World J Surg (2014) 38:68–79 DOI 10.1007/s00268-013-2224-1

Thyroid Lobectomy for Papillary Thyroid Cancer: Long-term Follow-up Study of 1,088 Cases Kenichi Matsuzu • Kiminori Sugino • Katsuhiko Masudo • Mitsuji Nagahama • Wataru Kitagawa • Hiroshi Shibuya • Keiko Ohkuwa • Takashi Uruno • Akifumi Suzuki • Syunsuke Magoshi • Junko Akaishi • Chie Masaki • Michikazu Kawano • Nobuyasu Suganuma • Yasushi Rino • Munetaka Masuda Kaori Kameyama • Hiroshi Takami • Koichi Ito

•

Published online: 1 October 2013 Ó Socie´te´ Internationale de Chirurgie 2013

Abstract Background Total thyroidectomy is well accepted as initial surgery for papillary thyroid cancer (PTC), but the extent of the thyroidectomy remains a matter of controversy. This study was designed to investigate the long-term clinical outcome of PTC patients who had undergone thyroid lobectomy and to elucidate the indications of lobectomy as initial surgery. Methods The cases of 1,088 PTC patients who underwent thyroid lobectomy with curative intent at Ito Hospital between 1986 and 1995 were analyzed retrospectively in this study. None of the patients had received postoperative radioactive iodine (RAI) ablation therapy. The median follow-up period was 17.6 years. All clinical outcomes,

including recurrence and death as a result of PTC or other reasons, were evaluated. To establish the indications for lobectomy as initial surgery for PTC, the potential risk factors, such as age, sex, primary tumor size, extrathyroidal invasion, and clinical lymph node metastasis at the time of the initial surgery, were assessed statistically for associations with recurrence and disease-related death. Results The remnant-thyroid recurrence-free survival (RT-RFS) rate, the regional- lymph-node recurrence-free survival (L-RFS) rate, and the distant-recurrence-free survival (D-RFS) rate as of 25 years after surgery were 93.5, 90.6, and 93.6%, respectively. The cause-specific survival (CSS) rate at 25 years was 95.2%. Univariate and multivariate analyses showed that none of the factors assessed were significantly associated with the RT-RFS rate. Tumor

K. Matsuzu (&)  K. Sugino  M. Nagahama  W. Kitagawa  H. Shibuya  K. Ohkuwa  T. Uruno  A. Suzuki  S. Magoshi  J. Akaishi  C. Masaki  M. Kawano  H. Takami  K. Ito Department of Surgery, Ito Hospital, 4-3-6 Jingumae, Shibuya-ku, Tokyo 150-8308, Japan e-mail: [email protected]

S. Magoshi e-mail: [email protected]

K. Sugino e-mail: [email protected] M. Nagahama e-mail: [email protected] W. Kitagawa e-mail: [email protected] H. Shibuya e-mail: [email protected] K. Ohkuwa e-mail: [email protected] T. Uruno e-mail: [email protected] A. Suzuki e-mail: [email protected]

123

J. Akaishi e-mail: [email protected] C. Masaki e-mail: [email protected] M. Kawano e-mail: [email protected] H. Takami e-mail: [email protected] K. Ito e-mail: [email protected] K. Masudo  N. Suganuma  Y. Rino  M. Masuda Department of Surgery, Yokohama City University Hospital, 3-9 Fukuura, Kanazawa-ku, Yokohama, Kanagawa 236-0004, Japan e-mail: [email protected] N. Suganuma e-mail: [email protected]

World J Surg (2014) 38:68–79

size, clinical lymph node metastasis, and extrathyroidal invasion were significantly associated with the L-RFS rate. The D-RFS and CSS rates were both significantly lower in the group of patients who were aged 45 years old or older, the group whose tumors were larger than 40 mm, and the group with extrathyroidal invasion. Based on the above findings, we classified the patients into four groups according to age \45 or C45 years, tumor size B40 or [40 mm, whether clinical lymph node metastasis was present, and whether extrathyroidal invasion was present. None of the patients without any of these four risk factors died of PTC. On the other hand, 22 patients who died of PTC were positive for one or more of these four factors. Conclusions The long-term clinical outcome of the PTC patients who had been treated by lobectomy without RAI ablation was excellent. Based on the above results, we concluded that lobectomy is a valid alternative to total thyroidectomy for the treatment of PTC patients who are younger than aged 45 years, whose tumor diameter is 40 mm or less, and who do not have clinical lymph node metastasis or extrathyroidal invasion.

Introduction The incidence of thyroid cancer has been increasing sharply in both the United States and Japan [1, 2], and the most common subtype of thyroid cancer in both countries is papillary thyroid cancer (PTC) [3, 4]. Guidelines published in the United States and European countries recommend total or near-total thyroidectomy plus postoperative radioactive iodine (RAI) ablation as the standard initial treatment for PTC [5–8]. However, because most PTC patients have an excellent outcome, both the extent of the thyroidectomy required and the criteria of application for RAI ablation remain matters of controversy. In contrast to western countries, thyroid lobectomy has been the mainstay of primary surgery for PTC in Japan, and total thyroidectomy is reserved for advanced cases [2]. Consequently, a large volume of data on PTC patients who have been treated by lobectomy is available in Japan. The present study was designed to document one institution’s experience regarding the long-term clinical outcomes of PTC patients who had been treated by lobectomy and to evaluate the validity of lobectomy as an Y. Rino e-mail: [email protected]

69

initial procedure. We attempted to identify risk factors for recurrence and disease-related death to be able to use any risk factors that were identified as criteria for establishing the indications for lobectomy as an initial surgical treatment for PTC.

Materials and methods Patients The medical records of 2,036 consecutive PTC patients who underwent an initial surgery at Ito Hospital between 1986 and 1995 were reviewed retrospectively. All of the information used in this study, including the patient characteristics, operative findings, postoperative treatment, and follow-up, was collected from the patients’ original medical records. Reliable information about a familial history of thyroid cancer or radiation exposure was not available in the records. Tumor size was measured in the surgical specimen. Extrathyroidal invasion, corresponding to class T4a in the UICC TNM classification, was diagnosed by surgeons based on the macroscopic intraoperative findings. Because preoperative ultrasonography (US) screening for lymph node metastasis was not performed in all of the cases during the period of this study, clinical lymph node metastasis had been primarily assessed by preoperative palpation. Multicentricity was assessed based on the microscopic findings and was recorded as positive when multiple foci were detected in the resected lobe. Because patients with intrathyroidal metastasis to the contralateral lobe were treated by total thyroidectomy, such patients were not included in this study. Histological sections of the surgical specimens were reexamined by one of the authors (K.K.), and the diagnosis of classic PTC based on the recent World Health Organization classification system was confirmed in each case [9]. Patients with a follicular variant, an oxyphilic cell variant, a diffuse sclerosing variant, a tall cell variant, poorly differentiated thyroid cancer, or undifferentiated cancer were excluded from the analysis. In accordance with the surgical strategy described below, 1,223 of the 2,036 patients were treated by lobectomy with curative intent. After excluding the 132 patients (10.8%) who failed to return for a follow-up examination within 5 years after surgery, the cases of the remaining 1,088 PTC patients were analyzed in this study. The median follow-up period was 17.6 years, and the total follow-up time was more than 19,140 years.

M. Masuda e-mail: [email protected] K. Kameyama Division of Diagnostic Pathology, Keio University Hospital, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan e-mail: [email protected]

Surgical strategy Although the major objective of the initial surgery was macroscopically complete resection of the tumor, our

123

70

surgical strategy during this period was unique and differed from most of the current surgical strategies for PTC. When a tumor was restricted to one lobe and there was no evidence of intrathyroidal metastasis in the contralateral lobe, ipsilateral lobectomy was performed. Patients with multiple tumors in a single lobe also were treated by lobectomy, as long as all of the tumors could be removed. Total thyroidectomy was performed in patients with a tumor that involved both lobes or who had at least one tumor in each lobe. Whether clinical lymph node metastasis was present at the time of the operation did not affect the extent of the thyroidectomy, but patients with distant metastasis underwent total thyroidectomy. Whenever the diagnosis of PTC had been made preoperatively, thyroidectomy was followed, in principle, by routine dissection of the lymph nodes in the central compartment (level VI) and the lateral compartment (level II–V) regardless of whether there was any evidence of lymph node metastasis. When a tumor was suspected of being PTC but the results of the cytology were not determined, only the central compartment was dissected, and dissection of the lateral compartment was omitted. Lymph node dissection was not performed in patients with a preoperative diagnosis of benign disease. Postoperative treatment and follow-up None of the patients received postoperative RAI ablation therapy. TSH suppression therapy was performed in most of the cases postoperatively, but the patients’ TSH levels were not analyzed in this study. Patients underwent a routine US examination every 1 or 2 years postoperatively. Computed tomography (CT) examination of the chest was performed every 5 years postoperatively to detect lung metastasis. The serum thyroglobulin (Tg) level was measured every 6 months postoperatively. Because the serum Tg level is affected by the remnant thyroid gland, the absolute serum Tg level was not regarded as important. Whenever a persistent increase in serum Tg level was observed postoperatively, the patient was carefully examined by US and CT for evidence of a recurrence. Whenever a recurrence in the remnant thyroid gland or a regional lymph node was diagnosed, the diagnosis was confirmed by aspiration cytology or by histological examination after reoperation. Distant metastasis was diagnosed, in principle, by CT and RAI scintigraphy after completion total thyroidectomy. The date and cause of death of the patients who died were confirmed by checking their medical records or by sending an inquiry to the referred hospital. Statistical analysis The data were statistically analyzed by using a statistical software program (JMP 8.0; SAS Institute, Inc., Cary, NC).

123

World J Surg (2014) 38:68–79

The cumulative remnant-thyroid recurrence-free survival (RT-RFS) rate, regional-lymph-node recurrence-free survival (L-RFS) rate, distant-recurrence-free survival (DRFS) rate, and cumulative cause-specific survival (CSS) rate were calculated. To establish the indications for lobectomy, five potential risk factors for recurrence or disease-related death, i.e., age, sex, primary tumor size, extrathyroidal invasion, and clinical lymph node metastasis at the time of surgery, were subjected to univariate analyses by the Kaplan–Meier method and log-rank test and to multivariate analyses using the Cox proportional hazard model [10–17]. To adjust for the UICC staging system, the thresholds for age and tumor size for the statistical analyses were set at 45 years old (\45 vs. C45 years) and 40 mm (B40 vs. [40 mm), respectively. Because the lymph node dissection levels were not the same in every patient and the differences may have affected the clinical outcomes, the dissection levels also were analyzed statistically for associations with recurrence and disease-related death. The significance level in all of the statistical tests performed in this study was set at P \ 0.05. The factors that were identified as significant by the univariate and multivariate analyses are reported below with the hazard ratios. The protocol of the present study was reviewed and approved by our Institution’s Ethics Board, and the study was conducted in accordance with the declaration of Helsinki.

Results Patient characteristics The patients’ demographic and tumor characteristics are shown in Table 1. There were 109 men (10%) and 979 women (90%). Patient age at the time of surgery ranged from 7 to 81 years, and their median age was 46 years. Tumor size ranged from 1 mm, in a tumor that was diagnosed as an incidental finding during histological examination of the surgical specimen obtained during surgery of a benign thyroid disease, to 100 mm, and the median primary tumor diameter was 22 mm. It is noteworthy that the primary tumor in 85.7% of the patients measured more than 10 mm in diameter, and these patients would have been treated by total thyroidectomy in western countries. Extrathyroidal invasion was detected in 84 patients (7.7%). Based on the postoperative histological examination, 929 of the 1,088 subjects had a solitary tumor, and the other 159 patients (14.6%) had multiple tumors in the resected lobe. The PTC patients with intrathyroidal metastasis in the contralateral lobe had been treated by total thyroidectomy and were not included in the study population. At the time of the operation, 69 patients (6.3%) had clinical lymph node metastasis. In accordance with the surgical strategy described above, both central and

World J Surg (2014) 38:68–79

71

lateral compartment lymph node dissection was performed in 898 (82.5%) patients, regardless of whether there was evidence suggesting lymph node metastasis. Lymph node dissection in the central compartment alone was performed in the 60 patients (5.5%) whose tumor was suspected of being a PTC, but in whom the diagnosis had not been confirmed. In most of the 130 (11.9%) cases in which lymph node dissection was not performed the preoperative diagnosis had been a benign tumor and the postoperative histopathological diagnosis was PTC. It is noteworthy that histological examination of the dissected lymph nodes resulted in a pathological diagnosis of lymph node metastasis in 775 patients, i.e., in 86.3% of the patients in whom lymph node dissection was performed.

cumulative RT-RFS rates at 10, 15, 20, and 25 years after surgery were 97.4, 95.8, 94.6, and 93.5%, respectively (Fig. 1a), and the corresponding remnant thyroid recurrence rates at 10, 15, 20, and 25 years after surgery were 2.6, 4.2, 5.4, and 6.5%, respectively. None of the five potential risk factors that were assessed were found to be significantly associated with a lower RT-RFS rate according to the results of the univariate (Fig. 1b–e) and multivariate analyses (data not shown), meaning that these five factors cannot be used to estimate the risk of recurrence in the remnant thyroid gland. Nor were there any significant associations between the regional lymph node dissection levels and the RT-RFS rates (Fig. 1f). Recurrence in regional lymph nodes

Recurrence in the remnant thyroid gland

[20 to B40

456 (41.9)

[40

109 (10)

Regional lymph node recurrence was diagnosed in 92 of the 1,088 subjects. The cumulative L-RFS rates at 10, 15, 20, and 25 years after surgery were 94.6, 91.8, 91.0, and 90.6%, respectively (Fig. 2a), and the corresponding regional lymph node recurrence rates at 10, 15, 20, and 25 years after surgery were 5.4, 8.2, 9.0, and 9.4%, respectively. Tumors larger than 40 mm (P \ 0.0077; Fig. 2c), extrathyroidal invasion (P \ 0.0001; Fig. 2d), and clinical lymph node metastasis (P \ 0.0001; Fig. 2e) were significantly associated with a lower L-RFS rate according to both univariate and multivariate analyses (Table 2), whereas sex (P = 0.3203) and age (P = 0.2642; Fig. 2b) were not significantly associated with a lower L-RFS rate. The hazard ratios for lymph node recurrence associated with a tumor larger than 40 mm, extrathyroidal invasion, and clinical lymph node metastasis were 1.8 (P = 0.039), 3.5 (P \ 0.0001), and 3.3 (P \ 0.0001), respectively (Table 2). It is noteworthy that there were no significant differences in the L-RFS rate according to the regional lymph node dissection level (Fig. 2f).

Negative

1004 (92.3)

Distant recurrence

Positive

84 (7.7)

Remnant-thyroid recurrence was detected in 52 of the 1,088 subjects during the period of the study. The Table 1 Patient characteristics Characteristics

No. of patients (%)

Sex Male Female Age (year)

109 (10) 979 (90)

\45

483 (44.4)

C45

605 (55.6)

Serum thyroglobulin (Tg) 372 ± 2461 ng/ml Tumor size (mm) B10

156 (14.3)

[10 to B20

367 (33.7)

Extrathyroidal invasion

Clinical lymph node metastasis Negative Positive Pathological lymph node metastasis

1019 (93.7) 69 (6.3)

Negative

123 (13.7)

Positive

775 (86.3)

Multicentricity in the resected lobe Negative

929 (85.4)

Positive

159 (14.6)

Lymph node dissection None

130 (11.9)

Central compartment

60 (5.5)

Central and lateral compartment

898 (82.5)

Postoperative distant recurrence was diagnosed in 35 of the 1,088 subjects during the follow-up period. The lung was the most frequent site of distant recurrence (n = 31). More specifically, distant recurrence was diagnosed in the lung alone in 27 subjects, bone alone in 3 subjects, the brain alone in 1 subject, both the lung and bone in 3 subjects, and at multiple sites in the lung, bone, and the adrenal gland in 1 subject. The cumulative D-RFS rates at 10, 15, 20, and 25 years after surgery were 98.6, 97.8, 96.3, and 93.6%, respectively (Fig. 3a), and the corresponding distant recurrence rates at 10, 15, 20, and 25 years after surgery were 1.4, 2.2, 3.7, and 6.4%, respectively. Univariate analyses revealed that age 45 years or older (P \ 0.0001), tumor larger than 40 mm (P = 0.015), extrathyroidal

123

RT-RFS rate (%)

72

World J Surg (2014) 38:68–79

a RT-RFS rate

b Age

c Primary tumor size

100

100

100

80

80

60

60 RT-RFS rates 10-Year 97.4% 15-Year 95.8% 20-Year 94.6% 25-Year 93.5%

40 20

80

5

10

20

20

15

20

25

40 mm (n = 979) >40 mm (n = 109)

40

0

0

0

p = 0.5608

60 40 mm (n = 109)

20

20

15

p = 0.0077

40

0

0

80 60

40

0

L-RFS rate (%)

c Primary tumor size 100

0

5

Years after surgery

10

15

20

Years after surgery

25

0

5

10

15

20

25

Years after surgery

Fig. 2 a L-RFS curve for all patients. The 10-, 15-, 20-, and 25-year L-RFS rates were 94.6, 91.8, 91.0, and 90.6, respectively. b–f Univariate analysis of L-RFS rates. Patients with a tumor that was larger than 40 mm, who had extrathyroidal invasion, or who had clinical lymph node metastasis had significantly lower L-RFS rates. However, age and sex did not have a significant effect on the L-RFS rate. b Comparison of L-RFS rates between patients aged 45 years or older and patients younger than aged 45 years. L-RFS rates between

patients whose tumor measured up to 40 mm and those whose tumor was larger than 40 mm. d Comparison of L-RFS rates between patients with and those without extrathyroidal invasion. e Comparison of L-RFS rates between patients with and those without clinical lymph node metastasis at the time of surgery. f Comparison of L-RFS rates between patients with no lymph node dissection, those with central compartment dissection, and those with both central and lateral compartment dissection

Table 2 Risk factors for lower L-RFS rate

that were present: 0-factor group of 381 patients; 1-factor group of 574 patients; 2-factor group of 108 patients; 3-factor group of 23 patients; and 4-factor group of 2 patients. The clinical outcomes of each group are shown in Table 6. The RT-RFS, L-RFS, D-RFS, and CSS were analyzed among the groups in univariate analyses by the Kaplan–Meier method and log-rank test. Because the 4-factor group contained only two patients and a survival curve could not be drawn, the 3-factor group and 4-factor group were combined into a 3- or 4-factor group (n = 25). There were no significant differences in RT-RFS rate between the four groups (P = 0.0665; Fig. 5a). The L-RFS rate was highest in the 0-factor group (92.1% at 25 years) and 1-factor group (93.7% at 25 years), lowest in the 3- or 4-factor group (59.4% at 25 years), and intermediate in the 2-factor group (75.6% at 25 years; Fig. 5b). Similarly, the D-RFS rate was highest in the 0-factor group (99.6% at 25 years) and 1-factor group (92.1%, at 25 years), lowest in the 3- or 4-factor group (62.3% at 25 years), and intermediate in the 2-factor group (86.1% at 25 years; Fig. 5c). The CSS rate also was highest

P value Tumor size [40 mm

HR

95% CI

0.039

1.8

1.0

3.1

Extrathyroidal invasion

\0.0001

3.5

2.1

5.7

Clinical lymph node metastasis

\0.0001

3.3

1.9

5.6

repeat analyses were the same as the results of the original analyses in terms of statistical significance, although there were slight changes in the P values and HR values.

Analysis of cases classified according to the number of four significant risk factors present As shown earlier, age 45 years or older, tumor diameter greater than 40 mm, extrathyroidal invasion, and clinical lymph node metastasis were significant risk factors associated with poor outcomes. All 1,088 subjects were classified into five groups according to the number of these risk factors

123

74

World J Surg (2014) 38:68–79

D-RFS rate (%)

a D-RFS rate

b Age

100

100

80

80

c Primary tumor size 100 p < 0.0001

60

60 D-RFS rates 10-Year 98.6% 15-Year 97.8% 20-Year 96.3% 25-Year 93.6%

40 20

60 < 45 years old (n = 483) 45 years old (n = 605)

40

20 0

0 0

5

10

15

20

25

d Extrathyroidal invasion

0

5

10

15

20

25

e Clinical lymph node metastasis

100

40 mm (n = 979) > 40 mm (n = 109)

40

20

0

p = 0.015

80

0

5

10

15

f Lymph node dissection

D-RFS rate (%)

p < 0.0001

80

60

60

40

p = 0.0012

20

40 negative (n = 1019) positive (n = 69)

20

0 5

10

15

20

25

none (n = 130) central (n = 60) central and lateral (n = 898)

20

0 0

p = 0.0595

80 60

40

negative (n = 1004) positive (n = 84)

25

100

100

80

20

0 0

5

Years after surgery

10

15

20

Years after surgery

25

0

5

10

15

20

25

Years after surgery

Fig. 3 a D-RFS curve for all patients. The 10-, 15-, 20-, and 25-year D-RFS rates were 98.6, 97.8, 96.3, and 93.6%, respectively. b–f Univariate analysis of D-RFS rates. Patients who were aged 45 years or older, whose tumor was larger than 40 mm, who had tumors with extrathyroidal invasion, or who had clinical lymph node metastasis had significantly lower D-RFS rates. b Comparison of D-RFS rates between patients aged 45 years or older and patients younger than aged 45 years. c Comparison of D-RFS rates between patients whose tumor measured up to 40 mm and those whose tumor

was larger than 40 mm. d Comparison of D-RFS rates between patients with and those without extrathyroidal invasion. e Comparison of D-RFS rates between patients with and those without clinical lymph node metastasis at the time of surgery. f Comparison of D-RFS rates between patients with no lymph node dissection, those with central compartment dissection, and those with both central and lateral compartment dissection. No significant differences in the D-RFS rates were observed among the different dissection levels

Table 3 Risk factors for lower D-RFS rate

patient in the 0-factor group was diagnosed with lung metastasis 16.2 years after the initial surgery, she is still alive at 19.1 years after the initial surgery. Ultimately, none of the 0-factor patients died as a result of PTC.

C45 years of age Tumor size [40 mm Extrathyroidal invasion Clinical lymph node metastasis

P value

HR

95% CI

\0.0001

6.4

2.5

0.0314

2.6

1.1

5.6

\0.0001

7.5

3.7

14.9

0.1545

1.9

0.8

4.3

21.8

in the 0-factor group (100% at 25 years) and 1-factor group (95.1% at 25 years), lowest in the 3- or 4-factor group (64.1% at 25 years), and intermediate in the 2-factor group (82.5% at 25 years; Fig. 5d). Patients with a larger number of risk factors usually had a poorer clinical outcome in terms of regional lymph node recurrence, distant recurrence, and disease-related death. The excellent outcomes of the patients in the 0-factor group deserve particular attention. Remnant thyroid recurrence and regional lymph node recurrence developed in 23 patients and 24 patients, respectively, in the 0-factor group, but all of recurrent lesions were completely removed at reoperation. Although a 40-year-old female

123

Discussion Treatment guidelines published in western countries basically recommend total or near total thyroidectomy as the initial surgery for PTC, and although they allow lobectomy for limited ‘‘low-risk cases,’’ the definitions of ‘‘low-risk’’ are not consistent [2, 5–8, 18–21]. The controversy regarding the extent of thyroidectomy required persists, mainly because little is known about the long-term outcomes of PTC patients who have been treated by lobectomy. In the present study, we retrospectively analyzed 1,088 PTC patients who had been treated by unilateral lobectomy and lymph node dissection, in principle. The recurrence rates in the remnant thyroid gland, regional lymph nodes,

World J Surg (2014) 38:68–79

CSS rate (%)

a

75

b

CSS rate

100

100

80

80

60

60 CSS rates 10-Year 99.4% 15-Year 98.5% 20-Year 97.8% 25-Year 95.2%

40 20

p < 0.0001

5

10

60 < 45 years old (n = 483) 45 years old (n = 605)

20

25

20 0 0

e

Extrathyroidal invasion

100

40 mm (n = 979) > 40 mm (n = 109)

40

20

15

p = 0.0076

80

0 0

Primary tumor size

100

40

0

d

c

Age

5

10

15

20

25

0

f

Clinical lymph node metastasis

5

10

15

Lymph node dissection

CSS rate (%)

p < 0.0001 80

60

60

40

p < 0.0001

20

60 40 negative (n = 1019) positive (n = 69)

20

0 5

10

15

20

25

none (n = 130) central (n = 60) central and lateral (n = 898)

20 0

0 0

p = 0.8228

80

40 negative (n = 1004) positive (n = 84)

25

100

100

80

20

0

5

10

15

20

25

0

5

10

15

20

25

Years after surgery

Years after surgery

Years after surgery

Fig. 4 a CSS curve for all patients. The 10-, 15-, 20-, and 25-year CSS rates were 99.4, 98.5, 97.8, and 95.2, respectively. b–f Univariate analysis of CSS. Patients who were aged 45 years or older, whose tumor was larger than 40 mm, who had tumors with extrathyroidal invasion, or who had clinical lymph node metastasis had significantly lower CSS rates. b Comparison of CSS between patients aged 45 years or older and patients younger than aged 45 years. c Comparison of CSS between patients whose tumor measured up to 40 mm

and those whose tumor was larger than 40 mm. d Comparison of CSS between patients with and those without extrathyroidal invasion. e Comparison of CSS between patients with and those without clinical lymph node metastasis at the time of surgery. f Comparison of CSS rates between patients with no lymph node dissection, those with central compartment dissection, and those with both central and lateral compartment dissection. No significant differences in the CSS rates were observed among the different dissection levels

Table 4 Risk factors for lower CSS

Table 5 Summary of the results according to risk factor

C45 years of age Tumor size [40 Extrathyroidal invasion Clinical lymph node metastasis

P value

HR

95% CI

\0.0001

15.6

3.2

Factor

Recurrence in:

281.8

0.0292

3.2

1.1

8

\0.0001

13.8

5.7

34.6

0.1296

2.2

0.8

5.6

HR hazard ratio; CI confidence interval

and distant organs at 25 years after the initial surgery were 6.5, 9.4, and 6.4%, respectively. The cause-specific survival rate at 25 years after surgery was 95.2%. The patient population included some patients whose primary tumor was larger than 40 mm, with extrathyroidal invasion, or with clinical lymph node metastasis. Because PTC patients whose primary tumor was larger than 40 mm, who had extrathyroidal invasion, or had clinical lymph node metastasis have been and would be treated by total thyroidectomy in western countries, the data we obtained are valuable. The data obtained in our study showed that the PTC patients with tumors larger than 40 mm,

CSS

Remnant thyroid

Lymph node

Distant organ

Age (year)

C45

9

9

d

d

Sex

Male

9

9

9

9

Tumor size (mm) Extrathyroidal invasion

[40 Positive

9 9

d d

d d

d d

Clinical lymph node metastasis

Positive

9

d

4

4

d Significant in both the univariate and multivariate analyses; 4 significant in the univariate analysis, but not in the multivariate analysis; 9 not significant

extrathyroidal invasion, or clinical lymph node metastasis had significantly poorer outcomes than the patients who were negative for each of these factors. The results of the following studies support total thyroidectomy as the initial surgery for PTC patients.

123

76

World J Surg (2014) 38:68–79

Table 6 Clinical outcomes classified by numbers of the risk factors

a

Classification

n

Remnant thyroid recurrence

Regional lymph node recurrence

0-factors

381

23

24

1

1-factor

574

23

33

15

6

2-factors

108

3

24

11

9

3-factors

23

3

10

7

6

4-factors

2

0

0

1

1

b

RT-RFS

80

p = 0.0665

60 0-factors (n = 381) 1-factors (n = 574) 2-factors (n = 108) 3-or 4-factors (n = 25)

40 20

0

5

10

15

20

L-RFS

80 p < 0.0001 60 0-factors (n = 381) 1-factors (n = 574) 2-factors (n = 108) 3-or 4-factors (n = 25)

40 20

25

0

5

10

15

20

25

d CSS

D-RFS

100

100

80

CSS rate (%)

D-RFS Srate (%)

0

0

0

c

Disease-related death

100

L-RFS rate (%)

RT-RFS rate (%)

100

Distant recurrence

p < 0.0001

60 0-factors (n = 381) 1-factors (n = 574) 2-factors (n = 108) 3-or 4-factors (n = 25)

40 20

80

p < 0.0001

60 0-factors (n = 381) 1-factors (n = 574) 2-factors (n = 108) 3-or4-factors (n = 25)

40 20 0

0 0

5

10

15

20

Years after surgery

25

0

5

10

15

20

25

Years after surgery

Fig. 5 Univariate analysis in relation to RT-RFS, L-RFS, D-RFS, and CSS rates. a Comparison of RT-RFS rates according to the risk factor classification. No significant differences were observed among the four groups (P = 0.0665). b Comparison of L-RFS rates according to the risk factor classification. The 0-factor group and 1-factor group had highest L-RFS rates, whereas the 3- or 4-factor group had lowest L-RFS rate and the 2-factor group had intermediate L-RFS rate (P \ 0.0001). c Comparison of D-RFS rates according to

the risk factor classification. The 0-factor group and 1-factor group had highest D-RFS rate, whereas the 3- or 4-factor group had lowest D-RFS rate and the 2-factor group had intermediate D-RFS rate (P \ 0.0001). d Comparison of CSS rates according to the risk factor classification. The 0-factor group and 1-factor group had highest CSS rates, whereas the 3- or 4-factor group had lowest CSS rate and the 2-factor group had intermediate CSS rate (P \ 0.0001)

Bilimoria et al. [22] collected clinical data regarding 52,173 PTC patients from the National Cancer Database (NCDB) (1985–1998) and reported finding that total thyroidectomy had improved the survival rate of patients with tumors larger than 1 cm. Although this robust and persuasive study was adopted as a strong basis for the ATA guidelines [5], it contains some inconsistencies that open it to criticism, such as the fact that no detailed data were presented on extrathyroidal invasion, recurrence, or survival. The data in that study indicating that 18.4% of the patients who were treated by lobectomy underwent RAI ablation also are contradictory. Because these inconsistencies may have affected the conclusions, whether total thyroidectomy improves the survival outcome of patients

whose primary tumor is larger than 1 cm remains uncertain. Indeed, several groups have demonstrated that total thyroidectomy does not improve survival, even among high-risk patients [23, 24]. Another reason for performing total thyroidectomy as the initial surgical treatment for PTC is the possibility of intrathyroidal metastasis in the contralateral lobe, because bilateral cancers have been reported in 30–85% of PTC patients [25–29]. However, the recurrence rate in the remnant thyroid gland at 25 years in the present study was just 6.5%, and it was almost the same as the rates in previous reports [10, 29, 30]. Similar discrepancies between the clinical recurrence rate and the potential recurrence rate that could be expected if all contralateral microscopic

123

World J Surg (2014) 38:68–79

multicentric disease actually manifested itself have been reported in several papers [31–34]. Thus, the possibility of intrathyroidal metastasis in the contralateral lobe is not alone a sufficient reason to justify routine total thyroidectomy in all PTC cases. Another reason for performing total thyroidectomy as the initial surgical treatment for PTC is as a preparation for RAI ablation. However, the indications for postoperative RAI ablation in PTC patients have not yet been clearly defined. Some studies have reported an improved outcome after RAI ablation [35, 36], but others were unable to identify any benefit [37]. A recent report by Gepalakrishna et al. [38] demonstrated the absence of any benefit of RAI ablation and showed a higher risk of secondary carcinogenesis in PTC patients younger than age 45 years, suggesting that RAI ablation, especially in patients younger than age 45 years, must be carefully considered. According to a large study of data obtained from the NCDB [22], which was estimated to have covered 88% of all thyroid cancer patients treated in the United States, patients who underwent postoperative RAI ablation accounted for 56% of the patients who had undergone total thyroidectomy. Thus, in almost half the PTC patients who were treated by total thyroidectomy in the United States, preparation for RAI ablation was not the reason for performing the total thyroidectomy. An indisputable advantage of total thyroidectomy is that when performed in combination with postoperative RAI ablation, the patient’s serum Tg level can be used as a sensitive marker for recurrence. However, total thyroidectomy is followed by a higher rate of surgical complications than lobectomy, even though the postoperative rates of permanent bilateral laryngeal nerve injury and hypoparathyroidism are very low when total thyroidectomy is performed by experienced surgeons. However, many PTC patients are not treated by experienced surgeons [21], and permanent bilateral laryngeal nerve injury and hypoparathyroidism do not occur as postoperative complications of unilateral lobectomy. In addition, lifetime levothyroxine supplementation is inevitable after total thyroidectomy. Thus, total thyroidectomy is not justified unless it improves patient outcome sufficiently to compensate for the increased risk of complications. In other words, lobectomy may be valid as the initial surgery in PTC cases in which the clinical outcome cannot be improved by total thyroidectomy. The results of the present study demonstrated that age 45 years or older, tumor larger than 40 mm, extrathyroidal invasion, and clinical lymph node metastasis were significant risk factors for a poor outcome (Fig. 5a–d). Because no data regarding patients who underwent total thyroidectomy were included in the present study, whether total thyroidectomy plus RAI ablation improves the outcome of such patients remains unknown, and further study will be

77

needed to reach a conclusion. On the other hand, in the patient group without any of these risk factors, which corresponds to the 0-factor group in this study, no one died of PTC (Table 6; Fig. 5d). Because the CSS rate in the 0-factor group in our study has been 100% for 25 years, total thyroidectomy, even if followed by RAI ablation, would not have improved the survival outcome of this group. Although total thyroidectomy could have prevented the remnant thyroid recurrence, no available evidence indicates that total thyroidectomy reduces regional lymph node recurrence in PTC patients. We cannot rule out the possibility that total thyroidectomy plus RAI ablation could have prevented the distant recurrence without making a comparison with a control group treated by total thyroidectomy. However, because distant recurrence was diagnosed in only 1 of the 381 patients in the 0-factor group in this long-term follow-up study, after considering the risks and benefits, we concluded that unilateral lobectomy is valid initial surgical treatment in this group of patients. The present study had several limitations in addition to the absence of a control group treated by total thyroidectomy. First, no data were available regarding a family history of thyroid cancer or history of radiation exposure, both of which might have been associated with a poor outcome. Because the number of patients who have been positive for these factors is likely negligible among the 1,088 patients in our study, the lack of this information is unlikely to have altered the conclusions. Second, clinical lymph node metastasis was assessed by palpation in the present study, but, today, even lymph node metastasis under 5 mm can be detected by US and fineneedle aspiration cytology. Although PTC patients with a clinical lymph node metastasis, especially larger than 3 cm, have been reported to have a poor outcome [39, 40], the impact of clinical lymph node metastasis that is not palpable but is detected by US and aspiration cytology remains unclear. The possibility of a correlation existing between preoperative metastatic lymph node size and outcome should be investigated further. Third, the impact of lymph node dissection on patient outcome was not clarified in the present study. We analyzed recurrences and disease-related deaths according to the level of lymph node dissection, but there were no significant differences between clinical outcomes according to the dissection level. This result may be at least partly attributable to the strong selection bias that exists among the three different dissection levels, as described in the ‘‘Surgical Strategy’’ section. Indeed, paradoxically the D-RFS rate of the group of patients who underwent both central and lateral compartment dissection (92.2%) was lower than in the group who underwent dissection of only the central compartment (100%) or no lymph node dissection (99.1%), although the differences were not

123

78

World J Surg (2014) 38:68–79

statistically significant (P = 0.0595). Further study will be needed to determine the impact of lymph node dissection on patient outcome. The results of such a further study combined with the results of the present study may make it possible to determine the most valid indication for the extent of thyroidectomy and lymph node dissection for PTC patients.

12. 13.

14.

Acknowledgments The authors acknowledge the contribution that Dr. Kunihiko Ito, who passed away in November 2012, made to the present study. Dr. Ito had been the attending physician of most of the patients included in the study and had participated in the planning and execution of this study.

15.

The authors have no conflict of interests to

17.

Conflict of interest declare.

16.

18.

References 19. 1. American Cancer Society (2011) Cancer Facts & Figures 2011. Available via www.cancer.org/acs/groups/content/@epidemiology surveilance/documents/document/acspc-029771.pdf. Accessed 8 Feb 2012 2. Takami H, Ito Y, Okamoto T et al (2011) Therapeutic strategy for differentiated thyroid carcinoma in Japan based on a newly established guideline managed by Japanese society of thyroid surgeons and Japanese association of endocrine surgeons. World J Surg 35:111–121. doi:10.1007/s00268-010-0832-6 3. Hundahl SA, Fleming ID, Fremgen AM et al (1998) A national cancer data base report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985–1995. Cancer 83:2638–2648 4. Shigematsu N, Takami H, Ito N et al (2005) Nation survey on the treatment policy for well-differentiated thyroid cancer. Results of a questionnaire distributed at the 37th meeting of the Japanese Society of Thyroid Surgery. Endocr J 52:479–491 5. Cooper DS, Doherty GM, Haugen BR et al (2009) Revised American thyroid association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19:1167–1213 6. National Comprehensive Cancer Network (2010) NCCN clinical practice guidelines in oncology. Thyroid carcinoma, vol 1. Available via http://www.nccn.org/professionals/physician_gls/ pdf/thyroid.pdf. Accessed 1 Jan 2013 7. Cobin RH, Gharib H, Bergman DA, Thyroid Carcinoma Task Force et al (2001) AACE/AAES medical/surgical guidelines for clinical practice: management of thyroid carcinoma. American Association of Clinical Endocrinologists. American college of endocrinology. Endocr Pract 7:202–220 8. British Thyroid Association (2007) Guidelines for the management of thyroid cancer, 2nd edn. Available via www.britishthyroid-assiciation.org/news/Docs/Thyroid_cancer_guidelines_ 2007.pdf. Accessed 8 Feb 2012 9. DeLellis RA, Lloyd RV, Heitz PU et al (2004) WHO classification of tumours, pathology and genetics of tumours of endocrine organs. IRAC Press, Lyon, pp 73–76 10. Hay ID, Bergstralh EJ, Goellner JR et al (1993) Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1,779 patients surgically treated at one institution during 1940 through 1989. Surgery 114:1050–1058 11. Hay ID, Grant CS, Taylor WF et al (1987) Ipsilateral lobectomy versus bilateral lobar resection in papillary thyroid carcinoma: a

123

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31. 32.

retrospective analysis of surgical outcome using a novel prognostic scoring system. Surgery 102:1088–1094 Cady B, Rossi R (1988) An expanded view of risk-group definition in differentiated thyroid carcinoma. Surgery 104:947–953 Shaha JP, Loree TR, Dharker D et al (1992) Prognostic factors in differentiated carcinoma of the thyroid gland. Am J Surg 164:658–661 Simpson WJ, McKinney SE, Carruthers JS et al (1987) Papillary and follicular thyroid cancer: prognostic factors in 1,578 patients. Am J Med 83:479–488 Kingma G, van den Bergen HA, de Vries JE (1991) Prognostic scoring systems in differentiated thyroid carcinoma: which is the best? Neth J Surg 43:63–66 Peto R, Pike MC, Armitage P et al (1977) Design and analysis of randomized clinical trials requiring prolonged observation of each patients. II. Analysis and examples. Br J Cancer 35:1–39 Cox DR (1972) Regression models and life tables. J Roy Stat B 34:187–204 Cooper DS, Doherty GM, Haugen BR et al (2009) Revised American thyroid association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19:1167–1214 Gharib H, Papini E, Valcavi R, AACE/AME Task Force on Thyroid Nodules et al (2006) American association of clinical endocrinologists and associazione medici endocrinologi medical guidelines for clinical practice for the diagnosis and management of thyroid nodules. Endocr Pract 12:63–102 Pacini F, Schlumberger M, Dralle H, European Thyroid Cancer Taskforce et al (2006) European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur J Endocrinol 154:787–803 Sherman SI, Angelos P, Ball DW, National Comprehensive Cancer Network Thyroid Carcinoma Panel et al (2007) Thyroid Carcinoma. J Natl Compr Canc Netw 5:568–621 Bilimoria KY, Bentrem DJ, Ko CY et al (2007) Extent of surgery affects survival for papillary thyroid cancer. Ann Surg 246:375–381 discussion 381–384 Wanebo H, Coburn M, Teates D et al (1998) Total thyroidectomy does not enhance disease control or survival even in high-risk patients with differentiated thyroid cancer. Ann Surg 227:912–921 Haigh PI, Urbach DR, Rotstein LE et al (2005) Extent of thyroidectomy is not a major determinant of survival in low- or highrisk papillary thyroid cancer. Ann Surg Oncol 12:81–89 Mazzaferri EL, Young RL, Oertel JE et al (1977) Papillary thyroid carcinoma: the impact of therapy in 576 patients. Medicine (Baltimore) 56:171–196 Russel WO, Ibanz ML, Clark RL et al (1963) Thyroid carcinoma: classification, intraglandular dissemination and clinicopathological study based upon whole organ sections of 80 glands. Cancer 16:1425–1460 Clark RL, White EC, Russel WO (1959) Total thyroidectomy for cancer of the thyroid: significance of intraglandular dissemination. Ann Surg 149:858–866 Black BM, Kirk TA, Woolner LB (1960) Multicentricity of papillary adenocarcinoma of the thyroid: influence on treatment. J Clin Endocrinol Metab 20:130–135 Samaan NA, Maheshwari YK, Nader S et al (1983) Impact of therapy for differentiated carcinoma of the thyroid: an analysis of 706 cases. J Clin Endocrinol Metab 56:1131–1138 Grant CS, Hay ID, Gough IR et al (1988) Local recurrence in papillary thyroid carcinoma: is extent of surgical resection important? Surgery 104:954–962 Tollefsen H, Shaha J, Huvos A (1972) Papillary carcinoma of the thyroid. Am J Surg 124:468–472 Cohn KH, Ba¨ckdahl M, Forsslund G et al (1984) Biologic considerations and operative strategy in papillary thyroid carcinoma:

World J Surg (2014) 38:68–79

33.

34.

35.

36. 37.

arguments against the routine performance of total thyroidectomy. Surgery 96:957–971 Rose RG, Kelsy MP, Russel WO et al (1963) Follow-up study of thyroid cancer treated by unilateral lobectomy. Am J Surg 106:494–500 Vickery AL, Wang CA, Walker AM (1987) Treatment of intrathyroidal papillary carcinoma of the thyroid. Cancer 60:2587–2595 Mazzaferri EL, Jhiang SM (1994) Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 97:418–428 Mazzaferri EL (1999) An overview of the management of papillary and follicular thyroid carcinoma. Thyroid 9:421–427 Schvartz C, Bonnetain F, Dabakuyo S et al (2012) Impact on overall survival of radioactive iodine in low-risk differentiated thyroid cancer patients. J Clin Endocrinol Metab 97:1526–1535

79 38. Gepalakrishna N, Morris LGT, Tuttle RM et al (2011) Rising incidence of second cancers in patients with low-risk (T1N0) thyroid cancer who receive radioactive iodine therapy. Cancer 117:4439–4446 39. Ito Y, Ichihara K, Masuoka H et al (2010) Establishment of an intraoperative staging system (iStage) by improving UICC TNM classification system for papillary thyroid carcinoma. World J Surg 34:2570–2580. doi:10.1007/s00268-010-0710-2 40. Sugitani I, Kasai N, Fujimoto Y et al (2004) A novel classification system for patients with PTC: addition of the new variables of large (3 cm or greater) nodal metastases and reclassification during the follow-up period. Surgery 135:139–148

123