The aim of the current study was to identify demographic or clinical parameters associated with the greatest improvement of HRQOL scores in patients with typical AFL undergoing CTI catheter ablation. Thus we have analysed, for the first time, changes in HRQOL scores before ablation and at one-year follow-up, in terms of several patient sub-groups. Taken as a whole, the population of typical AFL patients that we studied showed a clinically significant improvement in HRQOL one year after undergoing CTI ablation . This result was notable given that the cohort included patients with concomitant AF (43%) and also only a single episode of AFL (44%). After adjusting for age, sex, and other cardiovascular-related parameters, the presence of structural cardiopathy was associated with an improvement in the physical summary component that was 5.8 greater than that of the absence of structural cardiopathy, and the only variable that made a difference that was statistically significant.
The recurrence rate of AFL at one year was 14.6%, a slightly higher level than reported for the majority of other patient series (20% at 1 year, [5, 15] 10% and 6.3% at 16 meses,  4.8% at 21 meses ). One potential partial explanation would be that while a 8–mm- tip catheter was used in 93 patients (97.9%), an irrigated-tip catheter was used in the remaining two patients (2.1%), since it is known that an irrigated tip produces more profound lesions than an 8- mm- tip catheter. However, to date no associated differences in outcome have been demonstrated either acutely [16, 19] or in the longer term  after treatment of CTI-dependent AFL with an irrigated-tip or 8-mm- tip catheter. We therefore conclude that our increased recurrence levels were probably due to factors other than catheter tip, and most likely related to CTI morphology.’
Patients who developed AF during follow-up also showed a significant improvement in HRQOL at one year, except in the dimension of Bodily Pain. Although the group who developed AF during follow-up showed a smaller improvement in HRQOL with respect to the group that did not develop AF, the difference between the two groups did not approach significance in any dimension. Without further analysis, this result contrasts with that of Lee et al  in which a history of AF was the only factor associated with reduced improvement of HRQOL in patients with typical AFL undergoing CTI ablation. However, when we took into account HRQOL at one-year follow-up we also found in our cohort that for most dimensions final scores were significantly better in patients who had not experienced AF during follow-up than in those who had. These data suggest that the differences seen in basal HRQOL (lower in patients with a history of AF) maintained and formed a component of the scores obtained at one year after CTI catheter ablation (smaller improvement in patients experiencing AF during follow-up), thus accounting for the differences seen at one year in most of dimensions approaching or reaching significance. Interpreted thus, our results for HRQOL at one-year follow-up agree more closely with those of Lee et al .
A long-term (27 months) benefit in quality of life in patients undergoing CTI ablation with a history of AF has previously been reported by Anselme . Those results are similar to ours with respect to the 8 dimensions of the SF36 questionnaire, showing a significant improvement in all items except for Bodily Pain. In that study, the improvement was smaller in patients with a history of AF than in those without, but the difference between those two groups was not significant. However, in our cohort, basal scores were significantly lower overall, and especially in patients who had presented AF during follow-up. In Anselme’s cohort, moreover, patients with a history of AF also had higher basal scores than those without, in the majority of items.
Whilst the presence of AF reduced the benefit of CTI ablation in typical AFL patients, the AF group did still show a significant improvement in HRQOL at one year. Moreover, the post-ablation rate of AF appears to be time-dependent with previous figures for incidence of AF during three-year follow-up varying widely, from 15%  to 82% . This suggests that typical-AFL populations differ with respect to incidence of AF, and raises the question of how CTI ablation in AFL can be so effective within a context of such a frequent secondary arrhythmia. Willems et al have found an improvement in quality of life in a cohort of patients undergoing CTI ablation for typical AFL and with an AF incidence of 60% at 2 years .
Undoubtedly, the explanation will ultimately be found to be complex, but the following factors should be considered.
First, AF is better tolerated than AFL. AFL normally presents with a conduction ratio of 2:1 or 3:1, producing heart rates between 150 lpm and 130 lpm. The fact that it is also more difficult to achieve adequate pharmacological control of heart rate in AFL than in AF could explain why symptoms in AFL patients are generally more severe. In a cohort of patients with AF being treated with anti-arrhythmic drugs the presence of AFL during follow-up was associated with a lower rate of return to sinus rhythm due to anti-arrhythmic drugs and with a higher number of hospitalisations than AF recurrence. In the same population there was an 8.5% incidence of AFL having a 1:1 conduction ratio, a highly symptomatic form of AFL . Therefore, AFL appears to be an arrhythmia with a higher impact on HRQOL than that of AF because it is more symptomatic, more refractory to pharmacological treatment – in terms of control of both heart rate and rhythm – and leads to a greater number of hospitalisations than AF.
Second, reduction of number of symptomatic palpitation episodes and AF burden following ablation for AFL. A reduction in the number of AF episodes post-ablation has been widely reported [16, 25, 26]. One explanation could be the elimination of AFL as a trigger of AF events. It has also been suggested that normalisation of the effective refractory period and inverse electrical remodelling in the atrium reduce the frequency of AF . In our patient cohort, the recorded incidence of AF at one-year follow-up was reduced from 43% basal to 25%. This reduction should, however, be interpreted with caution as AF diagnosis was based on symptoms and events registered by Holter monitoring over the one-year period. A more rigorous approach might be to monitor and diagnose oligosymptomatic episodes, and/or to have a longer follow-up period in which to gather data on the development of AF in post-ablation patients. Ellis et al  have reported an incidence in post-ablation AF of up to 82% at 3 years and 3 months. Similarly, Chinitz et al  have reported an occurence of AF of 50% at a follow-up of 2.5 years in a cohort of patients with typical isolated AFL undergoing CTI ablation. More recently, a recurrence of AF has been reported in 90% of patients receiving hybrid treatment (ablation for AFL + antiarrhythmic treatment for AF) at 5-year follow-up . It seems likely that AFL and AF constitute different electrocardiolographic expressions of a common atrial disease and that CTI ablation does not eliminate the risk of subsequent AF .
Patients who had experienced only a single AFL event showed a similar improvement in HRQOL to those with recurrent AFL, although the latter group showed a tendency towards increased improvement in the physical dimensions. The benefit of CTI ablation over amiodarone treatment in patients who had experienced only a single episode of AFL had been described previously . This datum is important because patients with a single AFL event are usually treated more conservatively, on the assumption that they will benefit less from ablation therapy and recurrence will be delayed.
No differences in HRQOL were seen with respect to ventricular cycle length of ≤500 ms or >500 ms. Neither were any differences seen in relation to presence or absence of tachycardiomyopathy, which makes sense in the absence of ventricular-cycle-related differences. Our observation that patients who had presented with persistent AFL and a ventricular cycle of ≤500 ms (heart rate ≥120 lpm) showed a significant improvement in four HRQOL dimensions (compared with patients with persistent AFL and ventricular cycle of >500 ms), however, demonstrates that when AFL is persistent ventricular cycle length can affect HRQOL.
Analysing differences in HRQOL with respect to combinations of characteristics (i. persistent AFL + ventricular cycle ≤500 ms; ii. structural heart disease + ventricular cycle ≤500 ms + recurrent AFL) revealed greater differences, particularly in relation to physical dimensions. While these differences are potentially important, the fact that the number of cases in each group was low would have reduced the capacity for statistical tests to show significant differences in other dimensions. Indeed, it seems possible that the combination of structural heart disease, ventricular cycle ≤500 ms and recurrent AFL will be a better predictor of improvement in HRQOL scores following CTI ablation than will each of those factors individually.
The majority of studies using the SF-36 questionnaire have observed ceiling effect (more than 15% of patients had the maximum value for a dimension) on the dimensions of Physical Role and Emotional Role. In our cohort we have observed a ceiling effect in the basal dimensions Physical Role, Bodily Pain, Social Function and Emotional Role and at one-year follow-up in the same dimensions and additionally in Physical Function. Therefore the presence of a ceiling effect might have caused an underestimation of the size of change in HRQOL over the one-year follow-up period. In three dimensions (General Health, Vitality and Mental Health) a ceiling effect was not seen, and thus the values obtained in those dimensions were particularly valuable in assessing improvement in HRQOL at one year.
Other limitations were the following. a.) Because of the advanced age and comorbidities in our population, we lost some patients during follow-up due to death or incapacity to complete the second questionnaire at one year. b.) The low number of patients we recruited. c.) The observational approach we used. Because we did not randomly assign patients to treatment and placebo groups, we could not make inferences about cause and effect. It will therefore be important to carry out controlled studies in the future. d.) We used only a general type of questionnaire to evaluate quality of life, and not one designed for the context of cardiac arrhythmias [32, 33] and e) A 7-day Holter recording is a rather limited method to assess AF burden. An implantable loop recorder would have provide more accuracy .