Brugada syndrome (BrS) is a rare genetic autosomal dominant disease with incomplete penetrance which affects ion channels of the cardiac conduction system and causes a coved ST segment elevation in the right precordial leads with a pseudo right bundle branch block pattern. Patients afflicted with this disease can develop malignant ventricular arrhythmias. This disorder raises specific concerns as anesthesiologists routinely administer drugs that interact with cardiac ion channels which theoretically could trigger the development of malignant arrhythmias. Propofol infusions have induced Brugada-like electrocardiogram (ECG) abnormalities.1-6 The administration of sodium channel blockers to diagnose BrS by unmasking/provoking a type 1 Brugada ECG pattern raises safety concerns regarding the use of local anesthetics.7 As BrS is uncommon, with an estimated prevalence of approximately 1:5,000, large prospective studies to define the risks of anesthesia are impractical. We searched the Mayo Clinic medical records database to identify patients with BrS who underwent anesthesia and reviewed their anesthetic course. The purpose of this case series was to document perioperative adverse events in BrS patients within a single institution and to determine any possible association between adverse events in patients with BrS and anesthetic drugs administered in the perioperative setting.

Methods

After obtaining Institutional Review Board approval, a computerized search of the Mayo Clinic Rochester electronic medical records (EMR) database from 2000 through 2010 was conducted to identify patients with clinically diagnosed BrS who underwent anesthetic care. We performed a keyword search using the terms “SCN5A”, “Brugada”, “Sudden unexpected death syndrome”, and “SUDS” and refined our search by selecting terms to eliminate patients in whom BrS was excluded. Our inclusion criteria consisted of adult patients (aged ≥ 18 yr), procedures involving anesthesia care, and a clinical diagnosis of BrS. Since SCN5A-BrS accounts for only 20-30% of the disease, a positive genetic test result was not required for inclusion in the study. The only data included in the study were from patients who provided research authorization for the use of their medical records [Minnesota Statute 144.335 Subd. 3a. (d)].

The electronic medical and anesthesia records of eligible patients were reviewed by one of the authors (B.K.). Data were entered into a standardized data collection form, and all questionable entries were discussed with the senior authors. As BrS is a relatively heterogeneous disorder that imposes diagnostic challenges, the EMR of each subject was reviewed with a genetic cardiologist (M.J.A.) to obtain the most specific description of each patient’s BrS diagnosis. The following data were collected for each anesthetic: 1) drug usage (type, amount, time point) with specific focus on medications that conceivably could increase the risk of ventricular arrhythmias in a BrS host; 2) ST segment measurement (value, time point); and 3) pre-, peri-, and postprocedural events (specifically, arrhythmias, defibrillations, and episodes of hyperthermia defined by a core body temperature ≥ 38°C). ST segment changes were analyzed during time increments of two minutes. A change ≥ 1 mm in leads II and V5 measured at J+60 during two consecutive measurements (four-minute time period) was considered significant. ST segment depressions were excluded from the analysis due to lack of association with BrS. Since ST segment changes can be prone to artifacts, we correlated significant ST segment events with the drug administration time points. We observed that ST segment monitoring data were obtained from recordings of leads II and V5, neither of which assesses the right precordium, i.e., the area of the most typical ST segment changes in BrS patients. However, these were the two leads available from the electronic medical record. Intra- and postoperative hemodynamic variables were recorded. Abnormal values were defined as hypertension (systolic BP ≥ 30% of preinduction baseline), hypotension (systolic BP ≤ 30% of preinduction baseline), tachycardia (HR > 110 beats·min−1), and bradycardia (HR < 50 beats·min−1).

Principal findings

During the ten-year period from 2000 through 2010, there were eight adult patients (seven males) with a clinical diagnosis of BrS at the Mayo Clinic who had one or more than one anesthetic exposure. According to our definition, all patients exhibited a type 1 Brugada ECG pattern either at rest or with provocation, and four of the eight patients had SCN5A-mediated BrS (i.e., type 1 BrS, Table 1). These eight BrS patients underwent a total of 17 surgical procedures. Table 2 summarizes the demographic and clinical data of the patients, the surgeries performed, the drugs used during anesthesia, and the intraoperative and postoperative complications, including ST segment changes. Of the 17 surgical procedures, seven were performed using general anesthesia, nine procedures involved monitored anesthesia care (patients receiving sedation provided by anesthesia personal, but the patients were not unconscious), and one procedure involved combined spinal/ epidural anesthesia. The medication profiles are detailed in Table 2.

Table 1 Brugada syndrome study patients
Full size table
Table 2 Demographic and clinical data of patients with Brugada syndrome undergoing anesthetic care
Full size table

During the perioperative and recovery time periods, no arrhythmias were observed except for sinus tachycardia and isolated premature ventricular contractions (PVCs). All patients tolerated anesthesia without adverse events or complications with the exception of 20 significant ST segment elevations that occurred in four patients. Table 3 shows the specific patients and the relationships of ST segment elevation to medication administration. Patient 3 had one significant ST segment elevation which was not in close temporal relationship (within ten minutes) to bolus drug administration. Patient 5 had multiple events related to automatic implantable cardioverter defibrillator testing. Patient 4 had two significant events; one was related to applying a topical application of oxymetazoline to the nose, and the other occurred shortly after induction with propofol, lidocaine, and succinylcholine. In patient 8, two events were noted to occur in close relationship to initiation of a propofol infusion with intravenous lidocaine bolus, and another event occurred during induction with etomidate.

Table 3 Significant ST segment elevations with time-relation to perioperative application of drugs
Full size table

Discussion

Our retrospective review of anesthetics performed in eight patients with diagnosed BrS showed that all patients tolerated anesthesia well. Some anesthetic agents (propofol, etomidate, lidocaine, succinylcholine, and nasal oxymetazoline) were noted to have a temporal association to ST segment elevations, a finding that is suspected to result in an increased risk of precipitating ventricular arrhythmias.

Brugada syndrome is an uncommon autosomal dominant genetic disease with incomplete penetrance and variable expressivity stemming from various mutations of ion channels in the cardiac conduction system.8,9 The disease is typically diagnosed during the fourth decade of life. Seven genotypes of BrS have been characterized, including mutations in cardiac sodium (types 1, 2, 5, and 7), potassium (type 6), and calcium (types 3 and 4) channels.8 Type 1 BrS, secondary to mutations in the SCN5A-encoded Nav1.5 sodium channel α-subunit, is the most common genetic subtype accounting for about 20-30% of BrS. The type 1 Brugada ECG pattern is the most specific ECG pattern for BrS. Type 1 is characterized by a coved-type ST segment elevation of at least 2 mm in the right precordial leads associated with a complete or incomplete right bundle branch block followed by a negative T wave.10 This ECG pattern can be either present at rest or inducible with sodium channel blockers, such as flecainide or procainamide. Patients may be asymptomatic, but they are at risk of developing ventricular tachycardia (VT) or fibrillation and sudden death.11,12 Type 2 Brugada ECG pattern presents with a saddleback appearance and either a positive or biphasic T wave. Type 3 Brugada ECG pattern can assume a saddleback or coved appearance, but it is characterized by less pronounced ST segment elevations. An observed type 1 ECG pattern is considered diagnostic when identified in conjunction with one of the following findings10,11 history of ventricular fibrillation or VT, family history of sudden cardiac death < 45 yr of age, coved-type ECGs in family members, inducibility of VT with programmed electrical stimulation, syncope or nocturnal agonal respiration. Type 2 and 3 ECG patterns are considered diagnostic when conversion to a type 1 ECG pattern is observed with the use of sodium channel blockers. Conversion of a type 3 ECG pattern to a type 2 ECG pattern is inconclusive.11

The pathophysiology behind the increased susceptibility to ventricular arrhythmias in BrS patients is beyond the scope of this article, but Morita et al. 9 present a comprehensive review. Physiologic stress,13 medications, and increased vagal activity14-16 can augment ST segment elevation. While a clear causal link between acute worsening of ST segment abnormalities and subsequent ventricular arrhythmias has not been established, multiple reports suggest such a relationship.17-20

It is difficult to formulate evidence-based guidelines for anesthetic management of these patients due to the absence of prospective studies combined with the low prevalence of BrS. Current guidelines are derived from theoretical models based on disease pathomechanism and observations from case reports and series.21 We conducted a literature search to compare our observations with previous reports regarding outcomes of patients with BrS who underwent anesthetic care. PubMed (1966-present) and EMBASE (1988-present) databases were searched using the following medical subject headings (MeSH): Brugada syndrome, anesthesia, anesthetics, and bupivacaine. Text words included: Brugada, anesthesia, anesthetics, anaesthesia, anaesthetics, bupivacaine, propofol, blockade, sympathomimetic, anticholinergic, perioperative, delivery, and caesarean. The search was limited to human subjects. In our search of anesthetic management of patients with BrS, we identified 21 case reports and four case series.

Articles in languages other than English were excluded,22-29 although an exception was made if an English abstract was provided that contained information about the surgical procedure and the medications used 30-33 or if the article was previously reviewed by another investigator (e.g., the case reports of Sugi et al. 34 and Lafuente et al.35 were reviewed in an article by Edge et al.).36

All case reports and case series were based on review of medical records. Including the present study, a cumulative experience with 52 anesthetics involving 43 patients has been published since 1966. A comprehensive list of patients, surgical cases, anesthetic drugs, and complications is shown in Table 4.

Table 4 Previous reports of anesthesia in Brugada syndrome patients
Full size table

Overall, our patients had unremarkable anesthetic courses. They exhibited normal responses to commonly used anesthetic agents, with the exception of two patients who experienced significant ST segment elevations in temporal relationship to administration of certain drugs, i.e., propofol, lidocaine, succinylcholine, oxymetazoline, or etomidate. All the documented ST segment elevations resolved spontaneously, and postanesthesia recovery times were not prolonged. The results support a small but consistent body of evidence describing relatively uneventful outcomes associated with general anesthesia and MAC sedation for patients with BrS.31-33,35-46 Despite the generally favourable outcomes, there remain several specific concerns that merit special consideration in patients with BrS.

Regional anesthesia and sodium-channel blockers

Regional anesthesia has been associated with complications in BrS. In a report by Fujiwara et al.,47 a patient with BrS developed polymorphic ventricular tachycardia 50 min following bilateral T8 paravertebral block using 40 mL of ropivacaine. However, concomitant use of a dopamine infusion to treat hypotension attributed to the block was a confounding factor which may have contributed to the development of the arrhythmia (see section “Vasoactive agents”). In a case reported by Phillips et al.,7 a previously asymptomatic patient developed Brugada-like ECG changes after a 14-hr infusion of bupivacaine in a thoracic epidural catheter. The bupivacaine infusion was discontinued and the ECG normalized over the ensuing 48 hr. The patient was given a provisional diagnosis of BrS based on the inducibility of a Brugada ECG pattern with sodium channel blockers, although no further evaluation was performed. Kaneda et al. 48 described a case in which a patient with diagnosed BrS developed ventricular fibrillation postoperatively after undergoing general and epidural anesthesia. Unfortunately, the article does not provide details regarding the medications used in the epidural (i.e., type of local anesthetic or dose). Interestingly, Kaneda et al.48 reported the successful use of intravenous lidocaine (intravenous bolus followed by continuous infusion) to control postoperative ventricular fibrillation in a BrS patient. In our series, one patient had a lumbar epidural placed for labour analgesia through which fentanyl-bupivacaine was applied by bolus and continuous infusion. No complications were noted.

Several authors reported uneventful regional or wound infiltration with local anesthetics in BrS patients.36-39,49 Local anesthetics are class Ib antiarrhythmics and thus block sodium channels. Intravenous lidocaine was reported to induce a Brugada ECG pattern which led to the diagnosis of BrS with subsequent identification of a V232I+L1308F double-missense mutation in the SCN5A gene.50 However, the effect of lidocaine may be dependent on specific ion channel mutation. For example, lidocaine exerted a beneficial effect in a patient with type 1 BrS secondary to a SCN5A-N406S missense mutation.51,52 Regardless, local anesthetics should be used cautiously, and if used, the dose should be minimized and the patient should be monitored closely. Four patients in our series received intravenous lidocaine without evidence of dysrhythmias, but an association with ST segment elevations was noted in three cases (Table 3).

Propofol and other induction agents

It has been suggested that propofol should be avoided in patients with BrS.21 Vernooy et al. 1 examined the relationship of ECG changes and sudden death in propofol infusion syndrome (PRIS). The group reported an index case in which PRIS induced a Brugada ECG pattern. A subsequent chart review revealed seven additional cases of PRIS; in six of those seven patients, a Brugada ECG pattern was recorded shortly before the occurrence of an electrical storm. Roberts et al. defined the features of PRIS in a large prospective incidence study published in 2009.53 The characteristics include development of metabolic acidosis and cardiac dysfunction along with at least one of rhabdomyolysis, hypertriglyceridemia, or renal failure after receiving intravenous propofol. In a report by Robinson et al.,4 an infant treated with continuous propofol infusion developed wide complex tachycardia with left bundle branch block morphology. This condition was cardioverted into a slower irregular rhythm with right bundle branch block morphology and pronounced ST segment elevations in V1. In a case reported by Riezzo et al.,3 a patient with long-term propofol abuse developed a Brugada ECG pattern with subsequent cardiovascular instability and death. The described presentation was reminiscent of PRIS with regard to cardiovascular instability and metabolic acidosis. An in vitro experiment on cardiac myocytes demonstrated that propofol exerts a dose-dependent blockade of whole cell sodium current and induces a hyperpolarizing shift in the voltage-dependence of the inactivation of sodium currents.54 Propofol has also been found to inhibit cardiac L-type calcium channels,55 attenuate beta-adrenergic signal transduction,56 and augment acetylcholine receptor activity.57

However, these adverse outcomes have occurred only in the setting of propofol abuse3 and PRIS.1,5,53 More importantly, none of the observed patients carried a diagnosis of BrS. Consequently, the only association between PRIS and BrS is the implication of propofol infusions in unmasking a Brugada ECG pattern.1-5 In a report by Weiner et al.,6 they described a Brugada ECG pattern induced by propofol infusion in a healthy young male who subsequently underwent testing and was found not to have BrS. Propofol has been used during anesthesia in BrS patients without incident.31,32,34,35,39-41,49 In our series, propofol was administered as a bolus in nine cases and as an infusion in three cases, and no dysrhythmias were observed. However, one patient who had received a propofol bolus and one patient who received a propofol infusion were noted to have significant ST segment elevations following propofol administration (within seven minutes after injection or start of infusion). The ST segments subsequently normalized spontaneously.

It is important to highlight that a type 1 Brugada ECG pattern is an electrocardiographic result which is not necessarily diagnostic of BrS in the absence of supporting findings. The differential diagnosis of a patient who exhibits a type 1 Brugada ECG pattern includes BrS, but it also encompasses various acquired diseases as well as cocaine58/marijuana59 use. In addition, multiple occurrences of drug-induced Brugada ECG patterns in previously healthy patients were reported,60-63 although those ECG changes were noted mostly in the setting of medication overdoses. Nevertheless, propofol has the potential to alter ion channel function. Despite theoretical concerns and various reports of adverse outcomes with propofol in a small subset of patients without known BrS, in our view, the clinical experience does not support the recommendation of avoiding bolus dosing for induction in BrS patients. Caution is advised for continuous infusions, however, as BrS patients may be more predisposed to cardiac arrhythmias due to their intrinsic ion channel malfunction.

Regarding other induction agents, thiopental use has been described in multiple case reports22,28,30,31 without problems, whereas no studies reported the use of etomidate. In our series, self-limited ST segment elevations were noted following etomidate administration, while no ST segment abnormalities were noted with the administration of thiopental.

Inhalational anesthetics, muscle relaxants, acetylcholine-esterase inhibitors, anticholinergics, and antiemetics

Inhalational anesthetics have not been associated with adverse events in BrS patients either in our series or in the literature.30-33,35-38,41-44,49 Both depolarizing and nondepolarizing neuromuscular blocking agents have been utilized clinically in our series and in previous case studies without incident.31,36,40,42-45 In our series as well as in the literature, nondepolarizing muscle relaxants were reversed with neostigmine without incident.36,40,42-45

Atropine31,40,43,49 and glycopyrrolate35,38,44,45 were used by other authors without problems. In our series, glycopyrrolate and scopolamine were used without adverse effects. Interestingly, in one study of patients with BrS, ST segment elevation was found in the right precordial leads after intracoronary acetylcholine (3/3 tested patients) and intravenous edrophonium administration (2/3 patients).20 In theory, given that increased vagal tone may increase arrhythmia susceptibility in a patient with BrS, the anticholinergic action of atropine, glycopyrrolate, and scopolamine drugs could exert a beneficial effect on ST segment changes. A variety of opioids has been used in our series and reported in the literature31-33,35-38,41,42,44,49 without adverse events in patients with BrS.

Commonly used antiemetics, including ondansetron, granisetron, and dexamethasone, have not been associated with any reported adverse effects in BrS patients, and all these drugs were used in our series without incident. Droperidol was not used in our patients, and it has been recommended to avoid phenothiazine antipsychotics (trifluoperazine, thioridazine, perphenazine).21 Phenothiazine overdoses have resulted in Brugada ECG patterns.64-66 Perphenazine has produced a reversible blockade of the Nav1.5 sodium current and the Kv4.3 transient outward potassium current (Ito) in isolated rat right ventricular cardiomyocytes,67 and trifluoperazine and chlorpromazine have reduced calcium inward current, sodium currents, and inwardly and delayed potassium currents in isolated guinea pig myocytes and bovine portal veins.68 However, the use of prochlorperazine has not been reported in BrS patients.

Vasoactive agents

Modulation of adrenergic receptors can modify ECG tracings in BrS patients by altering the ST segments.12 Alpha-receptor agonists20 and β-receptor antagonists69 can worsen ECG patterns by increasing the magnitude of ST segment elevation or unmasking a Brugada ECG pattern (change from normal ST segments to ST segment elevations), whereas α-receptor antagonists20 and β-receptor agonists20 improve ECG patterns by returning elevated ST segments back to baseline or decreasing the magnitude of ST segment elevations. The β1 and β2 receptor agonistic activity of isoproterenol increases calcium current and has been used to reduce ST segment elevation20 and suppress arrhythmic events in patients with BrS.70-73 Several approaches have been proven to successfully abrogate electrical storms. Joshi et al. 72 and Jongman et al.71 reported suppression of arrhythmias with an infusion of isoproterenol 1 μg·min−1. Watanabe et al.73 used a bolus/maintenance approach with isoproterenol with an initial bolus of 1-2 μg followed by a continuous infusion adjusted from 0.15-0.30 μg·min−1. Ohgo et al.70 showed an effect at lower doses of 0.002 μg·kg−1·min−1 with titration to 0.004 μg·kg−1·min−1. Besides the use of isoproterenol, the basic life support and advanced life support protocol should be no different for patients with BrS; however, data regarding the safety of other commonly used vasopressors are almost non-existent. Theoretically, the predominant β-agonist action of dobutamine could be beneficial, while other vasopressors with dual α- and β-agonist effect may have unpredictable effects. Epinephrine in conjunction with procainamide has been used to unmask Brugada,74 and norepinephrine can augment ST segment elevations. However, the use of dopamine75 and ephedrine49 has been described, and ephedrine was used in our series without complication. The selective α1-agonist activity of phenylephrine has theoretic potential to be deleterious, but it was used in our series without incident.

An overdose of the β-antagonist, propanolol, has been reported to unmask a Brugada ECG pattern.69 This change may have resulted from bradycardia and β-antagonism. However, propanolol at high doses binds to cardiac sodium channels inhibiting sodium uptake, and this action could have been the mechanism.63 In our series, one patient received esmolol and one patient received labetalol without incident.

Limitations

In this small case series of BrS patients, the right precordial leads (V1-V3) were not monitored in any of our patients to assess for dynamic ST segment changes. Thus, subtle undetected changes may have occurred. However, ST segment elevations in the inferior leads76 and the left precordial leads77 have also been observed in Brugada-like syndromes. If the surgical field allows, monitoring the right precordial leads may enhance sensitivity to detect dynamic ST segment changes in BrS patients, and this might enable the anesthesiologist to notice a possible unmasking of a Brugada-type ECG pattern in patients with normal resting ECGs. Furthermore, the ST segment changes were recorded only as numerical values without the possibility to assess morphology, thus precluding the potential to identify “coved” ST segments (Figure).

Figure
figure 1

Precordial lead tracing (Patient 3). Typical right precordial lead tracings (V1-V3) showing T wave abnormalities typically observed in patients with Brugada syndrome

Full size image

Another limitation of this retrospective case series is the small number of patients, which precluded a definitive proof of safety of any anesthetic agent or technique.

Conclusions

Patients with BrS in our series tolerated anesthesia without any untoward events other than episodes of spontaneously resolving ST segment elevation. However, with the exception of sodium channel blockers, clear data is clearly limited regarding the drugs that can exacerbate ST segment elevations and/or facilitate the development of ventricular arrhythmias in BrS patients. For a comprehensive review, we refer the reader to the current consensus statement published in 200921 or to the website, http://www.brugadadrugs.org. Our case series and a limited body of existing evidence report safe anesthetics using agents that theoretically are associated with possible problems when used in BrS patients, e.g. propofol (especially when given as a continuous infusion) and local anesthetics. In multiple studies, the β-agonist, isoproterenol, has been shown to be effective in normalizing ST segment elevations and in preventing electrical storm in BrS patients.