3.1.1 Symptoms

Symptoms of blurred vision or an inability to see well enough for certain tasks (reading the whiteboard, schoolbooks or a newspaper, watching TV, driving, etc.) all suggest reduced vision due to ametropia or ocular disease. These symptoms should be explained by reductions in visual acuity and/or contrast sensitivity. If a patient has symptoms of poor vision, but the visual acuity is normal, contrast sensitivity and visual fields should be measured. Symptoms including unexplained headaches, unexplained visual acuity or contrast sensitivity loss, positive or scintillating scotoma, bumping into things on one side and symptoms consistent with neurological disease, such as cluster headache, acquired migraine, dizziness, tingling of limbs all suggest the need for visual field testing. Obviously if a patient complains of colour vision changes, their colour vision should be assessed.

3.1.2 Ocular, family and medical history

The ocular history may indicate an ocular condition that requires monitoring using visual acuity, contrast sensitivity, colour vision and/or visual fields, such as cataract or glaucoma. The family history may indicate a hereditary condition, such as open-angle glaucoma, that should be carefully checked using visual field assessment. The medical history could indicate a systemic condition that requires particular assessment, such as diabetes, or systemic medication that could affect some aspect of vision, such as chloroquine.

3.2.1 When should distance visual acuity be measured?

Visual acuity (VA) is a measure of the patient’s ability to resolve fine detail. It is the most commonly used measurement of visual function you will make. Distance VA is used to assess the adequacy of spectacle corrections and as a key indicator of ocular health. VA is also used to assess a person’s fitness to drive or enter into some professions such as the police force and to enable registration as a partially sighted or blind person. The need to measure VA accurately is obvious.

There are three principal measures of VA:

VA after objective refraction is also often recorded. Either vision and/or habitual VA should be measured immediately after the case history for legal reasons, to document the VA level prior to your examination. Habitual and optimal distance VA are routine measurements. Measuring distance unaided VA (vision) is optional, and should be measured with patients who:

3.2.2 LogMAR versus Snellen

LogMAR visual acuity (VA) charts (Figure 3.1) use the design principles suggested by Bailey and Lovie, including 0.1 logMAR progression of letter size from −0.3 to 1.0 logMAR (equivalent to 6/3 to 6/60 or 20/10 to 20/200), five letters per line, letters of similar legibility and per-letter scoring.¹ Snellen charts were devised by the German ophthalmologist Hermann Snellen in 1862 and have been widely used ever since. There is not a standard Snellen chart and the letter size sequences, number of letters and/or numbers, varieties of letters, etc., vary from manufacturer to manufacturer. They typically contain one letter at a VA level of 6/60 (20/200; 0.1 logMAR) and increasing number of letters at smaller letter sizes with a typical bottom line of about 6/5 or 20/15 (~ −0.1 logMAR).

The majority of Snellen charts have only one 6/60 or 20/200 letter, two 6/36 or 20/125 letters and three 6/24 or 20/80 letters. LogMAR charts typically have five letters on each of these lines and additional lines of letters at 6/30 or 20/100 (0.70 logMAR) and 6/48 or 20/160 (0.90 logMAR). Most logMAR charts have a ‘bottom line’ of −0.3 logMAR (6/3 or 20/10), whereas many Snellen charts have a bottom line of 6/5 or 20/15 and thus provide truncated data given that the average visual acuity of a young adult is about 6/4 (Table 3.1).²

LogMAR charts are widely recognised as providing the most reliable and discriminative VA measurements and are standard for clinical research or clinical trials of ophthalmic devices or drugs.3,4 Visual acuity measurements using a logMAR chart have been shown to be twice as repeatable as those from a Snellen chart and over three times more sensitive to inter-ocular differences in VA and therefore substantially more sensitive to amblyopic changes for example.^3,5 The Bailey–Lovie or ETDRS charts are the most commonly used logMAR charts for adults and the Glasgow Acuity Cards (commercially available as the Keeler logMAR crowded charts) have been designed specifically for children.^4,5

Many Snellen charts do not contain lines of small letters and are truncated to 6/4, 6/4.5, 6/5 or even 6/6. For some patients, this takes the approach of measuring ‘distance vision adequacy’ (i.e. determining whether distance VA is adequate for a patient’s daily needs, similar to the approach used for near VA) rather than distance VA (a threshold measurement). This makes the detection of slightly reduced VA due to eye disease or uncorrected refractive error in patients with good VA impossible. For example, if your chart is truncated to 6/5 or 20/15, you will not be able to detect a VA loss from 6/3 (or 6/4) to 6/5 or from 20/10 to 20/15.

The major disadvantage of logMAR charts is that although they are available in printed, and projector chart form, they are not as widely available as Snellen charts. However, this is slowly changing. In particular, logMAR VA measurements are being promoted on computer-based visual assessment systems. The new generation of flat panel displays appear especially useful as they are light (easy to wall mount), have excellent resolution, luminance and contrast and are flicker-free. VA measurements on these systems have the advantage of allowing randomisation of letters, calculation of VA scores and conversion of VA into different notations. One practical disadvantage of printed panel logMAR charts is that patients may not be able to see the ‘bottom line’ of the chart (−0.3 logMAR, 6/3, 20/10). If they have been used to having their VA measured on a truncated VA chart and being able to see the bottom line, this may upset some patients. You should explain that the new chart includes lines with smaller text than the old one (‘this new bottom line is just for superman/superwoman’) and indicate, or even highlight, the 6/6 (20/20) line.

Snellen charts have a major advantage at present in that they are widely available and Snellen notation of VA is universally understood. From a practical viewpoint, these charts can generally be produced in a smaller format as considerably fewer large letters are presented and this allows easier display on projector systems and the easy addition of other targets such as a duochrome and spotlight, etc., to wall charts. The charts provide a good target for refraction of patients with good VA as the number of lines between approximately 20/15 to 20/40 (6/5 to 6/12) is similar to logMAR charts and these lines typically have as many, if not more letters than logMAR charts. Although the truncation of Snellen charts (to say 6/5 or 20/15 when some patients can see 6/3 or 20/10) can be a disadvantage, it can speed up the measurement and allows most patients to read your ‘bottom line’.

3.2.3 Procedure

See online video 3.1.

3.2.4 Alternative procedures to assess VA in children

Amblyopia can be missed if single letters are used rather than a letter chart because of the lack of contour interaction. Ideally logMAR-based charts with contour bars at the end of lines, such as the logMAR crowded charts, should be used when measuring VA in children (Figure 3.2).⁸ The logMAR crowded charts have been shown to be over three times more sensitive to inter-ocular differences in VA than single letter Snellen charts and therefore substantially more sensitive to amblyopic changes.⁵ Crowded and standard and logMAR charts can even be used in children who do not know their letters by providing them with a key card that includes a selection of the letters from the chart. You then point to a letter on the chart and ask the child to identify the letter on their key card. For children that are unable to use a key card, charts are now available in logMAR format that include pictures rather than letters (Figure 3.2), such as the Kay crowded picture test, which has been shown to provide comparable results to the logMAR crowded charts.⁹

3.2.5 Adaptation of VA measurement with older patients

A hyperopic shift is common in older patients and some varifocal wearers adapt to this change by habitually raising their chin to improve distance vision by viewing through the additional plus power in the intermediate section of the lens. This can be seen when assessing habitual VA as these varifocal wearers will raise their chin during measurements to improve VA. VAs should be measured through the distance portion of the lens. You should make a mental note that these patients will likely require additional plus (or less minus) power in their distance correction.

3.2.6 Recording for logMAR and Snellen charts

VA measurements can be scored in logMAR notation, using the visual acuity rating (VAR) score or converting scores to an equivalent Snellen value. Computer-based systems will typically convert VA values for you. LogMAR or VAR could be used on your own record cards. However, equivalent Snellen values should generally be provided when writing referral letters and reports, as Snellen notation is universally understood, whereas logMAR is not at present. In all cases, it is preferable to score using a by-letter system rather than measuring the lowest line at which the majority of letters were correctly read as it provides more repeatable and discriminative measurements.¹⁰ Comparisons of various logMAR scores with Snellen and other recording notations are shown in Table 3.2.

The fact that logMAR VAs better than 6/6 or 20/20 are negative is counterintuitive. The VAR score provides a simpler method for scoring logMAR charts. VAR = 100 − 50 logMAR. Therefore 0.00 logMAR = 100 VAR and each letter has a score of 1. For example if a patient reads all the letters down to the 100 row and gets one letter wrong on this row, their score is 99, two letters wrong 98, etc. If they read all of the letters on the 100 row and one letter on the row below, their score is 101, two letters on the line below 102, etc. A disadvantage of the VAR score is that it suggests that 100 (6/6) is normal VA, which is far from true for many patients with healthy eyes.²

The Snellen fraction is defined as:

Vision or visual acuity is recorded as the smallest line in which the majority of the letters are seen, irrespective of subjective blur. Errors are recorded by appending a minus one, two or three to the Snellen fraction or decimal notation. If additional letters are seen on the following line, the Snellen fraction or decimal notation can be appended by a plus (usually up to no more than 3). If the patient could not see the 6/60 letter at 6 m, but could at 2 m, record 2/60. Similarly if the patient could not see the 20/200 letter at 20 feet, but could see the 20/120 letter at 5 feet, record 5/120.

Vision or ‘VA s Rx’ or VAsc or Vsc all mean visual acuity measured without a correction.

VA c Rx or VAcc or Vcc all mean visual acuity measured with a correction and generally refer to the VA with the patient’s spectacles. VA measured with the patient’s contact lenses would typically be recorded as VA c CLs or similar.

VAs are recorded for the right eye (RE or OD), left eye (LE or OS) and binocular (BE or OU).

Examples (logMAR charts):

Examples (Snellen charts):

This last example is given in decimal format, which is commonly used in parts of Europe. It should not be confused with logMAR.

3.2.7 Interpretation

Any deviation from normal age-matched results, as shown in Table 3.1, should be noted. Note that the average VA for patients less than 50 is about −0.14 logMAR (6/4.5, 20/15) and that 6/6 or 20/20 represents reduced VA for the vast majority of patients. 6/6 or 20/20 only becomes the average VA for patients over 70 years of age with healthy eyes. You must also check whether there has been any change in VA from the previous examination. In patients with normal or near normal VA, a significant change in VA is more than 0.1 logMAR or one line.^5,11 Also note any inter-ocular asymmetry of a line or more, or a binocular result that is worse than the monocular response.⁵

By comparing vision and optimal VA or habitual and optimal VA and using the one line of VA is equivalent to −0.25DS rule, an estimate of the mean spherical correction can be gained (section 4.1.5). If this estimate is widely different from the actual subjective refraction result, an error may be suspected and the subjective (and/or spectacle power) rechecked.

3.2.8 Most common errors

3.3.1 Comparison of different near vision charts

Near vision cards typically present sentences or paragraphs of words rather than isolated letters and can incorporate examples of near vision tasks such as sheet music, technical drawings and telephone directories (Figure 3.3). They are also now available on e-tablet and e-phones. There are five main types of notations used to measure near VA: logMAR, N-point, M-scale, equivalent Snellen, and Jaeger. LogMAR near charts have all the advantages of distance logMAR charts (section 3.2.2) and should be used whenever an accurate, non-truncated measurement of near VA is needed and particularly when distance VA and near VA may differ, such as with multifocal implants/contact lenses and patients with subcapsular cataract. Some near logMAR charts use isolated letters rather than words, but word charts are preferred, particularly in patients with conditions such as age-related macular degeneration as it relates better to reading performance and is typically worse than near letter VA.¹² N-point uses the Times New Roman font and is the standard test in the UK and Australia. It is based on the ‘point size’ used by printers and word processing packages on modern computers, in which 1 point = 1/72 inch (~0.353 mm). This can be a useful aid when indicating to a patient the level of vision that would be provided for computer use by new lenses. N8 is approximately equal to 1.0 M and this eight times conversion holds for all print sizes. M-units are widely used in North America and indicate the distance in metres at which the height of a lower case ‘x’ subtends 5 minutes of arc. A 1.0 M letter ‘x’ is therefore 1.45 mm high. Equivalent Snellen notation is a confusing notation, especially when near VAs are not measured at the standard 16′: What does 20/20 near VA at 8′ mean? In addition, near vision adequacy should indicate what patients can see at their own near working distance, not an arbitrary standard of 16′ (working distances for reading in patients with normal vision range from 10′ to 20′¹³). Cards using Jaeger notation should not be used as there is no standardisation of what J1 or J5, etc., means and different charts can give totally different sizes of print with the same J-value.

Many non-logMAR based reading charts are truncated as the smallest print sizes often provided are N5 (at 40 cm equivalent to ~6/9 distance VA) and 0.4 M (at 40 cm equivalent to ~20/20 distance VA) so that many patients could read sentences of text smaller than this if given the chance and a threshold is not measured. Thus ‘near vision adequacy’ is a more appropriate description of the measurements than near visual acuity. Near vision adequacy measurements have the added advantage that the measurement is quicker than a threshold measurement and most patients will be able to see the ‘bottom line’ of text. It is left to distance VA measurements, with its many advantages in this regard (fixed measurement distance, same letter format, letters of similar legibility, etc.) to provide an accurate measurement of a patient’s resolution.

3.3.2 Procedure: logMAR near VA and M or N-notation near vision adequacy

See online video 3.2.

