Use of ultrasound in childhood thyroid disorders

Use of ultrasound in childhood thyroid disorders Thyroid ultrasound findings in 55 patients aged 6 days to 19 years were reviewed to assess the importance o f this technique in evaluating childhood thyroid disorders. Findings were correlated with the available clinical, radionuclide, and pathologic data. In 25 patients with diffuse thyroid lesions (thyroiditis, Graves disease, euthyroid goiter, iodine-induced goiter, goitrous cretinism), ultrasound revealed only homogeneous thyroid enlargement or a nonspecific patchy echo pattern. In two infants with poorly visualized glands on radionuclide scans, ultrasound confirmed the presence o f anatomically normal thyroid tissue. Twenty patients had at least one focal thyroid lesion seen by ultrasound, including nodules not detected by palpation in one child or by technetium scan in three. Thyroid malignancies were found in four o f 13 patients with solitary thyroid nodules, occurring in two o f four patients with echogenic nodules, two o f five children with complex lesions, and none o f four with echofree nodules. Thyroid ultrasound is a sensitive, noninvasive means o f evaluating thyroid anatomy. Because it can detect thyroid tissue in the neck not seen on radionuclide scan in patients at all ages and can define the number and consistency o f focal lesions, this technique offers definite advantages in assessing a variety o f childhood thyroid disorders. (J PEDIATR 103:547, 1983)

Laura K. Bachrach, M.D., F.R.C.P.(C), Denis Daneman, M.B., B.Ch., F.R.C.P.(C), Alan Daneman, M.B., B.Ch., F.R.C.P.(C), and David J. Martin, M.B., B.S., F.R.C.P.(C) T o r o n t o , Ont., C a n a d a

IN ADULTS, THYROID ULTRASOUND is accepted as the most sensitive noninvasive method of defining the internal anatomy of the thyroid gland? The technique has been used successfully to differentiate solid and mixed thyroid nodules from purely cystic lesions, to distinguish solitary from multiple nodules, to measure thyroid gland size, and to differentiate extrathyroidal from intrinsic lesions of the thyroid gland. 1-4 The diagnostic value of thyroid ultrasound studies in childhood thyroid disorders has not been reviewed. Therefore, we analyzed our experience with 55 children referred for thyroid ultrasound examination at The Hospital for Sick Children, Toronto. Our aims were to describe the ultrasound appearance of a variety of thyroid abnormali-

From the Departments o f Paediatrics (Endocrine Div&ion) and Radiology, The Hospital for Sick Children and University o f Toronto. Presented in part at the 58th Meeting o f the American Thyroid Association, Quebec City, September 23, 1982. Reprint requests: Dr. Denis Daneman, The Hospital ['or Sick Children, 555 University Ave., Toronto, Ont., Canada M5G 1X8.

ties and to correlate ultrasound findings with the available clinical, radionuclide, and pathologic data. PATIENTS

A N D METHODS

We reviewed the charts of 43 girls and 12 boys who underwent thyroid ultrasound examination between August 1981 and December 1982. At the time of ultrasound examination, patients ranged in age from 6 days to 19 years (mean 12.4 years). The clinical indications for thyroid ultrasound study were as follows: 26 patients had diffuse thyroid lesions by palpation with abnormal thyroid gland size or consistency; 21 had palpable thyroid nodules; four infants with congenital hypothyroidism or poorly visualized thyroid glands on radionuclide scan were studied to assess the presence of thyroid tissue; and four children without thyroid disease underwent examination as part of research protocols. Two of the four patients without thyroid disease were to begin treatment with amiodarone, and the other two had clinical features characteristic of polyostotic fibrous dysplasia (McCune-Albright syndrome). Amiodarone therapy and McCune-Albright syndrome have both been associated with an increased incidence of thyroid dysfunction? ,6

The Journal o f P E D I A T R I C S

547

548

Bachrach et al.

The Journal o f Pediatrics October 1983

| Fig. 1. A, Normal transverse thyroid ultrasound scan. Note normal, homogeneous, finely stippled echo pattern of triangular thyroid lobes. B, Normal transverse scan in neonate. Thyroid lobes (arrows) are easily visualized, i, Isthmus; j, jugular vein; T, trachea; s, sternomastoid muscle.

