Drug- and non-drug-associated QT interval prolongation

Abstract

Sudden cardiac death is among the most common causes of cardiovascular death in developed countries. The majority of sudden cardiac deaths are caused by acute ventricular arrhythmia following repolarization disturbances. An important risk factor for repolarization disturbances is use of QT prolonging drugs, probably partly explained by gene–drug interactions. In this review, we will summarize QT interval physiology, known risk factors for QT prolongation, including drugs and the contribution of pharmacogenetics. The long QT syndrome can be congenital or acquired. The congenital long QT syndrome is caused by mutations in ion channel subunits or regulatory protein coding genes and is a rare monogenic disorder with a mendelian pattern of inheritance. Apart from that, several common genetic variants that are associated with QT interval duration have been identified. Acquired QT prolongation is more prevalent than the congenital form. Several risk factors have been identified with use of QT prolonging drugs as the most frequent cause. Most drugs that prolong the QT interval act by blocking hERG-encoded potassium channels, although some drugs mainly modify sodium channels. Both pharmacodynamic as well as pharmacokinetic mechanisms may be responsible for QT prolongation. Pharmacokinetic interactions often involve drugs that are metabolized by cytochrome P450 enzymes. Pharmacodynamic gene–drug interactions are due to genetic variants that potentiate the QT prolonging effect of drugs. QT prolongation, often due to use of QT prolonging drugs, is a major public health issue. Recently, common genetic variants associated with QT prolongation have been identified. Few pharmacogenetic studies have been performed to establish the genetic background of acquired QT prolongation but additional studies in this newly developing field are warranted.

Introduction

Sudden cardiac death is among the most common types of mortality in developed countries. Sudden death from cardiac causes accounts for approximately 50% of all deaths from cardiovascular diseases and 20% of total mortality [1–3]. In the general population, sudden cardiac death mostly occurs in individuals who are unrecognized to be at risk [4]. The majority (80–85%) of sudden cardiac deaths is caused by acute ventricular arrhythmia [5]. An important potential cause of ventricular arrhythmia is prolongation of ventricular repolarization, for instance, as is observed in the rare and genetically-determined ‘congenital long QT syndrome’ (LQTS) [6]. In addition, QT prolongation in the general population can be due to common genetic variants or the acquired long QT syndrome. Prolongation of ventricular repolarization may result in early after depolarizations (EAD), which in turn may induce re-entry and thereby provoke torsade de pointes and fatal ventricular arrhythmia [7–11].

The incidence of acquired long QT syndrome is much higher than the incidence of congenital LQTS. An important risk factor for acquired long QT syndrome is use of certain cardiac and non-cardiac QT prolonging drugs, probably caused by gene–drug interactions. In this review, we will give a short description of QT interval physiology, known non-drug related and drug-related risk factors of QT prolongation, including the role of pharmacogenetic interactions.

Ventricular repolarization disturbances

The QT interval, which is the traditional measurement for assessing the duration of ventricular de- and repolarization, is measured in milliseconds (ms) on the body surface electrocardiogram (ECG) from the Q-top, the beginning of the QRS complex, until the end of the T wave. Although several methods are used to adjust for the effect of heart rate (e.g. Fridericia formula (), Bazett's formula () or linear adjustment for RR interval), Bazett's formula is most often used (Figure 1) [12]. According to European regulatory guidelines, QT_c prolongation can be distinguished into three clinically-relevant categories. For men, the cut-off points are less than 430 ms (normal), 430–450 ms (borderline) and more than 450 ms (prolonged), and for women less than 450 ms (normal), 450–470 ms (borderline), and more than 470 ms (prolonged) [13].

Figure 1.

Corrected QT interval using Bazett's formula. Source: Al-Khatib [10]

The electrical activity of the heart is mediated through channels that regulate the flow of ions in and out of cardiomyocytes. The inward depolarizing currents, mainly through Na⁺ and Ca²⁺ channels, result in normal depolarization, and outward repolarizing currents, mainly through K⁺ channels, result in repolarization (Figure 2). Malfunction of ion channels may lead to an intracellular excess of positively charged ions, i.e. either by an inadequate outflow of potassium ions or by an excessive inflow of sodium ions. The intracellular excess of positively charged ions prolongs ventricular repolarization and results in QT interval prolongation on the ECG [14].

Figure 2.

The cardiac action potential. A) action potential showing the five phases of cardiac depolarization and repolarization with ion current directions during activation of the different ion channels; B) ECG. Phase 0: Rapid depolarization is caused by a large inward current of sodium ions (INa). Repolarization consists of three phases [1–3]: Phase 1: Rapid repolarization phase is caused by inactivation of I_Na and the transient efflux of potassium ions (I_t0); Phase 2: Plateau phase, which is a reflection of a balance between the influx of calcium ions through l-type calcium channels (I_Ca) and outward repolarizing potassium currents (I_K); Phase 3: Late repolarization phase results from the efflux of potassium (I_Kr, I_Ku, I_Ks). Phase 4: Resting potential is maintained by the inward rectifier potassium current (I_K1). I_Ca = calcium current; I_K = potassium current; I_K1 = inwardly rectifying potassium current; I_Na = depolarizing sodium current; I_t0 = transient outward potassium current; I_Kr = rapidly activating delayed rectifier potassium current; I_Ks = slowly activating delayed rectifier potassium current; I_Ku = ultra rapidly activating delayed rectifier potassium current. Source: Titier et al. [64]

Risk factors for repolarization disturbances

Risk factors for QT_c prolongation can be divided into two main categories, i.e. congenital and acquired abnormalities, but repolarization disturbances often require additional gene-environment interactions [10].

Congenital long QT syndrome

The congenital long QT syndrome (LQTS) is an inherited disease characterized by prolongation of ventricular myocyte repolarization, which is manifested by QT prolongation on the surface ECG, syncopal episodes, malignant ventricular tachycardia and fibrillation [6]. The prevalence of LQTS is estimated to be approximately 1 in 2000–2500 live births [15, 16]. The autosomal dominant inherited LQTS (Romano-Ward syndrome) is far more common than the recessive inherited LQTS, which is associated with deafness (Jervell and Lange-Nielsen syndrome) [17]. Most LQTS patients are asymptomatic and are either discovered incidentally on an ECG, by family history or after surviving an episode of syncope or severe ventricular arrhythmia. The prognosis of untreated patients is considered to be quite poor. It is estimated that approximately 20% of untreated patients presenting with syncope die within 1 year and 50% within 10 years [18].

LQTS is genetically and phenotypically heterogeneous and is caused by mutations in a variety of ion channel subunit or regulatory protein coding genes. To date there are eleven known subtypes of the Romano-Ward syndrome and two subtypes of the Jervell and Lang-Nielsen syndrome [19]. The most prevalent dominant disorders are LQT1 (KCNQ1) and LQT2 (KCNH2) due to mutations in potassium channels, and LQT3 (SCN5A) due to sodium channel mutations (Table 1). Most LQT1 events are triggered by exercise or stress, LQT2 events by emotional stress such as auditory stimuli upon a low heart rate, while LQT3 events most often occur during sleep or at rest [20]. Although the risk of cardiac events is higher among patients with LQT1 and LQT2, the frequency of fatal cardiac events is higher in LQT3 patients [21].

Table 1.

Common forms of the long QT syndrome (LQTS)

	Genetic subtype
	LQT1	LQT2	LQT3
Disease-associated gene	KCNQ1	KCNH2 / hERG	SCN5A
Ion current affected	IKs	IKr	INa
Pathophysiology	Decreased potassium outward currents	Decreased potassium outward currents	Excessive sodium inward current
Trigger of arrhythmia	Exercise (diving, swimming), stress	Emotional stress (acoustic)	Rest, sleep
Occurrence	>50%	35–40%	10–15%

I_Ks = rectifier potassium current, slow component; I_Kr = rectifier potassium current, rapid component; I_Na = inward sodium current.

Genetic variants associated with QT interval variation in the general population

Also in those without congenital LQTS, the QT interval is a quantitative trait with approximately 30% heritability in the general population [22–24]. Common genetic variants that are associated with QT interval duration have been identified. Several of these are located within genes known from the LQT syndromes (KCNQ1, KCNH2 (hERG), SCN5A, KCNE1). More recently, common variants in genes previously unknown to be relevant for repolarization were identified (NOS1AP, NDRG4 and GINS3, PLN, RNF207, LITAF, LIG3 and RFFL). The fourteen genetic variants in 10 loci identified in the QTGEN Study explain 5.4–6.5% of the population variation in the QT interval [25–30].

Acquired QT interval prolongation

There are many factors that predispose to QT prolongation including, e.g. age, female gender, left ventricular hyperthrophy, heart failure, myocardial ischaemia, hypertension, diabetes mellitus, increased thyroid hormone concentrations, elevated serum cholesterol, high body mass index, slow heart rate and electrolyte abnormalities (including hypokalaemia and hypomagnesaemia) [10, 31–35]. However, one of the most common causes of acquired QT_c prolongation is the use of specific drugs [36].

In the past decade, a frequent cause of withdrawal or restriction of marketed drugs has been prolongation of the QT_c interval [8, 37]. An increasing number of antipsychotic, antihistaminic, gastrointestinal and anti-infective drugs (e.g. thioridazine, astemizole, cisapride, grepafloxacin) have been withdrawn from the market due to delay of cardiac repolarization and reports of torsade de pointes. Several other drugs (e.g. terfenadine, haloperidol, sertindole) were restricted in use because of this potential adverse reaction [9, 38]. Some of the QT_c prolonging drugs withdrawn were associated with a QT_c interval prolongation of only 5–10 ms in patient populations [9]. However, this was an average and the possibility of more extreme values in subgroups of individuals should be taken into account. If such drugs are used on a large scale, it becomes evident that non-cardiac drugs associated with a pro-arrhythmogenic potential can be a considerable public health problem [11]. In a recent study, it was shown that doctors who overrule high level drug–drug interaction alerts on QT_c prolongation rarely record an ECG as a safety measure. If ECGs were recorded before and after following such an alert, clinically relevant QT_c prolongation was found in one-third of the cases with an average change in QT_c interval of 31 ms [39].

Virtually all QT_c prolonging drugs act by blocking the rapid component of the delayed rectifier potassium channel (I_kr) encoded by the human ether a go-go related gene (hERG) (Figure 3) [9]. The clinical implications of drug-induced QT_c prolongation are not completely clear. It is known that some drugs associated with QT_c prolongation are devoid of torsadogenic effects, whereas others seem to be associated with cardiac arrhythmia without QT_c prolongation [9, 40, 41]. Blocking of the I_kr current is not specific, since not all drugs which block this current cause torsade de pointes (e.g. amiodarone). On the other hand, some drugs that prolong the QT_c interval by only a few milliseconds were nevertheless implicated in the occurrence of cardiac arrhythmias and torsade de pointes (e.g. the antihistamine terfenadine) [40, 41]. Terfenadine is a potent I_Kr blocker but usually it does not prolong the QT_c interval because it is readily transformed into its metabolite fexofenadine which is without an effect on QT interval. Terfenadine has been withdrawn from the market in several countries because of the possible occurrence of interactions with diseases and drugs that inhibit metabolism by cytochrome P450 (CYP) 3A4 resulting in higher concentrations of terfenadine, and also due to the existence of alternative therapies [9, 11].

Figure 3.

Mechanisms of sudden cardiac death with hERG blockade. Drug blockade of the hERG channel (left) can result in QT interval prolongation (middle) and torsade de pointes (right; upper panel), which can develop into to ventricular fibrillation (VF) (right; lower panel). Source: Roden et al. [65]

In 2003, it was shown that drugs with a small margin (i.e. drugs that block the potassium channels in concentrations close to the therapeutic plasma concentration) had a high risk of serious cardiac arrhythmia while drugs with a high margin (i.e. drugs that block the potassium channels in concentrations exceeding the therapeutic level) had a lower risk. In that review, sertindole and thioridazine were shown to have a low margin, suggesting a high risk of cardiac arrhythmias, while amiodarone had a very high margin, which is in line with its known low risk of torsade de pointes [40]. Recently, it was demonstrated in a population-based case-control study that drugs with a high hERG-encoded potassium channel inhibiting capacity had a higher risk of causing sudden cardiac death than drugs with a low potassium channel inhibiting capacity [42].

Although most QT_c prolonging drugs act by blocking I_kr, few drugs are known to prolong the QT_c interval by modifying I_Na. Given the structural similarity between Na⁺ channels and K⁺ channels, it is possible that Na⁺ channel blocking drugs also bind to K⁺ channels [43]. Nevertheless, sodium channel blocking drugs can cause slowed intraventricular conduction, with the development of a re-entrant circuit, resulting in ventricular tachycardia or ventricular fibrillation. Examples of drugs with sodium channel blocking activity are certain antihistamines, β-adrenoceptor blockers, tricyclic antidepressants and phenothiazines [43, 44].

In recent years, several lists have been published of non-cardiac drugs associated with QT_c prolongation and cardiac arrhythmias [36, 37, 40, 45]. The internet based registry of R.L.Woosley (http://www.qtdrugs.org/medical-pros/drug-lists/drug-lists.cfm) contains drugs that are known to prolong the QT_c interval [36]. The QT_c prolonging drugs are classified into four categories, varying from drugs that are generally accepted to have a risk of causing torsade de pointes (list 1) (Table 2) to drugs that, in some reports, have a weak association with torsade de pointes and are unlikely to increase the risk when used in therapeutic dosages (list 4). This list includes drugs based on information from the International Registry of Drug-Induced Arrhythmias, literature, change in labels of (new) drugs and the Food and Drug Administration's (FDA) database for adverse drug event reports [36]. In addition, De Ponti et al. have published a list of non-anti-arrhythmic drugs with pro-arrhythmogenic effects, based on a structured literature search including published (non-) clinical evidence and official warnings in the labelling [37, 45].

Table 2.

Drugs that may cause torsade de pointes

Anti-arrhythmics

Amiodarone

Disopyramide

Dofetilide

Ibutilide

Procainamide

Quinidine

Sotalol

Antihistamines

Astemizole

Terfenadine

Anti-infectives

Clarithromycin

Erythromycin

Pentamidine

Sparfloxacin

Antimalarials

Chloroquine

Halofantrine

Antipsychotics

Chlorpromazine

Haloperidol

Mesoridazine

Pimozide

Thioridazone

Gastro-intestinal drugs

Cisapride

Domperidone

Opiate agonists

Levomethadyl

Methadone

Other

Arsenic trioxide

Bepridil

Droperidol

Probucol

Source:http://www.qtdrugs.org/medical-pros/drug-lists/drug-lists.cfm [36]

Drug–drug interactions

Both pharmacodynamic and pharmacokinetic drug effects may lead to QT_c prolongation. A pharmacodynamic interaction of concomitantly used drugs can lead to a prolonged QT_c interval if the individual QT_c prolonging drugs have an additive or potentiating effect [46]. A pharmacokinetic effect may occur if a drug reduces the clearance of a concomitantly used QT_c prolonging drug, leading to increased plasma and tissue concentrations [47]. Pharmacokinetic interactions often involve drugs which are both metabolized by specific CYP iso-enzymes [48]. Patients using two or more drugs concomitantly metabolized by CYP3A4 or CYP2D6, can develop QT_c prolongation due to increased plasma concentrations. For example, QT_c prolonging drugs such as cisapride and terfenadine are metabolized by CYP3A4 and haloperidol and thioridazine are metabolized by CYP2D6 [49].

Gene–drug interactions

Gene–drug interactions probably exist at different levels: rare ion-channel mutations that increase the risk of QT prolongation by drug use, common genetic variants that potentiate the QT prolonging effect of drugs and variation within drug metabolizing and transporting proteins that influence drug pharmacokinetics. The known LQTS genes and drug metabolizing enzyme coding genes are obvious candidates for these studies. The recently identified loci previously unsuspected to play a role in the population variation of the QT interval duration are novel candidates for drug-gene interaction effects [49].

