- Email: [email protected]
Strokes after cardioversion of atrial fibrillation — The FibStroke study
International Journal of Cardiology 203 (2016) 269–273
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Strokes after cardioversion of atrial fibrillation — The FibStroke study☆ Antti Palomäki a, Pirjo Mustonen b, Juha E.K. Hartikainen c, Ilpo Nuotio d, Tuomas Kiviniemi a, Antti Ylitalo e, Päivi Hartikainen f, Heidi Lehtola b, Riho Luite c, K.E. Juhani Airaksinen a,⁎ a
Heart Center, Turku University Hospital, University of Turku, Hämeentie 11, PO Box 52, 20521 Turku, Finland Department of Medicine, Keski-Suomi Central Hospital, Keskussairaalantie 19, 40620 Jyväskylä, Finland Heart Center, Kuopio University Hospital, Puijonlaaksontie 2, 70210 Kuopio, Finland d Department of Medicine, Turku University Hospital,University of Turku, Hämeentie 11, PO Box 52, 20521 Turku, Finland e Heart Center, Satakunta Central Hospital, Sairaalantie 3, 28500 Pori, Finland f NeuroCenter, Neurology, Kuopio University Hospital, Puijonlaaksontie 2, 70210 Kuopio, Finland b c
a r t i c l e
i n f o
Article history: Received 16 July 2015 Received in revised form 1 October 2015 Accepted 19 October 2015 Available online 20 October 2015 Keywords: Atrial fibrillation Cardioversion Stroke Transient ischemic attack
a b s t r a c t Background: Cardioversion of atrial fibrillation (AF) is associated with an increased risk for stroke. We identified all cardioversions during the 30 days preceding stroke or transient ischemic attack (TIA) in patients with a previously diagnosed AF, and sought to assess the characteristics of cardioversions leading to stroke or TIA. Methods: FibStroke is a cross-sectional observational multicenter registry that included AF patients with an ischemic stroke or intracranial bleed identified from a discharge registry of four Finnish hospitals. In total 3677 consecutive AF patients suffered 3252 strokes and 956 TIA episodes during 2003–2012. This pre-specified analysis focused on the 1644 events that occurred to patients with paroxysmal or persistent AF at the time of stroke/TIA. Results: A total of 78 strokes and 22 TIA episodes were preceded by a cardioversion. Post-cardioversion strokes accounted for 6.4% of strokes in patients with paroxysmal/persistent AF. Of the 100 cardioversions leading to an ischemic event, 77 were acute and 23 were elective, 63 events occurred in patients not using anticoagulation, and 5 patients had periprocedural INR b 2. Importantly, 21 patients were in low risk of stroke, i.e. CHA2DS2-VASc score b 2. The median delay from cardioversion to event was 2 days. All nine patients who after an unsuccessful cardioversion developed a stroke had a spontaneous cardioversion prior to stroke. Conclusions: Every sixteenth stroke of patients with paroxysmal/persistent AF is preceded by a cardioversion. Most post-cardioversion strokes occur in patients not using oral anticoagulation before cardioversion of acute AF. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Cardioversions are an essential part of rhythm-control strategy in the treatment of atrial fibrillation (AF). Every cardioversion is, however, associated with a temporary increase in stroke risk, although the risk has been less than 1% in patients with adequate anticoagulation or
Abbreviations: AF, atrial fibrillation; CHA2DS2-VASc, congestive heart failure, hypertension, age ≥ 75 (doubled), diabetes mellitus, and prior Stroke, transient ischemic attack or thromboembolism (doubled), vascular disease, Age 65 to 74, sex category (female, unless b65 years and no other risk factors); INR, international normalized ratio; IQR, inter-quartile range; TIA, transient ischemic attack. ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Heart Center, Turku University Hospital, PO Box 52, 20521 Turku, Finland. E-mail addresses: ajpalo@utu.fi (A. Palomäki), pirjo.mustonen@ksshp.fi (P. Mustonen), juha.hartikainen@kuh.fi (J.E.K. Hartikainen), ilpo.nuotio@tyks.fi (I. Nuotio), tuoski@utu.fi (T. Kiviniemi), [email protected] (A. Ylitalo), paivi.hartikainen@kuh.fi (P. Hartikainen), heidi.lehtola@ksshp.fi (H. Lehtola), riho.luite@kuh.fi (R. Luite), juhani.airaksinen@tyks.fi (K.E.J. Airaksinen).
http://dx.doi.org/10.1016/j.ijcard.2015.10.168 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
after cardioversion of acute AF [1–4]. Although rhythm control is a tempting primary treatment option, this strategy has offered no mortality benefit compared to rate-control in randomized trials [5,6]. Majority of strokes occurred in these trials in patients with subtherapeutic anticoagulation or in whom warfarin had been stopped after restoring the sinus rhythm. These results highlight the importance of long-term therapeutic anticoagulation in patients with paroxysmal and persistent AF in whom sinus rhythm is pursued. In this pre-specified analysis of the FibStroke study, we sought to assess the clinical characteristics and significance of cardioversion as a risk factor for cerebral thromboembolism in a large population of AF patients who suffered an ischemic stroke or transient ischemic attack (TIA). We hypothesized that cardioversion is a common risk factor for thromboembolic complications in patients with AF and that antithrombotic medication is underused in this population. The main objective of the study was to assess the typical characteristics of cardioversions leading to thromboembolic complication, with the ultimate goal of identifying potential targets for improving clinical practice and increasing patient safety.
270
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273
2. Methods 2.1. Study population The FibStroke study (www.ClinicalTrials.gov, identifier NCT02146040) is part of a wider protocol in progress to assess thrombotic and bleeding complications of AF in Finland [7–9]. The FibStroke registry included all consecutive patients with a previously diagnosed AF who suffered a stroke or intracranial bleeding during the study period. Data for the study was collected from two university hospitals and two central hospitals in Finland. The initial screening was conducted by identifying all patients from the hospital discharge records with the following criteria: 1) The patient had been diagnosed with AF or atrial flutter and 2) the patient had been diagnosed with ischemic stroke, TIA or intracranial bleed between the years 2003–2012 (in one central hospital 2006–2012). A comprehensive list of the ICD 10 codes used for the screening is provided in the Supplementary methods. After the initial screening, all patient files were reviewed individually, and all procedures and cardioversions performed during the 30 days preceding the stroke, TIA or intracranial bleeding were identified. Detailed information about the cardioversion and the patient characteristics at the time of cardioversion and at the time of the stroke, TIA or intracranial bleeding was recorded. Data was collected in a structured electronic case report form. There were a total 3677 patients with a history of AF and who after the diagnosis of AF developed a total of 3252 ischemic strokes and 956 TIA episodes. For this pre-specified analysis we focused on patients in whom an elective or acute cardioversion for paroxysmal or persistent AF was performed during the 30 days preceding the stroke or TIA. As a control for the ischemic events, we also identified 767 patients with a previously diagnosed AF who suffered 794 intracranial bleeds during the study period, and preceding cardioversions for AF were recorded similarly in this subgroup. 2.2. Exclusion We excluded one patient who underwent cardioversion of AF during coronary artery bypass operation. The stroke was discovered on the first postoperative day. One of the included patients developed two strokes after a single cardioversion. The first stroke was hemorrhagic and warfarin was stopped followed by another ischemic stroke 12 days later. The second stroke was excluded from this analysis. 2.3. Ethical issues The study protocol was approved by the Medical Ethics Committee of the Hospital District of Southwest Finland and the ethics committee of the National Institute for Health and Welfare. Informed consent was not required because of the registry nature of the study. The study conforms to the Declaration of Helsinki. 2.4. Definitions Diagnosis of AF was confirmed by 12-lead electrocardiogram according to the standard criteria. Cardioversion was considered successful if sinus rhythm was obtained and the patient was discharged in sinus rhythm. The diagnosis of stroke and TIA were confirmed from the patient records, as diagnosed by the treating neurologist. Only strokes and TIAs considered as definite by the treating physician were included in our study. All patients were imaged by computed tomography or magnetic resonance imaging. 2.5. Statistical analysis Continuous variables were reported as mean ± standard deviation if they were normally distributed, and as median [inter-quartile range
(IQR)] if they were skewed unless stated otherwise. Categorical variables were described with absolute and relative (percentage) frequencies. Comparisons between study subgroups were performed with Mann–Whitney test for continuous and Chi-square test or Fisher's exact test as appropriate for categorical variables. All tests were twosided and statistical significance was set at 5%. Statistical analysis was performed using IBM SPSS Statistics software version 22.0 and SAS version 9.3. This manuscript was written following STROBE guidelines for the reporting of observational studies [10]. 3. Results 3.1. Characteristics of the patients and the ischemic events A total of 3252 ischemic strokes and 956 TIA episodes occurred during the study period. The type of AF was permanent in 2188, paroxysmal or persistent in 1644, and undefined in 376 at the time of the event. We identified a total of 78 ischemic strokes and 22 TIAs in 99 patients that were preceded by a cardioversion of AF. One patient had two ischemic events that were both preceded by a cardioversion. The first was a TIA and second was an ischemic stroke which occurred over two years after the first episode. Clinical characteristics of the patients are depicted in Table 1. Since cardioversions were only performed in patients with paroxysmal or persistent AF, post-cardioversion events accounted for 6.1% of all ischemic events and 6.4% of strokes in this patient group at risk, while 2.4% of all events and 2.4% of all strokes were preceded by a cardioversion. 3.2. Characteristics of the cardioversions Of the 100 cardioversions leading to an ischemic event, 77 were acute and 23 were elective, and 63 of the events occurred to patients not using anticoagulation (Table 2). Similarly, of the 78 cardioversions complicated by a stroke, 61 (78.2%) were acute and 17 (21.8%) were elective, and 51 (65.4%) of the strokes occurred to patients not using anticoagulation. Of the 10 patients who had been using warfarin prior to cardioversion of acute AF leading to stroke, 4 had INR less than 2. The majority of the cardioversions leading to cerebral thromboembolism were successful in terms of restoring sinus rhythm. All nine patients who after an unsuccessful cardioversion developed a stroke had a spontaneous cardioversion before the onset of stroke. Seven of these Table 1 Clinical characteristics of 99 patients who suffered 78 ischemic strokes and 22 transient ischemic attacks in 30 days after cardioversion of atrial fibrillation or atrial flutter. N (%) N Female Age, median (range) Hypertension Diabetes Coronary artery disease Previous myocardial infarction Heart failure Previous stroke or TIA HAS-BLED-score, median [IQR]a 0–2 N2 Mechanical heart valve Pacemaker
99 53 (53.5) 71.1 (47.0–91.8) 66 (66.7) 14 (14.1) 29 (29.3) 17 (17.2) 9 (9.1) 8 (8.1) 2 [1–3] 65 (65.7) 34 (34.3) 3 (3.0) 6 (6.1)
In the one patient suffering two events (one TIA and one stroke over two years apart), characteristics at the time of the first event are depicted. HAS-BLED = hypertension, abnormal renal or liver function (1 point each), stroke, bleeding, labile INRs, elderly (age N 65 years), drugs or alcohol (1 point each); IQR = inter-quartile range; TIA = transient ischemic attack. a Modified HAS-BLED calculated without labile international normalized ratio (INR).
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273 Table 2 Data of 78 cardioversions (CV) leading to ischemic stroke and 22 CVs leading to TIA.
N Stroke TIA Method of CV Electrical Pharmacological Successful CV Spontaneous CV after unsuccessful CV TEE-guided CV Warfarin prior to CV INR at the time of CV, range INR b 2 Periprocedural LMWH Antiplatelet treatment (aspirin or clopidogrel) Warfarin at the time of stroke or TIA INR at the time of stroke or TIA, range INR b 2 CHA2DS2-VASc, median [IQR] 0 1 ≥2 Days from CV to stroke/TIA, median [IQR] 0–2 days 3–6 days N6 days
Acute CV
Elective CV
77 61 (79.2) 16 (20.8)
23 17 (73.9) 6 (26.1)
67 (88.3) 9 (11.7) 68 (88.3) 9/9 3 (3.9) 14 (18.2) 1.2–4.0 5/14 (35.7) 5 (6.8) 45 (60.8) 24 (31.2) 0.9–3.6 13/24 (54.2) 3.0 [2.0–4.0] 8 (10.4) 7 (9.1) 62 (80.5) 2 [1.0–3.0] 56 (72.7) 6 (7.8) 15 (19.5)
23 (100) 0 21 (91.3) 1/2 0 23 (100) 2.0–5.4⁎ 0⁎ 0 3 (15.0) 22 (95.7) 1.3–5.4 5/22 (22.7) 3.0 [1.0–4.0] 2 (8.7) 4 (17.4) 17 (73.96) 2 [2.0–5.0] 12 (52.2) 8 (34.8) 3 (13.0)
271
93% during the first 14 days following the cardioversion (Fig. 1). The median delay from cardioversion to stroke or TIA was 2 days (range 0–28). The delay was slightly longer in patients using warfarin prior to cardioversion than in patients without warfarin (3.0 vs. 2.0 days, p b 0.01). 3.5. Intracranial bleeds
CV = cardioversion; CHA2DS2-VASc = congestive heart failure, hypertension, Age ≥ 75 (doubled), diabetes mellitus, and prior Stroke, transient ischemic attack or thromboembolism (doubled), vascular disease, age 65 to 74, sex category (female, unless b65 years and no other risk factors); INR = international normalized ratio; IQR = interquartile range; LMWH = low molecular weight heparin; TEE = transesophageal echocardiography; TIA = transient ischemic attack. ⁎ Data missing for 3 patients.
