- Email: [email protected]
Plasma lead levels in normal and lead-intoxicated children
./unua~:y 1974
The Journal q/' P E D I A T R I C S
45
Plasma lead levels in normal and lead-intoxicated children
Lead reaches tissues through the plasma, where the concentration of this ion is characteristically low. Little information exists o~l the mechanisms which determine the dynamics of lead in plasma. Measurement of plasma lead levels have been determined in normal and lead-intoxicated children, newborn infants, and children with sickle-cell disease. The results in all groups were remarkably similar and constant o vet a wide range of blood lead concentration and regardless of hematocrit. These results lend further support to the postulate that the red cell represents a repositoryfor lead, maintaining plasma lead concentration within closely defined limits, and that methods other than an isolated measurement of plasma lead will be necessary to uncover a presumably dynamic transport system between red cell and plasma.
John F. Rosen, M.D.*, C. Zarate-Saivador, M.D., a n d E m m a E. T r i n i d a d , B . S . , N e w York, N . Y .
I N PREVIOUS reports, the concentration of plasma lead (Pb) has been summarized as varying from 0 to less than 5/.~g per 100 ml. in lead-intoxicated children, 1-3although the specific data supporting the latter statement have never been published. Recently McIntyre and Angle, 4 based upon theoretical regression lines, have estimated Pb concentration in serum (or plasma) of a magnitude several times greater than previous estimates. In addition, these authors indicated that serum Pb levels were lower in a n e m i c children who were d e f i c i e n t in glucose-6-phosphate dehydrogenase than in nondeficient children. To dispel these differences, to characterize plasma Pb levels before studying the dynamics of Pb transport, to examine the usefulness of Pb levels in plasma as an indicator of Pb metabolism and/or transport, and to evaluate possible differences in the parFrom the Department qf Pediatrics, Mont(fiore Hospital and Medical Center and The A Ibert Einstein College o[' Medicine. This study was supported in part by a grant from The John A. Hartjbrd Foundation. Inc. *Reprint address: Mont~fiore Hospital and Medical Center, The Albert Einstein College qf MedichTe, New York, N. Y., 1046Z
tition of Pb in low and high hematocrit blood, we have measured plasma Pb levels in 165 children.
SUBJECTS, MATERIALS, AND METHODS All subjects, except newborn infants, were patients at Montefiore Hospital and ranged in age from one to ten years. None had undergone chelation therapy prior to study. By venipuncture with a heparinized, plastic syringe, approximately 1.5 ml. of whole blood was See Editor's column, p. 163. drawn and divided as follows: (1) 0.2 to 0.4 ml. was placed in a heparinized, Unopette macrosampler, which was then centrifuged at 2,000 r.p.m, for 10 minutes. The plasma thus obtained was stored at - 2 0 ~ C. until analysis for lead. (2) 0.2 to 0.4 ml. of whole blood was placed in a macrosampler, which was stored at 4 ~ C. until determination of whole blood lead concentration. (3) A bematocrit reading was obtained on venous blood by routine techniques. Specimens of blood were taken from newborn infants born at Morrisania City Hospital within 24 hours after delivery. According to previous collection techniques, 5
Vol. 84, No. 1, pp. 45-48
46
Rosen Zarate-Salvador.
