Errata
Erratum to: Unexpectedly low pulse oximetry measurements
associated with variant hemoglobins: A systematic reviewy
Madeleine Verhovsek,1,2* Matthew P.A. Henderson,3 Gerard Cox,2 Hong-yuan Luo,1
Martin H. Steinberg,1 and David H.K. Chui1,4
Pulse oximetry estimates arterial blood oxygen saturation based on light absorbance of oxy- and deoxy-hemoglobin at 660 and 940 nm wavelengths. Patients with unexpectedly low SpO2 often undergo cardio-pulmonary testing to ascertain the cause of their hypoxemia. However, in a subset of patients, a variant hemoglobin
is responsible for low SpO2 measurements. The extent of this problem is unclear. We performed a systematic
literature review for reports of low SpO2 associated with variant hemoglobins. We also reviewed unpublished
cases from an academic hemoglobin diagnostic reference laboratory. Twenty-five publications and four
unpublished cases were identified, representing 45 patients with low SpO2 and confirmed variant hemoglobin.
Fifty-seven family members of patients had confirmed or suspected variant hemoglobin. Three low oxygen affinity variant hemoglobins had concordantly low SpO2 and SaO2. Eleven variant hemoglobins were associated
with unexpectedly low SpO2 measurements but normal SaO2. Hemoglobin light absorbance testing was
reported in three cases, all of which showed abnormal absorption spectra between 600 and 900 nm. Seven
other variant hemoglobins had decreased SpO2, with unreported or uncertain SaO2. Twenty-one variant hemoglobins were found to be associated with low SpO2. Most variant hemoglobins were associated with spuriously low SpO2. Abnormal absorption spectra explain the discrepancy between SpO2 and SaO2 for some variants. The differential diagnosis of possible variant hemoglobin ought to be considered in asymptomatic
patients found to have unexpectedly low SpO2. The correct diagnosis will help to spare patients from
C 2011 Wiley-Liss, Inc.
unnecessary investigations and anxiety. Am. J. Hematol. 86:722–725, 2011. V
Introduction
The pulse oximeter is a simple, noninvasive instrument
used to estimate a patient’s arterial blood oxygen saturation, sometimes known as ‘‘the fifth vital sign’’ [1]. Pulse
oximetry is used extensively in peri-operative monitoring,
the emergency room, in-patient wards, selected ambulatory
care settings, and during labor and delivery. Sometimes,
patients are found unexpectedly to have low oxygen saturation by pulse oximetry (SpO2). These patients may undergo
extensive cardio-pulmonary investigations in search of the
cause of their ‘‘hypoxemia.’’ However, in some of these individuals, arterial blood gas measurements (SaO2) are normal, and low SpO2 readings are spurious due to the presence of a variant hemoglobin, rather than cardiac or pulmonary diseases [2,3].
Laboratory testing for a variant hemoglobin is usually
undertaken due to anemia, polycythemia, cyanosis, abnormal erythrocyte indices like low mean cell volume (MCV) or
mean cell hemoglobin (MCH), a suspicion of illness such
as sickle cell disease or thalassemia, or a positive family
history of a variant hemoglobin [4]. More than 1,000 variant
hemoglobins have been identified [3], the majority of which
are not associated with abnormal SpO2 readings. However,
in some patients, low SpO2 is the finding that triggers further evaluation leading to identification of an underlying variant hemoglobin [5]. Patients who have inherited low oxygen affinity variant hemoglobin can have decreased arterial
blood oxygen saturation, low SpO2, and low SaO2. In a
subset of patients, a variant hemoglobin is responsible for
artifactually low SpO2 measurements with no pathophysiologic significance. They have normal arterial blood oxygen
saturation and normal SaO2 readings.
We have (1) performed a systematic review of the literature for reports of low SpO2 readings associated with variant hemoglobins; (2) categorized these variants according
to the potential mechanism for low SpO2; and (3) devised
an algorithm to guide the evaluation of a patient with unexplained low SpO2, with particular attention to hemoglobin-
opathy testing so as to spare the patient unnecessary
investigations and anxiety.
Methods
Literature search. We performed a systematic literature search
which included the following databases: Ovid MEDLINE (1950 to May
Week 2 2010), Ovid MEDLINE in-process and other nonindexed citations (May 20, 2010) and EMBASE (1980 to 2010 Week 19). The keyword search included the terms ‘‘variant hemoglobin,’’ ‘‘hemoglobinopathy,’’ ‘‘methemoglobin,’’ ‘‘methemoglobinemia,’’ ‘‘hemoglobin M,’’ ‘‘Hb
M,’’ ‘‘pulse oximetry,’’ and ‘‘oximetry.’’ We supplemented the formal literature review with a review of reference lists and contact with content
experts. We also reviewed unpublished cases from the Hemoglobin
Diagnostic Reference Laboratory (HDRL) at the Boston Medical Center.
Selection of reports. We included all case reports published in English
on patients with SpO2 of less than 95% who were found to have a variant
hemoglobin. Reports of patients with severe respiratory disorders or other illnesses causing hypoxemia during evaluation were excluded. Cases of
acquired or hereditary methemoglobinemia due to cytochrome b5 reductase
deficiency, or without further definitive investigations, were also excluded.
Additional Supporting Information may be found in the online version of this
article.
1
Section of Hematology and Oncology, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts 02118; 2Department of
Medicine, McMaster University Faculty of Health Sciences, Hamilton, Ontario, Canada L8S 4L8; 3Department of Pathology and Molecular Medicine,
McMaster University Faculty of Health Sciences, Hamilton, Ontario, Canada
L8S 4L8; 4Department of Pathology and Laboratory Medicine, Boston University School of Medicine, Boston, Massachusetts 02118
Conflict of interest: Nothing to report.
Contract grant sponsor: NIH, NIDDK; Contract grant number: RO1 DK069646.
*Correspondence to: Dr. Madeleine Verhovsek, McMaster University, Faculty
of Health Sciences, Department of Medicine, 1200 Main Street West HSC
3W10, Hamilton, Ontario, Canada L8N 3Z5. E-mail: verhovm@mcmaster.ca
y
There were numerous errors in the citing of references in text published in the
Journal 85(11):882–885, Nov. 2010 DOI 21810. Therefore the corrected article
has been reprinted in its entirety.
Received for publication 1 June 2010; Revised 22 June 2010; Accepted 23
June 2010
Am. J. Hematol. 86:722–725, 2011.
Published online in Wiley Online Library (wileyonlinelibrary.com).
DOI: 10.1002/ajh.22074
C 2011 Wiley-Liss, Inc.
V
American Journal of Hematology
722
http://wileyonlinelibrary.com/cgi-bin/jhome/35105
We collected data including: SpO2, arterial blood gas (SaO2 and
PaO2), details about the variant hemoglobin (name, globin gene mutation, oxygen affinity characteristics, stability), patient characteristics
(age, gender; and clinical data), and information about other affected
family members. Other laboratory results collected when available
were: hemoglobin level; variant hemoglobin as a percent of total hemoglobin; P50 testing; absorption spectrum; stability testing; and methemoglobin level by co-oximetry.
When multiple SpO2 and/or arterial blood gases (ABG) were
reported, we selected the set of data that was most complete (i.e.,
SpO2 and ABG done concurrently; co-oximetry reported, including
SaO2). If multiple complete data sets were reported, we selected the
set closest to the patient’s initial SpO2 evaluation.
For cases where both SpO2 and SaO2 were reported, we categorized
each as ‘‘concordant’’ (low SpO2 and low SaO2) or ‘‘discordant’’ (low SpO2
and SaO2 that was greater than or equal to 5% higher than SpO2).
Results
Cases of low SpO2 and variant hemoglobins
We identified 396 potential publications: 275 from
EMBASE and 121 from Ovid MEDLINE. We excluded 357
articles after screening titles and abstracts by using our
predefined inclusion and exclusion criteria. Supplemental
review of references, and consultation with content experts
yielded an additional nine publications for review [6–14].
Forty-eight publications were retrieved for detailed evaluation, of which 23 were excluded for the following reasons:
not a case report (n 5 5); SpO2 not reported (n 5 5); normal O2 saturation by SpO2 (n 5 1); severe respiratory illness during evaluation (n 5 1); and methemoglobinemia
without documented congenital Hb M variant (n 5 11).
In summary, we included 25 publications, with data for a
total of 41 patients with low SpO2 and confirmed variant hemoglobin. Four additional unpublished cases from HDRL
met inclusion criteria. Patients ranged in age from the neonate to age 63 years and were distributed approximately
equally between male and female. In 17 patients, SaO2
measurements were done and reported.
Thirteen relatives of index cases were reported to have
low SpO2 with a confirmed variant hemoglobin. An additional 17 relatives had a confirmed variant hemoglobin but
their SpO2 data were not provided. Twenty-seven other relatives had unexplained low SpO2 or other clinical reason to
suspect inheritance of the variant hemoglobin, but testing
had not been done.
Cases with concordantly low SpO2 and SaO2
There were five publications of patients who were carriers of either Hb Bassett [10,15], Hb Rothschild [2,5], or
Hb Canebiere [16] (Supporting Information—Table I). These
are low oxygen affinity variant hemoglobins [15–17]. There
was one report on each of these three variant hemoglobins
showing low SpO2 and similarly low SaO2. In three of these
cases [5,15,16], the PaO2 was normal, indicating that there
was no apparent tissue hypoxemia.
Hb Bassett accounted for 16% of the total hemoglobin in
one report, consistent with this being an a1-globin gene mutation [10]. In this instance, the P50 of the whole blood was normal (28 mm Hg), likely due to the fact that only a small proportion of the total hemoglobin was Hb Bassett. In another report,
Hb Bassett accounted for 30% of the total hemoglobin [15],
suggesting that the mutation resided on an a2-globin gene
[18]. In this study, P50 measurement on isolated Hb Bassett
was high (22 mm Hg; normal p50 for Hb A 10.5 mm Hg), consistent with the variant hemoglobin having low oxygen affinity.
Cases with low SpO2 and discordant SaO2
Six variant hemoglobins due to a-globin gene missense
mutations (Hbs Lansing [45], Titusville [32], Bonn [21], Delaware, and M-Iwate [20], and a novel hemoglobin [22]),
and five variant hemoglobins due to b-globin gene missense mutations (Hbs Hammersmith [23], Cheverly [29],
American Journal of Hematology
Figure 1. Spectrophotometric analysis of oxyhemoglobin from normal individual
and Hb Bonn. Note that oxyhemoglobin Bonn (HbO2 Bonn) has higher absorbance
at 660 nm wavelength, which is the sole wavelength the pulse oximeter utilizes to
measure deoxyhemoglobin, thus resulting in erroneously low SpO2 for patients
who have Hb Bonn. Figure from Zur et al. [21].
Okazaki, Regina, and Köln [26,30]) are tabulated in Table II
(Supporting Information). Carriers of these variant hemoglobins presented with unexpectedly low SpO2 but normal SaO2.
Several reports of cases with Hbs Cheverly [28], Hammersmith [24], Köln [35], and Titusville [7] did not include SaO2
values. These cases are also included in this group.
There is no clear trend in oxygen affinity among these
variants. Hb Titusville has low oxygen affinity [19,41]. Hb
Okazaki [25,33] and Hb Regina [34] have high oxygen affinity). The other variant hemoglobins in this group either
have normal affinity or their oxygen affinity has not been
reported. Many of the variants are unstable, or are presumed to be unstable based on unexpectedly low levels on
hemoglobin quantitation. Hb Köln (b Val98Met) is the most
common unstable variant hemoglobin known, and found in
many unrelated individuals. Hb M-Iwate (a His87Tyr) is a
congenital form of methemoglobin.
Absorption spectra of variant hemoglobins
Absorption spectra of variant hemoglobins were detailed
in three reports—Hb Bonn [21], Hb Cheverly [28], and Hb
Koln [26]—all hemoglobins with discordant SpO2 and
SaO2. Hb Bonn demonstrated an additional absorption
maximum of oxyhemoglobin at 668 nm, but no changes in
the absorption curve of deoxyhemoglobin (see Fig. 1). Hb
Cheverly showed increased absorbance between 600 and
660 nm compared with normal control hemoglobin. For Hb
Köln, all the absorbance curves for oxyhemoglobin, deoxyhemoglobin, and carboxyhemoglobin were shifted upward
in the range of 600 to 900 nm.
Cases with undefined relationship between
SpO2 and SaO2
There are four other variant hemoglobins (Hbs Sunshine
Seth [9], Louisville [14], Chico [27], and Nishinomiya [46])
and three methemoglobins (Hbs M-Milwaukee [47], F-Circleville [11] and FM-Fort Ripley [8]) all of which have
decreased SpO2, with unreported or uncertain SaO2 (Supporting Information—Table III).
Discussion
Respiratory disease associated with impaired gas
exchange is the most common cause of low SpO2 measurements. In much less common instances, individuals with
variant hemoglobins can have spuriously low SpO2 read-
723
10968652, 2011, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajh.22074, Wiley Online Library on [23/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Errata
Figure 2. Algorithm for evaluation of low SpO2. MetHb 5 methemoglobin. # Arterial blood gas should be done on room air and with simultaneous SpO2 measurement. * Investigations could include hemoglobin analysis by various methods,
and if necessary, DNA-based genotyping. [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
Figure 3. Hemoglobin extinction curve of normal adult oxyhemoglobin and normal adult deoxyhemoglobin. Figure from Sinex [36].
ings caused by the different absorption spectrum of their
abnormal hemoglobin. The differential diagnosis of a possible variant hemoglobin should be considered while investigating a patient who is found to have an unexpectedly low
SpO2 but is otherwise clinically asymptomatic. By making
the correct diagnosis promptly, the patient might be spared
unnecessary investigations and anxiety. A normal SaO2
measurement based on arterial blood gas studies with cooximetry will document the lack of hypoxemia, and suggest
that testing for variant hemoglobin is indicated. A summary
of our recommended algorithm is presented in Fig. 2.
Through our systematic review of the literature and unreported cases from our Laboratory, we have summarized
the reports of 45 patients with 21 different variant hemoglobins that are associated with low SpO2. When family members with documented or suspected variant hemoglobin are
considered, the number is likely to be over 100. The majority of the patients studied had no symptoms—low SpO2
was noted incidentally during routine vital signs monitoring.
In many of these cases, in vitro co-oximetry on arterial
blood revealed a normal hemoglobin oxygen saturation,
confirming the spurious result of pulse oximetry. Nonetheless, investigations for hemoglobinopathy were frequently
undertaken only after extensive cardio-pulmonary evaluation failed to reveal a cause for presumed hypoxemia.
Pulse oximetry uses two wavelengths of light (660 and
940 nm) emitted across a vascular bed to calculate the relative proportions of oxy- and deoxyhemoglobin, based on
their different absorption spectra. Normal deoxyhemoglobin
has relatively higher absorbance at 660 nm, whereas normal oxyhemoglobin has relatively higher absorbance at 940
nm (see Fig. 3). Many factors can interfere with the accuracy of pulse oximetry measurements such as motion artifact, poor perfusion, venous pulsation, nail polish, and signal artifact from ambient light [36]. Because pulse oximetry
calculates hemoglobin oxygen saturation values based on
light absorption by oxy- and deoxyhemoglobin at only two
wavelengths, accuracy is compromised in the presence of
variant hemoglobins that have abnormal absorption spectra. Moreover, carboxyhemoglobin has similar absorbance
to oxyhemoglobin at 660 nm. Therefore, when carboxyhemoglobin levels are elevated, pulse oximetry gives falsely
elevated arterial oxygen saturation. Similarly, methemoglobin causes a large absorbance at both 660 and 940 nm,
making SpO2 unreliable in a patient with methemoglobinemia [37]. Thus, individuals with congenital Hb M variants,
including those with Hbs M-Iwate, M-Milwaukee, and FMFort Ripley included in our review, are expected to have
discordantly low SpO2.
The early co-oximeters, such as the IL 282, used four
wavelengths (535.0, 585.2, 594.5, and 626.6 nm) to determine the fractional oxygen saturation [38]. Current co-oximeters measure absorbance at over 100 wavelengths to
construct a continuous absorption spectrum from 450 to
700 nm [39]. By using multiple wavelengths of light, in vitro
analysis of arterial blood with co-oximetry measures levels of
oxyhemoglobin, deoxyhemoglobin, carboxyhemoglobin, and
methemoglobin, thereby improving the accuracy of measurements made in the presence of these variant hemoglobins.
SaO2 is sometimes estimated from a measured pO2 and an
empirical equation for the oxyhemoglobin dissociation curve.
These calculations normally include correction for pH, temperature, and pCO2; however, other factors such as low O2
affinity variant hemoglobins, other hemoglobins with abnormal absorption spectra and altered intracellular erythrocyte
bisphosphoglycerate concentration, affect the oxygen-hemglobin equilibrium and invalidate the algorithm [40].
Among the cases found in our literature review, Hbs Bassett, Rothschild, and Canebiere are known to have low oxygen affinity (Supporting Information—Table I). Individuals
who are carriers of these variant hemoglobins had low
SpO2 and low SaO2. Oxygen delivery however may not
necessarily be adversely affected in these individuals as
low oxygen affinity variant hemoglobins might unload more
oxygen in peripheral tissues than normal Hb A [41]. Hb
Rothschild has unique properties. In the liganded state, the
variant hemoglobin dissociates into ab dimers. It generally
has low oxygen affinity [31]. On the other hand, the T state
of the Hb Rothschild has higher affinity for ligands than normal Hb A, thus producing a molecule with alternative low
and high oxygen affinity [17,42].
Many more cases that presented with low SpO2 were later
found to have normal SaO2 by co-oximetry. Absorption spectrum
testing on several of these variant hemoglobins (Hbs Bonn,
Cheverly and Köln) revealed increased absorption near 660 nm
which led to the erroneous measurement of increased proportion of deoxy-hemoglobin by pulse oximetry (e.g., Fig. 3).
Limitations
Although we included any case of congenital Hb M that
was identified to have low SpO2, we have not addressed
acquired methemoglobinemia, which is likely the most common cause of ‘‘discordant’’ oximetry readings (previously la-
724
American Journal of Hematology
10968652, 2011, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajh.22074, Wiley Online Library on [23/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Errata
beled as the ‘‘saturation gap’’ [43,44]), but similar principles
of investigation apply. Methemoglobinemia can be caused
by a variety of toxins and drugs, and may be specifically
identified with co-oximetry. Further investigations are
needed to determine whether the methemoglobinemia is
hereditary or acquired.
Why some variant hemoglobins with low oxygen affinity
have discordant oxygen saturation readings whereas others
have concordantly low SpO2 and SaO2 is unknown. For
example, both Hb Canebiere and Hb Titusville are low oxygen affinity variants, however the former results in low SpO2
and low SaO2, suggesting a true decrease in the concentration of oxyhemoglobin, whereas in the later the normal SaO2
suggests that the low SpO2 is artifactual. Perhaps the low
proportion of Hb Titusville (15%) versus the higher concentration of Hb Canebiere (50%) in the heterozygote results
in differing magnitude of effect on oxygen-binding.
Limited data for many of the variant hemoglobins is available, such as those in Table III (Supporting Information)
which we were unable to characterize. In future, patients
identified with low SpO2 and variant hemoglobin should
have SaO2 measured simultaneously, as well as hemoglobin absorption spectrum testing to further our understanding in this area.
Author Contributions
M. Verhovsek and D.H.K. Chui contributed to concept
and design of study; literature review; data gathering and
interpretation and preparation of the manuscript for submission. M.P.A. Henderson, G. Cox, H.Y. Luo, and M.H. Steinberg contributed to concept and design of study; data interpretation; and preparation of the manuscript for submission.
References
1. Neff TA. Routine oximetry. A fifth vital sign? Chest 1988;94:227.
2. Hladik A, Lynshue K. Dyshemoglobinemias and pulse oximetry: A therapeutic
challenge. J Pediatr Hematol Oncol 2008;30:850–852.
3. Hardison RC, Chui DH, Giardine B, et al. HbVar: A relational database of
human hemoglobin variants and thalassemia mutations at the globin gene
server. Hum Mutat 2002;19:225–233.
4. Chui DHK, Steinberg MH. Laboratory diagnosis of hemoglobinopathies and
thalassemias. In: Hoffman R, Benz EJ Jr, Shattil SJ, Furie B, Siberstein LE,
McGlave P, Heslop H, editors. Hematology: Basic Principles and Practice,5th
ed. Philadelphia: Churchill Livingstone Elsevier; 2009. pp 525–533.
5. Bruns CM, Thet LA, Woodson RD, et al. Hemoglobinopathy case finding by
pulse oximetry. Am J Hematol 2003;74:142–143.
6. Molchanova TP, Wilson JB, Gu LH, et al. A second observation of the fetal
methemoglobin variant Hb F-M-Fort Ripley or alpha 2G gamma
2(92)(F8)His----Tyr. Hemoglobin 1992;16:389–398.
7. Luo HY, Irving I, Prior J, et al. Hemoglobin Titusville, a low oxygen affinity variant hemoglobin, in a family of Northern European background. Am J Hematol
2004;77:384–386.
8. Hain RD, Chitayat D, Cooper R, et al. Hb FM-Fort Ripley: Confirmation of
autosomal dominant inheritance and diagnosis by PCR and direct nucleotide
sequencing. Hum Mutat 1994;3:239–242.
9. Bard H, Peri KG, Gagnon C. The biologic implications of a rare hemoglobin
mutant that decreases oxygen affinity. Pediatr Res 2001;49:69–73.
10. Das A, Sinha S, Hoyer JD. Hemoglobin Bassett produces low pulse oximeter
and co-oximeter readings. Chest 2007;131:1242–1244.
11. Dainer E, Shell R, Miller R, et al. Neonatal cyanosis due to a novel fetal hemoglobin: Hb F-Circleville [Ggamma63(E7)His-->Leu, CAT>CTT]. Hemoglobin 2008;32:596–600.
12. Kedar PS, Nadkarni AH, Phanasgoankar S, et al. Congenital methemoglobinemia caused by Hb-MRatnagiri (beta-63CAT-->TAT, His-->Tyr) in an Indian
family. Am J Hematol 2005;79:168–170.
13. Urabe D, Li W, Hattori Y, Ohba Y. A new case of Hb F-M-Osaka [G gamma
63(E7)His-->Tyr] showed only benign neonatal cyanosis. Hemoglobin
1996;20:169–173.
14. Wu Y, Ramani GV, Gai Q, et al. Rare hemoglobinopathy presenting as progressive dyspnea. Am J Hematol 2010;85:355–357.
15. Abdulmalik O, Safo MK, Lerner NB, et al. Characterization of hemoglobin
bassett (alpha94Asp-->Ala), a variant with very low oxygen affinity. Am J
Hematol 2004;77:268–276.
16. Thuret I, Mely L, Kister J, et al. Association of HbS and a new low oxygen
affinity variant, Hb Canebiere, [beta102(G4)Asn->Lys] in a healthy child. Haematologica 2004;89:ECR31.
American Journal of Hematology
725
10968652, 2011, 8, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ajh.22074, Wiley Online Library on [23/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Errata
17. Sharma VS, Newton GL, Ranney HM, et al. Hemoglobin Rothschild (beta
37(C3)Trp replaced by Arg): A high/low affinity hemoglobin mutant. J Mol Biol
1980;144:267–280.
18. Moradkhani K, Préhu C, Old J, et al. Mutations in the paralogous human
a-globin genes yielding identical hemoglobin variants. Ann Hematol
2009;88:535–543. Available at: http://resolver.scholarsportal.info/resolve/
09395555/v88i0006/535_mitphgyihv. 10.1007/s00277–008-0624–3.
19. Schneider RG, Atkins RJ, Hosty TS, et al. Haemoglobin Titusville: Alpha94
Asp replaced by Asn. A new haemoglobin with a lowered affinity for oxygen.
Biochim Biophys Acta 1975;400:365–373.
20. Kuji A, Satoh Y, Kikuchi K, et al. The anesthetic management of a patient with
hemoglobin M(Iwate). Anesth Analg 2001;93:1192–1193; table of contents.
21. Zur B, Hornung A, Breuer J, et al. A novel hemoglobin, Bonn, causes falsely
decreased oxygen saturation measurements in pulse oximetry. Clin Chem
2008;54:594–596.
22. Holbrook SP, Quinn A. An unusual explanation for low oxygen saturation. Br J
Anaesth 2008;101:350–353.
23. Mariette S, Leteurtre S, Lambilliotte A, Leclerc F. Pulse oximetry and genetic
hemoglobinopathies. Intensive Care Med 2005;31:1597; author reply 1598.
24. Lang SA, Chang PC, Laxdal VA, Huisman TH. Haemoglobin Hammersmith
precludes monitoring with conventional pulse oximetry. Can J Anaesth
1994;41:965–968.
25. Harano K, Harano T, Shibata S, et al. Hb Okazaki [beta 93(F8) Cys----Arg], a
new hemoglobin variant with increased oxygen affinity and instability. FEBS
Lett 1984;173:45–47.
26. Katoh R, Miyake T, Arai T. Unexpectedly low pulse oximeter readings in a boy
with unstable hemoglobin Koln. Anesthesiology 1994;80:472–474.
27. Starkey CR, Davies L, Hoyer JD, et al. Clinical manifestations of hemoglobin
Chico at high altitude. J Pediatr Hematol Oncol 2006;28:760–762.
28. Hohl RJ, Sherburne AR, Feeley JE, et al. Low pulse oximeter-measured hemoglobin oxygen saturations with hemoglobin Cheverly. Am J Hematol
1998;59:181–184.
29. Yodfat UA, Vaida SJ. Unexpected falsely low pulse oximeter measurements in
a child with hemoglobin Cheverly. Anesth Analg 2006;103:259–260.
30. Cheung VY, Silverman JA. Hemoglobin koln and pregnancy. J Obstet Gynaecol Can 2008;30:907–909.
31. Danish EH, Harris JW, Ahmed F, Anderson H. Hb Rothschild (beta 37 (C2)
Trp leads to Arg): Clinical studies. Hemoglobin 1982;6:51–55.
32. Deyell R, Jackson S, Spier S, et al. Low oxygen saturation by pulse oximetry
may be associated with a low oxygen affinity hemoglobin variant, hemoglobin
Titusville. J Pediatr Hematol Oncol 2006;28:100–102.
33. Grimaldi E, Esposito P, De Caterina M, Ferranti P. Further characterisation of
Hb Okazaki [b93 (F8) Cys-Arg], a rare hemoglobin variant found in a family
from Naples, Italy. Haematologica 2004;89:ECR24.
34. Devaraj R, Wilson JB, Huisman TH. Hb Regina or alpha 2 beta 2
96(FG3)Leu----Val: A high oxygen affinity variant discovered by cationexchange HPLC. Am J Hematol 1985;19:195–200.
35. Gottschalk A, Silverberg M. An unexpected finding with pulse oximetry in a
patient with hemoglobin Koln. Anesthesiology 1994;80:474–476.
36. Sinex JE. Pulse oximetry: Principles and limitations. Am J Emerg Med
1999;17:59–67.
37. Reynolds KJ, Palayiwa E, Moyle JT, et al. The effect of dyshemoglobins on
pulse oximetry: Part I, Theoretical approach and Part II, Experimental results
using an in vitro test system. J Clin Monit 1993;9:81–90.
38. Zwart A, Buursma A, Oeseburg B, Zijlstra WG. Determination of hemoglobin
derivatives with IL 282 CO-oximeter as compared with a manual spectrophotometric five-wavelength method. Clin Chem 1981;27:1903–1907.
39. Pamidi PVA, DeAbreu M, Kim D, Mansouri S. Hydroxocobalamin and cyanocobalamin interference on co-oximetry based hemoglobin measurements.
Clinica Chimica Acta 2009;401:63–67.
40. Dimeski G. Blood Gas and pH Analysis and Related Measurements;
Approved Guideline, 2nd ed. Wayne, PA: Clinical Laboratory Standards Institute; 2009.
41. Shirasawa T, Izumizaki M, Suzuki Y, et al. Oxygen affinity of hemoglobin regulates O2 consumption, metabolism, and physical activity. J Biol Chem
2003;278:5035–5043.
42. Nagel RL.Disorders of hemoglobin function and stability. In:Steinberg MH,Forget BG,Higgs DR,Nagel RL, editors. Disorders of Hemoglobin: Genetics,
Physiology and Clinical Management. Cambridge, UK: Cambridge University
Press; 2001. pp 1167.
43. Ash-Bernal R, Wise R, Wright SM. Acquired methemoglobinemia: A retrospective series of 138 cases at 2 teaching hospitals. Medicine (Baltimore)
2004;83:265–273.
44. Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology 1989;70:112–117.
45. Sarikonda KV, Ribeiro RS, Herrick JL, Hoyer JD. Hemoglobin lansing: A novel
hemoglobin variant causing falsely decreased oxygen saturation by pulse oximetry. Am J Hematol 2009;84:541.
46. Naito Y, Takahashi T, Matsunashi T, et al. Hb Nishinomiya [Leu-Gly-inserted
between codons 69(E13) and 70(E14) of beta]: A novel unstable hemoglobin
with reduced oxygen affinity found in a patient with spherocytic hemolysis. Int
J Hematol 2002;76:146–148.
47. Stucke AG, Riess ML, Connolly LA. Hemoglobin M (Milwaukee) affects arterial oxygen saturation and makes pulse oximetry unreliable. Anesthesiology
2006;104:887–888.