Werner & Ingbar's The Thyroid: A Fundamental & Clinical Text, 9th Edition

23B.Ophthalmopathy

Petros Perros

A. Jane Dickinson

Ophthalmic abnormalities, better known as ophthalmopathy, are the principal extrathyroidal manifestations of Graves' disease. The numerous synonyms for Graves' ophthalmopathy (orbitopathy, thyroid eye disease) reflect its multifaceted clinical expression and the uncertainties about its pathogenesis, natural history, and treatment.

EPIDEMIOLOGY

Graves' ophthalmopathy is clinically evident in about a third of patients with Graves' disease, although it can be demonstrated by orbital imaging in nearly all (1,2,3). Its incidence seems to have declined in recent years for reasons that are unclear (3). Most patients have mild eye disease not requiring specific treatment, while approximately 5% have severe potentially sight-threatening ophthalmopathy (2). The incidence of Graves' ophthalmopathy in the United States, based on a population study, was 16/100,000 per annum for women and 2.9/100,000 per annum for men (4). The age distribution at the time of presentation shows two peaks, one at 40 to 44 years and a later one at 60 to 64 years for women, and 65 to 69 years for men (1). The female to male ratio in patients attending specialist centers is approximately 2:1 (2).

PATHOGENESIS

Current evidence favors an autoimmune pathogenesis with important genetic and environmental influences, particularly smoking (5). Orbital muscle, connective tissue, and adipose tissue are infiltrated by lymphocytes and macrophages (2). The extracellular compartment of extraocular muscles and orbital fibroadipose tissue becomes edematous, secondary to deposition of hydrophilic glycosoaminoglycans, while the muscle cells themselves are unaffected (1,5).

Anatomic Considerations

The orbit resembles a rigid cone with its base open anteriorly, where it is bounded by the semidistensible anterior orbital septum. Edematous expansion of orbital tissue has predictable consequences: dysfunction of affected muscles, increased orbital pressure, and proptosis (exophthalmos, forward displacement of the eye). The clinical expression relating to each of these factors depends on the site and severity of inflammation and the potential for forward displacement. Several extraocular muscles, including the levator palpebrae superioris, are usually affected, although with variable frequency (2,6,7). This leads to restriction of the eye (dysmotility), lid lag, and incomplete eyelid closure. The latter may provoke sight-threatening corneal exposure, often compounded by proptosis. The degree of proptosis is limited by the length of the rectus muscles and the tightness of the anterior orbital septum. If the muscles are unable to stretch and the septum is tight, proptosis is minimal, but orbital pressure and venous congestion increase, and the rectus muscles may then compress the optic nerve at the orbital apex, with resulting visual loss. Conversely, if the septum is lax and the muscles are able to stretch, then proptosis increases, occasionally allowing subluxation of the eyeball. Acute inflammation may also cause erythema and swelling of the conjunctivae and eyelids, compounded by venous and lymphatic congestion. Muscle inflammation gives way to fatty degeneration and scarring, sometimes with further tethering and restriction. Various stages of this process may coexist in one or more muscles.

Immunology

The expansion of orbital tissues is primarily due to the edema that results from deposition of the very hydrophilic glycosaminoglycans. Orbital fibroblasts appear to synthesize and secrete these glycosoaminoglycans in response to cytokines produced by infiltrating immune cells and macrophages and by the fibroblasts themselves (5). Cytokines also stimulate orbital fibroblasts, vascular endothelium, and macrophages to produce other immunomodulatory and inflammatory mediators. These in turn aid recruitment of T cells into the orbit and antigen recognition and presentation, and therefore perpetuate the local inflammatory response (5). A cell-mediated (Th1-type) immune response appears prominent early in the evolution of Graves' ophthalmopathy, whereas humoral immunity (Th2-type), or both, may be relevant later in the course of the disease (5,6).

Evidence of autoimmune responses to thyroid antigens is invariably present in patients with Graves' ophthalmopathy, including those with euthyroid Graves' ophthalmopathy (2). An autoantigen shared by thyroid and orbit would explain the specific targeting of the latter, and the close association between Graves' thyroid disease and Graves' eye disease. The most studied and promising candidate autoantigen is the thyrotropin (TSH) receptor. This is expressed in orbital connective tissue, orbital fat (5,8,9), and extraocular muscle fibers (10), but at higher levels in patients with Graves' ophthalmopathy than normal subjects (5). Orbital connective tissue contains adipocyte precursors (preadipocytes), which under appropriate conditions can differentiate into adipocytes (11). In vitro, expression of TSH receptors increases as fibroblasts differentiate to preadipocytes and adipocytes (12), and both differentiation and TSH-receptor expression are stimulated by cytokines (5). These are likely events in the evolution of Graves' ophthalmopathy, because there is often an increase in orbital fat content (11). A role for antibodies to the TSH receptor is also possible (13,14). Other candidate antigens shared between the thyroid and orbital tissue include a surface antigen known as G2s, thyroglobulin (5), and the insulin-like growth factor (IGF-1) receptor (15). Two animal models for Graves' ophthalmopathy have been described. Mice immunized with the cDNAs for TSHR or G2s and TSHR appear to develop orbital infiltration similar to Graves' ophthalmopathy (5).

Genetic and Other Influences

The genetic contribution to Graves' disease is substantial (see Chapter 20). The same risks and associations apply to Graves' ophthalmopathy (16). Several studies examining genes that may distinguish between patients with Graves' disease with and without eye disease have focused on associations between ophthalmopathy and alleles at the major histocompatibility complex (MHC), cytotoxic lymphocyte-associated antigen (CTLA)-4 (17), and other loci (18). These studies have yielded contradictory results, which may be partly due to differences in allelotyping methodology, disease definition, race/ethnicity of the study subjects, and small sample sizes (16). Other factors that are associated with severe eye disease include advanced age, male sex, smoking, and persistent thyroid dysfunction, particularly hypothyroidism (1,5).

NATURAL HISTORY

The onset of eye disease usually coincides with that of the thyrotoxicosis however, exceptions are not uncommon, and Graves' ophthalmopathy can precede or follow thyrotoxicosis by months or even years (Fig. 23B.1) (2). The severity of ophthalmopathy follows a phasic pattern: a phase of progressive deterioration lasting several months; a short period of peak severity; a phase of spontaneous improvement lasting up to a year or longer; and a quiescent phase when inflammatory signs disappear and clinical features stabilize, although they do not usually resolve completely. This pattern of change in severity over time was first described years ago (Fig. 23B.2) and has been confirmed by later studies (19). A related concept is disease activity, which relates to the presence of an acute inflammatory process within the orbit. Change in activity is implied by change in severity over time; its assessment is discussed later. Graves' ophthalmopathy rarely becomes active again once it has become quiescent, but the activity and course may vary in individual eyes (20).

FIGURE 23B.1. Relationship between onset of Graves' ophthalmopathy and thyrotoxicosis. Most patients who ultimately have clinically important Graves' ophthalmopathy have some mild features of ophthalmopathy at the time of presentation with thyrotoxicosis (0 months). In a few patients ophthalmopathy precedes, in slightly more it follows the onset of thyrotoxicosis by 1 to 12 months, and in occasional patients the interval is much longer. (From Bartalena L, Pinchera A, Marcocci C. Management of Graves' ophthalmopathy: reality and perspectives. Endocr Rev 2000;21:168, with permission.)

FIGURE 23B.2. The natural history of the changes in severity (bold line) and activity (thin line) of Graves' ophthalmopathy as a function of time. The phases of severity of ophthalmopathy are progressive deterioration, plateau, slow improvement, and, lastly, stability. The phases of activity of ophthalmopathy relate to changes in severity but are more prolonged. (From Rundle FF, Wilson CW. Development and course of exophthalmos and ophthalmoplegia in Graves' disease with special reference to the effect of thyroidectomy. Clin Sci 1945;5:177, with permission.)

CLINICAL PRESENTATION

Graves' ophthalmopathy is usually bilateral, but it may be asymmetric. The onset can be rapid or insidious. Typical initial symptoms relate most commonly to discomfort of the surfaces of the eyes and changes in appearance, particularly periorbital swelling (Table 23B.1). These symptoms usually develop gradually over weeks to months, and although itching is absent, misdiagnosis as allergy is common. Other common symptoms are retroorbital pain, pain provoked by gazing in particular directions, and diplopia (21). Some patients describe slight diplopia as “blurring”; however, the blurring clears with monocular occlusion (2,21). Significant dysmotility may be accompanied by a compensatory head tilt (Fig. 23B.3A). The myopathy of Graves' ophthalmopathy relates to failure of relaxation, thus impeding the action of ipsilateral antagonist muscles. Rectus muscle involvement is usually asymmetrical, with restriction most evident in the inferior and medial rectus muscles and rare in the lateral rectus muscles. Hence, upward gaze and abduction are commonly restricted. Although dysmotility affects up to 60% of patients with Graves' ophthalmopathy (2), many have no diplopia due to symmetric involvement of both orbits or amblyopia, or because the restriction affects extremes of gaze irrelevant to daily life. Optic neuropathy is uncommon, but such patients may notice blurred vision unaffected by blinking or closing one eye, reduced color appreciation, or an awareness of gray areas of field loss (2,21). Older men with diabetes and vascular disease are most at risk for optic neuropathy (21).

TABLE 23B.1. EVALUATION OF SYMPTOMS OF GRAVES' OPHTHALMOPATHY


Patients are asked whether any of the following symptoms are present, whether the symptom has changed in the last 1 to 2 months, and its severity on a scale of 0 to 10.

Overall appearance of the eyes

   Normal

   Abnormal

Surface symptoms

   Grittiness

   Excessive watering

   Aversion to bright light

Swelling of the eyelids

Protrusion of the eyes

Pain or ache in or behind the eye

   Spontaneous orbital pain

   Pain only on gaze in a particular direction

Double vision

   Intermittent: apparent on waking but resolves, apparent only when tired, or apparent intermittently at other times.

   Inconstant: always present in certain directions of gaze

   Constantly present unless a particular head posture is adopted

   Constantly present regardless of head posture

Blurred vision

   Clears with blinking

   Clears on covering one eye

   Persistent blurring or awareness of gray areas in field of vision

Reduced color intensity or difference between eyes


FIGURE 23B.3. Photographs of eyes of patients with Graves' ophthalmopathy. A: Extraocular muscle dysmotility associated with compensatory head tilt and chin-up position. B: Concealed proptosis (exophthalmos): acuity was 20/100 with loss of color vision. C: Lid retraction in a patient with inactive ophthalmopathy. D: Lateral flare of the contour of the upper eyelid. E: Proptosis and lid swelling caused by forward displacement of orbital fat in a patient with inactive ophthalmopathy. F: Lid swelling due to active ophthalmopathy. G: Eyelid festoons (marked edema of lids). H: Dilated superficial vessels over insertion of the lateral rectus muscle in a patient with active ophthalmopathy disease. I: Inflamed caruncle (arrowhead) and plica (arrow) in a patient with active ophthalmopathy. J: Chemosis in a patient with active ophthalmopathy (slit-lamp examination). K: Superior limbic keratoconjunctivitis. L: Corneal ulcer. (From Dickinson AJ, Perros P. Controversies in the clinical evaluation of active thyroid-associated orbitopathy: use of a detailed protocol with comparative photographs for objective assessment. Clin Endocrinol (Oxf) 2001;55:283, with permission; and Perros P, Dickinson AJ, Kendall-Taylor P. Clinical presentation and natural history of Graves' ophthalmopathy. In: Bahn RS, ed. Thyroid eye disease. Boston: Kluwer Academic Publishers, 2001:119, with permission.)

Most patients with ophthalmopathy develop thyroid dysfunction concurrently or several months before onset ophthalmopathy (Fig. 23B.1), but 6% to 10% have characteristic ophthalmopathy without discernible thyroid dysfunction, termed euthyroid Graves' ophthalmopathy or ophthalmic Graves' disease (2). Most of these patients have onset of thyrotoxicosis within 18 months. About 5% of patients with Graves' ophthalmopathy have primary hypothyroidism rather than thyrotoxicosis. Concealed proptosis (Fig. 23B.3) is a clinical variant in which ophthalmopathy is obscured by only minimal proptosis. It is not uncommon, and is important to recognize because of the high risk of optic neuropathy; these patients almost always have tense orbits and dysmotility, and often aching eye pain.

SEVERITY AND ACTIVITY OF EYE DISEASE

Severity is defined as the degree of functional or cosmetic deficit at any time point in the course of the disease (21). Thus, variations in periorbital edema, proptosis, diplopia, corneal integrity, and optic nerve function are measures of severity. In contrast, activity implies the presence of acute inflammation, and therefore potential for change either spontaneously or in response to medical treatment. Activity can be inferred directly when symptoms and signs of acute inflammation are present, or by demonstrating change in measures of severity via sequential assessments.

CLINICAL ASSESSMENT

The aim of initial assessment is to verify the diagnosis, ascertain severity, and determine the phase of the ophthalmopathy (Fig. 23B.2). A detailed assessment of symptoms (Table 23B.1) helps to highlight the patient's concerns and priorities, and informs mutual understanding and counselling. Ascertaining severity allows the clinician to determine the need for interventional treatment, while ascertaining activity and disease phase dictates the therapeutic options then available. Disease phase may be partly determined by a trend in symptoms, but reassessment in two to three months is sometimes required.

Diagnosis

Most signs of Graves' ophthalmopathy are nonspecific, reflecting inflammation or volume changes in extraocular muscles or other tissue within the confined retrobulbar space; hence other causes of ophthalmopathy should be considered. These include myasthenia gravis, retroorbital tumors, carotid-cavernous fistula, and especially other inflammatory orbitopathies (2,22,23). Nevertheless, Graves' ophthalmopathy is the most common cause of not only bilateral but also unilateral proptosis (2,22). Typically, the combination of ophthalmic symptoms and signs plus the presence of thyrotoxicosis at that time or in the recent past will confirm the diagnosis without the need for supplementary studies (24). In patients with eyelid retraction (Fig. 23B.3C), Graves' ophthalmopathy is considered to be present if the patient has one of the following: abnormal thyroid function, proptosis, optic neuropathy, or restrictive myopathy. If eyelid retraction is absent, then thyroid dysfunction must be present together with proptosis, optic neuropathy, or restrictive myopathy to reach a confident diagnosis. Other signs that are highly suggestive of Graves' ophthalmopathy are lid lag and a change in the contour of the upper eyelid known as lateral flare (Fig. 23B.3D).

Evaluation of Disease Severity

Recommendations for objective assessment of Graves' ophthalmopathy have been proposed (25), but there is no agreement as to precisely what to score and how to define it (21). A more detailed protocol with a soft tissue atlas was published (21) and adopted by the European Group on Graves' ophthalmopathy (EUGOGO) (26); however, worldwide consensus remains elusive. Given the variable presentation of Graves' ophthalmopathy, summary descriptions are of little use, and each feature should be assessed individually. The NOSPECS scheme (Table 23B.2) provides a useful mnemonic to assist this, although as a scoring system it has been justifiably criticized for poor definitions and reproducibility, and for failing to record important change (2,21).

