The three-dimensional localization of internal mammary lymph nodes by radionuclide lymphoscintigraphy.

In breast cancer patients, radiation therapy planning must account for individual anatomy to ensure optimal coverage of tumor and internal mammary nodes. To achieve this, three-dimensional radionuclide lymphoscintigraphy (RNLS) was performed in 167 patients by obtaining two images of the nodes using a 30-degree slant hole collimator rotated 180 degrees between images. Analysis of 768 nodes (mean 4.6/patient) visualized from the level of rib 1 through interspace 5 was performed. The number of nodes seen was not a function of patient age. Cross-communication to the contralateral node chain occurred in 13.8% of cases. Eighty-two percent of nodes were located near the first three ribs and interspaces; 23% were located beyond 3.0 cm from the mid-sternal line. At the level of the radiation beam match line (second rib or interspace), 4.5% of nodes were deeper than 3.0 cm. From rib 3 through interspace 5, 3.9% were deeper than 3.0 cm. Using an idealized tangential field, at least one node would have been missed in 16.2% of patients. Three-dimensional RNLS allows definition and localization of normal sized nodes and ensures that radiation therapy portals can be tailored for each individual under treatment.

Clinical comparison of two radiocolloids for internal mammary lymphoscintigraphy.

Recent studies suggest that 99mTc-labeled radiocolloid (SC) compounded with hydrogen sulfide can be used to visualize lymph channels and nodes. Our study prospectively compared SC with 99mTc antimony sulfide (SbS) colloid, in 28 patients undergoing internal mammary lymphoscintigraphy. Images were recorded on a scintillation camera and computer at 0.5, 1.0, and 3.0 hr. Quantitative analysis included assessment of percent (%) injected dose in nodes, the percent remaining at the injection site, and the relative intensity of the most cephalad node compared to a 57Co standard. The mean (means) % injected dose of both radiocolloids within visualized nodes was less than 1% at each time interval, with no significant differences between means's. The means % injected dose remaining at the injection site at 3.0 hr was 83 for SbS and 76 for SC not statistically significant (N.S.). The means of the ratio of counts within the most cephalad node at 3.0 hr to counts within a 57Co standard was 0.98 for SbS and 1.03 for SC (N.S.). Clinical assessment of number of nodes visualized and extent of radiocolloid migration showed no difference between the two agents. The biological and clinical parameters for the two colloids appear similar when used for internal mammary lymphoscintigraphy.