Long-term Follow-up Sonography of Benign Cystic Thyroid Nodules after a Percutaneous Ethanol Injection:

Transcription

1 Long-term Follow-up Sonography of Benign Cystic Thyroid Nodules after a Percutaneous Ethanol Injection: The Incidence of Malignancy-mimicking Nodules 1 Ji Sung Park, M.D., Dong Wook Kim, M.D., Choong Ki Eun, M.D., Seok Jin Choi, M.D., Myung Ho Rho, M.D. 2 Purpose: To evaluate the incidence of malignancy-mimicking sclerosed thyroid nodules, from long-term follow-up ultrasonography (US) after an US-guided percutaneous ethanol injection (PEI). Materials and Methods: We examined 86 benign cystic thyroid nodules from 80 patients. The nodules were classified into two groups based on whether an aspiration (Group A, n=26) or non-aspiration (Group B, n=60) of infused ethanol was performed. The final follow-up US over 12 months was performed in all patients. Results: Of the 86 nodules, the cystic portion of 82 (95.3%) cases, from 76 patients, completely disappeared subsequent to the first follow-up US (Group A, n=24, Group B, n=58) (Chi-square test, p>0.05). Moreover, 46 sclerosed thyroid nodules showed two or more of the five sonographic criteria upon a follow-up US (Group A, n=13, 50.0%, Group B, n=32, 53.3%). A higher ratio of the cystic portion of the nodules was associated with a higher incidence of the five sonographic criteria for malignancies detected via a follow-up US (p<0.01; Student s t-test). Conclusion: For the long-term follow-up US, the five sonographic criteria were observed in half the patients who received US-guided PEI. Also, by acknowledging the possibility that sonographic findings mimic a malignancy, since the sclerosed thyroid nodule, patients may avoid an unnecessary biopsy. Index words : Thyroid nodule Ethanol sclerotherapy Ultrasonography Sclerotherapy 1 Department of Radiology, Busan Paik Hospital, Inje University School of Medicine, Busan, South Korea. 2 Department of Radiology, Masan Samsung Hospital, Sungkyunkwan University School of Medicine, South Korea. This work was supported by the Inje Research and Scholarship Foundation in Received August 18, 2007 ; Accepted November 7, 2007 Address reprint requests to : Dong Wook Kim, M.D., Department of Radiology, Busan Paik Hospital, Inje University School of Medicine, Busan, South Korea, , Gaegeum-dong, Busanjin-gu, Busan, South Korea Tel Fax kim_dongwook@yahoo.com 21

2 Ultrasonography (US)-guided percutaneous ethanol injection (PEI) is a popular treatment method for cystic or solid benign thyroid nodules because it is simple, inexpensive, highly successful, and safe. In fact, several reports exist describing the efficacy and safety of USguided PEI for the treatment of benign cystic thyroid nodules (1 8). In spite of this, several unnecessary USguided fine-needle aspiration biopsy (FNAB) have been performed for the treatment of sclerosed thyroid nodules which mimick malignancy on follow-up US. Hence, the goal of this study is to determine the incidence of malignancy-mimicking sonographic findings in sclerosed thyroid nodules after US-guided PEI and evaluate the difference in the incidence of malignancy-mimicking sonographic findings in the sclerosed nodules based on the cystic percent make-up of the thyroid nodules. To the best of our knowledge, no previous studies have examined the possibility of malignancy-mimicking in benign cystic thyroid nodules by long-term follow-up US after an US-guided PEI has not been reported. In addition, we compared the incidence of malignancy-mimicking sonographic findings in the nodules that were or were not subjected to the aspiration of infused ethanol during US-guided PEI. Equation 1: (Volume of the cystic portion: length width height /6) / (Volume of the entire cystic thyroid nodule: length width height /6) 100 Under ultrasonographic guidance, an 18- to 23- gauge needle, which was selected according to the fluid viscosity of the cystic nodule, was inserted into the lumen of the cystic thyroid nodule to aspirate as much cystic fluid as possible. A local anesthetic was not routinely applied unless requested by the patient. Moreover, an appropriate amount of 99.9% ethanol was slowly instilled into the nodule cavity to a volume of % of the aspirated fluid without removing the needle. The thyroid nodules were classified into two patients groups based on whether aspiration (Group A, n=26) or non-aspiration (Group B, n=60) of infused ethanol was performed. For the Group A nodules (n=26), the instilled ethanol was nearly completely evacuated from the nodule using a second needle after 10 min. For the Group B nodules (n=60), an US-guided PEI was completed within 10 min, (i.e. the instilled ethanol was not evacuated). An US-guided PEI was not performed in the solid portions of the thyroid nodules in this study. Subjects and Methods Eighty-six biopsy-proven benign cystic thyroid nodules were examined in 80 patients (13 males, 67 females, age range years, mean 42.9 years), who have received US-guided PEI. As for the complex cystic thyroid nodules, we performed an US-guided FNAB for the cystic and solid components of the nodule. Moreover, for the purely cystic thyroid nodules, we aspirated all of the fluid, and immediately smeared it, and sent the remaining fluid into the pathologic department for the preparation of cell blocks. We obtained informed consent from all patients for all the US-guided FNA and PEI. The US-guided PEI was performed with a 128 XP/10 scanner (Acouson, Mountain View, CA U.S.A.) using a 7-MHz linear probe. Furthermore, the follow-up US was executed with an HDI 3000 scanner (Advanced Technology Laboratories, Bothwell, WA U.S.A.) or a 128 XP/10 scanner. US-PEI The cystic percentage of the thyroid nodules was determined using the following formula: 22 Follow-up US Multiple follow-up sonographic examinations were performed after US-guided PEI. We arbitrarily defined a successful US-guided PEI, for a benign cystic thyroid nodule, as the complete loss of the cystic portion of a thyroid nodule following the first follow-up sonogram. All cases with a complete disappearance of the cystic portion of the thyroid nodule after US-guided PEI revealed a purely solid nodule following the first followup US, a gradual decrease in size for a subsequent follow-up US and disappearance without recurrence for a long-term follow-up US. The recently accepted sonographic findings for malignant thyroid nodules demonstrate positive predictive value, despite low sensitivity, and make up the five sonographic criteria, which include hypoechogenecity, an irregular margin, microcalcification, intranodule vascularity and a more tall than wide formation (9 13). We arbitrarily adopted these songraphic findings as sonographic criteria and examined their incidence in postinstilled thyroid nodules for follow-up US. Over the course of more than 12 months of follow-up USs, the success rate and incidence of the five sonographic criteria in the study patients was investigated.

3 The incidence of five sonographic criteria was correlated with the cystic percentage of the initial thyroid nodules. Also, the incidence of the five sonographic criteria was compared for Groups A and B thyroid nodules. Results Of the 86 thyroid nodules, 68 were complex cysts and 18 were pure cysts. In a single nodule, the cystic portion was greater than half of the volume of the nodule in all cases. A simultaneous sclerotic therapy was performed in four cases (two nodules in three cases and four nodules in one case) for multiple benign cystic thyroid nodules. Table 1 and 2 lists the clinical data, the results of the sclerotherapy, and the incidence of the five sonographic criteria in all of the treated patients. US-PEIT For the thyroid nodules belonging to Group A, the initial cystic volumes ranged from 2 to 20 ml (mean, 5.5 ml) before sclerotherapy. Moreover, the infused ethanol volumes ranged from 2.0 to 12.0 ml (mean, 4.7 ml). Further, the cystic percentage ranged from 50 to 100% (mean, 82.5%). For Group B nodules, the initial cystic volumes, before sclerotherapy, ranged from 1.0 to 96.0 ml (mean, 9.8 ml), whereas the infused ethanol volumes ranged from 1.0 to 20.0 ml (mean, 6.4 ml). The cystic portion of the thyroid nodules ranged from 50.0 to 100.0% (mean, 86.9%). US-guided PEIs were well tolerated in all patients, with no significant complications related with the procedure. Follow-up US The first follow-up US was performed between 1 and 12 months after a US-guided PEI (mean, 6.6 months for Group A and 4.4 months for Group B), whereas the last follow-up was performed between 12 and 45 months after US-guided PEI (mean, 25.1 months for Group A and 19.6 months for Group B). We obtained an excellent success rate for the US-guided PEI in both groups (95.3%; 92.3% in Group A and 96.7% in Group B) as Table 1. Clinical Data, Results of US-guided PEI and Incidence of Five sonographic criteria in Group A Sex Age Cystic First Infused First follow-up Last follow-up Five sonographic (yrs) portion (%) aspirate (ml) ethanol (ml) (month) (month) criteria Complications F a,b,c mild pain F a,b F mild pain F a,b,c,d F mild pain F c,d F F a,b mild pain F F a,b,c F F F F a,b,c,e F M a,b,c,e F mild dizziness F a,b,c mild pain & facial palsy F a,b M a,b mild pain F F b mild pain F F b,d F F a,b,e Mean five sonographic criteria * - a: hypoechogenecity, b: irregular margin, c: microcalcification, d: intranodule vascularity, e: a shape taller than wide 23

4 Table 2. Clinical data, Results of US-guided PEI and Incidence of the five sonographic criteria in Group B Sex Age Cystic First Infused First follow-up Last follow-up Five sonographic (yrs) portion (%) aspirate (ml) ethanol (ml) (month) (month) criteria Complications M a,b F a,b,d mild pain M a,b,e M F b,c headache F a, F a, F mild pain F b M b,c F a,b,c F a,b F mild pain F F a,b,c mild pain M a,b,c mild pain M a,b,c F a,b F F a,b,e moderate pain F b moderate pain F a F a,b mild pain F mild pain M a,b,c M F a,b,c mild pain F F F b,c,d mild pain F a,b,c F a,b,c F F a,b,c mild pain & headache F a,b mild pain & headache F mild pain & headache F mild pain & headache F a,b,c F F M a,b F mild pain F F c,e mild pain F a,b M a,b mild pain F mild pain F a,b F a,b,c F a,b F a,b,e F a,c mild pain F mild pain F F F moderate pain M a,c M a F F a,b,e Mean five sonographic criteria * - a: hypoechogenecity, b: irregular margin, c: microcalcification, d: intranodule vascularity, e: a shape taller than wide 24

5 J Korean Radiol Soc 2008;58:21-28 well as successfully sclerosed nodules for the purely solid nodule without any remnant cystic portion after a follow-up US (Figs. 1, 2). Further, we found no thyroid nodule recurrence for the successfully post-instilled thyroid nodules in our study. However, US-guided PEIs failed to obliterate the cystic portion of the four thyroid nodules (4.7%; two nodules in Group A and two nodules in Group B). The greater the cystic portion of the thyroid nodule before US-guided PEI begins, the higher the incidence of the five sonographic criteria, which are detected by long-term follow-up US (p<0.01; Student s t-test) (Table 3). Six nodules (6.9%; two in Group A and four in Group B) showed a complete disappearance as seen on longterm follow-up US. A For Groups A and B, three or more five sonographic criteria were revealed in 7 (26.9%) and 17 cases (28.3%), respectively. Two or more of the five sonographic criteria were detected on US in 13 (50%) and 32 cases (53.3%), respectively. One or more of the five sonographic criteria were revealed by a US in 14 (53.8%) and 38 cases (63.3%), respectively, and no five sonographic criteria were observed following an US in 12 (46.2%) and 22 cases (36.7%), respectively (Table 4). The results indicates a high incidence rate of five sonographic criteria in the sclerosed thyroid nodules in the two groups, and there was no statistical difference in the incidence rate of five sonographic criteria between the two groups (p>0.05, chi-square test) (Table 4 & 5). B C D E F Fig. 1. A 55-year-old man with a cystic thyroid nodule in the right lobe. A. A transverse US obtained before sclerotherapy shows a pure thyroid cyst filled with colloid material. B. A transverse US obtained one month after sclerotherapy reveals the loss of the cystic portion of the thyroid nodule and replacement with hypoechoic and other echogenic portions. C-E. A transverse, sagittal and color Dopper US obtained 15 months after sclerotherapy shows hypoechogenecity, an irregular margin, microcalcification, and a taller than wide thyroid nodule. However, this nodule shows no vascularity on a color Doppler study. F. A transverse US obtained 22 months after sclerotherapy shows a moderate decrease in nodule size. 25

6 Ji Sung Park, et al : Long-term Follow-up Sonography of Benign Cystic Thyroid Nodules after a Percutaneous Ethanol Injection Discussion Instead of the surgical removal of benign cystic thyroid nodules, several therapeutic methods, such as percutaneous simple aspiration, thyroid hormone supprestable 3. Incidence of the Five Sonographic Criteria according to the Percentage of Cystic Portion in Thyroid Nodules Cystic portion (%) Group A (n=26) Group B (n=60) Total (n=86) /5 (0%) 0/0 (0%) 0/1 (0%) 3/6 (50%) 0/2 (0%) 0/4 (0%) 2/5 (40%) 5/11 (45.5%) 0/7 (0%) 0/4 (0%) 2/6 (33.3%) 8/17 (47.1%) A sion therapy, percutaneous tetracycline instillation, or US-guided PEI, have been utilized (14, 15). At present, a US-guided PEI has become the first line treatment method for benign cystic or solid thyroid nodules because of its ease and safety (1-8). The instilled ethanol in the nodule, induces complex Table 4. Incidence of the Five Sonographic Criteria in Group A, Group B, and Total Cases (Chi-square test p>0.05 between Group A and B). No. of Malignant sonographic findings Group A (n=26) Group B (n=60) Total cases (n=86) (26.9%) 13 (50.0%) 14 (53.8%) 12 (46.2%) 17 (28.3%) 32 (53.3%) 38 (63.3%) 22 (36.7%) 24 (27.9%) 45 (52.3%) 52 (60.5%) 34 (39.5%) 10/14 (71.4%) 23/38 (60.5%) 33/52 (63.5%) B C D E F Fig. 2. A 46-year-old man with a cystic thyroid nodule in the left lobe A. A transverse US obtained before sclerotherapy shows a large hemorrhagic thyroid nodule with a 90% cystic portion and without vascularity. B. A transverse US obtained 3 months after sclerotherapy reveals a complete disappearance of the previous cystic portion of the thyroid nodule filled with a non-vascularized, hypoechoic component. C-E. A transverse, sagittal and color Dopper US obtained 26 months after sclerotherapy shows hypoechogenecity, microcalcification, some blurred margins, a taller than wide sclerosed thyroid nodule with no vascularity. F. A transverse US obtained 34 months after sclerotherapy shows a mild decrease in nodule size. 26

7 Table 5. Incidence of the Five Sonographic Criteria in the Sclerosed Thyroid Nodules Five sonographic Group A Group B Chi-square criteria (n=26) (n=60) test Hypoechogenicity 11 (42.3%) 32 (53.3%) p>0.05 Irregular margin 13 (50%).0 31 (51.7%) p>0.05 Microcalcification 07 (26.9%) 17 (28.3%) p>0.05 Intranodule vascularity 03 (11.5%) 2 (3.3%) p>0.05 Shape: more tall than wide 03 (11.5%) 5 (8.3%) p> thyroid tissue damage, including coagulative necrosis, vascular thrombosis, and a hemorrhagic infarction. Moreover, the treated areas were replaced by granulation tissue, which causes scarring and progressive shrinkage of the nodules (16, 17). In the present study, we defined a recurrence as a new cyst formation in the sclerosed portion of a thyroid nodule, following a long-term follow-up US, and a successful US-guided PEI because we focused the loss of the cystic portion in thyroid nodule. However, although an US-guided PEI, which failed to obliterate the cystic portion of the four thyroid nodules (4.7%; two nodules in Group A and two nodules in Group B), we did not observe any recurring cases. Also, among the sclerosed thyroid nodules observed in our study, 86 were observed after a first follow-up US over 12 months with a total of 45 cases revealing two or more of the five sonographic criteria (52.3%). The high rate for the five sonographic criteria in known benign cystic thyroid nodules, after a US-guided PEI, has a direct relationship with the high percentage of cystic proportion of these nodules. The reason for the high incidence of five sonographic criteria stems from purely cystic or non-vascularized thyroid nodules, with a 90% or greater percentage of the cystic component after the US-PEIT is uncertain. However, this result strongly suggests that the main component for a possible malignancy-mimicking, is fibrosis. Also, because of the sclerosed thyroid nodules, with a purely cystic component, appear we have a high incidence of the five sonographic criteria on US. We evaluated the differences in success rate, complications, and incidence of the five sonographic criteria,based on the aspiration or non-aspiration of infused ethanol, and found that no significant differences was found between the two groups. As a result, we performed several unnecessary USguided FNAB of benign sclerosed thyroid nodules, associated with the presence of the five sonographic criteria as detected by long-term follow-up USs. In two cases, we lost the patient history of the US-guided PEI. Also, in four of the cases, we performed US-guided FNAB to rule out the presence of a new malignancy in the known sclerosed thyroid nodule. In all six cases, no malignant cells were present, despite the complete performance of the consecutive US-guided FNAB. In our hospital, a US-guided FNAB for a thyroid nodule is performed according to Yokozawa s method (18). In addition, we have achieved a high percentage of the satisfactory cytological results (over 90% in the first US-guided FNAB) for thyroid nodules, except for the purely cystic thyroid nodule. We are convinced that the possibility of a newly developed malignancy after a US-guided PEI was ruled out because most sclerosed nodules, with their aberrant sonographic findings, showed a gradual decrease in size after an initial follow-up US. We confirmed several malignancy-mimicking nodules as a benign nodule by at least one or more satisfactory cytologic results by consecutive US-guided FNAB. In addition, over the four years of this study, we observed several benign cystic thyroid nodules that appeared as malignancy-mimicking nodules after the spontaneous collapse on long-term follow-up US. The limitations of this study include access to regular and close long-term follow-up US for the identification of the sclerosed thyroid nodules. The laboratory correlations were insufficient for the purely solid or predominantly solid thyroid nodules after the US-guided PEI were not included in this analysis Conclusion We confirmed that the long-term follow-up US of sclerosed thyroid nodules was made in (i.e. in our hospital). Next, the fiive sonographic criteria were observed in half of the patients. received US-guided PEI for known benign cystic thyroid nodules. By acknowledging the possibility that sonographic findings mimic a malignancy the sclerosed thyroid nodule, patients may avoid an unnecessary biopsy. References 1. Verde G, Papini E, Pacella C, Gallotti C, Delpiano S, Strada S, et al. Ultrasound guided percutaneous ethanol injection in the treatment of cystic thyroid nodules. Clin Endocrinol 1994;41: Cho YS, Lee HK, Ahn IM, Lim SM, Kim DH, Choi CG, et al. Sonographically guided ethanol sclerotherapy for benign thyroid cysts: results in 22 patients. AJR Am J Roentgenol 2000;174: Kim DW, Rho MH, Kim HJ, Kwon JS, Sung YS, Lee SW. Percutaneous ethanol injection for benign cystic thyroid nodules:

8 is aspiration of ethanol-mixed fluid advantagenous? AJNR Am J Neuroradiol 2005;26: Kim JH, Lee HK, Lee JH, Ahn IM, Choi CG. Efficacy of sonographically guided percutaneous ethanol injection for treatment of thyroid cysts versus solid thyroid nodules. AJR Am J Roentgenol 2003;180: Yasuda K, Ozaki O, Sugino K, Yamashita T, Toshima K, Ito K, et al. Treatment of cystic lesions of the thyroid by ethanol instillation. World J Surg 1992;16: Monzani F, Lippi F, Goletti O, Del-Guerra P, Caraccio N, Lipolis PV. Percutaneous aspiration and ethanol sclerotherapy for thyroid cysts. J Clin Endocrinol Metab 1994;78: Bennedbak FN and Hegedus L. Treatment of recurrent thyroid cysts with ethanol: a randomized double-blind controlled trial. J Clin Endocrinol Metab 2003;88: Lee SJ, Ahn IM. Effectiveness of Percutaneous Ethanol Injection Therapy in Benign Nodular and Cystic Thyroid Diseases: longterm follow-up experience. Endocrine Journal 2005;52: Papini E, Guglielmi R, Bianchini A, Crescenzi A, Taccogna S, Nardi F, et al. Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-doppler features. J Clin Endocrinol Metabol 2002;87: Kim EK, Park CS, Chung WY, Oh KK, Kim DI, Lee JT, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of non-palpable solid nodules of the thyroid. AJR Am J Roentgenol 2002;178: Frates MC, Benson CB, Doubilet PM, Cibas ES, Marqusee E. Can color Doppler sonography aid in the prediction of malignancy of thyroid nodules? J Ultrasound Med 2003;22: Iannuccilli JD, Cronan JJ, Monchik JM. Risk of Malignancy of thyroid nodules as assessed by sonographic criteria: the need for biopsy. J Ultrasound Med 2004; 23: Silver RJ, Parangi S. Management of thyroid incidentalomas. Surg Clin North Am 2004;84: Antonelli A, Campetelli A, Di Vito A, Alberti B, Baldi V, Salvioni G. Comparison between ethanol sclerotherapy and emptying with injection of saline in the treatment of thyroid cysts. Clin Invest 1994;72: Hegedus L, Hansen JM, Karstrup S, Torp-Pedersen S, Juul N. Tetracycline for sclerosis of thyroid cysts. Arch Intern Med 1988;148: Crescenzi A, Papini E, Pacella CM, Rinaldi R, Panunzi C, Petrucci L, et al. Morphological changes in a hyperfunctioning thyroid adenoma after percutaneous ethanol injection: histological, enzymatic and sub-microscopical alterations. J Endocrinol Invest 1996;19: Livraghi T, Paracchi A, Ferrari C, Reschini E, Macchi RM, Bonifacino A. The treatment of autonomous thyroid nodules with percutaneous ethanol injection: 4-year experience. Radiology 1994;190: Yokozawa T, Miyauchi A, Kuma K, Sugawara M. Accurate and Simple Method of Diagnosing Thyroid Nodules by the Modified Technique of Ultrasound-Guided Fine Needle Aspiration Biopsy. Thyroid 1995;5: