Clinic-Based Ultrasound Can Predict Malignancy in Pediatric Thyroid Nodules

Abstract

Background:

Thyroid nodules in pediatric patients may carry a greater risk for malignancy than in adults. Most nodules >1 cm in patients ≤21 years of age may require thyroidectomy for definitive diagnosis and treatment. Although clinic-based ultrasound (CBUS) has been shown useful in the evaluation of thyroid nodules in adults, its utility in evaluating nodules in the pediatric population remains unclear.

Methods:

Prospectively collected data regarding 50 patients ≤21 years who underwent preoperative CBUS and initial thyroidectomy at a single institution were retrospectively reviewed. All CBUS were performed by endocrine surgeons certified in basic and cervical ultrasonography. Preoperative CBUS characteristics of pediatric thyroid nodules were analyzed with respect to final pathology.

Results:

Of the 50 patients ≤21 years of age who underwent surgical resection for a dominant thyroid nodule, there were 45 females and 5 males with an average age of 17.4 years (range: 10–21 years). On univariate analysis, microcalcifications (p<0.001), abnormal lymph nodes (p<0.001), and dimensions taller more than wide (p=0.033) were individual CBUS characteristics predictive of thyroid malignancy. All nine patients with abnormal lymph nodes on CBUS had malignant disease on final pathology. Multiple thyroid nodules, a cystic component, and echogenicity did not predict malignancy; regular borders trended toward predicting a benign nodule (p=0.066). When malignant ultrasound features were considered (i.e., hypoechoic, irregular borders, microcalcifications, abnormal lymph nodes, and shape taller more than wide), having one malignant feature predicted malignancy with an odds ratio of 2.0 while having ≥2 features held even greater significance (p=0.004, OR 4.0). All patients with ≥3 malignant ultrasound features had thyroid cancer on final pathology.

Conclusion:

CBUS is a useful diagnostic modality in determining malignancy status of thyroid nodules in patients ≤21 years of age. CBUS should be employed as part of an initial assessment of any pediatric patient presenting with thyroid nodules to help further guide management and treatment.

Introduction

T hyroid cancer ranks among the top five cancers in adolescents, and it is the most common carcinoma in the pediatric population (1,2). In this age group, the incidence of thyroid cancer has had one of the greatest percent increases over the last 40 years, with close to 10% of all thyroid carcinomas occurring in patients ≤21 years of age (1 –6). The malignancy rate of a given thyroid nodule in this pediatric age group is four to five times that of an adult, and thus, whether discovered on physical exam or incidentally on an imaging study, can raise a significant concern for patients, parents, and clinicians (7 –10).

Over the past decade, clinic-based neck ultrasound (CBUS) has become an extension of the physical exam for many clinicians, especially among endocrine surgeons, as a cost-effective and accurate means of evaluating thyroid nodules (11,12). The potential inaccuracy of fine-needle aspiration (FNA) may make CBUS, with its high reliability, an integral component of the diagnostic evaluation of thyroid nodules in pediatric patients. Interestingly, neck ultrasound has actually been used since the 1980s to evaluate thyroid nodules in pediatric and adolescent patients with a history of head and neck radiation therapy or exposure (13 –15).

While FNA is safe and can be easily performed in pediatric patients, other factors should also be taken into consideration (16,17). Some authors report a high sensitivity and specificity for FNA in this pediatric population, whereas, others suggest the opposite; one group had a reported FNA sensitivity as low as 70% (7,15,18 –22). Additionally, multiple studies have described a malignancy rate of up to 50% in patients of all ages with “indeterminate” FNA biopsies with malignancy determined only on final pathology (17,19,23). Clinicians must also take into consideration the potential inability or immaturity of younger children to sit still for the procedure, an increased sensitivity to or fear of needle sticks, and a smaller space to maneuver both the ultrasound and FNA needle with precision. Although CBUS is a proven, useful tool for evaluation of thyroid nodules in adults, its utility in evaluating nodules in the pediatric population remains unclear. The goal of this study was to determine whether CBUS can help further guide surgical management and treatment of thyroid nodules in the pediatric population.

