The Optic Nerve Head Office Interpretation

GR Douglas
Professor Emeritus, Victoria, British Columbia, Canada

Introduction

Examination of the optic nerve head (ONH) is one of the most neglected skills in ophthalmology and general medicine, mainly due to time constraints in the clinic and poor acquaintance with up to date knowledge regarding this familiar bit of anatomy. Epidemiology has taught us that ONH assessment is superior to tonometry and perimetry for diagnosis and follow-up of chronic glaucoma when all 3 methods are used individually. That is not to say that any of these techniques should be eliminated, rather to emphasise the importance of a detailed but efficient ONH examination. The purpose of this article is to present an examination paradigm to clinicians so that early pathology, as well as progressive changes, may be recognised using available tools and requiring only slightly more time.

Since most practitioners in the world will not have reasonable access to the new ONH analysers, it remains the responsibility of each of us to become acquainted with simple, cost-effective, diagnostic methods that are readily available. Even with the inevitable decrease in price of these machines over time, costs will remain too high for all but the most affluent situations. With practice, the concepts and suggestions found in this article can be applied with minimum equipment and time expenditure.

As mentioned, if only a single examination modality is used, most glaucomatologists would choose ONH assessment rather than the other 2 options. Intraocular pressure (IOP) monitoring and perimetry both have a role when available. However, in many settings, this is not possible thus the need for an article such as this. Tonometry and ophthalmoscopy are often the only practical means of detecting and monitoring chronic glaucoma and, when properly used, can be most efficient for the clinic setting.

The challenge presented here is to acquire and apply new understanding and correlations amongst various parameters of the ONH. This increase in diagnostic acuity will help clinical knowledge to evolve beyond the simplistic and inadequate cup/ disc (C/D) ratio approach, which often prevails today.

Methods of Examination


Direct Ophthalmoscopy
Ophthalmoscopy remains the instrument of choice in most countries, if for no other reason than economy. With this instrument, we can vary the illumination, aperture size, image shape, and access red-free light. For our purposes, the small light is used for undilated pupils and the large one for dilated pupils. These, plus the red-free light that enhances our view of the retinal nerve fibre layer (RNFL) are the most useful options. An overly bright light may be disadvantageous both for the patient (photosensitivity) and the examiner (miosis), therefore, this should be adjusted for optimal viewing. The ophthalmoscope offers portability, magnification, and economy but is marred by a narrow field of view and a marginal appreciation of any topographical change in the neuroretinal rim (NRR).

Indirect Ophthalmoscopy
Indirect ophthalmoscopy offers a wide-angle stereoscopic image to the examiner but, depending on the condensing lens used, magnification may not be adequate to see sufficient detail on the ONH. The main problems with this apparatus are magnification, expense, and the need to dilate all but a few eyes.

Condensing Lens/Slit Lamp
A 60 to 90 D lens in front of the eye, used with the slit lamp, also gives good quality stereoscopic images of the ONH. This method requires dilation and, of course, an additional lens.

Contact Lens
Examination of the ONH with the central lens of an indirect Goldmann-style gonioscope provides magnification, stereopsis and, when coupled with the scale on the adjustable vertical light beam of a slit lamp, a means of quantitating the size of the ONH and other fundal findings. Although not absolute units, they may be noted for future changes of cup and NRR width if the same equipment is used. This method requires dilation, a slit lamp, and the lens.

Summary
If ease of use, economy, and availability are important factors, the ophthalmoscope is the clear winner for method. A Goldmann-style gonioscope, with its central lens, also has a definite advantage, especially in Asia where angle closure glaucoma is so prevalent. In larger or better-equipped clinics, alternatives may be present and should be used to supplement the ophthalmoscope.

