Ablation of atrial fibrillation in an ambulatory outpatient setting

Abstract

Background

The safety of atrial fibrillation (AF) ablation in an ambulatory outpatient center has not previously been reported.

Objective

The aim of this study is to report the feasibility and safety of AF ablation in an ambulatory setting.

Methods

We identified all AF ablations performed at the Alaska Heart and Vascular Institute’s ambulatory center since program initiation to current day using billing records. Procedural complications, postoperative utilization of hospital services, and emergency room (ER) utilization were captured by chart review.

Results

A total of 476 patients underwent pulmonary vein isolation in the ambulatory setting over a 6.3-year period. Patients’ average age was 58 ± 9.3 years, body mass index was 32.9 kg/m², and the CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age 65–74 years, sex category) score was 1.7. For 85%, this was the first AF ablation, and 55% had paroxysmal AF. Cryoablation was used in 85%. A combined primary safety outcome capturing potentially unstable perioperative safety events occurred in 1.5% of patients, all of whom were stabilized prior to hospital transfer. A total of 1.5% of patients required same-day hospital services, with another 1.5% returning to the ER within 24 hours. A total of 96% of patients did not require hospital services within 24 hours of ablation. The 30-day ER utilization was 13.7%, similar to published data of same-day discharge of AF ablation done in the hospital setting. There were no emergent cardiac surgical interventions and no mortality events.

Conclusion

Catheter ablation for AF in the ambulatory setting is both feasible and safe in this large single-center experience. More studies are needed to confirm this next frontier in catheter ablation for AF.

Key Findings.

• ▪Ablation of atrial fibrillation in an office-based lab is safe and feasible.
• ▪Major adverse complication rates in this large cohort were similar to published historical data.
• ▪All serious complications were stabilized on site prior to hospital transfer.
• ▪Considerations for a safe and hospital independent outpatient ablation program are discussed.

Introduction

The burden and prevalence of atrial fibrillation (AF) is increasing in the United States.^1,2 It is estimated that 12.1 million people in the United States will have AF in 2030.^2,3 AF accounts for one-third of hospitalizations for cardiac dysrhythmias,⁴ and with more than 750,000 AF-related hospitalizations each year, inpatient care of AF constitutes a significant burden to the U.S. healthcare system.⁵

Catheter ablation for AF is a well-established therapy for the management of AF after failing antiarrhythmic drug therapy, and emerging data support the superiority over antiarrhythmic therapy.^6,7 Improvements in ablative technology with both cryoablation and force-sensing radiofrequency catheters have improved the safety and reduced intraprocedural and early postoperative complications.^8,9 Unsurprisingly, the utilization of catheter ablation for treating AF has increased over time,¹⁰ and so has hospital utilization, as most AF ablation has been performed in a hospital-based setting.

Several studies support the safety of same-day discharge (SDD) following AF ablation,^11,12 and this practice has become commonplace, particularly in the post–COVID-19 era.^13,14 The Alaska Heart and Vascular Institute (AHVI) has been practicing SDD for hospital-based uncomplicated AF ablation since 2015. This is now routine practice, and has prompted an interest in operating outside of the hospital.

At the moment, the Centers for Medicare and Medicaid Services (CMS) only reimburses facility fees for ablation when performed in a designated facility (i.e. not an office-based or ambulatory cath lab) and does not reimburse ablation in an ambulatory surgery center (which is considered a facility, but the Current Procedural Terminology codes for ablation are not on the approved covered procedure list of a surgery center). Thus, neither ambulatory cath labs nor surgery centers are able to offer ablation to Medicare patients.

AHVI owns and operates a CMS-designated office-based laboratory, which consists of a single-room cardiac catheterization lab with full anesthesia support and an 8-bed pre- and postoperative area. The lab is equipped for diagnostic and interventional coronary procedures and electrophysiology (EP) procedures. This space also functions as an ambulatory surgery center 1 day a week, and is used for cardiac device implantation. Since equipping the lab with mapping and recording systems in late 2016, we have been able to offer patients ablation with SDD for a variety of arrhythmias including AF. Unsurprisingly, private payers were enthused to contract for these procedures in this space, given the better rates than the hospital.

The lab is physically located on hospital grounds with rapid access to emergency room (ER), intensive care unit, and operating room (OR) services, including cardiac surgical backup. Transfer protocols to the hospital are similar to those used in our nearby clinic and utilize lab staff to transfer stable patients, and emergency medical services (EMS) for patients who are too unstable for staff to escort to the hospital. Arrangements are in place with the local hospital for bed prioritization for direct admission if needed, cardiac surgery for emergent intervention if needed, hospital inpatient pharmacy for medications not on formulary at the ambulatory surgery center, and the blood bank for blood products if needed.

