- Email: [email protected]
Nucleated red blood cells: A marker for fetal asphyxia?
Phelan et al.
2. Benirschke K, Driscoll SG. The pathology of the human placenta. New York: Springer-Verlag, 1967. 3. D'Alton ME, Dudley DK. The ultrasound prediction of chorionicity in twin gestation. AMJ OBSTETGYNECOL1989; 160:557-61. 4. Winn HN, Gabrielli S, Reece EA, Roberts JA, Salafia C. Diagnosis of placental chorionicity in twin gestations. AMJ OBSTE~rGVNECOL1989;161:1540-2. 5. Townsend R, Simpson G, Filly R. Membrane thickness in ultrasound prediction of chorionicity of twin gestations. J Ultrasound Med 1988;7:327-32. 6. Watson WJ, Fidel AV, Seeds JW. Sonographic evaluation of growth discordance and chorionicity in twin gestation. Am J Perinatol 1992;8:342-6. 7. Mahony BS, Filly RA, Callen PW. Amnionicity and chori-
November 1995 Am J Obstet Gynecol
8. 9. 10. 11.
onicity in twin pregnancies: prediction using ultrasound. Radiology 1985;155:205-9. Barss VA, Benacerraf BR, Frigoletto FD. Uhrasonographic determination of chorion type in twin gestation. Obstet Gynecol 1985;66:779-83. Finberg HJ. The twin peak sign: reliable evidence of dichorionic twinning. J Ultrasound Med 1992;11:571-7. Multiple gestations. In: Gabbe S, Niebyl J, Simpson J, eds. Obstetrics normal and problem pregnancies. 2nd ed. New York: Churchill Livingstone, 1991:881-921. Multifetal pregnancies: epidemiology, clinical characteristics, and management. In: Reece EA, Hobbis JC, Mahoney MJ, Petrie RH, eds. Medicine of the fetus and mother. 1st ed. Philadelphia: JB Lippincon, 1992:266-83.
Nucleated red blood cells: A marker for fetal asphyxia? Jeffrey P. Phelan, MD,= Myoung Ock Ahn, MD, PhD, MPH," Lisa M. Korst, MD,b and Gilbert I. Martin, MD c' d Pomona, Pasadena, West Covina, and Orange, California, and Seoul, Korea OBJECTIVE: Our purpose was to determine whether a relationship exists between the presence of nucleated red blood cells, hypoxic ischemic encephalopathy, and long-term neonatal neurologic impairment. STUDY DESIGN: Nucleated red blood cell data from 46 singleton term neurologically impaired neonates were compared with cord blood nucleated red blood cells of 83 term nonasphyxiated newborns. The neurologically impaired neonates group was also separated as follows: nonreactive, nonreactive fetal heart rate from admission to delivery; tachycardia, reactive fetal heart rate on admission followed by tachycardia with decelerations; rupture, uterine rupture. The first and highest nucleated red blood cells value and the time to nucleated red blood cells disappearance were assessed. RESULTS: The neurologically impaired neonates group exhibited a significantly higher number of nucleated red blood cells per 100 white blood cells (34.5 -+ 68) than did the control group (3.4 -+ 3.0) (p < 0.00001). When the neurologically impaired neonates are separated as to the basis for the neurologic impairment, distinct nucleated red blood cell patterns were observed. Overall, the nonreactive group exhibited the highest mean nucleated red blood cell (51.4 +- 87.5) count and the longest clearance times (236 +_ 166 hours). CONCLUSION: In this limited population, nucleated red blood cell data appear to aid in identifying the presence of fetal asphyxia. When asphyxia was present, distinct nucleated red blood cells patterns were identified that were in keeping with the observed basis for the fetal injury. In general, the closer the birth was to the asphyxial event, the lower was the number of nucleated red blood cells. Thus our data suggest that cord blood nucleated red blood cells could assist in the timing of fetal neurologic injury. (AM J OBSTETGYNECOL1995;173:1380-4.)
Key words- Nucleated red blood cells, asphyxia, u t e r i n e rupture, fetal heart rate, neurologically i m p a i r e d infant, cerebral palsy
From the Department of Obstetrics and Gynecology, Pomona Valley Hospital Medical Center," Pasadena/' the Department of Neonatology, Queen of the Valley Hospital, ( the Department of Pediatrics, University of California, hwine, School of Medicine, d and the Department of Obstetrics and Gynecology, Cha Women's Hospital of Seoul. ~ Received for publication June 6, 1994; revised February 8, 1995; accepted March 6, 1995. Reprint requests:Jeffrey P. Phelan, MD, Suite 200, 1030 S. Arroyo Parkway, Pasadena, CA 91105. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/1/64560 1380
Nucleated red blood cells are n o t an u n c o m m o n finding in the circulating blood of n e w b o r n infants. H° T h e n u m b e r of nucleated red blood cells per 100 white blood cells is quite variable b u t is rarely m o r e t h a n 10." 4. 6. v In those instances where the nucleated red blood cells are in excess of 10, the most f r e q u e n t explanations for the increase are prematurity, 3" 6-~ Rh sensitization, 2' 7 maternal diabetes mellitus, 4' ,0 a n d i n t r a u t e r i n e growth retardation. ~,
Phe[an et al. 1381
Volume 173, Number 5 Am J Obstet Gynecol
Asphyxia has also been suggested as inducing a rise in the number of nucleated red blood cells in the circulating blood of newborn infants.'-" 4. 2. 7 Fox'-' commented that the n u m b e r of nucleated red blood cells is " . . . a rough guide to the degree of oxygen deprivation." To a certain extent, Soothill et al.' were also able to demonstrate among a group of growth-impaired fetuses a correlation between the number of nucleated red blood cells per 100 white blood cells and the severity of fetal hypoxemia. Unfortunately, none of these investigators has shown a relationship between the nucleated red blood cell count, hypoxic ischemic encephalopathy, and long-term neurologic impairment. Therefore the purpose of this investigation was to retrospectively analyze the nucleated red blood cell counts of neurologically impaired infants in the neonatal period and compare those findings with a group of normal nonasphyxiated newborn infants. Material and methods
T h e study population consisted of 83 normal nonasphyxiated newborns and 46 neurologically impaired neonates with hypoxic ischemic encephalopathy." The entry criteria for these 83 normal nonasphyxiated newborns were as follows: appropriate-for-gestational-age neonate at >-37 weeks' gestation, birth weight > 2800 gin, Apgar score > 7 at both 1 and 5 minutes, normal intrapartum fetal heart rate (FHR) pattern, clear amniotic fluid, normal neurologic evaluation at discharge, and hematocrit _>45%. In contrast, the neurologically impaired neonate population consisted of 46 singleton term infants. All infants had evidence of hypoxic ischemic encephalopathy with long-term neurologic impairment confirmed by a pediatric neurologist. The data from this population were obtained by retrospective chart review. Within this population the infants were separated into three groups on the basis of the F H R pattern and/or pattern of neurologic i m p a i r m e n t . " These groups were as follows: nonreactive, nonreactive FHR pattern from admission to delivery; tachycardia, reactive FHR pattern on admission followed by a prolonged and sustained FHR tachycardia with decelerations; rupture, infants injured as a consequence of uterine rupture, v_,In the uterine ruture group a normal FHR pattern on admission to the hospital was required. Pregnancies known to be associated with an elevated nucleated red blood cell count, such as Rh sensitization, fetal anemia, diabetes mellitus, ABO incompatibility, twins, p r e t e r m births, or evidence of intrauterine growth retardation according to California or Lubchenco curves,':' were excluded from the present investigation. In the normal newborn group the nucleated red blood cell count was obtained prospectively from mixed umbilical cord blood at birth. The nucleated red blood
451 130-
120
110
100 '
B C
90
/
80'
1 0 0
70"
W B C
50.
60'
40"
8 8 o8o
30-
20-
o8o 10-
0
NORMAL
IMPAIRED
Fig. 1. Distribution of nucleated red blood cells (NRBC) obtained from peripheral blood of 47 neurologically impaired neonates and cord blood of 83 normal newborns. WBC, White blood cells.
cell counts from the neurologically impaired neonates were obtained on peripheral and/or umbilical blood during the neonatal period. T h e neurologically impaired neonates data were obtained from a retrospective review of the neonatal records. The nucleated red blood cell data were calculated in the following manner: A complete blood cell count was performed in the hospital laboratory in the nonasphyxiated group from the cord blood obtained at delivery. The total white blood cell count was initially determined. The number of nucleated red blood cells were determined by an examination of the blood smear from the differential white blood cell count. T h e white blood cell count was then corrected for the nucleated red blood cell count. In the neurologically impaired group the number of nucleated red blood cells per 100 white blood cells was derived directly from the complete blood counts as recorded in each newborn's medical records. If a corrected white blood cell count had not been determined, a corrected white blood cell count was calculated in the manner previously described. For purposes o f this report the number of nucleated red blood ceils per I00 white blood cells is expressed as the nucleated red blood cell count. In the neurologically impaired infant group the nucleated red blood cells were determined on the basis
1382
Phelan eta[.
November 1995 Am J Obstet Gynecol
451130 ~2~ 110
N R B C
too 9o
/
so
I 0 0
7o 60
W B
so
C
40
30,
20 "TO.
o
8
8 oo
8 8 [~
o
oo oo
ooo
0
[ NONREACTIVE
TACHYCARDIA
RUPTURE
Fig. 2. Distribution of first nucleated red blood cell (NRBC) count obtained from peripheral blood count of 47 neurologically impaired neonates separated by pattern of neurologic impairment. WBC, White blood cells.
