Eur J Pediatr (1992) 151 : 851-854
European Journal of
Pediatrics
9 Springer-Verlag1992
Exchange transfusion: evaluating the use of a mixture of citrated red cells and heparinized plasma W. J. C.Valk 1, K . D . L i e m 1, and B . A . van Dijk 2 1Department of Paediatrics, 2Blood Transfusion Department, University Hospital Nijmegen, P.O. Box 9101, NL-6500 HB Nijmegen, The Netherlands Received March 3, 1991 / Accepted after revision February 28, 1992
Abstract. In 1987 a mixture for exchange transfusion was introduced in the Netherlands. It was composed of citrated red cells, heparinized plasma and third-party platelets if necessary. Selected biochemical and haematological properties of this mixture were compared to fresh heparinized whole blood, which was at that time the blood product of choice for exchange transfusion. The parameters of the mixture were less physiological than fresh heparinized whole blood. In addition, retrospective analysis of the same parameters was performed upon 149 blood samples from newborn infants who had undergone exchange transfusion. The bilirubin decreasing capacity of the mixture was adequate. Most other parameters did not change considerably and remained within the physiological range. This mixture may be an adequate product for exchange transfusion. However, to be certain of its safety and suitability, several other biochemical and haematological aspects must be studied in addition to the immunological and infectious risks. Key words: Exchange transfusion - Newborn infants Hyperbilirubinaemia - Component therapy
Introduction Exchange transfusion (ET) is an adequate haemotherapy for hyperbilirubinaemia, haemolytic disease of the newborn and in some cases of sepsis or disseminated intravascular coagulation. Initially, fresh heparinized whole blood with a restricted shelflife of 24 h at most was used [7]. Due to the introduction of citrate as an anticoagulant, the 24-h availability of donors and transfusionists is no longer necessary. The major disadvantage of citrated blood is its less physiological composition [21]. Several complications due to its acidity and concentration of Correspondence to: K. D. Liem Abbreviations: DIC = disseminated intravascular coagulation;
ET = Exchange transfusion(s)
sodium [8], potassium [23], calcium [15, 18] and glucose [22] have been documented in the literature. Despite these negative aspects, citrated blood is the blood product most often used for ETs throughout the world. In the Netherlands the use of fresh heparinized blood was preferred until 1987 because of its more physiological composition. Although not proven beyond doubt [1, 3, 16], beneficial anti-infectious [5, 6, 25] and coagulating properties were attributed to its higher leucocyte and platelet count [12]. Because of the risk of infectious diseases e.g. AIDS and hepatitis-B, the use of fresh heparinized blood is no longer recommended. A blood product which combines the positive facilities of both heparinized and citrated blood has been introduced. It is an A B O and rhesus (D) compatible mixture, composed of citrated red cells from one donor and heparinized fresh frozen plasma from another. If necessary, platelets can be added to create a platelet enriched mixture. Although this mixture has been described in the literature since 1988 [9, 14], only a few results about its use have been published [4, 19]. Consequently, this retrospective study was performed to accomplish to following aims: 1. To compare selected biochemical and haematological properties of the standard and platelet enriched mixture to those of fresh heparinized whole blood. 2. To investigate the influence of this mixture upon the bilirubin level as well as the same selected biochemical and haematological parameters in newborn blood following ET.
Materials and methods During the period from 1 November 1987 to 1 January 1990, 167 ETs were performed in 92 newborns in our hospital. Based upon the blood product used, three subgroups can be distinguished: Group 1 (n = 18): fresh heparinized whole blood Group 2 (n = 87): standard mixture Group 3 (n = 62): platelet enriched mixture Fresh heparinized whole blood was used in newborns with suspected sepsis or disseminated intravascular coagulation (DIC).
852
Table 1. Indications for ET (n = 167)
Performance of exchange transfusion
Group1 Hyperbilirubinaemia Haemolytic disease of the newborn Sepsis Disseminated intravascular coagulation
Group2
Group3
6
74
53
10
13 -
-
2
-
6
3
Sepsis, suspected by deterioration of clinical status and haematological evaluation (leucocytosis or leucocytopenia and/or increased number of band neutrophils in the differential white blood count) was confirmed by a blood culture. DIC was confirmed by increased fibrin degradation products, prolonged partial thromboplastin and prothrombin time and decreased platelet count. Both diagnoses were confirmed in 12 of 18 cases. If the ET was indicated due to hyperbilirubinaemia or haemolytic disease of the newborn, the new mixture was used. An ET was performed when the bilirubin level exceeded the exchange limit for the infant's birth weight category [10]. When the platelet count of the newborn was below 100 • 109/1, a platelet enriched mixture was used. This mixture was also used if the clinician expected platelet depletion due to earlier ETs (see Table 1).
Preparation of blood products Blood for donation was tested for hepatitis-B surface antigen and antibodies against syphilis and human immune deficiency virus. The serum of all blood donors was tested for irregular red cell antibodies. Citrated red cells were prepared from ABO-compatible donor blood, gathered in a 4-bag-system with a citrate sodiumphosphate and dextrose anticoagulant. The bully coat and plasma were transferred to satellite bags for further preparation. Red cells were stored for no longer than 72 h. For the preparation of heparinized plasma 500ml of ABrhesus(D) negative donor blood was collected in 1500 I.U. hepatin. After centrifugation, the plasma was stored at -30~ When plasma was needed a single donor portion was thawed in a water bath of 37~ The precipitate of fibronectin was centrifuged or filtered off. A single donor portion of platelets was produced out of the bully coat. If relatively fresh blood was used, both red cells and platelets were from the same donor. If not, a platelet concentrate from the regular blood bank stock, stored for no longer than 72h, was used [17], thereby introducing an additional donor. Both mixtures were made by mixing the single donor portions of red cells, plasma and if necessary platelets. To prepare fresh heparinized whole blood 500 ml of ABO compatible blood was collected in a bag containing 1500 I.U. of heparin.
ET was performed by means of the "push and pull" method exchanging a total blood volume corresponding to 200 ml/kg of body weight via a single lumen catheter in the umbilical vein. Upon completion of the ET, protamine-chloride (lmg/1000 I.U. of heparin) was administered to neutralize the heparin. No serious complications related to the procedure were observed. Minor complications included: hypothermia (n = 1); catheter occlusion (n = 6); blood leakage ( n = 1), bradycardia (n = 9 ) and unstable blood pressure (n = 4). These minor complications were easily treated.
Evaluation of exchange transfusion The three subgroups of donor blood used for ET were evaluated for the following biochemical and haematological properties: pH, sodium, potassium, calcium, magnesium, glucose, haemoglobin, haematocrit, leucocyte and platelet count. For these measurements samples were directly taken from the blood bag, immediately prior to the ET. To evaluate the effects of the standard mixture and the platelet enriched mixture on these parameters in the newborn, samples were taken from the first blood sample withdrawn (pre-ET value) and the last blood sample withdrawn (post-ET value). In addition to the parameters cited above bilirubin was measured as well. Changes in the parameters in Group 1 patients are not reported here, because the population is too small and the newborns were more severely affected (sepsis/DIC). In addition, mean pre-ET values deviate so much from those in groups 2 and 3, that a reliable comparison cannot be made. Demographics of newborns in groups 2 and 3 did not differ with respect to the parameters cited in Table 2 with the following exceptions: the number of hospital days and the sequential number of ET. The pre-ET values were not different, except for the platelet count (not presented). Statistical analysis was performed by means of a paired t-test and the Kruskal-Wallis test using SAS-statistics. A difference was assumed to be significant if P was < 0.05.
Results
Types of blood used for ET T h e t h r e e s u b g r o u p s d i f f e r e d for m o s t of t h e b i o c h e m i cal a n d h e m a t o l o g i c a l v a l u e s a s s e s s e d ( T a b l e 3). T h e most important differences were: (1) t h e p l a t e l e t c o u n t was v e r y l o w i n t h e s t a n d a r d mixture; (2) l e u c o c y t e c o u n t was d e c r e a s e d in b o t h m i x t u r e s ; (3) i n b o t h m i x t u r e s t h e p H a n d t h e c o n c e n t r a t i o n s of sodium and glucose were not within the physiological range;
Table 2. Exchange transfusion demographics. Values represent mean (range) Birth weight (g) Gestational age (days) Age at time of ET (h) Number of hospital days Number of death (not related to ET) Number of 2nd or later ETs (mean sequential number)
Group 2 (n = 87)
Group 3 (n = 62)
1717 (750-4000) 219 (172-286) 101 (3-952) 26 (2-139) 11
1435 (750-3120) 211 (172-264) 108 (23-236) 36 (2-154) 9
n.s. n.s. n.s. P < 0.05 n.s.
20 1.3
43 2.1
P < 0.05
853 Table 3. Selected biochemical and haematological values of the three types of blood products used for ET
Data
Group i
Group 2
Group 3
Haemoglobin (mmol/1) Haematocrit (1/l) Leucocytes (• 109/1) Platelet(xl09/l) pH Sodium (mmol/1) Potassium(mmol/1) Calcium (mmol/1) Magnesium (mmol/1) Glucose (mmol/1)
9.82 (1.34) 0.48 (0.06) 5.0 (1.4) 155 (54) 7.35 (0.03) 143 (3) 3.2 (0.3) 2.10 (0.12) 0.76 (0.07) 4.4 (1.9)
9.11 (1.35) 0.45 (0.07) 2.4 (1.8)*-** 41 (40)*'** 7.07 (0.07)* 151 (4)*' ** 3.8 (0.5)* 2.09 (0.14)** 0.78 (0.08) 13.4 (2.1)*.**
8.86 (1.56)* 0.45 (0.07) 3.7 (1.8)*'** 169 (65)** 7.06 (0.06)* 154 (3)*' ** 3.7 (0.5)* 1.99 (0.12)*'** 0.75 (0.06) 15.4 (2.9)*.**
Values represent: mean (S.D.) *P< 0.05, vs. group 1; **P< 0.05, group 2 vs. group 3 (Kruskal-Wallis test)
Table 4. Changes of several biochemical and haematological parameters in the blood of the newborn (post-ET value minus pre-ET value) Data
Group 2
Group 3
Haemoglobin (mmol/1) Haematocrit (1/i) Leucocytes (• 109/1) Platelet ( • 109/1) pH Sodium (mmol/l) Potassium (mmol/1) Calcium (mmoI/1) Magnesium (mmol/1) Glucose (mmol/1) Bilirubin (mmol/1) % Bilirubin (%)
-0.1 (1.5) -0.02 (0.08) -6.5 (9.1)* -164 (123)*,** -0.03 (0.06)* +2 (5)*' ** -0.4 (0.7)* +0.07 (o.18)* -0.04 (O.ll)* +1.0 (3.3) -128 (45)* -50.1 (10.0)*
-0.6 (1.4)* -0.03 (0.07) -7.0 (6.1)* +4 (76)** -0.04 (0.06)* +4 (5)*' ** -0.5 (0.6)* +0.08 (o.17)* -o.o3 (0.07)* -0.2 (4.8) -123 (41)* -51.8 (9.0)*
Values represent: mean (S.D.) *P < 0.05, post-ET vs. pre-ET value (paired t-test) **P < 0.05, group 2 vs. group 3 (Kruskal-Wallis test)
(4) calcium concentration was decreased in the platelet enriched mixture. The two mixtures differed in the platelet and leucocyte counts and the concentration of sodium, calcium and glucose.
Changes in selected parameters in newborn blood The changes in the biochemical and haematological parameters assessed in newborn blood were calculated for both mixtures and compared to the pre-ET value (Table 4). Statistical significance was seen for most values measured. The effect on bilirubin level was dependent upon the pre-ET value, therefore, the change is also reported as a percentage. The post-ET bilirubin level was approximately 50% of the p r e - E T value. The standard mixture differed in effect from the platelet enriched mixture only in platelet count and sodium concentration.
Discussion
Measured (biochemical and haematological) parameters in the mixture of citrated red cells and heparinized plasma deviated more from the physiological range than those in fresh heparinized whole blood. The anticoagulant, composed of Citrate, Phosphate-salt and Dextrose, elevates the acidity and the concentration of sodium and glucose of blood. Elevation of the potassium concentration is minimized by the use of blood stored for less than 72 h [20]. Leucocytes and platelets are decreased in number due to the preservation technique. Augmenting the mixture with additional platelets increases the platelet count to an adequate level. However, the addition of platelets also increases the amount and thus the side-effects of CPD. In addition, if a buffy coat was used for this purpose ( + 50%), leucocytes were introduced as well. The risks of cytomegalovirus infection and graft-versus-host reaction are consequently also increased. Using the mixtures for E T resulted in minor changes in mean values of several parameters. Although, changes were statistically significant, they were too small to have clinical relevance. Besides, they were usually directed in such a way that the post-ET values became more physiological (regression to the mean) and homeostasis was not disturbed. These results are in accordance with those of others [4, 19]. The bilirubin decreasing capacity of the mixtures was 50% which is similar to other blood products studied [26]. Comparing both mixtures, meets with the objection that the newborns in group 2 and 3 differ in the number of hospital days and in the sequential number of ETs. The first is a reflection of the degree of illness and potentially the vulnerability of electrolyte imbalances. However, except for sodium, which is ordinarily equilibrated quickly [24], no electrolyte differences were seen between the two groups. Therefore, we conclude that the two mixtures are comparable. A high sequential number of ETs might influence platelet and leucocyte-counts due to marrow exhaustion [13] or an altered maturation of multipotent progenitors [11]. These side-effects are, however, overcome by the administration of platelet concentrates which frequently
854 introduced leucocytes as well. Using this platelet enriched mixture, the m e a n platelet count was kept at the p r e - E T level. The critical p r e - E T value at which extra platelets should be a d d e d is still arbitrarily determined. A level of 100 x 109/1 is considered desirable. T h e r e are no reliable data in the literature about the influence of E T on leucocyte count. Despite these encouraging results, some precautionary notations should be m a d e regarding the use of the mixture of citrated red cells and heparinized plasma (and platelets) for ETs. In our study several specific p a r a m e t ers have b e e n reported. O t h e r parameters including acidbase-balance, fatty acids, amino acids, p h o s p h o r u s and insulin concentrations were not measured. F o r example, h y p o g l y c a e m i a due to reactive hyperinsulinaemia cannot be excluded. Additionally, no further coagulation studies were p e r f o r m e d . T h e r e f o r e the viability of administered platelets and their efficacy are u n k n o w n . The quantity of o t h e r coagulation factors is p r o p o s e d to be at an a d e q u a t e level [2]. A n o t h e r disadvantage of the mixture is the increased risk of infection due to additional d o n o r exposure, despite a d e q u a t e testing. F u r t h e r research on these matters is warranted. In our opinion, the risk of h a e m o r r h a g i c complications as a c o n s e q u e n c e of heparin administration is low. The a m o u n t of heparin in the mixture is minimal, and it is neutralized by protamine-chloride. N o h a e m o r r h a g i c complications were seen in the population studied.
In conclusion, the influnce of the mixture of citrated red cells and heparinized plasma (and platelets) on selected biochemical and hematological p a r a m e t e r s m e a s u r e d during E T is encouraging. Accordingly, the mixture described herein m a y be an a d e q u a t e p r o d u c t for ET. Further analysis is warranted to evaluate additional factors related to its suitabiltiy and safety for ET.
Acknowledgements. We thank Dr. L. Sanchez for her linguistic support and A. Theeuwes BSc. for his assistance in statistical analysis.
References 1. Bailey JE, Stork EK, Warkentin PI, Shurin SB (1987) Bully coat transfusion in neutropenic neonates with presumed sepsis: a prospective, randomized trial. Pediatrics 80 : 712-720 2. Barnard DR, Chapman RG, Simmons MA, Hathaway WE (1980) Blood for use in exchange transfusion in the newborn. Transfusion 20: 401-408 3. Bossi E, Meister B, Pfenninger J (1981) Exchange transfusion for severe neonatal septicemia (letter) Pediatrics 67 : 941 4. Brink MA, Derks MJM, Pietersz RNI, Reesink HW (1990) The use of a mixture of citrated erythrocytes and heparinized plasma for exchange transfusion (in Dutch). Ned Tijdschr Geneeskd 134 : 1856-1859 5. Cairo MG (1987) Granulocyte transfusion in neonates with presumed sepsis. Pediatrics 80 : 738-740 6. Courtney SE, Hall RT, Harris DJ (1979) Effect of blood transfusion on mortality and early onset Group B streptococcal septicaemia. Lancet II : 462-463
7. Diamond LK, Allen FH Jr, Thomas WO Jr (1951) Erythroblastosis fetalis. VII. Treatment with exchange transfusion. N EngI J Med 244: 39 8. Doyle PD, Eidelman AI, Lee K, Daum C, Gartner LM (1978) Exchange transfusion and hypernatremia. Possible role in intracranial hemorrhage in very-low-birth-weight infants. J Pediatr 92 : 848-849 9. Gaedicke G, Jonatha WD, Mueller-Eckhardt C (1988) P~diatrische Transfusionsmedizin. In: Mueller-Eckhardt C (ed) Transfusionsmedizin. Springer-Verlag, Berlin Heidelberg New York, pp 496-532 10. Gartner LM, Lee KS (1983) Unconjugated hyperbilirubinemia In: Fanaroff AA, Martin RJ (eds) Behrman's neonatal-perinatal medicine. CV Mosby, St. Louis, pp 754-771 11. Koenig JM, Christensen RD (1989) Neutropenia and thrombocytopenia in infants with Rh hemolytic disease. J Pediatr 114 : 625-631 12. Koppe JG (1985) Indications for exchange-transfusion in newborn infants (in Dutch) Ned Tijdschr Geneeskd 129:18171820 13. Kreuger A, Blomback M (1974) Exchange transfusion with frozen blood: effects on blood coagulation in the newborn. Haemostasis 3 : 329-339 14. Loos JA, Aken WG van (1988) Herstellung von Blutkomponenten. In: Mueller-Eckhardt C (ed) Transfusionsmedizin. Springer-Verlag, Berlin Heidelberg New York, pp 232-244 15. Maisels MJ, Ting-Kai Li BC, Piechocki JT, Werthman MW (1974) The effect of exchange transfusion on serum ionized calcium. Pediatrics 53 : 683-686 16. Pelet B (1979) Exchange transfusion in newborn infants: effects on granulocyte function. Arch Dis Child 54 : 687-690 17. Pietersz RNI, Loos JA, Reesink HW (1985) Platelet concentrates stored in plasma for 72 hours at 22~ prepared from buffycoats of C.P.D.-blood collected in a quadruple-bag S.A.G.M.system. Vox Sang 49 : 81-85 18. Radde IC, Parkinson DK, H6ffken B, Appiah KE, Hantey WB (1972) Calcium ion activity in the sick neonate: effect of bicarbonate administration and exchange transfusion. Pediatr Res 6 : 43-49 19. Samsom JF, Groenendijk MG, Lei J van der, Okken A (1991) Exchange transfusion in the neonate, a comparison between citrate-, heparinized- and reconstituted whole blood. (Letter) Eur J Hematol 47 : 152-154 20. Scanlon JW, Krakaur R (1980) Hyperkalemia following exchange transfusion. J Pediatr 96:108-110 21. Schiff D, Aranda JV, Chan G, Colle E, Stern L (1971) Metabolic effects of exchange transfusion. I. Effect of citrated and of heparinized blood on glucose, nonesterified fatty acids, 2- (4 hydroxybenzeneazo) benzoic acid binding and insulin. J Pediatr 78: 603-609 22. Schiff D, Aranda JV, Colle E, Stern L (1971) Metabolic effects of exchange transfusion. II. Delayed hypoglycaemia following exchange transfusion with citrated blood. J Pediatr 79:589593 23. Shortland D, Trounce JQ, Levene MI (1987) Hyperkalaemia, cardiac arrhythmias, and cerebral lesions in high risk neonates. Arch Dis Child 62:1139-1143 24. Tan KL, Tan IK (1975) Plasma potassium, sodium and chloride levels during and after exchange transfusion. Aust Paediatr J 11 : 165-168 25. Vain NE, Mazlumian JR, Swarner W, Cha CC (1980) Role of exchange transfusion in the treatment of severe septicaemia. Pediatrics 66 : 693-697 26. Zipursky A (1987) Isoimmune hemolytic disease. In: Nathan DG, Oski FA (eds) Hematology of infancy and childhood. WB Saunders, Philadelphia, pp 44-73
European Journal of
Pediatrics
9 Springer-Verlag1992
Exchange transfusion: evaluating the use of a mixture of citrated red cells and heparinized plasma W. J. C.Valk 1, K . D . L i e m 1, and B . A . van Dijk 2 1Department of Paediatrics, 2Blood Transfusion Department, University Hospital Nijmegen, P.O. Box 9101, NL-6500 HB Nijmegen, The Netherlands Received March 3, 1991 / Accepted after revision February 28, 1992
Abstract. In 1987 a mixture for exchange transfusion was introduced in the Netherlands. It was composed of citrated red cells, heparinized plasma and third-party platelets if necessary. Selected biochemical and haematological properties of this mixture were compared to fresh heparinized whole blood, which was at that time the blood product of choice for exchange transfusion. The parameters of the mixture were less physiological than fresh heparinized whole blood. In addition, retrospective analysis of the same parameters was performed upon 149 blood samples from newborn infants who had undergone exchange transfusion. The bilirubin decreasing capacity of the mixture was adequate. Most other parameters did not change considerably and remained within the physiological range. This mixture may be an adequate product for exchange transfusion. However, to be certain of its safety and suitability, several other biochemical and haematological aspects must be studied in addition to the immunological and infectious risks. Key words: Exchange transfusion - Newborn infants Hyperbilirubinaemia - Component therapy
Introduction Exchange transfusion (ET) is an adequate haemotherapy for hyperbilirubinaemia, haemolytic disease of the newborn and in some cases of sepsis or disseminated intravascular coagulation. Initially, fresh heparinized whole blood with a restricted shelflife of 24 h at most was used [7]. Due to the introduction of citrate as an anticoagulant, the 24-h availability of donors and transfusionists is no longer necessary. The major disadvantage of citrated blood is its less physiological composition [21]. Several complications due to its acidity and concentration of Correspondence to: K. D. Liem Abbreviations: DIC = disseminated intravascular coagulation;
ET = Exchange transfusion(s)
sodium [8], potassium [23], calcium [15, 18] and glucose [22] have been documented in the literature. Despite these negative aspects, citrated blood is the blood product most often used for ETs throughout the world. In the Netherlands the use of fresh heparinized blood was preferred until 1987 because of its more physiological composition. Although not proven beyond doubt [1, 3, 16], beneficial anti-infectious [5, 6, 25] and coagulating properties were attributed to its higher leucocyte and platelet count [12]. Because of the risk of infectious diseases e.g. AIDS and hepatitis-B, the use of fresh heparinized blood is no longer recommended. A blood product which combines the positive facilities of both heparinized and citrated blood has been introduced. It is an A B O and rhesus (D) compatible mixture, composed of citrated red cells from one donor and heparinized fresh frozen plasma from another. If necessary, platelets can be added to create a platelet enriched mixture. Although this mixture has been described in the literature since 1988 [9, 14], only a few results about its use have been published [4, 19]. Consequently, this retrospective study was performed to accomplish to following aims: 1. To compare selected biochemical and haematological properties of the standard and platelet enriched mixture to those of fresh heparinized whole blood. 2. To investigate the influence of this mixture upon the bilirubin level as well as the same selected biochemical and haematological parameters in newborn blood following ET.
Materials and methods During the period from 1 November 1987 to 1 January 1990, 167 ETs were performed in 92 newborns in our hospital. Based upon the blood product used, three subgroups can be distinguished: Group 1 (n = 18): fresh heparinized whole blood Group 2 (n = 87): standard mixture Group 3 (n = 62): platelet enriched mixture Fresh heparinized whole blood was used in newborns with suspected sepsis or disseminated intravascular coagulation (DIC).
852
Table 1. Indications for ET (n = 167)
Performance of exchange transfusion
Group1 Hyperbilirubinaemia Haemolytic disease of the newborn Sepsis Disseminated intravascular coagulation
Group2
Group3
6
74
53
10
13 -
-
2
-
6
3
Sepsis, suspected by deterioration of clinical status and haematological evaluation (leucocytosis or leucocytopenia and/or increased number of band neutrophils in the differential white blood count) was confirmed by a blood culture. DIC was confirmed by increased fibrin degradation products, prolonged partial thromboplastin and prothrombin time and decreased platelet count. Both diagnoses were confirmed in 12 of 18 cases. If the ET was indicated due to hyperbilirubinaemia or haemolytic disease of the newborn, the new mixture was used. An ET was performed when the bilirubin level exceeded the exchange limit for the infant's birth weight category [10]. When the platelet count of the newborn was below 100 • 109/1, a platelet enriched mixture was used. This mixture was also used if the clinician expected platelet depletion due to earlier ETs (see Table 1).
Preparation of blood products Blood for donation was tested for hepatitis-B surface antigen and antibodies against syphilis and human immune deficiency virus. The serum of all blood donors was tested for irregular red cell antibodies. Citrated red cells were prepared from ABO-compatible donor blood, gathered in a 4-bag-system with a citrate sodiumphosphate and dextrose anticoagulant. The bully coat and plasma were transferred to satellite bags for further preparation. Red cells were stored for no longer than 72 h. For the preparation of heparinized plasma 500ml of ABrhesus(D) negative donor blood was collected in 1500 I.U. hepatin. After centrifugation, the plasma was stored at -30~ When plasma was needed a single donor portion was thawed in a water bath of 37~ The precipitate of fibronectin was centrifuged or filtered off. A single donor portion of platelets was produced out of the bully coat. If relatively fresh blood was used, both red cells and platelets were from the same donor. If not, a platelet concentrate from the regular blood bank stock, stored for no longer than 72h, was used [17], thereby introducing an additional donor. Both mixtures were made by mixing the single donor portions of red cells, plasma and if necessary platelets. To prepare fresh heparinized whole blood 500 ml of ABO compatible blood was collected in a bag containing 1500 I.U. of heparin.
ET was performed by means of the "push and pull" method exchanging a total blood volume corresponding to 200 ml/kg of body weight via a single lumen catheter in the umbilical vein. Upon completion of the ET, protamine-chloride (lmg/1000 I.U. of heparin) was administered to neutralize the heparin. No serious complications related to the procedure were observed. Minor complications included: hypothermia (n = 1); catheter occlusion (n = 6); blood leakage ( n = 1), bradycardia (n = 9 ) and unstable blood pressure (n = 4). These minor complications were easily treated.
Evaluation of exchange transfusion The three subgroups of donor blood used for ET were evaluated for the following biochemical and haematological properties: pH, sodium, potassium, calcium, magnesium, glucose, haemoglobin, haematocrit, leucocyte and platelet count. For these measurements samples were directly taken from the blood bag, immediately prior to the ET. To evaluate the effects of the standard mixture and the platelet enriched mixture on these parameters in the newborn, samples were taken from the first blood sample withdrawn (pre-ET value) and the last blood sample withdrawn (post-ET value). In addition to the parameters cited above bilirubin was measured as well. Changes in the parameters in Group 1 patients are not reported here, because the population is too small and the newborns were more severely affected (sepsis/DIC). In addition, mean pre-ET values deviate so much from those in groups 2 and 3, that a reliable comparison cannot be made. Demographics of newborns in groups 2 and 3 did not differ with respect to the parameters cited in Table 2 with the following exceptions: the number of hospital days and the sequential number of ET. The pre-ET values were not different, except for the platelet count (not presented). Statistical analysis was performed by means of a paired t-test and the Kruskal-Wallis test using SAS-statistics. A difference was assumed to be significant if P was < 0.05.
Results
Types of blood used for ET T h e t h r e e s u b g r o u p s d i f f e r e d for m o s t of t h e b i o c h e m i cal a n d h e m a t o l o g i c a l v a l u e s a s s e s s e d ( T a b l e 3). T h e most important differences were: (1) t h e p l a t e l e t c o u n t was v e r y l o w i n t h e s t a n d a r d mixture; (2) l e u c o c y t e c o u n t was d e c r e a s e d in b o t h m i x t u r e s ; (3) i n b o t h m i x t u r e s t h e p H a n d t h e c o n c e n t r a t i o n s of sodium and glucose were not within the physiological range;
Table 2. Exchange transfusion demographics. Values represent mean (range) Birth weight (g) Gestational age (days) Age at time of ET (h) Number of hospital days Number of death (not related to ET) Number of 2nd or later ETs (mean sequential number)
Group 2 (n = 87)
Group 3 (n = 62)
1717 (750-4000) 219 (172-286) 101 (3-952) 26 (2-139) 11
1435 (750-3120) 211 (172-264) 108 (23-236) 36 (2-154) 9
n.s. n.s. n.s. P < 0.05 n.s.
20 1.3
43 2.1
P < 0.05
853 Table 3. Selected biochemical and haematological values of the three types of blood products used for ET
Data
Group i
Group 2
Group 3
Haemoglobin (mmol/1) Haematocrit (1/l) Leucocytes (• 109/1) Platelet(xl09/l) pH Sodium (mmol/1) Potassium(mmol/1) Calcium (mmol/1) Magnesium (mmol/1) Glucose (mmol/1)
9.82 (1.34) 0.48 (0.06) 5.0 (1.4) 155 (54) 7.35 (0.03) 143 (3) 3.2 (0.3) 2.10 (0.12) 0.76 (0.07) 4.4 (1.9)
9.11 (1.35) 0.45 (0.07) 2.4 (1.8)*-** 41 (40)*'** 7.07 (0.07)* 151 (4)*' ** 3.8 (0.5)* 2.09 (0.14)** 0.78 (0.08) 13.4 (2.1)*.**
8.86 (1.56)* 0.45 (0.07) 3.7 (1.8)*'** 169 (65)** 7.06 (0.06)* 154 (3)*' ** 3.7 (0.5)* 1.99 (0.12)*'** 0.75 (0.06) 15.4 (2.9)*.**
Values represent: mean (S.D.) *P< 0.05, vs. group 1; **P< 0.05, group 2 vs. group 3 (Kruskal-Wallis test)
Table 4. Changes of several biochemical and haematological parameters in the blood of the newborn (post-ET value minus pre-ET value) Data
Group 2
Group 3
Haemoglobin (mmol/1) Haematocrit (1/i) Leucocytes (• 109/1) Platelet ( • 109/1) pH Sodium (mmol/l) Potassium (mmol/1) Calcium (mmoI/1) Magnesium (mmol/1) Glucose (mmol/1) Bilirubin (mmol/1) % Bilirubin (%)
-0.1 (1.5) -0.02 (0.08) -6.5 (9.1)* -164 (123)*,** -0.03 (0.06)* +2 (5)*' ** -0.4 (0.7)* +0.07 (o.18)* -0.04 (O.ll)* +1.0 (3.3) -128 (45)* -50.1 (10.0)*
-0.6 (1.4)* -0.03 (0.07) -7.0 (6.1)* +4 (76)** -0.04 (0.06)* +4 (5)*' ** -0.5 (0.6)* +0.08 (o.17)* -o.o3 (0.07)* -0.2 (4.8) -123 (41)* -51.8 (9.0)*
Values represent: mean (S.D.) *P < 0.05, post-ET vs. pre-ET value (paired t-test) **P < 0.05, group 2 vs. group 3 (Kruskal-Wallis test)
(4) calcium concentration was decreased in the platelet enriched mixture. The two mixtures differed in the platelet and leucocyte counts and the concentration of sodium, calcium and glucose.
Changes in selected parameters in newborn blood The changes in the biochemical and haematological parameters assessed in newborn blood were calculated for both mixtures and compared to the pre-ET value (Table 4). Statistical significance was seen for most values measured. The effect on bilirubin level was dependent upon the pre-ET value, therefore, the change is also reported as a percentage. The post-ET bilirubin level was approximately 50% of the p r e - E T value. The standard mixture differed in effect from the platelet enriched mixture only in platelet count and sodium concentration.
Discussion
Measured (biochemical and haematological) parameters in the mixture of citrated red cells and heparinized plasma deviated more from the physiological range than those in fresh heparinized whole blood. The anticoagulant, composed of Citrate, Phosphate-salt and Dextrose, elevates the acidity and the concentration of sodium and glucose of blood. Elevation of the potassium concentration is minimized by the use of blood stored for less than 72 h [20]. Leucocytes and platelets are decreased in number due to the preservation technique. Augmenting the mixture with additional platelets increases the platelet count to an adequate level. However, the addition of platelets also increases the amount and thus the side-effects of CPD. In addition, if a buffy coat was used for this purpose ( + 50%), leucocytes were introduced as well. The risks of cytomegalovirus infection and graft-versus-host reaction are consequently also increased. Using the mixtures for E T resulted in minor changes in mean values of several parameters. Although, changes were statistically significant, they were too small to have clinical relevance. Besides, they were usually directed in such a way that the post-ET values became more physiological (regression to the mean) and homeostasis was not disturbed. These results are in accordance with those of others [4, 19]. The bilirubin decreasing capacity of the mixtures was 50% which is similar to other blood products studied [26]. Comparing both mixtures, meets with the objection that the newborns in group 2 and 3 differ in the number of hospital days and in the sequential number of ETs. The first is a reflection of the degree of illness and potentially the vulnerability of electrolyte imbalances. However, except for sodium, which is ordinarily equilibrated quickly [24], no electrolyte differences were seen between the two groups. Therefore, we conclude that the two mixtures are comparable. A high sequential number of ETs might influence platelet and leucocyte-counts due to marrow exhaustion [13] or an altered maturation of multipotent progenitors [11]. These side-effects are, however, overcome by the administration of platelet concentrates which frequently
854 introduced leucocytes as well. Using this platelet enriched mixture, the m e a n platelet count was kept at the p r e - E T level. The critical p r e - E T value at which extra platelets should be a d d e d is still arbitrarily determined. A level of 100 x 109/1 is considered desirable. T h e r e are no reliable data in the literature about the influence of E T on leucocyte count. Despite these encouraging results, some precautionary notations should be m a d e regarding the use of the mixture of citrated red cells and heparinized plasma (and platelets) for ETs. In our study several specific p a r a m e t ers have b e e n reported. O t h e r parameters including acidbase-balance, fatty acids, amino acids, p h o s p h o r u s and insulin concentrations were not measured. F o r example, h y p o g l y c a e m i a due to reactive hyperinsulinaemia cannot be excluded. Additionally, no further coagulation studies were p e r f o r m e d . T h e r e f o r e the viability of administered platelets and their efficacy are u n k n o w n . The quantity of o t h e r coagulation factors is p r o p o s e d to be at an a d e q u a t e level [2]. A n o t h e r disadvantage of the mixture is the increased risk of infection due to additional d o n o r exposure, despite a d e q u a t e testing. F u r t h e r research on these matters is warranted. In our opinion, the risk of h a e m o r r h a g i c complications as a c o n s e q u e n c e of heparin administration is low. The a m o u n t of heparin in the mixture is minimal, and it is neutralized by protamine-chloride. N o h a e m o r r h a g i c complications were seen in the population studied.
In conclusion, the influnce of the mixture of citrated red cells and heparinized plasma (and platelets) on selected biochemical and hematological p a r a m e t e r s m e a s u r e d during E T is encouraging. Accordingly, the mixture described herein m a y be an a d e q u a t e p r o d u c t for ET. Further analysis is warranted to evaluate additional factors related to its suitabiltiy and safety for ET.
Acknowledgements. We thank Dr. L. Sanchez for her linguistic support and A. Theeuwes BSc. for his assistance in statistical analysis.
References 1. Bailey JE, Stork EK, Warkentin PI, Shurin SB (1987) Bully coat transfusion in neutropenic neonates with presumed sepsis: a prospective, randomized trial. Pediatrics 80 : 712-720 2. Barnard DR, Chapman RG, Simmons MA, Hathaway WE (1980) Blood for use in exchange transfusion in the newborn. Transfusion 20: 401-408 3. Bossi E, Meister B, Pfenninger J (1981) Exchange transfusion for severe neonatal septicemia (letter) Pediatrics 67 : 941 4. Brink MA, Derks MJM, Pietersz RNI, Reesink HW (1990) The use of a mixture of citrated erythrocytes and heparinized plasma for exchange transfusion (in Dutch). Ned Tijdschr Geneeskd 134 : 1856-1859 5. Cairo MG (1987) Granulocyte transfusion in neonates with presumed sepsis. Pediatrics 80 : 738-740 6. Courtney SE, Hall RT, Harris DJ (1979) Effect of blood transfusion on mortality and early onset Group B streptococcal septicaemia. Lancet II : 462-463
7. Diamond LK, Allen FH Jr, Thomas WO Jr (1951) Erythroblastosis fetalis. VII. Treatment with exchange transfusion. N EngI J Med 244: 39 8. Doyle PD, Eidelman AI, Lee K, Daum C, Gartner LM (1978) Exchange transfusion and hypernatremia. Possible role in intracranial hemorrhage in very-low-birth-weight infants. J Pediatr 92 : 848-849 9. Gaedicke G, Jonatha WD, Mueller-Eckhardt C (1988) P~diatrische Transfusionsmedizin. In: Mueller-Eckhardt C (ed) Transfusionsmedizin. Springer-Verlag, Berlin Heidelberg New York, pp 496-532 10. Gartner LM, Lee KS (1983) Unconjugated hyperbilirubinemia In: Fanaroff AA, Martin RJ (eds) Behrman's neonatal-perinatal medicine. CV Mosby, St. Louis, pp 754-771 11. Koenig JM, Christensen RD (1989) Neutropenia and thrombocytopenia in infants with Rh hemolytic disease. J Pediatr 114 : 625-631 12. Koppe JG (1985) Indications for exchange-transfusion in newborn infants (in Dutch) Ned Tijdschr Geneeskd 129:18171820 13. Kreuger A, Blomback M (1974) Exchange transfusion with frozen blood: effects on blood coagulation in the newborn. Haemostasis 3 : 329-339 14. Loos JA, Aken WG van (1988) Herstellung von Blutkomponenten. In: Mueller-Eckhardt C (ed) Transfusionsmedizin. Springer-Verlag, Berlin Heidelberg New York, pp 232-244 15. Maisels MJ, Ting-Kai Li BC, Piechocki JT, Werthman MW (1974) The effect of exchange transfusion on serum ionized calcium. Pediatrics 53 : 683-686 16. Pelet B (1979) Exchange transfusion in newborn infants: effects on granulocyte function. Arch Dis Child 54 : 687-690 17. Pietersz RNI, Loos JA, Reesink HW (1985) Platelet concentrates stored in plasma for 72 hours at 22~ prepared from buffycoats of C.P.D.-blood collected in a quadruple-bag S.A.G.M.system. Vox Sang 49 : 81-85 18. Radde IC, Parkinson DK, H6ffken B, Appiah KE, Hantey WB (1972) Calcium ion activity in the sick neonate: effect of bicarbonate administration and exchange transfusion. Pediatr Res 6 : 43-49 19. Samsom JF, Groenendijk MG, Lei J van der, Okken A (1991) Exchange transfusion in the neonate, a comparison between citrate-, heparinized- and reconstituted whole blood. (Letter) Eur J Hematol 47 : 152-154 20. Scanlon JW, Krakaur R (1980) Hyperkalemia following exchange transfusion. J Pediatr 96:108-110 21. Schiff D, Aranda JV, Chan G, Colle E, Stern L (1971) Metabolic effects of exchange transfusion. I. Effect of citrated and of heparinized blood on glucose, nonesterified fatty acids, 2- (4 hydroxybenzeneazo) benzoic acid binding and insulin. J Pediatr 78: 603-609 22. Schiff D, Aranda JV, Colle E, Stern L (1971) Metabolic effects of exchange transfusion. II. Delayed hypoglycaemia following exchange transfusion with citrated blood. J Pediatr 79:589593 23. Shortland D, Trounce JQ, Levene MI (1987) Hyperkalaemia, cardiac arrhythmias, and cerebral lesions in high risk neonates. Arch Dis Child 62:1139-1143 24. Tan KL, Tan IK (1975) Plasma potassium, sodium and chloride levels during and after exchange transfusion. Aust Paediatr J 11 : 165-168 25. Vain NE, Mazlumian JR, Swarner W, Cha CC (1980) Role of exchange transfusion in the treatment of severe septicaemia. Pediatrics 66 : 693-697 26. Zipursky A (1987) Isoimmune hemolytic disease. In: Nathan DG, Oski FA (eds) Hematology of infancy and childhood. WB Saunders, Philadelphia, pp 44-73
