Sentinel Lymph Node Detection in Early-Stage Oral Squamous Cell Carcinoma Using Magnetic Resonance Lymphography: A...

Objectives: To assess the efficacy of magnetic resonance (MR) lymphography with gadobutrol contrast for sentinel lymph node (SLN) mapping in early-stage oral squamous cell carcinoma (OSCC). Methods: This pilot study compared the identification of SLNs by MR lymphography using a gadolinium-based contrast agent (gadobutrol) to conventional [^99mTc]Tc-nanocolloid lymphoscintigraphy (including single-photon emission computed tomography/computed tomography (SPECT/CT)) in 10 early-stage OSCC patients undergoing SLN biopsy. The patients initially underwent conventional lymphoscintigraphy following the peritumoral administration of indocyanine green [^99mTc]Tc-nanocolloid (120 megabecquerel; ~0.5 mL). Subsequently, 0.5-1.0 mL gadobutrol was peritumorally injected, and MR imaging was acquired for 30 min. The following day, the identified SLNs were harvested and subjected to a histopathological assessment. The MR lymphography and [^99mTc]Tc-nanocolloid lymphoscintigraphy results were evaluated and compared with respect to those of the SLN identification. The reference standard consisted of a histopathological evaluation of the harvested SLNs, complementary neck dissection specimens, and follow-up data. Results: The MR lymphography detected 16 out of 27 SLNs identified by [^99mTc]Tc-nanocolloid lymphoscintigraphy, revealing an additional SLN that did not harbor metastasis. MR lymphography failed to identify any SLNs in one patient. Of the seven histopathologically positive SLNs detected by [^99mTc]Tc-nanocolloid lymphoscintigraphy, three were identified by MR lymphography. All patients remained disease-free after a median follow-up of 16 months. Compared to [^99mTc]Tc-nanocolloid lymphoscintigraphy, MR lymphography using gadobutrol achieved an SLN identification rate of 59%, a sensitivity of 75%, and a negative predictive value of 86%. Conclusions: MR lymphography using gadobutrol demonstrates limited reliability for SLN mapping in early-stage OSCC.