Aim To evaluate optic nerve head, retinal nerve fibre layer (RNFL) and ganglion cell complex (GCC) thickness measurements in children operated for primary congenital glaucoma (PCG) using spectral domain optical coherence tomography (SDOCT).
Methods In a case-control study, 45 eyes of 37 children operated for PCG and 72 eyes of 41 normal children underwent optic nerve head, RNFL and GCC imaging with SDOCT. Differences in SDOCT parameters between PCG and control group, correlation between SDOCT parameters and a range of clinical variables, namely preop corneal diameter, intraocular pressure, degree of corneal oedema and age at which surgery was performed in PCG eyes, were evaluated.
Results Mean (±SD) age of children in PCG group was 10.1±3.6 years and control group was 13.6±3.2 years (p<0.001) at the time of SDOCT imaging. Visual fields, whenever possible were unreliable in 20 of 23 PCG and 30 of 46 normal eyes. All SDOCT parameters were significantly different (p<0.001) in PCG compared with control group. All global SDOCT parameters (rim area, average RNFL and GCC thickness) correlated significantly with the clinical cup to disc ratio measurements (correlation coefficients better than −0.70) in children with PCG. Age at which surgery was performed was inversely related to SDOCT parameter thickness but was not statistically significant.
Conclusions All SDOCT parameters were significantly different in children operated for PCG compared with normal children. Future research should evaluate the test-retest variability of SDOCT and its ability to diagnose progression in children unable to perform reliable visual field tests.
Statistics from Altmetric.com
Primary congenital glaucoma (PCG) is a rare inherited disorder that accounts for 0.01–0.04% of all blindness.1 The incidence of the condition is higher in a few geographical areas of the world. PCG accounted for 4.2% of all childhood blindness in a population-based study conducted in Southern India.2 PCG has a male preponderance and is bilateral in 70% of children. The condition is generally noticeable at birth or the latest by 3 years of age. Nearly 80% of PCG is detected by the 1st year of life. Diagnosis is made clinically. Surgery is the primary mode of intervention.3
Spectral domain optical coherence tomography (SDOCT) is a non-contact technique used to image ocular tissues at near microscopic resolution (5 µm).4 ,5 SDOCT has been shown to be useful in detecting the structural changes that happen in the optic nerve head (ONH), retinal nerve fibre layer (RNFL) and the macula in adult glaucoma.6 ,7 However, there are no studies on the utility of SDOCT in children with PCG. The purpose of our study was to evaluate the utility of SDOCT to detect the structural changes in the ONH, the RNFL and the macula that occur in children with PCG.
This was a case-control study conducted at a tertiary eye care centre in South India between July 2010 and July 2011. The institutional review board of the LV Prasad Eye Institute approved the study and all methods adhered to the tenets of the Declaration of Helsinki for research involving human subjects. Informed consent was obtained from the parent or the guardian of all the children involved in the study.
Children in the age group of 5–18 years, who had been diagnosed with and had undergone surgery for PCG in the past, by a single surgeon (AKM), were recruited for the study. The examination of children with PCG was performed under general anaesthesia using the operating microscope. Examination procedure in a child with PCG has been detailed earlier.3 Clinical features suggestive of PCG included increased intraocular pressure (IOP) of ≥21 mm Hg, increased horizontal corneal diameter (≥11 mm within the first year of life) and the presence of corneal oedema with Haab's striae.1 IOP was checked immediately after induction using Perkin's hand held applanation tonometer. Horizontal corneal diameter was measured with callipers. A direct and/or indirect ophthalmoscope with a +20D lens was used to examine ONH changes in eyes with clearer media. Children with thus confirmed PCG underwent primary combined trabeculotomy and trabeculectomy in the same sitting. The procedure of surgery has been described earlier.8–11
Children in the age group of 5–18 years, who were brought to the paediatric clinic by their parents or guardians for a general check-up or for refractive error correction, were recruited as the control group. Inclusion criteria for the control group children were an age between 5 years and 18 years, normal anterior segment examination, spherical refraction within ±6 D and astigmatism within ±3 D, IOP <20 mm Hg on office examination under topical anaethesia with 0.5% paracain drops (Paracain, Sunways (I) Pvt Ltd Mumbai, India) and normal posterior segment examination with normal ONH appearance. All the examinations in the control group were performed by two trained observers (SS, UKA). A single examiner (AKM) performed posterior segment examination. Children who were mentally challenged, or had inability to fixate were excluded from the study.
All normal children underwent a comprehensive ophthalmic examination including review of medical history, visual acuity testing, slit-lamp biomicroscopy, IOP measurement using Goldmann applanation tonometry and dilated funduscopic examination with a 90-D lens. Visual fields were assessed using Humphrey Visual Field Analyser (Carl Zeiss Meditec, Inc. Dublin, California, USA) where possible. Visual fields were considered unreliable if the fixation losses were over 20% or false positive or false negative responses were over 33%. SDOCT examination was performed using a table top RTVue SDOCT (Model RT-100, Optovue Inc, V.3.0, Fremont, California, USA).