Table I. Correlation of thyroid ultrasound and clinical findings in 55 children Ultrasound findings

Normal thyroid gland

No thyroid tissue in neck Diffuse lesions

Multiple focal lesions Solitary nodules Echogenic Echofree Complex

n '

8

2 25

7 13 4 4 5

Clinical diagnosis

n

Normal

4

Hypothyroidism (antibody negative) ? Palpable nodule Thyroid gland not seen on radionuclide scan congenital hypothyroidism Hashimoto thyroiditis Graves disease Other Multinodular Multicystic

1

15 6 4 2 5

Solitary nodules

13

1 2 2

The ultrasound examination was performed in all patients without sedation or other preparation. Patients were placed in the supine position with the neck hyperextended. Thyroid ultrasound was performed in longitudinal (parasagittal) and transverse directions, with additional

Fig. 2. A, Transverse thyroid ultrasound scan of lower neck in neonate, showing absence of thyroid tissue (A) lateral to the trachea (T). B, Lateral view of technetium 99m pertechnetate scan showing activity at base of tongue (arrow). Diagnosis is lingual thyroid gland. C, Chin marker. oblique views or transducer angulation if required. A Unirad EDP 1000 static B scanner with 5 or 7 m H z transducers was used for all the studies. Ultrasound studies were interpreted by two of us (A.D., D.J.M.) who knew the clinical history but did not know the radionuclide scan findings. At least one radionuclide scan was done in 32 of the patients. The technetium 99m-pertechnetate scans (n = 26) were performed as previously described. 7 The patients were sca~med using a g a m m a counter fitted with a pinhole collimator 20 to 30 minutes after injection of the radionuclide. Iodine 123 or 131 scans (n = 8) were also performed using similar apparatus. 7,8 Needle aspiration 2 was done i n nine children. Brief general anesthesia was required in one boy because of extreme anxiety, but the remaining procedures were perf o r g e d using only local anesthesia. With the patient supine and the neck extended, the skin was washed with povidone iodine and alcohol. An 18- to 21-gauge needle fitted to a syringe was used to aspirate cystic lesions, and a 25- to 27-gauge needle for solid nodules. Aspirated material was smeared immediately onto glass slides and fixed with Cytoprep spray (Winley-Morris, Montreal). Specimens were stained with a modification of the Papanicolaou stain.

Volume 103 Number 4

Ultrasound in thyroid disorders

549

Fig. 4. Longitudinal thyroid ultrasound scan in patient with McCune-Albright syndrome and thyrotoxicosis. Right lobe (between markers) is replaced by multiple cysts of varying sizes.

Fig. 3. Transverse thyroid ultrasound scan in child with Hashi moto thyroiditis. Enlarged thyroid gland (arrows) shows diffuse, coarse, nonhomogeneous echo pattern. T, Trachea; i, isthmus.

Table II. ComParison of thyroid ultrasound, radionuclide, and pathologic findings in 13 children

with solitary thyroid nodules Radionuclide findings (n) Ultrasound findings

Function of nodule

Isotope

n

99mTc

4

Cold

Echofree (cystic)

1311

Complex (mixed)

1231or 13tI 99mTc i31I 99mTc 99roTe

1 2 1 1 1 2

Hot Cold Warm Hot Warm Cold

13~I

1

Cold

Echogenic (solid)

RESULTS The ultrasound findings in the 55 Children studied are summarized in Table I. Normal thyroid gland. The thyroid gland appeared normal in eight patients: four who had no clinical evidence of thyroid disease (Fig. 1, A), one with acquired hypothyroidism, one who had a questionable nodule by palpation, and two infants in whom the thyroid gland was visualized poorly or not at all on technetium scan. In the two infants, ultrasound showed thyroid glands that were normal in size, location, and homogeneity (Fig~ 1, B). One infant was biochemically euthyroid, and the other had low thyroxine and elevated thyroid-stimulating hormone values at the time of scanning. In two infants found to have a lingual thyroid on technetium scan, no paratracheal thyroid tissue was detected by ultrasound (Fig. 2, A and B). Diffuse thyroid lesions. Diffuse lesions were demonstrated in 25 patients, including 15 with Hashimoto thyroiditis, Six with Graves disease, two with euthyroid goiter (thyroid antibody negative), and one each with iodine-induced goiter and goitrous cretinism. The ultra-

