Chapter 108 – Retinitis Pigmentosa and Related Disorders
PAUL A. SIEVING
• Retinitis pigmentosa and the related rod-cone and cone-rod dystrophies constitute a broad set of disorders that generally result in progressive visual dysfunction because of death of the photoreceptors.
• Progressive photoreceptor dysfunction and death.
• Clinical degeneration of the outer retina.
• Intraneural retinal “bone-spicule” pigment.
• Visual field constriction.
• Night blindness.
• Loss of rod and cone electroretinography responses.
• Variable degrees of initial visual acuity loss.
• Poor correlation between acuity and extent of “tunnel vision.”
• Slowed adaptation to decreased or increased lighting.
• Retinal arteriolar narrowing.
• “Waxy nerve pallor” from reactive gliosis.
• Progressive chorioretinal atrophy.
• Highly variable geographic involvement.
• Posterior subcapsular cataracts in middle age.
• Frequently, but not always, a family history.
• Associated systemic abnormalities uncommon but important.
The term retinitis pigmentosa encompasses a set of diverse hereditary disorders that affect the photoreceptors and retinal pigment epithelium (RPE) diffusely across the entire fundus but begin with initial geographic involvement in either the periphery or the macula. These conditions typically, but not always, progress over many years to an advanced stage and result in global reduction or loss of vision. As a group, the majority of forms of retinitis pigmentosa lead to death of the rod photoreceptors, which impairs vision in dim light and causes loss of peripheral vision, that is, “tunnel vision.” However, some of the allied forms primarily cause cone photoreceptor loss and initially manifest with a reduction in central visual acuity.
Such conditions are determined genetically, with rare exception, and are inherited within families. All genetic types are represented. Not infrequently, a patient represents an isolated case in a kindred with no known affected relatives, which makes the condition difficult to differentiate from inflammatory or infectious retinal insults. The possibility of genetic diagnosis is rapidly becoming feasible, with identification of individual genes and the various defects within the genes that result in these conditions. New treatment options are becoming available and others are anticipated in the years ahead because of a new understanding of the pathophysiology of the disease.
EPIDEMIOLOGY AND PATHOGENESIS
Familial retinal degeneration with intraneural retinal pigmentation was described as early as 1855 by Donders. Although “retinitis” implies an inflammatory or infectious cause, histopathology shows no evidence of macrophage invasion or other inflammatory response in the photoreceptor layer or elsewhere in the retina. Now it is understood that the majority of cases have a genetic basis and involve photoreceptor cell death through apoptosis. No racial or ethnic predisposition exists. Men may be affected slightly more than women because of X-linked conditions.
The term retinitis pigmentosa refers to a broad category of disease that includes many different forms of primary photoreceptor abnormality—some affect rods first and cones later (termed rod-cone dystrophy) or the reverse (cone-rod dystrophy). The key features of the rod-cone type of disease are progressive night blindness and tunnel vision, symptoms that become more severe as more rods die. Because the maximal rod density occurs in the midperiphery, the appearance of midperipheral ring scotomas during progressive stages of visual field loss is not uncommon. Typically, both eyes are affected similarly.
Specific diagnosis of disease subtype results in the best targeted prognoses. In the absence of such diagnosis, however, careful electroretinography (ERG) studies frequently identify cone-rod or rod-cone dystrophy. The cone-rod type of dystrophy causes day-vision problems of reduced acuity, color-vision impairment, and photoaversion. It is more likely that at least some peripheral vision is retained in patients who have cone-rod dystrophy as opposed to those who have rod-cone dystrophy.
The incidence of primary photoreceptor degeneration is in the range 1:3000–1:5000. The carrier state for recessive retinitis pigmentosa is approximately 1:100, based on the incidence of recessive retinitis pigmentosa. These numbers are very approximate and elastic because of the complexity of the many different forms of retinitis pigmentosa now identified by gene cloning. In most cases, these diseases are thought to be simple mendelian traits that result from DNA alteration in single genes.
Typical Retinitis Pigmentosa
The key features of typical retinitis pigmentosa normally found are:
• “Bone-spicule” intraneural retinal pigment;
• Thinning and atrophy of the RPE in the mid- and far-peripheral retina;
• Relative preservation of the RPE in the macula;
• Gliotic “waxy pallor” of the optic nerve head;
• Attenuation of retinal arterioles.
The extent of bone-spicule pigmentation is highly variable—many involved retinas have some, even if very little, of this pigment. Normally, pigment clumping is not found in typical retinitis pigmentosa. The severity of the features increases with age, such as the amount of pigment, the extent of disc gliosis, and the degree of arteriolar narrowing. Major deviations from this clinical picture suggest atypical retinitis pigmentosa. In particular, several of the X-chromosome retinal dystrophies (described subsequently) deviate widely from this standard picture. The identification of typical retinitis pigmentosa is particularly important for clinical trials because this diagnosis implies the exclusion of other retinitis pigmentosa subtypes that may have unusual rates of progression.
Symptoms of typical retinitis pigmentosa include a prolonged time to adjust to dim lighting. Many retinitis pigmentosa patients are able to drive at night on well-lit streets. The worst problem for driving is at dusk or in rain or fog. Patients complain of problems in dimly lit restaurants and theaters and are symptomatic for slow adaptation when they come indoors from bright sunlight. Dark stairwells cause difficulty. By the midstage of the disease, visual field constriction results, for example, in bruised shins because coffee tables are bumped into at knee height. Patients may appear clumsy because they collide with a door frame or a friend who walks alongside because of unrecognized tunnel vision.
Typically, both eyes are affected to a comparable extent, although some degree of difference between each eye is expected normally. Highly asymmetric differences are described as “unilateral retinitis pigmentosa,” in which one eye lags in degeneration by the equivalent of many years, although both eyes invariably show involvement on careful testing. Such apparent unilateral retinitis pigmentosa cases may also result from postinfectious causes or blunt trauma to one eye.
Later manifestations of the disease include cataracts, especially of the posterior subcapsular variety, and cystoid macular edema. Both these complications may reduce the central acuity, even when the underlying disease process affects the peripheral retina only.
Molecular identification of the causative gene is possible for some cases, as for the pro-23-his rhodopsin mutation ( Fig. 108-1 ). A long-sought goal is to correlate a specific gene with a specific phenotype, but this is accomplished rarely because most retinitis pigmentosa mutations result in a similar phenotype. Some unique conditions exist, such as the mutations in
Figure 108-1 “Typical” retinitis pigmentosa changes in a 73-year-old woman who had a pro-23-his rhodopsin autosomal dominant mutation. Visible are extensive, intraneural retinal, bone-spicule pigmentation, severely constricted retinal arteries, waxy pallor of the disc, and extensive retinal pigment epithelium atrophy in the macula and midperiphery (which reveals underlying choroidal vessels). Her visual acuity was 20/50 (6/15), but she made no errors on Ishihara color testing; her fields were severely constricted to 17° tunnel vision with the Goldmann V4e target.
the RDS/peripherin gene, which may cause a peculiar maculopathy in addition to peripheral retinal degeneration. In general, however, it is not yet possible to predict the specific causative gene from the clinical presentation.
