Pulmonary Vein Stenosis After Catheter Ablation of Atrial Fibrillation Using a Cryoballoon, Hot Balloon, or Laser Balloon

Abstract

Background: Pulmonary vein stenosis (PVS) after PV isolation (PVI) for atrial fibrillation (AF) is a severe complication that requires angioplasty. This study aimed to compare the reduction of the cross-sectional PV area (PVA) and the incidence of PVS after cryoballoon (CB)-PVI, hot balloon (HB)-PVI, or laser balloon (LB)-PVI.

Methods and Results: A total of 320 patients who underwent an initial catheter ablation procedure for AF using a CB, HB, or LB in 2 hospitals were included. They underwent contrast-enhanced multidetector CT before and 3 months after the procedure. In all 4 PVs, the reduction in PVA was more significant in the LB group than in the CB or HB groups, respectively. Moderate (50–75%) and severe (>75%) PVS were observed in 5.3% and 0.5% of the PVs, respectively. Although moderate PVS was more frequently observed in the LB group than in the CB or HB groups (8.2%, 3.8%, and 5.0%; P=0.03), the incidence of severe PVS was similar in the LB, CB, and HB groups (0.3%, 0.5%, and 1.0%; P=0.46). Symptomatic PVS requiring intervention occurred in 1 (0.3%) patient.

Conclusions: Although the reduction in cross-sectional PVA and the incidence of moderate PVS after LB-PVI was more significant than after CB-PVI or HB-PVI, it rarely led to severe PVS. Symptomatic PVS requiring intervention was rare after the balloon ablation of AF.

Although pulmonary vein isolation (PVI) has been established as a cornerstone treatment for atrial fibrillation (AF),¹ PV stenosis (PVS) after PVI remains a major complication requiring intervention and is associated with significant morbidity.^2–5 As the performance of wide encircling PVI using radiofrequency (RF) catheters spread around the world, the incidence of PVS has decreased. However, with the recent development of various balloon devices (e.g., cryoballoon [CB], hot balloon [HB], and laser balloon [LB]), PVS has increased again, because segmental PVI is required. After CB-PVI, a previous meta-analysis reported that the incidence of significant PVS resulting in symptoms or requiring intervention was 0.17%.⁶ The incidence of moderate-severe (>50% reduction in cross-sectional PV area [PVA]) and severe (>70–75% reduction in cross-sectional PVA) PVS after CB-PVI was reported to be 1.9–5.2% and 0.6–1.3%, respectively.^7–9 After HB-PVI, the incidence of PVS (reduction of PV diameter >70%) was 3.6–5.2%.^10,11 After LB-PVI, the incidence of moderate-severe and severe PVS was 27% and 2.1%, respectively.¹² The incidence of PVS varied among the types of ablation devices. However, no study has used a uniform measurement method to compare the incidence of PVS among several balloon devices. This study aimed to compare the PVA reduction and PVS incidence after catheter ablation of AF using a CB, HB, or LB.

Methods

Study Population

From 2017 to 2021, 320 patients underwent initial catheter ablation for AF using a CB, HB, or LB for AF in 2 university hospitals. Patients underwent contrast-enhanced multidetector computed tomography (MDCT) before and 3 months after the procedure. We did not use the balloon devices for patients with a common PV trunk or right middle PV. The decision to use each ablation device was based on the preference of each operator. All antiarrhythmic drugs were discontinued at least 5 half-lives before the procedure. All patients provided written informed consent before undergoing the procedure. Major complications were defined as procedure-related adverse events that required therapeutic intervention causing long-term disability or resulting in prolonged hospitalization. Clinical investigations were conducted according to the principles expressed in the Declaration of Helsinki. The studies and data collection were conducted according to protocols approved by the Human Research Committee of the Jikei University School of Medicine (34-048(11193)).

CB Ablation

The details of CB ablation were previously reported.¹³ PVI was performed for 180 s with a single-freeze technique using a 28-mm CB (Arctic Front Advance pro; Medtronic). A spiral mapping catheter (Achieve; Medtronic) was used to advance the CB and map the PV potentials. If electrical isolation was not achieved with 3 CB applications per vein, additional touch-up ablation was performed with an RF catheter. RF energy was delivered using an open irrigated-tip ablation catheter with a 30–35 W power limit.

HB Ablation

An HB (SATAKE HotBalloon; Toray Industries) with an inner lumen and J-tip guidewire was inflated at each PV ostium through a deflectable guiding sheath (Treswaltz; Toray Industries). For PV occlusion, the HB was inflated to 26–33 mm in diameter with 10–20 mL of contrast medium diluted 1 : 1 with saline. Once optimal PV occlusion, as assessed by contrast injection, was achieved, an RF current of 1.8 MHz was applied. The target internal balloon temperature of 70℃ was maintained by delivering vibratory waves through the lumen into the balloon to agitate the fluid inside. RF-generated thermal energy was applied to the left superior (LS) PV antrum for 240 s the left inferior (LI) PV antrum for 150 s, the right superior (RS) PV antrum for 240 s, and the right inferior (RI) PV antrum for 150 s.

