The Longitudinal Impact of Macular Telangiectasia (MacTel) Type 2 on Vision-Related Quality of Life

Abstract

Purpose.: To determine the longitudinal impact of idiopathic macular telangiectasia (MacTel) type 2 on vision-specific quality of life (QoL).

Methods.: Participants with MacTel type 2 and controls with no vision impairment were recruited at baseline. All underwent a full ophthalmic examination and the interview-administered 28-item Impact of Vision Impairment (IVI) questionnaire at baseline, to gather information on sociodemographic factors and vision-specific QoL. The MacTel patients were reassessed at 24 months.

Results.: For the MacTel participants (n = 22), the median (range) age and visual acuity were 64 years (45–87) and 20/32 (20/13–20/63) in the better eye, respectively. The corresponding median values in the control group (n = 38) were 57 years (41–68) and 20/25 or better in both eyes, respectively. Rasch analysis showed that the IVI and its three subscales had sufficient psychometric validity and possessed interval level estimates. The control group had almost twice the level of vision-specific QoL than did the MacTel group at baseline on all IVI scores (P < 0.001 for all). At 24 months, we found no significant change in any of the IVI scores in the MacTel group. Similarly, there was no significant difference in visual acuity in the better or worse eyes in that group after 2 years.

Conclusions.: Persons with MacTel type 2 had poorer overall vision-related QoL than did healthy controls. Several aspects of vision-related QoL and distance visual acuity did not significantly change after 24 months. Longer follow-up assessment periods are needed to determine the longitudinal impact of this condition on vision-related QoL.

Idiopathic macular telangiectasia (MacTel) type 2 is a bilateral eye disease that is characterized by progressive vision loss. In the early, nonproliferative phase, clinical changes in the macula consist of a reduction in retinal transparency and the formation of telangiectatic capillaries, predominantly temporal to the fovea. ^{1 –3} As the disease advances to the proliferative stages, these telangiectatic changes encompass the entire juxtafoveal area, resulting in the development of subretinal neovascularization in some cases. ^{1 –3} Most patients with MacTel type 2 have some degree of bilateral visual loss, and a small number have severe visual impairment due to the development of neovascularization. ⁴  

Little is known about the cause, pathogenesis, and anatomic basis of MacTel type 2, ⁵ and information about the impact of the condition from the patient's perspectives is almost nonexistent. The International MacTel Consortium was set up in 2005 to provide answers to these questions. Clemons et al. ⁴ from the consortium have described the impact of MacTel type 2 on vision-targeted functioning by using the National Eye Institute Visual Functioning Questionnaire (NEI-VFQ-25). They found that patients with MacTel type 2 reported reduced visual functioning compared with that reported by normal controls. ⁴ To date, however, our knowledge of the impact of MacTel type 2 on aspects of daily life other than vision-related functioning is limited, as is any information about changes in vision-specific quality of life (QoL) over time. Monitoring changes in visual function or QoL in MacTel patients over time is important, and such information may be useful in intervention trials to determine the efficacy of new treatment options. 

Although MacTel type 2 has traditionally been considered a rare eye condition, Klein et al. ⁶ recently reported its prevalence in the Beaver Dam Eye study to be 0.1%, thus affecting approximately 70,000 people in the United States. With a higher than previously thought prevalence, it is therefore important to assess the impact of this condition from the patients' perspective. The Impact of Vision Impairment (IVI) Questionnaire is a vision-specific QoL tool that has been developed to measure participation in activities of daily living in patients with vision impairment. ^{7 –10} The IVI has been extensively validated by modern psychometric theory, ^{7 –10} and its use is recommended by the American Academy of Ophthalmology to assess the impact of vision impairment on QoL. ¹¹ Therefore, the purpose of this study was to use the Melbourne site of the MacTel Consortium to determine with the IVI the baseline and longitudinal impact of MacTel type 2 on patients' vision-specific QoL. We hypothesized that the deterioration in vision induced by MacTel would be associated with a worsening in vision-specific QoL parameters at the 24-month follow-up examination. 

Methods

The MacTel Study is an ongoing 5-year observational study conducted in 30 clinical centers in seven countries. Individuals 18 years of age or older with a clinical diagnosis of MacTel type 2 are invited to participate in the study. Each center gained approval to conduct the study from the institutional review board or independent ethics committee, and the protocol complied with the guidelines of the Declaration of Helsinki. The clinical diagnosis of MacTel type 2 is initially determined by the principal investigator at each clinical center by fundus photography, ocular coherence tomography (OCT), fluorescence angiogram, and auto fluorescence. Diagnosis is later confirmed by image grading performed at the Fundus Reading Centre of Moorfield's Eye Hospital in London. After providing consent, all patients undergo a full ophthalmic examination, including best corrected visual acuity (BCVA) and indirect ophthalmoscopy. Participants also undergo interview-administered general questionnaires to collect information on sociodemographic factors, lifestyle parameters, and medical and family histories. 

The IVI Questionnaire

In the MacTel Melbourne Cohort, vision-specific QoL was assessed with the 28-item IVI questionnaire (Table 1). The development of the IVI is described elsewhere. ^{12 –16} The three subscales are mobility and independence (11 items), emotional well-being (8 items), and reading and accessing information (9 items). ¹³ The IVI was administered by experienced interviewers to MacTel type 2 participants at a baseline and 24-month follow-up visit. A healthy control group from Melbourne (n = 38) with no vision impairment or eye condition (logMAR <0.2) also completed the IVI questionnaire. 

Table 1.

 

The Specific Items of the IVI Questionnaire

Questions	Domain
1. Your ability to see and enjoy TV?	Reading and accessing information
2. Taking part in recreational activities such as bowling, walking, or golf?	Mobility and independence
3. Shopping? (finding what you want and paying for it)	Reading and accessing information
4. Visiting friends or family?	Mobility and independence
5. Recognizing or meeting people?	Reading and accessing information
6. Generally looking after your appearance?	Reading and accessing information
7. Opening packaging?	Reading and accessing information
8. Reading labels or instructions on medicines?	Reading and accessing information
9. Operating household appliances and the telephone?	Reading and accessing information
10. How much has your eyesight interfered with getting about outdoors? (on the pavement or crossing the street)	Mobility and independence
11. In the past month, how often has your eyesight made you go carefully to avoid falling or tripping?	Mobility and independence
12. In general, how much has your eyesight interfered with travelling or using transport? (bus and train)	Mobility and independence
13. Going down steps, stairs, or curbs?	Mobility and independence
14. Reading ordinary size print? (for example newspapers)	Reading and accessing information
15. Getting information that you need?	Reading and accessing information
16. Your general safety at home?	Mobility and independence
17. Spilling or breaking things?	Mobility and independence
18. Your general safety when out of your home?	Mobility and independence
19. In the past month, how often has your eyesight stopped you doing the things you want to do?	Mobility and independence
20. In the past month, how often have you needed help from other people because of your eyesight?	Mobility and independence
21. Have you felt embarrassed because of your eyesight?	Emotional well-being
22. Have you felt frustrated or annoyed because of your eyesight?	Emotional well-being
23. Have you felt lonely or isolated because of your eyesight?	Emotional well-being
24. Have you felt sad or low because of your eyesight?	Emotional well-being
25. In the past month, how often have you worried about your eyesight getting worse?	Emotional well-being
26. In the past month how often has your eyesight made you concerned or worried about coping with everyday life?	Emotional well-being
27. Have you felt like a nuisance or a burden because of your eyesight?	Emotional well-being
28. In the past month, how much has your eyesight interfered with your life in general?	Emotional well-being

 Three rating scales were used.

Statistical Analysis

To determine the validity of the IVI in this sample, Rasch analysis was conducted with the Andrich single rating scale model ¹⁷ with Winsteps software (ver. 3.68). ¹⁸ Rasch analysis states that the probability that an individual will choose a response for a particular item depends on both person (dis)ability and item difficulty. Thus, Rasch analysis provides a strict model that the structure of the responses should satisfy, rather than a simple statistical description of the responses. It converts the categorical data into a linear interval scale. ^{19 –23} We used three Andrich rating scale models, ²⁴ one for each type of rating scale found in the IVI questionnaire, to assess the pattern of response category thresholds. The presence of disordered thresholds indicates that participants have difficulty discriminating between the scale response options, and the collapsing of categories may be necessary. The person separation index is the ratio of the true variance in the estimated measures to the observed variance and indicates the ability of the scale to distinguish distinct levels of participant ability. A person separation index of 2.0 with a person reliability score of 0.8 represents three distinct levels of participant QoL. ¹⁹ Targeting of the scale items to the participants' ability is assessed by inspecting the person-item maps, where the person and item measures are displayed (in log of the odd units, or logits) on the same calibration ruler. ^19,25 The unidimensionality of the scale is assessed by its ability to measure a single underlying construct, which is measured using two parameters: item fit statistics and principal component analysis (PCA). Item fit determines how each item fits the underlying trait, vision-related QoL. Items with an infit mean square value ranging between 0.7 and 1.3 were considered acceptable. According to the PCA of the residuals, the variance explained by the Rasch measures for the empiric calculation should be comparable to that of the model (>50% for an acceptable model). Furthermore, the unexplained variance by the residuals in the first contrast should be <3.0 Eigen value units, which is close to that seen with random data. Finally, we assessed for differential item functioning (DIF), which indicates whether different groups within the sample (e.g., sex and age) systematically respond differently, despite the equal levels of the trait being assessed. We used the following criteria for DIF assessment: small or absent if the difference was <0.5 logits, 0.5 to 1 logit as minimal DIF, and <1.0 logit as notable DIF. If significant and meaningful DIF is found, it may indicate that the interpretation of the scale differs by group and that the scale is be influenced by confounding factor(s). 

It was important to establish that differences between the IVI scores at baseline and follow-up are valid indicators of changes over time. ²⁶ Consequently, the baseline and follow-up data were stacked, and the absence of DIF was used to establish invariance over time. As suggested earlier, DIF occurs when groups of scores within the sample (e.g., baseline and follow-up), despite equal levels of the underlying characteristic being measured, respond in a different manner to an individual item. This technique has been described previously. ²⁶  

Vision-specific QoL values for the overall IVI and the three subscale scores were the main outcomes examined statistically (SPSS statistical software, version 17.0; SPSS Science, Chicago, IL). Mean positive logit values denoted high vision-specific QoL, and negative scores indicated poor QoL. Descriptive statistical analyses were performed to characterize the participants' sociodemographic and clinical characteristics and IVI responses. At baseline, a one-way ANOVA was used to determine differences between the MacTel and control participants with regard to the dependent measures. At 24 months, paired t-tests were used to determine changes in QoL scores over time in the MacTel group. A level of P < 0.05 was chosen as the criterion for significance. 

Results

At baseline, 23 participants (14 women and 9 men) with MacTel type 2, aged between 45 and 87 years (median, 64) were enrolled in the study (Table 2). Most (n = 20; 87.0%) were of white/Caucasian ethnic background. Self-reported medical conditions at baseline included hypertension (39.1%), diabetes (17.4%), cancer (17.4%), coronary artery disease (8.7%), thyroid disease (4.3%), and cerebral vascular accident (4.3%). Of those with diabetes, all had type II with a minimum disease duration of 10 years. The median (range) visual acuity in the MacTel group was 20/32 (20/13–20/63) and 20/40 (20/20–20/160) in the better and worse eyes, respectively. Mean visual acuity in the better and worse eyes did not differ by age, sex, or racial category. All MacTel type 2 participants were reassessed at 24 months. 

Table 2.

 

Baseline Demographic Characteristics of MacTel Study Participants (n = 23)

Characteristics
Age, y
Mean (±SD)	64.1 (10.9)
Median (range)	64 (45–87)
Sex, n (%)
Female	14 (60.9)
Male	9 (39.1)
Race, n (%)
Caucasian	20 (87.0)
Non-Caucasian*	3 (13.0)
General health, n (%)
Hypertension	9 (39.1)
Diabetes (only type II)	4 (17.4)
Mean duration, y	10
Cancer	4 (17.4)
Coronary artery disease	2 (8.7)
Thyroid disease	1 (4.3)
Cerebral vascular accident	1 (4.3)
Age in years at diagnosis, mean (±SD)	56.4 (10.3)
Better eye visual acuity, median (range)	20/32 (20/13–20/63)
Worse eye visual acuity, median (range)	20/40 (20/20–20/160)

*  Asian Indians.

The control group consisted of 24 women and 14 men with a median (range) age of 57 years (41–68) and presenting visual acuity of 20/25 or better in both eyes. As anticipated, we found a statistically significant difference (P < 0.001) in visual acuities between the control and MacTel group at baseline in the better and worse eyes. 

Validation of the IVI

The psychometric validation of the IVI and its subscales showed no disordered thresholds (suggesting that the IVI response categories were working as intended) or item misfit. The person separation reliability of the IVI scores was adequate and ranged from 0.78 to 0.92, indicating that the scale could effectively differentiate three and four distinct levels of participant QoL. There was no evidence of multidimensionality, which indicated that each IVI score represented a single underlying construct. Targeting of the items showed a floor effect, as the mean person ability (2.22 ± 1.74 logits) was higher than the required ability of the items (0.00 logits). This finding suggests that MacTel type 2 participants had good vision-specific QoL and thus could benefit by including items of higher difficulty. There was no evidence of DIF regarding age and sex in this sample for the four different IVI scores. This result implies that the IVI scores were not influenced by sex and age but by the level of vision-specific QoL. In addition, we found no evidence of DIF for time (baseline-follow-up) for the IVI and its three subscales, suggesting that potential changes at follow-up were related, not to noise or artifact but to changes in person level of vision-specific QoL. 

Impact of MacTel on Vision-Specific QoL

Mean person measures (in logits) at baseline for the overall IVI and three subscale scores are shown in Table 3. The positive scores for the four IVI measures suggest that the MacTel group had little difficulty undertaking the items in the questionnaire. The most difficult items at baseline were item 14, reading ordinary print, and item 8, reading labels or instructions from medicines. The two least difficult items were item 17, spilling or breaking things, and item 16, general safety at home. As anticipated, there were statistically significant differences between the control and MacTel groups at baseline on all four IVI scores (P < 0.001 for all). With high mean positive values ranging between 4.6 and 6.6 logits, the control group had almost twice the level of vision-specific QoL as the MacTel group had. These findings demonstrate that the IVI has substantial discriminative validity. We found a significant relationship between visual acuity and the four IVI scores at baseline (P < 0.05; Spearman correlation; Table 4). The association was marginally stronger in the worse eye than in the better eye. We noted a higher association of visual acuity in both eyes with the reading and accessing information domain than with the other IVI scores. 

Table 3.

 

Person Measures for IVI Overall and Three Subscale Scores for the Control and MacTel groups

Person Measures	Control Group Baseline	MacTel Group
		Baseline	24 Months
Overall score	5.49 ± 1.09	2.22 ± 1.74*	2.09 ± 2.04
Reading and accessing information	4.58 ± 0.98	1.49 ± 1.87*	1.47 ± 2.08
Mobility and independence	5.24 ± 0.43	2.95 ± 1.79*	3.01 ± 2.13
Emotional well-being	6.58 ± 0.88	3.41 ± 2.55*	2.68 ± 2.91

 Data are expressed at the mean ± SD.

*  Significant differences between these two groups at baseline (P < 0.001).

Table 4.

 

Spearman Rank Correlations of Visual Acuity in the Better- and Worse-Seeing Eyes and the IVI Scale for the Overall and Three Subscale Scores

	Overall IVI VSQoL	Reading and Accessing Information	Mobility and Independence	Emotional Well-Being
Better eye	−0.367*	−0.387*	−0.282†	−0.245†
Worse eye	−0.388*	−0.410*	−0.302†	−0.265†

*  Correlation is significant at the 0.01 level (two-tailed).

†  Correlation is significant at the 0.05 level (two-tailed).

The IVI scores in the MacTel group at the 24-month time point are shown in Table 3. We found no statistically significant difference between follow-up person measures and baseline IVI scores (P > 0.05 for all). As vision-specific QoL is strongly associated with visual acuity, we compared baseline and follow-up visual acuity levels in the MacTel group. We found no statistically significant difference between the better eye (20/32 [20/16–20/80) vs. 20/32 [20/13–20/63]; P = 0.23) and the worse eye (20/63 [20/25–20/800] vs. 20/40 [20/20–20/160]; P = 0.09) for the mean (range) values in that group. 

Discussion

This is the first longitudinal study to report the impact of idiopathic MacTel type 2 on vision-specific QoL according to a Rasch-calibrated instrument. At baseline, we found that, compared with healthy control subjects, people living with MacTel have poorer overall vision-related QoL as well as specific aspects of vision-related QoL. After 24 months, we discovered no significant change in the MacTel group with regard to the overall vision-specific QoL score or in the specific aspects of daily living associated with the three life domains: mobility and independence, emotional well-being, and reading and accessing information. In our group of affected individuals, we also did not find any significant longitudinal alteration in distance vision in the better and worse eyes over the 24 months. Notwithstanding our small sample size, our findings indicate that the impact of MacTel on visual acuity did not change significantly over a 2-year period and, as the IVI scores correlated moderately with visual acuity, our finding of no significant longitudinal change in QoL parameters would therefore be expected. 

Although our sample was small and comprised only MacTel patients from Melbourne (Australia), our clinical and sociodemographic data were very similar to those recently reported by Clemons et al. ²⁷ in a large multicenter study involving 310 MacTel type 2 patients. For example, they reported mean visual acuities of approximately 20/32 (Snellen equivalents) and 20/50 for the better and worse eyes, respectively. These values are very similar to our median values of 20/32 and 20/40, respectively. These values suggest that our data should be valid for the larger group of MacTel patients. 

The only other study to investigate the health-related QoL in MacTel type 2 is reported from the entire MacTel Consortium where the National Eye Institute Visual Function Questionnaire (NEI-VFQ-25) was applied to all participants in the study. ⁴ Similar to the findings from the entire group of MacTel participants, we found that compared with people with no ocular co-morbidity, people with MacTel type 2 had significantly poorer overall vision-specific QoL and poorer participation in daily living tasks associated with mobility and independence, emotional well-being, and reading and accessing information. Our finding that the two most difficult items experienced by this group are associated with reading is not surprising, considering that this condition tends to affect the macula, an ocular area responsible for processing sharp and clear vision-dependent activities. 

Interestingly, we found that compared with no ocular condition, MacTel type 2 has a significant impact on vision-specific emotional well-being. This is a novel finding for this particular group, as Clemons et al. ⁴ did not find a significant association when using the NEI-VFQ-25. One reason for this discrepancy could be that the mental domain of the NEI-VFQ-25 has only four items, compared with the IVI, which has an eight-item emotional domain. Item sufficiency is an important prerequisite to reliably assess a trait (i.e., QoL), and many of the NEI-VFQ-25 subtraits (including the mental one) contain too few items to provide an adequate and reliable measure. Two recent studies have highlighted this structural limitation of the NEI-VFQ, and both have empirically suggested that a two-domain structure may provide a better understanding of the impact of vision loss from the patient's perspective. ^28,29  

The use of Rasch analysis to validate the IVI in this sample and generate person measures with interval-level properties is a significant strength of this study. Previous work in this disease group has used Likert or summary scoring to assess impact on patient reported outcomes. ⁴ This method assumes erroneously that assigning numbers to responses represents fundamental measurement, and that all items in an instrument are equal in difficulty. Modern psychometric methods have categorically rejected both hypotheses. By showing that our MacTel data fit the Rasch model, we empirically demonstrate that our IVI scores possess estimates of interval-level measurement properties (similar to a ruler). This transformation is important, as it improves measurement precision and enables valid parametric statistical analysis of the output. ³⁰ By using Rasch analysis, we also demonstrated that our four IVI scores are unidimensional and that each score measures a single underlying construct, which is a key indicator of instrument validity. However, our Rasch validation of the IVI also indicated that the targeting of the scale items to this visually able group is suboptimal, suggesting that items of greater visual difficulty are needed to optimally assess the impact of MacTel on vision-related QoL in this group. 

In conclusion, we observed MacTel type 2 patients for a 24-month period and found no significant deterioration in four vision-specific QoL indices in patients living with MacTel type 2 after the initial assessment. Importantly, there was also no significant deterioration in visual acuity which, to an extent, substantiates our finding of no significant change in vision-specific QoL scores, as assessed by the IVI questionnaire. Knowledge about the progression and impact of MacTel is scarce, and longitudinal data are therefore critical. Longer follow-up assessment time points may provide more elucidative information about the longitudinal impact of this condition on vision-specific QoL. 

Footnotes

 Supported by The Lowy Medical Research Institute Limited, a National Health and Medical Research Council (NHMRC) Career Development Award Fellowship (ELL), an NHMRC Early Career Fellowship (MD), an NHMRC Clinical Fellowship (RHG), and the Victorian Government Operational Infrastructure.

Footnotes

 Disclosure: E.L. Lamoureux, None; R.M. Maxwell, None; M. Marella, None; M. Dirani, None; E. Fenwick, None; R.H. Guymer, None