The Rough Collie holds a unique and somewhat unfortunate distinction in veterinary ophthalmology: no other breed has been studied as extensively for a single inherited eye condition as the Rough Collie has been for Collie Eye Anomaly. This intensive scrutiny is partly a consequence of the breed's popularity following the Lassie films and television series, and partly because the mutation frequency is simply so high. When I began my career in the early 1990s, prevalence figures of 70-90% affected or carrier were considered normal for this breed. Understanding this history and what has changed since is essential context for any Rough Collie enthusiast.
Why the Rough Collie? Historical Prevalence
The NHEJ1 mutation responsible for CEA almost certainly arose in British herding breeds many generations ago, before modern registries and selective breeding records existed. As the Rough Collie breed was formalised and its population grew in the 19th and early 20th centuries, the mutation spread. Because affected dogs are not visually impaired in the vast majority of cases — most having only choroidal hypoplasia — there was no selection pressure against the mutation. Dogs with CEA competed and won at the highest levels of dog shows, and the mutation was propagated through influential sires without anyone realising it carried an inherited eye condition.
By the time systematic ophthalmoscopic examination began in the 1960s and 1970s, the mutation had reached extraordinary frequency. Studies in the UK and North America documented affected and carrier rates that would be considered alarming in any other breed context. The mutation was, and in many populations remains, almost ubiquitous.
Clinical Presentation in Rough Collies
The Rough Collie generally presents with a somewhat more benign clinical picture than some other affected breeds. In my practice, approximately 80% of affected Rough Collies have choroidal hypoplasia as their only finding. The lesions are typically lateral to the optic disc, bilateral, and of modest extent. Vision is unaffected and quality of life is entirely normal.
Colobomas occur in perhaps 15-18% of affected Rough Collies, a lower proportion than in Shetland Sheepdogs or Border Collies. Severe colobomas causing significant visual impairment are relatively uncommon in the breed, though they do occur. Retinal detachment and intraocular haemorrhage are rare but not unknown. I have seen perhaps three or four Rough Collies with retinal detachment attributable to CEA in nearly three decades of practice — a small number in a breed with such high affected prevalence, reflecting the relative mildness of typical Rough Collie presentations.
Understanding the full spectrum of CEA severity from mild choroidal hypoplasia through to serious complications helps Rough Collie owners place their dog's specific diagnosis in appropriate context.
Typical Rough Collie Presentation
Bilateral choroidal hypoplasia, 1-3 disc diameters in size, temporal to optic disc. Vision normal. Excellent prognosis. No management required beyond excluding from breeding or pairing with clear dogs.
Intermediate Presentation
Choroidal hypoplasia plus small colobomas. Found in approximately 10-15% of affected Rough Collies. Annual monitoring recommended. Vision usually preserved.
Severe Presentation
Large colobomas with risk of retinal complications. Uncommon in Rough Collies. Requires specialist monitoring and owner education about warning signs.
The "Go Normal" Challenge in Rough Collies
The Rough Collie is particularly prone to the "go normal" phenomenon, where affected dogs become difficult or impossible to diagnose ophthalmoscopically as adults. This is partly because Rough Collie CEA lesions tend to be mild and partly because the breed's typically heavy coat pigmentation genes may influence retinal pigment deposition.
I have examined Rough Collies at breed club screening events where a significant minority of adults returned abnormal genetic test results despite appearing entirely normal ophthalmoscopically. Without genetic testing, these dogs would have been incorrectly recorded as clear and could have been bred without appropriate consideration of their carrier or affected status.
This underscores why every Rough Collie being considered for breeding must have a DNA test regardless of their clinical appearance. Ophthalmoscopic examination remains important, particularly for establishing severity grade in known affected dogs, but it cannot substitute for genetic testing in adults.
Breed Club Requirements and Registration
The Rough Collie breed clubs in the UK, USA, and several European countries have developed health testing recommendations that address CEA. The specifics vary by country and organisation, but the general principle is consistent: breeding dogs should be tested, and results should be recorded with breed registries.
In the UK, the Kennel Club's Breed Health and Conservation Plan for Rough Collies includes CEA as a priority condition. The KC's DNA testing service allows results to be automatically registered, creating a publicly accessible health database. American breeders working within AKC-registered programmes increasingly use OFA health testing protocols that include CEA alongside other herding breed conditions.
The puppy screening examination at 6-8 weeks, conducted by a board-certified veterinary ophthalmologist, is the cornerstone of responsible Rough Collie breeding practice. Many breed clubs host organised litter screening events that make this accessible and affordable for breeders.
Breeding Strategy for Rough Collie Breeders
Given the extraordinary prevalence of the CEA mutation in Rough Collies, the question of breeding strategy requires careful thought. Simply excluding all affected and carrier dogs from breeding would eliminate the vast majority of the breed's genetic pool and cause enormous damage to diversity. This is not what I recommend, and it is not what breed health organisations advise.
The sensible approach, detailed in my article on breeding strategies to reduce CEA incidence, is to use the genetic test to inform, rather than dictate, mating decisions. The guiding principle is: never produce affected puppies when it can be avoided without compromising other health and temperament goals.
Practical Mating Rules
- Clear x Clear: All puppies will be clear. Ideal where such pairings are possible without compromising genetic diversity.
- Clear x Carrier: 50% clear, 50% carrier offspring. No affected puppies produced. Acceptable for maintaining genetic diversity whilst reducing mutation frequency.
- Carrier x Carrier: 25% clear, 50% carrier, 25% affected. Produces affected puppies; generally avoidable and therefore generally inadvisable.
- Clear x Affected: 100% carrier offspring. No affected puppies. Sometimes used when an otherwise exceptional affected dog contributes valuable traits or diversity.
- Carrier x Affected and Affected x Affected: Produce affected offspring; avoid unless exceptional circumstances justify it with full informed owner disclosure.
The "Go Normal" Dog in the Breeding Programme
A particular challenge arises with dogs who were never examined as puppies, or for whom puppy records have been lost. An adult Rough Collie who appears ophthalmoscopically normal might be genuinely clear, a carrier, or an affected dog who has "gone normal." Only genetic testing distinguishes these possibilities.
I have sat across the consultation table from breeders devastated to learn that their ophthalmoscopically normal adult champion, who has already produced multiple litters, is genetically homozygous affected. This discovery does not mean previous breeding decisions were irresponsible — they were made in good faith with available information — but it does mean adjusting plans going forward. The research into modifier genes that influence how severely the mutation is expressed may eventually help us understand why such dogs appear entirely normal clinically despite carrying two affected alleles.
Famous Rough Collies and CEA
It would be surprising if the breed's most celebrated representatives were immune to a mutation affecting 70-90% of the population. Several Lassie descendants and influential show champions have been retrospectively identified as carriers or affected dogs through genetic testing of their progeny. This is not a reflection on the quality of these animals — they were exceptional in every other respect — but it illustrates how deeply embedded the mutation became before testing was available.
The lesson is that excellence in conformation, temperament, and working ability does not correlate with genetic status for CEA. We must test every dog regardless of pedigree prestige.
Looking at the International Picture
The Rough Collie is bred worldwide, and the prevalence of CEA varies considerably between national populations depending on the history of testing and breeding management. Some Scandinavian populations with decades of mandatory health testing show meaningfully lower affected rates than populations where testing has been inconsistent. The international data on CEA prevalence changes over time provides encouraging evidence that coordinated breed-wide testing programmes genuinely move the needle over generations.
Finnish and Norwegian Rough Collie clubs in particular have demonstrated what consistent, mandatory testing combined with transparent recording can achieve. These success stories should inspire breeders worldwide who might feel their individual efforts are insufficient to change population-level statistics. They matter enormously — every clear mating reduces the mutation frequency in the next generation, and these incremental gains accumulate into meaningful population change over decades.