SDOCT detects interferometric signals as a function of optical frequency and provides 26 000 A-scans per second, with a depth resolution of 5 μm and a transverse resolution of 15 μm as compared with 400 A-scans of time-domain OCT. The scan protocols used for imaging were ONH, 3D optic disc and ganglion cell complex (GCC) protocols.
The scan protocols have been described elsewhere.6 The ONH scan is a combination of circular scans for RNFL thickness analysis and radial scans for ONH shape analysis, both centred on the ONH. Thirteen circles of diameters 1.3–4.9 mm are used to create peripapillary nerve fibre layer thickness map. Twelve radial lines 3.7 mm in length are used to calculate the ONH shape parameters.
3D optic disc
The three dimensional (3 D) disc protocol uses 101 horizontal lines to cover a 5 mmE×5 mm square region and provides a total of 51 813 A-scans in 2.2 s. The software algorithms automatically detect the edges of retinal pigment epithelium or Bruch's membrane at the disc margin. However, the operator can modify the data points if necessary.
The GCC is a composite of the inner plexiform layer, the ganglion cell layer and the RNFL. It consists of 15 vertical lines and one horizontal line covering a 7 mm×7 mm region, centred at 1 mm temporal to the fovea. It captures 15 000 data points in 0.6 s.
Only well-centred scans with signal strength index better than 35 were considered for analysis. If the desired signal strength index was not reached, the particular scan protocol was repeated. One acceptable image with highest signal strength index for each scan type was chosen for the analysis.
Inclusion was based on eyes, and both eyes of children, if eligible were considered for the analysis. Generalised estimating equations were used to adjust for the correlations between the two eyes of the same child during estimation as well as comparison of parameters between the control and the case groups. Bivariate correlations were assessed between OCT parameters and a range of clinical variables, namely, preoperative corneal diameter, IOP, degree of corneal oedema and age at which surgery was performed. Statistical analyses were performed using commercial software (Stata V.11.0; StataCorp, College Station, Texas, USA). A p value of ≤0.05 was considered statistically significant.
After excluding four eyes with poor fixation, two eyes with low signal strength index on OCT and seven eyes with high refractive error, belonging to one of the two groups, there were 45 eyes of 37 children (20 boys and 17 girls) operated for PCG and 72 eyes of 41 normal children (22 boys and 19 girls) eligible for the study and analysis. There was one child with PCG who was premature at birth (by 16 days). All children with PCG were diagnosed and had undergone primary combined trabeculotomy and trabeculectomy during the first year of life. At the time of diagnosis, horizontal corneal diameter in these children was 13.3±1.1 mm (mean±SD, range: 11.0–15.5 mm) and the mean IOP was 30.2±5.9 mm Hg (range, 21–44 mm Hg). The cornea showed severe oedema in 21, moderate oedema in six and mild oedema in two eyes at the time of diagnosis. The remaining 16 eyes with PCG had no corneal oedema. Corneal oedema cleared completely after surgery. Visual function could not be assessed before surgery.
At the time of SDOCT examination, children with PCG were significantly (p<0.001) younger (10.1±3.6 years) than normal children (13.6±3.2 years). The best-corrected visual acuity in children with PCG ranged from 6/6 to counting fingers at 50 cms and that in normal children ranged from 6/6 to 6/15. Median cup to disc ratio measured clinically at the SDOCT scanning visit in children with PCG was 0.6 (range, 0.2–1.0). In the PCG group, the mean spherical error was −1.68±2.7 D, and the mean cylindrical error was −1.54±1.25 D. In the normal group, the mean spherical error was −0.65±1.5 D and the mean cylindrical error was −0.86±1.0 D. There was a statistically significant difference (p=0.015) between the mean spherical equivalent in children with PCG (mean±SD, −2.46±2.7 D) and normal children (−0.93±1.6 D). Children with PCG were significantly more myopic compared with normal children. Visual field responses were found to be unreliable in 20 of 23 PCG and 30 of 46 normal eyes where field examination was possible.
Mean values for ONH parameters in PCG and control eyes are summarised in Otable 1. Total rim area, temporal, superior, nasal and inferior rim area, rim volume and nerve head volume were significantly lower in PCG eyes as compared with normal eyes (p<0.001). Cup area and the cup volume were significantly higher in the PCG eyes. Disc area was similar in the two groups (p=0.71).
Mean values of RNFL thickness parameters in PCG and control eyes are summarised in Otable 2. The average thickness, temporal, nasal, superior and inferior quadrant RNFL thicknesses were significantly lower in PCG eyes as compared with normal eyes (p<0.001).
Mean values of GCC parameters in PCG and control eyes are summarised in Otable 3. The average, superior and inferior GCC thicknesses were significantly lower in PCG eyes.
