Thyroid Malignancy Markers on Sonography Are Common in Patients With Benign Thyroid Disease and Previous Iodine Deficiency

Transcription

1 ORIGINAL RESEARCH Thyroid Malignancy Markers on Sonography Are Common in Patients With Benign Thyroid Disease and Previous Iodine Deficiency Anne Krejbjerg, MD, Lucia Brilli, MD, Arunas Pikelis, MD, Henrik Baymler Pedersen, MD, Peter Laurberg, MD, DMSc Received February 20, 2014, from the Departments of Endocrinology (A.K., P.L.) and Ear, Nose, and Throat Head and Neck Surgery (A.P., H.B.P.), Aalborg University Hospital, Aalborg, Denmark; and Department of Endocrinology, University of Siena, Siena, Italy (L.B.). Revision requested March 24, Revised manuscript accepted for publication May 14, We thank Ingelise Leegaard for carefully performing sonography. This work was supported by the Foundation of Karen Elise Jensen, the Danish Council for Independent Research, the Foundation of Music Publishers Agnes and Knut Mørk, and the Scientific Exchange Grant from the European Endocrine Society. Address correspondence to Anne Krejbjerg, MD, Department of Endocrinology, Aalborg University Hospital, Søndre Skovvej 15, 9000 Aalborg, Denmark. doi: /ultra Objectives The purpose of this study was to evaluate the characteristics of benign thyroid nodules on sonography and ultrasound elastography in a population exposed to iodine deficiency. Methods We conducted a prospective systematic evaluation of preoperative thyroid sonography and elastography in patients assigned for surgical excision of benign thyroid nodules. Two experienced sonographers performed all sonographic and elastographic examinations. Thyroid nodules were evaluated by 7 generally accepted sonographic malignancy risk markers and assigned an elasticity score on elastography. The final diagnosis of a benign thyroid nodule was based on histopathologic analysis of resected thyroid gland tissue. Results We evaluated 232 thyroid nodules in 105 patients (86 women and 19 men). In total, 57% of the examined nodules had 1 or 2 malignancy risk markers present, and 24% did not have any markers present. A solid nodule larger than 15 mm was the most common malignancy risk marker observed (63%), followed by low elasticity (33%), microcalcifications (26%), and hypoechogenicity (15%). In an analysis stratified according to the number of nodules (solitary versus multiple), low elasticity was described more frequently in solitary nodules (61.9% versus 30.4%; P =.004). A large nodular volume was a predictor (P <.05) of microcalcifications and intranodular vascularization, whereas an absent halo sign and a solid nodule were found less frequently in nodules with larger volumes. Conclusions Our results show that routine preoperative malignancy risk evaluation of presumably benign thyroid nodules is of little value when performed on patients exposed to iodine deficiency. Key Words benign nodule; elastography; head and neck ultrasound; prospective; sonography; thyroid Thyroid nodules are common, especially in populations that have been exposed to iodine deficiency. 1,2 Although most thyroid nodules are benign, an increase in the incidence of thyroid cancer has been observed over the last decades in many countries. 3 5 In Denmark, the increase in the thyroid cancer incidence was mainly caused by incidental papillary microcarcinomas. 6 Thus, there is a need for effective preoperative malignancy risk evaluation to identify relevant nodules for fine-needle aspiration and to help the surgeon plan the optimal type of surgery for the individual patient by the American Institute of Ultrasound in Medicine J Ultrasound Med 2015; 34:

2 International guidelines recommend the use of sonography as the primary method for evaluating the risk of malignancy before referring patients for fine-needle aspiration or surgery. 7 Ultrasound elastography is a diagnostic technique that evaluates tissue stiffness by measuring the amount of deformation that occurs when a nodule is exposed to external pressure. 8 A stiff nodule is classified as suspicious for malignancy, and several studies have published encouraging elastographic results in prostate lesions, mammary lesions, and, most recently, thyroid nodules Previous studies have focused on the diagnostic accuracy and validity of sonography and elastography in differentiating between benign and malignant thyroid nodules and have included selected groups of patients assigned for thyroid cancer surgery Consequently, the results may be at variance in a broader group of patients investigated in a real-life clinical setting; particularly in a region where the population has been exposed to iodine deficiency, such as the Danish, 18 and therefore has a high frequency of thyroid nodules. 19 Thus, there is a growing demand for preoperative malignancy risk evaluation by sonography in a population in whom benign thyroid nodules are common due to low iodine intake. However, little is known about the sonographic and elastographic features of histologically verified benign thyroid nodules in a population exposed to iodine deficiency. Therefore, we conducted a blinded prospective preoperative evaluation of presumably benign thyroid nodules on sonography and elastography and included nodules that subsequently were proven benign by histopathologic analysis. The thyroid nodules detected on sonography were evaluated by generally accepted sonographic malignancy risk markers 7 and assigned an elasticity score on elastography. 12 The Department of Ear, Nose, and Throat Head and Neck Surgery is a primary referral center for patients undergoing thyroid surgery in the hospital uptake population of 580,000 residents. Information on smoking habits and prior neck radiotherapy or surgery was obtained by a questionnaire, and patients height and weight were measured. Two experienced sonographers performed all sonographic and elastographic examinations. Before the study was initiated, professional instructors from the manufacturer of the new sonographic and elastographic apparatus visited and gave instructions regarding the recommended use of the equipment. The sonographers had a 1-month training period in which sonographic and elastographic examinations were performed by both sonographers. During the examinations, the sonographers discussed the sonographic and elastographic features of the thyroid nodules to establish a baseline consensus. After the training period, 1 of the 2 sonographers performed both sonography and elastography on a given patient depending on which of the sonographers was available. The sonographers had no knowledge of the indication for surgery. The final diagnosis of a benign thyroid nodule was based on histopathologic analysis performed by the authorized personnel responsible for all histopathologic examinations at Aalborg University Hospital. General patient data, including fine-needle aspiration and histologic reports, in conjunction with sonographic and elastographic features, were recorded as part of Figure 1. Patients and thyroid nodules included in the final study population. Materials and Methods Study Population We conducted a prospective evaluation between November 2009 and May Patients assigned for thyroid surgery by 2 surgeons at the Department of Ear, Nose, and Throat Head and Neck Surgery of Aalborg University Hospital were consecutively referred for preoperative sonography and elastography during this period and were included in our study. The primary exclusion criterion was planned surgery for presumably malignant thyroid disease based on fine-needle aspiration and clinical appearance. Patients with no thyroid nodules and patients in whom histopathologic analysis of surgically removed tissue showed malignant thyroid disease were also excluded (Figure 1). 310 J Ultrasound Med 2015; 34:

3 routine clinical care. Data were recorded and stored as approved by the Danish Data Protection Agency ( ). Conventional Sonography Conventional sonography was performed with a Hi Vision 900 ultrasound system (Hitachi Medical Corporation, Tokyo, Japan) equipped with a linear 10 5-MHz probe. Patients were positioned lying on their back with the neck slightly extended over a pillow. The thyroid gland volume was calculated as maximal length width depth π/6 for each lobe 20 and was evaluated for echogenicity (compared to the strap muscles of the neck), structure, and tissue calcifications. Clearly defined thyroid nodules larger than 5 mm were registered and systematically evaluated for malignancy risk markers. In the event of more than 4 nodules in a single lobe, only the 4 largest nodules were registered and evaluated. The thyroid nodule size was assessed by the maximal depth, width, and length, and the nodule echo texture was classified as either homogeneous or heterogeneous. The malignancy risk markers evaluated were solid nodule, hypoechogenicity (defined according to individual surrounding thyroid parenchymal echogenicity), microcalcifications, an absent halo sign, irregular margins, taller-than-wide shape, and intranodular vascularization. 