Printer Friendly

Visual fields: where are we now?


Visual field evaluation is an important component of an optometric examination and can be used to identify both ocular and neurological disorders. This article discusses the different strategies for assessing the visual field and considers its role in the future as imaging technologies come to the fore.


Standard automated perimetry (SAP) is a well-established clinical test used in both optometric practice and the hospital eye service to aid the identification of a wide range of ocular and neurological disorders. It generally refers to the measurement of sensitivity to static circular stimuli of 0.43 degrees diameter (Goldmann III) and 200ms duration in a grid pattern across the central visual field. A thresholding algorithm, such as the Swedish Interactive Thresholding Algorithm (SITA) is generally used in the measurement of sensitivity. Sequential examinations are used to follow the course of various conditions and to monitor the effect of treatment. Much of its use is to aid the detection and management of glaucoma and this will be the focus of this article.


In 1873, an article in the British Medical Journal by Jeaffreson described a new form of perimeter that would not only provide accurate measurements, but do so in a rapid manner. (1) It consisted of a gas lamp and mirror, which projected a fixation point and moving target onto a white, hemispherical background at a test distance of 30cm. Many similar features, albeit more controlled, can be found in the perimeters of today. Until the introduction of the Goldmann perimeter (a manual, kinetic perimeter) in 1945, there was much dispute over the best technique to use to identify visual field loss, (2-5) and the lack of standardisation made it difficult to distinguish 'abnormal' from 'normal'. One of the reasons the Goldmann perimeter is credited as being one of the most important contributions to modern perimetry is that it achieved a level of standardisation. (5) Many of the parameters used in Goldmann perimetry, such as the standard stimulus sizes (Goldmann I-V), background luminance, and test distance, have been carried over into several modern, automated instruments.

Static perimetry, rather than kinetic perimetry, has since become the accepted standard, and the introduction of the automatic perimeter brought less reliance on the perimetrist for reliable results. (6) It is worth noting that static techniques were employed not because they were thought to be superior to kinetic techniques, but rather because technical limitations at the time made it difficult to successfully manufacture an automated kinetic perimeter. (5,7,8) During the age of automated static perimetry, research has focussed on the development of more accurate stimuli for the detection and monitoring of disease, suprathreshold techniques for disease screening, as well as algorithms for determining sensitivity with greater speed and efficiency. SAP is perhaps most commonly recognised on the Humphrey Field Analyzer (HFA, Carl Zeiss Meditec, Dublin, CA). It is on this instrument that the majority of research into stimulus sensitivity and variability has been performed, but the same stimuli are also used on other instruments such as the Octopus perimeters (Haag-Streit Ltd., Koeniz, Switzerland) (see Figure 1), Henson perimeters (Tinsley Ophthalmic, Redhill, Surrey), and Kowa perimeters (Kowa Ltd., Tokyo, Japan).

Recent reports have called into question the scientific basis for using the stimulus parameters currently employed in SAP, however. (9-11) While some investigations indicated that the Goldmann III stimulus was more resistant to the effects of optical blur than the Goldmann I stimulus commonly used in Goldmann perimetry, (12,13) there was little scientific or clinical reason for incorporating this parameter into SAP during its development. (9,11)


Visual field testing is an important part of an eye examination and is an essential test in the investigation of glaucoma. Indeed, while tonometry and optic nerve assessment are useful for assessing the risk of glaucoma (and/or progression) and detecting optic nerve damage, it is the visual field test that tells the clinician how well the patient sees; perhaps the main concern of a patient undergoing investigation. Despite its widespread use and obvious utility, it has some important limitations. First, it has been reported to have poor sensitivity to early glaucomatous damage. (14) Second, there is high variability in the measurement from one visit to the next when damage is moderate (see Figure 2), (15) such that identification of true progression is difficult, and third, it has an inadequate measurement range for testing patients with severe sight impairment. (16) Several studies have shown that variability with SAP (both within and between visits) increases with depth of defect; (15,17-19) this has the effect of making it more difficult to ascertain whether or not progression has occurred, once glaucoma has been detected. In order to monitor the field of patients with advanced disease, a large range of stimulus contrast (or brightness) is required. With SAP, OdB is reported either when a patient has no vision in that region of the visual field, or when the instrument cannot produce a stimulus with high enough contrast to elicit a 'seen' response. Therefore, in reality, it might be that the patient has some residual, immeasurable vision in that region of the field.


Research into the most effective means of visual field assessment has been ongoing for many years, with approaches from many different angles. Theories were proposed in the mid-1990s that particular subsets of retinal ganglion cells were preferentially damaged in glaucoma, (20-22) along with the idea that glaucoma might be detected earlier if it were possible to isolate and measure their function. As such, different clinical techniques were developed, in an attempt to isolate and test specific visual pathways. Perhaps the two most well known of these techniques are short-wavelength automated perimetry (SWAP) and frequency-doubling technology (FDT).

Short-wavelength automated perimetry (SWAP)

Reports of impaired colour vision in patients with early glaucoma and ocular hypertension, (23,24) led to the exploration of colour (specifically short-wavelength) perimetry, (25) in the hope that this would lead to earlier detection of glaucoma. SWAP was designed to preferentially stimulate the retinal ganglion cells that relay signals from blue cone bipolar cells to the lateral geniculate nucleus (LGN), by using a bright yellow background to depress the sensitivity of the long- and medium -wavelength-sensitive cones, and a blue target to stimulate the short-wavelength sensitive cones. (25,26) Early studies reported that SWAP could detect glaucomatous damage earlier than SAP. (25-27-28) However, a difficulty in comparing SWAP and SAP is that the decibel scales are not equivalent. For example, 30dB on SWAP is not equivalent to 30dB on SAP. Studies have attempted to overcome this by identifying the point at which sensitivity falls outside normal limits; (29) the more sensitive test being that which shows sensitivity falling outside normal limits earliest. A problem here is that there is high within-test 29, 30 and between-test (30) variability in SWAP, in glaucoma patients and in normals. Also, age-related lens yellowing causes absorption of blue light, meaning that higher target luminance is required to be seen by older patients (i.e. an apparently lower sensitivity). The normal range spans a greater proportion of the dynamic range of SWAP than the equivalent in SAP, making identification of glaucoma and its progression more difficult. This, together with the extended test time (including time taken to adapt to the yellow light) limited the utility of the technique in clinical practice. Although this method of preferentially stimulating the short-wavelength-sensitive cones is a classic method used in elegant studies of the physiology of vision since the 1940s, (31) its translation to clinical practice has been less successful.

Frequency-doubling technology (FDT)

Histological studies in the 1980s reported that larger diameter optic nerve fibres were selectively damaged in glaucoma. (21,32) It was further hypothesised that these were the retinal ganglion cells that formed part of the magnocellular pathway. (22) This led to the development of perimetry that attempted to preferentially stimulate and test the sensitivity of this pathway. (33)

'Frequency doubling' is the term given to the illusion of a doubling in spatial frequency of a low spatial frequency grating undergoing high temporal counterphase flicker over a certain frequency range. (34) These frequency characteristics were believed to be specifically identified by the magnocellular pathway. (35)

Although initial experiments attempted to ensure the frequency doubling illusion had been identified by observers, (35) this was not the case in later adaptations of the technique, which would become FDT as we know it. (33) As such, the term 'frequency doubling technology' is a misnomer, as the test is not based on the illusion, nor is visualisation of the illusion part of the task of the patient; FDT attempts to determine the contrast threshold for a grating of a fixed spatial frequency and fixed rate of flicker.

The decibel scale in FDT is not equivalent to that in SAP or SWAP, so results cannot be easily compared between these tests. However, various analytical methods have been used to overcome this problem in order to investigate which test identifies more cases of early glaucoma. (36-38) One study calculated a 'signal/noise' ratio (a ratio of hemifield differences in total deviation to between-test variability) for both FDT2 (the second generation of FDT instruments) and SAP. (38) As instrument-specific units apply to both the disease signal and the noise, these are cancelled out, so that tests can then be directly compared. It was noted that the overall signal/noise ratio was higher with FDT2, suggesting that this technique is at least as sensitive as SAP in identifying visual field damage. Another study found between-test variability to be more uniform across the sensitivity range for FDT2 compared with that for SAP. (18) These findings suggested that FDT2 also held promise for accurately identifying true disease progression. A subsequent study used an analysis technique known as 'Permutation of Pointwise Linear Regression' (PoPLR) to find the number of instances that FDT2 and SAP identified true overall visual field deterioration, (39) while overcoming the problem of non-equivalent units and reliance on normative databases (see later section). The study found that FDT2 identified deterioration in fewer patients than SAP when comparing total deviation. As such, it was concluded that FDT2 did not have notable advantages over SAP when investigating progression.

More recent evidence suggests that the Goldmann size III stimulus used in SAP is actually superior to the FDT stimulus in preferentially stimulating the magnocellular pathway. (40) Furthermore, a histological study on monkeys indicated a shrinkage of retinal ganglion cells (before cell death), (41) which may help to explain the earlier observation of a preferential loss of larger optic nerve fibres in early glaucoma.

FDT is quick, (33-42) relatively inexpensive and fairly robust to optical blur, (43) though it has also been found to be less robust to reduced retinal illuminance, (44) a limitation when examining those with pupil miosis or cataract.


Reports of substantial retinal ganglion cell loss before identifiable visual field loss in glaucoma spurred a great deal of research into alternative diagnostic techniques for the condition, (45) in the form of functional tests (as described above) and measurements of retinal structure. However, it is worth noting that in one of the most widely cited of these studies, (45) it was sensitivity from kinetic manual perimetry, rather than SAP, that was compared with histological counts of retinal ganglion cell axons.

Histological studies report large variations in the number of retinal nerve fibres, even in healthy eyes, (21,32,45-47) making it difficult to identify an abnormality. A study conducted in 2013 measured peripapillary retinal nerve fibre layer (RNFL) thickness with OCT and found a mean RNFL thickness of 114.8pm (SD: 13.3pm) in their healthy participants. (48) Using these figures in a schematic (see Figure 3), one can see that if a patient had a baseline RNFL thickness of 141.4pm (2 SD above the mean), and over a given time period has a reduction in thickness of 30%, they would still have a value considered to be 'within normal limits' (99pm). However, if the patient had a thinner RNFL thickness at baseline, the same amount of loss would be flagged sooner. Of course, much of this problem can also be attributed to the use of normative databases, and can be equally applied to functional tests such as SAP. It is logical to expect that functional tests should be able to detect glaucomatous damage before tests of retinal structure; after all, it is not possible to have function without neurons, but it is possible to have the remains of neurons without function. As previously stated, more recent histological studies have indicated a general ganglion cell shrinkage before ganglion cells are lost. (49)

There are, of course, distinct advantages to using imaging techniques; mainly that these techniques are objective, and therefore, not reliant on patient responses. As such, it is very tempting to expect that imaging is a good substitute for visual field testing. However, it should be borne in mind that the main reason for patients visiting an optometrist's practice and for subsequent investigations into possible glaucoma is out of concern for sight, not their retinal thickness nor the number of retinal ganglion cells they have. The reason for treating glaucoma is to prevent sight loss, therefore it is important to be able to accurately and precisely measure what a patient can see. Imaging is undoubtedly useful in that it provides extra information to aid clinical investigation, but it should be used in combination with measurements of visual function, rather than a substitute for them. (50)

At present, NICE guidelines specify SAP should be used for both diagnosing and monitoring glaucoma, and name the 24-2 SITA standard program on the Humphrey Field Analyzer as being the 'most accurate'.50 Of HRT and OCT, the guidelines state 'there may be a role for these technologies in detection of progressive change through sequential monitoring but evidence is as yet inadequate to support a recommendation in this regard'. Optometrists are expected to follow these guidelines, while research into more accurate and precise investigations are ongoing; replacing SAP with other forms of perimetry or imaging techniques may not bring significant benefits to clinical practice. If these tests are to be used, they should be used in conjunction with SAP, rather than in replacement of it.


Current research into efficient, accurate and precise methods of detecting and following visual field damage is concerned mainly with establishing the most appropriate testing strategies (including threshold/suprathreshold algorithms) stimuli, and techniques for analysis of clinical findings.

Motion perimetry, for example, the Moorfields Motion Displacement Test, (51) is one such technique currently undergoing clinical development. This technique, performed on a laptop computer, assesses the visual field with moving line stimuli. Early reports are that it has good diagnostic performance, (52) and that the suprathreshold screening strategy enables identification of more widespread visual field damage than SAP. (53)

Spatial summation describes the way in which the visual system 'adds up' the amount of light energy across the area of a stimulus. (9) Greater spatial summation has been found in early glaucoma, (10) suggesting that more consideration should be given to the area of the stimulus used in SAP. Some studies have suggested that varying stimulus area during a SAP test may boost the disease signal. (9,10)

Classic staircase algorithms were originally employed in SAP, but tests were time-consuming and tiring for patients, often leading to unreliable results. The past 20 years has seen a growing interest in finding alternative, more efficient algorithms, which would enable the clinician to make accurate and precise measurements within a shorter time frame. Arguably the most well known of these are SITA, employed on the Humphrey Field Analyzer, and the Zippy Estimation by Sequential Testing (ZEST), used in FDT2. Newer algorithms currently undergoing research include those that give more attention to the edges of scotomata, for example the Gradient-Oriented Automated Natural Neighbour Approach (GOANNA), (54) and those that make use of measurements from previous visits in an attempt to reduce variability. (55)

Clinicians are used to combining clinical data from multiple sources in their investigation of glaucoma. A decision is usually reached through clinical experience, knowledge and intuition. Automated analysis techniques are being developed that combine information from different sources such as RNFL thickness and visual field sensitivity, in order to ascertain the likelihood of true damage or deterioration. A recent study using Bayesian linear regression on visual field sensitivity and neuroretinal rim area of patients with ocular hypertension has shown that a combination of measurements provides a more accurate indication of the rate of deterioration than visual field sensitivity alone. Sophisticated techniques for identifying visual field deterioration over time are also being developed. (56,57)


It is clear that visual field tests serve an important purpose in clinical practice. Alternative forms of commercially available perimetry should be used in conjunction with SAP, but not replace it. In the absence of more accurate and precise techniques, it is important to perform as many visual field tests as possible on each individual with glaucoma or at risk of the disease, in order to increase the likelihood of identifying true deterioration. While it might be perceived that visual field testing hasn't evolved much in recent years, current research into improved measurement and analysis techniques is highly promising.

Course: code: C-41835 Deadline: September 26, 2015


To be able to explain to patients about the importance of visual field testing (Group 1.2.4)

To understand the different strategies for assessing visual field integrity (Group 3.1.5)

To understand the different approaches to visual field testing for the detection and monitoring of glaucoma (Group 6.1.8)


To be able to explain to patients about the importance of visual field testing (Group 1.2.4)

To understand the different strategies for assessing visual field integrity (Group 3.1.5)


To understand the most appropriate visual field test strategies for identifying glaucoma (Group 2.1.5)


Having completed this CET exam, consider whether you feel more confident in your clinical skills--how will you change the way you practise? How will you use this information to improve your work for patient benefit?


Lindsay Rountree is currently a PhD student at Cardiff University researching the use of stimulus parameters in perimetric testing. She is an optometrist with experience working in practice and within community ophthalmology teams.

Dr Tony Redmond is a senior lecturer at the School of Optometry and Vision Sciences at Cardiff University. His research interests include visual psychophysics and perimetry.

Exam questions

Under the enhanced CET rules of the GOC, MCQs for this exam appear online at Please complete online by midnight on September 26, 2015. You will be unable to submit exams after this date. Answers will be published on and CET points will be uploaded to the GOC every two weeks. You will then need to log into your CET portfolio by clicking on 'MyGOC' on the GOC website ( to confirm your points.


Visit, click on the article title and then on 'references' to download
COPYRIGHT 2015 Ten Alps Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:CET: VISUAL FIELDS; research on standard automated perimetry
Author:Rountree, Lindsay; Redmond, Tony
Publication:Optometry Today
Article Type:Report
Date:Aug 29, 2015
Previous Article:Keratoconus: signs, symptoms and management.
Next Article:Welcome.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters