Scottie Lion Jaw (CMO) and SLC37A2: What CT Scans and DNA Data Show

SamOur Scottie puppy yelps when he yawns and leaves his kibble. Could this be “lion jaw”?

Elena MarshIt could be one of several things, and only your vet can tell them apart. A 2024 imaging study shows why that visit should not wait for the puppy to “grow out of it”.

SamThe breeder said lion jaw is just a puppy phase. Is that true for Scotties?

Elena MarshBone growth often settles once the skeleton matures, but not always. In the same 2024 CT series of 20 dogs, five puppies had surgery, and the Scotties showed the heaviest changes.

SamThe litter’s paperwork says “CMO: one copy”. Is that just a carrier, like other DNA tests?

Elena MarshThat assumption is worth questioning. The 2016 Helsinki and Bern study that found the SLC37A2 variant concluded CMO behaves as dominant with incomplete penetrance.

Bottom line: Craniomandibular osteopathy (CMO, or “lion jaw”) is a painful overgrowth of bone in the jaw and skull of growing puppies, usually between about 4 and 8 months (reported range 3 to 12 months). In the largest CT study so far, 20 dogs seen at referral hospitals in South Africa and the UK (2024), Scottish Terriers made up 13 of the 20 cases and had the most severe changes, and five dogs had surgery. In the 2016 Helsinki–Bern genetic study, all 10 Scotties with two copies of the SLC37A2 c.1332C>T variant had CMO reported, as did 3 of 43 with one copy, and one affected Scottie had no copy. The authors suggest dominant inheritance with incomplete penetrance, so a one-copy dog is at risk, not a silent carrier. Since January 1, 2025 the Scottish Terrier Club of America requires an OFA-registered CMO DNA result for a CHIC number; the UK has no such requirement. A DNA result is risk information, not a diagnosis: a puppy that will not open its mouth, stops eating, has a swollen jaw or a fever needs a veterinarian, who diagnoses CMO by examination and imaging.

Not every jaw grows out of it: what CT scans of 20 puppies showed

The 2016 University of Helsinki paper summarises the classic picture: signs appear between 4 and 8 months, with jaw swelling, recurring fever, poor appetite, pain, difficulty opening the mouth and trouble swallowing, and “signs of the disease usually resolve with time, when the growth period is finished.” The Merck Veterinary Manual calls the prognosis favorable for mild to moderate cases and guarded for severe ones.

The study that complicates this for Scottish Terrier owners was published in 2024 by Pérez López, Almansa Ruiz, Steenkamp and Holdsworth in Frontiers in Veterinary Science. It reviewed every CMO case with a CT scan at two centres between 2011 and 2023: 13 dogs from the University of Pretoria in South Africa and 7 from Davies Veterinary Specialists in the UK. The puppies were 6 weeks to 12 months old (mean 6.5 months), with no sex difference.

Thirteen of the 20 were Scottish Terriers and 16 were terriers of some kind; the rest were single dogs of other breeds, including a Cairn, a Jack Russell, a Labrador and an Akita. All 13 Pretoria cases were Scotties, and they were the most severely affected. The combined CT severity score was significantly higher in Scotties than in the other dogs (median 13 versus 7, p = 0.038), and older puppies tended to score higher.

That 65% figure is easy to misread. It describes who reached two referral CT services, not how often Scotties get CMO. Earlier work puts the highest frequency in West Highland White Terriers, and a cluster of related dogs at one hospital can tilt a small sample. The Pretoria team raised the possibility of “unknown common ancestors” among their Scotties, but they did not genotype them, so whether those dogs carried the SLC37A2 variant is unknown.

What the paper does establish is that in this breed CMO is not reliably short and mild. Every dog had received pain relief before referral, with no or only mild improvement. Five of the 20, all from Pretoria, went on to surgery, and one dog was euthanised after surgery brought only temporary relief.

SamSo a Scottie could need an operation for something that looks like teething trouble?

Elena MarshIn the 2024 series, five of 20 puppies had surgery, all at one hospital. A referral group skews to hard cases, so it is not a rate for all Scotties, but it shows the ceiling.

Ear bulla, airway and jaw joint: how far the bone can spread

“Lion jaw” suggests a lower-jaw problem. The CT images show something wider. New bone affected the mandible in every dog (both sides in 95%), the cranium in 90% and the tympanic bulla, the bony chamber of the middle ear, in 60%. All 12 dogs with bulla changes also had narrowing of the nasopharynx, the airway behind the nose, which was almost closed in two. The ear canal was narrowed in 55%, and the authors suspected interference with the jaw joint in 10 of those 12 dogs.

The jaw joint matters most for the long term. The paper notes that bone overgrowth generally stops at skeletal maturity, around one year, but that dogs whose jaw joint fuses have a poor outlook. Merck lists the severe complications in the same direction: a mouth that cannot open fully, weight loss from poor nutrition, chronic ear infection, airway narrowing and neurological deficits if bone compresses the brain.

The signs that brought these dogs in were less dramatic. In 19 of 20, the main complaint was pain when opening the mouth, eating or having the lower jaw touched. The lymph nodes under the jaw were enlarged in 90%, a quarter could not open the mouth easily, and a fifth had a fever.

A much older case from Japan fits the same pattern. In 1985 Hashimoto described a seven-month-old female Scottish Terrier with four months of worsening jaw pain, a temperature of 40.1°C, difficulty opening and closing the mouth and marked swelling of both lower jaw branches on X-ray. She recovered gradually over five months.

One copy of SLC37A2 is not a “carrier” result

Most breed DNA tests follow a recessive logic: two copies cause disease, one copy makes a healthy carrier. CMO does not fit that template, and reading a CMO result as if it did is the most common mistake an owner can make.

The variant was described in 2016 by Hytönen, Lohi and colleagues from the University of Helsinki, the Folkhälsan Institute and the University of Bern in PLoS Genetics. It is a single-letter change, c.1332C>T, in exon 15 of SLC37A2 on canine chromosome 5. It does not change an amino acid; instead it disrupts how the gene’s message is spliced, so many transcripts lose 79 bases and code for a shortened protein. The team called it “leaky”, because affected dogs still make some normal transcript, and one-copy dogs showed a milder loss than two-copy dogs.

The pattern of affected dogs did not match a recessive disease: some one-copy dogs were affected, and some two-copy dogs had no signs reported. The authors wrote: “Collectively, our results suggest that CMO is inherited as dominant disease with incomplete penetrance.” They added that “alternative models of inheritance cannot be completely ruled out yet.” Incomplete penetrance means the variant raises the risk without deciding the outcome.

Labs phrase this differently. Wisdom Panel’s CMO condition page says autosomal dominant, that dogs with one or two copies are at risk, and rates the likelihood “low-moderate”. Embark’s page describes “incomplete dominance”; LABOKLIN writes “autosomal dominant with variable penetrance”. None treats a one-copy dog as risk-free. If a breeder labels a one-copy Scottie a “carrier”, ask what they mean.

SamShould a one-copy puppy worry me more than a recessive carrier would?

Elena MarshMore attention, not alarm. In the 2016 study, 3 of 43 one-copy Scotties had CMO reported, so most did not. Know the result and watch closely through the growth months.

Inside the genotype table: Scotties beside Westies and Cairns

The 2016 study genotyped 1,052 dogs: 695 West Highland White Terriers, 249 Scottish Terriers and 108 Cairn Terriers. Its supplementary table is the only published source on how often CMO was reported in each genotype group.

Among Scotties, 10 of 249 (4%) had two copies, and all 10 had CMO reported. Forty-three (17%) had one copy, and 3 of them (7%) had CMO reported. One affected Scottie had no copy, and no other change was found in its SLC37A2 coding sequence. For comparison, 48% of 123 two-copy Westies and 3.6% of 275 one-copy Westies had CMO reported; in Cairns, all 9 two-copy dogs and 3 of 15 one-copy dogs did.

Three cautions apply. First, the sample was Europe-centred and enriched for affected dogs, so 4% and 17% are not frequencies for US or UK Scotties; none has been published. Second, the table notes that “clinically confirmed phenotype information in regard to CMO was not available for population controls”. A dog listed as unaffected may have had a mild, unrecorded episode, so 7% is not a penetrance estimate. Third, ten two-copy Scotties is a small group: every one had CMO reported, but that does not make it a rule.

