- Email: [email protected]
Antenatal maternal administration of phenobarbital for the prevention of exchange transfusion in neonates with hemolytic disease of the fetus and newborn
American Journal of Obstetrics and Gynecology (2005) 192, 478–82
www.ajog.org
Antenatal maternal administration of phenobarbital for the prevention of exchange transfusion in neonates with hemolytic disease of the fetus and newborn Thomas N. Trevett, Jr, MD,a,* Karen Dorman, RN, MS,a Georgine Lamvu, MD, MPH,b Kenneth J. Moise, Jr, MDa Divisions of Maternal-Fetal Medicinea and Advanced Laparoscopy and Gynecologic Surgery,b Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, NC Received for publication May 5, 2004; revised July 22, 2004; accepted August 17, 2004
KEY WORDS Red cell alloimmunization Fetus Hemolytic disease of the newborn infant Phenobarbital Bilirubin encephalopathy Exchange transfusion
Objective: Hemolytic disease of the fetus and newborn infant (HDFN) can be associated with bilirubin encephalopathy, which is usually averted through neonatal exchange transfusions (EXT). However, EXT can be associated with significant procedure-related morbidity. We hypothesized that maternal oral administration of phenobarbital (PB) to women with HDFN would reduce the rate of EXT. Study design: Cases of HDFN from January 1985 to June 2003 were reviewed. All patients who underwent serial intrauterine transfusions (IUTs) for red cell alloimmunization were included. Patients were offered oral phenobarbital (30 mg 3 times a day) after their last IUT in an effort to enhance fetal hepatic maturity. Variables studied included gestational age and hemoglobin multiples of the median at first transfusion and delivery, peak neonatal bilirubin, need for exchange transfusion, and maternal PB administration. Multivariate regression analysis was applied to determine relative risks and 95% CIs. Results: Seventy-one patients met study criteria; 29% of the neonates underwent EXT. The use of antenatal PB was associated with a decreased incidence of EXT, 9% versus 52% (P ! .01). After controlling for confounding variables, the relative risk for EXT after antenatal PB administration was 0.23 (95% CI: 0.06-0.76). Conclusion: Maternal administration of PB reduces the need for neonatal EXTs in HDFN. Ó 2005 Elsevier Inc. All rights reserved.
Hemolytic disease of the fetus and newborn infant (HDFN) can lead to significant perinatal morbidity, including encephalopathy secondary to hyperbilirubinemia. Bilirubin is liberated from the heme molecule as * Reprint requests: Thomas N. Trevett, Jr, MD, 214 MacNider Hall, CB 7516, Chapel Hill, NC 27599. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.08.016
a result of immune-mediated hemolysis. It is transported to the hepatocyte via albumin where it then undergoes glucoronidation by a family of enzymes known as uridine-diphospho-glucuronosyltransferases (UGT). In humans, the major enzyme in this pathway is bilirubinUDP-glucuronosyltransferase (UGT1A1). After conjugation, the bilirubin diglucuronide, which is highly water soluble, is then actively excreted into bile and
Trevett et al eliminated by way of the urinary and gastrointestinal tracts. The rate-limiting step in this process is the concentration of UGT1A1.1 Phenobarbital (PB) has been shown to enhance the capability of the neonatal liver to conjugate and eliminate bilirubin. Studies by Brown and Zeulzer2 and Gartner and Arias3 in guinea pigs, and by Catz and Yaffe4 in mice provide evidence that PB given to pregnant animals can stimulate the induction of the glucuronyl transferase enzymatic pathway in their offspring. The first human studies by Trolle5 indicated a similar reduction in neonatal jaundice in humans from 22% to 5%. PB has been shown to increase the expression of the UGT1A1, a mechanism that has only recently been elucidated. Sugatani et al6 discovered a 290 base pair (bp) enhancer sequence on the gene for UGT1A1 that binds PB, leading to increased production of the enzyme. After birth, some infants affected by HDFN are unable to manage the increased production of bilirubin, leading to unconjugated hyperbilirubinemia. If untreated, this can lead to kernicterus with long-term neurologic consequences and even death. Phototherapy is usually the first line of treatment. However, if unsuccessful, neonates must undergo an exchange transfusion (EXT) to reduce the bilirubin load and the concentration of the maternal anti-red cell antibody. In a recent study, the incidence of major morbidities caused by this procedure, including transfusion reactions, line sepsis, and death after EXT, was 5%.7 Therefore, a reduction in the need for EXTs would further improve neonatal outcomes for HDFN. We hypothesized that maternal administration of PB before delivery in women with red cell alloimmunization undergoing serial intrauterine transfusions (IUTs) before delivery would induce neonatal hepatic maturity and reduce the need for EXT.
Material and methods Study design This was a retrospective case-control study of all women from January 1985 to June 2003 with documented HDFN undergoing cordocentesis and IUTs at the University of North Carolina School of Medicine in Chapel Hill, NC, and the Baylor College of Medicine in Houston, Texas. Before initiating this study, we obtained consent from the University of North Carolina institutional review board to review the medical records of the patients and neonates of interest. Cases were identified through a computerized database maintained by 1 of the authors (K.M.). All women included in this sample had either the presence of an anti-red cell antibody associated with HDFN that titered above the critical threshold or a history of a previous pregnancy affected by HDFN.
479 Table I
Maternal and fetal characteristics
Variable Maternal age (y) Gravidity* Number of IUTs* Gestational age at first IUT (wks) Gestational age at delivery (wks) Hemoglobin at first IUT (MoMs) Hemoglobin at delivery (MoMs) Hydrops at first IUT Antibody type D Kell D, C D, C, E D, C, M D, E c
29.1 G 5.1 4 (1-8) 3 (1-6) 25.9 G 3.4 34.4 G 4.2 0.58 G 0.24 0.97 G 0.26 41% (N, %) 49 (69) 7 (10) 10 (14) 2 (3) 1 (1) 1 (1) 1 (1)
All parameters expressed as means G SD unless otherwise indicated. * Expressed as median and range.
Pregnancies at risk for HDFN were followed intensively with surveillance for the development of fetal anemia. Before the year 2000, patients underwent serial amniocenteses for determination of their DOD450, using graphs created by Liley et al8 to determine the severity of anemia. After 2000, patients were monitored with serial middle cerebral artery (MCA) peak systolic velocities (PSV) measured weekly.9 In the first affected pregnancy, weekly MCA PSV measurements were initiated when a critical titer for the putative anti-red cell antibody was detected; in previously affected pregnancies, the MCA PSV was measured starting at 18 weeks’ gestation. Fetuses with an MCA PSV greater than 1.5 multiples of the median (MoMs) underwent cordocentesis with combined intravascular and intraperitoneal transfusion if necessary. After documentation of subsequent bone marrow suppression through fetal cell stains at the time of IUT, the interval of transfusion was lengthened to 3 weeks. IUTs were repeated serially until the 35th week of gestation. After the last IUT, the patient was offered PB, 30 mg orally 3 times a day for 10 days. Delivery was planned after the treatment period. The neonate was monitored intensively in the nursery for signs of anemia and hyperbilirubinemia. Standard institutional-specific protocols that were based on the current guidelines at the time were used to determine when to initiate phototherapy or EXT.10-12 Medical records were extracted for the following parameters: estimated gestational age at the time of first IUT, fetal hematocrit at first IUT, total number of IUTs, gestational age at delivery, birth weight at delivery, Apgar scores, maternal PB use, levels of cord hematocrit and percentage of fetal cells at delivery, peak neonatal total and direct bilirubin, neonatal need for
480 Table II
Trevett et al Bivariate analysis of study variables on need for neonatal exchange transfusion
Variable
No exchange
Exchange
P value
Number EGA at first IUT (wks) Hemoglobin at first IUT (MoMs) Hemoglobin at delivery (MoMs) Time from last IUT to delivery (d) Delivery EGA (wks) 1-min Apgar* 5-min Apgar* Neonatal peak total bilirubin (mg/dL) Neonatal direct bilirubin (mg/dL) Neonatal cord total bilirubin (mg/dL)
50 26.4 G 3.5 0.57 G 0.27 0.96 G 0.26 13.1 G 16.1 35.4 G 3.1 7 (0-9) 9 (1-10) 12.9 G 8.8 0.67G .08 6.2 G 3.4
21 25.7 G 3.4 0.62 G 0.21 1.0 G 0.29 19.5 G 11.2 33.0 G 3.9 6 (0-9) 9 (1-10) 19.4 G 4.7 0.71 G 0.86 7.6 G 7.07
d .22 .80 .71 .13 ! .01 .94 .97 ! .01 .83 .39
EGA, Estimated gestational age. All parameters expressed as mean G SD unless otherwise indicated. * Expressed as median and range.
Table III Results of binomial regression analysis, adjusting for all possible confounding variables on the relationship between the maternal administration of PB and the need for neonatal EXT Risk ratio for the association between EXT and administration of PB
Risk ratio
95% CIs
Crude Adjusted for: gestational age at delivery Adjusted for: gestational age at delivery, neonatal peak total bilirubin
0.24 0.18 0.23
0.08-0.75 0.05-0.59 0.06-0.76
blood transfusions or EXTs, and neonatal survival. Hematocrit was converted to approximate hemoglobin by dividing by a factor of 3. With the use of the equation developed by Mari et al,9 e(2.84–8.55/GA), median hemoglobin for each gestational age was calculated. Multiples of the median hemoglobin values were then calculated and used in the analyses. A univariate analysis of the data was conducted to identify frequencies, means, standard deviations, ranges, missing data, and outliers. A bivariate analysis was then undertaken to look for crude associations between the prenatal PB exposure and the covariates and for associations between the outcome and the covariates. Lastly, binomial regression analysis was used to model the effect of PB administration on EXTs. Gestational age at delivery and neonatal peak total bilirubin among other variables were selected as possible confounders and introduced into the regression models to estimate the final effect of PB on transfusions. A P value of !.05 was considered statistically significant. Where appropriate, 95% CIs for odds ratios were calculated.
Results Seventy-one women and neonates who underwent serial IUTs over the study period were identified. The
maternal and fetal characteristics of this cohort are shown in Table I. Thirty-three women received PB before delivery; all of their neonates survived. Thirtyeight women did not receive PB. The most common reason for not receiving PB was nonreassuring antenatal testing requiring delivery, followed by mature fetal lung profile at the last IUT. In this group, the overall perinatal survival was 76%; there were 9 neonatal deaths. There were 21 (29%) neonates who required EXTs for treatment of severe hyperbilirubinemia (median 1, range 1-3). Of the 33 women who received PB before delivery, only 3 (9%) of their neonates required an exchange transfusion. Of the 31 surviving infants that were delivered of mothers who did not receive PB, 16 (52%) required an EXT (P ! .01). A bivariate analysis of the data was performed to demonstrate which variables other than maternal PB had an impact on the primary outcome, need for EXT. These results are shown in Table II. The only variables with a statistically significant difference between the groups of neonates were the neonatal peak bilirubin (P ! .01) and the gestational age at delivery (P ! .01). A binomial regression analysis was undertaken to control for these variables. After adjusting for the gestational age at delivery and neonatal peak total bilirubin, administration of PB decreased the relative risk of the need for EXT to 0.23 (95% CI: 0.06-0.76). Table III represents the results of the multivariate regression analysis.
Comment The cohort of women in the current study represents a population with severe HDFN as evidenced by the early mean gestational age and mean fetal hematocrit at the time of first transfusion, total number of IUTs, and
Trevett et al the percentage of fetuses with hydrops at the time of the first transfusion. Despite the severity of disease, the perinatal outcome was excellent with an overall survival of 87%. The majority of the neonates who did require an EXT only required 1 or 2 procedures. The 29% incidence of EXT in the current study is consistent with previous work by Filbey et al8 who noted an incidence of 29% in neonates in Sweden who were affected by HDFN. Maternal PB late in gestation was, however, associated with more than a 75% reduction in the need for EXT. The major limitation of our finding is the confounding influence of preterm delivery on the need for an EXT. Most of the preterm deliveries that occurred were undertaken on an emergent basis and thus the patients were unable to receive PB. Premature infants are at greater risk of requiring an EXT for hyperbilirubinemia because of their immaturity of the hepatic microsomal enymatic conjugation pathways. Further, the level of unconjugated bilirubin at which the premature brain will be damaged is lower than the term brain because of the immaturity of the blood-brain barrier.13 Therefore, providers are more likely to initiate EXT when dealing with the premature infant. The mean gestational age at delivery of the pregnancies that underwent EXTs in our study was 2.6 weeks less than in pregnancies in which the infant did not require EXT, a statistically significant difference. For this reason, we controlled for gestational age in our multivariate analysis. After this analysis, the protective effect of maternal PB persisted. The dosing regimen of 30 mg 3 times a day of oral PB used in our study was chosen on the basis of the work by Halpin et al14 who performed a randomized controlled trial of providing PB or placebo to healthy pregnant women starting at 32 weeks of gestation in an effort to decrease hyperbilirubinemia. These authors showed a significant reduction in the mean neonatal unconjugated bilirubin from 8.5 to 6.4 mg/dL in the women treated with 1 g of PB before delivery. They also showed that the incidence of infants with levels exceeding 10 mg/dL at 96 hours of life was significantly reduced from 24.6% in controls to 7.3% in treated individuals. In our cohort, each woman received PB for a period of 7 to 10 days, with the maximum cumulative amount being 900 mg. The decision to use any drug during pregnancy should include consideration of perinatal safety. Multiple investigators have evaluated fetal exposure to PB during the latter stages of pregnancy. PB is known to freely cross the placenta throughout pregnancy.15-17 Usage of PB has been investigated as a means to reduce the occurrence and severity of neonatal intracranial hemorrhage in preterm infants. These studies have not shown a significant decrease in rates of intraventicular hemorrhage, however, in follow-up studies at both 2 and 7 years of age, Thorp et al18,19 found no difference in mean intelligence, achievement, and behavioral scores.
481 Concerns have been raised over the administration of a barbituate, a sedative, to the fetus. However, there have been no findings to date in multiple trials using antenatal PB regarding an increased incidence of neonatal depression.5,14,20,21 In conclusion, this retrospective study suggests that a brief course of maternal oral PB before delivery may reduce the need for neonatal EXT. Further study in a randomized controlled trial is necessary to confirm these results.
References 1. Yamamoto K, Sato H, Fujiyama Y, Doida Y, Bamba T. Contribution of two missense mutations (G71R and Y486D) of the bilirubin UDP glycosyltransferase (UGT1A1) gene to phenotypes of Gilbert’s syndrome and Crigler-Najjar syndrome type II. Biochim Biophys Acta 1998;1406:267-73. 2. Brown AK, Zuelzer WW. Studies on the neonatal development of the glucuronide conjugating system. J Clin Invest 1958;37: 332-40. 3. Gartner LM, Arias IM. The transfer of bilirubin from blood to bile in the neonatal guinea pig. Pediatr Res 1969;3:171-80. 4. Catz C, Yaffe SJ. Barbiturate enhancement of bilirubin conjugation and excretion in young and adult animals. Pediatr Res 1968;2:361-70. 5. Trolle D. A possible drop in first-week-mortality rate for lowbirth-weight infants after phenobarbitone treatment. Lancet 1968;2:1123-4. 6. Sugatani J, Kojima H, Ueda A, Kakizaki S, Yoshinari K, Gong QH, et al. The phenobarbital response enhancer module in the human bilirubin UDP-glucuronosyltransferase UGT1A1 gene and regulation by the nuclear receptor CAR. Hepatology 2001;33:1232-8. 7. Keenan WJ, Novak KK, Sutherland JM, Bryla DA, Fetterly KL. Morbidity and mortality associated with exchange transfusion. Pediatrics 1985;75:417-21. 8. Liley AW. Liquor amnil analysis in the management of the pregnancy complicated by resus sensitization. Am J Obstet Gynecol 1961;82:1359-70. 9. Mari G, Deter RL, Carpenter RL, Rahman F, Zimmerman R, Moise KJ Jr, et al. Noninvasive diagnosis by Doppler ultrasonography of fetal anemia due to maternal red-cell alloimmunization. Collaborative Group for Doppler Assessment of the Blood Velocity in Anemic Fetuses. N Engl J Med 2000;342:9-14. 10. Practice parameter: management of hyperbilirubinemia in the healthy term newborn. American Academy of Pediatrics. Provisional Committee for Quality Improvement and Subcommittee on Hyperbilirubinemia. Pediatrics 1994;94:558-65. 11. Maisels MJ. Jaundice. In: Glenn Brown Avery MAF, Mary Glenn MacDonald, editors. Neonatology: pathophysiology and management of the newborn. Philadelphia: JB Lippincott; 1999. p. 765-819. 12. Maisels MJ, Watchko JF. Treatment of jaundice in low birthweight infants. Arch Dis Child Fetal Neonatal Ed 2003;88:F459-63. 13. Watchko JF, Oski FA. Kernicterus in preterm newborns: past, present, and future. Pediatrics 1992;90:707-15. 14. Halpin TF, Jones AR, Bishop HL, Lerner S. Prophylaxis of neonatal hyperbilirubinemia with phenobarbital. Obstet Gynecol 1972;40:85-90. 15. Melchior JC, Svensmark O, Trolle D. Placental transfer of phenobarbitone in epileptic women, and elimination in newborns. Lancet 1967;2:860-1.
482 16. Arpino C, Brescianini S, Robert E, Castilla EE, Cocchi G, Cornel MC, et al. Teratogenic effects of antiepileptic drugs: use of an International Database on Malformations and Drug Exposure (MADRE). Epilepsia 2000;41:1436-43. 17. De Carolis MP, Romagnoli C, Frezza S, D’Urzo E, Muzii U, Mezza A, et al. Placental transfer of phenobarbital: what is new? Dev Pharmacol Ther 1992;19:19-26. 18. Thorp JA, O’Connor M, Jones AM, Hoffman EL, Belden B. Does perinatal phenobarbital exposure affect developmental outcome at age 2? Am J Perinatol 1999;16:51-60.
Trevett et al 19. Thorp JA, O’Connor M, Belden B, Etzenhouser J, Hoffman EL, Jones PG. Effects of phenobarbital and multiple-dose corticosteroids on developmental outcome at age 7 years. Obstet Gynecol 2003;101:363-73. 20. Rayburn W, Donn S, Piehl E, Compton A. Antenatal phenobarbital and bilirubin metabolism in the very low birth weight infant. Am J Obstet Gynecol 1988;159:1491-3. 21. Shankaran S, Cepeda EE, Ilagan N, Mariona F, Hassan M, Bhatia R, et al. Antenatal phenobarbital for the prevention of neonatal intracerebral hemorrhage. Am J Obstet Gynecol 1986;154:53-7.
www.ajog.org
Antenatal maternal administration of phenobarbital for the prevention of exchange transfusion in neonates with hemolytic disease of the fetus and newborn Thomas N. Trevett, Jr, MD,a,* Karen Dorman, RN, MS,a Georgine Lamvu, MD, MPH,b Kenneth J. Moise, Jr, MDa Divisions of Maternal-Fetal Medicinea and Advanced Laparoscopy and Gynecologic Surgery,b Department of Obstetrics and Gynecology, University of North Carolina School of Medicine, Chapel Hill, NC Received for publication May 5, 2004; revised July 22, 2004; accepted August 17, 2004
KEY WORDS Red cell alloimmunization Fetus Hemolytic disease of the newborn infant Phenobarbital Bilirubin encephalopathy Exchange transfusion
Objective: Hemolytic disease of the fetus and newborn infant (HDFN) can be associated with bilirubin encephalopathy, which is usually averted through neonatal exchange transfusions (EXT). However, EXT can be associated with significant procedure-related morbidity. We hypothesized that maternal oral administration of phenobarbital (PB) to women with HDFN would reduce the rate of EXT. Study design: Cases of HDFN from January 1985 to June 2003 were reviewed. All patients who underwent serial intrauterine transfusions (IUTs) for red cell alloimmunization were included. Patients were offered oral phenobarbital (30 mg 3 times a day) after their last IUT in an effort to enhance fetal hepatic maturity. Variables studied included gestational age and hemoglobin multiples of the median at first transfusion and delivery, peak neonatal bilirubin, need for exchange transfusion, and maternal PB administration. Multivariate regression analysis was applied to determine relative risks and 95% CIs. Results: Seventy-one patients met study criteria; 29% of the neonates underwent EXT. The use of antenatal PB was associated with a decreased incidence of EXT, 9% versus 52% (P ! .01). After controlling for confounding variables, the relative risk for EXT after antenatal PB administration was 0.23 (95% CI: 0.06-0.76). Conclusion: Maternal administration of PB reduces the need for neonatal EXTs in HDFN. Ó 2005 Elsevier Inc. All rights reserved.
Hemolytic disease of the fetus and newborn infant (HDFN) can lead to significant perinatal morbidity, including encephalopathy secondary to hyperbilirubinemia. Bilirubin is liberated from the heme molecule as * Reprint requests: Thomas N. Trevett, Jr, MD, 214 MacNider Hall, CB 7516, Chapel Hill, NC 27599. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.08.016
a result of immune-mediated hemolysis. It is transported to the hepatocyte via albumin where it then undergoes glucoronidation by a family of enzymes known as uridine-diphospho-glucuronosyltransferases (UGT). In humans, the major enzyme in this pathway is bilirubinUDP-glucuronosyltransferase (UGT1A1). After conjugation, the bilirubin diglucuronide, which is highly water soluble, is then actively excreted into bile and
Trevett et al eliminated by way of the urinary and gastrointestinal tracts. The rate-limiting step in this process is the concentration of UGT1A1.1 Phenobarbital (PB) has been shown to enhance the capability of the neonatal liver to conjugate and eliminate bilirubin. Studies by Brown and Zeulzer2 and Gartner and Arias3 in guinea pigs, and by Catz and Yaffe4 in mice provide evidence that PB given to pregnant animals can stimulate the induction of the glucuronyl transferase enzymatic pathway in their offspring. The first human studies by Trolle5 indicated a similar reduction in neonatal jaundice in humans from 22% to 5%. PB has been shown to increase the expression of the UGT1A1, a mechanism that has only recently been elucidated. Sugatani et al6 discovered a 290 base pair (bp) enhancer sequence on the gene for UGT1A1 that binds PB, leading to increased production of the enzyme. After birth, some infants affected by HDFN are unable to manage the increased production of bilirubin, leading to unconjugated hyperbilirubinemia. If untreated, this can lead to kernicterus with long-term neurologic consequences and even death. Phototherapy is usually the first line of treatment. However, if unsuccessful, neonates must undergo an exchange transfusion (EXT) to reduce the bilirubin load and the concentration of the maternal anti-red cell antibody. In a recent study, the incidence of major morbidities caused by this procedure, including transfusion reactions, line sepsis, and death after EXT, was 5%.7 Therefore, a reduction in the need for EXTs would further improve neonatal outcomes for HDFN. We hypothesized that maternal administration of PB before delivery in women with red cell alloimmunization undergoing serial intrauterine transfusions (IUTs) before delivery would induce neonatal hepatic maturity and reduce the need for EXT.
Material and methods Study design This was a retrospective case-control study of all women from January 1985 to June 2003 with documented HDFN undergoing cordocentesis and IUTs at the University of North Carolina School of Medicine in Chapel Hill, NC, and the Baylor College of Medicine in Houston, Texas. Before initiating this study, we obtained consent from the University of North Carolina institutional review board to review the medical records of the patients and neonates of interest. Cases were identified through a computerized database maintained by 1 of the authors (K.M.). All women included in this sample had either the presence of an anti-red cell antibody associated with HDFN that titered above the critical threshold or a history of a previous pregnancy affected by HDFN.
479 Table I
Maternal and fetal characteristics
Variable Maternal age (y) Gravidity* Number of IUTs* Gestational age at first IUT (wks) Gestational age at delivery (wks) Hemoglobin at first IUT (MoMs) Hemoglobin at delivery (MoMs) Hydrops at first IUT Antibody type D Kell D, C D, C, E D, C, M D, E c
29.1 G 5.1 4 (1-8) 3 (1-6) 25.9 G 3.4 34.4 G 4.2 0.58 G 0.24 0.97 G 0.26 41% (N, %) 49 (69) 7 (10) 10 (14) 2 (3) 1 (1) 1 (1) 1 (1)
All parameters expressed as means G SD unless otherwise indicated. * Expressed as median and range.
Pregnancies at risk for HDFN were followed intensively with surveillance for the development of fetal anemia. Before the year 2000, patients underwent serial amniocenteses for determination of their DOD450, using graphs created by Liley et al8 to determine the severity of anemia. After 2000, patients were monitored with serial middle cerebral artery (MCA) peak systolic velocities (PSV) measured weekly.9 In the first affected pregnancy, weekly MCA PSV measurements were initiated when a critical titer for the putative anti-red cell antibody was detected; in previously affected pregnancies, the MCA PSV was measured starting at 18 weeks’ gestation. Fetuses with an MCA PSV greater than 1.5 multiples of the median (MoMs) underwent cordocentesis with combined intravascular and intraperitoneal transfusion if necessary. After documentation of subsequent bone marrow suppression through fetal cell stains at the time of IUT, the interval of transfusion was lengthened to 3 weeks. IUTs were repeated serially until the 35th week of gestation. After the last IUT, the patient was offered PB, 30 mg orally 3 times a day for 10 days. Delivery was planned after the treatment period. The neonate was monitored intensively in the nursery for signs of anemia and hyperbilirubinemia. Standard institutional-specific protocols that were based on the current guidelines at the time were used to determine when to initiate phototherapy or EXT.10-12 Medical records were extracted for the following parameters: estimated gestational age at the time of first IUT, fetal hematocrit at first IUT, total number of IUTs, gestational age at delivery, birth weight at delivery, Apgar scores, maternal PB use, levels of cord hematocrit and percentage of fetal cells at delivery, peak neonatal total and direct bilirubin, neonatal need for
480 Table II
Trevett et al Bivariate analysis of study variables on need for neonatal exchange transfusion
Variable
No exchange
Exchange
P value
Number EGA at first IUT (wks) Hemoglobin at first IUT (MoMs) Hemoglobin at delivery (MoMs) Time from last IUT to delivery (d) Delivery EGA (wks) 1-min Apgar* 5-min Apgar* Neonatal peak total bilirubin (mg/dL) Neonatal direct bilirubin (mg/dL) Neonatal cord total bilirubin (mg/dL)
50 26.4 G 3.5 0.57 G 0.27 0.96 G 0.26 13.1 G 16.1 35.4 G 3.1 7 (0-9) 9 (1-10) 12.9 G 8.8 0.67G .08 6.2 G 3.4
21 25.7 G 3.4 0.62 G 0.21 1.0 G 0.29 19.5 G 11.2 33.0 G 3.9 6 (0-9) 9 (1-10) 19.4 G 4.7 0.71 G 0.86 7.6 G 7.07
d .22 .80 .71 .13 ! .01 .94 .97 ! .01 .83 .39
EGA, Estimated gestational age. All parameters expressed as mean G SD unless otherwise indicated. * Expressed as median and range.
Table III Results of binomial regression analysis, adjusting for all possible confounding variables on the relationship between the maternal administration of PB and the need for neonatal EXT Risk ratio for the association between EXT and administration of PB
Risk ratio
95% CIs
Crude Adjusted for: gestational age at delivery Adjusted for: gestational age at delivery, neonatal peak total bilirubin
0.24 0.18 0.23
0.08-0.75 0.05-0.59 0.06-0.76
blood transfusions or EXTs, and neonatal survival. Hematocrit was converted to approximate hemoglobin by dividing by a factor of 3. With the use of the equation developed by Mari et al,9 e(2.84–8.55/GA), median hemoglobin for each gestational age was calculated. Multiples of the median hemoglobin values were then calculated and used in the analyses. A univariate analysis of the data was conducted to identify frequencies, means, standard deviations, ranges, missing data, and outliers. A bivariate analysis was then undertaken to look for crude associations between the prenatal PB exposure and the covariates and for associations between the outcome and the covariates. Lastly, binomial regression analysis was used to model the effect of PB administration on EXTs. Gestational age at delivery and neonatal peak total bilirubin among other variables were selected as possible confounders and introduced into the regression models to estimate the final effect of PB on transfusions. A P value of !.05 was considered statistically significant. Where appropriate, 95% CIs for odds ratios were calculated.
Results Seventy-one women and neonates who underwent serial IUTs over the study period were identified. The
maternal and fetal characteristics of this cohort are shown in Table I. Thirty-three women received PB before delivery; all of their neonates survived. Thirtyeight women did not receive PB. The most common reason for not receiving PB was nonreassuring antenatal testing requiring delivery, followed by mature fetal lung profile at the last IUT. In this group, the overall perinatal survival was 76%; there were 9 neonatal deaths. There were 21 (29%) neonates who required EXTs for treatment of severe hyperbilirubinemia (median 1, range 1-3). Of the 33 women who received PB before delivery, only 3 (9%) of their neonates required an exchange transfusion. Of the 31 surviving infants that were delivered of mothers who did not receive PB, 16 (52%) required an EXT (P ! .01). A bivariate analysis of the data was performed to demonstrate which variables other than maternal PB had an impact on the primary outcome, need for EXT. These results are shown in Table II. The only variables with a statistically significant difference between the groups of neonates were the neonatal peak bilirubin (P ! .01) and the gestational age at delivery (P ! .01). A binomial regression analysis was undertaken to control for these variables. After adjusting for the gestational age at delivery and neonatal peak total bilirubin, administration of PB decreased the relative risk of the need for EXT to 0.23 (95% CI: 0.06-0.76). Table III represents the results of the multivariate regression analysis.
Comment The cohort of women in the current study represents a population with severe HDFN as evidenced by the early mean gestational age and mean fetal hematocrit at the time of first transfusion, total number of IUTs, and
Trevett et al the percentage of fetuses with hydrops at the time of the first transfusion. Despite the severity of disease, the perinatal outcome was excellent with an overall survival of 87%. The majority of the neonates who did require an EXT only required 1 or 2 procedures. The 29% incidence of EXT in the current study is consistent with previous work by Filbey et al8 who noted an incidence of 29% in neonates in Sweden who were affected by HDFN. Maternal PB late in gestation was, however, associated with more than a 75% reduction in the need for EXT. The major limitation of our finding is the confounding influence of preterm delivery on the need for an EXT. Most of the preterm deliveries that occurred were undertaken on an emergent basis and thus the patients were unable to receive PB. Premature infants are at greater risk of requiring an EXT for hyperbilirubinemia because of their immaturity of the hepatic microsomal enymatic conjugation pathways. Further, the level of unconjugated bilirubin at which the premature brain will be damaged is lower than the term brain because of the immaturity of the blood-brain barrier.13 Therefore, providers are more likely to initiate EXT when dealing with the premature infant. The mean gestational age at delivery of the pregnancies that underwent EXTs in our study was 2.6 weeks less than in pregnancies in which the infant did not require EXT, a statistically significant difference. For this reason, we controlled for gestational age in our multivariate analysis. After this analysis, the protective effect of maternal PB persisted. The dosing regimen of 30 mg 3 times a day of oral PB used in our study was chosen on the basis of the work by Halpin et al14 who performed a randomized controlled trial of providing PB or placebo to healthy pregnant women starting at 32 weeks of gestation in an effort to decrease hyperbilirubinemia. These authors showed a significant reduction in the mean neonatal unconjugated bilirubin from 8.5 to 6.4 mg/dL in the women treated with 1 g of PB before delivery. They also showed that the incidence of infants with levels exceeding 10 mg/dL at 96 hours of life was significantly reduced from 24.6% in controls to 7.3% in treated individuals. In our cohort, each woman received PB for a period of 7 to 10 days, with the maximum cumulative amount being 900 mg. The decision to use any drug during pregnancy should include consideration of perinatal safety. Multiple investigators have evaluated fetal exposure to PB during the latter stages of pregnancy. PB is known to freely cross the placenta throughout pregnancy.15-17 Usage of PB has been investigated as a means to reduce the occurrence and severity of neonatal intracranial hemorrhage in preterm infants. These studies have not shown a significant decrease in rates of intraventicular hemorrhage, however, in follow-up studies at both 2 and 7 years of age, Thorp et al18,19 found no difference in mean intelligence, achievement, and behavioral scores.
481 Concerns have been raised over the administration of a barbituate, a sedative, to the fetus. However, there have been no findings to date in multiple trials using antenatal PB regarding an increased incidence of neonatal depression.5,14,20,21 In conclusion, this retrospective study suggests that a brief course of maternal oral PB before delivery may reduce the need for neonatal EXT. Further study in a randomized controlled trial is necessary to confirm these results.
References 1. Yamamoto K, Sato H, Fujiyama Y, Doida Y, Bamba T. Contribution of two missense mutations (G71R and Y486D) of the bilirubin UDP glycosyltransferase (UGT1A1) gene to phenotypes of Gilbert’s syndrome and Crigler-Najjar syndrome type II. Biochim Biophys Acta 1998;1406:267-73. 2. Brown AK, Zuelzer WW. Studies on the neonatal development of the glucuronide conjugating system. J Clin Invest 1958;37: 332-40. 3. Gartner LM, Arias IM. The transfer of bilirubin from blood to bile in the neonatal guinea pig. Pediatr Res 1969;3:171-80. 4. Catz C, Yaffe SJ. Barbiturate enhancement of bilirubin conjugation and excretion in young and adult animals. Pediatr Res 1968;2:361-70. 5. Trolle D. A possible drop in first-week-mortality rate for lowbirth-weight infants after phenobarbitone treatment. Lancet 1968;2:1123-4. 6. Sugatani J, Kojima H, Ueda A, Kakizaki S, Yoshinari K, Gong QH, et al. The phenobarbital response enhancer module in the human bilirubin UDP-glucuronosyltransferase UGT1A1 gene and regulation by the nuclear receptor CAR. Hepatology 2001;33:1232-8. 7. Keenan WJ, Novak KK, Sutherland JM, Bryla DA, Fetterly KL. Morbidity and mortality associated with exchange transfusion. Pediatrics 1985;75:417-21. 8. Liley AW. Liquor amnil analysis in the management of the pregnancy complicated by resus sensitization. Am J Obstet Gynecol 1961;82:1359-70. 9. Mari G, Deter RL, Carpenter RL, Rahman F, Zimmerman R, Moise KJ Jr, et al. Noninvasive diagnosis by Doppler ultrasonography of fetal anemia due to maternal red-cell alloimmunization. Collaborative Group for Doppler Assessment of the Blood Velocity in Anemic Fetuses. N Engl J Med 2000;342:9-14. 10. Practice parameter: management of hyperbilirubinemia in the healthy term newborn. American Academy of Pediatrics. Provisional Committee for Quality Improvement and Subcommittee on Hyperbilirubinemia. Pediatrics 1994;94:558-65. 11. Maisels MJ. Jaundice. In: Glenn Brown Avery MAF, Mary Glenn MacDonald, editors. Neonatology: pathophysiology and management of the newborn. Philadelphia: JB Lippincott; 1999. p. 765-819. 12. Maisels MJ, Watchko JF. Treatment of jaundice in low birthweight infants. Arch Dis Child Fetal Neonatal Ed 2003;88:F459-63. 13. Watchko JF, Oski FA. Kernicterus in preterm newborns: past, present, and future. Pediatrics 1992;90:707-15. 14. Halpin TF, Jones AR, Bishop HL, Lerner S. Prophylaxis of neonatal hyperbilirubinemia with phenobarbital. Obstet Gynecol 1972;40:85-90. 15. Melchior JC, Svensmark O, Trolle D. Placental transfer of phenobarbitone in epileptic women, and elimination in newborns. Lancet 1967;2:860-1.
482 16. Arpino C, Brescianini S, Robert E, Castilla EE, Cocchi G, Cornel MC, et al. Teratogenic effects of antiepileptic drugs: use of an International Database on Malformations and Drug Exposure (MADRE). Epilepsia 2000;41:1436-43. 17. De Carolis MP, Romagnoli C, Frezza S, D’Urzo E, Muzii U, Mezza A, et al. Placental transfer of phenobarbital: what is new? Dev Pharmacol Ther 1992;19:19-26. 18. Thorp JA, O’Connor M, Jones AM, Hoffman EL, Belden B. Does perinatal phenobarbital exposure affect developmental outcome at age 2? Am J Perinatol 1999;16:51-60.
Trevett et al 19. Thorp JA, O’Connor M, Belden B, Etzenhouser J, Hoffman EL, Jones PG. Effects of phenobarbital and multiple-dose corticosteroids on developmental outcome at age 7 years. Obstet Gynecol 2003;101:363-73. 20. Rayburn W, Donn S, Piehl E, Compton A. Antenatal phenobarbital and bilirubin metabolism in the very low birth weight infant. Am J Obstet Gynecol 1988;159:1491-3. 21. Shankaran S, Cepeda EE, Ilagan N, Mariona F, Hassan M, Bhatia R, et al. Antenatal phenobarbital for the prevention of neonatal intracerebral hemorrhage. Am J Obstet Gynecol 1986;154:53-7.