SULFHEMOGLOBIN is a rare cause of cyanosis that is usually drug induced. The effects of sulfhemoglobin on pulse oximetry have not been reported widely. We present the case of a 48-yr-old woman who was scheduled to undergo palliative surgery. Before anesthesia, she had an oxygen saturation, measured by pulse oximetry (Spo2), of 85%. This apparent desaturation subsequently was discovered to be the result of a blood sulfhemoglobin concentration of 28%. She had been prescribed metoclopramide daily for more than 1 yr. Long-term ingestion of metoclopramide is a cause of drug-induced sulfhemoglobinemia.

The patient, a 48-yr-old woman with locally advanced carcinoma of the cervix, presented for palliative surgery for a rectovaginal fistula. At admission, she was being administered 100 mg slow-release morphine (MST) twice a day and 25 mg amitriptyline at night. For the previous year, she had also been prescribed 10 mg metoclopramide three times a day for persistent nausea and vomiting. At preoperative assessment, her cardiorespiratory function was normal, although she was anemic, with a hemoglobin concentration of 8.2 g/dl.

The patient was premedicated with 2 mg oral lorazepam 2 h preoperatively. At arrival in the operating room, a pulse oximeter (Hewlett Packard 54S; Hewlett Packard, Avondale, CA) was connected to her. Her Spo2while breathing room air was recorded as 85%, with the sensor on her index finger. At close observation, the patient was cyanotic but not dyspneic. A radial arterial cannula was inserted, and arterial blood was analyzed. Blood gas analysis (CIBA-Corning 288; CIBA-Corning Diagnostics, Medfield, MA) showed a pH of 7.41, a partial pressure of carbon dioide (Pco2) of 38 mmHg, and a partial pressure of oxygen (Po2) of 99 mmHg, with the patient breathing room air. A diagnosis of methemoglobinemia was presumed, so the sample was sent for analysis by hemoximeter (Chiron Rapidlab 865; Bayer Diagnostics, Tarrytown, NY). This showed a methemoglobin concentration of 0.3% and a carboxyhemoglobin concentration of 0.3%, but a sulfhemoglobin concentration greater than 1.5%. Because the hemoximeter was able to indicate only the presence of sulfhemoglobin, rather than an absolute concentration, anesthesia was postponed for further investigation.

A sample of arterial blood was analyzed by use of a manual assay with a variable wavelength spectrophotometer (Cecil Series 2; Cecil Instruments, Cambridge, UK); this showed a sulfhemoglobin concentration of 28%. 1The patient’s surgery was planned for the week after blood transfusion. However, the patient decided against operative treatment. She was discharged to outpatient hospice care and declined further investigation of her sulfhemoglobinemia.

In the case described, the presence of significant pulse oximetry desaturation associated with a normal arterial oxygen tension (Pao2) alerted us to the possibility of an abnormal hemoglobin species interfering with the pulse oximeter. Pulse oximeters use two light wavelengths (660 and 940 nm) to determine the ratio of pulse-added absorbencies. 2This ratio is associated with Spo2by means of a table derived from data from healthy volunteers. Dyshemoglobin molecules that have light absorbance peaks at 660 or 940 nm affect the ratio of light absorbencies at these wavelengths and lead to spurious Spo2readings. Methemoglobin has significant absorbencies at both of these wavelengths and has been reported widely to interfere with pulse oximetry. 3,4Sulfhemoglobin has a greater absorbance at 660 nm than do oxyhemoglobin, deoxyhemoglobin, and methemoglobin. 5,6We could not find data relating to the absorbance of sulfhemoglobin at 940 nm. Therefore, it is difficult to predict how increasing concentrations of sulfhemoglobin might interfere with pulse oximetry. The effects of increasing concentrations of methemoglobin on pulse oximetry have been determined experimentally in dogs. 4There is no similar report available for sulfhemoglobin, possibly because of the difficulty of inducing sulfhemoglobinemia in vivo . In a literature search, we could find only two cases of sulfhemoglobinemia in which the use of pulse oximetry was reported. The first report described an Spo2of 88%, with a sulfhemoglobin concentration of more than 1.5%, as indicated by hemoximetry. 7Unfortunately, although the presence of sulfhemoglobin was proven by gas chromatography, no absolute concentration was reported. The second report described an Spo2of 92–94%, with a sulfhemoglobin concentration of 16%, as measured by spectrophotometry. 8This compares with an Spo2of 85% associated with a sulfhemoglobin concentration of 28%, as seen in the current patient.

