Author + information
- Received May 21, 2018
- Revision received June 21, 2018
- Accepted July 9, 2018
- Published online October 15, 2018.
- Jawad H. Butt, MDa,∗ (, )@uni_copenhagen,
- Jonas B. Olesen, MD, PhDb,
- Eva Havers-Borgersen, MBa,
- Anna Gundlund, MDb,
- Charlotte Andersson, MD, PhDb,
- Gunnar H. Gislason, MD, PhDb,c,d,
- Christian Torp-Pedersen, MD, DMSce,
- Lars Køber, MD, DMSca and
- Emil L. Fosbøl, MD, PhDa
- aDepartment of Cardiology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
- bDepartment of Cardiology, Herlev and Gentofte University Hospital, Hellerup, Denmark
- cThe Danish Heart Foundation, Copenhagen, Denmark
- dThe National Institute of Public Health, University of Southern Denmark, Odense, Denmark
- eDepartment of Health Science and Technology, Aalborg University, Aalborg, Denmark
- ↵∗Address for correspondence:
Dr. Jawad H. Butt, Department of Cardiology, Rigshospitalet, Copenhagen University Hospital, Blegdamsvej 9, 2100 København Ø, Denmark.
Background The long-term risk of thromboembolism in patients developing new-onset post-operative atrial fibrillation (POAF) following noncardiac surgery is unknown, and data on stroke prophylaxis in this setting are lacking.
Objectives The purpose of this study was to assess the long-term risk of thromboembolism in patients developing new-onset POAF following noncardiac surgery relative to patients with nonsurgical, nonvalvular atrial fibrillation (NVAF).
Methods Using Danish nationwide registries, the authors identified all patients who developed POAF following noncardiac surgery from 1996 to 2015. These were matched by age, sex, heart failure, hypertension, diabetes, previous thromboembolism, ischemic heart disease, and year of diagnosis to patients with nonsurgical NVAF in a 1:4 ratio. Comparative long-term risk of thromboembolism was examined by multivariable Cox regression models.
Results In patients undergoing noncardiac surgery, 6,048 (0.4%) developed POAF during hospitalization, with the highest incidences following thoracic/pulmonary, vascular, and abdominal surgery. A total of 3,830 patients with POAF were matched with 15,320 patients with NVAF. Oral anticoagulation therapy was initiated within 30 days post-discharge in 24.3% and 41.3% of these patients, respectively (p value <0.001). The long-term risk of thromboembolism was similar in patients with POAF and NVAF (31.7 events vs. 29.9 events per 1,000 person years; hazard ratio [HR]: 0.95; 95% confidence interval [CI]: 0.85 to 1.07). Anticoagulation therapy during follow-up was associated with a comparably lowered risk of thromboembolic events in patients with POAF (HR: 0.52; 95% CI: 0.40 to 0.67) as well as NVAF (HR: 0.56; 95% CI: 0.51 to 0.62) compared with no anticoagulation therapy.
Conclusions New-onset POAF following noncardiac surgery was associated with a long-term risk of thromboembolism similar to NVAF.
New-onset post-operative atrial fibrillation (POAF) is a common complication in cardiac surgery (1–3). However, little attention has been paid to POAF in the setting of noncardiac surgery. Although the reported incidence of new-onset POAF in patients undergoing noncardiac surgery ranges from 0.3% to 4.1% (1,4–8), the large number of noncardiac surgical procedures performed worldwide makes it important to elucidate the clinical burden of and long-term risks associated with this complication. It is well-established that patients with nonvalvular atrial fibrillation (NVAF) carry a 5-fold increased risk of ischemic stroke and systemic embolism (9,10). However, it is unknown whether new-onset atrial fibrillation (AF) secondary to noncardiac surgery differs from NVAF in terms of long-term thromboembolic risk. In addition, data on practice patterns and outcomes associated with use of anticoagulation in new-onset POAF following noncardiac surgery are lacking, and international guidelines on the management of patients with AF do not provide any recommendations regarding oral anticoagulation (OAC) therapy in this setting (11,12). These gaps in knowledge prompted us to conduct a Danish nationwide retrospective cohort study to examine stroke prophylaxis and the long-term risk of thromboembolism in patients with new-onset POAF following noncardiac surgery relative to a matched cohort of patients with NVAF.
All residents in Denmark are assigned a unique and permanent civil registration number allowing accurate linkage of nationwide administrative registries at an individual level over time. For this study, data from the following nationwide administrative registries were collected: The Danish National Patient Registry, which holds information on all hospital admissions and outpatient contacts according to the International Classification of Diseases and all surgical procedures according to the NOMESCO Classification of Surgical Procedures (13); the Danish Registry of Medicinal Product Statistics (the Danish National Prescription Registry), which contains detailed information on dispensing date, strength, and quantity on all claimed drug prescriptions in Denmark (14); and the Danish Civil Registration System, which holds information on vital status, birth date, and sex (15).
All Danish residents above 30 years of age undergoing first-time surgery between January 1, 1996, and June 30, 2015, were identified. Patients were included in the study if they: 1) underwent noncardiac, nonobstetrical surgery (see Online Table 1 for NOMESCO Classification of Surgical Procedures codes); 2) had no history of AF (i.e., those who did not have a primary or secondary in-hospital or outpatient diagnosis of AF or prescriptions on antiarrhythmic drugs [i.e., digoxin, flecainide, amiodarone, and dronedarone] any time prior to hospitalization for surgery); 3) developed POAF (defined as a primary or secondary hospital discharge diagnosis of AF during hospitalization); 4) were not diagnosed with cancer within 1 year prior to admission or during admission; 5) had not redeemed any OAC prescriptions 6 months prior to surgery; and 6) were alive at discharge (Figure 1). To compare the risk of thromboembolism between patients with POAF and NVAF, we identified a nationwide cohort of patients diagnosed with nonsurgical NVAF during hospitalization who: 1) were not prescribed antiarrhythmic drugs any time prior to diagnosis; 2) were not prescribed OAC 6 months prior to diagnosis; 3) had not undergone any cardiac surgery prior to diagnosis; and 4) were not diagnosed with cancer within 1 year prior to diagnosis. Using risk-set matching, patients with POAF were matched by age (up to 2 years difference), sex, heart failure, hypertension, diabetes, previous thromboembolism, ischemic heart disease, and year of index date to patients with NVAF in a 1:4 ratio. The index date was defined as the discharge date for both POAF and NVAF patients.
Comorbidity was obtained using hospital discharge diagnoses any time prior to index (Online Table 2 for ICD-8 and -10 codes). Patients with diabetes and hypertension were identified using claimed drug prescriptions, as done previously (16,17). Alcohol abuse was defined from related prescription fills and ICD-10 diagnosis codes. Pharmacotherapy at baseline was defined as claimed prescriptions within 180 days prior to date of surgery or first diagnosis for patients with POAF and NVAF, respectively (see Online Table 3 for ATC codes). The estimated risk of stroke (CHADS2- and CHA2DS2-VASc score [congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke/transient ischemic attack, vascular disease, age 65 to 75 years, and sex category]) and bleeding (HAS-BLED score [hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile international normalized ratio, elderly (age >65 years), drugs/alcohol concomitantly]) was calculated as done previously (18,19).
Post-discharge anticoagulation therapy
OAC therapy (vitamin K or nonvitamin K antagonist OAC) was assessed continuously for each individual during follow-up using claimed prescriptions, taking dosage and packing size into account, as done previously (20,21). We defined exposure to OAC therapy as the point at which patients had medication available and defined discontinuation as the point at which patients had no more medication available for at least 30 days. Patients were allowed to change exposure status during the study period according to claimed OAC prescriptions.
The primary outcome was thromboembolism (a composite of ischemic stroke, transient cerebral ischemia, and thrombosis or embolism in peripheral arteries). Secondary outcomes included AF rehospitalization, defined as a new hospital admission with AF as the primary diagnosis, and all-cause mortality. The diagnoses of AF and ischemic stroke in the Danish National Patient Registry have previously been validated with a positive predictive value of 92.6% and 97%, respectively (22,23). Patients were followed from the index date until occurrence of the outcome of interest, emigration, 10 years after index, or end of the study (December 31, 2015), whichever came first.
Descriptive data were reported as frequencies and percentages or median (interquartile range [IQR]) as appropriate. Differences in baseline characteristics were tested by applying the chi-square test for categorical variables and the Mann-Whitney U test for continuous variables. Crude incidence rates were calculated as number of events per 1,000 person-years. The cumulative incidence of thromboembolism and AF rehospitalization according to groups was estimated using the Aalen-Johansen estimator, while taking into account the competing risk of death, and differences between groups were assessed using Gray’s test. Survival curves according to groups were constructed by the Kaplan-Meier method, and differences between groups were assessed using the log-rank test. Cox proportional hazard regression models conditional on the matching (i.e., comparing cases with their matched control subjects) were used to estimate hazard ratios (HRs) with 95% confidence intervals (CIs), adjusted for comorbidity, concomitant pharmacotherapy, and OAC therapy as a time-dependent covariate. Patients with NVAF served as the reference group in all models. The proportional hazards assumption was tested; because the assumption was not met in the analysis of death, this analysis was stratified in 2 time periods. Clinically relevant interactions, including age, sex, several comorbidities, and OAC treatment during follow-up, were tested for and found insignificant, unless otherwise stated. All statistical analyses were performed with SAS statistical software version 9.4 (SAS Institute, Cary, North Carolina). A 2-sided p value <0.05 was considered statistically significant.
This study was approved by the Danish Data Protection Agency (No. 2007-58-0015; internal reference: GEH-2014-013, I-Suite no. 02731). In Denmark, ethical approval is not required for register-based studies in which individuals cannot be identified.
From January 1, 1996, to June 30, 2015, 1,520,109 patients with no history of AF underwent noncardiac surgery, of which 6,048 (0.4%) patients developed POAF during hospitalization. After exclusion criteria were applied, 3,830 patients with POAF were matched with 15,320 patients with NVAF. A flow chart of the study population selection process is presented in Figure 1. The majority of patients with NVAF in the matched cohort had a primary discharge diagnosis of AF (65.4%). For patients who had AF registered as a secondary diagnosis (i.e., 34.6%), pneumonia (24%) and arterial thromboembolism (17%) were the most common primary diagnoses. Baseline characteristics according to groups are summarized in Table 1. The median age of the study population was 77 years (IQR: 69 to 84 years) and 43.2% were men. The mean CHA2DS2-VASc score in the population was 3.0 ± 1.7. The POAF group was characterized by a greater prevalence of noncardiovascular comorbidities, with the exception of a history of cancer and chronic obstructive pulmonary disease, compared with the NVAF group.
POAF following noncardiac surgery
Baseline characteristics of patients with no history of AF undergoing noncardiac surgery according to the development of POAF are presented in Online Table 4. Overall, the most common type of surgery was orthopedic (30.2% of all procedures). Patients who developed POAF were older and characterized by greater prevalence of cardiovascular and noncardiovascular comorbidities compared with those who did not develop POAF.
Post-discharge antiarrhythmic and anticoagulation therapy
In the POAF group, 109 (2.9%), 1,348 (35.2%), and 3 (0.1%) patients received amiodarone, digoxin, and flecainide, respectively, within 30 days after the index date. Correspondingly, within 30 days, 423 (2.8%), 5,913 (38.6%), and 55 (0.4%) patients in the NVAF group received amiodarone, digoxin, and flecainide, respectively.
In the POAF group, 859 (24.4%) patients initiated OAC therapy (of whom 76.4% received warfarin) within 30 days after the index date. Correspondingly, 6,234 (41.5%) patients initiated OAC therapy (of whom 80.3% received warfarin) in the NVAF group within 30 days. The proportion of patients receiving OAC therapy within 30 days according to CHA2DS2-VASc score, HAS-BLED score, and year of surgery or diagnosis in the POAF and NVAF group are displayed in Online Tables 5A and 5B and Online Figure 1, respectively. Among patients with POAF who initiated OAC therapy within 30 days after index and were alive, 844 (98.6%) continued treatment after 1 month, 548 (65.5%) after 3 months, 405 (50.1%) after 6 months, and 264 (35.5%) after 1 year.
The median follow-up time from the index date until occurrence of a thromboembolic event, death, emigration, or end of the study period was 3.2 years (IQR: 0.9 to 7.0 years) and 3.8 years (IQR: 1.6 to 7.6 years) for patients with POAF and NVAF, respectively. During follow-up, 496 (13.0%) patients with POAF and 2,086 (13.6%) patients with NVAF experienced a thromboembolic event. The crude incidence rates of thromboembolism were 31.7 events (95% CI: 29.0 to 34.6 events) and 29.9 events (95% CI: 28.7 to 31.2 events) per 1,000 person-years for patients with POAF and NVAF, respectively. The cumulative incidences of thromboembolism according to groups are displayed in Figure 2A. In multivariable Cox proportional hazard analysis, POAF was associated with a similar risk of thromboembolism compared with NVAF (HR: 0.95; 95% CI: 0.85 to 1.07). No interaction between OAC treatment during follow-up and groups was found (p for interaction = 0.80). In both patients with POAF and NVAF, OAC therapy during follow-up was associated with a significantly lower risk of thromboembolism (Figure 3).
AF rehospitalization and all-cause mortality
The crude incidence rates of AF rehospitalization were 48.2 events (95% CI: 44.6 to 52.0 events) and 89.8 events (95% CI: 87.2 to 92.4 events) per 1,000 person-years for POAF and NVAF patients, respectively. The cumulative incidences of AF rehospitalization in POAF and NVAF patients are illustrated in Figure 2B. In multivariable Cox proportional hazard analysis, patients with POAF had a significantly lower risk of AF rehospitalization compared with NVAF patients (HR: 0.58; 95% CI: 0.53 to 0.63).
The crude incidence rates of all-cause mortality were 133.0 events (95% CI: 127.6 to 138.7 events) and 108.5 events (95% CI: 106.2 to 110.9 events) per 1,000 person years for POAF and NVAF patients, respectively. Figure 2C depicts Kaplan-Meier curves for death in POAF and NVAF patients. In multivariable Cox proportional hazard analysis, POAF was associated with a significantly higher risk of all-cause mortality compared with NVAF during the first year (HR: 1.83; 95% CI: 1.67 to 2.01). After 1 year, POAF was associated with a similar risk of all-cause mortality compared with NVAF (HR: 1.00; 95% CI: 0.93 to 1.07). OAC therapy during follow-up was associated with a significantly lower risk of all-cause mortality in both patients with POAF and NVAF (Figure 3).
A number of sensitivity analyses were performed to test the robustness of our findings:
1. We restricted the primary outcome of thromboembolism to ischemic stroke and found a similar association as the main analysis (HR: 0.94; 95% CI: 0.84 to 1.07).
2. We compared the risk of thromboembolism in patients with POAF with an unmatched cohort of patients with NVAF and found consistent results (HR: 1.04; 95% CI: 0.96 to 1.13).
3. We restricted the POAF population to patients who were prescribed beta-blockers (i.e., only those who were not receiving beta-blockers prior to admission), digoxin, flecainide, or amiodarone within 30 days after discharge. The 30-day blanking period was set to ensure that patients had time to claim their prescriptions at a pharmacy, and follow-up was therefore started from 30 days post-discharge in this analysis. As in the main analysis, we found that patients with POAF who were prescribed digoxin, flecainide, amiodarone, or beta-blockers within 30 days after discharge (n = 1,812) had a similar associated risk of thromboembolism compared with a matched cohort of patients with NVAF (n = 7,248) (adjusted HR: 0.93; 95% CI: 0.79 to 1.09).
4. We compared the risk of AF rehospitalization among groups from 90 days after index (90 days grace period). As in the main analysis, this analysis demonstrated that patients with POAF had a significantly lower risk of AF rehospitalization compared with patients with NVAF (HR: 0.60; 95% CI: 0.55 to 0.66).
5. Finally, we compared the risk of outcomes in patients developing and not developing POAF following noncardiac surgery. Patients developing POAF were matched by the components of the CHA2DS2-VASc score, type of surgery, and year of diagnosis to patients not developing POAF in a 1:1 ratio (Online Table 6). Patients with POAF following noncardiac surgery had a higher associated risk of thromboembolism (adjusted HR: 1.89; 95% CI: 1.56 to 2.29), AF rehospitalization (adjusted HR: 5.26; 95% CI: 4.01 to 6.89), and all-cause mortality (adjusted HR: 1.67; 95% CI: 1.53 to 1.83) compared with those who did not develop POAF following noncardiac surgery (Online Figure 2).
In this nationwide cohort study, we examined the long-term risk of thromboembolism associated with new-onset POAF following noncardiac surgery relative to a matched cohort of patients with NVAF. The study yielded 3 major findings: first, among patients with no history of AF undergoing noncardiac surgery, 0.4% developed POAF during hospitalization, with the highest incidences following thoracic/pulmonary, vascular, and abdominal surgery; second, POAF was associated with a similar risk of thromboembolism despite a lower risk of AF rehospitalization compared with NVAF (Central Illustration); and third, OAC therapy was associated with a comparably lowered risk of thromboembolic events in both the POAF and NVAF group.
The pathophysiology of POAF is complex and not fully understood. In cardiothoracic surgery, direct manipulation of the heart, local inflammation, and elevation in atrial pressure from post-operative ventricular stunning may contribute to alterations in atrial refractoriness and/or local re-entry leading to POAF (24). In noncardiothoracic surgery, a combination of several mechanisms and factors, including sympathetic activation, systemic inflammation, electrophysiological disturbances, metabolic imbalances, hypoxia, and hypervolemia, have been proposed to provoke POAF in predisposed patients (25,26). The pathophysiology may likely explain the higher incidence of POAF in the setting of thoracic/pulmonary, vascular, and abdominal surgery compared with other types of noncardiac surgery.
Although traditionally thought to be transient and benign, mounting evidence suggest that new-onset POAF is associated with a greater risk of perioperative complications and ischemic stroke following cardiac surgery (4,27–31). Further, the association between POAF and adverse outcomes has been less studied in the setting of noncardiac surgery. Gialdini et al. (4) found that new-onset AF in conjunction with noncardiac surgery was associated with an increased long-term risk of ischemic stroke compared with no development of AF. Based on these administrative data, the authors suggested that perioperative AF may be similar to other etiologies of AF in regard to future thromboembolic risk. However, the study did not account for post-discharge OAC therapy due to the lack of data on medication use. To our knowledge, our study is the first to evaluate the long-term risk of thromboembolism in POAF relative to NVAF with data on post-discharge OAC therapy. After adjusting for comorbidity, concomitant pharmacotherapy, and post-discharge OAC treatment, we found that patients with new-onset POAF following noncardiac surgery carried a similar risk of thromboembolism compared with patients with primary NVAF. These data support that new-onset POAF should be regarded as primary NVAF in terms of long-term thromboembolic risk.
Despite a similar long-term risk of thromboembolism, we found that patients with POAF had a significantly lower risk of AF rehospitalization compared with patients with primary NVAF. Interestingly, in a study including patients without a history of AF with recently implanted pacemakers or defibrillators, short episodes of subclinical AF were associated with a significantly increased risk of subsequent ischemic stroke or systemic embolism (32). Although speculative, this may in part explain the similar long-term risk of thromboembolism in patients with POAF and NVAF.
Although patients with POAF following noncardiac surgery have a similar long-term risk of thromboembolism as compared with patients with NVAF, the question remains whether these patients should initiate OAC therapy for the prevention of thromboembolic events. The most recent guidelines for the management of patients with AF briefly address the role of anticoagulation in POAF. However, these guidelines do not provide any recommendations regarding OAC therapy in the setting of noncardiac surgery. In both the American and European guidelines, OAC therapy in POAF following cardiac surgery is a Class IIa recommendation (Level of Evidence: B) (11,12). As suggested by these guidelines, high-quality evidence regarding the role of OAC therapy in patients developing POAF is sparse. In our study, only 24.3% of patients developing POAF following noncardiac surgery initiated OAC treatment within 30 days of discharge despite a high predicted stroke risk and a moderate predicted bleeding risk. In comparison, 37.1% of patients with primary NVAF initiated OAC therapy within 30 days of discharge. Although the proportion of patients initiating OAC therapy was low in both groups, a steep increase was observed after 2010 (in 2015, 67.4% patients in the NVAF group and 48% patients in the POAF group initiated OAC therapy within 30 days). Interestingly, we found that OAC therapy during follow-up was associated with a comparably lowered risk of thromboembolic events and all-cause mortality in both the POAF and NVAF group, suggesting a similar effectiveness of OAC for the prevention of stroke. Thus, our data indicate the need for guidelines regarding OAC therapy for patients developing POAF following noncardiac surgery. However, to support our findings, studies specifically addressing the role of OAC therapy in the setting of POAF are warranted to examine the efficacy and safety as well as the timing and duration of OAC therapy, preferably in a randomized, controlled fashion.
Although the primary outcome in this study was thromboembolism, mortality is of great importance. We found that POAF was associated with a higher risk of all-cause mortality within the first year post-discharge—particularly within the first months—compared with NVAF. However, after 1 year, the risk of all-cause mortality was similar among groups. These findings most likely reflect the indication for surgery (i.e., a higher comorbidity burden) and possibly the surgical procedure itself.
Study strengths and limitations
The main strength of this study is the completeness of data in a large nationwide unselected cohort of patients undergoing noncardiac surgery and patients with primary NVAF followed in a real-world setting. The Danish health care system, funded by taxes, provides equal access to health care services for all residents regardless of socioeconomic or insurance status. In Denmark, OACs can be purchased only through prescription. Due to partial reimbursement of drug expenses by the Danish health care system, pharmacies are required to register all redeemed prescriptions ensuring complete and accurate registration. The findings of this study should be viewed in the context of a number of limitations. The observational nature of this study precludes the assessment of cause-effect relationships. The possibility of residual confounding cannot be excluded despite adjustment for potential confounders in the Cox regression models. Moreover, no causal inference can be drawn from the OAC effectiveness analyses. Although our data suggest that OAC therapy is associated with a similar lower thromboembolic risk in both groups, this association may be susceptible to confounding by indication (i.e., healthier patients receiving OAC therapy) and thus, does not necessarily indicate a clinical benefit of OAC therapy in patients developing POAF following noncardiac surgery. These findings underscore the need for future randomized clinical trials investigating the effectiveness of OAC therapy in POAF. Clinicians’ may not record episodes of POAF that are deemed clinically insignificant; thus, short episodes of POAF may have been missed. Further, patients are not necessarily monitored with continuous electrocardiography following noncardiac surgery. In line with other population-based studies, our study may likely have underestimated the proportion of patients developing POAF following noncardiac surgery, as some small-scale prospective studies have reported an incidence of POAF up to 10.2% following noncardiac surgery (33) and 26% following colorectal surgery (34). However, patients included in our study were at least 30 years of age and thus younger compared with other studies. Further, we did not have any information on duration of POAF or number of episodes with AF during admission, discharge rhythm, indication for OAC therapy, or factors affecting the clinicians’ decision to prescribe OACs. In addition, data on important clinical parameters, such as plasma creatinine levels, international normalized ratio, body mass index, and smoking habits, as well as electrocardiograms, echocardiograms, urgency of surgery (i.e., elective or emergency), and AF symptom burden during follow-up were not available. The lack of data on international normalized ratio during follow-up precludes the assessment of the quality of anticoagulation in either of the groups and may potentially have an effect on outcomes. Further, we did not have data on duration or recurrence of AF during admission in patients with NVAF.
New-onset POAF following noncardiac surgery was associated with a similar long-term thromboembolic risk compared with NVAF. Although OAC therapy was associated with a comparably lowered risk of thromboembolic events in patients with POAF and NVAF, more studies addressing the role of OAC therapy in POAF in the setting of noncardiac surgery are warranted to examine the efficacy and safety as well as the timing and duration of OAC therapy.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: The long-term risk of thromboembolism in patients developing AF following noncardiac surgery is similar to that in patients with other forms of nonvalvular AF.
TRANSLATIONAL OUTLOOK: Future studies should address the timing of initiation, safety, efficacy, and duration of anticoagulation therapy for patients who develop post-operative AF.
Dr. Olesen has served as a speaker for Bristol-Myers Squibb, Boehringer Ingelheim, Bayer, and AstraZeneca; has served as a consultant for Bristol-Myers Squibb, Boehringer Ingelheim, Novartis Healthcare, and Novo Nordisk; and has received funding for research from Bristol-Myers Squibb and The Capital Region of Denmark, Foundation for Health Research. Dr. Gundlund has received funding from Bristol-Myers Squibb. Dr. Torp-Pedersen has received a grant and personal fees from Bayer; and has received a grant from Biotronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- International Classification of Diseases
- nonvalvular atrial fibrillation
- oral anticoagulation
- post-operative atrial fibrillation
- Received May 21, 2018.
- Revision received June 21, 2018.
- Accepted July 9, 2018.
- 2018 American College of Cardiology Foundation
- Walkey A.J.,
- Benjamin E.J.,
- Lubitz S.A.
- Wolf P.A.,
- Abbott R.D.,
- Kannel W.B.
- January C.T.,
- Wann L.S.,
- Alpert J.S.,
- et al.
- Schramm T.K.,
- Gislason G.H.,
- Kober L.,
- et al.
- Olesen J.B.,
- Lip G.Y.,
- Hansen M.L.,
- et al.
- Gislason G.H.,
- Jacobsen S.,
- Rasmussen J.N.,
- et al.
- Joshi K.K.,
- Tiru M.,
- Chin T.,
- Fox M.T.,
- Stefan M.S.
- Bessissow A.,
- Khan J.,
- Devereaux P.J.,
- Alvarez-Garcia J.,
- Alonso-Coello P.
- Villareal R.P.,
- Hariharan R.,
- Liu B.C.,
- et al.
- Steinberg B.A.,
- Zhao Y.,
- He X.,
- et al.
- Mariscalco G.,
- Klersy C.,
- Zanobini M.,
- et al.
- Kosmidou I.,
- Chen S.,
- Kappetein A.P.,
- et al.