What Clinicians Should Know About the QT Interval

CLINICAL CARDIOLOGY CLINICIAN’S CORNER

Sana M. Al-Khatib, MD, MHS Context Of the several factors implicated in causing QT interval prolongation and Nancy M. Allen LaPointe, PharmD torsades de pointes, errors in the use of medications that may prolong this interval deserve special attention. Judith M. Kramer, MD, MS Objective To systematically summarize the available clinical data on the QT interval Robert M. Califf, MD and to offer improved recommendations for the use of QT-prolonging medications. HE QT INTERVAL ON THE ELEC- Data Sources We searched MEDLINE from 1966 through 2002 for all English- trocardiogram (ECG) has language articles related to the QT interval. Additional data sources included bibliog- gained clinical importance, raphies of articles identified on MEDLINE, a survey of experts, and data presented at primarily because prolonga- a meeting of experts on long QT syndrome. tionT of this interval can predispose to Study Selection We selected for review registries and case series examining clini- a potentially fatal ventricular arrhyth- cal outcomes of patients with prolonged QT interval and the effect of different meth- mia known as torsades de pointes. Mul- ods of measurement of the QT interval on patient outcomes. Ten studies were identified, of which 6 were included in the analysis. tiple factors have been implicated in causing QT prolongation and torsades Data Extraction Data quality was determined by publication in the peer-reviewed de pointes. Among these, improper use literature. of QT interval–prolonging medica- Data Synthesis Optimal measurement of the QT interval is problematic because of tions deserves special attention. Re- lack of standardization and lack of data regarding the best way to adjust for heart rate. cently, cisapride and grepafloxacin were Reliable information on the proper use of QT-prolonging medications is scarce. Al- removed from the US drug market be- though a QT interval of at least 500 milliseconds generally has been shown to correlate with a higher risk of torsades de pointes, there is no established threshold below cause of the risk for QT prolongation which prolongation of the QT interval is considered free of proarrhythmic risk. The 1,2 and fatal arrhythmias. The need to re- risk of torsades de pointes should be assessed in patients who are about to begin tak- move these agents from the market was ing a QT-prolonging medication. Although inadequate clinical studies preclude pre- related not just to the inherent prop- diction of absolute risk for individual patients, particularly high-risk situations can be erties of the drugs but also to the dem- defined based on clinical variables. We propose recommendations on proper moni- onstrated failure of government- toring of the QT interval in patients receiving QT-prolonging medications. mandated black box warnings and Conclusion Although the use of QT-prolonging medications can predispose to tor- “Dear Doctor” letters to mitigate inap- sades de pointes, there is a relative paucity of information that can help clinicians and propriate prescribing by physicians.3 patients make optimal informed decisions about how best to minimize the risk of this To reduce the risk of torsades de serious complication. pointes, health care providers must un- JAMA. 2003;289:2120-2127 www.jama.com derstand what is known about the QT interval. In this article, we address the METHODS 2002 related to the QT interval (search meaning and measurement of the QT Literature for this review was system- terms: long QT syndrome, death, out- interval, describe factors that affect the atically identified by searching comes, registries, case series, QT inter- QT interval, and assess the balance of MEDLINE for all English-language ar- val, and measurement), reviewing bib- risks and benefits of QT-prolonging ticles published from 1966 through liographies of articles identified on medications. We also evaluate the steps Author Affiliations: Duke Center for Education and Johnson & Johnson, Pfizer, and Eli Lilly, which all sell that have been taken to enhance proper Research on Therapeutics, Duke Clinical Research In- QT-prolonging drugs described in the article. None management of risk emanating from stitute, Durham, NC. of these research grants pertain to QT-prolonging Financial Disclosures: AstraZeneca and GlaxoSmith- drugs. the use of QT interval–prolonging Kline (GSK) have contributed unrestricted funds to Corresponding Author and Reprints: Sana M. Al- medications. projects of the Duke Center for Education and Re- Khatib, MD, MHS, Duke Clinical Research Institute, search on Therapeutics. Dr Al-Khatib is involved in a PO Box 17969, Durham, NC 27715 (e-mail: alkha001 research project sponsored and funded by GSK. This @mc.duke.edu). See also p 2041 and Patient Page. project is not related to the QT-prolonging drugs made Clinical Cardiology Section Editor: Michael S. Lauer, by GSK. Dr Califf has received research funding from MD, Contributing Editor.

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

MEDLINE, surveying experts, and re- Figure. Measuring the QT Interval in Different Clinical Scenarios viewing data presented at a meeting of experts on long QT syndrome (LQTS). A Normal Sinus Rhythm B Atrial Fibrillation We selected for review registries and case series examining clinical out- RR RR comes of patients with prolonged QT 1 2 interval and the effect of different methods of measurement of the QT inter- RR

val on patient outcomes. Ten studies R

were identified by the search, of which P T 6 were included in the analysis.4-9 Data Q quality was determined by publica- S QT 1 QT 2 tion in the peer-reviewed literature. QT

