You are here

Article Text

Article menu

Download PDFPDF

Clinical science
Original article
Early diabetic changes in the nerve fibre layer at the macula detected by spectral domain optical coherence tomography
  1. Hae Young-Lopilly Park1,
  2. In Tae Kim2,
  3. Chan Kee Park1
  1. 1Department of Ophthalmology and Visual Science, Seoul St Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
  2. 2Department of Ophthalmology and Visual Science, St Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
  1. Correspondence to Dr Chan Kee Park, Department of Ophthalmology and Visual Science, Seoul St Mary's Hospital, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Seocho-ku, Seoul 137-701, Korea; ckpark{at}catholic.ac.kr

Abstract

Aim To detect early nerve fibre layer (NFL) changes around the optic disc and macula in diabetic patients using Cirrus HD-optical coherence tomography (OCT).

Methods Forty normal patients without any optic nerve or retinal disease, 37 patients with diabetes with no diabetic retinopathy (NDR) and 89 patients with diabetic retinopathy (DR) of differing severity were enrolled. The NFL thickness around the optic disc was measured using Cirrus HD-OCT. The NFL thickness at the macula was also determined by scanning the macula with the optic disc scanning technique.

Results The NFL thickness around the optic disc differed statistically among all groups and tended to become thinner as the degree of DR progressed. The mean, superior and inferior peripapillary NFL thickness differed among groups. As the severity of DR progressed, the mean, superior, temporal, inferior and nasal macular NFL thickness tended to become thinner. However, only the macular NFL thickness of the superior sector differed significantly among the groups and especially between the control and NDR groups.

Conclusion The difference in NFL was first detected in the superior macular region, which differed significantly between the control group and diabetic group without clinical DR. This could be detected simply by modifying the Cirrus HD-OCT scan technique to detect the NFL thickness in the macular area.

  • Diabetic retinopathy
  • retinal nerve fibre layer thickenss
  • optical coherence tomography

https://doi.org/10.1136/bjo.2010.191841

Statistics from Altmetric.com

    Request Permissions

    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.

    Introduction

    Several studies have investigated changes in the nerve fibre layer (NFL) in diabetic eyes. Using diabetic rats, the loss of retinal ganglion cells and axons resulting in the thinning of the ganglion cell layers has been detected.1 In addition, atrophy of the retinal neural cells in human cadaver eyes of patients with diabetes has been reported.2 Clinical devices such as scanning laser polarimetry and optical coherence tomography (OCT) have been used to show the thinning of the peripapillary NFL in diabetic retinopathy (DR), and the peripapillary NFL thinning increases with disease severity.3–6 Using advanced technologies, such as the electroretinogram,7 short-wavelength perimetry8 and blue-on-yellow perimetry,9 retinal changes at the microvascular level have been reported in cases with no diabetic retinopathy (NDR). The detection of these changes in NDR should offer new perspectives for the early diagnosis and understanding the mechanisms of DR.

    Recently, several companies have developed newer versions of OCT, such as the Cirrus HD spectral-domain OCT (Carl Zeiss Meditec, Dublin, California), which use spectral domain technology. These OCT machines have a higher axial resolution and scan speed than conventional time-domain OCT. The higher sampling rates of the newer OCTs allow more data to be collected and shorter scan times with a high reproducibility. The Optic Disc Cube mode of the Cirrus HD-OCT consists of 200 A-scans that are derived from 200 B-scans and covers a 6 mm2 area centred on the optic disc. An NFL thickness map is created from the three-dimensional data, and an NFL thickness profile is acquired by resampling data points from the three-dimensional dataset.10–12

    In this study, we used Cirrus HD-OCT to detect the NFL changes in patients with diabetes. By modifying the scan technique, we could easily measure the NFL thickness at the macula and found that the macular NFL could detect subclinical changes in NDR patients. To our knowledge, this is the first report on the changes in the macular NFL in patients with diabetes; detecting these early changes in patients with diabetes without retinopathy may provide information about the mechanism of DR.

