A 17-yr-old female was anaesthetized for removal of a cervical carcinoma using sevoflurane, sufentanil and rocuronium with monitoring consisting of electrocardiogram, pulse oximetry, end-tidal CO2 (PETCO2) and invasive arterial pressure (AS3®, Datex, Helsinki, Finland). A blood-gas analysis (Table 1) performed at the beginning of the operation was normal (Radiometer® ABL 625, Copenhagen, Denmark). During the operation, patent blue V dye (Guerbet BP 50400, Roissy, France), 100 mg (in 4 mL), was injected into the tumour for sentinel node detection. Shortly after the injection, the pulse oximetry readings decreased from 99 to 90% and afterwards to 85%, although arterial pressure, heart rate, PETCO2 and airway pressure were unchanged. Increasing the fraction of inspired oxygen (FiO2) to 1.0 did not improve the pulse oximetry readings. Blood-gas analysis revealed that PaO2, pH and lactate concentration were within the normal range, but that arterial saturation was only 82% and methaemoglobin was 12% (Table 1). Methaemoglobin analysis was then repeated in the main laboratory using a multifrequency co-oximeter (LS 500®, Lange, Berlin, Germany). This analyser also detected methaemoglobinaemia, albeit at the lower level of 3%. Since the patient showed no clinical signs of hypoxaemia and because lung ventilation, haemodynamic and blood-gas status, and lactate concentration were all normal, the operation was continued. Blood-gas analysis performed 120 min after the patent blue V dye injection still showed signs of interference (Table 1). Emergence from anaesthesia and the postoperative period were uneventful. Skin and urine discoloration occurred and persisted for 24-36 h.
We found that the injection of patent blue V dye, used intraoperatively for sentinel node identification, interfered with pulse oximetry and co-oximetry during anaesthesia. Of interest was the high level of methaemoglobinaemia measured with co-oximetry. Dye injection as a means of facilitating sentinel node identification has been used since the early 1990s. The method is especially useful for identification of sentinel nodes for malignant melanoma and breast carcinoma, but it is currently also used to identify other carcinomas. Any dye used during the procedure can potentially interfere with pulse oximetry devices which use two wavelengths (660 and 940 nm) to differentiate between oxygenated and deoxygenated haemoglobin. Dyes that absorb light near either wavelength will potentially interfere with pulse oximetry . Patent blue V dye has its maximum absorbance at 640 nm and its presence in blood leads to increased absorbance of the 660 nm light emitted by the pulse oximeter. This increased absorbance of the 660 nm light leads to falsely low pulse oximetry readings .
In our patient, patent blue V dye also led to an increase in methaemoglobin - identified by co-oximetry, which detected methaemoglobin concentrations >12%. Only a multifrequency analysis of oxygenated blood showed this to be a falsely high amount, but even with the multifrequency analyser, the concentration of methaemoglobin was still reported as 3%.
Methaemoglobin is the result of oxidation of the iron atom in haemoglobin - an increase in methaemoglobin concentration during an anaesthetic procedure is usually a consequence of oxidation of the iron atom through drugs such as prilocaine . Methaemoglobin is measured with a co-oximeter at wavelengths near 635 nm - the maximum absorbance of methaemoglobin . Since the peak absorbance of patent blue dye is at 640 nm, it is not surprising the co-oximeter falsely detected increased amounts of ‘methaemoglobin’ in blood. Larsen and colleagues found that patent blue V dye could lead to an overestimation of methaemoglobin by 56% . However, whether any portion of the increase in detected methaemoglobin can be attributed to a true methaemoglobin increase cannot be discerned from laboratory analysis: the proximity of peak absorbance between patent blue V dye and methaemoglobin interferes with most analysers. A recent Editorial in this journal stressed that the anaesthesiologist should be aware of this dye and its interference with pulse oximetry , and our case report proves that point.
*Klinikum der Friedrich-Schiller-Universitaet, Klinik für Anaesthesiologie und Intensivtherapie, Jena, Germany
†Zentralklinik Bad Berka, Klinik für Anaesthesie und Intensivtherapie, Bad Berka, Germany
1. Rizzi RR, Thomas K, Pilnik S. Factious desaturation due to isosulfan dye injection. Anesthesiology
2. McEwan D, Lam D. Oximetry and patent blue five dye. Anaesth Intens Care
3. Rudlof B, Lambert R, Brandt L. Untersuchungen zum Einsatz der Pulsoxymetrie bei prilocaininduzierter Methämoglobinämie. Anaesthesist
4. Thomas L. Hämoglobine. In: Thomas L, ed. Labor und Diagnose: Indikation und Bewertung von Laborbefunden für die medizinische Diagnostik
. Frankfurt am Main, Germany: TH-Books, 2000: 487-491.
5. Larsen VH, Freudendal-Pedersen A, Fogh-Andersen N. The influence of patent blue V on pulse oximetry and haemoximetry. Acta Anaesthesiol Scand
1995; 107 (Suppl)
6. Cashman JN. Editorial: Sentinel lymph node biopsy: anaesthetic implications. Eur J Anaesthesiol