Barry Benson Goldberg MD, Presenter: Research support, General Electric Company

Daniel Arthur Merton, Abstract Co-Author: Nothing to Disclose

Ji-Bin Liu MD, Abstract Co-Author: Nothing to Disclose

Mohan Aruva PhD, Abstract Co-Author: Nothing to Disclose

Traci B. Fox BS, Abstract Co-Author: Nothing to Disclose

See-Ying Chiou MD, Abstract Co-Author: Nothing to Disclose

Mathew L. Thakur PhD, Abstract Co-Author: Nothing to Disclose

Laurence Parker PhD, Abstract Co-Author: Nothing to Disclose

Flemming Forsberg PhD, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

To identify the variability in the number of SLNs that receive lymphatic drainage from melanoma tumors located in the head and neck compared to other anatomical regions in a swine model and to determine how tumor location influences SLN detection using contrast-enhanced ultrasound imaging (US) of lymphatic channels (LCs) and SLNs (i.e., lymphosonography).

Twenty-four Sinclair swine having a total of 39 melanomas were studied using wide-band grayscale harmonic US and nuclear medicine (Tc99m). Ten tumors were located on the head /neck (Group 1) and 29 tumors were in other anatomical areas (Group 2). After 1ml peritumoral injections of Sonazoid (GE Healthcare, Oslo, Norway) the contrast-filled LCs were followed with US to their respective SLNs. The thoracic duct was also evaluated with US in an effort to detect Sonazoid draining from the lymphatic system to the venous system. Tc99m studies were performed using established techniques. The number of SLNs detected was compared to peritumoral injections of blue dye with dye-guided surgical dissections as the gold standard. Variability of the number of SLNs in each Group and detection rates of the 2 imaging modalities were compared with an unpaired t-test. The ability of the 2 modalities to correctly detect SLNs was determined using McNemar’s test.

Blue dye-guided surgery identified 47 SLNs in Group 1 and 94 SLNs in Group 2 with more SLNs per tumor in Group 1 than Group 2 (4.7 + 1.88 vs 3.1 + 1.32; p<0.0109). SLN detection rates for US were 77% + 26.3 and 94% + 13.4% (p=0.0089), while Tc99m identified 58% + 22% and 73% + 35% (p=0.21) from Groups 1 and 2 respectively. Lymphosonography performed better than Tc99m for SLN detection in Group 1 (p=0.0045) and Group 2 (p<0.0001). The flow of Sonazoid through the thoracic duct and into the venous system only was detected in 3 Group 1 tumors.

Melanomas located in the head and neck region had more SLNs than tumors located in other areas. Lymphosonography performed statistically better than nuclear medicine for the detection of SLNs in both Groups.

Improved detection of SLNs could enhance the ability to accurately stage cancers and improve the effectiveness of SLN resections.

Goldberg, B, Merton, D, Liu, J, Aruva, M, Fox, T, Chiou, S, Thakur, M, Parker, L, Forsberg, F, et al, , Variability in Lymphatic Drainage From Melanoma Tumors: The Effect of Tumor Location on the Detection of Sentinel Lymph Nodes.  Radiological Society of North America 2006 Scientific Assembly and Annual Meeting, November 26 - December 1, 2006 ,Chicago IL. http://archive.rsna.org/2006/4435803.html