Obstetric High Complexity Unit of the Fundacion Valle del Lili, Cali – ColombiaDivision of Obstetrics y Gynecology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, ChileCenter for Research and Innovation in Maternal-Fetal Medicine (CIMAF), Department of Obstetrics and Gynecology, Sótero del Río Hospital, Santiago, ChileRepresentative for Latin America at the FIGO Committee for Safe Motherhood and Newborn Health for the International Federation of Gynecology and Obstetrics (FIGO)Professor at the Department of Gynecology and Obstetrics at the University ICESI, Cali – ColombiaGynecology and Obstetrics Service and Obstetric High Complexity Unit of the Fundacion Valle del Lili, Cali – ColombiaSpecialist in Critical Medicine and Intensive Care, Universidad del Valle, Cali – ColombiaMasters in Clinical Epidemiology, Universidad de la Frontera, Temuco – ChileObstetrician Gynecologist, Universidad del Valle, Cali – Colombia

INTRODUCTION

In 2017, the World Health Assembly (WHA), the World Health Organization’s decision-making body, adopted a resolution on improving the prevention, diagnosis, and management of sepsis. The WHA resolution recognized sepsis as a major threat to patient safety and global health with the potential to save millions of lives if a proper approximation is made.¹ Indeed, the systematic analysis conducted in 2014, including 416 databases from 115 countries, reported a total of 60,799 maternal deaths; of those, sepsis was the cause of death in 10.7% of the cases.² The reduction of maternal deaths is a priority for achieving the Sustainable Development Goals, implementing the UN Global Strategy for Women’s, Children’s and Adolescents’ Health and critical for the Strategies toward Ending Preventable Maternal Mortality (EPMM),² the third most common direct cause of maternal mortality, maternal sepsis, received less attention, research, and programming.

Undetected or poorly managed maternal infections can lead to sepsis, death, or disability for the mother as well as an increased likelihood of early neonatal infection and other adverse outcomes. Infections cause approximately 11% of maternal deaths and are also a significant contributor to many deaths attributed to other conditions. The risk of early neonatal sepsis increases when maternal infection is present, and early neonatal sepsis causes about 8% of all neonatal deaths.³

Compared to other pregnancy complications, the case fatality rate of maternal sepsis is very high.^4,5 Approximately 8 to 12% of admissions of obstetric patients to intensive care units (ICUs) are due to sepsis.⁴ Sepsis with acute organ dysfunction had a mortality between 20 and 40% in high-income countries in the early 2000s, but more recent data shows an overall rate between 8% and 14% in women with septic shock. The UK United Kingdomreported that 19.5% of pregnant women with confirmed sepsis evolved to septic shock and, of those, 1.4% evolved to death.⁵

The reasons to explain this high mortality and morbidity in maternal sepsis are related to delays in the identification of cases and in a non-standardized management. For example, in obstetric population, there is no agreement with the definitions of sepsis and septic shock of the The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) in 2016.⁶ The effective prevention, early identification, and adequate management of maternal and neonatal infections and sepsis can contribute to reduce the burden of infection as an underlying and contributing cause of morbidity and mortality. The objective of this chapter is to review the evidence related to the recognition of sepsis during pregnancy.

PHYSIOLOGICAL CHANGES OF PREGNANCY

Physiological, immunological, and mechanical changes in pregnancy make pregnant women more susceptible to infections compared to non-pregnant women, particularly during the postpartum period. Furthermore, physiological adaptations to pregnancy, maternal efforts during the second stage of labourlabor, interventions during labourlabor, or blood loss, may obscure signs and symptoms of infection and sepsis. This may result in a delay in the recognition and treatment of sepsis.