    Methods

    Study participants

    This was a cross-sectional, case–control study. We recruited diabetic mellitus (DM) patients with NDR or non-proliferative diabetic retinopathy (NPDR) from the Department of Ophthalmology, Catholic University of Korea, St Mary's Hospital between November 2009 and January 2010. The study was approved by the Catholic University St Mary's Hospital institutional review board and conformed to the Declaration of Helsinki. Informed consent was obtained from all subjects.

    Participants underwent a complete ophthalmological examination, including visual acuity, refraction, slit-lamp biomicroscopy, Goldmann applanation tonometry, dilated stereoscopic examination of the optic disc, stereocolour fundus photography, Cirrus HD-OCT and standard automated perimetry using the Swedish Interactive Threshold Algorithm 24-2 test of the Humphrey Field Analyzer (Carl Zeiss Meditec). One eye per participant was selected randomly for inclusion.

    All participants who were included in the study had a best-corrected visual acuity of 20/40 or better, an intraocular pressure less than 21 mm Hg, a spherical equivalent refractive error within ±5.00 dioptres, both vertical and horizontal cup-to-disc ratios within 0.6 and asymmetry within 0.2, and clear vitreous without clinically significant cataracts. We excluded eyes with coexisting glaucomatous optic disc changes or visual-field defects, neuro-ophthalmological diseases, uveitis, macular diseases that included macular degeneration, retinal artery or vein occlusion, and degenerative myopia that could affect the NFL thickness.

    The patients with diabetes mellitus were divided into four stages based on the Early Treatment Diabetic Retinopathy Study classification as determined by a retinal specialist through stereocolour fundus photography. NDR was defined as the absence of all features of DR; mild NPDR was defined as the presence of microaneurysms, retinal haemorrhages and hard exudates; moderate NPDR was defined as the features of mild NPDR plus cotton wool spots or intraretinal microvascular abnormalities; severe NPDR was defined as having any one of (1) severe intraretinal haemorrhages and microaneurysms in all four quadrants, (2) venous beading in two or more quadrants or (3) moderate intraretinal microvascular abnormalities in at least one quadrant. According to the ETDRS classification using stereocolour fundus photography, eyes with clinically significant macular oedema were classified.13 14 The type of clinically significant macular oedema was classified by OCT; diffuse macular oedema; cystic macular oedema; focal oedema with minimal exudates; and significant hard exudates.11 15 16

    Optical coherence tomography scanning procedure

    The Cirrus HD-OCT (software version 3.0.0.50) uses spectral domain technology of an Optic Disc Cube obtained from a three-dimensional dataset composed of 200 A-scans from each of 200 B-scans. The software automatically determined the centre of the disc and then extracted a circumpapillary circle with a radius of 1.73 mm from the cube dataset for the NFL thickness measurement. After measuring the NFL thickness around the optic disc, the glaucoma software was applied to the macular region. We centred the fovea at the centre of the circumpapillary circle and the Optic Disc Cube program automatically measured the NFL thickness on a circle of 1.73 mm radius from the fovea. Then, macular images were acquired using Macular Cube protocols. The macular cube protocol acquires all 128 OCT B-scans in a continuous, automated sequence. A mean retinal thickness map of nine zones, including a 1 mm central zone and the mean macular thickness over a 6 mm scan diameter, was measured automatically. The macular thickness of the four sectors was defined as the mean of the retinal thickness of 3 mm diameter and 6 mm diameter circles. Poor-quality scans were defined as those with a signal strength of <6, the presence of overt misalignment of the surface detection algorithm on ≥15% of consecutive A-scans of 20% of cumulative A-scans or overt decentration of the circle location assessed subjectively. These were excluded. All images were acquired by one well-trained technician. Pharmacological dilation was performed if the pupil was small.

    Statistical analysis

    Statistical analyses were performed with the Statistical Package for the Social Sciences II software version 12.0.0 (SPSS, Tokyo, Japan). A probability value of <0.05 was considered statistically significant. Differences in the analysis of the results of the baseline datas and OCT RNFL parameters among groups were analysed using a one-way analysis of variance and Scheffé's post hoc tests. Nonlinear values were compared with the χ2 test.

