Cardiac resynchronization therapy (CRT) is an established treatment for heart failure patients with myocardial dysfunction and delayed ventricular activation, but approximately 25% to 40% of patients do not respond to CRT. Left ventricular (LV) multisite pacing (MSP) has been proposed as a tool to improve CRT response. The goal of this study is to examine the safety and efficacy of LV MSP in CRT nonresponders. Between January 2018, and September 2019, the Strategic Management to Improve CRT Using Multi-Site Pacing trial prospectively enrolled 584 CRT-defibrillator recipients for established indications at 52 sites across the United States and evaluated their response at 6 months using the clinical composite score (CCS). Of the nonresponders, 102 patients had the LV MSP feature turned on and 78 patients completed the 12-month CCS evaluation. The LV MSP feature-related complication-free rate was 99.0% with a lower 95% confidence interval limit of 94.9%, which was higher than the performance goal of 90%. The proportion of nonresponders with an improved CCS from 6 to 12 months was 51.3% with a lower 95% confidence interval limit of 41.4%, which was higher than the performance goal of 5%. The estimated mean reduction in battery longevity with the LV MSP feature was about 3.6 months (estimated battery longevity of 8.87 ± 2.08 years at 6 months and 8.07 ± 2.23 years at 12 months). In conclusion, in CRT nonresponders, the use of the LV MSP feature is safe and associated with a ∼50% conversion rate with a small projected reduction in CRT-defibrillator battery longevity. LV MSP should be considered in the management of CRT nonresponders.
Cardiac resynchronization therapy (CRT) is an established treatment for heart failure (HF) patients with reduced left ventricular ejection fraction (LVEF). CRT improves the functional status and myocardial function of patients, and decreases HF events and all-cause mortality. , Still, a significant proportion of eligible patients do not clinically benefit from CRT. Patient baseline clinical characteristics and the specific region of left ventricular (LV) pacing, away from the scar , and close to the site of the latest mechanical , , or electrical activation, have been implicated in patient response to CRT. Recently, pacing at multiple sites in the LV has been associated with improved response rates in smaller studies, but larger trials have not substantiated these findings or had mixed results. The present multicenter, observational trial was therefore designed to examine the safety and efficacy of LV multisite pacing (MSP) in improving symptomatic HF as measured by the clinical composite score (CCS) in nonresponders to 6 months of conventional CRT.
The Strategic Management to Improve CRT Using Multi-Site Pacing (SMART-MSP) trial was a prospective, single-arm, multicenter, postapproval study sponsored by Boston Scientific Corporation (Marlborough, Massachusetts) and developed in collaboration with the Food and Drug Administration. The protocol was approved by the institutional review boards of all participating institutions and all participants provided written, informed consent before enrollment. A Clinical Events Committee adjudicated safety and prespecified HF events. Furthermore, a physician steering committee provided oversight of key study decisions, final data interpretation, and publication. The objective of the SMART-MSP trial was to demonstrate the effectiveness of the LV MSP to improve CRT nonresponse rate and confirm the safety of the feature in an aftermarket setting. The trial was registered at ClinicalTrials.gov (NCT03257436).
The SMART-MSP study targeted 586 subject enrollments at approximately 60 centers. Subjects were followed in-clinic at enrollment, 6- and 12-month visits. Subjects were programmed to conventional CRT from the point of enrollment to the 6-month visit. At the 6-month visit, subjects were evaluated with the CCS and classified as Responder or Nonresponder. Responders exited the study at the 6-month visit. LV MSP was enabled in active Nonresponders who were followed for an additional 6 months. Active Nonresponders with LV MSP enabled (LV MSP Group) were classified again by way of CCS at the 12-month visit to evaluate their response to LV MSP. The study design is illustrated in Figure 1 .
Patients 18 years of age or older who received a Resonate, Momentum, or Vigilant device with an Acuity quadripolar LV lead (Boston Scientific, Massachusetts), for established clinical indications were considered for participation in this trial. Enrollment was limited to patients with: moderate to severe HF (New York Heart Association [NYHA] Class III to IV), a LVEF less than or equal to 35%, and QRS duration ≥120 ms; patients with left bundle branch block (LBBB) with QRS duration ≥130 ms, LVEF ≤30%; or mild (NYHA Class II) ischemic or nonischemic HF or asymptomatic (NYHA Class I) ischemic HF. Patients were excluded if they received previous LV pacing; had permanent atrial fibrillation or permanent complete atrioventricular block; had a life expectancy of fewer than 12 months, or were women that were pregnant or planned to become pregnant over the course of the study.
The CCS was calculated by the investigator or delegated study staff for active subjects at the 6- and 12-month visits and assessed the following components: (1) all-cause mortality; (2) HF events; (3) patient global assessment; and (4) NYHA HF classification. Qualified site staff at participating centers performed NYHA assessments and were blinded to the study design, subject clinical status, and programmed device settings. Subjects were classified as either improved, unchanged, or worsened based on CCS calculation. Subjects were considered improved if they experienced a favorable change by at least 1 NYHA functional class or in the patient global assessment, or both while remaining alive and free of HF events. Subjects were considered worsened if they died from any cause, experienced an HF event, or reported worsening of at least 1 NYHA functional class, or reported worsening of the patient global assessment. Subjects were considered unchanged if they were neither improved nor worsened. Subjects with an improved CCS were classified as Responders whereas subjects with an unchanged or worsened CCS were classified as Nonresponders. An HF event was defined as an adverse event with a primary cause of HF. All reported hospitalizations and HF events were adjudicated by the Clinical Event Committee for relatedness to HF.
