Comparator cohort In order to compare the Micra pacing thresholds with transvenous thresholds, we analyzed transvenous pacing thresholds

Transcription

1 Long-term outcomes in leadless Micra transcatheter pacemakers with elevated thresholds at implantation: Results from the Micra Transcatheter Pacing System Global Clinical Trial Jonathan P. Piccini, MD, MHS, FHRS, * Kurt Stromberg, MS, Kevin P. Jackson, MD, * Verla Laager, MA, Gabor Z. Duray, MD, PhD, Mikhael El-Chami, MD, FHRS, Christopher R. Ellis, MD, FHRS, John Hummel, MD, D. Randy Jones, MD, # Robert C. Kowal, MD, PhD, FHRS, ** Calambur Narasimhan, MD, Razali Omar, MD, FHRS, Philippe Ritter, MD, Paul R. Roberts, MD, Kyoko Soejima, MD, PhD, Shu Zhang, MD, FHRS, ## Dwight Reynolds, MD, FHRS *** for the Micra Transcatheter Pacing Study Group From the * Duke University Medical Center, Durham, North Carolina, Medtronic plc, Mounds View, Minnesota, Clinical Electrophysiology, Department of Cardiology, Medical Centre, Hungarian Defence Forces, Budapest, Hungary, Emory University Hospital, Atlanta, Georgia, Vanderbilt University Medical Center, Nashville, Tennessee, Ohio State University, Columbus, Ohio, # Providence Health & Services, Portland, Oregon, ** BHVH at Baylor University Medical Center, Dallas, Texas, CARE Hospitals, CARE Foundation, Hyderabad India, Electrophysiology and Pacing Unit, National Heart Institute, Kuala Lumpur, Malaysia, Hôpital Cardiologique du Haut-Lévêque, CHU Bordeaux, Université Bordeaux, IHU LIRYC, Bordeaux, France, University of Southampton, Southampton, United Kingdom, Department of Cardiology, Kyorin University Hospital, Tokyo, Japan, ## Clinical EP Lab and Arrhythmia Center, Fuwai Hospital, Beijing, China, and *** Cardiovascular Section, University of Oklahoma Health Sciences Center, OU Medical Center, Oklahoma City, Oklahoma. BACKGROUND Device repositioning during Micra leadless pacemaker implantation may be required to achieve optimal pacing thresholds. OBJECTIVE The purpose of this study was to describe the natural history of acute elevated Micra vs traditional transvenous lead thresholds. METHODS Micra study VVI patients with threshold data (at 0.24 ms) at implant (n ¼ 711) were compared with Capture study patients with de novo transvenous leads at 0.4 ms (n ¼ 538). In both cohorts, high thresholds were defined as 41.0 V and very high as 41.5 V. Change in pacing threshold (0 6 months) with high (1.0 to 1.5 V) or very high (41.5 V) thresholds were compared using the Wilcoxon signed-rank test. RESULTS Of the 711 Micra patients, 83 (11.7%) had an implant threshold of 41.0 V at 0.24 ms. Of the 538 Capture patients, 50 (9.3%) had an implant threshold of 41.0 V at 0.40 ms. There were no significant differences in patient characteristics between those with and without an implant threshold of 41.0 V, with the exception of left ventricular ejection fraction in the Capture cohort (high vs low thresholds, 53% vs 58%; P ¼.011). Patients with an implant threshold of 41.0 V decreased significantly (P o.001) in both cohorts. Micra patients with high and very high thresholds decreased significantly (P o.01) by 1 month, with 87% and 85% having 6-month thresholds lower than the implant value. However, when the capture threshold at implant was 42V,only 18.2%had a threshold of 1 V at 6 months and 45.5% had a capture threshold of 42 V. CONCLUSIONS Pacing thresholds in most Micra patients with elevated thresholds decrease after implant. Micra device repositioning may not be necessary if the pacing threshold is 2 V. KEYWORDS Bradycardia; VVI; Pacemaker; Leadless pacemaker; Capture threshold; Clinical trial; Outcomes (Heart Rhythm 2017;14: ) I 2017 The Authors. Published by Elsevier Inc. on behalf of Heart Rhythm Society. All rights reserved. Dr Piccini has received grants for clinical research from AHRQ, ARCA biopharma, Boston Scientific, Gilead, Johnson & Johnson, ResMed, St. Jude Medical, and Spectranetics and serves as a consultant to BMS/Pfizer, GlaxoSmithKline, Laguna Pharmaceuticals, Medtronic, and Spectranetics. Dr Stromberg and Ms Laager are employees of Medronic. Dr Hummel serves as a consultant to Medtronic. Dr El-Chami serves as a consultant to Boston Scientific and Medtronic. Address reprint requests and correspondence: Dr Jonathan P. Piccini, Electrophysiology Section, Duke Clinical Research Institute, Duke University Medical Center, PO Box 17969, Durham, NC address: jonathan.piccini@duke.edu /$-see front matter B 2017 The Authors. Published by Elsevier Inc. on behalf of Heart Rhythm Society. All rights reserved /j.hrthm

2 686 Introduction Permanent pacing has been a long-standing effective therapy for symptomatic bradycardia, with 4350,000 procedures performed each year in the United States alone. 1 However, pacemaker implantation is associated with several risks including an 11% lead complication rate and a pocket complication rate of nearly 7% at 5 years. 2 Furthermore, lead failures are associated with significant morbidity and a 16% risk of subsequent major complications. 3 In contrast to traditional transvenous pacemakers, the leadless transcatheter Micra pacemaker is a miniaturized (0.8 ml) single-chamber ventricular pacemaker that is implanted directly in the right ventricle. The Micra leadless pacemaker is associated with a 51% lower risk of complications in the first 6 months after implant compared with transvenous pacemakers, including a lower risk of infection. 4 The Micra leadless pacemaker has been shown to be effective with low and stable pacing thresholds. The device s nitinol tines, which are separate from the electrode, are used to anchor the device and stimulate less fibrosis, allowing for lower pacing thresholds. Implant of the Micra leadless pacing system is safe and efficient; however, device capture, repositioning, and redeployment may be required to obtain an optimal pacing threshold. Multiple repositioning is associated with a small but increased risk of adverse events, including pericardial bleeding. While thresholds may be elevated after initial deployment, the natural history of capture thresholds in Micra leadless devices are not well defined. More specifically, it is not known what percentage of patients with an elevated (41 V) threshold at implant will have a lower threshold at follow-up. Finally, the differences between threshold progression in leadless devices relative to traditional transvenous devices have not been well described. The objective of this analysis was to test the hypothesis that acute elevated Micra pacing thresholds would improve after implant in a manner comparable to transvenous leads. Methods Micra study cohort `The rationale and design of the Micra pivotal investigational device exemption (IDE) study has been described previously. 5 Pertinent national regulatory authorities and ethics committees at each participating site approved the protocol. In brief, patients with a class I or II guideline indication for de novo ventricular pacing were eligible for enrollment. There was no prespecified exclusion based on patient comorbidity (eg, chronic obstructive pulmonary disease). The study enrolled 745 patients, of whom 726 (97.4%) underwent implant attempts at 56 centers in 19 countries between December 2013 and May All Micra patients with a successful implant and pacing threshold data measured at a pulse duration of 0.24 ms at implant (711 of the 720 successfully implanted patients) were included in this analysis. Comparator cohort In order to compare the Micra pacing thresholds with transvenous thresholds, we analyzed transvenous pacing thresholds Heart Rhythm, Vol 14, No 5, May 2017 from the Capture study. 6 The Capture study was a contemporary study designed to assess pacing thresholds in EnPulse dualchamber devices. Right ventricular leads in Capture patients included multiple lead models and both active and passive fixation models (Medtronic: 4074, 4076, 4092, 5054, 5076, 5092; St. Jude Medical: 1388TC, 1470T, 1688TC; others). The devices in Capture were programmed to DDD or DDD(R) since both atrial and ventricular pacing capture thresholds were being analyzed. The Capture cohort also served as a good comparator study because the EnPulse pacing thresholds could be measured bythedevicein1/8vincrements(thesameasmicra).capture study patients who had their right ventricular lead implanted on the same day as their pulse generator and had a ventricular pacing threshold measured at 0.40 ms at implant were included as the comparator group (n ¼ 538). Follow-up capture thresholds in the Capture cohort were taken from capture management testing performed during clinic visits. Statistical analysis For the purposes of this analysis, in both studies, high pacing thresholds were defined as 41.0 V and very high thresholds as 41.5 V. Patient demographic characteristics, comorbidities, and pacing electrode location were compared within pacing systems for patients with high and low thresholds using the t test or Fisher exact test. We analyzed changes in pacing threshold among those patients with a high (1.0 to 1.5 V) or very high (41.5 V) threshold with paired implant and 6- month pacing threshold measurements within pacing systems using the Wilcoxon signed-rank test and between pacing systems using the Wilcoxon rank-sum test. In addition, the Wilcoxon signed-rank test was used to compare implant and postimplant pacing threshold measurements within pacing systems among patients with a pacing threshold of 41.0 V. Mean Micra pacing thresholds and percentage of patients with a pacing threshold of 41 V by number of device deployments were analyzed using linear and logistic regression, respectively. Multivariable logistic regression was used to identify factors associated with an implant threshold of 41 V at 0.24 ms. Candidate variables included the following baseline variables: pacing indication associated with permanent or persistent atrial fibrillation, hypertension, congestive heart failure, coronary artery disease, prior myocardial infarction, diabetes, renal dysfunction, age 475 years, female sex, body mass index o25 kg/m 2, and history of pulmonary disorder (including chronic obstructive pulmonary disease). Candidate variables also included the following procedural variables: use of heparin bolus during implant, number of Micra deployments, and apical location. Finally, correlations between the change in pacing threshold and impedance and sensing amplitude were assessed using the Pearson correlation coefficient. Significance was determined with a 2-tailed α value of o.05 for all analyses. Results Baseline characteristics The baseline characteristics, comorbidities, and electrode location of the Micra (n ¼ 711) and Capture (n ¼ 538) study

3 Piccini et al Leadless Thresholds 687 Table 1 Baseline assessments for high and low threshold patients by pacing system Micra Capture Characteristic High PCT (n ¼ 83) Low PCT (n ¼ 628) P * High PCT (n ¼ 50) Low PCT (n ¼ 488) P Age (y) Mean ± SD 76.0 ± ± ± ± Median th 75th percentile Minimum maximum No. (%) of patients with measure available 83 (100.0) 628 (100.0) 50 (100.0) 488 (100.0) Sex, n (%) Male 52 (62.7) 368 (58.6) (50.0) 291 (59.6).23 Female 31 (37.3) 260 (41.4) 25 (50.0) 197 (40.4) LVEF (%) Mean ± SD 60.8 ± ± ± ± Median th 75th percentile Minimum maximum No. (%) of patients with measure available 65 (78.3) 538 (85.7) 29 (58.0) 308 (63.1) Other comorbidities, n (%) Diabetes 25 (30.1) 177 (28.2).70 Not collected COPD 6 (7.2) 82 (13.1).16 Not collected Renal dysfunction 18 (21.7) 125 (19.9).66 Not collected LBBB 11 (13.3) 87 (13.9) (6.0) 50 (10.2).46 Vascular disease 7 (8.4) 48 (7.6).83 1 (2.0) 46 (9.4).11 CAD 22 (26.5) 174 (27.7) (38.0) 158 (32.4).43 AF 62 (74.7) 457 (72.8) (34.0) 145 (29.7).52 CHF 14 (16.9) 109 (17.4) (6.0) 64 (13.1).18 Hypertension 63 (75.9) 495 (78.8) (74.0) 320 (65.6).27 Valvular disease 27 (32.5) 271 (43.2).08 6 (12.0) 85 (17.4).43 Pacing electrode location, n (%) Apex 61 (73.5) 416 (66.2) (94.0) 425 (87.1).52 Septum 21 (25.3) 208 (33.1) 2 (4.0) 41 (8.4) Other 1 (1.2) 4 (0.6) 1 (2.0) 22 (4.5) Pacing electrode type, n (%) Active 29 (58.0) 199 (40.8).023 Passive 20 (40.0) 278 (57.0) Unknown 1 (2.0) 11 (2.3) AF ¼ atrial fibrillation; CAD ¼ coronary artery disease; CHF ¼ congestive heart failure; COPD ¼ chronic obstructive pulmonary disease; LBBB ¼ left bundle branch block; LVEF ¼ left ventricular ejection fraction; PCT ¼ pacing capture threshold. *P value from the t test for continuous variables or Fisher exact test for categorical variables comparing high threshold and low threshold Micra patients. P value from the t test for continuous variables or Fisher exact test for categorical variables comparing high threshold and low threshold Capture patients. patients are summarized in Table 1 according to their implant pacing threshold values: those with an implant threshold of 41.0 V (high) vs those with an implant threshold of 1.0 V (low). Of the 711 Micra patients with pacing threshold data measured at implant, 83 (11.7%) patients had an implant pacing threshold of 41.0 V at 0.24 ms. Of the 538 Capture patients included in the analysis, 50 (9.3%) had an implant pacing threshold of 41.0 V at 0.40 ms. There were no significant differences in patient demographic characteristics, comorbidities, or pacing electrode location between patients with and without an implant threshold of 41.0 V. However, in the Capture study, active fixation leads were associated with an implant threshold of 41.0 V (P ¼.023). The mean age of patients with elevated pacing thresholds was 76.0 ± 9.0 years in the Micra cohort compared with 72.5 ± 12.5 years in the Capture cohort. In the Capture study, patients with an implant pacing threshold of 41.0 V had a significantly lower ejection fraction than those without an implant pacing threshold of 41.0 V (53.2% vs 58.3%; P ¼.011), while the converse was true for Micra patients (60.8% vs 58.6%; P ¼.06). Overall, the prevalence of atrial fibrillation was much higher in the Micra cohort (73% vs 30%; P o.001). In both cohorts, the majority of electrodes were implanted in an apical location (Table 1). More electrodes were implanted on the septum in the Micra cohort vs the Capture cohort (32% vs 8%; P o.001). Changes in pacing thresholds after implant Table 2 details the change in pacing thresholds across follow-up (implant, 1 month, and 6 months) in both cohorts. Of the 83 Micra patients with a threshold of 41 V at implant, 1 died before their 6-month visit, 8 were awaiting their 6-month visit, and 2 had missed their 6-month visit at the time of the analysis. The patient who died had a threshold of 1.5 V at their 3-month visit. This left 72 Micra patients with high implant thresholds available for analysis. Among the 45 of 52 Micra patients with a capture threshold of 41.0

4 688 Heart Rhythm, Vol 14, No 5, May 2017 Table 2 Change from implant in pacing threshold for high threshold patients by study Study * High PCT category Visit Patients Mean ± SD Median (range) % Decrease P value vs implant P value comparison Capture 41 to 1.5 V Implant ± ( ) 1 mo ± ( ) 88.5 o mo ± ( ) 80.8 o V Implant ± ( ) 1 mo ± ( ) 89.5 o mo ± ( ) o.001 Micra 41 to 1.5 V Implant ± ( ) Discharge ± ( ) 68.4 o mo ± ( ) 85.7 o mo ± ( ) 86.7 o V Implant ± ( ) Discharge ± ( ) mo ± ( ) mo ± ( ) 85.2 o PCT ¼ pacing capture threshold. *Thresholds were measured at a pulse duration of 0.4 ms for Capture patients and 0.24 ms for Micra patients. Patients with paired implant and 6-mo threshold data. % decrease is the percentage of patients with a lower threshold at follow-up relative to implant value. P value from the Wilcoxon signed-rank test testing whether change from implant is different from zero. P value from the Wilcoxon rank-sum test testing whether changes from implant are different between studies. to 1.5 V at 0.24 ms at implant who had 6-month data available, 87% had a lower pacing threshold at 6 months, with the median pacing threshold decreasing from 1.25 to 0.63 V (P o.001). Among the 27 of 31 Micra patients with a pacing threshold of 41.5 V at 0.24 ms at implant with 6- month data available, 85% had a lower pacing threshold at 6 months, with the median pacing threshold decreasing from 2.0 to 1.0 V (P o.001), though 1 patient without a 6-month pacing threshold available had their Micra device programmed to OOO mode 32 days postimplant for the elevated threshold. This patient had an implant threshold of 2 V and a discharge threshold of 3.88 V followed by intermittent loss of capture. As shown in Online Supplemental Figure 1, changes in pacing impedance from implant to 6 months were not correlated with changes in pacing threshold (r ¼ 0.01; P ¼.84). Similarly, changes in R-wave amplitude were not correlated with changes in pacing threshold (r ¼ 0.08; P ¼.083). Mean thresholds at 6 months did not vary by electrode location (apical vs septum) in either study or by baseline threshold category (41 to 1.5 V and 41.5 to 2 V) (Online Supplemental Table 1). As shown in Table 2, similar changes were observed in the Capture study transvenous lead cohort with a pacing threshold of 41.0 to 1.5 V at 0.40 ms at implant, where 81% of patients had a lower 6-month pacing threshold, with median pacing thresholds decreasing from 1.25 V at implant to 0.75 V at 6 months (P o.001). All the 19 Capture study patients (100%) with a pacing threshold of 41.5 V at 0.40 ms at implant had a lower pacing threshold at 6 months, with median thresholds decreasing from 2.25 to 0.75 V (P o.001). Patients with an implant threshold of 41.0 V at implant tended to decrease after implant as displayed in Figures 1 and 2. An increasing number of Micra deployments was associated with higher pacing thresholds and a higher percentage of patients with a pacing threshold of 41.0 V at implant (Figure 3). Nearly all Micra patients (94.4%) with an implant threshold of 41.0 V had a 6-month pacing threshold that was the same or lower than the implant value (Figure 2). Pacing thresholds at 6 months (by categories of output) are shown in Figure 4. Among those Micra patients with a pacing threshold of 41.0 to 1.5 V at implant, 82.2% had a threshold of 1 V and 97.8% had a threshold of 1.5 V at 6 months. Among those Micra patients with a capture threshold of 41.5 to 2 V at implant, 75% had a threshold of Figure 1 Pacing thresholds for high threshold patients with paired implant and 6-month thresholds by study visit and pacing system. The red line represents pacing threshold at 0.24 ms for 72 Micra patients with an implant pacing threshold of 41.0 V and 6-month threshold measurement. The black line represents pacing threshold at 0.40 ms for 45 Capture patients with an implant pacing threshold of 41.0 V and 6-month threshold measurement. Error bars represent mean ± SD. Asterisk represents values significantly (P o.05) different from the implant value.

5 Piccini et al Leadless Thresholds 689 Figure 2 Micra pacing threshold measurements at discharge (A) and 6 months (B) vs implant. Pacing thresholds at prehospital discharge and 6 months compared to implant thresholds for Micra patients with an implant threshold of 41.0 V at 0.24 ms. Values below the diagonal line represent values lower at the follow-up visit relative to implant. Plotting points are scattered to show the distributional density. 1 V and 87.5% had a threshold of 1.5 V at 6 months. However, when the capture threshold at implant was 42 V, only 18.2% had a threshold of 1 V at 6 months and 45.5% had a capture threshold of 42 V. Predictors of elevated Micra thresholds As shown in Table 3, multivariable logistic regression identified the number of device deployments as the only factor independently associated with an elevated implant threshold (odds ratio [OR] 1.38; 95% confidence interval [CI] ; P o.001), although apical location had a marginally significant relationship with elevated implant thresholds (OR 1.76; 95% CI ; P ¼.053). Discussion We compared pacing thresholds at implant and subsequent follow-up between leadless pacemakers in the pivotal IDE Micra study and traditional transvenous electrodes in the Capture study. There are 3 main findings in our analysis. First, similar to traditional transvenous leads, Micra pacing thresholds decrease after implant. Second, the vast majority patients with a pacing threshold of o2 V had a pacing threshold of 1 V at follow-up. Finally, patients with a pacing threshold of 2 V had a significant risk of persistently elevated pacing thresholds of 42 V at follow-up. The Micra transcatheter leadless pacing system uses 4 self-expanding nitinol tines to anchor the device to the Figure 3 A: Distribution of Micra pacing thresholds at implant according to the number of device deployments. B: Proportion of Micra patients with an implant threshold of 1Vvs41Vat 0.24 ms at implant. The upper and lower bounds of the boxes in panel A represent the 1st and 3rd quartiles; the horizontal black line is the median. The red line connects the mean. The P value is from a linear regression model. In panel B, the odds ratio is the relative increase in the odds of an implant pacing threshold of 41 V per deployment attempt as computed using a logistic regression model. CI ¼ confidence interval. Figure 4 Micra capture thresholds at 6-month follow-up according to implant threshold. This figure displays the 6-month follow-up pacing thresholds in the vertical bars according to the baseline pacing capture threshold at implant (on the x-axis).

6 690 Table 3 Factors associated with a Micra pacing threshold of 41 V at 0.24 ms at implant Variable Odds ratio (95% CI) * P Heparin bolus use 0.67 ( ).124 No. of deployments 1.38 ( ) o.001 Apical device location 1.76 ( ).053 Indication associated with 1.56 ( ).110 persistent or permanent AF Hypertension 0.81 ( ).485 Congestive heart failure 0.89 ( ).739 Coronary artery disease 0.86 ( ).608 Prior MI 1.29 ( ).518 Diabetes 1.11 ( ).701 Renal dysfunction 1.26 ( ).470 Age 475 y 0.90 ( ).666 Female sex 0.92 ( ).733 BMI o25 kg/m ( ).212 Pulmonary disorder 0.76 ( ).338 AF ¼ atrial fibrillation; BMI ¼ body mass index; CI ¼ confidence interval; MI ¼ myocardial infarction. *Odds ratio from multivariable logistic regression model with implant threshold 41 as response and factors in the table included as predictors. Odds ratio per additional Micra deployment. One patient was missing height and weight at baseline and was excluded from the analysis. Total n ¼ 710. ventricular myocardium. At implant, device capture, repositioning, and redeployment may be required to obtain an optimal pacing threshold. The electrodes are TiN coated and are located directly on the pacemaker capsule. The tip or cathode is steroid eluting and has a surface area of 2.5 mm 2, while the ring or anode has a surface area of 27 mm 2. 7 While there appear to be fewer major complications with the Micra leadless pacemaker than with transvenous systems (hazard ratio 0.49; 95% CI ), major complications can occur. 4 In the Micra IDE study, there were 13 cases of pericardial effusion and/or perforation (rate 1.8%) associated with the device, comparable to the rate observed with transvenous lead placement in clinical practice (1.2%). 8 Each additional Micra device repositioning was associated with a 1.35 numerically higher odds of effusion (OR 1.35; 95% CI ; P ¼.052). Because lead repositioning is often influenced by elevated pacing thresholds, we sought to describe the natural history of high implant pacing thresholds in the Micra device. We hypothesized that Micra leadless pacing thresholds would decline over time. The evolution of the electrode-tissue interface in traditional transvenous pacemakers has been well described. The acute phase of threshold progression is marked by a 30% 40% decrease in the pacing threshold within minutes to hours of implant and is then followed by an elevation in threshold due to inflammation and foreign-body response that resolves within months and is limited with the use of steroid-eluting leads. 9,10 In this analysis, we found that the capture thresholds with the Micra leadless pacemaker decreased after implant in a fashion that was similar to traditional transvenous leads. Although the decrease in the pacing capture threshold in the Heart Rhythm, Vol 14, No 5, May 2017 acute phase is more marked with more traumatic transvenous leads (ie, active fixation), a significant decrease was still observed with the Micra transcatheter leadless pacemaker and suggests that the lead-tissue interface evolves favorably after implant. More specifically, in high implant threshold patients, median Micra pacing thresholds decreased by 50% between implant and 6 months resulting in improved battery longevity. For example, a pacing threshold decrease from 2 to o1 V corresponding to pacing outputs of 2.5 to o1.5 V would result in an approximate 3-year increase in device service (Online Supplemental Figure 2) dependent on pacing percentage, rate, and impedance. 11 One potential explanation for the decrease in pacing thresholds observed with the Micra leadless pacemaker is that there is an element of mechanical injury at the site of Micra electrode contact, despite the relative distance of the cathode from the electrically inactive nitinol tines. It is also interesting to note that the active fixation leads in the Capture study had higher thresholds than did the passive fixation transvenous leads. Another potential explanation for the decrease in Micra thresholds postimplant is that the Micra nitonol tines continue to provide active albeit small forces that draw the cathode to the tissue interface. This is not true of transvenous leads. What implications do these data have for clinical practice and Micra implantation procedures? Beyond the reassurance of observing an evolution of pacing thresholds similar to traditional transvenous devices, these data suggest that aggressive repositioning may not be necessary to achieve a reasonable pacing threshold if the initial threshold is below 2 V. Implant values in this range were rarely associated with higher pacing thresholds at follow-up. While lower pacing thresholds will lead to longer battery longevity, the risk of cardiac perforation increases with multiple positioning attempts. This concern is most notable in patients with risk factors for perforation, such as low body mass index, elderly women, and those on long-term corticosteroids. 8,12 However, if the threshold at implant is 2 V, clinicians should strongly consider recapturing and redeploying the device, since an implant pacing threshold of 42 was associated with high capture thresholds in almost half of the patients at 6 months. There are several potential reasons why a capture threshold of 42 V has such a stronger association with elevated thresholds in the future. One possibility might be that either poor contact or intrinsic fibrosis with poor capture to this degree marks a device position or substrate problem that does not improve with maturation of the device-tissue interface. Study limitations There are several limitations that need to be kept in mind when reviewing these data. First and foremost, the comparison was not randomized and thus may be subject to baseline confounding. In addition, the pulse duration in the transvenous Capture cohort was longer; thus, for any given threshold, the total energy was lower in Micra patients. However, these data provide the first quantified comparison

7 Piccini et al Leadless Thresholds 691 of short-term and long-term thresholds and their evolution in traditional transvenous leads vs Micra leadless pacemakers. Conclusion Only 12% of Micra patients had an implant threshold of 41.0 V at 0.24 ms. Similar to traditional transvenous pacemakers, pacing thresholds in most Micra transcatheter leadless pacemakers decrease after implant. In Micra patients with a threshold of 41 to 2 V at implant, approximately three-quarters or more had an optimal threshold ( 1 V)by6 months. Based on these findings, Micra device repositioning may not be necessary if the pacing threshold is o2 V. Appendix Supplementary data Supplementary data associated with this article can be found in the online version at References 1. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics 2012 update: a report from the American Heart Association. Circulation 2012;125:e2 e Udo EO, Zuithoff NP, van Hemel NM, de Cock CC, Hendriks T, Doevendans PA, Moons KG. Incidence and predictors of short- and long-term complications in pacemaker therapy: the FOLLOWPACE study. Heart Rhythm 2012;9: Hauser RG, Hayes DL, Kallinen LM, Cannom DS, Epstein AE, Almquist AK, Song SL, Tyers GF, Vlay SC, Irwin M. Clinical experience with pacemaker pulse generators and transvenous leads: an 8-year prospective multicenter study. Heart Rhythm 2007;4: Reynolds D, Duray GZ, Omar R, et al. A leadless intracardiac transcatheter pacing system. N Engl J Med 2016;374: Ritter P, Duray GZ, Zhang S, Narasimhan C, Soejima K, Omar R, Laager V, Stromberg K, Williams E, Reynolds D, Micra Transcatheter Pacing Study Group. The rationale and design of the Micra Transcatheter Pacing Study: safety and efficacy of a novel miniaturized pacemaker. Europace 2015;17: Rosenthal LS, Mester S, Rakovec P, Penaranda JB, Sherman JR, Sheldon TJ, Zeng C, Wang P, CAPTURE Trial Investigators. Factors influencing pacemaker generator longevity: results from the complete automatic pacing threshold utilization recorded in the CAPTURE Trial. Pacing Clin Electrophysiol 2010;33: Ritter P, Duray GZ, Steinwender C, et al. Early performance of a miniaturized leadless cardiac pacemaker: the Micra Transcatheter Pacing Study. Eur Heart J 2015;36: Mahapatra S, Bybee KA, Bunch TJ, Espinosa RE, Sinak LJ, McGoon MD, Hayes DL. Incidence and predictors of cardiac perforation after permanent pacemaker placement. Heart Rhythm 2005;2: Danilovic D, Ohm OJ. Pacing impedance variability in tined steroid eluting leads. Pacing Clin Electrophysiol 1998;21: Schwaab B, Schwerdt H, Heisel A, Frohlig G, Schieffer H. Chronic ventricular pacing using an output amplitude of 1.0 volt. Pacing Clin Electrophysiol 1997;20: Medtronic. Micra TM MC1VR01 Clinician Manual. Page medtronic.com/wcm/groups/mdtcom_sg/@emanuals/@era/@crdm/documents/ documents/contrib_ pdf. Accessed June 18, Cano O, Andres A, Alonso P, Osca J, Sancho-Tello MJ, Olague J, Martinez-Dolz L. Incidence and predictors of clinically relevant cardiac perforation associated with systematic implantation of active-fixation pacing and defibrillation leads: a single-centre experience with over 3800 implanted leads. Europace 2017;19: