August 2015
Volume 56, Issue 9
Free
Eye Movements, Strabismus, Amblyopia and Neuro-ophthalmology  |   August 2015
Ten-Year Incidence of Retinal Nerve Fiber Layer Defects: The Beijing Eye Study 2001/2011
Author Affiliations & Notes
  • Ren Jie
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
  • Liang Xu
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
  • Ya Xing Wang
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
  • Li Zhang
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
  • Qi Sheng You
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
  • Hua Yang
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
  • Jost B. Jonas
    Beijing Institute of Ophthalmology, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Capital Medical University, Beijing, China
    Department of Ophthalmology, Medical Faculty Mannheim of the Ruprecht-Karls-University Heidelberg, Heidelberg, Germany
Investigative Ophthalmology & Visual Science August 2015, Vol.56, 5118-5124. doi:10.1167/iovs.15-16682
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Ren Jie, Liang Xu, Ya Xing Wang, Li Zhang, Qi Sheng You, Hua Yang, Jost B. Jonas; Ten-Year Incidence of Retinal Nerve Fiber Layer Defects: The Beijing Eye Study 2001/2011. Invest. Ophthalmol. Vis. Sci. 2015;56(9):5118-5124. doi: 10.1167/iovs.15-16682.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: We examined the 10-year incidence of localized retinal nerve fiber layer defects (RNFLDs) and associated factors in adult Chinese.

Methods: The population-based longitudinal Beijing Eye Study, which included 4439 subjects (age, 40+ years) in 2001, was repeated in 2011 with 2695 subjects participating (66.4% of the survivors). Color fundus photographs taken at baseline and 10 years later were examined for localized RNFLDs, defined as wedge-shaped defects running toward the optic disc.

Results: Fundus photographs were available for 2422 (89.9%) subjects. Incident-localized RNFLDs were detected in 114/2422 persons (4.7% ± 0.2%) or 139/4867 eyes (2.9% ± 0.2%). A total of 25 (22%) persons had incident-localized RNFLDs bilaterally, and 89 (78%) persons unilaterally. Among the 114 persons with incident RNFLDs, 78 persons showed newly developed localized RNFLDs, while 36 subjects showed an enlargement of a preexisting localized RNFLD. In binary logistic regression analysis, a higher 10-year incidence of localized RNFLDs was significantly associated with a higher prevalence of arterial hypertension (P = 0.002; odds ratio [OR], 2.31; 95% confidence interval (CI), 1.35, 3.97), higher diastolic blood pressure (P = 0.02; OR, 1.02; 95% CI, 1.004,1.04), higher prevalence of previous cerebrovascular infarctions (P = 0.007; OR, 2.61; 95% CI, 1.30, 5.22), younger age (P = 0.01; OR, 0.97; 95% CI, 0.94, 0.99), higher prevalence of diabetic retinopathy (P = 0.001; OR, 4.70; 95% CI, 1.92, 11.5), higher prevalence of open-angle (P < 0.001; OR, 39.8; 95% CI, 19.7, 80.5) and angle-closure (P < 0.001; OR, 21.3; 95% CI, 7.90, 57.5) glaucoma, and thinner retinal nerve fiber layer (P < 0.001; OR, 0.97; 95% CI, 0.95, 0.98).

Conclusions: The 10-year incidence of localized RNFLDs in adult Chinese is 4.7% ± 0.2%. It is strongly associated with arterial hypertension and higher prevalence of cerebrovascular infarcts, in addition to ocular reasons, such as glaucoma and diabetic retinopathy. Ophthalmologically normal subjects with incident-localized RNFLDs may be screened for arterial hypertension and cerebrovascular infarcts.

