Napovedniki antitrombotične terapije pri ambulantnih bolnikih s kronično atrijsko fibrilacijo, ki odstopa od smernic zdravljenja Predictors for guideline-nonadherence with antithrombotic therapy in outpatients with chronic atrial fibrillation Avtor / Author Ustanova / Institute Reinhild Bücheler12, Christoph Meisner3, Ruth Bösel3, Richard Fux1, Christine Jägle1, Lusine Danielyan1, Katrin Schwarz1, Christoph H. Gleiter1,4, Klaus Mörike1 ^University Hospital Tübingen, Institute of Pharmacology and Toxicology, Department of Clinical Pharmacology, Tübingen, Germany; 2Medical Service of the Public Health Assurance (MDK) in Baden-Württemberg, Lahr, Germany; ^University Hospital Tübingen, Institute of Medical Information Processing, Tübingen, Germany; 4CenTrial GmbH, Tübingen, Germany Ključne besede: atrijska fibrilacija, antitrombotično zdravljenje, smernice Key words: Atrial fibrillation, antithrombotic, guideline Članek prispel / Received 12.03.2009 Članek sprejet / Accepted 06.05.2009 Naslov za dopisovanje / Correspondence Prof. Christoph H. Gleiter, MD University Hospital Tübingen Department of Clinical Pharmacology Otfried-Müller-Strasse 45 D-72076 Tübingen, Germany Tel. +49 7071 29 78277 Fax +49 7071 29 5035 E-mail: christoph.gleiter@med.uni-tuebingen.de Izvleček Namen: Razlogi za neustreznost antitrombotičnega zdravljenja pri bolnikih z atrijsko fibrilacijo (AF) so dobro znani. Naši cilj je bil določiti kvaliteto antitrombotičnega zdravljenja pri skupini bolnikov s kronično AF in določiti omejitve ustreznosti antitrombotičnega zdravljenja. Metode: Z multicentričnimi raziskavami smo analizirali podatke o zdravljenju, dejavnikih tveganja za možgansko kap in psihosocialne dejavnike. S pomočjo logistične regresije smo ugotavljali napovedne dejavnike za neustreznost antitrombotičnega zdravljenja, ki odstopa od ustaljenih smernic pri ambulantnih bolnikih s kronično atrijsko fibrilacijo Abstract Purpose: Our aims were to determine the prevalence of physicians' adherence to antithrombotic guidelines in the management of outpatients with chronic atrial fibrillation (AF) and to identify risk factors for nonad-herence to treatment guidelines. Methods: Data on drug treatment, stroke risk factors and psychosocial variables were analyzed descriptively in a cross-sectional community-based multicentre study. Predictors for nonadherence with guidelines for antithrombotic prophylaxis of AF patients were identified using logistic regression. Results: Of 413 outpatients with chronic AF attending the offices of family physicians, 184 (44.6 Rezultati: Od 413 ambulantnih bolnikov s kronično AF, ki so bili napoteni iz ambulant družinskih zdravnikov, je bilo 184 (44,6%) zdravljenih po priporo~enih smernicah (American College of Chest Physicians - ACCP 2001). V skupini bolnikov z visokim tveganjem za možgansko kap (n=387), je bilo 178 bolnikov (46,0%) zdravljenih po smernicah. 31 bolnikov z visokim tveganjem je prejelo ku-marinsko terapijo, ~eprav je bila pri njih prisotna vsaj ena kontraindikacija. Multivariantna analiza je pokazala, da je prisotnost absolutne kontraindikacije za kumarin neodvisni napovednik za zdravljenje, ki odstopa od smernic. Zaključek: Za izboljšanje priporo~enih smernic anti-trombotičnega zdravljenja ambulatnega bolnika z AF, je dejavnike tveganja potrebno upo'tevati za vsakega bolnika posebej. were treated according to the American College of Chest Physicians (ACCP) 2001 guidelines. In the group of patients with a high risk of stroke (n=387), 178 patients (46.0%) received guideline-adherent treatment. 31 of the high-risk patients received coumarin although they had at least one contraindication. Multivariate analysis showed the presence of absolute contraindications to coumarins to be an independent predictor of guideline-nonadherent treatment. Conclusion: To improve guideline-adherence in the antithrombotic treatment in AF outpatients, strategies involving individual assessment of the risks and benefits will need to be established and implemented. INTRODUCTION Atrial fibrillation (AF) is associated with a substantial increase in morbidity and mortality, with stroke being the most serious complication. To prevent thromboembolic events, adjusted-dose treatment with coumarin-type oral anticoagulants (e.g., warfarin or phenprocoumon) is the mainstay of anti-thrombotic treatment in AF patients with moderate to high risk of stroke. Aspirin is reserved for patients at lower stroke risk or for higher-risk patients with contraindications to coumarins (1). Although current treatment guidelines recommend adjusted-dose oral anticoagulants for the majority of AF patients (2-5) and health outcome studies support the effectiveness of this strategy (6-10), observational studies show that anticoagulant therapy is still substantially underused or inappropriately used thus imposing preventable risks of thromboembo-lism on AF patients (11-15). Predictors of guideline nonadherence, however, are largely unknown. We analyzed the antithrombotic therapy of a cohort of AF outpatients and assessed individual stroke risk, potential contraindications to cou- marins and aspirin, and the appropriateness of stroke prophylaxis according to the 2001 guidelines of the American College of Chest Physicians (ACCP) (2) and identified predictors of guideline nonadherence. MATERIAL AND METHODS Setting and study design Full details of the method have been described in a previous article (16). In brief, all family physicians in the study regions were invited to participate as study centers. This target group included all primary-care doctors and specialists in internal medicine with a family medicine focus registered with the regional Physicians in Public Health Insurance Board (Kassenärztliche Vereinigung). The study region included districts within the state of Baden-Württemberg, Southern Germany, with urban (Tübingen, Reutlingen, Freiburg, Offenburg) and surrounding rural areas. Participating physicians submitted a list of their patients diagnosed with chronic AF, i.e. an ICD-10 code of I48 or I49.8, to the Division of Clinical Pharmacology at the University Hospital Tübingen (the steering center). If a physician reported having more than 10 AF patients, the steering center randomly selected 10 patients from among them. The limit of 10 patients was set to avoid over-representation of single offices and their policies in AF management. Patients from 18 to 85 years were included if chronic non-valvular AF was diagnosed. Chronic AF included recurrent (intermittent) AF, defined as two or more episodes of AF, or permanent AF (17). A physician at the steering center re-analyzed a recent electrocardiogram (ECG) from each potential study patient to confirm the diagnosis of AF. The exclusion criteria were another condition requiring oral anticoagulation (e.g., pulmonary embolism, mitral stenosis, prosthetic heart valve); being scheduled for cardioversion, electroablation or cardiac surgery in the next 4 weeks; having a life expectancy of less than 1 year; or being unable to give informed written consent for study participation. The ethical committees of the Medical Faculty of the University of Tübingen and of the Physicians' Chamber (Landesärztekammer) Baden-Württemberg approved the study protocol. The authors certify that all applicable institutional and governmental regulations covering the ethical use of human volunteers were followed during this research. Target variables The primary target variable was the percentage of enrolled patients whose antithrombotic treatment was compatible with the ACCP 2001 recommendations (2,18) (Figure 1). These guidelines were used because they were state of the art when the study was initiated in 2003. On two occasions physicians at the steering center assessed whether each patient's management adhered to the guidelines by comparing data from the case report form (CRF) with the text of the reference ACCP guidelines. In addition, a computerized decision algorithm was used (Figure 1). If the physicians and computer assessment results differed, guideline adherence was re-assessed by the steering center. Cases where there was no detectable difference between the care being offered and the recommendations of the guidelines were judged to be compliant with guidelines. Secondary target variables included the following quality indicators: percentage of high-risk patients receiving a coumarin anticoagulant; anticoagulation level within the target INR range of 2.0-3.0; and whether the results of an echocardiogram were made available. To find predictors of inappropriate antithrombotic treatment, 47 potential variables (including physicians' specialization and patients' demographic characteristics, medical history, concomitant medication, ischaemic and bleedings risks, and social data) were selected from the findings of previous observational studies (see sections B and C of Table IV for the most important variables). Univariate analysis of the data was performed. Factors found to have a predictive potential by univariate analysis (p<0.1) were selected to undergo multivariate analysis. Variables predicting a high risk that AF patients would receive inappropriate antithrombotic treatment were identified by the multivariate analysis. Data acquisition Participating physicians completed a CRF with 124 variables for each patient. The variables included current health status, medication, various throm-boembolic and haemorrhagic risk factors, and psy-chosocial variables, such as mental and physical activity, family situation and compliance. Generally accepted contraindications to anti-thrombotic agents are listed in Table I. The table ÜH < -Ü u ■■13 S J\ ja M o ein u U < -c 13 Jd O ■u •iH (S v 0 M 0 ^ JO v -T3 rt r^ o ^z o " ni e3 -73 v -T3 * <ü > g -la o ^ s o ^ -T3 « o C O ^ . ^ JS il ^ iS -a "S C o ^ o o g ^ (rt ,J3 ib ^ ■jd -a rt "3 iT M " C rt -73 o ^ W -fi y M .Jd oo (rt d O -O (rt * Table I: Presence of contraindications to coumarins or aspirin in 413 patients with chronic AF The contraindications are derived from the summary of product characteristics. Some patients had more than one contraindication. Contraindication N Absolute contiaindications to coumaiins. The presence of at least one of the following variables precludes the use of a coumarin. Severe haematoma after oral anticoagulation Current faecal or urinary microbleeding Other bleeding episodes after oral anticoagulation requiring medical intervention Chronic use of an NSAID Hepatic disease and alcohol abuse Vascular malformation posing a bleeding risk History of proliferative diabetic retinopathy History of intracranial hemorrhage or recent CNS surgery Gastrointestinal or genitourinary bleeding during the preceding 6 months Hypersensitivity to or intolerance of coumarins Active peptic ulcer Thrombocytopenia (<100,000 Systolic blood pressure >180 mm Hg or diastolic blood pressure >95 mm Hg Severe renal dysfunction (serum creatinine >3.0 mg/dl) 0 Total of absolute contraindications to coumarins 81 Relative contiaindications to coumaiins. If one of the following variables is present, oral anticoagulation may be withheld in AF patients at an intermediate or low risk ( stroke, as defined by the ACCP 2001 guidelines (2). of Alcohol abuse (without hepatic disease) 24 Hepatic disease (without alcohol abuse) 20 Dementia 23 Poor patient compliance (as indicated by the physician) 17 Falls in the preceding 12 months 16 Decline of anticoagulant therapy by the patient 8 Total of relative contraindications to coumarins 116 Absolute contiaindications to aspirin Hypersensitivity to or intolerance of aspirin or NSAIDs 4 Active peptic ulcer 2 Haemorrhagic diathesis or thrombocytopenia (<100,000 ,ul-1) 2 Total of absolute contraindications to aspirin 8 reflects the exclusion criteria used in the SPAF I-III and SPINAF trials (19-23) and in the summaries of the product characteristics (24) of phenprocoumon as indicated in the written product material provided by the manufacturers (e.g., Marcumar®). Anticoagulation intensity was assessed using the results of the most recent test of the international normalized ratio (INR) by the participating physician. A specialist in internal medicine at the steering center assessed left ventricular function using clinical and echocardiographic documents, if available. Table IV: Factors with the potential to predict guideline nonadherence in antithrombotic management (from 387 chronic AF outpatients at a high risk of stroke). The relative risk (risk ratio, RR, with confidence intervals, CI) for factors predicting guideline nonadherence and the p values were estimated using univariate analysis. For factors found to have no predictive value (p>0.1), no RR is given (section C). Factors found to have a predictive potential (p<0.1; section B) were selected to undergo multivariate analysis. One of these factors, namely, "having an absolute contraindication to coumarins", was found to be a predictor in multivariate analysis (section A). Factor RR (95%-CI) A) Piedicting nonadheience by multivaiiate analysis (adjusted RR) Having an absolute contraindication to coumarins (Table I) 51.73 (6.82-392.58) B) Predicting nonadherence by univariate analysis: p<0.1 Treatment by a general practitioner 1.23 (0.98-1.54) Having no echocardiogram performed 1.32 (1.08-1.60) Having diabetes mellitus 1.22 (1.01-1.47) History of non-life-threatening bleeding 1.59 (1.31-1.93) Having a relative contraindication to coumarins (Table I) 1.40 (1.15-1.69) History of falls 1.33 (1.07-1.64) Needing assistance to see the doctor 1.37 (1.11-1.69) Barthel Index* score <95 1.29 (1.03-1.62) Having at least one absolute contraindication to coumarin use (Table I) 2.14 (1.89-2.42) C) Variables not predicting nonadherence: p>0.1 in univariate analysis Male sex Age >75 years Body mass index AF duration >5 years Permanent or intermittent AF Regular daily use of >5 drugs History of heart failure History of hypertension Presenting with systolic blood pressure >140 mm Hg and/or diastolic blood pressure >90 mm Hg History of heart valve disease History of stroke or thromboembolism History of hyperthyroidism History of coronary heart disease including myocardial infarction, stable angina, bypass surgery or angioplasty Alcohol abuse Vascular malformations History of cerebral or other severe haemorrhage HbA1C >6.5% in diabetic patients Fasting blood glucose >130 mg/dL * The Barthel Index reflects functional abilities in daily life on a scale of scores ranging from 0 to 100. It includes variables such as ambulation, stair climbing, transfers, personal hygiene, feeding, excretion and dressing. A score of 95 was chosen as a cut-off for a reduced Barthel Index. Upon submission of a CRF to the steering center, completeness and plausibility were checked and any issues were clarified with the study centers. Data were entered into an electronic database on two separate occasions by different investigators (double entry). Relative contraindications include dementia, poor compliance, and alcohol abuse without hepatic disease. The clinical significance of these relative contraindications to coumarins is highly variable among patients and can be assessed only on an individual basis. We therefore accepted any antithrombotic treatment for these patients as being guideline-adherent. Data analyses The statistical analysis involved all study patients and was made for descriptive purposes. Continuous variables are expressed as mean values +/- standard deviations or as median and quantiles, depending on their distribution. Discrete variables are expressed as counts and percentages. For identifying barriers to guideline adherence, univariate Mantel-Haenszel statistical analyses, presenting p-values and relative risks for potential barrier factors, were performed. These factors were chosen a priori (Table IV) and were believed to have potential effects on the rate of guideline adherence. Next, a multivariate logistic regression for the primary outcome was performed. The model includes all variables with missing values in the CRFs of less than 10% of patients and includes a minimum of 5% of patients remaining in the risk group (risk for nonadherence) of the respective variable . The final model was the result of a stepwise backward procedure based on the full model, which included every barrier factor with an entry level of p<0.1. The results of the logistic regression are presented with p-values and odds ratios with 95% confidence intervals. The SAS software package version 8.0 for Windows (SAS Institute, Cary, NC) was used for the statistical analysis. Contraindications to antithrombotic drugs Absolute contraindications to coumarins or aspirin (Table I), such as an active peptic ulcer or chronic use of non-steroidal anti-inflammatory drugs, justified withholding anticoagulation treatment with a coumarin. RESULTS 1022 patients with chronic AF were reported to the steering center by a total of 94 local study centers (Figure 2) between July 2001 and June 2003. From these 1022 patients, the steering center randomly selected 510 patients. From this group, 97 patients were excluded for various reasons (details in Figure 2), with the absence of confirmation of AF (43 patients) and the presence of other conditions requiring anticoagulant therapy Table II: Patient characteristics (N=413) Number (%) Demogiaphic chaiacteiistics Age <65 years 66 (16.0) Age >75 years 182 (44.1) Male gender 231 (55.9) Clinical characteristics Type of chronic AF: -recurrent (intermittent) AF -chronic AF 62 (15.0) 351 (85.0) History of hypertension 281 (68.0) Congestive heart failure functional class NYHA II-IV 185 (44.8) Reduced systolic left ventricular function in EC (performed in 179 patients) 81 (45.3) Diabetes mellitus 126 (30.5) Coronary heart disease History of myocardial infarction History of coronary artery bypass grafting surgery or angioplasty 98 (23.7) 41 (9.9) 26 (6.3) History of ischaemic stroke or TIA 92 (22.3) Valvular heart disease 67 (16.2) History of non-cerebral embolism (pulmonary embolism, deep venous thrombosis, peripheral arterial embolism) 60 (14.5) History of hyperthyroidism 22 (5.3) Thromboembolism risk groups according to ACCP 2001 guidelines (2) High risk 387 (93.7) Intermediate risk 14 (3.4) Low risk 12 (2.9) Coumarin treatment 302 (73.1) Psychosocial characteristics Alcohol abuse or other addiction 24 (5.8) Dementia (as indicated by the family doctor) 23 (5.6) Barthel Index of activities of daily living Severely reduced (^70) Not impaired (>95) 14 (3.4) 361 (87.4) Living alone at home 104 (25.2) Living in institutionalized care (nursing home, old people's home, care at home by a welfare agency) 14 (3.4) Needing assistance to see the physician or requiring home visits 52 (12.6) (21 patients) being the most frequent reasons for exclusion. Eventually, 413 patients from 73 study centers were enrolled. years old. 44.1% of the patients were 76 to 85 years old (Table II). All patients were of Caucasian origin. 351 (85.0%) patients had a history of chronic AF. Patient characteristics The mean (+/- SD) age was 73.0 (8.1) years (median, 74.0 years). 347 (84.1%) patients were >65 387 (93.7%) of the enrolled patients had a high risk (i.e., >4% per year) of systemic thromboembolism (Figure 1, step 1). The most frequently encountered risk factors were a history of hypertension (68.0%), Table III: Indicators for the quality of antithrombotic therapy in 413 outpatients with chronic AF. Number (%) Guideline-adherent antithrombotic therapy given to All patients* (N=413) 184 (44.6) a) patients at high risk of stroke (N=387) 178 (46.0) b) patients at intermediate risk of stroke (N=14) 5 (35.7) c) patients at low risk of stroke (N=12) 1 (8.3) Other quality indicators ECG performed 179 (43.3) Oral anticoagulation with a coumarin 297 (71.9) INR of patients on a coumarin (N=297) INR in target range (2.0-3.0) INR <2.0 INR >3.0 no INR available INR test not older than 4 weeks 208 (70.0) 53 (17.8) 25 (8.4) 11 (3.7) 240 (80.8) Patients on a coumarin also receiving aspirin 2 (0.7) * This group includes 11 patients with unique combinations of clinical conditions: 9 had both risk factors for stroke and for haemorrhage and 2 patients with active peptic ulcer and thrombocytopenia (platelet count <100,000 ^l-1) had absolute contraindications to both aspirin and coumarins. Appropriateness of antithrombotic treatment in such patients cannot be assessed on the basis of the ACCP guidelines (2). congestive heart failure (44.8%), age >75 years (44.1%), diabetes mellitus (30.5%) and coronary heart disease (23.7%) (Table II). A history of hyperthyroidism was recorded in 22 (5.3%) patients. Patients receiving coumarin treatment Of the 413 patients analyzed, 297 (71.9%) were on coumarin treatment. Of patients at a high risk of stroke, 283 (73.1%) were treated with a coumarin. Among the 334 high-risk patients eligible for coumarin therapy, 248 (74.3%) actually did received it (Figure 3). Contraindications to antithrombotic drugs 104 (25.2%) patients had at least one contraindication to coumarin or aspirin, with 54 (13.1%) patients having at least one absolute contraindication to coumarins, even given a high risk of throm-boembolism . For these patients, the ACCP 2001 guidelines recommend the use of aspirin 325 mg daily. Only five patients had absolute contraindications to aspirin. Four of these patients had active peptic ulcers and thrombocytopenia (<100,000 ^l-1), respectively , and were not eligible for a coumarin as an alternative drug. As no recommended alternative treatments for these situations are available, the antithrombotic treatment of these patients was accepted as being appropriate. Overall guideline adherence 302 (73.1%) patients received a coumarin (Table II). Ten patients were prescribed a thienopyridine (clopidogrel or ticlopidine) with or without concomitant aspirin. One patient received enoxaparin. Two patients received a combination of aspirin plus phenprocoumon. 31 (7.5%) patients were on no an-tithrombotic agent at all. Only 184 (44.6%) patients were treated according to the recommendations of the 2001 ACCP guidelines (Table III). Of all 387 patients (93.7%) with a high Fi*^ret3: /inmhrc^rr^boticg eaud^entin the uuougrpaüer^tsatšigdiiuk ošutroi<£.. '^r^c^tn^^n^ (^s ^^cnmmenrdenythr oredl AC^^1 contraindication to coumarins, but not to aspirin 4 patients with contraindications to coumarins and to aspirin adequate coumarin therapy (INR 2.0-3.0) insufficient antithrombotic therapy excessive anti-thrombotic therapy no INR available INR <2.0 antiplatelet drug* 177 INR >3.0 aspirin + coumarin no anti-thrombotic therapy 46 66 23 2 7 (13.8%) (16.5%) (6.9%) (0.6%) (2.1%) aspirin 250-325 mg/d 24 (7.2%) aspirin <250 mg/d n 14 (28.6%) other (thieno-pyridine) 31 (63.3%) 3 (6.1% risk of stroke, only 178 (46.0%) were treated according to the guidelines (bold frame in Figure 3). cases: 23 coumarin patients had an INR >3.0 and 2 patients were on a coumarin and aspirin concomitantly. Guideline adherence in AF patients with a high risk of stroke Of the 387 high-stroke-risk patients, 334 were eligible for coumarin treatment. However, only 177 patients (53.0%) in this group both received it and were in the target INR range (2.0-3.0) (Figure 3). Guideline violations were found in 157 coumarin-eligible patients. Of these, 46 patients (13.8%) were on coumarin treatment with an INR <2.0, 55 patients (16.5%) received an antiplatelet drug or enoxaparin instead of a cou-marin, and 24 patients (7.5%) were on no antithrombotic therapy whatsoever. The INR was unavailable in 7 patients (2.1%). Excessive treatment was found in 25 Forty-nine high-stroke-risk patients had at least one absolute contraindication to coumarin treatment. The right therapy, namely, a correct dose of aspirin (250-350 mg/d), was prescribed to only one patient (bold frame in Figure 3). However, in the majority of cases (31 patients, 63.3%), patients received a coumarin despite their absolute contraindication and against the recommendations of the guidelines; furthermore, two of these patients had an INR >3.0. The dose of aspirin was too low in 14 patients (28.6%). Three patients were on a thienopyridine antiplatelet drug. Four patients who had a contraindication to both coumarins and aspirin nevertheless received a coumarin. 4 Guideline adherence in AF patients with an intermediate risk of stroke Fourteen patients had an intermediate risk of stroke (Table III). Of these, five patients received guideline-adherent treatment (4 coumarin, 1 aspirin). There were, therefore, nine guideline violations: two patients were on aspirin <250 mg/d , three patients were on a coumarin and had an INR <2.0 (in one case despite the presence of a contraindication to coumarin use) and three patients received no an-tithrombotic treatment. Guideline adherence in AF patients with a low risk of stroke Among the 12 low-stroke-risk cases (Table III), one patient was on a correct dose of aspirin. Of the other 11 cases, three were on aspirin <250 mg/d and four were on a coumarin while four received no anti-thrombotic therapy . Second-line quality indicators of antithrombotic management 297 patients from the total study population received a coumarin. 208 (70.0%) had recent INR test results within the target range of 2.0 to 3.0. However, 17.8% of the INR test results were subtherapeutic (INR <2.0) (Table III). Echocardiography had been performed in 179 (43.3%) patients. Predictors of inappropriate antithrombotic therapy in AF patients From 47 variables, nnine (Table IV, section B) were identified by univariate analysis (p<0.1) as potential predictors of inappropriate antithrombotic treatment in AF patients with a high risk of stroke. Following subsequent multivariate analysis of these nine variables, one turned out to be associated independently with nonadherence to the 2001 ACCP guidelines, namely, having one or more absolute contraindications to oral anticoagulant use (Table IV, section A). DISCUSSION We found that less than half of high-risk patients received adequate treatment. Underuse of anticoagulant treatment in patients with chronic AF has been investigated by several authors. They found rates of anticoagulation of 20% (25), 27% (13), 45% (26), 50.4% (12) or 23% to 31% (9), respectively, with the variability of the rates being due to the differing settings of the studies. In one survey, more than 90% of 312 office-based physicians reported to regularly prescribe a coumarin to AF patients (27). In view of these rates, our finding of 71.9% of AF patients on a coumarin seems to be favourable. However, further analysis showed that crude antico-agulation rate appears to be an insufficient indicator of quality. This is because in substantial numbers of coumarin patients therapy was either inadequate (INR out of the target range, INR unavailable or concomitant antiplatelet therapy) or contraindicat-ed. In fact, having a contraindication to coumarin therapy predicted guideline nonadherence. This overuse of coumarins is the major finding that the present study adds to current knowledge. A limitation of our study may be a potential selection bias. Physicians who chose to participate may be more aware of guidelines than others who declined, and overall guideline nonadherence rates may therefore have been underestimated. Guideline adherence was analysed against the ACCP guidelines of 2001 because they were the most recent ones at the time of use. The ACCP guidelines were updated in 2004 (3) and new guidelines for the management of patients with AF were also published by the American College of Cardiology (ACC), American Heart Association (AHA) and European Society of Cardiology (ESC) in 2006 (29). Using these two newer guidelines, 45.5% (ACCP) (3) or 54.2% (ACC/AHA/ESC) (29), respectively, of our patients received guideline-adherent antithrombotic treatment. This is not fundamentally different from our finding using the 2001 ACCP guidelines. Additionally, the ACCP published new guidelines in 2008; essentially, they recommend aspirin at a dose of 75 to 325 mg/d for low-risk patients (30). In conclusion, efforts should be directed to improving guideline implementation. Improved communication of guidelines will be needed. On a physician-patient level, a management algorithm may be help clinicians select appropriate individualized antithrombot-ic treatment. Currently, the variety of types of software used in physicians' offices makes implementing an electronic decision support system, including alert signals, difficult. The present data illustrate that every AF patient needs individual assessment of the risk of stroke and risk of bleeding. This analysis is complex and many factors need to be taken into account. Funding: Bundesministerium für Bildung und Forschung (Federal Ministry of Education and Research, BMBF) (FK 01 EC 001), Berlin, Germany. A IWLEDGEMENTS The cooperation of the participating physicians is gratefully acknowledged (alphabetical order): E. and F. Ailinger (Lichtenstein), S. Albert (Offenburg), M. Bach (Eningen), T. Batz (Freiburg), T. Bausch (Freiburg), G. Bihlmaier (Römerstein), B. Bohl (Freiburg), W. Bösch (Horb), R. Burr (Hir-rlingen), W. Deissler (Freiburg), M. Dinkel (Reutlingen), R. Dorff (Freiburg), H.-M. Dürr (Eningen), T. Eisele (Waldkirch), W. Fengels (Simonswald), F. Feurer (Reutlingen), E. Fiechtner (Rottenburg), U. Föhrenbacher (Freiburg), H. Fricke (Freiburg), A. Gammel (Mössingen), J. Geldmacher (Emmendingen), G. Genz (Freiburg), U. Gessner (Waldkirch), B. Graf (Freiburg), G. Gregor (Tübingen), B. Greiner (Reutlingen), M. Greuter (Offenburg), D. Grünholz (Titisee-Neustadt), U.-F. Gundel (Reutlingen), A. and J. Häcker (Waldenbuch), G. Haffner (Reutlingen), R. Hagemann (Freiburg), M. Hagner (Rottenburg), P. Harosky (Freiburg), H. Hartmann (Wyhl), A. Heinemann (Reutlingen), U. Helber (Rottenburg), W. Helm (Freiamt), D. Henniges (Freiburg), M. Hitz-Bergau (Freiburg), T. Horstmann (Denzlingen), W. Hüther (Freiburg), E. Kapp (Tübingen), W.D. Kilchling (Freiburg), H. Kister (St. Johann), I. Kleiber-Greuter (Offenburg), R.C. Knöll (Walddorfhäslach), E. Krause (Freiburg), B. Kühnert (Freiburg), E. Mauthe (Freiburg), I. Lang-Mergner (Dusslingen), H. Lenzer (Freiburg), M. Ludwig (Freiamt), J. Mehrer (Emmendingen), T. Meyer (Endingen), C. Mohrmann (Reutlingen), M. Musch (Freiburg), W. Niebling (Titisee-Neustadt), B. and C. Nübel (Reutlingen), H. Pech (Pfullingen), U. Peuckert (Freiburg), A. Podmaniczky (Engstingen), H. Prautzsch (Tro-chtelfingen), A. Rager (Rottenburg), P. ReetzeBonorden (Freiburg), H. Renner (Tübingen), K. Rosset (Freiburg), K. Roth (Freiamt), M.G. Schef-fczyk (Freiburg), C. Schmidt (Waldkirch), C. Schmitthenner (Freiburg), M. Schöll (Offenburg), A. and K.-H. Schönleber (Hülben), P. Schröder (Freiburg), W. Schulz-Weiling (Freiburg), K. Secker (Reutlingen), T. Seyfferth (Reutlingen), M. Simon (Dusslingen), M. Stass (Freiburg), H. Straub (Wendlingen), K. Streier (Trochtelfingen), T. Szczeponik (Freiburg), T. Thum (Freiburg), E. Un-teregger (Freiburg), W. Vees (Starzach), E. Vogt (Offenburg), H. Wagner (Engstingen), M. Wepler (Freiburg), A. Ziegler (Lichtenstein), S. and U. Ziegler (Nehren), U. Zimmermann (Reutlingen). Conflict of Interest: none. Abbreviations: ACCP - American College of Chest Physicians AF - atrial fibrillation CRF - case report form EC - transthoracic echocardiogram ECG - electrocardiogram INR - international normalized ratio NSAID - nonsteroidal anti-inflammatory NYHA - New York Heart Association functional class TIA - transient ischaemic attack REFERENCES: 1. Lip GYH, Boos CJ. Antithrombotic treatment in atrial fibrillation. Heart 2006;92:155-161. 2. Albers GW, Dalen JE, Laupacis A, Manning WJ, Petersen P, Singer DE. Antithrombotic therapy in atrial fibrillation. Chest 2001;19(1 Suppl):194S-206S. 3. Singer D, Albers GW, Dalen JE, Go AS, Halperin JL, Manning WJ. Antithrombotic therapy in atrial fibrillation: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126(3 Suppl):429S-456S. 4. McNamara RL, Tamariz LJ, Segal JB, Bass EB. Management of atrial fibrillation: review of the evidence for the role of pharmacologic therapy, electrical cardioversion, and echocardiography. Ann Intern Med 2003;139:1018-1033. 5. Fuster V, Rydén LE, Asinger RW, Cannom DS, Cri-jns HJ, Frye RL, Halperin JL, Kay GN, Klein WW, Lévy S, McNamara RL, Prystowsky EN, Wann LS, Wyse DG. ACC/AHA/ESC guidelines for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines and Policy Conferences (Committee to Develop Guidelines for the Management of Patients With Atrial Fibrillation). Circulation 2001;104:2118-2150. 6. Caro JJ, Flegel KM, Orejuela ME, Kelley HE, Speck-man JL, Migliaccio-Walle K. Anticoagulant prophylaxis against stroke in atrial fibrillation: effectiveness in actual practice. CMAJ 1999;161:493-497. Erratum: CMAJ 2000;162:973. 7. Kalra L, Yu G, Perez I, Lakhani A, Donaldson N. Prospective cohort study to determine if trial efficacy of anticoagulation for stroke prevention in atrial fibrillation translates into clinical effectiveness. BMJ 2000;320:1236-1239. 8. Evans A, Kalra L. Are the results of randomized controlled trials on anticoagulation in patients with atrial fibrillation generalizable to clinical practice? Arch Intern Med 2001;161:1443-1447. 9. Frost L, Johnsen SP, Pedersen L, Toft E, Husted S, Sorensen HT. Atrial fibrillation or flutter and stroke: a Danish population-based study of the effectiveness of oral anticoagulation in clinical practice. J Intern Med 2002;252:64-69. 10. Go AS, Hylek EM, Chang Y, Phillips KA, Henault LE, Capra AM, Jensvold NG, Selby JV, Singer DE. Anticoagulation therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical practice? JAMA 2003;290:2685-2692. 