- The 15.8% skip metastasis rate reported by Byers et al. (1997):
- Is widely considered an overestimate due to several important methodological issues that have been carefully dissected in subsequent literature:
- Most notably by Warshavsky et al. in their 2019 JAMA Otolaryngology meta-analysis [1][2]
- Is widely considered an overestimate due to several important methodological issues that have been carefully dissected in subsequent literature:
- What Byers Reported:
- Byers et al. reviewed 277 previously untreated patients with oral tongue SCC (1970 to 1990):
- Who underwent glossectomy and neck dissection
- They reported that 15.8% of all patients had either:
- Level IV metastasis as the only manifestation of neck disease
- Level III as the only positive node without disease in levels I to II
- Subsequent level IV recurrence after initial dissection that did not include level IV
- Based on this, they recommended routine dissection of levels I to IV for all oral tongue SCC [2]
- Byers et al. reviewed 277 previously untreated patients with oral tongue SCC (1970 to 1990):
- Why the 15.8% Figure Is Misleading:
- Warshavsky et al. performed a careful reanalysis of the Byers data and identified several critical flaws that inflated the rate [1]:
- Conflation of skip metastasis definitions:
- Byers combined true level IV skip metastasis with level III skip metastasis (level III positive without levels I to II involvement):
- These are fundamentally different clinical scenarios — level III disease is already captured by a standard supraomohyoid neck dissection (levels I to III)
- By lumping both together, the rate was artificially elevated
- Byers combined true level IV skip metastasis with level III skip metastasis (level III positive without levels I to II involvement):
- Inclusion of neck recurrences as “skip metastases”:
- Nine patients (9.9%) who developed level IV recurrences after an initial dissection that did not include level IV were counted toward the 15.8% figure:
- As Warshavsky et al. noted, counting neck recurrence as a missed pathological lymph node is problematic because the neck has lost its normal anatomical lymphatic drainage and, in many cases, has been irradiated
- This makes it impossible to determine whether these were true skip metastases or recurrences from altered lymphatic flow [1]
- Nine patients (9.9%) who developed level IV recurrences after an initial dissection that did not include level IV were counted toward the 15.8% figure:
- True level IV skip metastasis rate was only 5.5%:
- When the data are restricted to patients with cN0 disease who had level IV metastasis found in the initial neck dissection specimen (i.e., true pathologic skip metastasis):
- The rate drops to 5.5% — roughly one-third of the reported figure [1]
- Even the most generous calculation yields only 4.8%:
- When accounting for all cases mentioned in the study (both initial pathologic findings and subsequent recurrences), the combined incidence of skip metastasis or subsequent recurrence in level IV was only 4.8% (13 of 270), not 15.8% [1]
- When the data are restricted to patients with cN0 disease who had level IV metastasis found in the initial neck dissection specimen (i.e., true pathologic skip metastasis):
- Conflation of skip metastasis definitions:
- Mixed cN0 and cN+ populations:
- The study included patients across all clinical N stages, and the data were not clearly stratified by preoperative nodal status, making it difficult to isolate the true elective (cN0) skip metastasis rate
- Era of the study (1970 to 1990):
- Preoperative imaging was far less sophisticated, meaning some patients classified as cN0 may have had undetected nodal disease, further confounding the results
- Warshavsky et al. performed a careful reanalysis of the Byers data and identified several critical flaws that inflated the rate [1]:
- What Modern Data Shows:
- The Warshavsky et al. meta-analysis of 13 studies (1,359 cN0 patients):
- Found the true skip metastasis rate to level IV is only 0.50% (95% CI, 0.09%–1.11%):
- With an overall level IV involvement rate of 2.53% [1]
- Even for oral tongue specifically — the highest-risk subsite:
- The level IV involvement rate was 3.60% [1]
- Found the true skip metastasis rate to level IV is only 0.50% (95% CI, 0.09%–1.11%):
- A large Tata Memorial audit of 761 early-stage cN0 patients:
- Found skip metastasis to level IV in only 0.3% [3]
- The Warshavsky et al. meta-analysis of 13 studies (1,359 cN0 patients):
- In summary, the Byers 15.8% figure resulted from a broad definition of “skip” that included level III skips (already addressed by SOHND), conflation of initial pathologic findings with subsequent recurrences, and lack of stratification by preoperative nodal status:
- Modern evidence overwhelmingly supports that true skip metastasis to level IV in cN0 oral tongue SCC is a rare event (~0.5%), and supraomohyoid neck dissection (levels I to III) remains adequate for the elective setting
- References:
- 1. Assessment of the Rate of Skip Metastasis to Neck Level IV in Patients With Clinically Node-Negative Neck Oral Cavity Squamous Cell Carcinoma: A Systematic Review and Meta-analysis. Warshavsky A, Rosen R, Nard-Carmel N, et al. JAMA Otolaryngology– Head & Neck Surgery. 2019;145(6):542-548. doi:10.1001/jamaoto.2019.0784.
- 2. Frequency and Therapeutic Implications of “Skip Metastases” in the Neck From Squamous Carcinoma of the Oral Tongue. Byers RM, Weber RS, Andrews T, et al. Head & Neck. 1997;19(1):14-9. doi:10.1002/(sici)1097-0347(199701)19:13.0.co;2-y.
- 3. Incidence and Impact of Skip Metastasis in the Neck in Early Oral Cancer: Reality or a Myth?. Gurmeet Singh A, Sathe P, Roy S, et al. Oral Oncology. 2022;135:106201. doi:10.1016/j.oraloncology.2022.106201.
Rodrigo Arrangoiz MS, MD, FACS, FSSO 