Mummified Thyroid Syndrome

Transcription

1 Neuroradiology/Head and Neck Imaging linical Perspective Neuroradiology/Head and Neck Imaging linical Perspective lexis Lacout 1 arole hevenet 2 Pierre-Yves Marcy 3 Lacout, hevenet, Marcy PY Keywords: cystic-like thyroid nodule, Doppler ultrasound, mummy, thyroid cancer, thyroid nodule shrinkage, thyroid nodule, TIRDS DOI: /JR Received July 2, 2015; accepted after revision October 27, entre d Imagerie Médicale, 47 lvd dupont Rouge, urillac, France. ddress correspondence to. Lacout (lacout.alexis@wanadoo.fr). 2 entre de Pathologie, urillac, France. 3 Interventional Radiology Department, Polyclinique Les Fleurs, Ollioules, France. JR 2016; 206: X/16/ merican Roentgen Ray Society OJETIVE. The purpose of this article is to highlight the various sonographic characteristics that should help to differentiate a restructured benign collapsed thyroid nodule from histologically proven thyroid carcinoma by different imaging means, including Doppler sonography, and fine-needle aspiration cytologic analysis. ONLUSION. enign thyroid nodules may display morphologic changes over time, which can have misleading sonographic features suggestive of malignancy. Precise knowledge of certain sonographic imaging features, such as regular eggshell calcifications, peripheral hypoechoic or hypoechoic rim, posterior shadowing, and absence of intranodular vascularization, and meticulous comparison with previous images showing thyroid nodule shrinkage over time are useful for reaching the correct final diagnosis. Fine-needle aspiration cytologic assessment of such initially suspicious thyroid nodules and sonographic follow-up contribute to establishing the final diagnosis of benign thyroid findings. Knowledge of the elements described should help to identify the so-called mummified thyroid nodule and avert surgical excision. I ncidental diagnosis of thyroid nodules is common in sonographic examinations of the neck in adults. In such cases, extensive knowledge of the signs of thyroid nodules at high-frequency Doppler sonography, as described in the thyroid imaging reporting and database system (TIRDS) nodule classification, is mandatory [1, 2]. Thyroid nodules with at least one sonographically suspicious pattern of malignancy, that is, strong hypoechogenicity, microcalcifications, tallerthan-wide shape, ill-defined margins, and high stiffness index at elastography, should be evaluated by fine-needle aspiration cytologic analysis (FN) according to the ethesda 2010 classification regardless of the target thyroid nodule size [2, 3]. Furthermore, Doppler sonographic examination may help to depict supplementary suspicious features, including anarchic tangled central vasculature and a high resistance index (> 0.75) [4]. However, benign thyroid nodules may also display morphologic changes over time, which could be misleading as sonographic features suggestive of malignancy. These changes include strong hypoechogenicity, microcalcifications, and irregular margins secondary to nodule necrosis, desiccation, and subsequent collapse (Fig. 1). The so- called thyroid nodule mummification process may occur spontaneously over time or result from thyroid nodule FN, percutaneous ethanol injection, or radiofrequency ablation [5, 6]. Interestingly, Kim et al. [7] reported that the typical features of papillary thyroid carcinoma are similar to the typical sonographic aspects of malignancy. The purpose of this article is to highlight the various sonographic characteristics that should help to differentiate a restructured benign collapsed thyroid nodule from histologically proven thyroid carcinoma by using different imaging means, including Doppler ultrasound, and FN. Two of us have 23 and 10 years of experience in the sonographic assessment of thyroid nodules. From this experience we retrospectively reviewed suspicious TIRDS scores 4 and 5 thyroid nodules among approximately 2000 consecutively evaluated thyroid nodules over a 5-year period. requirement for the review was that the nodules display initial typically benign features and have a cytologically proven benign evolution. This particular type of evolution of such nodules presenting suspicious ultrasound features at end-stage diagnosis was arbitrarily called a mummification process because one of the key features of the diagnosis JR:206, pril

