Where Are We Going

Using lasers to image the retina

BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7220.1303 (Published 13 November 1999) Cite this as: BMJ 1999;319:1303
  1. James E Morgan, senior lecturer (morganje3{at}cardiff.ac.uk)
  1. Cardiff Eye Unit, University Hospital of Wales, Cardiff CF4 4XW

    High quality ophthalmic care depends on the accurate assessment of ocular disease. Conventional direct ophthalmoscopy is used widely and provides good two dimensional views of the retina. However, the true nature of retinal disease is apparent only in three dimensions. In diabetic maculopathy, direct ophthalmoscopy can reveal retinal exudates, but the degree of macula oedema, which usually underlies the decision to treat by laser photocoagulation,1 is less clear. In glaucoma, the earliest damage can be seen as thinning of the retinal nerve fibre layer with increased cupping of the optic disc.2 These changes are best viewed stereoscopically. In both cases, disease of the retinal or optic nerve head will change the surface contour of the retina, either elevating or depressing the retinal surface by up to several hundred micrometres. Clearly, the objective quantification of these changes would be of immense benefit in diagnosing disease and monitoring disease progression and response to treatment.

    Summary points

    In diseases such as diabetic maculopathy and glaucoma changes in retinal structure precede visual symptoms

    Earlier detection of these changes allows early intervention and improves the prognosis

    Scanning laser ophthalmoscopes provide rapidly acquired views of the retina that enable the detection of these early changes

    Clinical studies have shown the value of these devices in the diagnosis of glaucoma and diabetic maculopathy

    The costs of these devices is falling and serious consideration should be given to their introduction into hospital based eye services

    Techniques for retinal imaging

    The three dimensional imaging of the retina has been facilitated by two technical developments. The first is the availability of cheap and powerful computers. The second is the production of affordable optoelectronics such as digital cameras and diode lasers. The incorporation of some of these technologies into clinical practice has been relatively straightforward and will be familiar to doctors in other specialties. For example, digitised photography of the retina has improved the detection of serious retinopathy in diabetes, 3 4 and when linked to telemedicine has allowed the rapid referral of patients for treatment. This type of imaging is a progression of existing practice in that most ophthalmology departments have been using conventional (non-digital) fundus cameras to photograph the retina for many years (fig 1).

    Fig 1.
    Fig 1.

    Digitised fundus photograph of optic disc with advanced glaucoma

    By contrast, scanning laser ophthalmoscopes 5 6 offer a radically different view of eye disease since they provide three dimensional views of the retina7; potentially, they represent an important step forward in the assessment of retinal disease. The devices use laser light rather than conventional white light to image the fundus. A low powered diode laser beam scans the retinal surface to build up an image of the retina line by line, analogous to the formation of images on a television screen.5 Several variants of this scanning technology exist, each providing slightly different views of retinal structure. In tomographic scanning laser ophthalmoscopes the focal plane is adjusted to generate a series of images that span the retinal thickness and can be used to derive a topographic plot of the shape of retinal surface or optic disc cup (fig 2).

    Fig 2.
    Fig 2.

    Tomogram of optic disc obtained with Heidelberg retina tomograph scanning laser ophthalmoscope. A cursor line has been placed across the optic disc to plot the retinal surface

    One drawback of this technique is that it gives only an indirect measure of changes within the retina.8 Other scanning laser devices have been developed to address this limitation. The scanning laser polarimeter measures the thickness of the retinal nerve fibre layer based on its birefringent properties (ability to polarise incident light).9 Plots of retinal nerve fibre layer birefringence approximate the variation in thickness of the layer around the optic disc10 (fig 3) and can be used to detect changes that are characteristic of glaucoma Perhaps the most dramatic development has been optical coherence tomography, in which interference patterns generated from the reflection of partially coherent laser light are used to construct an optical cross section of the retina (fig 4). 11 12

    Fig 3.
    Fig 3.

    Retardation image of normal optic disc taken with scanning laser polarimeter Brighter pixels correspond to regions of increased retardation (increased retinal nerve fibre layer thickness). The thickest parts of the retinal nerve fibre layer are at the superior and inferior borders of the optic disc

    Fig 4.
    Fig 4.

    Image from optical coherence tomograph. Arrows show approximate thickness of retinal nerve fibre layer. The image has been artificially colour coded

    Diagnosing and monitoring disease

    Several studies have examined the ability of these devices to detect glaucoma1315 and diabetic macula oedema.16 The tomographic scanning laser ophthalmoscope has been shown to detect glaucoma with high sensitivity (over 80%) and specificity (over 95%).14 These impressive statistics have been achieved without expert ophthalmic assessment and are based on the computerised analysis of retinal structure by the scanning laser ophthalmoscope. These ophthalmoscopes may therefore have a role in detection of glaucoma in primary care.

    This technology may also help detect progressive disease in patients who have already had glaucoma or diabetes diagnosed. The ability to detect structural change is an important advantage since these changes usually occur before the onset of clinically detectable visual deficits such as a reduction in visual acuity or loss of visual field. The delay between the structural and visual changes reflects the redundancy of neural components that is built into the visual system. Thus, in glaucoma, it has been estimated that up to 50% of the retinal ganglion cells at any particular location can be lost before a visual field defect is detectable using currently available clinical methods.17 Consequently, if we rely on tests of visual acuity or visual field, significant retinal damage may have already occurred by the time that disease is detected, leading to a poorer visual prognosis.

    Early detection also gives clinicians greater flexibility in managing patients. In glaucoma, quantification of the rate of optic disc cupping allows clinicians to estimate the onset of serious visual field loss, which can help when discussing the timing and possible outcomes of treatment. In diabetic maculopathy it may help improve targeting of focal laser treatment. These factors are important since the diseases have mild symptoms in the early stages, and treatment in the form of eye drops, laser, or surgery can have a greater effect on a patient's quality of life than the disease itself.

    The other major advantage of these imaging technologies is that they require little patient interaction. This contrasts with commonly used clinical tests such as automated perimetry, which can be arduous for some elderly patients. Furthermore, it is likely that fewer laser images will be needed than visual field tests to detect progression of disease since the noise in a scanning laser ophthalmoscope image is much less than that seen in perimetric tests. 1819 Finally, these devices provide important documentation of the retina and optic disc, which can be valuable when discussing the prognosis or considering the medicolegal aspects of a case.

    Implementation of laser technology

    The main barrier to the use of scanning laser ophthalmoscopes is that they are expensive. In addition to the initial capital costs, they require experienced staff to operate them and need specialist maintenance Taking images can, in some cases, be trying for the patient. The eye needs to be relatively immobile while the image is taken and, although the process is rapid for a single image (1.6 seconds for the scanning laser tomograph), three images are usually required to generate a clinically useful topographic map of the retinal surface. Indeed, a recent report of scanning laser ophthalmoscopy in an unselected patient population showed that up to 19% of patients could not provide satisfactory images.20 Patients may also be anxious about the new technology since lasers are often portrayed as powerful agents of destruction; their use as diagnostic tools requires careful explanation.

    Further evidence is required to justify the widespread clinical use of scanning laser ophthalmoscopes. Given that the role of the NHS is to deliver a uniform high standard of patient care, the evidence that scanning laser ophthalmoscopes help diagnose diseases such as glaucoma and diabetic maculopathy argues for their installation in most ophthalmic units. However, in many units in the United Kingdom the appearance of the optic disc and macula is still documented by hand, which provides a poor objective record of retinal disease. The introduction of simpler techniques such as stereoscopic optic disc photography, which uses existing fundus cameras, may therefore provide the best value for money since these images have been shown to be of value in distinguishing normal from glaucomatous eyes.21 Similarly, the institution of free eye tests for people aged over 60 is probably a more useful first step than the widespread introduction of scanning laser ophthalmoscopes.22 When so many basic steps have yet to be taken in clinical assessment it may be premature to consider the large scale introduction of such advanced imaging equipment.

    Despite these caveats, scanning laser ophthalmoscopy holds great promise for the diagnosis of ophthalmic disease, and it is important that we research its clinical application. In considering the benefits of this technology, we must not conclude that the relentless accrual of data always leads to improvements in patient care. Most patients want to spend as little time in clinic as possible and to receive the minimum necessary investigation and treatment. That said, if detailed topographic images can be acquired rapidly and with minimal discomfort, the wishes of both patient and clinician will be met. The cost of these devices remains a difficult issue. However, as with other electronic devices, this is likely to fall greatly over the next decade as development costs are recouped and computing and electronic costs are reduced. If these developments continue, the widespread use of laser imaging technology in routine clinical practice seems likely.

    Footnotes

    • Funding Welsh Office of Research and Development.

    • Competing interests None declared.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial

    Subscribe