|Year : 2017 | Volume
| Issue : 1 | Page : 28-31
Diagnostic accuracy of ultrasound imaging in Hashimoto's thyroiditis
Anuradha Kapali, Jaipal Beerappa, P Raghuram, Ravindra Bangar
Department of Radiodiagnosis, Kidwai Memorial Institute of Oncology, Bengaluru, Karnataka, India
|Date of Web Publication||20-Feb-2017|
C4, Subiksha Courette, Thulasi Theatre Main Road, Marathahalli, Bengaluru, Karnataka
Source of Support: None, Conflict of Interest: None
Context: There are studies describing the ultrasound (USG) features of Hashimoto's thyroiditis in literature; however, we have not come across studies determining the accuracy of USG in diagnosing Hashimoto's thyroiditis. Aims: We evaluated the cases referred to our institute with suspected thyroid abnormalities and studied in them the accuracy of USG in diagnosing Hashimoto's thyroiditis and also studied the associated malignancies and their USG characteristics. Settings and Design: The patients referred to our department with suspected thyroid abnormalities were included in the prospective study. The study period was of 1 year; we included 28 patients with Hashimoto's thyroiditis. Materials and Methods: We evaluated the USG features of the cases namely echogenicity, echotexture, micronodules, and increased vascularity and followed them up for final diagnosis by fine needle aspiration cytology, histopathology, or antithyroglobulin and thyroid peroxidase tests, other 60 cases were used as a control. The results were analyzed. Statistical Analysis Used: Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Results: Hashimoto's thyroiditis was present in 28 patients. The most sensitive parameter in diagnosing Hashimoto's thyroiditis was hypoechogenicity and increased vascularity. The most specific parameter was micronodules. Nodules were seen in 13 patients, out of which malignant nodules was present in six patients. Microcalcification, thick halo, and internal vascularity increase the likelihood of nodules being malignant. Conclusions: The most sensitive parameter in diagnosing Hashimoto's thyroiditis was hypoechogenicity and increased vascularity. The most specific parameter was micronodules. Coarsened echo texture had an intermediate sensitivity and specificity. The USG is a specific modality for diagnosing Hashimoto's thyroiditis with a good sensitivity. Microcalcification, thick halo, and internal vascularity also increase the likelihood of nodules being malignant in the background of Hashimoto's thyroiditis. Hence, these nodules must be subject to FNA.
Keywords: Accuracy, Hashimoto's thyroiditis, papillary carcinoma, sensitivity, ultrasound
|How to cite this article:|
Kapali A, Beerappa J, Raghuram P, Bangar R. Diagnostic accuracy of ultrasound imaging in Hashimoto's thyroiditis. Thyroid Res Pract 2017;14:28-31
|How to cite this URL:|
Kapali A, Beerappa J, Raghuram P, Bangar R. Diagnostic accuracy of ultrasound imaging in Hashimoto's thyroiditis. Thyroid Res Pract [serial online] 2017 [cited 2017 Apr 23];14:28-31. Available from: http://www.thetrp.net/text.asp?2017/14/1/28/200567
| Introduction|| |
Hashimoto's thyroiditis is an antibody and cell-mediated autoimmune thyroiditis which commonly affects middle-aged females in their fourth and fifth decade there is a female to male ratio of 10–15:1.
The definitive diagnosis of Hashimoto's thyroiditis is by pathologic assessment. Antithyroglobulin antibodies and thyroid peroxidase (TPO) antibodies are used as serological markers, which are positive in approximately 70 and 90–95% cases, respectively.
There are studies describing the ultrasound (USG) features of Hashimoto's thyroiditis in literature; however, we have not come across studies determining the accuracy of USG in diagnosing Hashimoto's thyroiditis. We evaluated the cases referred to our institute with suspected thyroid abnormalities and studied in them the accuracy of USG in diagnosing Hashimoto's thyroiditis and also studied the associated malignancies and their USG characteristics.
| Materials and Methods|| |
The patients referred to our department with suspected thyroid abnormalities were included in the study. The study period was for the duration of 1 year, we included 28 patients with Hashimoto's thyroiditis, and other 60 cases were used as a control, in whom there were no USG and no histopathologic examination (HPE) features of thyroiditis, but they underwent cyto/histopathological analysis due to suspected malignancy. Hashimoto's thyroiditis was confirmed by fine needle aspiration cytology (FNAC) or histopathology for operated cases for coexistent malignant nodules or by TPO and antithyroglobulin positivity. The sensitivity, specificity, positive, and negative predictive values were assessed for the individual USG features, and the overall accuracy of USG was determined.
