Thyroid cancer (TC) is an increasingly common malignancy, of which papillary thyroid cancer (PTC) is the most common histologic type, with a reported incidence of 80% to 85% [1
]. Papillary thyroid microcarcinoma (PTMC) is defined as a tumor less than 1 cm in size and has a good prognosis, with a 10-year survival rate over 95% and a recurrence rate of 10% [3
]. Central lymph node (CLN) metastasis has been reported in 20.7–62% of clinically lymph node negative (cN0) PTMC cases [6
]. Previous studies reported that CLN metastasis was associated with recurrence [11
]. Although patients with CLN metastasis have a risk of recurrence, the effectiveness of completion total thyroidectomy in patients diagnosed with pathological CLN metastasis (pCLNM) after surgery is still controversial. According to the 2015 American Thyroid Association (ATA) guidelines, completion total thyroidectomy is necessary in cases of unclear diagnosis after lobectomy, to provide complete resection of multicentric disease, and to allow for efficient radioactive iodine (RAI) therapy; however, the guidelines do not specify pCLNM [15
Prophylactic central compartment neck dissection (CCND) can affect the pCLNM rate. The mean size and number of metastatic lymph nodes removed during prophylactic CCND were reported as 0.35 cm and 2.6 ± 3 out of 13 ± 5 lymph nodes, respectively [16
]. As pCLNM was reported to be a micrometastasis, performing prophylactic CCND in patients with cN0 PTMC has been continually under controversy. Multiple studies have compared thyroidectomy alone to thyroidectomy with CCND in cN0 PTMC and reported varying results in complications, recurrence rates, and patient prognosis [18
]. Although opinions vary, most head and neck surgeons and endocrine surgeons in Korea, China, and Japan perform prophylactic CCND even in cases of cN0 PTMC.
Our previous study demonstrated that there was no significant difference in prognosis between CLN-positive and CLN-negative patients with PTMC [21
]. As the number of patients with recurrence in the previous study was small, it is necessary to confirm the benefit of completion total thyroidectomy in a CLN-positive group based on more cases and long-term observation. This retrospective study was designed to evaluate the necessity of completion total thyroidectomy in low-risk PTMC patients with pCLNM who underwent thyroid lobectomy with prophylactic CCND at a single medical center.
As the incidence of TC increases worldwide, the physicians’ interest in its treatment and prognosis is increasing [22
]. The long-term recurrence and mortality rates of PTMC are low; thus, there has been constant controversy with regard to the definite treatment guidelines for TC [24
]. Ito et al. suggested active surveillance (AS) as a therapeutic option for low-risk PTMC [26
]. AS can help reduce the risk of surgical complications by avoiding immediate surgery, reducing costs, and improving the quality of life [27
]. However, the tumor progressed unexpectedly in about 2–15% of patients who were considered suitable for AS, delaying treatment, and these patients’ cancer eventually progressed to an aggressive disease [29
]. According to previous studies, it was reported that the loco-regional recurrence rate of patients who underwent lobectomy because of PTMC was 2–6%, the distant metastasis rate was 1–2%, and the disease-specific mortality rate was less than 1% [34
]. Since the clinical characteristics or molecular factors that distinguish low-risk PTMC patients at risk of progression have not been clearly identified, whether surgery or AS has a therapeutic advantage remains controversial.
Since pCLNM had been reported to be relatively high in PTMC, the presence or absence of pCLNM may be estimated as a factor that can affect prognosis and treatment [36
]. Completion total thyroidectomy was recommended for an unclear diagnosis after lobectomy, complete resection of multicentric disease, and efficient RAI therapy in the 2015 ATA guidelines [15
]. However, there is little information on the effectiveness of completion total thyroidectomy in patients who were diagnosed as having pCLNM after thyroid lobectomy.
To demonstrate the usefulness of completion total thyroidectomy in patients diagnosed as having pCLNM after surgery, we followed up low-risk PTMC patients who had undergone thyroid lobectomy with prophylactic CCND for a long time period. During the 13 years of follow-up, 48 patients (5.5%) experienced recurrence. The recurrence rate of the CLN-positive group was higher than that of the CLN-negative group, but there was no significant between-group difference. Our multivariate analysis showed that pCLNM was not associated with recurrence, and DFS was not significantly different between the two groups. Furthermore, there was no distant metastasis or disease-specific mortality. The recurrence rate of PTMC was reported to be 1.96–6% in previous studies. The most common recurrence site was the contralateral lobe, followed by the lateral LN [25
]. In our study, 33 patients (33/48, 68.8%) experienced recurrence only in the contralateral lobe and 11 patients (11/48, 22.9%) only in the lateral LN.
Since the most common recurrence site of PTMC was the contralateral lobe, we were faced with a dilemma of whether to perform completion total thyroidectomy. Several studies have reported that 2–7% of patients experience transient recurrent laryngeal nerve (RLN) palsy; 0.5–4.4%, permanent RLN palsy; 7–20%, transient hypoparathyroidism; and 2.5–5.8%, permanent hypoparathyroidism after completion total thyroidectomy [44
]. Although the incidence of complications is low, the risk of complications can be completely circumvented by avoiding unnecessary surgery. In addition, because there was no significant difference in long-term prognosis between the CLN-positive and CLN-negative groups and the recurrence rate was very low, we could suggest that completion total thyroidectomy in PTMC with pCLNM was not necessary.
An additional consideration is that pCLNM can affect lateral LN recurrence. Several previous studies have revealed that pCLNM is associated with lateral LN metastasis [37
]. These studies indicated the possibility of occult lateral LN metastasis in TC with pCLNM. However, there were few studies that proved the relationship between pCLNM and lateral LN recurrence of PTMC after long-term follow up. Ryu et al. reported that the pCLNM rate of TC patients who underwent total thyroidectomy was 33.2%, and lateral LN recurrence was associated with pCLNM [52
]. Our analysis showed similar results. Lateral LN recurrence in the CLN-positive group was significantly higher than in the CLN-negative group (odds ratio = 3.893, CI 1.391–10.894, p
= 0.012). The multivariate analysis found that lateral LN recurrence was associated with male sex, tumor size over 5 mm, and pCLNM. Although pCLNM can develop into lateral LN recurrence, the lateral recurrence rate was very low (7/165, 4.2%). Therefore, active and frequent follow-up is thought to be sufficient for patients with pCLNM. In addition, lateral LN metastasis without pCLNM, so-called skip metastasis, was reported in 6.8–27.8% of PTC patients [48
]. Combining this finding with our results, we carefully question the effectiveness of prophylactic CCND for cN0 PTMC; however, additional prospective randomized studies will be needed in the future.
In several previous studies, the complication rate after thyroidectomy was reported in 5.1% to 26.0% of cases [58
]. Our study showed a relatively low complication rate (44/876, 5.0%). In particular, the rate of hypocalcemia occurrence was 1.3% (11/876) and RLN injury was 0.4% (4/876), which were very low. This result was likely due to the fact that our institution is a high-volume center where specialized endocrine surgeons perform thyroidectomy. Previous studies showing that high volume surgeons had better outcomes relating to complications and prognosis support this [58
]. Therefore, we can suggest that experienced, specialized endocrine surgeons play a significant part in reducing surgical complications.
Our research showed that the rate of RLN injury was higher in the CLN-positive group. Postoperative complications can occur in any surgery. Since our hospital is an academic institution, trainees frequently participate in surgeries. Although the number of RLN injuries was very small, they did occur. It is difficult to explain why RLN injuries occurred more frequently in the CLN-positive group. Since the surgical range of both groups was determined under the same pre-surgical conditions and indications, we do not think that microscopic CLN metastasis affected the occurrence of surgical complications. According to previous studies, there was no significant difference in complication and clinical outcome when trainees participated in surgery [63
]. Therefore, we do not believe that the involvement of trainees influenced the RLN injury rate.
When the patients were divided into two periods and analyzed, the recurrence rate of the patients in the first period was significantly higher. However, this may be the result of the significant difference between the two periods in mean follow-up duration (24.2 ± 2.4 vs. 12.7 ± 3.6, p < 0.001). Furthermore, since there was no difference in the mean recurrence-free survival duration, we believe that the difference in recurrence rates between the two periods would not have a significant effect on our study’s results.
There are some limitations to this study. First, as the quality of ultrasonography was not good in the 1980s and 1990s, it could not detect very tiny nodules, which may have been mistaken for recurrence at a later time point. This may have affected the recurrence rate of patients. In our subgroup analysis by period, the fact that the rate of recurrence 5 years after surgery was significantly higher in the first period indicates this possibility (13.0% vs. 2.8%, p
= 0.003). However, since the number of recurrence patients in the first period was very small, further research is needed. Second, incidental multifocality or microscopic extrathyroidal extension that was undetectable before surgery could have affected the recurrence rates, although multivariate analysis revealed that the two factors were not related to recurrence rates. Third, recent studies have shown that BRAFV600E
promoter mutations could affect aggressive features and the prognosis of PTC [67
]. However, the effects of BRAFV600E
promoter mutations on prognosis could not be analyzed in this study because of limited data. Additional research is needed on this in the future.