This study was performed in accordance with the Declaration of Helsinki. This human study was approved by the GangNeung Asan Hospital Institutional Review Board (2021-08-010). Due to the retrospective nature of this study, the need for informed consent was waived.

Between March 2017 and August 2021, 118 consecutive patients with thyroid nodules underwent biopsy for 141 suspicious cervical lesions showing suspicious US features of metastatic LNs on preoperative FNA or core needle biopsy (CNB) for thyroid nodules. Twenty-one consecutive patients underwent biopsy for 32 suspicious metastatic lesions during postoperative surveillance of thyroid cancer. Therefore, during the study period, 139 consecutive patients underwent FNA for 173 suspicious metastatic lesions presenting suspicious US features of cervical metastatic LNs in thyroid cancer. A lesion was classified as suspicious for metastasis if any suspicious US feature of metastatic LNs (hyperechogenicity, calcification, cystic change, and abnormal vascularity) was identified [17]. Among 173 suspicious metastatic lesions, 21 parathyroid lesions of 21 patients (20 preoperative and one postoperative [hemithyroidectomy]) were identified by the high level of measured FNA-washout PTH (>100 pg/mL) [18-20]. Serum PTH and calcium levels were evaluated in most patients with high levels of tissue PTH.

A normal PTG was defined as a parathyroid lesion without a typical US feature of parathyroid cyst or neoplasm [8,21] in patients with normal serum PTH and calcium levels. Eleven parathyroid lesions were excluded from the 21 identified parathyroid lesions due to normocalcemic hyperparathyroidism (n=4), incidental parathyroid adenomas (n=4), incidental parathyroid cysts (n=2), and parathyroid lesions without preoperative serum PTH measurements (n=1). Thus, the final number of participants with normal PTGs was 10 (group 1; seven women and three men; mean age, 49.1 years) (Fig. 1).

During the same study period, 398 consecutive patients underwent preoperative thyroid US examinations and thyroid surgery for malignant thyroid tumors. The group of patients with metastatic LNs (group 2) only included patients with metastatic LNs that exhibited US features suspicious for metastatic LNs on preoperative thyroid US. Group 2 was used to compare the US features between normal PTGs and metastatic LNs. After a review of the electronic medical records and US reports of all patients, 351 patients without metastatic LNs showing suspicious US features on preoperative US were excluded. The exclusion criteria were as follows: (1) no metastatic LNs (n=245) in the postoperative pathology report, (2) microscopic metastatic LNs (n=66), and (3) macroscopic metastatic LNs without suspicious US features on preoperative US (n=40). Based on a retrospective level-by-level radiologic-pathologic correlation of metastatic LNs, 47 patients (37 women and 10 men; mean age, 58.8 years) with 95 metastatic LNs showing suspicious US features were finally included in the group of patients with metastatic LNs (group 2) (Fig. 1). The metastatic LNs in this study were diagnosed by both preoperative biopsy (FNA or CNB) and surgery (n=43) or by surgery alone (n=52). There was no overlap of patients between the study populations of groups 1 and 2. That is, none of the patients with normal PTGs (group 1) simultaneously had metastatic LNs and only one patient with papillary thyroid cancer in group 1 underwent thyroid surgery at the authors’ affiliated institution.

US-guided FNA was performed in one to two suspicious metastatic LNs in most patients, while CNB was selectively performed for suspicious metastatic LNs (large LNs, suspected high-grade thyroid cancer or medullary carcinoma, and suspected metastatic LNs from non-thyroid organs). Freehand technique US-guided FNA was performed using a 21- or 23-gauge needle and a 5-mL disposable syringe. The combined method of a capillary and aspiration technique was employed. SurePath (BD, Franklin Lakes, NJ, USA) liquid-based cytology was used to establish a cytopathologic diagnosis via FNA. US-guided CNB procedures were performed using a disposable 1.1-cm excursion, 18-gauge, single- or double-action spring-activated needle (Stericut [TSK, Tokyo, Japan] or Ace-cut [Create Medic, Yokohama, Japan]). The CNB needle notch was positioned within the target lesions. Strict vessel mapping along the approach route was performed using color Doppler US during the procedure to avoid vessel injury. FNA-washout thyroglobulin was routinely measured for patients with lateral neck lesions and selectively for those with central neck lesions. FNA-washout PTH measurements were selectively obtained for central neck lesions to rule out parathyroid lesions. The parathyroid lesions were suspected when the lesions showed typical US features of parathyroid cyst or parathyroid adenoma [8,19] and when well-defined, round or oval, and isoechoic to hyperechoic small lesions were observed at the anatomical location of the PTG [10,11]. The FNA needle sample was rinsed using a 23-gauge needle with 1 mL saline and was assayed for thyroglobulin and PTH levels.

