- Email: [email protected]
Role of hemolysis in neonatal jaundice associated with glucose-6-phosphate dehydrogenase deficiency
804
Seidman et al.
The Journal of Pediatrics November 1995
Role of hemolysis in neonatal jaundice associated with glucose-6-phosphate dehydrogenase deficiency Daniel S. Seidman, MD, Michal Shiloh, MD, David K, Stevenson, ME), Hendrik J. Vreman, PhD, Ido Paz, MD, and Rena Gale, MD From the Department of Neonatology, Bikur-Cholim Hospital, Jerusalem; the Department of Obstetrics and Gynecology, Sheba Medical Center, TeI-Hashomer, Israel; and the Department of Pediatrics, Stanford University School of Medicine, Stanford, California
End-tidal carbon monoxide was measured in 108 newborn infants who had been screened for glucose-6-phosphate dehydrogenase (G6PD) deficiency. The mean ± SD end-tidal carbon monoxide did not differ significantly between the G6PD-deficient and the normal neonates, 2.1 ± 0.6 IJI/L and 2.0 ± 0.5 IJI/L, respectively, within 12 hours of birth and 1.9 ± 1.4 IJI/L and 1.5 ± 0.7 IJI/L, respectively, at 48 to 72 hours after birth. On the basis of these measurements, hemolysis is not a sufficient explanation for jaundice in G6PD-deficient newborn infants in the transitional period. (J PEDIATR1995;127:804-6)
The pathophysiologic basis for excessive jaundice in infants with glucose-6-phosphate dehydrogenase deficiency has not been established. Increased erythrocyte breakdown and reduced glucuronidation of bilirubin caused by defective G6PD activity in the hepatocyte may have a role in the pathogenesis. 1 Measuring bilirubin production in infants with G6PD deficiency may allow us to determine the role of hemolysis in the pathophysiology of the hyperbilirubinemia. The degradation of heme through the heine oxygenase pathway results in equimolar production of bilirubin and carbon monoxide, which is subsequently expelled during breathing. 2 End-tidal breath CO correlates with bilirubin production, and can therefore be used in its measurement. 3 This study examined whether early hemolysis, as reflected by increased bilirubin production, is more frequent or severe in G6PD-deficient infants. METHODS We prospectively screened for G6PD deficiency 108 term infants who were physically appropriate for their gestational ages and who were born between August and DecemSupported by National Institute of Child Health and Human Development grants HD 14426 and RR00070 and by the Mary L. Johnson Research Fund. Reprint requests: Daniel S. Seidman, MD, Division of Neonatology, Department of Pediatrics, Room $226, M/C 5119, Stanford University School of Medicine, Stanford, CA 94305-5119. Submitted for publication May 1, 1995; accepted June 23, 1995. Copyright © 1995 by Mosby-Year Book, Inc. 0022-3476/95/$5.00 + 0 9/24/67323
ber 1993 to mothers from areas known to have high incidences of G6PD deficiency (i.e., Kurdistan, Iraq, Iran, Syria, Turkey, and Buchara). 4 Exclusion criteria included exposure to any known hemolytic agent, blood group incompatibility, insmmaental delivery, neonatal asphyxia, congenital malformations, fever or any evidence of an infectious disease, cephalohematoma or signs of bruising, or maternal illness such as diabetes mellitus or preeclampsia. Maternal informed consent was obtained in all cases in accordance with the institution's human investigation review board. I
ETCOc End-tidal breath carbon monoxide corrected for inhaled carbon monoxide Glucose-6-phosphatedehydrogenase G6PD
All neonates had three consecutive end-tidal carbon monoxide measurements performed within 12 hours of birth and repeated at 48 to 72 hours of life. We performed end-tidal carbon monoxide measurements with a portable automated ETCOc analyzer (Baby's Breath carbon monoxide analyzer, Natus Medical Inc., San Carlos, Calif.).5 We collected and simultaneously analyzed triplicate samples of breath exhaled from the quietly resting (not crying) infants via a 5F catheter placed 2 cm into the anterior nares. We calculated the average error for ETCOc by subtracting the minimal value from the maximal value and dividing by the mean of the three measurements. The average margin of error for the triplicate ETCOc measurements was 44%. Blood group, direct Coombs test, and hematocrit and reticulocyte counts were determined by standard techniques.
The Journal of Pediatrics Volume 127, Number 5
Seidman et al.
805
Table. ETCOc and serum bilirubin measurements in healthy term infants according to maternal smoking and G6PD status Serum total bilirubin (maximum level)
ETCOc (IJI/L) Subjects (n)t Smoking mother (14) (range) Nonsmokingmother (85) (range) G6PD deficiency(17):~(range) G6PD normal (68)$ (range)
<12 hr
48-72 hr
(IJmol/U)
6.9 -+ 7.3§ (1.7-30.7) 2.0 -+ 0.6 (1.1-4.9) 2.1 -+ 0.6 (1.3-3.5) 2.0 -+ 0.5 (1.1-4.9)
1.4 -+ 0.5 (0.7-2.3) 1.5 -+ 0.7 (0.3-3.8) 1.9 -+ 1.4 (0.5-3.8) 1.5 -+ 0.7 (0.3-3.2)
140 -+45 (60-269) 176 -+ 71 (26-339) 179 -+ 78 (41-339) 173 -+ 70 (26-275)
Values are mean_+SD, with range in parentheses *To convertto milligramsper deciliter,divideby 17.1. "~Excludinginfantswith ABO blood groupincompatibility. 5/Infantsof smokingmothers were excluded. §p <0.03, comparedwith those infantswith a nonsmokingmother.
We measured serum total bilirnbin by determining optical absorbance at 460 nm, with correction for hemoglobin contamination at 550 nm (Unistat Bilirubinometer, AO Scientific Instruments, Buffalo, N.Y.). We performed screening for G6PD activity in erythrocytes using kit No. 400K (Sigma Chemical Co., St. Louis, Mo.), which is a colorimetric method based on the reduction of the blue dye dichlorophenol indophenol to a colorless state by reduced nicotinamideadenine dinucleotide phosphate formed in the presence of G6PD. Statistical analysis was performed with a nonpaired Student t test for continuous variables and with chi-square analysis or the Fisher Exact Test for dichotomous data.
RESULTS Of the 108 infants included in our study, 22 were G6PD deficient. The G6PD-deficient neonates did not differ significantly from the control group infants in maternal age, parity, and smoking; use of epidural anesthesia; frequency of oxytocin augmentation; cesarean section rate; 1- and 5-minute Apgar scores; gestational age; or birth weight. Breast-feeding was significantly less frequent among mothers of G6PD-deficient neonates, 59.1% versus 87.2% of control group mothers (p <0.006). Neonatal jaundice caused by ABO incompatibility was identified in nine infants (one study infant and eight control infants); the mean ETCOc level was 4.0 -+ 1.1 p1/L (range, 3.1 to 5.9 Ial/L) within 12 hours of birth, reflecting the hemolytic component of the hyperbilirubinemia. All were treated by phototherapy. In the two patients who required exchange transfusions, the ETCOc concentrations were 5.1 and 5.9/al/L, respectively. These values decreased to 0.55 and 2.2/al/L after exchange transfusions. These nine infants were excluded from the analysis.
Because ETCOc concentrations are highly sensitive to the effects of maternal smoking, those infants whose mothers smoked (4 study infants and 10 control infants) were also excluded from the analysis. However, their ETCOc values at 48 to 72 hours were no longer affected by maternal smoking (Table). The mean ETCOc measurements we obtained within 12 hours of birth and at 48 to 72 hours after birth did not differ significantly (p >0.05) between the 17 neonates found to be G6PD deficient and the 68 infants with normal G6PD activity (Table). The maximal total serum bilirubin levels were also similar in both groups. The mean _+ SD ETCOc was not significantly different for the four G6PD-deficient infants who had hyperbilirubinemia (>220 gmol/L) compared with the 11 jaundiced infants in the control group: 2.1 _+ 0.4 ~I/L and 2.4 -+ 1.0 ~I/L, respectively, within 12 hours of birth; and 2.1 -+ 1.1/al/L and 1.9 _+ 0.8 gl/L, respectively, at48 to 72 hours after birth.
DISCUSSION Our results indicate that marked hemolysis in term healthy newborn infants with Mediterranean-type G6PD deficiency is not sufficient to explain hyperbilirubinemiain this cohort. This is in agreement with previous studies6, 7 based on hematologic indexes that could not demonstrate evidence for acute hemolysis in G6PD-deficient infants. This study used a much more sensitive method for determining erythrocyte cell breakdown,5 and yet only a slightly increased erythrocyte destruction was suggested. Importantly, the pattern in bilirubin production after birth differed between the control infants and G6PD-deficient infants; a higher ETCOc persisted in the latter group. This lack of decrease in postnatal bilirubin production suggests a pathologic circumstance such as hemolysis (although to a limited degree).
806
Seidman et al.
A recent study from Nigeria8 concluded, in contrast to our findings, that hemolysis may be the overwhelming cause of neonatal hyperbilirubinemia in G6PD-deficient neonates. This conclusion was based on the correlation found between elevated carboxyhemoglobin levels and bilirubin-related morbidity and mortality rates. The discrepancy in findings may be attributed to a number of factors. First, the Nigerian G6PD variant may be different from the Mediterranean variant. Thus, although an increased incidence of hyperbilirubinemia has been reported in Israeli infants, this was not associated with the need for exchange transfusions or the occurrence of kemicterus.7' 9 Second, an environmental factor exposing the infants to oxidant injury, such as infection or the application of dyes to the umbilicus and skin, may affect the Nigerian population. Nutrition may also account for these differences. Our study did not reveal a significant difference in rates of hyperbilirubinemiain the G6PD-deficient neonates (22.2%) and in the control infants (14.5%). This observation may be attributed to the significantly lower rate of breast-feeding in the study population. However, it is consistent with previous reports 1° showing that neonatal jaundice in G6PD-deficient black Americans was not associated with the high morbidity rates and kemicterus found in G6PD-deficient infants born in Africa. l° It is not clear whether elevated ETCOc (i.e., a more significant degree of hemolysis) would have been demonstrated had our study group included additional jaundiced infants. However, the severity of G6PD deficiency does not correlate with the severity of jaundice, l° Additional study is therefore needed to elucidate the factors that lead to the development of jaundice in G6PD-deficient infants. We conclude that, on the basis of breath ETCOc measurements, an increased bilirubinproduction rate is not sufficient to explain hyperbilirubinemia in healthy G6PD-deficient newborn infants in the transitional period.
The Journal of Pediatrics November 1995
The Baby's Breath carbon monoxide analyzer, described in U.S. Patent No. 5,293,875, and necessary supplies were kindly provided by Natus Medical, Inc., San Carlos, Calif. REFERENCES
1. Beutler E. Glucose-6-phosphatedehydrogenase deficiency. N Engl J Med 1991;324:169-74. 2. Smith DW, Inguillo D, Martin D, Vreman HJ, Cohen RS, Stevenson DK. Use of noninvasive tests to predict significant jaundice in full-term infants: preliminary studies. Pediatrics 1985;75:278-80. 3. Smith DW, Hopper AO, Shahin SM, et al. Neonatal bilirubin production estimated from "end-tidal" carbon monoxide concentration. J Pediatr Gastroenterol Nutr 1984;3:77-80. 4. RamotB, Ben-BassatI, Shchory M. New glucose-6-phosphate dehydrogenasevariants observed in Israel and their association with congenital nonspherocytichemolytic disease. J Lab Clin Med 1969;74:895-901. 5. Vreman HJ, Baxter L, Stone RT, StevensonDK. Evaluation of an automated end-tidal carbon monoxide instrument for infant breath analysis [Abstract]. J Invest Med 1995;43:174A. 6. Meloni T, Costa S, Cutillo S. Haptoglobin,hemopexin, hemoglobin and hematocritin newbornswith erythrocyteof glucose6-phosphate dehydrogenase deficiency. Acta Haematol 1975; 54:284-8. 7. Kaplan M, Abramov A. Neonatal hyperbilirubinemia associated with glucose-6-phosphate dehydrogenase deficiency in Sephardic-Jewish neonates: incidence, severity, and the effect of phototherapy. Pediatrics 1992;90:401-5. 8. Slusher TM, Vreman HJ, McLaren DW, Lewison LJ, Brown AK, Stevenson DK. Glucose-6-phosphatedehydrogenase deficiency and carboxyhemoglobin concentrations associated with bilimbin-relatedmorbidity and death in Nigerian infants. Pediatrics 1995;126:102-8. 9. Ashkenazi S, Mimouni F, Merlob P, Reisner SH. Neonatal bilirubin levels and glucose-6-phosphate dehydrogenase deficiency in preterm and low birth weight neonates in Israel. Isr J Med Sci 1983;19:1056-8. 10. Karaylcin G, Acs H, Lanzkowski P. G-6-PD deficiency and hyperbilirubinemiain black American full-term neonates. NY State J Med 1979;79:22-3.
Seidman et al.
The Journal of Pediatrics November 1995
Role of hemolysis in neonatal jaundice associated with glucose-6-phosphate dehydrogenase deficiency Daniel S. Seidman, MD, Michal Shiloh, MD, David K, Stevenson, ME), Hendrik J. Vreman, PhD, Ido Paz, MD, and Rena Gale, MD From the Department of Neonatology, Bikur-Cholim Hospital, Jerusalem; the Department of Obstetrics and Gynecology, Sheba Medical Center, TeI-Hashomer, Israel; and the Department of Pediatrics, Stanford University School of Medicine, Stanford, California
End-tidal carbon monoxide was measured in 108 newborn infants who had been screened for glucose-6-phosphate dehydrogenase (G6PD) deficiency. The mean ± SD end-tidal carbon monoxide did not differ significantly between the G6PD-deficient and the normal neonates, 2.1 ± 0.6 IJI/L and 2.0 ± 0.5 IJI/L, respectively, within 12 hours of birth and 1.9 ± 1.4 IJI/L and 1.5 ± 0.7 IJI/L, respectively, at 48 to 72 hours after birth. On the basis of these measurements, hemolysis is not a sufficient explanation for jaundice in G6PD-deficient newborn infants in the transitional period. (J PEDIATR1995;127:804-6)
The pathophysiologic basis for excessive jaundice in infants with glucose-6-phosphate dehydrogenase deficiency has not been established. Increased erythrocyte breakdown and reduced glucuronidation of bilirubin caused by defective G6PD activity in the hepatocyte may have a role in the pathogenesis. 1 Measuring bilirubin production in infants with G6PD deficiency may allow us to determine the role of hemolysis in the pathophysiology of the hyperbilirubinemia. The degradation of heme through the heine oxygenase pathway results in equimolar production of bilirubin and carbon monoxide, which is subsequently expelled during breathing. 2 End-tidal breath CO correlates with bilirubin production, and can therefore be used in its measurement. 3 This study examined whether early hemolysis, as reflected by increased bilirubin production, is more frequent or severe in G6PD-deficient infants. METHODS We prospectively screened for G6PD deficiency 108 term infants who were physically appropriate for their gestational ages and who were born between August and DecemSupported by National Institute of Child Health and Human Development grants HD 14426 and RR00070 and by the Mary L. Johnson Research Fund. Reprint requests: Daniel S. Seidman, MD, Division of Neonatology, Department of Pediatrics, Room $226, M/C 5119, Stanford University School of Medicine, Stanford, CA 94305-5119. Submitted for publication May 1, 1995; accepted June 23, 1995. Copyright © 1995 by Mosby-Year Book, Inc. 0022-3476/95/$5.00 + 0 9/24/67323
ber 1993 to mothers from areas known to have high incidences of G6PD deficiency (i.e., Kurdistan, Iraq, Iran, Syria, Turkey, and Buchara). 4 Exclusion criteria included exposure to any known hemolytic agent, blood group incompatibility, insmmaental delivery, neonatal asphyxia, congenital malformations, fever or any evidence of an infectious disease, cephalohematoma or signs of bruising, or maternal illness such as diabetes mellitus or preeclampsia. Maternal informed consent was obtained in all cases in accordance with the institution's human investigation review board. I
ETCOc End-tidal breath carbon monoxide corrected for inhaled carbon monoxide Glucose-6-phosphatedehydrogenase G6PD
All neonates had three consecutive end-tidal carbon monoxide measurements performed within 12 hours of birth and repeated at 48 to 72 hours of life. We performed end-tidal carbon monoxide measurements with a portable automated ETCOc analyzer (Baby's Breath carbon monoxide analyzer, Natus Medical Inc., San Carlos, Calif.).5 We collected and simultaneously analyzed triplicate samples of breath exhaled from the quietly resting (not crying) infants via a 5F catheter placed 2 cm into the anterior nares. We calculated the average error for ETCOc by subtracting the minimal value from the maximal value and dividing by the mean of the three measurements. The average margin of error for the triplicate ETCOc measurements was 44%. Blood group, direct Coombs test, and hematocrit and reticulocyte counts were determined by standard techniques.
The Journal of Pediatrics Volume 127, Number 5
Seidman et al.
805
Table. ETCOc and serum bilirubin measurements in healthy term infants according to maternal smoking and G6PD status Serum total bilirubin (maximum level)
ETCOc (IJI/L) Subjects (n)t Smoking mother (14) (range) Nonsmokingmother (85) (range) G6PD deficiency(17):~(range) G6PD normal (68)$ (range)
<12 hr
48-72 hr
(IJmol/U)
6.9 -+ 7.3§ (1.7-30.7) 2.0 -+ 0.6 (1.1-4.9) 2.1 -+ 0.6 (1.3-3.5) 2.0 -+ 0.5 (1.1-4.9)
1.4 -+ 0.5 (0.7-2.3) 1.5 -+ 0.7 (0.3-3.8) 1.9 -+ 1.4 (0.5-3.8) 1.5 -+ 0.7 (0.3-3.2)
140 -+45 (60-269) 176 -+ 71 (26-339) 179 -+ 78 (41-339) 173 -+ 70 (26-275)
Values are mean_+SD, with range in parentheses *To convertto milligramsper deciliter,divideby 17.1. "~Excludinginfantswith ABO blood groupincompatibility. 5/Infantsof smokingmothers were excluded. §p <0.03, comparedwith those infantswith a nonsmokingmother.
We measured serum total bilirnbin by determining optical absorbance at 460 nm, with correction for hemoglobin contamination at 550 nm (Unistat Bilirubinometer, AO Scientific Instruments, Buffalo, N.Y.). We performed screening for G6PD activity in erythrocytes using kit No. 400K (Sigma Chemical Co., St. Louis, Mo.), which is a colorimetric method based on the reduction of the blue dye dichlorophenol indophenol to a colorless state by reduced nicotinamideadenine dinucleotide phosphate formed in the presence of G6PD. Statistical analysis was performed with a nonpaired Student t test for continuous variables and with chi-square analysis or the Fisher Exact Test for dichotomous data.
RESULTS Of the 108 infants included in our study, 22 were G6PD deficient. The G6PD-deficient neonates did not differ significantly from the control group infants in maternal age, parity, and smoking; use of epidural anesthesia; frequency of oxytocin augmentation; cesarean section rate; 1- and 5-minute Apgar scores; gestational age; or birth weight. Breast-feeding was significantly less frequent among mothers of G6PD-deficient neonates, 59.1% versus 87.2% of control group mothers (p <0.006). Neonatal jaundice caused by ABO incompatibility was identified in nine infants (one study infant and eight control infants); the mean ETCOc level was 4.0 -+ 1.1 p1/L (range, 3.1 to 5.9 Ial/L) within 12 hours of birth, reflecting the hemolytic component of the hyperbilirubinemia. All were treated by phototherapy. In the two patients who required exchange transfusions, the ETCOc concentrations were 5.1 and 5.9/al/L, respectively. These values decreased to 0.55 and 2.2/al/L after exchange transfusions. These nine infants were excluded from the analysis.
Because ETCOc concentrations are highly sensitive to the effects of maternal smoking, those infants whose mothers smoked (4 study infants and 10 control infants) were also excluded from the analysis. However, their ETCOc values at 48 to 72 hours were no longer affected by maternal smoking (Table). The mean ETCOc measurements we obtained within 12 hours of birth and at 48 to 72 hours after birth did not differ significantly (p >0.05) between the 17 neonates found to be G6PD deficient and the 68 infants with normal G6PD activity (Table). The maximal total serum bilirubin levels were also similar in both groups. The mean _+ SD ETCOc was not significantly different for the four G6PD-deficient infants who had hyperbilirubinemia (>220 gmol/L) compared with the 11 jaundiced infants in the control group: 2.1 _+ 0.4 ~I/L and 2.4 -+ 1.0 ~I/L, respectively, within 12 hours of birth; and 2.1 -+ 1.1/al/L and 1.9 _+ 0.8 gl/L, respectively, at48 to 72 hours after birth.
DISCUSSION Our results indicate that marked hemolysis in term healthy newborn infants with Mediterranean-type G6PD deficiency is not sufficient to explain hyperbilirubinemiain this cohort. This is in agreement with previous studies6, 7 based on hematologic indexes that could not demonstrate evidence for acute hemolysis in G6PD-deficient infants. This study used a much more sensitive method for determining erythrocyte cell breakdown,5 and yet only a slightly increased erythrocyte destruction was suggested. Importantly, the pattern in bilirubin production after birth differed between the control infants and G6PD-deficient infants; a higher ETCOc persisted in the latter group. This lack of decrease in postnatal bilirubin production suggests a pathologic circumstance such as hemolysis (although to a limited degree).
806
Seidman et al.
A recent study from Nigeria8 concluded, in contrast to our findings, that hemolysis may be the overwhelming cause of neonatal hyperbilirubinemia in G6PD-deficient neonates. This conclusion was based on the correlation found between elevated carboxyhemoglobin levels and bilirubin-related morbidity and mortality rates. The discrepancy in findings may be attributed to a number of factors. First, the Nigerian G6PD variant may be different from the Mediterranean variant. Thus, although an increased incidence of hyperbilirubinemia has been reported in Israeli infants, this was not associated with the need for exchange transfusions or the occurrence of kemicterus.7' 9 Second, an environmental factor exposing the infants to oxidant injury, such as infection or the application of dyes to the umbilicus and skin, may affect the Nigerian population. Nutrition may also account for these differences. Our study did not reveal a significant difference in rates of hyperbilirubinemiain the G6PD-deficient neonates (22.2%) and in the control infants (14.5%). This observation may be attributed to the significantly lower rate of breast-feeding in the study population. However, it is consistent with previous reports 1° showing that neonatal jaundice in G6PD-deficient black Americans was not associated with the high morbidity rates and kemicterus found in G6PD-deficient infants born in Africa. l° It is not clear whether elevated ETCOc (i.e., a more significant degree of hemolysis) would have been demonstrated had our study group included additional jaundiced infants. However, the severity of G6PD deficiency does not correlate with the severity of jaundice, l° Additional study is therefore needed to elucidate the factors that lead to the development of jaundice in G6PD-deficient infants. We conclude that, on the basis of breath ETCOc measurements, an increased bilirubinproduction rate is not sufficient to explain hyperbilirubinemia in healthy G6PD-deficient newborn infants in the transitional period.
The Journal of Pediatrics November 1995
The Baby's Breath carbon monoxide analyzer, described in U.S. Patent No. 5,293,875, and necessary supplies were kindly provided by Natus Medical, Inc., San Carlos, Calif. REFERENCES
1. Beutler E. Glucose-6-phosphatedehydrogenase deficiency. N Engl J Med 1991;324:169-74. 2. Smith DW, Inguillo D, Martin D, Vreman HJ, Cohen RS, Stevenson DK. Use of noninvasive tests to predict significant jaundice in full-term infants: preliminary studies. Pediatrics 1985;75:278-80. 3. Smith DW, Hopper AO, Shahin SM, et al. Neonatal bilirubin production estimated from "end-tidal" carbon monoxide concentration. J Pediatr Gastroenterol Nutr 1984;3:77-80. 4. RamotB, Ben-BassatI, Shchory M. New glucose-6-phosphate dehydrogenasevariants observed in Israel and their association with congenital nonspherocytichemolytic disease. J Lab Clin Med 1969;74:895-901. 5. Vreman HJ, Baxter L, Stone RT, StevensonDK. Evaluation of an automated end-tidal carbon monoxide instrument for infant breath analysis [Abstract]. J Invest Med 1995;43:174A. 6. Meloni T, Costa S, Cutillo S. Haptoglobin,hemopexin, hemoglobin and hematocritin newbornswith erythrocyteof glucose6-phosphate dehydrogenase deficiency. Acta Haematol 1975; 54:284-8. 7. Kaplan M, Abramov A. Neonatal hyperbilirubinemia associated with glucose-6-phosphate dehydrogenase deficiency in Sephardic-Jewish neonates: incidence, severity, and the effect of phototherapy. Pediatrics 1992;90:401-5. 8. Slusher TM, Vreman HJ, McLaren DW, Lewison LJ, Brown AK, Stevenson DK. Glucose-6-phosphatedehydrogenase deficiency and carboxyhemoglobin concentrations associated with bilimbin-relatedmorbidity and death in Nigerian infants. Pediatrics 1995;126:102-8. 9. Ashkenazi S, Mimouni F, Merlob P, Reisner SH. Neonatal bilirubin levels and glucose-6-phosphate dehydrogenase deficiency in preterm and low birth weight neonates in Israel. Isr J Med Sci 1983;19:1056-8. 10. Karaylcin G, Acs H, Lanzkowski P. G-6-PD deficiency and hyperbilirubinemiain black American full-term neonates. NY State J Med 1979;79:22-3.