Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Computerized high-density mapping of the pulmonary veins: new insights into their electrical activation in patients with atrial fibrillation

Computerized high-density mapping of the pulmonary veins: new insights into their electrical... Europace (2004) 6,97e108 Computerized high-density mapping of the pulmonary veins: new insights into their electrical activation in patients with atrial fibrillation a, b a Roberto De Ponti , Massimo Tritto , Marcelo E. Lanzotti , b a b Giammario Spadacini , Raffaella Marazzi , Paolo Moretti , Jorge A. Salerno-Uriarte University of Insubria, Department of Cardiovascular Sciences ‘‘Ospedale di Circolo e Fondazione Macchi’’ Varese, Italy Department of Cardiology ‘‘Mater Domini’’, Castellanza, Italy Submitted 25 March 2003, and accepted after revision 9 November 2003 KEYWORDS Abstract Aim To report the method and findings of computerized high-density high-density mapping of pulmonary veins (PVs) in patients undergoing their electrical isolation mapping; for atrial fibrillation (AF). pulmonary veins; Methods and results In 17 consecutive patients (8 M, age 55G11 years), a 64 elec- atrial fibrillation trode basket catheter was placed in the target PVs and 56 bipolar electrograms were recorded, analyzed and isochronal maps were generated. PVs were mapped during sinus rhythm, left-sided pacing and ectopic activity. The sites of earliest activation at the veno-atrial junction were defined as the atrium to vein conduction break- throughs. PV activation pattern was classified as predominantly longitudinal or transverse, according to the direction of the impulse from the breakthroughs. The ectopic pattern was defined as multifocal, when distant areas in the PV had activa- tion times within 10 ms. Thirty-one PVs were mapped. The activation pattern was predominantly longitudinal in 13 PVs and transverse in 18 PVs. Two breakthroughs were identified in 22 PVs and three in nine. All the breakthroughs were evident si- multaneously in sinus rhythm and left-sided pacing changed only the predominance of the breakthrough. Ectopies were mapped in 10 PVs: eight showed a multifocal and two a monofocal pattern; six ectopies originated from the proximal tract of the PV. Conclusion High-density mapping of PV identifies a typical activation pattern. Multiple and discrete breakthroughs are simultaneously identified in sinus rhythm. The majority of the mapped ectopies has a multifocal pattern and proximal origin. ª 2003 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. Corresponding author. Tel.: D39-0332-278394; fax: D39-0332-393309. E-mail address: [email protected] (R. De Ponti). 1099-5129/$30 ª 2003 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.eupc.2003.11.004 98 R. De Ponti et al. physiological procedure, all patients were on oral Introduction anticoagulants, which were withdrawn and substi- tuted by heparin 24 h before the procedure. A In patients with atrial fibrillation (AF), the pulmon- preprocedure transoesophageal echocardiogram ary veins (PVs) are considered a critical area for revealed no thrombus in the left atrium and normal arrhythmogenesis and therefore they represent blood flow velocity in the left atrial appendage. All a target for ablation [1,2]. Over the last years, sev- antiarrhythmic drugs were discontinued one week eral reports [3e7] focussed on the anatomy and the before the procedure; no patient was receiving histology of the myocardial extensions surrounding amiodarone. the PVs, in patients with or without AF, providing detailed description of their complex morphology, where fibrous and fatty tissue are interspersed Electrophysiological procedure and data among the myocardial fibres, forming a mesh-like acquisition arrangement [5,6]. On the other hand, little is known about the electrophysiological properties All patients underwent the electrophysiological and electrical activation of the myocardial tissue procedure after having signed the consent form surrounding the PVs. Both in the past [8,9] and re- for the study, approved by the Institutional Ethical cently [10,11], limited data have been gathered, Committee. Double transseptal puncture was per- which provide evidence of heterogeneous and dec- formed in all cases. Two long sheaths were, then, remental conduction properties [10] and of shorter positioned in the left atrium: an 8F 62 cm long effective refractory period [11] of these myocardial sheath (Preface, Biosense Webster, Inc., USA) was extensions, compared with the proper left atrial used to introduce the angiographic catheter or, musculature. High-density mapping of the PVs may alternatively, the ablation catheter in the left contribute to increase our knowledge of their func- atrium, whereas a 9F 60 cm long sheath (Boston Sci- tional aspects and electrical activation. This may entific Corporation, USA) with different curves (55, be useful not only precisely to locate the break- 90 or 120() was used to position the mapping cathe- throughs of atrium to vein conduction, an essential ter. After transseptal puncture, intravenous hepa- prerequisite for their ablation in patients with AF rin was administered to maintain the activated [12], but also to identify patterns of electrical acti- clotting time between 250 and 300 s. A 64 elec- vation during different heart rhythms and, espe- trode 31 mm diameter basket catheter (Constella- cially, during ectopic activity triggering AF. tion, Boston Scientific Corporation, USA) was used The aim of this study was to define by computer- to map the PVs; in this catheter eight electrodes ized high-density mapping the electrical activation are uniformly distributed on each of the eight of the PVs during sinus rhythm, pacing from differ- splines. In every patient, after venography of all ent left atrial sites and ectopic beats, in patients PVs, every effort was made to identify the arrhyth- with AF. mogenic PV(s). For this purpose, if the patient was in AF, repeated cardioversions during temporary sedation with propofol were performed. If the pa- tient was in sinus rhythm, prolonged observation Methods and provocative manoeuvers were used to detect spontaneous ectopic activity. Once the arrhythmo- Patient population genic PVs were identified, they were targeted for electrical isolation. In case no ectopic activity Seventeen consecutive patients (8 M, 9 F; mean age was observed, the superior PVs were isolated. In 55G11 years) with paroxysmal, frequently recur- every case, the basket catheter was deployed in rent AF, undergoing an electrophysiological proce- the target vein (Fig. 1A) by withdrawing the long dure for electrical isolation of the PVs, have been sheath, previously advanced in the PV over a included. Patients who previously underwent a PV 0.035 in. guidewire. After deployment, minimal ro- isolation procedure were excluded from the study. tations of the basket helped to distribute uniformly In all patients, repeated 12-lead Holter monitoring the splines of the basket inside the vein and opti- showed that the clinical palpitation episodes were mize the electrode/tissue contact. A second veno- related to AF, reproducibly initiated by atrial ecto- graphy (Fig. 1B) was then performed to assess the pies. All patients were refractory to 2:5G1:3 anti- position of the proximal electrodes at or very close arrhythmic drugs, previously tested. No patient to the os of the vein. By using two fluoroscopic pro- had structural heart disease, as assessed by echo- jections (Fig. 2), the position of the ‘‘A’’ and ‘‘B’’ cardiogram. For at least 45 days before the electro- splines (identified by radiopaque markers) in the Mapping of pulmonary veins 99 Figure 1 (A, B) Fluoroscopic image of the basket catheter position in a right superior pulmonary vein in antero- posterior view. In panel A, the basket and the angiographic catheters are both positioned in the vein through transseptal catheterization. It is noteworthy that the adaptable geometry of the basket fits the anatomy of the pulmonary vein. The arrows indicate the radiopaque markers (very similar to regular electrodes), which identify spline ‘‘A’’ and ‘‘B’’. When dye is injected (panel B), satisfactory electrode to vein contact is assessed. Data analysis and definitions vein was recorded and classified according to the clock quadrant, where 12 and 3 h are superior and anterior, respectively, for the left PVs, while Each segment acquired by the QMS2 system was an- they are superior and posterior, respectively, for alyzed in order to (1) define the pattern of electri- the right PVs. Through a junction box, the bipolar cal activation in the PV, (2) identify the sites of the signals from the basket catheter were simulta- breakthroughs of atrium to vein conduction, and neously recorded on a computerized mapping sys- (3) evaluate the pattern and site of earliest activa- tem (QMS2 System, Boston Scientific Corporation, tion during the ectopic activity. On the system, USA) for data analysis and on the conventional each electrogram was manually annotated to de- electrophysiology system for on-line continuous termine the local activation time, according to monitoring. In the QMS2 system, 56 bipolar electro- the following criteria. grams were recorded from seven contiguous longi- The analysis of the activation pattern of the PV tudinal bipoles on each spline. Signals were during sinus rhythm or pacing was performed at processed with filter and gain settings of a sweep speed of 200 mm/s. The local activation 30e290 Hz and 8e64, respectively. For each map- time was measured at the onset of each PV poten- ped PV, 30 s recordings were made in three differ- tial and the activation time of the earliest acti- ent rhythms: sinus rhythm, pacing from distal vated PV potential was considered as time 0. The coronary sinus and left atrial appendage for the left following two criteria were used to discriminate PVs and sinus rhythm, pacing from proximal coro- PV potentials from atrial electrograms, which may nary sinus and roof of the left atrium for the right be recorded as a far field activity even in the distal PVs. For the evaluation of the ectopic