All US examinations and biopsy procedures for the thyroid nodules and cervical LNs were performed using a 5-12 MHz linear-array transducer (EPIQ7, Philips Healthcare, Bothell, WA, USA) by two radiologists with 22 and 4 years of experience, respectively, in performing thyroid US imaging and interventions. The US images of all suspicious US features found in suspicious metastatic lesions were routinely obtained during the US examinations. The US features of normal PTGs and metastatic LNs were retrospectively assessed by an experienced radiologist (D.G.N.) blinded to the patients’ clinical histories and final diagnoses. The reviewer assessed the echogenicity, echotexture, echogenic foci (calcifications), cystic change, abnormal vascularity, size, shape, and location of the lesions. The echogenicity was categorized as marked hyperechogenicity (hyperechoic relative to the normal thyroid gland), moderate hyperechogenicity (isoechoic relative to that of the normal thyroid gland), mild hyperechogenicity (hyperechoic relative to the neck muscles and hypoechoic relative to the normal thyroid gland), and hypoechogenicity (isoechoic or hypoechoic relative to the neck muscles). In patients with diffuse thyroid disease and decreased parenchymal echogenicity, the homogenous hyperechoic echogenicity of the presumed normal thyroid gland was used as the reference structure. The echotexture of PTGs and metastatic LNs was categorized as homogeneous or heterogeneous. The homogeneous echotexture was defined as uniform echogenicity with one category of echogenicity. In contrast, heterogeneous echotexture was defined as mixed echogenicity with more than one category of echogenicity. If the lesion had a mixed pattern, echogenicity was determined using the predominant echogenicity. Echogenic foci (calcifications) were classified as punctate echogenic foci (microcalcification) (punctate [≤1 mm] hyperechoic foci with or without posterior acoustic artifacts) or large echogenic foci (large [>1 mm] hyperechoic foci with or without posterior acoustic artifact), including macrocalcification and rim calcification [16]. The presence of cystic changes was determined if any focal anechoic area was found within the lesion, or if the lesion was entirely anechoic. Abnormal vascularity was characterized by a peripheral or chaotic diffuse vascularity on color Doppler US images. The shape of the lesion was assessed using the long axis/ short axis (L/S) ratio and was categorized as round (L/S ratio ≤1.5) or ovoid (L/S ratio >1.5). The classifications of the location of lesions included the central or lateral neck. The central neck was classified as midline (prelaryngeal and pretracheal) and paramedian (superior thyroid pole, superior perithyroidal, inferior perithyroidal, inferior thyroid pole, and infrathyroidal) central neck. Whether the lesion was located at the superior or inferior thyroid pole was determined when the epicenter of the lesion was outside the thyroid gland but partially attached to the thyroid pole. The lesions located at the superior and inferior perithyroidal paramedian central neck were subcategorized as anterior, posterior, medial, or lateral according to the site of the lesion around the thyroid gland. The lateral neck was classified by the LN levels according to the eighth edition of the American Joint Committee on Cancer TNM classification system [22].

Categorical variables are reported as frequencies and percentages for each category. Continuous variables are presented as the median (interquartile range) or the mean±standard deviation, according to whether they had a parametric or nonparametric distribution, respectively. The Pearson chi-square test was used to compare the US findings between normal PTGs and metastatic LNs. The Mann–Whitney U Test was used to compare the long and short diameters for size and L/S ratios between normal PTGs and metastatic LNs. The statistical analysis was performed using SPSS version 24.0 for Windows (IBM Corp., Armonk, NY, USA). A significant difference was defined as one with a P-value <0.05.

The demographic data are summarized in Table 1. There was no significant difference in sex (P=0.551) between normal PTGs and metastatic LNs, but the mean age of participants with normal PTGs was significantly younger than those with metastatic LN (P=0.049). The long and short diameters of metastatic LNs were significantly larger than those of normal PTGs (P=0.009 and P<0.001, respectively). The L/S ratios of normal PTGs were significantly larger than those of metastatic LNs (P=0.022). In the 47 patients with metastatic LNs, the postoperative histologic types of thyroid cancers were papillary carcinoma (n=44), including follicular variant papillary carcinoma (n=4), follicular carcinoma (n=1), medullary thyroid carcinoma (n=1), and anaplastic thyroid carcinoma (n=1).

The clinical characteristics of the 10 identified normal PTGs are summarized in Table 2. The measured FNA-washout PTH levels were very high (>1,350-1,800 pg/mL), whereas the serum PTH levels were within the normal range (23.9-59.3 pg/mL). The measured FNA-washout thyroglobulin levels were ≤0.26 ng/mL in five patients in whom FNA washout thyroglobulin was measured without contamination of thyroid tissue. The total serum calcium levels were within the normal range (7.9-9.9 mg/dL).

The US features of normal PTGs and metastatic LNs are summarized in Table 3. The echogenicity of normal PTGs was significantly different from that of metastatic LNs (P<0.001). All normal PTGs consistently showed marked hyperechogenicity, whereas marked hyperechogenicity was not found in any metastatic LNs (Figs. 2, 3). The echogenicity of 95 metastatic LNs was classified as moderately hyperechoic in 15 (15.8%), mildly hyperechoic in 51 (53.7%), hypoechoic in 27 (28.4%) LNs, and not assessable in two (2.1%) entirely calcified metastatic LNs. Mild or moderate hyperechogenicity alone without concurrent other suspicious US feature was found in 16 (16.8%) metastatic LNs with suspicious US features (Fig. 4).

Homogeneous echotexture was found in eight of 10 normal PTGs (80%), while heterogeneous echotexture in two of 10 normal PTGs (20%). The two normal PTGs with heterogeneous echotexture showed a focal, linear, relatively hypoechoic lesion in the center of the PTG and a focal, round hypoechoic lesion with suspicion of anechoic cystic change (Fig. 5). The focal round hypoechoic lesion within the normal PTG was confirmed as a small parathyroid cyst with clear fluid by FNA. Metastatic LNs showed a lower frequency (42.1%, 40/95) of homogeneous echotexture than normal PTGs, without statistical significance (P=0.072).

Focal cystic change was found in one case (10%), and echogenic foci (calcification) or abnormal vascularity on color Doppler images was not found in any of the normal PTGs. Meanwhile, echogenic foci (calcifications) were found in 60 (63.2%) and abnormal vascularity in 29 (30.5%) of 95 metastatic LNs (P<0.001 and P=0.011, respectively) (Fig. 6). Cystic change and abnormal vascularity were more frequently found in lateral than in central metastatic LNs (P=0.042 and P=0.002, respectively). However, there were no significant differences in echogenicity, echotexture, and echogenic foci between central and lateral metastatic LNs. Eight (20%) of 10 normal PTGs demonstrated an ovoid shape. There was no significant difference in the frequency of ovoid shape between normal PTGs and metastatic LNs (P=0.821).

The distribution of normal PTGs and metastatic LNs in the central neck are summarized in Table 4. All identified normal PTGs were located in the left paramedian central neck, particularly the infrathyroid level VI area (n=4, 40.0%), inferior perithyroidal region (n=4, 40.0%; posterior 3 and lateral 1) and inferior thyroid pole (n=2, 20.0%) (Figs. 2, 3). Metastatic LNs were located in the midline (n=11, 35.5%) and paramedian region (n=20, 64.5%). Metastatic LNs found in the paramedian central neck were most frequently located in the infrathyroid level VI (n=11, 35.5%), followed by the inferior perithyroidal region (n=7, 22.6%; all posterior) and inferior pole (n=2, 6.5%).