The accurate assessment of visual fields has been a stalwart of vision assessment for decades. Along with pupil function, it offers information about the health or otherwise of the visual pathway, allowing practitioners to be able to interpret any field loss that respects the vertical meridian in terms of where within the brain a lesion might be presenting.
More useful still in everyday practice, the health of the retina may be better understood, with the anatomical representation of retinal nerve fibres being reflected in visual field loss whenever damage occurs. Lesions causing retinal damage will show as loss of the ability to see stimuli in the opposite visual space and so be recordable as field loss.
This is particularly important in disease with a slow insidious impact upon nerve fibre health, open angle glaucoma being the obvious candidate. Death of ganglion cells and their respective nerve fibres will eventually be discovered when developing paracentral field loss (around 10 to 15 degrees’ eccentricity from fixation), differences in stimulus sensitivity above and below the horizontal meridian in the nasal field (a nasal step) and eventual extension of loss along an arcuate pattern.
Harper and Reeves, in a well-known paper nearly 20 years back1 established that, when screening for glaucoma, a combination of central visual fields (they used a multiple stimulus program), optic disc analysis and tonometry was the best approach to offer the flexibility for screening the variable population and suggested fields assessment as the most robust of the three options.
Current market
In the UK, fields screening is ubiquitous in optometry practices with many using the multiple stimulus suprathreshold strategy offered by the Henson machines. The Humphrey Visual Field Analyser is usually cited as the ‘gold standard’ as it offers a range of single stimulus strategies including some that offer a faster (and therefore better specificity because ‘normal’ people are less likely to fail) full threshold (and therefore greater sensitivity) testing of areas of field most likely to be damaged in early glaucoma.
These SITA programs have analogous options in other instruments too, such as the ZATA algorithms in later Henson models and the TOP options in the Octopus (an instrument more commonly found in mainland Europe2). The Humphrey VFA uses an extensive normative database in helping it to predict whether results are suspicious and this has led to its adoption in most UK glaucoma clinics.
What to look for
I always suggest that a central fields assessment is appropriate for all patients at their first visit. Assuming most are normal, a suprathreshold strategy should be most appropriate, but where any risk factors exist (including being over 40 years old), then a full threshold approach would be best; ideally an adapted glaucoma-specific program such as SITA. So, a useful fields screener should have the following properties:
- Easy to use – can be pre-programmed to use an appropriate strategy or selection of strategies so that auxiliary staff (with appropriate training and supervision) might undertake the task.
- Appropriate ergonomic profile – minimal footprint is helpful.
- Wide range of strategies – Estermann for DVLA work, kinetic strategies, early glaucoma detection options (SITA algorithm or the like, possibly short wavelength strategies).
- Easy to interpret data output – the Humphrey style output showing reliability indices, threshold plots, mean and standard deviation plots are perhaps most familiar and helpful, and should include other statistical indicators such as mean deviation and glaucoma hemi field data.
- Easy data transfer, output and storage – all compliant with regulations on clinical data access and safety.
The Kowa AP-7000
I came across the Kowa-7000 fields analyser (figure 1) last year when visiting the UK supplier, Sense Medical. The latest version was recent to the UK and had some functionality that made me surprised that there was such little awareness of the machine over here. Last month I was loaned the instrument and put it through its paces.
The AP-7000 has a typical large bowl surface allowing kinetic isopter assessment and is operated via a touch screen (figure 2).
Figure 2
The user-friendliness is established when first switching on – while the machine is calibrating and adapting to ambient light levels, the screen tells you to ‘go and enjoy a coffee.’ Patient correction with a mean sphere is via an internal lens holder (figure 3) and the patient has fixation options including for when there is a central scotoma (figure 4).
Figure 3
Figure 4
Patient data input is simple, with options for importing a good deal of clinical information if required, and fixation monitoring and position adjustment to support the Heijl-Krakau fixation monitoring is helped by an internal camera (figure 5).
Figure 5
The machine boasts a wide range of assessment strategies (see table 1 for a comparison with the Humphrey and the Octopus) and includes speedier options for full threshold glaucoma assessment, analogous to DITA and ZATA, called Quick 1 and Quick 2. It also includes a short wavelength option for early selective ganglion cell damage.
Figure 6 shows a familiar data display after completion of a central threshold strategy and the reliability indices and statistical indicators are all present.
Figure 6
Output can be as a printed PDF (figure 7) or electronic transfer to an external database as required.
Figure 7
The internal computer has the capacity for storing data from 20,000 patients. Binocular summaries (figure 8) and sequential displays (figure 9) are all easy to specify. The machine can also produce predictive data plots.
Figure 8
Figure 9
With memories of cracking a patient on the back of the head with the swinging arm of the Goldmann kinetic perimeter in mind, I found the kinetic strategies easy to use (figure 10). It is possible to do colour isopters and, where a scotoma is found, the output can be easily enhanced with outlines and shading to highlight areas of loss or encroachment. Similarly, all programs can be customised, including the meridians for kinetic approach.
Figure 10
Structure and function
An interesting option for the AP-7000 is the ability to import a retinal image or an appropriate extent OCT or scanning laser ophthalmoscopy scan over which perimetric data can be superimposed. This may be as a JPEG image from a memory pen (figure 11a) or via a network.
Figure 11a
It can also receive direct input from a Kowa retinal camera. You then highlight the fovea (number 1 on figure 11b) and the disc (number 2).
Figure 11b
The instrument will then superimpose the selected threshold data pattern (figure 11c) or a combination of data outputs (figure 11d). I believe this ability to clearly represent physical structure with functional ability is helpful and is very easily done.
Figure 11c
Figure 11d
Useful addition
The AP-7000 works well, is easy to operate and programme, and meets the criteria I would expect from such an instrument. The normative database, often cited as a reason for staying with a particular instrument, is based on data acquisition from several thousand oriental eyes.
Though structural differences are known,3 the manufacturers are confident that the database is robust in predictive capability regarding field loss. As I have stated before,2 it pays to question any ‘gold standard’ which might be set for any length of time without due consideration of developments in flexibility and usability. I believe this new instrument is worth readers hearing about.
References
1 Harper RA, Reeves BC. Glaucoma screening: the importance of combining test data. Optom Vis Sci. 1999 Aug;76(8):537-43.
2 Harvey W. Finger on the pulse. Optician 31.01.2014, pp21-23.
3 Alasil, T et al. Analysis of Normal Retinal Nerve Fiber Layer Thickness by Age, Sex, and Race Using Spectral Domain Optical Coherence Tomography. Journal of Glaucoma: September 2013, Volume 22, Issue 7, pp 532-541.
Further information at www.sensemedical.co.uk