nine strokes occurred within two days of the attempted cardioversion, one after eight days, and one after 28 days. Three patients underwent transesophageal echocardiography before the cardioversion. No thrombi were detected in the left atrial appendage, nor were any other contraindications to cardioversion discovered. Two of the patients had been previously treated with ablation. Both suffered an ischemic stroke after acute cardioversion. The first patient was not using warfarin at the time of cardioversion, since it had been stopped after seemingly successful procedure. The second patient had continued warfarin after the ablation, and suffered a stroke despite INR 2.7 at the time of cardioversion. No differences were noted between central and university hospital settings except that pharmacological CV was more often performed in central hospitals (23.3% vs. 2.9%, p b 0.01). 3.3. Risk factors for stroke The CHA2DS2-VASc score was ≥2 in 79 (79%) of the patients who developed any ischemic event, and in 65/78 (83%) of the patients who developed a stroke after cardioversion. Five patients developed a stroke without any risk factors for thromboembolism as measured with the CHA2DS2-VASc score. All of these patients underwent an acute cardioversion and were not receiving peri- or postcardioversion anticoagulation. Age range of the patients was 59 to 64 years, and two were females. Nine patients who developed an ischemic stroke or TIA after the cardioversion had another predisposing factor (operation, myocardial infarction or major bleeding) during the 30 days preceding the stroke. Characteristics of these patients and the competing causes for stroke or TIA are depicted in Supplementary Table 1. 3.4. Timing of cerebral thromboembolism after cardioversion Most ischemic events occurred shortly after the cardioversion, as 68% occurred during the first two days, 83% during the first 7 days and
In the control group of 767 patients and 794 intracranial bleeds, three patients had a history of cardioversion performed within 30 days preceding the bleed, accounting for 0.4% of all intracranial bleedings. Two patients presented with a traumatic bleed and one patient with a spontaneous subdural hematoma 3, 27 and 4 days after the cardioversion, respectively. 4. Discussion The main finding of our study is that cardioversion is an important background factor for stroke in patients with paroxysmal or persistent AF and every sixteenth stroke in these relatively healthy patients was preceded by a cardioversion. The great majority of strokes occurred after cardioversion of an acute AF attack in patients not using oral anticoagulation prior to cardioversion. Importantly, patients with failed cardioversion are not safe from embolic complications, since spontaneous cardioversions were relatively common after attempted cardioversion of acute AF. 4.1. Acute cardioversion Most of the strokes occurred after cardioversion of acute (b 48 h) AF episode, and most of these complications were observed in patients not using anticoagulation. Until recent years, cardioversion of acute AF was considered safe even without anticoagulation. In our FinCV Study reporting data on 5116 successful acute cardioversions performed without anticoagulation the overall risk of thromboembolic events was 0.7% [4]. Thus, the mean risk of stroke in acute cardioversion performed without anticoagulation is in the same range as the risk of elective cardioversion performed with optimal guideline-recommended anticoagulation. Importantly, however, the risk of thromboembolic complications varied from a low risk of 0.2% to a very high risk of 9.8% depending on clustering of clinical stroke risk factors [4]. It is reasonable to assume that the thromboembolic complications could be prevented also in this acute setting with effective anticoagulation, although there is no research evidence on this so far. The current guidelines recommend intravenous heparin or low molecular weight heparin, or factor Xa or direct thrombin inhibitor before or immediately after cardioversion in patients with high stroke risk. In patients with low thromboembolic risk, these therapies may be considered [11]. One important new finding in FibStroke study was that 21% of the patients who developed an ischemic stroke or TIA after cardioversion had a CHA2DS2-VASc score b 2. We have recently shown that a delay of cardioversion exceeding 12 h from the symptom onset caused a N 3-fold increase in the risk of stroke when compared to later cardioversions of acute AF, and the risk of thromboembolism in the low risk patients with CHA2DS2VASc b 2 was as high as 0.9% if no anticoagulation was used [12]. In the light of these findings, it is possible that even low-risk patients might benefit from short term oral anticoagulation started at the time of cardioversion. 4.2. Elective cardioversion Only 23% of ischemic events occurred after elective cardioversion. This finding was not unexpected, since elective cardioversion is always performed during therapeutic anticoagulation. It is, however, noteworthy that even with this strategy the risk of thromboembolic complications has been about 0.5% within 30 days after the cardioversion
272
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273
Fig. 1. Timing of embolic complications after cardioversion of AF. Data of 100 cardioversions leading to 78 ischemic strokes and 22 TIA. AF: atrial fibrillation; TIA: transient ischemic attack.
[13–15]. On the other hand, the average risk of stroke in recently published large randomized multicenter studies has decreased to 0.1% to 0.2% per month during long-term anticoagulation [16–18]. In the light of these figures, it is evident that when a patient is referred to an elective cardioversion of AF we predispose the patient for an increased risk of stroke during the post-cardioversion month even when using the recommended effective anticoagulation. The increased risk is supported by the present findings on high ratio of strokes in relation to intracranial bleeds during the post-cardioversion month (2.4% vs 0.4%) compared with the respective yearly event rates in Re-Ly trial of 1.57% vs. 0.74% [16]. 4.3. Timing of cerebral thromboembolism after cardioversion In line with previous observations [4] the vast majority of strokes and TIAs occurred shortly after the successful cardioversion, with the median time from cardioversion to event of 2 days and over 80% of events occur during the first week after the cardioversion. Only 9 patients developed a stroke after a failed cardioversion. Of note, all these patients had, however, spontaneous cardioversion to sinus rhythm prior to the stroke supporting the view that the conversion of AF to sinus rhythm is the principal factor responsible for stroke in these patients. 4.4. Transesophageal echocardiography Three patients who suffered a stroke had a transesophageal echocardiography performed prior to the cardioversion. In previous studies, the risk of stroke of stroke after cardioversion guided by transesophageal echocardiography has been 0.8% [13,19]. This highlights the fact that exclusion of thrombus before the cardioversion does not guarantee safety, because of the risk of thrombus formation after cardioversion due to left atrial stunning [20]. 4.5. Limitations of the study Some limitations need to be addressed. Retrospective design carries always some limitations in the accuracy of data collection. This is compensated by the large number of events in our study. To our knowledge, this is the largest material of cardioversions leading to stroke or TIA published so far. Only a minority of patients who developed a stroke or TIA had a competing cause for the ischemic event in addition to
cardioversion showing, however, the multitude of clinical scenarios in real-life. The strengths of the study include the identification of all consecutive stroke and TIA patients with a diagnosis of AF from reliable hospital discharge records and the thorough individual case by case review of patient records. 5. Conclusions Every sixteenth stroke of patients with paroxysmal or persistent AF is preceded by a cardioversion of AF. Majority of strokes occur in patients not using oral anticoagulation before cardioversion of acute AF and typically 2–3 days after successful cardioversion. Unsuccessful cardioversion is not without risk for thromboembolism, since later spontaneous cardioversions are not uncommon. Significant proportion of the strokes was observed in relatively young patients with a low CHA2DS2-VASc score supporting the view that at least short-term anticoagulation should always be considered after cardioversion. Additional research is, however, needed to improve the safety of rhythm strategy in the treatment of AF. Disclosure of conflicts of interest Dr. Mustonen has given lectures for Orion, Boehringer Ingelheim, Bayer, Pfizer, Bristol-Myers Squibb, Sanofi-Aventis and Leo Pharma, and has been a member in the advisory boards for Boehringer Ingelheim, Bayer, Pfizer, Bristol-Myers Squibb and Leo Pharma. Dr. Juha Hartikainen has received research grants from the Finnish Foundation for Cardiovascular Research and the European Union Seventh Framework Program, has given lectures for Cardiome, St. Jude Medical and Biotronic, and has been member in the advisory boards for Astra Zeneca, Amgen and Bayer. Dr. Kiviniemi has given lectures for Bayer, Medicines Company, Astra Zeneca and St. Jude Medical, and received research grants from the Finnish Foundation for Cardiovascular Research and Finnish Cardiac Society. Dr. Päivi Hartikainen has received honoraria from Genzyme, Novartis, Biogen Idec, TEVA, and has given lectures for Sanofi-Aventis. Dr. Luite has received honoraria from Webster-Biosense, Biotronik, Bayer and St. Jude Medical, and has given lectures for Bristol-Myers Squibb. Dr. Airaksinen has received research grants from the Finnish Foundation for Cardiovascular Research, has given lectures for Bayer, Cardiome and Boehringer Ingelheim, and has been a member in the advisory boards for Bayer, Astra Zeneca and Boston Scientific. The other authors report no disclosures.
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273
Astra Zeneca and Boston Scientific. The other authors report no disclosures. Acknowledgment of grant support This work was supported by the Finnish Foundation for Cardiovascular Research. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2015.10.168. References [1] C.J. Bjerkelund, O.M. Orning, The efficacy of anticoagulant therapy in preventing embolism related to D.C. electrical conversion of atrial fibrillation, Am. J. Cardiol. 23 (1969) 208–216. [2] D.M. Weinberg, J. Mancini, Anticoagulation for cardioversion of atrial fibrillation, Am. J. Cardiol. 63 (1989) 745–746. [3] A.Z. Arnold, M.J. Mick, R.P. Mazurek, F.D. Loop, R.G. Trohman, Role of prophylactic anticoagulation for direct current cardioversion in patients with atrial fibrillation or atrial flutter, J. Am. Coll. Cardiol. 19 (1992) 851–855. [4] K.E. Airaksinen, T. Gronberg, I. Nuotio, et al., Thromboembolic complications after cardioversion of acute atrial fibrillation: the FinCV (Finnish CardioVersion) study, J. Am. Coll. Cardiol. 62 (2013) 1187–1192. [5] D.G. Wyse, A.L. Waldo, J.P. DiMarco, et al., A comparison of rate control and rhythm control in patients with atrial fibrillation, N. Engl. J. Med. 347 (2002) 1825–1833. [6] I.C. Van Gelder, V.E. Hagens, H.A. Bosker, et al., A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation, N. Engl. J. Med. 347 (2002) 1834–1840. [7] K.E. Airaksinen, F. Biancari, P. Karjalainen, et al., Safety of coronary artery bypass surgery during therapeutic oral anticoagulation, Thromb. Res. 128 (2011) 435–439.
273
[8] K.E. Airaksinen, P. Korkeila, J. Lund, et al., Safety of pacemaker and implantable cardioverter-defibrillator implantation during uninterrupted warfarin treatment—the FinPAC study, Int. J. Cardiol. 168 (2013) 3679–3682. [9] R. Virtanen, V. Kryssi, T. Vasankari, M. Salminen, S.L. Kivela, K.E. Airaksinen, Selfdetection of atrial fibrillation in an aged population: the LietoAF study, Eur. J. Prev. Cardiol. 21 (2014) 1437–1442. [10] E. von Elm, D.G. Altman, M. Egger, et al., Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies, BMJ 335 (2007) 806–808. [11] C.T. January, L.S. Wann, J.S. Alpert, et al., 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American college of cardiology/American heart association task force on practice guidelines and the heart rhythm society, J. Am. Coll. Cardiol. 64 (2014) e1–76. [12] I. Nuotio, J.E. Hartikainen, T. Gronberg, F. Biancari, K.E. Airaksinen, Time to cardioversion for acute atrial fibrillation and thromboembolic complications, JAMA 312 (2014) 647–649. [13] A.L. Klein, R.A. Grimm, R.D. Murray, et al., Use of transesophageal echocardiography to guide cardioversion in patients with atrial fibrillation, N. Engl. J. Med. 344 (2001) 1411–1420. [14] R. Cappato, M.D. Ezekowitz, A.L. Klein, et al., Rivaroxaban vs. vitamin K antagonists for cardioversion in atrial fibrillation, Eur. Heart J. 35 (2014) 3346–3355. [15] S. Apostolakis, K.G. Haeusler, M. Oeff, et al., Low stroke risk after elective cardioversion of atrial fibrillation: an analysis of the flec-SL trial, Int. J. Cardiol. 168 (2013) 3977–3981. [16] S.J. Connolly, M.D. Ezekowitz, S. Yusuf, et al., Dabigatran versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 361 (2009) 1139–1151. [17] M.R. Patel, K.W. Mahaffey, J. Garg, et al., Rivaroxaban versus warfarin in nonvalvular atrial fibrillation, N. Engl. J. Med. 365 (2011) 883–891. [18] C.B. Granger, J.H. Alexander, J.J. McMurray, et al., Apixaban versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 365 (2011) 981–992. [19] K. Seidl, M. Rameken, A. Drogemuller, et al., Embolic events in patients with atrial fibrillation and effective anticoagulation: value of transesophageal echocardiography to guide direct-current cardioversion. Final results of the Ludwigshafen observational cardioversion study, J. Am. Coll. Cardiol. 39 (2002) 1436–1442. [20] R.A. Grimm, W.J. Stewart, J.D. Maloney, et al., Impact of electrical cardioversion for atrial fibrillation on left atrial appendage function and spontaneous echo contrast: characterization by simultaneous transesophageal echocardiography, J. Am. Coll. Cardiol. 22 (1993) 1359–1366.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Strokes after cardioversion of atrial fibrillation — The FibStroke study☆ Antti Palomäki a, Pirjo Mustonen b, Juha E.K. Hartikainen c, Ilpo Nuotio d, Tuomas Kiviniemi a, Antti Ylitalo e, Päivi Hartikainen f, Heidi Lehtola b, Riho Luite c, K.E. Juhani Airaksinen a,⁎ a
Heart Center, Turku University Hospital, University of Turku, Hämeentie 11, PO Box 52, 20521 Turku, Finland Department of Medicine, Keski-Suomi Central Hospital, Keskussairaalantie 19, 40620 Jyväskylä, Finland Heart Center, Kuopio University Hospital, Puijonlaaksontie 2, 70210 Kuopio, Finland d Department of Medicine, Turku University Hospital,University of Turku, Hämeentie 11, PO Box 52, 20521 Turku, Finland e Heart Center, Satakunta Central Hospital, Sairaalantie 3, 28500 Pori, Finland f NeuroCenter, Neurology, Kuopio University Hospital, Puijonlaaksontie 2, 70210 Kuopio, Finland b c
a r t i c l e
i n f o
Article history: Received 16 July 2015 Received in revised form 1 October 2015 Accepted 19 October 2015 Available online 20 October 2015 Keywords: Atrial fibrillation Cardioversion Stroke Transient ischemic attack
a b s t r a c t Background: Cardioversion of atrial fibrillation (AF) is associated with an increased risk for stroke. We identified all cardioversions during the 30 days preceding stroke or transient ischemic attack (TIA) in patients with a previously diagnosed AF, and sought to assess the characteristics of cardioversions leading to stroke or TIA. Methods: FibStroke is a cross-sectional observational multicenter registry that included AF patients with an ischemic stroke or intracranial bleed identified from a discharge registry of four Finnish hospitals. In total 3677 consecutive AF patients suffered 3252 strokes and 956 TIA episodes during 2003–2012. This pre-specified analysis focused on the 1644 events that occurred to patients with paroxysmal or persistent AF at the time of stroke/TIA. Results: A total of 78 strokes and 22 TIA episodes were preceded by a cardioversion. Post-cardioversion strokes accounted for 6.4% of strokes in patients with paroxysmal/persistent AF. Of the 100 cardioversions leading to an ischemic event, 77 were acute and 23 were elective, 63 events occurred in patients not using anticoagulation, and 5 patients had periprocedural INR b 2. Importantly, 21 patients were in low risk of stroke, i.e. CHA2DS2-VASc score b 2. The median delay from cardioversion to event was 2 days. All nine patients who after an unsuccessful cardioversion developed a stroke had a spontaneous cardioversion prior to stroke. Conclusions: Every sixteenth stroke of patients with paroxysmal/persistent AF is preceded by a cardioversion. Most post-cardioversion strokes occur in patients not using oral anticoagulation before cardioversion of acute AF. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Cardioversions are an essential part of rhythm-control strategy in the treatment of atrial fibrillation (AF). Every cardioversion is, however, associated with a temporary increase in stroke risk, although the risk has been less than 1% in patients with adequate anticoagulation or
Abbreviations: AF, atrial fibrillation; CHA2DS2-VASc, congestive heart failure, hypertension, age ≥ 75 (doubled), diabetes mellitus, and prior Stroke, transient ischemic attack or thromboembolism (doubled), vascular disease, Age 65 to 74, sex category (female, unless b65 years and no other risk factors); INR, international normalized ratio; IQR, inter-quartile range; TIA, transient ischemic attack. ☆ All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Heart Center, Turku University Hospital, PO Box 52, 20521 Turku, Finland. E-mail addresses: ajpalo@utu.fi (A. Palomäki), pirjo.mustonen@ksshp.fi (P. Mustonen), juha.hartikainen@kuh.fi (J.E.K. Hartikainen), ilpo.nuotio@tyks.fi (I. Nuotio), tuoski@utu.fi (T. Kiviniemi), [email protected] (A. Ylitalo), paivi.hartikainen@kuh.fi (P. Hartikainen), heidi.lehtola@ksshp.fi (H. Lehtola), riho.luite@kuh.fi (R. Luite), juhani.airaksinen@tyks.fi (K.E.J. Airaksinen).
http://dx.doi.org/10.1016/j.ijcard.2015.10.168 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.
after cardioversion of acute AF [1–4]. Although rhythm control is a tempting primary treatment option, this strategy has offered no mortality benefit compared to rate-control in randomized trials [5,6]. Majority of strokes occurred in these trials in patients with subtherapeutic anticoagulation or in whom warfarin had been stopped after restoring the sinus rhythm. These results highlight the importance of long-term therapeutic anticoagulation in patients with paroxysmal and persistent AF in whom sinus rhythm is pursued. In this pre-specified analysis of the FibStroke study, we sought to assess the clinical characteristics and significance of cardioversion as a risk factor for cerebral thromboembolism in a large population of AF patients who suffered an ischemic stroke or transient ischemic attack (TIA). We hypothesized that cardioversion is a common risk factor for thromboembolic complications in patients with AF and that antithrombotic medication is underused in this population. The main objective of the study was to assess the typical characteristics of cardioversions leading to thromboembolic complication, with the ultimate goal of identifying potential targets for improving clinical practice and increasing patient safety.
270
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273
2. Methods 2.1. Study population The FibStroke study (www.ClinicalTrials.gov, identifier NCT02146040) is part of a wider protocol in progress to assess thrombotic and bleeding complications of AF in Finland [7–9]. The FibStroke registry included all consecutive patients with a previously diagnosed AF who suffered a stroke or intracranial bleeding during the study period. Data for the study was collected from two university hospitals and two central hospitals in Finland. The initial screening was conducted by identifying all patients from the hospital discharge records with the following criteria: 1) The patient had been diagnosed with AF or atrial flutter and 2) the patient had been diagnosed with ischemic stroke, TIA or intracranial bleed between the years 2003–2012 (in one central hospital 2006–2012). A comprehensive list of the ICD 10 codes used for the screening is provided in the Supplementary methods. After the initial screening, all patient files were reviewed individually, and all procedures and cardioversions performed during the 30 days preceding the stroke, TIA or intracranial bleeding were identified. Detailed information about the cardioversion and the patient characteristics at the time of cardioversion and at the time of the stroke, TIA or intracranial bleeding was recorded. Data was collected in a structured electronic case report form. There were a total 3677 patients with a history of AF and who after the diagnosis of AF developed a total of 3252 ischemic strokes and 956 TIA episodes. For this pre-specified analysis we focused on patients in whom an elective or acute cardioversion for paroxysmal or persistent AF was performed during the 30 days preceding the stroke or TIA. As a control for the ischemic events, we also identified 767 patients with a previously diagnosed AF who suffered 794 intracranial bleeds during the study period, and preceding cardioversions for AF were recorded similarly in this subgroup. 2.2. Exclusion We excluded one patient who underwent cardioversion of AF during coronary artery bypass operation. The stroke was discovered on the first postoperative day. One of the included patients developed two strokes after a single cardioversion. The first stroke was hemorrhagic and warfarin was stopped followed by another ischemic stroke 12 days later. The second stroke was excluded from this analysis. 2.3. Ethical issues The study protocol was approved by the Medical Ethics Committee of the Hospital District of Southwest Finland and the ethics committee of the National Institute for Health and Welfare. Informed consent was not required because of the registry nature of the study. The study conforms to the Declaration of Helsinki. 2.4. Definitions Diagnosis of AF was confirmed by 12-lead electrocardiogram according to the standard criteria. Cardioversion was considered successful if sinus rhythm was obtained and the patient was discharged in sinus rhythm. The diagnosis of stroke and TIA were confirmed from the patient records, as diagnosed by the treating neurologist. Only strokes and TIAs considered as definite by the treating physician were included in our study. All patients were imaged by computed tomography or magnetic resonance imaging. 2.5. Statistical analysis Continuous variables were reported as mean ± standard deviation if they were normally distributed, and as median [inter-quartile range
(IQR)] if they were skewed unless stated otherwise. Categorical variables were described with absolute and relative (percentage) frequencies. Comparisons between study subgroups were performed with Mann–Whitney test for continuous and Chi-square test or Fisher's exact test as appropriate for categorical variables. All tests were twosided and statistical significance was set at 5%. Statistical analysis was performed using IBM SPSS Statistics software version 22.0 and SAS version 9.3. This manuscript was written following STROBE guidelines for the reporting of observational studies [10]. 3. Results 3.1. Characteristics of the patients and the ischemic events A total of 3252 ischemic strokes and 956 TIA episodes occurred during the study period. The type of AF was permanent in 2188, paroxysmal or persistent in 1644, and undefined in 376 at the time of the event. We identified a total of 78 ischemic strokes and 22 TIAs in 99 patients that were preceded by a cardioversion of AF. One patient had two ischemic events that were both preceded by a cardioversion. The first was a TIA and second was an ischemic stroke which occurred over two years after the first episode. Clinical characteristics of the patients are depicted in Table 1. Since cardioversions were only performed in patients with paroxysmal or persistent AF, post-cardioversion events accounted for 6.1% of all ischemic events and 6.4% of strokes in this patient group at risk, while 2.4% of all events and 2.4% of all strokes were preceded by a cardioversion. 3.2. Characteristics of the cardioversions Of the 100 cardioversions leading to an ischemic event, 77 were acute and 23 were elective, and 63 of the events occurred to patients not using anticoagulation (Table 2). Similarly, of the 78 cardioversions complicated by a stroke, 61 (78.2%) were acute and 17 (21.8%) were elective, and 51 (65.4%) of the strokes occurred to patients not using anticoagulation. Of the 10 patients who had been using warfarin prior to cardioversion of acute AF leading to stroke, 4 had INR less than 2. The majority of the cardioversions leading to cerebral thromboembolism were successful in terms of restoring sinus rhythm. All nine patients who after an unsuccessful cardioversion developed a stroke had a spontaneous cardioversion before the onset of stroke. Seven of these Table 1 Clinical characteristics of 99 patients who suffered 78 ischemic strokes and 22 transient ischemic attacks in 30 days after cardioversion of atrial fibrillation or atrial flutter. N (%) N Female Age, median (range) Hypertension Diabetes Coronary artery disease Previous myocardial infarction Heart failure Previous stroke or TIA HAS-BLED-score, median [IQR]a 0–2 N2 Mechanical heart valve Pacemaker
99 53 (53.5) 71.1 (47.0–91.8) 66 (66.7) 14 (14.1) 29 (29.3) 17 (17.2) 9 (9.1) 8 (8.1) 2 [1–3] 65 (65.7) 34 (34.3) 3 (3.0) 6 (6.1)
In the one patient suffering two events (one TIA and one stroke over two years apart), characteristics at the time of the first event are depicted. HAS-BLED = hypertension, abnormal renal or liver function (1 point each), stroke, bleeding, labile INRs, elderly (age N 65 years), drugs or alcohol (1 point each); IQR = inter-quartile range; TIA = transient ischemic attack. a Modified HAS-BLED calculated without labile international normalized ratio (INR).
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273 Table 2 Data of 78 cardioversions (CV) leading to ischemic stroke and 22 CVs leading to TIA.
N Stroke TIA Method of CV Electrical Pharmacological Successful CV Spontaneous CV after unsuccessful CV TEE-guided CV Warfarin prior to CV INR at the time of CV, range INR b 2 Periprocedural LMWH Antiplatelet treatment (aspirin or clopidogrel) Warfarin at the time of stroke or TIA INR at the time of stroke or TIA, range INR b 2 CHA2DS2-VASc, median [IQR] 0 1 ≥2 Days from CV to stroke/TIA, median [IQR] 0–2 days 3–6 days N6 days
Acute CV
Elective CV
77 61 (79.2) 16 (20.8)
23 17 (73.9) 6 (26.1)
67 (88.3) 9 (11.7) 68 (88.3) 9/9 3 (3.9) 14 (18.2) 1.2–4.0 5/14 (35.7) 5 (6.8) 45 (60.8) 24 (31.2) 0.9–3.6 13/24 (54.2) 3.0 [2.0–4.0] 8 (10.4) 7 (9.1) 62 (80.5) 2 [1.0–3.0] 56 (72.7) 6 (7.8) 15 (19.5)
23 (100) 0 21 (91.3) 1/2 0 23 (100) 2.0–5.4⁎ 0⁎ 0 3 (15.0) 22 (95.7) 1.3–5.4 5/22 (22.7) 3.0 [1.0–4.0] 2 (8.7) 4 (17.4) 17 (73.96) 2 [2.0–5.0] 12 (52.2) 8 (34.8) 3 (13.0)
271
93% during the first 14 days following the cardioversion (Fig. 1). The median delay from cardioversion to stroke or TIA was 2 days (range 0–28). The delay was slightly longer in patients using warfarin prior to cardioversion than in patients without warfarin (3.0 vs. 2.0 days, p b 0.01). 3.5. Intracranial bleeds
CV = cardioversion; CHA2DS2-VASc = congestive heart failure, hypertension, Age ≥ 75 (doubled), diabetes mellitus, and prior Stroke, transient ischemic attack or thromboembolism (doubled), vascular disease, age 65 to 74, sex category (female, unless b65 years and no other risk factors); INR = international normalized ratio; IQR = interquartile range; LMWH = low molecular weight heparin; TEE = transesophageal echocardiography; TIA = transient ischemic attack. ⁎ Data missing for 3 patients.
nine strokes occurred within two days of the attempted cardioversion, one after eight days, and one after 28 days. Three patients underwent transesophageal echocardiography before the cardioversion. No thrombi were detected in the left atrial appendage, nor were any other contraindications to cardioversion discovered. Two of the patients had been previously treated with ablation. Both suffered an ischemic stroke after acute cardioversion. The first patient was not using warfarin at the time of cardioversion, since it had been stopped after seemingly successful procedure. The second patient had continued warfarin after the ablation, and suffered a stroke despite INR 2.7 at the time of cardioversion. No differences were noted between central and university hospital settings except that pharmacological CV was more often performed in central hospitals (23.3% vs. 2.9%, p b 0.01). 3.3. Risk factors for stroke The CHA2DS2-VASc score was ≥2 in 79 (79%) of the patients who developed any ischemic event, and in 65/78 (83%) of the patients who developed a stroke after cardioversion. Five patients developed a stroke without any risk factors for thromboembolism as measured with the CHA2DS2-VASc score. All of these patients underwent an acute cardioversion and were not receiving peri- or postcardioversion anticoagulation. Age range of the patients was 59 to 64 years, and two were females. Nine patients who developed an ischemic stroke or TIA after the cardioversion had another predisposing factor (operation, myocardial infarction or major bleeding) during the 30 days preceding the stroke. Characteristics of these patients and the competing causes for stroke or TIA are depicted in Supplementary Table 1. 3.4. Timing of cerebral thromboembolism after cardioversion Most ischemic events occurred shortly after the cardioversion, as 68% occurred during the first two days, 83% during the first 7 days and
In the control group of 767 patients and 794 intracranial bleeds, three patients had a history of cardioversion performed within 30 days preceding the bleed, accounting for 0.4% of all intracranial bleedings. Two patients presented with a traumatic bleed and one patient with a spontaneous subdural hematoma 3, 27 and 4 days after the cardioversion, respectively. 4. Discussion The main finding of our study is that cardioversion is an important background factor for stroke in patients with paroxysmal or persistent AF and every sixteenth stroke in these relatively healthy patients was preceded by a cardioversion. The great majority of strokes occurred after cardioversion of an acute AF attack in patients not using oral anticoagulation prior to cardioversion. Importantly, patients with failed cardioversion are not safe from embolic complications, since spontaneous cardioversions were relatively common after attempted cardioversion of acute AF. 4.1. Acute cardioversion Most of the strokes occurred after cardioversion of acute (b 48 h) AF episode, and most of these complications were observed in patients not using anticoagulation. Until recent years, cardioversion of acute AF was considered safe even without anticoagulation. In our FinCV Study reporting data on 5116 successful acute cardioversions performed without anticoagulation the overall risk of thromboembolic events was 0.7% [4]. Thus, the mean risk of stroke in acute cardioversion performed without anticoagulation is in the same range as the risk of elective cardioversion performed with optimal guideline-recommended anticoagulation. Importantly, however, the risk of thromboembolic complications varied from a low risk of 0.2% to a very high risk of 9.8% depending on clustering of clinical stroke risk factors [4]. It is reasonable to assume that the thromboembolic complications could be prevented also in this acute setting with effective anticoagulation, although there is no research evidence on this so far. The current guidelines recommend intravenous heparin or low molecular weight heparin, or factor Xa or direct thrombin inhibitor before or immediately after cardioversion in patients with high stroke risk. In patients with low thromboembolic risk, these therapies may be considered [11]. One important new finding in FibStroke study was that 21% of the patients who developed an ischemic stroke or TIA after cardioversion had a CHA2DS2-VASc score b 2. We have recently shown that a delay of cardioversion exceeding 12 h from the symptom onset caused a N 3-fold increase in the risk of stroke when compared to later cardioversions of acute AF, and the risk of thromboembolism in the low risk patients with CHA2DS2VASc b 2 was as high as 0.9% if no anticoagulation was used [12]. In the light of these findings, it is possible that even low-risk patients might benefit from short term oral anticoagulation started at the time of cardioversion. 4.2. Elective cardioversion Only 23% of ischemic events occurred after elective cardioversion. This finding was not unexpected, since elective cardioversion is always performed during therapeutic anticoagulation. It is, however, noteworthy that even with this strategy the risk of thromboembolic complications has been about 0.5% within 30 days after the cardioversion
272
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273
Fig. 1. Timing of embolic complications after cardioversion of AF. Data of 100 cardioversions leading to 78 ischemic strokes and 22 TIA. AF: atrial fibrillation; TIA: transient ischemic attack.
[13–15]. On the other hand, the average risk of stroke in recently published large randomized multicenter studies has decreased to 0.1% to 0.2% per month during long-term anticoagulation [16–18]. In the light of these figures, it is evident that when a patient is referred to an elective cardioversion of AF we predispose the patient for an increased risk of stroke during the post-cardioversion month even when using the recommended effective anticoagulation. The increased risk is supported by the present findings on high ratio of strokes in relation to intracranial bleeds during the post-cardioversion month (2.4% vs 0.4%) compared with the respective yearly event rates in Re-Ly trial of 1.57% vs. 0.74% [16]. 4.3. Timing of cerebral thromboembolism after cardioversion In line with previous observations [4] the vast majority of strokes and TIAs occurred shortly after the successful cardioversion, with the median time from cardioversion to event of 2 days and over 80% of events occur during the first week after the cardioversion. Only 9 patients developed a stroke after a failed cardioversion. Of note, all these patients had, however, spontaneous cardioversion to sinus rhythm prior to the stroke supporting the view that the conversion of AF to sinus rhythm is the principal factor responsible for stroke in these patients. 4.4. Transesophageal echocardiography Three patients who suffered a stroke had a transesophageal echocardiography performed prior to the cardioversion. In previous studies, the risk of stroke of stroke after cardioversion guided by transesophageal echocardiography has been 0.8% [13,19]. This highlights the fact that exclusion of thrombus before the cardioversion does not guarantee safety, because of the risk of thrombus formation after cardioversion due to left atrial stunning [20]. 4.5. Limitations of the study Some limitations need to be addressed. Retrospective design carries always some limitations in the accuracy of data collection. This is compensated by the large number of events in our study. To our knowledge, this is the largest material of cardioversions leading to stroke or TIA published so far. Only a minority of patients who developed a stroke or TIA had a competing cause for the ischemic event in addition to
cardioversion showing, however, the multitude of clinical scenarios in real-life. The strengths of the study include the identification of all consecutive stroke and TIA patients with a diagnosis of AF from reliable hospital discharge records and the thorough individual case by case review of patient records. 5. Conclusions Every sixteenth stroke of patients with paroxysmal or persistent AF is preceded by a cardioversion of AF. Majority of strokes occur in patients not using oral anticoagulation before cardioversion of acute AF and typically 2–3 days after successful cardioversion. Unsuccessful cardioversion is not without risk for thromboembolism, since later spontaneous cardioversions are not uncommon. Significant proportion of the strokes was observed in relatively young patients with a low CHA2DS2-VASc score supporting the view that at least short-term anticoagulation should always be considered after cardioversion. Additional research is, however, needed to improve the safety of rhythm strategy in the treatment of AF. Disclosure of conflicts of interest Dr. Mustonen has given lectures for Orion, Boehringer Ingelheim, Bayer, Pfizer, Bristol-Myers Squibb, Sanofi-Aventis and Leo Pharma, and has been a member in the advisory boards for Boehringer Ingelheim, Bayer, Pfizer, Bristol-Myers Squibb and Leo Pharma. Dr. Juha Hartikainen has received research grants from the Finnish Foundation for Cardiovascular Research and the European Union Seventh Framework Program, has given lectures for Cardiome, St. Jude Medical and Biotronic, and has been member in the advisory boards for Astra Zeneca, Amgen and Bayer. Dr. Kiviniemi has given lectures for Bayer, Medicines Company, Astra Zeneca and St. Jude Medical, and received research grants from the Finnish Foundation for Cardiovascular Research and Finnish Cardiac Society. Dr. Päivi Hartikainen has received honoraria from Genzyme, Novartis, Biogen Idec, TEVA, and has given lectures for Sanofi-Aventis. Dr. Luite has received honoraria from Webster-Biosense, Biotronik, Bayer and St. Jude Medical, and has given lectures for Bristol-Myers Squibb. Dr. Airaksinen has received research grants from the Finnish Foundation for Cardiovascular Research, has given lectures for Bayer, Cardiome and Boehringer Ingelheim, and has been a member in the advisory boards for Bayer, Astra Zeneca and Boston Scientific. The other authors report no disclosures.
A. Palomäki et al. / International Journal of Cardiology 203 (2016) 269–273
Astra Zeneca and Boston Scientific. The other authors report no disclosures. Acknowledgment of grant support This work was supported by the Finnish Foundation for Cardiovascular Research. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ijcard.2015.10.168. References [1] C.J. Bjerkelund, O.M. Orning, The efficacy of anticoagulant therapy in preventing embolism related to D.C. electrical conversion of atrial fibrillation, Am. J. Cardiol. 23 (1969) 208–216. [2] D.M. Weinberg, J. Mancini, Anticoagulation for cardioversion of atrial fibrillation, Am. J. Cardiol. 63 (1989) 745–746. [3] A.Z. Arnold, M.J. Mick, R.P. Mazurek, F.D. Loop, R.G. Trohman, Role of prophylactic anticoagulation for direct current cardioversion in patients with atrial fibrillation or atrial flutter, J. Am. Coll. Cardiol. 19 (1992) 851–855. [4] K.E. Airaksinen, T. Gronberg, I. Nuotio, et al., Thromboembolic complications after cardioversion of acute atrial fibrillation: the FinCV (Finnish CardioVersion) study, J. Am. Coll. Cardiol. 62 (2013) 1187–1192. [5] D.G. Wyse, A.L. Waldo, J.P. DiMarco, et al., A comparison of rate control and rhythm control in patients with atrial fibrillation, N. Engl. J. Med. 347 (2002) 1825–1833. [6] I.C. Van Gelder, V.E. Hagens, H.A. Bosker, et al., A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation, N. Engl. J. Med. 347 (2002) 1834–1840. [7] K.E. Airaksinen, F. Biancari, P. Karjalainen, et al., Safety of coronary artery bypass surgery during therapeutic oral anticoagulation, Thromb. Res. 128 (2011) 435–439.
273
[8] K.E. Airaksinen, P. Korkeila, J. Lund, et al., Safety of pacemaker and implantable cardioverter-defibrillator implantation during uninterrupted warfarin treatment—the FinPAC study, Int. J. Cardiol. 168 (2013) 3679–3682. [9] R. Virtanen, V. Kryssi, T. Vasankari, M. Salminen, S.L. Kivela, K.E. Airaksinen, Selfdetection of atrial fibrillation in an aged population: the LietoAF study, Eur. J. Prev. Cardiol. 21 (2014) 1437–1442. [10] E. von Elm, D.G. Altman, M. Egger, et al., Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies, BMJ 335 (2007) 806–808. [11] C.T. January, L.S. Wann, J.S. Alpert, et al., 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American college of cardiology/American heart association task force on practice guidelines and the heart rhythm society, J. Am. Coll. Cardiol. 64 (2014) e1–76. [12] I. Nuotio, J.E. Hartikainen, T. Gronberg, F. Biancari, K.E. Airaksinen, Time to cardioversion for acute atrial fibrillation and thromboembolic complications, JAMA 312 (2014) 647–649. [13] A.L. Klein, R.A. Grimm, R.D. Murray, et al., Use of transesophageal echocardiography to guide cardioversion in patients with atrial fibrillation, N. Engl. J. Med. 344 (2001) 1411–1420. [14] R. Cappato, M.D. Ezekowitz, A.L. Klein, et al., Rivaroxaban vs. vitamin K antagonists for cardioversion in atrial fibrillation, Eur. Heart J. 35 (2014) 3346–3355. [15] S. Apostolakis, K.G. Haeusler, M. Oeff, et al., Low stroke risk after elective cardioversion of atrial fibrillation: an analysis of the flec-SL trial, Int. J. Cardiol. 168 (2013) 3977–3981. [16] S.J. Connolly, M.D. Ezekowitz, S. Yusuf, et al., Dabigatran versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 361 (2009) 1139–1151. [17] M.R. Patel, K.W. Mahaffey, J. Garg, et al., Rivaroxaban versus warfarin in nonvalvular atrial fibrillation, N. Engl. J. Med. 365 (2011) 883–891. [18] C.B. Granger, J.H. Alexander, J.J. McMurray, et al., Apixaban versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 365 (2011) 981–992. [19] K. Seidl, M. Rameken, A. Drogemuller, et al., Embolic events in patients with atrial fibrillation and effective anticoagulation: value of transesophageal echocardiography to guide direct-current cardioversion. Final results of the Ludwigshafen observational cardioversion study, J. Am. Coll. Cardiol. 39 (2002) 1436–1442. [20] R.A. Grimm, W.J. Stewart, J.D. Maloney, et al., Impact of electrical cardioversion for atrial fibrillation on left atrial appendage function and spontaneous echo contrast: characterization by simultaneous transesophageal echocardiography, J. Am. Coll. Cardiol. 22 (1993) 1359–1366.