The USG features considered were echogenicity, echotexture, micronodules, and vascularity. The echogenicity was compared to the strap muscles; thyroid was classified as hypoechoic when its echogenicity was equal to or less than the strap muscles. Normal thyroid has fine echoes if it was coarse then classified as coarsened echotexture [Figure 1]. Nodules were called micronodules when their size was ≤6 mm, there were other similar nodules in the thyroid, both hyper- and hypoechoic nodules were included. Solitary nodules and dissimilar nodules were not classified as micronodules. Subjective assessment of vascularity was made based on color Doppler and thyroid vascularity was categorized as increased or normal.
|Figure 1: Normal thyroid is homogenous with fine echoes and is more echogenic than strap muscles|
Click here to view
| Results|| |
Out of 31 patients with Hashimoto's thyroiditis, 27 were female and 4 were male.
Hypoechoic thyroid was seen in 23 patients and isoechoic thyroid in 8 patients. Three of the patients who had hypoechoic thyroid had diffuse adenomatous hyperplasia and were false positive. Eighteen patients had coarsened echotexture with one of them being false positive. Normal echotexture was present in 13 cases. Micronodules were present in seven cases, and all of these cases had Hashimoto's thyroiditis. Increased vascularity was seen in 23 patients and was false positive in three cases.
Thyromegaly was seen in 14 cases with normal sized thyroid in 17 cases. An anteroposterior length of 2 cm was considered as normal, and a length above this was considered enlarged. Thyroid stimulating hormone (TSH) was increased in 11 patients and normal in 17 patients. The sensitivity, specificity, positive, and negative predictive values of each feature have been provided in [Table 1]. The USG was 78.5% sensitive and 95.2% specific in diagnosing Hashimoto's thyroiditis. The positive and negative predictive values were 88% and 90.9%, respectively. The overall accuracy was 90.1%.
|Table 1: Sensitivity, specificity, positive predictive value, and negative predictive value of ultrasound features in thyroiditis|
Click here to view
Out of 28 cases of thyroiditis, nodules were seen in 13 cases. Four nodules were benign one of them was colloid cyst, three other nodules were hyperechoic with one showing thick halo, significant central vascularity, other two nodules had thin halo with peripheral vascularity, and none of the nodules had calcification, on FNAC theses nodules were diagnosed as adenomatous hyperplasia. Three nodules were diagnosed as nodular Hashimoto's on FNAC, two of these nodules were hypoechoic and one nodule isoechoic, no calcification or internal vascularity was seen. Six nodules were diagnosed by FNAC/HPE as malignant nodules, (papillary carcinoma) of these three were echogenic, one was hypoechoic, and two were heterogeneous, four nodules had significant internal vascularity, one nodule had only peripheral vascularity and one nodule had no vascularity. Three of the nodules had microcalcification. Four of the nodules had a thick halo.
| Discussion|| |
Thyroiditis refers to inflammation of thyroid gland; it includes chronic lymphocytic thyroiditis (Hashimoto's thyroiditis), de Quervain thyroiditis, acute thyroiditis, and Riedel's thyroiditis. Each of this thyroiditis has different clinical presentations and can be differentiated from each other clinically. They can have similar appearances on the USG.
In our study, the female to male ratio was 6–7:1.
The sonographic features of Hashimoto's thyroiditis include enlarged gland, which is hypoechoic, with coarsened parenchyma and is often hypervascular [Figure 2].
|Figure 2: Thyroiditis is imaged in the right lobe which has heterogenous coarsened echoes is isoechoic to strap muscles (anterior to the cursor) and is enlarged (normal anteroposterior dimension is 2 cm)|
Click here to view
Hypoechogenicity is a sensitive parameter in suspecting Hashimoto's thyroiditis. Hypoechogenicity has a positive and a negative predictive value of 88% and 93%, respectively, in diagnosing autoimmune thyroiditis. In our study, hypoechogenicity of thyroid and vascularity were the two most sensitive parameters in detecting thyroiditis (71%). The positive and negative predictive values were 87 and 88%, respectively. The decreased echogenicity is possibly due to lymphocytic infiltration of thyroid and has been shown to correlate with hypothyroidism in other studies. In our study, 73% (8) of patients with raised TSH had hypoechoic gland (11 patients), and three patients had normal echogenicity. In 11 patients (39%), USG could detect thyroiditis even before the hormonal abnormalities.
Color Doppler demonstrates slight to markedly increased vascularity of thyroid parenchyma possibly due to trophic stimulation of TSH. In our study, increased vascularity was seen in 82% of patients with a specificity of 95%.
In our study, coarsened echotexture had an intermediate sensitivity of 60% and a specificity of 98%.
Micro nodulation is a specific feature of Hashimoto's thyroiditis with a positive predictive value of 95% approximately. In our study, micro nodulation had a positive predictive value of 100%.
It is now known that Hashimoto's thyroiditis is a risk factor for developing papillary carcinoma and lymphoma.
The features suspicious for malignant thyroid nodules in a background of normal thyroid are microcalcifications, markedly reduced echogenicity, solid composition, and central vascularity.
In a study by Anderson et al., it was noted that the features of benign and malignant nodules in Hashimoto's thyroiditis remains the same as in the general population. However, any type of calcification pointed toward malignancy; margins; and vascularity did not play such an important role, as it was not well assessable in view of background changes.
In our study, 13 patients had nodules, which were discreet. The benign nodules included three nodular forms of Hashimoto's thyroiditis, three adenomatous hyperplastic nodules, and one colloid cyst. None of these nodules had any microcalcification. One nodule showed thick halo and internal vascularity. Rest of the nodules showed no significant internal vascularity with thin or no halo. Three of the nodules were hyperechoic, two were hypoechoic, and one nodule was isoechoic.
We identified six malignant nodules in our study; all of them were papillary carcinoma out of which four nodules had microcalcification (none of the benign nodules had calcification), four nodules had significant internal vascularity (one benign nodule had internal vascularity), four nodules had thick halo (one benign nodule had thick halo), three nodules were echogenic, two were heterogeneous, and one nodule was hypoechoic (findings somewhat similar to benign nodules). Hence, we conclude that calcification is the most specific feature, which may predict the malignant nature of nodule in a background of Hashimoto's thyroiditis. Thick halo and internal vascularity also increase the likelihood of nodules being malignant. Hypoechoic nodules raise the possibility of the malignant nodule in the normal thyroid; however, in our study, a significant proportion of malignant nodules were echogenic in the background of hypoechoic thyroid. Thus, in the background of Hashimoto's thyroiditis nodule echogenicity may not be an important feature to predict malignancy, this might be because the echogenicity of thyroid itself is altered in thyroiditis.
| Conclusion|| |
The most sensitive parameter in diagnosing Hashimoto's thyroiditis was hypoechogenicity and increased vascularity. The most specific parameter was micronodules. Coarsened echo texture had an intermediate sensitivity and specificity. The USG is a specific modality for diagnosing Hashimoto's thyroiditis with a good sensitivity.
Microcalcification, thick halo, and internal vascularity also increase the likelihood of nodules being malignant in the background of Hashimoto's thyroiditis; hence, these nodules must be subject to FNA.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Nordmeyer JP, Shafeh TA, Heckmann C. Thyroid sonography in autoimmune thyroiditis. A prospective study on 123 patients. Acta Endocrinol (Copenh) 1990;122:391-5.
Chaudhary V, Bano S. Thyroid ultrasound. Indian J Endocrinol Metab 2013;17:219-27.
Patel SB, Shah SR, Goswami KG, Patel HB. Pictorial essays: Ultrasound features of thyroid and parathyroid lesions. Indian J Radiol Imaging 2005;15:211-6.
Pedersen OM, Aardal NP, Larssen TB, Varhaug JE, Myking O, Vik-Mo H. The value of ultrasonography in predicting autoimmune thyroid disease. Thyroid 2000;10:251-9.
Hayashi N, Tamaki N, Konishi J, Yonekura Y, Senda M, Kasagi K, et al.
Sonography of Hashimoto's thyroiditis. J Clin Ultrasound 1986;14:123-6.
Yeh HC, Futterweit W, Gilbert P. Micronodulation: Ultrasonographic sign of Hashimoto thyroiditis. J Ultrasound Med 1996;15:813-9.
Holm LE, Blomgren H, Löwhagen T. Cancer risks in patients with chronic lymphocytic thyroiditis. N Engl J Med 1985;312:601-4.
Hoang JK, Lee WK, Lee M, Johnson D, Farrell S. US features of thyroid malignancy: Pearls and pitfalls. Radiographics 2007;27:847-60.
Anderson L, Middleton WD, Teefey SA, Reading CC, Langer JE, Desser T, et al.
Hashimoto thyroiditis: Part 2, sonographic analysis of benign and malignant nodules in patients with diffuse Hashimoto thyroiditis. AJR Am J Roentgenol 2010;195:216-22.
[Figure 1], [Figure 2]