Hemodynamic Impact and Outcome of Permanent Pacemaker Implantation Following Transcatheter Aortic Valve Implantation Simon Biner, MD, Yoav Michowitz, MD*, Eran Leshem-Rubinow, MD, Yan Topilsky, MD, Eyal Ben-Assa, MD, Jason Shimiaie, BS, Shmuel Banai, MD, Gad Keren, MD, Arie Steinvil, MD, and Ariel Finkelstein, MD Transcatheter aortic valve implantation (TAVI) frequently requires postprocedural permanent pacemaker (PPM) implantation. We evaluated clinical and hemodynamic impact of PPM after TAVI. Clinical and echocardiographic data were retrospectively analyzed in 230 consecutive patients who underwent TAVI and echocardiography at baseline and after 6 months. Echocardiographic parameters included left ventricular ejection fraction (LVEF), left ventricular (LV) stroke volume, early mitral velocity/annulus velocity ratio (E/e0 ), right ventricular index of myocardial performance, systolic pulmonary artery pressure (SPAP), and aortic, mitral, and tricuspid regurgitation grades. Clinical outcomes included 2-year survival and cardiovascular and PPM-related event-free survival. The Medtronic CoreValve and Edwards Sapien prosthesis were used in 201 and 29 patients, respectively. PPM was required in 58 patients (25.4%). Two-year and event-free survival rates were similar between patients with and without PPM. At 6 months, patients with PPM demonstrated attenuated improvement in LVEF (L0.9 – 8.7% vs 2.3 – 10.8%, respectively, p [ 0.03) and LV stroke volume (L2 – 16 vs 4 – 10 ml/m2, respectively, p [ 0.015), a trend toward smaller reduction in systolic pulmonary artery pressure (L1 – 12 vs L6 – 10 mm Hg, respectively, p [ 0.09), and deterioration of right ventricular index of myocardial performance (L3 – 17% vs 5 – 26%, respectively, p [ 0.05). The differences in post-TAVI aortic, mitral, and tricuspid regurgitation grades were insignificant. In conclusion, PPM implantation after TAVI is associated with reduced LVEF and impaired LV unloading. However, this unfavorable hemodynamic response does not affect the 2-year clinical outcome. The maintenance of clinical benefit appears to be driven by TAVI-related recovery of LV and right ventricular performance that mitigates unfavorable impact of PPM. Ó 2014 Elsevier Inc. All rights reserved. (Am J Cardiol 2014;113:132e137) The close anatomical relation between the aortic valve and the branching atrioventricular (AV) bundle may lead to conduction system abnormalities during transcatheter aortic valve implantation (TAVI).1 Up to 40% of patients undergoing TAVI, particularly those using a self-expanding prosthesis, require permanent pacemaker (PPM) implantation.2e6 Right ventricular (RV) pacing results in desynchronization of left ventricular (LV) contraction, which in turn may lead to LV remodeling and reduced LV ejection fraction (LVEF).7 Numerous randomized clinical trials have demonstrated deleterious effects of RV pacing including increased risk of atrial fibrillation, ventricular arrhythmia, and heart failure.8e11 These studies evaluated patients with primary conduction system abnormalities with or without LV systolic dysfunction unrelated to aortic stenosis and TAVI. Furthermore, TAVI-related left bundle branch block (LBBB) or need of PPM was shown to Department of Cardiology, Tel-Aviv Medical Center, Sackler Faculty of Medicine, Tel-Aviv University, Tel Aviv, Israel. Manuscript received July 13, 2013; revised manuscript received and accepted September 17, 2013. Drs. Biner and Michowitz contributed equally to this work. See page 136 for disclosure information. *Corresponding author: Tel: (þ972) 3 6973311; fax: (þ972) 3 6974428. E-mail address: [email protected] (Y. Michowitz). 0002-9149/13/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2013.09.030
adversely affect LV improvement.12 However, a recent study evaluating clinical outcomes among patients with or without PPM implantation after TAVI showed similar mortality rates in both groups, along with similar cardiac and cerebrovascular morbidity rates.13 The mechanism of maintenance of clinical benefit of TAVI despite adverse hemodynamic effect of PPM in this subgroup of patients is unclear. Therefore, the purpose of the present study was to evaluate the hemodynamic consequences of permanent pacing in patients undergoing TAVI and further define its relation to clinical outcome. Methods Our study population consisted of consecutive patients undergoing percutaneous transfemoral TAVI for treatment of symptomatic aortic valve stenosis using the Edwards Sapien/Sapien XT prosthesis (Edwards Lifesciences, Irvine, California) or the Medtronic CoreValve prosthesis (MCV; Medtronic, Minneapolis, Minnesota) at a tertiary university hospital. The institutional ethics committee approved the study. Inclusion criteria were logistic EuroSCORE >20, porcelain aorta, previous chest radiation, logistic EuroSCORE >10%, and 1 of the following: cachexia,14 need for assistance in basic daily activities,15 or previous open heart www.ajconline.org
Valvular Heart Disease/Hemodynamics and Outcome of PPM After TAVI
surgery. The Institutional Heart Team including an interventional cardiologist, echocardiologists, and a cardiothoracic surgeon determined eligibility for TAVI. Temporary pacemakers were implanted in all patients unless they were already carrying PPM. For the immediate 36 hours after procedure interval, the patients were monitored in the intensive care unit; the temporary pacemaker was extracted after 24 hours unless AV conduction disturbances were detected. The criteria for PPM implantation included pre-TAVI right bundle branch block (BBB), post-TAVI complete AV block (AVB), type II second-degree AV block, alternating BBB, or new-onset LBBB with PR-interval prolongation 280 ms. Patients undergoing TAVI have been enrolled in a prospective registry recording procedural information as well as clinical, echocardiographic, and pacemaker data collected at baseline 30 days, 6 months, 12 months, and then at regular 6- to 12-month intervals. Clinical end points were overall survival and cardiovascular event-free survival. Cardiovascular events included hospitalization for heart failure, stroke, myocardial infarction, new-onset atrial fibrillation, and AVB occurring after discharge. At each follow-up visit, we recorded pacemaker pacing mode, battery and electrode status, capture threshold, impedance and sensing as well as cumulative percentage of RV pacing. Devices were programed to minimize ventricular pacing by activating ventricular pacing suppression algorithms, increasing AV delay, or lowering pacing rate. Transthoracic echocardiographic data were obtained with a standardized protocol. LV linear dimensions were obtained on the parasternal long-axis view. LV mass was estimated based on LV modeling as a prolate ellipse.16 LV end-diastolic volume, LV end-systolic volume, and LVEF were calculated using the Simpson rule16 or Quinones method.17 LV stroke volume (LVSV) was calculated as previously described.18 Early (E) transmitral filling peak velocity was obtained with pulsed wave Doppler sample at the tip of the mitral valve leaflets.19 Early velocity of the septal and lateral aspects of the mitral annulus was measured using tissue Doppler imaging. Mitral annulus velocity (e0 ) was defined as the average of septal and lateral velocities.19 Left atrial volume was determined using the biplane arealength or dimension-length methods.20 RV index of myocardial performance (RIMP) was used for global estimates of RV systolic and diastolic functions. RIMP was defined as the ratio of isovolumic time and ejection time.20 RV ejection time was measured using pulsed wave Doppler of the RV outflow. The isovolumetric time was calculated as the difference between the duration of flow of the tricuspid regurgitant jet and the ejection time.21 Systolic pulmonary artery pressure (SPAP) and right atrial pressure were estimated as previously described.21 Echo Doppler indexes of aortic stenosis severity included the maximal velocity across the aortic valve, the mean pressure gradient, and the aortic valve area (AVA) calculated using the standard continuity equation.19 Severe aortic stenosis was defined as an AVA 40 mm Hg.18 Aortic, mitral, and tricuspid regurgitation grades were classified semiquantitatively on a scale of 0 to 4 according to established criteria.22 AVA, left atrial volume, LV enddiastolic volume, LV end-systolic volume, and LVSV were indexed by body surface area.
133
Table 1 Baseline demographic and clinical characteristics of entire cohort and patients with and without permanent pacemaker implantation Variable
Age (yrs) Men Height (cm) Weight (kg) Body surface area (m2) Diabetes mellitus Dyslipidemia Hypertension Peripheral vascular disease Past stroke Heart failure Previous coronary bypass grafting Atrial fibrillation Logistic EuroSCORE
Total (n ¼ 230), %
Pacemaker Implanted, %
p
No (n ¼ 172)
Yes (n ¼ 58)
83 5 38.3 162 9 72 15 1.64 0.4 32.6 78.3 84.8 11.3
83 6 34.9 162 9 71 15 1.62 0.4 30.2 79.7 86 12.8
83 5 48.3 163 10 75 15 1.7 0.4 39.7 74.1 81 9
0.91 0.086 0.61 0.15 0.18 0.198 0.462 0.399 0.337
7.8 34.8 18.3
7.6 37.8 19.8
8.6 25.9 13.8
0.781 0.112 0.432
18.7 26 14
19.8 26 13
15.5 26 13
0.562 0.9
Table 2 Baseline echo Doppler characteristics of entire cohort and patients with and without permanent pacemaker implantation Variable
Aortic valve Vmax (cm/s) Aortic valve mean gradient (mm Hg) AVA index (cm2/m2) Left atrial volume index (ml/m2) LV mass index (g/m2) LV end-diastolic volume index (ml/m2) LV end-systolic volume index (ml/m2) SV index (ml/m2) LVEF (%) E wave (cm/s) e0 (cm/s) E/e0 SPAP (mm Hg) Tricuspid regurgitation time (ms) RV ejection time (ms) RIMP Mitral regurgitation grade 2 (%) Tricuspid regurgitation grade 2 (%) Aortic regurgitation grade 2 (%)
Total
428 66 47 15
Pacemaker Implanted No
Yes
439 66 47 66
428 66 46 15
p
0.26 0.86
0.43 0.12 0.42 0.12 0.44 0.15 0.4 47 17 48 18 45 21 0.50 139 56 63 20
134 65 63 24
137 32 65 26
0.77 0.7
28 10
28 16
28 15
0.74
0.67 0.54 0.67 0.59 0.64 0.12 0.66
33 58 99 5.1 21 44 402
12 10 31 1.6 8 15 45
33 58 100 5 21 45 401
15 11 32 1.7 8 15 50
34 58 98 5.2 20 41 404
18 10 30 1.4 8 13 45
306 41 305 32 308 40 0.55 0.32 0.15 0.32 0.15 0.32 0.15 0.99 44 (19) 38 (22) 6 (10) 0.054 23 (10)
18 (10)
5 (9)
0.76
18 (8)
15 (9)
3 (5)
0.57
The implantation depth of the bioprosthesis was assessed using aortography immediately after TAVI. For each patient, the distance between lower edges of the noncoronary cusp and ventricular end of prosthesis frame were
134
The American Journal of Cardiology (www.ajconline.org)
Table 3 The changes between baseline and 6-month follow-up echo Doppler parameters in entire cohort and patients with and without permanent pacemaker implantation Variable
Aortic valve Vmax (cm/s) Aortic valve mean gradient (mm Hg) Left atrial volume index (ml/m2) LV mass index (g/m2) LV end-diastolic volume index (ml/m2) LV end-systolic volume index (ml/m2) SV index (ml/m2) LVEF (%) E wave (cm/s) e0 (cm/s) E/e0 SPAP (mm Hg) Tricuspid regurgitation time (ms) RV ejection time (ms) RIMP Mitral regurgitation grade reduced from 2 to 1 (%) Tricuspid regurgitation grade reduced from 2 to 1 (%) Aortic regurgitation grade 2 (%)
Total
Pacemaker Implanted
235 118 38 15 2 12 6 32 4 14 1 10 28 18 4.2 30 0.5 1.8 0.14 7.7 4 10 9 50 7 35 0.01 0.13 28 (56) 8 (35) 18 (8)
obtained twice. Implantation depth was defined as an average of these 2 measurements. Continuous normally distributed parameters are presented as the mean SD and were compared using the Student or paired t test, as appropriate. Ordinal and/or non-normally distributed data are presented by the median and the first and third quartiles and were compared using the Wilcoxon rank sum or Wilcoxon signed-rank test. Categorical data were analyzed using the Fisher’s exact test. Event distributions were calculated according to the Kaplan-Meier method and were compared using the logrank test. All p values were 2-sided, and values of
adversely affect LV improvement.12 However, a recent study evaluating clinical outcomes among patients with or without PPM implantation after TAVI showed similar mortality rates in both groups, along with similar cardiac and cerebrovascular morbidity rates.13 The mechanism of maintenance of clinical benefit of TAVI despite adverse hemodynamic effect of PPM in this subgroup of patients is unclear. Therefore, the purpose of the present study was to evaluate the hemodynamic consequences of permanent pacing in patients undergoing TAVI and further define its relation to clinical outcome. Methods Our study population consisted of consecutive patients undergoing percutaneous transfemoral TAVI for treatment of symptomatic aortic valve stenosis using the Edwards Sapien/Sapien XT prosthesis (Edwards Lifesciences, Irvine, California) or the Medtronic CoreValve prosthesis (MCV; Medtronic, Minneapolis, Minnesota) at a tertiary university hospital. The institutional ethics committee approved the study. Inclusion criteria were logistic EuroSCORE >20, porcelain aorta, previous chest radiation, logistic EuroSCORE >10%, and 1 of the following: cachexia,14 need for assistance in basic daily activities,15 or previous open heart www.ajconline.org
Valvular Heart Disease/Hemodynamics and Outcome of PPM After TAVI
surgery. The Institutional Heart Team including an interventional cardiologist, echocardiologists, and a cardiothoracic surgeon determined eligibility for TAVI. Temporary pacemakers were implanted in all patients unless they were already carrying PPM. For the immediate 36 hours after procedure interval, the patients were monitored in the intensive care unit; the temporary pacemaker was extracted after 24 hours unless AV conduction disturbances were detected. The criteria for PPM implantation included pre-TAVI right bundle branch block (BBB), post-TAVI complete AV block (AVB), type II second-degree AV block, alternating BBB, or new-onset LBBB with PR-interval prolongation 280 ms. Patients undergoing TAVI have been enrolled in a prospective registry recording procedural information as well as clinical, echocardiographic, and pacemaker data collected at baseline 30 days, 6 months, 12 months, and then at regular 6- to 12-month intervals. Clinical end points were overall survival and cardiovascular event-free survival. Cardiovascular events included hospitalization for heart failure, stroke, myocardial infarction, new-onset atrial fibrillation, and AVB occurring after discharge. At each follow-up visit, we recorded pacemaker pacing mode, battery and electrode status, capture threshold, impedance and sensing as well as cumulative percentage of RV pacing. Devices were programed to minimize ventricular pacing by activating ventricular pacing suppression algorithms, increasing AV delay, or lowering pacing rate. Transthoracic echocardiographic data were obtained with a standardized protocol. LV linear dimensions were obtained on the parasternal long-axis view. LV mass was estimated based on LV modeling as a prolate ellipse.16 LV end-diastolic volume, LV end-systolic volume, and LVEF were calculated using the Simpson rule16 or Quinones method.17 LV stroke volume (LVSV) was calculated as previously described.18 Early (E) transmitral filling peak velocity was obtained with pulsed wave Doppler sample at the tip of the mitral valve leaflets.19 Early velocity of the septal and lateral aspects of the mitral annulus was measured using tissue Doppler imaging. Mitral annulus velocity (e0 ) was defined as the average of septal and lateral velocities.19 Left atrial volume was determined using the biplane arealength or dimension-length methods.20 RV index of myocardial performance (RIMP) was used for global estimates of RV systolic and diastolic functions. RIMP was defined as the ratio of isovolumic time and ejection time.20 RV ejection time was measured using pulsed wave Doppler of the RV outflow. The isovolumetric time was calculated as the difference between the duration of flow of the tricuspid regurgitant jet and the ejection time.21 Systolic pulmonary artery pressure (SPAP) and right atrial pressure were estimated as previously described.21 Echo Doppler indexes of aortic stenosis severity included the maximal velocity across the aortic valve, the mean pressure gradient, and the aortic valve area (AVA) calculated using the standard continuity equation.19 Severe aortic stenosis was defined as an AVA 40 mm Hg.18 Aortic, mitral, and tricuspid regurgitation grades were classified semiquantitatively on a scale of 0 to 4 according to established criteria.22 AVA, left atrial volume, LV enddiastolic volume, LV end-systolic volume, and LVSV were indexed by body surface area.
133
Table 1 Baseline demographic and clinical characteristics of entire cohort and patients with and without permanent pacemaker implantation Variable
Age (yrs) Men Height (cm) Weight (kg) Body surface area (m2) Diabetes mellitus Dyslipidemia Hypertension Peripheral vascular disease Past stroke Heart failure Previous coronary bypass grafting Atrial fibrillation Logistic EuroSCORE
Total (n ¼ 230), %
Pacemaker Implanted, %
p
No (n ¼ 172)
Yes (n ¼ 58)
83 5 38.3 162 9 72 15 1.64 0.4 32.6 78.3 84.8 11.3
83 6 34.9 162 9 71 15 1.62 0.4 30.2 79.7 86 12.8
83 5 48.3 163 10 75 15 1.7 0.4 39.7 74.1 81 9
0.91 0.086 0.61 0.15 0.18 0.198 0.462 0.399 0.337
7.8 34.8 18.3
7.6 37.8 19.8
8.6 25.9 13.8
0.781 0.112 0.432
18.7 26 14
19.8 26 13
15.5 26 13
0.562 0.9
Table 2 Baseline echo Doppler characteristics of entire cohort and patients with and without permanent pacemaker implantation Variable
Aortic valve Vmax (cm/s) Aortic valve mean gradient (mm Hg) AVA index (cm2/m2) Left atrial volume index (ml/m2) LV mass index (g/m2) LV end-diastolic volume index (ml/m2) LV end-systolic volume index (ml/m2) SV index (ml/m2) LVEF (%) E wave (cm/s) e0 (cm/s) E/e0 SPAP (mm Hg) Tricuspid regurgitation time (ms) RV ejection time (ms) RIMP Mitral regurgitation grade 2 (%) Tricuspid regurgitation grade 2 (%) Aortic regurgitation grade 2 (%)
Total
428 66 47 15
Pacemaker Implanted No
Yes
439 66 47 66
428 66 46 15
p
0.26 0.86
0.43 0.12 0.42 0.12 0.44 0.15 0.4 47 17 48 18 45 21 0.50 139 56 63 20
134 65 63 24
137 32 65 26
0.77 0.7
28 10
28 16
28 15
0.74
0.67 0.54 0.67 0.59 0.64 0.12 0.66
33 58 99 5.1 21 44 402
12 10 31 1.6 8 15 45
33 58 100 5 21 45 401
15 11 32 1.7 8 15 50
34 58 98 5.2 20 41 404
18 10 30 1.4 8 13 45
306 41 305 32 308 40 0.55 0.32 0.15 0.32 0.15 0.32 0.15 0.99 44 (19) 38 (22) 6 (10) 0.054 23 (10)
18 (10)
5 (9)
0.76
18 (8)
15 (9)
3 (5)
0.57
The implantation depth of the bioprosthesis was assessed using aortography immediately after TAVI. For each patient, the distance between lower edges of the noncoronary cusp and ventricular end of prosthesis frame were
134
The American Journal of Cardiology (www.ajconline.org)
Table 3 The changes between baseline and 6-month follow-up echo Doppler parameters in entire cohort and patients with and without permanent pacemaker implantation Variable
Aortic valve Vmax (cm/s) Aortic valve mean gradient (mm Hg) Left atrial volume index (ml/m2) LV mass index (g/m2) LV end-diastolic volume index (ml/m2) LV end-systolic volume index (ml/m2) SV index (ml/m2) LVEF (%) E wave (cm/s) e0 (cm/s) E/e0 SPAP (mm Hg) Tricuspid regurgitation time (ms) RV ejection time (ms) RIMP Mitral regurgitation grade reduced from 2 to 1 (%) Tricuspid regurgitation grade reduced from 2 to 1 (%) Aortic regurgitation grade 2 (%)
Total
Pacemaker Implanted
235 118 38 15 2 12 6 32 4 14 1 10 28 18 4.2 30 0.5 1.8 0.14 7.7 4 10 9 50 7 35 0.01 0.13 28 (56) 8 (35) 18 (8)
obtained twice. Implantation depth was defined as an average of these 2 measurements. Continuous normally distributed parameters are presented as the mean SD and were compared using the Student or paired t test, as appropriate. Ordinal and/or non-normally distributed data are presented by the median and the first and third quartiles and were compared using the Wilcoxon rank sum or Wilcoxon signed-rank test. Categorical data were analyzed using the Fisher’s exact test. Event distributions were calculated according to the Kaplan-Meier method and were compared using the logrank test. All p values were 2-sided, and values of
