Effect of Continuous Electrocardiogram Monitoring on Detection of Undiagnosed Atrial Fibrillation After Hospitalization for Cardiac Surgery: A Randomized Clinical Trial | Atrial Fibrillation | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Visual Abstract. Effect of Continuous Electrocardiogram Monitoring on Detection of Undiagnosed Atrial Fibrillation After Hospitalization for Cardiac Surgery
Effect of Continuous Electrocardiogram Monitoring on Detection of Undiagnosed Atrial Fibrillation After Hospitalization for Cardiac Surgery
Figure 1.  Randomization and Patients in the Post-Surgical Enhanced Monitoring for Cardiac Arrhythmias and Atrial Fibrillation (SEARCH-AF) Trial
Randomization and Patients in the Post-Surgical Enhanced Monitoring for Cardiac Arrhythmias and Atrial Fibrillation (SEARCH-AF) Trial

A total of 2803 patients were screened for the study and 336 patients underwent randomization. In the group that was assigned to receive continuous monitoring, 8 patients did not undergo any monitoring and 2 patients completed the monitoring but the data were not received by the monitoring center. In the usual care group, no patient underwent continuous monitoring and 4 patients underwent non–protocol-mandated Holter monitoring. During follow-up, 6 patients withdrew consent in both groups. There were 6 and 11 patients who were lost to follow-up in the intervention and usual care group, respectively. A total of 163 patients in the monitoring group and 173 patients in the usual care group were included in the primary analysis. A detailed list of reasons for screen failure is provided in eTable 3 in Supplement 2. CHA2DS2-VASc indicates congestive heart failure, hypertension, age 75 years or older, diabetes, prior stroke or transient ischemic attack, vascular disease, age 65 to 74 years, and female sex.

Figure 2.  Time to First Event Curves for the Primary End Point
Time to First Event Curves for the Primary End Point

Kaplan-Meier estimates of detection of cumulative atrial fibrillation (AF) or atrial flutter (AFL) lasting 6 minutes or longer or documentation of AF or AFL by a single 12-lead electrocardiogram within 30 days after randomization. The primary end point occurred in 32 patients (19.6%) in the monitoring group and in 3 patients (1.7%) in the usual care group for an absolute difference of 17.9% (95% CI, 11.5%-24.3%; P < .001).

Figure 3.  Cumulative Duration of Detected Atrial Fibrillation (AF) or Atrial Flutter (AFL) Episodes (per Patient) Within the First 30 Days of Randomization
Cumulative Duration of Detected Atrial Fibrillation (AF) or Atrial Flutter (AFL) Episodes (per Patient) Within the First 30 Days of Randomization

Box and whisker plots of the cumulative duration of detected AF or AFL durations for patients randomized to continuous cardiac rhythm monitoring within the first 30 days of randomization. The upper and lower ends of the box represent the first and third quartile. The line in the box represents the median. In week 2, the median was 0.2 minutes. In week 1, the whisker above the upper quartile is drawn up to the largest observed point that falls within 1.5 times of the interquartile range (IQR). In week 2, the whisker is 1.5 times of the IQR. The whiskers below the lower quartile are not illustrated because they cross zero. Outliers are presented as circles located outside the whiskers. There were 10 patients who had AF lasting 6 minutes or longer in multiple monitoring weeks.

Table 1.  Baseline Demographic and Clinical Characteristics
Baseline Demographic and Clinical Characteristics
Table 2.  Primary and Secondary Outcomes by Intention-to-Treat Analysis
Primary and Secondary Outcomes by Intention-to-Treat Analysis
1.
Lin  MH, Kamel  H, Singer  DE, Wu  YL, Lee  M, Ovbiagele  B.  Perioperative/postoperative atrial fibrillation and risk of subsequent stroke and/or mortality.   Stroke. 2019;50(6):1364-1371. doi:10.1161/STROKEAHA.118.023921PubMedGoogle ScholarCrossref
2.
Kaw  R, Hernandez  AV, Masood  I, Gillinov  AM, Saliba  W, Blackstone  EH.  Short- and long-term mortality associated with new-onset atrial fibrillation after coronary artery bypass grafting: a systematic review and meta-analysis.   J Thorac Cardiovasc Surg. 2011;141(5):1305-1312. doi:10.1016/j.jtcvs.2010.10.040PubMedGoogle ScholarCrossref
3.
Ambrosetti  M, Tramarin  R, Griffo  R,  et al; ISYDE and ICAROS Investigators of the Italian Society for Cardiovascular Prevention, Rehabilitation and Epidemiology (IACPR-GICR).  Late postoperative atrial fibrillation after cardiac surgery: a national survey within the cardiac rehabilitation setting.   J Cardiovasc Med (Hagerstown). 2011;12(6):390-395. doi:10.2459/JCM.0b013e328346a6d3PubMedGoogle ScholarCrossref
4.
Cioffi  G, Mureddu  G, Cemin  C,  et al.  Characterization of post-discharge atrial fibrillation following open-heart surgery in uncomplicated patients referred to an early rehabilitation program.   Ital Heart J. 2001;2(7):519-528.PubMedGoogle Scholar
5.
Giaccardi  M, Macchi  C, Colella  A,  et al.  Postacute rehabilitation after coronary surgery: the effect of preoperative physical activity on the incidence of paroxysmal atrial fibrillation.   Am J Phys Med Rehabil. 2011;90(4):308-315. doi:10.1097/PHM.0b013e31820f9535PubMedGoogle ScholarCrossref
6.
Funk  M, Richards  SB, Desjardins  J, Bebon  C, Wilcox  H.  Incidence, timing, symptoms, and risk factors for atrial fibrillation after cardiac surgery.   Am J Crit Care. 2003;12(5):424-433. doi:10.4037/ajcc2003.12.5.424PubMedGoogle ScholarCrossref
7.
Lowres  N, Mulcahy  G, Gallagher  R,  et al.  Self-monitoring for atrial fibrillation recurrence in the discharge period post-cardiac surgery using an iPhone electrocardiogram.   Eur J Cardiothorac Surg. 2016;50(1):44-51. doi:10.1093/ejcts/ezv486PubMedGoogle ScholarCrossref
8.
Axtell  AL, Moonsamy  P, Melnitchouk  S,  et al.  Preoperative predictors of new-onset prolonged atrial fibrillation after surgical aortic valve replacement.   J Thorac Cardiovasc Surg. 2020;159(4):1407-1414. doi:10.1016/j.jtcvs.2019.04.077PubMedGoogle ScholarCrossref
9.
Ayoub  K, Habash  F, Almomani  A,  et al.  Long term risk of recurrent atrial fibrillation and ischemic stroke after post-operative atrial fibrillation complicating cardiac and non-cardiac surgeries.   J Atr Fibrillation. 2018;10(6):1660. doi:10.4022/jafib.1660PubMedGoogle Scholar
10.
Thorén  E, Wernroth  ML, Christersson  C, Grinnemo  KH, Jidéus  L, Ståhle  E.  Compared with matched controls, patients with postoperative atrial fibrillation (POAF) have increased long-term AF after CABG, and POAF is further associated with increased ischemic stroke, heart failure and mortality even after adjustment for AF.   Clin Res Cardiol. 2020;109(10):1232-1242. doi:10.1007/s00392-020-01614-zPubMedGoogle ScholarCrossref
11.
Wang  EY, Hulme  OL, Khurshid  S,  et al.  Initial precipitants and recurrence of atrial fibrillation.   Circ Arrhythm Electrophysiol. 2020;13(3):e007716. doi:10.1161/CIRCEP.119.007716PubMedGoogle Scholar
12.
El-Chami  MF, Merchant  FM, Smith  P,  et al.  Management of new-onset postoperative atrial fibrillation utilizing insertable cardiac monitor technology to observe recurrence of AF (MONITOR-AF).   Pacing Clin Electrophysiol. 2016;39(10):1083-1089. doi:10.1111/pace.12949PubMedGoogle ScholarCrossref
13.
Lomivorotov  VV, Efremov  SM, Pokushalov  EA,  et al.  Randomized trial of fish oil infusion to prevent atrial fibrillation after cardiac surgery: data from an implantable continuous cardiac monitor.   J Cardiothorac Vasc Anesth. 2014;28(5):1278-1284. doi:10.1053/j.jvca.2014.02.019PubMedGoogle ScholarCrossref
14.
Chen  YL, Zeng  M, Liu  Y,  et al.  CHA2DS2-VASc score for identifying patients at high risk of postoperative atrial fibrillation after cardiac surgery: a meta-analysis.   Ann Thorac Surg. 2020;109(4):1210-1216. doi:10.1016/j.athoracsur.2019.07.084PubMedGoogle ScholarCrossref
15.
Kirchhof  P, Benussi  S, Kotecha  D,  et al; ESC Scientific Document Group.  2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS.   Eur Heart J. 2016;37(38):2893-2962. doi:10.1093/eurheartj/ehw210PubMedGoogle ScholarCrossref
16.
January  CT, Wann  LS, Alpert  JS,  et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines.  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. 2014;64(21):e1-e76. doi:10.1016/j.jacc.2014.03.022PubMedGoogle ScholarCrossref
17.
Verma  A, Cairns  JA, Mitchell  LB,  et al; CCS Atrial Fibrillation Guidelines Committee.  2014 Focused update of the Canadian Cardiovascular Society Guidelines for the management of atrial fibrillation.   Can J Cardiol. 2014;30(10):1114-1130. doi:10.1016/j.cjca.2014.08.001PubMedGoogle ScholarCrossref
18.
Yagdi  T, Nalbantgil  S, Ayik  F,  et al.  Amiodarone reduces the incidence of atrial fibrillation after coronary artery bypass grafting.   J Thorac Cardiovasc Surg. 2003;125(6):1420-1425. doi:10.1016/S0022-5223(02)73292-3PubMedGoogle ScholarCrossref
19.
Maesen  B, Nijs  J, Maessen  J, Allessie  M, Schotten  U.  Post-operative atrial fibrillation: a maze of mechanisms.   Europace. 2012;14(2):159-174. doi:10.1093/europace/eur208PubMedGoogle ScholarCrossref
20.
Kerwin  M, Saado  J, Pan  J,  et al.  New-onset atrial fibrillation and outcomes following isolated coronary artery bypass surgery: a systematic review and meta-analysis.   Clin Cardiol. 2020;43(9):928-934. doi:10.1002/clc.23414PubMedGoogle ScholarCrossref
21.
Gialdini  G, Nearing  K, Bhave  PD,  et al.  Perioperative atrial fibrillation and the long-term risk of ischemic stroke.   JAMA. 2014;312(6):616-622. doi:10.1001/jama.2014.9143PubMedGoogle ScholarCrossref
22.
Chao  TF, Liu  CJ, Chen  SJ,  et al.  CHADS2 score and risk of new-onset atrial fibrillation: a nationwide cohort study in Taiwan.   Int J Cardiol. 2013;168(2):1360-1363. doi:10.1016/j.ijcard.2012.12.011PubMedGoogle ScholarCrossref
23.
Saliba  W, Gronich  N, Barnett-Griness  O, Rennert  G.  Usefulness of CHADS2 and CHA2DS2-VASc scores in the prediction of new-onset atrial fibrillation: a population-based study.   Am J Med. 2016;129(8):843-849. doi:10.1016/j.amjmed.2016.02.029PubMedGoogle ScholarCrossref
24.
Reiffel  JA, Verma  A, Kowey  PR,  et al; REVEAL AF Investigators.  Incidence of previously undiagnosed atrial fibrillation using insertable cardiac monitors in a high-risk population: the REVEAL AF Study.   JAMA Cardiol. 2017;2(10):1120-1127. doi:10.1001/jamacardio.2017.3180PubMedGoogle ScholarCrossref
25.
Steinberg  BA, Zhao  Y, He  X,  et al.  Management of postoperative atrial fibrillation and subsequent outcomes in contemporary patients undergoing cardiac surgery: insights from the Society of Thoracic Surgeons CAPS-Care Atrial Fibrillation Registry.   Clin Cardiol. 2014;37(1):7-13. doi:10.1002/clc.22230PubMedGoogle ScholarCrossref
26.
Butt  JH, Xian  Y, Peterson  ED,  et al.  Long-term thromboembolic risk in patients with postoperative atrial fibrillation after coronary artery bypass graft surgery and patients with nonvalvular atrial fibrillation.   JAMA Cardiol. 2018;3(5):417-424. doi:10.1001/jamacardio.2018.0405PubMedGoogle ScholarCrossref
27.
Kaplan  RM, Koehler  J, Ziegler  PD, Sarkar  S, Zweibel  S, Passman  RS.  Stroke risk as a function of atrial fibrillation duration and CHA2DS2-VASc score.   Circulation. 2019;140(20):1639-1646. doi:10.1161/CIRCULATIONAHA.119.041303PubMedGoogle ScholarCrossref
28.
Kosmidou  I, Chen  S, Kappetein  AP,  et al.  New-onset atrial fibrillation after PCI or CABG for left main disease: the EXCEL Trial.   J Am Coll Cardiol. 2018;71(7):739-748. doi:10.1016/j.jacc.2017.12.012PubMedGoogle ScholarCrossref
29.
Piccini  JP.  Postoperative and nonvalvular atrial fibrillation-not so different after all.   JAMA Cardiol. 2019;4(11):1147-1148. doi:10.1001/jamacardio.2019.3826PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    Continuous Monitoring Post-Cardiac Surgery
    Stephen Friedman, MD, MPH | Rutgers NJ School of Medicine
    Have the authors considered continuous monitoring post-cardiac surgery in other groups such as post-mitral/tricuspid valve repair? Mild or intermittent atrial fibrillation prior to discharge may predict more severe arrhythmia after discharge. Continuous monitoring post-discharge may alert the physician and patient to impending atrial fibrillation and/or complete heart block in these cases.
    CONFLICT OF INTEREST: None Reported
    Views 5,729
    Citations 0
    Original Investigation
    Surgery
    August 27, 2021

    Effect of Continuous Electrocardiogram Monitoring on Detection of Undiagnosed Atrial Fibrillation After Hospitalization for Cardiac Surgery: A Randomized Clinical Trial

    Author Affiliations
    • 1Peter Munk Cardiac Centre, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada
    • 2Department of Cardiac Surgery, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
    • 3Department of Anesthesiology, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
    • 4University Institute of Cardiology and Respirology of Québec, Quebec City, Quebec, Canada
    • 5Division of Cardiac Surgery, New Brunswick, Saint John, New Brunswick, Canada
    • 6Division of Surgery, Cardiac Science Program, St Boniface General Hospital, Winnipeg, Manitoba, Canada
    • 7Department of Anesthesiology and Perioperative Medicine, Kingston General Hospital, Kingston, Ontario, Canada
    • 8Southlake Regional Health Center, University of Toronto, Newmarket, Ontario, Canada
    • 9University of Ottawa Heart Institute, Ottawa, Ontario, Canada
    • 10Division of Cardiology, St Paul’s Hospital, Vancouver, British Columbia, Canada
    • 11Applied Health Research Centre, Li Ka Shing Knowledge Institute of St Michael’s Hospital, Toronto, Ontario, Canada
    • 12Heart and Vascular Center, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts
    JAMA Netw Open. 2021;4(8):e2121867. doi:10.1001/jamanetworkopen.2021.21867
    Visual Abstract. Effect of Continuous Electrocardiogram Monitoring on Detection of Undiagnosed Atrial Fibrillation After Hospitalization for Cardiac Surgery
    Effect of Continuous Electrocardiogram Monitoring on Detection of Undiagnosed Atrial Fibrillation After Hospitalization for Cardiac Surgery
    Key Points

    Question  Can continuous cardiac rhythm monitoring beyond hospital discharge enhance atrial fibrillation (AF) detection among cardiac surgical patients?

    Findings  In this randomized clinical trial of 336 cardiac surgical patients with risk factors for stroke, use of continuous cardiac rhythm monitoring with wearable sensors increased the rate of AF detection by 17.9% within 30 days of hospital discharge compared with usual care.

    Meaning  Among cardiac surgical patients with risk factors for stroke and AF lasting less than 24 hours postoperatively, continuous cardiac rhythm monitoring significantly improved the rate of AF detection during the first 30 days after hospital discharge compared with usual care.

    Abstract

    Importance  Postoperative atrial fibrillation (POAF) occurring after cardiac surgery is associated with adverse outcomes. Whether POAF persists beyond discharge is not well defined.

    Objective  To determine whether continuous cardiac rhythm monitoring enhances detection of POAF among cardiac surgical patients during the first 30 days after hospital discharge compared with usual care.

    Design, Setting, and Participants  This study is an investigator-initiated, open-label, multicenter, randomized clinical trial conducted at 10 Canadian centers. Enrollment spanned from March 2017 to March 2020, with follow-up through September 11, 2020. As a result of the COVID-19 pandemic, enrollment stopped on July 17, 2020, at which point 85% of the proposed sample size was enrolled. Cardiac surgical patients with CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes, prior stroke or transient ischemic attack, vascular disease, age 65-74 years, female sex) score greater than or equal to 4 or greater than or equal to 2 with risk factors for POAF, no history of preoperative AF, and POAF lasting less than 24 hours during hospitalization were enrolled.

    Interventions  The intervention group underwent continuous cardiac rhythm monitoring with wearable, patch-based monitors for 30 days after randomization. Monitoring was not mandated in the usual care group within 30 days after randomization.

    Main Outcomes and Measures  The primary outcome was cumulative AF and/or atrial flutter lasting 6 minutes or longer detected by continuous cardiac rhythm monitoring or by a 12-lead electrocardiogram within 30 days of randomization. Prespecified secondary outcomes included cumulative AF lasting 6 hours or longer and 24 hours or longer within 30 days of randomization, death, myocardial infarction, ischemic stroke, non–central nervous system thromboembolism, major bleeding, and oral anticoagulation prescription.

    Results  Of the 336 patients randomized (163 patients in the intervention group and 173 patients in the usual care group; mean [SD] age, 67.4 [8.1] years; 73 women [21.7%]; median [interquartile range] CHA2DS2-VASc score, 4.0 [3.0-4.0] points), 307 (91.4%) completed the trial. In the intent-to-treat analysis, the primary end point occurred in 32 patients (19.6%) in the intervention group vs 3 patients (1.7%) in the usual care group (absolute difference, 17.9%; 95% CI, 11.5%-24.3%; P < .001). AF lasting 6 hours or longer was detected in 14 patients (8.6%) in the intervention group vs 0 patients in the usual care group (absolute difference, 8.6%; 95% CI, 4.3%-12.9%; P < .001).

    Conclusions and Relevance  In post–cardiac surgical patients at high risk of stroke, no preoperative AF history, and AF lasting less than 24 hours during hospitalization, continuous monitoring revealed a significant increase in the rate of POAF after discharge that would otherwise not be detected by usual care. Studies are needed to examine whether these patients will benefit from oral anticoagulation therapy.

    Trial Registration  ClinicalTrials.gov Identifier: NCT02793895

    Introduction

    Postoperative atrial fibrillation (POAF) after cardiac surgery occurs in 30% to 50% of patients during their hospital stay, with the incidence peaking at 3 to 5 days and decreasing afterwards.1 However, the incidence of POAF after discharge from cardiac surgery is not well defined. Most studies2-11 have been limited to the hospitalization phase only and were small, nonrandomized, or included patients with antecedent atrial fibrillation (AF) and used limited monitoring for AF. Among cardiac surgical patients who experienced little to no POAF during hospitalization, their risk of experiencing AF after discharge is unknown.12,13

    Quantifying the risk of ongoing POAF after hospitalization is an important but unanswered issue because many patients have elevated stroke risk (ie, CHA2DS2-VASc [congestive heart failure, hypertension, age ≥75 years, diabetes, prior stroke or transient ischemic attack, vascular disease, age 65-74 years, female sex] score ≥2); hence, diagnosing AF in this population may be actionable.14 In fact, POAF is associated with a 2-fold increase in early adverse outcomes such as stroke and death,1-3 and the risk of adverse events is known to persist even during long-term follow-up.1,2 Because there are sparse clinical trial data, guidelines provide little direction for clinicians on the optimal duration of monitoring, follow-up, or treatment for patients after hospitalization for cardiac surgery, particularly if they are in sinus rhythm at discharge.15,16 Randomized data guiding detection and management of POAF after hospitalization for cardiac surgery are lacking. Accordingly, we conducted a randomized clinical trial comparing continuous 30-day electrocardiography (ECG) monitoring with usual care for the detection of AF or atrial flutter (AFL) after hospitalization for cardiac surgery.

    Methods
    Trial Design and Setting

    The Post-Surgical Enhanced Monitoring for Cardiac Arrhythmias and Atrial Fibrillation (SEARCH-AF) trial was an investigator-initiated, prospective, open-label, multicenter, randomized clinical trial at 10 tertiary care cardiac surgical centers in Canada. Each site’s research ethics board approved the study, and all participants provided written informed consent. The trial protocol, protocol amendments, and statistical analysis plan are available in Supplement 1. The statistical analysis plan was written without knowledge of outcome data. This report was prepared in accordance with the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline for randomized trials.

    Study Population

    Patients were eligible if they were aged 18 years or older, underwent cardiac surgery (coronary artery bypass grafting [CABG] or valve repair or replacement with or without CABG), had no history of AF or AFL before surgery, had AF or AFL lasting less than 24 hours during hospitalization after surgery, and had a CHA2DS2-VASC score of greater than or equal to 4 or greater than or equal to 2 plus at least 1 additional factor associated with the risk developing POAF (chronic obstructive pulmonary disease, sleep apnea, impaired renal function, left atrial enlargement, elevated body mass index, or combined CABG with valve repair or replacement). Patients were excluded if they had a presurgical history of AF or AFL, had AF or AFL lasting 24 hours or longer after cardiac surgery, were in AF or AFL at the time of randomization, were hospitalized for 10 days or longer at the time of randomization, or received a mechanical valve or had other reason to be on oral anticoagulation. Race/ethnicity was assessed to examine its potential association with the incidence of POAF. Data on race/ethnicity were collected from review of medical records. Detailed inclusion and exclusion criteria are listed in eTable 1 in Supplement 2.

    Trial Procedures

    Patients were randomly assigned in a 1:1 ratio to the monitoring group (continuous 30-day ECG monitoring) or to usual care with no mandated monitoring. Randomization lists were prepared by the trial statistician (K.E.T.) and were computer-generated using random permuted blocks. Randomization was stratified by study site and by the type of cardiac surgery performed (isolated CABG or valve replacement or repair with or without CABG). Randomization occurred between the third postoperative day and discharge from hospital. In each site, study coordinators (with oversight from the site principal investigator) enrolled patients in the trial and administered the trial interventions.

    Patients randomized to the monitoring group received up to 30 days of continuous ECG monitoring with a wearable, adhesive patch monitor. Until September 30, 2018, the SEEQ system (Medtronic) was used. Afterward, the CardioSTATsystem (Icentia) was used because SEEQ manufacture was halted. Maximum monitoring duration for a single SEEQ or CardioSTAT patch was 7.5 and 14 days, respectively. Patients in the monitoring group received 4 SEEQ or 2 CardioSTAT patches, which provided up to 30 and 28 days of ECG monitoring, respectively. Before hospital discharge, the patch was applied on the anterior left chest, away from the sternotomy site. Patients were provided instruction on monitor use and were contacted weekly (SEEQ) or biweekly (CardioSTAT) to ensure compliance. Used patches were returned to the manufacturer for analysis. A report including all arrhythmic episodes with ECG strips was provided to site investigators and subsequently forwarded to the patient’s primary care physician, cardiologist, and cardiac surgeon at the end of the 28-day or 30-day monitoring period.

    Patients randomized to usual care did not undergo any protocol-mandated ECG monitoring within the first 30 days after randomization. This practice is consistent with current guidelines, which make no recommendations for routine monitoring.15-17 However, if clinically indicated, patients in the usual care group could undergo standard ECG or Holter monitoring within the first 30 days of randomization. If AF or AFL was documented on a 12-lead ECG, performance of additional monitoring, such as Holter monitoring, was suggested. Any AF or AFL detected by an ECG, Holter, or any other type of monitoring performed in the usual care group was included for end point assessment to avoid negatively biasing the event rate in this group. In both groups, therapeutic decisions, such as initiation of oral anticoagulation therapy, were left to local physicians’ discretion.

    Patients were followed at 1 to 3 months and 6 months after discharge from surgery on an in-person basis. A telephone-based follow-up was conducted at 9 months after discharge from surgery. At 6 months, patients in both study groups underwent 14 days of continuous cardiac rhythm monitoring (SEEQ or CardioSTAT patch).

    Outcomes

    The primary outcome was documentation of cumulative AF or AFL duration of 6 minutes or longer or documentation of AF or AFL by a single 12-lead ECG within 30 days after randomization. A list of prespecified secondary outcomes is provided in eTable 2 in Supplement 2. Rhythm-based outcomes, major adverse cardiovascular events, and major bleeding were adjudicated by a committee of physicians blinded to treatment assignment.

    Statistical Analysis

    The trial was designed to test the hypothesis of superiority of continuous ECG monitoring over usual care in the detection of the primary outcome within 30 days after randomization. The rate of the primary outcome was estimated to be 3% during 30 days in the usual care group.18 To detect a 7% absolute increase in primary outcome detection with a strategy of continuous cardiac rhythm monitoring, a total of 388 patients (194 in each group) would provide the trial with 80% power at a 2-sided α = .05. Assuming a 2% attrition rate, the final sample size was 396 patients.

    Outcome analyses were performed on the intention-to-treat population, which analyzed patients according to their randomization status. The difference in the primary outcome between the 2 study groups was reported as the absolute rate difference with associated 95% CIs. Comparison of the proportion of patients between the 2 groups with the primary outcome was performed with the Pearson χ2 test. The Pearson χ2 test was used to compare the proportions between groups for secondary outcomes. A subgroup analysis was performed involving 5 prespecified subgroups (sex, age, CHA2DS2-VASc score, surgery type [isolated CABG vs valve repair or replacement with or without CABG], and left atrial size) to assess for effect modification between the 2 randomization groups. A Kaplan-Meier curve of the primary outcome was generated for illustrative purposes.

    A per-protocol analysis was prespecified and consisted of all randomized patients who received their allocated intervention without significant deviations in the assigned treatment during the first 30 days after randomization. For patients randomized to the continuous cardiac rhythm monitoring group, the per-protocol cohort was defined by patients who wore the sensor for 24 hours or longer within the first 30 days after randomization. For patients randomized to the usual care group, the per-protocol cohort was defined by patients who did not wear a continuous cardiac rhythm monitor (SEEQ or CardioSTAT monitor) during the first 30 days after randomization. Statistical analyses were performed with R statistical software version 4.0.2 (R Project for Statistical Computing).

    Results
    Study Population

    Between March 2017 and March 2020, 336 patients (mean [SD] age, 67.4 [8.1] years; 73 women [21.7%]) were randomized to continuous monitoring (163 patients) or usual care (173 patients) at 8 Canadian sites (Figure 1). No patients were randomized at 2 sites. As a result of the COVID-19 pandemic, all sites mandated institutional restrictions on surgical procedures and research recruitment. On July 17, 2020, enrollment in the study was terminated, at which time 336 patients were enrolled, accounting for 85% of the planned sample size. The last protocol-mandated clinical follow-up occurred on August 31, 2020, and the last follow-up related to results of cardiac rhythm monitoring occurred on September 11, 2020. The median (IQR) time from surgery to randomization was 5 (4-6) days, and the median (IQR) time from randomization to discharge was 0 (0-1) days.

    There was no statistically significant difference in the baseline characteristics of the 2 assigned groups (Table 1). The median (IQR) CHA2DS2-VASc score was 4.0 points (3.0-4.0 points); 255 patients (75.9%) underwent CABG alone, 39 patients (11.6%) underwent valve repair or replacement alone, and 42 patients (12.5%) underwent both CABG and valve repair or replacement. At hospital discharge, amiodarone was prescribed for 13 patients (7 in the monitoring group and 6 in the usual care group). None of the patients had a presurgical history of AF or AFL by design, and 318 patients (94.6%) had no documented AF during hospitalization. The duration of in-hospital AF in the remaining 18 patients (5.4%) was less than 24 hours.

    All study-related follow-up visits were attended by 307 patients (91.4%). The median (IQR) follow-up was 276 (267-293) days in the continuous cardiac rhythm monitoring group and 277 (267-292) days in the usual care group. Among the 163 patients in the monitoring group, 8 patients (4.9%) did not undergo continuous cardiac rhythm monitoring and 2 patients (1.2%) completed continuous cardiac rhythm monitoring but the data were not received by the monitoring center (Figure 1). Of the remaining 153 patients in the intervention group, 79 patients (51.6%) wore the monitor for at least 28 days and 30 patients (19.6%) wore the monitor for 21 to 27 days. Reasons for screen failure are shown in eTable 3 in Supplement 2, and reasons for premature termination of monitoring are listed in eTable 4 in Supplement 2. In the continuous cardiac rhythm monitoring group, 17 patients (10.4%) experienced a device-related adverse event due to skin irritation from the adhesive material of the wearable patch within 30 days of randomization. None of these device-related adverse events resulted in a serious adverse event. Four patients (2.3%) in the usual care group underwent Holter monitoring within 30 days of randomization.

    Primary Outcome

    In the continuous monitoring group, the primary outcome was detected in 32 of 163 patients (19.6%) compared with 3 of 173 patients (1.7%) in the usual care group, with an absolute difference of 17.9% (95% CI, 11.5%-24.3%; P < .001) (Table 2 and Figure 2).The number needed to screen to detect the primary outcome was 6 patients (95% CI, 4-9 patients). Of the 32 patients in the monitoring group in whom the primary outcome occurred, 30 of them had at least 1 episode of detected AF or AFL lasting for 6 minutes or longer.

    In the monitoring group, the primary outcome was detected by continuous monitoring in 30 of 32 patients. For the other 2 patients, AF was detected by 12-lead ECG. The first episode of AF or AFL lasting for 6 minutes or longer was detected in 22 patients (73.3%) during the first week of monitoring, in 6 patients (20.0%) during the second week, and in 2 patients (6.7%) during the third week. The cumulative duration of detected AF or AFL episodes and the number of patients with detected AF or AFL lasting for 6 minutes or longer, stratified by monitoring weeks, is shown in Figure 3. Cumulative AF or AFL lasting 6 minutes or longer was detected in 10 patients in multiple monitoring weeks. Additional details of detected AF durations are described in eTable 5 in Supplement 2. Among the 3 patients with detected AF in the usual care group, AF was documented by a 12-lead ECG within 2 days of randomization.

    Secondary Outcomes

    Within the first 30 days, cumulative AF or AFL lasting 6 hours or longer was detected in 14 patients (8.6%) in the monitoring group and 0 patients in the usual care group (absolute difference, 8.6%; 95% CI, 4.3%-12.9%; P < .001) (Table 2). Cumulative AF or AFL lasting 24 hours or longer within the first 30 days was detected in 5 patients (3.1%) in the monitoring group and 0 patients in the usual care group (absolute difference, 3.1%; 95% CI, 0.4%-5.7%) (Table 2). Other descriptive outcomes of episode durations are provided in eTable 6 in Supplement 2.

    During the first 45 days after discharge, a total of 11 patients (3.3%) were prescribed oral anticoagulation therapy (7 patients [4.3%] in the monitoring group and 4 patients [2.3%] in the usual care group) (Table 2). Between 46 days after discharge and the end of follow-up, oral anticoagulation therapy was prescribed for 11 patients (3.3%) (6 patients [3.7%] in the monitoring group and 5 patients [2.9%] in the usual care group) (Table 2).

    After discharge, major adverse cardiovascular events occurred in 6 patients (1.8%), 2 patients (1.2%) in the monitoring group and 4 patients (2.3%) receiving usual care (Table 2). Two patients experienced an ischemic stroke during study follow-up, and AF was detected in 1 of those patients.

    At 6 months, 189 patients (56.3%) underwent protocol-mandated continuous cardiac rhythm monitoring with a 2-week, patch-based monitor. Among these 189 patients, cumulative AF or AFL lasting for 6 minutes or longer was detected in 3 patients (1.6%) at 6 months after surgery. Of the entire cohort of 336 patients, cumulative AF or AFL lasting for 6 minutes or longer by continuous cardiac rhythm monitoring or AF or AFL documented by a 12-lead ECG was detected among 7 patients (2.1%) between 31 days after randomization and end of study follow-up (eTable 7 in Supplement 2). The complete list of secondary outcomes is available in eTable 8 in Supplement 2.

    Subgroup Analysis and Per-Protocol Analysis

    The treatment effect of the study intervention was similar between the prespecified subgroups. Evaluation of effect modification could not be assessed between men and women because there were 0 events among women in the control group (eFigure in Supplement 2). Results of the per-protocol analysis were similar to the intent-to-treat analysis (eTable 9 in Supplement 2). An exploratory analysis examining the association between clinical factors and the primary end point is provided in eTable 10 in Supplement 2. Results of 2 additional exploratory analyses (detection of AF with 30 days of randomization according to occurrence of POAF before randomization and according to type of wearable sensor used) are provided in eTable 11 and eTable 12 in Supplement 2.

    Discussion

    In this randomized clinical trial, we found that continuous monitoring during the first 30 days after hospitalization for cardiac surgery detected significantly more POAF than usual care among patients at high risk of stroke with no presurgical history of AF who had AF for less than 24 hours while hospitalized. Similar findings were observed for detection of longer AF durations of 6 hours or longer and 24 hours or longer. Most patients in this study had no POAF detected during hospitalization after cardiac surgery. These findings demonstrate that POAF after cardiac surgery is not confined to the hospitalization period. A high rate of AF can be detected in the first month after surgery.

    POAF has long been considered an early, transient phenomenon after cardiac surgery because of pericardial inflammation, cardiac ischemia, hemodynamic fluctuations, and high adrenergic state, which is assumed to resolve in days.19 Although POAF after cardiac surgery is a common clinical problem, most studies20,21 have only described its incidence during hospitalization. The incidence of POAF occurring weeks to months after surgery is not well-defined; prior studies3-11 are few, retrospective, or nonrandomized. This represents an important unknown for clinicians managing these patients. In our study, continuous cardiac rhythm monitoring detected AF in nearly 1 of 5 patients within 30 days after discharge and was markedly higher than the rate of detected AF in the usual care group. This indicates that a substantial proportion of POAF is subclinical and would not be diagnosed without the use of continuous cardiac rhythm monitoring. In this study, patients had AF lasting less than 24 hours during hospitalization and would likely not have prompted any further monitoring or treatment under normal clinical circumstances. In this patient population, continuous ECG monitoring was essential to uncover the ongoing risk and extent of outpatient POAF. In particular, the occurrence of POAF was clustered within the first 2 weeks after discharge in our study cohort.

    In our study, the rate of AF detection with continuous cardiac rhythm monitoring was substantially higher than that in patients in the nonsurgical setting and with a similar risk profile. The CHA2DS2-VASc score is associated with development of AF in the nonsurgical setting.22,23 Among patients with a CHA2DS2-VASc score similar to that for our cohort (median, 4.0 points), AF incidences of 0.9 to 1.5 cases per 100 patient-years were reported, using limited modalities to ascertain AF.22,23 In a study24 using continuous monitors in a nonsurgical population with a CHA2DS2-VASc score similar to that of our population, the detection rate of new-onset AF was 6.2% after 30 days. Compared with these studies, our 30-day AF detection rate of 19.6% was much higher than that in a general, matched population.

    The rate of oral anticoagulation therapy in our patients was low. This was consistent with published studies25,26 showing that rates of oral anticoagulation therapy after cardiac surgery were low despite a high incidence of in-hospital POAF. Reluctance to prescribe anticoagulation therapy may be related to concerns over bleeding, the belief that POAF is self-limiting, or uncertainty regarding the duration of AF that would justify the use of oral anticoagulation for stroke prevention in this patient population. In our population with a median CHA2DS2-VASc score of 4 and AF lasting 6 minutes or longer, the annual stroke risk is estimated to be 1.3%,27 which would meet the threshold for oral anticoagulation therapy. Reluctance to prescribe oral anticoagulation therapy may reflect uncertainty in clinical practice guidelines as current recommendations for oral anticoagulation therapy after discharge are limited by lack of randomized trial data, especially if patients are in sinus rhythm.15,16 Our data may help inform these guidelines in terms of the optimal method for POAF detection after discharge.

    Our findings suggest that the incidence of POAF may decrease rapidly over time because the rate of AF detection markedly declined after the first 2 weeks of hospital discharge. However, previous studies1,13,14,20,28 found an association between POAF and cardiovascular risk over long-term follow-up, when, presumably, POAF had largely resolved. Future research is needed to assess whether early anticoagulation may mitigate adverse short-term and long-term cardiovascular outcomes.29

    Limitations

    This study has limitations that should be addressed. The primary end point is limited to 30 days and does not estimate the long-term, ongoing risk of POAF. Among patients with POAF lasting less than 24 hours, we did not collect the exact duration of AF and, hence, could not assess the association between in-hospital AF burden and the primary end point. Even with limited continuous monitoring, some patients did not complete it and reported a lack of interest as the primary reason. Our trial, by design, excluded patients with longer episodes of AF (≥24 hours) while hospitalized and those with prolonged hospital stay. The rates of AF occurrence and cardiovascular events might be higher if these patients were included in this study. We did not perform preoperative cardiac monitoring to rule out the presence of AF before surgery; subclinical AF could be prevalent among certain patient subsets, such as those with mitral valve disease. This was beyond the scope of this study. Lack of adherence with the monitoring by some patients led to underestimation of AF. The planned sample size and follow-up duration were not powered to detect for differences in major adverse cardiovascular outcomes, stroke rates in comparison with the presence of POAF, or whether oral anticoagulation therapy would alter the risk of such ischemic outcomes.

    Conclusions

    Among cardiac surgical patients with an elevated risk of stroke and no history of preoperative or predischarge AF, a strategy of continuous cardiac rhythm monitoring unveiled a significant burden of unrecognized AF compared with usual care, specifically within the first 2 weeks after discharge. Future studies are required to assess whether oral anticoagulation therapy is safe and effective in this patient population.

    Back to top
    Article Information

    Accepted for Publication: June 11, 2021.

    Published: August 27, 2021. doi:10.1001/jamanetworkopen.2021.21867

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Ha ACT et al. JAMA Network Open.

    Corresponding Authors: Atul Verma, MD, Southlake Regional Health Center, University of Toronto, 602-581 Davis Dr, Newmarket, ON L3Y 2P6, Canada (atul.verma@utoronto.ca); Subodh Verma, MD, PhD, Department of Cardiac Surgery, St. Michael’s Hospital, University of Toronto, 30 Bond St, Toronto, ON M5B 1W8 Canada (subodh.verma@unityhealth.to).

    Author Contributions: Drs Ha and A. Verma had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Ha, S. Verma, Mazer, and A. Verma contributed equally to this work.

    Concept and design: Ha, S. Verma, Mazer, Quan, Yanagawa, Latter, Singal, Wilson, Cheung, A. Verma.

    Acquisition, analysis, or interpretation of data: Ha, S. Verma, Mazer, Yanagawa, Yau, Jacques, Brown, Singal, Yamashita, Saha, K. H. Teoh, Lam, Deyell, Hibino, Cheung, Kosmopoulos, Garg, Brodutch, H. Teoh, Zuo, Thorpe, Jüni, Bhatt, A. Verma.

    Drafting of the manuscript: Ha, S. Verma, Yanagawa, Singal, Wilson, A. Verma.

    Critical revision of the manuscript for important intellectual content: Ha, S. Verma, Mazer, Quan, Yanagawa, Latter, Yau, Jacques, Brown, Yamashita, Saha, K. H. Teoh, Lam, Deyell, Hibino, Cheung, Kosmopoulos, Garg, Brodutch, H. Teoh, Zuo, Thorpe, Jüni, Bhatt, A. Verma.

    Statistical analysis: Hibino, Zuo, Thorpe, Jüni, A. Verma.

    Obtained funding: Ha, S. Verma, A. Verma.

    Administrative, technical, or material support: Ha, Mazer, Quan, Yanagawa, Jacques, Brown, Singal, Yamashita, Saha, Wilson, Kosmopoulos, Garg, Brodutch, A. Verma.

    Supervision: Ha, S. Verma, Mazer, Yau, Jacques, Saha, Deyell, Wilson, Thorpe, A. Verma.

    Conflict of Interest Disclosures: Dr Ha reported receiving lecture fees from Bayer, Bristol Myers Squibb/Pfizer, and Servier. Dr S. Verma reported holding a Tier 1 Canada Research Chair in Cardiovascular Surgery; reported receiving research grants and speaking honoraria from Amarin, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, EOCI Pharmacomm, Ltd, HLS Therapeutics, Janssen, Merck, Novartis, Novo Nordisk, PhaseBio, Sanofi, Sun Pharmaceuticals, and the Toronto Knowledge Translation Working Group; and reported being the President of the Canadian Medical and Surgical Knowledge Translation Research Group, a federally incorporated not-for-profit physician organization. Dr Mazer reported receiving advisory board fees from Amgen, AstraZeneca, Boehringer Ingelheim, and Octapharma. Dr Jacques reported receiving nonfinancial support from Medtronic and Icentia, grants and nonfinancial support from Allergan and Abbvie, and personal fees from LivaNova. Dr Deyell reported receiving personal fees from Medtronic and Servier. Ms Wilson reported receiving personal fees from HLS Therapeutics and Pfizer. Dr H. Teoh reported receiving honoraria from Boehringer Ingelheim and HLS Therapeutics. Dr Jüni reported serving as an unpaid steering committee member for Astra Zeneca, Biotronik, Biosensors, Abbott, and the Medicines Company, and reported receiving consulting fees from Amgen, Ava, and Fresenius. Dr Bhatt reported receiving advisory board fees from Cardax, CellProthera, Cereno Scientific, Elsevier Practice Update Cardiology, Janssen, Level Ex, Medscape Cardiology, MyoKardia, NirvaMed, Novo Nordisk, PhaseBio, PLx Pharma, and Regado Biosciences; being a member of the Board of Directors for the Boston VA Research Institute, Society of Cardiovascular Patient Care, and TobeSoft; being inaugural chair for the American Heart Association Quality Oversight Committee; serving on data monitoring committees for Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute, for the PORTICO trial, funded by St Jude Medical, now Abbott), Cleveland Clinic (including for the ExCEED trial, funded by Edwards), Contego Medical (chair, PERFORMANCE 2), Duke Clinical Research Institute, Mayo Clinic, Mount Sinai School of Medicine (for the ENVISAGE trial, funded by Daiichi Sankyo), and Population Health Research Institute; receiving honoraria from the American College of Cardiology (ACC; senior associate editor of Clinical Trials and News and chair of the ACC accreditation oversight committee), Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute; RE-DUAL PCI clinical trial steering committee funded by Boehringer Ingelheim; AEGIS-II executive committee funded by CSL Behring), Belvoir Publications (editor in chief, Harvard Heart Letter), Canadian Medical and Surgical Knowledge Translation Research Group (clinical trial steering committees), Duke Clinical Research Institute (clinical trial steering committees, including for the PRONOUNCE trial, funded by Ferring Pharmaceuticals), HMP Global (editor in chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (guest editor and associate editor), K2P (cochair, interdisciplinary curriculum), Level Ex (continuing medical education steering committees), Medtelligence/ReachMD, MJH Life Sciences, Population Health Research Institute (for the COMPASS operations committee, publications committee, steering committee, and US national coleader, funded by Bayer), Slack Publications (chief medical editor, Cardiology Today’s Intervention), Society of Cardiovascular Patient Care (secretary and treasurer), and WebMD (continuing medical education steering committees); being deputy editor of Clinical Cardiology, chair of the NCDR-ACTION Registry Steering Committee, and chair of the VA CART Research and Publications Committee; receiving grant support from Abbott, Afimmune, Amarin, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Cardax, CellProthera, Cereno Scientific, Chiesi, CSL Behring, Eisai, Ethicon, Ferring Pharmaceuticals, Forest Laboratories, Fractyl, Garmin, HLS Therapeutics, Idorsia, Ironwood, Ischemix, Janssen, Lexicon, Lilly, Medtronic, MyoKardia, NirvaMed, Novartis, Novo Nordisk, Owkin, Pfizer, PhaseBio, PLx Pharma, Regeneron, Roche, Sanofi Aventis, Synaptic, The Medicines Company, and 89Bio; receiving royalties from Elsevier (editor, Cardiovascular Intervention: A Companion to Braunwald’s Heart Disease); being a co–site investigator for Abbott, Biotronik, Boston Scientific, CSI, St. Jude Medical (now Abbott), Philips, and Svelte; being a trustee for the ACC; and conducting unfunded research with FlowCo, Merck, and Takeda. Dr A. Verma reported receiving grant support from Biotronik, Bristol Myers Squibb, and Boehringer Ingelheim; receiving grant support, advisory board fees, and lecture fees from Bayer and Biosense Webster; receiving advisory board fees and lecture fees and serving as principal investigator (PULSED AF and DIAMOND II trial) from Medtronic; receiving consulting fees for serving on steering committees for Boston Scientific, Kardium, Medlumics, and Thermedical; and receiving lecture fees from Servier. No other disclosures were reported.

    Funding/Support: This study was funded by the Heart and Stroke Foundation of Canada (grant G-17-0018661), Canadian Stroke Prevention Intervention Network, and unrestricted investigator-initiated grants from Bristol-Myers Squibb-Pfizer Alliance and Boehringer Ingelheim. Medtronic provided in-kind support with SEEQ devices.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Group Members: Members of the SEARCH AF CardioLink-1 Investigators are listed in Supplement 3.

    Meeting Presentation: This study was presented as a Late-Breaking Clinical Trial Abstract at the 2020 American Heart Association Scientific Sessions; November 16, 2020; Dallas, Texas.

    Data Sharing Statement: See Supplement 4.

    Additional Contributions: Samantha Saes, BSc (Applied Health Research Centre [AHRC], Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Toronto, Canada) assisted with study coordination. Ms Saes is an employee of AHRC, and a portion of her salary related to this study is supported by the study funds.

    References
    1.
    Lin  MH, Kamel  H, Singer  DE, Wu  YL, Lee  M, Ovbiagele  B.  Perioperative/postoperative atrial fibrillation and risk of subsequent stroke and/or mortality.   Stroke. 2019;50(6):1364-1371. doi:10.1161/STROKEAHA.118.023921PubMedGoogle ScholarCrossref
    2.
    Kaw  R, Hernandez  AV, Masood  I, Gillinov  AM, Saliba  W, Blackstone  EH.  Short- and long-term mortality associated with new-onset atrial fibrillation after coronary artery bypass grafting: a systematic review and meta-analysis.   J Thorac Cardiovasc Surg. 2011;141(5):1305-1312. doi:10.1016/j.jtcvs.2010.10.040PubMedGoogle ScholarCrossref
    3.
    Ambrosetti  M, Tramarin  R, Griffo  R,  et al; ISYDE and ICAROS Investigators of the Italian Society for Cardiovascular Prevention, Rehabilitation and Epidemiology (IACPR-GICR).  Late postoperative atrial fibrillation after cardiac surgery: a national survey within the cardiac rehabilitation setting.   J Cardiovasc Med (Hagerstown). 2011;12(6):390-395. doi:10.2459/JCM.0b013e328346a6d3PubMedGoogle ScholarCrossref
    4.
    Cioffi  G, Mureddu  G, Cemin  C,  et al.  Characterization of post-discharge atrial fibrillation following open-heart surgery in uncomplicated patients referred to an early rehabilitation program.   Ital Heart J. 2001;2(7):519-528.PubMedGoogle Scholar
    5.
    Giaccardi  M, Macchi  C, Colella  A,  et al.  Postacute rehabilitation after coronary surgery: the effect of preoperative physical activity on the incidence of paroxysmal atrial fibrillation.   Am J Phys Med Rehabil. 2011;90(4):308-315. doi:10.1097/PHM.0b013e31820f9535PubMedGoogle ScholarCrossref
    6.
    Funk  M, Richards  SB, Desjardins  J, Bebon  C, Wilcox  H.  Incidence, timing, symptoms, and risk factors for atrial fibrillation after cardiac surgery.   Am J Crit Care. 2003;12(5):424-433. doi:10.4037/ajcc2003.12.5.424PubMedGoogle ScholarCrossref
    7.
    Lowres  N, Mulcahy  G, Gallagher  R,  et al.  Self-monitoring for atrial fibrillation recurrence in the discharge period post-cardiac surgery using an iPhone electrocardiogram.   Eur J Cardiothorac Surg. 2016;50(1):44-51. doi:10.1093/ejcts/ezv486PubMedGoogle ScholarCrossref
    8.
    Axtell  AL, Moonsamy  P, Melnitchouk  S,  et al.  Preoperative predictors of new-onset prolonged atrial fibrillation after surgical aortic valve replacement.   J Thorac Cardiovasc Surg. 2020;159(4):1407-1414. doi:10.1016/j.jtcvs.2019.04.077PubMedGoogle ScholarCrossref
    9.
    Ayoub  K, Habash  F, Almomani  A,  et al.  Long term risk of recurrent atrial fibrillation and ischemic stroke after post-operative atrial fibrillation complicating cardiac and non-cardiac surgeries.   J Atr Fibrillation. 2018;10(6):1660. doi:10.4022/jafib.1660PubMedGoogle Scholar
    10.
    Thorén  E, Wernroth  ML, Christersson  C, Grinnemo  KH, Jidéus  L, Ståhle  E.  Compared with matched controls, patients with postoperative atrial fibrillation (POAF) have increased long-term AF after CABG, and POAF is further associated with increased ischemic stroke, heart failure and mortality even after adjustment for AF.   Clin Res Cardiol. 2020;109(10):1232-1242. doi:10.1007/s00392-020-01614-zPubMedGoogle ScholarCrossref
    11.
    Wang  EY, Hulme  OL, Khurshid  S,  et al.  Initial precipitants and recurrence of atrial fibrillation.   Circ Arrhythm Electrophysiol. 2020;13(3):e007716. doi:10.1161/CIRCEP.119.007716PubMedGoogle Scholar
    12.
    El-Chami  MF, Merchant  FM, Smith  P,  et al.  Management of new-onset postoperative atrial fibrillation utilizing insertable cardiac monitor technology to observe recurrence of AF (MONITOR-AF).   Pacing Clin Electrophysiol. 2016;39(10):1083-1089. doi:10.1111/pace.12949PubMedGoogle ScholarCrossref
    13.
    Lomivorotov  VV, Efremov  SM, Pokushalov  EA,  et al.  Randomized trial of fish oil infusion to prevent atrial fibrillation after cardiac surgery: data from an implantable continuous cardiac monitor.   J Cardiothorac Vasc Anesth. 2014;28(5):1278-1284. doi:10.1053/j.jvca.2014.02.019PubMedGoogle ScholarCrossref
    14.
    Chen  YL, Zeng  M, Liu  Y,  et al.  CHA2DS2-VASc score for identifying patients at high risk of postoperative atrial fibrillation after cardiac surgery: a meta-analysis.   Ann Thorac Surg. 2020;109(4):1210-1216. doi:10.1016/j.athoracsur.2019.07.084PubMedGoogle ScholarCrossref
    15.
    Kirchhof  P, Benussi  S, Kotecha  D,  et al; ESC Scientific Document Group.  2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS.   Eur Heart J. 2016;37(38):2893-2962. doi:10.1093/eurheartj/ehw210PubMedGoogle ScholarCrossref
    16.
    January  CT, Wann  LS, Alpert  JS,  et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines.  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. 2014;64(21):e1-e76. doi:10.1016/j.jacc.2014.03.022PubMedGoogle ScholarCrossref
    17.
    Verma  A, Cairns  JA, Mitchell  LB,  et al; CCS Atrial Fibrillation Guidelines Committee.  2014 Focused update of the Canadian Cardiovascular Society Guidelines for the management of atrial fibrillation.   Can J Cardiol. 2014;30(10):1114-1130. doi:10.1016/j.cjca.2014.08.001PubMedGoogle ScholarCrossref
    18.
    Yagdi  T, Nalbantgil  S, Ayik  F,  et al.  Amiodarone reduces the incidence of atrial fibrillation after coronary artery bypass grafting.   J Thorac Cardiovasc Surg. 2003;125(6):1420-1425. doi:10.1016/S0022-5223(02)73292-3PubMedGoogle ScholarCrossref
    19.
    Maesen  B, Nijs  J, Maessen  J, Allessie  M, Schotten  U.  Post-operative atrial fibrillation: a maze of mechanisms.   Europace. 2012;14(2):159-174. doi:10.1093/europace/eur208PubMedGoogle ScholarCrossref
    20.
    Kerwin  M, Saado  J, Pan  J,  et al.  New-onset atrial fibrillation and outcomes following isolated coronary artery bypass surgery: a systematic review and meta-analysis.   Clin Cardiol. 2020;43(9):928-934. doi:10.1002/clc.23414PubMedGoogle ScholarCrossref
    21.
    Gialdini  G, Nearing  K, Bhave  PD,  et al.  Perioperative atrial fibrillation and the long-term risk of ischemic stroke.   JAMA. 2014;312(6):616-622. doi:10.1001/jama.2014.9143PubMedGoogle ScholarCrossref
    22.
    Chao  TF, Liu  CJ, Chen  SJ,  et al.  CHADS2 score and risk of new-onset atrial fibrillation: a nationwide cohort study in Taiwan.   Int J Cardiol. 2013;168(2):1360-1363. doi:10.1016/j.ijcard.2012.12.011PubMedGoogle ScholarCrossref
    23.
    Saliba  W, Gronich  N, Barnett-Griness  O, Rennert  G.  Usefulness of CHADS2 and CHA2DS2-VASc scores in the prediction of new-onset atrial fibrillation: a population-based study.   Am J Med. 2016;129(8):843-849. doi:10.1016/j.amjmed.2016.02.029PubMedGoogle ScholarCrossref
    24.
    Reiffel  JA, Verma  A, Kowey  PR,  et al; REVEAL AF Investigators.  Incidence of previously undiagnosed atrial fibrillation using insertable cardiac monitors in a high-risk population: the REVEAL AF Study.   JAMA Cardiol. 2017;2(10):1120-1127. doi:10.1001/jamacardio.2017.3180PubMedGoogle ScholarCrossref
    25.
    Steinberg  BA, Zhao  Y, He  X,  et al.  Management of postoperative atrial fibrillation and subsequent outcomes in contemporary patients undergoing cardiac surgery: insights from the Society of Thoracic Surgeons CAPS-Care Atrial Fibrillation Registry.   Clin Cardiol. 2014;37(1):7-13. doi:10.1002/clc.22230PubMedGoogle ScholarCrossref
    26.
    Butt  JH, Xian  Y, Peterson  ED,  et al.  Long-term thromboembolic risk in patients with postoperative atrial fibrillation after coronary artery bypass graft surgery and patients with nonvalvular atrial fibrillation.   JAMA Cardiol. 2018;3(5):417-424. doi:10.1001/jamacardio.2018.0405PubMedGoogle ScholarCrossref
    27.
    Kaplan  RM, Koehler  J, Ziegler  PD, Sarkar  S, Zweibel  S, Passman  RS.  Stroke risk as a function of atrial fibrillation duration and CHA2DS2-VASc score.   Circulation. 2019;140(20):1639-1646. doi:10.1161/CIRCULATIONAHA.119.041303PubMedGoogle ScholarCrossref
    28.
    Kosmidou  I, Chen  S, Kappetein  AP,  et al.  New-onset atrial fibrillation after PCI or CABG for left main disease: the EXCEL Trial.   J Am Coll Cardiol. 2018;71(7):739-748. doi:10.1016/j.jacc.2017.12.012PubMedGoogle ScholarCrossref
    29.
    Piccini  JP.  Postoperative and nonvalvular atrial fibrillation-not so different after all.   JAMA Cardiol. 2019;4(11):1147-1148. doi:10.1001/jamacardio.2019.3826PubMedGoogle ScholarCrossref
    ×