    Results

    A total of 138 eligible eyes of 138 patients were entered into the scanning protocol, and 126 eyes of 126 patients with diabetes were included in the subsequent analysis. Twelve eyes were not analysed, as their scans did not meet the criteria for being of high quality. Of these, two eyes had overt misalignment of the NFL detection, and all of two eyes had cystic macular oedema. Two eyes from the control group, two from the NDR group, three from the mild NPDR group, two from the moderate NPDR group and two from the severe NPDR group were excluded.

    Among the eyes included, one had diffuse macular oedema, and two had focal oedema with minimal exudates in the moderate NPDR group. Among the severe NPDR eyes, one had focal oedema with minimal exudates, three had diffuse macular oedema, and two had significant hard exudates. In these eyes, no overt misalignment of the NFL detection was checked once again by looking through the cross-sectional images of the scanned macular region on the scan circle.

    Table 1 lists the demographic data from 126 patients with DM and 40 normal control patients. Age, intraocular pressure and refractive error that might affect the NFL thickness were not statistically different among the groups. No abnormality of optic disc appearance was detected, and the cup-to-disc ratio did not differ significantly among the groups. HbA1c were statistically different between groups, since the disease severity was different (p=0.018). A visual-field examination showed different retinal sensitivity between groups, which correlated with the stage of DR (p<0.001).

    View this table:
    Table 1

    Demographic features of the control subjects and subjects with diabetes mellitus

    The NFL thickness around the optic disc measured using the HD-OCT Optic Disc Cube is shown in table 2 and figure 1. The mean thickness differed statistically among all groups (p=0.031) and tended to become thinner as the state of the DR progressed. The peripapillary NFL thickness was thinner in the DM group superior and inferior to the optic disc (p=0.012 and p=0.027, respectively). Comparing the control group with the NDR group, no significant differences were found (p=0.712). By contrast, comparing the NDR and mild NPDR groups, significant differences were observed for the mean, superior and inferior peripapillary NFL thicknesses (p=0.041, p=0.018 and p=0.034, respectively). The superior peripapillary NFL thickness was 132.33±11.20 μm in the NDR group and 109.50±13.52 μm in the mild NPDR group; this was the greatest difference among the NFL parameters around the optic disc.

    View this table:
    Table 2

    Retinal nerve fibre layer thickness around the optic disc measured using HD-optical coherence tomography

    Figure 1
    Figure 1

    Comparison of the retinal nerve fibre layer (RNFL) thickness around the optic disc. NDR, no diabetic retinopathy; NPDR, non-proliferative diabetic retinopathy.

    The total retinal thickness and NFL thickness around the macula measured by HD-OCT are shown in table 3. The total retinal thickness at the macular area was significantly different among all groups (p=0.043) and tended to increase with the progression of DR. However, a comparison of the macular thickness in the four sectors revealed no statistical differences, as shown in figure 2. The mean macular NFL thickness was 39.92±9.3 μm in the control group. In the control group, the macular NFL thickness was 49.88±8.4 μm in the superior sector, 35.73±7.0 μm in the temporal sector, 44.12±7.7 μm in the inferior sector and 40.11±6.9 μm in the nasal sector. As the severity of DR progressed, the mean, superior, temporal, inferior and nasal macular NFL thickness tended to become thinner. However, the mean, temporal, inferior and nasal macular NFL thickness were not significantly different among groups (p=0.680; p=0.622; p=0.730; p=0.680; respectively). Only the macular NFL thickness of the superior sector was significantly different between the control and NDR groups (p=0.046). The NFL thickness at the superior macular area was 49.88±8.4 μm in the control group and 39.14±8.2 μm in the NDR group (figure 3).

    View this table:
    Table 3

    Macular and macular retinal nerve fibre layer (NFL) thickness measured using HD-optical coherence tomography

    Figure 2
    Figure 2

    Comparison of the total retinal thickness at the macula. NDR, no diabetic retinopathy; NPDR, non-proliferative diabetic retinopathy.

