Controlling recurrent papillary thyroid carcinoma in the neck by ultrasonographyguided

Eur Radiol (2008) 18: 835 842 DOI 10.1007/s00330-007-0809-5 HEAD AND NECK Byung Moon Kim Min Jung Kim Eun-Kyung Kim Sung Il Park Cheong Soo Park Woong Youn Chung Controlling recurrent papillary thyroid carcinoma in the neck by ultrasonographyguided percutaneous ethanol injection Received: 10 February 2007 Revised: 30 August 2007 Accepted: 23 October 2007 Published online: 27 November 2007 # European Society of Radiology 2007 B. M. Kim Department of Radiology, Sungkyunkwan University School of Medicine, Kangbuk Samsung Hospital, 108 Pyung-Dong, Jongro-Ku, Seoul, 110-746, South Korea M. J. Kim. E.-K. Kim (*) Department of Radiology, Yonsei University College of Medicine, 134 shinchon-dong, Seodaemoon-gu, Seoul, 120-752, South Korea e-mail: ekkim@yumc.yonsei.ac.kr Tel.: +82-2-2228-7400 Fax: +82-2-393-3035 S. I. Park Department of Radiology, Soonchunhyang University College of Medicine, Bucheon Hospital, Bucheon, South Korea C. S. Park. W. Y. Chung Department of Surgery, Yonsei University College of Medicine, 134 Shinchon-dong, Seodaemoon-gu, Seoul, 120-752, South Korea Abstract The purpose of this study was to retrospectively evaluate the efficacy of ultrasonography-guided percutaneous ethanol injection (PEI) for neck recurrence of papillary thyroid carcinoma (NR-PTC). Twenty-seven patients (19 80 years old; mean, 53.2) with 47 NR-PTCs were treated by PEI and were followed-up (14 38 months; mean, 28.2). Ethanol (99%) was repeatedly injected with adjusting needle position until the entire volume of NR-PTC was ablated. All patients received follow-up ultrasonography at 3 6-month intervals, and the percent volume decrease was measured. The NR-PTCs with decreased volume and no tumor vascularity on power-doppler study were regarded as treatment-effective. The NR-PTCs with stable or enlarged volume or with tumor vascularity were considered as treatment-failed, in which case PEI was repeated. The number of sessions and the total volume of ethanol per NR-PTC, and the volume of ethanol per session per NR-PTC were evaluated. All NR-PTCs significantly decreased in volume (range, 30 100%; mean, 93.6%). The mean number of sessions, the total volume of ethanol per NR-PTC, and the mean volume of ethanol per session per NR- PTC were 2.1 sessions (range 1 6), 2.4 ml (range 0.3 10.1), and 1.1 ml/ session (range 0.3 3.0), respectively. PEI appears to be an alternative option for controlling NR-PTCs in the selected patients who are poor surgical candidates. Keywords Percutaneous ethanol injection. Ultrasonography. Interventional. Papillary thyroid carcinoma. Lymph node. Cervical Introduction Thyroid cancer is the most common endocrine malignancy, among which papillary thyroid carcinoma (PTC) is the most common subtype, representing more than 80% of all thyroid cancers [1]. Well-differentiated papillary thyroid carcinoma follows an indolent course, with a 10-year cause-specific survival rate of 93% for patients with any stage [2]. However, neck recurrence occurs in up to 20% of patients with low-risk PTC and in 59% of patients with high-risk PTC [3 5]. Surgical treatment of the patients with neck recurrence of PTC (NR-PTC) can be challenging because of the scar formation from previous surgery, and these patients are at higher risk of complications, especially in those with repeated neck dissections [6]. Ultrasonography-guided percutaneous ethanol injection (PEI) is a minimally invasive technique that has been described as a method for treating hyperfunctioning thyroid

836 nodules [7] and sclerosing cystic thyroid nodules [8]. A small series reported that PEI might be a treatment option for patients with limited cervical lymph node metastases from PTC [9]. The purpose of this study is to retrospectively evaluate the efficacy of PEI for controlling NR- PTCs in patients who are poor surgical candidates. Materials and methods This retrospective study was approved by our institutional review board, and informed consent was not required. Twenty-seven patients were treated by PEI for NR-PTCs between September 2003 and December 2005, and were followed up to January 2007 (mean, 28.2±6.3 months; range, 14 38). The patients consisted of 5 men and 22 women, and the patient s average age was 53.2 years (range, 19 80 years). All NR-PTCs were confirmed by ultrasonography-guided fine needle aspiration prior to the procedure. All patients had undergone thyroidectomy with central lymph node resection, followed by