Establishment and test of graded diagnostic model of thyroid cancer in children and adolescents

doi:10.19401/j.cnki.1007-3639.2022.05.004

Abstract

Abstract:

Background and Purpose: The incidence rate of thyroid cancer in children and adolescents is increasing year by year, and there is a high rate of metastasis and recurrence. The application of diagnostic criteria for thyroid nodules based on adult characteristics in children has a high rate of missed diagnosis rate, misdiagnosis rate and unnecessary biopsy. This study aimed to establish a graded diagnostic model of thyroid nodules in children and adolescents, and to compare its diagnostic efficacy with American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Methods: One Hundred and forty-four patients with thyroid nodules aged 0-18 years in Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College from January 2000 to October 2017 were included as the training set, and 41 patients with thyroid nodules aged 0-18 years in China-Japan Friendship Hospital and Civil Aviation General Hospital from November 2015 to January 2022 were included as the testing set. Taking pathological diagnosis as the gold standard, the statistically significant ultrasonic characteristics were analyzed by multivariate binary logistic regression. The ultrasonic grading diagnostic model was established and brought into the testing set to compare its diagnostic efficiency with ACR TI-RADS. Results: There was significant difference in solidity, hypoechoic, taller-than-wide shape, irregular edge and microcalcification between the two groups (P<0.05). Logistic regression analysis showed that hypoechoic, solidity, irregular margin and calcification were independent predictors for the diagnosis of thyroid cancer in children and adolescents. When the ultrasonic grade was established based on the above independent predictors, this graded model had higher diagnostic accuracy in the testing set compared with ACR TI-RADS (92.7% vs 82.9%). Conclusion: The graded diagnostic model based on solidity, hypoechoic, irregular margin and microcalcification can be used in the diagnosis of thyroid nodules in children and adolescents. Compared with ACR TI-RADS, it has higher diagnostic accuracy.

Key words: Thyroid cancer, Children, Adolescents, Ultrasound, Graded diagnotic model

CLC Number:

R736.1

LIU Jia, LI Guanghan, GAO Luying, ZHANG Zijie, WANG Ying, XIONG Ying, ZHANG Bo. Establishment and test of graded diagnostic model of thyroid cancer in children and adolescents[J]. China Oncology, 2022, 32(5): 397-403.

Figures/Tables 8

Tab. 1

Tab. 2

Fig. 1

Fig. 2

Fig. 3

Tab. 3

Fig. 4

Tab. 4

References 21

[1]	SIEGEL R L, MILLER K D, JEMAL A. Cancer statistics, 2018[J]. CA A Cancer J Clin, 2018, 68(1): 7-30. doi: 10.3322/caac.21442
[2]	国家儿童医学中心, 国家儿童肿瘤监测中心, 中华医学会小儿外科学分会, 等. 中国儿童甲状腺结节及分化型甲状腺癌专家共识[J]. 中华实用儿科临床杂志, 2020, 35(20): 1521-1530.
	National Center for Children’s Health, National Children’s Cancer Surveillance Center, Pediatric Surgery Branch of Chinese Medical Association, et al. Expert consensus on thyroid nodules and differentiated thyroid cancer for Chinese children[J]. Chin J Appl Clin Pediatr, 2020, 35(20): 1521-1530.
[3]	王嘉丽, 任潇亚, 倪鑫, 等. 儿童甲状腺癌62例临床分析[J]. 中华儿科杂志, 2018, 56(8): 597-600.
	WANG J L, REN X Y, NI X, et al. Clinical analysis of thyroid cancer in 62 children[J]. Chin J Pediatr, 2018, 56(8): 597-600.
[4]	RIVKEES S A, MAZZAFERRI E L, VERBURG F A, et al. The treatment of differentiated thyroid cancer in children: emphasis on surgical approach and radioactive iodine therapy[J]. Endocr Rev, 2011, 32(6): 798-826. doi: 10.1210/er.2011-0011
[5]	HAUGEN B R, ALEXANDER E K, BIBLE K C, et al. 2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer[J]. Thyroid, 2016, 26(1): 1-133. doi: 10.1089/thy.2015.0020
[6]	TESSLER F N, MIDDLETON W D, GRANT E G, et al. ACR Thyroid Imaging, Reporting and Data System (TI-RADS): white paper of the ACR TI-RADS committee[J]. J Am Coll Radiol, 2017, 14(5): 587-595. doi: 10.1016/j.jacr.2017.01.046
[7]	KIM P H, YOON H M, HWANG J, et al. Diagnostic performance of adult-based ATA and ACR TI-RADS ultrasound risk stratification systems in pediatric thyroid nodules: a systematic review and meta-analysis[J]. Eur Radiol, 2021, 31(10): 7450-7463. doi: 10.1007/s00330-021-07908-8
[8]	PICCARDO A, FIZ F, BOTTONI G, et al. Facing thyroid nodules in paediatric patients previously treated with radiotherapy for non-thyroidal cancers: are adult ultrasound risk stratification systems reliable?[J]. Cancers, 2021, 13(18): 4692. doi: 10.3390/cancers13184692
[9]	LIM-DUNHAM J E. Ultrasound guidelines for pediatric thyroid nodules: proceeding with caution[J]. Pediatr Radiol, 2019, 49(7): 851-853. doi: 10.1007/s00247-019-04391-8
[10]	LIM-DUNHAM J E, TOSLAK I E, REITER M P, et al. Assessment of the American College of Radiology Thyroid Imaging Reporting and Data System for thyroid nodule malignancy risk stratification in a pediatric population[J]. Am J Roentgenol, 2019, 212(1): 188-194. doi: 10.2214/AJR.18.20099
[11]	GUPTA A, LY S, CASTRONEVES L A, et al. A standardized assessment of thyroid nodules in children confirms higher cancer prevalence than in adults[J]. J Clin Endocrinol Metab, 2013, 98(8): 3238-3245. doi: 10.1210/jc.2013-1796
[12]	DERMODY S, WALLS A, HARLEY E H Jr. Pediatric thyroid cancer: an update from the SEER database 2007-2012[J]. Int J Pediatr Otorhinolaryngol, 2016, 89: 121-126. doi: 10.1016/j.ijporl.2016.08.005
[13]	LIU C X, XIAO C, CHEN J J, et al. Risk factor analysis for predicting cervical lymph node metastasis in papillary thyroid carcinoma: a study of 966 patients[J]. BMC Cancer, 2019, 19(1): 622. doi: 10.1186/s12885-019-5835-6
[14]	LU Z H, MU Y M, ZHU H Q, et al. Clinical value of using ultrasound to assess calcification patterns in thyroid nodules[J]. World J Surg, 2011, 35(1): 122-127. doi: 10.1007/s00268-010-0827-3
[15]	POLAT Y D, ÖZTÜRK V S, ERSOZ N, et al. Is thyroid imaging reporting and data system useful as an adult ultrasonographic malignancy risk stratification method ın pediatric thyroid nodules?[J]. J Med Ultrasound, 2019, 27(3): 141-145.
[16]	陈继来. 未成年人与成年人甲状腺癌超声特征差异性分析[J]. 临床军医杂志, 2018, 46(2): 221-223.
	CHEN J L. Analysis on the difference of ultrasonic characteristics between minors and adults with thyroid cancer[J]. Clin J Med Off, 2018, 46(2): 221-223.
[17]	AL NOFAL A, GIONFRIDDO M R, JAVED A, et al. Accuracy of thyroid nodule sonography for the detection of thyroid cancer in children: systematic review and meta-analysis[J]. Clin Endocrinol (Oxf), 2016, 84(3): 423-430. doi: 10.1111/cen.12786
[18]	付鹏, 陈文, 崔立刚, 等. 不同纵横比算法对预测甲状腺结节良恶性风险的价值[J]. 中国医学科学院学报, 2019, 41(5): 663-666.
	FU P, CHEN W, CUI L G, et al. Values of longitudinal and transverse plane ratios in predicting the risk of malignant thyroid nodules[J]. Acta Acad Med Sin, 2019, 41(5): 663-666.
[19]	席雪华. 甲状腺结节超声多模态成像价值研究[D]. 北京: 北京协和医学院, 2019.
	XI X H. Evaluated the value of ultrasonographic multimodality diagnostic model of thyroid nodules[D]. Beijing: Peking Union Medical College, 2019.
[20]	LIM-DUNHAM J E, ERDEM TOSLAK I, ALSABBAN K, et al. Ultrasound risk stratification for malignancy using the 2015 American Thyroid Association management guidelines for children with thyroid nodules and differentiated thyroid cancer[J]. Pediatr Radiol, 2017, 47(4): 429-436. doi: 10.1007/s00247-017-3780-6
[21]	PARK C S, KIM S H, JUNG S L, et al. Observer variability in the sonographic evaluation of thyroid nodules[J]. J Clin Ultrasound, 2010, 38(6): 287-293.

Clinical characteristic	Pathological result		P value
Clinical characteristic	Benign (N=63)	Malignant (N=113)	P value
Age/year median (range)	16 (10-18)	16 (6-18)	0.885
Gender n			0.035
Male	7	28
Female	43	66
Number of nodules n			0.053
Single	22	57
Multiple	28	34
Diffuse calcification	0	3
Diffuse glandular lesions n (%)			0.095
With	23 (36.5)	56 (49.6)
Without	40 (63.5)	57 (50.4)
Size n (%)			0.001
<1 cm	11 (17.5)	27 (23.9)
1-2 cm	9 (14.3)	38 (33.6)
2-3 cm	14 (22.2)	27 (23.9)
>3 cm	29 (46.0)	21 (18.6)

Ultrasonic characteristic	Pathological result		P value
Ultrasonic characteristic	Benign (N=63)	Malignant (N=113)	P value
Composition			<0.010
Solid	18 (28.6)	101 (89.4)
Mixed cystic	45 (71.4)	12 (10.6)
Cystic	0 (0.0)	0 (0.0)
Echogenicity			<0.010
Hyperechoic	15 (23.8)	4 (3.5)
Isoechoic	35 (55.6)	16 (14.2)
Hypoechoic	13 (20.6)	93 (82.3)
Very hypoechoic	0 (0.0)	0 (0.0)
Anechoic	0 (0.0)	0 (0.0)
Margin			<0.010
Rule	57 (90.5)	18 (15.9)
Irregular	6 (9.5)	95 (84.1)
Calcification			<0.010
Microcalcification	0 (0.0)	83 (73.5)
Other calcification	2 (3.2)	1 (0.9)
None	61 (96.8)	29 (25.7)
Shape			<0.026
Wider-than-tall	62 (98.4)	95 (84.1)
Taller-than-wide	1 (1.6)	18 (15.9)

Independent variable	Partial regression coefficient	Standard error	Wals	P value	OR	95% CI
Composition (solid)	1.782	0.594	8.987	0.003	5.942	1.853-19.05
Echogenicity (hypoechoic)	1.414	0.612	5.333	0.021	4.113	1.239-13.66
Margin (irregular)	1.779	0.669	7.067	0.008	5.924	1.596-21.994
Calcification (with calcification)	3.051	0.905	11.356	0.001	21.129	3.584-124.579
Shape (taller-than-wide)	-0.842	1.252	0.452	0.501	0.431	0.037-5.015
Constant	-2.873	0.552	27.071	0.000	0.057	NA

Item	Sensitivity	Specificity	PPV	NPV	Accuracy	Jordan index
Training set
ACR TI-RADS	75.2	96.8	97.7	68.5	83.0	0.72
The graded diagnostic model	83.2	95.2	96.9	75.9	87.5	0.78
Testing set
ACR TI-RADS	73.1	100.0	100.0	68.2	82.9	0.73
The graded diagnostic model	96.2	86.7	92.6	92.9	92.7	0.83