10.1111@pace.13216

Heart Failure Severity, Inappropriate ICD therapy, and

Novel ICD Programming: A MADIT-RIT Sub-Study

Usama A. Daimee, MD, Katherine Vermilye, BS, Spencer Rosero, MD,

Claudio D. Schuger, MD, James P. Daubert, MD, Wojciech Zareba, MD, PhD,

Scott McNitt, MS, Bronislava Polonsky, MS, Arthur J. Moss, MD,

Valentina Kutyifa, MD, PhD

Heart Research Follow-Up Program, University of Rochester Medical Center, Rochester, NY; Cardiology Division, Duke University Medical Center, Durham, NC; and Division of Cardiology, Henry Ford Hospital, Detroit, MI

Address for correspondence:

Valentina Kutyifa, MD, PhD

Heart Research Follow-up Program

Cardiology Division, University of Rochester Medical Center

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi:

10.1111/pace.13216.

This article is protected by copyright. All rights reserved.

265 Crittenden Blvd., PO Box 653, Rochester, NY 14642

E-mail: valentina.kutyifa@heart.rochester.edu

Running title: NYHA and inappropriate ICD therapy

Clinical trial registration: https://clinicaltrials.gov/ct2/show/NCT00947310

ABSTRACT

Background: The effects of heart failure (HF) severity on risk of inappropriate implantable cardioverter-defibrillator (ICD) therapy have not been thoroughly investigated. We aimed to study the association between HF severity and inappropriate ICD therapy in MADIT-RIT.

Methods: MADIT-RIT randomized 1500 patients to three ICD programming arms: conventional (Arm A), high-rate cut-off (Arm B: ≥200 bpm), and delayed therapy (Arm C: 60second delay for ≥170 bpm). We evaluated the association between New York Heart

Association (NYHA) Class III (n=256) versus Class I-II (n=251) and inappropriate ICD therapy in Arm A patients with ICD-only and cardiac resynchronization therapy with defibrillator (CRT-D). We additionally assessed benefit of novel ICD programming in Arms B and C versus Arm A by NYHA Classification.

Results: In Arm A, the risk of inappropriate therapy was significantly higher in those with NYHA III vs. NYHA I-II for both ICD (HR=2.55, CI: 1.51-4.30, p<0.001) and CRT-D patients

This article is protected by copyright. All rights reserved. 2

(HR=3.73, CI: 1.14-12.23, p=0.030). This was consistent for inappropriate ATP and inappropriate ICD therapy < 200 bpm, but not for inappropriate shocks. Novel ICD programming significantly reduced inappropriate therapy in patients with both NYHA III (Arm B vs. A: HR=0.08, p<0.001; Arm C vs. A: HR=0.17, p<0.001) and NYHA I-II (Arm B vs. A: HR=0.25, p<0.001; Arm C vs. A: HR=0.28, p<0.001).

Conclusion: Patients with more severe HF are at increased risk for inappropriate ICD therapy, particularly ATP due to arrhythmias < 200 bpm. Novel programming with high-rate cut-off or delayed detection reduces inappropriate ICD therapies in both mild and moderate HF.

Keywords: ICD programming; HF severity; Inappropriate ICD therapy

INTRODUCTION

Heart failure (HF) with severely impaired left ventricular function leads to increased risk of ventricular arrhythmias and sudden cardiac death. Implantable cardioverter-defibrillators (ICDs) have been shown to successfully treat life-threatening ventricular arrhythmias and improve survival.1-3 However, supraventricular arrhythmias often co-exist with HF and may trigger inappropriate ICD therapies.4-6

Risk stratification for inappropriate ICD shocks may help identify high-risk patients. Prior studies have reported smoking, history of atrial fibrillation, younger age, and appropriate shock as predictors of inappropriate ICD shocks, which may further be associated with impaired outcomes.7-13 Two small single-center studies have suggested a link between more advanced HF symptoms and inappropriate shocks in ICD-only patients.14,

This article is protected by copyright. All rights reserved. 3

15 However, there is a scarcity of data on inappropriate ICD therapy based on HF severity in ICD-only and cardiac resynchronization therapy with defibrillator (CRT-D) patients from a contemporary era when biventricular pacing is more common. Moreover, while strategies utilizing high-rate cut-off and prolonged detection have been shown to reduce inappropriate ICD therapy, the effect of HF severity on benefit from such novel ICD programming has not previously been reported.16-18

Accordingly, the objectives of our study were (1) to examine the effect of HF severity on inappropriate ICD therapy in ICD-only and CRT-D patients, and (2) to assess the impact of HF severity on benefit derived from high-rate and duration-delay programming to reduce the risk of inappropriate ICD therapy.

METHODS

Study Protocol and Treatment

The protocol and results of the Multicenter Automatic Defibrillator Implantation Trial – Reduce Inappropriate Therapy (MADIT-RIT) have been described previously. In summary, MADIT-RIT enrolled 1500 patients at 98 centers from the United States, Canada, Europe, Israel, and Japan who met guideline-based criteria for primary prevention ICD-only or CRT-D devices. Exclusion criteria included NYHA Class IV and patients with second or third degree heart block. Patients were randomized to conventional ICD programming (“Arm A”), high-rate therapy (“Arm B”), or delayed therapy (“Arm C”). The institutional review board

This article is protected by copyright. All rights reserved. 4

at each of the participating centers approved the protocol. All patients provided written informed consent.

Device Programming, Interrogation, and Follow-up

All patients enrolled in MADIT-RIT had dual-chamber ICD or CRT-D devices, thus including atrial discrimination for inappropriate ICD therapy detection. Ventricular tachycardia (VT) and fibrillation detection and therapy programming were according to randomization arm, with the three programming arms differing in the heart rate cut-off for treatment of VT and delay in the treatment of a VT episode. Briefly, conventional programming, Arm A, involved a VT detection zone ≥ 170 bpm (detection delay 2.5 seconds; treatment with anti-tachycardia pacing (ATP) and shock) and a ventricular fibrillation (VF) zone ≥ 200 bpm (detection delay 1 second; treatment with Quick Convert ATP and shock). High-rate VT therapy programming in Arm B comprised of a monitor-only VT zone ≥ 170 bpm and a VF zone ≥ 200 bpm

(detection delay 2.5 seconds; treatment with Quick Convert ATP and shock). Delayed VT therapy programming in Arm C comprised of a VT zone ≥ 170 bpm (detection delay 60 seconds; treatment with ATP and shock), a second VT zone 200 – 249 bpm (detection delay

12 seconds; treatment with Quick Convert ATP and shock), and a VF zone ≥ 250 bpm

(detection delay 2.5 seconds; treatment with shock).

MADIT-RIT patients were followed for an average duration of 1.4 years. Patients had follow-up visits every 3 months during the first year and every 6 months thereafter. During each visit, a brief interim history, physical examination, and ICD interrogation were performed. Data were transmitted to the study Coordination and Data Center at the University of Rochester, Rochester, NY. For all patients, episodes from device interrogations were independently reviewed by the adjudication committee, which was blinded to the programming arm, based on pre-specified criteria.

This article is protected by copyright. All rights reserved. 5

Sub-study Population and End Points

The present study is a retrospective analysis of 1486 (99.1%) patients for whom data related to baseline New York Heart Association (NYHA) Functional Classification were available. The first occurrence of any inappropriate ICD therapy, the primary end point, was defined as any ICD or CRT-D therapy delivered for supraventricular rhythms (sinus tachycardia, atrial fibrillation, atrial flutter, and regular supraventricular tachycardia including atrial tachycardia, atrioventricular reentry tachycardia, and atrioventricular tachycardia) or non-arrhythmic events such as electromechanical interference, ICD lead noise, oversensing, or myopotentials. Secondary end points consisted of the first occurrences of inappropriate ATP and shock, and we further assessed inappropriate therapy episodes stratified by the heart rate of the non-ventricular arrhythmic event (< 200 bpm or ≥ 200 bpm) at onset. We have chosen first inappropriate ICD therapy as the primary end point because the MADIT-RIT investigators were allowed to change device programming after the first occurrence of any inappropriate ICD therapy.

The effects of HF severity on different types of inappropriate ICD therapy were assessed only in those patients with conventional programming (Arm A, n=507), who served as the control group relative to the novel programming arms (Arms B and C). Since patients with greater HF severity were more often prescribed CRT-D vs. ICD therapy based on guidelines at the time of MADIT-RIT 20, Arm A patients were divided into four sub-groups: NYHA III patients with ICD, NYHA III patients with CRT-D, NYHA I-II patients with ICD, and NYHA I-II patients with CRT-D. In a secondary analysis, the benefit from novel programming was assessed in relation to conventional programming for the primary end point.

Statistical analysis

This article is protected by copyright. All rights reserved. 6

Continuous variables were expressed as mean ± standard deviation (SD). Categorical data are summarized as frequencies and percentages. Baseline clinical characteristics were compared between patients with heart failure of NYHA Class I-II and Class III severity using the non-parametric Wilcoxon rank-sum test for continuous variables and chi-squared test or

Fisher’s exact test for dichotomous variables.

The cumulative probabilities of any inappropriate therapy, inappropriate ATP, inappropriate shock, inappropriate therapy for heart rate < 200 bpm, and inappropriate therapy for heart rate ≥ 200 bpm by NYHA Classification in ICD-only and CRT-D patients from Arm A were estimated using the Kaplan-Meier method. Additionally, the cumulative probabilities of any inappropriate therapy in novel relative to conventional programming were determined for NYHA Class I-II and NYHA Class III patients. Comparisons of cumulative event rates were performed using the log-rank test.

Multivariable Cox proportional hazards regression analyses were used to assess the relationship of NYHA Classification with the risk of various end points in ICD-only and CRT- D patients from Arm A. For each of the end points, the Cox models assessed the four patient groups simultaneously. The Cox models were adjusted for relevant clinical covariates, which were determined using best subset selection with the added stipulation of significance level at p < 0.05. Cox proportional hazards models analyzing benefit of novel relative to conventional programming were also performed, and further adjusted for ICD-only vs. CRT- D device type.

All statistical tests were two-sided, and a p-value of < 0.05 was considered statistically significant. The analyses were carried out with SAS software (version 9.3, SAS institute, Cary, North Carolina).

This article is protected by copyright. All rights reserved. 7

RESULTS

Baseline Clinical Characteristics

Among the 1486 patients with baseline NYHA class data, there were 706 (47.5%) with NYHA Class I-II (mild HF) and 780 patients (52.5%) with NYHA Class III (moderate HF). In Arm A, there were 251 (49.5%) patients with NYHA I-II HF and 256 (50.5%) with NYHA III HF.

Baseline characteristics of NYHA I-II and NYHA III patients are presented in Table 1. Those with moderate HF symptoms were more likely to be older and female and carry diagnoses of diabetes and hypertension. Patients with moderate HF had lower ejection fraction. However, mild HF patients had greater likelihood of ischemic cardiomyopathy and previous myocardial infarction. Additionally, mild HF patients more commonly received prior treatment for atrial arrhythmias and were more often prescribed beta-blockers, though the diagnoses of atrial fibrillation and flutter were similar in the two patient categories. Importantly, because NYHA III patients were more likely to receive CRT-D devices, further analysis was presented separately for CRT-D and ICD-only patients.

Effects of Heart Failure Severity on Inappropriate Therapy in ICD and CRT-D Patients

In analyzing Arm A, patients with NYHA III implanted with an ICD-only had the highest probability of any inappropriate therapy relative to all other patient groups (Figure 1, p for overall difference between four sub-groups < 0.001). The likelihood of inappropriate ATP was greatest in NYHA III patients with ICD and lowest in NYHA I-II patients with CRT-D (Figure 2A, p for overall difference between four sub-groups < 0.001), as found for any

This article is protected by copyright. All rights reserved. 8

inappropriate therapy. No significant difference was observed for inappropriate shock (Figure 2B, p for overall difference between four sub-groups = 0.904).

Patients with NYHA III implanted with an ICD-only had the highest probability of inappropriate therapy for HR < 200 bpm in comparison to all other patient groups (Figure 3A, p for overall difference between four sub-groups < 0.001). No significant separation was noted for inappropriate therapy for HR ≥ 200 bpm (Figure 3B, p for overall difference between four sub-groups = 0.248).

Multivariable analysis confirmed that, among patients receiving an ICD-only, NYHA Class III relative to NYHA Class I-II was associated with a significantly greater risk of any inappropriate ICD therapy (Table 2A, HR: 2.55, p<0.001), inappropriate ATP (Table 2B, HR=2.57, p<0.001), and inappropriate ICD therapy < 200 bpm (Table 2C, HR=2.89, p<0.001).

Similarly, among CRT-D patients, multivariable adjustment revealed that NYHA Class III vs. NYHA Class I-II was related to significantly increased risk of any inappropriate ICD therapy (Table 2A, HR=3.73, p<0.001), inappropriate ATP (Table 2B, HR=3.62, p<0.001), as well as inappropriate ICD therapy < 200 bpm (Table 2C, HR=3.32, p<0.001).

Event numbers were low to conduct multivariable analyses on inappropriate shock and inappropriate therapy for HR ≥ 200 bpm.

Relationship of Heart Failure Severity with Benefit from Novel Programming to

Reduce Risk of Inappropriate Therapy

In both NYHA Class III and NYHA Class I-II patients, high-rate and delayed programming were associated with a significantly lower likelihood of any inappropriate ICD therapy,

This article is protected by copyright. All rights reserved. 9

relative to conventional programming (Figure 4A-B). Consistently, multivariable analyses revealed that, among NYHA Class III patients, Arms B and C, in comparison to Arm A, showed significantly lowered risks of any inappropriate therapy by 92% and 83% respectively (Table 3). For NYHA I-II patients, the risk reduction was 75% for Arm B and 72% for Arm C relative to conventional programming.

Sensitivity Analyses

We conducted sensitivity analysis to test the robustness of our findings. Based on the differences in clinical characteristics between NYHA III and NYHA I-II patients, we adjusted our models gender, current smoking, ischemic cardiomyopathy, hypertension, ejection fraction, and medications including beta-blockers and amiodarone, receiving similar results. In MADIT-RIT, the overall risk of mortality was low. However, to account for potential competing risk of death, the combined end point of death and inappropriate ICD therapy was assessed, and no significant change in the results was found.

DISCUSSION

In the present study, we report that moderate HF patients, characterized by NYHA class III, experienced an increased risk of inappropriate ICD therapy, primarily driven by a higher risk of inappropriate ATP and inappropriate ICD therapy for non-ventricular arrhythmias < 200 bpm, relative to those with mild HF as defined by NYHA class I or II. Despite the higher risk in moderate HF patients, novel ICD programming with high-rate cut-off or delayed detection was associated with reduction in inappropriate ICD therapies in both mild and moderate HF sub-groups.

This article is protected by copyright. All rights reserved. 10