and Trinidad
capillary sampling was carried out of the arch of the foot, With a No. 419 microlance (Becton, Dickinson and Co.), after sequentially cleansing this area with pHisoHex, alcohol, 0.3N nitric acid, and alcohol. The first drop of blood was wiped offthe skin with a laboratory tissue; 2 to 3 drops of blood were collected in a heparinized macrosampler, which was capped and stored at 4 ~ C.; the next 3 to 5 drops of blood were collected in a similar sampling cup that was centrifuged as above. The plasma sample thus obtained was stored as previously noted; lastly, a few drops of blood were used to perform a hematocrit determination. Lead determinations were carried out on 0.5 Ixl aliquots of whole blood by a~ flameless atomic absorption spectrophotometric technique with the carbon rod atomizer (CRA) as previously d e s c r i b e d ) D u p l i c a t e analyses were performed on a Varian-Techtron A A - 5 R atomic absorption spectrophotometer, with the C R A model No. 61 replacing the burner for air/acetylene at the 2,170 A line. The sample, surrounded by 0.3 and 0.2 t~ 1 of xylene, in a sheathing atmosp here of nitrogen, was sequentially dried, ashed, and atomized according to the preset voltage/time cycles on the power supply. As previously reported, this method has a sensitivity of 0.5 txg of Pb per 100 ml. and a standard deviation-of 0.91 txg of Pb per 100 ml. s For plasma samples, the same analytical procedure was used, except that 1.0 ~ I aliquots without xylene were injected into C R A model No. 63 with a micropipettor (Unimetrics No. 8010M).* In addition, different standards were prepared by diluting the working stock solution with appropriate volumes of a mixed salt solution that contained, per liter, 0.48 Gin. of KH 2 PO4, 0.28 Gm. of CaC12 2.90 Gm. of NaCI, 0.07 Gm. K 2 SO4, and 8.10 Gin. of E D T A (disodium salt). These standards were used to account for ionic interferences present in plasma. The composition of this salt solution was arrived at by closely (but not absolutely) matching the chemical matrix of plasma, so that (1) measurements were precise and reproducible and (2) the analytical signal was quenched only to the exact extent that allowed for virtually complete recoveries of different concentrations of added lead to plasma (see below). Phosphate, which may be present in relatively high concentration in the plasma of toddlers, markedly decreased the quenching effect of chloride ion on the signal. Hence, the concentrations of chloride and phosphate in these standards *Results of blood Pb measurements in 25 specimens run on both the CRA model No. 6I and No. 63 had a correlation coefficient(r) of 0.995. Thus, if whole blood Pb measurements had been carried out on a CRA No. 63, one wouldexpectsuch levelsto verycloselyreflectthoseObtained with the CRA No. 61 workhead.
The Journal of Pediatrics January 1974
were somewhat lower and higher, respectively, compared to those generally found in plasma. Even though this m i x ed salt solution fully compensated for apparent chemical interferences and competing ions, we do not claim to have recognized completely all such substances present in plasma. The settings on the power supply for the dry, ash, and atomized cycles were 7/18, 8.5/15, and 8/3, respectively. The sensitivity was 0.35/xg of Pb per 100 ml. and the. standard deviation of 165 determinations was 0.92/~g of Pb per 100 ml. Recoveries of Pb at concentrations of 1, 4, and 7/xg per 100 rrd., when added to plasma, averaged 0.8, 4.1, and 7.2 /xg per 100 ml., respectively. Experiments with radioactive 210Pb yielded similar recoveries. With the above methods no difference between serum and plasma values of the same blood sample was discerned* RESULTS Table I lists the concentration of plasma Pb in 31 normal children, 48 children with elevated levels of blood Pb in whom excessive exposure appeared likely, 56 children with levels of Pb in the toxic range, 17 children with sickle-cell disease with and without elevated blood Pb levels, and 13 newborn infants sampled within 24 hours of delivery. In all groups, the mean concentration of Pb in plasma was remarkably constant and similar, within groups and between groups, regardless of whole blood Pb levels, hemat0crit, and symptoms. The range in plasma Pb, encompassing all groups, was 1 to 7/xg per 100 ml. and, as summarized in Table II, none of the mean values differed from the normal group. On the other hand, the concentration of Pb in plasma, expressed as a percentage of the whole blood Pb, varied from a high of 15 per cent at 20/xg per 100 ml. to a low of 2 per cent at 136/xg per 100 ml. (Table II). DISCUSSION
AND SPECULATION
The results have shown a notable constancy in Pb concentration in the plasma of children over a wide renge of hematocrit and whole blood Pb levels (Table I). No sig~Attwo separate centers, samples of wholebloodwe~'edrawn from 12 children. After hematocrit measurements (these values were not sent to us), 50/xl aliquotsof whole blood,packed red cells,and plasma from each sample were sent to this laboratory. After our results on all three aliquots from each sample were returned, theoreticalplasma concentrations were calculated, using red cell and wholeblood values as the two knowns (see formula in Discussion section). Subsequently these values were forwarded to us by mail. With the use of these 12 theoreticalvaluesas duplicates for plasma Pb determinations made by methods described,the standard deviation was 1.25 ~g per 100 ml. In addition, several plasma samples were divided and sent to EnvironmentalScienceAssociates, where plasma Pb levelswere measured by anodic stripping voltammetry.With the use of their results as duplicates for our values,the standard deviation was 0.66 ,~g per 100 ml.
Volume 84 Number 1
Plasma lead levels
47
Table I. Results of plasma lead determinations, whole blood lead concentration, and hematocrit, in all patient groups
Children with sicklecell disease
Neonates
90-99 100-136
18-136
:12-34 (mean, 19)
No. of subjects 14 17 24 24 14 14 10 8 10 Plasma Pb (mean,/zg% 3.0 3.2 3.0 3.3 3.2 3.10 3.0 3.0 _IS.E.M.) +0.44 __+0.33 -+0.37 • -+0.34 -+0.31 • ___0.33 • Range plasma Pb (/zg%) 1-7 1-6 1-7 1-7 1-5 1-7 2-4 1-3 1-5 Mean hematocrit (%) 36 36 34 34 33 31 29 28 26 Range of hematocrit (%) 35-37 35-37 34-36 34-36 31-35 30-33 28-30 27-30 25-28
17 3.1 _+0.35 1-6 20 18-24
13 3.0 :t:0.40 1-4 60 48-75
"Normals " ~ange o f whole blood Pb* (Ixg%) 20-29
30-39
Undue exposure likely
40-49
50-59
Undue exposure certain
60-69
70-79
80-89
*Whole bloodPb levels have not been corrected for hematocrit.
nificant statistical difference in these plasma values could be established fof~children with sickle-cell disease or lead intoxication, or for newborn infants, when compared with normal children (Table II). Though these results have agreed, generally, with two preceding reports, v e r y few subjects previously reported fell within the pediatric age group and none of these had either symptomatic Pb intoxication and/or sickle-cell disease. 2, 3 In view of the constancy of the data reported herein, the only way to elicit differences--probably artificial o n e s - - i n the patient groups, would be to express plasma Pb concentration as a percentage of whole blood Pb concentration, i.e., the percentage of plasma Pb in whole blood decreases as whole blood Pb increases. In regard to these strikingly uniform plasma Pb levels, we would speculate that, even though the determination of plasma Pb has turned out to be a relatively "static" measurement, saturation of plasma and/or overflow into plasma by red cell-bound Pb have not occurred. Stated differently, at the extreme of high blood Pb concentration and low hematocrit in children with sickle-cell disease, we have been unable to demonstrate an increase in plasma Pb concentration. Therefore, our observations have provided further support for the hypothesis that the red cell represents a rich and "protective" repository for Pb, 6 maintaining plasma Pb levels within very closely defined limits. Previous studies of adult human blood in vitro have supported this postulate, and have demonstrated constant levels of lead in plasma in the face of rising levels of lead in whole blood. 7 However, in a unique group of vitamin D-deficient, lead-intoxicated rats, apparent saturation of the red cell reservoir has been demonstrated above a blood Pb level of 225 to 250/xg per 100 ml. 8 In these hypocalcemic animals, concentration
Table II. Summary of data
Subject groups *
"Normals" (20 to 39/xg%) Undue exposure likely (40 to 59~zg%) Undue exposure certain (60 to 136/zg%) Children with sickle-cell disease (18 to 136/zg%) Neonates (12 to 34/zg%)
Mean plasma Pb (Izg% +_ S.E.M.)
3.10--+ 0.31 3.20-+0.29"t" 3.10-+ 0.26"t" 3.10+ 0.28t 3.0 _+ 0.31t
*Values in parentheses indicatewhole blood Pb concentration,not corrected for hematocrit. 1"Noneof these valuesdiffersignificantlyfrom the "normal"group. of lead in plasma may reach 30/xg per 100 ml. 8Thus, if our speculation proves correct, then "protection" by this red cell repository, even in low hematocrit blood, may be relatively unlimited and, accordingly, correction of whole blood Pb levels for hematocrit may no longer be justified (see below). The presented data have diverged considerably from Pb levels recently estimated from regression-coefficient equations based upon measurement of whole blood and red cell Pb, corrected for hematocrit. 4 In the cited report, Pb levels in serum were estimated at 22 to 43 txg per 100 ml., the lower figure in anemic children deficient in glucose-6-phosphate dehydrogenase (G-6-PD).4 The data presented here do not confirm the magnitude of these serum levels, nor have we confirmed a significant difference among normal, high, or low hematocrit groups. Admittedly, the anemic patients studied in this report all had homozygous sickle-cell disease and were not known to be deficient in G-6-PD. However, we
48
Rosen, Zarate-Salvador, and Trinidad
would suggest that these discrepancies are based upon t w o factors: (1) An indirect measurement of serum Pb has been calculated--a direct measurement was not determined and (2) the authors cited above have corrected whole blood and red cell Pb values for hematocrit which may have been the artificial source of their high estimate. In this regard, we would speculate that the red cell capacity to bind Pb may not have been exhausted within the wide gamut of hematocrit and whole blood Pb values reported here and, until such time as a clear difference has been demonstrated in the uptake, partition, and/or transport of Pb between normal compared to low or high hematocrit blood, correction of whole blood Pb levels for hematocrit appears not to be warranted. Thus, if Mclntyre and associates 4 had not corrected whole blood and red cell Pb values for hematocrit, perhaps their high estimates of serum Pb would have come significantly closer to those measured and reported herein. Moreover, if we were to correct Pb levels of newborn i n f a r l ~ r ~ a t o c r i t , several of these values would fall into the undue exposure category. Though this possibility has not been entirely excluded, 9 the calculated concentration of/xg of Pb per 100 ml. of red blood cells (RBC) at a whole blood Pb of 19/xg per 100 ml., in these newborn infants, was considerably less than that of a normal child or adult with a similar whole blood Pb level. This calculation was based upon: Volume x Concentration = (Whole blood Pb) Volume x Concentration + Volume x Concentration (Plasma Pb) (Red cell Pb) or
/xg Pb/100 ml. RBC = /xg Pb/100 ml. -- [/xg Pb/100 ml. • (1 - Hematocrit)] (Whole blood) (Plasma) Hematocrit We conclude, therefore, that the results of this study have suggested: (1) Correction of whole blood Pb for
The Journal of Pediatrics January 1974
hematocrit does not appear justified, until a difference in the uptake, transport, and/or partition of Pb between normal compared to low, or high hematocrit blood has been demonstrated. (2) Despite its relatively constant concentration in plasma, we propose that plasma-associated Pb is still the more biologically (rapidly exchangeable) fraction of blood Pb, l~ H and dynamic methods ~2 other than an isolated measurement of plasma Pb will likely be necessary to uncover the kinetics of Pb transport between red cell, plasma, and soft and hard tissues.
We are very grateful to Mr. P. Shaffer, who collected the newborn samples, and to Mr. Y. Greener, who performed some of the determinations included in this study.
REFERENCES
1. Robinson, M., Karpinski, F., and Brieger, H.: The concentration of lead in plasma, whole blood and erythrocytes of infants and children, Pediatrics 21: 793, 1958. 2. Butt, E. M., Nusbaum, R. E., Gilmour, T., Didi0, S., and Maciano, Sr.: Trace metal levels in human serum and blood, Arch. Environ. Health 8: 52, 1964. 3. Farrelly, R. O., and Pybus, J.: Measurement of lead in blood and urine by atomic absorption spectropl$otometry, Clin. Chem. 15: 566, 1969. 4. McIntyre, M., and Angle, C.: Air lead: Relation to lead in blood of black school children deficient in glucose-6phosphate dehydrogenase, Science 177: 520, 1972. 5. Rosen, J. F.: The microdetermination of blood lead in children by flameless atomic absorption: The carbon rod atomizer, J. Lab. Clin. Med. 80: 567, 1972. 6. Goyer, R. A.: Lead toxicity--A problem in environmental pathology, Am. J. PathoL 64: 167, 1971. 7. Clarkson, T. W., and Kench, J. E.: Uptake of lead by human erythrocytes in vitro, Biochem. J. 69: 432, 1958. 8. Rosen, J. F., and Roginsky, M.: Lead-intoxicated children: Plasma levels of 25-hydroxycholecalciferol (25-HCC), Pediatr. Res. 7: 393, 1973. 9. Rajegowda, B. K., Glass, L., and Evans, H. E.: Lead concentrations in the newborn infant, J. Pediatr. 80: 116, 1972. 10. Rosen, L F.: Actions of calcitonin in lead-intoxicated rats, Clin. Res. 19: 653, 1971. 11. Rosen, J. F.: Actions of calcitonin and parathyroid hormone in lead-intoxicated rats, Clin. Res. 20: 755, 1972. 12. Rosen, J. F., and Haymovits, A.: Lead intoxication: Displacement of lead from rat erythrocytes by ionized calcium in vitro, Pediatr. Res. 7: 393, 1973.
The Journal q/' P E D I A T R I C S
45
Plasma lead levels in normal and lead-intoxicated children
Lead reaches tissues through the plasma, where the concentration of this ion is characteristically low. Little information exists o~l the mechanisms which determine the dynamics of lead in plasma. Measurement of plasma lead levels have been determined in normal and lead-intoxicated children, newborn infants, and children with sickle-cell disease. The results in all groups were remarkably similar and constant o vet a wide range of blood lead concentration and regardless of hematocrit. These results lend further support to the postulate that the red cell represents a repositoryfor lead, maintaining plasma lead concentration within closely defined limits, and that methods other than an isolated measurement of plasma lead will be necessary to uncover a presumably dynamic transport system between red cell and plasma.
John F. Rosen, M.D.*, C. Zarate-Saivador, M.D., a n d E m m a E. T r i n i d a d , B . S . , N e w York, N . Y .
I N PREVIOUS reports, the concentration of plasma lead (Pb) has been summarized as varying from 0 to less than 5/.~g per 100 ml. in lead-intoxicated children, 1-3although the specific data supporting the latter statement have never been published. Recently McIntyre and Angle, 4 based upon theoretical regression lines, have estimated Pb concentration in serum (or plasma) of a magnitude several times greater than previous estimates. In addition, these authors indicated that serum Pb levels were lower in a n e m i c children who were d e f i c i e n t in glucose-6-phosphate dehydrogenase than in nondeficient children. To dispel these differences, to characterize plasma Pb levels before studying the dynamics of Pb transport, to examine the usefulness of Pb levels in plasma as an indicator of Pb metabolism and/or transport, and to evaluate possible differences in the parFrom the Department qf Pediatrics, Mont(fiore Hospital and Medical Center and The A Ibert Einstein College o[' Medicine. This study was supported in part by a grant from The John A. Hartjbrd Foundation. Inc. *Reprint address: Mont~fiore Hospital and Medical Center, The Albert Einstein College qf MedichTe, New York, N. Y., 1046Z
tition of Pb in low and high hematocrit blood, we have measured plasma Pb levels in 165 children.
SUBJECTS, MATERIALS, AND METHODS All subjects, except newborn infants, were patients at Montefiore Hospital and ranged in age from one to ten years. None had undergone chelation therapy prior to study. By venipuncture with a heparinized, plastic syringe, approximately 1.5 ml. of whole blood was See Editor's column, p. 163. drawn and divided as follows: (1) 0.2 to 0.4 ml. was placed in a heparinized, Unopette macrosampler, which was then centrifuged at 2,000 r.p.m, for 10 minutes. The plasma thus obtained was stored at - 2 0 ~ C. until analysis for lead. (2) 0.2 to 0.4 ml. of whole blood was placed in a macrosampler, which was stored at 4 ~ C. until determination of whole blood lead concentration. (3) A bematocrit reading was obtained on venous blood by routine techniques. Specimens of blood were taken from newborn infants born at Morrisania City Hospital within 24 hours after delivery. According to previous collection techniques, 5
Vol. 84, No. 1, pp. 45-48
46
Rosen Zarate-Salvador.
and Trinidad
capillary sampling was carried out of the arch of the foot, With a No. 419 microlance (Becton, Dickinson and Co.), after sequentially cleansing this area with pHisoHex, alcohol, 0.3N nitric acid, and alcohol. The first drop of blood was wiped offthe skin with a laboratory tissue; 2 to 3 drops of blood were collected in a heparinized macrosampler, which was capped and stored at 4 ~ C.; the next 3 to 5 drops of blood were collected in a similar sampling cup that was centrifuged as above. The plasma sample thus obtained was stored as previously noted; lastly, a few drops of blood were used to perform a hematocrit determination. Lead determinations were carried out on 0.5 Ixl aliquots of whole blood by a~ flameless atomic absorption spectrophotometric technique with the carbon rod atomizer (CRA) as previously d e s c r i b e d ) D u p l i c a t e analyses were performed on a Varian-Techtron A A - 5 R atomic absorption spectrophotometer, with the C R A model No. 61 replacing the burner for air/acetylene at the 2,170 A line. The sample, surrounded by 0.3 and 0.2 t~ 1 of xylene, in a sheathing atmosp here of nitrogen, was sequentially dried, ashed, and atomized according to the preset voltage/time cycles on the power supply. As previously reported, this method has a sensitivity of 0.5 txg of Pb per 100 ml. and a standard deviation-of 0.91 txg of Pb per 100 ml. s For plasma samples, the same analytical procedure was used, except that 1.0 ~ I aliquots without xylene were injected into C R A model No. 63 with a micropipettor (Unimetrics No. 8010M).* In addition, different standards were prepared by diluting the working stock solution with appropriate volumes of a mixed salt solution that contained, per liter, 0.48 Gin. of KH 2 PO4, 0.28 Gm. of CaC12 2.90 Gm. of NaCI, 0.07 Gm. K 2 SO4, and 8.10 Gin. of E D T A (disodium salt). These standards were used to account for ionic interferences present in plasma. The composition of this salt solution was arrived at by closely (but not absolutely) matching the chemical matrix of plasma, so that (1) measurements were precise and reproducible and (2) the analytical signal was quenched only to the exact extent that allowed for virtually complete recoveries of different concentrations of added lead to plasma (see below). Phosphate, which may be present in relatively high concentration in the plasma of toddlers, markedly decreased the quenching effect of chloride ion on the signal. Hence, the concentrations of chloride and phosphate in these standards *Results of blood Pb measurements in 25 specimens run on both the CRA model No. 6I and No. 63 had a correlation coefficient(r) of 0.995. Thus, if whole blood Pb measurements had been carried out on a CRA No. 63, one wouldexpectsuch levelsto verycloselyreflectthoseObtained with the CRA No. 61 workhead.
The Journal of Pediatrics January 1974
were somewhat lower and higher, respectively, compared to those generally found in plasma. Even though this m i x ed salt solution fully compensated for apparent chemical interferences and competing ions, we do not claim to have recognized completely all such substances present in plasma. The settings on the power supply for the dry, ash, and atomized cycles were 7/18, 8.5/15, and 8/3, respectively. The sensitivity was 0.35/xg of Pb per 100 ml. and the. standard deviation of 165 determinations was 0.92/~g of Pb per 100 ml. Recoveries of Pb at concentrations of 1, 4, and 7/xg per 100 rrd., when added to plasma, averaged 0.8, 4.1, and 7.2 /xg per 100 ml., respectively. Experiments with radioactive 210Pb yielded similar recoveries. With the above methods no difference between serum and plasma values of the same blood sample was discerned* RESULTS Table I lists the concentration of plasma Pb in 31 normal children, 48 children with elevated levels of blood Pb in whom excessive exposure appeared likely, 56 children with levels of Pb in the toxic range, 17 children with sickle-cell disease with and without elevated blood Pb levels, and 13 newborn infants sampled within 24 hours of delivery. In all groups, the mean concentration of Pb in plasma was remarkably constant and similar, within groups and between groups, regardless of whole blood Pb levels, hemat0crit, and symptoms. The range in plasma Pb, encompassing all groups, was 1 to 7/xg per 100 ml. and, as summarized in Table II, none of the mean values differed from the normal group. On the other hand, the concentration of Pb in plasma, expressed as a percentage of the whole blood Pb, varied from a high of 15 per cent at 20/xg per 100 ml. to a low of 2 per cent at 136/xg per 100 ml. (Table II). DISCUSSION
AND SPECULATION
The results have shown a notable constancy in Pb concentration in the plasma of children over a wide renge of hematocrit and whole blood Pb levels (Table I). No sig~Attwo separate centers, samples of wholebloodwe~'edrawn from 12 children. After hematocrit measurements (these values were not sent to us), 50/xl aliquotsof whole blood,packed red cells,and plasma from each sample were sent to this laboratory. After our results on all three aliquots from each sample were returned, theoreticalplasma concentrations were calculated, using red cell and wholeblood values as the two knowns (see formula in Discussion section). Subsequently these values were forwarded to us by mail. With the use of these 12 theoreticalvaluesas duplicates for plasma Pb determinations made by methods described,the standard deviation was 1.25 ~g per 100 ml. In addition, several plasma samples were divided and sent to EnvironmentalScienceAssociates, where plasma Pb levelswere measured by anodic stripping voltammetry.With the use of their results as duplicates for our values,the standard deviation was 0.66 ,~g per 100 ml.
Volume 84 Number 1
Plasma lead levels
47
Table I. Results of plasma lead determinations, whole blood lead concentration, and hematocrit, in all patient groups
Children with sicklecell disease
Neonates
90-99 100-136
18-136
:12-34 (mean, 19)
No. of subjects 14 17 24 24 14 14 10 8 10 Plasma Pb (mean,/zg% 3.0 3.2 3.0 3.3 3.2 3.10 3.0 3.0 _IS.E.M.) +0.44 __+0.33 -+0.37 • -+0.34 -+0.31 • ___0.33 • Range plasma Pb (/zg%) 1-7 1-6 1-7 1-7 1-5 1-7 2-4 1-3 1-5 Mean hematocrit (%) 36 36 34 34 33 31 29 28 26 Range of hematocrit (%) 35-37 35-37 34-36 34-36 31-35 30-33 28-30 27-30 25-28
17 3.1 _+0.35 1-6 20 18-24
13 3.0 :t:0.40 1-4 60 48-75
"Normals " ~ange o f whole blood Pb* (Ixg%) 20-29
30-39
Undue exposure likely
40-49
50-59
Undue exposure certain
60-69
70-79
80-89
*Whole bloodPb levels have not been corrected for hematocrit.
nificant statistical difference in these plasma values could be established fof~children with sickle-cell disease or lead intoxication, or for newborn infants, when compared with normal children (Table II). Though these results have agreed, generally, with two preceding reports, v e r y few subjects previously reported fell within the pediatric age group and none of these had either symptomatic Pb intoxication and/or sickle-cell disease. 2, 3 In view of the constancy of the data reported herein, the only way to elicit differences--probably artificial o n e s - - i n the patient groups, would be to express plasma Pb concentration as a percentage of whole blood Pb concentration, i.e., the percentage of plasma Pb in whole blood decreases as whole blood Pb increases. In regard to these strikingly uniform plasma Pb levels, we would speculate that, even though the determination of plasma Pb has turned out to be a relatively "static" measurement, saturation of plasma and/or overflow into plasma by red cell-bound Pb have not occurred. Stated differently, at the extreme of high blood Pb concentration and low hematocrit in children with sickle-cell disease, we have been unable to demonstrate an increase in plasma Pb concentration. Therefore, our observations have provided further support for the hypothesis that the red cell represents a rich and "protective" repository for Pb, 6 maintaining plasma Pb levels within very closely defined limits. Previous studies of adult human blood in vitro have supported this postulate, and have demonstrated constant levels of lead in plasma in the face of rising levels of lead in whole blood. 7 However, in a unique group of vitamin D-deficient, lead-intoxicated rats, apparent saturation of the red cell reservoir has been demonstrated above a blood Pb level of 225 to 250/xg per 100 ml. 8 In these hypocalcemic animals, concentration
Table II. Summary of data
Subject groups *
"Normals" (20 to 39/xg%) Undue exposure likely (40 to 59~zg%) Undue exposure certain (60 to 136/zg%) Children with sickle-cell disease (18 to 136/zg%) Neonates (12 to 34/zg%)
Mean plasma Pb (Izg% +_ S.E.M.)
3.10--+ 0.31 3.20-+0.29"t" 3.10-+ 0.26"t" 3.10+ 0.28t 3.0 _+ 0.31t
*Values in parentheses indicatewhole blood Pb concentration,not corrected for hematocrit. 1"Noneof these valuesdiffersignificantlyfrom the "normal"group. of lead in plasma may reach 30/xg per 100 ml. 8Thus, if our speculation proves correct, then "protection" by this red cell repository, even in low hematocrit blood, may be relatively unlimited and, accordingly, correction of whole blood Pb levels for hematocrit may no longer be justified (see below). The presented data have diverged considerably from Pb levels recently estimated from regression-coefficient equations based upon measurement of whole blood and red cell Pb, corrected for hematocrit. 4 In the cited report, Pb levels in serum were estimated at 22 to 43 txg per 100 ml., the lower figure in anemic children deficient in glucose-6-phosphate dehydrogenase (G-6-PD).4 The data presented here do not confirm the magnitude of these serum levels, nor have we confirmed a significant difference among normal, high, or low hematocrit groups. Admittedly, the anemic patients studied in this report all had homozygous sickle-cell disease and were not known to be deficient in G-6-PD. However, we
48
Rosen, Zarate-Salvador, and Trinidad
would suggest that these discrepancies are based upon t w o factors: (1) An indirect measurement of serum Pb has been calculated--a direct measurement was not determined and (2) the authors cited above have corrected whole blood and red cell Pb values for hematocrit which may have been the artificial source of their high estimate. In this regard, we would speculate that the red cell capacity to bind Pb may not have been exhausted within the wide gamut of hematocrit and whole blood Pb values reported here and, until such time as a clear difference has been demonstrated in the uptake, partition, and/or transport of Pb between normal compared to low or high hematocrit blood, correction of whole blood Pb levels for hematocrit appears not to be warranted. Thus, if Mclntyre and associates 4 had not corrected whole blood and red cell Pb values for hematocrit, perhaps their high estimates of serum Pb would have come significantly closer to those measured and reported herein. Moreover, if we were to correct Pb levels of newborn i n f a r l ~ r ~ a t o c r i t , several of these values would fall into the undue exposure category. Though this possibility has not been entirely excluded, 9 the calculated concentration of/xg of Pb per 100 ml. of red blood cells (RBC) at a whole blood Pb of 19/xg per 100 ml., in these newborn infants, was considerably less than that of a normal child or adult with a similar whole blood Pb level. This calculation was based upon: Volume x Concentration = (Whole blood Pb) Volume x Concentration + Volume x Concentration (Plasma Pb) (Red cell Pb) or
/xg Pb/100 ml. RBC = /xg Pb/100 ml. -- [/xg Pb/100 ml. • (1 - Hematocrit)] (Whole blood) (Plasma) Hematocrit We conclude, therefore, that the results of this study have suggested: (1) Correction of whole blood Pb for
The Journal of Pediatrics January 1974
hematocrit does not appear justified, until a difference in the uptake, transport, and/or partition of Pb between normal compared to low, or high hematocrit blood has been demonstrated. (2) Despite its relatively constant concentration in plasma, we propose that plasma-associated Pb is still the more biologically (rapidly exchangeable) fraction of blood Pb, l~ H and dynamic methods ~2 other than an isolated measurement of plasma Pb will likely be necessary to uncover the kinetics of Pb transport between red cell, plasma, and soft and hard tissues.
We are very grateful to Mr. P. Shaffer, who collected the newborn samples, and to Mr. Y. Greener, who performed some of the determinations included in this study.
REFERENCES
1. Robinson, M., Karpinski, F., and Brieger, H.: The concentration of lead in plasma, whole blood and erythrocytes of infants and children, Pediatrics 21: 793, 1958. 2. Butt, E. M., Nusbaum, R. E., Gilmour, T., Didi0, S., and Maciano, Sr.: Trace metal levels in human serum and blood, Arch. Environ. Health 8: 52, 1964. 3. Farrelly, R. O., and Pybus, J.: Measurement of lead in blood and urine by atomic absorption spectropl$otometry, Clin. Chem. 15: 566, 1969. 4. McIntyre, M., and Angle, C.: Air lead: Relation to lead in blood of black school children deficient in glucose-6phosphate dehydrogenase, Science 177: 520, 1972. 5. Rosen, J. F.: The microdetermination of blood lead in children by flameless atomic absorption: The carbon rod atomizer, J. Lab. Clin. Med. 80: 567, 1972. 6. Goyer, R. A.: Lead toxicity--A problem in environmental pathology, Am. J. PathoL 64: 167, 1971. 7. Clarkson, T. W., and Kench, J. E.: Uptake of lead by human erythrocytes in vitro, Biochem. J. 69: 432, 1958. 8. Rosen, J. F., and Roginsky, M.: Lead-intoxicated children: Plasma levels of 25-hydroxycholecalciferol (25-HCC), Pediatr. Res. 7: 393, 1973. 9. Rajegowda, B. K., Glass, L., and Evans, H. E.: Lead concentrations in the newborn infant, J. Pediatr. 80: 116, 1972. 10. Rosen, L F.: Actions of calcitonin in lead-intoxicated rats, Clin. Res. 19: 653, 1971. 11. Rosen, J. F.: Actions of calcitonin and parathyroid hormone in lead-intoxicated rats, Clin. Res. 20: 755, 1972. 12. Rosen, J. F., and Haymovits, A.: Lead intoxication: Displacement of lead from rat erythrocytes by ionized calcium in vitro, Pediatr. Res. 7: 393, 1973.