Full text of DOI:10.1007/BF03019812

38
GEN ERAL AN ETH ESIA
Intraoperative decrease
in pulse oximeter read-
ings following injection
of isosulfan blue
Robert W. H oskin MD PhD ,*
Robert Granger MD†
Purpose: Sentinel lymph node mapping, involving injection of isosulfan blue dye around a tumour, is beginning to
be used in patients with carcinoma of the breast. Absorption of the dye into the circulation may interfere with
pulse oximetry, causing falsely low readings. This report describes changes in pulse oximeter readings following
injection of isosulfan blue for sentinel lymph node mapping in a patient with carcinoma of the breast.
Clinical Features: An 83-yr-old female patient underwent sentinel node biopsy of the axilla followed by partial
mastectomy for carcinoma of the left breast. Isosulfan blue was injected in the area of the tumour in the left breast.
The SpO 2 began to decrease 15 min after dye injection, reaching a nadir of 89-90% 30 min after injection. Arterial
blood gas analysis showed normal arterial partial pressure of oxygen. Pulse oximeter readings did not return to
normal until more than six hours after dye injection.
Conclusion: Review of the literature reveals a small number of case reports of similar occurrences of low pulse
oximeter readings following injection of isosulfan blue or patent blue dye for lymphatic mapping. Data from these
reports and the case described here suggest that the latency, magnitude and duration of effect on pulse oximeter
readings following injection of these dyes is highly variable. It is important to rule out other causes of low pulse
oximeter readings when this effect occurs; normal oxygenation can be verified with arterial blood gas analysis. Co-
oximetry can be done to rule out methemoglobinemia as a cause of decreased SpO 2.
Objectif : La cartographie des ganglions lymphatiques de Troisier, comportant l’injection de colorant bleu iso -
sulfane autour d’une tumeur, est utilisée depuis peu dans les cas de cancer du sein. L’absorption du colorant dans
la circulation peut interférer sur l’oxymétrie de pouls en causant de fausses valeurs à la baisse. Le présent article
décrit ce genre de changements survenus à la suite de l’injection de colorant bleu isosulfane pour la cartographie
d'un ganglion de Troisier chez une patiente atteinte d’un cancer du sein.
Éléments cliniques : Une femme de 83 ans a subi, à l’aisselle, une biopsie d’un ganglion de Troisier suivie d’une
mastectomie partielle pour un cancer du sein gauche. Du bleu isosulfane a été injecté autour de la tumeur. La
SpO2 a commencé à baisser 15 min après l’injection du colorant et a atteint un creux de 89-90 %, 30 min après
l’injection. La gazométrie artérielle a montré une SpO2 normale. Les valeurs de l’oxymétrie pulsée sont revenues
à la normale plus de six heures après l’injection de colorant.
Conclusion : On retrouve peu de cas documentés de baisse semblable des valeurs d’oxymétrie pulsée après
l’injection de bleu isosulfane ou de colorant bleu spécial pour la cartographie lymphatique. Les données recueil-
lies laissent croire que le temps de latence, l’importance et la durée de l’effet des valeurs de l’oxymétrie pulsée
réalisée après une injection de ces colorants sont très variables. Il faut exclure d’autres causes de la baisse des
relevés d’oxymétrie quand cet effet se manifeste; l’oxygénation normale peut se vérifier par la gazométrie
artérielle. La co-oxymétrie peut permettre d’écarter la possibilité de méthémoglobinémie comme cause de la
baisse de SpO 2.
From the Department of Anaesthesia,* and Department of General Surgery,† St. Mary’s Hospital, 220 Royal Ave., New Westminster,
B.C., V3L 1H6 Canada.
Address correspondence to: Dr. R. Hoskin. Phone: 604-527-3251; Fax: 604-527-3223; E-mail: rhoskin@home.com
Accepted for publication October 9, 2000.

H oskin & Granger: I SO SU L F AN BLU E AND PU LSE O XIMETRY
39
ULSE oximetry relies on the differing absorp-
tion of red light by oxyhemoglobin and deoxy-
hemoglobin, allowing calculation of the
ration (SaO ) of 99.3% with inspired oxygen fraction of
0.4. Co-oximetry was not available.
2
At the completion of surgery 90 min after the dye
injection, SpO2 was measured as 93%. The patient was
stable in the recovery room and was transferred to the
surgical ward on oxygen via nasal cannula at 2 l·min-1.
Pulse oximetry measurement on the ward showed that
the patient’s SpO2 returned to 99% by seven hours fol-
lowing dye injection. The greyish discolouration of
her skin was beginning to fade by this time.
P
percentage of hemoglobin that is saturated
with oxygen (SpO2 ). Any substance which absorbs light
in the red wavelengths may interfere with the ability of a
pulse oximeter to measure oxygen saturation accurately.
Dyes such as methylene blue and patent blue five have
been reported to cause artificially low SpO2 readings.1 – 5
Recently, isosulfan blue (Lymphazurin 1%) has
begun to be used to facilitate sentinel lymph node
biopsy for carcinoma of the breast. In this technique,
dye is injected into the area surrounding the tumour.
Dye is absorbed into the lymphatics and causes visible
discolouration of the lymphatic channels and nodes
draining the area of the tumour. Selective biopsy and
histopathological analysis of lymph nodes stained by
the dye (“sentinel nodes”) may provide diagnostic
information of equal value to that provided by more
extensive axillary node biopsy. One report has been
published of intraoperative decrease in SpO2 following
injection of isosulfan blue into the uterine cervix.6
Reports of this dye’s ability to affect pulse oximetry
during breast surgery have not appeared in the litera-
ture to date. Therefore, report of a case in which iso-
sulfan blue was used intraoperatively for sentinel node
biopsy for carcinoma of the breast is presented below.
Discussion
Pulse oximeters use two wavelengths of light, usually
660 and 940 nm, to determine oxyhemoglobin content
of blood via a spectrophotometric process. Pulse oxime-
ters may be unable to accurately determine oxyhemo-
globin content in the presence of dyes which absorb red
light (for example, isosulfan blue which has its peak
absorption at 646 nm),6 or in the presence of carboxy-
hemoglobin or methemoglobin. Measurement of arte-
rial oxygen saturation (SaO2) and PaO2 of an arterial
blood sample using a blood gas analyzer permits differ-
entiation between decreases in measured SpO2 due to
absorption of the dye into the circulation and true
decreases in SpO2. In addition, some blood gas analyz-
ers incorporate co-oximeters which use four or more
light wavelengths in order to differentiate among oxy-
hemoglobin, deoxyhemoglobin, carboxyhemoglobin
and methemoglobin. Co-oximetry would allow methe-
moglobinemia to be ruled out as a cause of spurious
decreases in SpO2. This may be relevant as methylene
blue may cause methemoglobinemia when adminis-
tered in high doses, although there is no mention in the
literature of isosulfan blue causing this effect.
Most previous reports of changes in SpO2 resulting
from dyes have described transient decreases in SpO2
which resolved in minutes,1,4 although Morell et al.
reported a decrease in SpO2 of several hours’ duration
following intradermal injection of patent blue dye.5 The
only other report of changes in SpO2 after injection of
isosulfan blue described a decrease in SpO2 from 98% to
89-90% beginning five minutes after injection of the
dye into the uterine cervix and lasting for 10 min.6 In
the patient described in the present report, SpO2 did
not return to normal until more than six hours after the
dye had been injected. The reason for the longer dura-
tion of the dye’s effect on SpO2 in the present case is
uncertain. Coleman et al.6 suggested that, as isosulfan
blue is eliminated primarily by biliary excretion, patients
with impaired hepatobiliary function may demonstrate
prolonged duration of action of effects of the dye when
absorbed intravascularly. While liver function tests were
not done on the patient described in the present report,
Case report
An 83-yr-old woman with biopsy-proven adenocarcino-
ma of the left breast was booked to have a sentinel node
biopsy of the left axilla followed by a partial left mastec-
tomy. General anesthesia was induced with propofol and
fentanyl and maintained with desflurane, oxygen and
nitrous oxide. Inspired fraction of oxygen was main-
tained at 0.40. Muscle relaxation was maintained with
rocuronium. The patient was hemodynamically stable
and SpO2, measured with an Ohmeda RGM 5250 ana-
lyzer, remained at 98-99%, The surgeon injected 5 ml of
1% isosulfan blue (Lymphazurin 1%, U.S. Surgical
Corporation, Norwalk, CT) around the tumour mass,
aspirating carefully to avoid intravascular injection.
Fifteen minutes after the dye injection, the pulse oxime-
ter reading began to decrease, reaching a minimum of
89-90% 30 min after the injection. Blood pressure and
heart rate remained stable. Other causes of intraopera-
tive desaturation such as pneumothorax, malposition of
the endotracheal tube or bronchospasm were ruled out
clinically. A portion of the patient’s face was uncovered,
revealing a greyish discolouration of the skin. Arterial
blood gas analysis performed 35 min after dye injection
revealed a pH of 7.427, PaCO2 of 39.9 mm H g, PaO2
of 190.2 mm H g, H CO3 of 26.3 mEq·l- 1 and O2 satu-

40
C AN AD I AN JO U RN AL O F AN ESTH ESIA
she had no history of hepatic or biliary disease and no
clinical findings to suggest that this may have been the
case. It is possible that in previous reports the absorp-
tion of dye may have occurred over a brief period of
time, producing in effect an intravenous bolus of dye,
whereas in the present case absorption may have con-
tinued over a more prolonged period of time.
