January 2011
Volume 52, Issue 1
Free
Retina  |   January 2011
Correlations between M-CHARTS and PHP Findings and Subjective Perception of Metamorphopsia in Patients with Macular Diseases
Author Affiliations & Notes
  • Eiko Arimura
    From the Department of Ophthalmology, Kinki University Faculty of Medicine, Sakai Hospital, Osaka, Japan; and
  • Chota Matsumoto
    the Department of Ophthalmology, Kinki University Faculty of Medicine, Osaka-Sayama, Japan.
  • Hiroki Nomoto
    the Department of Ophthalmology, Kinki University Faculty of Medicine, Osaka-Sayama, Japan.
  • Shigeki Hashimoto
    the Department of Ophthalmology, Kinki University Faculty of Medicine, Osaka-Sayama, Japan.
  • Sonoko Takada
    the Department of Ophthalmology, Kinki University Faculty of Medicine, Osaka-Sayama, Japan.
  • Sachiko Okuyama
    the Department of Ophthalmology, Kinki University Faculty of Medicine, Osaka-Sayama, Japan.
  • Yoshikazu Shimomura
    the Department of Ophthalmology, Kinki University Faculty of Medicine, Osaka-Sayama, Japan.
  • Corresponding author: Eiko Arimura, Department of Ophthalmology, Kinki University Faculty of Medicine, Sakai Hospital, Harayamadai, Minamiku, Sakai City, Osaka 590-0132, Japan; arimura@ganka.med.kindai.ac.jp
Investigative Ophthalmology & Visual Science January 2011, Vol.52, 128-135. doi:10.1167/iovs.09-3535
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Eiko Arimura, Chota Matsumoto, Hiroki Nomoto, Shigeki Hashimoto, Sonoko Takada, Sachiko Okuyama, Yoshikazu Shimomura; Correlations between M-CHARTS and PHP Findings and Subjective Perception of Metamorphopsia in Patients with Macular Diseases. Invest. Ophthalmol. Vis. Sci. 2011;52(1):128-135. doi: 10.1167/iovs.09-3535.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose.: To assess the correlations between a patient's subjective perception of metamorphopsia and the clinical measurements of metamorphopsia by M-CHARTS and PreView PHP (PHP).

Methods.: The authors designed a 10-item questionnaire focusing on the symptoms of metamorphopsia and verified its validity with a Rasch analysis. M-CHARTS measured the minimum visual angle of a dotted line needed to detect metamorphopsia, and PHP used the hyperacuity function for detection. Subjects were 39 patients with idiopathic epiretinal membrane (ERM), 22 patients with idiopathic macular hole (M-hole), 19 patients with age-related macular degeneration (AMD), and 51 healthy controls.

Results.: Rasch analysis suggested the elimination of one question. The nine-item questionnaire score significantly correlated to the M-CHARTS score in ERM (r = 0.59; P = 0.0004) but not in M-hole and to the PHP result in AMD (r = −0.29; P = 0.04) but not in ERM. Eighty percent of ERM patients with greater horizontal M-CHARTS score subjectively perceived horizontal metamorphopsia more often. M-CHARTS showed better sensitivities than PHP in both ERM (89% vs. 42%) and AMD (74% vs. 68%) and better specificity (100% vs. 71%) in healthy controls. Rasch analysis indicated that the present form of the questionnaire is better suited for moderate to severe cases of metamorphopsia than for mild cases.

Conclusions.: The questionnaire appears to be a valid assessment of patient subjective perception of metamorphopsia and can be used to supplement the clinical measurements of metamorphopsia by M-CHARTS and PHP in patients with macular diseases.

In addition to visual acuity loss and central scotoma, metamorphopsia is an important symptom often observed in patients with macular diseases such as idiopathic epiretinal membrane (ERM), macular hole (M-hole), and age-related macular degeneration (AMD). Because of residual metamorphopsia, patients with metamorphopsia often report poor vision quality even after they have undergone successful operations and experienced improved visual acuity. 
Studies on clinical detection and quantification of metamorphopsia have been reported. 1 6 Amsler's Chart has been widely used as a conventional method 7 to detect metamorphopsia. However, it cannot quantify the severity of metamorphopsia. In 1999, we developed new assessment charts for metamorphopsia (M-CHARTS; Inami Co., Tokyo, Japan) that can quantify the degree of metamorphopsia in patients with macular diseases. 8 11 In 2003, a computer screen test (PreView PHP [PHP]; Carl Zeiss Meditec, Dublin, CA) based on the hyperacuity phenomenon was marketed to inspect the metamorphopsia fields of patients with AMD. 12 Although these are the available methods to detect and quantify metamorphopsia, it is not clear whether findings from them can really reflect a patient's subjective perception of metamorphopsia in daily life. 
Patients with metamorphopsia can be cognizant of this abnormal vision function through the appearance of outlines in many outdoor and indoor objects (e.g., objects with a straight-line pattern, the patient's own face or other people's faces, any printed words). Studies on the visual functioning evaluated by questionnaires have been previously published. 13 15 However, to our knowledge, no studies have been conducted to evaluate subjective metamorphopsia using a questionnaire. By obtaining a score for the patient's subjective impression of metamorphopsia with a questionnaire, ophthalmologists can compare the severity assessed by M-CHARTS or PHP with the level of subjective metamorphopsia evaluated by the questionnaire and the patient's level of difficulty in performing daily activities. They can, therefore, be better informed of the patient's conditions, which would be beneficial to treatment. For this purpose, we designed a metamorphopsia questionnaire to evaluate the patient's subjective perception of metamorphopsia and investigated the concordance between the results of the questionnaire and the assessments by M-CHARTS and PHP. 
Subjects and Methods
Subjects
Subjects were 39 patients with ERM (23 women; mean age, 64.1 ± 8.4 years; stage 1, 11 eyes; stage 2, 14 eyes; stage 3, 14 eyes; by Nakajima classification 16 ), 22 patients with M-hole (10 women; mean age, 63.1 ± 12.1 years; stage 1, 8 eyes; stage 2, 4 eyes; stage 3, 10 eyes; by Gass classification 17 ), 19 patients with AMD (3 women; mean age, 67.6 ± 7.8 years; dry type, 6 eyes; wet type, 13 eyes), and 51 healthy subjects (29 women; mean age, 68.0 ± 8.7 years). Their visual acuity (VA, logMAR) ranged from −0.1 to 0.52 in ERM, 0 to 1.3 in M-hole, 0 to 0.7 in AMD, and 0 in healthy subjects. All subjects were first tested by Amsler's Chart and had to be verified to have metamorphopsia. They also underwent a set of ophthalmic examinations including best-corrected visual acuity testing, slit-lamp biomicroscopy, applanation tonometry, dilated funduscopy, fundus photography, scanning laser ophthalmoscopy (SLO, Rodenstock, Germany), optical coherence tomography (Stratus OCT; Carl Zeiss Meditec), M-CHARTS test, and PHP test. Only the patients with AMD were also examined by fluorescein angiography (60 UVi fundus camera; Canon Inc., Tokyo, Japan). After all these clinical tests, the subjects completed the metamorphopsia questionnaire. Inclusion criteria for the subjects were as follows: pupil diameter of >3.0 mm, intraocular pressure of <21 mm Hg, refractive errors within the ranges of −6.0 D to + 3.0 D (spherical) and ± 3.0 D (cylindrical), clear optical media, and no systemic diseases that were likely to affect visual function. Informed consent was obtained from the subjects. All experiments were performed in accordance with the Declaration of Helsinki for research involving human subjects. 
Metamorphopsia Questionnaire
When we first constructed the questionnaire, we equally considered including items that could be viewed from a distance, at a close range, or somewhere in between, as well as items with patterns of horizontal or vertical lines. Items such as store signs on the street, a clock dial, a calendar, letter pads (types with horizontal or vertical lines), a chest of drawers in a room, and a fence or a wall of a house were originally included. To limit the number of items in the final questionnaire, we further selected items that were likely to be applicable to most people. How often the item would be used or would appear in a daily setting was also one of our considerations. For instance, the chance of having a TV set at home would be higher than the chance of having pictures or photographs displayed in the house. In addition, people may no longer use letter pads or may use them less often than in the past. For reasons such as these, we eliminated some items and devised a questionnaire with 10 questions (Q1-Q10) focusing on the symptoms of metamorphopsia regarding the patient's quality of vision and four questions (QA-QD) with designated aims (Fig. 1). Because metamorphopsia can be easily perceived in patterns with straight lines, Q1 to Q7 used targets with straight-line patterns that could be easily noticed in daily surroundings; two questions had an outdoor setting (Q1 and Q2) and the rest had an indoor setting. Q8 to Q10 used targets that were likely to frequently get close attention (people's faces, one's own face in the mirror, printed words). QA was designed to evaluate whether the patient had macropsia or micropsia. QB was designed to identify the direction of the patient's subjective metamorphopsia. QC was designed to determine the frequency of the distortion that occurred. QD was designed to assess the degree of difficulty caused by metamorphopsia in daily life activities. 
Figure 1.
 
The metamorphopsia questionnaire we designed included 10 questions on the symptoms of subjective metamorphopsia in patient's daily life and four other questions with designated aims. The average score of Q1 to Q10 was used as the patient's questionnaire score for the assessment of subjective metamorphopsia.
Figure 1.
 
The metamorphopsia questionnaire we designed included 10 questions on the symptoms of subjective metamorphopsia in patient's daily life and four other questions with designated aims. The average score of Q1 to Q10 was used as the patient's questionnaire score for the assessment of subjective metamorphopsia.
For the scoring of the questionnaire, the average score from Q1 to Q10 was used as the patient's questionnaire score. If the patient answered all 10 questions (“not at all,” 0 points; “a little,” 1 point; “moderately,” 2 points; “a great deal,” 3 points), his or her total score was divided by 10 to obtain the questionnaire score. If the question did not apply to the patient's lifestyle or situation because of reasons other than the patient's vision (e.g., no curtain rails in the patient's house, the patient hardly paid any attention to the lines of a crosswalk), the patient was advised to select response 4 for “none of the above,” and no score was given to that question. In that case, the total score was only divided by the number of questions with a score. Depending on the question, a patient could select “none of the above” if he or she was not sure of an appropriate answer. Therefore, we considered that a question with this response should be excluded from the scoring of the questionnaire. This scoring method was similar to that used in the famous VF-14 study by Steinberg, 15 in which an examiner read the questions one by one to the patient and the patient would select the appropriate answers. The questionnaire took approximately 10 minutes for each patient. 
Quantification of Metamorphopsia by M-CHARTS
We used M-CHARTS to quantify the severity of metamorphopsia in this study. In patients with metamorphopsia, a straight line projected onto the retina is recognized as a curved or an irregular line. If the straight line is replaced with a dotted line and the dot interval is changed from fine to coarse, the distortion of the line decreases with increasing dot interval until finally the dotted line appears straight. Based on this, we designed M-CHARTS that consist of 19 dotted lines with dot intervals ranging from 0.2° to 2.0° visual angles. 2 There are two types of M-CHARTS: type I with one dotted line and type II with two dotted lines on each chart. Type II is designed for patients with central scotomas such as M-hole. The examination distance was 30 cm, and the refraction of the eye was adjusted to this distance. In the first part of the M-CHARTS test, a vertical straight line (0°) was shown to the patient. If the patient recognized the straight line as straight, his metamorphopsia score was 0. If the patient recognized the straight line as irregular or curved, the following chart, which had a dotted line with a larger dot interval, would be shown. Dot intervals changed from fine to coarse, and the charts were shown one after another until the patient recognized the dotted line as straight (Fig. 2). The visual angle that denoted the dot interval of the vertical line seen straight was considered to be the patient's vertical M-CHARTS score. In the second part of the test, the M-CHARTS were rotated 90°, and the same test was carried out using the horizontal lines to obtain the patient's horizontal M-CHARTS score. For the reproducibility of the test, both vertical and horizontal tests were repeated three times for each subject, and the average of the three test scores was used. The patient's fundus information was completely masked from the examiner during the test. Each M-CHARTS test took approximately 5 minutes. 
Figure 2.
 
M-CHARTS have 19 charts and 2 types: type I with one dotted line for patients with general macular diseases and type II with two dotted lines for patients with M-hole. The visual angle that denoted the dot interval of the line seen straight was used as patient's M-CHARTS score. Both vertical and horizontal tests were performed to obtain separate scores.
Figure 2.
 
M-CHARTS have 19 charts and 2 types: type I with one dotted line for patients with general macular diseases and type II with two dotted lines for patients with M-hole. The visual angle that denoted the dot interval of the line seen straight was used as patient's M-CHARTS score. Both vertical and horizontal tests were performed to obtain separate scores.
Metamorphopsia Assessment by PHP
PHP technology uses hyperacuity, which can be defined as the ability to detect misalignment of objects, to evaluate metamorphopsia. Reportedly, the sensitivity of hyperacuity can detect an abnormality in any distorted area of the retina. 18,19 With the examiner's assistance, patients can perform this test by themselves on a computer screen. The examination distance was 50 cm, and the refraction of the eye was adjusted to this distance. In the PHP test, a single straight dotted line with a few dots out of alignment was flashed across different macular loci over a macular field of 14° × 14°. The patient used a stylus to touch the screen where he or she had experienced distortion in the line. Any distortion perceived by the patient was automatically recorded and analyzed, and a macular map showing the area of distortion and intensity of metamorphopsia was displayed. The PHP result for patients with AMD also included a denotation of “yes” or “no” to indicate whether the detected visual disturbance was consistent with the AMD progression compared with the normative database. For patients with ERM, PHP evaluated the total hyperacuity disturbance to detect metamorphopsia. Although a “yes/no” indication showing consistency with AMD progression was also displayed for the patients with ERM in this study, we disregarded this part of the test result. 
In this study, an M-CHARTS score greater than 0.2 was considered a positive detection. A PHP result was positive for the ERM group if an abnormal area with a total hyperacuity disturbance was shown on the macular map and positive for the AMD group if an abnormal area with a hyperacuity disturbance plus an indication of “yes” showing consistency with the AMD progression was shown on the macular map. Given that only two subjects from the M-hole group and none from the AMD group had an M-CHARTS score equal to or greater than 1.0, the severity of metamorphopsia measured by M-CHARTS was stratified to two levels (M-CHARTS score <0.5 and ≥0.5) for these two groups and to three levels (<0.5; ≥0.5 and <1.0; and ≥1.0) for the ERM group. 
Three approaches were taken to examine the relationships among the results of these three methods. We first compared the questionnaire and M-CHARTS scores in each disease group regarding the detection and severity of metamorphopsia. Because we suspected that ERM might have a closer association with the direction of metamorphopsia than AMD and M-hole, we further compared the answer to QB that showed the direction of subjective metamorphopsia to either the patient's vertical or horizontal M-CHARTS score in the ERM group. In addition, the correlation between the patient's M-CHARTS score and the answer to QD that indicated the level of difficulty in performing daily activities was also investigated in all three groups. This showed how metamorphopsia influenced vision quality in patients with various macular diseases. We also examined the correlation between the PHP result and the questionnaire score in the AMD and ERM groups. Because metamorphopsia in patients with M-hole is usually limited to the fixation area beyond the limit of PHP, we excluded the M-hole group when comparing PHP findings with the results of the metamorphopsia questionnaire or the M-CHARTS. Finally, we compared M-CHARTS and PHP by evaluating their sensitivities in the ERM and AMD groups and their specificities for detecting metamorphopsia in healthy subjects. 
Results
Rasch Analysis
For the validity of the questionnaire, we performed a Rasch analysis (RUMM2020; RUMM Laboratory, Perth, Australia) on our data. The Rasch analysis indicated one misfit item (Q8: Do people's faces appear distorted to you? Do the parts of the faces also appear missing?) in the original 10-item questionnaire. We then deleted Q8 and obtained improved fit to the model with item-trait interaction total χ2 probability (χ2 = 20.22; df = 18; P = 0.32). Questionnaire results in this study were based on the nine-item questionnaire. 
All nine items showed fit residuals of <2.5. The overall residual mean value and SD were 0 and 0.84, where the optimal fit of data to the model had a mean of 0 and an SD of 1. Table 1 shows the individual item fit statistics for all the items. Based on our Bonferroni-adjusted alpha value of 0.006 (0.05/9), none of the items showed a significant interaction. For item measures, the mean was 0 and the logit scores of the nine items in order by value were 1.64 (Q9), 0.43 (Q2), 0.40 (Q7), 0.25 (Q3), −0.20 (Q1), −0.30 (Q6), −0.31 (Q4), −0.47 (Q10), and −1.45 (Q5). The most difficult question (1.64 logits) was “Does your face appear distorted to you in the mirror?” and the least difficult (−1.45 logits) was “Do the frames of windows or bookshelves appear distorted to you?”  
Table 1.
 
Individual Item-Fit Statistics for Q1 to Q10
Table 1.
 
Individual Item-Fit Statistics for Q1 to Q10
Item Location Standard Error Item Fit Residuals Degrees of Freedom x 2 Probability
Q1 −0.197 0.334 −0.527 18.81 1.165 0.558
Q2 0.431 0.320 −0.244 23.33 4.607 0.100
Q3 0.252 0.305 2.143 23.33 1.370 0.504
Q4 −0.307 0.289 −0.681 25.59 2.022 0.363
Q5 −1.453 0.282 −1.472 30.10 1.088 0.580
Q6 −0.298 0.291 −0.844 24.08 3.021 0.220
Q7 0.404 0.259 1.774 27.09 2.121 0.346
Q9 −0.471 0.323 1.460 23.33 3.480 0.175
Q10 −0.471 0.278 −0.047 23.33 1.342 0.511
Disordered thresholds were also assessed, and no disordered thresholds were found in the categories used in the questionnaire. The four response categories tended to be used with relatively equal distribution across the nine items. The person separation reliability was 0.97. Determined by the Rasch analysis software, a person separation reliability index of 0.9 is the equivalent of a G value of 3 with the ability to discriminate 4 strata of person ability. Therefore, the nine-item metamorphopsia questionnaire could discriminate among at least four groups of people with different levels of metamorphopsia, and excellent power was indicated for test-of-fit. 
Given that Q4 (Do the columns in your Japanese style rooms appear distorted or tilted to you?) was culturally relevant only to participants in Japan, we also analyzed the eight-item questionnaire without Q4 and the item-trait interaction showed χ2 = 16.85, df = 16, and P = 0.40. If the questionnaire is to be used for other cultures, Q4 can be removed without affecting the fit of the questionnaire to the Rasch model. 
For person measures, the person fit residuals for our subjects ranged from −7.5 to 1.87. Four patients showed residuals exceeding the limit of ± 2.5 (−7.5, −2.84, −2.84, and −2.76). The person with residual of −7.5 had many missing responses, and this could explain why the person was an outlier. The person location ranged from −6.13 to 5.83, and the distribution had a right-sided skew. The mean ± SD for item and person location logit values were 0 ± 0.84 and −1.26 ± 3.41, respectively. 
The Differential Item Functioning (DIF) was investigated for sex, age, and disease type. The probability values ranged from 0.003 to 0.98 for sex, 0.048 to 0.98 for age, and 0.066 to 0.90 for disease type. Compared with the Bonferroni-adjusted P value of 0.006 (0.05/9), none of the probability values were significant. All the DIF values were less than 1, indicating that subjects with equal amounts of metamorphopsia did not tend to respond differently to an item based on their sex, age, or disease type and that the items of the questionnaire appeared to have good reliability. 
Group Comparison
M-CHARTS and the Metamorphopsia Questionnaire.
Overall, the questionnaire and M-CHARTS scores correlated well in all three disease groups (Figs. 3A–C). In the ERM group, the percentage of patients who subjectively perceived any metamorphopsia (i.e., the patients with a questionnaire score >0) significantly increased between the patients with an M-CHARTS score of <0.5 and those with a score ≥1.0 (P < 0.01; Mann-Whitney U test: U = 20; P = 0.005, Fig. 3A; Spearman's rank correlation coefficient: r = 0.59; P = 0.0004). Although not statistically significant, similar positive correlations were also observed in the M-hole (Fig. 3B; r = 0.17; P = 0.48) and AMD (Fig. 3C; r = 0.47; P = 0.05) groups except for those patients whose measurements could not be obtained because of the existence of central scotoma. 
Figure 3.
 
(AC) M-CHARTS and questionnaire scores were compared in the ERM (A), M-hole (B), and AMD (C) groups. In the ERM group, 58.9% of the subjects with an M-CHARTS score <0.5 had a questionnaire score >0. This percentage significantly increased to 100% in those patients with an M-CHARTS score ≥1.0 (P < 0.01; Mann-Whitney U test). A similar tendency was observed in the other two groups. (D, E) PHP and questionnaire results were compared in the ERM (D) and AMD (E) groups. Although the percentage of the subjects with a questionnaire score >0 did not differ significantly between the patients with and without an area of distortion in the ERM group, this percentage significantly increased (from 17% to 85%) between the patients without and with AMD progression in the AMD group. Results of both methods significantly correlated in the AMD group (P < 0.05; Mann-Whitney U test), but not in the ERM group.
Figure 3.
 
(AC) M-CHARTS and questionnaire scores were compared in the ERM (A), M-hole (B), and AMD (C) groups. In the ERM group, 58.9% of the subjects with an M-CHARTS score <0.5 had a questionnaire score >0. This percentage significantly increased to 100% in those patients with an M-CHARTS score ≥1.0 (P < 0.01; Mann-Whitney U test). A similar tendency was observed in the other two groups. (D, E) PHP and questionnaire results were compared in the ERM (D) and AMD (E) groups. Although the percentage of the subjects with a questionnaire score >0 did not differ significantly between the patients with and without an area of distortion in the ERM group, this percentage significantly increased (from 17% to 85%) between the patients without and with AMD progression in the AMD group. Results of both methods significantly correlated in the AMD group (P < 0.05; Mann-Whitney U test), but not in the ERM group.
Regarding the direction of metamorphopsia (vertical or horizontal M-CHARTS score vs. QB), we found that 80.0% of the patients with a greater horizontal M-CHARTS score and 62.5% of the patients with equivalent horizontal and vertical M-CHARTS scores subjectively felt that metamorphopsia occurred horizontally more often (Fig. 4). There was reasonable concordance between the results of both methods regarding direction. 
Figure 4.
 
Of the 39 patients with ERM, 25 patients could select 1 of 3 responses for QB. The correlation in the direction of metamorphopsia between M-CHARTS and questionnaire scores was investigated in the ERM group. For the patients whose horizontal M-CHARTS score was bigger than or equivalent to their vertical M-CHARTS score, more often they subjectively perceived metamorphopsia horizontally.
Figure 4.
 
Of the 39 patients with ERM, 25 patients could select 1 of 3 responses for QB. The correlation in the direction of metamorphopsia between M-CHARTS and questionnaire scores was investigated in the ERM group. For the patients whose horizontal M-CHARTS score was bigger than or equivalent to their vertical M-CHARTS score, more often they subjectively perceived metamorphopsia horizontally.
For the correlation between level of difficulty in a patient's daily life (score of QD) and M-CHARTS score, the patients who reported some level of difficulty in daily life (QD score >0) significantly increased in the ERM group between the patients with an M-CHARTS score of < 0.5 and those with a score ≥1.0 (P < 0.05; Mann-Whitney U test: U = 27.5; P = 0.025; Fig. 5A; Spearman's rank correlation coefficient: r = 0.39; P = 0.019). This indicated that in the ERM group, the patients who had higher M-CHARTS scores also experienced a higher level of difficulty in daily life. However, some patients in the M-hole and the AMD groups reported a great deal of difficulty in daily life (QD score = 3 or 4), even though their metamorphopsia was not so severe (M-CHARTS score <0.5; Figs. 5B, 5C). 
Figure 5.
 
(AC) Of the 39 patients with ERM, 37 patients could select 1 of 3 responses for QD. The relationship between M-CHARTS score and QD (level of difficulty in performing daily activities) in the three groups. In the ERM group, the percentage of the subjects reporting some level of difficulty (QD score >0) significantly increased from 19% in the patients with an M-CHARTS score <0.5 to 75% in those with a score ≥1.0 (P < 0.05; Mann-Whitney U test). However, in the other two groups, at least 10% of the subjects with less severe metamorphopsia (M-CHARTS score <0.5) reported a great deal of difficulty (QD score ≥3). (D, E) Relationship between PHP and QD results in the ERM (D) and AMD (E) groups. No significant difference in the subjects with some level of difficulty (QD score >0) was seen between the patients with and without an area of distortion in the ERM group. In the AMD group, all the patients who were not detected with AMD progression were indicated as free of difficulty.
Figure 5.
 
(AC) Of the 39 patients with ERM, 37 patients could select 1 of 3 responses for QD. The relationship between M-CHARTS score and QD (level of difficulty in performing daily activities) in the three groups. In the ERM group, the percentage of the subjects reporting some level of difficulty (QD score >0) significantly increased from 19% in the patients with an M-CHARTS score <0.5 to 75% in those with a score ≥1.0 (P < 0.05; Mann-Whitney U test). However, in the other two groups, at least 10% of the subjects with less severe metamorphopsia (M-CHARTS score <0.5) reported a great deal of difficulty (QD score ≥3). (D, E) Relationship between PHP and QD results in the ERM (D) and AMD (E) groups. No significant difference in the subjects with some level of difficulty (QD score >0) was seen between the patients with and without an area of distortion in the ERM group. In the AMD group, all the patients who were not detected with AMD progression were indicated as free of difficulty.
PHP and the Metamorphopsia Questionnaire.
The PHP and questionnaire results correlated significantly in the AMD group (r = −0.29; P < 0.05; Mann-Whitney U test: U = 62; P = 0.04) but not in the ERM group (Fig. 3D, 3E; Mann-Whitney U test: U = 166; r = −0.005; P = 0.98). We further compared the PHP results with the level of difficulty in performing daily activities (QD score). ERM patients both with and without a detected area of distortion indicated some level of difficulty in daily life (non-zero QD score), whereas 100% of AMD patients with no consistency with AMD progression indicated no difficulty (Figs. 5D, 5E). This demonstrated that the PHP findings could reflect the level of difficulty in daily life for the patients with AMD but not for those with ERM. 
In this study, we used the patients' total questionnaire scores for comparison with the M-CHARTS and PHP findings. We also compared the questionnaire's Rasch scaled scores with the PHP and M-CHARTS findings, and the correlations were similar to those using patients' total questionnaire scores. The conclusions did not change with the use of Rasch scaled scores compared with the questionnaire total scores. 
The sensitivities of PHP and M-CHARTS for the detection of metamorphopsia noted with Amsler's chart were 41.7% and 88.9% in ERM and 68.4% and 73.7% in AMD, respectively. The respective specificities of PHP and M-CHARTS in the healthy subjects were 71% and 100%; healthy subjects also had a questionnaire score of zero. M-CHARTS showed better sensitivities than PHP in both ERM and AMD groups and a better specificity in the healthy controls (Fig. 6; P < 0.01; Wilcoxon signed-rank test: T = 0; P = 0.0007). 
Figure 6.
 
(A, B) Sensitivities of M-CHARTS and PHP for detecting metamorphopsia in the ERM and AMD groups were compared. M-CHARTS showed better sensitivity in both groups, with significance seen in the ERM group (P < 0.01; Wilcoxon signed-rank test). Specificities of M-CHARTS and PHP for detecting metamorphopsia in the controls were also compared. M-CHARTS clearly demonstrated a superior specificity to PHP (P < 0.01; Wilcoxon signed-rank test).
Figure 6.
 
(A, B) Sensitivities of M-CHARTS and PHP for detecting metamorphopsia in the ERM and AMD groups were compared. M-CHARTS showed better sensitivity in both groups, with significance seen in the ERM group (P < 0.01; Wilcoxon signed-rank test). Specificities of M-CHARTS and PHP for detecting metamorphopsia in the controls were also compared. M-CHARTS clearly demonstrated a superior specificity to PHP (P < 0.01; Wilcoxon signed-rank test).
Discussion
Our results showed that patients' subjective perceptions of metamorphopsia evaluated by the questionnaire correlated well with the assessments by M-CHARTS and PHP in patients with macular diseases. The questionnaire and M-CHARTS scores showed good concordance in detection, severity, and direction of metamorphopsia, especially in patients with ERM, whereas the questionnaire and PHP results significantly correlated in the AMD group (but not in the ERM group). 
Over the visual field, the central field is regarded as the most influential area in our performance of daily life activities. Given that scotomatous lesions often occur in the central field in AMD and M-hole, patients with AMD or M-hole are more likely than patients with ERM to experience additional difficulty in performing daily activities, regardless of the severity of their metamorphopsia. This explained our current result in which some patients in the M-hole and AMD groups had higher questionnaire scores than did patients in the ERM group, even though the M-CHARTS scores were about the same in all three groups (Figs. 3A–C). 
We also showed that the PHP results significantly correlated with the questionnaire scores in the AMD group but not in the ERM group. In the ERM group, the percentage of subjects with a questionnaire score greater than 0 did not significantly differ between the patients with and without an area of distortion indicated by PHP (Fig. 3D), whereas a significant increase was observed in the AMD group between the patients with and without consistency with AMD progression (Fig. 3E). Similarly, ERM patients with and without an area of distortion did not show any significant difference in the level of difficulty in daily activities (Fig. 5D), whereas 100% of AMD patients who were confirmed to have no AMD progression by PHP indicated that they had no difficulties in performing their daily activities (Fig. 5E). Given that PHP was initially designed for the detection of metamorphopsia in patients with AMD, our current results confirmed that PHP could reflect subjective metamorphopsia more closely in patients with AMD than in patients with ERM. Based on these observations, we speculated that PHP might be more sensitive for local visual disturbances such as AMD than for diffuse visual disturbances such as ERM. 
In addition to detection and severity, we investigated the direction of metamorphopsia. There are reports showing that recognition ability is more sensitive horizontally than vertically in the human visual system. 20,21 We have validated that in patients with ERM, the patient's horizontal M-CHARTS score becomes greater than the vertical M-CHARTS score as the stage of the disease advances. 9 Another of our reports further showed the positive correlations in ERM patients between horizontal M-CHARTS score and vertical retinal contraction and between vertical M-CHARTS score and horizontal retinal contraction caused by ERM progression. 10 We therefore considered that ERM might have a closer association with the direction of metamorphopsia than AMD and M-hole. Our current results indicated that in the ERM group, the patients with higher horizontal M-CHARTS score or equivalent vertical and horizontal scores subjectively felt that metamorphopsia occurred more often horizontally. This was in agreement with previous results regarding the direction of metamorphopsia. 
Compared with PHP, M-CHARTS showed better sensitivity in the ERM group and better specificity in the healthy controls (Fig. 6). One possible explanation was that the unit size used in M-CHARTS (0.1°) was smaller than the unit size used in PHP (0.3°). This might have given M-CHARTS superiority over PHP. In addition, M-CHARTS are portable and less expensive than PHP, and the tasks are very easy for patients. However, M-CHARTS and PHP seemed to be sensitive to different types of visual disturbances. They can be effective tools for the detection and quantification of metamorphopsia in patients with various types of macular diseases. 
In this study, the mean ± SD for item location logit value (0 ± 0.84) was higher than for person location logit value (−1.26 ± 3.41). This suggested that the average difficulty of the questionnaire was higher than the patients' average ability and that, overall, the items did not cover the participants of all severity levels. In its present form, the questionnaire is more useful for patients with moderate or severe metamorphopsia. Based on the Rasch analysis findings, 7 of 9 items in our questionnaire appeared to represent tasks with relatively equal and average difficulty. Given that patients could have various lifestyles, we tried to select questions that were applicable to the lifestyles of most patients. For this reason, we would not consider eliminating some of the items of equal difficulty. However, we would recommend adding some items of less difficulty in the future to increase the range and usefulness in a wider range of patients, especially in patients with mild metamorphopsia. 
Although we have demonstrated that the questionnaire could be an effective device to evaluate a patient's subjective perception of metamorphopsia, this study has some potential limitations. We selected subjects with metamorphopsia detected by Amsler's Chart, but patients with a mild form of metamorphopsia that could not be detected by Amsler's Chart but could possibly be detected by M-CHARTS or PHP were not included in this study. Furthermore, because scotoma, reduced visual acuity, and contrast sensitivity often coexist with metamorphopsia, it is difficult for some patients to distinguish between the effects of metamorphopsia and these other visual defects. Despite these limitations, our questionnaire can be informative about a patient's subjective perception of metamorphopsia. 
In conclusion, we have shown that the metamorphopsia questionnaire can supplement PHP in patients with AMD and can supplement M-CHARTS in patients with ERM and AMD but not in patients with M-hole. Metamorphopsia can be assessed only by clinical measures because no methods have been available for the subjective evaluation of metamorphopsia. Using the questionnaire as a supplement to other clinical testing, ophthalmologists can obtain a score that measures the patient's subjective perception of metamorphopsia and can compare the score with that of other clinical measurements. In future, this questionnaire could be further modified for people of other cultures and revalidated to attempt to assist patients with less severe metamorphopsia. 
Footnotes
 Supported by the Ministry of Education of the Japanese Government (Grant 12671731, 2000) and the Osaka Medical Research Foundation for Incurable Diseases.
Footnotes
 Disclosure: E. Arimura, None; C. Matsumoto, P; H. Nomoto, None; S. Hashimoto, None; S. Takada, None; S. Okuyama, None; Y. Shimomura, None
The authors thank Reiyo Tahara for her editorial support. 
References
Hollones M Bunn KW . The relation between convergence micropsia and retinal eccentricity. Vision Res. 1977;17:403–408. [CrossRef] [PubMed]
Freisen L Freisen M . Micropsia and visual acuity in macula edema: a study of the neuro-retinal basis of visual acuity. Albrecbt Von Graefes Arcb Klin Exp Ophthalmol. 1979;210:69–77. [CrossRef]
Linblom B . Measuring micropsia in retinal swelling: a monocular approach. Chibret Int J Ophthalmol. 1987;48–54.
Lakshminarayanan V . Quantification of metamorphopsia using hyperacuity techniques. Optom Vis Sci. 1991;68:942–945. [CrossRef] [PubMed]
Shinoda K Ishida S Kawashima S Matszaki T Yamada K Katsura H . A new method for quantification of metamorphopsia in patients with epiretinal membrane. Jpn J Ophthalmol. 2000;44:424–427. [CrossRef] [PubMed]
Matsumoto C Tsuboi S Okuyama S Uyama K Otori T . Quantification of metamorphopsia: method of evaluation [in Japanese]. Ganka Rinsho Iho. 1985;79:663–667.
Amsler M . Earliest symptoms of diseases of the macula. Br J Ophthalmol. 1953;37:521–537. [CrossRef] [PubMed]
Matsumoto C Arimura E Hashimoto S Takada S Okuyama S Shimomura Y . A new method for quantification of metamorphopsia using M-CHARTS. Rinsho Ganka. 2000;54:373–377.
Matsumoto C Arimura E Hashimoto S Takada S Okuyama S Shimomura Y . Quantification of metamorphopsia in patients with epiretinal membranes. Invest Ophthalmol Vis Sci. 2003;44:4012–4016. [CrossRef] [PubMed]
Arimura E Matsumoto C Okuyama S Takada S Hashimoto S Shimomura Y . Retinal contraction and metamorphopsia scores in eyes with idiopathic epiretinal membrane. Invest Ophthalmol Vis Sci. 2005;46:2961–2966. [CrossRef] [PubMed]
Arimura E Matsumoto C Okuyama S Takada S Hashimoto S Shimomura Y . Quantification of metamorphopsia in a macular hole patient using M-CHARTS. Acta Ophthalmol Scand. 2007;85:55–59. [CrossRef] [PubMed]
Loewenstein A Malach R Goldstein M . Replacing the Amsler grid: a new method for monitoring patients with age-related macular degeneration. Ophthalmology. 2003;110:966–970. [CrossRef] [PubMed]
Bernth-Peterson P . Visual functioning in cataract patients: methods of measuring and result. Acta Ophthalmol (Copenh). 1981;59:198–205. [CrossRef] [PubMed]
Bernth-Peterson P . Outcome of cataract surgery: visual functioning in aphakic patients. Acta Ophthalmol (Copenh). 1982;60:243–251. [CrossRef] [PubMed]
Steinberg EP Tielsch JM Schein OD . The VF14: an index of functioning in aspects. Arch Ophthalmol. 1994;112:630–638. [CrossRef] [PubMed]
Nakajima M . Clinical study of epiretinal membrane: classification of idopathic epiretinal membrane [in Japanese]. Ganka Rinsho Iho. 1985;79:633–667.
Gass JDM . Reappraisal of biomicroscopic classification of stages of development of a macular hole. Am J Ophthalmol. 1995;119:752–759. [CrossRef] [PubMed]
Enoch JM Williams RA Essock EA Barricks M . Hyperacuity perimetry: assessment of macular function through ocular opacities. Arch Ophthalmol. 1984;102:1164–1168. [CrossRef] [PubMed]
Baraldi P Enoch JM Raphael S . Visual through nuclear and posterior subcapsular cataract. Int Ophthalmol. 1986;9:173–178. [CrossRef] [PubMed]
Hughe A . The topography of vision in mammals of contrasting life style: comparative optics and retinal organization. Handbook of Sensory Physiology. Berlin: Springer-Verlag; 1984;613–756.
Azuma N . Monocular cell biology on morphogenesis of the fovea and evolution of the central vision. Nippon Ganka Gakkai Zasshi. 2000;104:960–985. [PubMed]
Figure 1.
 
The metamorphopsia questionnaire we designed included 10 questions on the symptoms of subjective metamorphopsia in patient's daily life and four other questions with designated aims. The average score of Q1 to Q10 was used as the patient's questionnaire score for the assessment of subjective metamorphopsia.
Figure 1.
 
The metamorphopsia questionnaire we designed included 10 questions on the symptoms of subjective metamorphopsia in patient's daily life and four other questions with designated aims. The average score of Q1 to Q10 was used as the patient's questionnaire score for the assessment of subjective metamorphopsia.
Figure 2.
 
M-CHARTS have 19 charts and 2 types: type I with one dotted line for patients with general macular diseases and type II with two dotted lines for patients with M-hole. The visual angle that denoted the dot interval of the line seen straight was used as patient's M-CHARTS score. Both vertical and horizontal tests were performed to obtain separate scores.
Figure 2.
 
M-CHARTS have 19 charts and 2 types: type I with one dotted line for patients with general macular diseases and type II with two dotted lines for patients with M-hole. The visual angle that denoted the dot interval of the line seen straight was used as patient's M-CHARTS score. Both vertical and horizontal tests were performed to obtain separate scores.
Figure 3.
 
(AC) M-CHARTS and questionnaire scores were compared in the ERM (A), M-hole (B), and AMD (C) groups. In the ERM group, 58.9% of the subjects with an M-CHARTS score <0.5 had a questionnaire score >0. This percentage significantly increased to 100% in those patients with an M-CHARTS score ≥1.0 (P < 0.01; Mann-Whitney U test). A similar tendency was observed in the other two groups. (D, E) PHP and questionnaire results were compared in the ERM (D) and AMD (E) groups. Although the percentage of the subjects with a questionnaire score >0 did not differ significantly between the patients with and without an area of distortion in the ERM group, this percentage significantly increased (from 17% to 85%) between the patients without and with AMD progression in the AMD group. Results of both methods significantly correlated in the AMD group (P < 0.05; Mann-Whitney U test), but not in the ERM group.
Figure 3.
 
(AC) M-CHARTS and questionnaire scores were compared in the ERM (A), M-hole (B), and AMD (C) groups. In the ERM group, 58.9% of the subjects with an M-CHARTS score <0.5 had a questionnaire score >0. This percentage significantly increased to 100% in those patients with an M-CHARTS score ≥1.0 (P < 0.01; Mann-Whitney U test). A similar tendency was observed in the other two groups. (D, E) PHP and questionnaire results were compared in the ERM (D) and AMD (E) groups. Although the percentage of the subjects with a questionnaire score >0 did not differ significantly between the patients with and without an area of distortion in the ERM group, this percentage significantly increased (from 17% to 85%) between the patients without and with AMD progression in the AMD group. Results of both methods significantly correlated in the AMD group (P < 0.05; Mann-Whitney U test), but not in the ERM group.
Figure 4.
 
Of the 39 patients with ERM, 25 patients could select 1 of 3 responses for QB. The correlation in the direction of metamorphopsia between M-CHARTS and questionnaire scores was investigated in the ERM group. For the patients whose horizontal M-CHARTS score was bigger than or equivalent to their vertical M-CHARTS score, more often they subjectively perceived metamorphopsia horizontally.
Figure 4.
 
Of the 39 patients with ERM, 25 patients could select 1 of 3 responses for QB. The correlation in the direction of metamorphopsia between M-CHARTS and questionnaire scores was investigated in the ERM group. For the patients whose horizontal M-CHARTS score was bigger than or equivalent to their vertical M-CHARTS score, more often they subjectively perceived metamorphopsia horizontally.
Figure 5.
 
(AC) Of the 39 patients with ERM, 37 patients could select 1 of 3 responses for QD. The relationship between M-CHARTS score and QD (level of difficulty in performing daily activities) in the three groups. In the ERM group, the percentage of the subjects reporting some level of difficulty (QD score >0) significantly increased from 19% in the patients with an M-CHARTS score <0.5 to 75% in those with a score ≥1.0 (P < 0.05; Mann-Whitney U test). However, in the other two groups, at least 10% of the subjects with less severe metamorphopsia (M-CHARTS score <0.5) reported a great deal of difficulty (QD score ≥3). (D, E) Relationship between PHP and QD results in the ERM (D) and AMD (E) groups. No significant difference in the subjects with some level of difficulty (QD score >0) was seen between the patients with and without an area of distortion in the ERM group. In the AMD group, all the patients who were not detected with AMD progression were indicated as free of difficulty.
Figure 5.
 
(AC) Of the 39 patients with ERM, 37 patients could select 1 of 3 responses for QD. The relationship between M-CHARTS score and QD (level of difficulty in performing daily activities) in the three groups. In the ERM group, the percentage of the subjects reporting some level of difficulty (QD score >0) significantly increased from 19% in the patients with an M-CHARTS score <0.5 to 75% in those with a score ≥1.0 (P < 0.05; Mann-Whitney U test). However, in the other two groups, at least 10% of the subjects with less severe metamorphopsia (M-CHARTS score <0.5) reported a great deal of difficulty (QD score ≥3). (D, E) Relationship between PHP and QD results in the ERM (D) and AMD (E) groups. No significant difference in the subjects with some level of difficulty (QD score >0) was seen between the patients with and without an area of distortion in the ERM group. In the AMD group, all the patients who were not detected with AMD progression were indicated as free of difficulty.
Figure 6.
 
(A, B) Sensitivities of M-CHARTS and PHP for detecting metamorphopsia in the ERM and AMD groups were compared. M-CHARTS showed better sensitivity in both groups, with significance seen in the ERM group (P < 0.01; Wilcoxon signed-rank test). Specificities of M-CHARTS and PHP for detecting metamorphopsia in the controls were also compared. M-CHARTS clearly demonstrated a superior specificity to PHP (P < 0.01; Wilcoxon signed-rank test).
Figure 6.
 
(A, B) Sensitivities of M-CHARTS and PHP for detecting metamorphopsia in the ERM and AMD groups were compared. M-CHARTS showed better sensitivity in both groups, with significance seen in the ERM group (P < 0.01; Wilcoxon signed-rank test). Specificities of M-CHARTS and PHP for detecting metamorphopsia in the controls were also compared. M-CHARTS clearly demonstrated a superior specificity to PHP (P < 0.01; Wilcoxon signed-rank test).
Table 1.
 
Individual Item-Fit Statistics for Q1 to Q10
Table 1.
 
Individual Item-Fit Statistics for Q1 to Q10
Item Location Standard Error Item Fit Residuals Degrees of Freedom x 2 Probability
Q1 −0.197 0.334 −0.527 18.81 1.165 0.558
Q2 0.431 0.320 −0.244 23.33 4.607 0.100
Q3 0.252 0.305 2.143 23.33 1.370 0.504
Q4 −0.307 0.289 −0.681 25.59 2.022 0.363
Q5 −1.453 0.282 −1.472 30.10 1.088 0.580
Q6 −0.298 0.291 −0.844 24.08 3.021 0.220
Q7 0.404 0.259 1.774 27.09 2.121 0.346
Q9 −0.471 0.323 1.460 23.33 3.480 0.175
Q10 −0.471 0.278 −0.047 23.33 1.342 0.511
×
×

This PDF is available to Subscribers Only

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

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

×