Preeclamptic heart failure — perioperative concerns and management: a narrative review

Hypertensive disorders in pregnancy are an important cause of maternal morbidity and mortality. In addition to being an important risk factor for long-term maternal cardiovascular disease, it also contributes to a significant risk of fetal prematurity and mortality (Mehta et al. 2020). With a significant variation in racial and geographic distribution, hypertensive disorders in pregnancy disproportionately affect women of Black, African-American, and Alaskan Native women (Umesawa and Kobashi 2017). Hypertensive disorders in pregnancy comprise gestational hypertension, preeclampsia (with and without severe features), preeclampsia superimposed on chronic hypertension, and eclampsia (Rana et al. 2019; Say et al. 2014).

Cardiovascular diseases complicate up to 4% of all pregnancies with the incidence likely to be higher if hypertensive disorders are also included (Regitz-Zagrosek et al. 2018). Among all the cardiovascular diseases complicating pregnancy, heart failure (HF) remains the most common complication, regardless of the cause, and the incidence has been increasing over the past decades (Stergiopoulos et al. 2019; Mogos et al. 2018). Preeclampsia is an important risk factor for the development of HF during pregnancy and the postpartum period (Anthony and Sliwa 2016). The prevalence of HF in preeclamptic parturients is relatively underexplored and underreported probably owing to the challenges in accurate diagnosis and the resulting misclassifications. Preeclamptic parturients can develop HF of two distinct types. While some preeclamptics develop a form of HF characterized by left ventricular (LV) systolic dysfunction, referred to as HF with reduced ejection fraction (HFrEF), others exhibit variable degrees of LV diastolic dysfunction, resulting in HF with preserved ejection fraction (HFpEF) (Bright et al. 2021). The pathophysiology, cardiovascular mechanics, therapeutic options, and peripartum management strategies are different for these two forms of HF in preeclamptic parturients. It is, hence, crucial that clinicians be able to effectively screen for HF, correctly identify the sub-type, and risk-stratify preeclamptic parturients with HF, so that appropriate targeted interventions can then be initiated during the peripartum period. This review attempts to summarize the pathophysiology of preeclamptic HF, followed by a detailed description of the current evidence base and clinical considerations regarding the peripartum anesthetic and critical care management of preeclamptic parturients with HF.

During the course of a normal pregnancy, the maternal cardiovascular system undergoes extensive adaptations in order to maintain an adequate uteroplacental perfusion throughout pregnancy (Ramlakhan et al. 2020). By 8 weeks of gestational age, maternal systemic vascular resistance decreases by 10–30% and reaches a nadir between 20th and 26th weeks of pregnancy. This decline in SVR is accompanied by a decline in mean arterial pressure, which reverses from 26th to 28th weeks, reaching prepregnancy values at full term (Ouzounian and Elkayam 2012; Duvekot et al. 1993). The inotropic and chronotropic effects of pregnancy occur secondary to the effect of stimulation of baroreceptors in the cardiopulmonary and renal vasculature, with a subsequent increase in the contractility and heart rate by 15–25% and stroke volume by 20–30%. This brings about an attendant increase in cardiac output by 30–50%. Anatomically, the increase in preload and myocardial contractility manifests in LV hypertrophy and a larger left atrial diameter (Hunter and Robson 1992; Savu et al. 2012; Meah et al. 2016; Gaasch and Zile 2011). These cardiovascular changes are usually benign and fully recover within 1 year of delivery (Gaasch and Zile 2011; Melchiorre et al. 2016).

Preeclampsia results from the failure of embryonic cytotrophoblasts to invade and remodel the myometrial spiral arteries (Rana et al. 2019). The resultant placental bed comprises of arteries that are small and hyper-responsive to vasomotor stimuli, resulting in a mismatch between uteroplacental supply and fetal demands (Burton et al. 2019). This sets off a cascade of pro-inflammatory responses and endothelial dysfunction in the maternal circulation leading to increased maternal systemic vascular resistance (Orabona et al. 2018). This “arterial stiffness,” and the subsequent lower plasma volume expansion and reduced cardiac output, triggers maternal neurohormonal hyperactivity culminating in a vicious cycle of sustained maternal hypertension (Orabona et al. 2018). While these observations are consistent in untreated preeclamptics, they are contrasting to results in treated preeclamptics, who present with more heterogeneous cardiovascular variables. Furthermore, the pattern and degree of cardiac remodeling also seem to be influenced by the time of onset of preeclampsia (early onset vs late onset). In preeclamptics, the LV remodeling presents as concentric LV hypertrophy leading to left ventricular systolic dysfunction, impaired myocardial contractility, and a reduced LV ejection fraction, resulting in HFrEF (Melchiorre et al. 2014; Castleman et al. 2016). LV diastolic dysfunction, on the other hand, characterized by reduced regional myocardial relaxation, is a direct consequence of myocardial stiffening due to LV hypertrophy, which is reflected in left atrial remodeling by means of elevated left-sided filling pressure (Rafik Hamad et al. 2009; Melchiorre et al. 2011; Ponikowski et al. 2016a). About 50% of patients develop myocyte loss, subsequent fibrosis, and LV remodeling resulting in HFrEF, manifesting as reduced pump efficiency and impaired peripheral perfusion (Metra and Teerlink 2017). On the other hand, the pathophysiology of HFpEF is still not clear, but it seems to be related to a pro-inflammatory state resulting in endothelial dysfunction, myocyte hypertrophy, and collagen deposition together with fibrosis and reduced LV compliance. Ineffective calcium metabolism within the sarcoplasmic reticulum and altered mitochondrial density within the myocardium have also been implicated in the pathophysiology of diastolic dysfunction in preeclamptics (Maaten et al. 2016).

However, not all preeclamptics with features of maladaptive cardiac remodeling during pregnancy develop acute cardiac failure (Melchiorre et al. 2014). Acute decompensation of the cardiovascular system could occur in the presence of one or more precipitating factors, resulting in overt HF (Table 1).

Preeclamptic parturients present with hypertension after 20 weeks of gestation along with new onset proteinuria, with or without severe features. Severe features include blurring of vision, headache, seizures (eclampsia), dyspnea/hypoxia, right upper quadrant pain, nausea/vomiting, oligohydramnios, intrauterine growth restriction, and in some cases multiorgan dysfunction (Burton et al. 2019). In addition to the abovementioned spectrum of symptoms, a preeclamptic parturient in HF could present with sudden onset dyspnea, orthopnea, paroxysmal nocturnal dyspnea, increased fatigue, and syncopal attacks. Clinical examination could reveal tachycardia, pedal edema, elevated jugular venous pressures, hypo/hypertension, displacement of the apical impulse, audible third heart sound, and variable degree of pulmonary crepitation (Anthony and Sliwa 2016; Jensen et al. 2007). Several of these clinical features can occur as part of the cardiovascular and respiratory physiological changes that occur with normal pregnancy. Hence, a high degree of clinical suspicion and a low threshold for diagnostic investigations are warranted.

The differential diagnosis of obstetric acute HF is quite wide, and a clear understanding of these various entities is essential for an obstetric anesthesiologist (Table 2). An important and distinct cause of HFrEF in the peripartum period is peripartum cardiomyopathy (PPCM) and is a diagnosis by exclusion (Bauersachs et al. 2019).

Increased total leukocyte count, low platelets, elevated liver enzymes, serum creatinine, and uric acid are part of the spectrum of preeclampsia (Burton et al. 2019). While CKMB levels are unreliable during pregnancy and labor, cardiac-specific troponin levels are generally elevated in preeclamptics even in the absence of overt HF, leading to diagnostic uncertainty (Dockree et al. 2021). Recently, studies have evaluated the role of brain natriuretic peptide (BNP) and N-terminal pro-BNP (NT-BNP) in the diagnosis and prognostication of HF in obstetrics. BNP and NT-BNP are peptides released by the atria and ventricles in response to increased wall tension. They remain normal throughout pregnancy but are mildly elevated in preeclampsia (Kampman et al. 2014). However, they are markedly increased in PPCM and preeclamptic HF (Sliwa et al. 2010). A value of BNP < 100 pg/ml and NT-proBNP < 300 pg/ml usually rules out HF. An electrocardiogram is recommended in all preeclamptics with suspected HF. The presence of arrhythmias, ST-T segment changes, or block patterns could help in diagnosing the etiology and complications of HF (Ramlakhan et al. 2020). The presence of B lines on lung ultrasonogram is also suggestive of pulmonary edema or fluid overload with HF (Muniz et al. 2018; Maw et al. 2019).

Echocardiography involves the ultrasonographic assessment of cardiac structure and function, and transthoracic (TTE) has been increasingly used in the perioperative management of sick parturients. Echocardiography is particularly useful in the differential diagnosis of acute HF symptomatology (Table 2). A “focused TTE,” also known as “goal-directed TTE,” is an abridged version of a more comprehensive 2-dimensional TTE, and it is a point of care and noninvasive bedside tool. The main aim of this focused study is the prompt recognition of gross abnormalities using the standard echocardiographic views, i.e., parasternal, subcostal, and apical, so as to facilitate early stabilization and institution of treatment (Griffiths et al. 2018).

The Rapid Obstetric Screening Echocardiography (ROSE) is an obstetric-specific focused TTE, which takes into consideration the anatomic and physiological changes occurring during pregnancy (Dennis 2011). It comprises of the basic TTE views to identify gross abnormalities, but could also include a more detailed assessment of cardiac output using Doppler and an assessment of diastolic function by more experienced clinicians. The basic principles of the ROSE scan are described in Table 3. A detailed review of a comprehensive echocardiographic examination is beyond the scope of this article and can be found elsewhere (Castleman et al. 2016; Rafik Hamad et al. 2009).

In view of the paucity of research performed specifically in preeclamptic parturients with HF, most of the recommendations regarding clinical management are derived from clinical studies on HF in non-obstetric patients or expert opinion (Mehta et al. 2020; Regitz-Zagrosek et al. 2018). There exists a broad consensus that treatment modalities should avoid adverse fetal effects in parturients, and treating clinicians must be cognizant of the fetal safety profile of drugs used for the management of symptoms of HF in women who are pregnant and breastfeeding (Table 4). Obstetric patients with heart failure are best managed in a high dependency or critical care unit due to increased morbidity and mortality.

In contrast to treatment of HFrEF, there are even fewer studies done on patients with HFpEF. Since most studies of HFpEF have been performed on elder patients with more complex cardiac comorbidities including coronary artery disease, it is challenging to extrapolate these data onto preeclamptic obstetric patients with HFpEF. Currently, diuretics are the only class of drugs that are recommended for treating fluid overload in parturients with HFpEF, keeping in mind the fetal effects of the drugs. There are currently no recommendations regarding the use of other groups of drugs for the treatment of HFpEF in preeclamptic parturients (Henning 2020; Deshmukh et al. 2020).

Preeclamptics with HF are at a risk of cardiac dysrhythmias, with atrial fibrillation being the most common arrhythmia. Cardio-selective beta-blockers, digoxin, and calcium channel blockers are safe and effective antiarrhythmics for use in pregnancy (Joglar and Page 1999). In unstable patients, cardioversion/defibrillation, wherever applicable, is safe and should be performed without any delay. The decision to perform cardioversion/defibrillation under intravenous sedation or after securing the airway with an endotracheal tube must be taken with due consideration to the fasting status of the patient and the potential risk of gastric regurgitation and pulmonary aspiration (Ferrero et al. 2004). Whenever possible, it may be prudent to use fetal monitoring during cardioversion/defibrillation, due to the theoretical risk of inducing secondary arrhythmia in the fetus.

Ventricular dysfunction is an important risk factor for ventricular tachyarrhythmias and sudden cardiac death. However, given the higher rate of recovery of LV function with delivery and drug therapy, the early implantation of implantable cardioverter-defibrillators is generally not recommended. Wearable cardioverter-defibrillators have been proposed as a bridge to full cardiac recovery (Kober et al. 2016; Elming et al. 2017; Duncker et al. 2017).

Pregnancy is a procoagulant state, and HF and atrial fibrillation are independent risk factor for systemic thromboembolism (Dunkman et al. 1993; Refuerzo et al. 2003). The increased procoagulant state persists into the postpartum period, making these patients susceptible to peripheral arterial and venous thromboembolism. In this context, prophylactic administration of unfractionated or low molecular weight heparin is recommended. Patients with persistent atrial fibrillation, evidence of systemic thromboembolism, and those with an intracardiac thrombus, however, require therapeutic anticoagulation (Nelson-Piercy et al. 1997; Howie 1986).

Termination of pregnancy is recommended in cases of severe preeclampsia and eclampsia. Multidisciplinary team input is necessary for planning for delivery, and a coordinated liaison between obstetricians, anesthesiologists, cardiologists, intensivists, and neonatologists is crucial for shared and individualized decision-making in these patients (Easter et al. 2020). Severe preeclampsia and eclampsia are maternal indications for urgent termination of pregnancy, and for a majority of parturients with HF, vaginal delivery remains the safest option, in the absence of obstetric indications for cesarean delivery (Regitz-Zagrosek et al. 2018). For vaginal delivery, it is important to note that intense and prolonged Valsalva maneuver during straining and bearing down may be poorly tolerated in preeclamptic parturients with HF. This is due to rapid reduction in preload and increase in afterload that occurs normally during the Valsalva’s maneuver. Moreover, the overshoot in the cardiac output that occurs upon release of the Valsalva could exacerbate symptoms of HF in these parturients (Nishimura and Tajik 1986, 2004). Urgent cesarean delivery regardless of the gestational age should be considered in rapidly worsening parturients with hemodynamic instability despite ongoing medical therapy for HF. In significantly high-risk parturients, a planned cesarean delivery may offer the advantage of avoiding emergent delivery permitting the availability of senior clinicians.

Effective labor analgesia preserves hemodynamic stability and cardiopulmonary function during labor. By abolishing the pain-induced release of catecholamines, good neuraxial labor analgesia could attenuate the increase in myocardial oxygen demand and increased cardiac output and fluctuations in preload and afterload during labor (Shnider et al. 1983; Jouppila et al. 1982; Ramos-Santos et al. 1991). Therefore, in the absence of any contraindications, most obstetric anesthesiologists prefer to initiate epidural analgesia as early as possible in these patients. Epidural-only, combined spinal-epidural (CSE), and intrathecal opioid-only CSE techniques have all been successfully employed to improve analgesic efficacy, quality of block, and sacral root coverage (Ankichetty et al. 2013; Dennis 2012; Pascual-Ramirez et al. 2011). Whatever the technique used, it is important to gradually titrate the epidural drugs to avoid rapid reductions in systemic vascular resistance and preload, which can cause precipitous hypotension.

All preeclamptic parturients with HF must have appropriate cardiac monitoring during labor to predict and prevent serious maternal cardiac events. Waveform pulse plethysmography, electrocardiogram, and noninvasive blood pressure monitoring at regular intervals are minimum monitoring standards for these high-risk parturients. An invasive arterial line placement could be considered in sicker laboring parturients, particularly those who have a higher chance of requiring an urgent cesarean section. The indwelling arterial line could be invaluable during the rapid conversion of epidural labor analgesia into epidural anesthesia for cesarean delivery and in instances when general anesthesia is required for emergency cesarean delivery. Since laboring parturients often move and strain during the course of labor and delivery, measurement of central venous pressure is often of low clinical utility (Nishimura and Tajik 2004). However, insertion of a central venous catheter may be considered in instances where continuous infusion of vasoactive drugs is anticipated during labor and in the immediate postpartum period.

Currently, there is no consensus regarding the superiority of any single mode of anesthesia in terms of improved maternal and fetal outcomes in preeclamptic parturients with HF undergoing cesarean delivery. Regardless of the mode of anesthesia, the hemodynamic goals are common to all approaches. Sudden fall in the systemic vascular resistance should be avoided, as it can lead to maternal hypotension and uterine hypoperfusion. Preventing bradycardia and strict attention to intravenous fluid balance are other important aspects of perioperative care during operative delivery. Noninvasive cardiac output monitoring using pulse waveform analysis, suprasternal Doppler, and bioreactance have been used to guide the administration of intravenous fluids, vasopressors, and oxytocin in preeclamptics, and are valuable strategies to optimize hemodynamics of preeclamptics in HF (Armstrong et al. 2011; Ackerman-Banks et al. 2023).

Whether cesarean delivery is being performed under neuraxial or general anesthesia, an intra-arterial catheter for beat-to-beat blood pressure monitoring should be strongly considered. Additionally, in parturients requiring infusion of vasopressors/inotropes, central venous access should be established.

The postpartum phase is a high-risk period for maternal decompensation in the case of preeclamptic patients. Regardless of the mode of delivery, the “autotransfusion” of blood from the retracting uterus and relief of pressure over the inferior vena cava can result in a significant increase in cardiac preload (Arendt and Lindley 2019). This increase in preload can place undue strain over the dysfunctional left ventricle, predisposing to the development of pulmonary edema and HF. This is particularly a problem in patients with significant diastolic dysfunction. Inadequate pain management in the immediate postpartum period can increase the heart rate, cardiac output, and maternal myocardial oxygen consumption. A combination of these factors can increase the risk of maternal cardiac decompensation in the immediate postpartum period (Melchiorre et al. 2014). Oxygen supplementation, adequate analgesia, attention to intravenous fluids, judicious use of oxytocin infusions, and cardiac monitoring are all important aspects of postpartum care in these patients.

An intensivist must be involved early in the peripartum management of women who are at high risk for rapid deterioration (Banerjee and Cantellow 2021a). All preeclamptic parturients who are inotrope/vasopressor dependent should be transferred early to a high dependency or critical care unit for further management, in view of their high morbidity and mortality. The specific principles of obstetric critical care for such patients are outlined in Table 5. Management specific to acute HF in pregnancy is similar to the nonpregnant state and includes optimization of preload, afterload, myocardial contractility, rhythm, maintenance of oxygenation, and organ perfusion (Banerjee and Cantellow 2021a, 2021b).

Preeclampsia is an important risk factor for the development of HF during pregnancy and postpartum, which can present as either systolic or diastolic failure. It is important for anesthesiologists to have a good understanding of the underlying pathophysiology and the current evidence base for early diagnosis and stabilization of these high-risk patients. The anesthetic management of these patients for vaginal and operative delivery is challenging and requires multidisciplinary team inputs. Early referral of these high-risk parturients to higher centers with obstetric anesthesiology, cardiology, and critical care expertise is crucial for successful obstetric and neonatal outcomes. More research is needed to develop risk stratification models, perioperative anesthesia protocols, and practice guidelines for the anesthetic and critical care management of this high-risk population subgroup.