    Figure 3
    Figure 3

    Comparison of the nerve fibre layer (NFL) thickness at the macula. NDR, no diabetic retinopathy; NPDR, non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy.

    Figure 4 illustrates representative cases of a 62-year-old healthy normal control and a 55-year-old patient with diabetes with NDR. Disc and red-free photographs show no NFL defects or glaucomatous optic disc changes in both cases. The Cirrus HD-OCT around the optic disc also shows normal findings in both cases, although the mean, superior, inferior, temporal and nasal NFL thicknesses are decreased in the NDR eye. The Optic Disc Cube program was applied to the macular region with the fovea centred at the centre of the 1.73 mm radius circle. The normative data for around the optic disc are applied to the macular NFL measurements, and so the normative classification is not correct. However, the superior macular NFL thickness is 41 μm in the right eye, which is thinner than the superior macular NFL thickness in the control eye (55 μm).

    Figure 4
    Figure 4

    Cases of a 62-year-old healthy control and a 55-year-old patient with diabetes without diabetic retinopathy (NDR). Optic disc cube scanning of Cirrus optical coherence tomography around the optic disc shows normal findings in the control eye (A) and NDR eye (C). Optic disc cube scanning of Cirrus optical coherence tomography was applied around the macula. The centre of a 1.73 mm radius circle was placed on the fovea, and the scanning automatically measured the nerve fibre layer (NFL) around the macula. The normative data for around the optic disc are applied to the macular retinal nerve fibre layer (RNFL) measurements, and so the normative classification is not correct. However, the superior macular NFL thickness is decreased in the NDR eye (D) compared with control eye (B). INF, inferior; NAS, nasal; SUP, superior; TEMP, temporal; TSNIT, temporal–superior–nasal–inferior–temporal.

    Discussion

    Diabetes-associated NFL loss develops before the onset of visible vascular retinopathy, and RNFL thinning increases with disease severity.2 5 17–20 This is why some reports suggest the usefulness of OCT for the early detection of DR.19 Using HD-OCT, we expected to find more delicate changes in the NFL with DM. However, problems arise in using the OCT to measure the reduced NFL thickness in DM patients. The reduced NFL thickness is not detected in the macula as the stage of DR progresses, because increased vascular permeability masks the effects of NFL loss when total retinal thickness is compared. By applying the HD-OCT Optic Disc Cube software for NFL measurements around the optic disc to the macular area, we expected to find changes in the macular NFL thickness of DM patients, despite the total macular thickness.

    Our NFL thickness measurements around the optic disc using Cirrus HD-OCT showed a similar trend to that of previous reports.4 6 17 The peripapillary NFL thickness was decreased in clinically detectable DR, and the thinning of peripapillary NFL increased with worsening disease severity. The changes in NFL were significant in mild NPDR patients in our study, compared with the healthy control and NDR patients. A reduction in the NFL in diabetic eyes without glaucoma has been reported from the early stage of vasculopathy.20 21 The apoptosis of retinal ganglion cells (RGCs) is enhanced in the sensory retina in diabetes, and the death of the RGCs occurs early in diabetic eyes.22 23 The potential cause of diabetes-associated NFL loss is mainly explained by ischaemia, which is caused by retinal vasculopathy.17 24 In addition, some reports suggest that neuronal dysfunction or neuropathy precedes the vascular abnormalities in the early stage of DM.25–27 However, the NFL change was found mainly in the superior region in our study. In addition, an animal model of diabetes showed twice the number of microaneurysms and cellular capillaries in the superior retina compared with the inferior retina.28 Lower perfusion in the superior retina and the ONH may cause greater ischaemia, and the RGCs in the superior area may tend to be damaged structurally, as reflected in the superior NFL loss in the diabetic retina. Still, there is some debate as to whether NFL changes in diabetic eye are a result of the effect of vascular diabetic retinopathy or whether they are primarily caused by direct neurological damage from chronic hyperglycaemia.