TABLE 23B.2. NOSPECS CLASSIFICATION OF THE OCULAR CHANGES IN GRAVES' DISEASE


Classes

Grades

Ocular Symptoms and Signs


0

 No symptoms or signs

1

 Only (signs limited to upper lid retraction and stare, with or without lid lag and proptosis)

2

 Soft tissue involvement (symptoms of excessive lacrimation, sandy sensation, retrobulbar discomfort, and photophobia, but not diplopia); objective signs as follows:

 0

Absent

 a

Minimal (edema of conjunctivae and lids, conjunctival injection, and fullness of lids, often with orbital fat extrusion, palpable lacrimal glands, or swollen extraocular muscle palpable beneath lower lids)

 b

Moderate (above plus chemosis, lagophthalmos lid fullness)

c

Marked

3

Proptosis associated with classes 2 to 6 only (specify if inequality of 3 mm or more between eyes, or if progression of 3 mm or more under observation)

0

Absent (20 mm or less)

a

Minimal (21–23 mm)

b

Moderate (24–27 mm)

 c

Marked (28 mm or more)

4

Extraocular muscle involvement (usually with diplopia)

0

Absent

 a

Minimal (limitation of motion, evident at extremes of gaze in one or more directions)

b

Moderate (evident restriction of motion without fixation of position)

 c

Marked (fixation of position of a globe or globes)

5

Corneal involvement (primarily due to lagophthalmos)

0

Absent

a

Minimal (stippling or cornea)

b

Moderate (ulceration)

 c

Marked (clouding, necrosis, perforation)

6

Sight loss (due to optic nerve involvement)

 0

Absent

a

Minimal (disc pallor or choking, or visual field defect; vision 20/20 to 20/60)

 b

Moderate (disc pallor or choking, visual field defect, 20/70 to 20/200)

 c

Marked (blindness, i.e. failure to perceive light; vision less than 20/200)


From Werner SC. Modification of the classification of the eye changes of Graves' disease: recommendations of the Ad Hoc committee of the American Thyroid Association. J Clin Endocrinol Metab 1977; 44:203.

Assessment of Soft Tissue Changes

A complete examination protocol and photographic color atlas for assessment of the soft tissue changes that occur in patients with Graves' ophthalmopathy can be found at http://www.EUGOGO.org.

Eyelid and Periorbital Swelling

Orbital fat or lacrimal gland displacement can cause visible discrete eyelid swelling during any phase of ophthalmopathy (Fig. 23B.3E), whereas acute inflammation with subdermal fluid accumulation or tense skin thickening suggests active ophthalmopathy (Fig. 23B.3F). The edema of the eyelids, especially the lower lids, may be very marked (festoons) (Fig. 23B.3G), and it can persist long-term. Like many soft tissue features, lid swelling is best graded by photographic comparison to assess trend and treatment response (21).

Lid Erythema

Erythema of the eyelids implies active disease.

Conjunctival Erythema

Erythema of the bulbar conjunctivae is associated with active disease (21). Dilated superficial vessels overlying the insertion of the lateral rectus muscle (Fig. 23B.3H) may be especially prominent in patients with active ophthalmopathy (21).

Inflammation of the Caruncle or Plica

The caruncle is the triangular-shaped tissue inside the inner corner of the eyelids rich in sebaceous glands, while the plica is the vertical fold of conjunctiva just lateral to the caruncle. Inflammation in either of these structures suggests active ophthalmopathy (Fig. 23B.3I).

Chemosis

Chemosis denotes conjunctival edema (Fig. 23B.3J), which implies acute orbital congestion, a surrogate marker of retroorbital inflammation.

Superior Limbic Keratoconjunctivitis

This describes inflammation of the upper bulbar and eyelid conjunctiva where they move over one another. It occurs in about 3% of patients with active ophthalmopathy (Fig. 23B.3K) and is associated with upper lid retraction (27).

Palpebral Aperture

Widening of the palpebral fissure is usually due to retraction of both upper and lower eyelids, and is caused by multiple factors (2,21). For the upper lids these include contraction of both the levator and Müller's muscles, proptosis, scarring of the lacrimal fascia, and tightness of the inferior rectus muscle, while lower lid retraction correlates strongly with proptosis (Fig. 23B.3E). Standardized measurements require a consistent technique (distance fixation in a relaxed state, standardized head position, and, if vertical strabismus is present, occlusion of the contralateral eye). The width of the mid-pupil vertical fissure is measured, noting the distance between each lid and the adjacent limbus, plus the presence or absence of the upper lid contour abnormality known as lateral flare (Fig. 23B.3D).

Proptosis (Exophthalmos)

Proptosis is the forward displacement of the eyeball, measured clinically with an exophthalmometer. This is a measure of the position of the cornea relative to a fixed bony point, which is usually the lateral margin of the bony orbit. Normal values depend on race, age, sex, extent of myopia, and the instrument used (21). Accuracy to within two millimeters requires a consistent method with the same instrument and ideally by the same observer (21,28).

Motility

Motility can be assessed either subjectively or objectively (21). Those methods that compare the two eyes (Maddox rod test, Lancaster red-green test, prism cover test, and Hess Lees screen test) have limited use as outcome measures for patients with a bilateral disease such as Graves' ophthalmopathy, but the latter two still guide the use of prisms to improve vision and surgical management. The field of binocular single vision describes the area in which the patient has no double vision and is valuable for assessment, treatment monitoring, and surgical planning. In recent years, uniocular fields of fixation plotted by perimetry have been used as outcome measures (21,29), as they have proved the most reliable method for quantifying the excursion of each eye in different directions of gaze, thus permitting sequential analysis regardless of the muscles involved.

Cornea

Lid retraction causes relative corneal exposure, often with punctate epithelial breakdown that can be highlighted by fluorescein staining. Provided the eyelids can close to cover the cornea there is no real risk of ulceration, but if closure is incomplete, then ulceration and even perforation can occur. This usually happens when there is levator dysfunction with poor lid movement, and an absent Bell's phenomenon (no reflex upward eyeball rotation on attempted lid closure) caused by a tight inferior rectus muscle.

Intraocular Pressure

High intraocular pressure is caused by orbital venous congestion and globe compression by a tight inferior rectus, and increases further during upward gaze. This rarely progresses to true glaucoma.

Visual Function

Optic neuropathy is almost always caused by compression of the optic nerve at the apex of the orbit by grossly swollen muscles (21), but occasionally it is due to nerve stretching by extreme proptosis (30) or prolonged subluxation of the eyeball. Nearly all patients with optic neuropathy have associated dysmotility; the orbit is usually tense (assessed by palpation) and intraocular pressure is relatively high. Corrected visual acuity, color vision (blue discrimination is most sensitive), pupillary responses (swinging light test), and the appearance of the optic disc should be determined. Disc swelling is diagnostic of optic neuropathy in the absence of other causes, but other features are less reliable. Most patients with optic neuropathy have a visual acuity of 20/40 or better, 50% have normal discs, and 75% have bilateral compression with no afferent pupil defect (2,21). Perimetry, contrast sensitivity, and visual-evoked potential recordings help support the diagnosis, but the results of these tests should not be used as a basis for treatment decisions in the absence of clinical evidence of optic neuropathy (21). Perimetry is abnormal in 35% of patients with optic neuropathy, but the results are influenced by normal variations in test results and other pathology such as cataract or glaucoma, and are grossly unreliable if visual acuity is ≤20/80 (31). Confounding pathology such as cataract and age-related macular degeneration affect contrast sensitivity, while thyroid dysfunction and age affect visual-evoked potentials (21,32).

Clinical Evaluation of Disease Activity

The inflammatory soft-tissue changes that imply activity were described earlier. The Clinical Activity Score (CAS) provides a way to quantitate these changes. This score consists of two scores for pain, five for inflammatory features, and three for change in severity of proptosis, diplopia and, visual acuity over two months (Table 23B.3) (33,34). Although partly subjective, it is a simple, useful tool for everyday patient management. A score of 4 or higher signifies active disease.

TABLE 23B.3. CLINICAL ACTIVITY SCORE


Pain

   Painful, oppressive feeling on or behind the globe

   Pain on attempted up-, side-, or down-gaze

Redness

   Redness of the eyelids

   Diffuse redness of the conjunctiva (Fig. 23.3H)

Swelling

   Chemosis (Fig. 23.3J)

   Swollen caruncle (Fig. 23.3I)

   Edema of the eyelid (s) (Fig. 23.3F,23.3G)

   Increase in proptosis of 2 mm or more during 1- to 3-month period

Impaired Function

   Decrease in visual acuity of 1 or more lines on the Snellen chart (using a pinhole) during a 1- to 3-month period

   Decrease of eye movements in any direction equal to or more than 5 degrees during a 1- to 3-month period


One point is given for each sign present. The sum of the points defines the activity score. (From Mourits MP, Koornneef L, Wiersinga WM, et al. Clinical criteria for the assessment of disease activity in Graves' ophthalmopathy: a novel approach. Br J Ophthalmol1989;73:63, with permission of the publisher.)

Quality of Life and Self-Assessment

Patients with newly diagnosed Graves' ophthalmopathy describe impairment in psychological health comparable to patients with other chronic diseases (35), and the impairment persists for several years (36). Approximately half of the patients describe anxiety and depression (37), which are strongly correlated with their altered facial appearance and visual dysfunction, particularly diplopia. A validated disease-specific quality of life questionnaire is available (38) and is an essential outcome measure for clinical therapeutic trials (25).

Diagnostic Imaging

Most patients with Graves' ophthalmopathy do not need to undergo orbital imaging, whether ultrasonography, computed tomography (CT), magnetic resonance imaging (MRI), or octreotide scintigraphy (39). The exceptions include diagnostic uncertainty, in which case CT or MRI is mandatory, particularly for excluding orbital tumors (2); further evaluation of disease activity (A-mode ultrasonography, MRI and octreotide scintigraphy) (34); identifying patients with only expansion of orbital adipose tissue (CT or MRI); planning of surgical decompression or orbital radiation (CT); and identifying patients at high risk for optic neuropathy, in whom either CT or MRI will show “apical crowding” (40). Apical crowding denotes swollen muscles effacing the optic nerve at the orbital apex. If orbital fat also prolapses through the superior orbital fissure, then these combined features are highly sensitive and specific for optic neuropathy (41).

Ultrasonography

B-mode ultrasonography is of limited value because of technical factors that limit its accuracy and also prevent any imaging of the posterior orbit (42). A-mode ultrasonography provides qualitative information, with high muscle reflectivity (hyperechogenicity) implying edema, and therefore active Graves' ophthalmopathy. However, this result does not predict the response to medical treatment, whereas low reflectivity (hypoechogenicity) predicts a poor response (42).

Computed Tomography

Computed tomography (CT) is widely used to evaluate ophthalmopathy in patients with Graves' disease. The large differences in contrast between bone, muscle, and fat allow recognition of proptosis, the dimensions of fat and esepcially muscle, and the risk of optic neuropathy (40). CT is ideal for bone and sinus imaging before decompression, but cannot reliably distinguish closely adjacent muscles, e.g., the levator and superior rectus muscles.

Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is expensive and less widely available than CT. The indications for this test are the same as those for CT, except that it is not appropriate for predecompression bony imaging. Its advantages include lack of radiation, improved detection of optic neuropathy, and the potential for more accurate measurements of retroorbital tissues for treatment monitoring (43). Additionally, more sophisticated tissue characterization allows the potential for qualitative tissue analysis, which may help identify active disease. T2 relaxation times (34,42) and STIR (Short Tau Inversion Recovery) sequences help to identify tissue edema and have some correlation with the Clinical Activity Score, but do not accurately predict the response to treatment (44).

Orbital Scintigraphy

Orbital scintigraphy using radiolabeled somatostatin analogs (45) may be helpful in identifying patients with active ophthalmopathy, however its place in routine management is uncertain.

Biochemistry and Immunology

No serum tests can confirm, refute, or predict Graves' ophthalmopathy. Biochemical assessment of thyroid function is mandatory in all newly presenting patients, and should be repeated regularly to ensure euthyroidism is rapidly achieved and maintained (see Chapter 13). Measurements of serum TSH-receptor antibodies confirm the presence of Graves' disease. Approximately 75% of patients with Graves' ophthalmopathy have high serum concentrations of antithyroid peroxidase antibodies (2).

TREATMENT OF GRAVES' OPHTHALMOPATHY

Choice of Treatment for Thyrotoxicosis

The effect of different treatments for thyrotoxicosis on the natural history of Graves' ophthalmopathy has been a subject of debate, with much of the debate focused on the question of whether radioiodine treatment results in onset or exacerbation of ophthalmopathy (see Chapter 45). In a randomized prospective study, the risk of onset or exacerbation of ophthalmopathy was higher in patients treated with radioiodine, as compared with either antithyroid drug therapy or subtotal thyroidectomy (46). Thyroxine therapy initiated two weeks after radioiodine was given, thus preventing hypothyroidism, reduced the appearance or worsening of ophthalmopathy (47). A further, larger randomized study compared antithyroid drug therapy with radioiodine therapy, with and without the addition of prednisone (48). Radioiodine was associated with a small but significantly higher risk of onset or worsening of ophthalmopathy, as compared with antithyroid drug therapy; there was no increase in risk in those patients treated with radioiodine and prednisone. In a prospective study of patients with thyrotoxicosis who had inactive Graves' ophthalmopathy, radioiodine therapy combined with early thyroxine therapy to prevent hypothyroidism was not associated with onset or exacerbation of ophthalmopathy (49). The adverse effect of radioiodine in patients with ophthalmopathy is probably at least partly due to radioiodine-induced hypothyroidism. However, it is likely that radioiodine therapy itself has a detrimental effect on the eyes during the active phase of the disease. Antithyroid drugs and thyroidectomy are neutral with regards to the course of Graves' ophthalmopathy (46,50), and therefore are the preferred therapy for patients who have active ophthalmopathy.

Indications for Referral to an Ophthalmologist or Specialist Clinic

Most patients who develop Graves' ophthalmopathy present initially to a primary care physician or endocrinologist. Most of them have mild ophthalmopathy, and referral to a specialist is usually unnecessary. A brief history and a basic ophthalmologic examination, which can be rapidly performed by any physician, should identify those patients who would benefit from referral (Table 23B.4).

TABLE 23B.4. INDICATIONS FOR REFERRAL OF PATIENTS WITH GRAVES' OPHTHALMOPATHY TO AN OPHTHALMOLOGIST OR SPECIALIST CLINIC


Inability to close the eyelids (lagophthalmos), leaving the cornea visible*

Corneal ulceration or abscess* (Fig. 23.3l)

Diminished corrected visual acuity*

Awareness of diminished color vision*

Afferent pupillary defect*

Swollen optic disc*

Severe orbital pain

Symptomatic deterioration of preexisting Graves' ophthalmopathy

Troublesome diplopia, particularly affecting forward gaze or reading position or associated with compensatory head tilt (Fig. 23.2A)

Surface symptoms, unresolved after 1 to 2 weeks of regular use of topical lubricants

History of subluxation of the globe

Patients severely distressed by their appearance

Severe periorbital edema (Fig. 23.3B,23.3F,23.3G)

Severe proptosis (Fig. 23.3E)


* Urgent referral required if there is suspicion of optic neuropathy or corneal ulceration.

Sequence and Planning of Treatment

The first priority in patients with active Graves' ophthalmopathy is to ensure that the patient's vision is not threatened because of optic neuropathy or corneal ulceration. The presence of either of these findings mandates urgent treatment, usually with glucocorticoids. For other patients with active ophthalmopathy, the indications for glucocorticoid therapy are less well defined, but may include orbital pain, marked soft tissue involvement, or dysmotility that substantially reduces the patient's quality of life. Once ophthalmopathy becomes inactive, rehabilitative reconstructive surgery should be considered (Table 23B.5).

TABLE 23B.5. PLANNING AND SEQUENCE OF TREATMENTS FOR PATIENTS WITH GRAVES' OPHTHALMOPATHY


Vision threatened by:

Action:

   Optic nerve compression

   Urgent medical or surgical decompression and/or orbital irradiation

   Severe corneal exposure (eyelids cannot cover cornea)

   Urgent immunosuppression to improve closure; may also need urgent surgical decompression or occasionally lid lengthening

   Globe subluxation

   Semi-urgent surgical decompression

Active disease of moderate or worse severity:

Action:

   Soft tissue signs and/or dysmotility

   Consider immunosuppressive treatment

Inactive eye disease with marked proptosis

Action:

   Consider surgical decompression

Inactive disease with important dysmotility

Action:

    Consider strabismus surgery

Inactive disease with important lid problems

Action:

    Consider lid surgery, but first ensure that lid problems will not be addressed by prior decompression or strabismus surgery


Simple Treatments

The corneal exposure that results from lid retraction frequently induces surface-related symptoms, but responds well to topical lubricants, although overuse of preservative-containing preparations should be avoided. Upper-lid retraction may be reduced by botulinum toxin injection (51). Use of sunglasses often helps photophobia and tearing. Sleeping semi-recumbent rather than flat may improve morning tissue congestion. Temporary (fresnel) prisms often relieve diplopia in patients with active ophthalmopathy and may be permanently incorporated into glasses when dysmotility stabilizes. Diplopia unrelieved by prisms requires monocular occlusion, which may enable the patient to continue to drive a car. Patients who have subluxation of the eyes should be taught how to reposition the globe while awaiting decompression surgery (52).

Major Treatments

Glucocorticoids

Glucocorticoids have multiple immunosuppressive actions, including inhibition of leukocyte chemotaxis and cytokine release, reduction in glycosaminoglycan synthesis by orbital fibroblasts, and down-regulation of adhesion molecules (53). Prospective randomized studies have demonstrated a beneficial effect of glucocorticoids, as compared with no treatment, in thyrotoxic patients with ophthalmopathy who were treated with radioiodine (48,54).

The benefit of glucocorticoids in other patients with ophthalmopathy is less well documented; overall, about 60% to 75% of patients treated with glucocorticoids have improved, but all the studies were uncontrolled (53). Glucocorticoids can be administered orally, intravenously, or by retrobulbar injection; the latter route has no advantage and has largely been abandoned. They are usually given orally in high doses (e.g. prednisone, 60 to 80 mg daily) for 2 to 4 weeks, then gradually reduced; disease exacerbation is common during dose reduction. High doses of intravenous glucocorticoids, for example, 500 mg of methylprednisolone given daily for 3 days, may be more effective than oral glucocorticoids, with reported response rates of 73% to 100% (20,55,56). In one study this type of therapy resulted in a favorable clinical response in 88% of patients, as compared with a response rate of 63% in patients given oral glucocorticoid therapy (56), and there were fewer side effects in the intravenous-therapy group than in the oral-therapy group, even though the total dose was higher in the intravenous-therapy group.

Overall, glucocorticoid therapy is most appropriate for patients with active ophthalmopathy in whom sight is threatened; those with substantial dysmotility, especially diplopia affecting primary gaze or reading position; and those with moderate to severe inflammatory features. Glucocorticoid therapy is unlikely to be beneficial in patients whose principal problem is proptosis, and the risk of side effects far outweighs their value in patients with mild ophthalmopathy. Whenever glucocorticoids are given, the duration of therapy should be as short as possible, but attempts to reduce or discontinue glucocorticoid therapy are often followed by exacerbation of ophthalmopathy, and therefore the need for higher doses or reinstitution of therapy. Thus, patients may develop Cushing's syndrome. Osteoporosis prophylaxis should be considered, although in one study repeated pulses of high-dose intravenous glucocorticoid therapy did not accelerate bone loss (56). This therapy, however, can induce rises in serum concentrations of hepatic enzymes, and three deaths from liver failure have been reported in approximately 800 patients with ophthalmopathy treated with high intravenous doses of glucocorticoids (56,57).

Orbital Radiation

Orbital radiation may be of benefit in patients with Graves' ophthalmopathy. The benefit is presumed to be due to a suppressive effect on activated intraorbital lymphocytes and fibroblasts, leading to a decrease in the production of cytokines and glycosaminoglycans by these cells (58). The standard protocol consists of 20 Gy, delivered in 10 fractions over 2 weeks using well-collimated beams generated by a supervoltage linear accelerator (58). Planning includes CT and preparing a plastic mold for head immobilization. A short course of glucocorticoids (10 to 20 mg of prednisone daily) is sometimes given concurrently to avoid transient conjunctivitis. An alternative regimen delivering the same radiation dose over 20 weeks yielded slightly better results (59), however, most efficacy and safety data derive from studies using the standard protocol. The efficacy of radiation was similar to that of glucocorticoids in uncontrolled studies (53), but with a slower onset and peak effect at about 6 months.

There have been two prospective randomized placebo-controlled studies of the efficacy of orbital radiation in patients with moderately severe Graves' ophthalmopathy. In one study, 60% of patients treated with radiation improved, as compared with 31% of those who received sham radiation; the main improvement was in increased motility, especially in upgaze (60). In the other study, patients were randomly assigned to receive radiation to one eye and sham radiation to their other eye (61). There was no difference in the two eyes six months later. Patient selection may account for these discordant findings.

Given the safety of orbital radiation therapy, it should be considered in patients with active Graves' ophthalmopathy (62), particularly those in whom the ophthalmopathy is of recent onset and in whom dysmotility is a problem. The indications for glucocorticoid and radiation therapy are similar, and the two treatments should be viewed as complementary. Orbital radiation is appropriate as monotherapy in patients whose ophthalmopathy is not severe, and those in whom glucocorticoids are contraindicated. With modern equipment the radiation can be focused on the retroorbital tissues, and the radiation dose to surrounding tissues is very low. The estimated risk of fatal cancer following any radiation is 7 per 1000 (63), but no cases have ever been reported in patients with Graves' ophthalmopathy treated with orbital radiation. Nevertheless, it is not recommended for patients younger than 35 years of age, because few data are available for this group. Radiation-induced cataracts are rare in patients treated with radiation generated by a linear accelerator. So also is radiation-induced retinopathy; however, it is more likely in patients with retinal vascular disease or previous exposure to chemotherapeutic drugs (64,65,66).

Combined Therapy with Glucocorticoids and Orbital Radiotherapy

The combination of orbital radiation and glucocorticoids is more effective than either treatment alone (53), and in patients treated with both it may be possible to discontinue the glucocorticoid within three months.

Other Medical Treatments

Numerous other medical treatments have been described (2,53,67), but all should be considered experimental. The combination of cyclosporine and an oral glucocorticoid was marginally better than either alone, but cyclosporine therapy requires close monitoring. Intravenous immunoglobulin was reported to be as effective as glucocorticoids in one small study, but its expense and unknown long-term effects are strong arguments against its use. Octreotide appeared useful in a case-controlled study (68), but not in a randomized prospective study in patients with moderate Graves' ophthalmopathy (69). Plasmapheresis, often combined with a glucocorticoid, was effective in small uncontrolled studies. The combination of two antioxidants (allopurinol and nicotinamide) was beneficial in a small study (70). Pentoxifylline (71), colchicine, methotrexate, bromocriptine, metronidazole, chlorambucil, and cyclophosphamide have been thought to be bene ficial in a few patients, but they have substantial side effects (53); given the variable natural history of Graves' ophthalmopathy, the benefit of any of these drugs may well have been due to chance alone.

Prognosis and Prediction of Response to Medical Treatment

Among patients with thyrotoxicosis caused by Graves' disease who do not have clinically evident ophthalmopathy at the time antithyroid treatment is initiated, the risk of developing ophthalmopathy is approximately 20% in the next 18 months but is negligible later (2). Patients most likely to develop ophthalmopathy are those who have severe thyrotoxicosis (46), a large goiter, high serum TSH-receptor antibody concentrations (1), or persistent thyroid dysfunction (2); those who are smokers (8); and possibly those treated with radioiodine (48,72). Similarly, these same risk factors are associated with severe ophthalmopathy and, by inference, with possible progression of mild ophthalmopathy. In addition, old age and male sex are adverse prognostic factors (5). Based only on clinical signs during a single assessment, progression of ophthalmopathy could be predicted accurately in 69% of eyes using a neural network (73), however, this approach has not been validated prospectively. The small risk of onset or exacerbation of ophthalmopathy associated with radioiodine therapy may be prevented by glucocorticoid therapy (48); there appears to be no risk of radioiodine therapy in patients with inactive ophthalmopathy (49). There is circumstantial evidence that cessation of smoking leads to improvement in ophthalmopathy (74). Additionally, non-smokers have less severe ophthalmopathy of shorter duration, and therapy is more effective, as compared with smokers (48,75). Therefore, smoking cessation should be considered for all patients with Graves' disease.

Glucocorticoid therapy or orbital radiation therapy are more effective in patients with ophthalmopathy of recent onset than in those with chronic ophthalmopathy (53); therefore, the prognosis is better in patients treated promptly. The Clinical Activity Score, measurements of cytokines in serum, ultrasonography, MRI, and octreotide scintigraphy have all been reported to predict response to medical treatment (34,44,45,53). Multivariate logistic regression analysis including clinical, biochemical, and imaging measures was only marginally better than the Clinical Activity Score for predicting response to medical treatment (76). At present, the duration of eye disease and clinical assessment of disease activity are the only practical guides to predicting outcome. In patients treated with high intravenous doses of glucocorticoids, reassessment within a week usually gives an accurate indication as to response, which even at this early stage is predictive of longer-term outcome (2).

Surgical Treatment

Surgical therapy for patients with Graves' ophthalmopathy has expanded in recent years (77), due to the development of safer surgical techniques coupled with higher patient expectations. Three types of surgery are available, orbital decompression, correction of strabismus, and eyelid adjustment. While few patients require all procedures, the order should virtually always be decompression, then strabismus surgery, and then lid surgery. Premorbid photographs and supportive counselling help determine the functional and cosmetic concerns of the patient, and should allow development of a realistic plan, which is especially important when the scope for improvement is limited. Despite this, surgery can contribute very significantly to an improved outcome in patients with Graves' ophthalmopathy.

Orbital Decompression

Orbital decompression entails expansion of the bony boundaries of the orbits, removal of orbital fat, or both, thereby reducing the effects of the expanded orbital fibroadipose and muscle tissue. In patients with active ophthalmopathy, decompression may be indicated urgently to prevent visual loss due to optic neuropathy or corneal exposure. It may also be indicated in patients with active or inactive ophthalmopathy who have subluxation of their eyes. In patients with inactive ophthalmopathy, it may be indicated to improve appearance if eyelid surgery alone is unlikely to result in a satisfactory result.

The choice of decompression procedure depends on the indications for surgery and the desired amount of proptosis reduction, balanced against the risks, particularly of induced dysmotility, which is the most frequent complication (78,79,80,81). Hence, the surgical team should offer a range of procedures differing in approach, and with clear objectives: which walls of the bony orbits are to be opened, and by how much; how much periorbital tissue should be disrupted; and how much retroorbital fat should be removed (79,82,83,84). Four bony surfaces are available for decompression: the medial wall adjacent to the ethmoid sinuses; the floor of the orbit, which is the roof of the maxillary sinus; the anterolateral wall of the orbit adjacent to the temporal fossa above and buccal fat below; and the deep lateral wall comprising most of the greater wing and part of the lesser wing of the sphenoid bone (85). Decompression of the roof of the orbit has little effect on proptosis, induces orbital pulsation, and is virtually obsolete. In patients with optic neuropathy the posteromedial wall should at least be removed (86,87), either by an external or endoscopic approach. For rehabilitation, however, there is controversy as to which walls to remove first (78,80,86), and this debate has been further hampered by the lack of standardized reporting of both pre- and postoperative dys motility (81). Dysmotility occasionally improves after decompression, particularly in patients in whom muscle thickness is not increased but the muscles are stretched from proptosis. Fat decompression is often done in conjunction with bony decompression; performed alone, decompression of fibroadipose tissue rarely worsens dysmotility, but proptosis isn't reduced much (78,88). Other complications of decompression, in addition to dsysmotility, include blindness, orbital cellulitis, cerebrospinal fluid leak, cerebral hematoma, cutaneous sensory loss, sinus obstruction, anosmia, nasolacrimal outflow obstruction, and lid malposition.

Surgery for Strabismus

Strabismus surgery should be performed only when the ocular alignment has been stable for 6 months, although it may be done earlier if the ophthalmopathy is inactive or as a sequel to decompression (89,90). The goal of strabismus surgery is to improve ocular alignment and eliminate any abnormal head posture, and thereby create an optimal binocular field for straight-ahead vision and reading. Preoperative planning should include assessment of occupation, driving status, and hobbies to determine goals for that patient, as it is seldom possible to achieve binocular vision in all directions or accurately correct tiny deviations (91). Strabismus surgery in patients with Graves' ophthalmopathy for the most part involves loosening tight muscles (20,91). Multiple operations may be required (20,92). Most patients achieve useful binocular vision, but 10% have to resort to monocular occlusion. Surgical complications include lid malposition and perforation of the globe.

Eyelid Surgery

Eyelid surgery is rarely indicated in patients with active ophthalmopathy, because decompression is usually sufficient to prevent corneal perforation or globe subluxation. Lid surgery is usually the final stage of rehabilitation. Several procedures may be needed. They include adjustment of upper and lower eyelid positions; cautious removal of excess skin; repositioning of a prolapsed lacrimal gland; and occasionally a small lateral tarsorrhaphy. The latter is obsolete as a stand-alone procedure, as the esthetic outcome is poor. Upper eyelid surgery addresses lid retraction and eyelid swelling, and often improves ocular comfort and tearing as well as appearance, but lid retraction due to a tight inferior rectus muscle should first be addressed by strabismus surgery. Upper lid lengthening techniques abound (93), but spacer material is not required. In contrast, lower eyelid lengthening does require spacer material, ideally autogenous. Cautious removal of excess skin may also be required but should never precede lid lengthening.

Organizational Aspects of Treating Patients with Ophthalmopathy

Multidisciplinary teams comprising physicians and surgeons who are interested and experienced in management of patients with Graves' ophthalmopathy achieve the best outcome. A critical mass of patient referrals is necessary to develop and maintain these skills. For a disease like Graves' ophthalmopathy with a low incidence (4), streamlining of referrals to a few centers with experience and expertise is imperative. The multidisciplinary team should ideally include an endocrinologist and an ophthalmologist, with access to expert orbital surgery, eye muscle surgery, oculoplastic surgery, and radiation therapy. The goals of the team include rapid achievement of euthyroidism and maintenance of the euthyroid state, and selection, timing, and delivery of most appropriate treatment for ophthalmopathy.

REFERENCES

1. Wiersinga WM, Prummel MF. Pathogenesis of Graves' ophthalmopathy-current understanding. J Clin Endocrinol Metab 2001;86:501.

2. Burch HB, Wartofsky L. Graves' ophthalmopathy: current concepts regarding pathogenesis and management. Endocr Rev 1993;14:747.

3. Kendall-Taylor P, Perros P. Clinical presentation of thyroid associated orbitopathy. Thyroid 1998;8:427.

4. Bartley GB, Fatourechi V, Kadrmas EF, et al. The incidence of Graves' ophthalmopathy in Olmsted County, Minnesota. Am J Ophthalmol 1995;120:511.

5. Bahn RS. Pathophysiology of Graves' ophthalmopathy: the cycle of disease. J Clin Endocrinol Metab 2003;88:1939.

6. Pappa A, Lawson JM, Calder V, et al. T cells and fibroblasts in affected extraocular muscles in early and late thyroid associated ophthalmopathy. Br J Ophthalmol 2000;84:517.

7. Cockerham KP, Hidayat AA, Brown HG, et al. Clinicopathologic evaluation of the Mueller muscle in thyroid-associated orbitopathy. Ophthal Plast Reconstr Surg 2002;18:11.

8. Wakelkamp IM, Bakker O, Baldeschi L, et al. TSH-R expression and cytokine profile in orbital tissue of active vs. inactive Graves' ophthalmopathy patients. Clin Endocrinol (Oxf) 2003;58:280.

9. Agretti P, Chiovato L, De Marco G, et al. Real-time PCR provides evidence for thyrotropin receptor mRNA expression in orbital as well as in extraorbital tissues. Eur J Endocrinol 2002;147: 733.

10. Busuttil BE, Frauman AG. Extrathyroidal manifestations of Graves' disease: the thyrotropin receptor is expressed in extraocular, but not cardiac, muscle tissues. J Clin Endocrinol Metab 2001;86:2315.

11. Munsakul N, Bahn RS. Adipogenesis and TSH receptor expression. In: Bahn RS, ed. Thyroid eye disease. Boston: Kluwer Academic Publishers, 2001:37.

12. Starkey KJ, Janezic A, Jones G, et al. Adipose thyrotrophin receptor expression is elevated in Graves' and thyroid eye diseases ex vivo and indicates adipogenesis in progress in vivo. J Mol Endocrinol 2003;30:369.

13. Kaspar M, Archibald C, De BA, et al. Eye muscle antibodies and subtype of thyroid-associated ophthalmopathy. Thyroid 2002;12:187.

14. Gerding MN, van der Meer JW, Broenink M, et al. Association of thyrotrophin receptor antibodies with the clinical features of Graves' ophthalmopathy. Clin Endocrinol (Oxf) 2000;52: 267.

15. Pritchard J, Han R, Horst N, et al. Immunoglobulin activation of T cell chemoattractant expression in fibroblasts from patients with Graves' disease is mediated through the insulin-like growth factor I receptor pathway. J Immunol 2003;170:6348.

16. Vaidya B, Kendall-Taylor P, Pearce SH. The genetics of autoimmune thyroid disease. J Clin Endocrinol Metab 2002;87:5385.

17. Bednarczuk T, Hiromatsu Y, Fukutani T, et al. Association of cytotoxic T-lymphocyte-associated antigen-4 (CTLA-4) gene polymorphism and non-genetic factors with Graves' ophthalmopathy in European and Japanese populations. Eur J Endocrinol 2003;148:13.

18. Kretowski A, Wawrusiewicz N, Mironczuk K, et al. Intercellular adhesion molecule 1 gene polymorphisms in Graves' disease. J Clin Endocrinol Metab 2003;88:4945-.

19. Perros P, Kendall-Taylor P. Natural history of thyroid eye disease. Thyroid 1998;8:423.

20. Kalman R, Mourits MP. Late recurrence of unilateral Graves' orbitopathy on the contralateral side. Am J Ophthalmol 2002;133: 727.

21. Dickinson AJ, Perros P. Controversies in the clinical evaluation of active thyroid-associated orbitopathy: use of a detailed protocol with comparative photographs for objective assessment. Clin Endocrinol (Oxf) 2001;55:283.

22. Rootman J, Chang W, Jones D. Distribution and differential diagnosis of orbital disease. In: Rootman J, ed. Diseases of the orbit a multidisciplinary approach, 2nd ed. Philadelphia: Lippincott Williams and Wilkins, 2003:53.

23. Lacey B, Chang W, Rootman J. Nonthyroid causes of extraocular muscle disease. Surv Ophthalmol 1999;44:187.

24. Bartley GB, Gorman CA. Diagnostic criteria for Graves' ophthalmopathy. Am J Ophthalmol 1995;119:792.

25. Pinchera A, Wiersinga W, Glinoer D, et al. Classification of eye changes of Graves' disease. Thyroid 1992;2:235.

26. Prummel MF, Bakker A, Wiersinga WM, et al. Multi-center study on the characteristics and treatment strategies of patients with Graves' orbitopathy: the first European Group on Graves' Orbitopathy experience. Eur J Endocrinol 2003;148:491.

27. Setty R, Varma D, O'Keeffe M, et al. Superior limbic keratoconjunctivitis (SLK) in Graves' orbitopathy (GO). In: Abstracts of the European Society of Ophthalmic Plastic and Reconstructive Surgery, 2003.

28. Sleep TJ, Manners RM. Interinstrument variability in Hertel type exophthalmometers. Ophthal Plast Reconstr Surg 2002;18: 254.

29. Haggerty H, Richardson S, Mitchell K, et al. A modified method for the reliable assessment of uniocular fields of fixation (UFOF). In: Proceedings of the European Strabismological Association, 2001.

30. Kazim M, Trokel SL, Acaroglu G, et al. Reversal of dysthyroid optic neuropathy following orbital fat decompression. Br J Ophthamol 2000;84:600.

31. Henson DB, Chaudry S, Artes PH, et al. Response variability in the visual field: comparison of optic neuritis, glaucoma, ocular hypertension, and normal eyes. Invest Ophthalmol Vis Sci 2000;41:417.

32. Tamburini G, Tacconi P, Ferrigno P, et al. Visual evoked potentials in hypothyroidism: a long-term evaluation. Electromyogr Clin Neurophysiol 1998;38:201.

33. Mourits MP, Koornneef L, Wiersinga WM, et al. Clinical criteria for the assessment of disease activity in Graves' ophthalmopathy: a novel approach. Br J Ophthalmol 1989;73:639.

34. Mourits MP. Assessment of disease activity. In: Bahn RS, ed. Thyroid eye disease. Boston: Kluwer Academic Publishers, 2001: 185.

35. Gerding MN, Terwee CB, Dekker FW, et al. Quality of life in patients with Graves' ophthalmopathy is markedly decreased: measurement by the Medical Outcomes Study instrument. Thyroid 1997;7:885.

36. Terwee C, Wakelkamp I, Tan S, et al. Long-term effects of Graves' ophthalmopathy on health-related quality of life. Eur J Endocrinol 2002;146:751.

37. Kahaly GJ, Hardt J, Petrak F, et al. Psychosocial factors in subjects with thyroid-associated ophthalmopathy. Thyroid 2002;12: 237.

38. Terwee CB, Gerding MN, Dekker FW, et al. Development of a disease specific quality of life questionnaire for patients with Graves' ophthalmopathy: the GO-QOL. Br J Ophthalmol 1998;82:773.

39. Kahaly GJ. Imaging in thyroid-associated orbitopathy. Eur J Endocrinol 2001;145:107.

40. Giaconi JA, Kazim M, Rho T, et al. CT scan evidence of dysthyroid optic neuropathy. Ophthal Plast Reconstr Surg 2002;18:177.

41. Birchall D, Goodall KL, Noble JL, et al. Graves' ophthalmopathy: intracranial fat prolapse on CT images as an indicator of optic nerve compression. Radiology 1996;200:123.

42. Prummel M, Wiersinga W, Mourits MP. Assessment of disease activity in Graves' ophthalmopathy. In: Prummel M, ed. Recent developments in Graves' ophthalmopathy. London: Kluwer Academic Publishers, 2000:59.

43. Firbank MJ, Harrison RM, Williams ED, et al. Measuring extraocular muscle volume using dynamic contours. Magn Reson Imaging Clin N Am 2001;19:257.

44. Prummel, MF, Gerding MN, Zonneveld FW, et al. The usefulness of quantitative orbital magnetic resonance imaging in Graves' ophthalmopathy. Clin Endocrinol (Oxf) 2001;54:205.

45. Krassas GE. Octreoscan in thyroid-associated ophthalmopathy. Thyroid 2002;12:229.

46. Tallstedt L, Lundell G, Torring O, et al. Occurrence of ophthalmopathy after treatment for Graves' hyperthyroidism. N Engl J Med 1992;326:1733.

47. Tallstedt L, Lundell G, Blomgren H, et al. Does early administration of thyroxine reduce the development of Graves' ophthalmopathy after radioiodine treatment? Eur J Endocrinol 1994;130:494.

48. Bartalena L, Marcocci C, Bogazzi F, et al. Relation between therapy for hyperthyroidism and the course of Graves' ophthalmopathy. N Engl J Med 1998;338:73.

49. Perros P, Neoh C, Frewin S, et al. Prophylactic steroids are unnecessary in patients with thyroid-associated ophthalmopathy receiving radioiodine therapy. Endocrine Abstracts 2003;5:OC36.

50. Marcocci C, Bruno-Bossio G, Manetti L, et al. The course of Graves' ophthalmopathy is not influenced by near total thyroidectomy: a case-control study. Clin Endocrinol (Oxf) 1999;51:503.

51. Uddin JM, Davies PD. Treatment of upper eyelid retraction associated with thyroid eye disease with subconjunctival botulinum toxin injection. Ophthalmology 2002;109:1183.

52. Tse DT. A simple maneuver to reposit a subluxed globe. Arch Ophthalmol 2000;118:410.

53. Bartalena L, Pinchera A, Marcocci C. Management of Graves' ophthalmopathy: reality and perspectives. Endocr Rev 2000;21: 168.

54. Bartalena L, Marcocci C, Bogazzi F, et al. Use of corticosteroids to prevent progression of Graves' ophthalmopathy after radioiodine therapy for hyperthyroidism. N Engl J Med 1989;321: 1349.

55. Kauppinen-Makelin R, Karma A, Leinonen E, et al. High dose intravenous methylprednisolone pulse therapy versus oral prednisone for thyroid-associated ophthalmopathy. Acta Ophthalmol Scand 2002;80:316.

56. Marcocci C, Bartalena L, Tanda ML, et al. Comparison of the effectiveness and tolerability of intravenous or oral glucocorticoids associated with orbital radiotherapy in the management of severe Graves' ophthalmopathy: results of a prospective, single-blind, randomized study. J Clin Endocrinol Metab 2001;86:3562.

57. Marino M, Morabito E, Brunetto MR, et al. Acute and severe liver damage associated with intravenous glucocorticoid pulse therapy in patients with Graves' ophthalmopathy. Thyroid 2004;14:203.

58. Kahaly GJ, Roesler HP, Kutzner J, et al. Radiotherapy for thyroid-associated orbitopathy. Exp Clin Endocrinol Diabetes 1999;107[Suppl 5]:S201.

59. Kahaly GJ, Rosler HP, Pitz S, et al. Low- versus high-dose radiotherapy for Graves' ophthalmopathy: a randomized, single blind trial. J Clin Endocrinol Metab 2000;85:102.

60. Mourits MP, van Kempen-Harteveld ML, Garcia MB, et al. Radiotherapy for Graves' orbitopathy: randomised placebo- controlled study. Lancet 2000;355:1505.

61. Gorman CA, Garrity JA, Fatourechi V, et al. A prospective, randomized, double-blind, placebo-controlled study of orbital radiotherapy for Graves' ophthalmopathy. Ophthalmology 2001;108:1523.

62. Bartalena L, Marcocci C, Gorman CA, et al. Orbital radiotherapy for Graves' ophthalmopathy: useful or useless? Safe or dangerous? J Endocrinol Invest 2003;26:5.

63. Akmansu M, Dirican B, Bora H, et al. The risk of radiation-induced carcinogenesis after external beam radiotherapy of Graves' orbitopathy. Ophthalmic Res 2003;35:150.

64. Kahaly GJ, Gorman CA, Kal HB, et al. Radiotherapy for Graves' ophthalmopathy. In: Prummel MF, ed. Recent developments in Graves' ophthalmopathy. Boston: Kluwer Academic Publishers, 2000:115.

65. Robertson DM, Buettner H, Gorman CA, et al. Retinal microvascular abnormalities in patients treated with external radiation for Graves' ophthalmopathy. Arch Ophthalmol 2003;121:652.

66. Marcocci C, Bartalena L, Rocchi R, et al. Long-term safety of orbital radiotherapy for Graves' ophthalmopathy. J Clin Endocrinol Metab 2003;88:3561.

67. Perros P, Kendall-Taylor P. Medical treatment for thyroid-associated ophthalmopathy. Thyroid 2002;12:241.

68. Krassas GE, Dumas A, Pontikides N, et al. Somatostatin receptor scintigraphy and octreotide treatment in patients with thyroid eye disease. Clin Endocrinol (Oxf) 1995;42:571.

69. Kendall-Taylor P, Dickinson AJ, Vaidya BJ et al. Double blind placebo controlled trial of octreotide LAR in thyroid-associated orbitopathy: clinical outcomes. Thyroid 2003;7:671.

70. Bouzas EA, Karadimas P, Mastorakos G, et al. Antioxidant agents in the treatment of Graves' ophthalmopathy. Am J Ophthalmol 2000;129:618.

71. Balazs C, Kiss E, Vamos A, et al. Beneficial effect of pentoxifylline on thyroid associated ophthalmopathy (TAO): a pilot study. J Clin Endocrinol Metab 1997;82:1999.

72. Bonnema SJ, Bartalena L, Toft AD, et al. Controversies in radioiodine therapy: relation to ophthalmopathy, the possible radioprotective effect of antithyroid drugs, and use in large goitres. Eur J Endocrinol 2002;147:1.

73. Salvi M, Dazzi D, Pellistri I, et al. Classification and prediction of the progression of thyroid-associated ophthalmopathy by an artificial neural network. Ophthalmology 2002;109:1703.

74. Wiersinga WM, Bartalena L. Epidemiology and prevention of Graves' ophthalmopathy. Thyroid 2002;12:855.

75. Eckstein A, Quadbeck B, Mueller G, et al. Impact of smoking on the response to treatment of thyroid associated ophthalmopathy. Br J Ophthalmol 2003;87:773.

76. Gerding MN, ed. Assessment of disease activity in Graves' ophthalmopathy. Amsterdam, NL: L. van der Velde BV, 1999.

77. Kalman R, Mourits MP, van der Pol JP, et al. Coronal approach for rehabilitative orbital decompression in Graves' ophthalmopathy. Br J Ophthalmol 1997;81;41.

78. Goldberg RA. The evolving paradigm of orbital decompression surgery. Arch Ophthalmol 1998;116:95.

79. Metson R. Reduction of diplopia following endoscopic orbital decompression: the orbital sling technique. Laryngoscope 2002;112:1753.

80. Kacker A, Kazim M, Murphy M, et al. “Balanced” orbital decompression for severe Graves' orbitopathy: technique with treatment algorithm. Otolaryngol Head Neck Surg 2003;128: 228.

81. Paridaens D, Hans K, van Buitenen S, et al. The incidence of diplopia following coronal and translid orbital decompression in Graves' orbitopathy. Eye 1998;12:800.

82. Seiff SR, Tovilla JL, Carter SR et al. Modified orbital decompression for dysthyroid orbitopathy. Ophthal Plast Reconstr Surg 2000;16:62.

83. Abramoff MD, Kalman R, de Graaf MEL, et al. Rectus extraocular muscle paths and decompression surgery for Graves' orbitopathy: mechanism of motility disturbances. Invest Ophthalmol Vis Sci 2002;43:300.

84. Kikkawa DO, Pornpanich K, Cruz RC Jr, et al. Graded orbital decompression based on severity of proptosis. Ophthalmology 2002;109:1219.

85. Goldberg RA, Perry JD, Hortaleza V, et al. Strabismus after balanced medial plus lateral wall versus lateral wall only orbital decompression for dysthyroid orbitopathy. Ophthal Plast Reconstr Surg 2000;16:271.

86. Schaefer SD, Soliemanzadeh P, Della Rocca DA, et al. Endoscopic and transconjunctival orbital decompression for thyroid related orbital apex compression. Laryngoscope 2003;113:508.

87. Kazim M, Trokel SL, Acaroglu G, et al. Reversal of dysthyroid optic neuropathy following orbital fat decompression. Br J Ophthamol 2000;84:600.

88. Olivari N. Thyroid-associated orbitopathy: transpalpebral decompression by removal of intraorbital fat. Experience with 1362 orbits in 697 patients over 13 years. Exp Clin Endocrinol Diabetes 1999;107[Suppl 5]:S208.

89. Yolar M, Oguz V, Pazarli H, et al. Early surgery for dysthyroid orbitomyopathy based on magnetic resonance imaging findings. J Pediatr Ophthalmol Strabismus 2002;39:336.

90. Bradley EA, Bartley GB, Garrity JA. Surgical management of Graves' ophthalmopathy. In: Bahn RS, ed. Thyroid eye disease. Boston: Kluwer Academic Publishers, 2001:219.

91. Esser J, Eckstein A. Ocular muscle and eyelid surgery in thyroid-associated orbitopathy. Exp Clin Endocrinol Diabetes 1999;107 [Suppl 5]:S214.

92. Prendiville P, Chopra M, Gauderman WJ, et al. The role of restricted motility in determining outcomes for vertical strabismus surgery in Graves' ophthalmology. Ophthalmology 2000;107: 545.

93. Mourits MP, Sasim IV. A single technique to correct various degrees of upper lid retraction in patients with Graves' orbitopathy. Br J Ophthalmol 1999;83:81.