Abstract
Objective:
To investigate whether pre-operative ultrasound elastography (USE) can be used to predict aggressive features of solitary papillary thyroid carcinomas (PTCs).
Methods:
Clinical and USE indices were retrospectively analyzed in 487 patients with surgically confirmed solitary PTCs. The patients were grouped per aggressive features on pathologic testing. Univariate and binary logistic regression analyses were performed to explore independent risk factors of aggressive features.
Results:
Univariate analysis revealed standard deviation (SD) values of the tumor shear-wave velocity (SWV) were associated with capsular invasion (p < 0.05). Further, shear-wave elasticity and SWV ratios correlated with extrathyroidal extension (all p < 0.05). The tumor shear-wave elasticity and SWV SD values were associated with cervical lymph node metastasis (CLNM) (all p < 0.05). Binary logistic regression analysis identified location and capsule contact as independent predictive risk factors for capsular invasion (all p < 0.05); size for extrathyroidal extension (all p < 0.05); and sex, age, margin, and suspected CLNM for CLNM (all p < 0.05). However, pre-operational USE indexes were not independent predictors of aggressive features (all p > 0.05).
Conclusion:
Pre-operative USE indices were not independent risk factors of aggressive features of solitary PTCs. Thus, USE may have a limited value for predicting the aggressive features of PTC.
Advances in knowledge:
Pre-operative USE indices may have a limited value for predicting the aggressive features of PTC.
Introduction
The incidence of papillary thyroid cancer (PTC) has been increasing in recent years. 1 However, the majority of PTCs are mostly indolent with an excellent prognosis. 2 Given this context, the clinical guidelines for PTC treatment have increasingly emphasized limited surgery to prevent overtreatment. 3,4 Recently, the minimally invasive ultrasound-guided thermal ablation technique, is increasingly being used in clinical practice. 5–7 Active surveillance, as a conservative treatment strategy for PTC, has also been receiving increasing amounts of attention. 8,9 The greatest challenge for these changing treatment strategies is to ensure precise pre-operative stratification of the aggressive features of PTC. Only by accurately stratifying the aggressive features of PTCs can clinicians choose individualized treatment options.
Currently, pre-operative conventional ultrasound (CUS) is the most used imaging modality for differentiating between benign and malignant thyroid nodules, determining cervical lymph node metastasis (CLNM), and assisting in fine needle aspiration (FNA) examinations. 10–14 However, CUS has a limited value in the pre-operative assessment of the aggressive features of PTC. For example, although CUS has high sensitivity and specificity for the pre-operative diagnosis of lateral CLNM, it is difficult to accurately assess it in the central compartment owing to the interference by gas and bone. 15–17 In addition, CUS has low sensitivity for determining minimal extrathyroidal extension (ETE). 18
Ultrasound elastography (USE) is an additional tool to the CUS and can be used to assess tissue stiffness. It has been shown to play a substantial role in the differential diagnosis of benign and malignant thyroid nodules, as well as in the reduction of unnecessary FNA. 19–26 However, USE has been used in relatively few studies to predict the aggressive features of PTCs, and some findings are inconsistent. 27–31 Therefore, we performed a retrospective study to determine if pre-operative USE could predict aggressive features in patients with pathologically confirmed solitary PTCs.
Patients and methods
Patients
The study was approved by the ethics committee of Shanghai General Hospital and the informed consent was waived. Between July 2019 and December 2021, 768 patients with suspected PTC based on FNA examination results were included in this study. All patients were scheduled to undergo surgical treatment. The inclusion criteria were as follows: (a) pre-operative FNA results indicating PTC and (b) CUS and USE performed before the FNA examination. The exclusion criteria were as follows: (a) patients who did not undergo surgery due to various reasons; (b) patients who received thermal ablation treatment; and (c) patients with multiple PTCs confirmed by surgery. Finally, a total of 487 patients with a solitary PTC were enrolled (125 men and 362 women). The mean patient age was 43.09 ± 0.54 years (range, 17–80 years; median, 42 years), and the mean tumor size was 8.89 ± 0.28 mm (range, 2.00–44.10 mm; median, 5.30 mm). The flow chart of patients’ enrollment is shown in Figure 1.
Figure 1.
Flow chart of selection with PTC. FNA, fine needle aspiration; PTC, papillary thyroid carcinoma
Ultrasound examination
Both CUS and USE were performed using a commercially available ultrasound instrument (Aplio i900, Canon Medical Systems, Tochigi, Japan) with a linear array probe (frequency range, 5–18 MHz). Ultrasound examinations were performed by the same radiologist (with 10 years of experience in thyroid ultrasound and 5 years of experience in thyroid USE). Patients were examined in the supine position with a pillow under the shoulders to maintain neck extension. The focus, gain, and time intensity compensation were set to display the tumor and the surrounding thyroid parenchyma (STP) clearly. First, grayscale ultrasound was performed. After a slow, continuous dynamic scan of the entire thyroid gland in both cross-sectional and longitudinal views, detailed, multisectional ultrasound scanning of the thyroid nodules was initiated when a thyroid nodule was identified. The grayscale ultrasound scan of the thyroid nodule was stored as a cine clip. During the dynamic scan of a thyroid nodule, the still images showing the largest diameter of the thyroid nodule, and the transverse and sagittal plane images of the nodules were routinely stored. Second, color doppler ultrasound was performed on the thyroid and thyroid nodules. Still images showing the typical color doppler flow of the thyroid and nodules were obtained and stored. In addition to the thyroid and nodules, careful scans of the bilateral cervical lymph nodes of levels I through VI were performed with attention to the lymph node size, echogenic changes, morphology, clarity of the lymphatic portals, and color flow perfusion of the lymph nodes. All cine lopes and still images of the thyroid, nodules and cervical lymph nodes were stored in our ultrasound Imaging and Reporting System.
CUS index evaluation
On CUS exanimation, tumor size was defined as the largest diameter measured on multiple views. The location was classified as right, left, or isthmus. Echogenicity was classified as hypo-, hyper-, and isoechoic according to the tumor and the STP echo differences. Margin was recorded as clear or unclear, while the shape was classified as regular or irregular. The aspect ratio was calculated as the anteroposterior diameter divided by the left-right diameter in the transverse plane. On color doppler ultrasound, vascularity was classified into four types: Type I, no doppler signal in the tumor and STP; Type II, rich peritumoral signal and scattered intratumoral signal; Type III, slight signal in the tumor and the STP; and Type IV, marked intratumoral signal. The calcification was categorized as follows: Type I, no calcification; Type II, microcalcifications (punctate calcifications); Type III, coarse macrocalcifications; and Type IV, mixed type (including micro- and macrocalcifications). Capsule contact (CC) was defined as the absence of normal thyroid parenchyma between the tumor and the capsule. Discontinuous capsule echo (DCE) was noted when there was a discontinuity of the capsule echo. The criteria for suspected CLNM (SCLNM) were as follows: round shape, loss of fatty hilum, microcalcifications, cystic change, and chaotic or peripheral blood flow. 14,32,33 Information on the presence of SCLNM was obtained from the patients’ pre-operative ultrasound report in our ultrasound Imaging and Reporting System.
USE examination and evaluation
USE examination was performed in the longitudinal views before the FNA examination. The shear-wave elastography (SWE) modality was used. During SWE, the probe was gently placed on the skin. The “two-dimensional” and “one-shot” modes were used. A rectangle sampling box as large as possible was placed to include the tumor and STP. The scale was set from 0 to 145 KPa. SWE was performed five times for each patient. Five SWE images were obtained and stored for the subsequent measurements. Circular ROIs of the tumor were determined along with the STP at a similar depth avoiding the calcified and cystic areas. The diameter of ROIs was 3 mm for all tumors with size ≥3 mm, while an ROI of 1 mm was used for tumors with diameters of <3 mm. The following indices were calculated automatically: SWE_tumor, the SWE value of the tumor; SWE_SD_tumor, one standard deviation of the tumor SWE; SWE_STP, the SWE value of the STP; SWE_SD_STP, one standard deviation of the SWE_STP; and SWE ratio, the SWE ratio of the tumor to that of the STP. Since there is a fixed relationship between the Young modulus and the shear-wave velocity (SWV), the ultrasound instrument also provided the following corresponding SWV indices (Figure 2): SWV_tumor, SWV_SD_tumor, SWV_STP, SWV_SD_STP, and SWV ratio. The average value of each parameter was calculated based on five measurements based on five different locations of the tumor on five images stored previously and was finally recorded as the elasticity value.
Figure 2.
The SWE images. A patient with a suspected papillary thyroid carcinoma on fine needle aspiration underwent the SWE examination. Two ROIs with the same size and similar location were placed in the tumor (T1) and STP (T2), respectively. The average value and one standard deviation of elastic modulus and of the tumor (Ave.T1, SD.T1) and the STP (Ave.T2, SD.T2) were obtained (2A), while the average value and one standard deviation of the shear-wave velocity of the tumor (Ave.T1, SD.T1) and the STP (Ave.T2, SD.T2) were presented simultaneously (2B). Ave., average value; ROI, region of interest; SD, one standard deviation; STP, surrounding thyroid parenchyma; SWE, shear wave elasticity.
Interrater reliability and measurement consistency of CUS and USE indices
In CUS, seven qualitative indicators such as echogenicity, margin, shape, vascularity, calcification, CC, and DCE were independently assessed by two radiologists with more than 15 years of experience in thyroid ultrasound examination. The indices of calcification, CC and DCE were evaluated based on the previously stored cine clips, while other indicators were evaluated using the still images. Negotiated consensus results were considered for any inconsistent assessments. All radiologists were blinded to aggressive features. The interrater agreement for these qualitative indicators and measurement consistency of USE indicators were evaluated.
Acquisition of patients’ clinical datan
Patients' clinical data, such as age, sex, and Hashimoto’s thyroiditis (HT), were obtained from the medical records. HT was based on pathological results. Ipsilateral central lymph node dissection (CLND) was performed in patients who underwent hemithyroidectomy, while bilateral CLND was performed with total or near-total thyroidectomy. Lateral lymph node dissection (LLND) was performed based on pre-operative FNA results and intraoperative findings. Pathologic testing revealed the aggressive features were capsular invasion, ETE, and CLNM.
Statistical analysis
Data were analyzed using SPSS software (v. 23.0; SPSS, Inc.) Quantitative data were expressed as the mean ± standard deviation or medians and interquartile range. Categorical variables were expressed as numbers and percentages. T-tests or non-parametric tests were used for quantitative data comparisons; categorical variables were compared using χ2 tests or Fisher’s exact probability method. To explore the independent risk factors for predicting aggressive features, univariate analysis and binary logistic analysis were performed. The interclass correlation coefficient (ICC) was used to confirm the measurement consistency according to the following criteria: excellent (ICC >0.75), good (ICC: 0.60–0.75), fair (ICC: 0.40–0.60), and poor (ICC ≤0.40). Cohen’s κ (κ) values were obtained to measure interrater reliability. A two-way random model with absolute agreement method was used. κ < 0.20 indicated poor agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, substantial agreement; and 0.81–1.00, almost perfect agreement. p < 0.05 indicated that the difference was statistically significant.
Results
Surgical and pathological results
364 patients underwent hemithyroidectomy and ipsilateral CLND, including 4 patients who underwent ipsilateral LLND; 123 patients underwent total or near-total thyroidectomy and bilateral CLND, including 15 patients who underwent ipsilateral LLND and 1 patient who underwent bilateral LLND. Of the 487 patients enrolled, 180 (37.0%) patients showed capsular invasion, 61 (12.5%) patients had ETE, and 192 (39.4%) patients had CLNM.
Patients’ characteristics
The baseline characteristics are summarized in Table 1. Sex was not significantly different in relation to capsular invasion (p = 0.966) and ETE (p = 0.173). However, sex differed significantly between the CLNM and non-CLNM groups (p < 0.001). No statistically significant difference in age was observed in groups categorized by the presence of capsular invasion and ETE (p = 0.392 and 0.817, respectively). However, patients in the CLNM group were younger than those in the non-CLNM group (p < 0.001). Larger tumor size was more frequently observed in the capsular invasion, ETE, and CLNM groups (all p < 0.001). PTC with capsular invasion was more frequently observed in the right lobe (p = 0.026), but the location was not significantly different in relation to the incidence of ETE and CLNM (p = 0.769). No significant differences were found for HT in capsular invasion, ETE, and CLNM across groups (p = 0.335, 0.088, and 0.166, respectively).
