Sex Difference in Risk of Torsade de Pointes With d,l-Sotalol
Background The present study was undertaken to test the hypothesis that women are more prone than men to develop torsade de pointes (TdP) in a defined cohort of patients exposed to the QT-prolonging antiarrhythmic drug d,l-sotalol.
Methods and Results In a database derived from 22 clinical trials involving 3135 adult patients who received oral d,l-sotalol (median follow-up, 164 days), TdP developed in 44 (1.9%) of 2336 men and in 33 (4.1%) of 799 women (P<.001). Logistic regression analysis identified female sex (P<.0001), presenting arrhythmia of sustained ventricular tachycardia or fibrillation (P<.0001), history of congestive heart failure (P<.001), and d,l-sotalol dose >320 mg/d (P<.001) as factors most predictive of TdP; in addition to these, a serum creatinine >1.4 mg/dL in women and >1.6 mg/dL in men was weakly predictive (P<.05). After adjustment for these risk factors, women had threefold greater odds of developing TdP than men. The sex difference in TdP risk was age independent and could not be explained by differential dose-related bradycardic responses in women versus men.
Conclusions Women are at increased risk of developing TdP during administration of d,l-sotalol. This finding needs to be taken into account, together with other TdP risk factors, when patients are treated with this antiarrhythmic agent. Given the consistency between the present and other recent observations, greater caution in women regarding use of QT-prolonging drugs, in general, is advisable.
The drug d,l-sotalol, a class III antiarrhythmic agent with β-adrenergic blocking activity,1 has a suppressive effect on ventricular2 and supraventricular3 arrhythmias and is currently approved in the United States as a therapy for the treatment of sustained VT and VF. The recently published results of the Electrophysiologic Study Versus Electrocardiographic Monitoring trial demonstrated superior effectiveness of d,l-sotalol, compared with several other antiarrhythmic drugs, in reducing arrhythmia recurrence in patients with sustained VT or VF.4
As in the case of virtually all drugs that prolong the QT interval, however, electrophysiological effects of d,l-sotalol predispose to the development of TdP, a potentially lethal ventricular tachyarrhythmia.5 6 7 8 9 Currently, efforts to limit the occurrence of this proarrhythmia during drug administration rely primarily on attention to QT interval and avoidance of hypokalemia and hypomagnesemia. Makkar et al10 recently marshaled quantitative evidence from the literature implicating female sex as another risk factor for drug-induced TdP. Given the potential clinical and pathophysiological implications of such an association, it would be important to test the hypothesis that women are at increased risk for developing TdP.
Using a database accumulated prospectively from more than 3000 patients enrolled in various clinical trials involving d,l-sotalol, MacNeil et al11 previously identified a number of risk factors for TdP, but they did not examine the role of sex. This large aggregate clinical experience with d,l-sotalol, therefore, provided us with a unique opportunity to test the hypothesis that women are more likely than men to develop TdP after exposure to a QT-prolonging drug.
From 1984 through early 1991, clinical information from 3135 patients who received oral d,l-sotalol in 22 trials conducted at multiple centers in North America and Western Europe was collected on case report forms and electronically entered into a computerized composite database. All patients received at least one drug dose in prospective clinical trials of various designs. The database included double-blind, randomized, placebo and active control trials in addition to open-label studies. Sustained VT or VF was the presenting arrhythmia in 41% of the patients. The remaining 59% were enrolled in trials investigating non–life-threatening arrhythmias, including premature ventricular beats, atrial fibrillation, nonsustained VT, and paroxysmal supraventricular tachycardia. Overall, 2336 patients (75%) were men and 799 (25%) were women, with mean ages of 60±12 and 56±16 years, respectively (minimum, 15 years for both sexes). The median duration of follow-up was 164 days (range, 1 to 3330 days).
During the course of the trials, investigators were required to report all cases of proarrhythmia, defined as a new arrhythmia or aggravation of an existing arrhythmia. So as to encompass the widest range of potential problems, a precise definition was not used. TdP was reported by investigators if the rhythm was polymorphic VT associated typically with a prolonged QT interval (described further below). Other proarrhythmias that were reported but are not the subject of the present study included an increase in the frequency or duration of premature ventricular beats, the development of nonsustained VT, aggravation of VT, and the development of VF.11 Heart rates measured during the course of the trials were provided to the database by participating investigators.
