Article Text
PDF
Abstract
Background A prospective, national population-based cross-sectional study to enable understanding of the burden and management in the UK of hereditary retinal disorders presenting in childhood.
Methods Children aged <16 years with a new diagnosis of an inherited retinal disorder made between September 2006 and February 2008 in the UK were identified through two national active surveillance schemes. Clinical and socio-demographic information was collected on each child at diagnosis and 9 months later using standardised questionnaires.
Results 241 patients were reported with 24 distinct diagnoses. 14% had additional systemic disorders and 13% had dual sensory impairment. Annual incidence was 1.4/100 000 children (aged 0–15 years) and the cumulative incidence by age 16 years was 22.3/100 000 children. The most common mode of inheritance was autosomal recessive. A significantly higher rate was seen in males than females (relative rate (RR) 1.53), in children of Asian compared with White ethnicity (RR 7.12) and in those in the worst quintile of socio-economic deprivation compared with those in the best (RR 1.43). Parents most commonly detected a problem with their child's vision. Up to seven different health professionals were involved in a child's early management, and variations were noted in the proportion of eligible children having assessments for low vision aids, statement of educational needs and certification as sight-impaired.
Conclusions These findings illustrate the highly heterogeneous nature of childhood retinal dystrophies and provide previously unavailable data on disease incidence, distributions and management, which are important for service provision and for planning future treatment programmes, particularly as novel therapies become available.
- Retinal disorders
- childhood
- incidence
- retinal dystrophy
- retina
- dystrophy
- epidemiology
- child health (paediatrics)
Statistics from Altmetric.com
- Retinal disorders
- childhood
- incidence
- retinal dystrophy
- retina
- dystrophy
- epidemiology
- child health (paediatrics)
Introduction
Childhood-onset hereditary retinal disorders comprise a group of uncommon inherited diseases that cause retinal dysfunction and varying degrees of visual impairment during infancy or childhood. Retinal dystrophies are known to account for a significant proportion of childhood blindness in all regions of the world,1–3 with recent data from the UK indicating that such disorders account for 14% of all children newly diagnosed as blind each year4 and about 13% of certifications for blindness in children.5
To date there have been no national population-based epidemiological studies of these disorders. Thus there is limited knowledge of the all-cause and disorder-specific incidences and any variations in these by age, sex, ethnicity, socio-economic status and geographical location. Further, knowledge of the proportion of affected children with associated systemic features, their patterns of detection and presentation, prevailing management practices, natural history and broader functional outcomes is lacking. Such data are essential for assessing aetiological trends and the planning and evaluation of health, educational and social services for affected children and their families. In particular, these data are necessary for the planning of trials of novel therapeutic interventions for retinal dystrophies.
From a national population-based study, we report the incidence, additional clinical features and patterns of detection and management of childhood-onset hereditary retinal disease in the UK and how these vary by socio-demographic factors.
Methods
Case definition
Children were eligible for inclusion if they were aged <16 years, living in the UK, and newly diagnosed as having a stationary or progressive hereditary retinal disorder, irrespective of the level of visual function and whether the condition was isolated or part of a systemic disorder. Children with albinism, aniridia, hereditary optic nerve disorders and inherited red-green colour blindness were excluded from the study.
Case ascertainment
National active surveillance was undertaken through two sources. First, we used the British Ophthalmological Surveillance Unit (BOSU),6 the long-established national scheme for epidemiological studies of uncommon ophthalmic disorders in the UK. Second, and independently, we established a national network of other specialists involved in the management of patients with retinal disorders following a postal questionnaire survey. Nineteen of 35 genetics centres in the UK indicated they managed patients with retinal disorders and thus received a monthly report card to either notify cases or confirm they had no cases to report.
Surveillance was undertaken between August 2006 and February 2008, comprising a 1-month ‘lead in’ period and subsequent 18 months of surveillance. Follow-up data collection concluded in November 2008.
Data collection
Standardised data collection questionnaires were developed for the study to collect detailed clinical and socio-demographic information. These were sent to reporting clinicians at case notification and 9 months later. Two reminders were sent to non-responding clinicians.
The British Childhood Hereditary Retinal Disorders Special Interest Group
A clinical research network was established at the outset by inviting individuals with an interest in childhood-onset hereditary retinal disorders to participate and advise at various points in the study. This approach had previously been of benefit in similar studies.4 7
Analysis
All-cause and disorder-specific annual population incidence rates of new diagnosis in the UK for children aged <16 years with an inherited retinal disorder were calculated using the Office for National Statistics (ONS) 2007 mid-year population estimates8 (denominator multiplied by 1.5 to account for 18-month study period). In addition, incidence rates by age, sex, ethnic group, socio-economic status, UK country of residence and Strategic Health Care Authority (SHCA) were calculated. Cumulative incidence and disorder-specific incidence rates of new diagnosis were also calculated. Descriptive analysis was undertaken of the detection and presentation of the child, and their subsequent management.
Results
Two hundred and forty-one eligible patients were reported in the 18-month study ascertainment period. Fifty-nine per cent (n=143) were male.
Co-morbidity
Fourteen per cent of children had additional systemic abnormalities to the ocular condition. Fifteen per cent of all children had developmental delay and 13% had a hearing impairment.
Incidence
Children were diagnosed at different ages throughout childhood, with half diagnosed by the age of 7 years (figure 1).
Age at diagnosis.
All-cause incidence of diagnosis of childhood-onset hereditary retinal disorders in the UK was 1.39/100 000 children per year (table 1). The cumulative incidence (ie, conventionally interpreted as lifetime risk by 16th birthday) was 22.32/100 000 children. Twenty-four distinct diagnoses were reported, retinitis pigmentosa being the most frequent with an annual incidence of 3.01 per million children (table 1). Twenty-four per cent of the children reported had a retinal disorder with an associated systemic problem, for example, Usher syndrome (table 1).
Disorder-specific incidences of childhood-onset hereditary retinal disorders
Four age bands had slightly higher than average incidence of diagnosis, corresponding to clusters of specific disorders (figure 1, table 2).
Most frequent reported disorders at peaks in age of diagnoses
Rates were significantly higher in males than females and in children of South Asian (especially Pakistani) and of Black African (but not Black Caribbean) ethnicity compared with White ethnicity (table 3). The incidence was higher among those in the lowest (worst) quintile of deprivation, with a suggestion of a trend across the quintiles (table 3).
Variation in incidence by age, sex, ethnicity, socio-economic status and geography
The annual incidence was significantly higher in England than in the other countries of the UK (table 3). The SHCA with the highest incidence of new diagnosis was London with all other regions except the East of England having significantly lower rates (table 3).
Mode of inheritance
Fifty per cent of children had an autosomal recessive inheritance pattern, 15% X linked, 7% autosomal dominant, but in 28% it was undetermined at follow-up. Overall 26% of children were reported to have a known family history of a hereditary retinal disorder. Parental consanguinity was positively reported in 23% of families of which 78% were of South Asian ethnicity.
Patterns of detection
As shown in table 4, an ocular problem was first suspected most commonly by the parents in nearly half of all cases (49%). Seven different groups of health professionals were reported as then being the child's first point of contact in the healthcare system, most commonly an optometrist (28%).
Professionals involved in detection and management of children with retinal disorders
Most children presented or first detected because of poor vision or abnormal visual behaviour (58%). However 14% of children were detected through some form of screening examination (table 5).
Reason for presentation/detection of children with retinal disorders
Importantly, we found no significant variations in patterns of detection and presentation by ethnicity, socio-economic status or SHCA.
Management
In addition to ophthalmic professionals and family practitioners, 22 different groups of professionals were involved in the early management (table 4). Those patients with a purely ocular disorder saw on average two non-ophthalmic professionals (range 0–7) and those with an associated systemic disorder saw three (range 0–6).
As part of the diagnostic procedure 88% (range by SHCA 71–100%) of children had electrophysiology assessments, 35% (range by SHCA 18–67%) had molecular genetic investigation and 26% (range by SHCA 10–50%) had some form of imaging. There were no variations in the assessments undertaken by ethnicity or socio-economic status.
At follow-up 34% of children retained sufficiently good vision that they were not eligible to be certified as sight-impaired. However, only 54% of those eligible had been certified, 31% as severely sight-impaired and 23% as sight-impaired.
Nine months after diagnosis, 29% of those children in whom it was considered appropriate had had a low vision assessment; a further 21% had a planned referral. Of school-age children, 76% were in mainstream education and 6% in special schools (placement unknown in 18%); the majority of those in mainstream education were taught using print/sighted methods (82%). By follow-up, a statement of educational needs was in place or in progress in 32% of children, considered unnecessary in 38% and unknown in 30%.
