A case study of a 42-year-old woman with an unexpected cardiac arrest on induction of general anaesthesia

Intended learning outcomes

1. To describe the features of Takotsubo cardiomyopathy and its relevance to cardiovascular collapse in the perioperative setting.

2. To consider the causes of perioperative cardiac arrest.

3. To explore the perioperative management of Takotsubo cardiomyopathy.

Introduction

A 42-year-old woman attended hospital for laparoscopic oophorectomy in the context of a long-standing pelvic mass. An anaesthetic pre-assessment identified a background of migraine, sciatica, previous peptic ulceration and no other significant co-morbidities. It was noted that the patient had not previously been given general anaesthetics and had a low body mass index of 18. She was deemed to have an American Society of Anesthesiologist Physical Status (ASA-PS) classification of 1, with consent provided for a general anaesthetic. Pre-operative observations demonstrated a heart rate (HR) of 60 beats per minute (bpm), blood pressure (BP) 108/68 mmHg and an oxygen saturation of 100% on room air.

Report

On induction of anaesthesia the patient was given 100 µg of fentanyl, 200 mg of propofol and 30 mg of atracurium. An iGel laryngeal mask was inserted without complication. Approximately 60 seconds after induction the patient was noted to have hypotension with an initial BP reading of 59/36 mmHg and an HR of 57 bpm; 1.25 mg metaraminol was administered with no improvement in blood pressure. The patient subsequently became bradycardic with the lowest HR of 25 bpm, in a narrow complex sinus rhythm. Asystole was observed on cardiac monitoring and cardiopulmonary resuscitation was commenced immediately. Adrenaline (0.5 mg) was administered, along with 600 µg of atropine. The return of spontaneous circulation was achieved in fewer than 5 minutes. However, the patient then developed a narrow complex tachyarrhythmia with an HR of 170 bpm. Two boluses of 6 mg of adenosine were administered to achieve restoration of a sinus rhythm; 4 g of intravenous magnesium sulphate was also administered. An initial 12-lead electrocardiogram showed a sinus rhythm with no significant ST-segment depression. A decision was made to terminate anaesthesia and awaken the patient in the theatre. The patient was transferred awake to the Intensive Care Unit for further monitoring and investigation.

Following admission to the Intensive Care Unit, the patient did not require ongoing respiratory or cardiovascular support. A chest X-ray (Fig. 1) showed pulmonary vascular congestion and lower zone opacifications with no gross cardiomegaly.

Figure 1

A chest X-ray performed on the day of cardiac arrest shows pulmonary vascular congestion.

A transthoracic echocardiogram was performed the following day that showed severe left ventricular (LV) systolic dysfunction with akinetic mid and apical LV segments and an estimated ejection fraction of 15–20%. Impaired LV diastolic function (with normal filling pressures at rest) was seen. In addition, impaired right ventricular systolic function was observed with preserved contraction only present in the basal segments. Serial electrocardiograms (ECGs) showed progressive and new widespread T-wave inversion during the subsequent 72 hours (Figs. 2 and 3). As the differential diagnosis included perioperative anaphylaxis, mast cell tryptases were measured and the levels were not elevated above the normal range.

Figure 2

An ECG performed on the day of cardiac arrest shows a sinus rhythm with no significant ST-segment deviation.

Figure 3

An ECG performed on day 3 after cardiac arrest shows a sinus rhythm with T-wave inversion to leads I, II, III, aVF, aVL, V2, V3, V4, V5 and V6.

The patient was reviewed by the cardiology team and treatment for acute severe heart failure was initiated with dapagliflozin, bisoprolol and spironolactone. A coronary angiogram demonstrated no significant vessel obstruction. The patient’s symptoms improved with pharmacological management, and she was discharged home after 7 days. Regular cardiology follow-up occurred during the subsequent 10 months and the patient reported improvement in breathlessness. Improved LV function was observed on repeat echocardiogram 4 months after the event (LVEF 50–54%). A cardiac MRI scan showed a non-ischaemic cardiomyopathy with borderline LV dilatation and no hypertrophy.

Given the history of recovered LV function and unobstructed coronary arteries, a most likely explanation for her episode of decompensation was determined to be a Takotsubo-like cardiomyopathy (also known as ‘stress cardiomyopathy’). Her regular cardiac medications were stopped after 10 months. At time of writing this case report, she is yet to undergo her laparoscopic oophorectomy; however, on her last review by the cardiology team, it was felt it would be possible for her to have her procedure, with a suggestion made for the administration of pre-operative beta-blockers.

Question to be answered

What is Takotsubo cardiomyopathy and its relevance to the patient in a perioperative setting?

Discussion

Takotsubo cardiomyopathy (TCM) (or ‘stress cardiomyopathy’) is a form of temporary ventricular dysfunction, with a classical apical ballooning of the left ventricle on cardiac imaging (giving the appearance of an octopus trap, translated as ‘takotsubo’ in Japanese) [1]. TCM was first identified in 1990 [2]. It is currently thought to account for 1–3% of all patients presenting with acute coronary syndrome [3]. Most cases are sporadic with genetically inherited forms being much less common. The population prevalence of TCM is unclear; however, TCM was diagnosed in an estimated 0.02% of hospital admissions in the United States in 2008 [4]. Women are more commonly affected compared with men, with most cases occurring in those over the age of 50. Work by Hessel suggested that the perioperative incidence of TCM was approximately 1 in 6700 cases [5].

The primary hypothesis for the pathophysiology of TCM is a catecholamine storm caused by excessive sympathetic stimulation that leads to myocardial stunning, possibly through microvascular dysfunction or direct catecholamine toxicity to myocytes [6]. The most used diagnostic criteria are the Mayo criteria (Table 1), with key features including transient hypokinesis/akinesis of LV mid segments (with or without apical involvement), absence of obstructive coronary disease, new ECG abnormalities (ST elevation/T-wave inversion with modest troponin rise) and an absence of phaeochromocytoma or myocarditis [7].

Cardiac arrest during general anaesthesia remains an uncommon occurrence, with the 7th National Audit Project (NAP7) demonstrating an incidence of approximately 1 in 3000 cases (not including ASA class or the nature of procedure) [8]. The most common causes of cardiac arrest were haemorrhage, bradyarrhythmia and cardiac ischaemia. Patients suffering a cardiac arrest in theatres had better outcomes compared with other in-hospital cardiac arrests, with 75% of patients surviving their initial event. Overall, 28% of cardiac arrests occurred at the induction of anaesthesia but before the start of surgery, as illustrated in this case study and 27% of cardiac arrests were in those individuals who were ASA-PS 1 or 2. One percent of cardiac arrests in NAP7 were associated with isolated severe hypotension and 15% of patients identified had an initial rhythm of asystole. No cases of TCM were reported as part of NAP7 [8].

There have been several case studies of patients suffering perioperative cardiac arrest due to suspected TCM or in response to drugs used during the perioperative period [91011]. A systematic review by Agarwal et al. [12] identified no definitive risk factors for perioperative TCM; however, in the perioperative setting, this appears to occur in younger patients and ST elevation is less prevalent than in TCM outside the perioperative setting. In the absence of definitive risk factors, TCM should be considered as a differential diagnosis in any patient who suffers cardiovascular compromise in the perioperative period.

Common complications of TCM include cardiogenic shock (approximately 20% of cases) and atrial or ventricular tachyarrhythmias. Ventricular thrombus has also been reported due to ventricular segmental akinesis. In the context of TCM, the estimated 30-day mortality is 5.9% and the incidence of major adverse cardiovascular events is 7.1% [13]. There are limited data for outcomes focused on perioperative TCM. Patients may have persistent symptoms despite the recovery of cardiac function on imaging.

As illustrated in this case report, the first presentation of TCM may be on induction of anaesthesia or during a general anaesthetic. There are no definitive guidelines on the perioperative management of TCM. The best available guidance has been presented in review articles by Pillitteri et al. [14] and Agarwal et al. [15]. If a patient has known TCM, elective surgery should be postponed if there is ongoing cardiac compromise, until transient cardiac dysfunction has recovered [14]. If surgery cannot be avoided, or for elective cases in patients with prior TCM, disease severity can be assessed by ECG, echocardiogram and by measurement of cardiac enzyme levels. A coronary angiogram should be considered to determine coexisting ischaemic heart disease. The cardiology team should be involved in the patient’s perioperative care and informed early of the intended procedure. Anaesthetic techniques should aim to reduce stress (both psychological and physical) and the consequent catecholamine release . Depth of anaesthesia should be balanced to reduce sympathetic output in the context of potentially reduced myocardial reserve, particularly at the induction of anaesthesia. Bradycardia, tachycardia and arrhythmia should be avoided to preserve cardiac output [14]. Regional anaesthesia may result in lower catecholamine release by avoiding the stimulation of tracheal intubation [10, 16] and, when appropriate, a combined general and regional technique should be used to reduce any exposure of the cardiovascular system to excess catecholamine release. There have been several case studies describing patients experiencing TCM with spinal anaesthesia [17, 18], suggesting that a regional approach alone may not prevent cardiac compromise. Enhanced intraoperative monitoring, including invasive BP monitoring with or without cardiac output monitoring, should be considered. In association with invasive monitoring, pre-emptive central venous catheter insertion can permit the prompt administration of vasoactive drugs. As with any patient at high risk for postoperative instability, the postoperative destination must be considered, including enhanced-care or intensive-care areas.

In the event of new-onset intraoperative cardiomyopathy, the surgical procedure should be stopped if possible and any sequelae of cardiogenic shock or arrhythmia treated as necessary. Early cardiology input should be sought; investigations including troponin measurement, 12-lead ECG and an echocardiogram undertaken. The patient should be transferred to an enhanced-care or intensive-care environment to facilitate ongoing monitoring and intervention [14]. It is important to consider other differential diagnoses such as acute coronary syndrome, electrolyte abnormalities (in the context of arrhythmia) and adverse drug reactions including anaphylaxis. In the event of cardiac arrest, as with any other cause, Advanced Life Support should be delivered in keeping with Resuscitation Council UK guidelines. There is no high-quality evidence available to guide the medical management of TCM. Many patients will receive antiplatelet therapy prior to coronary angiogram that is discontinued if coronary artery disease is excluded. The main principles include management of arrhythmia and cardiogenic shock, as with other causes of cardiomyopathy. As with other causes of heart failure, angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptor blockers may be administered to prevent adverse ventricular remodelling and beta-receptor antagonists may be used to reduce adrenergic drive. Mineralocorticoid receptor antagonists are often used for patients with significant LV dysfunction [19] and SGLT2 inhibitors may also play a role in treatment (similar to their increasing use in other causes of systolic heart failure) [20]. There has been some suggestion that alpha- and beta-blockade may prevent stress-induced cardiac dysfunction [21]; however, there have been no definitive trials to support this. Most patients will recover ventricular function with supportive care alone, although the approximate recurrence rate is 11% [22].

With regards to the patient in our case report, whilst a diagnosis of a Takotsubo-like episode was made, the underlying cause for decompensation remains unclear. A possible cause may have been an initial anaesthesia-induced hypotension with a subsequent Takotsubo-like episode triggered by exposure to exogenous catecholamine and a sympathomimetic agent. There was also the possibility that the patient had pre-existing cardiac dysfunction not identified in pre-assessment that was exacerbated by several perioperative factors, including pre-operative anxiety and administration of myocardial depressive agents. We believe the uncertainty in this case report reflects the rarity of TCM and the relative lack of literature on its prevalence and risk during the perioperative period.

In conclusion, this case describes a patient suffering a perioperative cardiac arrest of initially unclear aetiology. We highlighted Takotsubo cardiomyopathy as a rare, but important, differential diagnosis of sudden cardiovascular collapse in the perioperative period. This case emphasises the importance of early diagnosis and supportive care in the event of decompensation, with our patient recovering ventricular function within several months of the initial event. Whilst the optimal anaesthetic management of TCM remains unclear, emphasis remains on avoiding anaesthetic exposure during a known episode of TCM and taking measures to reduce catecholamine release by both pharmacological and non-pharmacological measures.

Artificial intelligence declaration

No AI-assisted technology was used in the authorship of this article.

Declarations and conflicts of interest

Artificial intelligence declaration

No AI-assisted technology was used in the authorship of this article.