April 1976 TheJournalofPEDIATRICS
609
Variance in albumin loading in exchange transfusions To assess the rationale of albumin priming prior to exchange transfusions, 42 hyperbilirubinemic infants who required exchange transfusions were randomly assigned to one of two groups. Group I consisted of 15 infants who were given intravenously 1 g m / k g of salt-poor human serum albumin one hour before the exchanges. Group II, which consisted of 27 infants, received simple exchanges. No statistical differences were found in variations in reserve albumin-binding capacity, bilirubin, albumin, or red cell bilirubin at pre and one-hour post albumin infusion in the primed infants. The amount of bilirubin removed per kilogram is directly correlated to plasma bilirubin concentration (r = 0.87). No significant difference in efficiency on bilirubin removal was seen between the two groups. Beneficial effects of albumin therapy was apparent only in those infants with low R A B C as determined by the sephadex gel filtration technique.
George Chan, B.Sc., M.Se., and David Schiff, M.D., Ph.D., F.R.C.P.(C),* Edmonton, Alberta, Canada
ALBUMIN LOADING prior to exchange transfusions has
been used routinely in many newborn centers as an adjunct to the more eff• removal of bilirubin. The rationale for this approach is that the resultant increase in albumin concentration will enhance the bilirubin-albumin-binding capacity in the intravascular compartment, thus promoting diffusion of bilirubin from extravascular tissues into the circulation. The protective effect of human albumin in the prevention of bilirubin encephalopathy is well documented in experimental animals. '~ However, conclusive evidence as to its advantages is less clear-cut in human newborn infants. Odell and associates4 demonstrated improved efficiency in bilirubin removal with albumin priming; this finding has been contested by Comley and Wood? Using the HBABA dye [2-(4'hydroxybenzeneazo)benzoic acid] as an estimate of reserve albumin-binding capacity, G Wood and associates 7 found From the Departments of Pediatrics and Obstetrics & Gynecology, University of Alberta. Supported by grant No. 609 1023 209from the Department of National Health and Welfare, Ottawa, Canada. *Reprint address: Department of Pediatrics, University of Alberta, Edmonton, Alberta T5G 2G3, Canada.
an improvement both in albumin and binding capacity in infants receiving albumin infusion. Lucey and associates8 demonstrated a poor correlation between clinical kernicterus and reserve dye-binding capacity in jaundiced neonates. This observation was later confirmed in our laboratory by in vitro experiments, suggesting that the HBABA dye binding method is nonspecific for bilirubin or its competing anions, hence yielding unreliable results? Abbreviation used RABC: reserve albumin-binding capacity
The present study is designed to clarify some of the above findings utilizing a sephadex gel filtration method TM to measure reserve albumin-binding capacity and the variance in bilirubin, RABC, albumin, red cell bilirubin, and efficiency in bilirubin removal during exchange transfusions with albumin primed and nonalbumin primed hyperbilirubinemic infants. MATERIALS AND M E T H O D S Forty-two hyperbilirubinemic infants requiring exchange transfusions were randomly assigned to one of two
Vol. 88, No. 4, part 1, pp. 609-613
6 10
Chan and Schiff
The Journal of Pediatrics April 1976
Table I. C o m p a r a t i v e clinical data in albumin p r i m e d and n o n a l b u m i n p r i m e d infants
Table
IU. Bilirubin
and
albumin
concentrations in
a l b u m i n primed and n o n a l b u m i n primed exchange transfusions
Albumin primed Number Gestational age (wk) Mean Birth weight (kg) Mean Sex Diagnosis
Nonalbumin. primed
15 32-40
27 31-40
35.9 _+ 0.7 1.20 - 3.49 2.41 + 0.17 9M, 6 F 9 (Rh or ABO) 6 (others)
36.1 _+ 0.7 1.30 - 3.49 2.48 • 0.14 16M, 11 F 18 (Rh or ABO) 9 (others)
Table II. Variance in bilirubin, albumin, reserve album i n - b i n d i n g capacity, saturation point, m o l a r ratio and red blood cell bilirubin in pre a n d one hour post a l b u m i n h y p e r b i l i r u b i n e m i c plasma Pre
albumin Bilirubin (mg/dl) Mean • SEM Albumin (gm/dl) Mean • SEM RABC Mean • SEM Saturation point Mean • SEM Molar ratio Mean • SEM RBC-bilirubin Mean • SEM
10.3-28.5 16.7 • t.4 2.6-4.2 3.2 + 0.1 2.0-13.5 4.8 • 1.1 15.0-30.0 21.4 • 1.2 0.46-1.t0 0.82 • 0.5 0.73-1.9 1.33 _+ 0.15
1 hour post albumin 10.8-28.0 17.1 _+ 1.2 2.6-4.7 3.5 • 0.1 0.0-12.5 6.7 • 0.8 16.5-30.5 23.7 _+ 1.0 0.60-1.07 0.82 • 0.04 0.47-1.8 0.96 + 0.16
p
Albumin primed Pre exchange Bilirubin (mg/dl) Mean • SEM Albumin (gm/dl) Mean _+ SEM Post exchange Bilirubin (mg/dl) Mean • SEM Albumin (gm/dl) Mean _+ SEM 4 hours post exchange Bilirubin (mg/dl) Mean _+ SEM Albumin (gm/dl) Mean • SEM
Nonalbumin primed
10.8-29 16.5 _+ 1.3 2.7-4.7 3.5 _+ 0.2
10.5-28.5 14.8 • 0.9 2.2-3.9 3.2 • 0.1
4.0-10.0 6.7 • 0.5 2.6-3.9 3.3 _+ 0.1
3.5-11.5 7.3 +_ 0.4 2.0-3.8 3.2 • 0.1
6.5-17.0 12.1 • 1.2 2.5-3.6 3.1 • 0.1
7.5-17.5 11.6 _+ 0.6 2,4-3.8 3.1 +_ 0.1
p
> 0.3 > 0.1
> 0.3 > 0.5
> 0.5 n.s.
Table IV. Efficiency o f bilirubin removal in a l b u m i n a n d n o n a l b u m i n p r i m e d exchange transfusions
> 0.5 > 0.1 < 0.2 > 0.2 > 0.5 < 0.1
groups. Group I consisted o f 15 infants who were given 1 g m / k g o f salt-poor h u m a n a l b u m i n intravenously one hour prior to the exchange. Group II was c o m p o s e d o f 27
Albumin primed Volume (ml) 200-500 Mean • SEM 437 _+ 26 Hematocrit (%) 39-50 Mean + SEM 41.1 • 0.9 Bilirubin removed mg/bag Red blood cell 0.6-3.9 bilirubin Mean + SEM 1.6 • 0.3 Plasma bilirubin 7.9-32~8 Mean • SEM 24.3 _+ 2.2 Total bilirubin 8.5-42 Mean + SEM 26.0 _+ 2.5 Bilirubin/kg 5.6-14.5 Mean + SEM 10.8 + 0.7
Nonalbumin primed 100-500 402 • 22
> 0.3
36-55 41.6 • 0.9
> 0.5
0.3-4.7 1.2 _+ 0.2 4.7-40.6 22.2 • 2.0 5.4-51.1 23.4 _+ 2.1 4.2-18.6 8.9 • 0.6
> 0.2 > 0.3 > 0.3 < 0.3
infants who were given simple exchange transfusions. All babies received a two volume exchange with A C D b l o o d less than 48 hours old. T h e clinical data outlined in T a b l e I shows that the two groups are c o m p a r a b l y m a t c h e d with
exchange samples. In addition, all the blood r e m o v e d during the exchange transfusion was saved in a bag
regards to gestational age, birth weight, sex, a n d etiology
containing 1,500 units o f h e p a r i n . The bags were analyzed
o f the hyperbilirubinemia.
for volume, hematocrit, plasma bilirubin, and red b l o o d
Blood samples from infants in group I were d r a w n into heparinized tubes before receiving a l b u m i n a n d one h o u r
cell bilirubin. The total mass o f bilirubin r e m o v e d is calculated as:
after completion o f the a l b u m i n infusion. A bilirubin profile which included bilirubin, 11 albumin, ~2 R A B C by
1-hematocrit 100 x volume of exchanged blood x plasma bilirubin +
the sephadex gel filtration m e t h o d , 1~ a n d the erythrocytes for red cell bilirubin 14 was o b t a i n e d o n the specimens. A
heinatocrit 9. x volume of exchanged blood • red cell bilirubin. 100
bilirubin profile was also obtained o n both groups o f
Saturation point is defined as the total c o n c e n t r a t i o n o f
infants at pre, post exchange, and four hours post
bilirubin required to saturate all the p r i m a r y a l b u m i n
Volume 88 Number 4, part 1
Variance in albumin loading in exchange transfusions
6 11
-- 30' E c~ E Z m
~, 20 84 e~ B
.-1 m
< 10 D.
