Sepsis in Pregnancy
Maternal sepsis, especially puerperal sepsis, is a common pregnancy-related condition; in the United States, it is a leading cause of maternal mortality, accounting for up to 28% of maternal deaths and up to 15% of maternal admissions to the intensive care unit (ICU; Bauer, Lorenz, Bauer, Rao, & Anderson, 2015; Chang et al., 2003; Oud, 2015a, 2015b; Pollock, Rose, & Dennis, 2010). More concerning is that sepsis has been increasingly reported as the cause of maternal death, rising by up to 10% per year between 2000 and 2010 (Bauer, Bateman, Bauer, Shanks, & Mhyre, 2013; Oud, 2015a). This is due, in part, to a greater than 200% increase in the incidence of pregnancy-associated severe sepsis over that same period (Oud & Watkins, 2015). One contributing and modifiable factor to these deaths is failure to recognize sepsis, leading to delays in treatment (Bauer et al., 2015; Cantwell et al., 2011). Therefore, rapid and accurate diagnosis and initial management of sepsis in pregnancy in the emergency department (ED) is paramount.
Women may present to the obstetric triage unit or ED at any stage of illness. In a review of maternal deaths due to sepsis, almost a third of the women died at home without seeking medical care, demonstrating that many women do not present to the hospital and implying that many others may present late in the course of the disease (Bauer et al., 2015). Additionally, because there are many possible underlying infectious causes of sepsis, each clinical presentation will be different. However, given that the underlying inflammatory and immune processes are similar in many infections, vital signs and laboratory derangements are often present regardless of the specific underlying process. In the study mentioned previously that reviewed maternal deaths due to sepsis, among patients who presented to the hospital with sepsis, 9 out of 12 demonstrated one or more of the following vital sign abnormalities: heart rate greater than 120 beats per minute, respiratory rate higher than 30 breaths per minute, systolic blood pressure below 90 mmHg, peripheral oxygen saturation of less than 95% on room air, or temperature greater than 38°C or lower than 36°C (Bauer et al., 2015).
In pregnancy, the most common causes of sepsis include endometritis and chorioamnionitis, urinary tract infections, and pneumonia; up to 60% of maternal sepsis may be classified as obstetric, meaning it originates from the 266genital tract (Kramer et al., 2009; Paruk, 2008). In contrast to sepsis that occurs outside of pregnancy, in which gram-positive bacteria account for approximately 50% of cases, and polymicrobial infections account for only 5%, infections that result in sepsis in pregnancy tend to be polymicrobial, reflecting the anatomic continuity with the vaginal flora (Martin, Mannino, Eaton, & Moss, 2003). However, there has been an increase in severe beta-hemolytic streptococci group A (GAS) infections in pregnancy, leading to increased morbidity and mortality (Kramer et al., 2009; Schuitemaker et al., 1998). Despite its rarity, and because management of sepsis caused by GAS must be aggressive and requires a specific management algorithm, identification or exclusion of this organism is a priority (Rimawi, Soper, & Eschenbach, 2012).
HISTORY AND DATA COLLECTION
A standard history needs to be taken for any woman presenting with signs or symptoms of an infection, with a focus on attempting to elucidate the source of infection. This must include the (a) timing of onset and severity of symptoms and (b) medications taken before presentation to the hospital. Exposures before the onset of symptoms may be contributory, including exposures to sick adults or children, exposures to mosquitos or insects, and food-borne illnesses. A recent travel history may also be instructive. In addition, a thorough medical and surgical history is critical in order to evaluate for risk factors, including prior infections, current treatment for infection, and immunosuppression (e.g., medication, malignancy).
Additional modifiable and nonmodifiable risk factors for sepsis and septic shock include non-White race, public insurance or no insurance, delivery at a low-volume hospital (<1,000 births per year), medical comorbidities such as diabetes and hypertension, and pregnancy-related complications such as preeclampsia and postpartum hemorrhage (Acosta et al., 2013; Mohamed-Ahmed, Nair, Acosta, Kurinczuk, & Knight, 2015). Although many of these conditions cannot be changed, increased vigilance in the setting of these risk factors is warranted.
The initial component of the physical examination is a complete set of vital signs because these are the main diagnostic criteria for sepsis. A comprehensive physical examination can then be performed expeditiously in the pregnant woman with suspected sepsis, focusing on the most common causes of sepsis in pregnancy. A thorough heart and lung examination assesses for cardiopulmonary infections such as pneumonia. An abdominal and back examination assesses for intra-abdominal infections such as cholecystitis, appendicitis, and pyelonephritis. A pelvic examination evaluates for chorioamnionitis or endometritis. Finally, a thorough skin inspection looks for insect or animal bites, cellulitis, and wound infection.
LABORATORY AND IMAGING STUDIES
Initial laboratory studies include a complete blood count with differential, a comprehensive metabolic panel including a blood lactate level, two sets of blood cultures, and a urinalysis and urine culture. Additional cultures of amniotic fluid, wound, or abscess can be collected if the clinical situation warrants. Imaging 267studies should focus on the likely source of infection and may include a chest x-ray to evaluate for pneumonia, a computed tomography (CT) scan to evaluate for appendicitis, or an ultrasound to evaluate for cholecystitis. In general, a renal ultrasound is not necessary in the initial management of pyelonephritis but may be considered in cases where the symptoms are refractory to standard medical management.
DIAGNOSIS OF SEPSIS
In 1992, the American College of Chest Physicians and the Society of Critical Care Medicine introduced definitions for the systemic inflammatory response syndrome (SIRS), sepsis, severe sepsis, and septic shock, and in 2001, an international group of critical care specialists met to solidify these definitions (Bone et al., 2009; Levy et al., 2003). The goal of the SIRS criteria was to describe a clinical response to a nonspecific insult of either infectious or noninfectious origin (Muckart & Bhagwanjee, 1997).
In 2016, the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) were released (Singer et al., 2016). The objective of this consensus was to evaluate and update the definitions for sepsis and septic shock. The consensus notes that the original conceptualization of sepsis as an infection with at least two of the four SIRS criteria focused solely on inflammatory excess. A new definition of sepsis was proposed as life-threatening organ dysfunction caused by a dysregulated host response to infection and proposed the use of the Sequential [Sepsis-related] Organ Failure Assessment (SOFA) score to quantify organ dysfunction. The SOFA score is composed of scores from six organ systems (central nervous, cardiovascular, respiratory, gastrointestinal, genitourinary, and coagulation), graded from 0 to 4 according to the degree of dysfunction/failure (Vincent et al., 1996). Organ dysfunction from sepsis, therefore, can be identified as an acute change in the total SOFA score of greater than or equal to 2 points consequent to the infection, with a baseline SOFA score that can be assumed to be zero.
Any nonspecific SIRS criteria will continue to aid in the general diagnosis of infection. However, SIRS can simply reflect an appropriate and adaptive host response, whereas sepsis indicates organ dysfunction plus the accompanying inflammatory response. The definition of septic shock remained overall unchanged and was defined as a subset of sepsis in which underlying circulatory and cellular metabolic abnormalities are profound enough to substantially increase mortality. The term severe sepsis was thought to be redundant and was removed from the nomenclature because sepsis itself warrants greater levels of monitoring and intervention. Table 23.1 lists the definitions for SIRS, sepsis, severe sepsis, and septic shock as defined in 2001 and 2016.
The diagnosis of sepsis in pregnancy can be difficult because there is considerable overlap between the SIRS criteria and the SOFA score and normal physiologic changes during pregnancy (Abbassi-Ghanavati, Greer, & Cunningham, 2009; Bauer et al., 2014; Clark et al., 1989; Colditz & Josey, 1970; Cunningham et al., 2013; Guinn, Abel, & Tomlinson, 2007). See Table 23.2 for how the pregnant state differs from the nonpregnant state.
Although patients with fever, leukocytosis, and hypotension are often considered septic unless another diagnosis is evident, it is necessary to consider a differential diagnosis. Several noninfectious conditions may mimic sepsis and are sometimes 268referred to as pseudosepsis. These conditions are listed in Table 23.3 along with differentiating features from sepsis. Discriminating sepsis from pseudosepsis can be difficult, as clinically both can present with fever, leukocytosis, and hypotension, and central monitoring may demonstrate increased cardiac output and decreased peripheral resistance. When the diagnosis is unclear, do not delay sepsis therapy in order to rule out causes of pseudosepsis because timely management of sepsis is among the most critical factors preventing mortality.
Presence of viable bacteria in the blood
Systemic inflammatory response defined by two or more of the following:
• Temperature >38°C or <36°C
• Heart rate >90 bpm
• Respiratory rate >20/min or PaCO2 <32 mmHg
• White blood cell count >12 or <4/mcL or >10% bandemia
• Not used for the diagnosis of sepsis
• Will continue to aid in the general diagnosis of infection
SIRS + source of infection
Life-threatening organ dysfunction caused by dysregulated host response to infection
• SOFA score ≥2
Sepsis + evidence of organ dysfunction, tissue hypoperfusion, or hypotension
No longer in use
Sepsis + hypotension despite adequate fluid resuscitation
Sepsis + persistent hypotension requiring vasopressors to maintain MAP ≥65 mmHg and serum lactate >2 mmol/L despite adequate volume resuscitation
bpm, beats per minute; MAP, mean arterial pressure; PaCO2, partial pressure of arterial carbon dioxide; SIRS, systemic inflammatory response syndrome; SOFA, Sequential [Sepsis-related] Organ Failure Assessment.
Sources: Definitions from Bone et al. (2009) and Singer et al. (2016).
CLINICAL MANAGEMENT AND FOLLOW-UP
Severity-of-Illness Scoring Systems
In order to enable early detection of cases of sepsis at risk for rapid clinical deterioration, many disease severity scoring systems related to sepsis have been developed and validated for the general population. Among them, the SOFA score consistently performs the best with regard to predictive value in critically ill obstetric patients. However, it has not been evaluated specifically in pregnant women presenting with sepsis (Jain, Guleria, Suneja, Vaid, & Ahuja, 2016; Kallur, Patil Bada, Reddy, Pandya, & Nirmalan, 2014; Oliveira-Neto, Parpinelli, Cecatti, Souza, & Sousa, 2012).
The Sepsis-3 consensus statement recommends use of the quick SOFA (qSOFA) score (range 0–3 points, with 1 point each for systolic hypotension [≤100 mmHg], tachypnea [≥22/min], or altered mentation) rather than 269the standard SIRS criteria for rapid diagnosis (Seymour et al., 2016; Singer et al., 2016). The studies on which these recommendations were based did not include pregnant women, and the qSOFA score has yet to be studied specifically in the obstetric population. In addition, many pregnant women will have a systolic blood pressure ≤100 mmHg and a respiratory rate ≥22/min upon presentation, but very few will have altered mentation. The score is not likely to be useful in the rapid evaluation of pregnant women given that only two of the three components are likely to be used.
CHANGE FROM NONPREGNANT STATE
Diastolic: Decreased by 5–10 mmHg in second trimester, return to normal by third trimester
Increased by 15%–20% (83 ± 10 beats per minute)
Central venous pressure
Nonpregnant: 9.0 cmH2O (7.8–11.2)
First trimester: 7.5 cmH2O (6.5–8.2)
Second trimester: 4.0 cmH2O (3.6–4.6)
Third trimester: 3.8 cmH2O (2.0–4.4)
Decreased to 0.3–0.8 mg/dL
Increased (GFR increased 50%)
5.7–16.9/mcL by third trimester (up to 30/mcL in labor)
% Immature neutrophils
PaO2, arterial oxygen partial pressure; FIO2, fractional inspired oxygen; GFR, glomerular filtration rate.
*Indicates value used for quick Sequential [Sepsis-Related] Organ Failure Assessment (qSOFA) score.
†Indicates value used for SOFA score.
Source: Reference values from Abbassi-Ghanavati et al. (2009), Bauer et al. (2014), Clark et al. (1989), Colditz and Josey (1970), Cunningham et al. (2013), and Guinn et al. (2007).
Clearly, because of the considerable overlap between the SIRS or qSOFA criteria and the normal physiologic parameters during pregnancy and the postpartum period, an efficient, accurate, and reliable predictor of sepsis-related morbidity and mortality is a goal not entirely realized. Therefore, identification and prompt treatment of sepsis in pregnancy will remain imprecise until either alternative criteria or a pregnancy-specific sepsis scoring system is developed. One such scoring system has been proposed, the Sepsis in Obstetrics score, but has not yet been validated (Albright, Ali, Lopes, Rouse, & Anderson, 2014b). Until there are pregnancy-specific algorithms for the diagnosis of sepsis, it is recommended to use the current general population algorithm as described by the Sepsis-3 consensus statement. This algorithm initially utilizes the quick Sequential (Sepsis-Related) Organ Failure Assessment (qSOFA) score to triage patients with a suspected infection followed by the SOFA score to make a diagnosis of sepsis. Sepsis plus the need for vasopressors to maintain a mean arterial pressure greater or equal to 65 mmHg, or a serum lactate greater than 2 mmol/L, gives a diagnosis of septic shock (Singer et al., 2016).
DISTINGUISHING CLINICAL FEATURES
DISTINGUISHING LABORATORY INVESTIGATIONS
Acute myocardial infarction
• Risk factors (may not be present in pregnant women)
• Typical chest pain on exertion
• Serum troponin
Acute pulmonary embolus
• Risk factors
• Personal or family history
• Sudden dramatic onset with little prodrome
• Hypoxia should be prominent
• CT angiogram
• Upper abdominal and/or back pain
• Nausea and vomiting
• History of alcohol, cholelithiasis, or hypertriglyceridemia
• Serum amylase
• Consider CT abdomen
Acute adrenal insufficiency
• History of steroid use in past year, especially if Cushingoid in appearance
• Personal or family history of autoimmune disease
• Hyperpigmentation, especially in any new scars
• Serum electrolytes (high potassium and low sodium)
• ACTH stimulation test
• Empiric stress dose steroid administration
• Usually hematemesis or melena but not always present initially
• History suspicious for PUD, gastritis, or varices
• CBC (although may not be reflective of acute status)
• Rectal examination for blood
• NG tube for gastric lavage
Acute intra-abdominal hemorrhage
• Usually sudden-onset abdominal pain with peritoneal signs
• Ultrasound or CT abdomen for free fluid
• Lesions can include ruptured splenic aneurysm, abdominal aortic aneurysm
Overzealous diuresis or extensive third spacing
• Clinical context (diuretic administration or surgery associated with massive fluid shifts)
• Review of fluid balance
• Trial of fluid replacement
• Clinical context
• Reactions can include antigen/antibody mismatch but also leukoagglutination in the lungs
• Laboratory investigations by blood bank
Adverse drug reactions
• Clinical context following a potentially precipitating exposure
• 60% of time, the precipitant is not identified
• Itching/burning/hives on skin present in ~90% of cases
• Treat with IM epinephrine 0.5–1.0 mg (0.5–1.0 mL of a 1:1,000 [1 mg/mL] solution), fluids, diphenhydramine, ranitidine, and steroids
• Confirm diagnosis with serial measurements of levels of histamine and tryptase A
Amniotic fluid embolism
• Clinical context: typically intrapartum
• Sudden onset
• Associated with hypoxia, hypotension, and bleeding
• DIC screen (INR, aPTT, fibrin degradation products, fibrinogen)
• Usually proceeds to full resuscitative efforts with intubation
Fat emboli syndrome
• Typically in the setting of trauma, fracture, TPN, or pancreatitis
• DIC screen
• Usually proceeds to full resuscitative efforts with intubation
• Steroids may be helpful