Abstract
Objective: 1) Determine the ability of PET scan to predict pathologically positive nodes in patients with persistent nodal disease following radiation with or without chemotherapy. 2) Evaluate whether performing a single level neck dissection is feasible in patients with persistent nodal disease.
Method: Study design: Retrospective analysis. Setting: Tertiary care center. Patients were evaluated with PET after definitive (chemo) radiation and before neck dissection. The SUV of PET-positive nodes were recorded and compared to pathology reports of subsequent neck dissections. Data regarding follow-up, recurrence rates, and survival of these patients was recorded.
Results: When compared to pathology reports, post-radiation PET scans failed to identify positive nodes in 4/11 separate cervical levels (sensitivity of 63.6%, P = .005). It successfully identified 33/40 negative neck levels (specificity of 82.5%, P = .005). This data suggest that PET scan alone may be inadequate for predicting pathologically positive nodal levels for persistent neck disease after (chemo) radiation. It was also shown that 6/13 (46.2%) nodal levels with SUV greater than 3 on PET scan were pathologically positive. Conversely, a PET negative nodal level with SUV less than 3 had a 27.3% chance (3/11) of being found positive upon pathologic evaluation.
Conclusion: While the relatively high specificity demonstrated here may help support the case for selective neck dissection when PET shows positive nodes, the low sensitivity makes it difficult to exclude PET negative levels from the planned dissection. Larger prospective studies are needed to confirm these findings.
Get full access to this article
View all access options for this article.