- Email: [email protected]
Influence of the Extent of Myopia on the Progression of Normal-Tension Glaucoma
Influence of the Extent of Myopia on the Progression of Normal-Tension Glaucoma SAE WOON SOHN, JAE SEOK SONG, AND CHANGWON KEE ● PURPOSE:
To evaluate the influence of the extent of myopia on the progression rate of normal-tension glaucoma (NTG). ● DESIGN: Retrospective, observational case series. ● METHODS: One hundred forty-three eyes of 143 patients with NTG who were treated from 1994 through 2006 and followed up with standard automated perimetry were evaluated in this study. The participants were divided into 4 groups: mild myopia (ⴚ0.76 to ⴚ2.99 diopters [D]), moderate myopia (ⴚ3 to ⴚ5.99 D), severe myopia (ⴚ6 D or less), and nonmyopia (emmetropia and hyperopia, ⴚ0.75 D or more) groups. The change in mean deviation, corrected pattern standard deviation, mean thresholds of 10 zones corresponding to the glaucoma hemifield test, and thresholds of 52 points of the nonmyopia group were compared with those of the other myopia groups. Additionally, we controlled each analysis for age and posttherapeutic intraocular pressure to preclude the possibility of these covariates influencing the analysis of the effect of myopia on the progression of glaucoma. ● RESULTS: There was no statistically significant difference between the nonmyopia group and each of the myopia groups in terms of mean deviation, corrected pattern standard deviation, mean thresholds of 10 zones corresponding to the glaucoma hemifield test, and the thresholds of 52 point changes against refraction. Moreover, with the control of the other covariates (age and posttherapeutic intraocular pressure), no statistically significant differences were noted (multivariate analysis using mixed model, P > .1). ● CONCLUSIONS: Although a high incidence of openangle glaucoma among myopic patients has been reported previously, myopia did not influence the progression rate of NTG after treatment. (Am J Ophthalmol 2010; 149:831– 838. © 2010 by Elsevier Inc. All rights reserved.)
Accepted for publication Dec 18, 2009. From the Department of Ophthalmology, Gangneung Asan Hospital, Ulsan University College of Medicine, Gangneung, Korea (S.W.S.); the Department of Preventive medicine, Kwandong University College of medicine, Gangneung, Korea (J.S.S.); and the Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea (C.K.). Inquiries to Changwon Kee, Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul 135-710, Korea; e-mail: [email protected] 0002-9394/10/$36.00 doi:10.1016/j.ajo.2009.12.033
©
2010 BY
T
HE CORRELATION BETWEEN MYOPIA AND GLAUCOMA
has been a focus of interest in glaucoma research for a long time. Several studies already have proved the relatively high incidence rate of glaucoma in myopic patients. The Beaver Dam Eye Study1 reported that the incidence rate of glaucoma in myopic patients was 1.6 times higher than in those with emmetropia. The Malmo Eye Study2 reported that glaucoma developed in 1.5% of individuals with myopia and in 0.9% of those with emmetropia. Thus, several explanations are possible. One of them asserts that myopia includes an anatomic weakness of the disc, which manifests as increased cup size,3 unusually large or skewed canal shape,4 – 6 and thin sclera and lamina in the disc periphery.7 Furthermore, reduced blood flow8 and low ocular pulse amplitude9 –11 have been reported in myopic eyes. For these reasons, the optic disc in myopic patients may be subject to ischemia or damage at any level of intraocular pressure (IOP). Although it has been demonstrated previously that myopia is associated with increased incidence of glaucoma, the effect of myopia on the progression of glaucoma has yet to be elucidated. Quigley and associates12 and Perkins and Phelps13 reported that visual field (VF) defects were observed more frequently in myopic patients with untreated ocular hypertension. Chihara and Sawada14 reported that myopia was a risk factor in the progression of POAG. However, NouriMahdavi and associates15 and Phelps16 reported that myopia did not affect the progression of primary open-angle glaucoma (POAG) under treatment. These previous studies simply assessed the progression or nonprogression of glaucoma, considering myopia as one of various risk factors. Additionally, no studies have been conducted yet regarding the progression rate of glaucoma in association with the degree of myopia, nor have any studies been conducted with the other risk factors being controlled. This study particularly evaluated the effect of myopia on the progression rate of normal-tension glaucoma (NTG) in an effort to answer the question as to whether myopia should be considered in the treatment or prognosis of NTG.
METHODS ● STUDY SUBJECTS:
Among 874 patients diagnosed with NTG at the Department of Ophthalmology of the Samsung Medical Center from 1994 through 2006 and who
ELSEVIER INC. ALL
RIGHTS RESERVED.
831
were treated only with a topical hypotensive agent, 143 ultimately were selected. Inclusion criteria. In a retrospective study involving the records of 874 patients, 198 were selected for study based on the inclusion criteria described below. A diagnosis of NTG was made when a patient with an IOP of 21 mm Hg or less without treatment had findings of glaucomatous optic disc damage and corresponding VF defects, an open angle observed by gonioscopic examination, and no underlying cause for the optic disc damage aside from glaucoma. Each patient had to have been followed up in the outpatient clinic for 5 years or more and to have a VF test twice yearly, for a total of 10 or more VF tests. In patients with bilateral NTG, one eye was selected randomly and analyzed. Among 874 total subjects, 198 subjects were considered eligible.
FIGURE 1. Diagram showing the definition of 10 zones of the glaucoma hemifield test (GHT) and 52 points.
Exclusion criteria. Among the 198 selected patients, we further excluded any patient with (1) another ocular disorder (11 patients), (2) any history of ocular surgery and laser including cataract surgery (21 patients), (3) lens opacities more severe than C2,N2,P2 according to lens opacities classification system III criteria17 (13 patients), and (4) visual acuity of less than 20/40 (10 patients). Besides cataract formation, we were particularly careful to consider factors such as large peripapillary atrophy and myopic maculopathy, both of which are common in patients with high myopia and can influence VF. We performed a careful fundus examination, stereo disc photography, retinal nerve fiber layer photos, and optical coherence tomography semiannually to observe for any myopic maculopathy or large peripapillary atrophy. If either of these two conditions developed, the subject then was excluded and these 4 patients were included in the fourth exclusion criterion, visual acuity of less than 20/40 (10 patients).
