← Product Code [LWP](/productcode/LWP) · P010012S230

# BOSTON SCIENTIFIC CARDIAC RESYNCHRONIZATION THERPY DEFIBRILLATORS (CRT-DS) (P010012S230)

_Boston Scientific Corp · LWP · Sep 16, 2010 · Cardiovascular · APPR_

**Canonical URL:** https://fda-staging.innolitics.com/device/P010012S230

## Device Facts

- **Applicant:** Boston Scientific Corp
- **Product Code:** [LWP](/productcode/LWP.md)
- **Decision Date:** Sep 16, 2010
- **Decision:** APPR
- **Device Class:** Class 3
- **Review Panel:** Cardiovascular
- **Attributes:** Therapeutic, Expedited Review

## Indications for Use

These Boston Scientific Cardiac Resynchronization Therapy Defibrillators (CRT-Ds) are indicated for patients with heart failure who receive stable optimal pharmacologic therapy (OPT) for heart failure and who meet any one of the following classifications: - Moderate to severe heart failure (NYHA Class III-IV) with EF ≤ 35% and QRS duration ≥ 120 ms - Left bundle branch block (LBBB) with QRS ≥ 130 ms, EF ≤ 30%, and mild (NYHA Class II) ischemic or nonischemic heart failure or asymptomatic (NYHA Class I) ischemic heart failure

## Device Story

Implantable pulse generator (CRT-D) combining ICD and cardiac resynchronization therapy; senses electrical activity via atrial, coronary venous, and RV leads; delivers biphasic/monophasic shocks and biventricular pacing to synchronize ventricular contractions; features accelerometer-based adaptive-rate bradycardia therapy. Used in clinical settings by physicians to treat life-threatening ventricular arrhythmias and heart failure symptoms. Output includes antitachycardia pacing and defibrillation shocks. Benefits include reduced heart failure events and improved cardiac function in LBBB patients.

## Clinical Evidence

Prospective, randomized, multicenter MADIT-CRT study (N=1820). Primary effectiveness endpoint: combined all-cause mortality or heart failure event. In LBBB sub-population (N=1281), CRT-D reduced relative risk of primary endpoint by 57% (HR 0.43, 95% CI 0.33-0.56, p<0.001) vs ICD. Safety endpoint (system-related complication-free rate at 3 months) met (83.4% in LBBB cohort, >70% threshold).

## Technological Characteristics

Implantable pulse generator; metallic housing (titanium) as active electrode; Triad electrode system; IS-1/LV-1/DF-1 lead compatibility; accelerometer-based adaptive-rate pacing; biphasic/monophasic waveform energy delivery.

## Regulatory Identification

These devices treat bradycardia (slow heartbeats) with RA and/or RV pacing therapy as necessary.

## Reference Devices

- Cognis Cardiac Resynchronization High Energy Defibrillator Models N118, N119 ([P010012](/device/P010012.md)/S165)
- Livian Cardiac Resynchronization Therapy Defibrillators Models H220, H225, H227 and H229 ([P010012](/device/P010012.md)/S154)
- Contak Renewal 3 RF HE CRT-D Models H210, H215, H217, H219 ([P010012](/device/P010012.md)/S031)

## Submission Summary (Full Text)

> This content was OCRed from public FDA records by [Innolitics](https://innolitics.com). If you use, quote, summarize, crawl, or train on this content, cite Innolitics at https://innolitics.com.
>
> Innolitics is a medical-device software consultancy. We help companies design, build, and clear FDA-regulated software and AI/ML devices, including [a PMA](https://innolitics.com/services/regulatory/), [a 510(k)](https://innolitics.com/services/510ks/), [a SaMD](https://innolitics.com/services/end-to-end-samd/), [an AI/ML medical device](https://innolitics.com/services/medical-imaging-ai-development/), or [an FDA regulatory strategy](https://innolitics.com/services/regulatory/).

{0}

# SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)

## I. GENERAL INFORMATION

Device Generic Name: Cardiac Resynchronization Therapy Defibrillator (CRT-D)

Device Trade Name: Cognis Cardiac Resynchronization High Energy Defibrillator Models N118, N119

(P010012/S165, approved 5/08/2008)

Livian Cardiac Resynchronization Therapy Defibrillators Models H220, H225, H227 and H229

(P010012/S154, approved 2/15/2008)

Contak Renewal 3 RF HE CRT-D Models H210, H215, H217, H219

(P010012/S031, approved 2/09/2005)

Applicant's Name and Address: Guidant Corporation, a wholly-owned subsidiary of Boston Scientific Corporation, Cardiac Rhythm Management 4100 Hamline Avenue North

St. Paul, Minnesota 55112-5798

Date of Panel Recommendation: March 18, 2010

Premarket Approval Application (PMA) Number: P0100012 / S230

Date of FDA Notice of Approval: September 16, 2010

Expedited: Granted expedited review status on January 11, 2010, because the device provides a specific public health benefit or meets the need of a well-defined patient population

The original PMA P010012, Contak CRT-D, was approved on May 2, 2002, and is indicated for:

patients who are at high risk of sudden cardiac death due to ventricular arrhythmias and who have moderate to severe heart failure (NYHA Class III/IV) including left ventricular dysfunction (EF ≤ 35%) and QRS duration ≥ 120 ms and remain symptomatic despite stable, optimal heart failure drug therapy.

Patient populations at high risk of sudden cardiac death due to ventricular arrhythmias include, but are not limited to, those with:

- Survival of at least one episode of cardiac arrest (manifested by the loss of consciousness) due to a ventricular tachyarrhythmia.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 1

7

{1}

- Recurrent, poorly tolerated sustained ventricular tachycardia (VT).

NOTE: The clinical outcome of hemodynamically stable, sustained-VT patients is not fully known. Safety and effectiveness studies have not been conducted.

- Prior myocardial infarction, left ventricular ejection fraction of ≤ 35%, and a documented episode of nonsustained VT, with an inducible ventricular tachyarrhythmia. Patients suppressible with IV procainamide or an equivalent antiarrhythmic (drug) have not been studied.

The SSED to support the indication is available on the CDRH website and is incorporated by reference here.

PMA supplement P010012 / S014, (Contak CD, Contak CD 2, Renewal, Renewal 3), was approved on October 21, 2003, and the indication stated above was further expanded to include:

- Patients who may benefit from prophylactic treatment due to a prior myocardial infarction and an ejection fraction ≤ 30%.

PMA supplement P010012 / S017, (Ventak AV, Prizm, Vitality, Vitality AVT, Ventak PRx, Ventak Mini, Contak CD, Renewal), was approved on February 6, 2004, and the indications were changed to:

Guidant (BSC) cardiac resynchronization therapy defibrillators (CRT-Ds) are intended to provide ventricular antitachycardia pacing and ventricular defibrillation for automated treatment of life threatening ventricular arrhythmias. Guidant CRT-Ds are also indicated for reduction of symptoms of moderate to severe heart failure (NYHA III/IV) in patients who remain symptomatic despite stable, optimal heart failure drug therapy, and have left ventricular dysfunction (EF ≤ 35%) and QRS duration ≥ 120 ms.

PMA supplement P010012 / S026, (Contak CD, Contak CD 2, Renewal, Renewal 3), was approved on September 14, 2004, and the indications were changed to:

Guidant (BSC) cardiac resynchronization therapy defibrillators (CRT-Ds) are indicated for patients with moderate to severe heart failure (NYHA III/IV) who remain symptomatic despite stable, optimal heart failure drug therapy and have left ventricular dysfunction (EF ≤ 35%) and QRS duration ≥ 120 ms.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 2

8

{2}

The current supplement was submitted to expand the indications for use for the Cognis, Livian, and Contak Renewal 3 RF HE CRT-D's to include patients with left bundle branch block (LBBB) with QRS ≥ 130 ms, EF ≤ 30%, and mild (NYHA Class II) ischemic or nonischemic heart failure or asymptomatic (NYHA Class I) ischemic heart failure

# II. INDICATIONS FOR USE

These Boston Scientific Cardiac Resynchronization Therapy Defibrillators (CRT-Ds) are indicated for patients with heart failure who receive stable optimal pharmacologic therapy (OPT) for heart failure and who meet any one of the following classifications:

- Moderate to severe heart failure (NYHA Class III-IV) with \(\mathrm{EF} \leq 35\%\) and QRS duration \(\geq 120\) ms
- Left bundle branch block (LBBB) with QRS \(\geq 130\) ms, EF \(\leq 30\%\), and mild (NYHA Class II) ischemic or nonischemic heart failure or asymptomatic (NYHA Class I) ischemic heart failure

# III. CONTRAINDICATIONS

There are no contraindications for this device.

