Article Text

Download PDFPDF
Visual field defects and myopic macular degeneration in Singapore adults with high myopia
  1. Carla Lanca1,
  2. Chen Hsin Sun2,
  3. Rachel Chong1,3,4,5,
  4. Yee Ling Wong6,
  5. Monisha Esther Nongpiur1,3,4,
  6. Hla M Htoon1,4,
  7. Sahil Thakur1,3,
  8. Debra Q Y Quek4,7,
  9. Ching-Yu Cheng1,4,5,
  10. Quan V Hoang1,3,4,5,
  11. Charumathi Sabanayagam1,4,5,
  12. Seang-Mei Saw1,4,8,
  13. Chee Wai Wong1,3,4
  1. 1 Singapore Eye Research Institute, Singapore
  2. 2 Department of Ophthalmology, National University Hospital, Singapore
  3. 3 Singapore National Eye Centre, Singapore
  4. 4 Duke-NUS Medical School, Singapore
  5. 5 Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore
  6. 6 R&D Vision Sciences AMERA, Essilor International, Singapore
  7. 7 Ministry of Health Holdings Pte Ltd, Singapore
  8. 8 Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, Singapore
  1. Correspondence to Dr Chee Wai Wong, Singapore National Eye Centre, 168751, Singapore; wong.chee.wai{at}singhealth.com.sg

Abstract

Aims To characterise the association between visual field (VF) defects and myopic macular degeneration (MMD) in highly myopic adults without glaucoma.

Methods Participants (n=106; 181 eyes) with high myopia (HM; spherical equivalent ≤−5.0 D or axial length (AL) ≥26 mm), after excluding glaucoma and glaucoma suspects, from the Singapore Epidemiology of Eye Diseases-HM study were included in this cross-sectional study. Humphrey VF (central 24–2 threshold), cup-disc ratio (CDR) and intraocular pressure (IOP) measurements were performed. Mean deviation (MD) and pattern SD (PSD), VF defects (normal or abnormal; p<0.05 in ≥3 non-edge contiguous locations) and pattern (eg, generalised sensitivity loss) were analysed. MMD presence was diagnosed from fundus photographs. Generalised estimating equations were used for analysing factors (MD, PSD, VF defects, CDR and IOP) associated with MMD.

Results Mean age was 55.4±9.9 years and 51.9% were women (AL=26.7±1.1 mm). MMD eyes had lower MD (−3.8±2.9 dB vs −1.1±1.4 dB) and higher PSD (2.8±1.7 dB vs 1.7±0.6 dB). A higher percentage of MMD eyes (n=48) had abnormal VF (62.5% vs 28.6%; p<0.001) compared with no MMD (n=133 eyes). VF pattern in MMD eyes was significantly different from eyes without MMD (p=0.001) with greater generalised sensitivity loss (53.3% vs 10.5%) and arcuate defects (16.7% vs 10.5%). In multivariate analyses, MD (OR=1.52) and PSD (OR=1.67) were significantly (p=0.003) associated with MMD, but VF defects were not associated with MMD.

Conclusion Highly myopic adults with MMD may have VF loss when compared with highly myopic patients without MMD even in adults without glaucoma.

  • diagnostic tests/investigation
  • degeneration
  • epidemiology
  • field of vision

Data availability statement

All data relevant to the study are included in the article or available on reasonable request.

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

The global prevalence of myopia defined as spherical equivalent (SE) of −0.5 diopter (D) or less is increasing.1 Approximately 1406 million people (22.9% of the world population) are affected, of whom 163 million have high myopia (HM) defined as −5.0 D or less.1 Patients with HM are at risk of developing myopic macular degeneration (MMD), a sight-threatening disease and a major cause of visual impairment globally.2–5

Ophthalmologists frequently face a diagnostic dilemma due to the challenges of accurately detecting either MMD or glaucoma in highly myopic eyes with visual field (VF) loss. Data on the characteristics of VF defects due to MMD alone in non-glaucomatous eyes are scarce in the current literature.6 A recent case-control study7 has shown that patchy atrophy (n=32) in patients with HM without glaucoma may be an indicator of significant alterations in visual function, such as patchy scotoma and reduced retinal sensitivity in regions beyond the atrophic lesion on microperimetry. At present, it is difficult to attribute VF changes in HM eyes to underlying glaucomatous disease or MMD only. As such, glaucoma may be overdiagnosed in this group of patients, leading to unnecessary and lifelong use of glaucoma medications with potential side effects. Thus, there is a need for well-designed studies to clarify and differentiate VF changes due solely to MMD from VF changes due to glaucoma.