Changes in the cardiovascular system during pregnancy are profound and begin as early as eight weeks of gestation. Throughout gestation, the intravascular volume gradually increases approximately 30 to 50% due to the effect of the renin-angiotensin-aldosterone system. In response, the maternal heart rate increases, reaching values of 10 to 20 beats per minute faster than the normal rate for an adult. By Frank–-Starling lLaw, the cardiac output (CO) in that situation can increase by up to 50%. This is achieved predominantly via an increase in stroke volume, but also to a lesser extent, via an increase in heart rate. An increase in stroke volume is possible due to the early increase in ventricular wall muscle mass and end-diastolic volume (but not end-diastolic pressure) observed during pregnancy. The heart is physiologically dilated, and the myocardial contractility is increased. The peripheral vasodilatation is mediated by endothelium-dependent factors, including nitric oxide synthesis and upregulated by estradiol and possibly by vasodilatory prostaglandins (e.g. PGI2). Peripheral vasodilation leads to a 25–30% fall in systemic vascular resistance causing a decrease in blood pressure between 12 and 26 weeks, increasing again around the 36th week.⁷

The adequacy of tissue oxygenation depends on the rate of oxygen delivered (DO₂) to the tissues and the rate of oxygen consumed [or oxygen uptake (VO₂)] by the tissues. Oxygen delivery is the volume of oxygen delivered per minute to the systemic vascular bed and is the product of CO and arterial oxygen concentration (CaO₂). Oxygen uptake is the amount of oxygen that diffuses from capillaries to the mitochondria. Tissue oxygenation is adequate when tissues receive sufficient oxygen to meet their metabolic needs. When the tissues do not receive enough oxygen, cellular injury could potentially occur. Oxygen uptake (VO₂) remains independent of DO₂ over a wide range of values, because oxygen extraction (O₂ER), which is the ratio of VO₂ over DO₂, can readily adapt to the changes in DO₂ until DO₂ falls below a critically low threshold (DO₂crit). An abrupt increase in blood lactate concentrations then occurs, indicating the development of anaerobic metabolism.

Sepsis and septic shock are characterized by peripheral vasodilation associated with excessive release of pro-inflammatory mediators resulting in a decrease in systemic vascular resistance, decreased effective intravascular volume, and tissue hypoperfusion. In the presence of inflammatory markers, oxygen extraction capabilities are altered. In these conditions, VO₂ can become dependent on DO₂. In addition, sepsis causes a generalized response that is overexpressed by the host in case of infection. In the recognition of bacterial products such as endotoxins and exotoxins, the immune system activates a cascade of pro-inflammatory mediators (e.g. cytokines by macrophages), recruitment of inflammatory cells, and complement activation.^8,9 These events lead to widespread cellular injury with ischemia, mitochondrial dysfunction, apoptosis, immunosuppression, organ dysfunction, and death.

RISK FACTORS AND PREDISPOSING FACTORS FOR MATERNAL SEPSIS

Between 2005 and 2008, the Center for Maternal and Child Inquiries (CMACE) identified risk factors for the development of maternal sepsis, including: obesity, carbohydrate intolerance or diabetes, alterations of immunity or immunosuppressive medication, anemia, history of pelvic infection, amniocentesis and other invasive procedures, cervical cerclage, prolonged spontaneous rupture of membranes, vaginal trauma, caesareancesarean section, abortion, Streptococcus agalactiae infection, black race or ethnic group minority, and infection in close or family contacts.¹⁰

Regarding the microorganisms linked to maternal mortality due to sepsis, the most common are beta-hemolytic Streptococcus and Escherichia coli, and mixed infections with both Gram-positive and Gram-negative organisms are common, especially in cases of chorioamnionitis.^11,12 Several women who died had one or more risk factors, with beingurinary tract infection, respiratory infection, and chorioamnionitis being the most common infections associated with septic shock.¹³ A prospective case-control study¹⁴ including 365 confirmed cases of severe maternal sepsis and 757 controls from all UK obstetrician-led maternity units demonstrated that, for most women with severe sepsis, there were less than 24 hours between the first sign of systemic inflammatory response syndrome (SIRS) and the diagnosis of severe sepsis, and for most women with a group A Streptococcus infection there were less than 9 h between the first sign of SIRS and severe sepsis, with approximately half of them having less than 2 h between the first signs of sepsis and its diagnosis.

