Making QT Correction Simple is Complicated NOEL G. BOYLE, M.D., PH.D., and JAMES N. WEISS, M.D. From the Division of Cardiology, UCLA School of Medicine, Los Angeles, California Editorial Comment The QT interval and its correction formulas gained clin- ical importance with the description of the long QT syn- drome and its association with sudden cardiac death in 1957 by Jervell and Lange-Nielsen. 1 In 1978, Schwartz and Wolf 2 rst reported an increase in the incidence of sudden death by 2.2-fold in postmyocardial infarction patients when the QTc was prolonged to .0.44 seconds. With the association of QT interval prolongation with torsades de pointes and sudden cardiac death in many clinical situa- tions, measurement of the QT interval has become an im- portant tool for assessing the risk of life-threatening arrhyth- mias. QT prolongation is found in diverse clinical situations, from the congenital long QT syndromes to isch- emia, cardiomyopathy, hypokalemia, hypocalcemia, hypo- thermia, complete heart block, and as a drug effect. Most recently, QT prolongation has been reported as a factor in sudden infant death syndrome. 3 Many different drugs (mostly with potassium channel-blocking effect), including Class I and Class III antiarrhythmic drugs, prolong the QT interval and may induce torsades de pointes (www.torsades. org) . Although the amount of QT prolongation in itself is not directly predictive of proarrhythmia, it has been shown that the associated early afterdepolarizations are the mech- anisms of drug-induced torsades de pointes. 4 In addition, QT prolongation in itself may not be sufcient to cause arrhythmias and may require other factors, such as an in- crease in dispersion of repolarization and refractoriness. 5 However, in the area of drug testing and development, it is not clinically possible to study the incidence of torsades de pointes as a function of drug levels; hence, QT interval prolongation has become accepted as a surrogate marker for torsades de pointes risk. 6 The QT interval represents a surface ECG measurement of the action potential duration, i.e., the time for ventricular depolarization and repolarization to occur. Prolongation of the QT interval may reect changes due to abnormal depo- larization or repolarization, or it may be due to a combina- tion of both. Many factors affect the QT interval, including heart rate, autonomic nervous activity, electrolyte disorders, cardiac diseases, drugs, and congenital long QT syndromes. To compare QT intervals in individuals with different heart rates, some method of “correcting” the QT interval to a baseline rate, by convention 60 beats/min, is needed. The correction formula is a mathematical description of the relationship of QT to RR. If this formula does not accurately reect the actual relationship, then it will lead to an inac- curate value for QTc, the “corrected” QT interval, which may be an overestimate or underestimate. QT interval dependence on the heart rate was recognized early in the development of electrocardiography, when Ba- zett 7 published his formula for the corrected QT interval (QTc ) in 1920. Bazett examined the mathematical relation- ship of QT to RR interval and found that QT divided by the square root of the RR interval (in seconds) was a constant, K, which he gave as 0.37 for men and 0.40 for women. The constant K subsequently was replaced by the term QTc, the “corrected” QT, related to a “normalized” rate of 60/min, i.e., cycle length 1 second. Bazett’s formula was calculated based on a detailed analysis of the QT intervals in 15 male subjects aged 18 to 40 years and 18 female patients aged 21 to 30 years, with a single female subject aged 53 years. The effect of exercise on K was evaluated in three male subjects. Although the shortcomings of this correction have long been recognized, it demonstrated a remarkable enduring power, mostly attributable to its simplicity and ease of use. Unfortunately, with the widespread application of Ba- zett’s formula to different clinical and pathologic and phar- macologic states, the term “corrected” essentially has be- come a misnomer, as the “correction” formula has itself introduced further error into the QT normalization process. Milne et al. 8 showed that when Bazett’s formula is applied in the case of rate changes due to ventricular pacing or exercise, it undercorrects QTc at low heart rates (i.e., leads to shorter QTc) and overcorrects at high heart rates (i.e., leads to longer QTc intervals). In clinical situations, for single patients, the formula may yield acceptable values for QTc in an intermediate heart rate range, e.g., 50 to 70/min. 9 Since the appearance of Bazett’s article in 1920, many alternative methods to correct the QT interval have been proposed. In applying the various correction formulas to QT measurements in patients with myocardial infarction, Ah- nve 10 concluded that no other formulas (then derived) had been proved to be denitely superior to Bazett’s formula. Funck-Bretrano and Jaillon 11 pointed out that when consid- ering QT correction formulas, it is important to consider whether the QT correction formula is applied to a large population or to an individual. The formula that best de- scribes the QT versus RR relation for a population does not necessarily apply to an individual. In addition, even for the individual patient, there are multiple factors that may affect how the QT interval responds to heart rate changes. Auto- nomic nervous system effects can markedly inuence the QT interval. Some studies showed that vagal stimulation shortens the QT interval, independent of its bradycardic effect. 11-13 Others are consistent with a prolongation of ventricular refractoriness during vagal stimulation. 14-16 Ace- tylcholine, on the other hand, has no effect on the QT interval in normal subjects, but it lengthens the QT interval in long QT subjects. 17 In normal subjects with xed rate atrial pacing, atropine produced rate-independent shorten- ing of the QT interval, whereas beta blockers produced no signicant change. 18 The manner in which heart rate changes affect QT depends on the mechanism altering the heart rate. Milne at al. 8 showed that increasing the heart rate with ventricular pacing shortens the QT interval, but less J Cardiovasc Electrophysiol, Vol. 12, pp. 421-423, April 2001. Address for correspondence: Noel G. Boyle, M.D., Cardiology Division, Room 47-123 CHS, UCLA Medical Center, Los Angeles, CA 90095-1679. Fax: 310-206-5777; E-mail: nboyle@mednet.ucla.edu 421 Reprinted with permission from JOURNAL OF CARDIOVASCULAR ELECTROPHYSIOLOGY, Volume 12, No. 4, April 2001 Copyright ©2001 by Futura Publishing Company, Inc., Armonk, NY 10504-0418