Chapter 14 Acquired Macular Disorders

INTRODUCTION 594

CLINICAL EVALUATION OF MACULAR DISEASE 595

Symptoms 595

Slit-lamp biomicroscopy 595

Visual acuity 596

Contrast sensitivity 598

Amsler grid 598

FUNDUS FLUORESCEIN ANGIOGRAPHY 601

INDOCYANINE GREEN ANGIOGRAPHY 608

OPTICAL COHERENCE TOMOGRAPHY 611

AGE-RELATED MACULAR DEGENERATION 611

Introduction 611

Drusen 613

Prophylactic antioxidant supplementation in AMD 615

Non-exudative (dry) AMD 616

Retinal pigment epithelial detachment 616

Retinal pigment epithelial tear 619

Choroidal neovascularization 620

Haemorrhagic AMD 627

Retinal angiomatous proliferation 627

POLYPOIDAL CHOROIDAL VASCULOPATHY 628

AGE-RELATED MACULAR HOLE 629

MACULAR MICROHOLE 631

CENTRAL SEROUS CHORIORETINOPATHY 632

CYSTOID MACULAR OEDEMA 633

EPIMACULAR MEMBRANE 635

DEGENERATIVE MYOPIA 637

ANGIOID STREAKS 641

Ocular considerations 641

Systemic associations 641

CHOROIDAL FOLDS 643

HYPOTONY MACULOPATHY 644

VITREOMACULAR TRACTION SYNDROME 645

IDIOPATHIC CHOROIDAL NEOVASCULARIZATION 645

SOLAR RETINOPATHY 645

Page 594

Introduction

Anatomical landmarks

1 The macula (Fig. 14.1A) is a round area at the posterior pole, lying inside the temporal vascular arcades. It measures between 5 and 6 mm in diameter, and subserves the central 15–20° of the visual field. Histologically, it shows more than one layer of ganglion cells, in contrast to the single ganglion cell layer of the peripheral retina. The inner layers of the macula contain the yellow xanthophyll carotenoid pigments lutein and zeaxanthin in far higher concentrations than the peripheral retina (hence the full name ‘macula lutea’ – yellow plaque).

2 The fovea is a depression in the retinal surface at the centre of the macula, with a diameter of 1.5 mm (Fig. 14.1B and Fig. 14.2), about the same as the optic disc.

3 The foveola forms the central floor of the fovea and has a diameter of 0.35 mm (Fig. 14.1C). It isthe thinnest part of the retina and is devoid of ganglion cells, consisting only of a high density of cone photoreceptors and their nuclei, together with Müller cells.

4 The umbo is a depression in the very centre of the foveola which corresponds to the foveolar light reflex, loss of which may be an early sign of damage.

5 The foveal avascular zone (FAZ), a central area containing no blood vessels but surrounded by a continuous network of capillaries, is located within the fovea but extends beyond the foveola. The exact diameter varies with age and in disease, and its limits can be determined with accuracy only by fluorescein angiography (FA); an average is 0.6 mm.

Fig. 14.1 Anatomical landmarks. (A) Normal foveal light reflex; (B) OCT shows the foveal depression; (C) fovea (yellow circle); foveal avascular zone-approximate (red circle); foveola (lilac circle); umbo (central white spot)

Fig. 14.2 Cross-section of the fovea

Retinal pigment epithelium

1 Structure

• The retinal pigment epithelium (RPE) is composed of a single layer of cells, hexagonal in cross-section. The cells consist of an outer non-pigmented basal element containing the nucleus, and an inner pigmented apical section containing abundant melanosomes.

• The cell base is in contact with Bruch membrane, and at the cell apices multiple thread-like villous processes extend between the outer segments of the photoreceptors.

• At the posterior pole, particularly at the fovea, RPE cells are taller and thinner, more regular in shape and contain more numerous and larger melanosomes than in the periphery.

2 Functions

• RPE cells and intervening tight junctional complexes (zonula occludentes) constitute the outer blood–retinal barrier, preventing extracellular fluid leaking into the subretinal space from the choriocapillaris, and actively pump ions and water out of the subretinal space.

Page 595

• Its integrity, and that of Bruch membrane, is important for continued adhesion between the two, thought to be due to a combination of osmotic and hydrostatic forces, possibly with the aid of hemidesmosomal attachments.

• Facilitation of photoreceptor turnover by the phagocytosis and lysosomal degradation of outer segments following shedding.

• Preservation of an optimal retinal milieu. Maintenance of the outer blood–retinal barrier is a key factor, as are the inward transport of metabolites (mainly small molecules such as amino acids and glucose) and the outward transport of metabolic waste products.

• Storage, metabolism, and transport of vitamin A in the visual cycle.

• The dense RPE pigment serves to absorb stray light.

Bruch membrane

1 Structure. Bruch membrane separates the RPE from the choriocapillaris and on electron microscopy consists of five distinct elements:

• The basal lamina of the RPE.

• An inner collagenous layer.

• A thicker band of elastic fibres.

• An outer collagenous layer.

• The basal lamina of the inner layer of the choriocapillaris.

2 Function. The RPE utilizes Bruch membrane as a route for the transport of metabolic waste products out of the retinal environment. Changes in its structure are thought to be important in the pathogenesis of many macular disorders – for example, its integrity may be important in the suppression of choroidal neovascularization (CNV).

Clinical evaluation of macular disease

Symptoms

1 Blurred vision and difficulty with close work may be an early symptom. Onset can be rapid in some conditions such as CNV.

2 A positive scotoma, in which patients complain of something obstructing their central vision, is a symptom of more severe disease. This is in contrast to optic neuropathy which typically causes a missing area in the visual field (a negative scotoma).

3 Metamorphopsia (distortion of perceived images) is a common symptom that is not present in optic neuropathy.

4 Micropsia (decrease in image size) is caused by a spreading apart of foveal cones, and is less common.

5 Macropsia (increase in image size) is due to crowding together of foveal cones, and is uncommon.

6 Colour discrimination may be disturbed, but is generally less evident than in even relatively mild optic neuropathy.

7 Difficulties related to dark adaptation such as poor vision in dim light and persistence of after-images may occur.

Slit-lamp biomicroscopy

Indirect slit-lamp ophthalmoscopy (Fig. 14.3A) utilizes high power convex lenses designed to obtain a wide field of view of the fundus which is vertically inverted and laterally reversed (Fig. 14.3B). The technique is as follows:

Page 596

a The slit beam is adjusted to a width about

of its full round diameter.

b The illumination is set at an angle coaxial with the slit lamp viewing system.

c The magnification and light intensity are adjusted to the lowest settings.

d The light beam should be centred to pass directly through the pupil.

e The lens is held directly in front of the cornea, just clearing the lashes.

f The fundus is examined by moving the joystick and vertical adjustment mechanism of the slit-lamp whilst keeping the lens still.

g Magnification is increased as necessary.

h To view the peripheral retina the patient is instructed to direct gaze accordingly.

Fig. 14.3 (A) Indirect slit-lamp biomicroscopy; (B) fundus view

(Courtesy of B Tompkins – fig. B)

Visual acuity

Snellen visual acuity

Distance visual acuity (VA) is directly related to the minimum angle of separation (subtended at the nodal point of the eye) between two objects that allow them to be perceived as distinct. It is usually carried out using black letters or symbols (optotypes) of a range of sizes set on a white chart at a standard distance.

1 Normal VA equates to 6/6 (metric notation; 20/20 in non-metric ‘English’ notation) on Snellen testing (see below, Fig. 14.4). This should be regarded as only a reference or screening standard because normal corrected VA in healthy young adults is usually superior (6/4 – roughly 20/12) and then drops to 6/6 (20/20) by around the 7th decade.

2 Best-corrected VA denotes the level achieved with optimal refractive correction.

3 Pinhole VA. A pinhole (PH) aperture consists of an opaque occluder perforated by one or more holes of about 1 mm diameter (Fig. 14.5). It compensates for the effect of refractive errors. However, PH acuity in patients with macular disease and posterior lens opacities may be worse than with spectacle correction.

Fig. 14.4 Snellen visual acuity chart

Fig. 14.5 Pinhole occluder

Very poor visual acuity

If the patient is unable to read any letters at any distance, VA is recorded as follows:

1 Counting fingers (CF) denotes that the patient is able to tell how many fingers the examiner is holding at a specified distance (Fig. 14.6).

2 Hand movements (HM) is the ability to distinguish whether the examiner’s hand is moving when held just in front of the patient.

3 Perception of light (PL) means that the patient can discern only light. The quadrant from which the light can be perceived (projected) should be noted on a chart (Fig. 14.7).

Fig. 14.6 Testing of ‘counting fingers’ visual acuity

Fig. 14.7 Notation for the projection of light test (right eye); the patient cannot detect light directed towards the superior and temporal quadrants

LogMAR acuity

Page 597

LogMAR charts address many of the deficiencies of the Snellen chart (Table 14.1), and are the standard means of VA measurement in research and increasingly in clinical practice.

• LogMAR is an acronym for the base-10 logarithm of the minimum angle of resolution, and refers to the ability to resolve the elements of an optotype. Thus, if a letter on the 6/6 (20/20) equivalent line subtends 5′ of arc, and each limb of the letter has an angular width of 1′, an MAR of 1′ is needed for resolution. For the 6/12 (20/40) line, the MAR is 2′, and for the 6/60 (20/200) line it is 10′.

• The logMAR score is simply the base-10 log of the MAR, so as the log of the MAR value of 1′ is zero, 6/6 is equivalent to logMAR 0.00. The log of the 6/60 MAR of 10′ is 1, so 6/60 is equivalent to logMAR 1.00. The log of the 6/12 MAR of 2′ is 0.301, giving a logMAR score of 0.30. Scores better than 6/6 have a negative value.

• As letter size changes by 0.1 logMAR units per row and there are five letters in each row, each letter can be assigned a score of 0.02. The final score can therefore take account of every letter that has been read correctly and the test should continue until half of the letters on a line are read incorrectly.

Table 14.1 Comparison of Snellen and logMAR visual acuity testing

Snellen	LogMAR
Shorter test time	Longer test time
More letters on the lower lines introduces an unbalanced ‘crowding’ effect	Equal numbers of letters on different lines controls for crowding effect
Fewer larger letters reduces accuracy at lower levels of VA	Equal numbers of letters on low and higher acuity lines increases accuracy at lower VA
Variable readability between individual letters	Similar readability between letters (particularly in later ETDRS charts using Sloan optotypes)
Lines not balanced with each other for consistency of readability	Lines balanced for consistency of readability (particularly in later ETDRS charts)
Smaller chart so relatively portable	Larger chart so less portable
6 m testing distance: longer testing lane (or a mirror) required	4 m testing distance on many charts: smaller testing lane (or no mirror) required
Letter and row spacing not systematic	Letter and row spacing set to optimize contour interaction
Lower accuracy and consistency so relatively unsuitable for research	Higher accuracy and consistency so appropriate for research (but optimally three versions of ETDRS chart and standardized lighting required)
No negative scoring	Better than 6/6 equivalent VA gives a counterintuitive negative score
Straightforward scoring system; little or no mental arithmetic needed	Slightly more complex scoring; some mental arithmetic required (easier with VAR system)
Easy to use; many clinicians perceive as satisfactory in standard clinical setting	Despite advantages, slightly less user-friendly

LogMar charts

1 The Bailey–Lovie chart (Fig. 14.8) is the best-known, and is designed to be used at 6 m. Each line of the chart comprises five letters and the spacing between each letter and each row is related to the width and the height of the letters.

• The distance between two adjacent letters on the same row is equal to the width of a letter from the same row, and the distance between two adjacent rows is the same as the height of a letter from the lower of the two rows.

• Snellen VA values and logMAR VA are listed to the right and left of the rows respectively.

• VA can also be recorded on the Bailey–Lovie chart using the Visual Acuity Rating (VAR) score in which the 6/6 equivalent line read correctly gives a score of 100, with one point taken away or added for each letter fewer or more than this.

Page 598

2 Other charts are calibrated for 4 m. The early treatment diabetic retinopathy study (ETDRS) charts utilize balanced rows comprising Sloan optotypes, developed to confer equivalent legibility between individual letters and rows. ETDRS letters are square, based on a 5 × 5 grid, i.e. 5′ × 5′ for the 6/6 equivalent letters at 6 m. In the Bailey–Lovie chart, a 6/6 letter is 5′ in height by 4′ in width.

Fig. 14.8 Bailey–Lovie chart

Contrast sensitivity

1 Principles. Contrast sensitivity is a measure of the ability of the visual system to distinguish an object against its background. A target must be sufficiently large to be seen, but must also be of high enough contrast with its background; a light grey letter will be less well seen against a white background than a black letter. Contrast sensitivity represents a different aspect of visual function to that tested by the spatial resolution tests described above, which all use high-contrast optotypes.

• Many conditions reduce both contrast sensitivity and visual acuity, but under some circumstances (e.g. amblyopia, optic neuropathy, some cataracts, and higher order aberrations), visual function measured by contrast sensitivity can be reduced whilst VA is preserved.

• Hence, if patients with good VA complain of visual symptoms (typically evident in low illumination), contrast sensitivity testing may be a useful way of objectively demonstrating a functional deficit. Despite its advantages, it has not been widely adopted in clinical practice.

2 The Pelli–Robson contrast sensitivity letter chart (Fig. 14.9) is viewed at 1 metre and consists of rows of letters of equal size (spatial frequency of 1 cycle per degree) but with decreasing contrast of 0.15 log units for groups of three letters. The patient reads down the rows of letters until the lowest-resolvable group of three is reached.

Fig. 14.9 Pelli–Robson contrast sensitivity letter chart

Amsler grid

Page 599

The Amsler grid evaluates the 20° of the visual field centred on fixation (Fig. 14.10). It is principally useful in screening for and monitoring macular disease, but will also demonstrate central visual field defects originating elsewhere. Patients with a substantial risk of choroidal neovascularization should be provided with an Amsler recording chart for regular use at home.

Fig. 14.10 Amsler grid superimposed on the macula

(Courtesy of A Franklin)

Charts

There are seven charts, each consisting of a 10 cm square (Figs 14.11 and 14.12).

Chart 1 consists of a white grid on a black background, the outer grid enclosing 400 smaller 5 mm squares. When viewed at about one-third of a metre, each small square subtends an angle of 1°.

Chart 2 is similar to chart 1 but has diagonal lines that aid fixation in patients unable to see the central spot as the result of a central scotoma.

Chart 3 is identical to chart 1 but has red squares. The red-on-black design aims to stimulate long wavelength foveal cones. It is used to detect subtle colour scotomas and desaturation that may occur in toxic maculopathies, optic neuropathies and chiasmal lesions.

Chart 4 consisting only of random dots is used mainly to distinguish scotomas from metamorphopsia, as there is no form to be distorted.

Chart 5 consists of horizontal lines and is designed to detect metamorphopsia along specific meridians. It is of particular value in the evaluation of patients describing difficult reading.

Chart 6 is similar to chart 5 but has a white background and the central lines are closer together enabling more detailed evaluation.

Chart 7 exhibits a fine central grid, each square subtending an angle of a half degree, and is therefore more sensitive.

Fig. 14.11 Amsler grid chart

(Courtesy of A Franklin)

Fig. 14.12 Amsler charts 2–7

(Courtesy of A Franklin)

Technique

Page 600

The pupils should not yet have been dilated, and in order to avoid a photostress effect the eyes should not yet have been examined on the slit-lamp. A presbyopia correction should be worn if appropriate. The chart should be well illuminated and held at a comfortable reading distance. One eye is covered.

a The patient is asked to look directly at the central dot with the uncovered eye, to keep looking at this, and to report any distortion or waviness of the lines.

b Reminding the patient to maintain fixation on the central dot, ask if there are blurred areas or blank spots anywhere on the grid. Patients with macular disease often report that the lines are wavy whereas those with optic neuropathy often remark that some of the lines are missing or faint but not distorted.

c The patient is asked if he can see all four corners and all four sides of the square – a missing corner or border should raise the possibility of causes other than macular disease such as glaucomatous field defects or retinitis pigmentosa.

d The patient is given a recording sheet and pen and asked to draw any anomalies on a recording chart (Fig. 14.13).

e The other eye is tested.

Fig. 14.13 Amsler recording sheet shows wavy lines indicating metamorphopsia and a dense scotoma

Page 601

In clinical practice, testing is very commonly carried out simply using the recording chart, upon which patients can then illustrate any abnormality directly.

fundus Fluorescein angiography

Principles

1 Fluorescence is the property of certain molecules to emit light of a longer wavelength when stimulated by light of a shorter wavelength. The excitation peak for fluorescein is about 490 nm (in the blue part of the spectrum) and represents the maximal absorption of light energy by fluorescein. Molecules stimulated by this wavelength will be excited to a higher energy level and will emit light of about 530 nm (yellow-green; Fig. 14.14).

2 Fluorescein (sodium fluorescein) is an orange water-soluble dye that, when injected intravenously, remains largely intravascular and circulates in the blood stream. It undergoes both renal and hepatic metabolism and is excreted in the urine over 24–48 hours.

3 Fluorescein angiography (FA) involves photographic surveillance of the passage of fluorescein through the retinal and choroidal circulations following intravenous injection (Fig. 14.15).

4 Fluorescein binding. On intravenous injection, 70–85% of fluorescein molecules bind to serum proteins, the residue remaining unbound.

5 Outer blood–retinal barrier. The major choroidal vessels are impermeable to both bound and free fluorescein. However, the walls of the choriocapillaris contain multiple fenestrations through which free fluorescein molecules escape into the extravascular space. They then pass across Bruch membrane but on reaching the retinal pigment epithelium (RPE) are blocked by intercellular complexes termed tight junctions or zonula occludentes (Fig. 14.16).

6 Inner blood–retinal barrier is composed principally of the tight junctions between retinal capillary endothelial cells, across which neither bound nor free fluorescein can pass (Fig. 14.17A); the basement membrane and pericytes play only a minor role in this regard. Disruption of the inner blood–retinal barrier will permit leakage of both bound and free fluorescein into the extravascular space (Fig. 14.17B).

7 Filters of two types are used to ensure that blue light enters the eye and only yellow-green light enters the camera (Fig. 14.18).

a A cobalt blue excitation filter through which passes white light from the camera. The emerging blue light enters the eye and excites the fluorescein molecules in the retinal and choroidal circulations, which then emit light of a longer wavelength (yellow-green).

b A yellow-green barrier filter then blocks any reflected blue light from the eye, allowing only the emitted yellow-green fluorescent light to pass through.

8 Image capture in modern devices tends to be via the charge-coupled device (CCD) of a digital camera, with older cameras using fast black-and-white film. Digital image capture permits immediate picture availability, easy storage and access, image manipulation and enhancement.

Fig. 14.14 Excitation and emission of fluorescein

Fig. 14.15 Injection of fluorescein into the antecubital vein and its passage into the eye

Fig. 14.16 The outer blood–retinal barrier (ZO = zonula occludentes; BM = Bruch membrane)

Fig. 14.17 Inner blood–retinal barrier. (A) Intact; (B) disrupted (E = endothelial cell; B.M. = basement membrane; P = pericyte)

Fig. 14.18 Principles of fluorescein angiography

(Redrawn from PG Watson, BL Hazelman, CE Pavésio and WR Green, from The Sclera and Systemic Disorders, Butterworth-Heinemann, 2004)

Page 602

It must be emphasized that FA should only be performed if the findings are likely to influence management.

Technique

1 Preliminaries. A good quality angiogram requires adequate pupillary dilatation and clear media. The patient is asked about contraindications to FA.

• Fluorescein allergy is an absolute contraindication, and a history of a severe reaction to any allergen is a strong relative contraindication.

• Other relative contraindications include renal failure (lower the fluorescein dose if angiography is necessary), pregnancy, moderate-severe asthma and significant cardiac disease.

• It should be noted that allergy to iodine and seafood allergies are not contraindications to FA – fluorescein contains no iodine – but are absolute contraindications to indocyanine green (ICG) angiography as ICG contains iodine.

• Facilities and arrangements must be in place to address possible adverse events. This includes adequate staffing, resuscitation trolley, drugs for treatment of anaphylaxis, couch (or reclining chair) and a receiver in case of vomiting.

• The procedure is explained and formal consent taken. It is important to mention the common and serious adverse effects (Table 14.2), particularly the invariable skin and urine staining and the very common occurrence of nausea immediately following fluorescein injection.

2 Technique

Page 603

a The patient is seated comfortably in front of the fundus camera, and an intravenous cannula inserted. A standard venous cannula should be used rather than a less secure ‘butterfly’ winged infusion set. After cannulation, the line should be flushed with normal saline to check patency and exclude extravasation.

b Fluorescein, usually 5 mL of a 10% solution, is drawn up into a syringe. In eyes with opaque media, 3 mL of a 25% solution may afford better results.

c If not already obtained, colour photographs are taken.

d A ‘red-free’ image is captured (Table 14.3).

e If indicated, a pre-injection study is performed to detect autofluorescence (see below), with both the excitation and barrier filters in place.

f Fluorescein is injected over the course of a few seconds.

g Images are taken at approximately 1 second intervals, beginning 5–10 seconds after injection and continuing through the desired phases.

h If the pathology is monocular, control pictures of the opposite eye should still be taken, usually after the transit phase has been photographed in one eye.

i If appropriate, late photographs may be taken after 10 minutes to show leakage, and occasionally after 20 minutes.

j Stereo images may be helpful to demonstrate elevation, and are usually taken by manually repositioning the camera sideways or by using a special device (a stereo separator) to adjust the image; such images are actually pseudostereo, true stereo requiring simultaneous pictures from differing angles.

Table 14.2 Adverse events in fluorescein angiography

• Discolouration of skin and urine (invariable)

• Extravasation of injected dye (painful local reaction)

• Nausea very common, vomiting relatively uncommon

• Itching, rash

• Sneezing, wheezing

• Vasovagal episode or syncope (usually due to anxiety but sometimes to ischaemic heart disease)

• Anaphylactic and anaphylactoid reactions (1 : 2000 angiograms)

• Myocardial infarction (extremely rare)

• Death (1 : 220 000 in the largest study)

Table 14.3 Red-free fundus photography

• Image captured prior to fluorescein injection

• Taken with the yellow-green barrier filter in place, blocking red light

• Red structures appear black, heightening contrast

• Vasculature and haemorrhages easy to identify

• Visibility of retinal nerve fibre layer defects and other retinal details enhanced

Phases of the angiogram

Fluorescein enters the eye through the ophthalmic artery, passing into the choroidal circulation through the short posterior ciliary arteries and into the retinal circulation through the central retinal artery. Because the route to the retinal circulation is slightly longer than that to the choroidal, the latter is filled about 1 second before the former (Fig. 14.19). In the choroidal circulation, precise details are often not discernible, mainly because of rapid leakage of free fluorescein from the choriocapillaris and also because the melanin in the RPE cells blocks choroidal fluorescence. The angiogram consists of the following overlapping phases (Fig. 14.20).

1 The choroidal (pre-arterial) phase typically occurs 9–15 seconds after dye injection (longer in patients with poor general circulation) and is characterized by patchy lobular filling of the choroid due to leakage of free fluorescein from the fenestrated choriocapillaris. A cilioretinal artery, if present, will fill at this time because it is derived from the posterior ciliary circulation (Fig. 14.21).

2 The arterial phase starts about a second after the onset of choroidal fluorescence, and shows retinal arteriolar filling and the continuation of choroidal filling (Fig. 14.22A).

Page 604

3 The arteriovenous (capillary) phase shows complete filling of the arteries and capillaries with early laminar flow in the veins in which the dye appears to line the venous wall leaving an axial hypofluorescent strip (Fig. 14.22B). This phenomenon reflects initial drainage from posterior pole capillaries filling the venous margins, as well as the small-vessel velocity profile, with faster plasma flow adjacent to vessel walls where cellular concentration is lower.

4 The venous phase. Laminar venous flow (Fig. 14.22C) progresses to complete filling (Fig. 14.22D), with late venous phase featuring reducing arterial fluorescence. Maximal perifoveal capillary filling is reached at around 20–25 seconds in patients with normal cardiovascular function, and the first pass of fluorescein circulation is generally completed by approximately 30 seconds.

5 The late (recirculation) phase demonstrates the effects of continuous recirculation, dilution and elimination of the dye. With each succeeding wave, the intensity of fluorescence becomes weaker although the disc shows staining (Fig. 14.22E). Fluorescein is absent from the retinal vasculature after about 10 minutes.

6 The dark appearance of the fovea (Fig. 14.23A) is caused by three factors (Fig. 14.23B):

• Absence of blood vessels in the FAZ.

• Blockage of background choroidal fluorescence due to the high density of xanthophyll at the fovea.

• Blockage of background choroidal fluorescence by the RPE cells at the fovea, which are larger and contain more melanin and lipofuscin than elsewhere in the retina.

Fig. 14.19 Entry of fluorescein into the choroidal and retinal circulations

Fig. 14.20 Four phases of the fluorescein angiogram

Fig. 14.21 Choroidal phase shows patchy choroidal filling as well as filling of a cilioretinal artery

Fig. 14.22 Normal fluorescein angiogram. (A) Arterial phase shows filling of the choroid and retinal arteries; (B) arteriovenous (capillary) phase shows complete arterial filling and early laminar venous flow; (C) early venous phase shows marked laminar venous flow; (D) mid-venous phase shows almost complete venous filling; (E) late (elimination) phase shows weak fluorescence with staining of the optic disc

Fig. 14.23 Reasons for the dark appearance of the fovea

Causes of hyperfluorescence

Increased fluorescence may be caused by (a) enhanced visualization of a normal density of fluorescein, or (b) an increase in the fluorescein content of the tissues.

1 Autofluorescence compounds absorb blue light and emit yellow-green light in a similar fashion to fluorescein. It is imaged much more effectively by scanning laser ophthalmoscopy but can also be detected on standard fundus photography in exposed optic nerve head drusen (see Fig. 19.24B) and sometimes with lipofuscin in retinal drusen and other abnormalities such as astrocytic hamartoma (see Fig. 12.42D) and angioid streaks.

2 Pseudofluorescence (false fluorescence) refers to non-fluorescent reflected light visible prior to fluorescein injection; this passes through the filters due to the overlap of wavelengths passing through the excitation then the barrier filters. It is more evident when filters are wearing out.

3 A ‘window defect’ is caused by atrophy or absence of the RPE (Fig. 14.24A) as in atrophic age-related macular degeneration, full-thickness macular holes, RPE tears and some drusen. This results in unmasking of normal background choroidal fluorescence, characterized by very early hyperfluorescence which increases in intensity and then fades without changing size or shape (Fig. 14.24B and C).

4 Pooling in an anatomical space occurs due to breakdown of the outer blood–retinal barrier (RPE tight junctions):

a In the subretinal space as in central serous chorioretinopathy (Fig. 14.25A). This is characterized by early hyperfluorescence which, as the responsible leak tends to be only small, slowly increases in intensity and area, the maximum extent remaining relatively well-defined (Fig. 14.25B and C).

b In the sub-RPE space as in pigment epithelial detachment (PED – Fig. 14.26A). This is characterized by early hyperfluorescence which increases in intensity but not in size (Fig. 14.26B and C).

5 Leakage of dye is characterized by fairly early hyperfluorescence which increases in both area and intensity. It occurs as a result of breakdown of the inner blood–retinal barrier due to:

a Dysfunction or loss of existing vascular endothelial tight junctions as in background diabetic retinopathy, retinal vein occlusion, cystoid macular oedema (Fig. 14.27A) and papilloedema.

b Primary absence of vascular endothelial tight junctions as in choroidal neovascularization, proliferative diabetic retinopathy (Fig. 14.27B), tumours and some vascular anomalies such as Coats disease.

6 Staining is a late phenomenon consisting of the prolonged retention of dye in tissue such as drusen, fibrous tissue, exposed sclera, and the normal optic disc (see Fig. 14.22E), and is seen in the later phases of the angiogram, particularly after the dye has left the choroidal and retinal circulations.

Fig. 14.24 Hyperfluorescence caused by a transmission (window) defect associated with dry age-related macular degeneration

Fig. 14.25 Hyperfluorescence caused by pooling of dye in the subretinal space in central serous chorioretinopathy

Fig. 14.26 Hyperfluorescence caused by pooling of dye in the sub-RPE space in detachment of the RPE

Fig. 14.27 Causes of hyperfluorescence due to leakage. (A) Proliferative diabetic retinopathy; (B) cystoid macular oedema

(Courtesy of P Gili – fig. B)

Causes of hypofluorescence

Page 605

Page 606

Page 607

Reduction or absence of fluorescence may be due to: (a) optical obstruction (‘masking’ or blockage) of normal density of fluorescein in a tissue (Fig. 14.28) or (b) inadequate perfusion of tissue (‘filling defect’).

Page 608

1 Masking of retinal fluorescence. Pre-retinal lesions such as blood will block all fluorescence (Fig. 14.29). Deeper retinal lesions such as intraretinal haemorrhages and hard exudates will block only capillary fluorescence, sparing that from the larger retinal vessels.

2 Masking of background choroidal fluorescence is caused by all conditions that block retinal fluorescence as well as those which block only choroidal fluorescence:

a Subretinal or sub-RPE lesions such as blood.

b Increased density of the RPE that may be caused by congenital hypertrophy (Fig. 14.30).

c Choroidal lesions such as naevi.

3 Filling defects may result from:

a Vascular occlusion, which may involve the retinal arteries, veins or capillaries (‘capillary drop-out’ – Fig. 14.31A), or the choroidal circulation. FA is sometimes used to demonstrate optic nerve head filling defects as in anterior ischaemic optic neuropathy.

b Loss of the vascular bed as in myopic degeneration and choroideremia (Fig. 14.31B).

Fig. 14.28 Causes of blocked fluorescence

Fig. 14.29 Hypofluorescence caused by blockage of all fluorescence by a pre-retinal haemorrhage

Fig. 14.30 Hypofluorescence caused by blockage of background fluorescence by congenital hypertrophy of the retinal pigment epithelium

Fig. 14.31 Hypofluorescence caused by filling defects. (A) Capillary drop-out in diabetic retinopathy; (B) choroideremia

(Courtesy of C Barry – fig. B)

Systematic approach to reporting angiograms

A fluorescein angiogram should be interpreted methodically to optimize diagnostic accuracy. A suggested scheme:

a Note the clinical findings, including the patient’s age and gender, before assessing the angiogram.

b Indicate whether images of right, left or both eyes have been taken.

c Comment on any colour and red free images and on any pre-injection demonstration of pseudo- or autofluorescence.

d Looking at the post-injection images, indicate whether the overall timing of filling, especially arm-to-eye transit time, is normal.

e Briefly scan through the sequence of images in time order for each eye in turn, initially concentrating on the eye with the greatest number of shots as this is likely to be the one about which there is greater concern. On the first review, look for any characteristic major diagnostic/pathognomonic features; examples might include a lacy filling pattern or a smoke-stack (see later).

f Go through the run for each eye in greater detail, noting the evolution of any major features found on the first scan and then providing a description of any other findings using a methodical consideration of the causes of hyper- and hypofluorescence set out above.

Indocyanine green angiography

Principles

Page 609

1 Advantages over FA. Whilst FA is an excellent method of studying the retinal circulation, it is not helpful in delineating the choroidal vasculature, due principally to masking by the RPE. In contrast, the near-infrared light utilized in indocyanine green (ICG) angiography penetrates ocular pigments such as melanin and xanthophyll, as well as exudate and thin layers of subretinal blood, making this technique eminently suitable. An additional advantage is that about 98% of ICG molecules bind to serum proteins (mainly albumin) on entering the circulation.

2 Physiology. As the fenestrations of the choriocapillaris are impermeable to larger protein molecules, most ICG is retained within choroidal vessels, enhancing definition. Infrared light is also scattered less than visible light, making ICGA superior to FA in eyes with media opacities.

3 Image capture. ICG fluorescence is only 1/25th that of fluorescein so modern digital ICGA uses high-sensitivity videoangiographic image capture by means of a modified camera with infrared excitation (805 nm) and emission (835 nm) filters (Fig. 14.32). Alternatively, scanning laser ophthalmoscopy (SLO) systems provide high contrast images, with less scattering of light and fast image acquisition rates facilitating high quality ICG video.

4 The technique is similar to that of FA, but with an increased emphasis on the acquisition of later images (up to about 45 minutes) than FA. A dose of 25–50 mg in 1–2 mL water for injection is used.

5 The phases of a normal ICG angiogram are shown in Figure 14.33.

Fig. 14.32 Principles of indocyanine green angiography

(Redrawn from PG Watson, BL Hazelman, CE Pavésio and WR Green, from The Sclera and Systemic Disorders, Butterworth-Heinemann, 2004)

Fig. 14.33 Normal indocyanine green angiogram. (A) Early phase (up to 60 seconds post-injection) shows prominent choroidal arteries and poor early perfusion of the ‘choroidal watershed’ zone; (B) early mid-phase (1–3 minutes) shows more prominence of choroidal veins as well as retinal vessels; (C) late mid-phase (3–15 minutes) shows fading of choroidal vessels but retinal vessels are still visible; diffuse tissue staining is also present; (D) late phase (15–45 minutes) shows hypofluorescent choroidal vessels and gradual fading of diffuse hyperfluorescence

(Courtesy of S Milewski)

Adverse effects

ICGA is generally better tolerated than FA although the following problems may occur:

• Nausea, vomiting and urticaria are uncommon although anaphylaxis probably occurs with approximately equal incidence to FA.

• Serious reactions, including death are exceptionally rare. ICG contains iodide and so should not be given to patients allergic to iodine and possibly shellfish – newer iodine-free preparations such as infracyanine green are available.

Page 610

• It is also relatively contraindicated in liver disease (excretion is hepatic), and as with FA, in patients with a history of a severe reaction to any allergen, moderate or severe asthma and significant cardiac disease. The safety of ICG in pregnancy has not been established.

Page 611

Causes of hyperfluorescence

1 A ‘window defect’ as in FA.

2 Leakage from retinal or choroidal vessels, the optic nerve head or the RPE. This will give rise to tissue staining or to pooling.

3 Abnormal retinal or choroidal vessels with an anomalous morphology and/or exhibiting greater fluorescence than normal.

Hypofluorescence

1 Blockage (masking) of fluorescence. Pigment and blood are self-evident causes, but fibrosis, infiltrate, exudate and serous fluid also block fluorescence. A particular phenomenon to note is that in contrast to its FA appearance, a pigment epithelial detachment appears predominantly hypofluorescent on ICGA.

2 A ‘filling defect’ due to obstruction or loss of choroidal or retinal circulation.

Clinical indications

1 Exudative age-related macular degeneration (AMD). Conventional FA remains the primary method of diagnosis and assessment, but ICGA is a useful adjunctive investigation.

2 Polypoidal choroidal vasculopathy (PCV) in which ICGA is far superior to FA.

3 Chronic central serous chorioretinopathy in which it is often difficult to interpret areas of leakage on FA. However, ICGA shows choroidal leakage and the presence of dilated choroidal vessels. Previously unidentified lesions elsewhere in the fundus are also frequently visible using ICGA.

4 Posterior uveitis. ICGA can provide useful information beyond that available from FA in relation to diagnosis and to the extent of disease involvement.

5 Choroidal tumours may be imaged effectively but ICGA is inferior to clinical assessment for diagnosis.

6 Breaks in Bruch membrane such as lacquer cracks and angioid streaks are more effectively defined on ICGA than on FA.

7 When FA is contraindicated.

Optical coherence tomography

Definition

Optical coherence tomography (OCT) is a non-invasive, non-contact imaging system which provides high resolution cross-sectional images of the retina, vitreous and optic nerve head. Imaging of the anterior segment (AS-OCT) is also possible using the same technique although at present modified apparatus must be employed.

Principles

OCT is analogous to B-scan ultrasonography but uses near-infrared light interferometry rather than sound waves. Interferometry involves studying the pattern of interference created by the superposition of waves.

1 Low (short) coherence light is used, in which interference occurs over only micrometers. The imaging beam is split into a sampling path directed onto the tissue being imaged, and a reference path reflected from a mirror, and an image is constructed by analyzing the intensity of reflected reference light in combination with the intensity of reflectivity of different target tissue structures. Tissue reflecting more light will create more intense interference. Scattered light is excluded from the image.

2 In ‘time domain’ OCT, the position of the reference mirror is shifted towards and away from the source, essentially providing an axial scanning or A-scan function. Cross-sectional images are completed by scanning the sampling beam laterally across the target, yielding a two-dimensional data set usually displayed as a false-colour image.

3 Newer OCT instruments utilize ‘spectral/Fourier domain’ analysis, in which mechanical movement has been eliminated and the information for each point on the A-scan is collected simultaneously, speeding image acquisition and improving resolution. Spectral OCT also permits the ready construction of three-dimensional images and the study in relief of different retinal layers.

Indications

1 Diagnosis of cystoid macular oedema, macular holes, epiretinal membrane and vitreomacular traction, central serous chorioretinopathy, and to distinguish between long-standing retinal detachment and retinoschisis.

2 Monitoring progression of disease processes and response to treatment e.g. AMD, diabetic macular oedema, pre- and post-macular hole surgery.

3 Analysis of the optic nerve head and peripapillary retinal nerve fibre layer thickness, particularly in glaucoma diagnosis and monitoring.

4 Anterior segment OCT has an expanding range of clinical applications such as imaging the anterior chamber angle in glaucoma, the cornea (pachymetry, pre- and post-corneal refractive procedures, disease diagnosis and monitoring) and the lens.

Normal appearance

High reflectivity structures are depicted as red, intermediate as green-yellow and low reflectivity as blue-black (Fig. 14.34). High resolution OCT (Fig. 14.34B) has the ability to identify fine retinal structures such as the external limiting membrane and ganglion cell layer which are not visualized as clearly with standard resolution (Fig. 14.34A). Detailed quantitative information on retinal thickness can be displayed numerically and in a false-colour topographical map (Fig. 14.35).

Fig. 14.34 OCT displays. (A) Standard resolution of a normal macula in which most of the major retinal layers can be visualized; (B) high-resolution improves visualization of smaller structures such as the external limiting membrane (ELM) and ganglion cell layer (GCL); INL = inner nuclear layer; IPL = inner plexiform layer; IS/OS = photoreceptor inner and outer segment junction; NFL = nerve fibre layer; ONL = outer nuclear layer; OPL = outer plexiform layer; RPE = retinal pigment epithelium

(Courtesy of J Fujimoto)

Fig. 14.35 Stratus OCT numerical and false colour display of macular thickness in both eyes

(Courtesy of S Milewski)

Age-related macular degeneration

Introduction

Page 612

Age-related macular degeneration (AMD), also known as age-related maculopathy (ARM), is a degenerative disorder affecting the macula. It is characterized by the presence of specific clinical findings including drusen and RPE changes as early features with no evidence the signs are secondary to another disorder. Later stages of the disease are associated with impairment of vision.

Classification

1 Conventionally AMD is divided into two main types:

a Dry (non-exudative) AMD is the most common form, comprising around 90% of diagnosed disease; geographic atrophy (GA) is the advanced stage of dry AMD.

b Wet (exudative) AMD is much less common than dry, but is associated with more rapid progression to advanced sight loss. The main manifestations are (CNV) and pigment epithelial detachment (PED). Occasionally, the dry form can develop into the wet.

2 The International Age-Related Maculopathy Epidemiological Study Group (IARMESG) published guidance in 1995 in an attempt to standardize terminology, including referring to all signs of age-related macular change as age-related maculopathy (ARM), but usage continues to vary.

a Early AMD (‘early ARM’ in the IARMESG classification) is characterized by medium-large drusen, RPE hyperpigmentation and/or hypopigmentation. This stage is frequently known simply as ‘ARM’.

b Advanced AMD (IARMESG: ‘late ARM’ or just ‘AMD’) is more severe, with GA and/or CNV.

Epidemiology

• AMD is the most common cause of irreversible visual loss in industrialized countries. In the USA, it is responsible for around 54% of severe sight loss (better eye worse than 6/60) in Caucasian, 14% in Hispanic and 4% in black individuals. The prevalence increases with age and symptoms are rare in patients less than 50 years of age.

• In the UK, significant visual impairment (binocularly 6/18 or worse) from AMD affects about 4% of the population aged over 75 years and 14% of those over 90, with 1.6% over 75 having binocular acuity of less than 6/60.

• Patients with advanced AMD (late ARM) in one eye, or even moderate vision loss due to non-advanced AMD in one eye, have about a 50% chance of developing advanced AMD in the fellow eye within 5 years.

Page 613

Risk factors

AMD is multifactorial in aetiology, and is thought to involve a complex interaction between genetic and environmental factors.

1 Age is the major risk factor.

2 Race. Late ARM is more common in Caucasians than other races, despite a similar prevalence of early ARM.

3 Heredity. Family history is important. Variants in many genes have been implicated in AMD risk and protection such as chromosome 1q32 for complement factor H (CFH), which helps to protect cells from complement-mediated damage, the hemicentin gene on 1q24-25, and the ABCR gene on chromosome 1p (also important in Stargardt disease/fundus flavimaculatus).

4 Smoking roughly doubles the risk of AMD.

5 Hypertension and other cardiovascular risk factors are likely to be associated.

6 Dietary factors. High fat intake and obesity may promote AMD, with high antioxidant intake having a protective effect in some groups (see below).

7 Other factors such as cataract surgery, blue iris colour, high sunlight exposure, and female gender are suspected, but their influence remains less certain.

Drusen

Histopathology

1 Definition. Drusen (singular: druse) are extracellular deposits located at the interface between the RPE and Bruch membrane. The material of which they are composed has a broad range of constituents, and is thought to be derived from immune-mediated and metabolic processes in the RPE.

2 Role in pathogenesis of AMD is unclear. Age-related drusen are rare prior to the age of 40, but are common by the 6th decade. The distribution is highly variable, and they may be confined to the fovea, may encircle it or form a band around the macular periphery. They may also be seen in the peripheral and mid-peripheral fundus.

3 A distinction between ‘hard’ and ‘soft’ drusen is useful clinically and evident histopathologically (Fig. 14.36), although the underlying pathophysiological processes may be similar. Features associated with an increased risk of subsequent visual loss include large soft and/or confluent drusen, and associated focal hyperpigmentation of the RPE.

Fig. 14.36 Histology of drusen. (A) Hard drusen are discrete, homogeneous, eosinophilic nodular deposits lying between the RPE and the inner collagenous layer of Bruch membrane; (B) soft drusen are non-homogeneous, eosinophilic deposits with ill-defined margins

(Courtesy of J Harry)

Page 614

Clinical features

1 Hard drusen are well-defined and less than half a retinal vein width (< 63 µm) in diameter (Fig. 14.37A). Their presence as the only finding probably carries little increased risk of visual loss, and so they frequently are not included in definitions of ARM.

2 Soft drusen are less distinct and generally substantially larger than hard drusen (Fig. 14.37B). It has been suggested that the presence of more than five soft drusen might be taken as a defining feature of ARM. As soft drusen enlarge and become more numerous, they may coalesce giving a localized elevation of the RPE, a ‘drusenoid RPE detachment’ (Fig. 14.37C) – see below. Dystrophic calcification may develop in both types of drusen (Fig. 14.37D).

Fig. 14.37 Drusen. (A) Hard drusen; (B) soft drusen; (C) coalescence of soft drusen; (D) calcified drusen

Fluorescein angiography

Fluorescein angiographic findings depend on the state of the overlying RPE and on the affinity of the drusen for fluorescein. Hyperfluorescence can be caused by a window defect due to atrophy of the overlying RPE, or by late staining (Fig. 14.38). Hypofluorescent drusen masking background fluorescence are hydrophobic, with a high lipid content, and tend not to stain.

Fig. 14.38 (A) Soft drusen; (B) FA shows late hyperfluorescence

Lesions related to drusen

A number of conditions feature lesions similar to age-related drusen, and at least some may have a similar pathophysiological basis.

Page 615

1 Doyne honeycomb retinal dystrophy (malattia leventinese, autosomal dominant radial drusen) is an uncommon condition in which drusen appear during the 2nd–3rd decades (see Ch. 15); the genetic basis has been established for the majority of cases.

2 ‘Cuticular’ drusen, also known as ‘grouped early adult-onset’ or ‘basal laminar’ drusen (not to be confused with basal linear deposits, nodular thickenings of Bruch membrane seen in ARM), tend to be seen in relatively young adults. The lesions consist of small (25–75 µm) yellowish nodules (Fig. 14.39A) which tend to cluster and increase in number with time and can progress to serous PED. FA may give a ‘stars in the sky’ appearance (Fig. 14.39B). The condition has been linked to a variant of the CFH gene.

3 Type 2 membranoproliferative glomerulonephritis is a chronic renal disease that occurs in older children and adults. A minority of patients develop bilateral diffuse drusen-like lesions. The CFH gene has again been implicated.

Fig. 14.39 (A) Cuticular drusen; (B) FA shows hyperfluorescent spots – ‘stars in the sky’ appearance

(Courtesy of C Barry)

Prophylactic antioxidant supplementation in AMD

There is substantial evidence, notably from the Age-Related Eye Disease Study (AREDS), that taking high dose antioxidant vitamins and minerals on a regular basis can decrease the risk of AMD progression.

Recommendations for use

The recommendation was made in AREDS that individuals aged over 55 should undergo examination for the following high risk characteristics, and if one or more are present should consider antioxidant supplementation:

• Extensive intermediate-sized drusen.

• At least one large (≥ 125 µm) druse.

• Geographic atrophy in one or both eyes.

• Advanced AMD in one eye (greatest benefit in AREDS).

The reduction in risk of progression to further visual loss at 5 years is in the order of 25% for those taking supplements with the more advanced of these signs at baseline; supplements did not discernibly reduce progression in those with early or no AMD.

Page 616

Regimen

The regimen used in AREDS is set out below:

• 500 mg of vitamin C.

• 400 IU of vitamin E.

• 15 mg of beta-carotene.

• 80 mg of zinc, with 2 mg of copper (cupric oxide) to prevent zinc-induced copper deficiency.

Other considerations

Although supplementation is generally very safe, possible adverse effects include an increased risk of lung cancer with beta-carotene in smokers and ex-smokers, genitourinary tract problems with high zinc intake, and heart failure in people with vascular disease or diabetes (vitamin E).

• Macular xanthophylls (lutein and zeaxanthin) and omega-3 fatty acids may also be of benefit and many of the preparations available commercially now contain these. Trials such as AREDS2 are ongoing to try to establish the optimal nutrient combination.

• A liberal green leafy vegetable intake confers a lower risk of AMD, and for individuals with a strong family history of AMD and those with early AMD who do not meet the AREDS criteria, this may be a prudent lifestyle choice.

• Cessation of smoking should definitely be advised in these circumstances, and protective measures against exposure to excessive sunlight might be considered.

Non-exudative (dry) AMD

Diagnosis

1 Symptoms consist of gradual impairment of vision over months or years. Both eyes are usually affected, but often asymmetrically. Vision may fluctuate, and is often better in bright light.

2 Signs in approximately chronological order:

• Numerous intermediate-large soft drusen (Fig. 14.40A) which can become confluent.

• Focal hyper- and/or hypopigmentation of the RPE.

• Sharply circumscribed areas of RPE atrophy associated with variable loss of the retina and choriocapillaris (Fig. 14.40B and C).

• Enlargement of atrophic areas, within which larger choroidal vessels may become visible and pre-existing drusen disappear (GA; Fig. 14.40D). Visual acuity may be severely impaired if the fovea is involved. Rarely, CNV may develop in an area of GA.

• ’Drusenoid’ RPE detachment (see below). Many authorities regard the development of any form of PED as defining conversion to ‘wet’ AMD, regardless of evidence of CNV.

3 FA of atrophic areas shows a window defect due to unmasking of background choroidal fluorescence (Fig. 14.40E and F), if the underlying choriocapillaris is still intact. Exposed sclera may exhibit late staining.

Fig. 14.40 Atrophic age-related macular degeneration. (A) Drusen and mild RPE changes; (B) drusen and moderate retinal atrophy; (C) drusen and geographic atrophy; (D) geographic atrophy and disappearance of drusen; (E) FA arteriovenous phase shows slight hyperfluorescence; (F) FA late phase shows intense hyperfluorescence (window defects)

Management

1 Prophylaxis

• Antioxidant supplementation if indicated.

• Treatable risk factors should be addressed such as smoking, assessment of cardiovascular and dietary considerations, ocular sun protection measures and possibly a higher threshold for cataract surgery.

2 An Amsler grid should be provided for home use, with advice to self-test on a regular basis (perhaps weekly), seeking appropriate professional advice urgently in the event of any change.

3 Provision of low vision aids, and for patients with significant visual loss, certification as visually impaired where this is available may facilitate access to social and financial support.

4 Experimental surgery

• Miniature intraocular telescope implantation may be of benefit in selected cases.

• Retinal translocation surgery involving 360° retinotomy with retinal rotation, in conjunction with extraocular muscle surgery to correct torsion. There is a high risk of subsequent retinal detachment complicated by PVR. Accelerated degeneration of the ‘new’ macula may occur.

5 Laser photocoagulation of drusen leads to their disappearance, but does not seem to reduce the risk of progression to AMD.

Retinal pigment epithelial detachment

Pathogenesis

Pigment epithelial detachment (PED) from the inner collagenous layer of Bruch membrane is caused by disruption of the physiological forces maintaining adhesion. The basic mechanism is thought to be the reduction of hydraulic conductivity of a thickened and dysfunctional Bruch membrane, thus impeding movement of fluid from the RPE towards the choroid. Immune-mediated processes may also be important. The different types are discussed below.

Serous PED

1 Symptoms consist of blurred central vision and metamorphopsia. In some cases the former may be partly due to induced hypermetropia.

2 Signs

• An orange dome-shaped elevation with sharply delineated edges, often with a paler margin of subretinal fluid (Fig. 14.41A). Multiple lesions of diverse size may occur.

• A pigment band on the dome is thought to indicate chronicity.

• Associated subretinal fluid in an irregular distribution raises the suspicion of underlying CNV, as does the presence of chorioretinal folds.

• Sub-RPE or subretinal blood and retinal lipid are particularly suggestive of CNV.

• If no drusen are seen, polypoidal choroidal vasculopathy (PCV – see below) should be suspected.

Page 617

3 FA shows a well demarcated oval area of hyperfluorescence which increases in intensity but not in area with time: ‘pooling’ (Fig. 14.42B, and see Fig. 14.26). A notch in the circumscribed area may signify the presence of CNV.

4 ICGA demonstrates an oval area of hypofluorescence with a faint ring of surrounding hyperfluorescence (Fig. 14.42C). CNV is detected in more than 90% of cases as focal CNV (‘hot spot’) or diffuse ‘plaque’ CNV, or a combination of both.

5 OCT shows separation of the RPE from Bruch membrane by an optically empty area (Fig. 14.42D). CNV may be indicated in a serous PED by a notch between the main elevation and a second small mound.

Page 618

6 Natural course

• Patients over the age of 60 tend to have a worse prognosis. The eventual outcome tends to be poor (acuity 6/60 or less) whatever the specific course, although the speed of visual loss varies.

• RPE tear formation (see below) is associated with particularly rapid visual loss, as is haemorrhage from CNV.

• Chronicity with increasing atrophy and gradually worsening vision, or spontaneous resolution leaving an area of GA can occur.

• Spontaneous resolution without significant permanent visual loss is more common in younger patients.

• Up to about a third of patients develop clinical CNV within 2 years of diagnosis, although a much larger proportion has CNV on angiography.

7 Management

• Observation may be appropriate in patients without detectable CNV, especially those younger than 60 years of age.

• Intravitreal injection of vascular endothelial growth factor (VEGF) inhibitors, particularly ranibizumab and bevacizumab, may stabilize vision and improve the morphological features of vascularized PED, although this carries about a 10% risk of RPE tear formation.

• Combined PDT and intravitreal anti-VEGF has also been effective at stabilizing vision, though the RPE tear risk persists at around 10%.

• Combined photodynamic therapy (PDT) and intravitreal triamcinolone injection (IVTA) may be beneficial in some cases.

Fig. 14.41 Detachment of the RPE. (A) Clinical appearance; (B) FA shows hyperfluorescence; (C) ICGA shows hypofluorescence with a faint ring of surrounding hyperfluorescence; (D) OCT shows separation of the RPE from Bruch membrane

(Courtesy of P Gili – figs A and B; fig. C – A Bolton; C Barry – fig. D)

Fig. 14.42 Drusenoid detachment of the RPE. (A) Clinical appearance; (B) FA late phase shows moderate hyperfluorescence due to staining

Page 619

Fibrovascular PED

By definition (Macular Photocoagulation Study classification) fibrovascular PED represents a form of ‘occult’ CNV (see below).

1 Signs. The PED is much more irregular in outline and elevation than in serous PED.

2 FA shows markedly irregular granular or ‘stippled’ hyperfluorescence, with uneven filling of the PED, leakage and late staining.

3 ICGA demonstrates CNV more effectively.

4 OCT will demonstrate the PED, which will be optically denser than a serous PED in places, fibrous proliferation being shown as deeper scattered reflections. Subretinal fluid will be demonstrated.

5 Management is essentially as for serous PED with CNV.

Drusenoid PED

Drusenoid PED develops from confluent large soft drusen (Fig. 14.42A).

1 Signs. Shallow elevated pale areas with irregular scalloped edges that are often bilateral.

2 FA shows early diffuse hypofluorescence with patchy relatively faint early hyperfluorescence, progressing to moderate irregular late staining (Fig. 14.42B).

3 ICGA shows predominantly hypofluorescence.

4 OCT shows homogeneous hyperreflectivity within the PED, in contrast to the optically empty appearance of a serous PED. There will usually be no subretinal fluid.

5 Natural course. The outlook is usually better than other forms of PED, with only gradual visual loss, though probably around 75% still progress to develop GA and 25% CNV by 10 years from diagnosis. They often remain stable for long periods: at 3 years, only about one-third will have GA or CNV.

6 Management consists of observation in most cases, with no evidence to support the efficacy of any intervention.

Haemorrhagic PED

Virtually all eyes with haemorrhagic PED have underlying CNV or polypoidal choroidal vasculopathy (PCV): consider the latter if no drusen are seen.

1 Symptoms consist of sudden impairment of central vision.

2 Signs

• Elevated dark red dome-shaped lesion with a well-defined outline (Fig. 14.43A).

• The blood may break through into the subretinal space assuming a more diffuse outline and a lighter red colour (Fig. 14.43B).

3 FA exhibits dense masking of background fluorescence, but overlying vessels are visible.

4 Management of large haemorrhagic lesions is described below (‘haemorrhagic AMD’) but the prognosis for central vision is generally poor. CNV associated with a small haemorrhagic PED can be managed conventionally (see below for management of PCV).

Fig. 14.43 (A) Haemorrhagic detachment of the RPE; (B) blood has broken through into the subretinal space

Retinal pigment epithelial tear

1 Pathogenesis. An RPE tear may occur at the junction of attached and detached RPE if tangential stress becomes sufficient. Tears may occur spontaneously, following laser photocoagulation or PDT of a PED or associated CNV, or after intravitreal injection of anti-VEGF agent or steroid for any form of wet AMD. Older patients and large, irregular PEDs associated with CNV are at higher risk for this complication.

2 Presentation when the fovea is involved is with sudden worsening of central vision.

3 Signs. A crescent-shaped pale area of RPE dehiscence is seen, next to a darker area corresponding to the retracted and folded flap (Fig. 14.44A).

4 FA late phase shows hypofluorescence over the flap due to the thickened folded RPE, with adjacent hyperfluorescence over the exposed choriocapillaris where the RPE is absent. The two areas are separated by a well-defined linear border (Fig. 14.44B).

Page 620

5 OCT shows loss of the normal dome-shaped profile of the RPE in the PED, with hyper-reflectivity adjacent to the folded RPE (Fig. 14.44C).

6 The prognosis in subfoveal tears is poor although a minority of eyes maintain good visual acuity, particularly if the fovea is spared.

Fig. 14.44 Tear of the RPE. (A) A pale triangular area with surrounding blood and an adjacent darker area; (B) FA late phase shows relative hypofluorescence of the folded flap with adjacent hyperfluorescence where the RPE is absent; (C) OCT shows hyperreflectivity adjacent to the fold

(Courtesy of Moorfields Eye Hospital – figs A and B; C Barry – fig. C)

Choroidal neovascularization

Pathogenesis

1 Causative factors. Wet AMD is associated with CNV which comprises abnormal growth of a blood vessel complex through Bruch membrane from the choriocapillaris. The following interactive factors are thought to be important.

• The integrity of Bruch membrane and the RPE, inflammatory pathway components, localized hypoxia and the accumulation of metabolic products.

• The importance of locally-active cytokines involved with the promotion and inhibition of vessel growth has been realized, with particular attention focusing on vascular endothelial growth factor (VEGF), which binds to endothelial cell receptors, promoting angiogenesis and vascular leakage.

Page 621

• Complement factor H (CFH) is also thought to play a key role, as may the anti-angiogenic pigment epithelium-derived factor (PEDF).

2 Morphology. CNV membranes that remain sub-RPE are termed type 1, and when growth extends subretinally are known as type 2.

Clinical features

1 Presentation is with relatively rapid onset (often over days) of painless blurring of central vision including metamorphopsia. A positive scotoma may be described, particularly if haemorrhage has occurred.

2 Signs. Although the CNV itself can on occasion be visualized as a grey-green or pinkish-yellow lesion, most signs are caused by complications as follows:

• Localized subretinal fluid and occasionally CMO.

• Intra- and subretinal lipid deposition, sometimes extensive (Fig. 14.45A).

• Haemorrhage (Fig. 14.45B) which may be subretinal, preretinal, vitreous and associated with PED.

• Retinal and subretinal cicatrization (’disciform scar’ – Fig 14.45C) in an evolved or treated lesion.

• Exudative retinal detachment which may be extensive (Fig. 14.45D), resulting in total visual loss.

• The prognosis of untreated CNV is often very poor, with reduction of vision to the ‘hand movements’ level not uncommon.

Fig. 14.45 Complications of choroidal neovascularization. (A) Extensive lipid deposition; (B) bleeding; (C) ‘disciform’ scarring; (D) extensive exudative retinal detachment in end-stage disease

Fluorescein angiography

Page 622

FA is used primarily to confirm a suspected diagnosis of CNV prior to committing the patient to anti-VEGF treatment. Precise localization is now much less important and the distinction between classic and occult CNV membranes less relevant than formerly. Monitoring now consists predominantly of serial retinal thickness or volume assessment with OCT. The terminology commonly used to describe the FA appearances of CNV is derived from the Macular Photocoagulation Study (MPS) as follows:

1 Classic CNV is a well-defined membrane which fills with dye in a ‘lacy’ pattern during the early phases of dye transit (Fig. 14.46B), fluoresces brightly during peak dye transit (Fig. 14.46C), and then leaks into the subretinal space and around the CNV over 1–2 minutes. The fibrous tissue within the CNV then stains with dye exhibiting late hyperfluorescence (Fig. 14.46D). It is typically seen when CNV is subretinal rather than sub-RPE. Classic CNV is classified topographically according to its relation to the centre of the FAZ on FA as follows:

• Extrafoveal; 200–1500 µm from the centre.

• Juxtafoveal; 1–200 µm.

• Subfoveal (see Fig. 14.46); most membranes are subfoveal at presentation.

2 Occult CNV is used to describe CNV when its limits cannot be fully defined on FA (Fig. 14.47), typically when growth is between the RPE and Bruch membrane. Variants distinguished in the MPS classification are fibrovascular PED (see above) and ‘late leakage of an undetermined source’ (LLUS), areas of leakage in the late phase of the angiogram without classic CNV or fibrovascular PED. CNV may be said to be ‘predominantly’ or ‘minimally’ classic when the classic component is greater or less than 50% of the total lesion. The MPS classification also refers to ‘components’ of the lesion such as adjacent blood, serous PED or pigment which may be obscuring part of the CNV, and which are counted as part of the total lesion.

Fig. 14.46 FA of classic subfoveal CNV. (A) A few specks of blood at the fovea; (B) FA arteriovenous phase shows ‘lacy’ hyperfluorescence; (C) venous phase shows more intense hyperfluorescence; (D) late phase shows persistent hyperfluorescence due to staining

Fig. 14.47 FA of occult CNV. (A) Specks of blood at the fovea; (B–D) FA shows diffuse hyperfluorescence but the limits of the membrane cannot be defined

Indocyanine green angiography

ICGA demonstrates CNV as a focal hyperfluorescent ‘hot spot’ or ‘plaque’, and can be an extremely useful adjunct to FA for the following reasons:

• Increased sensitivity in the detection of CNV, for instance when the presence of low-density haemorrhage, fluid or pigment precludes adequate FA visualization (Fig. 14.48A–D).

• The distinction of CNV from other diagnoses having a similar clinical presentation, particularly PCV, retinal angiomatous proliferation (RAP) and central serous chorioretinopathy (CSR).

Page 623

• The delineation of occult CNV is now much less important since the introduction of anti-VEGF treatment. It may still have utility for combined modality treatment, and for patients who refuse intravitreal therapy.

• The identification of vascular feeder complexes supplying areas of CNV, again less important with the advent of anti-VEGF treatment.

• The management of RAP, when the identification of feeder vessels may facilitate their photoablation.

• The management of PCV, especially the localization of vascular complexes.

Fig. 14.48 ICGA of CNV. (A) Blood and fluid at the macula surrounded by hard exudates; (B–D) shows a small area of increasing hyperfluorescence (‘hot spot’) from underlying CMV

Optical coherence tomography

The major use of OCT in the management of CNV is in monitoring the response to treatment, for which it provides an accurate quantitative assessment. OCT has to date been of limited help in the diagnosis of CNV, though ongoing improvements in structural definition with newer high resolution devices, including three-dimensional imaging and the ability to construct separate images of different retinal layers, is likely to lead to increasing utility. Typically, CNV is shown as a thickening and fragmentation of the RPE/choriocapillaris high-reflectivity band. Subretinal and sub-RPE fluid (Fig. 14.49A), blood and scarring (Fig. 14.49B) are demonstrated.

Fig. 14.49 OCT. (A) CNV and subretinal fluid; (B) subretinal scarring

(Courtesy of C Barry)

Treatment with anti-VEGF agents

1 Principles. These agents prevent the VEGF-A form of the cytokine interacting with the relevant receptors on the endothelial cell surface and so retard or reverse CNV. They have become the predominant means of treatment for CNV, dramatically improving the visual prognosis. Intravitreal injection is the standard method of administration, notable risks including retinal detachment, damage to the lens, RPE tears and endophthalmitis. Elevated intraocular pressure and sterile uveitis may also occur. Systemically, there is a suggestion of a slightly increased incidence of stroke.

2 Indications. Predominantly classic, minimally classic and occult CNV subtypes all respond to anti-VEGF therapy, but benefit is only likely in the presence of active disease. Evidence for active CNV includes fluid or haemorrhage, leakage on FA, an enlarging CNV membrane, or deteriorating vision judged likely to be due to CNV activity. An eye with almost any level of vision may benefit, although better VA at the outset is associated with a better final VA and patients with only ‘hand movements’ should be assessed on an individual basis.

Page 624

3 Contraindications. The presence of a fibrotic disciform scar makes treatment extremely unlikely to be useful, even in the presence of active CNV. RPE tears may be a relative contraindication.

Ranibizumab (Lucentis)

Ranibizumab is a humanized monoclonal antibody fragment developed specifically for use in the eye. It non-selectively binds to and inhibits all isoforms of VEGF-A. The optimal timing of intravitreal injections is not yet clearly defined. The usual dose is 0.5 mg in 0.05 mL. Three main treatment strategies are currently adopted:

1 Regular monthly injection is the regimen adopted in initial major trials. Overall, around 95% of patients maintain vision regardless of lesion type, and 35–40% significantly improved, most markedly during the first 3 months.

2 Three initial monthly injections followed by monthly review with re-injection when deterioration occurs as assessed by VA (e.g. loss of 5 letters or more) and OCT (e.g. retinal thickness increase of 100 µm or more).

3 ’Treat and extend’ entails administering three initial injections at monthly intervals and then gradually increasing the period between injections until deterioration is evident. If possible a tailored interval is determined for each patient.

Bevacizumab (Avastin)

In contrast to ranibizumab, bevacizumab (Avastin) is a complete antibody and is very much cheaper; at present its use for AMD is ‘off label’.

Limited results suggest that it is effective – perhaps comparably to ranibizumab – and safe for intravitreal injection and trials are ongoing to compare the two. As bevacizumab is a larger molecule than ranibizumab, it may be retained in the vitreous for a longer period, so may need to be given less frequently. Speculatively it might also be associated with fewer systemic side-effects, at least in comparison with the 0.5 mg dose of ranibizumab; the bevacizumab 1-year stroke rate is around 0.5%, similar to the normal population rate. The dose of bevacizumab is usually 1.25 mg/0.05 mL or 2.5 mg/0.1 mL.

Pegaptanib (Macugen)

Page 625

Pegaptanib sodium was the first anti-VEGF agent approved by regulatory authorities for ocular treatment. Although offering visual outcomes superior to photocoagulation, the results are similar to outcomes with PDT (see below), and use of pegaptanib is considerably less widespread than other anti-VEGF agents.

Technique of intravitreal injection

1 Preparation

a The environment should be appropriate: an operating room or dedicated ‘clean room’ with adequate illumination.

b The procedure and its risks should be explained to the patient and appropriate consent obtained.

c Topical anaesthetic and mydriatic agents are instilled. Subconjunctival lidocaine 1% may be used to supplement the topical agent, particularly if a wider-gauge needle is used (no larger than 27-gauge).

d Some authorities recommend pre-injection topical antibiotics, typically for 3 days.

e Five per cent povidone iodine is applied to the ocular surface and at least three minutes allowed prior to injection (if allergic to iodine an alternative such as chlorhexidine can be used).

f Hands are washed using a standard surgical procedure, and sterile gloves donned.

g The periocular skin, eyelids and lashes are cleaned with 5–10% povidone iodine.

h As for other forms of intraocular surgery, an adhesive clear plastic sterile drape may be advisable, though is not universally used.

i A sterile speculum is placed in the eye.

2 Technique

a The patient is instructed to look away from the injection site – this is most commonly inferotemporal because of ease of access.

b A gauge is used to mark an injection site 3.5–4.0 mm posterior to the limbus (pars plana).

c The sterile pouch containing a pre-prepared syringe is opened, or a sterile syringe is used to draw up the appropriate volume of drug from a vial of ready-prepared drug. A needle (typically 30-gauge) on the syringe is primed to expel any air.

d Forceps can be used to stabilize the eye (and if wished to apply anterior traction to the conjunctiva so that the hole in the conjunctiva does not overlie the scleral injection site; a sterile cotton applicator is an alternative for this purpose).

e The needle is advanced perpendicularly through the sclera towards the centre of the eyeball, and the required volume of drug (0.5–1.0 mL) is slowly injected into the vitreous cavity. For larger gauge needles, stepping the entry site should be considered.

f The needle is removed and discarded. To minimize reflux including vitreous prolapse (‘vitreous wick syndrome’), a sterile cotton-tipped applicator can be rolled over the entry site as the needle is withdrawn.

g Broad-spectrum antibiotic drops are instilled immediately after the injection, and continued four times daily for at least 3 days.

h Elevated IOP can occlude the central retinal artery, and it is important routinely to ensure this remains perfused after the procedure by checking the patient’s vision (subjectively is adequate), directly visualizing the artery, or checking the IOP; the latter should always be considered in glaucoma patients. If the artery occludes, urgent paracentesis should be carried out; it may be helpful for the patient to lie down as this may improve blood flow.

Patients should be able to return to normal activity after 24 hours, but should be warned to seek advice urgently should they experience any deterioration in their vision or symptoms of inflammation.

Photodynamic therapy (PDT)

1 Principles. Verteporfin (Visudyne®) is a light-activated compound that is preferentially taken up by dividing cells including neovascular tissue. It is injected intravenously and is then activated focally by illumination with relatively low-energy light from a diode laser source at a peak absorption wavelength of the compound, leading to thrombosis. The main advantage of PDT is its sparing of healthy tissue (Fig. 14.50). The availability of anti-VEGF treatments has been associated with a dramatic reduction in the use of PDT. However, it remains useful in certain circumstances such as patient refusal of intravitreal treatment and as a component of combination therapy (see below).

2 Indications. In eyes with subfoveal predominantly classic CNV not larger than 5400 µm and a visual acuity of 6/60 or better. Other categories, particularly small occult and larger predominantly classic lesions, may also stabilize with PDT.

Page 626

3 Technique involves the intravenous infusion of verteporfin (6 mg/kg body weight) over 10 minutes, followed 5 minutes later by laser to an area 1000 µm larger than the greatest dimension of the CNV membrane for 83 seconds. Re-treatment is applied to areas of persistent or new leakage at 3-monthly intervals, with a mean of about five treatments in the first 2 years.

4 Side-effects include transient lower backache during infusion, transient decrease in vision, injection site reaction, and sensitivity to bright light for 24–48 hours. Preceding PDT with intravitreal steroid injection may confer an improved outcome.

Fig. 14.50 Photodynamic therapy. (A) Small dirty grey lesion at the fovea surrounded by blood; (B) FA venous phase shows hyperfluorescence from classic subfoveal CNV surrounded by a hypofluorescent ring; (C) greatest linear dimension of the lesion; (D) FA 3 months following successful treatment shows hypofluorescence at the site of the lesion

(Courtesy of S Milewski)

Combination therapies

Although anti-VEGF therapy has revolutionized the management of CNV, further investigation is attempting to achieve outcomes which are better still. A principal goal is a reduction in the frequency of intravitreal injections, in view particularly of the rare but potentially severe adverse effects. It is hoped that the use of more than one treatment modality in combination will facilitate this. Regimens include the combination of PDT with anti- VEGF, PDT with intravitreal steroid, steroid and anti-VEGF and ‘triple therapy’: with steroid/anti-VEGF/PDT.

Argon laser photocoagulation

Thermal laser ablation of CNV is now rarely used, though may still be suitable for treatment of classic extrafoveal membranes and some cases of PCV and RAP.

Experimental treatments

A range of additional treatment modalities are under investigation. Examples include:

• Brachytherapy with low-intensity strontium-90.

• VEGF Trap-Eye: an inhibitor binding all forms of VEGF-A to D as well as Placental Growth Factor, given intravitreally.

Page 627

• Inhibitors of other cytokines such as platelet-derived growth factor and integrin.

• Small interfering RNAs using gene-specific RNA strands to modify gene expression.

• VEGF receptor tyrosine kinase inhibitors.

• Sustained-release anti-VEGF systems, including microsphere encapsulation.

• Gene therapy utilizing adenoviral vectors; it is envisaged that delivery via this method would obviate the need for repeated intravitreal injections.

• Artificial retina implants.

Haemorrhagic AMD

The visual prognosis for most eyes with extensive subretinal or sub-RPE haemorrhage is poor, although the following should be considered.

1 Stop coumarin anticoagulant therapy, if appropriate, after liaising with the prescribing physician. Antiplatelet agents usually do not need to be discontinued.

2 Intravitreal anti-VEGF injection may be beneficial in some patients with thin (<1 mm) haemorrhage.

3 For massive haemorrhage in an eye with previously good vision, options include:

• Observation.

• Intravitreal anti-VEGF alone.

• Intravitreal recombinant tissue plasminogen activator (rTPA) and pneumatic (e.g. SF⁶ gas) haemorrhage displacement, with or without intravitreal VEGF inhibitor.

• Vitrectomy with subretinal rTPA combined with the above.

Retinal angiomatous proliferation

Retinal angiomatous proliferation (RAP) is a variant of exudative AMD in which the major component of the neovascular complex is initially located within the retina. The process may originate within the deep retinal capillary plexus or within the choroid, in the latter case with the early formation of retinal–choroidal anastomoses (RCA) without an underlying type 1 neovascular membrane. The disease is often bilateral and symmetrical and visual deterioration can be rapid and profound.

Diagnosis

1 Presentation is similar to that of AMD but PED and exudate are more common. Haemorrhages are also more common and tend to be superficial and multiple.

2 Stages

a Stage I shows intraretinal neovascularization (IRN) – intraretinal angiomatous proliferation. Dilated retinal vessels develop, typically accompanied by intra-, sub- and preretinal haemorrhage, oedema and exudate (Fig. 14.51A).

b Stage II manifests subretinal neovascularization (SRN) – proliferation extends posteriorly into the subretinal space and is associated with increasing oedema and exudate. An RCA and serous PED may be present.

Page 628

c Stage III in which CNV is clearly evident clinically or angiographically. A vascularized PED (V-PED), RPE tear or demonstrable RCA may be present. A disciform scar will often form.

3 OCT demonstrates the neovascularization as a hyper-reflective area. There will typically be cystoid macular oedema, subretinal fluid, and underlying elevation of the RPE.

4 FA is usually similar to purely occult or minimally classic CNV (Fig. 14.51B), but may show focal intraretinal hyperfluorescence.

5 ICGA is diagnostic in most cases, showing a hot spot in mid or late frames (Fig. 14.51C), and sometimes a characteristic ‘hairpin loop’.

Fig. 14.51 Retinal angiomatous proliferation. (A) Macular drusen and a small intraretinal haemorrhage at the macula; (B) FA early venous phase shows faint hyperfluorescence from a small frond of intraretinal neovascularization; (C) ICGA late phase shows hyperfluorescence of the frond (‘hot spot’)

(Courtesy of Moorfields Eye Hospital)

Treatment

Optimal management has yet to be determined. VEGF has been identified in excised RAP lesions, suggesting a useful role for anti-VEGF therapy, and reports to date show promising results. Limited success has been reported for other treatments including focal photocoagulation of the intraretinal component, PDT, intravitreal steroid and surgical section of supplying vessels. The condition may occasionally resolve spontaneously.

Polypoidal choroidal vasculopathy

Overview

Polypoidal choroidal vasculopathy (PCV), also known as posterior uveal bleeding syndrome, is an idiopathic choroidal vascular disease characterized by a dilated network of inner choroidal vessels with multiple terminal aneurysmal protuberances. It is more common in patients of African and East Asian ethnic origin than in whites and more common in women than men (5 : 1). The disease is often bilateral but asymmetrical in severity.

Diagnosis

1 Presentation is usually in late middle age (average age 60) with sudden onset unilateral visual impairment.

2 Signs

• Terminal swellings are frequently visible as reddish-orange nodules beneath the RPE in the peripapillary or macular area, and less commonly the periphery.

• Multiple recurrent serosanguineous retinal and RPE detachments (Fig. 14.52A).

Page 629

• Deterioration can be slow with intermittent bleeding and leakage, resulting in macular damage and visual loss.

• Up to 50% may have a favourable outlook, with eventual spontaneous resolution of exudation and haemorrhage.

3 ICGA is essential for diagnosis.

• Early stages show a network of large choroidal vessels with surrounding hypofluorescence.

• Polyp-like swellings (Fig. 14.52B and C) then appear on the larger vessels and rapidly begin to leak.

• The previously darker surrounding region becomes hyperfluorescent by the late phase (Fig. 14.52D).

• A cluster of grapes-like lesion may carry a higher risk of severe visual loss.

4 Differential diagnosis is mainly with AMD; the two conditions sometimes coexist.

Fig. 14.52 Polypoidal choroidal vasculopathy. (A) Haemorrhagic RPE detachment and macular exudate; (B–C) ICGA shows blockage by blood and hyperfluorescence of a polyp-like frond nasal to the fovea; (D) late hyperfluorescence due to staining

Treatment

The favourable prognosis without treatment in a substantial proportion of cases should be borne in mind.

• Asymptomatic polyps should usually be observed without treatment.

• Anti-VEGF agents appear to be less effective than in the CNV of AMD.

• PDT is more effective in PCV than AMD, although recurrence is common.

• Laser photocoagulation of feeder vessels or leaking polypoidal lesions may be effective in selected cases.

Age-related macular hole

Overview

Idiopathic age-related macular hole is a relatively common cause of central visual loss, with a prevalence of approximately 3 : 1000 individuals; with peak incidence of onset in females in the 7th decade. Presentation may be with impairment of central vision in one eye, or as a relatively asymptomatic deterioration, first noticed when the fellow eye is closed or at a routine sight test. The risk of involvement of the fellow eye at 5 years is around 10%.

Pathogenesis

The pathogenesis is incompletely defined, but current hypotheses suggest a role for the following:

• Oblique/anteroposterior traction via a persistent vitreofoveolar attachment following perifoveal vitreous separation.

• Tangential vitreoretinal traction.

• Predisposing involutional change of the inner retinal layers at the fovea.

Stages

1 Stage 1a: ‘Impending’ macular hole

a Signs: flattening of the foveal depression with an underlying yellow spot.

b Pathology: inner retinal layers (‘Müller cell cone’) detach from the underlying photoreceptor layer, with the formation of a schisis cavity.

2 Stage 1b: Occult macular hole

a Signs: a yellow ring (Fig. 14.53A) that may be associated with metamorphopsia or a mild decrease in visual acuity.

b Pathology: loss of structural support causes the photoreceptor layer to undergo centrifugal displacement (Fig. 14.54B).

3 Stage 2: Small full-thickness hole

a Signs: full-thickness hole less than 400 µm in diameter (Fig. 14.53B). The defect may be central, slightly eccentric or crescent-shaped.

b Pathology: a dehiscence develops in the roof of the schitic cavity, often with persistent vitreofoveolar adhesion (Fig. 14.54C).

4 Stage 3: Full-size macular hole

a Signs: full-thickness hole greater than 400 µm in diameter with a red base in which yellow-white dots may be seen. A surrounding grey cuff of subretinal fluid is usually present (Fig. 14.53C), and an overlying operculum (sometimes called a pseudo-operculum) may be visible. Visual acuity is often reduced to 6/60, although it is occasionally better, particularly in patients able to use eccentric fixation.

b Pathology: avulsion of the roof of the cyst (Fig. 14.54D) with an operculum (Fig. 14.54E) and persistent parafoveal attachment of the vitreous cortex.

5 Stage 4: Full-size macular hole with complete PVD

a Signs: as above.

b Pathology: the posterior vitreous is completely detached, often suggested (but not confirmed) by the presence of a Weiss ring.

Fig. 14.53 Macular hole (A) Occult – stage 1b; (B) small full-thickness – stage 2; (C) full-sized – stage 3

(Courtesy of J Donald M Gass, from Stereoscopic Atlas of Macular Diseases, Mosby 1997 – fig. A; S Milenkov – fig. B)

Fig. 14.54 High resolution OCT of macular hole. (A) Normal; (B) stage 1b shows attachment of the posterior hyaloid to the fovea, separation of a small portion of the sensory retina from the RPE in the foveolar region and intraretinal cystic changes; (C) eccentric stage 2 shows attachment of the vitreous to the lid of the hole and cystic changes; (D) stage 3 shows a full-thickness hole with intraretinal cystic spaces at its border; (E) stage 4 shows a full-thickness macular hole with intraretinal cystic spaces and an overlying pseudooperculum; (F) stage 4 after surgical closure

(Courtesy of J Fujimoto)

Investigations

Slit-lamp biomicroscopy alone is usually sufficient to make the diagnosis.

1 The Watzke–Allen test is performed by projecting a narrow slit beam over the centre of the hole both vertically and horizontally, preferably using a Goldmann fundus contact lens. A patient with a macular hole will report that the beam is thinned or broken. In contrast, patients with a pseudo- or lamellar hole, or a cyst, usually see a distorted beam of uniform thickness.

2 OCT is extremely useful in the diagnosis and staging (see Fig. 14.54).

3 FA in a full-thickness hole shows an early well-defined window defect due to xanthophyll displacement and RPE atrophy. Late frames may show the surrounding subretinal fluid as a hyperfluorescent halo (Fig. 14.55).

4 Amsler grid testing will show non-specific central distortion rather than a scotoma.

Fig. 14.55 (A) Stage 4 macular hole; (B) FA shows corresponding hyperfluorescence

(Courtesy of S Milewski)

Surgery

About 50% of stage 1 holes resolve following spontaneous vitreofoveolar separation, so these are managed conservatively. About 10% of full-thickness holes also close spontaneously, with variable visual improvement.

Page 630

1 Indications are stage 2 or worse holes provided visual acuity is less than 6/9. Superior results are usually achieved in holes which have been present for less than 6 months, but substantial visual improvement has been reported in long-standing cases.

2 Technique:

• Vitrectomy and peeling of the internal limiting membrane (ILM).

• Peeling is facilitated by staining the ILM with indocyanine green, trypan blue or triamcinolone. Although indocyanine green may be the most effective agent surgically, dose-related toxicity has been reported.

• Vitreomacular traction must be relieved, either by induction of a total PVD if not already present or by removal of the perifoveal vitreous.

• Gas tamponade is usual, but the necessity for strict extended postoperative face-down posturing (e.g. 50 minutes per hour for 7–14 days) is under review.

• Adjunctive agents such as autologous serum or platelets may be used.

• Because a cataract often develops following vitrectomy, combined surgery (‘phacovitrectomy’) may be considered.

Page 631

3 Results. The hole is closed in up to 100% of cases (Fig. 14.54F) and visual improvement occurs over the course of many months in 80–90% of eyes, with a final visual acuity of 6/12 or better in approximately 65%. Worsening of visual acuity occurs in up to 10% of eyes.

4 Complications are essentially those of vitrectomy. Occasionally, the hole may enlarge. Visual field defects may develop secondary to prolonged intraoperative retinal exposure to dry air.

Differential diagnosis of age-related macular hole

1 Other causes of full-thickness macular hole

a High myopia in the presence of posterior staphyloma may be associated with macular hole formation which can lead to retinal detachment. The subretinal fluid is generally confined to the posterior pole.

b Blunt ocular trauma may cause a macular hole as a result of either vitreous traction or commotio retinae (see Ch. 21).

2 Lesions with a similar appearance

a Pseudohole in a macular epiretinal membrane.

b Lamellar hole resulting from an abortive process of macular hole formation or in long-standing severe CMO (see Fig. 14.61).

c Foveal pseudocyst, typically idiopathic; in at least some patients may correspond to stage 1 macular hole.

d Vitreomacular traction syndrome.

e Solar retinopathy.

f Macular microhole (see below).

Macular microhole

Macular microhole is uncommon and may be easily overlooked without a careful history and examination. It is usually unilateral and has a favourable prognosis.

1 Presentation is with minimal symptoms, usually a central scotoma or reduced reading vision.

2 Signs. A very small, red, well-demarcated intraretinal foveal or juxtafoveal defect that remains stationary with long-term follow up (Fig. 14.56A).

Page 632

3 Higher-resolution OCT shows a well localized subtle defect gap in the photoreceptors and/or the RPE (Fig. 14.56B).

4 Differential diagnosis includes stage 1a age-related macular hole, solar retinopathy and blunt trauma.

Fig. 14.56 Macular microhole. (A) Small red foveal lesion; (B) OCT shows a subtle defect in the sensory retina

(Courtesy of C Barry – fig. B)

Central serous chorioretinopathy

Overview

Central serous chorioretinopathy (CSCR) is an idiopathic disorder characterized by a localized serous detachment of the sensory retina at the macula secondary to leakage from the choriocapillaris through focal, or less commonly diffuse, hyperpermeable RPE defects. CSCR typically affects one eye of a young or middle-aged Caucasian man; women with CSCR tend to be older. Imperfectly defined additional risk factors include psychological stress, type A personality, steroid administration, Cushing syndrome, systemic lupus erythematosus and pregnancy.

Clinical features

1 Presentation is with unilateral metamorphopsia that may be associated with micropsia, mild dyschromatopsia and decreased contrast sensitivity.

2 VA is usually reduced to 6/9–6/18, but often correctable to 6/6 with a weak convex lens, because elevation of the sensory retina gives rise to an acquired hypermetropia.

3 Signs

• A round or oval detachment of the sensory retina is present at the macula (Fig. 14.57A).

• The subretinal fluid may be clear (particularly in early lesions), turbid or fibrinous, and precipitates may be present on the posterior retinal surface.

• One or more abnormal depigmented RPE foci (sometimes small PEDs) of variable size may be visible within the neurosensory detachment.

• Small patches of RPE atrophy and hyperplasia elsewhere in the posterior pole may indicate the site of previous lesions.

• The optic disc should be examined to exclude a congenital pit.

Fig. 14.57 (A) Central serous chorioretinopathy; (B) OCT shows separation of the sensory retina from the RPE

(Courtesy of C Barry – fig. B)

Page 633

Investigations

1 Amsler grid confirms metamorphopsia corresponding to the neurosensory detachment.

2 OCT shows an optically empty neurosensory elevation (Fig. 14.57B). A RPE detachment or a deficit in the RPE may also be seen.

3 FA may show the following patterns:

• ’Smokestack’ manifests as an early hyperfluorescent spot (Fig. 14.58A) that progresses to form a vertical column by the late venous phase (Fig. 14.58B), followed by diffusion throughout the detached area.

• ‘Ink blot’ (most common) shows an early hyperfluorescent spot (Fig. 14.58C) that gradually enlarges (Fig. 14.58D).

• An underlying PED may be demonstrated.

• Multiple leakage points or diffuse leakage can be evident, particularly in chronic or recurrent disease.

4 ICGA. The early phase may show dilated or compromised choroidal vessels at the posterior pole. The mid-stage shows areas of hyperfluorescence due to choroidal hyperpermeability.

Fig. 14.58 FA in central serous chorioretinopathy. (A and B) ‘Smokestack’ appearance; (C and D) ‘ink blot’ appearance

(Courtesy of S Milewski)

Course

1 Spontaneous resolution occurs in most patients within 3–6 months, with return to near-normal or normal vision in more than 80% but recurrences occur in up to 50% of cases.

2 Chronic course lasting more than 12 months constitutes a minority and typically affects older patients.

• Prolonged detachment is associated with gradual photoreceptor and RPE degeneration with resultant visual impairment.

• Multiple recurrent attacks may also give a similar clinical picture.

• FA shows granular hyperfluorescence with one or more leaks (Fig. 14.59).

• CMO, CNV or RPE tears may develop in a minority of cases.

3 Bullous CSCR is characterized by large, single or multiple, serous retinal and RPE detachments that should not be misdiagnosed as rhegmatogenous retinal detachment or exudative detachment from another cause.

Fig. 14.59 FA in chronic central serous chorioretinopathy. (A) Venous phase shows granular hyperfluorescence; (B) late stage shows fading of fluorescence

(Courtesy of Moorfields Eye Hospital)

Management

1 Observation is appropriate in most cases.

2 Discontinue any corticosteroid treatment if possible, particularly in chronic, recurrent or severe cases.

3 Lifestyle change to reduce stress in selected cases.

4 Laser photocoagulation to the RPE leak speeds resolution but does not influence the final visual outcome or recurrence rate. It is advisable to wait at least 4 months before considering treatment of a first attack, and 1–2 months for recurrences. Thermal laser treatment is probably inadvisable if the leak is within the FAZ. Two or three low-intensity burns (200 µm, 0.2 second) are applied to the leakage site (Fig. 14.60) to produce mild greying of the RPE.

5 PDT may be considered in subfoveal leaks or chronic disease. Only 30% of the dose of verteporfin used for CNV, in conjunction with 50% light intensity is used by some authorities as first-line treatment.

6 Intravitreal anti-VEGF agents show some promise.

Fig. 14.60 Laser treatment of central serous chorioretinopathy. (A and B) Prior to treatment; (C and D) after successful treatment

(Courtesy of C Barry)

Cystoid macular oedema

Pathogenesis

Cystoid macular oedema (CMO) results from the accumulation of fluid in the outer plexiform and inner nuclear layers of the retina with the formation of cyst-like changes (Fig 14.61A). Fluid may initially accumulate intracellularly in Müller cells, with subsequent rupture. In long-standing cases, smaller microcystic spaces coalesce into larger cavities and may progress to lamellar hole formation at the fovea (Fig. 14.61B) with irreversible impairment of central vision. CMO is a non-specific manifestation of any type of macular oedema.

Fig 14.61 Cystoid macular oedema. (A) Histology shows cystic spaces in the outer plexiform and inner nuclear layer; (B) progression to lamellar hole formation

(Courtesy of J Harry and G Misson, from Clinical Ophthalmic Pathology, Butterworth-Heinemann 2001 – fig. A)

Diagnosis

Page 634

1 Presentation is with blurring and distortion. There is typically pre-existing disease such as diabetes indicating the cause, although CMO may be the presenting feature of a causative condition such as branch retinal vein occlusion.

Page 635

2 Signs

• Loss of the foveal depression, thickening of the retina and multiple cystoid areas in the sensory retina (Fig. 14.62A), best seen with red-free light using a fundus contact lens (Fig. 14.62B).

• Optic disc swelling is sometimes present.

• A lamellar hole may be visible.

• Features of associated disease.

3 Amsler chart testing demonstrates central blurring and distortion.

4 FA shows early hyperfluorescent spots due to leakage that progress to a characteristic petaloid pattern (Fig. 14.62C) of dye accumulation within microcystic spaces in the outer plexiform layer.

5 OCT shows cyst-like hyporeflective spaces within the retina, with retinal thickening and loss of the foveal depression (Fig. 14.62D). It is effective in demonstrating vitreoretinal traction and the presence of a lamellar hole. Serial examination is commonly used to assess the response to treatment.

Fig. 14.62 (A) Cystoid macular oedema; (B) red-free image; (C) FA late phase shows a ‘flower-petal’ pattern of hyperfluorescence; (D) OCT shows hyporeflective spaces within the retina, macular thickening and loss of the foveal depression

(Courtesy of J Donald M Gass, from Stereoscopic Atlas of Macular Diseases, Mosby 1997 – fig. A; P Gili – fig. B; C Barry – fig. D)

Causes

1 Ocular surgery and laser

• Clinically significant CMO occurs in probably 1–2% of eyes following phacoemulsification (see Ch. 9) but is higher following extracapsular extraction, particularly if associated with complications.

• Nd:YAG laser capsulotomy may cause CMO but the risk can be reduced if capsulotomy is delayed for 6 months or more after cataract surgery.

• Panretinal photocoagulation, if aggressive, may occasionally cause CMO.

• Other procedures such as retinal detachment surgery, glaucoma filtration surgery and corneal transplantation are sometimes followed by the development of CMO.

2 Retinal vascular disease such as diabetic retinopathy, retinal vein occlusion, hypertensive retinopathy, idiopathic retinal telangiectasis, retinal artery macroaneurysm and radiation retinopathy.

3 Inflammation such as intermediate uveitis, sarcoidosis, scleritis, birdshot retinochoroidopathy, multifocal choroiditis with panuveitis, toxoplasmosis, cytomegalovirus retinitis and Behçet syndrome.

4 Drug-induced due to topical prostaglandin derivatives and topical adrenaline derivatives in aphakic eyes.

5 Retinal dystrophies including retinitis pigmentosa, gyrate atrophy and dominant CMO.

6 Conditions involving vitreomacular traction such as epimacular membrane and vitreomacular traction syndrome (VMT).

7 CNV.

8 Fundus tumours such as retinal capillary haemangioma and choroidal haemangioma even when located well away from the macula.

9 Systemic disease such as multiple myeloma, leukaemia and chronic renal failure.

Epimacular membrane

Pathogenesis

An epimacular membrane (EMM), also known as a macular epiretinal membrane, cellophane maculopathy and macular pucker, is a sheet-like fibrocellular structure which develops on or above the surface of the retina. Proliferation of the cellular component and contraction of the membrane leads to visual symptoms, primarily due to retinal wrinkling, obstruction and localized elevation with or without CMO.

Classification

1 Idiopathic

• No apparent cause such as previous retinal detachment, surgery, trauma or inflammation.

• About 10% are bilateral.

• Predominant cellular constituent is glial cells, probably derived from the indigenous posterior hyaloid membrane (PHM) cell population (‘laminocytes’); these may be stimulated by the process of posterior vitreous detachment (PVD).

• Tend to be milder than secondary.

Page 636

asmitha636

2 Secondary

• Occur following retinal detachment surgery (most frequent cause of secondary EMM), retinal break, panretinal photocoagulation, retinal cryotherapy, retinal vascular disease, inflammation and trauma.

• Binocularity dependent on whether both eyes affected by causative factors.

• Cell type more varied; pigment cells are prominent – thought to be derived from the RPE.

• Tend to be more severe than idiopathic EMM.

Diagnosis

1 Presentation is with blurring and metamorphopsia although mild cases are often asymptomatic.

2 Signs

• VA is highly variable, depending on severity, but 6/9 is typical.

• An irregular translucent sheen is present at the macula in early EMM, often best detected using ‘red-free’ light (Fig. 14.63A).

• As the membrane thickens and contracts it becomes more obvious and typically causes mild distortion of blood vessels (Fig. 14.63B).

• Advanced EMM may give severe distortion of blood vessels, marked retinal wrinkling and striae and may obscure underlying structures (Fig. 14.63C).

• Associated findings may include macular pseudohole (Fig. 14.63D), CMO and small haemorrhages.

4 Amsler grid testing typically shows distortion.

5 OCT shows a highly reflective (red) surface layer associated with retinal thickening (Fig. 14.63E). OCT is useful in the exclusion of VMT, especially if CMO is present.

6 FA has been superseded by OCT for routine assessment of EMM, but highlights vascular tortuosity (Fig. 14.63F) and demonstrates any leakage.

Fig. 14.63 Epimacular membrane. (A) Translucent membrane seen with red-free light; (B) more obvious membrane with mild vascular distortion; (C) advanced membrane; (D) macular pseudohole seen with red-free light; (E) OCT shows high reflectivity and retinal thickening; (F) FA early venous phase highlights the vascular tortuosity

(Courtesy of L Merin – fig. A; C Barry – figs B, E and F; P Gili – fig. D)

Management

1 Observation if the membrane is mild and non-progressive. Spontaneous resolution of visual symptoms sometimes occurs, typically due to separation of the EMM from the retina.

2 Surgical removal of the membrane by vitrectomy to facilitate peeling usually improves or eliminates distortion (the main benefit), with an improvement in visual acuity of at least two lines in around 75% or more; in about a quarter VA is unchanged, and around 2% get worse.

Page 637

Degenerative myopia

Pathogenesis

Myopia is the result of complex hereditary and environmental factors. A refractive error of more than −6 dioptres constitutes high myopia, in which axial length is usually greater than 26 mm. ‘Pathological’ or degenerative myopia is characterized by progressive anteroposterior elongation of the scleral envelope associated with a range of secondary ocular changes, principally thought to relate to mechanical stretching of the involved tissues. It is a major cause of legal blindness, with maculopathy the most common cause of visual loss.

Diagnosis

1 A pale tessellated (‘tigroid’) appearance is due to diffuse attenuation of the RPE with visibility of large choroidal vessels (Fig. 14.64A).

2 Focal chorioretinal atrophy is characterized by visibility of the larger choroidal vessels and eventually the sclera (Fig. 14.64B).

3 Anomalous optic nerve head which may appear unusually small, large or anomalous with a ‘tilted’ conformation (Fig. 14.64B). Peripapillary chorioretinal atrophy is very common, and in milder cases taking the form of a temporal crescent of thinned or absent RPE exposing the choroid and/or sclera.

4 ‘Lacquer cracks’ are ruptures in the RPE-Bruch membrane-choriocapillaris complex characterized clinically by fine, irregular, yellow lines, often branching and criss-crossing at the posterior pole (Fig. 14.64C). They are present in around 5% of highly myopic eyes and can precede the development of CNV, retinal haemorrhages without CNV, and geographic atrophy. It is thought that the flaw in the RPE barrier constituted by the cracks provides a route through which choriocapillaris tissue can grow.

5 Lattice degeneration (see Ch. 16).

6 Subretinal ‘coin’ haemorrhages, which may be intermittent, may develop from lacquer cracks in the absence of CNV (Fig. 14.64D).

7 A Fuchs spot is a raised, circular, pigmented lesion at the macula developing after a subretinal haemorrhage has absorbed (Fig. 14.64E).

8 A staphyloma is an ectasia or bulging of the posterior sclera due to focal expansion and thinning (Fig. 14.64F). It occurs in about a third of eyes with pathological myopia, and is virtually always peripapillary or involves the macula. Staphyloma development can be associated with macular hole formation.

Fig. 14.64 High myopia. (A) Tessellated fundus; (B) focal chorioretinal atrophy and tilted disc; (C) lacquer cracks; (D) Fuchs spot; (E) ‘coin’ haemorrhage; (F) axial CT shows a left posterior staphyloma

Complications

1 Rhegmatogenous retinal detachment (RD) is much more common in high myopia, the pathogenesis including increased frequency of posterior vitreous detachment, lattice degeneration, asymptomatic atrophic holes, macular holes (Fig. 14.65A) and occasionally giant retinal tears. The prevalence of retinal detachment appears to be related to the severity of myopia.

2 CNV (Fig. 14.65B)

a Incidence 5–10% of highly myopic eyes develop CNV which may occur de novo, or more commonly in association with lacquer cracks or chorioretinal atrophy (see Fig. 14.64B and C).

b The visual prognosis is influenced by age, older patients tending to have a poorer outcome; the involved area tends to be smaller and shallower in younger patients with myopia-related CNV than in AMD.

c Treatment

• PDT has been the mainstay of therapy for subfoveal lesions, although with disappointing long-term results.

• Anti-VEGF therapy shows very promising results, with significant visual benefit; a lower injection frequency may be needed than for AMD but RD risk may be higher.

• Combination therapy with PDT is under investigation.

3 Foveal retinoschisis and macular retinal detachment without macular hole formation may occur in highly myopic eyes with posterior staphyloma, probably as a result of vitreous traction. Retinoschisis may be mistaken clinically for CMO, and is better characterized by OCT than biomicroscopy.

Page 638

4 Macular hole may occur spontaneously or after relatively mild trauma, and is associated with the development of retinal detachment very much more commonly than age-related idiopathic macular hole. As with foveoschisis, macular staphyloma may predispose; it is possible that myopic foveoschisis and myopic macular hole may be part of the same pathological process. Vitrectomy may be effective for both, but the best surgical technique remains undefined.

5 Peripapillary detachment is an asymptomatic, innocuous, yellow-orange, elevation of the RPE and sensory retina at the inferior border of the myopic conus (anomalous optic nerve head complex).

6 Cataract, which may be either posterior subcapsular or early onset nuclear sclerotic.

7 Glaucoma. There is an increased prevalence of primary open-angle glaucoma, pigmentary glaucoma and steroid responsiveness.

8 Amblyopia is uncommon but may develop when there is a significant difference in myopia between the two eyes.

9 Dislocation of the lens (natural or artificial) is a rare but well-recognized risk.

Page 639

Page 640

Page 641

Fig. 14.65 (A) shallow retinal detachment confined to the posterior pole, caused by a macular hole; (B) subretinal haemorrhage associated with CNV

(Courtesy of M Khairallah – fig. A)

Table 14.4 Systemic associations of high myopia

• Down syndrome

• Stickler syndrome

• Marfan syndrome

• Prematurity

• Noonan syndrome

• Ehlers–Danlos syndrome

• Pierre–Robin syndrome

Angioid streaks

Ocular considerations

Histology

Angioid streaks are crack-like dehiscences in brittle thickened and calcified Bruch membrane, associated with atrophy of the overlying RPE (Fig. 14.66A).

Fig. 14.66 Angioid streaks. (A) Histology shows a break in thickened Bruch membrane; (B) ‘peau d’orange’ and subtle angioid streaks; (C) advanced angioid streaks; (D) FA arteriovenous phase shows hypofluorescence of angioid streaks and three foci or CNV (arrows); (E) angioid streaks and optic disc drusen; (F) subretinal haemorrhage caused by a traumatic choroidal rupture

(Courtesy of J Harry and G Misson, from Clinical Ophthalmic Pathology, Butterworth-Heinemann 2001 – fig. A; P Saine – fig. C; S Milewski – fig. D)

Diagnosis

1 Signs

• ‘Peau d’orange’ (orange skin) mottled retinal pigmentation is common, and subtle associated angioid streaks may be overlooked (Fig. 14.66B).

• Grey or dark-red linear lesions with irregular serrated edges that lie beneath the normal retinal blood vessels that intercommunicate in a ring-like fashion around the optic disc and then radiate outwards from the peripapillary area, sometimes giving a ‘spider’s web’ appearance (Fig. 14.66C).

• The streaks tend to increase in width and extent slowly over time.

2 FA shows hyperfluorescent window defects due to RPE atrophy overlying the streaks, associated with variable associated hypofluorescence corresponding to RPE hyperplasia. FA is generally only indicated if CNV is suspected (Fig. 14.66D). The streaks can sometimes exhibit autofluorescence.

3 ICGA demonstrates the streaks as hyperfluorescent bands with brighter hyperfluorescent ‘pinpoints’ distributed within these.

4 Optic disc drusen are common (Fig. 14.66E).

Complications

Though angioid streaks are typically asymptomatic at first, the prognosis is guarded because visual impairment occurs in over 70% of patients as a result of one or more of the following:

1 CNV is by far the most common cause of visual loss (see Fig. 14.66D). Conventional thermal laser photocoagulation may be successful in extrafoveal lesions although there is a high risk of aggressive recurrence. Intravitreal injection of anti-VEGF agents stabilizes vision but CNV commonly recurs or develops at a new site, and studies are ongoing.

2 Choroidal rupture, which may occur following relatively trivial ocular trauma and result in a subretinal haemorrhage (Fig. 14.66F). Because eyes with angioid streaks are very fragile, patients should be warned against participating in contact sports and advised to use protective spectacles when appropriate.

3 Foveal involvement by a streak.

Systemic associations

Approximately 50% of patients with angioid streaks have one of the following conditions:

Pseudoxanthoma elasticum

Page 642

Page 643

Pseudoxanthoma elasticum (PXE) is by far the most common association of angioid streaks. Approximately 85% of patients develop ocular involvement, usually after the second decade of life. PXE is a hereditary disorder of connective tissue in which there is progressive calcification, fragmentation and degeneration of elastic fibres in the skin (Fig. 14.67A), eye and cardiovascular system. In almost all cases it is caused by homozygosity or compound heterozygosity for mutations of the ABCC6 gene, encoding for cellular transport protein; it is therefore an AR condition in which ocular manifestations are common but of variable severity. The signs are as follows.

• ‘Plucked chicken’ appearance of the skin resulting from small, yellowish macules, papules or plaques most commonly on the neck (Fig. 14.67B), axillae (Fig. 14.67C), antecubital fossae, groins and paraumbilical area, usually develops in children.

• Involved skin becomes progressively loose, thin and delicate.

• Calcification of elastic media and intima of arteries and heart valves, resulting in renal artery stenosis, intermittent claudication and mitral valve prolapse.

• Gastrointestinal haemorrhage, usually gastric in origin, is due to bleeding from fragile calcified submucosal vessels.

• Occasional bleeding from the urinary tract or cerebrovascular system.

• Severe angioid streaks (Grönblad–Strandberg syndrome).

Fig. 14.67 Pseudoxanthoma elasticum. (A) Histology shows thickened fragmented fibres in the dermis; (B) ‘chicken skin’ papules on the neck; (C) loose axillary skin

(Courtesy of J Harry and G Misson, from Clinical Ophthalmic Pathology, Butterworth-Heinemann 2001 – fig. A)

Occasional associations

1 Paget disease is a chronic, progressive metabolic bone disease characterized by excessive and disorganized resorption and formation of bone. Angioid streaks occur in only about 2%; it is thought that calcium binds to the elastin of Bruch membrane.

2 Haemoglobinopathies occasionally associated with angioid streaks are: homozygous sickle-cell disease (HbSS), sickle-cell trait (HbAS), sickle-cell thalassaemia (HbS thalassaemia), sickle-cell haemoglobin C disease (HbSC), haemoglobin H disease (HbH), homozygous beta-thalassaemia major, beta-thalassaemia intermedia and beta-thalassaemia minor. In these the reason for the abnormally brittle Bruch membrane is thought to be iron deposition.

3 Miscellaneous associations include familial hyperphosphataemia, idiopathic thrombocytopenic purpura, lead poisoning, haemochromatosis, Marfan syndrome and Ehlers–Danlos syndrome.

Choroidal folds

Pathogenesis

Choroidal folds are parallel grooves or striae involving the inner choroid, Bruch membrane, the RPE and sometimes the retina (chorioretinal folds). They are likely to develop in association with any process that induces sufficient compressive stress within the choroid, Bruch membrane, and retina. Primary mechanisms include choroidal congestion and scleral compression, and occasionally tissue contraction. Choroidal folds should be distinguished from retinal folds, which have a different pathogenesis (usually epimacular membrane).

Causes

Page 644

1 Idiopathic (‘congenital’) folds may be present in healthy, often hypermetropic, individuals in whom visual acuity is typically unaffected. The folds are usually bilateral. A syndrome of idiopathic acquired hypermetropia with choroidal folds has been described – in these patients elevated intracranial pressure should always be excluded even without evident papilloedema (see below) although a constricted scleral canal causing optic disc congestion has been proposed as an alternative mechanism in some patients.

2 Papilloedema. Choroidal folds may occur in patients with chronically elevated intracranial pressure such as idiopathic intracranial hypertension, and may be associated with reduced visual acuity which can occasionally be permanent.

3 Orbital disease such as retrobulbar tumours and thyroid ophthalmopathy may cause choroidal folds associated with impaired visual acuity.

4 Ocular disease such as choroidal tumours, posterior scleritis, scleral buckling for retinal detachment, in association with hypotony maculopathy, and as an early sign of underlying CNV.

Diagnosis

1 Presentation. The effect on vision is variable and dependent on the cause; many patients are asymptomatic, but some develop irreversible secondary degenerative retinal changes.

2 Signs

• Parallel lines, grooves or striae typically located at the posterior pole. The folds are usually horizontally orientated (Fig. 14.68A).

• The crest (elevated portion) of a fold is yellow and less pigmented as a result of stretching and thinning of the RPE and the trough is darker due to compression of the RPE.

• Clinical examination should be directed towards the exclusion of optic disc swelling, as well as other ocular or orbital pathology.

3 OCT allows differentiation between choroidal, chorioretinal and retinal folds.

4 FA shows hyperfluorescent crests as a result of increased background choroidal fluorescence showing through the stretched and thinned RPE and hypofluorescent troughs due to blockage of choroidal fluorescence by the compressed and thickened RPE (Fig. 14.68B); this facilitates distinction from retinal folds.

5 Additional imaging. Ultrasound, CT or MR scanning of the orbits or brain may be indicated. In elevated intracranial pressure or acquired hypermetropia with choroidal folds, an enlarged perineural space may be noted on B-scanning and MR of the optic nerves.

Fig. 14.68 (A) Choroidal folds; (B) FA shows alternating hypofluorescent and hyperfluorescent streaks

(Courtesy of JS Schuman, V Christopoulos, DK Dhaliwal, MY Kahook and RJ Noecker, from Lens and Glaucoma, in Rapid Diagnosis in Ophthalmology, Mosby 2008 – fig. B)

Hypotony maculopathy

Pathogenesis

Maculopathy is common in eyes developing hypotony, defined as IOP less than 5 mmHg. The most common cause is excessive drainage following glaucoma filtration surgery (higher risk with adjunctive antimetabolite), other causes including trauma (cyclodialysis cleft, penetrating injury), chronic uveitis (by directly impairing ciliary body function and by tractional ciliary body detachment due to cyclitic membrane), and retinal detachment. The development of secondary choroidal effusion may act to perpetuate the hypotony. With time the hypotonous process can itself lead to further damage, including sclerosis and atrophy of ciliary processes. Prolonged severe hypotony may lead to phthisis bulbi and loss of the eye. Treatment is aimed at restoring normal IOP.

Diagnosis

• Vision is variably affected. Delayed normalization of IOP may result in permanent visual impairment, though substantial improvement has been reported following hypotony reversal after several years.

Page 645

• Chorioretinal folds that may radiate outwards in branching fashion from the optic disc (Fig. 14.69); these are due to scleral collapse with resultant chorioretinal redundancy.

• Fine retinal folds radiating outwards from the foveola which may also show CMO.

• Other features that depend on aetiology, including a shallow anterior chamber, choroidal effusion, cataract, corneal decompensation, optic disc oedema, uveitis, wound leak or an inadvertent filtering bleb adjacent to a wound.

• Ultrasound biomicroscopy may show a cyclitic membrane or cyclodialysis cleft if there is clinical reason to suspect this.

• B-scan ultrasonography will demonstrate choroidal effusions.

• A-scan will show reduced axial length (compared with pre-operatively or with the fellow eye).

Fig. 14.69 Radial chorioretinal folds due to chronic hypotony

(Courtesy of P Gili)

Vitreomacular traction syndrome

1 Pathogenesis. In vitreomacular traction syndrome (VMT), the vitreous cortex remains attached to the fovea but is detached from the perifoveal region, with resultant exertion of persistent anteroposterior traction on the fovea. Changes at the posterior hyaloid membrane and retina similar to those involved in EMM and macular hole formation may be involved in the pathogenesis.

2 Presentation is usually in adult life with decreased vision, metamorphopsia, photopsia and micropsia. A mild variant of VMT has been reported following cataract surgery, when it may be mistaken for pseudophakic CMO.

3 Signs. The macula may show retinal surface wrinkling, distortion, EMM or CMO.

4 OCT shows incomplete posterior vitreous separation with persistent attachment of vitreous to the fovea (Fig. 14.70). The posterior vitreous surface often gives rise to a prominent signal.

5 Treatment in marked or progressive disease involves pars plana vitrectomy to relieve macular traction, usually with good results. Spontaneous resolution may also occur.

Fig. 14.70 OCT in vitreomacular traction shows incomplete posterior vitreous separation with persistent attachment at the fovea

(Courtesy of C Barry)

Idiopathic choroidal neovascularization

Idiopathic CNV is an uncommon condition which affects patients under the age of 50 years and is usually unilateral. The diagnosis is one of exclusion of other possible associations of CNV in younger patients such as angioid streaks, high myopia and chorioretinal inflammatory conditions such as presumed ocular histoplasmosis, MEWDS or PIC. The condition carries a better visual prognosis than that associated with AMD and in some cases spontaneous resolution may occur. The CNV lies predominantly above the RPE (type 2), often encircled by reactive RPE growth. PDT has met with variable results, but anti-VEGF agents show promise.

Solar retinopathy

1 Pathogenesis. Retinal injury is caused by photochemical effects of solar radiation by directly or indirectly viewing the sun (eclipse retinopathy).

2 Presentation is within 1–4 hours of solar exposure with unilateral or bilateral impairment of central vision and a small central scotoma.

3 VA is variable according to severity.

4 Fundus

• A small yellow or red foveolar spot which fades within a few weeks (Fig. 14.71A).

Page 646

• The spot is replaced by a sharply defined foveolar defect with irregular borders (Fig. 14.71B) or a lamellar hole.

5 OCT shows foveal thinning with a focal hyporeflective area, the depth of which correlates with the extent of visual acuity loss but which generally includes the photoreceptor inner and outer segments.

6 Treatment is not available.

7 Prognosis is good in most cases with improvement of visual acuity to normal or near-normal levels within 6 months; in a minority, significantly reduced vision persists.

Fig. 14.71 Solar maculopathy. (A) Yellow foveolar spot; (B) foveolar defect