Infections during pregnancy are relatively prevalent, and the majority of cases are managed well in the community. Occasionally, however, infections may be life-threatening. Sepsis may be associated with multiple organ dysfunction and a high mortality. The treatment of sepsis is time critical and requires early fluid resuscitation and antibiotics. Early involvement of other specialties and allied health-care professionals to provide a multidisciplinary approach to patient care is important. Continuous monitoring of maternal vital signs and provision of supportive care for multiple organ dysfunction are best done within the intensive care unit. Despite advances in patient care, the mortality rate associated with maternal sepsis remains high. Health-care services in low-income countries face particular problems that account for an increased incidence of puerperal sepsis and maternal mortality. These include lack of access to health care, septic abortions and a greater incidence of human immunodeficiency virus. The key to management of sepsis is early recognition, aggressive resuscitation, antibiotic administration and source control.
Introduction
Over five million new cases per year of maternal sepsis occur globally with an estimated 62,000 maternal deaths resulting . In high-income countries, the incidence of sepsis-related maternal morbidity is reported to be 0.1–0.6 per 1000 deliveries and accounts for 2.1% of all maternal deaths. However, in low-income countries, sepsis-related mortality rate is significantly higher and accounts for up to 11.6% of maternal deaths . Although the incidence of maternal sepsis is relatively low compared to other obstetric emergencies, the relative risk of mortality is significant. The mortality rate associated with maternal sepsis approaches 10% in developed countries . There has been an increase in deaths related to genital tract sepsis, despite a decline in the overall UK maternal mortality rate. This is related primarily to community-acquired Group A streptococcal disease. Sepsis is now the most common cause of direct maternal death in the UK, with an increase in the maternal mortality rate from 0.85 deaths per 100,000 maternities in 2003–05 to 1.13 deaths in 2006–08 . A similar rise in the rates of maternal sepsis has been described in a recent large study in Maryland, USA . Maternal deaths relating to sepsis are often associated with failure to recognise the severity of illness .
In low-income countries, the mortality related to puerperal sepsis is approximately 33% . Deaths due to sepsis are 2–2.7-fold higher in Africa, Asia, Latin America and the Caribbean than in developed countries . The long-term morbidity associated with maternal sepsis is significant and includes chronic pelvic inflammatory disease, chronic pelvic pain, bilateral tubal occlusion and infertility. The impact on neonatal mortality is also significant, with over one million infection-related neonatal deaths every year .
‘Sepsis’, as defined by the International Surviving Sepsis Campaign, is the presence of infection (either suspected or proven) along with features of systemic inflammation . These features are characterised by criteria defining the systemic inflammatory response syndrome (‘SIRS’) . However, many physiological changes associated with pregnancy, such as tachycardia and hypotension (Chapter 1), overlap with sepsis. Cut-off values for cardiorespiratory variables, including blood pressure and respiratory rate, for the definition of sepsis may not be directly applicable in the parturient. Nevertheless, any physiological value more than two standard deviations from the expected should be considered pathological. ‘Sepsis-induced hypotension’ is defined as a systolic blood pressure (SBP) <90 mmHg or mean arterial pressure (MAP) <70 mmHg or a SBP decrease >40 mmHg or more than two standard deviations below normal for age in the absence of other causes of hypotension. ‘Septic shock’ is defined as sepsis-induced hypotension persisting despite adequate fluid resuscitation. When accompanied by signs of infection-induced hypotension, organ dysfunction or tissue hypoperfusion (e.g., lactataemia or oliguria), ‘severe sepsis’ is said to have occurred.
While the ‘SIRS’ criteria are a sensitive but nonspecific criteria of generalised inflammation, they are an oversimplification of the physiological changes associated with sepsis. As such, criteria for the systemic manifestations of infection have been elaborated further ( Table 1 ). It must be emphasised that these diagnostic criteria do not specifically address maternal sepsis.
| Variable | Infection (documented or suspected) and ‘some’ of the following: |
|---|---|
| General | Fever (>38.3 °C) or hypothermia (core temperature <36 °C) Heart rate >90 beats/min Tachypnea Altered mental status Significant oedema or positive fluid balance (>20 ml/kg over 24 hrs) Hyperglycaemia (plasma glucose >7.7 mmol/L) in the absence of diabetes |
| Inflammatory | Leukocytosis (WBC count >12,000/ml) Leukopenia (WBC <4000/mL) Normal WBC count with greater than 10% immature forms Plasma C-reactive protein >2 SD above the normal value Plasma procalcitonin >2 SD above the normal value |
| Haemodynamic | Arterial hypotension (SBP <90 mmHg, MAP <70 mmHg, or an SBP decrease >40 mmHg in adults or more than 2 SD below normal for age |
| Organ dysfunction | Arterial hypoxaemia (PaO2/FIO2 < 300) Acute oliguria (urine output <0.5 mL/kg/hr for at least 2 hrs despite adequate fluid resuscitation) Creatinine rise >0.5 mg/dL or 44.2micromol/L Coagulation abnormalities (INR >1.5 or aPTT >60 s) Ileus (absent bowel sounds) Thrombocytopenia (platelet count <100,000/mL) Hyperbilirubinaemia (plasma total bilirubin > 4 mg/dl or 70 mmol/L) |
| Tissue hypoperfusion | Hyperlactaemia >1 mmol/L Decreased capillary refill or mottling |
Aetiology and presentation
Infectious complications in pregnancy may be divided into the following:
- 1.
Pregnancy-related infection
The causes of maternal sepsis vary in relation to the stage of pregnancy . Sepsis in early pregnancy may be associated with a miscarriage or follow a termination of pregnancy. In the second or third trimester, pre-labour rupture of membranes (PROM) is associated with an increased risk of chorioamnionitis. Perineal infections, endometritis, wound infections and mastitis should be considered in the post-delivery period.
- 2.
Non-pregnancy-related infection
Most women of childbearing age are healthy with no chronic co-morbidity. However, a small proportion have predisposing factors, including human immunodeficiency virus (HIV), co-morbid illness (e.g., cystic fibrosis), or may be on medications (e.g., steroids or other immunosuppressants) that may predispose to infection.
- 3.
Nosocomial infection
Prolonged hospital stay, indwelling lines or catheters and overcrowding may predispose to hospital-acquired infections.
Fever, diarrhoea, vomiting, abdominal pain, generalised maculopapular rash (staphylococcal or streptococcal sepsis), offensive vaginal discharge and visible evidence of infection in caesarean wounds are the common symptoms of puerperal sepsis .
In severe cases of puerperal pyrexia, group A beta-haemolytic streptococcus ( Streptococcus pyogenes , GAS) should be suspected. This organism has been the leading infective cause of puerperal deaths and has an attributable mortality greater than other invasive bacteria . Streptococcal throat infections are relatively common in the community, particularly among young children. Up to 30% of the population are carriers of GAS and are easily able to transmit the bacteria via a droplet spread. Precautions against hand to genital tract transmission should be taken, including hand washing before and after using the lavatory and changing sanitary towels. This is particularly important when the mother has had a recent close contact with someone with a sore throat or upper respiratory tract infection . GAS causes a wide spectrum of illness ranging from bacteraemia without a focus of infection (46%) to endometritis (28%) and peritonitis (8%). ‘Invasive GAS’ is associated with necrotising fasciitis (3%) and toxic shock syndrome (3%) . Though less common than other clinical manifestations of GAS, invasive GAS is associated with a mortality of 3.5–14.3% and is the leading cause of sepsis-related deaths in the UK .
The single most important risk factor for post-partum infection is caesarean section . Retained products of conception may also become embedded, confounded by poor wound healing due to tissue oedema, resulting in endometritis. Endometritis is most commonly associated with GAS infection, though Staphylococcus, coliforms and anaerobes may also be implicated. Patients often present with uterine tenderness, abdominal pain, purulent foul-smelling lochia and features of systemic infection. Cervical and high vaginal swabs must be obtained for culture. The rate of endometritis is approximately threefold higher in non-elective caesarean sections compared to elective sections (28.6% vs. 9.2%, respectively). Similarly, the risk of major puerperal infection is threefold higher in a low-risk planned caesarean delivery compared to a planned vaginal delivery at term . The use of prophylactic antibiotics prior to section is associated with a significant reduction in the rates of post-partum fever, wound infection and endometritis .
Ascending bacterial colonisation of the genital tract may result in uterine contractions and/or membrane weakening that in turn results in pre-term pre-labour rupture of membranes (PPROM). This may result in chorioamnionitis, which typically presents with abdominal tenderness and fever following PPROM. The use of prophylactic antibiotics is associated with prolongation of pregnancy, fewer neonatal infections and a reduction in rates of chorioamnionitis . There are concerns regarding the use of co-amoxiclav and its association with a relative increase in the risk of neonatal necrotising enterocolitis. It is therefore recommended that a 10-day course of erythromycin is administered following PPROM. If group B streptococcus is isolated, penicillin should be administered (or clindamycin to women who are allergic to penicillin) . A long-term follow-up of children demonstrates that the use of maternal antibiotics is not associated with adverse outcomes .
Prolonged use of antibiotics for PPROM may result in antibiotic resistance among commensal microorganisms, particularly group B streptococci . The selective use of antibiotics in patients with evidence of sub-clinical infection based on amniotic fluid sampling has been argued. However, not all centres may have the facilities to perform this, and there is no compelling evidence to start antibiotics guided by amniocentesis .
Other risk factors for maternal sepsis include maternal anaemia, obesity, poor nutrition, induced labour, prolonged labour (>12 hr) and frequent (>5) vaginal examination .
Aetiology and presentation
Infectious complications in pregnancy may be divided into the following:
- 1.
Pregnancy-related infection
The causes of maternal sepsis vary in relation to the stage of pregnancy . Sepsis in early pregnancy may be associated with a miscarriage or follow a termination of pregnancy. In the second or third trimester, pre-labour rupture of membranes (PROM) is associated with an increased risk of chorioamnionitis. Perineal infections, endometritis, wound infections and mastitis should be considered in the post-delivery period.
- 2.
Non-pregnancy-related infection
Most women of childbearing age are healthy with no chronic co-morbidity. However, a small proportion have predisposing factors, including human immunodeficiency virus (HIV), co-morbid illness (e.g., cystic fibrosis), or may be on medications (e.g., steroids or other immunosuppressants) that may predispose to infection.
- 3.
Nosocomial infection
Prolonged hospital stay, indwelling lines or catheters and overcrowding may predispose to hospital-acquired infections.
Fever, diarrhoea, vomiting, abdominal pain, generalised maculopapular rash (staphylococcal or streptococcal sepsis), offensive vaginal discharge and visible evidence of infection in caesarean wounds are the common symptoms of puerperal sepsis .
In severe cases of puerperal pyrexia, group A beta-haemolytic streptococcus ( Streptococcus pyogenes , GAS) should be suspected. This organism has been the leading infective cause of puerperal deaths and has an attributable mortality greater than other invasive bacteria . Streptococcal throat infections are relatively common in the community, particularly among young children. Up to 30% of the population are carriers of GAS and are easily able to transmit the bacteria via a droplet spread. Precautions against hand to genital tract transmission should be taken, including hand washing before and after using the lavatory and changing sanitary towels. This is particularly important when the mother has had a recent close contact with someone with a sore throat or upper respiratory tract infection . GAS causes a wide spectrum of illness ranging from bacteraemia without a focus of infection (46%) to endometritis (28%) and peritonitis (8%). ‘Invasive GAS’ is associated with necrotising fasciitis (3%) and toxic shock syndrome (3%) . Though less common than other clinical manifestations of GAS, invasive GAS is associated with a mortality of 3.5–14.3% and is the leading cause of sepsis-related deaths in the UK .
The single most important risk factor for post-partum infection is caesarean section . Retained products of conception may also become embedded, confounded by poor wound healing due to tissue oedema, resulting in endometritis. Endometritis is most commonly associated with GAS infection, though Staphylococcus, coliforms and anaerobes may also be implicated. Patients often present with uterine tenderness, abdominal pain, purulent foul-smelling lochia and features of systemic infection. Cervical and high vaginal swabs must be obtained for culture. The rate of endometritis is approximately threefold higher in non-elective caesarean sections compared to elective sections (28.6% vs. 9.2%, respectively). Similarly, the risk of major puerperal infection is threefold higher in a low-risk planned caesarean delivery compared to a planned vaginal delivery at term . The use of prophylactic antibiotics prior to section is associated with a significant reduction in the rates of post-partum fever, wound infection and endometritis .
Ascending bacterial colonisation of the genital tract may result in uterine contractions and/or membrane weakening that in turn results in pre-term pre-labour rupture of membranes (PPROM). This may result in chorioamnionitis, which typically presents with abdominal tenderness and fever following PPROM. The use of prophylactic antibiotics is associated with prolongation of pregnancy, fewer neonatal infections and a reduction in rates of chorioamnionitis . There are concerns regarding the use of co-amoxiclav and its association with a relative increase in the risk of neonatal necrotising enterocolitis. It is therefore recommended that a 10-day course of erythromycin is administered following PPROM. If group B streptococcus is isolated, penicillin should be administered (or clindamycin to women who are allergic to penicillin) . A long-term follow-up of children demonstrates that the use of maternal antibiotics is not associated with adverse outcomes .
Prolonged use of antibiotics for PPROM may result in antibiotic resistance among commensal microorganisms, particularly group B streptococci . The selective use of antibiotics in patients with evidence of sub-clinical infection based on amniotic fluid sampling has been argued. However, not all centres may have the facilities to perform this, and there is no compelling evidence to start antibiotics guided by amniocentesis .
Other risk factors for maternal sepsis include maternal anaemia, obesity, poor nutrition, induced labour, prolonged labour (>12 hr) and frequent (>5) vaginal examination .
Management
The Surviving Sepsis Campaign recommends protocolised ‘care bundles’ in the management of severe sepsis and septic shock . The protocolised care of patients with severe sepsis and septic shock is associated with a significant decrease in mortality ; however, it is unclear whether this is directly related to the protocol per se or simply due to earlier recognition and intervention. The management of the patient is divided into the initial resuscitation phase, antimicrobial therapy (including obtaining blood cultures and source control) and subsequent supportive therapies. However, it must be emphasised that these management guidelines do not specifically address maternal sepsis. The emphasis of these guidelines should be on the institution of early aggressive treatment and a multidisciplinary approach.
Resuscitation
The principle is to promptly restore adequate organ perfusion with fluid resuscitation and, if needed, drug support. One strategy currently being promoted is that of ‘early goal-directed therapy’ (EGDT), whereby a central venous oxygen saturation value >70% is targeted in addition to the standard cardiovascular markers such as blood pressure and urine output. This is based upon a single open, randomised emergency room study performed in Detroit, where significant mortality benefit was seen when this approach was applied in the first 6 h after presentation . Fluid loading and, in some cases, dobutamine and blood transfusion were given to reach the predefined haemodynamic goals. The validity and generalisability of this approach is currently being tested in three multicentre studies in Australasia, the USA and the UK. As mentioned above, the applicability of EGDT to the pregnant patient is also uncertain. A similar approach is likely to be less effective once organ dysfunction is established .
Tissue hypoxia may develop due to low blood flow and/or perfusion pressure. Fluid resuscitation should be initiated promptly in patients with haemodynamic instability or evidence suggestive of tissue hypoperfusion, such as an elevated lactate, to achieve an adequate oxygen delivery. Controversy rages as to the best type of fluid (colloid vs. crystalloid, type of colloid (e.g., albumin vs. gelatin vs. starch), balanced vs. unbalanced solutions, the optimal filling target, etc.). The literature is not particularly clear other than an excessive amount of starch-based solutions should be avoided as this has been associated with increased rates of acute kidney injury and, in some cases, higher mortality rates .
Consensus guidelines, based on a limited-existent evidence base, recommend targeting a mean arterial pressure of >65 mmHg . Whether this applies to young, pregnant women who, in health, are often completely asymptomatic at lower blood pressures is questionable. A reasonable compromise is to achieve a blood pressure compatible with adequate organ perfusion, as manifested by normal conscious level, good urine output, normal lactate level, etc. Foetal well-being also needs to be confirmed. Early delivery, if appropriate, may be indicated.
In critically ill patients, lactate is used as a surrogate marker of inadequate tissue oxygenation. Though this is not strictly accurate, a good relationship is nevertheless seen between the degree of hyperlactaemia and mortality . Lactate has therefore been promoted as a target in the goal-directed therapy for early sepsis. An EGDT guided by early lactate clearance (within the first 6 h after sepsis presentation) was associated with a reduction in 60-day mortality . As serum lactate does not change significantly in pregnancy, this is likely to be a reasonable guide to resuscitation.
The assessment of adequate fluid loading is more problematic in pregnant patients. The physiological effects of pregnancy, especially in the third trimester, including alterations in circulating blood volume, cardiac output, oxygen consumption, haemoglobin, vascular compliance, blood pressure and venous compression, form the gravid uterus. In severe sepsis/septic shock, there is a decreased vascular tone and decreased responsiveness to exogenous catecholamines, increased capillary leak and a high incidence of myocardial depression. Thus, a critically ill pregnant patient warrants close haemodynamic monitoring, preferably with a reliable measure of cardiac output, to titrate fluid and drugs more precisely. Arbitrary target values of central venous pressure (CVP), for example, recommended by the Surviving Sepsis Campaign guidelines as 8–12 mmHg in non-ventilated patients and >12 mmHg in mechanically ventilated patients, may not be applicable in the pregnant patient. A poor correlation between CVP values and fluid responsiveness is well recognised .
Fluid resuscitation should be titrated to achieve an adequate tissue perfusion. It is not possible to be prescriptive about the precise amount of fluid that should be administered. The volume of fluid required to achieve haemodynamic stability may vary significantly between patients. However, if the patient remains haemodynamically unstable or there is evidence of persistent organ hypoperfusion despite 20–30 ml kg −1 of intravenous fluid, a vasoactive therapy may be required. In such circumstances, haemodynamic monitoring may be useful to assess the patient’s response (in stroke volume) to a fluid challenge (‘functional haemodynamic monitoring’). A cumulative positive fluid balance over days is associated with a worse outcome in sepsis , and therefore over-zealous fluid resuscitation should also be avoided.
Resuscitation should aim to achieve adequate global oxygen delivery. Oxygen delivery (DO 2 ) depends on the haemoglobin concentration (Hb), arterial oxygen saturation (SaO 2 ) and cardiac output (CO):
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
