Concurrent Association of Multifocality, Bilaterality, and Recurrence in Pediatric Papillary Thyroid Cancer Patients

Abstract

Purpose

Pediatric papillary thyroid cancer (PTC) is recommended to perform aggressive surgery to reduce the risk of recurrence. This study was designed to evaluate the concurrent association between multifocality, bilaterality, and the risk of recurrence in pediatric PTC.

Materials and Methods

This retrospective cohort study included pediatric patients (age <19 years) who underwent total thyroidectomy for PTC between 1996 and 2014 in a single tertiary center. Clinicopathological parameters were analyzed to evaluate the prevalence of multifocality, bilaterality, recurrence, and their association.

Results

We analyzed 58 pediatric patients with PTC. There was no factor related to the presence of multifocality or bilaterality in multivariate analysis. Also, in univariate analysis, multifocality and bilaterality were not independent risk factors of each other’s presentation (p=0.061 and p=0.061, respectively). Recurrence was observed in 19 (32.8%) patients. In multivariate analysis of recurrence, clear cell subtype, multifocality, and gross extrathyroidal extension (ETE) were independent risk factors (p=0.027, p=0.035, and p=0.038, respectively). Most recurrences (68.4%) happened during the first 4 years of follow-up after the initial thyroidectomy.

Conclusion

Multifocality and bilaterality were not independent risk factors of each other’s presentation; however, multifocality was the risk factor for recurrence in pediatric PTC. For pediatric PTC, close monitoring for recurrence within the initial 4 years is recommended, particularly in patients with clear cell subtype, multifocality, and gross ETE.

Graphical Abstract

INTRODUCTION

Pediatric thyroid cancer is a rare disease, although it is the most common endocrine malignancy in children.¹,²,³ Pediatric thyroid cancer has recently gained growing interest due to its continuously increasing incidence.³,⁴,⁵ This disease mainly appears in the form of well-differentiated thyroid cancer (WDTC), and papillary thyroid cancer (PTC) accounts for more than 90% of cases of WDTC, followed by follicular thyroid cancer (FTC).⁶ Pediatric thyroid cancer often presents distinct characteristics compared to adult thyroid cancer, including larger tumor sizes, higher prevalence of extrathyroidal extension (ETE), and increased incidence of lymph node (LN) or distant metastasis.⁷,⁸,⁹ Moreover, certain studies have reported that pediatric cases often present a higher rate of multiplicity and bilaterality.¹⁰,¹¹ Therefore, the latest guidelines for pediatric thyroid cancer recommend total thyroidectomy as an initial surgical approach.⁶

Factors such as multifocality and bilaterality are not only related to recurrence but also could determine the extent of surgery.¹¹,¹² Although multifocality without ETE or LN metastasis is classified as low-risk according to the American Thyroid Association risk stratification guidelines for adults,¹³ some studies on pediatric thyroid cancer have reported that the prognosis of multifocal disease is significantly worse than that of unifocal disease.¹⁰,¹¹,¹⁴,¹⁵,¹⁶ Moreover, Cherella, et al.¹⁷ proposed that immediate completion thyroidectomy should be considered for pediatric patients with multifocal disease due to its association with bilateral disease. Despite these findings, the relationship between multifocality and bilaterality in pediatric PTC and their clinical implications remain insufficiently explored.

Given this background, this study aimed to identify clinical factors influencing multifocality and bilaterality in pediatric PTC and to examine their interrelationship. Furthermore, we sought to investigate how these factors are concurrently associated with recurrence, with the goal of providing surgical management and follow-up strategies for pediatric patients with PTC.

MATERIALS AND METHODS

Study population

This retrospective study analyzed pediatric patients with PTC who underwent total thyroidectomy in a tertiary center between January 1996 and December 2014. Current guidelines define PTC and FTC as well-differentiated types.⁶,¹²,¹³,¹⁷ Minimally invasive FTC has a favorable prognosis similar to PTC; however, their clinicopathological features and recurrence pathway may differ.¹⁸,¹⁹ Thus, we only included PTC patients in this study. All patients received subsequent radioactive iodine remnant ablation. This study was approved by the Institutional Review Board of Asan Medical Center (2015-0844). The requirement for informed consent was waived due to the anonymity and retrospective nature of this study. The exclusion criteria included persistent disease after initial treatment, distant metastasis at initial diagnosis, a postoperative follow-up period of <12 months, and lack of medical records. The associations between clinicopathological risk factors and multifocality, bilaterality, and structural recurrence were evaluated. All treatment methods were performed in accordance with the relevant guidelines and regulations.