We hypothesise that MMD, a degenerative disease that predominantly affects the outer retina, will manifest visual defects and functional parameters that are distinct from those seen in glaucoma, a disease that affects the retinal nerve fibre layer. We aimed to characterise the association between VF defects and MMD in highly myopic Singapore adults without glaucoma.

Methods

Participants with highly myopic eyes, aged ≥ 35 years, were included in this cross-sectional study from the Singapore Epidemiology of Eye Disease-HM (SEED-HM) and the Singapore Prospective Study Programme (SEED, 2004–2011; SP2, 2004–2007) studies. Study methodologies for both cohorts, which comprise Chinese, Malay and Indian participants, have been previously reported.8–10

Inclusion and exclusion criteria

Eyes with the following conditions that could result in VF defects were excluded:

  1. Clinical glaucoma diagnosis or classification as glaucoma suspects, defined as the presence of any of the following: (a) intraocular pressure (IOP) >21 mm Hg; (b) vertical cup-disc ratio (CDR) >0.6 or vertical CDR asymmetry >0.2 measured using an eyepiece graticule, etched in 0.1 U; (c) anterior segment deposit consistent with pseudoexfoliation or pigment dispersion syndrome; (d) morphological features, such as disc haemorrhage, notching of the neuroretinal rim and defects of the retinal nerve fibre layer; (e) narrow anterior chamber angle during gonioscopy or (f) known history of glaucoma. Final identification, adjudication, and classification of glaucoma cases were reviewed by a glaucoma specialist.

  2. Cataract, diagnosed using the Wisconsin Cataract Grading System11 and defined as the presence of nuclear (grade ≥4), cortical (≥25% of total lens area), or posterior subcapsular (≥5% of total lens area) cataract.12

  3. Age-related macular degeneration (AMD), diabetic retinopathy (DR), retinal detachment or neurological diseases that affect the visual pathway. AMD diagnosis was performed using the Wisconsin Age-related Maculopathy grading system.13 Neovascular AMD included serous or haemorrhagic detachment of the retinal pigment epithelium (RPE) or sensory retina, and the presence of subretinal or sub-RPE haemorrhages or subretinal fibrous scar tissue. Specifically, neovascular AMD was diagnosed when there were differentiating features from myopic choroidal neovascularisation, including presence of drusen, thick choroid and presence of sub-RPE fluid or pigment epithelial detachment. The grading for DR was performed using a modification of the Airlie House classification system for the Early Treatment Diabetic Retinopathy Study.14

Participants with past cataract surgery or Lasik were not excluded as axial length (AL) measures were available.

Grading of MMD

After cycloplegia, colour fundus photographs centred on the optic disc and fovea were captured for each eye using standardised settings with a non-mydriatic retinal camera (Canon CR-DGi with 10D SLR back; Canon, Tokyo, Japan). The fundus photographs were graded using the International Photographic Classification and Grading System for Myopic Maculopathy (Meta-PM) protocol.15 Based on fundus photograph grading, an eye was considered to have MMD if Meta-PM category 2, 3, 4 or any ‘plus’ lesion, was observed.16 The presence of optic disc abnormalities (optic disc tilt, peripapillary atrophy (PPA) and peripapillary intrachoroidal cavitation) was also graded. Optic disc tilt was defined by an oval optic disc with a tilt ratio (minimum diameter to maximum diameter) of <0.75. PPA was defined using the classification by Curtin and Karlin.17 Peripapillary intrachoroidal cavitation was defined as an elevated, well-circumscribed, dome-shaped, yellow-orange lesion adjacent to the optic disc and PPA.

