Relationship between coughing and bispectral index–guided sedation during pulsed field ablation for atrial fibrillation and its impact on mid-term outcomes

Abstract

Background

Coughing can occur during pulsed field ablation (PFA) for pulmonary vein (PV) isolation (PVI) of atrial fibrillation (AF) under deep sedation, and the bispectral index (BIS) is commonly used to monitor sedation depth.

Objective

We aimed to investigate the relationship between coughing and BIS-guided sedation depth during PFA-based PVI and to evaluate the impact of coughing on clinical outcomes after AF catheter ablation.

Methods

We retrospectively studied 198 consecutive patients undergoing an initial AF catheter ablation procedure using a 31-mm pentaspline PFA catheter under deep sedation. The BIS value before PFA for each PV, maximum BIS value among all PVs for each patient, and the presence or absence of coughing were assessed.

Results

Coughing occurred in 260 PVs (32.8%) among 95 patients (48.0%). The incidence of coughing at maximum BIS <40, 40–70, and >70 was 0%, 37.2%, and 55.5%, respectively (P = .013). In a multivariate analysis, higher body mass index and higher maximum BIS value were significant positive predictors of coughing (odds ratios 1.188 and 1.052; P = .002 and P = .012), whereas older age was a negative predictor (odds ratio 0.962; P = .044). After propensity score matching, the arrhythmia-free survival rate at 6 months was 93.3% in patients with coughing and 96.7% in those without, respectively (P = .306).

Conclusion

Coughing during pentaspline PFA catheter–based PVI under deep sedation was associated with higher BIS values, indicating lighter sedation, but had no adverse effect on the mid-term outcomes after the ablation procedure.

Graphical abstract

Key Findings.

• ▪During pentaspline pulsed field ablation (PFA) catheter–based pulmonary vein (PV) isolation under deep sedation, higher bispectral index (BIS) values, reflecting lighter sedation, were associated with an increased incidence of coughing, whereas deeper sedation (maximum BIS 40–70) was associated with fewer coughing episodes.
• ▪Younger age, higher body mass index, and higher maximum BIS value among all PVs were identified as independent predictors of coughing.
• ▪Coughing during PFA did not adversely affect mid-term freedom from atrial tachyarrhythmia recurrence in a propensity score–matched analysis.

Introduction

Pulsed field ablation (PFA) has recently emerged as a novel tissue-selective energy modality for catheter ablation of atrial fibrillation (AF), minimizing collateral injury while effectively isolating the pulmonary veins (PVs).1, 2, 3, 4 In contrast, PFA introduces unique procedural challenges, one of which is the occurrence of coughing during PV isolation (PVI) when performed under deep sedation instead of general anesthesia.4, 5, 6, 7 Coughing during PFA applications can disrupt catheter stability and compromise the precision of ablation, potentially affecting the procedural outcomes. The bispectral index (BIS) is a well-established tool for monitoring the depth of sedation using electroencephalogram-based signals.^8,9 To date, the relationship between coughing and BIS-guided sedation depth during PFA for PVI has not yet been fully investigated. Further, the clinical relevance of coughing during PFA-based PVI remains unclear. The goals of this study were to investigate the relationship between coughing and sedation depth guided by the BIS during PVI using a pentaspline PFA catheter (FARAWAVE, Boston Scientific) and to evaluate the potential impact of coughing on mid-term clinical outcomes after AF ablation procedures.

Methods

Study population

This study retrospectively enrolled 198 consecutive patients who underwent an initial AF catheter ablation procedure using a pentaspline PFA system (FARAPULSE, Boston Scientific) at Gunma Prefectural Cardiovascular Center and Dokkyo Medical University from November 2024 to March 2025. All patients underwent contrast-enhanced computed tomography (CT) or magnetic resonance imaging to evaluate PV and left atrial (LA) anatomy. Transesophageal echocardiography and/or contrast-enhanced CT ruled out any intracardiac thrombi before the ablation procedure. The exclusion criteria for this study included redo ablation procedures, lack of BIS monitoring during the procedure, and ablation performed under general anesthesia. This study was conducted in accordance with the Declaration of Helsinki and was approved by the local ethics committee. An opt-out method was used to obtain consent from the study patients.

Sedation during the ablation procedure

All patients received deep sedation throughout the procedure. At the beginning of the procedure, a bolus of propofol (1.0–1.5 mg/kg) or an initial loading dose of dexmedetomidine (6 μg/kg/h for 10 minutes), along with pentazocine (15 mg), was administered intravenously. Subsequently, continuous infusions of propofol (4.0–6.0 mg/kg/h) and/or dexmedetomidine (0.2–0.7 μg/kg/h) were initiated. If the sedation became inadequate, repeated boluses of propofol (0.2–0.5 mg/kg) or thiamylal (1.0–1.5 mg/kg) were administered. The BIS was continuously monitored on the forehead using a BIS monitoring system (Covidien/Medtronic, Inc.) to assess sedation depth, and the target BIS range was ∼40–70.^8,9 Arterial oxygen saturation was continuously monitored using a pulse oximeter. Supplemental oxygen was administered to maintain an arterial oxygen saturation of ≥95% using either a mask with an oxygen flow rate of 5–10 L/min, with or without an oral airway, or noninvasive positive pressure ventilation.

Ablation procedure

Oral anticoagulants were continued throughout the periprocedural period, and the activated clotting time was maintained at >300 seconds by bolus administration and continuous infusion of heparin during the ablation procedure.¹⁰ At the operator’s discretion, some patients were administered 0.5 mg of intravenous atropine to minimize sinus arrest and/or atrioventricular block caused by a vagal response and 1–2 mg/kg of intravenous lidocaine to reduce coughing ∼5 minutes before PVI.¹¹

A PFA catheter was introduced into the LA through a single transseptal sheath (FARADRIVE, Boston Scientific). All procedures were guided by fluoroscopy and a 3-dimensional electroanatomic mapping system (Rhythmia, Boston Scientific or EnSite X, Abbott), with or without intracardiac echocardiography. PVI was performed exclusively using a 31-mm pentaspline PFA catheter according to the standard workflow previously described: (1) using a basket configuration, 2 PFA applications were delivered at the same location of the PV ostium and another 2 were delivered after rotating the catheter ∼30°–40°; and (2) using a flower configuration, 2 PFA applications were delivered at the same location of the PV ostium and another 2 were delivered after rotating the catheter ∼30°–40°.^1,4,12 In patients with a left common PV, PVI was performed separately for the left superior PV and the left inferior PV (LIPV). Each application consisted of 5 trains of biphasic pulses with an electric field of 2000 V. Additional PFA applications within the PVs, at the PV antra, or on the carina region were delivered at the operator’s discretion. After completing PVI, additional linear or regional ablation, extra-PV trigger ablation, and atrial tachycardia ablation were performed using a PFA or radiofrequency ablation catheter, if necessary.

Assessment of coughing and BIS during PFA for PVI

The presence or absence of coughing was evaluated during PFA at each PV ostium. The position of the PFA catheter during PVI was monitored using various combinations of fluoroscopy, intracardiac echocardiography, and the 3-dimensional mapping system.

The BIS value was assessed immediately before the initial PFA application for each PV. On the basis of the BIS values, the PVs were classified into 3 categories—<40, 40–70, and >70—and the incidence of coughing was compared among those categories. In addition, to enable a patient-level analysis, the maximum BIS value immediately before PFA among all PVs was determined for each patient.

Postablation follow-up

The study patients were followed up in the outpatient clinics every 1–3 months. Follow-up examinations including 12-lead electrocardiogram recording (ECG), 24-hour Holter monitoring, and ambulatory ECG monitoring were performed at each follow-up visit or emergency visit for any symptoms including palpitations, chest discomfort, or dizziness. Any episodes of atrial tachyarrhythmias lasting >30 seconds after a 60-day blanking period were regarded as atrial tachyarrhythmia recurrences.^13,14 Any symptoms lasting >30 seconds even without ECG recordings, which were similar to those during AF before ablation, were also regarded as arrhythmia recurrences. The decision to continue or discontinue antiarrhythmic drugs was left up to the patient’s preference and the discretion of the physician in charge.

Statistical analysis

Continuous variables were expressed as the median (interquartile range [IQR]), and a comparative analysis was performed using the Mann-Whitney U test. Categorical variables were expressed as the number and percentage of patients or PVs, and a comparative analysis was performed using the χ² or Fisher exact test. For post hoc residual analyses performed using the χ² test, adjusted standardized residuals with an absolute value of ≥1.96 were considered to indicate statistical significance. A multivariate logistic regression model with forced entry of covariates was used to assess the association between coughing and maximum BIS values, adjusting for potential confounders selected on the basis of clinical relevance and univariate analyses (P < .05). A receiver operating characteristic (ROC) curve analysis evaluated the ability of the maximum BIS value to discriminate between patients with and without coughing. As a sensitivity analysis, a multivariate analysis was repeated.

A propensity score–matched analysis was applied in the following manner. A multivariate logistic regression model with simultaneous forced entry was used to calculate the propensity scores in the study patients. The model included 30 baseline covariates and 3 intraprocedural covariates, including the intraprocedural administration of atropine and lidocaine and the maximum BIS value before PFA among all PVs. One-to-one matching without replacement was performed using the nearest-neighbor algorithm on the logit of the propensity score for the type of group, with a caliper width set to 0.2 times the standard deviation of the logit of the propensity score. After propensity score matching, a Kaplan-Meier analysis and log-rank test were used to compare the survival rate free from any atrial tachyarrhythmia recurrence at 6 months after a single ablation procedure between patients with and without coughing. A P value of <.05 was considered statistically significant. Statistical analyses were performed using SPSS Statistics, version 22 (IBM Corporation).

Results

Coughing was observed in 95 of the 198 patients studied (48.0%) and occurred during PFA of 260 PVs (32.8%). The baseline and procedural characteristics of the overall and propensity score–matched cohorts are summarized in Tables 1 and 2. The area under the ROC curve (C statistic) for the propensity score model was 0.895. Overall, the following parameters significantly differed between patients with and without coughing: age; sex; body weight; body mass index (BMI); AF type (paroxysmal or nonparoxysmal); current smoking status; Brinkman index (calculated as [cigarettes per day] × [years of smoking]); hypertension; LA diameter and left ventricular ejection fraction (LVEF) on echocardiography; LA volume, LA volume index, and maximum ostial diameters of the right superior and inferior PVs (RSPV and RIPV) on CT or magnetic resonance imaging; use of class I antiarrhythmic drugs before admission; and use of atropine and lidocaine, dosages of propofol and dexmedetomidine, number of PFA applications, maximum BIS value among all PVs, BIS values for each PV, and performance of LA posterior wall and cavotricuspid isthmus ablation during the procedure. All baseline and procedural characteristics other than the number of PFA applications in the LIPV were comparable between the propensity score–matched cohorts.

Table 1.

Baseline characteristics of the overall and propensity score–matched cohorts

Variable	All (N = 198)	Overall cohort			Propensity score–matched cohort
		Patients with coughing (n = 95)	Patients without coughing (n = 103)	P	Patients with coughing (n = 30)	Patients without coughing (n = 30)	P
Age (y)	70 (61–76)	69 (60–75)	73 (64–77)	.011	70 (62–76)	75 (63–77)	.336
Male sex	146 (73.7)	77 (81.1)	69 (67.0)	.025	23 (76.7)	22 (73.3)	.766
Body weight (kg)	65.9 (57.3–75.7)	69.8 (64.1–79.5)	61.0 (53.7–70.0)	<.001	65.6 (57.4–72.6)	65.5 (56.3–75.4)	.859
BMI (kg/m2)	24.1 (21.7–26.5)	25.3 (22.9–27.3)	22.9 (20.9–25.3)	<.001	23.4 (21.7–25.7)	24.8 (21.3–26.2)	.723
Type of AF				.001			>.99
Paroxysmal	99 (50.0)	36 (37.9)	63 (61.2)		16 (53.3)	16 (53.3)
Nonparoxysmal	99 (50.0)	59 (62.1)	40 (38.8)		14 (46.7)	14 (46.7)
Structural heart disease	48 (24.2)	24 (25.3)	24 (23.3)	.748	9 (30.0)	8 (26.7)	.774
Ischemic heart disease	9 (4.5)	3 (3.2)	6 (5.8)	.291	0 (0)	1 (3.3)	.500
Hypertrophic cardiomyopathy	4 (2.0)	2 (2.1)	2 (1.9)	.658	1 (3.3)	1 (3.3)	.754
Dilated cardiomyopathy	6 (3.0)	3 (3.2)	3 (2.9)	.620	0 (0)	1 (3.3)	.500
Moderate-severe valvular disease	15 (7.6)	7 (7.4)	8 (7.8)	.916	4 (13.3)	1 (3.3)	.177
Tachycardia-induced cardiomyopathy	19 (9.6)	9 (9.5)	10 (9.7)	.955	3 (10.0)	5 (16.7)	.353
Atrial septal defect	2 (1.0)	2 (2.1)	0 (0)	.229	2 (6.7)	0 (0)	.246
Others	1 (0.5)	1 (1.1)	0 (0)	.480	1 (3.3)	0 (0)	.500
Smoking history
Nonsmoking	97 (50.3)	41 (44.1)	56 (56.0)	.098	15 (50.0)	15 (50.0)	>.99
Past smoking	65 (33.7)	32 (34.4)	33 (33.0)	.836	11 (36.7)	11 (36.7)	>.99
Current smoking	31 (15.7)	20 (21.1)	11 (10.7)	.045	4 (13.3)	4 (13.3)	.647
Brinkman index	0 (0–400)	89 (0–436)	0 (0–300)	.048	17 (0–400)	20 (0–410)	.881
Respiratory disease	18 (9.1)	8 (8.4)	10 (9.7)	.753	1 (3.3)	1 (3.3)	.754
Bronchial asthma	2 (1.0)	1 (1.1)	1 (1.0)	.731	0 (0)	0 (0)	–
COPD	8 (4.0)	2 (2.1)	6 (5.8)	.168	0 (0)	1 (3.3)	.500
Interstitial lung disease	2 (1.0)	1 (1.1)	1 (1.0)	.731	0 (0)	0 (0)	–
Sleep apnea syndrome	5 (2.5)	3 (3.2)	2 (1.9)	.462	1 (3.3)	0 (0)	.500
Lung tumor/post–lung resection	2 (1.0)	1 (1.1)	1 (1.0)	.731	0 (0)	0 (0)	–
Hypertension	118 (59.6)	68 (71.6)	50 (48.5)	.001	19 (63.3)	21 (70.0)	.584
Diabetes mellitus	37 (18.7)	20 (21.1)	17 (16.5)	.412	4 (13.3)	6 (20.0)	.488
Heart failure	39 (19.7)	19 (20.0)	20 (19.4)	.918	6 (20.0)	7 (23.3)	.754
Stroke/transient ischemic attack	10 (5.1)	5 (5.3)	5 (4.9)	.575	1 (3.3)	1 (3.3)	.754
CHA2DS2-VASc score	2 (1–3)	2 (1–3)	3 (1–3)	.420	2 (1–4)	3 (2–4)	.366
Creatinine (mg/dL)	0.91 (0.78–1.06)	0.95 (0.82–1.05)	0.87 (0.76–1.11)	.243	0.88 (0.75–1.03)	0.91 (0.79–1.08)	.574
eGFR (mL/min per 1.73 m2)	59 (50–70)	59 (51–67)	59 (47–72)	.717	63 (53–69)	58 (53–71)	.631
BNP (pg/mL)	97.4 (34.5–205.7)	99.0 (39.3–212.8)	84.4 (33.7–204.3)	.693	89.3 (29.6–190.9)	119.9 (39.1–251.6)	.274
Transthoracic echocardiographic findings
LA diameter (mm)	40 (35–44)	41 (37–46)	38 (33–42)	<.001	40 (35–44)	40 (37–42)	.947
LVEF (%)	61 (55–65)	60 (50–65)	65 (60–65)	<.001	60 (55–65)	62 (54–65)	.769
CT or MR imaging findings
LA volume (mL)	108.3 (84.8–146.0)	127.6 (94.3–163.9)	96.5 (66.2–127.9)	<.001	113.6 (87.1–143.3)	108.9 (86.0–153.8)	.859
LA volume index (mL/m2)	63.2 (49.1–84.0)	68.1 (54.2–87.5)	57.4 (39.2–79.9)	.003	61.9 (53.0–81.6)	65.3 (52.6–87.2)	.871
Left common PV	12 (6.1)	7 (7.4)	5 (4.9)	.459	3 (10.0)	2 (6.7)	.500
Maximum PV ostial diameter (mm)
LSPV	21.2 (18.9–24.7)	21.7 (19.2–25.5)	20.6 (18.5–24.0)	.061	22.6 (19.2–27.3)	23.6 (19.6–27.5)	.610
LIPV	18.4 (16.5–20.3)	18.7 (17.0–20.3)	18.1 (15.8–20.4)	.104	18.3 (19.8–20.7)	18.7 (17.2–21.1)	.544
RSPV	24.3 (20.3–27.3)	26.2 (22.4–28.9)	21.9 (19.3–25.8)	<.001	25.5 (23.0–28.0)	23.9 (21.1–27.4)	.297
RIPV	19.5 (16.9–23.1)	20.8 (18.0–24.4)	18.9 (15.6–21.7)	<.001	19.4 (17.6–25.1)	20.6 (18.3–24.0)	.994
Medical therapy before admission
Warfarin	3 (1.5)	1 (1.1)	2 (1.9)	.530	0 (0)	0 (0)	–
Direct oral anticoagulants	195 (98.5)	94 (98.9)	101 (98.1)	.530	30 (100)	30 (100)	–
Class I antiarrhythmic drugs	23 (11.6)	4 (4.2)	19 (18.4)	.002	3 (10.0)	2 (6.7)	.500
Amiodarone	18 (9.1)	12 (12.6)	6 (5.8)	.096	1 (3.3)	1 (3.3)	.754
Bepridil	26 (13.1)	13 (13.7)	13 (12.6)	.825	2 (6.7)	3 (10.0)	.500

Values are presented as median (interquartile range) or n (%).

AF = atrial fibrillation; BMI = body mass index; BNP = B-type natriuretic peptide; Brinkman index = (cigarettes per day) × (years of smoking); CHA₂DS₂-VASc score = congestive heart failure, hypertension, diabetes mellitus, vascular disease (myocardial infarction, aortic plaque, and peripheral vascular disease), age 65–74 y, female sex (1 point for the presence of each), age ≥75 y, and stroke and transient ischemic attack (2 points); COPD = chronic obstructive pulmonary disease; CT = computed tomographic; eGFR = estimated glomerular filtration rate; LA = left atrial; LIPV = left inferior pulmonary vein; LSPV = left superior pulmonary vein; LVEF = left ventricular ejection fraction; MR = magnetic resonance; PV = pulmonary vein; RIPV = right inferior pulmonary vein; RSPV = right superior pulmonary vein.

Table 2.

Procedural characteristics of the overall and propensity score–matched cohorts

Variable	All (N = 198)	Overall cohort			Propensity score–matched cohort
		Patients with coughing (n = 95)	Patients without coughing (n = 103)	P	Patients with coughing (n = 30)	Patients without coughing (n = 30)	P
Intraprocedural medication
Atropine	104 (52.5)	40 (42.1)	64 (62.1)	.005	13 (43.3)	13 (43.3)	>.99
Lidocaine	67 (33.8)	20 (21.1)	47 (45.6)	<.001	6 (20.0)	6 (20.0)	>.99
Dosage of propofol (mg/kg/h)	3.8 (3.2–4.8)	3.7 (3.1–4.5)	4.2 (3.4–5.0)	.007	3.7 (3.0–4.5)	4.0 (3.2–4.5)	.416
Dosage of dexmedetomidine (μg/kg/h)	0.67 (0.48–0.72)	0.70 (0.60–0.72)	0.61 (0.46–0.71)	.001	0.68 (0.60–0.72)	0.70 (0.58–0.72)	.595
Total procedure time (min)	91 (78–110)	95 (82–108)	87 (73–112)	.069	86 (79–107)	93 (77–119)	.959
Total fluoroscopy time (min)	30 (24–37)	31 (24–38)	29 (23–36)	.229	29 (24–36)	28 (22–39)	.673
LA dwell time (min)	48 (40–59)	50 (40–59)	46 (40–59)	.571	48 (38–57)	44 (40–70)	.594
Total number of PFA applications	52 (48–60)	54 (48–64)	51 (46–58)	.034	55 (48–59)	53 (48–59)	.796
Number of PFA applications for PVI	48 (42–54)	48 (46–56)	47 (40–53)	.009	50 (46–53)	52 (46–56)	.505
LSPV	12 (10–14)	12 (10–14)	12 (10–14)	.044	14 (12–16)	13 (12–14)	.335
LIPV	12 (10–13)	12 (10–14)	10 (10–12)	.127	10 (10–12)	12 (10–14)	.045
RSPV	12 (10–14)	12 (10–14)	12 (10–14)	.069	12 (11–14)	12 (12–14)	.927
RIPV	12 (10–14)	12 (10–14)	12 (10–14)	.041	12 (10–14)	12 (12–14)	.182
Maximum BIS before PFA among all PVs	74 (64–78)	76 (69–79)	71 (56–77)	.002	76 (69–77)	74 (66–78)	.609
BIS before PFA for each PV
LSPV	56 (44–65)	60 (47–69)	52 (42–63)	.005	61 (46–69)	57 (52–66)	.853
LIPV	70 (53–76)	72 (59–74)	64 (48–75)	.004	73 (56–77)	70 (57–73)	.446
RSPV	66 (47–73)	68 (56–74)	64 (42–71)	.005	63 (49–72)	68 (50–75)	.328
RIPV	62 (50–72)	65 (54–73)	60 (46–71)	.030	62 (51–71)	66 (51–73)	.539
Catheter ablation procedure
PVI	100 (100)	100 (100)	100 (100)	–	30 (100)	30 (100)	–
LAPW ablation	91 (46.0)	36 (37.9)	55 (53.4)	.029	11 (36.7)	11 (36.7)	>.99
CTI ablation	78 (39.4)	46 (48.4)	32 (31.1)	.013	14 (46.7)	13 (43.3)	.795
SVC isolation	13 (6.6)	3 (3.2)	10 (9.7)	.063	1 (3.3)	3 (10.0)	.306
Others	8 (4.0)	5 (5.3)	3 (2.9)	.317	2 (6.7)	2 (6.7)	.694
Complications	4 (2.0)	1 (1.1)	3 (2.9)	.342	0 (0)	0 (0)	–
Cardiac tamponade	0	0	0		0	0
Stroke	1 (0.5)	0	1 (1.0)		0	0
Coronary artery spasm	1 (0.5)	0	1 (1.0)		0	0
Air embolism	0	0	0		0	0
Pseudoaneurysm	0	0	0		0	0
Groin hematoma	2 (1.0)	1 (1.1)	1 (1.0)		0	0

Values are presented as median (interquartile range) or n (%).

BIS = bispectral index; CTI = cavotricuspid isthmus; LAPW = left atrial posterior wall; PFA = pulsed field ablation; PVI = pulmonary vein isolation; SVC = superior vena cava; other abbreviations as in Table 1.

Figure 1 presents the incidence of coughing according to the maximum BIS values among all PVs (<40, 40–70, and >70). The incidence of coughing was 0%, 37.2%, and 55.5%, respectively (P = .013), and was significantly lower than expected at BIS 40–70 and higher at BIS >70 on the basis of adjusted standardized residuals. Figure 2 presents the incidence of coughing in patients with BIS values of <40, 40–70, and >70 at each PV site. The incidence of coughing differed significantly across the BIS categories for the LIPVs and RSPVs (P = .003 and P = .013, respectively). Adjusted standardized residuals revealed a significantly lower incidence of coughing in the LIPVs at BIS <40 and a higher incidence in the LIPVs and RSPVs at BIS >70. The incidence of coughing also differed significantly depending on the PV site (P = .012) and was significantly lower than expected in the RIPV (Figure 3). The results of comparisons between patients with and without coughing at each PV site are presented in Supplemental Table 1.

Figure 1.

Incidence of coughing stratified by maximum BIS values immediately before PFA among all PVs for each patient. The blue and green bars represent the percentage of patients with coughing at maximum BIS values of 40–70 and >70, respectively. BIS = bispectral index; PFA = pulsed field ablation; PV = pulmonary vein.

Figure 2.

Incidence of coughing stratified by BIS values immediately before PFA at each PV site. The red, blue, and green bars represent the percentage of patients with coughing at BIS values of <40, 40–70, and >70, respectively. LIPV = left inferior pulmonary vein; LSPV = left superior pulmonary vein; RIPV = right inferior pulmonary vein; RSPV = right superior pulmonary vein; other abbreviations as in Figure 1.

Figure 3.

Proportion of patients with coughing at each PV site. The red and blue bars represent the number and percentage of patients with and without coughing, respectively. Abbreviations as in Figures 1 and 2.

Procedure-related complications occurred in 4 patients (2.0%): an atherothrombotic cerebral infarction in 1 patient, which was diagnosed by experienced neurosurgeons and treated with argatroban and clopidogrel; a coronary artery spasm in 1 patient, which occurred during cavotricuspid isthmus ablation with radiofrequency energy and was relieved by nitroglycerin; and a femoral hematoma in 2 patients, which was managed with the local application of pressure. None of the patients in this series developed an LA–esophageal fistula, symptomatic gastric hypomotility, or persistent phrenic nerve palsy.

Predictors of coughing during PFA for PVI

A multivariate logistic regression analysis was performed including age, male sex, BMI, nonparoxysmal AF, current smoking status, LA diameter, intraprocedural atropine and lidocaine use, and maximum BIS value before PFA among all PVs. Higher BMI and higher maximum BIS value were significant positive predictors of coughing (odds ratios 1.188 and 1.052; 95% confidence intervals [CIs] 1.063–1.328 and 1.011–1.094; P = 0.002 and P = 0.012), while older age was a negative predictor (odds ratio 0.962; 95% CI 0.927–0.999; P = .044) (Figure 4). An ROC analysis for the maximum BIS value demonstrated an area under the curve of 0.629 (95% CI 0.552–0.706), and the sensitivity and specificity for predicting coughing were 84.2% and 36.9%, respectively, at an optimal cutoff of 65 (Supplemental Figure 1).

Figure 4.

Forest plot showing ORs and 95% CIs derived from a multivariate logistic regression model for the predictors of coughing. BMI = body mass index; CI = confidence interval; LA = left atrial; OR = odds ratio; other abbreviations as in Figure 1.

Mid-term clinical outcomes

All patients had a follow-up period of >6 months after the ablation procedure. The median follow-up period did not significantly differ between the propensity score–matched cohorts with and without coughing (254 days [IQR 206–282 days] vs 263 days [IQR 218–295 days]; P = .464). A Kaplan-Meier analysis and log-rank test showed that the survival rate free from any atrial tachyarrhythmia recurrence at 6 months after a single ablation procedure was 93.3% in patients with coughing and 96.7% in those without, respectively (P = .306) (Figure 5). Antiarrhythmic drugs were continued at 6-month follow-up in 13.3% of patients with coughing and 23.3% of those without (P = .317). The mid-term clinical outcomes in the overall cohort are presented in Supplemental Figure 2.

Figure 5.

Kaplan-Meier curves for atrial tachyarrhythmia recurrence–free survival after a single ablation procedure in the propensity score–matched cohorts. The red and blue lines represent the matched patients with and without coughing, respectively.

Discussion

This retrospective observational study investigated the relationship between coughing and sedation depth guided by the BIS during PVI using a pentaspline PFA catheter and evaluated the impact of coughing on post-AF ablation clinical outcomes by using propensity score matching. We demonstrated that (1) overall, coughing occurred in 32.8% of PVs and 48.0% of patients experienced at least 1 episode of coughing during PFA for PVI; (2) deeper sedation (maximum BIS values of 40–70) was associated with a lower incidence of coughing, while lighter sedation (maximum BIS values of >70) was associated with a higher incidence; (3) younger age, higher BMI, and higher maximum BIS value were independent risk factors for coughing; and (4) the propensity score–matched analysis revealed comparable clinical outcomes after the ablation procedure between patients with and without coughing.

Relationship between coughing and sedation depth during PFA for PVI

Coughing during PFA under deep sedation is a well-recognized clinical phenomenon. However, to our knowledge, few studies have quantitatively assessed the impact of sedation depth monitored by the BIS on the occurrence of coughing during PFA-based PVI. In the present study, higher BIS values were associated with a higher incidence of coughing, and the maximum BIS value for each patient was identified as an independent risk factor for coughing in the multivariate analysis. The overall difference across the maximum BIS categories (<40, 40–70, and >70) was statistically significant, and the adjusted standardized residual analysis revealed a lower-than-expected incidence of coughing in the maximum BIS 40–70 category and a higher-than-expected incidence in the maximum BIS >70 category. However, the maximum BIS <40 category exhibited no statistically significant deviation, likely because this category included only 1 patient. Nevertheless, those results support the concept that maintaining an adequate depth of sedation, as objectively assessed by BIS monitoring, may help reduce coughing during PFA.

Our result that deeper sedation was associated with a lower incidence of coughing during PFA is concordant with evidence from a randomized trial with postoperative extubation.¹⁵ Zhang et al¹⁵ demonstrated that BIS-guided sedation (maintaining BIS 60–70) with dexmedetomidine and propofol significantly reduced cough occurrence during extubation, compared with no sedation, which suggests that controlled sedation depth reduces airway hyperresponsiveness and raises the cough threshold.

Relationship between coughing and PV location

Coughing was significantly less frequent in the RIPV and tended to be more frequent in the superior PVs in the present study. The superior PVs have been reported to be anatomically closer to the bronchi than the inferior PVs, as demonstrated by multislice CT.¹⁶ This anatomical proximity between the bronchi and PFA application sites may partly explain why coughing was more commonly observed in the superior PVs.

Factors associated with coughing during PFA for PVI

One possible mechanism of coughing is the direct stimulation of the bronchial smooth muscle near the PFA application sites. Younger age and male sex have been reported to be associated with thicker airway walls relative to older age and female sex.^17,18 Thus, we speculated that younger patients with a higher BMI may have a greater mass of bronchial smooth muscle than do older patients with a lower BMI. The greater the amount of bronchial smooth muscle, the more muscle is directly stimulated by PFA, which may increase the susceptibility to coughing.

Intravenous lidocaine (0.5–2 mg/kg) has been reported to be effective in preventing intubation-induced coughing.¹¹ The mechanism of coughing caused by laryngeal or tracheal stimulation from laryngoscopy or the endotracheal tube during intubation may differ from that caused by bronchial stimulation during PFA. However, one possible explanation for the cough-suppressing effect of lidocaine during PFA is that lidocaine reduced the direct stimulation of the bronchial smooth muscle by PFA. Another possible explanation is that lidocaine reduced coughing originating from the larynx or trachea, which was provoked by stimulation of the recurrent laryngeal or vagus nerve due to PFA.¹⁹ In the present study, lidocaine was administered to 33.8% of patients to suppress coughing during PFA, and patients receiving intravenous lidocaine had less coughing than did those who did not receive it. However, intravenous lidocaine was not an independent risk factor for coughing in the multivariate analysis. Those findings may be explained by patient-related factors, such as age, BMI, and BIS values, playing a more substantial role in determining cough occurrence than intravenous lidocaine alone. Nevertheless, because this study was not a randomized controlled trial designed to evaluate the effectiveness of lidocaine in reducing coughing, we cannot draw definitive conclusions regarding its efficacy in suppressing coughing during PFA based on our results. Given the potential suppressive effect of lidocaine on coughing, we performed a sensitivity analysis restricted to patients who did not receive intraprocedural lidocaine (Supplemental Table 2). The results were largely consistent with those of the primary analysis, with the association between coughing and the maximum BIS value remaining statistically significant.

Although patients with coughing had a significantly lower LVEF than did those without, both groups had a median LVEF of ≥60%. Therefore, LVEF was not included in the primary multivariate model. When LVEF was additionally included in the multivariate model to perform a sensitivity analysis, the association between coughing and the maximum BIS value remained statistically significant, while the effect of age was attenuated (Supplemental Table 3).

Clinical impact of coughing during PFA for PVI

The propensity score–matched analysis revealed that patients with and without coughing had comparable clinical outcomes, suggesting that coughing episodes during PFA for PVI are generally well tolerated and do not adversely affect AF ablation efficacy.

Coughing, as well as respiratory movements, can compromise the ablation catheter stability and cause map shifts in 3-dimensional mapping systems, potentially affecting the durability of PVI and procedural outcomes. In radiofrequency ablation, maintaining the catheter tip at the same location for a certain duration is critical for creating durable ablation lesions. Therefore, stabilizing the catheter tip and maintaining adequate contact force under general anesthesia appear to result in more durable lesion formation.^20,21 In contrast, since pulsed field energy is delivered over a much shorter duration than radiofrequency energy, PFA may considerably reduce the impact of cough-induced catheter displacement on lesion formation. Further, in the present study, the majority of patients received additional PFA applications distal to the PV ostium, as proposed in the “olive” strategy by Schaack et al.²² Compared with thermal ablation–based PVI, PFA-based PVI carries an extremely low risk of PV stenosis,^1,5 which allows ablation not only at the PV ostium but also directly on the PV sleeves. Performing PFA from the PV ostium into the PV sleeves may reduce the impact of cough-induced catheter displacement on procedural outcomes.

Clinical implications

Deep sedation, instead of general anesthesia, is often used in AF catheter ablation.23, 24, 25 Our results highlight that maintaining an adequate sedation depth, as monitored by the BIS, may be associated with fewer coughing episodes during PFA for PVI. In contrast, not only BIS values themselves but also patient characteristics, such as age and BMI, appear to be important for tailoring optimal sedation strategies, including deeper sedation or general anesthesia.

Coughing can interfere with PFA procedures, and suppressing coughing may facilitate procedural performance. Nevertheless, the impact of coughing on clinical outcomes may be limited. Prospective studies randomizing BIS targets in populations undergoing PFA are warranted to establish optimal sedation strategies that maximize procedural safety and efficiency.

Study limitations

The present study had several limitations. First, this study was retrospectively designed, and the data were derived from only 2 centers with a limited sample size. Second, anesthesia dosing was adjusted to maintain BIS values within a target range of 40–70. However, the observed median BIS was 62 (IQR 48–83), and PFA applications were delivered to some PVs at BIS values outside the target range. Third, this study focused on the mid-term outcomes of PFA-based AF catheter ablation under deep sedation. Further understanding of the impact of coughing on clinical outcomes will require an investigation in a prospective multicenter study with a larger sample size and long-term follow-up. Finally, it remains unclear whether the results of this study can be applied to other PFA systems.

Conclusion

During pentaspline PFA catheter–based PVI under deep sedation with BIS monitoring, PFA of approximately one-third of the PVs triggered coughing, and nearly half of patients experienced at least 1 coughing episode. Higher BIS values, indicating lighter sedation, were associated with a higher incidence of coughing, while lower BIS values, indicating deeper sedation, were associated with a lower incidence. This study suggests that younger age, higher BMI, and higher maximum BIS values among all PVs for each patient may be independent risk factors for coughing during PFA for PVI. However, coughing may not adversely affect mid-term freedom from atrial tachyarrhythmia recurrence after the ablation procedure.

Disclosures

The authors have no conflicts of interest to disclose.