Continuing Education
Eye essentials 2
Advanced instrumentation
Advances in technology have led to improvements in methods of measuring and assessing ocular structures. The increasing use of the laser has developed alongside microprocessor improvements allowing miniaturisation of hardware and greater software capability. Increasingly, much of the optometric assessment may be automated and digital encoding of data allows accurate storage, measurement and analysis. Dr Sandip Doshi and Bill Harvey provide a brief overview of the technology (CET module 1776)
SCANNING LASER OPHTHALMOSCOPY
The confocal scanning technique was introduced around 1980 by Petran and Boyde and has been adopted in many clinical and biological fields.
| Figure 1. The improved view of the peripheral fundus, in an undilated eye, has heightened the clincian's ability to detect and manage peripheral retinal disease |
The basic principle involves the production of a collimated polarised laser beam which is deflected stepwise across the two x and y directions by a scanning unit. The continually deflecting beam is then reflected by a dichroic mirror (the beam splitter) to pass through the objective lens, which focuses the light to the point at the level to be viewed.
Light emitted from the target reaches the beam splitter again, which remains transparent to the selected wavelength from the target point, while eliminating all out-of-focus light (from points above or below the target point). The xy deflection allows this to be done for all points in a plane at the established level and, by doing this for a variety of planes, a sequence of 'slices' through the object viewed may be imaged. The wavelength of the beam will be important in terms of its penetration of whatever is being viewed.
The main advantage of this system is that it can be used to carry out retinal investigations in ways that enable the ophthalmologist or optometrist to gain more information about the retina than can be obtained from conventional imaging techniques.
The Optos Panoramic 200 is designed to enable an area of up to 200 degrees of the fundus to be produced as one image and without the use of collaging or pupil dilation. The SLO tends to be superior to standard imaging methods when cataract is present and can be used for both fluorescein and indocyanine green angiography. The large field image formed by the Panoramic, combined with the unfamiliar colouring, often takes some experience to interpret. Indeed, the instrument initially led to an increase in referrals of eyes subsequently found to be within healthy bounds. However, the improved view of the peripheral fundus (Figure 1), remarkably in an undilated eye, has improved the clinician's ability to detect and manage peripheral retinal disease.
SCANNING LASER POLARIMETRY
By its simplest definition, a polarimeter is an instrument capable of measuring either the polarity (uni-directional property of light) or rotations to that polarity. As light passes through a structure, the nature of the change to the light gives information about the properties of that structure.
| Figure 2. Print-out from a GDx. Note the difference in the nerve fibre layer densities between the two eyes |
Layered structures may exhibit properties not found when the individual materials are assessed. An example of this is when waves traversing a medium, with their electric fields perpendicular to the medium, experience a different index of refraction than the waves parallel to the plane of the medium.
This difference is due to the varying boundary conditions imposed on the electric fields parallel and perpendicular to the interfaces between the different layers. This property is known as birefringence and forms the basis of the operation of the instrument known as the GDx (the GDx Access is the model used in modern practices).
The retinal nerve fibre layer shows birefringence of an incident laser and the influence that it has upon the incident confocal laser beam from the instrument is interpreted to infer the thickness of the layer. This results in the nerve fibre layer thickness profiles shown in Figure 2.
Because the assessment is made with measurement compared to a reference plane, the technique is not affected by refractive error.
As might be expected, one problem in the development of such a technique has been that birefringence occurs as light passes through the cornea and this must be compensated for if an accurate assessment of the nerve fibre layer is to be achieved. The GDx and GDx Access include such a compensator and this is usually set at a default appropriate for a standard cornea (an 80 per cent setting is typical). If the compensation is not accurate, this may be predicted by unusual data that is not borne out by other assessment.
Much has been published in recent years concerning the accuracy of the GDx unit and nerve fibre analysis. The fact that changes to the NFL may be a useful early indicator of glaucomatous change is well-established and many ophthalmologists have for years used bright or red-free light assessment of the peri-papillary area to look for deviations in the way the nerve fibres reflect light. A wedge-shaped area of less reflection would, for example, be treated as suspicious.
The use of NFL analysis as a tool in the effective screening of glaucoma has been found to have a high sensitivity and specificity. More recently, the use of data about the superior to nasal and the inferior to nasal NFL data ratios has been confirmed as effective and sensitive in differentiating between glaucomatous and non-glaucomatous eyes.
| Figure 3. The HRT II was introduced in 1999 |
Bearing this in mind, it seems as if the GDx provides a useful screening assessment of the NFL which makes it potentially valuable to an optometrist wanting a sensitive indicator of potential for glaucoma. It is also useful for monitoring glaucomatous NFL changes over a period of time. It does not, however, replace accurate disc analysis, nor does it remove the need for pressures and fields to be monitored.
SCANNING LASER TOMOGRAPHY
The accuracy of observation of an ocular structure, such as the optic disc, will be related to the skill and experience of the observer. Accurate photography of the disc allows a more accurate quantitative measurement of two-dimensional data, but the ability to accurately represent a three-dimensional model of the disc is yet more valuable.
An instrument which might somehow store an accurate three-dimensional representation of the disc architecture could usefully detect any parameters which are outside an expected norm and would also be able to detect accurately any changes to the disc over a period of time.
When the Heidelberg Retinal Tomograph was introduced in 1991, it was described as 'the first scanning laser system for routine glaucoma exam'. Its cost and bulky size proved prohibitive to eye care professionals in general practice, but as a hospital-based research tool able to analyse and store accurate topographic information it formed the basis of multiple research papers.
Topographical analysis using a confocal technique, of course, meant the HRT was useful not only in glaucoma research but also in the investigation of other structural change diseases such as macular holes, macular oedema, retinal detachments and neoplasms. There was also some application in the investigation of the atrophic changes in non-exudative macular degeneration.
| Figure 4. Reading taken from a diagnosed glaucoma patient |
In effect, the laser is focused and scanned rapidly across a common plane running through the optic nerve head. Elevation or depression of the optic disc relative to this flat plane is measured, so allowing an accurate topographic representation of the head to be stored. The software may then predict whether the topography is normal or suspicious of glaucoma, and sequential assessment of a disc over a time period may very sensitively detect changes in disc topography, as would occur with a chronic optic neuropathy.
In 1998 a team at Moorfields completed a study which aimed to define the HRT parameters that best allow it to separate patients with early glaucoma from normal subjects. A total of 131 patients were examined, 51 of whom had glaucoma leading to field loss (averaging a mean deviation of-3.6dB). Data was taken from the normals regarding the relationship between the neuroretinal rim area and the optic disc area, and the cup/disc ratio and the optic disc area. The normal ranges for these two parameters were defined mathematically (in terms of the 99 per cent prediction intervals of the linear regression between the parameter and the optic disc area) for the whole disc and for each of six predefined segments of disc. It was found that a high sensitivity (96.3 per cent) and specificity (84.3 per cent) in separating the normals from the glaucoma patients was achieved, and that the technique of linear regression provided a good separation method, the best figures being found when the log of the area of the neuroretinal rim was compared to the optic disc area.
A year later the same team published a study with the aim of trying to determine if an analysis of sequential optic disc images using the HRT would be able to demonstrate disc changes in ocular hypertensives before the development of reproducible field defects. They found that there was evidence of disc data recorded by the HRT, showing changes over a year period before any field loss was established. This suggested the machine might usefully play a role in glaucoma detection in an at-risk group prior to any field loss.
The HRT II was introduced in 1999 with the aim of providing the benefits of accurate laser scanning techniques in a more clinically friendly form and the instrument itself is similar to a small table-mounted slit-lamp unit with a separate computer and monitor (Figure 3). The basic unit itself can be dismantled easily and hence is readily portable.
| Figure 5. The Zeiss Stratus OCT |
Figure 4 shows a reading taken from a diagnosed glaucoma patient. The six red crosses clearly demonstrate that the disc profile fall outside the expected norm in all six sectors.
OPTICAL COHERENCE TOPOGRAPHY
As far back as 1878, Albert Michelson hypothesised that, if it were possible to split a beam of light into two parts and then transmit them along perpendicular paths, it might be possible upon receiving the returning beams to detect any differences in phase between them. This difference or interference would give valuable data about the surfaces from which the light has been reflected. This was the basis of the Michelson Interferometer, and is now the basis of the Zeiss Stratus OCT (Figure 5).
In this instrument a light beam is sent simultaneously to the eye and a reference mirror. The light penetrates through retinal tissues and is reflected back. The returning light is compared to the reference and this allows software to reconstruct a representation of the underlying
| Figure 6. A cross-section through the retina showing a macular hole |
ANTERIOR CONOFOCAL MICROSCOPY
Confocal microscopy has been a familiar technique in university and hospital research departments, allowing a practitioner to carry out cell counts, and monitor minor structural changes.
Such work on the endothelium is the basis of much of our knowledge about how this structure changes with age or with external insult (such as hypoxic stress) and much significant knowledge has been gained. Other research, such as looking at the effects of topical therapeutic agents, contact lenses and developmental variations may all be assessed in vivo with a significant advantage over in vitro techniques.
Like many other areas of optometric technology, the advances in refractive surgery have led to an increased demand for good resolution yet easy to achieve confocal corneal images. One of the best indications of postoperative corneal recovery (either for a photorefractive treatment such as Lasik or phototherapeutic intervention such as graft surgery) is to look for changes to cellular structure within the cornea, or to visualise debris build-up. For example, a high resolution confocal view of the stroma post-Lasik may allow changes to the keratocytes to be detected, as well as increases in cell debris or corneal nerve changes.
Doing this at intervals after surgery gives an excellent indication of corneal recovery and is an integral part of most refractive surgeons' follow-up routine. The technique also has a more non-iatrogenic use, however, in the detection and monitoring of corneal disease. For example, the Langerhans cell population in the plane of the basal cell layer near the limbus is useful in assessing the keratoconic, particularly after a full thickness graft.
Cellular changes and guttata may be detected at an early stage in Fuch's dystrophy. Microcystic response, either related to hypoxic insult or keratitis, may be monitored with much greater accuracy than a slit-lamp would afford. It is even possible to analyse erythrocyte flow in a neovascular response.
| Figure 7. Conofocal adapter |
Recently, an adapter to the HRT II (see earlier and Figure 7) allows the confocal laser previously focused in the plane of the optic disc to instead be focused at the cornea, so providing a useful image of any plane throughout the cornea.
TOPOGRAPHERS
In 1847 the English physician Henry Goode used the reflection of a square target from the cornea to decide upon corneal shape in what is claimed to be the first keratoscope. Independently, in 1880 the Portuguese oculist Antonio Placido introduced the now-famous black and white concentric ring target with a viewing hole to analyse the distortion of the reflected rings from the cornea; the first photokeratoscope.
It is a variation of the Placido's disc which is the basis of most modern computerised topographers. At this stage, assessment of corneal contour was primarily qualitative, with subjective assessment being made of distortion of the reflected image.
The introduction of computer analysis of the reflected targets began in earnest in the late 1980s and allowed the development of the quantification of reflected image changes. By sampling changes to the image at many points, a very accurate profile of the corneal contour may be constructed, and then represented in a variety of representations or 'maps'.
| Figure 8. Maps are usually colour-coded; red areas show where the cornea is steepest and blue areas where it is flattest |
The various maps are usually colour coded; red areas show where the cornea is steepest and blue areas where it is flatter (Figure 8).
Difference maps may allow a change in profile, for example after contact lens wear, to be shown.
More recent topographers, such as the Orbscan, use a slit-beam scanning device which allows not only the anterior corneal surface profile to be measured, but also the posterior surface. Furthermore, topographers may also represent the tear profile underneath a contact lens, so providing a useful alternative to the far more subjective fluorescein assessment.
The increasing use of topographers in contact lens practice has led to some software to generate the ideal contact lens parameter for corneal fitting. It is also the basis of the resurgence in popularity of orthokeratology, where the deliberate remoulding of the cornea by selective lens fitting to overcome refractive error, where vastly improved accuracy and predictability has been made possible.
The increased use of topographers as part of the preoperative assessment before refractive surgery has also resulted in increasing numbers of early corneal dystrophies, such as keratoconus, to be detected. This shows that the machines have a diagnostic as well as biometric role.
ABERROMETRY
When describing the very great amount of aberration that affects light passing through the human eye, Helmholtz once remarked: 'If an optician wanted to sell me an instrument which had all these defectsÉI should give him back his instrument'.
Refractive correction may compensate for the so-called lower-order aberrations, but the final image is reduced by an array of higher-order aberrations. The impact of these may be measured using an aberrometer. This relies upon an analysis of how a wavefront passing through the eye is altered. If the eye is aberration free, then an incident plane wave will be refracted to form a regular wave. The eye, however, is not aberration-free and hence the resultant wave is distorted. It is possible to measure the change from the regular wave at many points so recording wavefront aberration.
Most aberrometers use a system called the Hartmann-Shack. This employs an array of tiny lenses (lenslets) through which the aberrant wavefront passes. Each lenslet forms a point image which would be a perfect point for each were the wavefront to be aberration-free. However, the point spread at each point focus reveals the overall aberration. The accurate analysis of this gives useful information which may be used, for example, in programming the laser used in refractive surgery, such that the postoperative corneal profile has less aberrations, or in specifying contact lens design with minimal aberration for the cornea to which it is to be fitted.
 | Providing exclusive eye care news, information and educational needs every week, including a FREE CET programme. Subscribe to Optician Print Edition. |
