Evaluation of a Large-Scale Frozen Blood Program I. 0. SZYMANSKI A N D E. J. CARRINGTON From the University of Massachusetts Medical Center. Department of Laboratory Medicine, Worcester. Massachusetts and the Northeast Regional Red Cross Blood Program, Central Massachusetts Location, Worcester. Massachusetts.
The characteristicsof previously frozen red blood cells, prepared in a large-scale frozen blood program using the Red Cross method were evaluated. The use of the method as originally described resulted in approximately 91 per cent freeze-thaw-wash recovery of red blood cells. When the glycerolization step was modified by adding the partially glycerolized erythrocytes into 300 ml of 6.2M glycerol, freeze-thaw-wash recoveries were decreased. However, gradient addition of glycerol to the red blood cells without the use of stylet, resulted in acceptable in vino recoveries. Thawing frozen units in waterbath, to which no antiseptic was added, could introduce bacteria into units of previously frozen red blood cells. Therefore,it seems advisable to use dry heat thawing procedures. Previously frozen red blood cells prepared in the large scale maintained normal levels of ATP and 2,3 DPG. Therapeutic transfusions had acceptable 24-hour survival in vivo.
Methods
Blood Collection, Testing, and Storage From each volunteer donor approximately 450 ml of whole blood was collected into either 67.5 ml of acid-citrate-dextrose (ACD, N.I.H. formula A) or into 63 ml of citrate-phosphate-dextrose (CPD). The donor acceptability was determined with aid of the American National Red Cross Guidelines.2 The blood was processed according to the standards of the American National Red Cross. Blood group and Rh-factor of donor red blood cells were determined by Auto Analyzer,* and the serum was screened for the presence of atypical antibodies by saline, enzyme, and antiglobulin methods.’ Serology was determined in Auto Analyzer using ART reagent or by means of RPR card test,l5 and radioimmune assay was used SINCE HAYNES, TULLIS E T A L . ~first to detect hepatitis B surface antigen (HBsAg) in donor sera.I3 All units to be frozen were negative described a practical method to preserve for syphilis and HBsAg, but less than one per cent whole units of blood by freezing, several of the units to be frozen contained atypical blood different freezing methods have been intro- group antibodies. The red blood cells were stored duced and tested in clinical practice.8~’2.’’.18at 4 C for up to five days prior to freezing.
The success in these approaches has made possible red blood.cell freezing in large scale. A large scale red blood cell freezing program, initiated by the American National Red Cross was introduced to the Massachusetts Center in 1970.6 The processing of blood by freezing increased in this center each year to the present level, which is about five per cent of all bloods distributed to hospital blood banks. In our report w e are paying attention to the percentage of red blood cells recovered following processing, to other in vitro characteristics of the previously frozen erythrocytes, and to the survival in vivo.
Procedure of Glycerolization and Deglycerolization The units of whole blood were sedimented in an RC-3 refrigerated centrifuget a t 4,200 R P M either for three minutes to obtain platelet rich plasma (PRP) or for ten minutes to obtain plasma. All visible plasma was removed resulting in hematocrit of approximately 75 per cent after P R P separation or about 90 per cent after plasma separation. Prior to glycerolization, the red blood cells were kept at room temperature for at least thirty minutes. T h e glycerolization procedure of Meryman and Hornblower’2 is referred to as the Standard Method and was as follows: 400 ml of 6.2M glycerol containing 1.6 g/dl sodium lactate, and 30 mg/dl potassium chloride, buffered to pH
Received for publication June 28, 1976; accepted September 9, 1976.
*Technicon Corp., Tarrytown, NY. t h a n Sorval, Norwalk, CT.
43 1 Transfusion Sep1.-Ocl. 1911
Volumc 17 Numbcr 5
432
Transfusion ScpI.-Ocl. 1977
SZY MANSKl A N D CARRINGTON
6.8 with disodium phosphate were added to each unit of red blood cells. The glycerol was gravity fed through a special stylet to each unit of packed cells. The first 100 ml of glycerol was added rapidly at a rate of about 35 ml per minute while the red blood cells were agitated in a wrist action shaker1 at a speed of about nine cycles per second. After about five minutes the remaining glycerol was added to the red blood cells in the collection bag and the red blood cells and glycErol were mixed manually. Subsequently, the following variations in the method of glycerolization were introduced: Method I. The first I 0 0 ml of glycerol was added to the packed red blood cells without stylet at a rate of about 85 ml per minute while shaking the red blood cells in the wrist action shaker. The remaining glycerol was added as above. Method If. The first 100 ml ofglycerol was added as above. The partially glycerolized red blood cells were then added slowly to the remaining glycerol which had been introduced to the freezing bag. The glycerolized red blood cells were frozen slowly to -80 C and stored at that temperature for various periods of time. Thawing was done by immersing the freezing canister into 37 C waterbath. The thawed red blood cells were washed free from glycerol using solutions of NaCl as described before. I n most cases, a Haemonetics Blood Processor Model IS** was used, wash solutions being added by gravity feed. In some cases, an IBM Red Cell Processor No 2 9 9 l t t was used for deglycerolization. Bacteriological Studies A radiometric (Bactec)SS method was used to detect microorganisms in blood. This method m e a s u r e s t h e quantity of radioactive C O , g e n e r a t e d by t h e bacterial metabolism o f “C-glucose. The final segment of the deglycerolizing harness was isolated and approximately 2 ml of blood was obtained from the segment after cleaning the surface of the segment with 2 per cent iodine tincture and then with 70 per cent isopropyl alcohol. One ml of deglycerolized blood was added into each of two Bactec vials 6A, Tryptic Soy Broth, and 7A, Prereduced Tryptic Soy Broth. The vials were incubated at 37 C dry incubator for up to nine days. The release of “CO, was measured on or before the third day, usually the second measurement was done at the end of the incubation period. The quantity of “CO, was expressed as “growth index” from 0 to 100. When
$Burrell Corp., Pittsburgh, PA. **HaemoneticsCorp., Natick, MA. t t l B M , Princeton, NJ. $$Johnston Laboratories, Inc., Cockeysville, MD.
the growth index exceeded 30, the blood samples were considered to contain microorganisms.” However, the growth index of all positive cultures eventually reached 100. The bloods were recultured if another segment was available. Identity of the microorganisms was determined in Peter Bent Brigham Hospital Bacteriology Laboratory using standard bacteriological methods. Measurements in vitro of Previously Frozen Red Blood Cells The degree of hemolysis due to freezing and thawing was evaluated visually by comparing the intensity of red color in the initial effluent during deglycerolization to a set of colored tubes encased in plastic (hemoglobin Comparator).** Each tube, coded numerically from one to eight represented supernatant hemoglobin concentrations varying from 25 to 3,000 mg/dl. Recovery in vitro was calculated as follows:
% Freeze-thaw recovery Hb (Glyc) - ( I
=
Hct) x sup. Hb (Glyc.) H b (Glyc.) % Freeze-thaw-wash recovery Total cellular H b (Deglyc.) Total H b (Deglyc.) + Total H b (waste) Total cellular H b (Deglyc.) % Total recovery = Total H b (Glyc.) -
Hemoglobin concentration in the glycerolized and deglycerolized units and liquid waste was determined by cyanmethemoglobin m e t h ~ d . ~ Hematocrit was determined in triplicate by m i c r ~ m e t h o d .Supernatant ~ Hb was also determined by cyanmethemoglobin method. An electronic particle counter*** was used to determine the number of white blood cells in the units of deglycerolized blood.14 ATP and 2,3 DPG were determined with Ruorometric procedures on neutralized perchloric acid filtrates. io.Li Measurements in vivo The survival in vivo was performed by either radioactive technic3 or by automated differential agglutination method.I6 In the radioactive method, double label technique of Button et u I . ~ was used with a slight modification. A 10 ml aliquot of deglycerolized red blood cells, suspended to Hct of about 40 per cent and labeled with 20p Ci of “Cr (Chromitope)ttt was injected into the patient, together with about 2 p Ci of l Z 5 l (Albumitope).ttt The 24-hour survival in vivo was taken ***Coulter Electronics, Inc., Hialeah, FL.
t t t E . R. Squibb & Sons, Burlington, MA.
Volume 17 Number 5
433
FROZEN BLOOD
Table 2. Recovery of Erythrocytes In vitro Following Freeze Preservation of CPD Blood (Method 11).
Table 1. Recovery of Erythrocytes In vitro Following Freeze Preservation of Both ACD and CPD Blood (The Standard Method)
Total cellular hemoglobin before freezing, g Total cellular hemoglobin after processing. g Freeze-thaw recovery. % Freeze-thaw-wash recovery. % Total recovery. % Supernatant hemoglobin in the processed unit. mg/dl 24 hours after processing
ACD Blood (N = 77)
CPD Blood (N = ~ 6 2 )
M SD
63.3 6.8
65.6 8.6
M SD
53.6 8.3
54.8 7.1
M SD M SD M SD M SD
96.5 2.6 92.0 3.0 84.0 7 174.8 97
95.3 3.7 91 .o 3.6 84.0 6 216 93
Total cellular hemoglobin before freezing, g Total cellular hemoglobin after processing, g Freeze-thaw recovery. % Freeze-thaw-wash recovery, % Total recovery. %
Supernatant hemoglobin in the processed units, m u d l
M SD N M SD N M SD N M SD N M SD N M SD N
Haemonetics
IBM
64.95 6.75 38 50.94 6.06 38 96.05 3.5 38 84.5 5.5 38 78.5 6.16 38 164.7 56.1 38
64.48 8.51 21 51.90 8.1 21 94.55 5.4 8 84.5 8.0 19 79.8 8.76 20 145.8 83.4 21
Note: M = mean; SD = standard deviation; N = number of units tested.
Note: M = mean; SD = standard deviation; N = number of units tested.
as a measurement oi'the viability of the transfused red blood cells.
erolization by Method I 1 resulted in approximately 6.5 per cent lower freeze-thaw-wash recoveries than did glycerolization by the Standard Method Table 3 shows that excessive freezethaw-wash hemolysis was associated with Method I I but not with Method I. The addition of glycerol into partially glycerolized cells promoted good red blood cell recovery whereas, addition of partially glycerolized red blood cells to 6.2M glycerol caused excessive hemolysis. The special stylet did not seem to have a significant effect on the recoveries, and freezing random donor bloods with
Results The units of red blood cells to be glycerolized varied in weight between 230 and 320 g. Table I shows the effect of the Standard Method of Meryman and Hornblower'' on red blood cell recovery in vitro. The recoveries were similar regardless of. the type of anticoagulant used to collect whole blood. Table 2 shows that glycTable 3.
Recovery of Erythrocytes In vitro Following Freeze Preservation of CPD Blood
Method I
Total Cellular hemoglobin before freezing, g Total Cellular hemoglobin after processing, g Freeze-thaw recovery. % Freeze-thaw-wash recovery, % Total recovery, % Free hemoglobin in the processed unit. mg/dl
Method II
Pooled Units (N = 10)
Donor Units (N = 27)
Pooled Units ( N = 10)
M SD
51.3 1.5
64.9 7.1
50.3 1.4
M
SD
43.1 2.04
52.8 5.56
38.5 1.5
M SD M SD M SD M SD
98.7 0.11 89.0 2.4 84.1 3.0 97.3 24.6
97.4 2.1 5 88.9 6.95 81.8 6.74 122.8 34.8
99.0 0.14 79.6 3.38 76.7 4.0 135.0 38.8
Note: M = mean; SD = standard deviation: N = number of units tested
Transfusion 1977
SZYMANSKI AND CARRINGTON
y
+
: 2.23
Sep1:Ocl.
FIG. I . Relationship between freeze-thaw hemolysis and wash-hue code.
0.362~
0.738
r
N m 176
p
The characteristicsof previously frozen red blood cells, prepared in a large-scale frozen blood program using the Red Cross method were evaluated. The use of the method as originally described resulted in approximately 91 per cent freeze-thaw-wash recovery of red blood cells. When the glycerolization step was modified by adding the partially glycerolized erythrocytes into 300 ml of 6.2M glycerol, freeze-thaw-wash recoveries were decreased. However, gradient addition of glycerol to the red blood cells without the use of stylet, resulted in acceptable in vino recoveries. Thawing frozen units in waterbath, to which no antiseptic was added, could introduce bacteria into units of previously frozen red blood cells. Therefore,it seems advisable to use dry heat thawing procedures. Previously frozen red blood cells prepared in the large scale maintained normal levels of ATP and 2,3 DPG. Therapeutic transfusions had acceptable 24-hour survival in vivo.
Methods
Blood Collection, Testing, and Storage From each volunteer donor approximately 450 ml of whole blood was collected into either 67.5 ml of acid-citrate-dextrose (ACD, N.I.H. formula A) or into 63 ml of citrate-phosphate-dextrose (CPD). The donor acceptability was determined with aid of the American National Red Cross Guidelines.2 The blood was processed according to the standards of the American National Red Cross. Blood group and Rh-factor of donor red blood cells were determined by Auto Analyzer,* and the serum was screened for the presence of atypical antibodies by saline, enzyme, and antiglobulin methods.’ Serology was determined in Auto Analyzer using ART reagent or by means of RPR card test,l5 and radioimmune assay was used SINCE HAYNES, TULLIS E T A L . ~first to detect hepatitis B surface antigen (HBsAg) in donor sera.I3 All units to be frozen were negative described a practical method to preserve for syphilis and HBsAg, but less than one per cent whole units of blood by freezing, several of the units to be frozen contained atypical blood different freezing methods have been intro- group antibodies. The red blood cells were stored duced and tested in clinical practice.8~’2.’’.18at 4 C for up to five days prior to freezing.
The success in these approaches has made possible red blood.cell freezing in large scale. A large scale red blood cell freezing program, initiated by the American National Red Cross was introduced to the Massachusetts Center in 1970.6 The processing of blood by freezing increased in this center each year to the present level, which is about five per cent of all bloods distributed to hospital blood banks. In our report w e are paying attention to the percentage of red blood cells recovered following processing, to other in vitro characteristics of the previously frozen erythrocytes, and to the survival in vivo.
Procedure of Glycerolization and Deglycerolization The units of whole blood were sedimented in an RC-3 refrigerated centrifuget a t 4,200 R P M either for three minutes to obtain platelet rich plasma (PRP) or for ten minutes to obtain plasma. All visible plasma was removed resulting in hematocrit of approximately 75 per cent after P R P separation or about 90 per cent after plasma separation. Prior to glycerolization, the red blood cells were kept at room temperature for at least thirty minutes. T h e glycerolization procedure of Meryman and Hornblower’2 is referred to as the Standard Method and was as follows: 400 ml of 6.2M glycerol containing 1.6 g/dl sodium lactate, and 30 mg/dl potassium chloride, buffered to pH
Received for publication June 28, 1976; accepted September 9, 1976.
*Technicon Corp., Tarrytown, NY. t h a n Sorval, Norwalk, CT.
43 1 Transfusion Sep1.-Ocl. 1911
Volumc 17 Numbcr 5
432
Transfusion ScpI.-Ocl. 1977
SZY MANSKl A N D CARRINGTON
6.8 with disodium phosphate were added to each unit of red blood cells. The glycerol was gravity fed through a special stylet to each unit of packed cells. The first 100 ml of glycerol was added rapidly at a rate of about 35 ml per minute while the red blood cells were agitated in a wrist action shaker1 at a speed of about nine cycles per second. After about five minutes the remaining glycerol was added to the red blood cells in the collection bag and the red blood cells and glycErol were mixed manually. Subsequently, the following variations in the method of glycerolization were introduced: Method I. The first I 0 0 ml of glycerol was added to the packed red blood cells without stylet at a rate of about 85 ml per minute while shaking the red blood cells in the wrist action shaker. The remaining glycerol was added as above. Method If. The first 100 ml ofglycerol was added as above. The partially glycerolized red blood cells were then added slowly to the remaining glycerol which had been introduced to the freezing bag. The glycerolized red blood cells were frozen slowly to -80 C and stored at that temperature for various periods of time. Thawing was done by immersing the freezing canister into 37 C waterbath. The thawed red blood cells were washed free from glycerol using solutions of NaCl as described before. I n most cases, a Haemonetics Blood Processor Model IS** was used, wash solutions being added by gravity feed. In some cases, an IBM Red Cell Processor No 2 9 9 l t t was used for deglycerolization. Bacteriological Studies A radiometric (Bactec)SS method was used to detect microorganisms in blood. This method m e a s u r e s t h e quantity of radioactive C O , g e n e r a t e d by t h e bacterial metabolism o f “C-glucose. The final segment of the deglycerolizing harness was isolated and approximately 2 ml of blood was obtained from the segment after cleaning the surface of the segment with 2 per cent iodine tincture and then with 70 per cent isopropyl alcohol. One ml of deglycerolized blood was added into each of two Bactec vials 6A, Tryptic Soy Broth, and 7A, Prereduced Tryptic Soy Broth. The vials were incubated at 37 C dry incubator for up to nine days. The release of “CO, was measured on or before the third day, usually the second measurement was done at the end of the incubation period. The quantity of “CO, was expressed as “growth index” from 0 to 100. When
$Burrell Corp., Pittsburgh, PA. **HaemoneticsCorp., Natick, MA. t t l B M , Princeton, NJ. $$Johnston Laboratories, Inc., Cockeysville, MD.
the growth index exceeded 30, the blood samples were considered to contain microorganisms.” However, the growth index of all positive cultures eventually reached 100. The bloods were recultured if another segment was available. Identity of the microorganisms was determined in Peter Bent Brigham Hospital Bacteriology Laboratory using standard bacteriological methods. Measurements in vitro of Previously Frozen Red Blood Cells The degree of hemolysis due to freezing and thawing was evaluated visually by comparing the intensity of red color in the initial effluent during deglycerolization to a set of colored tubes encased in plastic (hemoglobin Comparator).** Each tube, coded numerically from one to eight represented supernatant hemoglobin concentrations varying from 25 to 3,000 mg/dl. Recovery in vitro was calculated as follows:
% Freeze-thaw recovery Hb (Glyc) - ( I
=
Hct) x sup. Hb (Glyc.) H b (Glyc.) % Freeze-thaw-wash recovery Total cellular H b (Deglyc.) Total H b (Deglyc.) + Total H b (waste) Total cellular H b (Deglyc.) % Total recovery = Total H b (Glyc.) -
Hemoglobin concentration in the glycerolized and deglycerolized units and liquid waste was determined by cyanmethemoglobin m e t h ~ d . ~ Hematocrit was determined in triplicate by m i c r ~ m e t h o d .Supernatant ~ Hb was also determined by cyanmethemoglobin method. An electronic particle counter*** was used to determine the number of white blood cells in the units of deglycerolized blood.14 ATP and 2,3 DPG were determined with Ruorometric procedures on neutralized perchloric acid filtrates. io.Li Measurements in vivo The survival in vivo was performed by either radioactive technic3 or by automated differential agglutination method.I6 In the radioactive method, double label technique of Button et u I . ~ was used with a slight modification. A 10 ml aliquot of deglycerolized red blood cells, suspended to Hct of about 40 per cent and labeled with 20p Ci of “Cr (Chromitope)ttt was injected into the patient, together with about 2 p Ci of l Z 5 l (Albumitope).ttt The 24-hour survival in vivo was taken ***Coulter Electronics, Inc., Hialeah, FL.
t t t E . R. Squibb & Sons, Burlington, MA.
Volume 17 Number 5
433
FROZEN BLOOD
Table 2. Recovery of Erythrocytes In vitro Following Freeze Preservation of CPD Blood (Method 11).
Table 1. Recovery of Erythrocytes In vitro Following Freeze Preservation of Both ACD and CPD Blood (The Standard Method)
Total cellular hemoglobin before freezing, g Total cellular hemoglobin after processing. g Freeze-thaw recovery. % Freeze-thaw-wash recovery. % Total recovery. % Supernatant hemoglobin in the processed unit. mg/dl 24 hours after processing
ACD Blood (N = 77)
CPD Blood (N = ~ 6 2 )
M SD
63.3 6.8
65.6 8.6
M SD
53.6 8.3
54.8 7.1
M SD M SD M SD M SD
96.5 2.6 92.0 3.0 84.0 7 174.8 97
95.3 3.7 91 .o 3.6 84.0 6 216 93
Total cellular hemoglobin before freezing, g Total cellular hemoglobin after processing, g Freeze-thaw recovery. % Freeze-thaw-wash recovery, % Total recovery. %
Supernatant hemoglobin in the processed units, m u d l
M SD N M SD N M SD N M SD N M SD N M SD N
Haemonetics
IBM
64.95 6.75 38 50.94 6.06 38 96.05 3.5 38 84.5 5.5 38 78.5 6.16 38 164.7 56.1 38
64.48 8.51 21 51.90 8.1 21 94.55 5.4 8 84.5 8.0 19 79.8 8.76 20 145.8 83.4 21
Note: M = mean; SD = standard deviation; N = number of units tested.
Note: M = mean; SD = standard deviation; N = number of units tested.
as a measurement oi'the viability of the transfused red blood cells.
erolization by Method I 1 resulted in approximately 6.5 per cent lower freeze-thaw-wash recoveries than did glycerolization by the Standard Method Table 3 shows that excessive freezethaw-wash hemolysis was associated with Method I I but not with Method I. The addition of glycerol into partially glycerolized cells promoted good red blood cell recovery whereas, addition of partially glycerolized red blood cells to 6.2M glycerol caused excessive hemolysis. The special stylet did not seem to have a significant effect on the recoveries, and freezing random donor bloods with
Results The units of red blood cells to be glycerolized varied in weight between 230 and 320 g. Table I shows the effect of the Standard Method of Meryman and Hornblower'' on red blood cell recovery in vitro. The recoveries were similar regardless of. the type of anticoagulant used to collect whole blood. Table 2 shows that glycTable 3.
Recovery of Erythrocytes In vitro Following Freeze Preservation of CPD Blood
Method I
Total Cellular hemoglobin before freezing, g Total Cellular hemoglobin after processing, g Freeze-thaw recovery. % Freeze-thaw-wash recovery, % Total recovery, % Free hemoglobin in the processed unit. mg/dl
Method II
Pooled Units (N = 10)
Donor Units (N = 27)
Pooled Units ( N = 10)
M SD
51.3 1.5
64.9 7.1
50.3 1.4
M
SD
43.1 2.04
52.8 5.56
38.5 1.5
M SD M SD M SD M SD
98.7 0.11 89.0 2.4 84.1 3.0 97.3 24.6
97.4 2.1 5 88.9 6.95 81.8 6.74 122.8 34.8
99.0 0.14 79.6 3.38 76.7 4.0 135.0 38.8
Note: M = mean; SD = standard deviation: N = number of units tested
Transfusion 1977
SZYMANSKI AND CARRINGTON
y
+
: 2.23
Sep1:Ocl.
FIG. I . Relationship between freeze-thaw hemolysis and wash-hue code.
0.362~
0.738
r
N m 176
p
