Maternal outcomes according to mode of delivery in women with severe preeclampsia: a cohort study.

http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, Early Online: 1–7 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.928689

ORIGINAL ARTICLE

Maternal outcomes according to mode of delivery in women with severe preeclampsia: a cohort study

J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 12/10/14 For personal use only.

Melania M. R. Amorim1,2, Leila Katz1, Amanda S. Barros3, Tainara S. F. Almeida3, Alex Sandro R. Souza1,4, and Anı´bal Fauńdes5 1

Department of Maternal and Child Health, Instituto de Medicina Integral Prof. Fernando Figueira (IMIP), Recife, Pernambuco, Brazil, Department of Obstetrics, Universidade Federal de Campina Grande (UFCG), Campina Grande, Paraı´ba, Brazil, 3Faculdade Pernambucana de Sau´de (FPS), Recife, Pernambuco, Brazil, 4Department of Maternal and Child Health, Universidade Federal de Pernambuco (UFPE), Recife, Pernambuco, Brazil, and 5Department of Obstetrics and Gynecology, Universidade de Campinas (UNICAMP), Campinas, Saõ Paulo, Brazil,

2

Abstract

Keywords

Objective: To determine the association between mode of delivery and maternal complications in patients with severe preeclampsia. Methods: A prospective cohort study was conducted with 500 pregnant women with severe preeclampsia. The mode of delivery, vaginal or caesarean section, was considered the exposure, while the postpartum maternal complications and severe maternal morbidity were the outcomes. Logistic regression analysis was performed to determine the adjusted risk and 95% confidence intervals (95% CI) of maternal morbidity. Results: Labour was spontaneous in 22.0% and induced in 28.2%, while 49.8% had an elective caesarean section. Ninety-five (67.4%) of the patients in whom labour was induced delivered vaginally. Total Caesarean rate was 68.2%. The risk of severe maternal morbidity was significantly greater in patients submitted to Caesarean section (54.0% versus 32.7%) irrespective of the presence of labour. Factors that remained associated with severe maternal morbidity following multivariate analysis were a diagnosis of HELLP syndrome after delivery (OR ¼ 3.73; 95% CI: 1.55–9.88) and having a caesarean (OR ¼ 1.91; 95% CI: 1.52–4.57). Conclusions: Caesareans are often performed in patients with severe preeclampsia and are associated with significant postpartum maternal morbidity. Induction of labour should be considered a feasible option in these patients.

Caesarean, complications, delivery, labour, obstetrics, pre-eclampsia, prognosis

Introduction The ideal mode of delivery in patients with severe preeclampsia still remains to be established [1]. Preeclampsia is one of the most common reasons for the therapeutic interruption of pregnancy, since delivery constitutes the only definitive treatment for the process [1]. As preeclampsia often occurs in preterm pregnancies [2], interruption is frequently indicated when labour is absent [3,4]. The global rate of caesarean sections is estimated to be high (around 70% or more in preterm pregnancies) because many obstetricians prefer performing Caesareans in these patients, even when foetal vitality is good [5,6]. Nevertheless, it has yet to be proven that Caesareans improve neonatal outcome and they may be associated with poorer maternal outcome [7]. Observational studies suggest that vaginal delivery may be better in the long term for Address for correspondence: Melania M. R. Amorim, MD, PhD, Department of Obstetrics, Universidade Federal de Campina Grande (UFCG), Rua Neuza Borborema de Souza, 300, 58406-120 Campina Grande, Paraı´ba, Brazil. Tel: +558333212695. Fax: +558333212695. E-mail: [email protected]

History Received 27 August 2013 Revised 15 May 2014 Accepted 23 May 2014 Published online 27 June 2014

premature babies, with lower complication rates and a shorter duration of hospitalization. Furthermore, the rate of cerebral palsy of the newborn remains the same [6,8]. The therapeutic decision regarding mode of delivery is not easy, since obstetricians often have to deal with pregnancies that are still remote from term and with unfavourable cervical conditions [6]. Additionally, both maternal clinical conditions and foetal well-being may be compromised [9]. Nonetheless, most medical society guidelines recommend vaginal delivery due to the benefits to the mother of this mode of delivery, the high likelihood of success and the lower rates of complications [10–14]. Women with preeclampsia have a low tolerance for blood loss and a greater risk of haemorrhagic complications; therefore, in these women vaginal delivery may confer significant benefits [15]. The present study was carried out to evaluate the association between mode of delivery and maternal outcomes among patients with severe preeclampsia.

Methods A prospective, cohort study was conducted in women with severe preeclampsia receiving care at Instituto de Medicina


2

M. M. R. Amorim et al.

Integral Prof. Fernando Figueira (IMIP), a tertiary referral center located in the city of Recife in northeastern Brazil, which is a major referral center for high risk obstetrics care in the region and the only one that possesses an obstetric ICU. The caesarean rate of the unit is 40%. The study was conducted between August 2008 and July 2009 after having been approved by the institution’s Institutional Review Board under reference # 1199 on 26 June 2008. All the patients included in the study voluntarily agreed to participate and gave their signed informed consent. STROBE recommendations for reporting cohort studies were followed. Data were extracted by the authors from medical charts and when there were missing data, charts were again checked and the patient consulted about disagreement in information. Sample size was calculated with open source software (Openepi version 2.3, Atlanta,), assuming a rate of haemorrhagic complications of 8% among patients submitted to Caesarean sections and 1% among patients who had vaginal deliveries (data from a previous pilot study carried out in our institution), with an alpha error of 5% and a power of 90%. Sample size was thus defined at 420 patients and this number was increased to 500 patients to cater for possible losses and exclusions. All pregnant women with confirmed diagnosis of severe preeclampsia admitted in our hospital were included in the study and followed after delivery until hospital discharge. Cases of haemorrhagic emergencies (placenta praevia and placental abruption), eclampsia, coma, inability to provide informed consent to participate in the study and patients with a diagnosis of other associated clinical conditions (diabetes, collagen diseases or thrombophilia) were excluded from the study. Associated clinical conditions were considered present when the patient had at admission previous diagnosis of the condition or when during hospital stay the diagnosis was made. The mode of delivery, vaginal or Caesarean section, was considered the independent variable, while the following postpartum maternal complications constituted the dependent variables: haemorrhagic complications, puerperal infection, blood transfusion, hypertensive crisis, need for antihypertensive drugs, oliguria, acute pulmonary oedema, thromboembolism, eclampsia, HELLP syndrome, severe maternal morbidity and duration of hospitalization longer than 7 d. The following control variables (potential confounding factors) were taken into consideration: maternal age, parity, body mass index (BMI), clinical form of preeclampsia (pure or superimposed), blood pressure levels, prepartum complications, severity of hypertension and HELLP syndrome diagnosed prior to delivery) and induction of labour. Criteria defined by the National High Blood Pressure Education Program Working Group (2000) were taken into consideration for the diagnosis of severe preeclampsia, i.e. presence of any one of the following signs or symptoms in women with hypertension diagnosed after 20 weeks of pregnancy and proteinuria: blood pressure 160 mmHg systolic or 110 mmHg diastolic; proteinuria 2.0 g in 24 h (2+ or 3+ on qualitative examination); increased serum creatinine level (41.2 mg/dL); platelet count 5100 000 cells/ mm3, evidence of microangiopathic haemolytic anemia (with increased lactic acid dehydrogenase concentration), or

J Matern Fetal Neonatal Med, Early Online: 1–7

both; elevated hepatic enzyme activities (either alanine aminotransferase, aspartate aminotransferase, or both); patient report of persistent headache or other cerebral or visual disturbances; patient report of persistent epigastric pain [16]. Preeclampsia was considered to be pure when hypertension and proteinuria developed after 20 weeks in previously normotensive patients, whereas preeclampsia was considered to be superimposed when it was diagnosed following the appearance of proteinuria (300 mg in 24 h) in patients with a diagnosis of hypertension prior to 20 weeks or in patients with hypertension and proteinuria prior to 20 weeks, in accordance with any of the following findings: sudden increase in proteinuria; sudden increase in blood pressure in a woman whose hypertension had previously been well controlled; thrombocytopenia (platelet count 5100 000 cells/ mm3) or increase in alanine aminotransferase or aspartate aminotransferase to abnormal levels [16]. All patients were given standard care provided at the hospital for cases of severe preeclampsia [1] based on the recommendations outlined in the guidelines of the pertinent international and national societies [10–13,17], including use of magnesium sulfate for prevention of eclampsia [10–12,16]. Investigators did not interfere in any way with the management adopted. Conservative management was indicated for women with less than 34 weeks of pregnancy and antenatal corticosteroid therapy with betamethasone (two 12 mg doses with a 24-h interval) was administered to accelerate foetal lung maturity. Conservatively managed patients received oral antihypertensive drugs, while foetal well-being was monitored by ultrasonography, Doppler velocimetry and cardiotocography. The criteria for interrupting pregnancy were: gestational age 434 weeks, spontaneous labour, impaired foetal wellbeing or maternal complications (HELLP syndrome, uncontrolled hypertension). Very high blood pressure occurring prior to delivery were treated with injectable hydralazine (5 mg given as an intravenous bolus, up to a maximum dose of 20 mg) and after delivery with captopril (25 mg orally) with the objective of reducing pressure by approximately 20%. The decision regarding mode of delivery was made by the attending team, although the institution’s regulation manual suggests inducing labour if there are no contraindications to vaginal delivery [1], in agreement with the recommendations of the international and national societies [12]. Labour induction was defined as the use of any method to artificially induce labour: either by pharmacological (misoprostol or oxytocin) or mechanical (Foley catheter) means. Induction was considered to have been successful when the outcome was vaginal delivery. Elective Caesarean section was defined as a Caesarean performed in the absence of labour, whereas an intrapartum Caesarean section was performed in the presence of uterine contractions, either spontaneous or induced. The indications for performing a Caesarean section are described in this study exactly as they were recorded on the patient’s chart by the physician on duty responsible for the procedure. They were later classified into groups for the purposes of analysis. If more than one indication had been recorded, the main reason for performing the Caesarean section was taken into consideration. All patients were followed during hospitalization until discharge and/or 30 d

Maternal outcomes in severe preeclampsia


DOI: 10.3109/14767058.2014.928689

postpartum in order to identify postpartum complications. Duration of hospitalization from delivery to discharge was determined. With respect to complications during pregnancy and postpartum, uncontrolled hypertension or hypertensive crisis was defined as the 24-h presence of at least one very high blood pressure urgently requiring administration of antihypertensive drugs (systolic pressure 180 mmHg or diastolic pressure 120 mmHg). HELLP syndrome was defined in accordance with the criteria established by Sibai [17]. Acute pulmonary oedema was diagnosed according to a clinical finding of severe dyspnea, production of pink frothy sputum from the airways, excessive sweating and cyanosis, as well as moist rales present in lung fields [18]. Oliguria was diagnosed when urinary output was less than 400 ml in 24 h, based on the criteria of acute kidney failure [19]. Thromboembolism was defined as the presence of any thromboembolic event confirmed by imaging tests and requiring anticoagulant therapy. The criteria defining postpartum severe maternal morbidity were modified from World Health Organization (WHO) in 2009 [20]. As severe preeclampsia per se is defined like ‘‘potentially life-threatening conditions’’ and we wanted to discriminate the most severe cases, we considered severe maternal morbidity the presence of other indicators such as severe hypertension, hypertensive encephalopathy, postpartum HELLP syndrome, endometritis, pulmonary oedema, respiratory failure, seizures, sepsis, shock, thrombocytopenia less than 100 000/ mm3, thyroid crisis and severe management indicators (blood transfusion, central venous access, hysterectomy, prolonged hospital stay, no anaesthetic intubation, return to operating room and surgical intervention [20]. Postpartum haemorrhage was defined as blood loss 1000 ml or haemorrhage that justified blood transfusion in the opinion of the attending physician. Post-surgical haematoma was defined as an accumulation of blood in the surgical wound requiring drainage. The patient was considered to have an infection when a clinical and laboratorial diagnosis of infection was made postpartum (infection at the surgical site, endometritis, urinary infection, lung infection or any other infection). Infection of the surgical site was defined in accordance with the criteria established by the Centers for Disease Control and Prevention (CDC) and was classified as an incisional infection (in the surgical wound itself) or related to organs and spaces (intracavitary or deep infection) [20]. A diagnosis of postpartum endometritis was based on the patient having a temperature 438 C in the absence of any other cause other than in the first 24 h postpartum or a temperature of 38.7 C in the first 24-h postpartum. Uterine sensitivity, purulent or foul-smelling lochia and leukocytosis with left shift were considered adjuvant criteria [21]. Statistical analysis was performed using the public domain Epi Info software program, version 3.5.3 (Centers for Disease Control and Prevention, Atlanta, GA). Measures of central trend and dispersion were calculated for numerical variables, and frequencies for categorical variables. The statistical tests used were Pearson’s chi-square test and Fisher’s exact test, as appropriate, considering only two-tailed values of p. A significance level of 5% was adopted throughout the statistical analysis. Risk ratio (RR) was calculated as a measure

3

of the relative risk of maternal complications according to the type of delivery and a 95% confidence interval was established. Multivariate analysis with multiple logistic regression was performed to determine the risk of severe maternal morbidity according to the mode of delivery, while controlling for potential confounding factors. A hierarchical model of analysis was adopted, as recommended by Victora [22], which included the following variables: maternal age, parity, BMI, clinical form of preeclampsia (pure or superimposed), gestational age at delivery, induced labour, spontaneous labour, type of Caesarean section, (elective or intrapartum), prepartum complications (HELLP syndrome and uncontrolled hypertension) and prepartum blood pressure levels. Adjusted odds ratios (OR) were calculated, together with their 95% confidence intervals (95% CI) for the variables that continued to be significantly associated with maternal morbidity.

Results During the study period, 831 cases of severe preeclampsia and eclampsia were admitted to this hospital. Of these, 614 patients were addressed for inclusion in the study, 525 of whom were considered eligible for the study and 89 ineligible. The reasons for exclusion were: haemorrhagic emergencies (n ¼ 22), eclampsia (n ¼ 52), associated clinical conditions (n ¼ 13), coma or inability to provide informed consent to participate in the study (n ¼ 2). Of the eligible patients, 500 agreed to participate and were evaluated. The mean age of the patients included in the study was 24.7 years, with 55% being under 25 years of age. Median parity was zero, with 65% of the participants being primiparas. Around 18% of the patients had had a previous caesarean section. Preeclampsia was considered pure in 77% of participants and superimposed in 23%. Mean gestational age, both at admission and at delivery, was around 36 weeks. The percentage of premature deliveries was 42.4%. Mean prepartum blood pressure levels were around 173 mmHg for systolic pressure and 113 mmHg for diastolic pressure. With respect to prepartum complications 28% of patients had uncontrolled hypertension. Mean duration of hospitalization (from delivery to discharge) was 6.6 ± 4.3 d. Patients were older, had higher BMI, gestational age was lower and systolic and diastolic blood pressure was higher in the caesarean group (Table 1). Labour was spontaneous in 110 patients (22%) and induced in 141 (28.2%), 95 of whom went on to have vaginal deliveries (67.4%). The total caesarean section rate was 68%; with 249 patients (50%) having elective and 92 (18.4%) intrapartum caesarean sections. The caesarean section rate following spontaneous labour was 42% (Table 2). The methods used to induce labour were vaginal misoprostol in 64 patients (45.4%), oxytocin alone in 71 patients (50.3%), a Foley catheter alone in 4 patients (2.8%) and a Foley catheter with oxytocin in 2 patients (1.4%). The main indications for performing a caesarean section were severe preeclampsia (57%), chronic foetal distress (15%), breech presentation (5.6%), dystocias and/or cephalopelvic disproportion (6%), non-reassuring foetal heart rate (4.4%), macrosomia (4%) and having had two or more previous caesarean sections (3.5%).

4



An increased risk of various postpartum complications was found in patients submitted to Caesarean section. There was a greater incidence of postpartum haemorrhage (4.1% versus 0.6%; p ¼ 0.024), post-surgical haematoma (8.2% versus 0.6%; p ¼ 0.00009) and any haemorrhagic complication (12.3% versus 1.3%; p ¼ 0.00005). The risk of any haemorrhagic complication was almost 10 times higher in patients submitted to caesarean section (RR ¼ 9.8; 95% CI: 2.4–40.0) and the need for blood transfusion was significantly higher following caesarean section (8.2% versus 2.5%; p ¼ 0.006) (Table 3).


Table 1. Patient’s characteristics.

Characteristics Maternal age (years) (Mean, SD) Parity (Median, IQR) Gestational age (weeks) (Mean, SD) BMI (kg/m2) (Mean, SD) SBP (mmHg) (Mean, SD) DBP (mmHg) (Mean, SD)

Caesarean (n ¼ 341)

Vaginal delivery (n ¼ 159)

p

25.1

6.5

23.7

6.6

0.03*

1.0 36.1

1–2 3.5

1.0 36.8

1–2 3.4

0.44y 0.04*

31.4 175.2 114.2

5.4 18.9 13.1

29.6 169.6 111.0

5.3 16.2 11.7

0.003* 0.001* 0.009*

SD, standard deviation; IQR, interquartile range; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure. *Student t-test. yMann–Whitney test.

Table 2. Characteristics of labour and delivery. Labour Spontaneous labour Induced labour Elective caesarean Vaginal delivery Intrapartum caesarean Caesarean following induction Caesarean following spontaneous labour Total number of Caesarean sections

N

%

110 141 249 159 92 46/141 46/110 341

22.0 28.2 49.8 31.8 18.4 32.6% 41.8% 68.2

Infection of surgical site (2.6%) occurred only in patients submitted to Caesarean section (p ¼ 0.002). The groups were not significantly different with respect to the risk of endometritis (p ¼ 0.55) or of any kind of puerperal infection (RR ¼ 1.59; 95% CI: 0.60–4.22; p ¼ 0.35) (Table 3). Thromboembolic complications were rare, with one case occurring in each group (p ¼ 0.54). Oliguria was significantly more common in women submitted to caesarean section (9.0% versus 4.0%; p ¼ 0.03). Very high blood pressure and the need for antihypertensive drugs were more common following caesarean section (47.8% versus 30.2%; p ¼ 0.0002 and 80.9% versus 62.3%; p ¼ 0.000006, respectively). There was no significant difference in the rate of acute pulmonary oedema (p ¼ 0.46) or postpartum HELLP syndrome (p ¼ 0.12) as a function of the type of delivery. Duration of hospitalization was longer after caesarean delivery, with a median of 7 d compared to 5 d for those who had a vaginal delivery. Hospitalization beyond 7 d was more common following caesarean section than after vaginal delivery (53% versus 23%; p50.0001) (Table 3). The risk of severe maternal morbidity was 1.7 times higher in patients submitted to caesarean section (54.0% versus 32.7%; RR ¼ 1.65; 95% CI: 1.29–2.11) (Table 3). There was no difference in the rate of severe maternal morbidity between elective and intrapartum caesarean sections (55.0% versus 51.1%, respectively; p ¼ 0.52). When multiple logistic regression analysis was performed, the variables that remained significantly associated with postpartum severe maternal morbidity were caesarean section (OR ¼ 1.91; 95% CI: 1.52–4.57) and prepartum presence of HELLP syndrome (OR ¼ 3.91; 95% CI: 1.55–9.88) (Table 4).

Discussion In the present study, a high caesarean section rate of almost 70% was found in patients with severe preeclampsia. Caesarean sections were associated with increased maternal morbidity, raising the risk of haemorrhagic and infectious complications and the rate of postpartum hypertensive crises, and prolonged hospitalization. The risk of severe maternal

Table 3. Postpartum maternal complications according to mode of delivery. Caesarean (n ¼ 341) Complication Postpartum haemorrhage Post-surgical haematoma Any haemorrhagic complication Blood transfusion Surgical site infection Endometritis Any puerperal infection Thromboembolic disease Oliguria Hypertensive crisis Need for antihypertensive drugs Acute pulmonary oedema HELLP syndrome 47 d of hospitalizationy Severe maternal morbidity

Vaginal delivery (n ¼ 159)

N

%

N

%

RR

95% CI

p

14 28 42 28 9 12 17 1 31 163 276 2 9 180 184

4.1 8.2 12.3 8.2 2.6 3.5 5.0 0.3 9.1 47.8 80.9 0.6 2.6 52.8 54.0

1 1 2 4 – 4 5 1 6 48 99 – 1 36 52

0.6 0.6 1.3 2.5 – 2.5 3.1 0.6 3.8 30.2 62.3 – 0.6 22.6 32.7

6.5 13.1 9.79 3.26 – 1.40 1.59 0.47 2.41 1.58 1.30 – 4.20 2.33 1.65

0.86–49.2 1.79–95.1 2.40–39.95 1.16–9.15 – 0.46–4.27 0.60–4.22 0.03–7.4 1.03–5.66 1.22–2.06 1.14–1.48 – 0.53–32.8 1.72–3.16 1.29–2.11

0.024 50.0001 50.0001 0.006 0.002 0.55 0.35 0.54 0.03 0.0002 50.0001 0.46 0.12 50.0001 50.0001

RR, relative risk; CI, confidence interval; NNH, number necessary to harm. *NE, Not estimated. yDuration of hospitalization (median): Caesarean ¼ 8 d; Vaginal delivery ¼ 5 d.

NNH (CI 95%) 29 13 9 18

19 6 5 3 5

(100–20) (25–10) (14–7) (100–11) NE* – – – (100–11) (13–4) (10–4) – – (5–3) (8–3)


DOI: 10.3109/14767058.2014.928689

Table 4. Risk factors for postpartum severe maternal morbidity (multivariate analysis).

Variable Caesarean Prepartum HELLP syndrome

Coefficient

Standard error

p

OR

95% CI

1.06 1.36

0.33 0.47

0.0019 0.0040

1.91 3.91

1.52–4.57 1.55–9.88


Variables entering the model: maternal age, parity, body mass index, clinical form of preeclampsia (pure or superimposed), gestational age at delivery, induced labour, spontaneous labour, type of caesarean section (elective or intrapartum), uncontrolled hypertension, HELLP syndrome antepartum blood pressure levels. OR, odds ratio; CI, confidence interval.

morbidity persisted even after controlling for potential confounding factors and was 65% higher in patients submitted to caesarean sections. Patients were older and had higher BMI in the caesarean group, although the difference found is not clinically significant. Patients did not differ regarding parity, but gestational age was lower in the c-section group. This finding can be explained because of the time from induction to delivery in the patients that achieve vaginal delivery. Systolic and diastolic blood pressure was higher in the caesarean group, and despite the difference not being clinically important, may be explained by the tendency that care takes usually demonstrate to be less tolerant and waiting for vaginal delivery in patients with uncontrolled blood pressure. One can argue that these groups are very differents once women who underwent caesarean section were older, at lower gestational week, with higher BMI and pre delivery also systolic and diastolic blood pressure levels. All these differences together could explain at least in part why clinicians decided to perform a caesarean section, although none of these findings are per se true indications for caesarean section. However, our research question does not adress the risk factors for caesarean section in these patients, but the consequences of performing a caesarean section in women with severe preeclampsia. We intend to evaluate the factors associated with caesarean section in these women in a later study. This high caesarean section rate is compatible with rates described by other authors [23]; however, it is higher than rates reported in more recent studies [24]. Differences in the study population and in the mode of management of these patients at each institution may explain these findings. Nevertheless, caesarean rates were high even when patients went into labour, either spontaneous (41.8%) or induced (32.6%). The high proportion of cases in which indication for caesarean section was only severe preeclampsia suggests that in many cases there was no real obstetric indication. On the other hand, the 67% success rate of vaginal delivery after labour induction in patients with severe preeclampsia achieved in the present study was higher than that described in the literature, which ranges from 48% to 60% in this specific clinical situation [6,7,9]. Few studies have specifically analysed the association between postpartum maternal complications and mode of delivery in patients with preeclampsia. Caesarean section constitutes a recognized risk factor for haemorrhagic and

5

infectious complications in the general population [25,26], although more recent studies have suggested a reduction in these risks when caesareans are scheduled in low-risk patients [27]. In the present study, in addition to the risk of haemorrhagic and infectious complications, an increased risk of oliguria, very high blood pressure and a need for antihypertensive drugs were found in women with preeclampsia who were submitted to a caesarean section, resulting in a significant increase in the duration of hospitalization. The present results are unsurprising, since preeclampsia increases the risk of haemorrhagic complications [28] and caesarean sections may aggravate this risk further, particularly in patients with thrombocytopenia [25,26]. Our results support the notion already reported that the best mode of delivery in cases of preeclampsia and is by the vaginal route, precisely because of the increased risk of haemorrhagic complications associated to caesarean delivery [10–12]. The risk of infection is increased by both caesarean sections and haemorrhage [29]. The more frequent finding of oliguria may reflect the increased amount of blood loss in caesarean sections [30] and any form of manipulation used to treat these complications, in addition to the increased infusion of liquids during the caesarean section, may explain the increase in very high blood pressure episodes and in the need for antihypertensive drugs following delivery. Studies suggest that prognosis is better with elective rather than intrapartum caesarean sections in the general population [26]. Although the proportion of complications was lower in the case of elective C-section as compared with intrapartum C-section, we did not find statistically significant association in the present analysis of women with preeclampsia. The sample size, however, was not calculated aiming to identify such difference. This finding strengthens the recommendation to attempt induction of labour in cases of preeclampsia since it is often successful and may reduce the risks of complications if delivery is vaginal. However, if an intrapartum caesarean section is necessary, the risk appears not to be higher than that found with an elective caesarean section. Relatively few studies have been published regarding the mode of delivery in women with severe preeclampsia and they generally fail to emphasize postpartum morbidity and to provide a clear definition of the analysed maternal outcomes [6,7,9]. To the best of our knowledge, the present study is the first to compare specific parameters of postpartum morbidity according to the mode of delivery in patients with severe preeclampsia. Since this is a prospective study, it has more strength than retrospective studies, permitting the adoption of rigid diagnostic criteria both for severe preeclampsia and for the various morbidities analysed. Furthermore, an expressive number of cases were included, with sufficient power to confirm the outcomes evaluated, thus representing a significant contribution to obstetric practice and reinforcing the recommendations of the medical societies and the consensus on vaginal delivery in cases of severe preeclampsia. Nevertheless, since this is an observational, cohort study, the level of evidence presented here is not as strong as that of a randomized clinical trial. As the selection of patients for vaginal delivery or caesarean section was not randomized but


6


was the responsibility of the team of attending physicians, various confounding factors may be involved. For example, the decision to perform a caesarean section may be associated with the presence of prepartum maternal complications and these complications in themselves may be associated with a greater risk of postpartum maternal morbidity. Such possible bias was controlled using a multivariate analysis and caesarean section remained as an independent risk factor for postpartum complications after controlling for potentially confounding factors. The only other variable that persisted in the model was a prepartum diagnosis of HELLP syndrome, which is a recognized risk factor for maternal complications in cases of preeclampsia; however, the effect of the caesarean section in itself continued to be significant, resulting in an almost two-fold adjusted risk of complications. The next step would be to perform controlled clinical trials, which will provide better quality evidence and clarify any doubts that may persist following observational studies. Although labour induction may represent a challenge in patients with severe preeclampsia, particularly in cases that are remote from term and in whom cervical conditions are unfavourable, we believe that it would be possible to perform a randomized clinical trial to compare labour induction with elective caesarean section in these cases. Until this clinical trial is carried out and based on the present findings together with other observational studies [6,7,9], we would suggest that, in addition to permitting labour to progress when spontaneous contractions are present, labour should be induced in cases of severe preeclampsia with an indication for interrupting the pregnancy, as long as there are no contraindications to vaginal delivery. Indicating caesarean section only on the basis of a diagnosis of preeclampsia should be avoided in patients with no associated maternal or foetal risks, thus preventing the increased risk of complications associated with caesarean sections among women in whom induction of labour could be considered.

Acknowledgements The authors thank National Council for Scientific and Technological Development (CNPq) for the financial support.

Disclosure of interest National Counsel of Technological and Scientific Development (CNPq), Brazil. Melania Amorim conceived the study, developed the initial protocol, analysed the data, jointly drafted the article and is guarantor. Leila Katz reviewed and amended the study protocol, contributed to enrolment of patients, interpreting data and writing the final version of the article. Amanda de Souza Barros and Tainara Sa´ Freire de Almeida participated in the development of the initial protocol, enrolled patients and collected data. Alex Sandro Rolland de Souza reviewed the study protocol and contributed to statistical analysis and writing the final version of the article. Anibal Fauńdes reviewed the study protocol and the final version of the article. The study was approved by the institution’s Institutional Review Board under reference # 1199 on 26 June 2008.


References 1. Noronha Neto C, Souza ASR, Amorim MMR. Tratamento da preéclaˆmpsia baseado em evideˆncias [Evidence-based treatment of preeclampsia]. Rev Bras Ginecol Obstet 2010;32:459–68. 2. Sibai BM. Chronic hypertension in pregnancy. Obstet Gynecol 2002;100:369–77. 3. Mozurkewich E, Chilimigras J, Koepke E, et al. Indications for induction of labour: a best-evidence review. BJOG 2009;116: 626–36. 4. Souza AS, Amorim MM, Feitosa FE. Comparison of sublingual versus vaginal misoprostol for the induction of labour: a systematic review. BJOG 2008;115:1340–9. 5. Catanzarite V, Quirk JG, Aisenbrey G. How do perinatologists manage preeclampsia? Am J Perinatol 1991;8:7–10. 6. Alanis MC, Robinson CJ, Hulsey TC, et al. Early-onset severe preeclampsia: induction of labour vs elective Caesarean delivery and neonatal outcomes. Am J Obstet Gynecol 2008;199:262.e1–6. 7. Coppage KH, Polzin WJ. Severe preeclampsia and delivery outcomes: is immediate Caesarean delivery beneficial? Am J Obstet Gynecol 2002;186:921–3. 8. Drife J. Mode of delivery in the early preterm infant (528 weeks). BJOG 2006;113:S81–5. 9. Nassar AH, Adra AM, Chakhtoura N, et al. Severe preeclampsia remote from term: labour induction or elective Caesarean delivery? Am J Obstet Gynecol 1998;179:1210–13. 10. National Collabourating Centre for Women’s and Children’s Health. Caesarean section. London (UK): National Institute for Clinical Excellence (NICE); 2004:142. 11. ACOG Committee on Practice Bulletins-Obstetrics. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. Number 33. Obstet Gynecol 2002;99:159–67. 12. Magee LA, Helewa M, Moutquin JM, von Dadelszen P, for the Hypertension Guideline Committee. Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy. SOGC Clinical Practice Guideline, No. 206. J Obstet Gynaecol Can 2008; 30:S1–48. 13. Blackwell SC, Redman ME, Tomlinson M, et al. Labour induction for the preterm severe pre-eclamptic patient: is it worth the effort? J Matern Fetal Med 2001;10:305–11. 14. Xenakis EM, Piper JM, Field N, Conway D, Langer O. Preeclampsia: is induction of labour more successful? Obstet Gynecol 1997;89:600–3. 15. Kuklina EV, Ayala C, Callaghan WM. Hypertensive disorders and severe obstetric morbidity in the United States. Obstet Gynecol 2009;113:1299–306. 16. Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol 2000;183:S1–22. 17. Sibai BM. The HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets): much ado about nothing? Am J Obstet Gynecol 1990;162:311–16. 18. Gandhi SK, Powers JC, Nomeir AM, et al. The pathogenesis of acute pulmonary edema associated with hypertension. N Engl J Med 2001;344:17–22. 19. Lianõ F, Gallego A, Pascual J, et al. Prognosis of acute tubular necrosis: an extended prospectively contrasted study. Nephron 1993;63:21–31. 20. Say L, Souza JP, Pattinson RC. WHO working group on Maternal Mortality and Morbidity classifications: maternal near miss – towards a standard tool for monitoring quality of maternal health care. Best Pract Res Clin Obstet Gynaecol 2009;23:287–96. 21. Mangram AJ, Horan TC, Pearson ML, et al. Guideline for Prevention of Surgical Site Infection, 1999. Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee. Am J Infect Control 1999;27:97–132. 22. Victora CG, Huttly SR, Fuchs SC, Olinto MT. The role of conceptual frameworks in epidemiological analysis: a hierarchical approach. Int J Epidemiol 1997;26:224–7. 23. Dissanayake VH, Samarasinghe HD, Morgan L, et al. Morbidity and mortality associated with pre-eclampsia at two tertiary care hospitals in Sri Lanka. J Obstet Gynaecol Res 2007;33:56–62. 24. Gowri V, Al-Zakwani I. Prevalence of Caesarean delivery in preeclamptic patients with elevated uric acid. Hypertens Pregnancy 2010;29:231–5.

DOI: 10.3109/14767058.2014.928689


25. Villar J, Carroli G, Zavaleta N, et al. World Health Organization 2005 Global Survey on Maternal and Perinatal Health Research Group. Maternal and neonatal individual risks and benefits associated with caesarean delivery: multicentre prospective study. Br Med J 2007;335:1025. 26. Pallasmaa N, Ekblad U, Aitokallio-Tallberg A, et al. Caesarean delivery in Finland: maternal complications and obstetric risk factors. Acta Obstet Gynecol Scand 2010;89:896–902. 27. Geller EJ, Wu JM, Jannelli ML, et al. Neonatal outcomes associated with planned vaginal versus planned primary Caesarean delivery. J Perinatol 2010;30:258–64.


7

28. Eskild A, Vatten LJ. Abnormal bleeding associated with preeclampsia: a population study of 315,085 pregnancies. Acta Obstet Gynecol Scand 2009;88:154–8. 29. Machado Junior LC, Sevrin CE, Oliveira E, et al. [Association between mode of delivery and maternal complications in a public hospital in Greater Metropolitan Saõ Paulo, Brazil]. Cad Saude Publica 2009;25: 124–32. 30. Ng SY, Ithnin F, Sia AT, Ng CC. Ergometrine administration for post-partum haemorrhage in an undiagnosed pre-eclamptic. Anaesth Intensive Care 2008;36:113–15.

Maternal postpartum complications according to delivery mode in twin pregnancies.

Induction, labor length and mode of delivery: the impact on preeclampsia-related adverse maternal outcomes.

Maternal complications associated with severe preeclampsia.

Uric Acid as a predictor of adverse maternal and perinatal outcomes in women hospitalized with preeclampsia.

The impact of maternal obesity, age, pre-eclampsia and insulin dependent diabetes on severe maternal morbidity by mode of delivery-a register-based cohort study.

Cross sectional study of mode of delivery and maternal and perinatal outcomes in mainland China.

Mode of Delivery according to Leisure Time Physical Activity before and during Pregnancy: A Multicenter Cohort Study of Low-Risk Women.

Changes in Maternal Plasma Adiponectin from Late Pregnancy to the Postpartum Period According to the Mode of Delivery: Results from a Prospective Cohort in Rio de Janeiro, Brazil.

Adverse perinatal outcomes related to the delivery mode in women with monochorionic diamniotic twin pregnancies.

Maternal Factors and Adverse Perinatal Outcomes in Women with Preeclampsia in Maceió, Alagoas.

eclampsia.

How do maternal views of delivery outcomes vary by demographics and preferred mode of delivery?

Effect of mode of delivery on perceived risks of maternal health outcomes among expectant parents: a cohort study in Beijing, China.

Maternal serum anti-Müllerian hormone in Sudanese women with preeclampsia.

[Maternal and perinatal outcomes in nulliparous gestations with late onset preeclampsia: Comparative study with gestations without preeclampsia].

The influence of hours worked prior to delivery on maternal and neonatal outcomes: a retrospective cohort study.

Maternal, placental and fetal exposure to bisphenol A in women with and without preeclampsia.

Pregestational excess weight, maternal obstetric complications and mode of delivery in the Rhea cohort in Crete.

Comparison of perinatal and maternal outcomes of severe preeclampsia, eclampsia, and HELLP syndrome.

Cohort analysis of antenatal care and delivery outcomes in pregnancy: a basis for improving maternal health.

Maternal and Neonatal Outcomes in Indicated Preterm Births Based on the Intended Mode of Delivery.

Severe Adverse Maternal Outcomes among Women in Midwife-Led versus Obstetrician-Led Care at the Onset of Labour in the Netherlands: A Nationwide Cohort Study.

Severe Preeclampsia versus HELLP Syndrome: Maternal and Perinatal Outcomes at <34 and ≥34 Weeks' Gestation.

Snoring during pregnancy and delivery outcomes: a cohort study.