Portal de Periódicos da CAPES

The optic nerve head assessed with HRT in 5-16-year-old normal children: normal values, repeatability and interocular difference

2009; Wiley; Volume: 89; Issue: 8 Linguagem: Inglês

10.1111/j.1755-3768.2009.01818.x

1755-3768

Eva Larsson, Eva Nuija, Albert Alm,

Neurological Disorders and Treatments

Purpose: To investigate the normal values, repeatability and interocular difference of the optic nerve head, using Heidelberg retina tomography (HRT), in 5–16-year-old full-term children with normal birth weights. Methods: Fifty-six children with normal visual acuity and refraction were examined with HRT-II/III. Three examinations were performed on each eye. One eye was randomized for analyses of normal values and repeatability, and 54 eyes could be evaluated. The coefficient of variance and the intraclass correlation (ICC) were calculated, and the ISNT rule was noted. The correlation between right and left eyes and the limits of difference were determined in 50 pair of eyes. Results: The mean values of disc and rim areas were 2.16 (SD 0.47) and 1.75 (SD 0.39) mm2 respectively. The coefficients of variance varied between optic nerve parameters from 1.8% to 21%, and the ICCs were >0.88. All parameters except retinal nerve fibre layer thickness correlated with the disc area. The ISNT rule was fulfilled in 56% of the eyes. The interocular difference was large but not statistically significant. Conclusion: We conclude that HRT can be used in children between 5 and 16 years of age, and normal values presented in the study can be used for comparison of children with optic nerve diseases. Because the assessment of rim area varied the least, it may be the parameter to use for follow-up. The normal large interocular difference should be taken into account when comparing eyes in the individual child. Several methods are used for evaluating and following diseases that affect the optic nerve. One of these methods is scanning laser ophthalmoscope, Heidelberg retina tomography (HRT), in which a three-dimensional image of the optic nerve head is built from a serial of optical sections. The HRT technique has mainly been used in adults and in patients with ocular hypertension or glaucoma (Zangwill et al. 1996; Vizzeri et al. 2009). Other authors have shown that HRT also can be useful in the evaluation of optic nerve hypoplasia (Pang & Frantz 2008) and in the analyses of optic cups in patients with cerebral visual impairment (CVI) (Ruberto et al. 2006). The HRT technique is non-invasive and fairly quick, and therefore suitable in the assessment of the optic nerve head in children. However, normal values are based on adults, and only few studies have provided normal values in children (Ruberto et al. 2006; Tong et al. 2007). Repeatability and interocular difference of HRT measurements have been analysed in adults (Gherghel et al. 2000; Strouthidis et al. 2005), but not in children. Such information is necessary for the use of this instrument also in children. We have previously compared HRT with optical coherence tomography (OCT) when assessing the retinal nerve fibre layer (RNFL) in a population-based study of 5–16 year-old children (Larsson et al. 2009). Here, we report the normal values of the optic nerve head from the same group, together with the interocular difference and the repeatability of HRT assessments. A detailed description of the included children has been reported previously (Larsson et al. 2009). In short, 56 full-term children, randomly selected from the Swedish National Board of Health and Social Welfare, between 5 and 16 years of age were examined. The 28 girls and 28 boys had a mean age of 10.1 (SD 3.0, range 5–16) years (Fig. 1). Best corrected visual acuity (VA) was assessed with a logMAR chart. The refraction was assessed with retinoscopy during cycloplegia. Spherical equivalent was calculated, and the astigmatism was noted. Only children with a VA >0.65 (<0.2 logMAR) and/or spherical equivalent between −3 and +3 diopters (D) and/or astigmatism <2 D were included. On ophthalmoscopy, all eyes had clear media and normal fundi, except for one child with a small choroidal naevus. The distribution of ages in the full-term children assessed with Heidelberg retina tomography. The HRT measurements were performed through dilated pupils in both eyes, right before left, by the same examiner. HRT is a confocal scanning laser ophthalmoscope using diode laser. Consecutive two-dimensional sections are integrated to a three-dimensional image of the optic nerve head, and a topographical image is automatically obtained from three scans centred on the optic nerve head. The optic disc margin is defined by a contour line and was drawn by one of the authors (EN). When the contour line was set, the hrt software calculated a series of variables that describe the morphology of the optic nerve head. The hrt-ii was used during