SINCE its initial report in the literature in 1993, sentinel lymph node biopsy has become a common technique for determining stage for patients with breast cancer. 1The surgery involves the injection of the ipsilateral breast with isosulfan dye or radioactive technetium sulfur colloid to locate the node or nodes for biopsy. We report two cases of factious desaturation during this type of operation that probably occurred because of interference with pulse oximetry by isosulfan dye absorption.

Case Reports

Case 1

A 58-yr-old woman was scheduled to undergo a left lumpectomy and sentinel node biopsy. The patient was injected with radioactive technetium colloid approximately 2 h previously in the nuclear medicine department. General anesthesia was provided with oxygen, nitrous oxide, and desflurane, supplemented by fentanyl and rocuronium. The intraoperative course was uneventful until 8 min after the injection of 5 ml isosulfan dye into the subcutaneous tissue of the left breast, at which time the oxygen saturation, evaluated with an Ohmeda 5250 respiratory gas monitor (Ohmeda, Austell, GA), rapidly decreased to 91%. An arterial blood gas sample (Chiron 855; Ciba-Corning Diagnostics Corp., Medfield, MA) showed a pH of 7.59, partial pressure of arterial carbon dioxide (Paco2) was 27 mmHg, and partial pressure of arterial oxygen (Pao2) was 212 mmHg. The cooximeter reading was 92%. The operation proceeded without further problems, and oxygen saturation returned to baseline 45 min later.

Case 2

A 50-yr-old woman was scheduled to undergo a left-sided reexcision of breast tissue and sentinel node biopsy. The patient was injected with radioactive technetium colloid approximately 1.5 h before proceeding to the operating room. She was administered oxygen, nitrous oxide, and isoflurane, supplemented by fentanyl, midazolam, and rocuronium for general anesthesia. The operation proceeded uneventfully until 12 min after the injection of 5 ml isosulfan dye into the subcutaneous tissue of the left breast, at which time, oxygen saturation decreased to 89%. No cause could be deduced, and no corrective action successfully increased oxygen saturation. An arterial blood gas sample showed a pH of 7.42, a Paco2of 31 mmHg, and a Pao2of 189 mmHg. The cooximetry reading was 91%. The surgery was completed without further problems, and oxygen saturation returned to baseline 25 min after the initial event.


Axillary lymph node dissection has significant morbidity for patients with breast cancer, and it is often substituted with sentinel node biopsy. 2Isosulfan dye or radioactive technetium sulfur colloid is injected into the ipsilateral breast tissue to identify the node or nodes appropriate for excision. There are no reports in the literature of particular anesthesia concerns related to this procedure. There is one report of desaturation after injection of 3 ml lymphazurin dye during a lymphatic mapping procedure for cervical cancer. 3 

We experienced two episodes of factious desaturation in the initial 31 cases of sentinel node biopsy at our institution. These two cases seem to represent interference with pulse oximetry, because the Pao2remained in the normal range, rather than an absolute reduction in the oxygen transport by hemoglobin of the patient. Pulse oximetry uses spectrophotometric analysis and plethysmography to calculate oxygen saturation of hemoglobin at two wavelengths—usually red 660 and near-infrared 940. 4It has been reported previously that the accuracy of this monitor can be affected by indigo carmine, indocyanine green, nail polish, and methylene blue. 5,6The presence of any substance in blood that absorbs light in the red or infrared spectrum can alter oxygen saturation readings. We believe that isosulfan dye used in sentinel node biopsy can, in some instances, be absorbed in sufficient quantities as to present this problem for the anesthesiologist. After the occurrence of the two cases described, one of the authors volunteered to participate in an informal experiment. A peripheral venous indwelling catheter was inserted, and at 45 min intervals, various quantities of isosulfan dye were injected. An injection of 0.1 ml did not cause a change in Spo2. A dose of 0.25 ml resulted in a reduction in Spo2to 97% from 99% in less than 2 min. An Spo2of 94% was obtained after 0.5 ml isosulfan dye was administered. Last, when 1.0 ml dye was injected, the Spo2decreased to 89%.

Why these two particular cases showed pulse oximetry interference, probably because of greater absorption of dye, is unclear. In all 31 procedures, isosulfan dye volume was equal, and the surgeon reported that it was injected into the subcutaneous tissue. The time course to desaturation from injection of dye and from recovery to baseline in the two reported cases argues against an intravascular injection. Our informal investigation also seems to rule out intravascular injection, but it supports the hypothesis that small quantities of absorbed isosulfan dye can interfere with oxygen saturation monitoring. As anesthesiologists gain further experience with this new procedure, it may become evident which factors increase the risk for interference, and recommendations regarding the proper evaluation of similar episodes of desaturation after dye injection can be determined.


Krag DN, Weaver, DL, Alex JC: Surgical resection and radiolocalization of the sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993; 2: 335–40
Veronesi U, Paganelli G, Galimberti V: Sentinel node biopsy to avoid axillary dissection in breast cancer with clinically negative lymph nodes. Lancet 1997; 349: 1864–7
Coleman RL, Whitten CW, O’Boyle J, Sidhu B: Unexplained decrease in measured oxygen saturation following injection of lymphazurin 1% during lymphatic mapping. J Surg Oncol 1999; 70 (2): 126–9
Kelloher JF: Pulse oximetry. J Clin Monit 1989; 5: 37–40
Scheller MS, Unger RJ, Kelner MJ: Effects of intravenously administered dyes on pulse oximetry. A nesthesiology 1986; 65: 550–5
Cote CJ, Goldstein A, Fuchman WH: The effect of nail polish on pulse oximetry. Anesth Analg 1988; 67: 683–7