Article Text
PDF
Abstract
Aim To identify the clinical features of optic nerve hypoplasia (ONH) and prevalence within a population of New Zealand children with severe visual impairment.
Methods Retrospective review of medical records of children with severe visual impairment registered with Blind and Low Vision Educational Network New Zealand.
Results Of 1500 children with severe visual impairment, 94 (6.3%) exhibited ONH, and 91 (97%) cases were bilateral. Of these 94 cases, 52 (55%) were males and ethnicities were European Caucasian (52%), Maori (40%), Pasifika (6%) and other (2%). Most children with ONH had poor vision, with 60% having ≤6/60 Snellen visual acuity equivalent. The median maternal age was 20.0 years old with 52% ≤20 years. There was a statistically significant over-representation of Maori ethnicity (40%) and young maternal age with age less than 20 years old (44%) in our cohort compared to the general population (14.6% and 7.4%, respectively; p<0.0001). Half had hypopituitarism, while neuroimaging abnormalities were detected in 60% cases. Cerebral neuroradiographic abnormalities were found to be associated with higher rate of developmental delay (OR 9.764 95% CI 3.246 to 29.373).
Conclusions This is the first major study of visual impairment in New Zealand children, and it demonstrates that ONH is an important cause of severe visual disability; with an over-representation of Maori children and younger maternal age.
- Optic Nerve
- Imaging
Statistics from Altmetric.com
Introduction
Optic nerve hypoplasia (ONH) is one of the major contributors worldwide to childhood blindness, occurring in approximately 1 in 10 000 children less than age 14 years.1 The incidence of ONH is estimated to be 7.2 in Sweden2 and 10.9 in England1 per 100 000. It overtook retinopathy of prematurity in Sweden as the leading cause of childhood blindness in 1997.3 ONH is a congenital, non-progressive, disease of unknown aetiology—though younger maternal age, primiparity and maternal recreational drug intake have been indicated as possible risk factors. Unfortunately, there is no effective treatment for ONH to date.4
ONH is characterised clinically by a morphologically small optic disc (figure 1), with or without a double ring sign and/or presence of retinal vascular tortuosity and/or abnormally straight retinal vessels.5 ONH usually involves optic nerves bilaterally (75%) though often asymmetrically, but may also occur unilaterally.3 ,6 ,7 Clinical expression is highly variable, ranging from a child with normal development and near-normal vision to a child with severe visual impairment with or without associated systemic disorders, such as endocrinological and neurological abnormalities.3 ,7
Optic nerve hypoplasia demonstrated on a clinical fundus photograph with a ratio of horizontal disc diameter (DD) and the distance between the temporal margin of the disc and macula (DM) of less than 1:3.
There were fewer than 20 cases of ONH in the scientific literature prior to the 1970s, but increased reporting has occurred since the first reported association with pituitary gland deficiency by Hoyt et al.8 However, controversy still remains in respect to the relationship between ONH and its association with midline structure anomalies, endocrinopathy and developmental delay. Recent reviews have suggested that ‘Septo-optic dysplasia’ (SOD), defined historically by de Morsier as the association of ONH with absence of septum pellucidum, is a misnomer as numerous studies had found that (1) the absence of the septum pellucidum imparts no increased risk for conditions associated with ONH and (2) most neuroradiographic abnormalities associated with ONH are not midline.9 ,10 SOD is a phenotypically variable disease that poses significant diagnostic challenges.4 The difficulty in the management of SOD is the various associations with other systemic disorders and the large clinical spectrum of the disease—with no definite predictor as yet available to identify high-risk patients. This problem contributes further to the burden on a child with severe visual impairment.
This study attempts to estimate the prevalence of ONH in a population of New Zealand children with severe visual impairment, to determine the clinical characteristics including ocular features of this condition in our population and to provide more information about this relatively uncommon yet debilitating condition.
Methods
A retrospective review was conducted of the comprehensive clinical medical records of children with severe visual impairment registered with the Blind and Low Vision Educational Network New Zealand (BLENNZ) between 2010 and 2012. BLENNZ is a national organisation responsible for the allocation of educational and developmental support for all visually impaired children in New Zealand. A complete registry is ensured through governmental mandated compulsory referrals by health providers and educational services for children with severe visual impairment. Referral criteria include best corrected vision of 6/18 or less in both eyes with or without other systemic disorders. BLENNZ follows a multidisciplinary approach consisting of a team of paediatric ophthalmologist, paediatrician, optometrist, special educational teachers, physiotherapists and other health professionals—with the aim to help learners who are blind, and those with low vision to access and participate in the regular curriculum, as well as develop independence skills.11 This study was part of a larger overall project involving the authors of data transfer and consolidation from medical records of BLENNZ into a live database. The ophthalmic diagnosis for each case was verified contemporaneously by a paediatric ophthalmologist.
The demographic and clinical data collected included age at diagnosis, gender, ethnicity (self-identified), neuroimaging and endocrinological abnormalities, visual acuity in Snellen decimal and developmental milestones. The visual acuity was recorded in each eye, but the reported vision in this study was the vision of the better-seeing eye. Individuals with acuity recorded as counting fingers or less were assigned as acuity <6/60 for the purpose of data analysis. The diagnosis of ONH was based on ophthalmoscopic examinations by at least two independent ophthalmologists including the referral ophthalmologist and BLENNZ paediatric ophthalmologist (SD). A few cases of ONH were initially diagnosed based on MRI scan findings of a smaller optic nerve as part of the systemic work-up for children with failure to thrive; all of which were subsequently confirmed by ophthalmoscopic examinations. The presence of developmental delay was diagnosed by local paediatricians based on their expert observations using the expected normal developmental milestones that encompass motor (gross and fine) skills, language, mental and psychosocial status. The vast majority of those cases were also confirmed by a developmental paediatrician who has more than 20 years’ experience working at BLENNZ with children suffering from severe visual disabilities. Maternity data collected included age at pregnancy and parity, time and type of delivery as well as the drug and/or alcohol intake and smoking status.
For the purpose of this study, SOD is defined as ONH with midline radiographic abnormalities and/or hypopituitarism, while isolated ONH is defined as ONH without systemic or neuroradiographic abnormalities.12
Statistical analysis was performed using software SPSS V.17 for Windows (Chicago, Illinois, USA) and SAS (North Carolina, USA). Basic descriptive statistics were calculated on all data gathered and are reported as mean±SD or median±IQR or percentage (%). Fisher's exact tests were applied as appropriate for testing associations between categorical variables. Most recent census data was obtained directly or indirectly via Infoshare software from the Statistics New Zealand at (http://www.stats.govt.nz), and was used for comparison with the study cohort.
This study was approved by the Northern Regional Ethics Committee for health research in New Zealand and adhered to the tenets of the declaration of Helsinki.
Results
Ninety-four children, less than 16 years of age, with ONH were identified from 1500 subjects enrolled in the BLENNZ registry. Other common diagnoses apart from ONH (6.3%) include cortical visual impairment (30.1%), optic nerve atrophy (16.3%), albinism (13.2%), various retinal dystrophies (11.2%), retinopathy of prematurity (6.4%), visual impairment from non-accidental injuries (3.6%) and other causes (12.9%). Median age at diagnosis was 5.0 months (IQR 3.0–8.0). Fifty-two children (55%) were male and forty-two (45%) female. Ethnicity included: European Caucasian 43 (52%), Maori 33 (40%), Pasifika 5 (6%) and other 2 (2%).
Bilateral optic nerve involvement occurred in 91 (97%) cases. Visual acuity data were available for 78 (83.0%) patients with a mean and median Snellen decimal value of 0.18 (Snellen equivalent 6/30–6/36; SD±0.305, range 0.0013–2.00) and 0.09 (Snellen equivalent 6/60–6/76; IQR 0.0016–0.25), respectively. The majority of children had poor vision (60% ≤6/60 best corrected Snellen equivalent). The detailed breakdown of the binocular visual acuities was: 50.0% with <6/60, 38.5% with 6/60 or better, but <6/12 and 11.6% with ≥6/12 (2 of 9 had unilateral involvement). Nystagmus and roving eye movements were observed in 74 (78.7%) patients; 47.5% and 31.2% respectively. Approximately two-thirds (63.6%) had strabismus; 39.8% had esotropia while 23.9% had exotropia.
Endocrinological work-up included fasting cortisol (an indirect measure of adenocorticotrophic-releasing hormone (ACTH)), thyroid stimulating hormone (TSH), free T4 and antidiuretic hormone (ADH) for all age groups, while gonadotrophin releasing hormone (GnRH) was only included for those who had reached puberty. In our cohort, 82 (87.2%) children had the complete work-up based on clinical indications as determined by their primary physician. Among those with endocrinopathy (50%), GH (39.5%) was the most commonly affected pituitary gland hormone, followed by ACTH (35.8%), TSH (33.3%), ADH (14.8%), and GnRH (8.6%). Developmental delay was recorded in 51 (60%) patients.
Seventy-eight (83%) patients had at least one form of neuroimaging. Forty-one (53%) patients had MRI while the rest had CT scan. As illustrated in table 1, of those who had neuroimaging, 47 (60.3%) had abnormalities which included 27 (34.6%) and 13 (16.7%), respectively, with absent or hypoplastic septum pellucidum or corpus callosum. Fourteen (17.9%) exhibited pituitary gland abnormalities, and 18 (23.1%) some form of hemispheric and/or ventricular abnormality. Table 1 also demonstrates the different neuroimaging abnormalities and their association with endocrinopathy and developmental delay. ONH in the presence of at least one or more neuroimaging abnormalities was found to be associated with high rate of developmental delay (OR 9.764 95% CI 3.246 to 29.373). The absence of the septum pellucidum (OR 2.940 95% CI 1.001 to 8.631) but not hypoplastic corpus callosum (OR 3.929 95% CI 0.800 to 19.297), was significantly associated with developmental delay; 19 of the 27 patients (70.3%) with absent septum pellucidum were found to have other malformations, such as schizencephaly, pituitary gland abnormality, absent corpus callosum, abnormalities of their ventricles and other cerebral abnormalities. As expected, pituitary hypoplasia or aplasia or ectopia was found to be associated with pituitary deficiency (OR 6.519 95% CI 1.328 to 32.007). However, pituitary hormone deficiency at diagnosis (OR 2.132 95% CI 0.835 to 5.441) was not associated with developmental delay. In the presence of neuroimaging abnormalities, hypopituitarism and developmental delay were more likely to occur—albeit only significantly in the latter (OR 2.455 95% CI 0.929 to 6.483 and OR 9.764 95% CI 3.246 to 29.373, respectively).
The different neuroimaging abnormalities in patients with ONH, and their association with endocrinopathy and developmental delay
The mean maternal age was 22.38 years (SD±6.04, range 15–40), while the median was 20.0 (IQR 19–24.5); 52% were found to be 20 years of age or younger. Forty-eight (65.8%) were primiparous; of which 53.3% were Maori, 44.4% Caucasian and 2.2% Asian with a median age of 19.0, 19.0 and 23.0 years old, respectively. Fifty (77.0%) had undergone spontaneous vaginal delivery, 7 (10.8%) were induced while 15 (23.1%) had caesarean section, and only three of these were elective. The maternal history included 14.5% (8/55) and 15.4% (10/65) of alcohol and/or recreational drug use and smoking, respectively, during the time of pregnancy. There was a statistically significant over-representation of Maori ethnicity (40%), and young maternal age with age less than 20 years (44%) in our cohort compared with the general population (14.6% and 7.4%, respectively; p<0.0001).