radioactive iodine ablation for PTC. Lateral neck node dissection was also performed in ten patients. According to the AJCC (American Joint Committee on Cancer) staging, TNM stages were I in 2, II in 5, III in 10, and IV-A in 10 patients. The mean number of neck surgeries was 2.1 (range, 1 4). Overall, 47 NR-PTCs were treated by PEI. Among them, 37 NR-PTCs were primarily detected, and 10 NR-PTCs in 7 patients were newly detected foci of tumor in the neck apart from the primary NR-PTCs during follow-up. Seven of 47 NR-PTCs were located in the surgical bed, and the remaining 40 NR-PTCs were in the lateral neck. The inclusion criteria for PEI were as follows: (1) the presence of three or fewer NR-PTCs and no recurrence beyond the neck at the PEI point in time, (2) the NR-PTCs were not close to vessels to avoid the potential risk of ethanol intravasation this is a precaution just based on the authors opinion and (3) the patients with NR-PTCs were poor surgical candidates (the patients with high risk for general anesthesia due to a medical condition or for surgery due to repeated neck dissection) and/or preferred not to have further surgery. The medical records and the ultrasonography images during the procedure were reviewed by one staff radiologist who did not perform the procedure. PEI technique All patients who underwent PEI for NR-PTC were treated on an outpatient basis. Platelet counts and coagulation times were not routinely checked. None of the patients was premedicated. Following PEI, the patients were monitored in the radiology-postprocedure recovery room for 1 to 2 h with continuous observation of all vital signs and then were discharged. A 12-MHz linear transducer of the ultrasonography scanner (ATL HDI 5000; Philips-Advanced Technology Laboratories, Bothell, WA) was used in all procedures. Ultrasonography and PEI were performed by one of the two experienced staff radiologists. The longitudinal, transverse, and anteroposterior diameters of each NR- PTC were carefully measured to document the baseline volume, and then PEI was performed. The technique of PEI was referenced to the method described by Lewis BD et al. [9], with some modifications. A 3-cm long, 25-G needle was attached to a 1-ml syringe containing 99% ethanol. The skin and the soft tissue between the skin and NR-PTC were anesthetized with 1% lidocaine. The NR-PTC was targeted under ultrasonographic guidance using the free-hand technique. Ethanol injection was started from one end of the periphery of the NR-PTC and progressed toward the other end. At first, the ethanol was injected by a rate of 0.1 ml with real-time ultrasonographic monitoring. Immediately after the injection, the injected area became hyperechogenic, and the needle tip was obscured. After a short period of time, this intense echogenicity diminished, thus allowing visualization of the needle. The needle was then finely repositioned to the adjacent region without withdrawing the needle tip beyond the skin for further ethanol injection. In this way, ethanol injection was repeated until the entire volume of the NR-PTC seemed to be ablated (Fig. 1). The tip of the needle was kept under the skin until all the ethanol contained in the syringe was used. In cases in which all the ethanol in the first syringe was used and additional ethanol was needed, the first needle was retrieved, and a second needle was used to target the NR-PTC, and the procedure was continued. After the procedure, the volume of the ethanol injected and the procedure-related complications were recorded. Complications were investigated and classified into major and minor complications according to the reporting standards of Society of Interventional Radiology. Follow-up and post-treatment evaluation Ultrasonography was performed at 3 6-month intervals for evaluation of PEI effectiveness. The longitudinal, transverse, and anteroposterior diameter of the NR-PTCs were carefully measured, and tumor vascularity within the NR-PTCs was evaluated by power-doppler study. The measurements of the NR-PTCs and tumor perfusion within the NR-PTCs were saved in PACS and recorded into the radiological report. The pretreatment volume and the posttreatment follow-up volumes of each NR-PTC were calculated from the anteroposterior (a), transverse (b), and longitudinal (c) measurements as: Volume ¼ 4=3 π a=2 b=2 c=2: And the percent volume decrease of each NR-PTC was calculated as: VD (%) = (V 1 V 2 )/V 1 100 (VD = percent volume decrease, V 1 = pretreatment volume, V 2 = posttreatment volume).

837 Fig. 1 A 41-year-old woman with a recurrent mass in the right thyroid bed. (a) Pretreatment transverse ultrasonogram revealed a large recurrent mass (measured volume = 3,278 mm 3 ) in the right thyroid bed. (b), (c), (d) Ethanol was injected into the multiple different sites within the mass with adjusting needle position. Immediately following the injection, the area within the mass becomes hyperechogenic with posterior shadowing as it disseminates throughout the lesion (arrows). (e) The final follow-up ultrasonogram obtained 32 months after six sessions of percutaneous ethanol injection showed that the mass has markedly decreased in volume (measured volume = 56.7 mm 3, volume decrease = 98.3%). C = Carotid artery, m = mass, T = trachea Decreased volume of the lesion and absence of intralesional power-doppler signals were considered a sign of successful results of the treatment. Successfully treated NR-PTCs were followed-up as scheduled. In case of persistence of intralesional power-doppler signals and/or enlargement of the lesion, additional PEI sessions were performed in order to achieve a successful ablation. The number of sessions per NR-PTC for achieving treatment-effective results, the total volume of ethanol used per NR-PCT, and the volume of ethanol used per session per NR-PTC were evaluated. Statistical analysis The mean and standard deviation of data for all measurements were determined by using an Excel 2000 (version 9.0; Microsoft, Redmond, WA) spreadsheet. The relationship

838 Table 1 The results of neck recurrences of papillary thyroid carcinoma (NR-PTCs) treated by ultrasonography-guided percutaneous ethanol injection (PEI) No/sex Age N Surg Loc. (level) Type V Eth (ml) N Ses V Eth/ses (ml) D 1 (mm) V 1 (mm 3 ) D 2 (mm) V 2 (mm 3 ) VD (%) 1/F 72 1 Rt bed Mass 0.5 1 0.5 5.2 39.8 0 0 100.0 20 2/F 49 2 Rt (3) LN 0.9 1 0.9 5.7 72.7 3.9 18.4 74.7 15 3/F 61 2 Rt (2) LN 7.5 5 1.5 20.0 1839.1 5.0 47.1 97.4 26 Lt (2) LN 2.7 3 0.9 12.4 161.6 5.0 34.5 78.7 26 4/F 36 2 Rt bed Mass 2.2 3 0.7 6.9 130.1 0 0 100.0 25 *Rt (4) LN 0.5 1 0.5 6.6 51.1 0 0 100.0 19 5/F 41 3 Lt (4) LN 1.9 3 0.6 11.0 518.1 3.0 6.3 98.8 28 Lt (4) LN 4.5 3 1.5 16.3 844.5 2.0 4.2 99.5 28 6/F 40 2 Lt (3) LN 0.6 1 0.6 10.0 130.8 7.0 91.6 30.0 9 7/F 44 2 Lt bed Mass 0.7 1 0.7 11.1 418.2 7.0 73.3 82.5 9 8/M 61 2 Lt bed Mass 3.2 2 1.6 17.0 1334.5 5.0 27.9 97.9 8 9/M 32 3 Lt (4) LN 1.4 3 0.5 8.0 234.6 3.0 6.3 97.3 26 10/F 62 2 Rt (3) LN 1.8 2 0.9 8.0 83.7 2.0 1 98.8 20 Lt (4) LN 1.2 2 0.6 7.5 90.3 0 0 100.0 20 *Rt (3) LN 0.3 1 0.3 6.0 47.1 4.0 18.8 60.1 9 *Rt (2) LN 0.6 1 0.6 10.0 251.2 5.0 31.4 87.5 4 11/F 51 2 Lt (4) LN 1.2 2 0.6 8.0 196.4 0 0 100.0 24 12/F 55 3 Rt (3) LN 1.3 2 0.7 8.0 146.5 0 0 100.0 31 Rt (3) LN 0.3 1 0.3 3.0 14.1 0 0 100.0 31 13/F 40 2 Rt bed Mass 10.1 6 1.7 21.0 3297 5.4 56.7 98.3 32 Lt (4) LN 2.2 3 0.7 15.0 446.5 0 0 100.0 32 *Rt (4) LN 3.0 4 0.8 13.0 476.2 5.0 20.9 95.6 24 *Lt (3) LN 1.6 3 0.5 11.0 322.4 0 0 100.0 23 14/F 72 1 Lt (4) LN 3.3 3 1.1 12.0 452.2 0 0 100.0 21 15/F 51 3 Rt (4) LN 2.6 3 0.9 12.0 263.8 4.0 8.4 96.8 24 16/F 64 2 Lt (4) LN 7.5 3 2.5 29.0 3278.2 0 0 100.0 27 Rt (4) LN 0.3 1 0.3 5.0 41.9 0 0 100.0 27 17/F 61 2 Rt (4) LN 1.5 1 1.5 9.0 94.2 0 0 100.0 16 18/M 64 2 Lt (3) LN 4.7 2 2.4 22.0 2072.4 6.0 28.3 98.6 25 *Lt (4) LN 0.6 1 0.6 7.0 131.9 4.0 18.8 85.7 6 19/F a 80 1 Rt (2) LN 5.5 4 1.4 15.0 1224.6 5.0 24.0 98.1 14 Rt (4) LN 8.1 4 2.0 20.0 2041 5.0 31.4 98.5 14 Rt (4) LN 2.5 2 1.3 14.0 886.5 0 0 100.0 14 20/M 44 4 Lt (3) LN 1.8 2 0.9 10.0 183.2 0 0 100.0 24 *Lt (3) LN 0.5 1 0.5 6.0 77.5 3.0 8.1 89.5 20 *Rt (4) LN 0.5 1 0.5 6.0 78.5 3.0 6.3 92.0 4 21/F 70 1 Rt bed Mass 2.2 2 1.0 14.0 578.7 6.0 62.8 89.1 15 *Rt bed Mass 1.5 1 1.5 14.0 732.7 8.0 100.5 86.3 4 22/F 52 2 Lt (2) LN 1.0 1 1.0 14.0 410.3 0 0 100.0 19 Lt (5) LN 0.4 1 0.4 9.0 141.3 0 0 100.0 19 23/F 68 1 Lt (3) LN 4.3 3 1.4 13.0 544.3 7.0 54.9 89.9 25 24/F 59 2 Rt (4) LN 5.2 2 2.6 20.0 1962.5 0 0 100.0 22 25/F 46 2 Rt (4) LN 1.2 3 0.4 6.0 78.5 0 0 100.0 22 26/M 42 3 Lt (3) LN 3.1 2 1.6 13.0 476.2 2.0 2.1 99.6 27 *Lt (4) LN 0.7 1 0.7 13.0 333.4 7.0 73.3 78.0 6 F/U (months)

839 Table 1 (continued) No/sex Age N Surg Loc. (level) Type V Eth (ml) N Ses V Eth/ses (ml) D 1 (mm) V 1 (mm 3 ) D 2 (mm) V 2 (mm 3 ) VD (%) F/U (months) 27/F a 19 2 Rt (3) LN 1.1 1 1.1 12.0 263.8 0 0 100.0 24 Rt (3) LN 0.7 1 0.7 8.1 124 0 0 100.0 24 Mean± SD 53.2± 13.9 2.1± 0.7 2.4±2.3 2.1± 1.2 1.1±0.6 11.6± 5.3 587.6± 788.7 2.6± 2.6 18.2± 26.5 93.6± 12.6 26.0±8.0 No = Number of patient, N Surg = number of neck surgery, Loc. = location of the NR-PTC, N Ses = number of sessions per NR-PTC (times), V Eth = total volume of ethanol used per NR-PTC (ml), V Eth/Ses = the volume of ethanol used per session per NR-PTC (ml/session), D 1 = pretreatment maximum diameter of the NR-PTC (mm), D 2 = maximum diameter of the lesion on the last follow-up ultrasonography (mm), V 1 = pre-treatment volume of the lesion (mm 3 ), V 2 = volume of the lesion on the last follow-up ultrasonography (mm 3 ), VD (volume decrease) = (V1 V2)/V1 100 (%), F/U = follow-up period of each lesion (months), Rt = right, Lt = left, LN = lymph node *Newly detected foci of recurred PTC in the neck apart from the primary NR-PTCs during the follow-up a Patients who developed multiple lung metastases graph between the volume of each NR-PTC and the followup months were plotted with trend lines using an Excel. Statistical analyses were performed by using the SPSS for Windows (version 10.0; SPSS INC, Chicago, IL). The comparison graph between the pretreatment volume and the last follow-up volume of all NR-PTCs was made by SPSS for Windows. The pretreatment volume and the last follow-up volume of the NR-PTC were compared, using t-test for paired data. The relationship among the pretreatment volume of NR-PTC, the total volume of ethanol used and the number of sessions per NR-PTC, and the volume of ethanol used per session per NR-PTC were evaluated, using multiple regression analysis. The findings in the lymph node metastasis (n=40) group were compared with the mass in the surgical bed group (n= 7) in the pretreatment volume, the number of sessions, and the total volume of ethanol per NR-PTC, and the volume of ethanol per session per NR-PTC, using t-test for unpaired data. The frequency of pain and other complications between the two groups was compared using Fisher s exact test. Statistical significance was determined by p<0.05 for 95% confidence interval. decrease of all NR-PTCs was 83.8±18.7% (range 15 100%) at 6 months and 90.3±13.2% (range 30 100%) at 12 months. The primary NR-PTCs (n=37) decreased in volume from a mean of 678.8±859.6 mm 3 to a mean of 15.6±24.3 mm 3 (average volume decrease, 95.3±12.5%; range, 30 100%) during the follow-up (mean, 28.2± 6.3 months; range, 14 38 months,) and the newly detected NR-PTCs (n=10) decreased in volume from a mean of 250.2±212.2 mm 3 to a mean of 27.8±31.6 mm 3 (average volume decrease, 87.5±11.2%; range 60.1 100%) during the follow-up (mean, 17.9±8.1; range, 4 24 months) (Table 1, Fig. 2). In the 47 NR-PTCs, 21 (44.7%) NR-PTCs in 15 of the 27 patients were completely ablated and did not reappear until the last follow-up. In 11 (40.7%) of the Results The patient s characteristics and the outcomes of PEI treatment are detailed in Table 1. Volume change of NR-PTCs and treatment outcome All 47 NR-PTCs (n=47) significantly decreased in volume from a mean of 678.8±871.4 mm 3 to a mean of 15.7± 24.7 mm 3 (average volume decrease, 93.6±12.6%; range 30 100%; p<0.05) during the follow-up (mean, 26±8; range, 10 38 months) (Fig. 2). The average volume Fig. 2 The mean volume change of all neck recurrences of papillary thyroid carcinoma after percutaneous ethanol injection

840 Fig. 3 Graph reveals the relationship between the volume of the each recurrent neck lesion treated with percutaneous ethanol injection and the follow-up months, which progressively decreases during the follow-up period 27 patients, no NR-PTC could be detected on the last follow-up ultrasonography for those who were followed up for a mean of 23.2±4.2 months (range, 16 31). However, 1 of the 11 patients developed multiple lung metastases. Twenty-six (55.3%) NR-PTCs in 16 patients remained visible on the last follow-up ultrasonography. They showed a progressive decrease in volume without tumor vascularity on power-doppler study, or decreased volume without tumor vascularity, and then stable volume to the last follow-up ultrasonography (Figs. 3, 4). PEI was effective in controlling all the primarily detected (n=37) and new foci of NR-PTCs (n=10) in all 27 patients. Two patients developed multiple lung metastases during the follow-up period, but none of the 27 patients died. The number of PEI sessions and the volume of ethanol used The mean number of PEI sessions per NR-PTC was 2.1± 1.2 sessions (range, 1 6). In 21 (44.7%) of all 47 NR-PTCs and in 8 (29.6%) of all 27 patients, only one PEI session was required to achieve treatment-effective results. Two or more PEI sessions were required to achieve treatmenteffective results for 26 (56.3%) of 47 NR-PTCs and in 19 (70.4%) of 27 patients. In 5 of 26 NR-PTCs, repeat PEI was conducted due to persistent intra-lesional power- Doppler signals in spite of decreased size of the NR-PTC. The total volume of ethanol per NR-PTC was 2.4±2.3 ml (range 0.3 10.1 ml). The average volume of ethanol used per session for each NR-PTC was 1.1±0.6 ml/session (range 0.3 3.0 ml/session). When the relationships among the pretreatment volume of NR-PTC, the number of sessions, the total volume of Fig. 4 A 41-year-old woman with lymph node metastasis in the left neck level 4. (a) A pretreatment longitudinal ultrasonogram showed an enlarged lymph node with focal psammomatous calcification (arrow) in level 4 of the left neck (measured volume =518.1 mm 3 ). (b) A follow-up ultrasonogram obtained 3 months after percutaneous ethanol injection shows that the lymph node has decreased in volume (65.7 mm 3 ). (c) The final follow-up ultrasonogram obtained 28 months after percutaneous ethanol injection showed that the lymph node had further decreased in volume (measured volume = 6.3 mm 3, volume decrease =98.8%)

841 ethanol used per NR-PTC, the volume of ethanol used per session for each NR-PTC, and the percent volume decrease after PEI treatment were evaluated by using multiple regression analysis, the pretreatment volume of NR-PTC showed statistically significant positive correlation with the number of sessions per NR-PTC and the total volume of ethanol used per NR-PTC (p<0.05). The pretreatment volume of NR-PTC showed a stronger positive correlation with the total volume of ethanol used per NR-PTC (coefficient, 0.907) than with the number of sessions per NR-PTC (coefficient, 0.597). There were no significant differences between the mass in the surgical bed group and the lymph node metastasis group in the pretreatment volume, the total volume of ethanol per NR-PTC, the number of sessions per NR-PTC, and the volume of ethanol per session per NR-PTC (p>0.05) (Table 2). Complications There were no procedure-related major complications. The minor complications after the PEI procedure were localized pain at the injection site and transient hoarseness. Localized pain at the injection site was recorded in 73 (73%) of the total of 100 PEI sessions. The pain occurred in 12 (75%) of 16 sessions for the mass in the surgical bed group (n=7) and in 61 (72.6%) of 84 sessions for the lymph node metastasis group (n=40). There was no significant difference in the frequency of the pain occurrence between the two groups (p>0.05) (Table 3). In all of the cases that pain occurred, the pain spontaneously subsided in a few hours without requiring analgesics. In one patient in whom PEI was performed for the mass in the surgical bed, transient hoarseness occurred immediately after the procedure, but completely recovered within 1 week. Table 2 Recurred mass in thyroid bed vs. lymph node metastasis Mass (6 patients, 7 NR-PTCs) LN (23 patients, 40 NR-PTCs) p-value V Pre (mm 3 ) 933.0±1,127.2 527.2±727.3 0.218 V Eth (ml) 2.9±3.3 2.3±2.1 0.496 N Ses (times) 2.3±1.8 2.1±1.1 0.712 V Eth/Ses (ml/session) 1.1±05 1.0±0.6 0.592 Sessions = Total number of sessions for each group, V Pre = pretreatment volume, N Ses = number of sessions per NR-PTC, V Eth = total volume of ethanol, V Eth/Ses = volume of ethanol per session, Mass = mass in thyroid bed group, LN = lymph node metastasis group Table 3 Complications of mass vs. lymph node metastasis Mass LN Total P Sessions (times) 16 84 100 Pain (%) 12 (75) 61 (72.6 73 (73) 0.56 Transient hoarseness (%) 1 (6.3) 0 (0) 1 (1) 0.16 Mass = Mass in thyroid bed group, LN = lymph node metastasis group, Pain = frequency of the pain at the injection site for each group, Transient hoarseness = frequency of transient hoarseness for each group Discussion Given the indolent nature of the disease, surgery may be an overly aggressive treatment for some patients with NR- PTC. Reoperation may be associated with complications such as injury to the recurrent laryngeal nerve, hypoparathyroidism, palsy of the spinal accessory nerve, and the cosmetic concerns related to cervical incisions [4, 10, 11]. Our study has shown that PEI effectively controlled all of the NR-PTCs, primarily detected (n=37) and newly detected foci during follow-up (n=10) in all of the 27 patients, with a significant decrease in the volume (p< 0.05) (Fig. 2). As in other neoplasia, the mechanism of PEI for treatment of NR-PTC was thought to causes ischemic tumor necrosis by inducing thrombosis of intralesional vessels [7]. Pain or discomfort at the injection site was the most frequent complication, likely related to local leakage of a small amount of ethanol into the surrounding soft tissue. The pain spontaneously subsided in a few hours without requiring analgesic in all of the cases. A single case of transient laryngeal nerve damage was observed among the patients with NR-PTC located in thyroid bed. None of the patients with lymph node metastasis showed this complication. Although this difference is not statistically significant, when NR-PTC is centrally located along the presumptive course of the laryngeal nerve, the ethanol should be fractionated into smaller volumes in each of its injections to avoid the risk of ethanol leakage outside the NR-PTC. As the number of NR-PTC increases, so will the difficulty of PEI. The time required for the procedure will increase, as well as the discomfort of the patient. Therefore, PEI was performed in patients with three or fewer NR-PTCs in this study. Another technical difficulty that can be encountered is treating a large NR-PTC. The injected ethanol will not diffuse throughout the entire volume of the lesion, and multiple sessions of PEI will be required for complete ablation. PEI has many advantages in treating NR-PTC over other treatment options. PEI is far less invasive than surgical neck reexploration and can be repeated without increased

842 technical difficulty. There may be a limit to how many surgical reexplorations can be safely attempted because of the increased technical difficulty, but not in PEI. Also, PEI has little or no morbidity so it can be done on an outpatient basis. The patients could resume their normal activities after 1 to 2 h of observation in the radiology-postprocedure recovery room. Limitations of this study include the small number of patients, the short follow-up period, and the retrospective nature of the study. Given the indolent nature of PTC, no real inference can be made concerning the effect of PEI on the long-term outcome or efficacy for treatment of patients with NR-PTC. In conclusion, although with some limitations, PEI was effective for controlling the NR-PTCs. The results of this study suggest that PEI may be an alternative treatment option for control of NR-PTCs in selected patients who are poor surgical candidates. References 1. Greenlee RT, Hill-Harmon MB, Murray T, Thun M (2001) Cancer statistics. Cancer J Clin 51:15 36 2. Hundahl SA, Fleming ID, Fremgen AM, Menck HR (1998) A national cancer data base report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985 1995. Cancer 83:2638 2648 3. Cady B, Rossi R (1988) An expanded view of risk-group definition in differentiated thyroid carcinoma. Surgery 104:947 953 4. Mazzaferri EL (1999) NCCN thyroid carcinoma practice guidelines. Oncology (Huntingt) 13:391 442 5. Hay ID, McConahey WM, Goellner JR (2002) Managing patients with papillary thyroid carcinoma: insights gained from the Mayo Clinic s experience of treating 2,512 consecutive patients during 1940 through 2000. Trans Am Clin Climatol Assoc 113:241 260 6. Samaan NA, Schultz PN, Hickey RC et al (1992) The results of various modalities of treatment of well differentiated thyroid carcinoma: a retrospective review of 1,599 patients. J Clin Endocrinol Metab 75:714 720 7. Tarantino L, Giorgio A, Mariniello N et al (2000) Percutaneous ethanol injection of large autonomous hyperfunctioning thyroid nodules. Radiology 214:143 148 8. Cho YS, Lee HK, Ahn IM et al (2000) Sonographically guided ethanol sclerotherapy for benign thyroid cysts: results in 22 patients. AJR 174:213 216 9. Lewis BD, Hay ID, Charboneau JW, McIver B, Reading CC, Goellner JR (2002) Pecutaneous ethanol injection for treatment of cervical lymph node metastases in patients with papillary thyroid carcinoma. AJR 178:699 704 10. Mazzaferri EL, Young RL (1981) Papillary thyroid carcinoma: a 10 year follow-up report of the impact of therapy in 576 patients. Am J Med 70:511 518 11. Esnaola NF, Cantor SB, Sherman SI, Lee JE, Evans DB (2001) Optimal treatment strategy in patients with papillary thyroid cancer: a decision analysis. Surgery 130:921 930