Case 23-2004 — A 50-Year-Old Woman with Low Oxygen Saturation
http://www.100md.com
《新英格兰医药杂志》
Presentation of Case
A 50-year-old woman was evaluated in the pulmonary clinic because of low oxygen saturation.
Five years earlier, dysfunctional uterine bleeding and dysmenorrhea had developed. A pelvic ultrasonographic examination at that time had revealed uterine fibroids. Four years later, the woman was found to be anemic; an endometrial biopsy showed no evidence of cancer. A repeated pelvic ultrasonographic examination revealed a slightly enlarged uterus, 11 cm in length, and a posterior submucous uterine fibroid, 35 by 30 by 41 mm, that displaced the endometrium anteriorly. Her symptoms were controlled for a time with ibuprofen. However, the pain increased, and after she did not have a response to treatment with medroxyprogesterone acetate, she was scheduled for dilatation and curettage, hysteroscopy, and possible resection of the fibroid.
A tonsillectomy had been performed 27 years earlier and a tubal ligation 6 years after that. The patient was noted to have coughing and cyanosis in the recovery room after the tubal ligation. She also had hypertension of recent onset; a history of a heart murmur, with a report of an echocardiogram revealing trace mitral regurgitation; a history of atypical chest pain (with a normal electrocardiogram) that had not recurred; and asthma, diagnosed two years previously, after she had wheezing with a respiratory infection. Her medications included atenolol (25 mg daily), ibuprofen, and an iron supplement. There was a family history of asthma and coronary artery disease, including a myocardial infarction in her 72-year-old mother, but there was no family history of any adverse reactions to general anesthesia. She had no allergies, did not smoke cigarettes, and did not abuse drugs or alcohol.
The patient was seen in the outpatient preadmission testing area. The blood pressure was 162/89 mm Hg, the heart rate 97 beats per minute, the respiratory rate 26 breaths per minute, the temperature 37.2°C, the height 153 cm, and the weight 75 kg. Auscultation of the chest revealed prolonged expiration, with clear lungs. The cardiac rhythm was regular, and there was a systolic ejection murmur (grade 2 to 3) at the base. The remainder of the findings on physical examination were normal. Oxygen saturation by digital pulse oximetry (SpO2) was 93 percent bilaterally while the patient was breathing ambient air. The results of laboratory tests are shown in Table 1. An electrocardiogram showed a normal sinus rhythm with clockwise rotation; there was no change from the electrocardiogram that had been obtained 14 years earlier. Chest radiography disclosed clear lungs; a heart of normal size and contour, with no evidence of congestive heart failure; and no abnormalities of the hilar, mediastinal, pleural, or bony structures.
Table 1. Results of Laboratory Tests.
Pulmonary-function tests performed five days after the outpatient examination showed normal lung volumes, normal single-breath carbon monoxide diffusion capacity, and a reduced SpO2 (91 percent) while the patient was breathing room air. An echocardiogram showed normal left and right ventricular function; there was no evidence of a shunt after an intravenous injection of agitated saline. Slight calcification of the posterior mitral annulus and left atrial dilatation, as well as trace mitral regurgitation (without evidence of mitral-valve prolapse), trace tricuspid insufficiency, and trace pulmonary insufficiency were observed on color and spectral Doppler imaging. No evidence of aortic insufficiency or pericardial effusion was seen. The estimated ejection fraction was 70 percent. The planned operation was postponed, and the patient was referred for a pulmonary evaluation.
Eight weeks later, she was evaluated in the outpatient pulmonary clinic. Additional information on her history was obtained; it was learned that the diagnosis of asthma had been made after she had moved into a dusty office and that her symptoms had resolved after she had moved out of the office. She had not had any other episodes of shortness of breath, and she was physically active. She was originally from Puerto Rico, had two children, and had no other history of occupational exposures to toxins or dust. On physical examination, she appeared well and was not in acute distress. Her lungs and heart were normal on auscultation, her fingers and toes showed no cyanosis or clubbing, and her arms and legs showed no edema. Her SpO2 while she was breathing ambient air in the supine and standing positions was 91 percent, and it remained in the 91 to 93 percent range with exertion. It increased to 97 to 98 percent while the patient was receiving 6 liters of oxygen by nasal cannula. The results of arterial blood gas studies and oximetry are shown in Table 2.
Table 2. The Results of Arterial Blood Gas Studies and Oximetry.
A diagnostic test was performed.
Differential Diagnosis
Dr. William E. Hurford: The key findings in this patient's preoperative evaluation were unexpectedly low SpO2 values on multiple occasions in the absence of known pulmonary or cardiac disease.
Preoperative Evaluation with Pulse Oximetry
The usefulness of pulse oximetry as a preoperative screening tool has never been examined prospectively. In their recent advisories on preoperative evaluation, neither the American Society of Anesthesiologists nor the American College of Cardiology recommends pulse oximetry for routine use.1,2 Nevertheless, the presumption that hypoxemia may contribute to perioperative morbidity and mortality has considerable empiric validity. The patient under discussion had a history of hypertension, a heart murmur, atypical chest pain, and asthma. She had become cyanotic in the recovery room after a previous operation. She also had a family history of asthma and coronary artery disease. Accordingly, the inclusion of pulse oximetry in her cardiopulmonary assessment was reasonable.
Causes of Low Oxygen Saturation on Pulse Oximetry
The low SpO2 value (93 percent) while the patient was breathing ambient air is unexpected, given that she did not have severe signs or symptoms of cardiac or pulmonary disease. What might explain this value? Hypoxemia is the most clinically important cause of a low SpO2, and pulmonary disease is the most common cause of hypoxemia (Table 3). The usual causes of hypoxemia include hypoventilation, mismatching of ventilation and perfusion, right-to-left shunting of blood, and diffusion abnormalities (Table 4). On initial examination, the patient was found to be slightly tachypneic and to have a prolonged expiratory time, but the lung fields were clear. The evaluation proceeded appropriately with chest radiography and tests of pulmonary function. Her chest radiograph was normal, and both spirometric values and the results of testing of the carbon monoxide diffusion capacity of the lung were normal. These tests effectively rule out common diseases such as emphysema or asthma as a cause of hypoxemia.
Table 3. Possible Causes of a Low Oxygen Saturation on Pulse Oximetry.
Table 4. Possible Causes of Hypoxemia.
Cardiac disease can also lead to hypoxemia, particularly when right-to-left shunting of blood or pulmonary edema is present. There was no clinical or radiologic evidence of pulmonary edema in this patient. An echocardiogram was obtained to assess the possibility of ventricular dysfunction or intracardiac shunting, and the images were normal. Right-to-left shunting of blood may also be due to pathologic intrapulmonary shunting, such as that occurring in pulmonary arteriovenous malformations or hepatopulmonary syndrome, but such processes would be unlikely in this patient, given her normal diffusing capacity and echocardiogram. Intrapulmonary shunting would have been detected by early opacification of the left ventricle after the intravenous injection of agitated saline. An explanation of her reduced SpO2 must lie elsewhere.
Limitations of Pulse Oximetry
Pulse oximetry does not directly measure oxygen saturation. Diodes on the oximeter probe emit two wavelengths of light (660 nm, which is red, and 940 nm, which is infrared) that pass through tissue (usually a finger or an earlobe). Light that is not attenuated by the tissue bed is detected. The oximeter calculates the ratio of the pulsatile and mean light absorbances at each wavelength to create a pulse-added absorbance signal, which is assumed to reflect changes in the arterial blood volume in the tissue. The SpO2 value is derived from the ratio of the pulse-added absorbances at the two wavelengths, which is then compared with a table of arterial oxygen saturations and absorbance ratios derived from volunteers exposed to varying degrees of hypoxia (Table 5). Because the device uses only two wavelengths, it can determine values for only two hemoglobin species: oxyhemoglobin and reduced hemoglobin (Figure 1). The values may be invalid for patients who have hemoglobin species that have different absorbance spectra, since the SpO2 is based on data derived from normal volunteers with low levels of carboxyhemoglobin and methemoglobin.3
Table 5. Terms Used to Denote Oxygen Saturation of Arterial Blood.
Figure 1. Extinction Curves for Various Hemoglobin Species.
Oxyhemoglobin and reduced hemoglobin have different absorptions at 660 and 940 nm, whereas methemoglobin has similar absorption at the two wavelengths. The extinction coefficient is shown on a logarithmic scale.
The extinction coefficient of methemoglobin at 660 nm is similar to that at 940 nm, resulting in a red-to-infrared ratio of 1:1 (Figure 1). The corresponding SpO2 value for this ratio is approximately 85 percent. Hence, as the methemoglobin level increases, the SpO2 will tend toward this value. When methemoglobin levels are in excess of 30 percent, the SpO2 will plateau at 85 percent and will be relatively unaffected by the oxygenation status4,5,6 (Figure 2).
Figure 2. Effect of Methemoglobin on Oxygen Saturation as Measured by Pulse Oximetry.
The ratio of absorbance of methemoglobin at 660 and 940 nm is approximately 1:1, corresponding to a value for oxygen saturation, as measured by pulse oximetry (SpO2), of approximately 85 percent. As the methemoglobin level increases, the SpO2 approaches 85 percent. At a methemoglobin level of 30 percent, the SpO2 plateaus at 85 percent and is unaffected by oxygenation status. Blue circles represent SpO2, and red squares functional saturation (SaO2). Adapted from Barker et al.4
Co-Oximetry
The oxygen saturation value often reported with arterial blood gas values is derived mathematically from the arterial partial pressure of oxygen (PaO2), pH, and temperature, and it reflects the concentration of oxyhemoglobin divided by the sum of oxyhemoglobin and reduced hemoglobin concentrations. This saturation value is sometimes termed the functional saturation of oxygen (SaO2) (Table 5). It does not take into account the presence of carboxyhemoglobin, methemoglobin, or abnormal hemoglobin species. Co-oximeters are multiple-wavelength spectrophotometers that measure the levels of hemoglobin, oxyhemoglobin, carboxyhemoglobin, and methemoglobin by using at least four wavelengths, and they can report values for fractional saturation. Fractional saturation reflects the concentration of oxyhemoglobin as compared with the total concentration of all measured hemoglobin species, rather than just the concentrations of oxyhemoglobin and reduced hemoglobin (Table 5).
In this patient, the fractional saturation value on co-oximetry was 89 percent and the value on pulse oximetry was 94 percent. Surprisingly, measurement of the PaO2 failed to confirm a diagnosis of hypoxemia. What could explain the differences among a fractional saturation of 89 percent determined by co-oximetry, an SaO2 of 98 percent calculated from the arterial blood gas measurement, and the intermediate SpO2 values? The most likely explanation is that an abnormality of hemoglobin, such as methemoglobinemia, was present, and that it altered the absorbance characteristics of the patient's blood.
Methemoglobinemia
Methemoglobin is formed by oxidation of the iron moiety of hemoglobin, which changes from the normal, ferrous state to a ferric state. Ferric heme is incapable of binding oxygen and causes an allosteric change in the remaining heme moieties of the molecule, which impairs the release of oxygen and shifts the oxyhemoglobin-dissociation curve to the left. Cyanosis may be evident at a methemoglobin concentration of 1.5 g per deciliter (about 10 percent of the hemoglobin).3,7 Dark skin pigmentation and poor ambient-light conditions can make the detection of cyanosis difficult.
Normally, an equilibrium exists between the proportions of hemoglobin and methemoglobin (methemoglobin usually constitutes approximately 1 percent of the total hemoglobin). For clinically significant methemoglobinemia to occur, at least one of the following factors must be present: a greatly increased production of methemoglobin; an abnormal hemoglobin that, once oxidized, is resistant to reduction; or decreased activity of erythrocytic NADH–cytochrome-b5 reductase, the primary enzyme responsible for methemoglobin reduction.8,9
Acquired Methemoglobinemia
Exposure to many chemicals and drugs can, directly or indirectly, cause methemoglobinemia, which occasionally can be severe and life-threatening. Rarely, sulfhemoglobinemia can also be induced by exposure to drugs or environmental agents. The clinical picture is similar to that of methemoglobinemia, and pulse-oximetry values can be inaccurate.10
Hemoglobin Variants
Hemoglobin M variants have an amino acid substitution at or near the heme group, such that methemoglobin becomes unusually resistant to reduction. Inheritance typically follows an autosomal dominant pattern. The presence of hemoglobin M may be suspected when there is a history of parent-to-child transmission of long-standing, unexplained cyanosis.9 Since the absorption spectrum of hemoglobin M is abnormal, pulse oximetry is an unreliable assessment tool in such patients. In one reported case of anesthetic management in a patient with hemoglobin MIwate (287Tyr2), a pulse oximeter indicated either no value or a reading of 31 to 36 percent, despite a normal PaO2 value and a normal calculated oxygen saturation.11 Definitive diagnosis of the presence of hemoglobin M can be made by electrophoresis of the methemoglobin and amino acid analysis.12 Unexpectedly low SpO2 readings have been noted in patients with unstable hemoglobin K?ln (2298Met).13,14 The low values were also attributed to the altered absorbance spectrum of the abnormal hemoglobin as compared with the spectrum of normal hemoglobin. Finally, other rare hemoglobin variants that have decreased oxygen affinity, which could lead to cyanosis or have decreased rates of methemoglobin reduction, have been described.9,12
Recessive Congenital Methemoglobinemia
NADH–cytochrome-b5 reductase is present in a membrane-bound isoform that is found in all cells and in a soluble form that is present mainly in erythrocytes. The membrane-bound isoform is involved in the desaturation and elongation of fatty acids, in the biosynthesis of cholesterol, and in drug metabolism. The soluble isoform is responsible for methemoglobin reduction. Both isoforms are encoded by a single gene, and more than 30 mutations that affect enzyme activity have been described.9,15,16 Inheritance of deficiencies of cytochrome-b5 reductase typically follows an autosomal recessive pattern and leads to recessive congenital methemoglobinemia.9 Hereditary deficiencies of cytochrome-b5 reductase have been divided into two types. In type 1, the enzyme deficiency is limited to the soluble form in erythrocytes. A cyanotic appearance usually is the only clinical abnormality. Treatment is usually unnecessary, but patients in whom this deficiency is diagnosed are more susceptible to the effects of methemoglobin-inducing agents.17
Ten to 15 percent of patients with recessive congenital methemoglobinemia have deficiencies of both the soluble and bound isoforms of the enzyme (type 2 of the disorder) and have not only methemoglobinemia but also progressive, severe neurologic abnormalities, for which there is no effective therapy.9,15 Congenital methemoglobinemia can also be caused by a cytochrome-b5 deficiency, but only a single case has been reported.18
Summary and Conclusions
This patient had low SpO2 values on repeated evaluations by pulse oximetry and a decreased fractional oxygen saturation as measured by co-oximetry, despite her healthy appearance and normal PaO2 and calculated (functional) oxygen saturation. She did not appear cyanotic, but she had had a previous episode of unexplained cyanosis. Overall, the clinical picture is consistent with the presence of methemoglobinemia. Oximetry values and the absence of cyanosis suggest a methemoglobin level of approximately 10 to 15 percent. Acquired causes can be reasonably ruled out by her history of cyanosis, persistently low SpO2 values, and absence of exposure to methemoglobin-inducing agents. An abnormal hemoglobin is unlikely given the negative family history as well as the normal hemoglobin pattern found on high-performance liquid chromatography. Therefore, the probable diagnosis is a congenital deficiency of erythrocytic cytochrome-b5 reductase activity (recessive congenital methemoglobinemia, type 1). Appropriate further diagnostic testing would include multiple-wavelength co-oximetry to quantify the degree of methemoglobinemia, and this test could be followed by the measurement of erythrocytic cytochrome-b5 reductase activity if an elevated methemoglobin level was detected.
Discussion of Management
There are several case descriptions of the perioperative care of patients with hereditary methemoglobinemia.8,19,20 Reasonable preoperative recommendations include the maintenance of an increased oxygen concentration in the perioperative period; reliance on arterial blood gas tensions and values obtained by co-oximetry, rather than pulse oximetry,20 to determine the adequacy of oxygenation; avoidance of methemoglobin-inducing drugs9; and treatment with methylene blue if clinically significant methemoglobinemia occurs.
Methylene blue (given intravenously at a dose of 1 to 2 mg per kilogram of body weight) can be used to treat severe cases of methemoglobinemia (in which methemoglobin levels are above 30 percent or there are signs or symptoms of hypoxia). Methylene blue acts as a cofactor that increases the rate of NADPH-dependent methemoglobin reduction. Since the action of methylene blue is dependent on NADPH, the treatment will be ineffective if NADPH levels are low (as they are in glucose-6-phosphate dehydrogenase deficiency) or if NADPH–methemoglobin reductase deficiency is present. High concentrations of methylene blue can act as an oxidant and worsen methemoglobinemia and can cause hemolysis, which may be severe in patients with glucose-6-phosphate dehydrogenase deficiency.17 Ascorbic acid and riboflavin have been used to reduce methemoglobin levels gradually and may be considered for cosmetic treatment of chronic methemoglobinemia caused by a NADH–cytochrome-b5 reductase deficiency.9 Exchange transfusion and hyperbaric oxygen can be considered for refractory life-threatening cases.
In this patient, awareness of her condition, avoidance of drugs known to produce methemoglobin, and the use of methylene blue if necessary should have been sufficient precautions to preclude complications during her operation.
A Physician: Do you think that her initial episode of cyanosis at surgery was caused by her deficiency and a drug that was given?
Dr. Hurford: It is a possibility. She would have been more sensitive to desaturation than someone with normal enzyme function because of the shift in her oxyhemoglobin curve and the increased amount of methemoglobin.
Dr. Nancy Lee Harris (Pathology): Dr. Medoff, you saw this patient in the clinic. Please summarize your thoughts before she had the diagnostic procedure.
Dr. Benjamin D. Medoff (Pulmonary and Critical Care): In my office, the most remarkable thing was how normal she appeared. A normal chest radiograph, normal results on initial pulmonary-function tests, and an unremarkable echocardiogram had ruled out intracardiac or intrapulmonary shunts. I thought that she might have a shunt that was affected by position. I measured her oxygen saturation while she was in various positions, and it never changed. At that point, I considered a hemoglobinopathy or an enzyme deficiency leading to methemoglobinemia as possible diagnoses. I referred her for a co-oximeter test and hemoglobin electrophoresis.
Clinical Diagnosis
Methemoglobinemia due to congenital deficiency of erythrocytic cytochrome-b5 reductase.
Dr. William E. Hurford's Diagnosis
Methemoglobinemia due to congenital deficiency of erythrocytic cytochrome-b5 reductase (recessive congenital methemoglobinemia, type 1).
Pathological Discussion
Dr. Alexander Kratz: The diagnostic procedure was co-oximetry, which showed a methemoglobin concentration of 8.6 percent (reference range, 0.4 to 1.5). This methemoglobin level could account for the variability among the values from finger oximetry, calculated saturation, and co-oximetry saturation. To determine the cause of the methemoglobinemia, a specimen of blood was sent to a reference laboratory for evaluation for methemoglobinemia.
The reference laboratory repeated the spectrophotometric measurement of methemoglobin and found a methemoglobin concentration of 5.5 percent. Since methemoglobin levels can decline by 40 percent per day and the specimen had been approximately one day in transit, this result was consistent with the original measurement at our hospital. The reference laboratory then performed a series of tests to determine the cause of the methemoglobinemia. These included determination of the hemoglobin M ratios, hemoglobin electrophoresis, repeated evaluation by high-performance liquid chromatography, and an assay for the presence of cytochrome-b5 reductase.
Methemoglobinemia due to hemoglobin M does not have the absorbance peak in the range of 630 to 635 nm that is associated with other forms of methemoglobinemia.21 By converting all hemoglobin to methemoglobin by the addition of potassium ferricyanide and assessing absorbances at 500, 600, and 630 nm, it is possible to determine the presence or absence of hemoglobin M.21 In this patient, both the A630:A600 ratio and the A500:A600 ratio were in the normal range. This finding argued strongly against the presence of hemoglobin M. To rule out the presence of hemoglobin M definitively, hemoglobin electrophoresis and repeated high-performance liquid chromatography were performed. Hemoglobin electrophoresis showed a pattern of hemoglobins normal for adults; quantification of the hemoglobins by high-performance liquid chromatography indicated that they were present in normal proportions.
Finally, an assay to determine cytochrome-b5 reductase activity was performed. Cytochrome-b5 reductase has very strong NADH–ferricyanide reductase activity. The reaction can be summarized by this formula: K3Fe(CN)6 + NADH + H+ ——cytochrome-b5 reductase— K3HFe(CN)6 + NAD+. The results of this assay can be followed spectrophotometrically at 340 nm.22,23 This method has been advocated as the easiest way to measure enzyme activity.21 The level of cytochrome-b5 reductase in the patient under discussion was determined to be 6.9 IU per gram of hemoglobin (reference range, 10.1 to 19.4), consistent with reduced activity of the enzyme.
In summary, on the basis of the elevated methemoglobin level, the normal methemoglobin M absorbance ratios, the normal hemoglobin electrophoretic pattern, the results of high-performance liquid chromatography, and the reduced cytochrome-b5 reductase activity, the diagnosis of cytochrome-b5 reductase deficiency was made.
Dr. Harris: If the same gene encodes both the membrane-bound and soluble forms, how can a mutation result in an abnormal soluble form but a normal membrane-bound form?
Dr. Kratz: Some mutations lead to a decrease in the catalytic activity of the protein, which affects both isoforms and causes type 2 disease. Other mutations leave the catalytic activity of the protein intact but cause impaired stability of the enzyme. Since mature erythrocytes (in contrast to other tissues) do not have the ability to synthesize new enzymes, these mutations will be manifested as type 1 disease.9,24
Dr. Harris: Did the patient have the planned operation?
Dr. Medoff: With this determination, the anesthesiologist was consulted and the patient underwent surgery. During surgery, methylene blue was kept on hand, and any agents that might cause methemoglobinemia were avoided. There were no complications during the operation, and the patient recovered uneventfully.
Dr. Harris: Is this the kind of inherited susceptibility you would want patients to know about so they can avoid exposures that might be a problem?
Dr. Medoff: This woman now wears a bracelet that says "methemoglobinemia." Her children have begun to undergo the screening process, and her son has been found to have normal oxygen saturation.
Pathological Diagnosis
Methemoglobinemia due to congenital deficiency of erythrocytic cytochrome-b5 reductase.
Source Information
From the Departments of Anesthesia and Critical Care (W.E.H.) and Pathology (A.K.), Massachusetts General Hospital and Harvard Medical School.
References
American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Practice advisory for preanesthesia evaluation: a report by the American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Anesthesiology 2002;96:485-496.
Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery -- executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Circulation 2002;105:1257-1267.
Alexander CM, Teller LE, Gross JB. Principles of pulse oximetry: theoretical and practical considerations. Anesth Analg 1989;68:368-376.
Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology 1989;70:112-117.
Reynolds KJ, Palayiwa E, Moyle JT, Sykes MK, Hahn CE. The effect of dyshemoglobins on pulse oximetry. I. Theoretical approach. II. Experimental results using an in vitro test system. J Clin Monit 1993;9:81-90.
Watcha MF, Connor MT, Hing AV. Pulse oximetry in methemoglobinemia. Am J Dis Child 1989;143:845-847.
Finch CA. Methemoglobinemia and sulfhemoglobinemia. N Engl J Med 1948;239:470-478.
Chisholm DG, Stuart H. Congenital methaemoglobinaemia detected by preoperative pulse oximetry. Can J Anaesth 1994;41:519-522.
Jaffé ER, Hultquist DE. Cytochrome b5 reductase deficiency and enzymopenic hereditary methemoglobinemia. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease. 8th ed. New York: McGraw-Hill, 2001:4555-70.
Aravindhan N, Chisholm DG. Sulfhemoglobinemia presenting as pulse oximetry desaturation. Anesthesiology 2000;93:883-884.
Kuji A, Satoh Y, Kikuchi K, Satoh K, Joh S. The anesthetic management of a patient with hemoglobin M(Iwate). Anesth Analg 2001;93:1191-1193.
Elghetany MT, Davey FR. Erythrocytic disorders. In: Henry JB, ed. Clinical diagnosis and management by laboratory methods. Philadelphia: W.B. Saunders, 1996:636-63.
Katoh R, Miyake T, Arai T. Unexpectedly low pulse oximeter readings in a boy with unstable hemoglobin K?ln. Anesthesiology 1994;80:472-474.
Gottschalk A, Silverberg M. An unexpected finding with pulse oximetry in a patient with hemoglobin K?ln. Anesthesiology 1994;80:474-476.
Vieira LM, Kaplan JC, Kahn A, Leroux A. Four new mutations in the NADH-cytochrome b5 reductase gene from patients with recessive congenital methemoglobinemia type II. Blood 1995;85:2254-2262.
Percy MJ, Gillespie MJ, Savage G, Hughes AE, McMullin MF, Lappin TR. Familial idiopathic methemoglobinemia revisited: original cases reveal 2 novel mutations in NADH-cytochrome b5 reductase. Blood 2002;100:3447-3449.
Curry S. Methemoglobinemia. Ann Emerg Med 1982;11:214-221.
Hegesh E, Hegesh J, Kaftory A. Congenital methemoglobinemia with a deficiency of cytochrome b5. N Engl J Med 1986;314:757-761.
Sugahara K, Sadohara T, Kawaguchi T, Hirano T. NADH-diaphorase deficiency identified in a patient with congenital methaemoglobinaemia detected by pulse oximetry. Intensive Care Med 1998;24:706-708.
Baraka A, Ayoub CM, Kaddoum RN, Maalouli JM, Chehab IR, Hadi UM. Severe oxyhemoglobin desaturation during induction of anesthesia in a patient with congenital methemoglobinemia. Anesthesiology 2001;95:1296-1297.
Fairbanks VF, Klee GG. Biochemical aspects of hematology. In: Burtis CA, Ashwood ER, eds. Tietz textbook of clinical chemistry. 3rd ed. Philadelphia: W.B. Saunders, 1999:1642-710.
Tanishima K, Takeshita M, Yubisui T, Yoneyama Y. A colorimetric method for the specific determination of methemoglobin reductase activity in red blood cells. Nippon Ketsueki Gakkai Zasshi 1978;41:695-704.
Board PG. NADH-ferricyanide reductase, a convenient approach to the evaluation of NADH-methemoglobin reductase in human erythrocytes. Clin Chim Acta 1981;109:233-237.
Shirabe K, Yubisui T, Borgese N, Tang CY, Hultquist DE, Takeshita M. Enzymatic instability of NADH-cytochrome b5 reductase as a cause of hereditary methemoglobinemia type I (red cell type). J Biol Chem 1992;267:20416-20421.(William E. Hurford, M.D.,)
A 50-year-old woman was evaluated in the pulmonary clinic because of low oxygen saturation.
Five years earlier, dysfunctional uterine bleeding and dysmenorrhea had developed. A pelvic ultrasonographic examination at that time had revealed uterine fibroids. Four years later, the woman was found to be anemic; an endometrial biopsy showed no evidence of cancer. A repeated pelvic ultrasonographic examination revealed a slightly enlarged uterus, 11 cm in length, and a posterior submucous uterine fibroid, 35 by 30 by 41 mm, that displaced the endometrium anteriorly. Her symptoms were controlled for a time with ibuprofen. However, the pain increased, and after she did not have a response to treatment with medroxyprogesterone acetate, she was scheduled for dilatation and curettage, hysteroscopy, and possible resection of the fibroid.
A tonsillectomy had been performed 27 years earlier and a tubal ligation 6 years after that. The patient was noted to have coughing and cyanosis in the recovery room after the tubal ligation. She also had hypertension of recent onset; a history of a heart murmur, with a report of an echocardiogram revealing trace mitral regurgitation; a history of atypical chest pain (with a normal electrocardiogram) that had not recurred; and asthma, diagnosed two years previously, after she had wheezing with a respiratory infection. Her medications included atenolol (25 mg daily), ibuprofen, and an iron supplement. There was a family history of asthma and coronary artery disease, including a myocardial infarction in her 72-year-old mother, but there was no family history of any adverse reactions to general anesthesia. She had no allergies, did not smoke cigarettes, and did not abuse drugs or alcohol.
The patient was seen in the outpatient preadmission testing area. The blood pressure was 162/89 mm Hg, the heart rate 97 beats per minute, the respiratory rate 26 breaths per minute, the temperature 37.2°C, the height 153 cm, and the weight 75 kg. Auscultation of the chest revealed prolonged expiration, with clear lungs. The cardiac rhythm was regular, and there was a systolic ejection murmur (grade 2 to 3) at the base. The remainder of the findings on physical examination were normal. Oxygen saturation by digital pulse oximetry (SpO2) was 93 percent bilaterally while the patient was breathing ambient air. The results of laboratory tests are shown in Table 1. An electrocardiogram showed a normal sinus rhythm with clockwise rotation; there was no change from the electrocardiogram that had been obtained 14 years earlier. Chest radiography disclosed clear lungs; a heart of normal size and contour, with no evidence of congestive heart failure; and no abnormalities of the hilar, mediastinal, pleural, or bony structures.
Table 1. Results of Laboratory Tests.
Pulmonary-function tests performed five days after the outpatient examination showed normal lung volumes, normal single-breath carbon monoxide diffusion capacity, and a reduced SpO2 (91 percent) while the patient was breathing room air. An echocardiogram showed normal left and right ventricular function; there was no evidence of a shunt after an intravenous injection of agitated saline. Slight calcification of the posterior mitral annulus and left atrial dilatation, as well as trace mitral regurgitation (without evidence of mitral-valve prolapse), trace tricuspid insufficiency, and trace pulmonary insufficiency were observed on color and spectral Doppler imaging. No evidence of aortic insufficiency or pericardial effusion was seen. The estimated ejection fraction was 70 percent. The planned operation was postponed, and the patient was referred for a pulmonary evaluation.
Eight weeks later, she was evaluated in the outpatient pulmonary clinic. Additional information on her history was obtained; it was learned that the diagnosis of asthma had been made after she had moved into a dusty office and that her symptoms had resolved after she had moved out of the office. She had not had any other episodes of shortness of breath, and she was physically active. She was originally from Puerto Rico, had two children, and had no other history of occupational exposures to toxins or dust. On physical examination, she appeared well and was not in acute distress. Her lungs and heart were normal on auscultation, her fingers and toes showed no cyanosis or clubbing, and her arms and legs showed no edema. Her SpO2 while she was breathing ambient air in the supine and standing positions was 91 percent, and it remained in the 91 to 93 percent range with exertion. It increased to 97 to 98 percent while the patient was receiving 6 liters of oxygen by nasal cannula. The results of arterial blood gas studies and oximetry are shown in Table 2.
Table 2. The Results of Arterial Blood Gas Studies and Oximetry.
A diagnostic test was performed.
Differential Diagnosis
Dr. William E. Hurford: The key findings in this patient's preoperative evaluation were unexpectedly low SpO2 values on multiple occasions in the absence of known pulmonary or cardiac disease.
Preoperative Evaluation with Pulse Oximetry
The usefulness of pulse oximetry as a preoperative screening tool has never been examined prospectively. In their recent advisories on preoperative evaluation, neither the American Society of Anesthesiologists nor the American College of Cardiology recommends pulse oximetry for routine use.1,2 Nevertheless, the presumption that hypoxemia may contribute to perioperative morbidity and mortality has considerable empiric validity. The patient under discussion had a history of hypertension, a heart murmur, atypical chest pain, and asthma. She had become cyanotic in the recovery room after a previous operation. She also had a family history of asthma and coronary artery disease. Accordingly, the inclusion of pulse oximetry in her cardiopulmonary assessment was reasonable.
Causes of Low Oxygen Saturation on Pulse Oximetry
The low SpO2 value (93 percent) while the patient was breathing ambient air is unexpected, given that she did not have severe signs or symptoms of cardiac or pulmonary disease. What might explain this value? Hypoxemia is the most clinically important cause of a low SpO2, and pulmonary disease is the most common cause of hypoxemia (Table 3). The usual causes of hypoxemia include hypoventilation, mismatching of ventilation and perfusion, right-to-left shunting of blood, and diffusion abnormalities (Table 4). On initial examination, the patient was found to be slightly tachypneic and to have a prolonged expiratory time, but the lung fields were clear. The evaluation proceeded appropriately with chest radiography and tests of pulmonary function. Her chest radiograph was normal, and both spirometric values and the results of testing of the carbon monoxide diffusion capacity of the lung were normal. These tests effectively rule out common diseases such as emphysema or asthma as a cause of hypoxemia.
Table 3. Possible Causes of a Low Oxygen Saturation on Pulse Oximetry.
Table 4. Possible Causes of Hypoxemia.
Cardiac disease can also lead to hypoxemia, particularly when right-to-left shunting of blood or pulmonary edema is present. There was no clinical or radiologic evidence of pulmonary edema in this patient. An echocardiogram was obtained to assess the possibility of ventricular dysfunction or intracardiac shunting, and the images were normal. Right-to-left shunting of blood may also be due to pathologic intrapulmonary shunting, such as that occurring in pulmonary arteriovenous malformations or hepatopulmonary syndrome, but such processes would be unlikely in this patient, given her normal diffusing capacity and echocardiogram. Intrapulmonary shunting would have been detected by early opacification of the left ventricle after the intravenous injection of agitated saline. An explanation of her reduced SpO2 must lie elsewhere.
