Author + information
- Received May 9, 2002
- Revision received July 24, 2002
- Accepted August 19, 2002
- Published online January 15, 2003.
- Carlo Pappone, MD, PhD*,* (, )
- Vincenzo Santinelli, MD*,
- Salvatore Rosanio, MD, PhD*,
- Gabriele Vicedomini, MD*,
- Stefano Nardi, MD*,
- Alessia Pappone, MD*,
- Valter Tortoriello, MD*,
- Francesco Manguso, MD, PhD*,
- Patrizio Mazzone, MD*,
- Simone Gulletta, MD*,
- Giuseppe Oreto, MD* and
- Ottavio Alfieri, MD*
- ↵*Reprint requests and correspondence:
Dr. Carlo Pappone, Department of Cardiology, Electrophysiology, and Cardiac Pacing Unit, San Raffaele University Hospital, Via Olgettina 60, 20132 Milan, Italy.
Objectives The aim of this study was to assess in a large cohort of asymptomatic subjects with Wolff-Parkinson-White (WPW) pattern the usefulness of invasive electrophysiologic testing (EPT) in predicting the occurrence of arrhythmic events over a five-year follow-up.
Background Sudden death may be the first clinical manifestation of the WPW syndrome in previously asymptomatic patients. Serial EPTs have been proposed to identify patients at risk.
Methods A total of 212 consecutive asymptomatic WPW patients were enrolled after a baseline EPT; patients were followed for five years, and 162 patients (115 noninducible and 47 inducible) patients underwent a second EPT.
Results After a mean follow-up of 37.7 months, 33 patients became symptomatic. Of the 115 noninducible patients, 18.2% lost anterograde accessory pathway (AP) conduction, 30% retrograde AP conduction, and only 4 (3.4%) developed symptomatic supraventricular tachycardia (SVT). Of the 47 inducible patients, 25 with sustained atrioventricular reciprocating tachycardia (AVRT) and atrial fibrillation (AF), and 4 with nonsustained AVRT and AF became symptomatic for SVT (n = 21) and AF (n = 8). They were younger, had shorter AP anterograde refractory periods, and multiple APs compared to patients who remained asymptomatic (for all comparisons, p < 0.0001). Of the eight patients with symptomatic episodes of AF and inducible sustained AF, two had a resuscitated cardiac arrest and one died suddenly; all three patients were inducible for AVRT and AF and had multiple APs.
Conclusions In asymptomatic WPW subjects, EPT may be a valuable tool to stratify the risk of symptomatic and fatal arrhythmic events.
The occurrence of sudden death as the first clinical manifestation of the Wolff-Parkinson-White (WPW) syndrome in previously asymptomatic individuals has stimulated interest in the possibility of identifying, within the large population of asymptomatic WPW subjects, the subset of patients at risk of developing fatal arrhythmic events (1–3). This challenge is justified because curative treatment with radiofrequency catheter ablation can be a reasonable therapy in individuals at definite risk (4). The purpose of this study was to prospectively examine, in a large cohort of asymptomatic WPW subjects, the usefulness of invasive electrophysiologic testing (EPT) in predicting the occurrence of symptomatic arrhythmic events over a five-year follow-up.
Between 1993 and 1996, a total of 212 consecutive subjects with asymptomatic WPW were recruited. All individuals were referred after WPW pattern was found either incidentally at routine examination (n = 160) or during a medical check-up before admission to a competitive sport (n = 35) or a high-risk occupation (n = 17). Age at enrollment ranged from 7 to 63 years (mean, 35.8 ± 20.5 years). According to electrocardiographic criteria (5), 99 patients had left-sided, 52 right-sided, 47 posteroseptal, and 14 anteroseptal accessory pathways (APs). Ten patients had structural heart diseases (mitral valve prolapse, n = 5; hypertrophic cardiomyopathy, n = 2; and hypertension, n = 3).
The EPT was performed by using at least four catheters positioned in the right atrium, right ventricular apex, His bundle region, and coronary sinus. The protocol included 1) atrial and ventricular incremental pacing up to the minimal cycle length (CL) maintaining atrioventricular or ventriculoatrial 1:1 conduction; and 2) programmed atrial and ventricular stimulation at drive CL of 400 and 350 ms. The anterograde effective refractory period (AERP) of AP was defined as the longest coupling interval at which anterograde block in the bypass tract was observed. If atrial fibrillation (AF) was not induced by atrial extrastimuli, atrial burst pacing at basic CLs of 300, 200, and 100 ms for 20 s was performed. Intravenous isoproterenol (1 to 4 μg/min) or atropine (0.02 to 0.004 mg/kg) was used to facilitate arrhythmia induction. Preexcited QRS morphology, mean ventricular rate, and the shortest RR interval between two consecutive preexcited QRS complexes (SPRRI) were calculated during sustained AF, when inducible. The same study protocol was used in follow-up studies. Arrhythmias were considered sustained if they lasted >1 min. Accurate endocardial mapping of retrograde atrial activation at the right and/or left atrioventricular groove was performed during right and/or left ventricular pacing and/or atrioventricular reciprocating tachycardia (AVRT). Multiple APs were defined as two or more distinct sites of early anterograde and/or retrograde activation.
The follow-up began at the time of the first EPT (baseline) and ended in December 2001. A second EPT was performed at five years, or earlier if symptoms or arrhythmias occurred. After baseline EPT, patients were closely followed, and any medication needed was given. All subjects were asked to report each of the following symptoms: palpitation, asthenia (fatigue at rest), rest dyspnea, effort dyspnea, dizziness, chest oppression, blurred vision, and syncope. Detailed reviews of clinical events, physical check-up, 12-lead surface electrocardiogram (ECG), 24-h ambulatory ECG monitoring, and exercise testing were performed annually. Baseline and follow-up, clinical, and electrophysiologic data were collected prospectively and stored in an electronic database.
Data are expressed as mean ± SD except when indicated. For categorical variables the chi-squared test was performed, unless the exact test was required for frequency tables when more than 20% of the expected values were <5. The independent-sample ttest procedure was used for continuous variables. Kaplan-Meier survival analysis was applied to estimate the cumulative risk of spontaneous arrhythmic events (supraventricular tachycardia [SVT], AF, and ventricular fibrillation [VF]) in patients with and without inducibility. Log-rank test was used to compare the survival functions. The binary logistic regression was used to examine the relationship between arrhythmic events (absence/presence = 0/1) as dependent variable and possible predictors (age, gender, AP–AERP at baseline and at follow-up both before and after isoproterenol, and inducibility) as independent variables. The model was estimated using block entry of variables. The coefficients obtained from the logistic regression analyses were also expressed in terms of odds of occurrence of an event. A p value ≤5% was considered statistically significant. All statistical analyses were performed by using the SPSS package for Windows (version 11.0.1, SPSS Inc., Chicago, Illinois).
Of the 212 study patients, 3 (1.4%) were lost to follow-up, 129 (60.8%) remained asymptomatic at the end of follow-up, whereas 33 (15.6%) developed arrhythmic events. Forty-seven (22.2%) patients refused the second EPT. No significant clinical and electrophysiologic differences were observed at baseline between the 162 subjects who had a follow-up EPT and the 50 who had not. The 47 patients who refused the five-year EPT remained asymptomatic during the entire follow-up period.
Clinical and electrophysiologic characteristics of the 162 subjects who completed the follow-up
Patients’ ages at the time of enrollment ranged from 7 to 63 years (mean, 33.6 ± 14.3 years); there were 105 (64.8%) male and 57 (35.2%) female patients (Table 1). The majority of subjects were in the third and fourth (50%) decade; 84 patients had left-sided, 37 right-sided, 37 posteroseptal, and 4 anteroseptal APs. Three patients had mitral valve prolapse, one had hypertrophic cardiomyopathy, and the remaining patients had no underlying heart disease. The surface ECG showed loss of anterograde preexcitation after five years in 21 patients. Of the 162 patients, 115 (71%) were not inducible and 47 (29%) inducible (Table 2); in the latter group, 17 patients had nonsustained AF (4 during isoproterenol), 19 had sustained (9 during isoproterenol) AVRT (mean CL, 244.8 ± 19 ms), and in the remaining 11 patients with induced-AVRT, the arrhythmia degenerated into totally preexcited sustained AF (mean SPRRI, 222.8 ± 17.9 ms). In 17 of 212 (8%) subjects (16 inducible and 1 not inducible) accurate atrial mapping during ventricular pacing showed different sites of early retrograde activation, suggesting the presence of multiple APs. Multiple APs did not participate in the re-entry circuit in all patients, and only in some subjects were both APs associated with anterograde conduction; in two patients AF exhibited two different preexcited morphologies. Subjects with multiple APs were younger (mean age, 15.4 ± 3.3 years; p < 0.001) than patients without. In all inducible patients, no evidence of cardiac disease was found.
Comparison of clinical and electrophysiologic characteristics of subjects who became symptomatic versus those who remained asymptomatic
During follow-up (37.7 ± 16.1 months; range 14 to 60 months), 33 (20%) of the 162 patients (mean age 20.1 ± 8.6 years) became symptomatic; arrhythmias were documented by standard ECG or Holter monitoring as a cause of symptoms, and included SVT in 25 patients and AF in 8 others (Table 3). Two of these eight patients subsequently had an aborted sudden death with documented ventricular fibrillation (VF) and one died suddenly owing to VF. Patients who became symptomatic were younger and had shorter AERP of AP, compared with those who remained asymptomatic; this difference persisted after isoproterenol (for all comparisons, p < 0.0001). Most of these patients (29; 87.8%; 21 males, 19.9 ± 8.8 years) were inducible for sustained AVRT, which in 11 subjects degenerated into preexcited sustained AF. Of note, of the 17 patients with inducible nonsustained AF, none developed symptoms. Only 4 (1 male, 21.7 ± 7.5 years) of the 33 patients who became symptomatic were never inducible. No patient who became symptomatic had an underlying heart disease. Patients who became symptomatic were advised to undergo radiofrequency catheter ablation of APs. Of the subjects who remained asymptomatic, 21 showed loss of anterograde AP conduction even during atrial pacing and isoproterenol administration, with intact retrograde conduction (age at loss of preexcitation 47 ± 15 years). These 21 patients had longer AP–AERP (317.3 ± 25 vs. 245.4 ± 34 ms; p < 0.001) and longer minimal 1:1 anterograde conduction CL (330.7 ± 38 vs. 265.5 ± 34 ms; p < 0.001) over AP during the baseline study than patients who had persistent preexcitation. Retrograde conduction over the AP was lost in 35 noninducible patients despite administration of isoproterenol, atropine, and adenosine. These 35 patients had longer minimal 1:1 retrograde conduction CL over AP than patients in whom ventriculoatrial conduction persisted (358.7 ± 33.2 and 365.7 ± 54.4 ms, vs. 270.3 ± 32.4 and 262.2 ± 43.4 ms, respectively; each p < 0.001). Of the 129 patients who remained asymptomatic, 111 (86%) were not inducible and 18 were inducible. Noninducible subjects were older than inducible ones (38.7 ± 13.3 vs. 27.1 ± 9.1 years; p < 0.001).
Clinical and electrophysiologic characteristics of subjects who had spontaneous AF
In the spontaneous AF group (eight patients), the presenting symptom was dizziness in three, sustained palpitations in one, and no important complaint despite extremely rapid ventricular response in four patients. All eight patients were inducible for sustained AVRT triggering AF. The four patients who presented with dizziness and palpitations were successfully ablated a few months after the first spontaneous episode of AF, and following ablation no arrhythmia recurred. The remaining four patients who had multiple APs and inducible AF refused ablation because AF was well tolerated, but unfortunately one of them died suddenly, and two others experienced an aborted sudden death due to documented VF. The first patient died while he was playing soccer and was found to have VF; the remaining two patients after resuscitation from VF had immediate successful ablation of four APs (Table 4).
Electrophysiologic and clinical predictors
The sensitivity, specificity, and positive and negative predictive values of short AERP (≤250 ms) and positive EPT testing in predicting arrhythmic events were 71.9%, 72.6%, 44.2%, 89.5%, and 87.9%, 86.0%, 61.7%, and 96.5%, respectively. The combination of short AERP and inducibility was associated with a sensitivity of 93.7% and a specificity of 67.6%, with positive and negative predictive values of 46.9% and 97.3%.
The predictive value of inducibility and age, by binary logistic regression analysis, is shown in Table 5. Inducibility and young age are associated with a significantly increased risk of arrhythmic events. Inducible subjects had a 17.9-fold risk (95% confidence interval [CI], 3.85–83.26) of events, whereas older individuals had a 0.91-fold risk (95% CI, 0.85–0.97). The Kaplan-Meier survival analysis (Fig. 1) showed that the cumulative risk of developing arrhythmias was significantly (p < 0.0001) higher for subjects with a positive EPT testing.
This study represents the largest and longest prospective electrophysiologic follow-up study in asymptomatic WPW subjects. The vast majority of these patients showed persistent ventricular preexcitation throughout a mean follow-up of >60 months; 20% became symptomatic for AVRT or AF. Of the eight patients who became symptomatic for AF, one died suddenly with documented VF and two were resuscitated from VF. Patients experiencing VF were young, and two of them had no symptoms during inducible-sustained AF despite extremely rapid ventricular rates. Our data confirm a large retrospective study by Timmermans et al. (3)in a population of 690 WPW patients in which 15 patients (2.2%) had an aborted sudden death, and VF was the first manifestation in 8 others (53%).
Serial electrophysiologic changes and arrhythmia inducibility
Of the 162 patients who completed the follow-up study, 115 (70.9%) were not inducible and 47 were inducible. Not-inducible individuals were older, had longer AP–AERP values compared with inducible ones, and only one had multiple APs. Patients who were not inducible at baseline EPT again showed noninducibility at follow-up study, suggesting that a single negative response is predictive of low risk for future arrhythmia induction with programmed stimulation. Of the noninducible group, 21 patients lost anterograde AP conduction, and 35 had retrograde AP conduction. Subjects who lost ventricular preexcitation were older and showed longer AERP of AP than patients in whom preexcitation persisted.
Predictors of arrhythmic events
Thirty-three patients (20%) became symptomatic during follow-up; they were younger and showed significantly shorter AERP of AP compared to patients who remained asymptomatic. Gender and location of AP were not able to predict arrhythmic events. Inducibility alone showed good sensitivity and specificity and higher positive predictive values than short AERP of AP in predicting the risk of symptomatic arrhythmias. Indeed, 17 of the 25 patients who became symptomatic for SVT were inducible for sustained AVRT, and all patients who became symptomatic for SVT were inducible for sustained AVRT, and all patients who became symptomatic for AF were inducible for sustained AF. In the majority of patients, AVRT was not inducible because of deficient retrograde AP conduction, which did not improve with isoproterenol-induced autonomic facilitation. Among the 115 noninducible patients, only 4 patients developed a sustained SVT, suggesting that a negative EPT is associated with a favorable prognosis.
Comparison with previous studies
Various rates of arrhythmic events have been reported for different follow-up periods, but no clinical or electrophysiological parameters have been found that could identify asymptomatic WPW subjects who are likely to become symptomatic (2,6–12). In a previous study of 75 asymptomatic WPW patients, only 6 (8%) patients became symptomatic and no patient died suddenly during a median follow-up of 4.3 years, and neither inducible sustained nor nonsustained AVRT predicted the development of subsequent symptomatic SVT (2). In contrast, our study performed in a large cohort of patients indicates that asymptomatic WPW patients have a significant risk of developing symptomatic and fatal events, and inducibility can predict which individuals will or will not become symptomatic.
Asymptomatic WPW and sudden death
It is very difficult to identify, within the large pool of WPW patients, the subset of asymptomatic subjects who are at risk for sudden death. A postmortem study concerning four patients with WPW and sudden death suggested that unpredictability of this event in such patients could be due to VF triggered by an atrial myocarditis leading to AF and then to VF (13). It is well known, however, that even in the absence of any underlying structural heart disease, VF can be triggered by sustained preexcited AF at extremely rapid ventricular rate, particularly in the presence of multiple APs. Multiple APs can be difficult to recognize, and they constitute a risk factor for VF in patients with the WPW syndrome (14). Both their true incidence and their clinical significance in this population have remained poorly investigated, because asymptomatic WPW subjects are usually evaluated by transesophageal pacing (15,16), and EPT has been performed only in a small number of subjects (2,8,9). In our large and lengthy follow-up EPT study, the overall incidence of multiple APs was 8%, and they were mainly found in younger patients. In addition, the three patients who experienced VF during the follow-up had multiple APs.
Management of asymptomatic WPW
The paucity of available data concerning parameters predictive of risk has suggested that indication to ablate asymptomatic subjects with APs needs to be individualized (4). Our study shows that in such patients EPT can be useful in risk stratification, and that both inducibility and the presence of multiple APs are associated with high risk of developing symptomatic and fatal arrhythmic events.
Although we followed patients for five years, this time period, however, is still relatively short with respect to individuals who have WPW potentially for a lifetime. The majority of our patients were in their 30s and 40s or older; therefore, the results could not be applicable to the pediatric population.
In the asymptomatic WPW population, a negative EPT with no AVRT or AF inducibility identifies subjects at very low risk for development of spontaneous arrhythmias. Inducibility of sustained preexcited AF with fast ventricular response, particularly in the presence of multiple APs, may help to select asymptomatic WPW subjects at definite risk for dying suddenly, and catheter ablation of APs appears as mandatory to prevent sudden death.
- anterograde effective refractory period
- atrial fibrillation
- accessory pathway
- atrioventricular reciprocating tachycardia
- confidence interval
- cycle length
- electrophysiologic testing
- shortest preexcited RR interval
- supraventricular tachycardia
- ventricular fibrillation
- Received May 9, 2002.
- Revision received July 24, 2002.
- Accepted August 19, 2002.
- American College of Cardiology Foundation
- Leitch J.W.,
- Klein G.J.,
- Yee R.
- Munger T.M.,
- Packer D.L.,
- Hammill S.C.
- Basso C.,
- Corrado D.,
- Rossi L.,
- Thiene G.
- Brembilla-Perrot B.,
- Ghawi R.