Neonatal periventricular-intraventricular hemorrhage after maternal ~-sympathomimetic tocolysis Lynn J. Groome, PhD, MD, Robert L. Goldenberg, MD, Suzanne P. Cliver, BA, Richard O. Davis, MD, Rachel L. Copper, RN, and the March of Dimes Multicenter Study Group Birmingham, Alabama OBJECTIVE: Our objective was to determine if the rate of periventricular-intraventricular hemorrhage is increased in the offspring of women who received a l3-sympathomimetic agent as part of the management of preterm labor. STUDY DESIGN: This retrospective study consists of 2827 women who were delivered of a singleton, live infant free of congenital neurologic anomalies between 25 and 36 completed weeks of gestation during a multicenter preterm birth prevention trial. The data were analyzed, adjusting for type of tocolytic agent, race, infant sex, gestational age, birth weight, health care center, route of delivery, indication for delivery, intrapartum fetal distress, respiratory distress syndrome, and neonatal sepsis. RESULTS: The overall incidence of periventricular-intraventricular hemorrhage in this population was 5.6%. In a univariate analysis in which no adjustment was made for potentially confounding variables, l3-sympathomimetic tocolysis was found to be associated with nearly a fourfold increase in the incidence of periventricular-intraventricular hemorrhage when compared with the use of either magnesium sulfate or no tocolytic agent. The results of a multivariate regression analysis revealed that l3-sympathomimetic agents were associated with a statistically significant increase in the overall incidence of periventricular-intraventricular hemorrhage (odds ratio 2.47, 95% confidence interval 1.34 to 4.56, P = 0.004) and a similar, but not significant, increase in the incidence of grades 3 and 4 periventricular-intraventricular hemorrhage (odds ratio 2.50, 95% confidence interval 0.96 to 6.48, P = 0.06). CONCLUSION: I3-Sympathomimetic tocolytic therapy may be associated with a more than twofold increase in the inCidence of neonatal periventricular-intraventricular hemorrhage. (AM J OBSTET GVNECOL 1992;167:873-9.)
Key words: Neonatal intraventricular hemorrhage, J3-sympathomimetic tocolytics, preterm Although J3-sympathomimetic agents are commonly used in an attempt to delay preterm delivery, it appears that the use of these drugs has not reduced the neonatal morbidity or mortality associated with premature birth.' Periventricular-intraventricular hemorrhage is a recognized complication of preterm birth and may be partly responsible for the lack of improvement in neonatal outcome after J3-sympathomimetic tocolysis. Factors associated with fluctuations in systemic blood pressure or cerebral blood flow in the germinal matrix have been implicated in the pathogenesis of periventricular-intraventricular hemorrhage. 2 . , J3-Sympathomimetic tocolytics are known to cross the placenta,' and
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Alabama at Birmingham. Supported in part by the March of Dimes. Presented at the Twelfth Annual Meeting of the Society of Perinatal Obstetricians, Orlando, Florida, February 3-8, 1992. Reprint requests: Lynn J. Groome, PhD, MD, Division of MaternalFetal Medicine, Department of Obstetrics (1 Gynecology, University of South Alabama, 2451 Fillingim St., Mobile, AL 36617. 6/6/39914
it is possible that the mechanical forces generated by an increase in fetal blood pressure or cerebral blood flow after the use of these drugs may place the infant at an increased risk for peri ventricular-intraventricular hemorrhage. The purpose of this retrospective study was to determine the rate of peri ventricular-intraventricular hemorrhage in the offspring of women who received a J3-sympathomimetic agent for preterm labor tocolysis and to compare this rate with that observed in infants whose mothers received no tocolytic agent, controlling for as many of the potentially confounding factors as possible. The study population consisted of 2827 infants who were delivered between 25 and 36 completed weeks of gestation during the time of a multicenter March of Dimes-sponsored preterm birth-prevention trial.
Methods The initial study population consisted of 32,982 women who were delivered of singleton, live infants without major congenital central nervous system anomalies during a multicenter pre term-birth prevention
874
Groome et al.
October 1992 Am J Obstet Gynecol
Table I. Demographic characteristics Characteristic
Overall
Study group
Gestational age (wk) No. Ethnic group White Black Hispanic Other Unknown Nulliparous Center
20-45
25-36
A
B C
D
E
32,982
2,827
11,674 15,232 5,439 611 26 13,045
(35.4%) (46.2%) (16.5%) (1.9%) (-) (39.6%)
938 (34.8%) 1,564 (55.3%) 249 (8.8%) 26 (0.1%) 5(-) 1,105 (39.1%)
12,938 4,813 6,871 4,311 4,049
(39.2%) (14.6%) (20.8%) (13.1%) (12.3%)
1,236 470 327 445 349
(43.7%) (16.6%) (11.6%) (15.7%) (12.3%)
Table II. Clinical characteristics of study population Characteristic
Sex Male Female Unknown Route of delivery Vaginal Cesarean section Clinical outcome Fetal distress in labor RDS Mechanical ventilation Neonatal sepsis Intraventricular hemorrhage Intraventricular hemorrhage, grades 3 and 4 Reason for pre term delivery Indicated Premature rupture of membranes Spontaneous
No. (%) 1384 (49.0) 1442 (51.0) 1 (-) 2101 (74.3) 726 (25.7) 535 623 616 151 157 48
(19.1) (22.0) (21.8) (5.3) (5.6) (1.7)
555 (19.6) 935 (33.1) 1337 (47.3)
trial sponsored by the March of Dimes between November 1982 and October 1986. Of these, 2827 infants were delivered between 25 and 36 completed weeks of gestation; they formed the major study group. The participating centers were the University of California at San Diego, Ohio State University, Vanderbilt University, Northwestern University, and the University of Alabama at Birmingham. The data base, definitions, and data collection verification methods are described elsewhere. 5. 7 From the available data base it was possible to determine the class of tocolytic agent used (i.e., f3-sympathomimetic, magnesium sulfate, a calcium channel blocker, or a prostaglandin synthetase inhibitor) and the timing of the use of this agent in relation to delivery. However, we could not identify the specific f3-sympatho mimetic agent used or the dosage of this drug. The
presence or absence of peri ventricular-intraventricular hemorrhage was assessed in terms of whether a tocolytic agent was ever used, whether it was used during the admission for delivery, or whether it was used in the 24 hours immediately before delivery. It was also possible to determine if the delivery was medically indicated (for maternal or fetal reasons), the result of spontaneous preterm labor, or if it followed preterm premature rupture of the fetal membranes. Route of delivery was categorized as either vaginal or cesarean section. Both obstetric and neonatal criteria were used to estimate gestational age. Neonatal gestational age was determined by a modified Dubowitz examination. All pregnancies complicated by preterm delivery, delivery of an infant weighing 1500 gm were determined by the staff neonatologist at each center, but the presence of symptoms was generally necessary before screening was instituted. 6 The data were analyzed by X~, the Student t test, and logistic regression. All calculations were performed with the SAS statistical program (Cary, N.C.). Results
Demographic characteristics for the overall population and for the subset of patients forming the study group are presented in Table I. As shown, the study group consisted of a higher percentage of black patients (55.3% vs 46.2%) and a lower percentage of Hispanic patients (8.8% vs 16.5%) as compared with the overall population from which it was drawn. Because we were primarily concerned with the outcome of the 2827 patients who delivered between 25 and 36 completed weeks of gestation, the clinical characteristics of this group is summarized in Table II. As shown, 157 (5.6%) of the infants had a diagnosis of periventricular-intraventricular hemorrhage; of these, 48 (30.5%) had grade 3 or 4 bleeding. Of the 2827
Volume 167 Number 4, Part I
Neonatal intraventricular hemorrhage after maternal j3-sympathomimetic tocolysis
875
Table III. Relationship between type of tocolytic agent and periventricular-intraventricular hemorrhage according to cause of preterm delivery Variable
Spontaneous, no tocol.ytic
Spontaneous, (3-sympathomimetic
Spontaneous, magnesium sulfate
Indicated no tocolytic
Other
No. (%) Intraventricular hemorrhage
1650 (58.4) 4.2
173 (6.1) 16.2*
155 (5.5) 3.9
462 (16.3) 6.9*
387 (13.7) 5.7
1.2
5.8*
1.9
1.7
1.8
2289 ± 648
1878 ± 664*
2028 ± 554*
1943 ± 667*
1921 ± 585*
34.1 ± 2.6
31.9 ± 3.4*
32.8 ± 2.6*
33.6 ± 2.5*
32.5 ± 2.8*
(%)
Intraventricular hemorrhage, grades 3 and 4 (%) Mean birth weight (gm) (±SD) Mean gestational age (wk) (±SD) *Difference significant at
p
Key words: Neonatal intraventricular hemorrhage, J3-sympathomimetic tocolytics, preterm Although J3-sympathomimetic agents are commonly used in an attempt to delay preterm delivery, it appears that the use of these drugs has not reduced the neonatal morbidity or mortality associated with premature birth.' Periventricular-intraventricular hemorrhage is a recognized complication of preterm birth and may be partly responsible for the lack of improvement in neonatal outcome after J3-sympathomimetic tocolysis. Factors associated with fluctuations in systemic blood pressure or cerebral blood flow in the germinal matrix have been implicated in the pathogenesis of periventricular-intraventricular hemorrhage. 2 . , J3-Sympathomimetic tocolytics are known to cross the placenta,' and
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Alabama at Birmingham. Supported in part by the March of Dimes. Presented at the Twelfth Annual Meeting of the Society of Perinatal Obstetricians, Orlando, Florida, February 3-8, 1992. Reprint requests: Lynn J. Groome, PhD, MD, Division of MaternalFetal Medicine, Department of Obstetrics (1 Gynecology, University of South Alabama, 2451 Fillingim St., Mobile, AL 36617. 6/6/39914
it is possible that the mechanical forces generated by an increase in fetal blood pressure or cerebral blood flow after the use of these drugs may place the infant at an increased risk for peri ventricular-intraventricular hemorrhage. The purpose of this retrospective study was to determine the rate of peri ventricular-intraventricular hemorrhage in the offspring of women who received a J3-sympathomimetic agent for preterm labor tocolysis and to compare this rate with that observed in infants whose mothers received no tocolytic agent, controlling for as many of the potentially confounding factors as possible. The study population consisted of 2827 infants who were delivered between 25 and 36 completed weeks of gestation during the time of a multicenter March of Dimes-sponsored preterm birth-prevention trial.
Methods The initial study population consisted of 32,982 women who were delivered of singleton, live infants without major congenital central nervous system anomalies during a multicenter pre term-birth prevention
874
Groome et al.
October 1992 Am J Obstet Gynecol
Table I. Demographic characteristics Characteristic
Overall
Study group
Gestational age (wk) No. Ethnic group White Black Hispanic Other Unknown Nulliparous Center
20-45
25-36
A
B C
D
E
32,982
2,827
11,674 15,232 5,439 611 26 13,045
(35.4%) (46.2%) (16.5%) (1.9%) (-) (39.6%)
938 (34.8%) 1,564 (55.3%) 249 (8.8%) 26 (0.1%) 5(-) 1,105 (39.1%)
12,938 4,813 6,871 4,311 4,049
(39.2%) (14.6%) (20.8%) (13.1%) (12.3%)
1,236 470 327 445 349
(43.7%) (16.6%) (11.6%) (15.7%) (12.3%)
Table II. Clinical characteristics of study population Characteristic
Sex Male Female Unknown Route of delivery Vaginal Cesarean section Clinical outcome Fetal distress in labor RDS Mechanical ventilation Neonatal sepsis Intraventricular hemorrhage Intraventricular hemorrhage, grades 3 and 4 Reason for pre term delivery Indicated Premature rupture of membranes Spontaneous
No. (%) 1384 (49.0) 1442 (51.0) 1 (-) 2101 (74.3) 726 (25.7) 535 623 616 151 157 48
(19.1) (22.0) (21.8) (5.3) (5.6) (1.7)
555 (19.6) 935 (33.1) 1337 (47.3)
trial sponsored by the March of Dimes between November 1982 and October 1986. Of these, 2827 infants were delivered between 25 and 36 completed weeks of gestation; they formed the major study group. The participating centers were the University of California at San Diego, Ohio State University, Vanderbilt University, Northwestern University, and the University of Alabama at Birmingham. The data base, definitions, and data collection verification methods are described elsewhere. 5. 7 From the available data base it was possible to determine the class of tocolytic agent used (i.e., f3-sympathomimetic, magnesium sulfate, a calcium channel blocker, or a prostaglandin synthetase inhibitor) and the timing of the use of this agent in relation to delivery. However, we could not identify the specific f3-sympatho mimetic agent used or the dosage of this drug. The
presence or absence of peri ventricular-intraventricular hemorrhage was assessed in terms of whether a tocolytic agent was ever used, whether it was used during the admission for delivery, or whether it was used in the 24 hours immediately before delivery. It was also possible to determine if the delivery was medically indicated (for maternal or fetal reasons), the result of spontaneous preterm labor, or if it followed preterm premature rupture of the fetal membranes. Route of delivery was categorized as either vaginal or cesarean section. Both obstetric and neonatal criteria were used to estimate gestational age. Neonatal gestational age was determined by a modified Dubowitz examination. All pregnancies complicated by preterm delivery, delivery of an infant weighing 1500 gm were determined by the staff neonatologist at each center, but the presence of symptoms was generally necessary before screening was instituted. 6 The data were analyzed by X~, the Student t test, and logistic regression. All calculations were performed with the SAS statistical program (Cary, N.C.). Results
Demographic characteristics for the overall population and for the subset of patients forming the study group are presented in Table I. As shown, the study group consisted of a higher percentage of black patients (55.3% vs 46.2%) and a lower percentage of Hispanic patients (8.8% vs 16.5%) as compared with the overall population from which it was drawn. Because we were primarily concerned with the outcome of the 2827 patients who delivered between 25 and 36 completed weeks of gestation, the clinical characteristics of this group is summarized in Table II. As shown, 157 (5.6%) of the infants had a diagnosis of periventricular-intraventricular hemorrhage; of these, 48 (30.5%) had grade 3 or 4 bleeding. Of the 2827
Volume 167 Number 4, Part I
Neonatal intraventricular hemorrhage after maternal j3-sympathomimetic tocolysis
875
Table III. Relationship between type of tocolytic agent and periventricular-intraventricular hemorrhage according to cause of preterm delivery Variable
Spontaneous, no tocol.ytic
Spontaneous, (3-sympathomimetic
Spontaneous, magnesium sulfate
Indicated no tocolytic
Other
No. (%) Intraventricular hemorrhage
1650 (58.4) 4.2
173 (6.1) 16.2*
155 (5.5) 3.9
462 (16.3) 6.9*
387 (13.7) 5.7
1.2
5.8*
1.9
1.7
1.8
2289 ± 648
1878 ± 664*
2028 ± 554*
1943 ± 667*
1921 ± 585*
34.1 ± 2.6
31.9 ± 3.4*
32.8 ± 2.6*
33.6 ± 2.5*
32.5 ± 2.8*
(%)
Intraventricular hemorrhage, grades 3 and 4 (%) Mean birth weight (gm) (±SD) Mean gestational age (wk) (±SD) *Difference significant at
p