The fundus photos were graded by one of two trained graders. Adjudication was performed by a retinal specialist. Grading of pathological lesions by the retinal specialist and two trained graders were compared; the kappa statistics showed high intergrader agreement (k coefficient of 0.92). All graders were masked to the subjects’ characteristics.

Visual field examination

VF tests were performed using Humphrey (Carl Zeiss Meditec, Dublin, California, USA) central 24–2 threshold, white-on-white automated perimetry (Swedish interactive threshold algorism fast 24–2). All VF tests were conducted in a single dark room (ambient light <5 lux). Reliable VF results meeting the reliability criteria of ≤33% false positives, ≤33% false negatives and ≤33% fixation losses were included in the analysis. Unreliable fields were repeated. Diagnoses of all the VF reports were determined by a glaucoma specialist and classified into three major types: normal, enlarged blindness spot (at least two abnormal edge points around the blind spot) and abnormal. An abnormal VF was defined as having defect of ≥3 non-edge contiguous locations with p value <0.05 (at least one location with p<0.01) on the same side of the horizontal meridian in the pattern deviation plot, with a Glaucoma Hemifield Test outside normal limits. The types of defects were further graded according to the Ocular Hypertension Treatment Study VF criteria16: nasal step, partial arcuate, arcuate, altitudinal and paracentral defect with additional myopic-related defects adapted from a previous study6 including generalised (widespread) sensitivity loss, vertical step, central defect, scatter defect and rim artefacts. The magnitude of VF damage was defined by the mean deviation (MD) and pattern SD (PSD).

Refraction and biometry measures

Autorefraction was performed using an auto-refractometer (model RK5; Canon, Tochigiken, Japan). Refraction was then subjectively refined until the best-corrected visual acuity was obtained. The results from subjective refraction were used in the analysis. SE was defined as sphere plus half cylinder. AL was measured using non-contact partial coherence interferometry (IOL Master V.3.01; Carl Zeiss Meditec, Jena, Germany). HM was defined as eyes with SE ≤−5.0 D or AL ≥26 mm.

Other measures

Detailed interviewer-administered questionnaires were used to collect demographic (age, sex and ethnicity), educational level, general medical and ophthalmic history. Education level was classified as low (primary/below education) and high (secondary/above education).

Statistical analysis

To determine whether the VF defects were associated with the presence of MMD in highly myopic eyes, we compared VF defects in highly myopic eyes with MMD (n=48 eyes) and highly myopic eyes without MMD (n=133 eyes) after excluding adults with glaucoma. The outcome ‘MMD’ was analysed as a categorical variable (presence/absence). Of the risk factors, while VF was analysed as a categorical variable (normal and abnormal), other factors including MD, PSD, CDR and IOP were analysed as continuous variables (per unit change). Both eyes were included for analysis; thus, generalised estimating equations (GEE) was used for analysing factors (MD, PSD, VF defect, CDR, optic disc tilt, PPA and IOP) associated with MMD presence, adjusting for confounders such age, sex, race, education level, AL and correlation between fellow eyes, with manual backward stepwise approaches. P value ≤0.05 was considered statistically significant. All statistical analyses were carried out with IBM SPSS Statistics for Windows (V.26.0, IBM, Armonk, New York, USA).

Results

Of 215 eyes with HM, 34 were excluded due to the presence of ocular comorbidities that could result in VF defects. Of the 34 excluded eyes, 19 eyes had glaucoma or suspected glaucoma. Other exclusions were eyes with cataract (n=5), retinal detachment or prior vitreoretinal surgery (n=4), macular hole (n=1), optic neuritis (n=1) and unreliable VF (n=4). None of the eyes included in the study had peripapillary intrachoroidal cavitations or macular atrophy (Meta-PM category 4).

