Related Experiment Video
Updated: Dec 4, 2025

Transoral Robotic Total Thyroidectomy and Bilateral Central Regional Lymph Node Dissection for Papillary Thyroid Carcinoma
Published on: September 15, 2023
A Retrospective Study of Lymph Node Yield in Lateral Neck Dissection for Papillary Thyroid Carcinoma
Khalil Issa1, Madelyn N Stevens1,2, Yuhui Sun3
1Department of Head and Neck Surgery & Communication Sciences, 12277Duke University Health System, Durham, NC, USA.
Objective:
Lateral neck dissection (LND) is important in managing papillary thyroid carcinomas (PTCs). This study aimed to evaluate the relationship between lymph node yield (LNY) for LND and patient outcomes, specifically postoperative serum thyroglobulin levels (sTG) and radioiodine uptake on thyroid scan, and to estimate a threshold LNY to signify adequate LND.
Methods:
Patients diagnosed with PTC who underwent LND from 2006 to 2015 at a single institution were included. Linear regression with restricted cubic splines was used to characterize the association of LNY with outcomes. Outcomes were log-transformed to achieve a more symmetric distribution prior to regression. For nonlinear associations, a Monte Carlo Markov Chain procedure was used to estimate a threshold LNY associated with postoperative outcome. This threshold was then used to define high LNY versus low LNY in the subsequent analyses.
Results:
In total, 107 adult patients were included. There was a significant relationship between LNY and postoperative sTG level (P = .004), but not radioiodine uptake (P = .64). An LNY of 42.96 was identified, which was associated with the maximum change in sTG level. No association was found between LNY groups (LNY ≥ 43 vs < 43) and radioiodine uptake, risk of complication, or longer operative times (all P > .05). High LNY was associated with a decrease in log(sTG) (estimate = -1.855, P = .03), indicating that adequacy of LND is associated with an 84.4% decrease in sTG.
Conclusion:
These results suggest an association between LNY and postoperative sTG level, with an estimated threshold of 43 nodes. This has implications for adequate therapeutic LND; additional work is needed to validate thresholds for clinical practice.

