Definition of high myopia by parapapillary atrophy. The Beijing Eye Study

Editor, The spectrum of myopia can be divided into moderate myopia and high myopia. The clinical importance of the differentiation between both types of myopia is based on findings that high myopia in contrast to moderate myopia is associated with a higher glaucoma susceptibility, a lower number, smaller size and less advanced type of macular drusen corresponding with a lower risk of early and late age-related macular degeneration, a higher risk for exudative myopic maculopathy, a larger optic nerve head, a decreased best corrected visual acuity and a higher frequency of visual field defects (Xu et al. 2007a). The ophthalmoscopic feature of high myopia is a large myopic crescent around the optic disc because of the myopia-associated stretching of the posterior pole of the fundus. Because the lower limit of the myopic refractive error in the definition of high myopia has not been clearly defined yet (Luo et al. 2006), it was the purpose of our study to assess in a population-based manner at which refractive error beta zone of parapapillary as surrogate for the highly myopic stretching of the posterior fundus starts to markedly increase in size. The answer to this question may help in detecting which eye has high myopia that is associated with a stretching of the posterior pole leading to an increased risk for an exudative myopic maculopathy and to an increased risk for glaucomatous optic neuropathy because of a stretching and thinning of the lamina cribrosa and peripapillary sclera (Xu et al. 2007b). The Beijing Eye Study is a population-based cross-sectional study in northern China as described in detail previously (Xu et al. 2007b; Wang et al. 2008; Jonas et al. 2009). In total, 4439 individuals participated in the eye examination (response rate of 83.4%). The Medical Ethics Committee of the Beijing Tongren Hospital had approved the study protocol, and all participants had given informed consent, according to the Declaration of Helsinki. Colour photographs of the optic disc were taken and morphometrically assessed. Parapapillary atrophy was differentiated in a peripheral alpha zone characterized by an irregular pigmentation and a beta zone at the optic disc border, characterized by visible sclera and visible large choroidal vessels (Jonas et al. 1989; Wang et al. 2008). The magnification by the optic media of the eye was corrected according to Littmann’s method taking into account the refractive error (Littmann 1982). The anterior corneal curvature radius was set at 7.8 mm which is the reported mean for Caucasians and Chinese (Cheung et al. 2000). To check the correction of the magnification of the optic disc photographs including the magnification by the fundus camera, the optic discs of 52 eyes were additionally imaged using a confocal laser scanning tomograph (Heidelberg Retina Tomograph HRT; Heidelberg Engineering, Dossenheim, Germany). Comparing the measurements obtained by planimetry of the digitized optic disc photographs and the measurements performed by confocal laser scanning tomography showed a linear difference of 1:1.02 (Wang et al. 2006). The statistical analysis was performed using a commercially available statistical software package (spss for Windows, version 17.0; SPSS, Chicago, IL, USA). Plotting the data of refractive error and the data of the size of beta zone of parapapillary atrophy against each other in a scatterplot, we determined at which refractive error the steep ascent of the correlation curve between refractive error and parapapillary atrophy occurred. From the 4439 individuals, readable optic disc photographs were available for 4027 (90.7%) subjects. Because glaucoma is associated with a secondary increase in beta zone of parapapillary atrophy (Jonas et al. 1989; Xu et al. 2007c), we excluded all eyes with an intraocular pressure >21 mmHg or any optic nerve disease such as the glaucoma, so that 3827 subjects were eventually included into the statistical analysis. As described in detail recently, glaucoma was defined by the appearance of the optic nerve, not taking into account the intraocular pressure measurement or the results of frequency doubling perimetry (Xu et al. 2007b). Only one eye per subject was taken for statistical analysis. The prevalence of beta zone was 19.8 ± 1.9%. The area of beta zone ranged from 0.10 to 15.0 mm2, with a median of 1.23 mm2. The area of beta zone was significantly (p < 0.001) correlated with myopic refractive error. The increase in beta zone of parapapillary atrophy as surrogate of a highly myopic crescent and myopic stretching of the posterior pole started at a refractive error of −7 to −8 D, with a steep increase in beta zone area towards a high myopic refractive error (Fig. 1). The area of beta zone corresponding to the start of the steep ascent of the regression line was about 2 mm2 which roughly represented the limit of the 95% confidence interval of the normal distribution of beta zone. Scattergram showing the correlation between the area of beta zone of parapapillary atrophy and refractive error in subjects without optic nerve disease and with normal intraocular pressure the Beijing Eye Study. The results suggest that if high myopia is defined as myopia-associated stretching of the posterior fundus pole with the surrogate of a large beta zone of parapapillary atrophy, the mean cut-off limit of high myopia is at about −7 to −8 D. The results agree with a previous investigation from the Beijing Eye Study, in which the prevalence of glaucomatous optic neuropathy was significantly higher for the group of subjects with a myopic refractive error exceeding −6 to −8 D than for the remaining subjects (Xu et al. 2007b). The prevalence of other ocular disorders such as retinal detachment was too low for a meaningful statistical analysis on the potential association between high myopia and these disorders. One may debate the decision to exclude glaucoma subjects from the analysis. Because, however, glaucoma leads to an enlargement of parapapillary atrophy which is not clearly distinguishable from the parapapillary atrophy in high myopia (Jonas et al. 1989; Xu et al. 2007a), we choose to exclude all glaucomatous eyes from the study. One has to take into account, however, that there were relatively many eyes with substantial parapapillary atrophy at −5 D (Fig. 1). It suggests that despite the strong relationship between parapapillary atrophy and refractive error, the mean cut-off point at about −7 to −8 D may vary markedly between individuals. To consider a myopic refractive error of about −7 to −8 D as criterion for the definition of high myopia agrees with previous investigations, in which the size of the optic disc in Chinese and in Caucasians started to markedly increase at a myopic refractive error of about −8 D (Jonas 2005;Wang et al. 2006). One may therefore infer that beyond the mean cut-off point of about −7 to −8 D, patients may be advised to get regular check-ups for diseases related with high myopia.