The optic nerve as cerebral fascicle has its origin in the retinal ganglion cell layer. Its axons extend from the inner retina through the optic disc, optic nerve, and optic tract to the lateral geniculate ganglion. Any optic nerve damage affects the retinal nerve fiber layer (RNFL) through the loss of optic nerve fibers. The RNFL has conventionally been examined by ophthalmoscopy or on color fundus photographs.15 The visibility of the RNFL shows a specific pattern with usually the best visibility in the temporal inferior arcade, followed by the temporal superior arcade, nasal superior arcade, and nasal inferior arcade.6 The distribution of the RNFL visibility correlates with the thickness of the RNFL as measured by histomorphometry or confocal scanning laser tomography, with the diameter of the retinal arterioles, and with the normal shape of the neuroretinal rim.711 A loss in the RNFL visibility occurs in a diffuse manner or in a localized pattern.25,12 Localized RNFL defects (RNFLDs) appear as wedge-shaped dark areas running toward, but not necessarily touching, the optic disc border. Since localized RNFLDs do not occur in normal eyes, they have a high diagnostic specificity for an optic nerve abnormality and, thus, a high importance in the clinical detection of an optic nerve damage.13,14 Evaluation of the RNFL, and in particular the search for localized RNFDs, has become an integral part of the routine ophthalmoscopic examination. 
Previous cross-sectional studies examined the prevalence of localized RNFLDs, and the factors associated with localized RNFLDs in hospital-based studies and cross-sectional population-based investigations. To our knowledge, the incidence of localized RNFLDs or the change of existing localized RNFLDs over time has been examined neither in hospital-based studies nor in a population-based manner. In view of the clinical importance of localized RNFLDs, we conducted this study to assess the incidence of localized RNFLDs and their change over time. We chose a population-based study design to avoid a potential selection artifact, inherent as a referral bias to any hospital-based study. 
Methods
The Beijing Eye Study is a population-based study that was performed in 2001 and repeated in 2011. The Medical Ethics Committee of the Beijing Tongren Hospital approved the study protocol and all participants gave informed written consent according to the Declaration of Helsinki. All data underlying the findings in our study are freely available in the manuscript and in data files available from the authors. Of 5324 eligible individuals, 4439 (83.4%) participated in the baseline study in 2001 with a mean age of 56.4 ± 10.4 years (range, 40–101 years). In 2011, the study was repeated by inviting all participants from the survey of 2001. The details of participants and nonparticipants at baseline, and at the 10-year follow-up examination in 2011 and the examinations techniques have been described previously.15,16 
In 2001 and 2011, all study participants underwent an interview with standardized questions on socioeconomic parameters, psychic depression, physical activity, known major systemic diseases, and intake of systemic medication. The ophthalmological examination included measurement of visual acuity, tonometry, slit-lamp–based biomicroscopy of the anterior ocular segment, ophthalmoscopy, and photography of the cornea, lens and fundus (fundus camera, type CR6-45NM; Canon, Inc., Lake Success, NY, USA) with the pupils medically dilated. The study technique already has been described in detail.1517 In 2011, we additionally measured the concentrations of glucose, cholesterol, low-density lipoproteins, and high-density lipoproteins, which were determined in fasting blood samples. Arterial hypertension was defined as a systolic blood pressure ≥ 140 mm Hg and/or a diastolic blood pressure ≥ 90 mm Hg, and/or self-reported current treatment for arterial hypertension with antihypertensive medication. Diabetes was defined as fasting glucose concentrations ≥ 7.0 mmol/L and self-reported diagnosis of diabetes. Dyslipidemia was defined as any of hypercholesterolemia (total cholesterol concentration ≥ 5.72 mmol/L [220 mg/dL]), hypertriglyceridemia (triglyceride concentration ≥ 1.70 mmol/L [150mg/dL]), low high-density lipoprotein-cholesterol (HDL-C concentration ≤ 0.91 mmol/L [35 mg/dL]), or a positive history for dyslipidemia. 
Glaucoma was defined according to the optic nerve head criteria of the International Society of Geographic and Epidemiological Ophthalmology ISGEO,18 and it was defined by a glaucomatous appearance of the optic disc.19 The optic nerve head was glaucomatous (1) if the inferior-superior-nasal-temporal (ISNT)–rule of the neuroretinal rim shape was not fulfilled in early glaucoma and in eyes with a normally shaped optic disc (it included a notch in the neuroretinal rim in the temporal inferior region and/or the temporal superior region), or (2) if an abnormally large cup was present in a small optic disc, which normally would not show cupping. The assessment of the optic disc photographs was carried in a masked manner. Each photograph of a glaucomatous optic disc was independently adjudicated by three senior graders (LX, YXW, JBJ). Besides the criteria described above, no other criteria were applied for the definition of glaucoma. The glaucoma definitions were used only for screening and they did not form the basis for the decision of a treatment of the patients. The application of the criteria for the definitions of glaucoma was based on a mydriatic fundus examination as the fundus photographs had been taken with the pupil medically dilated. As described in detail previously, the whole glaucoma group then was differentiated by gonioscopy or anterior segment optical coherence tomography (OCT) into subjects with open-angle glaucoma or angle-closure glaucoma.19 The minimum criterion for diagnosis of diabetic retinopathy was the presence of at least one microaneurysm. 
