Historical Background

Around 1786, Caleb Hillier Parry described 13 patients with goiter and tachycardia, one of whom had ocular protrusion.⁴ These observations did not appear in the medical literature until 1825 when Parry’s son published the work within a collection of his father’s medical writings. The first modern description of the clinical triad was published in 1835 by the Dublin surgeon Robert James Graves who described three women with goiter, violent palpitations, and bilateral ocular protrusion.⁵ Subsequent publications in 1840 and 1844 by Karl Adolph von Basedow described the same entity with more emphasis on the ocular component. Although the eponym Graves disease is popular in North America, von Basedow disease (and the Merseburger triad) is the favored term in Europe.

At the start of the twentieth century, the cause of Graves hyperthyroidism and TED was thought to be oversecretion of thyroid-stimulating hormone (TSH) by the pituitary gland. Consequently, irradiation of the pituitary gland was advocated for treatment of both conditions.⁶ Early animal studies identified an “exophthalmos-producing substance” purified from fractions of γ-globulin from patients with TED. This substance induced proptosis in a certain species of fish, and in later studies, proptosis was induced in guinea pigs or mice following injection with TSH or pituitary extract. In the mid-1950s, Adams et al. identified a substance found in the sera of most patients with GD, termed long-acting thyroid stimulator (LATS), which they believed to be the etiologic agent in both GD and TED.⁷ We know today that LATS is essentially immunoglobulin containing stimulatory anti-TSH-receptor antibodies, which are, indeed, the causative agent in Graves hyperthyroidism and also likely play a central role in TED development (see Pathogenesis).

The first descriptions of the natural history of TED were published by Rundle et al. between 1944 and 1960.³ By carefully examining and measuring various clinical parameters over time in a cohort of patients with untreated TED, they determined that the disease generally increases in severity following its inception until it reaches a dynamic plateau phase, with the subsequent static phase resulting in incomplete resolution of symptoms. He noted as well that the length of the dynamic phase varies widely from patient to patient as does the degree of severity reached before the static phase. These observations are still valid today, as understanding the phase of disease carries important therapeutic implications (see Management).

Although the majority of patients with TED have a history of Graves hyperthyroidism, others are either euthyroid or hypothyroid at presentation. Eye involvement is apparent in 30% to 50% of patients with Graves hyperthyroidism, depending on diagnostic criteria.⁸ TED symptoms and signs begin at approximately the same time that hyperthyroidism is diagnosed in almost a third of patients with TED; TED onset precedes the diagnosis of hyperthyroidism in 7.5% of patients and follows it in 63%.⁹ The dermal and bony manifestations of GD, dermopathy, and acropachy are relatively rare (seen in 4% and 1% of patients with TED, respectively) and almost never develop in patients with the indolent, mild form of the orbitopathy.

Epidemiology

The annual age-adjusted incidence of TED in the United States is 16 in 100,000 women and 3 in 100,000 men.¹⁷ The incidence peaks in a bimodal age distribution in women at 40 to 44 years of age and 60 to 64 years, and in men at 45 to 49 years of age and 65 to 69 years. Among incident cases in a large epidemiologic cohort study, TED affected females six times more frequently compared with males, which is similar to the female/male ratio for GD.¹⁸ Among patients diagnosed with TED, approximately 90% have Graves hyperthyroidism, 1% has primary hypothyroidism, 3% have HT with hypothyroidism, and 5% are euthyroid.

Epidemiologic, sibling, and twin studies have identified the major histocompatibility complex class II to be an important genetic factor involved in the development of GD. More recently, other immune regulatory or thyroid-specific genes have been implicated, including cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), PTPN22, FOXP3, CD40, CD25, thyrotropin receptor (TSHR), and thyroglobulin.¹⁹ However, no consistent genetic associations have been found to be more prevalent in patients with GD and TED than in those with GD alone.²⁰ Because TED risk is not accounted for by genetic susceptibility alone, it is likely that environmental factors, acting through epigenetic modifications, are important factors in disease development.²¹

Smoking has been identified as an important risk factor for TED in a multitude of studies.²² In a recent prospective trial of patients with newly diagnosed Graves hyperthyroidism and treated with either radioactive iodine or antithyroid drugs, the odds ratio among smokers versus nonsmokers was between 5.2 : 1 and 9.8 : 1, depending on criteria for TED diagnosis.²³ Overall, more that 40% of the smokers either developed TED or experienced deterioration in the disease, a rate almost double that seen in nonsmokers. Other studies have shown that the TED risk relative to active smoking is proportional to the number of cigarettes smoked per day and that former smokers have significantly lower risk than current smokers.²⁴ In addition, smokers experience less benefit from therapy aimed at improving their eye disease.²⁵ Smoking cessation appears to reduce the risk of disease development and its severity measurably and should be actively encouraged. Even if smoking cessation cannot be accomplished, meaningful reduction in smoking may be of benefit. Although mechanisms underlying the association between smoking and TED worsening are unclear, contributors may include orbital hypoxia or the free radicals found in tobacco smoke, both of which promote orbital fibroblast proliferation.²⁶ Further, smokers have lower levels of interleukin-1 receptor antagonists compared with nonsmokers, which could adversely affect the orbital disease process.²⁷ Data to define the risk related to e-cigarettes in TED are not available.

Pathogenesis

The clinical symptoms and signs of TED can be explained, in a mechanical sense, by the enlargement of the orbital soft tissues characteristically found within the fixed volume of the bony orbit.¹ The expanded orbital tissues displace the globe forward and additionally impede venous outflow from the orbit. Simultaneously, mononuclear cells and resident macrophages infiltrating the orbit secrete cytokines and other mediators of inflammation. These orbital tissue changes result in proptosis, periorbital edema, ocular pain, conjunctival and lid erythema and chemosis, varying from patient to patient. Although the majority of afflicted individuals exhibit enlargement of both orbital fat and extraocular muscles, either of these tissues may be enlarged in apparent isolation.³⁶ The expansion of the orbital adipose tissues is thought to be caused by the development of new fat cells from orbital fibroblast precursors, rather than enlargement of existing adipocytes, as well as by the local accumulation of hyaluronan, a hydrophilic glycosaminoglycan made by orbital fibroblasts.³⁷ The extraocular muscle bodies are enlarged in early, active stages of the disease, whereas the muscle cells themselves are intact and widely separated by locally secreted hyaluronan.³⁸ However, as the disease becomes inactive, the resolving inflammatory process within the muscles may result in fibrosis and strabismus.

The hyperthyroidism of GD is caused by autoantibodies directed against the TSHR on thyroid follicular cells.³⁹ The ocular component of GD is similarly thought to stem from an autoimmune process directed against TSHR. However, the orbital fibroblast is the TSHR-expressing cell against which the autoimmune process is targeted in TED.¹ In addition, although TSHR autoantibodies mediate hyperthyroidism, both cellular and humoral immunity are likely involved in initiating and propagating the orbital disease process. Activation of helper T cells recognizing TSHR peptides presented by orbital fibroblasts or resident antigen-presenting cells leads to the local secretion of inflammatory cytokines and chemokines. Further, ligation of TSHR on orbital fibroblasts by circulating autoantibodies results in increased hyaluronan production and enhanced adipogenesis within the orbital fibroblast population.^40,41 The ensuing connective tissue remodeling leads to varying degrees of extraocular muscle enlargement and orbital fat expansion. In addition, recent studies have described a subset of orbital fibroblasts that express CD34, are bone marrow derived, and reach the orbit and other sites of inflammation via the circulation.⁴² As these cells express particularly high levels of TSHR and are capable of producing copious cytokines and chemokines, they may represent a population of orbital fibroblasts of primary importance in the disease. In addition to TSHR, orbital fibroblasts from patients with TED express high levels of insulin-like growth factor 1R (IGF-1R).⁴³ Early studies suggested that these receptors may engage in cross-talk induced by TSHR ligation to synergistically enhance TSHR signaling, hyaluronan production, adipogenesis, and the secretion of inflammatory mediators.⁴⁴ No convincing evidence suggests that circulating antibodies directed against IGF-1R are specific to GD. It is possible, however, that physiologic IGF-1 produced by cells within the orbit in TED augment downstream TSHR signaling by binding to IGF-1R on these cells (Fig. 12.4).

TED is well known to be the most common cause of both unilateral and bilateral proptosis in the adult population.⁴⁵ No single clinical finding or laboratory test is diagnostic of TED. However, the diagnosis can often be made on the basis of a careful history, physical examination, and measurement of TSH and free thyroxine levels. Although the concordance of hyperthyroidism, goiter, and bilateral proptosis in a 50-year-old woman is easily diagnosed, milder or less typical cases may be challenging to diagnose. It is, therefore, helpful to consider in each patient three elements that aid in establishing the diagnosis: (1) typical clinical features, (2) current or past history of Graves hyperthyroidism, and (3) typical radiographic findings. The diagnosis can be reliably established when at least two of the three of these elements are present. Typical clinical features of TED include unilateral or bilateral upper eyelid retraction, proptosis either unilateral or bilateral (as evidenced by comparison to archival photographs), and extraocular motility restriction in a pattern consistent with TED (i.e., limitation of elevation or abduction). If neither current nor past hyperthyroidism can be identified, the presence of either TSHR-binding or TSHR-stimulating antibodies in a patient with a clinical presentation compatible with TED is suggestive of the diagnosis.¹³ In contrast, TPO antibodies do not support the diagnosis, as the prevalence of anti-TPO antibodies in the general population is high. The differential diagnosis is more extensive in a euthyroid patient with unilateral proptosis or other asymmetric findings. In this instance, computed tomography (CT) or magnetic resonance imaging (MRI) of the orbits may identify an orbital mass lesion, idiopathic orbital inflammation (pseudotumor), an infiltrative process, or other orbital or systemic pathology. If the imaging is compatible with TED, the finding of elevated antibodies directed against the TSHR is helpful in making the diagnosis in a euthyroid patient.¹³ Although the absence of elevated TSHR autoantibodies does not rule out the diagnosis of TED, it necessitates further evaluation and/or observation over time. A full discussion of TSHR antibody testing and thyroid function testing can be found in Chapter 4.

Imaging Studies

Inconstancy or inadequacy of the first two diagnostic criteria warrants further investigation with imaging. In addition, cases in which the initial evaluation is consistent with TED but diverges from that diagnosis over the course of time similarly deserve investigative imaging to exclude other confounding or concurrent diseases. The radiographic hallmarks of TED have been well described and include fusiform enlargement of the rectus muscles with relative sparing of the tendinous insertions (Fig 12.2A). The most commonly enlarged rectus muscles are, in descending order, superior rectus/levator complex, inferior rectus, medial rectus, and lateral rectus. Although rare, both the superior and inferior oblique muscles can be enlarged in cases of TED. It is important to examine the superior rectus/levator complex carefully. Although it is, in fact, the most commonly enlarged radiographically, accounting for the equally common upper eyelid retraction, the magnitude of enlargement is small in comparison with the other rectus muscles and is often overlooked in formal radiologic reviews⁴⁶ (Fig. 12.5). Diagnostically, CT or MRI are of equal value, and since contrast is not required, considerations of surgical planning, radiation exposure, claustrophobia, and the adequacy of a local MRI facility should guide the study choice. For surgical planning, CT is preferred, as the bony walls are more clearly defined for possible decompression.

Diagnostic Dilemmas

Despite the general utility of this approach described above, diagnosis can be difficult in a small subset of outliers in which the imaging results may be inconclusive or when there is coexisting disease.

Pediatric TED may be difficult to diagnose. In this setting, the clinical manifestations are often limited to progressive proptosis. The typical inflammatory signs and symptoms of the older phenotypes of the disease are strikingly absent⁴⁵ (Fig. 12.6). Not surprisingly, imaging also fails to be helpful diagnostically as the typical fusiform enlargement of the rectus muscles is absent. Proptosis in children results solely from orbital fat expansion, which is inconspicuous when imaged by CT or MRI, as there is no algorithm for assessing pathologic fat volume expansion. Similarly, very-early-phase TED can present clinically with nonspecific manifestations of inflammation (i.e., conjunctival or lid injection and ocular surface irritation), all of which are routinely seen in more common ocular conditions, including viral or allergic conjunctivitis. Interpretation of the orbital imaging can also be challenging, even when the extraocular muscles are enlarged, as similar enlargement may result from a cavernous sinus fistula, metastatic disease, lymphoma, or prior strabismus surgery. In these cases, a combined analysis of the clinical and radiographic features are required to establish the correct diagnosis. Moreover, the diagnostic picture can be further complicated by the coincidence of TED and other orbital diseases. These have included optic nerve meningioma, orbital tumors, ocular MG, primary open angle glaucoma, and retinal disorders.

Differential Diagnosis

Several attempts have been made to classify TED during the past 60 years. Each has been developed to address a particular clinical question but have failed to serve all clinical and research requirements. First among them is NOSPECS, described by Sydney Werner.⁴⁷ NOSPECS was initially purposed as a mnemonic to describe the clinical features of TED (N: no signs or symptoms, O: only signs, S: Soft tissue signs and symptoms, P: proptosis, E: extraocular muscle involvement, C: corneal involvement, S: sight loss) (Table 12.1). Although the features were listed in chronological order in which they appear clinically, not all patients suffer from all of the clinical manifestations, nor does the order of their appearance match the NOSPECS list. In its final version, each of signs was further subclassified by severity, from 0 to 3. Although of help to those learning about TED or to describe an individual case clinically, attempts to use it in clinical research studies failed. Unlike cancer grading systems (i.e., tumor–node–metastasis (TNM)), in which a higher grade correlates with a more advanced case, the NOPSPECS system numerically equated a corneal abrasion with mild optic neuropathy. Moreover, the NOSPECS system fails to define the stage of orbitopathy (active or quiescent).

Stay updated, free articles. Join our Telegram channel

Join