A hospital-based cross sectional study was carried out among 59 subjects referred to the Low Vision Clinic (LVC) from the Retina Clinic at B.P. Koirala Lions Centre for Ophthalmic Studies (BPKLCOS) in the period January to June 2010. Subjects having diabetic retinopathy and visual acuity equal to or less than 6/18 were enrolled in the study. Subjects having any other associated ocular pathology which was not attributable to diabetes, with currently uncontrolled blood sugar, or being treated with pan-retinal photocoagulation for less than 6 weeks, were excluded from the study.

Informed consent was obtained from all subjects and their attendants after explaining to them the purpose of the study. The subjects were requested to bring to the clinic their present optical correction, and the most recent blood sugar report to ensure control and stability of their blood sugar level. Only 38 subjects (64%) fitted the inclusion criteria to be assessed for the low vision service.

Clinical examination included the following procedures:

• (a)

  Presenting, and best-corrected distance visual acuities, were assessed using a Log MAR chart that was designed to be used at 10feet under normal room illumination. It was an illiterate E chart having a consistent number of five letters in each row. There was a geometric progression of 0.1log units in each line. If the subject was unable to read the top line at 10feet, the chart was brought nearer (8feet, 6feet, 4feet, 2feet or 1foot) until they were able to identify those letters.

• (b)

  Objective and subjective refraction was undertaken to achieve the best distance acuity. To determine the refractive error the power was estimated by the bracketing technique in which plus and minus lenses of equal powers was compared. To determine the interval of lenses to be compared, a concept of just noticeable difference (JND) was adopted. To calculate JND, the denominator of the acuity at 10feet was divided by 100.

• (c)

  Near vision was assessed with a continuous reading chart while wearing best refractive correction, and near addition in presbyopic subjects. The meter system was adapted to record near visual acuity. Accurate measurement of reading distance from the spectacle plane was also recorded.

• (d)

  Contrast sensitivity was assessed monocularly as well as binocularly using the Pelli–Robson chart at one meter distance, with best distance refractive correction and near addition of +0.75D in presbyopic subjects.

• (e)

  Color vision was assessed monocularly using the Farnsworth dichotomous D-15 test with best refractive correction, and near addition if necessary.

• (f)

  Central visual field was tested monocularly at 33cm distance, with an Amsler grid chart with best refractive correction, and near addition in presbyopic subjects. With the Amsler chart, a white grid on a black background was used. The grid represents a 10cm×10cm square containing 400 single squares equally distanced. If any scotoma, blurred lines, patchy defects, or metamorphopsia were detected the test was considered positive.

• (g)

  Peripheral visual field was assessed monocularly with the Bernell handheld disc perimeter under normal room illumination. The Bernell disc perimeter comes with a 9in. long wand and 1–2–3–4mm targets. The radius of the perimetric arc was 13in., and the arc was 2in. high. A white test object of 3mm size was used to test the field at a distance of 1/3m. This instrument has been found to have high reliability and repeatability when compared to the Humphrey Field Analyzer II (Syah, 2007, as cited in Haliza, 2010).18

To evaluate the visual disabilities of each subject in daily life, a set of 34 questions in 7 sections was used (Appendix 1). The reproducibility and validity of the data obtained by the questionnaire had been established by an investigation in glaucoma and retinitis pigmentosa patients.19,20 The seven categories included questions about the legibility of letters, the legibility of sentences, walking, going out, having a meal, choosing clothing/dressing, etc. Multiple questions on visual disabilities were present in each category.

The questionnaire was translated to Nepali and back to English to check the consistency in meaning. Few modifications were done to suit the environment and culture of our study population, e.g. “Can you read a fare table for trains and subways?” was removed as it would not be applicable in Nepal. “When you write sentences in vertical lines, does it lean in either direction?” was replaced with “When you write sentences between two horizontal lines, does it tilt in direction?” “Do you feel some difficulty in using chopsticks?” was replaced with “Do you feel some difficulty in using a knife, fork, or spoon?”

Each patient was also asked to write about visual disability in daily life, with respect to factors that were not covered in the questionnaire. Difficulty in taking insulin and doing a blood sugar evaluation were reported. They were all included in the questionnaire.

Each question had three types of response choices, which were scored as: greatly disabled (2 points), slightly disabled (1 point), and not disabled (0 point). The visual disability score was calculated as the mean score within each category and the sum was calculated for each subject.

All data were evaluated using the statistical package for social sciences (SPSS 17.0.). Presenting, and best corrected visual acuity, contrast sensitivity, peripheral visual field, and refractive error differences between the better and worse eyes, were analysed using parametric paired sample t-test. Color vision and central visual field tests were analysed using Wilcoxon sign rank test. Multiple regression analysis was also used to assess the correlation between each dependent variable (legibility of letters, legibility of sentences, walking, going out, having a meal, dressing, etc.), and independent variables (visual functions – visual acuity, color vision, contrast sensitivity, central and peripheral visual field). Confidence interval was considered at 95% level. p-Value was considered significant for less than 0.05.

The main characteristics of the sample in the study are given in Table 1. Most of subjects (42.1%) who were in the age group of 60–70 years attended to the low vision clinic. Male/female ratio was 2.4:1. Severe non-proliferative diabetic retinopathy was present in 38.2% of subjects. Fifty percent of subjects had diabetes for 10–20 years.

Results of visual function analysis are given in Table 2. Except color vision (p=0.5), all the components of visual functions were significantly different between the better eye (BE) and worse eye (WE). Color vision could not be assessed in 7.9% subjects due to poor visual acuity. Color vision was found to be normal in 13.2% of the BE group and 23.7% of the WE group. CVD did not show any specific line of defect in 26.3% in the BE, and 21% in the WE.

Fifteen subjects (39.5%) presented with spectacles during the assessment. One person had one aphakic eye (+9.00) and the other eye +2.00 hyperopia. Eight subjects (21%) were bilaterally pseudophakic and 24 subjects (63.1%) were monocular pseudophakic. Central vision defects comprised of metamorphopsia and patchy central scotoma. Peripheral visual field could not be assessed in 7.9% subjects who could only see hand movements.

Results of the assessments of visual disabilities are presented in Fig. 1. The most common disabilities reported were legibility of sentences (1.4±0.4), and legibility of letters (1.2±0.3). The least common visual disability reported was clothing/dressing (0.7±0.3). Reading letters in a dictionary (1.6), and reading/writing sentences (1.6) on “legibility of sentences”, were the most disabling conditions for the subjects.

Figure 1.

Visual disability index with standard deviation.

(0.07MB).

Table 3 contains the standard regression coefficients and adjusted R2 value of multiple regression analysis. Visual disability index (VDI) for “Legibility of letters” and “Legibility of sentences” was correlated with near visual acuity and peripheral visual field. Visual disability index for “Walking” was correlated with contrast sensitivity (−0.57, p=0.01). Visual disability index for “Going out” was correlated with contrast sensitivity (−0.56, p≤0.05) and peripheral visual field (−0.31, p≤0.05). Visual disability index for “having meals and dressing” was not significantly correlated with visual functions. The sum of the visual disability score was correlated with near visual acuity (0.33, p≤0.05) and contrast sensitivity (−0.38, p≤0.05).