Repeat Ultrasound-Guided Fine-Needle Aspiration for Thyroid Nodules 10 mm or Larger Can Be Performed 10.7 Months After Initial Nondiagnostic Results

Neuroradiology/Head and Neck Imaging Original Research Moon et al. Repeat US-Guided FNA of Thyroid Nodules After Nondiagnostic Results Neuroradiology/Head and Neck Imaging Original Research Hee Jung Moon 1 Eun-Kyung Kim Jin Young Kwak Jung Hyun Yoon Moon HJ, Kim EK, Kwak JY, Yoon JH Keywords: fine-needle aspiration, nondiagnostic cytology, thyroid nodules, ultrasound DOI:10.2214/AJR.15.15351 Received July 31, 2015; accepted after revision October 9, 2015. 1 All authors: Department of Radiology and Research Institute of Radiological Science, Severance Hospital, Yonsei University College of Medicine, Yonsei-ro 50, Seodaemun-gu, Seoul 120-752, Republic of Korea. Address correspondence to H. J. Moon (artemis4u@yuhs.ac). AJR 2016; 206:823 828 0361 803X/16/2064 823 American Roentgen Ray Society Repeat Ultrasound-Guided Fine-Needle Aspiration for Thyroid Nodules 10 mm or Larger Can Be Performed 10.7 Months After Initial Nondiagnostic Results OBJECTIVE. The purpose of this study is to investigate when repeat ultrasound (US) guided fine-needle aspiration (FNA) for thyroid nodules 10 mm or larger with initial nondiagnostic results due to inadequate or unsatisfactory specimen could be performed. MATERIALS AND METHODS. A total of 228 nodules 10 mm or larger with initial nondiagnostic results in 223 patients were classified into three groups according to the first follow-up US or US-guided FNA intervals after the initial US-guided FNA: within 3 months (group 1), 3 9 months (group 2), and more than 9 months (group 3). Nodules were classified according to change. The malignancy detection rate and clinicopathologic characteristics were compared among the three groups. RESULTS. Seven nodules (3.1%) were malignant, and their cancer stages were the same as that determined at the initial US-guided FNA. Malignancy detection rates, as well as tumor, extrathyroidal extension, and lymph node metastasis at pathologic analysis, were not significantly different among the three groups. Thirteen of 228 nodules (5.7%) showed increased at a mean (± SD) of 34.5 ± 25.1 months (range, 10.7 84.7 months) after initial US-guided FNA, and one of the 13 nodules (7.7%) was malignant, a minimally invasive follicular carcinoma without lymph node metastasis found at 63.2 months. Of 177 nodules without change, six papillary thyroid carcinomas (3.4%) were found at a mean of 10.4 months. None of the 38 nodules with decreased during a mean follow-up interval of 26.1 ± 19.8 months (range, 2.8 79.5 months) was malignant. CONCLUSION. Repeat US-guided FNA for initial nondiagnostic thyroid nodules after 10.7 months can reduce unnecessary repeat US-guided FNAs without progression of malignancy. U ltrasound (US) guided fine-needle aspiration (FNA) is the mainstay for selecting which thyroid nodules should undergo surgery; thus, the number of unnecessary thyroid surgeries has decreased with its use [1, 2]. Nondiagnostic results due to inadequate or unsatisfactory sampling is one of the limitations of FNA, and repeat FNA is recommended [3 6]. However, the optimal time interval for repeat FNA is not definitely described in current guidelines [3 6]. The only recommendation for the repeat FNA interval is 3 months after a previous FNA, because reactive atypical nuclear changes can affect the repeat FNA results [5, 7, 8]. In cases that are clinically under high suspicion for malignancy, repeat FNA for nondiagnostic nodules can be performed without further waiting at 3 months [6]. However, because of the lack of guidelines for a specific time interval, repeat FNAs for nodules with nondiagnostic findings are performed variably by institution and by the individual choices of the clinician [9]. Repeat FNA is empirically recommended within 6 12 months after the initial FNA and, in many cases, it is usually performed within 6 months [10]. To our knowledge, there has been no study on the optimal time interval for repeat FNA in thyroid nodules with initially nondiagnostic results. In the 2015 guidelines by the American Thyroid Association, US-guided FNA is recommended for nodules 10 mm or larger and is not recommended for nodules smaller than 10 mm, not even for those with a high suspicion for malignancy [6]. Therefore, we investigated when repeat FNA for initial nondiagnostic nodules 10 mm or larger could be performed. Materials and Methods Study Population The institutional review board of Severance Hospital, Yonsei University College of Medicine AJR:206, April 2016 823

Moon et al. approved this retrospective study. Neither patient approval nor informed consent was required for review of medical records. From February 2007 to February 2008, 5678 thyroid nodules underwent US-guided FNA, and 853 of them (15%) had nondiagnostic results due to inadequate or unsatisfactory sampling. Of 853 nodules, 343 were 10 mm or larger and had nondiagnostic results at the first US-guided FNA. Of the 343 nodules, 115 for which follow-up US or US-guided FNA was not performed were excluded. Two hundred twentyeight nodules in 223 patients that were followed up by US or repeat US-guided FNA were included in this study. Five patients had two nodules with initial nondiagnostic results, and 218 had one nodule with initial nondiagnostic results. Of the 223 patients, 188 were female and 35 were male. The mean (± SD) age of the female patients was 52 ± 11.5 years (range, 19 75 years) and the mean age of the male patients was 48.8 ± 15.4 years (range, 17 72 years). The mean of the 228 nodules was 21.6 ± 11.2 mm (range, 10 61 mm). The mean follow-up interval from the initial US-guided FNA to the last follow-up US or surgery was 42.7 ± 30 months (range, 1 94.8 months). Of the 228 nodules, 128 were included in a previous study that evaluated which nodules with nondiagnostic findings should undergo repeat FNA [11]. Ultrasound Examination and Ultrasound- Guided Fine-Needle Aspiration High-resolution US machines equipped with either a 5- or 12-MHz linear probe (iu22, Philips Healthcare) were used for US examinations and US-guided FNAs. US examinations and US-guided FNAs were performed by one of seven boardcertificated radiologists (with 1 13 years of experience) who were dedicated to thyroid imaging. TABLE 1: Clinicopathologic Characteristics of the Seven Malignancies Case No. Group Patient Age (y) TI-RADS Category Size at US (mm) Size Change Interval to Repeat FNA (mo) Findings of Repeat FNA Pathologic Finding Nodule was defined as the maximum diameter at US. US features of thyroid nodules with USguided FNA were prospectively recorded according to internal composition, echogenicity, margin, calcifications, shape, and vascularity [12 14]. Internal composition was classified as solid, solid portion 50% or more, and solid portion less than 50%. Echogenicity was classified as hyper-, iso-, or hypoechogenicity (i.e., when a nodule was hyper-, iso-, or hypoechoic compared with a normal thyroid gland), or marked hypoechogenicity (i.e., when a nodule was hypoechoic compared with the surrounding strap muscle). Echogenicity was classified according to the predominant pattern if the nodule was heterogeneous and according to the solid portion if the nodule had cystic portions. Margin was classified as well-defined or not well defined (i.e., microlobulated or irregular margins). Calcifications were classified as having microcalcifications (i.e., a nodule with both micro- and macrocalcifications was classified as having microcalcifications), macrocalcifications, or no calcifications. Shape was classified as wider than tall or taller than wide (i.e., greater in its anteroposterior dimension than in its transverse dimension). US-guided FNA was performed with the freehand technique with or without an aspirator according to the preference of the performing radiologist. A 23-gauge needle attached to a 2- or 20-mL disposable plastic syringe was used [12 14]. The needle was passed through until aspirated materials were seen in the needle hub. In nodules with cystic portions, US-guided FNAs were performed for the solid portions of the nodules after cystic fluid aspiration. Samples obtained in this manner were expelled onto two different glass slides, which were immediately placed in 95% alcohol for Papanicolaou staining. Afterward, with the use of a new 23-gauge needle and a 2- or 20- ml syringe, aspiration was repeated to make another two slides. After slide smearing, the remaining aspirated materials were rinsed with saline and processed for cell blocking. Cytopathologists were not on site during US-guided FNA. The criteria for a nondiagnostic result were defined as fewer than six groups of cells containing more than 10 cells [3 5]. Data and Statistical Analysis Of 228 nodules, 16 underwent surgery, 140 underwent repeat US-guided FNA, and 72 underwent follow-up US. Thyroid nodules with malignancy confirmed at surgery or repeat US-guided FNA were classified as malignant. Thyroid nodules with benign results at surgery or repeat FNA, nodules without malignant cytologic results at repeat USguided FNA and with no change, or nodules with decreased at follow-up US were considered benign, because the denominator in malignancy risk calculations is the total number of nodules, and the numerator is the number of confirmed malignancy cases according to the Bethesda system [3]. Size change was recorded after reviewing the original radiologic reports and US images. The follow-up interval was defined as the interval from the initial US-guided FNA to the last followup US or surgery. Initial US features were reclassified according to the Thyroid Imaging Reporting and Data System (TIRADS) [15]. In TIRADS, suspicious malignant US features included solidity, hypoechogenicity or marked hypoechogenicity, a margin that is not well-defined, microcalcifications, and a taller-than-wide shape [15]. Thyroid nodules without any suspicious US features were Pathologic Size (mm) Extrathyroidal Extension Lymph Node Metastasis 1 1 61 4c 15 No 1.0 PTC PTC 12 Minimal 0 2 1 33 3 25 No 1.8 Suspicious for PTC 18 No 1a PTC 3 2 47 4b 10 No 4.0 PTC PTC 7 No 0 4 2 59 4c 10 No 3.0 PTC PTC 9 Minimal 1a 5 2 35 4b 14 No 3.2 Nondiagnostic PTC 12 Minimal 0 6 3 58 4b 39 5 mm 63.2 Suspicious for Minimally invasive 38 No 0 follicular neoplasm follicular carcinoma 7 3 50 4c 25 No 33.3 Atypia Follicular variant PTC 19 No 1b a Note Groups 1, 2, and 3 included nondiagnostic nodules with the first follow-up occurring within 3 months, 3 9 months, and more than 9 months after the initial ultrasound (US) guided fine-needle aspiration (FNA), respectively. TI-RADS = Thyroid Imaging Reporting and Data System, PTC = papillary thyroid carcinoma. a Initial US-guided FNAs were performed for a thyroid nodule and a lateral lymph node, and both results were nondiagnostic. The thyroid nodule and lymph node showed no change during follow-up, and repeat FNAs were performed for both at 33.3 months. 824 AJR:206, April 2016

Repeat US-Guided FNA of Thyroid Nodules After Nondiagnostic Results classified as TIRADS category 3 [15]. Thyroid nodules with one, two, three or four, or five suspicious features were classified as category 4a, 4b, 4c, or 5, respectively [15]. Thyroid nodules were classified into three groups according to the time interval between the initial US-guided FNA and the first follow-up USguided FNA. Thyroid nodules with the first followup US or repeat US-guided FNA occurring within 3 months, 3 9 months, or more than 9 months after the initial US-guided FNA were classified into groups 1, 2, and 3, respectively. Malignancy detection rates of the three groups were compared. Malignancy risks for all nodules, nodules with increased, nodules without a change, nodules with decreased, and each TIRADS category were calculated as percentages. Clinicopathologic characteristics of the three groups were compared. Clinicopathologic characteristics according to the change were compared. Categoric variables were compared using the chi-square test or Fisher exact test, and continuous variables were compared using the independent t test or Kruskal-Wallis test. Statistical analysis was performed using SPSS statistical software (version 20.0, IBM). A twosided p < 0.05 was considered to indicate statistical significance. Results Of 228 thyroid nodules, seven (3.1%) were malignant and 221 (96.9%) were benign. Of seven malignancies, five were classic papillary carcinomas, one was follicular variant papillary carcinoma, and one was minimally invasive follicular carcinoma (Table 1). In the case of the minimally invasive follicular carcinoma that increased in by 5 mm after 63.2 months, there were no extrathyroidal extension and lymph node metastasis. Six papillary carcinomas without change were smaller than 20 mm at pathologic analysis. Three malignancies had minimal extrathyroidal extension. In the case of lateral lymph node metastasis, initial US-guided FNAs were performed for the thyroid nodule and the lateral lymph node, and both results were nondiagnostic. During the 33.3 months of follow-up, the thyroid nodule and lymph node showed no change. Seven malignancies had no change in cancer stage during the follow-up period, and the cancer stage was identical to the one assessed at the initial US-guided FNA. Of 228 nodules, 29, 71, and 128 were classified into groups 1, 2, and 3, respectively. Patient age and sex and lesion at US were not significantly different among the three groups (Table 2). The malignancy detection rates of the three groups were 6.9% (2/29), 4.2% (3/71), and 1.6% (2/128), which were not significantly different (p = 0.181). Nodules with TIRADS category 4c frequently underwent repeat FNA within 3 months, and nodules with category 3 underwent repeat FNA after 9 months (p = 0.012). TIRADS categories of the seven malignancies were not significantly different among the three groups (p = 0.829). Tumor (p = 0.084), extrathyroidal extension (p = 0.657), and lymph node metastasis at pathologic analysis (p > 0.999) were not significantly different. Of 228 nodules, 13 showed increased at a mean of 34.5 ± 25.1 months (range, 10.7 84.7 months). One nodule was malignant and the overall malignancy risk was 7.7%. Increased was not associated with malignancy (p = 0.396). One malignancy with increased was a minimally invasive follicular carcinoma without extrathyroidal extension and lymph node metastasis. Eleven nodules were benign at repeat US-guided FNA, and one was benign at surgery. One benign nodule in group 1 and two benign nodules in group 2 showed increased at 15.8, 42.9, and 85.1 months, respectively, after initial US-guided FNA (Fig. 1). The mean age of patients who had nodules with increased TABLE 2: Comparison of Clinicopathologic Characteristics of Three Groups of Nodules was 42.7 ± 12.5 years, which was younger than the 52.2 ± 12 years of patients who had nodules without change (p = 0.012). Patient sex (p = 0.412) and nodule (p = 0.278), composition (p = 0.299), echogenicity (p = 0.507), margin (p > 0.999), calcifications (p = 0.200), shape (p > 0.999), and TIRADS categories (p = 0.731) were not significantly different between nodules without change and those with increased. Of 228 nodules, 177 (77.6%) showed no change during a mean follow-up of 42 ± 30.7 months (range, 1 94.8 months). Six of 177 (3.4%) malignancies were found at a mean of 10.4 ± 12.6 months (range, 3 35.9 months). All six malignancies were papillary carcinomas and smaller than 20 mm. The malignancy risks of 177 nodules with TIRADS categories 3, 4a, 4b, 4c, and 5 were 1.7% (1/58), 0% (0/56), 3.3% (2/61), 21.4% (3/14), and 0% (0/0), respectively. Thirty-eight of 228 nodules (16.7%) showed decreased during a mean of 26.1 ± 19.8 months (range, 2.8 79.5 months). None of the 38 nodules was malignant. Initial nodule was larger in nodules with decreased (p = 0.033). Nodules with a solid portion less than 50% (p < 0.001) and nodules without calcifications (p = 0.003) were significantly associated with decreased. TIRADS categories were significantly different between nodules without change and nodules with decreased (p = 0.032). Patient sex (p = 0.408) and nodule echogenicity (p = 0.499), margin (p = 0.473), and shape (p > 0.999) were not significantly different between nodules without change and nodules with decreased. Discussion The malignancy rate was 3.1% in 228 nondiagnostic nodules 10 mm or larger, and the malignancy detection rates and cancer stages were not significantly different among the three Characteristic Group 1 (n = 29) Group 2 (n = 71) Group 3 (n = 128) p Patient characteristics Age (y), mean ± SD 50 ± 12.1 50.7 ± 12.6 52.5 ± 12.2 0.475 Sex 0.639 Female (n = 192) 26 (89.7) 58 (81.7) 108 (84.4) Male (n = 36) 3 (10.3) 13 (18.3) 20 (15.6) Final outcome 0.181 Benign (n = 221) 27 (93.1) 68 (95.8) 126 (98.4) Malignant (n = 7) 2 (6.9) 3 (4.2) 2 (1.6) Follow-up interval (mo), mean ± SD 30.7 ± 35.4 40.3 ± 30.4 46.3 ± 27.6 0.030 (Table 2 continues on next page) AJR:206, April 2016 825

Moon et al. TABLE 2: Comparison of Clinicopathologic Characteristics of Three Groups of Nodules (continued) Characteristic Group 1 (n = 29) Group 2 (n = 71) Group 3 (n = 128) p Ultrasound features for all nodules Size (mm) 21.1 ± 11.2 22.3 ± 11.5 21.4 ± 11.2 0.840 Composition 0.159 Solid (n = 122) 20 (69.0) 40 (56.3) 62 (48.4) Solid 50% (n = 63) 7 (24.1) 16 (22.5) 40 (31.2) Solid < 50% (n = 43) 2 (6.9) 15 (21.1) 26 (20.3) Echogenicity 0.067 Hyper- or isoechogenic (n = 126) 13 (44.8) 36 (50.7) 77 (60.2) Hypoechogenic (n = 98) 15 (51.7) 32 (45.1) 51 (39.8) Marked hypoechogenicity (n = 4) 1 (3.4) 3 (4.2) 0 (0) Margin 0.019 Well-defined (n = 215) 25 (86.2) 65 (91.5) 125 (97.7) Not well-defined (n = 13) 4 (13.8) 6 (8.5) 3 (2.3) Calcifications 0.061 Microcalcifications (n = 14) 2 (6.9) 7 (9.9) 5 (3.9) Macrocalcifications (n = 35) 7 (24.1) 14 (19.7) 14 (10.9) None (n = 179) 20 (69.0) 50 (70.4) 109 (85.2) Shape 0.059 Wider than tall (n = 222) 27 (93.1) 68 (95.8) 127 (99.2) Taller than wide (n = 6) 2 (6.9) 3 (4.2) 1 (0.8) TI-RADS category 0.012 3 (n = 79) 7 (24.1) 22 (31.0) 50 (39.1) 4a (n = 66) 8 (27.6) 18 (25.4) 40 (31.2) 4b (n = 67) 8 (27.6) 24 (33.8) 35 (27.3) 4c (n = 15) 6 (20.7) 7 (9.9) 2 (1.6) 5 (n = 1) 0 (0.0) 0 (0.0) 1 (0.8) Ultrasound features for malignant nodules only (n = 7) a TI-RADS category 0.829 3 (n = 1) 1 (50.0) 0 (0.0) 0 (0.0) 4b (n = 3) 0 (0.0) 2 (66.7) 1 (50.0) 4c (n = 3) 1 (50.0) 1 (33.3) 1 (50.0) Size (mm), median (range) 15 (10 18) 9 (7 12) 28.5 (19 38) 0.084 Extrathyroidal extension 0.657 Yes (n = 3) 1 (50.0) 2 (66.7) 0 (0.0) No (n = 4) 1 (50.0) 1 (33.3) 2 (100.0) Lymph node metastasis > 0.999 0 (n = 3) 1 (50.0) 2 (66.7) 0 (0.0) 1a (n = 2) 1 (50.0) 1 (33.3) 0 (0.0) 1b (n = 1) 0 (0.0) 0 (0.0) 1 (100) b Note Except where noted otherwise, data are number (%) of nodules or patients. Not all percentages total 100 because of rounding. Groups 1, 2, and 3 included nondiagnostic nodules with the first follow-up occurring within 3 months, at 3 9 months, and more than 9 months after the initial ultrasound-guided fine-needle aspiration (FNA). TI-RADS = Thyroid Imaging Reporting and Data System. a There were two malignant nodules in group 1, three in group 2, and two in group 3. b A nondiagnostic result was obtained at initial FNA for a suspicious metastatic lymph node. The lymph node showed no change during follow-up and repeat FNAs were performed for the thyroid nodule and lymph node at 33.3 months. 826 AJR:206, April 2016

Repeat US-Guided FNA of Thyroid Nodules After Nondiagnostic Results groups. Nodules showed an increase in at a minimum of 10.7 months after initial nondiagnostic results. None of the nodules with decreased was malignant, and decreased was observed at a mean of 26.1 months. Thus, repeat US-guided FNA after 10.7 months can reduce unnecessary repeat US-guided FNAs without progression of malignancy. The malignancy detection rate, pathologic tumor, extrathyroidal extension, and lymph node metastasis were not significantly different among the three groups. All extrathyroidal extension was minimal, which has been known to not be significantly associated with recurrence [16]. Only one malignancy increased in by 5 mm after 63.2 months and was found to be minimally invasive follicular carcinoma without extrathyroidal extension and lymph node metastasis. Six malignancies without change were papillary thyroid carcinomas smaller than 20 mm, which were known as indolent cancer [17]. 29 in group 1 228 Nodules 71 in group 2 128 in group 3 Repeat FNA for nondiagnostic nodules within 3 or 6 months of initial FNA is rationalized by the early detection of malignancy without progression. If repeat US-guided FNA can be performed before malignancy progresses, and if nodules can be selected for repeat FNA at that time, the number of unnecessary repeat US-guided FNAs can be reduced. Our results suggest that a follow-up interval of at least 10.7 months is not associated with malignancy progression because tumor and the frequencies of extrathyroidal extension and lymph node metastasis were not significantly different among the three groups, the increase was found after 10.7 months, and the cancer stages were the same as those observed at initial US-guided FNA. Recommending repeat US-guided FNA 10.7 months after initial FNA can reduce the number of unnecessary repeat US-guided FNAs while still maintaining the same stage of malignancy without progression. 28 No change 1 Decreased 67 No change 4 Decreased 10 Increased 27 Decreased The malignancy risk of 228 nondiagnostic nodules was 3.1%, which is comparable with rates found in previous studies and is within the range recommended by guidelines [3, 13, 18, 19]. Some authors have tried to triage nodules with nondiagnostic cytology and to select thyroid nodules for repeat FNA or clinical follow-up [11, 13]. A recent study has suggested that all nodules with nondiagnostic cytology can be followed up clinically without repeat FNA, regardless of US features, because the malignancy risk is 0.6% [19]. Our 3.1% malignancy risk was low, but slightly higher than the upper limit of 3% for clinical follow-up [3]. For nodules with nondiagnostic cytology with increased, the malignancy risk was 7.7%. Benign nodules can increase in during follow-up; thus, increased is not associated with malignancy [4, 20]. Increased, however, can be an indication for FNA because of the 7.7% malignancy risk [4]. Repeat FNA for nodules 2 Malignancy 1 Increased at 15.8 mo 1 Decreased at 56.8 mo 3 Malignancy 2 Increased at 42.9 and 85.1 mo 5 Decreased at 20.1, 40.6, 47.8, 69.8, and 70.1 mo 91 No change 1 Malignancy 1 Malignancy Fig. 1 Size changes and detected malignancies of nodules in groups 1, 2, and 3. Nodules were classified into groups 1, 2, and 3 according to first follow-up within 3 months, 3 9 months, and more than 9 months after initial nondiagnostic results. In group 1, one nodule showed decrease in at first follow-up within 3 months after initial nondiagnostic results, and 28 nodules showed no change. Two nodules without change were diagnosed as malignancy at first follow-up. Afterward, one nodule showed increased at 15.8 months and one showed decreased at 56.8 months. In group 2, four nodules showed decreased at first follow-up from 3 to 9 months and 67 showed no change. Three nodules were diagnosed as malignancy at first follow-up. After first follow-up, two nodules showed increased and five showed decreased. In group 3, 10 nodules showed increased at first follow-up after 9 months, 27 showed decreased, and 91 showed no change. One malignancy with increased and one malignancy without change were diagnosed at first follow-up. AJR:206, April 2016 827

Moon et al. with increased should be recommended regardless of US features. Thirteen nodules showed increased at least 10.7 months after initial US-guided FNA. Thus, repeat US-guided FNA after 10.7 months can detect malignancy without progression. Of 177 nodules without change, six were papillary thyroid carcinomas, which were found at a mean of 10.4 months. The overall malignancy risk of 177 nodules was 3.4%. The malignancy risks of nodules with nondiagnostic findings classified as TIRADS categories 3 and 4a at initial US were 1.7% and 0%, respectively; thus, clinical follow-up can be recommended [3]. The malignancy risks of nodules classified as categories 4b and 4c were 3.3% and 21.4%, respectively; thus, repeat FNA should be performed. Repeat FNA after 10.7 months can detect malignancies without progression. If nondiagnostic nodules classified as TIRADS categories 3 and 4a show no change, clinical follow-up can be recommended. If nondiagnostic nodules classified as TIRADS categories 4b or more show no change at follow-up US, repeat USguided FNA should be recommended. Thirty-eight nodules showed decreased at a mean follow-up of 26.1 months with a range of 2.8 79.5 months. None of the 38 nodules with nondiagnostic cytology that decreased in was malignant. Considering the zero malignancy rate and the wide range of intervals to decreased and considering that the nodules with increased only showed increased after a minimum of 10.7 months, a follow-up after 10.7 months can be recommended. If nodules with nondiagnostic cytology show decreased, repeat US-guided FNA is unnecessary. Repeat FNAs were frequently performed within 3 months for nodules with nondiagnostic cytology classified as TIRADS category 4c, whereas for nodules classified as TI- RADS category 3, repeat US-guided FNAs were performed after 9 months. Nodules with nondiagnostic findings with more suspicious features underwent repeat FNA within 3 months, consistent with the current guidelines [6]. The TIRADS categories of the seven malignancies, however, were not significantly different among the three groups. Considering that there were no significant differences in malignancy detection rates and that the cancer stages were the same as those observed at initial US-guided FNA, the first follow-up can be performed after 10.7 months regardless of initial US features. There are some limitations to our study. First, our study was retrospective. One hundred thirty-six nodules without any followup were excluded, and repeat FNAs were performed at variable time intervals according to the individual choices of the clinicians. Thus, a bias was present. Second, thyroid nodules without cytologic, histologic, or clinical evidence of malignancy were considered not malignant (i.e., benign). Malignancy risks might be underestimated. However, all lesions with increased were confirmed through surgery and repeat FNA in our study. The recommended malignancy risk for nondiagnostic nodules in the Bethesda system is 1 4%, in which the numerator is the number of confirmed malignancies and the denominator is the total number of nodules [3]. Third, 228 nodules in 223 patients were included. All analyses were performed on a per-nodule basis, not a per-patient basis. A generalized estimating equation analysis was not performed because there were data cells with a value of zero. In conclusion, repeat US-guided FNA for thyroid nodules with initial nondiagnostic cytology after 10.7 months can reduce unnecessary repeat US-guided FNAs without progression of malignancy. References 1. Richards ML, Bohnenblust E, Sirinek K, Bingener J. Nondiagnostic thyroid fine-needle aspiration biopsies are no longer a dilemma. Am J Surg 2008; 196:398 402 2. Yang J, Schnadig V, Logrono R, Wasserman PG. Fine-needle aspiration of thyroid nodules: a study of 4703 patients with histologic and clinical correlations. Cancer 2007; 111:306 315 3. Cibas ES, Ali SZ; NCI Thyroid FNA State of the Science Conference. The Bethesda system for reporting thyroid cytopathology. Am J Clin Pathol 2009; 132:658 665 4. Cooper DS, Doherty GM, Haugen BR, et al.; American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2009; 19:1167 1214 5. Gharib H, Papini E, Paschke R, et al. American Association of Clinical Endocrinologists, Associazione Medici Endocrinologi, and European Thyroid Association medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. Endocr Pract 2010; 16(suppl 1):1 43 6. Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2015 Oct 14 [Epub ahead of print] 7. LiVolsi VA, Merino MJ. Worrisome histologic alterations following fine-needle aspiration of the thyroid (WHAFFT). Pathol Annu 1994; 29:99 120 8. Recavarren RA, Houser PM, Yang J. Potential pitfalls of needle tract effects on repeat thyroid fine-needle aspiration. Cancer Cytopathol 2013; 121:155 161 9. Brito JP, Castro MR, Dean DS, Fatourechi V, Stan M. Survey of current approaches to non-diagnostic fine-needle aspiration from solid thyroid nodules. Endocrine 2015; 49:745 751 10. Jaume JC, Chen H. Inadequate cytology of thyroid nodules: repeat it or live with it. Indian J Surg Oncol 2011; 2:76 77 11. Yoon JH, Moon HJ, Kim EK, Kwak JY. Inadequate cytology in thyroid nodules: should we repeat aspiration or follow-up? Ann Surg Oncol 2011; 18:1282 1289 12. Kim EK, Park CS, Chung WY, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR 2002; 178:687 691 13. Moon HJ, Kim EK, Yoon JH, Kwak JY. Malignancy risk stratification in thyroid nodules with nondiagnostic results at cytologic examination: combination of thyroid imaging reporting and data system and the Bethesda system. Radiology 2015; 274:287 295 14. Moon HJ, Kwak JY, Kim MJ, Son EJ, Kim EK. Can vascularity at power Doppler US help predict thyroid malignancy? Radiology 2010; 255:260 269 15. Kwak JY, Han KH, Yoon JH, et al. Thyroid imaging reporting and data system for US features of nodules: a step in establishing better stratification of cancer risk. Radiology 2011; 260:892 899 16. Moon HJ, Kim EK, Chung WY, Yoon JH, Kwak JY. Minimal extrathyroidal extension in patients with papillary thyroid microcarcinoma: is it a real prognostic factor? Ann Surg Oncol 2011; 18:1916 1923 17. Davies L, Welch HG. Thyroid cancer survival in the United States: observational data from 1973 to 2005. Arch Otolaryngol Head Neck Surg 2010; 136:440 444 18. Al Maqbali T, Tedla M, Weickert MO, Mehanna H. Malignancy risk analysis in patients with inadequate fine needle aspiration cytology (FNAC) of the thyroid. PLoS One 2012; 7:e49078 19. Anderson TJ, Atalay MK, Grand DJ, Baird GL, Cronan JJ, Beland MD. Management of nodules with initially nondiagnostic results of thyroid fineneedle aspiration: can we avoid repeat biopsy? Radiology 2014; 272:777 784 20. Kim SY, Han KH, Moon HJ, Kwak JY, Chung WY, Kim EK. Thyroid nodules with benign findings at cytologic examination: results of long-term follow-up with US. Radiology 2014; 271:272 281 828 AJR:206, April 2016