of the first nucleated red blood cell count obtained, the peak nucleated red blood cell count and the time to nucleated red blood cell disappearance (i.e., number of nucleated red blood cells p e r 100 white blood cells = 0 to 1). Statistical analysis was performed with ×~ analysis with Yates' correction and Kruskal-Wallis one-way analysis of variance. Results
During the course of t h e investigation 83 normal, nonasphyxiated newborns underwent umbilical cord blood analysis to determine the nucleated red blood cell count at birth. In the neurologically impaired neonate group an initial nucleated red blood cell count was available in 46 infants. Overall, 29 infants had complete nucleated red blood cell d a t a - f i r s t or initial, peak, and time to disappearance. The distribution of nucleated red blood cells for the normal and asphyxiated population is illustrated in Fig. 1. There, the normal nonasphyxiated newborns had a mean __- SD of 3.4 --- 3.0 nucleated red blood cells with a range of 0 to 12 nucleated red blood cells. O f these, 10 (12%) neonates had no nucleated red blood cells at delivery. Moreover, 62 (75%) normal neonates had five
or fewer nucleated red blood cells in the cord blood. In contrast, the neurologically impaired neonate group had a mean + SD of 34.5 -4- 68.3 nucleated red blood cells with a range of 1 to 451 nucleated red blood cells. Moreover, 7 (15%) had a nucleated red blood cell count of five or fewer. When contrasted with the normal nonasphyxiated neonatal group, the neurologically impaired neonates had a significantly higher mean nucleated red blood cell count (p < 0.0001). Fig. 2 illustrates the data points and median values for the first nucleated red blood cell count obtained in these three neurologic injury groups. In contrast, the m e a n - + SD values for the nonreactive, tachycardiac, and rupture groups were 51.4 __- 87.5, 13.9 -+ 8.2, and 10.3 - 6.7 nucleated red blood cells, respectively. Although the nonreactive group demonstrated the highest nucleated r e d blood cell counts, there were no significant differences in mean nucleated red blood cell counts among these three groups. This was most likely because of the small sample size. Additionally, this reflects, in part, the retrospective nature of the study and the difficulty of being able to accurately identify the onset or timing of the asphyxial event in each patient. However, the nonreactive group was significantly more likely than the rupture group to have initial and peak nucleated red blood cells values > 11 (p < 0.05). When the study population was evaluated for nucleated red blood cell count > 10 at the initial and peak values, a statistically significant difference was observed between the asphyxia and normal neonate groups (p < 0.0001). Moreover, subclassification into the three subgroups also demonstrated a significant difference between each of these subgroups and the normal group for nucleated red blood cells > 10. T h e presence of decreased fetal movement before admission to the hospital or meconium was not statistically significant. With respect to maternal age, estimated gestational age at delivery, gravidity, parity, number of prenatal care visits, and mode of delivery, no significant differences were observed. Of the 46 impaired neonates, 29 neonates had information regarding time of disappearance of the nucleated red blood cells (Fig. 3). Of these 29 patients, the number of patients in each group was as follows: nonreactive, 15; tachycardia, 8; rupture, 6. Here, the mean _+ SD for the clearance time of nucleated red blood cells from the peripheral circulation of the neurologically impaired group was significantly longer in the nonreactive group at 236 -+ 166 hours. In comparison, the mean --- SD of the clearance rate for the tachycardia and rupture groups, respectively, were 86 _+ 101 hours and 56 -+ 37 hours. On comparison, the nonreactive group was significantly more likely than either the tachycardia or the rupture group, or both, to have a time to disappearance > 80 hours (p < 0.005).
Phelan et al.
Volume 173, Number 5 Am J Obstet Gynecol
1383
60"
N R B
50'
C 40 ¸
1 0 0 W B
30-
20-
10-
C
0
0
~°° i °°...°°°°.°...° °°°°°..°
w,.---D NONREACTIVE I . . . . 4. TACHYCARDIA | © ~ UTERINE RUPTURE I
°°°.°° °°°°°°°°°°° r'-, "'°°°°°°°'°°°'°°°°.°°. ''~'~ °°'°*°'°°.°°.,°°°°°.. ~~~ °°°°°° ~ °°.°°° 40
I : I'--~ I I I ', I I ', ', ', ', ', I ', i'~t i i I I 80
120
160
200
240
280
H O U R S AFTER BIRTH
Fig. 3. Graphic representations of nucleated red blood cell (NRBC) counts in 29 neurologically impaired neonates separated on basis of neurologic impairment (nonreactive, 15; tachycardia, 8; rupture, 6). WBC, White blood cells.
Comment
Nucleated red blood cells are commonly seen on the first day of life in the cord blood and peripheral blood of most newborns. "~° In the normal newborn the number of nucleated red blood cells is dependent on the gestational age of the fetus. For example, with advancing gestational age, there is a decline in the nucleated red blood cell count?' '~~ In the term nonasphyxiated infant the n u m b e r of nucleated red blood cells is variable but is rarely higher than lOJ' 4, ~, 7 Our results in nonasphyxiated term neonates are in keeping with those observations. Previous investigators have suggested that nucleated red blood cells increase in response to an asphyxial event.~. ~. 5. 7. o To our knowledge this is the first report to describe a relationship between the quantity of nucleated red blood cells and long-term neurologic impairment. As observed herein, the nucleated red blood cell count was significantly higher in the "asphyxiated" group. The time required to produce a rise in nucleated red blood cells count is unknown but appears to be relatively short in light of the rapidity of response observed in the uterine rupture group. In general, our results suggest that the closer the asphyxial event in time to the birth of the infant, the smaller will be the rise in the nucleated red blood cell count. As demonstrated by the nonreactive group, the further in time from the insult, the higher will be the nucleated red blood cell value. In the absence of an asphyxial event, preexisting intrauterine growth retardation, maternal diabetes mellitus, or prematurity could produce an elevated nucleated red blood cell count. This could obviously cloud or
confuse the clinical picture and make the determination of whether one of these fetuses was affected by asphyxia more difficult. Here, the clearance rate or time to disappearance of the nucleated red blood cells would be helpful. As demonstrated herein, the clearance rate appeared to be linked to the pattern or the type and/or timing of asphyxial insult. In this limited population nucleated red blood cells data appear to aid in identifying the presence of fetal asphyxia. When asphyxia was present, distinct nucleated red blood cell patterns were identified that were in keeping with the observed intrapartum FHR pattern. Moreover, our data suggest that cord blood nucleated red blood cell count would be helpful in identifying potential causes of fetal asphyxia. If the nucleated red blood cell count is elevated, frequent assessment of the nucleated red blood cells to determine the clearance time may be beneficial in characterization of the asphyxial pattern. In terms of precise timing of when the insult occurs, future studies would appear necessary. REFERENCES
l. Javert CT. The occurrence and significance of nucleated erythrocytes in the fetal vessels of the placenta. AM J OBSTETGYNE(;()L1939;37:184-94. 2. Fox H. The incidence and significance of nucleated erythrocytes in the fetal vessels of the mature human placenta. J Obstet Gynaecol Br Commonw 1967;74:40-3. 3. Ryerson CS, Sanes S. The age of pregnancy-histologic diagnosis from percentage of erythroblasts in chorionic capillaries. Arch Pathol 1934;17:648-54. 4. Green DW, Mimouni F. Nucleated erythrocytes in healthy infants and in infants of diabetic mothers. J Pediatr 1990; 116:129-31. 5. Merenstein GB, Blackmon LR, Kushner J. Nucleated red cells in the newborn. Lancet 1970;1:1293-4.
Copel, Buyon, and Kleinman
6. Philip AGS, Tito AM. Increased nucleated red blood cell counts in small for gestational age infants with very low birth weight. Am J Dis Child 1989;143:164-9. 7. Anderson GW. Studies on the nucleated red blood cell count in the chorionic capillaries and the cord blood of various ages of pregnancy. AMJ OBSTETGYNECOL1941;42: 1-14. 8. Hann IM, Gibson BES, Letsky EA. The normal blood picture in neonates. In: Fetal and neonatal hematology. Philadelphia: Balli/~reTindall, 1990:37. 9. Soothill PW, Nicolaides KH, Campbell S. Prenatal asphyxia, hyperlacticemia, hypoglycemia, and erythroblastosis in growth retarded fetuses. BMJ 1987;294:1051-3.
November 1995 Am J Obstet Gynecol
10. Mimouni E Miodvinik M, Siddiqi TA, et al. Neonatal polycythemia in infants of insulin dependent diabetic mothers. Obstet Gynecol 1986;68:370-2. 11. Phelan JP, Ahn MO. Perinatal observations in forty-eight neurologically impaired term infants. AM J OBSTET GVNECOL1994;171:424-31. 12. PhelanJP. Uterine rupture. Clin Obstet Gynecol 1990;33: 432-7. 13. Williams RL, Creasy RK, Cunningham GC, I-[awes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Ohstet Gynecol 1982;59:624-32.
Successful in utero therapy of fetal heart block Joshua A. Copel, MD," Jill P. Buyon, MDfl and Charles S. Kleinman, MD=' g c New Haven, Connecticut, and New York, New York OBJECTIVE: Congenital complete heart block is associated with maternal autoantibodies to Ro and La proteins,.which injure the fetal cardiac conduction system. We administered dexamethasone to the mothers of five fetuses with heart block caused by maternal antibodies. STUDY DESIGN: We diagnosed five cases of fetal heart block at 20 to 23 weeks and treated all mothers with dexamethasone 4 mg orally each day for the remainder of the pregnancy. All patients were positive for anti-SS-A (anti-Ro) and/or anti-SS-B (anti-La) antibodies. RESULTS: Four patients had complete heart block, and one had second-degree block. In two patients (one with complete heart block, one with second-degree heart block) the degree of block lessened with treatment. Hydrops in three complete heart block patients resolved after treatment. Maternal antibody levels did not change. Matched maternal and cord samples at delivery showed similar antibody levels. CONCLUSIONS: Complete heart block can respond to transplacental glucocorticoid therapy with improved cardiac conduction. Because there may be a concurrent myocarditis, treatment in utero may also improve cardiac contractility, leading to the observed rapid resolution of hydrops. Treatment with steroids that cross the placenta should be considered for newly diagnosed cases of complete heart block with positive antibody screens. (AM J OBSTETGYNECOL1995;173:1384-90.)
Key words: Fetal treatment, fetal arrhythmia, heart block, congenital heart disease
Congenital complete heart block is a rare cardiac conduction abnormality that results from anatomic or electric discontinuity in the conducting tissues connecting the atria and ventricles. Since cardiac muscle has electric pacemaker activity that is determined by spontaneous diastolic depolarization of the cell membrane, the heart beats at a rate governed by the fastest intrinsic pacemaker activity. The presence of heart block results
From The Yale Fetal Cardiovascular Center, Departments of Obstetrics and Gynecology,~ Pediatrics,~ and Diagnostic Imaging, c Yale University School of Medicine, and the Division of Rheumatolog~ Department of Medicine, New York University School of Medicine~ Supported in part by a grant to J.P.B. from the SLE Foundation, Inc., New York. Received for publication August 30, 1994; revised February 7, 1994; accepted March 14, 1995. Reprint requests:Joshua A. Copel, AID, Section of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine, PO Box 208063, New Haven, CT 06520-8063. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/1/64876 1384
in a ventricular rate that is lower than the normal intrinsic sinus node rate. Third-degree, or complete, heart block in fetuses is usually based on the auscultation of a sustained bradycardia, which is then confirmed by fetal echocardiography to be caused by atrioventricular block. Congenital heart block can be divided into cases associated with complex abnormalities of cardiac structure and those with structurally normal hearts. The cardiac malformations usually include abnormal anatomic connections at the atrioventricular junction, such as left atrial isomerism with a common atrioventricular orifice, and discordant atrioventricular connection.' The heart block results from abnormal formation of the atrioventricularjunction as a result of abnormal formation of the cardiac conduction tissue in this region. When congenital complete heart block is found in fetuses with structurally normal hearts, maternal autoantibodies are frequently identifiedfl-~ Virtually all mothers whose fetuses have complete heart block and structurally
2. Benirschke K, Driscoll SG. The pathology of the human placenta. New York: Springer-Verlag, 1967. 3. D'Alton ME, Dudley DK. The ultrasound prediction of chorionicity in twin gestation. AMJ OBSTETGYNECOL1989; 160:557-61. 4. Winn HN, Gabrielli S, Reece EA, Roberts JA, Salafia C. Diagnosis of placental chorionicity in twin gestations. AMJ OBSTE~rGVNECOL1989;161:1540-2. 5. Townsend R, Simpson G, Filly R. Membrane thickness in ultrasound prediction of chorionicity of twin gestations. J Ultrasound Med 1988;7:327-32. 6. Watson WJ, Fidel AV, Seeds JW. Sonographic evaluation of growth discordance and chorionicity in twin gestation. Am J Perinatol 1992;8:342-6. 7. Mahony BS, Filly RA, Callen PW. Amnionicity and chori-
November 1995 Am J Obstet Gynecol
8. 9. 10. 11.
