During the past 20 years, the incidence of thyroid cancer has increased sharply all over the world mainly due to increased detection using ultrasonography [1], and the incidence rate of thyroid cancer in Korea is the highest in the world [2]. The incidence of papillary thyroid carcinoma (PTC) accounts for the majority of this increase [3]. Although patients with PTC have excellent prognosis and survival, recurrence occurs in 5%–21% of patients with PTC [4,5]. Recurrence of PTC usually requires a second operation, which carries a higher complication rate, especially if central neck dissection (CND) is necessary. Therefore, the prevention of recurrence is important in the management of PTC to minimize treatment morbidity and improve both quality of life and survival rate.

Tumor size, extrathyroidal extension (ETE), age, lymph node (LN) metastasis, tumor multiplicity, extranodal spread, and male sex are all associated with increased risk of recurrence of PTC [6-8]. LN metastasis occurs in 20%–90% of patients with PTC [9-11], and is associated with increased recurrence and decreased survival in patients with PTC [12,13]. Recently, the number of metastatic LNs and the ratio of metastatic LNs to the total number of retrieved LNs (the LN ratio [LNR]) have been suggested as prognostic risk factors for recurrence in patients with PTC [14-21]. However, the clinical significance of these measures in patients with clinically node negative (cN0) PTC has not yet been established clearly. The aim of this study is to evaluate the prognostic significance of the number of metastatic LNs and the LNR of central compartment LNs in patients with cN0 PTC.

Prophylactic CND remains still controversial in patients with cN0 PTC. Argues against prophylactic CND include a lack of substantial evidence of the benefit and the potential morbidity of the procedure [11,23]. However, prophylactic CND may LN recurrence, and it provides more accurate staging for decisions regarding adjuvant RAI ablation [24,25]. The aim of this study was to evaluate the clinical significance of the number of metastatic central LNs and the LNR of metastatic LNs to the total number of retrieved LNs in patients with PTC who underwent total thyroidectomy and prophylactic CND. Prophylactic CND would be very useful to decide postoperative strategies if the information obtained after prophylactic CND such as the presence of LN metastasis, the number of positive LN, and LNR can be prognostic factors related to the recurrence.

In the current study, we found that a central LNR of 0.31, two or more metastatic central LNs, tumor size ≥20 mm and maximal ETE were independent risk factors for recurrence, whereas the presence of metastatic central LNs was not related to recurrence in multivariate analysis. The hazard ratio of LNR ≥0.31 was 11.23 (95% CI, 2.81 to 44.82), and it is higher than central LN metastasis ≥2, tumor size ≥20 mm, and maximal ETE (hazard ratio=3.43, 6.61, and 7.17, respectively). The LNR seems to be a promising prognostic factor for recurrence in patients with cN0 PTC.

The clinical significance of the number of metastatic LNs for patients with PTC remains unclear, especially in cN0 PTC. Some investigators suggested that the number of metastatic LNs is a significant predictive factor for disease recurrence [14,15]. Sugitani et al. [14] demonstrated that the risk of recurrence was significantly higher in PTC patients with five or more metastatic LNs (19%) that in those with fewer than five metastatic LNs (8%). Park et al. [15] also reported that the number of metastatic central compartment LNs (>3 metastatic LNs) in PTC was a significant predictor for recurrence conferring a hazard ratio of 1.36 (95% CI, 1.103 to 1.680). In the current study, two or more metastatic central LNs were an independent risk factor for recurrence in cN0 PTC by multivariate analysis. However, some authors have reported that the number of metastatic LNs was not associated with recurrence [21,26]. Wang et al. [26] reported that higher T stage, increased number of metastatic LNs, increased LN diameter, and presence of extranodal spread were related to recurrence in differentiated thyroid carcinoma in univariate analysis. However, in multivariate analysis, extranodal spread was the only statistically significant factor, and the number of metastatic LNs was not associated with recurrence [26]. Ryu et al. [21] also demonstrated by multivariate analysis that the number of metastatic LNs was not an independent risk factor for recurrence in pathologic N1a PTC. Further studies might be necessary to clarify the impact of number of LN metastases on the risk of recurrence in PTC considering all previous studies.