Discussion
Young maternal age and/or primiparity are the two known risk factors for the development of ONH.13 ,14 Indeed, ONH typically occurs more often in the first-born child to mothers aged 13–19 years.8 ,15 Almost half (44%) our cohort had a maternal age of less than 20 years compared to only 7.4% of mothers at the time of birth in the general New Zealand population (p<0.0001). Magalith et al and Garcia–Fillon et al in two independent studies have also highlighted that the maternal population is significantly younger (p<0.01 and p<0.001, respectively) when compared with the general population.16 ,17 The latter study also found that primiparity was significantly higher (p<0.001) than the general population.17 Although our study highlighted a majority with primiparity (65.8%), there are no available general population data on the prevalence of primiparity for comparison. Other possible risk factors include smoking,13 alcohol and recreational drug use 17 as well as caesarean section (OR=2.55 95% CI 1.63 to 3.99) and instrumental vaginal delivery (OR=1.00; 95% CI 0.43 to 2.34).12 We found no association between these postulated risk factors and ONH in our study cohort.
Unique to New Zealand, we identified an over-representation of (indigenous) Maori ethnicity (40%) in children with ONH compared with the general population (14.6%) (p<0.0001). Notably, the median maternal age for those of Maori ethnicity in our study cohort was found to be younger at 19.0 years than those of Caucasian (19.5), Pasifika (22.0) and Asian (23.0) ethnicities. Similar observations were demonstrated in the New Zealand 2006 Census with a median maternal age of 4.4 years younger in the Maori population compared with the general population (25.9 and 30.3, respectively). This observation leads us to conjecture that the younger maternal age among Maori women is related to the increased prevalence of ONH in the New Zealand Maori population. (http://www.stats.govt.nz/infoshare/ViewTable.aspx?pxID=b4f7bb75-23a1-4271-b666-89d6f891fd4f).
Hypopituitarism may be present at birth or develop later in patients' lives; therefore, it is vital for lifelong monitoring in these children. Signs or symptoms, such as recurrent and unexplained hypoglycaemia, jaundice, seizures, respiratory distress or electrolyte disturbances should prompt a high index of suspicion for possible hypopituitarism, and early full hormonal investigation should be instigated.18 This requirement was further reinforced when Brodsky et al reported sudden death in five children with SOD and corticotrophin deficiency.19 As in our study cohort, most studies concur that the presence of pituitary gland anomaly is predictive of hypopituitarism (OR 6.519 95% CI 1.328 to 32.007) but its absence does not necessarily exclude hypopituitarism.15 ,18–21 Contrary to traditional belief, absence of the septum pellucidum has not been associated with an increase in the risk of hypopituitarism in children with ONH6 ,22—an observation also confirmed in our cohort (p=0.600)
Half our study cohort had endocrinological dysfunction, fewer than reported (72–80%) in previous studies.10 ,21 However, the prevalence of GH (70%), TSH (46%), ACTH (27%) and ADH (5%) deficiencies, occurring in isolation or in combination, in previous studies was comparable to our findings (40%, 33%, 36% and 5%, respectively).3 ,10 Although we report a lower prevalence of hypopituitarism compared to other studies, since pituitary gland deficiency may evolve, this may increase over time to yet approximate previous studies.
Roughly 12–16% of the children in the USA and Australia have at least one form of developmental delay;23 ,24 much lower than the prevalence (71%) associated with ONH. 10 Garcia–Fillon et al, in a prospective study of 73 ONH patients, highlighted that hypoplasia of the corpus callosum and hypopituitarism are risk factors for developmental delay—but not the absence of septum pellucidum.6 ,10 ,20 By contrast, we found that the absence of the septum pellucidum was significantly associated with developmental delay (OR 2.940 95% CI 1.001 to 8.631) but not hypoplasia of the corpus callosum (OR 3.929 95% CI 0.800 to 19.297). This unexpected finding may be explained by (1) the relatively large number of patients who had CT scans (48%), a method known to have a poorer sensitivity (compared to MRI scans) in diagnosing corpus callosum hypoplasia and/or (2) the large proportion of patients with absent septum pellucidum were found to have other malformations (70.3%) suggesting that the absent septum pellucidum may be just a surrogate association.
Pituitary hormone deficiency at diagnosis (OR 2.132 95% CI 0.835 to 5.441) was not found to be associated with developmental delay. This could arguably be due to the early diagnosis of various hormonal deficiencies and commencement of appropriate hormone replacement therapy, which may be a result of increased awareness and understanding of the associations of ONH among paediatricians. Borchert and Garcia reported that most children with ONH had hypothalamic dysfunction and/or neurodevelopmental impairment regardless of the MRI findings or severity of ONH.3 ,7 Interestingly, Garcia et al demonstrated that patients with SOD had a lower rate of developmental delay (p=0.041) but a higher rate of endocrinological abnormality (p<0.001).7 In the presence of neuroimaging abnormalities, the current study demonstrated that hypopituitarism and developmental delay were more likely to occur—albeit only significant in the latter (OR 2.455 95% CI 0.929 to 6.483 and OR 9.764 95% CI 3.246 to 29.373, respectively). Although there is significant variability between the neuroimaging abnormalities and systemic associations between reported studies, these correlations are statistically significant within each study. This may be due to the disparity of the patient profile between studies, and some of the possible contributing factors to such variable results include: (1) different genetic composition (2) ethnicity and (3) the influence of the geographic, environmental and/or socioeconomic factors. Similarly, a direct comparison of developmental delay risks in ONH patients across studies may not be accurate due to the different definitions used, in addition to the retrospective nature of most of these studies.
Bilateral ONH is more likely to be associated with neurologic abnormalities such as hypothalamic/pituitary dysfunction (57%).7 ,15 and developmental delay (40%) 7 ,10 compared to unilateral ONH (32% and 8%, respectively). Notably, the current study cohort had fewer children with unilateral ONH (3.2%) compared to previous studies (75%).3 ,6 ,7 This is because BLENNZ typically accepts referral of children with Snellen equivalent of 6/18 or less in the better eye. This is likely to have skewed our data towards a higher incidence of bilateral ONH in comparison with unilateral ONH. The three unilateral cases were included in our registry due to pituitary hormone deficiency and/or developmental delay. Since unilateral ONH can be more challenging to diagnose due to late presentation, accurate diagnosis remains imperative because of the possible associations with endocrinopathy and developmental delay. Other investigative tools, such as visual-evoked potential (usually absent or delayed in ONH) may be vital in confirming the diagnosis in suspected cases.
Visual impairment is generally believed to be severe in children with ONH though it can vary from relatively normal vision to no light perception.25 Although in the current study, visual acuity data was available for 83% of the subjects, the authors found inconsistency in the quality of the visual acuity measurement. This is due to the retrospective nature of the study and the challenge in obtaining accurate visual acuity measurement in preverbal children as well as those with moderate-to-severe cognitive disabilities. As with many retrospective reviews of registries studies, caution must be used in extrapolating our findings.
Conclusion
This is the first study reporting New Zealand children with severe visual impairment, and it demonstrates that ONH is a major cause of severe visual impairment in New Zealand children. Similar to most studies, our study demonstrated young maternal age as a significant risk factor for the development of ONH, and of interesting note, unique to our country, there was a higher representation of ONH among Maori children. Pituitary hormone deficiency and developmental delay were common among children with OHN, and full paediatric assessment and endocrine testing are recommended for children with ONH.
References
Footnotes
-
Contributors All authors are given credit for (1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; and (3) final approval of the version to be published.
-
Competing interests None.
-
Ethics approval Northern Regional Ethics Committee.
-
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- At a glance