BACKGROUND: In breast cancer, lymphatic mapping with blue dye has been used intraoperatively to identify the sentinel lymph nodes. Several blue dyes are known to cause spurious desaturation when measured by a pulse oximeter. This study examined the differential effects of isosulfan blue (IB) and methylene blue (MB) on the level of oxygen saturation by pulse oximetry according to the sites measured. METHODS: Forty women with breast cancer were divided into two groups according to the dyes used. The level of oxygen saturation by pulse oximetry was measured on the index finger of the unoperated arm and on the 2nd toe, respectively, before injecting the dye and, 5, 10, 15, 20, 25, 30, 60, 90, 120, and 150 minutes after. The color of the skin was evaluated on a 4 point scale. RESULTS: In the IB group, the level of oxygen saturation on the finger was significantly lower immediately after injection than the level of oxygen saturation on the toe, and remained at a significantly lower level for up to 150 minutes. In the MB group, there was no significant reduction in the level of oxygen saturation at both sites after injecting the dye. The lowest oxygen saturation in each group was 95.8 +/- 1.9%, 97.5 +/- 1.2%, 99.0 +/- 1.0%, and 99.4 +/- 1.0% in the IB-finger, IB-toe, MB-finger, and MB-toe, respectively. The color change in the skin according to the dye injection was significant in the IB group but, not in MB group. CONCLUSIONS: IB decreased the level of oxygen saturation as measured by pulse oximetry, and its effects were more pronounced on the finger than toe. MB did not decrease the level of oxygen saturation as measured by pulse oximetry at both sites. Oxygen saturation monitoring on the toe is more desirable in a sentinel lymph node biopsy using IB than the finger.