Feline retinal degeneration: clinical experience and new findings (1994-1997)

A retrospective case series of 26 cats with diffuse retinal degeneration is presented. The most common presenting complaints included bumping into objects, dilated pupils, and reluctance to jump. Ophthalmic examination findings were consistent with those reported in dogs with progressive retinal atrophy. Breed predilection of the Siamese cat was observed. Cats with primary retinal degeneration presented late in the clinical course of their disease, when vision loss was severe. Early symptoms such as night blindness and secondary ocular complications (i.e., cataract and retinal detachment), reported in dogs with progressive retinal degeneration, were not observed in this study. All cats showed excellent adaptive capabilities to blindness.     

Investigation of fellow eye of unilateral retinal detachment in Shih‐Tzu

Objective To investigate disease in the fellow eye, and consider the relation to rhegmatogenous retinal detachment (RRD) in Shih‐Tzus. Animals studied The fellow eyes of 49 Shih‐Tzus (27 male, 22 female; median age: 6.8 years) with unilateral RRD diagnosed by funduscopy or ultrasonography at Rakuno Gakuen University Teaching Animal Hospital were assessed in this study. Procedures Ophthalmic examinations (including menace response, pupillary light reflex, slit‐lamp biomicroscopy, and funduscopy) were performed in the subjects. Electroretinography was performed in 12 eyes that developed retinal degeneration. Maximum follow‐up period was 42 months. Results Cataracts and vitreous opacity were observed in 26 (53%) and 32 eyes (65%), respectively, by slit‐lamp biomicroscopy. Retinal degeneration with various degrees of hyper‐reflectivity of the tapetal fundus and/or attenuation of retinal vessels was observed in 35 eyes (71%) on funduscopy. A reduction of amplitude in rod, standard combined and 30 Hz flicker electroretingram was detected in 5 (42%), 10 (83%), and 6 eyes (50%), respectively. During the follow‐up period, RRD was detected in six eyes. Conclusion Retinal degeneration was frequently detected by funduscopy and electroretingrams in the fellow eye in Shih‐Tzus with RRD. In our subjects, vitreous degeneration was also observed frequently. It has been reported that peripheral retinal degeneration is one of the causes of RRD associated with vitreous degeneration in humans. We assume that primary retinal degeneration with secondary vitreous degeneration is one of the causes of RRD in Shih‐Tzus.     

Retinal degeneration in the dog and cat

Retinal degenerations in the dog and cat are an important cause of blindness in these species. Particularly in the dog, many retinal degenerations, collectively called progressive retinal atrophy, seen in clinical practice are inherited. The clinical signs, electrophysiological findings, pathology, and underlying biochemical defects in the retina vary from breed to breed. Specific categories of inherited retinal degeneration are now recognized, and classified into early onset photoreceptor dysplasias, late-onset retinal degenerations, or retinal degenerations secondary to primary RPE dystrophy. As new inherited retinal degenerations are reported in different breeds they can generally be assigned to one these categories. Other causes of retinal degeneration include nutritional deficiencies, glaucoma, inflammation, ischemia, and toxins. Idiopathic retinal degeneration occurs in the dog with some frequency.     

Progressive retinal atrophy in Abyssinian and Somali cats in the Netherlands (1981-2001)

From 1981 to 2001, 248 Abyssinian and 127 Somali cats in the Netherlands were examined for hereditary eye disease. Distinct ophthalmoscopic signs consistent with hereditary progressive retinal atrophy (PRA) were observed in 11 Abyssinian cats, and subtle signs in 3 Abyssinian cats. A familial relationship was detected in 13 out of 14 of these cats, which supports a hereditary basis to the condition. Distinct funduscopic signs of retinal degeneration were observed at a median age of 4 years. One cat with advanced retinal degeneration was only 7 months old, whereas the remaining 10 cats were between 2 and 12 years old at the time of diagnosis. These differences in the age of onset are suggestive of at least two types of PRA occurring in Abyssinian cats in the Netherlands: a dysplastic, early-onset and a late-onset retinal degeneration. A large-scale and systematic examination programme for hereditary eye disease will be necessary to assess the incidence of PRA in the Dutch population of Abyssinian and Somali cats as a whole, and to provide a basis for a preventive breeding programme.     

Canine inherited retinal degenerations: update on molecular genetic research and its clinical application

Inherited retinal degenerations in the dog include generalised progressive retinal atrophy, retinal pigment epithelial dystrophy, congenital stationary night blindness and day blindness (hemeralopia). The clinical phenotype and pathology of these diseases closely resemble some types of human inherited retinal degeneration, in particular retinitis pigmentosa, one of the most common inherited causes of blindness in man. Molecular genetic investigations aim to identify the genetic mutations underlying the canine inherited retinal degenerations. Two major research strategies, candidate gene analysis and linkage analysis, have been used. To date, candidate gene analysis has definitively identified the genetic mutations underlying nine inherited retinal degenerations, each in a different breed of dog, and linkage studies have identified genetic markers for a further retinal degeneration which is found in at least six different breeds. This review outlines the research strategy behind candidate gene and linkage studies and summarises recent results in the search for genetic causes of canine inherited retinal degenerations. The aim is to increase awareness of this rapidly changing field and to show how the research can be used to develop genetic tests for these diseases and thereby reduce the incidence of inherited eye disease in dogs.     

Effects of hereditary retinal degeneration due to a CEP290 mutation on the feline pupillary light reflex

Objective To understand how progressive rod cone degeneration due to a mutation in CEP290 affects the pupillary light reflex (PLR) in domestic cats. Animals studied Domestic cats identified as either normal wildtype (WT; n = 6), or homozygous for the rdAc mutation in CEP290 and having early stage retinal degeneration (stage 2, S2; n = 4), or advanced retinal degeneration (S4; n = 6). Methods The effect of light on pupil size was measured over a series of 10‐s pulses of white and chromatic light in cats lightly sedated with medetomidine. Results In WT cats, the PLR was characterized by a pronounced initial constriction that rapidly re‐dilated during the stimulus (pupil escape), to a stable or sustained constriction. There was then a marked constriction at stimulus offset. Each component of the PLR was retained in affected cats, but with progressively reduced irradiance sensitivity from early to advanced retinal disease. Conclusions The PLR of cats had multiple phases, with a remarkably high‐amplitude ‘paradoxical’ off‐constriction even in the absence of retinal disease. In rdAc cats, reduced irradiance sensitivity was consistent with progressive loss of rod and cone function. Based on previously characterized retinal pathology, this suggests the visual streak of the retina has a proportionally large contribution to PLR input. These findings support the hypothesis that the efficacy of planned therapeutic trials can be determined by careful evaluation of the PLR in cats.     

Familial non-rcd1 generalised retinal degeneration in Irish setters

Four Irish setters were diagnosed with bilateral retinal degeneration and cataracts at an age ranging from six to 11 years. In three of these dogs, progressive night blindness was reported from an age of eight to 11 years. In the fourth dog, aged six, no signs of visual impairment had been noticed. In all four dogs, the rod-cone dysplasia type 1 (rcd1) mutation was excluded as a cause, using an allele-specific PCR. From their three-generation pedigrees, a familial relationship was detected in three out of four dogs, which were also related to four additional Irish setter dogs with a history and clinical signs suggestive of late-onset progressive retinal degeneration. These results suggest the existence of a possibly hereditary, late-onset, progressive retinal atrophy in the Irish setter breed, that is distinct from rcd1.     

N -ethyl- N -nitrosourea induces retinal photoreceptor damage in adult rats

Seven-week-old male Lewis rats received a single intraperitoneal injection of N-ethyl-N-nitrosourea (ENU) (100, 200, 400 or 600 mg/kg), and retinal damage was evaluated 7 days after the treatment. Sequential morphological features of the retina and retinal DNA damage, as determined by a TUNEL assay and phospho-histone H2A.X (γ-H2AX), were analyzed 3, 6, 12, 24 and 72 hr, 7 days, and/or 30 days after 400 mg/kg ENU treatment. Activation of the nuclear enzyme poly (ADP-ribose) polymerase (PARP) was analyzed immunohistochemically by poly (ADP-ribose) (PAR) expression in response to DNA damage of the retina. All rats that received ≥ 400 mg/kg of ENU developed retinal degeneration characterized by the loss of photoreceptor cells in both the central and peripheral retina within 7 days. In the 400 mg/kg ENU-treated rats, TUNEL-positive signals were only located in the photoreceptor cells and peaked 24 hr after ENU treatment. The γ-H2AX signals in inner retinal cells appeared at 24 hr and peaked at 72 hr after ENU treatment, and the PAR signals selectively located in the photoreceptor cell nuclei appeared at 12 hr and peaked at 24 hr after ENU treatment. However, degeneration was restricted to photoreceptor cells, and no degenerative changes in inner retinal cells were seen at any time points. Retinal thickness and the photoreceptor cell ratio in the central and peripheral retina were significantly decreased, and the retinal damage ratio was significantly increased 7 days after ENU treatment. In conclusion, ENU induced retinal degeneration in adult rats that was characterized by photoreceptor cell apoptosis through PARP activity.     

Retinal degeneration associated with the feeding of dog foods to cats

Retinal degeneration was observed in cats fed commerical dog food. The retinal degenerative lesions ranged in size from small areas of focal atrophy centered in the area centralis to generalized retinal atrophy. Blindness developed only in cats with generalized retinal atrophy. Analysis of the dog food diets, both dry and canned, revealed that taurine was absent or was present in very low concentrations when compared with control cat food diets. Plasma amino acid analysis also revealed taurine deficiency.     

Multifocal retinopathy of Great Pyrenees dogs

Forty-four related Great Pyrenees dogs were examined ophthalmoscopically. Focal retinal elevations, multiple gray-tan-pink subretinal patches, and discrete areas of tapetal hyper-reflectivity were seen in 19 dogs, ranging from 13 weeks to 10 years of age. These lesions varied in size from focal spots that were barely visible with the indirect ophthalmoscope to areas that were larger than the optic disc. Complete blood cell counts, serum biochemical profiles, urinalyses, and blood pressure measurements were completed on four affected dogs and all were within normal reference ranges. Photopic and scotopic electroretinography was completed and the a-wave and b-wave amplitudes and latencies were similar for affected and age-matched nonaffected Great Pyrenees and other normal dogs. Electroretinograms that were examined twice during a 3-year period on three affected adult dogs did not reveal significant progressive deterioration of the a or b-wave parameters. Fluorescein angiography was completed on four affected dogs of ages 1 (n = 2), 5, and 6 years. These angiograms were repeated in three of these dogs 1 year later. The blood ocular barrier was intact in these dogs but there was blocked choroidal fluorescence. Postmortem examination, light microscopy, scanning and transmission electron microscopy were performed on three affected puppies and two affected adult dogs. These examinations revealed that the lesions in the puppies were limited to bilateral multiple areas of retinal pigment epithelial vacuolation, hypertrophy, and apparent separation from Bruch's membrane, and multiple serous retinal detachments. The affected adult dogs had focal retinal degeneration and retinal pigment epithelial hypertrophy, hyperplasia and pigmentation. Pedigree analysis and test mating confirm that this condition is inherited, probably as an autosomal recessive trait. This condition develops at approximately 13 weeks of age and the focal areas of retinal detachment and retinal pigment epithelial vacuolation progress to permanent and stable focal areas of retinal degeneration, and retinal pigment epithelial hypertrophy and pigmentation.     

Morphometric analysis of the retina from horses infected with the Borna disease virus

Borna disease (BD) is a fatal disorder of horses, often characterized by blindness. Although degeneration of retinal neurons has been demonstrated in a rat model, there are controversial data concerning whether a similar degeneration occurs in the retina of infected horses. To investigate whether BD may cause degeneration of photoreceptors and possibly of other neuronal cells at least at later stages of the disease, we performed a detailed quantitative morphologic study of retinal tissue from Borna-diseased horses. BD was diagnosed by detection of pathognomonic Joest-Degen inclusion bodies in the postmortem brains. Paraffin sections of paraformaldehyde-fixed retinae were used for histologic and immunohistochemical stainings. Numbers of neurons and Müller glial cells were counted, and neuron-to-Müller cell ratios were calculated. Among tissues from 9 horses with BD, we found retinae with strongly altered histologic appearance as well as retinae with only minor changes. The neuron-to-Müller cell ratio for the whole retina was significantly smaller in diseased animals (8.5 +/- 0.4; P < .01) as compared with controls (17.6 +/- 0.8). It can be concluded that BD in horses causes alterations of the retinal histology of a variable degree. The study provides new data about the pathogenesis of BD concerning the retina and demonstrates that a loss of photoreceptors may explain the observed blindness in infected horses.     

Taurine deficiency retinopathy in the cat

The literature on feline central retinal degeneration is reviewed and an experiment reported which investigates whether taurine is essential in cats fed a purified diet. The development of taurine deficiency retinopathy is described and illustrated. The histopathological, ultrastructural and ERG changes are also described. Other retinal degenerations in the cat are discussed.     

Neuron-specific enolase antibodies in patients with sudden acquired retinal degeneration syndrome

Sudden acquired retinal degeneration syndrome (SARDS) is a disease characterised by sudden and bilateral vision loss of dogs. Previous studies failed to identify the underlying cause [Mattson, A., Roberts, S.M., Isherwood, J.M.E., 1992. Clinical features suggesting hyperadrenocorticism associated with sudden acquired retinal degeneration syndrome in a dog. J. Am. Anim. Hosp. Assoc. 28, 199-202; Van der Woerdt, A., Nasisse, M.P., Davidson, M.G., 1991. Sudden acquired retinal degeneration in the dog: clinical and laboratory findings in 36 cases. Prog. Vet. Comp. Ophthamol. 1, 11-18] and earlier investigations about the occurrence of anti-retinal antibodies in SARDS patients showed inconsistent results. To provide a novel approach to those findings we designed a more detailed study. Autoantibodies of SARDS patients and normal controls were tested against the purified autoantigens S-antigen and cellular retinaldehyde binding protein (CRALBP) that play a role in human autoimmune uveitis. Next we tested the autoantibody binding pattern to whole retinal lysate. No difference in the incidence of autoantibodies could be found between SARDS patients and healthy controls while testing the well-known autoantigens S-antigen and CRALBP. Potential novel, yet unknown autoantigens were identified by a screening test using the retinal proteome as an autoantigenic source. In SARDS patients and normal controls, several retinal proteins were bound by IgG antibodies, but one band was strongly marked by SARDS patients. That band was excised, subjected to mass spectrometry (matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF/TOF)) and identified as neuron-specific enolase. Binding of the IgG autoantibodies of SARDS-affected dogs to this protein was verified using purified NSE, revealing 25% of NSE autoantibody-positive SARDS patients and 0% of negative controls. Our findings indicate that at least some dogs with SARDS have autoantibodies against NSE, although it is unclear whether these play a causative role in SARDS or whether they are the result of retinal destruction by another mechanism.     

Fluoroquinolone-induced retinal degeneration in cats

Although the exact mechanism of fluoroquinolone-induced retinal degeneration in cats remains to be elucidated, it appears from the literature that a similar retinal degeneration can be reproduced from either direct intravitreal injection of high concentrations of drug or exposure to UVA light and drug in laboratory animals. (19,25) The fluoroquinolone molecular structure is also similar structurally to other drugs that are known to directly induce retinal degeneration, including the cinchona alkaloids and halogenated hydroquinolones. Experimental evidence suggests that both the parent compound and its breakdown products via metabolism and photodegradation are active inducers of retinal degeneration. (18,25) Development of toxicoses also appears to be dependent on the maximum concentration of active drug, metabolite, or both reaching the retina over time. (18) Evaluation of the literature suggests that risk factors predisposing cats to fluoroquinolone-induced retinal degeneration may include the following: 1) large doses or plasma concentrations of drug, 2) rapid IV infusion of the antibiotic, 3) prolonged courses of treatment, and 4) age. Theoretically, other risk factors may also be involved including the following: 1) prolonged exposure to UVA light while the antibiotic is being administered, 2) drug interactions, and 3) drug or metabolite accumulation from altered metabolism or reduced elimination. To date, there are no published reports suggesting that the dose of fluoroquinolones should be reduced in geriatric cats or those with renal or hepatic failure. However, accumulation of fluoroquinolone metabolites in dogs and of the parent compound in humans with decreased renal function has been reported. (8-10) In humans with decreased renal function has been reported. (8-10) humans, fluoroquinolone doses are typically decreased in response to the degree of renal impairment. (28) In general, all fluoroquinolone antibiotics should be reserved for severe or recurrent infections, and whenever possible their use should be based on results whenever possible their use should be based on results of culture and susceptibility tests. When indicated, the fluoroquinolones, including enrofloxacin, can be used with limited risk of developing retinal degeneration in cats, provided the manufacturer's guidelines are adhered to and dose reduction is considered in geriatric cats or those with renal impairment. Dosing on renal impairment. Dosing on exact body weight using split dosing (2.5 mg/kg, PO, q 12 h) and avoidance of rapid IV infusions, and drug interactions may help to reduce the risk of retinal degeneration in some cases. Furthermore, monitoring cats for mydriasis and avoidance of UVA light while undergoing treatment may also be of benefit. Further evaluation of the pharmacokinetics of enrofloxacin and the other fluoroquinolones is required in geriatric cats or those with mild to moderate renal or liver impairment to determine whether drug accumulation, elevated peak concentrations of drug, or both may be occurring in this subset of cats. Therapeutic monitoring of drug concentrations may not always be feasible because of time and cost, but renal panels with dose or frequency reduction in response to the degree of renal impairment and the site and severity of infection may help to reduce retinal toxicosis.     

Abnormal dark-adapted ERG in cats heterozygous for a recessively inherited rod-cone degeneration

Purpose To study retinal function in cats homozygous and heterozygous for a recessively inherited rod‐cone degeneration. Methods Dark‐adapted electroretinograms (ERGs) were performed on early affected, heterozygous (ophthalmoscopically normal), and clinically normal, nonrelated cats. Responses to blue stimuli over a 3.9‐log unit range were recorded. Results Lower b‐wave amplitudes than normal were observed in heterozygotes and early affected cats. The amplitudes of the heterozygotes took an intermediate position between normal and early affected cats. Normalized amplitude/intensity data suggest a normal dynamic range in carriers. B‐wave implicit times in carriers were comparable to those of normal cats. Conclusions These results show that heterozygotes have an altered retinal function, although they are ophthalmoscopically normal. It is difficult to electrophysiologically differentiate heterozygotes from affected cats with the very early stage of retinal degeneration.     

Clinical, electrophysiological and morphological changes in a case of hereditary retinal degeneration in the Papillon dog

An autosomal recessive retinal disease with a late onset in Swedish Papillon dogs has recently been described. A 7-year-old Papillon dog showed no obvious signs of visual impairment and only minor ophthalmoscopic changes. Cone ERG b-wave amplitudes were within normal limits, while rod responses were nonrecordable or severely abnormal. Ultrastructural examination showed a generalized retinal degenerative disease, most prominent in the peripheral areas. The inferior retina was more severely affected than the superior areas. Both rods and cones showed morphological changes. The Papillon dog is another dog breed affected by progressive rod-cone degeneration, with similarities to the canine retinal disease given the gene symbol prcd .     

Electrophysiologic differentiation of homozygous and heterozygous Abyssinian-crossbred cats with late-onset hereditary retinal degeneration

OBJECTIVE: To develop a method to electrophysiologically differentiate heterozygous-carrier Abyssinian-crossbred cats from homozygous-affected Abyssinian-crossbred cats before clinical onset of inherited rod-cone retinal degeneration. ANIMALS: 14 back-crossed Abyssinian-crossbred cats of unknown genotype (homozygous or heterozygous) for inherited rod-cone retinal degeneration, 24 age-matched mixed-breed control cats, 6 age-matched heterozygous Abyssinian-crossbred cats, and 6 homozygous Abyssinian cats. PROCEDURE: Electroretinography (ERG) of heterozygous and homozygous cats revealed differences, especially for scotopic recordings. Frequent ophthalmoscopy and ERG (2 to 5 times; at intervals of 3 to 6 months) of back-crossed cats were performed. Amplitudes and implicit times were analyzed by use of a graphic representation of results. Ratios for amplitudes of the b-waves to amplitudes of the a-waves (b-wave:a-wave) were compared. RESULTS: 8 back-crossed cats had decreased a-wave amplitudes, increased b-wave implicit times, and abnormal ERG waveforms. Values for the b-wave:a-wave for the highest scotopic light intensity were significantly higher for those same 8 cats. CONCLUSIONS AND CLINICAL RELEVANCE: The 8 back-crossed Abyssinian-crossbred cats with abnormal results developed fundus changes over time consistent with disease. A graphic representation of ERG results can be used to differentiate between genotypes prior to funduscopic changes. Values for the b-wave:a-wave ratio provide confirmation. These ERG analyses may be applied clinically in the diagnosis of retinal degenerations in various species. IMPACT FOR HUMAN MEDICINE: Cats with hereditary rod-cone degeneration may be a useful model for comparative studies in relation to retinitis pigmentosa in humans. Similar evaluations of ERG results could possibly be used for humans with suspected generalized retinal degeneration.     

Susceptibility to N-methyl-N-nitrosourea-induced retinal degeneration in different rat strains

To evaluate the potential role of genetic background in the susceptibility to retinal degeneration induced by N-methyl-N-nitrosourea (MNU), female rats of the Sprague-Dawley (SD), Long-Evans (LE) and Copenhagen (CH) strains were administered 50 mg/kg MNU or saline at 7 weeks of age. Retina morphology and morphometric analysis of all rats was performed 7 days after MNU administration. Atrophy of both the peripheral and central outer retina occurred in all rat strains exposed to MNU. Decreased photoreceptor cell ratio and increased retinal damage ratio were observed. The severities of the retinal atrophy were similar among all three rat strains. In conclusion, MNU-induced photoreceptor degeneration developed consistently in all three strains regardless of the absence (SD rats) or presence (LE and CH rats) of melanin in the retina, suggesting that genetic and melanin factors did not affect photoreceptor cell death after MNU.     

Retinal degeneration in nine Swedish Jämthund dogs

The Jämthund is the fourth most common breed in Sweden with approximately 1600 pups registered each year. Although it has been known that some adult dogs go blind, so they cannot hunt, the Jämthund dog has historically not been screened for hereditary eye diseases. This report describes nine Swedish Jämthund dogs with retinal degeneration. These dogs represent all Jämthund dogs diagnosed with progressive retinal atrophy (PRA) by the Swedish Eye Panel and registered with the Swedish Kennel Club from January 1998 to September 2008. The dogs were examined with indirect opthalmoscopy and slitlamp biomicroscopy. Additionally, electroretinograms (ERGs) following ECVO guidelines were performed in two dogs (one affected and one normal) and the eyes from three affected dogs were examined by light‐microscopy postmortem. Typical findings were bilateral symmetric generalized retinal degeneration with tapetal hyper‐reflectivity, attenuation of blood vessels and pigment clumping in the nontapetal fundus. These retinal findings progressed with time in two dogs after re‐examination. Visual impairment, especially under dim light conditions, was observed in the affected dogs. ERG from one affected dog showed profoundly reduced rod responses, whereas cone responses were better preserved. Microscopic changes in the eyes from three dogs were characterized by a severe diffuse predominantly outer retinal degeneration and atrophy. Re‐sequencing of the prcd‐gene for eight of the nine investigated dogs revealed that none of the individuals carried disease allele that has been associated with prcd‐PRA in other breeds. In conclusion, ophthalmoscopic, electroretinographic, and light‐microscopic alterations observed in nine Jämthund dogs were compatible with PRA. The prcd mutation was excluded as a cause of this retinopathy.     

Optic pathway degeneration in Japanese black cattle

Degeneration of the optic pathway has been reported in various animal species including cattle. We experienced a case of bilateral optic tract degeneration characterized by severe gliosis in a Japanese black cattle without any obvious visual defects. To evaluate the significance, pathological nature and pathogenesis of the lesions, we examined the optic pathway in 60 cattle (41 Japanese black, 13 Holstein and 6 crossbreed) with or without ocular abnormalities. None of these animals had optic canal stenosis. Degenerative changes with severe gliosis in the optic pathway, which includes the optic nerve, optic chiasm and optic tract, were only observed in 8 Japanese black cattle with or without ocular abnormalities. Furthermore, strong immunoreactivity of glial fibrillary acidic protein was observed in the retinal stratum opticum and ganglion cell layer in all 5 cattle in which the optic pathway lesions could be examined. As etiological research, we also examined whether the concentrations of vitamin A and vitamin B12 or bovine viral diarrhea virus (BVDV) infection was associated with optic pathway degeneration. However, our results suggested that the observed optic pathway degeneration was probably not caused by these factors. These facts indicate the presence of optic pathway degeneration characterized by severe gliosis that has never been reported in cattle without bilateral compressive lesions in the optic pathway or bilateral severe retinal atrophy.