Pulmonary Vein Morphology Before and After Segmental Isolation in Patients with Atrial Fibrillation

Marehiko Ueda; Hiroshi Tada; Kenji Kurosaki; Kazuhiro Itoi; Keiko Koyama; Shigeto Naito; Sachiko Ito; Issei Komuro; Shigeru Oshima; Koichi Taniguchi


Pacing Clin Electrophysiol. 2005;28(9):944-953. 

In This Article


This study included 30 patients with drug-resistant, paroxysmal AF (24 men, 6 women; mean age, 59 ± 8 years). The mean AF duration was 7 ± 7 years, and the mean number of symptomatic AF episodes per month was 20 ± 22. Echocardiography demonstrated a mean left ventricular ejection fraction of 0.63 ± 0.09. Two patients had coronary artery disease and the remaining 28 had no structural heart disease. In 10 patients (33%), linear ablation of the cavo-tricuspid isthmus was also performed for typical atrial flutter at the time of PV isolation. Multi-slice CT was performed within 1 week before and 3 ± 1 months after the PV isolation. At the time of acquisition of the CT before and after the PV isolation, all patients were in normal sinus rhythm. All patients gave informed consent for an electrophysiologic study, catheter ablation, and CT. As a control group, CT also was performed in 14 subjects with neither a history of AF nor structural heart disease (10 men, 4 women; mean age, 57 ± 16 years).

PV isolation was performed as previously described.[1,2,8] Electrograms were recorded near the PV ostia with a 7 Fr decapolar ring catheter (Lasso, Biosense Webster, Inc., Diamond Bar, CA, USA). RF energy was delivered with a 7 Fr quadripolar ablation catheter with a 4 mm distal electrode. PV isolation was performed by applying RF energy at the sites at which the earliest bipolar PV potentials and/or unipolar electrograms with the most rapid intrinsic deflection were recorded.[1,2,8] RF energy was delivered at a maximum power of 30 W and maximum temperature of 52°–C55°C for 60 seconds.[1,2,8] The endpoint for ablation was the elimination of the PV potentials at all Lasso catheter recording sites.

The images were acquired using a multi-slice CT scanner (GE Light Speed Ultra; GE Medical Image, Milwaukee, WI, USA) during intravenous injection of contrast dye (100 mL at 3 mL/s) in eight parallel slices (1.25 mm collimation). ECG gating was performed with a triggered delay set at 70% of the R-to-R interval to target the atrial end-diastolic phase. A processing workstation (Advanced Workstation 4.0; GE Medical Image) allowed for three-dimensional viewing of curved multi-planar reconstruction (MPR) images (Fig. 1A and B), virtual endoscopic images, reformatted cross-sectional images in discretionary directions (Fig. 1C), and volume rendering images (Figs. 2, upper panels, and 3). Curved MPR images were obtained using a cursor to trace a curved line along the center of the lumen of the PV manually on carefully selected cross-sectional source images, and orthogonal to the source images.[9]

(A) Measurement of the ostia of the pulmonary veins (PV) with curved MPR (upper panel: integrated coronal sectional view) and axial (lower panel: integrated axial sectional view) images of the left superior PVs. The black asterisk identifies the apex of the parabolic outline of the junction of the intervenal bridging wall. The trunk length of the PV was defined as the distance between the ostium and the onset of the first branch of the PV (arrow). The white asterisk identifies the apex of the parabolic outline of the junction to the left atrial appendage. (B) Coronal sectional (upper panel) and axial sectional (lower panel) images of the left superior PV demonstrating asymmetric stenosis after ablation. The full and dotted lines indicate the diameters of the perilesion references and the point of the maximal luminal reduction, respectively. This PV has a 48% luminal reduction in the supero-inferior direction (upper panel), but only a 14% luminal reduction in the antero-posterior direction (lower panel), indicating a heterogeneous pattern of stenosis in this vein. (C) Measurement of the left atrial diameters. The transverse (full line), antero-posterior (dotted line), and longitudinal diameters (arrow) were measured.

(A) Volume rendering images of the left inferior pulmonary vein (PV) in a patient with atrial fibrillation before ablation. The arrows indicate a narrowing of the vein (27% luminal reduction). (B) A volume rendering image postprocessing cutting the descending aorta (upper panel) and curved MPR (lower panel) of the left inferior PV before ablation. Dominant luminal reduction on the posterior side is shown. (C) A volume rendering image of the left inferior PV postprocessing cutting the spine (upper panel) and axial cross-sectional CT image (lower panel) in a control patient. The arrows indicate the narrowest portion of the vein (66% luminal reduction). A narrowed and kinked left inferior PV, positioned between the left atrium and the spine, is shown (black arrows).

Volume rendering images of the left inferior pulmonary vein (PV) before (left panel) and after ablation (right panel). The trunk length of the PV shortened after ablation (23% reduction compared with that before ablation).

In 30 patients with AF, quantitative measurements of the PVs, using electronic three-dimensional digital calipers, were performed in the right superior (n = 30; RS), right inferior (n = 30; RI), left superior (n = 26; LS), and left inferior (n = 26; LI) PVs (Fig. 1). Four obvious left common PVs and three right middle PVs that were found were excluded from the analysis. The PV ostial diameter was measured in two orthogonal directions (antero-posterior and supero-inferior directions; Fig. 1A). One of the ends of the digital caliper was set at the geometric junction (i.e., inflection or apex of the parabolic outline) between the PV and the intervenal bridging wall for the supero-inferior measurement or anterior wall of the left atrium (LA) or LA appendage for the antero-posterior measurement (Fig. 1A). The intersection of the opposite wall with a line perpendicular to the trunk axis running from the aforementioned junction was used as the other end for the measurement (Fig. 1A).

When a luminal reduction inside the PV was visualized on the curved MPR image, the severity of the luminal reduction was assessed as the percent diameter reduction inside the PV with a referential diameter that was the average of the distal and proximal peri-lesion diameters (Fig. 1B). After measuring the luminal reduction in two orthogonal directions, the larger of the percent diameter reduction values was used as the severity of the luminal reduction.We set 0% for the percent diameter reduction of the PVs without a luminal reduction. A luminal reduction ≥25% was defined as PV narrowing. We distinguished a de novo luminal reduction in the ablated PV from the pre-existent luminal reduction that was visualized both before and after ablation using the bidirectional curved MPR. In the PV without a pre-existent luminal reduction, we determined the luminal reduction that could be visualized after ablation to be a de novo luminal reduction. In the PVs with a pre-existent luminal reduction, an alteration ≥2 mm in the distance from the PV ostium to the narrowest point, conversion to tandem lesions from a single lesion, conversion to a bidirectional lesion from a unidirectional lesion, or increase ≥15% in the luminal reduction was defined as emergence of a de novo luminal reduction. The narrowest portion of the de novo luminal reduction was defined as the ablation site. The distance between the estimated ablation site and the PV ostium × before ablation was calculated as (distance between the point of the maximal diameter reduction and the PV ostiumlength of the PV trunk before ablation/length of the PV trunk after ablation). The diameter of the estimated ablation site before ablation was measured on the first CT image. The percent change in the diameter of the ablation site also was calculated as (100 – 100 × diameter of the ablation site after ablation/diameter of the estimated ablation site before ablation).

The length of the PV trunk, which was defined as the distance from the ostium to the first branching point of the PV, was measured in the coronal sectional view (Fig. 1A). The antero-posterior, supero-inferior, and transverse diameters of the LA were also measured before and after ablation (Fig. 1C). The transverse diameter of the LA was measured between the ostia of the superior PVs on the oblique coronal section. The antero-posterior diameter was measured at the level of the sino-tubular junction on the oblique sagittal section. The longitudinal diameter was measured from the roof to the mitral annulus on the oblique sagittal section. The percent diameter reduction of the PV ostium and LA was calculated as (100 – 100 × diameter before ablation/diameter after ablation).

All ablation sites were verified using multi-plane fluoroscopy, and each ablation site was assessed using the 12 divided segments of the PV wall (Fig. 4A), and the extent of the circumferential ablation area and number of RF energy deliveries in the four anatomic segments of the PV (superior, inferior, anterior, and posterior segments; Fig. 4B) were obtained.

(A)Arepresentation of the pulmonary vein (PV) divided into 12 segments for calculating the circumferential ablation region. (B) A representation of the PV divided into four segments to assess the number of RF energy deliveries at the four anatomic segments of the PV. (C) Total number of RF energy applications at the superior and inferior segments (S + I) and at the anterior and posterior (A + P) segments within the PV. LI (S) = left inferior (superior); RI (S) = right inferior (superior).

Continuous variables are expressed as mean ± SD. Continuous variables were compared with the t-test or one-way ANOVA coupled with Scheffe's test, as appropriate. Categorical variables were compared using the Fisher's exact test. Correlations between variables were assessed by Pearson's linear correlation and tested using Fisher's z transformation. All statistical analyses were carried out using the Stat-View statistical package, version 5.0 (Abacus Concept, Inc., Berkeley, CA, USA). A P value < 0.05 was considered statistically significant.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: