Curative catheter ablation in atrial fibrillation and typical atrial flutter: systematic review and economic evaluation

Krittayaphong et al., 200357

RCT (rated ‘satisfactory’) of RFCA versus amiodarone in patients with paroxysmal or persistent AF (70% paroxysmal, mean duration 4.6 years). All patients were refractory to at least one AAD but had never received amiodarone.

A total of 15 patients randomised to RFCA underwent circumferential PV ablation, with an additional line connecting the circular line to the mitral annulus. In addition, CTI ablation was performed, plus a horizontal line at the level of the mid right atrium. RFCA could not be performed in one patient because of failure of trans-septal puncture. Electrical cardioversion was performed in two patients still in AF following ablation, and all patients received amiodarone for 3 months post procedure.

A total of 15 patients were randomised to long-term amiodarone treatment. Cardioversion was performed in patients with persistent AF. When serious side effects occurred, amiodarone was discontinued in favour of class Ia or Ic agents.

At 12 months post randomisation, 11/15 patients in the RFCA group were free of AF compared with 6/15 patients in the amiodarone group.

Frequency of symptoms decreased from 42.8 (SD 22.6) attacks per month at baseline to 0.9 (2.8) attacks per month at 12 months in the RFCA group. There was a non-significant decrease over the same period in the amiodarone group. Differences between groups were not statistically significant.

SF-36 (Short Form 36) general health and physical functioning scores improved significantly at 12 months compared with baseline in the RFCA group but not in the amiodarone group. Between-group differences for general health score favoured ablation.

One stroke and one groin haematoma were associated with the ablation procedure. In the amiodarone group, seven patients experienced at least one adverse effect attributed to amiodarone: six had gastrointestinal (GI) adverse effects (mainly nausea), two each had corneal microdeposit, hypothyroidism and abnormal liver function test, and one had hyperthyroidism and sinus node dysfunction. Three patients in the RFCA group had amiodarone-attributed adverse effects during the 3-month amiodarone treatment period: two had GI adverse effects and one had sinus node dysfunction.

Pappone et al., 200361

Large non-randomised controlled study (rated ‘poor’) comparing RFCA with long-term AAD treatment in patients with symptomatic AF (70% paroxysmal, mean duration 4.6 years).

A total of 589 consecutive patients underwent circumferential PV ablation. In total, 19.5% of patients were given previously ineffective AADs if they had in-hospital AF episodes or required cardioversion. AADs were discontinued 3 months after ablation in the absence of recurrences.

A control group of 582 patients received AAD therapy (33% amiodarone, 17% propafenone, 15% flecainide, 13% sotalol, 9% quinidine, 6% disopyramide and 7% combined AAD therapy).

At 12 months, an estimated 84% of patients in the RFCA group and 61% of patients in the AAD group were free of AF. At 2 years these values were 79% and 47%, respectively, and at 3 years they were 78% and 37% respectively. There were a total of 120 AF recurrences in the ablation group and 340 in the AAD group.

After a median follow-up of 29.6 months there were more deaths overall in the AAD group (83 versus 38). The difference remained apparent for deaths due to cardiovascular causes (59 versus 18).

Four patients in the RFCA group (0.7%) required pericardiocentesis for cardiac tamponade. A total of 46 ablated patients (8%) and 98 AAD-treated patients (19%) were managed for a total of 54 and 117 adverse events respectively. Sinus rhythm was associated with significantly lower mortality rates [HR 0.24 (95% CI 0.16–0.37)] and adverse event rates [HR 0.22 (95% CI 0.15–0.31)].

SF-36 physical and mental composite scores from a subgroup of patients improved significantly from baseline to 1 year in ablated patients (n = 109) but not in medically treated patients (n = 102).

Wazni et al., 200560

RCT (rated ‘satisfactory’) comparing RFCA with long-term AAD treatment as first-line therapy in patients with symptomatic AF (96% paroxysmal, mean duration 5 months).

A total of 33 patients were randomised to receive segmental PV ablation. AADs were not given as part of the treatment protocol, but beta-blocker therapy was continued in 43% of patients. There were no repeat ablation procedures within the first year of follow-up. One patient was lost to follow-up.

A total of 37 patients were randomised to receive the maximum tolerable dose of AAD chosen by the treating physician (typically flecainide, propafenone or sotalol). Amiodarone was given only when two other drugs had previously failed. Two patients were lost to follow-up.

At 12 months, 87.5% of patients in the RFCA group were in sinus rhythm after a single procedure. For the same time period in the AAD group, 37% of patients were in sinus rhythm at follow-up. Sensitivity analyses accounting for patients lost to follow-up did not substantially change these results. There were no crossovers during the 12-month follow-up.

There was a significantly greater improvement in the ablation group than in the AAD group at 6 months on five of the eight subscales of the SF-36 questionnaire (general health, physical functioning, role physical, bodily pain, social functioning).

Two patients in the RFCA group developed mild to moderate PV stenosis. Three patients in the AAD group developed bradycardia. There were significantly more patients hospitalised in the AAD treatment group than in the RFCA group (54% versus 9%; p < 0.001).

Oral et al., 200656

RCT (rated ‘good’) comparing the effect of 3 months’ treatment with amiodarone and cardioversion with or without the addition of RFCA on long-term maintenance of sinus rhythm in patients with persistent AF (mean duration approximately 4.5 years).

A total of 77 patients were randomised to receive circumferential PV ablation followed by 3 months of amiodarone treatment, plus cardioversion if AF recurred within those 3 months. In total, 25 of these patients underwent a repeat ablation (20 for AF, five for flutter).

A total of 69 patients were randomised to receive amiodarone treatment for 3 months, plus cardioversion if AF recurred. In total, 53 of these patients (77%) ultimately elected to cross over into the RFCA group because of recurrent AF.

According to intention to treat analysis at 12 months, 74% of patients in the RFCA group were in sinus rhythm without AAD therapy, compared with 58% in the amiodarone group (p = 0.05). If the need for any treatment (ablation or drugs) after 3 months was considered a treatment failure, 74% of patients in the RFCA group were in sinus rhythm without AAD therapy, compared with 4% of those in the amiodarone group.

Patients in sinus rhythm had greater improvements in symptom severity score than those with recurrent AF or atrial flutter (p = 0.002).

In the RFCA group, five patients required further ablation for atypical atrial flutter. No other complications related to ablation or drug therapy were reported.

Pappone et al., 2006 (APAF)58

RCT (rated ‘satisfactory’) comparing RFCA with long-term AAD treatment in patients with drug-refractory paroxysmal AF (mean duration 6 years).

A total of 99 patients were randomised to receive circumferential PV ablation. Patients received AADs for 6 weeks post procedure, after which drug treatment was stopped. Nine of these patients underwent a repeat ablation (six for recurrent AF, of which five were successful, and three for atrial tachycardia, all successful). Five patients went on to have AAD therapy to maintain sinus rhythm.

A total of 99 patients were randomised to receive AADs (amiodarone, flecainide or sotalol, either singly or in combination) at the maximum tolerable doses. In total, 42 of these patients (42%) elected to cross over into the RFCA group after two failed AADs over 3 months. Of the 42 crossovers to RFCA, 36 were successful in terms of subsequent freedom from arrhythmia.

At 12 months, 86% of patients in the RFCA group were in sinus rhythm after a single procedure and without the need for AAD therapy. Including repeat ablations and patients on adjunctive AADs, 93% of patients in the RFCA group were in sinus rhythm at 12 months. For the same time period in the AAD group, 24% of patients were in sinus rhythm after the first tested AAD and 35% were in sinus rhythm after combination therapy.

No serious complications occurred in the RFCA group, although three patients developed post-ablation atrial tachycardia requiring further ablation. Significant adverse events leading to permanent drug withdrawal occurred in 23 patients. There were significantly more hospital admissions in the AAD treatment group than in the RFCA group (167 versus 24; p < 0.001).

Stabile et al., 200659

RCT (rated ‘satisfactory’) comparing long-term AAD treatment with or without RFCA in patients with paroxysmal or persistent AF for which AAD therapy had already failed (66% paroxysmal, mean duration 6.1 years).

A total of 68 patients were randomised to receive an AAD chosen by the physician (preferentially amiodarone, or class Ic if amiodarone could not be tolerated) and also undergo circumferential PV ablation, with an additional line connecting the left inferior PV to the mitral annulus. In addition, CTI ablation was performed when considered appropriate. There were no repeat ablations. Two patients were lost to follow-up.

A total of 69 patients were randomised to receive an AAD chosen by the physician without ablation. The AAD was only changed if the patient experienced a recurrence of arrhythmia. In total, 36 patients (52%) with AF relapses eventually crossed over to receive ablation while continuing the previously ineffective AAD regime. Two patients were lost to follow-up.

According to the authors’ analysis, at 12 months 56% of patients in the AAD plus ablation group were in sinus rhythm after a single procedure. For the same time period in the AAD therapy alone group, 9% of patients remained in sinus rhythm during follow-up.

Three major complications were related to the ablation procedure: one stroke (patient died 9 months later of a brain haemorrhage), one transient phrenic paralysis, and one pericardial effusion requiring pericardiocentesis. In the control group there was one transient ischaemic attack and two deaths (one sudden death, one cancer). There was no significant difference in the median number of hospitalisations needed between groups (one versus two; p = 0.34).

Jais et al., 200663

An unpublished RCT (rated ‘poor’, mainly because of the limited reporting of trial details) comparing RFCA with long-term AAD treatment in patients with drug-refractory paroxysmal AF.

A total of 53 patients were randomised to receive segmental PV ablation. A mean of 1.8 PV ablation procedures were conducted. In addition, 64% of patients were reported to have had CTI ablation, 30% had mitral isthmus ablation and 17% had ablation of the roof line.

A total of 59 patients were randomised to receive AADs (including beta-blockers and classes I, III and IV AADs, alone or in combination; amiodarone was used in 22 patients). In total, 37 patients (63%) elected to cross over into the RFCA group.

The primary end point was absence of AF for 3 minutes or more (either symptomatic or documented). At the 12-month follow-up, 75% of patients in the RFCA group and 6% of patients in the AAD group were AF free. RFCA was also favoured in terms of duration of recurrent AF episodes [8 minutes (SD 55) versus 150 minutes (SD 350)].

One tamponade and one PV stenosis (> 50%) occurred in the RFCA group. There was one case of amiodarone-induced hyperthyroidism.

Lakkireddy et al., 200662

An unpublished, non-randomised study (rated ‘poor’) comparing RFCA in older patients with AF (aged 60–80 years) with AV node ablation or direct current cardioversion (DCC) in age-matched control subjects.

The authors of this study concluded that PV ablation had significant mortality and morbidity benefits against the other therapeutic strategies; however, because of the inadequate and inconsistent data currently available for this study, it will not be discussed further here. For the sake of completeness, the available details are presented in Appendix 3.1.

Freedom from arrhythmia at 12 months

Figure 2 shows the RR and associated 95% confidence intervals for freedom from arrhythmia at 12 months for the six RCTs reporting sufficient data to calculate this outcome.

FIGURE 2.

Freedom from arrhythmia at 12 months: all atrial fibrillation randomised controlled trials.

The forest plots show that all of the included controlled studies favour ablation over AAD-based management, but the extremely high degree of statistical heterogeneity observed between the studies (I² 85.8–91.7%) indicates clearly that a pooled value for the group of studies as a whole is not considered appropriate. The heterogeneity can be partly explained by variation in patient populations and interventions: one study (Oral et al. 56) was limited to patients with persistent AF; one (Stabile et al. 59) evaluated combined RFCA/AAD treatment versus AADs alone (see Figure 2). The remaining randomised evidence included four RCTs57,58,60,63 comparing ablation with long-term AAD treatment (predominantly amiodarone, sotalol and class Ic agents, when reported) for the maintenance of sinus rhythm. However, when these four trials were pooled, highly significant statistical heterogeneity (I² 74%) remained (Appendix 4). One RCT63 intensively followed up arrhythmia in a way that differs from typical clinical practice and from follow-up protocols in other trials making the same comparison. This trial reported an unusually high rate of failures associated with AAD use, resulting in a very different relative effect from other RCTs. Removal of this study from the pooled comparison removed all statistical heterogeneity (I² 0%; see Figure 3).

FIGURE 3.

Freedom from arrhythmia at 12 months: RFCA vs long-term AAD therapy in paroxysmal atrial fibrillation.

Within the three remaining RCTs comparing ablation with long-term AAD treatment that could be pooled in a meta-analysis,57,58,60 97% of the patients (Figure 3) were diagnosed with paroxysmal AF (ten patients in the Krittayaphong et al. 57 study had chronic AF) and the majority of patients (75%, 87%, 93%, respectively) were free of structural heart disease in the three trials. Only nine repeat ablation procedures were reported (all in the APAF study). The pooled per protocol results showed that significantly more patients undergoing RFCA were free of arrhythmia (without AADs) at 12 months’ follow-up than those receiving AAD maintenance therapy alone [RR 2.36 (95% CI 1.89–2.95)]. There were only four patients lost to follow-up across the three studies; assuming that data from these patients would be unfavourable to ablation had little influence on the pooled estimate [RR 2.28 (95% CI 1.83–2.84)]. The pooled intention to treat analysis ignoring crossovers indicated a smaller but statistically significant effect in favour of ablation [RR 1.54 (95% CI 1.36–1.76)]. As the study of Wazni et al. 60 was limited to patients undergoing first-line therapy, the mean duration of AF was considerably shorter in this trial (5 months) than in the other two trials (4.6 years and 6 years). However, sensitivity analyses involving the removal of the Krittayaphong and Wazni studies had little influence on the overall estimate of effect (see Appendix 4).

One RCT by Oral et al. 56 compared the effect of adding ablation to a short-term (3 month) amiodarone/cardioversion treatment strategy on subsequently achieving sinus rhythm without the need for AADs in patients with persistent AF. The majority of patients in the amiodarone/cardioversion treatment arm (77%) crossed over into the ablation arm before the 12-month follow-up. Forest plots showing the effect on sinus rhythm without AADs at 12 months are given in Appendix 4. Disregarding the very large crossover and analysing the data by original treatment allocation gives a relatively small effect, of borderline statistical significance, in favour of ablation [RR 1.28 (95% CI 1.00–1.62)]. If the need for additional treatment for recurrent AF after the initial 3 months was considered a failure of short-term therapy, the effect was much larger and significantly favoured the group that had received ablation [RR 17.03 (95% CI: 5.59–51.90)].

Stabile et al. 59 evaluated the effect of adding ablation to a long-term AAD maintenance strategy in patients for whom previous AAD therapy had already failed. As in the Oral et al. trial, a high proportion of patients (52%) crossed over from the AADs alone group to receive ablation after recurrence of AF. For freedom from arrhythmia at 12 months, the intention to treat analysis reported by the study authors showed a large, statistically significant effect in favour of ablation [RR 6.43 (95% CI 2.91–14.21)]. Assuming that data from patients lost to follow-up would have been unfavourable to ablation gave a slightly smaller pooled estimate [RR 6.09 (95% CI 2.74–13.51)], and per protocol analysis increased the pooled estimate [RR 9.14 (95% CI 3.44–24.23); see Appendix 4). It is not entirely clear why the relative effect for this comparison (RFCA plus AADs versus AADs alone) should be noticeably larger than that typically seen in the RCTs comparing RFCA against AADs alone, although the very low success rate seen in the AADs alone arm of Stabile et al. may be attributable to the intensive follow-up (patients had daily ECG recordings for 3 months) undertaken in this study.

The preliminary findings of a trial by Jais et al. were reported in a conference abstract. 63 This study appeared to have intensive follow-up of arrhythmia, similar to the trial by Stabile et al. , and similarly reported a large, statistically significant effect in favour of RFCA over ongoing AAD therapy [RR 11.13 (95% CI 4.27–29.03)].

For reference, Table 9 shows these results for the meta-analysis and trials that could not be pooled when calculated as odds ratios and related confidence intervals. However, our discussion of the findings will be limited to the RR estimates as relative probabilities are more easily interpreted than relative odds.

Freedom from arrhythmia at other follow-up points

Of the studies reporting adequate freedom from arrhythmia data at other follow-up points, five were randomised and one was non-randomised. By far the largest of the studies was the non-randomised comparison of ablation against AAD therapy conducted by Pappone et al. 61 A lack of randomisation means that the two treatment arms are more likely to be unbalanced at baseline in terms of key participant characteristics. This appears to be true in the case of this particular study; at baseline the ablation group had longer duration of AF, a greater number of failed AADs and more frequent hospitalisation than the medical treatment group. However, survival analysis (mean follow-up of 900 days) indicated a significantly lower overall rate of AF recurrence in the ablation group [HR 0.30 (95% CI 0.24–0.37; p < 0.001)]. The difference in the estimated proportion of patients free from AF between the groups increased over time: 0.23 (95% CI 0.18–0.28) at 1 year, 0.32 (95% CI 0.28–0.38) at 2 years, and 0.41 (95% CI 0.33–0.49) at 3 years.

None of the randomised studies explicitly reported the number of patients free from arrhythmia before 12 months’ follow-up. When Kaplan–Meier analyses of AF recurrence were presented, these broadly indicated a steady rate of recurrences over time associated with AAD treatment, whereas recurrences associated with RFCA treatment tended to occur in the first 2–3 months post procedure before stabilising.

Complications, adverse events and mortality

Table 10 summarises the complications, adverse events and deaths reported in the controlled AF studies. A total of 1860 patients (932 ablation, 928 control) were included in these seven studies. Adverse events and complications were rarely reported; where data on major events/complications were not reported, we assumed that such an event/complication did not occur during the study. Some events were specific to ablation (tamponade, PV stenosis, groin haematoma) whereas others were specific to AADs (corneal microdeposit, thyroid dysfunction, pro-arrhythmia, sexual impairment).

A total of 30 strokes were reported. Eight were in patients receiving RFCA, and two of these were related to the ablation procedure itself. Seven patients undergoing ablation had cardiac tamponade or pericardial effusion and three had PV stenosis (one mild, two moderate).

Deaths were reported in three of the studies. 56,59,61 Brain haemorrhage caused death in one patient, 9 months after having had a stroke during an ablation procedure. 59 One patient who had received RFCA died from pneumonia. 56 The majority of deaths (n = 121) were reported in the single largest study61 with the longest follow-up (median 29.6 months).

Quality of life

Three controlled studies evaluated participants’ QoL using the SF-36 health survey. 57,60,61 This is made up of 36 questions that contribute to eight subscales: four within physical health (physical functioning, role physical, bodily pain and general health) and four within mental health (vitality, social functioning, role emotional and mental health).

Wazni et al. 60 and Krittayaphong et al. 57 found a significantly greater improvement on the general health subscale in patients randomised to RFCA compared with control subjects at 6 and 12 months respectively. In addition, Wazni et al. 60 reported that RFCA patients attained significantly higher scores on the physical functioning, role physical and bodily pain subscales.

In terms of within-group changes from baseline for RFCA treatment, Krittayaphong et al. reported a significant increase in both physical functioning and general health scores. Pappone et al. 61 reported a significant improvement in scores from baseline but did not state on which subscales.

Further QoL data, in relation to utility estimates, were considered in the development of the decision model (see Chapter 4, Decision model).

Populations

Details of the populations in the case series are presented in Table 11. The populations were relatively uniform in terms of average age and proportion of men, but other characteristics (proportion with paroxysmal AF and structural heart disease, duration of symptoms and number of unsuccessful AADs) varied substantially.

Of the 53 case series, 26 were from centres in Europe, 15 were from the USA, eight were from Asia and four reported experience from centres in more than one country. Several expert centres contributed multiple case series, including the Cleveland Clinic Foundation (ten publications), the Hôpital Haut-Leveque, France (eight publications) and the San Raffaele Hospital, Milan, Italy (four publications).

The number of treated patients ranged from 100 (the minimum required to meet inclusion criteria) to 1125. In total, the 53 case series included 11,908 patients (mean 224.7, median 158) but because of multiple publications from some centres it is likely that this overestimates the total number of unique patients. Mean age was in the 50s in 89% of the series. Age range was reported in 18 case series, with most series including a wide range of ages from young (in the teens or 20s) to elderly adults (in their 70s or 80s). Almost all case series (51/53) were made up of a majority of male participants; two case series did not report gender balance.

Duration of symptoms was reported in 40 case series, normally as the mean. The reported measure was 5 years or less in nine case series, 5–6 years in nine, 6–7 years in 13, 7–8 years in six and over 8 years in two. Based on reports in 36 case series, the mean number of unsuccessful AADs ranged from 1.5 to 5. The proportion of patients with paroxysmal AF (reported in 51 case series) ranged from 9% to 100%; 13 case series included only patients with paroxysmal AF. Presence of structural heart disease was reported in 40 case series and ranged from 0% to 65%; only four series had a majority of patients with structural heart disease. Hypertension was not included as structural disease when these were reported separately.

Interventions

Intervention characteristics varied widely between case series, the most fundamental division being between segmental and circumferential techniques for PV ablation. Of the 53 studies treated as case series, 24 used segmental PV isolation uniformly in all patients, eight used circumferential PV isolation, three used a combination of approaches, two used other PV isolation techniques and two used other approaches. The catheter sizes used, mapping technique, definition of end point, number of PVs ablated and other details varied across the case series. This variation of intervention characteristics needs to be interpreted in light of our decision to treat all techniques of RFCA as a single intervention. Of 14 studies (true case series and trials) that compared ablation techniques, seven compared different techniques within one basic approach (e.g. different catheter types or mapping systems) and seven compared results of two or more different approaches. Appendix 3.3 summarises details of the interventions in the included case series (including RCTs and CCTs that compared ablation techniques).

Relevant details regarding how AADs were used with RFCA were not well reported. For example, in 30 of the 53 case series it was unclear whether or not AADs were withdrawn before RFCA. Where it was reported most series did withdraw AADs. Similarly, details of AADs after the procedure were frequently not reported: in 26 out of 53 series it was unclear; in 18 out of 27 series patients did receive AADs. A total of 20 case series either did not use anticoagulants after the procedure or did not report details. When the time on anticoagulants was reported (32 series), it ranged from 1 to 6 months.

The UK series by Bourke and colleagues66 was fairly typical in terms of average age of participants (52 years) and percentage of male patients (81%) but it had a relatively high proportion of patients with chronic AF (64%) while excluding patients with structural heart disease. Mean duration of symptoms was 53 months (range 6–180) and patients had been treated unsuccessfully with an average of three AADs. The intervention was classified as segmental PV isolation using a 4-mm tip ablation catheter. The mean number of PVs isolated (2.41 ± 0.79) was relatively low in case series in which this outcome was reported. The authors adapted their technique later in the series by drawing additional ablation lines if necessary. Patients were returned to AAD treatment after RFCA and those with persistent AF were treated with anticoagulants for a minimum of 3 months after the procedure.

Freedom from arrhythmia at 12 months

Eleven of the 53 case series reported data for freedom from arrhythmia (AF) at 12 months. The rates reported ranged from 28% to 85.3%; 9/11 of the case series reported rates of over 60% (Table 12). The weighted pooled estimate of freedom from arrhythmia at 12 months was 76% (95% CI 74–77%) (Table 13).

Two case series64,90 reported on occurrence of arrhythmias other than AF, reporting rates of atrial tachycardia of 12% and 7% respectively.

Whether freedom from AF was achieved as a result of a single ablation procedure or whether it required repeat ablations was reported for 10 of the 11 case series reporting this outcome; for two the rates were achieved with repeat ablation and for seven they were achieved without. The rates achieved with repeat ablation were not necessarily higher than the rates achieved without repeat ablation. One case series reported on whether freedom from arrhythmia was achieved with or without AADs; 71/100 patients were free of AF without AADs, and a further 11/100 patients achieved sinus rhythm with additional AAD treatment.

Two series68,81 reported on freedom from both paroxysmal and chronic AF. In both series, freedom from paroxysmal AF (73.2% and 82.5% respectively) was higher than freedom from chronic AF (51% and 70.3% respectively).

Freedom from arrhythmia at mean follow-up

A total of 26 case series reported on freedom from arrhythmia (freedom AF) at a mean follow-up point (Appendix 3.7). Mean follow-up ranged from 5.5 to 30 months. Freedom from arrhythmia as reported in publications ranged from 52% to 98%, with freedom from arrhythmia above 70% achieved in 19 case series. The series with the longest mean follow-up (30 months) reported 66.2% freedom from AF. 99,116 The results were comparable with those reported in the survey by Cappato and colleagues25 (Table 14).

Whether freedom from AF at mean follow-up was achieved as a result of a single ablation procedure or whether it required repeat ablations was reported for only 10 of the 26 case series. Freedom from AF was based on numbers of patients undergoing a single ablation procedure in just two case series. 75,82 In eight other case series, substantial numbers of patients required a repeat procedure to achieve freedom from AF.

Only two case series that reported overall freedom from AF also explicitly reported the occurrence of other arrhythmias, essentially atrial flutter. Treating these arrhythmias as treatment failures, the rate of freedom from arrhythmia is reduced from 316/377 (83.8%) to 311/377 (82.5%) in one case series82 and from 911/1125 (81%) to 617/1125 (54.8%) in the other. 106 A subgroup of patients in this latter study had undergone previous cardiac surgery and had a higher incidence of atrial flutter during follow-up. Three case series that did not report overall freedom from AF reported the occurrence of atrial flutter at highly variable rates of 28/112 (25%),67 28/150 (18.7%)110 and 2/176 (1.1%). 83

Fifteen case series reported on whether AF-free patients were receiving AAD treatment. In all except one series the majority of patients did not receive AADs and in five series all patients who were AF free were also free of AAD treatment.

Freedom from paroxysmal and chronic AF was reported separately in eight case series. Freedom from paroxysmal AF ranged from 75% to 87.4% and was higher than freedom from chronic AF (range 57.6%–80.6%) in all but one series.

Quality of life

Only three studies treated as case series reported on QoL in patients with AF before and after RFCA. Hsu and colleagues101 administered the SF-36 questionnaire at baseline and at 1 and 12 months after ablation. SF-36 summary scores (physical and mental) increased significantly over time. In the series by Cha and colleagues,80 the SF-36 total score increased significantly from 63 ± 18 before ablation to 79 ± 17 3 months after ablation (p < 0.0001). In the case series by Chen and colleagues,82 193 patients (out of 377 treated patients) completed the SF-36 questionnaire before and after ablation and showed significant improvements on all subscales. Purerfellner and colleagues92 reported that standardised questionnaires were used to evaluate QoL but no results were presented.

It should be noted that, because the studies mentioned here present only SF-36 physical and mental component scores or an overall summary score, a separate review focused on obtaining utility-based QoL evidence was undertaken (see Appendix 7.4).

Complications and mortality

The main complications and mortality rates reported in the case series of RFCA in AF are summarised in Table 15. When case series specifically stated that there were no complications, events are inserted as zeros in the tables. The figures in this table also assume no duplication of patients across the included series and should therefore be treated cautiously. For comparison, the findings of Cappato and colleagues25 in their survey of procedures performed between 1995 and 2002 are included. Detailed results by time period of reporting are presented in Appendices 3.5–3.8.

Natale et al., 200029

RCT comparing RFCA with long-term drug treatment in patients with at least two episodes of symptomatic atrial flutter in the preceding 2 months.

A total of 31 patients were randomised to receive CTI ablation. All rate-control drugs were discontinued following ablation. After 12 months, two of these patients underwent a repeat ablation because of recurrence of atrial flutter.

A total of 30 patients were randomised to receive drug therapy. These patients tried a mean of 3.4 drugs, with 11 remaining on AADs at follow-up (eight on amiodarone, one on propafenone plus atenolol, and two on procainamide plus digoxin). In total, 16 patients received rate-control drugs after AADs failed to maintain sinus rhythm. Two patients crossed over into the ablation group before 12 months, and one required AV node ablation and pacing.

After a mean follow-up of 22 months, 25 (80%) patients in the RFCA group were in sinus rhythm without the need for AADs. For the same time period in the AAD group, 11 (36%) patients remained in sinus rhythm. Atrial flutter was seen in two patients (6%) in the RFCA group compared with 28 (93%) in the drug therapy group. AF was seen in nine patients (29%) in the RFCA group compared with 18 (60%) in the drug therapy group.

One patient experienced chest discomfort and another had a haematoma in the RFCA group. There were significantly more hospital admissions for severely symptomatic arrhythmia in the drug treatment group than in the RFCA group (19 versus 7; p < 0.01).

QoL and symptom scores significantly improved from baseline for all measures in the RFCA group at 6 and 12 months. Significant improvement over time in the drug therapy group was seen only for palpitations.

Da Costa et al., 2006121

RCT comparing RFCA with cardioversion plus long-term amiodarone treatment in older patients (mean age 78 years) after their first episode of symptomatic atrial flutter.

A total of 52 patients were randomised to receive CTI ablation. Seven patients went on to have AAD therapy following occurrence of AF (six patients) or ventricular tachycardia (one patient).

A total of 52 patients were randomised to receive cardioversion (after attempting intracardiac stimulation) followed by amiodarone treatment. One patient was excluded after being diagnosed with a left reentrant atrial tachycardia.

After a mean follow-up of 13 months (SD 6 months), 96% of patients randomised to the RFCA group were free from recurrence of atrial flutter. Of patients randomised to cardioversion plus amiodarone, 71% of patients were free from recurrence of atrial flutter. All patients with recurrent atrial flutter went on to be successfully treated with RFCA, with the exception of one patient who refused the procedure. For the same time period, the occurrence of significant symptomatic AF beyond 10 minutes did not significantly differ between the RFCA and the amiodarone groups (25% versus 18%, p = 0.3).

Five patients in the amiodarone group had adverse events requiring discontinuation. No procedural-related complications occurred in the RFCA group. Six patients in the RFCA group died, compared with eight in the amiodarone group.

Freedom from arrhythmia at 12 months

Neither of the atrial flutter RCTs reported freedom from arrhythmia at 12 months.

Freedom from arrhythmia at mean follow-up

The two RCTs evaluating CTI ablation for atrial flutter29,121 were conducted in clinically distinct populations and so were not combined statistically.

The trial by Natale et al. 29 found a statistically significant benefit favouring ablation in terms of freedom from arrhythmia without the need for AADs at a mean follow-up of 22 months [RR 2.2 (95% CI 1.33–3.63); see Figure 4]. Broken down by type of arrhythmia, this study suggested a very large statistically significant effect favouring ablation in terms of freedom from atrial flutter [RR 14.03 (95% CI 3.67–53.7)] and a smaller, but also significant, effect in terms of freedom from occurrence of AF during follow-up [RR 1.77 (95% CI 1.08–2.90)] (Figures 5 and 6).

FIGURE 4.

RFCA versus long-term AADs – freedom from arrhythmia. Refers to number of patients in sinus rhythm at follow-up, as reported by study authors.

FIGURE 5.

RFCA versus long-term AADs – freedom from atrial flutter. Refers to number of patients without occurrence of flutter during follow-up period.

FIGURE 6.

RFCA versus long-term AADs – freedom from atrial fibrillation. Refers to number of patients without occurrence of atrial fibrillation during follow-up period.

Da Costa et al. 121 reported a more modest effect favouring ablation in terms of freedom from atrial flutter at follow-up in older patients (mean age 78 years) after their first episode of flutter [ITT 1.35 (95% CI 1.13–1.62); per protocol 1.36 (95% CI 1.13–1.64); see Figure 7]. The groups did not significantly differ in terms of freedom from occurrence of significant AF [intention to treat RR 1.44 (95% CI 0.68–3.08); see Figure 8].

FIGURE 7.

RFCA versus intracardiac stimulation/amiodarone: freedom from atrial flutter. ‘Worst case’ analysis assumes that all six patients in the RFCA group who died had recurrence of flutter; per protocol analysis assumes that these patients did not have recurrence of flutter.

FIGURE 8.

RFCA versus intracardiac stimulation/amiodarone: freedom from atrial fibrillation.

The difference between the two RCTs in terms of treatment effects on freedom from atrial flutter results from a lower rate of recurrence in the AAD arm in the Da Costa et al. study and reflects the fact that this study recruited patients with a first episode of atrial flutter rather than the more drug-refractory patients recruited by Natale et al.

For reference, Table 19 shows the outcomes from the two RCTs when calculated as relative risks and odds ratios and related confidence intervals.

Complications, adverse events and mortality

Table 20 summarises the complications, adverse events and deaths reported in the two atrial flutter RCTs.

These studies included a total of 164 patients and reported a total of seven adverse events or complications. Natale et al. 29 reported one patient with groin haematoma and one with chest discomfort among 31 patients receiving ablation. Da Costa et al. 121 reported two occurrences of hypothyroidism, one of hyperthyroidism and two of symptomatic sick sinus syndrome among 51 patients receiving long-term amiodarone treatment.

The Da Costa et al. study reported six deaths in the ablation group (11.5%; one refractory heart failure, one sudden death, four non-cardiovascular causes) and eight deaths in the amiodarone group (15.7%; one pulmonary embolism, one death during coronary artery bypass graft surgery, four non-cardiovascular causes, two unknown).

Quality of life

In the study by Natale et al. ,29 QoL and symptom scores significantly improved from baseline for all measures in the RFCA group at 6 and 12 months. Significant improvement over time in the drug therapy group was seen only for palpitations.

Populations

Twenty-three case series with a total of 4238 participants were included (Table 21). The number of treated patients in each case series ranged from 100 to 417. Seven case series originated from the USA, 15 from Western Europe (mainly France, Germany and Italy) and one from Taiwan. Although obvious duplicate reports have been excluded, the case series included multiple publications from centres in France122,123,128,129 and Germany140,141 and so there may be some overlap of populations.

Of the 23 case series, 19 had an average patient age of 60 years or more. With one exception,136 all the case series had a majority of male patients. Duration of symptoms was reported in ten case series and ranged from 13 months to 4.1 years; this was shorter than for most of the AF case series. Patient populations were classified as drug refractory in nine of the 23 atrial flutter case series, mixed (drug refractory and first line) in nine and unclear in five. The number of unsuccessful AADs was reported for five of the case series with drug-refractory patient populations119,133,136,140,141 and ranged from 1.7 to 2.2. Prevalence of structural heart disease (excluding hypertension where possible) was reported for 18 case series and ranged from 22% to 72%.

Interventions

All of the case series used CTI ablation. A variety of different catheter types and mapping techniques were used but only three studies treated as case series were trials of different ablation techniques. 123,124,141 Most case series used bidirectional conduction block as the end point of the procedure, often requiring the block to last for at least 30 minutes. AAD treatment was stopped before the procedure in seven case series and continued in one; in the remaining series the situation was unclear. Patients returned to AADs after the procedure in one case series, AADs were withheld in three and in the other 19 series the situation was unclear. Of the eight case series that reported anticoagulant use, six used heparin and two used warfarin. Time on anticoagulants after ablation, when reported, ranged from 7 days to 3 months. Details are presented in Appendix 3.9.

Freedom from arrhythmia at 12 months

The results of the case series are reported as two outcomes: freedom from atrial flutter and occurrence of other arrhythmias, mainly AF. The results for freedom from atrial flutter and other arrhythmias at 12 months are summarised in Tables 22 and 23.

Freedom from atrial flutter at 12 months’ follow-up was reported to range from 72% to 95% with a weighted mean of 88% (95% CI 85–92%). None of the series reported whether the results at these time points were due to single or repeat procedures.

Freedom from arrhythmia at mean follow-up

Most studies reported data for a mean follow-up period (Table 24). Across all durations of follow-up freedom from atrial flutter was maintained in 68.3–97.8% of patients ablated. At the longest duration of mean follow-up (30 months) the reported rate of freedom from atrial flutter was 85.1%. 131 However, these values are not easily interpreted as we do not know how many patients were at risk over the follow-up periods. Although the success rates for freedom from atrial flutter appear high, the proportion of patients with AF or atrial tachycardia or atypical atrial flutter at the end of follow-up is noteworthy; ignoring the data from the study that included repeat ablation124 the proportion of patients with another arrhythmia ranged from 8.7% to 53.2%. Thus, the proportion of patients who achieve freedom from arrhythmia following RFCA for atrial flutter is calculated to range from 31.7% to 86.9%.

Quality of life

Only two case series reported on QoL before and after ablation of atrial flutter. Feld and colleagues120 used three health-related QoL instruments that were administered at baseline and 3 and 6 months after ablation. Significant improvements over baseline were seen in 10/13 items reported. In the series by Calkins and colleagues,126 the SF-36 questionnaire and a symptom checklist were administered at baseline and 6 months after ablation. Improvements were seen in six of the eight SF-36 domains, in general physical and mental health and in 13 out of 16 cardiac symptoms.

Complications and mortality

The reported complications and mortality rates are summarised in Table 25 with further details in Appendices 3.11 and 3.12.

Overview

The study was designed to compare the cost-effectiveness of left atrial catheter ablation (LACA) with the cost-effectiveness of two alternative treatment strategies for the management of AF. The two alternative strategies were (1) rhythm control with amiodarone therapy and (2) rate control with a combination of digoxin and atenolol therapy. The main outcome measure was the degree of stroke reduction evaluated according to different baseline risks of stroke (low or moderate). In addition, short- and long-term outcomes including haemorrhage, drug toxicity, adverse events and procedural complications for each treatment strategy were incorporated. In all of the strategies patients also received antithrombotic or anticoagulant therapy. Those at moderate risk of stroke received warfarin whereas those at low risk received warfarin or aspirin.

The study was based on a deterministic Markov decision-analytic model of AF in a hypothetical 65-year-old US population at risk of stroke, evaluated over a cycle length of 3 months with patients modelled until death. A moderate risk of stroke was defined as having one risk factor (hypertension, diabetes mellitus, coronary artery disease or congestive heart failure), a low risk of stroke was defined as having no risk factors, and a high risk of stroke (≥ two risk factors) was excluded from the analysis. An additional cohort of 55-year-old patients at moderate risk was also considered. The study reports a societal perspective, although the exclusion of productivity costs means that the analysis is closer to a US payer perspective.

Summary of effectiveness data

An efficacy rate for LACA of 80% was derived from a number of large studies including a worldwide survey of the methods, safety and efficacy of curative catheter ablation of AF. 25 The rate incorporates a 30% redo ablation rate from AF recurrences or post-ablation atrial flutters during the first year. Because of the absence of long-term data on the relapse rate to AF for patients successfully restored to sinus rhythm with LACA, a 2% annual rate was assumed. The model incorporated complication rates due to LACA for tamponade, stroke, atrioesophageal fistula, death and other events.

The rate of restoration to sinus rhythm with rate control therapy was based on findings from a 5-year follow-up of the AFFIRM trial. 145 The relapse rate back to AF was conservatively assumed to be 5% annually. Rates for amiodarone efficacy, adverse events and relapse to AF were derived from a large number of literature sources.

The annual baseline stroke risks (without antithrombotic therapy) were based on conservative estimates from two previous decision-analytic models and a more recent meta-analysis. 146,147 The additional use of aspirin or warfarin therapy in conjunction with AF treatment was modelled on the basis of an annual stroke risk reduction with aspirin therapy of 22% (for both risk groups), and an annual stroke risk reduction with warfarin therapy of 45% and 35% compared with aspirin therapy at moderate and low risk respectively. The annual risk of stroke in patients with AF restored to sinus rhythm was unknown. The study assumed a rate of 0.5% for the cohort at low risk of stroke, representing a 29% relative risk reduction compared with low-risk AF patients on warfarin therapy. A similar relative risk reduction in the moderate risk cohorts was assumed. The model incorporated differential mortality and disability rates associated with stroke severity for patients on aspirin and warfarin therapy, and the rates for these were derived from a number of published literature sources. Patients who experienced a stroke were modelled to be twice as likely to have a recurrent stroke.

The annual baseline risks of haemorrhage for aspirin and warfarin therapy were based on results from a recent meta-analysis of pooled data from six randomised clinical trials. 148 A 1.5 relative risk of rehaemorrhage was based on data from the SPAF I–III clinical trials. 149

Summary of resource utilisation and cost data

Costs associated with treatment (single event and annual costs), cardiac events, annual care, intracranial bleed, stroke and adverse events associated with treatment were based on Medicare reimbursement rates, hospital accounting information, previously published studies and the Red Book for wholesale drug costs. Costs were reported in US dollars for the year 2004 and discounted at an annual rate of 3%. Productivity and personal care costs were not included in the analysis.

