Outcome of catheter ablation in the very elderly‐insights from a large matched analysis

Abstract Background Ablation emerged as first line therapy in the treatment of various arrhythmias. Nevertheless, in older patients (pts), decision is often made pro drug treatment as more complications and less benefit are suspected. Hypothesis We hypothesized that different kind of ablations can be performed safely regardless of the pts age. Methods We enrolled all pts aged >80 years (yrs) who underwent ablation for three different arrhythmias (atrial flutter [AFL], atrioventricular nodal re‐entry tachycardia [AVNRT], ventricular tachycardia [VT]) between August 2002 and December 2018. Procedural data and outcome were compared with matched groups aged 60 to 80 years and 40 to 60 years, respectively. Periprocedural and in‐hospital complications were analyzed. Results The analysis included 1191 patients (397 pts per group: 63% AFL, 23% AVNRT, 14% VT) who underwent ablation. Acute success was high in all types of arrhythmias irrespective of age (>80, 60‐80, 40‐60 years: AFL 97%/98%/98%, AVNRT 97%/95%/97%, VT 82%/86%/93%). Rate of periprocedural complications were similar in all groups treated for AFL and AVNRT. For VT ablations significant differences were noted between pts > 80 or 60 to 80 years and those aged 40‐60 years (16.1%/14.3%/3.6%). Most complications were infections and groin haematoma. No strokes, iatrogenic atrioventricular blocks and deaths related to the ablation occurred. Conclusion Ablation appears safe in pts > 80 years. Success rates were comparable to matched younger cohorts. A significant difference was observed for VT patients.


| INTRODUCTION
As life expectancy is steadily increasing and quality of life remains high also in the very elderly, the health system is in a constant transition and challenges have changed over time. With increasing age arrhythmias are becoming clinically manifest in a growing number of patients, too. The need and wish for definitive therapy is also growing due to improved ablation techniques and consequently higher success rates. Catheter ablation has evolved as first line therapy for various arrhythmias. This is reflected in the recently updated ESC guidelines for the management of supraventricular tachycardias (SVT). 1 Of note, recommendations for ablation procedures are not limited to a special group of patients. While guidelines have devoted paragraphs to the specialized treatment of patients with congenital heart disease or during pregnancy, no specific recommendations are made for older patients. 1 Atrial fibrillation (AF) and atrial flutter (AFL) are very prevalent in the elderly. For typical AFL in particular, cavotricuspid isthmus ablation (CTI) has evolved as the gold standard since the 1990s in most patients as recurrence rate is high and treatment by CTI characterized by very high success rates. 2,3 Surprisingly, recent evidence found markedly higher complication rates related to CTI than previously reported. 4 Advanced age was found to be a major risk factor so that the authors postulated that patient selection for CTI should be made particularly careful in the elderly in whom outcome studies on CTI are scarce. In contrast to AFL, atrioventricular nodal re-entry tachycardia (AVNRT) manifests/presents most often in young and middle-aged adults. Interestingly, more recent studies showed, that effectiveness of slow-pathway modulation was greater in younger patients compared to patients >50 years. 5 In addition, age seems to be a relevant factor in socioeconomic aspects of AVNRT ablation.
Farkowski et al 6 showed that treatment costs were higher with increased patient age.
While SVT require treatment due to debilitating symptoms, ventricular tachyarrhythmias (VT) in the presence of structural heart disease are life-threatening and/or may result in adequate defibrillator interventions and therefore demand effective therapy. A contemporary meta-analysis underlined that VT ablation reduces ICD therapies in patients with coronary heart disease. 7 This is of importance as Sweeney et al 8 demonstrated an increased mortality in patients receiving ICD shocks. Of note, increasing age was associated with a higher rate of ablation-related complications and therefore implemented in a risk score predicting complications and in-hospital mortality on the basis of over 25.000 patients from a national US database. 9 Besides, Yousuf et al 10 showed a 1-year mortality of 15% after VT ablation and a rate of 7.5% of major adverse events as well as high rates of hospitalization for recurrent arrhythmias or heart failure.
In order to investigate the potential impact of age on efficacy and safety of catheter ablation in SVT and VT in the very elderly we performed a propensity matched analysis of ablation procedures in patients with AVNRT, AFL, and VT.

| PATIENTS AND METHODS
The study was conducted in accordance with the guidelines of the Declaration of Helsinki. In the present study, we analyzed our single center prospective ablation database for the period from august 2002 to December 2018. We included all patients >80 years (n = 387) who underwent catheter ablation for typical AFL, AVNRT, or VT and performed a matching for age and sex at time of ablation, and ensured that procedures were similarly distributed over the period of data acquisition. Three comparison groups for each arrhythmia were formed with age groups from 40-60 years, 60-80 years, and > 80 years of age. The AFL cohort consisted of 753 patients, the AVNRT cohort of 270 patients, and the VT cohort of 168 patients in total. Acute success rates, hospitalization time, and complications were recorded.

| Statistical analysis
Continuous data are reported as mean ± SD, and categorical data are reported as percentages. Statistical analysis was performed using Gra-phPad PRISM 6.0 (San Diego, CA) and the SPSS Statistics, version 20.0 (SPSS, Inc., IL). A P-value <.05 was considered statistically significant.
BMI was comparable in all age groups and arrhythmias and ranged from 25.5 to 28.0 kg/m 2 . Mean time of hospitalization ranged from 2.5 days for young AVNRT patients to 13.5 days for VT patients older than 80 years (see Table 2).
Complication rates were low for AFL and AVNRT patients of all age groups (2%-5%) and higher in VT patients (up to 15%, Table 2), mostly due to early recurrences of arrhythmia which were classified as a complication. There was no death during the ablation procedure.
Two patients died from cardiac arrest in cardiogenic shock the day after the VT ablation (one in age group 40-60 years and one

| DISCUSSION
In this study, we present data on the acute outcome of catheter ablations of typical AFL, AVNRT, and VT in patients >80 years and younger matched comparison groups. The study cohorts consisted of a rather typical population with different histories of structural heart disease and risk factors (see Table 1). As expected, more patients with with AFL and VT had cardiovascular risk factors compared to patients with AVNRT and prevalence of risk factors increased with age.
In AFL we found high acute success rates and fortunately, in our analysis there were fewer complications than reported in a recent registry data analysis by Steinbeck et al. 4 No in-hospital death or major adverse event occurred in our population. However, there was a certain risk for pacemaker implantation after successful ablation. In our data analysis, about 4% to 10% of AFL patients received a cardiac device during the same hospital stay (see Table 2). Similar results could be shown by an observational study from Taiwan. 11  were comparable in all groups. This underlines that CTI ablation is a procedure which can be performed safely and successfully also in older patients with more comorbidities.
Likewise, high success rates of ablation were also found for slow pathway modulation in AVNRT. While recurrence rates of AVNRT in the pediatric and adolescent population appear to be higher and range from 10% 12 up to over 20%, 13 success rates of slow pathway modulation in an adult population are high and well above 95% 14,15 and therefore in line with results from our trial. One may speculate that the ablation approach is performed more cautiously in children to minimize the risk of AV block and to avoid the necessity of pacemaker implantation. Interestingly, ablation seems to be more complicated in young patients than in older ones. AVNRT ablation is probably one of the most successful procedures with a high immediate and long-term success and seems to work equally efficient in older patients. This is in line with recent data from the German Ablation Registry. 16  A dutch working group underlined a different decisive factor for the outcome of VT ablation rather than age-namely the high impact of LV dysfunction in a study of patients with ischemic cardiomyopathy. 18 Nevertheless, higher age was associated with worse outcome in two contemporary risk scores predicting survival and recurrence risk in patients undergoing VT ablation. 19,20 As VT ablation is often not an elective procedure and as the reduction of events has direct impact on patients' outcome and prognosis, there is no alternative option in most cases. Consequently, data on these procedures are also of eminent importance to optimize procedure and success rates.

| CONCLUSION
Our data underline that various catheter ablation of AFL, AVNRT, and VT worked effectively and with low complication rates in this very elderly. Ablation of AFL as well as AVNRT were performed with very high acute success rates of >95% across all age groups although patients' comorbidities were clearly increasing with age. VT ablation also had a good acute success rate of over 80% in patients >60 while success rates were even higher in VT patients <60 years because of age-dependent decrease of idiopathic VT.

| Limitation
This study has several limitations, mostly associated to the retrospective nature of the data. Patients with AF were not included in our analysis as only a limited number of patients over 80 years underwent pulmonary vein isolation at our institution between August 2002 and December 2018. Furthermore, although propensity matching was carefully performed, there is always a risk for a certain selection bias and confounding factors.

ACKNOWLEDGMENTS
Open access funding enabled and organized by Projekt DEAL.