Radioiodine or thyroidectomy in children and adolescents with Graves’ disease: a structured literature review of efficacy and safety

Abstract.

Graves’ disease (GD) is the most common cause of hyperthyroidism in children and adolescents. When remission is not achieved with antithyroid drugs, definitive treatment with radioiodine therapy (RAI) or thyroidectomy is recommended. We conducted a structured literature review of studies indexed in PubMed/MEDLINE to evaluate the efficacy and safety of these approaches in the paediatric population. Twenty-one studies including 9,490 patients were analysed. RAI achieved high remission rates (85–96%), with hypothyroidism representing the expected therapeutic outcome. Thyroidectomy provided definitive disease control in the vast majority of patients, with recurrence mainly associated with subtotal procedures. The most frequent surgical complication was transient hypocalcaemia (16.9%), while permanent hypoparathyroidism was rare (0.79%). RAI was generally associated with infrequent reported adverse events, such as transient sialadenitis, and long-term data did not show increased risks of malignancy or reproductive complications. Overall, available observational evidence suggests that both RAI and total thyroidectomy are effective definitive treatments for paediatric GD. In the absence of randomized comparative studies, treatment choice should be individualized according to patient age, thyroid characteristics, institutional expertise, and patient and family preferences.

Highlights

● RAI and thyroidectomy show high efficacy in paediatric Graves’ disease.

● RAI: high remission with favourable long-term safety profile.

● Total thyroidectomy ensures definitive control with low complication risk.

Introduction

Graves’ disease (GD) is the most common cause of hyperthyroidism in developed countries. It has a higher prevalence in women (5–10:1) and an incidence peak between 30–60 yr of age (1). GD is an autoimmune disease caused by the production of antibodies to the TSH receptor (TRAb) which stimulates the production of thyroid hormones. The most common symptoms are weight loss, fatigue, heat intolerance, tremor, and palpitations, determined by the hormonal hyperproduction. Additionally, extrathyroidal manifestations such as orbitopathy (GO) and pretibial dermopathy may be present (2). Childhood GD is a relatively rare disease, representing approximately 5% of all GD cases, with an incidence in children and adolescents around 4.58/100,000 cases per year. GD shows a 3.27 Female/Male ratio; this sex disparity increases with age, particularly during the second decade of life (3, 4). As in adults, the first-line treatment of paediatric GD is medical therapy with antithyroid drugs (ATD), mostly carbimazole (CBZ) or its active metabolite methimazole (MMI). Because of the risk of hepatic failure Propylthiouracil (PTU) should not be used in children. Indications for definitive treatment include relapse after ATD treatment, serious or persistent side effects of ATD, poor compliance or obstructive symptoms determined by a large goitre. Definitive treatment can be provided either by radioiodine therapy (RAI) or thyroidectomy (5). However, in paediatric patients, treatment selection is additionally influenced by age-related considerations, since current European Thyroid Association guidelines consider RAI contraindicated in children younger than 5 yr and relatively contraindicated below 10 yr of age. Therefore, the choice between RAI and thyroidectomy is guided not only by local expertise, but also by patient age, thyroid characteristics, and safety considerations. To our knowledge the most recent meta-analysis evaluating the best treatment for GD in children and adolescents is the one conducted by Jawaad M. et al. in 2023, which explored the effectiveness of ATD and RAI in this population (6).

With the present structured literature review, we aimed to assess whether a difference in clinical outcome exists between RAI or Thyroidectomy in children and adolescent patients affected by GD, who did not respond to ATD.

Material and Methods

Search strategy and inclusion criteria

A literature search was performed in the PubMed/MEDLINE database to identify studies evaluating definitive treatments for Graves’ disease in children and adolescents.

The review question was structured according to the Population–Intervention–Comparison–Outcome (PICO) framework:• Population: children and adolescents with Graves’ disease;• Intervention: radioiodine therapy (RAI);• Comparison: thyroidectomy;• Outcome: treatment efficacy and safety.

The following search terms and their combinations were used: “Graves disease”, “paediatric”, “children”, “adolescents”, “radioiodine”, “thyroidectomy”, “surgery”, “hyperthyroidism”.

The search was updated until 31 January 2026. Only articles published in English were considered. Preclinical studies, conference abstracts, reviews, and editorials were excluded. All retrieved records were exported and managed using EndNote version 20.3.

Eligibility criteria

Eligibility criteria were defined according to the review question. Studies were included if they involved children or adolescents diagnosed with Graves’ disease, evaluated radioiodine therapy and/or thyroidectomy as definitive treatment, reported clinical outcomes related to efficacy or safety. Exclusion criteria for the review (qualitative analysis) were reviews, letters, comments, editorials, and conference abstracts on the topic of interest; case reports or small case series (fewer than 5 enrolled patients), as these articles are characterized by low-quality evidence and are typically affected by publication bias; preclinical studies; studies including adults populations; articles written in languages other than English and original articles addressing fields of interest different from the aim of the present review.

Study selection

I.S. and L.E. independently read the titles and abstracts of the records generated by the search algorithm; discrepancies were resolved through discussion with a third reviewer (E.G.). They then determined which studies were eligible based on predefined criteria.

Reporting

This structured literature review was conducted using predefined eligibility criteria and a transparent study selection process, following relevant PRISMA principles where applicable (7). Study quality was assessed using the Newcastle–Ottawa Scale (NOS), which is designed for observational cohort studies. NOS scores for included studies are reported in Table 1.

Table 1. Characteristics of the human studies considered for the review.

Data extraction

Data extraction was performed independently by the reviewers using a standardized data collection form. Information extracted from each study included general study characteristics (authors, year of publication, country, study design), patient characteristics (sample size, age, sex distribution), treatment modality (radioiodine therapy or thyroidectomy), efficacy outcomes (remission or recurrence rates), safety outcomes (short- and long-term complications), additional clinical information, including length of hospital stay and quality of life when available.

The main findings of the included studies are summarized in the Results section and in Tables 1 and 2.

Table 2. Results and main findings of the human studies considered for the review.

Results

Study selection and characteristics

The literature search identified 953 records. After removal of duplicates and title and abstract screening, 89 full-text articles were assessed for eligibility. Twenty-one studies fulfilled inclusion criteria and were included in the qualitative synthesis (Fig. 1), encompassing a total of 9,490 paediatric patients with Graves’ disease undergoing definitive treatment (8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28). The majority of included studies were retrospective cohort analyses (18 out of 21), reflecting the absence of randomized comparative trials in this clinical setting. Most studies were conducted in the United States (12/21), followed by Europe (6/21), Asia (2/21), and South America (1/21). Geographic variability in iodine nutritional status should also be considered when interpreting radioiodine outcomes, as iodine intake may influence thyroidal iodine uptake and consequently affect administered activity requirements and therapeutic response.

Fig. 1.

. Flowchart of the study selection process for eligible studies.

Overall, methodological quality according to the NOS was moderate to high across most included studies, although retrospective design, heterogeneous outcome definitions, and inconsistent reporting of adverse events and follow-up duration represented common limitations (Table 1).

Ten studies evaluated radioiodine therapy in 810 patients (10,11,12, 14, 15, 21, 23, 26,27,28), thirteen investigated thyroidectomy in 790 patients (8,9,10, 12, 13, 16,17,18,19,20, 22, 24, 25); of them two cohorts included patients treated with both modalities (10, 12). RAI regimens varied considerably across studies, ranging from fixed activities of approximately 8 mCi (300 MBq) to weight- or thyroid-volume–adjusted approaches, with higher activities administered in cases of retreatment. Patient age also varied substantially across cohorts, with several studies predominantly enrolling adolescents, whereas younger children represented a minority of treated patients. Moreover, radioiodine dosing strategies evolved considerably across the decades, ranging from relatively conservative fixed activities in earlier studies to contemporary thyroid volume–adjusted ablative approaches.

Surgical strategies evolved over time: Subtotal or near-total thyroidectomy was adopted in two earlier series (9, 16), total thyroidectomy was performed in four studies (18, 19, 24, 25), while three studies did not clearly specify the extent of resection (8, 12, 22).

Treatment efficacy and safety

Radioiodine therapy: Across the included cohorts, RAI consistently achieved high rates of hyperthyroidism resolution in children and adolescents with Graves’ disease. Remission rates after one or more RAI treatments generally ranged between 85% and 96% (11, 21, 23, 26). In the recent prospective study by Sheremeta et al., remission was achieved in 125 of 144 patients (86.8%), with hypothyroidism developing in 83% of treated individuals and euthyroidism persisting in a minority (4%) (26). Similarly, in a multicentre U.S. cohort, hypothyroidism was documented in 54 of 67 patients (80.6%) at six months after a mean administered radioiodine activity of 12.5 mCi, whereas only one patient developed recurrent hyperthyroidism at 12 mo (27).

Earlier long-term studies reported remission rates exceeding 90%, accompanied by a progressive increase in hypothyroidism over extended follow-up (10,11,12, 14, 15, 21, 23, 28). In particular, Freitas et al. documented remission in 49 of 51 patients (96%), with the prevalence of hypothyroidism increasing from 47% at six months to 92% at long-term follow (11). Similar long-term efficacy was reported by Hamburger and Read, further confirming the durability of disease control over decades (12, 15).

Similarly, Thompson et al. reported high remission rates in 27 paediatric patients treated with RAI, with most achieving disease control after a single dose and 74% developing hypothyroidism over a mean follow-up of 44.5 mo (10). Hamburger et al. confirmed these findings in a larger cohort (163/186 remissions after one dose), with additional responses after repeat treatments (12). Other smaller series and large retrospective cohorts from Iran and Brazil likewise reported remission rates exceeding 90%, although retreatment was sometimes required in cases of persistent hyperthyroidism (21, 23).

Retreatment was required in a subset of patients, particularly among those with larger baseline thyroid volumes (26, 27). Across studies, thyroid gland size and functional activity emerged as key determinants of treatment response. In the Sheremeta cohort, smaller baseline thyroid volume (< 45 mL), lower FT4 and TRAb levels, and higher iodine uptake were independently associated with a greater likelihood of remission (26). Consistently, Trout et al. showed that persistent hyperthyroidism at six months was associated with a lower administered activity per millilitre of thyroid tissue, adjusted for 24-hour iodine uptake (27).

Importantly, long-term observational studies did not identify a consistent increase in thyroid malignancy, leukaemia, gonadal dysfunction, or major reproductive complications attributable to RAI exposure. Freitas et al. followed 51 paediatric patients for a mean of 14.6 ± 7.9 yr (range 5–27 yr) after RAI treatment and reported no cases of thyroid cancer, leukaemia, or impaired reproductive outcomes (11). Read et al. subsequently reported outcomes from 107 patients followed for a mean of 26.1 yr and 98 patients followed for a mean of 36.2 yr, again without evidence of thyroid carcinoma or leukaemia and without excess congenital abnormalities or spontaneous abortions (15). Similarly, Cury et al. reported a mean follow-up of 11.5 ± 3.6 yr in 65 paediatric patients, without observed thyroid carcinoma, gonadal abnormalities, or haematological complications after RAI exposure (21). However, safety outcome definitions and follow-up durations varied substantially across studies, limiting direct comparisons between cohorts (11, 12, 15, 21, 23).

Adverse events were uncommon and generally mild. Transient sialadenitis was reported in a small cohort (1/14 patients) (28), and one case of hyperparathyroidism was described in a retrospective series of 107 patients (15). Overall, RAI therapy demonstrated a favourable safety profile, with hypothyroidism representing the predictable and intended therapeutic outcome rather than a complication.

Thyroidectomy: In the seven studies (10, 12, 13, 16, 17, 19, 20) reporting surgical efficacy outcomes, definitive control of hyperthyroidism was achieved in 575 out of 635 patients; however, treatment success varied according to the extent of thyroid resection. Recurrence was observed almost exclusively in patients who underwent subtotal or near-total thyroidectomy (10, 16). Sugino et al. reported that leaving a thyroid remnant smaller than 3 g reduced the likelihood of relapse, although it did not completely abolish the risk (16). In contrast, total thyroidectomy was consistently associated with durable disease control, with postoperative hypothyroidism representing the expected and intentional therapeutic endpoint.

Transient hypocalcaemia was the most frequently reported postoperative complication, described in 11 studies (8,9,10, 13, 16,17,18,19,20, 22, 25). Across the available cohorts, it occurred in 128 of 756 patients (16.9%), with individual study rates ranging from 4.8% to 35.4% (19, 20, 22). Permanent hypoparathyroidism was less common, being reported in 6 of 756 patients (0.79%) (13, 22).

Transient recurrent laryngeal nerve (RLN) injury was documented in 14 of 756 patients (17, 22). In the contemporary cohorts reporting predominantly total thyroidectomy, transient RLN dysfunction was described without cases of permanent paralysis (17, 19, 20, 22). Permanent RLN paralysis was rare and reported mainly in older series that included subtotal thyroidectomy (8,9,10).

Postoperative hematoma was reported in a limited number of cases in earlier cohorts involving subtotal thyroidectomy (8, 10), including one patient requiring temporary airway intervention (8). One postoperative death was described in an earlier surgical cohort that included subtotal procedures (9).

Finally, length of hospital stay ranged from same-day discharge in high-volume contemporary centers to approximately three days in earlier studies (19, 20, 22).

Quality of life outcomes

Only one prospective study evaluating disease-specific and psychosocial quality of life following thyroidectomy was identified. Significant improvement was observed in multiple ThyPRO domains, including goiter symptoms, hyperthyroid symptoms, tiredness, anxiety, cognitive complaints, and overall disease impact. Improvements were also detected in total and physical quality of life scores (24). No comparable quality-of-life data were available for RAI-treated cohorts. Therefore, direct comparisons between definitive treatment modalities regarding patient-reported outcomes remain limited.

Discussion

GD in childhood and adolescence represents a therapeutic challenge, particularly when relapse occurs after ATD therapy and a definitive treatment is required. According to the 2022 guidelines of the European Thyroid Association, RAI and total thyroidectomy are both appropriate definitive options in selected paediatric patients, with the choice largely driven by age, goitre size, comorbidities, and local expertise (5). In this structured review including 21 studies and 9,490 paediatric patients, both strategies demonstrated high efficacy in achieving durable control of hyperthyroidism, albeit with distinct safety profiles and practical implications.

Across RAI cohorts, remission rates ranged between approximately 85% and 96%, confirming the high effectiveness of this approach in children and adolescents (11, 12, 15, 23, 26, 27). Hypothyroidism developed in the majority of successfully treated patients and should be interpreted as the intended therapeutic endpoint rather than a complication. Importantly, thyroid volume emerged consistently as a key predictor of response, with larger glands associated with higher likelihood of persistent hyperthyroidism and retreatment (26, 27). Another relevant factor potentially influencing radioiodine efficacy is iodine nutritional status. Thyroidal radioiodine uptake may vary according to dietary iodine intake, with iodine-deficient populations generally demonstrating higher iodine avidity compared with iodine-replete regions. Since the included studies originated from geographically heterogeneous areas, including Europe, North America, and Asia, differences in background iodine intake may have contributed to variability in administered activities and treatment response. However, most studies did not systematically report urinary iodine status or population iodine sufficiency, preventing formal comparative analyses. Moreover, the predominance of U.S.-based cohorts may limit the generalizability of the findings to healthcare systems with different treatment practices and referral patterns. Future prospective studies should consider iodine nutritional status as a potentially relevant determinant of radioiodine treatment outcomes in paediatric Graves’ disease.

Long-term follow-up studies extending over decades did not demonstrate a clinically meaningful increase in thyroid malignancy, gonadal dysfunction, or adverse reproductive outcomes (11, 15, 21, 23), although these data are derived from observational cohorts and must be interpreted cautiously.

Similarly, thyroidectomy achieved definitive control of hyperthyroidism in the vast majority of patients (574/635 in studies reporting efficacy outcomes). However, treatment success was closely related to the extent of resection. Recurrence was observed almost exclusively after subtotal or near-total thyroidectomy, whereas total thyroidectomy was consistently associated with durable remission (10, 16). These findings support current surgical practice favouring total thyroidectomy in paediatric GD to minimize relapse risk.

The safety profiles of the two treatments differ substantially. RAI was associated with a low rate of short-term adverse events, which were generally mild and transient, such as sialadenitis (28). In contrast, thyroidectomy carries a well-defined risk of surgical morbidity. Across the available cohorts, transient hypocalcaemia occurred in 16.9% of reported cases and permanent hypoparathyroidism in 0.79% (13, 19, 20, 22). Transient recurrent laryngeal nerve palsy was reported in 13/756 patients, with permanent nerve injury mainly confined to older surgical series that included subtotal procedures (8,9,10, 17, 19, 20, 22). Postoperative hematoma and one postoperative death were described in historical cohorts, highlighting the importance of surgical expertise and referral to high-volume centres (13, 22).

Thyroid eye disease (TED) also deserves consideration when selecting definitive treatment. Although inflammatory manifestations of Graves’ orbitopathy are generally reported to be less severe in paediatric than in adult patients, radioiodine therapy may exacerbate active orbitopathy and should therefore be avoided in patients with active TED according to current European Thyroid Association guidelines (5). However, data specifically evaluating TED progression after RAI in children and adolescents were not systematically reported in the studies included in this review.

Age remains a critical determinant in therapeutic decision-making. According to current European Thyroid Association guidelines, RAI is contraindicated in children younger than 5 yr and relatively contraindicated below 10 yr of age because of concerns regarding increased radiation sensitivity in developing tissues, making thyroidectomy the preferred definitive treatment in younger patients (5). In the studies included in this review, younger children represented only a minority of treated patients, whereas most cohorts predominantly enrolled adolescents. Moreover, the studies spanned more than four decades, during which substantial changes occurred in radioiodine dosing strategies and therapeutic goals. Earlier cohorts often adopted more conservative approaches aimed at achieving euthyroidism while minimizing radiation exposure, whereas more contemporary protocols more frequently pursue complete thyroid ablation with subsequent hypothyroidism as the intended therapeutic outcome. This temporal heterogeneity likely contributed to differences in administered radioiodine activities and treatment outcomes across studies. Unfortunately, age-stratified efficacy and safety data were inconsistently reported, preventing formal comparisons between children younger than 10 yr versus ≥ 10 yr. In adolescents, both RAI and total thyroidectomy appear effective and acceptable options. Therefore, beyond strict clinical indications, shared decision-making involving patients and families becomes central, especially when considering differences in risk perception between immediate surgical complications and theoretical long-term radiobiological risks.

An important gap identified in our review concerns patient-reported outcomes. Only one prospective study systematically assessed quality of life after thyroidectomy, demonstrating significant improvements in several domains and in physical and school-related scores (24). No comparable quality-of-life data were available for RAI-treated cohorts. Evidence from adult Graves’ disease cohorts further supports the relevance of treatment-related quality-of-life outcomes. In a review comparing long-term antithyroid drugs, radioiodine, and surgery, El Kawkgi et al. highlighted that quality of life may differ across treatment modalities, with some adult data suggesting worse thyroid-specific quality-of-life scores after RAI compared with antithyroid drugs or surgery (29). Although these findings cannot be directly extrapolated to children and adolescents, they reinforce the need to include patient-reported outcomes in future paediatric studies. Given the psychosocial vulnerability of adolescents, especially in relation to body image and chronic disease burden, future studies should systematically incorporate validated patient-reported outcome measures when comparing definitive treatments.

This review has several limitations. First, no randomized controlled trials comparing RAI and thyroidectomy in paediatric GD were identified. Second, most included studies were retrospective and spanned several decades, introducing heterogeneity related to advances in surgical technique, anaesthesia, perioperative management, and radiation dosing. Third, RAI protocols varied considerably (fixed versus weight- or volume-adjusted dosing), and older surgical series included subtotal thyroidectomy, limiting direct comparability with contemporary total thyroidectomy outcomes. Given the marked heterogeneity in study design, patient selection, radioiodine dosing strategies, surgical approaches, outcome definitions, and follow-up duration, a formal quantitative synthesis was not considered methodologically appropriate. Finally, complication reporting was not uniform across studies, and follow-up duration differed substantially.

Despite these limitations, our findings indicate that available observational evidence suggests that both RAI and total thyroidectomy provide high rates of definitive disease control for paediatric GD. Total thyroidectomy offers immediate and definitive disease control at the expense of a quantifiable risk of transient and, rarely, permanent surgical complications. RAI provides a non-surgical alternative with excellent efficacy and a favourable short-term safety profile; available observational studies have not demonstrated a clear increase of long-term oncologic or reproductive risk. In adolescents, where both options are generally feasible, treatment choice should be individualized, integrating age, thyroid volume, orbitopathy status, comorbidities, institutional expertise, and patient and family preference.

Further prospective, multicentre studies with standardized outcome definitions and long-term follow-up, including quality-of-life assessment, are needed to better define the optimal definitive treatment strategy in children and adolescents with Graves’ disease.

Conclusion

Both radioiodine therapy and total thyroidectomy represent effective definitive treatment options for children and adolescents with Graves’ disease after failure of antithyroid drug therapy. Treatment selection should be individualized according to patient age, thyroid characteristics, orbitopathy status, comorbidities, institutional expertise, and patient and family preference. Further prospective studies with standardized outcomes, long-term safety assessment, and quality-of-life evaluation are needed to optimize treatment strategies in paediatric Graves’ disease.

Conflict of interests

The authors have no relevant financial or non-financial interests to disclose.