Cardiovascular disease

Cardiac disease is the leading cause of maternal death in the US (26.4%), with a cause-specific mortality ratio of 4.2 per 100,000 live births . Risk factors include pre-existing cardiovascular disease, obesity, substance misuse disorder, and African American race . Fifteen percent of the patients who present with cardiac disease in pregnancy have no pre-existing conditions . Timely recognition and response to symptoms of cardiopulmonary disease (e.g., shortness of breath, palpitations, chest pain, wheezing, or fatigue) was the single most important preventability factor identified in a recent review of deaths in California . Other management failures included insufficient use of antihypertensive medications and misdiagnosis of shortness of breath or wheezing as new-onset asthma or other respiratory illness.

Cardiomyopathy is the single largest subcategory of cardiac-related mortality (11.8%) and is therefore counted separately in surveillance reports . Maternal cardiomyopathy deaths are most frequently due to peripartum cardiomyopathy, other dilated cardiomyopathy, and hypertrophic heart disease . The majority of patients with peripartum cardiomyopathy present postpartum, 75% in the first month . The mortality rates range between 25% and 50%, and the majority of deaths occur within a few months of delivery. However, 30%–50% of patients will make a full recovery, and only 10% eventually require cardiac transplantation .

Congenital heart disease (CHD) accounts for more than half of all cardiac diseases in pregnancy. Advances in surgical and medical treatments now allow an increasing proportion of children to survive into adulthood . In the US, between 2000 and 2010, adult survivors of CHD increased by 30% (from 6 to 9 per 100,000 live births); the MMR was 178 per 100,000 hospitalizations for delivery, accounting for 2% of all in-hospital deaths . Adverse outcomes increased for those with concomitant pulmonary hypertension. Arrhythmias and heart failure are the most common complications during pregnancy, and 10% of women with CHD suffer a late cardiac event (beyond 6 months postpartum) .

Coronary artery disease is the leading cause of death in women in the US (one in every four female deaths) and an important etiology of maternal morbidity and mortality, given the increasing population prevalence of obesity, diabetes, and AMA. The physiologic stress of pregnancy can unmask previously silent disease processes such as valvular heart disease and ischemic conditions. The probability of myocardial infarction (MI) is increased two- to four-fold in pregnancy, and this risk increases six-fold in the immediate postpartum days compared with that in aged-matched nonpregnant women . Parturients aged >40 years have a 30-fold higher risk of MI than those aged <20 years ; moreover, the risk persists postpartum. Patients with a history of preeclampsia have a two-fold increase in the incidence of ischemic heart disease, CVA, and VTE 5–10 years postdelivery .

Multidisciplinary care for women with cardiac disease should address anticoagulation therapy, rhythm monitoring and management (e.g., pacemaker/implantable cardioverter defibrillator), delivery planning (e.g., route and timing of delivery, anesthetic technique), and prophylactic therapies to ensure optimal maternal and fetal outcomes. Elevated serum brain natriuretic peptide >100 pg/mL is abnormal in pregnancy but may be seen in women with pre-existing cardiac disease for whom serial measurements may help identify cardiac decompensation . Similarly, serial transthoracic echocardiography examinations may be used to evaluate cardiac output, right- and left-sided heart function, wall motion, and valvular function throughout pregnancy. The accuracy of echocardiography compares favorably with pulmonary arterial catheterization; moreover, echocardiography is noninvasive . In combination with symptom progression and fetal surveillance, the information can direct optimal delivery planning, and obstetric, anesthetic, and medical management.

Sepsis

Infection is the leading direct cause of maternal mortality in the US (2.2 per 100,000 live births), and genital tract sepsis is the second leading cause of direct maternal deaths in the UK . Pregnancy-associated severe sepsis (PASS) increased by 236% (11–26 per 100,000 estimated pregnancies) from 2001 to 2010 . Chronic liver disease, congestive heart failure in the Black race, poverty, substance misuse, and lack of health insurance increase the risk for PASS . Pneumonia/respiratory infection and genital infection are the leading causes of sepsis, followed by urinary and respiratory infection; gram-negative bacteria are the most common organism . Interestingly, 37.5% episodes of PASS occurred during hospitalization for delivery, and 24.5% and 24.1% occurred during antepartum and postpartum hospitalizations, respectively .

