Graves' Orbitopathy. Группа авторов

Incidence and Prevalence of Graves’ Orbitopathy?

Graves’ orbitopathy (GO) is a relatively rare disease. There are a few studies on its incidence, but prevalence data are lacking and can only be estimated. The incidence of GO in the general population of Olmsted County in Minnesota, USA, has been reported as 16 women and 3 men per 100,000 population per year [1]. Based on the incidence of Graves’ hyperthyroidism in a Swedish population, the incidence of GO was 42/100,000/year (32 and 10 per 100,000 per year for mild and moderate-to-severe GO, respectively) [2]. About two thirds of all GO patients have just mild GO. The incidence of moderate-to-severe GO will thus be much lower. In a population-based study in Denmark performed between 1992 and 2009, the incidence of moderate-to-severe GO was 1.61/100,000/year (2.67 in women, 0.54 in men); it was similar before and after mandatory salt iodization [3]. The incidence peaked in the age group of 40–60 years. The overall prevalence of GO is estimated to be about 90/100,000 population (60/100,000 for mild and 30/100,000 for moderate-to-severe GO) [4].

The occurrence of moderate-to-severe GO (including dysthyroid optic neuropathy) in patients with Graves’ hyperthyroidism is about 5% (4.9% in population-based surveys in Sweden and Denmark, and 6.1% in a single-centre study from Italy) [2–5]. It means that 1 out of 20 patients with Graves’ hyperthyroidism will have or develop significant GO. At diagnosis of Graves’ hyperthyroidism, the majority of patients (74%) has no GO, and in this group only 3% will develop moderate-to-severe GO during an 18-month follow-up (when treated with antithyroid drugs) [5]. Mild GO is present in 20% of patients when diagnosed with Graves’ hyperthyroidism, and, most interestingly in this group after 18 months of follow-up, mild GO is still present in 40% but GO has disappeared in 55% (Fig. 1) [5].

Fig. 1. Prevalence of Graves’ orbitopathy (GO) in 346 consecutive patients with newly diagnosed Graves’ hyperthyroidism, who were followed up for 18 months during treatment with antithyroid drugs. a Total prevalence of GO. b No GO at baseline (n = 194). c Mild GO at baseline (n = 43). Modified from Tanda et al. [5].

Has the Occurrence of Graves’ Orbitopathy Changed over the Last Few Decades?

Several studies indeed suggest that GO occurs less frequently and that GO is less severe now than a few decades ago. In a retrospective study from the UK, the first consecutive 100 patients presenting with the diagnosis of Graves’ disease at the beginning of each decade between 1960 and 1990 were examined. The proportion of GO in these patients with Graves’ disease declined from 57% in 1960 to 35% in 1990; the proportion of patients with severe GO (diplopia, optic nerve compression) also fell from 30.4 to 20.7% [6]. A questionnaire survey among members of the European Thyroid Association, published in 1998, reports that 43% of respondents thought GO was decreasing in frequency, 42% thought it unchanged, and 12% thought it to be increasing over the last 10 years [7]. In this respect it is noteworthy that all respondents from Hungary and Poland in this survey, where the proportion of smokers in the general population had increased since the fall of the Wall in 1989, indicated an increased incidence of GO. Lastly, a recent EUGOGO study compared characteristics of GO patients referred to EUGOGO centres over a 4-month period in 2012 (n = 269) with those referred over the same 4-month period in 2000 (n = 152) [8]. Smoking rates were 40% in both groups. Mild GO (60.5% in 2012 vs. 41.2% in 2000, p < 0.01) and inactive GO (63.2% in 2012 vs. 39.9% in 2000, p < 0.01) were more prevalent in 2012, suggesting a shift to less severe and less active GO in the first decade of the 21st century (Fig. 2). The secular trend to a lower incidence of GO and to less severe GO might be causally related to a decline in the prevalence of smokers, to an earlier diagnosis and treatment of Graves’ hyperthyroidism (facilitated by the introduction of sensitive TSH assays in the 1980s), and to a prudent use of ¹³¹I therapy (which carries a risk of about 15% for de novo development or worsening of GO).

Fig. 2. Characteristics of Graves’ orbitopathy (GO) patients referred to EUGOGO centres in 2012 (grey columns) and in 2000 (white columns), indicating a shift to less active and less severe GO. a GO activity. b GO severity. Reproduced with permission from Perros et al. [8].

Is the Age and Sex Distribution of Graves’ Orbitopathy Similar to That of Graves’ Hyperthyroidism?

Patients with Graves’ hyperthyroidism and GO are older than those with Graves’ hyperthyroidism without GO (mean age 46.4 and 40.0 years, respectively) [7–9]. The mean age of 152 GO patients referred to EUGOGO centres was 49 years [10].

      GO, like Graves’ hyperthyroidism, is more common in women than men. The female-to-male ratio was 9.3 in patients with mild orbitopathy, 3.2 in those with moderate orbitopathy, and 1.4 with severe orbitopathy [1–4].

      A sex-related difference in the severity of GO has been noted, with men comprising a relatively greater proportion of cases of severe orbitopathy. Eye disease tends to be more severe in older patients and in men. The reason for this effect of gender is not clear but the higher prevalence of smoking among men likely plays a role.

Are There Ethnic Differences?

The prevalence of GO among patients with Graves’ disease seems lower in Asians (7.5%) than in Caucasians (34%) [11]. In contrast, the prevalence of GO among 167 consecutive multiethnic Malaysian patients with Graves’ disease was 34.5%, not different from that in Caucasians [12]. Exophthalmos and lower lid retraction were the most common eye signs in this population. Corneal erosion secondary to acquired epiblepharon is a common sign in East Asian patients with GO. Orientals have a rounder and shallow orbit, black people have a rectangular orbit, and white people are in between [13]. These differences may explain the variation in exophthalmic readings between different ethnic groups. The upper normal limit of proptosis values (defined as 2 standard deviations above