Understanding and applying a bedside visual field assessment is an important skill for the neurologist. By appreciating some basic anatomical and physiological principles it is possible to target the examination appropriately, thus gaining important diagnostic information with the minimum of fuss. Specific patterns of visual loss are caused by damage at various sites within the visual pathway. This review focuses on techniques that can be used at the bedside to identify common visual field defects and cites examples by dividing the visual system into its component parts. We urge the use of an appropriately sized red pin and berate the well worn “waggling finger” technique.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The process of converting the photons that stimulate retinal rods and cones into our perception of a three-dimensional full colour environment is an extraordinary feat of evolutionary engineering. The foveate human visual system is adapted to provide both high quality central colour vision and sensitive movement detection. The basic “camera” functions are enhanced by a sophisticated eye movement system which provides image stabilisation, gaze shift and binocularity too. These produce the benefits of depth perception, acuity and colour vision. Most people of course take their vision for granted but “you don’t know what you’ve got ‘til it’s gone”.1 Moreover, even transient visual loss is frequently a frightening experience for patients. Many fear that they will go blind and this can both amplify and distort perceptions.
In this article, we will discuss the “bedside” assessment of the visual fields and their importance in both the localisation of disease and the identification of its absence. Not everything of course fits into the typical patterns that we describe but we hope that this article may at least whet the appetite to know more. In order to understand how particular defects in the area seen by a fixating eye can occur, first it is important to know something of the organisation of the visual sensory system.
Anatomy of the visual system
The functional organisation of the visual sensory system begins at the retina (fig 1) which has an inner layer—the neural retina—that contains the 95 million or so light sensitive receptor cells, or rods (responsible for vision in dim light) and cones (for fine detail and colour vision).2 The distribution of visual function across the retina is not uniform. The highest concentration of cones is in the central macular area, the fovea. The neural retina also contains the retinal ganglion cells whose axons pass across the surface of the retina, collect in a bundle at the optic disc and leave the eye to form the optic nerve. These cells can be divided into two main populations: M and P (relating to the layers of the lateral geniculate nucleus to which they project (M = magnocellular and P = parvocellular)). M cells function to detect motion and contrast sensitivity whereas P cells deal with fine spatial detail and colour vision. As a result of the convex lens in the eye, objects that stimulate the nasal retina are located in the temporal visual field and those that stimulate the inferior retina are located in the superior visual field.
More than 90% of retinal nerve fibres arise from the macula to form the papillomacular bundle passing to the optic nerve, taking up its central core (fig 2).3 Lesions of this bundle (either in the retina or in the optic nerve head) produce central or centrocaecal scotomas (ie, a central scotoma that connects with the physiological blind spot) (fig 3).
Retinal nerve fibres temporal to the macula must arc above or below the papillomacular bundle to enter the optic disc. Lesions of these arcuate bundles produce arcuate scotomas (fig 4). Nasal fibres pass directly to the nasal border of the disc. Lesions of the nasal retinal axons cause temporal field defects.
Once in the optic nerve, the macular fibres move to a central position with the superior retinal fibres above and the inferior retinal fibres below. Temporal and nasal fibres retain these positions within the optic nerve.
Lesions of the optic nerve tend to produce several different patterns of visual field abnormality depending on whether the lesion is in the anterior, middle or posterior portions. A good example of disease affecting the anterior optic nerve (the optic nerve head) is ischaemic optic neuropathy. The middle portion is often affected by demyelinating optic neuritis and the posterior optic nerve may be compressed by an enlarging pituitary tumour. The optic chiasm is usually situated directly above the pituitary gland but in about 15% it is “pre-fixed”, meaning that it is fixed anterior to the pituitary gland. When the chiasm is “post-fixed” (posterior to the pituitary) then a pituitary tumour may directly compress the optic nerves or the junction between the optic nerves and the chiasm causing either monocular field loss or a “junctional scotoma” which will be discussed later. Anterior pathologies tend to respect the horizontal meridian (including arcuate defects), lesions of the mid portion of the nerve cause central scotomas, and those involving the posterior portion close to the optic chiasm often show some respect to the vertical meridian.
Optic chiasm, tract and radiation
At the optic chiasm, the visual system becomes functionally divided by a vertical meridian through fixation. Visual information from the left hemifields of both eyes now continues through the visual pathway towards the right occipital cortex (and vice versa with the right hemifields). The crossed nasal fibres once joined with uncrossed temporal fibres proceed as the optic tract. These fibres in the optic tract then pass to the lateral geniculate nucleus of the thalamus where crossed and uncrossed retinal fibres are organised into retinotopic pairs (retinotopy refers to the projection of an image onto a part of the visual cortex in a way that preserves the spatial relations of the image on the retina). From the lateral geniculate nucleus the optic radiations pass posteriorly to the visual cortex. Nerve fibres originating from the superior retina (ie, lower visual field) travel directly to the visual cortex through the parietal lobe while those from the inferior retina (superior visual field) pass through the temporal lobe initially anteriorly and laterally. Homonymous quadrantanopias may thus be a feature of lesions of the optic radiations.
The primary visual cortex lies along the superior and inferior sides of the calcarine fissure. It contains a large macular projection area in the occipital pole (with peripheral optic tract fibres terminating more anteriorly). The occipital pole in particular has a rich anastamotic network between the terminal branches of the middle cerebral and posterior cerebral arteries. Activation of the primary visual cortex results in distribution of visual information into temporal, parietal and frontal lobes with increasing visual specialisation (colour, motion) and decreasing retinotopy. Broadly, a division into dorsal (parietal) and ventral (temporal) streams can be made but although this has some relevance to visual field assessment (for example, the use of static or moving targets and simultaneous stimulation) detailed discussion is beyond the scope of this article.
The visual field: definition and assessment
The visual field is the area seen by the steadily fixating eye and as such it is a slightly artificial construct. Traquair described it as “an island of vision surrounded by a sea of darkness”.4 The central peak of this island (fig 5) corresponds to the fovea (the centre of the retina where there is the highest concentration of cones) and is typically the area of highest sensitivity. Central field representation greatly dominates all levels of the visual pathway, including the visual cortex.5 In practice, nearly all of the pathological abnormalities fall within this central area (ie, virtually all visual field defects are detected within 20–30° of fixation).
Defects in the visual field may be relative or absolute. Large and dense visual field defects are easily identified but for the detection of small or relative defects, the choice of test is crucial. This applies in particular to confrontation tests at the bedside. Formal perimetry with either Goldmann or Humphrey perimeters requires trained operators and relatively expensive equipment and we will not discuss these here.
Taking the history
A visual field assessment should, like all elements of the neurological diagnosis, be a targeted procedure, and we hope the clinician has a fair idea of what might be going on from the history alone and so know what to expect and what to examine. Pertinent questions include:
What is the time course of the visual loss (eg, an acute onset suggests a vascular cause, progression over days followed by a plateau and then recovery is the rule for demyelinating optic neuritis, and a steadily progressive course is more in keeping with a compression)? The patient who suddenly notices a defect that may have been present for some time, when, say an insect flies into the other (good) eye does imply that the pathology is anterior to the chiasm, and except in the most unobservant, usually suggests an insidious onset of visual loss.
Is the visual loss monocular or binocular (has the patient tried covering one eye?) and is the defect homonymous (ie, loss of vision on the same side of the visual field in both eyes) in which case the pathology is retrochiasmal? Many patients with homonymous visual loss attribute their visual problems to the eye with the temporal field defect. It is useful in such circumstances, particularly if the disturbance was transient, to ask the patient what they can or could see (eg, is or was one half of faces missing, etc)? If so, this is not consistent with monocular loss and indicates a homonymous field defect
Is there pain, which may point to an inflammatory aetiology affecting the optic nerve?
Are there symptoms of raised intracranial pressure (including visual obscurations)?
Are there additional symptoms (eg, double vision or those suggesting a focal cortical disorder)?
The initial examination should include documenting the uncorrected and corrected visual acuity (important to know if the patient can see the targets used for field examination) and colour vision (with Ishihara test plates) which provides useful additional information on optic nerve function and underlying colour blindness which may influence the use of a red pin for field testing.
Confrontation testing of visual fields
The examination performed should be tailored to the particular clinical problem. For example, in the case of monocular visual loss associated with painful eye movement, one would look for a central scotoma to a red pin. If a patient has papilloedema, extra time should be spent examining for an enlarged blind spot. Clearly, when considering retrochiasmal pathology, looking for homonymous field defects is paramount.
Start any confrontation testing of the visual fields by asking the patient, viewing with both eyes, to describe any visual loss when looking at the examiner’s face and then to count (one or two) fingers in each of the four quadrants in each eye. Follow this by individual and then bilateral simultaneous stimulation of the superior and inferior quadrants. This is a quick way of picking up homonymous hemianopias and attentional field defects and neglect but cannot, of course, be expected to detect monocular or even chiasmal defects (because this is a binocular test it cannot necessarily pick up bitemporal defects because of the overlap between the visual fields in both eyes—a large enough target in the temporal field of one eye might be seen in the nasal field of the other eye).
The most sensitive method of detecting visual field defects by confrontation is to use a small (ideally 5 mm) red pin (this has a sensitivity of 73% compared with automated perimetry) (table 1). A “kinetic to finger” (ie, waggling finger) test to assess for field defects in the same study had a sensitivity of only 40%, while a 20 mm white pin was 48% sensitive.5 Assess each eye in turn, and move the target slowly in from the periphery while asking the patient when the colour red becomes detectable rather than when they first can see the pin; they will initially see it as black at the periphery where there are no colour sensitive cones. The same target should then be presented statically to multiple points within the central red field (usually representing about 20–30°). A red pin, by stimulating fewer cones than a white pin, detects more subtle defects. Although white pins are easier to find than red pins in a local shop, they can be painted red or ordered from www.hatpinsdirect.co.uk.
If a chiasm defect is suspected, then careful comparison of colour perception across the vertical meridian in each eye is important. The simultaneous presentation of two identical red pins on either side of the vertical meridian both above and then below the horizontal meridian is a sensitive method of doing this.
It is fairly common to see clinicians bring a waggling finger in from the periphery but this dramatically reduces the test sensitivity (see above) in the peripheral field and fails entirely to assess the central fields. This technique should not be used.
Visual field defects
There are a number of different patterns of visual field defects which we will consider in relation to the anatomical structures affected by pathology (table 2). Generally speaking, lesions of the retina, optic nerve, optic chiasm and visual pathways conform to a limited set of patterns:
altitudinal field defect
arcuate field defect
hemianopia (homonymous or bitemporal)
quadrantanopia (homonymous or bitemporal)
wedge shaped field defect
blind spot enlargement
Diseases of the retina can cause a variety of visual field defects but usually there is visible disease when viewing the retina with the ophthalmoscope. Macular lesions produce central (at fixation) scotomas that spare the periphery. These are mostly caused by ophthalmological disorders and a full discussion of retinal disease is beyond the scope of this review.
The optic nerve
Lesions at the optic disc typically produce visual defects which emerge from the blind spot and often respect the horizontal meridian.
With papilloedema, central and arcuate scotomas may be seen, and in chronic papilloedema inferior nasal defects are not infrequent. Generally the field defects are worse nasally than temporally where some vision may remain before progression to complete blindness. The increase in blind spot size in papilloedema is thought to be due to compression, detachment and displacement of the peripapillary retina or the elevation of that part of the retina by subretinal fluid (fig 6).6 The neuronal swelling in papilloedema is more extensive in the peripheral than the central optic nerve accounting for the additional constriction of the visual field (as all fibres carrying information from the peripheral field enter the edge of the optic disc).7 Minor enlargement of the blind spot is difficult to identify by bedside testing; formal perimetry is needed both in this situation and when sequential assessments are required to follow the progress of papilloedema.
When the anterior portion of the optic nerve (including the optic nerve head) is involved in a disease process, the defect usually respects the horizontal meridian. This is termed an “altitudinal” field loss. Ischaemic optic neuropathy typically produces an inferior defect (fig 7). It is postulated that impairment of posterior ciliary artery perfusion is responsible8 because the ciliary arteries supply a vascular anastamosis known as the circle of Zinn-Haller, which may have distinct upper and lower zones thus providing an anatomical basis for the altitudinal visual field defect observed in two-thirds of reported cases.9
Lesions at the optic disc may also produce defects which extend from the blind spot to the temporal periphery.
Glaucoma also affects the anterior optic nerve but typically causes arcuate defects, and these too respect the horizontal meridian (fig 4).
As the optic disc has no retinal photoreceptors it forms the blind spot (a physiological absolute scotoma) in the visual field. A central scotoma occurring in the absence of macular disease is the hallmark of a lesion involving the optic nerve (fig 8).
Defects are usually, but not always, monocular; hereditary optic atrophies and toxic/nutritional neuropathies are bilateral and optic neuritis may be bilateral and simultaneous (eg, in neuromyelitis optica). Retrobulbar disorders of the optic nerve (eg, optic neuritis, toxic neuropathies) especially depress the function of the central core of the nerve which carries fibres from the macula region (the papillomacular bundle) and so disease here causes a central scotoma (seen to some extent in over 90% of patients with demyelinating optic neuritis10). Central scotomas are classic in demyelinating optic neuritis but a broader spectrum of patterns commonly occurs, including diffuse defects (ie, generalised depression of the entire central 30° of the visual field) and focal defects.11
When a central field defect is found in one eye it is important to carefully examine the field of the other eye in case there is a junctional scotoma caused by compression at the meeting point of the optic nerve and chiasm (fig 9A, and see below).
The optic chiasm
The essential feature of visual field defects caused by damage to the optic chiasm is some form of bitemporal field defect with considerable variations on this depending on the exact site of the lesion. The lesions are usually insidiously progressive as most are the result of slow growing tumours, commonly pituitary adenomas, but craniopharyngiomas and meningiomas also occur.
Pathology at the anterior angle of the chiasm (ie, where the optic nerves join) causes a “junctional” scotoma because of separation of nasal crossed and temporal uncrossed fibres. What this means in practice is that small lesions may damage only the crossing fibres of the ipsilateral eye and can encroach on the ipsilateral optic nerve causing varying degrees of field defect. The contralateral field defect (caused by damage to the crossing fibres) may be overlooked unless the “normal” eye is carefully tested. So, a junctional lesion consists of a central scotoma in one eye with temporal (often superior) field loss in the other eye (fig 9A). The way in which the contralateral fibres are damaged is attributed to a structure known as “Wilbrand’s knee” where there is an anterior extension of these fibres into the affected optic nerve. This anatomical explanation has been challenged but, as Trobe points out, “the clinical entity persists”.12 Identifying junctional scotomas is important because it localises anterior chiasmal lesions. Less common is the “junctional scotoma of Traquair” (fig 9B) also caused by compression in a similar anatomical location but where only crossing nasal fibres are affected, producing a small temporal quadrantanopic scotoma.
Damage to the body of the optic chiasm
A bitemporal field defect (albeit sometimes quadrantic) is usually seen. Other symptoms may relate to problems with depth perception (eg, threading a needle). Figure 10 shows how a bitemporal field defect can cause blindness beyond a fixation point.
Other conditions that may mimic chiasmal field defects include congenitally tilted discs, nasal sector retinitis pigentosa and papilloedema with greatly enlarged blind spots. The first is a condition where the superotemporal optic disc is elevated and the inferonasal disc displaced posteriorly leading to an oval-shaped obliquely orientated disc (fig 11). Patients often have myopic astigmatism and can present with a bitemporal hemianopia suggesting a chiasmal syndrome. In these patients, however, the hemianopia is typically incomplete, does not respect the vertical meridian and is confined primarily to the superior quadrants.
Establishing that the vertical meridian forms the central border of the temporal defect is key to distinguishing chiasmal pathology from other defects that may mimic this.
Any lesion posterior to the chiasm produces a homonymous defect. The exception is a defect in the temporal periphery of the eye contralateral to a lesion of the anterior-most occipital lobe known as a “monocular crescent”.
The optic tract
As only the contralateral hemifield is represented in the optic tract then visual field defects are homonymous. Partial lesions cause contralateral homonymous defects but these may be incongruous (ie, defects are differing in shape and depth in each eye). There is a trend for more congruent homonymous hemianopias with lesions situated more posteriorly in the visual pathway but 50% of optic tract lesions produced congruent field defects in one study.14
The optic radiations
Superior homonymous quadrantic field defects (“pie in the sky”) may result from lesions located in the temporal (Meyer’s) loop of the optic radiation. The defect is usually, but not always, incongruous. If the optic radiation involved is in the parietal lobe, then a homonymous hemianopia usually results (often denser in the inferior quadrant). There are frequently other symptoms and signs of cerebral damage specific to the location of the pathology.
The occipital cortex
Focal lesions of the occipital cortex produce homonymous, contralateral visual hemifield defects which are highly congruent. A lesion at the posterior pole of the visual cortex will produce a homonymous, paracentral hemianopic scotoma. Such defects involve the central 5–10° of vision and may spare the remainder of the visual field if the rest of the striate cortex is normal. Alternatively, a homonymous hemianopia with “macular sparing” occurs with occipital lesions that spare the posterolateral striate cortex (fig 12). Partial lesions are frequently observed. In order to qualify as macular sparing, at least 5° of the macular field must be spared in both eyes on the side of the hemianopia (but this can be an artefact if the patient shifts fixation).
Functional visual field loss
Tests of the visual fields, either by confrontation or especially using perimetry, may be helpful in highlighting inconsistencies characteristic of functional field defects (figs 13, 14). Observing someone navigating around an unfamiliar environment with ease and then struggling to observe large bright targets on formal testing also helps. It is often suggested that, in this situation, it is helpful to test the visual fields at, for example, 1 m and 2 m. The concept is that in functional field loss the patient will maintain an equally constricted field at a greater distance while the normal field might be expected to enlarge. However, this technique does not recognise that in moving further away the angle subtended by the target at the retina diminishes and its luminance is reduced which may counteract any field enlargement. The technique can be useful using tangent (Bjerrum) screen testing at 1 and 2 m with appropriately sized targets but in practice we do not find this helpful, preferring to rely on the general principles above.
The visual pathway is complex but lesions along its course cause visual field defects that conform to certain patterns. Simple techniques make it possible to select a bedside examination appropriate to the clinical situation to aid localisation of the lesion. Lesions anterior to the chiasm may cause a variety of (usually but not exclusively) monocular defects. The classical bitemporal hemianopia of chiasmal compression may start as a subtle alteration of red perception across the vertical meridian. Retrochiasmal defects are homonymous. Armed with a grasp of the fundamental principles outlined above and a red pin, we hope that bedside visual field testing will become a more frequently performed and better understood aspect of the neurological examination.
The visual pathway consists of the retinas, optic nerves, optic chiasm, optic tracts, optic radiations and visual cortex; lesions in each area cause distinct patterns of visual loss.
The organisation of the retinal nerve fibre layer and the subsequent arrangement in the optic nerve head gives rise to arcuate defects, central/centrocaecal scotomas or temporal defects depending on the site of the lesion.
Lesions of the optic nerve may result in several different field defects depending on whether the anterior, middle or posterior portion of the nerve is affected; a central scotoma localises the pathology to the optic nerve.
At the chiasm vision becomes functionally divided by a vertical meridian.
Posterior to the chiasm visual field defects are homonymous (ie, involving the same hemifield in each eye).
Targeting the examination is important; beginning with asking for a description of the examiner’s face and then quadrant finger counting is useful.
A red pin is the most sensitive tool for assessing the visual fields at the bedside.
This article was reviewed by Christian Lueck, Canberra, Australia.
Competing interests None.
Provenance and peer review Commissioned; externally peer reviewed.