Patients presenting with an ophthalmological complaint may strike fear in some practitioners' hearts as they recall student days standing in dimly lit rooms looking at incomprehensible machines which obscure both patient and doctor, wondering what on earth was going on. History taking in ocular complaints is actually very similar to that taken with other problems and examination need not be difficult if you keep it simple and remember what it is you are looking for. The eye is one of the few organs of the body which can be looked at both outside and in and whose basic function can be assessed using simple desktop tools.
As with any physical examination, you need not carry out every test - functional ones should be directed by history and anatomical examination - but as with everything, practice will increase your confidence, improve your speed and you may well even derive some excitement at eliciting signs from a little organ that has a lot to say.
Putting it all together
Obviously, not all the examinations described below need be carried out on every patient. The table makes suggestions but be guided by the individual patient and examine more rather than less if uncertain.
|History and suggested examination in patients with different eye problems|
|All patients||History and visual acuity.|
|Red/painful eye||Lids, lacrimal system, conjunctiva, cornea, pupils, anterior chamber and intraocular pressure if possible (if not and no obvious cause, refer).|
|Foreign body||Lids, conjunctiva and cornea. If mechanism included high velocity, full anatomical examination of the eye is mandatory.|
|Reduced vision||Cornea, anterior chamber and beyond, functional testing of visual field, pupils, optic nerve and macula.|
|Double vision/orbital problems||Fundus, optic nerve function, extraocular muscle function.|
|Headache/neurological-sounding problems in absence of red eye||Fundus, optic nerve, pupillary functions, blood pressure, full neurological examination.|
The basic structure is no different than for other systems (follow usual structure and sections) but take particular note of the following:
- History of presenting complaint - the time and speed of onset, ocular associations (eg, red eye, pain, photophobia, blurry vision, etc), systemic associations (eg, headaches, nausea, rash on the forehead) and what the fellow eye is up to - a surprising number of patients fail to tell you about symptoms in there.
- Past ocular history - similar episodes, other episodes, surgery, 'lazy eye' (amblyopia).
- Social history - work (eg, nursery nurse susceptible to conjunctivitis), home help/carers (eg, will they be able to open a drop dispenser or self-administer drops), independence (eg, this may tip the balance one way or another when deciding on whether to operate on a borderline cataract).
Work systematically from front to back. See also separate article on External Eye Overview.
- Note position with regards to the fellow eye (ptosis), redness ± swelling (orbital cellulitis), lacerations (full thickness vs partial thickness, involvement of the puncta) and lumps/bumps (chalazion, sebaceous cysts): it is sometimes helpful to get the patient to close his or her lids; simultaneously, run the finger over the upper, then lower, lids to assess very small lumps. Note any skin abnormality - rashes (varicella zoster), ulcerations (basal cell carcinoma), ill-defined thickening (squamous cell carcinoma). Note eyelashes - if you have access to a slit lamp, look at them under magnification (blepharitis).
- Lid eversion:
- Indication: suspicion of foreign body (FB), examination of papillae/follicles.
- Requirements: drop of local anaesthetic (LA), two clean cotton buds.
- Procedure: explain to the patient what you are going to do and place an anaesthetic drop in the lower fornix. Ask them to look down, firmly hold eyelashes with index/thumb of one hand and place a cotton bud at the base of the tarsus (top of lid) with the other. The trick is to gently pull the lid down and towards you whilst keeping the cotton bud firmly in place, before lifting the lid directly up whilst reminding patient to keep looking down. Hold the lid in place with the 'eyelash hand' and examine the tarsus and fornix. Use your clean cotton bud to 'dust' off any foreign material (it is usually quite firmly embedded).
- Examine the lids as above and look at the puncta in particular (these are the openings to the canaliculi - tear drainage channels): are they sitting against the globe, turned in (entropion) or drooping out (ectropion)? Look for swellings medial to the canthus - where the lids meet (blocked tear ducts) and any evidence of redness, pain or discharge (dacryocystitis).
- Assessing for dry eye:
- Requirements: fluorescein stain (dilute drops), cobalt blue light.
- Procedure: instil a drop of fluorescein and look at the cornea using cobalt blue light. Ask the patient to close his or her eyes, then open them. For each eye, count the number of seconds it takes for the tear film (visualised as a hazy diffuse spread of fluorescein over the cornea) to break up. It should take at least 10 seconds. There is also Schirmer's test which involves strips of filter paper and waiting for several minutes for tear absorption but it is longer and can be difficult to get this test right so it is not always reliable.
- Look at its colour - is it injected (conjunctivitis), pale (anaemia), are there cysts (clear blebs), concretions (yellow deposits) or ulcerations?
- Look for FBs embedded up (or down) in the fornices or even hidden in folds (get the patient to look far left, then right). Note any discharge.
- Note any discharge.
- Look for presence of follicles or papillae (seen as little bumps in the conjunctival surface).
- Lid eversion (as above) may be necessary to assess the presence of follicles (raised, gelatinous pale bumps) or papillae (vascular bulges) and to rule out conjunctival FBs. Fluorescein staining of the conjunctiva will highlight small lacerations.
- Check first: is the patient wearing their contact lens? If you are suspecting a bacterial keratitis, the lens needs to be sent to the microbiology laboratories (check with your laboratory with regards to local policy about how to store the lens). Is the cornea hazy (all over - eg, acute angle-closure glaucoma - vs localised - eg, band keratopathy) or clear? Are there any white dots visible before fluorescein staining (infiltrates suggestive of infective keratitis).
- Before instilling fluorescein
- Check sensation (neuropathic keratopathy): twist a clean cotton bud/tissue to a tip and lightly touch the cornea - brisk reaction should immediately follow! Fluorescein drops are mixed with anaesthetic and this cannot be tested once instilled.
- If you suspect a penetrating injury, carry out Seidel's test (see 'Further examination', below). If this is positive, stop right there, apply a hard protective shield and send the patient (nil by mouth) to your closest eye unit. In the absence of a shield, pad the eye with gauze and tape and strict instructions not to rub the eye or the structures around it.
- Before instilling fluorescein
- Fluorescein staining
- A dab from the fluorescein strip applicator (ask the patient to blink a few times after this) or a single drop will do. If using the strip, apply it on the sclera or in the fornix, not on the highly sensitive cornea. Look for diffuse tiny spots (punctate epithelial erosion from dry eye) or presence of ulcers (eg, herpes simplex keratitis). If your suspicions are strong but you cannot see anything, refer up, as some of these lesions are tiny but must not go untreated.
- Seidel's test:
- Indication: suspicion of penetrating injury.
- Requirements: 10% fluorescein (this is dark orange - a dry fluorescein strip will do), cobalt blue light source, a rigid eye shield.
- Procedure: apply the fluorescein to the suspicious area, asking the patient not to blink. If it turns from a dark non-fluorescent orange to a swirly bright fluorescent yellow/green, aqueous is leaking out (diluting it). The patient should be made nil by mouth, an eye shield should be applied and an urgent referral made.
Assessment is limited without a slit lamp but a note can be made of the presence of hypopyon - a collection of pus sitting inferiorly (eg, endophthalmitis) or hyphaema - blood in the anterior chamber (eye trauma). If you have a slit lamp, further assess by narrowing the beam to 1 mm and putting it on its brightest light setting. Angle it at 30-45° to the cornea and focus in, past the cornea. If the iris comes into focus, you have focused too far, so go back again - you are effectively focusing into a space. Look for cells (like particles of dust passing through the shaft of light) and flare (slight cloudiness) of anterior uveitis.
Look at their relative size - if you think that there may be anisocoria (different-sized pupils), stand back from the patient, darken the room and look through the ophthalmoscope. You can elicit the red reflex in both eyes and compare the size of these directly rather than shifting from one to the other close up. Look for change in shape (typically oval in acute angle-closure glaucoma, asymmetry in a penetrating injury) and any abnormal oscillations (Adie's tonic pupil syndrome).
Lens and the red reflex
A cataract is not always easy to spot without a slit lamp unless it is very mature (brunescent or even white - rare these days). However, you may notice an opacity by examining the red reflex: this is best seen with a dilated pupil when the patient is looking at the light of the ophthalmoscope which is held about an arm's length away from the patient. Look through the ophthalmoscope and turn the dial until you see the reflex. This may be attenuated as a result of any opacity between the cornea and the fundus but a clue may lie in that a corneal opacity is visible externally and a vitreous opacity may be mobile.
The red reflex is part of the routine neonatal check. Use a direct ophthalmoscope in a dimly lit room and hold your ophthalmoscope about 2/3 of an arm's length away from the baby. If the baby is screwing their eyes shut, ask the mother to feed them and check it during the feed. Very occasionally, a neonate may need dilating to check the red reflex. If this is the case, it is best to refer to an ophthalmologist who may use cyclopentolate 0.5% in both eyes.
A basic fundus examination can be carried out with the ophthalmoscope. The key to success is practice and a systematic approach. In the first instance, become familiar with the ophthalmoscope. Try out the different light sources; most commonly used are the white and where available, the blue filter. The green filter highlights vessels more clearly. The dial changes the lens through which you look and what suits you best depends on whether you have a refractive error (and what type) and whether you keep your spectacles on during the examination. If the image appears to be getting increasingly blurry when dialling the ophthalmoscope, dial in the other direction.
When examining the patient, sit them comfortably, explain what you are going to do (including "I'm going to get very close to your face and shine a bright light in your eye. Don't look directly at the light unless I tell you to."). It helps to stand on the side you are going to do the examination. Always examine both eyes. Your view will be greatly improved by dilating the pupils (NB: make sure you have checked the relative afferent pupillary defect (RAPD) first if necessary - see 'Optic nerve function', below).
Elicit the red reflex and 'home in' through the pupil, looking for the disc. Once you have found it, adjust the focus by turning the dial until you are happy with what you see. Look for pallor (optic atrophy) or oedema (blurred margins - papilloedema). Then work your way along each of the four main vascular branches, looking for attenuation (embolus), aneurysms or exudates (diabetic eye disease). Look at the background retina - are there any haemorrhages (eg, retinal vein occlusion) or areas of pallor (eg, retinal artery occlusion) or unusual pale patches (eg, chorioretinitis)? Lesions are noted as being a number of disc diameters away (nasally or temporally) from the disc. Finally, ask the patient to look at the light to assess the macula which lies about two disc diameters temporally to the disc (it may be seen as a slightly darker area than the surrounding retina - more so in darker people - with blood vessels arching over and under it but not on top of it) - this part of the examination should be swift, as it may be uncomfortable. Look for the presence of haemorrhage or the loss of a pinpoint yellow foveal reflex (macular oedema).
Examination of function
The visual acuity is the most basic of functions that should be examined in every patient. Beyond this, functional testing is guided by the symptoms and findings so far. An outline of some of the further tests available in a specialised department has been given - there are, of course, a vast array that fill volumes of textbooks but here is a sample ...
This essential examination should be carried out on every patient presenting with an eye problem.
The most common method is using a Snellen chart which comprises random letters arranged in rows, decreasing in size in each row. Charts are designed to be read at 3 or 6 metres. The number indicated at the side of the row corresponds to the distance at which a normal eye could read that row. For example, the top row (marked 60) could be read by the normal eye 60 metres away. The patient should be tested one eye at a time using their normal distance glasses (or distance portion of their bifocals) and then using a pinhole - you cannot assume that their glasses are of the correct prescription and the pinhole will correct any refractory errors, unless there is media opacity - eg, corneal oedema.
