World J Surg (2014) 38:80–87 DOI 10.1007/s00268-013-2287-z
Predictive Factors of Thyroid Cancer in Patients with Graves’ Disease Meng Ren • Mu Chao Wu • Chang Zhen Shang Xiao Yi Wang • Jing Lu Zhang • Hua Cheng • Ming Tong Xu • Li Yan
•
Published online: 29 October 2013 Ó Socie´te´ Internationale de Chirurgie 2013
Abstract Background The best preoperative examination in Graves’ disease with thyroid cancer still remains uncertain. The objectives of the present study were to investigate the prevalence of thyroid cancer in Graves’ disease patients, and to identify the predictive factors and ultrasonographic features of thyroid cancer that may aid the preoperative diagnosis in Graves’ disease. Methods This retrospective study included 423 patients with Graves’ disease who underwent surgical treatment from 2002 to 2012 at our institution. The clinical features and ultrasonographic findings of thyroid nodules were recorded. The diagnosis of thyroid cancer was determined according to the pathological results. Results Thyroid cancer was discovered in 58 of the 423 (13.7 %) surgically treated Graves’ disease patients; 46 of those 58 patients had thyroid nodules, and the other 12 patients were diagnosed with incidentally discovered thyroid carcinomas without thyroid nodules. Among the 58 patients with thyroid cancer, papillary microcarcinomas were discovered in 50 patients, and multifocality and
M. Ren and M.C. Wu contributed equally to this work. M. Ren M. C. Wu X. Y. Wang J. L. Zhang H. Cheng M. T. Xu (&) L. Yan (&) Department of Endocrinology, Sun Yat-sen Memorial Hospital of Sun Yat-sen University, No.107 Yan jiang West Road, Guangzhou 510120, China e-mail: [email protected] L. Yan e-mail: [email protected] C. Z. Shang Department of Surgery, Sun Yat-sen Memorial Hospital of Sun Yat-sen University, Guangzhou 510120, China
123
lymph node involvement were detected in the other 8 patients. Multivariate regression analysis showed younger age was the only significant factor predictive of metastatic thyroid cancer. Ultrasonographic findings of calcification and intranodular blood flow in thyroid nodules indicate that they are more likely to harbor thyroid cancers. Conclusions Because the influencing factor of metastatic thyroid cancers in Graves’ disease is young age, every suspicious nodule in Graves’ disease patients should be evaluated and treated carefully, especially in younger patients because of the potential for metastasis.
Introduction The coexistence of thyroid nodules and Graves’ disease is common; however, the potential risk of malignancy of thyroid nodules in Graves’ disease remains unknown [1]. The prevalence of thyroid cancer in Graves’ patients varies greatly, ranging from 0.4 to 10 % [2–4]. The clinical manifestation of these concomitant carcinomas in Graves’ disease has been described as ranging from having no clinical significance [5] to being very aggressive [6]. In recent years, the prevalence of clinical thyroid cancer in Graves’ disease patients has increased. As the risk that carcinoma will develop within a nodule in Graves’ disease increases, the diagnosis and treatment of nodules and the risk of malignancy in Graves’ disease patients deserve more consideration [7–9]. However, no consensus on the best preoperative diagnosis of nodules in Graves’ disease is available. Thyroid ultrasonography is recognized the most sensitive method for detecting thyroid nodules [10]. The use of ultrasonography to assess the thyroid gland in Graves’ disease patients has resulted in the early detection of
World J Surg (2014) 38:80–87
thyroid nodules that were not found by routine physical examination [11]. The characteristics of thyroid cancer through ultrasonographic screening have been reported in many researches [12–14]; however, because the features of Graves’ disease vary on ultrasonography, it becomes difficult to differentiate benign and malignant nodules in that condition [15]. The present study was performed (1) to determine the prevalence and clinical influence of thyroid cancer in patients undergoing thyroidectomy for Graves’ disease, (2) to explore the differences in ultrasonographic findings between malignant and benign nodules in patients with Graves’ disease, and (3) to identify the risk factors of thyroid cancer in patients with Graves’ disease.
Subjects and methods Graves’ disease patients During the period 2002 to 2012, 423 patients with Graves’ disease underwent surgical treatment at our institute. The diagnosis of Graves’ disease was made as follows: history of Graves’ disease, signs of hyperthyroidism, elevated serum free thyroxine and free triiodothyronine levels, obviously suppressed thyroid stimulating hormone (TSH) levels, and the presence of TSH receptor antibody (TRAb) in serum. A total of 55 patients underwent scintiscan examination to confirm the diagnosis of Graves’ disease. None of the patients had previously received external irradiation to the neck and had a family history of thyroid cancer or autoimmune thyroid disease. Surgery criteria Of the 423 patients in this study, surgery was performed in 235 patients who had a poor response to medical treatment, who refused to use antithyroid drugs (i.e., methimazole or propylthiouracil), or who experienced relapse after antithyroid drug treatment over a long period. Surgery was also advised for 180 patients who had either obvious enlargement of the thyroid or large nodules associated with compressive symptoms. Furthermore, eight patients with nodules displaying clinical or ultrasonographic signs suggestive of malignancy also underwent thyroidectomy. Thyroid ultrasonography Before surgery, thyroid ultrasonography was performed in all patients. Ultrasonography was performed using a 13 MHz linear probe (HV-900, Hitachi, Tokyo, Japan). The size of nodule was expressed in its greatest dimension. The following ultrasonographic parameters were assessed:
81
nodule margin (clear or unclear), nodule shape (regular or irregular), echogenicity (hyperechoic, isoechoic, hypoechoic), nodule composition (completely solid, predominantly solid, mixed solid and cystic, predominantly cystic); homogeneity (homogeneous or heterogeneous); calcification (none, microcalcification, dense calcification, or rim calcification); and nodular vascularity (none, perinodular, intranodular, mixed vascularity). Ultrasonography was performed by a single operator and confirmed by another expert ultrasonographer. Preoperative treatment and histopathological diagnosis Before operation, the patients were treated with methimazole or propylthiouracil. Lugol solution was also given for 7 days before surgical intervention. All patients had total or near-total thyroidectomy. The criteria for surgical intervention were as follows: patient preference, failure of antithyroid drugs, failure of radioactive iodine treatment, obvious goiter with compressive symptoms, and severe ophthalmopathy. Tumor tissues (formalin-fixed and paraffin-embedded) were stained with hematoxylin and eosin. The pathological diagnosis was made by two independent pathologists blindly. Microcarcinoma was identified as carcinoma with diameter \10 mm. Statistical analysis All data are expressed as means ± SD. SPSS software was used to perform statistical analyses (SPSS Inc., Chicago, IL). The frequency distribution of categorical variables between subgroups was compared by the continuity corrected chi square test, Fisher’s exact test, or the Kruskal– Wallis test. Numerical variables between subgroups were compared by unpaired t tests. Logistic regression analysis was performed to determine the factor(s) predicting the presence of nodules or cancer. A p value less than 0.05 was considered statistically significant.
Results Clinical characteristics and ultrasonographic features of thyroid nodules in Graves’ patients Thyroid nodules were detected by ultrasonography in 22.7 % (96/423) of the patients with Graves’ disease (Table 1). The prevalence of nodules, especially multiple nodules, increased significantly with age (p \ 0.05). The prevalence of nodules detected by ultrasonography was slightly higher in women than in men; however, this comparison did not show statistical significance (25.0 %
123
82
World J Surg (2014) 38:80–87
Table 1 Clinical characteristics of 423 patients with Graves’ disease with or without thyroid nodules Diffuse goiter (n = 327)
Solitary nodule (n = 25)
Multiple nodules (n = 71)
p Value
Percentages
77.3 %
5.9 %
16.8 %
\0.05
Age (years)
32.07 ± 10.17
39.00 ± 13.01
46.41 ± 13.1
\0.001
Gender (M/F)
75/252
2/23
10/61
NS
Incidence of thyroid carcinoma
12 (3.6 %)
14 (56 %)
32 (45 %)
\0.001
NS not significant Table 2 Characteristics of begin and malignant nodules in patients with Graves’ disease Solitary nodule (n = 25) Benign (n = 11)
Multiple nodules (n = 71) Malignant (n = 14)
Benign (n = 39)
Malignant (n = 32)
Age (years)
44.09 ± 13.59
35.00 ± 11.46
48.08 ± 13.99
43.38 ± 11.84
Gender (M/F)
0/11
2/12
9/30*
1/31
Duration of Graves’ disease (years) B1
2
3
11
16
1–5
5
5
13
5
5–10
2
1
9
3
C10
1
3
7
8
B1
4
5
4
4
1–2
4
4
15
11
C2
2
5
26
17
Diameter (cm)
* p \ 0.05 versus malignant multiple nodules
[84/336] vs 13.8 % [12/87]; p [ 0.05.). Of 96 patients with thyroid nodules, 25 (26.0 %) had a solitary nodule, and 71 (74.0 %) had multiple nodules. In logistic regression analysis performed using patients age, gender, disease duration, and goiter diameter as independent variables, only age was a significant factor predicting presence of solitary and multiple thyroid nodule(s). In patients with Graves’ disease, the total incidence of thyroid cancer was 13.7 % (58/423). Among the 58 patients with thyroid cancer, 12 had diffuse goiter; that is to say, thyroid carcinomas occurred in 12 Graves’ disease patients without ultrasonographically identified nodules. In the other 46 patients with thyroid cancers, thyroid nodule(s) were detected. The incidence of incidentally discovered thyroid cancer was 3.6 % (12/327) in patients with diffuse goiter. Thyroid cancer was significantly more frequent in solitary nodules (14/25, 56.0 %) than in multiple nodules (32/71, 45.1 %; p \ 0.0001). Factor(s) predicting presence of cancer In order to determine the factors predictive of thyroid cancer in Graves’ patients with nodules, we further compared the clinical characteristics between those with cancer and without (Table 2). A multiple logistic regression
123
analysis analyzing age, gender, and type of nodularity was applied to determine the independent risk predictors of thyroid cancer based on the histology. This analysis did not show any significant difference in identifying thyroid cancer in solitary nodules and multiple nodules. At the same time, the disease duration and nodule diameter did not help predict the presence of thyroid cancer, whether in patients with solitary nodules or multiple nodules. Thyroid cancer histology was only positively associated with the female gender in malignant multiple nodules (p \ 0.05). Clinical and pathological characteristics of thyroid cancer in Graves’ patients Table 3 shows the overall clinical and pathological results of the 58 thyroid cancer patients. Overall, the mean age of patients with thyroid cancer in a solitary nodule was significantly lower than that of patients with cancer in multiple nodules (p \ 0.001). The male/female ratio was 1/11, 2/12, and 1/31, respectively in patients with diffuse goiter, solitary nodule, and multiple nodules. There was not any significance of disease duration or pathological type among these three groups. Thyroid carcinomas could also occur in Graves’ disease patients without ultrasonographically identified nodules. Of
World J Surg (2014) 38:80–87
83
Table 3 Clinical characteristics of 58 thyroid carcinomas in Graves’ patients Thyroid carcinoma
p Value
Diffuse goiter (n = 12)
Solitary nodule (n = 14)
Multiple nodules (n = 32)
Age (years)
36.33 ± 9.96
35.00 ± 11.46
44.38 ± 11.84
\0.001
Gender (M/F)
1/11
2/12
1/31
NS (0.308)
Duration of Graves’ disease (years) B1 3 (25.0 %)
5 (35.7 %)
16 (50.0 %)
–
NS (0.690)
1–5
5 (41.7 %)
4 (28.6 %)
5 (15.6 %)
–
5–10
1 (8.3 %)
3 (21.4 %)
3 (9.4 %)
–
3 (25.0 %)
2 (14.3 %)
8 (25.0 %)
–
Papillary
11 (91.7 %)
14 (100.0 %)
31 (96.9 %)
NS (0.497)
Follicular
0 (0.0 %)
0(0.0 %)
1 (3.1 %)
–
Undifferentiated
1 (8.3 %)
0(0.0 %)
0 (0.0 %)
–
Microcarcinoma (cases)
10 (83 %)
10 (71.4 %)
30 (93.75 %)
NS (0.372)
Metastasis (cases)
2 (17 %)
4 (28.6 %)
2 (6.25 %)
0.037
C10 Pathological type
NS not significant
the 58 thyroid cancer patients in the present study, 12 were diagnosed with incidentally discovered thyroid carcinomas without nodules. Ten of these 12 patients had papillary microcarcinomas with no invasion of surrounding tissues or distant metastases. In the other two patients with incidentally discovered thyroid cancer, the disease showed locally advanced characteristics. As for pathological results, 56 patients (96.6 %) were diagnosed with papillary thyroid carcinomas, one with follicular thyroid carcinoma, and 1 with undifferentiated carcinoma. The pathological results in 50 patients (86.2 %) revealed microcarcinomas with a diameter of 10 mm or less. The other eight patients had locally advanced thyroid cancer, two with diffuse goiter, four with solitary nodules, and the other two with multiple nodules. Patients with solitary nodules had a higher percentage of metastasis than those with diffuse goiter or multiple nodules.
hypoechoic echogenicity was lower in Graves’ patients with benign nodules than in those with malignant nodules; however, there was no difference in echogenicity between the two groups. With regard to calcification pattern, 30.4 % Graves’ patients with malignant nodules had ultrasonographic evidence of microcalcification, rim calcification, or dense calcification, with an incidence of calcification significantly higher than that of benign nodules (6.0 %, p \ 0.05). The frequencies of no nodular blood flow and intranodular blood flow in patients with benign nodules were 58.0 and 24.0 %, respectively, with statistical significance compared with the frequencies of those in malignant nodules (30.4 and 56.5 %, respectively; p \ 0.05).
Ultrasonographic features predicting benign and malignant nodules associated with Graves’ disease
Among the 58 patients with thyroid carcinoma, eight had evidence of distant metastasis. The clinical characteristics and histological features are listed in Table 5. All eight were women. Notably, with the exception of a 69-year-old woman, the other seven patients were all young, with a mean age of 29.4 ± 7.15 years (range: 13–34 years). Spearman analysis showed that young age was associated with the risk of metastatic thyroid carcinoma (r = -0.298, p \ 0.05). In the patients in the present study six of the eight metastatic tumors were papillary carcinomas, one of the follicular variant, and one case of the undifferentiated phenotype. Six tumors showed microscopic extrathyroid invasion, one exhibited thyroid capsule invasion, and one showed vascular invasion. Four patients had solitary
We further compared the ultrasonographic results of benign and malignant nodules in Graves’ patients (Table 4). Regular shape was seen in about 98.0 % and 93.5 % of benign and malignant nodules. The frequencies of clear border in patients with benign and malignant nodules were 72.0 and 67.4 %, respectively. Heterogeneous echogenicity was seen in about 90.0 % of benign nodules and in about 69.6 % of malignant nodules, with statistical difference. Most of the benign and malignant nodules in Graves’ patients were composed of completely or predominantly solid elements. The frequency of
Clinical and histological characteristics of Graves’ disease patients with metastatic thyroid carcinoma
123
84
World J Surg (2014) 38:80–87
Table 4 Ultrasonographic features predicting benign and malignant nodules associated with Graves’ disease Ultrasonographic findings Shape of nodule
Type
Benign nodules (n = 50)
Malignant nodules (n = 46)
Regular
49 (98.0 %)
43 (93.5 %)
Irregular
1 (2.0 %)
3 (6.5 %)
p Value NS (0.551) –
Border of nodule
Clear
36 (72.0 %)
31 (67.4 %)
Unclear
14 (28.0 %)
15 (32.6 %)
–
Homogeneity
Homogeneous
5 (10.0 %)
14 (30.4 %)
0.012
Heterogeneous
45 (90.0 %)
32 (69.6 %)
Composition
Completely solid
26 (52.0 %)
31 (67.4 %)
Predominantly solid
22 (44.0 %)
15 (32.6 %)
–
2 (4.0 %)
0 (0.0 %)
–
Mixed solid and cystic Predominantly cystic
0 (0.0 %)
0 (0.0 %)
34 (68.0 %) 3 (6.0 %)
33 (71.7 %) 4 (8.7 %)
NS (0.623)
– NS (0.100)
–
Echogenicity
Hypoechoic Isoechogenic Hyperechoic
13 (26.0 %)
9 (19.6 %)
–
Calcification
None
47 (94.0 %)
32 (69.6 %)
0.002
Microcalcification
3 (6.0 %)
10 (21.8 %)
–
Dense calcification
0 (0.0 %)
2 (4.3 %)
–
Rim calcification Vascularity
0(0.0 %)
2 (4.3 %)
None
29 (58.0 %)
13 (28.3 %)
0.049
Intranodular
12 (24.0 %)
27 (58.7 %)
–
Perinodular
3 (6.0 %)
2 (4.3 %)
–
Mixed vascularity
6 (12.0 %)
4 (8.7 %)
–
nodules, two had multiple nodules, and another two showed diffuse goiter. The nodules were all [2 cm in diameter. Ultrasonographic features showed that all eight cancers demonstrated completely or predominantly solid composition and hypoechoic echogenicity. In three patients the lesions had an ill-defined border and intranodular blood flow. Dense calcification was noted in one patient with diffuse goiter. Four patients showed calcification pattern in thyroid nodules.
Discussion In this retrospective study, we determined the prevalence of thyroid cancer in Graves’ disease patients, and also assessed the clinical and ultrasonographic characteristics of thyroid cancers in those patients. Thyroid nodules are commonly found in subjects with Graves’ disease. In fact, it is reported that the prevalence of thyroid nodules is higher in those with Graves’ disease than in the normal population when detected on palpation or via imaging modalities [2]. The prevalence of palpable thyroid nodules varies between 25 and 45 % in Graves’ disease patients [2, 16, 17]. In some cases, Graves’ disease itself may cause thyroid nodularity [18].
123
NS (0.613) –
–
In this study, thyroid nodules were detected by ultrasonography in 22.7 % of the Graves’ disease patients. Nodules with a diameter more than 1 cm were found in 33.6 % of these patients. Benign and malignant lesions often coexist in thyroid disease, such as Hashimoto’s thyroiditis and Graves’ disease. In recent years, the coexistence of thyroid cancer and Graves’ disease has been of great concern. Sometimes an unsuspected thyroid cancer was discovered incidentally on pathologic examination in a patient with Graves’ disease [5]. In our study, incidental thyroid carcinoma was detected in 12 of 327 Graves’ patients with diffuse goiter, with a frequency less than 4 %. Papillary microcarcinoma was diagnosed in ten of these 12 patients, with no invasion of surrounding tissues and no lymph node or distant metastases. It had been reported that incidental carcinomas in Graves’ disease patients were of no clinical consequence [5]; however, two patients with incidentally discovered thyroid cancer in our report had evidence of distant metastasis. Compared with a nodule in a euthyroid patient, a thyroid nodule diagnosed in a Graves’ disease patient has a higher risk of malignancy. In fact, the incidence of malignancy in the nodules of Graves’ disease patients is common. In our study, the total incidence of thyroid carcinoma in surgically treated Graves’ disease patients was 13.7 %. Furthermore,
None MicroHypoechoic Homogeneity Clear 25.0 9 39.0 Solitary Papillary 8 27 8
F
–
Intranodular Micro-
Dense–
Hypoechoic Heterogeneous
– –
Unclear
Completely solid Completely solid
– –
31.8 9 21.0 Multiple
Diffuse goiter Papillary
Papillary 8
1
69 7
F
16 6
F
None None Hypoechoic Heterogeneous Clear Predominantly solid 36.0 9 35.0 Solitary Papillary 8 27 5
F
Intranodular DenseHypoechoic Heterogeneous Unclear Predominantly solid 31.0 9 16.0 Solitary Papillary 1 13 4
F
Intranodular MicroHypoechoic Homogeneity Unclear Predominantly solid 30.2 9 21.0 Multiple Follicular 4 26 3
F
– – – – – – – Diffuse goiter Undifferentiated 3 34 2
F
Perinodular None Hypoechoic Homogeneity Clear Completely solid 24.6 9 10.1 Solitary Papillary 5 23 1
F
Margin of nodule Composition Nodules size (mm) Nodule number Pathology Duration (years) Age
Gender
85
Patient
Table 5 Clinical characteristics and histological features of Graves’ disease patients with metastatic thyroid carcinoma
Homogeneity
Echogenicity
Calcification
Vascularity
World J Surg (2014) 38:80–87
it has been reported that thyroid carcinoma is more frequent in solitary nodules than in multiple nodules. In our Graves’ patients with solitary thyroid nodules and multiple nodules, the incidence of malignancy reached 56 and 45 %, respectively. This result is in close agreement with other studies [19], whereas other authors [20–22] have reported that there is no difference in cancer prevalence between patients with solitary thyroid nodules and multiple nodular goiters. Criteria in the selection of Graves’ disease patients for thyroidectomy vary from time to time, which may explain the differences between studies carried out in different eras. Fifty years ago, thyroidectomy was the most frequent therapy for Graves’ disease; thereafter, anti-thyroid drugs and radioiodine therapy were the treatments of choice, and they were replaced in part with the traditional method of thyroidectomy [23]. In more recent years, patients with serious signs of Graves’ disease undergo surgical treatment, thus may not represent the whole population of Graves’ disease patients [1]. In the present study, many patients enrolled experienced a relatively longer disease duration, which may have led to the high incidence of nodule formation and the presence of thyroid carcinoma. An inverse relationship was observed between the risk of thyroid carcinoma and age. Patients with papillary thyroid cancer were younger. This finding is in agreement with Mittendorf’s research [24], in whose study the mean age was 35 and 43.3 years in patients with solitary nodules and multiple nodules, respectively. Our study demonstrated that male gender did not confer a higher risk of cancer in Graves’ patients, a finding that is not in agreement with other studies [25–27]. Perhaps this can be attributed to the finding that the prevalence of multiple nodules is less frequent in men than in women. We detected a high incidence of carcinoma in nodules from patients with Graves’ disease in the present study, and found that papillary thyroid cancer was the most common histologic subtype. More important, up to 84.5 % of all cases of papillary thyroid carcinoma were papillary microcarcinomas, as defined by the World Health Organization’s classification of thyroid cancer. It is reported that carcinoma in Graves’ disease may have a bad outcome, such as distant metastasis and patient death. Therefore, thyroidectomy is recommended in patients with the diagnosis of papillary microcarcinoma [28]. However, the progression and prognosis of thyroid cancer in Graves’ disease patients is a matter of debate. In the present study, eight thyroid carcinoma patients showed metastatic characteristics. Age may be the only factor predicting locally advanced cancers in Graves’ patients. With the exception of a single 69-year-old patient, the carcinoma patients were all younger, with a 13-year-old with a 1 year history of Graves’ disease the youngest.
123
86
It was once suggested that patients with Graves’ disease should be evaluated by ultrasonography for detection of thyroid nodules, and that thyroidectomy should be performed for control of hyperthyroidism and removal of potentially malignant nodules [16]. In fact, for all the Graves’ disease patients, conventional thyroidectomy for removal of nodules seems not to be appropriate in light of the surgical risks. Our study shows that younger subjects may have higher risk of thyroid carcinoma. This indicates that suspicious nodule detected in patients with Graves’ disease should be evaluated carefully, especially in younger patients. Malignant nodules usually had the following pathological features: hypoechogenicity, ill-defined borders, absence of a peripheral hypoechoic rim, no cystic changes, increased intranodular blood flow, and punctate calcifications. However, only heterogeneous echogenicity, calcification, and vascularity showed significant differences between benign and malignant nodules in Graves’ patients in the present study. Calcifications were more frequently detected in malignant nodules than in benign nodules. Frates et al. [29, 30] reported that thyroid carcinomas had increased nodular blood flow. The research of Cerbone et al. [31] showed that the large degree of cellular proliferation in malignant nodules could explain the dominant intranodular blood flow. In our series we found that malignant nodules, as compared to benign nodules, in patients with Graves’ disease tended to exhibit significantly increased intranodular blood flow. In conclusion, in this retrospective study, the prevalence of thyroid cancer in Graves’ disease was about 13.7 %, and the incidence of nodules in malignant Graves’ disease even reached 47.9 %. Compared to Graves’ disease patients with diffuse goiter, those with thyroid nodules are at higher risk of developing thyroid cancer. Although there is a high risk of carcinomas in Graves’ disease patients with nodules, most of them are micropapillary thyroid cancers. Age is one of the most important factors predictive of locally advanced cancers in Graves’ patients. According to the potential metastatic characteristics of the nodules, every suspicious nodule in Graves’ disease should be evaluated and treated seriously, especially in younger patients. As a kind of noninvasive and time-saving method, ultrasonography should be used routinely to find potential malignancies in Graves’ disease. Graves’ patients with calcification and intranodular blood flow in ultrasonographic findings are more likely to harbor thyroid cancers. Funding This work was supported in part by a grant from the National Natural Science Foundation of China (81000317) and by a Yat-Sen Scholarship for Young Scientist. Conflict of interest All authors declare that no conflicts of interest exist regarding the research reported.
123
World J Surg (2014) 38:80–87
References 1. Gerenova J, Buysschaert M, de Burbure CY et al (2003) Prevalence of thyroid cancer in Graves’ disease: a retrospective study of a cohort of 103 patients treated surgically. Eur J Intern Med 14:321–325 2. Belfiore A, Garofalo MR, Giuffrida D et al (1990) Increased aggressiveness of thyroid cancer in patients with Graves’ disease. J Clin Endocrinol Metab 70:830–835 3. Angusti T, Codegone A, Pellerito R et al (2000) Thyroid cancer prevalence after radioiodine treatment of hyperthyroidism. J Nucl Med 41:1006–1009 4. Gul K, Di Ri Koc A, Ki Yak G et al (2009) Thyroid carcinoma risk in patients with hyperthyroidism and role of preoperative cytology in diagnosis. Minerva Endocrinol 34:281–288 5. Phitayakorn R, McHenry CR (2008) Incidental thyroid carcinoma in patients with Graves’ disease. Am J Surg 195:292–297. discussion 297 6. Hales IB, McElduff A, Crummer P et al (1992) Does Graves’ disease or thyrotoxicosis affect the prognosis of thyroid cancer. J Clin Endocrinol Metab 75:886–889 7. Ozaki O, Ito K, Kobayashi K et al (1990) Thyroid carcinoma in Graves’ disease. World J Surg 14:437–440. doi:10.1007/BF01 658550. discussion 440–431 8. Chao TC, Lin JD, Jeng LB et al (1999) Thyroid cancer with concurrent hyperthyroidism. Arch Surg 134:130–134 9. Okamoto T, Iihara M, Obara T (2000) Management of hyperthyroidism due to Graves’ and nodular diseases. World J Surg 24:957–961. doi:10.1007/s002680010167 10. Terzioglu T, Tezelman S, Onaran Y et al (1993) Concurrent hyperthyroidism and thyroid carcinoma. Br J Surg 80:1301– 1302 11. Kim WB, Han SM, Kim TY et al (2004) Ultrasonographic screening for detection of thyroid cancer in patients with Graves’ disease. Clin Endocrinol (Oxf) 60:719–725 12. Bruneton JN, Balu-Maestro C, Marcy PY et al (1994) Very high frequency (13 MHz) ultrasonographic examination of the normal neck: detection of normal lymph nodes and thyroid nodules. J Ultrasound Med 13:87–90 13. Hegedus L (2004) Clinical practice. The thyroid nodule. N Engl J Med 351:1764–1771 14. Papini E, Guglielmi R, Bianchini A et al (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 87:1941–1946 15. Chan BK, Desser TS, McDougall IR et al (2003) Common and uncommon sonographic features of papillary thyroid carcinoma. J Ultrasound Med 22:1083–1090 16. Berker D, Isik S, Ozuguz U et al (2011) Prevalence of incidental thyroid cancer and its ultrasonographic features in subcentimeter thyroid nodules of patients with hyperthyroidism. Endocrine 39:13–20 17. Mishra A, Mishra SK (2001) Thyroid nodules in Graves’ disease: implications in an endemically iodine deficient area. J Postgrad Med 47:244–247 18. Kraimps JL, Bouin-Pineau MH, Mathonnet M et al (2000) Multicentre study of thyroid nodules in patients with Graves’ disease. Br J Surg 87:1111–1113 19. Carnell NE, Valente WA (1998) Thyroid nodules in Graves’ disease: classification, characterization, and response to treatment. Thyroid 8:647–652 20. Christensen SB, Ljungberg O, Tibblin S (1984) Thyroid carcinoma in Malmo, 1960–1977. Epidemiologic, clinical, and prognostic findings in a defined urban population. Cancer 53:1625–1633
World J Surg (2014) 38:80–87 21. Belfiore A, La Rosa GL, La Porta GA et al (1992) Cancer risk in patients with cold thyroid nodules: relevance of iodine intake, sex, age, and multinodularity. Am J Med 93:363–369 22. McCall A, Jarosz H, Lawrence AM et al (1986) The incidence of thyroid carcinoma in solitary cold nodules and in multinodular goiters. Surgery 100:1128–1132 23. Wartofsky L, Glinoer D, Solomon B et al (1991) Differences and similarities in the diagnosis and treatment of Graves’ disease in Europe, Japan, and the United States. Thyroid 1:129–135 24. Mittendorf EA, Khiyami A, McHenry CR (2006) When fineneedle aspiration biopsy cannot exclude papillary thyroid cancer: a therapeutic dilemma. Arch Surg 141:961–966. discussion 966 25. Rojeski MT, Gharib H (1985) Nodular thyroid disease. Evaluation and management. N Engl J Med 313:428–436 26. Gharib H (1994) Fine-needle aspiration biopsy of thyroid nodules: advantages, limitations, and effect. Mayo Clin Proc 69:44–49
87 27. Rago T, Fiore E, Scutari M et al (2010) Male sex, single nodularity, and young age are associated with the risk of finding a papillary thyroid cancer on fine-needle aspiration cytology in a large series of patients with nodular thyroid disease. Eur J Endocrinol 162:763–770 28. Belfiore A, Russo D, Vigneri R et al (2001) Graves’ disease, thyroid nodules and thyroid cancer. Clin Endocrinol (Oxf) 55:711–718 29. Taylor KJ, Ramos I, Carter D et al (1988) Correlation of Doppler US tumor signals with neovascular morphologic features. Radiology 166:57–62 30. Frates MC, Benson CB, Doubilet PM et al (2003) Can color Doppler sonography aid in the prediction of malignancy of thyroid nodules? J Ultrasound Med 22:127–131 quiz 132–124 31. Cerbone G, Spiezia S, Colao A et al (1999) Power Doppler improves the diagnostic accuracy of color Doppler ultrasonography in cold thyroid nodules: follow-up results. Horm Res 52:19–24
123
Predictive Factors of Thyroid Cancer in Patients with Graves’ Disease Meng Ren • Mu Chao Wu • Chang Zhen Shang Xiao Yi Wang • Jing Lu Zhang • Hua Cheng • Ming Tong Xu • Li Yan
•
Published online: 29 October 2013 Ó Socie´te´ Internationale de Chirurgie 2013
Abstract Background The best preoperative examination in Graves’ disease with thyroid cancer still remains uncertain. The objectives of the present study were to investigate the prevalence of thyroid cancer in Graves’ disease patients, and to identify the predictive factors and ultrasonographic features of thyroid cancer that may aid the preoperative diagnosis in Graves’ disease. Methods This retrospective study included 423 patients with Graves’ disease who underwent surgical treatment from 2002 to 2012 at our institution. The clinical features and ultrasonographic findings of thyroid nodules were recorded. The diagnosis of thyroid cancer was determined according to the pathological results. Results Thyroid cancer was discovered in 58 of the 423 (13.7 %) surgically treated Graves’ disease patients; 46 of those 58 patients had thyroid nodules, and the other 12 patients were diagnosed with incidentally discovered thyroid carcinomas without thyroid nodules. Among the 58 patients with thyroid cancer, papillary microcarcinomas were discovered in 50 patients, and multifocality and
M. Ren and M.C. Wu contributed equally to this work. M. Ren M. C. Wu X. Y. Wang J. L. Zhang H. Cheng M. T. Xu (&) L. Yan (&) Department of Endocrinology, Sun Yat-sen Memorial Hospital of Sun Yat-sen University, No.107 Yan jiang West Road, Guangzhou 510120, China e-mail: [email protected] L. Yan e-mail: [email protected] C. Z. Shang Department of Surgery, Sun Yat-sen Memorial Hospital of Sun Yat-sen University, Guangzhou 510120, China
123
lymph node involvement were detected in the other 8 patients. Multivariate regression analysis showed younger age was the only significant factor predictive of metastatic thyroid cancer. Ultrasonographic findings of calcification and intranodular blood flow in thyroid nodules indicate that they are more likely to harbor thyroid cancers. Conclusions Because the influencing factor of metastatic thyroid cancers in Graves’ disease is young age, every suspicious nodule in Graves’ disease patients should be evaluated and treated carefully, especially in younger patients because of the potential for metastasis.
Introduction The coexistence of thyroid nodules and Graves’ disease is common; however, the potential risk of malignancy of thyroid nodules in Graves’ disease remains unknown [1]. The prevalence of thyroid cancer in Graves’ patients varies greatly, ranging from 0.4 to 10 % [2–4]. The clinical manifestation of these concomitant carcinomas in Graves’ disease has been described as ranging from having no clinical significance [5] to being very aggressive [6]. In recent years, the prevalence of clinical thyroid cancer in Graves’ disease patients has increased. As the risk that carcinoma will develop within a nodule in Graves’ disease increases, the diagnosis and treatment of nodules and the risk of malignancy in Graves’ disease patients deserve more consideration [7–9]. However, no consensus on the best preoperative diagnosis of nodules in Graves’ disease is available. Thyroid ultrasonography is recognized the most sensitive method for detecting thyroid nodules [10]. The use of ultrasonography to assess the thyroid gland in Graves’ disease patients has resulted in the early detection of
World J Surg (2014) 38:80–87
thyroid nodules that were not found by routine physical examination [11]. The characteristics of thyroid cancer through ultrasonographic screening have been reported in many researches [12–14]; however, because the features of Graves’ disease vary on ultrasonography, it becomes difficult to differentiate benign and malignant nodules in that condition [15]. The present study was performed (1) to determine the prevalence and clinical influence of thyroid cancer in patients undergoing thyroidectomy for Graves’ disease, (2) to explore the differences in ultrasonographic findings between malignant and benign nodules in patients with Graves’ disease, and (3) to identify the risk factors of thyroid cancer in patients with Graves’ disease.
Subjects and methods Graves’ disease patients During the period 2002 to 2012, 423 patients with Graves’ disease underwent surgical treatment at our institute. The diagnosis of Graves’ disease was made as follows: history of Graves’ disease, signs of hyperthyroidism, elevated serum free thyroxine and free triiodothyronine levels, obviously suppressed thyroid stimulating hormone (TSH) levels, and the presence of TSH receptor antibody (TRAb) in serum. A total of 55 patients underwent scintiscan examination to confirm the diagnosis of Graves’ disease. None of the patients had previously received external irradiation to the neck and had a family history of thyroid cancer or autoimmune thyroid disease. Surgery criteria Of the 423 patients in this study, surgery was performed in 235 patients who had a poor response to medical treatment, who refused to use antithyroid drugs (i.e., methimazole or propylthiouracil), or who experienced relapse after antithyroid drug treatment over a long period. Surgery was also advised for 180 patients who had either obvious enlargement of the thyroid or large nodules associated with compressive symptoms. Furthermore, eight patients with nodules displaying clinical or ultrasonographic signs suggestive of malignancy also underwent thyroidectomy. Thyroid ultrasonography Before surgery, thyroid ultrasonography was performed in all patients. Ultrasonography was performed using a 13 MHz linear probe (HV-900, Hitachi, Tokyo, Japan). The size of nodule was expressed in its greatest dimension. The following ultrasonographic parameters were assessed:
81
nodule margin (clear or unclear), nodule shape (regular or irregular), echogenicity (hyperechoic, isoechoic, hypoechoic), nodule composition (completely solid, predominantly solid, mixed solid and cystic, predominantly cystic); homogeneity (homogeneous or heterogeneous); calcification (none, microcalcification, dense calcification, or rim calcification); and nodular vascularity (none, perinodular, intranodular, mixed vascularity). Ultrasonography was performed by a single operator and confirmed by another expert ultrasonographer. Preoperative treatment and histopathological diagnosis Before operation, the patients were treated with methimazole or propylthiouracil. Lugol solution was also given for 7 days before surgical intervention. All patients had total or near-total thyroidectomy. The criteria for surgical intervention were as follows: patient preference, failure of antithyroid drugs, failure of radioactive iodine treatment, obvious goiter with compressive symptoms, and severe ophthalmopathy. Tumor tissues (formalin-fixed and paraffin-embedded) were stained with hematoxylin and eosin. The pathological diagnosis was made by two independent pathologists blindly. Microcarcinoma was identified as carcinoma with diameter \10 mm. Statistical analysis All data are expressed as means ± SD. SPSS software was used to perform statistical analyses (SPSS Inc., Chicago, IL). The frequency distribution of categorical variables between subgroups was compared by the continuity corrected chi square test, Fisher’s exact test, or the Kruskal– Wallis test. Numerical variables between subgroups were compared by unpaired t tests. Logistic regression analysis was performed to determine the factor(s) predicting the presence of nodules or cancer. A p value less than 0.05 was considered statistically significant.
