Retinopathies--Old and New
World Small Animal Veterinary Association World Congress Proceedings, 2004
Peter G.C. Bedford, BVetMed, PhD, FRCVS, DVOphthal, DECVO, ILTM
Department of Veterinary Clinical Sciences, The Royal Veterinary College
North Mymms, Hatfield, Herts, UK


The combined use of the short-acting mydriatic tropicamide and the indirect ophthalmoscope puts the fundus within easy grasp of every practitioner. This simple technique, once practiced and polished, allows diagnostic examination of the retina and optic nerve in all but the most difficult of canine and feline patients. Interpretation of lesions in the posterior segment requires some education, but the combination of a transparent tissue stretched over a mirrored surface offers no hiding place for any pathological process involving the optic disc and tapetal fundus, while pigment disturbance is usually readily observed within the non-tapetal fundus. Many retinopathies are often detected as the result of routine screening in the absence of dramatic clinical features or noticeable disturbance of sight for, while our clients may notice the painful or discoloured eye, few recognise the subtle variation in pupil size and many do not suspect sight deficiency until it is well established.

The pathognomic features of retinal disease defined by ophthalmoscopic examination are changes in tapetal reflectivity and pigmentation, blood vessel congestion or attenuation, haemorrhage and retinal nonattachment. Thus, a combination of reduced tapetal reflectivity and intraretinal haemorrhage indicates active inflammation, whereas a zone of increased tapetal reflectivity or a patch of reduced pigmentation signifies postinflammatory degeneration. The ophthalmoscopic changes can only dictate inflammation and the precise cause of that inflammation may remain obscure until further investigations have been completed. Unfortunately, retinal pathology does not automatically flag its cause and, as such, dilemma in diagnosis is always possible. For example, the retinal degeneration which accompanies retinal dysplasia can be confused with post-inflammatory retinal degeneration and, without history, sudden acquired retinal degeneration (SARD) can look like progressive retinal atrophy (PRA).

The range of retinopathy in the dog and cat has been largely defined and new retinopathies are relatively few and far between. Central PRA appears to be a misnomer because the condition is a secondary photoreceptor degeneration due to abnormal retinal pigment epithelial activity, activity which appears to be governed by environmental or metabolic factors to some extent. Thus, a classical ocular diagnosis becomes an ocular feature of a condition of unspecified aetiology, yet one that demonstrates breed disposition. The diagnosis of SARD related to photoreceptor blowout has detracted from retrobulbar optic neuritis as a canine diagnosis and secondary hypertension is rapidly being established as the cause of retinal detachment in the aged cat. The aetiology of extensive bilateral retinal degeneration in this species remains a cause of considerable speculation whilst more canine breeds become involved in the PRA story of inherited photoreceptor degeneration. Similarly more breeds are becoming involved in retinal dysplasia; the multifocal and total forms of this condition have been joined by a geographical lesion in the Cavalier King Charles and Retriever breeds. Collie Eye Anomaly is still the commonest inherited ocular disease in dogs in the United Kingdom but its recent appearance in the Lancashire Heeler suggests that a change in nomenclature would be appropriate.


The term is applied to those inherited neuroretinal conditions which are seen clinically as either neuroretinal folds and rosettes or retinal non-attachment. However it should be remembered that within the progressive retinal atrophy complex there is both a rod and cone photoreceptor dysplasia in the Irish Setter and Rough Collie and a rod dysplasia accompanied by cone degeneration in the Norwegian Elkhound. The simplest manifestation of retinal dysplasia (R.D.) is a fold in the neuroretina, the affected dog demonstrating no associated visual impairment. Complicated folds in which there is proliferation of photoreceptor and RPE elements are also referred to as rosettes. This form of R.D. is inherited in the Cavalier King Charles Spaniel, the Hungarian Puli and the Rottweiler as a recessive trait. In the English Springer Spaniel the neuroretinal folds may be accompanied by retinal degeneration, these lesions taking on the appearance of post-inflammatory retinopathy due to the presence of melanin pigmentation. Occasionally retinal detachment complicates the clinical picture and both intraocular haemorrhage and cataract formation may be seen. The incidence is particularly high in the working strains of this breed and again inheritance is as a simple recessive trait. Total retinal non-attachment in which the neuroretina is seen lying within the vitreous has been described in the Sealyham Terrier, Bedlington Terrier and the Labrador Retriever, but the incidence is very low. In this latter breed congenital neuroretinal non-attachment also accompanies skeletal chondrodysplasia, the affected dogs demonstrating retarded growth of the radius, ulna and tibia.

