Aetiology and pathophysiology

Accessory pathways (APs) by definition traverse the AV groove, and are identified in all anatomical quadrants of the tricuspid and mitral annuli. Although WPW is prevalent, pathological demonstration is rare as this myocardial feature is microscopic and rarely systematically sought in postmortem study. Although APs are typically congenital, many well substantiated reports indicate that the clinical aspects of WPW may sometimes be acquired de novo, typically in patients who have undergone cardiac surgical procedures. An increased prevalence of WPW is noted among patients with Ebstein’s anomaly, in whom APs are typically anatomically associated with the abnormal tricuspid valve, and often multiple.

Among symptomatic patients, clinical presentation is typically with palpitations or a sustained episode of SVT. A bimodal age distribution is observed: a substantial number of patients present in the first month of life (in some cases, prenatally), and a secondary, more diffuse peak through the school age years. Prenatal and infantile presentation of SVT is often associated with signs of congestive heart failure, perhaps due in part to the relatively prolonged interval between onset and diagnosis. Although pre-excitation is typically a persistent phenomenon, it may in some cases be variably present, even after electrocardiographic diagnosis of WPW. Infants with SVT may demonstrate pre-excitation that emerges months or even years after their initial diagnosis of SVT. The reasons for this variability are poorly understood. The conduction velocity of the AV node as well as the refractory period of the APs change with developmental stage in childhood, factors that may be expected to affect the degree of pre-excitation (a balance between the speed of AV conduction and the percentage of the ventricular myocardium depolarised in advance of the normal activation sequence). However, they do not fully explain observations of “occult pre-excitation” or some certain aspects of AP behaviour during AF, and other mechanisms such as concealed retrograde conduction may need to be inferred to explain this phenomenon.1

Sudden death associated with WPW

Although the relationships of the bundle of Kent to both pre-excitation and reciprocating tachycardia were proposed soon after WPW was described in the 1930s, the rapid conduction of atrial fibrillation via the AP as a mechanism of sudden death in WPW was not identified until 1971 by Dreifus2 (fig 2). A North American population study suggested that the risk of this severe complication of WPW is about 3–4% over several decades, and this is conventionally quoted as an annual risk rate of ∼0.1%/year.3 These data were generated in an era which antedated both the widespread use of the ECG for screening of non-specific symptoms (fewer asymptomatic patients identified) and the modification of the population by ablation (longer periods to observe natural history). More recently, a long term follow-up of military personnel with WPW reported a mortality rate of 0.02%/year.4 Two additional studies comprising more than 4000 additional years of patient follow-up present estimated mortality rates between these two.5 6 In the aggregate, these studies suggest a consensus estimate of 0.05%/year. However, a widely noted report from Italy has also presented a WPW patient group that was prospectively followed for 3 years, and which recorded a severe event rate (defined as death or potentially lethal arrhythmia recorded on monitoring) that was 10 times higher—∼0.5%/year.7 This important, disparate finding requires further exploration and confirmation, as it notably alters the outcome of risk-to-benefit calculations which guide therapy decisions discussed below. Because it is difficult to resolve these numbers, it is also useful to check them against estimates of global sudden death rates in the “healthy” population (0.0013%/year),8 the fraction of such deaths attributable to WPW in pathological and clinical studies (2–4%),9 10 and an estimated prevalence of the disease in the general population of 1/1000. Utilisation of these statistics yield a projected mortality rate of 0.03–0.05%/year among WPW patients, tending to support the lower range of estimates for mortality presented above.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Pre-excited atrial fibrillation in Wolff–Parkinson–White syndrome, with shortest pre-excited RR interval (SPERRI) <200 ms.

Asymptomatic WPW

Symptoms associated with WPW are almost solely due to the occurrence of transient or sustained SVT. Frequent and/or severe SVT related symptoms may themselves prompt many patients to seek catheter ablation, especially if attempts to prevent episodes using drugs such as β-blockers prove unsuccessful. However, many patients with WPW are completely asymptomatic, and are diagnosed serendipitously at the time of cardiac evaluation proposed for an unrelated indication (for example, evaluation of murmur, chest pain, dizziness, screening for sports participation or before drug prescription). The proportion of patients with WPW who are truly asymptomatic is unknown, and its estimate is made difficult because of the increasing use of the ECG in general, and particularly for screening young populations (athletes, those receiving certain classes of prescription drugs). Counterbalancing this, there may often also be a desire of clinicians and patients to attribute retrospectively remembered episodes of transient palpitation to WPW after the disease has been serendipitously diagnosed. Nonetheless, there seems to be agreement within the literature that 50% or more of patients with WPW are asymptomatic.3 6 This also extends to the much smaller number of WPW patients who present with cardiac arrest, 40–50% of whom may have had no symptoms or knowledge of their diagnosis before the event.9 11

Risk stratification in WPW

Many investigators have sought to identify risk factors for cardiac arrest in WPW, studying both clinical and electrophysiological markers (box 1). Absence of symptoms in WPW patients is not in itself a marker of low risk for arrest, as the occurrence of sudden death in these patients is well documented. The affirmative (and more difficult to prove) converse of this statement—that symptomatic and asymptomatic WPW patients have an equivalent risk of cardiac arrest—has not been demonstrated. However, several studies reviewed below suggest that electrophysiological properties of the two groups may, in some ways, suggest that the asymptomatic WPW patient may in fact be at somewhat lower risk for cardiac arrest.

It was demonstrated in early clinical studies that patients who had suffered the most severe symptoms of tachycardia, particularly cardiac arrest, have shorter pre-excited RR intervals (SPERRI) during episodes of atrial fibrillation than those with more mild symptoms. Those with SPERRI ⩽220 ms (250 ms in some studies), and/or with other measurable properties of the AP such as its effective refractory period and rapid 1:1 conduction of atrial pacing are thus considered to be “high risk” for a life threatening event, relative to the general population with WPW. Conversely, longer SPERRI and/or longer AP refractory periods identify a lower risk patient subgroup.12

This association has been investigated by many independent observers. Milstein et al compared the electrophysiological properties of asymptomatic WPW patients to matched symptomatic controls and found that asymptomatic patients were at somewhat decreased risk for short SPERRI.13 Bromberg et al performed electrophysiological evaluation of 60 variably symptomatic children with WPW and showed that short pre-excited RR intervals were found in 35% of patients with SVT as their symptom, in 74% presenting with syncope, and in 100% of those who had a cardiac arrest; the overall incidence of low SPERRI was 53%.14 The severity of presentation was not associated with other clinical features of disease that might be used to predict risk. They concluded that presence of a SPERRI determined to be <220 ms was associated with an approximately threefold increase in risk of cardiac arrest, compared to the general WPW population. Dubin et al studied 119 children with WPW, divided into asymptomatic patients, those with SVT, and those with syncope. They found no difference between these groups with respect to SPERRI <270 ms, inducibility of SVT, the presence of multiple pathways and pathway location.15 Thus, the absence of symptoms did not predict the findings of intracardiac risk stratification, and therefore does not obviate need for testing. This finding intuitively corresponds to the observation made above that a significant fraction of patients experiencing cardiac arrest are asymptomatic before the event.

Less invasive risk stratification techniques have been proposed and studied in varying degrees as proxy measures which can be used to obviate the need for intracardiac electrophysiology study (EPS). These include the use of Holter monitoring and exercise testing, to determine whether pre-excitation disappears at high physiological heart rates. Loss of pre-excitation on graded exercise testing generally predicts longer AP effective refractory periods,16 and this is also assumed to be true for physiological sinus tachycardias recorded on ambulatory ECG monitoring. Thus, these tests are sensitive to those at risk, but relatively non-specific and of low positive predictive value. Although a formal SPERRI or pathway refractory period is not determined, the patient is considered “safe” in these cases, and more invasive study deferred. Bershader et al reviewed 88 patients who underwent exercise testing and subsequent intracardiac EPS and found that, although loss of pre-excitation on exercise was associated with longer average AP refractoriness, only 15% of patients had that finding, and among them many had pathways that were ultimately considered to be of “intermediate” risk.17

Oesophageal pacing has also been used to probe pathway behaviour. It is often possible to induce atrial fibrillation for formal assessment of SPERRI using this technique,18 and it has been shown that intracardiac and transoesophageal assessment of AP refractory period is well correlated.19 It must be born in mind, however, that induction of arrhythmia by this technique also carries the same intrinsic risks as those induced in intracardiac EPS, and ventricular fibrillation during oesophageal study has been reported.

It has also been suggested with less unanimity that other clinical associations may be important and useful markers: the nature and severity of clinical presentation,13 the presence of multiple pathways and an anatomical location of the AP in the right anteroseptal AV groove,11 20 and the inducibility of any SVT (including both AV reciprocating tachycardia and atrial fibrillation) on electrophysiological study.7 Identifying a reduced and reliable set of risk factors from these different studies is complicated by the fact that some are likely to be electrophysiologically associated.

Treatment for WPW

In the past two decades, the technique of catheter ablation (predominantly using radiofrequency energy) has been widely and successfully applied to the treatment of WPW. During the early development of this curative technique, patients most severely affected by symptoms of WPW—those who had experienced an aborted cardiac arrest, or who had frequent, disabling and/or drug refractory episodes of SVT—were selected for ablation. However, as the technique has matured and become more widely available, it has been increasingly applied to patients who are minimally symptomatic or asymptomatic.

It is established that effective catheter ablation of the AP in WPW eliminates pre-excitation of the ECG as well as the occurrence of AV reciprocating tachycardia. Although more difficult to prove in clinical study (due to the low frequency of the observed event), effective ablation also appears to eliminate the excess risk of cardiac arrest attributed to the disease.21 This finding is of particular importance, because sudden cardiac arrest (SCA) risk is the only indication for therapy in asymptomatic patients, and there is no indication that it can be reduced by antiarrhythmic drug treatment, which may be useful for some patients with symptomatic SVT. Interestingly, ablation also appears to reduce the elevated propensity of many WPW patients to the occurrence of atrial fibrillation.22

Unfortunately, catheter ablation also carries with it a low but finite risk of death or major adverse event. Registry and prospective studies in children and adults have been published over the last 15 years that allow very specific estimates of acute and chronic efficacy of ablation for AP mediated tachycardias, and the risk of procedural death, major and minor adverse complications. The Pediatric RF Registry provided acute outcome data on a multicentre series of 5383 AP ablations performed in the 1990s.23 The acute success rate for radiofrequency (RF) ablation in the “late era” (that is, post-learning curve) experience was 95.2%, with variation based on AP location. In this broad experience, one death occurred in 3187 “late era” ablations, and incidence of major acute complications of higher grade AV block, perforation and thrombus/thromboembolic event in this group were 0.6%, 0.5%, and 0.2%, respectively. Risk factors for adverse events in the RF Registry included small patient size, pathway location and presence of concomitant heart disease. The subsequent Prospective Assessment of Pediatric Cardiac Ablation (PAPCA) study looked closely at ablation outcomes in 481 patients with AP mediated SVTs.24 Acute success rate for ablations was 96%. Follow-up was exceptionally good over 12 months, and showed that recurrence rates averaged 10.7%, higher than had been suspected based on registry data.

Cost–benefit analysis of ablation therapy for asymptomatic WPW

The data presented above have fuelled a vigorous and longstanding debate as to whether patients with asymptomatic WPW should undergo prophylactic catheter ablation. Given these reasonably precise estimates of the risk of therapy and the risk of the disease untreated, a risk–benefit analysis may be performed to determine the social utility of ablation for this indication. Such an analysis is complicated by the fact that non-lethal complications of ablation such as stroke and AV block are difficult to compare directly against mortality risk (box 2).

One concept which has been applied to ablation therapy in the past and may be useful in this regard is “number needed to treat” analysis (NNT). NNT is the reciprocal of the absolute risk reduction afforded to a population by the treatment proposed. In the case of asymptomatic WPW, this number represents the number of patients who must be treated by prophylactic catheter ablation to prevent a single cardiac arrest, over some specified time period. The data presented above would support an uncontroversial estimate of the long term success of a single WPW ablation of 90%, and it can also be assumed that all successful ablations eliminate the risk of SCA associated with WPW. Given this relative risk reduction, the absolute risk reduction is dependent on: (1) the underlying annual risk of SCA in the population of asymptomatic WPW patients; and (2) the number of years over which that risk is integrated (that is, the “time horizon” over which doctor and patient look forward in determining whether or not an ablation is useful). The higher the annual risk of SCA assumed, and the greater the number of years over which the effect of therapy is considered, the more efficient will be a strategy of prophylactic ablation.

As outlined above, estimates of annual risk of SCA in WPW vary roughly by a factor of 10, between 0.05–0.5%. If one considers 20 years as a “risk reduction window” in the calculation of the value of prophylactic ablation, an assumption of 0.05% annual risk implies that ∼111 patients would need to undergo ablation to prevent one SCA, while an assumption of 0.5% annual risk implies that the same effect could be achieved with ∼11 ablations. Each of these ablated patients would incur the cost of the procedure in terms of health care expense and risk of sublethal adverse event, which can be used to perform other types of risk–benefit analyses. The pronounced difference between these two assumptions is mitigated somewhat by the fact that a proportion of asymptomatic patients over time develop symptoms and would then likely realise other treatment benefits. Conversely, although the benefits of successful ablation are life long, it is unlikely that healthy, asymptomatic patients, their families and physicians routinely discount their cardiac risks over several decades in deciding whether or not to assume smaller but more immediate risks and discomforts of an ablation procedure.

At present, asymptomatic WPW is considered to be a class 2b indication for catheter ablation in patients 5 years and older, and a class 3 indication for patients younger than 5 years.25 This is a strong indicator that, despite ongoing advances in ablation technique and clinical research in WPW, ablation in asymptomatic patients must be recommended only after careful consideration. Clinical equipoise still exists on this question, with valid expert opinion expressed on both sides of the question of indication for ablation. The divergence of academic opinion and acceptable clinical practice in this condition highlights the difficulties in quantifying the frequency of low frequency events in patient populations, the importance of tracking the evolution of procedural efficacy and adverse event rates, and the variability of risk interpretation among patients and physicians.

Activity restriction and prescription drug use in asymptomatic WPW

In patients with WPW, no evidence based data have been published indicating a clear association between either activities (sports participation) or prescription drug use, including stimulants and psychoactive drugs. However, the increased risk of sudden death associated with WPW in combination with anecdotal, highly publicised events of cardiac arrest in apparently healthy young people have driven policy regarding sports participation which is based on expert consensus. Recent recommendations are that all patients with untreated WPW undergo exercise test, echocardiogram (to exclude associated cardiovascular abnormalities) and, perhaps, ambulatory ECG recording during athletic activity. Asymptomatic patients may be allowed to participate if these are reassuring, but EPS may be recommended before allowing participation in “moderate to high intensity competitive sports”. Athletes with high risk pathways are prohibited from sports participation, until they have been successfully ablated.26

With respect to prescription drugs, it is known that some commonly used agents, such as stimulants, cause modest increases in heart rate and adrenergic tone, and others, particularly psychotropic agents, may for unclear reasons be associated with an increased risk of sudden death. To date, no expert guidance has been provided for the use of prescription drugs in patients with WPW.

It is useful to note that, in contrast to long QT syndrome, proscriptions against intense physical exertion and use of certain drugs are grounded in well researched understanding of specific mechanisms by which these activities or agents act to increase the risk of lethal arrhythmia. In contrast, the general risk associated with WPW has typically been conflated with the non-specific risks of these activities or agents, without proposal of any mechanistic hypotheses as to why they might be synergistic.

Recommendations

What, then, is the proper management of asymptomatic WPW? Our understanding of the intrinsic risk of this syndrome when left untreated is key to answering this question, and the answer involves balancing the small risk of an iatrogenic event at the time of treatment against the low and difficult to measure cumulative risk of the arrhythmia untreated over decades. Although the evidence for its efficacy is lacking, it is generally accepted that, unless clearly shown to be “low risk”, these patients should be restricted from varsity athletics and higher level competition, and stimulant medicines may only be used with heightened medical supervision and intermittent monitoring for arrhythmogenesis. There is no evidence that administration of antiarrhythmic medications of any form reduce the risk of sudden death over time. Thus the questions to be answered are: is ablation indicated for all, some, or none of these patients?

The brief analysis above has shown that the risk–benefit ratio for catheter ablation therapy of asymptomatic WPW is strongly dependent on one’s assumptions regarding perceived risk, and the extent to which the physician and patient are looking into the future when deciding on ablation. Although that leaves considerable latitude for individual decision making, it seems clear that patients in whom pre-excitation disappears at physiological heart rates tend to have pathway properties which do not place them in the higher risk group, and they can be followed and allowed to pursue normal lifestyle and activities with reasonable safety. Thus, an exercise stress test and ambulatory ECG should be a routine part of evaluation of patients determined to be asymptomatic WPW patients. It is our general practice that patients who do not have a reassuring pattern of clear loss of pre-excitation during one of these studies undergo risk stratification by intracardiac EPS. This is often combined with catheter ablation if it is deemed safe and necessary, based on patient size (preferably >20–25 kg, in our institution), pathway location (deferring ablation midseptal and para-Hissian pathways when possible), and any complicating medical or social issues (for example, desire to participate in varsity or elite athletic competition, need to treat with stimulant or other proarrhythmic medications).