The LNR is likely to be an important prognostic factor for risk of recurrence in the previous studies. Several authors have assessed the relationship between the LNR and recurrence in patients with PTC, and reported that LNR was an important risk factor for recurrence although, in fact, the cutoff values for the LNR range from 0.26 to 0.86 in different studies [16-21,27-30]. Vas Nunes et al. [16] reported that PTC patients with an LNR of 0.30 or higher had a 3.4 times higher risk of persistent or recurrent disease compared with patients with an LNR of 0.00. In other study, LNR ≥0.86 in the central compartment had significantly worse disease-free survival rates than patients with ratios below these threshold values [17]. Jeon et al. [18] reported that LNR (higher than 0.4) and size of metastatic nodes were significant prognostic factors in pathological N1a PTC. Yip et al. [19] also demonstrated that the LNR independently predicted PTC recurrence. Schneider et al. [20] found that a LNR ≥0.42 is the best cutoff value to divide those with LN metastasis based on disease-specific survival in the study of 10,995 PTC patients utilizing the surveillance, epidemiology, and end results (SEER) database. Choi et al. [27] reported that the LNR of 0.44 is a predictive factor for the loco-regional recurrence in in patients with papillary thyroid microcarcinoma. Lee et al. [28] analyzed both central and lateral compartment LNRs: it reported that cutoff values of 0.4 and 0.5 for central LNR and total LNR, respectively, were significant risk factors for recurrence of PTC. However, Beal et al. [29] analyzed 9,926 patients with differentiated thyroid carcinoma in the SEER database, and showed that the LNR was not associated with overall survival in the pathologic N1 patients.

There are few reports that evaluated the significance of LNR in prophylactic CND for PTC. Ryu et al. [21] demonstrated that an LNR >0.65 in the central compartment was the only independent factor for recurrence in pathologic N1a PTC after prophylactic CND. Lee et al. [30] also showed that an LNR ≥0.26 was an independent predictor of regional LN metastasis recurrence in patients with PTC who underwent total thyroidectomy and bilateral prophylactic CND. In the current study, an LNR of ≥0.31 was an independent predictor of recurrence in patients with cN0 PTC who underwent total thyroidectomy and prophylactic CND.

The differences of the LNR and the number of metastatic LNs between previous studies might be related to difference in study cohorts, cutoff values for metastatic LNs and LNR, LN yields, accuracy of pathologic examination, and follow-up period. Some studies analyzed only patients with central compartment LN metastasis, whereas other workers included patients with both central and lateral LN metastasis, which might be related to more advanced disease and worse prognosis. We excluded patients who underwent concurrent lateral compartment LN dissection and included only the homogenous prophylactic CND group. The total LN yield via CND might be influenced by the extent of LN dissection and accuracy of pathologic examination. Therefore, high and low yields of LNs might be defined differently in different studies, although the LNRs might be similar. In the current study, we performed compartment-oriented CND routinely, including pretracheal, paratracheal and prelaryngeal LN groups in most patients. The mean number of harvested LNs was 7.14±6.23, which we consider to be adequate, indicating that CND was carried out correctly. We also included only those patients from whom at least two central LNs were harvested. To minimize bias and differences between studies, standardized cutoff values for the LNR and number of metastatic LNs are needed. In addition, the follow-up period differed between the studies. In some studies, the follow-up period was in the region of 5 years; however, a long-term follow-up of more than 10–20 years is necessary for accurate evaluation of recurrence and survival in patients with PTC.

There are some limitations to the current study. It is a retrospective study performed in a single center and might be influenced by a selection bias. The diverse LN yield (range, 2 to 40) during CND might also cause bias in the results, although we routinely performed compartment-oriented CND. The relatively short follow-up period is also another drawback. The mean follow-up period was 82.2±26.4 months, which might not be long enough to determine all recurrences, although most recurrences occurred within 5–10 years. A further prospective cohort study including larger numbers of homogenous patients with long-term follow up is necessary to overcome the limitations of this study.

In conclusion, tumor size ≥20 mm, maximal ETE, two or more metastatic central LNs, and an LNR ≥0.31 are independent predictors for recurrence in patients with cN0 PTC. A LNR ≥0.31 and number of metastatic LNs greater than or equal to two might be used as prognostic indicators after prophylactic CND to guide postoperative adjuvant therapy and follow-up strategy.