Clinical Description of TdP Cases
Seventy-seven cases of investigator-identified TdP that occurred after initiation of treatment with d,l-sotalol form the basis of this report. TdP was diagnosed electrocardiographically by the treating physician in all cases (review of available tracings summarized below). There were 44 men (57%) and 33 women (43%), with mean ages of 64±9 and 59±14 years, respectively (P=.12). Sixty-five cases of TdP (84%) occurred during hospitalization; the 12 patients with out-of-hospital TdP consisted of 8 men and 4 women. Cardiac arrest necessitating external DC shock occurred in 12 men (27%) and 7 women (21%) with TdP (P=.60); 2 of these 19 events (both in men) occurred out-of-hospital. In 5 additional patients (4 men and 1 woman), implantable cardioverter-defibrillator shocks (occurring out-of-hospital in 3 patients) were witnessed during paroxysms of TdP (with syncope or near syncope in 3 cases). Of 31 patients with documented symptoms in the absence of external or internal DC shocks, 8 (4 men and 4 women) developed syncope as their most severe symptom during TdP.
At the time of TdP with d,l-sotalol, 7 patients (9%) (4 men and 3 women) were taking concomitant drugs capable of prolonging the QT interval (procainamide in 3, despiramine in 1, trimethoprim/sulfamethoxazole in 1, probucol in 1, and hydroxychloroquine in 1). Erroneous overdosing with d,l-sotalol occurred in 2 patients (1 man [1280 mg/d] and 1 woman [800 mg/d]). Four patients (2 men and 2 women) were known to have a history of hypothyroidism and were on thyroid replacement therapy at the time of TdP. Of 21 patients in whom serum potassium and/or magnesium values were available from the same day as the TdP episode, potassium values were normal (range, 3.6 to 4.8 mEq/L) in 12 of 13 men and in all of 8 women; hypokalemia (2.9 mEq/L) was documented in a single male patient. Serum magnesium values were normal (range, 1.6 to 2.5 mEq/L) in 7 of 8 men and in 4 of 5 women, with documented hypomagnesemia in 1 man (1.2 mEq/L) and 1 woman (1.4 mEq/L).
ECG Documentation of TdP
ECG recordings were available for review in 25 cases (32%) of investigator-identified TdP. Forty percent of these 25 cases were in women, versus 44% of the remaining 52 cases (P=.73). There were also no significant differences in patients with versus without ECG documentation in regard to prevalence of coronary artery disease, history of CHF, history of sustained VT or VF, baseline QTc12 or JTc,12 mean d,l-sotalol dose at time of TdP, or mean time to TdP from start of dose.
ECG tracings at the time of TdP in the 25 patients were analyzed independently by one of the authors (M.H.L.). Runs of polymorphic VT were documented in 24 patients (96%) and VF in the remaining patient (a man who probably had prodromal occurrences of nonsustained ventricular tachyarrhythmias, as suggested by three episodes of syncope in rapid succession before cardiac arrest; his QT interval had attained a duration of 0.60 second on drug). Of the 24 patients with documented self-terminating runs of polymorphic VT, a mean of 3.5 (range, 1 to 9) runs per patient were recorded, with a median maximum duration of 14 (range, 3 to ≥30) beats, where the mean rate of the longest run per patient was 183 bpm (range, 148 to 240 bpm). Degeneration to VF was documented in 5 patients (21%). Of 17 patients in whom sufficient numbers of beats were recorded at the onset of at least one run of polymorphic VT, a “short-long-short” initiating sequence classic for TdP13 14 was evident in 16 (94%). The median maximum QT interval associated with a conducted sinus beat recorded shortly before a run of polymorphic VT (or VF) was 0.64 second (range, 0.50 to ≥0.84 second); this QT interval was immediately preceded by a mean RR interval of 1.3 second (range, 0.84 to 1.76 second), representing a postextrasystolic pause in 16 of 22 patients. The median JT interval corresponding to the median maximum QT interval was 0.55 second (range, 0.34 to ≥0.72 second). There were no statistically significant differences between men and women with respect to mean maximum rate during polymorphic VT or median maximum QT interval, although mean corresponding prior RR interval was slightly shorter in women than in men (1.1 versus 1.4 second, respectively, P=.05).