In this article, we report our entire experience ablating AF in the ambulatory setting. This is, to our knowledge, the largest cohort of non–hospital-based AF ablation published. We seek to provide insight from our experience to help others perform ablation in this type of setting safely.

Methods

All AF ablation cases done at the AHVI ambulatory outpatient lab (between December 1, 2016, and March 22, 2023) were identified using billing records for a 93656 AF ablation code. No exclusion criteria were applied. Data were abstracted from the medical record including demographics, AF history, echocardiographic data, intraprocedural and postprocedural data, complications, discharge characteristics, and all ER records and hospitalization data in the month following the procedure. All data was maintained on the Alaska Heart Institute password-encrypted server in an Excel spreadsheet (Microsoft Corporation, Redmond, WA). Data analysis was performed by M.E.W. The sponsor did not participate in data abstraction, analysis, or manuscript preparation.

The research protocol was submitted to Advarra Institutional Review Board who determined that this research project was exempt from Institutional Review Board oversight using the Department of Health and Human Services regulations found at 45 CFR 46.104(d)(4). Patient consent was waived due to the use of retrospective and de-identified data. The study was overseen by the Alaska Cardiovascular Research foundation, with all researchers having undergone research training prior to accessing data.

Endpoints

We defined a primary safety outcome as the occurrence of any of below major safety events: (1) tamponade requiring pericardiocentesis; (2) major bleeding requiring blood product transfusion the same day as the procedure; (3) emergent cardiac or vascular surgery performed the same day as the procedure; (4) transient ischemic attack (TIA) or cerebrovascular accident (CVA); (5) pneumothorax requiring chest tube performed the same day as the procedure; (6) acute coronary syndrome requiring angiography the same day as the procedure; (7) death within 24 hours of procedure; and (8) EMS activation for transportation to ER, OR, or hospital of an unstable patient.

The secondary outcome was defined as the incidence of requiring same-day hospital services for any reason not listed previously prior to discharge from the ambulatory center.

The tertiary outcome was the incidence of ER utilization: (1) <24 hours postprocedure; (2) 24 hours to 1 week postprocedure; and (3) 1 week to 1 month postprocedure.

Results

A total of 476 patients were ablated between December 2, 2016, and March 22, 2023. Baseline characteristics can be seen in Table 1. Participants had an average age of 57.5 ± 9.3 years, a body mass index (BMI) of 32.9 ± 7.0 kg/m², and a CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age 65–74 years, sex category) score of 1.7 ± 1.3. The majority (55%) of patients had paroxysmal AF, while 45% had persistent or long-standing persistent AF. Prior antiarrhythmic use was 46% overall and was higher in persistent and long-standing persistent than paroxysmal AF. The average ejection fraction was 55.0% (range 10%–77%). Congestive heart failure history was present in 32.8% of patients with an average New York Heart Association functional class of 1.6. The average indexed left atrial volume was 34.8 mL/m². The majority of this cohort were undergoing their first AF ablation (85%), while 15% presented for repeat ablation.

Table 1.

Baseline Characteristics

Age, y	57.5 ± 9.3 (17–81)
BMI	32.9 ± 7.0 (18.7–61.0)
CHA2DS2-VASc	1.7 ± 1.3 (0–6)
Sex
Female	131
Male	345
HTN	274 (57.6)
TIA/CVA/PE	35 (7.4)
PAD/CAD	65 (13.7)
Congestive heart failure	156 (32.8)
Diabetes mellitus	69 (14.5)
Recent or current smoker	164 (34.5)
OSA	188 (39.5)
AF subtype
Paroxysmal	262 (55)
Persistent	202 (42)
Long-standing persistent	12 (3)
Prior antiarrhythmic use
Paroxysmal	99 (37.8)
Persistent	112 (55.4)
Long-standing persistent	8 (66.7)
Clinical heart failure	156 (32.8) with average NYHA functional class 1.6
LVEF preablation, %
Overall	55 ± 12.4
Paroxysmal	59.8 with 6.9 EF <50
Persistent	49.3 average with 44.6 EF <50
Long-standing persistent	52.0 with 41.7 EF <50
LA volume index, mL/m2
Paroxysmal	32.9
Persistent	38.4
Long-standing persistent	30.1
First AF ablation procedure	402 (85)
History of prior AF ablation	74 (15)

Values are mean ± SD (range) or n (%), unless otherwise indicated.

AF = atrial fibrillation; BMI = body mass index; CAD = coronary artery disease; CHA₂DS₂-VASc = congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age 65-74 years, sex category; CVA = cerebrovascular accident; EF = ejection fraction; HTN = hypertension; LA = left atrial; LVEF = left ventricular ejection fraction; NYHA = New York Heart Association; OSA = obstructive sleep apnea; PAD = peripheral artery disease; PE = pulmonary embolism; TIA = transient ischemic attack.

Pulmonary vein isolation (PVI) was performed with a 28-mm Arctic Front Advance cryoballoon (Medtronic, Minneapolis, MN) in 403 (85%) cases, with radiofrequency used as the sole ablative technology for the remainder. Most patients underwent additional ablation beyond the pulmonary veins, including routine superior vena cava isolation, and often cavotricsupid isthmus ablation; however, specific details of non-PVI ablation were not gathered for this article. Median fluoroscopy time was 4.1 minutes, with median contrast 30 mL. The median procedure duration was 2.25 hours. Lab area dwell time (from patient arrival to patient discharge) was a median of 9 hours.

Safety outcomes and rates of ER utilization are summarized in Table 2. Seven (1.5%) patients experienced the primary outcome of a major perioperative safety event. Four (0.8%) patients developed tamponade requiring percutaneous drainage. One of these events occurred after the AF ablation was completed and during (planned) pacemaker implantation but was included. All patients were stabilized in the cath lab and were transferred to the hospital in stable condition for overnight observation, and none required surgical intervention. Three (0.6%) patients were transferred due suspected bleeding for computed tomography imaging unavailable at the surgery center. Two (0.4%) were found to have a retroperitoneal bleed requiring transfusion but no surgical intervention, and in 1 (0.2%) no specific etiology was identified but transfusion was still administered given baseline anemia. No patients required emergent cardiac or vascular surgery, experienced a TIA/CVA, were transferred via EMS, or experienced an acute coronary syndrome, pneumothorax, or death.

Table 2.

Results

Safety endpoints	Data
Primary outcome: major perioperative safety event	7 (1.5)
Tamponade requiring pericardiocentesis	4 (0.8)
Major bleeding requiring blood product transfusion	3 (0.6)
Emergent cardiac or vascular surgery	0 (0)
TIA	0 (0)
CVA	0 (0)
Pneumothorax requiring chest tube	0 (0)
Acute coronary syndrome requiring angiography	0 (0)
EMS activation, ER, OR, or hospital of unstable patient	0 (0)
Death within 24 h of procedure	0 (0)
Secondary outcome: incidence of requiring same-day hospital services for reasons not listed above	7 (1.5)
Postdischarge ER utilization
ER utilization within 24 h postprocedure	7 (1.5)
ER utilization 24 h to 1 wk postprocedure	34 (7.1)
ER utilization 1 wk to 1 mo postprocedure	24 (5.0)
Freedom from hospital or ER during first 24 h	455 (95.6)
Freedom from hospital or ER during first 30 d	395 (83.0)

Values are n (%).

CVA = cerebrovascular accident; EMS = emergency medical services; ER = emergency room; OR = operating room; TIA = transient ischemic attack.

Seven patients (1.5%) met the secondary safety outcome, the incidence of ER or hospital services utilization prior to discharge. Two patients (0.4%) required prolonged groin care beyond our staffing structure, 1 of whom required expedited vascular ultrasound (not available on site). One patient (0.2%) was admitted for vascular surgery repair the following day of a radial arterial line complication. One patient (0.2%) was admitted for bradycardia managed with atropine and, observed overnight without need for permanent pacing. Two patients (0.2%) were slow to awaken from anesthesia and required ER or hospitalization for slow vasopressor wean. One (0.2%) patient was unable to be extubated and required transportation to the hospital and was found to have a previously undiagnosed pneumonia.

Finally, 7 (1.5%) patients required ER after discharge but within 24 hours of the procedure for a variety of reasons: shortness of breath (managed with fluids), groin bleeding (managed with pressure), visual changes (contact lens left in after procedure), 3 patients with chest pain (2 no specific diagnosis, 1 treated for pericarditis), bradycardia (spontaneously resolved). ER utilization from 24 hours to 7 days postablation occurred in 34 (7.1%) patients, and late postprocedure (1 week to 1 month) in 24 (5.0%) patients.

Discussion

This is the first and largest cohort of AF ablation in an ambulatory EP lab reported to our knowledge. We report every periprocedural clinical safety-related event, and ER or hospital utilization event from program inception to current day without exception, although it’s likely we failed to capture patients ablated for AF with durable PVI from a prior procedure where a 93656 was not billed.

The primary safety outcome was designed to capture complications while operating in an ambulatory center, where hospital services may be required for unstable patients. Overall these complications occurred in 1.5% of cases. Tamponade occurred in 0.8%, and was comparable to rates seen in the FIRE and ICE trial (0.6%)¹⁵; all these patients underwent pericardiocentesis in the ambulatory center and were stabilized prior to hospital transfer. Of the 3 patients transferred for concern for bleeding, none required surgical intervention. No patients required emergent cardiac or vascular surgery, experienced TIA or CVA, pneumothorax, acute coronary syndrome, death, or required EMS involvement to transfer an unstable patient. The location of the Alaska Heart lab as being on hospital campus is worth mentioning, as some of these patients may have required EMS transport in an off-site facility.

The secondary safety outcome sought to capture patients who required nonemergent hospitalization in that they could not be safely discharged but were never clinically unstable. Overall this rate was low at 1.5%. In an effort to identify patients discharged too soon, we defined a tertiary outcome of ER utilization by time and observed a low rate of ER utilization within 24 hours of the procedure at 1.5%.

Overall safety events and ER utilization rates in this cohort were comparable to published studies evaluating hospital-based SDD. Hospitalization for any reason prior to discharge was 2.9% (n = 14 of 476) in our cohort compared with a 2.5% hospitalization rate for complications following hospital-based AF ablation.¹² The nonemergent hospitalization rate was substantially lower in our cohort, at 1.5%, than in SDD cohorts (21% in Deyell and colleagues, 70% Ignacio and colleagues),^12,16 which is unsurprising, as we selected patients who we expected would not require hospitalization. It is hard to imagine that hospitalization rates would be higher with ambulatory lab-based ablation than hospital-based outpatient ablation. Finally, the 30-day ER visit rate was comparable, at 13.7% in our cohort, with SDD cohorts that ranged from 15.5% to 19.3%.^12,16

It is worth noting the wide breadth of surgery routinely performed in ambulatory centers across the United States, ranging in complexity from tympanostomy tubes to anterior cervical spine fusion. Diagnostic coronary and interventional procedures were previously only offered in hospitals with cardiac surgery backup, and are now performed frequently in the ambulatory setting. Ambulatory surgery centers, compared with inpatient surgeries or procedures, have been shown to reduce both cost and procedure times while providing similar outcomes.¹⁷

While all complications in this ambulatory cohort were stabilized prior to transfer, it remains possible to require surgery for persistent pericardial bleeding, intervention on an ischemic CVA, or coronary intervention, among other possible risks. Within and outside of hospital walls, established relationships with local cardiac and vascular surgery teams and ER physicians, combined with protocols for expedited care, minimize risk to patients. Specific to EP procedures, often done on anticoagulation, ensuring that the availability of anticoagulation reversal and blood products can also mitigate risk.

There are several limitations worth mentioning. This is a single-center retrospective observational study, which does not allow for direct comparison to hospital-based ablation of a similar cohort. Given the aforementioned reimbursement issues, the average age of patients in this cohort is relatively young, and patients were only offered ambulatory ablation if they satisfied the physician’s judgement for safety, and now our current operating protocol (BMI <50 kg/m², ejection fraction >20%, New York Heart Association functional class III or below). That said, a wide range of patient age, systolic function, and BMI were ablated. A total of 55 patients over 65 years of age were included in this cohort, 1 of whom experienced pericardial tamponade included previously, with an over-65 rate of the primary and secondary endpoints of 1.8% and 0%, respectively. The 11 patients with a BMI >50 kg/m² who were ablated did not have any primary or secondary outcome–related complications.

Finally, it is worth mentioning that with the additional availability of a 23-hour stay, ultrasound, and computed tomography imaging, all complications could have been managed without hospital or ER involvement, except for the single patient with a radial arterial line complication.

An ideal center for hospital independent ambulatory ablation in our experience would thus have the following characteristics: availability of critical equipment utilized in one’s current practice setting to include fluoroscopy, hemodynamic monitoring, an electroanatomic mapping system, ablative catheters (cryo, radiofrequency, etc.), intracardiac echocardiography, transseptal access systems, tamponade management kits; geographic proximity to a hospital with protocols in place to alert relevant services (ER, OR, stroke team); anesthesia support and airway management equipment; bradycardia management equipment including temporary or permanent pacemaker implantation capability; access to necessary medications for anesthesia, anticoagulant reversal, etc.; blood product availability; echocardiography, vascular ultrasound, and computed tomography imaging; and 23-hour observation staff and space.

Conclusion

In the largest cohort of AF ablation in an ambulatory EP lab reported to date, we observed a comparable rate of complications to published hospital-based cohorts, all of which were promptly stabilized on site. Safety protocols to manage complications are important within and outside of hospital walls. Design of an ideal surgical center as detailed previously could have allowed for all but 1 patient in this cohort to have been managed without hospital services.

We hope that the lessons learned from this cohort guide patient safety efforts by physician groups who pursue ablation in a similar setting. We also hope to spurn a conversation between professional societies and CMS to expand EP care of Medicare patients to facilities outside of hospital walls.