ORIGINAL ARTICLE
Evaluation of serum thyroid-stimulating hormone as indicator for fine-needle aspiration in patients with thyroid nodules Ji Soo Choi, MD, PhD,1,2 Chung Mo Nam, PhD,3 Eun-Kyung Kim, MD, PhD,1 Hee Jung Moon, MD, PhD,1 Kyung Hwa Han, MS,4 Jin Young Kwak, MD, PhD1* 1
Department of Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, Seoul, Korea, 2Department of Radiology, Samsung Medical Center, Seoul, Korea, 3Department of Biostatistics, Yonsei University College of Medicine, Seoul, Korea, 4Biostatistics Collaboration Unit, Medical Research Center, Yonsei University College of Medicine, Seoul, Korea.
Accepted 10 January 2014 Published online 9 April 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/hed.23616
ABSTRACT: Background. Recently, it has been reported that the risk of thyroid malignancy increases with increasing concentrations of serum thyroid-stimulating hormone (TSH). The purpose of this study was to determine whether or not serum TSH can be a predictor for thyroid malignancy when considering the relevant ultrasound features and clinical risk factors. Methods. This retrospective study included 1200 euthyroid patients with 1269 thyroid nodules who underwent ultrasound-guided fine-needle aspiration (FNA) biopsy between January and June 2009. Serum TSH, ultrasound feature, and clinical parameters were compared according to final diagnosis. Subgroup analyses were performed according to nodule size.
Results. Serum TSH did not show a positive association with malignancy for all nodules and the micronodule subgroup in multivariate analysis, although they showed significant association with thyroid malignancy for the macronodule subgroup. For all nodules and the 2 subgroups, suspicious ultrasound features and younger age were significantly associated with malignancy in univariate and multivariate analyses. Conclusion. Our study suggests that TSH alone is not as useful as ultrasound features in deciding whether or not to perform FNA in patients with microC 2014 Wiley Periodicals, Inc. Head Neck 37: 498–504, 2015 nodules. V
INTRODUCTION
ultrasound features, several guidelines apply clinical factors, such as a family history of thyroid cancer, radiation exposure, and nodule size, when considering FNA.4,6,7 Recently, it has been reported that the risk of thyroid malignancy increases with increasing concentrations of serum thyroid-stimulating hormone (TSH) in patients with nodular thyroid disease.14–21 Higher TSH values, even within normal ranges, have also shown association with a higher likelihood of and a more advanced stage of thyroid cancer.18,21,22 However, it is unclear whether the addition of TSH information to ultrasound features and other known predictors of malignancy can be helpful in decisions to perform FNA in patients with thyroid nodules. The purpose of our study was to determine whether or not serum TSH can be an indicator for performing FNA when considering the relevant ultrasound features and clinical risk factors.
Thyroid nodules are common clinical problems, which are increasingly detected by the widespread use of diagnostic imaging techniques.1 Epidemiologic studies have reported the estimated prevalence of nodular thyroid diseases to range from 4% to 7% with palpation and 19% to 67% with ultrasound.2,3 Thyroid malignancy, however, is less common and approximately 5% to 10% of detected nodules are malignant.4,5 Therefore, there is a need to establish guidelines to determine which thyroid nodules should undergo fine-needle aspiration (FNA). Thyroid ultrasound has been widely used not only to detect thyroid nodules but also to select the nodules that necessitate FNA, which has been accepted so far as the gold standard method in the diagnosis of thyroid nodules. Many reports have shown that malignant ultrasound features are useful in determining which nodules should undergo FNA.4,6–10 Malignant ultrasound features that predict a thyroid malignancy are hypoechogenicity or marked hypoechogenicity, irregular margins, taller than wide shape, and microcalcifications.4,8,10–13 In addition to
KEY WORDS: thyroid-stimulating hormone (TSH), ultrasound, thyroid nodule, thyroid malignancy, diagnosis
MATERIALS AND METHODS This retrospective study was approved by the institutional review board, and informed consent was waived. Informed consent for the ultrasound-guided FNA was obtained from all patients before each biopsy.
Patients *Corresponding author: J. Y. Kwak, Department of Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, 50 Yonseiro, Seodaemun-gu, Seoul 120-752, Korea. E-mail: [email protected]
498
HEAD & NECK—DOI 10.1002/HED
APRIL 2015
From January to June 2009, 1709 patients underwent ultrasound-guided FNA of thyroid nodules detected on
SERUM TSH
FOR FINE–NEEDLE ASPIRATION DECISION–MAKING
FIGURE 1. Diagram of the study population. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary. com.]
ultrasound at our institution (a referral center). At our hospital, scintigraphy is usually performed in patients whose serum TSH may suggest the presence of autonomous nodules. FNA is not recommended for autonomous thyroid nodules showing hot uptake on scintigraphy. Among the 1709 patients who underwent ultrasound-guided FNA, 1200 patients (1181 women and 194 men) with 1269 thyroid nodules were included in the study who fulfilled the following criteria: (1) patients with records available of serum TSH levels within 3 months before ultrasoundguided FNA; (2) patients positively diagnosed either in cytopathology or histopathology, (patients with unsatisfac-
tory [n 5 236], atypia [n 5 7], follicular neoplasm [n 5 14], or suspicious for malignancy [n 5 50] cytologic results and who did not undergo further evaluation, such as follow-up FNA or thyroid surgery, were excluded); (3) patients with malignant thyroid nodules diagnosed as differentiated thyroid carcinoma, (patients with a final diagnosis other than differentiated thyroid carcinoma [eg, medullary carcinoma or anaplastic carcinoma] were excluded); (4) euthyroid patients with no drug history of L-T4 or methimazole; and (5) patients with no history of known thyroid cancer or previous thyroid surgery (Figure 1). Of the 1200 patients, 1132 had 1 nodule, 67 had 2 HEAD & NECK—DOI 10.1002/HED
APRIL 2015
499
CHOI ET AL.
FIGURE 2. Diagnostic algorithm of 1269 thyroid nodules in 1200 patients. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary. com.]
nodules, and 1 patient had 3 nodules. Patients ranged in age from 20 to 79 years (mean, 48.6 years).
Ultrasound and ultrasound-guided fine-needle aspiration Real-time thyroid ultrasound was performed by 7 radiologists (4 faculties with 5–13 years of experience and 3 fellows with 1–2 years of experience). Ultrasound scans of the thyroid glands were obtained with a 5 to 12 MHz linear probe (iU22; Philips Medical Systems, Bothell, WA). Interpretation of ultrasound features of all thyroid nodules were prospectively recorded according to internal component, echogenicity, margins, calcifications, shape, and final assessment by the radiologists who had performed the ultrasound. Ultrasound features were classified according to previously published criteria.8 Suspicious malignant ultrasound features were defined as the following: marked hypoechogenicity (decreased echogenicity compared to the surrounding muscle), microlobulated or irregular margins, microcalcifications, and taller-than-wide shape. If a nodule showed 1 or more of these suspicious features on ultrasound, it was classified as suspiciously malignant. If a nodule showed no suspicious features, it was classified as probably benign. Nodule size was also measured in 3 dimensions and defined with the greatest diameter. After thyroid ultrasound, ultrasound-guided FNA was performed by the same radiologist who had performed the ultrasound. At our institution, ultrasound-guided FNA is performed in either the thyroid nodule with suspicious ultrasound features or the largest thyroid nodule if no suspicious features are detected. According to the American Thyroid Association guidelines, FNA for a thyroid nodule is not routinely recommended for nodules 5 mm at our institution unless a patient has specific risk factors (eg, family history of thyroid cancer or radiation exposure). However, we sometimes perform FNAs of very smallsized nodules at the request of the patient. Ultrasound500
HEAD & NECK—DOI 10.1002/HED
APRIL 2015
guided FNA was performed with a 23-gauge needle attached to a 2-mL disposable plastic syringe and each lesion was aspirated at least twice. Aspirated materials were expelled onto glass slides and smeared. All smears were immediately placed in 95% alcohol for Papanicolaou staining. The remaining aspirated materials were rinsed with saline for cell block processing. Additional special staining was done according to the cytopathologists’ needs on a case-by-case basis.
Laboratory test Serum TSH levels were measured by IRMA (TSHCTK-3, SORIN Biomedica, Saluggia, Italy) in every patient. The normal reference range for serum TSH was between 0.3 and 4.99 mIU/L. Among the 1709 patients who underwent ultrasound-guided FNA during the study period, 26 patients were excluded because they did not have records available of serum TSH levels within 3 months before ultrasound-guided FNA and 175 patients whose serum TSH levels were not within normal range were also excluded (Figure 1). Therefore, 1200 euthyroid patients were finally included in this study.
Histopathologic results Final diagnosis was based on cytologic or histopathologic results. Malignant nodules were confirmed with surgery (n 5 304) or FNA (n 5 38). Benign nodules were also confirmed with surgery (n 5 2) or FNA (n 5 925; Figure 2).
Data and statistical analysis The age and nodule size of the benign and malignant groups were compared using an independent 2-sample t test. Sex and ultrasound features of the 2 groups were compared using the chi-square test. The TSH levels were graded into 3 categories for statistical analysis (0.3–1.39; 1.4–2.49; and 2.5–4.99 mIU/L). The serum TSH grades
SERUM TSH
TABLE 1. Baseline characteristics of 1269 thyroid nodules in 1200 patients. Benign
Malignant
p value
All No. of patients* No. of nodules Mean age, y Sex (%) Male Female Mean nodule size, mm TSH value Mean TSH level, mIU/L TSH grade (%) 0.3–1.39 1.4–2.49 2.5–4.99 Ultrasound feature (%) Probably benign Suspiciously malignant Nodule size 1 cm No. of patients† No. of nodules Mean age Sex (%) Male Female Mean nodule size TSH value Mean TSH level TSH grade (%) 0.3–1.39 1.4–2.49 2.5–4.99 Ultrasound feature Probably benign Suspiciously malignant Nodule size >1 cm No. of patients† No. of nodules Mean age Sex (%) Male Female Mean nodule size TSH value Mean TSH level TSH grade (%) 0.3–1.39 1.4–2.49 2.5–4.99 Ultrasound feature (%) Probably benign Suspiciously malignant
901 312 927 342 50.9 6 11.6 47.6 6 11.7
Evaluation of serum thyroid-stimulating hormone as indicator for fine-needle aspiration in patients with thyroid nodules Ji Soo Choi, MD, PhD,1,2 Chung Mo Nam, PhD,3 Eun-Kyung Kim, MD, PhD,1 Hee Jung Moon, MD, PhD,1 Kyung Hwa Han, MS,4 Jin Young Kwak, MD, PhD1* 1
Department of Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, Seoul, Korea, 2Department of Radiology, Samsung Medical Center, Seoul, Korea, 3Department of Biostatistics, Yonsei University College of Medicine, Seoul, Korea, 4Biostatistics Collaboration Unit, Medical Research Center, Yonsei University College of Medicine, Seoul, Korea.
Accepted 10 January 2014 Published online 9 April 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/hed.23616
ABSTRACT: Background. Recently, it has been reported that the risk of thyroid malignancy increases with increasing concentrations of serum thyroid-stimulating hormone (TSH). The purpose of this study was to determine whether or not serum TSH can be a predictor for thyroid malignancy when considering the relevant ultrasound features and clinical risk factors. Methods. This retrospective study included 1200 euthyroid patients with 1269 thyroid nodules who underwent ultrasound-guided fine-needle aspiration (FNA) biopsy between January and June 2009. Serum TSH, ultrasound feature, and clinical parameters were compared according to final diagnosis. Subgroup analyses were performed according to nodule size.
Results. Serum TSH did not show a positive association with malignancy for all nodules and the micronodule subgroup in multivariate analysis, although they showed significant association with thyroid malignancy for the macronodule subgroup. For all nodules and the 2 subgroups, suspicious ultrasound features and younger age were significantly associated with malignancy in univariate and multivariate analyses. Conclusion. Our study suggests that TSH alone is not as useful as ultrasound features in deciding whether or not to perform FNA in patients with microC 2014 Wiley Periodicals, Inc. Head Neck 37: 498–504, 2015 nodules. V
INTRODUCTION
ultrasound features, several guidelines apply clinical factors, such as a family history of thyroid cancer, radiation exposure, and nodule size, when considering FNA.4,6,7 Recently, it has been reported that the risk of thyroid malignancy increases with increasing concentrations of serum thyroid-stimulating hormone (TSH) in patients with nodular thyroid disease.14–21 Higher TSH values, even within normal ranges, have also shown association with a higher likelihood of and a more advanced stage of thyroid cancer.18,21,22 However, it is unclear whether the addition of TSH information to ultrasound features and other known predictors of malignancy can be helpful in decisions to perform FNA in patients with thyroid nodules. The purpose of our study was to determine whether or not serum TSH can be an indicator for performing FNA when considering the relevant ultrasound features and clinical risk factors.
Thyroid nodules are common clinical problems, which are increasingly detected by the widespread use of diagnostic imaging techniques.1 Epidemiologic studies have reported the estimated prevalence of nodular thyroid diseases to range from 4% to 7% with palpation and 19% to 67% with ultrasound.2,3 Thyroid malignancy, however, is less common and approximately 5% to 10% of detected nodules are malignant.4,5 Therefore, there is a need to establish guidelines to determine which thyroid nodules should undergo fine-needle aspiration (FNA). Thyroid ultrasound has been widely used not only to detect thyroid nodules but also to select the nodules that necessitate FNA, which has been accepted so far as the gold standard method in the diagnosis of thyroid nodules. Many reports have shown that malignant ultrasound features are useful in determining which nodules should undergo FNA.4,6–10 Malignant ultrasound features that predict a thyroid malignancy are hypoechogenicity or marked hypoechogenicity, irregular margins, taller than wide shape, and microcalcifications.4,8,10–13 In addition to
KEY WORDS: thyroid-stimulating hormone (TSH), ultrasound, thyroid nodule, thyroid malignancy, diagnosis
MATERIALS AND METHODS This retrospective study was approved by the institutional review board, and informed consent was waived. Informed consent for the ultrasound-guided FNA was obtained from all patients before each biopsy.
Patients *Corresponding author: J. Y. Kwak, Department of Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, 50 Yonseiro, Seodaemun-gu, Seoul 120-752, Korea. E-mail: [email protected]
498
HEAD & NECK—DOI 10.1002/HED
APRIL 2015
From January to June 2009, 1709 patients underwent ultrasound-guided FNA of thyroid nodules detected on
SERUM TSH
FOR FINE–NEEDLE ASPIRATION DECISION–MAKING
FIGURE 1. Diagram of the study population. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary. com.]
ultrasound at our institution (a referral center). At our hospital, scintigraphy is usually performed in patients whose serum TSH may suggest the presence of autonomous nodules. FNA is not recommended for autonomous thyroid nodules showing hot uptake on scintigraphy. Among the 1709 patients who underwent ultrasound-guided FNA, 1200 patients (1181 women and 194 men) with 1269 thyroid nodules were included in the study who fulfilled the following criteria: (1) patients with records available of serum TSH levels within 3 months before ultrasoundguided FNA; (2) patients positively diagnosed either in cytopathology or histopathology, (patients with unsatisfac-
tory [n 5 236], atypia [n 5 7], follicular neoplasm [n 5 14], or suspicious for malignancy [n 5 50] cytologic results and who did not undergo further evaluation, such as follow-up FNA or thyroid surgery, were excluded); (3) patients with malignant thyroid nodules diagnosed as differentiated thyroid carcinoma, (patients with a final diagnosis other than differentiated thyroid carcinoma [eg, medullary carcinoma or anaplastic carcinoma] were excluded); (4) euthyroid patients with no drug history of L-T4 or methimazole; and (5) patients with no history of known thyroid cancer or previous thyroid surgery (Figure 1). Of the 1200 patients, 1132 had 1 nodule, 67 had 2 HEAD & NECK—DOI 10.1002/HED
APRIL 2015
499
CHOI ET AL.
FIGURE 2. Diagnostic algorithm of 1269 thyroid nodules in 1200 patients. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary. com.]
nodules, and 1 patient had 3 nodules. Patients ranged in age from 20 to 79 years (mean, 48.6 years).
Ultrasound and ultrasound-guided fine-needle aspiration Real-time thyroid ultrasound was performed by 7 radiologists (4 faculties with 5–13 years of experience and 3 fellows with 1–2 years of experience). Ultrasound scans of the thyroid glands were obtained with a 5 to 12 MHz linear probe (iU22; Philips Medical Systems, Bothell, WA). Interpretation of ultrasound features of all thyroid nodules were prospectively recorded according to internal component, echogenicity, margins, calcifications, shape, and final assessment by the radiologists who had performed the ultrasound. Ultrasound features were classified according to previously published criteria.8 Suspicious malignant ultrasound features were defined as the following: marked hypoechogenicity (decreased echogenicity compared to the surrounding muscle), microlobulated or irregular margins, microcalcifications, and taller-than-wide shape. If a nodule showed 1 or more of these suspicious features on ultrasound, it was classified as suspiciously malignant. If a nodule showed no suspicious features, it was classified as probably benign. Nodule size was also measured in 3 dimensions and defined with the greatest diameter. After thyroid ultrasound, ultrasound-guided FNA was performed by the same radiologist who had performed the ultrasound. At our institution, ultrasound-guided FNA is performed in either the thyroid nodule with suspicious ultrasound features or the largest thyroid nodule if no suspicious features are detected. According to the American Thyroid Association guidelines, FNA for a thyroid nodule is not routinely recommended for nodules 5 mm at our institution unless a patient has specific risk factors (eg, family history of thyroid cancer or radiation exposure). However, we sometimes perform FNAs of very smallsized nodules at the request of the patient. Ultrasound500
HEAD & NECK—DOI 10.1002/HED
APRIL 2015
guided FNA was performed with a 23-gauge needle attached to a 2-mL disposable plastic syringe and each lesion was aspirated at least twice. Aspirated materials were expelled onto glass slides and smeared. All smears were immediately placed in 95% alcohol for Papanicolaou staining. The remaining aspirated materials were rinsed with saline for cell block processing. Additional special staining was done according to the cytopathologists’ needs on a case-by-case basis.
Laboratory test Serum TSH levels were measured by IRMA (TSHCTK-3, SORIN Biomedica, Saluggia, Italy) in every patient. The normal reference range for serum TSH was between 0.3 and 4.99 mIU/L. Among the 1709 patients who underwent ultrasound-guided FNA during the study period, 26 patients were excluded because they did not have records available of serum TSH levels within 3 months before ultrasound-guided FNA and 175 patients whose serum TSH levels were not within normal range were also excluded (Figure 1). Therefore, 1200 euthyroid patients were finally included in this study.
Histopathologic results Final diagnosis was based on cytologic or histopathologic results. Malignant nodules were confirmed with surgery (n 5 304) or FNA (n 5 38). Benign nodules were also confirmed with surgery (n 5 2) or FNA (n 5 925; Figure 2).
Data and statistical analysis The age and nodule size of the benign and malignant groups were compared using an independent 2-sample t test. Sex and ultrasound features of the 2 groups were compared using the chi-square test. The TSH levels were graded into 3 categories for statistical analysis (0.3–1.39; 1.4–2.49; and 2.5–4.99 mIU/L). The serum TSH grades
SERUM TSH
TABLE 1. Baseline characteristics of 1269 thyroid nodules in 1200 patients. Benign
Malignant
p value
All No. of patients* No. of nodules Mean age, y Sex (%) Male Female Mean nodule size, mm TSH value Mean TSH level, mIU/L TSH grade (%) 0.3–1.39 1.4–2.49 2.5–4.99 Ultrasound feature (%) Probably benign Suspiciously malignant Nodule size 1 cm No. of patients† No. of nodules Mean age Sex (%) Male Female Mean nodule size TSH value Mean TSH level TSH grade (%) 0.3–1.39 1.4–2.49 2.5–4.99 Ultrasound feature Probably benign Suspiciously malignant Nodule size >1 cm No. of patients† No. of nodules Mean age Sex (%) Male Female Mean nodule size TSH value Mean TSH level TSH grade (%) 0.3–1.39 1.4–2.49 2.5–4.99 Ultrasound feature (%) Probably benign Suspiciously malignant
901 312 927 342 50.9 6 11.6 47.6 6 11.7
