OUP user menu

Gender-related differences in catheter ablation of atrial fibrillation

Giovanni B. Forleo , Claudio Tondo , Lucia De Luca , Antonio Dello Russo , Michela Casella , Valerio De Sanctis , Fabrizio Clementi , Rafael Lopes Fagundes , Roberto Leo , Francesco Romeo , Massimo Mantica
DOI: http://dx.doi.org/10.1093/europace/eum144 613-620 First published online: 18 July 2007


Aims Women have an increased risk for atrial fibrillation (AF)-related complications and there is evidence towards a reduced efficacy of the rhythm control strategy than men. A catheter-based strategy is therefore widely attractive, but the impact of gender on catheter ablation (CA) of AF remains undefined.

Methods and results We included 221 consecutive patients (150 men) who underwent CA of drug-refractory AF. Gender differences in clinical presentation and outcomes were compared. Women were older (P = 0.002), had a longer history of AF (P = 0.04), and were more likely to have hypertension (P = 0.04). Moreover, a concomitant valvular heart disease tended to be more common in women (32.4 vs. 23.3%; P = 0.28) and left atrium dimensions were significantly larger (P = 0.003). However, acute success rate and complications rate were similar between genders. After 22.5 ± 11.8 months of follow-up, the overall freedom from arrhythmia recurrences was similar (83.1 vs. 82.7% in men), and a similar improvement in SF-36 quality of life scores was achieved in both groups.

Conclusion Women are referred for AF ablation later with a more complex clinical pre-operative presentation. Despite this higher risk profile in women, no differences were detected in clinical outcomes. Our findings indicate that CA of AF appears to be safe and effective in women as in men.

  • Atrial fibrillation
  • Catheter ablation
  • Arrhythmia
  • Gender


Continuing advances in the field of catheter ablation (CA) for atrial fibrillation (AF) are leading more patients being offered this treatment option.16 Gender has proved important in the characterization and management of a variety of cardiovascular disorders including AF.7 Recent studies demonstrated gender-related differences in clinical presentation, outcome, and management of patients with AF. Women are significantly undertreated, have a worse outcome, and importantly a reduced efficacy of the rhythm control strategy that appears to enhance cardiovascular morbidity and mortality.810 Furthermore, a decreased quality of life (QoL) in women vs. men regardless of treatment strategy (rhythm vs. rate control) has been recently recognized.9 Therefore, a curative CA approach for AF appears a very attractive alternative in women.11 In contrast, anatomical gender differences due to smaller heart size and pulmonary vein (PV) antra in women might affect the ease of catheter manipulation inside the heart chambers as well as their outcome, including success and complication rates.12

However, no objective evidence to date has shown the impact that gender may have in referral patterns and long-term outcome for AF ablation. In the present investigation, we have addressed this question using a cohort of patients undergoing CA of symptomatic drug-refractory AF.


Study population

Consecutive patients referred for AF ablation to the St Ambrogio Clinical Institute in Milan and the St Camillo-Forlanini Hospital in Rome were analysed. The ablation was performed in patients with highly symptomatic and drug-refractory AF. Exclusion criteria were as follows: age < 18 or = 75 years, absence of informed patient consent, and any condition that would make survival for 1 year unlikely. Furthermore, patients in whom a previous CA for AF failed were excluded from the analysis. The final study population consisted of 221 patients (150 men; 67.9%). Baseline characteristics of the study subjects are described in Table 1.

View this table:
Table 1

Baseline clinical features of patients according to gender

Electrophysiological study and radiofrequency ablation

All patients had effective anticoagulation for = 1 month followed by subcutaneous fractionated heparin 3–5 days before the procedure. Antiarrhythmic drugs (AAD) including amiodarone were not discontinued. Immediately before the procedure, patients with a high embolic risk or presenting with AF underwent transoesophageal echocardiography to rule out the presence of atrial thrombi. Catheter electrodes were inserted with the use of one or both femoral veins. The left atrium (LA) was accessed by double transseptal puncture or via an open fossa ovalis. After transseptal access, intravenous heparin was infused to maintain an activated clotting time of at least 300 s.

Pre-ablation angiograms of all accessible PVs were performed. Non-fluoroscopic-guided mapping and isolation of PV vestibula were applied to all patients with the Nav-X mapping system (Endocardial Solution, Inc., USA).13 Radiofrequency energy was delivered using a 3.5 mm cooled-tip catheter at a target temperature set to 45°C and a maximal power output of 35 W (Figure 1).

Figure 1

Example of left atrial chamber reconstruction with the NaV-x system. Posterior view: orange and red dots identify left superior and inferior pulmonary veins, respectively. Green and yellow dots identify the right pulmonary veins. Brown dots indicate lesions created around the pulmonary vein vestibules and the linear lesion at the left mitral isthmus. The snapshot also illustrates the activation pattern of the left atrial lateral wall during distal coronary sinus pacing, after creation of the bidirectional block along the left mitral isthmus.

The goal of PV isolation was to abolish all PV potentials as measured by the loop-shaped multipolar mapping catheter at the PV ostium. In the patients in whom AF was still present after completion of ablation, either flecainide or electric cardioversion was performed to restore sinus rhythm.

Our policy is to perform cavo-tricuspid isthmus (CTI) ablation at the end of the LA ablation in a single procedure. In patients with prior CTI ablation, bidirectional block was re-assessed during the procedure. Moreover, at the discretion of the operator, a block at roofline joining the superior PVs and a block at the isthmus between the mitral annulus and the left inferior PV were performed. Left isthmus line was created with the target of bidirectional block, assessed by pacing from the two opposite sites of the line (i.e. left atrial appendage and the posterior mitral annulus).

Post-ablation management

After the procedure, each patient was evaluated with an intrahospital continuous ECG monitoring. Discontinuation of any antiarrhythmic treatment was considered in each patient according to the clinical pre-operative presentation and the clinical course during follow-up. Our policy is to discharge patients on AADs. Discontinuation takes place within 1 month in patients without structural heart disease and up to 3 months in the remaining patients. Seventy-seven patients (34.8%) received a trans-telephonic ECG recorder for at least 3 months (three ECG transmissions per week at fixed days and anytime patients perceived symptoms potentially related to arrhythmia relapse). Moreover, all patients were instructed to obtain a 12-lead ECG in the event of palpitations. Subcutaneous heparin was continued until the international normalized ratio was = 2. If no atrial arrhythmias were detected within the first 3–6 months, coumadin was discontinued. In the event of recurrent symptomatic arrhythmias, patients were offered an additional ablation after a trial of AADs.


After discharge, patients were periodically re-evaluated at 1, 3, 6 months and every 6 months thereafter. At each visit, they were asked whether medical events or symptoms suggestive of cardiac arrhythmias occurred and an ECG Holter monitoring was performed to detect the presence of asymptomatic arrhythmias. Because early recurrences of AF may be a transient phenomenon,14,15 a blanking period of 1 month after ablation was used. Freedom from AF and left atrial tachycardia (LAT) with or without AAD therapy, cardiac re-admission, death, stroke, repeat procedure, and complication rates was compared between the two groups. Furthermore, using the Medical Outcomes Study 36-Item Short-Form General Health Survey (SF-36), we assessed patients at baseline and 6 months after the procedure with respect to several indexes of QoL.1618

Statistical analysis

Categorical data are summarized using absolute values (percentage). Continuous data are presented as mean (SD) or, where shown, as median (interquartile range). Continuous variables were tested for normal distribution by the Kolmogorov–Smirnov test. Non-continuous variables expressed as proportions were compared with using χ2 analysis or Fisher's exact test. Comparison between groups was performed with either the Student's t-test or, when data were not normally distributed, the Wilcoxon rank sum or signed rank test. Baseline QoL and change in QoL over time between the groups were compared with the Wilcoxon test. Paired non-parametric exact methods were used to compare the change in QoL over time for each patient. Long-term survival was described with the Kaplan–Meier method, and comparisons were made by use of the log-rank statistic. Cox proportional hazard methods were used to adjust for baseline differences when investigating the associations between periprocedural drug use and outcomes. All P-values were two-sided, and a P-value of < 0.05 was considered to indicate statistical significance.


Patient characteristics

A comparison of baseline characteristics for men and women (Table 1) shows that there were significant differences between the two populations. All patients had symptomatic AF for a median of 48 months, despite a mean number of 2.8 ± 1.2 AADs. Paroxysmal AF was present in 92 men (61.3%) and 40 women (56.3%) (P = NS), persistent AF in 54 men (36.0%) and 27 women (38.0%) (P = NS), and permanent AF in 4 men (2.7%) and 4 women (5.6 %) (P = NS). Among the 221 study patients, there was a predominance of men (n = 150; 67.9%) when compared with women (n = 71; 32.1%). The average age of all of the patients was 58.2 ± 10.5 years. Compared with men, women were generally older (61.6 vs. 56.9; P < 0.002) and more likely to have hypertension (52.1 vs. 30.7%; P = 0.043), valvular heart disease (15.5 vs. 5.3%; P = 0.02), or a longer history of AF (60 vs. 47 months, median; P = 0.042) and were less likely to have lone AF (39.4 vs. 56.0%; P = 0.18) or paroxysmal AF (56.3 vs. 61.3%; P = NS). Moreover, a concomitant structural heart disease was more common in women (32.4 vs. 23.3%; P = 0.28) and LA dimensions were significantly larger (44.0 ± 6.5 vs. 40.6 ± 6.3; P = 0.0026). On discharge, women were more likely on statins (21.1 vs. 8.7%; P = 0.025) and on angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers (35.2 vs. 23.3%; P = 0.16) than men. No significant difference was found between groups in terms of the presence of coronary heart disease, left ventricular ejection fraction, and number of AADs used before ablation. The overall mean follow-up period was 22.5 ± 11.8 (range 5–43) months with no difference between genders.

Procedure-related parameters and initial results

A total of 250 ablation procedures in 221 patients (i.e. 29 redo procedures) were performed (Table 2). There were three initial ablation failures (1.4%; men, 2) related to early procedural complication. Pulmonary vein isolation was achieved in the remaining patients with no differences in radiofrequency (RF) delivery time (34.5 ± 11.3 vs. 36.3 ± 14.5 min) and fluoroscopy time (80.1 ± 20.4 vs. 80.7 ± 21.0 min) for men and women, respectively. The mean total procedure time was 228.8 ± 58.0 min for men and 221.4 ± 57.3 min for women (P = NS).

View this table:
Table 2

Procedural features and clinical outcome

Successful bidirectional CTI block was achieved with no differences between men and women (88.0 vs. 90.1%). Mitral isthmus ablation was performed in 59 men (39.3%) and 24 women (33.8%) (P = NS), most of whom required additional RF pulses from within the coronary sinus. The endpoint of mitral isthmus ablation was reached in 63 of 83 (76.0 %) patients. In addition, roofline ablation was performed in 21 men (14%) and 7 women (9.9%) (P = NS).


Four men (2.7%) and one woman (1.4%) (P = NS) had mild pericardial effusion managed medically (Table 2). Pericardial tamponade occurred in two men (1.3%) and in two women (2.8%) (P = NS). This was managed by surgical repair in a man and by percutaneous pericardiocentesis in the other three patients. Transient neurological events occurred in two men. All patients were discharged home without any sequelae. A computed tomography scan to evaluate PV stenosis was performed, at physician discretion, in 117 of 221 patients (52.9%). During long-term follow-up, two patients had moderate-to-severe PV stenosis ( = 50%) requiring stent placement in one case.


Before discharge, 15 patients (9 men) developed spontaneous atrial arrhythmia episodes. All but two (0.9%) were discharged because of sinus rhythm. Except in 41 men (27%) and 28 women (39%), AADs were permanently discontinued within 3 months after the procedure. Twenty-five men and 15 women received AADs after documentation of atrial arrhythmia, whereas the remaining patients maintained drugs according to the recommendations of the patient's attending physicians, independent of the results of the ablation.

Following a median of 258 days after the first procedure (Table 2), a repeat ablation was performed in 29 patients (13.1%): 12 patients for recurrent AF (9 men and 3 women; P = NS) and 17 patients for LAT (13 men and 4 women; P = NS). The repeat ablation was successful in 19 of 29 patients (14 men and 5 women; P = NS), remaining free from any recurrence of arrhythmias at the latest assessment.

At a mean follow-up of 22.5 ± 11.8 months after the last ablation procedure, 82.7% of men and 83.1 % of women were arrhythmia-free (hazard ratio 0.89, 95% CI 0.44–1.78; P = 0.75), with an 89.1% of 1-year arrhythmia-free survival. The success rate free of AAD was 74.0% in men and 67.6% in women (P = NS). Seventeen patients (7.7%) had recurrence of LAT and 21 patients (9.5%) had recurrence of AF. A schematic representation of clinical outcomes is depicted in Figure 2. A Kaplan–Meier curve of arrhythmia-free survival in men and women is shown in Figures 3 and 4.

Figure 2

Schematic representation of ablation results in the 221 study patients. LAT, left atrial tachycardia; AF, atrial fibrillation; SR, sinus rhythm; RF, radiofrequency ablation; AAD, antiarrhythmic drugs.

Figure 3

The Kaplan–Meier curve of survival free from recurrences after 1 month of the blanking period. Time until first recurrence was not different between male and female patients.

Figure 4

The Kaplan–Meier analysis of freedom from atrial fibrillation/left atrial tachycardia recurrence by sex in patients with or without structural heart disease. P = NS for comparisons between genders. P = 0.4 for comparing patients with or without structural heart disease (HR 1.96; IC 1.02–3.79).

Among the 15 patients with arrhythmia recurrences after paroxysmal AF ablation, 13 patients (87%) reported having significantly fewer symptoms than before the ablation. Interestingly, among the patients with arrhythmia recurrences after the ablation of a previously considered persistent or permanent AF, sinus rhythm was present at the latest assessment in 6 of 7 women (86%) and 10 of 16 men (62%) (P = NS). There was no difference between gender in the number of hospitalization (18.3 vs. 24.7% in men) recorded as being primarily or partially arrhythmia-related, including that required for redo ablations.

Additional ablation lines in LA were not associated with a higher success rate; indeed, there was no difference between patients with and without recurrences after the first procedure [19/40 (48%) in men with recurrence vs. 32/108 (30%) in men without recurrence and 6/17 (35%) in women with recurrence vs. 15/53 (28%) in women without recurrence; P = NS].

In a multivariate analysis controlling for age, history of hypertension, and structural heart disease, periprocedural drug use with statins or renin–angiotensin system blockers was not independently associated with a reduction in the risk of adverse outcomes.

A 66-year-old patient with a non-ischaemic cardiomyopathy died of cancer 7 months after RF ablation. Two patients developed a stroke after 16 and 25 months. The first was a 61-year-old male, known to have coronary heart disease. Stroke occurred 3 months after the recurrence of an LAT while he was on long-term anticoagulation. The second was a 67-year-old male with a history of hypertension, but as he had no evidence of AF after the procedure, he received only aspirin. A 62-year-old male with ablation failure and non-ischaemic-dilated cardiomyopathy underwent cardiac resynchronization therapy. After 26 and 15 months, two patients (one male) underwent implantation of a dual-chamber pacemaker for symptomatic sinus nodal dysfunction. Symptomatic bradycardia requiring withdrawal of AAD occurred in two men and one woman, all known to have hypertension or structural heart disease. No further AAD adverse effects were detected. No patient was lost to follow-up.

Changes in QoL before and after the ablation

Baseline or after procedure QoL surveys were missing in 51 patients (23%), including the patient who died. QoL was analysed in the remaining patients. There were no significant baseline differences between groups (Table 3). Of interest, one of the two groups experienced a similar significant improvement in QoL and arrhythmia-related symptoms. Overall, the evolution scores obtained with an SF-36 instrument were clearly positive (P < 0.05 vs. before ablation). Although a trend towards a better improvement in QoL scores was observed in women when compared with men, this was not statistically significant.

View this table:
Table 3

SF-36 quality-of-life scores, by sex, before and after the ablation


The present study provides a detailed analysis of the gender differences in referral pattern and clinical outcome of RF ablation in patients with drug-refractory AF. Despite a female having more complex and higher risk clinical profile before ablation, we detected similar outcomes in both genders, with no difference in success, complication, and recurrence rates.

Women under-representation and delayed referral

Although data are limited in two centres, there is evidence to suggest that women may be subject to delayed referrals or impaired access to care for symptomatic AF. The absolute number of males and females with AF in the general population is expected to be similar.19,20 In contrast, the present study documents an apparent under-representation of women; this is in line and supported by the observation from gender differences as described in other cardiac disorders such as coronary artery disease, arrhythmias, and heart failure.2127 Furthermore, Dagres et al.12 demonstrated that women are referred for ablation of supraventricular tachycardia (SVT) later than are men. Our data corroborate these findings suggesting the lack of attention towards early clinical signs and treatment in women when compared with men. Indeed, in the present study, women were older with more advanced disease and significantly more symptomatic than men. A possible explanation is that symptoms are more likely to be attributed to stress, panic, or anxiety in women than men. In addition, studies have shown that women suffering from AF are significantly older with a greater prevalence of structural heart disease. These characteristics may in part explain a lower women referral for AF ablation procedure.

Gender and clinical outcome

Recent studies demonstrated gender-related differences in referral and treatment policies and in the outcomes of patients with AF.8,9,28,29 Female sex is a well-known independent risk factor for thrombo-embolism30,31; furthermore, women are more symptomatic with a worse outcome and a higher rate of recurrence after cardioversion than men. A substudy of the Rate Control versus Electrical Cardioversion (RACE) study9 demonstrated that women, but not men, are at significantly increased risk of cardiovascular morbidity and mortality between the rhythm control vs. rate control arms. As it has been hypothesized that an increased sensitivity to adverse effects of AAD might account for this difference, a rate control strategy has been suggested in women avoiding AAD employment. Thus, curative CA of AF appears a very attractive alternative in women as antiarrhythmic therapy is usually ineffective in preventing both arrhythmia recurrences and clinical events related to them. Conversely, the anatomical gender differences due to smaller heart size and PV antra in women might affect the ease of catheter manipulation inside the heart chambers as well as their outcomes, including success and complication rates.12 This study provides data that could overcome these statements, showing that CA of AF provides uniform benefit in men and women in terms of effectiveness in maintaining sinus rhythm, free from AAD, and QoL improvement.

In the present study, despite women have more comorbidities than men, female sex was not associated with an increased risk of arrhythmia recurrences nor with an increased risk of CA complications. Maintenance of sinus rhythm has been achieved in the vast majority of patients; however, some patients were on AAD. Although no significant difference in AAD use was detected between genders, the greater percentage of women (39 vs. 27%) with AAD after 3 months in the absence of AF recurrences may reflect the higher risk pre-operative presentation and not the results of the ablations. We believe that this reflects the apparent lower success rate in females not treated with AAD (67.6 vs. 74.0%; P = NS) reported in Table 2. More importantly, no gender difference in adverse effects related to the underlying heart disease or AAD use was observed. This is an important issue because as mentioned before, women are at an increased risk of AAD adverse effect; besides in our study population, women had more comorbidities and therefore more prone to adverse effect than men.

The lack of a significant detection of AAD adverse effects in women could be related to the small number of women with structural heart disease and the use of low doses AAD over a short period of time.

The rate of AF recurrences after a mean of 22.5 ± 11.8 months of follow-up was 8.7%. Atrial fibrillation recurrences were equally distributed between men and women and our results compare favourably with previous series on CA of AF. Conversely, in the present study, the LAT recurrences were 7.7%, which appears to be higher than reported in some studies3234 and equally balanced between men and women. The reason for this higher LAT recurrence is not clear, but several hypotheses may explain this finding. We suspect that these differing results reflect differences in technique and a somewhat dissimilar patient population. Indeed, the magnitude of the LAT diagnosis in the study population prior to ablation is not known. In addition, because AADs can organize AF into LAT,35,36 the former could have been the primary arrhythmia in some patients. Therefore, a lower use of AAD after the procedure would have allowed AF resumption, leading to a final lower LAT recurrences. One can also speculate about the possible arrhythmogenic properties of RF applications and the role of additional lines in LA to prevent LAT recurrences. This issue was not evaluated in our study. It is noteworthy that in our study, arrhythmia recurrences were not prevented with additional lines in LA, although a potential bias is that linear lesions were relegated to more compromised patients.

As shown in Table 1, there is evidence that periprocedural drug use with statins or renin–angiotensin system inhibitors was more common in women than men. This is in line with the observation that the two groups are not comparable because women have more comorbidities than men. One could argue that although these drugs are not antiarrhythmic in the conventional sense, a beneficial role in AF patients has been described in several reports.37,38 Nevertheless, in our population, their use was not independently associated with a reduction in the risk of adverse outcomes in women, compared with men.

Surprisingly, no significant differences in baseline SF-36 QoL scores were detected between genders. However, six of eight baseline QoL scores were lower in women when compared with men. This observation, according to other reports showing lower QoL scores among women suffering from AF, could reflect a trend towards worse baseline scores in women. However, QoL was analysed in 170 patients (76.9%), to find any potential difference, a much larger number of patients are certainly required. Although not statistically significant, interestingly, women showed a better improvement in QoL scores when compared with men. This finding deserves additional studies, a major goal of AF ablation is to improve QoL, and our results suggest more benefits for women.


Our study reflects the experience of two centres. Of note, no significant differences were detected in referral patterns and outcomes of CA between the two institutions. Although clinical and demographic characteristics as well as ablation data and follow-up outcomes were prospectively collected, the study was not originally designed to prospectively evaluate the impact of gender on CA of AF. Additionally, the smaller sample size of women might account for a failure to detect important differences between groups, particularly for QoL. We may also have underestimated the recurrence rate because of asymptomatic undocumented arrhythmia episodes, but we suppose that these are likely to be equally distributed in both groups. Another important issue is that a decrease in PV size after ablation is common,39 and a computed tomography scan to evaluate PV stenosis was not routinely performed in all patients. We therefore could not evaluate whether gender-related differences in the LA and PV anatomy might predispose to PV stenosis; this issue requires further investigations.

Finally, the present study cannot be easily compared with others because the study population is inevitably dissimilar. RACE/AFFIRM patients were older with more severe underlying heart disease. Whether our results can be generalized and CA may become therapy of choice for a broader AF population remain controversial. A multicentre prospective study would strengthen the validity of the findings.


This study documents a variety of distinctions between male and female patients with regard to the clinical presentation, but with comparable outcome of AF ablation. Women are at higher risk than men for AF-related complications and the treatment is often delayed. Catheter ablation of AF provides a significant clinical benefit and appears to be as effective in women as in men. The results of our study highlight the need for heightened attention to the catheter-based treatment of AF in women.

Conflict of interest: none declared.


View Abstract