Adherence to guidelines for atrial fibrillation management of patients referred to cardiology departments: Studio Italiano multicentrico sul Trattamento della Fibrillazione Atriale (SITAF)

Abstract

Aims

The purpose of this study was to evaluate adherence to national guidelines on the non-pharmacologic (ablative) treatment of atrial fibrillation (AF).

Methods and results

This prospective, observational, transversal study enrolled 1256 consecutive in- and outpatients referred to 43 cardiology departments between 1 and 31 October 2008 for the management of AF as a primary diagnosis. A rhythm-control strategy (cardioversion, antiarrhythmic medication, pace-maker implantation, substrate ablation, alone or in combination) was prescribed in 865 (69%) of the patients and a rate-control strategy [drugs, atrioventricular junction ablation and pace-maker implantation (Ablate and Pace)] in 285 (23%). Specifically, substrate catheter ablation was indicated by the attending cardiologist in 187 (14.9%) patients and Ablate and Pace in 29 (2.3%). According to guideline indications, substrate catheter ablation would have been indicated in 183 (14.6%) patients, but only 105 (57%) of these were correctly identified by the attending cardiologist ( K statistics for agreement for indications 0.49). Atrioventricular junction ablation and pace-maker implantation would have been indicated in 108 (8.6%) patients, but only 29 (27%) of these were correctly identified by the attending cardiologist ( K statistics for agreement for indications 0.06).

Conclusion

About a quarter of patients referred to cardiology departments for AF management have potential indications for non-pharmacological treatment according to the guidelines. Substrate catheter ablation was offered by the attending cardiologist in a percentage similar to that expected, but concordance with guideline indications was moderate. Atrioventricular junction ablation and pace-maker implantation was largely underused.

Introduction

Atrial fibrillation (AF) is the commonest sustained cardiac arrhythmia in humans. ¹ Several population-based studies have provided information on the incidence, prevalence, and outcome of AF in the general population. Atrial fibrillation is associated with increased mortality ^{2 , 3} and causes a five-fold rise in stroke risk; it frequently coexists with heart failure, leading to an even further increase in mortality. The economic burden of AF has grown in the past decades and is expected to rise further as the trend in AF prevalence and hospitalizations increases. Therefore, an adequate treatment strategy for AF is warranted. Two arrhythmia strategies for AF treatment are currently offered to patients: rate control and rhythm control. Rate control aims at slowing the ventricular rate during ongoing AF, whereas rhythm control aims to terminate AF and maintain sinus rhythm. When pharmacologic antiarrhythmic treatment fails, reasonable non-pharmacologic alternatives are atrioventricular junction ablation and pace-maker implantation (Ablate and Pace), and substrate catheter ablation for rate control or rhythm control, respectively. Given the positive results of a large number of clinical studies, the international scientific societies (AHA/ACC/ESC, HRS/EHRA/ECAS) recently revised their guidelines on AF management; ^{4 , 5} in these documents substrate catheter ablation is included as a reasonable treatment, but the criteria for patient selection for non-pharmacologic therapy are not fully detailed. In 2006 the Associazione Italiana di Aritmologia e Cardiostimolazione (AIAC) produced guidelines ⁶ for the comprehensive management of AF, including detailed recommendations on non-pharmacologic (ablative) treatments. It subsequently sponsored the Studio Italiano multicentrico sul Trattamento della Fibrillazione Atriale (SITAF) in order to evaluate adherence to these guidelines among the Italian population. The primary endpoint of this survey was to evaluate adherence to current guidelines in the non-pharmacologic management of AF. The secondary endpoint was to assess the differences in treatment between centres that routinely perform ablative therapies of AF and those that do not.

Methods

This multicentre, observational, transversal study was prospectively performed in 43 cardiology departments homogenously distributed throughout Italy (see list of participating centres in Appendix 2). The centres were selected according to their common practice of performing both pharmacological and non-pharmacological ablative treatments [ablation centres (AC)], or pharmacological treatment only [non-ablation centres (NonAC)] with the aim of obtaining a balanced sample size of patients. The study was reviewed and approved by the local Institutional Review Boards.

Patients

Consecutive patients, older than 18 years, admitted to cardiology departments (urgent or planned hospitalization), or referred to their outpatient clinics for AF management as the primary clinical reason were included. Patients with AF evaluated for reasons unrelated to the arrhythmia, or those in whom AF manifested during hospitalization were excluded. Enrolment started on 1 October 2008 and ended on 31 October 2008. A daily record of enrolments was kept by the local investigators for each centre. The specifically designed report form documented the patient's characteristics, presence and type of underlying heart disease, standard transthoracic echocardiography data, antiarrhythmic treatment, anticoagulation status, and treatments offered. Diagnosis of hypertensive heart disease required a history of hypertension associated to echocardiographic evidence of left ventricular hypertrophy and/or of systolic or diastolic dysfunction. Coronary artery disease was diagnosed on the basis of a documented history of angina/myocardial infarction or of the presence of significant obstructive coronary lesions on angiography.

Atrial fibrillation was categorized according to the clinical pattern on presentation (paroxysmal, persistent, and long-lasting persistent/permanent), time of onset, aetiology, severity of symptoms, impairment of quality of life, and the presence of thrombo-embolic risk factors (see definitions in Appendix 1).

All treatments prescribed were included in the report, but only indications for non-pharmacologic therapies were considered for the analysis.

Data collection and validation

In each centre, data were collected daily by the investigator responsible on an electronic case-report form (CRF: https://clinic.air-tel.it/Sitaf/ ). Data were transferred through the Internet to the central database. Data were entered either on-line or off-line with daily data transfer. By means of a validation plan integrated into the data entry software, data were checked for missing or contradictory entries and values outside the normal range. Additional edit checks were performed by the coordinators of the study. Patient identification was registered in the participating centres, but was not transferred to the central database.

Study protocol

Comparison was made between the non-pharmacologic treatments actually prescribed by the attending cardiologist in accordance with his own clinical practice and those theoretically indicated on the basis of class I recommendations of the AIAC guidelines ( Table  1 ).

On completion of enrolment, the Endpoint Committee assigned the indications according to the recommendation of the AIAC guidelines. ⁶ Since substrate catheter ablation and Ablate and Pace were indicated in those guidelines for ‘relatively young’ and ‘old’ patients, respectively, an arbitrary age cut-off of 70 years was chosen.

Statistical analysis

Values are expressed as mean ±1 SD for continuous variables or median and 25th and 75th percentiles (interquartile range, IQR) for skewed distribution, and counts and percentages for categorical variables. Comparison between actual and theoretical indications for the non-pharmacological treatments was performed on the overall population, and in the AC and NonAC subgroups by means of cross-tabulation and Kappa statistics for measurements of agreement between indication for treatment according to the centre's practice and national guidelines. A multivariable logistic regression analysis was used to assess which variables (among baseline clinical characteristics and type of centre) drove the decision to undertake ablation therapy. On univariable analysis, differences in proportions were tested by means of the χ ² test, whereas differences in means were tested by applying the appropriate parametric or non-parametric test, according to distribution. The non-collinear variables with a P -value <0.1 were fitted into the multivariable model; the effect of aggregating clusters of 43 centres was also considered. A two-tailed P -value <0.05 was taken to indicate statistical significance. Statistical analyses were performed by means of the Statistica package (Stat Soft, Inc.).

Results

Patient enrolment

A total of 1285 AF patients from 43 centres met the inclusion criteria; 29 patients were excluded from the analysis owing to missing data. Therefore, 1256 patients (median 20 patients per centre, IQR 15–33) were considered for analysis. The recruiting centres roughly represented 6% of the Italian Cardiology Departments. Twenty centres were classified as ACs and 23 as NonACs. The ACs enrolled 672 patients (median 20 patients per centre, IQR 7–38) and the NonACs enrolled 584 patients (median 18 patients per centre, IQR 16–27).

Patient characteristics

Baseline clinical characteristics of the overall population and the comparison between patients enrolled at ACs or NonACs are summarized in Table  2 .

