Neonatal outcome in hypertensive disorders of pregnancy

Early Human Development 87 (2011) 445–449

Contents lists available at ScienceDirect

Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev

Neonatal outcome in hypertensive disorders of pregnancy Sergio Ferrazzani a, Rita Luciano b, Serafina Garofalo a, Vito D'Andrea b,⁎, Sara De Carolis a, Maria Pia De Carolis b, Valentina Paolucci b, Costantino Romagnoli b, Alessandro Caruso a a b

Department of Obstetrics and Gynecology, Catholic University of the Sacred Heart, Rome, Italy Division of Neonatology, Department of Pediatrics, Catholic University of the Sacred Heart, Rome, Italy

a r t i c l e

i n f o

Article history: Received 30 November 2010 Received in revised form 8 March 2011 Accepted 17 March 2011 Keywords: Hypertension Pregnancy Neonatal outcome

a b s t r a c t Background: Hypertensive disorders in pregnancy account for increased perinatal morbidity and mortality when compared to uneventful gestations. Aims: To analyze perinatal outcome of pregnancies complicated by different kinds of hypertension to uncomplicated pregnancies in a series of Italian women and to compare our data with series from other countries. Study design: The sample was divided into four groups of hypertensive women: chronic hypertension (CH), gestational hypertension (GH), preeclampsia (PE), and chronic hypertension complicated by preeclampsia (CHPE). One thousand normal pregnancies served as controls. Subjects: Neonatal features of the offspring of 965 Italian women with hypertension in pregnancy were evaluated. Measures: Gestational age, birthweight and the rate of small for gestational age were the outcomes. Perinatal asphyxia and mortality were also assessed. Results: Gestational age, the mean of birth weight and birth percentile were significantly lower in all groups with hypertensive complications when compared with controls. The rate of very early preterm delivery (b 32 weeks) was 7.8% in CH, 5.9% in GH, 21.2% in PE and 37.2% in CHPE while it was to 1.2% in the control group. The rate of SGA was globally 16.2% in CH, 22.8% in GH, 50.7% in PE, 37.2% in CHPE and 5% in controls. The rate of SGA in PE was much higher than reported in series from other countries. Conclusion: Comparing our data with those reported from other countries, it is evident that the rate of fetal growth restriction in PE we found in our center, is significantly higher even in the presence of a global lower incidence of PE. © 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Hypertensive disorders in pregnancy account for increased perinatal morbidity and mortality when compared with uneventful gestations. Hypertension may be pre-existent, may develop during pregnancy or both conditions may concur. The outcome of pregnancy depends on the type of disorder concerned [1–6]. The relationship between proteinuria, maternal hypertension and perinatal complications was already emphasized three decades ago [5]. An increased rate of still-birth and low birth-weight (LBW) infants was demonstrated to occur when the two symptoms are associated. We reported similar findings in a study concerning 444 pregnancies showing that the occurrence of proteinuria in hypertensive disorders was a negative factor for neonatal outcome and it could be considered as a clinical late marker of the vascular damage [6].

⁎ Corresponding author at: Division of Neonatology, Department of Pediatrics, Catholic University of the Sacred Heart, L.go A. Gemelli, 8, 00168, Rome, Italy. Tel.: + 39 0630154348; fax: + 39 063055301. E-mail address: [email protected] (V. D'Andrea). 0378-3782/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2011.03.005

The presence of proteinuria was considered the essential marker for poor pregnancy outcome and gestational and chronic hypertension were not included in most randomized trials designed to investigate preventive strategies for preeclampsia [7]. As a consequence, the frequency of adverse perinatal outcome in hypertensive disorders of pregnancy not complicated by proteinuria was less investigated in the past [8,9]. Recently, however, it was reported that gestational hypertension occurring early in pregnancy is associated with a higher rate of low birthweight infants and lower gestational age at delivery when compared with both late gestational hypertension and preeclampsia [10], raising the question whether other factors than proteinuria could be looked at as marker of adverse pregnancy outcome following maternal hypertension. Moreover, looking at series from different countries reported in the literature [6,11–16], strong dissimilarities were observed either in the prevalence of hypertension in pregnancy or in the associated perinatal outcome according to birth weight and rate of small-for-gestational age babies, suggesting a genetic polymorphism of the disorder. In the present study concerning a much larger series than our previous one, we selected only Italian pregnant women in order to compare the perinatal outcome after different kinds of hypertension with a control group of uncomplicated pregnancies in our institution, and also to make a comparison with series reported in the literature.

446

S. Ferrazzani et al. / Early Human Development 87 (2011) 445–449

2. Methods

Table 2 Perinatal outcome of study groups.

A series of 965 Italian women with hypertension in pregnancy admitted to the Department of Obstetrics and Gynecology of the Catholic University of Rome between 1986 and 1995 and registered on an electronic database were examined for the present study. This year interval was appropriately taken in order to have a homogeneous Italian white population sample before the beginning of the massive immigration of recent years. The sample was divided into four groups of hypertensive women according to the National High Blood Pressure Education Program Working Group Report on High Blood Pressure in Pregnancy [17]: chronic hypertension (CH), hypertension diagnosed before pregnancy and/or diastolic pressure ≥90 mm Hg and/or antihypertensive medication, each before the 20th week of pregnancy uncomplicated by de novo proteinuria; gestational hypertension (GH), women with diastolic pressure ≥90 mm Hg for the first time after 20 weeks of pregnancy and/or hospitalized during pregnancy because of hypertension and/or prescribed antihypertensive medication for the first time after 20 weeks without proteinuria; preeclampsia (PE), women different from GH because of the development of de novo proteinuria (≥0.3 g/24 h); and chronic hypertension complicated by preeclampsia (CHPE), women with chronic hypertension and de novo proteinuria after 20 weeks of pregnancy. Proteinuria was defined as one urine collection with a total protein excretion ≥300 mg/24 h or N1 g/l (or ++ with dipstick) in a random sample, without urinary infection. Blood pressure was taken with a standard mercury sphygmomanometer, using phases 1 and 5 of the Korotkoff sounds before delivery for systolic and diastolic blood pressure, respectively, and recording with the patient in a semi recumbent position. Gestational age (GA) was defined on the basis of the last maternal menstrual date confirmed by early ultrasound examination, patients with uncertain GA were not included. LBW was defined as a birth weight (BW) below 2500 g, very low birth weight (VLBW) was defined as a BW below 1500 g. A small for gestational age (SGA) infant was defined as a newborn infant with a BW below the 10th percentile according to a national standard curve for singleton birth. A large for gestational age (LGA) infant was defined as a newborn infant with a BW above the 90th percentile according to the national standard curve for singleton births as above. Perinatal mortality included stillbirths with a BW of 500 g or more and/or GA of 24 completed weeks or more, and deaths occurring within 7 days from delivery. Perinatal asphyxia was defined, in this context, as an Apgar score ≤3 at 1 min requiring oro-tracheal intubation. For the purpose of data analysis, information was extracted from the records and entered into a personal computer database. Information included aspects of prenatal care, previous medical history, complications, intra-partum care, delivery data, fetal outcome and post-partum course. Data entries were then accumulated and organized in tabular form on a spreadsheet database.

Controls CH GH PE CHPE

Perinatal asphyxia (%)

Perinatal mortality (%)

Days of hospitalization (mean ± SD)

0.6 11.9⁎ 8.5⁎ 20⁎ 19.6⁎

0.1 0.6 1.2⁎ 2.6⁎ 5.9⁎

3.6 ± 5.2 6.3 ± 11⁎ 6.9 ± 14.7⁎ 13.4 ± 20.4⁎ 16.9 ± 26.1⁎

The quantitative data are expressed as mean ± SD. ⁎ p b 0.0001 versus controls.

One thousand controls were randomly extracted from 20,000 selected healthy women who delivered in our department from 1986 to 1995. Statistical analysis included the Student t test for comparison of averages and the Chi-square or Fisher exact test when appropriate for comparison of frequencies. P-values b0.05 (two-tailed) were considered significant. 3. Results Table 1 shows the distribution of hypertension in the studied sample and the pregnancy outcome. Mean GA at birth, mean BW and birth percentile as well as the rate of preterm and very preterm deliveries, were significantly lower in each group of neonates born to hypertensive mothers compared to controls, being the difference more evident in the PE and CHPE groups. Perinatal outcome is shown in Table 2: according to prevalence of preterm delivery in the different groups, perinatal asphyxia was significantly more frequent and neonatal hospitalization was significantly longer in neonates born following any kind of maternal hypertensive disorder compared to the control group; perinatal mortality was significantly higher in all but CH hypertensive groups compared to controls. These perinatal complications were by far more evident in PE and CHPE groups. Table 3 shows the prevalence of preterm deliveries and SGA neonates in our series in comparison with previous reports. In Figs. 1 and 2 the rates of LBW and VLBW, respectively, among hypertensive groups are shown in comparison with controls. The rate of SGA infants was significantly higher in all the hypertensive groups when compared to controls. Moreover the prevalence of SGA neonates in all the hypertensive groups was related to the time of delivery whether or not it was before 37 weeks or at term. The rate of SGA neonates was not different in controls before 37 weeks or at term (4.4% versus 5.0%), while it was statistically different between term and preterm neonates in all the hypertensive groups (Fig. 3). Fig. 4 shows that the SGA rate b32 weeks was significantly higher in the whole hypertensive group compared to controls, but no significant difference was found between the various subgroups of hypertensive patients (CH = 43%, GH = 50%, PE = 54%, and CHPE = 39%).

Table 1 Distribution of hypertension and pregnancy outcome of study sample. n (%a) Controls CH GH PE CHPE

1000 154 (16a) 421 (44a) 339 (35a) 51 (5a)

GA

BW (g)

Birth percentile

Preterm delivery (%)

Very preterm delivery (%)

39.6 ± 1.7 37.8 ± 2.9⁎ 38.2 ± 2.7⁎ 35.4 ± 3.5⁎ 33.9 ± 3.8⁎

3380 ± 522 2914 ± 857⁎ 2831 ± 818⁎ 2065 ± 933⁎ 1778 ± 815⁎

55.9 ± 26.3 46.9 ± 30.5⁎ 40.9 ± 29.6⁎ 24.6 ± 26.6⁎ 20.7 ± 18.6⁎

7.2 33.1⁎ 27.4⁎ 68.4⁎ 76.5⁎

1.2 7.8⁎ 5.9⁎ 21.2⁎ 37.2⁎

The quantitative data are expressed as mean ± SD. a Percentages among hypertensive patients. ⁎ p b 0.0001 versus controls.

S. Ferrazzani et al. / Early Human Development 87 (2011) 445–449 Table 3 Comparative rate for small for gestational age (SGA) and preterm delivery from selected studies of pregnant woman with preeclampsia (PE). Primary author

Study design

Clausson et al. [12]

Countries

Ananth et al. [11] Ray et al. [13] Villar et al. [14]

Population based registry Retrospective cohort Prospective cohort Prospective cohort

Yucesoy et al. [15] Tranquilli et al. [16] Ferrazzani et al. [6] Current study

Retrospective Retrospective Retrospective Retrospective

cohort cohort cohort cohort

SGA in PE (%)

Preterm delivery in PE (%)

Sweden

14.7

–

USA Canada Argentina, Cuba, Saudi Arabia, and Thailand Turkey Italy Italy Italy

15.6 25 20

18 74 27

28 51.3 52 50.7

– – 58 78

4. Discussion According to previous studies, pregnancies complicated by hypertension, are characterized by an increased rate of preterm delivery, LBW and VLBW infants, compared with normal pregnancies [13,18–20]. This is true in all groups of hypertensive women independent on the presence or absence of proteinuria. Prematurity alone does not justify the whole amount of LBW and VLBW. Maternal hypertension accounts for a reduced utero-placental perfusion and, consequently, impaired fetal growth [21–23]. In fact, a high rate of SGA infants was found when GA at birth is taken into account and then either preterm birth or intrauterine growth restriction or the combination of both are responsible for the low and very low birth weight registered among these neonates. The

* 68

100

* 26

50

100 80

* 74

* 51

* 39

44

38 37

40 23

16

15

20

9

5 4 5

0 *p<0.0001 versus =37 wks

Fig. 3. SGA rate in the hypertensive group and in controls (total, b37 weeks, N37 weeks).

frequency of fetal growth restriction according to the different type of hypertension is, however, controversial [24,25]. The present study shows that pregnancies complicated by PE are associated with the highest rate of SGA confirming the strong influence of maternal proteinuria on fetal outcome in hypertensive diseases. It may be explained by the presence of a higher rate of placental insufficiency in PE [6]. Interesting enough, however, we found a significant higher rate of SGA infants even among pregnant women with chronic and gestational hypertension compared to controls. Analyzing the distribution of SGA neonates among the hypertensive pregnancies groups after stratification for preterm and term birth we support, according to previous studies [26,27], the occurrence of two subtypes of preeclampsia: preterm PE, with higher rate of SGA and consistent with the prevailing “hypoperfusion model” and term PE showing a normal placental function and a lower rate of SGA. It appears more evident observing the bimodal distribution rate of the weeks of delivery among the women with preeclampsia. This distributions, in fact, show the presence of two partially overlying Gaussian curves: the early one with the peak at 31 weeks (preterm PE), and the late one with the peak at 37 weeks (term PE). This bimodal pattern is also evident looking at the distribution in the other hypertensive groups (not shown). Actually, the rate of SGA infants among all forms of hypertension was similar before 32 weeks. We hypothesize that

100 CONTROLS CH

* 27

* 70

* 56

60

20

The LGA rate was similar in controls (5.8%), GH (5.5%) and PE (5.3%) while it was significantly higher in CH (13.8%) compared to controls and the other groups. Fig. 5 shows the distribution rate of the weeks of delivery among the women with preeclampsia. This distribution rate was bimodal, with two peaks at 31 weeks and at 37 weeks.

447

*

80

GH

60

PE

40

CHPE

43

*

50

CONTROLS

*

54

CH

*

39

GH PE CHPE

1

20

5

0 *p<0.0001 versus controls

0

Fig. 4. SGA rate in the hypertensive groups versus controls (b32 weeks).

* p<0.0001 versus Controls

Fig. 1. LBW rate in the hypertensive groups versus controls.

100

CONTROLS

80 60 40 20

1

* 5

* 8

* 32

* 31

CH GH PE CHPE

0 * p<0.0001 versus Controls

Fig. 2. VLBW rate in the hypertensive groups versus controls.

Fig. 5. Distribution rate of the weeks of delivery among the women with preeclampsia.

448

S. Ferrazzani et al. / Early Human Development 87 (2011) 445–449

hypertension complicated by severe early onset fetal growth restriction should be considered as a preeclampsia corresponding to the placental hypoperfusion model. In these cases when a preterm delivery is requested due to a severe fetal growth restriction, the preterm delivery anticipates the occurrence of proteinuria that would occur later on in the gestation [28]. As a consequence, one might conclude that the diagnostic criteria for PE should include fetal growth restriction even in the absence of proteinuria when a gestational age less than 32 weeks is considered [29]. Buchbinder et al. studied a cohort of women affected by preeclampsia in previous gestation: they found similar perinatal adverse outcomes in patients who developed severe preeclampsia and in patients who developed severe hypertension without proteinuria. Both these groups of women had a significantly increased risk of preterm delivery and of delivery of SGA infants with respect to women who developed mild preeclampsia. They argued that proteinuria might have developed if a preterm delivery would not have been necessary due to the maternal pathology [30]. Moreover, recent studies indicate that in most cases of term PE birth weight is normal and more infants than expected are LGA [27,31–33]. These findings suggest that placental dysfunction is absent or plays a minor role in a subset of term preeclamptic pregnancies. In our series a higher rate of LGA was only found in patients with uncomplicated chronic hypertension. Finally, our analysis also shows a higher rate of perinatal morbidity and mortality in hypertensive pregnant women compared to controls. It is well known that preeclampsia increases the risk of severe perinatal outcome, mostly by reducing GA and BW. However, in agreement with other recent studies, we found that also GH and CH increase the risk of fetal death, neonatal morbidity and mortality [13,34–37]. Disagreement between findings in this field could be at least partially justified by dissimilarities among hypertensive pregnancies from different countries. Comparing our data with those reported from other countries, it is evident that the rate of fetal growth restriction in PE we found in our center, was significantly higher even in the presence of a global lower incidence of PE (Table 3). This finding might be partially explained by a higher rate of preterm deliveries (Table 3) in our sample according to the preterm hypoperfusion model of PE. However, a significant lower rate of SGA infants is also found in studies with a similar rate of preterm births. We hypothesize that a subset of preeclampsia characterized by a higher rate of the “hypoperfusion model” might be more or less frequently encountered according to regional or racial distribution. In conclusion, the high number of subjects included in the study offers a contribution to the amount of data demonstrating that hypertension in pregnancy is a broad spectrum of diseases not only defined by the presence or absence of proteinuria but also characterized by a more or less precocious onset of growth retardation. Moreover, preeclampsia does not seem to be a disease with a uniform distribution in the world and different patterns of PE can be found in various countries. This concept should be always taken into account when comparison of results is done among studies or trials from different countries or populations. 5. Ethical statement I declare that I participated in the design, execution, and analysis of the paper by

and colleagues entitled:

that I have seen and approved the final version and that it has neither been published nor submitted elsewhere. I also declare that I have no conflict of interest, other than any noted in the covering letter to the editor. All authors should have made substantial contributions to all of the following: (1) the conception and design of the study, or acquisition of data, or analysis and interpretation of data, (2) drafting the article or revising it critically for important intellectual content, (3) final approval of the version to be submitted. References [1] Lin CC, Lindheimer MD, River P, Moawad AH. Fetal outcome in hypertensive disorders of pregnancy. Am J Obstet Gynecol 1982;142:255–60. [2] Sibai BM, Abdella TN, Anderson GD. Pregnancy outcome in 211 patients with mild chronic hypertension. Obstet Gynecol 1983;61:571–6. [3] Sibai BM, El-Nazer A, Gonzalez-Riuz A. Severe preeclampsia–eclampsia in young primigravid women. Subsequent pregnancy outcome and remote prognosis. Am J Obstet Gynecol 1986;155:1011–6. [4] Gleicher N, Boler LR, Norusis M, Del Granado A. Hypertensive diseases of pregnancy and parity. Am J Obstet Gynecol 1986;154:1044–9. [5] Page EW, Christianson R. Influence of blood pressure changes with and without proteinuria upon outcome of pregnancy. Am J Obstet Gynecol 1976;126:821–9. [6] Ferrazzani S, Caruso A, De Carolis S, Martino IV, Mancuso S. Proteinuria and outcome of 444 pregnancies complicated by hypertension. Am J Obstet Gynecol 1990;162:366–71. [7] Duley L, Henderson Smart DJ, Meher S, King JF. Antiplatelet agents for prevent preeclampsia and its complication. Cochrane Database Syst Rev 2007;18(2): CD004659. [8] Rey E, Couturier A. The prognosis of pregnancy in women with chronic hypertension. Am J Obstet Gynecol 1994;171:410–6. [9] Haelterman E, Bréart G, Paris-Llado J, Dramaix M, Tchobroutsky C. Effect of uncomplicated chronic hypertension on the risk of small-for-gestational age birth. Am J Epidemiol 1997;145:689–95. [10] Xiong X, Demianczuk NN, Saunders LD, Wang FL, Fraser WD. Impact of preeclampsia and gestational hypertension on birth weight by gestational age. Am J Epidemiol 2002;155(3):203–9. [11] Ananth CV, Peedicayil A, Savitz DA. Effect of hypertensive diseases in pregnancy on birthweight, gestational duration and small-for gestational age births. Epidemiology 1995;6:391–5. [12] Clausson B, Cnattingious S, Axelsson O. Preterm an term births of small for gestational age infants; a population-based study of risk factors among nulliparous women. Br J Obstet Gynecol 1998;105:1011–7. [13] Ray JG, Burrows RF, Burrows EA, Vermeulen MJ. MOS HIP: McMaster outcome study of hypertension in pregnancy. Early Hum Dev 2001;64:129–43. [14] Villar J, Carroli G, Wojdyla D, Abalos E, Giordano D, Ba'aqeel H, et al. World Health Organization Antenatal Care Trial Research Group: preeclampsia, gestational hypertension, and intrauterine growth restriction, related or independent conditions? Am J Obstet Gynecol 2006;194:921–31. [15] Yücesoy G, Ozkan S, Bodur H, Tan T, Calişkan E, Vural B, et al. Maternal and perinatal outcome in pregnancies complicated with hypertensive disorders of pregnancy: a seven year experience of a tertiary care center. Arch Gynecol Obstet 2005;273:43–9. [16] Tranquilli AL, Giannubilo SR. The weight of fetal growth restriction in 437 hypertensive pregnancies. Arch Gynecol Obstet 2004;270:214–6. [17] National High Blood Pressure Education Program Working Group report on high blood pressure in pregnancy. Am J Obstet Gynecol 2000;183:S1–S22.

S. Ferrazzani et al. / Early Human Development 87 (2011) 445–449 [18] Xiong X, Fraser WD. Impact of pregnancy-induced hypertension on birthweight by gestational age. Paediatr Perinat Epidemiol 2004;18:186–91. [19] Xiong X, Mayes D, Demianczuk N, Olson DM, Davidge ST, Newburn-Cook C, et al. Impact of pregnancy-induced hypertension on fetal growth. Am J Obstet Gynecol 1999;180:207–13. [20] Langenveld J, Jansen S, van der Post J, Wolf H, Mol BW, Ganzevoort W. Recurrence risk of a delivery before 34 weeks of pregnancy due to an early onset hypertensive disorder: a systematic review. Am J Perinatol 2010;27(7):565–71. [21] Friedman SA, Taylor RN, Roberts JM. Pathophysiology of preeclampsia. Clin Perinatol 1991;18:661–82. [22] van Beek E, Peeters LL. Pathogenesis of preeclampsia: a comprehensive model. Obstet Gynecol Surv 1998;53:233–9. [23] Misra DP. The effect of the pregnancy-induced hypertension on fetal growth: a review of the literature. Paediatr Perinat Epidemiol 1996;10:244–63. [24] Srinivas SK, Edlow AG, Neff PM, Sammel MD, Andrela CM, Elovitz MA. Rethinking IUGR in preeclampsia: dependent or independent of maternal hypertension? J Perinatol 2009;29(10):680–4. [25] Mabie WC, Pernoll ML, Biswas MK. Chronic hypertension in pregnancy. Obstet Gynecol 1986;67:197–205. [26] Von Dadelszen P, Magee LA, Roberts JM. Subclassification of preeclampsia. Hypertens Pregnancy 2003;22:143–8. [27] Vatten LJ, Skjaerven R. Is preeclampsia more than one disease? Br J Obstet Gynaecol 2004;111:298–302. [28] Jelin AC, Cheng YW, Shaffer BL, Kaimal AJ, Little SE, Caughey AB. Early-onset preeclampsia and neonatal outcomes. J Matern Fetal Neonatal Med 2010;23(5): 389–92.

449

[29] Thornton CE, Makris A, Ogle RF, Tooher JM, Hennessy A. Role of proteinuria in defining pre-eclampsia: clinical outcomes for women and babies. Clin Exp Pharmacol Physiol 2010;37(4):466–70. [30] Buchbinder A, Sibai BM, Caritis S, Macpherson C, Hauth J, Lindheimer MD, et al. National Institute of Child Health and Human Development Network of Maternal– Fetal Medicine Units. Adverse perinatal outcomes are significantly higher in severe gestational hypertension than in mild preeclampsia. Am J Obstet Gynecol 2002;186(1):66–71. [31] Rasmussen S, Irgens LM. Fetal growth and body proportion in preeclampsia. Obstet Gynecol 2003;101:575–83. [32] Xiong X, Deminianczuk NN, Buekens P, Saunders LD. Association of preeclampsia with high birth weight for age. Am J Obstet Gynecol 2000;183:148–55. [33] Odegard RA, Vatten LJ, Nilsen ST, Salvesen KA, Austgulen R. Preeclampsia and fetal growth. Obstet Gynecol 2000;96:950–5. [34] Sibai BM, Anderson GD. Pregnancy outcome of intensive therapy in severe hypertension in first trimester. Obstet Gynecol 1986;67:517–22. [35] Vigil-De Garacia P, Lasso M, Montufar-Rueda C. Perinatal outcome in women with severe chronic hypertension during the second half of pregnancy. Int J Gynecol Obstet 2004;85:139–44. [36] McCowan LM, Buist RG, North RA, Gamble G. Perinatal morbidity in chronic hypertension. Br J Obstet Gynecol 1996;103:123–9. [37] Hall DR, Odendaal HJ, Kirsten GF, Smith J, Grové D. Expectant management of early onset severe preeclampsia: perinatal outcome. Br J Obstet Gynecol 2000;107: 1258–64.