Further, even within the retinitis pigmentosa cases attributed to a particular gene, such as that which codes for rhodopsin, major variations in the clinical features and disease severity are caused by mutations at different positions within the gene. In rhodopsin retinitis pigmentosa, the pro-23-his mutation causes fairly typical retinitis pigmentosa ; the cys-187-tyr mutation follows a rapid course of degeneration ; and the thr-58-met mutation results in sectoral involvement and a slow and mild clinical course. The clinical course may vary even within a single family with a single genotype. Consequently, it is impossible to summarize retinitis pigmentosa as a single definition or clinical picture. In general, however, typical retinitis pigmentosa progresses slowly, such that a period of 1–3 years is needed to document changes.
X-Linked Recessive Retinal Dystrophies
X-linked recessive retinal dystrophies warrant separate descriptions because a positive family history directs attention to an X-chromosomal disease. Such patients are identified primarily by pedigree analysis, but, in addition, many of the dystrophies have fairly unique clinical features, which provide the clinical suspicion that results in careful determination of the family history.
X-Linked Retinitis Pigmentosa
X-linked retinitis pigmentosa (XLRP) has features of typical retinitis pigmentosa, although prominent parafoveal atrophy may be present. Affected boys shows subtle or modest RPE granularity, but frank, intraneural retinal bone-spicule pigment typically does not appear until the teenage years. Acuity is good during childhood, but by 20 years of age acuity loss and field constriction rule out the acquisition of a driver’s license, and dark-adapted thresholds are elevated by as much as 3 log units (1000-fold sensitivity loss). Night blindness is severe by the midteenage years. Rate of vision decline is rapid, and major functional loss is expected by 30 years of age; blindness by 40 years of age is common.
The clinical features of three different forms of XLRP (RP2, RP3, and RP15) overlap extensively, and all progress to severe vision disability by young adulthood. All those affected show typical characteristics of retinitis pigmentosa, although cone vision is affected to a greater degree than in many autosomal forms of retinitis pigmentosa, particularly in those who have RP15. ERG reveals a major reduction of both the rod and cone responses even in young boys and is essential to stage the disease.
Choroideremia results in a characteristic choriocapillaris loss with bare sclera and scalloped edges in the peripheral fundus. Wide pigmentary variations occur, but the pigment usually forms in clumps rather than in bone-spicule shapes. Despite only a small central island of vision, a surprising degree of local macular preservation and adequate acuity may occur well into middle age. Choroideremia may be confused with gyrate atrophy, a rare autosomal recessive condition that elevates serum ornithine levels 10- to 20-fold. Diagnosis is by clinical experience and may be confirmed by biochemical assay for Rab escort protein (REP-1) production in a blood sample. Choroideremia carriers show very coarse RPE granularity across major regions of the fundus and clinically are identified readily (see Chapter 110 ).
Congenital Stationary Night Blindness
X-linked congenital stationary night blindness (CSNB) may be confused with XLRP at first presentation in young boys because
both conditions cause complaints of difficult vision at night and show alterations in the fundus pigmentation (see Chapter 111 ). Differentiation between these two forms is critical because in CSNB vision remains “stationary.” In CSNB, night blindness is severe from birth, but visual acuity may be reduced only slightly and color discrimination is unaffected. High myopia is common. The fundus pigmentation is frequently mottled. Both ERG and visual field tests are critical in the diagnosis of CSNB. The photopic ERG provides a diagnostic clue—CSNB shows a characteristically wide cone a-wave trough. Visual fields are full for CSNB, whereas they are constricted to the Goldmann I4e target even in early XLRP and choroideremia.
Diagnosis is determined by the characteristic spoke wheel pattern of foveal and parafoveal intraretinal cysts and an X-linked family history. The ERG hallmark is a healthy rod a-wave but reduced rod b-wave with bright stimuli. Visual acuity is typically between 20/25 (6/7.5) and 20/80 (6/24) in teenagers but declines further to 20/200 (6/60) during the sixth decade because of secondary atrophy of the RPE in the macula. Despite the reduced rod b-wave, patients rarely complain of night blindness and dark-adapted visual thresholds may be nearly normal (see Chapter 112 )
Blue Cone Monochromatism
Boys and men affected by blue cone monochromatism (BCM) have normal nighttime rod vision but poor day vision as a result of the loss of both red and green cone function. Also, BCM causes small-amplitude nystagmus, reduced acuity, and photosensitivity. The acuity is between 20/80 (6/24) and 20/200 (6/60) and this may go unrecognized until school years. The fundus may show minimal RPE pigmentary mottling. Color vision is absent or at least severely limited, but differences in bluish hues are detectable. Some preschoolers affected by BCM correctly identify red-, green-, and blue-colored pencils, whereas achromats (rod monochromats) totally lack color vision. Because rod monochromacy is autosomal recessive, both sexes are affected. Both BCM patients and achromats fail the Ishihara and American Optical color plate tests and the Farnsworth D-15 and 100 Hue tests. Differentiation between achromats and BCM patients is aided by specially designed “blue arrow” color plate tests, which boys and men affected by BCM pass but achromats fail. The dark-adapted ERG b-wave amplitude is normal or slightly subnormal for BCM patients, but rod psychophysical threshold sensitivity remains normal, which differentiates BCM from degenerative rod disease cases. The light-adapted cone ERG single flash and flicker responses are reduced by >80–95%. Only 1% of cones are blue cones, and no blue cones are found within the human fovea, which accounts for the poor acuity in BCM patients. All other aspects of vision remain stable in BCM, and acuity may even improve slightly and reach 20/70 (6/21) by the age of 20 years, by which time the nystagmus is barely detectable. Some older BCM men have progressive macular atrophy. Only boys and men are affected by BCM; female carriers show no changes. Red and green opsins occur on the X chromosome, and BCM is an X-linked recessive trait. An X-linked family history helps differentiate BCM from autosomal recessive, early age, progressive cone dystrophy.
Congenital Red-Green Color Deficiency
Total red (protanopia) or green (deuteranopia) color blindness affects 2–3% of men. Partial forms are termed anomalous color perception (e.g., protanomaly). Tritanopia (total blue blindness) is exceedingly rare. For all forms together, 4–7% of men manifest some type of congenital color deficiency. Given this frequency, some men who are affected by other retinal degenerations are also “color blind,” and the congenital condition must be differentiated from the acquired disease. Progressive cone dystrophy also impairs color discrimination but is differentiated by abnormal visual acuity and/or peripheral fields, both of which are usually normal in the congenital color deficiencies. Female carriers manifest no clinical signs.
Ocular albinism occurs in several forms and follows all the inheritance patterns, including an X-linked recessive type. All forms are evident from birth and cause moderate-amplitude nystagmus, which the parents notice quite early. The fundus is hypopigmented and the iris may transilluminate. Acuity is between 20/70 (6/21) and 20/200 (6/60) but is difficult to test precisely during infancy. Color vision remains normal and nyctalopia does not occur. The foveal reflex may be muted or hypoplastic. The electroretinogram is normal or even supranormal because of enhanced intraocular light reflection. Cutaneous involvement accompanies systemic forms. Diagnosis is made by clinical examination. Vision remains stable. Confusion arises in the differentiation of ocular albinism from the “blond fundus” of patients who have pale skin and hair tones but normal acuity.
Female Carriers of X-Chromosomal Retinal Dystrophies
Female carriers of X-chromosomal retinal dystrophies may manifest retinal pigmentary changes and have functional vision impairment that present special difficulty in diagnosis. Recognition of the carrier state is important to establish the correct inheritance pattern for family genetic counseling. Some carriers have a severe vision abnormality, which may lead to a misdiagnosis of autosomal dominant disease. In female carriers, one of the two X chromosomes has a mutant gene. As a result of random X-chromosome inactivation, only one gene is active in each cell (the Lyon hypothesis). Because the mutant gene is retained in some retinal cells during early development, clusters of neighboring cells have the disease, and patches of clinical disease occur that mimic a mild form of the fully expressed male condition in choroideremia, XLRP, and X-linked ocular albinism. Carriers of juvenile retinoschisis, blue cone monochromacy, CSNB, and color-vision dichromacy show no fundus changes and experience no functional vision abnormality. Carriers of autosomal recessive disease rarely show retinal changes or have visual symptoms.
Female carriers of XLRP ( Fig. 108-2 ) show one or more small or large retinal patches of typical, intraneural retinal,
Figure 108-2 X-linked female retinitis pigmentosa carrier affected by lacunae of disease. Atrophy of retinal pigment epithelium and intraneural retinal bone-spicule pigment result. Acuity is 20/200 (6/60) because of macular atrophy.
Figure 108-3 Macular and parafoveal gross pigmentary retinal pigment epithelium mottling in a female choroideremia carrier. Despite such mottling, female choroideremia carriers are asymptomatic.
Figure 108-4 Punctate granularity and attenuation of the normally uniform pigmentation of the retinal pigment epithelium in an X-linked female carrier of ocular albinism.
bone-spicule pigmentation and atrophy of the underlying RPE and choriocapillaris in more than 50% of cases. Many carriers have myopic astigmatism at an oblique axis. Although vision is involved minimally in the majority, some are functionally blind by late middle age or older. Changes progress with time but generally are much slower than those in XLRP-affected men. An electroretinogram is very helpful, as amplitudes of one or more ERG components are reduced in 80–95% of XLRP carriers. Further, the ERG amplitudes generally correlate with the expected severity of overall vision loss in later years.
Choroideremia carrier females (see Fig. 108-3 ) show widespread retinal pigmentary disturbance of the RPE in the periphery and into the macula in 90% of cases. However, very few carriers have any visual symptoms beyond mild photoaversion in later age, and visual acuity remains normal. Electroretinograms are affected far less frequently than those of XLRP carriers; fundus examination is the most sensitive means of detection. Most choroideremia carriers are emmetropic or hyperopic, in contrast to the myopia typical of XLRP carriers. Progression has been observed in some choroideremia carriers.
X-linked ocular albinism female carriers ( Fig. 108-4 ) show punctate RPE pigmentation across the entire fundus and RPE thinning in the periphery (a mild version of changes found in the affected men),
Figure 108-5 Visual function test. For normal subjects, dark-adapted rod a- and b-waves result from bright, white flashes and, primarily, b-waves from dimmer blue flashes; cone responses are elicited by a single, light-adapted, white flash (a- and b-waves) and by 30?Hz flickers. For rod-cone patients, rod responses are reduced proportionally more than cone responses. For cone-rod patients, major losses in light-adapted and 30?Hz responses occur, with relative preservation particularly of the rod b-wave, to dim, dark-adapted, blue flash. Goldmann visual field tests using the small I4e target show major tunnel vision in rod-cone patients.
which may mimic the “salt-and-pepper” appearance of congenital rubella retinopathy. Visual acuity is not affected, and ERG is normal. Symptoms do not extend beyond mild photoaversion to bright light. Carriers of non–X-linked albinism rarely show fundus changes.
Rod-Cone and Cone-Rod Disease: When Subtyping Fails
Accurate subtyping of the disease provides the best information about prognosis, but if subtyping fails the patient is assessed for rod-cone disease or cone-rod disease, two conditions that carry quite different prognoses. Rod-cone degeneration is the more severe–the long-term prognosis is loss of most or all vision by later years. Cone-rod degeneration affects central vision quite early and peripheral vision only later. Because the human retina has 120 million rods and only 6 million cones, it may survive well without cones, as exemplified by achromatopsia. The prognosis in cone-rod dystrophy is good for at least some future vision, even though central vision is jeopardized. Retinal function studies, particularly ERG, are required to differentiate rod-cone from cone-rod degenerations.
Rod-cone dystrophy manifests clinically with typical retinitis pigmentosa and affects the rod photoreceptors earlier and more severely than it does the cones. Severe cone involvement occurs in the end stage of the disease, when total vision loss ensues. End-stage rod-cone disease results in loss of both peripheral and
Figure 108-6 The bull’s-eye maculopathy in this 5-year-old male who has a cone-rod dystrophy is not found in all cases of this entity.
central vision. Many patients have only barely functional central vision by late middle age and lose all vision later. However, the changes occur slowly, and the young patient may be reassured that some vision is most likely to persist for many years or decades, even though they must accept that the long-term prognosis is grim. Photoaversion to bright light occurs in end-stage rod-cone dystrophy, when the diseased cones saturate in bright light.
The functional profile of early-stage rod-cone dystrophy is shown in Figure 108-5 . The rod ERG amplitude loss is worse than that for cones (light-adapted and 30?Hz flicker). Goldmann visual fields are still relatively large with the large V4e target but considerably constricted to the small I4e target—such disparity between the ratio of field to V4e versus I4e is typical for rod-cone dystrophy. Visual acuity is initially affected minimally and typically remains near 20/20 (6/6) for many years despite progressive field loss to severe tunnel vision. Careful history taking reveals some night blindness, even in early-stage rod-cone disease. Tests of dark-adapted thresholds show 1–3 log units of rod sensitivity loss, which equates with 10–1000 times more light to see anything at night. The disease does not act uniformly across the retina, and rod threshold should be tested at multiple points of vision to give a sensitivity profile across the central and peripheral vision.
As visual fields constrict further, even to the large V4e target, dark-adapted thresholds become worse because only cones remain to mediate vision, even in dim light. At this stage the patient suffers from severe night blindness. With time, the cone ERG responses deteriorate further until eventually both the rod and cone responses are reduced profoundly, and all ERG responses are termed “nonrecordable.” By this time, visual acuity is typically less than 20/40 (6/12), color discrimination is impaired, and fields are greatly constricted.
Cone-rod patients complain of poor acuity, reduced color vision, and photoaversion to bright sunlight. Fundus changes are quite variable, and not all patients show “bull’s-eye” maculopathy ( Fig. 108-6 ). Particularly in the early stages, the fundus may appear benign, with minimal diffuse retinopathy and normal vessels and disc. Visual fields initially remain full with the Goldmann V4e target but may be constricted slightly with the I4e target. Electroretinograms show that cone amplitudes (30?Hz flicker and photopic single flashes) are reduced proportionally more than the dark-adapted rod b-wave (see Fig. 108-5 ). Although the rod b-wave amplitudes may be subnormal technically, rod dark-adapted threshold sensitivity remains nearly normal when tested using a Goldmann-Weekers dark adaptometer after 45 minutes in the dark. Such a combination of ERG and psychophysical rod threshold tests is the best way to diagnose cone-rod dystrophy. Patients who have the least rod involvement carry the best prognosis for intermediate-term vision. Infrequently, a cone-rod patient has an essentially normal rod electroretinogram but a very reduced cone electroretinogram and is considered to have “cone dystrophy.” Such patients have a good prognosis for vision into later age. Some cone-rod patients are first symptomatic after the age of 50 years or more and may progress quickly to considerable vision loss. Such patients are a diagnostic challenge because of the later age of presentation; initially, ERG results may be marginally normal, but the results of repeat ERGs after 6–12 months progress to subnormal.
DIAGNOSIS AND ANCILLARY TESTING
The preceding brief descriptions indicate that, for diagnosis, visual function tests are an important adjunct to retinal examination. Tests may also help identify correctly the clinical subtype of the disease. Accurate subtyping provides the basis for counseling about expectations for school and career choices and for the provision of genetic information to the extended family.
Before any tests are carried out, listen carefully to what the patient has to say. Complaints of “night blindness” may indicate a total lack of vision in very dim light or a diminished acuity as ambient light dims. Also, night blindness from rod photoreceptor disease:
• Prolongs the time required to adjust to changes in light, as when dark theaters or dim restaurants are entered or even when moving from sunlight into room light;
• Causes problems at dusk, when the normal switch from daytime to nighttime vision occurs; and
• Results in visual-field constriction in dimmer light.
Tunnel vision may be suggested by recent automobile accidents or clumsiness in the narrow spaces of elevators or doorways. Any family history of slowly progressive unexplained vision loss must be sought. The retina and macula must be examined carefully for RPE granularity that precedes intraneural retinal pigment accumulation or outright atrophy. When the history and clinical features have been assessed together, and if a retinal or macular dystrophy is suspected, formal retinal function tests must be considered, which include ERG, visual field tests, and dark adaptometry.
Retinal function tests have value far beyond simple diagnosis because they give insight into the nature of the disease, inform about the severity or stage, and provide information for genetic counseling. Results of the examination must be communicated to the patient because subtyping and staging the disease provide information about expectations for the course of future vision and the genetic implications for the family.
Results of functional tests must be integrated with knowledge of the clinical state. Although flagrant and advanced retinal degeneration may be diagnosed without extensive tests, tests serve to subtype the disease and establish the severity. Disease progression is monitored by yearly visual field measurements and by repeated ERG every few years. Future clinical trials to treat photoreceptor degenerations are likely, and for this, yearly ERG will be important to establish a baseline rate of progression.
The electroretinogram is quite sensitive to even mild photoreceptor impairment. Rod b-wave amplitudes are reduced in the earliest stages of disease, when the retina may appear clinically normal and vision complaints are minimal. For the diagnosis of genetically at-risk younger patients who have otherwise minimal retinal pigmentation, ERG is essential. Also, ERG helps in the diagnosis of patients of any age who have unclear visual symptoms. Finally, ERG helps in the diagnosis of the disease in relatives
when a retinitis pigmentosa pedigree is established. Good clinical sense and careful ophthalmoscopic examination prevail in most cases, but definitive ERG normal results may allay parents’ fears that their child is affected.
Analysis and Interpretation
Initially, ERG is used to test the rod system after 45 or more minutes in the dark (the “scotopic ERG”). The cone system is tested using single, bright flashes superimposed on a bright background that suppresses activity of rods (the “photopic ERG”) and also using a 30?Hz flicker to which cones respond but rods do not. Many complex and analytic schemes of ERG analysis are employed in special circumstances, but the most useful initial measures of ERG abnormality are :
• Scotopic (dark-adapted condition, rod driven) and photopic (light-adapted, cone driven) b-wave amplitudes—these provide the first index of disease severity and help differentiate rod-cone from cone-rod disease;
• Scotopic b-waves reduced by 50% or more—this indicates progressive disease rather than a variant of “stationary” disease;
• Early cone system disease—this frequently reduces the amplitudes of 30?Hz flicker before photopic b-wave responses to single flashes;
• Delayed flicker implicit time (from flash to response peak)—this is a highly sensitive measure of abnormality, and implicit times may be prolonged even with normal flicker amplitude; implicit times are very robust and relatively immune to artifacts caused when the patient squeezes on the ERG contact lens electrode (which reduces flicker amplitude but does not change timing);
• Photopic oscillatory potentials (high-frequency wavelets of small amplitude that originate in the proximal retina)—these are generally reduced earlier or by more than the photopic, single-flash b-wave, and oscillatory potentials may be reduced in retinal vascular diseases;
• Relative preservation of the scotopic a-wave amplitude (from rod photoreceptors) but reduced scotopic b-wave (from signaling by second-order bipolar cells)—this is highly suggestive of CSNB or X-linked juvenile retinoschisis (the ERG change indicates faulty synaptic signaling from rods to bipolar cells, deficient bipolar responsivity, or Müller cell disease);
• Broad and flat bottom trough to the photopic a-wave—this is highly suggestive of CSNB;
• The full-field (termed Ganzfeld) ERG—this reflects global retinal activity and is insensitive to macular scars; thus, it does not correlate with visual acuity determined solely by foveal function.
Staging of the disease is based on the current visual acuity, dark-adapted rod thresholds, peripheral fields, color discrimination, and rod and cone ERG status. Normally, reduced rod ERG amplitudes correspond to impaired, dark-adapted rod thresholds. If the rod b-wave is reduced greatly but the dark-adapted thresholds are elevated only slightly, it suggests that only a few functioning peripheral rods remain and that vision loss will progress rapidly when these few rods are lost. Rates of visual acuity change are less predictable than changes in peripheral field constriction.
Full-field ERG and visual acuity are quite different measures of vision. Full-field ERG is dominated by a large expanse of the peripheral retina rather than by the macula, whereas visual acuity is determined exclusively by foveal cones, which are only 1% of the total number of cones.
The peripheral retina outside the macular arcade vessels contains 60% of the rods and 60% of the cones. The macula contains about 40% of the total cone population. Loss of peripheral cones as retinitis pigmentosa progresses reduces the full-field cone electroretinogram even though acuity and color vision remain intact in the macula. Many people who have retinitis pigmentosa retain excellent visual acuity despite greatly reduced cone ERG responses that result from tunnel vision caused by extensive peripheral retinal pathology.
