Drugs and the QT Interval — Caveat Doctor
http://www.100md.com
《新英格兰医药杂志》
All drugs have the potential to cause adverse effects. Occasionally, serious adverse effects are not identified in preclinical studies and become apparent only when a drug is in widespread clinical use — a particular problem with adverse events that are rare and result in few symptoms. One example of relevance to clinicians and regulatory agencies is drug-induced prolongation of the QT interval. During the past few years, many drugs have been removed from the market or required to include "black box" warnings on their labels because of the potential for QT-interval prolongation. The list of causative medications is daunting, and the number and variety of interactions involving these drugs can seem overwhelming.
After adjustment for the heart rate, the corrected QT interval is considered to be prolonged if it is more than 450 msec in men or more than 460 msec in women (see Figure). This prolongation most often results from delayed ventricular repolarization, a process that is mediated by the efflux of intracellular potassium. The channels responsible for this current are susceptible to blockade by many drugs, producing a suitable environment for the development of torsades de pointes, a polymorphic form of ventricular tachycardia that is characterized by the twisting of the QRS axis around the baseline. Although it is often self-terminating, torsades de pointes can sometimes lead to sudden death.
Figure. Measurement of the QT Interval and Adjustment for Heart Rate.
In adults with a normal QRS duration, the corrected QT interval (QTc), after adjustment for the heart rate, is considered to be prolonged if it is more than 450 msec in men or more than 460 msec in women.
Non–drug-related factors associated with prolongation of the QT interval include female sex, advanced age, electrolyte disturbances such as hypokalemia and hypomagnesemia (which may also be medication-related), heart failure, bradycardia, ischemia, and the congenital long-QT syndromes.1 Although many drugs have been associated with prolongation of the QT interval,2 some are particularly important either because they are used by large numbers of patients or because common interactions between them and other drugs can increase the risk of torsades de pointes.
Antiarrhythmic agents, particularly the class IA agents quinidine, disopyramide, and procainamide, are well-known causes of QT-interval prolongation. Among class III agents, the risk of torsades de pointes is thought to be higher with sotalol than with amiodarone. Drug-induced torsades de pointes is not solely the concern of the cardiologist, however. Several commonly used classes of noncardiac drugs can directly prolong the QT interval, including antidepressants (particularly tricyclics), neuroleptics (such as haloperidol and thioridazine), some antibiotics (macrolides and quinolones), prokinetic agents, tamoxifen, and antifungal agents (a more comprehensive list can be found at www.torsades.org). Although any of these drugs can provoke torsades de pointes in a susceptible patient, the risk of this condition in patients with no other risk factors is almost certainly minimal. It is always reasonable, however, to ask whether the drug is really necessary or whether it can be replaced by a less arrhythmogenic alternative.
Unfortunately, even in patients who are otherwise healthy and free of other apparent predisposing factors, drug interactions can lead to life-threatening torsades de pointes. These interactions are of two sorts. The more straightforward type involves the use of two or more drugs, each with prolonging effects on the QT interval. This additive or synergistic effect on repolarization is a pharmacodynamic drug interaction. One example would be the simultaneous use of haloperidol and amitriptyline in a patient with mental illness. In this instance, the likelihood of QT-interval prolongation and torsades de pointes is at least partly a function of the dose of the drugs involved; arrhythmias are more likely when higher doses are prescribed. In general, the combined use of drugs with direct effects on repolarization should be avoided when possible.
The second type of drug interaction involves the simultaneous prescribing of a medication that prolongs the QT interval with another drug that increases the plasma concentration of the first drug, generally by interfering with its hepatic metabolism. Such pharmacokinetic drug interactions are particularly important when the QT-interval–prolonging drug has no other route of biotransformation. An example would be the simultaneous use of cisapride and verapamil. Cisapride is metabolized primarily by the cytochrome P-450 isoenzyme 3A4 (CYP3A4). Many commonly used drugs, including verapamil, block this enzyme and may result in the accumulation of cisapride, thereby enhancing its effects on repolarization and increasing the risk of torsades de pointes. Indeed, many of the drugs that are known to block CYP3A4 also have direct effects on repolarization — a pharmacologic double whammy that may cause a dramatic lengthening of the QT interval.
The substrates and inhibitors of the various cytochrome P-450 isoenzymes are numerous.3 For practical purposes, however, attention should be focused on five major classes of drugs: antifungal agents, antiretroviral drugs, calcium-channel blockers, selective serotonin-reuptake–inhibitor antidepressants, and antibiotics (particularly quinolones and macrolides other than azithromycin). Along with amiodarone, drugs in these classes cause most of the clinically significant cytochrome P-450 enzyme inhibition in clinical practice. Consequently, when prescribing these medications, physicians should always entertain the possibility of a drug interaction. The cytochrome P-450 enzymes inhibited by these classes of drugs and some of the QT-interval–prolonging drugs that are subsequently prone to accumulation are listed in the Table.
Table. Some Common Drugs That in Combination May Prolong the QT Interval.
From the clinical standpoint, what are the practical implications of drug-induced QT-interval prolongation? In the absence of other predisposing factors, the absolute risk of drug-induced torsades de pointes is probably extremely low when a single QT-interval–prolonging drug is prescribed in therapeutic doses, as evidenced by the millions of courses of erythromycin that have been taken safely during the past 30 years. However, for patients with other preexisting drug-related or non–drug-related risk factors, QT-interval–prolonging drugs should be used very cautiously and only after the risks and benefits have been weighed on a case-by-case basis.
Unfortunately, estimating the risk side of this equation is not always straightforward. Most of what is known about drug-induced QT-interval prolongation derives from spontaneous reporting mechanisms. Though instructive, these reports tell us little about what happens in the real world. How many people who are exposed to these drugs have no sequelae? How many avoidable sudden deaths might actually be misattributed to coronary disease? How do multiple risk factors act together in an individual patient? As illustrated by the study reported by Ray and colleagues in this issue of the Journal (pages 1089–1096), the discipline of pharmacoepidemiology has much to teach us about the consequences of drug interactions in clinical practice. As more of this information comes to light, a few short lists and an element of caution with the prescription pad will probably go a long way toward preventing torsades de pointes and other serious adverse effects of drug interactions.
Source Information
From the Department of Medicine, University of Toronto (B.A.L., D.N.J.); and the Divisions of Geriatric Medicine (B.A.L.), General Internal Medicine (D.N.J.), and Clinical Pharmacology (B.A.L., D.N.J.), Sunnybrook and Women's College Health Sciences Centre — both in Toronto.
References
Al-Khatib SM, LaPointe NMA, Kramer JM, Califf RM. What clinicians should know about the QT interval. JAMA 2003;289:2120-2127.
Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med 2004;350:1013-1022.
Dresser GK, Bailey DG. A basic conceptual and practical overview of interactions with highly prescribed drugs. Can J Clin Pharmacol 2002;9:191-198.
Related Letters:
Oral Erythromycin and the Risk of Sudden Death
Kaplan E. L., Winston A. P., Schoenholtz J. C., Harris J.-D., de Leon-Casasola O. A., Amory J. K., Amory D. W. Sr., Ray W. A., Murray K. T., Stein C. M., Liu B. A., Juurlink D. N.(Barbara A. Liu, M.D., and)
After adjustment for the heart rate, the corrected QT interval is considered to be prolonged if it is more than 450 msec in men or more than 460 msec in women (see Figure). This prolongation most often results from delayed ventricular repolarization, a process that is mediated by the efflux of intracellular potassium. The channels responsible for this current are susceptible to blockade by many drugs, producing a suitable environment for the development of torsades de pointes, a polymorphic form of ventricular tachycardia that is characterized by the twisting of the QRS axis around the baseline. Although it is often self-terminating, torsades de pointes can sometimes lead to sudden death.
Figure. Measurement of the QT Interval and Adjustment for Heart Rate.
In adults with a normal QRS duration, the corrected QT interval (QTc), after adjustment for the heart rate, is considered to be prolonged if it is more than 450 msec in men or more than 460 msec in women.
Non–drug-related factors associated with prolongation of the QT interval include female sex, advanced age, electrolyte disturbances such as hypokalemia and hypomagnesemia (which may also be medication-related), heart failure, bradycardia, ischemia, and the congenital long-QT syndromes.1 Although many drugs have been associated with prolongation of the QT interval,2 some are particularly important either because they are used by large numbers of patients or because common interactions between them and other drugs can increase the risk of torsades de pointes.
Antiarrhythmic agents, particularly the class IA agents quinidine, disopyramide, and procainamide, are well-known causes of QT-interval prolongation. Among class III agents, the risk of torsades de pointes is thought to be higher with sotalol than with amiodarone. Drug-induced torsades de pointes is not solely the concern of the cardiologist, however. Several commonly used classes of noncardiac drugs can directly prolong the QT interval, including antidepressants (particularly tricyclics), neuroleptics (such as haloperidol and thioridazine), some antibiotics (macrolides and quinolones), prokinetic agents, tamoxifen, and antifungal agents (a more comprehensive list can be found at www.torsades.org). Although any of these drugs can provoke torsades de pointes in a susceptible patient, the risk of this condition in patients with no other risk factors is almost certainly minimal. It is always reasonable, however, to ask whether the drug is really necessary or whether it can be replaced by a less arrhythmogenic alternative.
Unfortunately, even in patients who are otherwise healthy and free of other apparent predisposing factors, drug interactions can lead to life-threatening torsades de pointes. These interactions are of two sorts. The more straightforward type involves the use of two or more drugs, each with prolonging effects on the QT interval. This additive or synergistic effect on repolarization is a pharmacodynamic drug interaction. One example would be the simultaneous use of haloperidol and amitriptyline in a patient with mental illness. In this instance, the likelihood of QT-interval prolongation and torsades de pointes is at least partly a function of the dose of the drugs involved; arrhythmias are more likely when higher doses are prescribed. In general, the combined use of drugs with direct effects on repolarization should be avoided when possible.
The second type of drug interaction involves the simultaneous prescribing of a medication that prolongs the QT interval with another drug that increases the plasma concentration of the first drug, generally by interfering with its hepatic metabolism. Such pharmacokinetic drug interactions are particularly important when the QT-interval–prolonging drug has no other route of biotransformation. An example would be the simultaneous use of cisapride and verapamil. Cisapride is metabolized primarily by the cytochrome P-450 isoenzyme 3A4 (CYP3A4). Many commonly used drugs, including verapamil, block this enzyme and may result in the accumulation of cisapride, thereby enhancing its effects on repolarization and increasing the risk of torsades de pointes. Indeed, many of the drugs that are known to block CYP3A4 also have direct effects on repolarization — a pharmacologic double whammy that may cause a dramatic lengthening of the QT interval.
The substrates and inhibitors of the various cytochrome P-450 isoenzymes are numerous.3 For practical purposes, however, attention should be focused on five major classes of drugs: antifungal agents, antiretroviral drugs, calcium-channel blockers, selective serotonin-reuptake–inhibitor antidepressants, and antibiotics (particularly quinolones and macrolides other than azithromycin). Along with amiodarone, drugs in these classes cause most of the clinically significant cytochrome P-450 enzyme inhibition in clinical practice. Consequently, when prescribing these medications, physicians should always entertain the possibility of a drug interaction. The cytochrome P-450 enzymes inhibited by these classes of drugs and some of the QT-interval–prolonging drugs that are subsequently prone to accumulation are listed in the Table.
Table. Some Common Drugs That in Combination May Prolong the QT Interval.
From the clinical standpoint, what are the practical implications of drug-induced QT-interval prolongation? In the absence of other predisposing factors, the absolute risk of drug-induced torsades de pointes is probably extremely low when a single QT-interval–prolonging drug is prescribed in therapeutic doses, as evidenced by the millions of courses of erythromycin that have been taken safely during the past 30 years. However, for patients with other preexisting drug-related or non–drug-related risk factors, QT-interval–prolonging drugs should be used very cautiously and only after the risks and benefits have been weighed on a case-by-case basis.
Unfortunately, estimating the risk side of this equation is not always straightforward. Most of what is known about drug-induced QT-interval prolongation derives from spontaneous reporting mechanisms. Though instructive, these reports tell us little about what happens in the real world. How many people who are exposed to these drugs have no sequelae? How many avoidable sudden deaths might actually be misattributed to coronary disease? How do multiple risk factors act together in an individual patient? As illustrated by the study reported by Ray and colleagues in this issue of the Journal (pages 1089–1096), the discipline of pharmacoepidemiology has much to teach us about the consequences of drug interactions in clinical practice. As more of this information comes to light, a few short lists and an element of caution with the prescription pad will probably go a long way toward preventing torsades de pointes and other serious adverse effects of drug interactions.
Source Information
From the Department of Medicine, University of Toronto (B.A.L., D.N.J.); and the Divisions of Geriatric Medicine (B.A.L.), General Internal Medicine (D.N.J.), and Clinical Pharmacology (B.A.L., D.N.J.), Sunnybrook and Women's College Health Sciences Centre — both in Toronto.
References
Al-Khatib SM, LaPointe NMA, Kramer JM, Califf RM. What clinicians should know about the QT interval. JAMA 2003;289:2120-2127.
Roden DM. Drug-induced prolongation of the QT interval. N Engl J Med 2004;350:1013-1022.
Dresser GK, Bailey DG. A basic conceptual and practical overview of interactions with highly prescribed drugs. Can J Clin Pharmacol 2002;9:191-198.
Related Letters:
Oral Erythromycin and the Risk of Sudden Death
Kaplan E. L., Winston A. P., Schoenholtz J. C., Harris J.-D., de Leon-Casasola O. A., Amory J. K., Amory D. W. Sr., Ray W. A., Murray K. T., Stein C. M., Liu B. A., Juurlink D. N.(Barbara A. Liu, M.D., and)