Summary
Background
General practitioners (GPs) play a pivotal role in the long-term management of patients with atrial fibrillation (AF), including anticoagulant prophylaxis for stroke prevention.
Aims
To investigate the antithrombotic prescription behaviours of GPs in France and compare them with the European Society of Cardiology (ESC) guidelines for stroke prevention, and to identify the major determinants of use of antithrombotic therapy.
Methods
We conducted a cross-sectional survey, using data from the French Longitudinal Patient Database, on the use of antithrombotic treatments for stroke prevention in 15,623 patients (≥ 18 years of age) with AF who attended at least one GP consultation between July 2010 and June 2011. Data were collected on patient baseline characteristics, stroke risk factors, and prescription of antithrombotic drugs.
Results
The mean age was 74.6 ± 11.1 years, 59.5% were men, and 83.1% had a CHADS 2 score ≥ 1. Over half (52.6%) of the patients with a CHADS 2 score ≥ 1 received a vitamin K antagonist (alone or in combination with an antiplatelet), 19.3% received aspirin alone, and 23.4% received no antithrombotic therapy; 3.6% of the patients received dual antiplatelet therapy and 1.1% clopidogrel alone. Over half of the patients (56.3%) were treated in accordance with the ESC 2010 guidelines. Of the remaining patients, 19.4% received no treatment, 13.0% were inadequately treated, and 11.2% were over-treated. Factors associated with antithrombotic treatment were anti-arrhythmic therapy, higher stroke risk, presence of atherothrombotic disease, and use of non-steroidal anti-inflammatory drugs. Female gender was associated with a lower likelihood of antithrombotic treatment.
Conclusions
In this large French study, approximately 45% of thromboembolic high-risk patients were either not treated or inadequately treated. Better compliance with evidence-based guidelines is needed to reduce the burden of stroke in the AF population.
Résumé
Contexte
Les médecins généralistes jouent un rôle clé dans la prise en charge au long cours des patients présentant une fibrillation atriale (FA), notamment dans la gestion des traitements antithrombotiques.
Méthodes
Nous avons conduit une étude épidémiologique, à partir d’une base de données française avec suivi longitudinal, centrée sur la stratégie thérapeutique pour 15 623 patients (âgés ≥ 18 ans) présentant une FA, et bénéficiant d’au moins une consultation en médecine générale entre juillet 2010 et juin 2011. Les données recueillies concernaient les caractéristiques cliniques des patients au début de la période d’étude, les facteurs de risque d’accident vasculaire cérébral (AVC), et les prescriptions d’antithrombotiques.
Résultats
L’âge moyen était de 74,6 ± 11,1 ans, avec 59,5 % d’hommes, 83,1 % des patients ayant un score CHADS 2 ≥ 1. Plus de la moitié (52,6 %) des patients avec un score CHADS 2 ≥ 1 bénéficiait d’un traitement par un antivitaminique K (seul ou associé à un antiagrégant), 19,3 % étaient sous aspirine en monothérapie, et 23,4 % ne recevaient aucun traitement antithrombotique ; 3,6 % des patients étaient sous bithérapie antiagrégante plaquettaire, 1,1 % sous clopidogrel seul. Une faible majorité des patients (56,3 %) étaient traités selon les recommandations de 2010 de l’ESC. Pour les autres patients, 19,4 % ne recevaient aucun traitement, 13,0 % étaient insuffisamment traités, et 11,2 % étaient « surtraités » par rapport aux recommandations. Les facteurs associés à la prescription d’anticoagulants étaient la prescription d’anti-arythmiques, un risqué élevé d’AVC, la présence d’une atteinte athérothrombotique et l’emploi d’anti-inflammatoire non stéroïdien. Le sexe féminin était associé à une moindre prescription d’anticoagulants.
Conclusions
Au sein de cette large cohorte française, environ 45 % des patients en FA à haut risque thromboembolique ne sont, soit pas traités, soit traités sans respect des recommandations. Une meilleure application des recommandations sur la prise en charge de la FA semble indispensable pour réduire les accidents ischémiques cérébraux au sein de cette population.
Introduction
Atrial fibrillation (AF) is the most common cardiac arrhythmia and is frequently encountered in the primary care setting. The lifetime risk for developing AF or atrial flutter at 40 years of age is estimated to be 26% for men and 23% for women . AF affects between 600,000 and 1 million patients in France . Given the ageing population and the increasing prevalence of risk factors for AF, including hypertension, heart failure, older age, diabetes mellitus and vascular disease , the burden of AF in western populations is predicted to rise to epidemic proportions by 2050 .
Patients with AF are at increased risk of thrombotic events, including stroke and systemic embolism. Evidence-based guidelines recommend individualized risk stratification with validated tools and the use of oral anticoagulant (OAC) therapy for those identified to be at moderate or high risk of stroke . For more than six decades, the only oral anticoagulants available for long-term use have been the vitamin K antagonists (VKAs; e.g. warfarin) . While effective in preventing thromboembolism, VKAs have several drawbacks, including food and drug interactions, a narrow therapeutic range, and wide variations in response to treatment that necessitate close laboratory monitoring and frequent dose adjustments . Anticoagulant therapy also carries a risk of bleeding, with potentially catastrophic consequences in the event of intracranial haemorrhage . In an analysis from the United States, 65% of emergency hospitalizations (between 2007 and 2009) in older adults were due to unintentional drug overdoses, one-third of which were associated with warfarin therapy . Consequently, observational studies have shown that a large percentage of eligible patients at risk of stroke do not receive anticoagulant therapy . Conversely, a sizable proportion of AF patients at low risk receive inappropriate treatment with an oral anticoagulant, placing them at unnecessary risk of a bleed .
In France, anticoagulant prophylaxis for patients with AF is managed largely in the primary care setting. General practitioners (GPs) play a pivotal role in the long-term management of these patients. Adherence to practice guidelines for stroke prevention and effective collaboration between GPs and cardiologists, or other specialists, are therefore essential. In this analysis, we investigated the antithrombotic prescription behaviours of GPs in France and compared them with the recommendations in the 2010 and 2012 European Society of Cardiology (ESC) guidelines for stroke prevention in AF . We also identified the major determinants of use of antithrombotic therapy.
Methods
Using data from the French Longitudinal Patient Database (LPD; Cegedim Strategic Data, France), a medical records and prescriptions database , we conducted a cross-sectional survey on the use of antithrombotic treatments for stroke prevention in patients with AF. Since 1994, the LPD has collected anonymized data from more than 1.6 million patients though a computerized network of 1200 office-based GPs. The panel of physicians is representative of French GPs in terms of their age, sex, and geography. GPs do not receive direct compensation for participating in the database. The Commission Nationale de l’Informatique et des Libertés approved the use of the LPD data for analysis and patient informed consent was not required.
Study population and data extracted
The study population comprised adults (≥ 18 years of age) with a diagnosis of AF who attended at least one GP consultation between 1 July 2010 and 30 June 2011. During this period, data were collected on patient baseline characteristics, risk factors for stroke (i.e. congestive heart failure, left ventricular dysfunction, hypertension, diabetes mellitus, stroke, transient ischaemic attack, systemic embolism, vascular pathology), and prescription of VKAs and antiplatelet drugs.
Evaluation of stroke risk
Each patient’s risk of stroke was evaluated retrospectively using the CHADS 2 (congestive heart failure, hypertension, age > 75 years, diabetes mellitus, and prior stroke or transient ischaemic attack [doubled]) criteria. A CHADS 2 score of 0 was taken as indicative of low risk, 1 as moderate risk, and ≥ 2 as high risk. In addition, the patients’ CHA 2 DS 2 -VASc (cardiac failure or dysfunction, hypertension, age ≥ 75 [doubled], diabetes, stroke [doubled]-vascular disease, age 65–74, and sex category [female]) scores were calculated, with the caveat that this score was included in the 2012 ESC guidelines , and was not therefore incorporated into normal practice at the time the data were collected. Treatment with antithrombotic drugs for stroke prevention in AF (VKAs and antiplatelets) was evaluated according to the 2010/2012 ESC guideline recommendations . Inadequate treatment was defined as the use of aspirin instead of OAC in moderate- to high-risk patients and over treatment was defined as the use of OAC instead of aspirin in low-risk patients.
Statistical analysis
Continuous variables are expressed as mean ± standard deviation (SD) and categorical variables as frequency (percentage). The choice of treatment with VKA alone, aspirin alone, or VKA plus aspirin was studied in a binary manner (treatment vs. no treatment). Potential determinants of treatment (vs. no treatment) were identified from patient baseline and disease characteristics, medical history, and concomitant treatments. The analysis was performed using the GLIMMIX (SAS 9.2, SAS Institute Inc., Cary, NC, USA) procedure for multivariable analysis, with the variable “physician” taken as a random effect. Factors associated with treatment (vs. no treatment) that were significant at the 20% threshold in univariate analysis were included in a backwards elimination multivariable logistic regression model; factors significant at the 0.001 level were retained in the final model. All pairwise interactions were tested. CHADS 2 or CHA 2 DS 2 -VASc scores were introduced into the final models to determine whether they were major determinants of antithrombotic therapy use. Results are presented with odds ratios (ORs) and their 95% confidence intervals (CIs).
Results
Study population
A total of 15,623 patients with AF were identified. The mean age was 74.6 ± 11.1 years, 59.5% were men, and 5.1% had valvular disease ( Table 1 ). The overall mean body mass index (BMI) was 27.9 ± 5.3 kg/m 2 ; 27.9% of the patients had a BMI between 20 and 25 kg/m 2 and 4.1% had a BMI below 20 kg/m 2 . Of the patients in whom the type of AF was known, 77.4% had paroxysmal AF, 2.7% had persistent AF and 20.0% had permanent AF. The mean CHADS 2 score was 1.5 ± 1.1 and 83.1% had a CHADS 2 score ≥ 1. The mean CHA 2 DS 2 -VASc score was 2.9 ± 1.5 and 93.6% had a CHA 2 DS 2 -VASc score ≥ 1.
| Variable | All patients ( n = 15,623) | No antithrombotic therapy ( n = 3924) | Antithrombotic therapy ( n = 11,699) |
|---|---|---|---|
| Men | 9291 (59.5) | 2115 (53.9) | 7176 (61.3) |
| Age (years) | 74.6 ± 11.1 | 74.2 ± 13.1 | 74.7 ± 10.4 |
| Age ≥ 75 (years) | 8981 (57.5) | 2265 (57.7) | 6716 (57.4) |
| Body mass index (kg/m 2 ) a | 27.9 ± 5.3 | 26.9 ± 5.0 | 28.1 ± 5.4 |
| Systolic blood pressure (mmHg) b | 133 ± 15 | 133 ± 16 | 133 ± 15 |
| Diastolic blood pressure (mmHg) c | 76 ± 9 | 76 ± 10 | 76 ± 9 |
| Medical history | |||
| Arterial hypertension | 9246 (59.2) | 1934 (49.3) | 7312 (62.5) |
| Diabetes mellitus | 2683 (17.2) | 513 (13.1) | 2170 (18.5) |
| Heart failure | 1775 (11.4) | 347 (8.8) | 1428 (12.2) |
| Stroke | 763 (4.9) | 157 (4.0) | 606 (5.2) |
| TIA | 502 (3.2) | 82 (2.1) | 420 (3.6) |
| Systemic embolism | 97 (0.6) | 8 (0.2) | 89 (0.8) |
| Myocardial infarction | 665 (4.3) | 108 (2.8) | 557 (4.8) |
| Prosthetic heart valve | 309 (2.0) | 54 (1.4) | 255 (2.2) |
| Neurological disorders | 422 (2.7) | 134 (3.4) | 288 (2.5) |
| PAD | 827 (5.3) | 138 (3.5) | 689 (5.9) |
| Valvular disease | 795 (5.1) | 186 (4.7) | 609 (5.2) |
| Type of AF d | |||
| Paroxysmal | 1891 (77.4) | 513 (84.0) | 1378 (75.2) |
| Permanent | 488 (20.0) | 88 (14.4) | 400 (21.8) |
| Persistent | 65 (2.7) | 10 (1.6) | 55 (3.0) |
| Concomitant drugs | |||
| Non-steroidal anti-inflammatory drug | 7492 (48.0) | 1305 (33.3) | 6187 (52.9) |
| Anti-arrhythmic drug | 7425 (47.5) | 834 (21.3) | 6591 (56.3) |
| Injectable anticoagulants (e.g. heparin) | 482 (3.1) | 83 (2.1) | 399 (3.4) |
| Risk scores for stroke | |||
| CHADS 2 score (%) | |||
| 0 | 2638 (16.9) | 887 (22.6) | 1751 (15.0) |
| 1 | 5026 (32.2) | 1342 (34.2) | 3684 (31.5) |
| ≥ 2 | 7959 (50.9) | 1695 (43.2) | 6264 (53.5) |
| CHA 2 DS 2 -VASc score (%) | |||
| 0 | 998 (6.4) | 399 (10.2) | 599 (5.1) |
| 1 | 1774 (11.4) | 499 (12.7) | 1275 (10.9) |
| ≥ 2 | 12,851 (82.3) | 3026 (77.1) | 9825 (84.0) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