Univariate comparisons were performed with Pearson's χ2 test for categorical variables and Student's two-sample t test for continuous variables. Multivariate comparisons of risk factors for TdP were performed with logistic regression. Logistic regression was also used to calculate adjusted odds ratios and to estimate the probability of TdP. Logistic models were selected by stepwise procedures as well as by examining relevant models both a priori and a posteriori. When stepwise procedures were used, variables were stepped into the model if they were significant at the P=.05 level. Multivariate comparisons of continuous variables were performed with linear regression.
For risk factors that could be measured continuously, analyses were performed with the continuous factor, as well as cut points of the factor. Given that excretion of d,l-sotalol is almost exclusively via the kidney,15 special emphasis was given to measures of renal function, with the most predictive cut point presented here. A cut point of 50 mL/min was chosen for creatinine clearance, the latter estimated from a formula based on age, weight, sex, and serum creatinine.16 The cut point for serum creatinine was 1.4 mg/dL for women and 1.6 mg/dL for men. For d,l-sotalol dose, a cut point of 320 mg/d was chosen because it was the median dose prescribed and because previous clinical observations had suggested that there was a greater prevalence of proarrhythmia above that level11 ; it is also the highest dose usually recommended.17 The weight-normalized dose of 4.7 mg·kg−1·d−1 was also chosen as a cut point because it was near the median value. In the risk-factor analysis, dose was chosen as either dose at time of TdP (if that proarrhythmia occurred) or maximum dose received (for those patients who did not develop TdP). For QTc, a cut point of 465 ms (that includes values that round off to 0.47 second) was taken to demarcate clear QTc prolongation.18 19 A corresponding cut point of 365 ms was used for JTc on the basis of ECG findings in normal subjects.20
Heart rate responses to d,l-sotalol were analyzed in 1943 patients having ECGs both at baseline and on at least one drug dose. Mean heart rate on drug was determined by first averaging values from all ECGs available in each patient at the particular dose and then averaging those values for all patients with ECG data corresponding to that dose.
Median survival times free of TdP, for those patients developing this proarrhythmia, were compared by use of the median scores test. Survival functions were compared by the log-rank test.
Risk Factors for TdP
Table 1⇓ shows the number of patients with various baseline characteristics and the number of patients at different cut points of dosing, with the percentage of these patients who developed TdP. Women, patients who had presented with sustained VT or VF, and patients with a history of coronary artery disease or left ventricular dysfunction were most at risk for TdP. TdP was also more likely to develop at higher d,l-sotalol doses, which is expected because QT interval progressively prolongs as drug dose increases.21 Patients with elevated serum creatinine had a significantly greater prevalence of TdP; similarly, patients with low creatinine clearance (<50 mL/min) had a greater prevalence of TdP, but the difference was not significant. The increased prevalence of TdP among patients with a prolonged baseline QTc interval (≥465 ms) or JTc (≥365 ms) was not statistically significant.
Mean values of baseline characteristics and dosing for patients with and without TdP were compared. Patients with TdP had an average dose that was 87 mg/d greater than patients without TdP (P<.001). Mean baseline QTc was significantly longer in TdP patients (448 versus 433 ms in those without TdP, P<.01). However, this difference was largely attributable to a longer QRS duration (and slightly faster baseline heart rate) in TdP cases, because mean baseline JT and JTc intervals were not significantly different in patients with versus those without TdP. Creatinine clearance was found to be significantly lower in TdP patients (64 mL/min versus 73 mL/min in those without TdP, P<.01), but serum creatinine, magnesium, and potassium levels at baseline were not significantly different in patients with versus those without TdP.
Age and Risk of TdP in Women Versus Men
We investigated whether the increased likelihood of TdP in women might be limited to a certain age group. As shown in Fig 1⇓, however, within each age group analyzed, women had a risk of TdP typically exceeding or at least equaling that of men. With an age cut point of 50 years (approximate demarcation of menopause), the proportion of women developing TdP was slightly greater among those >50 versus ≤50 years (4.7% of 534 versus 3.0% of 265, respectively, P=.27) yet, in both of these age groups, exceeded by a factor of 2.3 the proportion of men developing TdP (2.0% and 1.3%, respectively).
Sex Differences in Prevalence of Concomitant TdP Risk Factors
In an attempt to explain the increased risk of TdP in women, we first explored whether the latter might have had a higher prevalence of other identified risk factors. On the contrary, women were significantly less likely to have a history of sustained VT or VF and had less evidence of structural heart disease than men (Table 2⇓). On the other hand, women received drug doses (normalized for body weight) higher than men, had a lower mean creatinine clearance, and had a greater prevalence of baseline JTc being ≥365 ms. Nonetheless, when patients were dichotomized according to a cut point close to the median dose or based on creatinine clearance, the proportion developing TdP was greater for women in both dose groups and in both creatinine clearance groups, eg, even in the low-dose/high-creatinine-clearance subset (Fig 2⇓). Similarly, female sex preponderance in propensity to TdP was consistent over both normal and high JTc interval ranges: TdP occurred in 3.7% and 6.8% of women with baseline JTc <365 and ≥365 ms, respectively, compared with 1.9% and 1.7% of men with corresponding JTc values.
Logistic Regression Analysis
After extensive evaluation of the risk factors, it was found that the logistic regression model most predictive of TdP was one that included terms for female sex (P<.0001), presentation with sustained VT or VF (P<.0001), history of CHF (P<.001), and dose >320 mg/d (P<.001). Weight-normalized dose was also a significant predictor of TdP (P<.01), but higher dosing, expressed either in mg/kg or mg/kg lean body mass,22 did not explain the increased incidence of TdP in women. When other baseline variables were added to a model that included female sex, history of sustained VT or VF, history of CHF, and dose, serum creatinine >1.4 mg/dL in women and >1.6 mg/dL in men was found to be weakly predictive of TdP (P<.05); no other baseline factors, including QTc (P=.13) and JTc (P=.07), added significant information predictive of TdP. Other measures of renal function (serum creatinine as a continuous variable, creatinine clearance as a continuous variable, and creatinine clearance dichotomized with a 50-mL/min cut point) were not statistically significant predictors.
Of the four principal risk factors for TdP, a history of sustained VT or VF was most predictive, present in 74% of the TdP cases. Female sex was the next most predictive risk factor. Among patients presenting with sustained VT or VF, 10.9% of women and 3.3% of men developed TdP (P<.0001); the corresponding TdP prevalences for women and men presenting with other arrhythmias were 1.8% and 0.7% (P<.05).
The unadjusted female/male odds ratio for developing TdP was 2.2 (P<.001). Various logistic regression models were analyzed to adjust for history of sustained VT or VF, history of CHF, dose, renal function, and baseline JTc. As shown in Table 3⇓, the risk of TdP in women actually increased (to three times the odds in men [P<.0005]) after these statistical adjustments, primarily because women had a lower prevalence of the concomitant risk factor of history of sustained VT or VF (Table 2⇑). Logistic regression was also used to estimate the probability of TdP for patients with particular combinations of the major risk factors, with demonstration of a consistently increased risk faced by women (Table 4⇓).
Adjustment for Possibly Unrecognized Cases of TdP
The foregoing analysis ultimately relied on a diagnosis of TdP that required ECG monitoring. Conceivably, some patients who developed TdP may have been included among the cases of “sudden death” observed in the overall cohort of d,l-sotalol–treated patients. We therefore analyzed prevalence of TdP by sex on a worst-case basis, ie, assuming that all sudden deaths actually represented cases of unrecognized TdP. Such an analysis had the effect of diluting the number of cases of women, since, as expected from other experiences,23 men predominated among cases of sudden death in the d,l-sotalol cohort (3.3% of 2336 men versus 2.8% of 799 women died suddenly, P=.45). Nonetheless, counting the two end points together, 6.8% of women had a known TdP or sudden death outcome, versus 5.1% for men (P=.08); after adjustment for a presenting arrhythmia of sustained VT or VF, history of CHF, and d,l-sotalol dose >4.7 mg·kg−1·d−1, the female/male odds ratio for TdP or sudden death was 1.7 (95% CI, 1.2 to 2.5; P<.003).
Comparison of Changes in Heart Rate With d,l-Sotalol in Women and Men
Since drug-induced TdP can be facilitated by bradycardia,6 7 24 we analyzed the heart rate–dose relationship by sex for the overall population in the d,l-sotalol database (Fig 3⇓). Although, as expected, heart rate progressively decreased with increasing drug dose in both sexes, women actually manifested slightly (1 to 2 bpm) higher mean heart rates than men over the course of this dose-response relationship. A similar sex difference was observed at baseline, before drug administration, with a mean heart rate of 76 bpm in women and 75 bpm in men (P=.07). At maximum drug dose per patient (mean, 5.4 mg·kg−1·d−1 in women and 5.2 mg·kg−1·d−1 in men, P=.25), the mean heart rate in women was significantly faster than in men (62 versus 60 bpm, respectively; P<.01).
