Review Article

Accepted Article
Maternal sepsis is an evolving challenge

Michael J. Turner

UCD Centre for Human Reproduction, Coombe Women and Infants University

Hospital, Dublin, Ireland.

Corresponding author: Michael J. Turner

UCD Centre for Human Reproduction, Coombe Women and Infants University

Hospital, Cork St, Dublin 8, Ireland D08 YW7X.

Email: [email protected]

Key words: Early neonatal sepsis; Maternal infection; Maternal infection

complications; Maternal morbidity; Maternal mortality; Maternal sepsis; Maternal

sepsis criteria; Obstetric early warning scores.

Synopsis: Sepsis remains an important cause of adverse pregnancy outcomes

worldwide. New agreed definitions of maternal sepsis are welcome, but further work

is required on diagnostic criteria.

This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/ijgo.12833
This article is protected by copyright. All rights reserved.
 Abstract
Accepted Article Despite major advances in the last century, particularly in high resource settings,

maternal sepsis remains a common and potentially preventable cause of direct

maternal death globally. A barrier to further progress has been the lack of consensus

on the definition of maternal sepsis. Publications from two recent multidisciplinary

consensus conferences, one on sepsis in the non-pregnant adult and the other on

sepsis in the pregnant woman, concluded that the criteria for diagnosing sepsis

should be clinically based, applicable at the bedside, and should not be laboratory-

based. Informed by reviews of the evidence, in 2017 WHO published a new

definition of maternal sepsis based on the presence of suspected or confirmed

infection. It also announced a Global Maternal and Neonatal Sepsis Initiative to

identify the diagnostic criteria for the early identification, epidemiology, and disease

classification of maternal sepsis. Standardizing the criteria for maternal sepsis

optimizes clinical audit and research. It may facilitate the evaluation of the role of

different clinical parameters and biomarkers in the diagnosis, earlier recognition and

management of maternal infection and sepsis. Further work is required to develop an

international consensus on the criteria for diagnosing maternal sepsis and any

associated organ dysfunction.

1 New guidelines on sepsis

In both low-income and high-income countries, sepsis is a common cause of death

and consumes considerable healthcare resources. Maternal infection, particularly if

left untreated or treated inappropriately, may lead to sepsis which is strongly

associated with critical illness and an increased risk of premature death [1].

This article is protected by copyright. All rights reserved.

 In the past there was little consensus internationally about the definition of sepsis
Accepted Article and, furthermore, the quality of clinical data was generally poor which hindered audit

and research, particularly research on the early recognition and optimum

management of sepsis. Over the last 30 years, two major factors have driven the

need for better definitions [2]. First, there have been major technical advances in

providing organ support in critical care units, particularly in well-resourced settings.

Second, there is an increased awareness that the human and financial costs of

critical illnesses due to sepsis are potentially preventable.

Following a Consensus Conference, in 1992 several North American and European

intensive care societies published recommendations on the definition of sepsis and

its management [1–4]. The Conference introduced the concept of the Systemic

Inflammatory Response Syndrome (SIRS) based on the patient having more than

one abnormal recording of the following parameters: temperature, heart rate,

respiratory rate or PaCO2, and white cell count. Sepsis was defined as SIRS plus

infection. A second Consensus Conference by the societies a decade later revisited

the definitions of sepsis and related conditions [3]. After discussion it was agreed

that, while unsatisfactory, the concepts of sepsis, severe sepsis, and septic shock

should remain unchanged. It was also agreed that, while SIRS remained a useful

concept, the diagnostic criteria for SIRS were overly sensitive and non-specific. The

use of biomarkers to diagnose sepsis was considered premature. Revised

international guidelines for the management of severe sepsis and septic shock were

subsequently published in 2004, 2008, and 2012 [1–4].

This article is protected by copyright. All rights reserved.

 In 2015–2016, a Third Consensus Conference of specialists representing 30
Accepted Article international organizations recommended a revised definition of sepsis [4]. The

Conference decided unanimously that the use of SIRS criteria was unhelpful and

should be abandoned. Studies had shown that SIRS is nearly ubiquitous in hospital

patients and occurs in many benign conditions which are unrelated to infection [2].

Abnormal SIRS parameters are too non-specific to be helpful in diagnosing sepsis.

Sepsis is now defined as “life-threatening organ dysfunction caused by a

dysregulated host response to infection” [4].

The Conference recognized that sepsis was a syndrome without a validated

standard diagnostic test, and this had been problematic in measuring incidence and

mortality rates [4]. The syndrome is also determined by pathogen factors and by host

factors which may evolve over time in an individual patient depending on the clinical

management.

One scoring system for assessing the severity of organ dysfunction in the setting of

an intensive care unit is the Sepsis Organ Failure Assessment (SOFA) scoring

system based on oxygenation, platelet count, bilirubin level, mean arterial blood

pressure, Glasgow Coma Score, and renal assessment (creatinine and urinary

output) [4]. An acute increase of >1 SOFA points could be used as a proxy for organ

dysfunction.

Outside the intensive care unit setting, it was suggested that a quick SOFA (qSOFA)

could be used to assess organ dysfunction [4]. This is based on two abnormalities

out of three parameters that could be assessed at the bedside, namely, altered

This article is protected by copyright. All rights reserved.

 mentation, a systolic blood pressure less than 101 mm Hg, or a respiratory rate
Accepted Article greater than 21/minute. The qSOFA can be calculated quickly without the need for

laboratory tests, particularly in resource-poor settings. However, SOFA or qSOFA

scores have not been widely used outside a critical care setting and other scoring

systems exist.

The Conference also recommended dropping the term “severe sepsis” and they

redefined severe septic shock. They suggested that the new definitions be

designated Sepsis-3. The latest consensus report from the Surviving Sepsis

Campaign international guidelines is comprehensive and makes 89

recommendations for the early management and resuscitation of patients with sepsis

or septic shock [1]. For the first time, pediatric guidelines will appear in a separate

document. It is notable that the report, like all previous editions, does not make any

recommendations specifically for pregnancy, but the guidelines are applicable for

sepsis complicating infections in gynecology [1].

2 Maternal Sepsis

Maternal sepsis was a common cause of death in the 18th and 19th centuries and

resulted in half of reported maternal deaths in Europe [5]. Improved living conditions,

the introduction of antibiotics, and advances in acute hospital care and medical

specialization led to major reductions in critical illnesses and death associated with

sepsis in the 20th century. Until recently, the main focus in preventing direct

maternal deaths worldwide has been on improving the management of obstetric

hemorrhage, pregnancy hypertension, and venous thromboembolism [6].

This article is protected by copyright. All rights reserved.

 At the start of the 21st century, however, maternal deaths due to sepsis in high-
Accepted Article income countries were highlighted by individual tragedies and by national obstetric

audits. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the

United Kingdom for the years 2006–2008 found that, since previous reports, there

has been a decline in the number of maternal deaths, predominantly as a result of a

reduction in deaths from thromboembolism and, to a lesser extent, hemorrhage [7].

Despite this overall decline, there had been an increase in deaths due to genital tract

sepsis, particularly from community acquired group A streptococcal disease [7].

Sepsis has emerged as the most common cause of direct maternal death in the UK

[7].

WHO estimated that the global prevalence of maternal sepsis is 4.4% among live

births, with an incidence of 9-49 per 100 000 deliveries in high-income countries

depending on the definition used and population studied [8]. There is a lack of data

from low-income countries, but it is estimated that sepsis accounts for one in 10

maternal deaths globally. Furthermore, maternal sepsis is strongly associated with

an increase in perinatal mortality and morbidity as a result of infection. These reports

led to a renewed focus on maternal sepsis with the publication of national clinical

guidelines, the development of obstetric early warning scores (EWS) to detect critical

illness, the application of care bundles, and a growing awareness of the need to treat

maternal infection early and appropriately [7,9,10]. Recognition of the need to reduce

maternal mortality and morbidity in the USA resulted in the establishment of the

National Partnership for Maternal Safety, which planned to develop Safety Bundles

that focus on sepsis identification and treatment, as well as supplemental bundles on

Maternal Early Warning Criteria [6]. The Partnership is collaborating with several

This article is protected by copyright. All rights reserved.

 national organizations including the American College of Obstetricians and
Accepted Article Gynecologists.

One of the barriers hindering progress in the prevention and management of

maternal sepsis has been the lack of consensus about definitions and the variety of

terminology often used interchangeably. Thus, a systematic review of 78 citations

and a multidisciplinary expert consultation were conducted in 2016, which led to the

development of a consensus definition [11]. As a result, there is a new WHO

definition of maternal sepsis: “a life-threatening condition defined as organ

dysfunction resulting from infection during pregnancy, childbirth, post-abortion, or

postpartum period” [11]. This definition has been endorsed by a number of

organizations including the Surviving Sepsis Campaign and the International

Federation of Gynecology and Obstetrics (FIGO). As part of the consultation

process, the majority of the experts were of the view that the definition of organ

dysfunction should be adapted from the Sepsis-3 consensus for adults, and based

on specific criteria for the obstetric population [11]. Experts preferred clinical markers

over laboratory investigations, and preferred thresholds adapted to the obstetric

population over those used for qSOFA.

In a previous systematic review and meta-analysis completed in June 2013 of 87

eligible studies, the normal range for each component of the SIRS in healthy

pregnant women was examined [12]. The review found that the normal range for

physiological and laboratory measurements prepartum and postpartum overlapped

substantially with SIRS criteria. In particular, the respiratory rate, PaCO2, heart rate,

and white cell count during a normal healthy pregnancy may meet the criteria for

This article is protected by copyright. All rights reserved.

 SIRS [12]. Thus, the SIRS criteria used to diagnose sepsis in the non-pregnant adult
Accepted Article are not appropriate for use in diagnosing maternal sepsis. In particular, the review

found a scarcity of data on the normal temperature during pregnancy, and there was

difficulty establishing a normal white blood cell count during labor. The authors

concluded that redefined criteria would be required to facilitate early diagnosis and to

prevent adverse outcomes in women with maternal sepsis. It is also important to

note that maternity EWS are designed to enable early identification of all critical

illnesses and not just sepsis. Various parameters may differ in their sensitivity and

specificity to identify abnormalities depending on the critical illness studied.

In Ireland, a Maternity Early Warning System (IMEWS) was successfully

implemented in all hospitals in April 2014 [10]. It was designated a system rather

than a score to emphasize that it was intended to complement clinical judgment and

not replace it. There was a concern that clinicians might focus solely on a score at

the expense of the traditional history and examination. As part of the IMEWS, the

SIRS criteria were adapted to allow for the physiological changes of pregnancy in

respiratory rate, heart rate, temperature, and white cell count [13].

In a retrospective evaluation of the customized criteria for the diagnosis of maternal

sepsis compared with the standard non-pregnant SIRS criteria, women with proven

bacteremia during 2009–2014 were reviewed [13]. Of the 93 women with bacteremia

out of 52 032 deliveries, 61 (66%) had sepsis based on the standard criteria, in

comparison with 52 (56%) based on the customized criteria (not significant).

However, the standard criteria placed more emphasis on the finding of a

This article is protected by copyright. All rights reserved.

 leukocytosis, which could be physiological, while the customized criteria relied more
Accepted Article on hyperthermia.

3 Maternal infection

Terminology, together with variations in the classification of infection, had been a

barrier hindering progress in optimizing the management and audit of maternal

sepsis. Some reports base classification on organisms, such as group B

streptococcus, and others on the organ or system infected, for example, the urinary

tract. Some base classification on the timing of infection in relationship to delivery—

pre-, peri- or postpartum—and the time limit for postpartum infection can vary from

10 to 42 days after delivery [7].

The diagnosis of maternal infection may not be confirmed microbiologically and also

the organism identified on microbiological culture of a high vaginal swab, for

example, may not be the organism responsible for infection. Isolation of an organism

on culture may be the result of contamination rather than infection, for example, on

blood culture. Some classifications are based on operative procedures such as

surgical site infection after cesarean delivery.

There are variations in the definition of infection. Chorioamnionitis, for example, may

be a clinical diagnosis or a histological diagnosis, and may or may not be

underpinned by a positive microbiological culture. These variations in the diagnosis

of maternal infection make it difficult to compare rates in different settings, and to

compare clinical outcomes either in practise or in a research setting. It hinders both

This article is protected by copyright. All rights reserved.

 the validation of obstetric EWS in identifying infection and the evaluation of the
Accepted Article effectiveness of therapeutic interventions.

The classification suggested is shown in Table 1. Maternal infection may be

classified into three categories: infections specific to pregnancy; infections

exacerbated by pregnancy; and infections incidental to pregnancy. These are

mutually exclusive categories which are aligned with the WHO classification used

worldwide of maternal mortality into direct, indirect, and incidental deaths.

The infections specific to pregnancy include chorioamnionitis and endometritis which

occur anatomically at the same site. As well as treating the chorioamnionitis,

consideration may have to be given to delivering the baby which may be challenging,

particularly around the age of fetal viability. As well as treating endometritis,

consideration may have to be given to exploring the uterine cavity for retained

placental products which risks traumatizing the soft uterine wall. The presentation of

chorioamnionitis and endometritis may be florid or subtle. They are both strongly

associated with a history of rupture of the membranes and/or cervical dilatation. As

the placental bed is highly vascular and closely related to the maternal circulation,

maternal bacteremia leading to sepsis can develop quickly. The site of infection may

not be obvious and is not visible to the eye. Microbiological cultures of the uterine

cavity are technically difficult to obtain without contamination by vaginal flora, and the

appropriate choice of antibiotics may not be immediately obvious.

This article is protected by copyright. All rights reserved.

 The other three infections specific to pregnancy, that is, breast, perineal, and
Accepted Article surgical site infections (for example, cesarean delivery site), are more common. The

source of infection is visible to the eye and more obvious by the presence of

localized pain, erythema, or discharge. Microbiological culture and the choice of

antibiotics is usually straightforward. These infections rarely lead to maternal

bacteremia or sepsis unless they are left untreated. Infections specific to pregnancy

are strongly associated with a breach in the physical integrity of the skin, or a breach

in the physical barriers of a closed cervix and intact gestational sac. When such

breaches occur, prevention with antibiotic prophylaxis or the early treatment of

infection is key to avoiding the onset of maternal sepsis.

It is known that pregnancy does not suppress the maternal immune system as

previously thought, but rather modulates immunity. As a result, maternal

susceptibility to certain infections are increased, for example, influenza, listeriosis,

hepatitis E virus, herpes simplex virus, and malaria [14]. The evidence for increased

susceptibility in pregnancy is more limited for measles and varicella.

Finally, maternal infections may be incidental to pregnancy, for example, lower

respiratory tract infections. There are additional challenges with emerging infections

incidental to pregnancy, particularly as women travel abroad more often to exotic

locations. Further work is also required, therefore, on the definition of individual

maternal infections [15]. It is suggested that a new classification of maternal infection

aligned with the WHO classification of maternal mortality will help determine the

optimum parameters for the early recognition of sepsis complicating maternal

infections.

This article is protected by copyright. All rights reserved.

 4 Conclusion
Accepted Article The recent WHO clinically based definition of maternal sepsis is a welcome step

forward. A standardized definition that can be applied at the bedside in all settings

without recourse to laboratory investigations will improve epidemiological

surveillance and evaluation of therapeutic interventions. There is still work to be

done, however, in determining the optimum maternal parameters for the diagnostic

criteria for both infection and sepsis which balances sensitivity with specificity. In

evaluating new criteria, a distinction also needs to be drawn between diagnostic

validation and prognostic validation. The early diagnosis and early commencement

of treatment are both key in preventing the onset of maternal sepsis globally [15].

Conflicts of interest

The author has no conflicts of interest.

Author contribution

The author is solely responsible for conceiving and writing the article.

This article is protected by copyright. All rights reserved.

 References
Accepted Article
[1] Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli MA, Ferrer RF et al.
Surviving Sepsis Campaign: International Guidelines for Management of
Sepsis and Septic Shock: 2016. Intensive Care Med 2017;43:304-377.

[2] Abraham E. New Definitions for Sepsis and Septic Shock. Continuing
Evolution but With Much Still to Be Done. JAMA 2016;315:757-9.

[3] Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D et al. 2001
SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference.
Crit Care Med 2003;31:1250-6.

[4] Singer M, Deutschman CS, Warren CW, Shankar-Hari M, Annane D, Bauer M

et al. The Third International Consensus Definitions for Sepsis and Septic
Shock (Sepsis-3).
JAMA 2016;315:801-10

[5] Bamfo JEAK. Managing the risks of sepsis in pregnancy. Best Pract Res Clin
Obstet Gynaecol 2013;27:583-5.

[6] D’Alton ME, Main EK, Menard MK, Levy BS. The National Partnership for
Maternal Safety. Obstet Gynecol 2014;123:973-7.

[7] Cantwell R, Clutton-Brock T, Cooper G, Dawson A, Drife J, Garrod D, et al.

Centre for Maternal and Child Enquiries (CMACE). Saving Mothers’ Lives:
reviewing maternal deaths to make motherhood safer: 2006-2008. The Eighth
Report into Maternal Deaths in the United Kingdom. BJOG 2011;118(Suppl.
1):1-203

[8] Bonet M, Nogueira-Pileggi V, Rijken MJ, Coomarasamy A, Lissauer D, Souza

JP et al.
Towards a consensus definition of maternal sepsis: results of a systematic
review and expert consultation. Reprod Health 2017;14:67-80.

[9] Turner MJ. Bacterial Infections Specific to Pregnancy. Clinical Practice

Guideline. Institute of Obstetricians and Gynaecologists. Royal College of
Physicians of Ireland. March 2015. https://rcpi-live-
cdn.s3.amazonaws.com/wp-content/uploads/2016/05/29.-Bacterial-Infection-
Specific-to-Pregnancy.pdf

[10] Irish Maternity Early Warning System. National Clinical Effectiveness

Committee National Clinical Guideline No.4 Version 2. Department of Health
2019. http://health.gov.ie/national-patient-safety-office/ncec/

[11] Statement on Maternal Sepsis. World Health

Organisation. 2017. http://www.who.int/reproductivehealth/publications/matern
al_perinatal_health/maternalsepsis-statement/en/. Accessed April 01, 2019.

This article is protected by copyright. All rights reserved.

 [12] Bauer ME, Bauer ST, Rajala B, MacEachren MP, Polley LS, Childers D et al.
Maternal Physiologic Parameters in Relationship to Systemic
Accepted Article Inflammatory Response Syndrome Criteria. A Systematic Review and
Meta-analysis. Obstet Gynecol 2014;124:535-41

[13] Maguire PJ, Power KA, Downey AF, O’Higgins AC, Sheehan SR, Turner MJ.
Evaluation of the Systemic Inflammatory Response Syndrome criteria for the
diagnosis of sepsis due to maternal bacteremia. Int J Gynecol Obstet
2016; 133:116-9.

[14] Kourtis AP, Read JS, Jamieson DJ. Pregnancy and infection. New Eng J
Med 2014;370:2211-8.

[15] O'Regan C, O'Malley EG, Power KA, Reynolds CME, Sheehan SR, Turner
MJ.Development of a novel bedside index for the early identification of severe
maternal infection. Eur J Obstet Gynecol Reprod Biol 2019;235:26-29.

This article is protected by copyright. All rights reserved.

 Table 1 Classification of maternal infection aligned with WHO classification of maternal death [9,
14].
Accepted Article
1. Pregnancy-specific infections
Chorioamnionitis
Endometritis
Lactational mastitis
Site of perineal trauma
Surgical site, e.g. cesarean
2. Infections exacerbated by pregnancy, including:
Urinary tract infection
Influenza
Listeriosis
Hepatitis E
Herpes simplex virus
Malaria
3. Incidental infections, including:
Lower respiratory trace infection
Tuberculosis
Sexually transmitted diseases

Adapted from [10].

This article is protected by copyright. All rights reserved.