- Email: [email protected]
Neonatal outcome of temporizing treatment in early-onset preeclampsia
European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211–215
www.elsevier.com / locate / ejogrb
Neonatal outcome of temporizing treatment in early-onset preeclampsia ¨ I.J. Withagen, Willy Visser*, Henk C.S. Wallenburg Mariella Erasmus University School of Medicine and Health Sciences, Institute of Obstetrics and Gynecology, Rotterdam, The Netherlands Received 24 December 1999; received in revised form 2 February 2000; accepted 13 March 2000
Abstract Objective: To assess the effect of prolongation of pregnancy on neonatal outcome by means of hemodynamic treatment in patients with early-onset preeclampsia. Study design: A retrospective case-controlled study of 222 liveborn infants of patients with early-onset (24–31 weeks) preeclampsia, who underwent temporizing hemodynamic treatment. Of the two control groups of liveborn preterm infants of non-preeclamptic mothers one group was matched with the study group for gestational age on admission (group I), one for gestational age at birth (group II). Primary outcome measures were neonatal and infant mortality and variables of neonatal morbidity. Results: Median gestation in the study group of preeclamptic patients was prolonged from 29.3 to 31.3 weeks. No difference in neonatal or infant mortality was observed between infants from preeclamptic mothers and in the control groups. The study population showed better results than control group I with regard to admission to NICU (P,0.01), mechanical ventilation (P,0.001) and intracranial hemorrhage (P,0.01). Control group II had better results than the study group with respect to birthweight (P,0.001), bronchopulmonary dysplasia (P,0.01), patent ductus arteriosus (P,0.01), and retinopathy (P,0.01). Conclusion: Prolongation of gestation in patients with early-onset preeclampsia may reduce neonatal morbidity, but neonates of the same gestational age without a preeclamptic mother still have a better prognosis. 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Preeclampsia; Neonatal morbidity; Neonatal mortality; Temporizing hemodynamic treatment
1. Introduction Preeclampsia complicates 1–5% of all pregnancies and is directly or indirectly responsible for a large proportion of maternal and perinatal mortality and morbidity [1]. Although the cause of preeclampsia is unknown, the placenta is considered a key factor [2], and delivery of fetus and placenta is the only effective treatment. Delivery may benefit the mother but is not in the interest of the second patient, the fetus, if remote from term, since the main cause of neonatal mortality and morbidity in patients with severe preeclampsia is prematurity [3]. Duration of pregnancy at delivery was shown to be the primary *Corresponding author. Corresponding address: Academic Hospital Rotterdam / Sophia, Room Sk 4130, Dr. Molewaterplein 60, P.O. Box 2060, 3000 CB Rotterdam, The Netherlands. Fax: 131-10-463-6815. E-mail address: [email protected] (W. Visser).
prognostic factor for neonatal death and impairment in cases of severe early preeclampsia [4]. In 1985, a protocol of temporizing management of severe early-onset preeclampsia was instituted in the Department of Obstetrics of the University Hospital Rotterdam (AZR). The protocol aims at prolongation of pregnancy to avoid severely preterm birth and is based on correction of the disturbed maternal circulation with plasma volume expansion and pharmacologic vasodilatation, usually under central hemodynamic monitoring [5]. It was shown that such hemodynamic treatment in preeclamptic patients with a gestational age of 32 weeks or less and no fetal indication for delivery within 48 h, may prolong pregnancy with an average of 17 days, generally improves maternal condition, and may reduce perinatal mortality [5–7]. However, the effect of temporizing hemodynamic treatment of preeclampsia on neonatal wellbeing was not assessed in previous studies. For that reason, we
0301-2115 / 01 / $ – see front matter 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0301-2115( 00 )00332-8
212
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215
performed a retrospective case-controlled study with the aim to evaluate the effect of temporizing management of severe early-onset preeclampsia on neonatal outcome.
Preterm delivery in the control group was caused by spontaneous idiopathic preterm labor, with or without preterm rupture of membranes. Patients admitted for suspected placental abruption, fetal distress or fetal growth retardation were excluded.
2. Patients and methods
2.1. Selection of patients
2.3. End points of study
The study population consisted of 222 consecutive liveborn infants of patients with severe preeclampsia between 24 and 31 completed weeks of gestation, who underwent temporizing hemodynamic treatment in the Department of Obstetrics of the AZR between January 1, 1985 and December 31, 1993. Patients with twin pregnancy or known vascular, renal, hepatic or hematologic disease were excluded. Severe preeclampsia was defined as the occurrence of a diastolic blood pressure$110 mm Hg (Korotkoff 4) with proteinuria ($ 0.3 g / l / 24 h) after 20 weeks’ gestation, or the occurrence of persistent diastolic blood pressures of $90 mm Hg with proteinuria in combination with either HELLP syndrome or eclampsia. The HELLP syndrome was defined as the simultaneous occurrence of hemolysis (defined by peripheral blood smear), raised (.30 U l 21 ) serum aspartate aminotransferase (ASAT) and serum alanine aminotransferase (ALAT) and a platelet count of less than 100310 9 l 21 [5]. Gestational age was based on the last menstrual period and confirmed by early ultrasonography. Temporizing hemodynamic treatment consisted of bed rest and correction of the maternal circulation by means of pharmacologic vasodilatation (dihydralazine) and plasma volume expansion (pasteurized plasma), usually under central hemodynamic monitoring with the use of the Swan–Ganz thermodilution catheter [5]. The target value of diastolic intra-arterial blood pressure was 90 mm Hg and that of the pulmonary capillary wedge pressure was 10 mm Hg. Fetal condition was monitored by means of cardiotocography and ultrasound. Decisions to deliver were taken by the attending obstetricians on the basis of their interpretation of the cardiotocogram or of maternal condition. The details of the management protocol are described elsewhere [5].