2 was a marked volume shrinkage due to thyroid nodule necrosis, desiccation, or scarring. One strength of our two-center study is that we gained insight into ultrasound features of misleading malignancy-like thyroid nodules, which to our knowledge has not precisely been reported to date. The other strength is that we report the sonographic evolution of initially likely benign nodules that became suspicious and hypothesize on the cytopathologic correlations with sonographic features of the mummification process. ecause different TIRDS classifications have been reported in the literature [1, 8], we assessed the nodules according to method of Russ et al. [1], as follows: TI- RDS score 4 includes mildly hypoechoic thyroid nodules and the absence of a highly suspicious pattern of malignancy. TIRDS score 4 includes one or two highly suspicious patterns. TIRDS score 5 includes at least three to five highly suspicious patterns, metastatic lymph nodes, or both. The equipment used was a Mylab 70 (Esaote) or a Z.one (Zonare) clinical ultrasound system. Pathologic Features In our experience, the thyroid nodule mummification process mostly concerns cystic or predominantly cystic TIRDS score 2 thyroid nodules rather than solid TIRDS score 3 nodules, and these nodules may occur either de novo or after FN or percutaneous ethanol injection (Fig. 2). entronodular thyrocyte hypoxia, nodule hemorrhage, and subsequent intracystic modification may play an important role in the initiation of a scarring process that leads to the mummy appearance seen at sonography. LiVolsi and Merino [9] reported worrisome histologic alterations after FN of the thyroid nodule and called the condition WHFFT syndrome. Interestingly, cytologic analysis revealed only acute and chronic benign changes. cute changes included hemorrhage, granulation tissue, giant cells and siderophages, mitoses and necrosis, nuclear clearing, poorly formed granuloma, capsular distortion, and rarely infarction. hronic changes included linear fibrosis near siderophage metaplasia (oncocytic, spindle cell, and squamous types), infarction, pseudoinvasion of the capsule, marked random nuclear atypia, cyst formation, papillary degeneration, and calcification. From a physiopathologic point of view, LiVolsi and Merino [9] speculated on the compromised vascular supply resulting from the extraction of large amounts of tissue, traumatic venous thrombosis, rough needle use, and multiple passes with vigorous aspiration after FN. Similarly, we hypothesize that interruption of the blood supply and the occurrence of venous thrombosis also may be related to the rapid growth of benign thyroid tumors and thus may explain the further cystic transformation found at sonographic assessment. However, some malignant thyroid nodules may also exhibit the worrisome histopathologic alterations [10]. Sonographic Features Shrinkage s long as the benign thyroid nodule that initially had benign ultrasound features will progressively shrink over time, ultrasound features resembling those of malignancy progressively appear (Figs. 1 5). s close follow-up was performed, we observed an almost 45% volume decrease of the examined mummified thyroid nodules over a mean time of 28 months (range, months). Thus, analyzing the sonographic changes in suspicious thyroid nodules over time by use of a PS is of prime importance. It is necessary to remember, however, that in rare cases genuine papillary thyroid carcinomas may exhibit volume stability over time at retrospective follow-up [11]. Furthermore, considerable tumor shrinkage has been reported in rare cases of micropapillary thyroid carcinoma. When comparison with previous examinations is impossible or the findings are doubtful and the patient s history remains unavailable, FN should be mandatory. Microcalcifications Typically, thyroid malignant microcalcification clusters correspond to so-called psammoma bodies. They represent one of the five strong criteria in favor of a diagnosis of papillary thyroid carcinoma or medullary thyroid carcinoma. However, echogenic foci also have been reported in benign conditions, namely, in dystrophic conditions and the benign thyroid nodule mummification process [5, 12] (Figs. 2, 6, and 7). Other Sonographic Features enign mummified thyroid nodules may also have the following suspicious sonographic features: ill-defined margins, tallerthan-wide shape, strong hypoechogenicity, and high stiffness index at elastographic examination [5]. Ill-defined margins can result from the collapse of an initially well-defined thyroid nodule (Figs. 2, 3, 5, 6, 8, and 9). The strong hypoechogenicity (Figs. 2, 3, 5, 6, 8, and 9) and high stiffness index (Fig. 3) can be explained by necrotic changes, cystic component shrinkage, and progressive dense fibrosis. Similarly, the thyroid nodule verticalization may be due to the asymmetric fibrous healing associated with transverse shrinking. Posterior sonographic shadowing also may contribute to a taller-than-wide appearance (Figs. 3, 5, 6, 8, and 9). lthough the aforementioned sonographic features may suggest thyroid malignancy, the association between posterior shadowing and a peripheral hypoechoic halo could suggest the diagnosis of benign necrotic thyroid nodule rather than malignancy [5]. Some necrotic nodules may have a double hypoechoic and hyperechoic rim (black-and-white halo), because the former corresponds to the development of tissue granulation and fibrous capsule initiation and the latter to eggshell calcification. We hypothesize that the hyperechoic layer underlying the hypoechoic thyroid nodule capsule corresponds to the granulation tissue that will calcify in some cases. This leads to the eggshell calcified nodule feature (Figs. 4, 7, 8, and 10). Eggshell macrocalcification typically suggests thyroid benignity, because it displays regular complete linear features. onversely, irregular or interrupted eggshell calcification, particularly when nodular tissue is present outside the calcification line, favors malignancy [13]. Doppler Sonography s expected, cystic necrotic mummified thyroid nodules frequently exhibit an absence of vasculature at Doppler sonographic examination (when an optimized protocol is used, reducing the pulse repetition frequency to MHz) (Figs. 2, 4, 5, 7, and 10). However, the lack of visualization of vasculature does not definitively rule out the possibility of thyroid malignancy, because a strong stromal desmoid reaction can preclude tumor neoangiogenesis [14]. onversely, the diagnosis of intranodular vasculature in this context goes against the diagnosis of thyroid nodule necrosis and thus should lead to prompt FN analysis of a likely malignant thyroid nodule. Fine-Needle spiration ytologic nalysis FN should especially be performed in cases in which comparison with a previous sonographic examination highlighting likely benign thyroid nodule shrinkage is not available. In cases of benign collapsed thyroid 838 JR:206, pril 2016

3 nodules, FN typically may display somewhat gross solid necrotic yellowish smear material with no malignant cells at microscopic examination (Fig. 8). ytopathologic examination may show histiocytes in the granulation tissue and altered cells corresponding to the apoptotic necrosis. However, in some challenging cases, cytopathologic examination may disclose atypia of unknown significance, which will be classified atypia of undetermined significance according to the ethesda 2010 classification [15] or Thy3 in the ritish Thyroid ssociation classification [16]. In those cases, precise knowledge of the sonographic features of the benign necrotic thyroid nodule and comparison with the previous sonographic examination is mandatory, leading to careful followup instead of inappropriate thyroid surgery. PET/T PET/T is not indicated in cases of suspicious thyroid nodules. However, in our experience, PET/T, when performed, shows no FDG uptake within the suspicious thyroid nodule owing to the absence of viable tissue (Fig. 7). Similarly, T shows no enhancement. However, some thyroid carcinomas may exhibit no FDG uptake on a PET scan, and papillary thyroid carcinoma usually has homogeneous low attenuation on T scans [17]. onversely, FDG uptake may be a strong argument for ruling out the diagnosis of benign mummification and thus should lead to prompt ultrasound-guided FN of thyroid nodules. Limitations Our study had important limitations. First, this two-center retrospective study focused on an arbitrary selection of TIRDS category 4 and 5 thyroid nodules that were eventually diagnosed as benign and thus had potential inherent bias. Second, close followup and FN were not systematically performed in all cases over the 5-year period of assessment. Third, although the cytologic examination was rigorous and helped in the final diagnosis, we did not perform a final histopathologic examination for benignity. more comprehensive multicenter prospective study is needed to complete our data and provide the specificity and sensitivity of the reported ultrasound findings. onclusion The diagnosis of benignity at ultrasound examination can be challenging because some mummified thyroid nodules that are desiccated and hemorrhagic after FN or percutaneous ethanol injection can mimic malignancy and misleadingly result in a high TIRDS score (> 3). Precise knowledge of certain typical imaging features, that is, regular eggshell calcifications, peripheral hyperechoic or hypoechoic rim, posterior shadowing, absence of intranodular vascularization, absence of ipsilateral suspicious lymph nodes [3, 4, 6], meticulous comparison with previous examinations (shrinkage), and FN and sonographic follow-up will contribute to establishing the final diagnosis of benignity and avoid unnecessary surgery. References 1. Russ G, Royer, igorgne, et al. Prospective evaluation of thyroid imaging reporting and data system on 4550 nodules with and without elastography. Eur J Endocrinol 2013; 168: Gharib H, Papini E, Paschke R, et al. merican ssociation of linical Endocrinologists, ssociazione Medici Endocrinologi, and European Thyroid ssociation medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. J Endocrinol Invest 2010; 33: Lacout, Isaac S, Marcy PY. Micro-medullary thyroid carcinoma: a diagnosis not to be missed. Postgrad Med J 2015; 91: De Nicola H, Szejnfeld J, Logullo F, et al. Flow pattern and vascular resistive index as predictors of malignancy risk in thyroid follicular neoplasms. J Ultrasound Med 2005; 24: Koo JH, Shin JH, Han K, et al. ystic thyroid nodules after aspiration mimicking malignancy: sonographic characteristics. J Ultrasound Med 2010; 29: Park NH, Kim DW, Park HJ, et al. Thyroid cysts treated with ethanol ablation can mimic malignancy during sonographic follow-up. J lin Ultrasound 2011; 39: Kim EK, Park S, hung WY, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. JR 2002; 178: Horvath E, Majlis S, Rossi R, et al. n ultrasonogram reporting system for thyroid nodules stratifying cancer risk for clinical management. J lin Endocrinol Metab 2009; 94: LiVolsi V, Merino MJ. Worrisome histologic alterations following fine-needle aspiration of the thyroid (WHFFT). Pathol nnu 1994; 29: Dali-Dannepond K, Fremy, Schoonberg. Worrisome histological alterations following FN of thyroid (WHFFT) [in French]. nn Endocrinol (Paris) 2013; 74: Simpson KW, lbores-saavedra J. Unusual findings in papillary thyroid microcarcinoma suggesting partial regression: a study of two cases. nn Diagn Pathol 2007; 11: Ellison E, Lapuerta P, Martin SE. Psammoma bodies in fine-needle aspirates of the thyroid: predictive value for papillary carcinoma. ancer 1998; 84: Park YJ, Kim J, Son EJ, et al. Thyroid nodules with macrocalcification: sonographic findings predictive of malignancy. Yonsei Med J 2014; 55: Lacout, Marcy PY. Highlights on power Doppler US of thyroid malignancy. Radiology 2010; 257: ochand-priollet, Vielh P, Royer, et al. Thyroid cytopathology: ethesda system nn Pathol 2012; 32: Lakhani R, Rourke T, Jefferis, et al. Thy3 cytology: what to do next? nn R oll Surg Engl 2011; 93: Kim DW. omputed tomography features of papillary thyroid carcinomas. J omput ssist Tomogr 2014; 38: (Figures start on next page) JR:206, pril

4 Fig. 1 Diagram shows sonographic assessment of thyroid nodule mummification., Predominantly cystic thyroid nodule with well-defined margins (thyroid imaging reporting and database system [TIRDS] category 2). Solid part exhibits some vasculature on Doppler ultrasound image., Sonographic feature of mummified nodule is that volume is less than at initial workup. Thyroid nodule has suspicious sonographic features (taller than wide, strongly hypoechoic, high stiffness index, and hyperechoic foci) suggesting psammoma bodies and leading to score of TIRDS 5. bsence of vasculature, presumed to be due to Doppler ultrasound finding of cell necrosis, is common. Regular peripheral eggshell macrocalcification is evident. bsence of any malignant lymph nodes at ipsilateral sites 3, 4, and 6 (Robbins classification) and of internal jugular vein invasion rules out invasive papillary thyroid carcinoma. Fig year-old woman with mummified thyroid nodule related to percutaneous ethanol injection (PEI). and, Doppler () and gray-scale () sonograms show 18-mm right subcapsular isthmolobar thyroid nodule. Doppler image shows strong heterogeneous pattern, global hypoechogenicity, some echogenic foci, ill-defined margins, and absence of nodule vasculature. Nodule has posterior reinforcement and no peripheral halo. Three years earlier, patient had undergone PEI of 54-mm partially cystic benign thyroid nodule for cosmetic reasons. Results of fine-needle aspiration cytologic and 3-year sonographic follow-up examinations ruled out malignancy. TRH = trachea. 840 JR:206, pril 2016

5 Fig year-old women with thyroid nodule mummification process related to fine-needle aspiration cytologic (FN) examination., Sonogram shows 25-mm predominantly cystic nodule (arrow) of right lobe, which was classified thyroid imaging reporting and database system category 2. FN analysis was performed on this predominantly cystic TIRDS 2 nodule., Gray-scale (left) and elastography (right) sonograms obtained 12 months after show 60% shrinkage of nodule, from 25 to 10 mm. Nodule was avascular; displayed hypoechogenicity, hyperechoic foci, ill-defined margins, high stiffness index, and was taller than wide (arrow). Fig year-old man with hemorrhagic thyroid nodule., Sonogram shows 20-mm predominantly cystic nodule of left lobe classified thyroid imaging reporting and database system category 2., Twenty-seven months after, patient reported neck pain and was referred to our institution. Sonogram shows increase up to 35 mm of long axis of nodule, whose heterogeneous cystic part suggests bleeding origin. (Fig. 4 continues on next page) JR:206, pril

6 Fig. 4 (continued) 52-year-old man with hemorrhagic thyroid nodule., Sonogram 5 months after shows nodule axis decreased to 19 mm. Nodule has no vasculature, and double black-and-white halo (arrow) is present at periphery. Fine-needle aspiration cytologic analysis showed poor cellularity and hemosiderin-containing histiocytes without malignancy. Fig year-old woman with thyroid nodule mummification process proven at fine-needle aspiration cytologic analysis (FN)., Sonogram shows 15-mm isoechoic nodule of right lobe. Tiny vascularized hypoechoic part (arrow) is visible. FN result was atypia of undetermined significance (ethesda system)., Sonogram obtained 19 months after shows almost 33% volume shrinkage of nodule, from 15 to 10 mm. Nodule was taller than wide, had poor peripheral vasculature, and displayed strong and diffuse hypoechogenicity and ill-defined margins but no peripheral halo. t follow-up, FN was performed twice and disclosed necrosis, poor cellularity, and background of hemosiderin-containing histiocytes. 842 JR:206, pril 2016

7 Fig year-old woman with spontaneous thyroid nodule mummification process., Sonogram shows 14-mm predominantly cystic left isthmolobar thyroid nodule with posterior reinforcement (star), classified thyroid imaging reporting and database system category 2., Sonogram obtained 4 years after shows almost 60% volume decrease of nodule, from 14 to 6 mm. Nodule is avascular and displays hypoechogenicity, microcalcifications, ill-defined margins, and posterior shadowing (star) but no peripheral halo. omparison with previous examination highlights nodule shrinkage and contributes to establishing final diagnosis. Results of fine-needle aspiration cytologic analysis confirmed diagnosis; colloidal background, glandular cells, histiocytes, and no suspicious or malignant cells were found. Fig year-old woman with right lobe mummified thyroid nodule., Doppler sonogram shows 10-mm thyroid nodule that is strongly hypoechoic and has ill-defined margins and tiny central hyperechoic foci (thin arrow) consistent with diagnosis of microcalcifications. Nodule was totally avascular. Posterior shadowing is evident. Discrete peripheral hyperechoic foci (thick arrow) may fit into first step of eggshell calcification process and suggest diagnosis of collapsed benign nodule., xial PET/T image shows incidental finding of no focal intrathyroid FDG uptake (arrow). (Fig. 7 continues on next page) JR:206, pril

8 Fig. 7 (continued) 60-year-old woman with right lobe mummified thyroid nodule. and D, Gross examination of fine-needle aspirate disclosed solid yellowish smear material (not shown). Photomicrographs (H and E, 40 []; H and E, 400 [D]) from cytologic examination show necrosis, macrophages, and microcalcifications (arrows). Fig year-old man with thyroid nodule mummification., Sonogram shows 17-mm left lobe thyroid nodule. This nodule displays absence of vasculature, strong hypoechogenicity, and ill-defined margins; seems taller than wide owing to posterior shadowing (star); and has black-and-white peripheral halo (arrow). Sonographic Doppler triad consisting of absence of vasculature, posterior shadowing, and peripheral halo suggested thyroid benignity and was confirmed at 2-year follow-up examination., Photograph of gross specimen from fine-needle aspiration cytologic analysis of left lobe thyroid nodule shows solid yellowish smear material., Photomicrograph (H and E, 400) shows background of necrosis and histiocytes and no suspicious or malignant cells. D 844 JR:206, pril 2016

9 Fig. 10 Diagram shows eggshell mummification process of capsule of thyroid nodule., apsule appears as continuous linear peripheral hypoechoic black halo., Initial appearance of capsular calcification as double black-and-white halo is due to underlying hyperechoic line., Final stage is mummification. apsule displays continuous eggshell calcification with posterior shadowing. Fig year-old man with thyroid nodule mummification process proved at fine-needle aspiration cytologic (FN) analysis., Sonogram shows 34-mm predominantly cystic nodule (cursors) of right lobe, classified thyroid imaging reporting and database system category 2. FN analysis was performed on this nodule that later appeared as mummified nodule., Sonogram 30 months after shows almost 75% volume decrease of nodule, from 34 to 9 mm. Nodule is avascular and displays hypoechogenicity, ill-defined margins, and posterior shadowing (star) but no peripheral halo., Gross examination of aspirate disclosed gross solid yellowish smear material. Photomicrograph (H and E, 400) shows necrosis, poor cellularity, and background of hemosiderin-containing histiocytes (arrow) with nonmalignant cells. JR:206, pril