A total of 106 participants (181 eyes) with HM were included in the analysis. There were 51 (48.1%) men and 55 (51.9%) women. The majority of the participants had high educational level (93.4%) and were Chinese (57.5%). The mean age (±SD) was 55.4±9.9 years (range 37–76 years). The mean AL was 26.7±1.1 mm. The mean IOP was 14.4±2.9 mm Hg (range 7–21 mm Hg). From the 181 included eyes, 20 (11.0%) had Meta-PM category 0, 113 (62.4%) had Meta-PM category 1, 46 (25.4%) had Meta-PM category 2 and 2 (1.1%) had Meta-PM category 3. In 1 of the 46 eyes with Meta-PM category 2, lacquer cracks were present (‘plus’ lesion). Individuals with HM and MMD were significantly older (61.8±6.8 years, compared with 53.9±9.3 years; p<0.001), were less likely to be Chinese (18.3%, compared with 81.7%; p<0.001) and had longer AL (27.5±1.3 mm, compared with 26.5±0.8 mm; p<0.001), compared with individuals with HM without MMD.

A significantly higher proportion of abnormal VF was seen in highly myopic eyes with MMD (62.5%) than in those without MMD (28.6%, p<0.001). The overall VF defect was worse with significantly higher MD (more negative value) in highly myopic eyes with MMD (−3.8±2.9 dB), compared with highly myopic eyes without MMD (−1.1±1.4 dB, p<0.001; figure 1A). A similar trend was observed for the PSD (dB; best quantifies the amount of loss in localised VF defects) with significantly (p<0.001) higher values for the highly myopic eyes with MMD (2.8±1.7 dB) compared with highly myopic eyes without MMD (1.7±0.6 dB; figure 1B).

Figure 1

Box plot of visual fields (VF) results. (A) Mean deviation (dB) and (B) pattern SD (dB) of highly myopic eyes with and without myopic macular degeneration (MMD) from Singapore Epidemiology of Eye Disease-HM after excluding glaucoma and glaucoma suspects (n=106 individuals; 181 eyes).

MD was worse with higher Meta-PM categories (category 0=−0.9±1.2; category 1=−1.2±1.5; category 2=−3.6±2.8; category 3=−7.8±2.8; p for trend <0.001) as well as the PSD (category 0=1.7±0.9; category 1=1.7±0.6; category 2=2.7±1.5; category 3=5.2±4.9; p for trend <0.001), except when comparing category 0 with 1 where no significant differences were found in post hoc comparisons (all p=1.00). The two eyes with Meta-PM category 3 had abnormal VF, showing a general sensitivity loss.

A significantly higher proportion of eyes with MMD had optic disc tilt (43.8%) compared with eyes without MMD (20.3%; p=0.002). They also had significantly higher proportion of PPA (97.9%) compared with eyes without MMD (76.5%; p=0.001). However, other optic disc morphological features, such as CDR (MMD, no=0.5±0.1, yes=0.5±0.2; p=0.26), and IOP (MMD, no=14.4±2.9 mm Hg, yes=14.4±3.0 mm Hg; p=0.91) were not significantly different between the eyes with and without MMD.

The pattern of VF defects in highly myopic eyes with MMD was significantly different from eyes without MMD (overall p=0.001) with greater generalised sensitivity loss (53.3% vs 10.5%) and arcuate defects (16.7% vs 2.6%) in eyes with MMD (figure 2). In the six eyes with arcuate defects, five had MMD and partial arcuate defects. From the five eyes with arcuate defects, one had extensive PPA and the other had superior temporal PPA. The pattern of VF defects in highly myopic eyes with MMD was in concordance with more negative values for MD (dB), representing general sensitivity loss, and higher PSD (dB), which is associated with localised defects. Figure 3 shows examples of VF defects in highly myopic eyes with MMD.

Figure 2

Visual field patterns of highly myopic eyes with and without myopic macular degeneration (MMD) from the Singapore Epidemiology of Eye Disease-HM study after excluding glaucoma and glaucoma suspects (n=106 individuals; 181 eyes).

Figure 3

Examples of visual field patterns of highly myopic eyes with myopic macular degeneration (MMD) from Singapore Epidemiology of Eye Disease-HM study after excluding glaucoma and glaucoma suspects. (A) Visual fields showing a general sensitivity loss (mean deviation (MD)=−5.79 dB, p<1%; pattern SD (PSD)=1.98 dB, p<5%) and fundus photo showing the International Photographic Classification and Grading System for Myopic Maculopathy (Meta-PM) category 2 in the left eye (axial length of 27.33 mm). (B) Visual fields showing arcuate defect and enlarged blind spot (MD=−10.17 dB, p<0.5%; PSD=7.92, dB p<0.5%) and fundus photo showing with Meta-PM category 2 and lacquer cracks in the right eye (axial length of 29.61 mm). (C) Visual fields showing a general sensitivity loss (MD=−5.42 dB, p<1%; PSD=4.01 dB, p<0.5%) and fundus photo showing with Meta-PM category 2 in the left eye (axial length of 27.52 mm).

In the multivariable logistic regression GEE analysis, adjusting for age, sex, race, education level, AL and accounting for correlation between fellow eyes, higher VF loss (represented by negative MD (dB) values) was associated with MMD (adjusted OR=1.52; table 1). For every −1 dB increase in the MD (dB), the odds of MMD were 1.52. Higher PSD (dB) was also associated with MMD (adjusted OR=1.67, table 1). Optic disc tilt (adjusted OR=3.36) and PPA (adjusted OR=18.53) were associated with MMD. Overall presence of VF defect (p=0.12), CDR (p=0.52) and IOP (p=0.96) were not associated with MMD. Model fit was good with smaller values of corrected Quasi-likelihood under Independence Model Criterion (QICC) for MD (QICC=144.6) and PPA (QICC=145.5) indicating better fit. However, the highest QICC was 152.9 for PSD, suggesting similar fit for all models. A supplementary analysis using a false positive threshold of 15% (excluding five eyes) showed similar results for MD (adjusted OR=1.49; p=0.005), PSD (adjusted OR=1.61; p=0.020), optic disc tilt (adjusted OR=2.66; p=0.028), PPA (adjusted OR=14.78; p=0.010), presence of VF defect (p=0.27), CDR (p=0.29) and IOP (p=0.66; data not shown). We also reverted the outcome and found similar results as MMD presence was not associated with presence of VF defect (adjusted OR=2.27; p=0.06; data not shown).

Table 1

VFs, optic disc features and intraocular pressure of highly myopic eyes with and without MMD from the SEED-HM after excluding glaucoma and glaucoma suspects (n=106 individuals; 181 eyes)

Discussion

In our SEED-HM study of adults without glaucoma, we found that highly myopic eyes with MMD, had more severe VF losses than highly myopic eyes without MMD. CDR and IOP were not associated with MMD.

In our study, highly myopic eyes with MMD had more negative MD and higher PSD values. A small cross-sectional study from the Kaohsiung Chang Gung Memorial Hospital with patients aged 53.2±12.4 years (n=52 highly myopic eyes) have shown that eyes with patchy chorioretinal atrophy (Meta-PM category 3) without glaucoma have higher functional loss due to patchy scotomas and in regions beyond the atrophic lesions on microperimetry.7 The use of MD and PSD parameters, together with the pattern of VF defects may be important to determine the severity of visual function loss due to MMD and to determine timing of clinical intervention when novel therapies for MMD become available in the future. Using MD as a clinical measure of functional loss has its pitfalls, as the MD decreases with higher degree of myopia,6 18 even in the absence of MMD. Monitoring of highly myopic eyes using regular VF testing may be important to accurately depict any changes in VF patterns over time, as MMD may be a source of false glaucoma diagnosis even in the absence of raised IOP.

