Prophylactic central neck dissection in clinically node-negative, low- to intermediate-risk papillary thyroid carcinoma: a meta-analysis and trial sequential analysis of randomized controlled trials
In brief
Prophylactic neck dissection left recurrence at 3.4% versus 4.5% in thyroid cancer
In six randomized trials of 1,303 patients with clinically node-negative, mostly small papillary thyroid cancers, adding neck dissection did not detectably reduce cancer recurrence. The evidence was too limited to rule out a small benefit, but it excluded a recurrence reduction of 50% or more; results do not establish whether the operation helps patients with larger tumors.
- Journal
- The British journal of surgery (Q1)
- Published
- 31 October 2026
- Study design
- Systematic review / meta-analysis of RCTs
- Evidence level
- Level 1, High (CEBM 1a)
- Authors
- Fabio Medas, Gian Luigi Canu, Federico Cappellacci, Eleonora Noli, Eleonora Lori, Salvatore Sorrenti, et al.
- PMID
- 43062043
- DOI
- 10.1093/bjs/znag168
Why clinicians should know about it
- Picked for Breast and Endocrine Surgery (paper of the day, 21 November 2026): Randomized evidence on de-escalating central neck dissection in low-risk thyroid
- Picked for Surgical Oncology (paper of the day, 21 November 2026): Randomized evidence directly informs prophylactic lymphadenectomy
Abstract
BACKGROUND: The value of adding prophylactic central neck dissection (pCND) to thyroidectomy in clinically node-negative (cN0) papillary thyroid carcinoma (PTC) is contested. Randomized evidence with the addition of a large RCT and a 10-year trial update was used to test its conclusiveness. METHODS: Five databases and two trial registries were searched for RCTs comparing thyroidectomy with versus without pCND in cN0 PTC. The primary outcome was structural recurrence; secondary outcomes were biochemical recurrence and parathyroid and nerve morbidity. Risk of bias was assessed in duplicate (RoB 2); risk differences (RD) were pooled with a random-effects model, and trial sequential analysis (TSA) estimated the information size required for a reliable conclusion. RESULTS: Six RCTs (1303 analysed patients) were included, enrolling predominantly T1-T2, low- to intermediate-risk tumours. pCND produced no statistically detectable reduction in structural recurrence (3.4% versus 4.5%; RD 0.00, 95% c.i. -0.02 to +0.01) or biochemical recurrence, with a small, non-significant harm-leaning signal for parathyroid morbidity. TSA showed that only about 13% of the required information had accrued, so a small absolute benefit could not be excluded, although a relative reduction of 50% or more was excluded. GRADE certainty was moderate for recurrence and low to very low for secondary outcomes. CONCLUSIONS: In predominantly T1-T2, low- to intermediate-risk cN0 PTC, current randomized evidence does not support routine pCND. There was no detectable oncological benefit and, at most, a small parathyroid morbidity cost. These findings should not be extrapolated to T3-T4 disease, as no stratified, adequately powered estimates exist for larger tumours. This should be the focus of future randomized trials.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.