Monitoring Disease Progression
After some period of time, the ERG is repeated for the following reasons:
• Confirmation of the diagnosis;
• Determination of the rate of progression;
• Monitoring of the effects of therapy, such as vitamin A administration;
• Provision of objective information about progression to help the patient cope with a disease that causes vision loss.
For adults, ERG may be repeated 1–2 years after the first test just to confirm the findings, or serial yearly tests may be used to estimate progression. More frequent ERGs are rarely warranted for adults. In younger children, however, retinal dysfunction may progress more rapidly, and yearly repeated ERGs are warranted to guide school decisions and anticipate future vision needs. Exceptionally rapid cone ERG loss in a child may raise suspicion of atypical forms of retinitis pigmentosa, which include neuronal ceroid lipofuscinosis and indicate the need for a neurological evaluation, particularly if seizure activity is reported.
Visual Field Testing
Goldmann perimetry is preferable for retinitis pigmentosa field testing out to 90° in the far temporal periphery, which is involved first in the rod-cone type of disease. Even moderate stages of rod disease typically show extensive peripheral field loss to the small, I4e Goldmann target, whereas cone-rod disease leaves peripheral fields more intact, the I4e target. Retinitis pigmentosa subjects may respond poorly to automated perimeters—careful studies using a Goldmann perimeter are more successful for obtaining reproducible fields. Macular dysfunction may be tracked excellently using the Humphrey Visual Field Analyzer 24-2 or 10-2 programs. Many retinitis pigmentosa patients are unaware of lost peripheral vision or a ring scotoma even after several suggestive events (i.e., recent automobile accidents). The patient will appreciate the physician who simply takes the time to explain visual field test results and may immediately recall instances of previous problems with everyday activities.
Night-vision symptoms occur early in the course of retinitis pigmentosa disease and must be evaluated using dark-adaptation studies. The most commonly used instrument is the Goldmann-Weekers dark adaptometer. The patient is placed in darkness and asked to detect the dimmest possible light that is made progressively dimmer as time proceeds. Final absolute threshold sensitivity is normally reached after 30–40 minutes in the dark. An alternative test strategy is to determine only the final thresholds after 45–60 minutes in the dark. Thresholds are tested in several different retinal locations to sample the distribution of disease. Some patients who complain of difficulty seeing at night are found to have normal dark-adapted thresholds. Such patients may have undercorrected myopia, and the complaint is really of blurred vision in dimmer light. Other patients may have maculopathy and notice worse acuity in dimmer light, even though normal rod, absolute dark sensitivity is maintained.
Color Vision Tests
In degenerative retinopathy, color testing is a useful adjunct to visual acuity tests because it provides additional information about the condition of the macular cones. Retinitis pigmentosa patients
rarely volunteer problems with color vision because nearly all can differentiate readily the major colors of red, green, and blue. However, in rod-cone dystrophies, tritanopic (“blue”) color discrimination loss on the Farnsworth D-15 panel is a sensitive index of early foveal cone involvement and may presage acuity loss within the next few years. In cone dystrophies, loss of color discrimination normally parallels visual acuity loss. The D-15 test, which consists of 15 color chips that must be arranged in color sequence, is simple, rapid, does not tire the patient, and is easy to score. More than two minor neighbor errors in the D-15 test indicates pathology. The Farnsworth 100-Hue test is more elaborate but seems to be no more sensitive for detecting maculopathy. The Ishihara and American Optical color plates were designed specifically to detect congenital red-green abnormal individuals and are less useful than the D-15 test for the evaluation of early macular dysfunction that results from a retinal dystrophy.
Fluorescein angiography rarely provides novel information in diffuse retinopathy beyond that found by careful retinal examination. It is most useful for hereditary maculopathies, such as Stargardt’s disease and Best’s disease, or in suspected toxic maculopathy from hydroxychloroquine, chloroquine, or psychotropic agents. A fluorescein angiogram should be obtained for patients who have diminished acuity without any clinically apparent maculopathy or when macular edema is suspected. Cone dystrophies frequently show subtle, foveal RPE window defects early in the course. A fluorescein angiogram provides objective verification of macular pathology in children who might otherwise be considered to suffer from psychogenic vision problems.
Electro-oculography (EOG) is abnormal whenever the ERG is abnormal and thus provides useful information only when the ERG is normal. Therefore, EOG is not performed automatically with every ERG examination. One of the very few current uses for EOG is to track the genetic pattern in Best’s vitelliform macular dystrophy, in which the expressivity is highly variable.
Visual-Evoked Cortical Potential
Visual-evoked cortical potential (VECP) monitors visual signals that reach the cortex and is dominated heavily by macular function, with a far smaller contribution from the peripheral retina. Any disturbance of retinal function, altered optic nerve conduction, or visual cortex processing alters the VECP. Retinal, and particularly macular, dystrophies affect the VECP, but these conditions are nearly always identified and followed better by other visual function tests.
Thioridazine retinal toxicity causes widespread retinal pigmentary degeneration and affects visual acuity; the visual symptoms overlap those of retinitis pigmentosa. Toxicity generally occurs within months of the initiation of thioridazine use, although retinal degeneration is also reported to occur late after use has ceased  (see Chapter 143 ).
Chloroquine causes bull’s-eye maculopathy ( Fig. 108-7 ); it binds to melanin in the RPE and causes cytotoxicity, which begins preferentially around the fovea. Visual acuity initially remains excellent, despite parafoveal metamorphopsia and difficulty in reading because of the paracentral scotoma. Because it is far less retinotoxic, hydroxychloroquine has supplanted chloroquine except for those who travel in malarial areas. Only a few cases of hydroxychloroquine toxicity have been reported, but these cause vision pathology similar to that seen in bull’s-eye
Figure 108-7 Chloroquine bull’s-eye parafoveal atrophy.
Figure 108-8 Rubella maculopathy in a young individual who has 20/60 (6/18) visual acuity.
maculopathy. The initial change is parafoveal and is picked up by macular field tests using a Humphrey Visual Field Analyzer Program 10-2, with careful attention paid to sensitivity threshold changes. Field sensitivity changes may even precede changes on fluorescein angiography. Careful examination of the parafoveal RPE using a 90D lens shows RPE disturbance with a sensitivity comparable to that of fluorescein angiography.
Congenital rubella infection causes coarse pigmentary spots and tiny clumps on the retina, termed salt-and-pepper retinopathy. However, visual fields remain full, and ERG remains essentially normal. The complaint at presentation may involve decreased visual acuity because of maculopathy from a yellowish, fibrotic macular scar ( Figs. 108-8 and 108-9 ). Peripheral vision remains stable, although visual acuity may worsen through young adulthood, and the patient may be predisposed to presenile macular degeneration.
Syphilitic retinitis may result in pigmentary retinopathy and progressive vision impairment that mimics retinitis pigmentosa. It may also cause asymmetric disease that mimics unilateral retinitis pigmentosa.  Screening serology tests should be carried out on patients who have retinitis pigmentosa but whose family history is negative.
Cancer-associated retinopathy syndrome results in acute onset of rapid and progressive bilateral vision loss without much initial
Figure 108-9 Fluorescein angiogram of patient in Fig. 108-8 who has rubella retinopathy. Marked granularity of the retinal pigment epithelium can be seen.
pigmentary retinopathy. Small-cell lung cancer is the most common cause, although many other cancers are also implicated, including endometrial, breast, and prostatic cancer. The mechanism involves production of autoantigens that enter the retina and cause apoptosis. The apparent target is the photoreceptor molecule recoverin, which is integral to the phototransduction cascade. The clinical course is rapid and inexorably progresses toward complete vision loss. Commercial tests for cancer-associated retinopathy antibodies are available.
Cutaneous Melanoma-Associated Retinopathy
Melanoma-associated retinopathy is a paraneoplastic syndrome of acute onset that causes bilateral night blindness without visible retinopathy. Antibodies generated against the cutaneous melanoma cross the blood-retinal barrier and target the retinal bipolar cells, which impairs function but apparently does not cause cell death. Visual symptoms include photopsias and shimmering patches of color. The scotopic ERG is reduced preferentially. Although the retinal course is usually otherwise benign with minimal progression, severe intraocular inflammation has been reported.
Retinitis pigmentosa is associated with many systemic conditions, of which the following warrant particular attention either because of their incidence (e.g., Usher’s syndrome) or because the diagnosis, which may be recognized by the ophthalmologist first, has major medical implications.
The hearing loss in Usher’s syndrome ranges from congenital and total to middle-age onset and partial. Usher’s syndrome is nearly always autosomal recessive, and typically no affected relatives are identified in previous generations. If the spouse has normal hearing, the risk of any affected children is small although never absolutely zero. The prevalence of Usher’s syndrome within the deaf population is estimated to be as high as 3–6%.
Type I Usher’s syndrome results in profound congenital deafness—hearing aids are ineffective. Vestibular dysfunction causes affected infants to walk late, at 18–20 months, because of balance problems; late walking enables the clinical diagnosis of siblings without any need for extensive tests. The vestibular dysfunction again causes problems during the teenage years when visual field loss combines with deafness to make the patient more clumsy at routine school sports that involve running or jumping. Vision impairment is severe by late teenage years, and functional blindness typically occurs before 40 years of age.
Type II Usher’s syndrome results in hearing impairment that may be corrected partially using hearing aids. Vision loss is less rapid than that in the type I disease but at any given age may be more severe than that in most retinitis pigmentosa patients. Vestibular function is normal, and walking begins at a normal age.
Type III Usher’s syndrome results in progressive hearing loss that is first apparent in middle age. Progression of retinitis pigmentosa is no faster than that in patients who have retinitis pigmentosa and whose hearing is not impaired.
Not all retinitis pigmentosa patients affected by impaired hearing have Usher’s syndrome. Possibly as many as 5–10% of retinitis pigmentosa patients also experience mild, progressive, high-tone hearing loss that begins in middle age, of which only some cases may represent type III Usher’s syndrome. Consequently, it cannot be presumed categorically that the inheritance pattern is autosomal recessive when counseling retinitis pigmentosa patients who have hearing loss. Familial hearing loss is not an uncommon genetic trait and may be autosomal recessive or dominant. Occasionally, families have autosomal dominant hearing loss and a sporadic case of retinitis pigmentosa in the same family that is merely coincidental. Any patient with retinal degeneration for whom there is a suspicion of even slight hearing loss must receive audiologic evaluation and treatment to minimize the effect of major sensory problems that arise from combined hearing and vision deficits.
Bardet-Biedl syndrome is an autosomal recessive entity that involves pigmentary retinopathy with polydactyly, renal dysfunction, and short stature in association with truncal obesity. The fingers are thick but taper to fine tips. Because extra digits are often removed surgically during infancy, finger surgery must be asked about specifically. Intelligence is frequently subnormal. Hypogenitalism occurs in more than half of patients. The peripheral fundus pigmentation may appear typical for retinitis pigmentosa or have only very coarse RPE granularity, and progressive parafoveal macular atrophy impairs visual acuity by the teenage years ( Fig. 108-10 ). Both visual acuity loss and field constriction are very severe by middle age. Renal disease frequently causes premature death by middle age, and patients may suffer renal failure that requires transplantation even during the teenage years. A full renal evaluation is suggested when the ophthalmic diagnosis is made. The similar Laurence-Moon syndrome, in addition, features paraplegia.
Kearns-Sayer syndrome is a mitochondrial myopathy that results in external ophthalmoplegia, lid ptosis, cardiac conduction block, and mild retinitis pigmentosa. Such patients present as young adults with modest night blindness and modest peripheral-field constriction. The fundus appearance is atypical for retinitis pigmentosa, as only very coarse RPE granularity is found with little intraneural retinal pigment accumulation. Visual field constriction is modest, but visual acuity may be impaired. The progression of visual symptoms is rather slow. Typically, ERG rod and cone amplitudes are reduced only modestly. Referral for cardiac evaluation is required because the conduction abnormalities may result in life-threatening arrhythmias.
Neuronal Ceroid Lipofuscinosis
Neuronal ceroid lipofuscinosis is a devastating condition that involves progressive neurological failure, mental deterioration, seizures, and visual acuity loss from cone degeneration that causes bull’s-eye maculopathy. Several subtypes affect different age groups, from infants to teenagers, and death occurs by the
Figure 108-10 Bardet-Biedl syndrome with extensive peripheral retinal pigment epithelium thinning and parafoveal retinal pigment epithelium atrophy.
second decade. The ophthalmologist may be among the first to suspect the diagnosis in the context of maculopathy, recent new seizure activity, and subtle mental deterioration judged by progressive difficulties at school (which initially may be blamed on the subnormal vision). Although the presentation is as a maculopathy, ERG shows widespread cone disease. Skin biopsy shows characteristic “fingerprint inclusions” under electron microscopy. If a neurologist is already involved, the diagnosis of a widespread cone dystrophy must be communicated because this may clarify the differential diagnosis. The vision loss is quite rapid and progresses to blindness within only a few years. Alternative terms include Batten-Mayou and Vogt-Spielmeyer disease.
Gene identification has progressed rapidly since the late 1980s, and currently more than 20 different altered genes have been identified that result in retinitis pigmentosa and allied diseases. Rhodopsin was the first major retinitis pigmentosa gene to be identified. As with rhodopsin, the majority of the retinitis pigmentosa genes identified thus far involve components of the phototransduction cascade within the rod photoreceptor, which include transducin, phosphodiesterase (a- and b-PDE), arrestin, recoverin, and the G protein–coupled Na+ /K+ light-activated channel on the rod membrane. A second set of genes code for structural proteins in rod cells and include RDS/peripherin  and ROM1. Developmental genes are also implicated, such as the homeobox gene CRX in the development of a cone-rod degeneration.
The molecular mechanisms by which these genetic mutations eventually cause rod-cell death are unclear, although ample evidence indicates that apoptosis is involved in the final pathway of cell death. Currently, three hypotheses are under investigation:
• Abnormal trafficking because of defective protein folding;
• Cellular metabolic disturbances; and
• “Constitutive activity” of transduction because of rhodopsin mutations.
