Author + information
- Received June 1, 1998
- Revision received May 18, 1999
- Accepted August 12, 1999
- Published online November 15, 1999.
- Philippe Chevalier, MD, PhD∗,
- Laurence Bontemps, PhD†,
- Marjaneh Fatemi, MD∗,
- Stephane Velon, MD∗,
- Eric Bonnefoy, MD∗,
- Gilbert Kirkorian, MD∗,
- Roland Itti, MD† and
- Paul Touboul, MD∗,*
- ↵*Reprint requests and correspondence: Dr. Paul Touboul, Hôpital Cardiovasculaire et Pneumologique Louis Pradel, BP Lyon-Montchat 69394 Lyon, Cedex 03, France
This study was designed to prospectively evaluate the effects of radiofrequency ablation in Wolff-Parkinson-White (WPW) syndrome by scintigraphic analysis.
The functional changes triggered by radiofrequency current ablation of atrioventricular accessory pathways are not fully known.
Forty-four patients with WPW syndrome were consecutively investigated before and 48 h after radiofrequency therapy. Fourteen patients had right sided atrioventricular pathways and 30 patients had left sided bypass-tracts. Planar gated imaging and gated blood pool tomography were performed in all of these patients.
A significant increase in the left ventricular ejection fraction (LVEF) was demonstrated in patients with left preexcitation (62.2 ± 7.9% before ablation against 64.4 ± 6.3% after ablation, p = 0.02) but not for those with right sided anomalous pathway. Phase analysis only gave significant differences following ablation of right sided pathways (left-to-right phase difference = 14.4 ± 13.8° before ablation versus 7.5 ± 7.2° after ablation, p < 0.05). Early abnormal ventricular contraction persisted in 12 patients with right accessory pathways and in 8 patients with left accessory pathways despite the complete disappearance of any abnormal conduction as proven electrophysiologically.
Following catheter ablation of atrioventricular accessory pathways: 1) an improvement of left ventricular function may be seen, particularly in patients with left sided accessory pathways, and 2) unexpected persistence of local ventricular preexcitation at the site of successful ablation may be detected.
For over 15 years radionuclide ventriculography has been used to localize the atrioventricular accessory pathways associated with the Wolff-Parkinson-White (WPW) syndrome (1–14). Analysis of the timing of the myocardial contraction based on Fourier phase histograms allows mapping of the kinetics of the different ventricular segments from functional phase pictures, and by extrapolation, mapping of the electrical activity of the heart. So far, studies in this field have only included a limited number of patients. Moreover, the phenomenon of successful radiofrequency ablation accessory pathways is not fully understood, and the exact mechanisms leading to disappearance of preexcitation still have not been worked out.
Using scintigraphic imaging techniques, this study was undertaken to assess the effects of atrioventricular accessory pathways radiofrequency ablation. The areas that were particularly examined for study were as follows: 1) the respective functions of the preexcited and controlateral ventricles quantitated by the measurement of the ejection fraction, 2) the overall preexcitation of one cavity compared with the other, as estimated from the mean values of phase histograms constructed separately for each ventricle, and 3) the evolution of the early ventricular contraction zone at the insertion of the treated accessory pathway.
The study was prospective and patient consent was obtained. Forty-four patients with overt preexcitation on the surface ECG were included (mean age of 40 ± 13 years, 13 women and 31 men). Eighty-eight tomograms were performed, each patient had this examination twice, before and within 48 h after radiofrequency ablation. Prior ablation, the localization of the accessory bundle by tomoscintigraphy was compared with the results of electrophysiological studies, these being considered as the method of reference. Depending on the electrophysiological localization of the accessory pathways, the patients were split into two groups: 14 patients with right sided connections (eight men and six women with a mean age of 33.6 ± 14.3 yr) and 30 patients with a left sided accessory conduction pathway (22 men and 8 women aged 42.5 ± 12.3 yr). Radiofrequency ablation was considered successful in all the patients. Those subjects with any abnormal electrocardiographic or electrophysiological findings following catheter ablation were excluded.
