3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 689
ORIGINAL RESEARCH
Sonographic Features of Thyroid Nodules That May Help Distinguish Clinically Atypical Subacute Thyroiditis From Thyroid Malignancy Fu-shun Pan, MD, Wei Wang, MD, Yan Wang, MD, Ming Xu, MD, Jin-yu Liang, MD, Yan-ling Zheng, MD, Xiao-yan Xie, MD, Xiao-xi Li, MD
Objectives—The purpose of this study was to evaluate sonographic features for distinguishing clinically atypical subacute thyroiditis from malignant thyroid nodules. Methods—A total of 165 hypoechoic thyroid nodules without calcification in 135 patients with histologic diagnosis were included in this study. These nodules were classified into 2 groups: a thyroiditis group (55 nodules in 36 patients) and a malignancy group (110 nodules in 99 patients). The sonographic features of the groups were retrospectively reviewed. Results—No significant differences were detected for the variables of marked echogenicity, a taller-than-wide shape, and mixed vascularity. However, a poorly defined margin was detected more frequently in the thyroiditis group than the malignancy group (P < .05); it yielded a high capability for differential diagnosis of atypical subacute thyroiditis, with sensitivity and specificity of 87.3% and 80.9%, respectively. Centripetal reduction echogenicity was observed exclusively in the thyroiditis group, with high specificity (100%) but low sensitivity (21.8%) for atypical subacute thyroiditis diagnosis. All of the thyroiditis nodules with a positive color signal showed noninternal vascularity (negative predictive value, 100%).
Received May 5, 2014, from the Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound (F.P., W.W., Y.W., M.X., J.L., Y.Z., X.X.), and Department of Vascular and Thyroid Surgery (X.L.), First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. Revision requested June 3, 2014. Revised manuscript accepted for publication July 8, 2014. This work was supported by the National Natural Science Foundation of China (grant 81301238). Address correspondence to Xiao-yan Xie, MD, Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, First Affiliated Hospital of Sun Yat-sen University, 510080 Guangzhou, Guangdong, China. E-mail: [email protected] doi:10.7863/ultra.34.4.689
Conclusions—There is a considerable overlap between the sonographic features of atypical subacute thyroiditis and thyroid malignancy. However, the margin, echogenicity, and vascularity type are helpful indicators for differential diagnosis of atypical subacute thyroiditis. Key Words—carcinoma; head and neck ultrasound; sonography; subacute thyroiditis
S
ubacute thyroiditis, also known as de Quervain thyroiditis, is the most common cause of thyroid pain. It is assumed to be postviral in origin, frequently preceded by an upper respiratory tract infection.1–2 The diagnosis of subacute thyroiditis is based on clinical and laboratory features, including cervical pain, swelling of the thyroid region, fever, occasional dysphasia, suppressed thyrotropin levels, and an elevated erythrocyte sedimentation rate, and pathologic diagnosis is rarely needed.2–4 Sonography is the first-line imaging modality for thyroid abnormalities, and several previous studies have revealed that sonography is a useful adjunctive tool for diagnosing and monitoring subacute thyroiditis. The characteristic sonographic features of sub-
©2015 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2015; 34:689–696 | 0278-4297 | www.aium.org
3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 690
Pan et al—Sonographic Features of Subacute Thyroiditis and Thyroid Malignancy
acute thyroiditis are thyroid enlargement and poorly defined and irregularly shaped hypoechoic thyroid lesions with an absence of color flow in the hypoechoic area.5–8 Thus, subacute thyroiditis cases with typical clinical and sonographic features may pose minimal diagnostic difficulties. However, some subacute thyroiditis cases are clinically atypical and do not feature cervical pain, fever, or serum thyroid dysfunction. In most of these atypical cases, which have been termed atypical subacute thyroiditis, the primary presentation consists of painless and hard thyroid nodules.2,9–13 An overlap in the clinical and sonographic features of atypical subacute thyroiditis and other types of thyroid nodules has been reported in several studies. Most frequently, atypical subacute thyroiditis nodules are mistaken for malignant lesions, which often results in unnecessary or improper therapy, including thyroid lobectomy.3,11,13,14 To date, the precise incidence of atypical subacute thyroiditis is not clear due to wide disparities in the various published studies.9,12,13 For example, Stein et al12 reported that 13 of 36 patients with subacute thyroiditis had no appreciable pain. However, with the broad application of sonographically guided fine-needle aspiration thyroid biopsy, we believe that the frequency of cytopathologic diagnosis of atypical subacute thyroiditis may be increasing. Thus, it will be an important concern for the differential diagnosis of this condition from other thyroid nodules, especially malignancy. To our knowledge, relatively few studies have described the sonographic features of atypical subacute thyroiditis.11,14 The purpose of our study was to evaluate sonographic features that distinguish atypical subacute thyroiditis from malignant thyroid nodules.