Table 1.
Comparison of clinical characteristics according to the aggressive features of PTC
| Characteristic | Capsular invasion | ETE | CLNM | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No (N = 307) |
Yes (N = 180) |
p-value | No (N = 426) |
Yes (N = 61) |
p-value | No (N = 295) |
Yes (N = 192) |
p-value | ||
| Sex | Male | 79 (25.7） | 46 (25.6） | 0.966 | 105 (24.6） | 20 (32.8） | 0.173 | 55 (18.6） | 70 (36.5） | < 0.001a |
| Female | 228 (74.3） | 134 (74.4） | 321 (75.4） | 41 (67.2） | 240 (81.4） | 122 (63.5） | ||||
| Age (years) | 42 (17.0） | 41 (20.0） | 0.392 | 42 (18.3） | 41 (20.0） | 0.817 | 44 (19.0) | 38 (16.5) | < 0.001a | |
| Size (mm) | 6.6 (4.9） | 8.3 (6.8） | < 0.001a | 6.9 (5.0） | 9 (7.2） | < 0.001 | 6.6 (4.3) | 8.1 (6.8) | < 0.001a | |
| Location | Right | 136 (44.3） | 101 (56.1） | 0.026a | 207 (48.6） | 30 (49.2） | 0.894 | 144 (48.8） | 93 (48.4） | 0.769 |
| Left | 160 (52.1） | 71 (39.4） | 203 (47.7） | 28 (45.9） | 141 (47.8） | 90 (46.9） | ||||
| Isthmus | 11 (3.6） | 8 (4.4） | 16 (3.8） | 3 (4.9） | 10 (3.4） | 9 (4.7） | ||||
| HT | No | 225 (73.3） | 139 (77.2） | 0.335 | 313 (73.5） | 51 (83.6） | 0.088 | 214 (72.5) | 150 (78.1) | 0.166 |
| Yes | 82 (26.7） | 41 (22.8） | 113 (26.5） | 10 (16.4） | 81 (27.5) | 42 (21.9) | ||||
PTC = papillary thyroid carcinoma;N = number of patients; ETE = extrathyroidal extension; CLNM = cervical lymph node metastasis; HT = Hashimoto’s thyroiditis.
Continuous variables are expressed as median values (interquartile range); categorical variables are expressed as numbers (percentages).
“a” indicates statistical differences.
Univariate analysis results
Echogenicity, CC, DCE, and SWV_tumor_SD were significantly higher in patients with capsular invasion (p = 0.026, < 0.001, 0.001, and 0.047, respectively). ETE was significantly associated with SCLNM, SWE_ratio, and SWV_ratio (p = 0.037, 0.005, and 0.007, respectively). CLNM was significantly correlated with an unclear margin, SCLNM, and higher values of SWE_tumor_SD and SWV_tumor_SD (p = 0.005, 0.001, 0.015, and 0.010, respectively) (Table 2).
Table 2.
Comparison of CUS and USE characteristics according to the aggressive features of PTC
| Characteristic | Capsular invasion | ETE | CLNM | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No (N = 307) |
Yes (N = 180) |
p-value | No (N = 426) |
Yes (N = 61) |
p-value | No (N = 295) |
Yes (N = 192） |
p- Value |
||
| Echogenicity | Hypo- | 136 (44.3） | 101 (56.1） | 0.026a | 381 (89.4） | 55 (90.2） | 0.097 | 267 (90.5） | 169 (88.0） | 0.530 |
| Hyper- | 160 (52.1） | 71 (39.4） | 6 (1.4） | 3 (4.9） | 4 (1.4） | 5 (2.6） | ||||
| Iso- | 11 (3.6） | 8 (4.4） | 39 (9.2） | 3 (4.9） | 24 (8.1） | 18 (9.4） | ||||
| Margin | Clear | 131 (42.7） | 77 (42.8） | 0.982 | 185 (43.4） | 23 (37.7） | 0.398 | 141 (47.8） | 67 (34.9） | 0.005a |
| Unclear | 176 (57.3） | 103 (57.2） | 241 (56.6） | 38 (62.3） | 154 (52.2） | 125 (65.1） | ||||
| Shape | Regular | 118 (38.4） | 72 (40） | 0.733 | 167 (39.2） | 23 (37.7） | 0.823 | 122 (41.4） | 68 (35.4） | 0.189 |
| Irregular | 189 (61.6） | 108 (60） | 259 (60.8） | 38 (62.3） | 173 (58.6） | 124 (64.6） | ||||
| Vascularity | Type I | 85 (27.7） | 46 (25.6） | 0.681 | 119 (27.9） | 12 (19.7） | 0.07 | 89 (30.2） | 42 (21.9） | 0.076 |
| Type II | 32 (10.4） | 16 (8.9） | 46 (10.8） | 2 (3.3） | 33 (11.2） | 15 (7.8） | ||||
| Type III | 166 (54.1） | 107 (59.4） | 230 (54.0） | 43 (70.5） | 154 (52.2） | 119 (62.0） | ||||
| Type IV | 24 (7.8） | 11 (6.1） | 31 (7.3） | 4 (6.6） | 19 (6.4） | 16 (8.3) | ||||
| Aspect Ratio | 1.03 (0.33） | 1.06 (0.29） | 0.744 | 1.04 (0.32） | 1.06 (0.28） | 0.910 | 1.05 (0.34) | 1.04 (0.29) | 0.292 | |
| Calcification | No | 128 (41.7） | 71 (39.4） | 0.936 | 178 (41.8） | 21 (34.4） | 0.497 | 89 (30.2) | 42 (21.9) | 0.076 |
| Micro- | 121 (39.4） | 75 (41.7） | 168 (39.4） | 28 (45.9） | 33 (11.2) | 15 (7.8) | ||||
| Macro- | 33 (10.7） | 18 (10.0） | 46 (10.8） | 5 (8.2） | 154 (52.2) | 119 (62.0) | ||||
| Mixed | 25 (8.1） | 16 (8.9） | 34 (8.0） | 7 (11.5） | 19 (6.4) | 16 (8.3) | ||||
| CC | No | 103 (33.6） | 29 (16.1） | < 0.001a | 120 (28.2） | 12 (19.7） | 0.163 | 82 (27.8) | 50 (26.0) | 0.670 |
| Yes | 204 (66.4） | 151 (83.9） | 306 (71.8） | 49 (80.3） | 213 (72.2) | 142 (74.0) | ||||
| DCE | No | 214 (69.7） | 98 (54.4） | 0.001a | 277 (65.0） | 35 (57.4） | 0.244 | 195 (66.1) | 117 (60.9) | 0.246 |
| Yes | 93 (30.3） | 82 (45.6） | 149 (35.0） | 26 (42.6） | 100 (33.9) | 75 (39.1) | ||||
| SCLNM | No | 267 (87.0） | 157 (87.2） | 0.936 | 376 (88.3） | 48 (78.7） | 0.037a | 269 (91.2) | 155 (80.7) | 0.001a |
| Yes | 40 (13.0） | 23 (12.8） | 50 (11.7） | 13 (21.3） | 26 (8.8) | 37 (19.3) | ||||
| SWE_tumor (KPa) | 33.64 (25.16) | 34.6 (25.35) | 0.673 | 33.64 (25.62） | 35.09 (23.27） | 0.785 | 33.64 (25.44） | 34.12 (25.62） | 0.712 | |
| SWE_tumor_SD (Kpa) | 6.34 (5.86) | 6.93 (7.19) | 0.057 | 6.340 (6.44） | 7.12 (7.72） | 0.099 | 6.34 (5.60） | 7.185 (6.99） | 0.015a | |
| SWE_STP (Kpa) | 26.08 (13.78) | 26.08 (16.78) | 0.652 | 26.08 (16.06） | 23.82 (15.22） | 0.158 | 26.08 (16.72） | 26.08 (14.51） | 0.581 | |
| SWE_STP_ SD (Kpa) | 4.68 (3.56) | 5.05 (3.10) | 0.153 | 4.92 (3.49） | 4.36 (3.52） | 0.573 | 4.82 (3.38） | 4.74 (3.49） | 0.945 | |
| SWE_ratio | 1.31 (0.62) | 1.295 (0.79) | 0.759 | 1.29 (0.65） | 1.54 (0.94） | 0.005a | 1.29 (0.63） | 1.325 (0.78） | 0.469 | |
| SWV_tumor (m/s) | 3.33 (1.05) | 3.34 (1.10) | 0.71 | 3.33 (1.09） | 3.37 (0.99） | 0.756 | 3.33 (1.08） | 3.365 (1.07） | 0.568 | |
| SWV_tumor_SD (m/s) | 0.31 (0.18) | 0.32 (0.28) | 0.047a | 0.31 (0.21） | 0.35 (0.25） | 0.067 | 0.31 (0.19） | 0.33 (0.25） | 0.010a | |
| SWV_STP (m/s) | 2.94 (0.68) | 2.94 (0.88) | 0.643 | 2.94 (0.77） | 2.77 (0.75） | 0.131 | 2.94 (0.84） | 2.94 (0.75） | 0.438 | |
| SWV_STP_SD (ms/) | 0.26 (0.14) | 0.26 (0.14) | 0.077 | 0.26 (0.14） | 0.24 (0.19） | 0.754 | 0.26 (0.14） | 0.26 (0.15） | 0.328 | |
| SWV_ratio | 1.13 (0.24) | 1.13 (0.29) | 0.870 | 1.13 (0.25） | 1.24 (0.30） | 0.007a | 1.13 (0.25） | 1.14 (0.28） | 0.343 | |
CUS = conventional ultrasound; USE = ultrasound elastography;PTC = papillary thyroid carcinoma;N = number of patients; ETE = extrathyroidal extension; CLNM = cervical lymph node metastasis; CC = capsule contact; DCE = discontinuous capsule echo; SCLNM = suspected cervical lymph nodes metastases; SWE_tumor = shear wave elasticity value of the tumor; SWE_tumor_SD = one standard deviation of shear wave elasticity value of the tumor; SWE_STP = shear wave elasticity value of the surrounding thyroid parenchyma; SWE_STP_ SD = one standard deviation of shear wave elasticity value of the surrounding thyroid parenchyma; SWE_ratio = the ratio of tumor shear wave elasticity value to that of parenchyma; SWV_tumor = shear wave speed of the tumor; SWV_tumor_SD = one standard deviation of shear wave speed of the tumor; SWV_STP = shear wave speed of the surrounding thyroid parenchyma; SWV_STP_SD = one standard deviation of shear wave speed of the surrounding thyroid parenchyma; SWV_ratio = the ratio of tumor shear wave speed to that of the parenchyma.
Continuous variables are expressed as median values (interquartile range); categorical variables are expressed as numbers (percentages).
“a” indicates statistical differences.
Multivariate analysis results
Multivariate analysis results are summarized in Table 3. Location (OR: 0.558; 95% CI: 0.367–0.830) and CC (OR: 0.470; 95% CI: 0.228–0.793) were independent factors for predicting capsular invasion (p = 0.004 and 0.005, respectively). Tumor size (OR: 1.055; 95% CI: 1.017–1.095) was an independent factor for predicting ETE (p = 0.005). Male sex (OR: 0.473, 95% CI: 0.305–0.733), younger age (OR: 0.969; 95% CI: 0.953–0.986), unclear margin (OR: 1.630; 95% CI: 1.096–2.442), and SCLNM (OR: 2.142; 95% CI: 1.198–3.828) were predictive factors for CLNM (p = 0.001, < 0.001, 0.016, and 0.010, respectively). None of the USE indicators was an independent predictor of aggressive behavior (all p > 0.05).
Table 3.
Multivariate logistic regression analysis of conventional ultrasound and ultrasound elastography indices for predicting the aggressive features of PTC
| Characteristic | Capsular invasion | ETE | CLNM | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | p- value | OR | 95% CI | p- value | OR | 95% CI | p- value | ||
| Sex | Male | Reference | ||||||||
| Female | 0.473 | 0.305,0.733 | 0.001a | |||||||
| Age | 0.969 | 0.953,0.986 | < 0.001a | |||||||
| Size | 1.033 | 0.993,1.074 | 0.107 | 1.055 | 1.017,1.095 | 0.005a | 1.020 | 0.982,1.058 | 0.309 | |
| Location | 0.016a | |||||||||
| Right | Reference | |||||||||
| Left | 0.558 | 0.367,0.830 | 0.004a | |||||||
| Isthmus | 0.704 | 0.262,1.890 | 0.486 | |||||||
| Echogenicity | 0.052 | |||||||||
| Hypo- | Reference | |||||||||
| Hyper- | 1.178 | 0.276,5.026 | 0.825 | |||||||
| Iso- | 0.367 | 0.161,0.840 | 0.018 | |||||||
| Margin | Clear | Reference | ||||||||
| Unclear | 1.630 | 1.096,2.422 | 0.016a | |||||||
| CC | No | 0.470 | 0.228,0.793 | 0.005a | ||||||
| Yes | Reference | |||||||||
| DCE | No | Reference | ||||||||
| Yes | 1.404 | 0.900,2.189 | 0.134 | |||||||
| SCLNM | No | Reference | Reference | |||||||
| Yes | 1.588 | 0.777,3.245 | 0.205 | 2.142 | 1.198,3.828 | 0.010a | ||||
| SWE_tumor_SD | 0.992 | 0.929,1.060 | 0.821 | |||||||
| SWE_ratio | 0.778 | 0.369,1.641 | 0.510 | |||||||
| SWV_tumor_SD | 1.221 | 0.502,2.974 | 0.695 | 3.007 | 0.410,22.053 | 0.279 | ||||
| SWV_ratio | 5.036 | 0.445,56.992 | 0.192 | |||||||
PTC = papillary thyroid carcinoma;ETE = extrathyroidal extension; CLNM = cervical lymph node metastasis; OR = odds ratio; CI = confident interval; CC = capsule contact; DCE = discontinuous capsule echo, SCLNM = suspected cervical lymph nodes metastases; SWE_tumor_SD = one standard deviation of shear wave elasticity value of the tumor; SWE_ratio = the ratio of tumor shear wave elasticity value to that of the parenchyma; SWV_tumor_SD = one standard deviation of shear wave speed of the tumor; SWV_ratio = the ratio of tumor shear wave speed to that of the parenchyma.
“a” indicates statistical differences.
Interrater reliability and measurement consistency of CUS and USE indicators
SWE_tumor, SWE_STP, SWV_tumor, and SWV_STP showed excellent repeatability, with an average ICC of 0.80 (range: 0.78–0.81). The ICCs for reproducibility were good for SWE_tumor_SD and SWV_ratio (ICC = 0.61 and 0.62, respectively). For SWE_STP_SD, SWE_ratio, SWV_tumor_SD, and SWV_STP_SD, fair reproducibility was determined, with an average ICC of 0.51 (range: 0.42–0.59). CUS indices such as echogenicity (κ: 0.93, 95% CI: 0.88–0.98), margin (κ: 0.89, 95% CI: 0.84–0.92), shape (κ: 0.84, 95% CI: 0.80–0.89), calcification (κ: 0.82, 95% CI: 0.78–0.86), CC (κ: 0.85, 95% CI: 0.80–0.90), and DCE (κ: 0.84, 95% CI: 0.79–0.89) showed almost perfect agreement, while the intrareader agreement was substantial for vascularity (κ: 0.79, 95% CI: 0.74–0.83).
Discussion
Aggressive PTC features affect disease recurrence and clinical outcome, and determination of these features preoperatively may facilitate the development of a more individualized and refined treatment strategy. Clinically, USE has been used to distinguish benign from malignant thyroid nodules; however, few studies have focused on assessing the aggressive PTC features by using USE. In this study, we evaluated the value of USE for predicting aggressive PTC features, and the results showed that the SWE indices were not independent risk factors. Therefore, USE may be of limited value in determining the aggressive features of solitary PTCs.
Capsular invasion is associated with PTC aggressiveness. Although capsular invasion does not affect the clinical outcome, 34 it may represent a primary stage for developing an extensive aggressive feature such as ETE. This study showed that CC was an independent risk factor for capsular invasion. The probability of capsular invasion was significantly higher in PTC with CC than in intrathyroidal cases. Location was another independent risk factor for capsular invasion, and PTC located in the right lobe was more likely to result in capsular invasion than that located in the left. The relationship between location and capsular invasion is unclear. In this study, the elasticity index SWV_tumor_SD was associated with capsular invasion in the univariate analysis; however, it was not an independent predictor in the multivariate analysis.
ETE is associated with tumor recurrence and patient prognosis. 3,35 In this study, tumor size was an independent risk factor for ETE. This was similar to the finding of a previous study. 36 As tumor size increases, it makes sense that the probability of soft-tissue infiltration will increase as well. Using univariate analysis, we found that USE indices such as SWE_ratio and SWV_ratio were significantly higher in the ETE group than in the non-ETE group, which suggested that tumor hardness may be associated with ETE. The USE indices Emean and Emin were independent predictors of ETE in micro-PTC. 29 However, this study showed that SWE_tumor, which represents the mean value of the tumor, was not significantly different between the ETE and non-ETE groups. It is possible that the inconsistent results are caused by the different ultrasound instruments used. Different instruments may produce systematic bias and yield different USE indices. For example, the instrument in this study did not provide the Emax and Emin indices. Another reason might be related to patient selection bias: Patients with micro-PTC were enrolled in the previous study, while patients were not grouped according to tumor size in this study. Although the univariate analysis showed a possible association between hardness and ETE, the multivariate analysis demonstrated none of USE indicators was an independent risk factor, and the result was similar to that reported previously. 30 More research is needed to confirm the value of USE in predicting ETE.
In this study, SCLNM was the strongest independent risk factor for predicting CLNM; this finding is consistent with that of a previous study. 37 Colakoglu et al 38 found that CUS had high sensitivity, specificity, and accuracy for evaluating lateral CLNM; however, the sensitivity and accuracy decreased significantly for the evaluation of central CLNM. Since CUS has an important role in CLNM evaluation, a careful examination of cervical lymph nodes should be performed, and patients showing SCLNM are indicated to undergo further FNA examination. In this study, sex was an independent risk factor for predicting CLNM. Male patients had a higher incidence of CLNM, which was similar to the findings reported previously. 39,40 There might be a sex predilection for CLNM. However, it was unclear why male patients are more likely to develop CLNM. Age, in this study, was an independent risk factor for predicting CLNM. Younger patients had a higher risk of CLNM; this finding was similar to that reported previously. 39,41 Although different age cut-off values were provided in previous reports, it seems that with regard to CLNM, more attention should be paid to younger PTC patients. The margin was another independent predictor of CLNM. PTC with an unclear margin was more frequently detected the CLNM. This finding was consistent with those of previous studies. 40,42 In this study, the CLNM group showed significantly higher SWE_tumor_SDs, and SWV_tumor_SDs than did the non-CLNM group in the univariate analysis. This result was inconsistent with that of Chang et al, which showed that the SWV ratio was an independent predictor of CLNM. 43 This might relate to the different patients enrolled; in addition, the outcome variable in this study was CLNM, whereas in the previous study, the variable was central CLNM. It has been found that other USE indicators including virtual touch tissue imaging area ratio 44 and Emax 28,45 are helpful in predicting CLNM. The predictive value of USE for CLNM in PTC requires further study.
This study had some limitations. First, this was a retrospective study; hence, a selection bias could exist. Some patients with a suspicious PTC on FNA examination did not undergo surgery but selected thermal ablation or follow-up observation. The predictive value of USE for aggressive features in these patients needs to be evaluated further. Second, operator bias was present in this study because USE results are strongly influenced by the operator’s technique. To minimize this impact, the same radiologist performs the USE examinations. Furthermore, bias existed in the USE index measurements owing to ROI location. Accordingly, the following methods are used to minimize this bias: (a) the tumor ROIs were placed such that calcified and cystic areas were avoided; (b) the ROIs of the thyroid tumor and the surrounding thyroid parenchyma were placed at similar depths; (c) USE indices of each tumor were measured at different locations; and (d) measurement consistency of the USE indicators was evaluated. Finally, this was a single-center study, and the sample size was small; hence, the results need to be confirmed by multicenter, prospective studies with a larger sample size.
Conclusion
The USE indices were not independent risk factors for predicting aggressive features of PTC, and ultrasound-based elastography may have limited value for predicting these aggressive features.
Contributor Information
Chao Jia, Email: wrjiachao@126.com.
Gang Li, Email: happy1314lg@163.com.
Qiusheng Shi, Email: sqs19631989@163.com.
Lianfang Du, Email: du_lf@163.com.
Rong Wu, Email: wurong7111@163.com.
REFERENCES
- 1. Lim H, Devesa SS, Sosa JA, Check D, Kitahara CM. Trends in thyroid cancer incidence and mortality in the United States, 1974-2013. JAMA 2017; 317: 1338–48. doi: 10.1001/jama.2017.2719 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Colonna M, Borson-Chazot F, Delafosse P, Schvartz C, Guizard A-V, FRANCIM network . Progression of incidence and estimate of net survival from papillary thyroid cancers diagnosed between 2008 and 2016 in France. Ann Endocrinol (Paris) 2020; 81: 530–38. doi: 10.1016/j.ando.2020.11.006 [DOI] [PubMed] [Google Scholar]
- 3. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al. 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: 1–133. doi: 10.1089/thy.2015.0020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Filetti S, Durante C, Hartl D, Leboulleux S, Locati LD, Newbold K, et al. Thyroid cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 2019; 30: 1856–83. doi: 10.1093/annonc/mdz400 [DOI] [PubMed] [Google Scholar]
- 5. Cao X-J, Liu J, Zhu Y-L, Qi L, Liu G, Wang H-L, et al. Efficacy and safety of thermal ablation for solitary T1Bn0M0 papillary thyroid carcinoma: A multicenter study. J Clin Endocrinol Metab 2021; 106: e573–81. doi: 10.1210/clinem/dgaa776 [DOI] [PubMed] [Google Scholar]
- 6. Cao X-J, Wang S-R, Che Y, Liu J, Cong Z-B, He J-F, et al. Efficacy and safety of thermal ablation for treatment of solitary T1N0M0 papillary thyroid carcinoma: A multicenter retrospective study. Radiology 2021; 300: 209–16. doi: 10.1148/radiol.2021202735 [DOI] [PubMed] [Google Scholar]
- 7. Mauri G, Hegedüs L, Bandula S, Cazzato RL, Czarniecka A, Dudeck O, et al. European thyroid Association and cardiovascular and Interventional radiological society of Europe 2021 clinical practice guideline for the use of minimally invasive treatments in malignant thyroid lesions. Eur Thyroid J 2021; 10: 185–97. doi: 10.1159/000516469 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Sugitani I, Ito Y, Takeuchi D, Nakayama H, Masaki C, Shindo H, et al. Indications and strategy for active surveillance of adult low-risk papillary thyroid Microcarcinoma: Consensus statements from the Japan Association of endocrine surgery task force on management for papillary thyroid Microcarcinoma. Thyroid 2021; 31: 183–92. doi: 10.1089/thy.2020.0330 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Sakai T, Sugitani I, Ebina A, Fukuoka O, Toda K, Mitani H, et al. Active surveillance for T1Bn0M0 papillary thyroid carcinoma. Thyroid 2019; 29: 59–63. doi: 10.1089/thy.2018.0462 [DOI] [PubMed] [Google Scholar]
- 10. Şahin M, Oguz A, Tuzun D, Akkus G, Törün GI, Bahar AY, et al. Effectiveness of TI-RADS and ATA classifications for predicting malignancy of thyroid nodules. Adv Clin Exp Med 2021; 30: 1133–39. doi: 10.17219/acem/139591 [DOI] [PubMed] [Google Scholar]
- 11. Russ G, Bonnema SJ, Erdogan MF, Durante C, Ngu R, Leenhardt L. European thyroid Association guidelines for ultrasound malignancy risk stratification of thyroid nodules in adults: The EU-TIRADS. Eur Thyroid J 2017; 6: 225–37. doi: 10.1159/000478927 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. García Pascual L, Surralles ML, Morlius X, González Mínguez C, Viscasillas G, Lao X. Ultrasound-guided fine needle aspiration of thyroid nodules with on-site cytological examination: Diagnostic efficacy, prevalence, and factors predicting for Bethesda category I results. Endocrinología, Diabetes y Nutrición (English Ed) 2019; 66: 495–501. doi: 10.1016/j.endien.2018.12.007 [DOI] [PubMed] [Google Scholar]
- 13. Chen Y, Zhan W, Zhou W, Dong Y. The application of ultrasound and fine-needle aspiration in low-volume lateral lymph nodes of papillary thyroid carcinoma patients. Ultrasound Q 2019; 35: 240–45. doi: 10.1097/RUQ.0000000000000397 [DOI] [PubMed] [Google Scholar]
- 14. Rosário PWS, de Faria S, Bicalho L, Alves MFG, Borges MAR, Purisch S, et al. Ultrasonographic differentiation between metastatic and benign lymph nodes in patients with papillary thyroid carcinoma. J Ultrasound Med 2005; 24: 1385–89. doi: 10.7863/jum.2005.24.10.1385 [DOI] [PubMed] [Google Scholar]
- 15. Khokhar MT, Day KM, Sangal RB, Ahmedli NN, Pisharodi LR, Beland MD, et al. Preoperative high-resolution ultrasound for the assessment of malignant central compartment lymph nodes in papillary thyroid cancer. Thyroid 2015; 25: 1351–54. doi: 10.1089/thy.2015.0176 [DOI] [PubMed] [Google Scholar]
- 16. Zhao H, Li H. Meta-analysis of ultrasound for Cervical lymph nodes in papillary thyroid cancer: Diagnosis of central and lateral compartment nodal metastases. Eur J Radiol 2019; 112: 14–21. doi: 10.1016/j.ejrad.2019.01.006 [DOI] [PubMed] [Google Scholar]
- 17. Hwang HS, Orloff LA. Efficacy of preoperative neck ultrasound in the detection of Cervical lymph node metastasis from thyroid cancer. Laryngoscope 2011; 121: 487–91. doi: 10.1002/lary.21227 [DOI] [PubMed] [Google Scholar]
- 18. Lamartina L, Bidault S, Hadoux J, Guerlain J, Girard E, Breuskin I, et al. Can preoperative ultrasound predict Extrathyroidal extension of differentiated thyroid cancer Eur J Endocrinol 2021; 185: 13–22: EJE-21-0091. doi: 10.1530/EJE-21-0091 [DOI] [PubMed] [Google Scholar]
- 19. Idrees A, Shahzad R, Fatima I, Shahid A. Strain Elastography for differentiation between benign and malignant thyroid nodules. J Coll Physicians Surg Pak 2020; 30: 369–72. doi: 10.29271/jcpsp.2020.04.369 [DOI] [PubMed] [Google Scholar]
- 20. Gregory A, Bayat M, Kumar V, Denis M, Kim BH, Webb J, et al. Differentiation of benign and malignant thyroid nodules by using comb-push ultrasound shear Elastography: A preliminary two-plane view study. Acad Radiol 2018; 25: 1388–97. doi: 10.1016/j.acra.2018.02.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Dighe M, Luo S, Cuevas C, Kim Y. Efficacy of thyroid ultrasound Elastography in differential diagnosis of small thyroid nodules. Eur J Radiol 2013; 82: e274–80. doi: 10.1016/j.ejrad.2013.01.009 [DOI] [PubMed] [Google Scholar]
- 22. Friedrich-Rust M, Vorlaender C, Dietrich CF, Kratzer W, Blank W, Schuler A, et al. Evaluation of strain Elastography for differentiation of thyroid nodules: Results of a prospective DEGUM multicenter study. Ultraschall Med 2016; 37: 262–70. doi: 10.1055/s-0042-104647 [DOI] [PubMed] [Google Scholar]
- 23. Xu J-M, Xu X-H, Xu H-X, Zhang Y-F, Zhang J, Guo L-H, et al. Conventional US, US elasticity imaging, and acoustic radiation force impulse imaging for prediction of malignancy in thyroid nodules. Radiology 2014; 272: 577–86. doi: 10.1148/radiol.14132438 [DOI] [PubMed] [Google Scholar]
- 24. Azizi G, Keller JM, Mayo ML, Piper K, Puett D, Earp KM, et al. Thyroid nodules and shear wave Elastography: A new tool in thyroid cancer detection. Ultrasound Med Biol 2015; 41: 2855–65. doi: 10.1016/j.ultrasmedbio.2015.06.021 [DOI] [PubMed] [Google Scholar]
- 25. Dobruch-Sobczak K, Zalewska EB, Gumińska A, Słapa RZ, Mlosek K, Wareluk P, et al. Diagnostic performance of shear wave Elastography parameters alone and in combination with conventional B-mode ultrasound parameters for the characterization of thyroid nodules: A prospective, dual-center study. Ultrasound Med Biol 2016; 42: 2803–11. doi: 10.1016/j.ultrasmedbio.2016.07.010 [DOI] [PubMed] [Google Scholar]
- 26. Dighe M, Kim J, Luo S, Kim Y. Utility of the ultrasound Elastographic systolic thyroid stiffness index in reducing fine-needle aspirations. J Ultrasound Med 2010; 29: 565–74. doi: 10.7863/jum.2010.29.4.565 [DOI] [PubMed] [Google Scholar]
- 27. Guo J-N, Song L-H, Yu P-Y, Yu S-Y, Deng S-H, Mao X-H, et al. Ultrasound elastic parameters predict central lymph node metastasis of papillary thyroid carcinoma. J Surg Res 2020; 253: 69–78. doi: 10.1016/j.jss.2020.03.042 [DOI] [PubMed] [Google Scholar]
- 28. Kim HJ, Choi IH, Jin S-Y, Park HK, Byun DW, Suh K, et al. Efficacy of shear-wave Elastography for detecting postoperative Cervical lymph node metastasis in papillary thyroid carcinoma. Int J Endocrinol 2018. doi: 10.1155/2018/9382649 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Park YJ, Kim JA, Son EJ, Youk JH, Park CS. Quantitative shear wave Elastography as a Prognostic implication of papillary thyroid carcinoma (PTC): Elasticity index can predict Extrathyroidal extension (ETE). Ann Surg Oncol 2013; 20: 2765–71. doi: 10.1245/s10434-013-2927-4 [DOI] [PubMed] [Google Scholar]
- 30. Yang X-Y, Xu L-F, Luo Y, Li N, Yang J-P, Zhu L-P, et al. Effects of age on the peak ratio of ophthalmic artery Doppler. Medicine (Baltimore) 2020; 99(): e23694. doi: 10.1097/MD.0000000000023694 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Hong Y-R, Yan C-X, Mo G-Q, Luo Z-Y, Zhang Y, Wang Y, et al. Conventional US, Elastography, and contrast enhanced US features of papillary thyroid Microcarcinoma predict central compartment lymph node metastases. Sci Rep 2015; 5: 7748. doi: 10.1038/srep07748 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Kuna SK, Bracic I, Tesic V, Kuna K, Herceg GH, Dodig D. Ultrasonographic differentiation of benign from malignant neck Lymphadenopathy in thyroid cancer. J Ultrasound Med 2006; 25: 1531–37. doi: 10.7863/jum.2006.25.12.1531 [DOI] [PubMed] [Google Scholar]
- 33. Na DG, Lim HK, Byun HS, Kim HD, Ko YH, Baek JH. Differential diagnosis of Cervical Lymphadenopathy: Usefulness of color Doppler Sonography. AJR Am J Roentgenol 1997; 168: 1311–16. doi: 10.2214/ajr.168.5.9129432 [DOI] [PubMed] [Google Scholar]
- 34. Furlan JC, Bedard YC, Rosen IB. Significance of tumor Capsular invasion in well-differentiated thyroid Carcinomas. Am Surg 2007; 73: 484–91. [PubMed] [Google Scholar]
- 35. Bortz MD, Kuchta K, Winchester DJ, Prinz RA, Moo-Young TA. Extrathyroidal extension predicts negative clinical outcomes in papillary thyroid cancer. Surgery 2021; 169: 2–6. doi: 10.1016/j.surg.2020.04.003 [DOI] [PubMed] [Google Scholar]
- 36. Kuo CY, Yang PS, Chien MN, Cheng SP. Preoperative factors associated with Extrathyroidal extension in papillary thyroid cancer. Eur Thyroid J 2020; 9: 256–62. doi: 10.1159/000506474 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Sun J, Jiang Q, Wang X, Liu W, Wang X. Nomogram for preoperative estimation of Cervical lymph node metastasis risk in papillary thyroid Microcarcinoma. Front Endocrinol 2021; 12: 613974. doi: 10.3389/fendo.2021.613974 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Colakoglu B, Alis D, Seymen H. Diagnostic accuracy of ultrasound for the evaluation of lateral compartment lymph nodes in papillary thyroid carcinoma. Curr Med Imaging 2020; 16: 459–65. doi: 10.2174/1573405615666190619093618 [DOI] [PubMed] [Google Scholar]
- 39. Liu C, Xiao C, Chen J, Li X, Feng Z, Gao Q, et al. Risk factor analysis for predicting Cervical lymph node metastasis in papillary thyroid carcinoma: A study of 966 patients. BMC Cancer 2019; 19(): 622. doi: 10.1186/s12885-019-5835-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Feng J-W, Hong L-Z, Wang F, Wu W-X, Hu J, Liu S-Y, et al. A Nomogram based on clinical and ultrasound characteristics to predict central lymph node metastasis of papillary thyroid carcinoma. Front Endocrinol (Lausanne) 2021; 12: 666315. doi: 10.3389/fendo.2021.666315 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Tao L, Zhou W, Zhan W, Li W, Wang Y, Fan J. Preoperative prediction of Cervical lymph node metastasis in papillary thyroid carcinoma via conventional and contrast-enhanced ultrasound. J Ultrasound Med 2020; 39: 2071–80. doi: 10.1002/jum.15315 [DOI] [PubMed] [Google Scholar]
- 42. Zhang Y, Luo Y, Zhang M, Li J, Li C, Tang J, et al. Values of ultrasound features and MMP-9 of papillary thyroid carcinoma in predicting Cervical lymph node metastases. Sci Rep 2017; 7: 6670. doi: 10.1038/s41598-017-07118-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Chang W, Tang L, Lu C, Wu M, Chen M. Shear wave Elastography in the evaluation of level VI lymph nodes in papillary thyroid carcinoma: Combined with gray-scale ultrasound ex vivo. BMC Cancer 2018; 18(): 1001. doi: 10.1186/s12885-018-4897-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Xu J-M, Xu X-H, Xu H-X, Zhang Y-F, Guo L-H, Liu L-N, et al. Prediction of Cervical lymph node metastasis in patients with papillary thyroid cancer using combined conventional ultrasound, strain Elastography, and acoustic radiation force impulse (ARFI) Elastography. Eur Radiol 2016; 26: 2611–22. doi: 10.1007/s00330-015-4088-2 [DOI] [PubMed] [Google Scholar]
- 45. Zhang C, Li BJ, Liu Z, Wang LL, Cheng W. Predicting the factors associated with central lymph node metastasis in clinical node-negative (Cn0) papillary thyroid Microcarcinoma. Eur Arch Otorhinolaryngol 2020; 277: 1191–98. doi: 10.1007/s00405-020-05787-1 [DOI] [PubMed] [Google Scholar]