The national cohort study with pregnant or recently pregnant women who had severe sepsis within the first 24 h of infection following critical care admission in the UK,¹⁴ evaluated the risk factors for severe sepsis morbidity on a national population level. The risk factors for maternal sepsis morbidity supported by this study are: young (maternal age < 20 years: rRelative risk (RR) 2.5, IC 95% 1.9 to 3.3) and old age (maternal age > 40 years: RR 1.8, IC 95% 1.2 to 2.6),¹⁴ multiple gestation (RR 4.4, IC 95% 3.1 to 6.3) and caesareancesarean section (RR 6.2, IC 95% 4.9 to 7.8). In this study, there was a clear trend in the risk of severe sepsis associated with decreasing socioeconomic status. This represents a striking example of health inequities within a high-income country.²

DEFINITION OF MATERNAL SEPSIS

The incidence of sepsis in pregnancy is difficult to determine since the diagnosis is not always made due to lack of medical suspicion and standardization in the diagnostic criteria. Since 1992, several strategies have been proposed to make its diagnosis universal and optimize response time, in order to attempt minimizing the damage and complications associated with this deregulated response.² Periodically, experts have reviewed such definitions but despite global efforts to generate advances in the diagnosis and management of sepsis, the mortality associated with sepsis has not changed.

In 1991 the American College of Chest Physicians (ACCP) and the Society of Critical Care Medicine (SCCM) established definitions for the spectrum of sepsis disease.¹⁵ The terms SIRS, sepsis, severe sepsis, septic shock, and multiple organ dysfunction syndrome began to be used in clinical practice. Sepsis was defined as the identification of two or more SIRS criteria, in addition to known or suspected infection (Table 1), while severe sepsis was defined as clinical sepsis accompanied by organ dysfunction, hypoperfusion, or hypotension. Septic shock is defined as a clinical condition in which fluid/vasopressor-resistant hypotension (mean arterial blood pressure ≤70 mmHg) and hypoperfusion is observed. In 2001, the SCCM, the European Society of Intensive Care Medicine (ESICM), the ACCP, the American Thoracic Society and the Surgical Infection Society held the second consensus meeting and updated the criteria for sepsis. The definitions of sepsis, severe sepsis, and septic shock were confirmed at the consensus.¹⁶

The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) was organized in 2016 because there were limitations in the previous definitions, with excessive focus around the concept of inflammation without having a continuum model from sepsis to shock, and with an inadequate sensitivity and specificity reported for SIRS diagnostic criteria.² A taskforce with expertise in sepsis pathobiology, clinical trials, and epidemiology was convened by the SCCM and the ESICM. Definitions and clinical criteria were generated through meetings, Delphi processes, analysis of electronic health record databases and voting, followed by circulation to international professional societies, requesting peer review and endorsement by 31 societies. This group generated the definitions for sSepsis and sSeptic sShock. According to this consensus, sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, and septic shock iwas defined as a sepsis accompanied by profound circulatory and cellular /metabolic abnormalities capable of substantially increasing mortality.

To identify maternal sepsis, a limitation has been established for a heterogeneous use of definitions as well as a diversity of recognition criteria in this population.¹⁷ This is why the WHO convened a consultation with experts on the field to analyze, formulate, and propose an update for the definition of maternal sepsis in order to be applied worldwide. With the information obtained from the bibliographic review and after consultation with specialists, the new definition of maternal sepsis reflects the concepts included in the definition of the consensus Sepsis-3 in adults. The proposed new definition of maternal sepsis is "a life-threatening condition defined as an organ dysfunction caused by an infection during pregnancy, delivery, puerperium, or after an abortion".¹⁸ This definition is useful for documenting confirmed cases of sepsis and allowing comparisons of the frequency of sepsis in different settings.

DIAGNOSIS CRITERIA

The Surviving Sepsis Campaign (SSC) is a joint collaboration of the SCCM and the ESICM committed to reduce the mortality and morbidity from sepsis and septic shock worldwide. The initial SSC guidelines were first published in 2004¹⁹ and revised in 2008,²⁰ 2012,²¹ and 2016.² The SSC criteria (Table 2) can be applied to the general population in the identification of serious cases, but they were not established for pregnant patients.²

In 2016, the task force of Sepsis-3 recognized that sepsis is a syndrome without, at present, a validated criterion standard diagnostic test. This group determined that there was an important need for features that can be identified and measured in individual patients and sought to provide such criteria to offer uniformity. Ideally, these clinical criteria should identify all the elements of sepsis (infection, host response, and organ dysfunction). The use of two or more SIRS diagnostic criteria to identify sepsis was unanimously considered unhelpful by the task force. Changes in white blood cell count, temperature, and heart rate reflect inflammation and the SIRS criteria do not necessarily indicate a dysregulated, life-threatening response.

The severity of organ dysfunction has been assessed with various scoring systems that quantify abnormalities according to clinical findings, laboratory data, or therapeutic interventions. Differences in those scoring systems have also led to inconsistency in the report of sepsis. The most common score used was the Sequential Organ Failure Assessment (SOFA) (originally the Sepsis-related Organ Failure Assessment) (Table 3). A high SOFA score is associated with an increased probability of mortality. The SOFA criteria evaluate platelet, liver, respiratory, kidney, cardiovascular, and central nervous system functions with scores ranging each from 0 (normal) to 4 (critical dysfunction).^11,6

To establish the diagnostic criteria for sepsis and validate the use of a scoring system, electronic health record data of 1.3 million encounters at 12 community and academic hospitals within the University of Pittsburgh Medical Center health system in southwestern Pennsylvania were studied. After this review, the task force of Sepsis-3 recommended using a change in baseline of the total SOFA score of ³≥2 points to represent organ dysfunction. The baseline SOFA score should be assumed to be 0 unless the patient is known to have preexisting (acute or chronic) organ dysfunction before the onset of infection. Patients with a SOFA score of ³≥2 had an overall mortality risk of approximately 10% in a general hospital population with presumed infection. The SOFA score is not intended to be used as a tool for patient management, but as a means to clinically characterize a septic patient. Components of SOFA require laboratory testing (such as creatinine or bilirubin level) and, thus, may not promptly capture dysfunction in individual organ systems.

A clinical model developed with multivariable logistic regression identified that two of any of the following three clinical variables – Glasgow Coma Scale score of 13 or less, systolic blood pressure of 100 mm Hg or less, and respiratory rate 22/min or greater – offered a predictive validity (AUROC = 0.81; 95% CI, 0.80–0.82) similar to that of the full SOFA score outside the ICU. This measure, termed qSOFA (for quick SOFA), provides simple bedside criteria to identify adult patients with suspected infection who are likely to have poor outcomes. Although the qSOFA is less robust than a SOFA score ³≥2 in the ICU, it does not require laboratory tests and can be assessed quickly and repeatedly. The task force suggested that the qSOFA criteria should be used to prompt clinicians to further investigate for organ dysfunction in order to initiate or escalate therapy as appropriate, and to consider increasinge the frequency of monitoring or referring to critical care if such actions have not been already undertaken.

The term severe sepsis is considered redundant, so it disappeared in the Third Consensus, and septic shock was now defined as the presence of sepsis with profound circulatory, cellular, and metabolic abnormalities that are associated with high mortality. Clinically it is identified by the requirement of vasopressors to maintain a mean blood pressure greater than 65 mmHg and/or the presence of serum lactate greater than 2 mmol/L (>18 mg/dL) in the absence of hypovolemia. This combination is associated with a mortality rate greater than 40%.^2,11 The task force recognized that serum lactate measurements are commonly, but not universally, available, especially in developing countries. Nonetheless, clinical criteria for septic shock were developed with hypotension and hyperlactatemia rather than either alone, because the combination encompasses both cellular dysfunction and cardiovascular compromise that is associated with a significantly higher risk-adjusted mortality rate.

