Abnormalities of Cone and Rod Function

Published on 09/03/2015 by admin

Filed under Opthalmology

Last modified 22/04/2025

Print this page

This article have been viewed 3281 times

Chapter 44 Abnormalities of Cone and Rod Function

David M. Wu, Amani A. Fawzi

For additional online content visit http://www.expertconsult.com

This chapter will review a genetically heterogeneous group of retinal disorders, which often have subtle clinical findings in the face of distinct visual symptoms. Clinical suspicion paired with the appropriate psychophysical and electrophysiological testing can lead to a correct diagnosis. Research toward identifying molecular causes for these diseases continues at a rapid pace. As the underlying genetic mutations responsible for each entity are identified, phenotype/genotype correlations may lead to improved understanding of disease pathophysiology and pave the way for new classifications based on molecular mechanisms. Also, as new therapies emerge for a subset of these previously untreatable disorders, identifying the causative mutation will be of utmost importance to patients.

Congenital color vision deficits as well as retinitis pigmentosa and its related disorders are covered in separate chapters (respectively, Chapter 10, Color vision and night vision, and Chapter 40, Retinitis pigmentosa and related disorders).

Disorders of the cone system

Cone disorders can be divided into congenital defects that have early onset and are usually stationary, or those with later onset that tend to be progressive. The stationary disorders include complete and incomplete achromatopsia and blue cone monochromatism. Progressive dystrophies include those that involve only the cones (cone dystrophies) and those that have a component of rod degeneration (cone–rod dystrophies).

Achromatopsia

Congenital achromatopsia is a rare disorder, with an incidence of roughly 1 in 30 000.¹ Individuals with this disorder have poor vision from birth and complain of poor color discrimination and photosensitivity, which relates to their visual acuity decreasing in bright light rather than actual discomfort. Patients with complete achromatopsia, also known as rod monochromatism, are generally considered to lack cones and have vision worse than 20/200. On the other hand, patients with incomplete achromatopsia, also known as atypical achromatopsia, may have slightly better visual acuity in the range of 20/80–20/200. Complete achromats have no color vision whereas incomplete achromats may have some residual color vision. Interestingly, because their color vision loss is congenital, even complete achromats may be able to identify colors. They generally perceive colors as lighter or darker shades of gray and may have learned that red is a darker shade of gray and yellow is a lighter shade of gray and so forth. Consequently, specific color vision tests (discussed below and in Chapter 10,) are designed to uncover this characteristic. Pendular nystagmus may be present, which can improve with age. Family history can be useful as achromatopsia displays an autosomal recessive inheritance pattern.

Diagnosis

On clinical examination, patients may have a normal fundus, or have subtle granularity or atrophy of the macula. The nerve may be normal or show some temporal pallor. Electrophysiologial testing is crucial to making the diagnosis in achromatopsia. Photopic and 30 Hz-flicker electroretinograms will show nearly or completely nonrecordable cone responses in the face of normal or near-normal rod responses (Fig. 44.1). Multifocal electroretinograms will be severely diminished. Full-field electroretinography (ERG) is critical to diagnosis as it helps differentiate achromatopsia from other entities that share a normal fundus exam and associated symptoms. For instance, an infant with Leber congenital amaurosis will also have nystagmus, extremely poor vision, and photosensitivity but will have extinguished photopic and scotopic ERGs. A blue cone monochromat will have poor vision, but will have recordable cone responses when stimulated with blue light on a yellow background. A patient with optic neuropathy would have optic nerve pallor with otherwise unremarkable fundus findings as well as decreased vision and color perception, but would show normal photopic and scotopic ERGs.

Fig. 44.1 Color fundus photographs, optical coherence tomography (OCT), and electroretinography (ERG) of a series of patients with achromatopsia (yo = years old). The OCTs display a range of the most common features seen in achromatopsia. These findings include foveal hypoplasia, illustrated by preservation of the inner retinal layers through the fovea, most prominent in the top and middle OCTs. The middle OCT demonstrates, in addition, disruption of the IS–OS junction and thinning of the retinal pigment epithelium within the fovea (arrows), which is manifesting as increased signal transmission and increased visibility of the vasculature in the underlying choroid. The bottom OCT demonstrates an optically empty hyporeflective zone involving the outer fovea (arrowheads) with loss of the IS–OS junction and relative preservation of the external limiting membrane and the bare retinal pigment epithelium. To the right of each panel, the corresponding ERG shows normal to near-normal rod-isolated and maximum-combined ERG but unrecordable cone-isolated and 30 Hz flicker ERGs.

(Courtesy of Dr Stephen Tsang, Columbia University.)

Other ancillary tests are also helpful in the diagnosis. Congenital achromats may be able to identify several of the pseudochromatic (Ishihara) color plates because they may have learned to distinguish colors as different shades of gray. Farnsworth D-15 testing may reveal a scotopic axis between the deutan and tritan axes. The Sloan achromatopsia test uses an achromat’s correlation of different shades of gray to various colors in order to distinguish them from normal individuals.² Dark-adaptation curves will show a lack of a cone–rod break, reflecting the lack of cone function (Fig. 44.2). Visual field testing may reveal a small central scotoma but the peripheral visual fields should be either mildly constricted or normal, and, most importantly, they remain stable over time. Progression on serial visual field testing should raise suspicion for other disorders such as a cone–rod dystrophy or retinitis pigmentosa with a cone–rod pattern.

Fig. 44.2 Rod monochromatism. The dark-adaptation curve (A) shows an absence of the initial (cone) segment with a normal rod segment.

The ERG changes precede the less specific and sometimes more subtle findings seen on optical coherence tomography (OCT) (Fig. 44.1). A recent report calculated the frequency of these findings in a large series of 77 eyes of achromatopsia patients.³ Among the most obvious and common is disruption of the inner/outer-segment (IS–OS) junction occurring in 70% of these eyes. A hyporeflective, optically empty cavity may be seen in the cone layer of the foveola (60%), preferentially weighted to the nasal side of the foveal center, where there is higher density of cone photoreceptors. Also present in the majority of patients is foveal hypoplasia (83%). Disruption of the RPE and outer retinal atrophy can be seen in a minority of patients (18%). Although achromatopsia has traditionally been thought to be stationary, the imaging studies suggest that this view should be reconsidered.⁴ Progressive cone death may occur in achromatopsia patients with age as suggested by comparing the patient’s OCT to age-matched controls as well as a preferential age-dependence of features such as the hyporeflective zone, thinning of the outer nuclear layer, and RPE atrophy.^3,⁵ Given the overlap in genetic and clinical findings, it is also plausible that some patients diagnosed with complete achromatopsia were incomplete achromats who later progressed.

Although the technique is not readily available in most clinics, adaptive optics scanning laser ophthalmoscopic imaging of the macular photoreceptor mosaic in a patient with complete achromatopsia has helped shed light on the pathophysiology.⁶ This case has revealed that the macular area has significant gaps within the photoreceptor mosaic, with a total absence of photoreceptors bearing the characteristics of normal healthy cones. These remaining photoreceptors appeared to have morphologic characteristics typical of rods.

Clinical molecular genetic testing for the four known genes responsible for the majority of achromatopsia cases is available from multiple laboratories (http://www.ncbi.nlm.nih.gov/sites/GeneTests/lab/clinical_disease_id/211350?db=genetests; accessed February 2012). A brief discussion of our molecular understanding of achromatopsia is available online.

Molecular basis of achromatopsia

Most cases of congenital achromatopsia are associated with four genes, which are primarily inherited in an autosomal recessive fashion. All mutations affect proteins involved in cone phototransduction. Two major mutations affect proteins that code for the alpha- and beta-subunit of cone cyclic-GMP gated (CNG) cation channels. CNGB3 codes for the beta-subunit and is responsible for approximately 45% of complete achromatopsia. It was originally identified in the Pingelapese people in the western Pacific nation of Micronesia.⁷ The second most-common mutation (25%) is in the CNGA3 gene, which codes for the alpha-subunit.^8,⁹ The third gene that is associated with complete achromatopsia (<2%) is GNAT2, which codes for the alpha-subunit of cone transducin. Transducin is involved in the signal transduction cascade, and acts to close the cGMP channels.^10,¹¹ Finally, most recently, mutations in PDE6C have also been identified as a cause of achromatopsia (<2%).¹² This gene, which has also been implicated in retinitis pigmentosa and congenital stationary night blindness, codes for the catalytic subunit of cone photoreceptor phosphodiesterase.

The vast majority of incomplete achromatopsia has been linked to missense mutations in CNGA3, which are inherited in an autosomal recessive fashion. At one time, it was felt that this was because mutations in CNGB3 and GNAT2 were primarily those leading to nonfunctional truncated products and thus only yielding the complete achromatopsia phenotype. However, the discovery of a frameshift mutation in GNAT1 that results in a truncated protein associated with an incomplete loss of color vision argues against this.¹³ As only a few of the mutations identified to date in CNGA3 are exclusive to incomplete achromatopsia while most mutations can also be associated with complete achromatopsia, additional epigenetic modifiers are being sought that may attenuate the color vision and visual acuity defects in incomplete compared to complete achromatopsia.

Treatment

There is currently no treatment for the underlying defects in achromatopsia. Photophobia can be reduced with tinted lenses, particularly orange or red lenses since rod photoreceptors are less sensitive to orange and red lights. Red-tinted soft contact lenses that transmit between 400 and 480 nm can reduce the stigma of having to wear dark glasses even indoors.¹⁴ Low vision aids may help to enhance visual acuity. Great strides have been made in research whereby adenoviral-mediated gene transfer has been used to correct the defect in all three genetic forms of human achromatopsia in corresponding animal models.¹⁵^–¹⁷ As these gene-based therapies eventually move to human trials, identification of mutations in patients with achromatopsia will become crucial.

Cone monochromatism and blue cone monochromatism

Cone monochromatism is a congenital disorder in which two of the three cone systems (long- [L], middle- [M], and short-wavelength [S]) are absent or nearly absent. Monochromatism involving preservation of either the red or the green cone system is extremely rare, with an incidence estimated at 1 in 100 million.¹⁸ These patients have a normal visual acuity, no nystagmus and a normal ERG, residual red and green sensitive pigments by fundus densitometry, and spectral sensitivity curves similar to normal patients.¹⁹^–²¹ Given its rarity, the mechanism remains poorly understood and is mentioned here mainly for completeness.

The remainder of this subsection will concentrate on the relatively more common entity of congenital blue cone monochromatism, which affects approximately 1 in 100,000 people. In blue cone monochromatism, there is preservation of normal S-cone function with absent L- and M-cone function. The signs and symptoms resemble congenital achromatopsia, where those affected have poor vision in the range of 20/80–20/200 from birth. They may also have a pendular nystagmus and photosensitivity. Family history is a useful clue to distinguish the two entities, since patients with blue cone monochromatism have an X-linked recessive, compared to an autosomal recessive inheritance pattern in congenital achromatopsia.

Diagnosis

Fundus examination may be normal, or may reveal progressive pigment irregularities and macular atrophy on ophthalmoscopy and autofluorescence.²²^–²⁴ By history, symptoms, and visual acuity, these individuals resemble achromatopsia and thus there are several tests that may help the clinician distinguish between the two. Because they have functional S cones, these patients will have rudimentary dichromatic color discrimination and they can be distinguished from achromats based on the presence of residual tritan discrimination on color testing. One can also use Berson plates that have been specifically developed for this purpose.²⁵^–²⁷ Although the photopic and 30 Hz cone flicker will be absent under normal testing conditions, an S-cone signal can be recorded with a blue light-flash on a yellow background.²⁸ Scotopic and bright-flash ERGs can be normal or attenuated.^22,^28,²⁹ The disorder has been reported to be either stable or progressive, and it is unclear whether this variability is due to the specific mutation, epigenetic influences, and/or environmental factors. Clinical molecular genetic testing is currently available for some of the mutations involved with blue cone monochromatism (see online for a discussion of molecular mechanisms).

Molecular basis of blue cone monochromatism

The genetic basis of blue cone monochromatism can be explained in light of the high degree of variability in the expression of L and M opsins. Humans have one L opsin gene and one or more M opsin genes, which sit in a tandem array on the X chromosome. The S opsin gene resides separately, on chromosome 7.^30,³¹ To date, the primary mechanisms by which blue cone monochromatism occurs involve either a one-step or two-step mechanism.^30,^32,³³ Approximately 40% of patients will have a one-step deletion that affects the locus control region, an upstream conserved sequence which regulates transcription such that only one opsin gene is expressed per cone photoreceptor.^23,^24,³² The other 60% have a two-step mutation process. The first step involves a nonhomologous recombination event between the L and M opsin genes, which reduces the number of genes in the opsin array on the X chromosome to one. A second mutation then occurs in the remaining opsin gene, rendering it ineffective.³⁴ Not all cases are explained by mutations in the structure of the opsin array and it is likely that additional mechanisms will continue to be identified.^32,³³

Treatment

There is currently no treatment for blue cone monochromatism, but visual acuity and contrast may be improved by blue cutoff filters.³⁵ Patients should be counseled regarding the X-linked inheritance mechanism of the disorder.

Progressive cone dystrophies

These are inherited retinal dystrophies that usually present within the first three decades of life, although there are rare reports of patients presenting after the fifth decade.³⁶^–³⁸ Decreased visual acuity and central scotomas, some degree of color vision loss, and photophobia are the usual symptoms. The visual defects become worse over time. Some are purely cone dystrophies in which rod function remains normal. In these individuals, night blindness is not a major complaint, nor is peripheral field loss. However, there are many patients who, sooner or later, will have diminished rod function and will manifest the accompanying symptoms at that time. Cone–rod dystrophies are primarily nonsyndromic, but can also be associated with syndromes including Bardet–Biedl and spinocerebellar ataxia, so attention must be paid to a review of systems and the medical history. All mechanisms of inheritance are possible with cone–rod dystrophies, and a careful collection of family history will assist with narrowing down the myriad of mutations responsible for this phenotype, as well as with genetic counseling.

Diagnosis

When confronted with a patient with a relatively normal-appearing fundus and the complaint of uncorrectable bilateral decreased acuity and photosensitivity, the diagnosis of a cone disorder should be entertained. When macular atrophy appears in cone dystrophies it is usually symmetric between the two eyes. Peripheral pigmentary deposits are a rare occurrence. The progressive history rules out stationary disorders of cone photoreceptors (achromatopsia or blue cone monochromatism) discussed earlier in this chapter, and Leber congenital amaurosis. Although the symptoms may be consistent with Stargardt disease, the absence of the characteristic Stargardt findings (fundus flavimaculatus, bull’s eye, and dark choroid on fluorescein angiography) can help distinguish between the two. Later in the course of cone and cone–rod dystrophies there may be changes in fundus appearance ranging from pigment deposition, macular atrophy, vessel attenuation, and optic disc pallor. At that time, it may be more difficult to use fundus appearance alone to distinguish from entities such as retinitis pigmentosa.

Electrophysiology continues to be the mainstay diagnostic test to confirm the diagnosis and differentiate between cone or cone–rod dystrophies. Patients are expected to have significantly decreased photopic and 30 Hz flicker ERGs with delays in the implicit times. Multifocal ERGs will be diminished. Early in the course of the disease, when the clinical examination may be normal, this is the most definitive way to confirm the diagnosis. Scotopic ERGs may be reduced as well, especially in later stages depending on the degree of rod involvement in cone–rod dystrophies. However, significant reduction in scotopic amplitudes of the ERG in the early stages would argue strongly in favor of retinitis pigmentosa.

Visual field examinations are useful adjuncts to the diagnostic workup and follow-up evaluations. Early in the course of the disease the fields can show central scotomas with sparing of the periphery. This will remain stable in those with cone dystrophies, whereas patients with cone–rod dystrophies will show progressive peripheral loss over time.

OCT may show thinning of the outer retinal layers, limited mostly to the central retina.³⁹ Fundus autofluorescence can show central or parafoveal areas of increased or decreased autofluorescence with variable involvement of the periphery. However, the OCT and autofluorescence changes are nonspecific and may be similar in a number of other retinal degenerations as well. It is therefore important to remember that it is the electrophysiological evaluation coupled with these imaging studies which will help make the definitive diagnosis.⁴⁰

Although not readily available in most clinics, adaptive optics imaging has been used to investigate eyes with cone dystrophy. With this technique the photoreceptor mosaic shows increased cone spacing in cone dystrophy compared to normal subjects or retinitis pigmentosa patients.⁴¹

The diversity of mutations responsible for cone and cone–rod dystrophies currently precludes widespread screening for mutations. However, this may change as more efficient screening mechanisms such as microarrays become more readily available.⁴² A brief discussion of the molecular mechanisms of cone–rod dystrophies is given online.

Molecular basis of cone–rod dystrophies

The progressive cone–rod dystrophies are extremely genetically heterogeneous and may be associated with autosomal dominant, autosomal recessive, and X-linked inheritance patterns. At the time of publication, RETNET (http://www.sph.uth.tmc.edu/Retnet/) listed 23 genes with mutations that have been associated with a cone–rod dystrophy. For some of these genes, known mutations predominantly result in cone–rod dystrophy phenotypes.⁴³ The principal gene in this category is CRX, a homeobox gene controlling rod and cone photoreceptor development, which may be associated with up to 10% of autosomal dominant cone–rod dystrophy. Mutations in CRX can also result in a retinitis pigmentosa or Leber congenital amaurosis phenotype, but as of yet the disease phenotype cannot be accurately predicted based on the mutation.⁴⁴ Furthermore, some CRX mutations have been associated with extensive phenotypic heterogeneity within the same family.⁴⁵ This suggests that there may be modifiers of CRX gene function that remain to be elucidated to help explain the wide range of clinical and genetic heterogeneity in this disorder.

Cone–rod dystrophy phenotype may also be seen with specific mutations in genes that are typically associated with other retinal diseases. For example, ABCA4 is most commonly associated with Stargardt disease, but is also responsible for 30–60% of cases of autosomal recessive cone–rod dystrophies.⁴⁶ PRPH2 and RPGR are usually associated with autosomal dominant and X-linked retinitis pigmentosa, respectively, but can also cause cone–rod phenotypes of the same inheritance pattern.^47,⁴⁸ Many of the genes associated with Leber congenital amaurosis, including GUCY2D, can also cause a cone–rod dystrophy phenotype.⁴⁹ Since less than half of cone–rod dystrophy cases are associated with known genetic mutations, much remains to be uncovered about the molecular mechanisms of this disease.

Treatment

There are currently no therapies available to reverse the retinal degeneration process in cone dystrophies. Low vision aids, lenses that reduce photosensitivity, and occupational and psychosocial support remain the primary treatment modalities.

Congenital stationary night blindness

Congenital stationary night blindness (CSNB) is generally associated with nonprogressive defects in scotopic vision and/or dark adaptation with otherwise normal visual function. The initial description of the Nougaret family by Belgian ophthalmologist Cunier,⁵⁰ which was later expanded by Nettleship,⁵¹ ultimately encompassed nine generations and 2121 patients, and established that this was indeed a nonprogressive “stationary” night blindness. In fact, this classic description still bears the name Nougaret and represents one of the largest pedigrees in ophthalmology and one of the first documentations of a dominantly inherited disorder.⁵²

Since these initial reports, CSNB is now recognized to have many variants, which are associated with various inheritance patterns (autosomal dominant, recessive, or X-linked). Although most are truly stationary, some mutations lead to progressive visual loss. Myopia is linked to some forms, but not all. Finally, while the original family had a normal fundus, CSNB variants with fundus abnormalities are well characterized. We now know that the heterogeneity of CSNB is secondary to the variety of mutations that cause this condition. Clinical testing of some of the CSNB genes is currently available (for a list of genes and laboratories offering testing, see the GeneReviews website http://www.ncbi.nlm.nih.gov/sites/GeneTests/?db=GeneTests; accessed February 2012). At the time of publication, a new gene array is being developed in order to facilitate rapid and economical screening of all known mutations for CSNB.⁵³ As emerging therapies become available, there is a continued need for clinicians to recognize the clinical characteristics to determine whether their patients may benefit from emerging treatments.⁵⁴

Clinical suspicion, careful clinical examination and testing, in conjunction with appropriate genetic testing will ultimately help clinicians make the diagnosis of the particular variant of CSNB. A molecular classification that parallels and supplements the conventional clinical classification was elegantly laid out by Dryja in his Jackson Memorial Lecture from 2000.⁵⁵

CSNB with normal fundi

CSNB patients with normal fundus examination may have normal visual acuity and may not complain of night blindness. Only some forms (X-linked variants) are associated with myopia and generally have subnormal vision (~20/50). On electrophysiologic testing, patients with CSNB and normal fundi generally have severely abnormal rod-mediated ERGs, and some forms have modestly reduced cone amplitudes. Even when reduced in amplitude, cone responses are expected to have a normal implicit time in these patients. A delay in the rod or cone peak implicit time with relatively high amplitudes is more likely a harbinger of progressive retinal degeneration.⁵⁶ Other ancillary tests are of less value in these patients, who have normal visual fields. Color vision as measured by conventional psychophysical testing is also typically normal.

In the approach to a patient with suspected CSNB with normal fundi, one must consider the differential diagnosis for those with poor night vision for other reasons. This may include vitamin A deficiency from dietary deficiency or malabsorption, paraneoplastic syndromes, retinitis pigmentosa, and other retinal degenerations, such as choroideremia. A careful history will help elucidate additional symptoms and identify some of these entities. During the examination, most of these other causes will reveal additional fundus findings, such as gray white spots (vitamin A deficiency), peripheral pigmentation (retinitis pigmentosa), or choroidal changes (choroideremia), that exclude them from further consideration.

Diagnosis

In a CSNB patient with a normal fundus, the ERG and dark-adaptation curve are very helpful in characterizing the disorder and determining the specific variant. These patients may broadly be divided into two groups depending on the presence or absence of an a-wave on ERG (Fig. 44.3). Riggs type (also known as type I) patients lack a scotopic ERG, lack both an a- and b-wave on maximum bright-field stimulation ERG, and lack a rod–cone break on their dark-adaptation curve. Without a scotopic ERG and a rod–cone break on the dark-adaptation curve, the molecular defect is predictably located at the rod photoreceptor level (Fig. 44.4).

Fig. 44.3 Congenital stationary night blindness (CSNB), normal fundi. Schematics of dark-adapted ERGs at varying stimulus intensities. Subject RC has type II (Schubert–Bornschein), and subject PM has type I (Nougaret).

Fig. 44.4 CSNB, normal fundus, showing dark-adaptation curve of subject PM in Fig. 44.3. There is no evidence of a rod–cone break.

Schubert–Bornschein (also known as type II) patients possess an a-wave on maximum bright-field stimulation ERG but no b-wave, hence exhibiting a negative waveform. They are further classified into those who possess a scotopic ERG (incomplete form) and those who do not (complete form). In the incomplete form there is a rod–cone break on the dark-adaptation curve whereas in the complete form there is not.⁵⁷ Because there is an intact a-wave (originating in the photoreceptors) but diminished b-wave (originating in the inner retina) the defect is expected to originate not in the photoreceptors but within the inner retina. Our molecular understanding of CSNB correlates with these clinical findings. More detail is available online.

Molecular basis of CSNB

Since type I (Riggs) patients do not have an a-wave and lack a rod–cone break on the dark-adaptation curve, the functional defect is localized within the rod photoreceptors themselves. Indeed, the mutations tend to involve proteins of the rod phototransduction cascade, including rhodopsin, rod transducin, and rod cGMP phosphodiesterase. Such mutations result in constant low-level stimulation of the rod photoreceptors. Because only one mutant copy is necessary to stimulate the phototransduction cascade, these mutations are inherited in an autosomal dominant fashion. The classic phenotype described in the Nougaret family is caused by mutations in the alpha subunit of rod transducin (GNAT1). This phenotype is also associated with mutations in rhodopsin (RHO), and cGMP-phosphodiesterase (PDE6B).

For type II (Schubert–Bornschein) patients, electrophysiologic and psychophysical testing imply that the complete and incomplete forms affect both rod and cone pathways and predict that the defects may be located within the inner retinal circuitry, downstream of the photoreceptors. Not surprisingly, the mutations identified to date affect the formation and function of retinal synapses. The incomplete form has been linked to mutations in CACNA1F, a gene that encodes calcium-channels. These are believed to disrupt signaling of cones to the OFF bipolar cells or rods to the ON bipolar cells, since the calcium channels are concentrated mostly in the outer nuclear layers.^58,^59,⁶⁰ The complete form has been linked to nyctalopin (NYX), which is an adhesion molecule involved in formation of synapses in the ON pathway of the retina.⁶¹ CACNA1F and NYX are both located on the X chromosome, causing X-linked inheritance. More recently, there have been reports of autosomal recessive forms of the Schubert–Bornschein type. These have been associated with mutations in CABP4, a calcium-binding protein located in the synaptic terminals.⁶² For the complete form, mutations identified include GRM6, which is a metabotropic glutamate receptor on bipolar dendrites, and TRPM1, which are transient receptor potential channels in ON bipolar cells.⁶³^–⁶⁸

CSNB with abnormal fundi

CSNB with abnormal fundi includes two entities: Oguchi disease and fundus albipunctatus.

Oguchi disease

Oguchi disease refers to cases of CSNB in which there is a particular fundus finding known as the Mizuo–Nakamura phenomenon.⁶⁹ In this variant, the retina appears normal following prolonged dark adaptation, but on exposure to light the retina displays a golden sheen with an unusually dark macula⁷⁰ (Fig. 44.5).

Fig. 44.5 Oguchi disease. (Left) Fundus appearance showing a metallic reflex with vessels standing out in bold relief against this background. (Right) After 3 hours of dark adaptation the retina reverts to normal color.

Diagnosis

Visual acuity and color vision are typically normal in this disorder. On clinical testing, in addition to the abnormal fundus appearance, patients have a dark-adaptation curve with a cone component (Fig. 44.6) but no rod–cone break, and exhibit gradual recovery of full rod sensitivity after prolonged dark adaptation of 1–2 hours. There is no scotopic ERG under normal testing conditions.⁷¹ A single-flash ERG following prolonged dark adaptation will be normal, but then subsequent flashes will not generate a response until the patient undergoes another prolonged period of dark adaptation.⁷²

Fig. 44.6 Dark-adaptation curve in Oguchi disease. Normal cone adaptation is obtained, but there is marked delay in obtaining a normal rod threshold.

The mechanism underlying the Mizuo–Nakamura phenomenon and fundus sheen seen in Oguchi disease remains a subject of speculation. Histopathological study suggests the presence of an abnormal layer between photoreceptor outer segments and the retinal pigment epithelium (RPE).⁷³ Recent OCT studies showed that the outer segments appeared shortened in areas with the golden sheen reflex seen in light and suggested that the sheen is related to accumulation of photoactivated rhodopsin in shortened rod outer segments.⁷⁴ Other authors have shown that the IS–OS junction was better visualized on OCT following prolonged dark adaptation (and disappearance of the golden sheen), suggesting that reversible changes in the rod photoreceptors related to this mutation may explain the golden sheen reflex.⁷⁵ More detail on the molecular basis is available online.

Molecular basis of Oguchi disease

Mutations in Oguchi disease have involved the rhodopsin kinase (GRK1) or arrestin (SAG).^76,⁷⁷ To date, all identified mutations are autosomal recessive requiring both copies to be affected in order to manifest a phenotype. Both rhodopsin kinase and arrestin deactivate photoactivated rhodopsin. Mutations in this gene are either less efficient or unable to inactivate rhodospin, desensitizing the rod photoreceptor. Thus, the rod photoreceptor supplies of chromophore are replenished by photoisomerized chromophore that is replaced during rhodopsin regeneration from the RPE, which takes 4–7 hours. This explains the kinetics of the dark-adaptation curve.⁷⁸

Fundus albipunctatus

Fundus albipunctatus describes a subgroup of CSNB in which white or yellow dots can be seen scattered through the fundus (Fig. 44.7). Individuals may complain of night blindness early in childhood without progression, though most patients remain asymptomatic until the characteristic flecks are detected incidentally on routine fundoscopy.

Fig. 44.7 Color fundus photographs of a 10-year-old girl with fundus albipunctatus. By history, she has had poor vision under scotopic conditions since a very young age. The scotopic ERG was borderline abnormal, while the final dark-adapted thresholds were elevated 2.0 log units above normal. (A) and (B) Fundus photographs showing the typical yellowish-white spots in the paramacular region of the right and left eyes. (C) Peripheral temporal view of the right eye illustrating the tendency of the deposits to become larger in the peripheral fundus compared to the paramacular region. (D) Higher magnification of the deposits nasal to the optic nerve.

(Courtesy of Dr Thomas O’ Hearn.)

Diagnosis

As with other forms of CSNB, visual acuity and color vision are typically normal in this disorder. A scotopic ERG can be recorded but only after unusually long dark adaptation, whereas the cone ERG is usually normal. The dark-adaptation curve of fundus albipunctatus features prolongation of both rod and cone sensitivities (Fig. 44.8). Consistent with this, rod and cone visual pigment regeneration is significantly slower in patients with fundus albipunctatus (Fig. 44.9).⁷⁹ Of note, there is also a form of fundus albipunctatus associated with cone dystrophy, in which there is progressive decrease in visual acuity and color vision with bull’s-eye maculopathy and significantly reduced cone ERGs.⁸⁰

Fig. 44.8 Dark adaptation in fundus albipunctatus compared with normal (N). There is marked prolongation of both cone and rod segments until normal thresholds are reached.

Fig. 44.9 Regeneration times of visual pigments in fundus albipunctatus. Sixteen degrees temporal retina (circles) shows the very prolonged regeneration times of rhodopsin (half-time = 6 minutes). The prolonged regeneration time of the cone pigments (squares) is shown in such patients (normal half-time of regeneration = 55 seconds).

Optical coherence tomography and autofluorescence imaging have recently been used to investigate the flecks in fundus albipunctatus with and without cone dystrophy.^81,⁸² In both reports, OCT demonstrated homogeneous dome-shaped hyperreflective deposits originating from the inner RPE that correlated with the white fundus spots and were noted to project into the outer retina, disrupting the IS–OS junction, the external limiting membrane, and the outer nuclear layer. Autofluorescence revealed areas of hyperautofluorescence corresponding to the elevated lesions on OCT. It is hypothesized that these lesions represent accumulations of retinoids secondary to the disruption of production of 11-cis retinal. The patient with cone dystrophy in conjunction with fundus albipunctatus also had a bull’s-eye lesion with corresponding loss of the IS–OS junction on OCT and abnormal circular hypoautofluoresence. More detail on the molecular basis is available online.

Molecular basis of fundus albipunctatus

Molecular studies have demonstrated that mutations in fundus albipunctatus largely affect the gene for 11-cis retinol dehydrogenase (RDH), which converts 11-cis-retinol to 11-cis retinal.^83,⁸⁴ This enzyme is located in the retinal pigment epithelium and is involved in production of 11-cis retinal that is then transported to photoreceptors for incorporation into the photochromophore. Recently, there has also been a report of fundus albipunctatus secondary to a mutation in RPE65, which is involved in the conversion of all-trans retinol into 11-cis retinol, a step upstream of the one catalyzed by 11-cis retinol dehydrogenase.⁸⁵ Not only does the molecular basis explain the kinetics of the dark-adaptation curve and the involvement of both the rods and cones, but because the enzymes are located in the retinal pigment epithelium, this may explain the imaging findings and how the flecks of fundus albipunctatus (presumed to be retinoids based on their hyperautofluorescence) appear to arise in the RPE.^81,⁸⁶ Since these mutations affect regeneration of 11-cis retinal in the RPE, thus affecting both rods and cones, it is not surprising that there is an element of cone dysfunction in some forms of fundus albipunctatus. It is interesting to note that the cone system may seem more affected than the rod system in some cases of fundus albipunctatus. In these patients there is often reduced visual acuity and bull’s-eye changes within the macula. One study found that 38% of patients with fundus albipunctatus and mutations in RDH5 had extensive, progressive cone dysfunction on ERG testing and, to a lesser degree, decreased rod dysfunction.⁸⁷ It is still unclear why the cones are potentially more functionally affected than rods in these patients.