University of Southern California Dissertations and Theses

Clinical significance of oxyhemoglobin dissociation curves on the interpretation of pulse oximetry in Sickle Cell disease

(USC Thesis Other)

Clinical significance of oxyhemoglobin dissociation curves on the interpretation of pulse oximetry in Sickle Cell disease

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

CLINICAL SIGNIFICANCE OF OXYHEMOGLOBIN
DISSOCIATION CURVES ON THE INTERPRETATION
OF PULSE OXIMETRY IN SICKLE CELL DISEASE
by
Lisa Katherine Morris

_______________________________________________________________

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(EXPERIMENTAL AND MOLECULAR PATHOLOGY)

August 2007

Copyright 2007 Lisa Katherine Morris
ii
Dedication

My thesis is dedicated to my parents, Darrell and Susan Morris.
iii
Acknowledgements
I would like to thank Thomas D. Coates, Robert R. Weihing, Janelle Miller, Mirna
Sweeney, Colleen McCarthy, Lingyun Ji, and Richard Sposto for the countless hours
they spent supporting this project. .
Funding
This work was supported in part by RO1-HL07180 (TDC), Grant Number MO1
RR00046 (TDC) , Children’s Hospital Los Angeles General Clinical Research
Center, and U27/CCU922106-01 from the CDC (TDC).

iv
List of Tables
Table 1: Population Demographics 7
Table 2: Fitting Parameters 11
v

List of Figures
Figure 1: Mean Oxyhemoglobin Dissociation Curves 10
Figure 2: Difference between the O2 saturations 13
Figure 3: Differences in predicted pO2 14
Figure 4: Non-linear behavior of the oxyhemoglobin dissociation 16

vi
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables v
List of Figures vi
Abstract vii
Introduction 1
Materials and Methods 3
Specimen Preparation 4
Oxyhemoglobin Dissociation Curves 5
Statistical Analysis 6

Results 7
Differences between curves 9
Derived errors in saturation 11
Effect of storage temperature 17

Discussion 18
References 20
Alphabetized Bibliography 21
Appendix 22

vii
Abstract
Some sickle cell patients have very abnormal SpO2 measurements yet appear to be
clinically well. The clinician’s interpretation of oxygenation status of a patient is
based on the relation of SpO2 to pO2 that has been calibrated through experience to
the HbAA oxyhemoglobin dissociation curve. To address the possibility that
apparent severe hypoxia in some SCD patients is due to large differences in their
individual oxyhemoglobin dissociation curves, we measured the oxyhemoglobin
dissociation curves in SCD subjects and controls, determined the within-subject and
population variance, and used the equations obtained from fitting population data
and data from individual subjects to derive the differences between observed and
corrected SaO2 in subjects with HbSS and HbSC phenotypes. There was excellent
agreement (< 5 saturation points difference) between observed SpO2 and derived
SaO2 for 87.5% of the measurements. However, about 12.5% of the saturation
measurements would be more than 10% higher if HbAA were present instead of
HbSS. This study confirms that SpO2 is a reasonable estimate of oxygenation status
in patients with sickle cell disease. However, in about 12% of the subjects, SpO2
values in the 80 to 90% range may significantly underestimate the oxygenation status
of the patient because of individual differences in the oxyhemoglobin dissociation
curve.
1
Introduction
Sickle cell disease, a genetic disorder that occurs primarily in people of African
descent, is characterized by the transformation of normally flexible, discoid red
blood cells into rigid crescent (sickle) shaped erythrocytes, which then cause vaso-
occlusion, tissue ischemia and subsequent organ damage. The production of
Hemoglobin S, an abnormal protein present in all forms of sickle cell disease, is
caused by a single point mutation in hemoglobin at position 6 of the β-globin chain
that results in a substitution of a valine residue for a glutamic acid
1
. If lysine is
substituted for glutamic acid at the same position, Hemoglobin C is produced
2
. In the
deoxy state, HbS polymerizes making red cells become rigid and take on a crescent
shape. This process, known as sickling, makes red blood cells unable to pass through
small blood vessels. HbSC patients often have similar, but milder, symptoms as
patients with HbSS
3
.
Hypoxia, which causes sickling, is monitored in patients using a pulse oximeter. This
commonly used clinical instrument measures the percent of oxyhemoglobin
saturation. One can determine the pO
2
, (the partial pressure of oxygen) from the
oxyhemoglobin saturation if the oxyhemoglobin dissociation curve is known
4
. A
clinician’s learned association between the severity of hypoxia and the pulse
oximeter reading is based on calibration of the pulse oximeter in the presence of
normal hemoglobin. Given that the oxyhemoglobin dissociation curve is different
for different types of hemoglobin, clinicians examining patients with a
hemoglobinopathy may misinterpret the severity of the patient’s hypoxia based on
2
pulse oximetry data alone.
We observed that some sickle cell patients have oxygen saturations that would be
interpreted as severe hypoxia, but they clinically appeared to be well. We suspected
that this paradox could be explained by large differences in the particular patient’s
oxyhemoglobin dissociation curve or by the measurement of the part of the
oxyhemoglobin dissociation curve that was highly non-linear. While it has been
known since 1954 that the oxyhemoglobin dissociation curve of HbSS is right
shifted
5
, there is no available data regarding the within patient variability or
population variability of the oxyhemoglobin dissociation curve nor is there a way to
convert the percent saturation measured by the pulse oximeter,SpO
2
, to pO
2
for an
individual patient based on his pulse oximetry data. Therefore we determined the
functional parameters based on population data that allow calculation of pO
2
from
SpO
2
, the within patient variability and population variability of the hemoglobin AA,
SS and SC oxyhemoglobin dissociation curves, and the sample storage conditions for
measurement of the oxyhemoglobin dissociation curves. When applied to a
population of sickle cell patients, we found that differences in the oxyhemoglobin
dissociation curve in certain patients introduce errors in the magnitude of the
estimation of pO
2
large enough to affect clinical judgments.

3
Materials and Methods
This study was carried out according to a protocol approved by the Committee on
Clinical Investigation, the Institutional Review Board of Children’s Hospital-Los
Angeles Caucasian, Hispanic, and Black males and females 13 years of age and older
who have HbSS, HbSC, HbS β
-
and HbAA were eligible for participation. None of
the subjects had a vaso-occlusive crisis within 2 weeks or had transfusions within
past 3 months.
4
Specimen Preparation
Blood samples were anti-coagulated with heparin and kept at either room
temperature or at 4
o
C until the oxyhemoglobin dissociation curve was measured.
Blood samples were obtained from each subject on at least two occasions 7 to 14
days apart and assayed at times varying from less than 1 hour to 24 hours after
phlebotomy according to the protocol described in the next section. Complete blood
counts, reticulocyte counts, fetal hemoglobin (HbF) levels, and hemoglobin
phenotype were determined in the clinical laboratory by standard methods.
Oxyhemoglobin Dissociation Curves
Oxyhemoglobin dissociation curves were measured at 37
o
C using a TCS Medical
Products Company Hemox-Analyzer model B (TCS Medical Products, New Hope,
PA). At the time of assay, a solution of 4 mL of hemox buffer, 20 µL of hemox
antifoaming agent, and 50-100 µL of heparinized blood was made and filled into the
Hemox-Analyzer. After allowing the pO
2
in the instrument to stabilize at 37.0
o
C,
and adjusting the O
2
potentiometer to the ambient barometric pressure, the blood
sample was deoxygenated by bubbling nitrogen through the suspension. The
hemoglobin saturation was measured spectrophotometrically while the sample was
aerated and the pO
2
rose from 0 to approximately 150 mmHg. The pO
2
and percent
saturation were recorded digitally along the entire range from 0 to 100% saturation.
5
Statistical analysis.
The relationship between partial oxygen pressure and the percentage of saturation of
hemoglobin followed a sigmoidal shape. PROC NLIN (SAS) was used to fit the
sigmoidal models. After the model parameters were estimated, pO
2
required to
achieve 30%-100% saturation of hemoglobin was calculated for all the experiments
for each patient. Linear mixed effect models
6
were then used to determine whether
there was any difference in the mean pO
2
levels required to achieve 30% to 100%
saturation of hemoglobin between patients with different hemoglobin phenotypes
(AA, SS, and SC), between blood samples stored for different durations of time after
blood draw and before being assayed ( <1 hour, 1-8 hours, or 23+ hours), or between
blood samples stored under room temperature and those stored under refrigerator
temperature. Given that data obtained from each patient was potentially correlated,
the linear mixed models were used to account for the within-patient correlation.
Pooled within patient standard deviation (SD)
7
was calculated for the following
three groups of blood samples: 1) blood samples stored at room temperature for less
than 1 hour; 2) blood samples stored at room temperature for up to eight hours; and
3) blood samples stored at 4
o
C. Pooled within patient SD and its 95% confidence
interval was provided for pO
2
required to reach 70%, 80%, 90%, 95% and 97%
saturation of hemoglobin. All p-values reported were two-sided. Data were analyzed
with SAS version 9.1 (SAS Institute, Cary, NC, USA).

6
Results
Patient population
Table 1 shows the characteristics of the study population. All subjects had studies
replicate studies done on the first day they came in and at a follow up appointment
fourteen days later. Many had between 2 and 4 pairs of measurements done. In total,
166 oxyhemoglobin dissociation curves were created from 13 normal controls and 21
patients with SCD.
7
Table 1. Patient Demographics

Control SS SC
N 13 17 4
Age

Mean 24.75 19.46 22
Median 25 18 22.5
Range 16 - 32 13 - 26 16 – 25
Sex

Male 4 5 1
Female 9 8 3
Hemoglobin

Mean 13.9 9.04 10.9
Median 13.8 9 11.35
Range 12.30-16.20 5.70 - 11.10 4.11 – 13.00
Hematocrit

Mean 40.72 26.2 29.92
Median 40.25 26.25 31.85
Range 37.10 – 45.20 15.40 - 32.60 10.90 - 34.90
Reticulocyte

Mean 1.12 8.19 3.38
Median 1.05 6.75 3.5
Range 0.80 - 1.80 1.50 – 22.00 2.10 - 5.30
Fetal
Hemoglobin*

Mean 0.74 10.82 1.88
Median 0.7 11.5 0.8
Range 0.30 - 2.10 2.30 - 21.30 0.60 – 5.30
MCV
Mean 88.34 92.32 72
Median 87.7 92.6 77.4
Range
80.70-
100.00 57.80-120.60 10.90-97.90
*The fetal hemoglobin in SS patients is higher than expected. This is because most
of the SS patients were on hydroxyurea which is known to raise the HbF levels
8
.
8
Differences between curves
Data from 34 subjects was available for analyses, and of the 34 patients 13 were of
AA Hb-phenotype, 4 SC, 15 SS, and 2 S β
−
thalassemia. S β
-
thalassemia patients
were grouped with the HbSS patients in the analyses. Figure 1 shows the mean
oxyhemoglobin dissociation curves derived from the population of normal controls
and patients with sickle cell disease. The oxyhemoglobin dissociation curve for
sickle for SS is significantly right shifted compared to as the in normal controls
indicating lower binding affinity. The oxyhemoglobin dissociation curve for HbSC
is not significantly different in comparison to the normal controls. Figure 1. Mean
oxyhemoglobin dissociation curves for 17 patients with HbSS (long dash), 4
subjects with HbSC (dashed) and 13 HbAA (solid). Every subject was studied on at
least two days.
9

When we examined the pO
2
at each of five specific saturation levels (70%, 80%,
90%, 95% and 97%), the hemoglobin of the patients with HbSS required
significantly higher pO
2
to achieve the same saturation level (p<0.02) than HbAA.
Controlling for temperature and duration of time for storing blood, pO
2
required for
SS patients to reach 70%, 80%, 90%, 95% and 97% saturation of hemoglobin was
5.6, 5.9, 6.2, 6.4, and 6.5 mmHg higher than for AA patients, respectively. After
adjusting for the HbF levels and reticulocytes, pO
2
levels required to achieve 70%,
80%, 90%, 95% and 97% saturation of hemoglobin did not significantly differ
between HbAA subjects and HbSC subjects. The within patient variability is
relatively small (3.0-4.2 mmHg) at 97% saturation indicating that there is good
reproducibility in the measurement.
10
Derived errors in saturation
We have measured the entire oxyhemoglobin dissociation curve in a population of
patients and have determined the variability characteristics. The equations that we
present here allow calculation of estimated pO
2
’s from saturation data for HbSS,
HbSC, and HbAA. We have included the fitting parameters for the means, upper and
lower 95% confidence intervals based upon population variability in the appendix.
Equation 1:
pO
2
=

0 ) 1
2
2 1
ln( x dx
A y
A A
+ −
−
−

Equation 2:
SpO
2
= A2 + _ A1 – A2_
1 + e
(x-x0)/dx)

Table 2: Fitting parameters

m
HbAA
UL
HbAA
LL
HbAA
mHbSS
UL
HbAA
LL
HbAA
m
HbSC
UL
HbSC
LL
HbSC
A1 -127.9 -130.7 -125.5 -50.05 -52.84 -47.11
-
55.92
-
62.05 -49.5
A2 100.1 100 100.2 99.88 99.79 99.99 99.96 99.78 100.2
x0 1.371 1.251 1.416 16.52 16.99 16.04 15.28 15.49 14.97
dx 17.8 18.6 17.01 17.25 18.33 16.18 15.84 17.46 14.25

The parameters for the above equations were derived from fits of approximately
1000 data points for each of 166 individual oxyhemoglobin dissociation curves. Each
individual data set fit the above equations with r
2
> 0.999, p < 0.00001. The
parameters in table 2 are derived by fitting all 166 sets of dissociation data combined
by hemoglobin type. Using the fitting parameters in the table above we are able to
11
determine the pO
2
for any SaO
2
(actual percent saturation) based upon the mean
characteristics of a particular population (HbSS, HbSC, HbAA). Similarly, we can
determine this relation for individual samples using the fitting parameters for the
individual patient. This approach allowed us to analyze the error in estimated pO
2
in
SaO
2
ranges that are uncommonly found in patients with sickle cell disease. To
construct figure 2, we determined the estimated pO
2
at discrete SaO
2
values of 80,
85, 90, 92.5, 95, 97.5, and 99 for 63 samples in 17 patients with HbSS. We then
entered the resulting pO
2
values into equation 2 using the derived mean HbAA fitting
parameters to model what the SpO
2
would be if the oximeter were looking at HbAA
instead of HbSS. Figure 2 shows the frequency of the differences between the SpO
2
-
SS and SpO
2-AA
at SaO
2-SS
values of 80, 85, 90, 92.5, 95, 97.5 and 99. The majority
of the patient’s estimated SpO
2
is only two saturation points away from the estimated
SaO
2
. However, 12.8% of the SpO
2
values in sickle patients are at least 5 saturation
units lower than the corresponding SpO
2
for HbAA In five patients, the SaO
2
was
between 9 and 13 saturation points lower than the corresponding saturation for
HbAA. Figure 2 also indicates that the largest differences occur at the lowest
saturation values, a range where clinical intervention is most likely to occur to
correct the apparent hypoxia. Differences of this magnitude are likely to affect
clinical decision-making and may lead to errors in as many as 12.8% of patients with
HbSS.
12
Figure 2.

Figure 2. Difference between the O
2
saturations of 80%, 90%, 95%, and 99% for
sickle cell subjects and the O
2
saturation corrected for HbAA.
13
Figure 3.

Figure 3.. Differences in predicted pO
2
based on individual oxyhemoglobin
dissociation curves for sickle cell subjects and pO
2
based on the mean
oxyhemoglobin dissociation curve derived from all HbAA subjects. At the 99%
saturation level, there is up to a 60mmHg difference in the pO
2
. At 85% saturation,
there are five patients who have a pO
2
mmHg higher than one would expect based on
the average.
Pulse oximetry in sickle cell disease correctly measures the SaO
2
. The problem is
that the SpO
2
from a sickle cell patient corresponds to a pO
2
is actually higher than
the clinician realizes because his reaction to pulse oximetry measurements is based
on the SpO
2
-pO
2
relation for HbAA. Figure 3 shows the derived differences in pO
2

14
for a group of sickle cell patients at selected SaO
2
values compared to the pO
2
at the
same SpO
2
derived from the HbAA oxyhemoglobin dissociation curve.
Underestimates of the pO
2
of up to 20 mmHg are seen at all saturation levels.
However, even greater differences are seen at higher saturations. Many of these pO
2

differences are highly clinically significant. The fact that the differences in pO
2

seem greater at high saturation (figure 3) while the differences in saturation seen
greater at lower saturations (figure 2) can be explained by the highly nonlinear
behavior of the oxyhemoglobin dissociation curve as depicted in figure 4. Clearly,
the slope of the dissociation curve becomes very high at saturations greater than 95
to 97%. Thus the difference in predicted pO
2
between HbSS and HbAA for a given
change in saturation for HbSS is very large in at high SaO
2
values. To illustrate the
effect of the two oxyhemoglobin dissociation curves of HbSS and HbAA, we have
plotted the mean curve for HbAA and the lower 95% confidence interval for HbSS.
The difference between these is seen in the inset. Clearly the error in the pO
2

estimate is very high at and above the inflection point in the curve. However, at
these absolute pO
2
values, there would be no adverse consequences to clinical
decisions making.
15
Figure 4.

Figure 4. Signficant non-linear behavior of the oxyhemoglobin dissociation curve in
the clinically important saturation range between 90 and 100%. Plotted are the mean
hemoglobin saturation for HbAA (N = 13, solid line) and the lower 95% confidence
interval for HbSS (N = 17; dashed line) calculated from the HbSS population
variance. The inset shows the difference in pO2 between HbAA and the lower 95%
confidence interval for HbSS as a function of SaO
2
.
16
Effect of storage temperature
Temperature affected the duration of time that the specimen could be stored and still
obtain an accurate determination of the oxyhemoglobin dissociation curve. At room
temperature, blood assayed more than 8 hours from time of phlebotomy resulted in a
dissociation curve that underestimated the pO
2
. This was true for 70%, 80%, 90%,
95% and 97% saturation of hemoglobin. For example, a saturation of 70% for blood
samples that were stored at room temperature for 23 hours was 7.5 mmHg lower than
for blood samples that were stored under room temperature for less than an hour.
Blood samples stored at 4
o
C for 23-94 hours did not show a significant difference in
pO
2
from blood samples stored under room temperature for less than an hour;
however, our data was limited for storage times between 24 and 96 hours. We
recommend limiting the time between phlebotomy and assay to 24 hours. Storage of
blood samples for less than 8 hours at room temperature had no effect on the assay
results. Based on this data, blood specimens can be stored for up to eight hours at
room temperature and 24 hours at 4°C prior to assay. These results held for both
HbSS and HbAA. However, we found that hemoglobin SS was more sensitive to
storage.

17
Discussion
For the past two decades, many have debated the accuracy of pulse oximetry in
sickle cell disease
9-13
. Overall, our results and those of others
12
show that the SpO
2