Neonatal data were obtained by review of charts over a period of 1 year. Primary outcome measures were mortality and morbidity. Mortality was divided into neonatal mortality, occurring within the first 28 days of life, and infant mortality, occurring within the first year of life. Morbidity was defined by Apgar scores, umbilical arterial acid and base values, neonatal hematocrit determined in the first 24 h after birth, admission and length of stay in the neonatal intensive care unit (ICU), admission and length of stay in the hospital, and ventilation characteristics. Smallfor-gestational age (SGA) was defined as a birthweight below the 10th percentile of the Dutch reference curve corrected for parity and gender [8]. Pneumothorax was diagnosed by radiographic findings. A diagnosis of hyaline membrane disease was made in the presence of progressive respiratory distress in combination with a chest radiogram showing fine granular infiltrates persisting until at least the third day after birth. Bronchopulmonary dysplasia was defined as chronic respiratory distress and oxygen requirement beyond 28 days of life, accompanied by a chest radiograph that showed persistent streaks of increased density in both lungs interspersed with normal or hyperlucent areas. Intracranial hemorrhage and periventricular leucomalacia were diagnosed by ultrasonographic examination and graded I–IV [9]. Hypotension was defined if the infant had to be treated with vasoactive substances or plasma infusion. The diagnosis of patent ductus arteriosus was based on the presence of clinical symptoms and confirmed by echocardiographic examination. The diagnosis of necrotizing enterocolitis was confirmed by an abdominal radiograph. A diagnosis of septicemia and pneumonia required a positive blood culture. Retinopathy of prematurity was defined as abnormal vasoproliferation and tortuosity, retinal ridge, scarring and retinal detachment.
2.2. Selection of controls Each infant of the study group was matched for gestational age, gender and year of birth in a blinded fashion, without knowledge of neonatal characteristics and clinical outcome, with two liveborn infants of non-preeclamptic mothers: one delivered at the gestational age (61 week) at which temporizing treatment of the preeclamptic patient was initiated (control group I) and one delivered at the gestational age (61 week) at which the preeclamptic patient was delivered (control group II). Each control infant was matched only once.
2.4. Statistical analysis Data analysis was based on intention-to-treat and included all patients, also when they were delivered before the actual start of temporizing treatment. Differences in categorical variables between groups were assessed with the x 2 test and the Mann–Whitney U test was used to compare quantitative variables. Because of multiple comparisons a P value of 0.01 or less was considered to represent statistical significance.
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215 Table 1 Obstetric characteristics a Study group
Nulliparous (n) HELLP (n) Eclampsia (n) Cesarean section (n) Maternal indication (n) Fetal indication (n) Combined indication (n) Antepartum corticosteroids (n)
nificantly higher in the study group than in both control groups. Of the 222 preeclamptic mothers 81 fulfilled the criteria of the HELLP-syndrome and four had eclampsia. The cesarean section rate was more than four times higher in the study group than in the control groups (P,0.0001), and was most often performed on the indication of fetal distress. A large proportion of mothers in the control groups, but not in the study group, received corticosteroids antepartum to accelerate fetal lung maturation (P,0.0001). There were no significant differences between both control groups.
Control groups I
II
n5222
n5222
n5222
165 (74.3)b 81 (36.5) 4 (1.8) 207 (93.7)b 19 (9.2) 167 (80.7) 21 (10.1) 4 (1.8)b
111 (50) 0 0 48 (21.6) 0 42 (87.5) 5 (12.5) 79 (36.1)
96 (43.2) 0 0 44 (19.8) 4 (9.1) 29 (65.9) 11 (25) 64 (29.1)
213
3.2. Neonatal outcome
a
Data are presented as numbers (%). b P,0.0001 between study group and control group I and between study group and control group II.
Table 2 compares neonatal outcome in the study group and the control groups. As previously defined, the study group and control group II were similar for gestational age at delivery (31.3 weeks). The median gestational age in control group I was 29.3 weeks, a difference of 2 weeks compared with the study group (P,0.001). Median birthweight in the preeclamptic group was approximately 600 g lower than that in control group II, delivered at the same gestational age (P,0.001). Of the infants of the study group 52% were small-for-gestational
3. Results
3.1. Obstetric characteristics Obstetric characteristics of the study population and the control groups are summarized in Table 1. As to be expected, the proportion of nulliparous women was sigTable 2 Neonatal outcome a Study group
Control groups I
n5222 Gestational age at birth (weeks) Male (n) Birth weight (g) Small-for-gestational age (n) Neonatal mortality (n) Infant mortality (n) Apgar score (5 min) Umbilical arterial pH Umbilical arterial BE Neonatal hematocrit (first 24 h) (l l 21 ) Admission to neonatal ICU (n) Stay in neonatal ICU (days) Hospital stay (days) Hypotension First 48 h (n) After 48 h (n) Total (n) Mechanical ventilation (n) Duration of mechanical ventilation (days) O 2 without mechanical ventilation (n) Duration O 2 without mechanical ventilation (days) Surfactant (n) a
b
n5222 d
31.3 (26–37) 88 (39.6) 1180 (480–2295)b,f 116 (52.3)b,f 18 (8.1) 29 (13.1) 8 (0–10) 7.20 (6.80–7.36)b,f 28.0 (224.5 to 17.0) 0.61 (0.36–0.81)b,f 181 (83)c,g 15 (,1–282)c,f 55 (,1–282)f
29.3 (24–32) 88 (39.6) 1300 (565–2650)d 7 (3.2) 27 (12.2) 30 (13.5) 8 (0–10) 7.27 (6.68–7.44) 26.0 (222.0 to 12.0) 0.51 (0.11–0.81)e 203 (92.7)d 17 (,1–455)d 51 (,1–751)d
31.3 (26–37) 88 (39.6) 1770 (750–4125) 7 (3.2) 28 (12.6) 33 (14.9) 8 (0–10) 7.26 (6.81–7.48) 26.7 (224.0 to 17.7) 0.54 (0.25–0.87) 157 (71.7) 10 (,1–150) 35 (,1–193)
28 (12.9) 24 (11.3) 41 (18.5) 127 (58.5)b 5 (,1–133)f 126 (56.8)c 5.5 (1–465)c 28 (12.8)g
39 (18.3) 23 (10.9) 50 (22.5) 187 (85)d 3 (,1–244)d 155 (69.8)d 12 (1–730)d 14 (6.5)
25 (11.5) 10 (4.6) 30 (13.5) 131 (59.5) 2 (,1–69) 102 (45.9) 8 (1–202) 12 (5.5)
Data are presented as median (range) or numbers (%). P value,0.001 between study group and control group I. c P value,0.01 between study group and control group I. d P value,0.001 between control group I and control group II. e P value,0.01 between control group I and control group II. f P value,0.001 between study group and control group II. g P value,0.01 between study group and control group II. b
II
n5222
214
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215
age, compared to 3.2% in both control groups (P,0.001). The study group had the lowest neonatal mortality, but the differences with the control groups did not reach statistical significance, and infant mortality was similar in all groups. The main causes of death were respiratory insufficiency, in combination with bronchopulmonary dysplasia, hyaline membrane disease, lung hypoplasia and circulatory problems. There were no differences in the distribution of causes of death between the groups. The significantly lower median values of umbilical arterial pH as well as the higher neonatal hematocrit in the study group in comparison with both control groups suggest chronic hypoxia in the fetuses of the study group. On the other hand the median Apgar scores after 5 min were not different between the three groups. The median hospital stays of the neonates of the study group and of control group I were not different, but they were significantly longer than that of control group II (P,0.001). Of the control group I infants 93% were admitted to the neonatal intensive care unit, and they stayed the longest time. No differences were found between groups with regard to the number of infants that became hypotensive. Significantly fewer infants from preeclamptic mothers than from mothers of control group I needed mechanical ventilation or oxygen without mechanical ventilation, and the duration of oxygen administration was shorter (P,0.01). More than twice as many infants in the study group than in control group II received surfactant treatment (P,0.01). Neonatal complications are summarized in Table 3. Bronchopulmonary dysplasia, patent ductus arteriosus and retinopathy were diagnosed more often in the study group
Table 3 Neonatal complications a
Hyaline membrane disease (n) Bronchopulmonary dysplasia (n) Pneumothorax (n) Patent ductus arteriosus (n) Sepsis (n) Pneumonia (n) Intracranial hemorrhage Total (n) Grade I–II (n) Grade III–IV (n) Grade unknown (n) Periventricular leucomalacia (n) Necrotizing enterocolitis (n) Retinopathy (n) a
Study group
Control groups I
II
n5222
n5222
n5222
79 33 11 63 28 11
(36.4) (14.9)e (5) (28.8)e (12.8) (5)
84 (38.7) 35 (15.8)d 22 (10.1) 60 (27.6)d 27 (12.4) 24 (11.1)d
60 (27.8) 16 (7.2) 17 (7.8) 37 (17) 27 (12.4) 9 (4.1)
19 (10.7)b 13 (5.9)b 5 (2.3) 1 13 (7.3) 5 (2.3) 10 (4.8)e
42 (22.8) 29 (13.1) 11 (5) 2 15 (8.2) 3 (1.4) 21 (10)c
26 (17) 17 (7.7) 7 (3.2) 2 9 (5.9) 2 (0.9) 0
Data are presented as numbers (%). P value,0.01 between study group and control group I. c P value,0.001 between control group I and control group II. d P value,0.01 between control group I and control group II. e P value,0.01 between study group and control group II. b
and control group I compared to control group II. A diagnosis of intracranial hemorrhage was significantly more frequent in control group I than in the study group, mainly caused by a more frequent occurrence of mild intracranial hemorrhage (grade I and II). No differences were found between groups with respect to the occurrence of hyaline membrane disease, pneumothorax, sepsis, pneumonia, periventricular leucomalacia and necrotizing enterocolitis.
4. Discussion Prolongation of pregnancy by 2 weeks in patients with early-onset preeclampsia has no apparent benificial effect on neonatal survival, although the power of the study may be insufficient to demonstrate a real difference. Neonatal mortality in our group of preeclamptic patients was similar to that reported in a group of selected preeclamptic patients with a duration of pregnancy of 29.1 weeks on admission and a neonatal mortality of 7.1% following temporizing management, but patients with growth-retarded fetuses were excluded; immediate delivery was associated with 17.1% mortality [10]. Sibai et al. [11] reported a randomized trial of aggressive (n546) vs. expectant (n549) management in preeclamptic patients between 28 and 32 weeks gestation. Expectant management was begun after a stable maternal condition was obtained during a 24-h observation period using magnesium sulfate, hydralazine or nifedipine. In that study no perinatal mortality occurred in both groups, but patients with fetal growth retardation and fetal distress were excluded. Because an appropriate control group of children of preeclamptic mothers who underwent immediate delivery was not available, we compared the infants of our study group with infants born from non-preeclamptic mothers. Only a small proportion of the neonates of the controls (7%) were growth retarded compared to a large proportion (52.3%) of the study group. Small-for-gestational age neonates have a lower rate of survival than appropriate-forgestational age babies, delivered at the same gestational age [12,13]. The high incidence of fetal growth retardation in our study is consistent with earlier reports [14,15]. Our study shows that 2 weeks prolongation of earlyonset preeclamptic pregnancy may reduce the need for assisted ventilation in the newborn, despite the fact that during the study period preeclamptic patients did not receive antenatal corticosteroids to accelerate fetal lung maturation, in contrast with a large proportion of the controls. Our initial reluctance to administer corticosteroids to hypertensive patients was based on the first randomized trial that showed an increased risk of fetal death in preeclamptic women treated with corticosteroids [16]. Later observational studies failed to confirm this risk [16] and because of convincing evidence of its efficacy in the prevention of the neonatal respiratory distress
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215
syndrome, antenatal corticosteroid therapy was introduced in the management of preeclamptic patients in our department in 1994. Although the preeclamptic patients did not receive antenatal corticosteroid treatment, the proportion of infants who needed mechanical ventilation was equal to that in the control group of infants delivered at the same gestational age by women who had received antenatal corticosteroids. A possible explanation could be that the preeclamptic fetus is subjected to ‘stress’ in utero and therefore has accelerated maturation [17]. It should be noted, however, that the study group needed a median of 5 days of mechanical ventilation, compared to 2 days in control group II. The longer duration of mechanical ventilation may explain the higher incidence of bronchopulmonary dysplasia in the study group compared with control group II. Although the gestational age was equal between the study population and control group II, patent ductus arteriosus occurred significantly more often in the study group, in agreement with findings in other studies [18,19]. The elevated incidence of patent ductus arteriosus may be explained by the higher number of SGA newborns in the study group, since there is evidence that the incidence of patent ductus arteriosus increases with decreasing birthweight within the same gestational age category [20]. Intracranial hemorrhage was most often diagnosed in the youngest gestational age group (control I). The 2 weeks prolongation of pregnancy by temporizing management was associated with a reduction in the risk of intraventricular hemorrhage, in agreement with previous studies indicating that a higher gestational age at birth reduces the risk of intraventricular hemorrhage [20,21]. Prolongation of pregnancy in preeclamptic women was associated with a reduction in the proportion of infants that needed intensive neonatal care, which was also found by Sibai et al. [11]. The newborns who did need intensive care treatment stayed longer in the intensive care unit than control newborns of similar gestational age. This is most likely explained by the lower birthweight of infants of preeclamptic patients. In conclusion, the results of this study indicate that prolongation of pregnancy in patients with early-onset preeclampsia may reduce neonatal morbidity, but infants born at the same gestational age without a preeclamptic mother still have a better prognosis. Further studies are necessary to evaluate the effects of prolongation of gestation in preeclamptic patients on the longterm development of the infant.
Acknowledgements We thank E. Rietveld for his contribution to our study and W.A. den Hollander-Huysman and P.C. Struijk for valuable advice.
215
References [1] Wallenburg HCS, Visser W. Pregnancy-induced hypertensive disorders. Current Opin Obstet Gynecol 1994;6:19–29. [2] Redman CWG. Pre-eclampsia and the placenta. Placenta 1991;12:301–8. [3] Romero R, Mazor M, Lockwood CJ et al. Clinical significance, prevalence, and natural history of thrombocytopenia in pregnancyinduced hypertension. Am J Perinatol 1989;6:32–8. [4] Derham RJ, Hawkins DF, de Vries LS, Aber VR, Elder MG. Outcome of pregnancies complicated by severe hypertension and delivered before 34 weeks; stepwise logistic regression analysis of prognostic factors. Br J Obstet Gynecol 1989;96:1173–81. [5] Visser W, Wallenburg HCS. Maternal and perinatal outcome of temporizing management in 254 consecutive patients with severe preeclampsia remote from term. Eur J Obstet Gynecol Reprod Biol 1995;63:147–54. [6] Visser W, Wallenburg HCS. Temporizing management of severe preeclampsia with and without the HELLP syndrome. Br J Obstet Gynaecol 1995;102:111–7. [7] Visser W, van Pampus MG, Treffers PE, Wallenburg HCS. Perinatal results of hemodynamic and conservative treatment in severe preeclampsia. Eur J Obstet Gynecol Reprod Biol 1994;53:175–81. [8] Kloosterman GJ. On intrauterine growth. Int J Gyneacol Obstet 1970;8:895–912. [9] Papile LA, Burnstein J, Burnstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34. ´ KS, Redman CWG, Gee H. Management of severe, early [10] Olah pre-eclampsia: is conservative management justified? Eur J Obstet Gynecol Reprod Biol 1993;51:175–80. [11] Sibai BM, Mercer BM, Schiff E, Friedman SA. Aggressive versus expectant management of severe preeclampsia at 28 to 32 weeks’ gestation: a randomized controlled trial. Am J Obstet Gynecol 1994;171:818–22. [12] Copper RL, Goldenberg RL, Creasy RK et al. A multicenter study of preterm birth weight and gestational age specific neonatal mortality. Am J Obstet Gynecol 1993;168:78–83. [13] Kok JH, den Ouden AL, Verloove-Vanhorick SP, Brand R. Outcome of very preterm small for gestational age infants: the first nine years of life. Br J Obstet Gynaecol 1998;105:162–8. [14] Long PA, Abell DA, Beischer NA. Fetal growth retardation and pre-eclampsia. Br J Obstet Gynaecol 1980;87:13–8. [15] Moore MP, Redman CWG. Case-control study of severe pre-eclampsia of early onset. Br Med J 1983;287:580–3. [16] Crowley PA. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol 1995;173:322–35. [17] Laatikainen TJ, Raisanen IJ, Salminen KR. Corticotropin-releasing hormone in amniotic fluid during gestation and labor and in relation to fetal lung maturation. Am J Obstet Gynecol 1988;159:891–5. [18] Brazy JE, Grimm JK, Little VA. Neonatal manifestations of severe maternal hypertension occuring before the thirty-sixth week of pregnancy. J Pediatr 1982;100:265–71. [19] Szymonowicz W, Yu VYH. Severe preeclampsia and infants of very low birth weight. Arch Dis Child 1987;62:712–6. [20] Robertson PA, Sniderman SH, Laros RK et al. Neonatal morbidity according to gestational age and birth weight from five tertiary care centers in the United States, 1983 through 1986. Am J Obstet Gynecol 1992;166:1629–44. [21] Allan WE, Philip AGS. Neonatal cerebral pathology diagnosed by ultrasound. Clin Perinatol 1985;12:195–218.
www.elsevier.com / locate / ejogrb
Neonatal outcome of temporizing treatment in early-onset preeclampsia ¨ I.J. Withagen, Willy Visser*, Henk C.S. Wallenburg Mariella Erasmus University School of Medicine and Health Sciences, Institute of Obstetrics and Gynecology, Rotterdam, The Netherlands Received 24 December 1999; received in revised form 2 February 2000; accepted 13 March 2000
Abstract Objective: To assess the effect of prolongation of pregnancy on neonatal outcome by means of hemodynamic treatment in patients with early-onset preeclampsia. Study design: A retrospective case-controlled study of 222 liveborn infants of patients with early-onset (24–31 weeks) preeclampsia, who underwent temporizing hemodynamic treatment. Of the two control groups of liveborn preterm infants of non-preeclamptic mothers one group was matched with the study group for gestational age on admission (group I), one for gestational age at birth (group II). Primary outcome measures were neonatal and infant mortality and variables of neonatal morbidity. Results: Median gestation in the study group of preeclamptic patients was prolonged from 29.3 to 31.3 weeks. No difference in neonatal or infant mortality was observed between infants from preeclamptic mothers and in the control groups. The study population showed better results than control group I with regard to admission to NICU (P,0.01), mechanical ventilation (P,0.001) and intracranial hemorrhage (P,0.01). Control group II had better results than the study group with respect to birthweight (P,0.001), bronchopulmonary dysplasia (P,0.01), patent ductus arteriosus (P,0.01), and retinopathy (P,0.01). Conclusion: Prolongation of gestation in patients with early-onset preeclampsia may reduce neonatal morbidity, but neonates of the same gestational age without a preeclamptic mother still have a better prognosis. 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Preeclampsia; Neonatal morbidity; Neonatal mortality; Temporizing hemodynamic treatment
1. Introduction Preeclampsia complicates 1–5% of all pregnancies and is directly or indirectly responsible for a large proportion of maternal and perinatal mortality and morbidity [1]. Although the cause of preeclampsia is unknown, the placenta is considered a key factor [2], and delivery of fetus and placenta is the only effective treatment. Delivery may benefit the mother but is not in the interest of the second patient, the fetus, if remote from term, since the main cause of neonatal mortality and morbidity in patients with severe preeclampsia is prematurity [3]. Duration of pregnancy at delivery was shown to be the primary *Corresponding author. Corresponding address: Academic Hospital Rotterdam / Sophia, Room Sk 4130, Dr. Molewaterplein 60, P.O. Box 2060, 3000 CB Rotterdam, The Netherlands. Fax: 131-10-463-6815. E-mail address: [email protected] (W. Visser).
prognostic factor for neonatal death and impairment in cases of severe early preeclampsia [4]. In 1985, a protocol of temporizing management of severe early-onset preeclampsia was instituted in the Department of Obstetrics of the University Hospital Rotterdam (AZR). The protocol aims at prolongation of pregnancy to avoid severely preterm birth and is based on correction of the disturbed maternal circulation with plasma volume expansion and pharmacologic vasodilatation, usually under central hemodynamic monitoring [5]. It was shown that such hemodynamic treatment in preeclamptic patients with a gestational age of 32 weeks or less and no fetal indication for delivery within 48 h, may prolong pregnancy with an average of 17 days, generally improves maternal condition, and may reduce perinatal mortality [5–7]. However, the effect of temporizing hemodynamic treatment of preeclampsia on neonatal wellbeing was not assessed in previous studies. For that reason, we
0301-2115 / 01 / $ – see front matter 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0301-2115( 00 )00332-8
212
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215
performed a retrospective case-controlled study with the aim to evaluate the effect of temporizing management of severe early-onset preeclampsia on neonatal outcome.
Preterm delivery in the control group was caused by spontaneous idiopathic preterm labor, with or without preterm rupture of membranes. Patients admitted for suspected placental abruption, fetal distress or fetal growth retardation were excluded.
2. Patients and methods
2.1. Selection of patients
2.3. End points of study
The study population consisted of 222 consecutive liveborn infants of patients with severe preeclampsia between 24 and 31 completed weeks of gestation, who underwent temporizing hemodynamic treatment in the Department of Obstetrics of the AZR between January 1, 1985 and December 31, 1993. Patients with twin pregnancy or known vascular, renal, hepatic or hematologic disease were excluded. Severe preeclampsia was defined as the occurrence of a diastolic blood pressure$110 mm Hg (Korotkoff 4) with proteinuria ($ 0.3 g / l / 24 h) after 20 weeks’ gestation, or the occurrence of persistent diastolic blood pressures of $90 mm Hg with proteinuria in combination with either HELLP syndrome or eclampsia. The HELLP syndrome was defined as the simultaneous occurrence of hemolysis (defined by peripheral blood smear), raised (.30 U l 21 ) serum aspartate aminotransferase (ASAT) and serum alanine aminotransferase (ALAT) and a platelet count of less than 100310 9 l 21 [5]. Gestational age was based on the last menstrual period and confirmed by early ultrasonography. Temporizing hemodynamic treatment consisted of bed rest and correction of the maternal circulation by means of pharmacologic vasodilatation (dihydralazine) and plasma volume expansion (pasteurized plasma), usually under central hemodynamic monitoring with the use of the Swan–Ganz thermodilution catheter [5]. The target value of diastolic intra-arterial blood pressure was 90 mm Hg and that of the pulmonary capillary wedge pressure was 10 mm Hg. Fetal condition was monitored by means of cardiotocography and ultrasound. Decisions to deliver were taken by the attending obstetricians on the basis of their interpretation of the cardiotocogram or of maternal condition. The details of the management protocol are described elsewhere [5].
