- Email: [email protected]
Upgrade and de novo cardiac resynchronization therapy: Impact of paced or intrinsic QRS morphology on outcomes and survival
Upgrade and de novo cardiac resynchronization therapy: Impact of paced or intrinsic QRS morphology on outcomes and survival Anita Wokhlu, MD, Robert F. Rea, MD, Samuel J. Asirvatham, MD, FHRS, Tracy Webster, RN, Kelly Brooke, RN, David O. Hodge, BA, MS, Heather J. Wiste, BA, YingXue Dong, MD, PhD, David L. Hayes, MD, FHRS, Yong-Mei Cha, MD From the Mayo Clinic, Rochester, Minnesota. BACKGROUND Cardiac resynchronization therapy (CRT) improves outcomes in patients with left bundle branch block (LBBB), but the benefits of CRT in patients with other QRS morphologies or previous pacing are uncertain. OBJECTIVE The purpose of this study was to describe outcomes in patients with prior right ventricular pacing and non-LBBB morphologies. METHODS We studied 505 patients who underwent de novo CRT (n ⫽ 338) or CRT upgrade (n ⫽ 167). De novo patients were categorized by underlying QRS morphology: LBBB (67%), right bundle branch block (RBBB; 11%), intraventricular conduction delay (IVCD; 13%), and QRS ⬍120 ms (9%). Upgrade patients were categorized by the percentage of previous ventricular pacing. RESULTS Patients were followed for death over a median of 2.6 years (interquartile range 1.6 – 4.0). New York Heart Association (NYHA) functional class and echocardiographic improvements were similar in de novo and upgrade patients. However, within the de novo group, NYHA improvements were less in patients with RBBB (0.3 ⫾ 0.8; P ⫽ .014) or IVCD (0.2 ⫾ 0.7; P ⫽ .001) than in those with LBBB (0.7 ⫾ 0.8). These patients had less left ventricular functional improvement as well. Survival was comparable after de novo versus upgrade CRT (61% vs 63% at 4 years; P
Introduction Cardiac resynchronization therapy (CRT) has been shown to improve outcomes in trials in which most patients with drug-refractory heart failure (HF) had intrinsic left bundle branch block (LBBB).1– 4 However, the benefits of CRT have not been conclusively demonstrated in patients with Dr. Wokhlu has received travel support from Medtronic, Boston Scientific, and St. Jude Medical to attend educational seminars. Dr. Hayes is an educational speaker and serves on the advisory boards for Medtronic, Boston Scientific, Sorin Medical Group, and St. Jude Medical, and he is a member of the steering committee for St. Jude Medical. Dr. Asirvatham receives honoraria and is on the speaker’s board for St. Jude Medical, Boston Scientific, and Medtronic. In addition, he is a co-patent holder for an alternative resynchronization therapy technique. Ms. Webster is on an advisory board for Boston Scientific. Address reprint requests and correspondence: Dr. Yong-Mei Cha, Division of Cardiovascular Diseases, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905. E-mail address: [email protected]. (Received March 29, 2009; accepted July 6, 2009.)
⫽ .906). No clinical or survival differences were noted in upgrade patients based on the percentage of previous pacing. However, survival in de novo CRT recipients with RBBB (32%) was lower than in those with LBBB (66%; P ⬍.001), and RBBB independently predicted death (hazard ratio 3.5, confidence interval 1.9 – 6.5; P ⬍.001). CONCLUSION RBBB and IVCD result in less clinical improvement or worsened survival after CRT. Additional selection criteria may be beneficial in identifying potential responders with RBBB, IVCD, or narrow QRS. KEYWORDS Biventricular pacing; Cardiac resynchronization therapy; Left bundle branch block; Right bundle branch block; QRS morphology; Upgrade; Ventricular pacing ABBREVIATIONS CI ⫽ confidence interval; CRT ⫽ cardiac resynchronization therapy; HF ⫽ heart failure; HR ⫽ hazard ratio; IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; LV ⫽ left ventricular; LVEDV ⫽ left ventricular end-diastolic volume; LVEF ⫽ left ventricular ejection fraction; LVESV ⫽ left ventricular end-systolic volume; RBBB ⫽ right bundle branch block; RV ⫽ right ventricular (Heart Rhythm 2009;6:1439 –1447) © 2009 Heart Rhythm Society. All rights reserved.
other native QRS morphologies or those who have undergone right ventricular (RV) pacing in the setting of a preexisting pacemaker or defibrillator due to limited representation within randomized controlled trials and variable outcomes. The impact of QRS morphology on HF outcomes has been examined in four randomized trials of CRT.1,2,4,5 In a substudy of CRT-D recipients from the MIRACLE (Multicenter InSync Randomized Clinical Evaluation) ICD study, RBBB was associated with less improvement in peak oxygen consumption than other QRS morphologies in univariate, but not multivariate, analysis.6 In a retrospective, pooled analysis from MIRACLE and CONTAK-CD that compared outcomes in 34 patients with RBBB who received CRT and 27 who received medical management, New York Heart Association (NYHA) functional class was the only clinical outcome that improved significantly more after active CRT. Improvements in other HF parameters, such as
1547-5271/$ -see front matter © 2009 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2009.07.009
1440 6-minute walk, peak oxygen consumption, and ejection fraction, did not differ significantly between the groups.7 In the Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure (COMPANION) trial, a survival benefit of CRT over pharmacologic therapy was observed in patients with LBBB but not in those with other QRS morphologies.4 Together, these data raise the concern that patients with non-LBBB morphologies somehow fare worse than their LBBB counterparts after CRT, possibly contributing to a significant nonresponse rate. The few retrospective studies that have attempted to clarify the prognostic impact of QRS morphology on survival after CRT have yielded mixed results.8,9 In a subset of potential CRT recipients who have previously undergone pacemaker or defibrillator placement, the underlying QRS morphology is a paced LBBB. In such patients, particularly those with AV nodal disease, chronic RV pacing can result in interventricular and intraventricular dyssynchrony in the short term and in adverse biventricular remodeling that predisposes to HF or atrial fibrillation in the long term.10 –13 Although upgrade to CRT has been recommended as a class IIa indication in these patients who otherwise meet CRT criteria, this recommendation is based largely on consensus and only on limited clinical data.14,15 The aim of this single-center, retrospective study was to assess the impact of native QRS morphologies or previous RV pacing on clinical and survival outcomes of de novo or upgrade CRT recipients.
Methods Study design This cohort consists of 505 consecutive patients who underwent de novo (n ⫽ 338; 39 CRT and 299 CRT-D) or upgrade CRT (n ⫽ 167; 10 CRT and 157 CRT-D) at Mayo Clinic from January 1, 2002, to December 31, 2006. Indications for CRT were HF symptoms despite optimal medical therapy (NYHA class II or greater), left ventricular ejection fraction (LVEF) ⱕ35%, and QRS duration ⱖ120 ms with the exception of select patients with QRS ⬍120 ms. Patients with QRS ⬍120 ms were included based on the combined assessment of an HF cardiologist and electrophysiologist. Among the 34 patients with narrow QRS (⬍120 ms) who were included in the study, 9 had echocardiographic evidence of dyssynchrony, 14 had first-time indications for long-term ventricular pacing such as atrial fibrillation with slow ventricular rates or bradycardia, 3 had HF exacerbated by previous RV pacing, and 8 had HF symptoms with a borderline QRS duration. The study was approved by the Mayo Institutional Review Board. Only patients who consented to the use of their records for research were included in the study.
Baseline evaluation Baseline evaluation was performed in all patients and included NYHA functional class, HF etiology, concomitant cardiovascular conditions (including hypertension, coronary artery disease, and diabetes), creatinine and hemoglobin
Heart Rhythm, Vol 6, No 10, October 2009 levels, ECG, and transthoracic echocardiogram. In addition, in the 167 patients with preexisting pacemakers, histograms retrieved at device interrogation were used to determine the percentage of ventricular pacing before upgrading to CRT.
Electrocardiography Patients were characterized according to native QRS morphologies on their pre-CRT ECGs. Standard criteria for LBBB included QRS duration ⱖ120 ms, QS or rS in lead V1, and broad R waves in lead I or V6. Criteria for RBBB included QRS duration ⱖ120 ms, qR or rSR= in lead V1, and wide S waves in leads I, V5 , and V6. Intraventricular conduction delay (IVCD) was defined as QRS ⱖ120 ms without typical LBBB or RBBB. Narrow QRS was defined as QRS ⬍120 ms. Among 338 patients undergoing de novo CRT, 67% had intrinsic LBBB (n ⫽ 228), 11% had RBBB (n ⫽ 36), 13% had IVCD (n ⫽ 43), and 9% had QRS ⬍120 ms (n ⫽ 30). The presence of concomitant left anterior or posterior fascicular blocks was noted in RBBB patients only.
Echocardiography Echocardiographic assessment included (1) LVEF; (2) RV systolic pressure; (3) mitral regurgitation severity (0 ⫽ none or trivial, 1 ⫽ mild, 2 ⫽ moderate, 3 ⫽ severe); (4) qualitative RV enlargement and systolic dysfunction (for both, 0 ⫽ normal, 1 ⫽ mild, 2 ⫽ moderate, 3 ⫽ severe) as determined by two-dimensional ultrasound and tissue Doppler velocity, with RV myocardial performance index performed if indicated; and (5) LV end-diastolic volume (LVEDV) and LV end-systolic volume (LVESV) calculated by Teichholz formula. Echocardiographic dyssynchrony studies in patients with QRS ⬍120 ms were performed as indicated. Intraventricular dyssynchrony criteria were met if either the maximum difference in time to peak systolic tissue velocity of any two basal segments was ⱖ100 ms or the standard deviation in time to peak tissue velocity was ⱖ34 ms.16
CRT therapy and follow-up During CRT implantation, coronary sinus venography was routine. Standard CRT settings included AV delay of 100 ms (sensed) and 130 ms (paced) with an interventricular delay of 0 ms in DDD or DDDR mode and lower (50 – 60 bpm) and upper (120 –130 bpm) pacing rates.17 Patients were recommended to have follow-up 3 to 6 months after device implant. The decision to pursue echocardiographic optimization or device reprogramming was discretionary based on CRT response. Clinical and echocardiographic follow-up was available in 379 (75%) and 361 patients (71%), respectively. Clinical “responder” status was assessed across QRS morphologies on the basis of improvement in NYHA class by 0.5 or LVEF ⱖ5%. Survival status as of May 15, 2007, was obtained using a national death and location database (Accurint).
Statistical analysis Continuous variables are expressed as mean ⫾ SD. Twosample t-tests or Wilcoxon rank sum tests were used to
Wokhlu et al
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
Results
assess the differences in variables between the upgrade and de novo groups. Comparisons across multiple groups were performed using analysis of variance (ANOVA) or KruskalWallis test when the data were skewed. Categorical variables are expressed as count (percent), and differences across the groups were assessed using Chi-square tests. Fisher exact test was used in some cases where sample sizes were small. Survival estimates were calculated by the Kaplan-Meier method. A log rank test was used to compare survival between groups. Univariate predictors of survival were identified using Cox proportional hazards regression methods. All variables with univariate significance (P ⬍.05) and ⱖ95% complete data were considered for a multivariate model. Relative risks are expressed as hazard ratio (HR) with 95% confidence interval (CI). Proportional hazards assumptions were tested using Schoenfeld residuals. For all variables, there was insufficient evidence (e.g., QRS morphology) or only slight evidence to reject the assumptions, so no modifications were made to the final model. Analyses were performed using SAS version 9.1.3 (SAS Insitute, Cary, NC, USA). Two-sided P ⬍.05 was considered significant. Table 1
1441
Patient characteristics Baseline characteristics in the de novo (n ⫽ 338) and upgrade groups (n ⫽ 167) are summarized in Table 1. The de novo group was younger, had less chronic atrial fibrillation, had less prolonged QRS duration, and had lower creatinine levels than the upgrade group. NYHA class, echocardiographic parameters, and pharmacologic HF regimens were similar (⬎80% on beta blockers, ⬎60% digoxin), except for the more frequent use of angiotensin-converting enzyme inhibitor or angiotensin receptor blocker in the de novo group (86% vs 77%; P ⫽ .014). Among the 338 de novo CRT recipients, baseline characteristics were comparable across QRS morphologies with the exceptions noted in Table 2. RV enlargement and dysfunction as well as left atrial enlargement were worse in patients with underlying RBBB or QRS ⬍120 ms. Programmed AV delays and LV lead site were comparable across groups. Among 167 patients with a preexisting device (54.5% defibrillator, 45.5% pacemaker), the average percentage of
Baseline characteristics in de novo and upgrade CRT groups
Characteristic Clinical and Laboratory Male Age at implant (years) Ischemic cardiomyopathy Atrial fibrillation NYHA class QRS duration (ms) Creatinine (mg/dL) Hemoglobin (mg/dL) Echocardiographic LV ejection fraction (%) LV end-diastolic volume (mL) LV end-systolic volume (mL) RV enlargement (0–3) RV dysfunction (0–3) Left atrial size (mm) Mitral regurgitation (0–3) RV systolic pressure (mmHg) Pharmacologic Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker Beta blocker Digoxin Device Lead location Lateral or posterior Anterolateral Middle cardiac Anterior interventricular Other Sensed AVD Paced AVD Biventricular pacing (%)
De novo (n ⫽ 338)
Upgrade (n ⫽ 167)
P value
253 (75) 67.7 ⫾ 11.8 204 (60%) 87 (26%) 3.0 ⫾ 0.5 158 ⫾ 31 1.5 ⫾ 0.6 12.9 ⫾ 1.8
146 (87) 70.1 ⫾ 10.3 110 (66%) 69 (41%) 3.0 ⫾ 0.5 184 ⫾ 32 1.6 ⫾ 0.6 12.8 ⫾ 1.8
.001 .026 .229 ⬍.001 .995 ⬍.001 .022 .409
23.1 235.6 171.7 0.8 1.0 59.0 1.5 46.8
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.3 73.7 64.0 0.9 1.0 9.8 0.8 15.0
23.2 225.0 167.5 1.1 1.1 61.1 1.6 46.9
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.2 69.6 62.2 1.0 1.0 10.2 0.9 16.0
.849 .154 .636 .018 .273 .055 .329 .925
282 (86%)
123 (77%)
.014
284 (87%) 207 (63%)
133 (84%) 104 (65%)
.476 .620
169 (50%) 93 (28%) 18 (5%) 50 (15%) 8 (2%) 100 (100–100) 130 (130–130) 92.7 ⫾ 14.5
77 (46%) 49 (29%) 11 (7%) 28 (17%) 2 (1%) 100 (100–100) 130 (130–130) 93.6 ⫾ 12.0
.410 .668 .567 .564 .509 .697 .681 .644
Values are given as number (percent) or mean ⫾ SD, unless otherwise indicated. AVD ⫽ atrioventricular delay [median (interquartile range)]; CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RV ⫽ right ventricular.
1442 Table 2
Heart Rhythm, Vol 6, No 10, October 2009 Baseline characteristics in de novo group based on QRS morphology
Characteristic Clinical and Laboratory Male Age at implant (years) Ischemic cardiomyopathy Atrial fibrillation NYHA class (1–4) QRS duration (ms) Creatinine (mg/dL) Hemoglobin (mg/dL) Echocardiographic LV ejection fraction (%) LV end-diastolic volume (mL) LV end-systolic volume (mL) RV enlargement (0–3) RV dysfunction (0–3) Left atrial size (mm) Mitral regurgitation (0–3) RV systolic pressure (mmHg) Pharmacologic Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker Beta blocker Digoxin Device Lead location Lateral or posterior Anterolateral Middle cardiac Anterior interventricular Other Biventricular pacing (%)
LBBB (N ⫽ 228)
RBBB (N ⫽ 36)
IVCD (N ⫽ 43)
QRS ⬍120 ms (N ⫽ 30)
P value
160 (70%) 68.7 ⫾ 11.6 134 (59%) 55 (24%) 3.0 ⫾ 0.5 168.9 ⫾ 25.3 1.5 ⫾ 0.6 12.9 ⫾ 1.7
29 (81%) 63.8 ⫾ 12.7* 20 (56%) 11 (31%) 3.0 ⫾ 0.4 168.9 ⫾ 21.2 1.4 ⫾ 0.3 13.0 ⫾ 2.0
35 (81%) 67.7 ⫾ 12.1 31 (72%) 9 (21%) 3.0 ⫾ 0.4 129.2 ⫾ 12.1* 1.4 ⫾ 0.4 12.7 ⫾ 1.8
28 (93%)* 64.4 ⫾ 10.7 18 (60%) 12 (40%) 3.0 ⫾ 0.7 104.7 ⫾ 19.2* 1.6 ⫾ 0.7 13.4 ⫾ 1.8
.022 .047 .383 .217 .895 ⬍.001 .305 .307
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
.311 .761 .550 ⬍.001 ⬍.001 .006 .641 .475
22.8 238.6 176.8 0.7 0.8 57.7 1.5 46.5
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.2 77.3 69.5 0.8 0.9 9.6 0.8 15.1
25.0 230.1 157.1 1.2 1.6 62.1 1.4 48.1
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.1 59.4 54.8 1.0* 1.1* 8.9* 0.9 13.0
22.8 231.5 170.1 0.7 1.0 59.9 1.6 49.7
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
6.2 62.6 48.8 0.8 1.1 8.9 0.8 14.7
24.1 225.3 159.6 1.4 1.5 64.1 1.4 44.0
10.0 76.1 60.1 1.0* 0.9* 10.7* 0.7 16.1
187 (85%)
32 (91%)
39 (93%)
23 (79%)
.285
193 (87%) 142 (64%)
30 (86%) 20 (57%)
37 (88%) 25 (60%)
23 (79%) 20 (69%)
.676 .725
119 (52%) 61 (27%) 9 (4%) 32 (14%) 7 (3%) 94.0 ⫾ 12.7
10 (28%) 14 (39%) 4 (11%) 8 (22%) 0 (0%) 87.0 ⫾ 22.4
24 (56%) 12 (28%) 3 (7%) 3 (7%) 1 (2%) 89.8 ⫾ 17.8
15 (50%) 6 (20%) 2 (7%) 7 (23%) 0 (0%) 92.5 ⫾ 12.0
.043 .355 .293 .139 .623 .168
Values are given as number (percent) or mean ⫾ SD, unless otherwise indicated. IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RBBB ⫽ bundle branch block; RV ⫽ right ventricular. *P ⬍.05 vs LBBB.
RV pacing was ⬍40% in 42 (25%) patients, 40%– 80% in 35 (21%) patients, and ⬎80% in 90 (54%) patients. The only baseline characteristic that differed significantly among patients with different degrees of RV pacing was left atrial size. Patients with ⬎80% RV pacing had a left atrial diameter of 64.0 ⫾ 9.5 mm compared to 57.0 ⫾ 10.5 mm in patients with 40%– 80% pacing and 58.9 ⫾ 9.9 mm in those with ⬍40% pacing (P ⫽ .006).
Improvement in heart failure Patients were paced in both ventricles 93% ⫾ 14% of the time. Outcomes in de novo and upgrade CRT groups are shown in Figure 1. Median time between implant and echocardiographic follow-up was 7.1 months (range 4.6 –12.6 months). No significant differences in follow-up times were observed in the upgrade versus de novo subgroup or among QRS morphology subgroups. No clinically significant differences were noted in patients with and those without echocardiographic follow-up. Improvements in NYHA class, LVEF, LVEDV, LVESV, mitral regurgitation, and RV systolic pressure after CRT were significant for both de novo and upgrade CRT groups. Mean improvement in NYHA
classification in patients with de novo versus upgrade CRT was the same (0.6 ⫾ 0.8; P ⫽ .690). Mean improvement in LVEF also was similar in the de novo and upgrade groups (6.6 ⫾ 10.0 vs 7.7 ⫾ 10.7; P ⫽ .308). LVESV had similar improvements in the de novo and upgrade groups (–21.9 ⫾ 44.9 vs –28.1 ⫾ 45.7; P ⫽ .454). Differences in improvement in LVEDV between de novo and upgrade groups did not achieve significance (–23.9 ⫾ 48.8 vs –12.7 ⫾ 49.2; P ⫽ .065).
Impact of previous pacing on CRT outcome Table 3 compares HF outcomes after upgrade CRT based on the percentage of previous RV pacing in the upgrade group. Although improvements in LVEF, LVEDV, and LVESV appeared more substantial in patients who were previously ⬎80% RV paced compared to patients in the 40%– 80% and ⬍40% paced groups, these differences were not significant.
Effect of QRS morphology HF outcomes across QRS morphologies in the de novo group are summarized in Figure 2. Compared to patients with LBBB (0.7 ⫾ 0.8), NYHA improvements were signif-
Wokhlu et al
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
1443
Figure 1 Heart failure outcomes before and after de novo and upgrade cardiac resynchronization therapy (CRT): New York Heart Association (NYHA) classification (A), left ventricular (LV) ejection fraction (B), LV end-diastolic volume (C), and mitral regurgitation severity (D).
icantly less in patients with RBBB (0.3 ⫾ 0.8; P ⫽ .014) or IVCD (0.2 ⫾ 0.7; P ⫽ .001). LVEF improvements also were significantly less in patients with RBBB (3.1 ⫾ 8.5 %; P ⫽ .039) or IVCD (2.7 ⫾ 7.3%; P ⫽ .010) compared to those with intrinsic LBBB (7.9 ⫾ 10.2%). Whereas improvements in LVEF were less in patients with RBBB and IVCD compared to those with LBBB, improvements in LVEDV differed substantially only between IVCD and LBBB. The differences between QRS morphology shown in Figure 2 were similar even after multivariate adjustment for baseline differences. Interestingly, RV enlargement overall worsened in patients with IVCD (0.5 ⫾ 0.7) in comparison to patients with LBBB (0.0 ⫾ 0.7; P ⫽ .004), RBBB (⫺0.1 ⫾ 0.6), or QRS ⬍120 ms (⫺0.2 ⫾ 1.0). Improvements in RV
Table 3
dysfunction, reductions in mitral regurgitation, and RV systolic pressure were not substantially different across QRS configurations. NYHA “responders” were among 69% of patients with LBBB, 61% with QRS ⬍120 ms (P ⫽ .438), 42% with RBBB (P ⫽ .008), and 33% with IVCD (P ⬍.001). LVEF “responders” were among 55% patients with LBBB, 52% with QRS ⬍120 ms (P ⫽ .818), 39% with IVCD (P ⫽ .122) and 38% with RBBB (P ⫽ .143). Patients with RBBB were additionally classified as those with RBBB but no additional left-sided conduction disease (n ⫽ 26) or those with additional left anterior or posterior fascicular block (n ⫽ 10). Four (36%) patients with RBBB alone were NYHA responders compared to 6 (67%) patients with mixed conduc-
Upgrade CRT outcomes stratified by percentage of previous pacing
Outcome (Post–Pre CRT)
⬍40% (N ⫽ 42)
40%– 80% (N ⫽ 35)
⬎80% (N ⫽ 90)
P value
⌬NYHA Biventricular pacing after CRT (%) ⌬LV ejection fraction (%) ⌬LV end-diastolic volume (mL) ⌬RV systolic pressure (mmHg) ⌬Mitral regurgitation (0–3) ⌬RV enlargement (0–3) ⌬RV dysfunction (0–3)
⫺0.6 90.8 5.4 1.5 ⫺3.5 ⫺0.3 ⫺0.1 0
⫺0.5 89.3 5.6 ⫺13.8 ⫺2.1 ⫺0.1 ⫺0.1 0
⫺0.6 95.7 9.5 ⫺18.7 ⫺3.1 ⫺0.3 0 ⫺0.1
.700 .031 .126 .245 .952 .557 .913 .892
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
0.9 12.8 9.2 37.7 14.0 0.4 0.8 0.8
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
0.7 16.0 9.9 38.7 11.0 0.7 0.8 0.4
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
0.9 10.2 11.3 55.9 15.4 0.7 0.9 0.9
CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RV ⫽ right ventricular.
1444
Heart Rhythm, Vol 6, No 10, October 2009
Figure 2 Heart failure outcomes in de novo patients based on QRS morphology. Positive changes demonstrate improvement in New York Heart Association (NYHA) classification (A), increase in left ventricular (LV) ejection fraction (B), increase in LV end-diastolic volume (C), and reduction in right ventricular (RV) systolic pressure (D). IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; RBBB ⫽ right bundle branch block.
tion disease (P ⫽ .370), and 4 (36%) patients with pure RBBB were LVEF responders compared to 3 (43%) patients with mixed conduction disease (P ⬎.999).
Survival outcomes There were 87 deaths in the de novo group and 41 deaths in the upgrade group over a median follow-up period of 2.6 years (interquartile range 1.6 – 4.0 years). Survival was similar between the two groups (61% de novo, 63% upgrade at 4 years; P ⫽ .906; Figure 3A). Within the upgrade group, no survival difference was noted based on the percent of previous pacing (70% in ⬍40% paced, 61% in 40%– 80%
paced, and 60% in ⬎80% paced at 4 years; P ⫽ .940). The number of deaths observed in each pacing group over the entire follow-up period was 10 in the ⬍40% group, 11 in the 40%– 80% group, and 20 in the ⬎80% group. Within the de novo group, there were 51 deaths in 228 patients with intrinsic LBBB, 16 deaths in 36 patients with RBBB, 12 deaths in 43 patients with IVCD, and 7 deaths in 30 patients with QRS ⬍120 ms. Four-year survival in patients with QRS ⬍120 ms (66%) was comparable to survival in patients with intrinsic LBBB (66%; P ⫽ .392; Figure 3B). However, survival in patients with RBBB (32%) was significantly lower than in patients with LBBB (P ⬍.001).
Figure 3 Survival after de novo versus upgrade cardiac resynchronization therapy (CRT) (A) and the impact of QRS morphology on survival after de novo CRT (B). IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; RBBB ⫽ right bundle branch block.
Wokhlu et al Table 4
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
1445
Univariate and multivariate mortality predictors after CRT Univariate
Characteristic Clinical and Laboratory Male Age at implant (per 10 years) Ischemic cardiomyopathy NYHA class Chronic atrial fibrillation Creatinine (per 1 mg/dL) Hemoglobin (per 1 mg/dL) CRT-P vs CRT-D implant Biventricular pacing (per 10%) Upgrade CRT (vs de novo) QRS Morphology LBBB RBBB IVCD QRS ⬍120 ms Paced Echocardiographic LV ejection fraction (per 10%) RV dysfunction (0–3) Mitral regurgitation (0–3) RV systolic pressure (per 10 mmHg increase) LV end-diastolic volume (per 50 mL) Pharmacologic Angiotensin-converting enzyme inhibitor/ angiotensin receptor blocker Beta blocker Digoxin Lead Location Lateral or posterior All other positions
Multivariate
Hazard ratio (CI)
P value
Hazard ratio (CI)
P value
1.7 1.4 2.4 1.6 1.1 1.9 0.7 1.0 1.1 1.0
(1.0–2.8) (1.2–1.7) (1.6–3.6) (1.1–2.2) (0.8–1.6) (1.6–2.3) (0.6–0.8) (0.6–1.7) (0.9–1.4) (0.7–1.5)
.040 ⬍.001 ⬍.001 .011 .599 ⬍.001 ⬍.001 .916 .403 .906
1.1 1.3 1.6 1.3 — 1.4 0.7 — — —
.678 .017 .035 .168 — .032 ⬍.001 — — —
1.0 2.3 1.5 1.5 1.2
(1.3–4.1) (0.8–2.6) (0.7–2.9) (0.8–1.9)
Reference .003 .196 .291 .346
1.0 3.6 1.7 1.3 1.3
0.9 1.3 1.3 1.0 1.0
(0.7–1.2) (1.1–1.6) (1.0–1.7) (1.0–1.0) (1.0–1.0)
.439 .006 .029 .002 .003
— * * * *
— — — —
1.1 (0.7–1.8)
.733
—
—
0.7 (0.5–1.1) 1.4 (0.9–2.1)
.126 .100
— —
1.0 1.7 (1.2–2.4)
Reference .005
— — — 1.0 1.2 (0.8–1.8)
(0.6–2.0) (1.0–1.5) (1.0–2.6) (0.9–1.9) (1.1–1.8) (0.7–0.8)
(2.0–6.8) (0.9–3.2) (0.6–2.8) (0.8–2.1)
Reference ⬍.001 .091 .520 .277
Reference .310
CI ⫽ confidence interval; CRT ⫽ cardiac resynchronization therapy; IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RBBB ⫽ bundle branch block; RV ⫽ right ventricular. *Not included; evaluated in separate multivariate analyses.
There was a trend toward lower 4-year survival in patients with IVCD as well (57%; P ⫽ .104).
Mortality predictors Table 4 lists the mortality predictors identified at baseline. Univariate predictors of mortality included male gender, age, NYHA classification, higher creatinine level, lower hemoglobin level, ischemic cardiomyopathy, nonlateral LV lead position, RV dysfunction, mitral regurgitation, larger LVEDV, higher RV systolic pressure, and RBBB. Narrow QRS, IVCD, and paced QRS morphologies (regardless of percentage of previous RV pacing) were not significant univariate predictors of differential survival from LBBB. A multivariate model of significant univariates except echocardiographic variables is given in Table 4. RBBB was found to be an independent predictor of worsened survival outcome compared to LBBB (HR 3.6, CI 1.9 – 6.6; P ⬍.001), in addition to age at implant (HR 1.3 per 10 years, CI 1.0 –1.5; P ⫽ .037), creatinine (HR 1.4, CI 1.1– 1.8; P ⫽ .018), hemoglobin (HR 0.7, CI 0.7– 0.8; P ⬍.001), and ischemic cardiomyopathy (HR 1.6, CI 1.0 –2.5; P ⫽ .045).
Of note, inclusion of multiple echocardiographic features within the multivariate analysis was precluded by the number of events in patients with complete datasets. Therefore, the effects of individual echocardiographic variables were investigated in separate multivariate models controlling for age, ischemic cardiomyopathy, creatinine level, hemoglobin level, and QRS morphology. In these models, only RV dysfunction (HR 1.3, CI 1.0 –1.6; P ⫽ .027) and mitral regurgitation (HR 1.3, CI 1.0 –1.7; P ⫽ .036) were significant, whereas RBBB retained the same degree of significance.
Discussion There were two main findings in this study of CRT in a clinical practice population. First, no evidence of a difference in outcomes between the upgrade and de novo implantation groups was observed. Second, native QRS morphology had clinical and prognostic value in recipients of de novo CRT. Patients with RBBB or nonspecific IVCD experienced less clinical improvement and less LV functional improvement. In addition, patients with RBBB experienced less survival benefit than did those with intrinsic LBBB. In
1446
Heart Rhythm, Vol 6, No 10, October 2009
contrast, select patients with narrow QRS ⬍120 ms benefited from CRT. These results provide information about CRT outcomes in populations that have been underrepresented in clinical trials.
to CRT despite LV functional improvements. In our study, greater RV dysfunction was observed in the RBBB population and independently predicted decreased survival benefit after CRT.
Comparable outcomes in paced and intrinsic LBBB
Impact of intraventricular conduction delay versus narrow QRS
We found that CRT upgrade was beneficial in patients with a previously implanted pacemaker. Although small crossover trials have suggested that patients with previous RV pacing and HF benefit from CRT, the magnitude of this effect is not fully understood. In fact, in previously RV paced patients, greater clinical response to CRT might have been expected due to the potential for reversibility in RV pacing–induced HF. In our study, retrospective identification of patients with pure RV pacing–induced HF was confounded by the high frequency of cardiac comorbidities in patients with a device. However, we were able to compare outcomes across upgrade patients based on degree of RV pacing and found no significant differences. Our study corroborates reports of comparable clinical outcomes in paced and intrinsic LBBB from smaller studies10,11,15,18,19 and provides new data demonstrating a common survival benefit. In contrast, a retrospective study of CRT in patients with previously paced versus intrinsic LBBB reported a trend toward decreased survival in previously RV–paced patients.20 These disparate survival outcomes may be due to population differences. Unlike the aforementioned study, our upgrade patients had varying degrees of RV pacing as opposed to ⬎90% pacing and experienced overall improved rather than worsened symptom and echocardiographic outcomes.
Less favorable clinical outcomes in RBBB Although de novo versus upgrade CRT implantation was not prognostically significant, QRS morphology was significant. Survival was markedly lower in CRT recipients with underlying RBBB than in those with LBBB, even after multivariate adjustments. There are a variety of potential explanations for this observation. First, the survival difference after CRT may simply reflect survival differences between these two HF subgroups. However, the impact of the interaction between QRS morphology and HF on our outcomes remains ill-defined given the inconsistent survival outcomes in RBBB versus LBBB across large HF cohorts.21–23 Second, some patients with RBBB may have dyssynchrony patterns that are ill-suited to CRT.23 For example, the RV rather than the LV may contract late, whereas the LV lateral free wall contracts earlier than the septum.11 Third, patients with RBBB may have sicker conduction systems than do routine CRT candidates with LBBB. In patients in whom RBBB is a late manifestation of extensive left- and right-sided conduction disease, the conduction system may be too diseased to respond to CRT. Our study contained too few patients with concomitant left- and right-sided conduction disease by ECG to test this possibility. Finally, RBBB may be a marker of RV dysfunction, which if severely symptomatic could limit clinical response
Similar to patients with RBBB, those with IVCD lacked the robust clinical response to CRT demonstrated by patients with LBBB and demonstrated a trend toward decreased survival compared to LBBB patients. Interestingly, in contrast, select patients with QRS ⬍120 ms had better clinical response to CRT than did patients with IVCD and had a survival that was comparable to that in patients with LBBB. These data add to the findings of a recent study of narrow QRS patients who were randomized to CRT versus medical therapy.24 Narrow QRS patients undergoing CRT improved echocardiographically and by symptoms from baseline, but the degree of improvement did not differ substantially from the medical therapy arm. One assimilation of these findings is that the improvement in our narrow QRS group represents the nonspecific effect of concomitant medical management. Alternatively, the select nature of our narrow QRS CRT recipients may provide a better explanation for outcomes similar to LBBB patients. Unlike the IVCD subgroup in this study, the subgroup with QRS ⬍120 ms either met dyssynchrony criteria or had other compelling circumstances such as a need for chronic RV pacing or other pacemaker indications to be offered CRT. Although no single echocardiographic parameter or combination thereof has been sufficiently validated for use in CRT selection,25 the finding of improved outcomes in the narrow QRS subgroup but not in the IVCD subgroup supports the need for further study of alternative criteria to QRS duration in the selection of CRT candidates.
Study limitations These data are limited by the retrospective nature of the study. Ideally, information about changes to CRT programming parameters during follow-up, quantitative assessments of functional status, and hospitalization history would provide complementary data about HF outcomes in patients with different QRS morphologies.
Conclusion Despite these limitations, our data identify the relative prognostic values of device upgrade and intrinsic QRS morphology in CRT in a clinical practice population. We established that CRT upgrade is comparable to, but not better than, de novo implantation in clinical and survival outcomes. We further identified less clinical and/or survival benefit for patients with RBBB and IVCD. Patients with widened QRS without LBBB appear less likely to respond to CRT and, therefore, may be better served by more aggressive medical management.
Wokhlu et al
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
References 1. Abraham WT, Fisher WG, Smith AL, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346:1845–1853. 2. Young JB, Abraham WT, Smith AL, et al. Combined cardiac resynchronization and implantable cardioversion defibrillation in advanced chronic heart failure: the MIRACLE ICD Trial. JAMA 2003;289:2685–2694. 3. Cleland JG, Daubert JC, Erdmann E, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: 1539 –1549. 4. Bristow MR, Saxon LA, Boehmer J, et al. Cardiac resynchronization therapy with or without an implantable defibrillator in advanced heart failure. N Engl J Med 2004;350:2140 –2150. 5. Lozano I, Bocchiardo M, Achtelik M, et al. Impact of biventricular pacing on mortality in a randomized crossover study of patients with heart failure and ventricular arrhythmias. Pacing Clin Electrophysiol 2000;23(11 Pt 2):1711– 1712. 6. Reynolds MR, Joventino LP, Josephson ME. Relationship of baseline electrocardiographic characteristics with the response to cardiac resynchronization therapy for heart failure. Pacing Clin Electrophysiol 2004;27:1513–1518. 7. Egoavil CA, Ho RT, Greenspon AJ, Pavri BB. Cardiac resynchronization therapy in patients with right bundle branch block: analysis of pooled data from the MIRACLE and CONTAK-CD trials. Heart Rhythm 2005;2:611– 615. 8. Iler MA, Hu T, Ayyagari S, et al. Prognostic value of electrocardiographic measurements before and after cardiac resynchronization device implantation in patients with heart failure due to ischemic or nonischemic cardiomyopathy. Am J Cardiol 2008;101:359 –363. 9. Adelstein EC, Saba S. Usefulness of baseline electrocardiographic QRS complex pattern to predict response to cardiac resynchronization. Am J Cardiol 2009;103:238 –242. 10. Eldadah ZA, Rosen B, Hay I, et al. The benefit of upgrading chronically right ventricle-paced heart failure patients to resynchronization therapy demonstrated by strain rate imaging. Heart Rhythm 2006;3:435– 442. 11. Byrne MJ, Helm RH, Daya S, et al. Diminished left ventricular dyssynchrony and impact of resynchronization in failing hearts with right versus left bundle branch block. J Am Coll Cardiol 2007;50:1484 –1490. 12. Wilkoff BL, Cook JR, Epstein AE, et al. Dual-chamber pacing or ventricular backup pacing in patients with an implantable defibrillator: the Dual Chamber and VVI Implantable Defibrillator (DAVID) Trial. JAMA 2002;288:3115–3123. 13. Sweeney MO, Hellkamp AS, Ellenbogen KA, et al. Adverse effect of ventricular pacing on heart failure and atrial fibrillation among patients with normal baseline
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
1447
QRS duration in a clinical trial of pacemaker therapy for sinus node dysfunction. Circulation 2003;107:2932–2937. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities. J Am Coll Cardiol 2008;51:e1– e62. Duray GZ, Israel CW, Pajitnev D, Hohnloser SH. Upgrading to biventricular pacing/defibrillation systems in right ventricular paced congestive heart failure patients: prospective assessment of procedural parameters and response rate. Europace 2008;10:48 –52. Yu CM, Gorscan J, Bleeker GB, et al. Usefulness of tissue Doppler velocity and strain dyssynchrony for predicting left ventricular reverse remodeling response after cardiac resynchronization therapy. Am J Cardiol 2007;100:1263–1270. Cha YM, Rea RF, Wang M, et al. Response to cardiac resynchronization therapy predicts survival in heart failure: a single-center experience. J Cardiovasc Electrophysiol 2007;18:1015–1019. Leclercq C, Cazeau S, Lellouche D, et al. Upgrading from single chamber right ventricular to biventricular pacing in permanently paced patients with worsening heart failure: the RD-CHF study. Pacing Clin Electrophysiol 2007;30:S23–S30. Witte K, Pipes R, Nanthakumar K, Parker JD. Biventricular pacemaker upgrade in previously paced heart failure patients—improvements in ventricular dyssynchrony. J Card Fail 2006;12:199 –204. Adelstein EC, Saba S. Usefulness of baseline electrocardiographic QRS complex pattern to predict response to cardiac resynchronization. Am J Cardiol 2009;103:238 –242. Baldasseroni S, Gentile A, Gorini M, et al. Intraventricular conduction defects in patients with congestive heart failure: left bundle not right bundle branch block is an independent predictor poor prognosis. A report from the Italian Network on Congestive Heart Failure (IN-CHF database). Ital Heart J 2003;4: 607– 613. Barsheshet A, Leor J, Goldbourt U, et al. Effect of bundle branch block patterns on mortality in hospitalized patients with heart failure. Am J Cardiol 2008;101: 1303–1308. Mueller C, Laule-Killian K, Klima T, et al. Right bundle branch block and long-term mortality in patients with acute congestive heart failure. J Intern Med 2006;260:421– 428. Garrigue S, Reuter S, Labeque JN, et al. Usefulness of biventricular pacing in patients with congestive heart failure and right bundle branch block. Am J Cardiol 2001;88:1436 –1441. Beshai JF, Grimm RA, Nagueh SF, et al. Cardiac resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med 2007;357:2461–2471.
Introduction Cardiac resynchronization therapy (CRT) has been shown to improve outcomes in trials in which most patients with drug-refractory heart failure (HF) had intrinsic left bundle branch block (LBBB).1– 4 However, the benefits of CRT have not been conclusively demonstrated in patients with Dr. Wokhlu has received travel support from Medtronic, Boston Scientific, and St. Jude Medical to attend educational seminars. Dr. Hayes is an educational speaker and serves on the advisory boards for Medtronic, Boston Scientific, Sorin Medical Group, and St. Jude Medical, and he is a member of the steering committee for St. Jude Medical. Dr. Asirvatham receives honoraria and is on the speaker’s board for St. Jude Medical, Boston Scientific, and Medtronic. In addition, he is a co-patent holder for an alternative resynchronization therapy technique. Ms. Webster is on an advisory board for Boston Scientific. Address reprint requests and correspondence: Dr. Yong-Mei Cha, Division of Cardiovascular Diseases, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905. E-mail address: [email protected]. (Received March 29, 2009; accepted July 6, 2009.)
⫽ .906). No clinical or survival differences were noted in upgrade patients based on the percentage of previous pacing. However, survival in de novo CRT recipients with RBBB (32%) was lower than in those with LBBB (66%; P ⬍.001), and RBBB independently predicted death (hazard ratio 3.5, confidence interval 1.9 – 6.5; P ⬍.001). CONCLUSION RBBB and IVCD result in less clinical improvement or worsened survival after CRT. Additional selection criteria may be beneficial in identifying potential responders with RBBB, IVCD, or narrow QRS. KEYWORDS Biventricular pacing; Cardiac resynchronization therapy; Left bundle branch block; Right bundle branch block; QRS morphology; Upgrade; Ventricular pacing ABBREVIATIONS CI ⫽ confidence interval; CRT ⫽ cardiac resynchronization therapy; HF ⫽ heart failure; HR ⫽ hazard ratio; IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; LV ⫽ left ventricular; LVEDV ⫽ left ventricular end-diastolic volume; LVEF ⫽ left ventricular ejection fraction; LVESV ⫽ left ventricular end-systolic volume; RBBB ⫽ right bundle branch block; RV ⫽ right ventricular (Heart Rhythm 2009;6:1439 –1447) © 2009 Heart Rhythm Society. All rights reserved.
other native QRS morphologies or those who have undergone right ventricular (RV) pacing in the setting of a preexisting pacemaker or defibrillator due to limited representation within randomized controlled trials and variable outcomes. The impact of QRS morphology on HF outcomes has been examined in four randomized trials of CRT.1,2,4,5 In a substudy of CRT-D recipients from the MIRACLE (Multicenter InSync Randomized Clinical Evaluation) ICD study, RBBB was associated with less improvement in peak oxygen consumption than other QRS morphologies in univariate, but not multivariate, analysis.6 In a retrospective, pooled analysis from MIRACLE and CONTAK-CD that compared outcomes in 34 patients with RBBB who received CRT and 27 who received medical management, New York Heart Association (NYHA) functional class was the only clinical outcome that improved significantly more after active CRT. Improvements in other HF parameters, such as
1547-5271/$ -see front matter © 2009 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2009.07.009
1440 6-minute walk, peak oxygen consumption, and ejection fraction, did not differ significantly between the groups.7 In the Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure (COMPANION) trial, a survival benefit of CRT over pharmacologic therapy was observed in patients with LBBB but not in those with other QRS morphologies.4 Together, these data raise the concern that patients with non-LBBB morphologies somehow fare worse than their LBBB counterparts after CRT, possibly contributing to a significant nonresponse rate. The few retrospective studies that have attempted to clarify the prognostic impact of QRS morphology on survival after CRT have yielded mixed results.8,9 In a subset of potential CRT recipients who have previously undergone pacemaker or defibrillator placement, the underlying QRS morphology is a paced LBBB. In such patients, particularly those with AV nodal disease, chronic RV pacing can result in interventricular and intraventricular dyssynchrony in the short term and in adverse biventricular remodeling that predisposes to HF or atrial fibrillation in the long term.10 –13 Although upgrade to CRT has been recommended as a class IIa indication in these patients who otherwise meet CRT criteria, this recommendation is based largely on consensus and only on limited clinical data.14,15 The aim of this single-center, retrospective study was to assess the impact of native QRS morphologies or previous RV pacing on clinical and survival outcomes of de novo or upgrade CRT recipients.
Methods Study design This cohort consists of 505 consecutive patients who underwent de novo (n ⫽ 338; 39 CRT and 299 CRT-D) or upgrade CRT (n ⫽ 167; 10 CRT and 157 CRT-D) at Mayo Clinic from January 1, 2002, to December 31, 2006. Indications for CRT were HF symptoms despite optimal medical therapy (NYHA class II or greater), left ventricular ejection fraction (LVEF) ⱕ35%, and QRS duration ⱖ120 ms with the exception of select patients with QRS ⬍120 ms. Patients with QRS ⬍120 ms were included based on the combined assessment of an HF cardiologist and electrophysiologist. Among the 34 patients with narrow QRS (⬍120 ms) who were included in the study, 9 had echocardiographic evidence of dyssynchrony, 14 had first-time indications for long-term ventricular pacing such as atrial fibrillation with slow ventricular rates or bradycardia, 3 had HF exacerbated by previous RV pacing, and 8 had HF symptoms with a borderline QRS duration. The study was approved by the Mayo Institutional Review Board. Only patients who consented to the use of their records for research were included in the study.
Baseline evaluation Baseline evaluation was performed in all patients and included NYHA functional class, HF etiology, concomitant cardiovascular conditions (including hypertension, coronary artery disease, and diabetes), creatinine and hemoglobin
Heart Rhythm, Vol 6, No 10, October 2009 levels, ECG, and transthoracic echocardiogram. In addition, in the 167 patients with preexisting pacemakers, histograms retrieved at device interrogation were used to determine the percentage of ventricular pacing before upgrading to CRT.
Electrocardiography Patients were characterized according to native QRS morphologies on their pre-CRT ECGs. Standard criteria for LBBB included QRS duration ⱖ120 ms, QS or rS in lead V1, and broad R waves in lead I or V6. Criteria for RBBB included QRS duration ⱖ120 ms, qR or rSR= in lead V1, and wide S waves in leads I, V5 , and V6. Intraventricular conduction delay (IVCD) was defined as QRS ⱖ120 ms without typical LBBB or RBBB. Narrow QRS was defined as QRS ⬍120 ms. Among 338 patients undergoing de novo CRT, 67% had intrinsic LBBB (n ⫽ 228), 11% had RBBB (n ⫽ 36), 13% had IVCD (n ⫽ 43), and 9% had QRS ⬍120 ms (n ⫽ 30). The presence of concomitant left anterior or posterior fascicular blocks was noted in RBBB patients only.
Echocardiography Echocardiographic assessment included (1) LVEF; (2) RV systolic pressure; (3) mitral regurgitation severity (0 ⫽ none or trivial, 1 ⫽ mild, 2 ⫽ moderate, 3 ⫽ severe); (4) qualitative RV enlargement and systolic dysfunction (for both, 0 ⫽ normal, 1 ⫽ mild, 2 ⫽ moderate, 3 ⫽ severe) as determined by two-dimensional ultrasound and tissue Doppler velocity, with RV myocardial performance index performed if indicated; and (5) LV end-diastolic volume (LVEDV) and LV end-systolic volume (LVESV) calculated by Teichholz formula. Echocardiographic dyssynchrony studies in patients with QRS ⬍120 ms were performed as indicated. Intraventricular dyssynchrony criteria were met if either the maximum difference in time to peak systolic tissue velocity of any two basal segments was ⱖ100 ms or the standard deviation in time to peak tissue velocity was ⱖ34 ms.16
CRT therapy and follow-up During CRT implantation, coronary sinus venography was routine. Standard CRT settings included AV delay of 100 ms (sensed) and 130 ms (paced) with an interventricular delay of 0 ms in DDD or DDDR mode and lower (50 – 60 bpm) and upper (120 –130 bpm) pacing rates.17 Patients were recommended to have follow-up 3 to 6 months after device implant. The decision to pursue echocardiographic optimization or device reprogramming was discretionary based on CRT response. Clinical and echocardiographic follow-up was available in 379 (75%) and 361 patients (71%), respectively. Clinical “responder” status was assessed across QRS morphologies on the basis of improvement in NYHA class by 0.5 or LVEF ⱖ5%. Survival status as of May 15, 2007, was obtained using a national death and location database (Accurint).
Statistical analysis Continuous variables are expressed as mean ⫾ SD. Twosample t-tests or Wilcoxon rank sum tests were used to
Wokhlu et al
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
Results
assess the differences in variables between the upgrade and de novo groups. Comparisons across multiple groups were performed using analysis of variance (ANOVA) or KruskalWallis test when the data were skewed. Categorical variables are expressed as count (percent), and differences across the groups were assessed using Chi-square tests. Fisher exact test was used in some cases where sample sizes were small. Survival estimates were calculated by the Kaplan-Meier method. A log rank test was used to compare survival between groups. Univariate predictors of survival were identified using Cox proportional hazards regression methods. All variables with univariate significance (P ⬍.05) and ⱖ95% complete data were considered for a multivariate model. Relative risks are expressed as hazard ratio (HR) with 95% confidence interval (CI). Proportional hazards assumptions were tested using Schoenfeld residuals. For all variables, there was insufficient evidence (e.g., QRS morphology) or only slight evidence to reject the assumptions, so no modifications were made to the final model. Analyses were performed using SAS version 9.1.3 (SAS Insitute, Cary, NC, USA). Two-sided P ⬍.05 was considered significant. Table 1
1441
Patient characteristics Baseline characteristics in the de novo (n ⫽ 338) and upgrade groups (n ⫽ 167) are summarized in Table 1. The de novo group was younger, had less chronic atrial fibrillation, had less prolonged QRS duration, and had lower creatinine levels than the upgrade group. NYHA class, echocardiographic parameters, and pharmacologic HF regimens were similar (⬎80% on beta blockers, ⬎60% digoxin), except for the more frequent use of angiotensin-converting enzyme inhibitor or angiotensin receptor blocker in the de novo group (86% vs 77%; P ⫽ .014). Among the 338 de novo CRT recipients, baseline characteristics were comparable across QRS morphologies with the exceptions noted in Table 2. RV enlargement and dysfunction as well as left atrial enlargement were worse in patients with underlying RBBB or QRS ⬍120 ms. Programmed AV delays and LV lead site were comparable across groups. Among 167 patients with a preexisting device (54.5% defibrillator, 45.5% pacemaker), the average percentage of
Baseline characteristics in de novo and upgrade CRT groups
Characteristic Clinical and Laboratory Male Age at implant (years) Ischemic cardiomyopathy Atrial fibrillation NYHA class QRS duration (ms) Creatinine (mg/dL) Hemoglobin (mg/dL) Echocardiographic LV ejection fraction (%) LV end-diastolic volume (mL) LV end-systolic volume (mL) RV enlargement (0–3) RV dysfunction (0–3) Left atrial size (mm) Mitral regurgitation (0–3) RV systolic pressure (mmHg) Pharmacologic Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker Beta blocker Digoxin Device Lead location Lateral or posterior Anterolateral Middle cardiac Anterior interventricular Other Sensed AVD Paced AVD Biventricular pacing (%)
De novo (n ⫽ 338)
Upgrade (n ⫽ 167)
P value
253 (75) 67.7 ⫾ 11.8 204 (60%) 87 (26%) 3.0 ⫾ 0.5 158 ⫾ 31 1.5 ⫾ 0.6 12.9 ⫾ 1.8
146 (87) 70.1 ⫾ 10.3 110 (66%) 69 (41%) 3.0 ⫾ 0.5 184 ⫾ 32 1.6 ⫾ 0.6 12.8 ⫾ 1.8
.001 .026 .229 ⬍.001 .995 ⬍.001 .022 .409
23.1 235.6 171.7 0.8 1.0 59.0 1.5 46.8
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.3 73.7 64.0 0.9 1.0 9.8 0.8 15.0
23.2 225.0 167.5 1.1 1.1 61.1 1.6 46.9
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.2 69.6 62.2 1.0 1.0 10.2 0.9 16.0
.849 .154 .636 .018 .273 .055 .329 .925
282 (86%)
123 (77%)
.014
284 (87%) 207 (63%)
133 (84%) 104 (65%)
.476 .620
169 (50%) 93 (28%) 18 (5%) 50 (15%) 8 (2%) 100 (100–100) 130 (130–130) 92.7 ⫾ 14.5
77 (46%) 49 (29%) 11 (7%) 28 (17%) 2 (1%) 100 (100–100) 130 (130–130) 93.6 ⫾ 12.0
.410 .668 .567 .564 .509 .697 .681 .644
Values are given as number (percent) or mean ⫾ SD, unless otherwise indicated. AVD ⫽ atrioventricular delay [median (interquartile range)]; CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RV ⫽ right ventricular.
1442 Table 2
Heart Rhythm, Vol 6, No 10, October 2009 Baseline characteristics in de novo group based on QRS morphology
Characteristic Clinical and Laboratory Male Age at implant (years) Ischemic cardiomyopathy Atrial fibrillation NYHA class (1–4) QRS duration (ms) Creatinine (mg/dL) Hemoglobin (mg/dL) Echocardiographic LV ejection fraction (%) LV end-diastolic volume (mL) LV end-systolic volume (mL) RV enlargement (0–3) RV dysfunction (0–3) Left atrial size (mm) Mitral regurgitation (0–3) RV systolic pressure (mmHg) Pharmacologic Angiotensin-converting enzyme inhibitor or angiotensin receptor blocker Beta blocker Digoxin Device Lead location Lateral or posterior Anterolateral Middle cardiac Anterior interventricular Other Biventricular pacing (%)
LBBB (N ⫽ 228)
RBBB (N ⫽ 36)
IVCD (N ⫽ 43)
QRS ⬍120 ms (N ⫽ 30)
P value
160 (70%) 68.7 ⫾ 11.6 134 (59%) 55 (24%) 3.0 ⫾ 0.5 168.9 ⫾ 25.3 1.5 ⫾ 0.6 12.9 ⫾ 1.7
29 (81%) 63.8 ⫾ 12.7* 20 (56%) 11 (31%) 3.0 ⫾ 0.4 168.9 ⫾ 21.2 1.4 ⫾ 0.3 13.0 ⫾ 2.0
35 (81%) 67.7 ⫾ 12.1 31 (72%) 9 (21%) 3.0 ⫾ 0.4 129.2 ⫾ 12.1* 1.4 ⫾ 0.4 12.7 ⫾ 1.8
28 (93%)* 64.4 ⫾ 10.7 18 (60%) 12 (40%) 3.0 ⫾ 0.7 104.7 ⫾ 19.2* 1.6 ⫾ 0.7 13.4 ⫾ 1.8
.022 .047 .383 .217 .895 ⬍.001 .305 .307
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
.311 .761 .550 ⬍.001 ⬍.001 .006 .641 .475
22.8 238.6 176.8 0.7 0.8 57.7 1.5 46.5
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.2 77.3 69.5 0.8 0.9 9.6 0.8 15.1
25.0 230.1 157.1 1.2 1.6 62.1 1.4 48.1
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
7.1 59.4 54.8 1.0* 1.1* 8.9* 0.9 13.0
22.8 231.5 170.1 0.7 1.0 59.9 1.6 49.7
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
6.2 62.6 48.8 0.8 1.1 8.9 0.8 14.7
24.1 225.3 159.6 1.4 1.5 64.1 1.4 44.0
10.0 76.1 60.1 1.0* 0.9* 10.7* 0.7 16.1
187 (85%)
32 (91%)
39 (93%)
23 (79%)
.285
193 (87%) 142 (64%)
30 (86%) 20 (57%)
37 (88%) 25 (60%)
23 (79%) 20 (69%)
.676 .725
119 (52%) 61 (27%) 9 (4%) 32 (14%) 7 (3%) 94.0 ⫾ 12.7
10 (28%) 14 (39%) 4 (11%) 8 (22%) 0 (0%) 87.0 ⫾ 22.4
24 (56%) 12 (28%) 3 (7%) 3 (7%) 1 (2%) 89.8 ⫾ 17.8
15 (50%) 6 (20%) 2 (7%) 7 (23%) 0 (0%) 92.5 ⫾ 12.0
.043 .355 .293 .139 .623 .168
Values are given as number (percent) or mean ⫾ SD, unless otherwise indicated. IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RBBB ⫽ bundle branch block; RV ⫽ right ventricular. *P ⬍.05 vs LBBB.
RV pacing was ⬍40% in 42 (25%) patients, 40%– 80% in 35 (21%) patients, and ⬎80% in 90 (54%) patients. The only baseline characteristic that differed significantly among patients with different degrees of RV pacing was left atrial size. Patients with ⬎80% RV pacing had a left atrial diameter of 64.0 ⫾ 9.5 mm compared to 57.0 ⫾ 10.5 mm in patients with 40%– 80% pacing and 58.9 ⫾ 9.9 mm in those with ⬍40% pacing (P ⫽ .006).
Improvement in heart failure Patients were paced in both ventricles 93% ⫾ 14% of the time. Outcomes in de novo and upgrade CRT groups are shown in Figure 1. Median time between implant and echocardiographic follow-up was 7.1 months (range 4.6 –12.6 months). No significant differences in follow-up times were observed in the upgrade versus de novo subgroup or among QRS morphology subgroups. No clinically significant differences were noted in patients with and those without echocardiographic follow-up. Improvements in NYHA class, LVEF, LVEDV, LVESV, mitral regurgitation, and RV systolic pressure after CRT were significant for both de novo and upgrade CRT groups. Mean improvement in NYHA
classification in patients with de novo versus upgrade CRT was the same (0.6 ⫾ 0.8; P ⫽ .690). Mean improvement in LVEF also was similar in the de novo and upgrade groups (6.6 ⫾ 10.0 vs 7.7 ⫾ 10.7; P ⫽ .308). LVESV had similar improvements in the de novo and upgrade groups (–21.9 ⫾ 44.9 vs –28.1 ⫾ 45.7; P ⫽ .454). Differences in improvement in LVEDV between de novo and upgrade groups did not achieve significance (–23.9 ⫾ 48.8 vs –12.7 ⫾ 49.2; P ⫽ .065).
Impact of previous pacing on CRT outcome Table 3 compares HF outcomes after upgrade CRT based on the percentage of previous RV pacing in the upgrade group. Although improvements in LVEF, LVEDV, and LVESV appeared more substantial in patients who were previously ⬎80% RV paced compared to patients in the 40%– 80% and ⬍40% paced groups, these differences were not significant.
Effect of QRS morphology HF outcomes across QRS morphologies in the de novo group are summarized in Figure 2. Compared to patients with LBBB (0.7 ⫾ 0.8), NYHA improvements were signif-
Wokhlu et al
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
1443
Figure 1 Heart failure outcomes before and after de novo and upgrade cardiac resynchronization therapy (CRT): New York Heart Association (NYHA) classification (A), left ventricular (LV) ejection fraction (B), LV end-diastolic volume (C), and mitral regurgitation severity (D).
icantly less in patients with RBBB (0.3 ⫾ 0.8; P ⫽ .014) or IVCD (0.2 ⫾ 0.7; P ⫽ .001). LVEF improvements also were significantly less in patients with RBBB (3.1 ⫾ 8.5 %; P ⫽ .039) or IVCD (2.7 ⫾ 7.3%; P ⫽ .010) compared to those with intrinsic LBBB (7.9 ⫾ 10.2%). Whereas improvements in LVEF were less in patients with RBBB and IVCD compared to those with LBBB, improvements in LVEDV differed substantially only between IVCD and LBBB. The differences between QRS morphology shown in Figure 2 were similar even after multivariate adjustment for baseline differences. Interestingly, RV enlargement overall worsened in patients with IVCD (0.5 ⫾ 0.7) in comparison to patients with LBBB (0.0 ⫾ 0.7; P ⫽ .004), RBBB (⫺0.1 ⫾ 0.6), or QRS ⬍120 ms (⫺0.2 ⫾ 1.0). Improvements in RV
Table 3
dysfunction, reductions in mitral regurgitation, and RV systolic pressure were not substantially different across QRS configurations. NYHA “responders” were among 69% of patients with LBBB, 61% with QRS ⬍120 ms (P ⫽ .438), 42% with RBBB (P ⫽ .008), and 33% with IVCD (P ⬍.001). LVEF “responders” were among 55% patients with LBBB, 52% with QRS ⬍120 ms (P ⫽ .818), 39% with IVCD (P ⫽ .122) and 38% with RBBB (P ⫽ .143). Patients with RBBB were additionally classified as those with RBBB but no additional left-sided conduction disease (n ⫽ 26) or those with additional left anterior or posterior fascicular block (n ⫽ 10). Four (36%) patients with RBBB alone were NYHA responders compared to 6 (67%) patients with mixed conduc-
Upgrade CRT outcomes stratified by percentage of previous pacing
Outcome (Post–Pre CRT)
⬍40% (N ⫽ 42)
40%– 80% (N ⫽ 35)
⬎80% (N ⫽ 90)
P value
⌬NYHA Biventricular pacing after CRT (%) ⌬LV ejection fraction (%) ⌬LV end-diastolic volume (mL) ⌬RV systolic pressure (mmHg) ⌬Mitral regurgitation (0–3) ⌬RV enlargement (0–3) ⌬RV dysfunction (0–3)
⫺0.6 90.8 5.4 1.5 ⫺3.5 ⫺0.3 ⫺0.1 0
⫺0.5 89.3 5.6 ⫺13.8 ⫺2.1 ⫺0.1 ⫺0.1 0
⫺0.6 95.7 9.5 ⫺18.7 ⫺3.1 ⫺0.3 0 ⫺0.1
.700 .031 .126 .245 .952 .557 .913 .892
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
0.9 12.8 9.2 37.7 14.0 0.4 0.8 0.8
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
0.7 16.0 9.9 38.7 11.0 0.7 0.8 0.4
⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾ ⫾
0.9 10.2 11.3 55.9 15.4 0.7 0.9 0.9
CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RV ⫽ right ventricular.
1444
Heart Rhythm, Vol 6, No 10, October 2009
Figure 2 Heart failure outcomes in de novo patients based on QRS morphology. Positive changes demonstrate improvement in New York Heart Association (NYHA) classification (A), increase in left ventricular (LV) ejection fraction (B), increase in LV end-diastolic volume (C), and reduction in right ventricular (RV) systolic pressure (D). IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; RBBB ⫽ right bundle branch block.
tion disease (P ⫽ .370), and 4 (36%) patients with pure RBBB were LVEF responders compared to 3 (43%) patients with mixed conduction disease (P ⬎.999).
Survival outcomes There were 87 deaths in the de novo group and 41 deaths in the upgrade group over a median follow-up period of 2.6 years (interquartile range 1.6 – 4.0 years). Survival was similar between the two groups (61% de novo, 63% upgrade at 4 years; P ⫽ .906; Figure 3A). Within the upgrade group, no survival difference was noted based on the percent of previous pacing (70% in ⬍40% paced, 61% in 40%– 80%
paced, and 60% in ⬎80% paced at 4 years; P ⫽ .940). The number of deaths observed in each pacing group over the entire follow-up period was 10 in the ⬍40% group, 11 in the 40%– 80% group, and 20 in the ⬎80% group. Within the de novo group, there were 51 deaths in 228 patients with intrinsic LBBB, 16 deaths in 36 patients with RBBB, 12 deaths in 43 patients with IVCD, and 7 deaths in 30 patients with QRS ⬍120 ms. Four-year survival in patients with QRS ⬍120 ms (66%) was comparable to survival in patients with intrinsic LBBB (66%; P ⫽ .392; Figure 3B). However, survival in patients with RBBB (32%) was significantly lower than in patients with LBBB (P ⬍.001).
Figure 3 Survival after de novo versus upgrade cardiac resynchronization therapy (CRT) (A) and the impact of QRS morphology on survival after de novo CRT (B). IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; RBBB ⫽ right bundle branch block.
Wokhlu et al Table 4
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
1445
Univariate and multivariate mortality predictors after CRT Univariate
Characteristic Clinical and Laboratory Male Age at implant (per 10 years) Ischemic cardiomyopathy NYHA class Chronic atrial fibrillation Creatinine (per 1 mg/dL) Hemoglobin (per 1 mg/dL) CRT-P vs CRT-D implant Biventricular pacing (per 10%) Upgrade CRT (vs de novo) QRS Morphology LBBB RBBB IVCD QRS ⬍120 ms Paced Echocardiographic LV ejection fraction (per 10%) RV dysfunction (0–3) Mitral regurgitation (0–3) RV systolic pressure (per 10 mmHg increase) LV end-diastolic volume (per 50 mL) Pharmacologic Angiotensin-converting enzyme inhibitor/ angiotensin receptor blocker Beta blocker Digoxin Lead Location Lateral or posterior All other positions
Multivariate
Hazard ratio (CI)
P value
Hazard ratio (CI)
P value
1.7 1.4 2.4 1.6 1.1 1.9 0.7 1.0 1.1 1.0
(1.0–2.8) (1.2–1.7) (1.6–3.6) (1.1–2.2) (0.8–1.6) (1.6–2.3) (0.6–0.8) (0.6–1.7) (0.9–1.4) (0.7–1.5)
.040 ⬍.001 ⬍.001 .011 .599 ⬍.001 ⬍.001 .916 .403 .906
1.1 1.3 1.6 1.3 — 1.4 0.7 — — —
.678 .017 .035 .168 — .032 ⬍.001 — — —
1.0 2.3 1.5 1.5 1.2
(1.3–4.1) (0.8–2.6) (0.7–2.9) (0.8–1.9)
Reference .003 .196 .291 .346
1.0 3.6 1.7 1.3 1.3
0.9 1.3 1.3 1.0 1.0
(0.7–1.2) (1.1–1.6) (1.0–1.7) (1.0–1.0) (1.0–1.0)
.439 .006 .029 .002 .003
— * * * *
— — — —
1.1 (0.7–1.8)
.733
—
—
0.7 (0.5–1.1) 1.4 (0.9–2.1)
.126 .100
— —
1.0 1.7 (1.2–2.4)
Reference .005
— — — 1.0 1.2 (0.8–1.8)
(0.6–2.0) (1.0–1.5) (1.0–2.6) (0.9–1.9) (1.1–1.8) (0.7–0.8)
(2.0–6.8) (0.9–3.2) (0.6–2.8) (0.8–2.1)
Reference ⬍.001 .091 .520 .277
Reference .310
CI ⫽ confidence interval; CRT ⫽ cardiac resynchronization therapy; IVCD ⫽ intraventricular conduction delay; LBBB ⫽ left bundle branch block; LV ⫽ left ventricular; NYHA ⫽ New York Heart Association; RBBB ⫽ bundle branch block; RV ⫽ right ventricular. *Not included; evaluated in separate multivariate analyses.
There was a trend toward lower 4-year survival in patients with IVCD as well (57%; P ⫽ .104).
Mortality predictors Table 4 lists the mortality predictors identified at baseline. Univariate predictors of mortality included male gender, age, NYHA classification, higher creatinine level, lower hemoglobin level, ischemic cardiomyopathy, nonlateral LV lead position, RV dysfunction, mitral regurgitation, larger LVEDV, higher RV systolic pressure, and RBBB. Narrow QRS, IVCD, and paced QRS morphologies (regardless of percentage of previous RV pacing) were not significant univariate predictors of differential survival from LBBB. A multivariate model of significant univariates except echocardiographic variables is given in Table 4. RBBB was found to be an independent predictor of worsened survival outcome compared to LBBB (HR 3.6, CI 1.9 – 6.6; P ⬍.001), in addition to age at implant (HR 1.3 per 10 years, CI 1.0 –1.5; P ⫽ .037), creatinine (HR 1.4, CI 1.1– 1.8; P ⫽ .018), hemoglobin (HR 0.7, CI 0.7– 0.8; P ⬍.001), and ischemic cardiomyopathy (HR 1.6, CI 1.0 –2.5; P ⫽ .045).
Of note, inclusion of multiple echocardiographic features within the multivariate analysis was precluded by the number of events in patients with complete datasets. Therefore, the effects of individual echocardiographic variables were investigated in separate multivariate models controlling for age, ischemic cardiomyopathy, creatinine level, hemoglobin level, and QRS morphology. In these models, only RV dysfunction (HR 1.3, CI 1.0 –1.6; P ⫽ .027) and mitral regurgitation (HR 1.3, CI 1.0 –1.7; P ⫽ .036) were significant, whereas RBBB retained the same degree of significance.
Discussion There were two main findings in this study of CRT in a clinical practice population. First, no evidence of a difference in outcomes between the upgrade and de novo implantation groups was observed. Second, native QRS morphology had clinical and prognostic value in recipients of de novo CRT. Patients with RBBB or nonspecific IVCD experienced less clinical improvement and less LV functional improvement. In addition, patients with RBBB experienced less survival benefit than did those with intrinsic LBBB. In
1446
Heart Rhythm, Vol 6, No 10, October 2009
contrast, select patients with narrow QRS ⬍120 ms benefited from CRT. These results provide information about CRT outcomes in populations that have been underrepresented in clinical trials.
to CRT despite LV functional improvements. In our study, greater RV dysfunction was observed in the RBBB population and independently predicted decreased survival benefit after CRT.
Comparable outcomes in paced and intrinsic LBBB
Impact of intraventricular conduction delay versus narrow QRS
We found that CRT upgrade was beneficial in patients with a previously implanted pacemaker. Although small crossover trials have suggested that patients with previous RV pacing and HF benefit from CRT, the magnitude of this effect is not fully understood. In fact, in previously RV paced patients, greater clinical response to CRT might have been expected due to the potential for reversibility in RV pacing–induced HF. In our study, retrospective identification of patients with pure RV pacing–induced HF was confounded by the high frequency of cardiac comorbidities in patients with a device. However, we were able to compare outcomes across upgrade patients based on degree of RV pacing and found no significant differences. Our study corroborates reports of comparable clinical outcomes in paced and intrinsic LBBB from smaller studies10,11,15,18,19 and provides new data demonstrating a common survival benefit. In contrast, a retrospective study of CRT in patients with previously paced versus intrinsic LBBB reported a trend toward decreased survival in previously RV–paced patients.20 These disparate survival outcomes may be due to population differences. Unlike the aforementioned study, our upgrade patients had varying degrees of RV pacing as opposed to ⬎90% pacing and experienced overall improved rather than worsened symptom and echocardiographic outcomes.
Less favorable clinical outcomes in RBBB Although de novo versus upgrade CRT implantation was not prognostically significant, QRS morphology was significant. Survival was markedly lower in CRT recipients with underlying RBBB than in those with LBBB, even after multivariate adjustments. There are a variety of potential explanations for this observation. First, the survival difference after CRT may simply reflect survival differences between these two HF subgroups. However, the impact of the interaction between QRS morphology and HF on our outcomes remains ill-defined given the inconsistent survival outcomes in RBBB versus LBBB across large HF cohorts.21–23 Second, some patients with RBBB may have dyssynchrony patterns that are ill-suited to CRT.23 For example, the RV rather than the LV may contract late, whereas the LV lateral free wall contracts earlier than the septum.11 Third, patients with RBBB may have sicker conduction systems than do routine CRT candidates with LBBB. In patients in whom RBBB is a late manifestation of extensive left- and right-sided conduction disease, the conduction system may be too diseased to respond to CRT. Our study contained too few patients with concomitant left- and right-sided conduction disease by ECG to test this possibility. Finally, RBBB may be a marker of RV dysfunction, which if severely symptomatic could limit clinical response
Similar to patients with RBBB, those with IVCD lacked the robust clinical response to CRT demonstrated by patients with LBBB and demonstrated a trend toward decreased survival compared to LBBB patients. Interestingly, in contrast, select patients with QRS ⬍120 ms had better clinical response to CRT than did patients with IVCD and had a survival that was comparable to that in patients with LBBB. These data add to the findings of a recent study of narrow QRS patients who were randomized to CRT versus medical therapy.24 Narrow QRS patients undergoing CRT improved echocardiographically and by symptoms from baseline, but the degree of improvement did not differ substantially from the medical therapy arm. One assimilation of these findings is that the improvement in our narrow QRS group represents the nonspecific effect of concomitant medical management. Alternatively, the select nature of our narrow QRS CRT recipients may provide a better explanation for outcomes similar to LBBB patients. Unlike the IVCD subgroup in this study, the subgroup with QRS ⬍120 ms either met dyssynchrony criteria or had other compelling circumstances such as a need for chronic RV pacing or other pacemaker indications to be offered CRT. Although no single echocardiographic parameter or combination thereof has been sufficiently validated for use in CRT selection,25 the finding of improved outcomes in the narrow QRS subgroup but not in the IVCD subgroup supports the need for further study of alternative criteria to QRS duration in the selection of CRT candidates.
Study limitations These data are limited by the retrospective nature of the study. Ideally, information about changes to CRT programming parameters during follow-up, quantitative assessments of functional status, and hospitalization history would provide complementary data about HF outcomes in patients with different QRS morphologies.
Conclusion Despite these limitations, our data identify the relative prognostic values of device upgrade and intrinsic QRS morphology in CRT in a clinical practice population. We established that CRT upgrade is comparable to, but not better than, de novo implantation in clinical and survival outcomes. We further identified less clinical and/or survival benefit for patients with RBBB and IVCD. Patients with widened QRS without LBBB appear less likely to respond to CRT and, therefore, may be better served by more aggressive medical management.
Wokhlu et al
Impact of Paced or Intrinsic QRS Morphology on CRT Survival
References 1. Abraham WT, Fisher WG, Smith AL, et al. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346:1845–1853. 2. Young JB, Abraham WT, Smith AL, et al. Combined cardiac resynchronization and implantable cardioversion defibrillation in advanced chronic heart failure: the MIRACLE ICD Trial. JAMA 2003;289:2685–2694. 3. Cleland JG, Daubert JC, Erdmann E, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: 1539 –1549. 4. Bristow MR, Saxon LA, Boehmer J, et al. Cardiac resynchronization therapy with or without an implantable defibrillator in advanced heart failure. N Engl J Med 2004;350:2140 –2150. 5. Lozano I, Bocchiardo M, Achtelik M, et al. Impact of biventricular pacing on mortality in a randomized crossover study of patients with heart failure and ventricular arrhythmias. Pacing Clin Electrophysiol 2000;23(11 Pt 2):1711– 1712. 6. Reynolds MR, Joventino LP, Josephson ME. Relationship of baseline electrocardiographic characteristics with the response to cardiac resynchronization therapy for heart failure. Pacing Clin Electrophysiol 2004;27:1513–1518. 7. Egoavil CA, Ho RT, Greenspon AJ, Pavri BB. Cardiac resynchronization therapy in patients with right bundle branch block: analysis of pooled data from the MIRACLE and CONTAK-CD trials. Heart Rhythm 2005;2:611– 615. 8. Iler MA, Hu T, Ayyagari S, et al. Prognostic value of electrocardiographic measurements before and after cardiac resynchronization device implantation in patients with heart failure due to ischemic or nonischemic cardiomyopathy. Am J Cardiol 2008;101:359 –363. 9. Adelstein EC, Saba S. Usefulness of baseline electrocardiographic QRS complex pattern to predict response to cardiac resynchronization. Am J Cardiol 2009;103:238 –242. 10. Eldadah ZA, Rosen B, Hay I, et al. The benefit of upgrading chronically right ventricle-paced heart failure patients to resynchronization therapy demonstrated by strain rate imaging. Heart Rhythm 2006;3:435– 442. 11. Byrne MJ, Helm RH, Daya S, et al. Diminished left ventricular dyssynchrony and impact of resynchronization in failing hearts with right versus left bundle branch block. J Am Coll Cardiol 2007;50:1484 –1490. 12. Wilkoff BL, Cook JR, Epstein AE, et al. Dual-chamber pacing or ventricular backup pacing in patients with an implantable defibrillator: the Dual Chamber and VVI Implantable Defibrillator (DAVID) Trial. JAMA 2002;288:3115–3123. 13. Sweeney MO, Hellkamp AS, Ellenbogen KA, et al. Adverse effect of ventricular pacing on heart failure and atrial fibrillation among patients with normal baseline
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
1447
QRS duration in a clinical trial of pacemaker therapy for sinus node dysfunction. Circulation 2003;107:2932–2937. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities. J Am Coll Cardiol 2008;51:e1– e62. Duray GZ, Israel CW, Pajitnev D, Hohnloser SH. Upgrading to biventricular pacing/defibrillation systems in right ventricular paced congestive heart failure patients: prospective assessment of procedural parameters and response rate. Europace 2008;10:48 –52. Yu CM, Gorscan J, Bleeker GB, et al. Usefulness of tissue Doppler velocity and strain dyssynchrony for predicting left ventricular reverse remodeling response after cardiac resynchronization therapy. Am J Cardiol 2007;100:1263–1270. Cha YM, Rea RF, Wang M, et al. Response to cardiac resynchronization therapy predicts survival in heart failure: a single-center experience. J Cardiovasc Electrophysiol 2007;18:1015–1019. Leclercq C, Cazeau S, Lellouche D, et al. Upgrading from single chamber right ventricular to biventricular pacing in permanently paced patients with worsening heart failure: the RD-CHF study. Pacing Clin Electrophysiol 2007;30:S23–S30. Witte K, Pipes R, Nanthakumar K, Parker JD. Biventricular pacemaker upgrade in previously paced heart failure patients—improvements in ventricular dyssynchrony. J Card Fail 2006;12:199 –204. Adelstein EC, Saba S. Usefulness of baseline electrocardiographic QRS complex pattern to predict response to cardiac resynchronization. Am J Cardiol 2009;103:238 –242. Baldasseroni S, Gentile A, Gorini M, et al. Intraventricular conduction defects in patients with congestive heart failure: left bundle not right bundle branch block is an independent predictor poor prognosis. A report from the Italian Network on Congestive Heart Failure (IN-CHF database). Ital Heart J 2003;4: 607– 613. Barsheshet A, Leor J, Goldbourt U, et al. Effect of bundle branch block patterns on mortality in hospitalized patients with heart failure. Am J Cardiol 2008;101: 1303–1308. Mueller C, Laule-Killian K, Klima T, et al. Right bundle branch block and long-term mortality in patients with acute congestive heart failure. J Intern Med 2006;260:421– 428. Garrigue S, Reuter S, Labeque JN, et al. Usefulness of biventricular pacing in patients with congestive heart failure and right bundle branch block. Am J Cardiol 2001;88:1436 –1441. Beshai JF, Grimm RA, Nagueh SF, et al. Cardiac resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med 2007;357:2461–2471.