the examinations, but the data were transferred for reanalysis by hrt-iii software. Only images with good quality (SD < 40) were accepted. Parameters analysed are presented in Table 1. The modified ISNT rule, as presented by the instrument, was noted – i.e., this rule assumes that the neuroretinal rim in a normal disc is the thickest in the following order: inferior > superior > nasal > temporal. Three separate measurements in each eye were performed to study repeatability, which is defined as the variation in measurements performed by the same person under the same conditions in the same subject. To provide the normal database, one of the two eyes was randomized to be included in the analyses. In these eyes, the repeatability was calculated, and the ISNT rule was noted. Interocular asymmetry was studied in pair of eyes in the same individual. One child was excluded because of the refractive exclusion criteria, and one child could not co-operate in the assessment. After randomization, 54 eyes could be included in the study. The normal values were based on the first examination in the 54 eyes, and the modified ISNT rule was noted in these measurements. Of the three repeated measurements in each of these eyes, three measurements were excluded because of poor quality (SD > 40) and two because of poor fixation. Five children co-operated in only two of three examinations and one child in only one examination. Therefore, of the 54 eyes, we had three measurements in 43 eyes, two in nine eyes and only one in two eyes, that is, the repeatability could be analysed in 52 eyes. The interocular asymmetry, right eye–left eye, was calculated in 50 pair of eyes. Informed consent was obtained from all care-givers before enrolment. The study was approved by the ethics committee in Uppsala, Sweden. Excel, Microsoft Software, was used for the randomization of eyes and spss software, version 13.0 (SPSS Inc, Chicago, IL, USA), for all other analyses. Pearsons correlation was used to bivariate correlations. The calculations of intrasession coefficient of variance (CV) were made by dividing each standard deviation for the repeated measurements by its mean. Intraclass correlation (ICC) was calculated, using the spss software. Paired t-test was used when comparing right and left eyes. Limits of difference were defined as the mean difference between the eyes (right eye − left eye) ± 1.96 SD. The mean VA of the 54 randomized eyes was 1.2 (SD 0.21, range 1.0–1.6). The mean spherical equivalent was 0.60 (SD 0.68, range −0.5 to +3) D, and the mean value of astigmatism was 0.22 (SD 0.32, range 0–1.0) D. There was no difference between the two eyes regarding VA or refraction. Normal values for HRT are presented in Table 1, together with the repeatability expressed as CV and ICC. There were no correlations between the spherical equivalent or astigmatism and the stereometric parameters. A weak correlation was found between age and cup/disc ratio (p = 0.04), rim/disc ratio (p = 0.04) and mean cup depth (p = 0.02), but no correlation was found with the other disc parameters. No difference between the genders was found. The topographical parameters positively correlated with the disc area except for the mean RNFL thickness, the height variation and the measurement of cup shape (Table 1). The modified ISNT rule, as presented by the instrument, was fulfilled in 30/54 (56.0%) eyes. The correlations and interocular differences between right and left eyes are given in Table 2. There was no statistical difference between right and left eyes, except for cup/disc area ratio (p = 0.03) and rim/disc area ratio (p = 0.03). In the present study, we present normal values of the optic nerve head topography, the repeatability of the measurements and the interocular difference assessed with HRT, in full-term children. We found that good quality images could be assessed also in children. Although HRTs have been used in children in a few studies (Martin & Lundvall-Nilsson 2007, Pang & Frantz 2008), there are only three that present normal values in children, and none of the authors discuss repeatability and interocular difference. Ruberto et al. (2006) reported normal values for HRT in comparison with children with CVI, and Tong et al. (2007) presented values of myopic 11–12-year-old Singaporean children. Recently, He et al. (2008) used HRT to analyse disc and cup area in a twin study of children. Apart from the disc area, which was larger in the study by Ruberto et al. (2006) [2.16 (present study) versus 2.48 mm2], our findings resembled theirs. In the Singaporean study, all children were myopic, and because our children had normal refractions, it is more difficult to compare. However, their values of the untilted discs, which were less myopic, were more similar to ours. The rim area was larger in our study [1.75 (present study) versus 1.57 mm2], although the disc area was almost the same [2.16 (present study) versus 2.15 mm2]. Chinese twins, studied by He et al. (2008), had smaller disc area [2.16 (present study) versus 1.97 mm2] but larger cup area [0.41 (present study) versus 0.50 mm2]. Differences among these studies may be as a result different populations. When comparing our results with the values in adults, the children had larger disc sizes and larger rim areas (Mistlberger et al. 1999; Miglior et al. 2002; Hawker et al. 2005). However, the normal distributions were broad and overlapped. We found no correlation with age regarding most of the parameters including disc size, cup area, rim area and mean retinal nerve fibre thickness. Also, Nakamura et al. (1999) found no correlation with age and HRT parameters in adults except for the mean retinal nerve fibre thickness. If there is a difference between children and adults, correlated with age, larger studies with broader age span are probably needed to prove this. In adults, the area of the disc correlates with most HRT topographical parameters (Nakamura et al. 1999; Bowd et al. 2002). This has previously not been evaluated in children but was evident also in the present study (Table 1). Disc size and RNFL thickness did not correlate, which also has been found in adults (Bowd et al. 2002). Some studies show a correlation between optic disc parameters and the refraction of the eye (Nakamura et al. 1999), whereas others do not (Bowd et al. 2002). In children, we found no correlation with refraction in accordance with Tong et al. (2007). However, in the present study all children had fairly normal refraction. According to the ISNT rule, the neuroretinal rim is the thickest inferiorly followed by the superior, nasal and temporal sectors, and may distinguish between normal and glaucomatous discs (Jonas et al. 1999). Still, Harizman et al. (2006) and Sihota et al. (2008) found that the ISNT rule, as evaluated by the HRT, was not fulfilled in a substantial number of normal adult eyes although there was a difference compared to eyes with glaucoma. Our study indicated that eyes that do not fulfil the modified ISNT rule were common also in normal children (44%). The repeatability, expressed as coefficient of variance, varied widely between optic nerve parameter (1.8–21%) in our study (Table 1). In adults, CVs between 4 and 157% have been reported (Sihota et al. 2002, Strouthidis et al. 2005). The rim area had the best repeatability in the present study. Good repeatability for rim area has also been observed in adults (Sihota et al. 2002; Strouthidis et al. 2005). In general, rim area and cup depth were among the measurements with the best repeatability among both children and adults. It is also important to know the normal difference between right and left eyes when evaluating monocular diseases or diseases that affect the eyes differently. Gherghel et al. (2000) used HRT and found a difference between normal right and left eyes in adults regarding RNFL thickness. In children, we found no statistically significant difference between the eyes in any of the topographical parameters except for cup/disc and rim/disc area ratios. However, when studying the limits of difference between the eyes (Table 2), we found a large interval of difference between right and left eyes in proportion to the mean values. The disc area and rim area had the smallest interocular differences. HRT has mostly been used in the diagnosis of glaucoma. There are some studies using the technique in the diagnosis of optic nerve hypoplasia (Pang & Frantz 2008), analysing the optic disc in children with CVI (Ruberto et al. 2006) and quantifying disc elevation in pseudotumor cerebri and pseudopapillary oedema (Trick et al. 2001). The large range of normal values and the great inter-individual variability (SD/mean) in the children, suggests difficulties in diagnosing optic nerve diseases. However, the repeatability was good regarding rim area and rim/disc area ratio, indicating the value of following optic nerve diseases, such as glaucoma, in children. In fact, the instrument may be a valuable tool in any progressive optic nerve disease in children, such as optic nerve glioma or any disease affecting the visual tract. In conclusion, we found that HRT can be used in children from at least 5 years of age, because the technique is quick and non-invasive. Of the optic nerve head parameters, assessed with HRT, the neuroretinal rim area had the best repeatability. Thus, for follow-up, rim area may be the best parameter to use. However, there was a rather large difference between right and left normal eyes, which should be taken into consideration when comparing the two eyes in children. We thank Börje Nordh for skilful and valuable help. This study was supported by the Crown Princess Margareta Foundation for Visually Impaired, the Swedish Society for Medicine and the Mayflower Charity Foundation for Children.