Neonatal data were obtained by review of charts over a period of 1 year. Primary outcome measures were mortality and morbidity. Mortality was divided into neonatal mortality, occurring within the first 28 days of life, and infant mortality, occurring within the first year of life. Morbidity was defined by Apgar scores, umbilical arterial acid and base values, neonatal hematocrit determined in the first 24 h after birth, admission and length of stay in the neonatal intensive care unit (ICU), admission and length of stay in the hospital, and ventilation characteristics. Smallfor-gestational age (SGA) was defined as a birthweight below the 10th percentile of the Dutch reference curve corrected for parity and gender [8]. Pneumothorax was diagnosed by radiographic findings. A diagnosis of hyaline membrane disease was made in the presence of progressive respiratory distress in combination with a chest radiogram showing fine granular infiltrates persisting until at least the third day after birth. Bronchopulmonary dysplasia was defined as chronic respiratory distress and oxygen requirement beyond 28 days of life, accompanied by a chest radiograph that showed persistent streaks of increased density in both lungs interspersed with normal or hyperlucent areas. Intracranial hemorrhage and periventricular leucomalacia were diagnosed by ultrasonographic examination and graded I–IV [9]. Hypotension was defined if the infant had to be treated with vasoactive substances or plasma infusion. The diagnosis of patent ductus arteriosus was based on the presence of clinical symptoms and confirmed by echocardiographic examination. The diagnosis of necrotizing enterocolitis was confirmed by an abdominal radiograph. A diagnosis of septicemia and pneumonia required a positive blood culture. Retinopathy of prematurity was defined as abnormal vasoproliferation and tortuosity, retinal ridge, scarring and retinal detachment.
2.2. Selection of controls Each infant of the study group was matched for gestational age, gender and year of birth in a blinded fashion, without knowledge of neonatal characteristics and clinical outcome, with two liveborn infants of non-preeclamptic mothers: one delivered at the gestational age (61 week) at which temporizing treatment of the preeclamptic patient was initiated (control group I) and one delivered at the gestational age (61 week) at which the preeclamptic patient was delivered (control group II). Each control infant was matched only once.
2.4. Statistical analysis Data analysis was based on intention-to-treat and included all patients, also when they were delivered before the actual start of temporizing treatment. Differences in categorical variables between groups were assessed with the x 2 test and the Mann–Whitney U test was used to compare quantitative variables. Because of multiple comparisons a P value of 0.01 or less was considered to represent statistical significance.
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215 Table 1 Obstetric characteristics a Study group
Nulliparous (n) HELLP (n) Eclampsia (n) Cesarean section (n) Maternal indication (n) Fetal indication (n) Combined indication (n) Antepartum corticosteroids (n)
nificantly higher in the study group than in both control groups. Of the 222 preeclamptic mothers 81 fulfilled the criteria of the HELLP-syndrome and four had eclampsia. The cesarean section rate was more than four times higher in the study group than in the control groups (P,0.0001), and was most often performed on the indication of fetal distress. A large proportion of mothers in the control groups, but not in the study group, received corticosteroids antepartum to accelerate fetal lung maturation (P,0.0001). There were no significant differences between both control groups.
Control groups I
II
n5222
n5222
n5222
165 (74.3)b 81 (36.5) 4 (1.8) 207 (93.7)b 19 (9.2) 167 (80.7) 21 (10.1) 4 (1.8)b
111 (50) 0 0 48 (21.6) 0 42 (87.5) 5 (12.5) 79 (36.1)
96 (43.2) 0 0 44 (19.8) 4 (9.1) 29 (65.9) 11 (25) 64 (29.1)
213
3.2. Neonatal outcome
a
Data are presented as numbers (%). b P,0.0001 between study group and control group I and between study group and control group II.
Table 2 compares neonatal outcome in the study group and the control groups. As previously defined, the study group and control group II were similar for gestational age at delivery (31.3 weeks). The median gestational age in control group I was 29.3 weeks, a difference of 2 weeks compared with the study group (P,0.001). Median birthweight in the preeclamptic group was approximately 600 g lower than that in control group II, delivered at the same gestational age (P,0.001). Of the infants of the study group 52% were small-for-gestational
3. Results
3.1. Obstetric characteristics Obstetric characteristics of the study population and the control groups are summarized in Table 1. As to be expected, the proportion of nulliparous women was sigTable 2 Neonatal outcome a Study group
Control groups I
n5222 Gestational age at birth (weeks) Male (n) Birth weight (g) Small-for-gestational age (n) Neonatal mortality (n) Infant mortality (n) Apgar score (5 min) Umbilical arterial pH Umbilical arterial BE Neonatal hematocrit (first 24 h) (l l 21 ) Admission to neonatal ICU (n) Stay in neonatal ICU (days) Hospital stay (days) Hypotension First 48 h (n) After 48 h (n) Total (n) Mechanical ventilation (n) Duration of mechanical ventilation (days) O 2 without mechanical ventilation (n) Duration O 2 without mechanical ventilation (days) Surfactant (n) a
b
n5222 d
31.3 (26–37) 88 (39.6) 1180 (480–2295)b,f 116 (52.3)b,f 18 (8.1) 29 (13.1) 8 (0–10) 7.20 (6.80–7.36)b,f 28.0 (224.5 to 17.0) 0.61 (0.36–0.81)b,f 181 (83)c,g 15 (,1–282)c,f 55 (,1–282)f
29.3 (24–32) 88 (39.6) 1300 (565–2650)d 7 (3.2) 27 (12.2) 30 (13.5) 8 (0–10) 7.27 (6.68–7.44) 26.0 (222.0 to 12.0) 0.51 (0.11–0.81)e 203 (92.7)d 17 (,1–455)d 51 (,1–751)d
31.3 (26–37) 88 (39.6) 1770 (750–4125) 7 (3.2) 28 (12.6) 33 (14.9) 8 (0–10) 7.26 (6.81–7.48) 26.7 (224.0 to 17.7) 0.54 (0.25–0.87) 157 (71.7) 10 (,1–150) 35 (,1–193)
28 (12.9) 24 (11.3) 41 (18.5) 127 (58.5)b 5 (,1–133)f 126 (56.8)c 5.5 (1–465)c 28 (12.8)g
39 (18.3) 23 (10.9) 50 (22.5) 187 (85)d 3 (,1–244)d 155 (69.8)d 12 (1–730)d 14 (6.5)
25 (11.5) 10 (4.6) 30 (13.5) 131 (59.5) 2 (,1–69) 102 (45.9) 8 (1–202) 12 (5.5)
Data are presented as median (range) or numbers (%). P value,0.001 between study group and control group I. c P value,0.01 between study group and control group I. d P value,0.001 between control group I and control group II. e P value,0.01 between control group I and control group II. f P value,0.001 between study group and control group II. g P value,0.01 between study group and control group II. b
II
n5222
214
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215
age, compared to 3.2% in both control groups (P,0.001). The study group had the lowest neonatal mortality, but the differences with the control groups did not reach statistical significance, and infant mortality was similar in all groups. The main causes of death were respiratory insufficiency, in combination with bronchopulmonary dysplasia, hyaline membrane disease, lung hypoplasia and circulatory problems. There were no differences in the distribution of causes of death between the groups. The significantly lower median values of umbilical arterial pH as well as the higher neonatal hematocrit in the study group in comparison with both control groups suggest chronic hypoxia in the fetuses of the study group. On the other hand the median Apgar scores after 5 min were not different between the three groups. The median hospital stays of the neonates of the study group and of control group I were not different, but they were significantly longer than that of control group II (P,0.001). Of the control group I infants 93% were admitted to the neonatal intensive care unit, and they stayed the longest time. No differences were found between groups with regard to the number of infants that became hypotensive. Significantly fewer infants from preeclamptic mothers than from mothers of control group I needed mechanical ventilation or oxygen without mechanical ventilation, and the duration of oxygen administration was shorter (P,0.01). More than twice as many infants in the study group than in control group II received surfactant treatment (P,0.01). Neonatal complications are summarized in Table 3. Bronchopulmonary dysplasia, patent ductus arteriosus and retinopathy were diagnosed more often in the study group
Table 3 Neonatal complications a
Hyaline membrane disease (n) Bronchopulmonary dysplasia (n) Pneumothorax (n) Patent ductus arteriosus (n) Sepsis (n) Pneumonia (n) Intracranial hemorrhage Total (n) Grade I–II (n) Grade III–IV (n) Grade unknown (n) Periventricular leucomalacia (n) Necrotizing enterocolitis (n) Retinopathy (n) a
Study group
Control groups I
II
n5222
n5222
n5222
79 33 11 63 28 11
(36.4) (14.9)e (5) (28.8)e (12.8) (5)
84 (38.7) 35 (15.8)d 22 (10.1) 60 (27.6)d 27 (12.4) 24 (11.1)d
60 (27.8) 16 (7.2) 17 (7.8) 37 (17) 27 (12.4) 9 (4.1)
19 (10.7)b 13 (5.9)b 5 (2.3) 1 13 (7.3) 5 (2.3) 10 (4.8)e
42 (22.8) 29 (13.1) 11 (5) 2 15 (8.2) 3 (1.4) 21 (10)c
26 (17) 17 (7.7) 7 (3.2) 2 9 (5.9) 2 (0.9) 0
Data are presented as numbers (%). P value,0.01 between study group and control group I. c P value,0.001 between control group I and control group II. d P value,0.01 between control group I and control group II. e P value,0.01 between study group and control group II. b
and control group I compared to control group II. A diagnosis of intracranial hemorrhage was significantly more frequent in control group I than in the study group, mainly caused by a more frequent occurrence of mild intracranial hemorrhage (grade I and II). No differences were found between groups with respect to the occurrence of hyaline membrane disease, pneumothorax, sepsis, pneumonia, periventricular leucomalacia and necrotizing enterocolitis.
4. Discussion Prolongation of pregnancy by 2 weeks in patients with early-onset preeclampsia has no apparent benificial effect on neonatal survival, although the power of the study may be insufficient to demonstrate a real difference. Neonatal mortality in our group of preeclamptic patients was similar to that reported in a group of selected preeclamptic patients with a duration of pregnancy of 29.1 weeks on admission and a neonatal mortality of 7.1% following temporizing management, but patients with growth-retarded fetuses were excluded; immediate delivery was associated with 17.1% mortality [10]. Sibai et al. [11] reported a randomized trial of aggressive (n546) vs. expectant (n549) management in preeclamptic patients between 28 and 32 weeks gestation. Expectant management was begun after a stable maternal condition was obtained during a 24-h observation period using magnesium sulfate, hydralazine or nifedipine. In that study no perinatal mortality occurred in both groups, but patients with fetal growth retardation and fetal distress were excluded. Because an appropriate control group of children of preeclamptic mothers who underwent immediate delivery was not available, we compared the infants of our study group with infants born from non-preeclamptic mothers. Only a small proportion of the neonates of the controls (7%) were growth retarded compared to a large proportion (52.3%) of the study group. Small-for-gestational age neonates have a lower rate of survival than appropriate-forgestational age babies, delivered at the same gestational age [12,13]. The high incidence of fetal growth retardation in our study is consistent with earlier reports [14,15]. Our study shows that 2 weeks prolongation of earlyonset preeclamptic pregnancy may reduce the need for assisted ventilation in the newborn, despite the fact that during the study period preeclamptic patients did not receive antenatal corticosteroids to accelerate fetal lung maturation, in contrast with a large proportion of the controls. Our initial reluctance to administer corticosteroids to hypertensive patients was based on the first randomized trial that showed an increased risk of fetal death in preeclamptic women treated with corticosteroids [16]. Later observational studies failed to confirm this risk [16] and because of convincing evidence of its efficacy in the prevention of the neonatal respiratory distress
M.I. J. Withagen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 94 (2001) 211 – 215
syndrome, antenatal corticosteroid therapy was introduced in the management of preeclamptic patients in our department in 1994. Although the preeclamptic patients did not receive antenatal corticosteroid treatment, the proportion of infants who needed mechanical ventilation was equal to that in the control group of infants delivered at the same gestational age by women who had received antenatal corticosteroids. A possible explanation could be that the preeclamptic fetus is subjected to ‘stress’ in utero and therefore has accelerated maturation [17]. It should be noted, however, that the study group needed a median of 5 days of mechanical ventilation, compared to 2 days in control group II. The longer duration of mechanical ventilation may explain the higher incidence of bronchopulmonary dysplasia in the study group compared with control group II. Although the gestational age was equal between the study population and control group II, patent ductus arteriosus occurred significantly more often in the study group, in agreement with findings in other studies [18,19]. The elevated incidence of patent ductus arteriosus may be explained by the higher number of SGA newborns in the study group, since there is evidence that the incidence of patent ductus arteriosus increases with decreasing birthweight within the same gestational age category [20]. Intracranial hemorrhage was most often diagnosed in the youngest gestational age group (control I). The 2 weeks prolongation of pregnancy by temporizing management was associated with a reduction in the risk of intraventricular hemorrhage, in agreement with previous studies indicating that a higher gestational age at birth reduces the risk of intraventricular hemorrhage [20,21]. Prolongation of pregnancy in preeclamptic women was associated with a reduction in the proportion of infants that needed intensive neonatal care, which was also found by Sibai et al. [11]. The newborns who did need intensive care treatment stayed longer in the intensive care unit than control newborns of similar gestational age. This is most likely explained by the lower birthweight of infants of preeclamptic patients. In conclusion, the results of this study indicate that prolongation of pregnancy in patients with early-onset preeclampsia may reduce neonatal morbidity, but infants born at the same gestational age without a preeclamptic mother still have a better prognosis. Further studies are necessary to evaluate the effects of prolongation of gestation in preeclamptic patients on the longterm development of the infant.
Acknowledgements We thank E. Rietveld for his contribution to our study and W.A. den Hollander-Huysman and P.C. Struijk for valuable advice.
215
References [1] Wallenburg HCS, Visser W. Pregnancy-induced hypertensive disorders. Current Opin Obstet Gynecol 1994;6:19–29. [2] Redman CWG. Pre-eclampsia and the placenta. Placenta 1991;12:301–8. [3] Romero R, Mazor M, Lockwood CJ et al. Clinical significance, prevalence, and natural history of thrombocytopenia in pregnancyinduced hypertension. Am J Perinatol 1989;6:32–8. [4] Derham RJ, Hawkins DF, de Vries LS, Aber VR, Elder MG. Outcome of pregnancies complicated by severe hypertension and delivered before 34 weeks; stepwise logistic regression analysis of prognostic factors. Br J Obstet Gynecol 1989;96:1173–81. [5] Visser W, Wallenburg HCS. Maternal and perinatal outcome of temporizing management in 254 consecutive patients with severe preeclampsia remote from term. Eur J Obstet Gynecol Reprod Biol 1995;63:147–54. [6] Visser W, Wallenburg HCS. Temporizing management of severe preeclampsia with and without the HELLP syndrome. Br J Obstet Gynaecol 1995;102:111–7. [7] Visser W, van Pampus MG, Treffers PE, Wallenburg HCS. Perinatal results of hemodynamic and conservative treatment in severe preeclampsia. Eur J Obstet Gynecol Reprod Biol 1994;53:175–81. [8] Kloosterman GJ. On intrauterine growth. Int J Gyneacol Obstet 1970;8:895–912. [9] Papile LA, Burnstein J, Burnstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34. ´ KS, Redman CWG, Gee H. Management of severe, early [10] Olah pre-eclampsia: is conservative management justified? Eur J Obstet Gynecol Reprod Biol 1993;51:175–80. [11] Sibai BM, Mercer BM, Schiff E, Friedman SA. Aggressive versus expectant management of severe preeclampsia at 28 to 32 weeks’ gestation: a randomized controlled trial. Am J Obstet Gynecol 1994;171:818–22. [12] Copper RL, Goldenberg RL, Creasy RK et al. A multicenter study of preterm birth weight and gestational age specific neonatal mortality. Am J Obstet Gynecol 1993;168:78–83. [13] Kok JH, den Ouden AL, Verloove-Vanhorick SP, Brand R. Outcome of very preterm small for gestational age infants: the first nine years of life. Br J Obstet Gynaecol 1998;105:162–8. [14] Long PA, Abell DA, Beischer NA. Fetal growth retardation and pre-eclampsia. Br J Obstet Gynaecol 1980;87:13–8. [15] Moore MP, Redman CWG. Case-control study of severe pre-eclampsia of early onset. Br Med J 1983;287:580–3. [16] Crowley PA. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol 1995;173:322–35. [17] Laatikainen TJ, Raisanen IJ, Salminen KR. Corticotropin-releasing hormone in amniotic fluid during gestation and labor and in relation to fetal lung maturation. Am J Obstet Gynecol 1988;159:891–5. [18] Brazy JE, Grimm JK, Little VA. Neonatal manifestations of severe maternal hypertension occuring before the thirty-sixth week of pregnancy. J Pediatr 1982;100:265–71. [19] Szymonowicz W, Yu VYH. Severe preeclampsia and infants of very low birth weight. Arch Dis Child 1987;62:712–6. [20] Robertson PA, Sniderman SH, Laros RK et al. Neonatal morbidity according to gestational age and birth weight from five tertiary care centers in the United States, 1983 through 1986. Am J Obstet Gynecol 1992;166:1629–44. [21] Allan WE, Philip AGS. Neonatal cerebral pathology diagnosed by ultrasound. Clin Perinatol 1985;12:195–218.