The fate of the fold is the source of much conjecture and hypothesis. Is a neuroretinal fold indicative of the presence of retinal dysplasia? Why do some folds disappear? Is the term "go normal" appropriate for this condition? How do we differentiate with the ophthalmoscope between focal degeneration of inflammatory origin and that which is associated with the fold?

I believe that all folds means a diagnosis of retinal dysplasia must be used. Diagnosis at 6 to 7 weeks of age is an essential control measure for a go normal state does exist. However, I believe that today's fold that has disappeared is tomorrow's focal degeneration. Tomorrow may be a long time in that there may be a "normal" ophthalmoscopic appearance for several months to a year, and possibly longer.


During organogenesis it is the cells of the posterior wall of the invaginating optic vesicle which form the retinal pigment epithelium. Failure to express growth hormone by these cells affects the subsequent differentiation of the ocular tissues. In Collie Eye Anomaly (CEA) the choroid remains hypoplastic in an area lateral to the optic disc and there may be failure of the foetal fissure to close leaving a colobomatous defect involving either papillary or peripapillary tissue. The degree of choroidal hypoplasia and the size of the colobomata vary considerably between affected individuals and even between the eyes of the same individual. All affected puppies demonstrate choroidal hypoplasia but by the age of twelve to sixteen weeks many may have masked the smaller lesions by melanin pigmentation. The estimates vary but in the U.K. it is likely that some thirty per cent of affected puppies demonstrate this masking procedure: somewhat confusingly this process is described as "go normal" status. The phenotype thus appears ophthalmoscopically normal but genetically these dogs are affected and must be avoided in disease control programmes. It is of considerable significance, underlying the necessity for screening all litters in the affected breeds.

Some thirty per cent of those demonstrating choroidal hypoplasia as adults also possess colobomatous defects. Papillary colobomata involve the optic disc and vary considerably in size, on occasion the whole disc being affected. The peripapillary colobomata are not as common but again they can vary in size. These defects are considered to be due to the failure of scleral tissue to develop in areas associated with the closure of the foetal fissure, but explanation of the atypical coloboma is less easy to ascribe to this process. Peripapillary colobomata are lined with degenerate neuroretinal tissue and those dogs with large papillary colobomata must have defective sight. The role of the coloboma in post-natal retinal detachment is not completely understood but they are considered to be involved in this process.

In a small percentage of affected dogs the neuroretina is non-attached at birth whilst in those with significant colobomatous defects the neuroretina may detach within the first three years of life. About one per cent of affected dogs present with intraocular haemorrhage but the aetiology of this complication remains open to speculation. It is considered that pre-retinal capillaries at the equatorial parts of the globe may develop as a response to retinal anoxia caused by choroidal hypoplasia. The hypothesis is that it is these unsupported vessels which rupture to cause the hyphaema. However haemorrhage can be associated with persistence of the hyaloid vasculature and unsupported blood vessels within the colobomata.

The diagnostic picture: is understood well and painted in the above terms it would appear to be straightforward. However, like most things in life the story is not as black and white as it may seem and we do see problems which, at the very least, should provide discussion. I believe that there is a possible ten per cent error in diagnosis due to several features:

1.  The small papillary coloboma in the six week old puppy

2.  The significance of the pale pink patch in the six week old fundus

3.  The reduction of peripapillary pigmentation in the sable and white or colour dilute dog

4.  The merle eye

Add to this the "go normal" phenomenon and the significance of the coloboma which is unaccompanied by choroidal hypoplasia. Thus life behind the ophthalmoscope can become difficult!