from pulse oximetry reflects pO
2
accurately enough for clinical purposes in sickle
cell patients. However, based on direct comparison to oxyhemoglobin saturation
measured by blood gas determination, there can be significant errors in estimation of
oxygenation status from SpO
2
in certain individuals with sickle cell disease
10
. These
studies are limited by the fact that comparisons can be made only at naturally
occurring saturation levels. Nonetheless, they documented significant errors in some
patients
10
. Since others have shown, based on direct comparison to blood gas
measurement, that the oxyhemoglobin dissociation curve from a particular patient
can be used to correct for the difference between SpO
2
measured by pulse oximetry
and SaO
2
by blood gas
4
, we used the oxyhemoglobin dissociation curves to assess
possible errors in interpretation of pulse oximetry results in sickle cell patients that
may occur due to differences in the HbSS and HbAA oxyhemoglobin dissociation
curves. Clinician’s interpretation of oxygenation status of a patient is “calibrated”
based on the relation of SpO2 to pO2 for HbAA, not HbSS. Using fitting
techniques, we were able to derive differences in the SpO2 observed for HbSS and
the SpO2 that would occur for HbAA at the same pO2 over a very wide range of
SpO2 values, including very low saturations that are not commonly seen clinically.
These calculations showed that there is excellent agreement (< 5 saturation points
difference) for 87.5% of the measurements (figure 2). However, about 12.5% of the
18
saturation measurements would be more than 10% higher if HbAA were present
instead of HbSS. Thus, some HbSS patients with clinically worrisome SpO2 values
in the 85 to 90% range would be in the 95 to 100% range if the oximeter were
reporting HbAA instead of HbSS. In summary, we have measured the
oxyhemoglobin dissociation curves in a large number of subjects with sickle cell
disease and normal controls and determined the sample storage conditions for the
assay system. We have determined the within subject and population variability and
provided the fitting constants that permit reconstruction of the mean saturation
curves and confidence intervals for HbAA, HbSS, and HbSC. Using these fitting
parameters, we have derived the SpO2 values for HbAA that correspond to SpO2
values for HbSS over a wide range of SaO2. We conclude that SpO2 is a reasonable
estimate of oxygenation status in patients with sickle cell disease. However, in about
10% of the subjects, SpO2 values in the 80 to 90% range may significantly
underestimate the oxygenation status of the patient leading the clinician to think the
patient is in a more serious pulmonary condition than is actually the case.

19
References

1. Ingram VM. A specific chemical difference between the globins of normal
human and sickle-cell anaemia haemoglobin. Nature 1956;178(4537):792-4.
2. Hunt JA, Ingram VM. Allelomorphism and the chemical differences of the
human haemoglobins A, S and C. Nature 1958;181(4615):1062-3.
3. Platt OS, Thorington BD, Brambilla DJ, et al. Pain in sickle cell disease--
Rates and risk factors Pain in sickle cell disease. Rates and risk factors.
NEnglJMed 1991;325(1):11.
4. Weston Smith SG, Glass UH, Acharya J, Pearson TC. Pulse oximetry in
sickle cell disease. Clin Lab Haematol 1989;11(3):185-8.
5. Becklake MR, Griffiths SB, Mc GM, Goldman HI, Schreve JP. Oxygen
dissociation curves in sickle cell anemia and in subjects with the sickle cell
trait. J Clin Invest 1955;34(5):751-5.
6. Singer J. Using SAS Proc Mixed to fit multilevel models, hierarchical
models, and individual growth models. J Educ Behav Stat 1998;24:323–55.
7. Dunn OJ, Clark VA. Applied statistics: analysis of variance and regression. 1
ed. New York: John Wiley & Sons; 1974.
8. Dover GJ, Humphries RK, Moore JG, et al. Hydroxyurea induction of
hemoglobin F production in sickle cell disease: relationship between
cytotoxicity and F cell production. Blood 1986;67(3):735-8.
9. Pianosi P, Charge TD, Esseltine DW, Coates AL. Pulse oximetry in sickle
cell disease. Arch Dis Child 1993;68(6):735-8.
10. Blaisdell CJ, Goodman S, Clark K, Casella JF, Loughlin GM. Pulse oximetry
is a poor predictor of hypoxemia in stable children with sickle cell disease.
ArchPediatrAdolescMed 2000;154(9):900.
11. Homi J, Levee L, Higgs D, Thomas P, Serjeant G. Pulse oximetry in a cohort
study of sickle cell disease. Clin Lab Haematol 1997;19(1):17.
12. Ortiz FO, Aldrich TK, Nagel RL, Benjamin LJ. Accuracy of pulse oximetry
in sickle cell disease. AmJRespirCrit Care Med 1999;159(2):447.
13. Rackoff WR, Kunkel N, Silber JH, Asakura T, Ohene-Frempong K. Pulse
oximetry and factors associated with hemoglobin oxygen desaturation in
children with sickle cell disease. Blood 1993;81:3422.

20
Alphabetized Bibliography
Becklake MR, Griffiths SB, Mc GM, Goldman HI, Schreve JP. Oxygen dissociation
curves in sickle cell anemia and in subjects with the sickle cell trait. J Clin
Invest 1955;34(5):751-5.

Blaisdell CJ, Goodman S, Clark K, Casella JF, Loughlin GM. Pulse oximetry is a
poor predictor of hypoxemia in stable children with sickle cell disease.
ArchPediatrAdolescMed 2000;154(9):900.

Dover GJ, Humphries RK, Moore JG, et al. Hydroxyurea induction of hemoglobin F
production in sickle cell disease: relationship between cytotoxicity and F cell
production. Blood 1986;67(3):735-8.

Dunn OJ, Clark VA. Applied statistics: analysis of variance and regression. 1 ed.
New York: John Wiley & Sons; 1974.

Hunt JA, Ingram VM. Allelomorphism and the chemical differences of the human
haemoglobins A, S and C. Nature 1958;181(4615):1062-3.

Homi J, Levee L, Higgs D, Thomas P, Serjeant G. Pulse oximetry in a cohort study
of sickle cell disease. Clin Lab Haematol 1997;19(1):17.

Ingram VM. A specific chemical difference between the globins of normal human
and sickle-cell anaemia haemoglobin. Nature 1956;178(4537):792-4.

Ortiz FO, Aldrich TK, Nagel RL, Benjamin LJ. Accuracy of pulse oximetry in sickle
cell disease. AmJRespirCrit Care Med 1999;159(2):447.

Pianosi P, Charge TD, Esseltine DW, Coates AL. Pulse oximetry in sickle cell
disease. Arch Dis Child 1993;68(6):735-8.

Platt OS, Thorington BD, Brambilla DJ, et al. Pain in sickle cell disease--Rates and
risk factors Pain in sickle cell disease. Rates and risk factors. NEnglJMed
1991;325(1):11.

Rackoff WR, Kunkel N, Silber JH, Asakura T, Ohene-Frempong K. Pulse oximetry
and factors associated with hemoglobin oxygen desaturation in children with
sickle cell disease. Blood 1993;81:3422.

Singer J. Using SAS Proc Mixed to fit multilevel models, hierarchical models, and
individual growth models. J Educ Behav Stat 1998;24:323–55.

Weston Smith SG, Glass UH, Acharya J, Pearson TC. Pulse oximetry in sickle cell
disease. Clin Lab Haematol 1989;11(3):185-8.
21

22
Appendix:

This is the equation to find the pO2 of any patient given a saturation level written in
a fomat that can be typed into a cell in an Excel spreadsheet..
To derive pO2 from saturation type the following formula into a cell:
= (LN((D2-E2)/(J2-E2)-1)*G2+F2)

Where
D2 = fitting parameter A1
E2 = fitting parameter A2
J2 = Saturation level
G2 = fitting parameter dx
F2 = fitting parameter x0

D2, E2 etc refer to cells that contain the respective fitting constants. If the saturation
is typed into cell J2, the pO2 will appear in the cell in which the formula was typed.

Use this equation to find what the pO2 would be translated to if the patient had
hemoglobin A. To derive SaO2 from pO2 type the following formula into a cell:

= N2 + (M2-N2)/(1 + EXP((Q2 -O2)/P2)))

Where
N2 = The mean HbAA A2
M2 = The mean HbAA A1
O2 = The mean HbAA x0
P2 = The mean HbAA dx
Q2 = pO2

M2, N2 etc refer to cells that contain the respective fitting constants. If the pO2 is
typed into cell Q2, the SaO2 will appear in the cell in which the formula was typed.

m
HbAA
UL
HbAA
LL
HbAA
mHbSS
UL
HbAA
LL
HbAA
m
HbSC
UL
HbSC
LL
HbSC
A1 -127.9 -130.7 -125.5 -50.05 -52.84 -47.11
-
55.92
-
62.05 -49.5
A2 100.1 100 100.2 99.88 99.79 99.99 99.96 99.78 100.2
x0 1.371 1.251 1.416 16.52 16.99 16.04 15.28 15.49 14.97
dx 17.8 18.6 17.01 17.25 18.33 16.18 15.84 17.46 14.25

Abstract (if available)

Abstract Some sickle cell patients have very abnormal SpO2 measurements yet appear to be clinically well. The clinician's interpretation of oxygenation status of a patient is based on the relation of SpO2 to pO2 that has been calibrated through experience to the HbAA oxyhemoglobin dissociation curve. To address the possibility that apparent severe hypoxia in some SCD patients is due to large differences in their individual oxyhemoglobin dissociation curves, we measured the oxyhemoglobin dissociation curves in SCD subjects and controls, determined the within-subject and population variance, and used the equations obtained from fitting population data and data from individual subjects to derive the differences between observed and corrected SaO2 in subjects with HbSS and HbSC phenotypes. There was excellent agreement (< 5 saturation points difference) between observed SpO2 and derived SaO2 for 87.5% of the measurements. However, about 12.5% of the saturation measurements would be more than 10% higher if HbAA were present instead of HbSS. This study confirms that SpO2 is a reasonable estimate of oxygenation status in patients with sickle cell disease. However, in about 12% of the subjects, SpO2 values in the 80 to 90% range may significantly underestimate the oxygenation status of the patient because of individual differences in the oxyhemoglobin dissociation curve.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A novel role for hypoxia-inducible factor-1alpha (HIF-1alpha) in the regulation of inflammatory chemokines and leukotriene expression in sickle cell disease

PDF

Modeling of cardiovascular autonomic control in sickle cell disease

PDF

Susceptibility-weighted MRI for the evaluation of brain oxygenation and brain iron in sickle cell disease

PDF

Cerebrovascular disease of white matter in patients with chronic anemia syndrome

PDF

The role of miRNA and its regulation in pulmonary hypertension in sickle cell disease

PDF

Photoplethysmogram-based biomarker for assessing risk of vaso-occlusive crisis in sickle cell disease: machine learning approaches

PDF

Placenta growth factor-miRNAs-lncRNAs axis in the regulation of ET-1 gene involved in pulmonary hypertension in sickle cell disease

PDF

Role of the autonomic nervous system in vasoconstriction responses to mental stress in sickle cell disease: a bioengineering perspective

$Quantitative MRI for the measurement of cerebral oxygen extraction fraction in sickle cell disease$

PDF

Quantitative MRI for the measurement of cerebral oxygen extraction fraction in sickle cell disease

PDF

Estimation of cognitive brain activity in sickle cell disease using functional near-infrared spectroscopy and dynamic systems modeling

Asset Metadata

Creator Morris, Lisa Katherine (author)

Core Title Clinical significance of oxyhemoglobin dissociation curves on the interpretation of pulse oximetry in Sickle Cell disease

School Keck School of Medicine

Degree Master of Science

Degree Program Experimental and Molecular Pathology

Publication Date 08/10/2007

Defense Date 05/19/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,pulse oximetry,sickle cell disease

Language English

Advisor Coates, Thomas D. (committee chair), Lawlor, Elizabeth R. (committee member), Wood, John C. (committee member)

Creator Email lkmorris@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m770

Unique identifier UC179566

Identifier etd-Morris-20070810 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-535862 (legacy record id),usctheses-m770 (legacy record id)

Legacy Identifier etd-Morris-20070810.pdf

Dmrecord 535862

Document Type Thesis

Rights Morris, Lisa Katherine

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu