Incidence and Predictors of Left Atrial Appendage Thrombus in Patients Treated With Nonvitamin K Oral Anticoagulants Versus Warfarin Before Catheter Ablation for Atrial Fibrillation

Accepted Manuscript Incidence and Predictors of Left Atrial Appendage Thrombus in Patients Treated with Non-Vitamin K Oral Anticoagulants Versus Warfarin Prior to Catheter Ablation for Atrial Fibrillation Joanne Wyrembak, MD, Kristen B. Campbell, PharmD, Benjamin A. Steinberg, MD, MHS, Tristram D. Bahnson, MD, James P. Daubert, MD, Eric J. Velazquez, MD, Zainab Samad, MD, Brett D. Atwater, MD PII:

S0002-9149(16)32004-5

DOI:

10.1016/j.amjcard.2016.12.008

Reference:

AJC 22324

To appear in:

The American Journal of Cardiology

Received Date: 19 October 2016 Revised Date:

6 December 2016

Accepted Date: 8 December 2016

Please cite this article as: Wyrembak J, Campbell KB, Steinberg BA, Bahnson TD, Daubert JP, Velazquez EJ, Samad Z, Atwater BD, Incidence and Predictors of Left Atrial Appendage Thrombus in Patients Treated with Non-Vitamin K Oral Anticoagulants Versus Warfarin Prior to Catheter Ablation for Atrial Fibrillation, The American Journal of Cardiology (2017), doi: 10.1016/j.amjcard.2016.12.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Incidence and Predictors of Left Atrial Appendage Thrombus in Patients Treated with Non-Vitamin K Oral Anticoagulants Versus

RI PT

Warfarin Prior to Catheter Ablation for Atrial Fibrillation

Joanne Wyrembak, MDa; Kristen B. Campbell, PharmDa,b; Benjamin A. Steinberg, MDa,b,

M AN U

Zainab Samad, MDa; Brett D. Atwater, MDa,b

SC

MHS; Tristram D. Bahnson, MDa,b; James P. Daubert, MDa,b; Eric J. Velazquez, MDa;

a

Division of Cardiology, Duke University Medical Center, Durham, NC, USA

bDuke

Center for Atrial Fibrillation, Duke University Health System, Durham NC, USA

Word Count:

TE D

Relationship with Industry: The co-authors have no industry relationships that constitute a conflict of interest for this work. The work was un-funded.

EP

Running Title: Pre-AF ablation LAA thrombus Address for Correspondence:

AC C

Brett D. Atwater, MD

7451 D, Duke North Hospital, Durham, NC 27710. Phone: 9192390127; Fax: 9196819260 E-mail: [email protected]

1

ACCEPTED MANUSCRIPT

The utility of routine TEE to exclude LAA thrombus prior to AF ablation in patients treated with NOAC therapy is unclear. This single center retrospective study sought to

RI PT

investigate the incidence of left atrial appendage (LAA) thrombus in patients undergoing routine transesophageal echocardiography (TEE) prior to atrial fibrillation (AF) ablation treated with warfarin or non-vitamin K oral anticoagulant (NOAC) therapy. We included

SC

937 routine pre-AF ablation TEE procedures performed in patients treated with warfarin (n=517) or NOAC (n=420). Patients were anticoagulated without interruption for at least

M AN U

4 consecutive weeks prior to the TEE. Patients treated with warfarin had lower LAA velocity and underwent TEE earlier in the study period than patients treated with NOAC (P<0.05). The incidence of LAA thrombus was higher in patients treated with warfarin (1.55%, 8/517) compared to patients treated with NOAC (0.24%, 1/420, P = 0.0473 for

TE D

difference). No LAA thrombus was identified in NOAC treated patients with a CHA2DS2VASC score <5 and in warfarin treated patients with a CHA2DS2-VASC score <2. TEE related complications occurred in 3/937 procedures (0.3%). In conclusion, LAA

EP

thrombus is detected rarely during pre-AF ablation TEE. Treatment with a NOAC is associated with a lower incidence of pre AF-ablation LAA thrombus compared to

AC C

warfarin.

Key Words: atrial fibrillation, catheter ablation, thrombus, transesophageal echocardiogram

2

ACCEPTED MANUSCRIPT

Despite routine use of peri-procedural anticoagulation, thromboembolic events occur in 1 - 5% of patients undergoing atrial fibrillation (AF) ablation.1,2 The risk of a thromboembolic event at the time of AF ablation varies depending on patient

RI PT

characteristics including the CHADS2 and CHA2DS2-VASC risk scores.3 To reduce the risk

of peri-procedural thromboembolism, transesophageal echocardiography (TEE) is used

SC

routinely to screen for LAA thrombus prior to AF ablation.4-6 Prior retrospective studies demonstrate that the incidence of LAA thrombus in patients taking chronic warfarin

M AN U

prior to AF ablation is between 0.25% and 5.3%.5,7,8 The incidence of LAA thrombus in patients taking chronic non-vitamin K oral anticoagulants (NOAC) prior to AF ablation is unknown. The 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of AF recommends performing TEE prior to AF ablation in patients on

TE D

warfarin without 3 consecutive weeks of therapeutic international normalized ratio (INR) values and considering TEE before AF ablation in all patients but has no recommendation for performing TEE before AF ablation in patients anticoagulated with

EP

NOACs.3 To better understand the usefulness of TEE to exclude LAA thrombus before planned AF ablation, we performed a comparative retrospective analysis of the

AC C

incidence of LAA thrombus in patients taking warfarin and NOAC anticoagulation. Methods: Cases were identified using the Duke Echocardiography Laboratory Database, (DELD).9 Clinical variables were extracted using the Duke Enterprise Data Unified Content Explorer (DEDUCE), an online query tool that organizes patient data for analysis.10 The institutional review board reviewed and approved this study and waived the need for informed consent.

3

ACCEPTED MANUSCRIPT

All consecutive patients who had a TEE at Duke University Medical Center between 1/1/2012 and 5/1/2015 were identified. (Figure 1) Patients were included if: 1)

RI PT

they were ≥ 18 years old, 2) the TEE was performed prior to a planned AF ablation, 3) they received anticoagulation with warfarin, rivaroxaban 20 mg or 10mg daily,

dabigatran 150 mg twice daily, or apixaban 5 mg or 2.5 mg twice daily dosed according

SC

to recommended guidelines and taken without interruption for at least 4 weeks prior to the TEE, and 4) patients receiving warfarin had an INR value measured within 4 weeks of

M AN U

the TEE. Patients were excluded if: 1) the TEE was performed for a reason other than ruling out LAA thrombus, 2) they had an arrhythmia other than AF, 3) the TEE was performed prior to a cardioversion without planned AF ablation, 4) the TEE findings were inconclusive, 5) the patient had prior surgical or percutaneous closure of the LAA,

TE D

or 6) the patient had prior mechanical aortic or mitral valve replacement, bioprosthetic mitral valve replacement, mitral valve repair, or moderate or severe mitral stenosis. Chart review was performed to determine the indication for TEE and the

EP

anticoagulant used at the time of TEE. Age and gender were available as demographic variables in DEDUCE. Other clinical variables were identified using ICD-9 diagnostic

AC C

codes. Comorbidities were defined as any inpatient or outpatient ICD-9 code of interest occurring between January 2002 and May 2015. ICD-9 codes for congestive heart failure (398.91, 402.01, 402.11, 402.91, 404.03, 404.11, 404.13, 404.91, 404.93, and 428.x), diabetes (250.x, 357.2, 362.0, and 366.41), hypertension (401.x, 402.x, 403.x, 404.x, and 405.x), stroke or transient ischemic attack (433.01, 433.11, 433.21, 433.31, 433.91, 434.x, 435.x, and v12.54), vascular disease (coronary artery disease 433.x, 410.x,

4

ACCEPTED MANUSCRIPT

411.x, 412.x, 413.x, and 414.x; other cerebrovascular disease 436.x, 437.x, 438.x, and 433.x; and atherosclerotic vascular disease 440.x, 447.1, 557.1, and 557.9) were chosen as described previously.11,12 The CHA2DS2-VASC score was calculated by adding risk

hypertension = 1, age 65-74 = 1 or age

RI PT

factor points.13 Patients are assigned points as follows: congestive heart failure = 1,

75 = 2, diabetes = 1, prior stroke, transient

SC

ischemic attack, or systemic embolism = 2, vascular disease = 1, female sex = 1.

Board-certified cardiologists with level 3 training in echocardiography performed

M AN U

TEE with an ie33 ultrasound system (Philips Healthcare, Andover, MA, USA). Cine loops of the LAA were acquired during stepwise rotation of the imaging sector from 0 to 180 degrees. LAA velocity was obtained using pulsed Doppler 1 cm into the LAA orifice. Thrombus was defined as an echodense intracavitary mass distinct from the underlying

TE D

endocardium. Microbubble contrast was administered as needed to improve visualization of the LAA. Acute TEE related complications were identified by the cardiologist prior to report finalization (< 24 hours) and were captured as free text fields

EP

in the DELD database. All complications underwent manual chart review. Major complications were defined as: 1) hemorrhage requiring transfusion, 2) respiratory

AC C

event requiring intubation, 3) any surgical intervention necessitated by attempted esophageal intubation, or 4) admission for monitoring post TEE. Minor complications were defined as all other complications reported at the time of TEE report finalization. Warfarin was not interrupted prior to TEE or AF ablation, and no heparin bridging was used. One dose of rivaroxaban was held the evening before the ablation while one dose of dabigatran and apixaban was held the morning of ablation. Adherence to prescribed

5

ACCEPTED MANUSCRIPT

anticoagulation was determined in a pre-ablation clinic visit, performed the day before or the day of the TEE. Patients who reported lack of adherence to the prescribed

RI PT

anticoagulant were excluded from the analysis. Ablation procedures were performed under general anesthesia with

administration of heparin to achieve a target activated clotting time > 300 seconds.

SC

Intracardiac echocardiography was used to assess atrial geometry and the LAA prior to trans-septal puncture. Double trans-septal access was then obtained under intracardiac

M AN U

echocardiographic guidance and an irrigated ablation catheter and a multipolar mapping catheter were used to perform left atrial mapping and ablation of the pulmonary vein antrum in all patients with paroxysmal AF. Additional ablation was performed in patients with persistent AF to isolate the left atrial posterior wall, coronary

TE D

sinus, and/or superior vena cava and to ablate atrial rotors according to the operator’s discretion. Cardioversion was performed when ablation did not restore sinus rhythm. The primary analyses compared the incidence of LAA thrombus in patients

EP

treated with warfarin versus NOAC. These were performed at both the TEE level and patient level to control for repeated measures. Secondary analysis compared the

AC C

frequency of patient characteristics among patients with versus without LAA thrombus. Multivariable logistic regression analysis was performed to identify independent predictors of LAA thrombus. Due to the small number of endpoints, only CHA2DS2-VASC score and NOAC versus warfarin therapy were included in the logistic regression analysis. Continuous variables are reported as mean (SD). Categorical data are reported as percentages with 2-sided 95% confidence limits. Normally distributed data were

6

ACCEPTED MANUSCRIPT

analyzed using Student’s t tests, Pearson chi-square tests or Fisher’s exact tests. Nonnormally distributed data were analyzed using Wilcoxon rank sum tests. Statistical

RI PT

analysis was performed using JMP Pro 11, SAS Institute, Cary, NC. For all comparisons, P < 0.05 was considered statistically significant.

Results: A total of 4013 TEE procedures were performed between 1/1/2012 and

SC

5/1/2015. Of these, a total of 864 patients were referred for 937 TEE procedures that met inclusion and exclusion criteria (Figure 1). At the time of TEE, 517 patients were

M AN U

treated with warfarin (55%), 90 were treated with apixaban (9.6%), 127 were treated with dabigatran (13.5%) and 203 were treated with rivaroxaban (21.7%). Patient characteristics are shown in Table 1. Patients were more frequently treated with warfarin than a NOAC at the beginning of the study period and were more frequently

TE D

treated with a NOAC than warfarin at the end (P < 0.01 for trend, Figure 2). Among patients treated with warfarin, 382/517 (74%) were therapeutic with an INR

2 with a

mean INR of 2.32 (0.59) at the time of the TEE (Figure 3).

EP

Microbubble echocardiographic contrast was administered in 22/517 (4.3%) patients treated with warfarin and 11/420 (2.6%) patients treated with NOAC (P = 0.21

AC C

for difference). TEE related events resulted in indeterminate imaging and exclusion from the study in 2 patients, one had pain with TEE probe insertion that prevented adequate imaging, one developed sustained ventricular tachycardia after lidocaine anesthetic was applied to the pharynx resulting in transfer to the emergency department without TEE imaging. Both patients underwent AF ablation without TEE imaging and had no AF ablation related complications. Among patients included in the analysis, TEE was

7

ACCEPTED MANUSCRIPT

performed without difficulty under moderate sedation in 929/937 patients (99%) but required laryngoscopy or general anesthesia in 8/937 (1%). No major TEE complications

RI PT

occurred. Minor TEE related complications occurred in 3/937 patients (0.3%); pharyngeal bleeding that did not require transfusion in 2 cases and bronchospasm in one case. No patient with a TEE related complication had LAA thrombus.

SC

The incidence of LAA thrombus was higher on TEEs performed in patients

treated with warfarin (1.55%, 8/517) compared to patients treated with NOAC (0.24%,

M AN U

1/420, P = 0.047 for difference). Among the patients treated with warfarin with LAA thrombus on TEE, 7/8 (88%) had an INR

2 at the time of the TEE while one patient had

an INR of 1.7; 3/8 patients with LAA thrombus had 4 consecutive therapeutic weekly INR values prior to TEE. The remaining warfarin treated patients with LAA thrombus did not

TE D

have weekly INR checks prior to TEE. The incidence of LAA thrombus was higher among patients therapeutic with warfarin at the time of TEE (1.8%, 7/399) compared to patients treated with a NOAC (0.24%, 1/420, P = 0.027 for difference). When analyses

EP

were repeated considering only a patient’s first TEE, the incidence of LAA thrombus remained higher in patients treated with warfarin (1.66%, 8/482) compared to patients

AC C

treated with NOAC (0.26%, 1/382, P = 0.044 for difference). Characteristics of patients without LAA thrombus are compared to patients with

LAA thrombus in Table 2. After adjusting for use of NOAC vs. warfarin, the CHA2DS2VASC score was a strong predictor of LAA thrombus identified on TEE in patients prior to AF ablation (odds ratio 1.4 for every 1 point increase in CHA2DS2-VASC score [95% CI 1.18-2.52 P = 0.02]) (Table 3). Detailed patient data from the 9 patients with LAA

8

ACCEPTED MANUSCRIPT

thrombus are presented in Table 4. Eliminating the use of routine pre-ablation TEE in patients treated with warfarin and a CHA2DS2-VASC score of 0-1 (n = 127/538, 25%) and

RI PT

in patients treated with NOAC and a CHA2DS2-VASC score 4 (n = 340/420, 81%) would have prevented 447/937 pre-AF ablation TEE procedures over the study period (a 48% reduction) and would have prevented 1/3 TEE related minor complications without

SC

missing any patients with LAA thrombus.

Discussion: We compared the incidence of LAA thrombus in patients treated with NOAC

M AN U

and warfarin anticoagulation prior to catheter ablation for AF. The major findings include (1) The incidence of LAA thrombus was lower in patients treated with NOAC compared to warfarin; (2) despite a high rate of therapeutic INR at the time of TEE, the incidence of LAA thrombus was 1.6% in patients anticoagulated with warfarin; (3) the

TE D

risk of LAA thrombus was predicted by the CHA2DS2-VASC score. Current expert statements recommend routine use of TEE in patients with persistent AF, left atrial enlargement, or an elevated CHA2DS2-VASC score however over

EP

70% of ablation centers report routine use of TEE in all patients prior to any AF ablation.4,14 A decision analysis examined routine use of TEE in an unselected population

AC C

anticoagulated with warfarin prior to AF ablation.15 They found that routine TEE lowered the risk of cerebral thromboembolic events but at a cost of $226,608 per qualityadjusted life year. This study assumed a positive TEE rate of 4%, more than twice the rate we found in patients treated with warfarin and 16 times the rate we found in patients treated with NOAC. Our results support prior work showing that the CHADS2 and CHA2DS2-VASC scores predict the presence of LAA thrombus in patients prior to AF

9

ACCEPTED MANUSCRIPT

ablation.7,8,16 While TEE provides high diagnostic accuracy; it is a semi-invasive procedure with associated risk. In our study, 1 patient could not complete the TEE due

RI PT

to procedure related complications (sustained VT), 9 required the use of general anesthesia to complete the TEE and 3 had minor TEE related complications. The

frequency of these events exceeded the frequency of LAA thrombus detected by TEE.

SC

The use of the CHA2DS2-VASC score to identify patients with a high pre-test probability for LAA thrombus who require pre-AF ablation TEE may improve the cost/benefit ratio

M AN U

of the procedure and reduce the frequency of TEE related complications. A recent study evaluated the safety of complete elimination of TEE prior to AF ablation in patients treated with at least 4 weeks of NOAC.17 Despite a mean CHA2DS2VASC score of 3 and inconsistent use of intracardiac echocardiography to rule out LAA

TE D

thrombus (71%), only 1/697 patients suffered a peri-procedural thromboembolic event. These results are consistent with our findings that patients treated with NOAC prior to AF ablation have a very low risk of LAA thrombus.

EP

This study was retrospective and therefore exploratory, dedicated prospective randomized controlled studies would be needed to conclusively determine CHA2DS2-

AC C

VASC score cutoffs where routine pre-AF ablation TEE could be safely avoided. However due to the low event rate, a randomized controlled trial of TEE prior to AF ablation may be impractical.15 This was a single-center study; a multicenter registry study would improve the ability to adjust for confounding variables using multivariable analysis. Weekly INR values before TEE were unavailable in some patients, limiting our ability to determine if sub-therapeutic anticoagulation was a potential cause of LAA thrombus in

10

ACCEPTED MANUSCRIPT

patients treated with warfarin. However, 3/8 patients treated with warfarin who had LAA thrombus identified on TEE had 4 consecutive weeks of therapeutic INR values

RI PT

documented prior to the TEE. We systematically asked patients if they had been compliant with their anticoagulation prescription without interruption for the 4 weeks before TEE and excluded patients who reported lack of compliance. The high rate of

SC

patient reported compliance with warfarin was confirmed by the fact that 74% of

patients had therapeutic INR at the time of TEE but there was no method to confirm the

M AN U

compliance of patients taking NOAC. Analyzing LAA velocity in both AF and sinus rhythm may have produced variability in the predictive performance of this variable. Patients with prior LAA occlusion were excluded from this study; future studies are needed

AC C

EP

TE D

exploring peri-AF ablation anticoagulation strategies in these patients.

11

ACCEPTED MANUSCRIPT

1. Scherr D, Sharma K, Dalal D, Spragg D, Chilukuri K, Cheng A, Dong J, Henrikson CA,

RI PT

Nazarian S, Berger RD, Calkins H, Marine JE. Incidence and predictors of periprocedural cerebrovascular accident in patients undergoing catheter ablation of atrial fibrillation. J Cardiovasc Electrophysiol 2009;20:1357-1363.

SC

2. Di Biase L, Burkhardt JD, Santangeli P, Mohanty P, Sanchez JE, Horton R, Gallinghouse GJ, Themistoclakis S, Rossillo A, Lakkireddy D, Reddy M, Hao S, Hongo R, Beheiry S,

M AN U

Zagrodzky J, Rong B, Mohanty S, Elayi CS, Forleo G, Pelargonio G, Narducci ML, Dello Russo A, Casella M, Fassini G, Tondo C, Schweikert RA, Natale A. Periprocedural stroke and bleeding complications in patients undergoing catheter ablation of atrial fibrillation with different anticoagulation management: results from the Role of Coumadin in

TE D

Preventing Thromboembolism in Atrial Fibrillation (AF) Patients Undergoing Catheter Ablation (COMPARE) randomized trial. Circulation 2014;129:2638-2644. 3. Calkins H, Kuck KH, Cappato R, Brugada J, Camm AJ, Chen SA, Crijns HJ, Damiano RJ,

EP

Jr., Davies DW, DiMarco J, Edgerton J, Ellenbogen K, Ezekowitz MD, Haines DE, Haissaguerre M, Hindricks G, Iesaka Y, Jackman W, Jalife J, Jais P, Kalman J, Keane D, Kim

AC C

YH, Kirchhof P, Klein G, Kottkamp H, Kumagai K, Lindsay BD, Mansour M, Marchlinski FE, McCarthy PM, Mont JL, Morady F, Nademanee K, Nakagawa H, Natale A, Nattel S, Packer DL, Pappone C, Prystowsky E, Raviele A, Reddy V, Ruskin JN, Shemin RJ, Tsao HM, Wilber D, Heart Rhythm Society Task Force on C, Surgical Ablation of Atrial F. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient

12

ACCEPTED MANUSCRIPT

management and follow-up, definitions, endpoints, and research trial design: a report of the Heart Rhythm Society (HRS) Task Force on Catheter and Surgical Ablation of Atrial

RI PT

Fibrillation. Developed in partnership with the European Heart Rhythm Association (EHRA), a registered branch of the European Society of Cardiology (ESC) and the

European Cardiac Arrhythmia Society (ECAS); and in collaboration with the American

SC

College of Cardiology (ACC), American Heart Association (AHA), the Asia Pacific Heart

Rhythm Society (APHRS), and the Society of Thoracic Surgeons (STS). Endorsed by the

M AN U

governing bodies of the American College of Cardiology Foundation, the American Heart Association, the European Cardiac Arrhythmia Society, the European Heart Rhythm Association, the Society of Thoracic Surgeons, the Asia Pacific Heart Rhythm Society, and the Heart Rhythm Society. Heart Rhythm 2012;9:632-696 e621.

TE D

4. Awad S, Mohajer K, Baranchuk A, Pal RS. Current clinical practice of transesophageal echocardiography and cardiac computed tomography prior to atrial fibrillation ablation in Canada. Int J Cardiol 2015;189:300-301.

EP

5. McCready JW, Nunn L, Lambiase PD, Ahsan SY, Segal OR, Rowland E, Lowe MD, Chow AW. Incidence of left atrial thrombus prior to atrial fibrillation ablation: is pre-

AC C

procedural transoesophageal echocardiography mandatory? Europace 2010;12:927-932. 6. Puwanant S, Varr BC, Shrestha K, Hussain SK, Tang WH, Gabriel RS, Wazni OM, Bhargava M, Saliba WI, Thomas JD, Lindsay BD, Klein AL. Role of the CHADS2 score in the evaluation of thromboembolic risk in patients with atrial fibrillation undergoing transesophageal echocardiography before pulmonary vein isolation. J Am Coll Cardiol 2009;54:2032-2039.

13

ACCEPTED MANUSCRIPT

7. Scherr D, Dalal D, Chilukuri K, Dong J, Spragg D, Henrikson CA, Nazarian S, Cheng A, Berger RD, Abraham TP, Calkins H, Marine JE. Incidence and predictors of left atrial

RI PT

thrombus prior to catheter ablation of atrial fibrillation. J Cardiovasc Electrophysiol 2009;20:379-384.

8. Khan MN, Usmani A, Noor S, Elayi S, Ching CK, Di Biase L, Patel D, Burkhardt JD,

SC

Cummings J, Schweikert R, Saliba W, Natale A. Low incidence of left atrial or left atrial

appendage thrombus in patients with paroxysmal atrial fibrillation and normal EF who

M AN U

present for pulmonary vein antrum isolation procedure. J Cardiovasc Electrophysiol 2008;19:356-358.

9. Velazquez EJ, Samad Z, Al-Khalidi HR, Sangli C, Grayburn PA, Massaro JM, Stevens SR, Feldman TE, Krucoff MW. The MitraClip and survival in patients with mitral regurgitation

1059 e1053.

TE D

at high risk for surgery: A propensity-matched comparison. Am Heart J 2015;170:1050-

10. Horvath MM, Winfield S, Evans S, Slopek S, Shang H, Ferranti J. The DEDUCE Guided

EP

Query Tool: Providing Simplified Access to Clinical Data for Research and Quality Improvement. J Biomed Inform 2011;44:266-276.

AC C

11. Navar-Boggan AM, Rymer JA, Piccini JP, Shatila W, Ring L, Stafford JA, Al-Khatib SM, Peterson ED. Accuracy and validation of an automated electronic algorithm to identify patients with atrial fibrillation at risk for stroke. Am Heart J 2015;169:39-44.e32. 12. Rothendler JA, Rose AJ, Reisman JI, Berlowitz DR, Kazis LE. Choices in the use of ICD9 codes to identify stroke risk factors can affect the apparent population-level risk factor prevalence and distribution of CHADS2 scores. Am J Cardiovasc Disease 2012;2:184-191.

14

ACCEPTED MANUSCRIPT

13. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-

RI PT

based approach: the euro heart survey on atrial fibrillation. Chest 2010;137:263-272. 14. Cappato R, Calkins H, Chen SA, Davies W, Iesaka Y, Kalman J, Kim YH, Klein G, Natale A, Packer D, Skanes A, Ambrogi F, Biganzoli E. Updated worldwide survey on the

Arrhythm Electrophysiol 2010;3:32-38.

SC

methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circ

M AN U

15. Gula LJ, Massel D, Redfearn DP, Krahn AD, Yee R, Klein GJ, Skanes AC. Impact of routine transoesophageal echocardiography on safety, outcomes, and cost of pulmonary vein ablation: inferences drawn from a decision analysis model. Europace 2010;12:1550-1557.

TE D

16. Wallace TW, Atwater BD, Daubert JP, Voora D, Crowley AL, Bahnson TD, Hranitzky PM. Prevalence and clinical characteristics associated with left atrial appendage thrombus in fully anticoagulated patients undergoing catheter-directed atrial fibrillation

EP

ablation. J Cardiovasc Electrophysiol 2010;21:849-852. 17. Di Biase L, Briceno DF, Trivedi C, Mohanty S, Gianni C, Burkhardt JD, Mohanty P, Bai

AC C

R, Gunda S, Horton R, Bailey S, Sanchez JE, Al-Ahmad A, Hranitzky P, Gallinghouse GJ, Reddy YM, Zagrodzky J, Hongo R, Beheiry S, Lakkireddy D, Natale A. Is transesophageal echocardiogram mandatory in patients undergoing ablation of atrial fibrillation with uninterrupted novel oral anticoagulants? Results from a prospective multicenter registry. Heart Rhythm 2016;13:1197-1202.

15

ACCEPTED MANUSCRIPT

Figure 1: Patient flow chart- AVR = aortic valve replacement; LAA = left atrial appendage; MS = mitral stenosis; MV = mitral valve; MVR = mitral valve replacement,

RI PT

TEE = transesophageal echocardiogram

Figure 2: Frequency of patients and type of anticoagulation during the study time

SC

period. Patients were more frequently treated with warfarin than non-vitamin K oral

anticoagulants (NOAC) at the beginning of the study period, but were more frequently

M AN U

treated with NOAC than warfarin at the end of the study period (P < 0.01 for trend).

Figure 3: Distribution of INR values at the time of TEE in patients treated with warfarin.

2 at the time of TEE. Frequencies < 0.5% are rounded down to

0%.

AC C

EP

warfarin with an INR

TE D

Mean INR = 2.3, SD = 0.59. A total of 382/517 (74%) patients were therapeutic on

16

ACCEPTED MANUSCRIPT

Table 1: Baseline patient characteristics. Patients on NOAC (n=420) 64 ± 10 286 (68%) 128 (30%) 77 (18%) 308 (73%) 67 (16%) 186 (44%) 2.9 ± 1.8 1.3 ± 0.3 95 ± 21 176 ± 17 30 ± 6 85 ± 26 56 ± 9 56 ± 24

P - value

1.4 ± 3.3

1.8 ± 4.5

0.20

18 (3.5%)

6 (1.4%)

0.06

AC C

EP

TE D

M AN U

Age, years (SD) 65 ± 10 Men 618 (66%) Heart Failure 301 (32%) Diabetes 199 (21%) Hypertension 701 (75%) Stroke 143 (15%) Vascular Disease 425 (45%) CHA2DS2-VASC (SD) 3.1 ± 2 INR (SD) 1.9 ± 0.7 Weight (kg, SD) 94 ± 22 Height (cm, SD) 174 ± 14 2 Body Mass Index (kg/m , SD) 30 ± 6.8 Pulse (beats per minute, SD) 86 ± 26 Ejection Fraction (%, SD) 56 ± 8 Left Atrial Appendage Velocity (cm/s, 54 ± 23 SD) Days from Transesophageal 1.6 ± 3.9 Echocardiogram to Atrial Fibrillation Ablation (mean, SD) Spontaneous Echocardiographic 24 (2.6%) Contrast Present Type of Atrial Fibrillation Paroxysmal 385 (41%) Persistent 478 (51%) Long Standing Persistent 74 (8%) Values are given as n (%) or as mean (SD), * P < 0.05

Patients on Warfarin (n=517) 65 ± 11 332 (64%) 173 (33%) 122 (24%) 393 (76%) 76 (15%) 239 (46%) 3.2 ± 2.1 2.3 ± 0.6 94 ± 23 173 ± 11 31 ± 7 87 ± 26 56 ± 9 52 ± 22

0.05 0.24 0.36 0.05 0.36 0.65 0.60 0.12 < 0.05* 0.31 < 0.05* 0.78 0.23 0.17 < 0.05*

RI PT

Overall (n=937)

SC

Variable

0.76 207 (40%) 269 (52%) 41 (8%)

178 (42%) 209 (50%) 33 (8%)

ACCEPTED MANUSCRIPT

Table 2: Baseline patient characteristics, patients without left atrial appendage thrombus versus those with left atrial appendage thrombus.

Overall (n=937)

Patients Without Thrombus (n=928) 65 ± 10 419 (45%) 612 (66%) 293 (32%) 194 (21%) 693 (74%) 141 (15%) 418 (45%) 3.1 ± 2 94 ± 22 174 ± 14 31 ± 7 56 ± 8 54 ± 23

P - value

65 ± 13 1 (11%) 6 (67%) 8 (89%) 5 (56%) 8 (89%) 2 (22%) 7 (78%) 4.8 ± 2 106 ± 31 176 ± 11 34 ± 8 35 ± 18 23 ± 9

0.85 <0.05* 1.0 <0.05* <0.05* 0.47 0.63 0.08 <0.05* 0.19 0.48 0.18 <0.05* <0.05*

4 (44%)

<0.05*

AC C

EP

TE D

M AN U

SC

Age, years (SD) 65 ± 10 Non-vitamin K Oral Anticoagulant 420 (45%) Men 618 (66%) Heart Failure 301 (32%) Diabetes 199 (21%) Hypertension 701 (75%) Stroke 143 (15%) Vascular Disease 425 (45%) CHA2DS2-VASC (SD) 3.1 ± 2 Weight (kg, SD) 94 ± 22 Height (cm, SD) 174 ± 14 2 Body Mass Index (kg/m , SD) 30 ± 6.8 Ejection Fraction, (%, SD) 56 ± 8 Left Atrial Appendage Velocity, 54 ± 23 (cm/s, SD) 24 (2.6%) 20 (2%) Spontaneous Echocardiographic Contrast Present Type of Atrial Fibrillation Paroxysmal 385 (41) 381 (41) Persistent 478 (51) 474 (51) Long Standing Persistent 74 (8) 73 (8) Values are given as n (%) or as mean (SD). *P < 0.05

Patients with Thrombus (n=9)

RI PT

Variable

0.89 4 (44) 4 (44) 1 (11)

ACCEPTED MANUSCRIPT

Table 3: Incidence of left atrial appendage thrombus by CHA2DS2-VASC score among patients treated with warfarin vs. non-Vitamin K oral anticoagulant. P = 0.02 for trend.

0/42 0/85 2/74 0/110 1/69 5/137

Non-Vitamin K Oral Anticoagulant 0/33 0/69 0/79 0/83 0/76 1/80

RI PT

Warfarin

AC C

EP

TE D

M AN U

SC

CHA2DS2-VASC Score 0 1 2 3 4 >4

ACCEPTED MANUSCRIPT

Table 4: Patient characteristics, patients with left atrial appendage thrombus.

<1

5 69 No Yes Yes Yes No Yes No

8 77 No Yes Yes Yes Yes Yes No

6 58 No Yes Yes Yes Yes Yes No

6 76 Yes Yes No Yes No Yes No

Mild Yes

Trivial No

N/A 16.6

2.7 34.2

35 Small No

20 Small No

8

11

4

2

2 66 No Yes No Yes No No No

2 77 No No No No No No No

4 67 No Yes No Yes No Yes Yes

5 42 Yes Yes Yes Yes No Yes No

5 49 Yes Yes Yes Yes No Yes No

No No

No Yes

Mild No

No No

Mild No

Mod Yes

1.7 30

2.0 20.4

2.0 28.7

2.2 17.1

3.1 34.2

1.6 13.3

2.3 11

15 Mod Yes

55 Small No

20 Large No

60 Small Yes

60 Mod No

50 Mod No

25 Large yes

TE D Mild Yes

2

SC

4

M AN U

12

RI PT

#2 #3 #4 #5 #6 #7 #8 #9 Warfarin Warfarin Warfarin Warfarin Warfarin Warfarin Warfarin Warfarin Persistent Persistent Paroxysmal Persistent Persistent Paroxysmal Paroxysmal Paroxysmal

EP

Years of Atrial Fibrillation CHA2DS2-VASC Age (years) Female Heart Failure Diabetes Hypertension Prior Stroke Vascular Disease Prior Left Atrial Appendage Thrombus Mitral Regurgitation Spontaneous Echocardiographic Contrast Lowest INR Left Atrial Appendage Velocity Ejection Fraction (%) Thrombus Size Thrombus Mobile

#1 Dabigatran Long Standing Persistent 2

AC C

Patient Anticoagulant Type of Atrial Fibrillation

1

4013 TEE Cases from 1/1/2012-5/1/2015

RI PT

ACCEPTED MANUSCRIPT

M AN U

SC

Exclude 3008 Cases Performed for non-AF Ablation Indication 1005 TEE Cases

TE D

962 TEE Cases

Exclude 43 Cases with Moderate or Severe MS, Mechanical AVR or MVR, Biological MVR, or MV ring

Exclude 15 Cases with Over-sewn LAA

AC C

EP

947 TEE Cases

Exclude 4 Cases Performed without Anticoagulation

943 TEE Cases Exclude 6 Cases with Indeterminate Results 937 TEE Cases

ACCEPTED MANUSCRIPT

100 90

RI PT

70 60 50

Warfarin

40

NOAC

30

SC

Number of Patients

80

20 10

AC C

EP

TE D

M AN U

0



RI PT

ACCEPTED MANUSCRIPT

M AN U

15% 14%

9%

8%

0%

0

0%

0%

0.5

0%

1

1.5

50

7%

EP AC C

1%

75

TE D

12%

3%

100

2

2.5

Number of Patients

125

SC

24%

25 2% 1%

3

INR

3.5

1% 0%

4

0%

0%

4.5

0%

0%

5

0%

0%

5.5

0%

6