Results Clinical characteristics and ultrasonographic features of thyroid nodules in Graves’ patients Thyroid nodules were detected by ultrasonography in 22.7 % (96/423) of the patients with Graves’ disease (Table 1). The prevalence of nodules, especially multiple nodules, increased significantly with age (p \ 0.05). The prevalence of nodules detected by ultrasonography was slightly higher in women than in men; however, this comparison did not show statistical significance (25.0 %
123
82
World J Surg (2014) 38:80–87
Table 1 Clinical characteristics of 423 patients with Graves’ disease with or without thyroid nodules Diffuse goiter (n = 327)
Solitary nodule (n = 25)
Multiple nodules (n = 71)
p Value
Percentages
77.3 %
5.9 %
16.8 %
\0.05
Age (years)
32.07 ± 10.17
39.00 ± 13.01
46.41 ± 13.1
\0.001
Gender (M/F)
75/252
2/23
10/61
NS
Incidence of thyroid carcinoma
12 (3.6 %)
14 (56 %)
32 (45 %)
\0.001
NS not significant Table 2 Characteristics of begin and malignant nodules in patients with Graves’ disease Solitary nodule (n = 25) Benign (n = 11)
Multiple nodules (n = 71) Malignant (n = 14)
Benign (n = 39)
Malignant (n = 32)
Age (years)
44.09 ± 13.59
35.00 ± 11.46
48.08 ± 13.99
43.38 ± 11.84
Gender (M/F)
0/11
2/12
9/30*
1/31
Duration of Graves’ disease (years) B1
2
3
11
16
1–5
5
5
13
5
5–10
2
1
9
3
C10
1
3
7
8
B1
4
5
4
4
1–2
4
4
15
11
C2
2
5
26
17
Diameter (cm)
* p \ 0.05 versus malignant multiple nodules
[84/336] vs 13.8 % [12/87]; p [ 0.05.). Of 96 patients with thyroid nodules, 25 (26.0 %) had a solitary nodule, and 71 (74.0 %) had multiple nodules. In logistic regression analysis performed using patients age, gender, disease duration, and goiter diameter as independent variables, only age was a significant factor predicting presence of solitary and multiple thyroid nodule(s). In patients with Graves’ disease, the total incidence of thyroid cancer was 13.7 % (58/423). Among the 58 patients with thyroid cancer, 12 had diffuse goiter; that is to say, thyroid carcinomas occurred in 12 Graves’ disease patients without ultrasonographically identified nodules. In the other 46 patients with thyroid cancers, thyroid nodule(s) were detected. The incidence of incidentally discovered thyroid cancer was 3.6 % (12/327) in patients with diffuse goiter. Thyroid cancer was significantly more frequent in solitary nodules (14/25, 56.0 %) than in multiple nodules (32/71, 45.1 %; p \ 0.0001). Factor(s) predicting presence of cancer In order to determine the factors predictive of thyroid cancer in Graves’ patients with nodules, we further compared the clinical characteristics between those with cancer and without (Table 2). A multiple logistic regression
123
analysis analyzing age, gender, and type of nodularity was applied to determine the independent risk predictors of thyroid cancer based on the histology. This analysis did not show any significant difference in identifying thyroid cancer in solitary nodules and multiple nodules. At the same time, the disease duration and nodule diameter did not help predict the presence of thyroid cancer, whether in patients with solitary nodules or multiple nodules. Thyroid cancer histology was only positively associated with the female gender in malignant multiple nodules (p \ 0.05). Clinical and pathological characteristics of thyroid cancer in Graves’ patients Table 3 shows the overall clinical and pathological results of the 58 thyroid cancer patients. Overall, the mean age of patients with thyroid cancer in a solitary nodule was significantly lower than that of patients with cancer in multiple nodules (p \ 0.001). The male/female ratio was 1/11, 2/12, and 1/31, respectively in patients with diffuse goiter, solitary nodule, and multiple nodules. There was not any significance of disease duration or pathological type among these three groups. Thyroid carcinomas could also occur in Graves’ disease patients without ultrasonographically identified nodules. Of
World J Surg (2014) 38:80–87
83
Table 3 Clinical characteristics of 58 thyroid carcinomas in Graves’ patients Thyroid carcinoma
p Value
Diffuse goiter (n = 12)
Solitary nodule (n = 14)
Multiple nodules (n = 32)
Age (years)
36.33 ± 9.96
35.00 ± 11.46
44.38 ± 11.84
\0.001
Gender (M/F)
1/11
2/12
1/31
NS (0.308)
Duration of Graves’ disease (years) B1 3 (25.0 %)
5 (35.7 %)
16 (50.0 %)
–
NS (0.690)
1–5
5 (41.7 %)
4 (28.6 %)
5 (15.6 %)
–
5–10
1 (8.3 %)
3 (21.4 %)
3 (9.4 %)
–
3 (25.0 %)
2 (14.3 %)
8 (25.0 %)
–
Papillary
11 (91.7 %)
14 (100.0 %)
31 (96.9 %)
NS (0.497)
Follicular
0 (0.0 %)
0(0.0 %)
1 (3.1 %)
–
Undifferentiated
1 (8.3 %)
0(0.0 %)
0 (0.0 %)
–
Microcarcinoma (cases)
10 (83 %)
10 (71.4 %)
30 (93.75 %)
NS (0.372)
Metastasis (cases)
2 (17 %)
4 (28.6 %)
2 (6.25 %)
0.037
C10 Pathological type
NS not significant
the 58 thyroid cancer patients in the present study, 12 were diagnosed with incidentally discovered thyroid carcinomas without nodules. Ten of these 12 patients had papillary microcarcinomas with no invasion of surrounding tissues or distant metastases. In the other two patients with incidentally discovered thyroid cancer, the disease showed locally advanced characteristics. As for pathological results, 56 patients (96.6 %) were diagnosed with papillary thyroid carcinomas, one with follicular thyroid carcinoma, and 1 with undifferentiated carcinoma. The pathological results in 50 patients (86.2 %) revealed microcarcinomas with a diameter of 10 mm or less. The other eight patients had locally advanced thyroid cancer, two with diffuse goiter, four with solitary nodules, and the other two with multiple nodules. Patients with solitary nodules had a higher percentage of metastasis than those with diffuse goiter or multiple nodules.
hypoechoic echogenicity was lower in Graves’ patients with benign nodules than in those with malignant nodules; however, there was no difference in echogenicity between the two groups. With regard to calcification pattern, 30.4 % Graves’ patients with malignant nodules had ultrasonographic evidence of microcalcification, rim calcification, or dense calcification, with an incidence of calcification significantly higher than that of benign nodules (6.0 %, p \ 0.05). The frequencies of no nodular blood flow and intranodular blood flow in patients with benign nodules were 58.0 and 24.0 %, respectively, with statistical significance compared with the frequencies of those in malignant nodules (30.4 and 56.5 %, respectively; p \ 0.05).
Ultrasonographic features predicting benign and malignant nodules associated with Graves’ disease
Among the 58 patients with thyroid carcinoma, eight had evidence of distant metastasis. The clinical characteristics and histological features are listed in Table 5. All eight were women. Notably, with the exception of a 69-year-old woman, the other seven patients were all young, with a mean age of 29.4 ± 7.15 years (range: 13–34 years). Spearman analysis showed that young age was associated with the risk of metastatic thyroid carcinoma (r = -0.298, p \ 0.05). In the patients in the present study six of the eight metastatic tumors were papillary carcinomas, one of the follicular variant, and one case of the undifferentiated phenotype. Six tumors showed microscopic extrathyroid invasion, one exhibited thyroid capsule invasion, and one showed vascular invasion. Four patients had solitary
We further compared the ultrasonographic results of benign and malignant nodules in Graves’ patients (Table 4). Regular shape was seen in about 98.0 % and 93.5 % of benign and malignant nodules. The frequencies of clear border in patients with benign and malignant nodules were 72.0 and 67.4 %, respectively. Heterogeneous echogenicity was seen in about 90.0 % of benign nodules and in about 69.6 % of malignant nodules, with statistical difference. Most of the benign and malignant nodules in Graves’ patients were composed of completely or predominantly solid elements. The frequency of
Clinical and histological characteristics of Graves’ disease patients with metastatic thyroid carcinoma
123
84
World J Surg (2014) 38:80–87
Table 4 Ultrasonographic features predicting benign and malignant nodules associated with Graves’ disease Ultrasonographic findings Shape of nodule
Type
Benign nodules (n = 50)
Malignant nodules (n = 46)
Regular
49 (98.0 %)
43 (93.5 %)
Irregular
1 (2.0 %)
3 (6.5 %)
p Value NS (0.551) –
Border of nodule
Clear
36 (72.0 %)
31 (67.4 %)
Unclear
14 (28.0 %)
15 (32.6 %)
–
Homogeneity
Homogeneous
5 (10.0 %)
14 (30.4 %)
0.012
Heterogeneous
45 (90.0 %)
32 (69.6 %)
Composition
Completely solid
26 (52.0 %)
31 (67.4 %)
Predominantly solid
22 (44.0 %)
15 (32.6 %)
–
2 (4.0 %)
0 (0.0 %)
–
Mixed solid and cystic Predominantly cystic
0 (0.0 %)
0 (0.0 %)
34 (68.0 %) 3 (6.0 %)
33 (71.7 %) 4 (8.7 %)
NS (0.623)
– NS (0.100)
–
Echogenicity
Hypoechoic Isoechogenic Hyperechoic
13 (26.0 %)
9 (19.6 %)
–
Calcification
None
47 (94.0 %)
32 (69.6 %)
0.002
Microcalcification
3 (6.0 %)
10 (21.8 %)
–
Dense calcification
0 (0.0 %)
2 (4.3 %)
–
Rim calcification Vascularity
0(0.0 %)
2 (4.3 %)
None
29 (58.0 %)
13 (28.3 %)
0.049
Intranodular
12 (24.0 %)
27 (58.7 %)
–
Perinodular
3 (6.0 %)
2 (4.3 %)
–
Mixed vascularity
6 (12.0 %)
4 (8.7 %)
–
nodules, two had multiple nodules, and another two showed diffuse goiter. The nodules were all [2 cm in diameter. Ultrasonographic features showed that all eight cancers demonstrated completely or predominantly solid composition and hypoechoic echogenicity. In three patients the lesions had an ill-defined border and intranodular blood flow. Dense calcification was noted in one patient with diffuse goiter. Four patients showed calcification pattern in thyroid nodules.
