What is visual acuity? What are its limitations?
Visual acuity describes the acuteness or “sharpness” of vision; that is the ability to perceive small details. The primary measurement tool is the letter chart introduced in 1862 by Donders and Snellen at the Eye Infirmary at Utrecht in the Netherlands.
Visual acuity measurement is so common that visual acuity measurement is often mistaken as a unique indicator for vision in general. This is a misconception. Visual acuity loss can detect many disorders, but not all. A primary example is glaucoma, which can do extensive and irreversible visual field damage before visual acuity is affected. Letter chart acuity tells us something about the very small retinal area onto which the letter seen is projected. When the image of that letter is blurred due to optical factors (opacities, refractive error) the surrounding image will be equally blurred. But when visual acuity loss is due to retinal factors, letter chart acuity tells us nothing about how the surrounding retina functions.
Visual acuity measurement is a good screening tool because normal visual acuity requires that all levels of the visual system function properly. The optical system of the eye must project a sharp image of the outside world onto the retina. The retina must then be able to translate this image into neural impulses. Finally, the neural impulses must travel to the brain, where they are analyzed and recognized. Therefore, a wide array of different visual disorders (but not all) can affect visual acuity.
Because visual acuity is so easily measured, it is often used as a primary eligibility criterion, such as 20/20 for a pilot’s license, 20/40 for a driver’s license, 20/200 for certain disability benefits. However, the ability to perform various tasks depends on many more factors than visual acuity alone.
For those disorders that affect visual acuity, it also is a good follow-up tool to document whether a condition is worsening or recovering or to record the effect of various interventions.
For all of these purposes it is desirable that the measurement methods are well standardized and the same from office to office. This was the primary incentive for producing a standardized test (see below) for the Early Treatment Diabetic Retinopathy Study (ETDRS).
Finally, letter charts are often used as an aid in subjective refraction. This is the least demanding application, since the usual question is only: which lens is better, A or B?
Most people are familiar with the notation of visual acuity as a fraction, but few understand what these fractions mean. Yet, the explanation is simple. If a subject needs letters that are twice as large or twice as close as those needed by a standard eye, that person’s visual acuity is said to be 1/2. If the letters needed are five times larger, the acuity is 1/5, if ten times larger: 1/10, etc. The value of those fractions can be expressed in different ways. E.g. 1/2 = 20/40 = 6/12 = 0.5 or 1/5 = 20/100 = 6/30 = 0.2. The MAgnification Requirement is also known as MAR, so that MAR = 1/V and V = 1/MAR. (see below)
Snellen insisted that the numerator of these fractions should indicate the testing distance; e.g. 20/… for 20 ft, 5/… for 5 meters, 6/… for 6 meters. Today, this convention is rarely followed. In the US the 20/… notation is routinely used, even if the testing distance is not 20 ft. In Europe the decimal notation is common; in Britain (and former British dominions) the 6/… notation is common.
In the section on low vision reading tests we will discuss a modified Snellen formula, which is easier to use for near vision.
According to the ICO Visual Acuity Measurement Standard (1984) a line is considered read if “more than half” of the characters are identified correctly. For an ETDRS chart with 5 letters per line, this means 3 or more correct.
Since the value of the visual acuity fraction compares a subject’s performance to the performance of a standard eye, that standard needs to be defined. Snellen chose to define it as the ability to recognize one of his letters when it subtends a visual angle of 5 min of arc. Louise Sloan later coined the name M-unit to describe this measurement unit. One M-unit subtends a visual angle of 5 min of arc at 1 meter. Simple geometry tells us that the same visual angle applies for 2 M-units at 2 meters, 3 M-units at 3 meters, etc. 1 M-unit also happens to be the size of average newsprint, but that is not the basis of its definition.
“Normal” visual acuity for healthy eyes is one or two lines better than 20/20. In population samples the average acuity does not drop to the 20/20 level until age 60 or 70. Always remember that the 20/20 reference standard does not refer to the average acuity of American eyes, just as the US standard foot is defined independently of the “normal” length of American feet.
Since letter charts contain only discreet letter sizes, the measurement accuracy depends on the step size between lines. Most traditional US charts have an irregular sequence of letter sizes (as did Snellen’s original chart). From 20/15 to 20/20 is a 33% increase, from 20/20 to 20/25 is 25%, from 20/60 to 20/70 is 17%, but from 20/100 to 20/200 is a 100% increase. It was soon recognized that equal step sizes would be desirable. The first chart with such a sequence was proposed in 1868 by John Green of St. Louis, who had spent time working with Snellen while touring Europe after his ophthalmology training. He proposed a geometric (logarithmic) sequence:
1.0 1.25 1.6 2.0 2.5 3.0 4.0 5.0 6.3 8.0 10.
Many different geometric progressions (progressions with a constant ratio between adjacent terms) are possible. The progression used in the ETDRS protocol and on all professional Precision Vision charts is known as the Preferred Numbers series. This series is defined in standard #3 of the International Standards Organization and is used in a wide variety of industrial standards. The preferred numbers series
(a) fits well with the decimal system, since each step represents the same 10√10 (100.1) ratio. Thus, 10 steps equal exactly 10x, and after 10 steps the same digits reappear with only a shift in the decimal position.
(b) Preferred numbers are convenient because, with only slight rounding, the series contains mostly whole numbers. Each step represents a 4:5 ratio (rounding error = +0.7%), 3 steps equal a factor 2x (rounding error = -0.2%). The same ratios are used to calculate decibels; 3 decibels = 2x.
(c) Being anchored at 1.0 (10, 100, 1000, etc.) the series is well suited for visual acuity, since the reciprocal of a preferred number and the product or quotient of two preferred numbers is again a preferred number. Thus, if the letter sizes (the denominator of the Snellen fraction) and the viewing distances (the numerator of the Snellen fraction) both follow this series, so will the resulting visual acuity values. This works well for the 20/… and decimal notations; in countries that use the 6/… notation, strict adherence to the series would require a test distance of 6.3 m instead of 6 m (6.3 and 3.2 are preferred numbers, 6 and 3 are not). Since the difference between 6 and 6.3 is only 5% (1/5 of a line, 1 ETDRS letter), this difference can be ignored for ordinary clinical measurements, where the measurement accuracy is in the order of one line. For research studies where multiple acuity measurements are averaged for greater accuracy, the difference may be significant.
In the US the 20/… notation is commonly used as an equivalent notation, meaning that the same notation is used regardless of the testing distance. In Europe, the decimal notation of visual acuity is prominent, while in Britain (and former dominions) the 6/... notation prevails. Others may want to use a true Snellen fraction, where the numerator specifies the test distance. The center part of Table 1 shows conversions between these different notations.
In addition, the left part of the table shows the ranges of vision loss defined in ICD-9-CM. Note again that the range of normal vision extends beyond 20/20.
Table 1 – Visual Acuity Ranges and Visual Acuity Notations
Traditional visual acuity values are well suited for calculations about letter size, magnification and viewing distance. If the visual acuity is 20/100 (1/5), the letter size has to be increased by 5x to achieve the equivalent of 20/20 (1/1) performance. Alternatively, one has to use a 5x magnifier or a 5x telescope, or bring the object 5x closer. This is commonly referred to as Kestenbaum’s rule.
For calculations and comparisons of the functional effects of different visual acuity levels, however, this geometric sequence cannot be used; we need a linear scale with constant increments. Clinicians have long used the expression that visual acuity has changed by a certain “number of lines”. This expression is only valid, if the steps between lines are equal. Equal steps imply a geometric (logarithmic) progression. Taking the logarithm of each value converts the geometric progression of letter sizes to a linear scale of functioning.
Two linear scales are available. One is the logMAR scale, which uses the logarithm directly; it is often found in scientific papers. LogMAR = log10(MAR). In the context of physiological optics, MAR is interpreted as Minimum Angle of Resolution; in a functional context the interpretation as MAgnification Requirement is more intuitive. As we have seen earlier, MAR is the basis for the definition of visual acuity: VA = 1/MAR or MAR = 1/VA.
Although logMAR is often presented as a measure of visual acuity, it actually is a measure of visual acuity loss. A logMAR value of 0 indicates “no loss”, i.e. visual acuity equal to the reference standard (20/20), while normal visual acuity (better than 20/20) is represented by negative logMAR values. Every increment of 0.1 logMAR indicates one line of loss.
The other scale is the Visual Acuity Score (VAS); it is an inversion of the logMAR scale, based on the formula VAS = 100 – 50x logMAR. It serves the same purpose, but is more intuitive since it avoids decimal values and since higher values indicate better acuity. For ETDRS-like charts with 5 letters on each line, the VAS increases by 1 point for every letter read correctly (5 points for every line). Similar letter count scores are used in many studies and surveys (including the ETDRS). VAS values are identified by square brackets; the scale is anchored at  = 20/20 , which means that  = 20/200 and  = 20/2000.
The Visual Acuity Score is well suited for applications with a functional emphasis. It is used as an estimate of visual functioning ability in the 5th and 6th edition of the AMA Guides for the Evaluation of Permanent Impairment. The traditional Snellen values (or the modified Snellen formula discussed in the Low Vision section) are better suited for clinical applications such as calculating the need for changes in magnification or of working distance.
Thus far we have discussed mainly distance visual acuity as measured with letters or symbols. We have seen that