Conventional ultrasound characteristics,TI-RADS category and shear wave speed measurement between follicular adenoma and follicular thyroid carcinoma

Abstract

The purpose of the study was to explore the differences of conventional ultrasound characteristics, thyroid imaging reporting and data system (TI-RADS) category and shear wave speed (SWS) measurement between follicular adenoma (FA) and follicular thyroid carcinoma (FTC). Twenty-eight FTCs and 67 FAs proven by surgery were retrospectively included for analysis. Conventional ultrasound and point-shear wave elastography (p-SWE) were performed in all of the included patients. The ultrasound features, American Thyroid Association (ATA) TI-RADS category and American College of Radiology (ACR) TI-RADS category, SWS measurement were compared between the two groups. Receiver operating characteristic (ROC) curve was performed and area under ROC curve (AUC) was obtained for significant features. There were no statistical differences in mean age (46.9±15.7years vs. 48.6±13.6years, P = 0.639), gender (9 males, 32.1% vs. 18 males, 29.0%, P = 0.766) and mean diameter (28.3±16.2 mm vs. 33.8±11.9 mm, P = 0.077) between FTCs and FAs. Hypoechogenicity, lobulated or irregular margin, macrocalcification were more common in FTCs than FAs (all P < 0.05). Mean SWS of FTCs (2.29±0.64 m/s) was slightly higher than that of FAs (1.94±0.68 m/s) (P = 0.023). The AUCs were 0.655, 0.744, and 0.744 with the cut-off SWS≥1.89 m/s, ACR TI-RADS category 4 and intermediate suspicion of ATA TI-RADS category. The sensitivity and AUC were 82.1% and 0.812 with combined ultrasound features of hypoechogenicity, lobulated or irregular margin and macrocalcification. In Conclusion, SWS measurement and TI-RADS categories were useful for the identification of FTCs from FAs.

Keywords

Follicular thyroid carcinoma (FTC)follicular adenoma (FA)ultrasound TI-RADS category Shear wave speed

1 Introduction

Follicular thyroid carcinoma (FTC) is the second malignancy in thyroid cancers (10–22%), followed by papillary thyroid carcinoma (PTC) (80–90%) [1, 2]. High resolution ultrasound examination is the preferred method for thyroid nodules [2]. The differential diagnosis for thyroid malignancy from all the thyroid nodules depends on the suspicious features on ultrasound images including solid, hypoechoic, irregular of microlobulated margin, taller than wide shape and microcalcifications with sensitivity of 26–87% and specificity of 53–93% [3, 4]. However, it remains difficult to distinguish FTC from follicular neoplasms (FN) before surgery in thyroid nodules management.

Thyroid imaging reporting and data system (TI-RADS) based on the above-mentioned features has been widely recognized in clinical practice. TI-RADS category has been widely used with sensitivity 75.0–89.6% for American Thyroid Association (ATA) TI-RADS category and 74.7–92.0% for American College of Radiology (ACR) TI-RADS category [5 –8]. However, these TI-RADS categories are extracted from overall thyroid carcinomas in which most of them are PTCs, it is unknown whether they are applicable for the diagnosis of FTC. Kim et al. reported that there were no statistically significant differences in tumor size and growth rate between FTC and FA, leading to that short-time follow up by US had no effect for differentiation of FTC and FA [9]. Another study reported that presence of calcifications, heterogeneous echo texture and solid component were risk factors for FTC, with odds ratios, 25.6, 24.9, and 9.4, respectively [10].

Radionuclide thyroid scan is recommended when the patient’s serum TSH is subnormal with or without thyroid nodules [2]. Focal [¹⁸F] fluorodeoxyglucose positron emission to mography (¹⁸FDG-PET) uptake is not recommended for those nodules >10 mm because of increased risk of thyroid cancer [2]. The utilization of SPECT/CT in FTC patients can detect the presence of metastases (lung, bone, adrenal, liver, and brain) and exclude potential false-positive lesions [11]. Contrast-enhanced ultrasound (CEUS) is a valuable supplemental method for thyroid carcinoma with 90% sensitivity and 86% specificity, while characteristics of FTC and FA had not been mentioned [12 –14].

Fine-needle aspiration (FNA) is the standard method to detect thyroid malignancy before surgery with sensitivities of 54–90% and specificities of 60–96% [2]. However, FNA is expensive and invasive and 17–22% of FNA results are indeterminate and unavailable. In addition, FNA cytology IV (follicular neoplasm or suspicious for a follicular neoplasm) indicates that FNA was unsuitable for FTC.

Elastography can be used to evaluate the hardness of tissue and hardness is considered to be related with malignancy. For shear wave elastography (SWE), transient pulses from the linear array probe are used to generate propagating shear waves in the region of interest (ROI) and shear wave speed (SWS) is calculated, which can quantitatively evaluate the tissue stiffness [15]. Point shear wave elastography (p-SWE) can obtain the mean SWS of the 6×5 mm sampling box by transmitting on point shear wave. P-SWE and/or combination of TI-RADS are proved to be useful in differential diagnosis between benign nodules and malignant nodules with sensitivity of 57–97% and specificity of 62–100% [16 –19]. Besides, it was reported that CEUS and SWE had similar diagnostic performance (accuracy 86.8% vs. 84.3%) and accuracy was increased to 93.1% from combination of CEUS and SWE [20].

Samir et al reported that SWE was helpful for thyroid malignancy in those nodules with fine needle aspiration (FNA) cytological results of FNA III (follicular lesion of undetermined significance or atypia of undetermined significance) and FNA IV (follicular neoplasm or suspicion for follicular neoplasm). The sensitivity and specificity was 82% and 88% with cut-off 22.3 kPa [21]. Therefore, we hypothesize that SWS measurement is statistically different between FTC and FA and SWS measurement is useful for the diagnosis of FTC.

Therefore, in this retrospective study, we assessed the differences of clinical features, conventional ultrasound features, ATA TI-RADS category, ACR TI-RADS category and SWS measurement between FTC and FA, the purpose of which was to find out the potential significant parameters and analyze their diagnostic performance for differentiation between FTC and FA.

2 Methods and materials

2.1 Patient selection

This retrospective study was approved by the Ethical Committee of the university affiliated hospital (approval number 19K100) and informed consent was waived. The inclusion criteria were as following: (a) patient age was 18–80 years old; (b) the patient received conventional ultrasound and p-SWE examination before surgery; (c) proven by surgery and pathological result was FTC or FA. 128 patients met the inclusion criteria. Patients were excluded when one of the following situations appeared: (a) the nodule was mostly cystic and the solid portion was smaller than 6×5 mm (n = 10); (b) the image data was incomplete (n = 17); (c) the pathological result of the target nodule was unclear or uncertain when the patient had multifarious pathological results (n = 11).

For one patient, only one thyroid nodule was included for analysis in the study. If nodules were multiple, the largest one was included for evaluation. Finally, 90 nodules (mean diameter, 32.1±13.6 mm; range, 8–63 mm) from 90 patients (27 males and 63 females; mean age, 48.0±14.2 years; range, 18–70 years) were included for analysis, including 28 FTCs (mean diameter, 28.3±16.2 mm; range, 9–58 mm) and 62 FAs (mean diameter, 33.8±11.9 mm; range, 8–63 mm).

2.2 Conventional ultrasound examination

All the incorporated patients underwent conventional ultrasound examination before surgery using a linear transducer (frequency: 10–12 MHz) from Siemens S2000 (Siemens Medical Solutions, Mountain View, Calif, USA) or Siemens S3000 (Siemens Medical Solutions, Mountain View, Calif, USA). The patients were lying in the examination couch and make sure the neck skin fully exposed. The coupling medium was used to reduce the interference of air between the linear transducer and skin. For each patient, transverse and longitudinal plane examination of gray scale and color Doppler flow image were performed. The diameter of tumor was the maximal diameter measured in longitudinal plane.

2.3 P-SWE examination

The target nodule was confirmed before p-SWE examination. ROI was placed on the solid area without macrocalcifications or cystic parts. The operator was asked to hold the transducer gently with minimum pressure on the patient neck skin. The patient was required to hold the breath during the acquisition of p-SWE. Generally, ROI was placed at the solid area without macrocalcifications. P-SWE was evaluated by SWS (m/s) measurement and effective SWS measurement was acquired at least 7 times. The SWS ranges from 0.5 to 8.4 m/s. The numbers displayed as “X.XX m/s” was replaced by 8.4 m/s corresponding to the solid portion and deleted when corresponding to the cystic portion. The mean SWS of the nodule was calculated after maximum and minimum were deleted. The whole process of conventional ultrasound and p-SWE examination lasted about 10 minutes.

2.4 Interpretation of conventional ultrasound features

The analyzable ultrasound features were as followed (classified according by thyroid ultrasound reporting lexicon from ACR): nodule composition (mixed cystic and solid, solid), echogenicity (hyperchoic, isoechoic, hypoechoic, mixehoic), shape (wider than tall shape, taller than wide shape), margin (smooth margin, ill-defined margin, lobulated or irregular margin), calcifications (none or large comet-tail artifacts, macrocalcifications, peripheral calcifications, microcalcifications). The echogenicity was compared with the surrounding thyroid tissue. Microcalcifications were defined as foci less than 1.5 mm. Vascularity was evaluated in the color Doppler flow imaging with velocity scale of 5–10 cm/s.

2.5 Interpretation of ATA and ACR TI-RADS category

ATA TI-RADS category [2] was classified as benign (risk of malignancy: <1%), very low suspicion (risk of malignancy: <3%), low suspicion (risk of malignancy: 5–10%), intermediate suspicion (risk of malignancy: 10–20%), high suspicion (risk of malignancy: 70–90%). ACR TI-RADS category [22] was classified as category 1 (0 point), category 2 (2 points), category 3 (3 points), category 4 (4–6 points), category 5 (≥7 points) according ultrasound features. The ultrasound features of thyroid nodules were divided into benign features (0 point each, including cystic or almost completely cystic, spongiform, large comet-tail artifacts and smooth margin), minimally suspicious features (1 point each, including hyperchoic or isoechoic, mixed cystic and solid, macrocalcifications), moderately suspicious features (2 points each, including solid or almost completely solid, hypoechoic, lobulated or irregular margin, peripheral calcifications) and highly suspicious features (3 points each, including markedly hypoechoic, extra-thyroidal extension, microcalcifications, taller than wide shape).

2.6 Statistical analysis

All the statistical analysis in the study was performed using SPSS 23.0 software. Continuous variables were expressed as mean±standard deviation (SD) and range (age, diameter, SWS). Enumeration data and categorical data were expressed as numbers (gender, ultrasound features, et al). The measurement data were compared using independent-samples t test while enumeration data using χ2 test (Fisher’s exact test was performed when one of numbers less than 5), and categorical data using non-parametric tests (echogenicity, calcifications, TI-RADS category). The receiver operating characteristic (ROC) curves of significant features were plotted and the the area under ROC curve (AUC) and its 95% confidence interval (CI) were obtained. The comparison of AUCs of different features was performed by using Z test. The cut-off value was acquired with the maximum of Youden index (sensitivity + specificity-1). Sensitivity, specificity, accuracy, positive predictive valve (PPV) and negative predictive value (NPV) of the significant features were calculated. A two-tailed P-value < 0.05 indicated statistically significant difference.

3 Results

3.1 Basic characteristics

The clinical information of FTC and FA was shown in Table 1. Finally, 28 FTCs (mean diameter, 28.3±16.2 mm) from 28 patients (9 males and 19 females) and 62 FAs (mean diameter, 33.8±11.9 mm) from 62 patients (27 males and 35 females) were included in the study. There were no statistically differences in mean diameter, gender and mean age (P = 0.077, 0.766, 0.639 respectively) between FTC and FA. Four FTCs had cervical lymph node metastasis and one of them also had distant metastasis (lung). 27 patients of the 28 FTC patients had single FTC while one patient (lung metastasis) had PTC and FTC in the two lobes respectively.

Table 1
Comparison of clinical features, ultrasound features and SWS between follicular thyroid carcinoma and follicular adenoma

Characteristics Follicular thyroid carcinoma Follicular adenoma P

No. of patients 28 62 /

No. of nodules 28 62 /

Location (left/right) 13/15 27/35 0.799

Mean diameter (mm) 28.3±16.2 33.8±11.9 0.077

Gender (male/female) 9/19 18/44 0.766

Mean age (years) 46.9±15.7 48.6±13.6 0.639

Nodule composition 0.166

Solid 25 (89.3%) 47 (75.8%)

Mixed cystic and solid 3 (10.7%) 15 (24.2%)

Echogenicity 0.001^*

Hyperchoic 1 (3.6%) 1 (1.6%)

Isoechoic 6 (21.4%) 25 (40.3%)

Hypoechoic 18 (64.3%) 13 (21.0%)

Mixehoic 3 (10.7%) 23 (37.1%)

Shape 0.073

Wider than tall 24 (85.7%) 60 (96.8%)

Taller than wide 4 (24.3%) 2 (3.2%)

Margin 0.001^*

Smooth 17 (60.7%) 59 (95.2%)

Ill-defined 2 (7.1%) 2 (3.2%)

Lobulated or irregular 9 (32.1%) 1 (1.6%)

Calcifications 0.001^*

None or large comet-tail artifacts 17 (60.7%) 56 (90.3%)

Macrocalcifications 10 (35.7%) 6 (9.7%)

Peripheral (rim) calcifications 0 (0.0%) 0 (0.0%)

Microcalcifications 1 (3.6%) 0 (0.0%)