# IV. WARNINGS AND PRECAUTIONS

The warnings and precautions can be found in the Cognis, Livian, and Contak Renewal 3 RF HE labeling (System Guide).

# V. DEVICE DESCRIPTION

There were no changes to the device description as compared to the previously approved devices. The following paragraphs provide a brief description of the devices.

These devices are implantable pulse generators, designed to sense electrical activity and to deliver electrical pulses. These devices are commonly referred to as implantable cardioverter defibrillators (ICD) that also delivery cardiac resynchronization therapy (CRT) and are commonly refer to as CRT-D (cardiac resynchronization therapy defibrillators). These devices provide ventricular tachyarrhythmia and cardiac resynchronization therapies. Ventricular tachyarrhythmia therapy is for the treatment of ventricular tachycardia (VT) and ventricular fibrillation (VF), rhythms that are associated with sudden cardiac death (SCD). Cardiac resynchronization therapy is for the treatment of heart failure (HF) and uses biventricular electrical stimulation to synchronize ventricular contractions. The devices also use accelerometer-based adaptive-rate bradycardia therapy. The pulse generators accept one IS-1 atrial lead, one LV-1 or one IS-1 coronary venous pace/sense lead, and one DF-1/IS-1 cardioversion / defibrillation lead.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 3

9

{3}

Cardioversion/defibrillation therapies include a range of low- and high-energy shocks using either a biphasic or monophasic waveform. The devices use the Triad electrode system for defibrillation energy delivery. By using the metallic housing of the pulse generator as an active electrode, combined with a two-electrode defibrillation lead, energy is sent via a dual-current pathway from the distal shocking electrode to the proximal electrode and to the pulse generator case. The devices also offer a wide variety of antitachycardia pacing schemes to terminate slower, more stable ventricular tachyarrhythmias. Bradycardia pacing with cardiac resynchronization therapy, including adaptive-rate features, is available to detect and treat bradyarrhythmias and to support the cardiac rhythm after defibrillation therapy.

## VI. ALTERNATIVE PRACTICES AND PROCEDURES

There are several other alternatives for the treatment of heart failure patients who have left ventricular dysfunction. These patients are routinely treated with medications. Medications may include those to treat arrhythmias as well as medications to treat heart failure. Additional medical treatments for heart failure include, but are not limited to, exercise and nutrition programs. Alternative therapies for treatment of ventricular arrhythmias, as deemed appropriate by the physician based upon electrophysiological testing and other diagnostic evaluation, include antiarrhythmic medication, electrical ablation, cardiac surgery, and electronic devices including pacemakers and other legally marketed implantable cardioverter defibrillators (ICD) systems, or a combination thereof.

Each alternative has its own advantages and disadvantages. A patient should fully discuss these alternatives with his/her physician to select the method that best meets expectations and lifestyle.

## VII. MARKETING HISTORY

Boston Scientific's Cognis, Livian, and Contak Renewal 3 RF HE CRT-Ds are currently available for commercial distribution in the U.S. and other countries including: Australia, Austria, Belgium, Canada, Chile, Czech Republic, Denmark, Dominican Republic, Finland, France, Germany, Greece, Guadeloupe, Guyana, Hong Kong, Iceland, India, Indonesia, Ireland, Israel, Italy, Jordan, Kuwait, Lebanon, Liechtenstein, Luxembourg, Malaysia, Martinique, Netherlands, New Caledonia, New Zealand, Norway, Portugal, San Marino, Saudi Arabia, Singapore, Slovenia, South Africa, Spain, Sweden, Switzerland, Thailand, Turkey, United Kingdom, and Venezuela. As of August 4, 2010, no Cognis, Livian, or Contak Renewal 3 RF HE CRT-Ds are currently withdrawn for safety issues from the market in any country.

## VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH

Below is a list of the potential adverse effects (e.g., observations and complications) associated with the use of the devices.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 4

10

{4}

The potential risks related to this study include those typical of implantation of a CRT-D system and those associated with the protocol. Based on published literature and pacemaker/lead implant experience, the following list includes possible physical effects from implantation of a CRT-D system:

- Air embolism
- Allergic reaction (e.g. titanium)
- Bleeding
- Cardiac tamponade
- Chronic nerve damage
- Component failure
- Conductor coil fracture
- Death
- Electrolyte imbalance/dehydration
- Elevated thresholds
- Erosion
- Excessive fibrotic tissue growth
- Extracardiac stimulation (muscle/nerve stimulation)
- Failure to convert an induced arrhythmia
- Foreign body rejection phenomena
- Formation of hematomas or seromas
- Inability to defibrillate or pace
- Inappropriate therapy (e.g., shocks where applicable, ATP, pacing)
- Incisional pain
- Incomplete lead connection with pulse generator
- Infection
- Insulating myocardium during defibrillation with internal or external paddles
- Lead dislodgment
- Lead fracture
- Lead insulation breakage or abrasion
- Lead tip deformation and/or breakage
- Myocardial infarction (MI)
- Myocardial necrosis
- Myocardial trauma (e.g., cardiac perforation, irritability, injury)
- Myocardial sensing
- Oversensing/undersensing
- Pacemaker-mediated tachycardia (PMT)
- Pericardial rub, effusion
- Pneumothorax
- Pulse generator migration
- Shunting current during defibrillation with internal or external paddles

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 5

11

{5}

- Tachyarrhythmias, which include acceleration of arrhythmias and early, recurrent atrial fibrillation
- Thrombosis/thromboemboli
- Valve damage
- Venous occlusion
- Venous trauma (e.g. perforation, dissection, erosion)
- Worsening heart failure

Patients may develop psychological intolerance to a pulse generator system that may include the following:

- Dependency
- Depression
- Fear of premature battery depletion
- Fear of shocking while conscious
- Fear that shocking capability may be lost
- Imagined shocking

In addition to the implantation of a pulse generator system, potential adverse events associated with implantation of a coronary venous lead system include:

- Allergic reaction to contrast media
- Breakage/failure of implant instruments
- Prolonged exposure to fluoroscopic radiation
- Renal failure from contrast media used to visualize coronary veins

These risks can be minimized through use of strict aseptic technique, compliance with technical implant procedures, adherence to the guidelines for selection of subjects, and close monitoring of the subject's physiologic status during the implant and follow-up procedures.

For the specific adverse events that occurred in the clinical studies, please see Section X below.

# IX. SUMMARY OF PRECLINICAL STUDIES

The Cognis, Livian, and Renewal 3 RF HE CRT-D's are commercially available systems. These systems were previously evaluated via non-clinical laboratory testing including bench testing (including hardware/software verification and validation), biocompatibility testing, and animal studies. Device design and system compatibility involved verification and validation of the system. The test procedures and results were previously reviewed and approved.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 6

12

{6}

These issues were reviewed in previous submissions as follows:

- Cognis Cardiac Resynchronization High Energy Defibrillator Models N118, N119 (P010012/S165, approved 5/08/2008)
- Livian Cardiac Resynchronization Therapy Defibrillators Models H220, H225, H227 and H229 (P010012/S154, approved 2/15/2008)
- Contak Renewal 3 RF HE CRT-D Models H210, H215, H217, H219 (P010012/S031, approved 2/09/2005)

# X. SUMMARY OF PRIMARY CLINICAL STUDY

Boston Scientific sponsored the Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy (MADIT-CRT) clinical study to demonstrate the safety and effectiveness of Boston Scientific CRT-Ds in heart failure patients with QRS ≥ 130 ms, EF ≤ 30%, and mild (NYHA Class II) ischemic or nonischemic heart failure or asymptomatic (NYHA Class I) ischemic heart failure. MADIT-CRT was a prospective, randomized, controlled, global multicenter study conducted at 110 investigational centers.

The primary safety and effectiveness endpoints were met. The primary effectiveness endpoint demonstrated a statistically significant reduction in the combined endpoint of all-cause mortality and heart failure events in the CRT-D group as compared to the ICD group. This reduction was driven entirely by heart failure events; there was no difference in mortality between the groups. The results were also analyzed by various subgroups that were pre-specified by the sponsor at the onset of the trial. However, an additional analysis, which was not pre-specified, looked at the subgroup of patients with left bundle branch block (LBBB) morphology on their ECG as compared to non-LBBB patients (including right bundle branch block (RBBB) and non-specific interventricular conduction delay).

