This retrospective study was approved by the Institutional Review Board of the University of Miami Miller School of Medicine and included all patients with documented retinochoroidal or optic nerve coloboma seen at the Bascom Palmer Eye Institute from 1976 to 2014. The study was conducted in compliance with the Declaration of Helsinki. Patients were identified by searching all patients for diagnosis of “retinochoroidal coloboma,” “optic nerve coloboma,” or “coloboma” in ultrasound logs. Patients with morning glory syndrome were excluded. Patient charts were reviewed and the data collected included: demographics, location, and extent of coloboma involvement, ultrasound measurement of maximal coloboma width and depth, presence of retrobulbar cysts, and, if present, their size, visual acuity, presence of retinal or choroidal detachment, structural abnormalities (microphthalmia, anophthalmia, microcornea, or any combination), other ocular conditions (amblyopia, glaucoma, cataract, anisometropia, strabismus, nystagmus), and any other nonocular associations (cardiac, ear, skeletal, or urogenital abnormalities, seizures, CHARGE syndrome).
When applicable, the Snellen decimal chart was used and expressed as a logarithm of the minimum angle of resolution (logMAR). Visual fractions were assigned for young children or those with severe visual impairments. Severe visual impairment (SVI) was defined as a visual acuity of 20/200 or worse as per the World Health Organization (WHO) Internal Classification of Diseases (ICD)–10 classification system.
10 Count fingers (CF) was equivalent to 1/200 or logMAR 2.3. Hand motion (HM) was equivalent to 0.5/200 or logMAR 2.6. Light perception (LP) was equivalent to 0.2/200 or logMAR 3.0. No light perception (NLP) was equivalent to 0.02/200 or logMAR 4.0. Eyes that were enucleated or eviscerated were considered to be no light perception. A
P value of <0.05 was statistically significant. The visual acuities of patients less than 2 years of age were included if they were measured by the Snellen system. Two patients less than 2 years old were excluded because their visual acuities were not measured by the Snellen system and an additional 15 patients did not have visual acuities recorded.
The axial length of the eye was measured with a portable ultrasound probe. The axial length was measured to the nearest millimeter with calipers and, when available, these measurements were taken directly from the image report. Care was taken to exclude the coloboma when measuring the axial length so as not to artificially increase this value. Microphthalmia was defined as an eye with axial length less than 2 SD below the mean for that age (axial length < 16 mm at birth, < 19 mm at 1 year).
11 The horizontal corneal diameter was measured to the nearest millimeter with calipers and, when available, these measurements were taken directly from the image report. Microcornea was defined as a cornea with horizontal diameter less than 10 mm.
12 Structural abnormality was a broad term that was defined as one or a combination of the following: microphthalmia, microcornea, or anophthalmia. When present, the maximum height and width of each retrobulbar cyst was measured with calipers from the available archived images (76 images from 1998–2014). Additionally, the maximum depth and maximum width of each coloboma was measured with calipers from the available archived images. The maximum depth of the coloboma was defined as the measurement between the apex of the coloboma and the extrapolated base of the retina. The volume of the coloboma used in the statistical analyses was an average of the coloboma volume calculated as a cylinder (3.147 *
r2 *
h) and as a cone (1/3 * [3.147 *
r2 *
h]). We believe this volume determination model provides a fair assessment of the coloboma given the limited measurement parameters that were retrospectively feasible. Relative coloboma excavation was defined as coloboma depth/axial length. This ratio was intended to normalize the coloboma over different axial lengths. When the distinction was made, the better eye was defined as the eye with the best-corrected visual acuity. In cases where visual acuity in both eyes was identical, the better eye was defined as the eye with the smallest coloboma volume. All measurements were performed by the same examiner. Hand-held calipers were used for printed images, while digital calipers were used for electronic images. An example of the measurements taken is provided in
Figure 1. Because the study consisted of patients seen over a 38-year period, multiple ultrasound machines were used by different technicians to take the photos.
All statistical analyses were performed with SPSS 22.0 (SPSS, Inc., Chicago, IL, USA) statistical package. ANOVA was used to compare categorical variables with logMAR visual acuity. Pearson correlation was used to compare coloboma and retrobulbar cyst measurements with logMAR visual acuity. χ2 test was used to compare categorical variables with the presence or absence of retinal detachment. Student's t-test was used to compare coloboma and retrobulbar cyst measurements with the presence or lack of retinal detachment. Cox regression was used in the analysis of risk factors for retinal detachment.