Visual acuity (VA) reflects the sharpness of our vision based on a measure of the angular size of the smallest detail that is resolvable by the observer.1 It is the most frequently tested visual function in evaluating the severity of ophthalmic disorders and the effectiveness of therapeutic interventions.2–5 In clinical practice, VA is measured with different acuity charts and design principles in different countries.6, 7 The Early Treatment of Diabetic Retinopathy Study (ETDRS) chart and Standard for Logarithmic Visual Acuity (SLVA) chart are the two most representative acuity charts in the world.
The ETDRS chart was designed by Dr. Bailey in the National Vision Research Institute of Australia in 1982 for the study of early diabetic retinal lesions.8 There are 14 lines with five letters per line in the ETDRS chart. The five Sloan letters (C, D, H, K, N, O, R, S, V, and Z) on each line are of the same size and approximately equal legibility.9 The spacing between letters is the width of a letter, and the spacing between lines is the height of the letters in the line below. The letters vary in size from − 0.3 to 1.0 logMAR with a 0.1 logMAR stepsize. The chart is designed to be used at a 4 m viewing distance, but VA as low as 1.6 logMAR could be measured with reduced test distances. Observers are usually required to read the chart letter by letter from the top line (the largest size) until he or she reaches a line where a minimum of 3 letters cannot be read correctly.10 Under the letter-by-letter scoring rule, visual acuity is scored by how many letters could be correctly identified by the observer. For an observer who correctly identified a total of n letters, the VA score is determined by the following formula: 1.1 − 0.02×n logMAR.11 The ETDRS chart is considered the gold standard in research, education, and clinics around the world.12, 13
The SLVA chart was designed by Prof Tian-rong Miao of Wenzhou Medical University in 1959.14 The SLVA chart also contains 14 lines, with each line containing Tumbling E letters in 4 orientations (upward, downward, leftward and rightward). This is useful for observers who cannot read the Latin alphabet.15 The number of Tumbling E letters is not identical across lines, starting from 2 on the top line and increasing to 8 on the bottom line. The spacing between lines is 24 mm. Letter size decreases with a 0.1 logMAR stepsize from the top to the bottom. Observers are required to identify the orientation of the Tumbling E letters, starting from the top line (1.0 logMAR) until a line in which less than 50% of the letters are correctly identified. The score corresponding to the line above it is recorded as the measured VA. The SLVA chart uses the ‘5-point recording method’ as the scoring rule. At a 5 m viewing distance, the VA scores range from 4.0 to 5.3, corresponding to 1.0 to -0.3 logMAR. According to the national compulsory standard issued by the Ministry of Health of China (GB 11533 − 2011), SLVA is the only legally approved acuity chart in clinical practice in China. This means that about 1 in 5 people around the world is tested with the SLVA chart.
These two most widely-used acuity charts have facilitated diagnosis and treatment of eye diseases. However, it has been unclear whether the scores from one chart could be generalized to the other, and vice versa. Actually, many details of the charts, including room illumination,16 optotypes,17 number of optotypes displayed on each line,18 etc, could affect the measured VA. In addition, observers’ VA scores are also determined by the different scoring and termination rules, which could also influence VA scores. Shah et al.19 compared the effect of scoring and termination rules (Letter-by-Letter observation with whole-chart stops by misidentifying over 5, 4, 3, 2, 1 letters) on VA scores in ETDRS testing. They found that the measured score (in logMAR) was higher if the test was terminated with one letter than with five or more letters. Mimouni et al.20 performed computer simulations to compare the effects of different scoring and termination rules on VA assessment. They found that the test-retest variability of 100% correct (i.e., all letters were correctly identified) rules was higher than that of 50% correct (i.e., the smallest line where at least 50% were correctly identified) rules. Our question is: Do the scores from the ETDRS and SLVA charts agree?
Our study is based on the acuity psychometric function (Fig. 1), which describes observer’s performance accuracy in the optotype identification task as a function of optotype size.21, 22 It can be best described by a sigmoidal function determined by acuity threshold and slope.23 In this study, we first measured the acuity psychometric function of ten adults across 12 viewing conditions: monocularly (with the left or right eye) and binocularly with refractive errors corrected or uncorrected, and with the dominant eye and 6 levels of Bangerter Occlusion Foils (i.e., 0.1, 0.2, 0.3, 0.4, 0.6, 0.8). We then conducted simulated visual acuity tests with the ETDRS and SLVA charts based on the fitted acuity psychometric functions to determine the observers’ VA scores and the probability correct values corresponding to the measured VA scores.