Abstract

Dexmedetomidine has predictable, complex, and negative cardiovascular effects that lead to additional adverse effects such as bradycardia and hypotension in up to 42% of patients and might cause profound left ventricular dysfunction and refractory shock. Usually, these temporary effects can be successfully counteracted with atropine, ephedrine, and volume supplementation. Clinicians need to be well informed about the potential of dexmedetomidine to cause bradycardia, which may progress to pulseless electrical activity, particularly in patients older than 50 years and patients with cardiac abnormalities. Here, we report the clinical characteristics of six patients who were scheduled for various neurosurgical procedures within a period of three months and suffered from cardiac arrest following dexmedetomidine administration. We urge clinicians to take caution against the negative effects of dexmedetomidine, especially when it is used in patients older than 50 years with underlying cardiac disease and in combination with cardiodepressant drugs.

Keywords

dexmedetomidine; fentanyl; heart arrest; propofol;

1. Introduction

Dexmedetomidine, like clonidine, is a potent alpha-2-adrenoceptor agonist with an 8-fold higher affinity for the alpha-2-adrenoceptor than clonidine. It causes a decrease in heart rate and a dose-dependent and predictable decrease in arterial blood pressure following an initial transient increase in pressure.¹ However, dexmedetomidine does not appear to have any direct effects on the myocardial contractility.² The incidence of bradycardia has been reported to be as high as 42% in patients who receive dexmedetomidine.³ There are also a few case reports describing cardiac arrest during the intraoperative period.4, 5, 6, 7, 8, 9 Here, we report six instances of intraoperative cardiac arrest following dexmedetomidine infusion.

2. Case reports

Here, we report the intraoperative events of six patients who were scheduled for neurosurgery during the 3-month period from February to April, 2011 at the neurosurgical operation theater (OT) of a tertiary medical college. These patients suffered cardiac arrest while under general anesthesia consisting of the combination of dexmedetomidine, fentanyl, and propofol. A total of 71 patients who underwent various neurosurgical procedures during the same period received dexmedetomidine as a part of the anesthetic regime. The clinical characteristics of these six patients who suffered cardiac arrest are shown in Table 1. All patients were 50 years of age or older, except one patient (case 3) who was 15 years old. Coincidentally, all patients were male and their body weights ranged from 46–70 kg. Four patients were classified as ASA physical status grade I. Case 2 was hypertensive, but he was successfully treated using a single drug (5 mg amlodipine). Case 4 had history of hypertension and diabetes mellitus for the last 12 years and underwent a post-CABG operation 9 years prior. He was prescribed an oral hypoglycemic along with an angiotensin converting enzyme (ACE) inhibitor for the control of blood glucose and hypertension, respectively. He was also on anti-anginal medications and was asymptomatic for the last 5 years with good exercise tolerance (New York Heart Association grade II). A recent echocardiograph of his heart showed no significant abnormalities and an ejection fraction of 56%. None of the patients had any history or electrocardiograph (ECG) evidence suggestive of heart block.

Table 1. Characteristics of patients who suffered a cardiac arrest after receiving an infusion of dexmedetomidine.

Download full-size image

A routine preanesthetic check-up and risk assessment were performed in all six patients. Proper optimization of the patients was performed before they underwent surgery. A standard anesthesia technique was followed for each case, as per our institutional protocol. Written informed consent was obtained from every patient before the surgical and the anesthetic procedures, and the risks related to the procedures were thoroughly explained to the patients and their escorts.

In case 1, in anticipation of difficult intubation, an awake fiberoptic bronchoscopy-guided intubation was planned. In cases 2, 3, and 4, tracheal intubation was performed in the usual manner after induction of general anesthesia. In cases 5 and 6, a laryngeal mask airway (LMA) was inserted after induction with propofol in order to deliver general anesthesia. Noninvasive blood pressure (NIBP), ECG, SpO2, heart rate, end-tidal carbon dioxide, and temperature were monitored. In case 3, central venous pressure, invasive arterial blood pressure, and peripheral neuromuscular blockades were also monitored.

Before the induction of anesthesia, all patients received a bolus intravenous injection of 1 μg/kg dexmedetomidine over a period of 10 minutes followed by a maintenance infusion that was delivered at a rate of 0.5 μg/kg-hour. Glycopyrrolate (0.2 mg), 2 μg/kg fentanyl, and 1 mg midazolam were also intravenously administered before induction.

In case 1, after proper positioning of the tracheal tube by fiberoptic bronchoscopy, 2 mg/kg propofol and 0.9 mg/kg rocuronium were intravenously administered. In cases 2, 3, and 4 following premedication, induction was performed using 2 mg/kg propofol and 0.9 mg/kg rocuronium followed by tracheal tube placement. In cases 5 and 6, LMA was inserted after induction with 2 mg/kg propofol. In all cases, induction of anesthesia was uneventful and proper placement of the airway devices was confirmed by capnography.

In all cases, the operative procedures were performed in the supine position, except case 1 who was operated on in the prone position. In case 1, cardiac arrest occurred before the patient was turned prone for the operation; thus, there was no problem performing cardiopulmonary resuscitation (CPR) despite cancellation of the surgical procedure. In the other cases, the surgical procedures were uneventfully continued after successful resuscitation following cardiac arrest. Anesthesia was maintained with an infusion of propofol (at a rate of 10 mg/kg-hour), dexmedetomidine (0.5 μg/kg-hour), and N₂O in O₂; top-up doses of fentanyl (1 μg/kg) and rocuronium (0.15 mg/kg) were administered as necessary; in cases 5 and 6, neuromuscular blocking drugs were not administered.

Cardiac arrest occurred before surgical incision, except cases 5 and 6 who underwent cardiac arrest at 20 and 30 minutes after the induction of anesthesia, respectively, and after the surgical procedures had already started. There was no significant surgical blood loss or episode of acute hypotension in cases 5 and 6. Significant bradycardia was noted prior to cardiac arrest in case 2, and in cases 1, 3, and 4 acute hypotension preceded cardiac arrest. In cases 1 and 4, pulseless electrical activity was noted, and in the other cases asystole was observed. Immediate CPR was started, as per advanced cardiovascular life support (ACLS) protocol.¹⁰ All patients were restored to life within 10 minutes after initiation of CPR. Five mg/kg amiodarone was used to control ventricular tachycardia in case 1. Cases 1, 2, and 4 required inotropic support (infusion of 10 μg/kg-minute dopamine) for 24 hours in order to maintain systolic blood pressure above 90 mmHg. Postoperative mechanical ventilatory support was not required in every case. Reversal of neuromuscular blockades was accomplished via the intravenous administration of 0.05 mg/kg neostigmine and 10 μg/kg glycopyrrolate.

The postoperative period was uneventful for all patients, and they were all peacefully discharged, except case 1 who was rescheduled for surgery after 7 days. However, case 1’s surgical and anesthetic procedures (propofol and dexmedetomidine were avoided) were uneventful.

3. Discussion

Dexmedetomidine can cause biphasic cardiovascular response. The administration of a bolus of 1 μg/kg dexmedetomidine initially causes an increase in blood pressure for a brief period, accompanied by a reflex decrease in heart rate and subsequent decrease in blood pressure during later part of the infusion.¹ However, the heart rate remains lower than the baseline level. Dyck et al found a 22% rise in the mean arterial pressure and 27% decline in the heart rate within 4–5 minutes after starting dexmedetomidine infusion.¹¹ However, they infused dexmedetomidine at a high dosage of 2 μg/kg over 5 minutes. Stimulation of alpha-2B-adrenoceptors in the vascular smooth muscle results in an initial increase of blood pressure. Slowing the infusion over 10 minutes or more can reduce the severity of the initial reaction but cannot abolish it totally, which was shown in the study by Hall et al; specifically, that a slow infusion of dexmedetomidine over 10 minutes could still increase the mean arterial pressure by 7% along with a 16–18% decrease in the heart rate.¹² This initial response, which usually lasts 5–10 minutes, is followed by a prolonged decrease (10–20%) in blood pressure with persistent bradycardia. Hypotension and bradycardia peak about 1 hour after the commencement of infusion.¹ Both effects result from the inhibition of the central sympathetic outflow. Stimulation of the presynaptic alpha-2-adrenoceptor by dexmedetomidine decreases norepinephrine release, which is considered another possible mechanism of the subsequent effects. It causes a dose-dependent decrease in the concentration of norepinephrine in plasma and can reduce norepinephrine release by up to 92% in young healthy volunteers.¹³ Hammer et al studied the effects of dexmedetomidine on cardiac conduction in pediatric patients and found that both sinus and atrioventricular (AV) nodal functions can be depressed.¹⁴ However, the baroreceptor reflex could be well preserved and the reflex augmented by pressor stimuli, according to the findings by Ebert et al.¹⁵

There are few case reports describing dexmedetomidine-induced bradycardia, which can culminate in cardiac arrest. Ingersoll-Weng et al first reported such an event in a 52-year-old woman with myasthenia gravis who was scheduled for a thymectomy and excisional biopsy of a mass in the right lung via median sternotomy; this patient eventually developed cardiac asystole following the infusion of dexmedetomidine.⁴ Shah et al described a 76-year-old woman who developed cardiac arrest from the use of dexmedetomidine during the extraction of a pacemaker lead.⁵ Sichrovsky et al reported another incident of cardiac arrest following the infusion of dexmedetomidine into a 50-year-old man for ablation of paroxysmal atrial fibrillation.⁶ However, the role of dexmedetomidine in the cardiac arrest was questioned by others. Gerlach and Murphy managed a case of a 74-year-old man who developed bradycardia followed by pulseless electrical activity upon receiving a dexmedetomidine infusion at the rate of 0.7 μg/kg-hour.⁷ Normal pulse activity was restored after 2 minutes of chest compressions and an intravenous bolus of 0.4 mg atropine. However, he suffered a postoperative myocardial infarction before dexmedetomidine infusion could be started. Nagasaka et al described an event in which a 64-year-old woman with hypertension, diabetes mellitus, and asymptomatic first-degree AV block, who was undergoing low anterior resection of the rectum, developed complete AV block followed by cardiac arrest upon receiving dexmedetomidine.⁸ Zhang et al described repeated dexmedetomidine-induced episodes of bradycardia and asystole in an 18-year-old double-lung transplant recipient.⁹ Discontinuation of the drug restored a regular sinus rhythm. Five of these six patients mentioned above were more than 50 years old and five had a cardiovascular system abnormality. Also in this case series, five of the six patients were more than 50 years of age. One patient had previously undergone coronary artery bypass graft surgery and was suffering from hypertension and diabetes, and one patient had isolated hypertension. The other four patients did not have any underlying systemic disorders. There are certain neurosurgical procedures, such as neuroendoscopy and the placement of extradural drains, that can produce bradycardia.¹⁶ The surgical procedures performed on our patients are not found to be associated with bradycardia. Space-occupying lesions that involve or compress the brain parenchyma, such as subdural hematoma and tumors, may induce bradycardia as a result of Cushing reflex, which can be found even before the operative procedure. Cushing reflex was preoperatively absent in our patients.

The sympatholytic action of dexmedetomidine could be deleterious to hypovolemic patients or patients with a fixed stroke volume. Preoperative screening and proper hydration may prevent such complications. Adverse cardiovascular events are generally found during the initial loading of dexmedetomidine. Also, four of our patients who sustained cardiac arrest within 10 minutes of anesthetic induction, received a loading dose of dexmedetomidine prior to the induction. We continued the maintenance dose of dexmedetomidine because we did not suspect dexmedetomidine as an etiologic agent for cardiac arrest. However, those patients did not suffer recurrent attacks of bradycardia or cardiac arrest during the maintenance period of dexmedetomidine. In these patients, anesthesia was induced using propofol and fentanyl, which might have potentiated the bradycardic action of dexmedetomidine (both propofol and fentanyl cause bradycardia). In the study by Peden et al, in which dexmedetomidine was combined with propofol for the induction of anesthesia in adults, two of their first four patients had brief and self-limiting sinus arrest after laryngoscopy.¹⁷ Initially, dexmedetomidine was administered as a bolus dose of 1 μg/kg over 15 minutes followed by an infusion of 0.4 μg/kg-hour prior to anesthetic induction with propofol. Later, the protocol was amended by decreasing the loading dose to 0.7 μg/kg dexmedetomidine over 15 minutes, followed by an infusion of 0.27 μg/kg-hour, which was found to be free of such adverse events. Therefore, the dose of dexmedetomidine may be adjusted when it is used along with other negative chronotropic and inotropic agents. Etomidate or thiopental sodium should be tried for induction in order to decrease adverse reactions. A randomized controlled study may be necessary to determine the loading and maintenance doses of dexmedetomidine that can be used in conjunction with propofol and fentanyl.

Therefore, dexmedetomidine must be used with exceptional caution, particularly when negative chronotropic and inotropic drugs are used concomitantly. Preferably, dexmedetomidine should be avoided in elderly patients who have an underlying cardiac disease or are at risk of developing cardiac complications. In such circumstances, relatively safe sedatives such as benzodiazepines may be used.¹⁸