    The total retinal thickness of the macular region differed significantly among groups. The total retinal thickness increased with disease severity. This was comparable with reported findings that the macular thickness increased gradually with the duration of DM because of an increase in vascular permeability in diabetic retina.29 30 The macular region has a characteristic mechanism that differs from other retinal regions. In the macular region, RGC bodies residing in the inner layer are multilayered and 10- to 20-fold thicker than their axons, and detection of RGC changes may be easier in the macular region than in the peripheral retina or peripapillary region.31 The Henle fibre layer consists of Muller cells and axons in the macular region. Ischaemic conditions might disrupt Muller cell function and cause macular oedema. Since ischaemia may be the main predisposing condition of diabetes-associated NFL loss, the macular region may be more fragile in terms of diabetic damage than the peripapillary region. This is why we focused on the change in the macular NFL thickness in patients with diabetes. Our results show that the macular NFL thickness could detect more earlier changes than the peripapillary NFL, especially in the superior macular region. No significant differences were found in the peripapillary NFL between the control and NDR groups, whereas the superior NFL thickness of the macula differed significantly. The superior macula was not significantly thicker in NDR patients compared with the control patients, although a difference was observed in the NFL thickness in that region. This shows that the altered NFL thickness should be considered independent of the changes in the total retinal thickness in the macular region. Measuring the macular NFL thickness, especially in the superior macular region, may be useful for detecting the early changes of DR.

    The measurement of the macular NFL thickness was only obtained by segmentation of the OCT A-scan profile in previous reports.32 33 In the previous reports, the macular NFL was defined by boundaries identified by automated computer program or by experts. Other than the scanning procedure, further image processing and analysis were needed. Even with these complicated methods, boundary detection failures occurred in 4%, and poor-quality images could not be analysed in 56%.32 The method that we used to measure the NFL thickness at the macula was simple and easy. This became possible with the advent of spectral domain OCT. The Optic Disc Cube scan of the Cirrus HD-OCT was applied to the macula with the scan circle centred at the fovea. Since there is no method to directly measure the macular NFL at present, our method could provide possibilities (figure 5A). Our scan length was 3.46 mm, and it is confined to the macular region representing more information about the NFL thickness of the macula. And more than only the average NFL thickness, we could obtain sector thicknesses (superior, inferior, nasal, and temporal) at the macula. We found that the superior sector of the macular NFL was the thickest, followed by the inferior sector of the macula. Additionally, it was shown to detect the NFL accurately, despite the presence of diffuse macular oedema or hard exudates (figure 5C,D). However, in eyes with cystic macular oedema, it failed to identify the NFL, and overt misalignment of the NFL detection occurred (figure 5B). Validation of this method may need further study, making comparisons with other methods to measure macular NFL.

    Figure 5
    Figure 5

    (A) Cross-section of the scanned macular region on the 1.73 mm radius circle. The program automatically identifies the nerve fibre layer (NFL), between the red and purple boundary. The boundary in red is the inner limiting membrane (ILM), and the boundary in purple is the border between the NFL and the ganglion cell layer. (B) In cystic macular oedema, the ILM is well demarcated. However, the purple boundary shows misalignment, which means failure of the automatic NFL detection. Diffuse macular oedema (C) and the presence of hard exudates (D) did not affect the alignment of boundaries in both the ILM and the border between the NFL and ganglion cell layer.

    In conclusion, we detected changes in the NFL thickness in patients with diabetes using Cirrus HD-OCT. The NFL in the superior macular region differed significantly between the healthy control patients and patients with diabetes without DR. Looking more closely at the superior macular NFL thickness can be useful for detecting the earliest changes of DR in patients with diabetes. Using the optic disc scanning technique of Cirrus HD-OCT for the macular region is a simple, easy way to measure the NFL thickness in the macula, and it may provide useful information to understand the changes in various retinal disease.