Discussion In this retrospective study, we determined the prevalence of thyroid cancer in Graves’ disease patients, and also assessed the clinical and ultrasonographic characteristics of thyroid cancers in those patients. Thyroid nodules are commonly found in subjects with Graves’ disease. In fact, it is reported that the prevalence of thyroid nodules is higher in those with Graves’ disease than in the normal population when detected on palpation or via imaging modalities [2]. The prevalence of palpable thyroid nodules varies between 25 and 45 % in Graves’ disease patients [2, 16, 17]. In some cases, Graves’ disease itself may cause thyroid nodularity [18].
123
NS (0.613) –
–
In this study, thyroid nodules were detected by ultrasonography in 22.7 % of the Graves’ disease patients. Nodules with a diameter more than 1 cm were found in 33.6 % of these patients. Benign and malignant lesions often coexist in thyroid disease, such as Hashimoto’s thyroiditis and Graves’ disease. In recent years, the coexistence of thyroid cancer and Graves’ disease has been of great concern. Sometimes an unsuspected thyroid cancer was discovered incidentally on pathologic examination in a patient with Graves’ disease [5]. In our study, incidental thyroid carcinoma was detected in 12 of 327 Graves’ patients with diffuse goiter, with a frequency less than 4 %. Papillary microcarcinoma was diagnosed in ten of these 12 patients, with no invasion of surrounding tissues and no lymph node or distant metastases. It had been reported that incidental carcinomas in Graves’ disease patients were of no clinical consequence [5]; however, two patients with incidentally discovered thyroid cancer in our report had evidence of distant metastasis. Compared with a nodule in a euthyroid patient, a thyroid nodule diagnosed in a Graves’ disease patient has a higher risk of malignancy. In fact, the incidence of malignancy in the nodules of Graves’ disease patients is common. In our study, the total incidence of thyroid carcinoma in surgically treated Graves’ disease patients was 13.7 %. Furthermore,
None MicroHypoechoic Homogeneity Clear 25.0 9 39.0 Solitary Papillary 8 27 8
F
–
Intranodular Micro-
Dense–
Hypoechoic Heterogeneous
– –
Unclear
Completely solid Completely solid
– –
31.8 9 21.0 Multiple
Diffuse goiter Papillary
Papillary 8
1
69 7
F
16 6
F
None None Hypoechoic Heterogeneous Clear Predominantly solid 36.0 9 35.0 Solitary Papillary 8 27 5
F
Intranodular DenseHypoechoic Heterogeneous Unclear Predominantly solid 31.0 9 16.0 Solitary Papillary 1 13 4
F
Intranodular MicroHypoechoic Homogeneity Unclear Predominantly solid 30.2 9 21.0 Multiple Follicular 4 26 3
F
– – – – – – – Diffuse goiter Undifferentiated 3 34 2
F
Perinodular None Hypoechoic Homogeneity Clear Completely solid 24.6 9 10.1 Solitary Papillary 5 23 1
F
Margin of nodule Composition Nodules size (mm) Nodule number Pathology Duration (years) Age
Gender
85
Patient
Table 5 Clinical characteristics and histological features of Graves’ disease patients with metastatic thyroid carcinoma
Homogeneity
Echogenicity
Calcification
Vascularity
World J Surg (2014) 38:80–87
it has been reported that thyroid carcinoma is more frequent in solitary nodules than in multiple nodules. In our Graves’ patients with solitary thyroid nodules and multiple nodules, the incidence of malignancy reached 56 and 45 %, respectively. This result is in close agreement with other studies [19], whereas other authors [20–22] have reported that there is no difference in cancer prevalence between patients with solitary thyroid nodules and multiple nodular goiters. Criteria in the selection of Graves’ disease patients for thyroidectomy vary from time to time, which may explain the differences between studies carried out in different eras. Fifty years ago, thyroidectomy was the most frequent therapy for Graves’ disease; thereafter, anti-thyroid drugs and radioiodine therapy were the treatments of choice, and they were replaced in part with the traditional method of thyroidectomy [23]. In more recent years, patients with serious signs of Graves’ disease undergo surgical treatment, thus may not represent the whole population of Graves’ disease patients [1]. In the present study, many patients enrolled experienced a relatively longer disease duration, which may have led to the high incidence of nodule formation and the presence of thyroid carcinoma. An inverse relationship was observed between the risk of thyroid carcinoma and age. Patients with papillary thyroid cancer were younger. This finding is in agreement with Mittendorf’s research [24], in whose study the mean age was 35 and 43.3 years in patients with solitary nodules and multiple nodules, respectively. Our study demonstrated that male gender did not confer a higher risk of cancer in Graves’ patients, a finding that is not in agreement with other studies [25–27]. Perhaps this can be attributed to the finding that the prevalence of multiple nodules is less frequent in men than in women. We detected a high incidence of carcinoma in nodules from patients with Graves’ disease in the present study, and found that papillary thyroid cancer was the most common histologic subtype. More important, up to 84.5 % of all cases of papillary thyroid carcinoma were papillary microcarcinomas, as defined by the World Health Organization’s classification of thyroid cancer. It is reported that carcinoma in Graves’ disease may have a bad outcome, such as distant metastasis and patient death. Therefore, thyroidectomy is recommended in patients with the diagnosis of papillary microcarcinoma [28]. However, the progression and prognosis of thyroid cancer in Graves’ disease patients is a matter of debate. In the present study, eight thyroid carcinoma patients showed metastatic characteristics. Age may be the only factor predicting locally advanced cancers in Graves’ patients. With the exception of a single 69-year-old patient, the carcinoma patients were all younger, with a 13-year-old with a 1 year history of Graves’ disease the youngest.
123
86
It was once suggested that patients with Graves’ disease should be evaluated by ultrasonography for detection of thyroid nodules, and that thyroidectomy should be performed for control of hyperthyroidism and removal of potentially malignant nodules [16]. In fact, for all the Graves’ disease patients, conventional thyroidectomy for removal of nodules seems not to be appropriate in light of the surgical risks. Our study shows that younger subjects may have higher risk of thyroid carcinoma. This indicates that suspicious nodule detected in patients with Graves’ disease should be evaluated carefully, especially in younger patients. Malignant nodules usually had the following pathological features: hypoechogenicity, ill-defined borders, absence of a peripheral hypoechoic rim, no cystic changes, increased intranodular blood flow, and punctate calcifications. However, only heterogeneous echogenicity, calcification, and vascularity showed significant differences between benign and malignant nodules in Graves’ patients in the present study. Calcifications were more frequently detected in malignant nodules than in benign nodules. Frates et al. [29, 30] reported that thyroid carcinomas had increased nodular blood flow. The research of Cerbone et al. [31] showed that the large degree of cellular proliferation in malignant nodules could explain the dominant intranodular blood flow. In our series we found that malignant nodules, as compared to benign nodules, in patients with Graves’ disease tended to exhibit significantly increased intranodular blood flow. In conclusion, in this retrospective study, the prevalence of thyroid cancer in Graves’ disease was about 13.7 %, and the incidence of nodules in malignant Graves’ disease even reached 47.9 %. Compared to Graves’ disease patients with diffuse goiter, those with thyroid nodules are at higher risk of developing thyroid cancer. Although there is a high risk of carcinomas in Graves’ disease patients with nodules, most of them are micropapillary thyroid cancers. Age is one of the most important factors predictive of locally advanced cancers in Graves’ patients. According to the potential metastatic characteristics of the nodules, every suspicious nodule in Graves’ disease should be evaluated and treated seriously, especially in younger patients. As a kind of noninvasive and time-saving method, ultrasonography should be used routinely to find potential malignancies in Graves’ disease. Graves’ patients with calcification and intranodular blood flow in ultrasonographic findings are more likely to harbor thyroid cancers. Funding This work was supported in part by a grant from the National Natural Science Foundation of China (81000317) and by a Yat-Sen Scholarship for Young Scientist. Conflict of interest All authors declare that no conflicts of interest exist regarding the research reported.
123
World J Surg (2014) 38:80–87
References 1. Gerenova J, Buysschaert M, de Burbure CY et al (2003) Prevalence of thyroid cancer in Graves’ disease: a retrospective study of a cohort of 103 patients treated surgically. Eur J Intern Med 14:321–325 2. Belfiore A, Garofalo MR, Giuffrida D et al (1990) Increased aggressiveness of thyroid cancer in patients with Graves’ disease. J Clin Endocrinol Metab 70:830–835 3. Angusti T, Codegone A, Pellerito R et al (2000) Thyroid cancer prevalence after radioiodine treatment of hyperthyroidism. J Nucl Med 41:1006–1009 4. Gul K, Di Ri Koc A, Ki Yak G et al (2009) Thyroid carcinoma risk in patients with hyperthyroidism and role of preoperative cytology in diagnosis. Minerva Endocrinol 34:281–288 5. Phitayakorn R, McHenry CR (2008) Incidental thyroid carcinoma in patients with Graves’ disease. Am J Surg 195:292–297. discussion 297 6. Hales IB, McElduff A, Crummer P et al (1992) Does Graves’ disease or thyrotoxicosis affect the prognosis of thyroid cancer. J Clin Endocrinol Metab 75:886–889 7. Ozaki O, Ito K, Kobayashi K et al (1990) Thyroid carcinoma in Graves’ disease. World J Surg 14:437–440. doi:10.1007/BF01 658550. discussion 440–431 8. Chao TC, Lin JD, Jeng LB et al (1999) Thyroid cancer with concurrent hyperthyroidism. Arch Surg 134:130–134 9. Okamoto T, Iihara M, Obara T (2000) Management of hyperthyroidism due to Graves’ and nodular diseases. World J Surg 24:957–961. doi:10.1007/s002680010167 10. Terzioglu T, Tezelman S, Onaran Y et al (1993) Concurrent hyperthyroidism and thyroid carcinoma. Br J Surg 80:1301– 1302 11. Kim WB, Han SM, Kim TY et al (2004) Ultrasonographic screening for detection of thyroid cancer in patients with Graves’ disease. Clin Endocrinol (Oxf) 60:719–725 12. Bruneton JN, Balu-Maestro C, Marcy PY et al (1994) Very high frequency (13 MHz) ultrasonographic examination of the normal neck: detection of normal lymph nodes and thyroid nodules. J Ultrasound Med 13:87–90 13. Hegedus L (2004) Clinical practice. The thyroid nodule. N Engl J Med 351:1764–1771 14. Papini E, Guglielmi R, Bianchini A et al (2002) Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab 87:1941–1946 15. Chan BK, Desser TS, McDougall IR et al (2003) Common and uncommon sonographic features of papillary thyroid carcinoma. J Ultrasound Med 22:1083–1090 16. Berker D, Isik S, Ozuguz U et al (2011) Prevalence of incidental thyroid cancer and its ultrasonographic features in subcentimeter thyroid nodules of patients with hyperthyroidism. Endocrine 39:13–20 17. Mishra A, Mishra SK (2001) Thyroid nodules in Graves’ disease: implications in an endemically iodine deficient area. J Postgrad Med 47:244–247 18. Kraimps JL, Bouin-Pineau MH, Mathonnet M et al (2000) Multicentre study of thyroid nodules in patients with Graves’ disease. Br J Surg 87:1111–1113 19. Carnell NE, Valente WA (1998) Thyroid nodules in Graves’ disease: classification, characterization, and response to treatment. Thyroid 8:647–652 20. Christensen SB, Ljungberg O, Tibblin S (1984) Thyroid carcinoma in Malmo, 1960–1977. Epidemiologic, clinical, and prognostic findings in a defined urban population. Cancer 53:1625–1633
World J Surg (2014) 38:80–87 21. Belfiore A, La Rosa GL, La Porta GA et al (1992) Cancer risk in patients with cold thyroid nodules: relevance of iodine intake, sex, age, and multinodularity. Am J Med 93:363–369 22. McCall A, Jarosz H, Lawrence AM et al (1986) The incidence of thyroid carcinoma in solitary cold nodules and in multinodular goiters. Surgery 100:1128–1132 23. Wartofsky L, Glinoer D, Solomon B et al (1991) Differences and similarities in the diagnosis and treatment of Graves’ disease in Europe, Japan, and the United States. Thyroid 1:129–135 24. Mittendorf EA, Khiyami A, McHenry CR (2006) When fineneedle aspiration biopsy cannot exclude papillary thyroid cancer: a therapeutic dilemma. Arch Surg 141:961–966. discussion 966 25. Rojeski MT, Gharib H (1985) Nodular thyroid disease. Evaluation and management. N Engl J Med 313:428–436 26. Gharib H (1994) Fine-needle aspiration biopsy of thyroid nodules: advantages, limitations, and effect. Mayo Clin Proc 69:44–49
87 27. Rago T, Fiore E, Scutari M et al (2010) Male sex, single nodularity, and young age are associated with the risk of finding a papillary thyroid cancer on fine-needle aspiration cytology in a large series of patients with nodular thyroid disease. Eur J Endocrinol 162:763–770 28. Belfiore A, Russo D, Vigneri R et al (2001) Graves’ disease, thyroid nodules and thyroid cancer. Clin Endocrinol (Oxf) 55:711–718 29. Taylor KJ, Ramos I, Carter D et al (1988) Correlation of Doppler US tumor signals with neovascular morphologic features. Radiology 166:57–62 30. Frates MC, Benson CB, Doubilet PM et al (2003) Can color Doppler sonography aid in the prediction of malignancy of thyroid nodules? J Ultrasound Med 22:127–131 quiz 132–124 31. Cerbone G, Spiezia S, Colao A et al (1999) Power Doppler improves the diagnostic accuracy of color Doppler ultrasonography in cold thyroid nodules: follow-up results. Horm Res 52:19–24
123