Data collection

Data were collected by manually reviewing the electronic medical records of the patients, including demographic data such as preoperative and clinicopathological information about the disease, as well as postoperative outcomes. Additionally, clinicopathological features, such as age, sex, primary tumor size, preoperative ultrasonography (US) finding, final pathology, multifocality, bilaterality, gross ETE, LN metastasis status, structural recurrence, and duration of no evidence of disease (NED), were evaluated. The NED period refers to the interval between the initial curative treatment and the detection of recurrence or the last known survival date. Follow-up US and thyroid function test after surgery were performed as a routine procedure; every 6 months for the first 2 years and every year after that. If recurrence was suspected, computed tomography was additionally performed. Moreover, recurrences were defined as structural recurrences confirmed by cytological or pathological reports. Preoperative US findings included the nodule content, calcifications, margins, echogenicity, and other nodules with a diameter of ≥0.5 cm. All specimens were pathologically reviewed. Here, multifocality was defined as the presence of 1) more than one focus of primary thyroid cancer in a unilateral thyroid lobe or 2) primary thyroid cancer involving the entire unilateral lobe.¹⁷ One focus of primary thyroid cancer presented in each bilateral lobes was considered unifocal in this study.

Statistical analysis

All statistical analyses were performed using SPSS statistics version 26 (IBM Corp., Armonk, NY, USA). Continuous variables are presented as means with standard deviations and ranges. Categorical variables are presented as numbers with percentages. Two-tailed p-values were calculated using Student’s t-test and Wilcoxon rank-sum test for continuous variables and Fisher’s exact test and the chi-square test for categorical variables. Univariate and multivariate logistic regression analyses were performed to find out the risk factors that were presumed to be related to multifocality and bilaterality. Only variables in univariate analysis with p-values of less than 0.05 were included in the multivariate analysis. Cox proportional hazards modeling analysis were performed to identify the risk factors for structural recurrence. Odds ratios with 95% confidence intervals (CI) were calculated in logistic regression analysis, and hazard ratios (HR) with 95% CIs were calculated in Cox proportional hazards modeling analysis. p-values of less than 0.05 were considered statistically significant.

RESULTS

Baseline features of pediatric patients with PTC

The baseline clinical and pathological characteristics of 90 pediatric patients with WDTC who underwent thyroidectomy were initially collected. Among the study population, 24 patients with lobectomy, three patients with persistent disease, two patients with distant metastasis in initial diagnosis, two patients with FTC, and one patient with less than 1 year follow-up period were excluded from this study. Finally, 58 pediatric PTC patients were analyzed, and their characteristics are presented in Table 1. No patient had a history of radiation exposure. Among the study population, six patients had a family history of thyroid cancer. However, information of thyroid cancer relatives was not obtainable from any of the patients for further investigation. The mean age of the study cohort was 15.5 years, and most patients were female (47, 81.0%). The mean primary tumor size was 2.9 cm, and 33 (56.9%) of the patients had tumor sizes >2.0 cm. The US findings of 48 patients were evaluable, excluding the US findings of 10 patients with insufficient information. Solid parenchyma was observed in 47 (97.9%) of the patients preoperatively, with a high incidence of calcifications (34, 70.8%) and margin irregularity (23, 47.9%). However, a taller than wide feature was seen in 3 (6.3%) of the patients. On preoperative US, additional thyroid nodules of a size ≥0.5 cm presenting on ipsilateral and contralateral sides were 12 (25.0%) and 28 (37.5%), respectively. As for final pathology, most cases of PTC subtype were classical (34, 58.7%), followed by diffuse sclerosing (9, 15.5%), tall cell (7, 12.1%), clear cell (4, 6.9%), follicular variant (2, 3.4%) and others. Multifocality was observed in 17 (29.3%), and bilaterality was observed in 17 (29.3%) as well. Gross ETE was observed in 12 (20.7%) of the patients. Central compartment LN dissection was performed in all patients, and central LN metastasis was observed in 52 (89.7%) of the patients. The mean numbers of central and lateral LN metastasis were 7.5 and 15.2, respectively. Structural recurrences were found in 19 (32.8%), and the mean NED period was 102.1 months. The mean follow-up duration for this study population was 11 years (134.8 months, range 15 to 291).

Table 1. Clinicopathological Characteristics of Pediatric Patients Who Underwent Total Thyroidectomy for PTC (n=58).

Characteristics		Value
Age (yr)		15.5±2.5 (6–18)
Sex
	Male	11 (19.0)
	Female	47 (81.0)
Primary tumor size (cm)		2.9±1.7 (0.8–9.8)
Primary tumor size >2 cm		33 (56.9)
Preoperative ultrasonography finding (n=48)
	Solid parenchyma	47 (97.9)
	Calcifications	34 (70.8)
	Irregular margins	23 (47.9)
	Taller than wide	3 (6.3)
	Hypoechoic	36 (75.0)
	Ipsilateral nodule present	12 (25.0)
	Contralateral nodule present	28 (37.5)
Final pathology
	PTC, classical	34 (58.7)
	PTC, diffuse sclerosing	9 (15.5)
	PTC, tall cell	7 (12.1)
	PTC, clear cell	4 (6.9)
	PTC, follicular variant	2 (3.4)
	PTC, others*	2 (3.4)
Multifocality		17 (29.3)
Bilaterality		17 (29.3)
Gross ETE		12 (20.7)
Extent of LN dissection
	Central	58 (100)
	Lateral	24 (41.4)
LN metastasis
	None	6 (10.3)
	Central	52 (89.7)
	Lateral	24 (41.4)
Number of retrieved central LN		13.5±7.3 (1–33)
Number of central LN metastasis		7.5±5.9 (0–26)
Number of retrieved lateral LN		61.8±34.3 (20–151)
Number of lateral LN metastasis		15.2±8.8 (3–37)
Structural recurrence		19 (32.8)
Period of no evidence of disease (months)		102.1±64.7 (12–289)
Follow up duration (months)		134.8±61.7 (15–291)

PTC, papillary thyroid cancer; ETE, extrathyroidal extension; LN, lymph node.

Data are presented as mean±standard deviation (range) or n (%).

^*Others consisted of one columnar cell type and one oncocytic type.

Risk factors associated with multifocality and bilaterality

The univariate analysis revealed that larger primary tumor size (p=0.018), gross ETE (p=0.018), and number of lateral neck LN metastasis (p=0.030) were significantly associated with multifocality than with unifocality. However, in the multivariate analysis, no independent factors were associated with multifocality (Table 2).

Table 2. Risk Factor Analysis of Pediatric PTC Patients Who Underwent Total Thyroidectomy, Comparing Unifocality and Multifocality.

Characteristics		Unifocality (n=42)	Multifocality (n=18)	p value	Univariate		Multivariate
					OR (95% CI)	p value	OR (95% CI)	p value
Age (yr)		15.8±1.9	14.7±3.4	0.245	0.84 (0.67–1.06)	0.149
Female sex		34 (82.9)	13 (76.5)	0.715	0.67 (0.17–2.67)	0.570
Primary tumor size (cm)		2.5±1.3	3.7±2.1	0.010	1.61 (1.08–2.40)	0.018	1.46 (0.74–2.88)	0.281
Primary tumor size >2 cm		22 (53.7)	11 (64.7)	0.564	1.58 (0.49–5.10)	0.441
Preoperative ultrasonography finding
	Solid parenchyma	35 (97.2)	12 (100)	0.999	N/A
	Calcifications	26 (72.2)	8 (66.7)	0.726	0.77 (0.19–3.13)	0.714
	Irregular margins	16 (44.4)	7 (58.3)	0.511	1.75 (0.47–6.57)	0.407
	Taller than wide	2 (5.6)	1 (8.3)	0.999	1.55 (0.13–18.73)	0.732
	Hypoechoic	28 (77.8)	8 (66.7)	0.462	0.57 (0.14–2.40)	0.445
	Ipsilateral nodule present	8 (22.2)	4 (33.3)	0.462	1.75 (0.42–7.35)	0.445
	Contralateral nodule present	13 (36.1)	5 (41.7)	0.743	1.26 (0.33–4.80)	0.731
Final pathology*
	PTC, classical	26 (63.4)	8 (47.1)	0.380	0.51 (0.16–1.61)	0.253
	PTC, diffuse sclerosing	6 (14.6)	3 (17.6)	0.999	1.25 (0.27–5.70)	0.773
	PTC, tall cell	4 (9.8)	3 (17.6)	0.407	1.98 (0.39–9.99)	0.407
	PTC, clear cell	2 (4.9)	2 (11.8)	0.573	2.60 (0.34–20.17)	0.361
	PTC, follicular variant	1 (2.4)	1 (5.9)	0.504	2.50 (0.15–42.44)	0.526
Bilaterality		9 (22.0)	8 (47.1)	0.067	3.16 (0.95–10.55)	0.061
Gross ETE		5 (12.2)	7 (41.2)	0.028	5.04 (1.31–19.34)	0.018	11.70 (0.56–44.95)	0.113
Number of central LN metastasis		6.4±5.7	10.2±5.7	0.025	1.09 (0.99–1.21)	0.076
Number of lateral LN metastasis		13.2±8.7	17.3±8.8	0.265	1.16 (1.01–1.32)	0.030	1.14 (0.95–1.36)	0.163

OR, odds ratio; CI, confidence interval; PTC, papillary thyroid cancer; ETE, extrathyroidal extension; LN, lymph node.

Data are presented as mean±standard deviation or n (%).

^*PTC, others were not included since no patient had multifocal disease.

In the univariate analysis, the risk factors for bilaterality relative to unilaterality were female sex (p=0.050), larger primary tumor size (p=0.024), and the number of central LN metastasis (p=0.034). However, in the multivariate analysis, no independent factors were related to the bilaterality (Table 3).

Table 3. Risk Factor Analysis of Pediatric PTC Patients Who Underwent Total Thyroidectomy, Comparing Unilaterality and Bilaterality.

Characteristics		Unilaterality (n=42)	Bilaterality (n=18)	p value	Univariate		Multivariate
					OR (95% CI)	p value	OR (95% CI)	p value
Age (yr)		15.8±1.9	14.7±3.4	0.247	0.84 (0.67–1.06)	0.149
Female sex		36 (87.8)	11 (64.7)	0.064	0.26 (0.07–0.99)	0.050	0.59 (0.11–3.24)	0.544
Primary tumor size (cm)		2.5±1.0	3.8±2.5	0.053	1.56 (1.06–2.29)	0.024	1.47 (0.96–2.25)	0.077
Primary tumor size >2 cm		20 (48.8)	13 (76.5)	0.080	3.41 (0.95–12.24)	0.060
Preoperative ultrasonography finding
	Solid parenchyma	33 (97.1)	14 (100)	0.999	N/A
	Calcifications	24 (70.6)	10 (71.4)	0.999	1.04 (0.26–4.12)	0.954
	Irregular margins	15 (44.1)	8 (57.1)	0.529	1.69 (0.48–5.93)	0.414
	Taller than wide	2 (5.9)	1 (7.1)	0.999	1.23 (0.10–14.78)	0.870
	Hypoechoic	23 (67.6)	13 (92.9)	0.081	6.22 (0.72–53.76)	0.097
	Ipsilateral nodule present	7 (20.6)	5 (35.7)	0.294	2.14 (0.54–8.46)	0.277
	Contralateral nodule present	11 (32.4)	7 (50.0)	0.330	2.09 (0.59–7.45)	0.255
Final pathology*
	PTC, classical	26 (63.4)	8 (47.1)	0.380	0.51 (0.16–1.61)	0.253
	PTC diffuse sclerosing	6 (14.6)	3 (17.6)	0.999	1.25 (0.27–5.71)	0.773
	PTC, tall cell	6 (14.6)	1 (5.9)	0.661	0.37 (0.04–3.28)	0.368
	PTC, clear cell	2 (4.9)	2 (11.8)	0.573	2.60 (0.34–20.17)	0.361
	PTC, others	1 (2.4)	1 (5.9)	0.504	2.50 (0.15–42.44)	0.526
Multifocality		9 (22.0)	8 (47.1)	0.067	3.16 (0.95–10.55)	0.061
Gross ETE		7 (17.1)	5 (29.4)	0.307	2.02 (0.54–7.60)	0.296
Number of central LN metastasis		6.6±6.1	9.7±4.9	0.067	1.12 (1.01–1.24)	0.034	1.07 (0.95–1.21)	0.255
Number of lateral LN metastasis		11.6±6.3	20.3±9.6	0.013	1.06 (0.96–1.17)	0.258

OR, odds ratio; CI, confidence interval; PTC, papillary thyroid cancer; ETE, extrathyroidal extension; LN, lymph node.

Data are presented as mean±standard deviation or n (%).

^*PTC, others were not included since no patient had multifocal disease.

Risk factors associated with structural recurrence and recurrence pattern analysis

To determine the concurrent association between multifocality, bilaterality and the risk of recurrence, 58 patients were initially classified according to their multifocality, and were then subdivided according to their laterality (Fig. 1). In this figure, classical refers to the subtypes except diffuse sclerosing subtype, which may have a more aggressive recurrence pattern. Of the 58 patients, structural recurrence was observed in 19 (32.8%) patients. Among 41 unifocal patients, the unifocal-unilateral type was presented in 32 (78.1%) patients, and 9 (21.9%) presented the unifocal-bilateral type. Within the unifocal-unilateral type, 26 (81.2%) patients presented with a classical type (one focus of primary thyroid cancer in unilateral lobe); the recurrence rate was 4 (15.4%). The other 6 (18.8%) patients with unifocal-unilateral type presented as diffuse sclerosing subtype without the involvement of a whole thyroid, and the recurrence rate was 1 (16.7%). The recurrence rate of unifocal-unilateral type was 5 (15.6%). All nine patients with unifocal-bilateral type had one focus of primary thyroid cancer in each lobe, and the recurrence rate was 3 (33.3%). Overall, 8 (19.5%) of patients with the unifocal type had structural recurrence.

Fig. 1. Diagram of pediatric patients with PTC according to primary tumor multifocality and bilaterality with the rate of structural recurrences. PTC, papillary thyroid cancer.

Among 17 multifocal patients, the multifocal-unilateral type was observed in 9 (52.9%) patients, and multifocal-bilateral type was observed in 8 (47.1%) patients. Within the multifocal-unilateral type, all nine patients presented with more than one focus of primary thyroid cancer in the unilateral lobe, and the recurrence rate was 6 (66.7%). In multifocal-bilateral type, primary thyroid cancer presented as bilateral diffuse sclerosing subtype without the involvement of a whole thyroid one lobe were found in 3 (37.5%) patients, and primary thyroid cancer presented in one entire lobe were found in 2 (25.0%) patients. The recurrence rate of the multifocal-bilateral type was 5 (62.5%). Overall structural recurrence was found in 11 (64.7%) patients with the multifocal type.

In all, unifocal-unilateral type was presented in half (32, 55.2%) of pediatric PTC patients; among them, unifocal-unilateral classical type was the most common (26, 44.8%), followed by unifocal-bilateral classical and multifocal-unilateral classical types with 9 (15.5%) patients each. However, the recurrence rate of unifocal-unilateral was 5 (8.6%). Within 30% of the patients with multifocal disease, eight had multifocal-bilateral disease, and the recurrence rate was 5 (8.6%).

The risk factors for structural recurrences were evaluated using univariate and multivariate analyses (Table 4). Clear cell type PTC, multifocality, and gross ETE were risk factors for recurrence in the multivariate analysis, and they were also independent risk factors for structural recurrence (HR, 4.43; p=0.027, HR, 2.99; p=0.035, HR, 3.04; p=0.038, respectively) (Table 4).

Table 4. Univariate and Multivariate Analyses of Risk Factors for Structural Recurrence in Pediatric Patients Who Underwent Total Thyroidectomy for PTC.

Related factors		Univariate		Multivariate
		HR (95% CI)	p value	HR (95% CI)	p value
Age (yr)		0.95 (0.80–1.13)	0.567
Female sex		0.91 (0.30–2.76)	0.863
Primary tumor size (cm)		1.21 (0.98–1.51)	0.083
Primary tumor size >2 cm		0.93 (0.37–2.36)	0.879
Final pathology
	PTC, classical	0.39 (0.15–1.02)	0.054
	PTC, diffuse sclerosing	0.92 (0.31–3.70)	0.920
	PTC, tall cell	1.98 (0.65–6.02)	0.230
	PTC, clear cell	4.22 (1.21–14.72)	0.024	4.43 (1.18–16.58)	0.027
Multifocality		4.64 (1.86–11.59)	0.001	2.99 (1.08–8.27)	0.035
Bilaterality		1.91 (0.76–4.81)	0.172
Gross ETE		4.41 (1.72–11.34)	0.002	3.04 (1.06–8.72)	0.038
Number of central LN metastasis		1.05 (0.98–1.13)	0.199
Number of lateral LN metastasis		1.08 (0.99–1.16)	0.070

HR, hazard ratio; CI, confidence interval; PTC, papillary thyroid cancer; ETE, extrathyroidal extension; LN, lymph node.

For further analysis, recurrence patterns were evaluated according to the postoperative follow-up months after the initial surgery. Most recurrences were found during the first 4 years of follow-up (68.4%) in the study group (Fig. 2). The rate of total thyroidectomy in pediatric patients with thyroid cancer recently increased from eight patients in 1996–2000 to 21 patients in 2011–2014. However, the recurrence rate reduced as the rate of operation increased in recent years, with 50% (4/8), 50% (5/10), 26.3% (5/19), and 23.8% (5/21) in 1996–2000, 2001–2005, 2006–2010, and 2011–2014, respectively (Fig. 3).

Fig. 2. Number of structural recurrences according to the postoperative follow-up months after the initial surgery.

Fig. 3. Number of total thyroidectomies performed and the number of structural recurrences according to the year of initial surgery.

DISCUSSION

In this retrospective study of 58 pediatric patients with PTC, we evaluated the clinicopathological features to determine their relationship with multifocality, bilaterality, and structural recurrence. The incidence rates of multifocality, bilaterality, and recurrence were 30% each. Multifocality and bilaterality were not related to each other’s presentation, but we found that multifocality was an independent risk factor for recurrence. In multivariate analysis, multifocality, clear cell variant, and gross ETE were identified as risk factors for recurrence. Recurrences occurred throughout the unifocal-unilateral, unifocal-bilateral, multifocal-unilateral, and multifocal-bilateral types, with the multifocal-unilateral type having the highest recurrence rate 6 (66.7%) and the unifocal-unilateral type having the lowest recurrence rate 4 (15.4%). Most recurrences happened during the first 4 years of follow-up.

Despite the small population size of this study, the rates of multifocality, bilaterality, and structural recurrence were high, and the results were similar to those reported in other studies.¹⁰,¹²,¹⁴,¹⁸ Due to the high prevalence of bilateral disease, the current guidelines recommend total thyroidectomy as an initial approach for pediatric patients with PTC, regardless of the tumor stage.⁶ However, several recent studies have suggested that less extensive surgery could be performed for localized disease.¹⁰,¹⁹,²⁰,²¹ According to Baumgarten, et al.,¹⁰ multifocality, widely invasive PTC on US, and lymphadenopathy were predictors of bilateral disease. Moreover, they reported that nodules >2 cm and the diffuse sclerosing variant of PTC were associated with bilateral disease. Herein, larger primary tumor size, higher proportion of male sex, and higher number of central LN metastasis were observed in the bilateral group compared to the unilateral group. However, in the multivariate analysis, no independent risk factors were found in relationship to bilaterality.

One study reported multifocality as a predictor of bilaterality, and another reported that if multifocal disease is confirmed after lobectomy or the entire lobe is occupied by the disease, the probability of residual disease in the contralateral lobe would be high.¹⁰,¹⁷ Herein, multifocality was a related factor of bilaterality in the univariate analysis. However, it was not statistically significant in the multivariate analysis. Within this study population, the bilaterality was observed only in half 8 (47.1%) of the multifocal patients.

Several studies reported that multifocality is an important risk factor for poor outcomes in adults.¹¹,¹⁵,¹⁷,²²,²³,²⁴ According to a retrospective study of a large pediatric thyroid cancer cohort by Lee, et al.,¹¹ the multifocality rate of pediatric thyroid cancer has not changed over 33 years, and multifocality was associated with poor recurrent-free survival in the multivariate analysis (HR, 5.4; p=0.005). However, in that study, the total thyroidectomy and lobectomy groups were not separated, and the definition of multifocality was different from that used in this study. We have adopted the definition of multifocality used in a study by Cherella, et al.¹⁷ to evaluate the occult bilaterality in the contralateral lobe and their effect on recurrence. Moreover, we considered that multifocality might be another manifestation of intraglandular metastasis, which is frequently observed in diffuse sclerosing PTC variants. However, no clear associations were found between multifocality, bilaterality, and diffuse sclerosing PTC variants in the univariate and multivariate analyses in this study.

Various clinicopathological factors affect recurrence. We identified the PTC clear cell variant, multifocality, and gross ETE as predictors of recurrence in the univariate analysis, and these findings were consistent in the multivariate analysis and also with those of previous studies.¹¹,¹⁸,²⁰,²⁵,²⁶,²⁷ However, the multifocality, which was thought to be a risk factor related to the intraglandular spread of the disease in the thyroid gland, was not associated with incidental bilaterality. Our findings suggest that the multifocality without bilaterality could be related to the aggressiveness and metastatic potential of cancer. Therefore, in patients with multifocal disease, even the lesions confined to unilateral lobe could be highly correlated with recurrence; therefore, careful follow-up strategies are required for the outpatient department.

Herein, most recurrences appeared during the first 4 years of follow-up, and the recurrence rate presented a decreasing trend as the rate of operation increased in recent years. This correlated with the findings reported in previous studies presenting early recurrences in pediatric patients.¹¹,¹⁸,²⁸,²⁹ However, based on our previous study, the detected frequency of malignancy in patients has recently increased, and the aggressiveness status of thyroid cancer has not increased.³⁰ Moreover, prompt detection of early-stage thyroid cancer through highly sophisticated US enabled us to perform minimal surgery at an early stage, which could have attributed to the decreased recurrence rate.

This study was limited in that it was a retrospective study conducted at a single center with a small sample size, having not enough strength to present a statistically strong analysis. However, the small sample size was due to the rarity of the disease. Although the single-center design limited the patient cohort size, we believe the number of enrolled patients was sufficient for our analysis. The strength of this study was that, even with a limited patient cohort size, the incidence rates of multifocality, bilaterality, and recurrence were similar to those reported in other studies. In addition, the time span of nearly two decades provides a comprehensive overview of the trends and outcomes in this patient cohort. Further studies with a larger sample size are needed to overcome these limitations.

In conclusion, in this study, 70% of the pediatric patients with PTC who underwent total thyroidectomy as initial therapy had unifocal disease, and >50% of the patients had unifocal-unilateral disease. The recurrence rate of unifocal-unilateral was 5 (8.6%). Within 30% of those with multifocal disease, eight had multifocal-bilateral disease, and the recurrence rate was 5 (8.6%). Based on our findings, multifocality and bilaterality were not independently correlated, but multifocality was a related factor of recurrence. Furthermore, patients presenting with clear cell variant, multifocality, and gross ETE at diagnosis should be closely observed for recurrence within the first 4 years after the initial surgery.