Interventions

A treatment was indicated by the attending cardiologist in 1184 out of 1256 (94%) patients ( Table  3 ). A rhythm-control strategy (cardioversion, antiarrhythmic medication, pace-maker implantation, substrate ablation, alone or in combination) was prescribed in 865 (69%) cases, and a rate-control strategy (drugs, Ablate and Pace) in 285 (23%). Specifically, substrate catheter ablation was prescribed by the attending cardiologist in 187 (14.9%) patients and Ablate and Pace in 29 (2.3%) patients.

According to class I guideline recommendations, substrate catheter ablation ( Table  4 ) should have been indicated in 183 (14.6%) patients, but was correctly offered by the attending cardiologist in only 105 (57%) cases ( K statistics for agreement for indications: 0.47). ACs prescribed substrate catheter ablation more frequently than NonACs (23.1 vs. 5.5%; P < 0.01) and with a better accordance with guideline indications (59 vs. 46%; K statistics for agreement for indications: 0.49 and 0.36, respectively). Nevertheless, when corrected for baseline variables, five parameters proved to be independently predictive of the referring physician's decision to perform substrate catheter ablation: refractoriness to medical therapy, younger age, paroxysmal or persistent form of AF and severity of symptoms, but not the type of centre ( Table  5 ).

According to class I guideline indications, Ablate and Pace ( Table  6 ) should have been prescribed in 108 (8.6%) patients, but was correctly offered by the attending cardiologist in only 29 (27%) cases ( K statistics for agreement for indications: 0.06 for all centres; 0.05 and 0.06 for ACs and NonACs, respectively). When corrected for baseline variables, only the permanent form of AF independently predicted the attending physician's decision to perform Ablate and Pace ( Table  5 ).

Discussion

This prospective multicentre survey shows that, among patients referred to a cardiology department for AF management as the primary clinical reason:

• a rhythm-control strategy was offered to more patients than a rate-control strategy;

• about a quarter of patients had an established indication for non-pharmacological therapy according to class I indications of the guidelines;

• nearly 15% of patients had an established indication for substrate ablation therapy according to class I indications of the guidelines;

• substrate catheter ablation was offered by the attending cardiologists in a percentage similar to that expected, but concordance with guideline indications was moderate;

• when corrected for clinical variables, the prescription of substrate catheter ablation was not significantly influenced by the type of the centre (AC or NonAC);

• Ablate and Pace treatment is largely underused.

After only a decade since the first publication demonstrating focal sources of AF in the pulmonary veins, ⁷ catheter-based therapy aimed at electrically isolating these structures is now the standard ablation procedure for the treatment of recurrent AF. Recent controlled, randomized clinical trials have shown that ablation therapy is superior to antiarrhythmic pharmacological treatment in maintaining sinus rhythm in patients with drug-refractory AF. ⁸ In particular, most of the published trials have reported a 70–85% success rate in patients with paroxysmal AF and without marked underlying cardiac disease, with a relatively safe risk profile (3–5% overall complication rate; 1–2% severe events). ^9–12 Nonetheless, information regarding the number of potential candidates for this therapy is not available yet. According to the results of the present study, 14.6% of the patients referred to a cardiology department for AF management have a class I indication for substrate ablation therapy. In contrast, if the criteria for patient selection proposed in the ACC/AHA/ESC ⁴ guidelines for AF management (i.e. paroxysmal/persistent AF, symptomatic, refractory to pharmacological therapy, with no or little left atrium enlargement) were taken into account, indications for ablation treatment in the present population would rise to 33.2%. This large discrepancy between AIAC and ACC/AHA/ESC guideline indications (14.6 vs 33.2%) is not surprising, since the latter guidelines do not consider any age cut-off for patient selection. Indeed, given the age-related prevalence of AF (greater in older patients), the exclusion of patients older than 70 years reasonably explains the markedly lower percentage of patients with an indication for substrate catheter ablation according to the Italian guideline recommendations.

In other published surveys, substrate catheter ablation was actually indicated in a lower percentage of cases than that observed in our study (14.9%). Indeed, in the Euro Heart Survey on AF, ¹³ which enrolled 5333 patients in 35 European countries, ablation was offered in only 2.8% of cases. However, this survey collected patients between 2003 and 2004, and was published in 2005, when AF catheter ablation was in an investigational phase and was not yet included in the guidelines. More recently, in a survey conducted in Switzerland, Meiltz et al . ¹⁴ reported that AF catheter ablation was prescribed for 4.8% of their study population. However, this study significantly differs from our survey, since only practice cardiologists and a selected AF population (i.e. ambulatory patients) were involved.

On univariate analysis, ACs offered substrate catheter ablation more frequently than NonACs. However, the population referred to ACs significantly differed from that enrolled at NonACs in terms of age, type and duration of AF, severity of symptoms, and refractoriness to medical treatment. When these clinical factors were taken into account in the multivariable analysis, the difference between ACs and NonACs in substrate catheter ablation prescription was no longer statistically significant. However, as the perception of the clinical state of the patient is largely subjective, a different attitude of the cardiologists working in ACs or in NonACs in evaluating the severity of symptoms and the response to medical therapy could have played a role in the treatment indications.

Among patients to whom substrate catheter ablation was offered, a proper indication according to guideline recommendations was given in 57% of the cases overall, with a slightly higher adherence in ACs than in NonACs (59 vs. 46%). Although we are unable to give an objective explanation of this inconsistency due to lack of specific items in the report form, we suppose that the following reasons are likely: (i) the relatively recent introduction of guidelines into clinical practice (incomplete penetration); (ii) different physicians' perception of the efficacy of the procedure (enthusiasm in someone vs. skepticism in some other); (iii) lack of established results of the substrate ablation especially with regard to long-term follow-up and hard clinical endpoints. Therefore, further efforts, including educational programs, should be made in order to improve adherence to guideline indications.

Although Ablate and Pace is a well-established, standardized therapy, especially in older patients with drug-refractory AF, ^{15 , 16} it was offered in fewer cases than expected according to guideline indications, with no difference emerging between ACs and NonACs. Probably some concerns remain with regard to the palliative nature of the treatment, and to the possible complications associated with cardiac device implantation. In addition, there is a lack of randomized trials regarding the use of Ablate and Pace with either conventional or biventricular pacing. ¹⁷

Randomized controlled trials have failed to demonstrate statistically significant differences in all-cause mortality between rhythm- and rate-control strategies in the selected patient populations studied. ¹⁸ In agreement with the Euro Heart Survey, ¹³ in our study a rhythm-control strategy was adopted in the majority of the patients (68 vs. 23%), regardless of the type of centre. In recent surveys performed on ambulatory patients ¹⁴ or patients admitted to a general hospital, ¹⁹ a near-equal distribution of the use of rhythm- and rate-control strategies in patients with primary diagnosis of AF was found. Nonetheless, cardiologists tended to offer a rhythm-control strategy more frequently than internal medicine specialists. ¹⁹ This observation may partially explain the preference for this therapy in our survey. The potential benefits of restoring and maintaining sinus rhythm, including a lower risk of heart failure and a better quality of life ^20–22 may also have played a role in prescribing a rhythm-control strategy.

Limitations

We intended to collect data from a broad sample of cardiology centres in Italy in order to evaluate the application of current AF management guidelines in clinical practice. We enrolled consecutive AF patients referred to cardiology departments primarily for the management of this arrhythmia. Since we were unable to control the criteria for referral, the characteristics of the patients might have differed among the centres. Thus, it might not be possible to replicate our results in other settings in which different referral practices are in use. For this reason, the study cannot be regarded as an epidemiological study. However, the mix of the 43 participating centres was thought to provide a sufficiently representative sample of current practice in Italy. Assuming that the recruiting centres account for 6% of the total Italian population (i.e. 3.6 million inhabitants), we estimate that 4180 patients per million inhabitants per year are referred to a cardiology department for AF management. Of these, 610 patients per million inhabitants per year have a class I indication for substrate catheter ablation, and 359 patients per million inhabitants per year have a class I indication for Ablate and Pace.

The Italian guidelines for AF management propose substrate ablation in ‘relatively young’ patients and Ablate and Pace in ‘old’ patients. The steering committee of this study arbitrarily applied a cut-off age of 70 years between relatively young and old patients.

Conclusions

The main results of this prospective multicentre study are that about a quarter of patients referred to a cardiology department for AF management have potential indications for non-pharmacological treatment according to national guidelines. Substrate catheter ablation was prescribed by the attending cardiologist in a similar percentage to that indicated by the guidelines, although concordance with guideline indications was moderate. Ablate and Pace therapy was largely underused.

Conflict of interest: none declared.

Funding

Funding to support the Electronic Data Capture (EDC AIRTEL) was supported by a grant from Boston Scientific Italia.

Appendix 1: Definitions

• Risk factors for stroke: age >75 years, previous transient ischaemic attack or stroke, hypertension, diabetes, heart failure, left ventricular ejection fraction <0.35.

• New-onset AF: episode of AF diagnosed for the first time by a physician.

• Idiopathic AF: atrial fibrillation occurring in patients without previously or currently diagnosed hypertension, coronary artery disease, heart failure, valvular heart disease, tachycardiomyopathy, sick sinus syndrome, chronic obstructive pulmonary disease, or thyroid disease.

• Paroxysmal AF: recurrent AF that terminates spontaneously and lasts <7 days (mostly <24–48 h).

• Persistent AF: recurrent AF or sustained AF that (regardless of duration) does not terminate spontaneously but only on pharmacological therapy or electrical cardioversion.

• Long-lasting persistent or Permanent AF: atrial fibrillation has been present for a long time (more than 1 year), cardioversion has not been indicated, or one or more attempts have failed to restore reliable sinus rhythm.

• Duration of AF: duration in years of the clinical history of AF.

• Symptomatic AF: the presence of one or more of the following symptoms in relation to AF: palpitations, dyspnea, chest pain, syncope, dizziness, fatigue, or other non-specified symptoms.

• Relevant symptoms: very intense symptoms which significantly impair quality of life in the judgment of the referring physician.

• Refractoriness to pharmacological treatment: inefficacy, contraindications, or drug intolerance to medical treatment in the judgment of the referring physician.

• Treatment indicated: oral anticoagulant therapy; cardioversion: pharmacological or electrical conversion; antiarrythmic drugs: class IA, IC, or III antiarrhythmic agents (Vaughan Williams classification) for the maintenance of sinus rhythm; rate control: digitalis, class II, or class IV antiarrhythmic drugs for the control of ventricular response; pace-maker implantation: implantation mainly for brady–tachy form of sick sinus syndrome; Ablate and Pace: atrioventricular node ablation followed by conventional or biventricular pacing; substrate AF catheter ablation: one of the currently used types of treatment, such as segmental PV ablation or circumferential PV ablation.

• Of importance, one or more type of treatment could be selected by the investigator.

Appendix B: Organization of the survey

Steering Committee: M.T. (Chairman), N.B. (co.Chairman), Maria Grazia Bongiorni, Domenico Catanzariti, Giuseppe Inama, Roberto Verlato.

Endpoint Committee: Carlo Menozzi, Jorge-Antonio Salerno, Massimo Santini, Sergio Sermasi, Paolo Della Bella.

Statistical support: Catherine Klersy, MD.

Acknowledgements: This study was endorsed by the Associazione Italiana Aritmologia e Cardiostimolazione (AIAC). Organizational support was provided by Boston Scientific Italia (Francesco Accardi, Giovanni Raciti). Database for data collection was provided by Airon Telematica S.r.l.

List of participating centres and investigators:

• Istituto Clinico Mater Domini, Castellanza (VA)—G.S., Paolo Moretti

• Azienda Ospedaliera S. Maria Nuova, Reggio Emilia—M.I., Fabio Quartieri

• Ospedale Cisanello, Pisa—Maria Grazia Bongiorni, Ezio Soldati

• Ospedale S. Maria del Carmine, Rovereto (TN)—Domenico Catanzariti, Massimiliano Maines

• Ospedale Civile, Camposampiero (PD)—Roberto Verlato, Maria Stella Baccillieri

• Ospedale S. Filippo Neri, Roma—R.R., Stefania Di Fusco

• Clinica Gavazzeni Humanitas, Bergamo—Giosuè Mascioli, Alessio Borrelli

• Centro Cardiologico Monzino, Milano—Gaetano Fassini, Benedetta Majocchi

• Policlinico S. Matteo, Pavia—Maurizio Landolina, Antonio Sanzo

• Ospedale di Circolo e Fondazione Macchi, Varese—R.D., Raffaella Marazzi

• Ospedale Cardinal Massaia, Asti—Fiorenzo Gaita, Marco Scaglione

• Ospedale S. Chiara, Trento—Maurizio Del Greco, Massimiliano Marini

• Ospedale Civile, Conegliano Veneto (TV)—N.S., Pietro Delise

• Ospedale Regionale, Treviso—Roberto Mantovan, Vittorio Calzolari

• Ospedale Umberto I, Mestre (VE)—Antonio Raviele, S.T.

• Ospedale Civile, Mirano (VE)—Franco Zoppo, Emanuele Bertaglia

• Clinica Mediterranea, Napoli—Giuseppe Stabile, Assunta Iuliano

• Ospedale Miulli, Acquaviva delle Fonti (BA)—Massimo Grimaldi, Grigorios Katsouras

• Ospedale Maggiore, Crema—Giuseppe Inama, Claudio Pedrinazzi

• Ospedale Civile, Lavagna (GE)—M.B., Francesco Croci

• Dipartimento di Cardiologia, Università di Foggia, Foggia—M.B., Pier Luigi Pellegrino

• Ospedale S. Giovanni di Dio e Ruggi d'Aragona, Salerno—Andrea Campana, Michele Manzo

• Ospedale Garibaldi-Nesima, Catania—Michele Gulizia, Daniele Mangiameli

• Ospedali Riuniti, Trieste—Massimo Zecchin, Aneta Aleksova

• Ospedale Civile, Imperia—Giacomo Musso, Silvia Negroni

• Ospedale Civile, Bentivoglio (BO)—Biagio Sassone, Luca Gabrieli

• Ospedale Civile, Cento (FE)—Paolo Alboni, Massimo Sciammarella

• Ospedale S. Orsola, Bologna—Giuseppe Boriani, Mauro Biffi

• Ospedale Civile, Piacenza—Alessandro Capucci, Giovanni Villani

• Ospedale Infermi, Rimini—Marco Marconi, Maurizio Mezzetti

• Ospedale G. Panico, Tricase (LE)—M.A., Gabriele De Luca De Masi

• Ospedale Civile, Casarano (LE)—Giacinto Pettinati, Fernando Desantis

• Ospedale Scillesi d'America, Scilla (RC)—Michele Musolino, Alfonso Marrari

• Ospedale E. Moscatello, Augusta (SR)—Giacomo Chiarandà, Giuseppe Busacca

• Nuovo Ospedale di Sassuolo, Sassuolo (MO)—Francesco Melandri, Mila Menozzi

• Ospedale B. Ramazzini, Carpi (MO)—Elia De Maria, Stefano Cappelli

• Policlinico, Modena—Vincenzo Malavasi, Mirza Becirovic

• Ospedale Morgagni-Pierantoni, Forlì—Alberto Bandini, Paolo Golia

• Ospedale S. Maria Delle Croci, Ravenna—Corrado Tomasi, Angelo Placci

• Ospedale Sant'Anna, Catanzaro—S.I., Fabiano Gennaro

• Ospedale Maggiore della Carità, Novara—Eraldo Occhetta, Luca Venegoni

• Spedali Civili, Brescia—Antonio Curnis, Luca Bontempi

• Ospedale S. Giovanni Bosco, Torino—Mauro Bensoni, Moballeghi Hossin

References