Atrial fibrillation and non‐ischaemic cardiomyopathy in the peripartum period

Summary

A 31‐year‐old primiparous woman with a history of bigeminy as a teenager developed atrial fibrillation with rapid ventricular response during elective caesarean section. Initial postoperative medical management was undertaken on the maternal high dependency unit and involved the administration of beta‐blockers and digoxin. On postoperative day 1 the patient was transferred to the coronary care unit where she subsequently required synchronised direct current cardioversion to restore sinus rhythm. The patient remained on the coronary care unit for 5 days before discharge. Magnetic resonance imaging undertaken 6 weeks postpartum showed non‐ischaemic cardiomyopathy. In this report, we discuss tachycardia‐induced and peripartum cardiomyopathies, along with their potential underlying pathologies, incidence and associated morbidity. We describe potential pharmacological therapies including beta‐blockers and angiotensin‐converting enzyme inhibitors, as well as the implications of such medications for breastfeeding mothers. Patients presenting with palpitations in the antenatal period should receive prompt investigation including electrocardiography with ambulatory monitoring considered for those with persistent symptoms. Anyone with a proven cardiac arrhythmia should undergo echocardiography. This report illustrates the importance of the investigation of the symptoms of arrhythmia during pregnancy and emphasises the role of multidisciplinary working in the management of obstetric patients with complex medical comorbidity.

Introduction

Palpitations and anxiety are a relatively common presentation in the obstetric population, with ectopic beats and non‐sustained tachycardias identified in more than 50% of pregnant women experiencing palpitations. Atrial fibrillation is, however, rare and often associated with underlying cardiac pathology [1]. We present a case report of a 31‐year‐old gravida 1 para 0 patient with anxiety, tocophobia and a history of bigeminy as a teenager. She had presented with palpitations on multiple occasions during this pregnancy. Appropriate obstetric consultant review had occurred during the antenatal period, but no interventions had been required. During her elective caesarean section, she was found to be in atrial fibrillation and was subsequently found to have non‐ischaemic cardiomyopathy. In this report, we consider whether the history of palpitations should have prompted further antenatal investigation; the possible triggers for her atrial fibrillation; the types of non‐ischaemic cardiomyopathy; and treatment considerations in the breastfeeding mother.

Report

A 31‐year‐old gravida 1 para 0 patient was assessed in the consultant antenatal clinic at 36 weeks’ gestation. She had extreme anxiety and tocophobia, and a long‐standing history of palpitations without chest pain, breathlessness or syncope. Investigations during adolescence at the onset of her palpitation symptoms included a normal transthoracic echo and atrial bigeminy on electrocardiogram (ECG). However she reported no current cardiac symptoms when attending antenatal clinic and was subsequently booked for an elective lower segment caesarean section.

At 37 weeks and 4 days of gestation, the patient reported palpitations without chest pain. A midwifery review indicated normotension and an irregular pulse, and the patient was discharged with no further investigations. At 38 weeks, the patient was reviewed again and was found to be clinically well with a heart rate of 68 beats.min⁻¹ (rhythm not documented).

On admission for elective caesarean section, routine pre‐operative observations showed a heart rate of 109 beats.min⁻¹ and arterial blood pressure of 148/96 mm Hg. Spinal anaesthesia was performed as is standard for our institution, with hyperbaric bupivacaine 12.5 mg and diamorphine 300 µg injected intrathecally at the L3–4 interspace. Standard monitoring was applied and an intravenous (i.v.) phenylephrine infusion was commenced at 1 mg.h⁻¹. Three‐lead ECG monitoring demonstrated an irregular, narrow complex tachycardia with a heart rate between 150 and 170 beats.min⁻¹, consistent with atrial fibrillation. Intravenous magnesium sulphate 2 g was administered with no effect. However, in view of a stable arterial blood pressure, the consultant anaesthetist and obstetrician decided it was safe to continue with caesarean delivery. The patient remained haemodynamically stable throughout the procedure and a healthy baby was delivered with no subsequent surgical or paediatric concerns. Blood loss was recorded as 424 ml.

Analysis of a postoperative 12‐lead ECG confirmed atrial fibrillation with rapid ventricular response. Full blood count, urea and electrolytes and thyroid function tests were within the reference ranges. She was discussed with the on‐call cardiologist who advised that in the context of haemodynamic stability, normal bloods and absence of infection her atrial fibrillation should resolve within 48 h. A single dose of i.v. metoprolol 5 mg was prescribed for heart rate control.

Overnight the patient was cared for in the maternal high dependency unit, her atrial fibrillation persisted but she remained otherwise stable. After multidisciplinary team review, including cardiologist involvement, a transthoracic echocardiogram was performed. This showed impaired left ventricular systolic function with globular dilatation and left atrial dilatation. The advice from the cardiology team was for further rate control with oral bisoprolol 2.5 mg and digoxin 500 µg, transfer to the coronary care unit, with further treatment plans including angiotensin‐converting enzyme (ACE) inhibitor administration and synchronised direct current (DC) cardioversion following 24 h of treatment‐dose low molecular‐weight heparin and transoesophageal echocardiography to rule out intracardiac thrombus. Transoesophageal echocardiography and synchronised DC cardioversion were booked on the following day.

Transoesophageal echocardiography and 150 J synchronised DC cardioversion were performed under general anaesthesia with successful cardioversion on the first shock. The patient was monitored for 3 further days on the coronary care unit with ongoing palpitations. Telemetry displayed premature ventricular complexes and short runs of bigeminy. A repeat transthoracic echocardiogram showed a moderately‐dilated left ventricle with good systolic function and mild‐to‐moderate mitral regurgitation. The beta‐blockers and ACE inhibitors were stopped before discharge.

Midwifery care was provided daily on the coronary care unit, and the patient initially planned to breastfeed but was concerned about the transmission of medications via breast milk. The risks and benefits were discussed, and she opted to bottle feed.

Cardiac magnetic resonance imaging (MRI) at 6 weeks postpartum followed by cardiology and obstetric review were arranged at discharge. The MRI demonstrated severe left ventricular systolic impairment (ejection fraction 23%) and mild right ventricular systolic impairment (ejection fraction 39%), suggestive of non‐ischaemic dilated cardiomyopathy. Differential diagnoses include peripartum cardiomyopathy and tachycardia‐associated cardiomyopathy.

Discussion

Palpitations during pregnancy are common, which is unsurprising given the prevalence of palpitations in the general population and the normal physiological changes in pregnancy that result in a sinus tachycardia, particularly in the third trimester. Non‐sustained arrhythmias and ectopic beats are recorded in the majority of pregnant women investigated for palpitations. Sustained tachycardias, however, are only found in 0.3%. Unless the frequency of palpitations is high, 24‐ to 48‐h monitoring is unlikely to capture an event. Without syncope or other associated symptoms a loop recorder is not routinely recommended. Echocardiography is, however, recommended in patients with a proven cardiac arrhythmia [1].

In contrast to palpitations, cardiac arrhythmias in pregnancy are rare. Presenting in 1.2 out of 1000 pregnancies and usually only in the third trimester, they are usually associated with structural heart disease [2]. Precipitants for atrial fibrillation also include drug toxicity, pulmonary embolism, re‐entrant pathways (e.g. Wolff–Parkinson–White syndrome) and electrolyte abnormalities [2].

This patient’s history of palpitations is not unusual. Her normal echocardiogram as a teenager was reassuring and supports her initial consultant‐led antenatal management. However, her subsequent presentation to the antenatal clinic with an irregular heartbeat should have prompted a 12‐lead ECG. If an arrhythmia had been identified on ECG when she presented to antenatal clinic with palpitations, repeat echocardiography would have been indicated.

During pre‐operative assessment on the day of her caesarean section, the patient described her heart rate as ‘normal for her’ when anxious. As such, no pre‐operative ECG was performed. During the spinal anaesthetic, non‐invasive blood pressure and pulse oximetry monitoring were attached but the three‐lead ECG was not. The exact time of the onset of atrial fibrillation is therefore unclear. This monitoring practice is standard in our hospital; however, the Association of Anaesthetists guidelines suggest that all patients should have ECG monitoring during induction of any anaesthetic, be it regional or general [3].

No case reports of atrial fibrillation triggered by spinal anaesthesia could be identified. The most common arrhythmia during this technique is bradycardia due to the loss of cardiac sympathetic tone. Atrial fibrillation is thought to be triggered by spontaneous firing of the parasympathetic and sympathetic autonomic nervous system, both of which show decreased activity with spinal anaesthesia [4]. Therefore, spinal anaesthesia as a direct trigger for atrial fibrillation is unlikely. Given this woman’s history, however, an ECG should have been performed before her going to theatre and three‐lead ECG monitoring should have been present before the commencement of spinal anaesthesia. Had atrial fibrillation been identified before commencing anaesthesia, discussions regarding optimisation would have been appropriate.

The cardiac MRI performed 6 weeks after her caesarean section identified non‐ischaemic cardiomyopathy. Acquired causes include peripartum, tachycardia‐induced and inflammatory (i.e. myocarditis). In this case, there was no evidence of infection or myocarditis, leaving tachycardia‐induced and peripartum cardiomyopathy as possible differential diagnoses.

Classically, tachycardia‐induced cardiomyopathy is described as the presence of cardiomyopathy in tandem with a tachyarrhythmia that should resolve with the rate control [5]. As such, the cardiac MRI should show some improvement from function as assessed by inpatient transthoracic echocardiogram.

In contrast, peripartum cardiomyopathy is a rare disorder occurring in previously healthy individuals who develop left ventricular dysfunction in either the last month of pregnancy or within 5 months of delivery [6]. About 50% of patients with peripartum cardiomyopathy recover spontaneously but persistent disease at 6 months is associated with a poor prognosis. Risk factors for peripartum cardiomyopathy include age over 30 years; multiparity; Afro‐Caribbean ethnicity; multiple gestations; obesity; pre‐eclampsia; and chronic hypertension. The only risk factors our patient exhibited was age over 30 years. Though peripartum cardiomyopathy is normally associated with a sinus tachycardia, it can also be present with other arrhythmias [7].

Decisions regarding the initial medical management of this patient’s tachyarrhythmia had to take account of the patient’s wish to breastfeed and concerns about the impact of beta‐blockers and ACE inhibitors on the infant. Though the patient opted to breastfeed initially, she became concerned that the baby appeared drowsy and as such the decision was made by the patient to switch to bottle feeding. A 2018 review of peripartum cardiomyopathy and considerations for breast feeding summarises treatment options and their impact; it emphasises that because breastfeeding conveys significant benefits to both infant and mother, abstaining from breastfeeding is not a risk‐free option [8].

The pertinent medications in this case are β‐blockers and ACE inhibitors. β‐blockers are typically not found in high levels in breast milk due to their protein binding and short half‐life. However, bisoprolol has been associated with neonatal hypotension, lethargy and poor feeding. Metoprolol and enalapril are considered safe first‐line therapies [8]. Obstetrics, midwifery, cardiology and pharmacy input should help support the mother to make informed decisions about breastfeeding, including providing what limited evidence is available.

Guidelines by the Obstetric Anaesthetists’ Association and the Association of Anaesthetists highlight the importance of multidisciplinary management of complex obstetric patients [9]. This patient had twice‐daily multidisciplinary reviews by obstetric, anaesthetic and midwifery personnel, with input from cardiology whilst on the maternity high dependency unit.

The decision to care for this patient in the cardiology setting (in a separate wing of the hospital) may have affected the quality of postnatal care delivered. Daily review by respective teams occurred, but there was a paucity of multidisciplinary discussion at the bedside. Midwifery visits were a planned daily event, but there was reduced access to some important services that are embedded in the postnatal ward such as infant monitoring and breastfeeding support [10]. This may have contributed to the patient’s decision to stop breastfeeding.

This case emphasises the need to consider pathological causes of tachycardia in anxious patients and the importance of antenatal review and appropriate peri‐operative monitoring. Electrocardiography and echocardiography are appropriate approaches to investigate tachyarrhythmia and underlying cardiac disease. Multidisciplinary working between obstetric, anaesthetic, midwifery and other specialty colleagues is paramount in coordinating the care of obstetric patients with complex comorbidities, particularly when women are transferred to other areas of the hospital to receive specialist clinical care.