11. Inoue H, Nozawa T, Okumura K, Iwasa A, Lee JD, Shimizu A, Hayano M, Yano K. Attitudes of Japanese cardiologists toward anticoagulation for non-valvular atrial fibrillation and reasons for its under-use. Circ J 2004;68:417-421. 12. Nilsson GH, Bjorholt I. Occurrence and quality of anticoagulant treatment of chronic atrial fibrillation in primary health care in Sweden: a retrospective study on electronic patient records. BMC Clin Pharmacol 2004;4:1. 13. Laguna P, Martin A, del Arco C, Gargantilla P. Investigators in the Spanish Atrial Fibrillation in Emergency Medicine Study Group (GEFAUR). Risk factors for stroke and thromboprophylaxis in atrial fibrillation: what happens in daily clinical practice? The GEFAUR-1 study. Ann Emerg Med 2004;44:3-11. 14. White S, Feely J, O'Neill D. Community-based study of atrial fibrillation and stroke prevention. Ir Med J 2004;97:10-12. 15. Osseby GV, Benatru I, Sochurkova D, Urbinelli R, Megherbi SE, Couvreur G, Moreau T, Wolf J, Giroud M. Trends in utilization of antithrombotic therapy in patients with atrial fibrillation before stroke onset in a community-based study, from 1985 through 1997. From scientific evidence to practice. Prev Med 2004;38:121-128. 16. Bücheler R, Meisner C, Mörike K, Fux R, Hauswald K, Lorenz G, Kühlkamp V, Gleiter CH. Qualität der antithrombotischen Therapie bei chronischem Vorhofflimmern: AFib-Studie. (Quality of antithrombotic therapy in patients with chronic atrial fibrilla- tion: the AFib Trial) Z Ärztl Fortbild Qualitätssich 2003;97:421-431. 17. Lévy S, Camm AJ, Saksena S, Aliot E, Breithardt G, Crijns HJ, Davies DW, Kay GN, Prystowsky EN, Sutton R, Waldo AL, Wyse DG, Working Group on Arrhythmias of European Society of Cardiology; Working Group of Cardiac Pacing of European Society of Cardiology; North American Society of Pacing and Electrophysiology. International consensus on nomenclature and classification of atrial fibrillation: A collaborative project of the Working Group on Arrhythmias and the Working Group of Cardiac Pacing of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. J Cardiovasc Electrophysi-ol 2003;14:443-445. 18. Hart RG, Halperin JL, Pearce LA., Anderson DC, Kronmal RA, McBride R, Nasco E, Sherman DG, Talbert RL, Marler JR. Stroke Prevention in Atrial Fibrillation Investigators. Lessons from the Stroke Prevention in Atrial Fibrillation trials. Ann Intern Med 2003;138:831-838. 19. Stroke Prevention in Atrial Fibrillation Study Investigators. Stroke Prevention in Atrial Fibrillation Study. Final results. Circulation 1991;84:527-539. 20. Anon. Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study. Lancet 1994;343:687-691. 21. Anon. Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial. Lancet 1996;348:633-638. 22. The SPAF III Writing Committee for the Stroke Prevention in Atrial Fibrillation Investigators. Patients with nonvalvular atrial fibrillation at low risk of stroke during treatment with aspirin: Stroke Prevention in Atrial Fibrillation III Study. JAMA 1998;279:1273-1277. 23. Ezekowitz MD, Bridgers SL, James KE, Carliner NH, Colling CL, Gornick CC, Krause-Steinrauf H, Kurtzke JF, Nazaria SM, Radford MJ, Rickles FR, Shabetai R, Deykin D, for the Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibril- lation Investigators.Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators. N Engl J Med 1992;327:1406-1412. 24. BPI, Service, GmbH. FachInfo. Fachinformationsverzeichnis Deutschland. 2001; http://www. fachinfo.de (1/2001). 25. Ono A, Fujita T. Stroke prevention in patients with atrial fibrillation. J Clin Neurosci 2003;10:71-73. 26. Scott PA, Pancioli AM, Davis LA, Frederiksen SM, Eckman J. Prevalence of atrial fibrillation and antithrombotic prophylaxis in emergency department patients. Stroke 2002;33:2664-2669. 27. Müller-Oerlinghausen B, Rostock T. Ambulante Therapiegewohnheiten der Berliner Ärzteschaft bei der medikamentösen Kardioversion von Vor-hofflimmern. (Pharmacologic management of atrial fibrillation by specialists in internal medicine, cardiologists, and general practitioners in Berlin). Med Klin (Munich) 2003;98:359-363. 28. Guyatt G, Schunemann H, Cook D, Jaeschke R, Pauker S, Bucher H. Grades of recommendation for antithrombotic agents. Chest 2001;119 (1 Suppl):3S-7S. 29. Fuster V, Rydén LE, Cannom DS, Curtis AB, Ellenbogen KA, Halperin JL, Le Heuzey J-Y, Kay GN, Lowe LE., Olsson SB., Prystowsky EN, Tamargo JL, Wann LS. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation -executive summary: a report of the American College of Cardiology/American Heart Association Task Force and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation). J Am Coll Cardiol 2006;48:854-906. 30. Singer DE, Albers GW, Dalen JE, Fang MC, Go AS, Halperin JL, Lip GYH, Manning WJ. Antithrombotic therapy in atrial fibrillation: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 2008;133:546-592.