An identical radioisotope examination was performed in eight normal young individuals aged 22 ± 3 years to provide the reference values of the ventricular parameters that were measured during the study.
Peripheral intravenous administration of technetium 99 m with an activity of 700 to 800 MBq was preceded by an injection of stannous pyrophosphate to allow in vivo labelling of the red blood cells. The examination was then performed in two phases.
Standard gamma angiocardiography
This examination was performed first because the laboratory references for cavity volumes and ejection fraction had been obtained earlier with planar imaging. The measurements were made with electrocardiographic synchronization from the left anterior oblique view, with a caudocranial tilt of the detector axis of 10°. Each heart cycle was divided into 16 images, with a matrix of 64 × 64 pixels. Semiautomatic analysis of the acquired data (ECCAP software) allowed the ejection fractions to be calculated and the wall kinetics to be viewed.
A Sopha DS7 gamma camera, equipped with a high resolution collimation with parallel holes, was used. For image processing a Sophy NXT computer system was used. Thirty-two projections were acquired with one projection every 30 s, while the detection head was turned through 180°, from the left posterior oblique to the right anterior oblique. These pictures were synchronized by the electrocardiogram and each heart cycle was divided into 16 segments. Transverse slices were reconstructed, using a filtered back-projection algorithm. Only the planes of the slices, which included the right and left ventricles simultaneously (horizontal long axis and short axis slices) were kept for analysis, which was based on comparison of the contraction of the two ventricles. A protocol for analysis of the radioisotope ventriculography, similar to that used in planar imaging, allowed the functional variables of the right and left ventricles to be evaluated and provided a phase image, with separate quantitative analysis for each ventricle in the form of phase histograms using the first harmonic of the Fourier transform. To represent the contraction time for all of the pixels of the image during one cardiac cycle, a color scale gave each point a color corresponding to the value of its phase. Also, histograms representing the distribution of the pixels for each ventricle according to their phases allowed the mean phase and its standard deviation to be calculated. The postablation sites of early contraction were only taken into consideration if they had the same location as the preablation sites and if they corresponded to a cluster of pixels and not to isolated pixels like noise.
All the planar and tomographic data were used in the analysis of seven quantitative variables for each phase of the investigation (before and after ablation): the ejection fraction of the left ventricle (LVEF), the ejection fraction of the right ventricle (RVEF), the mean phase of the left ventricle (LVMP), the mean phase of the right ventricle (RVMP), the difference of the mean phases of the left and right ventricles (L-RMP), the standard deviation of the mean left ventricular phases (LVPSD) and the standard deviation of the mean right ventricular phases (RVPSD).
The electrophysiological tests were performed while fasting and after stopping all antiarrhythmic drugs. All of the ablation sessions were performed under sedative treatment (midazolam ± phenoperidine) and anticoagulation (intravenous injection of 2,000 U of Heparin every 2 h). A baseline electrophysiological study was systematically performed in order to define the properties and the location of the accessory pathway, as well as the characteristics of any tachycardia that could be induced. The radiofrequency current was applied by means of a steerable Mansfield-Webster catheter by using a large distal electrode (4 mm) and a HAT 300 generator (Osypka) connected to an IBM computer. The ablation procedure was considered successful if anterograde and retrograde conduction of the accessory pathway was completely abolished, this being associated with the inability to induce atrioventricular reciprocating tachycardias.
After the ablation procedure all the patients were discharged without any antiarrhythmic drugs. A repeat electrophysiologic study was performed one to two months later and never showed resumption of abnormal conduction. None of the patients had a recurrence of atrioventricular tachycardia over a mean follow-up period of 17 ± 9 months.
The quantitative results are expressed as mean value ± standard deviation. Comparison of the data obtained before and after treatment in the same series of patients was made using the Student ttest, adapted to paired series. A p value <0.05 was considered as statistically significant.