Grading. Refractive status was measured via a manifest refraction test using a spherical equivalent. Emmetropia was defined as between ⫺0.75 and 0.75 diopters (D) and hyperopia as 0.76 D or more. Myopia was classified into mild myopia (⫺0.76 to ⫺2.99 D), moderate myopia (⫺3 to ⫺5.99 D), and severe myopia (⫺6 D or less).1 ● VISUAL FIELD TESTING: The Central 30-2 full-threshold program of the Humphrey Visual Field Analyzer (Carl Zeiss Ophthalmic System, Inc., Dublin, California, USA) was used for VF testing. A reliable VF had to fulfill 3 criteria: fixation loss less than 20%, a false-positive rate of 30% or less, and a false-negative rate of 30% or less. The results of the first 2 tests were excluded to remove the learning effect from the final analysis. Additionally, those with a mean deviation (MD) of ⫺25 dB or less, for whom glaucoma progression would be difficult to evaluate from a VF test, also were excluded. The corrected pattern standard deviation (CPSD) with mean deviation (MD) of ⫺12 dB or less was excluded from the analysis, because the CPSD may be decreased when the glaucoma progresses to severe status.18,19 VF data for left eyes were switched across the vertical meridian such that, for example, the nasal area for the left eyes corresponded with the nasal area for the right eyes in all analyses. Hence, all figures of VF are presented as right eyes, although they actually represent mixed results from right and left eyes. Using these VF results, the glaucoma progression rate and progression or nonprogression in myopic groups were analyzed statistically in comparison with those of the nonmyopic groups. The determination of time to progression is based on glaucoma change probability maps (GCPMs) of total deviation (TD), which is included in the STATpac software package of the Humphrey Visual Field Analyzer. To detect VF progression, all follow-up results of the VF tests were compared with an average of 2 baseline VF tests from the same eye using GCPMs. Definite VF progression was defined as at least 3 test points exhibiting significant
Treatment. When NTG was diagnosed, the patients received betaxolol 5 mg/mL (Alcon, Inc, Fort Worth, Texas, USA) twice daily. During the follow-up, if glaucoma progressed or did not reach the target IOP, betaxolol was replaced by latanoprost 50 g/mL (Pfizer, Inc., New York, New York, USA). The first target of IOP reduction was to reduce 30% of basal IOP. But if it did not reach its first target IOP, 14 mm Hg was set for the second target IOP. When latanoprost did not have sufficient effect, brimonidine 2 mg/mL (Allergan, Inc., Irvine, California, USA) usually was added. A Goldmann applanation tonometer was used for IOP measurement. The basal IOP was calculated as the mean IOP of each hospital visit before the administration of a topical hypotensive agent. Additionally, the posttherapeutic IOP was calculated as the mean IOP of each hospital visit after the administration of a topical hypotensive agent. 832
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MAY 2010
TABLE 1. Characteristics of Patients with Normal-Tension Glaucoma in Each Myopic and Nonmyopic Group
Characteristics
Nonmyopic Group (Hyperopia and Emmetropia ⫺0.75 D or more; n ⫽ 75)
Mild Myopia Group (⫺0.76 to ⫺2.99 D; n ⫽ 22)
Moderate Myopia Group (⫺3 to ⫺5.99 D; n ⫽ 22)
Severe Myopic Group (⫺6 D or less; n ⫽ 24)
Age Sex (male/female) Refraction (D) Basal IOP (mm Hg) Posttherapeutic IOP (mm Hg) Ratio of IOP lowering (%) No. of examinations Basal MD (dB) Basal CPSD (dB) Follow-up (mos) Diabetes mellitus (n)b Hypertension (n)c
61.8 ⫾ 8.9 35/40 0.39 ⫾ 0.75 16.1 ⫾ 2.4 14.4 ⫾ 1.9 13.2 ⫾ 7.3 8.4 ⫾ 3.1 ⫺6.7 ⫾ 4.9 6.9 ⫾ 4.1 67.4 ⫾ 21.6 10 13
59.7 ⫾ 10.8 8/14 ⫺1.92 ⫾ 0.55 17.7 ⫾ 2.5 14.8 ⫾ 2.0 14.8 ⫾ 8.4 8 ⫾ 2.7 ⫺6.7 ⫾ 4.1 7.5 ⫾ 3.6 63.8 ⫾ 23.5 5 3
65.5 ⫾ 10.5 7/15 ⫺4.57 ⫾ 0.76 17.1 ⫾ 2.9 15.1 ⫾ 1.8 14.2 ⫾ 8.9 9.2 ⫾ 3.6 ⫺7.1 ⫾ 5.0 8.5 ⫾ 4.9 67.7 ⫾ 25.0 3 4
60.7 ⫾ 8.6 17/7 ⫺9.37 ⫾ 0.43 16.9 ⫾ 3.2 15.0 ⫾ 1.9 13.6 ⫾ 6.8 9.5 ⫾ 3.6 ⫺7.1 ⫾ 5.4 7.1 ⫾ 3.6 69.4 ⫾ 24.3 3 4
P Valuea
.624 .230 0 .131 .419 .517 .307 .665 .524 .865 .957 .975
CPSD ⫽ corrected pattern standard deviation; D ⫽ diopters; IOP ⫽ intraocular pressure; MD ⫽ mean deviation. Data are presented as number and mean ⫾ standard deviation. a Analysis of variance. b Medical history. c Defined as systolic pressure more than 160 mm Hg, diastolic pressure more than 95 mm Hg, or medical history.
TABLE 2. Progression Rate of Mean Deviation and Corrected Pattern Standard Deviation in Patients with Normal-Tension Glaucoma in Each Myopic and Nonmyopic Group
MD (dB/year) CPSD (dB/year)
Nonmyopic Group
Mild Myopia Group
Moderate Myopia Group
Severe Myopia Group
P Valuea
⫺1.334 ⫾ 1.003 0.967 ⫾ 0.48
⫺1.055 ⫾ 0.328 1.525 ⫾ 0.7
⫺1.113 ⫾ 0.466 1.022 ⫾ 0.126
⫺0.912 ⫾ 0.313 1.259 ⫾ 0.748
.255 .106
CPSD ⫽ corrected pattern standard deviation; MD ⫽ mean deviation. Data are presented as mean ⫾ standard deviation. a P value based on multivariate analysis using a mixed model.
(P ⬍ .05) progression at the same locations on 3 consecutive tests as compared with the baseline.20 ● PROGRESSION RATE ANALYSIS:
Univariate analysis. The first method used for comparisons of the progression rate was univariate analysis, using linear regression, which concerned only the effects of myopia. The progression rate of each myopic and nonmyopic group was calculated by linear regression. The methods of univariate analysis were as follows. VF data (MD, CPSD, the thresholds of 52 points) were transferred to a personal computer, and univariate linear regression was conducted with time as the independent variable. Regression analysis was conducted with respect to 2 global indices (MD and CPSD), and similar analysis was conducted for the mean thresholds of the 10 zones corresponding to those of the glaucoma hemifield test and the thresholds of the 52 test points (Figure 1). Sixty-four linear regressions (1(MD) ⫹ 1(CPSD) ⫹ 10(zones) ⫹ 52(points) ⫽ 64) were conducted for each individual.
VOL. 149, NO. 5
MYOPIA
ON THE
Among them, the P ⱕ .025 for the slope of the MD and CPSD, P ⱕ .005 for the slope of 10 zones, and P ⱕ .001 for the slope of 52 points were considered to be statistically significant. Only such significant values were selected for the calculation of the mean values of the progression rate.21,22 Multivariate analysis. The second method used for the comparison of the progression rate was a multivariate analysis using a mixed model. Each progression rate of MD, CPSD, 10 zones, and 52 points over time was compared by multivariate analysis using the mixed model between the myopia and nonmyopia groups. Multivariate analysis was used to evaluate the association of the continuous dependent variable of the degree of myopia (each myopic and non-myopic groups) with the binary independent variable of progression rate. Additionally, covariates including age, sex, diabetes mellitus (DM), hypertension (HTN), basal IOP, postNTG PROGRESSION
833
FIGURE 2. Diagrams showing the progression rate of 10 zones in patients with normal-tension glaucoma in each myopic and nonmyopic group: (Top left) nonmyopic group, (Top right) mild myopia group, (Middle left) moderate myopia group, and (Middle right) severe myopia group. Data are presented as mean (dB/year). (Bottom) *P value based on multivariate analysis using a mixed model that controlled age and posttherapeutic intraocular pressure.
therapeutic IOP, basal MD, and basal CPSD were controlled using various covariate combinations. Thus, 64 (1(MD) ⫹ 1(CPSD) ⫹ 10(zones) ⫹ 52(point)) times of multivariate analyses using a mixed model were conducted with each covariate combination; therefore, more than hundreds of analyses were conducted in total.
the aforementioned progression rate. The decision-making process for progression is the same as explained above, using GCPMs of TD. The Student t test was used to analyze the degree of myopia between the progression and nonprogression groups, and a chi-square test was used to analyze the differences in the frequency of progression between the myopic and nonmyopic groups.
Survival analysis. The third method used for comparisons of the progression rate was Kaplan-Meier survival analysis of time to progression. The process of the progression decision was the same as the above explanations using GCPMs of TD.
Multivariate analysis. Additionally, using logistic regression, the association of VF progression with myopia was evaluated, and the influences of age and posttherapeutic IOP were explored. All statistical analyses were conducted using SAS software version 8.0 (SAS, Inc., Cary, North Carolina, USA) and SPSS software version 11.0 (SPSS, Inc., Chicago, Illinois, USA).
● PROGRESSION OR NONPROGRESSION ANALYSIS:
Univariate analysis. The final method used in this study was used to compare progression or nonprogression, and not
834
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MAY 2010
FIGURE 3. Diagrams showing the progression rate of 52 points in patients with normal-tension glaucoma in each myopic and nonmyopic group: (Top left) nonmyopic group, (Top right) mild myopia group, (Middle left) moderate myopia group, and (Middle right) severe myopia group. Data are presented as mean (dB/year). (Bottom) *P value based on multivariate analysis using a mixed model that controlled for age and posttherapeutic intraocular pressure.
RESULTS ● GENERAL ANALYSIS: Among 874 patients, 143 patients and 143 eyes (67 from males and 76 from females) fulfilled the inclusion criteria of this study. With regard to the classification of refractive error, (1) 22 eyes were categorized with mild myopia (⫺0.76 to ⫺2.99 D), (2) 22 eyes with moderate myopia (⫺3 to ⫺5.99 D), and (3) 24 eyes with severe myopia (⫺6 D or less). In the 75 nonmyopic eyes (⫺0.75 D or more), (1) 51 eyes were categorized with emmetropia (⫺0.75 to 0.75 D) and (2) 24 eyes with hyperopia (0.76 D or more). Age, refraction, basal IOP, posttherapeutic IOP, frequency of VF testing, basal MD, basal CPSD, and follow-up period for each group are listed in Table 1. Except for the refraction value, there were no significant differences in any variable used in comparisons between the nonmyopic group and any of the myopic groups. ● PROGRESSION RATE ANALYSIS:
Univariate and multivariate analysis. Changes in the linear regression values for MD for each group were as follows: ⫺1.334 dB/year for the
VOL. 149, NO. 5
MYOPIA
ON THE
nonmyopic group and ⫺1.055 dB/year, ⫺1.113 dB/year, and ⫺0.912 dB/year for the mild, moderate, and severe myopic groups, respectively. Results of multivariate analysis using the mixed model, in which age and posttherapeutic IOP were controlled, revealed no statistically significant differences between groups (P ⫽ .255). The change in the CPSD by linear regression analysis in the nonmyopic group was 0.967 dB/year and 1.525 dB/ year, 1.022 dB/year, and 1.259 dB/year for the mild, moderate, and severe myopic groups, respectively. There was no significant between-group differences by multivariate analysis using the mixed model (P ⫽ .106; Table 2). Comparisons of the 10 zones yielded the following results (Figure 2). There were no significant differences by multivariate analysis using the mixed model among 9 areas; zone 6 was the exception. In zone 6, the nonmyopic group lost ⫺0.613 dB/year and the mild, moderate, and severe myopic groups lost ⫺0.493 dB/year, ⫺0.656 dB/ year, and ⫺0.959 dB/year, respectively. In zone 6, there were significant differences as the result of multivariate analysis (P ⫽ .011) using the mixed model between the NTG PROGRESSION
835
TABLE 3. Multivariate Risk Factors Associated with Visual Field Progression in Patients with Normal-Tension Glaucoma Factors
Odds Ratio (95% CI)
P Valuea
Myopia (1-D increase) Age (1-year increase) Posttherapeutic IOP (1-mm Hg increase)
1.34 (0.452 to 3.667) 1.771 (0.465 to 4.129)
.458 .576
1.316 (0.782 to 5.796)
.461
CI ⫽ confidence interval; D ⫽ diopters; IOP ⫽ intraocular pressure. a P value based on logistic regression analysis.
7.0 months, respectively, and did not differ significantly from that of the nonmyopic group (P ⫽ .903, .672, respectively, log-rank test; Figure 4).
FIGURE 4. Kaplan-Meier curve showing the cumulative probability of visual field progression in patients with normaltension glaucoma in each myopic and nonmyopic group. The progression periods for the mild, moderate, and severe myopia group were 78.8 ⴞ 9.0 months, 73.4 ⴞ 9.0 months, and 76.1 ⴞ 7.0 months, respectively, and did not differ significantly from that of the nonmyopic group: 76.6 ⴞ 5.1 months (P ⴝ .882, .903, .672, respectively, log-rank test).
● PROGRESSION OR NONPROGRESSION ANALYSIS:
Univariate analysis. Among the 143 eyes, 79 showed progression. The mean refraction value of these 79 eyes was ⫺2.4 ⫾ 4.3 D. For the 64 eyes in which progression was not noted, the value was ⫺2.2 ⫾ 3.8 D. These figures were not statistically significantly different (P ⫽ .765, Student t test). With regard to exacerbation in each myopic group, 42 of the 75 nonmyopic eyes (56%) evidenced progression. Progression was detected in 11 of 22 mild myopia eyes (50%), 10 of 22 moderate myopia eyes (45.5%), and 10 of 24 severe myopia eyes (41.7%). There was no significant differences in progression between the nonmyopic and each myopic group (P ⫽ .599, chi-square test).
nonmyopic and each myopic group, in which both age and posttherapeutic IOP were controlled. There were no statistically significant differences between groups for each of the 52 points by multivariate analysis using the mixed model (Figure 3). Covariates such as age, sex, DM, HTN, basal IOP, posttherapeutic IOP, basal MD, and basal CPSD were controlled in the multivariate analysis using mixed method. Although we controlled all 8 of the covariates (age, sex, DM, HTN, basal IOP, posttherapeutic IOP, basal MD, basal CPSD) and any 8 covariate combinations, the results were not significant. However, because of the small number of the study subjects (143 patients) and the negative results, we selected the 2 most important factors, age and posttherapeutic IOP, and subsequently analyzed these. Zone 6 proved significant under the control of the age and posttherapeutic IOP combination only, but was not significant in the control of other factors (sex, DM, HTN, basal IOP, basal MD, basal CPSD).
Multivariate analysis. Logistic regression analysis was used to seek a possible correlation between myopia and VF progression. The results showed that the myopic patients evidenced VF progression of 1.34-fold that of the nonmyopic group. However, there were no statistically significant differences between the groups (P ⫽ .458, logistic regression analysis; Table 3).
DISCUSSION AS MENTIONED ABOVE, MANY PREVIOUS STUDIES HAVE
assessed the effects of myopia on the incidence and progression or nonprogression of POAG. However, this study focused on the effects of myopia on the progression rate of NTG, which has not been elucidated previously. The reasons for analyzing the progression rate are as follows. First, the results of previous studies regarding progression or nonprogression proved controversial. Thus, we expected to detect progression or nonprogression indirectly, via progression rate analysis. Second, the progression rate was our main concern and focus of interest because the progression rate can be more rapid than
Survival analysis. Disease progression was interpreted with GCPMs.20 In the survival analysis of VF progression for the nonmyopic and mild myopia groups using the Kaplan-Meier life table method, the progression periods were 76.6 ⫾ 5.1 months and 78.8 ⫾ 9.0 months, respectively. The differences between the 2 groups were not statistically significant (P ⫽ .882, log-rank test). The progression periods of the moderate myopia group and the severe myopia group were 73.4 ⫾ 9.0 months and 76.1 ⫾ 836
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MAY 2010
usual, because of anatomic and hemodynamic weaknesses of the myopic disc. In particular, this study assessed NTG only because the effects of IOP would be relatively smaller than those of POAG. Additionally, the prevalence of NTG in Asian countries such as Korea and Japan is quite high, and some recent studies also have reported a relatively higher prevalence of myopia in that region.23 The results of this study demonstrated that, except for the progression rate of the sixth zone, myopia did not affect either the progression or nonprogression or the progression rate of NTG. However, much more attention should be paid to the analysis of these negative results. This is because not just one factor, but many complicated factors, may be associated with the progression or nonprogression and the progression rate of NTG. Thus, we conducted analyses via 4 different methods, including multivariate analysis, in an effort to clarify these negative results. The reason that we analyzed a number of results (MD, CPSD, 10 zones, and 52 points) of VF is that the NTG is known to progress commonly in certain specialized regions (e.g., paracentral or nasal defects) as well as in generalized VF change. Therefore, we also conducted analyses of 10 zones and 52 points, because we thought that global index analysis only, such as MD and CPSD, were not appropriate. In the sixth zone of the glaucoma hemifield test, the progression rate was ⫺0.613 dB/year in the nonmyopic group and ⫺0.493 dB/year, ⫺0.659 dB/year, and ⫺0.959 dB/year, respectively, in the mild, moderate, and high myopia groups. There were significant differences (P ⫽ .011) in the multivariate analysis using a mixed model that controlled for age and posttherapeutic IOP. However, each point of nos. 30, 31, and 32 comprising zone 6 were not significant. The value of zone 6 was calculated as the average of these 3 points (nos. 30, 31, and 32). Because zone 6 may continue to progress even after the value of any of these 3 points (nos. 30, 31, and 32) falls to 0, the value of an individual point is considered to be more useful than that of the zone. Also, at the univariate analysis, zone 6 came out to be meaningless (P ⫽ .124, analysis of variance), and at the multivariate analysis, it had difference only under control of age and posttherapeutic IOP only. Therefore, zone 6, which does not match individual points, is not considered to be meaningful. The limitations of this study are as follows. First, it was basically a retrospective study, and second, the selection of subjects was limited. With regard to subject selection, cases of extremely advanced glaucoma (MD ⱕ ⫺25 dB) and any history of ocular surgery, including cataract surgery, were excluded from this study. Additionally, this study used only subjects who had been followed-up for 5 years or more and
who had undergone VF tests at least twice yearly, for a total of 10 VF tests or more. As a result, only 143 subjects ultimately were enrolled, and some patients without progression were lost to follow-up observations. Moreover, because the study was conducted in a tertiary medical care institution, rather than being conducted with new patients, many patients had been referred from primary and secondary medical institutions with continuous progression, despite having been treated. In fact, the change in MD in this study occurred more rapidly than in the Early Manifest Glaucoma Trial (EMGT) treatment group (⫺0.03 dB/month)24 and the Collaborative Normal Tension Glaucoma Study (⫺0.4992 dB/year),25 both of which were prospective studies. Additionally, the progression or nonprogression rate in this study was higher than the rate reported in the EMGT. The results of this study based on the GCPMs also revealed a high rate of progression (52%) even after treatment, whereas the EMGT study reported a rate of 45%24 and the Collaborative Normal Tension Glaucoma Study reported a rate of 18%.25 However, the EMGT used GCPMs of pattern deviation, which are known to be more conservative than our GCPMs of TD.26,27 In addition, the basal MD in the EMGT was ⫺5.0 ⫾ 3.7 dB, which is a less profound progression than found in our study, and the EMGT excluded MD ⱕ ⫺16 dB, which was defined as advanced glaucoma. However, in our study, MD ⱕ ⫺25 dB was defined as advanced glaucoma, and thus excluded. Therefore, simple numerical comparisons of these studies would be unreasonable. The Collaborative Normal Tension Glaucoma Study also differs from our study in the definition of progression and the subject selection process. Finally, our study used only the results of VF to judge progression. This made it easy to compare the results of VF statistically, because the results of the VF test are numerical. Because the subjects of the study were patients with myopic changes such as tilted discs and large peripapillary atrophies, other standards such as disc, retinal nerve fiber layer, and so forth could not be used to determine progression. Yet, the use of VF tests alone may prove insufficient to determine progression. In conclusion, the results of our study showed that myopia did not affect the progression rate or progression or nonprogression in NTG patients who had begun treatment. Our conclusions suggest that, in treated NTG patients, the myopic element is not the principal issue for treatment planning or prognosis. Therefore, regardless of the various limitations of this study, its results are useful and relevant. In the future, more prospective research will be necessary to expand on the results of this study.
THE AUTHORS INDICATE NO FINANCIAL SUPPORT OR FINANCIAL CONFLICT OF INTEREST. INVOLVED IN DESIGN OF THE STUDY (C.K.); Conduct of the study (S.S.W., J.S.S., C.K.); Collection and management of data (S.S.W., J.S.S., C.K.); Analysis and interpretation of data (S.S.W., J.S.S., C.K.); Preparation and writing of the manuscript (S.S.W.); Review and final approval of the manuscript (C.K.). This research was approved by Samsung Medical Center Institutional Review Board. An exemption from informed consent for research was granted because this was a retrospective research study.
VOL. 149, NO. 5
MYOPIA
ON THE
NTG PROGRESSION
837
REFERENCES 16.
1. Wong TY, Klein BK, Klein R, Knudtson M, Lee KE. Refractive errors, intraocular pressure, and glaucoma in a white population. Ophthalmology 2003;110:211–217. 2. Grodum K, Heijl A, Bengtsson B. Refractive error and glaucoma. Acta Ophthalmol Scand 2001;79:560 –566. 3. Tomlinson A, Phillips CI. Ratio of optic cup to optic disc. In relation to axial length of eyeball and refraction. Br J Ophthalmol 1969;53:765–768. 4. Bellezza AJ, Hart RT, Burgoyne CF. The optic nerve head as a biomechanical structure: initial finite element modeling. Invest Ophthalmol Vis Sci 2000;41:2991–3000. 5. Chihara E, Sawada A. Atypical nerve fiber layer defects in high myopes with high-tension glaucoma. Arch Ophthalmol 1990;108:228 –232. 6. Dichtl A, Jonas JB, Naumann GO. Histomorphometry of the optic disc in highly myopic eyes with absolute secondary angle closure glaucoma. Br J Ophthalmol 1998;82:286 –289. 7. Cahane M, Bartov E. Axial length and scleral thickness effect on susceptibility to glaucomatous damage: a theoretical model implementing Laplace’s law. Ophthalmic Res 1992; 24:280 –284. 8. Avetisov ES, Savitskaya NF. Some features of ocular microcirculation in myopia. Ann Ophthalmol 1977;9:1261–1264. 9. Shih YF, Horng IH, Yang CH, Lin LL, Peng Y, Hung PT. Ocular pulse amplitude in myopia. J Ocul Pharmacol 1991; 7:83– 87. 10. To’mey KF, Faris BM, Jalkh AE, Nasr AM. Ocular pulse in high myopia: a study of 40 eyes. Ann Ophthalmol 1981;13: 569 –571. 11. Perkins ES. The ocular pulse. Curr Eye Res 1981;1:19 –23. 12. Quigley HA, Enger C, Katz J, Sommer A, Scott R, Gilbert D. Risk factors for the development of glaucomatous visual field loss in ocular hypertension. Arch Ophthalmol 1994;112: 644 – 649. 13. Perkins ES, Phelps CD. Open angle glaucoma, ocular hypertension, low-tension glaucoma, and refraction. Arch Ophthalmol 1982;100:1464 –1467. 14. Chihara E, Liu X, Dong J, et al. Severe myopia as a risk factor for progressive visual field loss in primary open-angle glaucoma. Ophthalmologica 1997;211:66 –71. 15. Nouri-Mahdavi K, Hoffman D, Coleman AL, et al. Predictive factors for glaucomatous visual field progression in the
838
AMERICAN JOURNAL
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
OF
Advanced Glaucoma Intervention Study. Ophthalmology 2004;111:1627–1635. Phelps CD. Effect of myopia on prognosis in treated primary open-angle glaucoma. Am J Ophthalmol 1982;93:622– 628. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities Classification System III. The Longitudinal Study of Cataract Study Group. Arch Ophthalmol 1993;111:831– 836. Boland MV, Zhang L, Broman AT, Jampel HD, Quigley HA. Comparison of optic nerve head topography and visual field in eyes with open-angle and angle-closure glaucoma. Ophthalmology 2008;115:239 –245. Artes PH, Iwase A, Ohno Y, Kitazawa Y, Chauhan BC. Properties of perimetric threshold estimates from Full Threshold, SITA Standard, and SITA Fast strategies. Invest Ophthalmol Vis Sci 2002;43:2654 –2659. Leske MC, Heijl A, Hyman L, Bengtsson B. Early Manifest Glaucoma Trial: design and baseline data. Ophthalmology 1999;106:2144 –2153. Smith SD, Katz J, Quigley HA. Analysis of progressive change in automated visual fields in glaucoma. Invest Ophthalmol Vis Sci 1996;37:1419 –1428. Katz J, Gilbert D, Quigley HA, Sommer A. Estimating progression of visual field loss in glaucoma. Ophthalmology 1997;104:1017–1025. Sawada A, Tomidokoro A, Araie M, Iwase A, Yamamoto T. Tajimi Study Group Refractive errors in an elderly Japanese population: the Tajimi study. Ophthalmology 2008;115:363– 370. Heijl A, Leske MC, Bengtsson B, et al. Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Arch Ophthalmol 2002; 120:1268 –1279. Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Collaborative Normal-Tension Glaucoma Study Group. Am J Ophthalmol 1998;126:487– 497. Manassakorn A, Nouri-Mahdavi K, Koucheki B, Law SK, Caprioli J. Pointwise linear regression analysis for detection of visual field progression with absolute versus corrected threshold sensitivities. Invest Ophthalmol Vis Sci 2006;47: 2896 –2903. Katz J. A comparison of the pattern- and total deviationbased glaucoma change probability programs. Invest Ophthalmol Vis Sci 2000;41:1012–1016.
OPHTHALMOLOGY
MAY 2010
To evaluate the influence of the extent of myopia on the progression rate of normal-tension glaucoma (NTG). ● DESIGN: Retrospective, observational case series. ● METHODS: One hundred forty-three eyes of 143 patients with NTG who were treated from 1994 through 2006 and followed up with standard automated perimetry were evaluated in this study. The participants were divided into 4 groups: mild myopia (ⴚ0.76 to ⴚ2.99 diopters [D]), moderate myopia (ⴚ3 to ⴚ5.99 D), severe myopia (ⴚ6 D or less), and nonmyopia (emmetropia and hyperopia, ⴚ0.75 D or more) groups. The change in mean deviation, corrected pattern standard deviation, mean thresholds of 10 zones corresponding to the glaucoma hemifield test, and thresholds of 52 points of the nonmyopia group were compared with those of the other myopia groups. Additionally, we controlled each analysis for age and posttherapeutic intraocular pressure to preclude the possibility of these covariates influencing the analysis of the effect of myopia on the progression of glaucoma. ● RESULTS: There was no statistically significant difference between the nonmyopia group and each of the myopia groups in terms of mean deviation, corrected pattern standard deviation, mean thresholds of 10 zones corresponding to the glaucoma hemifield test, and the thresholds of 52 point changes against refraction. Moreover, with the control of the other covariates (age and posttherapeutic intraocular pressure), no statistically significant differences were noted (multivariate analysis using mixed model, P > .1). ● CONCLUSIONS: Although a high incidence of openangle glaucoma among myopic patients has been reported previously, myopia did not influence the progression rate of NTG after treatment. (Am J Ophthalmol 2010; 149:831– 838. © 2010 by Elsevier Inc. All rights reserved.)
Accepted for publication Dec 18, 2009. From the Department of Ophthalmology, Gangneung Asan Hospital, Ulsan University College of Medicine, Gangneung, Korea (S.W.S.); the Department of Preventive medicine, Kwandong University College of medicine, Gangneung, Korea (J.S.S.); and the Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea (C.K.). Inquiries to Changwon Kee, Department of Ophthalmology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul 135-710, Korea; e-mail: [email protected] 0002-9394/10/$36.00 doi:10.1016/j.ajo.2009.12.033
©
2010 BY
T
HE CORRELATION BETWEEN MYOPIA AND GLAUCOMA
has been a focus of interest in glaucoma research for a long time. Several studies already have proved the relatively high incidence rate of glaucoma in myopic patients. The Beaver Dam Eye Study1 reported that the incidence rate of glaucoma in myopic patients was 1.6 times higher than in those with emmetropia. The Malmo Eye Study2 reported that glaucoma developed in 1.5% of individuals with myopia and in 0.9% of those with emmetropia. Thus, several explanations are possible. One of them asserts that myopia includes an anatomic weakness of the disc, which manifests as increased cup size,3 unusually large or skewed canal shape,4 – 6 and thin sclera and lamina in the disc periphery.7 Furthermore, reduced blood flow8 and low ocular pulse amplitude9 –11 have been reported in myopic eyes. For these reasons, the optic disc in myopic patients may be subject to ischemia or damage at any level of intraocular pressure (IOP). Although it has been demonstrated previously that myopia is associated with increased incidence of glaucoma, the effect of myopia on the progression of glaucoma has yet to be elucidated. Quigley and associates12 and Perkins and Phelps13 reported that visual field (VF) defects were observed more frequently in myopic patients with untreated ocular hypertension. Chihara and Sawada14 reported that myopia was a risk factor in the progression of POAG. However, NouriMahdavi and associates15 and Phelps16 reported that myopia did not affect the progression of primary open-angle glaucoma (POAG) under treatment. These previous studies simply assessed the progression or nonprogression of glaucoma, considering myopia as one of various risk factors. Additionally, no studies have been conducted yet regarding the progression rate of glaucoma in association with the degree of myopia, nor have any studies been conducted with the other risk factors being controlled. This study particularly evaluated the effect of myopia on the progression rate of normal-tension glaucoma (NTG) in an effort to answer the question as to whether myopia should be considered in the treatment or prognosis of NTG.
METHODS ● STUDY SUBJECTS:
Among 874 patients diagnosed with NTG at the Department of Ophthalmology of the Samsung Medical Center from 1994 through 2006 and who
ELSEVIER INC. ALL
RIGHTS RESERVED.
831
were treated only with a topical hypotensive agent, 143 ultimately were selected. Inclusion criteria. In a retrospective study involving the records of 874 patients, 198 were selected for study based on the inclusion criteria described below. A diagnosis of NTG was made when a patient with an IOP of 21 mm Hg or less without treatment had findings of glaucomatous optic disc damage and corresponding VF defects, an open angle observed by gonioscopic examination, and no underlying cause for the optic disc damage aside from glaucoma. Each patient had to have been followed up in the outpatient clinic for 5 years or more and to have a VF test twice yearly, for a total of 10 or more VF tests. In patients with bilateral NTG, one eye was selected randomly and analyzed. Among 874 total subjects, 198 subjects were considered eligible.
FIGURE 1. Diagram showing the definition of 10 zones of the glaucoma hemifield test (GHT) and 52 points.
Exclusion criteria. Among the 198 selected patients, we further excluded any patient with (1) another ocular disorder (11 patients), (2) any history of ocular surgery and laser including cataract surgery (21 patients), (3) lens opacities more severe than C2,N2,P2 according to lens opacities classification system III criteria17 (13 patients), and (4) visual acuity of less than 20/40 (10 patients). Besides cataract formation, we were particularly careful to consider factors such as large peripapillary atrophy and myopic maculopathy, both of which are common in patients with high myopia and can influence VF. We performed a careful fundus examination, stereo disc photography, retinal nerve fiber layer photos, and optical coherence tomography semiannually to observe for any myopic maculopathy or large peripapillary atrophy. If either of these two conditions developed, the subject then was excluded and these 4 patients were included in the fourth exclusion criterion, visual acuity of less than 20/40 (10 patients).
Grading. Refractive status was measured via a manifest refraction test using a spherical equivalent. Emmetropia was defined as between ⫺0.75 and 0.75 diopters (D) and hyperopia as 0.76 D or more. Myopia was classified into mild myopia (⫺0.76 to ⫺2.99 D), moderate myopia (⫺3 to ⫺5.99 D), and severe myopia (⫺6 D or less).1 ● VISUAL FIELD TESTING: The Central 30-2 full-threshold program of the Humphrey Visual Field Analyzer (Carl Zeiss Ophthalmic System, Inc., Dublin, California, USA) was used for VF testing. A reliable VF had to fulfill 3 criteria: fixation loss less than 20%, a false-positive rate of 30% or less, and a false-negative rate of 30% or less. The results of the first 2 tests were excluded to remove the learning effect from the final analysis. Additionally, those with a mean deviation (MD) of ⫺25 dB or less, for whom glaucoma progression would be difficult to evaluate from a VF test, also were excluded. The corrected pattern standard deviation (CPSD) with mean deviation (MD) of ⫺12 dB or less was excluded from the analysis, because the CPSD may be decreased when the glaucoma progresses to severe status.18,19 VF data for left eyes were switched across the vertical meridian such that, for example, the nasal area for the left eyes corresponded with the nasal area for the right eyes in all analyses. Hence, all figures of VF are presented as right eyes, although they actually represent mixed results from right and left eyes. Using these VF results, the glaucoma progression rate and progression or nonprogression in myopic groups were analyzed statistically in comparison with those of the nonmyopic groups. The determination of time to progression is based on glaucoma change probability maps (GCPMs) of total deviation (TD), which is included in the STATpac software package of the Humphrey Visual Field Analyzer. To detect VF progression, all follow-up results of the VF tests were compared with an average of 2 baseline VF tests from the same eye using GCPMs. Definite VF progression was defined as at least 3 test points exhibiting significant
Treatment. When NTG was diagnosed, the patients received betaxolol 5 mg/mL (Alcon, Inc, Fort Worth, Texas, USA) twice daily. During the follow-up, if glaucoma progressed or did not reach the target IOP, betaxolol was replaced by latanoprost 50 g/mL (Pfizer, Inc., New York, New York, USA). The first target of IOP reduction was to reduce 30% of basal IOP. But if it did not reach its first target IOP, 14 mm Hg was set for the second target IOP. When latanoprost did not have sufficient effect, brimonidine 2 mg/mL (Allergan, Inc., Irvine, California, USA) usually was added. A Goldmann applanation tonometer was used for IOP measurement. The basal IOP was calculated as the mean IOP of each hospital visit before the administration of a topical hypotensive agent. Additionally, the posttherapeutic IOP was calculated as the mean IOP of each hospital visit after the administration of a topical hypotensive agent. 832
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MAY 2010
TABLE 1. Characteristics of Patients with Normal-Tension Glaucoma in Each Myopic and Nonmyopic Group
Characteristics
Nonmyopic Group (Hyperopia and Emmetropia ⫺0.75 D or more; n ⫽ 75)
Mild Myopia Group (⫺0.76 to ⫺2.99 D; n ⫽ 22)
Moderate Myopia Group (⫺3 to ⫺5.99 D; n ⫽ 22)
Severe Myopic Group (⫺6 D or less; n ⫽ 24)
Age Sex (male/female) Refraction (D) Basal IOP (mm Hg) Posttherapeutic IOP (mm Hg) Ratio of IOP lowering (%) No. of examinations Basal MD (dB) Basal CPSD (dB) Follow-up (mos) Diabetes mellitus (n)b Hypertension (n)c
61.8 ⫾ 8.9 35/40 0.39 ⫾ 0.75 16.1 ⫾ 2.4 14.4 ⫾ 1.9 13.2 ⫾ 7.3 8.4 ⫾ 3.1 ⫺6.7 ⫾ 4.9 6.9 ⫾ 4.1 67.4 ⫾ 21.6 10 13
59.7 ⫾ 10.8 8/14 ⫺1.92 ⫾ 0.55 17.7 ⫾ 2.5 14.8 ⫾ 2.0 14.8 ⫾ 8.4 8 ⫾ 2.7 ⫺6.7 ⫾ 4.1 7.5 ⫾ 3.6 63.8 ⫾ 23.5 5 3
65.5 ⫾ 10.5 7/15 ⫺4.57 ⫾ 0.76 17.1 ⫾ 2.9 15.1 ⫾ 1.8 14.2 ⫾ 8.9 9.2 ⫾ 3.6 ⫺7.1 ⫾ 5.0 8.5 ⫾ 4.9 67.7 ⫾ 25.0 3 4
60.7 ⫾ 8.6 17/7 ⫺9.37 ⫾ 0.43 16.9 ⫾ 3.2 15.0 ⫾ 1.9 13.6 ⫾ 6.8 9.5 ⫾ 3.6 ⫺7.1 ⫾ 5.4 7.1 ⫾ 3.6 69.4 ⫾ 24.3 3 4
P Valuea
.624 .230 0 .131 .419 .517 .307 .665 .524 .865 .957 .975
CPSD ⫽ corrected pattern standard deviation; D ⫽ diopters; IOP ⫽ intraocular pressure; MD ⫽ mean deviation. Data are presented as number and mean ⫾ standard deviation. a Analysis of variance. b Medical history. c Defined as systolic pressure more than 160 mm Hg, diastolic pressure more than 95 mm Hg, or medical history.
TABLE 2. Progression Rate of Mean Deviation and Corrected Pattern Standard Deviation in Patients with Normal-Tension Glaucoma in Each Myopic and Nonmyopic Group
MD (dB/year) CPSD (dB/year)
Nonmyopic Group
Mild Myopia Group
Moderate Myopia Group
Severe Myopia Group
P Valuea
⫺1.334 ⫾ 1.003 0.967 ⫾ 0.48
⫺1.055 ⫾ 0.328 1.525 ⫾ 0.7
⫺1.113 ⫾ 0.466 1.022 ⫾ 0.126
⫺0.912 ⫾ 0.313 1.259 ⫾ 0.748
.255 .106
CPSD ⫽ corrected pattern standard deviation; MD ⫽ mean deviation. Data are presented as mean ⫾ standard deviation. a P value based on multivariate analysis using a mixed model.
(P ⬍ .05) progression at the same locations on 3 consecutive tests as compared with the baseline.20 ● PROGRESSION RATE ANALYSIS:
Univariate analysis. The first method used for comparisons of the progression rate was univariate analysis, using linear regression, which concerned only the effects of myopia. The progression rate of each myopic and nonmyopic group was calculated by linear regression. The methods of univariate analysis were as follows. VF data (MD, CPSD, the thresholds of 52 points) were transferred to a personal computer, and univariate linear regression was conducted with time as the independent variable. Regression analysis was conducted with respect to 2 global indices (MD and CPSD), and similar analysis was conducted for the mean thresholds of the 10 zones corresponding to those of the glaucoma hemifield test and the thresholds of the 52 test points (Figure 1). Sixty-four linear regressions (1(MD) ⫹ 1(CPSD) ⫹ 10(zones) ⫹ 52(points) ⫽ 64) were conducted for each individual.
VOL. 149, NO. 5
MYOPIA
ON THE
Among them, the P ⱕ .025 for the slope of the MD and CPSD, P ⱕ .005 for the slope of 10 zones, and P ⱕ .001 for the slope of 52 points were considered to be statistically significant. Only such significant values were selected for the calculation of the mean values of the progression rate.21,22 Multivariate analysis. The second method used for the comparison of the progression rate was a multivariate analysis using a mixed model. Each progression rate of MD, CPSD, 10 zones, and 52 points over time was compared by multivariate analysis using the mixed model between the myopia and nonmyopia groups. Multivariate analysis was used to evaluate the association of the continuous dependent variable of the degree of myopia (each myopic and non-myopic groups) with the binary independent variable of progression rate. Additionally, covariates including age, sex, diabetes mellitus (DM), hypertension (HTN), basal IOP, postNTG PROGRESSION
833
FIGURE 2. Diagrams showing the progression rate of 10 zones in patients with normal-tension glaucoma in each myopic and nonmyopic group: (Top left) nonmyopic group, (Top right) mild myopia group, (Middle left) moderate myopia group, and (Middle right) severe myopia group. Data are presented as mean (dB/year). (Bottom) *P value based on multivariate analysis using a mixed model that controlled age and posttherapeutic intraocular pressure.
therapeutic IOP, basal MD, and basal CPSD were controlled using various covariate combinations. Thus, 64 (1(MD) ⫹ 1(CPSD) ⫹ 10(zones) ⫹ 52(point)) times of multivariate analyses using a mixed model were conducted with each covariate combination; therefore, more than hundreds of analyses were conducted in total.
the aforementioned progression rate. The decision-making process for progression is the same as explained above, using GCPMs of TD. The Student t test was used to analyze the degree of myopia between the progression and nonprogression groups, and a chi-square test was used to analyze the differences in the frequency of progression between the myopic and nonmyopic groups.
Survival analysis. The third method used for comparisons of the progression rate was Kaplan-Meier survival analysis of time to progression. The process of the progression decision was the same as the above explanations using GCPMs of TD.
Multivariate analysis. Additionally, using logistic regression, the association of VF progression with myopia was evaluated, and the influences of age and posttherapeutic IOP were explored. All statistical analyses were conducted using SAS software version 8.0 (SAS, Inc., Cary, North Carolina, USA) and SPSS software version 11.0 (SPSS, Inc., Chicago, Illinois, USA).
● PROGRESSION OR NONPROGRESSION ANALYSIS:
Univariate analysis. The final method used in this study was used to compare progression or nonprogression, and not
834
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MAY 2010
FIGURE 3. Diagrams showing the progression rate of 52 points in patients with normal-tension glaucoma in each myopic and nonmyopic group: (Top left) nonmyopic group, (Top right) mild myopia group, (Middle left) moderate myopia group, and (Middle right) severe myopia group. Data are presented as mean (dB/year). (Bottom) *P value based on multivariate analysis using a mixed model that controlled for age and posttherapeutic intraocular pressure.
RESULTS ● GENERAL ANALYSIS: Among 874 patients, 143 patients and 143 eyes (67 from males and 76 from females) fulfilled the inclusion criteria of this study. With regard to the classification of refractive error, (1) 22 eyes were categorized with mild myopia (⫺0.76 to ⫺2.99 D), (2) 22 eyes with moderate myopia (⫺3 to ⫺5.99 D), and (3) 24 eyes with severe myopia (⫺6 D or less). In the 75 nonmyopic eyes (⫺0.75 D or more), (1) 51 eyes were categorized with emmetropia (⫺0.75 to 0.75 D) and (2) 24 eyes with hyperopia (0.76 D or more). Age, refraction, basal IOP, posttherapeutic IOP, frequency of VF testing, basal MD, basal CPSD, and follow-up period for each group are listed in Table 1. Except for the refraction value, there were no significant differences in any variable used in comparisons between the nonmyopic group and any of the myopic groups. ● PROGRESSION RATE ANALYSIS:
Univariate and multivariate analysis. Changes in the linear regression values for MD for each group were as follows: ⫺1.334 dB/year for the
VOL. 149, NO. 5
MYOPIA
ON THE
nonmyopic group and ⫺1.055 dB/year, ⫺1.113 dB/year, and ⫺0.912 dB/year for the mild, moderate, and severe myopic groups, respectively. Results of multivariate analysis using the mixed model, in which age and posttherapeutic IOP were controlled, revealed no statistically significant differences between groups (P ⫽ .255). The change in the CPSD by linear regression analysis in the nonmyopic group was 0.967 dB/year and 1.525 dB/ year, 1.022 dB/year, and 1.259 dB/year for the mild, moderate, and severe myopic groups, respectively. There was no significant between-group differences by multivariate analysis using the mixed model (P ⫽ .106; Table 2). Comparisons of the 10 zones yielded the following results (Figure 2). There were no significant differences by multivariate analysis using the mixed model among 9 areas; zone 6 was the exception. In zone 6, the nonmyopic group lost ⫺0.613 dB/year and the mild, moderate, and severe myopic groups lost ⫺0.493 dB/year, ⫺0.656 dB/ year, and ⫺0.959 dB/year, respectively. In zone 6, there were significant differences as the result of multivariate analysis (P ⫽ .011) using the mixed model between the NTG PROGRESSION
835
TABLE 3. Multivariate Risk Factors Associated with Visual Field Progression in Patients with Normal-Tension Glaucoma Factors
Odds Ratio (95% CI)
P Valuea
Myopia (1-D increase) Age (1-year increase) Posttherapeutic IOP (1-mm Hg increase)
1.34 (0.452 to 3.667) 1.771 (0.465 to 4.129)
.458 .576
1.316 (0.782 to 5.796)
.461
CI ⫽ confidence interval; D ⫽ diopters; IOP ⫽ intraocular pressure. a P value based on logistic regression analysis.
7.0 months, respectively, and did not differ significantly from that of the nonmyopic group (P ⫽ .903, .672, respectively, log-rank test; Figure 4).
FIGURE 4. Kaplan-Meier curve showing the cumulative probability of visual field progression in patients with normaltension glaucoma in each myopic and nonmyopic group. The progression periods for the mild, moderate, and severe myopia group were 78.8 ⴞ 9.0 months, 73.4 ⴞ 9.0 months, and 76.1 ⴞ 7.0 months, respectively, and did not differ significantly from that of the nonmyopic group: 76.6 ⴞ 5.1 months (P ⴝ .882, .903, .672, respectively, log-rank test).
● PROGRESSION OR NONPROGRESSION ANALYSIS:
Univariate analysis. Among the 143 eyes, 79 showed progression. The mean refraction value of these 79 eyes was ⫺2.4 ⫾ 4.3 D. For the 64 eyes in which progression was not noted, the value was ⫺2.2 ⫾ 3.8 D. These figures were not statistically significantly different (P ⫽ .765, Student t test). With regard to exacerbation in each myopic group, 42 of the 75 nonmyopic eyes (56%) evidenced progression. Progression was detected in 11 of 22 mild myopia eyes (50%), 10 of 22 moderate myopia eyes (45.5%), and 10 of 24 severe myopia eyes (41.7%). There was no significant differences in progression between the nonmyopic and each myopic group (P ⫽ .599, chi-square test).
nonmyopic and each myopic group, in which both age and posttherapeutic IOP were controlled. There were no statistically significant differences between groups for each of the 52 points by multivariate analysis using the mixed model (Figure 3). Covariates such as age, sex, DM, HTN, basal IOP, posttherapeutic IOP, basal MD, and basal CPSD were controlled in the multivariate analysis using mixed method. Although we controlled all 8 of the covariates (age, sex, DM, HTN, basal IOP, posttherapeutic IOP, basal MD, basal CPSD) and any 8 covariate combinations, the results were not significant. However, because of the small number of the study subjects (143 patients) and the negative results, we selected the 2 most important factors, age and posttherapeutic IOP, and subsequently analyzed these. Zone 6 proved significant under the control of the age and posttherapeutic IOP combination only, but was not significant in the control of other factors (sex, DM, HTN, basal IOP, basal MD, basal CPSD).
Multivariate analysis. Logistic regression analysis was used to seek a possible correlation between myopia and VF progression. The results showed that the myopic patients evidenced VF progression of 1.34-fold that of the nonmyopic group. However, there were no statistically significant differences between the groups (P ⫽ .458, logistic regression analysis; Table 3).
DISCUSSION AS MENTIONED ABOVE, MANY PREVIOUS STUDIES HAVE
assessed the effects of myopia on the incidence and progression or nonprogression of POAG. However, this study focused on the effects of myopia on the progression rate of NTG, which has not been elucidated previously. The reasons for analyzing the progression rate are as follows. First, the results of previous studies regarding progression or nonprogression proved controversial. Thus, we expected to detect progression or nonprogression indirectly, via progression rate analysis. Second, the progression rate was our main concern and focus of interest because the progression rate can be more rapid than
Survival analysis. Disease progression was interpreted with GCPMs.20 In the survival analysis of VF progression for the nonmyopic and mild myopia groups using the Kaplan-Meier life table method, the progression periods were 76.6 ⫾ 5.1 months and 78.8 ⫾ 9.0 months, respectively. The differences between the 2 groups were not statistically significant (P ⫽ .882, log-rank test). The progression periods of the moderate myopia group and the severe myopia group were 73.4 ⫾ 9.0 months and 76.1 ⫾ 836
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MAY 2010
usual, because of anatomic and hemodynamic weaknesses of the myopic disc. In particular, this study assessed NTG only because the effects of IOP would be relatively smaller than those of POAG. Additionally, the prevalence of NTG in Asian countries such as Korea and Japan is quite high, and some recent studies also have reported a relatively higher prevalence of myopia in that region.23 The results of this study demonstrated that, except for the progression rate of the sixth zone, myopia did not affect either the progression or nonprogression or the progression rate of NTG. However, much more attention should be paid to the analysis of these negative results. This is because not just one factor, but many complicated factors, may be associated with the progression or nonprogression and the progression rate of NTG. Thus, we conducted analyses via 4 different methods, including multivariate analysis, in an effort to clarify these negative results. The reason that we analyzed a number of results (MD, CPSD, 10 zones, and 52 points) of VF is that the NTG is known to progress commonly in certain specialized regions (e.g., paracentral or nasal defects) as well as in generalized VF change. Therefore, we also conducted analyses of 10 zones and 52 points, because we thought that global index analysis only, such as MD and CPSD, were not appropriate. In the sixth zone of the glaucoma hemifield test, the progression rate was ⫺0.613 dB/year in the nonmyopic group and ⫺0.493 dB/year, ⫺0.659 dB/year, and ⫺0.959 dB/year, respectively, in the mild, moderate, and high myopia groups. There were significant differences (P ⫽ .011) in the multivariate analysis using a mixed model that controlled for age and posttherapeutic IOP. However, each point of nos. 30, 31, and 32 comprising zone 6 were not significant. The value of zone 6 was calculated as the average of these 3 points (nos. 30, 31, and 32). Because zone 6 may continue to progress even after the value of any of these 3 points (nos. 30, 31, and 32) falls to 0, the value of an individual point is considered to be more useful than that of the zone. Also, at the univariate analysis, zone 6 came out to be meaningless (P ⫽ .124, analysis of variance), and at the multivariate analysis, it had difference only under control of age and posttherapeutic IOP only. Therefore, zone 6, which does not match individual points, is not considered to be meaningful. The limitations of this study are as follows. First, it was basically a retrospective study, and second, the selection of subjects was limited. With regard to subject selection, cases of extremely advanced glaucoma (MD ⱕ ⫺25 dB) and any history of ocular surgery, including cataract surgery, were excluded from this study. Additionally, this study used only subjects who had been followed-up for 5 years or more and
who had undergone VF tests at least twice yearly, for a total of 10 VF tests or more. As a result, only 143 subjects ultimately were enrolled, and some patients without progression were lost to follow-up observations. Moreover, because the study was conducted in a tertiary medical care institution, rather than being conducted with new patients, many patients had been referred from primary and secondary medical institutions with continuous progression, despite having been treated. In fact, the change in MD in this study occurred more rapidly than in the Early Manifest Glaucoma Trial (EMGT) treatment group (⫺0.03 dB/month)24 and the Collaborative Normal Tension Glaucoma Study (⫺0.4992 dB/year),25 both of which were prospective studies. Additionally, the progression or nonprogression rate in this study was higher than the rate reported in the EMGT. The results of this study based on the GCPMs also revealed a high rate of progression (52%) even after treatment, whereas the EMGT study reported a rate of 45%24 and the Collaborative Normal Tension Glaucoma Study reported a rate of 18%.25 However, the EMGT used GCPMs of pattern deviation, which are known to be more conservative than our GCPMs of TD.26,27 In addition, the basal MD in the EMGT was ⫺5.0 ⫾ 3.7 dB, which is a less profound progression than found in our study, and the EMGT excluded MD ⱕ ⫺16 dB, which was defined as advanced glaucoma. However, in our study, MD ⱕ ⫺25 dB was defined as advanced glaucoma, and thus excluded. Therefore, simple numerical comparisons of these studies would be unreasonable. The Collaborative Normal Tension Glaucoma Study also differs from our study in the definition of progression and the subject selection process. Finally, our study used only the results of VF to judge progression. This made it easy to compare the results of VF statistically, because the results of the VF test are numerical. Because the subjects of the study were patients with myopic changes such as tilted discs and large peripapillary atrophies, other standards such as disc, retinal nerve fiber layer, and so forth could not be used to determine progression. Yet, the use of VF tests alone may prove insufficient to determine progression. In conclusion, the results of our study showed that myopia did not affect the progression rate or progression or nonprogression in NTG patients who had begun treatment. Our conclusions suggest that, in treated NTG patients, the myopic element is not the principal issue for treatment planning or prognosis. Therefore, regardless of the various limitations of this study, its results are useful and relevant. In the future, more prospective research will be necessary to expand on the results of this study.
THE AUTHORS INDICATE NO FINANCIAL SUPPORT OR FINANCIAL CONFLICT OF INTEREST. INVOLVED IN DESIGN OF THE STUDY (C.K.); Conduct of the study (S.S.W., J.S.S., C.K.); Collection and management of data (S.S.W., J.S.S., C.K.); Analysis and interpretation of data (S.S.W., J.S.S., C.K.); Preparation and writing of the manuscript (S.S.W.); Review and final approval of the manuscript (C.K.). This research was approved by Samsung Medical Center Institutional Review Board. An exemption from informed consent for research was granted because this was a retrospective research study.
VOL. 149, NO. 5
MYOPIA
ON THE
NTG PROGRESSION
837
REFERENCES 16.
1. Wong TY, Klein BK, Klein R, Knudtson M, Lee KE. Refractive errors, intraocular pressure, and glaucoma in a white population. Ophthalmology 2003;110:211–217. 2. Grodum K, Heijl A, Bengtsson B. Refractive error and glaucoma. Acta Ophthalmol Scand 2001;79:560 –566. 3. Tomlinson A, Phillips CI. Ratio of optic cup to optic disc. In relation to axial length of eyeball and refraction. Br J Ophthalmol 1969;53:765–768. 4. Bellezza AJ, Hart RT, Burgoyne CF. The optic nerve head as a biomechanical structure: initial finite element modeling. Invest Ophthalmol Vis Sci 2000;41:2991–3000. 5. Chihara E, Sawada A. Atypical nerve fiber layer defects in high myopes with high-tension glaucoma. Arch Ophthalmol 1990;108:228 –232. 6. Dichtl A, Jonas JB, Naumann GO. Histomorphometry of the optic disc in highly myopic eyes with absolute secondary angle closure glaucoma. Br J Ophthalmol 1998;82:286 –289. 7. Cahane M, Bartov E. Axial length and scleral thickness effect on susceptibility to glaucomatous damage: a theoretical model implementing Laplace’s law. Ophthalmic Res 1992; 24:280 –284. 8. Avetisov ES, Savitskaya NF. Some features of ocular microcirculation in myopia. Ann Ophthalmol 1977;9:1261–1264. 9. Shih YF, Horng IH, Yang CH, Lin LL, Peng Y, Hung PT. Ocular pulse amplitude in myopia. J Ocul Pharmacol 1991; 7:83– 87. 10. To’mey KF, Faris BM, Jalkh AE, Nasr AM. Ocular pulse in high myopia: a study of 40 eyes. Ann Ophthalmol 1981;13: 569 –571. 11. Perkins ES. The ocular pulse. Curr Eye Res 1981;1:19 –23. 12. Quigley HA, Enger C, Katz J, Sommer A, Scott R, Gilbert D. Risk factors for the development of glaucomatous visual field loss in ocular hypertension. Arch Ophthalmol 1994;112: 644 – 649. 13. Perkins ES, Phelps CD. Open angle glaucoma, ocular hypertension, low-tension glaucoma, and refraction. Arch Ophthalmol 1982;100:1464 –1467. 14. Chihara E, Liu X, Dong J, et al. Severe myopia as a risk factor for progressive visual field loss in primary open-angle glaucoma. Ophthalmologica 1997;211:66 –71. 15. Nouri-Mahdavi K, Hoffman D, Coleman AL, et al. Predictive factors for glaucomatous visual field progression in the
838
AMERICAN JOURNAL
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
OF
Advanced Glaucoma Intervention Study. Ophthalmology 2004;111:1627–1635. Phelps CD. Effect of myopia on prognosis in treated primary open-angle glaucoma. Am J Ophthalmol 1982;93:622– 628. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities Classification System III. The Longitudinal Study of Cataract Study Group. Arch Ophthalmol 1993;111:831– 836. Boland MV, Zhang L, Broman AT, Jampel HD, Quigley HA. Comparison of optic nerve head topography and visual field in eyes with open-angle and angle-closure glaucoma. Ophthalmology 2008;115:239 –245. Artes PH, Iwase A, Ohno Y, Kitazawa Y, Chauhan BC. Properties of perimetric threshold estimates from Full Threshold, SITA Standard, and SITA Fast strategies. Invest Ophthalmol Vis Sci 2002;43:2654 –2659. Leske MC, Heijl A, Hyman L, Bengtsson B. Early Manifest Glaucoma Trial: design and baseline data. Ophthalmology 1999;106:2144 –2153. Smith SD, Katz J, Quigley HA. Analysis of progressive change in automated visual fields in glaucoma. Invest Ophthalmol Vis Sci 1996;37:1419 –1428. Katz J, Gilbert D, Quigley HA, Sommer A. Estimating progression of visual field loss in glaucoma. Ophthalmology 1997;104:1017–1025. Sawada A, Tomidokoro A, Araie M, Iwase A, Yamamoto T. Tajimi Study Group Refractive errors in an elderly Japanese population: the Tajimi study. Ophthalmology 2008;115:363– 370. Heijl A, Leske MC, Bengtsson B, et al. Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Arch Ophthalmol 2002; 120:1268 –1279. Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Collaborative Normal-Tension Glaucoma Study Group. Am J Ophthalmol 1998;126:487– 497. Manassakorn A, Nouri-Mahdavi K, Koucheki B, Law SK, Caprioli J. Pointwise linear regression analysis for detection of visual field progression with absolute versus corrected threshold sensitivities. Invest Ophthalmol Vis Sci 2006;47: 2896 –2903. Katz J. A comparison of the pattern- and total deviationbased glaucoma change probability programs. Invest Ophthalmol Vis Sci 2000;41:1012–1016.
OPHTHALMOLOGY
MAY 2010