Prevention of Torsade de Pointes in Hospital Settings

HomeCirculationVol. 121, No. 8Prevention of Torsade de Pointes in Hospital Settings Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBPrevention of Torsade de Pointes in Hospital SettingsA Scientific Statement From the American Heart Association and the American College of Cardiology Foundation Barbara J. Drew, RN, PhD, FAHA, Chair, Michael J. Ackerman, MD, PhD, FACC, Marjorie Funk, RN, PhD, FAHA, W. Brian Gibler, MD, FAHA, Paul Kligfield, MD, FAHA, FACC, Venu Menon, MD, FAHA, FACC, George J. Philippides, MD, FAHA, FACC, Dan M. Roden, MD, FAHA, FACC, Wojciech Zareba, MD, PhD, FACC and Barbara J. DrewBarbara J. Drew , Michael J. AckermanMichael J. Ackerman , Marjorie FunkMarjorie Funk , W. Brian GiblerW. Brian Gibler , Paul KligfieldPaul Kligfield , Venu MenonVenu Menon , George J. PhilippidesGeorge J. Philippides , Dan M. RodenDan M. Roden , Wojciech ZarebaWojciech Zareba and and on behalf of the American Heart Association Acute Cardiac Care Committee of the Council on Clinical Cardiology, the Council on Cardiovascular Nursing, and the American College of Cardiology Foundation Originally published8 Feb 2010https://doi.org/10.1161/CIRCULATIONAHA.109.192704Circulation. 2010;121:1047–1060is corrected byCorrectionOther version(s) of this articleYou are viewing the most recent version of this article. Previous versions: February 8, 2010: Previous Version 1 Cardiac arrest due to torsade de pointes (TdP) in the acquired form of drug-induced long-QT syndrome (LQTS) is a rare but potentially catastrophic event in hospital settings. Administration of a QT-prolonging drug to a hospitalized population may be more likely to cause TdP than administration of the same drug to an outpatient population, because hospitalized patients often have other risk factors for a proarrhythmic response. For example, hospitalized patients are often elderly people with underlying heart disease who may also have renal or hepatic dysfunction, electrolyte abnormalities, or bradycardia and to whom drugs may be administered rapidly via the intravenous route.In hospital units where patients' electrocardiograms (ECGs) are monitored continuously, the possibility of TdP may be anticipated by the detection of an increasing QT interval and other premonitory ECG signs of impending arrhythmia. If these ECG harbingers of TdP are recognized, it then becomes possible to discontinue the culprit drug and manage concomitant provocative conditions (eg, hypokalemia, bradyarrhythmias) to reduce the occurrence of cardiac arrest.The purpose of this scientific statement is to raise awareness among those who care for patients in hospital units about the risk, ECG monitoring, and management of drug-induced LQTS. Topics reviewed include the ECG characteristics of TdP and signs of impending arrhythmia, cellular mechanisms of acquired LQTS and current thinking about genetic susceptibility, drugs and drug combinations most likely to cause TdP, risk factors and exacerbating conditions, methods to monitor QT intervals in hospital settings, and immediate management of marked QT prolongation and TdP.Characteristic Pattern of TdPThe term torsade de pointes was coined by Dessertenne in 1966 as a polymorphic ventricular tachycardia characterized by a pattern of twisting points.1 Several ECG features are characteristic of TdP and are illustrated in Figure 1. First, a change in the amplitude and morphology (twisting) of the QRS complexes around the isoelectric line is a typical feature of the arrhythmia; however, this characteristic twisting morphology may not be evident in all ECG leads. Second, episodes of drug-induced TdP usually start with a short-long-short pattern of R-R cycles consisting of a short-coupled premature ventricular complex (PVC) followed by a compensatory pause and then another PVC that typically falls close to the peak of the T wave.2 However, because of the underlying long-QT interval, this R-on-T PVC does not have the short coupling interval that is characteristic of idiopathic ventricular fibrillation. On the basis of experiments performed in isolated canine ventricular wedge preparations, this short-long-short sequence is thought to promote TdP by increasing heterogeneity of repolarization across the myocardial wall. Third, TdP episodes usually show a warm-up phenomenon, with the first few beats of ventricular tachycardia exhibiting longer cycle lengths than subsequent arrhythmia complexes. The rate of TdP ranges from 160 to 240 beats per minute, which is slower than ventricular fibrillation. Fourth, in contrast to ventricular fibrillation that does not terminate without defibrillation, TdP frequently terminates spontaneously, with the last 2 to 3 beats showing slowing of the arrhythmia. However, in some cases, TdP degenerates into ventricular fibrillation and causes sudden cardiac death. Download figureDownload PowerPointFigure 1. Onset of TdP during the recording of a standard 12-lead ECG in a young male with a history of drug addiction treated with chronic methadone therapy who presented to a hospital emergency department after ingesting an overdose of prescription and over-the-counter drugs from his parent's drug cabinet. Classic ECG features evident in this rhythm strip include a prolonged QT interval with distorted T-U complex, initiation of the arrhythmia after a short-long-short cycle sequence by a PVC that falls near the peak of the distorted T-U complex, "warm-up" phenomenon with initial R-R cycles longer than subsequent cycles, and abrupt switching of QRS morphology from predominately positive to predominately negative complexes (asterisk).The term torsade de pointes has also been used to describe polymorphic ventricular arrhythmias in which QT intervals are not prolonged. However, the term is better confined to those polymorphic tachycardias with marked (>500 ms) QT-interval prolongation and QT-U deformity, because they appear to be a distinct mechanistic and therapeutic entity.Premonitory ECG Signs of TdPLessons learned from research in large cohorts of individuals with congenital LQTS indicate that there is a gradual increase in risk for TdP as the heart rate–corrected QT interval (QTc) increases. Each 10-ms increase in QTc contributes approximately a 5% to 7% exponential increase in risk for TdP in these patients.3,4 Therefore, a patient with a QTc of 540 ms has a 63% to 97% higher risk of developing TdP than a patient with a QTc of 440 ms. There is no threshold of QTc prolongation at which TdP is certain to occur. Data from congenital LQTS studies5,6 indicate that a QTc >500 ms is associated with a 2- to 3-fold higher risk for TdP. Likewise, case reports and small series of patients with drug-induced TdP show similar increased risk when the threshold of QTc >500 ms is exceeded.7–9Although research in congenital LQTS indicates that the risk for syncope and sudden death varies directly with the duration of the QT interval,5 monitoring the QT/QTc intervals alone may be inadequate to accurately predict TdP.10 One reason QT monitoring alone may be inadequate is that it is difficult to measure this interval accurately in clinical practice and in clinical trials. Automated systems and human observers are reasonably adept at measuring QT intervals that have normal duration and morphology; however, establishing the end of the QT interval that is morphologically distorted is much more challenging and prone to interrater differences. The typical short-long-short sequence of R-R intervals seen before the initiation of TdP is associated with marked QT prolongation and T-U–wave distortion in the last sinus beat (terminating the long pause) before the episode. Distortion often involves changes in T-wave morphology such as T-wave flattening, bifid T waves, prominent U waves that are fused with T waves, and an extended and gradual sloping of the descending limb of the T wave, which makes it difficult to determine the end of the T wave. Some reports indicate that TdP is especially likely when the QT interval is prolonged because of an increase in the terminal portion of the T wave, from the peak of the T wave to its end (Tpeak-Tend).11,12In a patient with drug-induced LQTS, the QT interval may be prolonged during normal sinus rhythm without adverse effect, but after a pause (eg, after an ectopic beat or during transient atrioventricular block), QT-interval prolongation and T-U deformity become markedly exaggerated, and TdP is triggered. This beat-to-beat instability of the QT interval not only appears likely to influence the accuracy of measurement, but it may also be related to the underlying mechanism of the arrhythmia.13 In addition to an ever-increasing and distorted QT interval, another rare but ominous premonitory ECG sign of impending TdP is macroscopic T-wave alternans,14 as illustrated in Figure 2. In the future, it may be possible to assess risk by use of sophisticated T-U–wave morphology analysis; however, until such analysis becomes available, exaggerated QT-interval prolongation with T-U distortion after a pause should be considered a strong marker of risk for TdP. Download figureDownload PowerPointFigure 2. Top rhythm strip, TdP degenerating into ventricular fibrillation in an 83-year-old female hospitalized in the intensive care unit for pneumonia. She was started on intravenous erythromycin several hours before cardiac arrest. A ventricular couplet followed by a pause provided the short-long-short cycle sequence that triggered TdP. Bottom rhythm strip, ECG 1 hour before the onset of TdP shows extreme prolongation of the QT interval (QTc in cycles with larger T waves=730 ms), a ventricular couplet (asterisk), and macroscopic T-wave alternans (vertical arrows). If these signs of impending TdP had been recognized, discontinuation of the culprit drug and administration of magnesium most likely would have prevented the subsequent cardiac arrest.Cellular Mechanisms of Acquired LQTSProlongation of the QT interval, changes in T-U wave morphology, and subsequent TdP are results of abnormal function (and structure) of ion channels and related proteins involved in the repolarization process in ventricular myocytes. These abnormalities can be caused by mutations of genes that encode ion channels or associated proteins in congenital forms of LQTS; however, they can also be caused by the action of drugs in acquired LQTS. Drugs with the potential to cause TdP most frequently inhibit the rapid component of the delayed rectifier potassium current (IKr), which causes a reduction in the net repolarizing current and results in prolongation of the ventricular action potential duration and a prolonged QT interval on the ECG.15Experiments in canine ventricular wedge preparations have shown that in normal circumstances, there are differences in repolarization in the various layers of the myocardium, with the subepicardium having the shortest action potential duration, the subendocardium having an intermediate duration, and the mid myocardium (M cells) having the longest action potential duration.16,17 However, because the myocardial layers are tightly coupled in the intact human heart, such differences are small. Many reports indicate that the QT interval on the ECG represents the longest repolarization in the M-cell region. This physiological transmural dispersion of repolarization usually does not lead to TdP; however, proarrhythmic states may arise as a result of specific gene mutations or actions of medications that cause selective action potential prolongation in certain layers of the myocardium (usually the M-cell region) that lead to increased transmural repolarization gradients.17 This increased transmural gradient is thought to create the conditions for reentry and subsequent TdP.The trigger for TdP is thought to be a PVC that results from an early afterdepolarization generated during the abnormally prolonged repolarization phase of the affected myocardium.18 A long preceding pause increases the amplitude of early afterdepolarizations, which makes them more likely to reach the threshold necessary to produce a PVC or ventricular couplet. Because of the marked delay of repolarization in certain areas of the myocardium, conduction of the PVC is blocked initially in some directions but not in others, which sets up reentry that perpetuates TdP.Not all QT-prolonging drugs are associated with risk for TdP. Therefore, it appears that QT prolongation alone is insufficient and that heterogeneity of repolarization may also be necessary to produce an arrhythmogenic response. However, the mechanisms whereby not all QT prolongation confers the same degree of risk are not well established.Experts in electrocardiography, including members of this writing group, have been curious about the peculiar pattern of sine-wave QRS changes with TdP. El-Sherif et al19 provided an electrophysiological mechanism for the characteristic periodic transition of the QRS axis during TdP. In an experimental setting, they demonstrated that the initial beat of TdP arose as a subendocardial focal activity, whereas subsequent beats were due to reentrant excitation in the form of rotating scrolls. The arrhythmia ended when reentrant excitation was terminated. The transition in the QRS axis coincided with a transient bifurcation of the predominantly single rotating scroll into 2 simultaneous scrolls that involved both the right ventricle and left ventricle separately. The common mechanism for the initiation or termination of this bifurcation was the development of functional conduction block between the anterior or posterior right ventricular free wall and the ventricular septum.Genetic Susceptibility to Drug-Induced TdPIt is becoming increasingly evident that genetic susceptibility, whether due to the presence of rare LQTS-causing mutations or the presence of functional common polymorphisms, must be considered in the patient who manifests drug-induced QT prolongation and TdP. Since the sentinel discovery of congenital LQTS as a channelopathy with mutations identified in genes encoding voltage-gated potassium and sodium channels in 1995,20,21 nearly 1000 individually rare LQTS-causing mutations have now been detected in 12 distinct LQTS-susceptibility genes. Three of the 12 LQTS-susceptibility genes (KCNQ1-encoded IKs α-subunit [LQT1], KCNH2-encoded IKr α-subunit [LQT2], and SCN5A-encoded Nav1.5 α-subunit [LQT3]) are the major LQTS-susceptibility genes, accounting for nearly 75% of all congenital LQTS cases.22Approximately two thirds of LQTS stems from loss-of-function mutations in either KCNQ1 or KCNH2 whereby there is a perturbation in phase 3 repolarization that results in a prolongation in the action potential duration and hence QT-interval prolongation. These defects provide the pathogenic substrate on which an ill-timed PVC and its cellular early afterdepolarization can precipitate TdP. Besides the predominant mechanism of potassium channel loss of function, approximately 5% to 10% of LQTS stems from gain-of-function mutations in the sodium channel whereby the mutations (mostly missense, ie, single amino acid substitutions) produce a sodium channel with a marked accentuation in late sodium current. Rather than shutting down within the first 5 ms of a cardiac action potential, this persistent but relatively small influx of inward sodium current disrupts phase 2 of the fine-tuned balance of the action potential, which prolongs the cellular action potential duration and confers the substrate for TdP. In addition to these 3 major LQTS-susceptibility genes that account for 75% of congenital LQTS, 9 minor LQTS-susceptibility genes account for an additional 5%. The remaining 20% of congenital LQTS cases remain genotype negative.From 1995 to 2004, research-based LQTS genetic testing revealed a plethora of genotype-phenotype relationships, including genotype-suggestive ECG patterns, arrhythmogenic triggers, and genetically determined responses to pharmacotherapy. In 2004, LQTS genetic testing matured into a clinically available test because of its established diagnostic, prognostic, and therapeutic implications. Just as a period of time (eg, during swimming or during the postpartum period) can suggest the presence of congenital LQTS, drug-induced long QT and TdP may also signal the presence of an LQTS genetic defect. In fact, the yield from LQTS genetic testing with respect to the 3 major LQTS-susceptibility genes is approximately 10% to 15% in individuals with isolated drug-induced acquired LQTS.23–25In addition to these individually rare mutations that confer susceptibility for the primary channelopathy known as congenital LQTS, which affects approximately 1 in 2500 persons, numerous common polymorphisms in these same cardiac channel genes have been identified, and some are now known to contribute to a reduced repolarization reserve and confer a modifier effect.26 For example, SCN5A-S1103Y is one of the most common polymorphisms in black Africans, 10% to 15% of whom may be heterozygous for this common, nonsynonymous single-nucleotide polymorphism. SCN5A-S1103Y is now known to produce or acquire a cellular phenotype of accentuated late sodium current (LQT3-like) when exposed to cellular acidosis and confer clinical susceptibility to proarrhythmia and premature sudden death as early as infancy in African Americans.27–30 In addition, KCNE2-Q9E was published originally as a rare, LQT6-causing missense mutation after its identification in a 76-year-old African American female with profound QT prolongation and TdP who required defibrillation after 7 doses of intravenous erythromycin and 2 doses of oral clarithromycin, both of which are known IKr blockers.31 Functional studies demonstrated that an IKr complex containing Q9E in the KCNE2-encoded β-subunit resulted in a potassium channel with a marked increase in sensitivity to hERG (human ether-a-go-go) block by clarithromycin. However, in contrast to its initial impression of rarity (absence in more than 2000 control alleles of unspecified ethnicity), KCNE2-Q9E is a relatively black-specific common polymorphism present in approximately 3% to 5% of African Americans.32Case series of drug-induced TdP (usually involving antiarrhythmic agents) identify subclinical congenital LQTS in 5% to 20% of cases.23–25 However, the extent to which the congenital LQTS confers risk during administration of drugs is not well understood. To illustrate the lack of clarity about genetic susceptibility and drug risk, moxifloxacin is a drug that very rarely causes TdP; however, the risk does not appear to increase even in the presence of congenital LQTS for the following reason. The incidence of TdP with moxifloxacin is very low, 1:100 000 to 1:1 000 000 exposures. Moxifloxacin is pharmacokinetically well behaved, with no known drug interactions or organ dysfunction that severely alters plasma concentrations. Given the fact that the mutations associated with congenital LQTS occur in 1 in 2500 individuals in the population,33 it appears irrefutable that many patients with congenital LQTS have been exposed to the drug without adverse effects.This kind of logic points to a likely distinction between high- and low-risk drugs. For example, the high-risk drugs, such as antiarrhythmic agents, methadone, and haloperidol, may increase risk for TdP in individuals with genetic mutations, whereas the low-risk drugs, such as moxifloxacin, may require other risk factors such as electrolyte disorders.Drugs That Cause TdP: Incidence and Other FeaturesWhen sudden death occurs without autopsy evidence for an explainable cause of death, an arrhythmic death is assumed. However, the proportion of sudden arrhythmic deaths that are due to TdP is unclear, because few individuals are being monitored at the time of death. When TdP occurs in outpatient settings, the first responders who arrive on the scene with portable monitor-defibrillators are likely to observe ventricular fibrillation. In this situation, it is impossible to determine whether ventricular fibrillation was preceded by QT prolongation and TdP. In hospital settings, the same lack of clarity about the arrhythmia mechanism that caused the cardiac arrest may occur if a patient is not undergoing continuous ECG monitoring at the time of arrest. Postarrest ECG changes are not uncommon, and a link to LQTS may not be made. For example, the postarrest QT interval may be prolonged because of the hypoxic/anoxic insult, or it may be quite short, presumably due to elevated potassium in this setting.Preclinical and early-phase clinical testing of new drugs may reveal a QT-prolongation signal that may be identified by consulting the drug label. Use of a QT-prolonging drug must be based on risk-benefit analysis in individual patients, and where efficacy of alternatives is equivalent, the non–QT-prolonging agent should be preferred. Where benefit clearly outweighs risk, QT prolongation should not limit necessary therapy. QT prolongation is not necessarily equivalent to arrhythmogenicity. The only class of drugs for which reasonable TdP incidence data are available is the antiarrhythmic agents. Those known to prolong the QT interval and block sodium and potassium channels (older drugs such as quinidine, disopyramide, and procainamide), as well as those that block potassium channels (sotalol, dofetilide, ibutilide), appear to have a TdP incidence of 1% to 10%.34 For the older drugs, the numbers are derived from uncontrolled case series, whereas for the newer agents, summary data from clinical trials and new drug applications are available.35–40Many non-antiarrhythmic drugs have also been associated with TdP. For some drugs, multiple case reports and small case series confirm that the drug causes the arrhythmia. High-profile examples include methadone,41 thioridazine,42 and haloperidol.43 Although the absolute TdP incidence is difficult to establish from these reports of non-antiarrhythmic agents, it is generally believed to be less than that reported for antiarrhythmic agents.IKr inhibition is a very common effect of many drugs, and case reports and small series implicate the involvement of many such drugs with TdP. Some of these are widely and commonly used, such as erythromycin and droperidol. In addition, a number of drugs have been withdrawn from the market or relabeled because of the risk for TdP.26 The absolute incidence figures are difficult to establish but appear to be very small, even in these cases of banned drugs. Thus, for example, nearly one hundred million prescriptions for the antihistamine terfenadine had been written before a very small risk for TdP was recognized. The overall incidence of TdP with terfenadine is exceedingly small and appears to be confined largely to patients with specific risk factors related to metabolism of the drug.44For virtually all QT-prolonging drugs, risk increases as a function of dose and, more specifically, plasma drug concentration, with the exception of quinidine. Quinidine is a potent IKr blocker,45 so at low concentrations it may prolong action potentials, whereas this effect may be blunted (by the drug's sodium channel–blocking properties) at higher concentrations, which explains the clinical observation that quinidine-induced TdP often occurs at low concentrations.The high-potency IKr blocker terfenadine undergoes near-complete presystemic metabolism, mediated largely by a specific hepatic cytochrome P450 (CYP3A4). Both terfenadine and its metabolite fexofenadine are potent antihistamines, but fexofenadine does not block IKr8; nevertheless, there is 1 case report of fexofenadine-related TdP.46 The vast majority of cases of terfenadine-associated TdP were associated with inhibition of CYP3A4 due to advanced liver disease, overdose, or ingestion of specific inhibitor drugs, notably erythromycin and ketoconazole. Erythromycin itself can also cause TdP, almost always with high doses or with use of the intravenous route and often in patients with other risk factors.47The problem of dramatic drug accumulation due to use of high doses, dysfunction of organs of elimination, or interacting drugs applies to other situations. Dofetilide and sotalol are cleared by the kidneys, and the use of ordinary doses in patients with renal failure increases TdP risk with these drugs. Procainamide undergoes hepatic clearance to an active metabolite, N-acetylprocainamide (NAPA), which has IKr-blocking properties. NAPA itself is eliminated by the kidneys, so patients with renal dysfunction may develop NAPA-related TdP during procainamide therapy.37 Thioridazine is bioinactivated by CYP2D6, and subjects with deficient activity of this enzyme due to genetic factors (5% to 10% of white and black populations) or the use of CYP2D6-inhibiting drugs such as quinidine, fluoxetine, or paroxetine have higher plasma drug concentrations.48Case series of methadone-related TdP indicate that the use of high doses and/or recent dose increases are common clinical features of affected patients.41 Methadone is cleared by multiple pathways; although inhibiting drugs have been implicated, their precise role is unclear at this time. Nearly 1 million Americans use methadone for narcotic dependence or for chronic pain therapy.49 Recently published methadone clinical guidelines recommend a pretreatment ECG for QTc interval screening and a follow-up ECG within 30 days and then annually.49The risk for TdP should be evaluated in any patient who presents to the emergency department with an overdose of a QT-prolonging drug. However, because it is often unclear what drug or combination of drugs the patient may have taken, the ECG of all drug overdose victims should be assessed for signs of prolonged QT, QT-U distortion, and other signs of impending TdP (Figure 2). The tricyclic antidepressants such as amitriptyline can cause TdP, although the incidence is not well established and other arrhythmias due to sodium channel blocker toxicity (eg, wide QRS and sinusoidal ventricular tachycardia) may also be present. Less frequent use of these antidepressants for outpatient treatment of depression has decreased the presentation of patients with an overdose of these agents. Because depressed patients are the most susceptible to purposeful drug overdoses, pharmaceutical manufacturers have attempted to create multiple new antidepressants such as selective serotonin reuptake inhibitors for use in depression.50 Despite this, TdP has been reported in patients with overdoses of these medications, such as citalopram.50 Other nontricyclic antidepressants, such as trazodone, have also been implicated in TdP in patients with purposeful overdose.51 Moreover, a recent analysis of current users of older typical versus newer atypical antipsychotic agents revealed that both groups had a similar dose-related increased risk of sudden cardiac death compared with matched nonusers of antipsychotic drugs.52Chronic administration of amiodarone markedly prolongs the QT interval, yet it is very rarely associated with TdP.53 It has been postulated (although as yet unproven) that unlike high-risk drugs that selectively prolong repolarization in myocytes located in the mid myocardium (M cells), amiodarone uniformly delays repolarization in all layers of the myocardial wall. As a result, there is only QT prolongation and no transmural heterogeneity of repolarization, which is the necessary substrate for the development of a reentrant arrhythmia. Another theory regarding the low TdP risk nature of amiodarone suggests that the drug also inhibits the physiological late sodium currents that ultimately produce the arrhythmia.54This theory also applies to verapamil, a relatively potent IKr blocker55 that has never been associated with TdP, probably because it is a much more potent blocker of L-type calcium channels. The newer antianginal agent ranolazine also blocks IKr,56,57 but the extent of the QT prolongation appears limited during long-term therapy, probably because the drug also blocks the physiological late sodium current. In a large clinical trial, ranolazine was not associated with an increased incidence of TdP.58Intravenous administration can be associated with higher drug concentrations and greater cardiac exposure than corresponding oral dosing. Thus, the intravenous route may be a risk factor for TdP. In addition, there are provocative data from an animal model of TdP that suggest that rapid infusion may be more likely to cause the arrhythmia than slower infusion (of higher drug doses).59 The mechanism underlying this effect is unknown but may reflect differential drug delivery to various sites within the myocardium.The Arizona Center for Education & Research on Therapeutics maintains an updated list of drugs that have a risk of causing TdP on their World Wide Web site at www.qtdrugs.org. Table 1 shows a drug list from this World Wide Web site that has been modified to exclude amiodarone (regarded as low risk) and drugs that are no longer available in the United States. Table 1 represents the most common drugs that can be implicated in TdP, but it is not a complete list of all reported possible contributing substances. Importantly, the drugs listed in Table 1 are not equipotent in their risk of causing TdP. For example, the risk of TdP ranges from approximately 0.001% for Propulsid (cisapride) to approximately 8% for the antiarrhythmic quinidine. We also reemphasize that the use of these medications may be clearly indicated from a risk-benefit perspective despite the presence of the possibility of drug-induced TdP. For example, a recent analysis of a large number of surgical patients (>290 000) revealed no change in the incidence of TdP in patients who received antiemetic therapy with