Sonoelastographic evaluation for benign neck lymph nodes and parathyroid lesions
Article Category: original-paper
Online veröffentlicht: 30. Jan. 2019
Seitenbereich: 284 - 289
Eingereicht: 16. Aug. 2018
Akzeptiert: 31. Dez. 2018
DOI: https://doi.org/10.15557/jou.2018.0041
Schlüsselwörter
© 2018 Namik Kemal Altinbas et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Parathyroid adenomas are benign tumors and are the most common causes of primary hyperparathyroidism(1). Solitary benign adenoma (89%), double adenomas (4%), parathyroid hyperplasia (6%), and parathyroid carcinoma (<1%) are probable causes of primary hyperparathyroidism(2,3).
Hashimoto thyroiditis, a type of autoimmune thyroiditis, is one of the most common thyroid disorders. In these patients, prominent reactive cervical lymph nodes may be present especially in level VI and the “Delphian” node, just cephalic to the isthmus. These perithyroidal lymph nodes are also helpful in diagnosis(4,5).
High-resolution sonography is a sensitive imaging test for the detection of superficial masses, and currently one of the first-line imaging techniques for the evaluation of parathyroid gland lesions(1). Doppler ultrasonography with the detection of the extrathyroidal feeding vessel has improved the determination of parathyroid adenomas(6). Ultrasonography and technetium-99m-sestamibi scintigraphy with or without sestamibi single photon emission computed tomography (SPECT) are the dominant imaging modalities for preoperative localization of parathyroid adenomas(7,8).
As a relatively new method, sonoelastography is currently under investigation for tissue characterization of several anatomic sites. Elastography of a mass shows its elastic properties. It allows examination of changes in tissue elasticity features(9). Strain elastography, as performed in this study, is a non-invasive method involving manual compression that provides an evaluation of tissue stiffness differences.
The aim of this study was to determine the performance and additional values of real-time strain sonoelastography in the differential diagnosis of perithyroidal reactive lymph nodes of patients with Hashimoto thyroiditis, reactive jugular lymph nodes of healthy individuals and parathyroid gland lesions.
This prospective study was performed in the Department of Radiology between April 2016 and March 2018. All patients gave informed written consent for sonographic evaluation and for this work. All procedures were performed in accordance with the Declaration of Helsinki for human subjects, and our Institutional Review Board approved the study.
In total, 95 patients who had been referred to the Department of Radiology for sonographic evaluation, were included in the study. A total of 153 lesions were examined by ultrasonography, and strain sonoelastography was added.
The diagnosis in patients with parathyroid adenomas or hyperplasia was confirmed by parathyroidectomy or fineneedle aspiration biopsy with parathormone (PTH) washout or with a positive scintigraphy scan and high serum PTH level.
The diagnosis of Hashimoto thyroiditis was confirmed by demonstration of serum thyroid antibodies and antithyroglobulin antibodies. Patients with Hashimoto thyroiditis have lymph nodes around the thyroid gland like parathyroid glands. Thus, these lymph nodes located at level VI were used as one of the control groups. Patients with Hashimoto thyroiditis without district nodular lesions in the thyroid gland were included in the study to exclude probable metastatic lymph nodes. Additionally, we believe that there are insufficient numbers of level VI lymph nodes in healthy individuals to serve as a control group. Therefore, level III and IV lymph nodes of healthy individuals were used as the other control group.
Ultrasonography and strain elastography examinations were performed by a radiologist experienced in sonographic evaluations. Lesion dimensions on gray-scale ultrasound were evaluated and noted. Sonographic examinations were performed with a Logiq S7 Expert machine (GE Healthcare, Milwaukee, WI) equipped with a 9L-D linear-array probe. Slight repetitive manual compressions were performed, keeping the transducer perpendicular to the skin during strain elastography. The elastographic images were obtained with appropriate compression and decompression application according to a quality bar. The bar scale ranged from 1 to 7 on the screen. Images and measurements were obtained only when the optimal compression bar was in the range of 5 to 7. The elastogram was displayed as a real-time color map of relative elasticity, which was superimposed on the gray-scale image. Elastography and B-mode ultrasonography images were simultaneously presented as a two-panel image. The region of interest (ROI) was placed first on the reference tissue using a circle drawing tool and second on the mass using a free-hand drawing tool along to the lesion border. Elasticity score (E-index) was obtained by elastographic analysis. The described strain elastography method represents relative information about hardness(10). The relation between elastogram colors and the hardness of an area is given on the elastography color bar (at the left side of each elastogram). While red color reflects less stiff tissue, blue color reflects more stiff tissue. In addition, green color reflects medium stiff tissue of the entire area undergoing elastography. When the ROI drawing is completed, the sonography machine automatically gives a value, which reflects E-index, from 0 (softest) to 6 (hardest). E-index is a semi-quantitative value which represent the strain of the ROI relative to the entire area undergoing elastography. While a low E-index indicates soft tissue, a higher E-index indicates stiff tissue. E-index from real-time strain elastography was noted. The difference in E-index and diameter between parathyroid lesions and lymph nodes was evaluated. In calculating the elasticity ratio (E-ratio), which represents how many times stiffer a lesion is, we should use a reference tissue. However, some lesions were not appropriate for this calculation. Some lesions were not in the same depth or were not at the same level with standardized reference tissue. Therefore, E-ratio was not used to compare the groups(10).
Statistical analyses were performed using SPSS version 15.0 software for Windows (IBM Corporation, Armonk, NY). Descriptive statistics of continuous variables are given as mean ± standard deviation (SD) and median (interquartile range 25–75). Kolmogorov-Smirnov test was used to test normality. To compare groups, one-way Anova test was used.
In the study population (95 patients, 153 lesions), there were three groups. Group 1 (parathyroid lesions) had 37 patients with a total of 50 lesions. Group 2 (perithyroidal lymph nodes in Hashimoto thyroiditis patients) had 27 patients with a total of 52 lymph nodes. Group 3 (reactive jugular lymph nodes in healthy individuals) had 31 patients with a total of 51 lesions. Parathyroid hyperplasia was detected by parathyroidectomy in two patients (2/37; 5%) with a total of four lesions (4/50; 8%). The other parathyroid lesions were considered as adenomas.
The mean age of the patients was 48.7 ± 14.88 years (range 19–79), and 22% of the patients were male (
The mean sizes of lesions in Group 1, Group 2 and Group 3 were 13.46 ± 5.69 mm, 7.83 ± 3.35 mm and 11.60 ± 4.96 mm, respectively. There were statistically significant differences between Group 2 and Group 1, and between Group 2 and Group 3 in terms of the diameters of the lesions (both
The mean E-index of parathyroid lesions, lymph nodes of Hashimoto patients and jugular lymph nodes of healthy individuals were 2.30 ± 0.91, 2.70 ± 0.93 and 1.88 ± 0.59, respectively. There were statistically significant differences between the groups in terms of E-index via multiple comparisons (all
All results of the study are presented in Table 1. Images of a parathyroid adenoma, a lymph node of a Hashimoto patient and a jugular lymph node of a healthy individual, including gray-scale and elastography examinations, are shown in Fig. 1, Fig. 2 and Fig. 3.
Comparison of age of the patients, size of the lesions, and elasticity score (E-index) among the groups
| Group 1 | Group 2 | Group 3 |
|
|
|---|---|---|---|---|
|
|
54.38 ± 14.13 |
39.78 ± 11.97a
|
49.68 ± 14.77 |
<0.001 |
|
|
13.46 ± 5.69 |
7.83 ± 3.35b
|
11.60 ± 4.96 10; 4.00 (8–12) | <0.001 |
|
|
2.30 ± 0.91c
|
2.70 ± 0.93d
|
1.88 ± 0.59e
|
<0.001 |
Different from both Group 1 and Group 3 (
Different from both Group 1 and Group 3 (both p <0.001)
Different from both Group 2 and Group 3 (
Different from both Group 1 and Group 3 (
Different from both Group 1 and Group 2 (
Fig. 1.
Fig. 2.
Fig. 3.
The perithyroidal central compartment lymph nodes, as particularly prominent in patients with lymphocytic thyroiditis, can commonly be mistaken for parathyroid glands. However, several features have been described to distinguish lymph nodes from adenomas. While a benign reactive lymph node usually has an echogenic fatty hilum supplied by small hilar vessels, a parathyroid adenoma usually has a polar, peripheral vascular structure supplied by an extrathyroidal feeding vessel(1,6,11,12). High-resolution ultrasonography findings, together with laboratory findings and other imaging methods, have an impressive value for the diagnosis of parathyroid lesions. In the current study, strain sonoelastography was found to provide additional contribution to the differential diagnosis of parathyroid lesions and lymph nodes.
In our study, parathyroid hyperplasia was detected by parathyroidectomy in two patients (2/37; 5%) with a total of four lesions (4/50; 8%). One patient had a single lesion and the other patient had three lesions. The distribution of the patients with parathyroid adenoma was as follows: a solitary lesion in 27 patients (27/35; 77%), 2 lesions in six patients (6/35), 3 lesions in one patient (1/35), and 4 lesions in one patient (1/35). Among the Hashimoto patients with lymph nodes, 15 patients (15/27) had one lymph node, 4 patients (4/27) had 2 lymph nodes, 4 patients (4/27) had 3 lymph nodes, 3 patients (3/27) had 4 lymph nodes, and one patient (1/27) had 5 lymph nodes. Among the healthy individuals with reactive lymph nodes, 19 patients (19/31) had one lymph node, 8 patients (8/31) had 2 lymph nodes, 2 patients (2/31) had 3 lymph nodes, 1 patients (1/31) had 4 lymph nodes, and one patient (1/31) had 6 lymph nodes. Frequencies were in accordance with the frequency commonly reported in the literature(2,3).
In an article by Azizi
In an article by Isidori
The patients in Group 2 were selected from among patients without nodules in the thyroid parenchyma. The limitation of the study was that no tissue sampling of the lymph nodes was performed in the Hashimoto patient group. Actually, it does not need to be confirmed histologically if no nodules are detected in this type of patient unless there are additional doubts.
High-resolution ultrasonography has become a valuable diagnostic tool in the evaluation of superficial structures, such as parathyroid glands and lymph nodes. This study focused on one of ultrasonography techniques, i.e. realtime strain sonoelastography. The results, based on Hashimoto’s lymph nodes, jugular reactive lymph nodes and parathyroid lesions, show that this method could add additional value to differential diagnosis of parathyroid lesions and benign neck lymph nodes.