The Burden of Eclampsia: Results from a Multicenter

Study on Surveillance of Severe Maternal Morbidity in

Brazil
Juliana C. Giordano1, Mary A. Parpinelli1*, Jose G. Cecatti1,2, Samira M. Haddad1, Maria L. Costa1,
Fernanda G. Surita1, Joao L. Pinto e Silva1, Maria H. Sousa2
1 Department of Obstetrics and Gynecology, School of Medical Sciences, University of Campinas - UNICAMP, Campinas, São Paulo, Brazil, 2 Center for Studies on
Reproductive Health of Campinas (CEMICAMP), Campinas, São Paulo, Brazil

Abstract
Objective: Maternal mortality (MM) is a core indicator of disparities in women’s rights. The study of Near Miss cases is
strategic to identifying the breakdowns in obstetrical care. In absolute numbers, both MM and occurrence of eclampsia are
rare events. We aim to assess the obstetric care indicators and main predictors for severe maternal outcome from eclampsia
(SMO: maternal death plus maternal near miss).

Methods: Secondary analysis of a multicenter, cross-sectional study, including 27 centers from all geographic regions of
Brazil, from 2009 to 2010. 426 cases of eclampsia were identified and classified according to the outcomes: SMO and non-
SMO. We classified facilities as coming from low- and high-income regions and calculated the WHO’s obstetric health
indicators. SPSS and Stata softwares were used to calculate the prevalence ratios (PR) and respective 95% confidence
interval (CI) to assess maternal characteristics, clinical and obstetrical history, and access to health services as predictors for
SMO, subsequently correlating them with the corresponding perinatal outcomes, also applying multiple regression analysis
(adjusted for cluster effect).

Results: Prevalence of and mortality indexes for eclampsia in higher and lower income regions were 0.2%/0.8% and 8.1%/
22%, respectively. Difficulties in access to health care showed that ICU admission (adjPR 3.61; 95% CI 1.77–7.35) and
inadequate monitoring (adjPR 2.31; 95% CI 1.48–3.59) were associated with SMO.

Conclusions: Morbidity and mortality associated with eclampsia were high in Brazil, especially in lower income regions.
Promoting quality maternal health care and improving the availability of obstetric emergency care are essential actions to
relieve the burden of eclampsia.

Citation: Giordano JC, Parpinelli MA, Cecatti JG, Haddad SM, Costa ML, et al. (2014) The Burden of Eclampsia: Results from a Multicenter Study on Surveillance of
Severe Maternal Morbidity in Brazil. PLoS ONE 9(5): e97401. doi:10.1371/journal.pone.0097401
Editor: C. Mary Schooling, CUNY, United States of America
Received December 30, 2013; Accepted April 19, 2014; Published May 13, 2014
Copyright: ß 2014 Giordano et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study was financially supported by CNPq/DECIT (The National Research Council and the Department of Science and Technology of the Brazilian
Ministry of Health), grant number 402702/2008-5. The publication of this article was sponsored by Fapesp, Grant 2014/07890-4. The funders had no role in study
design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]

Introduction The case fatality rate (number of deaths/number of cases) of

eclampsia ranges from 0–1.8% in high-income countries up to
Eclampsia is a rare, however potentially life-threatening 17.7% in India, emphasizing a huge gap in the quality of maternal
complication of the hypertensive disorders (HD) of pregnancy, health care according to social and economic patterns [8]. Over a
accountable for large numbers in morbidity and deaths among one-year period, the Swedish Medical Birth Register identified no
women of reproductive age and their offspring [1–4]. The estimate maternal death due to eclampsia, whilst in India, in the same
of incidence and the burden of eclampsia is still a challenging period, only one hospital reported 11 eclampsia-related deaths [8–
pursuit worldwide; currently only seven countries have national 10].
data on the topic [5]. A systematic review on preeclampsia (PE) Reducing maternal mortality (MM) by three quarters is one of
and eclampsia, performed in 2013, indicated that the crude the United Nations’ Millennium Development Goals [11]. Nearly
incidence of eclampsia fluctuates from 0 to 0.1% in Europe and up the totality of women who die from pregnancy-related causes
to 4% in Nigeria; Brazilian studies showed a 0.6% incidence [5,6]. comes from LMIC [2,3]. According to the Brazilian Ministry of
Nonetheless, 94.6% of the data were collected in the USA, Health, there has been a substantial reduction of maternal deaths
highlighting a marked regionalization bias and, therefore, the need (MD) in the country from 1990 to 2010, i.e., a decrease from 141
for more studies, especially in low- and middle-income countries to 62 deaths for 100,000 live births (LB) [12]. Nevertheless, in
(LMIC) [5,7].

PLOS ONE | www.plosone.org 1 May 2014 | Volume 9 | Issue 5 | e97401

 Eclampsia as Severe Maternal Morbidity

order to achieve the MDG5 by 2015, Brazil would have to halve Main Outcomes
this number, what seems to be a very difficult mission to pursue. Maternal outcomes for eclampsia during pregnancy, childbirth
Recently, the World Health Organization (WHO) defined the or puerperium were considered in two different groups:
presence of organ dysfunction or failure during pregnancy, Non-Severe Maternal Outcome (non-SMO). All cases of
childbirth or postpartum as maternal near miss (NM). A woman eclampsia in the absence of organ failure/dysfunction were
who fulfills one of the clinical, laboratory or management criteria classified as non-SMO; this is the comparison group.
established by WHO is a NM case. From a theoretical perspective, SMO (Severe maternal outcome). All cases of maternal
the NM cases should be as similar to maternal deaths as possible death or maternal near miss.
[13–17].
Childbirth care in LMIC is usually associated with difficult N Maternal Near Miss (NM): cases that fulfilled at least one of the
access to adequate maternity services [2,3,7]. In Brazil, although clinical, laboratory or management criteria representing life-
98% of pregnant women do deliver their babies in hospitals, a threatening conditions (i.e., organ failure/dysfunction) and
large number of these facilities are not well equipped to deal with who survived this condition. Figure 1.
pregnancy-related complications. The shortage of intensive care N Maternal Death (MD): death during pregnancy or within 42 days
units (ICU) to where such women can be transferred is still a post-partum, regardless length or site of pregnancy, from any
worrying reality in several settings [12]. In addition the proportion cause related to or aggravated by the pregnancy or its
of facilities with adequately trained staff to deal with complications management, yet not from accidental or incidental causes.
is not known at all.
MM is amongst the worst-performing health indicators in
resource-poor settings. In absolute numbers, both maternal Covariates
mortality and the occurrence of eclampsia are rare events Information on age, skin color, marital status, schooling and
[2,3,16,17]. The only Brazilian national data on eclampsia is the parity were analyzed as possible predictors of SMO from
total number of deaths, 167 cases in 2010, with a maternal eclampsia, as they are already been identified in some studies as
mortality ratio (MMR) of 5.83 [18]. It is with the intent of filling predictors for eclampsia [4,22,23].
this epidemiological gap that our study aims to assess the obstetric Previous disease was defined as any pathological condition
care indicators and main predictors for severe maternal outcomes diagnosed before or during pregnancy, but not related to it.
from eclampsia (SMO: maternal death plus maternal near miss). Chronic hypertension was defined as the presence of high blood
pressure (BP) $140690 mmHg diagnosed before the 20th week of
pregnancy, after two measurements within a minimum interval of
Methods
4 hours, regardless of the use of medications [1].
Our study is a secondary analysis of The Brazilian Network for We selected the most frequent associated complications during
Surveillance of Severe Maternal Morbidity Study. The purpose of admission period: hemorrhage, HELLP syndrome, severe hyper-
this network was to identify cases of severe maternal morbidity/ tension, pulmonary edema and severe sepsis. Coagulation
near miss, using the criteria recently established by WHO to disorders, shock, jaundice concomitantly with preeclampsia and
characterize these conditions [16]. According to this definition, a cerebrovascular accident are part of the NM definition criteria,
maternal near miss case is a woman who experienced a very and were therefore excluded from the analysis because of their
serious complication during pregnancy and as a consequence behavior as interacting variables.
almost died, surviving at least until the 42nd day after childbirth. The use of magnesium sulphate (MgSO4) for the prevention and
The methods of the Brazilian Network have already been management of eclampsia was assessed as a dichotomous variable
described in details elsewhere [19,20]. (use and non-use) because the data collection form had no
Briefly, it was a cross-sectional multicenter study conducted information on the exact period of time when clinical events
from July 2009 to June 2010, involving 27 hospitals from all occurred or procedures were performed, Therefore the opportu-
different regions of Brazil, excluding the Federal District. From nity of its use could not be detailed assessed.
those 27 centers, 95% of cases were insured by SUS, the Brazilian Post-partum admission was regarded as a worse outcome, based
publicly funded health system. Brazil is geographically divided into upon the assumption that the eclamptic women who were
5 different regions and one Federal District: North (N), Northeast admitted after giving birth had to be transferred to a health
(NE), Midwest (MW), South (S) and Southeast (SE). We assembled center capable of delivering a better care. Bearing in mind that not
these regions into 2 major groups, according to their 2000, human all the Brazilian health facilities caring for pregnant women are
development index (HDI) [21]. According to this definition, S and equipped with an ICU or have an ICU bed promptly available
SE were high HDI regions and N, NE and MD were low HDI (most ICUs operate at their full capacity at any given time), we had
regions. We then calculated the indicators proposed by WHO to to assume that only the most complicated cases of eclampsia were
monitor the quality of obstetric care using maternal near miss and admitted to ICU. In addition to this inference, ICU availability
maternal death cases with eclampsia [16]. was also assessed by the variable ‘‘inadequate monitoring’’,
During this period, out of the 9,555 women who were translated into availability or not of ICU care.
diagnosed with severe maternal complications, 6,706 presented Our study had local coordinators who were trained to gather
with severe hypertensive disorders and 426 were admitted with, or accurate information from both health care providers involved
developed, eclampsia during hospitalization. Eclampsia was with the care at its initial phases and from medical records, aiming
to address as many aspects of care as possible. We classified and
identified by the occurrence of tonic-clonic seizures - including
defined the variables ‘‘lack of drug’’, ‘‘inadequate monitoring’’,
seizures and coma - that occurred during pregnancy, delivery or
‘‘delay for transfer’’, ‘‘lack of staff’’, ‘‘delay for diagnosis’’, ‘‘not
puerperium and that were not related to preexisting organic brain
opportune treatment’’ and ‘‘inadequate management’’ to evaluate
disorders [1].
the access to and quality of appropriate obstetric emergency care.
This was performed by both the local investigator and coordinator
and then checked by the study team of the coordinating center.

PLOS ONE | www.plosone.org 2 May 2014 | Volume 9 | Issue 5 | e97401

 Eclampsia as Severe Maternal Morbidity

Figure 1. WHO criteria for maternal near miss.

doi:10.1371/journal.pone.0097401.g001

We finally analyzed the perinatal outcomes and mode of Quality Control

delivery in SMO cases. The variables were defined as follows: cut- The network database was fed with information extracted from
off point for gestational age at delivery, determined by clinical the medical records, transcribed manually onto the data collection
criteria (less than 33 weeks and $34 weeks); mode of delivery, form by local investigators and later on transferred to the
listed as cesarean section or vaginal birth; perinatal outcome electronic forms. The technical procedures for case selection and
(stillbirth or live birth); birth weight (,2500 or $2500 grams); accurate form filling were detailed explained in the respective
neonatal outcome (defined as neonatal ICU admission, neonatal manual of operation. Local study coordinators performed
death, i.e., death until 28 days of life, or hospital discharge); fifth systematic quality control of data, so that possible incongruences
minute Apgar score (,7 and from 7 to 10 indicates, respectively could be identified. One of the investigators from the coordinating
low and high vitality score at birth); and perinatal death (stillbirth center visited the institutions taking part of this study, aiming to
plus neonatal death ,7 days). verify the consistency of the information retrieved from both
manually- and electronically-filled collection forms in light of the
Statistical Analysis case reports of the study subjects, randomly selecting such cases.
Bivariate analysis was performed to identify factors (predictors) The final quality control was performed by the application of
associated with SMO (maternal NM or MD) by estimating logical consistency and review of database.
prevalence ratios (PR) and their respective 95% confidence
intervals (CI), adjusted for cluster effect (maternal hospital or Ethical Statement
centers) [24]. Access to health care facilities, maternal character- This study is a secondary analysis; all records were obtained
istics, complications and procedures related to and/or used for through the database of the main study, the Brazilian Network for
management of eclampsia, other than those already used for NM Surveillance of Severe Maternal Morbidity. According to the rules
case definition according to the WHO criteria, were described of the sponsor agency the database is not of public domain and the
comparatively among women from both groups, with differences principal investigators are the owners of the data, being
assessed by a Chi-square test. Additionally the hazard of perinatal responsible for its use for scientific purposes. We followed all the
outcomes including the mode of delivery was also estimated for principles that regulate research on human beings defined by the
women who progressed towards an SMO, with adjusted PR and Brazilian National Health Council, as well as the Declaration of
their respective 95% confidence interval (CI). Finally, Poisson Helsinki. There was no need for an Informed Consent Form, since
multiple regression analysis was performed in 321 cases in which data were collected from medical records post–discharge or post-
all variables were available and adjusted by cluster and all other mortem and no contact occurred with the subjects. Local IRBs
predictors. The primary sampling units of our study were the (listed below) and the National Committee of Ethics in Research
health care facilities and therefore it was necessary to adjust the (CONEP, Brazilian Ministry of Health) approved the study, under
analysis by the cluster effect [25]. The softwares used for the the letter of approval 097/2009. The National Council for Ethics
analysis were SPSS version 17 (SPSS, Chicago, IL, USA) and in Research and the Institutional Review Boards of each site
Stata version 7.0 (StataCorp, College Station, TX, USA). granted a waiver of individual informed consent.

PLOS ONE | www.plosone.org 3 May 2014 | Volume 9 | Issue 5 | e97401

 Eclampsia as Severe Maternal Morbidity

The Review Boards of the following institutions reviewed and presented pregnancy-related severe complications and met the
approved this study: Maternidade Cidade Nova Dona Nazarina study inclusion criteria. Out of this population, 910 women
Daou (Manaus, AM), Maternidade Climério de Oliveira (Salva- progressed to SMO (770 maternal NM and 140 maternal deaths);
dor, BA), Hospital Geral de Fortaleza (Fortaleza, CE), Hospital 20% of cases of eclampsia, 4% of cases of other severe
Geral Dr. César Cals (Fortaleza, CE), Maternidade Escola Assis hypertensive disorders (excluding eclampsia) and 17% of other
Chateaubriand (Fortaleza, CE), Hospital Materno Infantil de morbidities (infectious and hemorrhagic disorders) developed
Goiania (Goiania, GO), Hospital Universitário da Universidade SMO. Respectively, almost 4% of cases of eclampsia and other
Federal do Maranhao (Sao Luis, MA), Maternidade Odete morbidities and 0.4% of cases of other severe hypertensive
Valadares (Belo Horizonte, MG), Instituto de Saúde Elıdio de disorders died. (Figure 2).
Almeida (Campina Grande, PB), Hospital Universitário Lauro The total prevalence of eclampsia was 5.2 (per 1000 LB) and a
Wanderley da Universidade Federal da Paraiba (Joao Pessoa, PB), specific MMR of 19.5 (per 100,000 LB). The total mortality index
Centro Integrado de Saúde Amaury de Medeiros (Recife, PE), was 18.6%, 2.7 times higher in the Brazilian regions of lower HDI:
Instituto de Medicina Integral Prof. Fernando Figueira (Recife, 22.6% gathering the Midwest, Northeast and North regions, and
PE), Hospital das Clınicas da Universidade Federal de Pernam- 8.3% for South and Southeast regions (Table 1).
buco (Recife, PE), Hospital das Clınicas da Universidade Federal Approximately 70% of all eclamptic women were primiparous.
do Paraná (Curitiba, PR), Hospital Maternidade Fernando The median age of the cases was 20 years, with the youngest being
Magalhaes (Rio de Janeiro, RJ), Instituto Fernandes Figueira 13 and the eldest 44 (data not showed in tables). Amongst the
(Rio de Janeiro, RJ), Hospital das Clinicas da Universidade maternal characteristics, obstetric background and medical
Federal do Rio Grande do Sul (Porto Alegre, RS), Faculdade de history, the only factors associated with the risk of SMO from
Medicina de Botucatu da Universidade Estadual Paulista eclampsia were any previous disease and chronic hypertension.
(Botucatu, SP), Hospital da Mulher da Universidade Estadual de Medical histories of any previous disease were present in 27% of
Campinas (Campinas, SP), Hospital e Maternidade Celso Pierro the cases and almost doubled the risk of SMO (PR 1.86; CI 1.35–
da Pontifıcia Universidade Católica (Campinas, SP), Hospital 2.57) (Table 2).
Israelita Albert Einstein (São Paulo, SP), Faculdade de Medicina Moreover, the adequacy of the prenatal care received, indirectly
de Jundiaı́ (Jundiaı, SP), Hospital das Clınicas da Faculdade de evaluated by the number of visits corrected for gestational age at
Medicina de Ribeirão Preto da Universidade de São Paulo birth, was appropriate in more than 67% of the total number of
(Ribeirão Preto, SP), Santa Casa de Limeira (Limeira, SP), Santa cases and the moment of hospital admission, if still during
Casa de São Carlos (São Carlos, SP), Casa Maternal Leonor pregnancy (80%) or after giving birth (20%), showed no
Mendes de Barros (São Paulo, SP), Hospital São Paulo da association with worse outcome amongst eclamptic women
Universidade Federal de São Paulo (São Paulo, SP). (Table 3).
Variables used to evaluate the access to health care demon-
Results strated a robust association with the risk of SMO from eclampsia
(Table 3). ICU admission (PR 4.70; 95% CI 2.81–7.84),
In the one-year study period, there were 82,388 women inadequacies of monitoring (PR 2.94; 95% CI 2.13–4.07); delay
admitted to the 27 maternity hospitals participating in the study; for transfer (PR 2.32; 95% CI 1.33–4.05); lack of trained staff (PR
these women gave birth to 82,144 live births (LB). 9,555 women 1.88; 95% CI 1.20–2.93); delay for diagnosis (PR 2.29; 95% CI

Figure 2. Distribution of Non-SMO, MNM and MD in women with eclampsia, others SHD and other morbidities [Non-SMO = women
without severe maternal outcomes (MNM or MD), SMO = Severe maternal outcome (MNM = maternal near miss or MD = maternal
death), SHD = severe hypertensive disorders].
doi:10.1371/journal.pone.0097401.g002

PLOS ONE | www.plosone.org 4 May 2014 | Volume 9 | Issue 5 | e97401

 Eclampsia as Severe Maternal Morbidity

Table 1. Obstetric health indicators for eclampsia to total of cases and according to the level of income from the Brazilian regions
where the facilities are located.

Obstetric Care Indicators LI Regions N/NE/MW HI Regions S/SE Total

Maternal Near miss 48 22 70

Maternal Death 14 2 16
Live Births 39,747 42,397 82,144
SMO 62 24 86
NMM ratio/1000 LB 1.2 0.5 0.85
SMO ratio/1000 LB 1.55 0.56 1.04
NM: MD 3.4:1 11:1 4.35:1
MDI (%) 22.6 8.33 18.6
Prevalence of eclampsia per 1000 LB 8.37 2.2 5.18

SMO = severe maternal outcome, NMM = Maternal Near miss, LB = live births, MDI = maternal death index (MDI = MD/MD+NM), LI: low income, HI: high income.
doi:10.1371/journal.pone.0097401.t001

1.42–3.69), not opportune treatment (PR 2.27; 95% CI 1.48–3.46) one quarter relies on the insurance and/or private health sector.
and inadequate management of the case (PR 1.86; 95% CI 1.33– The only center included in the Brazilian Network that provides
2.60) led to a two- to four-fold increased in the risk of SMO. care exclusively to high-income private patients did not have any
Poisson multiple regression analyses confirmed admission to case of eclampsia over this one-year period.
ICU, lack adequate monitoring, severe sepsis and any previous Our study has several strengths, including a broad geographical
disease as the main independent predictors for SMO. All distribution of the participating centers, assuring the representa-
complications of pregnancy were strongly associated with tiveness of all regions (with 55.6% LB from the South and
increased risk of SMO (data not presented in tables), despite the Southeast, and 48.4% from the North, Northeast, and Midwest); a
fact that only severe sepsis, on the multiple regression analysis, rigorous three-step check for control of data quality, as described;
predicted SMO (adjusted PR 2.75; 95% CI 1.35–5.61) (Table 4). and a large sample size due to the eligibility of many of the cases.
Nevertheless, the incidence of such complications among eclamp- We found a proportion of 9.5% of SMO for eclampsia among the
tic women was not similar, fluctuating from 1% (pulmonary 910 cases of SMO from the network, very close to that found by
edema) to 12% (HELLP). (Data not presented in tables.). the WHO Multicountry Survey (9.6%) undertaken in 29 countries
Gestational age at delivery reached a median of 36 weeks, and 357 health facilities, recently published [27].
fluctuating from 22 to 42 weeks; birth weight implied a median of Figure 1 demonstrates that SMO was present in 20% of cases of
2.410 grams, ranging from 520 to 4.900 grams (data not presented eclampsia and was responsible for the majority of SMO in the
in tables) and C-section accounted for 88% of all deliveries. severe hypertensive disorders group. One of our study’s strengths
Perinatal outcomes were also substantially worse in the SMO is the design of a severity scale over which eclampsia cases can be
group: stillbirth (PR 2.34, 95% CI 1.29–4.24), neonatal ICU split into two groups. This provides a clear view of the main
admission (PR 1.84; 95% CI 1.09–3.10), neonatal death (PR 2.68; predictors of worse outcomes. In other words, even though
95% CI 1.21–5.91), low 5-min Apgar score (PR 2.87; 95% CI eclampsia is a rare event, the percentage of life threatening
1.79–4.62) and perinatal death (PR 2.3; 95% CI 1.45–3.65) complications or death due to it is still extremely high in our
(Table 5). population.
Notwithstanding the well-established association between oc-
Discussion currence of eclampsia and maternal characteristics such as age,
ethnicity, marital status, years of education and parity, our findings
Our study presents an overview of the clinical morbidities and did not identify the same patterns when assessing the risk of SMO
the access to health care for women with eclampsia in selected from eclampsia [4,22,23]. One possible hypothesis is that no
obstetric units in the five geographical Brazilian regions. These matter how robust the association between the occurrence of
results confirm a prevalence of SMO for eclampsia five times eclampsia and low Human Development Index (HDI) and its
higher than for other severe hypertensive disorders of pregnancy indicators (e.g., low schooling and income) is, once a woman
group (excluding cases of eclampsia). In fact, eclampsia is a major seizures, the outcomes mainly rely on proper and timely care,
cause of morbidity and death in this group. Multiple regression irrespective of the social and economic background characteristics
analysis pointed out that the lack of emergency care facilities in [7].
obstetric units are predictors of worse outcomes among women All associated complications (hemorrhage, HELLP syndrome,
with eclampsia. In addition, any previous disease and severe sepsis severe hypertension, pulmonary edema and sepsis) were associated
were also main predictors of SMO from eclampsia. The higher with SMO. The study was originally designed to perform a
risk of dying found when ICU admission was present shows that surveillance of severe complications in all pregnancies from the
probably there was a pre-selection of most severe cases towards participating institutions during a fixed period of time. Therefore it
admission to ICU, and the already well-known shortage of ICU was a decision to keep the data collection form as short as possible
beds in many Brazilian health care facilities has a major impact on to facilitate its implementation. Thus specific questions to go
this result [26]. deeper in each cause or associated factor were decided not to be
Three quarters of the Brazilian population are insured by SUS, included. In addition, taking into account the fact that information
the Brazilian publicly funded health system, and the remaining was cross-sectionally collected after the women was discharged, the

PLOS ONE | www.plosone.org 5 May 2014 | Volume 9 | Issue 5 | e97401

 Table 2. Rate of SMO and estimated risk of SMO for eclampsia according to maternal characteristics, obstetric background and medical history.

Characteristics SMO rate N% Non-SMO N% PR 95% CI

Age (years)
#19 33 16.4 168 83.6 0.73 0.42–1.28
20–34 45 22.5 155 77.5 Ref.
$35 8 32.0 17 68.0 1.42 0.89–2.27

PLOS ONE | www.plosone.org

Ethnicity (a)
Non white 34 19.3 142 80.7 0.92 0.56–1.53
White 23 20.9 87 79.1 Ref.
Marital Status (b)
Without partner 23 14.9 131 85.1 0.78 0.52–1.17
With partner 35 19.2 147 80.8 Ref.
Schooling (c)
Elementary 21 14.2 127 85.8 1.00 0.46–2.17
. Elementary 19 14.2 115 85.8 Ref.
Number of pregnancies (d)
1 52 17.7 242 82.3 0.75 0.54–1.05
.1 29 23.6 94 76.4 Ref.

6
Any previous disease (e)
Yes 25 27.2 67 72.8 1.86 1.35–2.57
No 36 14.6 211 85.4 Ref.
Chronic hypertension (e)
Yes 10 30.3 23 69.7 1.82 1.14–2.90
No 51 16.7 255 83.3 Ref.

SMO = severe maternal outcome (maternal near miss plus maternal death), PR: prevalence ratio adjusted by cluster effect, CI: confidence interval.
Missing data for: (a) 140 cases, (b) 90 cases, (c) 144 cases, (d) 9 cases, (e) 87 cases.
Values in bold mean they are significant.
doi:10.1371/journal.pone.0097401.t002

May 2014 | Volume 9 | Issue 5 | e97401

Eclampsia as Severe Maternal Morbidity
 Eclampsia as Severe Maternal Morbidity

Table 3. Rate of SMO and estimated risk of SMO for eclampsia according to several characteristics concerning access to health
care facilities.

Characteristics SMO rate N% Non-SMO N% PR 95% CI

Prenatal adequacy (a)

no 1 5.9 16 94.1 0.34 0.06–1.96
yes 49 17.1 237 82.9
Post-partum admission
yes 22 26.8 60 73.2 1.44 0.97–2.15
no 64 18.6 280 81.4
ICU admission
yes 73 31.4 159 68.6 4.70 2.81–7.84
no 13 6.7 181 93.3
MgSO4 prescription
no 5 35.7 9 64.3 1.82 0.79–4.20
yes 81 19.7 331 80.3
Lack of drug (b)
yes 9 27.3 24 72,7 0.83 0.83–2.45
no 68 19.1 288 80.9
Inadequate monitoring (b)
yes 21 47.7 23 52.3 2.94 2.13–4.07
no 56 16.2 289 83.8
Delay for transfer (b)
yes 12 40.0 18 60.0 2.32 1.33–4.05
no 62 17.3 297 82.7
Lack of staff (b)
yes 17 33.3 34 66.7 1.88 1.20–2.93
no 60 17.7 278 82.3
Delay for diagnosis (b)
yes 19 37.2 32 62.8 2.29 1.42–3.69
no 55 16.3 283 83.7
Treatment opportunity (b)
no 24 35.3 44 64.7 2.27 1.48–3.46
yes 50 15.6 271 84.4
Inadequate management (b)
yes 28 29.2 68 70.8 1.86 1.33–2.60
no 46 15.7 247 84.3

SMO = severe maternal outcome (maternal near miss and maternal death), PR: prevalence ratio adjusted by cluster effect, CI: confidence interval, MgSO4: magnesium
sulphate.
Missing data from: (a) 123 cases, (b) 37 cases.
Values in bold mean they are significant.
doi:10.1371/journal.pone.0097401.t003

Table 4. Variables independently associated with severe maternal outcome by Poisson multiple regression analysis (n = 321).

Factors Adjusted PR* 95% CI p

Inadequate Monitoring 2.31 1.48–3.59 0.001

ICU admission 3.61 1.77–7.35 0.001
Any previous disease 1.82 1.26–2.64 0.003
Severe Sepsis 2.75 1.35–5.61 0.007

*Considering cluster design (center/hospital), PR: prevalence ratio, CI: confidence interval, ICU = intensive care unit.
Statistical model including variables: Age, ethnicity, marital status, schooling, number of pregnancies, any previous disease, chronic hypertension, post-partum
admission, ICU admission, magnesium sulphate use, lack of medication, inadequate monitoring, delay in transfer, lack of trained staff, diagnosis delay, treatment
opportunity, inadequate management, hemorrhagic complication, HELLP syndrome, severe hypertension, pulmonary edema, severe sepsis, gestational age at birth.
doi:10.1371/journal.pone.0097401.t004

PLOS ONE | www.plosone.org 7 May 2014 | Volume 9 | Issue 5 | e97401

 Table 5. Estimated risk of SMO according to gestational age, mode of delivery and perinatal outcomes in women with eclampsia.

SMO Non-SMO PR 95% CI

n % n %

Gestational age at delivery (a)

22 to 33 weeks 24 35.3 84 28.0 1.31 0.80–2.14
$34 weeks 44 64.7 216 72.0

PLOS ONE | www.plosone.org

Mode of delivery (b)
C-section 76 91.6 290 86.8 1.51 0.74–3.11
Vaginal birth 7 8.4 44 13.2
Perinatal outcome (c)
Still births 12 16.0 17 5.5 2.34 1.29–4.24
Live births 63 84.0 293 94.5
Birth weight (d)
,2500 g 40 62.5 150 50.7 1.49 0.93–2.39
$2500 g 24 37.5 146 49.3
Neonatal outcome (e)
NICU admission 24 40.7 74 26.6 1.84 1.09–3.10

8
Neonatal death 5 8.5 9 3.2 2.68 1.21–5.91
Hospital discharge 30 50.8 195 70.1
Apgar at 5th min (f)
,7 14 24.6 22 7.7 2.87 1.79–4.62
7 a 10 43 75.4 263 92.3
Perinatal death (g)
Yes 15 21.7 24 8.2 2.30 1.45–3.65
No 54 78.3 269 91.8

SMO = severe maternal outcome (maternal near miss plus maternal death), PR: prevalence ratio adjusted for cluster effect, CI: confidence interval, NICU = neonatal intensive care unit.
Missing data from: (a) 58, (b) 8 missing cases and 1 abortion, (c) 41, (d) 66, (e) 89, (f) 68, (g) 64 cases.
Values in bold mean they are significant.
doi:10.1371/journal.pone.0097401.t005

May 2014 | Volume 9 | Issue 5 | e97401

Eclampsia as Severe Maternal Morbidity
 Eclampsia as Severe Maternal Morbidity

information regarding the exact time on each event occurred or to describe a randomly peculiar characteristic that could not have
each procedure was performed is not available. That is the reason been contemplated by the form. Thus we were able to include
why our study could only assess use and non-use of MgSO4 and insights on health systems problems. As examples we could quote:
not its appropriateness. Being a key drug to prevent seizures in ‘‘after C-section in Cabedelo city a patient was transferred to state
situations of severe preeclampsia, we can make the inference that capital Joao Pessoa (25 km) for ICU admission at 7 pm, and died
its use on the study population was almost always delayed or after several seizures at 9 pm’’; or ‘‘MgSO4 administered after C-
inappropriate, as all the included cases had seizures [28,29]. section, patient had other seizure after procedure.’’
Fourteen women (3%) did not receive MgSO4 at all, which is Brazil is a member of the BRICS nations group, which also
noteworthy, considering it is a low-cost, effective, life-saving and includes Russia, India, China and South Africa. The current
well-known drug. Our study, however, did not find a significantly economic up growth, combined with a significant influence on
increased risk of SMO from eclampsia in cases to whom MgSO4 regional and global matters, bond these emerging nations. It is well
was not administered. The remaining 97% of cases in our study known that social and educational improvements do not always
were prescribed MgSO4 at any given time, what is a remarkably progress hand-in-hand with the economic boom, and this is still a
high figure, especially when compared to the finding of 89% challenge not only for Brazil but for the whole BRICS community.
MgSO4 use reported in the recently published WHO Multi- In conclusion, improvements in social and educational struc-
country Survey that evaluated 29 different LMIC countries [27]. tures alone will probably not lead to the needed changes on time
Only 8% of the cases were reported as having no access to the for the Millennium Development Goal number 5 to be achieved
drug in any given time before or after delivery. by 2015. Our findings point out clearly that lower income regions
We found a 5-fold increase in the risk of SMO among our in Brazil have a worse performance in all obstetric health care
population with eclampsia, mostly as a consequence of delays in indicators among women with eclampsia. The strengthening of
diagnosis, delay in transportation. inadequate management, lack
health systems might be a possible strategy to reduce morbidity
of well-trained staff and lack of intensive care unit These variables
and deaths in women of reproductive age and their offspring
quantify barriers and delays for proper obstetric care, and our
[28,34]. It is known that social and economic determinants are
findings reinforce studies that pointed those factors out as the main
associated with higher maternal and perinatal mortality [3,32].
challenges for improving maternal and perinatal health care in
Waiting for changes in those patterns in order to get better
LMIC [7,8,30,31]. The 3-fold higher risk of SMO in women that
obstetric and perinatal outcomes might not be the faster route to
had difficulties in accessing the obstetric ICU corroborates other
reduce SMO due to eclampsia. Instead, qualifying emergency
findings that had already pointed to the need for staff training and
better infrastructure for maternal facilities and obstetric ICUs for obstetric health care by promoting continued staff training and
the delivery of prompt and adequate care to severe obstetric increasing the number of well-equipped health care facilities
complications [26,28]. These categorizations were of course (especially obstetric ICU beds) are a more plausible and expedient
attributed with the knowledge of the SMO status. If a risk of pathway not only for Brazil, but also for all other LMIC and
information bias is likely to exist, this would probably be in the emerging nations who endeavor to relieve the burden of
way of diminishing, and not showing that these substandard care/ eclampsia.
delays in fact occurred, considering that both local investigators
and coordinators were also part of the clinical staff of each Acknowledgments
participating center. The authors acknowledge the involvement of the members of its Steering
To the best of our knowledge there are no published findings Committee and all the other investigators and coordinators from all the
concerning perinatal outcomes from two different groups of centres involved in the National Network for the Surveillance of Severe
severity (SMO and non-SMO) of eclampsia, making comparisons Maternal Morbidity. The authors acknowledge the contribution of the
difficult to be established. We found a 10% total prevalence of Brazilian Network for the Surveillance of Severe Maternal
perinatal mortality among cases of eclampsia. According to a Morbidity Group: João P Souza, Rodolfo C Pacagnella, Rodrigo S.
recent systematic review, a Nigerian study presenting perinatal Camargo, Vilma Zotareli, Lúcio T. Gurgel, Lale Say, Robert C Pattinson,
outcomes in eclampsia treated with MgSO4 found 30% incidence Marilza V Rudge, Iracema M Calderon, Maria V Bahamondes, Danielly S
Santana, Simone P Gonçalves, Eliana M Amaral, Olı́mpio B Moraes Filho,
of perinatal mortality, and a British one, 6%. Considering the 97%
Simone A Carvalho, Francisco E Feitosa, George N Chaves, Ione R Brum,
use of magnesium sulphate in our cases, we could argue that our Gloria C Saint’Ynes, Carlos A Menezes, Patricia N Santos, Everardo M
perinatal mortality amongst cases of eclampsia is more similar to Guanabara, Elson J Almeida Jr, Joaquim L Moreira, Maria R Sousa,
that of the UK than to that of Nigeria [32,33]. Frederico A Peret, Liv B Paula, Luiza E Schmaltz, Cleire Pessoni, Leila
Regarding the possible limitations of our study, the database Katz, Adriana Bione, Antonio C Barbosa Lima, Edilberto A Rocha Filho,
cannot bet understood as representative of the whole Brazilian Melania M Amorim, Debora Leite, Ivelyne Radaci, Marilia G Martins,
population. However it had a multicenter cross sectional design Frederico Barroso, Fernando C Oliveira Jr, Denis J Nascimento, Cláudio S
and an appropriate sample size. Secondly, some maternal Paiva, Moises D Lima, Djacyr M Freire, Roger D Rohloff, Simone M
characteristics are challenging to evaluate, for instance, numbers Rodrigues, Sergio M Costa, Lucia C Pfitscher, Adriana G Luz, Daniela
Guimaraes, Gustavo Lobato, Marcos Nakamura-Pereira, Eduardo Cor-
on skin color and years of education were missing in approxi-
dioli, Alessandra Peterossi, Cynthia D Perez, Jose C Peraçoli, Roberto A
mately a third of the database, marital status in a fifth - yet those Costa, Nelson L Maia Filho, Jacinta P Matias, Silvana M Quintana, Elaine
variables did not appear as predictors of SMO. At the same time, C Moises, Fátima A Lotufo, Luiz E Carvalho, Carla B Andreucci, Márcia
the key variables that predicted maternal SMO had less than 10% M Aquino, Maria H Ohnuma, Rosiane Mattar and Felipe F Campanharo.
of missing data. One of the particularities of studying NM cases is
the possibility to interview women after life-threatening events, Author Contributions
thus identifying breakdowns in health systems [16]. In our study
no interviews were undertaken however, but we developed a Conceived and designed the experiments: JCG MAP JGC SMH.
Performed the experiments: JGC MAP SMH MLC FGS JLPS. Analyzed
structured form and trained investigators to gather information on
the data: JGC MAP MHS. Contributed reagents/materials/analysis tools:
access to care not only from medical records, but also from MAP JCG SMH MHS. Wrote the paper: JCG MAP MLC.
hospital staff, and we included an open variable that could be used

PLOS ONE | www.plosone.org 9 May 2014 | Volume 9 | Issue 5 | e97401

 Eclampsia as Severe Maternal Morbidity

References
1. Report of the National High Blood Pressure Education Program Working 19. Cecatti JG, Souza JP, Parpinelli MA, Haddad SM, Camargo RS, et al. (2009)
Group on High Blood Pressure in Pregnancy (2000) American Journal of Brazilian Network for the surveillance of maternal potentially life-threatening
Obstetris and Gynecology 183(1): S1–S22. morbidity and maternal near-miss and a multidimensional evaluation of their
2. Duley L (2009) The global impact of pre-eclampsia and eclampsia. Seminars in long term consequences. Reproductive Health 6: 15.
Perinatology 33(3): 130–7. 20. Haddad SM, Cecatti JG, Parpinelli MA, Souza JP, Costa ML, et al. (2011) From
3. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF (2006) WHO planning to practice: building the national network for the Surveillance of Severe
analysis of causes of maternal death: a systematic review. Lancet 367(9516): Maternal Morbidity. BMC Public Health 11: 283.
1066–74. 21. PNUD Brazil (2010) Program of the United Nations for Development.
4. Fong A, Chau CT, Pan D, Ogunyemi DA (2013) Clinical morbidities, trends, Available: http://www.pnud.org.br/atlas/ranking/Ranking-IDHM-UF-2010.
and demographics of eclampsia: a population-based study. American Journal of aspx. Accessed Sep 10 2013.
Obstetris and Gynecology 209(3): 229.e1–7. 22. Morikawa M, Cho K, Yamada T, Yamada T, Sato S, et al. (2012) Risk factors
5. Abalos E, Cuesta C, Grosso A, Chou D, Say L (2013) Global and regional for eclampsia in Japan between 2005 and 2009. International Journal of
estimates of preeclampsia and eclampsia: a systematic review. European Journal Gynaecology and Obstetrics 117(1): 66–8.
of Obstetetrics and Gynecology 170(1): 1–7. 23. Knight M (2007) Eclampsia in the United Kingdom 2005. British Journal of
6. Souza JP, Cecatti JG, Parpinelli MA, Sousa MH, Lago TG, et al. (2010) Obstetrics and Gynaecology 114(9): 1072–8.
Maternal morbidity and near miss in the community: findings from the 2006 24. Haddad SM, Sousa MH, Cecatti JG, Parpinelli MA, Costa ML, et al. (2012)
Brazilian demographic health survey. BJOG 117(13): 1586–92. Intraclass correlation coefficients in the Brazilian Network for Surveillance of
7. Firoz T, Sanghvi H, Merialdi M, von Dadelszen P (2011) Pre-eclampsia in low Severe Maternal Morbidity study. BMC Pregnancy and Childbirth 12: 101.
and middle income countries. Best Practice & Research Clinical Obstetrics & 25. Donne A, Klar N (1994) Methods for comparing event rates in intervention
Gynaecology 25(4): 537–48. studies when the unit of allocation is a cluster. American Journal of
8. Ronsmans C, Campbell O (2011) Quantifying the fall in mortality associated Epidemiology 140(3): 279–89.
with interventions related to hypertensive diseases of pregnancy. BMC Public
26. Oliveira Neto AF, Parpinelli MA, Cecatti JG, Souza JP, Sousa MH (2009)
Health 11 Suppl 3: S8.
Factors associated with maternal death in women admitted to an intensive care
9. Kullberg G, Lindeberg S, Hanson U (2002) Eclampsia in Sweden. Hypertension
unit with severe maternal morbidity. International Journal of Gynaecology and
in Pregnancy 21(1): 13–21.
Obstetrics 105(3): 252–6.
10. Khosla AH, Dahiya K, Sangwan K (2000) Maternal mortality and ‘near-miss’ in
27. Souza JP, Gülmezoglu AM, Vogel J, Carroli G, Lumbiganon P, et al. (2013)
rural north India. International Journal of Gynaecology and Obstetrics 68(2):
Moving beyond essential interventions for reduction of maternal mortality (the
163–4.
WHO Multicountry Survey on Maternal and Newborn Health): a cross-
11. United Nations (2013) Millennium Development Goals, Available: http://www.
sectional study. Lancet 381 (9879): 1747–55.
un.org/millenniumgoals/maternal.shtml.Accessed Jun 2013.
12. Brazil (2010) Presidência da República [Internet]. [Millenium Development 28. Schutte JM, Schuitemaker N, Van Roosmalen J, Steegers E on behalf of the
Goals: national report of follow up]. Brası́lia (DF): IPEA. Accessed June 14th, Dutch Maternal Mortality Committee (2008) Substandard care in maternal
2013. Available: ,http://mds.gov.br/bolsafamilia/arquivos/relatorio-nacional- mortality due to hypertensive disease in pregnancy in the Netherlands. British
de acompanhamento.pdf/download$. Journal of Obstetrics and Gynaecology 115: 732–736.
13. Mantel GD, Buchmann E, Rees H, Pattinson RC (1998) Severe acute maternal 29. Duley L, Henderson-Smart DJ, Walker GJ, Chou D (2010) Magnesium sulphate
morbidity: a pilot study of a definition for a near-miss. British Journal of versus diazepam for eclampsia. Cochrane Database Systematic Reviews (12):
Obstetrics and Gynaecology 105(9): 985–90. CD000127.
14. Waterstone M, Bewley S, Wolfe C (2001) Incidence and predictors of severe 30. Knight HE, Self A, Kennedy SH (2013) Why are women dying when they reach
obstetric morbidity: case-control study. BMJ 322(7294): 1089–93. hospital on time? A systematic review of the ‘‘Third Delay’’. PloSOne 8(5):
15. Pattinson R, Say L, Souza JP, Broek N, Rooney C (2009) WHO maternal death e63846.
and near-miss classifications. Bulletim of the World Health Organization 87(10): 31. WHO. Evaluating the quality of care for severe pregnancy complications. The
734. WHO near-miss approach for maternal health. 2011. Available: http://
16. Say L, Souza JP, Pattinson RC, WHO working group on Maternal Mortality whqlibdoc.who.int/publications/2011/9789241502221_eng.pdf Accessed 24
and Morbidity classifications (2009) Maternal near miss-towards a standard tool Aug 2013.
for monitoring quality of maternal health care. Best Practice & Research Clinical 32. Ghulmiyyah L, Sibai B (2012) Maternal Mortality from Preclampsia/Eclampsia.
Obstetrics & Gynaecology 23(3): 287–96. Seminars in Perinatology 36: 56–59.
17. Sousa MH, Cecatti JG, Hardy EE, Serruya SJ (2008) Severe maternal morbidity 33. McDonald SD, Lutsiv O, Dzaja N, Duley L (2012) A systematic review of
(near miss) as a sentinel event of maternal death. An attempt to use routine data maternal and infant outcomes following magnesium sulphate for pre-eclampsia/
for surveillance. Reproductive Health 5: 6. eclampsia in real-world use. International Journal of Gynaecology and
18. DATASUS. [Information on health. Vital statistics]. Available: http://www2. Obstetrics 118(2): 90–6.
datasus.gov.br/DATASUS/index.php?area = 0205&VObj = http://tabnet. 34. WHO (2011) Evaluating the quality of care for severe pregnancy complications:
datasus.gov.br/cgi/deftohtm.exe?sinasc/cnv/nv. Accessed Aug 15 2013. The WHO near-miss approach for maternal health. Geneva: WHO.

PLOS ONE | www.plosone.org 10 May 2014 | Volume 9 | Issue 5 | e97401