Summary
Background
Limited French data are available for the different clinical types (paroxysmal, persistent and permanent) of atrial fibrillation and their comorbidities (AF).
Aims
To provide contemporary insights into the characteristics and management of outpatients with a history of or current AF in France.
Methods
EPHA is a national, observational, cross-sectional, multicentre descriptive study with retrospective data collection relating to the management, treatment and hospitalization of patients with AF.
Results
One thousand three hundred and thirty-one patients (mean age: 74 ± 11 years [55.7% ≥ 75 years]; 58.8% men) were included into the study between February 2009 and May 2009; their data were collected during the past 12 months. Of these, 38.2% had paroxysmal AF, 10.0% persistent AF and 51.8% permanent AF. Most patients had at least one cardiovascular risk factor (80.8%). Almost all patients (96.6%) had received an antiarrhythmic drug in the previous year, of which 59.6% received a rhythm control strategy (class I, class III) with or without rate control strategy (class II, class IV, digitalis) and 40.6% received a rate control strategy exclusively. Almost all (94.4%) patients were treated with an antithrombotic: 83.4% with a vitamin K antagonist and 21.9% with antiplatelet therapy. Almost one-fifth (18.4%) of patients had been hospitalized related to AF at least once in the previous year. Patients with paroxysmal and persistent AF were hospitalized more frequently (20.0% and 31.1%, respectively) than patients with permanent AF (14.8%).
Conclusions
About half of the patients had paroxysmal or persistent AF. Four-fifths of AF patients had at least one cardiovascular risk factor. The use of antiarrhythmic and antithrombotic treatments was very high. The rhythm control strategy was preferred in patients with paroxysmal or persistent AF.
Résumé
Introduction
Il existe peu de données françaises sur les différents types cliniques de fibrillation auriculaire (FA) (paroxystique, persistante ou permanente) et leurs comorbidités.
Objectif
Décrire les caractéristiques et la prise en charge des patients ambulatoires présentant un antécédent de FA ou actuellement en FA, en France.
Méthodes
L’étude EPHA est une étude française, observationnelle, transversale, multicentrique descriptive avec recueil rétrospectif des données de prise en charge, de traitement et d’hospitalisation des patients atteints de FA.
Résultats
Mille trois cent trente-et-un patients (âge moyen : 74 ± 11 ans [55,7 % ≥ 75 ans] ; 58,8 % de sexe masculin) ont été inclus dans l’étude entre février 2009 et mai 2009, les données ont été recueillies sur les 12 derniers mois. Parmi ces patients, 38,2 % avaient une FA paroxystique, 10 % une FA persistante et 51,8 % une FA permanente. La majorité de ces patients avait au moins un facteur de risque cardiovasculaire (80,8 %). La quasi-totalité des patients (96,6 %) étaient sous traitement antiarythmique dans l’année précédant l’inclusion. Parmi les patients traités, 59,6 % recevaient un traitement de contrôle du rythme (antiarythmiques de classe I ou III) avec ou sans traitement de contrôle de la fréquence cardiaque (antiarythmiques de classe II, de classe IV ou digitaliques) et 40,6 % recevaient exclusivement un traitement de contrôle de la fréquence cardiaque. La majorité des patients (94,4 %) recevaient un antithrombotique : 83,4 % un antivitamine K et 21,9 % un antiplaquettaire. Près d’un cinquième des patients (18,4 %) ont été hospitalisés au moins une fois pour raison cardiovasculaire liée à la FA dans l’année précédente. Les patients en FA paroxystique ou persistante ont plus souvent été hospitalisés (20,0 % et 31,1 %, respectivement) que les patients en FA permanente (14,8 %).
Conclusion
Environ la moitié des patients avaient une FA paroxystique ou persistante. Quatre patients sur cinq en FA avaient au moins un facteur de risque cardiovasculaire. L’utilisation de traitements antiarythmiques et antithrombotiques était très élevée. Une stratégie de contrôle du rythme était plus fréquent chez les patients en FA paroxystique ou persistante que chez les patients en FA permanente. Ces résultats soulignent la nécessité de traitements ou techniques avec un meilleur rapport bénéfice–risque afin d’optimiser la prise en charge de cette pathologie dont la prévalence ne cesse de croître.
Introduction
Atrial fibrillation (AF) is the most frequently observed cardiac arrhythmia in clinical practice. It is a major cause of long-term morbidity and mortality including major cardiovascular events as ischaemic stroke, heart failure and all-cause death . One in six strokes occurs in patients with AF , and mortality in these patients is about double that of subjects in sinus rhythm and is related directly to the severity of underlying heart disease .
The estimated prevalence of AF ranges from 0.4–1% in the general population, rising to 8% in patients older than 80 years . AF affects around 2.2 million people in the United States and 4.5 million in Europe ; between 600,000 and one million patients in France are estimated to have AF . These figures will escalate with the ageing of the population, leading to increasing healthcare expenses of AF management.
AF can be defined as paroxysmal (AF that terminates spontaneously and generally lasts ≤ 7 days), persistent (AF that does not terminate spontaneously and usually lasts > 7 days), and permanent (AF in which cardioversion has failed or has not been attempted). When a patient has had two or more episodes, AF is considered recurrent. The distribution of clinical AF types varies in France . Permanent AF is diagnosed in 30–50% of AF cases, with paroxysmal or persistent AF accounting for the remaining 50–70%. To the best of our knowledge, there is no recent French epidemiological study describing the different types of AF and age, comorbidities and treatment. The Euro Heart Survey recently described treatment and prognosis up to 1 year, according to type of AF, in subjects in 35 European countries, but included only 51 patients from France .
The primary objective of the present study is to describe the distribution of clinical AF types in patients with a history of or current AF who were being followed-up by cardiologists in France. The secondary objectives are: to determine the patients’ characteristics, cardiovascular risk factors, and comorbidities for each type of AF; to describe treatment strategies during the past year; to describe the cardiovascular hospitalizations related to AF and to its complications; and to estimate the healthcare resources consumed.
Methods
EPHA is a national, observational, cross-sectional, multicentre descriptive study with retrospective data collection relating to the management, treatment and hospitalization of patients with AF.
Physician selection
Three thousand cardiologists were identified at random from a comprehensive national database (TVF physician identification database) of all cardiologists in France. The aim was to enrol 300 physicians, 100 from hospital-based practices and 200 from office-based or both hospital- and office-based practices willing to participate in the study. Each cardiologist completed a standardized case report form that collected information about their age, sex, year of graduation, department of practice, type of practice (office- or hospital-based practice), mean number of patients seen each month, and mean number of patients seen each month with a history of or current documented AF.
The representativeness of the physician sample was checked by comparing the distribution of participating cardiologists with national statistics published by the Direction de la recherche, des études, de l’évaluation et des statistiques (DREES) , in order to verify the lack of selection biases. The physicians enrolled patients seen during the course of normal clinical practice and no changes were imposed by the study in the diagnosis and treatment of these individuals.
Patient selection
Investigating physicians undertook to include the first eight consecutive patients who fulfilled the inclusion criteria, until 1300 patients had been enrolled.
Men and women aged ≥ 18 years were eligible for inclusion if they had been seen in consultation, and monitored, by a cardiologist for at least 1 year; had a history of or current documented AF (on an electrocardiogram [ECG] or Holter ECG) for at least 1 year, irrespective of the rhythm on the day of inclusion; and provided written informed consent before participating in the study.
Patients were excluded from this study if they presented with AF due to a transient cause (e.g., thyrotoxicosis, excessive alcohol intake, myocarditis, pericarditis, acute myocardial infarction, pulmonary embolism, metabolic disturbances or electrocution), with AF following cardiac surgery within the past 3 months or had not been monitored regularly by the investigator (last visit > 12 months before inclusion).
Cardiologists completed case report forms that collected data on the patient’s demographics, cardiovascular risk factors, medical history, clinical examination, characteristics of AF, electrocardiographic and echocardiographic data, biological data, management of AF during the past 12 months, and healthcare resources related to AF management (including number of hospitalizations related to AF). Direct healthcare resources only were considered.
Statistical methods
The sample size was determined in order to have a degree of precision ≥ 2.5% to describe the different types of AF, 1300 patients needed to be enrolled. For the statistical analysis, prespecified descriptive analyses were performed. These analyses (percentages and means) are based on the number of non-missing values for each variable. All categorical variables were tested between the three types of AF using chi-square, and continuous variables using analysis of variance.
The study protocol was submitted to two national authorities, Comité consultatif sur le traitement de l’information en matière de recherche dans le domaine de la santé (CCTIRS) and Commission nationale de l’informatique et des libertés (CNIL), and was conducted in accordance with French law and the declaration of Helsinki.
Results
Study population
Of 416 cardiologists who agreed to participate, 234 enrolled at least one patient into the study. A total of 1470 patients were enrolled between 25 February and 6 May 2009. Of these, 41 did not provide written informed consent and were excluded from the study. A further 98 did not fulfil the inclusion criteria. The analysis is therefore based on data from 1331 patients.
Patient characteristics at consultation
The characteristics of the 1331 patients included in the study are shown in Table 1 . The mean age was 74 ± 11 years (55.7% were ≥ 75 years) and 58.8% were men. Over half of the 1331 patients enrolled had permanent AF (51.8%, 95% CI 49.2–54.5), 38.2% (95% CI 35.6–40.8) had paroxysmal AF and 10.0% (95% CI 8.4–11.6) had persistent AF. Younger patients were more likely to present paroxysmal or persistent AF, whereas older patients (≥ 75 years) more often presented permanent AF ( Fig. 1 ).
| Paroxysmal AF ( n = 508) | Persistent AF ( n = 133) | Permanent AF ( n = 690) | p value a | Total ( n = 1331) | |
|---|---|---|---|---|---|
| Demographics | |||||
| Men | 292 (57.5) | 80 (60.2) | 411 (59.6) | 0.73 | 783 (58.8) |
| Age, years | 71 ± 11 | 72 ± 11 | 76 ± 9 | < 0.001 | 74 ± 11 |
| ≥ 75 years | 221 (43.5) | 64 (48.1) | 457 (66.2) | < 0.001 | 742 (55.7) |
| Cardiovascular risk factors | |||||
| ≥ 1 risk factor | 406 (80.1) | 108 (81.2) | 560 (81.3) | 0.87 | 1074 (80.8) |
| Hypertension | 318 (78.3) | 93 (86.1) | 450 (80.4) | 0.19 | 861 (80.2) |
| Diabetes mellitus | 71 (17.5) | 23 (21.3) | 121 (21.6) | 0.27 | 215 (20.0) |
| Dyslipidaemia | 244 (60.1) | 61 (56.5) | 320 (57.3) | 0.64 | 625 (58.3) |
| Smoking history | |||||
| Current | 29 (7.1) | 6 (5.6) | 26 (4.7) | 0.30 | 61 (5.7) |
| Stopped < 3 years | 15 (3.7) | 3 (2.8) | 13 (2.3) | 31 (2.9) | |
| Stopped ≥ 3 years | 107 (26.4) | 38 (35.2) | 163 (29.2) | 308 (28.7) | |
| Family history of premature | |||||
| CVD | |||||
| Paternal b | 25 (7.4) | 9 (11.3) | 28 (6.6) | 0.35 | 62 (7.4) |
| Maternal c | 14 (4.1) | 4 (5.0) | 10 (2.4) | 0.28 | 28 (3.3) |
| Early stroke (< 45 years) | 4 (1.2) | 4 (5.0) | 1 (0.2) | 0.003 | 9 (1.1) |
| No family history of premature CVD | 301 (88.5) | 66 (82.5) | 386 (91.5) | 0.0448 | 753 (89.4) |
| Medical history and comorbidities | |||||
| ≥ 1 medical history/comorbidity | 365 (72.0) | 103 (77.4) | 598 (87.0) | < 0.001 | 1066 (80.3) |
| Heart failure | 65 (17.8) | 29 (28.2) | 250 (41.8) | < 0.001 | 344 (32.3) |
| NYHA class I | 15 (23.1) | 7 (24.1) | 31 (12.5) | 0.13 | 53 (15.5) |
| NYHA class II | 39 (60.0) | 15 (51.7) | 170 (68.5) | 224 (65.5) | |
| NYHA class III | 11 (16.9) | 6 (20.7) | 44 (17.7) | 61 (17.8) | |
| NYHA class IV | 0 | 1 (3.4) | 3 (1.2) | 4 (1.2) | |
| Ischaemic stroke | 29 (7.9) | 9 (8.7) | 45 (7.5) | 0.91 | 83 (7.8) |
| Transient ischaemic attack | 31 (8.5) | 5 (4.9) | 40 (6.7) | 0.37 | 76 (7.1) |
| Ischaemic heart disease | 82 (22.5) | 29 (28.2) | 165 (27.6) | 0.18 | 276 (25.9) |
| Non-ischaemic heart disease | 101 (27.7) | 25 (24.3) | 200 (33.4) | 0.058 | 326 (30.6) |
| Valvular disease | 108 (29.6) | 32 (31.1) | 258 (43.1) | < 0.001 | 398 (37.3) |
| Rheumatic | 18 (16.7) | 6 (18.8) | 71 (27.6) | 0.063 | 95 (23.9) |
| Pulmonary disease | 51 (14) | 20 (19.4) | 80 (13.4) | 0.26 | 151 (14.2) |
| Thyroid disease | 62 (17.0) | 14 (13.6) | 86 (14.4) | 0.49 | 162 (15.2) |
| Hypothyroidism | 51 (82.3) | 9 (69.2) | 60 (73.2) | 0.36 | 120 (76.4) |
| Hyperthyroidism | 11 (17.7) | 4 (30.8) | 22 (26.8) | 37 (23.6) | |
| Chronic renal insufficiency | 30 (8.2) | 7 (6.8) | 54 (9.0) | 0.73 | 91 (8.5) |
| Sleep apnea syndrome | 27 (7.4) | 10 (9.7) | 40 (6.7) | 0.54 | 77 (7.2) |
| Pacemaker | 65 (17.8) | 10 (9.7) | 138 (23.1) | 0.0032 | 213 (20.0) |
a Analysis of variance or chi-square.
b MI or sudden death before age 55 in father or parent of first-degree male relative.
c MI or sudden death before age 65 in mother or parent of first-degree female relative.
Cardiovascular risk factors, medical history and comorbidities
Most patients (80.8%) had at least one cardiovascular risk factor, including hypertension (80.2%), dyslipidaemia (58.3%), a history of or current tobacco consumption (37.3%), diabetes mellitus (20.0%) and a family history of premature cardiovascular disease (10.6%). The proportion of patients with at least one cardiovascular risk factor was similar irrespective of the type of AF (80.1% for paroxysmal, 81.2% for persistent, 81.3% for permanent) ( Table 1 ).
Almost one-third of patients with AF had a history of heart failure, which was most frequent among the group with permanent AF (41.8% vs. 28.2% for persistent AF and 17.8% for paroxysmal AF, p < 0.001). Patients with permanent AF more often had New York Heart Association class III–IV heart failure (6.8%) compared to patients with paroxysmal or persistent AF (2.8%). Eight per cent had a history of ischaemic stroke; 25.9% had ischaemic heart disease; 30.6% had non-ischaemic heart disease; and 37.3% had valvular disease. Patients with permanent AF more often had valvulopathy (43.1%) compared to patients with persistent (31.1%) or paroxysmal (29.6%) AF; one-fifth of patients had a pacemaker.
Overall, 15.2% had thyroid disease, most commonly hypothyroidism (76.4%), and 8.5% had chronic renal insufficiency, but there was no significant difference in the prevalences across AF types.
CHADS 2 score at study entry
The mean CHADS 2 (heart failure, hypertension, age ≥ 75 years, diabetes, stroke or transient ischaemic attack [TIA]) score was highest in patients with permanent AF, intermediate in those with persistent AF and lowest in patients with paroxysmal AF ( p < 0.001) ( Table 2 ). One-third of patients with permanent AF were at high risk of a stroke (CHADS 2 score > 2) vs. 27.8% of those with persistent AF and 17.1% with paroxysmal AF.
| Paroxysmal AF ( n = 508) | Persistent AF ( n = 133) | Permanent AF ( n = 690) | p value a | Total ( n = 1331) | |
|---|---|---|---|---|---|
| Reason for consultation | |||||
| AF monitoring | 367 (72.7) | 104 (78.8) | 480 (70.3) | 0.12 | 951 (72.0) |
| Other reason | 122 (24.2) | 23 (17.4) | 165 (24.2) | 310 (23.5) | |
| Both | 16 (3.2) | 5 (3.8) | 38 (5.6) | 59 (4.5) | |
| Mean time since diagnosis of AF (years) | 5.3 (5.0) | 4.9 (4.9) | 7.0 (5.5) | < 0.001 | 6.1 (5.3) |
| Symptoms | |||||
| None | 216 (43.1) | 48 (36.1) | 272 (40.1) | 0.29 | 536 (40.9) |
| Palpitations | 186 (37.1) | 48 (36.1) | 84 (12.4) | < 0.001 | 318 (24.2) |
| Dyspnoea | 143 (28.5) | 54 (40.6) | 354 (52.2) | < 0.001 | 551 (42.0) |
| Asthenia | 70 (14.0) | 35 (26.3) | 135 (19.9) | 0.0014 | 240 (18.3) |
| Chest pain | 27 (5.4) | 5 (3.8) | 20 (2.9) | 0.10 | 52 (4.0) |
| Lipothymia | 34 (6.8) | 2 (1.5) | 17 (2.5) | < 0.001 | 53 (4.0) |
| Syncope | 4 (0.8) | 0 | 1 (0.1) | 0.23 | 5 (0.4) |
| Other | 11 (2.2) | 2 (1.5) | 8 (1.2) | 0.39 | 21 (1.6) |
| Physical examination | |||||
| BMI (kg/m 2 ) | 26.6 ± 4.6 | 27.8 ± 5.7 | 27.04 ± 4.8 | 0.04 | 27.0 ± 4.9 |
| < 25 kg/m 2 | 203 (40.0) | 38 (28.8) | 244 (35.6) | 0.10 | 485 (36.6) |
| 25–30 kg/m 2 | 199 (39.3) | 60 (45.5) | 270 (39.4) | 529 (39.9) | |
| > 30 kg/m 2 | 105 (20.7) | 34 (25.8) | 172 (25.1) | 311 (23.5) | |
| Systolic BP (mmHg) | 136.8 ± 16.3) | 138.5 ± 17.5 | 134.0 ± 15.4 | 0.001 | 135.5 ± 16.0 |
| Diastolic BP (mmHg) | 77.4 ± 9.2 | 79.5 ± 9.7 | 77.3 ± 8.8 | 0.03 | 77.6 ± 9.0 |
| ECG characteristics | |||||
| Sinus rhythm | 423 (85.3) | 59 (45.0) | 0 | 482 (37.0) | |
| AF | 38 (7.7) | 65 (49.6) | 637 (94.1) | 740 (56.7) | |
| Atrial flutter | 5 (1.0) | 3 (2.3) | 20 (3.0) | 28 (2.1) | |
| Heart rate (bpm) | 66.4 ± 14.1 | 76.8 ± 20.6 | 74.5 ± 13.6 | < 0.001 | 71.6 ± 15.2 |
| QT (ms) | 380.7 ± 108.1 | 347.4 ± 136.1 | 346.0 ± 122.1 | < 0.001 | 360.3 ± 119.4 |
| LVH | 52 ± 10.9 | 12 ± 9.4 | 87 ± 13.8 | 0.21 | 151 ± 12.2 |
| Sokoloff index | 22.7 ± 8.8 | 23.1 ± 9.8 | 23.6 ± 9.5 | 0.32 | 23.2 ± 9.3 |
| Negative T wave | 94 (19.3) | 27 (21.6) | 223 (34.5) | < 0.001 | 344 (27.3) |
| Echocardiography | 386 (76.0) | 109 (82.0) | 519 (75.2) | < 0.001 | 1014 (76.3) |
| Mean time since echocardiography, days | 99.31 ± 110.32 | 93.86 ± 97.83 | 89.54 ± 104.21 | 0.40 | 93.71 ± 105.92 |
| Diameter of left atrium, mm (mean) | 40.99 ± 7.31 | 43.19 ± 7.47 | 47.81 ± 8.72 | < 0.001 | 44.72 ± 8.69 |
| Surface area of left atrium, cm 2 (mean) | 23.33 ± 7.87 | 25.56 ± 8.25 | 30.27 ± 10.42 | < 0.001 | 27.15 ± 9.87 |
| LVEF, % (mean) | 62.81 ± 11.08 | 60.74 ± 11.50 | 58.02 ± 12.45 | < 0.001 | 60.12 ± 12.05 |
| Shortening fraction, % (mean) | 36.37 ± 9.06 | 34.86 ± 8.14 | 33.14 ± 9.37 | < 0.001 | 34.55 ± 9.24 |
| Biological examination | |||||
| Creatininaemia (mean), μmol/L | 98.38 ± 38.93 | 96.15 ± 31.37 | 101.25 ± 43.34 | 0.40 | 99.65 ± 40.64 |
| Kalaemia, meq/L | 4.30 ± 0.40 | 4.28 ± 0.48 | 4.37 ± 0.44 | 0.04 | 4.33 ± 0.43 |
| INR (mean) | 2.33 ± 0.64 | 2.51 ± 0.79 | 2.53 ± 0.61 | < 0.001 | 2.47 ± 0.65 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
