Sentinel Lymph Node Imaging with Microbubble Ultrasound Contrast Material 1 Robert F. Mattrey, MD, Yuko Kono, MD, Kris Baker, Tom Peterson RATIONALE AND OBJECTIVES Lymphadenectomy is necessary to provide local con- trol and staging of breast cancer patients because the de- gree of nodal involvement remains the most important prognostic indicator (1). Because morbidity with axillary dissection occurs in as many as 20% of patients (2,3), attempts at limiting dissection have led to the develop- ment of sentinel node resection. The technique was popu- larized by Morton and associates (4) for staging mela- noma, and Giuliano and associates (5) applied it to breast cancer. They showed that, when the sentinel node was negative, the remainder of the downstream nodes were negative in 126 of 127 cases. When the node was posi- tive, it was the only positive node in over 60% of cases (4) and contained five times more micrometastasis than nonsentinel nodes (6). The difficulty of the procedure lies in the localization and identification of the sentinel node. As described by Giuliano and associates (5), the proce- dure begins with the injection of 3 to 5 mL of isosulfan blue, a water-soluble dye, in the breast mass and sur- rounding tissue. Approximately 5 min later, blunt dissec- tion is made to locate a blue lymphatic channel or a blue node. Although all blue nodes are removed, an attempt is made to follow the feeding lymphatic channel toward the mass to ensure the identification of the first and true sen- tinel node. The resected nodes are assessed histologically. If no cancer deposits are found, dissection is terminated; otherwise, classic axillary dissection follows. Much in this technique bears refinement, however, since Giuliano, the most experienced investigator, reported that the sentinel node was detected in 58% in the first 87 cases and 78% in the next 87 cases. The failure is that nodes are indistin- guishable from breast tissue unless colored blue, the dye has unpredictable and rapid clearance, and the drainage pattern varies among patients. The rapid clearance of the blue dye provides a short time window of only a few minutes between operating too early when no nodes are stained, or too late when too many nodes are stained. Radiopharmaceutical colloids are available that provide some preoperative localization as they flow through the lymphatic chain (7,8). Further, with the aid of a gamma pencil probe, pinpoint localization of radioactivity could lead to the nodes intraoperatively. Although radiolabeled colloids have a more delayed transit and provide a skin marking option, they are less than ideal. The fluid is in- visible intraoperatively, many nodes are enhanced, and, more importantly, the proximity of the injection site to the nodes decreases the target-to-background ratio, de- creasing sentinel node specificity (9,10). At present, most centers use both the blue dye and the radiolabeled colloid methods to gain sensitivity. Particles injected subcutaneously enter the lymph ves- sel through gaps between lymphatic endothelial cells or by transcellular endo- or exocytosis. On average, smaller particles (10 to 40 nm) are more likely to enter than larger particles. As particles approach 1 m, their uptake into lymphatics is very poor and must be carried away by phagocytes or reduced in size by local processes. In fact, over 95% of particles larger than 400 nm stay at the in- jection site, whereas 74% of particles 10 times smaller (40 nm) are absorbed (11). We hypothesized that 2- to Acad Radiol 2002; 9(suppl 1):S231–S235 1 From the Department of Radiology, University of California, San Diego, MRI Insti- tute, 410 Dickinson Street, San Diego, CA 92103-8756. Supported in part by ROI- CA36799 and Alliance Pharmaceutical Corp. Equipment loan from Siemens Ultra- sound. Address correspondence to R.F.M. R.F.M. is a consultant to APC. © AUR, 2002 S231