Materials and Methods This retrospective study was approved by the Institutional Review Board, with a waiver of informed consent. From January 2005 to June 2013, a total of 36 patients with 55 thyroid nodules were included in an atypical subacute thyroiditis group, which consisted of 8 male and 28 female patients ranging in age from 25 to 59 years (mean age, 46.5 years). All patients in the thyroiditis group denied having any cervical pain or tenderness during the last 6 months and had received histologic confirmation of subacute thyroiditis. Meanwhile, because all of the nodules in the thyroiditis group were solid and hypoechoic without calcification, another 110 thyroid nodules in 99 patients were randomly selected during the same period to comprise a malignancy group. The following inclusion criteria were used for the malignancy group: the presence of solid, noncalcified hypoechoic echogenicity on sonography and his-
690
tologic confirmation of the diagnosis. The malignancy group consisted of 19 male and 80 female patients ranging in age from 22 to 72 years (mean age, 45.0 years). The criteria for surgery were based on sonographic features that raised a suspicion of malignancy (taller-thanwide shape and marked hypoechogenicity), patients presenting with pressure symptoms, patient preference, and abnormal results of gun biopsies (either malignancy or other suspicious results, including atypical, inadequate, and nondiagnostic findings). Pathologic diagnoses for the thyroiditis group were obtained by specimens from gun biopsy (n = 5) and surgery (n = 50). All patients in the malignancy group underwent surgery with or without preceding gun biopsies. The pathologic diagnoses of 110 malignant nodules included papillary carcinoma (106 [96.4%]), squamous cell carcinoma (1 [0.9%]), thyroid lymphoma (2 [1.8%]), and anaplastic carcinoma (1 [0.9%]). Laboratory examination results including serum thyroid function, anti-thyroglobulin and antithyroid peroxidase antibodies, were available for all of the patients with atypical subacute thyroiditis within 2 weeks of surgery or biopsy. During the sonographic examination, each patient was in the supine position, and the anterior area of the neck was fully exposed. Thyroid sonography was performed with several pieces of equipment, including MyLab 90 (Esaote SpA, Genoa, Italy), Acuson Sequoia 512 (Siemens Medical Solutions, Mountain View, CA), and ProSound SSD-5000 (Aloka Co, Ltd, Tokyo, Japan) with linear array small-parts probes. Sonographic examinations were performed by 3 radiologists (Y.W., W.W., and M.X.) with more than 5 years of experience in thyroid sonography. Standard sonographic settings dedicated to the thyroid were used. The scanning protocol in all cases included both transverse and longitudinal real-time imaging of the thyroid nodules. Color Doppler sonography of all thyroid nodules included in this study was also performed before surgery or biopsy. The Doppler gain was set at an appropriate level to avoid any color noise. The interval from thyroid sonography to surgery was less than 2 weeks. The sonograms were independently analyzed by 2 other staff radiologists (J.L. and Y.Z.) in consensus, both of whom had more than 8 years of experience in thyroid sonography. Both specialists were blinded to the clinical history, previous imaging, and pathologic results corresponding to each sonogram. Grayscale sonographic features that were evaluated for this study included thyroid volume, lesion size, margin, shape, and echogenicity. The lesion size was assessed by the maximum diameter, and the lesions were classified into 3 subgroups: 1.0 cm or less, 1.0 cm to
J Ultrasound Med 2015; 34:689–696
3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 691
Pan et al—Sonographic Features of Subacute Thyroiditis and Thyroid Malignancy
2.5 cm, and greater than 2.5 cm. The thyroid gland was considered to be enlarged when the anteroposterior diameter was greater than 2.0 cm. The margin was classified as either poorly defined (defined as a thyroid nodule that could not be discriminated from the peripheral thyroid parenchyma) or well defined. The shape of the nodule was classified as ovoid to round (anteroposterior dimension equal to or less than transverse dimension), taller than wide (anteroposterior dimension greater than transverse dimension), or irregular (neither ovoid to round nor taller than wide). Echogenicity was classified as markedly hypoechoic and hypoechoic with respect to the adjacent strap musculature and thyroid tissue, respectively. In addition, lesions with gradually decreasing echoes from the peripheral to the central areas were designated as having centripetal reduction echogenicity. The color Doppler features of the lesions were assessed according to the vascularization distribution, which was classified as peripheral (peripheral blood flow signals only), internal (internal blood flow signals only), mixed (equal peripheral and internal blood flow signals), or none (no blood flow signals). All data analyses were performed with the SPSS version 16.0 statistical package (IBM Corporation, Armonk, NY). Data were presented as mean ± standard deviation and number (percentage). Statistical significance was defined as P < .05. χ2 and Fisher exact tests were used for comparisons of categorical variables, and Student t tests were used for comparisons of quantitative variables (ie, age and nodule size). To determine the independent sonographic predictors of atypical subacute thyroiditis from the sonographic characteristics that showed statistical significance, a multiple logistic regression model was used. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were also calculated.
Results The mean diameters of the nodules ± SD in the thyroiditis and malignancy groups were 1.9 ± 1.0 cm (range, 0.6–4.7 cm) and 1.1 ± 0.8 cm (range, 0.4–3.7 cm), respectively. The thyroiditis group had larger nodules than the malignancy group (P < .05). Of the 36 patients with atypical subacute thyroiditis, 6 had hyperthyroidism, and 7 had hypothyroidism. In addition, 11 patients had elevated anti-thyroglobulin and anti-thyroid peroxidase antibody levels. Thyroid gland enlargement was detected in 8 (22.2%) of the patients with atypical subacute thyroiditis in our study. The sonographic features of the groups are listed in Table 1. The nodule sizes of the thyroiditis group were primarily between 1.0 and 2.5 cm, and most of the sizes in
J Ultrasound Med 2015; 34:689–696
the malignancy group were less than 1.0 cm. No significant differences were detected for the variables of marked echogenicity and a taller-than-wide shape. The thyroiditis group was more prone to having poorly defined margins (Figure 1) than the malignancy group (87.3% versus 19.1%; P < .05). Irregular nodule shapes were observed in 32.7% of the thyroiditis group versus 17.3% of the malignancy group, and peripheral vascularity was detected in 34.5% of the thyroiditis group versus 17.3% of the malignancy group, but these results did not reach statistical significance. In the multiple logistic regression analysis to evaluate independent sonographic features for diagnosis of atypical subacute thyroiditis (Table 2), only the feature of a poorly defined margin showed a significant association with atypical subacute thyroiditis (P < .05). Importantly, 12 atypical subacute thyroiditis nodules showed a specific grayscale feature: ie, centripetal reduction echogenicity (Figures 2 and 3). This specific feature never appeared in the malignancy group. All of these 12 nodules were poorly defined, and 2 had an irregular shape. Three of these nodules were taller than wide, and 7 were round to ovoid in shape. The diameters of these 12 nodules were less than 1 cm in 3, 1 to 2.5 cm in 5, and greater than 2.5 cm in 4. Additionally, the frequency of positive color flow was similar in the thyroiditis and malignancy groups, and no significant differences were detected for mixed vascularity (Figure 4). Table 1. Sonographic Features of the Atypical Subacute Thyroiditis Malignancy Groups Feature
Thyroiditis (n = 55)
Malignancy (n = 110)
8 (14.5) 33 (60.0) 14 (25.5)
53 (48.2) 42 (38.2) 15 (13.6)
18 (32.7) 10 (18.2) 27 (49.1)
19 (17.3) 34 (30.9) 57 (51.8)
48 (87.3) 7 (12.7)
21 (19.1) 89 (80.9)
15 (27.3) 40 (72.7) 12 (21.8)
47 (42.7) 63 (57.3) 0 (0.0)
NA
27 (49.1) 19 (34.5) 0 (0.0) 9 (16.4)
51 (46.4) 19 (17.3) 27 (24.5) 13 (11.8)
>.99 .003 NA .429
Size, cm, n (%) 2.5 Shape, n (%) Irregular Taller than wide Ovoid to round Margin, n (%) Poorly defined Well defined Echogenicity, n (%) Markedly hypoechoic Hypoechoic Centripetal reduction Vascularity, n (%) None Peripheral Internal Mixed
P
ORIGINAL RESEARCH
Sonographic Features of Thyroid Nodules That May Help Distinguish Clinically Atypical Subacute Thyroiditis From Thyroid Malignancy Fu-shun Pan, MD, Wei Wang, MD, Yan Wang, MD, Ming Xu, MD, Jin-yu Liang, MD, Yan-ling Zheng, MD, Xiao-yan Xie, MD, Xiao-xi Li, MD
Objectives—The purpose of this study was to evaluate sonographic features for distinguishing clinically atypical subacute thyroiditis from malignant thyroid nodules. Methods—A total of 165 hypoechoic thyroid nodules without calcification in 135 patients with histologic diagnosis were included in this study. These nodules were classified into 2 groups: a thyroiditis group (55 nodules in 36 patients) and a malignancy group (110 nodules in 99 patients). The sonographic features of the groups were retrospectively reviewed. Results—No significant differences were detected for the variables of marked echogenicity, a taller-than-wide shape, and mixed vascularity. However, a poorly defined margin was detected more frequently in the thyroiditis group than the malignancy group (P < .05); it yielded a high capability for differential diagnosis of atypical subacute thyroiditis, with sensitivity and specificity of 87.3% and 80.9%, respectively. Centripetal reduction echogenicity was observed exclusively in the thyroiditis group, with high specificity (100%) but low sensitivity (21.8%) for atypical subacute thyroiditis diagnosis. All of the thyroiditis nodules with a positive color signal showed noninternal vascularity (negative predictive value, 100%).
Received May 5, 2014, from the Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound (F.P., W.W., Y.W., M.X., J.L., Y.Z., X.X.), and Department of Vascular and Thyroid Surgery (X.L.), First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China. Revision requested June 3, 2014. Revised manuscript accepted for publication July 8, 2014. This work was supported by the National Natural Science Foundation of China (grant 81301238). Address correspondence to Xiao-yan Xie, MD, Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, First Affiliated Hospital of Sun Yat-sen University, 510080 Guangzhou, Guangdong, China. E-mail: [email protected] doi:10.7863/ultra.34.4.689
Conclusions—There is a considerable overlap between the sonographic features of atypical subacute thyroiditis and thyroid malignancy. However, the margin, echogenicity, and vascularity type are helpful indicators for differential diagnosis of atypical subacute thyroiditis. Key Words—carcinoma; head and neck ultrasound; sonography; subacute thyroiditis
S
ubacute thyroiditis, also known as de Quervain thyroiditis, is the most common cause of thyroid pain. It is assumed to be postviral in origin, frequently preceded by an upper respiratory tract infection.1–2 The diagnosis of subacute thyroiditis is based on clinical and laboratory features, including cervical pain, swelling of the thyroid region, fever, occasional dysphasia, suppressed thyrotropin levels, and an elevated erythrocyte sedimentation rate, and pathologic diagnosis is rarely needed.2–4 Sonography is the first-line imaging modality for thyroid abnormalities, and several previous studies have revealed that sonography is a useful adjunctive tool for diagnosing and monitoring subacute thyroiditis. The characteristic sonographic features of sub-
©2015 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2015; 34:689–696 | 0278-4297 | www.aium.org
3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 690
Pan et al—Sonographic Features of Subacute Thyroiditis and Thyroid Malignancy
acute thyroiditis are thyroid enlargement and poorly defined and irregularly shaped hypoechoic thyroid lesions with an absence of color flow in the hypoechoic area.5–8 Thus, subacute thyroiditis cases with typical clinical and sonographic features may pose minimal diagnostic difficulties. However, some subacute thyroiditis cases are clinically atypical and do not feature cervical pain, fever, or serum thyroid dysfunction. In most of these atypical cases, which have been termed atypical subacute thyroiditis, the primary presentation consists of painless and hard thyroid nodules.2,9–13 An overlap in the clinical and sonographic features of atypical subacute thyroiditis and other types of thyroid nodules has been reported in several studies. Most frequently, atypical subacute thyroiditis nodules are mistaken for malignant lesions, which often results in unnecessary or improper therapy, including thyroid lobectomy.3,11,13,14 To date, the precise incidence of atypical subacute thyroiditis is not clear due to wide disparities in the various published studies.9,12,13 For example, Stein et al12 reported that 13 of 36 patients with subacute thyroiditis had no appreciable pain. However, with the broad application of sonographically guided fine-needle aspiration thyroid biopsy, we believe that the frequency of cytopathologic diagnosis of atypical subacute thyroiditis may be increasing. Thus, it will be an important concern for the differential diagnosis of this condition from other thyroid nodules, especially malignancy. To our knowledge, relatively few studies have described the sonographic features of atypical subacute thyroiditis.11,14 The purpose of our study was to evaluate sonographic features that distinguish atypical subacute thyroiditis from malignant thyroid nodules.
Materials and Methods This retrospective study was approved by the Institutional Review Board, with a waiver of informed consent. From January 2005 to June 2013, a total of 36 patients with 55 thyroid nodules were included in an atypical subacute thyroiditis group, which consisted of 8 male and 28 female patients ranging in age from 25 to 59 years (mean age, 46.5 years). All patients in the thyroiditis group denied having any cervical pain or tenderness during the last 6 months and had received histologic confirmation of subacute thyroiditis. Meanwhile, because all of the nodules in the thyroiditis group were solid and hypoechoic without calcification, another 110 thyroid nodules in 99 patients were randomly selected during the same period to comprise a malignancy group. The following inclusion criteria were used for the malignancy group: the presence of solid, noncalcified hypoechoic echogenicity on sonography and his-
690
tologic confirmation of the diagnosis. The malignancy group consisted of 19 male and 80 female patients ranging in age from 22 to 72 years (mean age, 45.0 years). The criteria for surgery were based on sonographic features that raised a suspicion of malignancy (taller-thanwide shape and marked hypoechogenicity), patients presenting with pressure symptoms, patient preference, and abnormal results of gun biopsies (either malignancy or other suspicious results, including atypical, inadequate, and nondiagnostic findings). Pathologic diagnoses for the thyroiditis group were obtained by specimens from gun biopsy (n = 5) and surgery (n = 50). All patients in the malignancy group underwent surgery with or without preceding gun biopsies. The pathologic diagnoses of 110 malignant nodules included papillary carcinoma (106 [96.4%]), squamous cell carcinoma (1 [0.9%]), thyroid lymphoma (2 [1.8%]), and anaplastic carcinoma (1 [0.9%]). Laboratory examination results including serum thyroid function, anti-thyroglobulin and antithyroid peroxidase antibodies, were available for all of the patients with atypical subacute thyroiditis within 2 weeks of surgery or biopsy. During the sonographic examination, each patient was in the supine position, and the anterior area of the neck was fully exposed. Thyroid sonography was performed with several pieces of equipment, including MyLab 90 (Esaote SpA, Genoa, Italy), Acuson Sequoia 512 (Siemens Medical Solutions, Mountain View, CA), and ProSound SSD-5000 (Aloka Co, Ltd, Tokyo, Japan) with linear array small-parts probes. Sonographic examinations were performed by 3 radiologists (Y.W., W.W., and M.X.) with more than 5 years of experience in thyroid sonography. Standard sonographic settings dedicated to the thyroid were used. The scanning protocol in all cases included both transverse and longitudinal real-time imaging of the thyroid nodules. Color Doppler sonography of all thyroid nodules included in this study was also performed before surgery or biopsy. The Doppler gain was set at an appropriate level to avoid any color noise. The interval from thyroid sonography to surgery was less than 2 weeks. The sonograms were independently analyzed by 2 other staff radiologists (J.L. and Y.Z.) in consensus, both of whom had more than 8 years of experience in thyroid sonography. Both specialists were blinded to the clinical history, previous imaging, and pathologic results corresponding to each sonogram. Grayscale sonographic features that were evaluated for this study included thyroid volume, lesion size, margin, shape, and echogenicity. The lesion size was assessed by the maximum diameter, and the lesions were classified into 3 subgroups: 1.0 cm or less, 1.0 cm to
J Ultrasound Med 2015; 34:689–696
3404jum553-720 copy_Layout 1 3/17/15 10:09 AM Page 691
Pan et al—Sonographic Features of Subacute Thyroiditis and Thyroid Malignancy
2.5 cm, and greater than 2.5 cm. The thyroid gland was considered to be enlarged when the anteroposterior diameter was greater than 2.0 cm. The margin was classified as either poorly defined (defined as a thyroid nodule that could not be discriminated from the peripheral thyroid parenchyma) or well defined. The shape of the nodule was classified as ovoid to round (anteroposterior dimension equal to or less than transverse dimension), taller than wide (anteroposterior dimension greater than transverse dimension), or irregular (neither ovoid to round nor taller than wide). Echogenicity was classified as markedly hypoechoic and hypoechoic with respect to the adjacent strap musculature and thyroid tissue, respectively. In addition, lesions with gradually decreasing echoes from the peripheral to the central areas were designated as having centripetal reduction echogenicity. The color Doppler features of the lesions were assessed according to the vascularization distribution, which was classified as peripheral (peripheral blood flow signals only), internal (internal blood flow signals only), mixed (equal peripheral and internal blood flow signals), or none (no blood flow signals). All data analyses were performed with the SPSS version 16.0 statistical package (IBM Corporation, Armonk, NY). Data were presented as mean ± standard deviation and number (percentage). Statistical significance was defined as P < .05. χ2 and Fisher exact tests were used for comparisons of categorical variables, and Student t tests were used for comparisons of quantitative variables (ie, age and nodule size). To determine the independent sonographic predictors of atypical subacute thyroiditis from the sonographic characteristics that showed statistical significance, a multiple logistic regression model was used. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were also calculated.