Methods

Clinical and pathologic data were prospectively collected in an institutional review board (IRB)–approved database at the University of Miami Health System for 50 consecutive patients, ≤21 years of age with solitary or dominant thyroid nodules, who underwent preoperative CBUS and initial thyroidectomy from March 2002 to March 2011. From this database, a retrospective review of the preoperative CBUS characteristics of pediatric thyroid nodules were analyzed with regard to final pathology.

All preoperative CBUS were performed by endocrine surgeons certified in basic and cervical ultrasonography by the American College of Surgeons using high-frequency linear array transducers 7.5–13 MHz. Prospectively collected CBUS characteristics of thyroid nodules included size (width, length, and height), echogenicity (hypoechoic, isoechoic, and hyperechoic), borders (regular and irregular), calcifications (micro vs. coarse/none), cystic component (vs. solid), shape (taller vs. wider on transverse view), number of nodules (single vs. multiple), location (unilobar vs. bilobar), and presence of abnormal lymph nodes (increased size, rounded shape, irregular margins, internal calcifications, loss of echogenic fatty hilum, and cystic replacement). Demographics collected included age, gender, ethnicity, or ancestral origin as reported by the patient or parent at the hospital, and body mass index.

Patients were subdivided into two groups based on final pathology: patients with benign thyroid disease (n=22) and those with thyroid malignancy (n=28). Indications for thyroidectomy included an FNA result of cancer, suspicious for cancer or indeterminate, nodule size > 4 cm, compression symptoms, hyperthyroidism with a toxic nodule, or suspicious nodule characteristics on CBUS in conjunction with a fixed nodule, suspicious lymph nodes, or a history of head and neck radiation therapy or exposure.

All data were analyzed using SPSS 18.0 (IBM Co., Somers, NY). To demonstrate the utility of CBUS for prediction of benign and malignant disease, data were analyzed in a case–control mode. Univariate analysis was performed for all potential predictors of malignancy using a Student's t-test for continuous data and chi-square analysis for categorical data. Following univariate analysis, all previously identified predictors of thyroid malignancy in adult patients were identified in this pediatric population. Ultrasound features of thyroid malignancy identified for analysis included: hypo- vs. iso- or hyperechogenicity, irregular versus regular borders, presence of microcalcifications, shape taller more than wide, and presence of abnormal lymph nodes. Patients were then scored for the number of classically malignant and benign ultrasound features identified. Odds ratios for malignancy/benignity were calculated based on the number of ultrasound characteristics present.

Results

Of 50 patients ≤21 years of age who underwent surgical resection for a dominant thyroid nodule, 45 were females and 5 were males with an average age of 17.4 years (range: 10–21 years). There were 25 patients (50%) who were ≤18 years of age. Of this study group, 22 patients were found to have benign thyroid disease while 28 patients had thyroid malignancy. On final pathology, 16 patients had papillary thyroid cancer (PTC), 10 had a follicular variant of papillary thyroid cancer (FVPTC), and 2 had medullary thyroid cancer (MTC). One patient with MTC was a 10-year-old female with MEN 2A and a 0.7 cm nodule with three malignant preoperative surgeon-performed ultrasound features, whereas the second patient with MTC was a 16-year-old female with a 1.7 cm nodule that had an equivocal surgeon-performed ultrasound and indeterminate FNA features. Standard demographics including age, gender, ethnicity, ancestral origin, or body mass index did not significantly differ between the groups with benign and malignant disease. On univariate analysis, microcalcifications (p<0.001), abnormal lymph nodes (p<0.001), and dimensions taller more than wide (p=0.033) were individual CBUS characteristics predictive of thyroid malignancy (Table 1).

Table 1.

Demographic and Clinic-Based Ultrasound Predictors of Thyroid Malignancy

	Malignant (%) (n=28)	Benign (%) (n=22)	p-Value
Demographics
Age (years)	17.2	17.7	0.521^a
Male gender	3.5	18.2	0.155^b
Caucasian	46.4	47.6	0.817^b
CBUS features
Single nodule	64.3	63.6	0.597^b
Cystic component	40.7	45.5	0.477^c
Microcalcifications	81.5	31.8	<0.001 ^c
Regular borders	51.9	77.3	0.066 ^c
Shape taller more than wide	18.5	0	0.033 ^c
Iso- or hyperechoic	40.7	54.5	0.336^c
Hypoechoic	51.9	36.4	0.335^c
Abnormal lymph nodes	59.1	0	0.018 ^b

Bold values indicate statistical significance (p>0.05).

Student's t-test.

Fischer's exact test.

Pearson chi-square test.

CBUS, clinic-based ultrasound.

When ultrasound features of malignant thyroid nodules were present (i.e., hypoechoic, irregular borders, microcalcifications, shape taller more than wide, and abnormal lymph nodes), preoperative CBUS was effective in predicting thyroid cancer. Overall, nine patients had three or more malignant thyroid nodule features on CBUS. When abnormal lymph nodes were present, there were 13 patients who had 3 or more malignant features on CBUS; all had thyroid malignancy on final pathology. Additionally, nine patients presented with abnormal lymph nodes on CBUS and all had malignant disease on final pathology (five were included in the nine with malignant nodule features only); one-third also had clinically palpable lymphadenopathy and two-thirds of these patients with abnormal lymphadenopathy were ≤18 years of age. Using the criteria of three malignant ultrasound features, CBUS in patients ≤21 years had a true positive rate of 100% and a false negative rate of 12.3% for predicting thyroid malignancy. The false negative patient had a 4.2-cm nodule with three benign features on CBUS that had FVPTC on final pathology. The presence of two malignant features represented an odds ratio of 4.0 (p=0.004), whereas the presence of only one malignant feature had an odds ratio of 2.0 (p=0.35).

When evaluated in the context of predicting benignity, CBUS that demonstrated three or more benign ultrasound features (i.e., iso- or hyperechoic, regular borders, no microcalcifications, and cystic component) supported a diagnosis of a benign thyroid lesion (p =< 0.001). Of the entire cohort, 18 patients demonstrated 3 or more benign features on CBUS; 14 had benign lesions and 4 were found to have a malignancy on final pathology. Of these four patients, two had FVPTC measuring >5 cm, and two had incidental cancers (one FVPTC and one PTC) found outside of the dominant index nodule. As two of four were actually incidental cancers, only two of 18 patients truly had malignant pathology on the final path of their dominant nodule. When considering individual ultrasound characteristics, however, only the presence of regular borders trended toward predicting benignity (p=0.066).

Discussion

CBUS is a noninvasive, easy to use, real-time imaging test that has become an integral part of the physical exam for evaluating thyroid nodules in adult patients. CBUS provides the most thorough assessment of nodule characteristics with simultaneous appraisal of the contralateral thyroid lobe and cervical lymph nodes. The information obtained with CBUS allows a trained clinician to accurately evaluate a thyroid nodule, and for a trained surgeon, not only direct operative planning including its extent, but also to have a detailed, informative discussion with the patient preoperatively.

This study is the first to demonstrate that CBUS assessment of thyroid nodules, previously described with significant power in adults, can be generally applied to pediatric patients. Certain combinations of ultrasound features have a predictive value for determining the benign or malignant nature of thyroid nodules (24,25). CBUS features highly suggestive of malignancy may help clinicians plan for total thyroidectomy with or without central neck dissection, or in the case of a very small nodule (<1 cm), advise the patient to undergo an FNA biopsy or diagnostic thyroid lobectomy.

With the growing number of thyroid nodules being discovered in younger patients, many patients ≤21 years of age with a thyroid nodule > 1 cm often undergo operations for definitive diagnosis and treatment (4,26). Though the prevalence of palpable thyroid nodules in patients ≤21 years of age is only 0.05%–1.8%, the true incidence may be greater than 13% based on autopsy studies (3,9,10,26,27 –29). Unlike the adult population, where only 5% of thyroid nodules are malignant, ∼25% of thyroid nodules in the pediatric population are malignant, with rates up to 50% in patients that present with suspicious-appearing lymph nodes (6,30).

In the adolescent population, the main risk factor for thyroid malignancy is a history of head and neck radiation therapy or exposure that is dose- and age-dependent (31,32). The most common cancer histories associated with thyroid cancer in this younger patient group are Hodgkin's disease (HD), non-Hodgkin's lymphoma (NHL), leukemia, retinoblastoma, or the patient having received radiation therapy in preparation for a bone marrow transplant. In a study of 16,500 leukemia survivors, thyroid carcinoma was the most common second malignancy in patients with a history of HD and NHL, and the third most common after leukemia (33). These patients represent a unique group that have classically been monitored with thyroid ultrasound, but solely for determining the presence or absence of a nodule or features of thyroiditis. One value of the authors' current study, specifically for clinicians caring for such cancer survivors, may be an improved utilization and understanding of CBUS as a monitoring tool already in place as standard practice in other areas of cancer surveillance.

In the largest study of ultrasound use in pediatric thyroid nodules to date, which did not utilize CBUS, pediatric patients with malignant thyroid nodules were more likely to have microcalcifications, lymphadenopathy, and altered nodular vascular pattern (8). However, each characteristic was only present in 47%–73% of patients respectively. Further, in this series, a subset of patients deemed to have benign ultrasound findings (i.e., nodules with regular borders, normal vasculature, no calcifications, and no suspicious lymph nodes) were followed up without any change in examination for at least one year. While the results of the authors' present study complement these aforementioned findings, further analyses with greater power are needed before diagnostic lobectomies can be safely foregone in this population based solely on CBUS features. However, if CBUS features are entirely benign and the patient has no other surgical indications including family history, obstructive symptoms, large >4 cm lesion, cosmetic concerns, and possibly personal history of head and neck radiation, the results of these two studies suggest that the clinician, in discussion with the patient and the family, may opt for further monitoring with repeat ultrasound at 6 months and then yearly thereafter for 3–5 years as recommended by current ATA guidelines (34). FNA or diagnostic thyroid lobectomy with isthmusectomy should be considered if ultrasound characteristics change (growth of nodule, increase in solid component of heterogeneous nodule, and presence of new abnormal features) or new clinical factors (obstructive symptoms and firmness to palpation) develop during the elected observation period.

The authors believe that ultrasound should be used routinely in evaluating thyroid nodules in pediatric patients. This viewpoint complements the few other reports in the literature suggesting its advantage in this patient population (35 –37). Additionally, the authors suggest ultrasound be routinely used in pediatric patients to identify those patients who require further FNA diagnosis. Although the authors are not advocating for the elimination of FNA in the pediatric population, the necessity of performing routine FNA in every pediatric patient should be reconsidered.

There are two main limitations to this study. First, there is a selection bias of thyroid nodules that were surgically resected and therefore had final pathology to review. Since patient selection was from a surgeon's practice, most patients were referred for surgical resection after undergoing evaluation for their thyroid nodule(s) by their pediatrician and/or pediatric endocrinologist, and therefore, were either large nodules or possibly worrisome for thyroid cancer by FNA and/or ultrasound. Second, due to the overall low incidence of detected thyroid nodules in this age group, the study's sample size was fairly small. A multi-institutional study would be needed to overcome this particular limitation.

In conclusion, CBUS may be useful in determining malignancy status of thyroid nodules in patients who are ≤21 years of age. CBUS should therefore be employed as part of a standard, initial assessment of any pediatric patient presenting with thyroid nodules to help guide further management and treatment.

Footnotes

Disclosure Statement

The authors declare that no competing financial interests exist.

References

Ries

LAG

, Harkins

, Krapcho

. 2006. SEER Cancer Statistic Review, 1975–2003. NCI: Bethesda, MD http://seer.cancer.gov/csr/1975–2003. 2011 September 1.

Waguespack

, Wells

, Ross

, Bleyer

. 2006. Thyroid cancer SEER AYA monograph. Bleyer

, O'Leary

, Barr

et al. Cancer Epidemiology in Older Adolescents and Young Adults 15–29 Years of Age, Including SEER Incidence and Survival 1975–2000NIH Pub No. 06–5767NCI: Bethesda, MD, 143–154.

American Cancer Society 2009 Cancer Facts and Figures. American Cancer Society: Atlanta, GA.

Machens

, Lorenz

, Thanh

, Lorenz

, Schmeil

, Stratmann

, Sekulla

, Brauckhoff

, Dralle

. 2010. Papillary thyroid cancer in children and adolescents does not differ in growth pattern and metastatic behavior. J Pediatrics, 157:648–652.

Hogan

, Zhuge

, Perez

, Koniaris

, Lew

, Sola

. 2009. Pediatric thyroid carcinoma: Incidence and outcomes in 1753 patients. J Surg Research, 156:167–172.

Pacini

. 2002. Thyroid cancer in children and adolescents. J Endocrinol Invest, 25:572–573.

The Canadian Pediatric Thyroid Nodule Study Group. 2008. The Canadian Pediatric Nodule Study: an evaluation of current management practices. J Pediatric Surg, 43:826–830.

Corrias

, Mussa

, Baronio

, Arrigo

, Salerno

, Segni

, Vigone

, Gastaldi

, Zirilli

, Tuli

, Beccaria

, Iughetti

, Einaudi

, Weber

, De Luca

, Cassio

. Study Group for Thyroid Diseases of Italian Society for Pediatric Endocrinology and Diabetology (SIEDP/ISPED) 2010 Diagnostic features of thyroid nodules in pediatrics. Arch Pediatr Adolesc Med, 164:714–719.

Wiersinga

. 2007. Management of thyroid nodules in children and adolescents. Hormones, 6:194–199.

10.

Niedziela

. 2006. Pathogenesis, diagnosis, and management of thyroid nodules in children. Endocr Relat Cancer, 13:427–453.

11.

Solorzano

, Carneiro

, Ramirez

, Lee

, Irvin

3rd . 2004. Surgeon-performed ultrasound in the management of thyroid malignancy. Am Surg, 70:576–580.

12.

Milas

, Stephen

, Berber

, Wagner

, Miskulin

, Siperstein

. 2005. Ultrasonography for the endocrine surgeon: a valuable clinical tool that enhances diagnostic and therapeutic outcomes. Surgery, 138:1193–1200.

13.

Crom

, Kaste

, Tubergen

, Greenwald

, Sharp

, Hudson

. 1997. Ultrasonography for thyroid screening after head and neck irradiation in childhood cancer survivors. Med Pediatr Oncol, 28:15–21.

14.

Solt

, Gaitini

, Pery

, Hochberg

, Stein

, Arush

. 2001. Comparing thyroid ultrasonography to thyroid function in long-term survivors of childhood lymphoma. Med Pediatr Oncol, 35:35–40.

15.

Corrias

, Einaudi

, Chioboli

, Weber

, Crinò

, Andreo

, Cesaretti

, de Sanctis

, Messina

, Segni

, Cicchetti

, Vigone

, Pasquino

, Spera

, de Luca

, Mussa

, Bona

. 2001. Accuracy of fine needle aspiration biopsy of thyroid nodules in detecting malignancy in childhood: comparison with conventional clinical, laboratory, and imaging approaches. J Clin Endocrinol Metab, 86:4644–4648.

16.

Stevens

, Lee

JKP

, Sadatsafavi

, Blair

. 2009. Pediatric thyroid fine-needle aspiration cytology: a meta-analysis. J Ped Surg, 44:2184–2191.

17.

Kim

, Nam-Goong

, Gong

, Hong

, Kim

, Shong

. 2003. Postoperative findings and risk for malignancy in thyroid nodules with cytological diagnosis of the so-called “follicular neoplasm.” Korean J Intern Med, 18:94–97.

18.

Amrikachi

, Ponder

, Wheeler

, Smith

, Ramzy

. 2005. Thyroid fine-needle aspiration biopsy in children and adolescents: experience with 218 aspirates. Diagn Cytopathol, 32:189–192.

19.

Bargren

, Meyer-Rochow

, Sywak

, Delbridge

, Chen

, Sidhu

. 2010. Diagnostic utility of fine-needle aspiration cytology in pediatric differentiated thyroid cancer. World J Surg, 34:1254–1260.

20.

Lugo-Vicente

, Ortiz

, Irizarry

, Camps

, Pagán

. 1998. Pediatric thyroid nodules: management in the era of fine needle aspiration. J Pediatr Surg, 33:1302–1305.

21.

Izquierdo

, Shankar

, Kort

, Khurana

. 2009. Ultrasound-guided fine-needle aspiration in the management of thyroid nodules in children and adolescents. Thyroid, 19:703–705.

22.

Willgerodt

, Keller

, Bennek

, Emmrich

. 2006. Diagnostic value of fine-needle aspiration biopsy of thyroid nodules in children and adolescents. J Pediatr Endocrinol Metab, 19:507–515.

23.

Brooks

, Shaha

, DuMornay

, Huvos

, Zakowski

, Brennan

, Shah

. 2001. Role of fine-needle aspiration biopsy and frozen section analysis in the surgical management of thyroid tumors. Ann Surg Oncol, 8:92–100.

24.

Jabiev

, Ikeda

, Reis

, Solorzano

, Lew

. 2009. Surgeon-performed ultrasound can predict differentiated thyroid cancer in patients with solitary thyroid nodules. Ann Surg Oncol, 16:3140–3145.

25.

Goldfarb

, Gondek

, Solorzano

, Lew

. 2001. Surgeon performed ultrasound can predict benignity in thyroid nodules. Surgery, 150:436–441.

26.

Guth

, Theune

, Aberle

, Galach

, Bamberger

. 2009. Very high prevalence of thyroid nodules detected by high frequency (13 MHz) ultrasound examination. Eur J Clin Invest, 39:699–706.

27.

Dinauer

, Breuer

, Rivkees

. 2008. Differentiated thyroid cancer in children: diagnosis and management. Curr Opin Oncol, 20:59–65.

28.

Gosepath

, Spix

, Talebloo

, Blettner

, Mann

. 2007. Incidence of childhood cancer of the head and neck in Germany. Ann Oncol, 18:1716–1721.

29.

Rallison

, Dobyns

, Meikle

, Bishop

, Lyon

, Stevens

. 1991. Natural history of thyroid abnormalities: prevalence, incidence, and regression of thyroid disease in adolescents and young adults. Am J Med, 91:363–370.

30.

Hay

, Gonzalez-Losada

, Reinalda

, Honetschlager

, Richards

, Thompson

. 2010. Long-term outcome in 215 children and adolescents with papillary thyroid cancer treated during 1940 through 2008. World J Surg, 34:1192–1202.

31.

Sklar

, Whitton

, Mertens

, Stovall

, Green

, Marina

, Greffe

, Wolden

, Robison

. 2000. Abnormalities of the thyroid in survivors of Hodgkin's disease: data from the childhood cancer survivor study. J Clin Endocrinol Metab, 85:3227–3232.

32.

Tucker

, Jones

, Boice

Jr. , Robison

, Stone

, Stovall

, Jenkin

, Lubin

, Baum

, Siegel

. 1991. Therapeutic radiation at a young age is linked to secondary thyroid cancer. The Late Effects Study Group. Cancer Res, 51:2885–2888.

33.

Maule

, Scelo

, Pastore

, Brennan

, Hemminki

, Tracey

, Sankila

, Weiderpass

, Olsen

, McBride

, Brewster

, Pompe-Kirn

, Kliewer

, Chia

, Tonita

, Martos

, Jonasson

, Merletti

, Boffetta

. 2007. Risk of second malignant neoplasms after childhood leukemia and lymphoma: an international study. J Natl Cancer Inst, 99:790–800.

34.

Cooper

, Doherty

, Haugen

, Kloos

, Lee

, Mandel

, Mazzaferri

, McIver

, Sherman

, Tuttle

. 2009. American Thyroid Association (ATA) guidelines taskforce on thyroid nodules and differentiated thyroid cancer. Revised American thyroid association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid, 19:1167–1214.

35.

Corrias

, Cassio

, Weber

, Mussa

, Wasniewska

, Rapa

, Gastaldi

, Einaudi

, Baronio

, Vigone

, Messina

, Bal

, Bona

, de Sanctis

. 2008. Thyroid nodules and cancer in children and adolescents affected by autoimmune thyroiditis. Arch Pediatr Adolesc Med, 162:526–531.

36.

Stulak

, Grant

, Farley

, Thompson

, van Heerden

, Hay

, Reading

, Charboneau

. 2006. Value of preoperative ultrasonography in the surgical management of initial and reoperative papillary thyroid cancer. Arch Surg, 141:489–494.

37.

Wada

, Sugino

, Mimura

, Nagahama

, Kitagawa

, Shibuya

, Ohkuwa

, Nakayama

, Hirakawa

, Yukawa

, Rino

, Masuda

, Ito

. 2009. Treatment strategy of papillary thyroid carcinoma in children and adolescents: clinical significance of the initial nodal manifestation. Ann Surg Oncol, 16:3442–2449.