    References

    1. ↵
      1. Zhang L,
      2. Ino-ue M,
      3. Dong K,
      4. et al
      . Retrograde axonal transport impairment of large- and medium-sized retinal ganglion cells in diabetic rat. Curr Eye Res 2000;20:131–6.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Lieth E,
      2. Gardner TW,
      3. Barber AJ,
      4. et al
      . Retinal neurodegeneration: early pathology in diabetes. Clin Exp Ophthalmol 2000;28:3–8.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Takahashi H,
      2. Chihara E
      . Impact of diabetic retinopathy on quantitative retinal nerve fiber layer measurement and glaucoma screening. Invest Ophthalmol Vis Sci 2008;49:687–92.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Takahashi H,
      2. Goto T,
      3. Shoji T,
      4. et al
      . Diabetes-associated retinal nerve fiber damage evaluated with scanning laser polarimetry. Am J Ophthalmol 2006;142:88–94.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Oshitari T,
      2. Hanawa K,
      3. Adachi-Usami E
      . Changes of macular and RNFL thicknesses measured by Stratus OCT in patients with early stage diabetes. Eye (Lond) 2009;23:884–9.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Sugimoto M,
      2. Sasoh M,
      3. Ido M,
      4. et al
      . Detection of early diabetic change with optical coherence tomography in type 2 diabetes mellitus patients without retinopathy. Ophthalmologica 2005;219:379–85.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Ghirlanda G,
      2. Di Leo MA,
      3. Caputo S,
      4. et al
      . Detection of inner retina dysfunction by steady-state focal electroretinogram pattern and flicker in early IDDM. Diabetes 1991;40:1122–7.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Remky A,
      2. Arend O,
      3. Hendricks S
      . Short-wavelength automated perimetry and capillary density in early diabetic maculopathy. Invest Ophthalmol Vis Sci 2000;41:274–81.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Nomura R,
      2. Terasaki H,
      3. Hirose H,
      4. et al
      . Blue-on-yellow perimetry to evaluate S cone sensitivity in diabetics. Ophthalmic Res 2000;32:69–72.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Kiernan DF,
      2. Hariprasad SM,
      3. Chin EK,
      4. et al
      . Prospective comparison of cirrus and stratus optical coherence tomography for quantifying retinal thickness. Am J Ophthalmol 2009;147:267–75.e2.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Forooghian F,
      2. Cukras C,
      3. Meyerle CB,
      4. et al
      . Evaluation of time domain and spectral domain optical coherence tomography in the measurement of diabetic macular edema. Invest Ophthalmol Vis Sci 2008;49:4290–6.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Keane PA,
      2. Mand PS,
      3. Liakopoulos S,
      4. et al
      . Accuracy of retinal thickness measurements obtained with Cirrus optical coherence tomography. Br J Ophthalmol 2009;93:1461–7.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Anonymous
      . Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study report number 1. Early Treatment Diabetic Retinopathy Study research group. Arch Ophthalmol 1985;103:1796–806.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Anonymous
      . Early Treatment Diabetic Retinopathy Study design and baseline patient characteristics. ETDRS report number 7. Ophthalmology 1991;98(5 Suppl):741–56.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Baskin DE
      . Optical coherence tomography in diabetic macular edema. Curr Opin Ophthalmol 2010;21:172–7.
      OpenUrlCrossRefPubMed
    16. ↵
      1. Yamamoto S,
      2. Yamamoto T,
      3. Hayashi M,
      4. et al
      . Morphological and functional analyses of diabetic macular edema by optical coherence tomography and multifocal electroretinograms. Graefes Arch Clin Exp Ophthalmol 2001;239:96–101.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Lopes de Faria JM,
      2. Russ H,
      3. Costa VP
      . Retinal nerve fibre layer loss in patients with type 1 diabetes mellitus without retinopathy. Br J Ophthalmol 2002;86:725–8.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Goebel W,
      2. Kretzchmar-Gross T
      . Retinal thickness in diabetic retinopathy: a study using optical coherence tomography (OCT). Retina 2002;22:759–67.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Lattanzio R,
      2. Brancato R,
      3. Pierro L,
      4. et al
      . Macular thickness measured by optical coherence tomography (OCT) in diabetic patients. Eur J Ophthalmol 2002;12:482–7.
      OpenUrlPubMedWeb of Science
    20. ↵
      1. Villain MA,
      2. Greenfield DS
      . Peripapillary nerve fiber layer thickness measurement reproducibility using optical coherence tomography. Ophthalmic Surg Lasers Imaging 2003;34:33–7.
      OpenUrlPubMedWeb of Science
    21. ↵
      1. Greenfield DS,
      2. Knighton RW,
      3. Feuer WJ,
      4. et al
      . Correction for corneal polarization axis improves the discriminating power of scanning laser polarimetry. Am J Ophthalmol 2002;134:27–33.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Asnaghi V,
      2. Gerhardinger C,
      3. Hoehn T,
      4. et al
      . A role for the polyol pathway in the early neuroretinal apoptosis and glial changes induced by diabetes in the rat. Diabetes 2003;52:506–11.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Barber AJ,
      2. Lieth E,
      3. Khin SA,
      4. et al
      . Neural apoptosis in the retina during experimental and human diabetes. Early onset and effect of insulin. J Clin Invest 1998;102:783–91.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Lim MC,
      2. Tanimoto SA,
      3. Furlani BA,
      4. et al
      . Effect of diabetic retinopathy and panretinal photocoagulation on retinal nerve fiber layer and optic nerve appearance. Arch Ophthalmol 2009;127:857–62.
      OpenUrlCrossRefPubMed
    25. ↵
      1. Greenstein VC,
      2. Shapiro A,
      3. Zaidi Q,
      4. et al
      . Psychophysical evidence for post-receptoral sensitivity loss in diabetics. Invest Ophthalmol Vis Sci 1992;33:2781–90.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Varkonyi TT,
      2. Peto T,
      3. Degi R,
      4. et al
      . Impairment of visual evoked potentials: an early central manifestation of diabetic neuropathy? Diabetes Care 2002;25:1661–2.
      OpenUrlFREE Full Text
    27. ↵
      1. Kern TS,
      2. Engerman RL
      . Vascular lesions in diabetes are distributed non-uniformly within the retina. Exp Eye Res 1995;60:545–9.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Oshima Y,
      2. Emi K,
      3. Yamanishi S,
      4. et al
      . Quantitative assessment of macular thickness in normal subjects and patients with diabetic retinopathy by scanning retinal thickness analyser. Br J Ophthalmol 1999;83:54–61.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Weinberger D,
      2. Axer-Siegel R,
      3. Landau D,
      4. et al
      . Retinal thickness variation in the diabetic patient measured by the retinal thickness analyser. Br J Ophthalmol 1998;82:1003–6.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Yasukawa T,
      2. Kiryu J,
      3. Tsujikawa A,
      4. et al
      . Quantitative analysis of foveal retinal thickness in diabetic retinopathy with the scanning retinal thickness analyzer. Retina 1998;18:150–5.
      OpenUrlPubMedWeb of Science
    31. ↵
      1. Leung CK,
      2. Chan WM,
      3. Yung WH,
      4. et al
      . Comparison of macular and peripapillary measurements for the detection of glaucoma: an optical coherence tomography study. Ophthalmology 2005;112:391–400.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Ishikawa H,
      2. Stein DM,
      3. Wollstein G,
      4. et al
      . Macular segmentation with optical coherence tomography. Invest Ophthalmol Vis Sci 2005;46:2012–17.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Tan O,
      2. Li G,
      3. Lu AT,
      4. et al
      . Mapping of macular substructures with optical coherence tomography for glaucoma diagnosis. Ophthalmology 2008;115:949–56.
      OpenUrlCrossRefPubMedWeb of Science
    View Abstract

    Footnotes

    • I-TK and H-YLP contributed equally.

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval Ethics approval was provided by the Catholic University St Mary's Hospital institutional review board.

    • Provenance and peer review Not commissioned; externally peer reviewed.