To investigate the natural clinical course and the risk factors of visual field (VF) progression in untreated normal-tension glaucoma (NTG) with myopia with a minimum 3-year follow-up.
Retrospective cohort study.
Kaplan-Meier survival analysis and log-rank test were used to compare survival experiences between the groups. The hazard ratios (HRs) for the associations between potential risk factors and glaucomatous progression were determined using Cox proportional hazards model.
The average follow-up period for the included 97 patients was 71.1 ± 29.7 months and 32 eyes showed progression during the observation period. In NTG with myopia, Kaplan-Meier survival curve showed progression cases in 10.3% at 36 months and 24.8% at 60 months. When age and disc tilt direction were classified into 2 groups, a higher cumulative probability of progression was observed in those aged ≤50 years than >50 years ( P = .001) and in those with disc tilt direction <45 degrees than disc tilt direction ≥45 degrees ( P = .002). In Cox proportional hazards model, younger age, presence of disc hemorrhage, and more temporally tilted disc or parapapillary atrophy were associated with progression.
NTG with myopia seems to have a slow progression rate even without glaucoma medications. Only 10.3% and 24.8% of patients showed VF progression at 3 years and 5 years of follow-up in untreated NTG with myopia. To predict VF progression in untreated NTG with myopia, it seems that age and disc morphology should be considered.
Myopia has been known as a risk factor for glaucoma in previous studies. However, controversy remains regarding the role of myopia in glaucoma progression. In some reports, myopia was associated with progression. However, there have been a number of other studies published claiming that myopia is not associated with glaucomatous progression. In 1 previous study, open-angle glaucoma (OAG) with myopia showed no glaucomatous progression even without the use of intraocular pressure (IOP)-lowering medications over 7 years.
Several earlier reports have suggested possible risk factors for glaucomatous progression in myopic glaucoma. In particular, optic nerve head (ONH) morphologies such as disc tilt or parapapillary atrophy (PPA) were suggested to be associated with progression in glaucomatous eyes. Among the studies investigating the association between disc tilt (or PPA) and glaucomatous progression, there were conflicting results. Some reports showed that ONH deformation is associated with glaucomatous progression. Glaucomatous progression was also linked with ONH and characteristics of the surrounding structures such as disc tilt direction, greater β-zone PPA, and the region of largest β-zone PPA area. However, other reports suggested different results indicating that ONH deformation may be associated with nonprogression or that there were no associations between ONH deformation and progression.
Importantly, these previous studies mostly analyzed myopic glaucomatous eyes with different levels of IOP or different numbers of glaucoma medications. In that IOP is the most important risk factor of glaucoma including normal-tension glaucoma (NTG), different numbers of medications might have affected the previous study results. Thus, in the present study, we included patients with NTG with myopia who did not use IOP-lowering medications until glaucomatous visual field (VF) progression was confirmed. We investigated the natural course of NTG with myopia with a minimum 3 years follow-up, and evaluated whether or not ONH morphology is associated with progression in untreated NTG with myopia.
This was a retrospective longitudinal study. Candidates of this study were patients with untreated NTG with myopia, who initially visited Samsung Medical Center (Seoul, South Korea) between January 2005 and July 2015. Patients with NTG with myopia with a minimum 3 years follow-up who had been subjected to more than 5 VF tests were included. During the initial eye examination, if the eyes were myopic and within the normal IOP range, then the possible natural course of NTG with myopia was explained to all the patients. After that, based on the VF severity, the clinicians discussed whether or not to use initial IOP-lowering treatments with the patients. In general, if the NTG eyes with myopia had a VF defect of more than -6 dB of mean deviation (MD) value and the patients also agreed to undergo observation without IOP-lowering treatments, then the eyes were observed without IOP-lowering treatments until VF progression was confirmed. Though NTG eyes with myopia had a worse VF defect of less than -6 dB of MD, if the patients did not want to use medication, then the patients were observed without IOP-lowering treatment until VF progression was confirmed. Informed consent was obtained from all included patients. This study followed all guidelines for experimental investigation in human subjects, was approved by the Samsung Medical Center Institutional Review Board, and adhered to the tenets of the Declaration of Helsinki.
Participants underwent a comprehensive ophthalmic assessment, including slit-lamp biomicroscopy, Goldmann applanation tonometry, manifest refraction, gonioscopic examination, dilated stereoscopic examination of the ONH and color fundus photography, automated perimetry using the central 30-2 Humphrey Field Analyzer (HFA model 640 or model 740; Humphrey Instruments Inc, San Leandro, California, USA), axial length (AL) measurement (IOLMaster; Carl Zeiss Meditec Inc, Dublin, California, USA), and ultrasound pachymetry (Tomey SP-3000; Tomey Ltd, Nagoya, Japan). The extent of VF defect was measured using MD, pattern standard deviation (PSD), and VF index (VFI). Reliable VF analysis was defined as a false-negative rate of less than 15%, a false-positive rate of less than 15%, and a fixation loss of less than 20%.
IOP at baseline was defined as the average IOP of 2 consecutive visits in the absence of IOP-lowering medication usage. The average IOP was calculated as the average of all records of IOP during the observation period without IOP-lowering medication. A disc hemorrhage was defined as an isolated flame-shape or splinter-like hemorrhage on the optic disc or parapapillary area extending to the border of the optic disc. If a disc hemorrhage was found at any time point during the observation period, the eye was regarded as the eye with disc hemorrhage. Exclusion criteria included eyes with media opacities such as a corneal opacity, cataract, or vitreous opacity; systemic disease or ocular disease and eyes with ocular surgery history such as cataract surgery or vitrectomy that can affect VF test; and a baseline MD of -12 dB or less, for which glaucoma progression would be difficult to evaluate on VF test. One eye in each patient was selected for inclusion in this study. When NTG with myopia existed in both eyes, only 1 eye was randomly selected.
Definition of Myopic Normal-Tension Glaucoma, Visual Field Progression, and Pattern
The following criteria were required for the diagnosis of NTG with myopia. First, the eye should exhibit myopia with less than -0.5 diopters (D). Second, glaucomatous optic disc changes such as increased cupping (vertical cup-to-disc ratio >0.7), diffuse or focal neural rim thinning, disc hemorrhage, or retinal nerve fiber layer (RNFL) defects needed to be present at the time of diagnosis. Third, glaucomatous VF defects positive for at least 2 of the following criteria were required to have been present during more than 1 reliable test: (1) a cluster of 3 points with a probability less than 5% on the pattern deviation map in at least 1 hemifield, including at least 1 point with a probability less than 1% or a cluster of 2 points with a probability less than 1%; (2) a glaucoma hemifield test result outside the normal limits; or (3) a PSD of 95% outside the normal limits. Fourth, an open angle on gonioscopic examination must have been confirmed with no identified cause of secondary glaucoma present. Fifth, the baseline IOP should be under 21 mm Hg.
Progression of glaucomatous VF was defined using the Glaucoma Progression Analysis (GPA; Carl Zeiss Meditec Inc) software, which supplies an event-based progression analysis. In the event-based analysis, results were regarded as significant ( P < .05) when deterioration was evident on the pattern deviation probability maps at the same 3 or more points on 3 consecutive follow-up tests, and the GPA flagged it as “likely progression.” This was considered as a significant VF progression. Ordinary least squares linear regression was applied to determine the VF progression rate for the series of fields in dB/year.
The VF patterns were classified according to the VF defect location. When the VF defect was involved in only 1 hemifield, it was regarded as a superior or inferior VF defect based on the VF defect location. Conversely, when the VF defect was involved in both hemifields, it was regarded as a dual VF defect. The existence of paracentral scotoma was defined as a glaucomatous VF defect in 1 hemifield within 10 degrees of fixation, with at least 1 point at P < 1% lying at the 2 innermost paracentral points irrespective of VF abnormality outside the central 10 degrees. In case the glaucomatous VF defect was involved in both hemifields, only the hemifield with the dominant VF defect was used to determine the existence of paracentral scotoma so as to minimize false-positive cases.
Extent and Direction of Disc Tilt
Color fundus photography (Topcon, Paramus, New Jersey, USA) was obtained for measurements of the optic disc and β-zone PPA. The photographs were acquired after dilation of the pupil. Images were evaluated by 2 investigators (J.C.H. and S.H.H.) in a masked fashion using ImageJ software (version 1.52; National Institutes of Health, Bethesda, Maryland, USA). The averages of the values measured by the 2 investigators were used in the final analysis. The extent of disc tilt was defined as the ratio between the longest and shortest diameters of the optic disc. The direction of disc tilt was defined as the deviation of the long axis of the optic disc from the vertical meridian. The vertical meridian was defined as the vertical line 90 degrees from the line that connects the fovea and the center of the optic disc. The presence of β-zone PPA was defined as the region of chorioretinal atrophy with both visible sclera and choroidal vessels adjacent to the optic disc. The areas of β-zone PPA were measured using ImageJ. In the present study, the β-zone PPA-to-disc-area ratio was used instead of just the β-zone PPA to minimize the effects of photographic magnification error ( Figure 1 ).
The included cases were divided into several groups according to disc tilt direction. Because of the median value of disc tilt direction was close to 45 degrees, 45 degrees was used to stratify the cases into 2 groups. In addition, the first and the third quartiles of disc tilt direction (θ) were 20 degrees and 75 degrees, and the cases were arbitrarily classified into the 4 groups of NTG with myopia with temporally tilted (NTG TT ; θ ≤ 20 degrees), slightly tilted (NTG sIT ; 20 degrees < θ ≤45 degrees), moderately tilted (NTG mIT ; 45 degrees < θ ≤ 75 degrees), and highly inferiorly tilted disc (NTG hIT ; 75 degrees < θ), respectively.
Interobserver (J.C.H. and S.H.H.) reproducibility was assessed by the calculation of intraclass correlation coefficients (ICCs). An independent t test or Mann-Whitney U test was used to compare continuous variables between untreated NTG with myopia with and without progression after data normality was assessed. The χ 2 test or Fisher exact test were used to compare categorical data. A paired t test and Wilcoxon signed rank test were used to compare VF-related continuous variables between initial and final VF test. Separately, the exact symmetry test and McNemar test were used to compare categorical data between the initial and final VF tests. Continuous variables were expressed as means ± standard deviations (ranges). Kaplan-Meier analysis and the log-rank test were used to compare survival experiences (time-to-progression event) between the different groups. The hazard ratios (HRs) for the associations between potential risk factors and glaucomatous progression were determined using Cox proportional hazards modeling. Factors with P < .1 on univariate analysis were included in the multivariate Cox proportional hazards model. However, there were correlations between PPA direction and disc tilt direction; thus we used 2 multivariate logistic regression models to divide these parameters in different models. Adjusted HRs with 95% confidence intervals (CIs) were calculated. Statistical analyses were performed with SPSS software version 24.0 (SPSS, Inc, Chicago, Illinois, USA) and R statistical packages version 3.5.2 (available in the public domain, www.R-project.org ). A P value of less than .05 was considered statistically significant.
A total of 97 eyes in 97 patients with untreated NTG with myopia were included in the present study. Intraobserver ICC values were 0.992 (95% CI; 0.983-0.996) for tilt ratio, 0.976 (95% CI; 0.983-0.996) for tilt direction, 0.994 (95% CI; 0.987-0.997) for PPA size ratio, and 0.977 (95% CI; 0.952-0.989) for PPA direction. During observational periods, 32 eyes showed significant progression, while the other 65 eyes did not. No differences were observed regarding sex, baseline IOP, mean IOP, and central corneal thickness (CCT) between the progression and nonprogression groups. However, the progression group presented younger age, longer follow-up periods, less spherical equivalent, and faster progression rate than nonprogression group ( P = .007, P = .002, P = .031, and P = .040, respectively). In optic disc morphology, the nonprogression group demonstrated more inferior direction of disc tilt and PPA ( P = .018 and P = .048, respectively). Disc hemorrhage occurred more frequently in the progression group than in the nonprogression group ( P < .001) ( Table 1 ).
|Total (N = 97)||Progression (N = 32)||Nonprogression (N = 65)||P Value|
|Age at initial visit (years)||53.8 ± 13.7 (25-80)||47.8 ± 11.7 (25-67)||55.4 ± 14.0 (31-80)||.007* , a|
|Follow-up periods (months)||71.1 ± 29.7 (37-124)||84.0 ± 31.5 (37-124)||64.8 ± 26.8 (37-120)||.002* , a|
|Male sex, n (%)||46 (47.4)||15 (46.9)||31 (47.7)||.940 b|
|Baseline intraocular pressure (mm Hg)||15.2 ± 2.3 (10-20)||14.8 ± 2.5 (11-20)||15.3 ± 2.2 (10-20)||.354 a|
|Mean intraocular pressure (mm Hg)||15.1 ± 2.1 (10.2-20.0)||15.2 ± 2.0 (11.8-19.8)||15.1 ± 2.2 (10.2-20.0)||.838 a|
|Central corneal thickness (μm)||531.9 ± 30.5 (436-624)||535.1 ± 35.6 (436-624)||530.4 ± 27.8 (456-601)||.476 a|
|Spherical equivalent (D)||−3.5 ± 2.6 (−10.0 to −0.5)||−4.3 ± 2.9 (−10.0 to −0.5)||−3.1 ± 2.4 (−9.5 to −0.5)||.031* , a|
|Initial MD (dB)||−4.2 ± 3.6 (−11.8 to 2.0)||−4.6 ± 3.8 (−11.6 to 1.6)||−4.1 ± 3.5 (−11.8 to 2.0)||.517 a|
|Final MD (dB)||−5.0 ± 4.2 (−17.4 to 1.8)||−6.3 ± 4.6 (−15.8 to −1.2)||−4.4 ± 3.9 (−17.4 to 1.8)||.040* , a|
|Initial PSD (dB)||5.5 ± 4.1 (1.4-15.7)||6.0 ± 4.2 (1.5-14.4)||5.2 ± 4.0 (1.4-15.7)||.376 a|
|Final PSD (dB)||6.4 ± 4.2 (1.6-17.0)||7.9 ± 4.1 (2.2-15.0)||5.7 ± 4.1 (1.6-17.0)||.015* , a|
|Progression rate (dB/year)||−0.13 ± 0.34 (−1.2 to 0.6)||−0.23 ± 0.32 (−1.2 to 0.4)||−0.08 ± 0.34 (−1.1 to 0.6)||.040* , a|
|Optic disc morphology|
|Disc tilt direction (degrees)||48.1 ± 31.3 (−8.5 to 98.4)||37.5 ± 29.5 (−8.5 to 92.4)||53.3 ± 31.0 (−8.1 to 98.4)||.018* , a|
|Tilt ratio||1.4 ± 0.2 (1.1-2.3)||1.5 ± 0.3 (1.2-2.3)||1.4 ± 0.2 (1.1-2.1)||.199 a|
|PPA location (degrees)||45.6 ± 31.3 (−34.5 to 118.7)||36.6 ± 31.1 (−34.5 to 105.6)||50.0 ± 30.7 (−9.3 to 118.7)||.048* , a|
|PPA-to-disc-area ratio||1.1 ± 0.8 (0.1-4.3)||1.2 ± 0.8 (0.4-4.3)||1.0 ± 0.7 (0.1-4.3)||.335 a|
|Disc hemorrhage, n (%)||12 (12.4)||10 (31.3)||2 (3.1)||<.001* , b|
The follow-up period for the 97 untreated NTG myopia cases was 71.1 ± 29.7 months and the number of VF tests performed was 7.7 ± 3.2. The average MD value was -4.2 ± 3.6 dB at the initial visit and -5.0 ± 4.2 dB at the final visit ( P = .001). PSD was 5.5 ± 4.1 dB at the initial visit and 6.4 ± 4.2 dB at the final visit ( P < .001). The initial VF test demonstrated that there were 90 superior VF involvement (92.8%), 2 inferior VF involvement (2.1%) and 5 dual VF involvement (5.2%) cases. In the final VF test, VF defect pattern analysis revealed that 89 superior VF involvement (91.8%), 2 inferior VF involvement (2.1%) and 6 dual VF involvement (6.2%) cases existed. Paracentral scotoma was found in 40 eyes (41.2%) initially and 44 eyes (45.4%) finally, but there was no significant difference between the 2 groups. In the 32 NTG with myopia with progression cases, the follow-up period was 84.0 ± 31.5 months and the number of VF tests was 8.8 ± 3.7. MD value was -4.6 ± 3.8 dB initially and -6.3 ± 4.6 dB finally ( P < .001). Separately, PSD was 6.0 ± 4.2 dB initially and 7.9 ± 4.1 dB finally ( P < .001). The average time to-progression was 55.8 ± 28.5 months and the average age at the time of confirmed progression was 52.5 ± 12.8 years. In the initial VF test, VF pattern analysis revealed 28 superior VF involvement (87.5%), 2 inferior VF involvement (6.3%), and 2 dual VF involvement (6.3%). In the final VF test, VF defect pattern showed that there were 27 cases of superior VF involvement (84.4%), 2 inferior VF involvement (6.3%), and 3 dual VF involvement (9.4%). Paracentral scotoma was found in 14 eyes (43.8%) initially and 18 eyes (56.3%) finally. No significant differences were found between the initial and final VF defect patterns and the presence of paracentral scotoma even in the progression group ( Table 2 ).