Pathologic findings

4

Papillary carcinoma Dermoid cyst Not done NOt done Benign Benign Follicular carcinoma Benign Follicular carcinoma Benign Benign

2 1 1 2

1 1

sound scans showed nonhomogeneous echo patterns without definite nodules in 47% of the patients with thyroiditis (Fig. 3) as well as in the patient with iodine-induced goiter. Thyroid ultrasounds appeared homogeneous in the remaining patients with thyroiditis and in all patients with Graves disease except one, who had small areas of decreased echogenicity one year after radioiodine therapy. Radionuclide scans of the thyroid performed in eight of the patients with thyroiditis were compared with ultrasound findings. In one g i r l the presence of nodules, suspected on technetium scanning, was not confirmed by ultrasound. Multiple focal lesions. Seven children had multiple focal lesions detected on ultrasound examination; in only one of these, a girl with thyroiditis, none of the multiple cysts had been detected on clinical examination. Multiple focal lesions were found by ultrasound examination in an additional two patients thought to have solitary lesions on palpation. Two patients with multiple thyroid cysts were thyrotoxic and had the typical skin and bony lesions of McCune-Albright syndrome (Fig. 4)~ Thyroid pathologic findings were benign in all six children who underwent

550

Bachrach et al.

The Journal of Pediatrics October 1983

| Fig. 5, Solitary thyroid nodules. A, Longitudinal thyroid ultrasound scan showing well-defined, echogenic mass (M) in lower pole of right lobe. Note difference in echo pattern from normal thyroid in upper pole (t). Diagnosis is papillary carcinoma. B Longitudinal scan showing echofree lesion (c) in rigl~t lobe. Diagnosis is benign cyst. needle aspiration (n = 4) and/or surgery (n = 3). In the five children with multiple lesions who had technetium scans, ultrasound detected cysts not apparent on radionuclide examination in three. Solitary thyroid nodules. Ultrasound examination confirmed the presence of solitary thyroid nodules in 13 children, all of whom were euthyroid (Table II). The nodules in four patients had an echogenic appearance (Fig. 5, A), and all were "cold" on technetium thyroid scanning. At surgery, two of these patients were found to have papillary carcinoma, and a third child had a dermoid cyst. The fourth patient refused further intervention. Four patients had echofree nodules (Fig. 5, B). Needle aspiration of these cystic nodules yielded benign cytologic findings in the three patients with "cold" or "warm" lesions on radionuelide scan. In the fourth patient, a small, "hot" nodule was detected by td~l~etium scanning two years before ultrasound examination. She remains euthyroid and her solitary 8 X 5 mm cyst has not been biopsied. Five children had solitary nodules with a complex (mixed) echo pattern on ultrasound. Three of these lesions

9

Fig. 6. A, Transverse thyroid ultrasound scan showing mass (arrowheads) in right lobe, with mixed echogenic and echofree areas. Left lobe (arrows) is normal. T, Trachea. B, 131I scan showing uptake only in mass on right. Diagnosis is autonomously functioning follicular carcinoma. were "cold," one was "hot," and one "warm" by radionuclide scan. Needle aspiration or surgery in three patients revealed benign pathologic findings. The other two children were found to have follicular carcinomas at surgery; one of these had a hot nodule on 13q scan (Fig. 6, A and B). Preoperative ultrasound examination confirmed the presence of thyroid~t~issueoutside the focal lesion in this patient without a TSH stimulation scan. None of the patients with proved thyroid carcinoma had a history of irradiation to the head or neck or a family history of thyroid malignancy. None had clinical evidence of vocal cord paralysis, tracheal invasion, rapid growth of the thyroid mass, or a fixed thyroid gland. Only one patient with metastatic papillary carcinoma had a suspiciously firm thyroid gland and lymph node enlargement. DISCUSSION Definition of the internal anatomy of the thyroid gland is useful in the management of a variety of thyroid disorders. Although ultrasound is accepted as the most sensitive technique to delineate thyroid gland anatomy in adults, 1'4 its use in children is not well described. The

Volume 103 Number 4

pediatric literature reports the findings from only a few children with focal thyroid lesions? We studied the findings from all children referred for thyroid ultrasound examination to assess the application of this technique in a variety of childhood thyroid disorders. Ultrasound appears to be as readily adaptable in children as in adults, We found ultrasound sufficiently sensitive to detect thyroid tissue in children from newborn infants to adolescents. Because the technique does not reIy on the functional" state of the thyroid gland, ultrasound can detect tissue not seen or only poorly visualized by radionuclide scan. In our study, ultrasound confirmed the presence of normally located thyroid tissue in two infants with equivocal results on technetium scans. The failure of technetium uptake by the thyroid glands of these infants is unexplained. Similar nonvisualization of thyroid tissue with 123I has been reported in a euthyroid newborn infant, l~ Conversely, no thyroid tissue was detected in the neck b y ultrasound in two infants with lingual thyroid glands. Thus ultrasound may prove more sensitive than radionuclide scanning in distinguishing hypothyroid neonates with athyreosis or ectopic glands from those with enzyme defects. Such differentiation may be important for both genetic counseling and establishing a prognosis for mental development.ll. ~ Thyroid ultrasound also detected extralesional thyroid tissue in a boy whose autonomously functioning thyroid nodule was the only tissue seen on I311scan. "Hot" nodules must be differentiated from hemiagenesis or other embryonic defects of thyroid structure preoperatively to avoid removal of normal thyroid tissue. Thyroid ultrasound provides this information without the added radiation associated with radioactive iodine uptake or scan studies after TSH stimulation. Thyroid ultrasound appears to be of limited benefit in the study of diffuse thyroid lesions. In only one of the 25 children studied in our series were multiple cysts detected that were not clinically apparent. In the remainder of the patients, ultrasound demonstrated diffuse enlargement with a homogeneous echo pattern or a nonspecific, patchy pattern that may reflect edema, fibrosis, lymphoid infiltrates, and possibly tiny cysts? Conversely, thyroid ultrasound proved to be a sensitive diagnostic tool in the management of palpable focal lesions of the thyroid gland in children. Thyroid nodules are uncommon in children; however, the reported incidence of malignancy in such nodules in childhood varies from 17% to 57%. 9't3-17 A recent series reports a trend toward a decreasing incidence of carcinoma that may reflect an end to the practice of irradiation to the head and neck in childhood for a variety of benign disorders. ~6 In the management of thyroid nodules in childhood, some authors have advised surgical removal of all focal

Ultrasound in thyroid disorders

55 1

lesions because of the risk of malignancy?. ". ~8Others have recommended a trial Of thyroxine suppression in patients without a history of irradiation or physical findings suggestive of malignancy? 9 As the incidence of malignancy in thyroid nodules declines, there is an increasing need for diagnostic tools to allow selection of the children at highest risk for thyroid carcinoma. The combination of thyroid ultrasound and needle aspiration in adults effectively increases the accuracy of selecting patients with detectable or highly suspect malignancies and reducing surgery for patients with benign lesions.t, 2 As in adults, our experience with children suggests that thyroid nodules that appear completely echofree on ultrasound study are most commonly benign, but that occasionally solitary cystic lesions may be malignant. 3.4,20,2, Therefore, we recommend conservative management initially, including needle aspiration of these cystic lesions for cytology, thyroid hormone suppression, and clinical followu p . 3"19"2~ Because the needle aspiration procedure may evoke more anxiety in children than in adults, confirmation of the presence and precise location of the nodule by ultrasound scan is particularly helpful. In contrast, thyroid carcinomas were found in four of nine children in our series who had echogenic or complex nodules on ultrasound examination; this group included one boy with a functioning solitary nodule. Although uncommon, malignancies have been reported previously in functioning thyroid nodules in children and adults) 2,23 Thyroid nodules that appear solid or mixed on thyroid ultrasound warrant pathologic examination, irrespective of the radionuclide status of the lesions. We conclude that thyroid ultrasound examination offers several advantages in the evaluation of childhood thyroid disorders. The procedure requires no preparation or sedation and does not expose the child to ionizing radiation. Ultrasound is able to define the anatomy of the thyroid gland independently of the functional state of the tissue, and thus may provide information not obtained by radionuclide scan. Finally, the technique distinguishes solitary from multiple thyroid nodules, and echofree from echogenic or complex lesions, to aid in the management of focal thyroid disorders. We thank Dr. R. M. Ehrlich, P. O. Goodwin, and the Department of Medical Publications, The Hospital for Sick Children, for their help in preparation of this manuscript. REFERENCES 1. Van Herie AJ, Rich P, Ljung BME, Ashcraft MW, Solomon DH, Keeler EB: The thyroid nodule. Ann Intern Med 96:221, 1982. 2. Walfish PG, Hazani E, Strawbridge HTG, Miskin M, Rosen IB: Combined ultrasound and needle aspiration cytology in

552

3. 4. 5. 6.

7.

8.

9.

10.

11.

12.

Bachrach et al.

the assessment and management of hypofunctioning thyroid nodule. Ann Intern Med 87:270, i977. Rosen IB, Walfish PG, Miskin M: The ultrasound of thyroid masses. Surg Clin North Am 59:19, 1979. Klonoff DC, Greenspan FS: The thyroid nodule. Adv Intern Med 27:101, 1982. Hamilton CR Jr, Maloof F: Unusual types of hyperthyroidism. Medicine 52:195, 1973. Burger A, Dinichert D, Nicod P, Jenny M, LemarchandB6raud T, Vallotton MB: Effeci~ of atniodarone on serum triiodothyronine, reverse triiodothyronine, thyroxin and thyrotropin: A drug influencing peripheral metabolism of thyroid hormones. J Clin Invest 58:255, 1976. Alderson PO, Gilday DL, Wagner HN, editors: Atlas of pediatric nuclear medicine, St. Louis, 1978, CV Mosby, p 236. Gulenchun KY, Brown RG, Green MD, Ash JM: 1-123 for pediatric thyroid studies: An evaluation of its role. J Radioanalytical Chem 65:251, 1981. Hung W, August GP, Randolph JG, Schisgall RM, Chandra R: Solitary thyroid nodules in children and adolescents. J pediatr Surg 17:225, 1982. Heyman S, Crigler JF Jr, Treves S: Congenital hypothyroidism: t23I thyroidal uptake and scintigraphy. J PEDIATR !01:571, 1982. Smith DW, Blizzard RM, Wilkins L: The menta ! prognosis in hypothyroidism in infancy and childhood: A review of 128 cases. Pediatrics 19:1011, 1957. Rimoin DL, Schimke RN, editors: Genetic disorders of the endocrine glands, St. Louis, 197I, CV Mosby, p 113.

The Journal o f Pediatrics October 1983

13. Hayles AB, Johnson LM, Beahrs OH, Woolner LB: Carcinoma of the thyroid in children. Am J Surg 106:735, 1963. 14. Kirkland RT, Kirkland JL, Rosenberg HS, Harberg FJ, Librik L, Clayton GW: Solitary thyroid nodules in 30 children and report of a Child with a thyroid abscess. Pediatrics 51:85, i973. 15. Rallison ML, Dobyns BM, Keating FR Jr, Rail JE, Tyler FH: Thyroid nodularity in children. JAMA 233:1069, 1975. 16. Scott MD, Crawford JD: Solitary thyroid nodules in childhood: Is the incidence of thyroid carcinoma declining? Pediatrics 58:521, 1976. 17. Silverman SH, Nussbaum M, Rausen AR: Thyroid nodules in children: A ten year experience at one institution. Mt Sinai J Med 46:460, 1979. 18. Bin:row GN: Truth and fancy in the management of the solitary thyroid nodule. Yale J Biol Med 53:325, 1980. 19. Fisher DA: Thyroid nodules in childhood and their management. J PEDIATR89:866, 1976. 20. Crile G Jr: Treatment of thyroid cysts by aspiration. Surgery 59:210, 1966. 21. Hammer M, Wortsman J, Folse R: Cancer in cystic lesions of the thyroid. Arch Surg 117:1020, 1982. 22. Hamburger JI: Solitary autonomously functioning thyroid lesions: Diagnosis, clinical features and pathogenetic considerations. Am J Med 58:740, 1975. 23. Hopwood N J, Carroll RG, Kenny FM, Foley TP Jr: Functioning thyroid masses in childhood and adolescence: Clinical, surgical, and pathologic correlations. J PEDIATR 89:710, 1976.