We also found that the pattern of VF defects in highly myopic eyes with MMD was distinct from what has been reported in eyes with glaucoma. This may suggest that the mechanisms that lead to VF defects in highly myopic eyes with MMD may be different from highly myopic eyes with glaucoma. The spatial differences in patterns of VF defect may be partially explained by focal injury affecting retinal ganglion cells in glaucoma, resulting in patterns of losses that follow optic nerve fibre bundle trajectories in the retina. Eyes with glaucoma commonly have a nasal step, followed by temporal sector defects in later stages which follows the spatial distribution of the affected retinal nerve fibre layers.19 Arcuate defects are also commonly seen in glaucoma, especially in those with moderate glaucoma with a greater predilection for the superior hemifield.19 In contrast, our study showed that non-glaucomatous HM eyes with MMD commonly presented with generalised sensitivity loss, which may reflect the mechanisms of MMD, a degenerative disease that predominantly affects the outer retina. However, early glaucoma can sometimes present generalised depression and this may occur before there are discrete nerve fibre-bundle defects, highlighting the importance of serial VF in the diagnosis of glaucoma, particularly in eyes with MMD.20 In a previous study, including young Chinese high myopes (n=487), the most common defects were enlarged blind spots and generalised reduction in sensitivity.6 In the same study, nasal step, early arcuate and advanced arcuate defects similar to glaucoma VF defects, were found in 16.1% of all field defects. In the same study, 2% of the eyes were further classified as high-risk glaucoma suspects and some of those defects may reflect undiagnosed glaucoma.

We also found that a relatively smaller proportion of non-glaucomatous HM eyes with MMD have arcuate defects. This pattern needs further clarification, as it may be due to (1) greater axial elongation resulting in thinning of the inner retina, that is, retinal ganglion cell layers or (2) disc tilt stress in the nerve fibre layer resulting in damage, related to an extension of the PPA, or (3) altered visual function due to changes in retinal curvature at the edges of a posterior staphyloma, where chorioretinal atrophy commonly occurs. The existence of two different patterns of VF loss in MMD may have important implications for the pathophysiology of the disease. Overall, due to overlap in the results of structural and functional testing seen in HM and glaucoma, differentiating between these two conditions can be challenging.

Further longitudinal studies are of paramount importance to understand how structural and functional tests can be used to assess and measure MMD progression, as documented progression may be important to decide on treatment strategies.

Strengths and limitations

The main strength of our study was the comparison of VF fields in adults with and without MMD, after excluding adults with glaucoma. However, our study was limited by its cross-sectional nature. We also acknowledge that we do not know whether some of the adults had, in fact, undiagnosed normal tension glaucoma or preglaucomatous lesions that can progress to glaucoma later. Thus, further follow-up is needed to confirm that the VF pattern loss is solely due to MMD alone in non-glaucomatous adults. Further studies with a longitudinal design, larger sample size and concurrent structural analyses such as measurements of peripapillary retinal nerve fibre thickness and/or macular ganglion cell-inner plexiform layer thickness are necessary.

Conclusion

In adults with HM and without glaucoma, individuals with MMD have greater visual function loss as measured by static automated perimetry when compared with individuals without MMD. We have identified the pattern of VF defects in eyes with MMD to be mainly generalised sensitivity loss. Thus, VF defects in older adults may be attributable to MMD and not glaucoma. Future longitudinal studies are important to further clarify the differences in VF defects attributed to MMD so as to facilitate and improve the clinical diagnosis of MMD and glaucoma in adults with VF loss.

Data availability statement

All data relevant to the study are included in the article or available on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

The ethics approval was obtained from the Singapore Eye Research Institute Institutional Review Board (CIRB Ref. No.: 2017/2454) and the study adhered to the Declaration of Helsinki. Written informed consent was obtained from all participants.

References

Footnotes

  • CL and CHS are joint first authors.

  • Correction notice This paper has been updated since it was first published. The authors noticed an error in the legend of figure 3 and this has now been corrected.

  • Contributors CL and CHS contributed equally to this work. S-MS and CWW contributed to the conception and design of the study. CL, CHS and HMH performed the data analyses. CL and CHS wrote the manuscript. Manuscript revision: RC, YLW, MEN, ST, DQYQ, C-YC, QVH and CS. Final approval of the manuscript: all authors.

  • Funding This work was supported by National Medical Research Council grant numbers 0796/2003, IRG07nov013, IRG09nov014, STaR/0003/2008, CG/SERI/2010 and CSA-MOH-000151.

  • Competing interests YLW is an employee of Essilor International, Singapore.

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

Linked Articles

  • Highlights from this issue
    Frank Larkin