3.3.3 Adaptation for older patients

You should have asked the patient whether they use additional lighting, such as an anglepoise or goose-neck light, to read in their home. If they do not and you subsequently find they cannot easily read N5 (or 0.4 M) at the end of the reading addition part of the refraction, they should be encouraged to obtain additional light for near tasks in their home. It is very useful to demonstrate how helpful such additional light can be. The main objective of people with vision impairment is to be able to read and a majority can be successfully helped in primary eye care using high reading additions, simple magnifiers and additional lighting.¹⁴ Note that although near VA levels can be associated with a range of near tasks (Table 3.3), the poorer near VAs are threshold measurements (measurements of N5, 0.4 M and 20/20 may be truncated and not thresholds as discussed earlier) so that patients would not be able to read print of that size comfortably for any length of time. For example, to allow somebody to read newspaper print comfortably requires a near VA better than N8 or 1.0 M which is the typical size of newspaper print (Table 3.3) as a ‘reading reserve’ of 2 : 1 is needed and a near VA of N4 or 0.5 M.¹⁵

3.3.4 Recording

Note the working distance and then record the smallest paragraph size seen by the right and left eyes and binocularly. For approximate equivalents to other notations see Table 3.3. Some clinicians do not note the working distance unless it is different from the ‘norm’. However, even patients with good vision present a wide range of normal working distances (22 cm to 50 cm for reading,¹³ and further away for computer use), and as stated earlier a reading acuity is meaningless without a working distance. It can also be useful to record the working distance(s) used to determine near VA with the patient’s own spectacles to allow a comparison with the one used to determine the reading add. This can also be useful for comparison if patients return to your practice dissatisfied with the near vision in any new spectacles. These cases are often due to problems with working distance determination rather than an incorrect refraction.

Near vision or ‘NVA s Rx’ or NVAsc or NVsc all mean near visual acuity measured without a correction. NVA c Rx or NVAcc or NVcc all mean near visual acuity measured with a correction and generally refer to the VA with the patient’s spectacles. Near VA measured with the patient’s contact lenses would typically be recorded as NVA c CLs or similar. Near VAs are recorded for the right eye (RE or OD), left eye (LE or OS) and binocular (BE or OU) and should include the near working distance in cm or inches. If the patient can only read a paragraph slowly or with difficulty, include this information.

Examples:

3.3.5 Interpretation

When determining a reading addition, do not assume you have the correct add just because a patient can see the smallest text on your near chart such as N5, 0.4 M or 20/20. At 40 cm 0.4 M is equivalent to about 20/20 or 6/6 distance VA and N5 is equivalent to 6/9 or 20/30. Therefore, a patient could have reduced distance VA and still read N5, 0.4 M or 20/20 at near. Apart from this truncation effect, near VA can be expected to be similar to distance VA in most cases provided that the eye is accommodating normally or that the reading addition is correct. Notable exceptions include patients with multifocal intra-ocular lenses (IOLs) or wearing multifocal contact lenses or patients with posterior sub-capsular cataract. Patients with some eye disorders, such as amblyopia, age-related macular degeneration (ARMD) and macular oedema, can have significantly worse reading VA than distance VA and isolated-letter near VA.¹²

3.3.6 Most common errors

3.4.1 Comparison of tests

The speed and accuracy of contemporary fast threshold estimation strategies have made several of the traditional screening techniques somewhat redundant. Fast thresholding strategies can produce an estimation of visual field sensitivity in a time (2.5–4 minutes per eye) similar to single stimulus, suprathreshold screeners. All of the fast central field analysis techniques have the advantage over suprathreshold screening techniques in that they are better able to detect early visual field defects, and can give an idea of defect depth and area. They have the disadvantage of taking longer than some suprathreshold techniques. They are similar in sensitivity and specificity for glaucoma as frequency doubling perimetry, but much better at detecting other types of visual field defect.17,18 When compared to techniques for full central field analysis they are quicker but less precise and with worse test-retest characteristics.¹⁹ The Humphrey Field Analyser (HFA) II, SITAFast (Swedish Interactive Thresholding Strategy), Central 24-2 tests 58 locations over the central 25° in a 6° grid pattern that straddles the horizontal and vertical mid-lines, i.e. targets are located 3° either side of the mid-lines. In addition there are targets located on the nasal field between 25° and 30°. The SITAFast, 24-2 programme rarely takes more than 3.5 minutes in a normal patient, and can be as quick as 2.5 minutes. Most modern perimeters have similar fast thresholding central visual field programmes.

3.4.2 Procedure

3.4.3 Recording

If no test locations are highlighted on the Total and Pattern Deviation probability plots then record ‘SITA-Fast: WNL (within normal limits) R and L’. If there is a field defect, repeat the test, print and store both fields of both eyes and attach to the record card. SITA-Standard (or equivalent) could be used instead of SITA-Fast for the repeated field. The single field analysis of the repeated field should accompany any referral to the secondary eye care system. The single field analysis printout illustrates the data as an interpolated greyscale, raw data in decibels, and Total and Pattern deviation plots. It also summarises the field using the Glaucoma Hemifield Test, Global Indices, Reliability Indices, and Gaze Tracking plots (section 3.5.4).

3.4.4 Interpretation

These are interpreted in exactly the same way as the strategies for full central field analysis (section 3.5.4). It is important to be aware of the possible causes of artefact in cases where it would appear that a new defect has been detected, particularly in a patient with no previous experience of visual field screening or no history of field loss. A new defect is not a defect until it is repeated, ‘if in doubt always repeat’. The effect of learning and fatigue can be dramatic.²⁰ Note that when the Glaucoma Hemifield Test classifies the field as having a ‘general reduction of sensitivity’, the Mean Defect/Deviation is abnormal and/or the Total Deviation probability plot shows a majority of test locations as being outside of normal limits, care should be taken when interpreting the results. There is usually an obvious clinical reason, with the most likely association being with cataracts or small pupils.

3.4.5 Most common errors

• Significant risk factors for glaucoma (compare with section 3.4) including, but not limited to, combinations of IOP >21 mmHg, old age, family history, narrow angles, vertical elongation of the optic nerve head, notching of the neural rim tissue of the optic nerve head, disc haemorrhage, nerve fibre layer defect, exfoliative syndrome, pigment dispersion, optic nerve head asymmetry.

• Abnormal screening test (e.g. positive screening test, confrontation or Amsler).

• Symptoms consistent with neurological disease (for example, headache including migraine, dizziness, tingling of limbs) or neuro-ophthalmic disease.

• Symptoms consistent with central field loss, e.g. non-refractive reduced vision, positive scotoma, scintillating scotoma.

3.5.1 Comparison of tests

The instrument should be capable of monitoring fixation, providing full threshold fields in less than 8 minutes, providing reliability indices and analysing the results. A rapid threshold estimation algorithm, such as the HFA’s SITAStandard or the Octopus Dynamic Strategy is recommended. These strategies take approximately 7 to 9 minutes per eye, without compromising the accuracy or repeatability of the result.22,23 The use of faster, less repeatable, thresholding strategies (e.g. HFA SITAFast and Octopus TOPs; section 3.4) may be considered as an alternative for some patients with a demonstrated history of fatigue. There has been discussion that SITA should not be used in patients suspected of conditions other than glaucoma as they are optimised specifically for glaucoma. However, clinically the advantage of the reduced test time makes such a compromise worthy of consideration, and no evidence has been presented that suggests a reduction in diagnostic capability for non-glaucomatous defects.

3.5.2 Procedure

The procedure for central visual field analysis is the same as that for central visual field screening (section 3.4.2) except that:

3.5.3 Recording

If no test locations are highlighted on the Total and Pattern Deviation probability plots then record ‘SITA-Standard (30-2): WNL (Within Normal Limits) R and L’. Print the fields for both eyes and attach to the record card. If a new defect has been detected, particularly in a patient with no previous experience of perimetry or no history of field loss, then repeat the field measurement. Confirmation of visual field abnormalities is essential for distinguishing reproducible visual field loss from long-term variability (section 1.2). The single field analysis printout illustrates the data as an interpolated greyscale, raw data in decibels, and Total and Pattern deviation plots (Figure 3.4). It also summarises the field using the Glaucoma Hemifield Test, Global Indices, Reliability Indices, and Gaze Tracking plots. When monitoring glaucoma, the Guided Progression Analysis (GPA), which is based upon the Early Manifest Glaucoma Trial, is designed to identify true glaucoma progression.²⁴

3.5.4 Interpretation

See online figures 3.4i to 3.4viii, with interpretation. Visual fields should be interpreted with respect to their reliability, as a single field and with respect to change over time.

3.5.5 Most common errors

See section 3.4.5.

3.6.1 Comparison of tests

The recommended approach to peripheral field testing is to first record a fast threshold central test (see section 3.4) followed by a peripheral suprathreshold screening programme that tests between 30 and 60 degrees. Similar programmes can be found on most modern bowl perimeters. Alternate methods would include a full field suprathreshold screening programme or combining a fast threshold central test with a fast threshold peripheral test. It should be noted that there has been no comparison of these different approaches.

Virtually all visual field defects, including those due to chiasmal or post-chiasmal lesions, are reflected within the central 30° visual field.²⁹ This is simply due to the anatomy of the visual pathway. There is a systematic bias towards representation of the central visual field, with over 80% of the visual pathway dedicated to processing the central 30° of vision.³⁰

3.6.2 Procedure

The example used is the Humphrey Field Analyser, Three-zone, Peripheral 60.

3.6.3 Recording

If all test locations are labelled as being ‘within normal limits’ on the printout then record ‘Peripheral 60, 3-zone: WNL (within normal limits) R and L’. If there is a defect evident then print the fields for both eyes, attach to the record card and consider a confirmatory field. The printout illustrates the data with symbols that designate the location as ‘within normal limits’, ‘relative defect’ and ‘absolute defect’. There is usually some indication of patient reliability, e.g. test time, catch trials and fixation losses. The combined printout will show the threshold 30-2 result as a greyscale, surrounded by the Peripheral 60 symbols.

3.6.4 Interpretation

Identify any clusters of relative or absolute defect. Repeatable clusters of three or more relative defects should be noted. Look for continuity of defect from the central to peripheral field. As with all visual field analysis the position and shape of a defect, along with additional clinical findings, will dictate the management of the patient.

3.6.5 Most common errors

See section 3.4.5.

3.7.1 Comparison of tests

Standard automated perimetry has been shown to be much more sensitive, specific, reliable and valid for detecting central visual field changes than Amsler charts.32,33 Standard Amsler charts are high contrast and even threshold adaptations of the chart (using cross-polarising filters) are poor at detecting scotomas smaller than 6°.³² Schuchard also found that more than half of the distortion reported in Amsler grids was at retinal locations that corresponded to the location of scotoma. Amsler charts rely heavily on subjective interpretation and may also be compromised by the ‘completion phenomena’, which perceptually fills small gaps in line stimuli.³³

The Amsler Grid is an alternative to 10° central visual field analysis if a quick assessment of macular function is required and it is particularly useful in cases with metamorphopsia or visual distortion. Amsler charts have the advantage that they are portable, so can be used for home visits. The recording sheets can be used for home monitoring, although compliance has been shown to be poor and it is likely that the white-on-black Amsler charts are more sensitive to macular changes than the black-on-white recording sheets.34,35

3.7.2 Procedure for the Humphrey Field Analyser 10-2 programme

The same as central visual field analysis but replace programme 30-2 with programme 10-2 (Section 3.5.2).

3.7.3 Procedure for Amsler charts

3.7.4 Recording

3.7.5 Interpretation

3.7.6 Most common errors

3.8.1 Advantages and disadvantages

The binocular Esterman grid is now available on several of the automated perimeters, including the Humphrey Field Analyser, and gives an automated score. As such it has gained in popularity for the evaluation of visual impairment and visual disability.²¹ It has not been validated for use with standard automated perimetry, but has become a standard of measurement in many circumstances, including the driving standard in several countries.

3.8.2 Procedure

The example used is the Humphrey Field Analyser binocular Esterman test. Other perimeters have similar testing procedures.

Standard visual field set-up applies, but in addition:

3.8.3 Recording

The printout uses a non-interpolated grey scale to illustrate those grid locations that were seen (open circle) at the 10 dB screening level, and those that were missed (black box). The number of seen and missed points are also stated as a proportion of the total 120 grid locations and an efficiency score is expressed as the percentage seen. The efficiency score is then used to judge the patient’s disability. At the end of the test print the result and record the efficiency score as a percentage.

3.8.4 Interpretation

The results are interpreted as a percentage of visual function, giving an indication of visual disability. For driving standards it is often necessary to assess the extent of the horizontal binocular visual field. Several jurisdictions consider 120° or more of continuous horizontal field to be the required standard. The percentage rate of false positives is an important check of the reliability of the test, as some patients can try to improve their chances of ‘passing’ this driving standard test by pressing the response button when a light was not seen. Typically, a false positive score above 20% means that the visual fields are unreliable and not acceptable to a driving standards agency, so that the test must be repeated.

3.8.5 Most common errors

These are the same as for visual field screening (section 3.4.5).

3.9.1 Comparison of tests

The prime use of simple visual field tests is during home (domiciliary) visits. Their only advantages are that they are portable and inexpensive compared to automated perimeters. The most sensitive method appears to be examination of the central visual field with a red target(s).37–39 From a visual field screening point of view, all of these tests have been shown to be insensitive to all but gross field defects such as homonymous hemianopias when compared to automated perimetry.^38,39 It is advisable for general medical practitioners who suspect a patient may have a visual field defect to refer such patients for automated field testing rather than relying on the results of a confrontation test.

3.9.2 Procedure: for red card test³⁷

3.9.3 Recording

Record whether any of the squares or parts of them were missed or looked a paler red than the others. A normal result could be recorded as ‘Fields grossly full to red card test’.

3.9.4 Interpretation

The red card test is useful in the detection and monitoring of large absolute defects (e.g. hemianopsia) as a simple, portable, ‘bedside’ test and appears superior to confrontation tests.37,39

• Matching coloured objects such as clothes, paints and materials used in crafts and hobbies.

• Differentiating differently coloured objects such as ripe and unripe fruit, school workbooks, features on maps.

• Judging when meat is cooked.

• Recognising skin rashes and sunburn.

3.10.1 Comparison of tests

The Ishihara test (Figure 3.7) is a very efficient screening test for red–green colour deficiency and provides quick and simple measurements.⁴¹ It is by far the most commonly used colour vision test and is a required entrance test for several professions throughout the world, including the armed forces, aviation and railway industries. Disadvantages include that the Ishihara plates do not assess tritan colour problems. In addition, it is designed as a screening test around the normal/abnormal boundary and is therefore less useful for grading the severity of a defect or for monitoring an acquired colour deficiency. To grade the severity of a congenital or acquired colour deficiency, for monitoring acquired colour defects, and detecting blue–yellow defects, either the City University or Farnsworth D-15 tests or similar tests are recommended. After several years, note that the colours on the Ishihara plates can fade and the test loses its validity. Tests similar to the Ishihara are often provided on computer-based eye examination systems, although their validity has yet to be confirmed.

3.10.2 Procedure: Ishihara test

See online video 3.3. The Ishihara test is made up of several plates that present various numbers made up of coloured dots of varying size embedded in a background of different coloured dots (Figure 3.7). The colours of the number and background dots are chosen so that they are confused by patients with red–green colour defects (i.e. they appear isochromatic to those with colour defects) but discriminated by patients with normal colour vision. Plate 1 is a demonstration plate that should be read by all literate patients and can be used to indicate malingerers. Different designs of pseudoisochromatic plates follow, and include transformation (plates 2–9), vanishing (10–17) and hidden digit (18–21) plates. Normal trichromats can see numbers on all but the hidden digit plates. Patients with red–green colour deficiency do not see a number on the vanishing plates, see a different number than normals on the transformation plates (Figure 3.7) and can see a number on the hidden digit plates. Classification plates, which attempt to differentiate protans and deutans, are found on plates 22–25. Two numbers are shown on each plate. The right hand number (blue-purple) is not seen or seen less well by deutans, and the left hand number (red-purple) is not seen or less well seen by protans (Figure 3.8). The rest of the plates contain pseudoisochromatic pathways and are used for patients who cannot read letters, such as young children. The patient’s task is to trace the pathway (Figure 3.9).

3.10.3 Recording

Record the number of plates correctly determined from the number of plates attempted. Patients with normal colour vision will make few, if any, errors. If a patient fails the test, attempt to categorise the defect using the result from the classification plates. You should also record any advice given to the patient and their family.

Examples:

3.10.4 Interpretation

Patients with red–green colour defects will make specific errors as indicated in Table 3.4. Generally, three or more errors constitutes a fail, although entrance requirements for some professions allow no errors.^41,43 Some clinicians do not present the hidden-digit plates, which are not very sensitive to colour deficiency, and just present transformation and vanishing plates.⁴¹ Mistakes that are NOT the specific mistakes that red–green colour defectives make should be viewed with caution, as they are much less likely to indicate red–green colour deficiency.⁴³

3.10.5 Most common errors

3.10.6 Alternative procedure: Standard Pseudoisochromatic Plates Part 2

The SPP-2 is another very efficient screening test for red–green colour deficiency and has the advantage over the Ishihara in that it can also be used to screen for blue–yellow defects.44,45 Its major disadvantage is that it is not used as a standard entrance requirement for certain professions like the Ishihara test and therefore is less commonly used. Table 3.5 is a modified score sheet for the SPP-2. Place an ‘X’ through the figures on each plate that were missed and record the total number of blue–yellow, red–green and scotopic errors. Asking the patient to identify which number is more distinct when two figures are seen on each page may only be helpful when testing for subtle differences between the two eyes. If the only BY error was on plate 4, then repeat the plate to rule out the error being caused by the patient’s expectation of seeing only one figure. When totalling the errors, ignore any mistakes of ‘2’ on plate 3 and ‘3’ on plate 6. The ‘2’ is very difficult to discern and almost everyone misses it. The ‘3’ is a reference that can be used to compare the visibility of the two numbers on the page. A patient over 60 years of age fails the test if they make two or more errors, a patient less than 20 years of age fails the blue–yellow part of the test with two or more errors. The failure criteria for all other patients is one or more errors. Classifying an error based on a fail on plate 12 can be confusing because the figure can be missed by individuals with either a protan or tritan defect. In this case, errors on other plates should be considered. With other red–green errors, but not blue–yellow, classify the patient as a protan. With no other red–green errors, then the error should be classified as blue–yellow. With acquired defects, both red–green and blue–yellow errors can occur along with failing plate 12 and additional testing should be carried out with either the Farnsworth-Munsell D-15 or the City University (TCU) test.

3.11.1 Comparison of tests

Because the Ishihara test is relatively poor at grading the severity of congenital and acquired colour defects and monitoring acquired colour defects, and cannot detect blue–yellow defects, an additional colour vision test is required. The Farnsworth-Munsell D-15 test consists of 16 caps that each contains a paper of a different colour (Figure 3.10). The differences between the colours are relatively large and the test was designed to separate patients into those with a mild colour defect who pass the test and those with a moderate to severe defect that fail the test. It can grade the severity of colour vision problems and can test for blue–yellow and red–green defects so that it can be used to detect and monitor all patients with acquired colour deficiency. It is more sensitive to protan loss than the City University test.⁴⁸

The City University test contains ten plates that each displays a central coloured dot surrounded by four coloured dots derived from the Farnsworth-Munsell D-15 test.⁴⁹ The patient’s task is to select the peripheral coloured dot that looks most similar in colour to the central dot. Three of the peripheral colours are chosen as typical isochromatic confusion colours for patients with a protan, deutan or tritan deficiency respectively. The fourth colour is very similar to the central coloured dot and is the one chosen by patients with normal colour vision. The 2nd edition of the TCU is preferred to the 3rd edition, which has not been independently evaluated and is substantially different from the 2nd edition.⁴⁹ It can grade the severity of colour vision problems and can test for blue–yellow and red–green defects so that it can be used to detect and monitor all patients with acquired colour deficiency.

However, the D-15 and TCU are not as sensitive to subtle colour vision defects as the Ishihara test and patients with a mild red–green defect could pass the D-15 and/or TCU and yet fail the Ishihara. Therefore the D-15 and TCU must never be used as a screening test, particularly given that passing the more stringent Ishihara test is a common requirement for some professions. The usefulness of any colour vision test is influenced by whether it is used as part of the entrance requirements to certain professions in the region you are working.

3.11.2 Procedure: The Farnsworth-Munsell D-15

See online video 3.4.

3.11.3 Procedure: The City University (TCU) test

3.11.4 Recording

Examples:

3.11.5 Interpretation

Patients with normal colour vision should make no errors.

3.11.6 Most common errors

3.12.1 Comparison of tests

Pelli-Robson CS is quickly and simply measured and provides a reliable measurement of low spatial frequency CS (0.5–1 cycles/degree) when measured at the standard 1 m. It provides significantly more repeatable measures than sine-wave grating charts such as the Vistech)⁵², FACT⁵³ or CSV-1000 charts.⁵⁴ Other ‘large letter’ contrast sensitivity charts such as the MARS test are similarly repeatable, but those provided on electronic test charts may not be due to issues with liquid crystal display screens at low contrast.⁵⁵ If high contrast visual acuity is used to assess the high spatial frequency end of the CS curve, then a combination of Pelli-Robson CS and standard VA provides an indication of the whole CS curve. For example, a patient with low frequency CS loss only would have reduced Pelli-Robson CS and normal VA; a patient with high frequency CS loss only would have normal Pelli-Robson CS and reduced VA and a patient with CS loss at all frequencies would have reduced Pelli-Robson CS and reduced VA.⁵⁰ The Pelli-Robson chart is ideal when determining functional vision loss in patients with low vision and moderate and dense cataract, when screening for low spatial frequency loss in patients with optic neuritis, multiple sclerosis or visual pathway lesions and when examining diabetics with little or no background retinopathy and patients with Parkinson’s or Alzheimer’s disease. The Pelli-Robson chart can be used at longer working distances such as 3 m, so that higher spatial frequencies are assessed and it becomes more sensitive to conditions such as early cataract. One disadvantage of the chart is that a variable endpoint can be gained depending on how long the patient is left to stare at the letters near threshold.⁵⁰

3.12.2 Procedure

3.12.3 Recording

Record the CS score in log units.

Examples:

3.12.4 Interpretation

For patients between 20 and 50 years old, monocular CS should be 1.80 log units and above; for patients less than 20 years old and older than 50 years, monocular CS should be 1.65 log units and above. It is best to obtain your own norm values. If the monocular scores are equal, the binocular score should be 0.15 log units higher (binocular summation). With increasingly unequal monocular CS, the binocular summation will reduce and in some patients, the best monocular score can be better than the binocular (binocular inhibition).

3.12.5 Most common errors

3.12.6 Useful additional techniques: Small letter CS and low contrast VA

Small-letter CS is more sensitive than traditional VA to several clinical conditions, such as early cataract and contact lens oedema and should be ideal when attempting to measure subtle losses of vision such as after refractive surgery.⁵⁶ CS of very small letters, such as 20/30, correlates very highly with VA and the ideal size for a small-letter test may be about 20/50.⁵⁷ The measurement procedure is very similar to that for the Pelli-Robson chart.

Low-contrast VA charts measure the smallest letter that can be resolved at a fixed contrast and do not measure CS. It is difficult to state which spatial frequencies the low-contrast letter charts are measuring, because this depends on the VA threshold. If only the large letters at the top of the chart can be seen, the score gives an indication of CS at intermediate spatial frequencies. If a patient can see the small letters at the bottom of the chart, the score gives an indication of higher spatial frequencies. Low-contrast VA scores are believed to indicate the slope of the high-frequency end of the CSF. It has been suggested that they can be used to indicate the CSF when used in combination with a low-frequency or peak CS measure such as the Pelli-Robson chart and a high-contrast VA measurement. The lower the contrast of the acuity charts, the more sensitive they become to subtle vision loss. For example, for detecting subtle vision losses in aviators or subtle changes after refractive surgery, 5–10% charts should be used. For greater losses in vision, such as cataract, even the large letters on these very low contrast charts cannot be seen by some patients, and a higher-contrast chart at about 25% is necessary. As with high contrast VA measurements, charts that follow the Bailey-Lovie (logMAR) design principles should be used and the measurement procedure is the same as for high contrast visual acuity (section 3.2).

3.13.1 Comparison of tests

For routine optometric practice, disability glare is mainly used to help determine functional vision loss in patients with cataract. In these situations, the amount of intraocular light scatter is large and disability glare is probably best determined by remeasuring visual acuity while directing a glare source, such as a penlight or ophthalmoscope, into the eye. The disadvantage of this technique is the lack of standardisation of the amount of glare light reaching the eye.⁵⁰ This can be remedied by using a standardised glare source such as the Brightness Acuity Tester (BAT), although this is a relatively expensive instrument for what it provides.⁵² For more specialised practices, more sensitive measures of disability glare can be obtained using a low contrast VA chart or letter contrast sensitivity chart with a BAT or using a straylight meter.^52,58 The latter is sensitive to relatively minor changes in intraocular light scatter, such as in contact lens wearers and pre- and post-refractive surgery.^58,59

3.13.2 Procedure: Simple penlight glare test

3.13.3 Recording

Record as visual acuity with glare. For example:

3.13.4 Interpretation

The amount of glare loss will differ depending on the test conditions such as the glare illumination and glare angle (disability glare is inversely proportional to the glare angle) and you should attempt to keep these constant and develop your own mean data for healthy eyes. Typically, most patients will show no change in visual acuity. A decrease in visual acuity of one line or less is normal. Some cataract patients can lose four lines of visual acuity and more. This test gives the patient’s visual acuity under bright light conditions, which can be reduced with media opacification: corneal scars, post-refractive surgery, cataracts, posterior capsular opacification or central vitreous floaters. A poor visual acuity in glare conditions can provide justification for early referral of patients with cataract or posterior capsular opacification who have good visual acuity in normal light conditions.

3.13.5 Most common errors

3.14.1 Comparison of tests

Although fundus biomicroscopy provides a much clearer view of the retina behind media opacity than direct ophthalmoscopy, subtle maculopathies can still provide inconclusive funduscopic findings. Schein et al. reported that 63% of patients who were predicted by an ophthalmic examination to have visual acuity of 6/12 (20/40, 0.5) or worse after surgery (the level at which cataract surgery is typically deemed ‘unsuccessful’) actually attained a visual acuity of 6/9 or better.⁶³ This suggests that potential vision tests are required in addition to standard clinical tests such as case history information, dilated fundus examination and the swinging flashlight test. Unfortunately, the most commonly used tests, the potential acuity meter (PAM) and the various interferometers, cannot penetrate dense or even moderate cataracts and suggest that potential vision is poor in these cases regardless of the state of the neural system.⁶⁴ In addition, the interferometers in particular can predict good post-operative vision in patients with certain retinal diseases that is not obtainable.⁶⁵ The ‘super’ pinhole test is a very simple potential vision test that has provided encouragingly accurate results, superior to the previous standard tests of the PAM and interferometers in moderate cataract.^64,66,67 This is simply a pinhole visual acuity test that incorporates adaptations to overcome the loss of light when a VA chart is viewed though a pinhole and cataract.

3.14.2 Procedure: Super pinhole test

3.14.3 Recording

Record the near VA obtained with the super pinhole test:

3.14.4 Interpretation

The near VA obtained with the super pinhole VA gives an indication of the possible near VA after uncomplicated cataract surgery. The test cannot bypass dense cataracts so that in such cases the super pinhole result is likely to be worse than the post-operative VA and just represents the minimum VA that is likely to be obtained after surgery.⁶⁴ Other results and assessments should be taken into account when considering the likely visual outcome after cataract surgery and these include the patient’s age, indications from the case history and results from a dilated fundus examination and swinging flashlight test. Visual acuity in the pseudophakic eye is another useful indicator for patients undergoing second eye surgery.

3.14.5 Most common error

Not allowing the patient an opportunity to move the multiple pinhole occluder around to get the best possible potential visual acuity reading.

3.15.1 Photostress recovery time

Long lasting after-images (longer than 15 seconds) produced by relatively brief flashes of light are due to photochemical changes in the receptors. Light absorption by rhodopsin (it is assumed that cone photopigments work in a similar fashion, although slightly faster) leads to the separation of the retinal chromophore from opsin. This process is called bleaching as it results in the loss of rhodopsin’s purple colour. While the photopigment is being regenerated, the patient will see an after-image. Photostress recovery time (PSRT) determines the speed at which the photopigment regenerates. Recovery time is solely dependent on the speed of regeneration of the photopigment and is therefore a test of retinal function. It should be independent of other disease within the visual pathway and to some extent will be independent of mild cataract as long as sufficient bleaching light can reach the retina. Only macular disease will cause a prolonged photostress recovery time. The main disadvantage of the test is that there is no standardisation of the procedure. Recovery times are primarily dependent on the brightness of the light and the length of time used, and it is best to obtain your own normal values with your own particular technique and instrumentation.⁶⁸

3.15.2 Procedure

3.15.3 Recording

Record the time taken in seconds to recover to within one line of pre-bleached visual acuity in seconds.

Examples:

3.15.4 Interpretation

When used with patients with unilateral visual acuity loss, the results are compared between the ‘good’ and the ‘bad’ eye. If the recovery times are similar in the two eyes, then the cause of the poor visual acuity in the ‘bad’ eye is likely to be an optic nerve lesion. If the PSRT is much longer in the bad eye compared to the fellow eye, then the cause of the poor visual acuity is likely retinal. It is generally suggested that any PSRT longer than 50 sec is abnormal, and suggests a macular disease rather than optic nerve abnormality. Of course, recovery times are primarily dependent on the brightness of the light and the length of time used, and it is best to obtain your own normal values with your own particular technique and instrumentation.⁶⁸

3.15.5 Most common errors

1. Bailey, IL, Lovie, JE. New design principles for visual acuity letter charts. Am J Optom Physiol Opt. 1976;53:740–745.

2. Elliott, DB, Yang, KC, Whitaker, D. Visual acuity changes throughout adulthood in normal, healthy eyes: seeing beyond 6/6. Optom Vision Sci. 1995;72:186–191.

3. Lovie-Kitchin, JE. Validity and reliability of visual acuity measurements. Ophthalmic Physiol Opt. 1988;8:363–370.

4. Ferris, FL, Bailey, I. Standardizing the measurement of visual acuity for clinical research studies: Guidelines from the Eye Care Technology Forum. Ophthalmology. 1996;103:181–182.

5. McGraw, PV, Winn, B, Gray, LS, Elliott, DB. Improving the reliability of visual acuity measures in young children. Ophthalmic Physiol Opt. 2000;20:173–184.

6. Carkeet, A. Modeling logMAR visual acuity scores: effects of termination rules and alternative forced-choice options. Optom Vision Sci. 2001;78:529–538.

7. Schulze-Bonsel, K, Feltgen, N, Burau, H, et al. Visual acuities ‘hand motion’ and ‘counting fingers’ can be quantified with the Freiburg visual acuity test. Invest Ophthalmol Vis Sci. 2006;47:1236–1240.

8. McGraw, PV, Winn, B. Glasgow Acuity Cards: a new test for the measurement of letter acuity in children. Ophthalmic Physiol Opt. 1993;13:400–404.

9. Jones, D, Westall, C, Averbeck, K, Adolell, M. Visual acuity assessment: a comparison of two tests for measuring children’s vision. Ophthalmic Physiol Opt. 2003;23:541–546.

10. Bailey, IL, Bullimore, MA, Raasch, TW, Taylor, HR. Clinical grading and the effects of scaling. Invest Ophthalmol Vis Sci. 1991;32:422–432.

11. Bailey IL. Visual acuity. In: Borish’s Clinical refraction. (WJ Benjamin, ed.). 2nd ed. Philadelphia: WB Saunders; 2006.

12. Cacho, I, Dickinson, CM, Smith, HJ, Harper, RA. Clinical impairment measures and reading performance in a large age-related macular degeneration group. Optom Vision Sci. 2010;87:344–349.

13. Millodot, M, Millodot, S. Presbyopia correction and the accommodation in reserve. Ophthalmic Physiol Opt. 1989;9:126–132.

14. Elliott, DB, Trukolo-Ilic, M, Strong, JG, et al. Demographic characteristics of the vision-disabled elderly. Invest Ophthalmol Vis Sci. 1997;38:2566–2575.

15. Latham, K, Tabrett, DR. Guidelines for predicting performance with low vision aids. Optom Vision Sci. 2012;89:1316–1326.

16. Burr, JM, Mowatt, G, Hernández, R, et al. The clinical effectiveness and cost-effectiveness of screening for open angle glaucoma: a systematic review and economic evaluation. Health Technol Assess. 2007;11:1–190.

17. Liu, S, Lam, S, Weinreb, RN, et al. Comparison of standard automated perimetry, frequency-doubling technology perimetry, and short-wavelength automated perimetry for detection of glaucoma. Invest Ophthalmol Vis Sci. 2011;52:7325–7331.

18. Noval, S, Contreras, I, Rebolleda, G, et al. A comparison between Humphrey and frequency doubling perimetry for chiasmal visual field defects. Eur J Ophthalmol. 2005;15:739–745.

19. Artes, PH, Iwase, A, Ohno, Y, et al. Properties of perimetric threshold estimates from full threshold, SITA standard, and SITA fast strategies. Invest Ophthalmol Vis Sci. 2002;43:2654–2659.

20. Hudson, C, Wild, JM, O’Neill, EC. Fatigue effects during a single session of automated static threshold perimetry. Invest Ophthalmol Vis Sci. 1994;35:268–280.

21. Anderson, DR, Patella, VM. Automated static perimetry. CV. St. Louis: Mosby; 1999.

22. Wild, JM, Pacey, IE, O’Neill, EC, Cunliffe, IA. The SITA perimetric threshold algorithms in glaucoma. Invest Ophthalmol Vis Sci. 1999;40:1998–2009.

23. Budenz, DL, Rhee, P, Feuer, WJ, et al. Sensitivity and specificity of the Swedish interactive threshold algorithm for glaucomatous visual field defects. Ophthalmology. 2002;109:1052–1058.

24. Leskea, MC, Heijl, A, Hyman, L, et al. Factors for progression and glaucoma treatment: the Early Manifest Glaucoma Trial. Curr Opin Ophthalmol. 2004;15:102–106.

25. Lalle P. The Humphrey visual field analyzer. In: Primary Care of the Glaucomas (M Fingeret, T Lewis, eds). 2nd ed. New York: McGraw-Hill; 2001. pp. 213–53.

26. Katz, J, Sommer, A. Reliability indices of automated perimetric tests. Arch Ophthalmol. 1988;106:1252–1254.

27. Heijl, A, Lindgren, G, Olsson, J. A package for the statistical analysis of visual fields. Doc Ophthalmol Proc Ser. 1987;49:153–168.

28. Heijl, A, Lindgren, A, Lindgren, G. Test re-test variability in glaucomatous visual fields. Am J Ophthalmol. 1989;108:130–135.

29. Wellings, PC. Detection and recognition of visual field defects resulting from lesions involving the visual pathways. Aust NZ J Ophthalmol. 1989;17:331–335.

30. Drasdo, N, Peaston, WC. Sampling systems for visual field assessment and computerised perimetry. Br J Ophthalmol. 1980;64:705–712.

31. Marmor MF, Kellner U, Lai TY, et al. Revised recommendations on screening for chloroquine and hydroxychloroquine retinopathy. Ophthalmology 2011;118:415-22.

32. Schuchard, RA. Validity and interpretation of Amsler grid reports. Arch Ophthalmol. 1993;111:776–780.

33. Achard, OA, Safran, AB, Duret, FC, Ragama, E. Role of the completion phenomenon in the evaluation of Amsler grid results. Am J Ophthalmol. 1995;120:322–329.

34. Fine, AM, Elman, MJ, Ebert, JE, et al. Earliest symptoms caused by neovascular membranes in the macula. Arch Ophthalmol. 1986;104:513–514.

35. Roper-Hall, MJ. The usefulness of the Amsler chart. Eye. 2006;20:508.

36. Esterman, B. Functional scoring of the binocular field. Ophthalmology. 1982;89:1226–1234.

37. Mandahl, A. Red square test for visual field screening. A sensitive and simple bedside test. Acta Ophthalmol. 1994;72:683–687.

38. Elliott, DB, North, I, Flanagan, J. Confrontation visual field tests. Ophthalmic Physiol Opt. 1997;17:S17–S24.

39. Pandit, RJ, Gales, K, Griffiths, PG. Effectiveness of testing visual fields by confrontation. Lancet. 2001;358(9295):1820.

40. Cole, BL. The handicap of abnormal colour vision. Clin Exp Optom. 2004;87:258–275.

41. Birch, J. Efficiency of the Ishihara test for identifying red–green colour deficiency. Ophthalmic Physiol Opt. 1997;17:403–408.

42. Long, GM, Lyman, BJ, Tuck, JP. Distance, duration and blur effects on the perception of pseudoisochromatic stimuli. Ophthalmic Physiol Opt. 1985;5:185–194.

43. Miyahara, E. Errors reading the Ishihara pseudoisochromatic plates made by observers with normal colour vision. Clin Exp Optom. 2008;91:161–165.

44. Hovis, JK, Cawker, CL, Cranton, D. Comparison of the standard pseudoisochromatic plates–Parts 1 and 2–as screening tests for congenital red-green color vision deficiencies. J Am Optom Assoc. 1996;67:320–326.

45. Vu, BL, Easterbrook, M, Hovis, JK. Detection of color vision defects in chloroquine retinopathy. Ophthalmology. 1999;106:1799–1803.

46. Ismail, GM, Whitaker, D. Early detection of changes in visual function in diabetes mellitus. Ophthalmic Physiol Opt. 1998;18:3–12.

47. Elliott, DB, Skaff, A. Vision of the famous: the artist’s eye. Ophthalmic Physiol Opt. 1993;13:82–90.

48. Oliphant, D, Hovis, JK. Comparison of the D-15 and City University (second) color vision tests. Vision Res. 1998;38:3461–3465.

49. Birch, J. Clinical use of the City University Test (2nd edition). Ophthalmic Physiol Opt. 1997;17:466–472.

50. Elliott, DB, Contrast sensitivity and glare testing. In: Borish’s Clinical Refraction (WJ Benjamin, ed.), 2nd ed. Philadelphia: WB Saunders, 2006.

51. Leat, SJ, Legge, GE, Bullimore, MA. What is low vision? A re-evaluation of definitions. Optom Vision Sci. 1999;76:198–211.

52. Elliott, DB, Bullimore, MA. Assessing the reliability, discriminative ability, and validity of disability glare tests. Invest Ophthalmol Vis Sci. 1993;34:108–119.

53. Pesudovs, K, Hazel, CA, Doran, RM, Elliott, DB. The usefulness of Vistech and FACT contrast sensitivity charts for cataract and refractive surgery outcomes research. Br J Ophthalmol. 2004;88:11–16.

54. Kelly, SA, Pang, Y, Klemencic, S. Reliability of the CSV-1000 in adults and children. Optom Vision Sci. 2012;89:1172–1181.

55. Thayaparan, K, Crossland, MD, Rubin, GS. Clinical assessment of two new contrast sensitivity charts. Br J Ophthalmol. 2007;91:749–752.

56. Rabin, J, Wicks, J. Measuring resolution in the contrast domain–the small letter contrast test. Optom Vision Sci. 1996;73:398–403.

57. Elliott, DB, Situ, P. Visual acuity versus letter contrast sensitivity in early cataract. Vision Res. 1998;38:2047–2052.

58. van der Meulen IJ, Engelbrecht LA, van Vliet JM, et al. Straylight measurements in contact lens wear. Cornea 2010;29:516–22.

59. Rozema, JJ, Coeckelbergh, T, van den Berg, TJ, et al. Straylight before and after LASEK in myopia: changes in retinal straylight. Invest Ophthalmol Vis Sci. 2010;51:2800–2804.

60. Casparis, H, Lindsley, K, Kuo, IC, et al. Surgery for cataracts in people with age-related macular degeneration. Cochrane Database Syst Rev. (6):2012.

61. Chakravarthy, U, Wong, TY, Fletcher, A, et al. Clinical risk factors for age-related macular degeneration: a systematic review and meta-analysis. BMC Ophthalmol. 2010;10:31.

62. Wang, JJ, Fong, CS, Rochtchina, E, et al. Risk of age-related macular degeneration 3 years after cataract surgery: paired eye comparisons. Ophthalmology. 2012;119:2298–2303.

63. Schein, OD, Steinberg, EP, Javitt, JC, et al. Variation in cataract surgery practice and clinical outcomes. Ophthalmology. 1994;101:1142–1152.

64. Vianya-Estopa, M, Douthwaite, WA, Funnell, CL, Elliott, DB. Clinician versus potential acuity test predictions of visual outcome after cataract surgery. Optometry. 2009;80:447–453.

65. AHCPR report: Cataract Management Guideline Panel. Cataract in Adults: Management of Real World Impairment. Clinical Practice Guideline, Number 4. Rockville, MD. U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. AHCPR Pub. No. 93-0542. Feb. 1993.

66. Hofeldt, AJ, Weiss, MJ. Illuminated near card assessment of potential acuity in eyes with cataract. Ophthalmology. 1998;105:1531–1536.

67. Melki, SA, Safar, A, Martin, J, et al. Potential acuity pinhole–A simple method to measure potential visual acuity in patients with cataracts, comparison to potential acuity meter. Ophthalmology. 1999;106:1262–1267.

68. Margrain, TH, Thomson, D. Sources of variability in the clinical photostress test. Ophthalmic Physiol Opt. 2002;22:61–67.