LB Ablation

As previously described,¹⁴ a 1st-generation (HeartLight, CardioFocus) or 3rd-generation (HeartLight X3, CardioFocus) LB was filled and continuously flushed with heavy water and introduced into the left atrium (LA) via a deflectable transseptal sheath. The LB was positioned at the ostium of the target PV, and the balloon was inflated. Laser energy was applied for 20 or 30 s depending on the preselected power (5.5–12 W). Encircling lesions around individual PVs were deployed point-by-point with 30–50% overlap, whereas a single ablation lesion covered 30° of the circle. For X3 procedures, RAPID mode, manual mode, or their combination was used. After placement of the initial anatomically guided encircling lesion set, the circular mapping catheter was used to assess for electrical isolation of the PV. If residual PV potential was observed after the HB or LB application, RF touch-up ablation was performed.

Evaluation of PV Morphology

MDCT was performed before and 3 months after the procedure to evaluate the PV morphology as previously described.⁷ Three-dimensional images of the LA and PVs were reconstructed from MDCT images on a 3D electroanatomical mapping system (Ensite NavX^TM, St. Jude Medical). The cross-sectional PVA was measured on the 3D electroanatomical mapping system by tracing the cross-sectional area within the PV plane at 5-mm intervals from the PV ostium in a distal direction until 20 mm (5 segments) or the bifurcation of each PV (Figure 1). The measurement was performed by 2 independent experts who were unaware of the procedural results. The PVA reduction was calculated at each segment as: (1-PVA_post / PVA_pre) × 100 (%). Moderate and severe PVS was defined as a 50–74% and 75–100% reduction, respectively, of the maximum PVA among 5 segments.^7,8 The average PVA was defined as the average value of the cross-sectional areas of the PVs at 5 locations.

Figure 1.

Evaluation of the cross-sectional PVA. Typical example of measurement and the method of calculating the reduction in PVA. The PVA was measured before and after catheter ablation at 5-mm intervals from the PV ostium in a distal direction for 20 mm. The PVA reduction was calculated at each point of the PV using the following formula: (1-PVA_post / PVA_pre) × 100 (%). The PVA reduction in each PV was defined as the most significant PVA reduction among the 0 mm (1), 5 mm (2), 10 mm (3), 15 mm (4), and 20 mm points (5) from the ostium. PV, pulmonary vein; PVA, PV area.

Statistical Analysis

Continuous variables are expressed as the mean±standard deviation or median (interquartile range). An unpaired Student’s t-test or the Mann-Whitney U test was used for continuous variables. Categorical variables, expressed as numbers or percentages, were analyzed using the chi-squared test unless the expected values in any cells were <5, in which cases Fisher’s exact test was used. P values <0.05 were considered to indicate statistical significance. A one-way ANOVA or Kruskal analyzed data in three or more groups–Wallis test, with the Tukey or Steel-Dwass post hoc test, as appropriate. The inter-rater agreement statistic (kappa coefficient) was used to evaluate the agreement between classifications. All statistical analyses were performed using the SPSS software program (version 27; SPSS, Chicago, IL, USA).

Results

Patients’ Characteristics

The baseline patient characteristics are shown in Table 1. The mean age was 61.2±10.3 years, the mean LA diameter was 37.6±5.4 mm, and 90.6% of patients had paroxysmal AF. The characteristics of the 3 groups were similar. Before the procedure, the average PVA in all PVs was similar among the 3 groups.

Table 1. Patients’ Characteristics and Details of the Procedures

	Total	Cryoballoon (n=153)	Hot balloon (n=75)	Laser balloon (n=92)	P value
Age (years)	61.2±10.3	61.7±10.9	61.0±10.1	61.2±10.3	0.71
Sex (male %)	248 (77.5)	117 (76.4)	54 (72.0)	77 (83.6)	0.18
Body mass index (kg/m2)	24.3±3.4	24.6±3.3	23.9±3.1	24.2±3.7	0.38
Paroxysmal AF, n (%)	290 (90.6)	133 (86.9)	71 (94.7)	86 (93.4)	0.09
AF history (years)	3.3±4.9	3.3±6.0	3.5±3.7	3.3±4.9	0.92
LA diameter (mm)	37.6±5.4	38.0±5.5	37.5±5.7	37.0±4.9	0.43
LV ejection fraction (%)	64.8±6.0	64.8±6.4	64.8±5.6	65.1±5.5	0.93
eGFR(mL/min/1.73 m2)	72.6±14.9	72.6±13.5	72.9±13.8	72.3±17.6	0.97
BNP (pg/mL)	56.2±63.9	64.9±177.1	56.2±76.6	42.1±41.5	0.13
Hypertension, n (%)	140 (43.8)	73 (47.7)	30 (40.0)	37 (40.2)	0.39
Diabetes mellitus, n (%)	29 (9.1)	16 (10.4)	8 (10.6)	5 (5.4)	0.36
CHA2DS2-VASc score	1 (0–2)	1 (0–2)	1 (0–2)	1 (0–2)	0.09
Average PVA before ablation (cm2)
LSPV	2.3±0.6	2.3±0.7	2.2±0.4	2.4±0.6	0.36
LIPV	1.8±0.4	1.7±0.4	1.7±0.4	1.8±0.4	0.17
RSPV	2.5±0.7	2.6±0.7	2.4±0.6	2.5±0.7	0.24
RIPV	2.2±0.6	2.2±0.6	2.2±0.5	2.1±0.5	0.71
Total procedure time (min)	164.4±51.6	132.3±36.9*,†	178.2±43.3*,‡	207.9±42.0†,‡	<0.001
Isolation area (%)	36.9±11.1	33.3±10.8	39.2±18.8	37.8±11.0	0.07