The cost associated with a single ablation was US$16,500. The cost of repeat ablation was assumed to be the same as the cost of the first procedure. The cost of complications from ablation was estimated to be the average of complication costs from tamponade and stroke with LACA at US$11,000. A rare atrioesophageal fistula complication with a 50% mortality rate was estimated to cost US$50,000 per event. The annual cost of warfarin therapy included the cost of regular 4-week serum monitoring and an office visit. Patients receiving rate control treatment were assigned atenolol and digoxin therapy. The cost of digoxin therapy included 6-month monitoring costs.

Summary of cost-effectiveness data

Quality of life for warfarin and aspirin therapies and individual health states were obtained from published studies. For clinical events (stroke, haemorrhage, drug toxicity and LACA complications) a disutility of 0.5 was applied for the duration of the event.

For 65-year-old patients with AF at moderate risk of stroke and on warfarin therapy, LACA (with an 80% efficacy rate) was estimated to be more effective but more costly than the alternative treatment strategies. The corresponding incremental cost-effectiveness ratio (ICER) was US$51,800 per QALY gained when compared with the use of rate control therapy. The use of amiodarone therapy was both less effective and more costly, and was dominated by rate control therapy. Similarly, for 55-year-old patients with AF at moderate risk of stroke, LACA was more costly and more effective with an ICER of US$28,700 per QALY gained relative to rate control therapy, which dominated amiodarone therapy. For 65-year-old patients at low risk for stroke and on aspirin therapy, LACA had an ICER of US$98,900 per QALY gained compared with rate control therapy, which dominated amiodarone therapy.

The ICERs above US$50,000 are driven largely by the significant upfront costs of LACA surgery. Any time horizon shorter than a lifetime would make LACA appear less cost-effective. The study estimated that for 65-year-old moderate stroke risk patients and an 80% LACA efficacy rate, relative stroke risk reductions with long-term sinus rhythm restoration of 42% and 11% would yield ICERs of less than US$50,000 and US$100,000 per QALY gained respectively. For the same patient population at low risk of stroke, LACA therapy could never be cost-effective unless the reduction in stroke risk was improbably large. The LACA efficacy rate is inversely related to the relative stroke risk reductions with sinus rhythm restoration, i.e. higher and lower LACA efficacy rates require correspondingly lower and higher stroke risk reductions to achieve the required cost-effective thresholds.

A series of univariate and multivariate sensitivity analyses were performed over a range of estimates for patients at moderate risk of stroke. The one-way analyses indicated that the ICER was most sensitive to the risk of stroke in AF with warfarin therapy, the discount rate, the cost of LACA, the utility and haemorrhage risk with warfarin therapy, the rate of recurrence of AF after LACA, and the conversion rate to sinus rhythm with rate control therapy. The multivariate sensitivity analyses were conducted using Monte Carlo simulation methods. The results indicate that, for 55-year-old moderate stroke risk patients, there is an 82% probability that the ICER comparing LACA treatment with rate control therapy is below US$50,000 per QALY gained. Among the 65-year-old moderate stroke risk population the cost-effectiveness is less certain with a 40% probability that the ICER falls below US$50,000 per QALY gained.

Systematic reviews

Economic evaluations

Quality of life

Prognosis

Adverse effects of amiodarone

Cochrane Database of Systematic Reviews (CDSR)

Cochrane Library – 2006 Issue 3. Searched 25 July 2006. URL: www.library.nhs.uk/

This search strategy retrieved two reviews (two completed and no protocols).

#1 MeSH descriptor Catheter Ablation, this term only

#2 catheter NEXT ablation*

#3 electric* NEXT ablation*

#4 fulguration*

#5 electrofulguration*

#6 cryoablation*

#7 radiofrequen* NEXT ablation*

#8 radio-frequen* NEXT ablation*

#9 RFA

#10 ablation NEXT catheter*

#11 atrial NEXT flutter NEXT ablation*

#12 transcatheter NEXT ablation*

#13 trans-catheter NEXT ablation*

#14 ablative NEXT cure*

#15 rf NEXT ablation*

#16 arrhythmia NEXT ablation*

#17 atrioventricular NEXT nod* NEXT ablation*

#18 av NEXT nod* NEXT modification*

#19 slow NEXT av NEXT nod* NEXT pathway* NEXT ablation*

#20 his NEXT bundle NEXT ablation*

#21 radiofrequen* NEXT linear NEXT ablation*

#22 radio-frequen* NEXT linear NEXT ablation*

#23 auricular NEXT fibrillation*

#24 MeSH descriptor Atrial Fibrillation, this term only

#25 atrial NEXT fibrillation*

#26 atrium NEXT fibrillation*

#27 MeSH descriptor Atrial Flutter, this term only

#28 auricular NEXT flutter*

#29 atrial NEXT flutter*

#30 atrial NEXT tachycardia*

#31 atrial NEXT tachyarrhythmia*

#32 atrium NEXT tachycardia*

#33 atrial NEXT arrhythmia*

#34 heart NEXT fibrillation*

#35 MeSH descriptor Tachycardia, Ectopic Atrial, this term only

#36 typical NEXT flutter*

#37 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22)

#38 (#23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35 OR #36)

#39 (#37 AND #38)

Database of Abstracts of Reviews of Effects (DARE)

CRD Internal Database – July 2006. Searched 26 July 2006.

This search strategy produced two records.

S catheter(w)ablation$or electric$(w)ablation$or fulguration$or electrofulguration$or cryoablation$or radiofrequen$(w)ablation$or radio-frequen$(w)ablation$or RFA or ablation(w)catheter$or atrial(w)flutter(w)ablation$or transcatheter(w)ablation$or trans-catheter(w)ablation$or ablative(w)cure$or rf(w)ablation$or arrhythmia(w)ablation$or atrioventricular(w)nod$(w)ablation$or av(w)nod$(w)modification$or slow(w)av(w)nod$(w)pathway$(w)ablation$or his(w)bundle(w)ablation$or radiofrequen$(w)linear(w)ablation$or radio-frequen$(w)linear(w)ablation$

S auricular(w)fibrillation$or atrial(w)fibrillation$or atrium(w)fibrillation$or auricular(w)flutter$or atrial(w)flutter$or atrial(w)tachycardia$or atrial(w)tachyarrhythmia$or atrium(w)tachycardia$or atrial(w)arrhythmia$or heart(w)fibrillation$or typical(w)flutter$

S s1 and s2

BIOSIS

DIALOG – 1993–present. Searched 28 July 2006. URL: http://library.dialog.com/

This search strategy resulted in 1142 records.

S catheter(w)ablation?

S electric?(w)ablation?

s fulguration?

s electrofulguration?

s cryoablation?

s radiofrequen?(w)ablation?

s radio-frequen?(w)ablation?

s RFA

s ablation(w)catheter?

s atrial(w)flutter(w)ablation?

s transcatheter(w)ablation?

s trans-catheter(w)ablation?

s ablative(w)cure?

s rf(w)ablation?

s arrhythmia(w)ablation?

s atrioventricular(w)nod?(w)ablation?

s av(w)nod?(w)modification?

s slow(w)av(w)nod?(w)pathway?(w)ablation?

s his(w)bundle(w)ablation?

s radiofrequen?(w)linear(w)ablation?

s radio-frequen?(w)linear(w)ablation?

s auricular(w)fibrillation?

s atrial(w)fibrillation?

s atrium(w)fibrillation?

s auricular(w)flutter?

s atrial(w)flutter?

s atrial(w)tachycardia?

s atrial(w)tachyarrhythmia?

s atrium(w)tachycardia?

s atrial(w)arrhythmia?

s heart(w)fibrillation?

s typical(w)flutter?

s s1:s21

s s22:s32

s s33 and s34

S S35/2000:2007

CENTRAL (Cochrane Central Register of Controlled Trials)

Cochrane Library – 2006 Issue 3. Searched 25 July 2006. URL: www.library.nhs.uk/

The same search strategy was used as for CDSR. This search resulted in 106 records.

Cumulative Index to Nursing and Allied Health Literature (CINAHL)

OvidWeb – 1982 to July Week 3 2006. Searched 25 July 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved 381 records.

1. catheter ablation/

2. catheter ablation$.ti,ab.

3. electric$ablation$.ti,ab.

4. fulguration$.ti,ab.

5. electrofulguration$.ti,ab.

6. cryoablation$.ti,ab.

7. radiofrequen$ablation$.ti,ab.

8. radio-frequen$ablation$.ti,ab.

9. RFA.ti,ab.

10. ablation catheter$.ti,ab.

11. atrial flutter ablation$.ti,ab.

12. transcatheter ablation$.ti,ab.

13. trans-catheter ablation$.ti,ab.

14. ablative cure$.ti,ab.

15. rf ablation$.ti,ab.

16. arrhythmia ablation$.ti,ab.

17. atrioventricular nod$ablation$.ti,ab.

18. av nod$modification$.ti,ab.

19. slow av nod$pathway$ablation$.ti,ab.

20. his bundle ablation$.ti,ab.

21. radiofrequen$linear ablation$.ti,ab.

22. radio-frequen$linear ablation$.ti,ab.

23. auricular fibrillation$.ti,ab.

24. Atrial Fibrillation$.ti,ab.

25. Atrium Fibrillation$.ti,ab.

26. auricular flutter$.ti,ab.

27. atrial tachycardia$.ti,ab.

28. atrial tachyarrhythmia$.ti,ab.

29. Atrium Tachycardia$.ti,ab.

30. atrial arrhythmia$.ti,ab.

31. heart fibrillation$.ti,ab.

32. typical flutter$.ti,ab.

33. Atrial Fibrillation/

34. Atrial Flutter/

35. Tachycardia, Atrial/

36. Atrial flutter$.ti,ab

37. or/1–22

38. or/23–36

39. 37 and 38

40. limit 39 to yr=“2000 – 2006”

EMBASE

OvidWeb – 1980 to 2006 Week 29. Searched 25 July 2006. URL: http://gateway.ovid.com/athens

This search strategy resulted in 2184 records.

1. catheter ablation/

2. catheter ablation$.ti,ab.

3. electric$ablation$.ti,ab.

4. fulguration$.ti,ab.

5. electrofulguration$.ti,ab.

6. cryoablation$.ti,ab.

7. radiofrequen$ablation$.ti,ab.

8. radio-frequen$ablation$.ti,ab.

9. RFA.ti,ab.

10. ablation catheter$.ti,ab.

11. atrial flutter ablation$.ti,ab.

12. transcatheter ablation$.ti,ab.

13. trans-catheter ablation$.ti,ab.

14. ablative cure$.ti,ab.

15. rf ablation$.ti,ab.

16. arrhythmia ablation$.ti,ab.

17. atrioventricular nod$ablation$.ti,ab.

18. av nod$modification$.ti,ab.

19. slow av nod$pathway$ablation$.ti,ab.

20. his bundle ablation$.ti,ab.

21. radiofrequen$linear ablation$.ti,ab.

22. radio-frequen$linear ablation$.ti,ab.

23. auricular fibrillation$.ti,ab.

24. Atrial Fibrillation$.ti,ab.

25. Atrium Fibrillation$.ti,ab.

26. auricular flutter$.ti,ab.

27. atrial tachycardia$.ti,ab.

28. atrial tachyarrhythmia$.ti,ab.

29. Atrium Tachycardia$.ti,ab.

30. atrial arrhythmia$.ti,ab.

31. heart fibrillation$.ti,ab.

32. typical flutter$.ti,ab.

33. Heart Atrium Fibrillation/

34. Heart Atrium Flutter/

35. Atrial flutter$.ti,ab

36. or/1–22

37. or/23–35

38. 36 and 37

39. limit 37 to yr=“2000 – 2007” (2184)

Health Technology Assessment (HTA) Database

CRD Internal Database – July 2006. Searched 26 July 2006.

The same search strategy was used as for the DARE database. This search produced nine records.

MEDLINE

OvidWeb – 1966 to July Week 2 2006. Searched 25 July 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved 2057 records.

1. catheter ablation/

2. 2 catheter ablation$.ti,ab.

3. electric$ablation$.ti,ab.

4. fulguration$.ti,ab.

5. electrofulguration$.ti,ab.

6. cryoablation$.ti,ab.

7. radiofrequen$ablation$.ti,ab.

8. radio-frequen$ablation$.ti,ab.

9. RFA.ti,ab.

10. ablation catheter$.ti,ab.

11. atrial flutter ablation$.ti,ab.

12. transcatheter ablation$.ti,ab.

13. trans-catheter ablation$.ti,ab.

14. ablative cure$.ti,ab.

15. rf ablation$.ti,ab.

16. arrhythmia ablation$.ti,ab.

17. atrioventricular nod$ablation$.ti,ab.

18. av nod$modification$.ti,ab.

19. slow av nod$pathway$ablation$.ti,ab.

20. his bundle ablation$.ti,ab.

21. radiofrequen$linear ablation$.ti,ab.

22. radio-frequen$linear ablation$.ti,ab.

23. auricular fibrillation$.ti,ab.

24. Atrial Fibrillation/

25. Atrial Fibrillation$.ti,ab.

26. Atrium Fibrillation$.ti,ab.

27. Atrial Flutter/

28. auricular flutter$.ti,ab.

29. Atrial Flutter$.ti,ab.

30. atrial tachycardia$.ti,ab.

31. atrial tachyarrhythmia$.ti,ab.

32. Atrium Tachycardia$.ti,ab.

33. atrial arrhythmia$.ti,ab.

34. heart fibrillation$.ti,ab.

35. tachycardia, ectopic atrial/

36. typical flutter$.ti,ab.

37. or/1–22

38. or/23–36

39. 37 and 38

40. limit 39 to yr=“2000 – 2006”

MEDLINE In-process and other non-indexed citations

OvidWeb – 24 July 2006. Searched 25 July 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved 109 records.

1. catheter ablation$.ti,ab.

2. electric$ablation$.ti,ab.

3. fulguration$.ti,ab.

4. electrofulguration$.ti,ab.

5. cryoablation$.ti,ab.

6. radiofrequen$ablation$.ti,ab.

7. radio-frequen$ablation$.ti,ab.

8. RFA.ti,ab.

9. ablation catheter$.ti,ab.

10. atrial flutter ablation$.ti,ab.

11. transcatheter ablation$.ti,ab.

12. trans-catheter ablation$.ti,ab.

13. ablative cure$.ti,ab.

14. rf ablation$.ti,ab.

15. arrhythmia ablation$.ti,ab.

16. atrioventricular nod$ablation$.ti,ab.

17. av nod$modification$.ti,ab.

18. slow av nod$pathway$ablation$.ti,ab.

19. his bundle ablation$.ti,ab.

20. radiofrequen$linear ablation$.ti,ab.

21. radio-frequen$linear ablation$.ti,ab.

22. auricular fibrillation$.ti,ab.

23. Atrial Fibrillation$.ti,ab.

24. Atrium Fibrillation$.ti,ab.

25. auricular flutter$.ti,ab.

26. Atrial Flutter$.ti,ab.

27. atrial tachycardia$.ti,ab.

28. atrial tachyarrhythmia$.ti,ab.

29. Atrium Tachycardia$.ti,ab.

30. atrial arrhythmia$.ti,ab.

31. heart fibrillation$.ti,ab.

32. typical flutter$.ti,ab.

33. or/1–21

34. or/22–32

35. 33 and 34

36. limit 36 to yr=“2000 – 2006”

Science Citation Index (SCI)

Web of Knowledge – 1956 to present. Searched 26 July 2006. URL: http://wos.mimas.ac.uk/

This search strategy retrieved 2889 records.

catheter ablation* OR electric* ablation* OR fulguration* OR electrofulguration* OR cryoablation* OR radiofrequen* ablation* OR radio-frequen* ablation* OR RFA OR ablation catheter* OR atrial flutter ablation* OR transcatheter ablation* OR trans-catheter ablation* OR ablative cure* OR rf ablation* OR arrhythmia ablation* OR atrioventricular nod* ablation* OR av nod* modification* OR slow av nod* pathway* ablation* OR his bundle ablation* OR radiofrequen* linear ablation* OR radio-frequen* linear ablation*

AND

auricular fibrillation* OR atrial fibrillation* OR atrium fibrillation* OR auricular flutter* OR atrial flutter* OR atrial tachycardia* OR atrial tachyarrhythmia* OR atrium tachycardia* OR atrial arrhythmia* OR heart fibrillation* OR typical flutter*

ISI Proceedings: Science and Technology

Web of Knowledge – 1990 to present. Searched 27 July 2006. URL: http://wos.mimas.ac.uk/

This database was searched with the same search strategy as for SCI and produced 597 records.

Zetoc Conferences

MIMAS – 1993 to present. Searched 26 July 2006. URL: http://zetoc.mimas.ac.uk/

After within-database deduplication this series of individual search strings retrieved 288 records.

“catheter ablation*” – 202 records

“electric* ablation*” – 1 record

fulguration – 14 records

electrofulguration – 0 records

cryoablation – 89 records

“radiofrequen* ablation*” – 201 records

“radio-frequen* ablation*” – 19 records

rfa – 29 records

“ablation catheter*” – 6 records

“atrial flutter ablation*” – 2 records

“transcatheter ablation*” – 2 records

“trans-catheter ablation*” – 0 records

“ablative cure*” – 1 record

“rf ablation*” – 26 records

“arrhythmia ablation*” – 0 records

“atrioventricular nod* ablation*” – 3 records

“av nod* modification*” – 2 records

“slow av nod* pathway ablation*” – 0 records

“his bundle ablation*” – 4 records

“radiofrequen* linear ablation*” – 0 records

“radio-frequen* linear ablation*” – 0 records

EconLit

WebSPIRS – 1969 to June 2006. Searched 26 July 2006. URL: http//arc.uk.ovid.com/

This search retrieved no records.

#1 catheter ablation* OR electric* ablation* OR fulguration* OR electrofulguration* OR cryoablation* OR radiofrequen* ablation* OR radio-frequen* ablation* OR RFA OR ablation catheter* OR atrial flutter ablation* OR transcatheter ablation* OR trans-catheter ablation* OR ablative cure* OR rf ablation* OR arrhythmia ablation* OR atrioventricular nod* ablation* OR av nod* modification* OR slow av nod* pathway* ablation* OR his bundle ablation* OR radiofrequen* linear ablation* OR radio-frequen* linear ablation*

#2 auricular fibrillation* OR atrial fibrillation* OR atrium fibrillation* OR auricular flutter* OR atrial flutter* OR atrial tachycardia* OR atrial tachyarrhythmia* OR atrium tachycardia* OR atrial arrhythmia* OR heart fibrillation* OR typical flutter*

#1 and #2

Health Economics Evaluation Database (HEED)

CD-ROM – July 2006. Searched 26 July 2006.

This search strategy retrieved 12 records.

ablation* OR fulguration* OR electrofulguration* OR cryoablation* OR ablative OR av OR rfa

AND

fibrillation* OR tachyarrhythmia* OR tachycardia* OR arrhythmia* OR flutter*

IDEAS

Current. Searched 26 July 2006. URL: http://ideas.repec.org/

This search strategy retrieved no records.

ablation* fulguration* electrofulguration* cryoablation* ablative av

NHS Economic Evaluation Database (NHS EED)

CRD Internal Database – July 2006. Searched 26 July 2006.

The same search strategy was used as for DARE. This search produced six records.

BMJ Clinical Evidence

URL: www.clinicalevidence.com/ceweb/index.jsp

No relevant articles (one forthcoming relevant article).

Health Evidence Bulletin Wales

URL: http://hebw.cf.ac.uk/

This database was browsed and no relevant guidelines were identified.

HSTAT

URL: www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=hstat

A search for catheter ablation retrieved two guidelines.

National Guidelines Clearing House

URL: www.guideline.gov/

A search for catheter ablation retrieved 22 guidelines.

NICE

URL: www.nice.org.uk/

A search for catheter ablation retrieved 124 guidelines.

NLH Guidelines Finder

URL: www.library.nhs.uk/guidelinesFinder/

A search for catheter ablation retrieved six guidelines.

Scottish Intercollegiate Guidelines Network (SIGN)

URL: www.sign.ac.uk/

This database was browsed and one potentially relevant guideline was identified.

TRIP

URL: www.tripdatabase.com/index.html

A search for catheter ablation retrieved 32 guidelines.

ClinicalTrials.gov

Searched 2 August 2006. URL: www.clinicaltrials.gov/

This search retrieved 35 records, which were all entered into the EndNote library.

(catheter ablation OR electric ablation OR fulguration OR electrofulguration OR cryoablation OR radiofrequency ablation OR radio-frequency ablation OR RFA OR ablation catheter OR atrial flutter ablation OR transcatheter ablation OR trans-catheter ablation OR ablative cure OR rf ablation OR arrhythmia ablation OR atrioventricular nod ablation OR av nod modification OR slow av nod pathway ablation OR his bundle ablation OR radiofrequency linear ablation OR radio-frequency linear ablation) AND (auricular fibrillation OR atrial fibrillation OR atrium fibrillation OR auricular flutter OR atrial flutter OR atrial tachycardia OR atrial tachyarrhythmia OR atrium tachycardia OR atrial arrhythmia OR heart fibrillation OR typical flutter) [ALL-FIELDS]

MetaRegister of Controlled Trials

Searched 9 August 2006. URL: www.controlled-trials.com/

All registers [except ClinicalTrials.gov and the National Research Register (NRR), which were searched directly] were selected. This search strategy retrieved 48 records (six studies were deemed potentially relevant and imported into the EndNote library).

The search was for any of these words:

ablation fulguration electrofulguration cryoablation RFA ablative

National Research Register (NRR)

2006 Issue 3. Searched 9 August 2006. URL: www.update-software.com/national/

This search retrieved 30 records, which were all entered into the EndNote library.

#1 catheter ablation* OR electric* ablation* OR fulguration* OR electrofulguration* OR cryoablation* OR radiofrequen* ablation* OR radio-frequen* ablation* OR RFA OR ablation catheter* OR atrial flutter ablation* OR transcatheter ablation* OR trans-catheter ablation* OR ablative cure* OR rf ablation* OR arrhythmia ablation* OR atrioventricular nod* ablation* OR av nod* modification* OR slow av nod* pathway* ablation* OR his bundle ablation* OR radiofrequen* linear ablation* OR radio-frequen* linear ablation*

#2 auricular fibrillation* OR atrial fibrillation* OR atrium fibrillation* OR auricular flutter* OR atrial flutter* OR atrial tachycardia* OR atrial tachyarrhythmia* OR atrium tachycardia* OR atrial arrhythmia* OR heart fibrillation* OR typical flutter*

#3 #1 and #2

Research Findings Electronic Register (ReFeR)

Searched 9 August 2006. URL: www.info.doh.gov.uk/doh/refr_web.nsf/Home?OpenForm

This search strategy retrieved five records (none of which were relevant and which were therefore not imported into the EndNote library).

ablation OR fulguration OR electrofulguration OR cryoablation OR RFA OR ablative

Cochrane Database of Systematic Reviews (CDSR)

Cochrane Library – 2006 Issue 3. Searched 17 October 2006. URL: www.library.nhs.uk/

This search strategy retrieved 22 reviews (18 completed and four protocols).

#1 Atrioventricular Node/

#2 (av near/2 node near/2 ablat*)

#3 (av near/2 nodal near/2 ablat*).

#4 (atrioventricular near/2 node near/2 ablat*)

#5 (atrioventricular near/2 nodal near/2 ablat*)

#6 (atrio-ventricular near/2 node near/2 ablat*)

#7 (atrio-ventricular near/2 nodal near/2 ablat*)

#8 (a-v near/2 node near/2 ablati*)

#9 (a-v near/2 nodal near/2 ablat*)

#10 Amiodarone/

#11 amiodarone

#12 cordarone

#13 exp Anti-Arrhythmia Agents/

#14 antiarrhythmia

#15 antiarrhythmic

#16 Anti-Arrhythmia

#17 Anti-Arrhythmic

#18 antifibrillatory

#19 cardiac next depressant*

#20 myocardial next depressant*

#21 (rate near/2 control*)

#22 (rhythm near/2 control*)

#23 auricular next fibrillation*

#24 Atrial Fibrillation/

#25 Atrial next Fibrillation*

#26 Atrium next Fibrillation*

#27 Atrial Flutter/

#28 auricular next flutter*

#29 Atrial next flutter*

#30 atrial next tachycardia*

#31 atrial next tachyarrhythmia*

#32 Atrium next tachycardia*

#33 atrial next arrhythmia*

#34 heart next fibrillation*

#34 tachycardia, ectopic atrial/

#35 typical next flutter*

#36 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22)

#37 (#23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35)

#38 (#36 AND #37)

Cumulative Index to Nursing and Allied Health Literature (CINAHL)

OvidWeb – 1982 to October Week 1 2006. Searched 17 October 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved 23 records.

1. Atrioventricular Node/

2. (av adj2 node adj2 ablat$).ti,ab.

3. (av adj2 nodal adj2 ablat$).ti,ab.

4. (atrioventricular adj2 node adj2 ablat$).ti,ab.

5. (atrioventricular adj2 nodal adj2 ablat$).ti,ab.

6. (atrio-ventricular adj2 node adj2 ablat$).ti,ab.

7. (atrio-ventricular adj2 nodal adj2 ablat$).ti,ab.

8. (a-v adj2 node adj2 ablati$).ti,ab.

9. (a-v adj2 nodal adj2 ablat$).ti,ab.

10. Amiodarone/

11. amiodarone.ti,ab.

12. cordarone.ti,ab.

13. exp Anti-Arrhythmia Agents/

14. antiarrhythmia.ti,ab.

15. antiarrhythmic.ti,ab.

16. Anti-Arrhythmia.ti,ab.

17. Anti-Arrhythmic.ti,ab.

18. antifibrillatory.ti,ab.

19. cardiac depressant$.ti,ab.

20. myocardial depressant$.ti,ab.

21. (rate adj2 control$).ti,ab.

22. (rhythm adj2 control$).ti,ab.

23. (pace or pacing or pacemaker$).ti,ab.

24. Pacemaker, Artificial/

25. auricular fibrillation$.ti,ab.

26. Atrial Fibrillation/

27. Atrial Fibrillation$.ti,ab.

28. Atrium Fibrillation$.ti,ab.

29. Atrial Flutter/

30. auricular flutter$.ti,ab.

31. Atrial Flutter$.ti,ab.

32. atrial tachycardia$.ti,ab.

33. atrial tachyarrhythmia$.ti,ab.

34. Atrium Tachycardia$.ti,ab.

35. atrial arrhythmia$.ti,ab.

36. heart fibrillation$.ti,ab.

37. tachycardia, ectopic atrial/

38. typical flutter$.ti,ab.

39. or/25–38

40. CARDIAC PACING, ARTIFICIAL/

41. Meta Analysis/

42. systematic review/

43. systematic review.pt.

44. (systematic adj4 (review$or overview$)).ti,ab.

45. (literature adj2 review$).ti,ab.

46. literature review/

47. or/41–46

48. auricular fibrillation.ti,ab.

49. Atrial Fibrillation/

50. Atrial Fibrillation$.ti,ab.

51. Heart Atrium Fibrillation/

52. Atrium Fibrillation.ti,ab.

53. Atrial Flutter/

54. auricular flutter.ti,ab.

55. Atrial Flutter.ti,ab.

56. atrial tachycardia.ti,ab.

57. atrial tachyarrhythmia$.ti,ab.

58. exp Tachycardia, Supraventricular/

59. Supraventricular Tachycardia$.ti,ab.

60. Supraventricular Tachycardia/

61. Atrium Tachycardia.ti,ab.

62. cardiac arrhythmia$.ti,ab.

63. atrioventricular junction$.ti,ab.

64. atrial arrhythmia$.ti,ab.

65. supraventricular arrhythmia$.ti,ab.

66. premature cardiac complex$.ti,ab.

67. atrial premature complex$.ti,ab.

68. paroxysmal tachycardia$.ti,ab.

69. heart arrhythmia$.ti,ab.

70. atrium arrhythmia$.ti,ab.

71. heart fibrillation$.ti,ab.

72. Arrhythmia/

73. or/48–72

74. (or/1–24) or 40

75. 39 and 47 and 74

76. limit 75 to (english and yr=“2000 – 2007”)

Database of Abstracts of Reviews of Effects (DARE)

CRD Internal Database – October 2006. Searched 17 October 2006.

This search strategy produced 33 records.

s atrioventricular node/kwo

s av(2w)node(2w)ablat$

s av(2w)nodal(2w)ablat$

s atrioventricular(2w)node(2w)ablat$

s atrioventricular(2w)nodal(2w)ablat$

s atrio(w)ventricular(2w)node(2w)ablat$

s atrio(w)ventricular(2w)nodal(2w)ablat$

s a(w)v(2w)node(2w)ablati$

s a(w)v(2w)nodal(2w)ablat$

s amiodarone

s cordarone

s antiarrhythmia

s antiarrhythmic

s anti(w)arrhythmia

s anti(w)arrhythmic

s antifibrillatory

s cardiac(w)depressant$

s myocardial(w)depressant$

s rate(2w)control$

s rhythm(2w)control$

s s1 or s2 or s3 or s4 or s5 or s6 or s7 or s8 or s9 or s10 or s11 or s12 or s13 or s14 or s15 or s16 or s17 or s18 or s19 or s20

s auricular(w)fibrillation$

s atrial(w)fibrillation$

s atrium(w)fibrillation$

s auricular(w)flutter$

s atrial(w)flutter$

s atrial(w)tachycardia$

s atrial(w)tachyarrhythmia$

s atrium(w)tachycardia$

s atrial(w)arrhythmia$

s heart(w)fibrillation$

s typical(w)flutter$

s s22 or s23 or s24 or s25 or s26 or s27 or s28 or s29 or s30 or s31 or s32

s s21 and s33

EMBASE

OvidWeb – 1996 to 2006 Week 41. Searched 17 October 2006. URL: http://gateway.ovid.com/athens

This search strategy resulted in 203 records.

1. Atrioventricular Node/

2. (av adj2 node adj2 ablat$).ti,ab.

3. (av adj2 nodal adj2 ablat$).ti,ab.

4. (atrioventricular adj2 node adj2 ablat$).ti,ab.

5. (atrioventricular adj2 nodal adj2 ablat$).ti,ab.

6. (atrio-ventricular adj2 node adj2 ablat$).ti,ab.

7. (atrio-ventricular adj2 nodal adj2 ablat$).ti,ab.

8. (a-v adj2 node adj2 ablati$).ti,ab.

9. (a-v adj2 nodal adj2 ablat$).ti,ab.

10. Amiodarone/

11. amiodarone.ti,ab.

12. cordarone.ti,ab.

13. exp Anti-Arrhythmia Agents/

14. antiarrhythmia.ti,ab.