Mean SWS (m/s) 2.29±0.64 1.94±0.68 0.023^*

Characteristics	Follicular thyroid carcinoma	Follicular adenoma	P
No. of patients	28	62	/
No. of nodules	28	62	/
Location (left/right)	13/15	27/35	0.799
Mean diameter (mm)	28.3±16.2	33.8±11.9	0.077
Gender (male/female)	9/19	18/44	0.766
Mean age (years)	46.9±15.7	48.6±13.6	0.639
Nodule composition			0.166
Solid	25 (89.3%)	47 (75.8%)
Mixed cystic and solid	3 (10.7%)	15 (24.2%)
Echogenicity			0.001^*
Hyperchoic	1 (3.6%)	1 (1.6%)
Isoechoic	6 (21.4%)	25 (40.3%)
Hypoechoic	18 (64.3%)	13 (21.0%)
Mixehoic	3 (10.7%)	23 (37.1%)
Shape			0.073
Wider than tall	24 (85.7%)	60 (96.8%)
Taller than wide	4 (24.3%)	2 (3.2%)
Margin			0.001^*
Smooth	17 (60.7%)	59 (95.2%)
Ill-defined	2 (7.1%)	2 (3.2%)
Lobulated or irregular	9 (32.1%)	1 (1.6%)
Calcifications			0.001^*
None or large comet-tail artifacts	17 (60.7%)	56 (90.3%)
Macrocalcifications	10 (35.7%)	6 (9.7%)
Peripheral (rim) calcifications	0 (0.0%)	0 (0.0%)
Microcalcifications	1 (3.6%)	0 (0.0%)
Mean SWS (m/s)	2.29±0.64	1.94±0.68	0.023^*

Note: SWS, shear wave speed; No., numbers;^*, statistical significance.

3.2 Conventional ultrasound features and SWS measurement

The ultrasound features and SWS of FTC and FA were presented in Table 1. There were statistical differences in echogenicity, margin and calcifications between FTC and FA. Hypoechogenicity, lobulated or irregular and macrocalcification were more common in FTCs than FAs (all P < 0.05) (Fig. 1). However, the highly suspicious characteristics such as taller than wide shape and microcalcification had no significant differences between FTCs and FAs (P = 0.073 and P = 0.311 respectively). Taller than wide shape and microcalcification were uncommon in FTCs. Mean SWS of FTCs was higher than that of FAs (2.29±0.64 m/s vs. 1.94±0.68 m/s, P = 0.023).

Fig.1

Ultrasound features of follicular thyroid carcinoma and follicular adenoma. A, the nodule is hypoechoic (FTC); B, the nodule is hypoechoic and irregular in shape (FTC); C, the hypoechoic nodule has marocalcifications (FTC); D, the mixed cystic and solid has smooth margin (FA); E, the nodule is solid and isoechoic (FA); F, the nodule is solid and hyperechoic (FA). FTC, follicular thyroid carcinoma; FA, follicular adenoma.

3.3 ATA TI-RADS category and ACR TI-RADS category

ATA TI-RADS category and ACR TI-RADS category of FTC and FA were presented in Table 2. There were statistical differences in ATA TI-RADS category and ACR TI-RADS category between FTCs and FAs (all P < 0.001) (Figs. 2 and 3). The highest frequency was low suspicion of ATA TI-RADS category (77.4%) and ACR TI-RADS category 3 (58.1%) in FAs while was high suspicion of ATA TI-RADS category (39.3%) and ACR TI-RADS category 4 (46.4%) for FTCs.

Fig.2

Follicular adenoma in a 59 years old man. A, the 37×24 mm nodule in left lobe of thyroid is solid and isoechoic on gray scale ultrosound; B, the nodule shows peripheral flow on color doppler flow imaging. The nodule is low suspicious of ATA TI-RADS category and is ACR TI-RADS category 3; C, SWS is 1.10 m/s on p-SWE measurement; D, the nodule is proven to be thyroid follicular adenoma pathologically (H&E staining, magnification, 10×10). ATA, American Thyroid Association; TI-RADS, thyroid imaging reporting and data system; ACR, American College of Radiology; SWS, shear wave speed; p-SWE, point shear wave elastography.

Fig.3

Follicular thyroid carcinoma in a 29 years old woman. A, the 13×8 mm nodule in right lobe of thyroid shows isoechogenicity, irregular margin and peripheral (rim) calcifications on gray scale ultrosound; B, the nodule shows peripheral flow on color doppler flow imaging. The nodule is intermediate suspicious on ATA TI-RADS category and is ACR TI-RADS category 4; C, SWS is 2.21 m/s on p-SWE measurement; D, the nodule is proven to be minimally invasive follicular thyroid carcinoma pathologically (H&E staining, magnification, 10×10). The capsule is incomplete and infiltrative by blood vessel (arrow). ATA, American Thyroid Association; TI-RADS, thyroid imaging reporting and data system; ACR, American College of Radiology; SWS, shear wave speed; p-SWE, point shear wave elastography.

Table 2

Comparison of ATA TI-RADS category and ACR TI-RADS category between follicular thyroid carcinoma and follicular adenoma

TI-RADS category	Follicular thyroid carcinoma (n = 28)	Follicular adenoma (n = 62)	P
ATA TI-RADS category			<0.001^*
Very low suspicion	0 (0.0%)	0 (0.0%)
Low suspicion	8 (28.6%)	48 (77.4%)
Intermediate suspicion	9 (32.1%)	12 (19.4%)
High suspicion	11 (39.3%)	2 (3.2%)
ATA category≥Intermediate suspicion	20/28 (71.4%)	14/62 (22.6%)	<0.001^*
ACR TI-RADS category			<0.001^*
TI-RADS 2	4 (14.3%)	12 (19.4%)
TI-RADS 3	4 (14.3%)	36 (58.1%)
TI-RADS 4	13 (46.4%)	12 (19.4%)
TI-RADS 5	7 (25.0%)	2 (3.2%)
TI-RADS category≥TI-RADS 4	20/28 (71.4%)	14/62 (22.6%)	<0.001^*
Number of significant features^#			<0.001^*
0	5 (17.9%)	44 (71.0%)
1	13 (46.4%)	17 (27.4%)
2	6 (21.4%)	1 (1.6%)
3	4 (14.3%)	0 (0.0%)
Number of significant features≥1	23/28 (82.1%)	18/62 (29.0%)	<0.001^*

Note: ACR, American College of Radiology; ATA, American Thyroid Association; TI-RADS, Thyroid Imaging Reporting and Data System;^#, significant features were hypoechogenicity, lobulated or irregular margin and macrocalcifications;^*, statistical significance.

3.4 The diagnostic performance of significant features

The diagnostic performance of conventional ultrasound features, SWS measurement and TI-RADS categories is presented in Table 3. Lobulated or irregular shape, macrocalcifications and SWS≥1.89 m/s had moderate diagnostic values with AUCs ranging 0.5–0.7. Hypoechogenicity, ATA TI-RADS category≥intermediate suspicion and ACR TI-RADS category≥4 had good diagnostic values with AUCs ranging 0.7–0.9.

Table 3
Diagnostic performance of significant parameters

Features Sensitivity, % Specificity, % Accuracy, % PPV, % NPV, % AUC

Hypoechogenicity 64.3 79.0 74.4 58.1 83.1 0.717

Lobulated or irregular shape 32.1 98.4 77.8 90.0 76.3 0.661

Macrocalcification 35.7 90.3 73.3 62.5 75.7 0.630

Number of significant features≥1 82.1 71.0 74.4 56.1 89.8 0.812

Mean SWS≥1.89 m/s 75.0 53.2 60.0 42.0 82.5 0.655

ATA TI-RADS category≥ 71.4 77.4 75.6 58.8 85.7 0.744

Intermediate suspicion

ACR TI-RADS category≥4 71.4 77.4 75.6 58.8 85.7 0.744

Features	Sensitivity, %	Specificity, %	Accuracy, %	PPV, %	NPV, %	AUC
Hypoechogenicity	64.3	79.0	74.4	58.1	83.1	0.717
Lobulated or irregular shape	32.1	98.4	77.8	90.0	76.3	0.661
Macrocalcification	35.7	90.3	73.3	62.5	75.7	0.630
Number of significant features≥1	82.1	71.0	74.4	56.1	89.8	0.812
Mean SWS≥1.89 m/s	75.0	53.2	60.0	42.0	82.5	0.655
ATA TI-RADS category≥	71.4	77.4	75.6	58.8	85.7	0.744
Intermediate suspicion
ACR TI-RADS category≥4	71.4	77.4	75.6	58.8	85.7	0.744

Note: SWS, shear wave speed; ACR, American College of Radiology; ATA, American Thyroid Association; TI-RADS, Thyroid Imaging Reporting and Data System; PPV, positive predictive value; NPV, negative predictive value; AUC, area under receiver operating characteristic curve.

The AUC of combined application in ultrasound significant feature (one or more suspicious features appeared) was statistically higher than that of macrocalcifications (0.812 vs. 0.630, P = 0.024), lobulated or irregular shape (0.812 vs. 0.661, P = 0.046), and SWS≥1.89 m/s (0.812 vs. 0.655, P = 0.034), while no statistical difference was found for hypoechogenicity (0.812 vs. 0.717, P = 0.200), ATA TI-RADS category≥intermediate suspicion (0.812 vs. 0.744, P = 0.354) and ACR TI-RADS category≥4 (0.812 vs. 0.744, P = 0.354). The highest valve belonged to the combined application of hypoechogenicity, lobulated or irregular shape and macrocalcifications with sensitivity 82.1% and AUC 0.812.

4 Discussion

The present study compared the clinical features, conventional ultrasound characteristics, SWS measurement, ATA and ACR TI-RADS category between FTC and FA, and explored the diagnostic performance of significant parameters in distinguishing FTC from FA. There were statistical differences in hypoechogenicity, lobulated or irregular margin, macrocalcifications, SWS measurement, ATA and ACR TI-RADS category between FTC and FA. The combined application of hypoechogenicity, lobulated or irregular margin, macrocalcifications had the highest sensitivity and AUC.