The table can also be read in reverse. Of 96 affected dogs across the three breeds, 78 had two copies, 16 had one and 2 had none. That describes affected dogs, not the share of each genotype that became ill.

Thirty years of searching before the 2016 answer

CMO has appeared in veterinary journals since at least the 1960s; Riser and colleagues published a radiology paper titled “Canine craniomandibular osteopathy” in 1967. By 1979 Thornburg, in the American Journal of Pathology, presented it as an animal model of Caffey disease in human infants.

The first genetic study came in 1986. Padgett and Mostosky described CMO as a disease “principally” of West Highland White and Scottish terriers, analysed one Westie family and concluded that “the condition is an autosomal recessive trait.” A North American database study gives a sense of scale: LaFond, Breur and Austin counted only 35 CMO diagnoses over 1986–1995 across ten veterinary teaching hospitals, the fewest of the twelve developmental bone diseases they examined.

American clubs paid for much of the search. The AKC Canine Health Foundation’s grant directory lists a 1998–2000 project toward a CMO DNA test in Cairn, Westie and Scottish Terriers, which ruled out candidate genes after examining “about 30 percent of the genome”. A 2006–2007 linkage grant to Patrick Venta at Michigan State University, for $18,605.84, was sponsored by the Foundation of the Cairn Terrier Club of America, the Scottish Terrier Club of America and the Westie Foundation of America. Its summary notes that finding carriers by test breeding was possible but had “many undesirable aspects”.

The answer came through a European collaboration, with the Westie Foundation of America sending stored samples to Switzerland. A 2012 joint announcement said a direct DNA test would launch there that December, and the full analysis appeared in 2016: a genome-wide scan of 51 Westies pointed to chromosome 5, fine mapping in 88 dogs narrowed the region, and sequencing found the SLC37A2 change (p = 6.62 × 10^-303 across 96 cases).

Not every paper has caught up. The 2024 CT study still says in its introduction that recessive inheritance “has been proven” in Westies. When older articles or breeders talk about CMO “carriers”, they are often working from the 1986 model.

SamWhy did one gene take thirty years to find in breeds this well known?

Elena MarshThe 1986 family study pointed toward a recessive pattern, and the 2016 paper notes that mild cases can resolve unseen, which blurs any family tree.

A clear result is not a guarantee: other variants, other causes

The 2016 result answered the question for most affected terriers, not all. One Scottie and one Cairn had CMO without the variant, and the authors listed three explanations: a look-alike condition, a misdiagnosis or a different genetic cause.

Later work supports the third. In 2020 Letko and colleagues at the University of Bern, with the Helsinki group, sequenced the genomes of seven dogs with CMO from seven breeds and one American Staffordshire Terrier with skull hyperostosis, against 584 reference genomes. No variant was shared. A Basset Hound carried a different change in the same gene, SLC37A2 c.1446+1G>A, judged “most likely pathogenic”; variants in other genes in an Australian Terrier and a Weimaraner remain of uncertain significance. The authors concluded: “Our study indicates allelic and locus heterogeneity for canine CMO”. In Deutsch Drahthaar dogs, a 2018 Hannover analysis did not find the terrier variant and called the inheritance “more complex than a simple Mendelian model”.

For a Scottie owner, a test reporting only c.1332C>T says nothing about the Basset variant or undiscovered causes, so a clear result lowers known genetic risk without ruling CMO out. And a puppy with jaw pain needs a diagnosis whatever the report says. Researchers also debate where CMO ends: a 2011 New Zealand case report and a 2025 study of a Belgian Malinois litter both suggest grouping it with related juvenile skull conditions as “idiopathic canine juvenile cranial hyperostosis”.

Paperwork for a Scottie puppy: the STCA rule and the UK contrast

The Scottish Terrier Club of America took an unusual step in 2024. Its CHIC requirements, approved on February 11, 2024 and effective January 1, 2025, make three items mandatory for a Canine Health Information Center number: von Willebrand’s disease type III, patellar luxation and “a DNA based CMO test from an approved lab, results registered with OFA”, plus one elective test. The document states that “all Scottish Terriers should have a CHIC number.” The AKC’s terrier-group list of parent-club recommendations includes CMO for the Scottie and the Westie (added by the Westie club in 2019), but not the Cairn.

Scotties ranked No. 57 of 202 breeds in the AKC’s 2025 popularity table, so buyers may not see many litters. A CHIC number shows the breeder completed the club’s screening and published results; it does not mean the parents are free of the variant. Ask for each parent’s CMO genotype and check it in the OFA database.

The OFA does not run a CMO test itself; CMO is absent from its OFA/University of Missouri list. Owners test at an approved lab and register the result, which costs $15 per test for an individual dog (OFA fee schedule, checked October 10, 2026). Embark says its results are accepted wherever the OFA has a DNA registry, but leaves it to customers to confirm what their club accepts.

Prices change often; these are from October 10, 2026. Embark’s breed and health kit was $129 on a fall sale (regularly $159), and its breeder kit listed at $139. Wisdom Panel Premium was $127.99 on sale (regularly $159.99) in the US and £144.99 in the UK. Wisdom Panel Premium lists CMO (the Scottish Terrier variant) among its screened conditions, and Embark lists it on its condition page; check which variant a report covers before relying on it. Single tests cost less: LABOKLIN UK charges £48.00 including 20% VAT (service 8348), or £132.00 for a Scottish Terrier bundle that adds CDPA/CDDY and vWD type III; Genomia in the Czech Republic lists $56.00 excluding VAT, with about 12 business days’ turnaround.

The UK is different. The Royal Kennel Club says the Scottish Terrier “does not have any recommended pre-breeding health screening in place”, and LABOKLIN UK’s catalogue does not mark CMO as part of the Kennel Club DNA testing scheme. Testing is voluntary. The legal backstop is general: licensed breeders in England may not keep a dog for breeding if its “genotype, phenotype or state of health” means breeding could harm it or its offspring (SI 2018/486, Schedule 6). The rule does not name CMO, and how it applies to a one-copy dog is a judgement, not a settled fact.

SamThe breeder’s dogs have CHIC numbers. Does that mean my puppy cannot get CMO?

Elena MarshIt means the parents were tested and results are public. The 2016 study found CMO in one-copy Scotties and in one with no copy, so ask to see each parent’s genotype.

When a Scottie puppy stops wanting to chew

The signs to act on are concrete: a puppy of roughly 3 to 12 months that cries when yawning or picking up food, drops kibble, drools, resists having the lower jaw touched, has a hard swelling along the jaw or runs a fever. Any of these warrants a prompt visit to your regular vet rather than waiting.

Your vet will usually start with an examination and skull X-rays. The 2024 study argues that CT shows the ear bulla, airway and jaw joint far better, so if those look involved your vet may refer you to a teaching hospital or specialist with CT. In the US, the ACVS directory of board-certified surgeons, the American Veterinary Dental College and the American College of Veterinary Radiology list specialists; in the UK, the RCVS Find a Vet service has a specialist search. Bring the DNA report, but the diagnosis rests on examination and imaging.

Not every swollen puppy jaw is CMO. A Swedish study of 12 Irish Setters with canine leucocyte adhesion deficiency, an inherited immune disorder, found bone lesions consistent with CMO among other changes, and all were euthanised before six months because of severe, incurable infections. Skull hyperostosis in young Bullmastiffs and periostitis ossificans in Belgian Malinois puppies are further look-alikes.

Published reports describe treatment as pain control and supportive care, with surgery in severe cases. Anti-inflammatories, opioids and corticosteroids appear in the literature; the choice and dose belong to your vet. Embark’s guidance adds soft food when chewing hurts and avoiding hard chews. No treatment has been shown to cure or prevent CMO, and no controlled trial has compared options. The 2016 authors float vitamin D as an idea to explore, but it is an untested hypothesis, not advice.

Insurance timing follows the age of onset: cover started before any signs, ideally at collection, is what counts. Embrace says it covers hereditary and congenital conditions not pre-existing at enrolment; Nationwide states that pre-existing conditions are not covered. In the UK, Agria treats inherited conditions as pre-existing if signs or a diagnosis came before cover began. Some US states have puppy lemon laws; New York’s General Business Law § 753 covers a vet-certified “congenital malformation” within 180 days of sale from a pet dealer, but whether CMO qualifies is not settled by any source we found.