That the cone photoreceptors ultimately die from a disease that begins with rod-cell disease remains a puzzle. One hypothesis invokes common elements of the RPE that are involved intimately in the diurnal cycle of phagocytosis of the outer-segment discs shed daily by both rods and cones. In the case of rhodopsin-mutation retinitis pigmentosa, rhodopsin is the major protein in the rod outer segments and the diurnal process of phagocytosis of the shed rod-disc membrane by the RPE may eventually result in secondary RPE pathology. With time, the RPE cannot properly service the cone photoreceptors, which subsequently die as “innocent bystanders.”
Figure 108-11 Histology of the parafoveal retina of a 73-year-old woman who had a pro-23-his rhodopsin mutation (same patient and eye as in Fig. 108-1 ). Only the macula retained any photoreceptors. The eyes were fixed about 1 hour postmortem. INL, Inner nuclear layer; ONL, outer nuclear layer; IS, inner segments; OS, outer segments; RPE, retinal pigment epithelium.)
Histological examination of the retina from a 73-year-old woman who had autosomal dominant retinitis pigmentosa from a pro-23-his rhodopsin mutation showed major loss of the photoreceptors ( Fig. 108-11 ). She had 20/50 (6/15) visual acuity several months before death, and her fields were only 17°. Her fundus had typical retinitis pigmentosa changes of heavy bone-spicule pigmentation across the entire 360° periphery, and the underlying RPE was atrophic. Tissue from the parafoveal region of the left eye in the region of relative preserved retina showed:
• Photoreceptor outer segments shortened greatly, such that they are nearly absent, and the inner segments shortened;
• Number of photoreceptor nuclei (outer nuclear layer) decreased greatly, the majority of those left being multinucleated cone nuclei—nearly no rod photoreceptors remain in this end-stage retinitis pigmentosa retina;
• RPE swollen grossly by intraretinal debris, with loss of melanosomes and dispersion of pigment granules.
Current treatments for retinitis pigmentosa are not highly effective, although new research developments suggest that it may be possible to slow disease progression, possibly to the extent that vision may persist for a lifetime.
A long-term study of oral vitamin A palmitate supplementation (15,000?IU daily) administered to 600 patients who had typical retinitis pigmentosa showed a modest but positive slowing of vision loss. For all cases in aggregate, vision loss slowed to a decline of 8.3% per year compared with 10% per year in controls. Such a slight slowing of degeneration is rarely noticeable to an
individual patient over a short period, but it may provide additional years of vision when spread over a lifetime. The rescue mechanism is unknown, but vitamin A is essential for the formation of light-sensitive rhodopsin. Opsin alone, in the absence of vitamin A, may exhibit a small degree of toxicity and possibly cause photoreceptor demise over a lifetime. In the same study, the administration of vitamin E (400?IU daily) without vitamin A speeded the degeneration by a small but statistically significant amount. However, when combined with vitamin A, vitamin E did not substantially alter the slowing of progression afforded by vitamin A palmitate alone, and thus the modest amount of vitamin E in multivitamin formulations may not be detrimental when taken along with vitamin A. If this treatment is suggested, the patient is advised to use vitamin A for the long term and to expect no immediate benefit in vision. Yearly checks of serum liver enzymes and/or vitamin A levels are advisable while vitamin A is taken in high dosage, and discontinuation is necessary if pregnancy is expected.
Acetazolamide for Cystoid Macular Edema
In some patients, retinitis pigmentosa results in cystoid macular edema (CME), possibly because the efficiency with which fluid is pumped across the RPE is compromised or because of slow retinal vascular leakage. Some studies show that treatment with acetazolamide may be of benefit, with an initial dose of 250?mg daily, increased to 500?mg daily if no effect is apparent. A trial of several weeks is warranted, with successful outcome judged by improved visual acuity on careful repeated measurements or by decreased CME on fluorescein angiography. If decreased CME is observed by fluorescein angiography after several weeks of use, the continuation of acetazolamide may be considered even if visual acuity has not improved, provided the patient is able to tolerate the drug.
Docosahexaenoic acid (DHA), a 22:6 fatty acid, is the major lipid component of rod photoreceptor membranes and is important for the maintenance of membrane fluidity required for rods to function. Abnormal cholesterol and serum lipid levels have been reported in some retinitis pigmentosa patients, and DHA levels are particularly and somewhat consistently low in XLRP patients. On the basis that insufficient DHA may affect photoreceptor survival, trials are under way to determine whether DHA dietary supplementation slows progression in retinal degeneration. Because neither the benefit nor the possible detriment of any substance can be known without formal and rigorous scientific evaluation of efficacy, it is not wise to advise patients to use DHA until the outcome of these ongoing studies is known.
A report in 1990 showed that intraocular injection of basic fibroblast growth factor effectively slowed photoreceptor degeneration in the RCS rat model of retinal degeneration. Although no practical current therapy employs growth factors, the results demonstrate that effective therapies may be developed to slow the rate of degeneration radically in these diseases.
Disproved “Treatment” Strategies
Many treatments for retinitis pigmentosa have been tried and discarded now that they have been proved ineffective. Such discredited attempts at therapy are:
• Placental implants along the sclera;
• ENCAD (daily periocular and intramuscular injections of mushroom RNA extract) treatment in Russia;
• Vasodilator drugs; and
• Cuban “treatment,” which includes vasodilators, hyperbaric oxygen, and surgical insertion of periorbital fat into the subchoroidal space.
COURSE AND OUTCOMES
Projections about future vision are always difficult in degenerative disease, particularly because the retinitis pigmentosa subtypes do not have a single clinical course. XLRP typically affects visual acuity by young adulthood, and visual acuity of some XLRP female carriers also becomes severely impaired. Thus, a simple summary of vision loss in the various forms of retinal degeneration is not possible, particularly for visual acuity. In all cases, functional tests using ERG and visual thresholds best establish the current stage of retinal cell function in aggregate and thus provide an initial basis for any prognostic statement. Prognostic statements depend upon careful disease subtyping. When subtyping is elusive, analysis of whether the patient has rod-cone disease or cone-rod disease provides vision estimates that may be used for general prognosis.
1. Donders FC. Beiträge zur pathologischen Anatomie des Auges. Graefes Arch Clin Exp Ophthalmol. 1855;1:106–18.
2. Nettleship E. On retinitis pigmentosa and allied diseases. R Lond Ophthalmol Hosp Rep. 1907;17:1–56.
3. Boughman JA, Conneally PM, Nance WE. Population genetic studies of retinitis pigmentosa. Am J Hum Genet. 1980;32:223–35.
4. Massof RW, Finkelstein D, Starr SJ. Bilateral symmetry of vision disorders in typical retinitis pigmentosa. Br J Ophthalmol. 1979;63:90–6.
5. Henkes HE. Does unilateral retinitis pigmentosa really exist? An ERG and EOG study of the fellow eye. In: Burian HM, Jacobson JH, eds. Clinical electroretinography. Proceedings 3rd ISCERG Symposium. London: Pergamon Press; 1966:327–50.
6. Cogan DG. Pseudoretinitis pigmentosa. Arch Ophthalmol. 1969;81:45–53.
7. Weleber RG, Carr RE, Murphy WH, et al. Phenotypic variation including retinitis pigmentosa, pattern dystrophy, and fundus flavimaculatus in a single family with a deletion of codon 153 or 154 of the peripherin/RDS gene. Arch Ophthalmol. 1993;111:1531–42.
8. Berson EL, Rosner B, Sandberg MA, Dryja TP. Ocular findings in patients with autosomal dominant retinitis pigmentosa and a rhodopsin gene defect (pro-23-his). Arch Ophthalmol. 1991;109:92–101.
9. Richards JE, Scott KM, Sieving PA. Disruption of conserved rhodopsin disulfide bond by Cys187Tyr mutation causes early and severe autosomal dominant retinitis pigmentosa. Ophthalmology. 1995;102:669–77.
10. Richards JE, Kuo C-Y, Boehnke M, Sieving PA. Rhodopsin Thr58Arg mutation in a family with autosomal dominant retinitis pigmentosa. Ophthalmology. 1991;98:1797–805.
11. Berson EL, Sandberg MA, Rosner B, et al. Natural course of retinitis pigmentosa over a three-year interval. Am J Ophthalmol. 1985;99:240–51.
12. McCulloch C, McCulloch RJP. A hereditary and clinical study of choroideremia. Trans Am Acad Ophthalmol Otolaryngol. 1948;542:160–90.
13. Simmel O, Takki K. Raised plasma-ornithine and gyrate atrophy of the choroid and retina. Lancet. 1973;2:1031–3.
14. Carr RE. Congenital stationary night blindness. Trans Am Ophthalmol Soc. 1974;LSXXII:448–85.
15. Murayama K, Sieving PA. Different rates of growth of human and monkey photopic ERG suggests two sites of light adaptation. Clin Vis Sci. 1992;7:385–92.
16. George NDL, Yates JRW, Moore AT. X-linked retinoschisis. Br J Ophthalmol. 1995;79:679–702.
17. Nathans J, Davenport CM, Maumenee IH, et al. Molecular genetics of human blue cone monochromacy. Science. 1989;245:831–8.
18. Berson EL, Sandberg MA, Rosner B, Sullivan PL. Color plates to help identify patients with blue cone monochromatism. Am J Ophthalmol. 1983;95:741–7.
19. Berson EL, Rosen JB, Simonoff EA. Electroretinographic testing as an aid in detection of carriers of X-chromosome–linked retinitis pigmentosa. Am J Ophthalmol. 1979;87:460–8.
20. Sieving PA, Niffennager J, Berson EL. The electroretinogram in selected pedigrees with choroideremia. Am J Ophthalmol. 1986;101:361–7.
21. Rowe SE, Trobe JD, Sieving PA. Idiopathic photoreceptor dysfunction causes unexplained visual acuity loss in later adulthood. Ophthalmology. 1990;97:1632–7.
22. Marmor MF, Arden GB, Nilsson SE, Zrenner E. International Standardization Committee: standards for clinical electroretinography. Arch Ophthalmol. 1989;107:816–19.
23. Berson EL, Guras P, Hoff M. Temporal aspects of the electroretinogram. Arch Ophthalmol. 1969;81:207–14.
24. Meredith TA, Aaberg TM, Willerson D. Progressive chorioretinopathy after receiving thioridazine. Arch Ophthalmol. 1978;96:1172–6.
25. Yanoff M. The retina in rubella. In: Tasman W, ed. Retinal disease in children. New York: Harper & Row; 1971:223–32.
26. Boldi FC, Hervouet F. Syphilitic chorioretinitis. Arch Ophthalmol. 1968;79:294–6.
27. Skalka W. Asymmetric retinitis pigmentosa, luetic retinopathy and the question of unilateral retinitis pigmentosa. Acta Ophthalmol. 1979;57:351–7.
28. Matsubara S, Yamaji Y, Sato M, et al. Expression of a photoreceptor protein, recoverin, as a cancer-associated retinopathy autoantigen in human lung cancer cell lines. Br J Cancer. 1996;74:1419–22.
29. Milam AH, Saari JC, Jacobson SG, et al. Autoantibodies against retinal bipolar cells in cutaneous melanoma-associated retinopathy. Invest Ophthalmol Vis Sci. 1993;34:91–100.
30. Kim RY, Retsas S, Fitzke FW, et al. Cutaneous melanoma-associated retinopathy. Ophthalmology. 1994;101:1837–43.
31. Kellner U, Bornfeld N, Foerster MH. Severe course of cutaneous melanoma associated paraneoplastic retinopathy. Br J Ophthalmol. 1995;79:746–52.
32. Boughman JA, Vernon M, Shaver KA. Usher syndrome: definition and estimate of prevalence from two high-risk populations. J Chronic Dis. 1983;36:595–603.
33. Dryja TP, McGee TL, Reichel E, et al. A point mutation of the rhodopsin gene in one form of retinitis pigmentosa. Nature. 1990;343:364–6.
34. Farrar GJ, Kenna P, Jordan SA, et al. A three-base-pair deletion in the peripherin-RDS gene in one form of retinitis pigmentosa. Nature. 1991;354:478–80.
35. Kajiwara K, Berson EL, Dryja TP. Digenic retinitis pigmentosa due to mutations at the unlinked peripherin/RDS and ROM1 loci. Science. 1994;264:1604–8.
36. Swain PK, Wang Q-L, Chen S, et al. Mutations in the cone-rod homeobox gene are associated with the cone-rod dystrophy photoreceptor degeneration. Neuron. 1997; 19:1329–36.
37. Chang G-Q, Hao Y, Wong F. Apoptosis: final common pathway of photoreceptor death in rd, rds, and rhodopsin mutant mice. Neuron. 1993;11:595–605.
38. Berson EL, Rosner B, Sandberg MA, et al. A randomized trial of vitamin A and vitamin E supplementation for retinitis pigmentosa. Arch Ophthalmol. 1993;111: 761–72.
39. Steinmertz RL, Fitzke FW, Bird ZC. Treatment of cystoid macular edema with acetazolamide in a patient with serpiginous choriodopathy. Retina. 1991;11:412–15.
40. Converse CA, McLachlan T, Hammer HM. Hyperlipidemia in retinitis pigmentosa. In: LaVail MM, Anderson RE, Hollyfield JG, eds. Retinal degenerations. New York: Alan R Liss; 1985:63–74.
41. Hoffman DR, Birch DG. Docosahexaenoic acid in red blood cells of patients with X-linked retinitis pigmentosa. Retina. 1995;36:1009–18.
42. Faktorovich EG, Steinberg RH, Yasumura D, et al. Photoreceptor degeneration in inherited retina dystrophy delayed by basic fibroblast growth factor. Nature. 1990; 347:83–6.
43. Steinberg RH. Survival factors in retinal degenerations. Curr Opin Neurobiol. 1994;4:515–24.