Cardiac Arrest Upon Induction of General Anesthesia

Abstract

There are numerous causes of cardiac arrest in the perioperative period, including hypoxia, hypovolemia, and vagal response to medications or procedures during routine anesthetics. Initiation of adequate cardiopulmonary resuscitation, administration of epinephrine, and application of a defibrillator, with shocking when applicable, are all essential steps in achieving return of spontaneous circulation. Knowledge and utilization of monitoring equipment can alert the provider to problems leading to cardiac arrest as well as ensure proper resuscitative efforts during the event. Polypharmacy is quite common with many of today's special needs patients. It is important to understand the medications they are taking as well as the potential interactions that may occur with drugs given during sedation and general anesthesia. The following is a case report of cardiac arrest including asystole and pulseless electrical activity in a 27-year-old man with autism and behavioral problems who presented for restorative dentistry under general anesthesia in the ambulatory surgery setting.

Cardiac arrest in the perioperative period can be caused by many mechanisms, including hypoxia, hypovolemia, drug interactions, and vagal response to surgical stimulation medications or procedures. Essential steps in managing arrest include initiation of adequate cardiopulmonary resuscitation (CPR), administration of epinephrine, and use of defibrillator when applicable if return of spontaneous circulation is to occur. Utilization of standard monitoring equipment supported by proper knowledge of monitor interpretation will alert the perioperative anesthesia provider of events leading to arrest and instruct for proper resuscitative efforts during the event. In the case presented below, there were a multitude of factors that likely contributed to the eventual pulseless electrical activity arrest.

CASE PRESENTATION

A 27-year-old Caucasian man (height 175 cm [5 ft 9 in], weight 91 kg [200 lb], and body mass index 29.5 kg/m²) presented with his mother for his scheduled full-mouth dental rehabilitation under general anesthesia in the dental clinic. The patient had previously undergone general anesthesia for dentistry in 2013 and 2015 and tolerated the procedures well. His medical history was significant for autism spectrum disorder (approximate mental development of a 5-year-old) with behavioral problems, grand mal seizures, asthma, complex Tourette disorder, obsessive-compulsive disorder, gastroesophageal reflux disease, and neurogenic bladder, all of which were well controlled with his medications. He also had an episode of serotonin syndrome, which was attributed to mismanagement of his home medications 2 years prior, but this information was not disclosed to the anesthesia team until after the incident. The patient's last seizure was 9 months prior to the appointment. His home medications consisted of the following: albuterol inhaler (twice a day [BID] as needed), fluvoxamine (150 mg, 3 times a day [TID]), paroxetine (40 mg, every night at bedtime [qhs]), extended-release guanfacine (1 mg, TID), haloperidol (5 mg, BID), olanzapine (10 mg, BID), ziprasidone (60 mg, BID), lacosamide (100 mg, BID), lamotrigine (200 mg, BID), zonisamide (100 mg, BID), propranolol (20 mg, TID), black cohosh (qhs), marshmallow liquid (qhs), juniper tar oil (qhs), hemp oil (qhs), and melatonin (3 mg, qhs). Because of his large size and combative behavior, his mother was instructed to give his normal morning medications with a sip of water 2 hours prior to his scheduled appointment time. The patient had no reported known drug allergies.

Upon presentation, a preoperative assessment revealed no new findings, and the patient was appropriately NPO, having taken his medications as instructed. The patient was transported into the operatory, and standard American Society of Anesthesiologists monitors were applied. An intramuscular injection of ketamine (300 mg) was administered into the patient's right deltoid secondary to poor compliance/inability to cooperate for intravenous (IV) cannulation. His vital signs after onset of the intramuscular ketamine were heart rate 76 bpm, end tidal carbon dioxide (EtCO₂) 40 mm Hg, and systolic blood pressure 120 mm Hg. Supplemental oxygen (12 L/min) was administered via the anesthetic face mask. Sevoflurane (1.9%) was initiated ∼5 minutes after the ketamine to further deepen the patient during placement of a 22-gauge IV catheter in the dorsum of the left hand. His heart rate had dropped to 57 bpm, so the sevoflurane was discontinued. In addition, IV boluses of glycopyrrolate (100 μg), fentanyl (50 μg), and midazolam (1 mg) were administered. The patient's heart rate stabilized at 58 bpm, and anesthesia was induced with an IV bolus of propofol (100 mg). The blood pressure cuff was cycled and read a systolic of 106 mm Hg, while his heart rate remained stable at 57 bpm. An additional IV bolus of propofol (50 mg) was administered after noting a lack of apnea. Airway patency was confirmed, and IV succinylcholine (200 mg) was administered for paralysis, facilitating easy mask ventilation. Directly after administration of succinylcholine, his heart rate began to decline to 45 to 49 bpm. Additional glycopyrrolate (100 μg) was administered while atropine was prepared; however, before the atropine could be administered, the recycled blood pressure produced a systolic of 76 mm Hg. At that point, the decision to use atropine was aborted, IV epinephrine (10 μg) was administered, and IV fluids were fully opened. The heart rate continued to decline to 40 to 44 bpm, so 2 additional IV boluses of epinephrine (10 μg) were given. When the heart rate dropped to 36 bpm, an IV bolus of epinephrine (50 μg) was administered, but the heart rate continued to decline to 22 bpm. A pulse check was performed revealing no radial or carotid pulses, which was rapidly confirmed by a second provider.

CPR was initiated immediately while calls for additional assistance and 911 were made. An attempt at oral intubation was made during chest compressions but was unsuccessful. Upon arrival of additional help with a crash cart, automated external defibrillator pads were applied, and a rhythm check was performed. However, the patient was in asystole, so no shock was warranted. There were no pulse oximetry or capnography waveforms present. Chest compressions immediately resumed, and an IV bolus of epinephrine (1 mg) was administered followed by a 10-mL saline flush. Oral intubation was attempted a second time by a different provider and was again unsuccessful likely because of ongoing chest compressions. A flexible size 5 laryngeal mask airway (LMA) was placed to facilitate asynchronous ventilation and continuous chest compressions. At this point, the heart rate was 117 bpm, coinciding with the rate of chest compressions, and the EtCO₂ was 37 mm Hg, reflecting the adequacy of CPR.

Emergency medical services arrived 8 minutes after the 911 call. They immediately performed a pulse check, confirmed lack of a pulse, and applied their monitors, which confirmed asystole. Concurrently, another round of epinephrine (1 mg) was administered followed by a 10-mL flush and the resumption of CPR. Three minutes after emergency medical services arrival and 15 minutes after the start of CPR, return of spontaneous circulation (ROSC) was achieved as confirmed by the presence of a carotid pulse, heart rate of 76 bpm, SpO₂ of 99%, and positive EtCO₂. The flexible LMA was exchanged for a size 5 iGel LMA, and the patient was transported to the emergency department. The total time the patient spent in asystole and pulseless electrical activity was approximately 15 minutes.

Postemergency Care

Upon arrival to the emergency department, an initial blood pressure measurement was unable to be obtained despite palpable central pulses. However, after a bolus of phenylephrine (100 μg), the mean arterial pressure improved to 70 mm Hg. The patient withdrew to painful stimuli on examination. He was intubated after receiving fentanyl and an additional dose of succinylcholine (120 mg). His body temperature was measured as 97.5°F Differentials at that time included underlying cardiac abnormality (such as long QT or structural disease), toxins (serotonin syndrome or neuroleptic malignant syndrome), and hypotensive mediated arrest. Labs were drawn and were grossly normal aside from the elevated troponins, which was to be expected following CPR (Table). Although the patient was hemodynamically stable, he was admitted to the intensive care unit for further monitoring and a cardiac workup. He remained sedated and intubated during his evaluations, as there were concerns regarding his lack of compliance otherwise. An electrocardiogram (Figure 1) and transthoracic echocardiogram (Figure 2) were performed within 6 hours of the event but demonstrated no abnormal findings. Chest radiograph and computed tomography of the head also showed no abnormal findings. He was extubated that night after his workup came back negative. He was discharged 2 days later with no deficits.

Table.

Laboratory Results With Grossly Negative Findings*

Chemistry	Value	Hematology (CBC)	Value
Sodium	142	WBC	12.72
Potassium	4.2	RBC	4.97
Chloride	110 (high)	Hemoglobin	15.0
Carbon dioxide	21 (low)	Hematocrit	44.1
BUN	16	Platelet count	153
Creatinine serum	1.08
BUN/creatine ratio	15
Estimated GFR	>60
Ionized calcium	4.54 (low)
ALT	403 (high)
AST	253 (high)
Troponin†	0.48 (high)	0.41 (high)	0.28 (high)

^*ALT indicates alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CBC, complete blood cell count; GFR, glomerular filtration rate; RBC, red blood cell; and WBC, white blood cell.

^†Elevated troponin likely due to cardiopulmonary resuscitation.

Figure 1.

Electrocardiogram with normal sinus rhythm.

Figure 2.

Transthoracic echocardiogram results. Normal left ventricular size and function, mild hypokinesis in basal segments. Ejection fraction 55–60%, normal right ventricular size and function, no hemodynamically significant valvular abnormalities.

DISCUSSION

Cardiac Arrest in the Perioperative Period

Cardiac arrest in the perioperative period is vastly different from arrests outside of the hospital, as it is almost always witnessed, and the personnel involved with treatment are typically trained to handle such situations. Vagal responses from stimulation such as laryngoscopy, vagotonic anesthetics, sympatholytic anesthetic agents, and β-blockers are common causes of bradycardia, which can result in cardiac arrest if left uncorrected. Hypoxia, generally resulting from airway loss, is another common cause of arrest, while hypovolemia, often compounded by hemorrhage, can result in pulseless electrical activity.¹ The National Anesthesia Clinical Outcomes Registry recently revealed that the incidence of cardiac arrest associated with anesthesia is approximately 5.6 per 10,000 cases and that the risk of cardiac arrest increases with increased age and American Society of Anesthesiologists classification.¹ For patients who experienced arrest within 24 hours of the surgery, asystole was the most common arrest rhythm, with a survivorship of 30.5–80%.¹

The first steps when treating a patient in crisis, such as cardiac arrest, is to identify and correctly diagnose the problem, to initiate proper corrective measures. Potential cardiac arrest signs can include electrocardiogram abnormalities or dysrhythmias, hypotension or an apparently nonfunctioning blood pressure cuff, and loss of capnography and pulse oximetry waveforms. EtCO₂ is typically seen as the most reliable monitor indicating a crisis and cardiac arrest.¹ Although it would appear easy to identify these issues, there are many factors that can confound the situation and obscure the presence of a problem from providers. These include the inability to monitor the patient's mental status, use of controlled ventilation, patient positioning, and coverage of the patient's body with surgical drapes. Sudden, severe bradycardia is often caused by physical manipulation of the patient (ie, laryngoscopy) triggering an increase in vagal tone, which can be potentiated by vagotonic/sympatholytic anesthetic agents.¹ When arrest or lethal arrhythmia is suspected, no more than 10 seconds should be taken to assess for a pulse before initiating chest compressions at a rate of 100–120 per minute and a depth of 5 cm (2 in.) in adults, ensuring full chest recoil to optimize preload.^2,3 Interruptions in CPR should be minimized as much as possible. An EtCO₂ of 20 mm Hg or higher during chest compressions indicates adequate CPR efforts and is associated with improved survival. If CPR is initiated within 4 minutes of cardiovascular collapse, the chance of survival to discharge from the hospital doubles.⁴ Capnography is typically the most reliable indicator of return of spontaneous circulation, with a sudden rise in EtCO₂ (>35–40 mm Hg in an intubated patient), suggesting return of spontaneous circulation has been achieved.¹ In this case, the patient was generally anesthetized, paralyzed, and in a semireclined position, all of which were likely contributing factors. However, CPR in this case was performed well despite the slight delay (>10 seconds) before starting compressions, which was attributed to 2 separate providers palpating for a pulse. During attempts at resuscitation, there were several intervals captured on the monitor in which the heart rate was noted to be between 110 and 120 bpm, revealing an adequate pace of compressions.

Hyperventilation during cardiac arrest should be avoided. A rate of 10 bpm is preferred to 20 bpm as the survival rate is higher due to the reduced time spent with increased intrathoracic pressures secondary to higher ventilation rates, tidal volumes, and inspiratory times. With the potential for delayed chest compressions, these factors are inversely proportional to coronary artery perfusion.¹ Emergent placement of an LMA is an acceptable alternative should intubation not be possible.⁴ The compression-to-ventilation ratio should not exceed 30:2 in a patient without a definitive airway. For patients with a definitive airway, the ventilation rate should be no more than 10 bpm, with a total inspiratory time of 1 second and a tidal volume producing a chest rise or approximating 500 mL.^1,4 Because positive-pressure ventilation decreases venous return and hypoventilation has less severe consequences, it may be reasonable to ventilate patients to an SpO₂ of 90% or greater.¹ In this case, oral intubation was attempted via video laryngoscopy twice, first by a resident and then by an attending, but neither attempt was successful. While a flexible LMA was successfully placed, having a back-up iGel would have been advisable, as it is easier to seat and secure after placement.

According to the 2015 American Heart Association Advanced Cardiovascular Life Support Bradycardia algorithm, the treatment of symptomatic bradycardia includes atropine, transcutaneous pacing, and/or epinephrine. Our initial thought was to administer IV atropine (0.5 mg); however, that was abandoned when the recycled blood pressure demonstrated a systolic of 76 mm Hg. It was felt that atropine would address only the bradycardia while epinephrine would address both the declining heart rate as well as the hypotension. Pacing was not attempted, as the patient was in pulseless electrical activity arrest when the crash cart arrived. Epinephrine is a nonselective adrenergic agonist that displays preferential binding at different dosages. In low doses, epinephrine has a higher predilection for the β-receptors in the heart and smooth muscle.⁸ At higher concentrations, epinephrine preferentially binds to α-receptors, which can dominate the vasodilatory effects of β₂-receptor stimulation, producing a net result of vasoconstriction. The overall physiologic effect of epinephrine administration is variably affected by the relative amounts of α- and β-receptor activation produced.⁹ Although the initial bradycardia and hypotension failed to respond adequately to epinephrine in small boluses (10 μg), systemic vasoconstriction due to larger doses (1 mg IV) was sufficient to support CPR efforts.

Contributing Factors

Perioperative bradycardia, asystole, and pulseless electrical activity have 16 potential causes (8H's and 8T's) that can occur individually or work synergistically to worsen the hemodynamic status of the patient. Electrolyte abnormalities, acidosis, cardiac tamponade, hyper- and hypothermia, thrombus, and tension pneumothorax were ruled out upon admission to the hospital emergency department. However, the cardiac arrest occurring in this case was most likely multifactorial.^1–4 Hypovolemia was a likely causative factor, as the patient had been NPO for more than 15 hours, excluding his morning medications taken with a small sip of water. Medication overdose was another potential cause, specifically propranolol and guanfacine, as the patient had a history of serotonin syndrome attributed to overmedication for behavioral problems by his caregiver. Hypoxemia was unlikely a contributing cause, as the patient was verified to have an end-tidal O₂ of 90% prior to administration of propofol.

Nonselective Beta-Blockers

Nonselective beta-adrenergic receptor blocking agents such as propranolol antagonize β₁-receptors in the heart as well as the β₂-receptors in the lungs.⁷ Classically, these drugs are prescribed for the treatment of hypertension, angina, and arrhythmias. As opposed to selective beta-blockers, which have become more widely used, nonselective agents are generally less common for hypertension management and can cause bronchoconstriction via β₂-blockade in the lungs.^7–9 However, propranolol is now being prescribed for behavioral management in some patients with special needs as well as for migraine headache prevention because of its antagonism of the β₂-receptors in the cerebral vasculature, which may reduce migraines triggered by vasospasm. Beta-blockers will typically exhibit little effect on the heart during relaxed states but have more of an impact when there is increased sympathetic activity such as during times of stress or exercise. Twice-daily administration of propranolol is sufficient to produce antihypertensive effects in some patients, which may have contributed in this case.¹⁰

Guanfacine

Guanfacine is an α_2A-agonist used predominantly for the treatment of attention-deficit hyperactivity disorder in children. It has 15–20 times higher affinity for α_2A-receptor subtype compared with the α_2B- and α_2C-receptor subtypes; clonidine has similar affinity for all 3 receptor subtypes.¹¹ Originally, Food and Drug Administration approval for guanfacine was for the treatment of hypertension, as the mechanism of action is selective stimulation of the α₂-adrenoreceptors in the brainstem, which causes reduced sympathetic outflow, resulting in decreased vasomotor tone and heart rate.¹² Common adverse reactions to the medication include sedation (76.8%), bradycardia (30%), and hypotension (25.8%).^11,13 The manufacturer's drug label advises patients to avoid dehydration while taking guanfacine; it also cautions against the coadministration of antihypertensive medications.¹¹ The treatment of attention-deficit hyperactivity disorder is achieved through α₂-agonist stimulation regulating the subcortical activity in the prefrontal cortex, improving attention and impulsivity.¹² Dosage recommendations are 1–4 mg 4 times a day, and the dosage is dependent on patient response and tolerability.¹¹ Dose-dependent decreases in heart rate may be observed during the first 12 hours after administration when blood plasma has maximal concentration. Studies have shown that a 4-mg dose can decrease heart rate by 13 bpm.¹¹ Stimulation of α₂-receptors within the central nervous system has a sympatholytic effect caused by suppression of norepinephrine through a negative-feedback loop. In turn, selective α₂ stimulation can produce reductions in heart rate and blood pressure.⁵ If taken in conjunction with CYP3A4 receptor inhibitors, the plasma concentration of guanfacine will be increased, increasing the risk of events such as hypotension and bradycardia. The patient was reportedly taking fluvoxamine 150 mg (TID), a moderate CYP3A4 inhibitor.

Fentanyl and Propranolol

It has long been demonstrated that lung tissue takes up a variety of endogenous and exogenous substances via pulmonary circulation. The lungs are unique in that they receive all the cardiac output and possess a large capillary surface area that permits a high degree of drug diffusion from the blood plasma into the tissues. One particular chemical property that would promote lung tissue uptake includes basic amines with a pKa greater than 8, with moderate to high lipid solubility.¹⁴ First-pass uptake by the pulmonary tissues of basic lipophilic amines including lidocaine, propranolol, fentanyl, and meperidine have been demonstrated.¹⁴ Roerig et al¹⁴ demonstrated that for patients on no other drugs, the first-pass uptake of fentanyl into the lungs was 80%, while for patients taking propranolol (average 30–120 mg/d), the first-pass uptake decreased to 52%. The concurrent use of fentanyl and propranolol may have been a contributing factor in this patient.

Succinylcholine and Bradycardia

Succinylcholine is a depolarizing muscle relaxant associated with bradycardia with repeated dosing.¹⁶ Asystole after a single dose of succinylcholine is a rare occurrence but has been reported. The patient in this case was taking propranolol, which could have potentiated the bradycardic response to succinylcholine. In addition, he was given fentanyl (50 μg), which can cause vagomimetic effects and enhance bradycardia.¹⁷ Pretreatment with glycopyrrolate typically mitigates the bradycardia associated with fentanyl; however, that was not the case with this patient.¹⁸ Although the mechanism of bradycardia caused by succinylcholine is not fully understood, there are several possible mechanisms, including (1) biphasic action similar to acetylcholine, (2) direct stimulation of the carotid baroreceptors resulting in a reflex bradycardia, and (3) accumulation of acetylcholine due to competition for cholinesterase sites in the presence of succinylcholine.^16,19 Measures can be taken to prevent or limit bradycardia associated with succinylcholine. Administering a small dose of a nondepolarizing agent (ie, rocuronium) 3 minutes prior to succinylcholine has been shown to reduce bradycardia that occurs with repeat doses.¹⁹ Alternatively, anticholinergics can be used either intramuscularly, during premedication, or intravenously, prior to succinylcholine. Doses of atropine 6–10 μg/kg are insufficient to prevent bradycardia, while doses of atropine of 15 μg/kg IV 5 minutes prior to induction have been shown to be effective.¹⁹

SUMMARY

Cardiac arrest in the perioperative period can occur for numerous reasons. Common causes include hypoxia, hypovolemia, and increased vagal activity due to medications routinely used during general anesthesia or surgical stimulation. The patient in this report was taking 2 medications, propranolol and guanfacine, which likely contributed to the bradycardia and eventual cardiac arrest. There is a drug-drug interaction with propranolol and fentanyl regarding uptake by the lungs, essentially increasing the blood plasma concentration of fentanyl. In addition, bradycardia associated with administration of IV succinylcholine has been known to occur and can be limited by premedication with atropine. Prompt recognition of cardiac arrest, initiation of adequate CPR, administration of epinephrine, and application/use of a defibrillator when applicable are all essential steps in achieving return of spontaneous circulation and an optimal outcome. Knowledge and utilization of proper monitoring equipment can alert the anesthesia provider to problems that may lead to cardiac arrest as well as guide proper resuscitative efforts during the event, such as obtaining an EtCO₂ of 20 mm Hg or higher during CPR.