Description of the condition
Epidemiology and pathogenesis of pediatric Graves' disease
Graves' disease (GD) is an autoimmune disease caused by thyroid‐stimulating hormone (TSH) receptor antibodies, which stimulate the gland to synthesize and secrete excess amounts of thyroid hormones. GD is characterized by thyrotoxiema, developed as a result of a complex interaction between predisposing genes and environmental triggers. Most children with thyrotoxicosis suffer from GD accounting for 10% to 15% of all childhood thyroid diseases (Zimmerman 1998). Pediatric GD has a peak incidence at 11to 15 years of age. The female to male preponderance is approximately 5:1. Graves' disease can be familial and associated with other autoimmune diseases (Reid 2005). Most pediatric GD patients have a positive family history of some type of autoimmune thyroid disease. Graves' ophthalmopathy (GO) is an organ specific autoimmune process strongly linked to Graves' hyperthyroidism. GO occurs in 33% of patients with pediatric GD; its prevalence is higher in countries with a higher prevalence of smoking among teenagers (Krassas 2005).GO varies in different population studies and is age‐related. Pathogenesis of GO is not yet fully understood.

Clinical manifestations
The classic clinical manifestations of pediatric GD include nervousness, emotional lability, fatigue, weight loss, restless sleep, and rapid heart beat, all of which result in deteriorating school performance. The signs of pediatric GD include the diffuse goiter, tachycardia, exophthalmos and pretibial myoxedema. However, symptoms and signs of pediatric GD are typically insidious in onset and sometimes confusing, even mimicking hypothyroidism (Nordyke 1988).
GO has a variable clinical presentation such as eyelid retraction, proptosis, periorbital edema, chemosis, and disturbances of ocular motility. In some cases, GO may progress to visual loss as a result of exposure to keratopathy or compressive optic neuropathy (Chan 2002). Fortunately, the severity of childhood GO appears to be less than that of adulthood GO, and more likely to remit completely, presumably explained by the lower prevalence of smoking in children. Therefore, most patients do not require specific therapy (Chan 2002; Durairaj 2006; Krassas 2005; Gruters 1999).

Diagnosis
Diagnosis of GD was based on classic clinical manifestations, common clinical signs, elevated free thyroxine (FT4), free tri‐iodothyronine (FT3), suppressed plasma TSH concentration, and positive immunological markers. Laboratory tests including the assessment of serum total or free T4 or T3, and TSH can confirm the presence of hyperthyroidism. TSH receptor antibodies are the most frequently used immunological markers for the diagnosis of Graves' disease (78%) (Escobar 1998). High thyroperioxidase (TPO) and thyroglobulin (Tg) antibody titers are often observed in pediatric GD (Lavard 2000). Nonspecific laboratory findings can occur in Graves' disease, including anemia, granulocytosis, lymphocytosis, hypercalcemia, transaminase elevations, and alkaline phosphatase elevations.

Description of the intervention
Once the diagnosis of pediatric GD is established, therapy should be adopted to control hyperthyroidism to avoid the untoward effects on growth and pubertal development. Three treatment modalities are available for pediatric GD: antithyroid drugs (ATD), surgery and radioiodine (Fisher 1994; LeFranchi 1991). The goal of therapy for pediactric GD is to correct the hypermetabolic state with the fewest side effects and the lowest incidence of hypothyroidism. Up to date, the optimal therapy remains controversial.

Radioiodine ( ¹³¹ I)
Radioiodine is a g,b‐emitting radionuclide with a physical half‐life of 8.1 days, a principal g‐ray of 364 keV, and a principal b‐particle with a maximum energy of 0.61 Mev, an average energy of 0.192 Mev, and a range in tissue of 0.8 mm. At present, radioiodine is the most commonly used treatment for adult GD.
Radioiodine was introduced for the treatment of GD more than 50 years ago (Chapman 1983). While a number of studies have indicated a dose‐response effect, no clear threshold dose has been identified. Early but not late hypothyroidism is possibly associated with radioiodine dosage. A gland‐specific dosage (50 to 200 mCi or 1.85 to 7.4 MBq radioiodine per gram of thyroid tissue) based on the estimated weight of the gland and the 24‐hour uptake may allow a lower dosage and result in a lower incidence of hypothyroidism but may have a higher recurrence rate (Harper 2003). A gland‐specific dosage is mostly used in China. A higher‐dose ablative therapy (13.8 to 15.6 mCi) is an effective therapy in nearly all children with Graves' disease (Nebesio 2002). Early hypothyroidism may result from this regimen and the children have to be closely monitored. Two studies (Weetman 2000; Allahabadia 2001) have shown that the eventual incidence of hypothyroidism is comparable regardless of the radioactive iodine dosage. Therefore, the responses to treatment as related to the iodine‐131 dose in children are not well‐defined.

Adverse effects of the intervention
Historically, radioiodine was associated with a considerable increase in the risk of thyroid cancer in young children exposed to external radiation. Radioiodine has traditionally been used if major side effects were experienced or if the hyperthyroidism did not remit after several years of ATD treatment. There is now a tendency to advocate radioiodine as a choice of treatment also for children (Clark 1995; Franklyn 1992; Leu 2003). Radioiodine has been increasingly used in some Canadian centres as a first‐line therapy for adolescents and for patients who have trouble adhering to the medication schedule (Ward 1999). In a recent review on the experience in 587 American children aged between one and 18 years, no serious complications have been observed during a follow‐up period of up to 23 years (Clark 1995). Therefore, radioiodine is considered as a safe and effective therapy for pediatric GD.
Side effects of radioiodine include vomiting and radiation‐induced thyroiditis as well as the development of thyroid nodules (Clark 1995). In the pooled analysis of seven studies, the risk of developing thyroid carcinoma after irradiation below the age of five years is twofold higher than in children treated with irradiation between five and nine years of age, and fivefold higher than in children treated between 10 and 14 years of age (Ron 1995). However, no increase in the overall cancer mortality after ¹³¹I‐therapy compared with standard US mortality rates was observed over a mean follow‐up period of 21 years (Ron 1998). In general, no significant increased risk of thyroid malignancy has been reported so far after radioiodine treatment for approximately 1000 children with Graves' disease (Clark 1995; Freitas 1979; Hamburger 1985; Hayek 1970; Kogut 1965; Safa 1975; Starr 1969; Zimmerman 1998). The duration of follow‐up in these studies ranged from less than five years to 15 years, with only some children followed up for more than 20 years. Studies of the offspring born to patients who received radioiodine for childhood GD revealed a 3% incidence of congenital anomalies, similar to the general population. Furthermore, there are no reports of an increase in miscarriage rate or fetal loss because of genetic abnormalities (Holm 1991). Permanent hypothyroidism occurred in 60 to 70% of the patients, leading to the same long‐term problems after subtotal or total thyroidectomy. The appearance or worsening of ophthalmopathy following radioiodine remains controversial. There is very limited experience with the radioiodine treatment of very young children (Gruters 1998).

Other interventions to treat pediatric Graves' diesease
Antithyroid drugs (ATD)
Methimazole and propylthiouracil (PTU) are mostly used agents available for GD. The starting dosage of methimazole is 15 to 30 mg per day (Cooper 2005). Methimazole is adjusted according to clinical status and monthly free T4 or free T3 levels, toward an eventual euthyroid (i.e., normal T3 and T4 levels) maintenance dosage of 5 to 10 mg per day (Ginsberg 2003; Woeber 2000). The starting dosage of PTU is 100 mg three times per day with a maintenance dosage of 100 to 200 mg daily (Cooper 2005). In most cases, a dosage of 80 to 320 mg per day is sufficient. Beta‐blockers are mostly used as treatment adjuncts to offer prompt relief of the adrenergic symptoms of hyperthyroidism such as tremor, palpitations, heat intolerance, and nervousness. Calcium channel blockers such as diltiazem can be used to reduce heart rate in patients who cannot tolerate beta‐blockers (Ginsberg 2003). There is no clear evidence in favour of giving thyroid hormone supplementation following the initial treatment of GD with antithyroid medication (Abraham 2005; Ward 1999).
Effects of dose, regimen and duration of ATD therapy for Graves' disease on the recurrence of hyperthyroidism, course of ophthalmopathy and adverse effects have been widely studied. The comparison of a block‐replace regimen (where a higher dose of antithyroid drug is used with a replacement dose of thyroid hormone) with a titration regimen (where the antithyroid drug dose is reduced by titrating treatment against thyroid hormone concentrations) suggested that there was no significant difference between the regimens for relapse of hyperthyroidism (relative risk (RR) 0.93, 95% confidence interval (CI) 0.84 to 1.03) (Allannic 1990; Allannic 1991; Garcia‐Mayor 1992; Maugendre 1999; Weetman 1994). Remission rates vary with the length of treatment, but rates of 60 percent have been reported when therapy is continued for two years (Harper 2003). A longer period of methimazole treatment was needed to normalize serum thyroid hormone levels and to restore normal thyroidal triiodothyronine suppressibility in aged patients (Yamada 1994). A systemic review by Abraham indicated that the titration regimen appeared as effective as the block‐replace regimen, and was associated with fewer adverse effects. Limited evidence suggested 12 to 18 months of antithyroid drug treatment should be used (Abraham 2005).
Many pediatric endocrinologists currently recommend ATDs as a first‐line treatment (Bergman 2001) because ATD treatment frequently results in subsequent remission (Cheetham 1998). However, only 20% to 30% of pubertal and 15% of prepubertal individuals treated with ATD as the first‐line treatment will experience long‐term remission (Hamburger 1985; Lazar 2000; Rivkees 1998; Shulman 1997). Additionally, ATD medications are associated with side effects and a high relapse rate. Serious and fatal side effects of ATDs like liver toxicity leading to organ failure have been observed in single cases (Williams 1997). ATD medications also need frequent monitoring of blood cells with a high risk for recurrence of GD due to poor compliance during adolescence. Long‐term quality of life assessments as well as follow‐up investigations after ATD treatment of pediatric GD are seem to be lacking completely.
Relapse of GD after stopping treatment was ascertained either by hyperthyroxinema with re‐appearance of overt or persistent (more than two months) thyrotoxic symptoms such as palpitation, excessive sweating, tremor, fatigue, weight loss, or by worsening biochemical thyrotoxicosis at two consecutive visits regardless of symptoms. When necessary, the possibility of superimposing painless thyroiditis was ruled out by perchenatate uptake test (Misaki 2003).

Thyroidectomy
Surgery still plays an important role when patients cannot tolerate ATD therapy, when medical treatment is rejected by patients, or when surgery is deemed the fastest and safest route in managing the patient (Schussler‐F. 2006). A subtotal or total thyroidectomy remains controversial. A subtotal thyroidectomy is performed most commonly. Contrarily, given that subtotal thyroidectomy provides an unpredictable outcome and that the risk of permanent complications is not greater than with total thyroidectomy, total thyroidectomy is the preferred option for the surgical management of Graves' disease (Barakate 2002). In Europe, subtotal or total thyroidectomy is the therapy of choice after recurrence during or after ATD treatment in pediatric GD (Cheetham 1998). Surgery also has been suggested as the primary choice of treatment. However, thyroidectomy has the disadvantages of hospitalization, surgical complications such as nerve palsy, postoperative hypoparathyroidism and postoperative hypothyroidism. Still thyroidectomy requires careful medical preparation and experienced anesthesiologists and surgeons to avoid operative morbidity. Postoperative recurrence of hyperthyroidism is observed in a significant number of patients (2 to 16%) (Witte 1997). Postoperative hypothyroidism, which occurs in up to 80% of the patients followed postoperatively, results in the need of lifelong thyroid hormone replacement therapy and lifelong monitoring of thyroid function.

Why it is important to do this review
In summary, strategies for pediatric GD vary from country to country. In Europe, ATD remains the initial treatment of choice in almost all the medical centers, with surgery being used mainly to deal with antithyroid failures, while radioiodine is preferred by only a small percentage of physicians for this group of patients (Perrild 1994). On the other hand, radioiodine therapy for pediatric GD has strong advocates in the USA, who emphasize the safety, simplicity and economic advantages of radioiodine ablation therapy , which should be considered more commonly in children (Halnan 1985). Recently, also in the UK and the Netherlands radioiodine treatment has been recommended as a primary choice or second‐line therapy for pediatric GD (Franklyn 1992).
There are currently no systematic reviews on studies of radioiodine treatment for pediatric GD.