7 Solid nodules were defined as nodules containing greater than 75% solid tissue and were classified as a malignancy risk marker when the largest nodule diameter was greater than 15 mm. 21 Microcalcifications were defined as highly echogenic foci inside the thyroid nodule that resembled microcalcifications. A tallerthan-wide shape was measured on the transverse view. Intranodular vascularization was defined as marked intranodular blood flow and absent or slight perinodular blood flow evaluated by the color flow Doppler pattern. 22 Elastography Elastography was performed by a real-time freehand technique using the same Hitachi Hi Vision 900 ultrasound system and probe. The elastogram was displayed over the B-mode image in a color scale depending on the tissue elasticity: red and green were soft/elastic tissue, and blue was hard/stiff tissue. The color coding of the elastographic image was classified into 1 of 4 elasticity score 12 groups, as shown in Figure 2. The elasticity score was assessed according to the overall pattern on more than 10 consecutive frames. We used the lowest pressure necessary to obtain an elastographic record. The pressure level was indicated on the screen by a numeric scale (1 5), and during the examinations, the pressure was constantly maintained between 3 and 4. The sonographer performed real-time elastography. Thus, elastograms were made technically acceptable according to the manufacturer s instructions and subsequently scored by the sonographer as part of the same procedure. In case of a partially cystic lesion, the solid component of the nodule was examined, reducing artifacts known to be caused by the fluid. Definition of Variables The nodule volume was calculated as maximal length width depth π/6 and analyzed in 3 categories: (1) less than 1 ml, (2) 1 to 6 ml, and (3) greater than 6 ml. The total number of nodules examined in a single thyroid gland was categorized in 3 groups: (1) 1 to 3 nodules, (2) 4 nodules, and (3) 5 to 7 nodules. Thyroid enlargement was defined as a thyroid volume exceeding 18 ml for women and 25 ml for men, which corresponds to the mean + 3 SD in iodine-sufficient populations, 23 and analyzed in 4 groups: (1) normal thyroid volume, (2) 1.1 to 2.0 upper reference limit, (3) 2.1 to 4.0 upper reference limit, and (4) greater than 4.0 upper reference limit. Multinodularity was defined as 2 or more nodules in the thyroid gland. Statistical Analysis All data processing was done with Stata version 11.0 software (StataCorp, College Station, TX). Two-sided P <.05 was considered significant. Patients with malignant thyroid disease and patients without nodules detected by sonography were excluded (Figure 1). Some data were Figure 2. Elasticity scores (ES) in thyroid nodules (marked by stippled outlines). For each investigation, the conventional sonogram is shown on the right, and the corresponding elastogram is shown on the left. An elasticity score of A indicates elasticity in the entire nodule; B, elasticity in most of the nodule ( 50%); C, stiffness in most of the nodule (>50%); and D, stiffness and no elasticity in the entire nodule. Scores C and D were considered malignancy risk markers. J Ultrasound Med 2015; 34:

4 missing for the following malignancy risk markers: halo (n = 6), solid nodule (n = 1), and elasticity score (n = 7). In a stratified analysis, we used a χ 2 test to compare frequencies, and φ 4-fold correlation was used to describe the relationship between the 8 risk markers evaluated. To determine predictors of malignancy risk markers, a multivariate logistic regression model was developed for each individual risk marker, and robust standard error was used to account for dependency between multiple nodules. Included in the models were nodule volume, number of nodules, and thyroid volume. Comparisons between continuous variables were made with the Mann-Whitney test, and a χ 2 test was used to compare categorical variables. among the other 7 sonographic malignancy risk markers in the stratified analysis (data not shown). When investigating correlations between the presence of malignancy risk markers, we found 6 positive correlations out of 21 possible. As depicted in Figure 3, 6 conventional sonographic risk markers correlated with some degree, but with low correlation coefficients. Contrary to this, elasticity scores C and D and a taller-than-wide nodule did not correlate with any of these conventional sonographic risk markers. Results Study Population We prospectively examined 338 thyroid nodules in 105 patients (86 women and 19 men) assigned for thyroid surgery at the Department of Ear, Nose, and Throat Head and Neck Surgery of Aalborg University Hospital. Hemithyroidectomy was performed on 85 of the examined patients; thus, histopathologic results were not available for 104 nonresected nodules, and these were excluded from further analyses. Furthermore, we excluded 2 nodules that were not clearly delineated, leaving 232 benign thyroid nodules for analyses (Figure 1). Characteristics of the study population, including both patients and nodules, are presented in Table 1. Patients included in our study had either hemithyroidectomy (n = 85) or total thyroidectomy (n = 20) performed for presumably benign thyroid disease due to pressure symptoms or minor conditions such as cosmetic indications. Indications for surgery are described in Table 1. Malignancy Risk Markers We used 8 different sonographic malignancy risk markers, including elasticity scores C and D on elastography, to evaluate all examined nodules and observed that 57% of the nodules had 1 or 2 malignancy risk markers present, and 24% did not have any markers present (Table 2). A solid nodule larger than 15 mm was the most common malignancy risk marker observed, followed by elasticity scores C and D, microcalcifications, and hypoechogenicity. Of the 232 nodules, 23 (9.9%) were solitary, and 209 (90.1%) were part of a multinodular thyroid gland. In an analysis stratified according to the number of nodules (solitary versus multiple), elasticity scores C and D were described more frequently in solitary nodules (61.9% versus 30.4%; P =.004). There was no significant difference in frequency Table 1. Characteristics of the Study Population Characteristic Value Patients (n = 105) Women/men 86/19 Age, y 52 ± 12 Body mass index, kg/m ( ) Smoking status Current smoker 27 (25.7) Previous smoker 33 (31.4) Never smoker 45 (42.9) Thyrotropin, mu/l 0.97 ( ) Thyroid function a Hypothyroid 1 (1.0) Euthyroid 91 (86.7) Hyperthyroid 13 (12.4) Nodules per patient, n 3 ± 1.7 Indication for surgery Solitary euthyroid nodule 26 (24.8) Multinodular euthyroid goiter 53 (50.5) Multinodular toxic goiter 11 (10.5) Graves hyperthyroidism 1 (1.0) Thyroid cyst 14 (13.3) Nodules (n = 232) Maximal diameter, mm 22.4 ( ) Volume, ml 3.0 ( ) Echogenicity Hypoechoic 36 (15.5) Isoechoic 186 (80.2) Hyperechoic 10 (4.3) Cysticity of nodule 0% 32 (13.8) 1% 25% 114 (49.1) 26% 50% 40 (17.2) 51% 75% 12 (5.2) 76% 100% 34 (14.7) Data are presented as number (percent), mean ± SD, and median (25th 75th percentiles) where applicable. a All patients with hypothyroidism/hyperthyroidism received medical treatment and were biochemically euthyroid at the time of the study. 312 J Ultrasound Med 2015; 34:

5 Nodular Size and Malignancy Risk Markers A large nodule volume was a predictor of microcalcifications and intranodular vascularization, whereas an absent halo sign and a solid nodule were less frequent in nodules with large volumes (Table 3). The number of nodules in a gland and the thyroid gland volume were not consistent predictors of malignancy risk markers. In supplementary analyses, we considered all solid nodules (>5 mm) as suspicious (n = 146 [63%]). In such analyses, only 9% of the examined thyroid nodules did not have any risk markers present. The overall results of the correlation and logistic regression analyses were unaltered (data not shown). Discussion Principal Findings We conducted a prospective systematic evaluation of benign thyroid nodules on sonography and elastography in Danish patients assigned for thyroid surgery. In Denmark, where the population has been exposed to iodine deficiency and where benign thyroid nodules are common, the finding of thyroid malignancy in patients undergoing surgery for presumably benign thyroid disease has been increasing in recent years. 