A retrospective observational descriptive study conducted in a fourth-level clinic in Colombia compared SIRS diagnostic criteria and SOFA among pregnant women who met the inclusion criteria (two SIRS criteria plus infection). Patients with systemic compromise were admitted to the high-complexity obstetric unit (HCOU), whereas those with multiorgan dysfunction were admitted to the ICU. The study included 688 patients with diagnosis of maternal sepsis according of two SIRS criteria in the presence of infection. The four diagnostic criteria for sepsis according to SIRS at admission was positive among 431 patients (63%); 279 (65%) in the HCOU group and 152 (35%) in the ICU group. The SOFA test at admission was positive in 69 (39%) of 179 patients with complete data. The concordance – measured using the κ statistic – between SIRS and SOFA was low (0.016). According to this study, the SIRS sepsis criteria have greater sensitivity than SOFA for the diagnosis of sepsis in pregnant women, which could aid identification of patients requiring admission to the HCOU or ICU for early initiation of clinical management strategies. This is very important given the high mortality rate associated with sepsis in the obstetric population. The use of the SOFA scale to diagnose maternal sepsis requires additional analysis with a clinical and public-health approach, especially among low-income and middle-income countries where the availability of laboratory tests can be limited. In this study, only 39% of the cohort with the required data for calculation had a SOFA score ³≥2. This scale is more specific and showed a more accurate correlation with the severity of the case than SIRS among patients with severe clinical dysfunction, especially those admitted to the ICU. In other words, a SOFA score ³≥2 in an obstetric patient definitely identifies a severely ill patient with a higher probability of extreme maternal morbidity and death. From a clinical perspective, this population comprised patients who were truly ill and for whom investing clinical resources is mandatory for improving their survival. However, owing to the diagnostic difficulties mentioned above and the importance of early management of pregnant women with sepsis, delaying the start of treatment for obstetric patients with positive SOFA could have detrimental results.²³ The identification of patients with sepsis under non-critical conditions through clinical systems with clear reproducibility in low- and middle-income countries may be the key to reducing mortality in maternal sepsis.

In order to help health professionals, several tools have been designed, including clinical, laboratory, and treatment indicators (early warning systems) to identify septic pregnant women at risk for complications.²⁴ These tools employ different variables and thresholds to foresee the need for specialized attention or to predict mortality. However, these tools do not work well to predict the risk of maternal sepsis or to identify women who may need early treatment or intensive care because of an infection.^25,26 In addition, the lack of information regarding the validation and normalization of data from current or puerperal women limits the possibility of using these tools, particularly in low-resource settings.^25,26 Therefore, applicable criteria are urgently needed to identify "possible severe cases of maternal infection" (presumed maternal sepsis) with sufficient anticipation in the clinical evolution to allow timely treatment and better outcomes. Clinical criteria are also needed to confirm maternal sepsis (an infection with organ dysfunction).

In order to establish criteria to identify women with possible serious maternal infections (presumed maternal sepsis) and confirmed cases of maternal sepsis, the Global Maternal Sepsis Study (GLOSS) was undertook in order to establish and validate diagnostic criteria for possible severe cases of maternal infection, and to evaluate the frequency of recommended essential practices for the prevention, early identification, and management of maternal sepsis. This may contribute to the understanding of mother-–child transmission of bacterial infections, to evaluate the level of awareness of maternal and neonatal sepsis among health workers, and to establish a network of health institutions to implement strategies aimed at improving the quality of care for better identification and management of early maternal and neonatal sepsis.³ The results of GLOSS will be published this year.

EARLY WARNING SYSTEMS

In the general population, the use of early warning systems has been established for the detection of the degree of illness during the first 24 hours of hospital admission, which favors a rapid redirection of patients to the respective services and ensures early interventions, trying to rapidly define whether or not they require admission to the ICU.²⁷ The rapid identification and early guided therapy of critically ill patients have shown better outcomes in terms of mortality if they are initiated from the emergency department.^28,29

The physiological changes of pregnancy can complicate the early identification of sepsis in pregnant women; in addition, some of the diagnostic criteria of SIRS raised in the previous definition can be difficult to identify during pregnancy and puerperium, causing a delay in the diagnosis of obstetric sepsis, increasing not only maternal but also perinatal mortality.^30,31 In a systematic review conducted by Bauer et al.³², the authors tried to define the normal ranges for pregnancy and puerperium of each of the SIRS criteria in 8,834 healthy pregnant patients. The study showed that all current SIRS criteria are present in healthy pregnant women during the second and third trimesters of pregnancy, as well as during labor and puerperium except for high temperature, suggesting that patients with a temperature >38°C persistent for more than 1 hour require complementary studies.^31,32,33

The aim of these scoring systems, specifically designed for groups of obstetric patients receiving non-ICU care, is to reduce maternal morbidity and mortality. While some of these newly developed maternal warning systems are specific for particular maternal conditions (sepsis, eclampsia, venous thromboembolism),³⁴ others are more comprehensive.^35,36 Most of these maternal early warning scores have been developed retrospectively and need to be implemented prospectively.^37,38 Table 4 summarizes a comparison of some of these obstetric emergency classification scales.

Another criterion that has been studied in cases of sepsis is the Shock Index (SI). The calculation of the simple ratio between heart rate and systolic blood pressure generates a value that provides a rapid diagnosis and response to hypovolemic, cardiogenic, distributive, and obstructive shock events. Values between 0.5 and 0.7 normally represent patients outside of shock risk in the general population. However, due to the physiological hyperdynamic state of pregnancy, it could be common to have a SI higher than the established normal value. To date it is still unclear whether or not the SI can be an adequate tool for predicting the severity of the disease.³⁹ In a prospective study,⁴⁰ the SI and conventional vital signs were compared to identify acute critical illness in the emergency department. In total, 275Two hundred and seventy-five consecutive adults who showed up to receive urgent medical attention were included. Upon arrival at the emergency department patients had vital signs, SI, and priority of triage recorded. Two groups were identified according to the value of the SI: group one (41 patients) had an SI of more than 0.9, and group two (234 patients) had an SI of less than 0.9. Although both groups had apparently stable vital signs on admission, group one1 had a significantly greater proportion of patients who were assigned to a priority that required immediate treatment (23 vs. 45; p < 0.01), who required hospital admission (35 versus. 105; p < 0.01) and admission to continuous therapy in the ICU (10 vs. 13; p < 0.01) than patients with an SI < 0.9. Therefore, it was concluded that even with apparently stable vital signs, an abnormal elevation of the SI to more than 0.9 was associated with a condition that needed to be treated immediately, with management including the need for admission to ICU from the beginning.⁴⁰

Recently, normal SI values during pregnancy have been established. The variation of the cut-off values of the SI, as gestation progresses, comes in 0.756 (± 0.127) in pregnant women with 12 or less weeks of gestation, 0.795 (± 0.132) from 13 to 20 weeks, 0.825 (± 0.149) from 21 to 27 weeks, 0.831 (± 0.144) between 28 and 32 weeks, 0.821 (± 0.140) from 33 to 36 weeks, and 0.790 (± 0.139) above 37 weeks.⁴¹ However, it is not clear if the SI is an adequate tool for predicting the severity of sepsis in pregnant women. The SI was already studied as a form of rapid action to shock and septic shock widely in areas such as pPediatrics and iIntensive mMedicine obtaining promising results in the composition of a tool for rapid response and as a predictor of severity.^42,43 However, the search for diagnostic methods to trigger faster and more efficient responses in potentially severe cases of maternal sepsis is necessary due to the high morbidity and mortality related to the disease, which is most of the time treatable and curable. The SI has been studied and disseminated in the general clinic, but is still little addressed in obstetrics, and can be a marker of severity for the pregnant woman who arrives at emergency units with sepsis.

CONCLUSIONS

Maternal sepsis is an entity associated with high mortality, which is very difficult to diagnose due to the physiological changes that occur during pregnancy, delaying the initiation of determinant interventions for the reduction of mortality. One of the main challenges is to create and validate tools for evaluation and rapid response in obstetric emergencies. The incorporation into clinical practice of clear and standardized diagnostic criteria of maternal sepsis and septic shock, as well as early warning systems, can be highly effective strategies to reduce the impact of this entity on maternal health worldwide.

PRACTICE RECOMMENDATIONS