Author + information
- Received February 10, 2016
- Revision received April 6, 2016
- Accepted April 27, 2016
- Published online November 1, 2016.
- Daniel W. Kaiser, MDa,b,
- Jun Fan, MSb,
- Susan Schmitt, PhDb,
- Claire T. Than, MPHb,
- Aditya J. Ullal, BAb,
- Jonathan P. Piccini, MD, MHScc,
- Paul A. Heidenreich, MD, MSa,b and
- Mintu P. Turakhia, MD, MASa,b,∗ ()
- aStanford University School of Medicine, Stanford, California
- bVeterans Affairs Palo Alto Health Care System, Palo Alto, California
- cDuke University Medical Center, Durham, North Carolina
- ↵∗Reprint requests and correspondence:
Dr. Mintu P. Turakhia, Veterans Affairs Palo Alto Health Care System, Stanford University School of Medicine, 3801 Miranda Avenue, 111C, Palo Alto, California 94304.
Objectives This study sought to explore gender differences in real-world outcomes after catheter ablation of atrial fibrillation (AF).
Background Compared with men, women with AF have greater thromboembolic risk and tend to be more symptomatic. Catheter ablation is generally more effective than antiarrhythmic drug therapy alone. However, there are limited data on the influence of gender on AF ablation outcomes.
Methods We analyzed medical claims of 45 million U.S. patients enrolled in a variety of employee-sponsored and fee-for-service plans. We identified patients who underwent an AF ablation from 2007 to 2011 and evaluated 30-day safety and 1-year effectiveness outcomes.
Results Of the 21,091 patients who underwent an AF ablation, 7,460 (29%) were female. Women, compared with men, were older (62 ± 11 years vs. 58 ± 11 years), had higher CHADS2 (1.2 ± 1.1 vs. 1.0 ± 1.0), higher CHA2DS2-VASc (2.9 ± 1.5 vs. 1.6 ± 1.4), and higher Charlson comorbidity index scores (1.2 ± 1.3 vs. 1.0 ± 1.2) (p < 0.001 for all). Following ablation, women had higher risk of 30-day complications of hemorrhage (2.7% vs. 2.0%; p < 0.001) and tamponade (3.8% vs. 2.9%; p < 0.001). In multivariable analyses, women were more likely to have a rehospitalization for AF (adjusted hazard ratio: 1.12; p = 0.009), but less likely to have repeat AF ablation (adjusted hazard ratio: 0.92; p = 0.04) or cardioversion (adjusted hazard ratio: 0.75; p < 0.001).
Conclusions Women have increased hospitalization rates after AF ablation and are more likely to have a procedural complication. Despite the higher rate of hospital admissions for AF after ablation, women were less likely to undergo repeat ablation or cardioversion. These data call for greater examination of barriers and facilitators to sustain rhythm control strategies in women.
Atrial fibrillation (AF) is the most common sustained arrhythmia in the adult population with estimates that 1 in 4 women develop the arrhythmia at some point in their lifetime (1). In 2014, female gender was incorporated into risk stratification guidelines for stroke prevention in AF based on the CHA2DS2-VASc score, highlighting the importance of gender in AF management (2). Not only are women at an increased risk of thromboembolic complications, they tend to be more symptomatic than men are, and less likely to respond to pharmacologic therapy (3–6). Catheter ablation may therefore seem particularly attractive in women. However, there are limited data on the influences of gender on AF ablation procedures. We therefore sought to explore gender influences on long-term outcomes of catheter ablation of AF using data from multiple insurance health plans throughout the United States.
