giant cell arteritis.

Examination

Examination of the patient with decreased vision involves detection, quantification, and localization in order to determine etiology. The diagnostic process relies on a full neuro-ophthalmic evaluation.

Best-Corrected Visual Acuity

Best-corrected visual acuity (BCVA) should be obtained with refraction. Improvement with pinhole viewing indicates a refractive component, whereas worsening with pinhole viewing suggests a retinal or lenticular contribution. For patients with visual acuity levels worse than 20/400, the clinician should quantify BCVA using a standard 200 optotype E. The distance at which the patient discerns the letter orientation is recorded using standard Snellen notation (eg, “5/200”). This test provides a more accurate and reproducible measurement than does the finger-counting method.

Distance and near BCVA should be similar, and disparity may suggest a specific pathology. Better visual acuity at near may indicate nuclear sclerotic cataract or macular disease, because near magnification may overcome small scotomata. Better visual acuity at distance may imply posterior subcapsular or polar cataracts. The clinician should document the presence of eccentric fixation (possible central scotoma), tendency to read half of the eye chart (possible hemianopic field defect), or improvement in BCVA when reading single optotypes (possible amblyopia).

Color Vision Testing

Testing of color vision complements the assessment of visual acuity. Optic nerve disease, particularly demyelinating optic neuritis, may disproportionately affect color vision compared with BCVA. In macular disease, visual acuity and color vision tend to decline correspondingly. Thus, the clinician should prioritize optic neuropathy higher than macular disease in the differential diagnosis for an eye with 20/30 visual acuity but severe loss of color vision. Persistent dyschromatopsia can occur even after recovery of visual acuity in optic neuropathy.

Color vision should be tested separately in each eye to detect unilateral disease. The wide availability of pseudoisochromatic color plate testing makes it a commonly used evaluation of color vision. The test screens for congenital red-green color deficiencies but may miss mild cases of acquired dyschromatopsia. Fortunately, most optic neuropathies manifest red-green defects. Bilateral, symmetric color vision loss in men may signify congenital dyschromatopsia rather than bilateral optic neuropathy. The Hardy-Rand-Rittler (HRR) plates can screen for tritan (ie, blue-yellow) defects as well as red-green defects. Blue-yellow color defects often accompany optic neuropathy but also can occur in macular disease and glaucoma. Although Lanthony tritan plates can detect blue-yellow defects, few ophthalmology practices have these test plates.

More detailed color testing under standardized lighting conditions comprehensively characterizes a color vision defect and may help distinguish between acquired and congenital abnormalities. The Farnsworth panel D-15 test requires the patient to arrange 15 colored discs in order of hue and intensity. Desaturation of the color chips (as in the Lanthony desaturated 15-hue test) can improve

sensitivity. The Farnsworth-Munsell 100-hue test, using 85 colored discs, is the most detailed test and provides the best discrimination. It requires a substantial amount of time to test and score, however, which limits its use in routine clinical testing. A shortened version, using only 21 color chips in the set, may be effective for discriminating among optic neuropathies. Color vision testing is discussed further and illustrated in BCSC Section 12, Retina and Vitreous.

Melamud A, Hagstrom S, Traboulsi E. Color vision testing. Ophthalmic Genet. 2004;25(3):159–187.

Nichols BE, Thompson HS, Stone EM. Evaluation of a significantly shorter version of the Farnsworth-Munsell 100-hue test in patients with three different optic neuropathies. J Neuroophthalmol. 1997;17(1):1–6.

Figure 3-1 Assessing for relative afferent pupillary defect (RAPD) in a patient with left traumatic optic neuropathy; the left pupil is pharmacologically dilated. A, The right pupil constricts in response to light directed at the right eye only. B, The right pupil dilates in response to light directed at the left eye only, indicating a left RAPD. (Courtesy of Michael S. Lee, MD.)

Pupillary Testing

Normally, light directed at 1 pupil causes constriction of both pupils (see Chapters 1 and 10 for pupillary anatomy), a response that should be symmetric between the eyes. However, impaired conduction of the afferent pupillomotor signal along 1 optic nerve results in pupillary constriction (direct and consensual) in the affected eye that is reduced compared with the eye with normal optic nerve conduction. This difference in pupillary responses after light stimulation of the right versus left eye is known as a relative afferent pupillary defect (RAPD).

The most popular clinical method for detecting an RAPD is the swinging flashlight test (see Practical Tips in Testing for a Relative Afferent Pupillary Defect). In this test, the patient stares into the distance with the lights dimmed. The examiner shines a bright focal light along the visual axis into 1 pupil for 2–3 seconds and then rapidly swings the light to shine into the other pupil for 2–3 seconds. This action is repeated 4–5 times. The amplitude and velocity of pupillary constriction should be symmetric regardless of which eye is stimulated. In the patient with impaired optic nerve conduction in 1 eye, light stimulation of the affected eye will produce slower pupillary constriction of lower amplitude. With significant asymmetry, the pupils redilate immediately during the 3 seconds of light stimulation (Fig 3-1). When the light stimulus swings to the unaffected eye, the pupils constrict with greater speed and amplitude. To reduce interfering corneal reflections and improve visualization of the pupillary response, the examiner can have the patient stare upward while directing the light to shine at the inferior limbus.

PRACTICAL TIPS IN TESTING FOR ARELATIVE AFFERENT PUPILLARY

DEFECT

1.Dim the ambient lighting; it is easier to evaluate pupillary movement when the pupil size is larger.

2.Ask the patient to fixate at distance to avoid pupillary miosis from accommodation.

3.Use a well-charged, bright, steady light source, such as a standard “muscle light.” A light source that is too dim or too bright may produce false-positive or false-negative results, respectively.

4.Stimulate 1 eye for 2–3 seconds, and quickly move across the bridge of the nose to stimulate the other eye for 2–3 seconds. Make several alternations and mentally average the pupil responses. Do not rely on a single observation.

5.Observe the initial pupillary constriction (velocity and amplitude) and any pupillary dilation. A dense RAPD is easily detected when the affected eye’s pupil dilates when stimulated. A mild to moderate RAPD is more difficult to detect because the abnormal pupil may still constrict, only less vigorously than the normal pupil.

6.If 1 pupil does not react (eg, iris trauma, synechiae, pharmacologic mydriasis or miosis), evaluation of the direct and consensual responses of the functioning pupil may demonstrate the asymmetry of responses and indicate the side of the RAPD (see Fig 3-1).

7.Bilateral optic neuropathy, when fairly symmetric, may show sluggish pupillary responses but not a relative difference (and therefore no RAPD) between the 2 eyes when pupillary responses are compared.

8.The RAPD may be graded 1–4+ in increasing severity or may be quantified using neutral-density filters. The filters, placed in front of the normal eye, decrease the intensity of the light stimulus. Beginning with the lowest 0.3 log10 unit, the swinging flashlight test is repeated. If an RAPD is still detected, increasingly stronger filters are placed over the normal eye until an RAPD disappears. At this balance point, the light input from the normal eye with the filter matches the light input from the abnormal eye. The RAPD is quantified by the strength of the neutral-density filter needed over the normal eye to reach the balance point.

9.The RAPD magnitude correlates with the degree of damage to retinal ganglion cells and their axons. The magnitude may not parallel visual acuity if the papillomacular bundle is spared, and a prominent RAPD can occur with normal visual acuity.

10.Optic neuropathy and RAPD do not cause anisocoria. Although the affected pupil reacts poorly, it does not dilate because the consensual input from the normal eye equalizes the pupil size.

An RAPD is an extremely reliable and sensitive indicator of optic nerve dysfunction. Its absence generally indicates either a lack of an optic neuropathy or bilateral optic nerve involvement. Less commonly, an RAPD may also result from a retinal disorder such as central retinal artery occlusion or retinal detachment. These disorders typically require substantial retinal involvement to produce an RAPD. Chiasmal lesions may produce an RAPD secondary to asymmetric optic nerve involvement. A mild contralateral RAPD can result from optic tract lesions because each tract contains more crossed

than uncrossed fibers, and a lesion will damage more fibers crossing from the affected eye. In rare cases, an RAPD may result from media opacities such as cataract or vitreous hemorrhage or from amblyopia.

Fundus Examination

The fundus examination may reveal media opacities or fundus abnormalities that explain a patient’s decreased visual acuity. Direct ophthalmoscopy, unlike indirect ophthalmoscopy and slit-lamp biomicroscopy, does not permit a view through a media opacity to allow the examiner to evaluate the visibility of the fundus. The direct ophthalmoscope examination can screen for opacities or irregularities in the cornea, lens, or vitreous using the red reflex (opacities appear black on the contrasting red background). The clarity of the view suggests how much visual impairment the media opacities might cause. Using either direct ophthalmoscopy or slit-lamp biomicroscopy, the clinician can assess the appearance of the optic disc and macula. The disc is examined for evidence of atrophy, edema, excavation, or other abnormality; the macula is examined for pigmentary disturbance, edema, scar, or other disruption of structural integrity.

Optic disc pallor indicates optic atrophy, the hallmark of damage to the retinal ganglion cells. Although the examiner observes atrophy of the neuroretinal rim only, disc pallor can occur from damage to any portion of the ganglion cells, from the cell bodies to their synapses at the lateral geniculate nucleus. Optic atrophy does not occur immediately after injury but takes at least 4–6 weeks from the time of axonal damage. Severe damage causes the optic disc to appear chalky white (Fig 3- 2). The disc margins become sharper compared with the dulled appearance of the peripapillary retina because the normal softening effect of the overlying nerve fiber layer (NFL) disappears. Milder forms of atrophy remain more difficult to detect, and therefore the examiner must pay close attention to the following aspects:

Comparison of the color between discs. In some cases, subtle pallor is appreciated only in relation to the normal fellow eye. (Such comparison may be difficult after unilateral cataract extraction.)

Evaluation of the surface vasculature of the disc. Normally, this capillary net is easily visible with the direct ophthalmoscope or with increased magnification on slit-lamp biomicroscopy. The net becomes thin or absent in early atrophy, even when pallor is still very mild.

Assessment of the peripapillary NFL. Dropout of the NFL fibers may precede visible optic atrophy. The fine defects appear as dark bands among normal striations and initially affect the thickest portions of the NFL, the superior and inferior arcades. Such defects are called rake defects owing to their similarity to rake marks in soil. The normal translucent, glistening quality of the retina may disappear. Such loss produces a dull red appearance, which may appear in broad or fine radial patches (Fig 3-3).