Intra-operative Injection Improves Sentinel Node Accuracy In Auricular Melanoma
Seth Noorbakhsh1, Raysa Cabrejo, MD2, Darko Pucar, MD, PhD2, Kelly Olino, MD2, Sarah Weiss, MD2, James Clune, MD2.
1West Virginia University School of Medicine, Morgantown, WV, USA, 2Yale School of Medicine, New Haven, CT, USA.
Purpose: Sentinel lymph node mapping for invasive auricular melanoma is standard of care, yet lymphatic drainage from the ear is poorly understood. Studies that have examined the lymphatic drainage patterns of the ear are few, and conclusions have varied. Uncertain drainage patterns and discrepancies in the literature make surgical planning difficult. The goal of this study is to evaluate lymphatic mapping of the ear before and during surgery and to compare sentinel node biopsy sites to locations of future recurrence. Methods: In this study, 177 cases of auricular melanoma from 1997 to 2018 were reviewed. Fifty-five cases were excluded as melanoma in-situ. One hundred twenty-two patients were included in the study; lesion site, pathological findings, and demographic information were analyzed for each. Pre-operative lymphoscintigraphy was performed in 68 cases to map the draining nodal basin(s). Lymphatic drainage regions of the neck were classified into five surgically-relevant regions based on lymphoscintigram and/or SPECT/CT imaging and description. Lymphatic mapping was also performed intra-operatively by the surgeon injecting technetium Tc 99m-sulfur colloid, and a gamma probe was used to identify "hot" nodes. Intra-operative lymph node biopsy sites were compared to pre-operative maps. Sites of future recurrence were also compared to the pre-operative maps and intra-operative biopsy sites. Results: Sixty-three patients had draining nodes identified on pre-operative lymphoscintigram, with a total of 95 draining basins. Five patients did not have identifiable lymph nodes on the pre-operative lymphoscintigram but had lymph nodes identified by the intra-operative injection (3 parotid, 1 anterior neck and 1 not recorded). A total of eight patients (13 total nodes) had micrometastatic disease on SLNB. Drainage to the parotid/pre-auricular region on pre-operative lymphoscintigram was found in 31.1% of patients, while 45.9% of patients had a SLNB within the parotid/pre-auricular region intra-operatively during a second injection of T-99 (p= 0.07). This corresponded to 16 patients with parotid/pre-auricular region sampling intra-operatively that did not show evidence of parotid/pre-auricular drainage in the pre-operative lymphoscintigram. Five of these 16 patients (31.3%) missed by pre-operative lymphoscintigraphy subsequently had a sentinel node biopsy positive for melanoma in the parotid/pre-auricular basin. In total, 16 of 68 patients that had sentinel node biopsies experienced loco-regional recurrence. Only two patients experienced recurrence in a nodal basin that was negative on initial sentinel node biopsy in this study, a false negative rate of 2.9%. Conclusions: Lymphatic drainage from the ear is difficult to predict. Pre-operative lymphoscintigraphy followed by intra-operative T-99 injection is an accurate method for detecting lymphatic spread in auricular melanoma. Relying on pre-operative lymphoscintigraphy alone would have resulted in five missed positive sentinel nodes of the 13 total positive nodes in the study, an error rate of 38%. Intra-operative T-99 injection was able to rectify these pre-operative lymphatic mapping errors. Most positive nodes were located in the parotid basin, which was also the most likely basin to be missed pre-operatively. Utilizing this dual pre-operative and intra-operative lymphatic mapping strategy, only two patients (2.9%) had a nodal recurrence in a basin that was previously biopsied.
Back to 2020 Posters