We analyzed data from the Truven Health MarketScan Commercials Claims and Encounters and Medicare Supplemental Databases (Truven Health Analytics Inc., Ann Arbor, Michigan). This data source contains deidentified medical and pharmacy claims data on inpatient, outpatient, and prescription drug experience of more than 45 million enrolled employees, dependents, and retirees covered under a variety of fee-for-service and managed care health plans, including health maintenance organizations, preferred and exclusive provider organizations, and point-of-service and consumer-directed health plans (7). Linked datasets include the Inpatient Admissions file, which contains all diagnoses and procedures in the service records related to a hospital admission; the Outpatient Services file, which contains information on encounters and claims for services delivered at a doctor’s office, hospital outpatient facility, emergency room, or other outpatient facility; and the Outpatient Pharmaceutical Claims file, which includes complete records of mail-order or card program prescription drug claims. Medical claims are linked to outpatient prescription drug claims and person-level enrollment information. These data sources have been used extensively in health services research of heart rhythm disorders, including for catheter ablation (8–10).
We identified all adult patients with nonvalvular AF who received a catheter ablation between January 1, 2007 and December 31, 2011 (Figure 1). Patients were required to have a primary diagnosis of AF (International Classification of Diseases-9th Revision [ICD-9] code 427.31) associated with any inpatient or outpatient encounter and a documented catheter ablation procedure (ICD-9 code 37.34 or Current Procedural Terminology-4th Edition code 93651). To avoid bias in ascertainment of baseline comorbidities, drug prescriptions, and in post-ablation outcomes, we excluded patients without continuous health plan enrollment for at least 6 months before and 1 month after the index ablation procedure date.
Baseline comorbidities were determined using comorbidity-specific ICD-9 codes up to 1 year before the catheter ablation date, based on the Agency for Healthcare Research and Quality Clinical Classification System. We additionally assessed comorbidity using the Charlson and Selim Comorbidity Indices using validated methods (11). CHADS2 and CHA2DS2-VASc scores were calculated by summing the component comorbidities. Pharmacological therapies received in the 120 days before AF ablation were ascertained using the MarketScan Outpatient Pharmaceutical Claims file.
We ascertained 30-day post-procedural complications based on hospital discharge diagnoses or procedure codes that would be used to manage a complication (e.g., pericardiocentesis). This approach has been previously applied to AF treatment studies using the same data source (8,12). One-year clinical outcomes were similarly ascertained and included hospitalization for AF, cardioversion (inpatient or outpatient), or repeat catheter ablation of AF. For the endpoint of AF hospitalization, we considered any inpatient encounter associated with a primary diagnosis for AF, which occurred after the initial inpatient or outpatient AF ablation procedure. We evaluated cause-specific hospitalization for acute coronary syndrome or myocardial infarction (MI), heart failure, supraventricular tachycardia, ventricular arrhythmia, syncope, stroke or transient ischemic attack (TIA), and intracranial hemorrhage.
We compared differences in baseline characteristics and 30-day complications between females and males using Student t tests for continuous variables and chi-square tests for categorical variables. Cox proportional hazards regression was used to examine the association between gender and each outcome at 1 year (hospitalization for AF, cardioversion, repeat catheter ablation of AF), while adjusting for potential confounders. We included the following baseline covariates in multivariate models: age, sex, Charlson Comorbidity Index score, congestive heart failure, hypertension, diabetes, stroke/TIA, prior MI, anemia, peripheral artery disease or vascular disease, year of ablation, insurance plan, and receipt of concomitant drug therapies (warfarin, dabigatran, clopidogrel, class I and class III antiarrhythmic drugs, amiodarone, beta-blockers, and calcium channel blockers). The Cox model assumption of proportional hazards was found to be valid by using the Schoenfeld test. We also used the Kaplan-Meier method and log-rank tests to calculate differential probability of event-free survival using the composite endpoint of AF hospitalization, cardioversion, or repeat AF ablation within 1 year.
The study was approved by the local Institutional Review Board. All analyses were performed using SAS version 9.1 (Cary, North Carolina) and STATA version 11.0 (College Station, Texas). The funding agencies were not involved in any way with study design, data analysis, interpretation of results, or manuscript preparation.
We identified 21,091 patients with an index AF ablation meeting inclusion criteria. Although procedure volume increased over the study period (2007 to 2011), 30% of the ablations were performed in women, and this proportion did not substantively change over the observation period. At the time of catheter ablation, women, compared with men, were older (62 ± 11 years vs. 58 ± 11 years; p < 0.001), and had higher CHADS2 (1.2 ± 1.1 vs. 1.0 ± 1.0; p < 0.001), CHA2DS2-VASc (2.9 ± 1.5 vs. 1.6 ± 1.4; p < 0.001), and Charlson comorbidity index scores (1.2 ± 1.3 vs. 1.0 ± 1.2; p < 0.001) (Table 1). Women, compared with men, also had a higher prevalence of hypertension (66.2% vs. 61.8%; p < 0.001), diabetes (22.8% vs. 20.7%; p < 0.001), prior stroke or TIA (5.7% vs. 3.5%; p < 0.001), and anemia (13.6% vs. 8.2%; p <0 .001) but had a lower prevalence of prior MI (4.5% vs. 5.6%; p < 0.001). There was no significant difference in prevalence of heart failure (33.4% vs. 33.6%; p = NS) or chronic kidney disease (8.0% vs. 7.4%; p = NS).
Before ablation, women were more likely to be treated with rate-control agents (70.6% vs. 63.0%; p < 0.001) and class I antiarrhythmic agents (28.0% vs. 23.8%; p < 0.001) but less likely to be treated with amiodarone (13.9% vs. 16.0%; p < 0.001). Women had similar use of class III antiarrhythmic agents (20.4% vs. 19.7%, p = NS) and oral anticoagulation (66.0% vs. 67.8%, p = NS).
In the 30 days following the ablation, women had an increased risk of complications, including vascular complications (2.7% vs. 2.0%; p < 0.001), hematoma or hemorrhage (2.3% vs. 1.6%; p < 0.001), and perforation or tamponade (3.8% vs. 2.9%; p < 0.001) (Table 2). Women demonstrated a trend for increased all-cause hospitalization (9.4% vs. 8.6%; p = 0.07) and stroke or TIA (0.85% vs. 0.64%; p = 0.09). There were 7 in-hospital deaths (0.07% vs. 0.02%; p = NS).
In the year following ablation, women were less likely to receive a cardioversion (17% vs. 21%; adjusted hazard ratio [HR]: 0.75; p < 0.001) or repeat ablation procedure (13% vs. 15%; adjusted HR: 0.92; p < 0.001) (Table 3). However, unadjusted and adjusted risk of 1-year all-cause hospitalization was higher in women compared with men (32% vs. 27%; p < 0.001; adjusted HR: 1.14; 95% confidence interval [CI]: 1.07 to 1.20; p < 0.001) as was risk of hospitalization for AF (13% vs. 12%; adjusted HR: 1.12; 95% CI: 1.03 to 1.22; p < 0.001). In secondary analyses, we found that women were less likely to be rehospitalized for MI (0.7% vs. 1.2%; adjusted HR: 0.84; 95% CI: 0.42 to 0.84; p < 0.05) and were more likely to be admitted with congestive heart failure over the following year (2.5% vs. 1.8%; adjusted HR: 3.45; 95% CI: 2.68 to 4.45; p < 0.001) (Table 4). Cumulative event-free survival, defined as freedom from AF rehospitalization, was decreased among women compared with men (86.6% vs. 89.0%; adjusted HR: 0.88; 95% CI: 0.78 to 0.97; p < 0.01) as shown in Figure 2.
We evaluated the 30-day safety and 1-year effectiveness outcomes in more than 20,000 patients undergoing catheter ablation procedures for AF using medical claims data from a variety of representative healthcare plans throughout the Unites States. We found that gender was significantly associated with certain aspects of the procedure, including the baseline clinical characteristics of the patients, the safety profile, and long-term outcomes. Adjusting for risk factors, women revealed a paradoxical discordance between increased rehospitalizations for AF and decreased subsequent use of cardioversion or repeat ablation. Our data call for greater examination of barriers and facilitators to sustain rhythm control strategies in women with AF.
In agreement with prior studies, we found women undergoing catheter ablation of AF were older with increased comorbidities compared with men (13–16). This pattern may be indicative of gender influences on the natural history of AF. Population studies, such as Framingham, have demonstrated that new-onset AF tends to occur at an age approximately 5 years older in women and are more likely to have comorbidities, such as hypertension and diabetes, at the time of diagnosis (17,18). Although women display a similar overall prevalence of AF as men (19), the older age at diagnosis may decrease the likelihood of undergoing catheter ablation. We also found that women were more likely to be treated with class I antiarrhythmic medications, which may have been driven by a lower prevalence of prior MI, a contraindication to this class of medication.
Compared with men, women were found to have a low, but modestly higher, risk of complications within 30 days of the catheter ablation procedures. These findings are consistent with prior studies and confirm the increased procedural risk for women in our large study population, which is representative of most patients receiving AF ablation today (20). Prior studies have reported 30-day complication rates that range from 5% to 8% in women, roughly 1% to 3% higher compared with men (12,21,22). The higher complication rate seen in women may be partially explained by the increased comorbidity burden. However, because of the low frequency of events, we did not perform multivariate adjustment. Other potential explanations may include differences in pharmacokinetics and body size. For example, women tend to require a lower heparin dose to obtain adequate anticoagulation, even after adjusting for weight (23). Concordantly, women have demonstrated higher activated clotting times during catheter ablation procedures (24). Fortunately, the overall rates of both stroke (<1%) and death (<0.1%) were low and similar between genders.
With respect to gender differences with longer-term procedural success, prior studies have not revealed consistent findings (13,14,25–27). Several studies have reported that women have a higher recurrence rate after the ablation procedure (13,14,25), whereas other studies have found no significant difference between men and women (26,27). Our study, which is the largest AF ablation study examining procedural outcomes by gender in the United States, found that women were significantly more likely to be rehospitalized with AF within 1 year after an ablation procedure but less likely to undergo cardioversion or repeat ablation. Although the nature of the claims data does not allow for evaluation of any AF recurrence, our findings, in context, may be indicative of potential barriers to optimal or sustained rhythm control strategies in women. For example, women are more likely than men to report having delayed care because of nonmedical barriers, including time logistics and cost (28). Furthermore, women are more likely to have their cardiac symptoms and risk of heart disease downplayed by their doctors (29); and are less likely to receive implantable defibrillators (30). In the case of AF, there is evidence that women exhibit a delayed referral pattern compared with men and fail more antiarrhythmic medications before proceeding to catheter ablation (15,16). Delaying the time to ablation could promote a higher AF or comorbidity burden at the time of ablation, which may have resulted in increased electrical and structural remodeling. However, it is unclear from our data if women were less likely to be offered rhythm-control strategies, or if women were more likely to decline the procedures. The underlying motivations driving clinical decision-making are complex and may vary by gender (31). For example, a single-center study from Japan suggested women were more likely to refuse catheter ablation (14). In context, these data highlight the importance of communication between patients and clinicians to ensure that fears, motivations, and perspectives of patients and caregivers are adequately addressed to deliver optimal and patient-centered care.
First, we used deidentified administrative claims data. Although claims data have the advantages of performing large-scale analyses across multiple health care systems or payers, claims-based ascertainment of ablation procedures and outcomes may be imperfect. AF ablation ascertainment based on procedural and diagnosis could lead to misclassification and inappropriate inclusion of other ablations, such as atrial tachycardia, which tend to be more prevalent in women, although cohort inclusion was conditioned on a concurrent AF diagnosis (32,33). However, because AF ablation has higher risk than atrial tachycardia or supraventricular tachycardia procedures (34), this misclassification would bias complication risk in women toward the null. Second, unidentified confounders and covariates influencing patient outcomes, such as repeat ablation, or cardioversion, such as frailty or body mass index, may not be identified. Third, we are unable to measure differences in AF recurrence, AF burden, or quality of life, which are the key endpoints used in randomized trials of ablation. Fourth, we are unable to classify AF severity (e.g., paroxysmal vs. persistent), AF duration before ablation, heart failure severity, or ejection fraction from administrative data. The ICD-10 medical classification system has separate diagnoses codes for AF severity and may prove useful for future work. Finally, although the baseline differences and outcomes provide insight, the clinical circumstances or patient preferences associated with care cannot be identified. However, these findings provide the rationale for other study designs that may allow more granular and qualitative examination of these issues.
In 21,091 patients undergoing catheter ablation of AF across the United States, women, compared with men, were older, with increased comorbidity burden, and had a low but modestly increased risk of procedural complications. Adjusting for risk factors, women who underwent catheter ablation revealed a paradoxical discordance between increased rehospitalizations for AF and decreased subsequent use of cardioversion or repeat ablation. These data call for greater examination of barriers and facilitators to sustain rhythm control strategies in women.
COMPETENCY IN MEDICAL KNOWLEDGE: Similar to stroke risk, women with atrial fibrillation have a low, but modestly higher rate of procedural complications in the 30 days following atrial fibrillation ablation.
TRANSLATIONAL OUTLOOK: The higher risk of 30-day complications in women may be mitigated by improved patient selection or tailoring aspects of the procedure to minimize complication rates. Additionally, a greater examination of barriers and facilitators of care and use of shared decision-making tools could help to minimize potential disparities in care and ensure that treatment decisions are in alignment with the clinical evidence and patient preferences.
Dr. Kaiser is supported by a grant from the National Heart, Lung and Blood Institute (1T32Hl098049). Dr. Piccini receives grants for clinical research from Boston Scientific, Johnson & Johnson, and St. Jude Medical; and serves as a consultant to Johnson & Johnson and Medtronic. Dr. Turakhia is supported by a Veterans Health Services Research & Development Career Development Award (CDA09027-1), an American Heart Association National Scientist Development Grant (09SDG2250647), a VA Health Services and Development MERIT Award (IIR 09-092), the Gilead Sciences Cardiovascular Scholars Program, and the Stanford Center for Health Research on Women and Sex Differences in Medicine; and is a consultant for Medtronic, St. Jude, and Boehringer-Ingelheim. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- confidence interval
- hazard ratio
- International Classification of Diseases-9th Revision
- myocardial infarction
- transient ischemic attack
- Received February 10, 2016.
- Revision received April 6, 2016.
- Accepted April 27, 2016.
- American College of Cardiology Foundation
- Lloyd-Jones D.M.,
- Wang T.J.,
- Leip E.P.,
- et al.
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Rienstra M.,
- Van Veldhuisen D.J.,
- Hagens V.E.,
- et al.
- Kerr C.R.,
- Humphries K.
- ↵Truven Health MarketScan® Commercial and Medicare Supplemental Databases. Available at: http://img.en25.com/Web/TruvenHealthAnalytics/PH_11609_0213_MarketScanResearchDatabases_SS_web.pdf. Accessed February 1, 2016.
- Reynolds M.R.,
- Gunnarsson C.L.,
- Hunter T.D.,
- et al.
- Deshmukh A.,
- Patel N.J.,
- Pant S.,
- et al.
- Humphries K.H.,
- Kerr C.R.,
- Connolly S.J.,
- et al.
- Piccini J.P.,
- Hammill B.G.,
- Sinner M.F.,
- et al.
- Piccini J.P.,
- Sinner M.F.,
- Greiner M.A.,
- et al.
- Kummer B.R.,
- Bhave P.D.,
- Merkler A.E.,
- Gialdini G.,
- Okin P.M.,
- Kamel H.
- Dagres N.,
- Clague J.R.,
- Breithardt G.,
- Borggrefe M.
- Forleo G.B.,
- Tondo C.,
- De Luca L.,
- et al.
- U.S. Department of Health and Human Services,
- Health Resources and Services Administration,
- Maternal and Child Health Bureau