Clinical Therapeutics/Volume 36, Number 9, 2014
Vitamin K Antagonist Treatment in Patients With Atrial Fibrillation and Time in Therapeutic Range in Four European Countries Franc¸ois-Emery Cotte´, PharmD, MPH, PhD1; Hicham Benhaddi, MSc1; Isabelle Duprat-Lomon, MD1; Adam Doble, PhD2; Nick Marchant, BA, MSc3; Alexia Letierce, PhD4; and Michael Huguet, MSc4 1
Health Economics & Outcomes Research, Bristol-Myers Squibb, Rueil-Malmaison, France; Foxymed, Paris, France; 3Pfizer, Walton Oaks, United Kingdom; and 4Cegedim Strategic Data, Boulogne-Billancourt, France 2
ABSTRACT Purpose: Patients with atrial fibrillation are at increased risk for stroke and thus require anticoagulant prophylaxis with vitamin K antagonists. However, many such patients fail to achieve target coagulation status. The objective of this study was to evaluate time in the therapeutic range and its relationship to clinical outcomes in patients with nonvalvular atrial fibrillation prescribed a vitamin K antagonist in everyday clinical practice in 4 European countries (France, Germany, Italy and the United Kingdom). Methods: Data were extracted from the European electronic primary care database, the Longitudinal Patient Database. Included in the analysis were 6250 adult patients for whom data on monitoring of coagulation time and international normalized ratio were available. The time within the therapeutic range was estimated by using the Rosendaal method. Patients spending 470% of time within the therapeutic range were considered to have well-controlled treatment. Data on stroke and bleeding events occurring during the study period were taken from patient records. Stroke risk was calculated by using the CHA2DS2-VASc score (i.e. 2 points for a history of stroke or TIA and age 475years, and 1 point for age between 65 and 74 years, hypertension, diabetes mellitus, a recent cardiac failure, vascular disease and female sex).
patients. After adjustment for stroke risk factors, the odds ratio was 1.38 (95% CI, 0.93–2.06; P ¼ 0.110). The incidence of hemorrhage was 1.1 and 1.3 events/ 100 person-years, respectively (odds ratio, 0.91 [95% CI, 0.72–1.16]). Implications: Many patients receiving prophylaxis with vitamin K antagonists in everyday community care have poorly controlled anticoagulation treatment with vitamin K antagonists. Their international normalized ratio is frequently outside the therapeutic range, and they are thus exposed to an unnecessary risk of stroke or bleeding complications. (Clin Ther. 2014;36:1160–1168) & 2014 The Authors. Published by Elsevier HS Journals, Inc. Key Words: vitamin K antagonist, warfarin, atrial fibrillation, stroke, hemorrhage, anticoagulation.
Findings: The proportion of patients with poorly controlled treatment varied from 34.6% in the United Kingdom to 55.8% in Germany. The incidence of stroke was 0.5/100 person-years in well-controlled patients, compared with 1.0/100 in poorly controlled
Accepted for publication July 23, 2014. http://dx.doi.org/10.1016/j.clinthera.2014.07.016 0149-2918/$ - see front matter
1160
INTRODUCTION Atrial fibrillation (AF) is the most frequently observed arrhythmia in clinical practice, with an estimated prevalence in the general population between 1% and 2%.1–3 The prevalence increases with age, from o0.5% between 40 and 50 years, to 5% to 15% at 480 years of age,4 with the incidence doubling for each successive decade. In France, the number of subjects with AF has been estimated to be between 600,000 and 1 million, with a median age at diagnosis
& 2014 The Authors. Published by Elsevier HS Journals, Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Volume 36 Number 9
F-E. Cotte´ et al. of 75 years.5 AF affects at least 1% of the population in Germany6 and 1.3% in Italy.7 Epidemiologic studies in the United Kingdom have reported prevalence rates for AF of 0.65% in subjects aged 45 to 64 years in the general population8 and 4.7% in subjects 465 years old in primary care.9 AF is associated with high morbidity and mortality,10 in particular with an elevated risk of stroke.11,12 After controlling for age and other risk factors, the risk for stroke is between 2 and 7 times higher than in subjects without AF.13,14 Strokes related to AF account for 1 in 5 strokes15 and are associated with a high recurrence rate and worse prognosis.16 To prevent the occurrence of stroke in patients with AF, prophylaxis with oral anticoagulant drugs is recommended. The American College of Chest Physicians14,17 recommends using the CHADS2 score18 to stratify stroke risk and that all patients with a CHADS2 score Z1 receive longterm prophylaxis with an oral anticoagulant. Guidelines from the European Society of Cardiology, published in 20104 and updated in 2012,19 provide similar advice, although they recommend the use of the CHA2DS2-VASc as a more sensitive tool for risk stratification. Nevertheless, several previous studies have shown that many patients with AF do not receive anticoagulant prophylaxis according to treatment guidelines and that lack of compliance with these guidelines is associated with a higher risk of stroke and with higher mortality.21–23 Historically, vitamin K antagonists (VKAs) have been the mainstay of oral anticoagulant prophylaxis, but coagulation status needs to be monitored carefully to avoid major hemorrhagic complications, which can be fatal.24 Coagulation status is expressed as the international normalized ratio (INR), the target therapeutic range being defined as an INR between 2 and 3. An INR o2.0 is associated with an increased risk of stroke, and an INR Z4 is associated with an increased risk of major bleeding.25 The proportion of time in the therapeutic range (TTR) thus provides a useful measure of the level of control of cardiovascular risk. This parameter is a major determinant of the efficacy and safety of VKAs, with maximum benefits evident when the TTR is Z70%.26–28 Despite the importance of maintaining an acceptable INR, it has been estimated in randomized controlled trials that patients spend more than one third of their time outside the TTR.29 In a recent Italian study of 112 patients with AF starting VKA prophylaxis at the time of
September 2014
hospital discharge, only 58% achieved an adequate TTR.30 It may be expected, however, that coagulation status in everyday practice is less well controlled than in randomized controlled trials.31 For this reason, we have performed a large retrospective observational cohort (Real Evidence of AntiCoagulation Treatment in Atrial Fibrillation) of patients with nonvalvular AF in 4 European countries that provides an opportunity to collect data on patients treated with VKAs. The primary objective of the present analysis was to assess TTR in everyday clinical practice in these patients. The relationship between TTR and clinical outcomes was also evaluated, but the analysis was purely exploratory.
PATIENTS AND METHODS This was a retrospective observational cohort study of patients with a diagnosis of nonvalvular AF conducted in 4 European countries (France, Germany, Italy and the United Kingdom) between May 2010 and May 2012. The study included a cross-sectional phase and a longitudinal phase to assess the occurrence of stroke.
Data Source Data were extracted from electronic medical records in the Longitudinal Patient Database (LPD) managed by Cegedim Strategic Data (Boulogne Billancourt, France) and owned by Cegedim Group, a company specialised in developing databases and software in the field of health. LPD is a computerized network of general practitioners (GPs) in different European countries who contribute extensive anonymous data on patient consultations and treatment to a centralized database. The database reflects the regular activity of a national sample of independent practitioners, and it enables collection and longitudinal analysis of data taken from patient records related to prescription practices and health care consumption in everyday clinical practice. Currently, 4500 primary care practitioners contribute to the LPD in the 4 countries of the study, providing data from routinely collected records of 45 million patients. The LPD database has been shown to be a reliable source of information in numerous previous studies in several disease areas,32–34 including AF.35,36
Study Population All adult patients with a diagnosis of nonvalvular AF in the LPD database who had consulted a physician at least once during the 2-year study period
1161
Clinical Therapeutics and had been prescribed a VKA at least once were identified. From this group, we selected those patients for whom data on monitoring of coagulation time were available and who had at least 2 documented measures of INR separated by o90 days, at least once each year since the beginning of treatment. These patients constituted the analyzable population. For all patients in the analyzable population, the TTR was estimated by using the Rosendaal method.37 Patients with a TTR 470% were considered to have wellcontrolled VKA treatment.38 The analyzable population was thus divided into 2 subgroups of well-controlled (TTR 470%) and poorly controlled (TTR r70%) patients.
Statistical Analysis The population evaluated in this article corresponded to all patients for whom data were available for least 2 measures of INR separated by o90 days. The analysis of the data is principally descriptive. Data are described by using mean (SD) and median (range) values for continuous variables and frequency counts and percentages for categorical variables. For selected variables, 95% CIs were calculated. Risk of stroke and hemorrhage in patients with suboptimal TTR was evaluated as odds ratios with their 95% CIs. A multivariate logistic regression was performed to adjust for potential confounding factors, namely age, sex, CHA2DS2-Vasc score, and VKA treatment duration. Patient follow-up data were censored after discontinuation of VKA.
Data Collection At the index consultation, data were extracted from the electronic medical records on age, sex, body mass index, date of AF diagnosis, comorbidities, history of complications, treatments, and smoking status. Timedependent variables such as age were also documented at the date of treatment initiation and treatment discontinuation for treated patients, at the end of the follow-up period, and at the visit before occurrence of stroke (for patients who experienced strokes). For patients receiving VKA medication, coagulation status (determined by using the INR) was documented. Stroke risk was calculated with the CHA2DS2VASc score.20 This is calculated by assigning 1 point each for the presence of congestive heart failure, hypertension, diabetes mellitus or vascular disorder, for female sex, and for age 65 to 74 years; 2 points were assigned each for age Z75 years and for a history of stroke or transient ischemic attack. The presence of comorbidities was ascertained from the International Classification of Diseases, 10th Revision, disease codes documented in the database.
Ethics The study was performed in accordance with Good Epidemiological Practice guidelines and relevant international and national guidelines for pharmacoepidemiologic studies. Ethics board approval was not required because the data were not nominative and were collected retrospectively.
RESULTS Patients Overall, 32,075 patients prescribed a VKA were included from the 4 participating countries (Figure 1). Of these, sufficient data on INR to determine the TTR were available for 6250 patients (19.5%), who thus constituted the analyzable population.
63,244 patients with atrial fibrillation
32,075 patients prescribed VKA
6250 patients analyzable
Incidence of Stroke and Bleeding Occurrence of first stroke and bleeding was defined as the first recording of a diagnosis of stroke in the LPD database after the index date. The index date was defined as the date of the first consultation during the study period for patients with a prior diagnosis of AF and the date of the consultation at which a diagnosis of AF was made for newly diagnosed patients. The observation period extended from the index date to October 31, 2012.
1162
France n = 1110
Germany n = 1208
Italy n = 1230
UK n = 2702
Figure 1. Patient enrollment and anticoagulant prophylaxis. VKA ¼ vitamin K antagonist.
Volume 36 Number 9
September 2014 Table I. Demographic characteristics and stroke risk factors in patients with nonvalvular atrial fibrillation at the index consultation treated with vitamin K antagonists (VKAs). France (N ¼ 16,447) Characteristic Age, y Mean (SD) Median, range o75 y Z75 y Sex Male Female Medical sroke risk factors Diabetes Arterial hypertension Coronary heart disease Heart failure Vascular disease History of stroke or TIA CHA2DS2-VASc score* Mean (SD Median [range] Z1 Z2
All VKA (n ¼ 9184)
74.9 77 3766 5418
(10.5) [18–103] (41.0%) (59.0%)
5714 (62.2%) 3470 (37.8%)
Germany (N ¼ 13,917)
Analyzable (n ¼ 1110)
74.4 76 480 630
(9.6) [20–94] (43.2%) (56.8%)
636 (57.3%) 474 (42.7%)
All VKA (n ¼ 6386)
73.9 75 3059 3327
(9.5) [22–100] (47.9%) (52.1%)
3460 (54.2%) 2926 (45.8%)
Italy (N ¼ 22,447)
Analyzable (n ¼ 1208)
71.7 73 681 527
(9.5) [37–95] (56.4%) (43.6%)
622 (51.5%) 586 (48.5%)
All VKA (n ¼ 8774)
76.2 78 3160 5614
(9.7) [21–112] (36.0%) (64.0%)
United Kingdom (N ¼ 10,433)
Analyzable (n ¼ 1230)
74.4 75 571 659
(8.4) [23–111] (46.4%) (53.6%)
All VKA (n ¼ 7731)
74.8 76 3305 4426
(10.5) [18–102] (42.7%) (57.3%)
4456 (50.8%) 4318 (49.2%)
580 (47.2%) 650 (52.8%)
4434 (57.4%) 3297 (42.6%)
Analyzable (n ¼ 2702)
72.1 73 1520 1182
(9.7) [29–98] (56.3%) (43.7%)
1527 (56.5%) 1175 (43.5%)
2090 6502 1437 1454 1178 1042
(22.8%) (70.8%) (15.6%) (15.8%) (12.8%) (11.3%)
280 808 201 1262 806 998
(25.2%) (72.8%) (18.1%) (20.2%) (12.9%) (16.0%)
2724 5316 2834 1966 1239 998
(42.7%) (83.2%) (44.4%) (30.8%) (19.4%) (15.6%)
467 1008 493 340 236 200
(38.7%) (83.4%) (40.8%) (28.1%) (19.5%) (16.6%)
2263 6395 1955 1730 947 1118
(25.8%) (72.9%) (22.3%) (19.7%) (10.8%) (12.7%)
359 935 290 262 153 181
(29.2%) (76.0%) (23.6%) (21.3%) (12.4%) (14.7%)
1154 2255 1696 1339 724 1245
(14.9%) (29.2%) (21.9%) (17.3%) (9.4%) (16.1%)
410 796 545 438 246 463
(15.2%) (29.5%) (20.2%) (16.2%) (9.1%) (17.1%)
2.7 3 8323 6912
(1.6) [0–9] (90.6%) (75.3%)
3.3 3 1065 950
(1.6) [0–8] (95.9%) (85.6%)
3.1 3 5994 5090
(1.8) [0–9] (93.9%) (79.7%)
3.0 3 1118 953
(1.8) [0–9] (92.5%) (78.9%)
3.0 3 8246 7057
(1.7) [0–9] (94.0%) (80.4%)
3.5 3 1087 143
(1.6) [0–9] (88.4%) (11.6%)
2.3 2 6702 5154
(1.5) [0–9] (86.7%) (66.7%)
2.5 2 2449 1970
(1.5) [0–8] (90.6%) (72.9%)
1163
F-E. Cotte´ et al.
TIA ¼ transient ischemic attack. CHA2DS2-VASc score : 2 points for a history of stroke or TIA and age >75years, and 1 point for age between 65 and 74 years, hypertension, diabetes mellitus, a recent cardiac failure, vascular disease and female sex. * At the time of diagnosis of atrial fibrillation or the time that treatment was initiated, whichever came later.
Clinical Therapeutics Demographic data for all patients receiving VKAs, as well as for the 6250 patients in the analyzable population, are provided in Table I, together with information on the CHA2DS2-VASc stroke risk scores. The median CHA2DS2-VASc score was 3 in all countries except the United Kingdom (median, 2). Some differences were observed between the analyzable population and all the study patients prescribed a VKA. In all countries, the analyzable population tended to be younger, with a higher proportion of women.
VKA Treatment and Monitoring The most widely prescribed VKAs were warfarin in Italy and the United Kingdom, fluindione in France, and phenprocoumon in Germany (Table II). The median duration of treatment was 4 years in all countries except the United Kingdom (5 years). On average, the INR was determined at least 9 times per year in all countries. The mean TTR ranged from 66.0% in Italy to 72.6% in the United Kingdom. In France, Germany, and Italy, o50% of the patients were considered to have wellcontrolled VKA treatment (TTR 470%) and up to 20% spent o50% of their time outside the therapeutic range. The proportion of patients spending more TTR was higher in the United Kingdom.
TTR and Stroke and Bleeding Risk The association between the time spent in the therapeutic range and the incidence of stroke and hemorrhage was investigated (Figure 2). To optimize the power of the analysis, data were pooled across the 4 participating countries. Overall, 105 ischemic strokes were documented for 14,836 patient-years of VKA prescription. These strokes were more frequent in patients who had a TTR r70% (1.0 stroke/100 person-years) than in patients whose coagulation status was well controlled (0.5 stroke/100 personyears). The unadjusted odds ratio for stroke was 1.60 (95% CI, 1.08–2.36; P ¼ 0.019). After adjustment for age, sex, treatment duration, and CHA2DS2-Vasc score, the adjusted odds ratio was 1.38 (95% CI, 0.93 – 2.06), which was no longer significant (P ¼ 0.110). In terms of hemorrhage, 291 bleeding events were documented, including 55 gastrointestinal hemorrhages and 24 hemorrhagic strokes. The frequency of events was somewhat lower in patients who had well-controlled VKA treatment (1.1 vs 1.3 events/100 person-years). No significant association was observed between hemorrhage rate and TTR, either in the unadjusted analysis (odds ratio, 1.00 [95% CI, 0.79–1.27]; P o 0.098) or the adjusted analysis (odds ratio, 0.91 [95% CI, 0.72–1.16]; P ¼ 0.45).
Table II. Prescription of vitamin K antagonists (VKAs). Variable Individual VKA* Warfarin Acenocoumarol Fluindione Phenprocoumon Phenindione Duration of VKA treatment, y Median [range] Determination of INR Mean no. of measures/patient Mean no. of measures/patient/year Median no. of measures/patient/year Time spent in therapeutic range, % Mean (SD) INR in therapeutic range o50% of the time 50%–70% of the time 470% of the time
France (n ¼ 1110)
Germany (n ¼ 1208)
Italy (n ¼ 1230)
97 (8.7%) 68 (6.1%) 945 (85.1%) — —
7 (0.6%) — — 1201 (99.4%) —
1091 (88.7%) 139 (11.3%) — — —
4 [1–13] 37.5 9.61 9.14 66.2
(32.4) (5.86) [1.0–39.0] (21.1)
209 (18.8%) 370 (33.3%) 531 (47.8%)
4 [1–10] 38.4 9.48 9.25 65.3
(30.0) (4.49) [1.5–23.0] (19.6)
232 (19.2%) 441 (36.5%) 535 (44.2%)
4 [1–14] 49.0 9.78 10.0 66.0
United Kingdom (n ¼ 2702)
2697 (99.8%) 2 (0.1%) — — 3 (0.1%) 5 [1–13]
(40.7) (5.40) [1.0–42.7] (18.4)
192 (15.6%) 471 (38.3%) 567 (46.1%)
74.7 13.07 12.36 72.6
(48.8) (5.46) [1.5–51.5] (13.7)
156 (5.8%) 779 (28.8%) 1767 (65.4%)
INR ¼ international normalized ratio. * Because certain patients were prescribed 41 VKA, these treatments are not mutually exclusive.
1164
Volume 36 Number 9
F-E. Cotte´ et al.
Frequency (Events/100 Person-Years)
2.0
Stroke
Hemorrage
1.5
1.0
0.5
0
≤70%
>70% ≤70% Time in Therapeutic Range
>70%
Figure 2. Rate of occurrence of stroke or bleeding as a function of time spent in therapeutic range. Data are presented as mean values with their 95% CIs.
DISCUSSION This study evaluated patients with nonvalvular AF treated with VKA in everyday care in 4 European countries. We observed that a majority of the patients in France (52%), Germany (56%), and Italy (54%) had a TTR r70%. The situation in the United Kingdom was somewhat better, with approximately one third of patients presenting with a TTR o70% (35%). Compared with the other countries, patients with AF in the United Kingdom differed in having a lower stroke risk, more frequently evaluated INR, and a higher TTR. These differences may be explained by the care paradigm in the United Kingdom: such patients are managed by physicians in specialized “warfarin clinics,” where INR is frequently controlled and the dose of VKA is adapted. It is also noteworthy that different VKAs were used in different countries: warfarin accounted for the large majority of VKA use in Italy and the United Kingdom, phenprocoumon in Germany, and fluindione in France. Comparative data on the effectiveness of these VKAs in everyday practice are extremely limited, and it is possible that differences in the VKAs prescribed may contribute to some of the between-country variation in TTR observed. The relation between TTR and clinical outcome was also evaluated, although it should be noted that this analysis was purely exploratory, and the study was neither designed nor powered to address this factor.
September 2014
When the data were pooled across countries, we observed a higher frequency of stroke in these poorly controlled patients (1.0 stroke/100 person-years) compared with those with a TTR 470% (0.5 stroke/100 person-years). However, when adjusted on stroke risk factors, the odds ratio did not reach statistical significance. Sample attrition may have provided less power to our analysis, and the observed trend should be evaluated in a larger population. The absence of a similar trend for hemorrhage is surprising but could be explained by the fact that patients with uncontrolled VKA treatment may be more often below the intended INR therapeutic range than above. This phenomenon has been observed previously39 and may reflect a tendency of GPs to underdose VKAs because of possible adverse effects related to concerns about VKA-associated hemorrhage. This study has several limitations, principally linked to the use of electronic general practice databases. These databases provide information on a large number of patients gathered in real-life conditions by a representative national sample of GPs. However, because the data are collected in everyday clinical practice, their availability depends on the assiduity of the GP in using the medical software. Whereas basic data on patient characteristics, diseases, and drug prescriptions are generally collected comprehensively by all GPs, information on paraclinical tests such as INR assays may only be documented in the electronic database by some GPs some of the time. Calculation of the TTR requires regular assay and documentation of the INR, which could explain why only 20% of the patients in our survey were considered analyzable in the 4 countries considered. The low proportion of analyzable patients may have introduced some inclusion bias with respect to all patients with AF who consult their GP. Indeed, in all 4 countries, the analyzable populations tended to be younger and more frequently women, although their stroke risks were similar to those of the entire study sample. Generally speaking, the characteristics of the AF patients evaluated were consistent with previous findings from real-life cohorts in France,40 Germany,41 and Italy.42 With such databases, it is impossible to ensure that all medical events are exhaustively documented, particularly if they are not the specific objective of the GP consultation. Thus, it is useful to compare the event rates documented in our study with those reported in dedicated patient cohorts in real-life conditions of care. For example, in a cohort of 3302 patients with AF in Italy,39 the rates of stroke (0.76 event/100 patient-years)
1165
Clinical Therapeutics and bleeding events (1.24 events/100 patient-years) were similar to those reported in the present study. Conversely, in the United Kingdom and France, although the incidence of stroke is consistent with what has been observed previously in real-life studies (0.9 and 1.2 events/100 patient-years, respectively), the incidence of hemorrhage in the present study appeared to be lower than expected from these earlier studies (3.8 and 3.2 events/100 patient-years, respectively).43,44 This discrepancy could be attributable, at least in part, to the fact that bleeding events may be managed elsewhere than in primary care, and events that are not serious or without consequences may go undocumented.
2.
3.
4.
5.
CONCLUSIONS This study of the management of AF in primary care in 4 European countries indicates that many patients receiving VKA prophylaxis did not achieve their target coagulation status, with their INR frequently outside the therapeutic range, and are thus exposed to an unnecessary risk of stroke or bleeding complications. A trend toward a higher incidence of stroke in patients who fail to achieve target levels was observed.
6.
7.
8.
ACKNOWLEDGMENTS Funding for the study was provided by Bristol-Myers Squibb and Pfizer, purveyors of apixaban, which is indicated for stroke prevention in patients with AF.
CONFLICTS OF INTEREST Dr. Cotté, Mr. Benhaddi and Dr. Duprat-Lomon are employees of Bristol-Myers Squibb and Mr. Marchant is an employee of Pfizer. Drs. Letierce and Mr. Huguet are employees of Cegedim, a contract research organization that was mandated by Bristol-Myers Squibb and Pfizer to implement the REACT AF study, for which they received financial compensation. Dr. Doble is an employee of Foxymed, a medical communication and consultancy company which participated in the exploration and interpretation of the results of the REACT AF study on behalf of Bristol-Myers Squibb and Pfizer and prepared the manuscript for publication. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
9.
10.
11.
12.
13.
14.
15.
REFERENCES 1. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the
1166
16.
AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA. 2001;285:2370–2375. Wilke T, Groth A, Mueller S, et al. Incidence and prevalence of atrial fibrillation: an analysis based on 8.3 million patients. Eur Soc Cardiol. 2013;15:486–493. Kannel WB, Wolf PA, Benjamin EJ, et al. Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: population-based estimates. Am J Cardiol. 1998;82:2N–9N. Camm AJ, Kirchhof P, Lip GY, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J. 2010;31:2369–2429. Charlemagne A, Blacher J, Cohen A, et al. Epidemiology of atrial fibrillation in France: extrapolation of international epidemiological data to France and analysis of French hospitalization data. Arch Cardiovasc Dis. 2011;104:115–124. Meinertz T, Kirch W, Rosin L, et al. Management of atrial fibrillation by primary care physicians in Germany: baseline results of the ATRIUM registry. Clin Res Cardiol. 2011; 100:897–905. Piccinocchi G, Laringe M, Guillaro B, et al. Diagnosis and management of atrial fibrillation by primary care physicians in Italy: a retrospective, observational analysis. Clin Drug Invest. 2012;32:771–777. Stewart S, Hart CL, Hole DJ, et al. Population prevalence, incidence, and predictors of atrial fibrillation in the Renfrew/Paisley study. Heart. 2001;86:516–521. Sudlow M, Thomson R, Thwaites B, et al. Prevalence of atrial fibrillation and eligibility for anticoagulants in the community. Lancet. 1998;352:1167–1171. Benjamin EJ, Wolf PA, D’Agostino RB, et al. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998;98:946–952. McManus DD, Rienstra M, Benjamin EJ. An update on the prognosis of patients with atrial fibrillation. Circulation. 2012;126:e143–e146. Tendera M, Syzdol M, Parma Z. ARISTOTLE RE-LYs on the ROCKET. What’s new in stroke prevention in patients with atrial fibrillation? Cardiol J. 2012;19:4–10. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22:983–988. Singer DE, Albers GW, Dalen JE, et al. Antithrombotic therapy in atrial fibrillation: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6 Suppl):546S–592S. Marini C, De Santis F, Sacco S, et al. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: results from a population-based study. Stroke. 2005;36:1115–1119. Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial
Volume 36 Number 9
F-E. Cotte´ et al.
17.
18.
19.
20.
21.
22.
23.
fibrillation. Arch Intern Med. 1994; 154:1449–1457. You JJ, Singer DE, Howard PA, et al. Antithrombotic therapy for atrial fibrillation: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):e531S–e5375. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA. 2001;285: 2864–2870. Camm AJ, Lip GY, De Caterina R, et al. 2012 Focused update of the ESC guidelines for the management of atrial fibrillation: an update of the 2010 ESC guidelines for the management of atrial fibrillation. *Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J. 2012;33:2719–2747. Olesen JB, Torp-Pedersen C, Hansen ML, et al. The value of the CHA2DS2VASc score for refining stroke risk stratification in patients with atrial fibrillation with a CHADS2 score 0-1: a nationwide cohort study. Thromb Haemost. 2012;107:1172–1179. Gorin L, Fauchier L, Nonin E, et al. Prognosis and guideline-adherent antithrombotic treatment in patients with atrial fibrillation and atrial flutter: implications of undertreatment and overtreatment in real-life clinical practice; the Loire Valley Atrial Fibrillation Project. Chest. 2011;140:911–917. Lin LJ, Cheng MH, Lee CH, et al. Compliance with antithrombotic prescribing guidelines for patients with atrial fibrillation—a nationwide descriptive study in Taiwan. Clin Ther. 2008;30:1726–1736. Nieuwlaat R, Olsson SB, Lip GY, et al. Guideline-adherent antithrombotic treatment is associated with improved outcomes compared with undertreatment in high-risk patients with atrial fibrillation. The Euro Heart Survey on Atrial
September 2014
24.
25.
26.
27.
28.
29.
30.
31.
32.
Fibrillation. Am Heart J. 2007;153: 1006–1012. Fang MC, Go AS, Chang Y, et al. Death and disability from warfarinassociated intracranial and extracranial hemorrhages. Am J Med. 2007;120: 700–705. Mohr JP, Thompson JL, Lazar RM, et al. A comparison of warfarin and aspirin for the prevention of recurrent ischemic stroke. N Engl J Med. 2001;345:1444–1451. Gallego P, Vilchez JA, Lane DA. Apixaban compared with warfarin for stroke prevention in atrial fibrillation: implications of time in therapeutic range. Circulation. 2013;127: 2163–2165. De Caterina R, Husted S, Wallentin L, et al. General mechanisms of coagulation and targets of anticoagulants (Section I). Position Paper of the ESC Working Group on Thrombosis—Task Force on Anticoagulants in Heart Disease. Thromb Haemost. 2013;109:569–579. Gallagher AM, Setakis E, Plumb JM, et al. Risks of stroke and mortality associated with suboptimal anticoagulation in atrial fibrillation patients. Thromb Haemost. 2011;106: 968–977. Phillips KW, Ansell J. Outpatient management of oral vitamin K antagonist therapy: defining and measuring high-quality management. Exp Rev Cardiovasc Ther. 2008;6:57–70. Pereira de Sousa L, Burba I, Ruperto C, et al. Vitamin K antagonists in patients with nonvalvular atrial fibrillation: appropriateness and quality of treatment in an Italian cohort. J Cardiovasc Med (Hagerstown). 2013;14:534–540. Wan Y, Heneghan C, Perera R, et al. Anticoagulation control and prediction of adverse events in patients with atrial fibrillation: a systematic review. Circulation Cardiovascular Quality and Outcomes. 2008;1:84–91. Bouee S, Charlemagne A, Fagnani F, et al. Changes in osteoarthritis management by general practitioners in the COX2-inhibitor era-concomitant
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
gastroprotective therapy. Joint Bone Spine. 2004;71:214–220. Cotte FE, Fardellone P, Mercier F, et al. Adherence to monthly and weekly oral bisphosphonates in women with osteoporosis. Osteoporos Int. 2010;21:145–155. Cea Soriano L, Bateman BT, Garcia Rodriguez LA, et al. Prescription of antihypertensive medications during pregnancy in the UK. Pharmacoepidemiol Drug Safe. 2014. Mazzaglia G, Filippi A, Alacqua M, et al. A national survey of the management of atrial fibrillation with antithrombotic drugs in Italian primary care. Thromb Haemost. 2010;103:968–975. Sabouret P, Depret-Bixio L, Cotte FE, et al. Sex differences in stroke prevention in atrial fibrillation in French primary care. Results of the AFIGP (Atrial Fibrillation In General Practice) Database. Clin Res Cardiol. 2014. Rosendaal FR, Cannegieter SC, van der Meer FJ, et al. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost. 1993;69:236–239. Lader E, Martin N, Cohen G, et al. Warfarin therapeutic monitoring: is 70% time in the therapeutic range the best we can do? J Clin Pharm Ther. 2012;37:375–377. Poli D, Testa S, Antonucci E, et al. Bleeding and stroke risk in a realworld prospective primary prevention cohort of patients with atrial fibrillation. Chest. 2011;140:918–924. Cohen A, Dallongeville J, DurandZaleski I, et al. Characteristics and management of outpatients with history of or current atrial fibrillation: the observational French EPHA study. Arch Cardiovasc Dis. 2010;103:376–387. Reinhold T, Rosenfeld S, MullerRiemenschneider F, et al. [Patients suffering from atrial fibrillation in Germany. Characteristics, resource consumption and costs]. Herz. 2012;37: 534–542. Zoni-Berisso M, Filippi A, Landolina M, et al. Frequency, patient characteristics, treatment strategies, and
1167
Clinical Therapeutics resource usage of atrial fibrillation (from the Italian Survey of Atrial Fibrillation Management [ISAF] study). Am J Cardiol. 2013;111:705–711. 43. Gallagher AM, van Staa TP, MurrayThomas T, et al. Population-based cohort study of warfarin-treated patients with atrial fibrillation: incidence of cardiovascular and bleeding outcomes. BMJ Open. 2014;4:e003839. 44. Cotte FE, Chaize G, Kachaner I, et al. Incidence and cost of stroke and hemorrhage in patients diagnosed with atrial fibrillation in France. J Stroke Cerebrovasc Dis. 2014; 23:e73–e83.
Address correspondence to: François-Emery Cotté, Bristol-Myers Squibb, 3, rue Joseph Monier, 92500 RUEIL MALMAISON, France. E-mail: [email protected]
1168
Volume 36 Number 9
Vitamin K Antagonist Treatment in Patients With Atrial Fibrillation and Time in Therapeutic Range in Four European Countries Franc¸ois-Emery Cotte´, PharmD, MPH, PhD1; Hicham Benhaddi, MSc1; Isabelle Duprat-Lomon, MD1; Adam Doble, PhD2; Nick Marchant, BA, MSc3; Alexia Letierce, PhD4; and Michael Huguet, MSc4 1
Health Economics & Outcomes Research, Bristol-Myers Squibb, Rueil-Malmaison, France; Foxymed, Paris, France; 3Pfizer, Walton Oaks, United Kingdom; and 4Cegedim Strategic Data, Boulogne-Billancourt, France 2
ABSTRACT Purpose: Patients with atrial fibrillation are at increased risk for stroke and thus require anticoagulant prophylaxis with vitamin K antagonists. However, many such patients fail to achieve target coagulation status. The objective of this study was to evaluate time in the therapeutic range and its relationship to clinical outcomes in patients with nonvalvular atrial fibrillation prescribed a vitamin K antagonist in everyday clinical practice in 4 European countries (France, Germany, Italy and the United Kingdom). Methods: Data were extracted from the European electronic primary care database, the Longitudinal Patient Database. Included in the analysis were 6250 adult patients for whom data on monitoring of coagulation time and international normalized ratio were available. The time within the therapeutic range was estimated by using the Rosendaal method. Patients spending 470% of time within the therapeutic range were considered to have well-controlled treatment. Data on stroke and bleeding events occurring during the study period were taken from patient records. Stroke risk was calculated by using the CHA2DS2-VASc score (i.e. 2 points for a history of stroke or TIA and age 475years, and 1 point for age between 65 and 74 years, hypertension, diabetes mellitus, a recent cardiac failure, vascular disease and female sex).
patients. After adjustment for stroke risk factors, the odds ratio was 1.38 (95% CI, 0.93–2.06; P ¼ 0.110). The incidence of hemorrhage was 1.1 and 1.3 events/ 100 person-years, respectively (odds ratio, 0.91 [95% CI, 0.72–1.16]). Implications: Many patients receiving prophylaxis with vitamin K antagonists in everyday community care have poorly controlled anticoagulation treatment with vitamin K antagonists. Their international normalized ratio is frequently outside the therapeutic range, and they are thus exposed to an unnecessary risk of stroke or bleeding complications. (Clin Ther. 2014;36:1160–1168) & 2014 The Authors. Published by Elsevier HS Journals, Inc. Key Words: vitamin K antagonist, warfarin, atrial fibrillation, stroke, hemorrhage, anticoagulation.
Findings: The proportion of patients with poorly controlled treatment varied from 34.6% in the United Kingdom to 55.8% in Germany. The incidence of stroke was 0.5/100 person-years in well-controlled patients, compared with 1.0/100 in poorly controlled
Accepted for publication July 23, 2014. http://dx.doi.org/10.1016/j.clinthera.2014.07.016 0149-2918/$ - see front matter
1160
INTRODUCTION Atrial fibrillation (AF) is the most frequently observed arrhythmia in clinical practice, with an estimated prevalence in the general population between 1% and 2%.1–3 The prevalence increases with age, from o0.5% between 40 and 50 years, to 5% to 15% at 480 years of age,4 with the incidence doubling for each successive decade. In France, the number of subjects with AF has been estimated to be between 600,000 and 1 million, with a median age at diagnosis
& 2014 The Authors. Published by Elsevier HS Journals, Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Volume 36 Number 9
F-E. Cotte´ et al. of 75 years.5 AF affects at least 1% of the population in Germany6 and 1.3% in Italy.7 Epidemiologic studies in the United Kingdom have reported prevalence rates for AF of 0.65% in subjects aged 45 to 64 years in the general population8 and 4.7% in subjects 465 years old in primary care.9 AF is associated with high morbidity and mortality,10 in particular with an elevated risk of stroke.11,12 After controlling for age and other risk factors, the risk for stroke is between 2 and 7 times higher than in subjects without AF.13,14 Strokes related to AF account for 1 in 5 strokes15 and are associated with a high recurrence rate and worse prognosis.16 To prevent the occurrence of stroke in patients with AF, prophylaxis with oral anticoagulant drugs is recommended. The American College of Chest Physicians14,17 recommends using the CHADS2 score18 to stratify stroke risk and that all patients with a CHADS2 score Z1 receive longterm prophylaxis with an oral anticoagulant. Guidelines from the European Society of Cardiology, published in 20104 and updated in 2012,19 provide similar advice, although they recommend the use of the CHA2DS2-VASc as a more sensitive tool for risk stratification. Nevertheless, several previous studies have shown that many patients with AF do not receive anticoagulant prophylaxis according to treatment guidelines and that lack of compliance with these guidelines is associated with a higher risk of stroke and with higher mortality.21–23 Historically, vitamin K antagonists (VKAs) have been the mainstay of oral anticoagulant prophylaxis, but coagulation status needs to be monitored carefully to avoid major hemorrhagic complications, which can be fatal.24 Coagulation status is expressed as the international normalized ratio (INR), the target therapeutic range being defined as an INR between 2 and 3. An INR o2.0 is associated with an increased risk of stroke, and an INR Z4 is associated with an increased risk of major bleeding.25 The proportion of time in the therapeutic range (TTR) thus provides a useful measure of the level of control of cardiovascular risk. This parameter is a major determinant of the efficacy and safety of VKAs, with maximum benefits evident when the TTR is Z70%.26–28 Despite the importance of maintaining an acceptable INR, it has been estimated in randomized controlled trials that patients spend more than one third of their time outside the TTR.29 In a recent Italian study of 112 patients with AF starting VKA prophylaxis at the time of
September 2014
hospital discharge, only 58% achieved an adequate TTR.30 It may be expected, however, that coagulation status in everyday practice is less well controlled than in randomized controlled trials.31 For this reason, we have performed a large retrospective observational cohort (Real Evidence of AntiCoagulation Treatment in Atrial Fibrillation) of patients with nonvalvular AF in 4 European countries that provides an opportunity to collect data on patients treated with VKAs. The primary objective of the present analysis was to assess TTR in everyday clinical practice in these patients. The relationship between TTR and clinical outcomes was also evaluated, but the analysis was purely exploratory.
PATIENTS AND METHODS This was a retrospective observational cohort study of patients with a diagnosis of nonvalvular AF conducted in 4 European countries (France, Germany, Italy and the United Kingdom) between May 2010 and May 2012. The study included a cross-sectional phase and a longitudinal phase to assess the occurrence of stroke.
Data Source Data were extracted from electronic medical records in the Longitudinal Patient Database (LPD) managed by Cegedim Strategic Data (Boulogne Billancourt, France) and owned by Cegedim Group, a company specialised in developing databases and software in the field of health. LPD is a computerized network of general practitioners (GPs) in different European countries who contribute extensive anonymous data on patient consultations and treatment to a centralized database. The database reflects the regular activity of a national sample of independent practitioners, and it enables collection and longitudinal analysis of data taken from patient records related to prescription practices and health care consumption in everyday clinical practice. Currently, 4500 primary care practitioners contribute to the LPD in the 4 countries of the study, providing data from routinely collected records of 45 million patients. The LPD database has been shown to be a reliable source of information in numerous previous studies in several disease areas,32–34 including AF.35,36
Study Population All adult patients with a diagnosis of nonvalvular AF in the LPD database who had consulted a physician at least once during the 2-year study period
1161
Clinical Therapeutics and had been prescribed a VKA at least once were identified. From this group, we selected those patients for whom data on monitoring of coagulation time were available and who had at least 2 documented measures of INR separated by o90 days, at least once each year since the beginning of treatment. These patients constituted the analyzable population. For all patients in the analyzable population, the TTR was estimated by using the Rosendaal method.37 Patients with a TTR 470% were considered to have wellcontrolled VKA treatment.38 The analyzable population was thus divided into 2 subgroups of well-controlled (TTR 470%) and poorly controlled (TTR r70%) patients.
Statistical Analysis The population evaluated in this article corresponded to all patients for whom data were available for least 2 measures of INR separated by o90 days. The analysis of the data is principally descriptive. Data are described by using mean (SD) and median (range) values for continuous variables and frequency counts and percentages for categorical variables. For selected variables, 95% CIs were calculated. Risk of stroke and hemorrhage in patients with suboptimal TTR was evaluated as odds ratios with their 95% CIs. A multivariate logistic regression was performed to adjust for potential confounding factors, namely age, sex, CHA2DS2-Vasc score, and VKA treatment duration. Patient follow-up data were censored after discontinuation of VKA.
Data Collection At the index consultation, data were extracted from the electronic medical records on age, sex, body mass index, date of AF diagnosis, comorbidities, history of complications, treatments, and smoking status. Timedependent variables such as age were also documented at the date of treatment initiation and treatment discontinuation for treated patients, at the end of the follow-up period, and at the visit before occurrence of stroke (for patients who experienced strokes). For patients receiving VKA medication, coagulation status (determined by using the INR) was documented. Stroke risk was calculated with the CHA2DS2VASc score.20 This is calculated by assigning 1 point each for the presence of congestive heart failure, hypertension, diabetes mellitus or vascular disorder, for female sex, and for age 65 to 74 years; 2 points were assigned each for age Z75 years and for a history of stroke or transient ischemic attack. The presence of comorbidities was ascertained from the International Classification of Diseases, 10th Revision, disease codes documented in the database.
Ethics The study was performed in accordance with Good Epidemiological Practice guidelines and relevant international and national guidelines for pharmacoepidemiologic studies. Ethics board approval was not required because the data were not nominative and were collected retrospectively.
RESULTS Patients Overall, 32,075 patients prescribed a VKA were included from the 4 participating countries (Figure 1). Of these, sufficient data on INR to determine the TTR were available for 6250 patients (19.5%), who thus constituted the analyzable population.
63,244 patients with atrial fibrillation
32,075 patients prescribed VKA
6250 patients analyzable
Incidence of Stroke and Bleeding Occurrence of first stroke and bleeding was defined as the first recording of a diagnosis of stroke in the LPD database after the index date. The index date was defined as the date of the first consultation during the study period for patients with a prior diagnosis of AF and the date of the consultation at which a diagnosis of AF was made for newly diagnosed patients. The observation period extended from the index date to October 31, 2012.
1162
France n = 1110
Germany n = 1208
Italy n = 1230
UK n = 2702
Figure 1. Patient enrollment and anticoagulant prophylaxis. VKA ¼ vitamin K antagonist.
Volume 36 Number 9
September 2014 Table I. Demographic characteristics and stroke risk factors in patients with nonvalvular atrial fibrillation at the index consultation treated with vitamin K antagonists (VKAs). France (N ¼ 16,447) Characteristic Age, y Mean (SD) Median, range o75 y Z75 y Sex Male Female Medical sroke risk factors Diabetes Arterial hypertension Coronary heart disease Heart failure Vascular disease History of stroke or TIA CHA2DS2-VASc score* Mean (SD Median [range] Z1 Z2
All VKA (n ¼ 9184)
74.9 77 3766 5418
(10.5) [18–103] (41.0%) (59.0%)
5714 (62.2%) 3470 (37.8%)
Germany (N ¼ 13,917)
Analyzable (n ¼ 1110)
74.4 76 480 630
(9.6) [20–94] (43.2%) (56.8%)
636 (57.3%) 474 (42.7%)
All VKA (n ¼ 6386)
73.9 75 3059 3327
(9.5) [22–100] (47.9%) (52.1%)
3460 (54.2%) 2926 (45.8%)
Italy (N ¼ 22,447)
Analyzable (n ¼ 1208)
71.7 73 681 527
(9.5) [37–95] (56.4%) (43.6%)
622 (51.5%) 586 (48.5%)
All VKA (n ¼ 8774)
76.2 78 3160 5614
(9.7) [21–112] (36.0%) (64.0%)
United Kingdom (N ¼ 10,433)
Analyzable (n ¼ 1230)
74.4 75 571 659
(8.4) [23–111] (46.4%) (53.6%)
All VKA (n ¼ 7731)
74.8 76 3305 4426
(10.5) [18–102] (42.7%) (57.3%)
4456 (50.8%) 4318 (49.2%)
580 (47.2%) 650 (52.8%)
4434 (57.4%) 3297 (42.6%)
Analyzable (n ¼ 2702)
72.1 73 1520 1182
(9.7) [29–98] (56.3%) (43.7%)
1527 (56.5%) 1175 (43.5%)
2090 6502 1437 1454 1178 1042
(22.8%) (70.8%) (15.6%) (15.8%) (12.8%) (11.3%)
280 808 201 1262 806 998
(25.2%) (72.8%) (18.1%) (20.2%) (12.9%) (16.0%)
2724 5316 2834 1966 1239 998
(42.7%) (83.2%) (44.4%) (30.8%) (19.4%) (15.6%)
467 1008 493 340 236 200
(38.7%) (83.4%) (40.8%) (28.1%) (19.5%) (16.6%)
2263 6395 1955 1730 947 1118
(25.8%) (72.9%) (22.3%) (19.7%) (10.8%) (12.7%)
359 935 290 262 153 181
(29.2%) (76.0%) (23.6%) (21.3%) (12.4%) (14.7%)
1154 2255 1696 1339 724 1245
(14.9%) (29.2%) (21.9%) (17.3%) (9.4%) (16.1%)
410 796 545 438 246 463
(15.2%) (29.5%) (20.2%) (16.2%) (9.1%) (17.1%)
2.7 3 8323 6912
(1.6) [0–9] (90.6%) (75.3%)
3.3 3 1065 950
(1.6) [0–8] (95.9%) (85.6%)
3.1 3 5994 5090
(1.8) [0–9] (93.9%) (79.7%)
3.0 3 1118 953
(1.8) [0–9] (92.5%) (78.9%)
3.0 3 8246 7057
(1.7) [0–9] (94.0%) (80.4%)
3.5 3 1087 143
(1.6) [0–9] (88.4%) (11.6%)
2.3 2 6702 5154
(1.5) [0–9] (86.7%) (66.7%)
2.5 2 2449 1970
(1.5) [0–8] (90.6%) (72.9%)
1163
F-E. Cotte´ et al.
TIA ¼ transient ischemic attack. CHA2DS2-VASc score : 2 points for a history of stroke or TIA and age >75years, and 1 point for age between 65 and 74 years, hypertension, diabetes mellitus, a recent cardiac failure, vascular disease and female sex. * At the time of diagnosis of atrial fibrillation or the time that treatment was initiated, whichever came later.
Clinical Therapeutics Demographic data for all patients receiving VKAs, as well as for the 6250 patients in the analyzable population, are provided in Table I, together with information on the CHA2DS2-VASc stroke risk scores. The median CHA2DS2-VASc score was 3 in all countries except the United Kingdom (median, 2). Some differences were observed between the analyzable population and all the study patients prescribed a VKA. In all countries, the analyzable population tended to be younger, with a higher proportion of women.
VKA Treatment and Monitoring The most widely prescribed VKAs were warfarin in Italy and the United Kingdom, fluindione in France, and phenprocoumon in Germany (Table II). The median duration of treatment was 4 years in all countries except the United Kingdom (5 years). On average, the INR was determined at least 9 times per year in all countries. The mean TTR ranged from 66.0% in Italy to 72.6% in the United Kingdom. In France, Germany, and Italy, o50% of the patients were considered to have wellcontrolled VKA treatment (TTR 470%) and up to 20% spent o50% of their time outside the therapeutic range. The proportion of patients spending more TTR was higher in the United Kingdom.
TTR and Stroke and Bleeding Risk The association between the time spent in the therapeutic range and the incidence of stroke and hemorrhage was investigated (Figure 2). To optimize the power of the analysis, data were pooled across the 4 participating countries. Overall, 105 ischemic strokes were documented for 14,836 patient-years of VKA prescription. These strokes were more frequent in patients who had a TTR r70% (1.0 stroke/100 person-years) than in patients whose coagulation status was well controlled (0.5 stroke/100 personyears). The unadjusted odds ratio for stroke was 1.60 (95% CI, 1.08–2.36; P ¼ 0.019). After adjustment for age, sex, treatment duration, and CHA2DS2-Vasc score, the adjusted odds ratio was 1.38 (95% CI, 0.93 – 2.06), which was no longer significant (P ¼ 0.110). In terms of hemorrhage, 291 bleeding events were documented, including 55 gastrointestinal hemorrhages and 24 hemorrhagic strokes. The frequency of events was somewhat lower in patients who had well-controlled VKA treatment (1.1 vs 1.3 events/100 person-years). No significant association was observed between hemorrhage rate and TTR, either in the unadjusted analysis (odds ratio, 1.00 [95% CI, 0.79–1.27]; P o 0.098) or the adjusted analysis (odds ratio, 0.91 [95% CI, 0.72–1.16]; P ¼ 0.45).
Table II. Prescription of vitamin K antagonists (VKAs). Variable Individual VKA* Warfarin Acenocoumarol Fluindione Phenprocoumon Phenindione Duration of VKA treatment, y Median [range] Determination of INR Mean no. of measures/patient Mean no. of measures/patient/year Median no. of measures/patient/year Time spent in therapeutic range, % Mean (SD) INR in therapeutic range o50% of the time 50%–70% of the time 470% of the time
France (n ¼ 1110)
Germany (n ¼ 1208)
Italy (n ¼ 1230)
97 (8.7%) 68 (6.1%) 945 (85.1%) — —
7 (0.6%) — — 1201 (99.4%) —
1091 (88.7%) 139 (11.3%) — — —
4 [1–13] 37.5 9.61 9.14 66.2
(32.4) (5.86) [1.0–39.0] (21.1)
209 (18.8%) 370 (33.3%) 531 (47.8%)
4 [1–10] 38.4 9.48 9.25 65.3
(30.0) (4.49) [1.5–23.0] (19.6)
232 (19.2%) 441 (36.5%) 535 (44.2%)
4 [1–14] 49.0 9.78 10.0 66.0
United Kingdom (n ¼ 2702)
2697 (99.8%) 2 (0.1%) — — 3 (0.1%) 5 [1–13]
(40.7) (5.40) [1.0–42.7] (18.4)
192 (15.6%) 471 (38.3%) 567 (46.1%)
74.7 13.07 12.36 72.6
(48.8) (5.46) [1.5–51.5] (13.7)
156 (5.8%) 779 (28.8%) 1767 (65.4%)
INR ¼ international normalized ratio. * Because certain patients were prescribed 41 VKA, these treatments are not mutually exclusive.
1164
Volume 36 Number 9
F-E. Cotte´ et al.
Frequency (Events/100 Person-Years)
2.0
Stroke
Hemorrage
1.5
1.0
0.5
0
≤70%
>70% ≤70% Time in Therapeutic Range
>70%
Figure 2. Rate of occurrence of stroke or bleeding as a function of time spent in therapeutic range. Data are presented as mean values with their 95% CIs.
DISCUSSION This study evaluated patients with nonvalvular AF treated with VKA in everyday care in 4 European countries. We observed that a majority of the patients in France (52%), Germany (56%), and Italy (54%) had a TTR r70%. The situation in the United Kingdom was somewhat better, with approximately one third of patients presenting with a TTR o70% (35%). Compared with the other countries, patients with AF in the United Kingdom differed in having a lower stroke risk, more frequently evaluated INR, and a higher TTR. These differences may be explained by the care paradigm in the United Kingdom: such patients are managed by physicians in specialized “warfarin clinics,” where INR is frequently controlled and the dose of VKA is adapted. It is also noteworthy that different VKAs were used in different countries: warfarin accounted for the large majority of VKA use in Italy and the United Kingdom, phenprocoumon in Germany, and fluindione in France. Comparative data on the effectiveness of these VKAs in everyday practice are extremely limited, and it is possible that differences in the VKAs prescribed may contribute to some of the between-country variation in TTR observed. The relation between TTR and clinical outcome was also evaluated, although it should be noted that this analysis was purely exploratory, and the study was neither designed nor powered to address this factor.
September 2014
When the data were pooled across countries, we observed a higher frequency of stroke in these poorly controlled patients (1.0 stroke/100 person-years) compared with those with a TTR 470% (0.5 stroke/100 person-years). However, when adjusted on stroke risk factors, the odds ratio did not reach statistical significance. Sample attrition may have provided less power to our analysis, and the observed trend should be evaluated in a larger population. The absence of a similar trend for hemorrhage is surprising but could be explained by the fact that patients with uncontrolled VKA treatment may be more often below the intended INR therapeutic range than above. This phenomenon has been observed previously39 and may reflect a tendency of GPs to underdose VKAs because of possible adverse effects related to concerns about VKA-associated hemorrhage. This study has several limitations, principally linked to the use of electronic general practice databases. These databases provide information on a large number of patients gathered in real-life conditions by a representative national sample of GPs. However, because the data are collected in everyday clinical practice, their availability depends on the assiduity of the GP in using the medical software. Whereas basic data on patient characteristics, diseases, and drug prescriptions are generally collected comprehensively by all GPs, information on paraclinical tests such as INR assays may only be documented in the electronic database by some GPs some of the time. Calculation of the TTR requires regular assay and documentation of the INR, which could explain why only 20% of the patients in our survey were considered analyzable in the 4 countries considered. The low proportion of analyzable patients may have introduced some inclusion bias with respect to all patients with AF who consult their GP. Indeed, in all 4 countries, the analyzable populations tended to be younger and more frequently women, although their stroke risks were similar to those of the entire study sample. Generally speaking, the characteristics of the AF patients evaluated were consistent with previous findings from real-life cohorts in France,40 Germany,41 and Italy.42 With such databases, it is impossible to ensure that all medical events are exhaustively documented, particularly if they are not the specific objective of the GP consultation. Thus, it is useful to compare the event rates documented in our study with those reported in dedicated patient cohorts in real-life conditions of care. For example, in a cohort of 3302 patients with AF in Italy,39 the rates of stroke (0.76 event/100 patient-years)
1165
Clinical Therapeutics and bleeding events (1.24 events/100 patient-years) were similar to those reported in the present study. Conversely, in the United Kingdom and France, although the incidence of stroke is consistent with what has been observed previously in real-life studies (0.9 and 1.2 events/100 patient-years, respectively), the incidence of hemorrhage in the present study appeared to be lower than expected from these earlier studies (3.8 and 3.2 events/100 patient-years, respectively).43,44 This discrepancy could be attributable, at least in part, to the fact that bleeding events may be managed elsewhere than in primary care, and events that are not serious or without consequences may go undocumented.
2.
3.
4.
5.
CONCLUSIONS This study of the management of AF in primary care in 4 European countries indicates that many patients receiving VKA prophylaxis did not achieve their target coagulation status, with their INR frequently outside the therapeutic range, and are thus exposed to an unnecessary risk of stroke or bleeding complications. A trend toward a higher incidence of stroke in patients who fail to achieve target levels was observed.
6.
7.
8.
ACKNOWLEDGMENTS Funding for the study was provided by Bristol-Myers Squibb and Pfizer, purveyors of apixaban, which is indicated for stroke prevention in patients with AF.
CONFLICTS OF INTEREST Dr. Cotté, Mr. Benhaddi and Dr. Duprat-Lomon are employees of Bristol-Myers Squibb and Mr. Marchant is an employee of Pfizer. Drs. Letierce and Mr. Huguet are employees of Cegedim, a contract research organization that was mandated by Bristol-Myers Squibb and Pfizer to implement the REACT AF study, for which they received financial compensation. Dr. Doble is an employee of Foxymed, a medical communication and consultancy company which participated in the exploration and interpretation of the results of the REACT AF study on behalf of Bristol-Myers Squibb and Pfizer and prepared the manuscript for publication. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
9.
10.
11.
12.
13.
14.
15.
REFERENCES 1. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the
1166
16.
AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA. 2001;285:2370–2375. Wilke T, Groth A, Mueller S, et al. Incidence and prevalence of atrial fibrillation: an analysis based on 8.3 million patients. Eur Soc Cardiol. 2013;15:486–493. Kannel WB, Wolf PA, Benjamin EJ, et al. Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: population-based estimates. Am J Cardiol. 1998;82:2N–9N. Camm AJ, Kirchhof P, Lip GY, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J. 2010;31:2369–2429. Charlemagne A, Blacher J, Cohen A, et al. Epidemiology of atrial fibrillation in France: extrapolation of international epidemiological data to France and analysis of French hospitalization data. Arch Cardiovasc Dis. 2011;104:115–124. Meinertz T, Kirch W, Rosin L, et al. Management of atrial fibrillation by primary care physicians in Germany: baseline results of the ATRIUM registry. Clin Res Cardiol. 2011; 100:897–905. Piccinocchi G, Laringe M, Guillaro B, et al. Diagnosis and management of atrial fibrillation by primary care physicians in Italy: a retrospective, observational analysis. Clin Drug Invest. 2012;32:771–777. Stewart S, Hart CL, Hole DJ, et al. Population prevalence, incidence, and predictors of atrial fibrillation in the Renfrew/Paisley study. Heart. 2001;86:516–521. Sudlow M, Thomson R, Thwaites B, et al. Prevalence of atrial fibrillation and eligibility for anticoagulants in the community. Lancet. 1998;352:1167–1171. Benjamin EJ, Wolf PA, D’Agostino RB, et al. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998;98:946–952. McManus DD, Rienstra M, Benjamin EJ. An update on the prognosis of patients with atrial fibrillation. Circulation. 2012;126:e143–e146. Tendera M, Syzdol M, Parma Z. ARISTOTLE RE-LYs on the ROCKET. What’s new in stroke prevention in patients with atrial fibrillation? Cardiol J. 2012;19:4–10. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22:983–988. Singer DE, Albers GW, Dalen JE, et al. Antithrombotic therapy in atrial fibrillation: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6 Suppl):546S–592S. Marini C, De Santis F, Sacco S, et al. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: results from a population-based study. Stroke. 2005;36:1115–1119. Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial
Volume 36 Number 9
F-E. Cotte´ et al.
17.
18.
19.
20.
21.
22.
23.
fibrillation. Arch Intern Med. 1994; 154:1449–1457. You JJ, Singer DE, Howard PA, et al. Antithrombotic therapy for atrial fibrillation: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):e531S–e5375. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA. 2001;285: 2864–2870. Camm AJ, Lip GY, De Caterina R, et al. 2012 Focused update of the ESC guidelines for the management of atrial fibrillation: an update of the 2010 ESC guidelines for the management of atrial fibrillation. *Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J. 2012;33:2719–2747. Olesen JB, Torp-Pedersen C, Hansen ML, et al. The value of the CHA2DS2VASc score for refining stroke risk stratification in patients with atrial fibrillation with a CHADS2 score 0-1: a nationwide cohort study. Thromb Haemost. 2012;107:1172–1179. Gorin L, Fauchier L, Nonin E, et al. Prognosis and guideline-adherent antithrombotic treatment in patients with atrial fibrillation and atrial flutter: implications of undertreatment and overtreatment in real-life clinical practice; the Loire Valley Atrial Fibrillation Project. Chest. 2011;140:911–917. Lin LJ, Cheng MH, Lee CH, et al. Compliance with antithrombotic prescribing guidelines for patients with atrial fibrillation—a nationwide descriptive study in Taiwan. Clin Ther. 2008;30:1726–1736. Nieuwlaat R, Olsson SB, Lip GY, et al. Guideline-adherent antithrombotic treatment is associated with improved outcomes compared with undertreatment in high-risk patients with atrial fibrillation. The Euro Heart Survey on Atrial
September 2014
24.
25.
26.
27.
28.
29.
30.
31.
32.
Fibrillation. Am Heart J. 2007;153: 1006–1012. Fang MC, Go AS, Chang Y, et al. Death and disability from warfarinassociated intracranial and extracranial hemorrhages. Am J Med. 2007;120: 700–705. Mohr JP, Thompson JL, Lazar RM, et al. A comparison of warfarin and aspirin for the prevention of recurrent ischemic stroke. N Engl J Med. 2001;345:1444–1451. Gallego P, Vilchez JA, Lane DA. Apixaban compared with warfarin for stroke prevention in atrial fibrillation: implications of time in therapeutic range. Circulation. 2013;127: 2163–2165. De Caterina R, Husted S, Wallentin L, et al. General mechanisms of coagulation and targets of anticoagulants (Section I). Position Paper of the ESC Working Group on Thrombosis—Task Force on Anticoagulants in Heart Disease. Thromb Haemost. 2013;109:569–579. Gallagher AM, Setakis E, Plumb JM, et al. Risks of stroke and mortality associated with suboptimal anticoagulation in atrial fibrillation patients. Thromb Haemost. 2011;106: 968–977. Phillips KW, Ansell J. Outpatient management of oral vitamin K antagonist therapy: defining and measuring high-quality management. Exp Rev Cardiovasc Ther. 2008;6:57–70. Pereira de Sousa L, Burba I, Ruperto C, et al. Vitamin K antagonists in patients with nonvalvular atrial fibrillation: appropriateness and quality of treatment in an Italian cohort. J Cardiovasc Med (Hagerstown). 2013;14:534–540. Wan Y, Heneghan C, Perera R, et al. Anticoagulation control and prediction of adverse events in patients with atrial fibrillation: a systematic review. Circulation Cardiovascular Quality and Outcomes. 2008;1:84–91. Bouee S, Charlemagne A, Fagnani F, et al. Changes in osteoarthritis management by general practitioners in the COX2-inhibitor era-concomitant
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
gastroprotective therapy. Joint Bone Spine. 2004;71:214–220. Cotte FE, Fardellone P, Mercier F, et al. Adherence to monthly and weekly oral bisphosphonates in women with osteoporosis. Osteoporos Int. 2010;21:145–155. Cea Soriano L, Bateman BT, Garcia Rodriguez LA, et al. Prescription of antihypertensive medications during pregnancy in the UK. Pharmacoepidemiol Drug Safe. 2014. Mazzaglia G, Filippi A, Alacqua M, et al. A national survey of the management of atrial fibrillation with antithrombotic drugs in Italian primary care. Thromb Haemost. 2010;103:968–975. Sabouret P, Depret-Bixio L, Cotte FE, et al. Sex differences in stroke prevention in atrial fibrillation in French primary care. Results of the AFIGP (Atrial Fibrillation In General Practice) Database. Clin Res Cardiol. 2014. Rosendaal FR, Cannegieter SC, van der Meer FJ, et al. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost. 1993;69:236–239. Lader E, Martin N, Cohen G, et al. Warfarin therapeutic monitoring: is 70% time in the therapeutic range the best we can do? J Clin Pharm Ther. 2012;37:375–377. Poli D, Testa S, Antonucci E, et al. Bleeding and stroke risk in a realworld prospective primary prevention cohort of patients with atrial fibrillation. Chest. 2011;140:918–924. Cohen A, Dallongeville J, DurandZaleski I, et al. Characteristics and management of outpatients with history of or current atrial fibrillation: the observational French EPHA study. Arch Cardiovasc Dis. 2010;103:376–387. Reinhold T, Rosenfeld S, MullerRiemenschneider F, et al. [Patients suffering from atrial fibrillation in Germany. Characteristics, resource consumption and costs]. Herz. 2012;37: 534–542. Zoni-Berisso M, Filippi A, Landolina M, et al. Frequency, patient characteristics, treatment strategies, and
1167
Clinical Therapeutics resource usage of atrial fibrillation (from the Italian Survey of Atrial Fibrillation Management [ISAF] study). Am J Cardiol. 2013;111:705–711. 43. Gallagher AM, van Staa TP, MurrayThomas T, et al. Population-based cohort study of warfarin-treated patients with atrial fibrillation: incidence of cardiovascular and bleeding outcomes. BMJ Open. 2014;4:e003839. 44. Cotte FE, Chaize G, Kachaner I, et al. Incidence and cost of stroke and hemorrhage in patients diagnosed with atrial fibrillation in France. J Stroke Cerebrovasc Dis. 2014; 23:e73–e83.
Address correspondence to: François-Emery Cotté, Bristol-Myers Squibb, 3, rue Joseph Monier, 92500 RUEIL MALMAISON, France. E-mail: [email protected]
1168
Volume 36 Number 9