All the global parameters of SDOCT (rim area, average RNFL thickness and average GCC thickness) correlated significantly with the cup to disc ratio (correlation coefficients better than −0.70 and p<0.05 for all the parameters) in children with PCG. All OCT parameters were thinner in children operated later in life but the differences were not statistically significant. No other clinical variables were significantly related to any of the OCT parameters. There were six eyes of four subjects that required antiglaucoma medication even after surgery. However, the OCT parameters in these eyes were not different from the eyes which did not require antiglaucoma drops after surgery. In addition, we also analysed the differences in OCT parameters in children with unilateral disease in comparison with those with bilateral disease. Children with unilateral disease had thinner ONH rims, RNFL and GCC thickness parameters compared with children who had bilateral disease (p<0.05 for all parameters).
In our study to evaluate the SDOCT parameters in children operated for PCG, we found that all the SDOCT parameters were significantly different in PCG eyes compared with normal eyes. To our knowledge, this is the first study to evaluate the ONH, RNFL and GCC thickness measurements using SDOCT in children operated for PCG.
PCG is a condition, which is diagnosed clinically. It presents at an age when imaging of posterior segment is usually not possible. Our study was a basic exploratory study to evaluate the ONH, RNFL and GCC thicknesses in children operated for PCG using SDOCT. Our study was not aimed at evaluating the ability of SDOCT in diagnosing PCG.
As none of the imaging modalities in glaucoma currently have normative database for children less than 18 years of age, we recruited a control cohort of children with no posterior segment pathology to compare against the changes found in PCG eyes. In the control group, we found the mean RNFL thickness with RTVue SDOCT to be 114 µm, with the inferior quadrant RNFL being the thickest and the temporal quadrant being the thinnest. Our results are similar to the RNFL thickness recently reported with Spectralis SDOCT in normal Turkish children.12 RNFL thicknesses in our study are also similar to those measured using time domain OCT in normal children of Indian origin.13 To our knowledge, this is also the first study in literature to report the ONH and the GCC parameters in normal children. We found the ONH rim areas in normal children to be thickest in inferior, superior and nasal quadrants comparable with the ISNT rule (inferior rim > superior rim > nasal rim > temporal rim) found in normal adults. GCC thickness was similar in the superior and the inferior hemispheres.
We found the RNFL thickness in glaucomatous eyes to be significantly lesser compared with normal eyes, which is in parallel to that reported in a study on paediatric glaucoma.14 However, the above study examined superior and inferior quadrants only. Other studies on children with juvenile glaucoma also observed significantly reduced RNFL thickness only in the superior and inferior quadrants,15 ,16 in contrast to the concentric decrease seen in our study. Interestingly, in the ONH, we observed the neuroretinal rim to be reduced in a concentric fashion, similar to the RNFL thickness, with all the quadrants showing significant thinning, when compared with that of normal children. GCC thickness was significantly reduced in the superior and the inferior hemispheres in children with PCG, compared with that of normal children. This could possibly suggest a diffuse loss in PCG when compared with focal loss in juvenile glaucoma. Further studies are required to investigate the pattern of structural damage in PCG. In the PCG group, SDOCT parameters also correlated significantly with the clinical estimation of cup to disc ratio as reported in another study17 demonstrating significant decrease in all the global SDOCT parameters in eyes with larger cup to disc ratio. We observed that children with unilateral glaucoma had significantly thinner ONH rims, RNFL and GCC thicknesses when compared with children with bilateral glaucoma. One explanation for this could be related to the significantly early presentation (p=0.05) of the children with bilateral disease (median age at surgery: 5 months) compared with those with unilateral disease (median age at surgery: 7 months).
The limitation of our study was that the children with PCG were statistically significantly younger and more myopic than the control group, even though the actual differences were small. Previous studies have reported age18 and refractive error19 to affect the OCT measurements in adults. However, the effect of age on mean RNFL thickness is about 2 microns reduction per decade increase in age, at least in adults.18 The minimum between-group difference in RNFL and GCC thicknesses in our study was about 15 microns, which is much higher than that might be explained by a 3.5-year difference in age between the two cohorts. In contrast to studies in adults, recent studies conducted in children report different results regarding the influence of age on RNFL thickness. RNFL thickness was positively correlated with age in Chinese children <18 years.20 Another recent study among Indian children showed no correlation with age, in a population <18 years.21 The effect of age on RNFL thickness in children appears to follow a trend that is different from that observed in adults. The difference that we observed in our study probably reflects a true difference in RNFL thickness related to PCG, rather than that explained by small difference in age between the two groups. We observed that the GCC thickness is compromised in congenital glaucoma much like the adult glaucoma.
In conclusion, all SDOCT parameters were significantly compromised in eyes of children with PCG compared with normal children. SDOCT is a promising tool for evaluating the eyes of children operated for PCG in whom the visual fields are unreliable. Future research should examine the test-retest variability of SDOCT parameters and their ability to diagnose progression in these children who are unable to perform reliable visual field tests.
Contributors SS, UKA, HLR, CSG and AKM: design and conduct of study, collection, management, analysis and interpretation of the data preparation, review or approval of the manuscript.
Competing interests None.
Ethics approval Institutional Review Board of L. V. Prasad Eye Institute.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement No additional data. Data available on request from the corresponding author: AKM: firstname.lastname@example.org.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.