Start with the strangest number on this page. Screening x-rays held by the Orthopedic Foundation for Animals cover 1,256 Shih Tzu, and 21.7% of them were graded dysplastic in the hips. That is 43rd of the 219 breeds the registry ranks. The Rottweiler sits at 21.1% and the German Shepherd Dog at 20.4%.
A dog that weighs less than a bag of flour, level with the two breeds hip dysplasia is famous for. Only 1,256 films went into that figure, so read it as the same band rather than a win on points. The result is real, checkable, and almost never reported. This is the 21st most registered breed in the United States, so a lot of people are missing it.
What the standard asks for
The American breed standard is specific about size. It wants “9 to 10½ inches” at the shoulder and will take 8 to 11. Mature weight is “9 to 16 pounds.”
That is one range for both sexes. The standard never separates dogs from bitches, so the split between the two columns above is our dataset’s estimate.
Shape matters more than weight here. The standard asks for a body “slightly longer than height at withers”, then warns off both extremes: the dog “must never be so high stationed as to appear leggy, nor so low stationed as to appear dumpy or squatty.” Hold on to that. It comes back under hips.
Hips: a number that does not match the dog
Put the registry rows side by side and the oddity is obvious.
| Breed | Hips screened | Graded dysplastic | Rank of 219 |
|---|---|---|---|
| Bulldog | 1,654 | 70.8% | 3 |
| Pembroke Welsh Corgi | 17,398 | 21.8% | 42 |
| Shih Tzu | 1,256 | 21.7% | 43 |
| Rottweiler | 108,893 | 21.1% | 48 |
| German Shepherd Dog | 146,557 | 20.4% | 50 |
Three cautions before you use it.
These are volunteers. Owners choose whether to send a film. One study comparing registry results with a hospital’s own records found normal hips 8.2 times more likely to be submitted than abnormal ones. So 21.7% is a floor.
A grade is not a limp. Dysplasia is scored on how the joint looks on film. The Merck Veterinary Manual puts it flatly: “Clinical signs of hip dysplasia vary and do not always correlate with radiographic abnormalities.” A 14-pound dog also drives far less force through that joint, every step, for a whole life.
The samples are not equals. Rottweiler owners have sent 108,893 films. Shih Tzu owners have sent 1,256. On that few, 21.7% carries a margin of roughly 19 to 24%, and both of the big breeds fall inside it. So the honest reading is a tie, not a lead.
What survives all three cautions is the size of the number itself. That Merck entry allows only that hip dysplasia “is most common in large-breed dogs; however, it can also affect small-breed dogs and cats”, and this breed is well past “can also”.
Why a lapdog might carry hips like these
Here is our best explanation, labelled as ours.
The Shih Tzu is chondrodysplastic: short-legged by design, and the design is genetic. OMIA, the reference catalogue of inherited traits in animals, lists the breed among those carrying the FGF4 retrogene on chromosome 18, a spare copy of a growth gene that cuts the long bones short. The Dachshund, the Corgi and the Basset Hound carry the same thing. A leg that stops growing early does not stop evenly, and the hip is where the leg meets the body.
There is a muscle argument too. A review comparing the disease in dogs and people reports that “a larger amount of pelvic muscle mass is associated with a lower incidence” of hip dysplasia. Toy breeds carry very little over the pelvis.
Now the honest part. We could not find a study testing either idea in this breed, or in short-legged breeds as a group. We read the FCI’s grading rules, which allow only that “anatomic variation can occur between and within breeds” and never name chondrodysplasia. Neither that review nor a second one we read treats short legs as a risk factor either. Take the explanation above as an inference from anatomy, not as something the literature has confirmed.
What needs no hedge: a breeder holding hip results for both parents is giving you something checkable, and many never x-ray at all.
The knees go the other way
Small dogs are supposed to have bad kneecaps. Patellar luxation, the kneecap slipping out of its groove, is the classic toy-breed complaint. On the same registry’s figures, this breed mostly does not have it.
| Breed | Knees screened | Abnormal | Rank |
|---|---|---|---|
| Pomeranian | 3,481 | 29.1% | 1 |
| Yorkshire Terrier | 3,169 | 17.0% | 4 |
| Shih Tzu | 3,368 | 2.1% | 86 |
Those sample sizes are nearly identical, so this is no artefact. A separate dataset agrees. Across 210,824 dogs at 119 English practices, 1.30% overall had been diagnosed, on an interval of 1.21 to 1.39. The 4,151 Shih Tzu in that population came in at 1.4% (0.8–2.0), statistically ordinary. Pomeranians ran 6.5% (3.9–9.2), Yorkshire Terriers 5.4% (4.4–6.3).
Two methods, one voluntary and one from vet records, agreeing on a low figure. That is about as solid as small-breed orthopaedic evidence gets.
So: a toy breed that gets the hip wrong and the knee right. Whatever is happening in that hip is not a general weakness of small joints, or the knees would follow.
Elbows are the third joint and the thinnest data here: 10.8% abnormal, from only 167 dogs. A hint, not a prevalence.
A flat face, and a scheme that only just arrived
Shih Tzu are brachycephalic. The muzzle is shortened but the soft tissue inside it is not, so the airway is crowded. Brachycephalic obstructive airway syndrome (BOAS) is the name for what follows.
The breed has now been measured. In 2026 Cambridge researchers ran 898 dogs across fourteen flat-faced breeds through a standard assessment: listening at the throat before and after “a 3-minute exercise test whereby the dog is run at a fast trot to a gallop”. Forty-two were Shih Tzu.
The breed landed in the middle band. The authors set the cut-off explicitly: “Breeds with 25–50% Grade 0 dogs were labelled ‘moderate BOAS risk’.” So between a quarter and a half of those Shih Tzu were completely clear, and the rest showed something. The Pekingese and Japanese Chin sat in the worst band. The Pomeranian and the Chihuahua were among the six breeds in the best.
For scale, the same grading applied to 604 dogs of the three worst-hit breeds gave 64.6% affected among 189 Pugs, 58.9% among 214 French Bulldogs and 51.2% among 201 Bulldogs. This breed is not in that company, and it is not clear of the problem either.
Three body measurements came out significant for the Shih Tzu, and one is yours to change. Body condition score, the hands-on fat check your vet does, carried an odds ratio of 5.0 (95% CI 1.1–28.9, p=0.047). Read that interval: it runs from barely anything to enormous, because only 42 dogs went into it. Trust the direction, not the size. Our guide to spotting an overweight puppy covers checking without a scale.
The other two you are born with. Shorter dogs were more affected (p=0.01), and “Tail length was noted as significantly associated with BOAS for the Shih Tzu (adjusted p=0.04)”. Shorter tails carried the higher risk. That result showed up in only one other breed of the fourteen, and the authors do not claim to know why.
Worth knowing what the standard asks for: “Nostrils are broad, wide, and open.”
The certificate, and the contradiction on OFA’s page
In April 2026 the Orthopedic Foundation for Animals extended its breathing programme. Its own page says that “Based on updated research at the University of Cambridge, the following breeds have been added to the list of breeds eligible for exams and official RFGS certification”, and the Shih Tzu is on that list of thirteen.
Further down the same page, an older line reads: “The screening program is currently open to Pugs, Bulldogs, and French Bulldogs.”
Both are live at once. We take the April update as current because it is newer, and quote the other because you will find it and wonder which to believe.
Grades run 0 to III. Grade 0 means “The dog is clinically unaffected and free of any respiratory signs of BOAS”; grade III means it “is clinically affected and has severe respiratory signs of BOAS”. Eligibility is only months old, so the registry held four Shih Tzu results when we checked. Four is not a prevalence, but one graded parent is a checkable fact, and that beats a photograph.
Eyes: the part that needs you every day
If you take one section away, take this one.
Ophthalmologists put 50 Shih Tzu through a full eye exam and a battery of tests. What is startling is not that any one dog was in crisis. It is that the anatomy repeated across all of them.
- Every eye had caruncular trichiasis and medial lower lid entropion: hairs at the inner corner and a lower lid rolling inward, both rubbing the surface.
- 82% of eyes showed lagophthalmos: the lids do not fully meet.
- Tear film break-up time averaged 5.3 ± 2.4 seconds: how long the tear layer holds before it splits. Short means a cornea drying between blinks.
- Corneal touch sensitivity averaged 1.8 ± 0.7 cm on the test filament, where a lower reading means the surface registers less.
- Already visible: pigment in 27% of eyes, scarring in 20%, new blood vessels in 12%.
The authors put it plainly: “Qualitative tear film deficiency (low TFBUT), along with several anatomical abnormalities that promote ocular irritation and reduce globe protection, together help explain the concerningly high prevalence of ocular surface disease in the Shih Tzu breed.”
A prominent eye that dries fast, cannot close fully, is rubbed by its own hairs and feels less than it should. Four problems, one surface.
What that produces in real records
In a year of UK vet records covering 2,423 Shih Tzu, ulcerative keratitis, an open sore on the clear front of the eye, was logged in 3.5% (2.8–4.3). Dry eye reached 1.2% (0.9–1.8), eye discharge 2.1% (1.6–2.8). Grouped together, eye disorders hit 11.9% (10.7–13.3), third behind skin and teeth.
Against other breeds the gap is wider than that sounds. Among 104,233 dogs in England where 0.80% (0.75–0.86) had a corneal ulcer recorded, the Shih Tzu carried odds of 10.04 (7.30–13.80) against crossbreds. Only the Pug at 19.05 and the Boxer at 12.12 beat it. That paper also explains the mechanism: flat-faced dogs have “reduced numbers of corneal nerve endings”, and features such as “a large palpebral aperture and a shallow socket” make it worse “by promoting greater exposure.”
At the specialist end, a referral hospital reviewed 970 dogs of seven flat-faced breeds over ten years. Shih Tzu made up 333 of them, 34.3%, more than the Pug at 202 or the Boston Terrier at 161. “Shih Tzus were the most common breed and presented at a higher frequency for an ophthalmic examination than all other breeds.” Corneal ulcers topped the whole group’s list at 373 of 2,873 diagnoses (13%), then dry eye at 294 (10.2%).
Proptosis: the emergency to know about in advance
A shallow socket means the eye can be shoved forward past the lids, which then shut behind it and trap it. This is proptosis, and it is a same-hour emergency.
A study of 100 dogs treated for it reports that “The most commonly reported breeds were Chihuahua (20), Shih tzu (19), mixed breed dogs (17)”. The cause was usually another dog: 77 of the 100 came from bite wounds. Ninety-seven survived to discharge, but 54 lost the eye and 44 had the lids stitched shut to hold the globe in place.
Practical version: keep this dog out of scuffles with bigger ones, and know where the emergency clinic is before you need it.
The kidney claim, and the test behind it
Search this breed and you meet juvenile renal dysplasia, where the kidneys never finish developing. The American Shih Tzu Club says it “is found most commonly in Shih Tzu, Lhasa Apsos, and Soft-Coated Wheaten Terriers”, and steers owners toward tissue examination, because blood tests “are of little use in identifying mildly or moderately affected dogs.”
Then you meet a DNA test. It rests on a 2011 paper reporting variants in the COX-2 gene promoter (the switch region controlling how much of a gene gets read) in a Lhasa Apso family of 52 dogs, 35 biopsied. The correlation held in that pedigree (X² = 15.2, p<0.001). No Shih Tzu were in the authors’ own sample.
Two things happened next, and neither is in the marketing. The journal attached a formal Expression of Concern on 8 November 2012, listing four problems. Two matter most. The control group was a different breed, so “The validity of the control population employed in the study is compromised”. And “the evidence from other breeds suggests that this may be a neutral DNA variant.”
Then in 2015 another group sequenced the same region in 152 healthy dogs and 7 gray wolves and found nineteen versions of it sitting harmlessly in animals with no kidney disease. Their conclusion: “The wild-type sequence is not expected to increase the likelihood of RD, challenging the claim that this variant and/or other regional sequence variants harbor susceptibility to RD in dogs.” No Shih Tzu were in that sample either. Neither side has tested the breed the test is sold for.
The disease is genuine and the breed is genuinely named in it. The test is not settled, and a “clear” result does not mean what a buyer will assume. Ask your vet before spending on it. In that year of records on 2,423 Shih Tzu, kidney disease did not reach the common disorders, though renal disease was the recorded cause of death in 3 of the 89 dogs that died, 3.4%.
Living with one
Set the ceiling first. Researchers who matched 18,385 owner surveys against 2,155 sequenced dogs concluded that “Breed explains just 9% of behavioral variation in individuals.” Most traits were still heritable: the heritable part is simply not parcelled out by breed. A page like this sets expectations. It cannot describe your dog.
Then a gap worth knowing about. The 2008 C-BARQ study, the cross-breed survey people cite for temperament rankings, built two samples: 33 breeds recruited online and 11 through breed clubs, with a 45-dog minimum in both. The two lists overlap, so they are not 44 different breeds. We read both row by row. The Shih Tzu is in neither. A site that ranks this breed for aggression and footnotes that paper is pointing at something which does not contain it.
What exists is vet records, which measure something else: behaviour bad enough that somebody wrote it down. Of those 2,423 Shih Tzu, aggression was recorded in 2.4% (1.9–3.1) and undesirable behaviour in 2.2% (1.7–2.9). Low numbers, and a low bar: a dog has to reach a consulting room before it counts. We have no temperament ranking to offer, and would rather say so than invent one.
The first eighteen months
Adolescence is documented rather than folklore, and the study everyone cites holds two separate groups of dogs that are easy to run together. The first was 285 young dogs scored on questionnaires at 5, 8 and 12 months. The second was a different 93 dogs given a hands-on “sit” test at 5 and 8 months only: 82 at the earlier point, 80 at the later, 69 at both.
The finding people quote belongs to the smaller group. Those dogs “responded less to the ‘sit’ command during adolescence, but only when the command was given by their carer, not a stranger (the carer and stranger were the same people at both time points).” The teenage phase has your name on it.
Owner reports peak slightly later. A UK cohort followed puppies from 6 to 18 months. The share of owners flagging at least one unwanted behaviour topped out at 42.1%, or 513 of 1,219, at 12 months, then eased to 29.9%, or 204 of 684, by 18 months. An earlier report on the same puppies put the 6-month figure at 31.3% of 965. It mostly improves on its own.
No Shih Tzu took part in either, so read both as findings about dogs.
Growth restriction arguments belong to giant breeds and do not apply here. Two other clocks do. Baby teeth that should have dropped out were still present in 2.6% of dogs in that year of records, and a retained tooth crowds a small mouth further. And the daily eye check (a squint, a held-shut lid, a change in how your dog carries its face) is a habit worth building while a puppy still thinks handling is a game.
Other things to raise at a check-up
None of this is a diagnosis, and most Shih Tzu never meet most of it. In that year of UK records, 68.9% of the 2,423 dogs had at least one disorder logged, which also means nearly a third had none.
| Disorder | Dogs | Prevalence (95% CI) |
|---|---|---|
| Periodontal disease | 229 | 9.5% (8.4–10.7) |
| Anal sac impaction | 180 | 7.4% (6.5–8.5) |
| Ear disorder | 134 | 5.5% (4.7–6.5) |
| Otitis externa | 114 | 4.7% (3.9–5.6) |
| Overgrown nails | 103 | 4.3% (3.5–5.1) |
| Umbilical hernia | 94 | 3.9% (3.2–4.7) |
| Obesity | 86 | 3.5% (2.9–4.4) |
By category, skin came top at 16.6%, then teeth at 13.3%, eyes at 11.9%, gut at 10.5%, ears at 10.2%. The study’s own summary: “Shih Tzus had higher prevalence of anal sac impaction, umbilical hernias and eye problems than reported previously in dogs overall.”
Teeth. Gum disease at 9.5% is the most recorded problem, and the registry agrees from another angle: 11.8% of Shih Tzu assessed for dentition were abnormal, though from only 127 dogs. Start brushing early.
Heat. Across 905,543 UK dogs in one year, 395 heat-related illness events were confirmed, an incidence of 0.04% (0.04–0.05). Brachycephalic dogs ran at 2.10 times the odds of medium-muzzled ones, on an interval of 1.68 to 2.64. This breed came out well inside that: “Of the breeds analysed, the Lhaso Apso, Shih tzu and Chihuhua had the lowest odds of HRI despite all being brachycephalic breeds”, the spellings are the authors’. Stay careful anyway, because of what happens when it does occur. Of those events, 14.18% were fatal (11.08–17.96).
Degenerative myelopathy. The registry holds a SOD1 DNA result for only 62 Shih Tzu, of which 25.8% were homozygous. Sixty-two dogs would fit in a room. And homozygous is a genotype, not a disease: OMIA records that “some homozygous mutant dogs had no signs of degenerative myelopathy, which suggests incomplete penetrance” for this variant. We could not find the Shih Tzu among the breeds OMIA lists as having the clinical disease recorded.
Eye screening. Of 2,397 Shih Tzu given a registry eye exam, 4.5% were abnormal, 114th among the breeds listed for that exam, and we could not confirm how long that list is. Hold that against the eye section above. Registry exams hunt for inherited disease inside the eye. They are not built to catch a cornea being scraped by an eyelash.
How long they live
This breed does unusually well, and the two best figures differ by four years.
Records from eight Tokyo pet cemeteries covering 12,039 dogs include 962 Shih Tzu, with life expectancy at birth of 15.0 years (14.8–15.1), against 13.7 for the whole dataset. The highest figure in that dataset was the Shiba at 15.5, so this breed sits near the top of it.
UK life tables from vet records give 11.23 years for dogs overall, and 11.05 for this breed. The same paper prints both side by side: “Shih Tzu (UK = 11.05 and Japan = 15.0 years)”, so the gap is a known feature of the two datasets rather than an error in either. Japanese pet dogs are mostly indoor and mostly small, and a cemetery population is self-selecting: those owners chose and paid for a burial.
A third figure sits between them. In that year of UK primary-care records, median age at death was 12.7 years (IQR 8.7–14.3, range 2.0–19.9), from 89 deaths with an age recorded for 85. Median age at death and life expectancy at birth are different measurements, which is most of why it beats 11.05.
Our read: plan for a long-lived dog. Whichever figure you take, this is not a breed whose face costs it most of its life, and among flat-faced breeds that is unusual.
What we could not pin down
Why the hips read the way they do. The chondrodysplasia explanation is our inference from anatomy and genetics. No paper we found tests whether short-legged breeds get more hip dysplasia, and the FCI rules carve out no exception for them.
Any month-by-month Shih Tzu weight chart. Everything we opened was unsourced. The table on this page is model output for a dog of this adult weight, not a record of what real puppies of this breed weighed.
A daily exercise figure. No study has measured how much this breed needs. Since exertion is what brings out airway signs, a fixed minute count would be worse than nothing.
Where this breed sits on temperament against others. The standard cross-breed survey never included it. Vet records tell you what got written down.
Whether the renal dysplasia DNA test predicts anything here. The paper behind it carries an Expression of Concern, the refutation used other breeds, and neither study genotyped a Shih Tzu.
How common breathing trouble really is. Forty-two dogs put the breed in a middle band. A real measurement, and a small one.
Coat and grooming claims. Every “hypoallergenic” and “low-shedding” line we checked traced back to marketing rather than measurement, so we have left the subject alone.