so ask a local vet when your own risk window opens.
The tick has usually been on for days. Paralysis begins around the third day of feeding and worsens after that, which is what makes a daily tick search worth doing.
Finding no tick settles nothing. The review notes the offending tick “may be early in her feeding or have finished her feed and detached prior to the onset of significant clinical signs”, and that finding any tick, or the crater where one was attached, is diagnostic on its own.

The same review lists where to search, and one of the places is the answer to a question owners ask constantly. Attention goes to “the head and neck, thorax, areas of skin folds (axilla, vulva, and groin), and the interdigital spaces.” Ticks attach between the toes. A dog in tick season licking at one foot has earned a proper look at that foot, with the hair parted, under good light.

If you find one, take it off — the paralysis tick page covers how, and what not to do while you are doing it. Then go to a vet anyway. Removing the tick stops any more toxin going in, and the dog can still go downhill afterwards: in a 2024 prospective study of 506 Australian dogs across 42 practices from north Queensland to Victoria, “further deterioration after hospitalisation (‘lag phase’) was observed in 9% (37) cases before signs of improvement were seen.” Those dogs were already in a hospital when it happened. The same study put mortality at 6.9%. Keep searching after you have found one, too — a dog can be carrying more than one, and the review above is blunt that a good outcome is impossible until every tick is off.

Snake bite belongs in the same paragraph for the same reason. The review notes that depending on the dose of venom, the species and the age of the snake, dogs “may present with only weakness and ataxia consistent with lower motor neuron blockade.” Weakness and unsteadiness, with no wound anyone can find, is a presentation of envenomation. The snake bite page covers the trap where a collapsed dog gets up and seems fine.

So the first question at the kitchen door is whether the whole back end is working. Which leg comes second.

The limps that need a vet the same day

A dog that will not put the foot down at all

Complete non-weight-bearing lameness means a fracture, a dislocation, a deep wound or a level of pain that needs managing today. It also gives you the one situation where the subjective judgement of a limp is reliable, because the studies further down this page agree on that much: observers only agree with each other when the dog is very lame.

A limp with a swelling, heat, or an obvious wound

Swelling over a joint or a bone, heat you can feel by comparing the same spot on the other leg, or any puncture, cut or missing hair over the sore area. Cat scratches and dog bites over a limb look trivial and abscess quickly.

Sudden lameness in one leg with licking at a single spot

In spring and summer that is a grass seed until someone has looked between the toes. A grass seed travels one way only, and the window where it can be pulled out with forceps rather than dug out under anaesthetic closes fast.

A young, large-breed dog with a leg that broke for no reason

Osteosarcoma accounts for over 80% of bone tumours in dogs. The MSD Veterinary Manual lists the usual sites as the distal radius, proximal humerus, distal femur and proximal tibia, and the presentation as “lameness, bone swelling, and an acute, nontraumatic pathological bone fracture.” A leg that breaks without anything happening to it is a diagnosis in its own right. This is uncommon compared with everything else on this page, and it is the reason a single-leg lameness in a big dog should never be managed at home for months.

Any limp that has not improved after a week of genuine rest

Genuine rest means lead-only, no stairs, no jumping. A week of that with no change means an examination, and usually x-rays.

What is on a vet’s list, and why age and size narrow it fast

The Australian hazards above are the part an overseas page will miss. The orthopaedic list underneath them is much the same anywhere, and the useful thing about it is how quickly a dog’s age, size and which end is limping cut it down. Cornell’s Riney Canine Health Center groups it this way, and it is worth knowing what your vet is already thinking before you walk in.

Older dog, back leg. The most common cause is a torn cranial cruciate ligament, with hip arthritis, bone cancer, spinal disease and foot injuries behind it.
Older dog, front leg. Elbow arthritis, shoulder problems, neck or spinal pain referred down the leg, and bone cancer.
Young dog, back leg. Hip dysplasia leads, and it shows up as trouble with stairs, jumping and getting into the car. Then growing-pain conditions like panosteitis, a luxating patella, and soft tissue injuries.
Young dog, front leg. Elbow dysplasia, especially in large breeds, along with panosteitis, osteochondrosis and shoulder problems.

Two of those deserve a sentence of their own, because they produce the limps owners most often decide are nothing.

A luxating patella is the kneecap slipping out of its groove, and it is common in the small breeds that fill Australian suburbs. The MSD Veterinary Manual describes the classic picture as “intermittent non–weight-bearing lameness of the affected hindlimb that resolves within a few steps (often described as a ‘skipping’ gait)”, with signs generally starting before the age of one. A dog that skips for three strides, shakes the leg out and carries on has been telling you something for months.

Panosteitis moves. It is a self-limiting condition of growing large-breed dogs, and the lameness shifts from leg to leg over weeks. It is the single commonest innocent explanation for the history this page asks you to write down, which is why the question of whether the limp has changed legs is on the list.

And do not skip the feet. A split nail, a cut pad, a pad burnt on hot bitumen in an Australian summer, or a grass seed working its way in will all produce a dog that suddenly will not put a foot down, and all four are visible to an owner with a torch and ten minutes. Check the pads of a dog that goes lame in the afternoon on a hot day before you assume anything about its joints — and see the heat page for how hot a summer afternoon actually gets here.

The cases that limp the least are often the worst

Cranial cruciate ligament rupture is the most common serious orthopaedic problem in dogs, and almost everything owners believe about it is borrowed from human sport.

The Animal Referral & Emergency Network, one of the larger Australian specialist and emergency groups, puts the contrast directly: in people the equivalent ligament usually goes through “an accident involving excessive force, commonly a football or skiing injury”, while in dogs “in most cases this is a spontaneous and slowly progressive condition.” The ligament degenerates over months and then gives way during something ordinary. The afternoon at the beach was the last straw, and owners spend a lot of energy blaming it.

Their description of how it shows up covers a wide range: “Some dogs may show stiffness after rest which wears off within a few minutes”, some show “mild persistent lameness over several months”, and some are “completely unable to use the leg.”

And then the line that matters most. Dogs sometimes rupture both sides at once, and those dogs are “harder to recognise as the dogs may not limp” — what the owner sees is stiffness on rising, reluctance to exercise, pain after exercise, or trouble jumping into the car.

Read that next to what most owners do with a stiff older dog, which is to decide it is slowing down and buy a joint supplement. A dog with two ruptured cruciates walks stiffly on four legs and limps on none of them. So does a dog with arthritis. Telling those two apart takes a pair of hands on the stifles, and it changes the entire plan. We have written separately about the older dog that has slowed down without limping at all, which is the commoner version of this problem and the harder one to see.

Why “he seems a bit better” is the weakest evidence in the room

Three studies say something uncomfortable about how well anybody judges a limp by looking at it, and once you know what they found, the whole business of managing a lame dog changes shape.

Vets disagree with each other. Quinn and colleagues, publishing in Veterinary Surgery in 2007, ran an experimental study on 21 mixed-breed dogs that had all had the same procedure on the same leg — a right tibial osteotomy repaired with an external fixator — and had three small-animal veterinarians with orthopaedic training score the lameness before surgery and at four and eight weeks after, against force plate measurements taken at the same visits. “Agreement was generally low among observers”, the paper reports, and “no significant relationships between any observer’s scores and force plate data existed if very lame dogs were omitted.” Their conclusion includes a sentence worth quoting whole: “Agreement is low unless lameness is severe, and each observer uses an individually unique scale.”

Experience helped less than you would hope. Waxman and colleagues, in the same journal the following year, induced a temporary lameness in 24 healthy dogs weighing 25 to 35 kg, videoed 60 gait trials, and had them scored by three surgeons whose practice was limited to small animal orthopaedics and three first-year veterinary students, with the force platform data hidden from all six. The finding: “Concordance coefficients were low for all observers and were similar between students and surgeons.” Agreement with the force plate stayed low “even when observers only had to be within +/-10% of the ground reaction forces.” Experience did buy something, and the paper says what: “When repeat trials were evaluated surgeons had a much higher repeatability compared with students.” The surgeons were more consistent than the students. They were no closer to the truth.

Owners and vets both see improvement that is not there. Conzemius and Evans went through the placebo arm of an FDA-approved study — 58 dogs, lame from osteoarthritis, everybody blinded, force plate readings throughout. A caregiver placebo effect showed up in owners assessing their own dog 39.7% of the time. In the veterinarians it was 44.8% watching the dog walk, 44.8% at a trot, and 43.1% examining the joint by hand. Meanwhile the force plate found that of the 58 dogs, 46 were unchanged, 7 had improved by 5% or more, and 5 had got worse. The effect “was significantly enhanced with time”, which is to say the longer everyone watched the dog on a sugar pill, the better they thought it was doing.

What that does and does not mean

It would be easy to take this too far, so here is the other side of it, which the same literature supplies.

The most direct challenge comes from the same journal. Aulakh and colleagues, in Veterinary Surgery in 2020, took 18 client-owned dogs with radiographically confirmed elbow osteoarthritis, filmed them at a walk and a trot, randomly shuffled the clips, and had six observers score every one of them twice, at least a fortnight apart. The result: “The ICC for both scores for all observers was >0.9”, and the conclusion, “Interobserver and intraobserver agreement was high for both lameness scores.” High agreement, real dogs, real disease.

Reading those findings side by side is the useful bit. Aulakh’s observers were all scoring identical video clips of dogs with one confirmed condition, and the study measured whether they agreed with one another. Quinn’s and Waxman’s observers were judging dogs against a force plate, and those studies measured whether anyone was right. Agreeing and being right are separate questions, and a room full of people can do the first without the second. Aulakh’s paper also carries a published erratum, which is a reason to hold it a little loosely.

Consistency is the thing that survives. Waxman’s surgeons repeated themselves well. Quinn’s group turned the same finding into a clinical instruction: “Observers must stay the same during the duration of a study for accurate analyses.”

Two caveats belong here as well. Waxman’s dogs were healthy animals with a lameness induced for the purpose, which is a cleaner and more artificial thing than a dog with a diseased joint, and the statistical method that paper used drew a published letter of challenge in the same journal later the same year. The Conzemius paper drew three published comments of its own. This is a live argument among people who measure dogs for a living, and anyone who tells you it is settled has read one side of it.

What survives all of it is a short list of practical consequences, and they are the most useful thing on this page.

See the same vet for the same limp. The comparison that carries information is one person against their own earlier assessment of the same dog. Rotating through whoever is rostered on throws that away.
Film the dog on the same day of the week, at the same time, on the same surface. Your phone does not have a good day or a bad day. Two clips a fortnight apart, shot the same way, is real evidence, and it is evidence a vet can look at twice.
Write down the things that are countable. How many steps before the limp appears. Whether it can still get into the car. How long the stiffness lasts after getting up. Whether it lies down halfway through the usual walk, and at which corner.
Be suspicious of your own optimism about a treatment, particularly a supplement, and particularly after a few weeks. Roughly two owners in five, in a controlled trial, registered a placebo effect assessing their own dog, and the effect grew the longer it went on. Being fond of your dog is what makes you a bad instrument.

Cornell’s Riney Canine Health Center puts the video point from the other direction: videos of the dog at home “are extremely helpful and can provide valuable information that isn’t visible during the exam”, because “some dogs also move normally once they arrive at the clinic due to excitement or adrenaline.” Every owner who has driven forty minutes with a limping dog and watched it trot into the consulting room perfectly sound already knows this.

What your vet will want to know

Work through this before the appointment and write the answers down. An examination costs the same whether you arrive with this or without it.

When it started, and whether it came on over seconds, hours or weeks.
Which leg, and whether it has changed legs since it started.
Whether it is worse first thing in the morning, worse after exercise, or the same all day.
What the dog was doing in the hours before it started.
Whether the dog has been anywhere with long grass, bush, or a beach walk, and when it was last treated for ticks — with the product name and the date.
Whether anything has been given for it, including anything from the cupboard, and whether it seemed to help.
Whether the dog is eating, drinking and toileting normally.
Whether the bark has changed at all.

The five things not to do

1. Do not give human painkillers

This is the one that causes real harm, and the dangerous version is quieter than people expect.

The MSD Veterinary Manual puts acute paracetamol signs in dogs above 100 mg/kg, and methaemoglobinaemia above 200 mg/kg. Those sound like big numbers until you do the arithmetic on an actual Australian dog. A standard paracetamol tablet is 500 mg. In a 5 kg dog, one tablet is 100 mg/kg. In a toy poodle, a Maltese cross, a Chihuahua or a small terrier — which is a very large share of the dogs in this country — a single tablet from the bathroom cupboard is the threshold, and two is well past it.

Ibuprofen is where a limping dog gets hurt more quietly. Dogs given “8–16 mg/kg, PO, every 24 hours, for 30 days showed gastric ulceration or erosions.” A 200 mg Nurofen tablet in a 5 kg dog is 40 mg/kg, which is several times that daily figure, and a limp is exactly the kind of problem someone treats for weeks. Higher up the scale, a single dose of 100–125 mg/kg brings vomiting and abdominal pain, renal failure appears between 175 and 300 mg/kg, and seizures and coma above 400 mg/kg.

So there are two ways this goes wrong. A small dog and one tablet is the fast one. A dog of any size and a fortnight of daily anti-inflammatories is the slow one, and it arrives as a bleeding stomach in a dog whose owner was being kind.

2. Do not give another dog’s anti-inflammatories

Leftover meloxicam, carprofen or Previcox from a previous dog or a previous surgery is dosed for that dog’s weight and that dog’s kidneys, and it will flatten the signs your vet needs to see. It can also limit what can safely be given next, because anti-inflammatories are generally not stacked and a washout is needed before switching.

3. Do not go hunting for the sore spot

Checking the feet and running your hands over the coat for ticks is worth doing and is described above. Flexing the joint, pressing along the bone and bending the leg to find out how much it hurts is a different activity, and it does two unhelpful things: it hurts the dog, and a dog in pain that is handled hard by the person it trusts most will sometimes bite them. Leave the joint testing to the examination, where somebody can read the response properly.

4. Do not let a limping dog keep exercising

A partially torn cruciate that finishes tearing is a bigger operation. “He is fine once he warms up” describes a dog that has stopped protecting the joint, which is the opposite of what the joint needs.

5. Do not wait for it to show itself

Intermittent lameness is still lameness, and a dog that limps on Mondays has something wrong with it on Tuesdays as well. This is exactly what the video is for.

What it costs in Australia

Getting a limp looked at is the cheap part, and the figures are worth knowing before you decide to wait another week.

Walkerville Vet in Adelaide, one of very few Australian practices that publishes a price list, puts a standard consultation at $98.20, and x-rays under sedation, two views, at $520. That is one suburban practice’s published pricing, so treat it as a rough guide, and expect a capital-city referral centre to run higher.

For what sits on the other side of that decision, PetSure, which underwrites a large share of Australian pet insurance, publishes average and highest claim figures from its own book. On 2025 claims data as at March 2026, hip surgery averaged $9,268 with a highest claim of $31,582; fracture repair averaged $4,356 with a highest of $38,787; and intervertebral disc disease averaged $4,924 with a single claim of $72,732.

Cruciate surgery has two answers, and the gap between them is instructive. Bow Wow Insurance, citing PetSure claims data from 2022, reports an average cruciate claim of $2,408 and a highest of $25,687 across 10,993 dogs in that year. That average covers claims of every size, including dogs managed without surgery, so it sits well below what an operation costs. Brisbane Pet Surgery, which publishes its prices and says it does over 250 cruciate procedures a year, lists TPLO from $3,400 to $4,400, CBLO from $3,300 to $4,400, and MMP from $3,100 to $3,600, before optional pre-anaesthetic bloods. That is one knee. Muir and colleagues, following 380 dogs with a rupture on one side, found that “204 of 380 dogs (54%) developed subsequent contralateral rupture within the study period”, with a median survival for the second ligament of 947 days. Slightly better than even odds that you do this twice, and about two and a half years between them.

Every one of those numbers is larger than the consultation that might have caught the problem earlier. None of them is an argument for guessing at home.

The short version

Weakness in the back legs, a changed bark, gagging or retching is a paralysis tick or a snake bite until a vet says otherwise. Go now. Three-quarters of Australian tick paralysis cases fall between September and December.
Film the dog walking, and film it getting up. Thirty seconds each, flat surface, loose lead.
Check between every toe and under every pad, then go over the whole dog for ticks. Ticks attach between the toes.
Confine the dog to lead walks until it has been examined.
Go the same day for a leg the dog will not use at all, a swelling, a wound, sudden lameness with licking at one spot in grass seed season, or a big dog whose leg broke for no reason.
A stiff dog that limps on nothing can still be in serious trouble — dogs that rupture both cruciates at once may not limp at all.
Treat your own impression of improvement carefully, and your vet’s too. In a controlled trial, owners assessing their own dog showed a placebo effect 39.7% of the time and the veterinarians about 44%, while a force plate found 46 of 58 dogs unchanged. Use the same vet, the same camera and countable things.
No paracetamol, no ibuprofen, no aspirin, no leftovers from another dog. Daily ibuprofen at doses people think of as small ulcerates dogs’ stomachs inside a month.

Sources

This page is general information, not veterinary advice, and it cannot tell you why your dog is limping. Lameness is diagnosed by a veterinarian with hands on the dog, and often with x-rays. If your dog is limping, book an examination — and if it has gone weak or wobbly in the back legs over hours, go now.

Dog Limping — What to Check First, and When It Is an Emergency