Data are presented as the mean±standard deviation, median (interquartile range), or n (%). *^,†,‡Pairs of values with a significant difference in post hoc test. AF, atrial fibrillation; BNP, B-type natriuretic peptide; eGFR, estimated glomerular filtration rate; LA, left atrium; LI, left inferior; LS, left superior; LV, left ventricle; PV, pulmonary vein; PVA, PV area; RI, right inferior; RS, right superior.

Ablation Results

All PVs were successfully isolated from the LA during the procedure. RF touch-up ablation was required in 23/612 (3.8%) PVs in the CB group, 86/300 (28.7%) PVs in the HB group, and 65/368 (17.7%) PVs in the LB group. For the PVs requiring touch-up ablation, the mean number of touch-up ablations was 2.3±1.6/PVs. The total procedure time in the CB group was shorter than in the HB or LB group. Major complications occurred during the procedure in 3.8% of patients, with a similar incidence among the 3 groups (P=0.33). Neither pericardial effusion nor cardiac tamponade occurred. Phrenic nerve injury occurred in 6 (3.9%) patients in the CB group and in 3 (3.3%) in the LB group. Although all phrenic nerve injuries (6/6) due to CB-PVI recovered within 6 months after the procedure, 2 of 3 phrenic nerve injuries due to LB-PVI did not heal during the follow-up period. Cerebral embolism with transient minor symptoms occurred in 2 (1.3%) patients in the CB group and in 1 (1.3%) patient in the HB group.

Reduction of PVA Following Catheter Ablation

An excellent interobserver correlation was observed in the measurement of PVA (kappa coefficient, 0.89). The median reduction in PVA after CB-PVI, HB-PVI, and LB-PVI was 15 (8.3–23) %, 15 (6.8–25) %, and 25 (15–37) %, respectively (Figure 2). In all 4 PVs, the reduction in PVA was more significant in the LB group than in the CB or HB groups. After excluding 174 (14%) PVs with RF touch-up ablations, the reduction in PVA was still more significant in the LB group than in the CB or HB groups. The location of maximum PVA reduction was more proximal in the LB group than in the CB and HB groups (Table 2). At 3 months after the procedure, the LA diameter had decreased from 37.6±5.4 to 36.2±5.1 mm. There was no correlation between the reduction in LA diameter and the PVA in any of the 4 PVs (LSPV r=0.011, LIPV r=0.001, RSPV r=0.072, and r=0.016).

Figure 2.

Reduction in cross-sectional PVA. Box plots represent the interquartile range of values, horizontal lines represent the median, and whiskers indicate the maximum and minimum range, excluding outliers. Black dots are outliers at least twice the interquartile range (25–75%) from the median. There was a more significant reduction in PVA in the LB group than in the CB or HB groups in total and for each PV. CB, cryoballoon; HB, hot balloon; LB, laser balloon; PV, pulmonary vein; PVA, PV area.

Table 2. Location of the Maximum PVA Reduction

Distance from the PV ostium	Cryoballoon	Hot balloon	Laser balloon	P value
0 mm	309 (50.9)	156 (52.2)	241 (74.6)	<0.001
5 mm	171 (28.2)	83 (27.8)	68 (21.1)
10 mm	78 (12.9)	36 (12.0)	11 (3.4)
15 mm	37 (6.1)	15 (5.0)	3 (0.9)
20 mm	12 (2.0)	9 (3.0)	0 (0)

Data are presented as n (%). PV, pulmonary vein; PVA, PV area.

In the LB group, 24 patients underwent PVI using a 3rd-generation LB. The number of LB applications was less, the total LB application time was shorter, and the maximum power was higher in the 3rd-generation LB group than in the 1st-generation LB group (Supplementary Table). The median reduction in PVA was more significant, and the incidence of moderate PVS was higher in the 3rd-generation LB group than in the 1st-generation LB group [29 (20–45) % vs. 23 (13–35) %; P=0.03, 16.0% vs. 6.1%, P=0.004]. However, the incidence of severe PVS was not different between the LB groups.

Patients With Moderate and Severe PVS

Moderate (50–75% reduction in PVA) and severe PVS (>75% reduction in PVA) were observed in 5.3% and 0.5% of PVs, respectively. The incidence of moderate PVS was higher in the LB group than in the CB or HB groups (8.2%, 3.8%, and 5.0% in the LB, CB, and HB groups, respectively; P=0.03; Figure 3). On the other hand, the incidence of severe PVS was similar among the 3 groups (0.3%, 0.5%, and 1.0% in the LB, CB, and HB groups, respectively; P=0.46). All cases of severe PVS occurred within 5 mm of the PV ostium (Site 1 or 2 in Figure 1). The mean age of the 7 patients with severe PVS was 56.0±9.4 years and all were male. The patients characteristics were similar among those without PVS, with moderate PVS, and with severe PVS (Table 3). A voltage map was created in 127 patients. The % PVI area was calculated as previously reported¹⁵ and was not significantly different between patients with or without PVS.

Figure 3.

Incidence of severe and moderate PV stenosis. CB, cryoballoon; HB, hot balloon; LB, laser balloon; LI, left inferior; LS, left superior; PV, pulmonary vein; RS, right superior.

Table 3. Patients With and Without Severe PVS

	PVS (−) (<50%) (n=264)	Moderate PVS (50–75%) (n=48)	Severe PVS (75%<) (n=7)	P value
Age (years)	61.0±10.3	63.3±10.2	56.0±9.4	0.14
Sex (male)	202 (76.2)	39 (81.3)	7 (100)	0.26
Body mass index (kg/m2)	24.5±3.4	23.4±3.1	25.6±4.2	0.10
AF history (years)	3.2±4.9	3.8±4.2	5.1±8.9	0.49
LA diameter (mm)	37.7±5.4	36.9±5.3	36.6±5.2	0.57
LV ejection fraction (%)	64.7±6.1	65.4±5.0	65.8±6.0	0.69
eGFR(mL/min/1.73 m2)	72.7±15.1	71.9±15.0	74.2±14.9	0.91
BNP (pg/mL)	58.1±67.2	47.8±46.4	41.6±36.5	0.49
Hypertension, n (%)	114 (43.0)	23 (47.9)	3 (42.9)	0.82
Diabetes mellitus, n (%)	25 (9.4)	4 (8.3)	0 (0)	0.68
CHA2DS2-VASc Score	1.1±1.2	1.1±1.3	0.7±1.0	0.51
Total procedure time (m)	163.5±53.3	170.5±43.4	157.1±37.6	0.67
Isolation area (%)	36.8±11.1	36.8±11.5	40.2±9.0	0.84

Data are presented as mean±standard deviation, median (interquartile range), or n (%). PVS, pulmonary vein stenosis. Other abbreviations as in Table 1.

Severe LSPVS occurred in 2 (1.3%) patients in the CB group, 1 (1.3%) patient in the HB group, and 1 (1.1%) patient in the LB group (Figure 4). Severe RSPVS occurred in 1 (0.7%) patient in the CB group, and severe RIPVS occurred in 2 (2.7%) patients in the HB group. No patients developed severe PVS in the LIPV. Only 1 patient (0.3%) complained of symptoms associated with PVS. The patient had chest pain and hemoptysis 6 months after HB-PVI and underwent emergency PV intervention. The patient’s clinical course with PVS after HB ablation has been reported previously.^4,16 CT revealed severe LSPVS and balloon angioplasty (balloon size: 8*20 mm) was performed. However, the LSPV showed restenosis 4 months later. Balloon angioplasty with stenting (bare-metal stent: 6*18 mm) for LSPV restenosis was performed and 1 year after PV stenting, angioplasty was performed for the 3rd time (balloon size: 8*20 mm) because of intimal growth in the stent. After that, the patient continued with aspirin 100 mg QD and rivaroxaban 15 mg QD. MDCT performed 2 years after the final angioplasty revealed the absence of restenosis.

Figure 4.

Details of 7 patients with severe PV stenosis (PVS) on postoperative computed tomographic images. LI, left inferior; LS, left superior; PV, pulmonary vein; PVA, PV area; RS, right superior.

Repeat ablation was performed in 28 patients. The reduction in PVA was similar between the PVs with and without reconnection (18.4±18.3% vs. 12.9±15.0%, P=0.12).

Predictors of PVS

The periprocedural parameters of the CB, HB, and LB groups were compared among PVs without PVS, with moderate PVS, and with severe PVS (Table 4). RF touch-up ablation was not related to the incidence of PVS. In the CB group, preprocedural ostial PVA and freezing parameters were similar among the 3 groups. The total HB application time was longer for PVs with severe stenosis than for PVs without stenosis (P=0.03) or with moderate stenosis (P=0.01). In the LB group, a 3rd-generation LB was more frequently used in PVs with moderate stenosis than for those without stenosis. In the 3rd-generation LB group, the preprocedural ostial PVA in PVs with and without moderate PVS did not differ to a statistically significant extent (2.4 vs. 2.5 cm², P=0.39).

Table 4. Predictors of PVS

	PVS (−) (<50%) (n=1,205)	Moderate PVS (50–75%) (n=68)	Severe PVS (75%<) (n=7)	P value
Cryoballoon (%)	586 (95.8)	23 (3.8)	3 (0.5)
Ostial PVApre	2.5±0.8	2.2±1.2	1.6±0.8	0.09
RF touch-up (%)	24 (4.1)	1 (4.3)	0 (0)	0.94
No. of CB application	1.2±0.5	1.0±0.2	1.3±0.6	0.25
Nadir temperature (℃)	−48.9±6.2	−47.1±6.0	−50.7±6.2	0.39
Total freezing time (s)	202.1±65.9	195.2±46.1	240±60.0	0.54
Thawing time (s)	45.7±15.2	42.0±13.8	57.7±7.6	0.22
Time to isolation (s)	38.0±21.6	41.2±21.6	40.7±21.5	0.82
HB	282 (94.0)	15 (5.0)	3 (1.0)
Ostial PVApre	2.3±0.6	2.2±0.7	2.5±0.6	0.78
RF touch-ups (%)	80 (28.4)	6 (40.0)	0 (0)	0.34
No. of HB application	1.2±0.4	1.0±0.0*	1.7±1.2*	0.04
Total HB application time (s)	209.2±73.7†	167.4±60.8‡	323.0±234.7†,‡	0.007
HB Inflation volume (mL)	11.3±2.1	10.8±1.0	10.7±1.2	0.64
LB	337 (91.6)	30 (8.2)	1 (0.3)
Ostial PVApre	2.5±0.6	2.4±0.5	2.5	0.71
RF touch-up (%)	57 (16.9)	7 (23.3)	1 (100)	0.07
No. of LB application	18.4±8.3	15.9±10.9	26	0.21
Total LB application time (s)	389.6±193.4	362.3±184.7	500	0.64
LB Inflation volume	5.4±1.5	5.1±1.1	5	0.63
LB maximum power (W)	11.6±1.2	12.2±1.0	12	0.04
3rd-generation LB (%)	81 (24.0)¶	15 (50.0)¶	0 (0)	0.007

Data are presented as mean±standard deviation, median (interquartile range), or n (%). *^,†,‡,¶Pairs of values with a significant difference in post hoc test. CB, cryoballoon; HB, hot balloon; LB, laser balloon; PVS, pulmonary vein stenosis; RF, radiofrequency.

Discussion

Main Findings

This is the first study to compare the incidence of PVS after CB, HB, or LB ablation of AF. The following key findings were identified.

1) The reduction in PVA was more significant after LB-PVI than after either CB-PVI or HB-PVI.

2) Severe and moderate PVS occurred in 0.5% and 5.3% of the PVs, respectively. Although moderate PVS (50–75% reduction of PVA) was more frequent after LB-PVI than after CB-PVI or HB-PVI, the incidence of severe PVS (>75% reduction of PVA) was similar among the 3 groups.

3) Symptomatic PVS requiring intervention was rare (0.3%).

4) The total HB application time was associated with severe PVS. Using a 3rd-generation LB was related to moderate PVS but not severe PVS.

Measurement of Cross-Sectional PVA

Several previous studies had reported the incidence of PVS after PVI for AF; however, the assessment of PVS varies (e.g., the PV ostial diameter on PV angiography;¹⁷ the PV ostial diameter on MDCT;^10,11 major and minor axes of the PV ostium on MDCT^9,12). We previously reported a method of evaluating the PV morphology more accurately.^7,8 By measuring the PVA every 5 mm from the PV ostium, the exact location on pre- and postoperative CT images can be compared.

Incidence of PVS

PVS most frequently occurred in the LSPV. The number of touch-up ablations was similar among the 4 PVs (P=0.19). On review of previous studies, post-ablative PVS most frequently occurred in the LSPV,^2,4,8 but the mechanism has not been clarified. The larger amount of myocardium at the LSPV ostium than with the other PVs may result in a more robust postoperative local inflammatory response.

With a better understanding of the risks of PVS and ablation techniques, including wide circumferential ablation, and antral isolation, the incidence of severe PVS has decreased. A previous multicenter study demonstrated that moderate PVS after circumferential RF-PVI was identified in 0.9% of PVs and that none of the PVs had severe stenosis.¹⁸ In the current study, the incidence of moderate and severe PVS after CB-PVI was 5.3% and 0.5%, respectively, which is consistent with previous reports.^7–9 As was observed for RF-PVI, a more proximal PVI is desirable for preventing PVS. Several freezing parameters did not correlate with the incidence of PVS after CB-PVI.

After HB-PVI, the incidence of moderate and severe PVS was 5.0% and 1.0%, which aligned with previous reports.^10,11 The incidence rates were similar to those after CB-PVI. The extra damage due to the prolonged application time increased the risk of PVS. RF touch-up ablations were frequently performed in the HB group in this study. However, because the number of touch-up ablation procedures in each case was small, touch-up ablation may have had little effect on the results. The incidence of PVS was similar when PVs with RF touch-up ablation were excluded.

After LB-PVI, the incidence of moderate and severe PVS was 8.2% and 0.3%, respectively, which was lower than previously reported (27%).¹² The reduction in PVA in the LB group was more significant than in the CB or HB groups. Moderate PVS was relatively common after LB-PVI; however, cases leading to severe PVS were rare. The location of maximum PVA reduction was more proximal in the LB group than in the CB and HB groups. The CB and HB can unexpectedly form ablative lesions far distal to the PV, where tissue is more vulnerable. This may be a factor in why the incidence of severe stenosis with LB was comparable to that for the other balloons.

Mechanisms of PVS

The PV frequently demonstrate negative remodeling following AF ablation, with an average reduction of 5.1–9.8% in the orifice area.¹⁹ A previous study reported a significant correlation between the rate of PVA reduction and the rate of LA volume reduction.⁹ A subset of PVs with mild PVS on MDCT may represent this negative remodeling process or volume changes. However, an apparent “waist-like shrinkage” was typically seen in PVs with moderate or severe narrowing. The width of the stenotic segment was usually narrow. This was a different process from negative remodeling. There were 14 patients with moderate or severe PVS of ≥2 PVs. Most such patients were male and relatively young. Other than the intraoperative findings, predisposing factors, such as being prone to inflammation and connective tissue abnormalities, may also be risk factors for PVS.

The reduction in PVA after LB-PVI was more significant than after CB-PVI or HB-PVI. A previous study demonstrated that changes in inflammatory markers (white blood cell count and C-reactive protein level) after LB-PVI were significantly greater than after CB-PVI.²⁰ In particular, laser energy is presumed to form more definite lesions in the PV-LA junction compared with cryothermal energy.²¹ Such local and powerful energy delivery may cause the waist-like shrinkage that is thought to be related to fibrosis at the site of LB ablation. The use of a 3rd-generation LB was associated with the incidence of moderate PVS but not severe PVS. A previous study reported that lesion depth and width following ablation using a 3rd-generation LB in the RAPID mode were smaller than after ablation using a 1st-generation LB.²² Due to the smaller lesion size when the RAPID mode is used, the use of the LB can be expected to reduce the risk of PVS. The results of the present study were not in line with this expectation. A 3rd-generation LB can create a continuous lesion, and the postoperative PV shrinkage may be more apparent because fewer gaps are present. Moreover, the laser beam uniformly goes around any size of PV in 3 min in RAPID mode, greater amounts of local energy may be delivered in PVs with a shorter circumference. However, it should be noted that using a 3rd-generation LB did not increase the incidence of severe PVS. A 3rd-generation LB was used only in 24 patients, so the sample size was too small for any conclusions to be drawn from a comparison of 1st- and 3rd-generation LBs.

Prognosis of Severe PVS

Although severe PVS occurred in 7 (0.5%) PVs, PV intervention was required in only 1 symptomatic patient after HB-PVI. Symptomatic PVS was rare after CB, HB, or LB-PVI. However, once it occurs, without semi-urgent intervention it becomes a serious situation with severe, life-threatening symptoms. Restenosis often occurs repeatedly, even after intervention. Using stents provides superior long-term PV patency compared with balloon angioplasty alone.^2,3,23 A previous multivariate analysis demonstrated a significantly lower risk of restenosis in PVs treated with a stent (hazard ratio: 2.84).²³

How to Prevent PVS

During PVI, it is essential to minimize the trauma to the PVs, which includes avoiding excessive damage and limiting the number of ablations performed on the PVs. A thorough evaluation should be performed before PVI to identify the risk factors for PVS. This includes CT or MRI to assess the anatomy and size of the PVs. The choice of energy source can affect the incidence of PVS. The risk of PVS can be minimized by selecting the optimal device for the morphology and size of the PVs. The procedure should be performed by an experienced operator familiar with the potential complications and their management.

In cases involving use of a CB, the proximal sealing technique or the pre-freezing technique results in a more proximal balloon position.¹⁵ More proximal occlusions with decreased contact between the CB and PV antrum prevent excessive cryothermal application, which can be associated with a reduced risk of PVS. For large-diameter PVs, the CB tends to travel more deeply, so either other devices should be selected or segmental nonocclusive CB ablation should be performed.²⁴ Unlike CBs, HBs, and LBs have variable balloons and can adjusted for more proximal ablation. The novel function of measuring the surface temperature of HBs may reduce the risks of excessive ablation and thus PVS. Using appropriate energy settings during LB ablation is essential for preventing PVS. The energy settings should be selected based on the size of the PVs and the thickness of their walls.

Study Limitations

This was a retrospective and non-randomized study. The specific balloon used for each participant was left to the operator’s discretion, which may introduce potential selection bias. Although preprocedural PV size was similar among the 3 groups, preoperative CT scan results may have influenced device selection. The actual ablation site was challenging to assess with CT. MRI may be better for clarifying the relationship between the actual ablation and stenosis sites. Because the incidence of severe PVS was low, a multivariate analysis could not be performed to identify independent risk factors for PVS. Because the LB is a more recent device than the CB and HB, operators are likely to have less experience than with other devices. Further larger-scale multicenter studies are necessary to reveal the predictors of severe PVS after balloon ablation of AF.

Conclusions

The reduction in the cross-sectional PVA was more significant, and the incidence of moderate PVS was more frequent after LB-PVI than after CB- or HB-PVI. However, it rarely led to severe PVS. Symptomatic PVS requiring intervention was also rare.

Acknowledgment

We thank Dr. Brian Quinn (Japan Medical Communication Inc.) for his comments on the manuscript’s language.

Disclosures

M.T. received a speaker’s honorarium and consulting fees from Medtronic, and research funding from Japan Life Line. T.Y. received a speaker’s honorarium from Medtronic and Abbott Medical. M. Yoshimura is a member of the Circulation Journal’s Editorial Team. The other authors declare that there are no conflicts of interest.

IRB Information

The studies and data collection were conducted according to protocols approved by the Human Research Committee of the Jikei University School of Medicine (34-048(11193)).

Data Availability

The deidentified participant data will not be shared.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-23-0048

References

• 1.   Haïssaguerre M, Jaïs P, Shah DC, Takahashi A, Hocini M, Quiniou G, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998; 339: 659–666, doi:10.1056/NEJM199809033391003.
• 2.   Fender EA, Widmer RJ, Hodge DO, Cooper GM, Monahan KH, Peterson LA, et al. Severe pulmonary vein stenosis resulting from ablation for atrial fibrillation: Presentation, management, and clinical outcomes. Circulation 2016; 134: 1812–1821, doi:10.1161/CIRCULATIONAHA.116.021949.
• 3.   Fender EA, Widmer RJ, Hodge DO, Packer DL, Holmes DR Jr. Assessment and management of pulmonary vein occlusion after atrial fibrillation ablation. JACC Cardiovasc Interv 2018; 11: 1633–1639, doi:10.1016/j.jcin.2018.05.020.
• 4.   Ogawa T, Yamashita S, Oseto H, Yokoyama M, Itakura R, Kang R, et al. Pulmonary vein angioplasty for pulmonary vein stenosis after ablation therapy for atrial fibrillation: A report of 7 cases. Circ J 2022; 86: 1229–1236, doi:10.1253/circj.CJ-22-0162.
• 5.   Fink T, Vogler J, Proietti R, Sciacca V, Heeger CH, Rottner L, et al. Antithrombotic therapy after angioplasty of pulmonary vein stenosis due to atrial fibrillation ablation: A two-center experience and review of the literature. J Arrhythm 2022; 38: 1009–1016, doi:10.1002/joa3.12777.
• 6.   Andrade JG, Khairy P, Guerra PG, Deyell MW, Rivard L, Macle L, et al. Efficacy and safety of cryoballoon ablation for atrial fibrillation: A systematic review of published studies. Heart Rhythm 2011; 8: 1444–1451, doi:10.1016/j.hrthm.2011.03.050.
• 7.   Narui R, Tokuda M, Matsushima M, Isogai R, Tokutake K, Yokoyama K, et al. Incidence and factors associated with the occurrence of pulmonary vein narrowing after cryoballoon ablation. Circ Arrhythm Electrophysiol 2017; 10: e004588, doi:10.1161/CIRCEP.116.004588.
• 8.   Tokutake K, Tokuda M, Yamashita S, Sato H, Ikewaki H, Okajima E, et al. Anatomical and procedural factors of severe pulmonary vein stenosis after cryoballoon pulmonary vein ablation. JACC Clin Electrophysiol 2019; 5: 1303–1315, doi:10.1016/j.jacep.2019.08.003.
• 9.   Miyama H, Takatsuki S, Hashimoto K, Yamashita T, Fujisawa T, Katsumata Y, et al. Anatomical changes in the pulmonary veins and left atrium after cryoballoon ablation. Pacing Clin Electrophysiol 2020; 43: 1289–1294, doi:10.1111/pace.14092.
• 10.   Sohara H, Ohe T, Okumura K, Naito S, Hirao K, Shoda M, et al. HotBalloon ablation of the pulmonary veins for paroxysmal AF: A multicenter randomized trial in Japan. J Am Coll Cardiol 2016; 68: 2747–2757, doi:10.1016/j.jacc.2016.10.037.
• 11.   Nakamura Y, Sohara H, Ihara M. Pulmonary vein stenosis after HotBalloon pulmonary vein isolation of paroxysmal atrial fibrillation. Heart Vessels 2021; 36: 1739–1745, doi:10.1007/s00380-021-01862-7.
• 12.   Yamamoto T, Takahashi Y, Yamaguchi J, Sekigawa M, Shirai Y, Tao S, et al. Pulmonary vein narrowing after visually guided laser balloon ablation: Occurrence and clinical correlates. J Cardiovasc Electrophysiol 2020; 31: 1597–1605, doi:10.1111/jce.14525.
• 13.   Tokuda M, Matsuo S, Isogai R, Uno G, Tokutake K, Yokoyama K, et al. Adenosine testing during cryoballoon ablation and radiofrequency ablation of atrial fibrillation: A propensity score-matched analysis. Heart Rhythm 2016; 13: 2128–2134, doi:10.1016/j.hrthm.2016.08.018.
• 14.   Schmidt B, Petru J, Chun KRJ, Sediva L, Bordignon S, Chen S, et al. Pivotal study of a novel motor-driven endoscopic ablation system. Circ Arrhythm Electrophysiol 2021; 14: e009544, doi:10.1161/CIRCEP.120.009544.
• 15.   Tokuda M, Yamashita S, Sato H, Oseto H, Yokoyama M, Tokutake K, et al. Using the pre-freezing technique results in a wider area of antral isolation during cryoballoon pulmonary vein isolation in patients with atrial fibrillation. Heart Vessels 2023; 38: 413–421, doi:10.1007/s00380-022-02179-9.
• 16.   Yamashita S, Ogawa T, Yoshimura M, Yamane T. Severe pulmonary vein stenosis requiring angioplasty after hot balloon ablation for pulmonary vein isolation. Heart Rhythm Case Rep 2020; 6: 469–470, doi:10.1016/j.hrcr.2020.04.001.
• 17.   Nagase T, Bordignon S, Perrotta L, Bologna F, Weise FK, Konstantinou A, et al. Low risk of pulmonary vein stenosis after contemporary atrial fibrillation ablation: Lessons from repeat procedures after radiofrequency current, cryoballoon, and laser balloon. Circ J 2018; 82: 1558–1565, doi:10.1253/circj.CJ-17-1324.
• 18.   Samuel M, Khairy P, Mongeon FP, Andrade JG, Gomes S, Galvan Z, et al. Pulmonary vein stenosis after atrial fibrillation ablation: Insights from the ADVICE trial. Can J Cardiol 2020; 36: 1965–1974, doi:10.1016/j.cjca.2020.10.013.
• 19.   Merchant FM, Levy MR, Iravanian S, Weragoda RM, Clermont EC, Kelli HM, et al. Pulmonary vein remodeling following atrial fibrillation ablation: Implications for the radiographic diagnosis of pulmonary vein stenosis. J Atr Fibrillation 2016; 9: 1453, doi:10.4022/jafib.1453.
• 20.   Yano M, Egami Y, Ukita K, Kawamura A, Nakamura H, Matsuhiro Y, et al. Impact of myocardial injury and inflammation due to ablation on the short-term and mid-term outcomes: Cryoballoon versus laser balloon ablation. Int J Cardiol 2021; 338: 102–108, doi:10.1016/j.ijcard.2021.06.016.
• 21.   Khurram IM, Catanzaro JN, Zimmerman S, Zipunnikov V, Berger RD, Cheng A, et al. MRI evaluation of radiofrequency, cryothermal, and laser left atrial lesion formation in patients with atrial fibrillation. Pacing Clin Electrophysiol 2015; 38: 1317–1324, doi:10.1111/pace.12696.
• 22.   Nagase T, Asano S, Fukunaga H, Kasai Y, Inoue K, Sekiguchi Y, et al. Evaluation of linear lesion formation and thermodynamics by dragging ablation with the third-generation laser balloon. Heart Rhythm O2 2022; 3: 311–318, doi:10.1016/j.hroo.2022.04.001.
• 23.   Widmer RJ, Fender EA, Hodge DO, Monahan KH, Peterson LA, Holmes DR Jr, et al. Contributors toward pulmonary vein restenosis following successful intervention. JACC Clin Electrophysiol 2018; 4: 547–552, doi:10.1016/j.jacep.2017.10.001.
• 24.   Aryana A, Su W, Kuniss M, Okishige K, de Asmundis C, Tondo C, et al. Segmental nonocclusive cryoballoon ablation of pulmonary veins and extrapulmonary vein structures: Best practices III. Heart Rhythm 2021; 18: 1435–1444, doi:10.1016/j.hrthm.2021.04.020.