Hemoximeters use multiple wavelengths to determine concentrations of oxyhemoglobin, deoxyhemoglobin, carboxyhemoglobin, and methemoglobin. Some machines do not distinguish between methemoglobin and sulfhemoglobin because of their similar absorption peaks. 9This has caused inappropriate treatment of sulfhemoglobinemia with methylene blue. 7Some machines, such as the Chiron Rapidlab 865, indicate the presence of sulfhemoglobin but cannot quantify the concentration. The laboratory measurement of sulfhemoglobin relies on an absorption peak at 620 nm, which, unlike methemoglobin, persists after the addition of cyanide or dithionate. 1 

Sulfhemoglobinemia is recognized as a rare cause of cyanosis, including intraoperative cyanosis. 10Sulfhemoglobin is a stable, green-pigmented molecule that lasts the lifetime of the erythrocyte. Certain drugs and chemicals cause oxidation of hemoglobin, which, with the addition of a sulfur atom, forms sulfhemoglobin. The most common drugs known to cause sulfhemoglobinemia are phenacetin, dapsone, and the sulfonamides. It has also been described with occupational exposure to sulfur compounds and with drug abuse. 11Long-term metoclopramide ingestion has been reported as a cause of sulfhemoglobinemia. 12Short-term high-dose metoclopramide therapy combined with N -acetylcysteine has also caused sulfhemoglobinemia. 7,8Our patient had been prescribed 30 mg/day metoclopramide for more than 1 yr. She was not administered any other drugs known to cause sulfhemoglobinemia. Metoclopramide is structurally related to aniline dyes and prilocaine, both of which cause methemoglobinemia. Why the same drugs can cause methemoglobinemia in some patients and sulfhemoglobinemia in others is not clear.

Central cyanosis is surprisingly difficult to detect, especially if, as in this case, the patient is asymptomatic. Only 0.5 g/dl sulfhemoglobin is needed to cause clinically detectable cyanosis, as compared with 1.5 g/dl methemoglobin and 5 g/dl deoxygenated hemoglobin. 5Sulfhemoglobin cannot carry oxygen; however, high concentrations of sulfhemoglobin are well-tolerated, despite the resulting physiologic anemia. This is caused by a right shift in the hemoglobin–oxygen dissociation curve of the normal heme in the presence of sulfhemoglobin, thus facilitating tissue oxygenation. This is in contrast to methemoglobin, which causes a left shift in the curve. This can result in severely impaired tissue oxygenation at higher concentrations. There is no specific treatment for sulfhemoglobinemia, other than removing the suspected cause. The concentration of sulfhemoglobin decreases as erythrocytes are destroyed and replaced.

The authors thank Mr. Desmond Barton and Mr. Jhangir Iqbal, The Royal Marsden Hospital, London, United Kingdom, for their help with this case report.

Tietz NW: Textbook of Clinical Chemistry. Philadelphia, WB Saunders, 1986, pp 1536–40
Pologe JA: Pulse oximetry: Technical aspects of machine design. Int Anesthesiol Clin 1987; 25: 137–53
Eisenkraft JB: Pulse oximeter desaturation due to methemoglobinemia. A nesthesiology 1988; 68: 279–82
Barker SJ, Tremper KK, Hyatt J: Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. A nesthesiology 1989; 70: 112–7
Finch CA: Methemoglobinemia and sulfhemoglobinemia. New Engl J Med 1948; 239: 470–8
Siggard-Andersen O, Norgaard-Pedersen B, Rem J: Hemoglobin pigments. Spectrophotometric determination of oxy-, carboxy-, met-, and sulfhemoglobin in capillary blood. Clin Chim Acta 1972; 42: 85–100
Lu HC, Shih RD, Marcus S, Ruck B, Jennis T: Pseudomethemoglobinemia. Arch Pediatr Adolesc Med 1998; 152: 803–5
Langford SL, Sheikh S: An adolescent case of sulfhemoglobinemia associated with high-dose metoclopramide and N-acetylcysteine. Ann Emerg Med 1999; 34: 538–41
Wu C, Kenny M: A case of sulfhemoglobinemia and emergency measurement of sulfhemoglobin with an OSM3 CO-oximeter. Clin Chem 1997; 43: 162–6
Schmitter CR: Sulfhemoglobinemia and methemoglobinemia: Uncommon causes of cyanosis. A nesthesiology 1975; 43: 586–7
Park CM, Nagel RL: Sulfhemoglobinemia. New Engl J Med 1984; 310: 1579–84
Van Veldhuizen PJ, Wyatt A: Metoclopramide induced sulfhemoglobinemia. Am J Gastroenterol 1995; 90: 1010–1