QT1 QT2 QT QTc = QTc = WHAT IS THE QT INTERVAL QTc = 1 √ 2 √ RR1 RR2 AND HOW SHOULD IT BE √RR Bazett Formula MEASURED? QTc + QTc = 1 2 QTc The QT interval on the surface ECG is 2 measured from the beginning of the QRS complex to the end of the T wave. Thus, it is the electrocardiographic QTc indicates corrected QT interval; RR, R-R interval. A, Normal sinus rhythm; the Bazett formula is used to correct the QT interval for the heart rate. B, Atrial fibrillation; QT interval is calculated by taking the average of manifestation of ventricular depolar- QT intervals with shortest and longest preceding R-R intervals. ization and repolarization. This elec- trical activity of the heart is mediated through channels, complex molecular ity, resulting from variations in T-wave longed at slower heart rates and short- structures within the myocardial cell morphology, noisy baseline, and the ened at faster heart rates, many formulas membrane that regulate the flow of ions presence of U waves. Interobserver vari- have been proposed to adjust for these in and out of cardiac cells. The rapid ability also results from the lack of variations. Yet differences of opinion ex- inflow of positively charged ions (so- agreement among experts about stan- ist regarding the most useful correc- dium and calcium) results in normal dardizing approaches to measure the tion for heart rate.15-18 One of the com- myocardial depolarization. When this QT interval.11,12 Although experts on the monly used formulas is the Bazett inflow is exceeded by outflow of po- QT interval argue that intraobserver and formula, in which the QT interval is ad- tassium ions, myocardial repolariza- interobserver variability and measure- justed for heart rate by dividing it by tion occurs. Malfunction of ion chan- ment error are higher when the cor- the square root of the R-R interval nels leads to an intracellular excess of rected QT (QTc) interval is taken from (FIGURE, A). However, this formula has positively charged ions by way of an in- computerized ECG algorithms rather been criticized for being inaccurate at adequate outflow of potassium ions or than from careful high-resolution fast heart rates.19 Other formulas are the excess inflow of sodium ions. This in- manual measurements, automated Fridericia cube-root correction (QT tracellular excess of positively charged readings may be useful for rapid assess- interval divided by the cube root of ions extends ventricular repolariza- ment of patient safety.13 Unfortu- the R-R interval) and the Framingham tion and results in QT interval prolon- nately, there is no credible empirical linear regression equation.16,17 From gation.10 evidence to support this view. In addi- an epidemiological perspective, the In the clinical setting, it is now widely tion, as demonstrated by a recent sur- Framingham approach is the most recognized that typical measurement of vey of health care practitioners, many sound because it is based on empirical the QT interval is subject to substan- clinicians simply do not know how to data from a large population sample tial variability, which can cloud inter- measure the QT interval. Whereas 61% rather than on hypothetical reason- pretation.11,12 This variability in QT in- of respondents were able to identify ing. Unfortunately, none of these cor- terval measurement results from what the QT interval represented on an rections has been examined compar- biological factors, such as diurnal ef- ECG, only 36% correctly measured it.14 atively to determine the most effective fects, differences in autonomic tone, Although it is standard practice to formula in predicting which patients are electrolytes, and drugs; technical fac- measure the QT interval from the beat greatest risk for torsades de pointes. tors, including the environment, the ginning of the QRS complex to the end A group of experts on LQTS re- processing of the recording, and the ac- of the T wave, the actual methods of cently acknowledged the lack of em- quisition of the ECG recording; and in- measurement have not been standard- pirical data in determining the best ap- traobserver and interobserver variabil- ized. Because the QT interval is pro- proach to measuring the QT interval.

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

This group convened in August 2000 sure the QT interval in this setting; data equate statistical power. In the ab- to discuss the current knowledge of on the best way to make this measure- sence of such studies, practitioners must LQTS (see Acknowledgment). As a re- ment do not exist. The Pfizer Tikosyn be aware of the ongoing uncertainty sult of this meeting, the panel pro- program specifies that while the QTc about the best way to adjust the QT posed the following 4 guidelines13 for should be no more than 440 millisec- interval. measuring the QT interval, based on ex- onds (ms) to start dofetilide in the set- pert opinion: ting of a narrow QRS complex, the QTc FACTORS THAT AFFECT 1. The QT interval should be mea- should be no more than 500 ms in the THE QT INTERVAL sured manually, preferably by using one setting of ventricular conduction ab- Although it is convenient to think of QT of the limb leads that best shows the end normality.20 This guidance may be used prolongation as occurring because of ei- of the T wave on a 12-lead ECG. with other QT-prolonging medica- ther congenital or acquired abnormali- 2. The QT interval should be mea- tions until a standard method to mea- ties, the phenomenon probably most sured from the beginning of the QRS sure the QT interval in the setting of often involves a gene-environment in- complex to the end of the T wave and ventricular conduction abnormality is teraction. Pure congenital prolonga- averaged over 3 to 5 beats. U waves pos- identified. tion characterized by lifelong, ambi- sibly corresponding to the late repo- Some argue that the QT interval ent QT prolongation is rare but does larization of cells in the mid myo- should be measured only by cardiolo- carry a high risk of sudden death. Sev- cardium should be included in the gists, but this suggestion is impracti- eral forms of congenital LQTS have measurement only if they are large cal. In light of the number of medica- been reported, and 3 forms (LQT1, enough to seem to merge with the tions that could prolong the QT LQT2, and LQT3) have been well char- T wave. interval, other health care practition- acterized in previous studies.21 These 3. The QT interval should be mea- ers, especially internists, family prac- forms have been found to have dis- sured during peak plasma concentra- titioners, and psychiatrists, should ei- tinctly different clinical outcomes and tion of a QT-prolonging medication. ther learn how to measure the QT clinical manifestations, including fac- 4. The QT interval should be ad- interval or develop systematic ap- tors that trigger clinical events and ECG justed for heart rate. Because the best proaches to ensuring that accurate mea- features.6,22 For example, physical ac- way to adjust for heart rate has not been surements are being made at the ap- tivity tends to trigger events in LQT1, determined by prospective studies, the propriate time by specialists. Indeed, auditory stimuli in LQT2, and rest or panel could not make a definitive rec- nurses, physician assistants, and clini- sleep in LQT3.7,23,24 Each form has also ommendation in this regard. cal pharmacists may play an impor- been characterized electrocardiographi- Measurement of the QT interval is tant role in this regard; if properly cally by a specific pattern of T waves.25 particularly challenging if the patient trained, it is likely that they can be re- The T wave is of long duration in LQT1, is in atrial fibrillation because the QT lied on to measure the QT interval. is small and/or notched in LQT2, and interval varies from beat to beat de- However, the use of multiple health care has an unusually long onset in LQT3.22 pending on the interval between suc- practitioners in measuring the QT in- More important, the genotype of LQTS cessive R waves. Unfortunately, there terval for clinical decision making needs seems to have a significant impact on is no consensus on how to measure the to be tested in prospective studies. outcome.4 In a study using a large in- QT interval in this circumstance. Some It is important to realize that the ternational registry of LQTS, it was clinicians suggest using the same steps methods proposed to correct the QT in- noted that although the risk of cardiac in the aforementioned recommenda- terval have primarily been evaluated for events was significantly higher among tions but further advise averaging the their correlation with heart rate. The patients with LQT1 and LQT2 than with measured QT interval over 10 beats. formula with the best correlation with LQT3, the frequency of lethal cardiac Others prefer to measure the QT inter- heart rate is believed to be the most ac- events was significantly higher in the vals that follow the shortest and long- curate. It would be of great clinical im- LQT3 group.6 est R-R intervals and divide each by the portance if these formulas were vali- When exposed to QT-prolonging square root of the R-R interval preced- dated prospectively and then compared medications, individuals without life- ing it. The average of these intervals with each other in adequately sized prolong QT prolongation may develop QT would then be used as the adjusted QT spective studies. In this way, it could prolongation with or without torsades interval (Figure, B). be determined which one most strongly de pointes or may not develop QT pro- Measurement of the QT interval is correlates with an increased risk of ad- longation at all. Even after adjustment also difficult in the setting of a wide QRS verse clinical events (especially death). for other factors that could prolong QT complex related to either ventricular However, this endeavor would be chal- interval, some patients seem to be more conduction defects or a paced QRS lenging because it might be difficult to likely than others to have QT prolon- complex. This is primarily because of identify a patient population with an gation at a given dose of a drug. This the lack of a standard method to mea- event rate high enough to provide ad- observation led researchers to hypoth-

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

esize that patients with acquired QT Box. Potential of Selected Medications for Causing QT Prolongation prolongation may have a genetic pre- * Based on a Survey of Expert Opinion disposition for it.8,10,26 Recent investi- VERY PROBABLE IMPROBABLE gations suggest that such patients may Antiarrhythmics Anti-infectives have clinically silent gene mutations Amiodarone Fluconazole that lead to overt QT prolongation only Disopyramide Levofloxacin with exposure to QT-prolonging medi- 8,10,26 Dofetilide Trimethoprim-sulfamethoxazole cations. It is important to note that the ma- Ibutilide Antidepressants Procainamide jority of patients with documented Fluoxetine Quinidine acquired LQTS never experience tor- Paroxetine Sotalol sades de pointes, and many patients Antipsychotics Migraine Drugs with torsades de pointes have a nor- Thioridazine Sumatriptan mal QT interval shortly before the Zolmitriptan event. It appears that a variety of coin- PROBABLE Other cident circumstances, including ge- Antipsychotics Methadone netic predisposition and a prolonged QT interval (which may occur precipi- Pimozide VERY IMPROBABLE Ziprasidone tously and transiently), are required to Anti-infectives precipitate torsades de pointes. POSSIBLE IN HIGH-RISK PATIENTS Azithromycin Factors that predispose to QT prolongation and higher risk of torsades de Anti-infectives Ciprofloxacin Clindamycin pointes include older age, female sex, Clarithromycin low left ventricular ejection fraction, left Erythromycin Other ventricular hypertrophy, ischemia, slow Gatifloxacin Isradipine Pentamidine Nicardipine heart rate, and electrolyte abnormali- Sparfloxacin ties including hypokalemia and hypo- 5,27-34 UNKNOWN magnesemia. Certain drugs also Antipsychotics predispose to QT prolongation (BOX). Chlorpromazine Antipsychotics An extensive list of these drugs can be Haloperidol Mesoridazine found at http://www.torsades.org. Re- Olanzapine Quetiapine garding antiarrhythmic QT-prolonging Risperidone Antidepressants drugs, the risk of torsades de pointes Antidepressants Doxepin seems to be highest within the first few 35-37 Amitriptyline Other days of initiating therapy. For this Desipramine Chloroquine reason, physicians should consider ad- Imipramine Domperidone mitting patients to the hospital when Sertraline Felbamate starting such drugs, a practice most war- Venlafaxine Foscarnet ranted among patients with structural Other Fosphenytoin heart disease. Hospitalized patients can Droperidol Indapamide be better monitored for the warning Moexipril/hydrochlorothiazide signs that precede torsades de pointes. Octreotide In a rigorous study of patients with su- Ondansetron praventricular tachycardias, investiga- Quinine tors reported that a 72-hour hospital- Tacrolimus Tamoxifen ization for initiation of antiarrhythmic 38 Vasopressin therapy appeared to be cost-effective. Several noncardiac medications can *“Very probable” indicates more than 50% of respondents stated they would always check cause torsades de pointes, either by di- an electrocardiogram (ECG) when starting this medication; “probable,” 40%-49% of respondents stated they would always check an ECG when starting this medication; “possible in rectly blocking potassium currents or high-risk patients,” more than 40% of respondents stated they would always check an ECG by interacting with other medications in high-risk patients; “improbable,” 40%-49% of respondents stated they would never check (TABLE). These interactions could be an ECG when starting this medication; “very improbable,” more than 50% of respondents stated they would never check an ECG when starting this medication; and “unknown,” sur- purely pharmacodynamic (both drugs vey responses did not fit any of the other categories. block outward potassium currents), purely pharmacokinetic (one drug in-

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

tients with symptoms of chronic Table. Pharmacokinetic Interactions With Selected QT-Prolonging Drugs* arrhythmias, the risk may outweigh the QT-Prolonging Drugs Drugs Possibly Affecting Pharmacokinetics benefit because few studies have shown Antiarrhythmics Disopyramide Erythromycin a significant effect of antiarrhythmic 43 Dofetilide Cimetidine, ketoconazole, megesterol, prochlorperazine, therapy in this situation. The risks are trimethoprim, verapamil, thiazide diuretics particularly disconcerting with antiar- Procainamide Amiodarone, cimetidine, trimethoprim rhythmic medications that have been Quinidine Amiodarone, cimetidine, possibly erythromycin and verapamil shown to worsen survival.44-46 Antipsychotics The risk of torsades de pointes should Haloperidol Fluoxetine, venlafaxine be assessed for patients who are about Pimozide Erythromycin to begin taking a QT-prolonging medi- Thioridazine Paroxetine cation. Although inadequate clinical Ziprasidone Fluconazole, itraconazole, ketoconazole Antidepressants studies preclude prediction of abso- Amitriptyline Cimetidine, fluconazole, fluoxetine, ritonavir lute risk for individual patients, par- Desipramine Venlafaxine ticularly high-risk situations can be de- Anti-infectives fined based on clinical variables. This Erythromycin Ritonavir estimate requires knowledge of the Sparfloxacin Cisapride drug’s properties, including route of Other Bepridil Ritonavir elimination and drug interactions, fa- Cisapride Clarithromycin, erythromycin, fluconazole, indinavir, itraconazole, miliarity with factors that predispose to ketoconazole, nefazodone, ritonavir torsades de pointes, and baseline mea- *Drugs from the “very probable,” “probable,” and “possible in high-risk patients” categories of the Box are included in surement of the QT interval. For ex- this table. This is not an all-inclusive list of all pharmacokinetic drug-drug interactions with these agents but, rather, some interactions that could lead to increased serum concentrations of the QT-prolonging drug. New drug-drug ample, sotalol and dofetilide are re- interactions may be identified in the future. Pharmacodynamic interactions are not included in this table; however, nally cleared; thus, it is important to combinations of QT-prolonging drugs such as macrolide antibiotics and quinolones are strongly discouraged.39-42 monitor the renal function of patients starting these medications and to re- terferes with the clearance of another), minimized. Specifically, they should de- duce the dose if renal function is im- or of mixed pharmacodynamic and termine whether the likely benefit jus- paired. To avoid risk of torsades de pharmacokinetic origin. An example of tifies the potential risk and should do so pointes, physicians should be aware that a purely pharmacodynamic interac- in light of the treated condition, the spe- dofetilide has significant interactions tion is that of quinidine and sotalol— cific circumstances of the patient, and with commonly used drugs such as both block outward potassium cur- other available therapeutic options. For verapamil and trimethoprim-sulfa- rents. The interaction between cisapride example, a QT interval–prolonging methoxazole. and ketoconazole provides an ex- medication is the appropriate choice if Among the factors that could pre- ample of a purely pharmacokinetic in- it has a proven salutary effect on sur- dispose to torsades de pointes, hypo- teraction. Ketoconazole inhibits the cy- vival. However, the majority of these kalemia and hypomagnesemia are par- tochrome P-4503A4 isoenzyme that medications have not been proven to im- ticularly significant and remediable. metabolizes cisapride, and this inhibi- prove survival. Another strong reason to Physicians should monitor potassium tion results in increased cisapride lev- use such a medication is if it will sig- and magnesium levels in patients who els that may augment QT prolonga- nificantly improve symptoms and mor- start antiarrhythmic QT-prolonging tion and result in torsades de pointes. bidity relative to other treatment op- medications and supplement them as An example of a mixed interaction is tions. In this regard, medications that needed, especially in patients taking that of erythromycin and cisapride. Not commonly cause significant QT prolon- other medications that can cause hy- only do both drugs block potassium gation must be used only if no other pokalemia or hypomagnesemia. currents, but erythromycin also inhib- medications have a comparable benefi- Measurement of the baseline QT in- its cytochrome P-4503A4. cial effect in treating the same condi- terval may also be of critical impor- tion, if they are known to be or are po- tance when assessing the risk of tors- ASSESSING THE BALANCE tentially superior to other available ades de pointes in a particular patient. OF RISK AND BENEFIT OF medications, or if other medications However, with many drugs that can QT INTERVAL–PROLONGING carry other more significant risks. cause QT interval prolongation, the risk MEDICATIONS The benefit of antiarrhythmic therapy, of torsades de pointes is so low that the When drugs that can prolong the QT infor example, is clearest when it results majority of experts do not consider mea- terval are used, physicians should en- in the immediate termination of a sus- surement of the QT interval to be cost- sure that the potential benefits are clini- tained ventricular arrhythmia. When an- effective. For some of these drugs, the cally important and the risks are tiarrhythmic therapy is used for pa- QT interval must be measured in thou-

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

sands of patients to identify 1 person at have improbable or very improbable po- edge of these relationships should en- risk of significant QT prolongation. The tential for causing QT prolongation, able an astute clinician to make a semi- cost of this practice, in our opinion, out- checking an ECG before and after start- quantitative decision about the use and weighs the benefits. In a survey com- ing the drug, especially in low-risk pa- dosing of dofetilide by weighing the pleted by LQTS experts, the majority tients, may not be necessary (Box). risks and the proposed benefits of pre- would always check an ECG before and Despite these recommendations, un- venting recurrent atrial fibrillation or after starting an antiarrhythmic medi- certainty remains regarding the spe- flutter. Likewise, detailed dosing and cation, one third to half would always cific relationship between the degree of monitoring recommendations are part check an ECG before and after starting QT prolongation and the risk of life- of the product labeling for the 2 most an antipsychotic drug, and less than one threatening arrhythmias with each in- recently approved antiarrhythmic third would always check an ECG be- dividual drug. A QT interval of at least agents, dofetilide and sotalol. The prod- fore and after starting an anti-infective 500 ms generally has been shown to uct labeling for medications that could or antidepressant (Box). Based on the re- correlate with a higher risk of torsades prolong the QT interval contains warn- sults of this survey, we propose the fol- de pointes, but there is no established ings regarding their effect on the QT in- lowing recommendations on the proper threshold below which prolongation of terval that clinicians who prescribe monitoring of the QT interval in pa- the QT interval is considered free of these drugs should be aware of. tients receiving QT-prolonging medi- proarrhythmic risk.9,49 Thus, physi- Some antipsychotic drugs have been cations. First, an ECG should be rou- cians are left with great uncertainty re- shown to prolong the QT interval.51,52 tinely checked before and after starting garding when to stop a QT-prolong- A recent clinical epidemiological study an antiarrhythmic agent that can pro- ing medication. Respondents to the has demonstrated a direct relation- long the QT interval (Table 1). If the pa- survey on LQTS were more likely to ship between the dose of the older an- tient has a prolonged QTc at baseline stop a QT-prolonging medication for a tipsychotic drugs and the risk of sud- (Ͼ450 ms in men and Ͼ460 ms in QT of 520 ms than for a QT of 500 ms. den death, but the QT interval was not women in the absence of interventricu- However, it should be emphasized that measured in this study to examine its lar conduction defects), it is important there is no clear-cut consensus on the correlation with sudden death.53 More to try to avoid all QT-prolonging medi- degree of drug-induced QT prolonga- recently, a population-based study con- cations.47 Although this is not an abso- tion that should require drug discon- firmed that antipsychotic drugs known lute contraindication to starting QT- tinuation. to produce greater QT prolongation prolonging drugs, expert opinion should Despite the clinical importance of the than other antipsychotic drugs were as- be sought before such drugs are started. QT interval, definitive information on sociated with a higher risk of cardio- If the patient is already taking an anti- the clinical epidemiology of the QT in- vascular death.54 Of note, no consen- arrhythmic agent with this potential and terval and its prolongation by medica- sus exists about the relative likelihood another drug needs to be added, it is im- tions is surprisingly lacking. Few stud- of torsades de pointes among patients portant to know whether the new drug ies have evaluated the relationship treated with different antipsychotic may also prolong the QT interval (Box) between the QT interval and patient drugs, nor is there a prospective study or can interact with the antiarrhythmic outcomes. Even for commonly used an- delineating the absolute risk of arrhyth- medication (Table). If one of these events tiarrhythmic drugs like sotalol and mia. This paucity of data leaves the is a possibility, an alternative agent amiodarone, this relationship has not practitioner in a lurch concerning thera- should be considered. If no alternative been adequately explored. The newly peutics: the need for antipsychotic is available and the drug being added is marketed compound dofetilide is the therapy cannot be ignored, and the pa- necessary, an ECG should be per- only medication for which such a re- tients with the most severe psychoses formed to monitor the QT interval be- lationship has been extensively ex- may need the highest doses of drugs. fore and after starting the new drug. plored and reported.20,50 It has been These same patients are also at higher Indeed, concurrent prescribing of demonstrated that a linear relation- risk of nonadherence to prescribed QT-prolonging drugs is common in the ship exists between the plasma con- therapies (including omitting and tak- outpatient setting; however, the clini- centration of dofetilide and the mean ing excessive amounts of needed thera- cal consequences of this practice are not change from baseline QTc. It has also pies) and to the development of other known.48 been shown that a relationship exists conditions, such as electrolyte deple- Second, an ECG should be checked between the dose and concentration of tion or sympathetic overactivity, that before and after starting a drug if the dofetilide and its efficacy as well as the may predispose to torsades de pointes. drug is one that has been deemed by the risk of torsades de pointes associated Furthermore, no data exist on how LQTS experts to have very probable, with its use.20 A direct correlation be- and when to monitor the QT interval probable, or possible potential for caus- tween rate of torsades de pointes and when various antipsychotic drugs are ing QT prolongation (Box). If the drug increase from baseline QTc has also used. Until more data are obtained, the was deemed by the LQTS experts to been proven for dofetilide.20 Knowl- Box should provide some guidance in

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

this regard. Of note, because of the con- standardization of preclinical and clini- cardiac safety of new drugs. In Novem- cern raised about risk of QT prolonga- cal cardiovascular drug testing.57 For ex- ber 2001, the FDA announced that as of tion with ziprasidone, a trial has been ample, it was not specified whether drug the fall of 2002, sponsors must submit launched that will randomly assign testing should explore the effect of the ECG raw data in digital format with an- more than 15000 patients with schizo- drug on the QTc or the uncorrected QT notations to enable the FDA to inde- phrenia to ziprasidone and olanza- interval. The label for dofetilide relies on pendently assess the cardiac safety pine. This trial’s primary end point is the QTc (using the Bazett formula), profile of new drugs.58 The FDA is also all-cause mortality (Brian Strom, MD, whereas the label for sotalol uses the un- working with Health Canada to de- unpublished data, 2002). corrected QT interval. velop guidance on the assessment of QT Quinolone antibiotics also pose a dif- In February 2002, the ICHS7B guid- prolongation during clinical trials.47 The ficult dilemma. They are used on a wide ance was proposed and, although it has final guidance is still pending. Through scale for common infections without not been finalized, provides more spe- these initiatives, regulators are hoping checking the ECG, an approach gener- cific direction for cardiac safety test- to standardize preclinical and clinical ally agreed on by the aforementioned ing of new drugs.56 The ICHS7B speci- cardiovascular testing of all drugs, an im- LQTS experts (Box). Yet, sporadic epi- fies that new drugs should be tested in portant endeavor in enhancing efforts at sodes of torsades de pointes have been 3 preclinical assays: the human ether- risk management of QT-prolonging reported in association with quino- a-go-go–related gene channel assay to medications. Equally important, how- lones,55 and there is evidence that qui- check for blockade of the IKr channel, ever, is the clear dissemination of these nolones are commonly coprescribed the action potential duration (APD) as- guidelines and clear labeling of drugs with other QT-prolonging drugs.48 Be- say that uses canine Purkinje fibers to with QT-prolonging potential, specifi- cause of the immense uncertainly sur- check for significant APD prolonga- cally as they relate to interactions that rounding the effects of concurrent use tion, and an in vivo rodent ECG with could augment prolongation. of QT-prolonging drugs, no preventive a detailed description of test methods. In this article, we present an update approach is currently recommended The guidance also suggests that if the of the current knowledge of the QT in- other than trying to avoid quinolones in tested drug is shown in preclinical as- terval and proposed ways to enhance patients taking other QT-prolonging says to cause some blockade of the IKr risk management of QT-prolonging drugs or with other risk factors. channel or prolongation of the APD, its medications. As more knowledge about potential clinical risks should be evalu- this important topic is gained, it is criti- RISK MANAGEMENT OF ated in carefully designed clinical trials. cal that this knowledge be dissemi- QT INTERVAL–PROLONGING Those studies must have adequate nated in a timely fashion and in a style MEDICATIONS sample sizes and should ensure fre- that is easily comprehended by clini- We have thus far discussed the role of quent recording of ECGs. cians. Perhaps the most surprising find- health care practitioners in manage- Although the preliminary ICHS7B ing of our review is the relative pau- ment of the risk of QT-prolonging guidance lacks data on the standards of city of information that can help medications in treating individual pa- clinical evaluation, it is hoped that the clinicians and patients make informed tients. The scope of risk management, final version will provide specific infor- decisions about drugs that can pro- however, extends beyond health care mation on the clinical evaluation stan- long the QT interval. practitioners and includes regulators, dards for compounds with and with- pharmaceutical companies, and inves- out hazardous QT signals in preclinical Author Contributions: Study concept and design: tigators. For example, regulators came testing. It is well known that many com- Al-Khatib, Allen LaPointe, Kramer, Califf. Acquisition of data: Al-Khatib, Allen LaPointe. to realize the potential of noncardio- panies are screening compounds and Analysis and interpretation of data: Al-Khatib, Allen vascular drugs to prolong the QT discontinuing those in which a flag is LaPointe, Kramer. Drafting of the manuscript and obtaining funding: interval and potentially result in life- raised at the preclinical level. The po- Al-Khatib, Califf. threatening arrhythmias. Thus, regu- tential advantage of this approach is ob- Critical revision of the manuscript for important in- tellectual content: Allen LaPointe, Kramer, Califf. latory guidances on drug develop- vious: it prompts stopping the drug’s de- Administrative, technical, or material support: ment advise that all new drugs be velopment before the complex clinical Al-Khatib, Allen LaPointe, Califf. Study supervision: Kramer. evaluated for possible effects on car- manifestations become an issue. How- Funding/Support: This study was supported in part 56 diac repolarization. ever, the list of QT-prolonging drugs in- by grant U18HS10548 from the Agency for Health- Guidance in that regard had until re- cludes several that provide substantial care Research and Quality, Rockville, Md. Acknowledgment: The following persons partici- cently been sketchy. The US Food and health benefits, and it would be unfor- pated in a survey and convened in August 2000 to Drug Administration’s (FDA’s) Interna- tunate if drugs with the potential for a discuss the current knowledge of LQTS. Academic participants: Mark E. Anderson, John T. Bigger, Jr, Rob- tional Conference on Harmonization highly positive overall impact were ert M. Califf, Robert R. Fenichel, David A. Flockhart, (ICH) S7A guidance included only a dropped early in development. Augustus O. Grant, Ralph Lazzara, Kerry L. Lee, Marek Malik, Joel Morganroth, Arthur J. Moss, Milton Packer, general mention of cardiovascular test- The ICHS7B is one of many initia- Ileana L. Pina, Eric N. Prystowsky, Dan M. Roden, Jer- ing of new drugs, which led to a lack of tives recently developed to improve the emy N. Ruskin, Michael Sanguinetti, Gordon F. To-

Downloaded From: https://jamanetwork.com/ on 10/02/2021 THE QT INTERVAL

maselli, Raymond L. Woosley; participants from the 17. Aytemir K, Maarouf N, Gallagher MM, Yap YG, of antiarrhythmic therapy for supraventricular tachy- FDA: Stephen Fredd, Maryann Gordon, Abraham Waktare JE, Malik M. Comparison of formulae for heart cardias. Am J Cardiol. 1997;80:1551-1557. Karkowsky, John Koerner, Raymond J. Lipicky, An- rate correction of QT interval in exercise electrocar- 39. Clinical Pharmacology [database online]. Tampa, thony Proakis, Shari Targum, Robert J. Temple; par- diograms. Pacing Clin Electrophysiol. 1999;22:1397- Fla: Gold Standard Multimedia Inc; 2002. Available ticipants from European regulatory agencies: Ursula 1401. at: http://cp.gsm.com. Accessed September 17, 2002. Hoppe, Rashmi Shah; participants from industry: Leif 18. Malik M. Problems of heart rate correction in as- 40. USPDI advice for the patient. MICROMEDEX [da- Carlsson, Borje Darpo, Paul R. Eisenberg, Karen Frank, sessment of drug-induced QT interval prolongation. tabase online]. Stamford, Conn: Thomson Corp. Alan S. Hollister, J. Heyward Hull, Lewis Kinter, John J Cardiovasc Electrophysiol. 2001;12:411-420. Healthcare Series Vol 114. Expiration: December 2002. H. Lawrence, Leslie Lipka, Eric Michelson, Steven J. 19. Milne JR, Ward DE, Spurrell RA, Camm AJ. The ven- Available at: http://www.micromedex.com. Ac- Miller, Andrew Nicholls, Gunnar Olsson, Ihor W. Rak, tricular paced QT interval: the effects of rate and exer- cessed April 2, 2003. Stots B. Relle, David T. Roberts, Pierre A. Wicker. cise. Pacing Clin Electrophysiol. 1982;5:352-358. 41. Tatro DS, ed. Drug Interaction Facts. St Louis, Mo: 20. Dofetilide product monograph and confirmation Facts and Comparisons; 2002. of education. New York, NY: Pfizer Inc; March 2000. 42. Hansten PD, Horn JR. Hansten and Horn’s Drug REFERENCES 21. Chiang CE, Roden DM. The long QT syn- Interactions Analysis and Management. St Louis, Mo: 1. MedWatch. Glaxo Wellcome voluntarily with- dromes: genetic basis and clinical implications. JAm Facts and Comparisons; 2002. draws Raxar (grepafloxacin) [press release]. Rock- Coll Cardiol. 2000;36:1-12. 43. Pacifico A, Hohnloser SH, Williams JH, et al. Pre- ville, Md: US Food and Drug Administration; Octo- 22. Moss AJ, Zareba W, Benhorin J, et al. ECG T- vention of implantable-defibrillator shocks by treat- ber 1999. Available at: http://www.fda.gov/medwatch wave patterns in genetically distinct forms of the he- ment with sotalol. N Engl J Med. 1999;340:1855- /safety/1999/raxar.html. Accessed October 22, 2002. reditary long QT syndrome. Circulation. 1995;92: 1862. 2. FDA Talk Paper. Janssen Pharmaceutica Stops Mar- 2929-2934. 44. Morganroth J, Goin J. Quinidine-related mortal- keting Cisapride in the US. Rockville, Md: US Food 23. Wilde AA, Jongbloed RJ, Doevendans PA, et al. ity in the short-to-medium-term treatment of ven- and Drug Administration; March 2000. Available at: Auditory stimuli as a trigger for arrhythmic events dif- tricular arrhythmias: a meta-analysis. Circulation. 1991; http://www.fda.gov/bbs/topics/answers/ans01007 ferentiate HERG-related (LQTS2) patients from KV- 84:1977-1983. .html. Accessed October 22, 2002. LQT1-related patients (LQTS1). J Am Coll Cardiol. 45. Coplen SE, Antman EM, Berlin JA, Hewitt P, 3. Smalley W, Shatin D, Wysowski DK, et al. Con- 1999;33:327-332. Chalmers TC. Efficacy and safety of quinidine therapy traindicated use of cisapride: impact of Food and Drug 24. Schwartz PJ, Priori SG, Locati EH, et al. Long QT for maintenance of sinus rhythm after cardioversion: Administration regulatory action. JAMA. 2000;284: syndrome patients with mutations of the SCN5A and a meta-analysis of randomized control trials. Circu- 3036-3039. HERG genes have differential responses to Na+ chan- lation. 1990;82:1106-1116. 4. Moss AJ, Zareba W, Kaufman ES, et al. Increased nel blockade and to increases in heart rate: implica- 46. Waldo AL, Camm AJ, de Ruyter H, et al. Effect risk of arrhythmic events in long QT syndrome with tions for gene-specific therapy. Circulation. 1995;92: of d-sotalol on mortality in patients with left ventricu- mutations in the pore region of the human ether-a- 3381-3386. lar dysfunction after recent and remote myocardial in- go-go-related gene potassium channel. Circulation. 25. Zhang L, Timothy KW, Vincent GM, et al. Spec- farction. Lancet. 1996;348:7-12. 2002;105:794-799. trum of ST-T-wave patterns and repolarization param- 47. The Clinical Evaluation of QT/QTc Interval Pro- 5. Locati EH, Zareba W, Moss AJ, et al. Age- and sex- eters in congenital long-QT syndrome: ECG findings longation and Proarrhythmic Potential for Non- related differences in clinical manifestations in pa- identify genotypes. Circulation. 2000;102:2849- antiarrhythmic Drugs. Available at: http:// tients with congenital log QT syndrome: findings from 2855. www.fda.gov/cder/calendar/meeting/qt4jam.pdf. the international LQTS Registry. Circulation. 1998; 26. Donger C, Denjoy I, Berthet M, et al. KVLQT1 C- Accessed March 25, 2003. 97:2237-2244. terminal missense mutation causes a forme fruste 48. Curtis LH, Østbye T, Sendersky V, et al. Prescrip- 6. Zareba W, Moss AJ, Schwartz PJ, et al. Influence long-QT syndrome. Circulation. 1997;96:2778-2781. tion of QT-prolonging drugs in a large, national out- of the genotype on the clinical course of the long-QT 27. Reardon M, Malik M. QT interval change with patient population. Am J Med. 2003;114:135-141. syndrome. N Engl J Med. 1998;339:960-965. age in an overtly healthy older population. Clin Car- 49. Bednar MM, Harrigan EP, Anziano RJ, Camm AJ, 7. Schwartz PJ, Priori SG, Spazzolini C, et al. Genotype- diol. 1996;19:949-952. Ruskin JN. The QT interval. Prog Cardiovasc Dis. 2001; phenotype correlation in the long-QT syndrome: gene- 28. Ahnve S. QT interval prolongation in acute myo- 43:1-45. specific triggers for life-threatening arrhythmias. Cir- cardial infarction. Eur Heart J. 1985;6(suppl D):85-95. 50. Demolis JL, Funck-Brentano C, Ropers J, Ghad- culation. 2001;103:89-95. 29. Rebeiz AG, Al-Khatib SM. A case of severe ische- anfar M, Nichols DJ, Jaillon P. Influence of dofetilide 8. Priori SG, Napolitano C, Schwartz PJ. Low pen- mia-induced QT prolongation. Clin Cardiol. 2001; on QT-interval duration and dispersion at various heart etrance in the long-QT syndrome: clinical impact. Cir- 24:750. rates during exercise in humans. Circulation. 1996; culation. 1999;99:529-533. 30. Khan IA. Clinical and therapeutic aspects of con- 94:1592-1599. 9. Moss AJ, Schwartz PJ, Crampton RS, et al. The long genital and acquired long QT syndrome. Am J Med. 51. Reilly JG, Ayis SA, Ferrier IN, Jones SJ, Thomas QT syndrome: prospective longitudinal study of 328 2002;112:58-66. SH. QTc interval abnormalities and psychotropic drug families. Circulation. 1991;84:1136-1144. 31. Kay GN, Plumb VJ, Arciniegas JG, Henthorn RW, therapy in psychiatric patients. Lancet. 2000;355: 10. Viskin S. Long QT syndromes and torsade de Waldo AL. Torsade de pointes: the long-short initiat- 1048-1052. pointes. Lancet. 1999;354:1625-1633. ing sequence and other clinical features: observations 52. Glassman AH, Bigger JT. Antipsychotic drugs: pro- 11. Morganroth J, Brozovich FV, McDonald JT, Ja- in 32 patients. J Am Coll Cardiol. 1983;2:806-817. longed QTc interval, torsade de pointes, and sudden cobs RA. Variability of the QT measurement in healthy 32. Roden DM. Risks and benefits of antiarrhythmic death. Am J Psychiatry. 2001;158:1774-1782. men, with implications for selection of an abnormal therapy. N Engl J Med. 1994;331:785-791. 53. Ray WA, Meredith S, Thapa PB, Meador KG, Hall QT value to predict drug toxicity and proarrhythmia. 33. Roden DM. Mechanisms and management of pro- K, Murray KT. Antipsychotics and the risk of sudden Am J Cardiol. 1991;67:774-776. arrhythmia. Am J Cardiol. 1998;82:49I-57I. cardiac death. Arch Gen Psychiatry. 2001;58:1161- 12. Molnar J, Zhang F, Weiss J, Ehlert FA, Rosenthal 34. Makkar RR, Fromm BS, Steinmen RT, Meissner 1167. JE. Diurnal pattern of QTc interval: how long is pro- MD, Lehmann MH. Female gender as a risk factor for 54. Hennessy S, Bilker WB, Knauss JS, et al. Cardiac longed? possible relation to circadian triggers of car- torsades de pointes associated with cardiovascular arrest and ventricular arrhythmias in patients taking diovascular events. J Am Coll Cardiol. 1996;27:76- drugs. JAMA. 1993;270:2590-2597. antipsychotic drugs: cohort study using administra- 83. 35. Maisel WH, Kuntz KM, Reimold SC, et al. Risk of tive data. BMJ. 2002;325:1070-1074. 13. Anderson ME, Al-Khatib SM, Roden DM, Califf initiating antiarrhythmic drug therapy for atrial fibril- 55. Anderson ME, Mazur A, Yang T, Roden DM. Po- RM. Cardiac repolarization: current knowledge, criti- lation in patients admitted to a university hospital. Ann tassium current antagonist properties and proarrhyth- cal gaps, and new approaches to drug development Intern Med. 1997;127:281-284. mic consequences of quinolone antibiotics. J Pharma- and patient management. Am Heart J. 2002;144:769- 36. Chung MK, Schweikert RA, Wilkoff BL, et al. Is col Exp Ther. 2001;296:806-810. 781. hospital admission for initiation of antiarrhythmic 56. ICHS7B Safety Pharmacology Studies for Assess- 14. Allen LaPointe N, Al-Khatib SM, Kramer JM, Battle therapy with sotalol for atrial arrhythmias required? ing the Potential for Delayed Ventricular Repolariza- J, Califf RM. Deficits in knowledge related to the QT yield of in-hospital monitoring and prediction of risk tion (QT Interval Prolongation) by Human Pharma- interval that could impact patient safety [abstract]. for significant arrhythmia complications. J Am Coll Car- ceuticals. Washington, DC: US Food and Drug J Am Coll Cardiol. 2002;39:125A. diol. 1998;32:169-176. Administration; 2002. 15. Funck-Brentano C, Jaillon P. Rate-corrected QT 37. Torp-Pedersen C, Moller M, Bloch-Thomsen PE, 57. ICHS7A Safety Pharmacology Studies for Hu- interval: techniques and limitations. Am J Cardiol. 1993; et al. Dofetilide in patients with congestive heart fail- man Pharmaceuticals. Washington, DC: US Food and 72:17B-22B. ure and left ventricular dysfunction. N Engl J Med. Drug Administration; 2001. 16. Hnatkova K, Malik M. “Optimum” formulae for 1999;341:857-865. 58. Center for Drug Evaluation and Research home heart rate correction of the QT interval. Pacing Clin 38. Simons GR, Eisenstein EL, Shaw LJ, Mark DB, page. Available at: http://www.fda.gov/cder. Ac- Electrophysiol. 1999;22:1683-1687. Pritchett EL. Cost effectiveness of inpatient initiation cessed March 25, 2003.

Downloaded From: https://jamanetwork.com/ on 10/02/2021