In previously studies [8 , 24], malignant thyroid nodules were smaller than benign nodules and more common in females and younger person. However, there were no differences in gender, mean age and mean diameter between FTCs and FAs in this study. Cystic degeneration was considered as a characteristic of FA [10] and the similar result was present in our study while no statistical difference (15 of 62 in FAs vs. 3 of 28 FTCs, P = 0.114). Most FAs were solid or almost completely solid, so solid component was not a differential point for FTCs. Hypoechogenicity, lobulated or irregular margin, macrocalcifications were the significant features on conventional ultrasound for FTC in the present study. It can be hypothesized that out of control of follicular cells growth leading to reduction of echogenicity and lobulated or irregular margin. In ATA TI-RADS category, hypoechogenicity was intermediate suspicion, lobulated or irregular margin was high suspicion and macrocalcifications was not specified. In ACR TI-RADS category, hypoechogenicity and lobulated or irregular margin was intermediate suspicion (2 points), macrocalcifications was low suspicion (1 point). Therefore, ATA and ACR TI-RADS category had higher classification in FTCs than FAs (P < 0.001).

Microcalcifications and taller than wide shape, which were highly suspicious features in thyroid nodules were rare in FTC (3.6% and 24.3%), markedly less than in PTC (40.4% and 51.2%) [8]. Thus, 54.2% malignant thyroid nodules were high suspicion for ATA TI-RADS category while only 39.3% FTCs in our study. For ACR TI-RADS category, 52.2% malignant thyroid nodules were category 5 while only 25.0% in this study [5]. ATA and ACR RI-RADS category had equivalently diagnostic value for FTC and FA, the sensitivity was 71.4% and AUC was 0.744 with intermediate suspicion of ATA TI-RADS category and ACR TI-RADS category 4. ATA and ACR TI-RADS category were useful for FTC and FA (sensitivity 71.4%), while the diagnostic performance was inferior to the previous studies that TI-RADS categories were performed in overall nodules (Sensitivity 74.7–98.3%) [25 –28]. The sensitivity and AUC were improved to 82.1% and 0.812 when hypoechogenicity, lobulated or irregular margin and macrocalcifications were combined.

It was reported that 37.2% PTC had multifocal malignant tumors, multifocal PTC had higher invasiveness, cervical lymph node metastasis, TNM staging and recurrence [29, 30]. Literatures about multifocal FTC had not been found, likewise in our case there had no multifocal FTC. It is interesting to note that one patient with PTC in the left and FTC in the right resulted had lung metastasis. It was unclear whether multiple types of thyroid cancer promote the occurrence of metastasis.

SWS measurement of p-SWE can quantitatively evaluate the stiffness of thyroid nodules and the tissue stiffness is associated with malignancy. In this study, the mean SWS of FTCs were statistical higher than FAs, similar to the result of previous study that SWS of malignant nodules was larger than that of benign nodules [17]. However, the difference of mean SWS between FTC and FA (2.29±0.64 m/s vs. 1.94±0.68 m/s) was much smaller than that of malignant nodules and benign nodules (4.90±2.74 m/s vs. 2.10±1.04 m/s) [18]. Besides, the diagnostic performance was unsatisfactory with sensitivity 75.0%, specificity 53.2% and AUC 0.655 compared with the previous study with most of the malignancy being PTCs (sensitivity 82%, specificity 90% and AUC 0.86) [18]. The reason probably was that the pasmmoma bodies and fibrosis which markedly increase the stiffness of tissue were uncommon in FTC than PTC.

It was difficult to diagnose FTC from FN before surgery for that FNA cytology and frozen pathology had a large proportion of false negative rate [31 –33]. However, in the present study, there were statistical differences in hypoechogenicity, lobulated or irregular shape, macrocalcification, SWS measurement, ATA and ACR TI-RADS category between FTC and FA. Therefore, comprehensive evaluation of these significant ultrasound features will be helpful to preoperatively predict malignancy and surgical planning in FNs, meanwhile possible disputes with patients can be reduced or avoided.

There had several limitations in this study. Firstly, it was a retrospective study. Prospective studies with large sample and long time follow-up results were needed for further analysis. Secondly, the sample size was small (28 FTCs and 62 FAs). More cases and long-term follow-up were needed for further study. In addition, the study didn’t include some relevant technologies in thyroid management (contrast-enhanced ultrasound, fine needle aspiration, gene mutation detection, et al).

5 Conclusion

Hypoechogenicity, lobulated or irregular shape, macrocalcification, SWS measurement, ATA and ACR TI-RADS category were useful for differential diagnosis between FTC and FA. The combinations of suspicious features, SWS measurement, ATA and ACR TI-RADS category may provide differential diagnosis value and help surgeons in making decisions regarding with selection of thyroid surgery or operation methods.

Conflicts of interest

The authors declare that there are no conflicts of interest regarding the publication of this article.

Footnotes

Acknowledgments

This work was supported in part by the National Natural Science Foundation of China (Grants 81671695, 81601502, 81725008, and 81927801), Fundamental Research Funds for the Central Universities (Grants 22120190021 and 22120190137), Shanghai Hospital Development Center (Grants 16CR3061B and SHDC12014229), and the Science and Technology Commission of Shanghai Municipality (Grants 19441903200, 16411971100 and 14441900900).

References

Carling

, Udelsman

. Follicular neoplasms of the thyroid: what to recommend. Thyroid. 2005;15(6):583–7.

Haugen

, Alexander

, Bible

, et al. 2015 American thyroid association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American thyroid association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid. 2016;26(1):1–133.

Brito

, Gionfriddo

, Al

, et al. The accuracy of thyroid nodule ultrasound to predict thyroid cancer: systematic review and meta-analysis. J Clin Endocrinol Metab. 2014;99(4):1253–63.

Nam

, Shin

, Han

, et al. Preoperative ultrasonographic features of papillary thyroid carcinoma predict biological behavior. J Clin Endocrinol Metab. 2013;98(4):1476–82.

, Na

, Baek

, Sung

, Kim

, Kang

. US fine-needle aspiration biopsy for thyroid malignancy: diagnostic performance of seven society guidelines applied to 2000 thyroid nodules. Radiology. 2018;287(3):893–900.

Hoang

, Middleton

, Farjat

, et al. Reduction in thyroid nodule biopsies and improved accuracy with American college of radiology thyroid imaging reporting and data system. Radiology. 2018;287(1):185–93.

Trimboli

, Deandrea

, Mormile

, Ceriani

, Garino

, Limone

, Giovanella

. American thyroid association ultrasound system for the initial assessment of thyroid nodules: use in stratifying the risk of malignancy of indeterminate lesions. Head Neck. 2018;40(4):722–7.

Yoon

, Lee

, Kim

, Moon

, Kwak

. Malignancy risk stratification of thyroid nodules: comparison between the thyroid imaging reporting and data system and the 2014 American thyroid association management guidelines. Radiology. 2016;278(3):917–24.

Kim

, Han

, Lee

, et al. Tumour growth rate of follicular thyroid carcinoma is not different from that of follicular adenoma. Clin Endocrinol (Oxf). 2018;88(6):936–42.

10.

Zhang

, Hu

. Sonographic features of thyroid follicular carcinoma in comparison with thyroid follicular adenoma. J Ultrasound Med. 2014;33(2):221–7.

11.

Zhao

, Shen

, Liu

, Kuang

. Unusual adrenal and brain metastases from follicular thyroid carcinoma revealed by ¹³¹I SPECT/CT. Clin Nucl Med. 2016;41:e53–5.

12.

, Zhang

, Ling

, Liu

, Jia

, Zhu

, Wang

, Liu

, Huang

, Liu

. Contrast-enhanced sonography for the identification of benign and malignant thyroid nodules: Systematic review and meta-analysis. J Clin Ultrasound. 2016;44:199–209.

13.

Platz Batista da Silva

, Jung

, Schlitt

, Hornung

. VueBox® perfusion analysis of contrast-enhanced ultrasound (CEUS) examinations in patients with primary hyperparathyroidism for preoperative detection of parathyroid gland adenoma. Clin Hemorheol Microcirc. 2018;70:423–31.

14.

Zhan

, Diao

, Chen

, Wang

, Ding

. Predicting cervical lymph node metastasis in patients with papillary thyroid cancer (PTC) - Why contrast-enhanced ultrasound (CEUS) was performed before thyroidectomy. Clin Hemorheol Microcirc. 2019;72:61–73.

15.

Shiina

, Nightingale

, Palmeri

, Hall

, Bamber

, Barr

, Castera

, Choi

, Chou

, Cosgrove

, Dietrich

, Ding

, Amy

, Farrokh

, Ferraioli

, Filice

, Friedrich-Rust

, Nakashima

, Schafer

, Sporea

, Suzuki

, Wilson

, Kudo

. WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part basic principles and terminology. Ultrasound Med Biol. 2015;41(5):1126–47.

16.

Cosgrove

, Barr

, Bojunga

, Cantisani

, Chammas

, Dighe

, Vinayak

, Xu

, Dietrich

. WFUMB guidelines and recommendations on the clinical use of ultrasound elastography: Part 4. Thyroid. Ultrasound Med Biol. 2017;43(1):4–26.

17.

Liu

, Li

, Xu

, Guo

, Zhang

, Xu

, Liu

, Li

, Xu

, Qu

, Xing

. Quantitative shear wave velocity measurement on acoustic radiation force impulse elastography for differential diagnosis between benign and malignant thyroid nodules: a meta-analysis. Ultrasound Med Biol. 2015;41:3035–43.

18.

, Xu

, Zhang

, Guo

, Liu

, Zheng

. Conventional US, US elasticity imaging, and acoustic radiation force impulse imaging for prediction of malignancy in thyroid nodules. Radiology. 2014;272(2):577–86.

19.

, Hou

, Du

, Wu

, Wang

, Zhou

. Virtual touch tissue imaging and quantification (VTIQ) combined with the American college of radiology thyroid imaging reporting and data system (ACR TI-RADS) for malignancy risk stratification of thyroid nodules. Clin Hemorheol Microcirc. 2019;72:279–91.

20.

Chen

, Zhang

, Xu

, Zhang

, Cao

, Sun

. Shear wave elastography and contrast-enhanced ultrasonography in the diagnosis of thyroid malignant nodules. Mol Clin Oncol. 2016;5:724–30.

21.

Samir

, Dhyani

, Anvari

, Prescott

, Halpern

, Faquin

, Stephen

. Shear-wave elastography for the preoperative risk stratification of follicular-patterned lesions of the thyroid: diagnostic accuracy and optimal measurement plane. Radiology. 2015;277(2):565–73.

22.

Tessler

, Middleton

, Grant

, et al. ACR thyroid imaging, reporting and data system (TI-RADS): white paper of the ACR TI-RADS committee. J Am Coll Radiol. 2017;14(5):587–95.

23.

Singh Ospina

, Brito

, Maraka

, Espinosa de Ycaza

, Rodriguez-Gutierrez

, Gionfriddo

, Castaneda-Guarderas

, Benkhadra

, Al Nofal

, Erwin

, Morris

, Castro

, Montori

. Diagnostic accuracy of ultrasound-guided fine needle aspiration biopsy for thyroid malignancy: systematic review and meta-analysis. Endocrine. 2016;53:651–61.

24.

Friedrich-Rust

, Vorlaender

, Dietrich

, Kratzer

, Blank

, Schuler

, Broja

, Cui

, Herrmann

, Bojunga

. Evaluation of strain elastography for differentiation of thyroid nodules: results of a prospective DEGUM multicenter study. Ultraschall Med. 2016;37:262–70.

25.

Basha

, Alnaggar

, Refaat

, El-Maghraby

, Refaat

, Abd Elhamed

, Abdalla

, Aly

, Hanafy

, Mohamed

, Afifi

, Harb

. The validity and reproducibility of the thyroid imaging reporting and data system (TI-RADS) in categorization of thyroid nodules: multicentre prospective study. Eur J Radiol. 2019;117:184–92.

26.

Buda

, Wildman-Tobriner

, Hoang

, Thayer

, Tessler

, Middleton

, Mazurowski

. Management of thyroid nodules seen on US images: deep learning may match performance of radiologists. Radiology. 2019:181343.

27.

Chen

, Zhan

, Diao

, Liu

, Shi

, Chen

, Zhan

. Additional value of superb microvascular imaging for thyroid nodule classification with the thyroid imaging reporting and data system. Ultrasound Med Biol. 2019;45(8):2040–8.

28.

Jin

, Yu

, Mo

, Su

. Clinical study of the prediction of malignancy in thyroid nodules: modified score versus American college of radiology’s thyroid imaging reporting and data system ultrasound lexicon. Ultrasound Med Biol. 2019;45(7):1627–37.

29.

Wang

, Yu

, Shen

, Zhu

, Huang

, Liu

, Viola

, Elisei

, Puxeddu

, Fugazzola

, Colombo

, Jarzab

, Czarniecka

, Lam

, Mian

, Vianello

, Yip

, Riesco-Eizaguirre

, Santisteban

, O’Neill

, Sywak

, Clifton-Bligh

, Bendlova

, Sýkorová

, Wang

, Liu

, Zhao

, Xing

. The prognostic value of tumor multifocality in clinical outcomes of papillary thyroid cancer. J Clin Endocrinol Metab. 2017;102:3241–50.

30.

Kiriakopoulos

, Petralias

, Linos

. Multifocal versus solitary papillary thyroid carcinoma. World J Surg. 2016;40:2139–43.

31.

Liu

, Medici

, Kwong

, Angell

, Marqusee

, Kim

, Larsen

, Cho

, Nehs

, Ruan

, Gawande

, Moore

Jr , Barletta

, Krane

, Cibas

, Yang

, Alexander

. Bethesda categorization of thyroid nodule cytology and prediction of thyroid cancer type and prognosis. Thyroid. 2016;26(2):256–61.

32.

Romero-Rojas

, Diaz-Perez

, Mastrodimos

, Chinchilla

. Follicular thyroid carcinoma with signet ring cell morphology: fine-needle aspiration cytology, histopathology, and immunohistochemistry. Endocr Pathol. 2013;24(4):239–45.

33.

Sugino

, Ito

, Nagahama

, Kitagawa

, Shibuya

, Ohkuwa

, Yano

, Uruno

, Akaishi

, Suzuki

, Masaki

, Ito

. Diagnostic accuracy of fine needle aspiration biopsy cytology and ultrasonography in patients with thyroid nodules diagnosed as benign or indeterminate before thyroidectomy. Endocr J. 2013;60(3):375–82.