onicity in twin pregnancies: prediction using ultrasound. Radiology 1985;155:205-9. Barss VA, Benacerraf BR, Frigoletto FD. Uhrasonographic determination of chorion type in twin gestation. Obstet Gynecol 1985;66:779-83. Finberg HJ. The twin peak sign: reliable evidence of dichorionic twinning. J Ultrasound Med 1992;11:571-7. Multiple gestations. In: Gabbe S, Niebyl J, Simpson J, eds. Obstetrics normal and problem pregnancies. 2nd ed. New York: Churchill Livingstone, 1991:881-921. Multifetal pregnancies: epidemiology, clinical characteristics, and management. In: Reece EA, Hobbis JC, Mahoney MJ, Petrie RH, eds. Medicine of the fetus and mother. 1st ed. Philadelphia: JB Lippincon, 1992:266-83.
Nucleated red blood cells: A marker for fetal asphyxia? Jeffrey P. Phelan, MD,= Myoung Ock Ahn, MD, PhD, MPH," Lisa M. Korst, MD,b and Gilbert I. Martin, MD c' d Pomona, Pasadena, West Covina, and Orange, California, and Seoul, Korea OBJECTIVE: Our purpose was to determine whether a relationship exists between the presence of nucleated red blood cells, hypoxic ischemic encephalopathy, and long-term neonatal neurologic impairment. STUDY DESIGN: Nucleated red blood cell data from 46 singleton term neurologically impaired neonates were compared with cord blood nucleated red blood cells of 83 term nonasphyxiated newborns. The neurologically impaired neonates group was also separated as follows: nonreactive, nonreactive fetal heart rate from admission to delivery; tachycardia, reactive fetal heart rate on admission followed by tachycardia with decelerations; rupture, uterine rupture. The first and highest nucleated red blood cells value and the time to nucleated red blood cells disappearance were assessed. RESULTS: The neurologically impaired neonates group exhibited a significantly higher number of nucleated red blood cells per 100 white blood cells (34.5 -+ 68) than did the control group (3.4 -+ 3.0) (p < 0.00001). When the neurologically impaired neonates are separated as to the basis for the neurologic impairment, distinct nucleated red blood cell patterns were observed. Overall, the nonreactive group exhibited the highest mean nucleated red blood cell (51.4 +- 87.5) count and the longest clearance times (236 +_ 166 hours). CONCLUSION: In this limited population, nucleated red blood cell data appear to aid in identifying the presence of fetal asphyxia. When asphyxia was present, distinct nucleated red blood cells patterns were identified that were in keeping with the observed basis for the fetal injury. In general, the closer the birth was to the asphyxial event, the lower was the number of nucleated red blood cells. Thus our data suggest that cord blood nucleated red blood cells could assist in the timing of fetal neurologic injury. (AM J OBSTETGYNECOL1995;173:1380-4.)
Key words- Nucleated red blood cells, asphyxia, u t e r i n e rupture, fetal heart rate, neurologically i m p a i r e d infant, cerebral palsy
From the Department of Obstetrics and Gynecology, Pomona Valley Hospital Medical Center," Pasadena/' the Department of Neonatology, Queen of the Valley Hospital, ( the Department of Pediatrics, University of California, hwine, School of Medicine, d and the Department of Obstetrics and Gynecology, Cha Women's Hospital of Seoul. ~ Received for publication June 6, 1994; revised February 8, 1995; accepted March 6, 1995. Reprint requests:Jeffrey P. Phelan, MD, Suite 200, 1030 S. Arroyo Parkway, Pasadena, CA 91105. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/1/64560 1380
Nucleated red blood cells are n o t an u n c o m m o n finding in the circulating blood of n e w b o r n infants. H° T h e n u m b e r of nucleated red blood cells per 100 white blood cells is quite variable b u t is rarely m o r e t h a n 10." 4. 6. v In those instances where the nucleated red blood cells are in excess of 10, the most f r e q u e n t explanations for the increase are prematurity, 3" 6-~ Rh sensitization, 2' 7 maternal diabetes mellitus, 4' ,0 a n d i n t r a u t e r i n e growth retardation. ~,
Phe[an et al. 1381
Volume 173, Number 5 Am J Obstet Gynecol
Asphyxia has also been suggested as inducing a rise in the number of nucleated red blood cells in the circulating blood of newborn infants.'-" 4. 2. 7 Fox'-' commented that the n u m b e r of nucleated red blood cells is " . . . a rough guide to the degree of oxygen deprivation." To a certain extent, Soothill et al.' were also able to demonstrate among a group of growth-impaired fetuses a correlation between the number of nucleated red blood cells per 100 white blood cells and the severity of fetal hypoxemia. Unfortunately, none of these investigators has shown a relationship between the nucleated red blood cell count, hypoxic ischemic encephalopathy, and long-term neurologic impairment. Therefore the purpose of this investigation was to retrospectively analyze the nucleated red blood cell counts of neurologically impaired infants in the neonatal period and compare those findings with a group of normal nonasphyxiated newborn infants. Material and methods
T h e study population consisted of 83 normal nonasphyxiated newborns and 46 neurologically impaired neonates with hypoxic ischemic encephalopathy." The entry criteria for these 83 normal nonasphyxiated newborns were as follows: appropriate-for-gestational-age neonate at >-37 weeks' gestation, birth weight > 2800 gin, Apgar score > 7 at both 1 and 5 minutes, normal intrapartum fetal heart rate (FHR) pattern, clear amniotic fluid, normal neurologic evaluation at discharge, and hematocrit _>45%. In contrast, the neurologically impaired neonate population consisted of 46 singleton term infants. All infants had evidence of hypoxic ischemic encephalopathy with long-term neurologic impairment confirmed by a pediatric neurologist. The data from this population were obtained by retrospective chart review. Within this population the infants were separated into three groups on the basis of the F H R pattern and/or pattern of neurologic i m p a i r m e n t . " These groups were as follows: nonreactive, nonreactive FHR pattern from admission to delivery; tachycardia, reactive FHR pattern on admission followed by a prolonged and sustained FHR tachycardia with decelerations; rupture, infants injured as a consequence of uterine rupture, v_,In the uterine ruture group a normal FHR pattern on admission to the hospital was required. Pregnancies known to be associated with an elevated nucleated red blood cell count, such as Rh sensitization, fetal anemia, diabetes mellitus, ABO incompatibility, twins, p r e t e r m births, or evidence of intrauterine growth retardation according to California or Lubchenco curves,':' were excluded from the present investigation. In the normal newborn group the nucleated red blood cell count was obtained prospectively from mixed umbilical cord blood at birth. The nucleated red blood
451 130-
120
110
100 '
B C
90
/
80'
1 0 0
70"
W B C
50.
60'
40"
8 8 o8o
30-
20-
o8o 10-
0
NORMAL
IMPAIRED
Fig. 1. Distribution of nucleated red blood cells (NRBC) obtained from peripheral blood of 47 neurologically impaired neonates and cord blood of 83 normal newborns. WBC, White blood cells.
cell counts from the neurologically impaired neonates were obtained on peripheral and/or umbilical blood during the neonatal period. T h e neurologically impaired neonates data were obtained from a retrospective review of the neonatal records. The nucleated red blood cell data were calculated in the following manner: A complete blood cell count was performed in the hospital laboratory in the nonasphyxiated group from the cord blood obtained at delivery. The total white blood cell count was initially determined. The number of nucleated red blood cells were determined by an examination of the blood smear from the differential white blood cell count. T h e white blood cell count was then corrected for the nucleated red blood cell count. In the neurologically impaired group the number of nucleated red blood cells per 100 white blood cells was derived directly from the complete blood counts as recorded in each newborn's medical records. If a corrected white blood cell count had not been determined, a corrected white blood cell count was calculated in the manner previously described. For purposes o f this report the number of nucleated red blood ceils per I00 white blood cells is expressed as the nucleated red blood cell count. In the neurologically impaired infant group the nucleated red blood cells were determined on the basis
1382
Phelan eta[.
November 1995 Am J Obstet Gynecol
451130 ~2~ 110
N R B C
too 9o
/
so
I 0 0
7o 60
W B
so
C
40
30,
20 "TO.
o
8
8 oo
8 8 [~
o
oo oo
ooo
0
[ NONREACTIVE
TACHYCARDIA
RUPTURE
Fig. 2. Distribution of first nucleated red blood cell (NRBC) count obtained from peripheral blood count of 47 neurologically impaired neonates separated by pattern of neurologic impairment. WBC, White blood cells.
of the first nucleated red blood cell count obtained, the peak nucleated red blood cell count and the time to nucleated red blood cell disappearance (i.e., number of nucleated red blood cells p e r 100 white blood cells = 0 to 1). Statistical analysis was performed with ×~ analysis with Yates' correction and Kruskal-Wallis one-way analysis of variance. Results
During the course of t h e investigation 83 normal, nonasphyxiated newborns underwent umbilical cord blood analysis to determine the nucleated red blood cell count at birth. In the neurologically impaired neonate group an initial nucleated red blood cell count was available in 46 infants. Overall, 29 infants had complete nucleated red blood cell d a t a - f i r s t or initial, peak, and time to disappearance. The distribution of nucleated red blood cells for the normal and asphyxiated population is illustrated in Fig. 1. There, the normal nonasphyxiated newborns had a mean __- SD of 3.4 --- 3.0 nucleated red blood cells with a range of 0 to 12 nucleated red blood cells. O f these, 10 (12%) neonates had no nucleated red blood cells at delivery. Moreover, 62 (75%) normal neonates had five
or fewer nucleated red blood cells in the cord blood. In contrast, the neurologically impaired neonate group had a mean + SD of 34.5 -4- 68.3 nucleated red blood cells with a range of 1 to 451 nucleated red blood cells. Moreover, 7 (15%) had a nucleated red blood cell count of five or fewer. When contrasted with the normal nonasphyxiated neonatal group, the neurologically impaired neonates had a significantly higher mean nucleated red blood cell count (p < 0.0001). Fig. 2 illustrates the data points and median values for the first nucleated red blood cell count obtained in these three neurologic injury groups. In contrast, the m e a n - + SD values for the nonreactive, tachycardiac, and rupture groups were 51.4 __- 87.5, 13.9 -+ 8.2, and 10.3 - 6.7 nucleated red blood cells, respectively. Although the nonreactive group demonstrated the highest nucleated r e d blood cell counts, there were no significant differences in mean nucleated red blood cell counts among these three groups. This was most likely because of the small sample size. Additionally, this reflects, in part, the retrospective nature of the study and the difficulty of being able to accurately identify the onset or timing of the asphyxial event in each patient. However, the nonreactive group was significantly more likely than the rupture group to have initial and peak nucleated red blood cells values > 11 (p < 0.05). When the study population was evaluated for nucleated red blood cell count > 10 at the initial and peak values, a statistically significant difference was observed between the asphyxia and normal neonate groups (p < 0.0001). Moreover, subclassification into the three subgroups also demonstrated a significant difference between each of these subgroups and the normal group for nucleated red blood cells > 10. T h e presence of decreased fetal movement before admission to the hospital or meconium was not statistically significant. With respect to maternal age, estimated gestational age at delivery, gravidity, parity, number of prenatal care visits, and mode of delivery, no significant differences were observed. Of the 46 impaired neonates, 29 neonates had information regarding time of disappearance of the nucleated red blood cells (Fig. 3). Of these 29 patients, the number of patients in each group was as follows: nonreactive, 15; tachycardia, 8; rupture, 6. Here, the mean _+ SD for the clearance time of nucleated red blood cells from the peripheral circulation of the neurologically impaired group was significantly longer in the nonreactive group at 236 -+ 166 hours. In comparison, the mean --- SD of the clearance rate for the tachycardia and rupture groups, respectively, were 86 _+ 101 hours and 56 -+ 37 hours. On comparison, the nonreactive group was significantly more likely than either the tachycardia or the rupture group, or both, to have a time to disappearance > 80 hours (p < 0.005).
Phelan et al.
Volume 173, Number 5 Am J Obstet Gynecol
1383
60"
N R B
50'
C 40 ¸
1 0 0 W B
30-
20-
10-
C
0
0
~°° i °°...°°°°.°...° °°°°°..°
w,.---D NONREACTIVE I . . . . 4. TACHYCARDIA | © ~ UTERINE RUPTURE I
°°°.°° °°°°°°°°°°° r'-, "'°°°°°°°'°°°'°°°°.°°. ''~'~ °°'°*°'°°.°°.,°°°°°.. ~~~ °°°°°° ~ °°.°°° 40
I : I'--~ I I I ', I I ', ', ', ', ', I ', i'~t i i I I 80
120
160
200
240
280
H O U R S AFTER BIRTH
Fig. 3. Graphic representations of nucleated red blood cell (NRBC) counts in 29 neurologically impaired neonates separated on basis of neurologic impairment (nonreactive, 15; tachycardia, 8; rupture, 6). WBC, White blood cells.
Comment
Nucleated red blood cells are commonly seen on the first day of life in the cord blood and peripheral blood of most newborns. "~° In the normal newborn the number of nucleated red blood cells is dependent on the gestational age of the fetus. For example, with advancing gestational age, there is a decline in the nucleated red blood cell count?' '~~ In the term nonasphyxiated infant the n u m b e r of nucleated red blood cells is variable but is rarely higher than lOJ' 4, ~, 7 Our results in nonasphyxiated term neonates are in keeping with those observations. Previous investigators have suggested that nucleated red blood cells increase in response to an asphyxial event.~. ~. 5. 7. o To our knowledge this is the first report to describe a relationship between the quantity of nucleated red blood cells and long-term neurologic impairment. As observed herein, the nucleated red blood cell count was significantly higher in the "asphyxiated" group. The time required to produce a rise in nucleated red blood cells count is unknown but appears to be relatively short in light of the rapidity of response observed in the uterine rupture group. In general, our results suggest that the closer the asphyxial event in time to the birth of the infant, the smaller will be the rise in the nucleated red blood cell count. As demonstrated by the nonreactive group, the further in time from the insult, the higher will be the nucleated red blood cell value. In the absence of an asphyxial event, preexisting intrauterine growth retardation, maternal diabetes mellitus, or prematurity could produce an elevated nucleated red blood cell count. This could obviously cloud or
confuse the clinical picture and make the determination of whether one of these fetuses was affected by asphyxia more difficult. Here, the clearance rate or time to disappearance of the nucleated red blood cells would be helpful. As demonstrated herein, the clearance rate appeared to be linked to the pattern or the type and/or timing of asphyxial insult. In this limited population nucleated red blood cells data appear to aid in identifying the presence of fetal asphyxia. When asphyxia was present, distinct nucleated red blood cell patterns were identified that were in keeping with the observed intrapartum FHR pattern. Moreover, our data suggest that cord blood nucleated red blood cell count would be helpful in identifying potential causes of fetal asphyxia. If the nucleated red blood cell count is elevated, frequent assessment of the nucleated red blood cells to determine the clearance time may be beneficial in characterization of the asphyxial pattern. In terms of precise timing of when the insult occurs, future studies would appear necessary. REFERENCES
l. Javert CT. The occurrence and significance of nucleated erythrocytes in the fetal vessels of the placenta. AM J OBSTETGYNE(;()L1939;37:184-94. 2. Fox H. The incidence and significance of nucleated erythrocytes in the fetal vessels of the mature human placenta. J Obstet Gynaecol Br Commonw 1967;74:40-3. 3. Ryerson CS, Sanes S. The age of pregnancy-histologic diagnosis from percentage of erythroblasts in chorionic capillaries. Arch Pathol 1934;17:648-54. 4. Green DW, Mimouni F. Nucleated erythrocytes in healthy infants and in infants of diabetic mothers. J Pediatr 1990; 116:129-31. 5. Merenstein GB, Blackmon LR, Kushner J. Nucleated red cells in the newborn. Lancet 1970;1:1293-4.
Copel, Buyon, and Kleinman
6. Philip AGS, Tito AM. Increased nucleated red blood cell counts in small for gestational age infants with very low birth weight. Am J Dis Child 1989;143:164-9. 7. Anderson GW. Studies on the nucleated red blood cell count in the chorionic capillaries and the cord blood of various ages of pregnancy. AMJ OBSTETGYNECOL1941;42: 1-14. 8. Hann IM, Gibson BES, Letsky EA. The normal blood picture in neonates. In: Fetal and neonatal hematology. Philadelphia: Balli/~reTindall, 1990:37. 9. Soothill PW, Nicolaides KH, Campbell S. Prenatal asphyxia, hyperlacticemia, hypoglycemia, and erythroblastosis in growth retarded fetuses. BMJ 1987;294:1051-3.
November 1995 Am J Obstet Gynecol
10. Mimouni E Miodvinik M, Siddiqi TA, et al. Neonatal polycythemia in infants of insulin dependent diabetic mothers. Obstet Gynecol 1986;68:370-2. 11. Phelan JP, Ahn MO. Perinatal observations in forty-eight neurologically impaired term infants. AM J OBSTET GVNECOL1994;171:424-31. 12. PhelanJP. Uterine rupture. Clin Obstet Gynecol 1990;33: 432-7. 13. Williams RL, Creasy RK, Cunningham GC, I-[awes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Ohstet Gynecol 1982;59:624-32.
Successful in utero therapy of fetal heart block Joshua A. Copel, MD," Jill P. Buyon, MDfl and Charles S. Kleinman, MD=' g c New Haven, Connecticut, and New York, New York OBJECTIVE: Congenital complete heart block is associated with maternal autoantibodies to Ro and La proteins,.which injure the fetal cardiac conduction system. We administered dexamethasone to the mothers of five fetuses with heart block caused by maternal antibodies. STUDY DESIGN: We diagnosed five cases of fetal heart block at 20 to 23 weeks and treated all mothers with dexamethasone 4 mg orally each day for the remainder of the pregnancy. All patients were positive for anti-SS-A (anti-Ro) and/or anti-SS-B (anti-La) antibodies. RESULTS: Four patients had complete heart block, and one had second-degree block. In two patients (one with complete heart block, one with second-degree heart block) the degree of block lessened with treatment. Hydrops in three complete heart block patients resolved after treatment. Maternal antibody levels did not change. Matched maternal and cord samples at delivery showed similar antibody levels. CONCLUSIONS: Complete heart block can respond to transplacental glucocorticoid therapy with improved cardiac conduction. Because there may be a concurrent myocarditis, treatment in utero may also improve cardiac contractility, leading to the observed rapid resolution of hydrops. Treatment with steroids that cross the placenta should be considered for newly diagnosed cases of complete heart block with positive antibody screens. (AM J OBSTETGYNECOL1995;173:1384-90.)
Key words: Fetal treatment, fetal arrhythmia, heart block, congenital heart disease
Congenital complete heart block is a rare cardiac conduction abnormality that results from anatomic or electric discontinuity in the conducting tissues connecting the atria and ventricles. Since cardiac muscle has electric pacemaker activity that is determined by spontaneous diastolic depolarization of the cell membrane, the heart beats at a rate governed by the fastest intrinsic pacemaker activity. The presence of heart block results
From The Yale Fetal Cardiovascular Center, Departments of Obstetrics and Gynecology,~ Pediatrics,~ and Diagnostic Imaging, c Yale University School of Medicine, and the Division of Rheumatolog~ Department of Medicine, New York University School of Medicine~ Supported in part by a grant to J.P.B. from the SLE Foundation, Inc., New York. Received for publication August 30, 1994; revised February 7, 1994; accepted March 14, 1995. Reprint requests:Joshua A. Copel, AID, Section of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine, PO Box 208063, New Haven, CT 06520-8063. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/1/64876 1384
in a ventricular rate that is lower than the normal intrinsic sinus node rate. Third-degree, or complete, heart block in fetuses is usually based on the auscultation of a sustained bradycardia, which is then confirmed by fetal echocardiography to be caused by atrioventricular block. Congenital heart block can be divided into cases associated with complex abnormalities of cardiac structure and those with structurally normal hearts. The cardiac malformations usually include abnormal anatomic connections at the atrioventricular junction, such as left atrial isomerism with a common atrioventricular orifice, and discordant atrioventricular connection.' The heart block results from abnormal formation of the atrioventricularjunction as a result of abnormal formation of the cardiac conduction tissue in this region. When congenital complete heart block is found in fetuses with structurally normal hearts, maternal autoantibodies are frequently identifiedfl-~ Virtually all mothers whose fetuses have complete heart block and structurally