Results The mean diameters of the nodules ± SD in the thyroiditis and malignancy groups were 1.9 ± 1.0 cm (range, 0.6–4.7 cm) and 1.1 ± 0.8 cm (range, 0.4–3.7 cm), respectively. The thyroiditis group had larger nodules than the malignancy group (P < .05). Of the 36 patients with atypical subacute thyroiditis, 6 had hyperthyroidism, and 7 had hypothyroidism. In addition, 11 patients had elevated anti-thyroglobulin and anti-thyroid peroxidase antibody levels. Thyroid gland enlargement was detected in 8 (22.2%) of the patients with atypical subacute thyroiditis in our study. The sonographic features of the groups are listed in Table 1. The nodule sizes of the thyroiditis group were primarily between 1.0 and 2.5 cm, and most of the sizes in
J Ultrasound Med 2015; 34:689–696
the malignancy group were less than 1.0 cm. No significant differences were detected for the variables of marked echogenicity and a taller-than-wide shape. The thyroiditis group was more prone to having poorly defined margins (Figure 1) than the malignancy group (87.3% versus 19.1%; P < .05). Irregular nodule shapes were observed in 32.7% of the thyroiditis group versus 17.3% of the malignancy group, and peripheral vascularity was detected in 34.5% of the thyroiditis group versus 17.3% of the malignancy group, but these results did not reach statistical significance. In the multiple logistic regression analysis to evaluate independent sonographic features for diagnosis of atypical subacute thyroiditis (Table 2), only the feature of a poorly defined margin showed a significant association with atypical subacute thyroiditis (P < .05). Importantly, 12 atypical subacute thyroiditis nodules showed a specific grayscale feature: ie, centripetal reduction echogenicity (Figures 2 and 3). This specific feature never appeared in the malignancy group. All of these 12 nodules were poorly defined, and 2 had an irregular shape. Three of these nodules were taller than wide, and 7 were round to ovoid in shape. The diameters of these 12 nodules were less than 1 cm in 3, 1 to 2.5 cm in 5, and greater than 2.5 cm in 4. Additionally, the frequency of positive color flow was similar in the thyroiditis and malignancy groups, and no significant differences were detected for mixed vascularity (Figure 4). Table 1. Sonographic Features of the Atypical Subacute Thyroiditis Malignancy Groups Feature
Thyroiditis (n = 55)
Malignancy (n = 110)
8 (14.5) 33 (60.0) 14 (25.5)
53 (48.2) 42 (38.2) 15 (13.6)
18 (32.7) 10 (18.2) 27 (49.1)
19 (17.3) 34 (30.9) 57 (51.8)
48 (87.3) 7 (12.7)
21 (19.1) 89 (80.9)
15 (27.3) 40 (72.7) 12 (21.8)
47 (42.7) 63 (57.3) 0 (0.0)
NA
27 (49.1) 19 (34.5) 0 (0.0) 9 (16.4)
51 (46.4) 19 (17.3) 27 (24.5) 13 (11.8)
>.99 .003 NA .429
Size, cm, n (%) 2.5 Shape, n (%) Irregular Taller than wide Ovoid to round Margin, n (%) Poorly defined Well defined Echogenicity, n (%) Markedly hypoechoic Hypoechoic Centripetal reduction Vascularity, n (%) None Peripheral Internal Mixed
P