Optic Nerve Head Parameters

Optic Nerve Head Size
An important but often confusing point is where the outer limits of the disc are located. The scleral rim is commonly seen as a partial or total white circle peripheral to the normally pink NRR. Outside this rim is the peripapillary zone and inside is the ONH (Figure 1). It is now well established that the ONH (disc) area can vary by up to 7-fold amongst individuals.¹ Due to this factor, if the C/D ratio is used and the disc area is not properly interpreted, an examiner may easily over- or under-interpret the status of the NRR. Since the neuroretinal rim area (NRA) appears to be the most important clinical parameter for clinicians when looking for and monitoring glaucoma, estimation of the disc size becomes an essential part of ONH examination (Figures 1 and 2).^2-4 Of all the parameters of the ONH, NRA appears to be most highly correlated with damage.⁵ Even on the Heidelberg Retinal Tomograph (HRT), NRA is second in importance only to a computer-derived term called cup-shape measure.⁶ It has also been determined that the size of the cup is directly related to that of the disc.³ Therefore, small discs with small cups may hide damage whereas large discs with large cups may cause concern and lead to initiation of unnecessary treatment.

Figure 1. Normal optic nerve head scleral rim (arrows), neuroretinal rim thickness (open arrows), retinal nerve fibre layer seen as radiating lines against the darker choroid, and zone alpha is seen temporally.	Figure 2. Small optic nerve head compared with Figure 1 for relative size. The blood vessels appear larger relative to the disc.

Having shown the importance of disc size, how is it estimated? Clinicians may misinterpret the size of an ONH because of the optics of the eye. Littmann provided a formula by which we can compensate for the effects of optics,⁷ although the next section helps to retain this article's concept of simplicity. The image on many ophthalmoscopes subtends an angle of approximately 5° which is also the size of the 'average' disc. To confirm the size of a particular ophthalmoscopic image, shine the image onto a wall from 1 meter away.⁸ The diameter should be approximately 85 to 90 mm. Now, by observing the relative size between this image positioned beside the ONH and that of the ONH disc, the size can be estimated. A large disc will be much larger than the image while a small disc will be much smaller. Recorded notes of discs can then be more meaningful, as in small, average, or large, especially if the C/D ratio is also noted. Using this method, disc size asymmetry can also be discovered to explain cup asymmetry. This estimating technique is useful for all eyes except those in which extreme ametropia exists.

Neuroretinal Rim
As for the disc, rim margins may be difficult to identify. Colour change is often used to distinguish between the cup and the rim but this works only for normal ONHs, where contours (topography) and colour are often coincident. In looking for disease, it is important not to utilise a criterion that is typical only for the normal state. A better method is to observe the change in direction of blood vessels as they cross the NRR (Figures 1 and 3). This allows the clinician to detect a change in NRR topography when 'atrophy' s not yet apparent. In more obvious cases, the 'laminar dots' of the lamina cribrosa or extension of the central cup pallor may help as they are seen to be closer to the disc margin in one area rather than another (NRR thinning).

Relative Thickness
Relative thickness of the NRR varies in a fairly predictable pattern amongst healthy individuals.⁹ Due to the macular fibres, the inferior rim is the widest, followed by the superior rim. These are followed by the nasal and, finally, the temporal rim, which is the thinnest (Figure 1). In subtle cases, for example, even equality of the rim widths superiorly and inferiorly may indicate pathology. As already mentioned, in glaucoma, the NRR is the most important part of the ONH. Clinicians would gain more information looking at the NRR than what is not there the cup! Although it is not possible to routinely estimate the rim area, it can be observed and the relative thickness or width of it recorded. Generalised or localised thinning should be noted and correlated with perimetric findings whenever possible (Figures 3 and 4). If a correlation cannot be found, other aetiologies for abnormalities (retinal, neurological) must be considered. The crucial point in NRR examination is to concentrate on the topography of the NRR and disc and not the central pale area.

Figure 3. Pallor versus neuroretinal rim change the inferior rim is damaged as shown by a change in blood vessel direction close to the scleral rim if only pallor had been used to judge the cup size, the true size would have been missed. Note also the retinal nerve fibre layer between 4.30 and 5.30 o'clock on the retina as a darker area (arrows).

Figure 4. Generalised enlargement and notch formation blood vessels change direction directly over the scleral rim from 5 to 6 o'clock, indicating an absolute loss of neuroretinal rim. There is also a narrowing of the superior neuroretinal rim. Note the loss of retinal nerve fibre layer superiorly and inferiorly. Reflections of light, parallel to some blood vessels in these 2 areas, show the inner limiting membrane which is now draped over them rather than being flat on a healthy retinal nerve fibre layer.

Careful observers are already aware of slopes of the NRR rather than the sharply-demarcated rim edges implied above. These are more commonly seen in highly myopic discs but, a gradual slope away from the neuroretinal surface can be found in many eyes, both normal and diseased. A problem arises in trying to identify any margin under these circumstances. This is where stereoscopy (contact lens) can be of help. The term 'saucerisation' has been coined for this situation, to cover an inability to quantify or to adequately describe this appearance. In the absence of a contact lens or a condensing lens, parallax may be used to estimate relative depths, that is, horizontal or vertical movements within the pupil observing the relative 'with and against' movements of the retina and cup depths.

Pallor
Pallor of the NRR is difficult to identify until it is well-established, is asymmetrical, or is found only in a localised segment of the rim. There have been attempts to quantitate pallor but clinicians must use their judgement when making this finding. However, segmental pallor is often found following branch vascular occlusions or chorioretinal breaks following trauma, both of which can be the cause of nerve fibre bundle defects typical of glaucoma. This is not to be confused with glaucomatous loss of rim in which the pale deep tissue of the ONH is now exposed.

Translucency
Translucency of the NRR has also been described in glaucoma. This term is often used to describe transparent tissue below blood vessels that has not been displaced backward into a notch or thinning of the rim. Translucency may be suspected but its validity lies mostly in correlation with perimetric findings.

Haemorrhages
Haemorrhages (Figure 5) are most commonly found crossing the NRR (splinter) but they may also be found in the rim or disc substance (splinter or diffuse). They are elusive and are rarely found unless sought. First recognised by Donders in the 1860s, they were 'lost' by generations of ophthalmologists until rediscovered by Drance and Begg in 1970.¹⁰ Their typical flame shape is easily recognised, although many have been missed because of their resemblance to small blood vessels. They may also be large and have to be distinguished from vascular events in the retina. They resorb within 6 to 8 weeks and, in a significant number of patients, leave a perimetric nerve fibre bundle defect or a retinal nerve fibre layer defect.¹¹ For this reason, finding one or even multiple haemorrhages is considered by many to represent glaucomatous progression.¹¹ The ophthalmoscope is ideal for finding a haemorrhage because of its magnification. It is now known that, although more commonly found in low tension glaucoma, haemorrhages may be found in any chronic form of the disease.¹²

Parameters Outside the Optic Nerve Head

Chorioretinal Atrophy
Chorioretinal atrophy has now been firmly associated with changes in the ONH as well as age and the refractive state of the eye. Jonas et al have written extensively on this finding, calling the outer area zone alpha (irregular pigmentation derived from the retinal pigment cells) and the inner area zone beta (choroidal vessels and sclera).¹³ Zone alpha is often seen in healthy eyes, whereas zone beta is correlated with glaucoma. Studies have shown that localised atrophy may be associated with localised changes at the NRR a good clue when clinicians are concerned about an area of thinning adjacent to the atrophy. This can also be found in older, sclerotic ONHs and in high myopia separate from any glaucomatous process. The main value is found in cases of localised atrophy in eyes with a suspicious notch or loss of NRR. There are also some theories on why such findings occur in glaucoma. Observations made by Rader et al suggest that vasoconstriction of retinal arterioles as they cross the NRR may indicate a potential aetiology for glaucomatous chorioretinal atrophy.¹⁴

Retinal Nerve Fibre Layer Defects
Retinal nerve fibre layer defects (Figures 1, 3, 4, and 5) have been recognised for many years but have only been associated with glaucoma since the early 1980s.^15,16 Red-free light is useful for detection and dilation aids the clinician regardless of whether the ophthalmoscope or slit lamp is used. Defects may be discrete or diffuse but the latter are much harder to find. Figures 1 and 6 illustrate the normal pattern in which the fibres associated with superior and inferior vascular bundles meld into the less obvious nasal and temporal quadrants. As with haemorrhages, the observer must appreciate the normal pattern before expecting to 'see' the abnormal. Figures 3, 4, and 5 demonstrate localised defects, all of which are associated with localised NRR defects. When recognising these RNFL defects for the first time they are always seen because of associated notches. With practice, clinicians will be able to recognise some of the RNFL abnormalities first and either verify the presence of a subtle NRR change or recognise glaucomatous pathology prior to recognisable perimetric alteration.

Figure 5. Optic nerve head haemorrhages this large disc has a large, patho-logical cup plus a small haemorrhage on the neuroretinal rim at 7 o'clock. A retinal nerve fibre layer defect is also present peripheral to the rim.	Figure 6. Optic nerve head haemorrhage a more obvious haemorrhage than that in Figure 5 is seen at 7:30 on this disc.

Developing the Clinical Paradigm

Normal Optic Nerve Head
A normal ONH is one in which no changes have taken place over time, that is it remains static. It may be small, average, or large in size and should have rims with thickness as described above. Age and changes as-sociated with extreme refractive errors may be present and should be differentiated from changes associated with glaucoma. 'Normal' for an individual may also include abnormalities such as optic pit(s), coloboma, or a tilted disc, all of which may cause visual field changes consistent with glaucoma.

One method of approach for examination of the ONH is as follows:

• estimation of disc size
  
• rim thickness in each quadrant, colour (pallor / translucency), and an active search for haemorrhage(s)
• peripapillary halo pattern
• RNFL defects
• comparison with the other eye
• correlation with visual fields (where possible).

Abnormal Optic Nerve Head
An abnormal ONH is one that has changed over time. However, clinicians often lack the all-important earlier photograph or drawing that would indicate the starting point of the eye prior to the onset of the glaucoma. Clear pathological ONHs are not difficult to identify but a large number are 'borderline' and cannot be distinguished from a normal variant. For example, a large disc (with its large cup) may be a concern until disc size is estimated and the rim thickness pattern is found to be normal. A localised thinning of a NRR is highly suspicious, but the presence of a haemorrhage or an adjacent peripapillary halo / RNFL abnormality may help in judging whether or not it is significant. Clinicians can develop stronger opinions about ONHs that have recognisable abnormalities prior to development of visual field defects.

At least 4 patterns of ONH abnormality have been found focal ischaemic, myopic glaucomatous, senile sclerotic, and generalised enlargement.¹⁷ These examples are 'pure' in their appearance but there are many more hybrids of one or more of these categories. Focal ischaemic ONH abnormality has a localised notch whereas myopic glaucomatous ONH is defined by a tilted, cupped disc, temporal crescent, and deficiencies of the superior and inferior rims in a myope. Senile sclerotic ONH has a pale disc, a saucerised cup, and a large halo in an older person. Patients with generalised enlargement have large, deep cups without peripapillary halo formation.

Documentation of the Optic Nerve Head
Documentation of the ONH is extremely important, whether the patient has suspected or established glaucoma. Change over time is the hallmark of a progressive disease such as glaucoma, therefore, a baseline must be established for future reference. Many clinicians will write the disc size and the cup / disc ratio on a chart but a freehand drawing is preferred. If at all possible, photography will provide the best record. Often, 2 slides held side-by-side will provide stereopsis, which is the next step of documentation. These can be viewed using 2 single slide viewers hinged together with tape. The more thorough the documentation, the smaller the increment of change at the ONH that can be recognised.

Conclusion

This article is intended for those ophthalmologists who do not have access to expensive ONH analysis equipment and yet have the responsibility of caring for populations who are at risk of developing chronic glaucoma. Using simple tools and some coordinated knowledge it is possible for reasonable examination of the ONH, even outside of the clinic situation. A paradigm has been presented that should encourage all those interested in the detection and follow-up of patients with chronic glaucoma to do so with greater assurance that these diseases are neither being overlooked nor overtreated. A 'clinical' study to identify the 4 ONH patterns mentioned above has been reported by some of the authors mentioned. It was encouraging to see that up to 79% were correctly classified.¹⁸ For those who master these guidelines and who, without perimetric backup, can still diagnose chronic glaucoma congratulations! We can never be absolutely certain but we can do better than we are at present.

References

1. Jonas JB, Gusek GC, Naumann GOH. Optic disc, cup and neuroretinal rim size, configuration and correlations in normal eyes. Invest Ophthalmol Vis Sci 1988;29:1151-1158.
2. Britton RJ, Drance SM, Schulzer M. The area of the neuroretinal rim of the optic nerve in normal eyes. Am J Ophthalmol 1987;103:497-504.
3. Bengtsson B. The variation and covariation of cup and disc diameters. Acta Ophthalmologica (Kbh) 1976;54: 804-818.
4. Airaksinen PJ, Drance SM, Schulzer M. Neuroretinal rim area in early glaucoma. Am J Ophthalmol 1985;99:1-4.
5. Balazsi AG, Drance SM, Schulzer M, Douglas GR. The neuroretinal rim area in suspected glaucoma and early chronic open-angle glaucoma: correla-tion with parameters of visual function. Arch Ophthalmol 1984;102:1011-1014.
6. Iester M, Mikelberg FS, Courtright P, Drance SM. Correlation between the visual field indices and the Heidelberg retinal tomograph parameters. J Glaucoma 1997;6:78-82.
7. Littmann H. Zur bestimmung der wahren grose eines objektes auf dem hintergrund des lebenden auges. Klin Monatbl Augenheild 1982;180:286-289.
8. Gross P, Drance SM. Comparison of simple ophthalmoscopic and planimet-ric measurement of glaucomatous neuroretinal rim. J Glaucoma 1995;4: 314-316
9. Jonas JB, Fernandez MC, Sturmer J. Pattern of glaucomatous neuroretinal rim loss. Ophthalmology 1993;100: 63-68.
10. Drance SM, Begg IS. Section hemor-rhage a probable acute ischemic disc change in chronic simple glau-coma. Can J Ophthalmol 1970;5: 137-141.
11. Drance SM. Disc hemorrhages in the glaucomas. Surv Ophthalmol 1989;33: 331-337.
12. Jonas JB, Grundler AE. Correlation between mean visual field loss and morphometric optic disc variables in the open angle glaucomas. Am J Ophthalmol 1997;124:488-497.
13. Jonas JB, Fernandez MC, Naumann GOH. Glaucomatous peripapillary chorioretinal atrophy: occurrence and correlations. Arch Ophthalmol 1992; 110:214-222
14. Rader J, Feuer WJ, Anderson DR. Peripapillary vasoconstriction in the glaucomas and the ischemic optic neuropathies. Am J Ophthalmol 1994; 117:72-80.
15. Airaksinen PJ, Drance SM, Douglas GR. Visual field and retinal nerve fibre layer comparison in glaucoma. Arch Ophthalmol 1985;103:205-207.
16. Sommer A, Katz J, Quigley HA, et al. Clinically-detectable nerve fibre atrophy precedes the onset of glaucomatous field loss. Arch Ophthalmol 1991;109: 77-83.
17. Nicolela MT, Drance SM. Various glaucomatous optic nerve appearances: clinical correlations. Ophthalmology 1996;103:640-649.
18. Nicolela MT, Drance SM, Broadway DC, Chauhan BC, McCormick TA, LeBlanc RP. Agreement among clinicians in the recognition of patterns of optic disc damage in glaucoma. Am J Ophthalmol 2001;132:836-844.