Patrick Tchou

Cleveland Clinic, Cleveland, Ohio, United States

Are you Patrick Tchou?

Claim your profile

Publications (284)2128.11 Total impact

  • Mark J Niebauer · John Rickard · Patrick J Tchou · Niraj Varma
    [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: QRS characteristics are the cornerstone of patient selection in cardiac resynchronization therapy (CRT) and the presence of left bundle branch block (LBBB) and baseline QRS ≥150ms portends a good outcome. We previously showed that baseline QRS frequency analysis adds predictive value to LBBB alone and have hypothesized that a change in frequency characteristics following CRT may produce additional predictive value. Methods: We examined the QRS frequency characteristics of 182 LBBB patients before and soon after CRT. Patients were assigned to responder and nonresponder groups. Responders were defined by a decrease in left ventricular end systolic volume (LVESV) ≥15% following CRT. We analyzed the QRS in ECG leads I, AVF and V3 before and soon after CRT using the discrete Fourier transform algorithm. The percentage of total QRS power within discrete frequency intervals before and after CRT was calculated. The reduction in lead V3 power <10Hz was the best indicator of response. Results: Baseline QRS width was similar between the responders and nonresponders (162.2±17.2ms vs 158±22.1ms, respectively; p = 0.180). Responders exhibited a greater reduction in QRS power <10Hz (-17.0±11.9% vs -6.6±12.5%; p<0.001) and a significant AUC (0.743; p<0.001). A ≥8% decline in QRS power <10Hz produced the best predictive values (PPV = 84%, NPV = 59%). Importantly, when patients with baseline QRS <150ms were compared, the AUC improved (0.892, p<0.001). Conclusions: Successful CRT produces a significant reduction in QRS power below 10 Hz, particularly when baseline QRS <150ms. These results indicate that QRS frequency changes after CRT provide additional predictive value to QRS alone. This article is protected by copyright. All rights reserved.
    No preview · Article · Jan 2016 · Journal of Cardiovascular Electrophysiology
  • [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Various ablation strategies of persistent atrial fibrillation (PersAF) have had disappointing outcomes, despite concerted clinical and research efforts, which could reflect progressive atrial fibrillation-related atrial remodeling. METHODS AND RESULTS: Two-year outcomes were assessed in 1241 consecutive patients undergoing first-time ablation of PersAF (2005-2012). The time intervals between the first diagnosis of PersAF and the ablation procedures were determined. Patients had echocardiograms and measures of B-type natriuretic peptide and C-reactive protein before the procedures. The median diagnosis-to-ablation time was 3 years (25th-75th percentiles 1-6.5). With longer diagnosis-to-ablation time (based on quartiles), there was a significant increase in recurrence rates in addition to an increase in B-type natriuretic peptide levels (P=0.01), C-reactive protein levels (P<0.0001), and left atrial size (P=0.03). The arrhythmia recurrence rates over 2 years were 33.6%, 52.6%, 57.1%, and 54.6% in the first, second, third, and fourth quartiles, respectively (Pcategorical<0.0001). In Cox Proportional Hazard analyses, B-type natriuretic peptide levels, C-reactive protein levels, and left atrial size were associated with arrhythmia recurrence. The diagnosis-to-ablation time had the strongest association with the ablation outcomes which persisted in multivariable Cox analyzes (hazard ratio for recurrence per +1Log diagnosis-to-ablation time 1.27, 95% confidence interval 1.14-1.43; P<0.0001; hazard ratio fourth versus first quartile 2.44, 95% confidence interval 1.68-3.65; Pcategorical<0.0001). CONCLUSIONS: In patients with PersAF undergoing ablation, the time interval between the first diagnosis of PersAF and the catheter ablation procedure had a strong association with the ablation outcomes, such as shorter diagnosis-to-ablation times were associated with better outcomes and in direct association with markers of atrial remodeling.
    No preview · Article · Jan 2016 · Circulation Arrhythmia and Electrophysiology
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Introduction: Persistent atrial fibrillation (PerAF) ablation has been associated with significant recurrence rates which could reflect progressive AF-related atrial remodeling. We hypothesized that the first-diagnosis to ablation time for PerAF is a major determinant of success rates and in direct association with pathways of atrial remodeling. Methods: Two-year outcomes were assessed in 1241 patients undergoing first time ablation of PerAF between January 2005 and December 2012 at our institution. The time intervals between the first diagnosis of PerF and the ablation procedures were determined. Patients had echocardiograms and measures of B-type natriuretic peptide (BNP) and C-reactive protein (CRP) before the ablation procedures. During ablations, patients with atrial scarring by voltage were identified. Results: The median time-to-ablation since the first PerAF diagnosis was 3 years (interquartile range 1-6.5). With longer diagnosis-to-ablation time (based on quartiles), there was a significant increase in BNP levels (p=0.01), CRP levels (p<0.0001), left atrial size (p=0.03) and scarring (p=0.04). Atrial arrhythmia recurred after a single ablation in 555 patients (44.7%); and 364 (29.3%) underwent repeat ablations. At last follow-up, 1005 patients (81.0%, 390 on antiarrhythmic medications) were either arrhythmia free or had their arrhythmia controlled. In Cox Proportional Hazard analyzes, BNP levels, CRP levels, left atrial size and scarring were associated with arrhythmia recurrence. The diagnosis-to-ablation time had the strongest association with success rates which persisted in multivariate Cox analyzes (HR for recurrence per +1Log diagnosis-to-ablation time 1.25, 95%CI 1.11-1.42, p<0.0001; 4th vs. 1st quartile 2.27, 95%CI 1.52-3.47, p<0.0001). Conclusions: The success rates with PerAF ablation are highest with early intervention, that is ablation before the progression of atrial remodeling.
    Full-text · Article · Nov 2015 · Circulation
  • [Show abstract] [Hide abstract]
    ABSTRACT: Many patients with drug refractory atrial fibrillation (AF) have a history of a prior cerebrovascular event (CVE). These patients are considered to be at high procedural risk for catheter ablation but data are scant.
    No preview · Article · Nov 2015 · JACC Clinical Electrophysiology
  • Source

    Full-text · Article · Sep 2015 · JAMA Internal Medicine
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: Left atrial flutter following atrial fibrillation (AF) ablation is increasingly common and difficult to treat. We evaluated the safety and efficacy of ablation of the anteroseptal line connecting the right superior pulmonary vein (RSPV) to the anteroseptal mitral annulus (MA) for the treatment of perimitral flutter (PMF). Methods: We systematically studied patients who were previously treated with AF ablation and who presented to the electrophysiology laboratory with atrial tachyarrhythmias between January 2000 and July 2010. The diagnosis of PMF was confirmed by activation mapping and/or entrainment. After re-isolation of any recovered pulmonary vein, a linear radiofrequency (RF) ablation was performed on the line that connected the RSPV to the anteroseptal MA. In this analysis, we included only patients who were treated with an anteroseptal line for their PMF. Results: Ablation was performed at the anteroseptal line in 27 PMF patients (63±13 years; 9 women) who had undergone prior ablation for paroxysmal (n=3) or persistent (n=24) AF, using electroanatomic activation mapping (70% CARTO, 30% NavX). The anteroseptal ablation line was effective in 22/27 (81.5%) patients in the acute-care setting. Termination of AF to sinus rhythm occurred in 15/22 (68.2%) patients, and 7/22 (31.8%) patients׳ AF converted to another right or left atrial flutter. At the 6-month follow-up, 20% of patients demonstrated recurrent left atrial tachyarrhythmia. Only one patient required repeat ablation, and the remaining patients׳ condition was controlled with antiarrhythmic medications. No major procedural complications or heart block occurred. Conclusion: Ablation at the left atrial anteroseptal line is safe and efficacious for the treatment of PMF. Unlike ablation at the traditional mitral isthmus line, ablation at the left atrial anteroseptal line does not require ablation in the coronary sinus.
    Full-text · Article · Jul 2015 · Journal of Arrhythmia
  • Paul Egan · Bruce L Wilkoff · Patrick Tchou
    [Show abstract] [Hide abstract]
    ABSTRACT: Unnecessary ventricular pacing from Cardiac Implantable Electronic Devices has been associated with long-term risks (heart failure, atrial fibrillation, and possibly stroke). Several device programming strategies are available to minimize ventricular pacing, each potentially associated with unintended consequences. Occasionally, the only effective means is to program to the AAI(R) pacing mode. However, in one manufacturer's implantable cardioverter defibrillators (ICD), the AAI(R) mode has the potential risk of prolonged pacing cessation following a nonsustained ventricular tachycardia (NSVT). Patients with ICDs, managed through the Cleveland Clinic device clinic, follow the Heart Rhythm Society consensus document recommendations for device monitoring with remote interrogations (every three months) and yearly in-person evaluations. Clinically significant findings also trigger additional evaluations by the nurse and physician. Two patients having Boston Scientific ICDs (E110 Teligen 100), had asystole and marked bradycardia following untreated NSVT. These pauses in pacing were due to use of the AAI(R) pacing mode. In order to enhance ventricular tachycardia detection, by design atrial pacing is disabled during, and for a time after, an episode of ventricular tachycardia when the device operates in the "ventricular tachycardia response" (VTR) phase. Thus, following spontaneous termination of the NSVT, no pacing occurred in these patients until the VTR period ended. Non-conventional programming was utilized to permit AAI(R) pacing while avoiding these asystolic and bradycardic events during VTR. Unintended consequences can occur when complex VT detection parameters interact with specific pacing modes. At times, non-conventional programming can avoid these interactions while still achieving effective AAI(R) pacing. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
    No preview · Article · May 2015 · Pacing and Clinical Electrophysiology
  • Source

    Full-text · Conference Paper · May 2015
  • Source

    Full-text · Article · May 2015 · Heart Rhythm
  • [Show abstract] [Hide abstract]
    ABSTRACT: Limited data exist outcome of atrial fibrillation (AF) surgery and catheter ablation in patients with hypertrophic cardiomyopathy (HCM). Our aim was to evaluate the safety and efficacy of non-pharmacologic treatment of AF in HCM. 147 patients [age 55±11 years; 46 females; ejection fraction (EF) 58±8%] with symptomatic paroxysmal (58%), persistent (31%), and long-standing persistent AF (11%) refractory to anti-arrhythmic drugs who presented for their first catheter ablation (n=79) or AF surgery (n=68) were included. After a follow-up of 35 months (interquartile range:13,60), 29% of patients who underwent catheter ablation and 51% of those who had AF surgery had no documented recurrent atrial arrhythmia after a single procedure. Repeat ablation was preformed in 55% of patients with recurrent arrhythmia in the catheter group, and 24% in the surgery group, increasing the success rate to 39% and 53% respectively after one or more procedures. Predictors of success after the first procedure in a multivariable setting included higher baseline EF, and male gender. Persistent or long-standing AF, and log of AF duration were associated with lower success. Major complications occurred in 6% of the catheter ablation and 18% of the AF surgery groups. On follow-up, 16 patients died (9 in catheter and 7 in surgery group) and 1 underwent heart replacement. Lower baseline EF, and older age were independently associated with death. Catheter ablation and AF surgery are associated with symptomatic improvement in HCM patients. However, long-term success is lower and complications are higher than previously published. Copyright © 2015. Published by Elsevier Inc.
    No preview · Article · Mar 2015 · Heart rhythm: the official journal of the Heart Rhythm Society

  • No preview · Article · Mar 2015 · Journal of the American College of Cardiology

  • No preview · Article · Mar 2015 · Journal of the American College of Cardiology
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Cardiac Sarcoidosis (CS) can lead to life-threatening ventricular dysrhythmias and sudden death. Immunosuppressive medications, radiofrequency ablation (RFA), and implantable cardioverter defibrillators (ICD) have been utilized to manage ventricular dysrhythmias but their benefits remain poorly defined. Objective: The aim of this study is to assess the durability of RFA in CS population and to determine outcome predictors after RFA. Methods: We compared the CS patients who had RFA±ICD against those with only ICD placement and contemporaneous patients with arrhythmogenic right ventricular dysplasia (ARVD) who had RFA. We analyzed time to a composite first event of appropriate ICD therapy, subsequent RFA, cardiac transplantation or death. We also evaluated variables predicting recurrence of ventricular dysrhythmias, including LVEF, cardiac involvement on PET scan, percent of ventricular ectopic beats, number of inducible VT foci and success of the RFA procedure. We used propensity matching and multivariable regression to adjust for baseline differences between the groups to identify outcome predictors. Results: Thirty ablations for VT were performed in 20 CS patients (13 had concomitant ICD placement); 12 ablations were done in eight ARVD patients and 33 CS patients with only ICD placements were included in this cohort. The median follow-up period was 48 (9-173) months. Fourteen (70%) patients reached composite end points after RFA compared to 13 (63%) following ICD placement and five (87%) in the ARVD cohort. There was a significant time difference to reach composite end points (p=0.02) in favor of ICD only cohort. The median number of ICD therapies were higher in the CS-RFA group (p=0.01). The requirement for ICD therapy increased over time following RFA, especially after 12 months. Variables predicting earlier time-to-event were EF < 40% (OR=13.2) and unsuccessful RFA procedure (OR=7.9). The presence of more than one inducible VT morphology was associated with higher likelihood of unsuccessful RFA (p=0.03). Conclusion: RFA can be an effective modality for the short-term treatment of ventricular dysrhythmias in cardiac sarcoidosis; however, after more than 12 months, the number of appropriate therapies escalates. Accordingly, ICD placement is recommended for all patients who undergo RFA for VT associated with CS, whether it is successful or not. Low LVEF and unsuccessful ablation were strong predictors of future events.
    No preview · Article · Jan 2015 · Sarcoidosis, vasculitis, and diffuse lung diseases: official journal of WASOG / World Association of Sarcoidosis and Other Granulomatous Disorders
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background Baseline QRS duration (QRSd) ≥150 ms is a recognized predictor of clinical improvement by cardiac resynchronization therapy (CRT), particularly for those with left bundle branch (LBBB). Patients with QRSd <150ms, are considered less likely to respond. Objective We theorized that left ventricular dyssynchrony, while usually associated with wider QRSd, also exhibits lower QRS frequency characteristics and that low frequency content should predict CRT response in LBBB patients. Methods We retrospectively examined the QRS frequency content of 170 heart failure patients with LBBB and QRSd ≥120 ms using the Fourier transformation. Ninety-four responders to CRT (definition; reduction in left ventricular end-systolic volume by ≥15% from baseline) were compared to 76 nonresponders (<15% reduction). The analysis of three standard ECG leads (I, AVF and V3) representing the three dimensions of depolarization, was performed and V3 provided the best predictive value. Results The QRSd of responders (160.3±17.8ms) and nonresponders (161.8±21.1ms; p=0.604) were similar. We found that the percentage of total QRS frequency power <10Hz that exceeded 52% was most predictive of CRT response compared to other cutoff values. However, the percentage of patients with total QRS power >52% below 10 Hz was especially predictive of response in those with QRSd <150ms. In these patients, this power threshold was highly predictive of CRT response (PPV=85.7% and NPV=71.4%). Conclusions In this group of CRT recipients with LBBB, retrospective analysis of QRS frequency content below 10 Hz had greater predictive value for CRT response than baseline QRSd, particularly in those with QRSd<150ms.
    No preview · Article · Dec 2014 · Heart Rhythm
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Direct Electrical Cardioversion (DCC) has been known to increase the risk of thromboembolism (TE) in patients with Atrial Fibrillation (AF). However, guidelines and current practices are unclear if this risk exists in patients undergoing DCC < 48 hours after AF onset. We aim to assess TE risk in these patients with and without the use of therapeutic anti-coagulation. Methods: All DCC done at Cleveland Clinic between 1996 to 2012, <48 hours after AF onset were selected from Electrophysiology database. They were divided into two groups on the basis of their anti-coagulation status. We did extensive chart review in these patients to look for outcomes of major TE complications within a month of DCC. Results: Among 567 DCC in 510 patients without therapeutic anti-coagulation, the mean (SD) CHA2DS2-VASc score was 2.34(1.66). At the time of DCC, 54% of these patients were on aspirin, while 12% were on warfarin with INR ≤1.5. There were 7 reported cerebrovascular accidents (CVAs) (1.22%) within a month after DCC. In these 7 patients, the mean (SD) CHA2DS2-VASc score was 3.57(1.27) ranging between 2 to 5. In the second group, 901 DCC were done in 733 patients who were therapeutic on warfarin (63%) or heparin (37%) at the time of DCC. Mean (SD) CHA2DS2-VASc score was 2.61(1.73). There were 2 reported (0.22%, p= 0.016) CVAs within a month after DCC. These two had CHA2D2-VASc scores of 4 and 6. Of these two patients, one was sub- therapeutic (INR=1.14) when he presented with neurological deficits 18 days after the DCC. The other patient held anticoagulation 10 days after DCC for a surgical procedure which was followed by the TE stroke a week later. The non-treated group therefore had a 5.6 times greater odds of having a thromboembolic stroke within a month of DCC (95% CI: 1.16 to 27.14) . Conclusions: In patients with acute onset atrial fibrillation, odds of thromboembolic complications are over 5 times higher in patients who did not receive therapeutic anti-coagulation at the time of DCC, despite having a lower baseline stroke risk as defined by their CHA2DS2-VASc scores. In addition, the two patients in our study who did have a stroke in the therapeutically anti-coagulated group had ceased their anticoagulant prior to the stroke.
    No preview · Conference Paper · Jul 2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background QRS morphology and duration (QRSd) determine CRT candidate selection but criteria require refinement. Objective Assess CRT effect according to QRSd, treated by dichotomization vs a continuous function, and modulation by gender. Methods Patients selected were NYHA Class III/IV with LBBB and non-ischemic cardiomyopathy (to test “pure” CRT effect) with pre- and post-implant echocardiographic evaluations. Positive response was defined as improved LVEF post-CRT. Results In 212 patients (LVEF 19±7.1%; QRSd 160±23 ms; 105 females), CRT improved LVEF to 30±15% (p<0.001) during median 2 years follow-up. Positive response occurred in 150/212 (71%). Genders did not differ for QRSd, pharmacotherapy and comorbidities, but female CRT response was greater: incidence 84% (88/105) vs 58% (62/107) in males (p<0.001); LVEF improvement 15±14% vs 7.2±13%, respectively (p<0.001). Overall, response rate was 58% when QRSd<150ms vs 76% when QRSd≥150 ms (p=0.009). This probability differed between genders: 86% in women vs 36% in men (p<0.001) for QRSd <150 ms, and 83% vs 69% respectively when QRSd≥150ms (p=0.05). Thus, female response rates remained high whether QRSd was < or ≥150 ms (86 vs 83%, p=0.77) but differed in males (36 vs 69%, p<0.001). With QRSd as a continuum, the CRT-response relationship was nonlinear, and significantly different between genders. Female superiority at shorter QRSd inverted with prolongation >180 ms. Conclusions The QRSd-CRT response relationship in heart failure patients with LBBB and non-ischemic cardiomyopathy is better described by a sex-specific continuous function and not by dichotomization by 150 ms which excludes a large proportion of women with potentially favorable outcome.
    No preview · Article · Jul 2014 · Heart rhythm: the official journal of the Heart Rhythm Society
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Despite sparse clinical data, current atrial fibrillation (AF) guidelines favor amiodarone as a drug of choice for patients with left ventricular hypertrophy (LVH). This study tested the hypothesis that patients with persistent AF and LVH on nonamiodarone antiarrhythmics have higher mortality compared to patients on amiodarone. In an observational cohort analysis of patients who underwent cardioversion for AF, patients with LVH, defined as left ventricular wall thickness ≥1.4 cm, by echocardiogram prior to their first cardioversion, were included; clinical data, including antiarrhythmic drugs and ejection fraction (LVEF), were collected. Mortality, determined via the Social Security Death Index, was analyzed using Kaplan-Meier and Cox proportional hazards models to determine whether antiarrhythmic drugs were associated with higher mortality. In 3,926 patients, echocardiographic wall thickness was available in 1,399 (age 66.8 ± 11.8 years, 67% male, LVEF 46 ± 15%, septum 1.3 ± 0.4, posterior wall 1.2 ± 0.2 cm), and 537 (38%) had LVH ≥1.4 cm. Among 537 patients with LVH, mean age was 67.5 ± 11.7 years, 76.4% were males, and mean LVEF was 48.3 ± 13.3%. Amiodarone was associated with lower survival (log rank P = 0.001), including after adjusting for age, LVEF, and coronary artery disease (P = 0.023). In propensity-score matched cohorts with LVH treated with no drugs, nonamiodarone antiarrhythmic drugs (non-AADs), or amiodarone (N = 65 each group), there was early lower survival in patients on amiodarone (P = 0.05). Patients with persistent AF and LVH on non-AADs do not have higher mortality compared to patients on amiodarone. Importantly, these findings do not support amiodarone as a superior choice in patients with LVH.
    Full-text · Article · May 2014 · Pacing and Clinical Electrophysiology

  • No preview · Article · Mar 2014 · Journal of the American College of Cardiology
  • Source

    Full-text · Article · Mar 2014 · Journal of the American College of Cardiology
  • [Show abstract] [Hide abstract]
    ABSTRACT: Many patients eligible for cardiac resynchronization therapy (CRT) are over 80 years of age. Survival in this population and how it compares to the general octogenarian population has not been established. We extracted clinical data on a cohort of 800 consecutive patients undergoing the new implantation of a CRT device between April 15, 2004 and August 6, 2007. Patients over age 80, with class III-IV New York Heart Association heart failure symptoms on optimal medical therapy undergoing initial CRT implantation, were included in the final cohort. Using the United States Social Security Period Life Table for 2006, fractional survival for octogenarians in the general population was calculated and matched to our cohort based on age and gender. A comparison was then made between octogenarians undergoing CRT compared to the general population. A total of 95 octogenarians who met inclusion criteria were identified, of whom 86.3% received a biventricular defibrillator and the remainder a biventricular pacemaker. Over a mean follow-up of 3.6 ± 1.5 years, there were 47 deaths (47.4%). The mean survival time was 4.1 years (95% CI 3.7-4.5), and survival at 2 years was 78.9%. Compared to the general octogenarian population, octogenarians receiving CRT had only modestly worse survival over the duration of follow-up with the survival curves diverging at 2 years of follow-up (P = 0.03). Octogenarians with advanced heart failure have a reasonable mean survival time following CRT. All-cause mortality in this patient population is only modestly worse compared to the general octogenarian population. Therefore, in octogenarians deemed to be reasonable candidates, CRT should not be withheld based on age alone.
    No preview · Article · Jan 2014 · Pacing and Clinical Electrophysiology

Publication Stats

10k Citations
2,128.11 Total Impact Points


  • 1996-2015
    • Cleveland Clinic
      • • Department of Cardiovascular Medicine
      • • Center for Atrial Fibrillation
      • • Department of Cardiology
      Cleveland, Ohio, United States
  • 2009
    • Metropolitan Heart and Vascular Institute
      Minneapolis, Minnesota, United States
  • 1986-2008
    • University of Wisconsin - Milwaukee
      Milwaukee, Wisconsin, United States
    • Mount Sinai Medical Center
      New York, New York, United States
  • 2004-2007
    • University of Chicago
      Chicago, Illinois, United States
  • 2006
    • Philadelphia ZOO
      Philadelphia, Pennsylvania, United States
  • 2003
    • Baptist Hospital
      Nashville, Tennessee, United States
  • 2001-2002
    • Case Western Reserve University
      • Department of Biomedical Engineering
      Cleveland, Ohio, United States
    • University of Illinois at Chicago
      Chicago, Illinois, United States
  • 1998-2001
    • The Ohio State University
      • Division of Cardiology
      Columbus, Ohio, United States
  • 1991-1997
    • University of Pittsburgh
      • Division of Cardiology
      Pittsburgh, Pennsylvania, United States
    • Samaritan Medical Center
      ART, New York, United States
  • 1991-1993
    • Mount Sinai Hospital
      New York, New York, United States
  • 1989
    • University of Oklahoma Health Sciences Center
      Oklahoma City, Oklahoma, United States