SamIf my puppy is diagnosed, will it be in pain for months?

Elena MarshReports vary. A 1985 Japanese Scottie recovered over five months, yet in the 2024 CT series pain relief before referral helped little. Your vet adjusts care as it changes.

A puppy disease that mirrors one in human infants

Part of the scientific interest comes from a human counterpart. Caffey disease, or infantile cortical hyperostosis, causes episodes of new bone in babies within the first five months of life and usually resolves on its own. In 2005 Gensure and colleagues traced the dominant familial form to one change in COL1A1, the type I collagen gene, and noted that some carriers of it never have bone episodes: incomplete penetrance again. A 2014 review in Bone calls the disease “largely self-limiting” and notes that nobody yet knows why one point change causes bone overgrowth limited in time and place.

The canine gene is different. Hytönen and Lohi, in Rare Diseases in 2016, listed Caffey (SLC37A2) among their dog models of rare human disorders, and the main 2016 paper proposed SLC37A2 as a new candidate gene for human Caffey disease. No human case caused by SLC37A2 appeared in the sources reviewed here.

The gene’s job hints at why bone is affected. SLC37A2 codes for a sugar-phosphate transporter that is especially active in bone marrow, macrophages and osteoclasts, the cells that resorb bone. A 2023 study found that mice lacking it build up unusually high bone mass because resorption and formation fall out of balance. The 2016 authors suggest that disturbed sugar supply in osteoclasts could tip a growing skull toward excess bone, but this has not been shown in dogs.

Frequently asked questions

Q. Are Scottish Terriers the breed most prone to CMO?
The evidence does not say that. The 2016 genetic study reported the highest frequency in West Highland White Terriers. Scotties were 65% of the 2024 CT study, but that was a referral group in which one hospital saw many related cases. What the study does show is that its Scotties had the most severe changes.

Q. My Scottie has one copy of the CMO variant. Is it just a carrier?
No. Because CMO appears to be dominant with incomplete penetrance, a one-copy dog carries some risk itself. In the 2016 study, 3 of 43 one-copy Scotties had CMO reported, so most did not. Watch closely through the growth months and tell your vet the result.

Q. My puppy tested clear. Can it still develop lion jaw?
It is less likely, but possible. One affected Scottie in the 2016 study had no copy of the variant, and a different SLC37A2 variant has been found in a Basset Hound. A clear result covers the variant tested, not every cause.

Q. Will a two-copy Scottie definitely get CMO?
All 10 two-copy Scotties in the 2016 study had CMO reported, but ten dogs from a case-enriched sample cannot support a rule, and some two-copy Westies had no signs reported. Treat two copies as high risk and plan veterinary follow-up in the first year.

Q. Does CMO always go away once the puppy stops growing?
Often the bone growth stops around one year and signs settle, but not always. In the 2024 CT series, five of 20 dogs had surgery and one was euthanised, and dogs whose jaw joint fuses have a poor outlook. Your veterinarian is the right person to judge an individual dog’s prognosis.

Q. Is a CMO DNA test required for Scottish Terriers?
It is a breed-club requirement tied to the CHIC number. Since January 1, 2025 the Scottish Terrier Club of America requires an OFA-registered CMO DNA result for a CHIC number, alongside vWD type III and patellar screening. In the UK, the Royal Kennel Club has no recommended pre-breeding screening for the breed.

Q. Can a DNA test diagnose my puppy’s sore jaw?
No. It tells you about genetic risk. A veterinarian diagnoses CMO through examination and imaging such as X-rays or CT, and other conditions can look similar. If your puppy will not open its mouth, stops eating, has a swollen jaw or a fever, see your vet promptly.

