Gender and Cardiac Arrhythmias

Abstract

The incidence of certain clinical arrhythmias varies between and women. Clinical and experimental observations suggest the existence of true differences in electrophysiologic properties between the sexes. We review these differences, possible mechanisms, clinical implications, and therapeutic considerations in the treatment of various arrhythmias in women. (Tex Heart Inst J 2001;28:265–75)

Some reports have noted differences in the incidence of certain clinical arrhythmias according to the sex of those studied (Table I). ^1–4 Some of these differences are related to known variations in the frequency of underlying organic heart disease, such as coronary artery disease (CAD) and associated ventricular arrhythmias. ² However, clinical and experimental observations suggest that true differences in electrophysiologic properties exist between men and women. These differences have important clinical implications and raise special therapeutic considerations in the management of arrhythmias in women. In this review, we outline these differences, as well as the clinical implications and therapeutic considerations in the treatment of arrhythmias in women.

TABLE I. Gender-Related Frequency in Clinical Arrhythmias

Electrocardiographic and Electrophysiologic Properties: Differences between the Sexes

The exact pathophysiologic mechanisms by which the sex of the individual influences the occurrence of cardiac arrhythmias are far from being understood. Current knowledge stems from clinical and electrocardiographic observations and from surprisingly sparse experimental data on cellular electrophysiologic mechanisms; most studies have focused on repolarization in the ventricular myocardium. Nevertheless, the available data support 2 general mechanisms: hormonal effects on the expression or function of ion channels (either directly or indirectly), and differences in autonomic tone, which may act alone or in concert with hormonal effects (Table II).

TABLE II. Postulated Mechanisms for Gender Differences

Clinical and Electrocardiographic Observations

Heart Rate.

As early as 1920, Bazett ⁵ observed that women had a higher resting heart rate than did men. This observation was later corroborated in several other reports. ^6–10 In a large population-based study (n=5,116), the average heart rate for women was found to be 3 to 5 beats faster per minute than that of men. ⁶ These findings may be explained in part by differences in physique and exercise tolerance, but differences in intrinsic properties of the sinus node and autonomic modulation could have an effect. In a study of resting heart rate before and after double autonomic blockade (administration of propranolol and atropine intravenously to block any autonomic influence on the sinus node), women had significantly shorter sinus cycle lengths than those of men (Fig. 1). ¹⁰ Although these data detract greatly from the autonomic modulation hypothesis, they raise the possibility of intrinsic differences in sinus node automaticity. In a study of 100 consecutive male and female patients ¹¹ without structural heart disease, both corrected and uncorrected sinus node recovery times were shorter in women. ^1,11 Inappropriate sinus tachycardia, a rare but increasingly recognized clinical entity, is substantially more common in women. ^12–14

Fig. 1 Sinus cycle length before and after double autonomic blockade. In both instances, men had a significantly longer cycle length (slower heart rate) than did women.

*P <0.05

(Reprinted from American Journal of Medicine, Vol 100. Burke JH, Goldberger JJ, Ehlert FA, Kruse JT, Parker MA, Kadish AH. Gender differences in heart rate before and after autonomic blockade: evidence against an intrinsic gender effect. p. 537–43, copyright 1996, ¹⁰ with permission from Excerpta Medica Inc., and from the author.)

Heart Rate Variability.

Sex- and age-related differences in heart rate variability and baroreceptor sensitivity have been reported by several authors. ^15–17 Women tend to have a lower low-frequency power and a higher high-frequency to low-frequency ratio, suggesting a greater parasympathetic influence on their heart rate compared with that of men. In a study by Huikuri's group, ¹⁷ postmenopausal women given estrogen replacement therapy showed a restoration of baroreceptor sensitivity and an increase in power of both the high- and low-frequency domains (Fig. 2). These results strongly suggest that sexual differences in autonomic modulation exist and may be explained in part by hormonal influences.

Fig. 2 Heart rate variability between women with and without hormone (estrogen) replacement therapy.

HF = high frequency; HRT = hormone replacement therapy; LF = low frequency; pNN50 = percentage difference between successive RR intervals >50 msec

*P <0.05

(Data adapted with permission from Huikuri HV, et al. ¹⁷)

QT Interval.

Differences between men and women regarding the QT interval have also been described by Bazett ⁵ and confirmed by others. ^8,18 Women have a longer corrected QT interval compared with that of men, and the difference becomes more pronounced at lower heart rates. In a population-based Canadian study, Rautaharju and colleagues ¹⁹ reported that the QT difference between men and women was due to a drop in the corrected QT that occurred in males after puberty (when androgen levels are highest). However, the interval in men gradually increased with age until 50 years, at which point it paralleled that of women (Fig. 3). ¹⁹ Similar observations with regard to male QT shortening at puberty were made in families with genotypically characterized long QT syndrome. ²⁰ These findings support the hypothesis that actions of the sex hormones may explain these differences in the QT interval.

Fig. 3 Corrected QTI interval according to age (birth to 75 yr) in 14,379 children and adults. Note a decrease in the men's QT interval after puberty, which then increases over time but never reaches that of women.

(From Rautaharju PM, Zhou SH, Wong S, Calhoun HP, Berenson GS, Prineas R, Davignon A. Sex differences in the evolution of the electrocardiographic QT interval with age. Can J Cardiol 1992;8:690–5. ¹⁹ Reproduced with permission of Can J Cardiol.)

Clinical Distribution of Arrhythmias.

Sexual differences with regard to the epidemiology and the clinical distribution of arrhythmias have been described (Table I). Inappropriate sinus tachycardia is, for the most part, a disease of women. ^12–14 In 623 patients who underwent invasive electrophysiologic testing due to various tachycardias, atrioventricular node reentrant tachycardia (AVNRT) was twice as common in women, whereas atrial tachycardia affected both sexes equally. ³ The reverse was true for atrioventricular reentrant (circus-movement) tachycardia (AVRT), atrial fibrillation (AF), and ventricular fibrillation (VF), which occurred more often in men. Similar trends in the distribution of supraventricular tachycardias (SVTs), AF, VF, and sudden death were described in the Framingham cohort. ² A more detailed description of specific arrhythmias follows later in this review.

Menstrual Cycle Changes.

Burke and colleagues ^10,21 reported that a lower average heart rate was present during menses than during the follicular or luteal phases of the menstrual cycle, but that the response to double autonomic blockade was identical regardless of phase (Fig. 4). Myerburg and associates ²² found that the chance of inducing SVT during electrophysiologic testing was greatest at the onset of menses or during the premenstrual phase. Rosano and coworkers ²³ correlated the cyclic variation of spontaneous episodes of SVT to the temporal profile of ovarian hormones. There was a significant positive correlation between the frequency and duration of SVT episodes and progesterone levels, and an inverse correlation with β-estradiol levels (Fig. 5). An increase in the frequency and duration of SVT episodes was noted on day 28, when progesterone levels are higher, as opposed to day 7 of the menstrual cycle, when estrogen predominates. These observations argue that the SVT-prone state is hindered by estrogen and facilitated by progesterone. Whether these effects are a result of modulation of autonomic tone, of direct cellular electrophysiologic effects, of the 2 in combination, or of some factor not yet described, is unknown.

Fig. 4 Sinus cycle lengths before and after double autonomic blockade during the various menstrual phases. The sinus cycle length was significantly longer (*P <0.03) during the menstrual phase of the cycle at baseline. This difference was no longer evident after autonomic blockade.

(Reprinted from American Journal of Medicine, Vol 100. Burke JH, Goldberger JJ, Ehlert FA, Kruse JT, Parker MA, Kadish AH. Gender differences in heart rate before and after autonomic blockade: evidence against an intrinsic gender effect. p. 537–43, copyright 1996, ¹⁰ with permission from Excerpta Medica Inc., and from the author.)

Fig. 5 Incidence of supraventricular tachycardia (SVT) plotted against serum levels of plasma ovarian hormones on day 28 of the menstrual cycle. There was significant positive correlation between progesterone and SVT episodes and an inverse correlation between estradiol-17β and SVT.

(Reprinted with permission from Elsevier Science [The Lancet 1996;347:786–8 ²³] and from the author.)

Pregnancy and the Postpartum Period.