15. antiarrhythmic.ti,ab.

16. Anti-Arrhythmia.ti,ab.

17. Anti-Arrhythmic.ti,ab.

18. antifibrillatory.ti,ab.

19. cardiac depressant$.ti,ab.

20. myocardial depressant$.ti,ab.

21. (rate adj2 control$).ti,ab.

22. (rhythm adj2 control$).ti,ab.

23. (pace or pacing or pacemaker$).ti,ab.

24. Pacemaker, Artificial/

25. cardiac pacing, artificial/

26. auricular fibrillation$.ti,ab.

27. heart atrium Fibrillation/

28. Atrial Fibrillation$.ti,ab.

29. Atrium Fibrillation$.ti,ab.

30. heart atrium Flutter/

31. atrial flutter$.ti,ab.

32. auricular flutter$.ti,ab.

33. Atrial Flutter$.ti,ab.

34. atrial tachycardia$.ti,ab.

35. atrial tachyarrhythmia$.ti,ab.

36. Atrium Tachycardia$.ti,ab.

37. atrial arrhythmia$.ti,ab.

38. heart fibrillation$.ti,ab.

39. typical flutter$.ti,ab.

40. Meta Analysis/

41. systematic review/

42. (systematic adj4 (review$or overview$)).ti,ab.

43. (literature adj2 review$).ti,ab.

44. or/1–25

45. or/26–39

46. or/40–43

47. 44 and 45 and 46

48. limit 47 to (english language and yr=“2000 – 2007”)

Health Technology Assessment (HTA) Database

CRD Internal Database – October 2006. Searched 17 October 2006.

This search strategy produced one record.

s atrioventricular node/kwo

s av(2w)node(2w)ablat$

s av(2w)nodal(2w)ablat$

s atrioventricular(2w)node(2w)ablat$

s atrioventricular(2w)nodal(2w)ablat$

s atrio(w)ventricular(2w)node(2w)ablat$

s atrio(w)ventricular(2w)nodal(2w)ablat$

s a(w)v(2w)node(2w)ablati$

s a(w)v(2w)nodal(2w)ablat$

s amiodarone

s cordarone

s antiarrhythmia

s antiarrhythmic

s anti(w)arrhythmia

s anti(w)arrhythmic

s antifibrillatory

s cardiac(w)depressant$

s myocardial(w)depressant$

s rate(2w)control$

s rhythm(2w)control$

s s1 or s2 or s3 or s4 or s5 or s6 or s7 or s8 or s9 or s10 or s11 or s12 or s13 or s14 or s15 or s16 or s17 or s18 or s19 or s20

s auricular(w)fibrillation$

s atrial(w)fibrillation$

s atrium(w)fibrillation$

s auricular(w)flutter$

s atrial(w)flutter$

s atrial(w)tachycardia$

s atrial(w)tachyarrhythmia$

s atrium(w)tachycardia$

s atrial(w)arrhythmia$

s heart(w)fibrillation$

s typical(w)flutter$

s s22 or s23 or s24 or s25 or s26 or s27 or s28 or s29 or s30 or s31 or s32

s s21 and s33

MEDLINE

OvidWeb – 1996 to October Week 1 2006. Searched 17 October 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved 73 records.

1. Atrioventricular Node/

2. (av adj2 node adj2 ablat$).ti,ab.

3. (av adj2 nodal adj2 ablat$).ti,ab.

4. (atrioventricular adj2 node adj2 ablat$).ti,ab.

5. (atrioventricular adj2 nodal adj2 ablat$).ti,ab.

6. (atrio-ventricular adj2 node adj2 ablat$).ti,ab.

7. (atrio-ventricular adj2 nodal adj2 ablat$).ti,ab.

8. (a-v adj2 node adj2 ablati$).ti,ab.

9. (a-v adj2 nodal adj2 ablat$).ti,ab.

10. Amiodarone/

11. amiodarone.ti,ab.

12. cordarone.ti,ab.

13. exp Anti-Arrhythmia Agents/

14. antiarrhythmia.ti,ab.

15. antiarrhythmic.ti,ab.

16. Anti-Arrhythmia.ti,ab.

17. Anti-Arrhythmic.ti,ab.

18. antifibrillatory.ti,ab.

19. cardiac depressant$.ti,ab.

20. myocardial depressant$.ti,ab.

21. (rate adj2 control$).ti,ab.

22. (rhythm adj2 control$).ti,ab.

23. auricular fibrillation$.ti,ab.

24. Atrial Fibrillation/

25. Atrial Fibrillation$.ti,ab.

26. Atrium Fibrillation$.ti,ab.

27. Atrial Flutter/

28. auricular flutter$.ti,ab.

29. Atrial Flutter$.ti,ab.

30. atrial tachycardia$.ti,ab.

31. atrial tachyarrhythmia$.ti,ab.

32. Atrium Tachycardia$.ti,ab.

33. atrial arrhythmia$.ti,ab.

34. heart fibrillation$.ti,ab.

35. tachycardia, ectopic atrial/

36. typical flutter$.ti,ab.

37. or/23–36

38. (pace or pacing or pacemaker$).ti,ab.

39. Cardiac Pacing, Artificial/

40. Pacemaker, Artificial/

41. (or/1–22) or (or/38–40)

42. 37 and 41

43. meta-analysis.ti,ab.

44. meta-analysis.pt.

45. meta-analysis/

46. (systematic$adj4 (review$or overview$)).ti,ab.

47. “review literature”/

48. (literature adj2 review$).ti,ab.

49. or/43–48

50. 42 and 49

51. limit 50 to english language

52. limit 51 to yr=“2000 – 2006”

MEDLINE In-process and other non-indexed citations

OvidWeb – October 16 2006. Searched 17 October 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved two records.

1. (av adj2 node adj2 ablat$).ti,ab.

2. (av adj2 nodal adj2 ablat$).ti,ab.

3. (atrioventricular adj2 node adj2 ablat$).ti,ab.

4. (atrioventricular adj2 nodal adj2 ablat$).ti,ab.

5. (atrio-ventricular adj2 node adj2 ablat$).ti,ab.

6. (atrio-ventricular adj2 nodal adj2 ablat$).ti,ab.

7. (a-v adj2 node adj2 ablati$).ti,ab.

8. (a-v adj2 nodal adj2 ablat$).ti,ab.

9. amiodarone.ti,ab

10. cordarone.ti,ab.

11. antiarrhythmia.ti,ab.

12. antiarrhythmic.ti,ab.

13. Anti-Arrhythmia.ti,ab.

14. Anti-Arrhythmic.ti,ab.

15. antifibrillatory.ti,ab.

16. cardiac depressant$.ti,ab.

17. myocardial depressant$.ti,ab.

18. (rate adj2 control$).ti,ab.

19. (rhythm adj2 control$).ti,ab.

20. (pace or pacing or pacemaker$).ti,ab.

21. auricular fibrillation$.ti,ab.

22. Atrial Fibrillation$.ti,ab.

23. Atrium Fibrillation$.ti,ab.

24. atrial flutter$.ti,ab.

25. auricular flutter$.ti,ab.

26. Atrial Flutter$.ti,ab.

27. atrial tachycardia$.ti,ab.

28. atrial tachyarrhythmia$.ti,ab.

29. Atrium Tachycardia$.ti,ab.

30. atrial arrhythmia$.ti,ab.

31. heart fibrillation$.ti,ab.

32. typical flutter$.ti,ab.

33. (systematic adj4 (review$or overview$)).ti,ab.

34. (literature adj2 review$).ti,ab.

35. meta-analysis.pt.

36. or/1–20

37. or/21–32

38. or/33–35

39. 36 and 37 and 38

Results of search strategies for studies on the clinical effectiveness of selected comparators to catheter ablation

A total of 271 unique bibliographic records were retrieved (357 before deduplication).

MEDLINE

OvidWeb – 1966 to November Week 2 2006. Searched 21 November 2006. URL: http://gateway.ovid.com/athens

This search strategy retrieved 193 records.

1. Atrioventricular Node/

2. (av adj2 node adj2 ablat$).ti,ab.

3. (av adj2 nodal adj2 ablat$).ti,ab.

4. (atrioventricular adj2 node adj2 ablat$).ti,ab.

5. (atrioventricular adj2 nodal adj2 ablat$).ti,ab.

6. (atrio-ventricular adj2 node adj2 ablat$).ti,ab.

7. (atrio-ventricular adj2 nodal adj2 ablat$).ti,ab.

8. (a-v adj2 node adj2 ablati$).ti,ab.

9. (a-v adj2 nodal adj2 ablat$).ti,ab.

10. Amiodarone/

11. amiodarone.ti,ab.

12. cordarone.ti,ab.

13. exp Anti-Arrhythmia Agents/

14. antiarrhythmia.ti,ab.

15. antiarrhythmic.ti,ab.

16. Anti-Arrhythmia.ti,ab.

17. Anti-Arrhythmic.ti,ab.

18. antifibrillatory.ti,ab.

19. cardiac depressant$.ti,ab.

20. myocardial depressant$.ti,ab.

21. (rate adj2 control$).ti,ab.

22. (rhythm adj2 control$).ti,ab.

23. or/1–22

24. auricular fibrillation$.ti,ab.

25. Atrial Fibrillation/

26. Atrial Fibrillation$.ti,ab.

27. Atrium Fibrillation$.ti,ab.

28. Atrial Flutter/

29. auricular flutter$.ti,ab.

30. Atrial Flutter$.ti,ab.

31. atrial tachycardia$.ti,ab.

32. atrial tachyarrhythmia$.ti,ab.

33. Atrium Tachycardia$.ti,ab.

34. atrial arrhythmia$.ti,ab.

35. heart fibrillation$.ti,ab.

36. tachycardia, ectopic atrial/

37. typical flutter$.ti,ab.

38. or/24–37

39. 23 and 38

40. economics/

41. exp “costs and cost analysis”/

42. economic value of life/

43. exp economics,hospital/

44. exp economics, medical/

45. economics, nursing/

46. economics, pharmaceutical/

47. exp models,economic/

48. exp “fees and charges”/

49. exp budgets/

50. ec.fs.

51. (cost or costs or costly or costing$).ti,ab.

52. (economic$or price$or pricing or pharmacoeconomic$).ti,ab.

53. or/40–52

54. 39 and 53

55. 54

56. limit 55 to english language