Traditionally, sepsis was defined as a combination of infection and inflammation, which was recognized by documenting the systemic inflammatory response syndrome (SIRS) criteria. SIRS is a problematic concept in obstetrics because the parameters overlap with normal vital signs in pregnancy . SIRS is also problematic in nonobstetric settings, given its poor correlation with severe morbidity or mortality. Some of the most critically ill patients never mount an inflammatory response. Consequently, in 2015, the Society of Critical Care Medicine and the European Society of Intensive Care Medicine published Sepsis-3, the Third International Consensus Definitions for Sepsis and Septic Shock . According to Sepsis-3, sepsis is defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection. Septic shock is a subset of sepsis in which underlying circulatory and cellular/metabolic abnormalities are profound enough to substantially increase mortality . The Sequential Organ Failure Assessment (SOFA) is the most sensitive tool to identify end-organ injury in adult patients with suspected infection. Alternatively, the quick SOFA (qSOFA) is more specific and includes three parameters that can be rapidly assessed: respiratory rate ≥22 per min, altered mentation, and systolic blood pressure (SBP) ≤ 100 mmHg . Moderate tachypnea may be normal in the peripartum period, but altered mental status should always be considered abnormal. The combination of qSOFA ≥2 and infection indicates sepsis, and the additional need for vasopressors to maintain end-organ perfusion in the presence of elevated lactate levels indicates septic shock. Critics of sepsis-3 contend that multiple definitions lead to inaccurate reporting, resulting in misleading incidence and mortality data. In addition, the new definitions rely on evidence of end-organ injury and increase the risk for diagnostic delay.

Early diagnosis, fluid resuscitation, and appropriate anti-infective therapy are the most important factors to ensure survival from sepsis. The Surviving Sepsis Campaign (SSC) originated in 2002 with the aim to reduce sepsis mortality by 25% over 5 years. The resuscitation and management bundles should be completed within the first 6 and 24 h, respectively. A multicentered international study assessing institutional compliance with SSC guidelines over a 2-year period found that compliance with the resuscitation bundle increased from 10.9% to 31.3% ( P < 0.001), compliance with the management bundle increased from 18.4% to 36.1% ( P = 0.008), and the aOR for mortality decreased by 5.4% (95% confidence interval [CI], 2.5–8.4%) over the 2-year period . SSC bundles were designed for use in the emergency department and critical care areas but have been demonstrated to facilitate timely and appropriate critical care in areas with limited specialist skills on site . A revised resuscitation bundle referred to as “Sepsis Six” was implemented in the UK, and a recent study revealed reduced mortality (20% vs. 44.1%, P < 0.001) when the Sepsis Six bundle was used . Sepsis six includes delivery of high-flow oxygen, blood cultures, empiric intravenous antibiotic therapy, measurement of serum lactate, a full blood count, initiation of intravenous fluid resuscitation, and accurate assessment of urine output.

Major hemorrhage

Historically, hemorrhage was the leading cause of maternal death in the US, but over the past 20 years, improvement in care has decreased the cause-specific mortality ratio by approximately 30% to 1.8 per 100,000 live births . Moreover, current hemorrhage-related deaths are frequently preventable, and system solutions to improve readiness, recognition, and response to maternal hemorrhage have the potential to improve outcomes for future patients . Although maternal mortality from hemorrhage has decreased, the incidence of postpartum hemorrhage (PPH) and hemorrhage-related morbidity has increased, largely because of the increasing prevalence of uterine atony . The increase in PPH cannot be accounted for by adjusting for changes in maternal demographics, maternal comorbidity, or mode of delivery . Increasing associated morbidities include blood transfusion, renal failure, respiratory failure, and coagulopathy .

The reported incidence of hemorrhage depends on its definition. The ACOG recently defined PPH as “cumulative blood loss of at least 1000 mL or blood loss accompanied by signs/symptoms of hypovolemia within 24 h following the birth process” . Cumulative blood loss that exceeds 500 mL should trigger increased supervision and interventions but does not qualify as hemorrhage for the purpose of surveillance. Common definitions are also needed clinically to structure a coordinated interdisciplinary team response . Regardless of the mode of delivery, routine cumulative blood loss quantification is central to the prevention, diagnosis, and timely management of obstetric hemorrhage .

Abnormal placentation with placenta accreta is associated with maternal mortality, most commonly because of exsanguination at delivery or complications of massive blood transfusion. Outcomes are best when parturients are diagnosed antenatally and delivered electively in placenta accreta centers of excellence . Current imaging technology is imperfect; thus, unanticipated placenta accreta may complicate low-risk cesarean delivery. Other risk factors for massive hemorrhage and peripartum hysterectomy, such as chorioamnionitis, placental abruption, stillbirth, and antepartum hemorrhage, are difficult to predict . All centers, regardless of delivery volume, must be prepared to address massive PPH and peripartum hysterectomy . On average, small-volume delivery centers, particularly those with <1000 deliveries per year, have the greatest risk-adjusted rates of hemorrhage-related morbidity . Smaller institutions (≤1790 annual deliveries) also have the highest rate of PPH from all etiologies combined and from uterine atony alone .

A structured team-based response to obstetric hemorrhage improves maternal outcomes, regardless of the size or acuity of the delivery setting . Guidelines and protocols endorsed by maternal safety organizations emphasize early and aggressive management of obstetric hemorrhage, starting with risk factor identification, rapid diagnosis, timely management, and multidisciplinary review . Systems to accelerate the initial response include an obstetric emergency response team, a “PPH cart” to transport essential equipment to the bedside, and emergency hemorrhage medication packs that contain uterotonics, analgesics, and hemostatic agents . A unit-standard stage-based obstetric hemorrhage emergency management plan with checklists ensures a coordinated stepwise approach for diagnosis and management . Timely access to blood products, hemostatic agents, and surgical expertise must be ensured, even in small-volume delivery centers that focus on low-risk deliveries populations . A massive transfusion guideline commonly consists of a system to rapidly deliver six units of erythrocytes and four units of plasma; some centers add platelets or cryoprecipitate or combination of both . Although goal-directed therapy remains the gold standard for hemodynamic and hemostatic resuscitation, fixed-ratio transfusion (e.g., 1:1:1 ratio of erythrocytes, plasma, and platelets) allows clinical teams to respond to massive blood loss and minimize dilutional coagulopathy while awaiting the results of laboratory values . Finally, manual compression of the aorta may be a life-saving intraoperative maneuver to allow the surgeons to identify and control the sources of bleeding and, in the event of cardiac arrest, to improve the efficacy of cardiopulmonary resuscitation.

Hypertensive disorders of pregnancy

Hypertensive disorders of pregnancy is the sixth leading cause of maternal mortality in the US (9.4%) , with a cause-specific mortality ratio of 1.5 per 100,000, which is unchanged from previous reports . Intracranial hemorrhage is the leading cause of death in women with preeclampsia and eclampsia . Observational evidence suggests that uncontrolled hypertension (SBP >155–160 mmHg) is the most important risk factor for stroke in the setting of preeclampsia. Guidelines and protocols for acute management of hypertension focus on early diagnosis, prompt antihypertensive therapy, and seizure prophylaxis with magnesium sulfate.

The ACOG updated the definitions of all hypertensive disorders of pregnancy in 2013 to align nomenclature with maternal risk . The term “mild preeclampsia” was eliminated because endothelial disease severe enough to cause proteinuria or other end-organ injury is never “mild.” Instead, “preeclampsia without severe features” is distinguished from gestational hypertension by the presence of proteinuria. “Preeclampsia with severe features” is diagnosed if the patient demonstrates severe-range blood pressure elevation (two elevated [≥160/110 mmHg] blood pressure measurements at least 15 min apart) or evidence of other end-organ injury, even in the absence of any proteinuria. The California Maternal Quality Care Collaborative uses a lower threshold for severe-range blood pressure (155/105 mmHg) because of reports of adverse maternal outcomes when SBP was greater than 155 mmHg .

Bundles and toolkits have been developed for severe hypertension in pregnancy. In 2015, the ACOG added oral nifedipine to the recommendations for the treatment of acute-onset severe hypertension as an alternative to the first-line medications, and intravenous (IV) labetalol or IV hydralazine . Maintaining a “Severe Preeclampsia-eclampsia Box” containing essential and emergency medications is advisable for labor and delivery units to initiate early treatment . The role of the anesthesiologist may involve anesthesia risk assessment, blood pressure control, invasive monitoring, fluid management, eclampsia prophylaxis, and analgesia and anesthesia planning. It is highly advisable for patients to be counseled on the advantages of neuraxial analgesia for labor and not only for analgesic purposes. Neuraxial analgesia and anesthesia may decrease the requirement for general anesthesia. Failed airway management and hypertensive crisis with induction and emergence of general anesthesia are more common among women with preeclampsia.

Thromboembolism

VTE is the seventh leading cause of maternal mortality in the US (9.3%) . The cause-specific mortality ratio increased by 50% over the past 20 years to 1.5 per 100,000 . Pregnant or postpartum women have a four-fold increased risk of VTE compared with a nonpregnant control group . The annual incidence is five times higher (511.2 vs. 95.8 per 100,000 woman-years) in the postpartum period (up to 3 months) than that during pregnancy . The incidence of deep vein thrombosis (DVT) was three times higher than that of pulmonary thromboembolism (151.8 vs. 47.9 per 100,000 woman-years); however, it was 15 times more likely to occur in the postpartum period than during pregnancy (159.7 vs. 10.6 per 100,000 woman-years) . Pulmonary thromboembolism occurs in approximately 16% of patients with untreated DVT . Medical conditions associated with increased risk for pregnancy-related VTE include heart disease, thrombophilia, history of thrombosis, antiphospholipid syndrome, sickle cell disease, lupus, obesity, and smoking . Pregnancy and delivery complications associated with increased risk for VTE include preeclampsia, multiple gestation, anemia, hyperemesis, fluid/electrolyte/acid-base imbalance, antepartum hemorrhage, postpartum infection, PPH, transfusion, and cesarean delivery . AMA and Black race were also significant risk factors .

Thromboprophylaxis is the most important strategy to reduce maternal death from VTE, but the optimal combination of mechanical and/or pharmacologic prophylaxis depends on individual patient risk. A number of risk stratification tools have been proposed, including the Royal College of Obstetricians and Gynecologists criteria, the American College of Chest Physicians criteria, and the Caprini scoring system modified for obstetric patients . Routine risk assessment has been recommended during the first prenatal visit, during all antepartum admissions, immediately postpartum during a hospitalization for childbirth, and on discharge to home after birth . Pharmacologic prophylaxis with low-molecular-weight heparin or unfractionated heparin is recommended for those deemed to be at increased risk for VTE, including all antepartum patients on bedrest. The increasing use of antepartum pharmacologic prophylaxis has implications for the safe provision of neuraxial anesthesia and for the increased risk of PPH. Given the complex interactions between individual patient risk assessment and the dynamic nature of the delivery process, anesthesia providers should closely collaborate with obstetricians to ensure that women receive the optimal combination of thromboprophylaxis and anesthetic care.

Amniotic fluid embolism

Amniotic fluid embolism (AFE) is the ninth leading cause of maternal mortality in the US (5.3%) . Worldwide, the incidence varies considerably between 1.2 and 6.6 per 100,000 deliveries; this variation may be explained by the use of various definitions and criteria and the different reporting systems used internationally . Significant risk factors for AFE include eclampsia (aOR 26.5), placenta previa or abruption (aOR 17), cesarean delivery (aOR 15), cervical laceration or uterine rupture (aOR 7.4), instrumental vaginal delivery (aOR 6.6), AMA (aOR 2.7), diabetes (aOR 2.6), preeclampsia (aOR 2.5), other hypertensive disease (aOR 2.2), and medical induction of labor (aOR 1.7) .

There is great variability in the presentation of AFE. Classically, women present with a triad of hypotension, hypoxemia, and altered mental status followed by hemodynamic collapse, but many cases present with mild symptoms or atonic bleeding . Hemorrhage with disseminated intravascular coagulopathy (DIC) requires treatment with large volumes of blood products, including plasma, platelets, and cryoprecipitate. Plasma may be a particularly important component to reverse the DIC process . C1 esterase inhibitor (C1INH) has been found to be low in fatal cases than in nonfatal cases of AFE, and plasma not only supplies coagulation factors but also contains C1INH . Synthetic C1INH is a possible future treatment. Heparin has been suggested for the treatment of DIC; however, because of the high risk of hemorrhage associated with AFE, heparin administration is not recommended. Finally, case reports and case series describe the utility of hemofiltration to treat AFE, including extracorporeal membrane oxygenation, cardiopulmonary bypass, intra-aortic balloon pump, pulmonary artery thromboembolectomy, hemofiltration, and plasma exchange transfusions .

Anesthesia-related causes of mortality

Anesthesia-related mortality is the tenth leading cause of maternal mortality in the US. The cause-specific mortality ratio declined by 50% between 1987–1990 and 2006–2010, from 2 to 1 per million live births . In the most recent in-depth analysis, all cases were identified among women who delivered by cesarean (84%) or had unknown mode of delivery; none of the deaths were among women with a known vaginal delivery, suggesting that neuraxial labor analgesia is extremely safe .

The safety of general anesthesia for cesarean delivery has shown vast improvement since 1985, possibly because of advances in anesthesia equipment, monitoring, emergency procedures, and algorithms. In 1979–1984, general anesthesia had a 16.7-fold (95% CI 12.9–21.8) increased risk of maternal death compared with neuraxial anesthesia. By 1997–2002, this risk ratio decreased to 1.7 (95% CI 0.6–4.6), a difference that is not statistically significant .

Airway disasters account for most deaths from general anesthesia. Robust airway management algorithms have been proposed for the safe induction of general anesthesia . Careful planning, monitoring, and management are also needed for tracheal extubation and postoperative recovery. Between 1985 and 2003 in Michigan, there were eight anesthesia-related maternal deaths; five of these involved airway obstruction and/or hypoventilation, which occurred either on extubation (one case), in the postanesthesia care unit (PACU) (three cases), or in the postpartum ward (one case) . The latest data from the UK for 2010–2012 reported four anesthesia-related deaths; of these, two deaths were caused by hypoventilation related to general anesthesia for PPH (intraoperative [one case] and postoperative [one case]) .

Over this same period, the case fatality ratio for neuraxial anesthesia has not improved. This likely reflects the fact that an increasing proportion of high-risk cases now receive neuraxial anesthesia instead of general anesthesia. A serious complication occurs in approximately 1:3000 obstetric anesthetics, and the most frequent causes are high neuraxial block (1:4336), peripartum respiratory arrest (1:10,042), and unrecognized spinal catheter (1:15,435) . Unintentional dural puncture is often considered a benign complication of neuraxial blockade, but it can lead to lethal complications including brainstem herniation, central vein thrombosis, and intracranial hemorrhage. In the UK, there were two anesthesia-related deaths associated with unintentional dural puncture during attempted epidural catheter placement . One patient received an epidural blood patch and the other patient was treated conservatively; however, they suffered headaches for several weeks before presenting emergently and were diagnosed with cerebral vein thrombosis (one case) and subdural hematoma (one case) .

Safe anesthesia care requires interprofessional teamwork. Examples of medical errors and care management problems in anesthesia-related deaths include insufficient preanesthesia evaluation, patient factors (failure to disclose vital information), medication errors, inadequate trainee supervision, and inadequate postoperative monitoring . Obstetric PACU care should meet standard postanesthesia guidelines and standards, with an anesthesiologist readily available . In most settings, nurses require regular training opportunities to recognize and respond to rare events such as hypoventilation, airway obstruction, regurgitation, seizure, and cardiac arrest. The Stanford Obstetric Life Support course emphasizes cardiopulmonary resuscitation differences in pregnant and nonpregnant patients and reviews medical emergencies that can occur in an obstetric PACU .

Communication errors are an important cause of maternal morbidity and mortality, as documented in mortality reviews and closed claims reports . Two of the fatal Michigan cases had episodes of hypotension in more than one location (PACU and the postpartum ward, and PACU and ICU). Concise communication between the anesthesiologist and PACU staff is important, and defined criteria should be met prior to the patient being discharged from PACU, including cardiovascular and respiratory stability .

Anesthesia-related maternal deaths are extremely rare but are almost always iatrogenic and preventable. Further efforts to understand the epidemiology of anesthesia-related maternal death require population-level data collection with sufficient clinical detail to understand the contextual and clinical factors leading to the event. The Society for Obstetric Anesthesia and Perinatology published a report of serious complications related to obstetric anesthesia; at the time of data collection (2004–2009), there was no central repository for this type of data . In 2009, the American Society of Anesthesiologists created the Anesthesia Quality Institute, and in 2010, the National Anesthesia Clinical Outcomes Registry was formed.