Carboxyhemoglobin and Methemoglobin Levels in Banked Blood Ichiro Uchida, MD,* Chikara Tashiro, MD,? Yoo Hee Koo, MD,+ Takashi Mashimo, MD& Ikuto. Yoshiya, MD11 Deparrtnent Osaka,

of

Anesthesiology,

Carboxyhemoglobin

to the duration

globin

level decreased

2.0%

(SD)

In u separate hemoglobin to 2.4Sz

!iAssistant I( to 4.2%.

blood ,for 21 duys. The meun initial

2 1 .S%, and the mean hay-life

+- 0.6%

ut the end of’storage. hemoglobins

Anesthesia, methemoglobin.

Keywords:

increased

from an initial

The use of’bunked

could be a potential

transfusion;

le.1~~1of carboxy-

of carboxyhemog-lobin

was

I. 3% ? 0.2%

blood contain@

risk in critically

high

ill patients.

blood, stored, carboxyhemoglobin,

Introduction

Professor Professor

Address reprint requests to Dr. Uchida at the Department of Anesthesiology, Osaka University Medical School, Fukushima I- I50, Fukushima-ku, Osaka 553, Japan. Keceived for publication August 2 I, 1989; revised manuscript accepted for publication November 29, 1989. 0 1990 Butterworth

86

time increused,

47 days. Methemoglobin

bez~elsqf these ubnormal

Professor

were determined.

study, blood drawn from six -i)olunteur:\who had smoked two rigarettes was 4.4%

upproximutely

$Visiting

School.

OY, to 9.6%>. Methemoglobin

a meun level of’l.6Ye

euch was stored as bunked

tAssociate Professor, present address: Chairman, Department of AnesthesiaOsaka Medical Center and Research lnstitute for Maternal and Child Health, Osaka, Japan

Medical

levels in 3 12 units off bunked

of storage

as the storage

with a runge from

.storuge, showing

Fellow

University

and methemoglobin

their relationship

*Postdoctoral

0saka

,Japan.

J. Clin.

Publishers

Anesth.,

vol. 2, March/April

In recent years, the potential risk of carboxyhemoglobin in banked blood has been reported. i.BAronow et al.’ stated that the increased carboxyhemoglobin levels of banked blood would be a potential risk factor impairing cardiovascular function when a massive transfusion of such blood was performed in critically ill patients with severe cardiac and/or puhnonary disease. With regard to the amount of hemoglobin available for oxygen transport, increased levels of methemoglobin, which has no capacity to bind oxygen, in banked blood would be another potential risk factor for hypoxemia in those patients. Several types of spectrophotometric CO-oximeters have become available for the determination of the levels of oxyhemoglobin, reduced hemo1990

HbCO and MrtHb levels in banked blood: Uchida et al.

globin, carboxyhemoglobin, and methemoglobin in patients’ blood. They automatically measure the light absorbance of a sonicated blood sample using four or more light wavelengths and accurately denote the percentage fraction of each hemoglobin.” A CO-oximeter can therefore be used to determine the percentage of dyshemoglobins in banked blood. Using a CO-oximeter, this study was performed with the following aims: 1. To determine levels of carboxyhemoglobin and methemoglobin in citrate-phosphate-dextrose banked blood 2. To clarify the relationship between these levels and the number of storage days in a plastic bag 3. To discuss these data with respect to the potential risk posed to critically ill patients

In a separate study performed with the approval of the Osaka University Hospital Ethics Committee and written informed consent from volunteers, venous blood was collected from six healthy male volunteers (ages 22-30 years) immediately after they had smoked two cigarettes each. This blood was stored at 4°C in the same plastic bag containing citrate-phosphate-dextrose as is used to store banked blood in -Japan. A 2 ml sample was drawn anaerobically from the bag every 2 to 4 days, and the percentage fractions of the dyshemoglobins were determined during storage for 21 days. Statistical analysis of differences was performed by one-way analysis of variance and Student’s t-test. Linear regression and correlation coefficients were used to analyze the relationship between these levels and the storage day in the smoking study. A significance level of p < 0.05 was used to declare statistical significance.

Materials and Methods Three hundred twelve units of blood were randomly selected from citrate-phosphate-dextrose banked blood that had been administered to surgical patients in the operating room of Osaka University Hospital from September to November 1987. All of the blood examined was supplied by the Osaka Blood Center of the Japanese Ked Cross and was stored in plastic bags (BB-SC206J, Terumo, Tokyo, Japan) at 4°C before transfusion. After recording the sampling date and the storage period, a 2 ml sample was drawn into a heparinized syringe, and the percentage fractions of oxyhemoglobin, reduced hemoglobin, carboxyhemoglobin, and methemoglobin were determined with a calibrated CO-oximeter (Model 2500, Ciba-Corning Diagnostics Co., Medfield, MA). Simultaneously, pH, partial pressure of oxygen and carbon dioxide (PO, and PCO,, respectively), and oxygen content were determined at 37°C with a blood gas analyzer (Model 178, Ciba-Corning Diagnostics Co., Medfield, MA) connected to the CO-oximeter. Table 1. Distribution the Banked Blood

of Carboxyhemoglobin

Results Carboxyhemoglobin and Methemoglobin Levels in Banked Blood The mean carboxyhemoglobin level of banked blood was 1.42% & 1.97% (SD) (range 0% to 9.7%), and the mean methemoglobin level was 1.57% + 0.42% (range 0.5% to 4.1%). The numbers and percentages of units with various levels of carboxyhemoglobin and methemoglobin are shown in Table I.

Relationship of Percent Carboxyhemoglobin, Percent Methemoglobin, pH, PO,, PCO,, and Oxygen Content to the Duration qf Storage The blood samples were divided into four groups according to the duration of storage: 0 to 2 days (group and Methemo+bin

Methemoglobin

Carboxyhemoglobin Range

Levels in

Number

51.5% 5 3 .5 $6 < -. 5 5 %I S7.5%8 >7.5’%8

219 41 32 13 7

(70.1%) (13.1%) (10.3%) (4.2%) (2.2%)

Total

312

(100%)

Range

Number

50.75% 5 1.25%’ 51.75% 52.25% >2.25%

5 5X 153 79 17

(1.6%) (18.6%) (49.0%) (25.3%) (5.5%)

312 (loo%>)

J. Clin. Anesth.,

vol. 2, March/April

1990

87

Original Contributions Table 2. Percent Carboxyhemoglobin (HbCO), Percent Methemoglobin (MetHb), Carbon Dioxide (PO,, PCO,), and Oxygen Content (O&T) in Banked Blood

pH, Partial Pressure of Oxygen and

Group 1 (n = 81)

Storage: group

I = 0

(n 2 36)

to 2 days; group 2 = 3 to 7 days; group 3

3 (n = 154)

Significance (Student’s t-test)

(n =4 31)

8 to 14 das;

group

4 =

I.'to "0 days.

All values are mean r+_SD. *p < 0.05. +$!< 0.01. tt, < 0.00 1.

3 to 7 days (group 2), 8 to 14 days (group 3), and 15 to 20 days (group 4). Using one-way analysis of variance, the percent carboxyhemoglobin (p < 0.05), percent methemoglobin (p < O.Ol), pH (p < O.Ol), PCO,, ($J < O.Ol), PO, ($I < 0.05), and oxygen content (p < 0.05) showed significant differences among the four groups. The mean value of each group and the differences between the groups are summarized in Table I. When storage was prolonged, the carboxyhemoglobin level decreased and the methemoglobin level increased. During storage, the pH decreased and the PO,, PCOB, and oxygen content increased (Table I),

In this study, approximately 30% of the 312 units had carboxyhemoglobin levels higher than 1.5%, which is the level when 9 parts per million of carbon monoxide (Air Quality Standard) is equilibrated with

% HbCO

2).

The venous blood of six volunteers on the first day of storage showed a mean carboxyhemoglobin level of 4.4% -+ 1.6% (SD) (range 2.7% to 6.1%). During the storage period, carboxyhemoglobin values decreased exponentially, and the mean half-life calculated from the regressions was 46.7 +- 22.6 days (SD) (Figure 1). The mean initial methemoglobin level was 1.3% + 0.2% (range 1.1% to 1.6%), and the level increased gradually to 2.4% 2 0.6% (range 2. I % to 3.0%) at the end of storage. There was a significant correlation with the length of storage (Figure 2).

Discussion Decreased levels of 2,3-diphosphoglycerate and adenosine triphosphate in preserved red cells increase oxygen affinity and decrease oxygen delivery. Also, elevated methemoglobin levels in acid-citrate-dextrose banked blood4 result in decreasing oxygen transport. The potential risk of carboxyhemoglobin in banked blood also has been reported.‘,’ 88

J. Clin. Anesth., vol. 2, March/April 1990

0

5

10

STORAGE

15

20

DAY

Figure 1. The relation of percent carboxyhemoglobin (% HbCO) plotted as the logarithm scale to the length of stat-age in six volunteers’ blood. All data and the regression lines obtained from six volunteers are shown. All correlation coefficients (r) are significant (p < 0.05). The mean halflife calculated from the regressions are 16.7 F 22.6 (SD) days. Open circles = volunteer 1; open triangles = volunteer 2; open squares = volunteer 3; solid circles = volunteer 4; solid triangles = volunteer 5; solid squares y volunteer 6.

HbCO and MetHb levels in banked blood: Uchida et al.

1

I.

L

5

0

10

15

STORAGE

20

DAY

Figure 2. The relation of percent methemoglobin (Y? MetHb) plotted as the logarithm scale to the length of storage in six volunteers’ blood. All data and the regression lines obtained from six volunteers are shown. All correlation coefficients (r) are significant. Open circles = volunteer 1; open triangles = volunteer 2; open squares = volunteer 3; solid circles = volunteer 4; solid triangles = volunteer 5; solid squares = volunteer 6.

hemoglobin.” smoking globin

The

cigarettes

venous blood of the volunteers had a mean

level of 4.4%,

which

initial

was higher

after

carboxyhemothan

that of

blood examined (p < 0.01). Along with atmospheric carbon monoxide pollution, cigarette smoking is associated with increased levels of carboxyhemoglobin in the blood. Stewart et al.2 and Spieb et al.” showed that banked blood from smokers had relative high levels of carboxyhemoglobin compared with blood from nonsmokers and that levels of up to 13.5% have been recorded in donor blood. Regarding the adverse effects of carboxyhemoglobin alone, a 2% level of carboxyhemoglobin can aggravate angina pectoris due to coronary artery disease.7 A high carboxyhemoglobin level would be a potential risk when the banked blood containing a high level of carboxyhemoglobin is administered to patients with severe cardiac and/or pulmonary disease. Elevated carboxyhemoglobin, which might result from transfusion with a high level of carboxyhemoglobin,*Bg could consequently reduce the available hemoglobin for oxygen transport and be a critical factor for hypoxemia in such high-risk patients. Although the in viuo half-life of carboxyhemoglobin in healthy humans is approximately 40 minutes when breathing 100% oxthe banked

ygen and 4 hours when breathing air,‘(’ the half-life would be considerably prolonged in hypoxic patients. Moreover, carboxyhemoglobin induces a leftward shift of the oxyhemoglobin dissociation curve,1L inhibiting the release of oxygen at the tissue level. The current data showed that 30% of the 3 12 units had carboxyhemoglobin levels greater than 1.5%; the highest level was 9.7%. Carboxyhemoglobin levels gradually decreased during the storage period. Aronow et al.’ reported that 49% of banked blood examined had carboxyhemoglobin levels above 1.5%. They determined the level using the tubing segment of the bag and did not consider the relationship to storage because it had been reported that the length of storage of banked blood did not affect the carboxyhemoglobin leve1.6,7.12Although differences due to the population of smokers and atmospheric pollution cannot be ignored, this discrepancy of the carboxyhemoglobin level appears to be mainly caused by the difference in the gas permeability of the plastic bags used as blood containers.‘” Carbon monoxide escapes from the blood bag in exchange for oxygen. Therefore, the carboxyhemoglobin level in the banked blood in this study gradually decreased and the oxygen content increased (Table 1). The half-life of carboxyhemoglobin calculated from the volunteer study was approximately 47 days, which is long compared with the expected storage period of banked blood. Thus, it is likely that approximately one-third of all banked blood has abnormally high levels of carboxyhemoglobin. Methemoglobin levels increase in stored blood as storage is prolonged. I4 The level of methemoglobin in randomly collected banked blood had not been previously determined. The current data showed that the mean methemoglobin level in banked blood was 1.670, which is about three times higher than that of normal blood.‘” In the volunteer study, the initial mean methemoglobin was 1.3%, which also exceeded the normal value. However, the mean value of the same blood before preservation in the plastic bag with citrate-phosphate-dextrose solution was 0.5% -+ 0.1% (SD). This increase in the methemoglobin level immediately after preservation in the container has not been documented, and further investigation will be methemoglobin levels signifinecessary. Thereafter, cantly correlated with the length of storage and eventually increased to 2.5%. This increase resulted mainly from the inhibition of reduction enzymes due to the decrease in pH and adenosine-triphosphate levels in red blood cells during storage.‘” Dyshemoglobin levels of 13.8% could be attained by combining the highest values of carboxyhemoglobin and methemoglobin in this study. Met hemoglobin J. Glin. Anesth., vol. 2, March/April 1990

89

Original

Contributions

levels increased with storage. Carboxyhemoglobin levels decreased, but the half-life was long in comparison to the usual storage period. Therefore, with regard to available hemoglobin and its oxygen transport, the authors conclude that banked blood containing high levels of carboxyhemoglobin and methemoglobin may pose a potential risk when it is transfused to critically ill patients who have insufficient compensatory mechanisms. Since the in viuo half-life of carboxyhemoglobin when breathing air is approximately 4 hours,“’ smokers are advised to stop smoking at least 12 hours before donating blood to avoid high levels of carboxyhemoglobin in banked blood. Carbon monoxide is removed from blood containing relatively high carboxyhemoglobin levels in exchange for air. In order to reduce the high level of carboxyhemoglobin, it is recommended that the plastic container be made more permeable to air and that consideration be given to aerating preserved banked blood. In summary, one-third of the 312 units of banked blood had carboxyhemoglobin levels greater than 1.5%. Carboxyhemoglobin levels decreased as the storage period increased, in spite of its long half-life. The methemoglobin level in banked blood was about three times higher than that of normal blood, and the level increased during the storage period. If banked blood containing high levels of carboxyhemoglobin or methemoglobin is used in patients with severe cardiac and/or pulmonary disease, a critical reduction in hemoglobin available for oxygen transport will result.

Acknowledgments The authors wish to thank Dr. R. Hama for their advice in preparing this article.

and Dr. S. Naito

References 1. Aronow WS, O’Donohue WJ, Freygang J, Sketch Carboxyhemoglobin levels in banked blood. 1984;85:694-5.

90

J. Clin. Anesth.,

vol. 2, March/April

1990

MH: Chest

2. Stewart RD, Baretta ED, Platte LR, et al: Carboxyhemoglobin levels in American blood donors.JAMA 1974; 229: 1187-95. 3. Dennis RC:, Valeri (JR: Measuring percent oxygen saturation of hemoglobin, percent carboxyhemoglobin and methemoglobin, and concentration of total hemoglobin and oxygen in blood of man, dog, and baboon. C’lir~ Chem 1980;26: 1304-8. 4. Ioppolo C, Amiconi G, Currell DL, et al: Biochemical changes on storage of blood. Decrease in rate of methemoglobin reduction and increase in oxygen affinit) on storage of ACD blood. Vox Sang 1974;27:403-10. 5. Aronow WS, O’Donohue WJ: Carboxyhemoglobin levels in banked blood. Cheht 1985;87:409. 6. Spieb W, Weibhaar P, Vjdanovski B: KohlenmonoAN oxydbelastung bei Emf’angern von Blutkonserven. wsth Intensivthev Notfallmed 1980; 15:486-II?;. 7. Aronow WS: Aggravation of angina pectoris by 2% carboxyhemoglobin. RmJ Med 1977;63:904-8. 8. Middleman V, Poznak AV, Artusio JF, Smith SM: Garbon monoxide accumulation in closed circle anesthesia systems. Anesthesiology 1965;26:715-9. 9. Kandall SR, Landaw SA, Thalar MM: Carboxyhemoglobin exchange between donors and recipients of blood Pediatric.\ 1973:52:716-X. transfusions. 10. (Goldzimer EL, Parton ME, Kane EW: Elevated intraoperative carboxyhemoglobin levels. Anesth Anolg 1984;63:698. 1 1. Roughton FJW, Darling R