The RNFL was assessed on the fundus photographs and the photographs of the optic disc. The photographs had been taken at baseline of the study in the year 2001 and 10 years later. A localized RNFLD was defined as a wedge-shaped and not spindle-like defect, running toward or touching the optic disc border for not more than 60° of the optic disc circumference. The photographs were examined by an experienced ophthalmologist (RJ) in a masked manner without knowledge of the history, morphometric optic disc data, perimetric results, and other data. In case of doubt, the photographs were reexamined by a panel including several ophthalmologists (RJ, YXW, LX, JBJ). The technique already has been described in detail previously.12 Incident-localized RNFLDs were defined as absence of localized RNFLDs in both eyes at the baseline examination in 2001 and presence of a localized RNFLD in at least one eye at the follow-up examination in 2011, or incident-localized RNFLDs were defined by an enlargement of a RNFLD that already existed in 2001. 
In 2011, the thickness of the RNFL was additionally measured by spectral-domain OCT (SD-OCT, Spectralis HRA+OCT; Heidelberg Engineering, Heidelberg, Germany). We used the high-speed resolution mode to collect the images. Circular B-scans (3.4-mm diameter, 768 pixels, 1536 A-scans/s) centered at the optic disc were automatically averaged to reduce speckle noise. The pupil was dilated by the 1% tropicamide before examination. The Spectralis OCT software, Heidelberg Explorer (HEE, version 5.3) was used for the automatic segmentation of the RNFL and for the calculation of the RNFLT. Artificial adjustment of upper and lower boundary lines when position was misplaced was done automatically by the machine itself. 
The statistical analysis was performed using a commercially available statistical software package (SPSS for Windows, version 21.0; IBM-SPSS, Chicago, IL, USA). The results of only one randomly selected eye per subject were taken for statistical analysis. For that purpose, we randomly assessed for which study participant numbers the right or left eye was included into the statistical analysis. In a first step of the analysis, we examined the mean values (presented as mean ± SD). Frequencies were presented as mean ± SE. In a second step, we performed univariate analyses with the presence of localized RNFLDs as dependent parameter, and ocular and general parameters as independent parameters. In a third step, we performed binary regression analysis, with the presence of localized RNFLDs as dependent variable and all those parameters as independent variables that were significantly associated with the presence of localized RNFLDs in the univariate analysis. We then dropped step by step those parameters from the list of independent variables that no longer showed a significant association with the presence of localized RNFLDs, starting with the parameter with the highest P value. Odds ratios (OR) and 95% confidence intervals (95% CI) were presented. All P values were 2-sided and were considered statistically significant when the values were less than 0.05. 
Results
Of 4439 subjects who participated in the baseline examination in 2001, 2695 (1139 men [42.3%]) were reexamined in 2011, whereas 379 were dead and 1365 did not agree to be reexamined or had moved away (see Fig.). The response rate was 60.7% of the original cohort, or 66.4% of the survivors. Study participants were significantly (P < 0.001) younger than nonparticipants and lived more often in the rural region. Both groups did not differ in refractive error (P = 0.23) and sex (P = 0.06). Of the 2695 subjects reexamined in 2011, 185 (396 eyes) were excluded, since the quality of the fundus photographs did not allow a reliable assessment of the RNFL. Additionally, 88 subjects (129 eyes) were excluded since they showed a localized RNFLD in 2001 that did not change during the follow-up period. The study population eventually included 2422 subjects (Table 1). The participants excluded from the study (n = 273) compared to the persons included were significantly older (P < 0.001), lived more often in the rural region (P < 0.001), and had a lower cognitive function score (P < 0.001), higher systolic (P < 0.001) and diastolic (P < 0.001) blood pressure, and a thinner overall RNFL (P < 0.001, Table 1). Both groups did not differ in parameters, such as sex (P = 0.39), body mass index (P = 0.41), IOP (P = 0.70), and biochemical blood analyses. 
Figure
 
Flow chart of study participants in The Beijing Eye Study 2001/2011.
Figure
 
Flow chart of study participants in The Beijing Eye Study 2001/2011.
Table 1
 
Comparisons of Subjects Included or Excluded from the Study (Mean ± SD)
Table 1
 
Comparisons of Subjects Included or Excluded from the Study (Mean ± SD)
An incident-localized RNFLD from 2001 to 2011 was detected in 114/2422 persons (4.7% ± 0.2%) or 139/4867 eyes (2.9% ± 0.2%). Of these 114 persons, 25 (22%) had localized RNFLDs bilaterally, and 89 (78%) had localized RNFLDs unilaterally. Of the 89 persons with unilateral incident RNFLDs, 4 subjects showed localized RNFLDs in one eye, while the contralateral eye revealed a diffuse RNFL loss. Among the 114 persons with incident RNFLDs, 78 showed newly developed localized RNFLDs, while 36 subjects showed an enlargement of a preexisting localized RNFLD. 
Looking for ophthalmologic causes for the incident-localized RNFLDs revealed that of the 139 eyes (114 patients) with incident-localized RNFLDs, 61 (44%) eyes had glaucoma or were glaucoma suspects. In the case of a glaucoma suspect, the definition of glaucoma as outlined above was partially, but not fully, fulfilled. Seven (5%) eyes showed nonglaucomatous optic nerve damage with increased pallor of the neuroretinal rim of the optic disc, a normal appearance of the neuroretinal rim shape according to the ISNT-rule, and decreased retinal arteriolar diameter.11 In two (1%) eyes a retinal vascular disease, such as diabetic retinopathy or retinal vein occlusions, was assumed to be associated with the incident-localized RNFLDs. For 69 (50%) of the eyes, the cause for the incident-localized RNFLDs remained elusive. 
In univariate analysis, the 10-year incidence of localized RNFLDs was significantly associated with the systemic parameters of higher systolic (P = 0.003) and diastolic (P = 0.007) blood pressure, and higher prevalence of arterial hypertension (P < 0.001), heavier body weight (P = 0.002), longer waist circumference (P < 0.001), longer hip circumference (P = 0.04), higher body mass index (P = 0.02), higher prevalence of previous cerebral infarction (P = 0.046), higher prevalence of diabetes mellitus (P < 0.001) and diabetic retinopathy (P = 0.003), rural versus urban region of habitation (P = 0.01), lower cognitive function score (P = 0.03), and with the ocular parameters of shallower anterior chamber depth (P < 0.001), higher IOP (P = 0.008), higher prevalence of glaucoma (P < 0.001) overall, open-angle glaucoma (P < 0.001), angle-closure glaucoma (P < 0.001), and retinal vein occlusions (P = 0.008, Table 2). 
Table 2
 
Comparison of Participants With or Without Incident-Localized RNFLD in the Beijing Eye Study 2001/2011
Table 2
 
Comparison of Participants With or Without Incident-Localized RNFLD in the Beijing Eye Study 2001/2011
The binary logistic regression analysis included the 10-year incidence of localized RNFLDs as dependent variable and all parameters as independent variables, which were significantly associated with RNFLD incidence in the univariate analysis. We first dropped body weight, and circumference of hip and waist due to the collinearity with body mass index. We then dropped step by step all other parameters from the list of independent variables that were no longer significantly associated with incident-localized RNFLDs. In the final model, a higher 10-year incidence of localized RNFLDs was significantly associated with a higher prevalence of arterial hypertension (P = 0.002), higher diastolic blood pressure (P = 0.02), higher prevalence of previous cerebrovascular infarction (P = 0.007), younger age (P = 0.01), higher prevalence of diabetic retinopathy (P = 0.001), higher prevalence of open-angle (P < 0.001) and angle-closure (P < 0.001) glaucoma, and thinner RNFL (P < 0.001, Table 3). 
Table 3
 
Multivariate Analysis of Associations Between the 10-Year Incidence of Localized RNFLDs, and Ocular and Systemic Parameters in the Beijing Eye Study 2001/2011
Table 3
 
Multivariate Analysis of Associations Between the 10-Year Incidence of Localized RNFLDs, and Ocular and Systemic Parameters in the Beijing Eye Study 2001/2011
Discussion
In our population-based study, incidence of localized RNFLDs was 4.7% ± 0.2% per person or 2.9% ± 0.2% per eye. A higher 10-year incidence of localized RNFLDs was significantly (P < 0.05) associated with higher diastolic blood pressure; higher prevalences of arterial hypertension, previous cerebrovascular infarction, diabetic retinopathy, and open-angle and angle-closure glaucoma; younger age; and thinner RNFL. One may infer that the development of localized RNFLDs in patients without glaucoma and without diabetic retinopathy suggests screening for modifiable risk factors common to arterial hypertension and cerebrovascular infarcts. 
To our knowledge, this was the first study on the 10-year incidence of localized RNFLDs so that the results cannot be compared directly with the findings obtained in previous hospital studies or population-based investigations. The association between localized RNFLDs, and glaucoma and diabetic retinopathy also was described in cross-sectional studies.35,13,17,20,21 The association between localized RNFLDs and arterial hypertension or cerebrovascular stroke has not often been described yet.22,23 Kim et al.22 examined 4395 Korean subjects who underwent health checkups, including brain magnetic resonance imaging and fundus photography. Prevalence of RNFLDs, detected in in 5.4% of the study population, was significantly higher in hypertensive subjects (OR, 1.73) and in subjects with cerebral small vessel disease (OR, 1.58). In a recent hospital-based study by Xu et a.23 on 359 patients with arterial hypertension and 331 individuals without hypertension, presence of localized RNFLDs was significantly associated with arterial hypertension after adjusting for blood concentrations of glycosylated hemoglobin HbA1c and hyperlipidemia in a multivariate analysis. The association between arterial hypertension and localized RNFLDs was in addition to the association between arterial hypertension and the classic retinal microvascular abnormalities such arteriovenous nicking and generalized arteriolar narrowing. Also, the various categories of arterial blood pressure (optimal/normal/highly normal; mild, moderate, and severe hypertension) were significantly associated with localized RNFLDs in multivariate analysis. The ORs for the association between localized RNFLDs and arterial hypertension were higher than those for the association between retinal microvascular abnormalities and arterial hypertension. These investigations agreed with a study on arterial hypertensive rhesus monkeys, in which localized RNFLDs developed during the follow-up.24 In a hospital-based study, localized RNFLDs were significantly associated with acute or preceding cerebral stroke after adjusting for arterial blood pressure and other factors associated with stroke or localized RNFLDs.25 These preceding cross-sectional hospital-based studies, the experimental monkey study, and the present longitudinal population-based investigations agree on the association between localized RNFLDs and arterial hypertension and cerebral stroke. It suggests that localized RNFLDs should be searched for in any patients with arterial hypertension and/or at risk for stroke. This may hold true in particular in patients in whom a localized RNFLD cannot be explained by coexisting glaucoma, diabetic retinopathy, or any other ocular reason for localized optic nerve damage. The reason for the association between localized RNFLDs and stroke may be the anatomy with the retina and brain sharing the same vascular system stemming from the internal carotid artery. 
The detectability of localized RNFLDs increases with a greater thickness of the RNFL, which shows a typical distribution in normal eyes. It usually is thickest in the temporal inferior and temporal superior sectors, followed by the nasal superior and nasal inferior sectors.10 The regional differences in the ophthalmoscopic detectability of the RNFL is associated with the diameters of the retinal arterioles, which usually are thicker in the temporal inferior vessel arcade than in the temporal superior arcade, with the physiologic shape of the neuroretinal rim of the optic nerve head following the so called ISNT rule, and with the location of the foveola approximately 0.5 mm inferior to the horizontal optic nerve head axis.10,11 Depending on these physiological parameters of the RNFL, localized RNFLDs usually are detected more often in the temporal inferior fundus and temporal superior regions than in other fundus areas.12 
One reason for the association between a higher incidence of localized RNFLDs and younger age (P = 0.01; OR, 0.97; 95% CI, 0.94, 0.99) may have been a higher degree of opacities in the optic media of the eye, in particular of nuclear cataract, in older individuals compared to younger persons, leading to a decrease in the ophthalmoscopic visibility of the RNFL and in the detectability of localized RNFLDs.12 Another reason may have been that older individuals compared to younger ones have experienced a physiological loss of retinal ganglion cell axons of approximately 0.3% per year of life, parallel to a loss of 0.3% of their rods, cones, and retinal pigment epithelium cells.2628 Since localized RNDLs can be detected better the higher the contrast is between the depth of the defect and the height of the surrounding retinal nerve fiber layer tissue, any decrease in the general thickness of the RNFL may be associated with a generally decreased visibility of localized RNFLDs. Finally, a third reason why older age was not associated with an increased incidence of localized RNFLDs in our study was that the multivariate model included major age-related diseases that were associated per se with localized RNFLDs, so that after adjusting for the prevalence of these diseases, the incidence of localized RNFLDs was not associated with older age, but with younger age. The association between localized RNFLDs and diabetic retinopathy, as well as the association between localized RNFLDs and arterial hypertension may be due to preceding microinfarcts in the RNFL, with retinal cotton-wool spots as their ophthalmoscopic correlate.29 These retinal cotton-wool spots disappear after approximately 2 months and leave back a localized RNFLD, which remains visible as long as the surrounding RNFL tissue is thick enough to result in a sufficient spatial contrast between the depth of the localized RNFLD and the height of the surrounding tissue. As long as the RNFL remains sufficiently thick to allow the detectability of localized RNFLDs, the latter can be regarded as a lifelong marker for a previous event damaging the optic nerve in a localized manner. 
If the various parameters that were significantly associated with incident-localized RNFLDs in the multivariate analysis were compared to each other, the strongest associations were detected for open-angle glaucoma and angle-closure glaucoma, followed by diabetic retinopathy (Table 3). Higher diastolic blood pressure with an OR of 1.02 had a relatively weak association with incident RNFLDs (Table 3). If one takes into account, however, that also presence of arterial hypertension was included in the multivariate analysis and that it was associated positively with incident RNFLDs, the importance of arterial hypertension in general and of elevated diastolic blood pressure in particular for the development of localized RBFLDs gets more evident. In the multivariate model, every increase in diastolic blood pressure by 1 mm Hg increased the incidence of localized RNFLDs by 2%. 
This study is part of a whole series of investigations of the Beijing Eye Study, which also addressed aspects of the RNFL previously. These former studies by the Beijing Eye Study were focused on the presence of myelinated nerve fibers in the retina, on the ophthalmoscopic assessment of the RNFL in general, and the associations of the ophthalmoscopic visibility of the RNFL with other ocular and systemic parameters, on the thickness of the RNFL, and its associations with ocular and systemic parameters, on the detection of localized RNFLDs by OCT, and on the association between localized RNFLDs and stroke in a cooperation between the Beijing Eye Study and hospital-based study from a neurological department. The current study presented new data on the 10-year development of new localized RNFLDs. To our knowledge, such data from the Beijing Eye Study have not been presented and have not been discussed before. In a similar manner, the other longitudinal population-based studies in ophthalmology have not yet presented such data and findings.9,17,3032 Other recent studies by the Beijing Eye Study, such as the observed association between the estimated cerebrospinal fluid pressure and retinal vein diameter in arterial hypertension, and the 10-year mortality of ocular diseases, do not have a direct association with the topic of the current study.33,34 
From a practical point of view, one may raise the question whether brain imaging should be performed to exclude cerebral ischemic attacks in patients without glaucoma or diabetic retinopathy and who present with an incident RNFLD. The design of our study did not allow answering such a question. Since, however, stroke was one of the main factors associated with incident RNFLDs, in particular after exclusion of glaucoma and diabetic retinopathy, one may consider a neurologic examination of such patients. 
Interestingly, for approximately half of the incident-localized RNFLDs, an ophthalmological reason could not be named clearly. It may show the clinical importance of localized RNFLDs for the ophthalmological assessment of systemic diseases, such as arterial hypertension and stroke. This list of systemic diseases potentially leading to localized RNFLDs also may include autoimmune pathologies, which also can give rise to wedge-shaped defects of the RNFL.35,36 
Potential limitations of our study should be mentioned. First, we could not use OCT-based data on the incidence of localized RNFLDs, since at baseline of the study in 2001, the OCT device was not available for the study. Therefore, we had to use fundus photographs–based data on the incidence of localized RNFLDs. Previous study had revealed that localized RNFLDs as detected by assessment of fundus photographs were almost congruently detected by SD-OCT.3739 In 2011, however, we measured the RNFL thickness by SD-OCT so that the RNFL thickness could be used as one of the parameters in the multivariate analysis. Second, a multivariate analysis with so many parameters assessed as in the present study with a sample size of 114 subjects (with localized RNFLDs) runs the risk of an overestimation of the statistical significance of associations and differences, and has a marginal statistic power even it is significant. If a Bonferroni correction was applied for adjusting for performing multiple comparisons, most of the P values were, however, still smaller than 0.05. In the multivariate analysis, the associations between the 10-year incidence of localized RNFLDs and presence of arterial hypertension, diabetic retinopathy, open-angle glaucoma, and of angle-closure glaucoma, and thickness of the RNFL remained statistically significant after performing a Bonferroni correction. Third, as for any prevalence study nonparticipation or exclusion of subjects may have led to a confounding effect. In our 10-year follow-up study, of 4439 participants who participated in the baseline examination in 2001, 2422 subjects or 54.6% eventually were included into the present study. In view of the relatively long follow-up, this figure may be acceptable. Strength of our study was that it was the first population-based investigation that assessed the incidence of localized RNFLDs longitudinally, over a relatively long period of 10 years. 
In conclusion, the 10-year incidence of localized RNFLDs in adult Chinese (4.7% ± 0.2%) was strongly associated with arterial hypertension and higher prevalence of cerebrovascular infarcts, in addition to ocular reasons, such as glaucoma and diabetic retinopathy. Development of localized RNFLDs in patients without glaucoma, diabetic retinopathy or other ocular reasons for a localized RNFLD suggests arterial hypertension or cerebrovascular infarction. Ophthalmologically normal subjects with incident-localized RNFLDs may be screened for arterial hypertension and cerebrovascular infarcts. 
Acknowledgments
Supported by the State Natural Sciences Fund (81041018), Natural Sciences Fund of Beijing Government (7092021;7112031), and the National Key Technology R&D Program of the Ministry of Science and Technology (Nos. 2012BAH05F05 and 2013BAH19F04). The authors alone are responsible for the content and writing of the paper. 
Disclosure: R. Jie, None; L. Xu, None; Y.X. Wang, None; L. Zhang, None; Q.S. You, None; H. Yang, None; J.B. Jonas, None 
References
Vogt A. Herstellung eines gelbblauen Lichtfiltrates, in welchem die Makula centralis in vivo in gelber Farbe erscheint, die Nervenfasern der Netzhaut und andere feine Einzelheiten derselben sichtbar werden und der Grad der Gelbfärbung der Linse ophthalmoskopisch nachweisbar ist. Graefe's Arch Clin Exp Ophthalmol. 1913; 84: 293–311.
Hoyt WF, Schlicke B, Eckelhoff RJ. Funduscopic appearance of a nerve fiber bundle defect. Br J Ophthalmol. 1972; 568: 577–583.
Airaksinen PJ, Drance SM, Douglas GR, Mawson DK, Nieminen H. Diffuse and localized nerve fiber loss in glaucoma. Am J Ophthalmol. 1984; 98: 566–571.
Sommer A, Miller NR, Pollack I, Maumenee AE, George T. The nerve fiber layer in the diagnosis of glaucoma. Arch Ophthalmol. 1977; 95: 2149–2156.
Hitchings RA, Poinoosawamy D, Poplar N, Sheth GP. Retinal nerve fibre layer photography in glaucomatous patients. Eye. 1987; 1: 621–625.
Jonas JB, Nguyen NX, Naumann GO. The retinal nerve fiber layer in normal eyes. Ophthalmology. 1989; 96: 627–632.
Varma R, Skaf M, Barron E. Retinal nerve fiber layer thickness in normal human eyes. Ophthalmology. 1996; 103: 2114–2119.
Dichtl A, Jonas JB, Naumann GOH. Retinal nerve fiber layer thickness in human eyes. Graefes Arch Clin Exp Ophthalmol. 1999; 237: 474–479.
Wang YX, Pan Z, Zhao L, You QS, Xu L, Jonas JB. Retinal nerve fiber layer thickness. The Beijing Eye Study 2011. PLoS One. 2013; 8: e66763.
Jonas JB, Schiro D. Visibility of the normal retinal nerve fiber layer correlated with rim width and vessel caliber. Graefe's Arch Clin Exp Ophthalmol. 1993; 231: 207–211.
Jonas JB, Gusek GC, Naumann GO. Optic disc cup and neuroretinal rim size, configuration and correlations in normal eyes. Invest Ophthalmol Vis Sci. 1988; 29: 1151–1158.
Jonas JB, Dichtl A. Evaluation of the retinal nerve fiber layer. Surv Ophthalmol. 1996; 40: 369–378.
Chihara E, Matsuoka T, Ogura Y, Matsumura M. Retinal nerve fiber layer defect as an early manifestation of diabetic retinopathy. Ophthalmology. 1993; 100: 1147–1151.
Jonas JB, Schiro D. Localized retinal nerve fiber layer defects in nonglaucomatous optic nerve atrophy. Graefes Arch Clin Exp Ophthalmol. 1994; 232: 759–760.
