Year : 2018 | Volume
: 8 | Issue : 1 | Page : 50--53
Management of emergency caesarean section in a patient with decompensated critical aortic stenosis
Andrew C Leatherbarrow
Selwyn Crawford Department of Anaesthesia, Birmingham Women's Hospital, Birmingham Women's and Children's NHS Foundation Trust, Mindelsohn Way, Edgbaston, Birmingham, United Kingdom
Dr. Andrew C Leatherbarrow
Selwyn Crawford Department of Anaesthesia, Birmingham Women's Hospital, Birmingham Women's and Children's NHS Foundation Trust, Mindelsohn Way, Edgbaston, Birmingham - B15 2TG
Increasing numbers of congenital heart disease patients are now surviving to child bearing age. This complex patient cohort present a great challenge to the obstetric anaesthetist and multidisciplinary teams. This report describes a rare case of a patient undergoing emergency caesarean section at 31 weeks gestation due to decompensated critical congenital aortic stenosis. The physiological effects of pregnancy in patients with aortic stenosis are discussed along with principles of managing anaesthesia for caesarean section in the presence of this cardiac lesion.
|How to cite this article:|
Leatherbarrow AC. Management of emergency caesarean section in a patient with decompensated critical aortic stenosis.J Obstet Anaesth Crit Care 2018;8:50-53
|How to cite this URL:|
Leatherbarrow AC. Management of emergency caesarean section in a patient with decompensated critical aortic stenosis. J Obstet Anaesth Crit Care [serial online] 2018 [cited 2019 Dec 9 ];8:50-53
Available from: http://www.joacc.com/text.asp?2018/8/1/50/230061
Through significant advances in medical care over the last 50 years, the number of patients with congenital heart disease surviving to adulthood is consistently rising. This complex group of patients is now growing up and getting pregnant. These patients represent a growing challenge to the obstetric anaesthetist. This report describes a rare case of a patient undergoing emergency caesarean section at 31 weeks gestation due to decompensated critical congenital aortic stenosis.
A 17-year-old female was referred to our hospital during her second trimester due to a history of congenital aortic stenosis associated with a congenital bicuspid aortic valve. The patient had been under the care of the regional paediatric cardiology team since birth. Although the valve stenosis had been gradually worsening the patient had remained asymptomatic with normal exercise tolerance. Replacement with a bio prosthetic valve was planned once the patient reached adulthood. Following extensive risk counselling in relation to pregnancy, the patient became pregnant prior to surgical valve replacement. The patient had no other significant medical history. She smoked 10 cigarettes per day and had a healthy dentition.
From 20 weeks gestation, the patient began to complain of gradually worsening exertional dyspnoea. She was also unable to lie flat in bed and began sleeping on 3 pillows. Transthoracic echocardiography revealed critical aortic stenosis with a valve area of 0.58 cm 2 and a transvalvular gradient of 64 mmHg. Despite the severity of the aortic stenosis, the left ventricle remained normal with an ejection fraction of 72% with no evidence of hypertrophy or dilatation. The patient was commenced on Co-amilofruse 2.5 mg/20 mg following review by a cardiologist. Following the initiation of diuretic therapy the patient's symptoms improved. She was discharged with monthly review in a combined cardiology/obstetric clinic.
At 30 weeks gestation, the patient was admitted to the high dependency unit of the delivery suite due to severe dyspnoea at rest and episodes of palpitations. The patient had clinical and radiological features of pulmonary oedema. A 12 lead ECG was performed which showed a sinus tachycardia of 130 beats per minute. There were no other abnormalities present. A 24-hour ECG was recorded with a Holter Monitor. Monitoring revealed a persistent sinus tachycardia of 130 beats per minute but no evidence of tachyarrhythmias. There was one ventricular ectopic and one supraventricular ectopic recorded over a 24 hour period. The pulmonary oedema was treated with two 40 mg IV furosemide boluses administered over a 12 hour period and increasing the oral Co-Amilofruse to 5 mg/40mg daily. In view of the persistent tachycardia and palpitations Bisprolol 1.25 mg daily was started. This was increased to 2.5 mg daily after 1 week. A decision was made to admit the patient to hospital until delivery due to concerns about further decompensation.
At 31 weeks gestation, a decision was made to deliver the patient via emergency caesarean section due to worsening pulmonary oedema, which was refractory to IV furosemide boluses. The patient had a Pa02 of 10.4 kPa on room air and required supplemental oxygen via nasal cannula to maintain oxygen saturations greater than 92%. The multidisciplinary team considered it unsafe to allow the patient to continue further into pregnancy.
The team decided to deliver the patient by caesarean section in the cardiothoracic theatre suite with cardiopulmonary bypass facilities available. This was done for two reasons. Firstly, due to the severity of aortic stenosis and presence of decompensation there was a high risk of cardiovascular collapse during induction of anaesthesia. This would have required emergency sternotomy and institution of cardiopulmonary bypass to save the patient. Secondly, if the patient did require emergency bypass, the cardiothoracic surgical team planned to replace the aortic valve. This would avoid the need for further surgery.
Anaesthetic management began with premedication. The patient was administered Ranitidine 150 mg PO prior to transfer to theatre. Sodium citrate 30 ml PO was also administered immediately before pre-oxygenation and induction. Peripheral venous access was gained with the insertion of two 14G peripheral venous cannula. An arterial cannula was inserted into the left radial artery. A right internal jugular vein central venous catheter was also inserted. Initial arterial blood pressure was 130/60 and initial central venous pressure was recorded as 14 mmHg. Full standard monitoring including 3 lead ECG, pulse oximetry and waveform capnography was also utilised. A Transoesophageal echocardiography (TEE) probe was also available in theatre but we did not use TEE in this case as the patient remained haemodynamically stable. A modified rapid sequence induction was performed. A target controlled infusion of remifentanil was started (Minto model/effect side targeting) and gradually titrated up to an effect site level of 5 ng/ml. Remifentanil was used principally to prevent tachycardia as this would reduce cardiac output in the presence of aortic stenosis. This was followed by slow injection of sodium thiopental until loss of verbal communication with the patient. A total of 125 mg was required. Neuromuscular blockade was induced with Suxamethonium 100 mg. The patient had a grade I view on direct laryngoscopy and intubation was uneventful. Post induction, anaesthesia was maintained with a gas mix of 50% oxygen and 50% nitrous oxide with Sevoflurane at an end tidal concentration of 1.4%. A total of 50 mg of Rocuronium was given to maintain neuromuscular blockade throughout the case. Blood pressure was maintained with a noradrenaline infusion with additional intermittent boluses of phenylephrine. We use a 2 mg/50ml dilution of noradrenaline as standard in cardiac theatres. The infusion was started at 5 ml/hr (0.05 micrograms/kg/minute) at 5 minutes post induction as we had administered five 100 microgram boluses of phenylephrine in rapid succession to maintain blood pressure. Blood pressure then stabilised without requirement for further phenylephrine boluses.
Administration of the uterotonic drugs was the most challenging aspect of this case. It would be normal practice to administer oxytocin five units intravenously as a slow bolus post-delivery. Oxytocin, however, causes a significant reduction in systemic vascular resistance with could lead to cardiovascular collapse in a patient with aortic stenosis. We initially attempted to administer the five units as a slow infusion over 20 minutes however the uterine tone remained poor leading to on-going haemorrhage. We decided to administer the oxytocin in one unit incremental boluses. This still induced a significant fall in blood pressure requiring the noradrenaline infusion to be increased to 10 ml/hr (0.1 microgram/kg/minute) and a further four 100 microgram boluses of phenylephrine to be administered to maintain blood pressure. Total estimated blood loss for the case was 1.5 litres. The patient was transfused two units of red cells and received 1.5 litres of Hartmanns solution intra-operatively.
The patient remained stable throughout and was extubated a few hours later in the cardiothoracic critical care unit. The patient would have been suitable for extubation in theatre, however, it is policy in our institution to transfer all cardiac theatre patients to critical care for extubation due to the high theatre workload. The noradrenaline infusion was reduced to 5 ml/hr (0.05 micrograms/kg/minute) prior to transfer from theatre to critical care. This was rapidly weaned down and stopped when the blood pressure increased after switching off sedation in preparation for extubation. Post-op analgesia consisted of paracetamol 1g PO qds and Oramorph 10–20 mg PRN up to every 2 hours. The baby was delivered in good condition and transferred to the neonatal unit. Apgar scores were 7 at 1 minute and 8 at 5 minutes. The baby required a period of respiratory support with CPAP. One week post-partum, the patient was asymptomatic and had returned to normal exercise tolerance. She was scheduled for a valve replacement 6–12 months post delivery to allow sufficient bonding time with the baby.
Congenital aortic stenosis is rare, accounting for less than four per cent of congenital heart disease in the developed world. The stenotic lesion can be supra-valvular, valvular or infra-valvular. Valvular lesions are almost universally associated with a congenital bicuspid aortic valve.
Patients with aortic stenosis have a fixed cardiac output due to inability to increase stroke volume. Maintenance of systemic blood pressure and myocardial perfusion pressure is critically dependent on maintenance of systemic vascular resistance. Pregnancy presents a dangerous physiological challenge to this group of patients. This is firstly due to a progressive decline in systemic vascular resistance of between 30–70% in the second trimester. Secondly, cardiac output can increase up to 50% pre-pregnancy values. This leads to progressive cardiac decompensation which begins around 20 weeks gestation.
Regarding the perioperative management of this patient cohort, several key principles must be followed. Crucially, these patients will not tolerate hypotension. Invasive haemodynamic monitoring must be in place prior to commencing anaesthesia. This is to facilitate immediate detection of hypotension and enable rapid correction with vasopressors. We did not use TEE in this case, however, it does provide useful information about ventricular function and filling. For induction, cardiostable agents must be utilised. We used Remifentanil as our opioid of choice as it has several advantages in addition to being cardiostable. Remifentanil not only has a protective effect in aortic stenosis by preventing tachycardia but also its rapid clearance minimises any sedative effect on the neonate. Despite attempting to make induction as cardiostable as possible in our patient, the patient still became hypotensive a few minutes after induction. Systolic blood pressure fell to 85 mmHg. This was rapidly corrected by starting a noradrenaline infusion and administering boluses of phenylephrine until the noradrenaline took effect. Careful thought must be given to the choice of vasoactive drugs used in patients with aortic stenosis. The aim is to increase systemic vascular resistance whilst avoiding tachycardia. Tachycardia is detrimental in aortic stenosis as the reduces diastolic filling time and further worsens the already restricted cardiac output. For this reason, vasoactive drugs with strong positive chronotropic activity are best avoided. This includes dobutamine and ephedrine. Phenylephrine with its pure alpha agonist activity and noradrenaline, which has minimal beta agonist activity at lower doses are the agents of choice. There must also be careful consideration to the mode of anaesthesia. If there is a moderate to high risk of requiring sternotomy and emergency cardiopulmonary bypass then general anaesthesia is the preferred option. In this case, due to the severity of the aortic stenosis and features of decompensation there was a high risk of cardiovascular collapse and potential need for cardiopulmonary bypass. In view of this general anaesthesia was the most appropriate strategy in our patient. In patients with lesser degrees of valvular stenosis and who are haemodynamically stable with no features of decompensation, a neuraxial block may be considered as an alternative. Caesarean section requires a neuraxial block to the T4 dermatome. This may be achieved with a cautious epidural top-up; however, the block and associated vasodilatation will be irreversible for several hours. Epidural anaesthesia in the presence aortic stenosis is not without risk. Systemic vascular resistance will fall significantly, which may compromise the maternal myocardium and placental blood flow. Blood pressure must be tightly controlled to avoid hypotension. In addition, there is always a risk the epidural block may not be adequate for surgery. Tyagaraj et al. noted that the choice of anaesthetic technique must be made following a detailed haemodynamic assessment of the patient and not just based on severity of stenosis and transvalvular gradient. Although general anaesthesia is still considered the gold standard in patients with severe stenosis. Post-delivery, great care must be taken with the use of uterotonic drugs. Oxytocin causes a marked reduction in systemic vascular resistance and tachycardia. Overall, it is the safest uterotonic agent in the presence of aortic stenosis but must be administered as a slow infusion and never given as a bolus. The slow infusion does however increase the risk of post-partum haemorrhage. Ergometrine causes coronary artery vasospasm and can significantly impair myocardial perfusion in patients with aortic stenosis. This is especially true of left ventricular hypertrophy is present. It is therefore best avoided in patients with aortic stenosis.
Fortunately, in this case, the patient remained stable and did not require sternotomy and institution of emergency cardiopulmonary bypass. In event of this occurring attention must be given to the management of bypass. Datt et al. describe the management of a patient undergoing a dual procedure of caesarean section and aortic valve replacement on bypass. They advise number of recommendations, which include the avoidance of hypothermic perfusion and haemodilution, maintaining pump flow rates >2.5 L/min/m2 and maintenance of mean arterial blood pressure above 70 mmHg.
This case illustrates the challenge posed by this unique cohort of obstetric patients. Adopting a multidisciplinary approach and careful planning are the key to successful management. Maintenance of systemic vascular resistance and avoidance of hypotension are crucial in order to maintain cardiovascular stability and achieve a safe delivery in theatre.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
|1||Van der Bom T, Mulder BJ, Meijboom FJ, van Dijk AP, Pieper PG, Vliegen HW, et al. Contemporary survival of adults with congenital heart disease. Heart 2015;101:1989-95.|
|2||Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890-900.|
|3||Gelson E, Gatzoulis M, Johnson M. Valvular heart disease. BMJ 2007;335:1042-5.|
|4||Brown J, Morgan-Hughes NJ. Aortic stenosis and non-cardiac surgery. Contin Educ Anaesth Crit Care Pain 2005;5:1-4.|
|5||Whitfield A, Holdcroft A. Anaesthesia for caesarean section in patients with aortic stenosis: The case for general anaesthesia. Anaesthesia 1998;53:109-12.|
|6||Tyagaraj K, Gutman DA, Belliveau L, Sadiq A, Bhutada A, Feierman DE. A Multidisciplinary Approach to Anesthetic Management of a Parturient with Severe Aortic Stenosis. Case Rep Anesthesiol 2015;2015:489157.|
|7||Tamhane P, O'sullivan G, Reynolds F. Oxytocin in parturients with cardiac disease. Int J Obstet Anesth 2006;15:332-3.|
|8||Burt CC, Durbridge J. Management of cardiac disease in pregnancy. BJA CEACCP 2009;9:44-7.|
|9||Datt V, Tempe DK, Virmani S, Datta D, Garg M, Banerjee A, et al. Anesthetic management for emergency cesarean section and aortic valve replacement in a parturient with severe bicuspid aortic valve stenosis and congestive heart failure. Ann Card Anaesth 2010;13:64-8.|