|Year : 2023 | Volume
| Issue : 1 | Page : 32-37
Retrospective analysis of low vision assistive products – A 6-year review
Bharat Gurnani1, Kirandeep Kaur2, Priya Sivakumar3, Sahil Bhandari4
1 Department of Cataract, Cornea and Refractive Services, Aravind Eye Hospital, Puducherry, India
2 Department of Cataract, Pediatric Ophthalmology and Strabismus Services, Aravind Eye Hospital, Puducherry, India
3 Department of Cataract, Neuro-Ophthalmology and Low Vision Services, Aravind Eye Hospital, Puducherry, India
4 Department of Cataract and Vitreoretinal Services, Guru Hasti Chikitsalaya, Pipar City, Jodhpur, Rajasthan, India
|Date of Submission||19-Nov-2021|
|Date of Decision||01-Mar-2022|
|Date of Acceptance||09-Mar-2022|
|Date of Web Publication||09-Mar-2023|
Department of Pediatric Ophthalmology and Strabismus, Aravind Eye Hospital, Thavalakuppam, Cuddalore Main Road, Puducherry - 605 007
Source of Support: None, Conflict of Interest: None
PURPOSE: The purpose of the study was to analyze the demographics, visual acuity (VA), etiologies, recommended low vision assistive products (LVAP), and the acceptance rates of LVAP in various age groups.
METHODS: This was a long-term retrospective review of all the patients presenting to the low vision clinic of our tertiary eye care hospital from January 2011 to December 2016. Data obtained included age, gender, VA, visual fields, ocular pathology causing the low vision, and types of LVAP advised. The primary outcome was to analyze the type of LVAP prescribed in different age groups, and the secondary outcome was the acceptance rate of LVAP.
RESULTS: We analyzed the results of 8309 patients, out of which 2844 (34%. 2) were <15 years of age, 2425 (29.5%) were between 16-40 years, and 3013 (36.3%) were above 40 years. A total of 5522 (66.4%) had best-corrected visual acuity (BCVA) ranging from 6/18-3/60, and 2796 (33.6%) had BCVA from 3/60-No PL. Approximately 38% improved with LVAPs. The most common etiology was retinitis pigmentosa in 1545 (18.6%) patients, followed by congenital nystagmus in 1482 (17.8%), and the least was albinism 383 (4.6%). Maximum prescribed and accepted LVAP were hand and stand magnifiers among 1017 (44.3%) and 512 (52.6%) patients, respectively.
CONCLUSION: Products that are easy to use, require lesser adaptability, are cheap, and require lower maintenance have maximum acceptance rates. We suggest that great emphasis should be laid on training, education, and guidance for low vision rehabilitation centers.
Keywords: Acceptance rates, low vision assistive products, low vision, magnifiers
|How to cite this article:|
Gurnani B, Kaur K, Sivakumar P, Bhandari S. Retrospective analysis of low vision assistive products – A 6-year review. Saudi J Ophthalmol 2023;37:32-7
| Introduction|| |
The term visual impairment (VI) collectively refers to “low vision” and “blindness.” As per the 10th revision of the World Health Organization (WHO) International Statistical Classification of Diseases, “low vision” is defined as visual acuity (VA) between 6/18 and 3/60 in the better eye with the best possible correction, and “Blindness” is defined as VA of <3/60, in the better eye with the best possible correction. Approximately 253 million people worldwide are estimated to be living with VI, 36 million are blind, and 217 million have moderate-to-severe VI. Over 90% of all VI can be prevented or cured with cost-effective interventions. Approximately 826 million people are living with a near vision impairment. About 90% of patients with VI reside in developing countries, with more than half of them in Asia and a vast majority in rural areas. There are about 45 million visually handicapped people in India alone. Patients with low vision have difficulty doing daily activities and often have a low quality of life index (QLI). Low vision leads to impaired mobility, reading difficulties, functional impairment, and above all, loss of personal independence.
The causes of low vision can be categorized as congenital or acquired. The common congenital causes include hereditary dystrophies, amblyopia, congenital eye anomalies, and albinism. The common acquired causes include retinitis pigmentosa (RP), age-related macular degeneration (ARMD), diabetic retinopathy (DR), myopic degeneration, glaucoma, other forms of optic atrophy, corneal disease, endophthalmitis, acquired cataract, and higher visual pathway defects. Gao et al. in their study from China reported congenital cataract (21.61%), optic atrophy (14.20%), macular dystrophy (11.73%), nystagmus (9.88%), and congenital glaucoma (9.26%) as the leading causes of VI.
There are many techniques and devices which can aid those suffering from VI. The basic principles of various low vision techniques are (1) optical magnification to provide an enlarged image of an object, (2) improve lighting by using a gooseneck lamp aimed at the task or a high lumen pocket flashlight, (3) reduced glare, or (4) improved contrast. Low vision aids (LVAs) are assistive products that facilitate disability self-management and enable an independent and improved quality of life. Thus, these are now termed as low vision assistive products (LVAP) and include (1) optical LVAP which works on the principle of using magnifying lenses to make objects look larger and easier to see. These include magnifying spectacles, stand magnifiers, hand-held magnifiers, and telescopes, (2) nonoptical LVAP includes audible instruments, large print books, audible books, telephone, thermostat, telephones, thermostats, watches and remote controls with large-sized numbers and high-contrast colors, needle threaders, magnifying mirrors, and tactile (textured) labels, etc., (3) electronic devices in the form of video magnifiers, audio and electronic books, smartphone and tablets, and computers that can read aloud or magnify what is on the screen. Gao et al. reported that the most frequently prescribed low vision devices for distant and near vision were binocular telescopes (23.57%) and stand magnifiers (22.93%), respectively. Chotikavanich et al. found that glasses (49.5%) and hand-held magnifiers (40.4%) were most commonly prescribed LVAP among adults, and glasses (35.5%) and telescopes (26.9%) were the most frequently prescribed LVAP among the pediatric population.
The currently available service centers for patients with low vision can be broadly divided into low vision centers, online training groups, and support groups. Low vision services include vision rehabilitation, mobility training, sensory substitution, environment modification, and LVAP. The various management options help by efficient use of the residual vision, improve daily function, maximize patient's ability to perform daily activities, and thereby improve their QLI.,, De et al. in their analysis on the impact of low vision clinic and LVAPs on QLI found that the low vision clinic visit alone demonstrated a 4.55-point improvement in QLI (from 77.77 to 82.33 points, P = 0.001). Further, adjusting for age, gender, and education, the patients who received LVAPs experienced an even larger increase of 8.89 points. Gobeille et al., in their study, assessed utilization and abandonment of low vision devices in a mobile clinic versus standard outpatient department and found that digital magnification, optical magnifiers, and filters were most frequently recommended. At 3 months, 29% of participants abandoned at least one device.
As per a detailed literature review, majority of the previous studies on LVA have analyzed the common causes and treatment modalities for patients with low vision, but none of the studies have simultaneously analyzed the recommended and accepted LVA. This study aimed to analyze the age-wise distribution of patients presenting to low vision clinics, underlying causes of low vision, various LVAP recommended, and their acceptance rates in large sample sizes over 6 years. As per the literature search and to the best of our knowledge, this is the largest analysis of low vision and LVAP done in the South Indian population to date.
| Methods|| |
This was a long-term retrospective, observational analysis conducted at a tertiary eye care hospital over 6 years from January 2011 to December 2016. The study adhered to the tenets of the Declaration of Helsinki. The study was approved by the Ethics Committee of our hospital's Institutional Review Board. The case sheets of patients who presented to the low vision clinic of our hospital during the study period were retrospectively reviewed.
A total of 8309 patients were evaluated by trained optometrists and a medical officer in the low vision clinic. Characteristics such as age, presenting and best-corrected visual acuity (BCVA), causes of low vision, improvement with LVA, recommended and acceptance rates of LVAP were entered in an excel sheet.
The definitions of VI used for the estimates in this study followed the criteria of the International Classification of Diseases (ICD) update and revision 2010.
- Category 1 – Moderate VI: best-corrected distance visual acuity (DVA) in the better eye worse than 6/18 (0.50) and better than or equal to 6/60 (1.00)
- Category 2 – Severe VI: best-corrected DVA in the better eye worse than 6/60 (1.00) and better than or equal to 3/60 (1.30)
- Category 3 – blindness: best-corrected DVA in the better eye worse than 3/60 (1.30) and better than or equal to 1/60 (1.50) or the visual field of the better eye is <10° in a radius around central fixation
- Category 4 – blindness: best-corrected DVA in the better eye worse than 1/60 (1.50) and better than or equal to light perception
- Category 5 – blindness: No light perception.
As per the WHO-ICD 10 classification, categories three to five are categorized as blindness. Categories one to four were included in our study, based on the services available at our center. In cases where two or more disorders were responsible for the VI, the primary cause was recorded. The decision for the type of LVAP was based on each patient's visual status, visual needs, level of understanding of the patient, socioeconomic status, patient satisfaction, and ease of use assessed by the ophthalmologist, low-vision therapist, and after discussion with the patient and/or the guardian.
All participants were tested under binocular viewing conditions. Distance VA was measured using Lea's numbers log MAR chart at a 4-meter distance. The magnification required was calculated using the following formulae: Equivalent viewing power = BCVA × 100/working distance.
Low vision assistive products offered in our clinic include
- High power spectacles – Convex sphere, aspheric sphere, prismospheres, and clip-on loupes
- Magnifiers – Hand and stand magnifiers, pocket magnifiers, magnifiers with illumination
- Telescopes – Distance or near, handheld/spectacle mounted, monocular or binocular
- Electronic devices – Closed circuit television and video magnifiers
- Nonoptical devices – Signature card, notex, typoscopes, needle threader, and mobility cane.
Statistical data analysis
The results were analyzed using SPSS statistical package (version 10; SPSS Science, Chicago, IL, USA). Continuous variables were stated as median (minimum-maximum) values and categorical variables as number (n) and percentage (%). Chi-square and Fisher's exact test were used to identify differences in proportions. A P < 0.05 was considered statistically significant.
| Results|| |
A total of 8309 patients were examined in the low vision clinic at our hospital during the study period. Of these, 34% (2844) were <15 years of age, 30% (2452) were between 16 and 40 years, and 36% (3013) were more than 40 years of age. The year-wise age distribution of patients is described in [Table 1]. Based on the BCVA, 66.4% of patients with BCVA 6/18–3/60 were grouped under low vision while 33.6% patients with BCVA of 3/60-perception of light were grouped as blind. The year-wise distribution of patients in the two groups based on the BCVA range is listed in [Table 2]. Of the total patients screened, 3673 (44.2%) benefited from LVAPs. This was almost consistent over the 6 years duration of the study, except in 2013, where 57.1% of patients seemed to benefit from one or the other LVAPs. The detailed year-wise distribution of patients screened versus those who benefited is illustrated in [Table 3]. The most common underlying etiology for low vision in our study cohort was RP (18.6%), congenital nystagmus (17.8%), DR (15.7%), and ARMD (15.5%), respectively. A total of 63% of patients with congenital nystagmus and 56.2% with RP showed improvement with LVAPs. While patients with ARMD and congenital optic nerve pathologies, only 30.7% and 32.5%, respectively, showed improvement. [Table 4] highlights the number of patients referred to low vision clinic with different etiologies and the patients that were benefitted from LVAPs.
|Table 2: Correlation between number of patients seen year wise and best-corrected visual acuity range|
Click here to view
|Table 3: Correlation between number of patients screened year-wise versus improved with low vision aids|
Click here to view
|Table 4: Depicting varied etiologies of low vision, number of patients with percentage distribution along with the number of patients who improved with low vision assistive product|
Click here to view
In total, 2291 patients were prescribed different LVAPs, while others were deferred either because of lack of appreciation of improved vision by the patient/lack of interest/felt need. These patients were offered an option for the hospital-based trial before investing in any LVAPs. Out of 2291 patients prescribed LVAPs, 973 (42.5%) patients accepted these. Hand and stand magnifiers were most commonly recommended through all ages. Spectacle magnifiers were particularly found helpful in patients aged >40 years, and 55.6% of recommendations were in this group, and 65.9% of telescope recommendations were for children <15 years of age.
The maximum acceptance rate was seen with hand and stand magnifiers, which 512 (50.3%) patients accepted. This was followed by spectacle magnifiers accepted by 304 (45.2%) patients. On the other hand, the acceptance rates for telescopes and electronic devices were 18.95% and 8.8%, respectively. Overall, the acceptance rates for LVAPs increased with age. The proportion of patients who were recommended and who accepted the LVAPs in different age groups has been shown in [Table 5]a and [Table 5]b.
| Discussion|| |
VI is a global concern and likely to increase with an increased life span. Since there are limited trained personnel for LVAP dispensing in both developing and developed countries, increased demand would be an extra burden on the existing health-care resources. As a result, a large number of those who would benefit from the low vision services may be denied the opportunity because of the increasing demand for the provision of these services. Majority of the previous articles have highlighted only the causes of low vision and recommended options for LVAPs. Our study highlights the age-wise distribution of patients presenting to the low vision clinic, underlying pathology causing low vision, the proportion of patients with different etiologies, which improved versus acceptance rate in different age groups.
We found that 36.3% of the patients presenting to the low vision clinic were >40 years of age, and 34.2% were lesser than 15 years of age. This shows the wide distribution of pathologies, leading to low vision. Chotikavanich et al., in their study, found that of the total sample, 23.4% were <15 years of age, while 76.6% were >15 years of age. A significant proportion of the younger population presenting to these clinics highlights the importance and need for improving the availability of low vision services, as simple interventions like providing appropriate LVAPs/lifestyle modifications can significantly impact the quality-adjusted life year for these patients. Based on the BCVA, 66.4% of patients were categorized into the low vision group, while 33.6% fell into the blind group. In another study from India, similar results were obtained where they had 72.7% adult patients in the low vision group and 27.4% in the blindness group. Throughout the study duration, we had a continuous inflow of a similar number of patients, highlighting the need for low vision clinics in society. In 2013, we examined fewer patients because of a natural calamity (Tsunami floods) in the geographical area where the study was conducted. This further highlights the impact of natural disasters and difficulty accessing health care by patients with low vision.
The common underlying etiologies were found to be RP among 18.6% of patients, followed closely by congenital nystagmus (17.8%), DR (15.7%), and ARMD (15.5%). Mohidin and Yusoff also found RP as the most common cause of low vision in their study. Similar results were also seen in a study from Thailand, where RP was found as the most common cause of VI (28.3%), followed by AMD, glaucoma, and DR for 10.3%, 10.0%, and 9.6% of patients, respectively. These results align with our study. In contrast, studies from the United Kingdom, United States of America, and Germany had reported ARMD as the most common cause of low vision. The differences might be related to a higher prevalence of ARMD among Europeans than Asians and a higher prevalence of RP in the Asian than the Caucasian population., The enormous burden of pathologies like ARMD and DR can be attributed to longer life expectancy, adoption of western lifestyle, lack of education leading to ignorance for regular health-care visits, and limited access to health care. The greater RP prevalence can also be attributed to consanguineous marriages in the study population. This shows that genetic counseling educational programs related to lifestyle modifications are required to reduce the unwanted propagation of ocular diseases.
Out of 8309 patients examined at our low vision clinic, we found that 3673 (44.2%) patients were benefitted from LVAP. Maximum benefit was seen in patients with congenital nystagmus (63%) and RP (56.2%). Patients with congenital optic nerve pathologies were additionally prescribed visual and developmental stimulation exercises at home. Adult patients were also given mobility household work training and guided for appropriate job opportunities based on their residual vision. Our LVAP recommendation rates were similar to Bakkar et al., who gave LVAP to 43.7% of the total patients seen in low vision clinics. Chotikavanich et al. in their study recommended LVAPs to 51.8% of adults and 25.1% of the pediatric group.
The most common causes of VI in our study population as a whole differed markedly from the studies in Korea by Ji et al. (Optic Atrophy) and in Northern Jordan by Bakkar et al. (Albinotic patients). In the study conducted by Gao et al., the leading causes of VI were congenital and hereditary optic nerve pathologies in the Chinese population. Uncorrected refractive errors leading to amblyopia were the leading cause of low vision in a study by Wong et al. in East Asia. We believe that, due to extensive school screening camps and outreach activities, uncorrected refractive errors are relatively less common in South India.
In regards to management, hand and stand magnifiers were most commonly recommended, followed by spectacle magnifiers. Telescopes were the most frequently prescribed LVAP in the 0–15 years group. Whereas in >40 years age group, spectacle magnifiers were most commonly used. This can be justified with the need for distance vision in schools in younger children. Older patients tend to restrict their activities to near work and engage in more reading work, making magnifiers a better and more convenient option.
We found that hand and stand magnifiers (50.34%) were the most accepted LVAP, followed by spectacle magnifiers (45.24%). This was followed by walking sticks (nonoptical) and telescopes (optical). Electronic devices were the least accepted LVAP, probably because of the cost factor involved in buying and maintaining these. Electronic LVAP, like video magnifiers, require electricity input or rechargeable batteries, and wireless input adds to the cost of these devices. Audio and electronic books also require electrical input. Smartphone LVAP is costly compared to other optical and nonoptical LVAP, and computer magnifiers and closed-circuit televisions are expensive, bulky, and immovable, making these LVAPs less preferable than other available options. [Table 6] highlights the different types of recommended versus accepted LVAPs. Patients seeking low vision rehabilitation services at our center are predominantly the young population (<40 years) but had the least acceptance rate for LVAPs. The elder group (>40 years) had the maximum acceptance rates in our study. There is a need to educate, create awareness and at the same time make low vision services affordable and easily accessible to the masses in need. We found RP, congenital nystagmus, and DR as the leading underlying causes of VI. Unlike in other countries, extensive periodic school screening in this part of the country has significantly reduced the prevalence of uncorrected refractive errors and amblyopia as causes of VI.
|Table 6: Correlation between low vision aid type with number of recommended and accepted patients|
Click here to view
LVAP potentially restores independence and enhances the quality of life in visually impaired patients. However, the low vision services are in short supply and very expensive, making it unaffordable for patients to buy them from a limited number of optical centers offering such aids in the country. Thus, there is a need to train and educate and at the same time create awareness about low vision clinics and increase availability at an affordable cost. We hope that our paper will enhance understanding needs, varied pathologies, and acceptance of LVAPs in these patients. The strengths of our study include a large sample size of 8309 patients, long-term analysis of 6 years which is probably one of the largest analyses from South India and a fairly decent acceptance rate of LVAP in various age groups. The study being retrospective in nature has its inherent limitations. The intricate details of the level of visual improvement either on the Snellen chart or in terms of the visual field were not studied, and the cause of nonacceptance of the LVAPs was also not documented for all the patients.
| Conclusion|| |
With the ever-growing burden of VI, health-care providers need to lead by educating for lifestyle modifications and help in rehabilitation. Counseling is a critical component of low vision clinics which will help the patients to understand their disease process and have tangible expectations. Psychological factors predicting the use of LVAPs need to be duly considered while choosing therapy/LVAP. This will also improve the acceptance and directly enhance patients' quality of life. Follow-up strategies should focus on multidisciplinary approaches, dealing with patients' cognitive psychology, adaptation to LVAP, and motivating them for an independent life. In addition, great emphasis should be focused on training, education, guidance, and support for setting up more low vision rehabilitation centers from the policymakers.
VI should be treated as an issue of major public concern. Efforts should be made to improve quality of life assessment and raise the living standard in the care protocol for the visually impaired. This will help in effective rehabilitation of the visually impaired, maximize residual vision use, and enhance their ability to perform everyday tasks and lead an independent life. We further recommend future large-scale studies to assess the amount of vision improvement with LVAP and its effect on QLI using LVAP in day-to-day activities. This will provide an insight into the need for further government regulations and promote training and support in upbringing new low vision centers.
We would like to thank Mr. Shanmugam for helping with the statistical analysis.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
World Health Organization. International Statistical Classification of Diseases and Related Health Problems 10th Revision Current Version for 2003 Chapter VII H54 Blindness and Low Vision. Available from: http://www.who.int/classifications/icd/en/
. [Last accessed on 2021 Nov 19].
Bourne RR, Flaxman SR, Braithwaite T, Cicinelli MV, Das A, Jonas JB, et al.
Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: A systematic review and meta-analysis. Lancet Glob Health 2017;5:e888-97.
Burton MJ, Ramke J, Marques AP, Bourne RR, Congdon N, Jones I, et al.
The lancet global health commission on global eye health: Vision beyond 2020. Lancet Glob Health 2021;9:e489-551.
Fricke TR, Tahhan N, Resnikoff S, Papas E, Burnett A, Ho SM, et al.
Global prevalence of presbyopia and vision impairment from uncorrected presbyopia: Systematic review, meta-analysis, and modelling. Ophthalmology 2018;125:1492-9.
Gao G, Yu M, Dai J, Xue F, Wang X, Zou L, et al.
Demographic and clinical characteristics of a paediatric low vision population in a low vision clinic in China. Clin Exp Optom 2016;99:274-9.
Chotikavanich S, Chanvarapha N, Loket S, Yingyong R, Dongngam S, Nujoi W, et al.
A 5-year retrospective record review of hospital-based low-vision rehabilitation in Thailand. Clin Optom (Auckl) 2018;10:41-50.
Renieri G, Pitz S, Pfeiffer N, Beutel ME, Zwerenz R. Changes in quality of life in visually impaired patients after low-vision rehabilitation. Int J Rehabil Res 2013;36:48-55.
Stelmack J. Quality of life of low-vision patients and outcomes of low-vision rehabilitation. Optom Vis Sci 2001;78:335-42.
Wang BZ, Pesudovs K, Keane MC, Daly A, Chen CS. Evaluating the effectiveness of multidisciplinary low-vision rehabilitation. Optom Vis Sci 2012;89:1399-408.
Do AT, Ilango K, Ramasamy D, Kalidasan S, Balakrishnan V, Chang RT. Effectiveness of low vision services in improving patient quality of life at Aravind Eye Hospital. Indian J Ophthalmol 2014;62:1125-31.
] [Full text]
Gobeille MR, Malkin AG, Jamara R, Ross NC. Utilization and abandonment of low vision devices prescribed on a mobile clinic. Optom Vis Sci 2018;95:859-64.
Dougherty BE, Kehler KB, Jamara R, Patterson N, Valenti D, Vera-Diaz FA. Abandonment of low-vision devices in an outpatient population. Optom Vis Sci 2011;88:1283-7.
Culham LE, Ryan B, Jackson AJ, Hill AR, Jones B, Miles C, et al.
Low vision services for vision rehabilitation in the United Kingdom. Br J Ophthalmol 2002;86:743-7.
Silver JH. Low vision aids in the management of visual handicap. Br J Physiol Opt 1976;31:47-87.
Mohidin N, Yusoff S. Profile of a low vision clinic population. Clin Exp Optom 1998;81:198-202.
Pardhan S, Mahomed I. The clinical characteristics of Asian and Caucasian patients on Bradford's Low Vision Register. Eye (Lond) 2002;16:572-6.
Owsley C, McGwin G Jr., Lee PP, Wasserman N, Searcey K. Characteristics of low-vision rehabilitation services in the United States. Arch Ophthalmol 2009;127:681-9.
Nguyen NX, Weismann M, Trauzettel-Klosinski S. Spectrum of ophthalmologic and social rehabilitation at the Tübinger Low-Vision Clinic: A retrospective analysis for 1999-2005. Ophthalmologe 2008;105:563-9.
Wong WL, Su X, Li X, Cheung CM, Klein R, Cheng CY, et al.
Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: A systematic review and meta-analysis. Lancet Glob Health 2014;2:e106-16.
Zhang Q. Retinitis pigmentosa: Progress and perspective. Asia Pac J Ophthalmol (Phila) 2016;5:265-71.
Chizzolini M, Galan A, Milan E, Sebastiani A, Costagliola C, Parmeggiani F. Good epidemiologic practice in retinitis pigmentosa: From phenotyping to biobanking. Curr Genomics 2011;12:260-6.
Ji YH, Park HJ, Oh SY. Clinical effect of low vision aids. Korean J Ophthalmol 1999;13:52-6.
Bakkar MM, Alzghoul EA, Haddad MF. Clinical characteristics and causes of visual impairment in a low vision clinic in northern Jordan. Clin Ophthalmol 2018;12:631-7.
Gao G, Ouyang C, Dai J, Xue F, Wang X, Zou L, et al
. Baseline traits of patients presenting at a low vision clinic in Shanghai, China. BMC Ophthalmol 2015;15:16.
Wong TY, Zheng Y, Jonas JB, Flaxman SR, Keeffe J, Leasher J, et al
. Prevalence and causes of vision loss in East Asia: 1990-2010. Br J Ophthalmol 2014;98:599-604.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]