Page 311

Chapter 10 Glaucoma

INTRODUCTION 312
Aqueous secretion 312
Aqueous outflow 312
Intraocular pressure 313
Overview of glaucoma 313
TONOMETRY 313
Goldmann tonometry 313
Other types of tonometry 315
GONIOSCOPY 316
Introduction 316
Indirect gonioscopy 316
Direct gonioscopy 318
Identification of angle structures 319
Grading of angle width 320
Pathological findings 323
EVALUATION OF THE OPTIC NERVE HEAD 323
Normal optic nerve head 323
Changes in glaucoma 324
IMAGING IN GLAUCOMA 327
Stereo disc photography 327
Confocal scanning laser tomography 327
Scanning laser polarimetry 330
Optical coherence tomography 331
Anterior chamber depth measurement 331
PERIMETRY 331
Definitions 331
Types of perimetry 333
Sources of error 333
Humphrey Field Analyzer 334
Short-wave automated perimetry 338
Frequency-doubling contrast test 338
OCULAR HYPERTENSION 338
PRIMARY OPEN-ANGLE GLAUCOMA 340
Introduction 340
Screening 341
Diagnosis 341
Visual field defects 342
Management 343
NORMAL-PRESSURE GLAUCOMA 346
PRIMARY ANGLE-CLOSURE GLAUCOMA 348
Introduction 348
Diagnosis 350
Treatment 351
CLASSIFICATION OF SECONDARY GLAUCOMA 353
Open-angle 353
Angle-closure 354
PSEUDOEXFOLIATION 355
Pseudoexfoliation syndrome 355
Pseudoexfoliation glaucoma 355
PIGMENT DISPERSION 357
Pigment dispersion syndrome 357
Pigmentary glaucoma 359
NEOVASCULAR GLAUCOMA 359
Introduction 359
Rubeosis iridis 360
Secondary open-angle glaucoma 361
Secondary angle-closure glaucoma 361
INFLAMMATORY GLAUCOMA 361
Introduction 361
Angle-closure glaucoma with pupillary block 362
Angle-closure glaucoma without pupillary block 362
Open-angle glaucoma 362
Treatment 363
Posner–Schlossman syndrome 364
LENS-RELATED GLAUCOMA 364
Phacolytic glaucoma 364
Phacomorphic glaucoma 366
Lens dislocation into the anterior chamber 367
Incarcerated lens in the pupil 367
TRAUMATIC GLAUCOMA 367
Hyphaema 367
Angle recession glaucoma 368
IRIDOCORNEAL ENDOTHELIAL SYNDROME 368
GLAUCOMA IN INTRAOCULAR TUMOURS 369
GLAUCOMA IN EPITHELIAL INGROWTH 371
GLAUCOMA IN IRIDOSCHISIS 371
PRIMARY CONGENITAL GLAUCOMA 372
Introduction 372
Diagnosis 374
Management 374
Differential diagnosis 376
IRIDOCORNEAL DYSGENESIS 376
Posterior embryotoxon 376
Axenfeld–Rieger syndrome 377
Peters anomaly 378
Aniridia 378
GLAUCOMA IN PHACOMATOSES 382
Sturge–Weber syndrome 382
Neurofibromatosis type 1 382
GLAUCOMA MEDICATIONS 383
Beta-blockers 383
Alpha-2 agonists 384
Prostaglandin analogues 384
Topical carbonic anhydrase inhibitors 385
Miotics 385
Combined preparations 386
Systemic carbonic acid inhibitors 386
Osmotic agents 386
LASER THERAPY 387
Argon laser trabeculoplasty 387
Selective laser trabeculoplasty 388
Nd:YAG laser iridotomy 388
Diode laser cycloablation 390
Laser iridoplasty 390
TRABECULECTOMY 391
Technique 391
Shallow anterior chamber 391
Failure of filtration 394
Late bleb leakage 394
Bleb-associated bacterial infection and endophthalmitis 395
NON-PENETRATING SURGERY 396
ANTIMETABOLITES IN FILTRATION SURGERY 397
DRAINAGE SHUNTS 399
Page 312

Introduction

Aqueous secretion

Aqueous humour is produced in two steps:

Formation of a plasma filtrate within the stroma of the ciliary body.
Formation of aqueous from this filtrate across the blood-aqueous barrier.

Two mechanisms are involved:

1 Active secretion by the non-pigmented ciliary epithelium accounts for the vast majority, and involves a metabolic process that depends on several enzyme systems, especially the Na+/K+ ATPase pump which secretes sodium ions into the posterior chamber.
2 Passive secretion by ultrafiltration and diffusion, which are dependent on the capillary hydrostatic pressure, oncotic pressure (colloid osmotic pressure exerted by proteins in blood plasma) and the level of IOP, is thought to play a minor role in the genesis of aqueous humour under normal conditions.

Aqueous outflow

Anatomy

1 The trabecular meshwork (trabeculum) is a sieve-like structure at the angle of the anterior chamber, through which 90% of the aqueous humour leaves the eye (Fig. 10.1). It is made up of the following three portions (Fig. 10.2):
a The uveal meshwork is the innermost portion and consists of cord-like endothelial cell-covered strands arising from the iris and ciliary body stroma, and extending from the root of the iris to Schwalbe line. The intertrabecular spaces are relatively large and offer little resistance to the passage of aqueous.
b The corneoscleral meshwork forms the larger middle portion which extends from the scleral spur to Schwalbe line. The layers are sheet-like and composed of connective tissue strands also with overlying endothelial-type cells. The intertrabecular spaces are smaller than those of the uveal meshwork, conferring greater resistance to flow.
c The juxtacanalicular (cribriform) meshwork is the outer part of the trabeculum, and links the corneoscleral meshwork with the endothelium of the inner wall of the Schlemm canal. This offers the major proportion of normal resistance to aqueous outflow, consisting of cells embedded in a dense extracellular matrix with narrow intercellular spaces.
2 Schlemm canal is a circumferential channel in the perilimbal sclera, bridged by septa. The inner wall is lined by irregular spindle-shaped endothelial cells containing infoldings (giant vacuoles) which are thought to convey aqueous via the formation of transcellular pores. The outer wall is lined by smooth flat cells and contains the openings of the collector channels which leave the canal at oblique angles and connect directly or indirectly with episcleral veins.
image

Fig. 10.1 Scanning electron micrograph of the trabecular meshwork

image

Fig. 10.2 Anatomy of outflow channels. (A) Uveal meshwork; (B) corneoscleral meshwork; (C) Schwalbe line; (D) Schlemm canal; (E) connector channels; (F) longitudinal muscle of the ciliary body; (G) scleral spur

Physiology

Aqueous flows from the posterior chamber via the pupil into the anterior chamber, from where it exits the eye by two different routes (Fig. 10.3):

1 Trabecular (conventional) route accounts for approximately 90% of aqueous outflow. The aqueous flows through the trabeculum into the Schlemm canal and is then drained by the episcleral veins. This is a bulk flow pressure-sensitive route so that increasing the pressure head will increase outflow. Trabecular outflow can be increased by drugs (miotics, sympathomimetics), laser trabeculoplasty and filtration surgery.
Page 313
2 Uveoscleral (unconventional) route accounts for the remaining 10% in which aqueous passes across the face of the ciliary body into the suprachoroidal space and is drained by the venous circulation in the ciliary body, choroid and sclera. Uveoscleral outflow is decreased by miotics and increased by atropine, sympathomimetics and prostaglandin analogues. Some aqueous also drains via the iris.
image

Fig. 10.3 Routes of aqueous outflow. (A) Trabecular; (B) uveoscleral; (C) iris

Intraocular pressure

The IOP is determined by the balance between the rate of aqueous secretion and aqueous outflow. The latter is in turn related to the resistance encountered in the outflow channels and to the level of episcleral venous pressure. The rate of aqueous outflow is proportional to the difference between the intraocular and episcleral venous pressure.

Concept of normal intraocular pressure

The distribution of IOP within the general population has a range of 11–21 mmHg. Although there is no absolute pathological point, 21 mmHg is considered the upper limit of normal and levels above this are viewed with suspicion. However, in some patients glaucomatous damage occurs with IOPs less than 21 mmHg (normal-tension or normal-pressure glaucoma) whilst others remain unscathed with IOPs up to 30 mmHg (ocular hypertension). Although the actual level of IOP is important in the development of glaucomatous damage, other factors are also significant.

Fluctuation

Normal IOP varies with the time of day, heartbeat, blood pressure level and respiration. The diurnal pattern varies, with a tendency to be higher in the morning and lower in the afternoon and evening. Normal eyes manifest a mean diurnal pressure variation of 5 mmHg; ocular hypertensive or glaucomatous eyes, however, exhibit a wider fluctuation. A single normal reading, particularly if taken during late afternoon, may therefore be misleading and it may be necessary to take several readings at different times of day (‘phasing’). In clinical practice phasing during the morning hours may be sufficient because 80% of patients peak between 8.00 a.m. and noon.

Overview of glaucoma

Definition

It is difficult to define glaucoma precisely, as it encompasses a diverse group of disorders. All forms of the disease have in common a potentially progressive and characteristic optic neuropathy which is associated with visual field loss as damage progresses, and in which intraocular pressure is usually a key modifying factor. On a molecular level, glaucoma of diverse aetiology is linked by the presence of endothelial leucocyte adhesion molecule-1 (ELAM-1), which indicates activation of a stress response in trabecular meshwork cells.

Epidemiology

Glaucoma affects up to 2% of those over the age of 40 years globally, and up to 10% over the age of 80; 50% may be undiagnosed. In a population of European or African ethnic origin, primary open-angle glaucoma (POAG) is the most common form. On a worldwide basis, primary angle-closure constitutes up to half of cases, with particularly high prevalence in individuals of Far Eastern descent.

Classification

Glaucoma may be congenital (developmental) or acquired. Sub-classification into open-angle and angle-closure types is based on the mechanism by which aqueous outflow is impaired with respect to the anterior chamber angle configuration. Distinction is also made between primary and secondary glaucoma; in the latter a recognizable ocular or non-ocular disorder contributes to elevation of IOP.

Tonometry

Goldmann tonometry

Principles

Goldmann applanation tonometry (GAT) is based on the Imbert–Fick principle, which states that for an ideal, dry, thin-walled sphere, the pressure (P) inside the sphere equals the force (F) necessary to flatten its surface divided by the area (A) of flattening (i.e. P = F/A). Theoretically, average corneal rigidity and the capillary attraction of the tear meniscus cancel each other out when the flattened area has the 3.06 mm diameter contact surface of the Goldmann prism (Fig. 10.4A), which is applied to the cornea with a variable amount of measurable force from which the IOP is deduced. The Goldmann tonometer is shown in Figure 10.4B. Disposable tonometer prisms and tonometer caps have been introduced to counter fears of infection from reusable prisms.

image

Fig. 10.4 Goldmann tonometry. (A) Physical principles; (B) tonometer

(Courtesy of J Salmon – fig. B)

Page 314

Technique

a The patient is positioned at the slit-lamp with the forehead firmly against the headrest.
b Topical anaesthetic and fluorescein are instilled into the conjunctival sac.
c With the cobalt blue filter and the brightest beam projected obliquely at the prism, the prism is centred in front of the apex of the cornea.
d The dial is preset between 1 and 2 (i.e. 10–20 mmHg).
e The prism is advanced until it just touches the apex of the cornea (Fig. 10.5A).
f Viewing is switched to the ocular of the slit-lamp.
g A pattern of two semicircle mires will be seen, one above and one below the horizontal midline, which represent the fluorescein-stained tear film touching the upper and lower outer halves of the prism.
h The dial on the tonometer is rotated to align the inner margins of the semicircles (Fig. 10.5B, right).
i The reading on the dial, multiplied by 10, gives the IOP.
image

Fig. 10.5 Applanation tonometry. (A) Tonometer touching the cornea; (B) fluorescein-stained semicircles during tonometry (see text)

Sources of error

Page 315
1 Inappropriate fluorescein pattern. Excessive fluorescein will make the mires too thick and the radius too small (Fig. 10.5B, left), when IOP will be overestimated, whereas insufficient fluorescein will make the semicircles too thin and the radius too large (Fig. 1.5B, centre) with consequent IOP underestimation.
2 Pressure on the globe from the examiner’s fingers, the patient squeezing the eyelids or from restricted extraocular muscles (e.g. thyroid myopathy) may result in an artificially high reading.
3 Central corneal thickness (CCT). Calculations of IOP by GAT assume that central corneal thickness is 520 µm, with minimal normal variation. If the cornea is thinner, underestimation of IOP may result, and if thicker an overestimation. Individuals with ocular hypertension tend to have corneas thicker than average, whereas those with normal-pressure glaucoma tend to have thinner corneas. Following refractive surgery procedures the cornea is both thinner and structurally altered such that IOP is likely to be underestimated. New methods of IOP measurement (see below) have been developed with the intention of reducing the effect of the various structural confounding variables.
4 Corneal oedema may result in artificial lowering of IOP, presumably due to a boggy softening; notably the associated increased CCT seems to be more than offset.
5 Astigmatism, if significant, may give distorted mires. If over three dioptres, the average reading of two can be taken with the prism rotated 90° for the second, or preferably, the prism is rotated so that the red line on the tonometer housing is aligned with the prescription of the minus axis.
6 Incorrect calibration of the tonometer can result in a false reading. It is therefore important to check this before each clinical session using the calibration arm supplied.
7 Wide pulse pressure. It is normal for there to be a small oscillation in IOP in time with the rhythm of ocular perfusion. If this ‘pulse pressure’ is substantial, the mid-point is taken as the reading.
8 Repeated readings over a short period will often be associated with a slight fall in IOP due to the massaging effect on the eye.
9 Other factors that may be associated with over-estimation of IOP include a tight collar and breath-holding, both of which obstruct venous return.

Other types of tonometry

1 Pneumotonometers are also based on the principle of applanation but, instead of using a prism, the central part of the cornea is flattened by a jet of air. The time required to sufficiently flatten the cornea relates directly to the level of IOP. Contact is not made with the subject’s eye and topical anaesthesia is not required, so it is particularly useful for screening in the community. Its main disadvantage is that it is accurate only within the low-to-middle range. The jet of air can startle the patient both with its apparent force and noise. A pneumotonometer may be non-portable (Fig. 10.6) or portable (Fig. 10.7A).
2 Reichert ocular response analyzer is a recently-developed form of pneumotonometer which measures IOP whilst attempting to compensate for corneal biomechanical properties by using two sequential measurements to assess corneal hysteresis, a function of viscous damping.
3 Dynamic contour tonometry (‘Pascal’) uses a solid state sensor and a corneal contour-matching surface to measure IOP. The instrument has been designed with the aim of measuring IOP relatively independently of corneal mechanical factors such as central corneal thickness. It is used on the slit lamp in a similar fashion to the Goldmann tonometer.
4 Perkins applanation tonometer uses a Goldmann prism adapted to a small light source. It is hand-held (Fig. 10.7B), and can therefore be used in bed-bound or anaesthetized patients.
5 Tono-Pen® is a hand-held, self-contained, battery powered, portable, miniaturized electronic contact tonometer (Fig. 10.7C). The probe tip contains a transducer that measures applied force. The instrument correlates well with the Goldmann. Its main advantage is the facility to measure IOP in eyes with distorted or oedematous corneas, through a bandage contact lens and in supine patients.
6 iCare® tonometer is a recently developed small hand-held device based on a new measuring principle, rebound or dynamic tonometry, in which a very light probe makes momentary contact with the cornea. Because only a very small force is applied to the cornea a topical anaesthetic is not required. The instrument can be used for self-monitoring (Fig. 10.7D) and screening in the community.
7 Schiotz tonometer uses the principle of indentation tonometry, in which the extent of corneal indentation by a plunger of known weight is measured; it is now seldom used in clinical practice.
image

Fig. 10.6 Non-portable pneumotonometer

image

Fig. 10.7 Portable tonometers. (A) Keeler pneumotonometer; (B) Perkins; (C) TonoPen®; (D) iCare®

(Fig. D, Courtesy of Mainline Instruments Ltd)

Page 316

Gonioscopy

Introduction

Overview

1 Gonioscopy is a method of evaluating the anterior chamber angle to provide information regarding the type of glaucoma. It can also be utilized therapeutically for procedures such as laser trabeculoplasty and goniotomy.
2 Other means of anterior chamber angle assessment such as high-frequency ultrasound biomicroscopy (UBM), and anterior segment optical coherence tomography (OCT), offer advantages in some aspects of angle analysis and may be used to supplement visual gonioscopic findings.

Optical principles

The angle of the anterior chamber cannot be visualized directly through the intact cornea because light from angle structures undergoes ‘total internal reflection’ at the anterior surface of the precorneal tear film (Fig. 10.8). Because the refractive index of a goniolens is similar to that of the cornea, it eliminates total internal reflection by replacing the tear film-air interface with a new tear film-goniolens interface. Light rays can then be viewed as they exit the contact lens. The two main types of goniolenses are indirect and direct (see below).

image

Fig. 10.8 Optical principles of gonioscopy; n = refractive index; i = angle of incidence

Page 317

Indirect gonioscopy

Indirect goniolenses use a mirror to reflect rays from the angle such that they exit the lens at much less than the critical angle. They provide a mirror image of the opposite angle and can be used only in conjunction with a slit lamp.

Non-indentation gonioscopy

1 Goniolenses
The classic Goldmann lens consists of three mirrors, one of which is specifically for gonioscopy (Fig. 10.9A); some goniolenses have one (Fig. 10.9B), two or four mirrors (Fig. 10.9C).
Lenses of similar basic structure but with modifications include the Magna View, Ritch trabeculoplasty, Thorpe four-mirror and the Khaw direct view.
Because the curvature of the contact surface of the lens is steeper than that of the cornea, a viscous coupling substance of refractive index similar to the cornea is required to bridge the gap between cornea and lens.
2 Technique
a It is essential that the examination takes place in a room in which the ambient illumination is very low – completely dark if possible.
b The size and intensity of the slit beam should be reduced to the absolute minimum compatible with an adequate view, in particular avoiding any of the beam being directed through the pupil.
c The patient is seated at the slit-lamp and advised that the lens will touch the eye but will not usually cause discomfort.
d A drop of local anaesthetic such as proxymetacaine 0.5% is instilled.
e The forehead must be kept against the headband and both eyes should remain open.
f A drop or two of coupling fluid (an artificial tear preparation such as hypromellose 0.3%) is placed on the contact surface of the lens.
g The patient is asked to look upwards and the lens is inserted rapidly so as to avoid loss of the coupling fluid. The patient then looks straight ahead.
h Indirect gonioscopy gives an inverted view of the angle opposite to the mirror.
i Once the initial examination has been performed and the findings noted, increasing the level of illumination may help in defining the angle structures.
j When the view of the angle is obscured by a convex iris, it is possible to see ‘over the hill’ by asking the patient to look in the direction of the mirror. Only slight movement is permissible, otherwise the structures will be distorted and a closed angle may appear open.
k Excessive pressure with a non-indentation lens narrows the angle appearance (in contrast to the effect of pressure during indentation gonioscopy – see below). Excessive pressure also causes folds in the cornea which compromise the clarity of the view.
l In some eyes, suction on the cornea from the lens may artificially open the angle; awareness of the need to avoid retrograde as well as anterograde pressure on the lens will help to avoid.
image

Fig. 10.9 Goldmann goniolens. (A) Three mirror; (B) single mirror; (C) four mirror

Indentation gonioscopy

1 Goniolenses include the Zeiss (generally used along with the detachable Unger fork handle – Fig. 10.10), Posner (modified Zeiss with fitted handle) and Sussman (no handle), all of which have four-mirror gonioprisms.
The contact surface of the lenses has a curvature flatter than that of the cornea, negating the need for a coupling substance.
The lenses do not stabilize the globe and are not suitable for laser trabeculoplasty.
2 Technique
a The first stages are as set out above for non-indentation gonioscopy.
b Indentation is performed by gently pressing the lens posteriorly against the cornea (Fig. 10.11A); this forces aqueous into the angle, pushing the peripheral iris posteriorly.
c If the angle is closed only by apposition between the iris and cornea it will be forced open, allowing visualization of the angle recess (Figs. 10.11B).
d If the angle is closed by adhesions between the peripheral iris and cornea – peripheral anterior synechiae (PAS – Fig. 10.12A) it will remain closed (Fig. 10.12B).
e Dynamic gonioscopy can be invaluable in helping to define the structures in angles which are difficult to assess, such as in distinguishing an extensive or double highly-pigmented Schwalbe line from the pigmented trabecular meshwork.
image

Fig. 10.10 (A) Zeiss goniolens; (B) lens in place

image

Fig. 10.11 Indentation gonioscopy in appositional angle closure. (A) Total angle closure prior to indentation; (B) during indentation the entire angle becomes visible (arrow) and the cornea develops folds

(Courtesy of W Alward, from Color Atlas of Gonioscopy, Wolfe 1994)

image

Fig. 10.12 Indentation gonioscopy in partial synechial angle closure. (A) Total angle closure prior to indentation; (B) during indentation part of the angle becomes open (small arrow) and the remainder remains closed (large arrow) due to PAS

(Courtesy of W Alward, from Color Atlas of Gonioscopy, Wolfe 1994)

Direct gonioscopy

Direct goniolenses work by constructing the viewing surface of the lens in a domed or slanted configuration such that exiting light rays strike the contact lens/air interface at a steeper than critical angle so that they will pass through to the observer. This approach is called ‘direct’ because light rays from the angle are viewed directly, without reflection inside the lens. They do not require a slit-lamp and are used with the patient in the supine position, typically under general anaesthesia in the evaluation and surgical treatment of infantile glaucoma.

Page 319
1 Goniolenses
a Diagnostic lenses include the Koeppe, a dome-shaped direct diagnostic goniolens which comes in several sizes (Fig. 10.13).
b Surgical lenses (Fig. 10.14) used for angle surgery include the Medical Workshop, Barkan and Swan-Jacob.
2 Technique
a Gonioscopy is performed with the patient in the supine position (note that this may deepen the angle appearance) in conjunction with an operating or hand-held microscope (or magnifying loupes) and an additional illumination source if necessary.
b The technique cannot be used with a desktop slit-lamp so clarity, illumination and variable magnification are not comparable with indirect lenses.
image

Fig. 10.13 Koeppe goniolenses

image

Fig. 10.14 Surgical goniolenses. (A) Medical Workshop; (B) Barkan; (C) Swan-Jacob

Identification of angle structures

Figure 10.15 shows the anatomy of angle structures.

1 Schwalbe line is the most anterior structure, appearing as an irregular opaque line. Anatomically it demarcates the peripheral termination of Descemet membrane and the anterior limit of the trabeculum. It may be barely discernible, particularly in younger patients. In contrast, there may be pigment deposits on or anterior to Schwalbe line (Sampaolesi line) that may make interpretation of the angle structures difficult.
Page 320
2 The corneal wedge is useful in locating an inconspicuous Schwalbe line. Using a narrow slit beam, two distinct linear corneal reflections can be identified (see Fig. 10.15), one on the inner and one on the outer corneal surface; the outer reflection will arc round across the corneoscleral interface – due to the sclera being opaque – to meet the inner reflection at the apex of the corneal wedge which coincides with the Schwalbe line.
3 The trabeculum extends from Schwalbe line to the scleral spur and has an average width of 600 µm. In younger people it has a ground-glass appearance and appears to have depth. The anterior non-functional part lies adjacent to Schwalbe line and has a whitish colour. The posterior, pigmented functional part lies adjacent to the scleral spur and has a greyish-blue translucent appearance in the young. Trabecular pigmentation is rare prior to puberty, but in older eyes involves the posterior trabeculum to a variable extent, most marked inferiorly. Patchy trabecular pigmentation in a suspiciously narrow angle raises the possibility of intermittent iris contact.
4 Schlemm canal may be identified in the non-pigmented angle as a slightly darker line deep to the posterior trabeculum. Blood can sometimes be seen in the canal (Fig. 10.16), either physiologically (sometimes due to excessive pressure on the episcleral veins with a goniolens), or in the presence of low intraocular or raised episcleral venous pressure.
5 The scleral spur is the most anterior projection of the sclera and the site of attachment of the longitudinal muscle of the ciliary body. Gonioscopically it is situated immediately posterior to the trabeculum and appears as a narrow, dense, often shiny, whitish band.
6 The ciliary body stands out just behind the scleral spur as a pink to dull-brown to slate-grey band. Its width depends on the position of iris insertion and it tends to be narrower in hypermetropic eyes and wider in myopic eyes. The angle recess represents the posterior dipping of the iris as it inserts into the ciliary body.
7 Iris processes are small extensions of the anterior surface of the iris which insert at the level of the scleral spur and cover the ciliary body to a varying extent (see Fig. 10.16). They are present in about one-third of normal eyes and are most prominent during childhood and in brown eyes. The processes should not be confused with PAS which are generally broader.
8 Blood vessels running in a radial pattern at the base of the angle recess are often seen in normal eyes. Pathological blood vessels run randomly in various directions. As a general principle, any blood vessel that crosses the scleral spur onto the trabecular meshwork is abnormal.
image

Fig. 10.15 Normal angle structures

(Courtesy of W Alward, from Color Atlas of Gonioscopy, Wolfe 1994)

image

Fig. 10.16 Blood in Schlemm canal (arrow), and iris processes

(Courtesy of J Schuman, V Christopoulos, D Dhaliwal, M Kahook and R Noecker, from Lens and Glaucoma, in Rapid Diagnosis in Ophthalmology, Mosby 2008)

Page 321

Grading of angle width

Shaffer system

The Shaffer system records the angle in degrees between two imaginary lines tangential to the inner surface of the trabeculum and the anterior surface of the iris about one-third of the distance from its periphery. In practice, the angle is graded by many according to the visibility of various structures. The system assigns a numerical grade to each quadrant of the angle as below (Fig. 10.17); it should be borne in mind that most angles are narrowest superiorly.

1 Grade 4 (35–45°) is the widest angle, characteristic of myopia and aphakia, in which the ciliary body can be visualized with ease.
2 Grade 3 (25–35°) is an open angle in which at least the scleral spur can be identified.
3 Grade 2 (20°) is a moderately narrow angle in which only the trabeculum can be identified.
4 Grade 1 (10°) is a very narrow angle in which only Schwalbe line, and perhaps also the top of the trabeculum, can be identified.
5 Slit angle is one in which there is no obvious iridocorneal contact but no angle structures can be identified.
Page 322
6 Grade 0 (0°) is a closed angle due to iridocorneal contact and is recognized by the inability to identify the apex of the corneal wedge. Indentation gonioscopy will distinguish ‘appositional’ from ‘synechial’ angle closure (see Fig. 10.20).
image

Fig. 10.17 Grading of angle width

Other systems

1 Spaeth system is detailed but underused. It describes consideration of the position of the iris insertion, the angular approach and curvature of the peripheral iris.
2 Scheie classification refers to the angle structures visible and allocates a Roman numeral accordingly. In contrast to common clinical use, in the original system a higher numeral (e.g. IV) actually signifies a narrower angle.
3 The van Herick method (Table 10.1) uses the slit-lamp alone to estimate the anterior chamber angle width:
A thin bright slit beam is set approximately perpendicularly to the corneal surface (offset from the optics by about 60°) to the patient’s temporal side for each eye.
The beam is used to estimate the ratio of the corneal thickness to the most peripheral part of the anterior chamber (see Figs 10.40B and 10.41B).
This method provides a useful approximation in a majority of patients and has utility as a screening tool and as an aid to assessment in angles which are difficult to interpret on gonioscopy. However, it overestimates angle width in a proportion of patients, particularly those with a ‘plateau iris’ conformation.

Table 10.1 Van Herick method for anterior chamber angle assessment

image
image

Fig. 10.41 Primary angle-closure. (A) Closed angle; (B) van Herick grade 1

(Courtesy of L MacKeen fig. A; J Schuman, V Christopoulos, D Dhaliwal, M Kahook and R Noecker, from Lens and Glaucoma, in Rapid Diagnosis in Ophthalmology, Mosby 2008 – fig. B)

Pathological findings

1 Peripheral anterior synechiae
Primary angle-closure glaucoma.
Anterior uveitis.
Iridocorneal endothelial (ICE) syndrome.
2 Neovascularization
Neovascular glaucoma.
Fuchs heterochromic cyclitis.
Chronic anterior uveitis.
3 Hyperpigmentation
Pigment dispersion syndrome.
Pseudophakic pigment dispersion.
Pseudoexfoliation syndrome.
Blunt ocular trauma.
Anterior uveitis.
Following acute angle-closure glaucoma.
Following YAG laser iridotomy.
Iris melanoma.
Iris pigment epithelial cysts.
Naevus of Ota.
4 Trauma
Angle recession.
Trabecular dialysis.
Cyclodialysis.
Foreign bodies.
5 Blood in the Schlemm canal
Carotid-cavernous fistula and dural shunt.
Sturge–Weber syndrome.
Obstruction of the superior vena cava.
Physiological variant.

Evaluation of the optic nerve head

Normal optic nerve head

Neuroretinal rim

The neuroretinal rim (NRR) is the tissue between the outer edge of the cup and the optic disc margin. The normal rim has an orange or pink colour and a characteristic configuration in most healthy eyes: the inferior rim is the broadest followed by the superior, nasal and temporal (the ‘ISNT’ rule).

Optic disc size

Optic disc size is important in deciding if a cup-disc (C/D) ratio is normal. Normal median vertical diameter for non-glaucomatous discs is 1.50 mm in a Caucasian population. It can be assessed clinically as follows:

a A narrow slit beam is focussed on the disc using a fundus biomicroscopy lens.
b The height of the beam is adjusted until it matches the distance between the superior and inferior limits of the NRR (not the scleral rim surrounding the neural tissue), and the diameter in millimetres read from the slit-lamp graticule.
c A correction factor may be necessary, dependent on the lens used (Table 10.2). Refractive error affects measurement only minimally, although myopia above −8 dioptres may distort the result.

Table 10.2 Correction factors for estimating optic disc diameter

Lens Correction factor
Volk 60 D × 0.88–1.0
Nikon 60 D Around 1.0
Volk 90 D ×1.3
Volk 78 D ×1.1
Goldmann 3-mirror ×1.27

Cup–disc ratio

The C/D ratio indicates the diameter of the cup expressed as a fraction of the diameter of the disc; the vertical rather than the horizontal ratio is generally used in clinical practice. The NRR occupies a relatively similar cross-sectional area in different eyes.

Small discs have small cups with a median C/D ratio of about 0.35 (Fig. 10.18A)
Large discs have large cups with a median C/D ratio of about 0.55 (Fig. 10.18B).
Only 2% of the population have a C/D ratio greater than 0.7.
In any individual, asymmetry of 0.2 or more between the eyes should also be regarded with suspicion, though it is critical to exclude a difference in overall disc size.
image

Fig. 10.18 Normal discs. (A) Small disc with a low C/D ratio; (B) large disc with a higher C/D ratio

(Courtesy of S Farley, T Cole and L Rimmer)

Page 324

Changes in glaucoma

In many cases, it is not possible to decide with certainty whether an individual optic disc is glaucomatous. The clinical findings and results of investigation should be considered together to guide management. Glaucomatous damage results in characteristic signs involving (a) the optic nerve head, (b) the peripapillary area and (c) the retinal nerve fibre layer.

Optic nerve head

The spectrum of disc damage in glaucoma ranges from highly localized tissue loss with notching of the NRR to diffuse concentric enlargement of the cup, as well as changes in vasculature. Pathological cupping is caused by an irreversible decrease in the number of nerve fibres, glial cells and blood vessels. A documented increase in cup size is always significant. If an eye with a small optic disc and correspondingly small cup develops glaucoma, the cup will increase in size, but even in the presence of substantial damage may still be smaller than that of a large physiological cup, so overall disc diameter must be taken into account as discussed above. Assessment of the thickness, symmetry and colour of the NRR is of substantial importance (see ‘ISNT’ rule above).

Subtypes of glaucomatous damage

The appearance and pattern of disc damage may correlate with subtypes of glaucoma and provide clues as to the pathogenic mechanisms involved. Four ‘pure’ glaucomatous disc appearances have been described, although the majority of discs are unclassifiable.

1 Focal ischaemic discs are characterized by focal superior and/or inferior polar notching (Fig. 10.19A) which may be associated with localized field defects with early threat to fixation.
2 Myopic disc with glaucoma refers to a tilted (obliquely inserted), shallow disc with a temporal crescent of parapapillary atrophy, together with features glaucomatous damage (Fig. 10.19B). Discs with degenerative myopia are excluded. Dense superior or inferior scotomas threatening fixation are common. These discs tend to occur in younger male patients.
3 Senile sclerotic discs are characterized by a shallow, saucerized cup and a gently sloping NRR, variable peripapillary atrophy and peripheral visual field loss (Fig. 10.19C). Patients are older (both genders equally), and the disc type is associated with ischaemic heart disease and hypertension.
4 Concentrically enlarging discs (verified by serial monitoring) are characterized by uniform NRR thinning (Fig. 10.19D) and are frequently associated with diffuse visual field loss. At presentation IOP is often significantly elevated.
image

Fig. 10.19 Specific subtypes of glaucomatous damage. (A) Type 1 – focal ischaemic; (B) type 2 – myopic; (C) type 3 – senile sclerotic; (D) type 4 – concentrically enlarging

Non-specific signs of glaucomatous damage

Other disc signs of glaucomatous damage, though of variable specificity, include:

1 Baring of circumlinear blood vessels is a sign of early thinning of the NRR. It is characterized by a space between a superficial blood vessel that runs from the superior or inferior aspects of the disc towards the macula, and the disc margin (Fig. 10.20A). ‘Overpass cupping’, in which there is loss of NRR underlying vessels, leaving space between the bridging vessels and the remaining nerve tissue, is similar.
2 Bayoneting is characterized by double angulation of a blood vessel (Fig. 10.20B). With NRR loss, a vessel entering the disk from the retina may angle sharply backwards into the disk and then turn towards its original direction to run across the lamina cribrosa.
Page 325
Page 326
3 Collaterals between two veins at the disc (Fig. 10.20C), similar to those following central retinal vein occlusion are rare.
4 Loss of nasal NRR (Fig. 10.20D) is a sign of moderately advanced damage; a space may develop between the NRR and the central retinal vasculature.
5 Lamina dot sign occurs in advancing glaucoma. The gray dot-like fenestrations in the lamina cribrosa become exposed as the NRR recedes (Fig. 10.20E). The fenestrations sometimes appear linear, and this itself may be a sign of advanced damage, indicating distortion of the lamina. The dot sign is not specific for glaucomatous atrophy, and may be seen in normal eyes.
6 Disc haemorrhages often extend from the NRR onto the retina, most commonly inferotemporally (Fig. 10.20F). Their presence is a risk factor for glaucoma and they may also be a marker of inadequate control. They can also occur in healthy individuals as well as in patients with hypertension, diabetes and those taking antiplatelet agents.
7 ‘Sharpened edge’ or ‘sharpened rim’ is a sign of advancing damage. As NRR is lost adjacent to the edge of the disc, the disc margin contour assumes a sharper angle backwards. Bayoneting of vessels is often seen at a sharpened edge. This should not be confused with a ‘sharpened nasal polar edge’, which refers to the sharp angulation of the NRR at the nasal margin of a focal vertical polar notch.
image

Fig. 10.20 Non-specific signs of glaucomatous damage. (A) Inferior baring of circumlinear blood vessels; (B) inferior bayoneting; (C) collaterals; (D) loss of nasal neuroretinal rim; (E) lamellar dots; (F) disc haemorrhage

Page 327

Peripapillary changes

Peripapillary atrophy surrounding the optic nerve head may be of significance in glaucoma (Fig. 10.21) and may be a sign of early damage in patients with ocular hypertension.

1 Alpha (outer) zone is characterized by superficial retinal pigment epithelial changes. It tends to be larger and possibly more common in glaucomatous eyes.
2 Beta (inner) zone is characterized by chorioretinal atrophy. It is larger and more common in glaucoma.
image

Fig. 10.21 Parapapillary changes. Zone beta (black arrow); zone alpha (white arrow)

It is important to note the distinction from the scleral lip or rim, the white band of exposed sclera central to the beta zone

Retinal nerve fibre layer

In glaucoma subtle retinal nerve fibre layer (RNFL) defects precede the development of detectable optic disc and visual field changes; their onset often follows disk haemorrhages. Two patterns occur: (a) localized wedge-shaped defects (Fig. 10.22A) and (b) diffuse defects that are larger and have indistinct borders. Red-free (green) light increases the contrast between normal retina and defects on slit-lamp biomicroscopy and typically makes identification easier (Fig. 10.22B). Defects may be easier to detect on (black-and-white) photographs than during clinical examination. Optical coherence tomography (OCT) and scanning laser polarimetry are highly effective means of quantifying the RNFL. It should be noted that RNFL defects are not specific to glaucoma, and can be seen in a range of neurological disease, as well as apparently normal individuals.

image

Fig. 10.22 Retinal nerve fibre layer defects. (A) Superotemporal wedge-shaped defect; (B) same eye seen with a green filter

(Courtesy of P Gili)

Imaging in glaucoma

Stereo disc photography

Stereo photography has historically been regarded as the reference standard in optic disc imaging, and remains a valuable option. The images are taken by repositioning the camera slightly between shots, either manually or using a stereo separator built into the camera.

Page 328

Confocal scanning laser tomography

1 Physics. The scanning laser ophthalmoscope (SLO) produces images of the optic nerve head and retina by scanning a diode laser beam across tissues at progressively varying depths, utilizing the confocal principle to capture information from only a thin layer at a time and thereby building a three-dimensional image. The Heidelberg Retinal Tomograph (HRT) has been widely adopted in the assessment of glaucoma patients. A HRT3 version has been introduced, though the HRTII is still the most commonly used.
2 Indications
To distinguish normal from glaucomatous eyes by comparison against a normative database (Moorfields regression analysis).
To monitor disease progression in individual glaucoma patients.
The macula can also be examined, although the OCT has more commonly been adopted for this.
3 Technique. Keratometry values must be entered and significant (>1.0 dioptre) astigmatism corrected by means of a cylindrical lens. High quality images can usually be acquired without pupillary dilatation and through mild–moderate lens opacity. After image capture, the operator must manually mark the contour line defining the edge of the neuroretinal rim.
4 Display. Images, data and analysis can be examined on a computer screen or printed. Sample monocular printouts from the HRTII are shown (Figs 10.23 and 10.24).
Images of the disc and peripapillary retina are shown at the top of the display.
In the topographic image (top left) the cup is represented in red, the neuroretinal rim in green and the connecting slope in blue.
Page 329
The reflectivity false colour image (top right) is divided into six sectors. Both the neuroretinal rim (green and blue on the topographic image) and the disc area (green, blue and red) are assessed using Moorfields regression analysis, taking into account age and overall disc size. A green tick within a sector indicates it is within normal limits, a yellow exclamation mark borderline and a red cross outside normal limits.
The two cross-sectional images (top centre and middle left) show the amount of cupping in the vertical and horizontal planes. Two lines represent the edge of the optic disc and the single red line represents the arbitrary reference plane.
The mean height contour graph (centre right) displays the variation of the retinal surface height along the contour line (green). The reference line (red) below this shows the position of the reference plane, designated as the plane of separation between the cup below and the neuroretinal rim above. This reference plane is parallel to the peripapillary retinal surface and is located 50 µm below the retinal surface at the location of the papillomacular bundle on the contour line. It is thus approximately located at the lower extent of the RNFL.
The display of the retinal surface height variation along the contour line begins temporally at 0° (approximate centre of the papillomacular bundle). The height profile is plotted in a clockwise direction for a right eye and a counter-clockwise direction for a left eye. The graph largely corresponds to the course of the RNFL thickness along the disc margin.
The Moorfields regression analysis is depicted as seven colour bar graphs, one bar for each segment and one global bar (bottom right). If the top of the green bar lies above the 95.0% prediction interval then the corresponding disc segment is classified as within normal limits, if it lies between the 95.0% and 99.9% it is borderline, and if it lies below 99.9% it is outside normal limits.
Page 330
Detailed stereometric data are presented in a table (bottom left). Readings outside normal are indicated with an asterisk.
image

Fig. 10.23 HRT of a normal eye

image

Fig. 10.24 HRT of a glaucomatous eye

Scanning laser polarimetry

1 Physics. The GDx (Glaucoma Diagnosis) RNFL analyzer assesses the nerve fibre layer thickness by using its assumed ‘birefringent’ (resolving or splitting a light wave into two unequally reflected or transmitted waves) nature to change the polarization of incident polarized diode laser light; the amount of alteration is directly related to the thickness of the layer. The degree of polarization is assessed over an area of 1.75 disc diameters concentric to the disc and the profile of the density of the RNFL established; the thicker the RNFL the greater the polarization. The newer GDxVCC (Variable Corneal Compensation) version has eliminated many of the problems of the previous model which hindered its ready clinical acceptance.
2 Indications are similar to those of the SLO, although there is no macular facility.
3 Display provides colour images of the optic nerve head and RNFL maps in the four quadrants (Fig. 10.25):
Page 331
The fundus image of the left and right eyes at the top is useful in identifying image quality.
The thickness maps are presented in a colour-coded spectrum from blue to red. Red followed by yellow indicates a thick RNFL whereas blue followed by green shades are consistent with thin RNFL. The map has an hourglass appearance because the RNFL is thickest superiorly and inferiorly.
The deviation maps show the location and magnitude of RNFL defects as tiny colour coded squares (pixels).
The TSNIT (temporal-superior-nasal-inferior- temporal) graph is displayed at the bottom. It shows the actual values for that eye along with a shaded area that represents the 95% normal range for that age. The curve in a healthy eye should fall within the shaded area and has a double hump pattern because the superior and inferior fibres are thickest. The central printout shows the values for both eyes together.
Parameters for each eye are displayed in a table (top centre). The nerve fibre indicator (NFI) at the bottom of the table indicates a global value based on the entire thickness map and is the optimal parameter for discriminating normal from glaucoma. Normal is 1–30, borderline is 31–50 and abnormal is 51–100.
image

Fig. 10.25 GDx VCC shows reduction in retinal nerve fibre density in the right eye and abnormal parameters

(Courtesy of J Salmon)

Optical coherence tomography

OCT has become a routine part of the management of macular and other retinal disease, but is also widely used for the assessment of glaucoma. The principles are discussed in detail in Chapter 14. The following imaging strategies are applicable to glaucoma:

1 Peripapillary retinal nerve fibre layer. This involves the acquisition of a circular scan of diameter 3.4 mm of the retina around the optic nerve head. Retinal thickness is compared with normals. Sensitivity and specificity are around 90%.
2 Optic nerve head. Radial cross-sectional scans permit an objective and repeatable assessment of disc morphology, with reasonable discriminatory value. This function has tended to be less commonly used than RNFL analysis in practice.

Anterior chamber depth measurement

Objective measurement of the depth of the anterior chamber is sometimes clinically useful in glaucoma management. Indications include monitoring of progression in conditions where the anterior chamber is shallowed such as post-trabeculectomy hypotony and cilio-lenticular block, and as a diagnostic tool, including the comparison of the two eyes. Older methods involved using a slit-lamp with or without a special attachment, but an accurate and repeatable measurement can be obtained using ultrasonographic or optical interferometric methods (e.g. ACD function on Zeiss IOLMaster). Utility and accuracy is limited in pseudophakic eyes.

Perimetry

Definitions

1 The visual field can be represented as a three-dimensional structure akin to a hill of increasing sensitivity (Fig. 10.26A). The outer aspect extends approximately 50° superiorly, 60° nasally, 70° inferiorly and 90° temporally. Visual acuity is sharpest at the very top of the hill (i.e. the fovea) and then declines progressively towards the periphery, the nasal slope being steeper than the temporal. The ‘bottomless pit’ of the blind spot is located temporally between 10° and 20°, slightly below the horizontal.
2 An isopter is a line connecting points of the same sensitivity, and on a two-dimensional isopter plot encloses an area within which a stimulus of a given strength is visible. An isopter plot of the right eye is shown in Fig. 10.26B. When the field is represented as a hill, isopters resemble the contour lines on a map.
3 A scotoma is an area of reduced (‘relative’) or total (‘absolute’) loss of vision which is surrounded by a seeing area.
4 Luminance is the intensity or ‘brightness’ of a light stimulus, measured in apostilbs (asb). A higher intensity stimulus has a higher asb value; this is the inverse of sensitivity.
5 Logarithmic scale rather than a linear scale is used for stimulus intensity and sensitivity, so that for each log unit intensity changes by a factor of 10. With a log scale, greater significance is given to the lower end of the intensity range. The normal eye has a very large sensitivity range, and assessment of the lower end of the scale is of critical significance so that early damage can be detected. With a linear scale, the lower end would be reduced to a very small portion of a graphical chart axis. The visual system itself operates on close to a logarithmic scale, so using this method more closely matches the physiological situation.
6 Decibels. Simple log units are not used in clinical perimetry, but rather ‘decibels’ (dB), where 10 dB = 1 log unit. Decibels are not true units of luminance but a representation, and vary between visual field machines. Perimetry usually concentrates on the eye’s sensitivity rather than the stimulus intensity. Therefore, the decibel scale goes up as retinal sensitivity goes up, which obviously corresponds to reducing intensity of the perceived stimulus. This makes the assessment of visual fields more intuitive, as a higher number corresponds with higher retinal sensitivity. If the sensitivity of a test location is 20 dB (= 2 log units), a point with a sensitivity of 30 dB would be the more sensitive. The blind spot has a sensitivity of 0 dB. If, on a given machine, seeing a stimulus of 1000 asb gives a value of 10 dB, a stimulus of 100 asb will give 20 dB.
Page 332
7 Differential light sensitivity represents the degree by which the luminance of a target must exceed background luminance in order to be perceived. The visual field is therefore a three-dimensional representation of differential light sensitivity at different points.
8 Threshold at a given location in the visual field is the brightness of a stimulus at which it can be detected by the subject. It is defined as ‘the luminance of a given fixed-location stimulus at which it is seen on 50% of the occasions it is presented’. In practice we usually talk about an eye’s sensitivity at a given point in the field rather than the stimulus intensity. The threshold sensitivity is highest at the fovea and decreases progressively towards the periphery. After the age of 20 years the sensitivity decreases by about 1 dB per 10 years.
Page 333
9 Background luminance. The retinal sensitivity at any location varies depending on background luminance. Rod photoreceptors are more sensitive in dim light than cones and so due to their preponderance in the peripheral retina, at lower (‘scotopic’) light levels the peripheral retina becomes more sensitive in proportion to the central retina; the hill of vision flattens, with a central crater rather than a peak at the fovea due to the high concentration of cones, which have low sensitivity in scotopic conditions. It should be noted that it takes about 5 minutes to adapt from darkness to bright sunlight and 20–30 minutes from bright sunlight to darkness.
image

Fig. 10.26 (A) Hill of vision; (B) isopter plot

Types of perimetry

Kinetic

Kinetic perimetry is a two-dimensional assessment of the boundary of the hill of vision. It involves the presentation of a moving stimulus of known luminance or intensity from a non-seeing area to a seeing area until it is perceived (Fig. 10.27A). The stimulus is moved at a steady speed along various meridia (clock hours) and the point of perception is recorded on a chart. By joining these points along different meridians an isopter is plotted for that stimulus intensity. Using stimuli of different intensities a contour map of the visual field with several different isopters can be plotted. Kinetic perimetry can be performed by simple confrontation or by means of a perimeter such as the Goldmann.

image

Fig. 10.27 Principles of perimetry. (A) Kinetic; (B) static

Static

A method of assessing fields in which the location of a stimulus remains fixed at a certain location within the field, with the intensity increased until it is seen by the subject (or decreased until it is no longer seen). In other words, the target intensity is increased (or decreased) until threshold is reached (Fig. 10.27B). The most frequently used automated perimeter is the Humphrey Field Analyzer (HFA); others include the Henson, Dicon and Octopus. Automated static perimetry now constitutes the method used for the great majority of visual fields monitoring in the care of patients with glaucoma.

Suprathreshold

Suprathreshold perimetry involves testing with stimuli of luminance above the expected normal threshold levels for an age-matched population to assess whether these are detected; in other words, testing to check that a subject can see stimuli that would be seen by a normal person of the same age. It enables testing to be carried out rapidly to indicate whether function is grossly normal or not. However, it is not highly quantitative, and so is usually reserved for screening.

Threshold

Threshold perimetry is used for detailed assessment of the hill of vision by plotting the threshold luminance value in various locations in the visual field and comparing the results with age-matched ‘normal’ values. In Humphrey perimetry (see below), a stimulus of higher than expected intensity is presented; if seen, the intensity is decreased in 4 dB steps until it is no longer seen (‘staircasing’). The stimulus is then increased again in 2 dB steps until seen once more (Fig. 10.28). If the stimulus is not seen initially, its intensity is increased in 4 dB steps until seen, then decreased in 2 dB steps until no longer seen. Essentially, the threshold is crossed in one direction with large increments, then crossed again to ‘fine-tune’ the result with smaller increments. Threshold testing is quantitatively detailed and is therefore used for monitoring glaucomatous fields.

image

Fig. 10.28 Determination of threshold

Sources of error

The skill of the perimetrist in setting up the test, explaining the procedure to the patient, reassuring the patient and monitoring performance is fundamental to obtaining an accurate field. However, errors may still occur as a result of one or more of the following factors:

1 Poor performance by the patient.
Page 334
Page 335
2 Uncorrected refractive error can cause a significant decrease in central sensitivity. If a hypermetropic patient who usually wears contact lenses is tested wearing spectacles, this will have the effect of magnifying and enlarging any scotomas as compared with contact lenses. Most perimetry is performed with a stimulus at approximately reading distance, so a near correction should be used for presbyopic patients.
3 Spectacle rim artefact. Spectacles can cause rim scotomas if small aperture lenses are used or if incorrectly dispensed. Some (narrow-aperture) trial frame lenses are unsuitable for perimetry.
4 Miosis decreases sensitivity in the peripheral field and increases variability in the central field in both normal and glaucomatous eyes. Pupils less than 3 mm in diameter should therefore be dilated prior to perimetry; a consistent mydriatic should be used for serial tests.
5 Media opacities (usually cataract) can have a profound effect, exaggerated by miosis.
6 Ptosis, even if mild, can suppress the superior visual field. Similar effects result from dermatochalasis, prominent eyelashes, and deeply set eyes.
7 Inadequate retinal adaptation may also lead to error if perimetry is performed soon after ophthalmoscopy.

Humphrey Field Analyzer

The Humphrey Field Analyzer (HFA) consists of a hemispherical bowl onto which a target can be projected at any location in the visual field (Fig. 10.29).

A monitor on the side of the instrument presents a series of menus. Background luminance is set at 31.5 asb, considered to be at the lower end of the photopic illumination range.
Variation in stimulus intensity can be achieved by altering either target size or luminance. Size is set prior to the test; 4 mm2 is used routinely, the same as the Goldmann perimeter stimulus size III.
The other stimulus sizes available on the Humphrey correspond to the rest of the Goldmann stimulus size range (I, II, IV and V), but these are rarely utilized and usually only luminance is altered – this can be varied between 0.08 asb and 10 000 asb brighter than the background: between about 51 dB and 0 dB.
image

Fig. 10.29 Humphrey perimetry

Testing patterns

Most important defects in glaucoma occur within the central 30° radius of field, so this is the area most commonly tested.

The pattern of the points analyzed in a particular test is located at the top left on a standard HFA printout; 24–2 is an example in widespread routine use in which ‘24’ denotes the extent, in degrees, to which the field is tested on the temporal side (tested to 30° on the nasal side in the 24–2). The number after the dash (–2 or sometimes –1) describes the pattern of the points tested.
The –2 strategy involves a grid of test points spaced 6° apart, offset from the vertical and horizontal meridia whereas the –1 includes points along the vertical and horizontal meridia.
In another common glaucoma pattern, 30–2, the area tested extends to 30° temporally as well as nasally.
Other examples include 10–2 and FF-120; 10–2 is used to assess an area of central radius 10° – as defects here may threaten central vision careful monitoring is commonly required. FF (‘full field’) – 120 (120°) is used to assess neurological defects. The HFA can also be used to perform binocular field testing (e.g. Esterman strategy) to assess statutory driving entitlement.

Testing strategies

1 Suprathreshold strategies are rapid (6 minutes per eye) qualitative programs. An 88-point screening test using a three-zone strategy may be used initially as it is fast and less demanding than full threshold formats. An absolute defect is indicated with a black square and a relative defect with a cross.
2 Full-threshold strategy is now seldom used, mainly because of its long duration, frequently 15–20 minutes/eye including setup, making it difficult for patients to maintain concentration. Initially four points are tested to determine threshold levels which are then used to predict the levels for neighbouring points and so on until the entire field has been tested. Multiple additional threshold checks are performed.
3 SITA (Swedish Interactive Thresholding Algorithm) uses an extensive database of normal and typically glaucomatous fields to estimate threshold values, and takes the patient’s ongoing responses into account to arrive at adjusted estimates throughout the test, based on probability levels. It stops testing a given location when the margin of error is acceptable, and uses response time rather than false positive catch trials to estimate the false positive rate (there is a strong correlation between the two). The stimulus presentation rate is speeded up in fast responders. Standard and faster versions are available; SITA-Fast uses similar methods to SITA-Standard and is preferred by some practitioners, but may be less repeatable and slightly less sensitive.

Displays

1 The numerical display (numerical grid) is located to the left of the grey scale and to the right of the reliability indices. It gives the measured or estimated (depending on strategy) threshold in dB at each point. In a full-threshold strategy, where the threshold is rechecked either as routine or because of an unexpected (>5 dB) result, the second result is shown in brackets next to the first.
2 The grey scale represents the adjacent numerical display in graphical form and is the simplest display to interpret: decreasing sensitivity is represented by darker tones. Each change in grey scale tone is equivalent to a 5 dB change in sensitivity at that location.
3 Total deviation display (Fig. 10.30 left) represents the difference between the test-derived threshold at each point and the normal sensitivity at that point in the general population, correcting for age. Negative values indicate lower than normal sensitivity, positive values higher than normal.
Page 336
4 Pattern deviation (Fig. 10.30 right) is derived from the total deviation values adjusted for any generalized decrease in sensitivity in the overall field which might be caused by other factors such as lens opacities or miosis. It therefore demonstrates localized defects such as occur in glaucoma.
5 Probability displays are located below the numerical total and pattern deviation displays (Fig. 10.30 bottom). These constitute a graphical representation of the percentage (<5% to <0.5%) of the normal population in whom the measured defect at each point would be expected. Darker symbols represent a greater likelihood that a defect is significant.
image

Fig. 10.30 Total deviation, pattern deviation and probability indices

Reliability indices

Reliability indices reflect the extent to which the patient’s results are reliable and should be analyzed first. If grossly unreliable, further analysis of a visual field printout is of little value. In patients who consistently fail to achieve good reliability indices it may be useful to switch to a suprathreshold strategy or kinetic perimetry.

1 Fixation losses indicate steadiness of gaze during the test. They are detected in older HFA versions by presenting stimuli to the blind spot; if the patient responds, a fixation loss is recorded. The fewer the number of losses the more reliable is the test. A ‘gaze monitor’ is used on newer HFAs. A high fixation loss score may occur if the instrument has incorrectly plotted the blind spot.
2 False positives are detected when a stimulus is accompanied by a sound. If the sound alone is presented and the patient still responds, a false positive is recorded. False positive responses do not increase in damaged visual fields. With a high false positive score the grey scale printout appears abnormally pale (Fig. 10.31). Fixation losses are also frequently high and the glaucoma hemifield test shows abnormally high sensitivity. In SITA testing, false positives are estimated based on the response time, which correlates well with the false positive rate.
3 False negatives are detected by presenting a stimulus much (9 dB) brighter than threshold at a location where the threshold has already been determined. If the patient fails to respond a false negative is recorded. A high false negative score indicates inattention or tiredness. It may also be due to short-term fluctuation associated with glaucoma, and may be an indicator of disease severity rather than patient unreliability. The grey scale printout in individuals with high false negative responses tends to have a clover-leaf shape (Fig. 10.32).
4 Interpretation. It is important to note that there is relatively little research evidence in this area, with limited absolutes in branding a field as clearly reliable or unreliable. With SITA strategies, false negatives or false positives over about 15% should probably be regarded as highly significant, and with full-threshold strategies, fixation losses over 20% and false positives or negatives over 33%.
image

Fig. 10.31 High false positive score (arrow) with an abnormally pale grey scale display

image

Fig. 10.32 High false-negative score (arrow) with a clover-leaf-shaped grey scale display

Page 337

Glaucoma hemifield test

The glaucoma hemifield test (GHT) is a means, available in the 24–2 and 30–2 HFA testing patterns, of assessing the visual field for damage in a pattern commonly seen in glaucoma. The GHT compares five corresponding areas on the superior and inferior fields (as glaucomatous change is typically vertically asymmetrical). It also assesses overall sensitivity.

Global indices

Essentially, the global indices represent a statistical summary of the field in a single number; they are used principally used to monitor progression of glaucomatous damage rather than for initial diagnosis.

1 Mean deviation (MD) (elevation or depression) gives an indication of the overall sensitivity of the field. It is derived from averaging the total deviation values, with central points given more weight.
Page 338
2 Pattern standard deviation (PSD) is a measure of focal loss or variability within the field taking into account any generalized depression in the hill of vision. An increased PSD is therefore a more specific indicator of glaucomatous damage than MD.
3 Short-term fluctuation (SF) is an indication of the consistency of responses during a single test. It is derived from the difference in threshold at locations where it is tested twice. Abnormality may indicate poor concentration or tiredness; it tends to increase with ageing and in glaucoma.
4 Corrected pattern standard deviation (CPSD) consists of PSD corrected for SF to produce a value for focal field abnormality corrected for intratest variability.
5 Global indices in SITA. SF and CPSD cannot be calculated with SITA testing as the threshold is never tested twice at any location. Therefore, there is no data available about intratest SF and without SF CPSD cannot be derived.
6 Probability values. Abnormal global indices are followed by a probability (P) value. This represents the percentage likelihood that an abnormal value of this level will occur in a normal subject; in other words, the lower the P value, the more likely the result is abnormal.

Computer analysis of serial fields

Adoption into routine clinical practice of computer software for analysis of serial visual fields has, to date, been slow due to several factors. A large number of reliable fields need to be carried out (usually over an extended period) before analysis is effective. Clinical impression frequently differs from software-based interpretation, and different software packages have tended to show poor correlation in assessing the same data. Resource constraints may also have played a part. The quality of available software has been improving steadily, and there are signs that gradual integration may take place. Newer analysis programs include ‘Progressor’ and ‘PeriData’. In the former, each test location is subjected to linear regression analysis, and an indication of stability and of the speed of any deterioration is given using coloured graphical representation.

Short-wave automated perimetry

Short-wave automated perimetry (SWAP) uses a blue stimulus on a yellow background. Sensitivity to blue light (mediated by blue cone photoreceptors) is adversely affected relatively early in glaucoma. SWAP is more sensitive to early glaucomatous defects but has not been widely adopted because cataract decreases sensitivity to blue light (the brunescing lens acts as a yellow filter) and patients frequently dislike the lengthy test. It is available on newer HFA models.

Frequency-doubling contrast test

1 Physiological principles. Ganglion (M) cells with relatively large diameter axons comprise 25% of the ganglion cell population. They are particularly susceptible to glaucomatous damage and appear to be preferentially lost in early glaucoma. A loss of a small number of these cells has a considerable effect on visual function. Psychophysical tests have been devised to target visual function provided by these magnocellular pathways in the detection of early glaucoma.
2 Frequency-doubling illusion is produced when a low spatial frequency sinusoidal grating (less than one cycle per degree) undergoes high temporal frequency counter phase flicker (>15 Hz). The rapid alternation in which the light bars become dark and vice versa produces the illusion of the grating having doubled its frequency.
3 The perimeter is a tabletop instrument (Fig. 10.33 bottom) which can be used under normal room lighting and requires no patching, since the viewing canopy automatically covers the eye not being tested. The device requires minimal training and is relatively portable.
4 Stimuli are presented in 17 or 19 sectors in the central 20° or 30° depending on the program used, screening or full threshold.
5 Testing time is short with full threshold programs taking about 5 minutes per eye and screening procedures between 45 and 90 seconds per eye. Because of this most patients prefer the FDT test to conventional perimetry.
6 Results are displayed and printed together with reliability indices, probabilities, mean deviation and pattern standard deviation (Fig. 10.33 top). FDT has high sensitivity both in screening to differentiate healthy individuals from those with glaucoma and for quantifying glaucomatous damage. The results are minimally affected by refractive error of up to 6 D and not at all by pupil size. The device has an age-adjusted normative database, as well as a statistical analysis package for immediate evaluation of results.
7 The Humphrey Matrix is a more recently introduced FDT perimeter which allows extended testing of considerably larger areas of field than the basic screening version. It is thus proposed as being at least comparable to the HFA for refined diagnosis and monitoring.
image

Fig. 10.33 Frequency-double perimeter and display

Ocular hypertension

Definition

In the general population the mean IOP is 16 mmHg; two standard deviations on either side of this gives a ‘normal’ IOP range 11–21 mmHg. The distribution is Gaussian with the curve skewed to the right (Fig. 10.34).

In the elderly the mean IOP is higher, particularly in women, and the standard deviation greater than in younger individuals. This means that ‘normal’ IOP in elderly women may range up to 24 mmHg and not 21 mmHg.
Page 339
It is estimated that 4–7% of the population over the age of 40 years have IOPs >21 mmHg without detectable glaucomatous damage: ‘ocular hypertension’ (OHT).
An absence of angle-closure is implicit, and there should be no detectable cause of secondary glaucoma, though sometimes the term OHT is used to describe raised IOP in these contexts.
image

Fig. 10.34 Distribution of intraocular pressure in the general population

Risk factors for developing glaucoma

The Ocular Hypertension Treatment Study (OHTS) was a multicentre longitudinal trial. In addition to looking at the effect of treatment in ocular hypertensives (IOP <32 mmHg), invaluable landmark information was gained about the effect of a range of putative risks for conversion from OHT to glaucoma; the percentage of OHT patients likely to develop glaucoma taking key factors into account is set out in Tables 10.3 and 10.4 (median follow-up was 72 months). Additional considerations are discussed below. Limitations included the possibility that early glaucomatous damage was already present in some of the patients classified as having OHT. The fact that some percentages appear anomalous may be due to the relatively low numbers in different subcategories.

Table 10.3 Risk of developing glaucoma according to IOP (intraocular pressure) and CCT (central corneal thickness)

image

Table 10.4 Risk of developing glaucoma according to vertical C/D ratio and CCT

image

The following factors were significant on multivariate analysis:

1 Intraocular pressure. The risk increases with increasing IOP.
2 Age. Older age is associated with greater risk.
3 Central corneal thickness (CCT). The risk is greater in eyes with low CCT and lower in eyes with higher CCT. This is probably due to resultant under- and over-estimation of IOP although it has been proposed that associated structural factors, perhaps at the lamina cribrosa, might also be important.
4 Cup–disc (C/D) ratio. The greater the C/D ratio the higher the risk. This may be because an optic nerve head with a large cup is structurally more vulnerable, or it may be that early damage is already present.
5 Pattern standard deviation (PSD). A greater PSD result represented a significant risk. It is possible that this signified early glaucomatous field change.

The following factors were significant on univariate analysis only; they were not significant in isolation but were over-ridden when the factors considered above were taken into account.

Page 340
1 African-American race was associated with a higher glaucoma risk.
2 Gender. Males were more likely to convert.
3 Heart disease was found to be significant.

Factors examined in the OHTS but not found to be significant are listed below.

1 Myopia, although it is suspected that myopic discs are more susceptible to glaucomatous damage at a lower IOP than emmetropic discs.
2 Diabetes. An apparent protective effect of diabetes was initially found, but later analysis with refreshed data did not confirm this.
3 Family history of glaucoma was not found to be a risk factor for conversion.
4 Other factors which were not examined in the OHTS but may be important include retinal nerve fibre defects (though the presence of these may be taken to indicate pre-perimetric glaucoma – see below) and specific peripapillary atrophic changes.

Pre-perimetric glaucoma

This concept refers to glaucomatous damage, usually manifested by a suspicious optic disc and/or the presence of retinal nerve fibre layer defects, in which no visual field abnormality has developed. The field testing modality for this purpose is usually taken as standard achromatic automated perimetry.

Management

In the OHTS, untreated patients with ocular hypertension had a 9.5% cumulative risk of developing POAG after 5 years; treatment (which aimed to reduce IOP by 20% or more and to reach 24 mmHg or less) reduced this to 4.4%. Hence, when deciding on whether to start treatment it is important to take into account that it will be necessary to treat a large number of patients in order to prevent the development of glaucoma in a single individual.

Age, and so life expectancy, is a key point to consider. In general, only those at higher risk should be treated, although patient preference may be a decisive factor.
Most practitioners would treat every patient with an IOP of 30 mmHg or more. The decision to treat in patients with varying risk profiles is commonly less than straightforward, and has to be made on an individual basis.
Various guidelines exist, but there is a high level of disagreement even between glaucoma specialists. Careful monitoring is a reasonable alternative in many circumstances.
OHT almost certainly increases the risk of retinal venous occlusion, an additional point to take into account when considering whether to start treatment.
Drug choice is the same as for POAG, although a less aggressive pressure-lowering approach is frequently taken.

Primary open-angle glaucoma

Introduction

Definition

Primary open-angle glaucoma (POAG), also referred to as chronic simple glaucoma, is a generally bilateral disease of adult onset characterized by:

An IOP >21 mmHg at some stage.
Glaucomatous optic nerve damage.
An open anterior chamber angle.
Characteristic visual field loss as damage progresses.
Absence of signs of secondary glaucoma or a non-glaucomatous cause for the optic neuropathy.

POAG is the most prevalent type of glaucoma in individuals of European and African ethnic origin. It affects both sexes equally.

Risk factors

1 IOP. The higher the IOP, the greater the likelihood of glaucoma.
2 Age. It is more common in older individuals.
3 Race. It is significantly (perhaps four times) more common, develops at an earlier age, and may be more difficult to control in black individuals than in whites.
4 Family history of POAG. First-degree relatives of patients with POAG are at increased risk. An approximate risk to siblings is four times and to offspring twice the normal population risk.
Page 341
5 Diabetes mellitus. Many studies suggest a correlation between diabetes and POAG.
6 Myopia is associated with an increased incidence of POAG and myopic eyes may be more susceptible to glaucomatous damage.
7 Vascular disease. A range of systemic conditions linked to vascular compromise may be associated, though clear-cut relationships have proved difficult to demonstrate consistently. Systemic hypertension, cardiovascular disease, diabetes and vasospastic conditions such as migraine have all been implicated. Poor ocular perfusion may be a risk factor for glaucoma progression.

Genetics

Mutations at 15 loci in the human genome have so far been identified as associated with POAG and are designated primary open angle glaucoma-1A (GLC1A) to GLC1O. Four susceptible genes have been identified: the MYOC gene (chromosome 1q21-q31), coding for the glycoprotein myocilin that is found in the trabecular meshwork and other ocular tissues, the OPTN gene on chromosome 10p, which codes for optineurin, the WDR36 gene on chromosome 5q22, and the NTF4 gene on chromosome 19q13.3. Among them MYOC is the most frequently mutated gene in POAG: a study of unrelated POAG patients found myocilin mutations in at least 4% of the adults. A number of different mutations have been described in the MYOC gene, though the normal function of myocilin and its role in causing glaucoma is as yet undetermined. If a single family member develops glaucoma prior to age 35 years, the chances that the genetic defect is a mutation is the myocilin gene may be as high as 33%.

Steroid responsiveness

A proportion of the population develops an elevation in IOP in response to a course of topical steroid; more potent steroids have a greater propensity to elevate IOP, as does greater frequency of instillation. This tendency is more marked in patients with POAG and their close relatives. Intra- and periocular steroid administration, including periocular application of steroid skin cream and nasal administration, are also prone to elevate IOP. Systemic steroids are much less prone to cause elevation of IOP, but substantial, probably dose-dependent, rises can occur and some authorities have advocated screening for all patients on systemic steroids, perhaps those on dexamethasone in particular. The precise mechanism of the ‘steroid response’ is uncertain, but it may be mediated by an increase in trabecular meshwork cell myocilin production.

Pathogenesis of glaucomatous optic neuropathy

Retinal ganglion cell death in glaucoma occurs predominantly through apoptosis (programmed cell death) rather than necrosis. The preterminal event is calcium ion influx into the cell body and an increase in intracellular nitric oxide; glutamine metabolism is intrinsically involved. After initial injury, a cascade of events results in astrocyte and glial cell proliferation, and alterations in the extracellular matrix of the lamina cribrosa, with subsequent optic nerve head remodelling. Multiple factors are likely to be involved, but the mechanisms remain relatively speculative: the process of glaucomatous damage and the relationship with IOP and other potential influences is still poorly understood. One or both of the following mechanisms may be involved:

1 Direct mechanical damage to retinal nerve fibres at the optic nerve head, perhaps as they pass through the lamina cribrosa.
2 Ischaemic damage, possibly due partly to compression of blood vessels supplying the optic nerve head.

Insults via both mechanisms might lead to reduction in axoplasmic flow, interference with the delivery of nutrients or removal of metabolic products, deprivation of neuronal growth factors, oxidative injury and the initiation of immune-mediated damage.

Screening

Universal population screening for glaucoma has not been demonstrated to be cost-effective, and current practice restricts screening to high-risk groups, such as older individuals, those with a history of POAG in a close family member over the age of 40, and people of black ethnicity. In these groups, screening tends to be performed sporadically via routes such as commercial sight tests, which may lead to the relative exclusion of underprivileged economic groups. Population screening with tonometry alone is unsatisfactory, since it will label as normal a significant number of cases with other features of POAG such as cupping and visual field loss. Even with the added criterion of a vertical cup–disc ratio of >0.4, only a proportion of potential POAG patients will be identified. It is therefore prudent for routine screening eye examinations to include visual field examination as well as tonometry and ophthalmoscopy.

Diagnosis

History

1 Visual symptoms will usually be absent, unless damage is advanced. Sometimes symptomatic central field defects may occur at an early stage, in the presence of a relatively normal peripheral field.
2 Previous ophthalmic history. Specific enquiry should be made about:
Refractive status as myopia carries an increased risk of POAG, and hypermetropia of primary angle-closure glaucoma (PACG).
Causes of secondary glaucoma such as ocular trauma or inflammation; previous eye surgery, including refractive surgery may affect IOP reading.
3 Family history
POAG or related conditions such as OHT.
Other ocular disease in family members.
Page 342
4 Past medical history. Asking specifically about the following may be indicated.
Asthma, heart failure or block, peripheral vascular disease: contraindications to the use of beta-blockers.
Head injury, intracranial pathology including stroke that may cause optic atrophy or visual field defects.
Vasospasm: migraine and Raynaud phenomenon.
Diabetes, systemic hypertension and cardiovascular disease may increase the risk of POAG.
5 Current medication
Steroids including skin cream and inhalants.
Oral beta-blockers may lower IOP.
6 Social history including smoking and alcohol intake, especially if toxic/nutritional optic neuropathy is suspected.
7 Allergies especially to any drugs likely to be used in glaucoma treatment, particularly sulfonamides.

Examination

1 Visual acuity is likely to be normal except in advanced glaucoma.
2 Pupils. Exclude a relative afferent pupillary defect (RAPD); if absent then subsequently develops this constitutes an indicator of substantial progression.
3 Colour vision assessment such as Ishihara chart testing if there is any suggestion of an optic neuropathy other than glaucoma.
4 Slit-lamp examination. Exclude features of secondary glaucomas such as pigmentary and pseudoexfoliative.
5 Tonometry, prior to pachymetry, noting the time of day.
6 Pachymetry for CCT.
7 Gonioscopy.
8 Optic disc examination should always be performed with the pupils dilated, provided gonioscopy does not show critically narrow angles. Red-free light can be used to detect RNFL defects.
9 Perimetry should usually be performed prior to clinical examination.
10 Optic disc or peripapillary RNFL imaging as described above.

Visual field defects

1 The earliest changes suggestive of glaucoma consist of increased variability of responses in areas that subsequently develop defects. Alternatively there may be slight asymmetry between the two eyes.
2 Paracentral, small, relatively steep depressions (Fig. 10.35A and B) constitute approximately 70% of all early glaucomatous field defects. Since the defects respect the distribution of the retinal nerve fibre layer they terminate at the horizontal midline; defects above and below the horizontal therefore are not aligned with each other. Central/paracentral scotomata may be most appropriately monitored using the 10–2 Humphrey perimetry pattern.
3 Nasal (Rønne) step represents a difference in sensitivity above and below the horizontal midline in the nasal field. It is a common finding usually associated with other defects (Fig. 10.36A and B). A temporal wedge is less common but has similar implications.
4 Arcuate-shaped defects develop as a result of coalescence of paracentral scotomas. They typically develop between 10° and 20° of fixation in areas that constitute downward or, more commonly, upward extensions from the blind spot around fixation (Bjerrum area). With time, they tend to elongate circumferentially along the distribution of arcuate nerve fibres (Seidel scotoma) and may eventually connect with the blind spot (arcuate scotoma) reaching to within 5° of fixation nasally (Fig. 10.37A and B).
Page 343
5 Enlargement of scotomas due to damage to adjacent fibres.
6 Deepening of scotomas and development of fresh defects.
7 A ring scotoma develops when arcuate defects in upper and lower halves of the visual field join. Misalignment between the two often preserves the nasal step (Fig. 10.38A and B).
8 End-stage changes are characterized by a small island of central vision typically accompanied by a temporal island. The temporal island is usually extinguished before the central.
image

Fig. 10.35 Mild damage. (A) Minimal cupping; (B) small paracentral scotoma

image

Fig. 10.36 Moderate damage. (A) Moderate cupping; (B) arcuate scotoma and a nasal step

image

Fig. 10.37 Severe damage. (A) Marked cupping; (B) dense arcuate scotoma/nasal step connecting with the blind spot

image

Fig. 10.38 Very severe damage; (A) Gross cupping; (B) ring scotoma

Global indices should always be taken into account (Fig. 10.39); on average annual deterioration in mean total deviation of just over 1.0 dB can be expected in treated patients.

image

Fig. 10.39 Progression of visual field defects and deterioration of global indices over a period of 30 months

Management

The primary aim of treatment is to prevent functional impairment of vision within the patient’s lifetime by slowing the rate of ganglion cell loss closer to that of the normal population (approximately 5000/year). Currently the only proven method of achieving this is the lowering of IOP.

Patient instruction

Page 344

An explanation should be offered concerning the nature of the disease, and an explanatory booklet provided. The timing of medication use should be specified, and the patient educated in the technique of eye drop instillation. At follow-up visits the patient’s proficiency at instilling drops should be checked. In order to maximize drug contact time with the anterior segment and to minimize systemic absorption the patient should be instructed either to perform lacrimal sac occlusion (by applying fingertip pressure at the medial canthus) or to close the eyes for about 3 minutes after instillation. Common or severe potential adverse effects should be explained at the commencement of treatment and their occurrence enquired about at follow-up visits.

Treatment goals

1 Target pressure. It is assumed that the pre-treatment level of IOP has damaged the optic nerve and will continue to do so. An IOP level is identified below which further damage is considered unlikely (‘target pressure’). This is identified taking into account the severity of existing damage, the level of IOP, CCT, the rapidity with which damage occurred if known, as well as the age and general health of the patient. Therapy should maintain the IOP at or below the target level. If not achievable by conservative modalities, a decision is made regarding whether to proceed with surgery or to continue monitoring with an above-target IOP.
2 Proportional reduction. An alternative strategy is to aim for a reduction in IOP by a certain percentage – often 30% – and then monitor, aiming for a further reduction if progression occurs. There may be a smaller margin for error with this approach if advanced damage is present.
3 Monitoring of the optic nerve and visual fields is performed. In the event of further damage the target IOP is reset at a lower level. Although there is no ‘safe’ level, progression is uncommon if the IOP is <16 mmHg. It appears that each 1 mmHg reduction in IOP leads to a 10% reduction in the rate of nerve fibre loss. As the disease progresses the degree of redundancy or ‘reserve capacity’ within the visual system diminishes and the loss of each remaining ganglion cell has a greater proportional impact on visual function. Lower target pressures therefore tend to be set in patients with advanced disease.

Medical therapy

1 Commencing medical therapy
Any chosen drug should be used in its lowest concentration, instilled as infrequently as possible consistent with the desired therapeutic effect.
Ideally the drug with the fewest potential side-effects should be used.
Initial treatment is usually with one drug, usually a prostaglandin analogue or beta-blocker; a range of considerations is taken into account when deciding which to select.
2 Review
The interval to review after starting medication is set according to the individual patient, but is usually 4–8 weeks.
Response to the drug is assessed against the target IOP.
If the response is satisfactory subsequent assessment is generally set for a further 3–6 months.
If there has been little or no response the initial drug is withdrawn and another substituted.
If there has been an apparently incomplete response another drug may be added or a fixed combination substituted.
When two separate drugs are used the patient should be instructed to wait five minutes before instilling the second drug to prevent washout of the first.
Sometimes it may be worthwhile allowing a further month or two of treatment before altering a regimen, as response may improve over time.
Poor compliance or inadequate drop instillation technique should be considered as a cause of unsatisfactory response.
When drops are administered in the morning, it is good practice always to enquire about whether today’s dose has been used prior to the examination.
3 Perimetry. If IOP control is good and glaucomatous damage mild or moderate with no substantial threat to central vision, annual perimetry is generally sufficient.
4 Gonioscopy should be performed annually in most patients because the anterior chamber angle tends to narrow with age.
5 Optic disc examination should be performed at appropriate intervals. Serial imaging may also provide useful additional information.
6 Causes of treatment failure
Inappropriate target pressure. If the IOP is maintained in the upper part of the statistically normal range, progressive field loss is relatively common.
Poor compliance with therapy occurs in at least 25% of patients.
Wide fluctuations in IOP frequently occur in patients treated medically, and may be associated with progression.
Patients may deteriorate despite apparently good IOP control. Causes include occult compliance failure, undetected diurnal variation, and possibly impaired optic nerve perfusion. The possibility of an alternative pathology, particularly a compressive lesion, should always be considered.

Laser trabeculoplasty

In argon laser trabeculoplasty (ALT) or selective laser trabeculoplasty (SLT) argon or Nd:YAG laser is applied to the trabeculum to enhance aqueous outflow and lower IOP. The therapeutic effect is highly variable, and when effective tends to be transient, although typically lasting for months to years. The following are the main indications:

1 Intolerance of topical medication including allergy.
2 Failure of medical therapy, as a less aggressive treatment measure than surgery.
3 Avoidance of polypharmacy, usually with more than two preparations. In this situation laser therapy may be considered as a substitute for an additional drug.
4 Avoidance of surgery, for example in:
Patients in whom laser may defer the need for surgery beyond life expectancy.
Patients in whom filtration surgery carries a poorer prognosis.
5 Primary therapy in accordance with patient preference, or in patients who are unable or unwilling to comply with medical therapy. Since IOP reduction with laser is seldom greater than 30%, an IOP higher than 28 mmHg is unlikely to be adequately controlled by laser alone.

Surgery

Trabeculectomy is the surgical procedure most commonly performed for glaucoma, although non-penetrating surgery is gaining in popularity. If significant lens opacity is present, phacoemulsification alone may be associated with a fall in IOP; alternatively it can be combined with a filtration procedure (phacotrabeculectomy). Progressive damage is thought to be less likely after surgery than with medical therapy, probably because the resultant IOP is often significantly lower and less likely to fluctuate, and because compliance is no longer a factor. The following are the main indications:

1 Failed medical therapy when laser trabeculoplasty is likely to be inadequate or inappropriate.
2 Intolerance of or allergy to medical therapy when trabeculoplasty is inappropriate.
3 Avoidance of polypharmacy, though the option of using three or more medications should be discussed with the patient.
4 Progressive deterioration despite seemingly adequate IOP control.
5 Primary therapy. Advanced disease requiring a very low target pressure may achieve a superior long-term outcome from early surgery, though the risks must be carefully assessed on an individual basis.
6 Patient preference. Occasionally patients express a strong desire to be free of the commitment to chronic medical treatment.

Prognosis

The great majority of patients diagnosed with POAG will not become blind in their lifetime, but the rate of glaucoma progression varies considerably.

Untreated, the mean time for progression to blindness has been estimated at 20 years.
The average period from diagnosis to death is around 15 years.
Older 20 year follow-up data for functional blindness in one eye showed that this occurred in 25% of patients, with blindness in both eyes in 10%, but recent figures are considerably lower, with about 15% of patients’ worse eyes becoming blind; ongoing advances in management are likely to lead this to fall further.

Normal-pressure glaucoma

Definition

Normal-pressure glaucoma (NPG), also referred to as normal- or low-tension glaucoma, is a variant of POAG. It is characterized by:

IOP consistently equal to or less than 21 mmHg.
Signs of optic nerve damage in a characteristic glaucomatous pattern.
An open anterior chamber angle.
Visual field loss as damage progresses, consistent in pattern with the nerve appearance.
No features of secondary glaucoma or a non-glaucomatous cause for the neuropathy.

The distinction between NPG and POAG is based on an epidemiologically-derived range of normal IOP. It is essentially an arbitrary division and may not have significant clinical value, though there is the possibility that a spectrum exists in which, towards the NPG end, IOP-independent factors are of increasing relative importance.

Pathogenesis

Any aetiological factors distinct from those in POAG have not been conclusively determined although various mechanisms have been postulated including anomalies of local and systemic vascular function, structural optic nerve anomalies and autoimmune disease. With the introduction of widespread central corneal thickness (CCT) assessment, NPG in some patients has been explained by very low CCT, and overall CCT in patients with NPG is lower than in POAG. A small proportion of NPG patients have been found to have marked nocturnal IOP spikes, sometimes only detected on testing in the supine position.

Page 347

Risk factors

1 Age. Patients tend to be older than those with POAG, though this may be due to delayed diagnosis.
2 Gender. Some studies have found a higher prevalence in females.
3 Race. NPG occurs more frequently in Japan than in Europe or North America.
4 Family history. The prevalence of POAG is greater in families of patients with NPG than in the normal population. Mutations in the OPTN gene coding for optineurin have been identified in some patients with NPG as well as patients with POAG.
5 CCT is lower in patients with NPG than POAG.
6 Abnormal vasoregulation, particularly migraine and Raynaud phenomenon, have been found more commonly in NPG than POAG by some investigators; others have found abnormalities just as commonly in POAG.
7 Systemic hypotension including nocturnal blood pressure dips of >20%, particularly in those on oral hypotensive medication.
8 Obstructive sleep apnoea syndrome may be associated, perhaps via an effect on ocular perfusion.
9 Autoantibody levels have been found to be higher in NPG patients than the general population by some investigators.

Differential diagnosis

1 POAG presenting with apparently normal IOP because of wide diurnal fluctuation. Plotting a diurnal IOP curve over an 8-hour period (phasing) during office hours may detect daytime elevation, but detection of nocturnal IOP spikes requires substantial resource commitment.
2 Spontaneously resolved pigmentary glaucoma. The typical examination features of pigmentary glaucoma tend to become less evident with increasing age.
3 Previous episodes of raised IOP may have occurred as a result of ocular trauma, uveitis or local or systemic steroid therapy.
4 Masking by systemic treatment such as an oral beta-blocker, commenced after glaucomatous damage has already been sustained.
5 Progressive retinal nerve fibre defects not due to glaucoma such as may occur in myopic degeneration and optic disc drusen.
6 Congenital disc anomalies simulating glaucomatous cupping such as disc pits and colobomas.
7 Neurological lesions causing optic nerve or chiasmal compression can produce visual field defects which may be misinterpreted as glaucomatous, and neuroimaging should be performed if there is any suspicion; some practitioners routinely perform a cranial MRI in all cases of NPG.
8 Previous anterior ischaemic optic neuropathy (AION) may give rise to a disc appearance and visual field defect consistent with glaucoma. Non-arteritic AION often occurs in a ‘crowded’ disc, and the fellow eye should be examined for this.
9 Previous acute optic nerve insult, such as hypovolaemic or septicaemic shock, or head injury.

Diagnosis

History and examination are essentially the same as for POAG but specific points warrant attention:

1 History
Migraine and Raynaud phenomenon.
Episodes of shock.
Head injury.
Headache and other neurological symptoms (intracranial lesion).
Medication such as systemic steroids and hypotensives including beta-blockers.
2 IOP is usually in the high teens, but may rarely be in the low teens. In asymmetrical disease the more damaged disc typically corresponds to the eye with the higher IOP.
3 Optic nerve head
Although the optic nerve head may be larger in NPG than in POAG glaucomatous cupping is similar and acquired optic disc pits are possibly more common.
Peripapillary atrophic changes may be more prevalent.
Splinter haemorrhages (see Fig. 10.20F) may be more frequent than in POAG.
4 Visual field defects are essentially the same as in POAG although it has been suggested that they tend to be closer to fixation, deeper, steeper and more localized. In probably more than half of patients, field changes are non-progressive over a period of 5 years or more without treatment. However, possibly because of delayed diagnosis, patients tend to present with more advanced damage than those with POAG.
5 Other investigations are as for POAG although in selected patients the following can be considered.
Assessment of systemic vascular risk factors.
24-hour ambulatory BP monitoring to exclude nocturnal systemic hypotension.
Blood tests for other causes of non-glaucomatous optic neuropathy such as vitamin B12, red cell folate, full blood count, erythrocyte sedimentation rate/C-reactive protein, treponemal serology including Lyme disease, and serum ACE level, plasma protein electrophoresis and autoantibody screen.
Cranial MR.
Nail-fold capillaroscopy with cold provocation to detect blood flow abnormality. If present, calcium channel blockers may be more likely to be beneficial.

Treatment

Page 348

Further lowering of IOP is effective in reducing progression in some patients. However, as many untreated patients will not deteriorate, in most cases progression should be demonstrated before commencing treatment. Exceptions to this general principle include advanced damage, particularly if threatening central vision, and young age. Regular assessment including perimetry should be performed at 4–6 monthly intervals. With treatment that reduces IOP by 30% from baseline, 80% of patients are stable and 20% show progression.

1 Medical treatment in progressive cases may include betaxolol because of its beneficial effects on optic nerve blood flow in addition to its IOP-lowering properties. Prostaglandin derivatives tend to have a greater ocular hypotensive effect, which may be an over-riding consideration. It should be noted that topical beta-blockers can have a dramatic effect on BP in a minority of patients, and may contribute to nocturnal dips.
2 Laser trabeculoplasty can be effective.
3 Surgery should be considered if progression occurs despite IOP in the low teens.
4 Control of systemic vascular disease such as diabetes, hypertension and hyperlipidaemia may be important, in order theoretically to optimize optic nerve perfusion.
5 Systemic calcium-channel blockers have been advocated by some authorities to address vasospasm.
6 Antihypotensive measures. If significant nocturnal dips in BP are detected, it may be necessary to reduce antihypertensive medication (especially if taken at bedtime).

Primary angle-closure glaucoma

Introduction

Overview

The term ‘angle-closure’ refers to occlusion of the trabecular meshwork (TM) by the peripheral iris (iridotrabecular contact – ITC), obstructing aqueous outflow. Angle-closure can be primary, when it occurs in an anatomically predisposed eye, or secondary to another ocular condition. Primary angle-closure glaucoma may be responsible for up to half of all cases of glaucoma globally, with a particularly high prevalence in individuals of Far Eastern descent. It is typically associated with greater speed of progression and visual morbidity than POAG.

Classification

Recently classification has moved away from a symptom-based approach (acute, subacute and chronic), to reflect the stages in the natural history of the disease. This takes into account that the majority of patients are asymptomatic.

1 Primary angle-closure suspect (PACS)
Gonioscopy shows posterior meshwork ITC in three or more quadrants but no PAS.
Many patients with less ITC have evidence of intermittent angle-closure, and a lower threshold for diagnosis such as two quadrants of ITC or even a very narrow angle approach of 20° may be justified (Fig. 10.40).
Normal IOP, optic disc and visual field.
2 Primary angle-closure (PAC)
Gonioscopy shows three or more quadrants of ITC with raised IOP (Fig. 10.41) and/or PAS, or excessive pigment smudging on the TM.
Normal optic disc and field.
3 Primary angle-closure glaucoma (PACG)
Gonioscopy shows ITC in three or more quadrants.
Optic neuropathy.
image

Fig. 10.40 Primary angle-closure suspect. (A) Very narrow angle; (B) van Herick grade 3

(Courtesy of L MacKeen – fig. A)

Mechanism

The mechanisms involved in angle-closure can be categorized according to the anatomical level (anterior to posterior) at which causative forces act. In many patients more than one level is contributory.

1 Pupillary block
Failure of aqueous flow through the pupil (relative pupillary block – Fig. 10.42A), leads to a pressure differential between the anterior and posterior chambers, with resultant anterior bowing of the iris (Fig. 10.42B) and ITC (Fig. 10.42C).
Iridotomy equalizes anterior and posterior chamber pressure, provided the TM remains sufficiently functional.
2 Non-pupillary block relating to the iris
Specific anatomical factors include plateau iris (anteriorly positioned ciliary processes), and a thicker or more anteriorly-positioned iris.
An element of pupillary block is invariably present, but angle-closure is not fully relieved by iridotomy.
The term ‘mixed mechanism’ may be used to describe glaucoma in which both significant pupillary block and non-pupillary block iris-induced mechanism coexist.
Associated with a deeper anterior chamber than pure pupillary block.
Plateau iris configuration is characterized by a flat central iris plane in association with normal central anterior chamber depth. The angle recess is very narrow, with a sharp iris angulation over anteriorly positioned and/or orientated ciliary processes (Fig. 10.43).
Plateau iris syndrome describes the occurrence of angle-closure despite a patent iridotomy in a patient with morphological plateau iris.
3 Lens-induced angle-closure
Includes only those cases in which a sudden change in lens volume and/or position leads to an acute or subacute IOP rise.
Usually rapid progression of lens intumescence (phacomorphic glaucoma) or anterior lens subluxation.
The lens contributes to angle-closure in categories 1 and 2 and virtually all pupillary block can be said to have a phacomorphic element that increases with age as the lens enlarges.
4 Retrolenticular causes
Malignant glaucoma (‘ciliolenticular block’ – see below).
Posterior segment causes of secondary angle-closure (see below).
5 ‘Combined mechanism’ describes the combination of angle-closure and open-angle elements, although it is generally not possible to determine whether sustained elevation of IOP following successful anatomical opening of an angle is due to TM changes secondary to prior iris apposition.
image

Fig. 10.42 Mechanism of angle-closure. (A) Relative pupillary block; (B) anterior iris bowing; (C) iridocorneal contact

image

Fig. 10.43 Ultrasound biomicroscopy in plateau iris configuration shows loss of the ciliary sulcus due to anterior position of the ciliary processes

(Courtesy of J Schuman, V Christopoulos, D Dhaliwal, M Kahook and R Noecker, from Lens and Glaucoma, in Rapid Diagnosis in Ophthalmology, Mosby 2008)

Risk factors

1 Age
The average age at presentation is about 60 years and the prevalence increases thereafter.
Non-pupillary block forms of primary angle-closure tend to occur at a younger age.
2 Gender. Females are more commonly affected than males.
3 Race. As discussed above.
4 Family history. First degree relatives are at increased risk.
5 Refraction. Eyes with ‘pure’ pupillary block are typically hypermetropic, although this is not as clear-cut with non-pupillary block.
6 Axial length
Short eyes tend to have shallow anterior chambers (Fig. 10.44).
Eyes with nanophthalmos have a very short eye with a proportionally large lens and are at particular risk.
image

Fig. 10.44 Shallow anterior chamber

Diagnosis

Symptoms

Most patients with angle-closure are asymptomatic, including a majority of those with intermittently or chronically elevated IOP.
Some patients present acutely (congestive glaucoma) with haloes around lights due to corneal oedema, ocular pain and headache.
Other patients may have intermittent milder symptoms of blurring (‘smoke-filled room’) unassociated with pain.
Precipitating factors include watching television in a darkened room, reading, pharmacological mydriasis or miosis, acute emotional stress and rarely systemic medication: parasympathetic antagonists or sympathetic agonists (e.g. inhalers, motion sickness patches and cold remedies) and topiramate.

Signs

1 Chronic presentation
VA is normal unless damage is advanced.
Anterior chamber is shallower in pupillary block than non-pupillary block.
Optic nerve signs depend on severity of damage.
IOP elevation may be only intermittent.
Gonioscopic abnormalities are as described above.
‘Creeping’ angle-closure is characterized by a gradual band-like anterior advance of the apparent insertion of the iris. It starts in the deepest part of the angle and spreads circumferentially.
Episodic (intermittent) ITC is associated with the formation of discrete PAS, individual lesions having a pyramidal (‘saw-tooth’) appearance.
2 Acute (congestive) angle-closure
VA usually 6/60-HM.
IOP is usually very high (50–100 mmHg).
Conjunctival hyperaemia with violaceous circumcorneal injection.
Anterior chamber is shallow and aqueous flare may be present.
Corneal epithelial oedema (Fig. 10.45A).
Unreactive mid-dilated vertically oval pupil (Fig. 10.45B).
Fellow eye generally shows an occludable angle (see Fig. 10.40).
3 Resolved acute (post-congestive) angle closure
Folds in Descemet membrane (Fig. 10.46A), (if IOP has been reduced rapidly), optic nerve head congestion and choroidal folds.
Later there is iris atrophy with a spiral-like configuration, irregular pupil, posterior synechiae (Fig. 10.46B) and glaukomflecken (Fig. 10.46C).
The optic nerve may be normal or exhibit varying degrees of atrophy (Fig. 10.46D).
If PAS cover more than half of the TM, IOP control is unlikely with medical treatment alone.
4 Subacute angle closure in which signs between episodes are similar to those of the chronic presentation although occasionally post-congestive-type findings may be present indicating one or more episodes of very high IOP.
image

Fig. 10.45 Acute (congestive) angle-closure. (A) Corneal epithelial oedema; (B) mid-dilated vertically oval pupil

image

Fig. 10.46 Post-congestive angle-closure. (A) Stromal corneal oedema and folds in Descemet membrane; (B) glaukomflecken; (C) spiral-shaped atrophic iris, dilated pupil and posterior synechiae; (D) optic atrophy

Provocation testing

Although investigations are usually unnecessary provocation testing may aid decision-making in some circumstances. For example, in patients with only partially opened angles following laser iridotomy, to assess the propensity to develop a steep increase in IOP and so determine whether further intervention (e.g. iridoplasty) might be appropriate. In the dark room/prone provocation test the patient sits face down in a dark room for one hour. The IOP is checked and a rise of 8 mmHg or more is taken as being of significance but may also sometimes occur in normal eyes. A positive response is virtually always abolished following lens extraction.

Treatment

Primary angle-closure suspect (PACS)

1 Prophylactic laser iridotomy is recommended. Iridotomy often widens the angle by about two grades (Fig. 10.47A and B), although this may not necessarily prevent the later development of angle-closure or raised IOP with an open angle.
2 If significant ITC persists after iridotomy, optimal management remains undefined; options include observation, laser iridoplasty, prophylaxis with pilocarpine 1% and lens extraction.
image

Fig. 10.47 Effects of peripheral iridotomy. (A) Closed angle before treatment; (B) open angle after treatment

Chronic presentation of PAC and PACG

Management is as for PACS, but with a lower threshold for further intervention if there is inadequate angle widening after laser iridotomy, particularly if IOP remains elevated.
Urgency and intensity of treatment and frequency of review is tailored to the individual patient, taking into account IOP, extent of angle-closure and glaucomatous damage, if present.
Medical treatment as for POAG may be required for eyes with substantial synechial closure or with persistently elevated IOP despite an opened angle.

Acute and subacute presentation of PAC and PACG

Treatment intensity should be individualized dependent on severity. Hospital admission is usually required in an acute presentation, though not necessarily when subacute.

1 Initial treatment
The patient should assume a supine position to encourage the lens to shift posteriorly under the influence of gravity.
Acetazolamide 500 mg is given intravenously if IOP >50 mmHg, and orally (not slow-release) if IOP is <50 mmHg.
If treatment is intravenous an additional oral dose of acetazolamide 500 mg may be given.
Topical apraclonidine 1%, timolol 0.5%, prednisolone 1% or dexamethasone 0.1% to the affected eye, leaving 5 minutes between each instillation.
Pilocarpine 2–4% one drop to the affected eye, repeated after half an hour, and one drop of 1% as prophylaxis into the fellow eye.
Some practitioners omit pilocarpine in an acutely presenting eye with very high IOP until a significant IOP fall has taken place, as the associated ischaemia will compromise the action of pilocarpine on the pupillary sphincter.
Analgesia and an antiemetic may be required.
2 Subsequent medical treatment
Pilocarpine 2% q.i.d. to the affected eye and 1% q.i.d. to the fellow eye.
Topical steroid (prednisolone 1% or dexamethasone 0.1%) q.i.d. if the eye is acutely inflamed.
Any or all of the following should be continued as necessary according to response: timolol 0.5% b.d., apraclonidine 1% t.i.d. and oral acetazolamide 250 mg q.i.d. may be required.
3 If the above measures fail consider the following:
Central corneal indentation with a squint hook or indentation goniolens to force aqueous into the angle may break an attack.
Corneal oedema can be cleared with topical 50% glycerol first to improve visualization and avoid abrasion.
Further pilocarpine 2–4%, timolol 0.5%, apraclonidine 1% and topical steroid.
Mannitol 20% 1–2 g/kg intravenously over 1 hour or glycerol 50% 1 g/kg, having checked for contraindications.
Laser iridotomy or iridoplasty after clearing corneal oedema with glycerol.
Surgical options in resistant cases include peripheral iridectomy, lens extraction, goniosynechialysis, trabeculectomy and cyclodiode.
4 Following successful treatment with a clear cornea, reasonably quiet anterior chamber, and preferably a normalized IOP, bilateral laser iridotomy is performed. Topical steroids and any necessary hypotensives are continued for at least a week.
5 Subsequent management is as for post-iridotomy chronic PAC/PACG. Options including observation, treatment of persistently raised IOP as for POAG, iridoplasty or long-term low dose pilocarpine if appositional closure persists. A relatively low threshold may be adopted for cataract surgery, particularly if a significant phacomorphic element is suspected. Trabeculectomy is occasionally necessary for persistent IOP elevation despite a successfully opened angle.

Differential diagnosis of an acute elevation of IOP

A key indicator that PAC/PACG may not be responsible is an open angle in a fellow eye.

1 Lens-induced angle-closure due to an intumescent (swollen) or subluxated lens.
2 Malignant glaucoma, especially if intraocular surgery has recently taken place (usually trabeculectomy).
3 Other causes of secondary angle-closure, with or without pupillary block; see below.
4 Neovascular glaucoma may occasionally cause the sudden onset of pain and congestion.
5 Inflammatory elevation with an open angle. Iridocyclitis with trabeculitis (particularly herpetic), glaucomatocyclitic crisis (Posner–Schlossman syndrome), scleritis without angle-closure.
6 Pigment dispersion with sudden elevation of IOP.
7 Pseudoexfoliation with sudden elevation of IOP.
8 Orbital/retro-orbital lesions including orbital inflammation, retrobulbar haemorrhage and carotid-cavernous fistula.
9 Others include atypical presentation of secondary open-angle glaucoma (see below).

Classification of secondary glaucoma

Open-angle

Secondary open-angle glaucoma can be subdivided on the basis of the site of aqueous outflow obstruction as follows:

1 Pre-trabecular glaucoma in which aqueous outflow is obstructed by a membrane covering the trabeculum (Fig. 10.48A), which may consist of:
Fibrovascular tissue (neovascular glaucoma).
Endothelial cellular membranous proliferation (iridocorneal endothelial syndrome).
Epithelial cellular membranous proliferation (epithelial ingrowth).
2 Trabecular glaucoma in which the obstruction occurs as a result of ‘clogging up’ of the meshwork (Fig. 10.48B) by the following:
Pigment particles (pigmentary glaucoma).
Red blood cells (red cell glaucoma).
Degenerate red cells (ghost cell glaucoma).
Macrophages and lens proteins (phacolytic glaucoma).
Proteins (hypertensive uveitis).
Pseudoexfoliative material (pseudoexfoliation glaucoma).
image

Fig. 10.48 Pathogenesis of secondary glaucoma. (A) Pre-trabecular obstruction; (B) trabecular obstruction; (C) angle-closure with pupillary block; (D) angle-closure without pupillary block

Trabecular glaucomas may also be caused by alteration of the trabecular fibres themselves by:

Oedema (herpes zoster iritis/trabeculitis).
Scarring (post-traumatic angle recession glaucoma).
3 Post-trabecular glaucoma in which the trabeculum itself is normal but aqueous outflow is impaired as a result of elevated episcleral venous pressure due to conditions such as:
Carotid-cavernous fistula.
Sturge–Weber syndrome.
Obstruction of the superior vena cava.

Angle-closure

Secondary angle-closure is caused by impairment of aqueous outflow secondary to apposition between the peripheral iris and the trabeculum. Classification is based according to the presence or absence of pupillary block:

1 With pupillary block (Fig. 10.48C)
Seclusio pupillae (360° ‘ring’ posterior synechiae), usually secondary to recurrent iridocyclitis.
Subluxated lens.
Phacomorphic glaucoma.
Capsular block syndrome with 360° iris-capsule adhesion.
Aphakic pupillary block.
Anterior chamber lens implant without a patent iridotomy.
2 Without pupillary block (Fig. 10.48D)
Secondary causes of PAS such as advanced neovascular glaucoma and chronic anterior uveitis.
Cilio-choroidal effusion.
Capsular block syndrome without iris-capsule adhesion.
Page 355
Ciliary body/iris cyst or other ciliary body or posterior segment tumour.
Contraction of retrolenticular fibrovascular tissue such as in proliferative vitreoretinopathy and retinopathy of prematurity.
‘Malignant’ glaucoma may arbitrarily be considered a form of secondary rather than primary angle-closure.

Most of the above conditions are described later.

Pseudoexfoliation

Pseudoexfoliation syndrome

Introduction

The pseudoexfoliation syndrome (PXF), sometimes known as exfoliation syndrome, is a relatively common cause of chronic open-angle glaucoma, though subtle signs are easily overlooked. When an eye with PXF develops secondary open-angle glaucoma the condition is referred to as pseudoexfoliation glaucoma (PXG). PXF is more common in females but males appear to be at higher risk of developing glaucoma. The condition is particularly common in Scandinavia. A high risk of developing PXF and PXG is conferred by mutations in the LOXL1 gene at locus 15q22, coding for elastic fibre components of the extracellular matrix. The cumulative risk of glaucoma in eyes with PXF is 5% at 5 years and 15% at 10 years.

Pathogenesis

A grey-white fibrillary extracellular material composed of a protein core surrounded by glycosaminoglycans is produced by abnormal basement membranes of ageing epithelial cells in the trabeculum, equatorial lens capsule, iris and ciliary body. The material is then deposited on the anterior lens capsule (Fig. 10.49A), zonules, ciliary body, iris, trabeculum, anterior vitreous face and conjunctiva. In addition to its occurrence within the eye, exfoliative fibrillopathy has been reported in skin and visceral organs, suggesting that PXF may be an ocular manifestation of a systemic disorder; PXFS is associated with an increasing number of vascular disorders, hearing loss and Alzheimer disease.

image

Fig. 10.49 Pseudoexfoliation syndrome. (A) Christmas-tree like deposits of pseudoexfoliative material (PXF) on the lens capsule; (B) PXF on the pupillary margin; (C) transillumination defect corresponding to sphincter iris atrophy; (D) PXF on the lens; (E) gonioscopy shows patchy trabecular hyperpigmentation and Sampaolesi line

(Courtesy of J Harry and G Misson, from Clinical Ophthalmic Pathology, Butterworth-Heinemann 2001– fig. A; M Jager – fig. D; J Schuman, V Christopoulos, D Dhaliwal, M Kahook and R Noecker, from Lens and Glaucoma, in Rapid Diagnosis in Ophthalmology, Mosby 2008 – fig. F)

Diagnosis

1 Cornea occasionally shows PXF on the endothelium as well as pigment deposition that is usually diffuse although occasionally may take the form of a Krukenberg spindle.
2 Mild aqueous flare is sometimes seen, and results from breakdown of the iris blood-aqueous barrier.
3 Iris shows PXF on the pupillary margin (Fig. 10.49B) and sphincter atrophy characterized by ‘moth-eaten’ transillumination defects at the pupillary margin (Fig. 10.49C).
4 PXF on the anterior lens surface
The constant rubbing of the pupil scrapes the material off the midzone of the lens giving rise to a central disc and a peripheral band of PXF, with a clear zone between (Fig. 10.49D).
The peripheral band is granular and has a well-delineated inner border with multiple radial striations. It can be detected only after the pupil has been dilated.
Cataract surgery is more hazardous due to a combination of poorly dilating pupil, increased risk of zonular dialysis and capsular tear. Other problems include a postoperative pressure spike, corneal oedema, increased incidence of capsular opacification and contraction, and IOL subluxation.
6 Gonioscopy
Trabecular hyperpigmentation is common and is usually most marked inferiorly. It may antedate the appearance of PXF by several years. The pigment lies on the surface of the trabeculum and has a patchy distribution (Fig. 10.49E).
A scalloped band of pigment running on to or anterior to Schwalbe line (Sampaolesi line) is common.
PXF deposits in the trabeculum can give rise to a ‘dandruff-like’ appearance.
Narrow angles are present in some cases, and there is an increased risk of angle closure, probably due to zonular laxity.

Pseudoexfoliation glaucoma

Pathogenesis

Probable causes of elevation of IOP include trabecular blockage due to a combination of ‘clogging up’ of the trabeculum by PXF material and/or pigment released from the iris.

Diagnosis

1 Presentation is usually in the 7th decade.
2 Signs. The majority of patients have a chronic open-angle glaucoma which is usually unilateral. Occasionally the IOP may rise acutely despite a wide-open angle and may be confused with primary angle-closure. There is no apparent association between angle characteristics and the severity of glaucoma, unless angle-closure develops.

Treatment

1 Medical treatment is the same as for POAG. However, despite initial success in most cases, there is a high incidence of late failure and patients are more likely to require laser therapy or surgery.
2 Laser trabeculoplasty is particularly effective, possibly because of trabecular hyperpigmentation. However, following an initial good response a gradual late rise of IOP occurs so that after 4 years the results are the same as in POAG.
3 Trabeculectomy has the same success rate as in POAG.
4 Trabecular aspiration with light tissue contact confers at least a short-term benefit, and can be performed at the same time as cataract surgery or trabeculectomy.
Page 357

Prognosis

The prognosis is worse than in POAG; the IOP is often significantly elevated and may also exhibit great fluctuation. Severe damage may develop rapidly. It is therefore important to monitor patients closely, and some practitioners feel that review should take place at intervals of no more than 6 months for patients with PXF.

A patient with unilateral PXG and only PXF in the fellow eye is at high risk (50% in 5 years) of developing glaucoma in the fellow eye.
A patient with unilateral PXG who does not have PXF in the fellow eye has only a low risk of developing glaucoma in the normal eye.

Pigment dispersion

Pigment dispersion syndrome

Introduction

Pigment dispersion syndrome (PDS) is a usually bilateral condition characterized by the liberation of pigment granules from the iris pigment epithelium and their deposition throughout the anterior segment. PDS primarily affects whites and may be inherited as AD with variable penetrance. There is a significant linkage between the disease phenotype and genetic markers located on 7q35-36. Myopia predisposes to the phenotypical manifestations and the development of a secondary open-angle ‘pigmentary’ glaucoma. However, some manifestations of PDS may be extremely subtle and go undetected.

Pathogenesis

Pigment shedding is caused by the mechanical rubbing of the posterior pigment layer of the iris against packets of lens zonules as a result of excessive posterior bowing of the mid-peripheral portion of the iris. It is postulated that an increase in anterior chamber pressure (relative to the posterior chamber) occurs due to ‘reverse pupil block’, with resultant posterior bowing of the iris and iridozonular touch (Fig. 10.50A). This is supported by the observation that neutralization of reverse pupil block with peripheral iridotomy flattens the iris and decreases iridozonular contact (Fig. 10.50B). The pigment epithelium itself may be abnormally susceptible to shedding. In some patients strenuous exercise may precipitate episodes of pigment dispersion associated with a rise in IOP. Pigment granules are released into the aqueous humour, dispersed by aqueous currents and deposited on all anterior chamber structures, including the zonular fibres and ciliary body.

image

Fig. 10.50 High-frequency ultrasonography in pigment dispersion syndrome. (A) Very deep anterior chamber and posterior bowing of the peripheral iris; (B) flattening of the peripheral iris following laser iridotomy

(Courtesy of J Salmon)

Diagnosis

1 Cornea shows pigment deposition on the endothelium, in a vertical spindle-shaped distribution (Krukenberg spindle) (Fig. 10.51A). This finding, although common, is neither universal nor pathognomonic of PDS, and in long-standing cases may be more difficult to detect because it tends to become smaller and lighter in colour.
2 Anterior chamber is very deep (Fig. 10.51B) and melanin granules may be seen floating in the aqueous.
3 Iris
Fine surface pigment granules that may extend onto the lens; partial loss of the pupillary ruff (frill) (Fig. 10.51C).
Pigment epithelial atrophy due to shedding of pigment from the mid-periphery gives rise to characteristic radial slit-like transillumination defects (Fig. 10.51D).
4 Gonioscopy
The angle is wide open and there is a characteristic mid-peripheral iris concavity that may increase with accommodation.
Trabecular hyperpigmentation is most marked over the posterior trabeculum (Fig. 10.51E). The pigment is finer than in PXF and appears to lie both on and within the trabecular meshwork. It also has a more homogeneous appearance and forms a dense band involving the entire circumference of the meshwork uniformly. Pigment may also be seen on or anterior to Schwalbe line.
5 Lens may occassionally show a line or an annular ring of pigment on the peripheral posterior surface (Scheie line).
image

Fig. 10.51 Pigment dispersion syndrome. (A) Krukenberg spindle; (B) very deep anterior chamber; (C) pigment granules on the surface of the iris and partial loss of the pupillary ruff; (D) radial slit-like transillumination defects; (E) homogeneous trabecular hyperpigmentation

Page 359

Pigmentary glaucoma

Pathogenesis

Elevation of IOP appears to be caused by pigmentary obstruction of the intertrabecular spaces and damage to the trabeculum secondary to denudation, collapse and sclerosis.

Risk factors

About a third of patients with PDS develop ocular hypertension or chronic open-angle glaucoma after 15 years. Men are affected twice as frequently as women. The optic disc may be more susceptible to the effects of elevated IOP because of the underlying myopia. It is therefore important to regularly follow patients with the condition, particularly myopic males with Krukenberg spindles. However, initial IOP, cup–disc ratio and degree of trabecular hyperpigmentation are not helpful in identifying those who will eventually develop glaucoma. Patients with pigmentary glaucoma have an increased incidence of steroid responsiveness.

Diagnosis

1 Presentation is usually with chronic glaucoma most commonly in the 3rd and 4th decades although in women it tends to develop around 10 years later. Occasionally the sudden release of pigment granules spontaneously or following strenuous physical exercise may precipitate an acute rise in IOP, with corneal oedema and haloes.
2 IOP may initially be very unstable, so that a single normal reading does not exclude glaucoma. Some patients exhibit higher levels and wider fluctuations of IOP than in POAG, and at the time of diagnosis it is common to find advanced disease in one eye and relatively mild damage in the other.

Treatment

1 Medical treatment is similar to that of POAG. Miotics would theoretically be of particular benefit because they decrease iridozonular contact in addition to facilitating aqueous outflow. They have the disadvantages, however, of exacerbating the myopia common in these patients and also of a risk of precipitating retinal detachment in myopia. They are not well tolerated by young patients. Topical thymoxamine, a selective alpha-adrenergic antagonist, induces miosis without causing spasm of accommodation, but is also poorly tolerated as it causes irritation.
2 Laser trabeculoplasty is often initially effective although it is important not to over-treat eyes with heavily pigmented angles and to start at a relatively low power laser setting. At least one-third of patients will require trabeculectomy within 5 years of laser trabeculoplasty.
3 Laser iridotomy has been proposed to retard pigment liberation by reversing iris concavity and eliminating iridozonular contact (see Fig. 10.50). It may have utility in patients under the age of 40 years but benefit has not been conclusively demonstrated.
4 Trabeculectomy is indicated more commonly than in POAG although the results can be disappointing, perhaps at least partly because most patients are relatively young. The use of adjunctive antimetabolites may improve surgical outcome.

Prognosis

Over time the control of IOP becomes easier and occasionally the IOP may spontaneously revert to normal; this may or may not be associated with a decrease in trabecular pigmentation. Patients with undetected previous pigmentary glaucoma may later be erroneously diagnosed as having NPG.

Differential diagnosis

1 POAG may be associated with a hyperpigmented trabeculum. However, the pigment tends to be concentrated in the inferior sector of the angle, in contrast to the homogeneous distribution in PDS. Patients with POAG are also usually older and lack Krukenberg spindles and iris transillumination defects.
2 Pseudoexfoliation may exhibit trabecular hyperpigmentation and pigment dispersion. However, transillumination defects are evident at the margin of the pupil rather than in the periphery. Pseudoexfoliation glaucoma usually affects patients over the age of 60 years, is unilateral in 50% of cases and has no predilection for a myopic refractive error.
3 Pseudophakic pigmentary glaucoma occurs in the context of rubbing of the haptics and optics of a posterior chamber intraocular lens against the posterior surface of the iris, with resultant pigment dispersion and outflow obstruction.
4 Anterior uveitis may result in trabecular hyperpigmentation and iris atrophy. Clustered old pigmented keratic precipitates may be mistaken for a Krukenberg spindle on cursory examination.
5 Subacute angle-closure glaucoma may be associated with a heavily pigmented trabeculum where the iris root has been in contact with the angle.

Neovascular glaucoma

Introduction

Pathogenesis

Page 360

Neovascular glaucoma (NVG) is an aggressive condition which occurs as a result of iris neovascularization (rubeosis iridis). The common aetiological factor is severe, diffuse and chronic retinal ischaemia. It is postulated that hypoxic retinal tissue produces growth factors in an attempt to revascularize hypoxic areas; the most important of these is vascular endothelial growth factor (VEGF). Apart from inducing retinal neovascularization (proliferative retinopathy) such factors also diffuse into the anterior segment and initiate rubeosis iridis and neovascularization in the angle of the anterior chamber. The latter initially impairs aqueous outflow in the presence of an open angle and later contracts to produce a secondary angle-closure glaucoma which is usually severe and relentless (Fig. 10.52A).

image

Fig. 10.52 Neovascular glaucoma. (A) Rubeosis iridis and angle-closure by PAS; (B) tiny capillary tufts on the pupil margin; (C) invasion of angle structures by new vessels; (D) progressive synechial angle-closure

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

Causes of rubeosis iridis

1 Ischaemic central retinal vein occlusion accounts for about a third of cases. Approximately 50% of eyes develop NVG following ischaemic central retinal vein occlusion. Extensive peripheral retinal capillary non-perfusion on fluorescein angiography is the most valuable predictor of the risk of subsequent NVG, although in some patients non-ischaemic occlusion may subsequently become ischaemic. Glaucoma typically occurs 3 months after the occlusion (‘100-day glaucoma’) but intervals from 4 weeks to 2 years have been documented.
2 Diabetes mellitus accounts for a slightly smaller proportion. Patients with long-standing diabetes (10 years or more) with proliferative retinopathy are at particular risk. The risk of glaucoma is decreased by appropriate panretinal photocoagulation and increased by cataract extraction. Pars plana vitrectomy may also precipitate rubeosis iridis if inadequate laser therapy is applied or tractional retinal detachment remains.
3 Arterial retinal vascular disease such as central retinal artery occlusion and ocular ischaemic syndrome are uncommon causes.
4 Miscellaneous causes include intraocular tumours, long-standing retinal detachment and chronic intraocular inflammation.

Classification

Despite a degree of overlap it is convenient to divide NVG into the following three stages: (a) rubeosis iridis, (b) secondary open-angle glaucoma and (c) secondary synechial angle-closure glaucoma. Systemic investigation and treatment should be tailored to individual causes.

Rubeosis iridis

Diagnosis

In chronological order rubeosis develops as follows:

Tiny dilated capillary tufts or red spots develop at the pupillary margin and may be missed unless the iris is examined carefully under high magnification (Fig. 10.52B).
The new vessels grow radially over the surface of the iris towards the angle, sometimes joining dilated blood vessels at the collarette. At this stage the IOP may still be normal and the new vessels may regress either spontaneously or with treatment.
Angle neovascularization in the absence of pupillary involvement may occur, particularly after an ischaemic central retinal vein occlusion. It is therefore important to perform careful gonioscopy without mydriasis in eyes at high risk even when the pupillary border is uninvolved.
Page 361

Treatment

1 Panretinal photocoagulation (PRP), if performed early, is often effective in inducing regression of the new vessels and preventing subsequent progression to glaucoma.
2 Intravitreal vascular endothelial growth factor (VEGF) inhibitors such as bevacizumab (Avastin®) at a dose of 1.25 mg in 0.05 mL may decrease neovascularization at this stage and thus improve IOP control, although the duration of control is often limited, requiring further injections or definitive control with PRP.
3 Retinal surgery. If rubeosis develops or persists following vitrectomy in a diabetic patient with residual retinal detachment, reattachment should be attempted, since if successful, the rubeosis will frequently regress. Additional panretinal photocoagulation is also beneficial.

Secondary open-angle glaucoma

Diagnosis

Neovascular tissue proliferates across the face of the angle (Fig. 10.52C). Here the new blood vessels arbourize and form a fibrovascular membrane which blocks the trabeculum, giving rise to secondary open-angle glaucoma.

Treatment

1 Medical treatment is as for POAG but miotics should be avoided, and prostaglandin derivatives used with relative caution due to their inflammation-promoting potential. Topical atropine 1% and intensive topical steroids should be given if significant inflammation is present. Topical apraclonidine and/or oral acetazolamide may be required as short-term temporizing measures.
2 Intravitreal VEGF inhibitor injection may be effective if fibrovascular angle-closure has not yet supervened.
3 Cyclodiode should be performed if medical control of IOP is not possible, particularly if the eye is uncomfortable, has useful visual potential or corneal oedema prevents an effective retinal view for PRP.
4 PRP should still be performed even if the IOP is adequately controlled medically, although this will not reverse the fibrous component of the fibrovascular membrane. If the retinal view is poor, indirect ophthalmoscopic application may provide better access, if necessary in the operating theatre with iris hooks to open a small pupil caused by posterior synechiae. Trans-scleral cryotherapy or diode laser are options.

Secondary angle-closure glaucoma

Diagnosis

If rubeosis continues to progress the angle becomes progressively closed by contraction of fibrovascular tissue with pulling of the peripheral iris over the trabeculum (Fig. 10.52D and see Fig. 10.52A). The angle thus closes circumferentially in a zipper-like fashion resulting in very high IOP, severe visual impairment, congestion of the globe and pain. The prognosis for visual function is generally poor by this stage, though aggressive management can achieve comfort and retain useful sight in some cases.

Treatment

1 Medical treatment is as discussed above for the secondary open-angle stage. Steroids and atropine alone may be adequate if there is no potential for vision.
2 Intravitreal VEGF inhibitor injection is generally not thought to be effective once synechial angle-closure is present.
3 Cyclodiode should be considered in the circumstances discussed above.
4 PRP is performed if the fundus can be adequately visualized (see above). Eyes with opaque media can be treated by trans-scleral cryotherapy or cyclodiode, if appropriate.
5 Filtration surgery may be considered if vision is hand movements or better. The options are trabeculectomy with adjunctive mitomycin C and artificial filtering shunts (glaucoma drainage devices).
6 Retrobulbar alcohol injection is useful in relieving pain but it may cause permanent ptosis and does not relieve congestion.
7 Enucleation may be considered if all else fails.

Inflammatory glaucoma

Introduction

Overview

Elevation of intraocular pressure (IOP) secondary to intraocular inflammation frequently presents a diagnostic and therapeutic challenge. The elevation of IOP may be transient and innocuous, or persistent and severely damaging. The prevalence of secondary glaucoma increases with chronicity and severity of disease. Secondary glaucoma is particularly common in Fuchs uveitis syndrome and chronic anterior uveitis associated with juvenile idiopathic arthritis. Posterior uveitis is less likely to affect the aqueous outflow pathway and consequently less likely to lead to IOP elevation.

Classification

1 Angle-closure with pupillary block.
2 Angle-closure without pupillary block.
3 Open-angle.
4 Posner–Schlossman syndrome.

Diagnostic dilemmas

1 IOP fluctuation may be dramatic in uveitic glaucoma and phasing may be helpful in patients with borderline IOP.
Page 362
2 Ciliary body shutdown caused by acute exacerbation of chronic anterior uveitis is frequently associated with lowering of IOP that may mask the underlying tendency to glaucoma. Even eyes with considerably elevated IOP (30–35 mmHg) may become hypotonous during acute exacerbations of uveitis. Return of ciliary body function with subsidence of uveitis may be associated with a rise in IOP in the presence of permanently compromised outflow facility.
3 Pathogenesis of elevation of IOP may be uncertain; multiple mechanisms may be involved. Steroid-responders often represent a therapeutic challenge.
4 Assessment of glaucomatous damage may be hampered by a small pupil or opacities in the media. Poor visual acuity may also compromise accurate perimetry.
5 Iris vessels may give rise to diagnostic confusion with NVG.

Angle-closure glaucoma with pupillary block

Pathogenesis

Secondary angle-closure is caused by posterior synechiae extending for 360° (seclusio pupillae) which obstruct aqueous flow from the posterior to the anterior chamber (Fig. 10.53A). The resultant increased pressure in the posterior chamber produces anterior bowing of the peripheral iris (iris bombé – Fig. 10.53B) resulting in shallowing of the anterior chamber and apposition of the iris to the trabeculum and peripheral cornea (Fig. 10.53C). Such an inflamed iris easily sticks to the trabeculum and the iridocorneal contact may become permanent with the development of PAS.

image

Fig. 10.53 Secondary angle-closure with pupillary block. (A) Seclusio pupillae; (B) iris bombé; (C) iridocorneal contact

Diagnosis

1 Slit-lamp biomicroscopy shows seclusio pupillae, iris bombé and a shallow anterior chamber.
2 Gonioscopy shows angle-closure from iridotrabecular contact. Indentation may be used to assess the extent of appositional as opposed to synechial angle-closure.

Angle-closure glaucoma without pupillary block

1 Pathogenesis. Chronic anterior uveitis causes the deposition of inflammatory cells and debris in the angle (Fig. 10.54A and B). Subsequent organization and contraction pulls the peripheral iris over the trabeculum, thereby causing gradual and progressive synechial angle-closure (Fig. 10.54C) and eventual elevation of IOP. The eye with a pre-existing narrow angle may be at higher risk, as may one with granulomatous inflammation with inflammatory nodules in the angle.
2 Diagnosis. The anterior chamber is deep but gonioscopy shows extensive angle-closure by PAS.
image

Fig. 10.54 Secondary angle-closure without pupillary block. (A) Deposition of inflammatory cells in the angle; (B) gonioscopy shows inflammatory debris; (C) synechial angle-closure

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

Open-angle glaucoma

In acute anterior uveitis

Page 363

In acute anterior uveitis the IOP is usually normal or subnormal due to concomitant ciliary shutdown. Occasionally, however, secondary open-angle glaucoma develops due to obstruction of aqueous outflow, most commonly as the acute inflammation is subsiding and ciliary body function is returning. This effect, which is often transient and innocuous, may be steroid-induced or caused by a combination of the following mechanisms:

1 Trabecular obstruction by inflammatory cells and debris which may be associated with increased aqueous viscosity due to leakage of proteins from the inflamed iris blood vessels.
2 Acute trabeculitis involving inflammation and oedema of the trabecular meshwork with secondary diminution of intertrabecular porosity may result in a reduction in outflow facility. It is thought that this is especially relevant in anterior uveitis associated with herpes zoster, herpes simplex and toxoplasma retinitis.

In chronic anterior uveitis

In chronic anterior uveitis the main mechanism for reduced outflow facility is thought to be trabecular scarring and/or sclerosis secondary to chronic trabeculitis. The exact incidence and importance of this mechanism is, however, difficult to determine as most eyes also have some degree of synechial angle-closure. Because of the variable appearance of the angle on gonioscopy, definitive diagnosis of trabecular damage is difficult. Theoretically, the angle should be open and, in some eyes, a gelatinous exudate resembling ‘mashed potatoes’ is seen on the trabeculum. Treatment is as for secondary synechial angle-closure glaucoma.

Treatment

Medical

Medical control of IOP is more likely to be achieved if the angle is completely open.
The aim of therapy in terms of IOP level to be attained depends on the health of the optic nerve head; eyes with advanced damage require a low target IOP.
In steroid-reactors it is important not to sacrifice control of inflammation for fear of steroid-induced IOP elevation. Long-acting depot preparations should be used with great caution in patients with a history of a steroid response.
The IOP-lowering effect of ocular hypotensive drugs is less predictable in uveitis and some cases may be unexpectedly sensitive to topical carbonic anhydrase inhibitors (CAI).
The use of prostaglandin analogues in uveitic glaucoma is tempered by the small risk of precipitating a uveitic episode and CMO.
Page 364
A beta-blocker is therefore usually the drug of first choice.
The choice of a second line agent often depends on the IOP level. If the IOP is very high, systemic acetazolamide may be required in the short-term. If elevation of IOP is moderate (e.g. less than 35 mmHg on a beta-blocker) in the absence of significant glaucomatous damage, an alpha-adrenergic agonist or a topical CAI might be appropriate.
Miotics are contraindicated as they increase vascular permeability and may promote inflammation, and miosis enhances the formation of posterior synechiae.

Laser iridotomy

Laser iridotomy is performed to re-establish communication between the posterior and anterior chambers in eyes with pupillary-block angle-closure glaucoma. The resulting hole is usually quite small and likely to become occluded in the presence of active uveitis.
It is important to bear in mind that correction of pupillary block may not control the IOP if there is insufficient open-angle for drainage. In cases of progressive angle closure, iridotomy may nevertheless prevent further PAS formation.
Intensive topical steroid therapy should be used to minimize post-laser inflammation.
Surgical iridectomy is the definitive method of preventing further pupil block.

Surgery

1 Preoperative preparation
Control of uveitis for a minimum of 3 months before surgery is ideal but often impractical.
Preoperative topical steroids should be used, not only as prophylaxis against recurrent inflammation, but also to reduce the conjunctival inflammatory cell population.
In patients with particularly labile inflammatory disease systemic steroids should be considered (0.5 mg/kg/day of oral prednisolone).
2 Trabeculectomy is usually the procedure of choice.
Combined cataract and glaucoma surgery is not appropriate. Ideally cataract surgery should be deferred for about 6 months after trabeculectomy.
Adjunctive antimetabolites, particularly mitomycin C, are required since these eyes carry a high risk of failure.
Postoperative hypotony is a risk as a delicate balance may exist between reduced aqueous production and severely restricted aqueous outflow. If production drops in the early postoperative period, any filtration may be excessive.
After trabeculectomy steroids are tapered according to the level of inflammation and the appearance of the filtering bleb, and usually discontinued after 3–6 months, although earlier tapering may be necessary in cases of overfiltration.
3 Glaucoma drainage devices should be considered in cases where trabeculectomy, even with adjunctive antimetabolites, has a poor success rate. This includes aphakic eyes, children with chronic anterior uveitis, or a previously failed trabeculectomy.
5 Cyclodestructive procedures should be used with caution because they may not only exacerbate the intraocular inflammation but can result in profound hypotony, which may proceed to phthisis bulbi. Even eyes with seemingly intractable uveitic glaucoma may paradoxically develop ciliary body insufficiency in the longer term.
6 Angle procedures include trabeculodialysis and goniotomy may be successful in children. The former involves making an incision along Schwalbe line in order to establish communication between the anterior chamber and Schlemm canal (Fig. 10.55).
image

Fig. 10.55 Trabeculodialysis

Posner–Schlossman syndrome

Posner–Schlossman syndrome (glaucomatocyclitic crisis) is characterized by recurrent attacks of unilateral, acute secondary open-angle glaucoma associated with mild anterior uveitis. The cause of the raised IOP is presumed to be acute trabeculitis. There is some evidence that the herpes simplex virus may play a pathogenic role. Posner–Schlossman syndrome is a rare condition typically affecting young adults, 40% of whom are positive for HLA-Bw54. Males are affected more frequently than females. The IOP is elevated for between a few hours and several days. The attacks are unilateral, although 50% of patients have bilateral involvement at different times. The intervals between attacks vary and, with time, usually become longer. Patients should be followed even after the attacks have completely subsided, because a significant percentage will develop chronic open-angle glaucoma.

Diagnosis

1 Presentation is with mild discomfort, haloes around lights and slight blurring of vision.
2 Slit-lamp biomicroscopy shows corneal epithelial oedema due to a high IOP (40–80 mmHg), a few aqueous cells and fine white central keratic precipitates (Fig. 10.56).
3 Gonioscopy shows an open angle.
image

Fig. 10.56 Keratic precipitates in Posner–Schlossman syndrome

Treatment

Topical steroids are used to control the inflammation and aqueous suppressants for the raised IOP. Oral non-steroidal anti-inflammatory agents may also be beneficial.

Lens-related glaucoma

Phacolytic glaucoma

Pathogenesis

Page 365

Phacolytic glaucoma (lens protein glaucoma) is open-angle glaucoma occurring in association with a hypermature cataract. Trabecular obstruction is caused by high molecular weight lens proteins which have leaked through the intact capsule into the aqueous humour. Macrophages containing lens proteins may also contribute to trabecular blockage (Fig. 10.57A and B). Phacolytic glaucoma should not be confused with phacoanaphylactic (phacoantigenic) uveitis which is an autoimmune granulomatous reaction to lens proteins occurring in an eye with a ruptured capsule.

image

Fig. 10.57 Phacolytic glaucoma. (A) Lens protein-containing macrophages in the angle; (B) lens protein-containing macrophages on the corneal endothelium similar to keratic precipitates; (C) hypermature cataract and lens protein-containing macrophages floating in the aqueous; (D) neglected end-stage glaucoma with corneal vascularization and a small pseudohypopyon

(Courtesy of J Harry – figs A and B)

Diagnosis

1 Presentation is with pain; vision is already poor due to cataract.
2 Slit-lamp biomicroscopy shows corneal oedema, a hypermature cataract and a deep anterior chamber. The aqueous may manifest floating white particles (Fig. 10.57C), which may form a pseudohypopyon if very dense (Fig. 10.57D).
3 Gonioscopy shows an open-angle.

Treatment

Once the IOP is controlled medically, the proteinaceous material is flushed out and the cataract removed. Care should be taken not to rupture the zonules when performing anterior capsulotomy.

Page 366

Phacomorphic glaucoma

Pathogenesis

Phacomorphic glaucoma is an acute secondary angle-closure glaucoma precipitated by an intumescent cataractous lens. Equatorial age-related growth of the lens slackens the suspensory ligament and allows the lens to move anteriorly. Associated anteroposterior growth leads to increased iridolenticular contact and potentiates pupillary block and iris bombé.

Diagnosis

1 Presentation is similar to acute PACG with a shallow anterior chamber and dilated pupil; cataract is usually evident (Fig. 10.58).
2 Examination of the fellow eye may demonstrate a deep anterior chamber and an open angle, thus making PACG unlikely, although phacomorphic glaucoma is more likely in eyes with a shorter axial length and shallower anterior chamber.
image

Fig. 10.58 Intumescent cataract, shallow anterior chamber, dilated pupil and corneal oedema in phacomorphic glaucoma

Treatment

Treatment is initially similar to acute PACG, but miotics are omitted as they tend to increase iris-lens apposition and shift the lens anteriorly. Systemic hyperosmotic agents may be required more commonly than in PACG. Laser iridotomy may be worthwhile but is often not possible (due to corneal oedema or lens-cornea proximity) or ineffective. Definitive treatment consists of early cataract extraction, ideally when the IOP is normal and the eye quiet.

Page 367

Lens dislocation into the anterior chamber

Causes

1 Blunt ocular trauma, even if relatively trivial, may result in lens dislocation in eyes with a weak zonule as in pseudoexfoliation and homocystinuria (Fig. 10.59A).
2 Small lenses (microspherophakia) as in Weill–Marchesani syndrome.
image

Fig. 10.59 Lens-induced pupillary block glaucoma. (A) Lens dislocation into the anterior chamber; (B) lens incarceration in the pupil

Diagnosis

The dislocated lens causes acute pupillary block and a sudden and severe elevation of IOP with associated visual impairment. This constitutes an acute emergency because lenticulocorneal contact may cause permanent endothelial damage.

Treatment

The IOP is initially reduced with osmotic agents. Subsequent management is dependent on the absence or presence of some remaining zonular attachments and the hardness of the lens as follows:

1 Intact zonule. The patient is placed into a supine posture and the pupil dilated in an attempt to reposition the lens into the posterior chamber.
2 Soft lens without zonular attachments. A lensectomy is performed through a limbal incision. Lenses in patients over the age of 35 years are usually too hard to be removed by this technique.
3 Hard lens without zonular attachments. A pars plana vitrectomy and lensectomy is performed.

Incarcerated lens in the pupil

1 Pathogenesis. The rise in IOP is caused by pupillary block by a microspherical lens in which only part of the zonule has been disrupted so that the intact zonule acts as a hinge (Fig. 10.59B).
2 Treatment involves relieving pupillary block with mydriatics or Nd:YAG laser iridotomy. Miotics are contraindicated because they will worsen pupillary block. The fellow eye should undergo prophylactic laser iridotomy.

Traumatic glaucoma

Hyphaema

Pathogenesis

A traumatic hyphaema may be associated with IOP elevation due to trabecular blockage by red blood cells. Pupillary occlusion by a blood clot may be superimposed on an angle-closure component. Secondary haemorrhage, often more severe than the primary bleed, may develop within 3–5 days of the initial injury. Patients with sickle-cell haemoglobinopathies are at increased risk of developing complications associated with traumatic hyphaema.

Risk of glaucoma

Although most traumatic hyphaemas are relatively innocuous and transient, severe and prolonged elevation of IOP may damage the optic nerve and cause blood staining of the cornea; the latter can progress very rapidly. The size of a hyphaema is a useful indicator of visual prognosis and risk of complications:

Hyphaema involving less than half the anterior chamber (Fig. 10.60) is associated with a 4% incidence of raised IOP, a 22% incidence of complications and a final visual acuity of >6/18 in 78% of eyes.
Page 368
Hyphaema involving over half the anterior chamber is associated with an 85% incidence of raised IOP, 78% incidence of complications and a final visual acuity of >6/18 in only 28% of eyes.
image

Fig. 10.60 Small hyphaema with a low risk of glaucoma

Treatment

1 General
A coagulation abnormality, particularly a haemoglobinopathy, should be excluded.
Any current anticoagulant medication should be discontinued after liaison with a general physician to assess the risk; NSAIDs should not be used for analgesia.
Hospital admission may be required for large hyphaemas.
Strict bed rest is probably unnecessary, but substantially limiting activity is prudent, and the patient should remain in a sitting or semi-upright posture, including during sleep.
2 Medical
A beta-blocker and/or a topical or systemic CAI is administered (not in sickle haemoglobinopathies if possible) depending on the IOP. Miotics should be avoided as they may increase pupillary block and disrupt the blood-aqueous barrier, and prostaglandins as they may promote inflammation. Alpha-agonists may be useful, but are avoided in small children and sickling disorders.
Occasionally a hyperosmotic agent is needed.
Topical steroids should be used since they reduce inflammation and possibly the risk of secondary haemorrhage.
Mydriatics are controversial. Atropine is recommended by some authorities to achieve constant mydriasis rather than a mobile pupil, in order to minimize the chances of secondary haemorrhage.
Systemic antifibrinolysis (aminocaproic acid or tranexamic acid) is rarely given now; topical aminocaproic acid shows promise but remains investigational at present.
3 Surgical evacuation of the blood is indicated if it is judged there is a risk of permanent corneal staining (rare) or persistently intolerable IOP. If a total hyphaema persists for more than 5 days consider evacuation to prevent the occult development of peripheral anterior synechiae and chronic secondary glaucoma; a low threshold is required in haemoglobinopathy patients (even moderate pressure elevation can lead to optic atrophy) and in young children with a risk of amblyopia. A glaucoma filtration procedure may be necessary in some cases.
4 On discharge the patient should be advised to avoid any activity with a risk of even minor eye trauma for several weeks; symptoms of a rebleed should prompt immediate review.

Angle recession glaucoma

Pathogenesis

Angle recession involves rupture of the face of the ciliary body, the portion that lies between the iris root and the scleral spur, due to blunt trauma. Although a large percentage of eyes with traumatic hyphaema exhibit some degree of angle recession, only 6–9% develop glaucoma after 10 years. The rise in IOP is secondary to associated trabecular damage rather than from angle recession itself; however, the risk of glaucoma is directly related to the extent of angle recession. Since glaucoma may not develop until months or even years after the initial injury, angle recession mandates periodic review.

Diagnosis

1 Presentation is with unilateral chronic glaucoma.
2 Slit-lamp biomicroscopy may show signs of previous blunt trauma, which may be mild, such as a small sphincter rupture.
3 Gonioscopy may initially show irregular widening of the ciliary body (Fig. 10.61A). In long-standing cases, the cleft may become obscured by fibrosis and the angle may show hyperpigmentation (Fig. 10.61B).
image

Fig. 10.61 (A) Angle recession; (B) old angle recession with hyperpigmentation

(Courtesy of R Curtis – fig. A)

Treatment

1 Medical treatment is as for other types of secondary open-angle glaucoma but is frequently unsatisfactory and laser trabeculoplasty is ineffective.
2 Trabeculectomy with adjunctive antimetabolites is generally an effective procedure.
3 An artificial filtering shunt or cyclodiode should be considered if trabeculectomy fails.

Iridocorneal endothelial syndrome

Classification

Page 369

The iridocorneal endothelial (ICE) syndrome typically affects one eye of a middle-aged woman. It consists of the following three very rare and frequently overlapping disorders: (a) progressive iris atrophy, (b) iris naevus (Cogan–Reese) syndrome and (c) Chandler syndrome.

Pathogenesis

The common link between the three variants of ICE syndrome is an abnormal corneal endothelial cell layer which has the capacity to proliferate and migrate across the angle and onto the surface of the iris. The term ‘proliferative endotheliopathy’ has therefore been suggested to describe this disorder. The ICE syndrome may progress to glaucoma, corneal decompensation or both. Glaucoma is due to synechial angle closure secondary to contraction of this abnormal tissue. Polymerase chain reaction shows the presence of herpes simplex virus DNA in a substantial percentage of ICE syndrome corneal specimens, suggesting that the condition may be of viral origin.

General features

1 Slit-lamp biomicroscopy
Corectopia (malposition of the pupil – Fig. 10.62A).
Pseudopolycoria (supernumerary false pupils) in a previously normal iris (Fig. 10.62B).
Iris atrophy of varying severity (Fig. 10.62C and D).
2 Gonioscopy shows broad-based PAS that often extend anterior to Schwalbe line (Fig. 10.62E).
3 Glaucoma is present in about 50% of cases.
image

Fig. 10.62 Iridocorneal endothelial syndrome. (A) Corectopia; (B) pseudopolycoria; (C) iris atrophy; (D) very severe iris atrophy; (E) broad peripheral anterior synechiae; (F) iris nodule in Cogan–Reese syndrome

(Courtesy of L MacKeen – fig. E; R Martincova – fig. F)

Specific features

In their purest form, the three conditions are easily distinguished from each other. However, there is frequently considerable overlap and clear differentiation may be difficult. Occasionally, during follow-up one condition can be observed changing into another. The differentiation depends primarily on the iris changes.

1 Progressive iris atrophy is characterized by severe iris changes.
2 The iris naevus (Cogan–Reese) syndrome is characterized by either a diffuse naevus which covers the anterior iris or iris nodules (Fig. 10.62F). Iris atrophy is absent in 50% of cases and in the remainder it is usually mild to moderate although corectopia may be severe. It is important not to misdiagnose a diffuse iris melanoma as the iris naevus syndrome
3 Chandler syndrome is characterized by ‘hammered-silver’ corneal endothelial abnormalities (Fig. 10.63A) and frequently presents with blurred vision and haloes due to corneal oedema (Fig. 10.63B). Stromal atrophy is absent in about 60% of cases and in the remainder is variable in its severity; corectopia is mild to moderate. Glaucoma is usually less severe than in the other two syndromes, and at presentation the IOP may be normal.
image

Fig. 10.63 Chandler syndrome. (A) Hammered-silver endothelial changes; (B) corneal oedema due to endothelial decompensation

(Courtesy of J McAllister – fig. B)

Treatment of glaucoma

1 Medical treatment may be tried but is often ineffective.
2 Trabeculectomy, even when combined with adjunctive antimetabolites, is frequently unsuccessful because of late-onset bleb failure.
3 Artificial filtering shunt or cyclodiode is eventually required in many cases.

Glaucoma in intraocular tumours

Approximately 5% of eyes with intraocular tumours develop a secondary elevation of IOP. Depending on the location of the tumour one or more of the following mechanisms may be responsible:

Trabecular block

Trabecular block may be the result of one of the following:

Page 371
1 Angle invasion by a solid iris melanoma (Fig. 10.64A).
2 Trabecular infiltration by neoplastic cells originating from an iris melanoma (Fig. 10.64B). Rarely, tumour seeding from a retinoblastoma may also invade the trabeculum.
3 Melanomalytic glaucoma may occur in some eyes with iris melanoma; it is due to trabecular blockage by macrophages which have ingested pigment and tumour cells (Fig. 10.64C), similar to phacolytic glaucoma.
image

Fig. 10.64 Glaucoma in intraocular tumours. (A) Angle invasion by a solid iris melanoma; (B) melanoma cells infiltrating the trabeculum; (C) melanomalytic glaucoma; (D) angle-closure by a large ciliary body melanoma

(Courtesy of R Curtis – figs A and C; J Harry – figs B and D)

Secondary angle closure

Secondary angle closure may be the result of one of the following:

1 Neovascular glaucoma is the most common mechanism in eyes with choroidal melanoma or retinoblastoma.
2 Anterior displacement of iris-lens diaphragm may occur in an eye with a ciliary body melanoma (Fig. 10.64D) or a large tumour of the posterior segment.

Glaucoma in epithelial ingrowth

Pathogenesis

Epithelial ingrowth is a rare and potentially blinding complication of anterior segment surgery or trauma. Conjunctival or corneal epithelial cells migrate through the wound and proliferate in the anterior segment, in a cystic or diffuse manner. The latter is characterized by the proliferation of sheets of epithelial cells over the posterior cornea, trabeculum, iris and ciliary body (Fig. 10.65A) and is more commonly associated with secondary glaucoma than the cystic variety. Elevation of IOP is caused by a combination of often pre-existing PAS, obstruction of the trabeculum by the epithelial membrane, and desquamated epithelial and inflammatory cells.

image

Fig. 10.65 Diffuse epithelial ingrowth. (A) Stratified squamous epithelium lining the anterior iris surface and filtration angle; (B) translucent membrane with a scalloped border involving the posterior corneal surface

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

Diagnosis

Persistent postoperative anterior uveitis.
Diffuse epithelialization is characterized by a translucent membrane with scalloped border involving the posterior corneal surface (Fig. 10.65B) in the sector of the incision.
Pupillary distortion.

Treatment

The aim of treatment is to eradicate the invading epithelium to avoid recurrence or the conversion of epithelial cysts into diffuse epithelialization with consequent intractable glaucoma.

1 Block excision involves simultaneous excision of adjacent iris, pars plicata of the ciliary body, together with all layers of the sclera and cornea in contact with the lesion. The resultant defect is covered with a tectonic corneoscleral graft. The area of iris involvement may be delineated by applying argon laser burns which will cause whitening of the affected area.
2 Cryotherapy may be applied trans-sclerally to devitalize the epithelium remaining on the posterior surface of the cornea, in the angle and on the ciliary body. Intraocular air is used to insulate other tissues from the effects of the cryotherapy.
3 Artificial filtering shunts are of value for medically uncontrolled glaucoma associated with extensive epithelial ingrowth unsuitable for surgical excision.

Glaucoma in iridoschisis

Iridoschisis is a rare condition which typically affects the elderly and is often bilateral. It is associated with underlying angle-closure glaucoma in at least 90% of cases. It is thought that acute or intermittent angle-closure results in iris atrophy as a result of high IOP.

Page 372
Page 373
1 Slit lamp biomicroscopy
Shallow anterior chamber (Fig. 10.66A).
Iridoschisis usually involves the inferior iris (Fig. 10.66B).
The severity ranges from intrastromal atrophy to extensive splitting of the anterior leaf (Fig. 10.66C) and disintegrated iris fibrils.
2 Gonioscopy shows a narrow occludable angle which may be associated with PAS.
3 Treatment initially involves peripheral laser iridotomy. Subsequent treatment is aimed at limiting glaucomatous damage.
image

Fig. 10.66 Iridoschisis. (A) Shallow anterior chamber; (B) mild; (C) very severe

Primary congenital glaucoma

Introduction

Genetics

Most cases of primary congenital glaucoma (PCG) are sporadic. In approximately 10% inheritance is AR with incomplete penetrance. To date, PCG has been linked to three loci: 2p21 (GLC3A), for which the responsible gene is CYP1B1, 1p36 (GLC3B) and 14q24 (GLC3C), for which the genes are not yet identified.

Page 374

Pathogenesis

Impaired aqueous outflow in PCG is caused by maldevelopment of the angle of the anterior chamber, unassociated with any other major ocular anomalies (isolated trabeculodysgenesis). Clinically, trabeculodysgenesis is characterized by absence of the ciliary body band due to translucent amorphous material that obscures the trabeculum (Fig. 10.67B).

image

Fig. 10.67 (A) Normal infant angle shows the iris root, prominent ciliary body band but no discernible scleral spur and trabeculum; (B) angle in congenital glaucoma shows the iris root but not the ciliary body band due to translucent amorphous tissue that obscures the trabeculum

(Courtesy of K Nischal)

Classification

1 True congenital glaucoma (40%) in which IOP becomes elevated during intrauterine life.
2 Infantile glaucoma (55%) which manifests prior to the third birthday.
3 Juvenile glaucoma, the least common, in which the pressure rise develops after the third birthday but before the age of 16 years. Gonioscopy may be normal or reveal trabeculodysgenesis. Patients with normal findings are classed as having juvenile open-angle glaucoma that behaves like adult primary open-angle glaucoma.

Diagnosis

Although PCG is the most common of the congenital glaucomas, it is still a very rare condition, affecting 1 : 10 000 births; 65% of patients are boys. The clinical features depend on the age of onset and the level of IOP. Both eyes are affected in 75% of cases although involvement is frequently asymmetrical.

1 Corneal haze is often the first sign noticed by the parents (Fig. 10.68A). It is caused by epithelial and stromal oedema secondary to raised IOP and may be associated with lacrimation, photophobia and blepharospasm (Fig. 10.68B).
2 Buphthalmos is a large eye as a result of stretching due to elevated IOP prior to the age of 3 years (Fig. 10.68C). It is not usually reported by the parents unless advanced (Fig. 10.68D). As the sclera stretches it becomes thinner and translucent; the eye then takes on a blue appearance due to enhanced visualization of the underlying uvea. As ocular enlargement continues the anterior chamber deepens and in advanced cases the zonular fibres stretch and the lens may rarely subluxate. The increased axial length also causes axial myopia, which may give rise to anisometropic amblyopia.
3 Breaks in Descemet membrane secondary to corneal stretching may be associated with a sudden influx of aqueous into the corneal stroma. Haab striae represent healed breaks in Descemet membrane and appear as horizontal curvilinear lines (Fig. 10.68E). Chronic stromal oedema may lead to permanent scarring and vascularization (Fig. 10.68F).
4 Optic disc cupping in infants may regress once the IOP is normalized. Most normal infants exhibit no apparent cup; very few have a cup–disc ratio greater than 0.3, unlike a high percentage of infants with PCG. In contrast to the adult eye, the scleral canal in the infant eye enlarges as part of the generalized enlargement of the globe and the lamina cribrosa may bow posteriorly in response to elevated IOP. Cup size may therefore be increased from neuronal loss, enlargement of the scleral canal, or both.
image

Fig. 10.68 Congenital glaucoma. (A) Corneal haze; (B) photophobia and blepharospasm; (C) buphthalmos; (D) severe buphthalmos and scleral thinning; (E) Haab striae; (F) corneal scarring and vascularization

(Courtesy of M Parulekar – fig. A; U Raina – figs B, C and F)

Management

Initial evaluation

The initial evaluation should be performed under general anaesthesia with intravenous ketamine, since this lowers IOP less than other agents. Examination of the optic discs should be undertaken first, followed by measurement of IOP and corneal diameters, and finally gonioscopy.

1 IOP is measured with the Perkins tonometer (Fig. 10.69) or Tono-Pen® (see Fig. 10.7C).
2 Corneal diameter is measured in both the vertical and horizontal meridian with callipers. A diameter >11 mm prior to the age of one year or >13 mm at any age should be viewed with suspicion. Diameters of 14 mm are typical of advanced buphthalmos.
3 Gonioscopy is performed with a direct goniolens.
image

Fig. 10.69 Measurement of intraocular pressure with the Perkins tonometer

Surgery

Page 375
Page 376
1 Goniotomy is performed at the initial examination if the diagnosis is confirmed, provided there is sufficient corneal clarity and the angle can be visualized. The procedure involves making a horizontal incision at the midpoint of the superficial layers of the trabecular meshwork (Fig. 10.70). Although goniotomy may need to be repeated, the eventual success rate is about 85%. However, the results are poor if the corneal diameter is 14 mm or more because in such eyes the canal of Schlemm is obliterated.
2 Trabeculotomy may be necessary if corneal clouding prevents visualization of the angle or when repeated goniotomy has failed. In this procedure a partial thickness scleral flap is fashioned (Fig. 10.71A and B), the Schlemm canal is found (10.71C), a trabeculotome is inserted and then rotated into the anterior chamber (Fig. 10.71D). The technique is highly demanding and requires previous experience and good anatomical landmarks to achieve predictable results. In addition, the Schlemm canal may be difficult to canalize because of hypoplasia or angle anomaly.
3 Trabeculectomy is often successful, particularly when combined with adjunctive antimetabolites.
4 Combined trabeculectomy and trabeculotomy has been used but its superiority to trabeculectomy alone is debatable.
image

Fig. 10.70 Goniotomy – arrow shows the cleft

(Courtesy of K Nischal)

image

Fig. 10.71 Trabeculotomy

(Courtesy of K Nischal)

Follow-up

The patients should be reviewed 1 month after initial surgery. The IOP and corneal diameters should be monitored at regular intervals because progressive enlargement of the corneal diameter is as important a sign of uncontrolled congenital glaucoma analagous to progressive visual field loss in adult glaucoma. Cycloplegic refraction should be performed at 6-monthly intervals. About 50% of patients suffer visual loss from optic nerve damage, anisometropic amblyopia, corneal scarring, cataract and lens subluxation. A buphthalmic eye is also susceptible to traumatic damage.

Differential diagnosis

1 Cloudy cornea at birth
Birth trauma, which gives rise to corneal oedema due to breaks in Descemet membrane.
Intrauterine rubella, which results in a cloudy cornea due to keratitis. Ten percent of infants with the rubella syndrome also have congenital glaucoma due to an angle anomaly similar to that found in PCG. This may be missed because the eye may not appear significantly enlarged, due to pre-existing microphthalmos.
Metabolic disorders such as mucopolysaccharidoses and mucolipidoses.
Congenital hereditary endothelial dystrophy.
2 Large cornea due to megalocornea or very high myopia.
3 Lacrimation resulting from delayed canalization of the nasolacrimal duct.
4 Secondary infantile glaucoma
Tumours such as retinoblastoma and juvenile xanthogranuloma.
Persistent hyperplastic primary vitreous.
Retinopathy of prematurity.
Intraocular inflammation.
Trauma.
Ectopia lentis.

Iridocorneal dysgenesis

Posterior embryotoxon

Posterior embryotoxon is an isolated innocuous finding in 10% of the general population.

1 Signs
Thin grey-white, arcuate ridge on the inner surface of the cornea, adjacent to the limbus (Fig. 10.72A and B).
It comprises a prominent and anteriorly displaced Schwalbe line.
2 Associations
Page 377
a Axenfeld–Rieger anomaly is always associated with posterior embryotoxon.
b Alagille syndrome is associated with posterior embryotoxon in 95% of cases. It is characterized by paucity of intrahepatic bile ducts, cardiopulmonary malformations, peculiar facies and vertebral defects. Optic disc drusen are also common.
image

Fig. 10.72 Axenfeld anomaly. (A) Posterior embryotoxon; (B) magnified view; (C) gonioscopy shows of strands of peripheral iris tissue extending to the cornea

(Courtesy of P Gili – fig. A; L MacKeen – fig. B)

Axenfeld–Rieger syndrome

Pathogenesis and genetics

Axenfeld–Rieger syndrome is a spectrum of disorders designated in current nomenclature by the following eponyms: (a) Axenfeld anomaly, (b) Rieger anomaly and (c) Rieger syndrome. Gene loci have been mapped to 4q25 (PITX2 gene), 6p25 (FKHL7) and 13q14 (RIEG2). All patients with Axenfeld–Rieger syndrome, irrespective of ocular manifestations, share the following features:

Bilateral developmental ocular anomalies which are not necessarily symmetrical.
Frequent family history with AD inheritance.
No gender predilection.
Frequent presence of systemic developmental defects.
Associated glaucoma.

Axenfeld anomaly

This is characterized by posterior embryotoxon with attachment of strands of peripheral iris tissue (Fig. 10.72C).

Rieger anomaly

1 Slit-lamp biomicroscopy
Posterior embryotoxon.
Iris stromal hypoplasia (Fig. 10.73A and B).
Ectropion uveae (Fig. 10.73C).
Corectopia and full-thickness iris defects (Fig. 10.73D).
2 Gonioscopy in mild cases shows Axenfeld anomaly. In severe cases, broad leaves of the iris stroma adhere to the cornea anterior to Schwalbe line (Fig. 10.73E).
3 Glaucoma develops in about 50% of cases, usually during early childhood or early adulthood due to an associated angle anomaly or secondary synechial angle closure. The elevation of IOP should initially be managed medically, although surgery may be required subsequently.
Page 378
image

Fig. 10.73 Rieger anomaly and syndrome. (A) Mild iris stromal hypoplasia; (B) severe iris stromal hypoplasia; (C) ectropion uveae; (D) corectopia and full-thickness iris defects; (E) peripheral anterior synechiae; (F) facial and dental anomalies in Rieger syndrome

(Courtesy of U Raina – fig. F)

Rieger syndrome

Rieger syndrome is linked to the region of the epidermal growth factor gene on chromosome 4. It is characterized by Rieger anomaly in association with the following extraocular malformations:

1 Dental anomalies consisting of hypodontia (few teeth) and microdontia (small teeth – Fig. 10.73F).
2 Facial anomalies include maxillary hypoplasia, broad nasal bridge, telecanthus and hypertelorism (see Fig. 10.73F).
3 Other anomalies include redundant paraumbilical skin and hypospadias. Hearing loss, hydrocephalus, cardiac and renal anomalies and congenital hip dislocation are rare.

Peters anomaly

Peters anomaly is an extremely rare but serious condition which is bilateral in 80% of cases. It is the result of defective neural crest cell migration in the 6th to 8th weeks of fetal development, during which time the anterior segment of the eye is formed. It is not a homogeneous condition and may vary from mild to severe.

1 Inheritance. Most cases are sporadic, although AR inheritance and chromosomal defects have been described.
2 Signs
Central corneal opacity of variable density (Fig. 10.74A).
Underlying defect involving the posterior stroma, Descemet membrane and endothelium with or without iridocorneal (Fig. 10.74B) or lenticulocorneal (Fig. 10.74C) adhesions.
3 Investigations. In severe cases ultrasound biomicroscopy is used to determine associated pathology prior to contemplating penetrating keratoplasty.
4 Ocular associations occasionally present include Axenfeld–Rieger anomaly, aniridia, microphthalmos, persistent fetal vasculature and retinal dysplasia.
5 Glaucoma occurs in about 50% of cases as a result of an associated angle anomaly in which there is incomplete development of the trabecular meshwork and Schlemm canal. Elevation of IOP is usually evident in infancy but may occasionally develop in childhood or even later. Treatment of glaucoma is very difficult and the prognosis tends to be worse than that of primary congenital glaucoma.
6 Systemic associations include craniofacial anomalies, central nervous system anomalies, fetal alcohol syndrome, chromosome abnormalities, and ‘Peters plus’ syndrome (short-limbed dwarfism, cleft lip/palate and learning difficulties).
image

Fig. 10.74 Peters anomaly. (A) Corneal opacity; (B) iridocorneal adhesion; (C) lenticulocorneal adhesion

Aniridia

Genetics

Aniridia (AN) is a rare bilateral condition that may have life-threatening associations. It occurs as a result of abnormal neuroectodermal development secondary to a mutation in the PAX6 gene linked to 11p13. PAX6 is adjacent to gene WT1, mutation of which predisposes to Wilms tumour.

Classification

Page 379
Page 380
1 AD form accounts for about two-thirds of cases and has no systemic implications. Penetrance is complete (i.e. all patients with the genotype will have the phenotype) but expressivity (severity) is variable.
2 Sporadic, including WARG syndrome (Wilm tumour, Aniridia, Genitourinary abnormalities, mental Retardation), previously known as Miller syndrome, accounts for about a third of patients. Children with sporadic aniridia have about a 30% chance of developing Wilms tumour.
3 Gillespie syndrome accounts for only about 1% of cases. Inheritance is AR but not caused by PAX6 mutations. Cerebellar ataxia and mental handicap are features.

All patients with sporadic aniridia should have abdominal ultrasonography (to detect Wilms tumour) every 3 months until 5 years of age, every 6 months until 10 years of age and annually until 16 years of age or until molecular genetic analysis confirms an intragenic mutation without extragenic involvement.

Diagnosis

1 Presentation is typically at birth with nystagmus and photophobia. The parents may have noticed absence of irides or ‘dilated pupils’.
2 Aniridia is variable in severity, ranging from minimal, detectable only by retroillumination, to partial (Fig. 10.75A) and total (Fig. 10.75B).
3 Gonioscopy even in eyes with ‘total’ aniridia usually shows a hypoplastic or rudimentary frill of iris tissue (Fig. 10.75C).
4 Lids often show meibomian gland dysfunction.
5 Cornea
Tear film instability, dry eye and epithelial defects are common.
Limbal stem cell deficiency may result in ‘conjunctivalization’ of the peripheral cornea.
Total corneal central stromal scarring and vascularization may occur in end-stage disease.
Other lesions include opacity, epibulbar dermoids, microcornea, sclerocornea and kerato-lenticular adhesions.
6 Lens changes include cataract, subluxation (usually superiorly – Fig. 10.75D), congenital aphakia and persistent pupillary membranes.
7 Fundus may exhibit foveal hypoplasia (Fig. 10.75E), optic nerve hypoplasia and choroidal coloboma.
image

Fig. 10.75 Aniridia. (A) Partial; (B) total; (C) open angle and remnants of the iris root; (D) superior subluxation of a cataract; (E) foveal hypoplasia; (F) angle closed synechially by iris rudiments

(Courtesy of R Curtis – fig. C; L MacKeen – fig. D)

Glaucoma

Glaucoma occurs in approximately 75% of patients and usually presents in late childhood or adolescence. It is caused by synechial angle-closure secondary to the pulling forward of rudimentary iris tissue by contraction of pre-existing fibres that bridge the angle (Fig. 10.75F). Treatment is difficult and the prognosis guarded.

1 Medical treatment is usually the initial approach although it is usually eventually inadequate.
2 Goniotomy may prevent subsequent rise in IOP if performed before the development of irreversible synechial angle-closure.
Page 381
Page 382
3 Combined trabeculectomy-trabeculotomy may be successful although trabeculectomy alone is seldom beneficial.
4 Artificial filtering shunts may be effective in established cases.
5 Diode laser cycloablation may be necessary if other modalities fail.

Management of aniridia

1 Opaque contact lenses may be used to create an artificial pupil and improve vision and cosmesis.
2 Lubricants are frequently required for associated keratopathy.
3 Cataract surgery is often required. Care must be taken to minimize trauma to the limbus and preserve stem cell function.
4 Limbal stem cell transplantation with or without keratoplasty may be required.

Glaucoma in phacomatoses

Sturge–Weber syndrome

Sturge–Weber syndrome (encephalotrigeminal angiomatosis) is a congenital, sporadic phacomatosis (see Ch. 1).

Pathogenesis of glaucoma

Glaucoma develops in about 30% of patients ipsilateral to the facial haemangioma, especially if the lesion affects the upper eyelid. Elevation of IOP occurs within the first 2 years of life in 60% of glaucoma patients and may result in buphthalmos (Fig. 10.76A). In the remainder, glaucoma may develop at any time from infancy to adulthood. The pathogenesis is controversial and often obscure.

Isolated trabeculodysgenesis may be instrumental in infants.
Raised episcleral venous pressure (associated with an arteriovenous communication in an episcleral haemangioma (Fig. 10.76B) may be responsible in older patients.
image

Fig. 10.76 Glaucoma in Sturge–Weber syndrome. (A) Bilateral naevus flammeus and bilateral buphthalmos; (B) episcleral haemangioma

(Courtesy of R Bates – fig. A)

Treatment

1 Medical treatment with topical prostaglandin analogues may be successful.
2 Goniotomy may be successful in eyes with angle anomalies.
3 Combined trabeculotomy-trabeculectomy gives good results in early-onset cases. The rationale is that trabeculotomy addresses the barrier to aqueous outflow posed by a congenital angle anomaly, while trabeculectomy bypasses the episcleral veins. Surgery carries a high risk of choroidal effusion and suprachoroidal haemorrhage.

Neurofibromatosis type 1

Neurofibromatosis is a disorder that primarily affects cell growth of neural tissues. Inheritance is AD with irregular penetrance and variable expressivity (see Ch. 19). Glaucoma is relatively rare and, when present, is usually unilateral and congenital. About 50% of patients with glaucoma have an ipsilateral plexiform neurofibroma of the upper eyelid or exhibit facial hemiatrophy (Fig. 10.77A). One or more of the following may be the causative mechanism:

Obstruction of aqueous outflow by neurofibromatous tissue in the angle.
Developmental angle anomaly which may be associated with congenital ectropion uveae (Fig. 10.77B).
Secondary angle-closure caused by forward displacement of the peripheral iris associated with neurofibromatous thickening of the ciliary body.
Secondary synechial angle-closure caused by contraction of a fibrovascular membrane.
image

Fig. 10.77 Glaucoma in NF1. (A) Extensive neurofibromatosis and left facial hemiatrophy; (B) congenital ectropion uveae

(Courtesy of R Bates – fig. B)

Page 383

Glaucoma medications

Most glaucoma medications are administered topically. As a general rule, treatment is indicated whenever glaucomatous damage is deemed likely to occur. The decision on which medication to prescribe depends not only on the type of glaucoma, but also on the patient’s medical history (e.g. presence of asthma or bradycardia). This requires a detailed knowledge of the potential side-effects. To improve compliance, patients should be fully informed not only about their disease but also the medications used, how to administer the drug, and what side-effects to expect. The efficacy of therapy should be regularly evaluated and the regimen altered to improve efficacy, if appropriate, or to reduce adverse effects.

Beta-blockers

Pharmacology

Adrenergic neurones secrete noradrenaline at sympathetic postganglionic nerve endings. Adrenergic receptors are of the following four main types:

1 Alpha-1 receptors are located in the arterioles, dilator pupillae and Müller muscle. Stimulation gives rise to hypertension, mydriasis and lid retraction.
2 Alpha-2 inhibitory receptors located in the ciliary epithelium. Stimulation results in increase in the facility of aqueous outflow.
3 Beta-1 receptors are located in the myocardium and give rise to tachycardia and increased cardiac output when stimulated.
4 Beta-2 receptors are located in the bronchi and ciliary epithelium. Stimulation causes bronchodilatation and increased aqueous production.

Beta-blockers are drugs that antagonize the effects of catecholamines at beta receptors. Non-selective beta-blockers are equipotent at beta-1 and beta-2 receptors, while cardioselective are more potent at beta-1 receptors. The advantage of the latter, at least in theory, is that the bronchoconstrictive effect of beta-2 blockade is minimized. Betaxolol is the only cardioselective agent currently available for the treatment of glaucoma.

Mode of action

Beta-blockers reduce IOP by decreasing aqueous secretion and are therefore useful in all types of glaucoma, irrespective of the state of the angle. The exact pharmacological basis for this is unclear. However, in approximately 10% of cases the pressure response decreases with time: tachyphylaxis. This may occur within a few days of starting treatment (‘short-term escape’) or within a few months (‘long-term drift’). As a general rule, little additional effect is obtained if a topical beta-blocker is used in a patient who is already on a systemic beta-blocker. During sleep, aqueous flow is normally less than half the daytime flow and beta-blockers have limited effect. When a beta-blocker is used in combination with brimonidine or a topical carbonic anhydrase inhibitor, an additional 15% reduction in IOP may be achieved. When combined with a prostaglandin analogue, the reduction is even greater (20%).

Side-effects

1 Ocular side-effects include occasional allergy, corneal punctate epithelial erosions and reduced aqueous tear secretion.
2 Systemic side-effects tend to occur during the first week of administration. Although uncommon they may be serious.
Bradycardia and hypotension can result from beta-1 blockade. The patient’s pulse must be palpated before prescribing a beta-blocker.
Bronchospasm may be induced by beta-2 blockade and may be fatal in pre-existing asthma or severe chronic pulmonary obstruction.
Miscellaneous side-effects include sleep disorders, hallucinations, confusion, depression, fatigue, headache, nausea, dizziness, decreased libido and possible reduction of plasma high-density lipoprotein level.
3 Reduction of systemic absorption may be achieved by:
Lacrimal occlusion following instillation, by closing the eyes and applying digital pressure over the lacrimal sac area for about 3 minutes. Apart from obstructing lacrimal drainage and reducing systemic absorption this also prolongs eye–drug contact and increases therapeutic efficacy.
Merely closing the eyes for 3 minutes will reduce systemic absorption by about 50%.
4 Contraindications to beta-blockers include asthma and obstructive airways disease, bradycardia, congestive cardiac failure, and second or third degree heart block. Beta-blockers should not be instilled at bedtime because they may cause a profound drop in blood pressure while the individual is asleep, thus reducing optic disc perfusion and potentially causing visual field deterioration; as noted above their Iop effect is also lower.

Preparations

1 Timolol is available in three forms:
0.25% and 0.5% used b.d.
Timoptol LA 0.25% and 0.5% used once daily.
Nyogel (timolol 0.1% gel) used once daily.
2 Betaxolol (Betoptic) 0.5% b.d. has less hypotensive effect than timolol, but the effect on preservation of the visual field may be superior. Betaxolol may increase optic disc blood flow, probably because of a calcium-channel blocking effect on the microcirculation of the disc.
3 Levobunolol (Betagan) 0.5% daily or b.d. is similar to timolol.
4 Carteolol (Teoptic) 1%, 2% b.d. is similar to timolol and also exhibits intrinsic sympathomimetic activity. It has a more selective action on the eye than on the cardiopulmonary system and may therefore induce less bradycardia than timolol.
5 Metipranolol 0.1%, 0.3% b.d. is similar to timolol but it may occasionally cause a granulomatous anterior uveitis. It is available only in preservative-free units.

Alpha-2 agonists

Alpha-2 agonists decrease IOP by both decreasing aqueous secretion and enhancing uveoscleral outflow. Because the drugs cross the blood-brain barrier they should not be used in children.

1 Brimonidine (Alphagan) 0.2% b.d. is a highly selective alpha-2 agonist which also has a neuroprotective effect. Its efficacy in isolation is less than timolol but generally better than betaxolol. It exhibits an additive effect with beta-blockers. The major ocular side-effect is allergic conjunctivitis that may be delayed for up to 18 months after commencement of therapy (Fig. 10.78A). Acute granulomatous anterior uveitis has been reported. Systemic side-effects include xerostomia, drowsiness and fatigue.
2 Apraclonidine (Iopidine) 1% is mainly used after laser surgery on the anterior segment to offset an acute rise in IOP. The 0.5% concentration may be used short-term, typically whilst a patient is awaiting glaucoma surgery. It is not suitable for long term use because of tachyphylaxis (loss of therapeutic effect over time) and a high incidence of local side-effects.
image

Fig. 10.78 Side-effects of topical medication. (A) Allergic conjunctivitis due to brimonidine; (B) lengthening and hyperpigmentation of lashes due to prostaglandin analogues; (C) darkening of irides due to prostaglandin analogues; (D) blepharoconjunctivitis due to topical carbonic anhydrase inhibitors

(Courtesy of J Salmon – fig. A; P Watson – fig. C)

Prostaglandin analogues

This group of agents have a sustained IOP-lowering effect which probably extends for several days in most patients.

Pharmacology

Prostanoid receptors are located on many ocular tissues, with involvement in functions such as regulation of intraocular pressure and blood flow.

1 Latanoprost and travoprost are F2-alpha analogues that act as selective agonists of the FP prostanoid receptor; both of these agents enhance aqueous humour outflow through the uveoscleral route.
2 Bimatoprost is a synthetic prostamide analogue structurally similar to prostaglandins that selectively mimics naturally occurring prostamide. It lowers IOP by promoting outflow through both uveoscleral and trabecular routes.
3 Tafluprost is a synthetic analogue of the prostaglandin F2α, also acting through the FP receptor.

Preparations

1 Latanoprost (Xalatan®) 0.005% used once daily at bedtime is superior to timolol although a proportion of patients show no response. Latanoprost produces an additive reduction of IOP of 14–28% when combined with timolol but not with pilocarpine.
2 Travoprost (Travatan®) 0.004% once daily is similar to latanoprost except in black patients in whom it may be more effective. Conjunctival hyperaemia occurs in up to 50% of patients but tends to subside with time.
3 Bimatoprost (Lumigan®) 0.03% once daily at bedtime is similar to latanoprost but may cause more conjunctival hyperaemia but fewer headaches and perhaps also less iris hyperpigmentation. A newer 0.01% preparation seems to leave a comparable IOP-lowering effect but with less hyperaemia.
4 Tafluprost (Saflutan™, Taflotan®) 0.0015% once daily at bedtime is a newer prostaglandin derivative, and the first available in preservative-free form.

Side-effects

1 Ocular
Conjunctival hyperaemia and a foreign body sensation are common.
Eyelash lengthening, thickening, hyperpigmentation and occasionally increase in number (Fig. 10.78B).
Iris hyperpigmentation, which is irreversible, occurs in 11–23% of patients after 6 months (Fig. 10.78C). The highest incidence is in green-brown irides, less in yellow-brown irides and least in blue-grey/brown irides. Hyperpigmentation is caused by an increase in the number of pigmented granules within the superficial stroma rather than an increase in the number of melanocytes. Iris naevi and freckles are, however, not affected.
Hyperpigmentation of periorbital skin is common but reversable.
It is possible that these drugs increase the frequency of cystoid macular oedema after cataract surgery.
Anterior uveitis is very rare and usually responsive to steroid therapy. The drug should therefore be used with caution in uveitic glaucoma.
Increase in severity and recurrence of herpetic keratitis is also rare.
Conjunctival hyperpigmentation has also been reported.
2 Systemic side-effects include occasional headache, precipitation of migraine in susceptible individuals, skin rash and mild upper respiratory tract symptoms. These preparations should be avoided in pregnancy as animal studies have shown potential teratogenic effects.

Topical carbonic anhydrase inhibitors

The carbonic anhydrase inhibitors (CAIs) are chemically related to the sulphonamides. They lower IOP by inhibiting aqueous secretion.

1 Dorzolamide (Trusopt) 2% is used t.i.d. as monotherapy or b.d. as adjunctive treatment, and is similar in efficacy to betaxolol but inferior to timolol. The main side-effects are allergic blepharoconjunctivitis (Fig. 10.78D) and a transient bitter taste. The drug should be used with caution in patients with corneal endothelial dysfunction as it may precipitate decompensation.
2 Brinzolamide (Azopt) 1% b.d. or t.i.d. is similar to dorzolamide, but with a lower incidence of stinging and local allergy.

Miotics

Pharmacology

Miotics are parasympathomimetic drugs that act by stimulating muscarinic receptors in the sphincter pupillae and ciliary body.

Page 386
1 In POAG miotics reduce IOP by contraction of the ciliary muscle, which increases the facility of aqueous outflow through the trabecular meshwork.
2 In PACG contraction of the sphincter pupillae and the resultant miosis pulls the peripheral iris away from the trabeculum, thus opening the angle. It is often necessary to reduce IOP with systemic medication before miotics can take effect.

Ocular side-effects include miosis, brow ache, myopic shift and exacerbation of symptoms of cataract. Visual field defects appear denser and larger.

Preparations

1 Pilocarpine is equal in efficacy to beta-blockers and is available in the following forms:
Pilocarpine drops 0.5%, 1%, 2%, or 4% is used q.i.d. as monotherapy. When used in combination with a beta-blocker, b.d. administration is adequate.
Pilocarpine gel (Pilogel®) 4% consists of pilocarpine adsorbed on to a plastic gel, instilled once daily at bedtime so that the induced myopia and miosis last only during sleep. The main disadvantage is the development of a diffuse superficial corneal haze in 20% of users, although this rarely affects visual acuity.
2 Carbachol 3% t.i.d. is an effective alternative to pilocarpine in resistant or intolerant cases.

Combined preparations

Combined preparations with similar ocular hypotensive effects to the sum of the individual components improve convenience and patient compliance. They are also more cost effective. Examples include:

1 Cosopt (timolol + dorzolamide) b.d.
2 Xalacom (timolol + latanoprost) once daily.
3 TimPilo (timolol + pilocarpine) b.d.
4 Combigan (timolol + brimonidine) b.d.
5 DuoTrav (timolol + travoprost) once daily.
6 Ganfort) (timolol + bimatoprost) once daily.
7 Azarga (timolol + brinzolamide) b.d.

Systemic carbonic acid inhibitors

Preparations

1 Acetazolamide is available in the following forms:
Tablets 250 mg. The dose is 250–1000 mg daily in divided doses with onset of action within 1 hour, a peak at 4 hours and duration up to 12 hours.
Sustained-release capsules 250 mg used 250–500 mg daily with duration of up to 24 hours.
Powder 500 mg vials for injection. The onset of action is almost immediate, with a peak at 30 minutes and duration up to 4 hours. This is the only CAI preparation available for injection and is useful in acute angle-closure glaucoma.
2 Dichlorphenamide tablets 50 mg. The dose is 50–100 mg b.d. or t.i.d. with onset of action within 1 hour, peak at 3 hours and duration up to 12 hours.
3 Methazolamide tablets 50 mg. The dose is 50–100 mg b.d. or t.i.d. with onset of action within 3 hours, peak at 6 hours and duration of 10–18 hours. This is a useful alternative to acetazolamide with a longer duration of action but is currently not available in the UK.

Systemic side-effects

Systemically administered CAIs may be useful as short-term treatment, particularly in patients with acute glaucoma. Because of their systemic side-effects long-term use is reserved for patients at high risk of visual loss. The patient should always be warned of potential side-effects as this decreases anxiety and improves compliance.

1 Paraesthesia characterized by tingling of the fingers, toes, hands or feet, and occasionally at the mucocutaneous junctions, is a universal finding and usually innocuous. Compliance is suspect if the patient denies this symptom.
2 Malaise complex is characterized by a combination of malaise, fatigue, depression, weight loss and decreased libido. A supplemental 2-week course of sodium acetate may be helpful.
3 Gastrointestinal complex is characterized by gastric irritation, abdominal cramps, diarrhoea and nausea. This can occur independently of the malaise syndrome and is unrelated to any specific changes in blood chemistry.
4 Renal stone formation is uncommon.
5 Stevens–Johnson syndrome may rarely occur since CAIs are sulphonamide derivatives.
6 Blood dyscrasias are extremely rare and may be of two types:
Dose-related bone marrow suppression which usually recovers when treatment is stopped.
Idiosyncratic aplastic anaemia is not dose-related and has a mortality of 50%.
7 Hypokalaemia may occur with long-term treatment and blood potassium levels should be monitored.

Osmotic agents

Physiological principles

Osmotic pressure is dependent on the number rather than on the size of solute particles in a solution. Lower molecular weight solutes therefore exert a greater osmotic effect per gram. Osmotic agents remain intravascular, thus increasing blood osmolality. They lower IOP by creating an osmotic gradient between blood and vitreous so that water is ‘drawn out’ from the vitreous. The higher the gradient, the greater the reduction in IOP. To be effective in the eye, an osmotic agent must therefore be unable to penetrate the blood-aqueous barrier. If penetration occurs, an osmotic equilibrium is set up and any further effect is lost. Osmotic agents are therefore of limited value in the treatment of inflammatory glaucomas in which the integrity of the blood-aqueous barrier is compromised.

Clinical uses

When a temporary drop in IOP is required that cannot be achieved by other means.

Page 387
In acute angle-closure glaucoma.
Prior to intraocular surgery when the IOP is very high as may occur from dislocation of the lens into the anterior chamber.
These preparations should be given fairly rapidly and the patient should not subsequently be given fluids to quench thirst until a useful effect has been exerted.

Side-effects

1 Cardiovascular overload may occur as a result of increased extracellular volume. Osmotic agents should therefore be used with great caution in patients with cardiac or renal disease.
2 Urinary retention may occur in elderly men following intravenous administration. Catheterization may be necessary in those with prostatism.
3 Miscellaneous side-effects include headache, backache, nausea and confusion.

Preparations

1 Mannitol is the most widely used intravenous osmotic agent. The dose is 1 g/kg body weight or 5 mL/kg body weight (20% solution in water) over 30–60 minutes. Peak of action is achieved within 30 minutes, with duration up to 6 hours.
2 Glycerol is an oral agent with a sweet and sickly taste. Pure lemon (not orange) juice often needs to be added to avoid nausea. The dose is 1 g/kg body weight or 2 mL/kg body weight (50% solution). Peak action is within 1 hour, with duration up to 3 hours. Although glycerol is metabolized to glucose, it may be given to well-controlled diabetics.
3 Isosorbide is an oral agent with a minty taste. Metabolically inert, it may be given to diabetics without insulin cover. The dose is the same as for glycerol.

Laser therapy

Argon laser trabeculoplasty

Overview

Argon laser trabeculoplasty (ALT) involves the application of discrete laser burns to the trabecular meshwork. This enhances aqueous outflow and lowers IOP. ALT is performed in open-angle glaucomas, usually as an adjunct to medical therapy. It is believed that the procedure causes increased outflow facility by the following mechanisms: (a) mechanical tightening of the trabecular meshwork to open the adjacent, untreated trabecular spaces, and/or (b) inducing cell division and migration of macrophages to clear the trabecular meshwork of debris. ALT is ineffective in paediatric glaucoma and in most secondary glaucomas, with the exception of pigmentary and pseudoexfoliation.

Technique

a A drop of apraclonidine 1% is instilled to avert an early post-laser IOP rise.
b Two drops of a topical anaesthetic are instilled.
c A goniolens is inserted with the mirror at the 12 o’clock position to visualize the inferior angle (usually the easiest part to see).
d The scleral spur, Schwalbe line (which may be pigmented) and the three-dimensional ground-glass appearance of the trabecular meshwork are identified.
e The aiming beam is focused at the junction of the pigmented and non-pigmented trabecular meshwork ensuring that the spot is round and has a clear edge (Fig. 10.79A). An oval spot with an indistinct outline (Fig. 10.79B) means that the aiming beam is not perpendicular to the trabecular surface.
f Initial laser settings are commonly: 50 µm spot size, 0.1 sec and duration and 700 mW power.
g The laser is fired; the ideal reaction is a transient blanching (Fig. 10.80A) or appearance of a minute gas bubble (Fig. 10.80B) at the point of incidence. A large gas bubble (Fig. 10.80C) is excessive.
h If the reaction is inadequate, the power is increased by 200 mW. In a heavily pigmented meshwork, a power setting of 400 mW may suffice, whereas a non-pigmented meshwork may require up to 1200 mW (the average is about 900 mw).
i Twenty-five burns are applied at regularly spaced intervals from one end of the mirror to the other.
Page 388
j The goniolens is rotated clockwise by 90° and a further 25 burns applied making a total of 50 over 180° of the angle. It is important to be familiar with the rotational pattern of the mirror so that adjacent quadrants are treated systematically. With practice it is possible to perform ALT by continuously rotating the goniolens and applying each burn through the centre of the mirror. Some ophthalmologists initially treat 180° and later treat the other 180° if the response is unsatisfactory. Others, however, treat the entire circumference with 100 burns at the initial sitting.
k Apraclonidine 1% is instilled.
l Topical fluorometholone or prednisolone 0.5% q.i.d. daily for a week is prescribed; glaucoma medical therapy is continued.
image

Fig. 10.79 Laser trabeculoplasty. (A) Correct focus of aiming beam; (B) incorrect focus

image

Fig. 10.80 Laser trabeculoplasty. (A) Blanching of trabecular meshwork – appropriate; (B) small bubble – also appropriate; (C) large bubble – excessive; (D) peripheral anterior synechiae due to poor technique

Follow-up

Four to 6 weeks should be allowed for the treatment to take effect. If the IOP is reduced significantly by 6 weeks, gradual withdrawal of medication may be attempted, although total withdrawal is seldom possible. The main aim of ALT is to obtain a safe IOP; the reduction of medication is usually a secondary consideration. If IOP remains high and only 180° has been treated, the remaining 180° is treated. Following 360° ALT, re-treatment is less likely to be beneficial and filtration surgery merits consideration.

Complications

1 Peripheral anterior synechiae (Fig. 10.80D) may develop if the burns are applied too posteriorly or if the energy level is high. In the majority of cases this does not compromise aqueous outflow.
2 Small haemorrhages may develop if the blood vessels on the peripheral iris or ciliary body are inadvertently treated. Such bleeding is easily stopped by applying pressure on the globe with the goniolens.
3 Acute elevation of IOP may occur, especially if prophylactic apraclonidine or brimonidine is not used. The IOP should be monitored carefully in the subsequent weeks in patients with severe glaucomatous damage.
4 Anterior uveitis is fairly common but usually mild, transient and innocuous.
5 Adverse effect on subsequent filtration surgery. The incidence of encapsulated blebs following filtration surgery may be higher in eyes previously treated by ALT.

Results

1 In POAG the initial success rate is 75–85%. The average reduction in IOP is about 30% – eyes with initially high IOPs therefore manifest a greater reduction. Up to 50% of eyes are still controlled after 5 years and about 33% after 10 years. Failure occurs most frequently in the first year; therefore if the IOP is still controlled at 1 year, the probability of control after 5 years is 65% and after 10 years, 40%. If ALT is used as primary treatment, 50% of cases require additional medical therapy within 2 years. Following initially successful ALT, re-treatment carries a low success rate (30% after 1 year and only 15% after 2 years). In general, the results are worse in patients under the age of 50 years. Black patients respond as well as whites initially, but tend to have a more rapid loss of effect.
2 In NPG 50–70% of patients have a good response, but the absolute reduction in IOP is less than in POAG.
3 In pigmentary glaucoma results are generally good, although less so in older patients.
4 In pseudoexfoliation glaucoma initial results are excellent, although failure may occur earlier than in POAG and subsequent rise may be rapid.

Selective laser trabeculoplasty

Selective laser trabeculoplasty (SLT) is a relatively new procedure which uses a 532 nm frequency-doubled, Q-switched Nd:YAG laser, which selectively targets melanin pigment in the trabecular meshwork cells, leaving non-pigmented structures unscathed. Targeting is easier than with ALT, which may lead to more consistent results being achieved. It may be safer than ALT as there is no thermal tissue damage and it is thought that treatment can therefore be repeated. Initial results show that it is probably as effective as ALT.

Nd:YAG laser iridotomy

Indications

PACS, PAC and PACG.
Secondary angle-closure with pupillary block.

Technique

a A drop of apraclonidine 1% or brimonidine 0.2% is instilled 30–60 minutes prior to the procedure.
b The pupil should be miosed with topical pilocarpine, although this may not be possible in acute glaucoma.
c A topical anaesthetic is instilled.
d A special contact lens such as the Abraham iridotomy lens (Fig. 10.81A) is inserted.
e A site is selected, preferably in the superior iris, so that it is covered by the eyelid thus minimizing the risk of monocular diplopia or glare (highest risk if an iridotomy is half-covered by the lid margin). Radially, the iridotomy should be located within the outer third in order to minimize the risk of damage to the crystalline lens. Targeting an iris crypt is beneficial but not essential.
f Laser settings and effective power vary between machines. Most iridotomies are made with settings of 4–5 millijoules (mJ). For a thin blue iris the required energy level is 1–4 mJ per shot. Thick, velvety smooth, brown irides may necessitate higher energy levels.
g The beam is focused precisely and the laser fired. Successful penetration is characterized by a gush of pigment debris. Approximately 10 shots are generally required to produce an adequate iridotomy (Fig. 10.81B) although with an iris crypt this can be reduced to two or three.
h A second drop of apraclonidine 1% is instilled; oral acetazolamide may also be given if necessary.
i A strong topical steroid (e.g. dexamethasone) is prescribed every 10 minutes for 30 minutes and thereafter four times daily for 1 week.
Page 389
j The IOP should be checked 1–2 hours after the procedure to exclude an early spike. Routine review is usually at 1 or 2 weeks, with subsequent monitoring according to individual circumstances. Patients with marked glaucomatous damage may require extended ocular hypotensive cover and earlier review.
image

Fig. 10.81 Nd:YAG iridotomy. (A) Abraham lens; (B) appropriate size opening; (C) too small; (D) not covered by the eyelid and perhaps not sufficiently peripheral

Technical problems

1 Initial failure
Increasing the energy level may be sufficient.
Re-treat the same site after allowing a few minutes for pigment and debris to clear, or move to a different site.
Over-treatment should be avoided due to the risk of substantial postoperative inflammation and pressure spikes; further treatment can be applied after a few days.
In thick brown irides, relatively gentle pre-treatment with argon laser can be beneficial: 10 shots of 0.1 s duration, 200 µm spot size and 200 mW are approximately effective settings.
2 Opening too small (Fig. 10.81C). The optimal iridotomy diameter is 150–200 µm. It is sometimes easier to create an additional opening at a different site rather than to try to enlarge the opening.

Complications

1 Bleeding occurs in about 50% of cases. It is usually slight and stops after several seconds. Persistent bleeding can be terminated by pressing the contact lens against the cornea.
2 IOP elevation within one hour of treatment is common. It is mild and usually transient (see above).
Page 390
3 Iritis is common and usually mild. Severe iritis, which may result in the formation of posterior synechiae, is invariably caused by over-treatment and inadequate post-laser steroid therapy. It is more likely in darker irides, including that due to prostaglandin derivatives.
4 Corneal burns may occur if a contact lens is not used or if the anterior chamber is shallow; they usually heal very rapidly without problems.
5 Lens opacities which are localized and non-progressive occasionally develop at the treatment site; cataract formation may be accelerated by iridotomy.
6 Glare and diplopia may rarely occur if the iridotomy is not sited under the upper eyelid (Fig. 10.81D), particularly if at the eyelid margin.

Diode laser cycloablation

Diode laser ablation (cyclodiode) lowers IOP by destroying part of the secretory ciliary epithelium, thereby reducing aqueous secretion. In the past it was used mainly in uncontrolled end-stage secondary glaucoma with minimal visual potential, mainly to control pain. However, it is now apparent that it can safely be used in eyes with good vision which may be retained provided control of IOP is adequate. More than one treatment session is commonly required for adequate pressure control.

1 Technique
a A sub-Tenon or peribulbar anaesthetic is administered.
b Laser settings are 1.5–2 s and 1500–2000 mW; the spot size is not adjustable.
c The power is adjusted over sequential shots until a ‘popping’ sound is heard and then reduced to just below that level.
d Approximately 12–24 burns are placed posteriorly to the limbus over 360°, avoiding the neurovascular bundles at 3 and 9 o’clock (Fig. 10.82).
e A lower-intensity regimen can be used for eyes with good vision, in order to reduce the risk of over-treatment; more treatment sessions are likely to be required using this approach.
f A strong topical steroid is prescribed hourly on the day of treatment and then 2-hourly for 2 days and q.i.d. for at least 2 weeks. A topical antibiotic and a cycloplegic (cyclopentolate 1% b.d.) are used for 3 days.
g Pre-laser glaucoma treatment may be continued, or reduced slightly.
h Oral non-steroidal anti-inflammatory agents may be prescribed for 2 days.
i Review is generally after 3 or 4 days, because of the risk of significant reactive inflammation.
2 Complications. Mild pain and anterior segment inflammation are common. A temporary IOP rise is not uncommon during the first few weeks. Serious complications are rare and include chronic hypotony, phthisis bulbi, suprachoroidal haemorrhage, corneal decompensation and retinal detachment.
3 Results depend on the type of glaucoma; frequently the procedure has to be repeated. Pain relief is generally good, but does not appear to be solely related to pressure control.
image

Fig. 10.82 (A) Diode laser cycloablation; (B) close up of the probe

(Courtesy of J Salmon – fig. A; Krachmer, Mannis and Holland, from Cornea, Mosby 2005 – fig. B)

Laser iridoplasty

Laser iridoplasty is performed to widen the anterior chamber angle by causing contraction of the peripheral iris away from the angle recess. It can be used to attempt to break an episode of acute angle-closure, but is more commonly applied on an elective basis (see ‘Primary angle-closure).

a A topical anaesthetic is instilled.
b One drop each of 1% pilocarpine and 1% apraclonidine is instilled.
c Via an iridotomy lens, 1–2 burns per clock hour are applied to the periphery, 500 µm size, 100–200 mW, 0.5 s duration, aiming for slight visible iris contraction; overtreatment should be avoided as prolonged IOP spikes can occur.
d Post-procedure 1% apraclonidine is given (consider oral acetazolamide if significant glaucomatous optic neuropathy is present).
e Topical prednisolone 1% or dexamethasone 0.1% hourly for the first day then four times daily.
f Review 1–2 hours post-laser, then after 1 week and subsequently depending on progress and glaucomatous damage – patients with significant glaucomatous neuropathy may need regular review for the first few weeks to detect and treat an IOP spike.
g Altered accommodation is fairly common but almost always transient.
Page 391

Trabeculectomy

Trabeculectomy lowers IOP by creating a fistula, to allow aqueous outflow from the anterior chamber to the sub-Tenon space. The fistula is protected or ‘guarded’ by a superficial scleral flap (Fig. 10.83). The procedure is usually performed when medical therapy has failed to achieve adequate control of IOP.

image

Fig. 10.83 Trabeculectomy principles. (A) Pathway of aqueous egress following trabeculectomy; (B) appearance from inside the eye following completion

Technique

a The pupil is miosed.
b A bridle suture is inserted either into peripheral clear cornea superiorly or into the superior rectus muscle.
c A limbal or fornix-based flap of conjunctiva and Tenon’s capsule is fashioned superiorly.
d Episcleral tissue is cleared. An outline of the proposed superficial scleral flap is made with wet-field cautery.
e Incisions are made along the cautery marks through two-thirds of scleral thickness, to create a ‘trapdoor’ lamellar scleral flap (Fig. 10.84A). This flap may be rectangular (3 × 4 mm) or triangular, according to preference.
f The superficial flap is dissected forwards until clear cornea has been reached (Fig. 10.84B).
g A paracentesis is made in superotemporal peripheral clear cornea.
h The anterior chamber is entered along the width of the trapdoor flap.
i A block of deep sclera is excised with a punch (Fig. 10.84C).
j A peripheral iridectomy is performed in order to prevent blockage of the internal opening (Fig. 10.84D); some surgeons omit this step in pseudophakic eyes or those with a deep anterior chamber.
k The superficial scleral flap is sutured at its posterior corners so that it is lightly apposed to the underlying bed.
l Alternatively, the flap may be sutured tightly with releasable or lysable sutures to reduce the risk of postoperative scleral flap leakage and shallow anterior chamber.
m Balanced salt solution is injected into the anterior chamber through the paracentesis. This tests the patency of the fistula and facilitates the detection of any holes or leaks in the flap.
n Conjunctiva/Tenon’s capsule flap is sutured. Irrigation through the paracentesis is repeated to produce a bleb, which is then checked for leakage.
o A drop of atropine 1% is instilled; when no iridectomy has been performed, pilocarpine 2% may be used instead.
p A steroid and an antibiotic are injected under the inferior conjunctiva.
q Steroid-antibiotic drops are used four times daily for 1–2 weeks and then changed to prednisolone acetate 1% or dexamethasone 0.1% for a further 8–10 weeks.
image

Fig. 10.84 Trabeculectomy technique. (A) Outline of superficial scleral flap; (B) dissection of superficial scleral flap; (C) excision of deep scleral tissue with a punch; (D) peripheral iridectomy

Shallow anterior chamber

A shallow anterior chamber may be due to (a) pupillary block, (b) overfiltration or (c) malignant glaucoma (aqueous misdirection). Severe and sustained shallowing is uncommon (Fig. 10.85A and B), the chamber re-forming spontaneously in most cases. However, those that do not may develop severe complications such as peripheral anterior synechiae, corneal endothelial damage (Fig. 10.85C) and cataract (Fig. 10.85D).

image

Fig. 10.85 Shallow anterior chamber. (A) Peripheral iris-corneal apposition; (B) pupillary border-corneal apposition; (C) lenticulo-corneal apposition resulting in corneal oedema; (D) cataract following inappropriate management

(Courtesy of J Schuman, V Christopoulos, D Dhaliwal, M Kahook and R Noecker, from Lens and Glaucoma, in Rapid Diagnosis in Ophthalmology, Mosby 2008 – fig. A)

Pupillary block

Pupillary block may occur with a non-patent peripheral iridectomy.

1 Signs
High IOP and flat bleb.
Negative Seidel test.
Iris bombé with a non-patent iridectomy.
Page 392
2 Treatment involves YAG laser to the pigment epithelium at the iridectomy site if the anterior iris stroma appears to have been largely penetrated, or a new laser iridotomy is performed.

Overfiltration

Overfiltration may be caused by scleral flap leakage due to insufficient resistance to outflow by the lamellar scleral flap, but bleb leakage through an inadvertent buttonhole or due to inadequate closure of the conjunctiva and Tenon capsule is perhaps the most common cause.

1 Signs
Low IOP with a well formed bleb in a scleral flap leak and flat in a bleb leak.
Seidel test is negative in a scleral flap leak and a positive in a bleb leak (Fig. 10.86A).
The cornea may manifest signs of hypotony such as folds in Descemet membrane.
Choroidal detachment may be present (Fig. 10.86B).
2 Treatment depends on the cause and degree of shallowing.
a Initial conservative treatment in eyes without lens-corneal touch is observation, with atropine to prevent PAS formation and malignant glaucoma.
b Subsequent treatment if the above measures are ineffective involves temporary tamponade of the conjunctiva to enhance spontaneous healing by simple pressure patching, a large diameter soft bandage contact lens, a collagen shield or a Simmons shell designed for the purpose.
c Definitive treatment often consists of inserting additional conjunctival sutures, and if necessary placing a transconjunctival scleral flap suture. If potentially serious shallowing is present, the anterior chamber can be reformed with a viscoelastic. Choroidal detachments rarely require drainage.
image

Fig. 10.86 (A) Positive Seidel test; (B) choroidal detachment

Malignant glaucoma

Page 393

Malignant glaucoma is rare but serious. It is caused by anterior rotation of the ciliary processes and iris root, often with aqueous misdirection (ciliolenticular block); blockage of aqueous flow occurs in the vicinity of the pars plicata of the ciliary body, so that aqueous is forced backwards into the vitreous.

1 Signs
High IOP and absent bleb.
Negative Seidel test.
2 Treatment
a Initial treatment is with mydriatics (atropine 1% and phenylephrine 10%) to dilate the ciliary ring and increase the distance between the ciliary processes and the equator of the lens, thereby tightening the zonule and pulling the lens posteriorly into its normal position. Intravenous mannitol may be used if mydriatics are ineffective in order to shrink the vitreous gel and allow the lens to move posteriorly.
b Subsequent treatment if medical therapy fails is with Nd:YAG laser fired through the iridectomy in order to disrupt the anterior hyaloid face, reduce the vitreous volume and break any ciliary block. In pseudophakic eyes, laser posterior capsulotomy and disruption of the anterior hyaloid face should be performed. Cyclodiode may be effective in some cases. Pars plana vitrectomy is performed if laser therapy fails: sufficient vitreous gel is excised to allow free flow of aqueous to the anterior chamber.
Page 394

Failure of filtration

Diagnosis

A normally functioning bleb should be slightly elevated, avascular and show superficial microcysts (Fig. 10.87A). Poor filtration is indicated by increasing IOP and a bleb with one of the following appearances:

1 Flat without vascularization (Fig. 10.87B).
2 Vascularized bleb due to episcleral fibrosis (Fig. 10.87C).
3 Encapsulated bleb (Tenon cyst) which typically develops 2–8 weeks postoperatively. It is characterized by a localized, highly elevated, dome-shaped, firm, fluid filled cavity of hypertrophied Tenon capsule with engorged surface blood vessels (Fig. 10.87B).
image

Fig. 10.87 Filtering blebs. (A) Normal; (B) flat non-functioning; (C) vascularized; (D) encapsulated – Tenon cyst

Causes

Causes of failure can be classified according to the site of blockage:

1 Extrascleral causes include (a) subconjunctival and episcleral fibrosis and (b) occasionally bleb encapsulation.
2 Scleral causes include (a) an over-tight suturing of the scleral flap, and (b) gradual scarring in the scleral bed may lead to obstruction of the fistula at that level.
3 Intraocular causes are uncommon and include (a) blockage of the sclerostomy by vitreous, blood or uveal tissue and (b) obstruction of the internal opening by a variety of thin membranes derived from surrounding cornea or sclera.

Management

Management of filtration failure depends on the cause and may involve one or more of the following:

1 Ocular compression in an effort to force outflow through the surgical fistula may be performed by (a) digital compression through the lower lid with the eyes closed and the patient looking straight ahead or (b) at the slit-lamp with a moistened sterile cotton bud at the edge of the scleral flap in an attempt to promote outflow.
2 Suture manipulation may be considered 7–14 days postoperatively if the eye has high IOP, a flat bleb and a deep anterior chamber. Releasable sutures can be cut or released according to the technique of initial placement. Argon laser suture lysis is useful if releasable sutures have not been used. It may be performed through a suture lysis lens or a Zeiss four-mirror goniolens.
3 Needling of an encysted bleb may be performed at the slit-lamp or operating microscope under topical anaesthesia. It can be augmented with 5-fluorouracil to enhance the success rate.
4 Subconjunctival injection of 5-fluorouracil may be used in the first 7–14 days to suppress episcleral fibrosis; 5 mg (0.1 mL of 50 mg/mL solution) is injected approximately 10 mm away from the bleb and can be repeated as necessary.

Late bleb leakage

1 Cause is disintegration of conjunctiva overlying the sclerostomy following previous operative application of antimetabolites, particularly mitomycin C. Necrosis of the surface epithelium results in transconjunctival drainage of aqueous.
2 Complications of untreated leaks include infection and hypotony maculopathy (see Ch. 14).
3 Signs
Low IOP and an avascular cystic bleb.
Seidel testing may initially be negative with only multiple punctate staining areas (sweating) being seen. Later the formation of a hole results in gross leakage with a positive test.
Shallow anterior chamber and choroidal detachment may be present in severe cases.
Page 395
4 Treatment is difficult. The following are some of the methods used, none of which are universally successful.
a Initial treatment is as for early postoperative overfiltration but is seldom successful.
b Subsequent treatment depends on whether the leakage involves merely ‘sweating’ or is due to a hole.
Sweating blebs may be treated by injection of autologous blood into the bleb, ‘compression’ sutures or a transconjunctival scleral flap suture.
Full thickness holes usually require revisional surgery, such as conjunctival advancement to hood the existing bleb, free conjunctival patch autografts with removal of the existing bleb and scleral grafts to limit flow through the sclerostomy.

Bleb-associated bacterial infection and endophthalmitis

Glaucoma filtration-associated infection is classified as limited to the bleb (blebitis) or endophthalmitis, although there is some overlap. The incidence of blebitis following trabeculectomy with mitomycin has been estimated to be up to 5% per year but many studies show a for lower rate.

Pathogenesis

Page 396

Adjunctive antifibrotic agents (mitomycin C, 5-fluorouracil) are frequently used to increase the success of glaucoma filtration surgery. The use of these agents can lead to a very thin-walled drainage bleb (Fig. 10.88A) that significantly increases the risk of late-onset infection. The infection presumably gains access directly through the thin and avascular wall of the drainage bleb. All patients with such blebs should be warned of the possibility of late infection and strongly advised to report immediately should they develop a red and sticky eye, or blurred vision (RSVP – red, sticky, visual loss, pain).

1 Risk factors include blepharitis, higher does of mitomycin, long-term topical antibiotic use, an inferior or nasally placed bleb, and bleb leak. Late bleb leaks should be treated aggressively to reduce the risk of infection.
2 Pathogens. The most frequent are H. influenzae, Streptococcus spp., and Staphylococcus spp. The often poor visual prognosis is related to the virulence of these organisms.
image

Fig. 10.88 (A) Thin-walled bleb; (B) blebitis; (C) endophthalmitis with hypopyon

Blebitis

Blebitis describes infection without vitreous involvement.

1 Presentation is with mild discomfort and redness.
2 Signs
A white bleb that appears to contain inflammatory material (Fig. 10.88B).
Anterior uveitis may be absent.
The red reflex is normal.
3 Investigation. A conjunctival swab should be taken; a sample should not be aspirated from within the bleb.
4 Treatment
Topical ofloxacin and cefuroxime (or vancomycin 50 mg/mL) hourly.
Oral co-amoxiclav 500/125 mg t.i.d. and ciprofloxacin 750 mg b.d. for 5 days; azithromycin 500 mg daily for 5 days is an alternative.

Endophthalmitis

1 Presentation is with a short history of rapidly worsening vision, pain and redness.
2 Signs
White milky bleb containing pus.
Severe anterior uveitis that may be associated with hypopyon (Fig. 10.88C).
Vitritis and impairment of the red reflex.
3 Treatment involves topical and systemic therapy as for blebitis. Intravitreal antibiotics as for acute post-operative endophthalmitis following cataract extraction (see Ch. 9) should be considered if there is no early response.

Non-penetrating surgery

Overview

Page 397

In non-penetrating filtration surgery the anterior chamber is not entered and the internal trabecular meshwork is preserved, thus reducing the incidence of postoperative overfiltration and hypotony and its potential sequelae. Two lamellar scleral flaps are fashioned and the deep flap excised leaving behind a thin membrane consisting of trabeculum/Descemet membrane through which aqueous diffuses from the anterior chamber to the subconjunctival space. The surgery is technically challenging and requires meticulous dissection of a deep scleral flap without entering the delicate anterior trabecular meshwork.

Indications

The main indication for non-penetrating surgery is POAG, although other open-angle glaucomas may also be amenable. In general the IOP reduction is less than that achieved by trabeculectomy, so that topical medication often needs to be recommenced. Conventional filtration is therefore still the procedure of choice when the target IOP is in the low teens though it is probably associated with a lower risk of ‘snuffing out’ central vision when advanced damage is present.

Technique

1 Deep sclerectomy in which a Descemet window is created, that allows aqueous seepage from the anterior chamber (Fig. 10.89). Subsequent egress is subconjunctival resulting in a shallow filtration bleb, as well as along deeper suprachoroidal routes. The long-term results can be enhanced by using a collagen implant at the time of surgery and postoperative application of Nd:YAG laser to the meshwork at the surgical site using a gonioscope (goniopuncture).
2 Viscocanalostomy involves the creation of a filtering window, identification and dilatation of Schlemm canal with high density viscoelastic. The superficial scleral flap is sutured tightly so that subconjunctival fluid outflow and bleb formation are minimized. The procedure probably causes inadvertent microscopic ruptures in the juxtacanalicular tissue and meshwork. A variation on this procedure involves the cannulation of the entire circumference of Schlemm canal with a microcatheter (canaloplasty).
image

Fig. 10.89 Non-penetrating filtration surgery: deep sclerectomy. (A) Dissection of scleral flap; (B) dissection into clear cornea exposing Schlemm canal; (C) collagen implant; (D) shallow diffuse avascular bleb

(Courtesy of A Mermoud)

Trabectome

The Trabectome is a novel microelectrosurgical device which approaches the angle ab interno under direct vision using a gonioscopy lens, to remove a strip of trabecular meshwork and inner wall of Schlemm canal (‘trabeculotomy’). Whilst it does not seem to lower the intraocular pressure as effectively as trabeculectomy, the safety profile is better.

Antimetabolites in filtration surgery

Indications

Page 398

Adjunctive antimetabolites inhibit the natural healing response that may preclude successful filtration surgery. They should, however, be used with caution because of the serious nature of potential complications, and usually considered in the presence of known risk factors for failure of trabeculectomy. In uncomplicated glaucoma the use of low-dose antimetabolites may improve long-term control of IOP but this benefit should be weighed against possible complications such as corneal epithelial defects, chronic hypotony and late-onset bleb leakage.

1 High risk factors
Neovascular glaucoma.
Previous failed trabeculectomy or artificial filtering devices.
Certain secondary glaucomas (e.g. inflammatory, post-traumatic angle recession and iridocorneal endothelial syndrome).
2 Intermediate risk factors
Patients on topical medication (particularly sympathomimetics) for over 3 years.
Previous conjunctival surgery.
Previous cataract surgery.
3 Lower risk factors
Black patients.
Patients under the age of 40 years.

5-fluorouracil

5-fluorouracil (5-FU) inhibits DNA synthesis and is active on the ‘S’ phase (synthesis phase) of the cell cycle. Fibroblastic proliferation is inhibited, but fibroblastic attachment and migration are unaffected. It is the antimetabolite of choice in elderly patients who have risk factors for failure. The drug can be used in one or both of the following ways:

1 Intraoperative application is as follows:
a The conjunctival flap is dissected.
b A small cellulose sponge is soaked in a 50 mg/mL solution of 5-FU.
c The sponge is placed under the dissected flap of Tenon’s capsule at the site of filtration making sure that the edges of the conjunctival incision are not exposed to the drug.
d The sponge is removed after 5 minutes.
e The space between the conjunctiva and episclera is thoroughly irrigated with balanced salt solution.
f The trabeculectomy is completed.
2 Subconjunctival injection of 5 mg daily postoperatively for up to 7 days as follows:
a The eye is anaesthetized with a cotton pledget soaked in amethocaine.
b 0.5 mL of 5-FU (50 mg/mL) is drawn up into a tuberculin syringe.
c The 27-guage needle is exchanged for a 30-guage needle.
d The bubbles are shaken to the top of the syringe.
e 0.4 mL of 5-FU is expressed so than only 0.1 mL remains in the syringe.
f The contents of the syringe are injected subconjunctivally 180° away from the filtration site.
g Any reflux is caught on a dry cotton-tipped applicator or irrigated out.

Mitomycin C

Mitomycin C (MMC) is an alkylating agent rather than an antimetabolite, and selectively inhibits DNA replication, mitosis and protein synthesis. The drug inhibits proliferation of fibroblasts, suppresses vascular ingrowth and is much more potent than 5-FU. Optimum concentration and exposure time is not known and vary between 0.2–0.5 mg/mL and 1–5 minutes. In general, low or intermediate risk indicates use of a low concentration (0.2 mg/mL), whilst high risk implies the need for a higher concentration (0.4–0.5 mg/mL). Higher concentrations and extended exposure times are associated with an increased risk of complications. The technique of intraoperative application is the same as for 5-FU and great care should be taken to prevent contamination of the anterior chamber. MMC can also be applied externally to the bleb with a sponge in the postoperative period.

Complications

1 Corneal epithelial defects and postoperative wound leaks occur mainly after the use of 5-FU.
2 Cystic thin-walled blebs may occur following the use of both 5-FU and mitomycin C and may predispose to chronic hypotony, late-onset bleb leak and endophthalmitis.

Drainage shunts

Page 399

Shunts using episcleral explants

Types

These create a communication between the anterior chamber and sub-Tenon space. All such shunts consist of a tube attached to a posterior episcleral explant. Some contain pressure-sensitive valves for regulation of aqueous flow. Reduction of IOP is due to passive, pressure-dependent flow of aqueous across the capsular wall.

1 Molteno implant consists of a silicone tube connected to one or two polypropylene plates 13 mm in diameter (Fig. 10.90).
2 Baerveldt implant consists of a silicone tube connected to a large area silicone plate impregnated with barium.
3 Ahmed implant consists of a silicone tube connected to a silicone sheet valve held in a polypropylene body. The valve mechanism consists of two thin silicone elastomer membranes.
image

Fig. 10.90 (A) Molteno implant; (B) postoperative appearance

(Courtesy of P Gili – fig. B)

Indications

Uncontrolled glaucoma despite previous trabeculectomy with adjunctive antimetabolite therapy.
Secondary glaucoma where routine trabeculectomy, with or without adjunctive antimetabolites, is unlikely to be successful. Examples include neovascular glaucoma and glaucoma following traumatic anterior segment disruption.
Eyes with severe conjunctival scarring precluding accurate dissection of the conjunctiva.
Certain congenital glaucomas where conventional procedures have failed (i.e. goniotomy, trabeculotomy and trabeculectomy).

Complications

1 Excessive drainage may occur due to leakage around or down the tube if the occluding suture is loose and result in hypotony and a shallow anterior chamber.
2 Malposition that may result in endothelial or lenticular touch (Fig. 10.91A).
3 Tube erosion through the sclera and conjunctiva (Fig. 10.91B).
4 Early drainage failure may occur as a result of blockage of the end of the tube by vitreous, blood or iris tissue (Fig. 10.91C).
5 Late drainage failure occurs in about 10% of cases per year and is comparable to that following trabeculectomy.
image

Fig. 10.91 Complications of drainage implants. (A) Malposition; (B) tube erosion; (C) blockage by iris

(Courtesy of J Salmon – fig. B; R Bates – fig. C)

Page 400

Results

The results depend on the type of glaucoma. In general, an IOP in the mid-teens is achieved, though as with trabeculectomy, topical medication is typically required. The long-term success rate in neovascular glaucoma is often disappointing because of progressive retinal disease with loss of vision and late development of phthisis bulbi. Adjunctive mitomycin C may enhance the success rate of drainage shunt surgery but is associated with a higher complication rate.

Mini shunts

These are generally used in uncomplicated glaucoma.

1 ExPress™ Mini Shunt is a new device without a valve that is inserted under a scleral flap during a modified trabeculectomy. The technique is relatively straightforward in comparison with other shunts.
2 iStent® is another novel device consisting of a tiny titanium hooked tube inserted ab interno into Schlemm canal via the trabecular meshwork, and shows promise for moderate IOP reduction.