The reading is recorded as 6/60 - this means that the patient was tested at 6 metres (or equivalent if you used a reversed 3-metre chart and a mirror) and were able to read the top row only. If they score 6/4 (ie read the lowest row), they were tested at 6 metres but their eyesight was so good that they actually saw what a 'normal' person would usually need to be 4 metres away to read. If the patient is unable to read the top row, try counting fingers (CF) at 1 metre in a well-lit room, then hand movements (HM), then perception of light (PL). If the patient sees nothing at all, they are said to be NPL (no perception of light). If the patient reads most of a line right but gets one or two wrong at the 12 metre row for example, this is recorded as 6/12-2. If he or she gets more than two wrong, assume they patient can only read the line above. Similarly, if he or she could manage a couple of letters in the line below but not the whole line, it is recorded as 6/12+2.
Other visual acuity tests
There are variations of the Snellen chart for patients who are illiterate: capital 'E's are rotated in different directions which the patient has to identify. Children may use the Sheridan-Gardiner test where they have to match up letters or pictures of different sizes with those presented on a card in front of them. Very young children are examined by assessing their preferential looking of cards of various pictorial complexity (Cardiff card test) and babies may be assessed by their ability to pick up very small objects such as the 'hundreds and thousands' cake decorations. Infants are watched for tracking of a light source.
A number of other visual acuity tests exist which take other factors into account, such as contrast sensitivity and the crowding phenomenon (where the spacing between the letters affects the acuity in addition to the actual size of the letter, as seen in amblyopes for example) but these are the remit of ophthalmology departments.
The best way to examine these in the primary care setting is to do a confrontational visual field test. This only works with a 'co-operative' patient and, as it is a comparison with your own visual field, it is important to be aware of any defects of your own (which will limit the test). This means that there can be inter-examiner variability. There are various ways to carry out this test but the principle is always the same.
Sit yourself directly opposite the patient, about a metre away from each other. Ask the patient to cover one eye and cover your contralateral eye (so that, effectively, your covered eyes are opposite to each other). Tip: this test requires the patient to concentrate and some people get flustered over their left and right and possibly getting it wrong in front of the doctor. Just say "cover one eye with your hand" and then follow their cue as to which of your eyes you should cover.
Hold your arm out so that your hand is equidistant between you and the patient and place in one of the four quadrants. "I want you to keep your eye looking into mine and I'm going to test how well it can see out at the edges" - keep reminding the patient to look into your eye or the test is meaningless. Depending on the patient, either instruct them to say how many fingers you are holding up (do this 3 or 4 times for each quadrant) or - for a more accurate measurement - use a white hatpin. If the patient's visual acuity is very poor, use a moving finger. Progressively work your way in towards the centre.
Repeat for the fellow eye, giving time for the patient to rest between the two, particularly if they are elderly.
A very low intraocular pressure (IOP) may manifest itself as a soft eyeball on palpation of the globe over the closed lids and a very high IOP may feel hard. However, these are very crude measures (and a globe thought to be soft on account of perforation should not be palpated) and are not a substitute for proper tonometry where there is a concern over IOP.
IOP can be very easily measured using a slit lamp and an applanation tonometer. The principle is that the tonometer measures the degree of resistance provided by the cornea to gentle indentation and converts this into a figure. Most tonometers are mounted on the slit lamp and some use a disposable head, others a washable one. Apply a drop of fluorescein/anaesthetic and put the blue light filter. "Look at the blue light which is going to get very close to your eye. It won't hurt but it is important that you keep looking ahead with your eyes wide open." Tips: if you tell patients not to blink, they will and if you tell them that the tonometer will touch their eyeball, you can be sure that blepharospasm ensues! Gently move forwards and when the tonometer is just about touching the surface of the cornea, look through the eyepiece. You will see two semicircles. Rotate the dial on the side so that the semicircles are only just overlapping. Tell the patient to sit back and take the reading off the side of the tonometer (normal readings should be between 10 and 21 mm Hg).
Although this test is completely painless, many patients find this very difficult. Be patient!
It is important to get these right, as they will influence the results of your tests. Sit the patient in a dimly lit room (to avoid pupillary constriction from the room light over-riding that from your torch) and tell the patient to look at a far wall to overcome the accommodation reflex. Use a bright light source which should be directed from below to avoid the shadow from the nose.
Direct response to light
Light directly shone on the eye for three seconds should elicit a prompt pupillary constriction of the pupil. Failure to do so is known as an afferent pupillary defect and indicates severe optic nerve pathology (eg, transected nerve). There will also be failure of the fellow pupil to constrict. If there is no pupillary reaction but the fellow pupil does constrict, consider a traumatic iris paresis.
The swinging flashlight test
This identifies the presence of a relative afferent pupillary defect (RAPD). Shine the light source from one eye to the other in rapid succession. Stimulation of the normal eye should elicit a brisk constriction of both pupils but when the light is shone on the diseased eye, both pupils dilate. What happens is that the dilatation produced by withdrawing the light from the normal eye outweighs the weak constriction produced by shining light on the diseased eye - this is why it is called a relative afferent pupillary defect.
If the reactions to light are normal, you can go on to test for the accommodation reflex. The room light should be turned on again and the patient asked to gaze to a far wall. Tell the patient that as soon as they see the watch/pen/other object, they should focus straight on it. As they gaze to the distance, hold your object above the level of their eyes, making sure that your arm is not in the way. Then, drop it into their line of view and observe the pupillary reactions as they look at it - there should be a brisk constriction. Failure to do so is known as light-near dissociation.
If all these pupillary tests are normal, the patient can be said to have Pupils Equally Reacting to Light and Accommodation (PERLA).
Optic nerve function
There are several essential components to examining the function of the optic nerve if you suspect pathology at this level:
- Check visual acuity.
- Check for an RAPD.
- Check for colour impairment (dyschromatopsia). Ideally, this is done using Ishihara pseudo-isochromatic plates: cover the good eye first and flick through the booklet, allowing about five seconds per number, then compare with the fellow eye. If the booklet is not available, ask the patient to look at a bright red object (such as a child's toy) and compare the intensity of the colour when viewed with each eye separately - descriptions of things looking "washed out" should ring alarm bells.
- Assess brightness sensitivity: shine a light in each eye and ask the patient to compare the brightness. A useful measure is to suggest: "If I were to give you a pound for this brightness" (shine light in the good eye), "how much would you give me for this ..." (shine light in the bad eye).
- Do a confrontational field test and assess the blind spot: in the same examination position and conditions as for confrontational visual field, bring a bright red object horizontally across their central field of vision, asking them to tell you if/when it disappears and then when it re-appears - if the blind spot is any bigger than yours, examine its margins, moving the pin around until you have an idea of its size.
The easiest method of assessing macular function in an office setting is using an Amsler grid which effectively measures the central 10' to 20' of each eye's visual field. It consists of a piece of paper on which a 10 cm x 10 cm grid box is printed with a black dot in the centre. The patient is asked to cover one eye and fix their gaze on the central dot. The patient is asked if they can see the four corners of the box. They are then told to comment on any distortions or missing areas within the box. If able to, the patient can draw the areas of distortion on and this provides a record of disease progression. This should be repeated for the fellow eye. This gives a reasonable indication of macular function. This simple tool can be used by the patient who can self-test at home and report early if changes are detected.
Examination of the function of the eyelids is usually done in the context of assessing a ptosis. There are several simple measurements that can be made using a simple ruler with millimeter calibrations (a see-through ruler is ideal):
- The palpebral fissure (PF) - the distance between the upper and lower eyelid in vertical alignment with the centre of the pupil.
- The marginal reflex distance-1 (MRD-1) - the distance between the centre of the pupillary light reflex and the upper eyelid margin with the eye in primary gaze.
- MRD-2 - the distance between the centre of the pupillary light reflex and the lower eyelid margin with the eye in primary gaze.
- Levator function - the distance the eyelid travels from downgaze to upgaze while the frontalis muscle is held inactive at the brow: ask the patient to look down pressing your finger firmly on the eyebrow as they do so. Put the ruler near the eye and ask the patient to look as far up as possible. Measure the distance covered from down to upgaze by the lid margin. A normal adult value is typically 15-20 mm.
- The margin fold distance (MFD) - the distance from the upper eyelid margin to the fold of skin.
- Assess for lagophthalmos: gentle closure results in a residual show of upper sclera as the eye moves from up to downward gaze.
- Examine the pupils for evidence of Horner's syndrome.
There are a few more assessments that will additionally be carried out in an oculoplastic clinic to identify the problem or monitor postoperative results.
This is important, as a 3D image cannot be correctly formed if both foveas are not simultaneously fixed on the object. Hold a light source about an arm's length away from the patient and look at the position of the light reflection. This is usually in the centre of each pupil. If one side or the other is towards the outer edge, this indicates an inward deviation of the globe (esotropia) and if there is a reflex more towards the inner edge of the pupil, there is an outward deviation of the globe (exotropia).
The cover/uncover test helps further assess squints. An object to focus on is held in front of the patient who is instructed to focus on it. One eye is completely occluded for several seconds and the uncovered eye is observed for movement as it focuses on the object. This eye is then covered and the other eye is observed for movement. Movement of the eye outwards confirms that there is an esotropia (ie the eye was turned inwards initially) and vice versa for exotropia. The test is repeated for objects at six metres and far distance which may also reveal a vertical squint.
A modified version of this test (the alternate cover test) is used to detect a exo/endophoria (as opposed to exo/endotropia described above) which is a latent squint that only manifests itself in the absence of bifoveal stimulation. Most normal people have this to a very mild degree.
An epicanthus or facial asymmetry may mimic a squint.
This examination is necessary in a number of orbital problems (eg, orbital floor fracture) as well as neuromuscular problems (eg, myasthenia gravis). Sit the patient in front of you and explain that you want them to follow this (show a bright object) with their eyes only and that you will help them keep their head still. Gently but firmly place a hand on their forehead and with the other, test all the positions of gaze in that hemifield. Swap hands and do the same in the other hemifield. Look for limitation of globe movement, presence of nystagmus and ask for diplopia, blurring or loss of the image.
Some other tests carried out in the ophthalmology department
There are various other tests that are routinely performed in specialist units, examples of which are:
- Visual field assessment - using static and kinetic perimeters.
- Ultrasound scan - to visualise the lens, vitreous and retina.
- Exophthalmometer - to assess proptosis (eg, thyroid eye disease).
- Keratometry - to assess the gradient of the cornea, rather like an Ordnance Survey map.
- Hess chart - to map extraocular muscle movement accurately.
- Fluorescein angiography - to map retinal vessels and identify leaks.
- Optical Coherence Tomography (OCT).
Further reading & references
- Snellen Chart; Living Well with Low Vision
- Clinical Examination: Relative Afferent Pupillary Defect; Richmond Eye Associates
- Denniston AKO, Murray PI; Oxford Handbook of Ophthalmology (OUP), 2009
- Wilson FM; Practical Ophthalmology: A Manual for Beginning Residents, American Academy of Ophthalmology, 2005
- Madge SN, Kersey J, Hawker M, Lamont M; Clinical Techniques in Ophthalmology (2006)
Disclaimer: This article is for information only and should not be used for the diagnosis or treatment of medical conditions. EMIS has used all reasonable care in compiling the information but make no warranty as to its accuracy. Consult a doctor or other health care professional for diagnosis and treatment of medical conditions. For details see our conditions.
Dr Olivia Scott
Dr Colin Tidy
Dr Olivia Scott