Previously unrecognized LQTS can be identified in 5–20% of patients with drug-induced torsade de pointes [50–52]. These patients have ‘forme-fruste’ mutations in congenital LQTS genes and are in general asymptomatic with normal to borderline QT_c intervals. After exposure to QT_c prolonging drugs, these subjects appear to be more susceptible to blockade of hERG-encoded potassium channels and develop QT_c prolongation [49]. Several genetic variations have been identified in patients with severe QT_c prolongation, torsade de pointes, cardiac arrest or sudden cardiac death after drug exposure [51, 53, 54].

In a genome wide association study, polymorphisms in genes previously unknown to play a role in QT interval physiology or prolongation in response to medication were associated with QT_c prolongation during treatment with an atypical antipsychotic [55]. Recently, a population-based study has been published demonstrating that polymorphisms in the NOS1AP gene significantly potentiate the QT_c prolonging effect of verapamil [56].

In addition, single nucleotide polymorphisms in the drug metabolizing enzyme genes, such as CYP2D6, can lead to an altered function of the enzyme. Subjects with two non-functional CYP2D6 alleles are classified as ‘poor metabolizers’. Approximately 5–10% of the Caucasian population are ‘poor metabolizers’ [57]. ‘Poor metabolizers’ using QT_c prolonging drugs metabolized by CYP2D6 have an increased risk of developing QT_c prolongation or torsade de pointes. A similar effect is observed for drug transporters that show altered activity due to genetic variation. An altered activity of transporters can lead to a change in drug clearance or intracellular drug concentrations and thereby influence the QT_c interval [58, 59].

Sudden cardiac death

According to the most recent definition, sudden cardiac death is defined as i) a witnessed natural death attributable to cardiac causes, heralded by abrupt loss of consciousness, within 1 h of onset of acute symptoms or ii) an unwitnessed, unexpected death of someone seen in a stable medical condition <24 h previously with no evidence of a non-cardiac cause [60, 61].

Sudden cardiac death accounts for more deaths each year than the total number of deaths from AIDS, breast cancer, lung cancer and stroke together. It is estimated that worldwide more than 3 million people die annually from sudden cardiac death [5].

The incidence of sudden cardiac arrest in the general Dutch population is 9–10 per 10 000 inhabitants each year [3, 62]. This can be extrapolated to approximately 40 sudden cardiac arrests per day in the Netherlands. In Dutch patients who had undergone 24 h ambulatory electrocardiography for various indications, 3.7% experienced sudden death within 2 years after electrocardiography [63].

Future directions

Most studies have focused on the interaction of drugs or other environmental factors with the known rare LQTS genes. However, future studies should also focus on interactions with common genetic variants. A genome-wide association study on acquired LQTS or the continuous QT interval duration in people with and without exposure to potential QT prolonging drugs might provide important new insights into drug-gene interactions and mechanisms of action. This approach requires a large number of well phenotyped samples to overcome the consequences of multiple testing with its risk of false positive signals despite Bonferroni corrections. These efforts are currently conducted in international consortia of research groups, amongst others due to the availability of more economically affordable commercial chips.

Conclusions

The long QT syndrome can be congenital or acquired. However, repolarization disturbances often occur due to additional gene-environment interactions. The congenital long QT syndrome, which is caused by mutations in ion channel subunits or regulatory protein coding genes, is quite rare. Recently, more common genetic variants that are associated with QT interval duration have been identified, which explain approximately 6% of the population variation in the QT interval.

Acquired QT prolongation is more prevalent than the congenital form. Several risk factors have been identified with use of QT_c prolonging drugs as the most frequent cause, and often the reason for withdrawal from the market or for restriction of use. Most drugs that prolong the QT_c interval act by blocking hERG-encoded potassium channels, although some drugs modify sodium channels. Both pharmacodynamic as well as pharmacokinetic effects may lead to QT_c prolongation. Pharmacokinetic interactions often involve drugs that are metabolized by cytochrome P450 enzymes. Gene–drug interactions are due to genetic variants that increase the QT_c prolonging effect of drugs. Future studies are necessary to establish the genetic background of acquired QT_c prolongation.

Competing interests

There are no competing interests to declare.