Age at diagnosis of ocular coloboma in 33 children from Olmsted County, Minnesota, 1968 to 2007.
Nakamura KM, Diehl NN, Mohney BG. Incidence, Ocular Findings, and Systemic Associations of Ocular ColobomaA Population-Based Study. Arch Ophthalmol. 2011;129(1):69-74. doi:10.1001/archophthalmol.2010.320
To describe the incidence, ocular findings, and systemic associations of coloboma in a population-based cohort of children.
We retrospectively reviewed the medical records of pediatric (aged <19 years) patients diagnosed as having ocular coloboma from January 1, 1968, through December 31, 2007, as residents of Olmsted County, Minnesota.
Thirty-three children were newly diagnosed as having ocular coloboma (annual incidence, 2.4 per 100 000 residents <19 years old; prevalence, 1 in 2077 live births). Median patient age at diagnosis was 3.9 months (range, 2 days to 18.4 years), and 22 patients (67%) had unilateral involvement. Twelve patients (36%) had involvement of the anterior segment only, 13 (39%) of the posterior segment only, and 8 (24%) of both. During median ophthalmologic follow-up of 9.2 years (range, 13 days to 35.9 years), 19 patients (58%) had other ocular disorders, including amblyopia in 11 (33%) and strabismus in 10 (30%). During median medical follow-up of 16.8 years, 22 patients (67%) were diagnosed as having a nonocular disorder, including abnormal development in 12 (36%) and CHARGE (coloboma, heart defects, choanal atresia, retarded growth and development, genital abnormalities, and ear anomalies) syndrome in 4 (12%).
Ocular coloboma occurred in 1 in 2077 live births. More than half of the patients were diagnosed as having an ocular disorder other than coloboma, including strabismus and amblyopia in approximately one-third. Two-thirds of patients were diagnosed as having a nonocular disorder, including CHARGE syndrome in 1 in 8 patients.
Ocular coloboma is a rare congenital abnormality caused by defective embryogenesis. Most colobomas result from the failed closure of the embryonic or choroidal fissure during weeks 5 to 7 of fetal life.1 Colobomas may involve any of the ocular structures, including the iris, zonules and ciliary body, choroid, retina, and optic nerve.2 Colobomas are commonly associated with visual loss for which there is little treatment other than amblyopia management.3 Surgical interventions may be performed for cosmetic reasons or for associated complications, such as retinal detachment and cataracts.4- 7 Colobomas may be associated with other ocular disorders, such as microphthalmia,2,8 or with multisystem syndromes, such as CHARGE (coloboma, heart defects, choanal atresia, retarded growth and development, genital abnormalities, and ear anomalies) syndrome,3,9,10 or they may be found in isolation.
Although there are several population-based prevalence studies of coloboma from Europe,11- 14 we are unaware of any studies from the United States. The purpose of this study was to describe the incidence, ocular findings, and systemic conditions observed in a population-based cohort of children younger than 19 years diagnosed as having ocular coloboma in a 40-year period.
The institutional review boards of the Mayo Clinic and Olmsted Medical Group approved this study. The medical records of all patients 19 years or younger who had their condition diagnosed by an ophthalmologist as having any form of ocular coloboma as a resident of Olmstead County between January 1, 1968, and December 31, 2007, were retrospectively reviewed. Coloboma was defined generally in this study as having a notch, gap, hole, or fissure in any of the ocular structures. Potential cases were identified using the resources of the Rochester Epidemiology Project, a medical record linkage system designed to capture data on any patient-physician encounter in Olmsted County.15 The population of Olmstead County is relatively isolated from other urban areas, and almost all medical care is provided to its residents by the Mayo Clinic, Olmsted Medical Group, and their affiliated hospitals.
A diagnostic code search of these 2 institutions generated a list of 100 potential patients, including 25 individuals diagnosed as having unspecified congenital abnormalities that were determined to be noncoloboma disorders on review. Seventeen patients were excluded owing to diagnoses of other, noncoloboma congenital abnormalities, such as morning glory disc abnormality, corectopia, and ectropion uveae. Nine individuals were excluded owing to incorrect diagnoses, and 3 were eliminated as having eyelid coloboma. Of the remaining 46 patients, 12 were later found to be nonresidents of Olmstead County at the time of their diagnosis, and 1 who received a diagnosis before 1968. The remaining 33 patients were included in this study.
Demographic characteristics, including sex and age at diagnosis, were collected for each patient. Data on perinatal, developmental, and medical history and family history of coloboma were also collected. The ophthalmic record was carefully reviewed for the location of the coloboma, any other structural abnormalities, best-corrected visual acuity, ocular alignment, and refractive error. Any association with CHARGE syndrome and the results of genetic testing were also noted. The ophthalmic record of each patient was reviewed for progression of disease, type of management, and final outcome.
To determine the incidence of ocular coloboma in Olmsted County, annual age- and sex-specific incidence rates were constructed using the age- and sex-specific population figures for this county from the US Census Bureau. The 95% confidence intervals were calculated using assumptions based on the Poisson distribution. Because coloboma is a congenital disorder, the birth prevalence was also calculated from the number of births occurring from January 1, 1968, through December 31, 2007, in this county.
Thirty-three new cases of congenital coloboma in children were diagnosed during the 40-year study for an annual incidence of 2.4 (95% confidence interval, 1.6-3.2) per 100 000 patients younger than 19 years, or a live birth prevalence of 1 in 2077. There were 16 girls (48%) and 17 boys (52%). The demographic characteristics of the 33 patients are summarized in Table 1. Twelve of 33 patients (36%) had coloboma in the right eye only, 10 (30%) had coloboma in the left eye only, and 11 (33%) had bilateral involvement. Twelve patients (36%) had involvement of the anterior segment only, 13 (39%) of the posterior segment only, and 8 (24%) both the anterior and posterior segments (Table 2). The median age at diagnosis for the 33 patients was 3.9 months (range, 2 days to 18.4 years) (Figure). The median age at diagnosis by coloboma location was 1.5 months (range, 2 days to 9.4 months) for patients with anterior segment involvement only, 3.9 months (range, 0.9-18.3 months) for patients with both anterior and posterior segment involvement, and 9.1 months (range, 1.5-19.6 years) for patients with posterior segment involvement alone. Eight of 33 patients (24%) had decreased vision at the initial examination, all of which occurred in the eye(s) with coloboma.
Thirty-two of 33 study patients had at least 2 ophthalmic examinations and were observed for a median of 9.2 years (range, 13 days to 35.9 years). The type and number of ocular abnormalities other than coloboma are given in Table 3. Nineteen of the 33 study patients (58%) were diagnosed as having another ocular disorder. Best-corrected visual acuity at final examination was less than 20/60 in 9 of the 28 patients (32%) with sufficient data, including 8 patients with reduced visual acuity in the same eye(s) affected by coloboma and 1 with coloboma unilaterally and bilaterally reduced vision due to retinitis pigmentosa. Eleven of the 33 patients (33%) were diagnosed as having amblyopia: 8 were managed by patching and 3 by observation alone; 6 patients (18%) had amblyopia at the final examination. Ten of the 33 patients (30%) were diagnosed as having strabismus: 3 received strabismus surgery, and the remaining 7 were merely observed; 7 patients (21%) had strabismus at the final examination. Five of the 33 patients (15%) had microphthalmia or anophthalmia, and 3 (9%) had anisometropia; 7 other eye disorders were diagnosed in 5 additional patients (Table 3).
Table 4 provides the percentages of study patients with noncoloboma ocular disorders based on the location of their coloboma. Fifty percent of patients (6 of 12) with iris coloboma alone were diagnosed as having amblyopia compared with 33% of patients (3 of 9) with involvement of both the anterior and posterior segments and 17% of patients (2 of 12) with involvement of the posterior segment alone. None of the 4 study patients with isolated chorioretinal coloboma were diagnosed as having other ocular disorders, whereas 7 of the 9 patients (78%) with coloboma of multiple ocular structures (all of whom had chorioretinal involvement) were diagnosed as having at least 1 other ocular disorder.
The 33 patients were medically observed for a median of 16.8 years (mean, 17.4 years; range, 36 days to 51.3 years), and 22 patients (67%) were diagnosed as having a nonocular disorder. The nonocular medical conditions diagnosed in the 33 study patients are summarized in Table 5, including 12 (36%) with abnormal development, 7 (21%) with heart anomalies, 7 (21%) with ear anomalies, and 6 (18%) with skeletal anomalies. CHARGE syndrome was diagnosed in 4 patients, for a birth prevalence of 1 in 17 000 live births. All 4 patients with CHARGE syndrome were screened for chromosomal abnormalities with negative results; however, 1 of these 4 patients underwent specific testing of the CHD7 gene, and a frameshift mutation (c.1820insA into exon 3) was found.
This population-based study describes the incidence, ocular findings, and systemic conditions associated with coloboma. During the 40-year study, coloboma was diagnosed in 33 children for a birth prevalence of 1 in 2077. More than half of the study patients had other ocular disorders, including strabismus and amblyopia in one-third. Two-thirds of the patients were diagnosed as having a nonocular disorder, including developmental delay in one-third and CHARGE syndrome in 12%.
Population-based prevalence rates for coloboma range from 3.7 per 100 000 from a Hungarian national registry11 to 8 per 100 000 in Scotland.12 However, these rates are a factor of 10-fold less than the 4.8 per 10 000 live births diagnosed among Olmsted County residents. There are several possible explanations for this difference. First, incomplete capture of all patients is likely in the previous studies because they relied on national registry data. The national registry in Scotland is estimated to include less than two-thirds of all cases,12 and a 1-year verification study in Hungary suggests that the Hungarian registry is incomplete.13 Similarly, a Spanish study14 that reported a prevalence of 4.9 per 100 000 live births is based on a national registry limited to congenital malformations diagnosed in the first 3 days of life. Second, diagnostic criteria may vary across time and between countries. The previous studies11,12,14 are based on diagnoses in the past 20 years, whereas the present study includes diagnoses up to 40 years ago. Third, ethnic differences may contribute to different coloboma prevalence rates in different populations. Fourth, we should expect the birth prevalence of coloboma to be higher than the reported prevalence of CHARGE syndrome, at 1 in 8500 to 15 000,16,17 given that patients with CHARGE syndrome are a subset of the coloboma population.
Although some risk factors, such as increased paternal age, have been associated with the development of coloboma,18,19 other factors, including young gestational age and low birth weight, have not.11 The mean paternal age of the present patients was 32.4 years, although we did not determine the mean ages of Olmsted County parents during the 40-year study for comparison. Preterm birth does not seem to be a risk factor for coloboma in this population: 9% of study patients (n = 3) were born prematurely compared with 12% to13% of the general US population.20 Intrauterine exposure to alcohol has been implicated in coloboma development21; however, only 1 patient in this study had a documented history of fetal alcohol exposure. A positive family history of coloboma was reported in only 2 patients (6%), suggesting that genetic factors may be significant in a few patients. However, additional study patients may have family members with undiagnosed coloboma. We found no clear sex predominance for ocular coloboma. Finally, although coloboma is a congenital disorder as evidenced by the generally early age at diagnosis (Figure), a small proportion of patients seem to be diagnosed around the first decade of life, presumably after failed visual screening.
Coloboma occurred bilaterally in approximately one-third of the study patients. This finding is less than the reported percentages of patients with bilateral coloboma involvement in Scotland (42%) and Hungary (47.5%).12,13 This difference may be due to a reporting or referral bias favoring bilateral diagnoses. As described previously, incomplete capture is likely in the previous studies owing to their dependence on national registry data.12,13 Unilateral intraocular cases may be more likely to go unnoticed by patients, parents, and physicians.
In this study, 36% of patients had a coloboma of the anterior segment (iris) only, 39% of the posterior segment only, and 24% of both the anterior and posterior segments. Bermejo and Martínez-Frías14 reported 55 cases of coloboma, of which 47% involved the iris, 40% were chorioretinal, and 13% were unspecified. The higher prevalence of iris coloboma in that study may be explained by their 3-day diagnosis window compared with a mean diagnosis age of 4 months in the present study. According to the present findings, colobomas of the posterior segment are more likely than those of the anterior segment to be diagnosed at an older age.
