|Year : 2012 | Volume
| Issue : 2 | Page : 101-104
Comparison of visual acuity estimates using three different letter charts under two ambient room illuminations
Ai-Hong Chen, Fatin Nur Najwa Norazman, Noor Halilah Buari
Department of Optometry, Universiti Teknologi MARA, Puncak Alam Campus, 42300, Selangor, Malaysia
|Date of Submission||07-Sep-2009|
|Date of Acceptance||04-Jun-2011|
|Date of Web Publication||20-Mar-2012|
Department of Optometry, Faculty of Health Sciences, Universiti Teknologi MARA, Puncak Alam Campus, 42300 Bandar Puncak Alam, Selangor
Source of Support: UiTM Excellent Fund, Conflict of Interest: None
Background: Visual acuity is an essential estimate to assess ability of the visual system and is used as an indicator of ocular health status. Aim: The aim of this study is to investigate the consistency of acuity estimates from three different clinical visual acuity charts under two levels of ambient room illumination. Materials and Methods: This study involved thirty Malay university students aged between 19 and 23 years old (7 males, 23 females), with their spherical refractive error ranging between plano and −7.75D, astigmatism ranging from plano to −1.75D, anisometropia less than 1.00D and with no history of ocular injury or pathology. Right eye visual acuity (recorded in logMAR unit) was measured with Snellen letter chart (Snellen), wall mounted letter chart (WM) and projected letter chart (PC) under two ambient room illuminations, room light on and room light off. Results: Visual acuity estimates showed no statistically significant difference when measured with the room light on and with the room light off (F1,372 = 0.26, P = 0.61). Post-hoc analysis with Tukey showed that visual acuity estimates were significantly different between the Snellen and PC (P = 0.009) and between Snellen and WM (P = 0.002). Conclusions: Different levels of ambient room illumination had no significant effect on visual acuity estimates. However, the discrepancies in estimates of visual acuity noted in this study were purely due to the type of letter chart used.
Keywords: Illumination, letter charts, visual acuity
|How to cite this article:|
Chen AH, Norazman FN, Buari NH. Comparison of visual acuity estimates using three different letter charts under two ambient room illuminations. Indian J Ophthalmol 2012;60:101-4
|How to cite this URL:|
Chen AH, Norazman FN, Buari NH. Comparison of visual acuity estimates using three different letter charts under two ambient room illuminations. Indian J Ophthalmol [serial online] 2012 [cited 2020 May 29];60:101-4. Available from: http://www.ijo.in/text.asp?2012/60/2/101/90489
Visual acuity is an estimate of the visual system's ability to resolve fine details. Visual acuity estimates are taken by assessing and quantifying the eye's ability to resolve and recognize letters of varying sizes and shapes. This estimate is an essential component in the battery of tests conducted to determine the patient's refractive status, the adequacy of optical correction and as a key indicator of ocular health status. ,
With improvement in optical knowledge and visual perception studies, various forms of visual acuity test charts have been introduced and marketed, such as the conventional rotatable Snellen letter chart (Snellen), wall-mounted letter chart (WM) and projected letter chart (PC). Each one of these charts was constructed based on different pattern designs, interaction, contrast, visual cues, perception and surrounding lighting. ,,, The visual acuity chart measures the minimum angle of resolution of the visual system;  that is the smallest target estimate in angular subtense that an individual is capable of resolving. However, different letters of the same subtense (angular size) have different legibility. Letters with similar appearance, such as, "O" and "C", "P" and "F", "O" and "Q" are more difficult to distinguish than those that vary greatly in their appearance, such as, "T" and "S", "U" and "N" and "A" and "H". This is due to the different luminance profile of each letter. 
Factors that interfere with the luminance profile of the letters produce lower resolution limit, thus, a reduced estimate of visual acuity. ,,, Those factors that had been intensively investigated in visual acuity estimates included the luminance of the visual acuity chart and the ambient illumination of the examination room. Studies showed that when there was a reduction in chart luminance, the width of the point spread function of the letter's image increased on the retina, as a result of the rapid fall off of the letter contrast,  which in turn resulted in decreased visual resolution. 
It is necessary to consider not only the luminance of the visual acuity chart, but also the overall room illumination. Increases in ambient room illumination reduce the pupil size. A smaller pupil increases the depth of focus and also reduces the peripheral light rays that enter the eye. Therefore, reduced pupil diameter produces less spherical aberration and other higher order aberration, such as coma aberration. A clinician may obtain an artificially better estimate of visual acuity in a brightly-lit examination room. Considering the potential disadvantages of measuring visual acuity under bright room illumination with conventional Snellen chart, later visual acuity charts have been designed to be used under dim room illumination with a properly calibrated chart luminance.  This is also essential especially in visual acuity measurement and prescription modification for night myopic patient. 
The purpose of the study is to investigate the consistency of visual acuity estimates with three different letter charts under two ambient room illuminations in an optometric clinic. This is purely a clinical experiment which aims to investigate the reliability of visual acuity estimates taken with commonly used letter charts in optometric practice.
| Materials and Methods|| |
Subjects were recruited from the first and second year Optometry students. Thirty Malay students ages ranging between 19 and 23 years (mean: 20.71±0.69) (7 males and 23 females) participated in this study (based on power analysis calculation). Subjects' refractive error ranged from plano to −7.75D, astigmatism ranged from plano to −1.75D, anisometropia equal to 1D or less. Subjects reported no history of ocular injury or pathology. All subjects had monocular visual acuity of 20/30 or better with their habitual visual correction. The study adhered to the ethical standards with the Helsinki Declaration of 1975, as revised in 2000. This project was approved by the Ethics Committee in UiTM.
This study was carried out in three different visual examination rooms in Optometry Clinic of the Faculty of Health Sciences, Universiti Teknologi MARA. Each one of these rooms was equipped with one of the three letter charts: 1) Snellen letter chart (Snellen) [Clement Clarke, 4101510/4101590, Haag-Streit UK Ltd., Edinburgh Way, Harlow, Essex, CM20 2TT, UK], 2) wall-mounted letter chart (WM) [Tian Le, SR5,Ningbo Ming Sing Optical RandD Co Ltd., No702, Yinzhou District, China] and 3) projected letter chart (PC) [Topcon, ACP8EM,Topcon Corporation, 75-1, Hasunuma-cho, Itabashi-ku, Tokyo, 174-8580, Japan]. The angle subtended by 1.0 Snellen (0.00LogMAR) for Snellen, WM, and PC was 0.96, 1.02 and 0.90 respectively. In each examination room, the level of illumination was calibrated according to the manual. The chart illumination measurement, taken at 20 inches from the center of the chart with Luminometer LS100 Konica Minolta, was: for the Snellen 847.90 lux, WM 444.73 lux and PC 136.50 lux with the room light on and 715.27 lux, 375.97 lux and 123.27 lux respectively with the room light off. Room light off means no light beside the chart light. Windows and doors in the examination rooms were covered with black curtains to prevent variation in room illumination. Monocular visual acuity from the right eye was tested at 20 feet. All thirty subjects performed visual acuity estimation with Snellen, PC and WM under both illuminations.
Monocular visual acuity is recorded in Snellen notation and then converted into logMAR as the smallest line in the letter chart that the subject could read. Subjects were encouraged to read to the smallest line. Estimates of visual acuity were stopped at the line when subjects made at least two mistakes. Each subject read the same chart twice under two different illuminations. Order of testing conditions was randomized using a random number table. To avoid bias in chart memory and the effect of guessing in measurements using the same chart under two different ambient illumination levels, letters in each line were pointed to for identification by the examiner in a random sequence. An average of half an hour was required to complete all testing. Four optometrists performed the measurements. Inter-examiner variation in estimates of visual acuity was evaluated prior to commencement of the study. All examiners tested 30 subjects who were not involved in the current study using each of the three charts (Snellen, WM and PC). Inter-examiner variation in this pilot study was not significant (F = 0.26, P = 0.08).
The estimates of visual acuity were obtained with and without the room light using three different letter charts available in UiTM optometry clinic. In this analysis, the independent variables were the types of letter chart and two levels of ambient room illumination. The dependent variable was the visual acuity estimates.
| Results|| |
The data were determined to be normally distributed using the Kolmogrov-Smirnov test; therefore, data analysis was carried out using the two-way ANOVA with post-hoc comparisons. The level of significance was set to 0.01.
Visual acuity estimates obtained by different examiners and in different examination rooms were analyzed with one-way ANOVA to evaluate repeatability of the measures. This analysis showed no significant differences; therefore, visual acuity estimates were not affected by inter-examiner variability. Visual acuity estimates were not statistically different under the two room illumination levels (F1,372 = 0.26, P = 0.60). This was true for all three letter charts [Table 1]. High intraclass correlation (0.89, P < 0.01) was found in the study.
|Table 1: Comparison of visual acuity estimates with two-way ANOVA to investigate the effect of room illumination on three different letter charts|
Click here to view
Despite the fact that visual acuity estimates were not significantly different when the room light was turned on compared with the room light turned off, visual acuity estimates were statistically different across the type of letter chart (F2, 372 = 6.80, P = 0.001) [Table 1]. The interaction between the type of letter chart and levels of ambient illumination did not reach significance in this study (F2,372 = 0.78, P = 0.46). Post-hoc analysis with Tukey showed visual acuity estimates were significantly different between the Snellen and PC (P = 0.009) and between Snellen and WM (P = 0.002) [Table 2]. Differences of two letters were observed in both comparisons.
|Table 2: Comparison of visual acuity estimates with post-hoc test in two-way ANOVA|
Click here to view
| Discussion|| |
The effects of both illumination (with the room light vs without the room light) and chart types (Snellen, PC and WM) on visual acuity estimates were explored in this study. Different levels of ambient room illumination had no significant effect on the estimates of visual acuity. The discrepancy in visual acuity estimates noted in this study was purely due to the type of letter chart. Visual acuity estimates were significantly lower with both PC and WM charts than those estimates obtained using the Snellen chart.
Inconsistency in visual acuity estimates obtained in our study might be attributed to the distinct differences in chart design. , The Snellen chart uses eight optotypes without weighing the difference of relative legibility or readability level of optotypes. Some of the optotypes such as "A" and "H" are easily distinguishable even with some degree of defocus under bright illumination. Optotypes with similar relative legibility such as "H" with "N" and "O" with "D" reduce the possibility of guessing a correct answer, thus reducing estimates of visual acuity.  Furthermore, the construction of the Snellen chart consists of different numbers of optotypes from line 20/200 to line 20/13. There is only one optotype at the 20/200 line, while there are eight optotypes at the 20/13 line. A greater number of optotypes in each line has been reported to contribute to a more precise estimate of visual acuity with better reliability and repeatability. The greater number of optotypes at the lines approaching a subject's threshold (for line 20/20 and 20/13) means that the size progression is less than 0.02 log unit per optotype, which yields less variability in visual acuity measurement as compared to projected letter chart which always contains four optotypes at every level of acuity. , Hence, the visual acuity measurement recorded using the Snellen letter chart showed more consistencies and less variance compared to that recorded using the WM chart. Both charts show the same number of optotypes at line 20/20 and 20/13 (eight optotypes at each line). The major difference is the spacing of the optotypes. On the wall mounted letter chart, the spacing between optotypes at each line is equal to the width of the optotype; while on the Snellen letter chart, the spacing between optotypes is equal to 2.5 times the width of an optotype at line 20/20 and line 20/13. A two fold change in spacing had been reported to alter the visual acuity estimates by 0.03 log unit with British letters and 0.04 log unit with Sloan letters.  Visual acuity estimates would be higher when the spacing between adjacent optotypes was wider as a result of reduced contour interaction. Additionally, eye movement control and fixation tremor might contribute to the reduction of visual acuity estimates when the letters were closely spaced and might be greater when the threshold print size was smaller. Therefore, it was reasonable to expect higher visual acuity estimates with the WM chart compared to the Snellen chart.
The PC chart also showed a less consistent visual acuity estimate compared to the Snellen chart. The contrast level of the optotypes presented in the PC chart was very much influenced by the room lighting. Presence of glare source and focus of the projection instrument reduced the contrast of the optotypes. Visual acuity estimates the resolution ability of human eyes at 100% contrast. However, reduced contrast level of the optotypes as a result of the focus of the projector chart potentially reduced contrast sensitivity at higher spatial frequency.
Time factor was the major limitation in this study. The same subject must accomplish the visual acuity measurement in two different illumination levels and with all tested charts. The study was carried out before students had their lecture to avoid after effect of near work which might interfere with the test result. Recruiting the subjects during their free slot was difficult with their packed lecture schedule. Thus, we had decided to perform the visual acuity measure in one session. Another limitation of the study was that the subjects were optometry students and they might be well versed with these charts. Additionally, some of these subjects might have even memorized the chart inadvertently due to repeated use of them earlier.
Visual acuity estimates with Snellen chart showed greater consistency in UiTM optometry clinic. Visual acuity estimates with Snellen chart was unlikely to be affected by room illumination levels. Use of the PC chart and WM chart to produce more consistent visual acuity estimates in optometric clinical procedures required careful calibration of the instruments and room arrangement to avoid over- or under-estimation of visual acuity. These observed differences suggest that any estimate of visual acuity should be recorded with notation of the conditions of the measurements, chart-type and room illumination. This information will be critical for the patient's long-term follow-up care and changes in refractive error.
| Acknowledgement|| |
Special thanks to Mr Azmir Ahmad and Mdm Wan Elhami Wan Omar for helping in data collection.
| References|| |
Davidson DW. Visual acuity. In Eskridge JB, Amos JF, Bartlett JD, editors. Clinical procedures in optometry. Philadelphia: J.B. Lippincott Company; 1991. p. 17-29.
Bennett AG, Rabbetts RB. Visual acuity. In: Bennett AG, Rabbetts RB, editors. Clinical visual optics.
London: Butterworths; 1984. p. 21-57.
McMonnies CW. Analysis of errors acuity measurements. Clin Exp Optom 1996;79:144-51.
Grimm W, Rassow B, Wesemann W, Saur K, Hilz, R. Correlation of optotypes with the Landolt Ring: A fresh look at the comparability of optotypes. Optom Vis Sci 1994;71:6-13.
Raasch TW, Bailey IL, Bullimore MA. Repeatability of visual acuity measurement. Optom Vis Sci 1998;75:342-8.
Bailey IL. Visual Acuity. In: Benjamin WJ, editor. Borish's clinical refraction.
Philadelphia: W.B. Saunders Company; 1998. p. 179-202.
Strang NC, Atchison DA, Woods RL. Effects of defocus and pupil size on human contrast sensitivity. Ophthal Physiol Opt 1999;19:415-26.
Bennett AG, Rabbetts RB. Visual acuity. In: Bennett AG, Rabbetts RB, editors. Clinical visual optics.
London: Butterworths; 1984. p. 21-57.
Schwartz SH. Spatial vision. In: Schwartz SH, editor. Visual Perception. New York: Appleton and Lange; 1999a. p. 175-204.
Grosvenor T. The preliminary examination. In: Grosvenor T, editor. Primary care optometry. Boston: Butterworth-Heinemann; 1996. p. 309-40.
Grosvenor T. The preliminary examination. In: Grosvenor T, editor. Primary care optometry. Anomalies of refraction and binocular vision. 3 rd
ed. Boston: Butterworth-Heinemann; 1996. p. 155-87.
[Table 1], [Table 2]