isosulfan blue dye into the breast for sentinel lymph
node mapping. The most likely explanation for this
change is absorption of dye into the circulation from
the injection site. Arterial blood gases showed PaO2 and
SaO2 consistent with the oxygen delivery, confirming
the suspicion that the pulse oximeter was delivering
falsely low readings. H owever, when intraoperative
desaturation occurs, it is important to rule out possible
causes of desaturation such as pneumothorax, bron-
chospasm or endotracheal tube malpositioning. The
duration of the effect on SpO2 in this patient was
greater than six hours, much longer than the five min-
utes duration reported by Coleman et al.6 following iso-
sulfan blue injection into the cervix. Data from the
present case and others reported in the literature3,5,6
indicate that, following dye injection for lymphatic
mapping, the magnitude and duration of effect on
SpO2 are highly variable.
The value of sentinel node biopsy in predicting axil-
lary disease in patients with carcinoma of the breast is
uncertain. A multicentre NSABP (National Surgical
Adjuvant Breast and Bowel Project) study is underway
to evaluate this. If the sensitivity and specificity of sen-
tinel node biopsy prove to be high, it is not unreason-
able to expect that most axillary node dissections will be
replaced by sentinel node biopsies. Therefore, the type
of incident described in this report will probably
become increasingly common. H owever, the frequency
with which decreased SpO2 occurs following injection
of isosulfan blue into the breast has not been reported.
Finally, although dye absorption may be suspected as
the cause of an intraoperative decrease in SpO2, it is
important to rule out causes of real arterial hypoxemia.
In the case reported here, checking the FI O2 displayed
on the Ohmeda 5250 gas analyzer and the O2 analyzer
on the anaesthesia machine did not indicate delivery of a
hypoxic gas mixture. Increasing the FI O2 to 1.0 to rule
out delivery of a hypoxic gas mixture could also be con-
sidered, although this was not done in the case reported
here. Integrity of the anesthetic breathing circuit should
be checked. Inadvertent endobronchial intubation
should be ruled out by checking for bilateral breath
sounds and verifying that the endotracheal tube is appro-
priately positioned. A suction catheter can be passed
down the tube to rule out kinking or obstruction.
Pneumothorax should be ruled out by checking for
equal air entry bilaterally, absence of tracheal deviation,
and absence of hyperresonance on percussion of the
chest. Chest X-ray can be performed if necessary. If
pneumothorax is suspected, administration of nitrous
oxide should be discontinued and, if tension pneumoth-
orax is suspected, needle decompression of the affected
hemithorax should be performed immediately. Breath
sounds should be auscultated for expiratory wheeze or
prolongation of the expiratory phase, and peak inspira-
tory pressure and shape of the expired CO2 waveform
should be checked to rule out bronchospasm. Causes of
ventilation-perfusion mismatch such as atelectasis or pul-
monary edema must be ruled out. The pulse oximeter
probe may be replaced or moved to another site on the
patient to rule out probe malfunction or poor sensing
due to local hypoperfusion or vasoconstriction.
Addendum
During revision of this case report, another incident of
intraoperative desaturation following administration of iso-
sulfan blue for axillary sentinel node mapping was
observed in another hospital. The SpO decreased from
2
99% to 89% despite increasing inspired O from 40% to
2
100%. Aterial blood gas analysis with co-oximetry was
available at this hospital, and showed a PaO of 449 mm
2
Hg and methemoglobin, expressed as a fraction of total
hemoglobin, of 0.005 (normal range 0 to 0.020). This
demonstrates that co-oximetry was able to rule out methe-
moglobinemia as a cause of decreased SpO in this patient.
2
References
1 Scheller, MS, Unger RJ, Kelner MJ. Effects of intra-
venously administered dyes on pulse oximetry readings.
Anesthesiology 1986; 65: 550–2.
2 Kessler MR, Eide T, Humayun B, Poppers PJ. Spurious
pulse oximeter desaturation with methylene blue injec-
tion. Anesthesiology 1986; 65: 435–6.
3 McEwan, D, Lam K. Oximetry and patent blue five
dye (Letter). Anaesth Intensive Care 1997; 25: 587–8.
4 Saito S, Fukura H, Shimada H, Fujita T. Prolonged
interference of blue dye “patent blue” with pulse oxime-
try readings. Acta Anaesthesiol Scand 1995; 39: 268–9.
5 Morell RC, Heyneker T, Kashtan HI, Ruppe C. False
desaturation due to intradermal patent blue five dye.
Anesthesiology 1993; 78: 363–4.
6 Coleman RL, Whitten CW, O’Boyle J, Sidhu B.
Unexplained decrease in measured oxygen saturation by
pulse oximetry following injection of Lymphazurin 1%
(isosulfan blue) during a lymphatic mapping procedure.
J Surg Oncol 1999; 70: 126–9.
In summary, a case is presented in which intraopera-
tive decrease in SpO2 occurred following injection of