BILIRUBtN REMOVED mg/kg Fig. I. Correlation between plasma bilirubinconcentration and amount of bilirubin removed per kilogram body weight in both groups of infants receiving two volume exchange transfusions, r = 0.87; y = 2.14 + 1.32 x.
binding sites for bilirubin and is the sum of RABC and plasma bilirubin. Negative values denote oversaturation of primary binding sites. Molar ratios are calculated from the saturation points and albumin concentrations, using molecular weights of 585 for bilirubin and 69,000 for albumin. Variance analysis of unpaired t tests and regression coefficients were analyzed using an Olivetti-Underwood Programmer 101. RESULTS Effect of albumin priming. The change in bilirubin, albumin, RABC, saturation point, bilirubin-albumin molar ratio, and red cell bilirubin pre and one-hour post albumin infusion are shown in Table II. The mean increase in albumin concentration (0.3/dl) is associated with a slight elevation in bilirubin (0.4 mg/dl), an improvement in RABC (1.9 mg/dl) and in saturation point (1.3 mg/dl), and a decrease in red cell bilirubin (-0.3 mg/dl). The variance is not statistically significant in any of these determinations. Comparison of albumin and bilirubin concentrations in group I and II infants. The concentration of bilirubin and
albumin in groups I and II are shown in Table III. Analyses of either concentration show no statistical difference between the two groups at pre, post, or four-hour post exchange levels. Efficiency of exchange transfusions with albumin. Bilirubin removal data on the efficiency of bilirubin removal in groups I and II infants are shown in Table IV. No significant difference was obtained in any of the features studied. There is a positive correlation of the regression
line between the plasma bilirubin and the amount of bilirubin removed per kilogram body weight in both groups (Fig. 1). DISCUSSION The data presented showed no significant advantage in albumin priming in the group of infants studied. This seems to contradict earlier reports that albumin administration causes an increase in both serum bilirubin and albumin levels in artificially jaundiced rats. ~, a These experimental animals were loaded with bilirubin far in excess of the binding capacity of the albumin present. By contrast in the present study, the mean RABC of 4.8 _ 1.1 in our group of infants indicate the presence of adequate reserve binding sites prior to albumin therapy. The mean net increase in albumin of 0.3 gm/dl is in agreement with that reported by Odell and associates 4 of 0.4 g m / d l using the same albumin-priming procedure. However, the changes in bilirubin concentration one hour post albumin infusion were less predictable. Six of the 15 babies had a slight decrease with a mean net increase of ordy 0.4 mg/dl. There is a poor correlation of the changes in bilirubin concentration with RABC (r = 0,45), suggesting that the bilirubin increment is the net result of volume expansion caused by the infusion of albumin and the formation of new bilirubin rather than from an intracellular source. In two infants in whom the RABC was low (--2 and 0), indicating complete or almost complete saturation of ,the primary binding site on albumin, the beneficial effects of albumin priming is apparent. The RABC improved to 4
6 12
Chan and Schiff
and 7, respectively, while the bilirubin levels were increased by 2.5 and 4.5 mg/dl. Improvement in RABC reduced the fraction of diffusable bilirubin which if allowed to reach high levels may pose a threat to cellular toxicity. In fact. the high increment in bilirubin after albumin infusion in these infants may be an indication of redistribution Of the bilirubin pool; the direction of movement being from tissues to the bloodstream. It has been shown that red blood cell lipids bind bilirubin..... and that this property seemed to be influenced by the degree of saturation of the albumin molecule with bilirubin.1~ On this basis. Bratlid 17 suggested that erythrocyte bilirubin concentration may reflect the relative saturation of albumin with bilirubin: the range in our group of infants (n = 1l) is of the same magnitude as that reported by Bratlid. 17Administration of albumin decreases the eluable red cell bilirubin in all cases studied. This is not surPrising since albumin has a higher affinity for bilirubin than red celt membrane lipids with a resultant net efflux of bilirubin into the plasma as a new bilirubin albumin dynamic equilibrium is achieved. However. clinical interpretation of red cell bilirubin values is still in doubt. Although conclusions cannot be drawn because of our small sample size. preliminary data suggest that there is a fairly wide scatter of values (0.5 1.9 mg/dl) and seems to be poorly correlated with plasma bilirubin (r = 0.375), albumin (r = 0.320), and RABC (r = 0.668). The present findings with regards to efficiency of bilirubin removed is in agreement with that reported by Cromley and Wood2 No statistical difference in the amount of bilirubin removed per kilogram body weight was found in the two groups. In support of this observation, the albumin and bilirubin concentrations were comparable at pre, post, and four-hour post exchange levels (Table III). Erythrocyte-bound bilirubin was found to constitute 5% to 10% of plasma bilirubin in the exchanged blood and was therefore included as part of the total bilirubin removed. Our mean average of ( 8 . 9 - 10.8) mg/kg of total biiirubin removed in two volume exchanges is less impressive than those Obtained by Odell and associates~ and Comley and Wood 5 (12.3 and 14, respectively). The lower efficiency can be explained at least in part by the lower mean pre-exchange bilirubin concentrations in our group of infants. To further support this point, regression coefficients between the amount of bilirubin removed per kilogram body weight and plasma bilirubin levels were analyzed. Apositive correlation was obtained in these parameters in both groups of infants (Fig. 1). Thus it would appear that the amount of bilirubin removed bears a direct relationship to plasma bilirubin
The Journal of Pediatrics April 1976
concentration. In view of the small increment in bilirubin increase induced by albumin priming in the group of infants studied, it is not entirely unexpected that no improvement in efficiency in bilirubin removal was observed. It would seem therefore that the use of albumin as a priming agent does not improve the efficiency of removal of bilirubin during an exchange. In most instances the giving of albumin in the pre-exchange period does not significantly improve the reserve albuminTbinding capacity. This may be a reflection of the fact that these infants already have adequate reserve albumin-binding capacity and the addition of albumin did not change this situation. However, in two infants who had almost complete saturation of the albumin with endogenous bilirubin, the giving of the albumin seemed to improve the situation in a temporary fashion prior to the exchange transfusion. It would seem that the use of albumin, although safe, would o~ly have significant effect in those situations where the binding capacity is already compromised, and where there may be some delay, in achieving the exchange transfusion. It is recommended that reserve albuminbinding capacity would be a useful tool in the monitoring of infants prior to exchange transfusion, with regard to the timing of the exchange transfusion and also to the need of using albumin as a priming agent prior to the exchange transfusion. We thank Mrs. K. Merrills for technica! assistance, and the nursing staff of the neonatal intensive care unit, University of Alberta Hospital, for assistance with exchange transfusions. REFERENCES
1. BowenWR, Porter E, and Waters WJ: The protective action of albumin in bilirubin toxicity in newborn puppies, Am J Dis Child 98:568, 1959 (abstr). 2. SchmidR, Diamond I, Hammakar L, and Gundersen CB: Interaction of bilirubin with albumin, Nature 206:1041, t965. 3. Diamond I, and Schmid R: Experimental bilirubin encephalopathy. The mode of entry of bilirubin-C~ into the central nervous system, J Clin Invest 45:678, 1966. 4. Odell GB, Cohen SN, and Gordes EH: Administration of albumin in the management of hyperbilirubinemia by exchange transfusions, Pediatrics 30:613, 1962. 5. Comley A, and Wood B: Albumin administration in exchange transfusion for hyperbilirubinemia, Arch Dis Child 43:151, 1968. 6. Porter EG, and Waters WJ: A rapid micromethod for measuring the reserve albumin binding capacity in serum from newborn infants with hyperbilirubinemia, J Lab Clin Med 67:660, 1966. 7. Wood B, Comley A, and Sherwell J: Effect of additional albumin administration during exchange transfusion on plasma albumin binding capacity, Arch Dis Child 45:59, 1970.
Volume 88 Number 4, part 1
8. Lucey JF, Valaes T, and Doxiadis SA: Serum albumin reserve PSP dye binding capacity in infants with kernicterus, Pediatrics 39:876, 1967. 9. Chart G, Schiff D, and Stern L: Competitive binding of free fatty acids and bilirubin to albumin: differences in HBABA dye And Sephadex G-25 interpretation of results, Clin Biochem 4:205, 1971. 10. Schiff D, Chart G, and Stern L: Sephadex G-25 quantitative estimation of free bilimbin potential in jaundiced newborn infants' sera: a guide to the prevention of kernicterus, J Lab Clin Med 80:455, 1972. 11. Martinek RG: Improved micromethod for determination of serum bilirubin, Clin Chim Acta 13:161, 1966. 12. Gomall AG, Bardonwill CS, and David MM: Determination of serum proteins by the use of the biuret reaction, Biol Chem 177:751, 1949. Thin gel electrophoresis in agarose, Anal Inc Cat No 1-1000 (96) 1966.
Variance in albumin loading in exchange transfusions
6 13
13. Chan G, and Schiff D: Further clinical evaluation of the sephadex G-25 elution technique in the management of the hyperbilirubinemic infant, International Bilirubin Symposium, May 1974, Jerusalem, Israel. Birth Defect Series. In Press. 14. Bratlid D: Bilirubin binding of human erythrocytes, Scand J Clin Lab Invest 29:91, t972. 15. Watson D: The absorption of bilirubin by erythrocytes, Clin Claim Acta 7:733, 1962. 16. Oski FA, and Naiman JL: Red cell binding of bilirubin, J PEDIATR63:1034, 1963. 17. Bratlid D: Reserve albumin binding capacity salicylate saturation index, and red cell binding of bilirubin in neonatal jaundice, Arch Dis Child 48:393, 1973.
609
Variance in albumin loading in exchange transfusions To assess the rationale of albumin priming prior to exchange transfusions, 42 hyperbilirubinemic infants who required exchange transfusions were randomly assigned to one of two groups. Group I consisted of 15 infants who were given intravenously 1 g m / k g of salt-poor human serum albumin one hour before the exchanges. Group II, which consisted of 27 infants, received simple exchanges. No statistical differences were found in variations in reserve albumin-binding capacity, bilirubin, albumin, or red cell bilirubin at pre and one-hour post albumin infusion in the primed infants. The amount of bilirubin removed per kilogram is directly correlated to plasma bilirubin concentration (r = 0.87). No significant difference in efficiency on bilirubin removal was seen between the two groups. Beneficial effects of albumin therapy was apparent only in those infants with low R A B C as determined by the sephadex gel filtration technique.
George Chan, B.Sc., M.Se., and David Schiff, M.D., Ph.D., F.R.C.P.(C),* Edmonton, Alberta, Canada
ALBUMIN LOADING prior to exchange transfusions has
been used routinely in many newborn centers as an adjunct to the more eff• removal of bilirubin. The rationale for this approach is that the resultant increase in albumin concentration will enhance the bilirubin-albumin-binding capacity in the intravascular compartment, thus promoting diffusion of bilirubin from extravascular tissues into the circulation. The protective effect of human albumin in the prevention of bilirubin encephalopathy is well documented in experimental animals. '~ However, conclusive evidence as to its advantages is less clear-cut in human newborn infants. Odell and associates4 demonstrated improved efficiency in bilirubin removal with albumin priming; this finding has been contested by Comley and Wood? Using the HBABA dye [2-(4'hydroxybenzeneazo)benzoic acid] as an estimate of reserve albumin-binding capacity, G Wood and associates 7 found From the Departments of Pediatrics and Obstetrics & Gynecology, University of Alberta. Supported by grant No. 609 1023 209from the Department of National Health and Welfare, Ottawa, Canada. *Reprint address: Department of Pediatrics, University of Alberta, Edmonton, Alberta T5G 2G3, Canada.
an improvement both in albumin and binding capacity in infants receiving albumin infusion. Lucey and associates8 demonstrated a poor correlation between clinical kernicterus and reserve dye-binding capacity in jaundiced neonates. This observation was later confirmed in our laboratory by in vitro experiments, suggesting that the HBABA dye binding method is nonspecific for bilirubin or its competing anions, hence yielding unreliable results? Abbreviation used RABC: reserve albumin-binding capacity
The present study is designed to clarify some of the above findings utilizing a sephadex gel filtration method TM to measure reserve albumin-binding capacity and the variance in bilirubin, RABC, albumin, red cell bilirubin, and efficiency in bilirubin removal during exchange transfusions with albumin primed and nonalbumin primed hyperbilirubinemic infants. MATERIALS AND M E T H O D S Forty-two hyperbilirubinemic infants requiring exchange transfusions were randomly assigned to one of two
Vol. 88, No. 4, part 1, pp. 609-613
6 10
Chan and Schiff
The Journal of Pediatrics April 1976
Table I. C o m p a r a t i v e clinical data in albumin p r i m e d and n o n a l b u m i n p r i m e d infants
Table
IU. Bilirubin
and
albumin
concentrations in
a l b u m i n primed and n o n a l b u m i n primed exchange transfusions
Albumin primed Number Gestational age (wk) Mean Birth weight (kg) Mean Sex Diagnosis
Nonalbumin. primed
15 32-40
27 31-40
35.9 _+ 0.7 1.20 - 3.49 2.41 + 0.17 9M, 6 F 9 (Rh or ABO) 6 (others)
36.1 _+ 0.7 1.30 - 3.49 2.48 • 0.14 16M, 11 F 18 (Rh or ABO) 9 (others)
Table II. Variance in bilirubin, albumin, reserve album i n - b i n d i n g capacity, saturation point, m o l a r ratio and red blood cell bilirubin in pre a n d one hour post a l b u m i n h y p e r b i l i r u b i n e m i c plasma Pre
albumin Bilirubin (mg/dl) Mean • SEM Albumin (gm/dl) Mean • SEM RABC Mean • SEM Saturation point Mean • SEM Molar ratio Mean • SEM RBC-bilirubin Mean • SEM
10.3-28.5 16.7 • t.4 2.6-4.2 3.2 + 0.1 2.0-13.5 4.8 • 1.1 15.0-30.0 21.4 • 1.2 0.46-1.t0 0.82 • 0.5 0.73-1.9 1.33 _+ 0.15
1 hour post albumin 10.8-28.0 17.1 _+ 1.2 2.6-4.7 3.5 • 0.1 0.0-12.5 6.7 • 0.8 16.5-30.5 23.7 _+ 1.0 0.60-1.07 0.82 • 0.04 0.47-1.8 0.96 + 0.16
p
Albumin primed Pre exchange Bilirubin (mg/dl) Mean • SEM Albumin (gm/dl) Mean _+ SEM Post exchange Bilirubin (mg/dl) Mean • SEM Albumin (gm/dl) Mean _+ SEM 4 hours post exchange Bilirubin (mg/dl) Mean _+ SEM Albumin (gm/dl) Mean • SEM
Nonalbumin primed
10.8-29 16.5 _+ 1.3 2.7-4.7 3.5 _+ 0.2
10.5-28.5 14.8 • 0.9 2.2-3.9 3.2 • 0.1
4.0-10.0 6.7 • 0.5 2.6-3.9 3.3 _+ 0.1
3.5-11.5 7.3 +_ 0.4 2.0-3.8 3.2 • 0.1
6.5-17.0 12.1 • 1.2 2.5-3.6 3.1 • 0.1
7.5-17.5 11.6 _+ 0.6 2,4-3.8 3.1 +_ 0.1
p
> 0.3 > 0.1
> 0.3 > 0.5
> 0.5 n.s.
Table IV. Efficiency o f bilirubin removal in a l b u m i n a n d n o n a l b u m i n p r i m e d exchange transfusions
> 0.5 > 0.1 < 0.2 > 0.2 > 0.5 < 0.1
groups. Group I consisted o f 15 infants who were given 1 g m / k g o f salt-poor h u m a n a l b u m i n intravenously one hour prior to the exchange. Group II was c o m p o s e d o f 27
Albumin primed Volume (ml) 200-500 Mean • SEM 437 _+ 26 Hematocrit (%) 39-50 Mean + SEM 41.1 • 0.9 Bilirubin removed mg/bag Red blood cell 0.6-3.9 bilirubin Mean + SEM 1.6 • 0.3 Plasma bilirubin 7.9-32~8 Mean • SEM 24.3 _+ 2.2 Total bilirubin 8.5-42 Mean + SEM 26.0 _+ 2.5 Bilirubin/kg 5.6-14.5 Mean + SEM 10.8 + 0.7
Nonalbumin primed 100-500 402 • 22
> 0.3
36-55 41.6 • 0.9
> 0.5
0.3-4.7 1.2 _+ 0.2 4.7-40.6 22.2 • 2.0 5.4-51.1 23.4 _+ 2.1 4.2-18.6 8.9 • 0.6
> 0.2 > 0.3 > 0.3 < 0.3
infants who were given simple exchange transfusions. All babies received a two volume exchange with A C D b l o o d less than 48 hours old. T h e clinical data outlined in T a b l e I shows that the two groups are c o m p a r a b l y m a t c h e d with
exchange samples. In addition, all the blood r e m o v e d during the exchange transfusion was saved in a bag
regards to gestational age, birth weight, sex, a n d etiology
containing 1,500 units o f h e p a r i n . The bags were analyzed
o f the hyperbilirubinemia.
for volume, hematocrit, plasma bilirubin, and red b l o o d
Blood samples from infants in group I were d r a w n into heparinized tubes before receiving a l b u m i n a n d one h o u r
cell bilirubin. The total mass o f bilirubin r e m o v e d is calculated as:
after completion o f the a l b u m i n infusion. A bilirubin profile which included bilirubin, 11 albumin, ~2 R A B C by
1-hematocrit 100 x volume of exchanged blood x plasma bilirubin +
the sephadex gel filtration m e t h o d , 1~ a n d the erythrocytes for red cell bilirubin 14 was o b t a i n e d o n the specimens. A
heinatocrit 9. x volume of exchanged blood • red cell bilirubin. 100
bilirubin profile was also obtained o n both groups o f
Saturation point is defined as the total c o n c e n t r a t i o n o f
infants at pre, post exchange, and four hours post
bilirubin required to saturate all the p r i m a r y a l b u m i n
Volume 88 Number 4, part 1
Variance in albumin loading in exchange transfusions
6 11
-- 30' E c~ E Z m
~, 20 84 e~ B
.-1 m
< 10 D.
BILIRUBtN REMOVED mg/kg Fig. I. Correlation between plasma bilirubinconcentration and amount of bilirubin removed per kilogram body weight in both groups of infants receiving two volume exchange transfusions, r = 0.87; y = 2.14 + 1.32 x.
binding sites for bilirubin and is the sum of RABC and plasma bilirubin. Negative values denote oversaturation of primary binding sites. Molar ratios are calculated from the saturation points and albumin concentrations, using molecular weights of 585 for bilirubin and 69,000 for albumin. Variance analysis of unpaired t tests and regression coefficients were analyzed using an Olivetti-Underwood Programmer 101. RESULTS Effect of albumin priming. The change in bilirubin, albumin, RABC, saturation point, bilirubin-albumin molar ratio, and red cell bilirubin pre and one-hour post albumin infusion are shown in Table II. The mean increase in albumin concentration (0.3/dl) is associated with a slight elevation in bilirubin (0.4 mg/dl), an improvement in RABC (1.9 mg/dl) and in saturation point (1.3 mg/dl), and a decrease in red cell bilirubin (-0.3 mg/dl). The variance is not statistically significant in any of these determinations. Comparison of albumin and bilirubin concentrations in group I and II infants. The concentration of bilirubin and
albumin in groups I and II are shown in Table III. Analyses of either concentration show no statistical difference between the two groups at pre, post, or four-hour post exchange levels. Efficiency of exchange transfusions with albumin. Bilirubin removal data on the efficiency of bilirubin removal in groups I and II infants are shown in Table IV. No significant difference was obtained in any of the features studied. There is a positive correlation of the regression
line between the plasma bilirubin and the amount of bilirubin removed per kilogram body weight in both groups (Fig. 1). DISCUSSION The data presented showed no significant advantage in albumin priming in the group of infants studied. This seems to contradict earlier reports that albumin administration causes an increase in both serum bilirubin and albumin levels in artificially jaundiced rats. ~, a These experimental animals were loaded with bilirubin far in excess of the binding capacity of the albumin present. By contrast in the present study, the mean RABC of 4.8 _ 1.1 in our group of infants indicate the presence of adequate reserve binding sites prior to albumin therapy. The mean net increase in albumin of 0.3 gm/dl is in agreement with that reported by Odell and associates 4 of 0.4 g m / d l using the same albumin-priming procedure. However, the changes in bilirubin concentration one hour post albumin infusion were less predictable. Six of the 15 babies had a slight decrease with a mean net increase of ordy 0.4 mg/dl. There is a poor correlation of the changes in bilirubin concentration with RABC (r = 0,45), suggesting that the bilirubin increment is the net result of volume expansion caused by the infusion of albumin and the formation of new bilirubin rather than from an intracellular source. In two infants in whom the RABC was low (--2 and 0), indicating complete or almost complete saturation of ,the primary binding site on albumin, the beneficial effects of albumin priming is apparent. The RABC improved to 4
6 12
Chan and Schiff
and 7, respectively, while the bilirubin levels were increased by 2.5 and 4.5 mg/dl. Improvement in RABC reduced the fraction of diffusable bilirubin which if allowed to reach high levels may pose a threat to cellular toxicity. In fact. the high increment in bilirubin after albumin infusion in these infants may be an indication of redistribution Of the bilirubin pool; the direction of movement being from tissues to the bloodstream. It has been shown that red blood cell lipids bind bilirubin..... and that this property seemed to be influenced by the degree of saturation of the albumin molecule with bilirubin.1~ On this basis. Bratlid 17 suggested that erythrocyte bilirubin concentration may reflect the relative saturation of albumin with bilirubin: the range in our group of infants (n = 1l) is of the same magnitude as that reported by Bratlid. 17Administration of albumin decreases the eluable red cell bilirubin in all cases studied. This is not surPrising since albumin has a higher affinity for bilirubin than red celt membrane lipids with a resultant net efflux of bilirubin into the plasma as a new bilirubin albumin dynamic equilibrium is achieved. However. clinical interpretation of red cell bilirubin values is still in doubt. Although conclusions cannot be drawn because of our small sample size. preliminary data suggest that there is a fairly wide scatter of values (0.5 1.9 mg/dl) and seems to be poorly correlated with plasma bilirubin (r = 0.375), albumin (r = 0.320), and RABC (r = 0.668). The present findings with regards to efficiency of bilirubin removed is in agreement with that reported by Cromley and Wood2 No statistical difference in the amount of bilirubin removed per kilogram body weight was found in the two groups. In support of this observation, the albumin and bilirubin concentrations were comparable at pre, post, and four-hour post exchange levels (Table III). Erythrocyte-bound bilirubin was found to constitute 5% to 10% of plasma bilirubin in the exchanged blood and was therefore included as part of the total bilirubin removed. Our mean average of ( 8 . 9 - 10.8) mg/kg of total biiirubin removed in two volume exchanges is less impressive than those Obtained by Odell and associates~ and Comley and Wood 5 (12.3 and 14, respectively). The lower efficiency can be explained at least in part by the lower mean pre-exchange bilirubin concentrations in our group of infants. To further support this point, regression coefficients between the amount of bilirubin removed per kilogram body weight and plasma bilirubin levels were analyzed. Apositive correlation was obtained in these parameters in both groups of infants (Fig. 1). Thus it would appear that the amount of bilirubin removed bears a direct relationship to plasma bilirubin
The Journal of Pediatrics April 1976
concentration. In view of the small increment in bilirubin increase induced by albumin priming in the group of infants studied, it is not entirely unexpected that no improvement in efficiency in bilirubin removal was observed. It would seem therefore that the use of albumin as a priming agent does not improve the efficiency of removal of bilirubin during an exchange. In most instances the giving of albumin in the pre-exchange period does not significantly improve the reserve albuminTbinding capacity. This may be a reflection of the fact that these infants already have adequate reserve albumin-binding capacity and the addition of albumin did not change this situation. However, in two infants who had almost complete saturation of the albumin with endogenous bilirubin, the giving of the albumin seemed to improve the situation in a temporary fashion prior to the exchange transfusion. It would seem that the use of albumin, although safe, would o~ly have significant effect in those situations where the binding capacity is already compromised, and where there may be some delay, in achieving the exchange transfusion. It is recommended that reserve albuminbinding capacity would be a useful tool in the monitoring of infants prior to exchange transfusion, with regard to the timing of the exchange transfusion and also to the need of using albumin as a priming agent prior to the exchange transfusion. We thank Mrs. K. Merrills for technica! assistance, and the nursing staff of the neonatal intensive care unit, University of Alberta Hospital, for assistance with exchange transfusions. REFERENCES
1. BowenWR, Porter E, and Waters WJ: The protective action of albumin in bilirubin toxicity in newborn puppies, Am J Dis Child 98:568, 1959 (abstr). 2. SchmidR, Diamond I, Hammakar L, and Gundersen CB: Interaction of bilirubin with albumin, Nature 206:1041, t965. 3. Diamond I, and Schmid R: Experimental bilirubin encephalopathy. The mode of entry of bilirubin-C~ into the central nervous system, J Clin Invest 45:678, 1966. 4. Odell GB, Cohen SN, and Gordes EH: Administration of albumin in the management of hyperbilirubinemia by exchange transfusions, Pediatrics 30:613, 1962. 5. Comley A, and Wood B: Albumin administration in exchange transfusion for hyperbilirubinemia, Arch Dis Child 43:151, 1968. 6. Porter EG, and Waters WJ: A rapid micromethod for measuring the reserve albumin binding capacity in serum from newborn infants with hyperbilirubinemia, J Lab Clin Med 67:660, 1966. 7. Wood B, Comley A, and Sherwell J: Effect of additional albumin administration during exchange transfusion on plasma albumin binding capacity, Arch Dis Child 45:59, 1970.
Volume 88 Number 4, part 1
8. Lucey JF, Valaes T, and Doxiadis SA: Serum albumin reserve PSP dye binding capacity in infants with kernicterus, Pediatrics 39:876, 1967. 9. Chart G, Schiff D, and Stern L: Competitive binding of free fatty acids and bilirubin to albumin: differences in HBABA dye And Sephadex G-25 interpretation of results, Clin Biochem 4:205, 1971. 10. Schiff D, Chart G, and Stern L: Sephadex G-25 quantitative estimation of free bilimbin potential in jaundiced newborn infants' sera: a guide to the prevention of kernicterus, J Lab Clin Med 80:455, 1972. 11. Martinek RG: Improved micromethod for determination of serum bilirubin, Clin Chim Acta 13:161, 1966. 12. Gomall AG, Bardonwill CS, and David MM: Determination of serum proteins by the use of the biuret reaction, Biol Chem 177:751, 1949. Thin gel electrophoresis in agarose, Anal Inc Cat No 1-1000 (96) 1966.
Variance in albumin loading in exchange transfusions
6 13
13. Chan G, and Schiff D: Further clinical evaluation of the sephadex G-25 elution technique in the management of the hyperbilirubinemic infant, International Bilirubin Symposium, May 1974, Jerusalem, Israel. Birth Defect Series. In Press. 14. Bratlid D: Bilirubin binding of human erythrocytes, Scand J Clin Lab Invest 29:91, t972. 15. Watson D: The absorption of bilirubin by erythrocytes, Clin Claim Acta 7:733, 1962. 16. Oski FA, and Naiman JL: Red cell binding of bilirubin, J PEDIATR63:1034, 1963. 17. Bratlid D: Reserve albumin binding capacity salicylate saturation index, and red cell binding of bilirubin in neonatal jaundice, Arch Dis Child 48:393, 1973.