Discussion
Our findings suggest that for a child in the UK, the risk of being diagnosed with a hereditary retinal disorder by the age of 16 years is 22 in 100 000. The incidence is higher among boys, those of South Asian (Pakistani, Bangladeshi or Indian) ethnicity and those from the most socio-economically deprived groups. Retinitis pigmentosa is the most common disorder, with an annual incidence of 3 per million, which is more than 30-fold greater than the frequency of the least common disorders reported: Alstrom syndrome, incontinentia pigmenti, Norrie disease and Oguchi disease. Inheritance is most commonly autosomal recessive or X linked.
A significant proportion of affected children had serious co-morbidity, with 13% having dual vision and hearing impairment. There was a spectrum of visual function, reflecting the heterogeneity of the underlying disorders, but the majority of school-aged children were in mainstream education. Fifty-eight per cent of children presented or were first detected because of poor vision and parents suspected a problem prior to detection by a health professional in only 49% of cases, showing the value of case detection through targeted screening examination of those known to be at high risk. A diverse range of professionals are involved in the management from the earliest stages.
We used national active surveillance through two sources to maximise ascertainment of children with these uncommon disorders. Despite this, we recognise the potential for incomplete or biased ascertainment, which can occur in any study of uncommon disorders using similar methods. We were unable to use capture–recapture analysis to formally assess the level of ascertainment because our two sources were not completely independent. However, the well-established BOSU scheme with its high card return rate (79%),9 the diversity of individual diagnoses and the geographical spread of reported cases, combined with the participation in the study group of specialists within the UK who manage the majority of affected children, suggest that our study sample is sufficiently representative to allow meaningful analysis.
It was beyond the scope of this study to collect long-term data on reported children. Thus 7% of children did not have a definitive diagnosis, as investigation and characterisation were not complete by the study follow-up period 9 months after diagnosis. The length of the ascertainment period (standard for studies through BOSU) may have resulted, through chance alone, in under-representation of the rarest hereditary retinal disorders. As with all similar studies, we report incidence of diagnosis rather than true incidence of disease because we cannot be certain whether some children with hereditary retinal disorders went unidentified during the study period. Thus our estimates for disorder-specific incidence should be regarded as the minimum for each disorder.
Our study is the first population-based report of the incidence and early management of hereditary retinal dystrophies; there is therefore a limited literature for direct comparison of our findings. Most reports of frequency10–13 provide prevalence data on diverse populations derived from differing sources, many of which would not capture the full spectrum of inherited retinal disorders. Using data from the national registers of the visually impaired (acuity worse than 0.5 logarithm of the minimum angle of resolution (LogMAR)) in five Nordic countries, the annual rate of notification of retinal disorders can be estimated to be 0.76/100 000 in those aged 0–17 years.14 The overall annual incidence of disease we report here is considerably higher, but only a third of children in our study had vision that was sufficiently impaired to make them eligible for certification in the UK, which itself has a higher threshold (LogMAR 1.0). Thus a direct comparison is impossible, but our findings would be consistent with a higher rate of childhood-onset hereditary retinal disorders within the UK than Nordic countries. This may be due to differing ethnic composition and levels of consanguinity.15 16
The all-cause incidence of hereditary retinal disorders was greater in boys than girls, which is probably explained by the frequency (15%) of X linked disorders. Variations in incidence by ethnicity, those of South Asian ethnicity having a sevenfold increased rate, are likely to be at least partly attributable to an increased rate of autosomal recessive disease in communities with higher rates of parental consanguinity.3 13 15 The observed gradient of increased incidence in our study, being highest among those of Pakistani origin and lowest in children of Indian ethnicity, echoes the findings of previous studies.3 12 17
Our findings are suggestive of an inverse relationship between socio-economic status and the all-cause incidence of hereditary disorders, and this has not been reported previously. This may partly reflect the association between socio-economic deprivation and ethnicity in the UK. The largest proportion of those living in the least deprived socio-economic quintile in our study population were of White ethnicity (85%), while the largest proportion living in the most deprived quintile were of Asian ethnicity. Nevertheless, further investigation of this relationship is warranted but would require large-scale investigations of individual disorders to allow samples of sufficient size for the necessary multi-variable analysis. Importantly this finding adds to a growing literature showing that blinding eye disease in childhood is more common in children from the most socio-economically deprived backgrounds.18–20
Notably, we did not find ethnicity or socio-economic status to be associated with patterns of detection/presentation, as has been reported for other paediatric disorders.21–24 It may be that our sample size may have prevented true but small variations from being detected in the subgroup analysis. Given the existing literature suggesting that socio-demographic factors influence access to and use of healthcare services, it would of interest to explore this further in future studies, in particular to inform strategies to improve early detection.
The regional and national geographic variations in incidence of diagnosis in our study are likely due to a combination of two factors. First, there are significant differences in ethnic and socio-economic composition of the populations in each geographical region: for example, in London, which had the highest incidence, ethnic ‘minority’ groups comprise 30% of the population (12% Asian) and in the South East Coast region, with the lowest incidence, ethnic minority groups comprise 3.8% of the population (1.7% Asian).25 Second, specialist ophthalmic services for children with hereditary retinal disorders tend to be concentrated in regional medical centres, which may have impacted on geographical variations in incidence of diagnosis by reflecting referral patterns rather than incidence of disease. Significant geographical (by SHCA) variation was also seen in both imagining and molecular diagnostic assessments although little in electrophysiology testing, indicating that the latter is more readily accessible across the country.
The findings of this study characterise the complexity of the management of affected children, in particular the diverse professionals involved from the outset. This highlights the high level of resources that are required for the management of this group of children, most of whom have significant sight loss. It is arguable that these could be most efficiently utilised if affected children were managed by regional level clinical teams that could draw, when required, on multidisciplinary expertise. This could also enhance the experiences of affected children and their families by enabling better liaison, co-ordination and information provision.26
The findings of our study demonstrate the complex heterogeneity of the group of children newly diagnosed each year with hereditary retinal disorders. They serve to highlight the need for studies with sufficiently large and representative populations of children affected with individual disorders to allow the detailed analysis that may enable further elucidation of the role of any environmental factors that may modify risk or severity of these genetic disorders. Our findings provide the data necessary for the planning of specialist clinical services for affected children, to ensure equitable provision by identifying those at highest risk and for the planning of clinical intervention trials in the future.
Acknowledgments
Members of the British Childhood Onset Hereditary Retinal Disorders Network: G. Adams, J. Ainsworth, L. Allen, L. Amaya, S. Anwar, N. Astbury, S. Biswas, G. Black, J. Bradbury, J. Butcher, T. Chan, A. Chandna, B. Church, A. Churchill, M. Clarke, J. Dean, S. Downs, G. Dutton, J. Elston, B. Enoch, A. Evans, N. Evans, B. Fleck, T. Freeguard, A. Glenn, K. Gregory-Evans, C. Gregory-Evans, R. Gregson, R. Hennekam, P. Hodgkins, D. Josifova, N. Kayahi, T. Lavy, J. Leitch, C. Lloyd, V. Long, A. Lotery, R. Lumsden, R. Markham, J. Marr, V. McConnell, A. Morris, D. Morrison, A. Mulvihill, A. Nemeth, B. Newman, K. Nischal, M. Parker, A. Quinn, N. Ragge, A. Reddy, S. Rowe, I. Russell-Eggitt, C. Scott, A. Shafiq, G. Silvestri, I. Simmons, N. Sinclair, T. Sleep, J. Sloper, D. Taylor, R. Taylor, G. Thompson, J. Twomey, M. Votruba, E. Wakeling, C. Williams, D. Williams, H. Willshaw, G. Woodruff, E. Wraith, A. Wright. We thank all the clinicians who reported cases and completed questionnaires.
References
Footnotes
Funding ELH was supported by the Special Trustees of Moorfields Eye Hospital and UK Department of Health's NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust and the UCL Institute of Ophthalmology. Surveillance carried out in association with the British Ophthalmological Surveillance Unit which receives funding from the Guide Dogs for the Blind Association. JSR is a member of the Medical Research Council's (MRC) Centre of Epidemiology for Child Health and Department of Health's NIHR Biomedical Research Centre at Great Ormond Street Hospital and the UCL Institute of Child Health. The funders had no role in the design or conduct of this research.
Competing interests None declared.
Ethics approval Ethics committee approval was obtained from the London Multi-Centre Research Ethics Committee (MREC) and the research followed the tenets of the Declaration of Helsinki.
Provenance and peer review Not commissioned; externally peer reviewed.