Purpose: To evaluate the reproducibility of lymphoscintigraphic results in assessment of the location and number of sentinel nodes in patients with breast cancer.
Materials and methods: Twenty-five patients with breast cancer were prospectively enrolled in this study. Lymphoscintigraphy was performed after intratumoral injection of about 130 MBq of technetium 99m nanocolloid. Anterior and lateral images were obtained 20 minutes and 2 and 4 hours after injection. The following day, scintigraphy was repeated after a second injection of the radiolabeled colloid in an identical fashion and was preceded by acquisition of a starting image. Two observers evaluated the paired images independently, and count rates were calculated from the images. Correlation coefficient and Bland-Altman methods were used to analyze the paired count rates.
Results: At least one sentinel node was visualized at lymphoscintigraphy in all 25 patients. Drainage to the axilla was observed in 17 patients; drainage to the axilla and extraaxillary basins, in seven patients; and drainage exclusively to extraaxillary sentinel nodes, in one patient. The second scintigraphic study revealed the same drainage pattern in all 25 patients (reproducibility, 100%; 95% CI: 86%, 100%). The Pearson correlation coefficient of the paired count rates was 0.54 (P <.001). Count rates at repeat scintigraphy were 23%-417% of the count rates at first scintigraphy in 95% of cases.
Conclusion: Results of lymphoscintigraphy for lymphatic mapping in breast cancer are highly reproducible for assessment of the number of sentinel nodes.
Copyright RSNA, 2003.