The values measured in the group of normal individuals were as follows: LVEF: 62% ± 6%; RVEF: 44% ± 6%; LVMP: 134 ± 14°; RVMP: 131 ± 18°; L-RMP: 8 ± 4°; LVPSD: 9 ± 2°; RVPSD: 18 ± 3°.
Effects of ablation on the ventricular ejection fraction (Tables 1 and 2)
In the patients with right sided accessory pathways, no significant change of the RVEF was seen after treatment (36 ± 10% vs. 40 ± 9%, p = 0.16). However, in three cases the RVEF markedly increased by 10%, 23% and 11%, respectively (Patients 5, 7 and 8). The LVEF remained unaltered (60 ± 9 vs. 59 ± 9, p = 0.17).
After destruction of the left sided pathways, a modest but significant increase in the mean LVEF was noticed (from 62.2 ± 7.9% to 64.4 ± 6.3%, p = 0.02). The two patients with the greatest improvement in the LVEF showed an increase by 16% and 10% (Patients 14 and 16). There was no significant change in the RVEF following left accessory pathways ablation.
Effects of ablation on phase analysis (Tables 1 and 2)
In patients with right sided accessory pathways there was no difference in the LV phases and the RV phases before and after ablation (144.4 ± 21.6° vs. 143.4 ± 18.9°, NS and 132.3 ± 17.7° vs. 142.3 ± 13°, NS, respectively). As for the differences between the left and right values seen in each patient in the control state and following ablation, there was a statistically different decrease from 14.4 ± 13.8° to 7.5 ± 7.2° (p = 0.03). After right sided ablation, the standard deviations of the phases were unchanged in the left ventricle (LVPSD: 14.5 ± 5.9° versus 12.1 ± 2.7°, NS). The postablation reduction in the phase dispersion of the right ventricle was statistically significant (26.5 ± 11.4° vs. 19 ± 3.2°, p = 0.03).
Following ablation of left sided accessory pathways, the patients showed no significant changes in the LV phases (142.5 ± 16.9° vs. 143.0 ± 14.6°, NS) or the RV phases (144.7 ± 18.2° against 141.3 ± 14.7°, NS). Similarly, the left-right phases differences were unaltered (7.0 ± 9.1° before treatment and 7.5 ± 6.1° after, NS). Regarding the standard deviations for phase, there was no statistically significant differences before and after ablation for the mean LVPSD (13.3 ± 3.1° against 13.0 ± 2.4°, NS) and for the mean RVPSD (20.6 ± 8.4 against 21.0 ± 8.3, NS).
Comparison of scintigraphy and electrophysiological testing for the localization of accessory pathways (Tables 1 and 2)
In right sided pathways, gamma angiocardiography recognized the right location of the preexcitation in 86% of cases (12 of 14 cases) and found the exact insertion site in 64% of the cases (9 of 14 cases). For left sided pathways, the determination of the side was less accurate and was correctly found in 63% of the cases (19/30). The rate of concordance with the electrophysiological testing for locating the ventricular insertion zone was almost the same as in right sided preexcitation, that is 60% (18/30).
Surprisingly, the persistence of a zone of premature contraction was noted after ablation in 12 patients with right sided accessory pathways and in 8 with left accessory pathways. This abnormal preexcitation was restricted to the ventricular insertion of the accessory pathway and was not propagated (Fig 1 and 2)⇓. ⇓However, in all these patients, the surface ECG had returned to normal and no sign of preexcitation had been detected in the postprocedure electrophysiological study.
In this study involving WPW syndrome patients, scintigraphy was used to detect the changes after radiofrequency ablation. First of all, the concordance of scintigraphy and endocardial mapping for localizing accessory pathways was broadly confirmed. Secondly, the ipsilateral ventricular ejection fraction could be improved following radiofrequency ablation particularly in patients with left sided accessory pathways. However, the main finding was that, despite successful ablation of the accessory pathways, unexpected persistence of local ventricular preexcitation was occasionally unveiled by scintigraphy.
Most scintigraphic studies, which have so far been performed to assess preexcitation syndromes, have used planar imaging and have included a limited number of patients. In a study of 14 patients with right sided accessory pathways, Aliot et al. (6)found agreement between gamma angiocardiography and electrophysiological tests in 12 patients (85%). On the other hand in five patients with left sided WPW syndrome, gamma angiocardiography was negative. After atrial stimulation, the diagnosis of three of these five patients was proven on phase analysis. A more recent study by Santelli et al. (10)included five patients with left sided WPW syndrome, three of whom were recognized by using phase analysis after baseline planar imaging. Atrial stimulation by the transoesophageal route managed to unmask a fourth case of preexcitation. Silka et al. (11)studied 20 WPW patients with reference to the accessory pathway location and achieved a similar sucess rate (85%). The report by Nakajima et al. (7)dealing with tomography to localize the WPW syndrome is of interest. These authors examined each of their 14 WPW syndrome cases (eight right and six left sided) by planar ventriculography and then by tomography. In 57% of patients (8/14), the accessory pathway was found by the first technique, while tomography provided the right location in 86% of cases (12/14). In fact, tomography allows the extraction of the sites which would have been hidden in a plane projection imaging technique due to superimposition (9).
The increase in the ipsilateral ventricular ejection fraction after accessory pathway ablation is noteworthy. One can assume that some degree of myocardial depression may be ascribed to the development of ventricular asynchrony. In this view, the suppression of cardiac preexcitation should improve cardiac function mainly in the ipsilateral ventricle where the anomalous process is more prominent. However, in our study, the postablation changes were less apparent in the subjects with right sided accessory pathways despite the fact that major preexcitation is usually seen in this setting. The lesser amplitude of the right ventricular contraction may render any therapeutic effect more difficult to unmask. Interestingly, phase analysis identified significant changes in right ventricular contraction after radiofrequency ablation of right sided accessory pathways in accord with the pathophysiologic expectations.
The persistence of a zone of premature ventricular contraction as shown by scintigraphy, despite the disappearance of the delta wave on the surface electrocardiogram, has never been described. Such a phenomenon may indicate a persisting aborted conduction via the accessory pathway. One must postulate that the atrial insertion of the accessory connection is, at least partly, undamaged. On the other hand, it is apparent that the site of anterograde block as a result of ablation occurred near the ventricular interface.
To explain the conduction changes in the accessory pathways, the impedance mismatch hypothesis has been invoked. The role of this source-sink interaction has been mainly discussed in vitro. Mendez et al. (15)proposed a funnel-like system to explain that the safety margin for impulse propagation was different according to whether the impulse proceded from the myocardium to Purkinje fibers or in the opposite direction. One year after, in a model of preexcitation syndrome using atrial tissue where a narrow isthmus was created, De la Fuente et al. (16)demonstrated that conduction block occurred in any location where the source was much smaller than the sink. More recently, Cabo et al. (17)tested the funnel hypothesis by creating narrow paths in specimen of sheep epicardial muscle.
In humans, there is only one study that suggests that the source-sink interaction may be relevant in clinical arrhythmias. Ong et al. (18)used high-density endocardial mapping in eight patients undergoing surgical ablation of their accessory pathway. They noticed that during atrial fibrillation, collision or fractionation of the local wavefronts was likely to cause block in accessory pathways, whereas more organized and uniform atrial activation patterns favored conduction through the accessory pathways.
Regarding our data following destruction of accessory fibers proximally, the amount of current that reaches the relevant ventricular area may be dramatically diminished and thus rendered ineffective. Interestingly, factors limiting anterograde conduction in concealed accessory pathways, have been shown to be usually located at the interface of the accessory pathway and the ventricle (19). Finally, it is noteworthy that none of the patients with this aborted persisting preexcitation had a recurrence of tachycardia in the midterm, suggesting that such a phenomenon has no clinical relevance and is unable to set off tachycardias.
Tomography has certain limitations, which are either technical or linked to the electrophysiology of the WPW syndrome itself. The main technical drawback is a poor spatial resolution, which is currently rarely better than 10 to 15 mm. Also, detection of abnormalities in the septal and paraseptal regions is made difficult by the fact that these are areas with low amplitude contraction (7). Additionally, it may be arduous to identify the cause of variations in the heart with reference to the body morphology in certain patients. The poor diagnostic capability of scintigraphy in detecting left-sided accessory pathways is probably due to the fact that, in this situation, the distance between the sinus node and the site of the bundle of Kent is larger than with right sided accessory pathways. Ventricular preexcitation is, therefore, favored in the presence of a right accessory connection. This explains the results of phase analysis (especially the values of the phase differences) obtained in this study. In this view pharmacological sensitization techniques, using adenosine for example, might be helpful to accentuate the appearance of preexcitation. The same reservation also applies to the study of the effects of catheter ablation. In left sided accessory pathways, phase analysis did not show any difference after radiofrequency current ablation. On the other hand, the patients with right sided preexcitation exhibited a significant reduction in the phase dispersion of the right ventricle and thus in the heterogeneity of the ventricular contraction after treatment.
The identification of preexcitation zones in WPW syndromes can be better appreciated with tomography than with planar gamma angiocardiography. Results of tomography performed for the location of abnormal zones are well correlated with those of electrophysiological testing. Following ablation, both LVEF and RVEF can improve, due to the resumption of fast homogeneous ventricular activation. Phase histograms for right-sided accessory pathways showed, after ablation, a definite improvement involving both the overall synchronization of the two ventricles and the internal asynchrony of the right ventricle. A similar effect was not apparent in cases of left-sided accessory pathways, probably due to a lesser degree of ventricular preexcitation in the basal state. Finally, the persistence of premature activation at the preexcited site after successful ablation suggests that anterograde conduction block may be located near the ventricular interface.
- difference of the mean phases of the left and right ventricles
- left ventricular ejection fraction
- mean phase of the left ventricle
- standard deviation of the mean left ventricular phases
- right ventricular ejection fraction
- mean phase of the right ventricle
- standard deviation of the mean right ventricular phases
- Received June 1, 1998.
- Revision received May 18, 1999.
- Accepted August 12, 1999.
- American College of Cardiology
- Links J.M.,
- Douglass K.H.,
- Wagner H.N.
- Turner A.,
- Behren L.V.,
- Ruggie N.T.,
- et al.
- Chan W.C.,
- Calff V.,
- McDonald D.,
- et al.
- Botvinick E.,
- Frais M.,
- O’Connell W.,
- et al.
- Nakajima K.,
- Bunko H.,
- Tada A.,
- et al.
- Aliot E.,
- Laurens M.H.,
- Thouvenot P.,
- et al.
- Itti R.,
- Philippe L.,
- Casset-Senon D.,
- Bontemps L.,
- Cosnay P.,
- Fauchier J.P.
- Santinelli V.,
- Fazio S.,
- Turco P.,
- Depaoza M.,
- Santomauro M.,
- Chiariello M.
- Clausen M.,
- Weismuller P.,
- Weller R.,
- Adam W.E.,
- Henze E.
- Hombach V.,
- Kochs M.,
- Weismuller P.,
- et al.
- ↵Mendez C, Mueller WJ, Urguiaga X. Propagation of impulses across the Purkinje fiber-muscle junctions in the dog heart. Circ Res 26:135–50.
- Fuente D.,
- Sasyniuk B.,
- Moe G.K.
- Cabo C.,
- Pertsov A.M.,
- Baxter J.M.,
- Davidenko J.M.,
- Gray A.,
- Jalife J.
- Kuck K.H.,
- Friday K.,
- Kunze K.P.,
- Schlüter M.,
- Lazzara R.,
- Jackman W.M.