Nineteen of the 33 patients (58%) in this study were diagnosed as having another ocular disorder. The association of coloboma with other ocular disorders has been reported,2 but there are few data on the prevalence of specific ocular disorders in the coloboma population. In this study, 33% of patients were diagnosed as having amblyopia, and 30% were diagnosed as having strabismus. These percentages are significantly greater than the estimated prevalence rates of amblyopia and strabismus in the general population: 1.6% to 3.5% and 4% to 6%, respectively.22,23 Retinal detachment was a rare complication in patients with coloboma in this study, affecting only 6%. This is consistent with more recent studies24 and is less than early estimates of 20% to 40%,25,26 which were likely elevated because of referral bias.
It is difficult to derive conclusions about the prevalence of specific ocular diagnoses in patients with coloboma based on such a small sample. However, amblyopia was diagnosed in 50% of study patients with an isolated iris coloboma compared with 17% of patients with a strictly posterior lesion (Table 4). In addition, none of the patients with an isolated chorioretinal coloboma was diagnosed as having other ocular disorders, whereas 78% of patients with multilevel coloboma were diagnosed as having at least 1 other ocular disorder (Table 4).
The high comorbidity of coloboma and extraocular disorders in this study is consistent with previous studies.1,24,27,28 However, it is difficult to determine the prevalences of specific systemic abnormalities in patients with coloboma given its relatively rare occurrence. In a study of 48 consecutive patients with chorioretinal coloboma, Daufenbach et al24 reported systemic abnormalities in 38% of patients, including 6 (13%) with abnormal development. We report a higher percentage of patients (67% [22 of 33]) with at least 1 nonocular diagnosis and a higher percentage (36% [12 of 33]) with abnormal development. Several reasons may account for this difference. First, the mean follow-up age in the study by Daufenbach et al24 is 6.1 years compared with 17.4 years for the present study. In addition, Daufenbach et al included only patients with chorioretinal and multilevel colobomas, whereas the present study included all types of ocular coloboma diagnosed during the 40-year period.
The reported birth prevalence of CHARGE syndrome of approximately 1 in 17 000 in this cohort is lower than the 1 in 8500 to 15 000 from previous studies.16,17 This difference may be due to several reasons. First, because of the relatively rare occurrence of CHARGE syndrome in this study, adding or subtracting a single study patient would significantly alter the prevalence. Second, an accurate diagnosis of CHARGE syndrome is difficult because there is considerable phenotypic variability, with no single defining clinical feature.29 Moreover, patients in this study with several features of CHARGE syndrome but who did not achieve the diagnosis of CHARGE syndrome further illustrate the difficulty of accurately diagnosing this disorder. According to the criteria established by Blake et al,16 an individual must exhibit 4 major criteria or 3 major and 3 minor criteria to be diagnosed as having CHARGE syndrome. In addition to the 4 patients who were clinically diagnosed as having CHARGE syndrome, 1 patient exhibited 2 major and 3 minor criteria, and another displayed 1 major and 2 minor criteria.
There are several limitations to the findings in this study. Its retrospective nature is limited by nonstandardized data and incomplete follow-up. Colobomas are often asymptomatic, particularly intraocular lesions that do not affect visual acuity, and may go unnoticed by the patient or the physician. Although most patients in Olmsted County are managed by the 2 medical systems in the community, some residents may have sought care outside of Olmsted County, thereby underestimating the true incidence in this population. In addition, the small number of cases of coloboma in this study population makes characterizing the incidence of coloboma subtypes and associated ocular or systemic conditions difficult. Finally, the ability to generalize these findings nationally and internationally is limited by the demographics of Olmsted County, a relatively homogeneous semi-urban white population.
This study provides population-based data on ocular coloboma diagnosed during a 40-year period. Ocular coloboma occurred in approximately 1 in 2077 live births, a rate that is higher than that of all previous studies. More than half of the study patients were diagnosed as having an ocular disorder other than coloboma, including strabismus or reduced vision in approximately one-third. Two-thirds of the patients were diagnosed as having a nonocular disorder, including CHARGE syndrome in 1 in 8 study patients.
Correspondence: Brian G. Mohney, MD, Department of Ophthalmology, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (firstname.lastname@example.org).
Submitted for Publication: December 19, 2009; final revision received April 3, 2010; accepted April 5, 2010.
Financial Disclosure: None reported.
Funding/Support: This study was supported in part by the Rochester Epidemiology Project (grant R01-AR30582 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases) and by an unrestricted grant from Research to Prevent Blindness, Inc.