activity, only electrode pairs. The first criterion was the chronol- cases, in whom at least three recordings of the ogy of the PV potential from the surface P wave on- ectopic activity in the same vein were acquired, set: no signal was considered as a PV potential if were considered. In all cases, recordings of the ec- earlier than 75 and 35 ms in sinus rhythm and left topic activity were performed when no catheter atrial pacing, respectively, for the left PVs and 40 other than the basket was inside the vein. and 60 ms, in sinus rhythm/left atrial roof pacing 100 R. De Ponti et al. Figure 2 (A, B) Fluoroscopic image of the basket catheter position in the same vein and patient, in 30( right anterior oblique view, before (panel A) and during (panel B) venography. An oblique fluoroscopic projection allows further assessment of catheter positioning and contact inside the vein, the course of which is not rectilinear, but shows curves in different planes. A rather homogeneous distribution of the splines around the vein perimeter is observed and the position of spline A (indicated by the arrow and identified by a single radiopaque marker) is evident at 5 o’clock. and proximal coronary sinus pacing, respectively, visualized in the centre of the map, whereas the for the right PVs. These time intervals were deter- proximal electrodes are along the outer perimeter. mined on the basis of a previous activation study The position of spline ‘‘A’’ is identified on the map [13]. Second, for each spline, the discrimination by larger dots and the other splines are distributed of the atrial from PV potentials was done first in in a counter-clockwise fashion. For each map, the the proximal electrograms, where the onset of position of the spline ‘‘A’’ was adjusted according the PV potentials was annotated. Each subsequent to its position on the clock quadrant as assessed by annotation in a more distal dipole was made at the fluoroscopy. On each map, during sinus rhythm or onset of the first deflection later than the previous pacing a breakthrough of atrium to vein conduction annotation in the proximal dipole on the same was defined as the site of earliest activation along spline. Examples may be seen in Figs. 3 and 4.In the outer perimeter of the map. Comparison be- case satisfactory discrimination between atrial tween the site of the breakthrough and the site and vein potentials was not possible, the recording where ablation modified and abolished PV activity was not considered for analysis. was made for each breakthrough in every patient. During the ectopic activity, in order to evaluate The pattern of electrical activation inside the vein the site of earliest activation, each electrogram vi- was defined as predominantly longitudinal, if fast sualized at the maximum gain value was annotated propagation was observed from the breakthrough at the onset of the first sharp deflection. To avoid to the distal part of the vein with a direction longitu- the bias of artefact and misinterpretation of the dinal to the long axis of the vein. On the isochronal signals, the analysis was repeated for three ectopic map, red to yellow colours are distributed on the beats from each vein and comparison with the elec- same spline. Conversely, the propagation pattern trograms of a sinus beat not followed by concealed inside the vein was defined as predominantly trans- or manifest ectopy was made. verse, when the propagation proceeds from the After the annotation process was complete, a po- breakthrough with a direction perpendicular to lar isochronal map was generated by the system, the long axis of the vein. In this case, the red to yel- where red and blue indicate early and late activa- low colours are distributed in the proximal electro- tion, respectively, and the distal electrodes are des of adjacent splines. In the analysis of the site of Mapping of pulmonary veins 101 Figure 3 Isochronal polar map of a left superior pulmonary vein and display of the electrograms recorded in the spline C, E and F, during coronary sinus pacing. In the map, as well as in the following figures, the inner circle identifies the distal basket electrodes, whereas the outer circle corresponds to the proximal electrodes. Spline ‘‘A’’ (bigger dots) is located at 12 o’clock; red and purple identify earliest and latest activation, respectively, of the pulmonary vein potentials and the activation values are shown on the scale. The electrograms are displayed from top to bottom from the distal to the proximal electrode pair of the spline; the red line corresponds to the stimulation artefact preceding the surface P wave (yellow line on the top) and for each electrogram the number indicates the activation time. In this case, three breakthroughs, located at 9e10, 2 and 5 o’clock, respectively, are present, the inferior breakthroughs prevailing during coronary sinus stimulation. A predominantly longitudinal activation pattern is observed in the vein, since the activation proceeds fast in a longitudinal direction, from the breakthroughs along the corresponding splines, as assessed also by the activation values in splines ‘‘C’’ and ‘‘F’’. In a later activated area corresponding to spline ‘‘E’’, transverse conduction is observed, as assessed by very similar activation values from proximal to distal electrode pairs. earliest activation during ectopies, the pattern was in Table 1. The number of PVs mapped per patient defined as multifocal if activation times of different was 2G0:7. In two patients the left PVs were map- sites in non-adjacent splines were within 10 ms. All ped and isolated at their common os and, for the the recordings and maps were independently eval- purpose of this study, they have been considered uated by two electrophysiologists. as a single vein. In no case, was ectopy arising from the right inferior PV documented and, according to the protocol, this vein was not considered for abla- Statistical analysis tion. Ninety-three acquisitions were made during sinus rhythm, coronary sinus and left atrial stimula- Continuous variables are expressed as meanG 1 SD. tion, whereas the remaining 30 acquisitions (three Considering the limited value of finding statistically repeated acquisitions in 10 different veins) were significant differences in such a restricted num- performed during ectopies. Sixty-seven of 93 ber of patients and veins as those of the present (72%) acquisitions during sinus rhythm or pacing study, statistical analysis was not performed for could be analyzed and an isochronal map gener- the activation times and the distribution of activa- ated; the remaining 26 (28%) acquisitions per- tion patterns. formed during pacing were not considered for analysis, because the atrial and PV potentials could Results not be clearly discriminated. Anyhow, it is impor- tant to highlight that all the 31 veins could be eval- Data acquisition and analysis uated in sinus rhythm. Table 1 reports the percent of the recordings that could be analyzed in differ- A total of 123 acquisitions was made by the ent rhythms out of the overall number of acquisi- QMS2 system in 31 PVs, whose location is shown tions made for each vein. As also shown in Table 1, 102 R. De Ponti et al. Figure 4 Isochronal polar map of a right superior pulmonary vein and display of the electrograms recorded in the spline C, E and G during sinus rhythm. The map and the splines are as in the previous figure. In this case, the activation in the veins shows a predominant transverse activation pattern, since the propagation proceeds from the two breakthroughs at 10e12 and 3e5 o’clock following a direction perpendicular to the long axis of the vein. Consequently, the redeorange colour spreads from the breakthrough to the proximal dipoles of the adjacent splines, rather than proceeding longitudinally along spline ‘‘C’’ and ‘‘G’’, as in the case shown in the previous figure. In fact, there is a jump of 26 and 10 ms in the conduction along spline ‘‘C’’ and ‘‘G’’, respectively, between the third and the fourth proximal dipole. A later activated spline (‘‘E’’) shows as well a transverse pattern, with very similar activation values from proximal to distal dipoles. 7/31 (22.5%) veins were evaluated in a single artefacts and have been annotated. Concordance (sinus) rhythm, the vast majority of them (5/7 between the two independent observers was veins) being right superior PVs. In only two cases, obtained in 97% of the maps. Thirty of 31 PVs the basket catheter displaced minimally during were completely isolated by low power radio- the procedure, with a clockwise rotation of 15 frequency energy (14 patients) or cryothermal and 30(, respectively, which, in no case, rendered energy (three patients) delivery, whereas in one the evaluation of the map impossible. Overall, in vein a 90% reduction of the amplitude of the the 97 analyzed acquisitions a total of 5328 PV potentials was obtained. No complication was electrograms (96.4%) were free from technical observed during or after the procedure. Table 1 Number of veins mapped and their features RSPV LSPV LIPV Common os of left PVs No. of veins 12 12 5 2 No. of veins mapped in a single rhythm 5 1 1 e Percent of recordings analyzed during Sinus rhythm 100% 100% 100% 100% Coronary sinus pacing 50% 93% 80% 100% Left atrial stimulation 42% 43% 40% 50% No. of veins with a predominant Longitudinal activation pattern 4 6 3 e Transverse activation pattern 8 6 2 2 No. of veins with Two breakthroughs 8 8 4 2 Three breakthroughs 4 4 1 e Abbreviations: LIPV[ left inferior PV; LSPV[ left superior PV; RSPV[ right superior PV. Mapping of pulmonary veins 103 Activation patterns in the pulmonary veins Localization of atrium to vein conduction breakthroughs As shown in Table 1, a predominant longitudinal ac- tivation pattern (Fig. 3) was observed in 13 veins, In all the 31 veins, more than one breakthrough of while in the remaining 18 veins it was transverse atrium to vein conduction was observed and in no (Fig. 4), with a trend towards a more frequent case were more than three breakthroughs identi- transverse pattern in the right superior PV. Also in fied. The number of breakthroughs in the different veins classified as having a predominant longitudi- PVs is shown in Table 1. The location of the break- nal activation pattern, limited areas of transverse throughs was ubiquitous in the clock quadrants in conduction were present, as shown in Fig. 3.Inno mapped veins. In 24 PVs analyzed in more than case, in whom the PV was evaluated in more than one rhythm, the position of the breakthrough was one rhythm, was a rhythm-dependent change in not rhythm-dependent and, importantly, no addi- the activation pattern observed. Interestingly, tional breakthrough became evident by pacing among the 13 patients, in whom more than one vein at different sites. Nevertheless, a change in the was mapped, nine showed the same activation pat- predominance of a breakthrough depending on tern in all mapped veins, which was transverse in the change of the rhythm was observed in every five and longitudinal in four. During sinus rhythm, case, the breakthrough closer to the site of origin the time interval between the earliest and latest of the rhythm becoming predominant. Fig. 5 shows activated PV potential was 40G19 ms in the left su- an example of this finding. In all cases, in whom perior PV, 36G19 ms in the right superior, 31G8ms both the left superior and left inferior PVs in the left inferior and 49G27 ms in the common os were separately mapped, a strict anatomical rela- of left PVs. In sinus rhythm, the activation time of tionship was observed between the inferior veins with a predominantly longitudinal pattern breakthrough of the superior vein (identified at was very similar to one of the veins with a predom- 6e7 o’clock) and the superior breakthrough of the inantly transverse pattern (37G9ms vs 35G8 ms). inferior vein, localized between 11 and 1 o’clock Figure 5 Isochronal polar map of a left inferior pulmonary vein in sinus rhythm and pacing. In sinus rhythm (panel A) the two breakthroughs, located at 11e12 and 5e6 o’clock, respectively, are equally predominant and the propagation inside the vein proceeds longitudinally from both breakthroughs with similar activation times. During left atrial appendage (panel B) and coronary sinus (panel C) stimulation, the breakthrough closer to the pacing site becomes more evident than the opposite one and the activation times inside the vein modify accordingly, although the longitudinal activation pattern of the vein does not change. 104 R. De Ponti et al. (Fig. 6). Finally, in 29 out of 31 veins there was co- vein showed a reproducible activation pattern with incidence of the location of the splines identifying no significant change among the three subsequent the atrium to vein conduction breakthroughs and acquisitions. In eight veins, ectopies showed a mul- the site where energy application resulted in mod- tifocal activation pattern with areas of early simul- ification and abolition of the PV potentials. In two taneous activation widely distributed in the vein, PVs, this validation was rendered impossible by as paradigmatically shown in Fig. 7. Only in two a rotation of the basket catheter after the mapping veins in the same patient, ectopies had a monofocal phase. pattern, with a wide longitudinal area of early activation, corresponding to the area of a break- through of atrium to vein conduction (Fig. 8). In Evaluation of ectopic activity seven veins, at least one site of earliest activation during the ectopy corresponded or was very close to a breakthrough; in these veins the activation Recordings of the ectopic activity were repeatedly pattern was longitudinal in four veins and trans- acquired in 10 veins in seven patients. Data are verse in three. In the remaining three veins, the shown in Table 2. In all cases, the ectopy in a given Figure 6 Isochronal polar map of a left superior (panel A) and inferior (panel B) pulmonary veins in the same patient during left atrial appendage stimulation. In the superior vein, three breakthroughs (at 11, 2e3 and 6 o’clock, respectively) are present, whereas the inferior shows two breakthroughs at 12e1 and 4 o’clock, respectively. It is noteworthy that there is a strict anatomic relationship between the inferior breakthrough of the superior vein and the superior breakthrough of the inferior vein. Mapping of pulmonary veins 105 Table 2 Location and characteristics of the ectopic activity in the pulmonary veins RSPV LSPV LIPV Common os of left PVs No. of veins with ectopic activity 4 3 2 1 Pattern Multifocal 3 2 2 1 Monofocal 1 1 ee Location in the vein Mediumedistal 2 1 1 e Proximal 2 2 1 1 Activation pattern of the vein during sinus rhythm or pacing Longitudinal 2 3 1 1 Transverse 2 e 1 e Abbreviations as in Table 1. ectopy was not spatially related to a breakthrough. Discussion Of the 10 ectopies mapped, six showed at least one area of earliest activation recorded in the two PVs have been identified as structures playing a proximal dipoles of the basket catheter, and, major role in the genesis of AF in humans [1,2]. therefore, in the area corresponding to veno-atrial Nevertheless, comprehensive understanding of the junction. Seven ectopies occurred in a vein with arrhythmogenesis of AF and particularly of the un- a longitudinal activation pattern, whereas in the derlying mechanism and role of ectopic activity remaining three the activation pattern of the vein arising from the left atrium and PVs has not was transverse. yet been gathered [14,15]. Apparently, precise Figure 7 Isochronal polar map of ectopic activity in a right superior (panel A) and left inferior (panel B) pulmonary vein. In both panels a multifocal pattern is shown. In fact, two areas in panel A and three areas in panel B widely longitudinally and transversely distributed in the vein show a similar activation time (within 10 ms). In panel A, an additional area activated 29 ms later than the earliest is observed in a diametrically opposite site. In both veins, areas of very early activation during ectopy are located proximally in the vein. 106 R. De Ponti et al. Figure 8 Isochronal polar map of a right superior pulmonary vein during ectopic activity (panel A) and on sinus rhythm (panel B). In this case, the ectopic activity shows a monofocal pattern, since a single area of early activation (dark red area) is observed on the floor of the vein in a proximal position. From this site, the activation spreads distally in the vein, following the direction of the breakthrough observed in sinus rhythm in panel B. anatomical and histological analysis of the PVs in forms: a predominantly longitudinal activation pat- humans has found no major difference in subjects tern (observed in 13 veins) and a predominantly with and without AF, since in both cases the sub- transverse activation pattern (observed in 18 strate for non-uniform anisotropic conduction re- veins). In the first pattern, the impulse proceeds lated to fibrosis and complex fibre alignment was from the breakthrough distally in the vein, follow- observed [3,5,6]. On the other hand, it is possible ing a direction parallel to the long vein axis. Con- that the major difference between patients with versely, in the transverse pattern, the impulse and those without AF concerns functional aspects propagates from the breakthrough in a circular of the myocardial fibres surrounding the PVs. Re- fashion with a direction perpendicular to the long cent reports provided evidence of a higher degree vein axis. Although the type of activation pattern of decremental conduction [10] and a shorter ef- in each vein appears very clear in a qualitative fective refractory period [11] of these fibres com- analysis, limited areas of transverse activation pared with left atrial myocardium in patients were observed even in patients with a predomi- suffering from AF. To our knowledge, this is the first nantly longitudinal pattern, reflecting a complex study of computerized high-density mapping of the electrical activation of the vein. However, the dif- PVs during sinus rhythm, pacing and ectopic activ- ferent activation patterns apparently have no im- ity originating from the PVs, in patients with parox- pact on the duration of PV electrical activation ysmal AF. Although preliminary, these data may and, in fact, the time interval between the earliest contribute to increase our understanding of func- and the latest PV potential is very similar for both tional aspects of the veno-atrial junction and myo- patterns. This suggests that the two activation pat- cardial extensions of the PVs. terns are an expression of different propagation di- rections into the PVs, with no significant effect on duration, since later activated areas, although dif- Pulmonary vein activation patterns ferently distributed, can be observed in both pat- terns. These findings may be well correlated with In the present study, the evaluation of activation histological data of PVs, reporting the presence of inside the PVs led to identification of different pat- loops of fibres leaving the atrium and returning to terns, which can be grouped into two fundamental it after covering the venous wall [3] and of Mapping of pulmonary veins 107 a mesh-like arrangement with a combination of spi- electrophysiological procedure. Another interest- rally and longitudinally oriented bundles of myo- ing finding of the present study has to be high- cytes [6]. In the present study a trend towards lighted. In all the 24/31 veins evaluated in more a higher prevalence of the transverse activation than one rhythm, the use of left atrial pacing had pattern has been noted in the right superior PVs. only the ability to render more evident the break- This may reflect a more complex fibre orientation through closer to the site of origin of the rhythm in the right PVs, as they are close to interatrial con- and in no case did left-sided pacing reveal a latent nections observed in this area [4]. Interestingly, in breakthrough. This suggests that, when high- the majority of patients in whom more than one density mapping is used, analysis in a single rhythm vein was mapped, the same activation pattern (lon- could be enough to identify all the atrium to vein gitudinal or transverse) was observed in all the conduction breakthroughs. Some other considera- veins, suggesting in some cases a patient specific tions of the use of pacing to discriminate atrial to activation pattern of PVs. vein potentials have to be made. In this study acqui- sitions were made during sinus rhythm and pacing from different left sites, but 28% of them were High-density mapping and identification of not analyzed because of poor discrimination be- atrium to vein conduction breakthroughs tween atrial and venous potentials and five right and two left PVs were evaluated in a single rhythm. Nevertheless, sinus rhythm turned out to be the High-density mapping of the PVs identifies multiple ‘‘golden’’ rhythm for high-density mapping, since discrete areas of earliest activation along the pe- it allowed satisfactory discrimination between the rimeter of the vein os, which reflect the location atrial and venous potentials in all cases. Coronary of the atrium to vein conduction breakthroughs. sinus pacing can be alternatively and successfully According to the data presented in this study, all used in the vast majority of the left PVs, whereas the sites of breakthrough can be localized simulta- neously and precisely on the high-density map. This the usefulness of pacing from the roof and append- ability can be explained by the high resolution pro- age of the left atrium seems in doubt, since it al- vided by 56 bipolar recordings distributed along the lowed differentiation of the potentials only in perimeter of the vein os, which allow detailed anal- approximately half of the cases, both in the right ysis of the activation at the veno-atrial junction. In superior and left PVs. This could be in partial con- trast to previous data [16], reporting the need for the present study, no less than two and no more distal coronary sinus pacing to avoid the superim- than three discrete breakthroughs were found in position of atrial to PV potentials in the left PVs. the mapped veins. In a previous study [12],in34 Nevertheless, in the present study, the adaptability out of 162 (20%) of the considered PVs a single of the basket catheter favoured an orientation of breakthrough of atrium to vein conduction was the dipoles parallel to the long pulmonary vein found. Conversely, a study of PV anatomy [7] re- axis, virtually concordant with the direction of ported the presence of myocardial tissue along atrium to vein propagation and this may have con- the entire perimeter of the PV ostium, suggesting tributed to a better discrimination of potentials that ablation should be frequently extended to even in sinus rhythm. the complete circumference of the ostium to ob- tain electrical vein isolation. The discrepancy be- Evaluation of ectopic activity tween the first and our study could be explained both by the relevantly smaller number of patients considered in this paper and by the fact that in In the present study, spontaneous ectopic activity the present study the breakthroughs were evaluated was repeatedly observed and recorded in approxi- at the very proximal veno-atrial junction, where the mately one third of the considered veins. Although longitudinal myocardial fibres tend to be broader this is very early experience, it is important to un- and thicker, whereas they become thinner in a more derline that in the majority of the cases the ectopy distal position in the vein [6]. This might have led to showed multiple sites of earliest activation in dif- identification of a higher number of breakthroughs. ferent areas of the veins, a close spatial relation- As to the other study [7], the presence of muscular ship with the site of breakthrough and a proximal fibres all along the perimeter of the vein ostium position in the vein. Of course, further data are re- does not necessarily imply that they have a func- quired to confirm these observations and subclassi- tional relevance to the atrium to vein conduction fy the ectopic pattern. Nevertheless, the finding of and, moreover, the definition of the ostium posi- multiple foci in the same vein is in accordance with tion may be different in anatomical specimens a prior report by Haı¨ssaguerre et al. [1], who ob- compared with angiographic images during an served during ectopy, a synchronous local activation 108 R. De Ponti et al. in a large sector of the venous perimeter or differ- catheter in the PV was very accurate, in the major- ent activation times during ectopy, suggesting ity of the cases it is not possible to distribute in a various ectopy sources or activation courses. Al- geometrically homogeneous way the splines along though less probable, it cannot be excluded ‘‘a pri- the circumference of the vein, as they are repre- ori’’ that a multifocal pattern is the result of sented on the map. Consequently, the actual posi- simultaneous propagation in a distant part of the tion and the width of each breakthrough is better vein from a remote site, deep in the venous branch. defined referring to the position of the spline in the It is also remarkable that in the present study high- vein rather than to the clock quadrant on the map. density mapping showed an identical pattern of the earliest ectopic activation in the three subsequent References recordings, which were performed in each vein, ac- cording to the protocol. This suggests that, regard- [1] Haı¨ssaguerre M, Jaı¨s P, Shah DC, et al. Electrophysiological less of the subsequent course from the vein to the end point for catheter ablation of atrial fibrillation initiated atrium, each ectopic activity in each vein may have from multiple pulmonary venous foci. Circulation 2000;101: a precise and fixed ‘‘site of origin’’. 1409e17. [2] Hsieh MH, Tai CT, Tsai CF, et al. Pulmonary vein electrogram characteristic in patients with focal sources of paroxysmal Clinical implications atrial fibrillation. J Cardiovasc Electrophysiol 2000;11: 953e9. [3] Nathan H, Eliakim M. The junction between the left atrium The ability to locate simultaneously by high-density and the pulmonary veins. An anatomic study of the human mapping all the atrium to vein connections in a sin- hearts. Circulation 1966;34:412e22. gle sinus beat may have an important practical role, [4] Ho SY, Sanchez-Quintana D, Cabrera JA, Anderson RH. especially in cases showing AF recurrences early af- Anatomy of the left atrium: implications for radiofrequency ter DC cardioversion. In these as well as in the other ablation of atrial fibrillation. J Cardiovasc Electrophysiol 1999;10:1525e33. AF cases, this method allows a precise electrophys- [5] Saito T, Waki K, Becker AE. Left atrial myocardial iologically guided approach to produce discrete le- extensions onto pulmonary veins in humans: anatomic sions aimed at electrical disconnection of the PV. observations relevant for atrial arrhythmias. J Cardiovasc Moreover, localization of ectopic activity in the Electrophysiol 2000;11:888e94. proximal part of the vein may have a direct implica- [6] Ho SY, Cabrera JA, Tran VH, Farre ´ J, Anderson RH, Sanchez- Quintana D. Architecture of the pulmonary veins: relevance tion on the ablation strategy, since the operator has to radiofrequency ablation. Heart 2001;86:265e70. consistent data to decide how proximal the abla- [7] Weiss C, Gocht A, Willems S, Hoffmann M, Risius T, Meinertz tion lesion should be to neutralize also the proximal T. Impact of the distribution and structure of myocardium in PV foci. In so doing it may be hoped that the most the pulmonary veins for radiofrequency ablation of atrial clinically useful approach is obtained when an elec- fibrillation. Pacing Clin Electrophysiol 2002;1352e6. [8] Zipes DP, Knope RF. Electrical properties of the thoracic trophysiologically guided method is used. veins. Am J Cardiol 1972;29:372e6. [9] Spach MS, Barr RC, Jewett PH. Spread of excitation from the atrium into thoracic veins in human beings and dogs. Limitations Am J Cardiol 1972;30:844e54. [10] Tada H, Oral H, Ozadyn M, et al. Response of pulmonary These are preliminary data on a limited number of vein potentials to premature stimulation. J Cardiovasc PVs and patients, which allow evaluation of the Electrophysiol 2002;13:33e7. feasibility and the potential of high-density map- [11] Jaı¨s P, Hocini M, Macle L, et al. Distinctive electrophysio- logical properties of pulmonary veins in patients with atrial ping of the PVs. There are two major limitations fibrillation. Circulation 2002;106:2479e85. in this study. First, every electrogram was manually [12] Haı¨ssaguerre M, Shah DC, Jaı¨s P, et al. Electrophysiological annotated to define the local activation time ac- breakthroughs from the left atrium to the pulmonary veins. cording to the method described, since the current Circulation 2000;102:2463e5. version of the QMS2 system software has no algo- [13] De Ponti R, Ho SY, Salerno-Uriarte JA, Tritto M, Spadacini G. Electroanatomic analysis of sinus impulse propagation in rithm for the discrimination between atrial and normal human atria. J Cardiovasc Electrophysiol 2002;13: PV potentials and for the subsequent evaluation 1e10. of activation times. Even if an algorithm for auto- [14] Shah DC, Haıssaguerre M, Jaıs P. Toward a mechanism- ¨ ¨ matic analysis is developed, such a complex anno- based understanding of atrial fibrillation. J Cardiovasc tation of the signals will require further Electrophysiol 2001;12:600e1. [15] Prystowsky EN. When technology exceeds knowledge, is evaluation by the electrophysiologist to generate success a reasonable expectation? J Cardiovasc Electro- a correct map and this, as well as the manual anno- physiol 2001;12:1284e5. tation used in this study, is time consuming and may [16] Haı¨ssaguerre M, Jaı¨s P, Shah DC, et al. Catheter ablation of limit, in some cases, the use of this method. chronic atrial fibrillation targeting the reinitiating triggers. Second, although the deployment of the basket J Cardiovasc Electrophysiol 2000;11:2e10. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Europace Oxford University Press

Computerized high-density mapping of the pulmonary veins: new insights into their electrical activation in patients with atrial fibrillation

Download PDF
12 pages

Loading next page...
 
/lp/oxford-university-press/computerized-high-density-mapping-of-the-pulmonary-veins-new-insights-6zrdM4ryJa

References (16)

Publisher
Oxford University Press
Copyright
© 2004 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.
ISSN
1099-5129
eISSN
1532-2092
DOI
10.1016/j.eupc.2003.11.004
pmid
15018866
Publisher site
See Article on Publisher Site

Abstract

Europace (2004) 6,97e108 Computerized high-density mapping of the pulmonary veins: new insights into their electrical activation in patients with atrial fibrillation a, b a Roberto De Ponti , Massimo Tritto , Marcelo E. Lanzotti , b a b Giammario Spadacini , Raffaella Marazzi , Paolo Moretti , Jorge A. Salerno-Uriarte University of Insubria, Department of Cardiovascular Sciences ‘‘Ospedale di Circolo e Fondazione Macchi’’ Varese, Italy Department of Cardiology ‘‘Mater Domini’’, Castellanza, Italy Submitted 25 March 2003, and accepted after revision 9 November 2003 KEYWORDS Abstract Aim To report the method and findings of computerized high-density high-density mapping of pulmonary veins (PVs) in patients undergoing their electrical isolation mapping; for atrial fibrillation (AF). pulmonary veins; Methods and results In 17 consecutive patients (8 M, age 55G11 years), a 64 elec- atrial fibrillation trode basket catheter was placed in the target PVs and 56 bipolar electrograms were recorded, analyzed and isochronal maps were generated. PVs were mapped during sinus rhythm, left-sided pacing and ectopic activity. The sites of earliest activation at the veno-atrial junction were defined as the atrium to vein conduction break- throughs. PV activation pattern was classified as predominantly longitudinal or transverse, according to the direction of the impulse from the breakthroughs. The ectopic pattern was defined as multifocal, when distant areas in the PV had activa- tion times within 10 ms. Thirty-one PVs were mapped. The activation pattern was predominantly longitudinal in 13 PVs and transverse in 18 PVs. Two breakthroughs were identified in 22 PVs and three in nine. All the breakthroughs were evident si- multaneously in sinus rhythm and left-sided pacing changed only the predominance of the breakthrough. Ectopies were mapped in 10 PVs: eight showed a multifocal and two a monofocal pattern; six ectopies originated from the proximal tract of the PV. Conclusion High-density mapping of PV identifies a typical activation pattern. Multiple and discrete breakthroughs are simultaneously identified in sinus rhythm. The majority of the mapped ectopies has a multifocal pattern and proximal origin. ª 2003 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. Corresponding author. Tel.: D39-0332-278394; fax: D39-0332-393309. E-mail address: [email protected] (R. De Ponti). 1099-5129/$30 ª 2003 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.eupc.2003.11.004 98 R. De Ponti et al. physiological procedure, all patients were on oral Introduction anticoagulants, which were withdrawn and substi- tuted by heparin 24 h before the procedure. A In patients with atrial fibrillation (AF), the pulmon- preprocedure transoesophageal echocardiogram ary veins (PVs) are considered a critical area for revealed no thrombus in the left atrium and normal arrhythmogenesis and therefore they represent blood flow velocity in the left atrial appendage. All a target for ablation [1,2]. Over the last years, sev- antiarrhythmic drugs were discontinued one week eral reports [3e7] focussed on the anatomy and the before the procedure; no patient was receiving histology of the myocardial extensions surrounding amiodarone. the PVs, in patients with or without AF, providing detailed description of their complex morphology, where fibrous and fatty tissue are interspersed Electrophysiological procedure and data among the myocardial fibres, forming a mesh-like acquisition arrangement [5,6]. On the other hand, little is known about the electrophysiological properties All patients underwent the electrophysiological and electrical activation of the myocardial tissue procedure after having signed the consent form surrounding the PVs. Both in the past [8,9] and re- for the study, approved by the Institutional Ethical cently [10,11], limited data have been gathered, Committee. Double transseptal puncture was per- which provide evidence of heterogeneous and dec- formed in all cases. Two long sheaths were, then, remental conduction properties [10] and of shorter positioned in the left atrium: an 8F 62 cm long effective refractory period [11] of these myocardial sheath (Preface, Biosense Webster, Inc., USA) was extensions, compared with the proper left atrial used to introduce the angiographic catheter or, musculature. High-density mapping of the PVs may alternatively, the ablation catheter in the left contribute to increase our knowledge of their func- atrium, whereas a 9F 60 cm long sheath (Boston Sci- tional aspects and electrical activation. This may entific Corporation, USA) with different curves (55, be useful not only precisely to locate the break- 90 or 120() was used to position the mapping cathe- throughs of atrium to vein conduction, an essential ter. After transseptal puncture, intravenous hepa- prerequisite for their ablation in patients with AF rin was administered to maintain the activated [12], but also to identify patterns of electrical acti- clotting time between 250 and 300 s. A 64 elec- vation during different heart rhythms and, espe- trode 31 mm diameter basket catheter (Constella- cially, during ectopic activity triggering AF. tion, Boston Scientific Corporation, USA) was used The aim of this study was to define by computer- to map the PVs; in this catheter eight electrodes ized high-density mapping the electrical activation are uniformly distributed on each of the eight of the PVs during sinus rhythm, pacing from differ- splines. In every patient, after venography of all ent left atrial sites and ectopic beats, in patients PVs, every effort was made to identify the arrhyth- with AF. mogenic PV(s). For this purpose, if the patient was in AF, repeated cardioversions during temporary sedation with propofol were performed. If the pa- tient was in sinus rhythm, prolonged observation Methods and provocative manoeuvers were used to detect spontaneous ectopic activity. Once the arrhythmo- Patient population genic PVs were identified, they were targeted for electrical isolation. In case no ectopic activity Seventeen consecutive patients (8 M, 9 F; mean age was observed, the superior PVs were isolated. In 55G11 years) with paroxysmal, frequently recur- every case, the basket catheter was deployed in rent AF, undergoing an electrophysiological proce- the target vein (Fig. 1A) by withdrawing the long dure for electrical isolation of the PVs, have been sheath, previously advanced in the PV over a included. Patients who previously underwent a PV 0.035 in. guidewire. After deployment, minimal ro- isolation procedure were excluded from the study. tations of the basket helped to distribute uniformly In all patients, repeated 12-lead Holter monitoring the splines of the basket inside the vein and opti- showed that the clinical palpitation episodes were mize the electrode/tissue contact. A second veno- related to AF, reproducibly initiated by atrial ecto- graphy (Fig. 1B) was then performed to assess the pies. All patients were refractory to 2:5G1:3 anti- position of the proximal electrodes at or very close arrhythmic drugs, previously tested. No patient to the os of the vein. By using two fluoroscopic pro- had structural heart disease, as assessed by echo- jections (Fig. 2), the position of the ‘‘A’’ and ‘‘B’’ cardiogram. For at least 45 days before the electro- splines (identified by radiopaque markers) in the Mapping of pulmonary veins 99 Figure 1 (A, B) Fluoroscopic image of the basket catheter position in a right superior pulmonary vein in antero- posterior view. In panel A, the basket and the angiographic catheters are both positioned in the vein through transseptal catheterization. It is noteworthy that the adaptable geometry of the basket fits the anatomy of the pulmonary vein. The arrows indicate the radiopaque markers (very similar to regular electrodes), which identify spline ‘‘A’’ and ‘‘B’’. When dye is injected (panel B), satisfactory electrode to vein contact is assessed. Data analysis and definitions vein was recorded and classified according to the clock quadrant, where 12 and 3 h are superior and anterior, respectively, for the left PVs, while Each segment acquired by the QMS2 system was an- they are superior and posterior, respectively, for alyzed in order to (1) define the pattern of electri- the right PVs. Through a junction box, the bipolar cal activation in the PV, (2) identify the sites of the signals from the basket catheter were simulta- breakthroughs of atrium to vein conduction, and neously recorded on a computerized mapping sys- (3) evaluate the pattern and site of earliest activa- tem (QMS2 System, Boston Scientific Corporation, tion during the ectopic activity. On the system, USA) for data analysis and on the conventional each electrogram was manually annotated to de- electrophysiology system for on-line continuous termine the local activation time, according to monitoring. In the QMS2 system, 56 bipolar electro- the following criteria. grams were recorded from seven contiguous longi- The analysis of the activation pattern of the PV tudinal bipoles on each spline. Signals were during sinus rhythm or pacing was performed at processed with filter and gain settings of a sweep speed of 200 mm/s. The local activation 30e290 Hz and 8e64, respectively. For each map- time was measured at the onset of each PV poten- ped PV, 30 s recordings were made in three differ- tial and the activation time of the earliest acti- ent rhythms: sinus rhythm, pacing from distal vated PV potential was considered as time 0. The coronary sinus and left atrial appendage for the left following two criteria were used to discriminate PVs and sinus rhythm, pacing from proximal coro- PV potentials from atrial electrograms, which may nary sinus and roof of the left atrium for the right be recorded as a far field activity even in the distal PVs. For the evaluation of the ectopic activity, only electrode pairs. The first criterion was the chronol- cases, in whom at least three recordings of the ogy of the PV potential from the surface P wave on- ectopic activity in the same vein were acquired, set: no signal was considered as a PV potential if were considered. In all cases, recordings of the ec- earlier than 75 and 35 ms in sinus rhythm and left topic activity were performed when no catheter atrial pacing, respectively, for the left PVs and 40 other than the basket was inside the vein. and 60 ms, in sinus rhythm/left atrial roof pacing 100 R. De Ponti et al. Figure 2 (A, B) Fluoroscopic image of the basket catheter position in the same vein and patient, in 30( right anterior oblique view, before (panel A) and during (panel B) venography. An oblique fluoroscopic projection allows further assessment of catheter positioning and contact inside the vein, the course of which is not rectilinear, but shows curves in different planes. A rather homogeneous distribution of the splines around the vein perimeter is observed and the position of spline A (indicated by the arrow and identified by a single radiopaque marker) is evident at 5 o’clock. and proximal coronary sinus pacing, respectively, visualized in the centre of the map, whereas the for the right PVs. These time intervals were deter- proximal electrodes are along the outer perimeter. mined on the basis of a previous activation study The position of spline ‘‘A’’ is identified on the map [13]. Second, for each spline, the discrimination by larger dots and the other splines are distributed of the atrial from PV potentials was done first in in a counter-clockwise fashion. For each map, the the proximal electrograms, where the onset of position of the spline ‘‘A’’ was adjusted according the PV potentials was annotated. Each subsequent to its position on the clock quadrant as assessed by annotation in a more distal dipole was made at the fluoroscopy. On each map, during sinus rhythm or onset of the first deflection later than the previous pacing a breakthrough of atrium to vein conduction annotation in the proximal dipole on the same was defined as the site of earliest activation along spline. Examples may be seen in Figs. 3 and 4.In the outer perimeter of the map. Comparison be- case satisfactory discrimination between atrial tween the site of the breakthrough and the site and vein potentials was not possible, the recording where ablation modified and abolished PV activity was not considered for analysis. was made for each breakthrough in every patient. During the ectopic activity, in order to evaluate The pattern of electrical activation inside the vein the site of earliest activation, each electrogram vi- was defined as predominantly longitudinal, if fast sualized at the maximum gain value was annotated propagation was observed from the breakthrough at the onset of the first sharp deflection. To avoid to the distal part of the vein with a direction longitu- the bias of artefact and misinterpretation of the dinal to the long axis of the vein. On the isochronal signals, the analysis was repeated for three ectopic map, red to yellow colours are distributed on the beats from each vein and comparison with the elec- same spline. Conversely, the propagation pattern trograms of a sinus beat not followed by concealed inside the vein was defined as predominantly trans- or manifest ectopy was made. verse, when the propagation proceeds from the After the annotation process was complete, a po- breakthrough with a direction perpendicular to lar isochronal map was generated by the system, the long axis of the vein. In this case, the red to yel- where red and blue indicate early and late activa- low colours are distributed in the proximal electro- tion, respectively, and the distal electrodes are des of adjacent splines. In the analysis of the site of Mapping of pulmonary veins 101 Figure 3 Isochronal polar map of a left superior pulmonary vein and display of the electrograms recorded in the spline C, E and F, during coronary sinus pacing. In the map, as well as in the following figures, the inner circle identifies the distal basket electrodes, whereas the outer circle corresponds to the proximal electrodes. Spline ‘‘A’’ (bigger dots) is located at 12 o’clock; red and purple identify earliest and latest activation, respectively, of the pulmonary vein potentials and the activation values are shown on the scale. The electrograms are displayed from top to bottom from the distal to the proximal electrode pair of the spline; the red line corresponds to the stimulation artefact preceding the surface P wave (yellow line on the top) and for each electrogram the number indicates the activation time. In this case, three breakthroughs, located at 9e10, 2 and 5 o’clock, respectively, are present, the inferior breakthroughs prevailing during coronary sinus stimulation. A predominantly longitudinal activation pattern is observed in the vein, since the activation proceeds fast in a longitudinal direction, from the breakthroughs along the corresponding splines, as assessed also by the activation values in splines ‘‘C’’ and ‘‘F’’. In a later activated area corresponding to spline ‘‘E’’, transverse conduction is observed, as assessed by very similar activation values from proximal to distal electrode pairs. earliest activation during ectopies, the pattern was in Table 1. The number of PVs mapped per patient defined as multifocal if activation times of different was 2G0:7. In two patients the left PVs were map- sites in non-adjacent splines were within 10 ms. All ped and isolated at their common os and, for the the recordings and maps were independently eval- purpose of this study, they have been considered uated by two electrophysiologists. as a single vein. In no case, was ectopy arising from the right inferior PV documented and, according to the protocol, this vein was not considered for abla- Statistical analysis tion. Ninety-three acquisitions were made during sinus rhythm, coronary sinus and left atrial stimula- Continuous variables are expressed as meanG 1 SD. tion, whereas the remaining 30 acquisitions (three Considering the limited value of finding statistically repeated acquisitions in 10 different veins) were significant differences in such a restricted num- performed during ectopies. Sixty-seven of 93 ber of patients and veins as those of the present (72%) acquisitions during sinus rhythm or pacing study, statistical analysis was not performed for could be analyzed and an isochronal map gener- the activation times and the distribution of activa- ated; the remaining 26 (28%) acquisitions per- tion patterns. formed during pacing were not considered for analysis, because the atrial and PV potentials could Results not be clearly discriminated. Anyhow, it is impor- tant to highlight that all the 31 veins could be eval- Data acquisition and analysis uated in sinus rhythm. Table 1 reports the percent of the recordings that could be analyzed in differ- A total of 123 acquisitions was made by the ent rhythms out of the overall number of acquisi- QMS2 system in 31 PVs, whose location is shown tions made for each vein. As also shown in Table 1, 102 R. De Ponti et al. Figure 4 Isochronal polar map of a right superior pulmonary vein and display of the electrograms recorded in the spline C, E and G during sinus rhythm. The map and the splines are as in the previous figure. In this case, the activation in the veins shows a predominant transverse activation pattern, since the propagation proceeds from the two breakthroughs at 10e12 and 3e5 o’clock following a direction perpendicular to the long axis of the vein. Consequently, the redeorange colour spreads from the breakthrough to the proximal dipoles of the adjacent splines, rather than proceeding longitudinally along spline ‘‘C’’ and ‘‘G’’, as in the case shown in the previous figure. In fact, there is a jump of 26 and 10 ms in the conduction along spline ‘‘C’’ and ‘‘G’’, respectively, between the third and the fourth proximal dipole. A later activated spline (‘‘E’’) shows as well a transverse pattern, with very similar activation values from proximal to distal dipoles. 7/31 (22.5%) veins were evaluated in a single artefacts and have been annotated. Concordance (sinus) rhythm, the vast majority of them (5/7 between the two independent observers was veins) being right superior PVs. In only two cases, obtained in 97% of the maps. Thirty of 31 PVs the basket catheter displaced minimally during were completely isolated by low power radio- the procedure, with a clockwise rotation of 15 frequency energy (14 patients) or cryothermal and 30(, respectively, which, in no case, rendered energy (three patients) delivery, whereas in one the evaluation of the map impossible. Overall, in vein a 90% reduction of the amplitude of the the 97 analyzed acquisitions a total of 5328 PV potentials was obtained. No complication was electrograms (96.4%) were free from technical observed during or after the procedure. Table 1 Number of veins mapped and their features RSPV LSPV LIPV Common os of left PVs No. of veins 12 12 5 2 No. of veins mapped in a single rhythm 5 1 1 e Percent of recordings analyzed during Sinus rhythm 100% 100% 100% 100% Coronary sinus pacing 50% 93% 80% 100% Left atrial stimulation 42% 43% 40% 50% No. of veins with a predominant Longitudinal activation pattern 4 6 3 e Transverse activation pattern 8 6 2 2 No. of veins with Two breakthroughs 8 8 4 2 Three breakthroughs 4 4 1 e Abbreviations: LIPV[ left inferior PV; LSPV[ left superior PV; RSPV[ right superior PV. Mapping of pulmonary veins 103 Activation patterns in the pulmonary veins Localization of atrium to vein conduction breakthroughs As shown in Table 1, a predominant longitudinal ac- tivation pattern (Fig. 3) was observed in 13 veins, In all the 31 veins, more than one breakthrough of while in the remaining 18 veins it was transverse atrium to vein conduction was observed and in no (Fig. 4), with a trend towards a more frequent case were more than three breakthroughs identi- transverse pattern in the right superior PV. Also in fied. The number of breakthroughs in the different veins classified as having a predominant longitudi- PVs is shown in Table 1. The location of the break- nal activation pattern, limited areas of transverse throughs was ubiquitous in the clock quadrants in conduction were present, as shown in Fig. 3.Inno mapped veins. In 24 PVs analyzed in more than case, in whom the PV was evaluated in more than one rhythm, the position of the breakthrough was one rhythm, was a rhythm-dependent change in not rhythm-dependent and, importantly, no addi- the activation pattern observed. Interestingly, tional breakthrough became evident by pacing among the 13 patients, in whom more than one vein at different sites. Nevertheless, a change in the was mapped, nine showed the same activation pat- predominance of a breakthrough depending on tern in all mapped veins, which was transverse in the change of the rhythm was observed in every five and longitudinal in four. During sinus rhythm, case, the breakthrough closer to the site of origin the time interval between the earliest and latest of the rhythm becoming predominant. Fig. 5 shows activated PV potential was 40G19 ms in the left su- an example of this finding. In all cases, in whom perior PV, 36G19 ms in the right superior, 31G8ms both the left superior and left inferior PVs in the left inferior and 49G27 ms in the common os were separately mapped, a strict anatomical rela- of left PVs. In sinus rhythm, the activation time of tionship was observed between the inferior veins with a predominantly longitudinal pattern breakthrough of the superior vein (identified at was very similar to one of the veins with a predom- 6e7 o’clock) and the superior breakthrough of the inantly transverse pattern (37G9ms vs 35G8 ms). inferior vein, localized between 11 and 1 o’clock Figure 5 Isochronal polar map of a left inferior pulmonary vein in sinus rhythm and pacing. In sinus rhythm (panel A) the two breakthroughs, located at 11e12 and 5e6 o’clock, respectively, are equally predominant and the propagation inside the vein proceeds longitudinally from both breakthroughs with similar activation times. During left atrial appendage (panel B) and coronary sinus (panel C) stimulation, the breakthrough closer to the pacing site becomes more evident than the opposite one and the activation times inside the vein modify accordingly, although the longitudinal activation pattern of the vein does not change. 104 R. De Ponti et al. (Fig. 6). Finally, in 29 out of 31 veins there was co- vein showed a reproducible activation pattern with incidence of the location of the splines identifying no significant change among the three subsequent the atrium to vein conduction breakthroughs and acquisitions. In eight veins, ectopies showed a mul- the site where energy application resulted in mod- tifocal activation pattern with areas of early simul- ification and abolition of the PV potentials. In two taneous activation widely distributed in the vein, PVs, this validation was rendered impossible by as paradigmatically shown in Fig. 7. Only in two a rotation of the basket catheter after the mapping veins in the same patient, ectopies had a monofocal phase. pattern, with a wide longitudinal area of early activation, corresponding to the area of a break- through of atrium to vein conduction (Fig. 8). In Evaluation of ectopic activity seven veins, at least one site of earliest activation during the ectopy corresponded or was very close to a breakthrough; in these veins the activation Recordings of the ectopic activity were repeatedly pattern was longitudinal in four veins and trans- acquired in 10 veins in seven patients. Data are verse in three. In the remaining three veins, the shown in Table 2. In all cases, the ectopy in a given Figure 6 Isochronal polar map of a left superior (panel A) and inferior (panel B) pulmonary veins in the same patient during left atrial appendage stimulation. In the superior vein, three breakthroughs (at 11, 2e3 and 6 o’clock, respectively) are present, whereas the inferior shows two breakthroughs at 12e1 and 4 o’clock, respectively. It is noteworthy that there is a strict anatomic relationship between the inferior breakthrough of the superior vein and the superior breakthrough of the inferior vein. Mapping of pulmonary veins 105 Table 2 Location and characteristics of the ectopic activity in the pulmonary veins RSPV LSPV LIPV Common os of left PVs No. of veins with ectopic activity 4 3 2 1 Pattern Multifocal 3 2 2 1 Monofocal 1 1 ee Location in the vein Mediumedistal 2 1 1 e Proximal 2 2 1 1 Activation pattern of the vein during sinus rhythm or pacing Longitudinal 2 3 1 1 Transverse 2 e 1 e Abbreviations as in Table 1. ectopy was not spatially related to a breakthrough. Discussion Of the 10 ectopies mapped, six showed at least one area of earliest activation recorded in the two PVs have been identified as structures playing a proximal dipoles of the basket catheter, and, major role in the genesis of AF in humans [1,2]. therefore, in the area corresponding to veno-atrial Nevertheless, comprehensive understanding of the junction. Seven ectopies occurred in a vein with arrhythmogenesis of AF and particularly of the un- a longitudinal activation pattern, whereas in the derlying mechanism and role of ectopic activity remaining three the activation pattern of the vein arising from the left atrium and PVs has not was transverse. yet been gathered [14,15]. Apparently, precise Figure 7 Isochronal polar map of ectopic activity in a right superior (panel A) and left inferior (panel B) pulmonary vein. In both panels a multifocal pattern is shown. In fact, two areas in panel A and three areas in panel B widely longitudinally and transversely distributed in the vein show a similar activation time (within 10 ms). In panel A, an additional area activated 29 ms later than the earliest is observed in a diametrically opposite site. In both veins, areas of very early activation during ectopy are located proximally in the vein. 106 R. De Ponti et al. Figure 8 Isochronal polar map of a right superior pulmonary vein during ectopic activity (panel A) and on sinus rhythm (panel B). In this case, the ectopic activity shows a monofocal pattern, since a single area of early activation (dark red area) is observed on the floor of the vein in a proximal position. From this site, the activation spreads distally in the vein, following the direction of the breakthrough observed in sinus rhythm in panel B. anatomical and histological analysis of the PVs in forms: a predominantly longitudinal activation pat- humans has found no major difference in subjects tern (observed in 13 veins) and a predominantly with and without AF, since in both cases the sub- transverse activation pattern (observed in 18 strate for non-uniform anisotropic conduction re- veins). In the first pattern, the impulse proceeds lated to fibrosis and complex fibre alignment was from the breakthrough distally in the vein, follow- observed [3,5,6]. On the other hand, it is possible ing a direction parallel to the long vein axis. Con- that the major difference between patients with versely, in the transverse pattern, the impulse and those without AF concerns functional aspects propagates from the breakthrough in a circular of the myocardial fibres surrounding the PVs. Re- fashion with a direction perpendicular to the long cent reports provided evidence of a higher degree vein axis. Although the type of activation pattern of decremental conduction [10] and a shorter ef- in each vein appears very clear in a qualitative fective refractory period [11] of these fibres com- analysis, limited areas of transverse activation pared with left atrial myocardium in patients were observed even in patients with a predomi- suffering from AF. To our knowledge, this is the first nantly longitudinal pattern, reflecting a complex study of computerized high-density mapping of the electrical activation of the vein. However, the dif- PVs during sinus rhythm, pacing and ectopic activ- ferent activation patterns apparently have no im- ity originating from the PVs, in patients with parox- pact on the duration of PV electrical activation ysmal AF. Although preliminary, these data may and, in fact, the time interval between the earliest contribute to increase our understanding of func- and the latest PV potential is very similar for both tional aspects of the veno-atrial junction and myo- patterns. This suggests that the two activation pat- cardial extensions of the PVs. terns are an expression of different propagation di- rections into the PVs, with no significant effect on duration, since later activated areas, although dif- Pulmonary vein activation patterns ferently distributed, can be observed in both pat- terns. These findings may be well correlated with In the present study, the evaluation of activation histological data of PVs, reporting the presence of inside the PVs led to identification of different pat- loops of fibres leaving the atrium and returning to terns, which can be grouped into two fundamental it after covering the venous wall [3] and of Mapping of pulmonary veins 107 a mesh-like arrangement with a combination of spi- electrophysiological procedure. Another interest- rally and longitudinally oriented bundles of myo- ing finding of the present study has to be high- cytes [6]. In the present study a trend towards lighted. In all the 24/31 veins evaluated in more a higher prevalence of the transverse activation than one rhythm, the use of left atrial pacing had pattern has been noted in the right superior PVs. only the ability to render more evident the break- This may reflect a more complex fibre orientation through closer to the site of origin of the rhythm in the right PVs, as they are close to interatrial con- and in no case did left-sided pacing reveal a latent nections observed in this area [4]. Interestingly, in breakthrough. This suggests that, when high- the majority of patients in whom more than one density mapping is used, analysis in a single rhythm vein was mapped, the same activation pattern (lon- could be enough to identify all the atrium to vein gitudinal or transverse) was observed in all the conduction breakthroughs. Some other considera- veins, suggesting in some cases a patient specific tions of the use of pacing to discriminate atrial to activation pattern of PVs. vein potentials have to be made. In this study acqui- sitions were made during sinus rhythm and pacing from different left sites, but 28% of them were High-density mapping and identification of not analyzed because of poor discrimination be- atrium to vein conduction breakthroughs tween atrial and venous potentials and five right and two left PVs were evaluated in a single rhythm. Nevertheless, sinus rhythm turned out to be the High-density mapping of the PVs identifies multiple ‘‘golden’’ rhythm for high-density mapping, since discrete areas of earliest activation along the pe- it allowed satisfactory discrimination between the rimeter of the vein os, which reflect the location atrial and venous potentials in all cases. Coronary of the atrium to vein conduction breakthroughs. sinus pacing can be alternatively and successfully According to the data presented in this study, all used in the vast majority of the left PVs, whereas the sites of breakthrough can be localized simulta- neously and precisely on the high-density map. This the usefulness of pacing from the roof and append- ability can be explained by the high resolution pro- age of the left atrium seems in doubt, since it al- vided by 56 bipolar recordings distributed along the lowed differentiation of the potentials only in perimeter of the vein os, which allow detailed anal- approximately half of the cases, both in the right ysis of the activation at the veno-atrial junction. In superior and left PVs. This could be in partial con- trast to previous data [16], reporting the need for the present study, no less than two and no more distal coronary sinus pacing to avoid the superim- than three discrete breakthroughs were found in position of atrial to PV potentials in the left PVs. the mapped veins. In a previous study [12],in34 Nevertheless, in the present study, the adaptability out of 162 (20%) of the considered PVs a single of the basket catheter favoured an orientation of breakthrough of atrium to vein conduction was the dipoles parallel to the long pulmonary vein found. Conversely, a study of PV anatomy [7] re- axis, virtually concordant with the direction of ported the presence of myocardial tissue along atrium to vein propagation and this may have con- the entire perimeter of the PV ostium, suggesting tributed to a better discrimination of potentials that ablation should be frequently extended to even in sinus rhythm. the complete circumference of the ostium to ob- tain electrical vein isolation. The discrepancy be- Evaluation of ectopic activity tween the first and our study could be explained both by the relevantly smaller number of patients considered in this paper and by the fact that in In the present study, spontaneous ectopic activity the present study the breakthroughs were evaluated was repeatedly observed and recorded in approxi- at the very proximal veno-atrial junction, where the mately one third of the considered veins. Although longitudinal myocardial fibres tend to be broader this is very early experience, it is important to un- and thicker, whereas they become thinner in a more derline that in the majority of the cases the ectopy distal position in the vein [6]. This might have led to showed multiple sites of earliest activation in dif- identification of a higher number of breakthroughs. ferent areas of the veins, a close spatial relation- As to the other study [7], the presence of muscular ship with the site of breakthrough and a proximal fibres all along the perimeter of the vein ostium position in the vein. Of course, further data are re- does not necessarily imply that they have a func- quired to confirm these observations and subclassi- tional relevance to the atrium to vein conduction fy the ectopic pattern. Nevertheless, the finding of and, moreover, the definition of the ostium posi- multiple foci in the same vein is in accordance with tion may be different in anatomical specimens a prior report by Haı¨ssaguerre et al. [1], who ob- compared with angiographic images during an served during ectopy, a synchronous local activation 108 R. De Ponti et al. in a large sector of the venous perimeter or differ- catheter in the PV was very accurate, in the major- ent activation times during ectopy, suggesting ity of the cases it is not possible to distribute in a various ectopy sources or activation courses. Al- geometrically homogeneous way the splines along though less probable, it cannot be excluded ‘‘a pri- the circumference of the vein, as they are repre- ori’’ that a multifocal pattern is the result of sented on the map. Consequently, the actual posi- simultaneous propagation in a distant part of the tion and the width of each breakthrough is better vein from a remote site, deep in the venous branch. defined referring to the position of the spline in the It is also remarkable that in the present study high- vein rather than to the clock quadrant on the map. density mapping showed an identical pattern of the earliest ectopic activation in the three subsequent References recordings, which were performed in each vein, ac- cording to the protocol. This suggests that, regard- [1] Haı¨ssaguerre M, Jaı¨s P, Shah DC, et al. Electrophysiological less of the subsequent course from the vein to the end point for catheter ablation of atrial fibrillation initiated atrium, each ectopic activity in each vein may have from multiple pulmonary venous foci. Circulation 2000;101: a precise and fixed ‘‘site of origin’’. 1409e17. [2] Hsieh MH, Tai CT, Tsai CF, et al. Pulmonary vein electrogram characteristic in patients with focal sources of paroxysmal Clinical implications atrial fibrillation. J Cardiovasc Electrophysiol 2000;11: 953e9. [3] Nathan H, Eliakim M. The junction between the left atrium The ability to locate simultaneously by high-density and the pulmonary veins. An anatomic study of the human mapping all the atrium to vein connections in a sin- hearts. Circulation 1966;34:412e22. gle sinus beat may have an important practical role, [4] Ho SY, Sanchez-Quintana D, Cabrera JA, Anderson RH. especially in cases showing AF recurrences early af- Anatomy of the left atrium: implications for radiofrequency ter DC cardioversion. In these as well as in the other ablation of atrial fibrillation. J Cardiovasc Electrophysiol 1999;10:1525e33. AF cases, this method allows a precise electrophys- [5] Saito T, Waki K, Becker AE. Left atrial myocardial iologically guided approach to produce discrete le- extensions onto pulmonary veins in humans: anatomic sions aimed at electrical disconnection of the PV. observations relevant for atrial arrhythmias. J Cardiovasc Moreover, localization of ectopic activity in the Electrophysiol 2000;11:888e94. proximal part of the vein may have a direct implica- [6] Ho SY, Cabrera JA, Tran VH, Farre ´ J, Anderson RH, Sanchez- Quintana D. Architecture of the pulmonary veins: relevance tion on the ablation strategy, since the operator has to radiofrequency ablation. Heart 2001;86:265e70. consistent data to decide how proximal the abla- [7] Weiss C, Gocht A, Willems S, Hoffmann M, Risius T, Meinertz tion lesion should be to neutralize also the proximal T. Impact of the distribution and structure of myocardium in PV foci. In so doing it may be hoped that the most the pulmonary veins for radiofrequency ablation of atrial clinically useful approach is obtained when an elec- fibrillation. Pacing Clin Electrophysiol 2002;1352e6. [8] Zipes DP, Knope RF. Electrical properties of the thoracic trophysiologically guided method is used. veins. Am J Cardiol 1972;29:372e6. [9] Spach MS, Barr RC, Jewett PH. Spread of excitation from the atrium into thoracic veins in human beings and dogs. Limitations Am J Cardiol 1972;30:844e54. [10] Tada H, Oral H, Ozadyn M, et al. Response of pulmonary These are preliminary data on a limited number of vein potentials to premature stimulation. J Cardiovasc PVs and patients, which allow evaluation of the Electrophysiol 2002;13:33e7. feasibility and the potential of high-density map- [11] Jaı¨s P, Hocini M, Macle L, et al. Distinctive electrophysio- logical properties of pulmonary veins in patients with atrial ping of the PVs. There are two major limitations fibrillation. Circulation 2002;106:2479e85. in this study. First, every electrogram was manually [12] Haı¨ssaguerre M, Shah DC, Jaı¨s P, et al. Electrophysiological annotated to define the local activation time ac- breakthroughs from the left atrium to the pulmonary veins. cording to the method described, since the current Circulation 2000;102:2463e5. version of the QMS2 system software has no algo- [13] De Ponti R, Ho SY, Salerno-Uriarte JA, Tritto M, Spadacini G. Electroanatomic analysis of sinus impulse propagation in rithm for the discrimination between atrial and normal human atria. J Cardiovasc Electrophysiol 2002;13: PV potentials and for the subsequent evaluation 1e10. of activation times. Even if an algorithm for auto- [14] Shah DC, Haıssaguerre M, Jaıs P. Toward a mechanism- ¨ ¨ matic analysis is developed, such a complex anno- based understanding of atrial fibrillation. J Cardiovasc tation of the signals will require further Electrophysiol 2001;12:600e1. [15] Prystowsky EN. When technology exceeds knowledge, is evaluation by the electrophysiologist to generate success a reasonable expectation? J Cardiovasc Electro- a correct map and this, as well as the manual anno- physiol 2001;12:1284e5. tation used in this study, is time consuming and may [16] Haı¨ssaguerre M, Jaı¨s P, Shah DC, et al. Catheter ablation of limit, in some cases, the use of this method. chronic atrial fibrillation targeting the reinitiating triggers. Second, although the deployment of the basket J Cardiovasc Electrophysiol 2000;11:2e10.

Journal

Europace – Oxford University Press

Published: Jan 1, 2004

Keywords: high-density mapping; pulmonary veins; atrial fibrillation

There are no references for this article.