Comparison of Time to TdP in Women and Men
Given the existence of a sex difference in risk of developing TdP, we sought to determine whether there might also be a sex difference in time course to its appearance. In the 77 patients who had TdP, the median time to TdP from initiation of most recent dose was 3 days, and 77% of the TdP cases occurred within 7 days of a dose initiation. TdP occurred earlier in women than in men, with a median time to TdP of 2 versus 3 days, respectively, although the difference was not statistically significant (Fig 4⇓).
The present study demonstrates for the first time, in a prospectively generated clinical database of patients exposed to a QT-prolonging drug, that female sex is a statistically significant independent risk factor for TdP. Specifically, after treatment with d,l-sotalol, women were found to face a threefold increased risk of developing this potentially life-threatening arrhythmia, after other important risk factors, including sustained VT or VF as the presenting arrhythmia, dose, and history of CHF, were taken into account. This finding thus validates the existence of a sex difference in risk of drug-induced TdP, as suggested by our recent study in which a greater-than-expected prevalence of women was observed among published cases of TdP occurring on a wide variety of QT-prolonging drugs.10
Possible Confounding Factors
Although clinical data originating from the various trials involving d,l-sotalol were accumulated prospectively, those studies were designed principally to assess drug efficacy and safety rather than to test hypotheses regarding risk factors for TdP. The present analysis was thus performed in a post hoc fashion. Nevertheless, a number of steps were taken to address this limitation. First, ECG recordings, available in nearly one third of cases of investigator-reported TdP, were analyzed and confirmed to possess morphological and/or prodromal cycle length change characteristics, as well as QT-interval prolongation, indicative of a diagnosis of TdP. These findings, as well as the similarity in clinical characteristics of patients with versus those without ECG documentation available for review by an author, support the designation of our study group as a TdP cohort. Second, logistic regression analysis enabled us to correct statistically for a variety of confounding risk factors, as identified in a prior univariate analysis.11 None of these possible confounders, alone or in combination, were able to account for the high (P<.001) statistical significance of female sex as a risk factor for TdP. Third, we also considered the possibility that some cases of “sudden death” occurring during follow-up in various d,l-sotalol trials may have represented cases of fatal but unrecognized TdP. Yet even when all sudden deaths (predominantly male) were added to the pool of cases of investigator-identified TdP, female prevalence remained significantly greater than expected after correction for other confounding TdP risk factors.
One potentially important variable lacking from our analysis was plasma level of d,l-sotalol at the time of TdP. It should be noted, however, that the drug has high bioavailability, undergoes virtually no hepatic transformation, and is excreted almost entirely through the kidneys,15 so that plasma level varies linearly with dose.21 Thus, when both most recent drug dose and serum creatinine were incorporated into the logistic regression analysis, the major determinants of d,l-sotalol plasma level were taken into account. Even after the logistic regression analysis was repeated with calculated creatinine clearance or dose corrected for lean body weight (because d,l-sotalol is hydrophilic), female sex remained a highly statistically significant predictor of TdP. A related issue is that the serum creatinine value used in our analysis was obtained at baseline and, therefore, may not have accurately reflected the value prevailing on the day of the TdP event. There is no reason to believe, however, that day-to-day fluctuations in renal function and serum creatinine would exhibit, on average, a systematic sex-dependent deviation over time. It is relevant to note that among published cases of quinidine-associated TdP, the greater-than-expected prevalence of women could not be explained by sex disparities in plasma quinidine level.10
Conceivably, some unknown or unmeasured covariate of sex may have accounted for the differences we observed between men and women in regard to risk of TdP. On the basis of known factors potentially contributory to the occurrence of TdP, however, there is little evidence to support this contention. For example, the possibility that more exaggerated d,l-sotalol–induced heart-rate slowing occurred in women, thereby enhancing the risk of TdP,6 7 24 is unlikely. At each progressively higher drug dose, bradycardic responses of men and women were similar, with women actually exhibiting slightly faster mean heart rates. Similarly, the mean RR interval immediately preceding the longest measured QT interval on ECG recordings made just before the occurrence of TdP was slightly shorter in women than in men. Although serum electrolytes drawn on the day of occurrence of TdP were available in only a minority of patients, the presence of either hypokalemia or hypomagnesemia was uncommon, with no evident sex differences. Finally, concomitant drugs that might have prolonged the QT interval were present in only 9% of cases of TdP, with no significant difference in prevalence between the sexes.
Clinical Severity and Time to TdP
The occurrence of cardiac arrest requiring DC cardioversion in 25% of patients with d,l-sotalol–associated TdP emphasizes the life-threatening potential of this proarrhythmic response. The observed prevalence of cardiac arrest may actually represent an underestimate of the likelihood of this catastrophic complication, given that an additional 6% of patients who developed TdP on d,l-sotalol had implantable cardioverter-defibrillator discharges that, in some cases, may have aborted progression to VF. Moreover, the fact that 83% of cases of TdP unassociated with cardiac arrest occurred as in-hospital events allowed for prompt proarrhythmia recognition and drug discontinuation or dosage reduction, possibly averting progression to VF in these patients. Additionally, some TdP-related cardiac arrests may have gone unrecognized among the sudden deaths that occurred during the clinical trials.
Although more prone to developing TdP on d,l-sotalol, women were not found to be significantly more likely than men to experience cardiac arrest as a manifestation of this proarrhythmia, nor were ECG characteristics of TdP different between the sexes. In general, three quarters of all cases of TdP occurred within 1 week of the most recent dose initiation, and the median time to TdP was similar in women and men (2 and 3 days, respectively). Thus, in-hospital monitoring after initiation of a new dose of d,l-sotalol can permit identification of most cases of TdP in either sex.
Sex Differences in Duration of Cardiac Repolarization and Propensity to TdP: A Widely Manifest Phenomenon
The increased susceptibility of women to d,l-sotalol–associated TdP may be viewed as part of an emerging, consistent pattern of sex differences in cardiac repolarization under a wide variety of conditions. For example, even in the normal adult population, the absolute QT interval at comparable heart rates and the QTc interval are longer in women than in men.12 25 26 In the hereditary, autosomal dominant long-QT syndrome, which predisposes gene carriers to symptomatic or fatal TdP, the majority (rather than the expected 50%) of probands and other symptomatic family members are female27 28 ; women in these families also exhibit greater QTc prolongation.29 Moreover, among reported cases of TdP associated with complete heart block, a bradycardia-induced (ie, nondrug) model of long-QT syndrome, a significantly greater-than-expected proportion of patients are women.30 Of further interest, the sex disparity observed with respect to risk of d,l-sotalol–related TdP is paralleled by a greater degree of sotalol-induced JTc prolongation among women, even after adjustment for sex difference in baseline JTc.31 Women also display a greater degree of QTc prolongation induced by the antihyperlipidemic drug probucol.32
Taken together, these data suggest that in women, compared with men, cardiac repolarization is an intrinsically slower process and prone to greater delays, with an increased risk of TdP, in various predisposing settings (eg, inherited abnormality of repolarization, certain drugs, and slow heart rate). It must be emphasized, however, that the present study and prior analysis10 provide evidence only for the existence of a relatively greater susceptibility in women versus men to development of drug-induced TdP. This sex disparity could reflect the greater presence in women of a TdP-promoting factor or, instead, a relative dearth in women of some protective factor found more abundantly in men. Recent ECG,29 33 34 clinical,35 and experimental36 observations actually favor the latter explanation and point to a possible role for androgens in blunting QT prolongation in men. The finding in our study that women >50 and those ≤50 years old had a similar likelihood (both absolute and relative to men) of developing TdP with d,l-sotalol certainly argues against an estrogenic basis for the sex difference in propensity to this proarrhythmia. A more definitive understanding of underlying pathophysiological mechanisms must await additional basic research in this area.
The data presented here emphasize the need for greater caution in the use of QT-prolonging drugs in women. Specifically with respect to d,l-sotalol, when initiation of therapy (or dose increase) is contemplated, the patient's sex, presenting arrhythmia, and type and extent of compromised cardiac function all need to be taken into account as preexisting risk factors for TdP, together with the risk factor of dose (Table 4⇑). The weaker but possibly significant adverse contribution of renal dysfunction, as reflected in the serum creatinine, should also be considered in dose selection.17 If these various factors are taken into account in estimating risk of TdP, decisions regarding use of d,l-sotalol and extent of in-hospital ECG monitoring during drug therapy or dose adjustment can be made more rationally.
The existence of an increased tendency in women to development of TdP also has relevance to the design and interpretation of clinical trials involving QT-prolonging medication. Estimation of expected incidence of TdP in such trials should probably take into account this sex difference. Among unmonitored sudden deaths that may be observed with these drugs, a greater-than-expected female prevalence may imply an important contributory role of (unrecognized) TdP proarrhythmia. These considerations highlight the importance of adequate representation of women in clinical trials involving QT-prolonging drugs.
Selected Abbreviations and Acronyms
|CHF||=||congestive heart failure|
|JTc||=||rate-corrected QT-minus-QRS interval|
|TdP||=||torsade de pointes|
The authors thank James D. Marsh, MD, for his critical review, and Karen Beal and Diane Szubeczak for their assistance in the preparation of this manuscript.
Presented in part at the 43rd Annual Scientific Session of the American College of Cardiology, Atlanta, Ga, March 15, 1994.
- Received March 18, 1996.
- Revision received June 4, 1996.
- Accepted June 10, 1996.
- Copyright © 1996 by American Heart Association
Smith WM, Gallagher JJ. ‘Les torsades de pointes’: an unusual ventricular arrhythmia. Ann Intern Med. 1980;93:578-584.
Patterson E, Szabo B, Scherlag BJ, Lazzara R. Arrhythmogenic effects of antiarrhythmic drugs. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1995:496-511.
Arstall MA, Hii JTY, Lehman RG, Horowitz JD. Sotalol-induced torsade de pointes: management with magnesium infusion. Postgrad Med J. 1992;68:289-290.
Bazett HC. An analysis of the time-relations of electrocardiograms. Heart. 1920;7:353-357.
Betapace drug package insert. Berlex Laboratories, 300 Fairfield Rd, Wayne, NJ 07470-7358
Moss AJ, Robinson J. Clinical features of the idiopathic long QT syndrome. Circulation. 1992;85(suppl I):I-140-I-144.
Keating M. Linkage analysis and long QT syndrome: using genetics to study cardiovascular disease. Circulation. 1992;85:1973-1986.
Zareba W, Moss AJ, and the LQTS Study Group. Criteria for delayed repolarization in patients with wide QRS complex. J Am Coll Cardiol. 1994;37A. Abstract.
Cobb LA, Weaver WD, Fahrenbruch CE, Hallstrom AP, Copass MK. Community-based interventions for sudden cardiac death: impact, limitations, and changes. Circulation. 1992;85(suppl I):I-98-I-102.
Roden DM, Hoffman BF. Action potential prolongation and induction of abnormal automaticity by low quinidine concentrations in canine Purkinje fibers: relationship to potassium and cycle length. Circ Res. 1985;56:857-867.
MacFarlane PW, Lawrie TDV. The normal electrocardiogram and vectorcardiogram. In: MacFarlane PW, Lawrie TDV, eds. Comprehensive Electrocardiology: Theory and Practice in Health and Disease. Elmsford, NY: Pergamon Press Inc; 1989;1:407-463.
Moss AJ, Schwartz PJ, Crampton RS, Tzivoni D, Locati EH, MacCluer J, Hall WJ, Weitkamp L, Vincent GM, Garson A Jr, Robinson JL, Benhorin J, Choi S. The long QT syndrome: prospective longitudinal study of 328 families. Circulation. 1991;84:1136-1144.
Lehmann MH, MacNeil DJ. Sex differences in magnitude of JTc prolongation with d,l-sotalol. Circulation. 1995;92(suppl I):I-277. Abstract.
Lehmann MH, Timothy KW, Frankovich D, Fromm BS, Keating MA, Locati EH, Taggart RT, Towbin JA, Moss AJ, Schwartz PJ, Vincent GM. Age-sex influence on QTc and the QT-heart rate relationship in genotypically-characterized long QT syndrome families. J Am Coll Cardiol. In press.
Drici MD, Burklow TR, Haridasse V, Glazer RI, Woosley RL. Sex hormones prolong the QT interval and downregulate potassium channel expression in the rabbit heart. Circulation.. 1996;94:1471-1474.