[
Original Research Cardiovascular Disease
]
Stroke/Thromboembolism and Intracranial Hemorrhage in a Real-world Atrial Fibrillation Population The Complications of Atrial Fibrillation in the Bologna Area (CAFBO) Study Gualtiero Palareti, MD; Luisa Salomone, MD; Mario Cavazza, MD; Marcello Guidi, MD; Antonio Muscari, MD; Giuseppe Boriani, MD; Antonio Di Micoli, MD; Giordano Guizzardi, MD; Gaetano Procaccianti, MD; Angelo Guidetti, MD; Nicola Binetti, MD; Simona Malservisi, MD; Marco Masina, MD, Antonella Viola, MD; Vincenzo Bua, MD; Maurizio Ongari, MD; Giampaolo Diaspri, MD; and Gregory Y. H. Lip, MD
Ischemic events (IEs) and intracranial hemorrhages (ICHs) are feared complications of atrial fibrillation (AF) and of antithrombotic treatment in patients with these conditions.
BACKGROUND:
Patients with AF admitted to the EDs of the Bologna, Italy, area with acute IE or ICH were prospectively recorded over 6 months.
METHODS:
A total of 178 patients (60 male patients; median age: 85 years) presented with acute IE. Antithrombotic therapy was as follows: (1) vitamin K antagonists (VKAs) in 31 patients (17.4%), with international normalized ratio (INR) at admission of , 2.0 in 16 patients, 2.0 to 3.0 in 13 patients, and . 3.0 in two patients; (2) aspirin (acetylsalicylic acid) (ASA) in 107 patients (60.1%); and (3) no treatment in 40 patients (22.5%), mainly because AF was not diagnosed. Twenty patients (eight male patients; median age: 82 years) presented with acute ICH: 13 (65%) received VKAs (INR, 2.0-3.0 in 11 patients and . 3.0 in two patients), while six (30%) received ASA. Most IEs (88%) and ICHs (95%) occurred in patients aged . 70 years. A modeling analysis of patients aged . 70 years was used to estimate annual incidence in subjects anticoagulated with VKAs in our Network of Anticoagulation Centers (NACs), or those expected to have AF but not included in NACs. The expected incidence of IE was 12.0%/y (95% CI, 10.7-13.3) in non-NACs and 0.57%/y (95% CI, 0.42-0.76) in NACs (absolute risk reduction [ARR], 11.4%/y; relative risk reduction [RRR], 95%; P , .0001). The incidence of ICH was 0.63%/y (95% CI, 0.34-1.04) and 0.30%/y (95% CI, 0.19-0.44), respectively (ARR, 0.33%/y; RRR, 52.4%/y; P 5 .04).
RESULTS:
IEs occurred mainly in elderly patients who received ASA or no treatment. One-half of patients with IEs receiving anticoagulant treatment had subtherapeutic INRs. Therapeutic approaches to elderly subjects with AF require an effective anticoagulant treatment strategy. CONCLUSIONS:
CHEST 2014; 146(4):1073-1080
Manuscript received October 21, 2013; revision accepted March 14, 2014; originally published Online First May 8, 2014. ABBREVIATIONS: AF 5 atrial fibrillation; ARR 5 absolute risk reduction; ASA 5 aspirin (acetylsalicylic acid); GP 5 general practitioner; ICH 5 intracranial hemorrhage; IE 5 ischemic event; INR 5 international normalized ratio; LMWH 5 low-molecular-weight heparin;
journal.publications.chestnet.org
NACs 5 Network of Anticoagulation Centers; RRR 5 relative risk reduction; VKA 5 vitamin K antagonist AFFILIATIONS: From the Angiologia e Malattie della Coagulazione (Drs Palareti and Salomone), Medicina D’Urgenza e Pronto Soccorso (Drs Cavazza and Guidi), Stroke Unit, Medicina Interna (Dr Muscari), and Institute of Cardiology (Dr Boriani), Department of Experimental,
1073
Ischemic stroke and systemic embolism are the most feared complications of atrial fibrillation (AF), a condition whose prevalence sharply increases with age.1 One in five of all strokes is attributed to AF, with a risk for stroke that increases with age, reaching values of 23.5% in patients aged ⱖ 80 years.2 Furthermore, AF-related strokes are more often fatal and more disabling than those due to other causes.3 Anticoagulant treatment with vitamin K antagonists (VKAs) is extremely effective in reducing the risk for AF-associated stroke, but is still underused in clinical practice4 or not adequately monitored with subsequent insufficient protection.5 Many patients with AF receive aspirin (acetylsalicylic acid) (ASA), though its efficacy is
Materials and Methods Setting The area of Bologna, in Northern Italy, has an incident population of 866,160 (735,960 aged ⱖ 18 years) and covers an area of 2,915 km2 with 50 municipalities. Population-based epidemiologic studies are possible in this setting because medical care is completely and uniformly provided by (1) a network of hospitals comprising two major facilities—Ospedale Maggiore and the University Hospital, both in the city of Bologna—and other middle-size hospitals distributed in the area and (2) general practitioners (GPs). These health services are part of the regional health system. A major ED is present in each of two major hospitals in Bologna; both are approved for thrombolysis in acute stroke. A network of EDs is active in the other hospitals. In general, patients with acute stroke are urgently referred to one of the two largest EDs, so that thrombolytic treatment can be provided, if indicated. Monitoring of patients treated by oral anticoagulants (at the time of the study, only VKAs were used) is performed by a Network of Anticoagulation Centers (NACs), covering all the Bologna area, which use the same computerized system to help patient management and dosing (P.A.R.M.A.; Werfen Group). Most subjects receiving anticoagulant
Diagnostic and Specialty Medicine, University of Bologna, Sant’OrsolaMalpighi University Hospital, Bologna, Italy; Pronto Soccorso e Medicina D’Urgenza Ospedale Maggiore (Drs Di Micoli and Guizzardi), Bologna, Italy; IRCCS Institute of Neurological Sciences, Maggiore Hospital Stroke Unit (Dr Procaccianti), Bologna, Italy; Pronto Soccorso ed Emergenza Territoriale Area Nord (Dr Guidetti and Binetti), Bologna, Italy; Stroke Care, Geriatria, Ospedale di Bentivoglio (Drs Malservisi and Masina), Bologna, Italy; Pronto Soccorso ed Emergenza Territoriale Area Sud (Drs Viola and Bua), Bologna, Italy; UOC Medicina PorrettaVergato (Drs Ongari and Diaspri), Bologna, Italy; and University of Birmingham Centre for Cardiovascular Sciences (Dr Lip), City Hospital, Birmingham, England. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study. CORRESPONDENCE TO: Gualtiero Palareti, MD, Department of Angiology and Blood Coagulation, University Hospital Policlinico S. OrsolaMalpighi, Via Albertoni 15, 40138 Bologna, Italy; e-mail: gualtiero. [email protected] © 2014 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.13-2443
1074 Original Research
limited,6,7 while many others remain completely untreated. Nonadherence to guidelines and undertreatment in clinical practice are independently associated with a high risk for stroke and mortality.8 This observational study prospectively recorded all the patients with AF who, during a 6-month period, were consecutively admitted with an acute ischemic event (IE) (ie, stroke or systemic embolism) or intracranial hemorrhage (ICH) to the EDs of the area of Bologna, Italy, and examined the type (if any) and quality of treatment they were receiving. Using a specific modeling analysis, we estimated treatment outcomes in patients with AF aged . 70 years living in that area.
treatment living in the area are included in one of the NACs; however, some patients, prospectively very few, are referred to their GP for anticoagulation management. Design and Study Population All the EDs in the area agreed to participate in the study. At least one doctor in each participating clinical center was identified as point person for the study, and a doctor (L. S.), active in the Angiology and Blood Coagulation unit (coordinating center), was in charge of monitoring the study. The institutional review boards of all participating centers approved the study: Azienda Ospedaliero-Universitaria di Bologna Policlinico S.Orsola-Malpighi (110/2011/U/Oss); Azienda AUSL c/o Dipartimento Farmaceutico Ospedale Maggiore di Bologna (11052). Written informed consent for participation was obtained from the patients or from relatives when a patient was unable to give it. From January 22, 2012, to July 22, 2012 (183 days), all patients admitted to the EDs with IE or ICH were included in the study if AF was present at admission, there was a known diagnosis of paroxysmal AF preadmission, or AF was detected after hospitalization. Patients were excluded if the event was more likely attributable to other conditions than AF. The included patients were identified from the general database of the NACs, and results of anticoagulation control during the previous 4 months were analyzed and evaluated as “stable” patients with an international normalized ratio (INR) between 1.8 and 3.2 (with no variations in VKA doses), and “variable” in other cases. Statistical Analysis Differences between groups were assessed by the x2 test with Yates’ correction for categorical variables and by the Mann-Whitney test for continuous variables. The data were analyzed with Prism software version 3.0 (GraphPad Software Inc), and SPSS software version 11.0 (IBM). Specific modeling analyses were designed to calculate separately the incidence of IE or ICH in subjects with AF who were aged . 70 years included or not in NACs. The number of inhabitants in the Bologna area aged . 70 years was obtained from 2012-related data of the regional health system; the total numbers of subjects with AF in this population was calculated by estimating, in-line with more recent population studies,9,10 an AF prevalence of 7%. A rough estimate of elderly patients with AF with unknown treatment was obtained by subtracting the number of patients with AF included in NACs from the total calculated elderly population with AF. The modeling analysis was performed at the end of the 6-month study period.
[
146#4 CHEST OCTOBER 2014
]
Results During the 6-month period, 178 patients with AF (60 male patients; median age, 85 years [range, 58-99 years]) with acute IE and 20 patients (eight male patients; median age, 82 years [range, 62-96 years]) with acute ICH were included in the study, while 16 patients with IE and one patient with ICH were excluded because the events were more likely to be attributed to causes other than AF (Table 1). IEs occurred more frequently in women than in men (66.3% vs 33.7%, respectively; P , .0001). Table 2 summarizes characteristics of included patients and details of their antithrombotic treatment. In 16 of the 31 patients (51.6%) treated with VKAs (three had a temporary interruption of treatment), the INR was , 2.0 at the event, while in 13 (41.9%), the INR was within the therapeutic range. Overall, the quality of treatment in the previous 4 months was stable in one-
TABLE 1
half of the patients treated with VKAs and followed-up in NACs (n 5 12); seven patients were receiving VKAs but were not followed in NACs. Among the patients who presented with IE, 147 (82.6%) were treated with ASA or did not receive any antithrombotic therapy; some (n 5 32, 18.0%) had formerly been treated with VKAs, but treatment had been interrupted permanently for various reasons (Table 2) (Fig 1). Of the 20 patients who presented with ICH (Table 2), 13 (65%) were receiving VKA treatment and one was taking concomitant low-dose ASA. Of these, 11 had an INR result in the therapeutic range, and two had INR values . 3.0. Anticoagulation quality in the previous 4 months was stable in 10 cases (76.9%) and variable in only one. The quality of anticoagulation could not be assessed in one patient, whose VKA therapy was managed by the GP, and in another case in which the ICH occurred within the first 20 days of treatment. Six patients (30%)
] Patient Characteristics and Outcomes
Patients
Ischemic Events
Intracranial Hemorrhages
194
21
Mechanical valve prosthesis
4
…
Peripheral artery disease
3
…
Severe carotid stenosis
4
…
5
…
Screened, No. Excluded for, No.
Different final diagnosis Bleeding within a brain tumor Included, No. Age, median (range), y Male sexa Stroke Peripheral arterial emboli
… 178 85 (58-99) 60 (33.7)
1 20 82 (62-96) 8 (40)
159 (89.3)
…
19 (10.7)
…
Type of AF, No. (%; No. male patients) Chronic
84 (47.2; 29)
13 (65; 5)
Paroxysmal
61 (34.3; 16)
4 (20; 1)
New onset
21 (11.8; 9)
3 (15; 2)
Diagnosed during hospitalization
12 (6.7; 6)
…
Age distribution of patients and type of the events, No. (%; No. male patients); [stroke/peripheral arterial emboli), y ⱕ 70
21 (11.8; 13) [20/1]
71-80
40 (22.5; 19) [34/6]
8 (40; 4)
81-90
82 (46.1; 25) [74/8]
10 (50; 3)
. 90
35 (19.6; 3) [31/4]
In-hospital death
34 (19.1)
1 (5; 1)
1(5; 0) 9 (45.0)
Data given as No. (%) unless otherwise indicated. AF 5 atrial fibrillation. aP , .0001 vs female patients.
journal.publications.chestnet.org
1075
TABLE 2
] Patient Characteristics and Antithrombotic Treatment at Presentation With an Event
Characteristics Patients treated with VKAs, No. (%) 1 ASA
Ischemic Events (n 5 178) 31 (17.4)
Intracranial Hemorrhages (n 5 20) 13 (65)
1
1
CHA2DS2 VASc, median (range)
5 (2-9)
HAS-BLED, median (range)
3 (1-4)
VKA temporarily interrupted before the event, No. (% patients treated with VKA)
3 (9.7)
1a
Quality of VKA treatment INR value at the event 2-3 (in range)
13 (41.9)
, 2.0
16 (51.6)
. 3.0
2 (6.5)
b
11 (84.6) 0 2 (15.4)
Overall quality of treatment in the 4 previous mo Stable
12 (38.7)
10 (76.9)
Variable
12 (38.7)
1 1
NVc NA
d
Patients treated with antiplatelet drugs
7 (22.6) 107 (60.1)e
CHA2DS2 VASc, median (range)
5 (0-8)
HAS-BLED, median (range)
2 (0-5)
LMWH (enoxaparin) Patients who did not receive any antithrombotic treatment CHA2DS2 VASc, median (range) HAS-BLED, median (range) Patients formerly treated with VKAs who had permanently stopped VKAs before the event
…
1 6 (30)f
1 (5)g
40 (22.5)
0
4 (1-7) 2 (1-4) 32 (18.0)
0
Cause for permanent interruption, No. Bleeding
18
Cardioversion
5
Risk for fall
5
Patient’s decision
4
Antithrombotic treatment at the event, No. Antiplatelet drugs No treatment
25 7h
Data given as No. (%) unless otherwise indicated. ASA 5 aspirin (acetylsalicylic acid); CHA2DS2-VASc 5 congestive heart failure, hypertension, age ⱖ 75 y (2 points), diabetes mellitus, stroke (2 points), vascular disease, age 65 to 74 y, and sex category (female); HAS-BLED 5 hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly (. 65 y), drugs/alcohol concomitantly; INR 5 international normalized ratio; LMWH 5 low-molecular-weight heparin; NA 5 not applicable; NV 5 not valuable; VKA 5 vitamin K antagonist. aTreated with LMWH. bIn three of these patients, the INR was low because the VKA therapy was temporarily interrupted. cNot valuable because intracranial hemorrhage occurred during the first 20 d from beginning of VKA treatment; not applicable. dNot available because these patients were receiving VKA at the moment of the event but were not included in the Network of Anticoagulation Centers and, therefore, no information on the quality of anticoagulation control was available. eASA (n 5 81), thienopyridine (n 5 22), ASA plus clopidogrel (n 5 4). fASA (n 5 5), thienopyridine (n 5 1). gLMWH was given at therapeutic doses. hFour patients were receiving prophylactic doses of LMWH.
1076 Original Research
[
146#4 CHEST OCTOBER 2014
]
a diagnosis of acute IE or ICH attributable to AF. The type of antithrombotic treatment (if any) and the quality of anticoagulation monitoring was evaluated. This was possible due to the presence of NACs in the area that use the same general rules and the same VKA-dedicated electronic software that manage the vast majority of patients treated with VKAs.