The primary safety end point was the LV MSP feature-related complication-free rate between the date LV MSP was enabled and 180 days after, in nonresponders with LV MSP on for any duration. LV MSP feature-related complications were defined as those complications that were determined by the site principal investigator as “yes, related” or “possibly related” to the LV MSP feature. A sample of 61 subjects was required to evaluate the primary safety end point to achieve a power of 80%. The performance goal for the LV MSP feature-related complication-free rate was 90%.
The primary effectiveness end point for the SMART-MSP study was the proportion of the LV MSP group with an improved CCS at 12 months. The overall study sample size of 586 subjects was driven by the primary effectiveness end point. The primary effectiveness performance goal of 5% conversion from Nonresponder to Responder required a sample size of 110 subjects to achieve 95% power. Given the estimated overall attrition of 81%, driven in large part by the 6-month response rate, 586 subjects were required to obtain this sample size.
The LV MSP feature-related complication-free rate between the 6- and 12-month visits was calculated using Kaplan-Meier methodology. Each subject’s evaluation period began when LV MSP was turned on (at or after the 6-month visit) and continued through 180 days thereafter. The 95% one-sided lower confidence limit of the LV MSP feature complication-free rate was calculated using the log-log methodology and compared with the 90% performance goal. Subjects that had LV MSP enabled at 6 months but failed to reach 180 days of follow-up, without experiencing an end point event, were censored from the safety end point analysis.
The primary effectiveness end point (CCS Improved) was analyzed when the last 12-month visit was completed by subjects in the LV MSP Group. Analysis was limited to subjects with ≥93% LV MSP pacing with complete CCS data. Deceased subjects and those who withdrew from the study after an HF event were included in the effectiveness end point and classified as nonresponders. Subjects with the LV MSP feature on for less than 6 months were included in the analysis. The 95% one-sided lower confidence limit of the response rate was calculated using one-sided exact methodology for a single binomial proportion and compared with the performance goal of 5%. Subjects with incomplete CCS data from the 12-month follow-up visit were considered to have missing data. To understand the impact of various study components on the primary effectiveness end point, multiple sensitivity analyses were performed.
The baseline demographic and clinical characteristics of patients enrolled in SMART-MSP are listed in Table 1 . The SMART-MSP population consisted of typical CRT recipients. The mean age of participants at enrollment was 67 years, and most were White (81%) men (64%). The proportion of patients with ischemic etiology of cardiomyopathy was 41% and a significant majority (92%) underwent CRT-defribilator implantation for primary prevention of sudden cardiac death. The mean LVEF of patients was 25% and their QRS width 158 ms, overwhelmingly (81%) of LBBB morphology. Details on device settings are reported in Supplementary Table 1 , no differences were evident between total patients and MSP patients.
| Variable | Total Enrolled (n = 584) | MSP Group (n = 78) |
|---|---|---|
| Age (years) | 66.7 ± 11.0 | 66.7 ± 11.0 |
| Women | 211 (36%) | 27 (35%) |
| Left ventricular ejection fraction (%) | 25.0 ± 6.5 | 23.9 ± 5.9 |
| QRS Width (ms) | 157.9 ± 20.5 | 155.1 ± 17.7 |
| Ischemic cardiomyopathy | 241 (41%) | 30 (38%) |
| Nonischemic cardiomyopathy | 342 (59%) | 48 (61%) |
| Sudden cardiac death primary prevention | 535 (92%) | 67 (86%) |
| Atrial fibrillation | 131 (22%) | 19 (24%) |
| Atrial flutter | 31 (5%) | 5 (7%) |
| Bradycardia | 200 (34%) | 27 (35%) |
| Renal dysfunction | 120 (21%) | 17 (22%) |
| Hypertension | 442 (76%) | 55 (70%) |
| Diabetes Mellitus | 222 (38%) | 30 (39%) |
| Chronic pulmonary Disease | 254 (44%) | 34 (44%) |
| Left bundle branch block | 471 (81%) | 62 (79%) |
| Right bundle branch block | 58 (10%) | 12 (15%) |
| Intraventricular conduction delay | 28 (5%) | 3 (4%) |
| New York Heart Association I | 7 (1%) | 2 (3%) |
| New York Heart Association II | 226 (40%) | 34 (44%) |
| New York Heart Association III | 345 (59%) | 42 (54%) |
| New York Heart Association IV | 5 (1%) | 0 (0%) |
| Any renin-angiotensin medication | 448 (77%) | 60 (77%) |
| Angiotensin-converting enzyme inhibitor | 211 (36%) | 25 (32%) |
| Angiotensin receptor blocker | 109 (19%) | 21 (27%) |
| ARNI | 133 (23%) | 15 (19%) |
| Aldosterone antagonists | 212 (36%) | 24 (31%) |
| Anti-platelet therapy | 401 (69%) | 50 (64%) |
| Beta-blockers | 537 (92%) | 73 (94%) |
| Class I antiarrhythmics | 9 (1%) | 1 (1%) |
| Class III antiarrhythmics | 81 (14%) | 11 (14%) |
| Diuretics | 368 (63%) | 43 (55%) |
| Oral anticoagulant | 87 (15%) | 12 (15%) |
| Statin | 361 (62%) | 48 (61%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
