Optimal Extent of Neck Dissection for a Head and Neck Lymph Node Metastasis from a Remote Primary Site

Abstract

Background: Despite its rarity and limited documentation, therapeutic neck dissection (ND) for cervical lymph node (LN) metastases from distant primary sites is increasingly practiced, potentially enhancing survival rates. However, the optimal ND extent remains unclear. This study aimed to determine the safety of excluding upper neck levels from ND.

Methods: We retrospectively analyzed 25 patients who underwent ND for cervical LN metastases from remote primary tumors between 2015 and 2021 (12 with primary lung tumors, four with ovary, three with mammary gland, three with esophagus, two with thymus, and one with colon).

Results: Assessing clinical characteristics and occult metastasis rates, we observed LN metastases predominantly at levels III and IV. Occult metastases occurred in 14 out of 25 patients, primarily at neck levels III and IV (55.0% and 50.0%, respectively). The five-year disease-specific survival rate for all patients was 44.3%. While no statistically significant impact of occult metastasis on prognosis was confirmed, an association between the postoperative LN ratio and poor prognosis was revealed.

Conclusions: Our findings suggest that prophylactic NDs at levels I, II, and Va may not be essential for managing cervical LN metastases from remote primary malignancies. This could lead to a more tailored and less invasive therapeutic strategy.

Keywords: cervical lymph node metastasis; occult metastasis; optimal neck dissection extent; remote primary; survival.

Figure 1

Distribution of Pathological and Occult LN Metastases by Cervical Level. The figure presents the percentage of pathological (gray) and occult (blue) LN metastases at cervical levels Ia through Vb. Pathological involvement is confirmed through histopathology after dissection, while occult metastasis is undetected clinically but identified post-surgery. Notably, pathological involvement peaks at level IV (92%), while occult metastasis is most prevalent at level III (55%). Levels Ia, IIb, and Va show no pathological involvement.

Figure 2

Comparison of disease-specific survival between the lung cancer group and other cancer groups using the Kaplan–Meier method. The log-rank test was employed to determine statistical significance. No significant differences were observed between the two groups (p = 0.19).

Figure 3

Kaplan–Meier survival analysis comparing disease-specific survival among groups with adenocarcinoma, squamous cell carcinoma (SCC), and other histological types. The log-rank test indicated no significant survival differences between the groups (p = 0.43).

Figure 4

Kaplan–Meier analyses of disease-specific (DSS) and recurrence-free survival (RFS) across different prognostic factors. (A) For the number of p+N, no significant differences in DSS (p = 0.250) or RFS (p = 0.408) were observed between patients with n = 1 and those with n ≥ 2. (B) Analysis of occult lymph node (LN) metastasis showed similar DSS (p = 0.400) and RFS (p = 0.545) for patients with and without occult LN metastasis. (C) However, when stratified using a lymph node ratio (LNR) of 0.2, significant disparities emerged, with improved DSS (p = 0.041) and RFS (p = 0.02) in patients below the cutoff, highlighting the prognostic value of LNR. *, p < 0.05.