Xu L, Li Y, Zheng Y, Jonas JB. Associated factors for age-related maculopathy in the adult population in China. The Beijing Eye Study. Br J Ophthalmol. 2006; 90: 1087–1090.
Zhou JQ, Xu L, Wang S, et al. The 10-year incidence and risk factors of retinal vein occlusion: the Beijing Eye Study. Ophthalmology. 2013; 120: 803–808.
Zhang Y, Xu L, Zhang L, Yang H, Wang YX, Jonas JB. Ophthalmoscopic assessment of the retinal nerve fiber layer. The Beijing Eye Study. PLoS One. 2013; 8: e62022.
Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The definition and classification of glaucoma in prevalence surveys. Br J Ophthalmol. 2002; 86: 238–242.
Wang YX, Xu L, Yang H, Jonas JB. Prevalence of glaucoma in North China. The Beijing Eye Study. Am J Ophthalmol. 2010; 150: 917–924.
Kim J, Woo SJ, Ahn J, Park KH, Chung H, Park KH. Long-term temporal changes of peripapillary retinal nerve fiber layer thickness before and after panretinal photocoagulation in severe diabetic retinopathy. Retina. 2012; 32: 2052–2060.
Lee SB, Kwag JY, Lee HJ, Jo YJ, Kim JY. The longitudinal changes of retinal nerve fiber layer thickness after panretinal photocoagulation in diabetic retinopathy patients. Retina. 2013; 33: 188–193.
Kim M, Park KH, Kwon JW, Jeoung JW, Kim TW, Kim DM. Retinal nerve fiber layer defect and cerebral small vessel disease. Invest Ophthalmol Vis Sci. 2011; 52: 6882–6886.
Xu L, Zhou JQ, Wang S, et al. Localized retinal nerve fiber layer defects and arterial hypertension. Am J Hypertens. 2013; 26: 511–517.
Hayreh SS, Jonas JB. Appearance of the optic disk and retinal nerve fiber layer in atherosclerosis and arterial hypertension: an experimental study in rhesus monkeys. Am J Ophthalmol. 2000; 130: 91–96.
Wang D, Li Y, Wang C, et al. Retinal nerve fiber layer defect and ischemic stroke. Stroke. 2014; 45: 1651–1656.
Jonas JB, Schmidt AM, Müller-Bergh JA, Schlötzer-Schrehardt UM, Naumann GO. Human optic nerve fiber count and optic disc size. Invest Ophthalmol Vis Sci. 1992; 33: 2012–2018.
Panda-Jonas S, Jonas JB, Jakobczyk-Zmija M. Retinal photoreceptor density decreases with age. Ophthalmology. 1995; 102: 1853–1859.
Panda-Jonas S, Jonas JB, Jakobczyk-Zmija M. Retinal pigment epithelial cell count distribution, and correlations in normal human eyes. Am J Ophthalmol. 1996; 121: 181–189.
Zhang L, Xu L, Zhang JS, Zhang YQ, Yang H, Jonas JB. Cotton-wool spot and optical coherence tomography of a retinal nerve fiber layer defect. Arch Ophthalmol. 2012; 130: 913.
Zhao L, Wang Y, Chen CX, Xu L, Jonas JB. Retinal nerve fibre layer thickness measured by Spectralis spectral-domain optical coherence tomography. The Beijing Eye Study. Acta Ophthalmol. 2014; 92: e35–41.
You Q, Xu L, Jonas JB. Prevalence of myelinated retinal nerve fibers in a population-based study. Acta Ophthalmol Scand. 2007; 85: 631–632.
Zhao L, Wang YX, Zhang W, et al. Localized retinal nerve fiber layer defects detected by optical coherence tomography. The Beijing Eye Study. PLoS One. 2014; 228: e68998.
Jonas JB, Wang N, Wang S, et al. Retinal vessel diameter and estimated cerebrospinal fluid pressure in arterial hypertension: The Beijing Eye Study. Am J Hypertens. 2014; 27: 1170–1178.
Wang YX, Zhang JS, You QS, Xu L, Jonas JB. Ocular diseases and 10-year mortality: The Beijing Eye Study 2001/2011. Acta Ophthalmol. 2014; 92: e424–8e42.
Wax MB. The case for autoimmunity in glaucoma. Exp Eye Res. 2011; 93: 187–190.
Bell K, Gramlich OW, von Thun und Hohenstein-Blaul N, Does autoimmunity play a part in the pathogenesis of glaucoma? Prog Retin Eye Res. 2013; 36: 199–216.
Hwang JM, Kim TW, Park KH, Kim DM, Kim H. Correlation between topographic profiles of localized retinal nerve fiber layer defects as determined by optical coherence tomography and red-free fundus photography. J Glaucoma. 2006; 15: 223–228.
Ye C, To E, Weinreb RN, et al. Comparison of retinal nerve fiber layer imaging by spectral domain optical coherence tomography and scanning laser ophthalmoscopy. Ophthalmology. 2011; 118: 2196–2202.
Wu XS, Xu L, Jonas JB, Zhang L, Yang H, Chen CX. Agreement between spectral domain optical coherence tomography and retinal nerve fiber layer photography in Chinese. J Glaucoma. 2012; 21: 228–233.
Figure
 
Flow chart of study participants in The Beijing Eye Study 2001/2011.
Figure
 
Flow chart of study participants in The Beijing Eye Study 2001/2011.
Table 1
 
Comparisons of Subjects Included or Excluded from the Study (Mean ± SD)
Table 1
 
Comparisons of Subjects Included or Excluded from the Study (Mean ± SD)
Table 2
 
Comparison of Participants With or Without Incident-Localized RNFLD in the Beijing Eye Study 2001/2011
Table 2
 
Comparison of Participants With or Without Incident-Localized RNFLD in the Beijing Eye Study 2001/2011
Table 3
 
Multivariate Analysis of Associations Between the 10-Year Incidence of Localized RNFLDs, and Ocular and Systemic Parameters in the Beijing Eye Study 2001/2011
Table 3
 
Multivariate Analysis of Associations Between the 10-Year Incidence of Localized RNFLDs, and Ocular and Systemic Parameters in the Beijing Eye Study 2001/2011
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×