References

• Hytönen MK, Arumilli M, Lappalainen AK, Owczarek-Lipska M, Jagannathan V, Hundi S, Salmela E, Venta P, Sarkiala E, Jokinen T, Gorgas D, Kere J, Nieminen P, Drögemüller C, Lohi H (2016). Molecular characterization of three canine models of human rare bone diseases: Caffey, van den Ende-Gupta, and Raine syndromes. PLoS Genetics 12(5):e1006037.
• Pérez López LA, Almansa Ruiz JC, Steenkamp G, Holdsworth A (2024). Computed tomographic characteristics of craniomandibular osteopathy in 20 dogs. Frontiers in Veterinary Science 11:1436356.
• Letko A, Leuthard F, Jagannathan V, Corlazzoli D, Matiasek K, Schweizer D, Hytönen MK, Lohi H, Leeb T, Drögemüller C (2020). Whole genome sequencing indicates heterogeneity of hyperostotic disorders in dogs. Genes (Basel) 11(2):163.
• Vagt J, Distl O (2018). Complex segregation analysis of craniomandibular osteopathy in Deutsch Drahthaar dogs. The Veterinary Journal 231:30-32.
• Padgett GA, Mostosky UV (1986). Animal model: the mode of inheritance of craniomandibular osteopathy in West Highland White terrier dogs. American Journal of Medical Genetics 25(1):9-13.
• LaFond E, Breur GJ, Austin CC (2002). Breed susceptibility for developmental orthopedic diseases in dogs. Journal of the American Animal Hospital Association 38(5):467-477.
• Riser WH, Parkes LJ, Shirer JF (1967). Canine craniomandibular osteopathy. Veterinary Radiology 8(1):23-31.
• Thornburg LP (1979). Infantile cortical hyperostosis (Caffey-Silverman syndrome). Animal model: craniomandibular osteopathy in the canine. American Journal of Pathology 95(2):575-578.
• Hashimoto M (1985). Canine craniomandibular osteopathy successfully treated. Journal of the Japan Veterinary Medical Association 38(6):393-395.
• Thompson DJ, Rogers W, Owen MC, Thompson KG (2011). Idiopathic canine juvenile cranial hyperostosis in a Pit Bull Terrier. New Zealand Veterinary Journal 59(4):201-205.
• Bar Giora Y, Mann O, Kachtan I, Bar Am Y (2025). Periostitis ossificans in a litter of Belgian Malinois puppies. Journal of Veterinary Dentistry 43(3):262-277.
• Pastor KF, Boulay JP, Schelling SH, Carpenter JL (2000). Idiopathic hyperostosis of the calvaria in five young bullmastiffs. Journal of the American Animal Hospital Association 36(5):439-445.
• Trowald-Wigh G, Ekman S, Hansson K, Hedhammar A, Hård af Segerstad C (2000). Clinical, radiological and pathological features of 12 Irish setters with canine leucocyte adhesion deficiency. Journal of Small Animal Practice 41(5):211-217.
• Gensure RC, Mäkitie O, Barclay C, Chan C, Depalma SR, Bastepe M, Abuzahra H, Couper R, Mundlos S, Sillence D, Ala Kokko L, Seidman JG, Cole WG, Jüppner H (2005). A novel COL1A1 mutation in infantile cortical hyperostosis (Caffey disease) expands the spectrum of collagen-related disorders. Journal of Clinical Investigation 115(5):1250-1257.
• Nistala H, Mäkitie O, Jüppner H (2014). Caffey disease: new perspectives on old questions. Bone 60:246-251.
• Hytönen MK, Lohi H (2016). Canine models of human rare disorders. Rare Diseases 4(1):e1241362.
• Ng PY, Ribet ABP, Guo Q, Mullin BH, Tan JWY, Landao-Bassonga E, Stephens S, Chen K, Yuan J, Abudulai L, Bollen M, Nguyen ETTT, Kular J, Papadimitriou JM, Søe K, Teasdale RD, Xu J, Parton RG, Takayanagi H, Pavlos NJ (2023). Sugar transporter Slc37a2 regulates bone metabolism in mice via a tubular lysosomal network in osteoclasts. Nature Communications 14(1):906.
• Wisdom Panel (Mars Petcare). Craniomandibular Osteopathy (Discovered in Scottish Terrier breeds). Health condition page. Accessed 2026-10-10.
• LABOKLIN. Craniomandibular osteopathy (CMO), LABOKLIN Service ID 8348. Laboratory product page. Accessed 2026-10-10.

About the author

Elena Marsh

Editor & writer (not a veterinarian)

A writer with a molecular-biology background and a lifelong dog and cat owner. Not a veterinarian — she translates peer-reviewed genetics research and primary data into plain language, always as information rather than diagnosis.