Figure 1 – Different antithrombotic treatments in relation to age.
with ICH were receiving ASA, and one was treated with a therapeutic dose of low-molecular-weight heparin (LMWH). Modeling Analysis of Patients Aged . 70 Years
Very few events occurred in patients aged ⱕ 70 years (12% of IEs and 5% of ICHs). Therefore, we focused the analysis on the events that occurred in patients aged . 70 years vs the calculated number of subjects aged . 70 years who had AF in the population of Bologna area, by estimating a prevalence of AF of 7% in this population9,10 (Table 3). While 78% of all those subjects were receiving anticoagulant treatment and monitored in NACs, no information was available on treatment (if any) in the remaining elderly subjects expected to have AF. The expected incidence of IEs in subjects not included in NACs was 12.0%/y (95% CI, 10.7-13.3) and 0.57%/y (95% CI, 0.42-0.76) in those monitored in NACs, with an absolute risk reduction (ARR) of 11.4%/y and a relative risk reduction (RRR) of 95%/y (P , .0001). The expected incidence of ICH was 0.63%/y (95% CI, 0.34-1.04) and 0.30%/y (95% CI, 0.19-0.44) in subjects not included or included in NACs, respectively, with an ARR of 0.33%/y and a RRR of 52.4%/y (P 5 .040).
Discussion In the present study, we provide observational data on antecedents and associated antithrombotic therapy in patients presenting with IEs or ICH in a defined, local, AF population, and modeled this to estimate events in the area of Bologna. The present study is unique, as we prospectively recorded and examined all the patients aged ⱖ 18 years who were referred during consecutive 6 months to all the EDs in the area of Bologna (northern Italy, with a population of 735,960 aged ⱖ 18 years) with
journal.publications.chestnet.org
Our results suggest that the current therapeutic approach to patients with AF is much improved compared with the results of a previous Italian study,11 in which almost one-half of patients with AF with a first stroke did not receive any antithrombotic treatment and consistent with reviews.12 Only 22.5% of these patients with IE were untreated at admission, and one-half had previously nondiagnosed AF. This emphasizes the importance of a systematic screening of the population at risk (especially in elderly subjects). The best cost-effectiveness strategy seems to be opportunistic case finding, in which a health-care professional can assess a patient’s pulse during a consultation9 or use a specifically modified BP monitor.13,14 These simple and effective approaches should be strongly recommended.15,16 Less than one-fifth of all IEs in our study occurred in patients who were being treated with VKAs, and in some, the treatment had been temporarily interrupted before the event. This finding confirms the potential impact of even short, temporary withdrawals of treatment on the risk of complications in patients with AF,17,18 which cannot be adequately offset by prophylactic treatment with LMWH. About one-half of IEs during VKA treatment were associated with subtherapeutic INR values at the event, while in one-third of the cases, anticoagulation quality was variable. These observations emphasize the need for high-quality VKA anticoagulation management and are in-line with studies showing that suboptimal anticoagulation control is associated with an increased risk for stroke and mortality in patients with AF.5 The majority of IEs in our study (60%) occurred in patients treated with antiplatelet drugs (mainly low-dose ASA). ASA is significantly less effective than warfarin for prevention of stroke in patients with AF, with no difference in major bleeding or ICH compared with warfarin.19 Unfortunately, ASA is still largely given to patients with AF, especially the elderly.20 A more recent review reported that . 50% of patients are treated with antiplatelet drugs in clinical practice.12 Also, a systematic review of physicians’ attitudes regarding anticoagulant treatment in patients with AF reported that the most
1077
TABLE 3
] Subset of Inhabitants in Bologna Area Aged . 70 y: Modeling Analysis of the Expected Number of Subjects Affected by AF and Number of Events
Populations Total inhabitant population in Bologna area in 2012 aged . 70 y, No. (No. male patients; %)
No. of Subjects and Events
RR, %/ya
148,160 (59,477; 40.1)
…
71-80 y
88,472 (38,995)
. 80 y
59,688 (20,482)
Expected number of subjects aged . 70 y with AF, calculated according an estimated prevalence of AF of 7% in this population
10,370 (4,160)b
…
Anticoagulated patients with AF aged . 70 y monitored by NACs in February 2012
8,130 (3,315)
…
Expected number of subjects with AF aged . 70 y not included in NACs (thus, not receiving anticoagulation, or monitored by general practitioners, or receiving antiplatelet drugs, or not treated with antithrombotics)
2,240 (900)b
…
Ischemic events recorded during the 6 mo of the study in patients with AF aged . 70 y Monitored in NACs Calculated incidence per 100 patient-y, %/y (95% CI)
23
…
0.57 (0.42-0.76)
ARR: 11.4
Not included in NACs, No.
134
Calculated incidence per 100 patient-y, %/y (95% CI)
12.0 (10.7-13.3)
RRR: 95 P , .0001
Intracranial hemorrhages recorded in patients with AF aged . 70 y Monitored in NACs Calculated incidence per 100 patient-y, %/y (95% CI)
12
…
0.30 (0.19-0.44)
ARR: 0.41
Not included in NACs Calculated incidence per 100 patient-y, %/y (95% CI)
8c 0.71 (0.41-1.16)
RRR: 57.7 P 5 .0103
Data given as total No. (No. male patients) unless otherwise indicated. ARR 5 absolute risk reduction; NACS 5 Network of Anticoagulation Centers; RR 5 risk reduction; RRR 5 relative risk reduction. See Table 1 and 2 legends for expansion of other abbreviations. aIn patients monitored in NACs vs not included in NACs. bThese figures are calculated on the basis of the number of inhabitants of the Bologna area aged . 70 y by estimating a prevalence of AF of 7% in this population, with 40.1% of male patients. cSix patients received ASA, one received a therapeutic dose of LMWH, and one patient was treated with VKAs but not monitored in NACs.
important barriers to prescription of VKA anticoagulation were as follows: increasing age (the most important), bleeding risk or previous bleeding, fall risk, comorbidities, and lack of compliance.21 A 2011 study also reported a low risk of bleeding in very old patients treated with VKA and managed in anticoagulation centers, thus, emphasizing that age by itself should not be considered a contraindication to treatment, providing that these more frail patients are monitored by specifically trained professionals.22 In contrast with available evidence, many physicians are under the misconception that VKA treatment is harmful in elderly patients with AF and prefer to prescribe them ASA as a treatment they consider effective (though probably less than warfarin), but at much lower risk of bleeding and much easier for the patients. In our study, 13 of 20 ICH events occurred in patients treated with VKA. In all but two cases (one at the beginning of therapy and one with an INR . 3.0), the tempo1078 Original Research
rally related INR value was within the therapeutic range. Indeed, the occurrence of ICH is more often linked to specific predisposing conditions of the patients and interaction with VKAs, than to a high anticoagulation intensity.23 Six patients suffered from an ICH while receiving ASA, again confirming that this treatment is not free from serious complications.24 Our modeling analyses enable us to have some insights regarding the treatment conditions and results in the everyday real life in aged patients with AF living in the area of Bologna. For our modeling analysis, the figures of subjects . 70 years old living in Bologna area reported in Table 3 are the “real world” of that area, and the estimated number of subjects with AF depends on the prevalence of AF that can be attributed to that population. The number of those patients treated with VKAs followed in the NACs has been obtained from the informatic database of NACs. It is not known how the patients with
[
146#4 CHEST OCTOBER 2014
]
AF who are not included in NACs were treated; some of them may have received VKAs via their GP. Indeed, among the patients receiving anticoagulant treatment with VKAs and monitored in the NACs, the annual incidence of ICHs was 0.30%, which is lower than the expected incidence. VKA treatment is associated with an increased risk for ICH,25 and it is well known that advanced age and intensity of anticoagulation are the most important risk factors.26 One population-based study of incidence of VKA-associated ICH in Finland27 showed that despite a nearly fourfold increase in the proportion of VKAtreated subjects, the annual incidence of ICHs decreased during the study period, proving that this major complication can partly be prevented by a more accurate management of treated patients. The patients managed in our NACs were well monitored, with good quality anticoagulation control, and this may explain the relatively low rate of ICHs and why only one ICH event was associated with high INR value. In contrast, the rate of ICH events occurring in the elderly patients who were calculated to have AF and not followed in NACs was substantially higher than expected (annual incidence of 0.71%). As a limitation, we do not know the proportion in this group of subjects who were receiving antiplatelet drugs or for whom VKAs were monitored by their GP, or those who were not treated at all. The risk of bleeding associated with ASA (and especially ICH) is not negligible.19,28 Indeed, our data indicate that therapeutic approaches to elderly subjects with AF different from a well-monitored VKA
journal.publications.chestnet.org
therapy were not only markedly less effective in preventing thromboembolism but also at least equally associated with the risk of ICH. Limitations
First, the IE and ICH events that may have led to death before admission to an ED were not included in this study. Second, a longer duration of the study (eg, 1 year instead than 6 months) would lead to more events and could possibly have led to more robust evidence. Third, the estimate of 7% of AF prevalence in our population aged . 70 years can be questioned, although it is broadly consistent with the epidemiology of this arrhythmia. Finally, it was impossible to assess the specific conditions of patients not followed in NACs, such as how many of them were treated with VKAs, or received antiplatelet treatments, or were not treated at all. In conclusion, our study shows that AF-associated IEs occurred mainly in elderly patients who received antiplatelet drugs or were not treated because AF was undiagnosed. In patients receiving anticoagulant treatment, one-half of the IEs were associated with subtherapeutic INR levels, while all but one ICH occurred with INR in the therapeutic range. Elderly subjects with AF should be treated with well-monitored VKA anticoagulation or possibly the novel oral anticoagulants, because antiplatelet drugs were markedly less effective and at least equally associated with the risk for ICH. Further improvement in detection and diagnosis of AF and its appropriate treatment in the at-risk population is recommended.
1079
Acknowledgments Author contributions: G. Palareti had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. G. Palareti contributed to the study design, wrote the first draft of the manuscript, and served as principal author; L. S. contributed to the study design, monitored the collection of the patients in the participating EDs, collected data on anticoagulant therapy, and evaluated the quality of anticoagulation in the treated patients; M. C., M. G., A. M., G. B., A. D. M., G. G., G. Procaccianti, A. G., N. B., S. M., M. M., A. V., V. B., M. O., and G. D. contributed to patient recruitment and data acquisition and analysis; G. Y. H. L. contributed to review of the study results, analysis of the study, and the final version of the manuscript; and G. P., L. S., M. C., M. G., A. M., G. B., A. D. M., G. G., G. P., A. G., N. B., S. M., M. M., A. V., V. B., M. O., G. D., and G. Y. H. L. contributed to the revision of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Other contributions: A list of participating clinical centers and doctors is provided in e-Appendix 1. Additional information: The e-Appendix can be found in the Supplemental Materials section of the online article.
References 1. Camm AJ, Lip GY, De Caterina R, et al; ESC Committee for Practice Guidelines (CPG). 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(21):2719-2747. 2. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983-988. 3. Gattellari M, Goumas C, Aitken R, Worthington JM. Outcomes for patients with ischaemic stroke and atrial fibrillation: the PRISM study (A Program of Research Informing Stroke Management). Cerebrovasc Dis. 2011;32(4):370-382. 4. De Caterina R, Hylek EM. Stroke prevention in atrial fibrillation: current status and near-future directions. Am J Med. 2011;124(9):793-799. 5. Gladstone DJ, Bui E, Fang J, et al. Potentially preventable strokes in highrisk patients with atrial fibrillation who are not adequately anticoagulated. Stroke. 2009;40(1):235-240.
1080 Original Research
18. Raunsø J, Selmer C, Olesen JB, et al. Increased short-term risk of thromboembolism or death after interruption of warfarin treatment in patients with atrial fibrillation. Eur Heart J. 2012;33(15):1886-1892. 19. Mant J, Hobbs FDR, Fletcher K, et al; BAFTA investigators; Midland Research Practices Network (MidReC). Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet. 2007;370(9586):493-503. 20. Wilke T, Groth A, Mueller S, et al. Oral anticoagulation use by patients with atrial fibrillation in Germany. Adherence to guidelines, causes of anticoagulation under-use and its clinical outcomes, based on claims-data of 183,448 patients. Thromb Haemost. 2012;107(6):1053-1065. 21. Pugh D, Pugh J, Mead GE. Attitudes of physicians regarding anticoagulation for atrial fibrillation: a systematic review. Age Ageing. 2011;40(6):675-683. 22. Poli D, Antonucci E, Testa S, Tosetto A, Ageno W, Palareti G; Italian Federation of Anticoagulation Clinics. Bleeding risk in very old patients on vitamin K antagonist treatment: results of a prospective collaborative study on elderly patients followed by Italian Centres for Anticoagulation. Circulation. 2011;124(7):824-829. 23. Fang MC, Go AS, Chang Y, et al. A new risk scheme to predict warfarinassociated hemorrhage: the ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) study. J Am Coll Cardiol. 2011;58(4):395-401. 24. Friberg L, Rosenqvist M, Lip GY. Evaluation of risk stratification schemes for ischaemic stroke and bleeding in 182 678 patients with atrial fibrillation: the Swedish Atrial Fibrillation cohort study. Eur Heart J. 2012;33(12): 1500-1510. 25. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med. 1994;120(11):897-902. 26. Fang MC, Chang YC, Hylek EM, et al. Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med. 2004;141(10):745-752. 27. Huhtakangas J, Tetri S, Juvela S, Saloheimo P, Bode MK, Hillbom M. Effect of increased warfarin use on warfarin-related cerebral hemorrhage: a longitudinal population-based study. Stroke. 2011;42(9):2431-2435. 28. Connolly SJ, Eikelboom J, Joyner C, et al; AVERROES Steering Committee and Investigators. Apixaban in patients with atrial fibrillation. N Engl J Med. 2011;364(9):806-817.
6. Lip GY. The role of aspirin for stroke prevention in atrial fibrillation. Nat Rev Cardiol. 2011;8(10):602-606. 7. Alberts MJ, Eikelboom JW, Hankey GJ. Antithrombotic therapy for stroke prevention in non-valvular atrial fibrillation. Lancet Neurol. 2012;11(12):1066-1081. 8. Gorin L, Fauchier L, Nonin E, Charbonnier B, Babuty D, Lip GYH. 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(4):911-917. 9. Hobbs FD, Fitzmaurice DA, Mant J, et al. A randomised controlled trial and cost-effectiveness study of systematic screening (targeted and total population screening) versus routine practice for the detection of atrial fibrillation in people aged 65 and over. The SAFE study. Health Technol Assess. 2005;9(40): 1-74. 10. Zoni-Berisso M, Filippi A, Landolina M, et al. Frequency, patient characteristics, treatment strategies, and resource usage of atrial fibrillation (from the Italian Survey of Atrial Fibrillation Management [ISAF] study). Am J Cardiol. 2013;111(5):705-711. 11. Paciaroni M, Agnelli G, Caso V, et al. Atrial fibrillation in patients with firstever stroke: frequency, antithrombotic treatment before the event and effect on clinical outcome. J Thromb Haemost. 2005;3(6):1218-1223. 12. Ogilvie IM, Welner SA, Cowell W, Lip GY. Characterization of the proportion of untreated and antiplatelet therapy treated patients with atrial fibrillation. Am J Cardiol. 2011;108(1):151-161. 13. Stergiou GS, Karpettas N, Protogerou A, Nasothimiou EG, Kyriakidis M. Diagnostic accuracy of a home blood pressure monitor to detect atrial fibrillation. J Hum Hypertens. 2009;23(10): 654-658. 14. Wiesel J, Fitzig L, Herschman Y, Messineo FC. Detection of atrial fibrillation using a modified microlife blood pressure monitor. Am J Hypertens. 2009;22(8):848-852. 15. Lowres N, Neubeck L, Redfern J, Freedman SB. Screening to identify unknown atrial fibrillation. A systematic review. Thromb Haemost. 2013;110(2):213-222. 16. Willits I, Keltie K, Craig J, Sims A. WatchBP Home A for opportunistically detecting atrial fibrillation during diagnosis and monitoring of hypertension: a NICE medical technology guidance. Appl Health Econ Health Policy. 2014;12(3): 255-265. 17. Broderick JP, Bonomo JB, Kissela BM, et al. Withdrawal of antithrombotic agents and its impact on ischemic stroke occurrence. Stroke. 2011;42(9):2509-2514.
[
146#4 CHEST OCTOBER 2014
]
Original Research Cardiovascular Disease
]
Stroke/Thromboembolism and Intracranial Hemorrhage in a Real-world Atrial Fibrillation Population The Complications of Atrial Fibrillation in the Bologna Area (CAFBO) Study Gualtiero Palareti, MD; Luisa Salomone, MD; Mario Cavazza, MD; Marcello Guidi, MD; Antonio Muscari, MD; Giuseppe Boriani, MD; Antonio Di Micoli, MD; Giordano Guizzardi, MD; Gaetano Procaccianti, MD; Angelo Guidetti, MD; Nicola Binetti, MD; Simona Malservisi, MD; Marco Masina, MD, Antonella Viola, MD; Vincenzo Bua, MD; Maurizio Ongari, MD; Giampaolo Diaspri, MD; and Gregory Y. H. Lip, MD
Ischemic events (IEs) and intracranial hemorrhages (ICHs) are feared complications of atrial fibrillation (AF) and of antithrombotic treatment in patients with these conditions.
BACKGROUND:
Patients with AF admitted to the EDs of the Bologna, Italy, area with acute IE or ICH were prospectively recorded over 6 months.
METHODS:
A total of 178 patients (60 male patients; median age: 85 years) presented with acute IE. Antithrombotic therapy was as follows: (1) vitamin K antagonists (VKAs) in 31 patients (17.4%), with international normalized ratio (INR) at admission of , 2.0 in 16 patients, 2.0 to 3.0 in 13 patients, and . 3.0 in two patients; (2) aspirin (acetylsalicylic acid) (ASA) in 107 patients (60.1%); and (3) no treatment in 40 patients (22.5%), mainly because AF was not diagnosed. Twenty patients (eight male patients; median age: 82 years) presented with acute ICH: 13 (65%) received VKAs (INR, 2.0-3.0 in 11 patients and . 3.0 in two patients), while six (30%) received ASA. Most IEs (88%) and ICHs (95%) occurred in patients aged . 70 years. A modeling analysis of patients aged . 70 years was used to estimate annual incidence in subjects anticoagulated with VKAs in our Network of Anticoagulation Centers (NACs), or those expected to have AF but not included in NACs. The expected incidence of IE was 12.0%/y (95% CI, 10.7-13.3) in non-NACs and 0.57%/y (95% CI, 0.42-0.76) in NACs (absolute risk reduction [ARR], 11.4%/y; relative risk reduction [RRR], 95%; P , .0001). The incidence of ICH was 0.63%/y (95% CI, 0.34-1.04) and 0.30%/y (95% CI, 0.19-0.44), respectively (ARR, 0.33%/y; RRR, 52.4%/y; P 5 .04).
RESULTS:
IEs occurred mainly in elderly patients who received ASA or no treatment. One-half of patients with IEs receiving anticoagulant treatment had subtherapeutic INRs. Therapeutic approaches to elderly subjects with AF require an effective anticoagulant treatment strategy. CONCLUSIONS:
CHEST 2014; 146(4):1073-1080
Manuscript received October 21, 2013; revision accepted March 14, 2014; originally published Online First May 8, 2014. ABBREVIATIONS: AF 5 atrial fibrillation; ARR 5 absolute risk reduction; ASA 5 aspirin (acetylsalicylic acid); GP 5 general practitioner; ICH 5 intracranial hemorrhage; IE 5 ischemic event; INR 5 international normalized ratio; LMWH 5 low-molecular-weight heparin;
journal.publications.chestnet.org
NACs 5 Network of Anticoagulation Centers; RRR 5 relative risk reduction; VKA 5 vitamin K antagonist AFFILIATIONS: From the Angiologia e Malattie della Coagulazione (Drs Palareti and Salomone), Medicina D’Urgenza e Pronto Soccorso (Drs Cavazza and Guidi), Stroke Unit, Medicina Interna (Dr Muscari), and Institute of Cardiology (Dr Boriani), Department of Experimental,
1073
Ischemic stroke and systemic embolism are the most feared complications of atrial fibrillation (AF), a condition whose prevalence sharply increases with age.1 One in five of all strokes is attributed to AF, with a risk for stroke that increases with age, reaching values of 23.5% in patients aged ⱖ 80 years.2 Furthermore, AF-related strokes are more often fatal and more disabling than those due to other causes.3 Anticoagulant treatment with vitamin K antagonists (VKAs) is extremely effective in reducing the risk for AF-associated stroke, but is still underused in clinical practice4 or not adequately monitored with subsequent insufficient protection.5 Many patients with AF receive aspirin (acetylsalicylic acid) (ASA), though its efficacy is
Materials and Methods Setting The area of Bologna, in Northern Italy, has an incident population of 866,160 (735,960 aged ⱖ 18 years) and covers an area of 2,915 km2 with 50 municipalities. Population-based epidemiologic studies are possible in this setting because medical care is completely and uniformly provided by (1) a network of hospitals comprising two major facilities—Ospedale Maggiore and the University Hospital, both in the city of Bologna—and other middle-size hospitals distributed in the area and (2) general practitioners (GPs). These health services are part of the regional health system. A major ED is present in each of two major hospitals in Bologna; both are approved for thrombolysis in acute stroke. A network of EDs is active in the other hospitals. In general, patients with acute stroke are urgently referred to one of the two largest EDs, so that thrombolytic treatment can be provided, if indicated. Monitoring of patients treated by oral anticoagulants (at the time of the study, only VKAs were used) is performed by a Network of Anticoagulation Centers (NACs), covering all the Bologna area, which use the same computerized system to help patient management and dosing (P.A.R.M.A.; Werfen Group). Most subjects receiving anticoagulant
Diagnostic and Specialty Medicine, University of Bologna, Sant’OrsolaMalpighi University Hospital, Bologna, Italy; Pronto Soccorso e Medicina D’Urgenza Ospedale Maggiore (Drs Di Micoli and Guizzardi), Bologna, Italy; IRCCS Institute of Neurological Sciences, Maggiore Hospital Stroke Unit (Dr Procaccianti), Bologna, Italy; Pronto Soccorso ed Emergenza Territoriale Area Nord (Dr Guidetti and Binetti), Bologna, Italy; Stroke Care, Geriatria, Ospedale di Bentivoglio (Drs Malservisi and Masina), Bologna, Italy; Pronto Soccorso ed Emergenza Territoriale Area Sud (Drs Viola and Bua), Bologna, Italy; UOC Medicina PorrettaVergato (Drs Ongari and Diaspri), Bologna, Italy; and University of Birmingham Centre for Cardiovascular Sciences (Dr Lip), City Hospital, Birmingham, England. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study. CORRESPONDENCE TO: Gualtiero Palareti, MD, Department of Angiology and Blood Coagulation, University Hospital Policlinico S. OrsolaMalpighi, Via Albertoni 15, 40138 Bologna, Italy; e-mail: gualtiero. [email protected] © 2014 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.13-2443
1074 Original Research
limited,6,7 while many others remain completely untreated. Nonadherence to guidelines and undertreatment in clinical practice are independently associated with a high risk for stroke and mortality.8 This observational study prospectively recorded all the patients with AF who, during a 6-month period, were consecutively admitted with an acute ischemic event (IE) (ie, stroke or systemic embolism) or intracranial hemorrhage (ICH) to the EDs of the area of Bologna, Italy, and examined the type (if any) and quality of treatment they were receiving. Using a specific modeling analysis, we estimated treatment outcomes in patients with AF aged . 70 years living in that area.
treatment living in the area are included in one of the NACs; however, some patients, prospectively very few, are referred to their GP for anticoagulation management. Design and Study Population All the EDs in the area agreed to participate in the study. At least one doctor in each participating clinical center was identified as point person for the study, and a doctor (L. S.), active in the Angiology and Blood Coagulation unit (coordinating center), was in charge of monitoring the study. The institutional review boards of all participating centers approved the study: Azienda Ospedaliero-Universitaria di Bologna Policlinico S.Orsola-Malpighi (110/2011/U/Oss); Azienda AUSL c/o Dipartimento Farmaceutico Ospedale Maggiore di Bologna (11052). Written informed consent for participation was obtained from the patients or from relatives when a patient was unable to give it. From January 22, 2012, to July 22, 2012 (183 days), all patients admitted to the EDs with IE or ICH were included in the study if AF was present at admission, there was a known diagnosis of paroxysmal AF preadmission, or AF was detected after hospitalization. Patients were excluded if the event was more likely attributable to other conditions than AF. The included patients were identified from the general database of the NACs, and results of anticoagulation control during the previous 4 months were analyzed and evaluated as “stable” patients with an international normalized ratio (INR) between 1.8 and 3.2 (with no variations in VKA doses), and “variable” in other cases. Statistical Analysis Differences between groups were assessed by the x2 test with Yates’ correction for categorical variables and by the Mann-Whitney test for continuous variables. The data were analyzed with Prism software version 3.0 (GraphPad Software Inc), and SPSS software version 11.0 (IBM). Specific modeling analyses were designed to calculate separately the incidence of IE or ICH in subjects with AF who were aged . 70 years included or not in NACs. The number of inhabitants in the Bologna area aged . 70 years was obtained from 2012-related data of the regional health system; the total numbers of subjects with AF in this population was calculated by estimating, in-line with more recent population studies,9,10 an AF prevalence of 7%. A rough estimate of elderly patients with AF with unknown treatment was obtained by subtracting the number of patients with AF included in NACs from the total calculated elderly population with AF. The modeling analysis was performed at the end of the 6-month study period.
[
146#4 CHEST OCTOBER 2014
]
Results During the 6-month period, 178 patients with AF (60 male patients; median age, 85 years [range, 58-99 years]) with acute IE and 20 patients (eight male patients; median age, 82 years [range, 62-96 years]) with acute ICH were included in the study, while 16 patients with IE and one patient with ICH were excluded because the events were more likely to be attributed to causes other than AF (Table 1). IEs occurred more frequently in women than in men (66.3% vs 33.7%, respectively; P , .0001). Table 2 summarizes characteristics of included patients and details of their antithrombotic treatment. In 16 of the 31 patients (51.6%) treated with VKAs (three had a temporary interruption of treatment), the INR was , 2.0 at the event, while in 13 (41.9%), the INR was within the therapeutic range. Overall, the quality of treatment in the previous 4 months was stable in one-
TABLE 1
half of the patients treated with VKAs and followed-up in NACs (n 5 12); seven patients were receiving VKAs but were not followed in NACs. Among the patients who presented with IE, 147 (82.6%) were treated with ASA or did not receive any antithrombotic therapy; some (n 5 32, 18.0%) had formerly been treated with VKAs, but treatment had been interrupted permanently for various reasons (Table 2) (Fig 1). Of the 20 patients who presented with ICH (Table 2), 13 (65%) were receiving VKA treatment and one was taking concomitant low-dose ASA. Of these, 11 had an INR result in the therapeutic range, and two had INR values . 3.0. Anticoagulation quality in the previous 4 months was stable in 10 cases (76.9%) and variable in only one. The quality of anticoagulation could not be assessed in one patient, whose VKA therapy was managed by the GP, and in another case in which the ICH occurred within the first 20 days of treatment. Six patients (30%)
] Patient Characteristics and Outcomes
Patients
Ischemic Events
Intracranial Hemorrhages
194
21
Mechanical valve prosthesis
4
…
Peripheral artery disease
3
…
Severe carotid stenosis
4
…
5
…
Screened, No. Excluded for, No.
Different final diagnosis Bleeding within a brain tumor Included, No. Age, median (range), y Male sexa Stroke Peripheral arterial emboli
… 178 85 (58-99) 60 (33.7)
1 20 82 (62-96) 8 (40)
159 (89.3)
…
19 (10.7)
…
Type of AF, No. (%; No. male patients) Chronic
84 (47.2; 29)
13 (65; 5)
Paroxysmal
61 (34.3; 16)
4 (20; 1)
New onset
21 (11.8; 9)
3 (15; 2)
Diagnosed during hospitalization
12 (6.7; 6)
…
Age distribution of patients and type of the events, No. (%; No. male patients); [stroke/peripheral arterial emboli), y ⱕ 70
21 (11.8; 13) [20/1]
71-80
40 (22.5; 19) [34/6]
8 (40; 4)
81-90
82 (46.1; 25) [74/8]
10 (50; 3)
. 90
35 (19.6; 3) [31/4]
In-hospital death
34 (19.1)
1 (5; 1)
1(5; 0) 9 (45.0)
Data given as No. (%) unless otherwise indicated. AF 5 atrial fibrillation. aP , .0001 vs female patients.
journal.publications.chestnet.org
1075
TABLE 2
] Patient Characteristics and Antithrombotic Treatment at Presentation With an Event
Characteristics Patients treated with VKAs, No. (%) 1 ASA
Ischemic Events (n 5 178) 31 (17.4)
Intracranial Hemorrhages (n 5 20) 13 (65)
1
1
CHA2DS2 VASc, median (range)
5 (2-9)
HAS-BLED, median (range)
3 (1-4)
VKA temporarily interrupted before the event, No. (% patients treated with VKA)
3 (9.7)
1a
Quality of VKA treatment INR value at the event 2-3 (in range)
13 (41.9)
, 2.0
16 (51.6)
. 3.0
2 (6.5)
b
11 (84.6) 0 2 (15.4)
Overall quality of treatment in the 4 previous mo Stable
12 (38.7)
10 (76.9)
Variable
12 (38.7)
1 1
NVc NA
d
Patients treated with antiplatelet drugs
7 (22.6) 107 (60.1)e
CHA2DS2 VASc, median (range)
5 (0-8)
HAS-BLED, median (range)
2 (0-5)
LMWH (enoxaparin) Patients who did not receive any antithrombotic treatment CHA2DS2 VASc, median (range) HAS-BLED, median (range) Patients formerly treated with VKAs who had permanently stopped VKAs before the event
…
1 6 (30)f
1 (5)g
40 (22.5)
0
4 (1-7) 2 (1-4) 32 (18.0)
0
Cause for permanent interruption, No. Bleeding
18
Cardioversion
5
Risk for fall
5
Patient’s decision
4
Antithrombotic treatment at the event, No. Antiplatelet drugs No treatment
25 7h
Data given as No. (%) unless otherwise indicated. ASA 5 aspirin (acetylsalicylic acid); CHA2DS2-VASc 5 congestive heart failure, hypertension, age ⱖ 75 y (2 points), diabetes mellitus, stroke (2 points), vascular disease, age 65 to 74 y, and sex category (female); HAS-BLED 5 hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly (. 65 y), drugs/alcohol concomitantly; INR 5 international normalized ratio; LMWH 5 low-molecular-weight heparin; NA 5 not applicable; NV 5 not valuable; VKA 5 vitamin K antagonist. aTreated with LMWH. bIn three of these patients, the INR was low because the VKA therapy was temporarily interrupted. cNot valuable because intracranial hemorrhage occurred during the first 20 d from beginning of VKA treatment; not applicable. dNot available because these patients were receiving VKA at the moment of the event but were not included in the Network of Anticoagulation Centers and, therefore, no information on the quality of anticoagulation control was available. eASA (n 5 81), thienopyridine (n 5 22), ASA plus clopidogrel (n 5 4). fASA (n 5 5), thienopyridine (n 5 1). gLMWH was given at therapeutic doses. hFour patients were receiving prophylactic doses of LMWH.
1076 Original Research
[
146#4 CHEST OCTOBER 2014
]
a diagnosis of acute IE or ICH attributable to AF. The type of antithrombotic treatment (if any) and the quality of anticoagulation monitoring was evaluated. This was possible due to the presence of NACs in the area that use the same general rules and the same VKA-dedicated electronic software that manage the vast majority of patients treated with VKAs.
Figure 1 – Different antithrombotic treatments in relation to age.
with ICH were receiving ASA, and one was treated with a therapeutic dose of low-molecular-weight heparin (LMWH). Modeling Analysis of Patients Aged . 70 Years
Very few events occurred in patients aged ⱕ 70 years (12% of IEs and 5% of ICHs). Therefore, we focused the analysis on the events that occurred in patients aged . 70 years vs the calculated number of subjects aged . 70 years who had AF in the population of Bologna area, by estimating a prevalence of AF of 7% in this population9,10 (Table 3). While 78% of all those subjects were receiving anticoagulant treatment and monitored in NACs, no information was available on treatment (if any) in the remaining elderly subjects expected to have AF. The expected incidence of IEs in subjects not included in NACs was 12.0%/y (95% CI, 10.7-13.3) and 0.57%/y (95% CI, 0.42-0.76) in those monitored in NACs, with an absolute risk reduction (ARR) of 11.4%/y and a relative risk reduction (RRR) of 95%/y (P , .0001). The expected incidence of ICH was 0.63%/y (95% CI, 0.34-1.04) and 0.30%/y (95% CI, 0.19-0.44) in subjects not included or included in NACs, respectively, with an ARR of 0.33%/y and a RRR of 52.4%/y (P 5 .040).
Discussion In the present study, we provide observational data on antecedents and associated antithrombotic therapy in patients presenting with IEs or ICH in a defined, local, AF population, and modeled this to estimate events in the area of Bologna. The present study is unique, as we prospectively recorded and examined all the patients aged ⱖ 18 years who were referred during consecutive 6 months to all the EDs in the area of Bologna (northern Italy, with a population of 735,960 aged ⱖ 18 years) with
journal.publications.chestnet.org
Our results suggest that the current therapeutic approach to patients with AF is much improved compared with the results of a previous Italian study,11 in which almost one-half of patients with AF with a first stroke did not receive any antithrombotic treatment and consistent with reviews.12 Only 22.5% of these patients with IE were untreated at admission, and one-half had previously nondiagnosed AF. This emphasizes the importance of a systematic screening of the population at risk (especially in elderly subjects). The best cost-effectiveness strategy seems to be opportunistic case finding, in which a health-care professional can assess a patient’s pulse during a consultation9 or use a specifically modified BP monitor.13,14 These simple and effective approaches should be strongly recommended.15,16 Less than one-fifth of all IEs in our study occurred in patients who were being treated with VKAs, and in some, the treatment had been temporarily interrupted before the event. This finding confirms the potential impact of even short, temporary withdrawals of treatment on the risk of complications in patients with AF,17,18 which cannot be adequately offset by prophylactic treatment with LMWH. About one-half of IEs during VKA treatment were associated with subtherapeutic INR values at the event, while in one-third of the cases, anticoagulation quality was variable. These observations emphasize the need for high-quality VKA anticoagulation management and are in-line with studies showing that suboptimal anticoagulation control is associated with an increased risk for stroke and mortality in patients with AF.5 The majority of IEs in our study (60%) occurred in patients treated with antiplatelet drugs (mainly low-dose ASA). ASA is significantly less effective than warfarin for prevention of stroke in patients with AF, with no difference in major bleeding or ICH compared with warfarin.19 Unfortunately, ASA is still largely given to patients with AF, especially the elderly.20 A more recent review reported that . 50% of patients are treated with antiplatelet drugs in clinical practice.12 Also, a systematic review of physicians’ attitudes regarding anticoagulant treatment in patients with AF reported that the most
1077
TABLE 3
] Subset of Inhabitants in Bologna Area Aged . 70 y: Modeling Analysis of the Expected Number of Subjects Affected by AF and Number of Events
Populations Total inhabitant population in Bologna area in 2012 aged . 70 y, No. (No. male patients; %)
No. of Subjects and Events
RR, %/ya
148,160 (59,477; 40.1)
…
71-80 y
88,472 (38,995)
. 80 y
59,688 (20,482)
Expected number of subjects aged . 70 y with AF, calculated according an estimated prevalence of AF of 7% in this population
10,370 (4,160)b
…
Anticoagulated patients with AF aged . 70 y monitored by NACs in February 2012
8,130 (3,315)
…
Expected number of subjects with AF aged . 70 y not included in NACs (thus, not receiving anticoagulation, or monitored by general practitioners, or receiving antiplatelet drugs, or not treated with antithrombotics)
2,240 (900)b
…
Ischemic events recorded during the 6 mo of the study in patients with AF aged . 70 y Monitored in NACs Calculated incidence per 100 patient-y, %/y (95% CI)
23
…
0.57 (0.42-0.76)
ARR: 11.4
Not included in NACs, No.
134
Calculated incidence per 100 patient-y, %/y (95% CI)
12.0 (10.7-13.3)
RRR: 95 P , .0001
Intracranial hemorrhages recorded in patients with AF aged . 70 y Monitored in NACs Calculated incidence per 100 patient-y, %/y (95% CI)
12
…
0.30 (0.19-0.44)
ARR: 0.41
Not included in NACs Calculated incidence per 100 patient-y, %/y (95% CI)
8c 0.71 (0.41-1.16)
RRR: 57.7 P 5 .0103
Data given as total No. (No. male patients) unless otherwise indicated. ARR 5 absolute risk reduction; NACS 5 Network of Anticoagulation Centers; RR 5 risk reduction; RRR 5 relative risk reduction. See Table 1 and 2 legends for expansion of other abbreviations. aIn patients monitored in NACs vs not included in NACs. bThese figures are calculated on the basis of the number of inhabitants of the Bologna area aged . 70 y by estimating a prevalence of AF of 7% in this population, with 40.1% of male patients. cSix patients received ASA, one received a therapeutic dose of LMWH, and one patient was treated with VKAs but not monitored in NACs.
important barriers to prescription of VKA anticoagulation were as follows: increasing age (the most important), bleeding risk or previous bleeding, fall risk, comorbidities, and lack of compliance.21 A 2011 study also reported a low risk of bleeding in very old patients treated with VKA and managed in anticoagulation centers, thus, emphasizing that age by itself should not be considered a contraindication to treatment, providing that these more frail patients are monitored by specifically trained professionals.22 In contrast with available evidence, many physicians are under the misconception that VKA treatment is harmful in elderly patients with AF and prefer to prescribe them ASA as a treatment they consider effective (though probably less than warfarin), but at much lower risk of bleeding and much easier for the patients. In our study, 13 of 20 ICH events occurred in patients treated with VKA. In all but two cases (one at the beginning of therapy and one with an INR . 3.0), the tempo1078 Original Research
rally related INR value was within the therapeutic range. Indeed, the occurrence of ICH is more often linked to specific predisposing conditions of the patients and interaction with VKAs, than to a high anticoagulation intensity.23 Six patients suffered from an ICH while receiving ASA, again confirming that this treatment is not free from serious complications.24 Our modeling analyses enable us to have some insights regarding the treatment conditions and results in the everyday real life in aged patients with AF living in the area of Bologna. For our modeling analysis, the figures of subjects . 70 years old living in Bologna area reported in Table 3 are the “real world” of that area, and the estimated number of subjects with AF depends on the prevalence of AF that can be attributed to that population. The number of those patients treated with VKAs followed in the NACs has been obtained from the informatic database of NACs. It is not known how the patients with
[
146#4 CHEST OCTOBER 2014
]
AF who are not included in NACs were treated; some of them may have received VKAs via their GP. Indeed, among the patients receiving anticoagulant treatment with VKAs and monitored in the NACs, the annual incidence of ICHs was 0.30%, which is lower than the expected incidence. VKA treatment is associated with an increased risk for ICH,25 and it is well known that advanced age and intensity of anticoagulation are the most important risk factors.26 One population-based study of incidence of VKA-associated ICH in Finland27 showed that despite a nearly fourfold increase in the proportion of VKAtreated subjects, the annual incidence of ICHs decreased during the study period, proving that this major complication can partly be prevented by a more accurate management of treated patients. The patients managed in our NACs were well monitored, with good quality anticoagulation control, and this may explain the relatively low rate of ICHs and why only one ICH event was associated with high INR value. In contrast, the rate of ICH events occurring in the elderly patients who were calculated to have AF and not followed in NACs was substantially higher than expected (annual incidence of 0.71%). As a limitation, we do not know the proportion in this group of subjects who were receiving antiplatelet drugs or for whom VKAs were monitored by their GP, or those who were not treated at all. The risk of bleeding associated with ASA (and especially ICH) is not negligible.19,28 Indeed, our data indicate that therapeutic approaches to elderly subjects with AF different from a well-monitored VKA
journal.publications.chestnet.org
therapy were not only markedly less effective in preventing thromboembolism but also at least equally associated with the risk of ICH. Limitations
First, the IE and ICH events that may have led to death before admission to an ED were not included in this study. Second, a longer duration of the study (eg, 1 year instead than 6 months) would lead to more events and could possibly have led to more robust evidence. Third, the estimate of 7% of AF prevalence in our population aged . 70 years can be questioned, although it is broadly consistent with the epidemiology of this arrhythmia. Finally, it was impossible to assess the specific conditions of patients not followed in NACs, such as how many of them were treated with VKAs, or received antiplatelet treatments, or were not treated at all. In conclusion, our study shows that AF-associated IEs occurred mainly in elderly patients who received antiplatelet drugs or were not treated because AF was undiagnosed. In patients receiving anticoagulant treatment, one-half of the IEs were associated with subtherapeutic INR levels, while all but one ICH occurred with INR in the therapeutic range. Elderly subjects with AF should be treated with well-monitored VKA anticoagulation or possibly the novel oral anticoagulants, because antiplatelet drugs were markedly less effective and at least equally associated with the risk for ICH. Further improvement in detection and diagnosis of AF and its appropriate treatment in the at-risk population is recommended.
1079
Acknowledgments Author contributions: G. Palareti had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. G. Palareti contributed to the study design, wrote the first draft of the manuscript, and served as principal author; L. S. contributed to the study design, monitored the collection of the patients in the participating EDs, collected data on anticoagulant therapy, and evaluated the quality of anticoagulation in the treated patients; M. C., M. G., A. M., G. B., A. D. M., G. G., G. Procaccianti, A. G., N. B., S. M., M. M., A. V., V. B., M. O., and G. D. contributed to patient recruitment and data acquisition and analysis; G. Y. H. L. contributed to review of the study results, analysis of the study, and the final version of the manuscript; and G. P., L. S., M. C., M. G., A. M., G. B., A. D. M., G. G., G. P., A. G., N. B., S. M., M. M., A. V., V. B., M. O., G. D., and G. Y. H. L. contributed to the revision of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Other contributions: A list of participating clinical centers and doctors is provided in e-Appendix 1. Additional information: The e-Appendix can be found in the Supplemental Materials section of the online article.
References 1. Camm AJ, Lip GY, De Caterina R, et al; ESC Committee for Practice Guidelines (CPG). 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(21):2719-2747. 2. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983-988. 3. Gattellari M, Goumas C, Aitken R, Worthington JM. Outcomes for patients with ischaemic stroke and atrial fibrillation: the PRISM study (A Program of Research Informing Stroke Management). Cerebrovasc Dis. 2011;32(4):370-382. 4. De Caterina R, Hylek EM. Stroke prevention in atrial fibrillation: current status and near-future directions. Am J Med. 2011;124(9):793-799. 5. Gladstone DJ, Bui E, Fang J, et al. Potentially preventable strokes in highrisk patients with atrial fibrillation who are not adequately anticoagulated. Stroke. 2009;40(1):235-240.
1080 Original Research
18. Raunsø J, Selmer C, Olesen JB, et al. Increased short-term risk of thromboembolism or death after interruption of warfarin treatment in patients with atrial fibrillation. Eur Heart J. 2012;33(15):1886-1892. 19. Mant J, Hobbs FDR, Fletcher K, et al; BAFTA investigators; Midland Research Practices Network (MidReC). Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet. 2007;370(9586):493-503. 20. Wilke T, Groth A, Mueller S, et al. Oral anticoagulation use by patients with atrial fibrillation in Germany. Adherence to guidelines, causes of anticoagulation under-use and its clinical outcomes, based on claims-data of 183,448 patients. Thromb Haemost. 2012;107(6):1053-1065. 21. Pugh D, Pugh J, Mead GE. Attitudes of physicians regarding anticoagulation for atrial fibrillation: a systematic review. Age Ageing. 2011;40(6):675-683. 22. Poli D, Antonucci E, Testa S, Tosetto A, Ageno W, Palareti G; Italian Federation of Anticoagulation Clinics. Bleeding risk in very old patients on vitamin K antagonist treatment: results of a prospective collaborative study on elderly patients followed by Italian Centres for Anticoagulation. Circulation. 2011;124(7):824-829. 23. Fang MC, Go AS, Chang Y, et al. A new risk scheme to predict warfarinassociated hemorrhage: the ATRIA (Anticoagulation and Risk Factors in Atrial Fibrillation) study. J Am Coll Cardiol. 2011;58(4):395-401. 24. Friberg L, Rosenqvist M, Lip GY. Evaluation of risk stratification schemes for ischaemic stroke and bleeding in 182 678 patients with atrial fibrillation: the Swedish Atrial Fibrillation cohort study. Eur Heart J. 2012;33(12): 1500-1510. 25. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med. 1994;120(11):897-902. 26. Fang MC, Chang YC, Hylek EM, et al. Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med. 2004;141(10):745-752. 27. Huhtakangas J, Tetri S, Juvela S, Saloheimo P, Bode MK, Hillbom M. Effect of increased warfarin use on warfarin-related cerebral hemorrhage: a longitudinal population-based study. Stroke. 2011;42(9):2431-2435. 28. Connolly SJ, Eikelboom J, Joyner C, et al; AVERROES Steering Committee and Investigators. Apixaban in patients with atrial fibrillation. N Engl J Med. 2011;364(9):806-817.
6. Lip GY. The role of aspirin for stroke prevention in atrial fibrillation. Nat Rev Cardiol. 2011;8(10):602-606. 7. Alberts MJ, Eikelboom JW, Hankey GJ. Antithrombotic therapy for stroke prevention in non-valvular atrial fibrillation. Lancet Neurol. 2012;11(12):1066-1081. 8. Gorin L, Fauchier L, Nonin E, Charbonnier B, Babuty D, Lip GYH. 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(4):911-917. 9. Hobbs FD, Fitzmaurice DA, Mant J, et al. A randomised controlled trial and cost-effectiveness study of systematic screening (targeted and total population screening) versus routine practice for the detection of atrial fibrillation in people aged 65 and over. The SAFE study. Health Technol Assess. 2005;9(40): 1-74. 10. Zoni-Berisso M, Filippi A, Landolina M, et al. Frequency, patient characteristics, treatment strategies, and resource usage of atrial fibrillation (from the Italian Survey of Atrial Fibrillation Management [ISAF] study). Am J Cardiol. 2013;111(5):705-711. 11. Paciaroni M, Agnelli G, Caso V, et al. Atrial fibrillation in patients with firstever stroke: frequency, antithrombotic treatment before the event and effect on clinical outcome. J Thromb Haemost. 2005;3(6):1218-1223. 12. Ogilvie IM, Welner SA, Cowell W, Lip GY. Characterization of the proportion of untreated and antiplatelet therapy treated patients with atrial fibrillation. Am J Cardiol. 2011;108(1):151-161. 13. Stergiou GS, Karpettas N, Protogerou A, Nasothimiou EG, Kyriakidis M. Diagnostic accuracy of a home blood pressure monitor to detect atrial fibrillation. J Hum Hypertens. 2009;23(10): 654-658. 14. Wiesel J, Fitzig L, Herschman Y, Messineo FC. Detection of atrial fibrillation using a modified microlife blood pressure monitor. Am J Hypertens. 2009;22(8):848-852. 15. Lowres N, Neubeck L, Redfern J, Freedman SB. Screening to identify unknown atrial fibrillation. A systematic review. Thromb Haemost. 2013;110(2):213-222. 16. Willits I, Keltie K, Craig J, Sims A. WatchBP Home A for opportunistically detecting atrial fibrillation during diagnosis and monitoring of hypertension: a NICE medical technology guidance. Appl Health Econ Health Policy. 2014;12(3): 255-265. 17. Broderick JP, Bonomo JB, Kissela BM, et al. Withdrawal of antithrombotic agents and its impact on ischemic stroke occurrence. Stroke. 2011;42(9):2509-2514.
[
146#4 CHEST OCTOBER 2014
]