6 Thus, preoperative thyroid nodule evaluation has become increasingly important. Previous studies have concentrated on the diagnostic accuracy and validity of sonography and elastography in differentiating between benign and malignant thyroid nodules in patients assigned for thyroid cancer surgery. Our intention was to examine the appearance of benign thyroid nodules on sonography and elastography in a population exposed to iodine deficiency. In this group of Danish patients assigned for thyroid surgery, benign thyroid nodules with malignancy risk markers were very common. Many of the risk markers correlated, but notably, none of the conventional sonographic risk markers correlated with a taller-than-wide thyroid nodule or the elasticity score. On the other hand, the nodular volume predicted the presence of several of the malignancy risk markers. have investigated a patient cohort representative of the diverse group of patients with thyroid nodular disease seen in a clinic where the population has been exposed to iodine deficiency. Wienke et al 24 investigated sonographic features of primarily solitary benign thyroid nodules and found, in accordance with our results, that more than two-thirds of the examined nodules had at least 1 sonographic malignancy risk marker present. The study was limited in sample size (68 benign nodules), and the malignancy risk markers did not include the elasticity score. In a recent study by Mehta Table 2. Prevalence of Malignancy Risk Markers on Sonography in the Examined Thyroid Nodules (n = 232) Parameter n (%) No. of malignancy risk markers 0 55 (23.7) 1 73 (31.5) 2 59 (25.4) 3 33 (14.2) 4 7 (3.0) 5 5 (2.2) Malignancy risk markers Absent halo 20 (8.6) Microcalcifications 61 (26.3) Irregular margins 22 (9.5) Solid nodule >15 mm 95 (41.0) Hypoechogenecity 35 (15.1) Intranodular vascularization 15 (6.5) Taller than wide 19 (8.2) Elasticity scores C + D 76 (32.8) Data were missing for 14 nodules. Figure 3. Relationship between the 8 malignancy risk markers evaluated. Lines between risk markers indicate a statistically significant positive φ 4- fold correlation coefficient (r) and statistical significance (P). If no line is shown, the correlation was not statistically significant. Previous Studies Numerous studies have investigated malignancy prediction of thyroid nodules by sonography and elastography Most of these studies examined a selected group of patients assigned for thyroid cancer surgery and reported encouraging results. Thus, sonography and elastography may well give valuable information on the risk of thyroid nodule malignancy in such selected patients. However, fewer studies J Ultrasound Med 2015; 34:

6 et al, 25 mutation-negative nodules with follicular lesions that had cytologic findings of undertermined significance were examined with preoperative sonography, not including elastography. In this group of patients, at least 1 suspicious sonographic feature was identified in 32% of nodules, regardless of histologic findings. In addition, they found a solid nodule composition, microcalcifications, and hypo - echogenicity to be the most common sonographic risk markers, also regardless of histologic findings. In our preoperative examinations, we found the same pattern of risk marker frequency, despite the more heterogeneous group of patients. In a recent large study, Smith-Bindman et al 26 performed a retrospective case-control evaluation to quantify the risk of thyroid cancer associated with thyroid nodules based on conventional sonographic characteristics. That study included a large unsorted group of patients and found only a solid nodule composition, microcalcifications, and a nodule size of greater than 2 cm to be associated with malignancy in multiple-predictor models. Lippolis et al 27 investigated the potential role of elastography in the presurgical diagnosis of thyroid cancer. They examined a cohort of consecutive patients before surgery for a thyroid nodule with indeterminate cytologic findings and found that two-thirds of histologically verified benign nodules were classified as suspicious by elastography. Rivo-Vázquez et al 28 recently published a study examining patients who had sonography and elastography before undergoing thyroid surgery for nodular disease. Their study population included primarily multinodular thyroid glands (80% of the patient cohort), and in contrast to Lippolis et al, 27 they reported that only 13% of benign nodules were suspicious for malignancy on elastography. However, 60% of the verified malignant nodules were classified as benign or probably benign by elastography. Comparisons with and between these various studies are complicated by the different sonographic and elastographic apparatuses and software used, the interobserver variability, and the different elasticity scores and cutoff values Table 3. Predictors of Malignancy Risk Markers on Sonography in the Examined Thyroid Nodules (n = 232) Absent Halo Sign Microcalcifications Irregular Margins Solid Nodules >15 mm Parameter OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P Nodule volume <1 ml 1.00 (ref) 1.00 (ref) 1.00 (ref) NA 1 6 ml 0.52 ( ) ( ) ( ) (ref) >6 ml 0.23 ( ) ( ) ( ) ( ).04 No. of nodules (ref) 1.00 (ref) 1.00 (ref) 1.00 (ref) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ).22 Thyroid volume Normal size 1.00 (ref) 1.00 (ref) 1.00 (ref) 1.00 (ref) upper ref 1.63 ( ) ( ) ( ) ( ) upper ref 1.83 ( ) ( ) ( ) ( ).43 >4.0 upper ref 0.37 ( ) ( ) ( ) ( ).001 Hypoechogenicity Intranodular Vascularization Taller Than Wide Elasticity Scores C + D Parameter OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P Nodule volume <1 ml 1.00 (ref) 1.00 (ref) 1.00 (ref) 1.00 (ref) 1 6 ml 0.69 ( ) ( ) ( ) ( ).55 >6 ml 0.66 ( ) ( ) ( ) ( ).52 No. of nodules (ref) 1.00 (ref) 1.00 (ref) 1.00 (ref) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ).24 Thyroid volume Normal size 1.00 (ref) 1.00 (ref) 1.00 (ref) 1.00 (ref) upper ref 0.55 ( ) ( ) ( ) ( ) upper ref 0.65 ( ) ( ) ( ) ( ).50 >4.0 upper ref 0.72 ( ) ( ) ( ) ( ).68 CI indicates confidence interval; NA, not applicable; OR, odds ratio; and ref, reference limit. 314 J Ultrasound Med 2015; 34:

7 chosen for the individual studies. Our evaluation showed that elasticity scores C and D were significantly more frequent in solitary nodules than in the nodules of a multinodular gland. The reason for this difference may be that artifacts occur when using elastography on nodules with little surrounding normal thyroid tissue, as is often the case in multinodular thyroid glands, leading to stiffer color coding. Thus, the difference in the study populations investigated by Lippolis et al 27 and Rivo-Vázquez et al 28 may explain some of the discrepancy between frequencies of suspicious elasticity scores in benign nodules. Similar to the study of Rivo-Vázquez et al, 28 our evaluation included primarily patients with multinodular glands (76%), and we found elasticity scores C and D in 33% of all examined nodules. Strengths and Limitations In the stratified analysis, our evaluation was limited by the sample size, which resulted in a limited number of nodules in some groups. Furthermore, the sample size restricted the number of possible predictors in our multivariate regression models. We had 2 sonographers performing the examinations; thus, the sonographic and elastographic results were subject to interobserver variability, although both sonographers had years of experience and had a 1-month training period in which sonographic and elastographic examinations were performed by both sonographers together. Additionally, we did not exclude nodules with greater than 20% cystic content or nodules with eggshell or coarse calcifications, even though these may produce artifacts on elastography. However, we aimed at examining an unselected group of patients with benign thyroid nodules as they would be examined in a real-life preoperative clinical setting. The strengths of the evaluation were the prospective design and the unselected group of consecutive patients examined. Furthermore, the likelihood of a context bias was low because both sonographers were blinded to the type of disease. Conclusions We evaluated benign thyroid nodules by sonography and elastography in Danish patients assigned for thyroid surgery and found a high frequency of malignancy risk markers among these nodules. Our results show that routine preoperative malignancy risk evaluation of presumably benign thyroid nodules is of little value when performed on patients exposed to iodine deficiency. The study does not allow conclusions on the usefulness of routine preoperative sonographic and elastographic malignancy evaluation in populations with long-standing sufficient iodine intake. References 1. Delange F. The disorders induced by iodine deficiency. Thyroid 1994; 4: Laurberg P, Cerqueira C, Ovesen L, et al. Iodine intake as a determinant of thyroid disorders in populations. Best Pract Res Clin Endocrinol Metab 2010; 24: Londero SC, Krogdahl A, Bastholt L, et al. Papillary thyroid carcinoma in Denmark : an investigation of changes in incidence. Cancer Epidemiol 2013; 37:e1 e6. 4. Smailyte G, Miseikyte-Kaubriene E, Kurtinaitis J. Increasing thyroid cancer incidence in Lithuania in BMC Cancer 2006; 6: Reynolds RM, Weir J, Stockton DL, Brewster DH, Sandeep TC, Strachan MW. Changing trends in incidence and mortality of thyroid cancer in Scotland. Clin Endocrinol (Oxf) 2005; 62: Londero SC, Krogdahl A, Bastholt L, et al. Papillary thyroid microcarcinoma in Denmark : a national study of epidemiology and clinical significance. Thyroid 2013; 23: Gharib H, Papini E, Paschke R, et al. American Association of Clinical Endocrinologists, Associazione Medici Endocrinologi, and European Thyroid Association medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. J Endocrinol Invest 2010; 33(suppl): Lyshchik A, Higashi T, Asato R, et al. Elastic moduli of thyroid tissues under compression. Ultrason Imaging 2005; 27: Konig K, Scheipers U, Pesavento A, Lorenz A, Ermert H, Senge T. Initial experiences with real-time elastography guided biopsies of the prostate. J Urol 2005; 174: Itoh A, Ueno E, Tohno E, et al. Breast disease: clinical application of US elastography for diagnosis. Radiology 2006; 239: Rago T, Santini F, Scutari M, Pinchera A, Vitti P. Elastography: new developments in ultrasound for predicting malignancy in thyroid nodules. J Clin Endocrinol Metab 2007; 92: Asteria C, Giovanardi A, Pizzocaro A, et al. US-elastography in the differential diagnosis of benign and malignant thyroid nodules. Thyroid 2008; 18: Cappelli C, Castellano M, Pirola I, et al. Thyroid nodule shape suggests malignancy. Eur J Endocrinol 2006; 155: Papini E, Guglielmi R, Bianchini A, et al. Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-doppler features. J Clin Endocrinol Metab 2002; 87: Hong Y, Liu X, Li Z, Zhang X, Chen M, Luo Z. Real-time ultrasound elastography in the differential diagnosis of benign and malignant thyroid nodules. J Ultrasound Med 2009; 28: Vorlander C, Wolff J, Saalabian S, Lienenluke RH, Wahl RA. Real-time ultrasound elastography: a noninvasive diagnostic procedure for evaluating dominant thyroid nodules. Langenbecks Arch Surg2010; 395: Ding J, Cheng H, Ning C, Huang J, Zhang Y. Quantitative measurement for thyroid cancer characterization based on elastography. J Ultrasound Med 2011; 30: J Ultrasound Med 2015; 34:

8 18. Krejbjerg A, Bjergved L, Pedersen IB, et al. Iodine fortification may influence the age-related change in thyroid volume: a longitudinal populationbased study (DanThyr). Eur J Endocrinol 2014; 170: Knudsen N, Bülow I, Jorgensen T, Laurberg P, Ovesen L, Perrild H. Goitre prevalence and thyroid abnormalities at ultrasonography: a comparative epidemiological study in two regions with slightly different iodine status. Clin Endocrinol (Oxf) 2000; 53: Knudsen N, Bols B, Bülow I, et al. Validation of ultrasonography of the thyroid gland for epidemiological purposes. Thyroid1999; 9: Frates MC, Benson CB, Charboneau JW, et al. Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound consensus conference statement. Radiology 2005; 237: Rago T, Vitti P, Chiovato L, et al. Role of conventional ultrasonography and color flow-doppler sonography in predicting malignancy in cold thyroid nodules. Eur J Endocrinol 1998; 138: Gutekunst R, Becker W, Hehrmann R, Olbricht T, Pfannenstiel P. Ultrasonic diagnosis of the thyroid gland [in German]. Dtsch Med Wochenschr 1988; 113: Wienke JR, Chong WK, Fielding JR, Zou KH, Mittelstaedt CA. Sonographic features of benign thyroid nodules: interobserver reliability and overlap with malignancy. J Ultrasound Med 2003; 22: Mehta RS, Carty SE, Ohori NP, et al. Nodule size is an independent predictor of malignancy in mutation-negative nodules with follicular lesion of undetermined significance cytology. Surgery 2013; 154: Smith-Bindman R, Lebda P, Feldstein VA, et al. Risk of thyroid cancer based on thyroid ultrasound imaging characteristics: results of a population-based study. JAMA Intern Med 2013; 173: Lippolis PV, Tognini S, Materazzi G, et al. Is elastography actually useful in the presurgical selection of thyroid nodules with indeterminate cytology? J Clin Endocrinol Metab 2011; 96:E1826 E Rivo-Vázquez A, Rodríguez-Lorenzo Á, Rivo-Vázquez JE, et al. The use of ultrasound elastography in the assessment of malignancy risk in thyroid nodules and multinodular goitres. Clin Endocrinol (Oxf) 2013; 79: J Ultrasound Med 2015; 34: