Chemical and Hematological Changes in Stored CPD Blood D. N. BAILEYAND J. R. BOVE From the Department of Laboratory Medicine. Yale University School of Medicine. New Haven. Connecticut

the storage lesion of CPD bank blood have been published. The most recent and complete report is that of Limbird and Silver,'O who evaluated sequential changes of various formed blood elements, chemical components, coagulation factors, and blood gases in CPD blood during storage at 4 C for 35 days. The published studies, besides being few in number, are limited to units drawn, processed, and stored especially for the research study. Sequential changes in CPD blood processed, shipped, and handled in routine fashion have not been reported. This study was undertaken to evaluate sequential changes in CPD blood collected and processed by a large community blood center and used in routine blood bank operations, and to compare these changes with those occurring in units drawn especially for the study from healthy volunteers and stored at 4 C for 28 days.

Blood was drawn from ten healthy volunteer donors into citrate-phosphatedextrose (CPD) anticoagulant and placed on the quarantine shelf of the blood bank refrigerator. Plasma dextrose, sodium, potassium, chloride, bicarbonate, GOT, LDH, and hemoglobin as well as WBC, hematocrit, MCV, MCHC, whole blood pH, and ammonia were measured on all samples initially and at one, two, seven, 14, 21, and 28 days of storage at 4 C. Whole blood lactate also was analyzed serially on five of the units. An additional 27 units of CPD bank blood (two to 21 days of age), routinely processed, handled, and stored by the blood bank, were submitted to the same analyses on the day of administration to the patient. Five of these processed units, 21 days old, were resampled at 28 days. Results of the analyses are presented and discussed. The most pronounced changes were seen for dextrose, potassium, bicarbonate, lactate, LDH, ammonia, and hemoglobin. Plasma dextrose and bicarbonate declined in concentration while potassium, lactate, LDH, ammonia, and hemoglobin rose with storage. In general, changes in the regularly processed, singly sampled bank units were greater than those observed in the specially processed, quarantined units sampled serially. This study indicates that routine transportation, processing, and handling of bank blood may lead to increased biochemical alteration.

IN 1957, Gibson and his associates' introduced a citrate-phosphate-dextrose (CPD) solution for the preservation of human blood. This solution is less acid than the standard acid-citrate-dextrose (ACD) solution, contains 20 per cent less citrate ion, exerts less alteration in the mean corpuscular hemoglobin concentration, and maintains a more physiological pH, thereby improving recovery and survival of the stored blood. Although CPD has now largely replaced ACD as the standard solution for the preservation of human blood, few studies of

Materials and Methods Blood Collection, Storage, and Sampling Srudy A: Blood was drawn from ten healthy volunteer donors according to the guidelines of the American National Red Cross Blood Program. Approximately 450 ml of blood was collected into each plastic bag containing 63 ml of citrate-phosphate-dextrose anticoagulant.* The blood was shipped approximately 30 miles by refrigerated truck and stored on the quarantine shelf of the blood bank refrigerator at 4 C. After thorough mixing, samples were withdrawn using aseptic technique into evacuated tubes through a medication-injection site (Fenwal, AE-9B), within 24 hours of collection and after one, two, seven, 14,21, and 28 days of storage.

Received for publication September 6, 1974; accepted October 7, 1974. This work was supported in part by USPHS Grant #STOI GM 00696-14.

'Fenwal Laboratories, Inc., Morton Grove, Ill.

244 Transfusion May-June 1975

Volume 15 Number 3

Volume I 5

245

C H A N G E S IN S T O R E D C P D BLOOD

Number I

Study B. An additional 27 units of C P D blood (two to 21 days of age), processed, transported, and stored in routine fashion, were sampled immediately before administration. Five of these unselected units, sampled at 21 days, were resampled on day 28. Chemical and hematological values were measured as indicated below. For comparison, dextrose, sodium, potassium, chloride, bicarbonate, pH, and ammonia were also measured on the C P D solution in one Fenwal bag.

analyzed by the microdiffusion method of Seligson and Hirahara." Whole blood lactate was analyzed on five of the study A units' by a modifiation of the method of H o h o r ~ t . ~ Hematological Measurements

Plasma hemoglobin was determined by a modification of the method of Crosby and F ~ r t h . ~ Leukocyte count (WBC), hematocrit (Htc), mean corpuscular volume (MCV), and mean corpuscular hemoglobin concentration (MCHC) were obtained with a Coulter Counter Model S.t

Chemical Measurements

Plasma dextrose was determined by an automated glucose oxidase procedure, plasma sodium and potassium by flame photometry, plasma chloride by an automated modification of the amperometric titration of Cotlove,2 and plasma bicarbonate by automated coulometric back titration. Whole blood pH was measured with a digital pH meter.** Plasma glutamicoxalacetic transaminase (GOT) and plasma lactic dehydrogenase (LDH) activity were assayed kinetically by a modification of the methods of Henry et. al." Whole blood ammonia was **Orion Research, Inc., Cambridge, Mass.

Results For each variable measured in the serially sampled units of Study A, the mean and standard deviation were calculated from the ten values obtained initially, and at one, two, seven, 14,21, and 28 days of storage (Table 1). The mean values plus and minus one standard deviation were plotted against storage time (Figs. 1 and 2). The 32 values for each variable in Study B were displayed by a superimposed scattergram in order to facilitate comparison of the data from the two studies. tCoulter Electronics, Inc., Hialeah, Fla.

Table 1. Chemical and Hematological Changes in CPD Blood with Storage Time (Study A ) Mean f 1 S.D. Day Plasma dextrose (mg1100 ml) Plasma sodium (Meqbiter) Plasma potassium (Meq1liter) Plasma chloride (Meqlliter). Plasma bicarbonate (Meqhiter) Whole-blood pH Whole-blood lactate* (mg1100 ml) Plasma GOT Plasma LDH Whole-blood ammonia (ug/lOO ml) Plasma hemoglobin

(mg/100 ml) WBC ( x 103) Hematocrit (%) MCHC MCV

0

1

2

7

14

21

28

345i21

332i9

323r15

312524

282i33

231i24

230t36

168t3

163f3

166f4

166f4

163f4

156f3

154i 3

3.9f0.2

5.4f0.4

6.6f0.6 11.9il.5

17.2f2.1

21.0i2.6

22.5f2.3

80i2

81 i 2

80f2

78 f 3

77 t 4

77* 3

10.8i2.3 6.9iO.O

10.2i4.5 6.9iO.O

79f3

18.3t1.4 7.1tO.O

15.0f1.5 14.3f1.6 15.350.9 7.1fO.O 7.1iO.O 7.0fO.O

11.6t1.6 7.0kO.O

41f10 26f12 2142 13

55flO 32i4 213t21

64210 1923 267248

101k18 22f4 636f 185

145+20 179f32 23 f 6 31 2 8 7 4 3 ~ 2 1 8 10492327

282 f 126

178 f 76

207 f 32

300 f 61

447 t 82

500 f 98

705

12.5f6.8 4.1 f 0.9 36.5 f 3.8 32.6 f 0.4 87i5

19.1i13.2 3.2 f 0.5 34.7 3.0

28.9i21.7 2.9 f 0.6 35.7 i 3.8 32.8 f 0.5 87 ,5

1.7f1.0 4.9 f 1.2 36.3 t 3.0 33.5 f 0.7 90f 5

*Data obtained from five units.

3.1f1.6 2.9t1.4 7.8i4.0 5.3 f 1.0 4 . 8 t 1.2 4.4 f 1.1 36.3 i 3.4 35.9 f 3.5 35.8 f 3.5 33.1 f 0.5 33.3f 0.3 33.1 f 0.4 91i5 9225 86f4

*

86 i 5

196i32 40 f 6 19502 1014

* 98

*

246

Transfusion

BAILEY A N D BOVE

May-June 1975

16

DAYS

-

0

10 DAYS

IC

DAYS

20

30

10 20 DAYS

30

0

10 20 DAYS

30

20

30

10 DAY

s

FIG. I . Changes in CPD blood with storage time: A) dextrose, B) sodium, C) potassium, D) chloride, E) bicarbonate, F) p H . The open circles indicate the mean values derived from ten units of CPD blood sampled serially. The shaded area indicates = t l S.D. from the mean. The closed circles show values for routinely processed C P D blood sampled immediately before administration to the patient.

The mean plasma dextrose fell from an initial 345 to 230 mg/100 ml at 28 days for Study A. The units in Study B also showed a drop in dextrose, and values were usually lower than those observed in Study A. The baseline dextrose of the undiluted C P D solution was 2,230 mg/100 ml . The plasma sodium dropped from an initial mean of 168 to 154 Meq/liter at day 28. The routinely processed units (Study B) showed similar values. The sodium content of the undiluted C P D solution was 280 Meq/liter. Mean plasma potassium rose from 3.9 to 22.5 Meq/liter over the 28 days of storage. Study B gave similar values, although occasionally higher values were seen. The concentration of potassium in the undiluted C P D was zero. Mean plasma chloride remained essentially constant in the range of 75 to 85 Meq/liter for

both Study A and Study B. No chloride was found in the undiluted C P D solution. The mean plasma biarbonate dropped from an initial 18.3 to 10.2 Meq/liter at 28 days for the specially handled units (Study A) while the routinely processed and stored units (Study B) showed somewhat greater variation and lower bicarbonate with storage time. The bicarbonate content of the C P D solution was zero. The mean pH of whole blood remained essentially unchanged (6.8 to 7.2) for both series of C P D bloods. The pH of the C P D solution obtained from one Fenwal bag was 5.7. The mean lactate level was measured in five units from Study A and increased from 41 to 196 mg/100 ml at 28 days. Plasma GOT varied widely over the 28 days for Study B while showing a modest rise with time in Study A (26 initially to 40 at 28 days).

Volume I S

247

CHANGES IN STORED C P D BLOOD

Number 3

28

O'

.

I0 ' 20 DAYS

'

30

-

0

10 20 DAYS

30

Or

I0 ' 20 DAYS

'

36

2E

mz

60.0

0 J

W

0

40.0

I w I

2 0.0

10

0 - 0 DAYS

20

30

0 - 0 DAYS

DAYS-

FIG. 2. Changes in C P D blood with storage time: A) lactate, B) GOT, C ) LDH, D) ammonia, E) plasma hemoglobin, F) WBC. The open circles indicate the mean values derived from ten units of C P D blood sampled serially. The shaded area indicates =t1 S.D. from the mean. The closed circles show values for routinely processed C P D blood sampled immediately before administration to the patient.

Mean plasma LDH rose from an initial value of 214 to 1950 at 28 days. The routinely handled units (Study B) showed a somewhat greater rise. The mean ammonia content of whole blood increased from an initial 282 to 705 ug/IOO ml at 28 days of storage. Study B followed the same pattern, but values were frequently higher. No ammonia was found in the undiluted C P D solution. The mean plasma hemoglobin varied from 1.7 initially to 28.9 rng/100 ml a t 28 days. The 32 units in Study B also showed a progressive rise in plasma hemoglobin, and values were generally higher in these units. The WBC showed a progressive drop to approximately 40 per cent of the initial value over the 28-day period for Study A while Study B revealed only a modest drop. The hematocrit value remained essentially

constant at about 36 per cent in both studies for the 28 days of storage. The MCHC remained basically unchanged (32 to 34) in both Study A and Study B although the 32 units from Study B showed somewhat higher values. The MCV remained constant at 86 to 90 for both studies although bloods in Study B had somewhat higher values.

Discussion

Since Gibson and his associates' introduced citrate-phosphate-dextrose (CPD) solution for the preservation of human blood in 1957, it has become the standard anticoagulant in most blood banks. Few ob-

248

BAILEY A N D BOVE

servations on the variation of chemical and hematological components of CPD blood have been reported. The few relatively complete studiesIohave not evaluated C P D blood that had undergone routine processing and handling prior to administration to the patient. Rather, these studies have been confined to units drawn from selected donors, placed on the quarantine shelf throughout the period of serial sampling, and not subjected to the trauma of transportation, handling, and multiple refrigerations usually characteristic for bank blood. The data (Table 1) from ten units drawn from healthy volunteer donors and held for serial sampling (Study A) are generally consistent with the findings of the few reported studies of C P D blood and the well-established patterns for ACD. The fall in plasma dextrose agreed with that found by Limbird and Silver'O and also paralleled the decrease reported with acid-citrate-dextrose blood.'" The decline in plasma sodium as sodium entered the erythrocyte was more profound than that noted by Schechter and Swan in CPD bloodI3 although the values agreed more closely with those observed in ACD blood.'" The rapid rise in plasma potassium was similar to that observed by others for both CPD and ACD I 3 . l 5 The maintenance of a stable plasma chloride was also reported by Limbird and Silver.'O The fall in plasma bicarbonate has not been reported, although a decreasing CO, content has been observed.'O As previously r e p ~ r t e d , " . ' ~ -pH ' ~ changed only slightly with storage, dropping 0.1 to 0.2 units. The rise in blood lactate seen in the present study was also documented by Schechter and Swan.I3 A rise in plasma GOT activity has not been noted previously, as Limbird and Silverlo studying C P D blood and Crowley4 studying ACD blood found little change in this enzyme with storage. The rise in plasma LDH has been documented by others in both CPD'O and ACD4 bank blood although the elevation observed by us was

Transfusion May-June 1975

not as profound as in other studies. Blood ammonia rose to 705 pg/lOO ml at 28 days in this study. These values paralleled those reported by Schechter and Swant3for C PD blood. However, Limbird and Silver'O noted a more exaggerated elevation in C P D blood ammonia to 1,900 pg/lOO ml at 28 days of storage, although the range of normal by the method they used was lower than ours. Plasma hemoglobin showed a rise to 28.9 mg/100 ml at 28 days. However, much greater increases, reflecting relatively more hemolysis, were reported by Schechter and SwanI3 (to approximately 100 mg/100 ml), Orlina and Josephson'2 (to 38 mg/100 ml), Limbird and Silverlo (to approximately 120 mg/100 ml), and Gibson et al." (to 40 mg/ 100 ml) at 28 days. The fall in WBC observed in this study was also documented by Limbird and Silver,'O who noted an 80 per cent drop by 35 days of storage. These same authors also noted the maintenance of a stable hematocrit (35 to 34 per cent), as found in the present study. In this study, a moderate difference in certain variables was noted between the units placed on the quarantine shelf for serial sampling (Study A) and those processed, stored, and handled in routine fashion (Study B). The most notable differences were plasma dextrose (Study B lower), bicarbonate (Study B lower), LDH (Study B higher), and plasma hemoglobin (Study B higher). In Study B, blood ammonia and plasma potassium were also frequently higher, but this was not a consistent finding. These findings suggest that routine processing, multiple refrigerations, and transportation of bank blood increase biochemical deterioration of the unit. An alternative, although somewhat less likely, explanation might be based upon the fact that C PD units from Study A were mixed periodically (with each serial sampling) while units from Study B were knowingly mixed only once (at the single time of sampling). Mixing of blood has been shown to enhance the preservation of 2,3-diphosphoglycerate,'

Volume 15 Number 1

CHANGES I N S T O R E D C P D BLOOD

and, in the case of CPD-anticoagulant, may lead to improved posttransfusion red blood cell s ~ r v i v a l . ~ Acknowledgments The authors gratefully acknowledge the assistance provided by Mrs. Rosanne Flaks, Miss Fay Weirich, Mr. Robert Bonvini, Mr. Sirlester Parker, Mrs. Fran Sarris, and the staff of the Clinical Laboratories and the Blood Transfusion Service of Yale-New Haven Hospital, New Haven, Conn.

References I.

2.

3.

4.

5.

6.

7.

Beutler, E.: The maintenance of red cell function during liquid storage. I n Progress in Transfusion and Transplantation, P. J. Schmidt, Ed. Washington, D.C., American Association of Blood Banks, 1972, p. 289. Cotlove, E.: Chloride. In Standard Methods of Clinical Chemistry (Vol. 3), D. Seligson, Ed. New York, Academic Press, 1961, p. 81. Crosby, W. H., and F. W. Furth: A modification of the benzidine method for measurement of hemoglobin in plasma and urine. Blood 11:380, 1956. Crowley, L. V.: Variations in levels of transaminases and lactic dehydrogenase in bank blood. Clin. Chem. 8:626, 1962. Dern, R. J., J. J. Wiorkowski, and T. Matsuda: Studies on the preservation of human blood. V. The effect of mixing anticoagulated blood during storage on the poststorage erythrocyte survival. J . Lab. Clin. Med. 75:37, 1970. Gibson, J. G., 11, C . B. Gregory, and L. N. Button: Citrate-phosphate-dextrose solution for preservation of human blood: a further report. Transfusion 1:280, 1961. -, S. B. Rees, T. J. McManus, and W. A. Scheitlin: A citrate-phosphate-dextrose solution for the preservation of human blood. Am. J . Clin. Pathol. 28569, 1957.

249

8. Henry, R. J., N. Chiamori, 0. J. Golub, and S. Berkman: Revised spectrophotometric methods for the determination of glutamicoxalacetic transaminase, glutamic-pyruvic transaminase, and lactic acid dehydrogenase. Am. J. Clin. Pathol. 34:381, 1960. 9. Hohorst, H. J.: L-(1)-Lactate determination with lactic dehydrogenase and DPN. In Methods of Enzymatic Analysis, H. U. Bergrneyer, Ed., New York, Academic Press, 1963, p. 266. 10. Limbird. T. J., and D. Silver: Sequential changes in blood preserved with citrate-phosphatedextrose. Surg. Gynecol. Obstet. 138:401, 1974. 11. McCullough, J., and B. J. Weiblen: Citrate phosphate dextrose (CPD) anticoagulant in blood transfusion. Minn. Med. 56980, 1973. 2. Orlina, A. R., and A. M. Josephson: Comparative viability of blood stored in ACD and CPD. Transfusion 9:62, 1969. 3. Schechter, D. C., and H. Swan: Biochemical alterations of preserved blood. Arch. Surg. 84:269, 1962. 4. Seligson, D., and K. Hirahara: The measurement of ammonia in whole blood, erythrocytes, and plasma. J. Lab. Clin. Med. 49:962, 1957. 5. Simon, G. E., and J. R. Bove: The potassium load from blood transfusion. Postgrad. Med. 49:61, 1971. 16. Technical Methods and Procedures, 5th ed. American Association of Blood Banks. Chicago, Twentieth Century Press, Inc., 1970. p. 196.

D. N. Bailey, Trainee, National Institutes of Health, Postdoctoral Fellow, Department of Laboratory Medicine, Yale University School of Medicine, New Haven, CT. 06510. J. R. Bove, Director, Blood Bank and Professor of Laboratory Medicine, Yale University, School of Medicine, New Haven, CT. 065 10.