Supraventricular tachycardia is reported to be the most common arrhythmia during pregnancy and the postpartum period. This statement has drawn weak support from small case series ^24–27 and from 2 larger studies, ^28,29 one of which was retrospective. Tawam's group ²⁸ retrospectively studied 60 women who had SVT and found an increased risk (relative risk, 5.1; P <0.001) of onset or exacerbation of any SVT during pregnancy. The report from Lee and co-authors, ²⁹ however, is less convincing. From a study cohort of 207 women, the portion experiencing 1st-onset SVT during pregnancy was a mere 4%. The association between the 1st onset of SVT and pregnancy was not statistically significant in patients with an accessory pathway. In patients with AVNRT, a lower risk of 1st-onset SVT was seen. In both studies, however, the symptoms of SVT were increased during pregnancy. These observations support the argument that the progesterone-rich gravid state facilitates SVT. Conversely, since SVT is a very common arrhythmia in women of this age group, it is unclear whether a true association exists.

New-onset ventricular tachycardia during pregnancy has been reported. ^30,31 In most cases, there was no evidence of structural heart disease. Brodsky and associates ³⁰ have suggested that these tachycardias are catecholamine-mediated (that is, aggravated by physical and emotional stress) and are therefore responsive to β-blocker therapy. The number of patients studied, however, is low, and there has been no systematic validation of these observations. In an analysis of the International Long QT Syndrome Registry, Rashba and co-authors ³² reported an increased risk of death, sudden cardiac death (SCD), and aborted SCD during the postpartum period but not before or during pregnancy. More importantly, they found this risk to be significantly reduced by β-blocker therapy. The postulated mechanism involves the changing hormones surrounding delivery, coupled with autonomic modulation. The relative tachycardia of pregnancy is thought to shorten the QT interval and to provide some protection. During the postpartum period, there is a drop in the resting heart rate, which, added to the physical and emotional stress, may increase the transmural dispersion of repolarization and facilitate the generation of arrhythmias. Although the exact mechanisms are unclear, the possible effects of hormonal influence and autonomic modulation are supported by these observations.

Effects on Cellular Electrophysiology

The notion that sex hormones can act directly or indirectly on the cardiovascular system is supported by the presence of cytoplasmic and nuclear receptors specific for these hormones in the various tissues of the cardiovascular system and its neuroregulatory system. ³³ In a study by Johnson and coworkers, ³⁴ ventricular myocytes from estrogen-receptor-deficient mice, compared with controls, showed a higher expression of cardiac L-type calcium channels. The density of ion channels appears to be regulated by the estrogen receptor, with diminished estrogen receptor occupancy upregulating the number of functional L-type calcium channels. A higher level of functional calcium channels could explain a shorter sinus cycle length on the basis of a more rapid phase 0 of the action potential. However, if one were to invoke this mechanism, the relatively higher heart rate in women should be present across the various age groups and diminish during menopause. Moreover, one would expect the QT interval to shorten because of decreased opposition to the repolarizing currents during phase 3. However, this is clearly not what is observed clinically. ^5,8,18 Hence, other mechanisms must be operational.

Drici's group ³⁵ showed that estradiol administration in ovariectomized rats decreased the mRNA for 2 distinct classes of delayed rectifier potassium channels and prolonged the QT interval. They also found that dihydrotestosterone (DHT) had similar effects on mRNA levels and on the QT interval, although the administration of quinidine resulted in QT prolongation only in the estradiol-treated mice. These findings are consistent with those reported by Liu and colleagues, ³⁶ who studied sexual differences in the potassium currents and cycle-length-dependent QT intervals in rabbits. In female rabbits, 2 repolarizing potassium currents (I_KR and I_KI) were of significantly less magnitude than those in males. Moreover, there was significantly greater QT prolongation in accordance with corresponding increments in cycle length. Hara and associates ³⁷ administered estradiol and DHT to ovariectomized rabbits and found that DHT shortened the early part of ventricular repolarization, while estradiol, by exerting its effects on the latter part of repolarization, prolonged the action potential duration. The administration of an I_KR blocker resulted in a significantly greater prolongation of action potential, and even showed evidence of after- depolarizations in the estradiol-treated rabbits (Fig. 6).

Fig. 6 The effects of the I_KR blocker, E4031, on placebo-, estrogen-, and dihydrotestosterone-treated rabbit papillary muscles.

DHT = dihydrotestosterone; EAD = early afterdepolarization; EST = estrogen

(From Hara M, Danilo P Jr, Rosen MR. Effects of gonadal steroids on ventricular repolarization and on the response of E4031. J Pharmacol Exper Ther 1998;285:1068–72. ³⁷ Reproduced with permission from the publisher and from the author.)

The differences in potassium channels and currents support the clinical observations regarding longer QT intervals, the response to the administration of class-III agents, and the predisposition of women to proarrhythmia. The combined effect of reducing L-type calcium currents and prolonging repolarization via the effects of estrogen on the potassium currents may at least partly explain the less frequent occurrence of AF in women. ³⁸ Most of these studies were performed on in vivo preparations of ventricular myocardium. However, given the paucity of data on atrial myocardium, it seems reasonable to suggest that similar or related mechanisms might be operational in the atria. More investigation is necessary in order to find definitive answers.

Effects on Autonomic Tone

In an elegant review, Larsen and Kadish ³⁹ proposed that sex steroid-induced changes in autonomic tone might have an effect on the sex-related differences in cardiac arrhythmias. This hypothesis arose from early observations that the resting heart rate was consistently higher in women than in men. ^5,6 Subsequent experiments using double autonomic blockade demonstrated that intrinsic heart rate is higher in women than in men. ¹⁰ This suggested that the sex-related differences might be intrinsic to the sinus node with minimal input from the sympathetic or parasympathetic nervous systems. In the interim, other studies evaluating sex differences in heart rate variability and baroreceptor sensitivity (see above) suggested strongly that hormones do affect autonomic tone. ¹⁷ Ng and coworkers ⁴⁰ evaluated the influence of age and sex on sympathetic tone by measuring muscle sympathetic nerve activity (MSNA). They reported significantly more MSNA in men than in women and increasing disparity in MSNA between the two with advancing age. Both sympathetic and parasympathetic influences have been implicated in the genesis of arrhythmias in those patients with either acquired or congenital long QT syndromes and in those with Brugada syndrome. ^41–45 The existence of an M-cell region and heterogeneity within the myocardium provides a mechanism by which differential responses of the various myocardial cell types to sympathetic or parasympathetic excess or withdrawal could lead to a greater transmural dispersion of repolarization and the genesis of cardiac arrhythmias.

Rubart and von der Lohe ⁴⁶ suggested nitric oxide as another possible mechanism by which sex steroids might affect autonomic tone. Estradiol has been shown to augment nitric oxide release from endothelial cells, which in turn can modulate autonomic input by inhibiting neurotransmitter release or by antagonizing the actions of norepinephrine at the cellular level. ⁴⁷ The net effect, as demonstrated in the canine model, is modulation of sinus discharge and atrioventricular (AV) nodal conduction by augmentation of vagal tone and inhibition of sympathetic neurotransmission. ⁴⁸ These experimental findings may further explain the cyclic variation in paroxysmal SVT. ²³

Specific Cardiac Arrhythmias

Atrioventricular Node Reentrant Tachycardia.

The most frequent mechanism of AV node-dependent tachycardia is AV-node reentrant tachycardia (AVNRT), which constitutes approximately 60% of all narrow-complex tachycardias seen in clinical practice. ^49–51 A 2:1 female predominance of AVNRT has been reported in at least 2 large series. ^3,52 Reproducible initiation and termination of paroxysmal AVNRT with programmed stimulation supports the mechanism of reentry, as does the longitudinal physiologic dissociation of the AV node (“dual” AV node physiology) found in many patients with AVNRT. ^49,52,53 A shorter refractory period of the slow pathway, which has been described in women, ⁵⁴ increases the window of inducibility and may partially explain the increased incidence of AVNRT. Fortunately, there are no differences in response to the treatment of choice—radiofrequency ablation is equally successful (>95%), regardless of sex. ^52,55 Similar to the findings of others, the experience at our institution was that women constituted 64% of those ablated for AVNRT (247 of 371), and our success rates were comparable.^*

Atrioventricular Reentrant (Circus-Movement) Tachycardia.

The 2nd most common mechanism of paroxysmal SVT is reentry with the atrioventricular node as the antegrade limb and an accessory AV pathway as the retrograde limb (so-called circus-movement tachycardia or atrioventricular reentrant tachycardia [AVRT]). ^49–51,56 These accessory pathways can either be concealed (not evident on the resting electrocardiogram) or be manifest (demonstrating preexcitation or antegrade conduction on the resting electrocardiogram—the classic Wolff-Parkinson-White syndrome). In large series, the diagnosis of AVRT has been twice as frequent in men as in women. ^3,57 Accessory pathways are more common in men than in women and are more likely to be manifest. In contrast, spontaneous orthodromic AVRT is found more often in women. In a study of 690 patients evaluated for accessory pathways, ⁴ aborted sudden death due to AF and VF was less common in women, indicating a less favorable natural progression of Wolff-Parkinson-White syndrome in men. In our hospital registry, ablation for accessory pathways (n=273) occurred with equal frequency between men (138) and women (135). Accessory pathways in both men and women were more often manifest than not, but with a tendency toward more concealed accessory pathways in women.^* The treatment of choice is radiofrequency ablation, which is successful in more than 90% of cases. ^58–60 Outcomes again are similar, regardless of the sex of the patient. Given the increased incidence of SVT in women with accessory pathways during pregnancy, radiofrequency ablation should be considered before these patients attempt pregnancy.

Atrial Fibrillation.

Atrial fibrillation is the most common SVT, affecting approximately 0.4% of the overall population. ^2,61,62 Its incidence increases dramatically with age, and the development of AF is associated not only with increased morbidity rates but also with an approximate doubling of all-cause mortality. Men are at greater risk of AF than are women. In the Framingham Heart Study, ^2,62 men were shown to have a 1.5 times greater risk of developing the arrhythmia than did women. However, when AF occurred in women, their survival rates were lower and their risk of death was higher than that of men. This remained true even after results were adjusted for other risk factors. Data from the Coronary Artery Surgery Study showed that in patients with CAD, men had a 5.4 times greater risk of developing AF than did women. ⁶³ The prevalence of AF is higher in men than in women in all age groups. However, because there are almost twice as many women than men who are older than 75 years in the general population, the absolute number of women with AF in older age groups exceeds that of men. 2,62,64,65 Men are also more likely to develop AF after cardiothoracic surgery. ⁶⁶ Advanced age and male sex have been shown to be strong risk factors for postoperative AF. It remains unclear why men are more susceptible than are women to the development of this arrhythmia. Sexual differences in the prognosis of AF have also been found. Women are more likely than men to experience recurrent AF after successful pharmacologic or electrical cardioversion. ⁶⁷ In addition, there is some evidence to suggest that women are at higher risk for systemic embolization during AF than are men. ⁶⁸ Other studies, however, have shown no significant difference in this risk. The benefits of the various strategies aimed at the control of heart rate, the prevention of thromboembolic complications, and the conversion to and maintenance of sinus rhythm are similar regardless of the sex of the patient. However, caution must be used in the administration of antiarrhythmic agents to women, because women experience QT prolongation and proarrhythmia more frequently than do men.

Inappropriate Sinus Tachycardia.

Inappropriate sinus tachycardia is an uncommon clinical syndrome characterized by an elevated resting heart rate accompanied by an exaggerated response to stress or exercise. The underlying mechanisms are not yet completely defined but seem to involve enhanced automaticity of the sinus node or abnormal autonomic regulation of the sinus node (either excess sympathetic or reduced parasympathetic tone). ^14,69 Studies that have used autonomic blockade, autonomic maneuvers, and heart rate variability analysis provide support for one or both mechanisms and suggest the possibility of a multifactorial cause. Typically, patients with inappropriate sinus tachycardia are young (<40 years) and female (approximately 90%), 12–14,69,70 and there is a larger-than-expected proportion of health care workers. The reasons for the high percentage of women with this condition are unknown. The diagnosis of inappropriate sinus tachycardia is one of exclusion and is demonstrated by the following: 1) a resting heart rate ≥100 beats/min, or a heart rate that increases to ≥100 beats/min with minimal exertion (such as standing or walking slowly); 2) P-wave morphology identical to that exhibited during sinus rhythm; 3) endocardial atrial activation in a high-to-low and right-to-left pattern, indicating the origin to be in the sinus node; 4) tachycardia that cannot be induced or terminated using standard programmed stimulation; 5) a duration of the tachycardia that is chronic and nonparoxysmal; and 6) exclusion of secondary causes (such as fever, anemia, and hyperthyroidism). ^12,13,71 The therapy for inappropriate sinus tachycardia is empiric.

Torsades de Pointes and Congenital and Acquired Long QT Syndrome.

The high incidence of cardiac events in women, specifically torsades de pointes, has been described in connection with both congenital and acquired long QT syndromes. ^72,73 Results from the International Long QT Syndrome Registry and from trials involving antiarrhythmic agents have provided some insight into possible mechanisms.

In the Long QT Registry, 70% of the probands and 58% of the affected members were women. ⁷⁴ At any point in time, men had a shorter corrected QT interval than did women. Men were more prone to experience syncope, cardiac arrest, and unexplained SCD until puberty; thereafter, the predisposition shifted dramatically toward women. ⁷⁵ At puberty, QT prolongation did not occur in girls, but QT shortening was noted in boys (Fig. 7). As described earlier, women with congenital long QT were also found to be at a heightened risk of cardiac events during the postpartum period; however, these events could be controlled with β-blockers. ³²

Fig. 7 Percentage distribution of probands (top) and family members with corrected QT (QTc) interval >440 ms (bottom) by sex and age at baseline electrocardiography. Up until the age of 15 years, the proportion of males:females is approximately 1:1. After 15 years, there is a marked female predominance.

(From Locati HE, Zareba W, Moss AJ, Schwartz PJ, Vincent M, Lehmann MH, et al. Age- and sex-related differences in clinical manifestations in patients with congenital long-QT syndrome: findings from the International LQTS Registry. Circulation 1998;97:2237-44. ⁷⁴ Reproduced with permission from Lippincott Williams & Wilkins and from the author.)

In a pooled analysis of 322 patients with drug-induced torsades de pointes, 70% of the patients were women. ⁷⁶ This predominance was irrespective of underlying left ventricular function, electrolyte abnormalities, and baseline QT intervals. The SWORD trial, terminated early because of increased mortality, showed a 4.7-fold increase in the risk of proarrhythmia and sudden death in women, a finding that has been corroborated in multiple subsequent studies and their pooled analyses, and with the use of racemic d,l-sotalol in different patient populations. ^77–79

These observations, specifically with regard to QT shortening at puberty, suggest a protective effect of androgens, a notion that has been supported in several animal studies. ^35–37 Administration of quinidine, which has I_KR-blocking effects, to estrogen-treated rabbits resulted in QT prolongation, whereas no such prolongation was noted in testosterone-treated animals. ³⁵ Going further, other investigators showed that such I_KR-blocking effects were more pronounced in the epicardial myocardium than in the endocardial myocardium. ⁸⁰ This differential effect can result in a greater dispersion of repolarization and an increased propensity for arrhythmogenesis. At 1st glance, this appears to be consistent with earlier experiments, ^36,37 which revealed reduced potassium-channel mRNA with estradiol treatment in rabbits. Unfortunately, in the same experiments, the administration of DHT had similar effects on potassium-channel mRNA levels and on the QT interval. Subsequent quinidine administration, however, resulted in QT prolongation in the estrogen-treated rabbits only. Although this information improves our understanding of these clinical syndromes, it also makes it apparent that there are mechanisms operational other than potassium-channel modulation. Why men are less prone to proarrhythmia remains unclear. Perhaps this phenomenon is due to a combination of effects—calcium channel, potassium channel, and other channels not yet described—that are modified by varying degrees of autonomic influence. Another hypothesis is that each of the ion channels has a functionally different response when exposed to sex-steroid hormones. This would explain why a similar reduction in potassium-channel mRNA might result in functionally less rectifying currents in women and a predisposition to proarrhythmia.

Ventricular Tachycardia and Sudden Cardiac Death.

Significant sex-related differences in the origins of ventricular tachycardia and SCD were delineated in the Framingham Study. ^81–84 Over a 26-year follow-up period, the incidence of SCD increased with the age of the subjects, with a male predominance in all age groups (Fig. 8). ⁸⁵ Although this difference was partly explained by the epidemiology of CAD—women present 10 to 20 years later than men—the men remained more prone to sudden death in matched cohorts without CAD.

Fig. 8 Sex- and age-specific rates for sudden death, morbidity, and total coronary deaths in subjects aged 35 to 84 years: 26-year follow-up, Framingham study.

morb. = morbidity; mort. = mortality

(From Lerner DJ, Kannel WB. Patterns of coronary heart disease morbidity and mortality in the sexes: a 26-year follow-up of the Framingham population. Am Heart J 1986;111:383-90. ⁸⁵ Reproduced with permission from Mosby, Inc., A Harcourt Health Sciences Company.)

Differences between the sexes were also noted in the risk factors for CAD, ventricular tachycardia, and sudden death. ^81,82 The classic risk factors for CAD and ventricular tachycardia were predictive in men but not in women. Left ventricular hypertrophy, hyper-cholesterolemia, tobacco use, and obesity were risk factors in men, whereas hyperglycemia, an elevated hematocrit level, and diminished vital capacity were predictors in women. Diabetes mellitus was predictive in both but had a significantly greater influence in women. ⁸⁶ The presence of nonsustained ventricular tachycardia and premature ventricular contractions with or without underlying CAD correlated with an increased risk of sudden death in men but not in women. ⁸⁶ Likewise, the presence of premature ventricular contractions after a myocardial infarction led to an increased risk of death in men but not in women. ⁸⁷

A decade ago, studies suggested that women who survived SCD were less susceptible to arrhythmia inducibility during electrophysiologic testing than were men. ^88–89 However, in 1996, a larger study ⁹⁰ reported no significant differences after correcting for underlying CAD. All of these studies were retrospective analyses in relatively small populations. Therefore, the matter of arrhythmia inducibility in women remains largely unsettled. Female survivors of SCD are also less likely than men to have underlying CAD; they are more likely to have either normal hearts or other types of heart disease (Fig. 9). ⁹⁰ In a study of outcomes after treatment with implantable cardioverter-defibrillators, Kudenchuk and co-authors ⁹¹ reported that women were younger, had better ejection fractions, had lower defibrillation thresholds, were less likely to have organic heart disease, and were more likely to present with ventricular fibrillation. On 2-year follow-up, the incidence of ventricular events necessitating device therapy was lower in women than in men (37% vs 52%). ⁹¹

Fig. 9 Relative proportions of underlying heart disease in survivors of cardiac arrest: A) female and B) male.

CAD = coronary artery disease; DCM = dilated cardiomy-opathy; RV = right ventricular; VHD = valvular heart disease

(From Albert CM, McGovern BA, Newell JB, Ruskin JN. Sex differences in cardiac arrest survivors. Circulation 1996;93:1170–6. ⁹⁰ Reproduced with permission from Lippincott Williams & Wilkins and from the author.)

These observations carry important clinical implications. If SCD tends to occur more unexpectedly in women, risk stratification and primary prevention may be more difficult than it is in men. Nevertheless, the same therapeutic considerations should be applied equally to all who present with malignant ventricular tachyarrhythmias. Even if women have less structural heart disease, fewer arrhythmic events, and less inducibility during electrophysiologic testing, these factors should never preclude the use of implantable devices, especially in survivors of sudden death.

Conclusion

There are clear sexual differences in the incidence and prevalence of certain cardiac arrhythmias. The exact mechanisms responsible for these differences are as yet unknown. The scant evidence available suggests 2 general mechanisms: sex steroid hormone effects on ion channels and modulation of autonomic tone. Many questions remain, suggesting even more avenues for investigation, such as these:

• What are the physiologic effects of the gonadal hormones (estrogens, androgens, and progesterone) on human atrial myocytes?

• How do the gonadal hormones affect the various ion channels?

• Are receptors to gonadal hormones distributed differently in men and women?

• Is the distribution of ion channels the same in the atrium as it is in the ventricle?

• Are there gender-specific variations in the responsiveness of these ion channels?

• Is there an M-cell equivalent in atrial myocardium?

• Does hormone replacement or antagonism have antiarrhythmic or proarrhythmic effects?

The answers to these questions could clarify the basic mechanisms of these differences between the sexes and facilitate the development of newer, more effective methods of diagnosis and treatment. At present, despite differences in the epidemiology of supraventricular and ventricular tachyarrhythmias that have been described, therapeutic options are broad and generally effective regardless of the sex of the patient.