Limitations of Pulse Oximetry
Pulse oximetry does not directly measure oxygen saturation. Diodes on the oximeter probe emit two wavelengths of light (660 nm, which is red, and 940 nm, which is infrared) that pass through tissue (usually a finger or an earlobe). Light that is not attenuated by the tissue bed is detected. The oximeter calculates the ratio of the pulsatile and mean light absorbances at each wavelength to create a pulse-added absorbance signal, which is assumed to reflect changes in the arterial blood volume in the tissue. The SpO2 value is derived from the ratio of the pulse-added absorbances at the two wavelengths, which is then compared with a table of arterial oxygen saturations and absorbance ratios derived from volunteers exposed to varying degrees of hypoxia (Table 5). Because the device uses only two wavelengths, it can determine values for only two hemoglobin species: oxyhemoglobin and reduced hemoglobin (Figure 1). The values may be invalid for patients who have hemoglobin species that have different absorbance spectra, since the SpO2 is based on data derived from normal volunteers with low levels of carboxyhemoglobin and methemoglobin.3
Table 5. Terms Used to Denote Oxygen Saturation of Arterial Blood.
Figure 1. Extinction Curves for Various Hemoglobin Species.
Oxyhemoglobin and reduced hemoglobin have different absorptions at 660 and 940 nm, whereas methemoglobin has similar absorption at the two wavelengths. The extinction coefficient is shown on a logarithmic scale.
The extinction coefficient of methemoglobin at 660 nm is similar to that at 940 nm, resulting in a red-to-infrared ratio of 1:1 (Figure 1). The corresponding SpO2 value for this ratio is approximately 85 percent. Hence, as the methemoglobin level increases, the SpO2 will tend toward this value. When methemoglobin levels are in excess of 30 percent, the SpO2 will plateau at 85 percent and will be relatively unaffected by the oxygenation status4,5,6 (Figure 2).
Figure 2. Effect of Methemoglobin on Oxygen Saturation as Measured by Pulse Oximetry.
The ratio of absorbance of methemoglobin at 660 and 940 nm is approximately 1:1, corresponding to a value for oxygen saturation, as measured by pulse oximetry (SpO2), of approximately 85 percent. As the methemoglobin level increases, the SpO2 approaches 85 percent. At a methemoglobin level of 30 percent, the SpO2 plateaus at 85 percent and is unaffected by oxygenation status. Blue circles represent SpO2, and red squares functional saturation (SaO2). Adapted from Barker et al.4
Co-Oximetry
The oxygen saturation value often reported with arterial blood gas values is derived mathematically from the arterial partial pressure of oxygen (PaO2), pH, and temperature, and it reflects the concentration of oxyhemoglobin divided by the sum of oxyhemoglobin and reduced hemoglobin concentrations. This saturation value is sometimes termed the functional saturation of oxygen (SaO2) (Table 5). It does not take into account the presence of carboxyhemoglobin, methemoglobin, or abnormal hemoglobin species. Co-oximeters are multiple-wavelength spectrophotometers that measure the levels of hemoglobin, oxyhemoglobin, carboxyhemoglobin, and methemoglobin by using at least four wavelengths, and they can report values for fractional saturation. Fractional saturation reflects the concentration of oxyhemoglobin as compared with the total concentration of all measured hemoglobin species, rather than just the concentrations of oxyhemoglobin and reduced hemoglobin (Table 5).
In this patient, the fractional saturation value on co-oximetry was 89 percent and the value on pulse oximetry was 94 percent. Surprisingly, measurement of the PaO2 failed to confirm a diagnosis of hypoxemia. What could explain the differences among a fractional saturation of 89 percent determined by co-oximetry, an SaO2 of 98 percent calculated from the arterial blood gas measurement, and the intermediate SpO2 values? The most likely explanation is that an abnormality of hemoglobin, such as methemoglobinemia, was present, and that it altered the absorbance characteristics of the patient's blood.
Methemoglobinemia
Methemoglobin is formed by oxidation of the iron moiety of hemoglobin, which changes from the normal, ferrous state to a ferric state. Ferric heme is incapable of binding oxygen and causes an allosteric change in the remaining heme moieties of the molecule, which impairs the release of oxygen and shifts the oxyhemoglobin-dissociation curve to the left. Cyanosis may be evident at a methemoglobin concentration of 1.5 g per deciliter (about 10 percent of the hemoglobin).3,7 Dark skin pigmentation and poor ambient-light conditions can make the detection of cyanosis difficult.
Normally, an equilibrium exists between the proportions of hemoglobin and methemoglobin (methemoglobin usually constitutes approximately 1 percent of the total hemoglobin). For clinically significant methemoglobinemia to occur, at least one of the following factors must be present: a greatly increased production of methemoglobin; an abnormal hemoglobin that, once oxidized, is resistant to reduction; or decreased activity of erythrocytic NADH–cytochrome-b5 reductase, the primary enzyme responsible for methemoglobin reduction.8,9
Acquired Methemoglobinemia
Exposure to many chemicals and drugs can, directly or indirectly, cause methemoglobinemia, which occasionally can be severe and life-threatening. Rarely, sulfhemoglobinemia can also be induced by exposure to drugs or environmental agents. The clinical picture is similar to that of methemoglobinemia, and pulse-oximetry values can be inaccurate.10
Hemoglobin Variants
Hemoglobin M variants have an amino acid substitution at or near the heme group, such that methemoglobin becomes unusually resistant to reduction. Inheritance typically follows an autosomal dominant pattern. The presence of hemoglobin M may be suspected when there is a history of parent-to-child transmission of long-standing, unexplained cyanosis.9 Since the absorption spectrum of hemoglobin M is abnormal, pulse oximetry is an unreliable assessment tool in such patients. In one reported case of anesthetic management in a patient with hemoglobin MIwate (287Tyr2), a pulse oximeter indicated either no value or a reading of 31 to 36 percent, despite a normal PaO2 value and a normal calculated oxygen saturation.11 Definitive diagnosis of the presence of hemoglobin M can be made by electrophoresis of the methemoglobin and amino acid analysis.12 Unexpectedly low SpO2 readings have been noted in patients with unstable hemoglobin K?ln (2298Met).13,14 The low values were also attributed to the altered absorbance spectrum of the abnormal hemoglobin as compared with the spectrum of normal hemoglobin. Finally, other rare hemoglobin variants that have decreased oxygen affinity, which could lead to cyanosis or have decreased rates of methemoglobin reduction, have been described.9,12
Recessive Congenital Methemoglobinemia
NADH–cytochrome-b5 reductase is present in a membrane-bound isoform that is found in all cells and in a soluble form that is present mainly in erythrocytes. The membrane-bound isoform is involved in the desaturation and elongation of fatty acids, in the biosynthesis of cholesterol, and in drug metabolism. The soluble isoform is responsible for methemoglobin reduction. Both isoforms are encoded by a single gene, and more than 30 mutations that affect enzyme activity have been described.9,15,16 Inheritance of deficiencies of cytochrome-b5 reductase typically follows an autosomal recessive pattern and leads to recessive congenital methemoglobinemia.9 Hereditary deficiencies of cytochrome-b5 reductase have been divided into two types. In type 1, the enzyme deficiency is limited to the soluble form in erythrocytes. A cyanotic appearance usually is the only clinical abnormality. Treatment is usually unnecessary, but patients in whom this deficiency is diagnosed are more susceptible to the effects of methemoglobin-inducing agents.17
Ten to 15 percent of patients with recessive congenital methemoglobinemia have deficiencies of both the soluble and bound isoforms of the enzyme (type 2 of the disorder) and have not only methemoglobinemia but also progressive, severe neurologic abnormalities, for which there is no effective therapy.9,15 Congenital methemoglobinemia can also be caused by a cytochrome-b5 deficiency, but only a single case has been reported.18
Summary and Conclusions
This patient had low SpO2 values on repeated evaluations by pulse oximetry and a decreased fractional oxygen saturation as measured by co-oximetry, despite her healthy appearance and normal PaO2 and calculated (functional) oxygen saturation. She did not appear cyanotic, but she had had a previous episode of unexplained cyanosis. Overall, the clinical picture is consistent with the presence of methemoglobinemia. Oximetry values and the absence of cyanosis suggest a methemoglobin level of approximately 10 to 15 percent. Acquired causes can be reasonably ruled out by her history of cyanosis, persistently low SpO2 values, and absence of exposure to methemoglobin-inducing agents. An abnormal hemoglobin is unlikely given the negative family history as well as the normal hemoglobin pattern found on high-performance liquid chromatography. Therefore, the probable diagnosis is a congenital deficiency of erythrocytic cytochrome-b5 reductase activity (recessive congenital methemoglobinemia, type 1). Appropriate further diagnostic testing would include multiple-wavelength co-oximetry to quantify the degree of methemoglobinemia, and this test could be followed by the measurement of erythrocytic cytochrome-b5 reductase activity if an elevated methemoglobin level was detected.
Discussion of Management
There are several case descriptions of the perioperative care of patients with hereditary methemoglobinemia.8,19,20 Reasonable preoperative recommendations include the maintenance of an increased oxygen concentration in the perioperative period; reliance on arterial blood gas tensions and values obtained by co-oximetry, rather than pulse oximetry,20 to determine the adequacy of oxygenation; avoidance of methemoglobin-inducing drugs9; and treatment with methylene blue if clinically significant methemoglobinemia occurs.
Methylene blue (given intravenously at a dose of 1 to 2 mg per kilogram of body weight) can be used to treat severe cases of methemoglobinemia (in which methemoglobin levels are above 30 percent or there are signs or symptoms of hypoxia). Methylene blue acts as a cofactor that increases the rate of NADPH-dependent methemoglobin reduction. Since the action of methylene blue is dependent on NADPH, the treatment will be ineffective if NADPH levels are low (as they are in glucose-6-phosphate dehydrogenase deficiency) or if NADPH–methemoglobin reductase deficiency is present. High concentrations of methylene blue can act as an oxidant and worsen methemoglobinemia and can cause hemolysis, which may be severe in patients with glucose-6-phosphate dehydrogenase deficiency.17 Ascorbic acid and riboflavin have been used to reduce methemoglobin levels gradually and may be considered for cosmetic treatment of chronic methemoglobinemia caused by a NADH–cytochrome-b5 reductase deficiency.9 Exchange transfusion and hyperbaric oxygen can be considered for refractory life-threatening cases.
In this patient, awareness of her condition, avoidance of drugs known to produce methemoglobin, and the use of methylene blue if necessary should have been sufficient precautions to preclude complications during her operation.
A Physician: Do you think that her initial episode of cyanosis at surgery was caused by her deficiency and a drug that was given?
Dr. Hurford: It is a possibility. She would have been more sensitive to desaturation than someone with normal enzyme function because of the shift in her oxyhemoglobin curve and the increased amount of methemoglobin.
Dr. Nancy Lee Harris (Pathology): Dr. Medoff, you saw this patient in the clinic. Please summarize your thoughts before she had the diagnostic procedure.
Dr. Benjamin D. Medoff (Pulmonary and Critical Care): In my office, the most remarkable thing was how normal she appeared. A normal chest radiograph, normal results on initial pulmonary-function tests, and an unremarkable echocardiogram had ruled out intracardiac or intrapulmonary shunts. I thought that she might have a shunt that was affected by position. I measured her oxygen saturation while she was in various positions, and it never changed. At that point, I considered a hemoglobinopathy or an enzyme deficiency leading to methemoglobinemia as possible diagnoses. I referred her for a co-oximeter test and hemoglobin electrophoresis.
Clinical Diagnosis
Methemoglobinemia due to congenital deficiency of erythrocytic cytochrome-b5 reductase.
Dr. William E. Hurford's Diagnosis
Methemoglobinemia due to congenital deficiency of erythrocytic cytochrome-b5 reductase (recessive congenital methemoglobinemia, type 1).
Pathological Discussion
Dr. Alexander Kratz: The diagnostic procedure was co-oximetry, which showed a methemoglobin concentration of 8.6 percent (reference range, 0.4 to 1.5). This methemoglobin level could account for the variability among the values from finger oximetry, calculated saturation, and co-oximetry saturation. To determine the cause of the methemoglobinemia, a specimen of blood was sent to a reference laboratory for evaluation for methemoglobinemia.
The reference laboratory repeated the spectrophotometric measurement of methemoglobin and found a methemoglobin concentration of 5.5 percent. Since methemoglobin levels can decline by 40 percent per day and the specimen had been approximately one day in transit, this result was consistent with the original measurement at our hospital. The reference laboratory then performed a series of tests to determine the cause of the methemoglobinemia. These included determination of the hemoglobin M ratios, hemoglobin electrophoresis, repeated evaluation by high-performance liquid chromatography, and an assay for the presence of cytochrome-b5 reductase.
Methemoglobinemia due to hemoglobin M does not have the absorbance peak in the range of 630 to 635 nm that is associated with other forms of methemoglobinemia.21 By converting all hemoglobin to methemoglobin by the addition of potassium ferricyanide and assessing absorbances at 500, 600, and 630 nm, it is possible to determine the presence or absence of hemoglobin M.21 In this patient, both the A630:A600 ratio and the A500:A600 ratio were in the normal range. This finding argued strongly against the presence of hemoglobin M. To rule out the presence of hemoglobin M definitively, hemoglobin electrophoresis and repeated high-performance liquid chromatography were performed. Hemoglobin electrophoresis showed a pattern of hemoglobins normal for adults; quantification of the hemoglobins by high-performance liquid chromatography indicated that they were present in normal proportions.
Finally, an assay to determine cytochrome-b5 reductase activity was performed. Cytochrome-b5 reductase has very strong NADH–ferricyanide reductase activity. The reaction can be summarized by this formula: K3Fe(CN)6 + NADH + H+ ——cytochrome-b5 reductase— K3HFe(CN)6 + NAD+. The results of this assay can be followed spectrophotometrically at 340 nm.22,23 This method has been advocated as the easiest way to measure enzyme activity.21 The level of cytochrome-b5 reductase in the patient under discussion was determined to be 6.9 IU per gram of hemoglobin (reference range, 10.1 to 19.4), consistent with reduced activity of the enzyme.
In summary, on the basis of the elevated methemoglobin level, the normal methemoglobin M absorbance ratios, the normal hemoglobin electrophoretic pattern, the results of high-performance liquid chromatography, and the reduced cytochrome-b5 reductase activity, the diagnosis of cytochrome-b5 reductase deficiency was made.
Dr. Harris: If the same gene encodes both the membrane-bound and soluble forms, how can a mutation result in an abnormal soluble form but a normal membrane-bound form?
Dr. Kratz: Some mutations lead to a decrease in the catalytic activity of the protein, which affects both isoforms and causes type 2 disease. Other mutations leave the catalytic activity of the protein intact but cause impaired stability of the enzyme. Since mature erythrocytes (in contrast to other tissues) do not have the ability to synthesize new enzymes, these mutations will be manifested as type 1 disease.9,24
Dr. Harris: Did the patient have the planned operation?
Dr. Medoff: With this determination, the anesthesiologist was consulted and the patient underwent surgery. During surgery, methylene blue was kept on hand, and any agents that might cause methemoglobinemia were avoided. There were no complications during the operation, and the patient recovered uneventfully.
Dr. Harris: Is this the kind of inherited susceptibility you would want patients to know about so they can avoid exposures that might be a problem?
Dr. Medoff: This woman now wears a bracelet that says "methemoglobinemia." Her children have begun to undergo the screening process, and her son has been found to have normal oxygen saturation.
Pathological Diagnosis
Methemoglobinemia due to congenital deficiency of erythrocytic cytochrome-b5 reductase.
Source Information
From the Departments of Anesthesia and Critical Care (W.E.H.) and Pathology (A.K.), Massachusetts General Hospital and Harvard Medical School.
References
American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Practice advisory for preanesthesia evaluation: a report by the American Society of Anesthesiologists Task Force on Preanesthesia Evaluation. Anesthesiology 2002;96:485-496.
Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery -- executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Circulation 2002;105:1257-1267.
Alexander CM, Teller LE, Gross JB. Principles of pulse oximetry: theoretical and practical considerations. Anesth Analg 1989;68:368-376.
Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology 1989;70:112-117.
Reynolds KJ, Palayiwa E, Moyle JT, Sykes MK, Hahn CE. The effect of dyshemoglobins on pulse oximetry. I. Theoretical approach. II. Experimental results using an in vitro test system. J Clin Monit 1993;9:81-90.
Watcha MF, Connor MT, Hing AV. Pulse oximetry in methemoglobinemia. Am J Dis Child 1989;143:845-847.
Finch CA. Methemoglobinemia and sulfhemoglobinemia. N Engl J Med 1948;239:470-478.
Chisholm DG, Stuart H. Congenital methaemoglobinaemia detected by preoperative pulse oximetry. Can J Anaesth 1994;41:519-522.
Jaffé ER, Hultquist DE. Cytochrome b5 reductase deficiency and enzymopenic hereditary methemoglobinemia. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease. 8th ed. New York: McGraw-Hill, 2001:4555-70.
Aravindhan N, Chisholm DG. Sulfhemoglobinemia presenting as pulse oximetry desaturation. Anesthesiology 2000;93:883-884.
Kuji A, Satoh Y, Kikuchi K, Satoh K, Joh S. The anesthetic management of a patient with hemoglobin M(Iwate). Anesth Analg 2001;93:1191-1193.
Elghetany MT, Davey FR. Erythrocytic disorders. In: Henry JB, ed. Clinical diagnosis and management by laboratory methods. Philadelphia: W.B. Saunders, 1996:636-63.
Katoh R, Miyake T, Arai T. Unexpectedly low pulse oximeter readings in a boy with unstable hemoglobin K?ln. Anesthesiology 1994;80:472-474.
Gottschalk A, Silverberg M. An unexpected finding with pulse oximetry in a patient with hemoglobin K?ln. Anesthesiology 1994;80:474-476.
Vieira LM, Kaplan JC, Kahn A, Leroux A. Four new mutations in the NADH-cytochrome b5 reductase gene from patients with recessive congenital methemoglobinemia type II. Blood 1995;85:2254-2262.
Percy MJ, Gillespie MJ, Savage G, Hughes AE, McMullin MF, Lappin TR. Familial idiopathic methemoglobinemia revisited: original cases reveal 2 novel mutations in NADH-cytochrome b5 reductase. Blood 2002;100:3447-3449.
Curry S. Methemoglobinemia. Ann Emerg Med 1982;11:214-221.
Hegesh E, Hegesh J, Kaftory A. Congenital methemoglobinemia with a deficiency of cytochrome b5. N Engl J Med 1986;314:757-761.
Sugahara K, Sadohara T, Kawaguchi T, Hirano T. NADH-diaphorase deficiency identified in a patient with congenital methaemoglobinaemia detected by pulse oximetry. Intensive Care Med 1998;24:706-708.
Baraka A, Ayoub CM, Kaddoum RN, Maalouli JM, Chehab IR, Hadi UM. Severe oxyhemoglobin desaturation during induction of anesthesia in a patient with congenital methemoglobinemia. Anesthesiology 2001;95:1296-1297.
Fairbanks VF, Klee GG. Biochemical aspects of hematology. In: Burtis CA, Ashwood ER, eds. Tietz textbook of clinical chemistry. 3rd ed. Philadelphia: W.B. Saunders, 1999:1642-710.
Tanishima K, Takeshita M, Yubisui T, Yoneyama Y. A colorimetric method for the specific determination of methemoglobin reductase activity in red blood cells. Nippon Ketsueki Gakkai Zasshi 1978;41:695-704.
Board PG. NADH-ferricyanide reductase, a convenient approach to the evaluation of NADH-methemoglobin reductase in human erythrocytes. Clin Chim Acta 1981;109:233-237.
Shirabe K, Yubisui T, Borgese N, Tang CY, Hultquist DE, Takeshita M. Enzymatic instability of NADH-cytochrome b5 reductase as a cause of hereditary methemoglobinemia type I (red cell type). J Biol Chem 1992;267:20416-20421.(William E. Hurford, M.D.,)