Estimations of incidence vary but it may be that CEA still enjoys an inordinately high incidence in the Rough Collie, Smooth Collie and Shetland Sheepdog breeds in the U.K. CEA is also seen in the Border Collie but the incidence is low, probably between inherited as a simple autosomal recessive trait, but there are contra views. Without doubt the choroidal hypoplasia is a recessive trait but there is some evidence to suggest that the coloboma may occur as a condition in its own right. The high incidence of the choroidal hypoplasia means that colobomata are most often seen accompanying this lesion.

The latest complication relates to the fact that CEA can be described in non-collie breeds. The literature contains the odd reference to CEA with borzoi, beagle, dachshund, German Shepherd dog, Great Dane, Maltese terrier, miniature and toy poodles and the Siberian husky. These reports usually refer to individual cases or single lesions, and breed predispositions have not been adequately demonstrated. However a future publication will report CEA in the Lancashire Heeler breed of terrier, and this will put the cat among the pigeons! The author suggests that new terminology is required and suggests that "Congenital Posterior Segment Anomaly" (CPSA) might fit the bill.


Progressive retinal atrophy (PRA) is the umbrella term used to describe a number of inherited neuroretinal degenerations. Generalised PRA, or simply PRA, describes those degenerations in which the primary focus of disease is the photoreceptor unit. Such degenerations are characterised by a nyctalopia which progresses to total blindness and involves a high incidence of secondary cataract formation. All these diseases are inherited as simple autosomal recessive traits. Central PRA (CPRA) is retinal degeneration in which there is photoreceptor disease, but this is secondary to disease of the retinal pigment epithelium (RPE). In consequence CPRA has been reclassified as retinal pigment epithelial dystrophy (RPED). The inheritance pattern remains undetermined. Both PRA and RPED may be present in the same breed.

Dysplasia of the rod and cone photoreceptors has been described in the Irish Setter and Rough Collie breeds. As such this type of PRA is an early onset disease with severe impairment of vision being present at eight months of age and total blindness at twelve months. The photoreceptor defect is an enzyme abnormality within the phototransduction cascade. Specifically the retinal level of the nucleotide cyclic guanosine monophosphate (cGMP) is elevated to approximately ten times its normal value due to reduced cGMP-phosphodiesterase activity. Early onset means early diagnosis and this, together with a known pattern of inheritance, means that effective disease control is possible. It is of interest to note that an identical rod/cone dysplasia has been described in the Abyssinian cat in the United Kingdom.

A second form of PRA in which the rod photoreceptor unit is dysplastic and there is subsequent degeneration of normal cone photoreceptors has been described in the Norwegian Elkhound. The initiatory rod defect remains undetermined, as does the cause of a third form of PRA, a disease seen classically in the Miniature and Toy Poodle breeds, the English Cocker Spaniel and the Labrador Retriever. Here there is normal development of both photoreceptor units but blindness is caused by their premature degeneration in middle age. It is of interest to note that it is the same genetic abnormality which causes the disease in the Poodle and the Cocker Spaniel breeds, but a considerable research effort will be necessary to define the aetiology as completely as that of the photoreceptor dysplasia of the Irish Setter. Recent work has demonstrated that the genetic defect in the Irish Setter is abnormality of the beta sub unit of cGMP phosphodiesterase and that both disease and carrier state may be determined by a relatively simple DNA test. Obviously progress along these lines with the other forms of PRA is impatiently awaited for it will render the control and possible eradication of PRA a realistic goal.

Within the last twelve months two more breeds have been added to the long list of PRA affected breeds. Classical presumed rod/cone degeneration has been described for the Golden Retriever and the Lhasa Apso. Who knows what next year will bring, but there is a suspicion of PRA in the miniature smooth-haired dachshund and the Yorkshire terrier in the U.K. Of course you know it is present in the Papillon, but this one has not yet made an impact in the U.K.


Originally considered to be a primary photoreceptor degeneration, this disease is due to defect of the RPE. One of the many important functions of RPE cells is the degradation of utilised photoreceptor outer segment (POS). The process of phototransduction is localised to the visual pigment within the POS and there is rapid turnover of this material. Rod outer segment renewal takes approximately ten days, the basic materials for outer segment production being provided by the RPE cells as a result of their breakdown of POS. Dystrophic RPE cells can neither degrade utilised POS quickly enough nor effectively participate in POS production. Their cytoplasm accumulates phagocytosed POS material and their many other functions, in terms of neuroretinal support cease. Thus the rod and cone photoreceptors degenerate and sight is affected. However, melanin protects against this disease process and the neuroretina overlying the pigmented RPE cells of the non-tapetal fundus will continue to function normally long into the disease process. Affected dogs therefore lose their central field of vision but maintain peripheral sight. There is undoubtedly genetic predisposition to this disease as witnessed by specific breed involvement but many factors influence the course of degeneration.

Retinal pigment epithelial dystrophy (RPED) is characterised by an accumulation of autofluorescent lipopigment within RPE cells which is recognised ophthalmoscopically by the appearance of brown spots and patches, particularly within the tapetal fundus, and associated retinal degeneration. The ophthalmoscopic changes are very similar to those observed in dogs which have been subjected to dietary vitamin E deficiency.

During a five year period (1992-1997) plasma samples were obtained from twenty-seven RPED affected dogs presented to the Ophthalmology Service at the Royal Veterinary College, University of London. Plasma samples obtained from fifty ophthalmoscopically normal, healthy dogs of various breeds served as controls. All plasma samples were immediately frozen and stored at-80°C prior to analysis by High Performance Liquid Chromatography. Oral Vitamin E Tolerance Tests (OVETT's) were conducted in five RPED affected dogs and in four ophthalmoscopically normal dogs which served as controls.

Plasma vitamin E levels were found to be lower in RPED affected Briards (n=14) than in ophthalmoscopically normal controls (8.3±0.51µg/ml; normal range 11.6-30µg/ml). In RPED affected Nizzinis (n=3) and Cocker Spaniels (n=10), in which plasma vitamin E levels were found to be extremely low (1.31±0.284 μg/ml), this difference was extremely significant (p<0.0001). The OVETTs revealed that although peak plasma levels of vitamin E attained after the administration of a large oral dose of the vitamin tended to be lower than peak levels attained in control dogs, the differences between the two groups with respect to indices of absorption of vitamin E (such as maximum increase in plasma levels, and area under the curve to 12 hours) were not statistically significant. However, the subsequent decline in plasma vitamin E levels was consistently more rapid in RPED-affected dogs than in control dogs. A similar response to the oral administration of large doses of vitamin E is observed in human patients with familial or isolated vitamin E deficiency.

Neurological signs were present in at least 5 of the 13 RPED affected Nizzinis and Cocker Spaniels which were included in this study. The most consistent neurological findings were ataxia and proprioceptive deficits which were most pronounced in the hind limbs. In the two Cocker Spaniels which were available for post-mortem examination, histopathologic evidence of widespread peroxidative injury was observed in addition to the characteristic retinal pathology of RPED. In both cases intestinal lipofuscinosis and a degenerative/dystrophic encephalomyelopathy (consistent with the clinical neurological findings) were prominent findings.

Thus a classical primary retinal degeneration may be nothing more than the ocular feature of an underlying systemic deficiency of an important anti oxidant, vitamin E.


Now that the "retrobulbar optic neuritis" diagnosis has been firmly placed on the top shelve, the aetiology of Sudden Acquired Retinal Degeneration (SARD) demands explanation. The cause of blindness is photoreceptor degeneration but what is responsible for this. Answers on a post card would be more than welcome!

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Peter G.C. Bedford, BVetMed, PhD, FRCVS, DVOphthal, DECVO, ILTM
Department of Veterinary Clinical Sciences, The Royal Veterinary College
North Mymms, Hatfield, Herts, UK

MAIN : Ophthalmology : Retinopathies
Powered By VIN