|Home | About | Journals | Submit | Contact Us | Français|
The risk for lethal ventricular arrhythmias is increased in individuals who carry mutations in genes that encode cardiac ion channels. Loss-of-function mutations in SCN5A, the gene encoding the cardiac sodium channel, are linked to Brugada syndrome (BrS). Arrhythmias in BrS are often preceded by coved-type ST-segment elevation in the right-precordial leads V1 and V2. Loss-of-function mutations in KCNH2, the gene encoding the cardiac ion channel that is responsible for the rapidly activating delayed rectifying potassium current, are linked to long-QT syndrome type 2 (LQT-2). LQT-2 is characterised by delayed cardiac repolarisation and rate-corrected QT interval (QTc) prolongation. Here, we report that the risk for ventricular arrhythmias in BrS and LQT-2 is further increased during fever. Moreover, we demonstrate that fever may aggravate coved-type ST-segment elevation in BrS, and cause QTc lengthening in LQT-2. Finally, we describe molecular mechanisms that may underlie the proarrhythmic effects of fever in BrS and LQT-2. (Neth Heart J 2010;18:165–9.)
The risk for sudden death due to ventricular arrhythmias is increased in individuals who carry mutations in genes that encode cardiac ion channels and associated proteins. Inherited or de novo mutations in SCN5A, the gene that encodes the cardiac sodium channel, are linked to Brugada syndrome (BrS), a disease characterised by signs of conduction slowing on the ECG (prolonged PR and QRS intervals), and increased incidence of ventricular arrhythmias that are preceded by coved-type ST-segment elevation in the precordial leads V1 and V2.1 BrS-linked SCN5A mutations exert their proarrhythmic effects by causing sodium channel loss-of-function (i.e., reducing cardiac sodium current [INa]) by disrupting intracellular trafficking of mutant channel proteins from the endoplasmic reticulum to the sarcolemma, forming channels that conduct no or very small INa, or altering the opening and closing (gating) properties of mutant channels in such a manner that the channels are less frequently open.2,3 Anecdotal reports have suggested that fever may aggravate the typical ECG changes and increase the risk for sudden death in BrS.4,5 However, the association between fever and the risk for ventricular arrhythmias and preceding ECG changes has not been studied systematically. Recently, we first investigated this association in index patients who were diagnosed with BrS at our institution.6 In this retrospective study, we assessed the prevalence of fever-triggered ventricular tachycardia, ventricular fibrillation, or sudden death in 111 BrS patients, and compared these data with 41 control participants who were admitted after out-of-hospital cardiac arrest due to ventricular tachycardia or ventricular fibrillation. We found that, compared with control participants, BrS patients experienced cardiac arrest significantly more often during fever. Additionally, we analysed fever-induced ECG changes in 24 BrS patients for whom 12-lead ECGs during both fever and normothermia were available, and compared the data with ECG changes in ten control participants who were admitted for noncardiac reasons. We found that fever, irrespective of the cause, induced additional prolongation of PR and QRS intervals, and aggravated the coved-type ST-segment elevation in leads V1 and V2 in BrS patients. In contrast, the PR interval shortened, and the QRS interval and the ST-segment amplitude in leads V1 and V2 did not change during fever in control participants (figure 1A and B). We also investigated whether the use of antipyretic drugs during fever affected the prevalence of cardiac arrest in BrS patients. Indeed, a larger proportion of patients who had not used paracetamol (acetaminophen) experienced cardiac arrest during fever than those who had used acetaminophen. Thus, this systematic study indicated that fever increases the risk for ventricular arrhythmias and aggravates the typical ECG changes in BrS patients.
Despite the clinical evidence, the molecular mechanism underlying fever-triggered ventricular arrhythmias in BrS is not well understood. Cellular electrophysiological studies suggest that SCN5A mutations impair cardiac sodium channel gating more severely at higher temperatures, leading to further INa reduction during fever.7-9 Gating results from transitions of channel proteins between distinct conformational states. Upon membrane depolarisation, sodium channels change from their closed state to their open state (activation), thereby allowing influx of sodium ions into the cell. This causes more membrane depolarisation and triggers the cardiac action potential. Within milliseconds thereafter, sodium influx declines as sodium channels inactivate. Inactivation comprises multiple conformational states (fast, intermediate, and slow inactivation). To be able to open again for the next action potential, the channels must first return from the inactivated to the closed state (recovery from inactivation). This occurs during the diastolic interval.10 We and others have demonstrated that some SCN5A mutations cause sodium channels to enter their slow inactivation state more easily at higher temperatures (i.e., enhanced slow inactivation), thereby causing further INa reduction (figure 1C).7,8 Further INa reduction at higher temperatures is believed to cause additional conduction slowing, which corresponds with the prolongation of PR and QRS intervals in BrS patients during fever.6,11,12 Interestingly, increased dysfunction of mutant sodium channels at higher temperatures is not specific for SCN5A mutations, since mutations in genes encoding other organ-specific sodium channels are also related to diseases with temperature-dependent symptoms. Mutations in SCN1A, the gene that encodes the brain sodium channel, is linked to generalised epilepsy with fever-triggered seizures.13 Mutations in SCN9A, the gene that encodes a sodium channel predominantly expressed in sensory and sympathetic neurones, is linked to primary erythermalgia, a disease characterised by intermittent heat-triggered burning pain and skin redness in the extremities.14
Importantly, sinus tachycardia may also contribute to the development of arrhythmias during fever. During sinus tachycardia, the time between two consecutive depolarisations (i.e., heart beats) may become too short for mutant channels to full recover from inactivation. This may result in the accumulation of mutant channels in the slow inactivation state at fast heart rates, leading to further INa reduction. This hypothesis is supported by the aggravation of BrS-linked ECG changes in BrS patients during exercise-related tachycardia, and in right ventricular tissue preparations at higher stimulation frequencies.15,16
Finally, fever-induced ventricular arrhythmias are not limited to BrS. Recently, we demonstrated that individuals with type 2 long-QT syndrome (LQT-2) may be at increased risk for arrhythmias during fever.17 LQT-2 is caused by loss-of-function mutations in KCNH2 (formerly called HERG), the gene that encodes the channel carrying the rapidly activating delayed rectifying potassium current (IKr). IKr plays an important role in ventricular repolarisation, and IKr reduction delays ventricular repolarisation to cause action potential and QT-interval prolongation. We reported two related patients (father and son) with LQT-2 due to the heterozygous A558P HERG missense mutation who repeatedly experienced syncope and ventricular arrhythmias during fever. ECG analysis showed QTc lengthening in both LQT-2 patients. In contrast, QTc intervals in nine control participants shortened during febrile temperatures (figure 2A). When expressed in heterologous expression systems (HEK-293 cells), A558P mutant proteins displayed loss-of-function due to deficient intracellular trafficking (figure 2B). Because patients were heterozygous, carrying both normal and mutant allele, we co-expressed wild-type (WT) HERG proteins with A558P mutant proteins (WT+A558P), and found that A558P disrupts the intracellular trafficking of WT proteins (i.e., dominant-negative effect), causing 65% current reduction (figure 2B). Moreover, A558P reduced the normal increase of WT current at higher temperatures. In other words, the WT+A558P current failed to increase to the same extent as the WT current and caused larger reduction in current density at higher temperatures. A similar temperature-dependent phenotype was seen for co-expression of WT proteins with the F640V HERG mutation (figure 2C and D). Similar to A558P, F640V is linked to LQT-2, displays a trafficking-deficient phenotype, and exerts a dominant-negative effect.18 We postulated that the reduced increase in the current density in WT-mutant co-assembled channels at higher temperatures may underlie QTc lengthening and the development of ventricular arrhythmias during fever. Indeed, IKr reduction at febrile temperatures has been shown to prolong action potential duration in Purkinje fibre preparations, thereby increasing transmural differences in action potential duration, and facilitating the occurrence of early afterdepolarisations (which are believed to result from reactivation of L-type Ca2+ channels during phase 3 of the cardiac action potential).19,20 Early afterdepolarisations are believed to initiate ventricular arrhythmias in LQT-2.21
In summary, we have shown that individuals with BrS or LQT-2 may be at increased risk for ventricular arrhythmias during fever. Accordingly, we demonstrated that fever may aggravate the coved-type ST-segment elevation in leads V1 and V2 that often precedes arrhythmias in BrS, and may cause QTc lengthening in LQT-2. The possible underlying mechanisms involve the effects of fever on the function of mutant ion channels. Nevertheless, fever-triggered arrhythmias may not only be attributed to further dysfunction of mutant ion channels, since they may also occur in individuals without mutations in ion channel encoding genes.22-24 Importantly, early treatment of fever with antipyretic drugs may lower the risk for ventricular arrhythmias during fever in BrS, and perhaps also in LQT-2. Moreover, although evidence is lacking, vaccination against seasonal flu viruses, and in particular the 2009 H1N1 virus, may lower arrhythmia risk by preventing severe febrile episodes due to viral infections. Future research will focus on the in vivo effects of fever and its various aspects (e.g., higher body temperature, sinus tachycardia, and inflammatory cells and cytokines) on cardiac electrophysiology in transgenic mouse models carrying mutations in a specific gene of interest, and in individuals without mutations in ion channel encoding genes but with common acquired diseases (e.g., myocardial infarction, and congestive heart failure).
This study was supported by the Netherlands Organisation for Scientific Research grants NWO ZonMW-VICI 918-86-616 (Hanno L. Tan), and NWO 902-16-193 (Arthur A.M. Wilde), the Netherlands Heart Foundation grant NHS 2003T302 (Arthur A.M. Wilde), and the National Heart, Lung, and Blood Institute grant R01 HL60723 (Craig T. January).