If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Robert H Henderson BSc MBBS MD FRCOphth, Consultant Paediatric Ophthalmologist and Adult and Paediatric Vitreoretinal Surgeon, Great Ormond Street Hospital and Moorfields Eye Hospital, London, UK. Conflicts of interest: Advisory Board Member for the Novartis UK Voretigene-Neparvovec gene therapy programme
Inherited retinal dystrophies (IRD) are a diverse group of progressive blinding genetic diseases that can present from birth through to late middle age. Symptoms include loss of night vision, visual field, colour, and central acuity. Sophisticated imaging modalities and electrophysiology permit genotype–phenotype correlations. The genetics are increasingly understood and have led to the development of gene therapy programmes, one of which is now a licensed treatment. Identifying these patients and establishing a genotype early is now of greater significance given the potential for treatment. It is important for paediatricians to understand the risk of systemic associations and equally, the developmental impact of blindness; supporting patients and families together with ophthalmology is vital.
In 1857 Cornelius Franz Donders, using the then newly developed ophthalmoscope, identified patients with black pigment scattered throughout the peripheral retina. He postulated that this must be inflammatory in nature and so coined the term Retinitis Pigmentosa. RP is one of a diverse group of, what are now understood to be, inherited retinal dystrophies (IRD) that are estimated to affect 1:3500 people.
Retinal Dystrophies can be categorised by the subtype of cell that is predominantly affected e.g. in Rod-Cone dystrophies (such as RP), the rod photoreceptor is affected worse than the cone primarily, whilst with macular, cone, or cone-rod dystrophies the opposite is true. Dystrophies are by definition progressive. In RP, the primary symptom is loss of night time vision (nyctalopia) and peripheral visual field, before subsequently involving central macula function, with reducing central visual acuity, colour, and contrast sensitivity. Patients with cone, cone-rod or macula dystrophies tend to present with photophobia and loss of central vision with reduced colour and contrast sensitivity.
Inherited retinal dystrophies may present at birth or in the first 6 months of life at the severest end of the spectrum known as Leber's congenital amaurosis (LCA). Severe early onset retinal dystrophy (SEORD) affects young children, whilst juvenile onset RP is generally thought of as affecting older children and teenagers when another unrelated retinal degeneration, juvenile X-linked retinoschisis also presents. The commonest macula dystrophies presenting in childhood include Stargardt and Best disease. Stationary non-progressive cone and rod dysfunction syndromes such as achromatopsia (ACH) or Congenital Stationary Night Blindness (CSNB) also present in childhood.
There are a number of childhood onset IRDs that have systemic features including but not limited to Usher syndrome (sensorineural hearing loss); Alstrom (hearing loss, cardiomyopathy, obesity); Bardet Biedl (obesity, reduced renal function, polydactyly); Joubert (Cerebellar hypoplasia); Senior-Loken (nephronophthisis), Cohen (microcephaly, hyotonia, intellectual disability, characteristic facies) or Battens - neuronal ceroid Lipofuscinosis (neurodegeneration). It should also be remembered that there are four syndromes associated with a retinal degeneration for which a treatment exists: abetalipoproteinaemia – treated with vitamins A and E; Gyrate atrophy – with elevated plasma ornithine and treated with a low protein diet; ataxia with vitamin E deficiency (AVED); Refsum disease treated with dietary reduction in phytanic acid.
The genetics of retinal dystrophies and stationary dysfunction syndromes are increasingly understood and there are over 270 genes known to be responsible. These encode a wide variety of proteins responsible for numerous cellular functions in both photoreceptors the retinal pigment epithelium and other retinal cells. The majority of these conditions have a simplex, or autosomal recessive pattern, but all modes of inheritance are found (Figure 1).
The purpose of this article is to give the reader a broad introduction to some of the phenotype and genotypes of isolated retinal dystrophies and those with systemic associations. The investigation and management of IRD will be explained; finally, current and future therapeutic avenues will be explored.
This electrodiagnostic test provides a mainstay of IRD testing. Skin electrodes can be applied to the lower lids in young children, while more sensitive gold wire or contact lens electrodes can be used with older children and adults. A standardised testing algorithm of both flash, pattern, and flicker stimuli are used to isolate retinal components. The negative deflection to a standard flash of light provides the a-wave and corresponds to photoreceptor function; the subsequent hyper-polarisation forms the b-wave and is a measure of inner retinal - bipolar cell - function. These flashes are performed in both light and dark-adapted states to isolate rod and cone function. The 30 Hz flicker stimulus isolates cone function as rods cannot respond to a stimulus faster than 20 Hz. Older children and adults are capable of performing more in-depth testing that allows isolation of separate cone sub-classes, and detection of more subtle inner retinal dysfunction (Figure 2).
Fundus autofluorescence (FAF)
Lipofuscin, the break down product of outer segment disc phagocytosis, is stored in the RPE. It acts as a fluorophore and, when stimulated with appropriate wavelength light, provides a surrogate marker of RPE-photoreceptor activity. Hyper-autofluorescence suggests increased activity which may reflect photoreceptor cell stress; whilst hypo-autofluorescence indicates cell death. Frequently, in RP, a hyperfluorescent ring around the macula can be seen which represents a watershed zone between dead peripheral retina and surviving central macula cells; this can be monitored over time and will, as disease progresses, shrink in size corresponding with a reduction in peripheral visual field. Characteristic patterns of autofluorescence can provide definitive diagnoses e.g. Stargardt disease (Figure 3).
Optical coherence tomography (OCT)
OCT uses light reflection off tissues of differing density to build a cross sectional image in a manner analogous to ultrasound. This has transformed the understanding of retinal pathology, allowing clinicians to visualise the retina in almost histological detail. Software analysis and image registration allows for longitudinal measurement of retinal thickness, segmentation of retinal layers, and progress of disease.
Ophthalmology has been something of a vanguard specialty in terms of utilising modern genomic techniques to understand the molecular mechanisms behind retinal disease. Since mapping of the first gene for XLRP in 1984, over 270 retinal disease genes and additional loci have been identified. Next Generation Sequencing and whole exome panels are more commonly utilised over single gene testing. There is an increasing recognition that many of the older disease nomenclatures are being redefined as the full spectrum of phenotype–genotype correlations is better appreciated. Stargardt disease, for example, has a classic fundus appearance and is caused most commonly by mutations in the ABCA4 gene. It is now appreciated however, that ABCA4 retinopathy has a much wider range of phenotypes than was previously encompassed by Stargardt. Similarly, Best disease has a tight phenotypic description but the phenotypes associated with mutations in the Bestrophin gene are much more extensive and are termed Bestrophinopathies.
Following the Genomics England 100,000 Genome study, the provision of specialty genetic testing has been reorganised around genomic hubs. The provision of Next Generation Sequencing panel tests for ocular genetic disease is now provided by three centres in the UK – Great Ormond Street, Birmingham and Manchester. Turnaround times vary but will, once the reorganisation is complete, should come down to 6 months. With accurate phenotyping, and the correct panel test instituted, there is approximately a 50–60% possibility of detecting a causative gene mutation. Given the relatively high probability of detecting a gene, the increasing availability of treatments or treatment trials, and the importance of informing parents of the risk to future pregnancies, it is advised that genetic testing be undertaken early in the diagnostic process.
There is currently only one licensed treatment for any IRD – Voretigene-Neparvovec – a gene therapy from Novartis. Vitamin A or alternate retinoid supplementation has proponents though the evidence for its efficacy is limited and, moreover, is retino-toxic if given to patients with Stargardt disease. Carbonic Anhydrase inhibitors can be used either orally or topically to try to limit the macular oedema that sometimes accompanies advanced disease and further damages vision. Children and adults will often benefit from having a low visual aids assessment, and this should form part of the ophthalmic pathway. If vision is compromised, certification of visual impairment is advisable and referral to Visual Impairment services.
The retina is an ideal target tissue for genetic modification: it is easily accessible, visible, and it is immune-privileged. Adeno-Associated Viral (AAV) vectors are widely employed in a number of trials world-wide and is the vector used to deliver Voretigene-Neparvovec. The limitation of AAV remains the ‘carrying capacity’ allowing it to accommodate genes of only modest size. The two most common genes causing IRDs are USH2A and ABCA4 accounting for 25% of disease, but the cDNA of both are two to four times larger than the AAV packaging capacity. Lentiviral vectors have a larger capacity but do not yet have the same safety record or specificity for transfecting the outer retinal layers. In addition, some genes, when over expressed, can cause retinal toxicity. For many older patients who have advanced disease, gene therapy will be insufficient, and some form of replacement treatment will be necessary to restore sight.
Stem cell treatments: There are currently no clinically available stem cell treatments for IRDs though there are an increasing number of early phase studies. Replacing the RPE monolayer may well be a feasible treatment avenue in the reasonably near future, but stem cell replacement of photoreceptor and inner retinal layers is still largely in the domain of pre-clinical research. Nevertheless, the treatment of IRDs with induced pluripotent stem cells, coupled with CRISPR-Cas9 gene editing technologies offers the best long-term answer to the treatment of IRDs.
Artificial retina: Several retinal prostheses have been developed and one is now FDA/EMA approved (Argus II – Second Sight, New York). These are indicated in patients with minimal or no light perception, albeit with a precondition of having had previous visual experience. There is currently no literature on using the device in children. The Second Sight device consists of an electrode array that is tacked onto the inner retinal surface over the macula. It consists of a grid of stimulating electrodes that are individually tuneable; each generate phosphenes which are perceived as a pulse of light. The array is connected via a cable to the outside of the eye and is powered by an induction coil mounted on glasses, that also carry a camera. The pattern of stimulation comes from a belt worn processor unit that has input from the camera. The visual signal is crude but permits recognition of large letters on a screen. The same company are currently working on a system (OrionⓇ) to bypass the eye entirely with direct stimulation of the visual cortex.
Small molecules and pharmacogenomics
A number of small molecules, read through drugs, and SiRNA based treatments are in varying states of development, though none are as yet licensed for use.
Ciliary neurotrophic factor (CNTF) is a small protein belonging to a family that includes IL6 and Leukaemia inhibitory factor. It has high expression levels in the CNS and has well documented neuro-protective capabilities in animal models however clinical trials of encapsulated CNTF in the eye have failed to demonstrate similar effects. Rod Derived Cone Viability Factor (RdCVF) is a diffusely secreted neuroprotectant that, in animal models, appears to protect cones. Exogenously administered RdCVF promotes cone survival in RP mutant mice; clinical trials are awaited. Ataluren is one of a number of aminoglycoside drugs that promote read-through of certain nonsense mutations that would otherwise lead to nonsense mediated decay (NMD). It has been demonstrated in cell lines and mouse models of disease that these small molecules permit gene function by read-through of the premature termination codons. Subsequent early clinical work has shown promising results in NMD diseases varying from Cystic Fibrosis, to Duchenne Muscular Dystrophy. Trials are now planned for several retinal dystrophies including Choroideremia and Usher Syndrome.
IRD presentations in childhood
Leber's congenital amaurosis (LCA)
The incidence of LCA is 1/33,000 and accounts for 20% of children in blind schools. Children present at birth, or in early infancy, with a history of roving nystagmus. There is an absence of, or very limited, visual response and children have sluggish or amaurotic pupil reactions. There is no detectable electroretinogram. Children may be photophobic and often have a moderate to high long-sighted (hypermetropic) refractive error. The retinal appearances vary between being normal to widespread retinal pigment or macular atrophy.
LCA is inherited in an autosomal recessive manner. There are currently 18 genes associated with LCA, several of which display close genotype–phenotype correlations. Children with mutations in CRB1 have a characteristic ‘nummular’ pigment clumping at the level of the RPE and the retina is generally very thickened on OCT (Figure 4b). The retinal appearances associated with NMAT1 include macular atrophy. Patients with biallelic RPE65 mutations have a history of photo-attraction so strong that infants often ‘light stare’ and have profound night blindness. The retina is generally normal in appearance, but there is almost no fundus autofluorescence seen.
Children presenting with LCA and Early Onset Retinal Dystrophies in general should have a general paediatric workup including an MRI and renal evaluation as there are associations with syndromic conditions including Senior-Loken Syndrome with nephrocalcinosis, or Joubert Syndrome with the typical cerebellar features and intellectual disability.
Paediatricians should also be aware that children with low or no vision are at significantly greater risk of developmental delay and a direct referral to the specialist teaching service for children with Visual Impairment is warranted.
Blind children often poke or rub their eyes – the so called oculo-digital sign of Franceschetti. Older children and adults report two reasons for it: that it elicits the sensation of phosphenes, or makes them feel calmer, which may result from the production of endorphins. Eye-poking/rubbing may result in the development of keratoconus, a conical appearance to the cornea, cataract and loss of orbital fat giving enophthalmos - a sunken appearance. Trying to stop children from doing this seems fruitless however, and many clinicians assuage parental concern about this behaviour.
X-Linked retinitis pigmentosa (XLRP)
Over 300 mutations in the RPGR gene, most of which occur in the open reading frame exon (ORF15), cause an abnormally short protein that is expressed in the connecting cilium of photoreceptors, and is an important component of all ciliated cells in the body. A small subset of patients with mutations in RPGR develop primary ciliary dyskinesia with recurrent chest infections, otitis media, and sinusitis.
Together with mutations in RP2, these two genes account for the significant majority of cases of XLRP. The disease causes progressive visual loss in boys and young men with nyctalopia, loss of peripheral visual field loss, and eventual progression to legal blindness in the 4th or 5th decade. Cataracts are common in later disease, and many patients develop macula oedema.
Stargardt macular dystrophy
Stargardt is the most common macular dystrophy in both children and adults, affecting 1:8000. The inheritance is autosomal recessive with over 900 mutations throughout the ABCA4 gene, though mutations have also, more rarely, been identified in ELOVL4 and PROM1. ABCA4 is expressed on the outer segments of photoreceptors and functions to process the metabolites of Vitamin A in the visual cycle. Excessive build-up of these breakdown products accumulate within the RPE and are toxic to the overlying retina. The characteristic appearance is of pale flecks, or fine white dots, accumulating under the macula that cause progressive atrophy of RPE and retina, most easily visible on fundus autofluorescence (Figure 3, Figure 5). Generally, the earlier the onset of disease, the more severe the final outcome in terms of vision. Children can present from 5 years of age and older with loss of central vision, poor colour vision, reduced contrast sensitivity, and often complain of photophobia. Progression is slow over years but atrophy at the macula and severe reduction in central acuity is the most common final outcome. Children will frequently need visual impairment registration during their early teenage years and increasing support at school with text enlargement and dedicated VI teachers as the disease progresses.
Stargardt treatments: Patients are advised to avoid bright sunlight and wear good quality sunglasses. Additionally, excessive vitamin A consumption is likely to aggravate the disease process, though reducing dietary vitamin A is not likely to be helpful.
Vitelliform macular dystrophy or Best disease is caused by mutations in the BEST1 (VMD2) gene and is inherited in an autosomal dominant fashion; there is however variable penetrance. The prevalence is estimated at between 1 and 9/100,000. The gene product, Bestrophin, is a transmembrane protein expressed on the basolateral aspect of the RPE cells, and functions to facilitate chloride conductance across the RPE. Mutations affect fluid transport across the RPE and lead to an accumulation of debris between the photoreceptors and RPE. The classic presentation of Best disease is with an “egg yolk” appearance at the macula (Figure 6).
This breaks down with time leading to an atrophic macula scar and central visual loss. Mutations in BEST1 can also cause a number of other retinal diseases including retinitis pigmentosa and ADVIRC (Autosomal Dominant VitreoRetinoChoroidopathy); multifocal small egg yolk deposits may be caused by biallelic mutations in BEST1 giving Recessive Best Disease. The diagnosis is made clinically but performing an ElectroOculoGram (EOG) can sometimes be helpful where the diagnosis is in question.
The disease can be complicated by the development of choroidal neovascularisation with haemorrhage and leak into the retina; these can be successfully treated with intravitreal anti-VEGF agents.
Juvenile X-linked retinoschisis
The retinoschisin gene RS1 encodes a protein that is primarily involved in cell adhesion. The retina in male patients with RS1 mutations have cystic cavities through the middle of the retina which progressively enlarge with age and cause an associated reduction in visual acuity. With disease progression, in the retinal periphery, the inner ‘leaf’ of split retina can degenerate with large atrophic holes developing, leaving the residual retinal blood vessels appearing to hang in the vitreous cavity above the retina. This can be a cause of vitreous haemorrhage (Figure 7).
Patients are almost always long sighted in their refractive error. The prevalence ranges between 1:5000–1:25,000 worldwide. Central visual acuity is slowly lost and most children complete a fully sighted education, albeit with enlarged texts and Visual Impairment teacher support in main stream school.
Treatment of the retinal cysts is commonly with topical or oral carbonic anhydrase inhibitors, though these usually have only a modest effect.
Congenital stationary night blindness (CSNB)
This is a polygenic disease with 17 genes currently associated. It is most easily diagnosed on electroretinography. There are four subtypes - Schubert Bornstein (further divided into complete and incomplete type), Riggs, Fundus Albipunctatus and Oguchi Disease – the latter two of which have an abnormal fundus appearance. The classical clinical description is one of poor night or low illumination vision and some photophobia that starts at birth or in early infancy. CSNB is associated with myopia and a fast, low amplitude nystagmus and is non progressive. It is now recognised that many children with ‘incomplete’ CSNB are not aware of any symptoms, though their central visual acuity is usually reduced to a variable degree.
This rare autosomal recessive disease affects 1:30,000–40,000 people and is caused by mutations in one of 6 differing genes: CNGB3 and CNGA3, which account for around 75% of all cases, as well as GNAT2, PDE6C, PDE6H and ATF6. It causes complete colour blindness and reduced central visual acuity. Patients present with profound photophobia and nystagmus in the first months of life. The nystagmus tends to be horizontal or pendular which differs from the roving nystagmus seen in LCA. The diagnosis is made on electrophysiology where cone cell function is absent whilst rod function is normal. There are complete and, more rarely, incomplete forms - with a less severe phenotype in the latter. The disease is essentially stable. The glare and photophobia can be ameliorated with either a brown or red tint in glasses.
This dual sensory disease affects hearing, balance and vision. In Type 1 Usher syndrome, babies are born with profound sensorineural hearing loss and, most commonly, have vestibular hypofunction. Early onset retinitis pigmentosa is detectable on ERG in the first years of life, but the symptoms are not usually present until 5–10 years of age. Central visual loss is uncommon for many years and nearly half of type 1 Usher syndrome patients will maintain 6/12 visual acuity in their better seeing eye until the age of 50 years albeit with significant visual field deficits. The commonest genetic cause is MYO7A (USH1B) (Figure 8) which accounts for approx. 50% of Usher 1 patients, but there are 10 known loci (USH1 A-K) including 6 identified genes.
Children with Usher Syndrome type 2, have congenital mild to moderate sensorineural hearing loss in the low frequencies, and severe to profound hearing loss in the higher frequencies. The vestibular reflexes are intact. The retinitis pigmentosa does not start causing symptoms until teenage years and has a better visual prognosis than Type 1, with two thirds of patients retaining 6/12 vision in their better eye until 50 years of age. The commonest genetic cause is USH2A, a gene that is also the commonest cause of non-syndromic RP.
Leber's congenital amaurosis
The first ocular gene therapy trials were performed for patients with autosomal recessive RPE65 mutations and published in 2008. In 2018 the EMA granted a license for Luxturna (Voretigene-Neparvovec, Novartis) and NICE have recently agreed to fund this. Results indicate an improvement in scotopic (low light) visual function, and a slight improvement in photopic macula function that appears to be maintained, at least over the short to medium term, with one subretinal injection. There are gene therapy trials underway for patients with GUCY2D and CEP290, and further preclinical success in animal models of CRB1 and AIPL1 related disease. Antisense RNA oligonucleotide trials for the commonest mutation in CEP290 are also underway.
X-Linked RP (XLRP)
There are a number of gene therapy trials based both in the UK and overseas for patients with RPGR related XLRP. Phase I studies have produced some encouraging results. High dose vitamin A palmitate is suggested by some in the US to slow the rate of vision loss but is not commonly prescribed in the UK. Carbonic anhydrase inhibitors are commonly used to reduce the macula oedema.
Phase II studies of orally bioavailable drugs designed to prevent build up of toxic metabolites, A2E and lipofuschin, within the RPE are now commencing. There are also a number of gene therapy and stem cell approaches also in phase I and II trials worldwide. The ABCA4 gene is large, precluding use of the most commonly used vector – Adeno Associated Virus (AAV). One approach is to break the gene into two parts and package each separately with AAV but deliver both simultaneously, with intracellular DNA recombination of the gene appearing to occur in a proportion of transfected cells.
The cochlea implant program has transformed the hearing loss element of this disease, but the RP remains untreatable currently. USHSTAT a gene therapy trial from Sanofi, for patients with MYO7A mutations is currently underway in a number of countries. Small molecule drug trials are also in the pipeline with Ataluren, a read-through aminoglycoside for nonsense mutations; and anti-sense oligonucleotides for specific mutations within MYO7A.
There are ongoing gene therapy trials in the UK for patients with CNGA3 and CNGB3 but there are no other treatments currently available.
Congenital stationary night blindness
There is no treatment currently available, though a small study looking at oral administration of 9-cis BetaCarotene, was effective in improving visual fields and ERG amplitudes in patients with RDH5 mutations.
Juvenile X-linked retinoschisis
The only RS1 gene therapy trial involved an intravitreal injection of vector but, to date, has not produced evidence of significant improvement in the older patients that received treatment; there remains some hope that treating younger patients before the cysts enlarge excessively may prevent further progression.
Early diagnosis is crucial both for anxious parents who are most often aware that their child's vision is not normal, and to support childhood development and education. There are a number of tertiary centres in the UK where electrodiagnostic testing is available for children and subsequent genetic testing can be undertaken. Genetic Testing for Ophthalmic disorders is now based in three genomic hubs: Great Ormond Street, Birmingham, and Manchester.
Visual impairment registration, either as sight impaired (SI) or severely sight impaired (SSI), is very often warranted and should be undertaken by the diagnosing ophthalmologist. This registration process then informs local social services departments and associated the Local Educational Authority who, in turn, should provide Special Educational Needs assessment and access to Visual Impairment teachers.
Children with inherited retinal diseases should, in general, be reviewed on an annual basis both to observe any progress in the disease state, but also to provide the correct refraction and glasses prescription; monitor and treat macula oedema or cataract; ensure that children are getting adequate support at school; and provide access to low visual aid services as well as providing reports to the local VI teachers and community paediatricians of a child's visual needs.
Increasingly, gene therapy and other novel therapeutic avenues are becoming a reality. Where previously no real hope existed, there is now cautious optimism that at least some inherited retinal diseases may be amenable to treatments that will stabilise if not improve the visual prognosis.
A baby with irregular, large amplitude or roving nystagmus may have severe visual impairment and early referral to a paediatric ophthalmologist is warranted.
A diagnosis of Leber Congenital Amaurosis or early onset retinal dystrophy may be associated with a wider syndrome: consider a renal ultrasound, blood pressure and bloods, as well as an MRI to rule out Senior-Loken or Joubert Syndromes.
A severely visually impaired child will not have the cues to reach out and explore the world; this can often be a cause for developmental delay and should be assessed for.
Early referral to the Specialist teacher for Visual Impairment (QTVI) is important.
If no genetics have been requested by the ophthalmologist, then consider a referral to clinical genetics for NGS panel testing.
Ensure family support/followup/EHCP plans are in place.