Although the analysis of LBBB was post-hoc, the results were consistent across a variety of other demographic and clinical variables. In addition, the subgroup analysis of patients with LBBB was compelling and consistent with previous observations regarding LBBB and cardiac resynchronization therapy. LBBB patients had a greater reduction in heart failure events as compared to non-LBBB patients. Further discussions resulted in a restriction in the final indications for use to those patients with LBBB. As a result, the following clinical summary refers to both the full MADIT-CRT population and to the cohort of LBBB patients enrolled in the MADIT-CRT clinical study. Additional information about the rationale for this restriction to LBBB patients is provided in both the advisory panel and conclusion sections of this document.

A total of 1820 patients were enrolled and randomized in a 3:2 ratio to receive CRT-D (1089) or ICD (731). Of the 1820 patients, 1281 (70.4%) had LBBB; 761 received CRT D and 520 received ICD. Randomization was stratified by clinical center and ischemic status. Each randomized patient remained counted as a member of the original randomized assignment (intention-to-treat) regardless of subsequent crossover or protocol adherence.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 7

13

{7}

# A. Study Design

Patients were implanted from December 22, 2004 through April 23, 2008. The database for this PMA supplement reflected data collected through December 31, 2009 and included 1820 patients. There were 110 investigational sights.

MADIT-CRT was a prospective, multi-center, randomized clinical study conducted in the United States (US), Europe, Canada, and Israel. Patients were randomized in a 3:2 ratio to receive either a CRT-D or implantable cardioverter defibrillator (ICD). Randomization was stratified by clinical center and ischemic status.

The study design was a Wang-Tsiatis group-sequential design, with two-sided significance level of 5% with 95% power to detect a hazard ratio of 0.75. The study was designed to allow for one of the following conclusions (A) CRT-D is superior, (B) ICD is superior, or (C) there is no real difference between the two treatment regimens. However, this design did not allow early stopping for a conclusion of no difference.

MADIT-CRT used an Executive Committee, Data Safety Monitoring Board (DSMB), Heart Failure Event Committee, and Mortality Event Review Committee (MERC) for study strategy, safety, heart failure event adjudication, and mortality adjudication, respectively.

Data continued to be collected and events were adjudicated until December 31, 2009.

# 1. Clinical Inclusion and Exclusion Criteria

# Inclusion Criteria

Patients who met the following inclusion criteria were given consideration for inclusion in the MADIT-CRT clinical investigation:

- NYHA Class I or II patients for the three calendar months prior to and at the time of enrollment with ischemic heart disease defined as:
- one or more clinically documented (q wave or enzyme positive) prior myocardial infarction, but not within three calendar months of enrollment and/or
- one or more prior coronary artery bypass graft surgeries or percutaneous coronary intervention (balloon and/or stent angioplasty), but not within three calendar months of enrollment;

OR

- NYHA Class II patients for the three calendar months prior to and at the time of enrollment with non-ischemic heart disease defined as including dilated cardiomyopathy characterized by a low ejection fraction and increased ventricular volume, with ventricular compliance that is normal or increased;

AND all of the following:

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 8

14

{8}

- Stable Optimal Pharmacological Therapy (OPT) (including ACEs (Angiotensin Converting Enzymes) / ARBs (Angiotensin Receptor blockers), Beta Blockers Diuretics). Ischemic patients were required to have statin therapy.
- An ejection fraction ≤ 0.30 by angiographic, radionuclide, or echocardiographic methods within one year prior to enrollment and measured during the enrollment echocardiogram
- Resting QRS duration ≥ 130 ms on print-out of a current ECG using a market-approved electrocardiographic recorder
- Sinus rhythm by ECG (including RBBB and first degree heart block with PR< 250 ms.)
- Men and women 21 years of age or older (no upper-age cut off)

# Exclusion Criteria

Patients were excluded from the MADIT-CRT clinical investigation if any of the following conditions applied:

- Existing indication for CRT therapy
- Implanted pacemaker
- Existing ICD or CRT device
- NYHA Class I with non-ischemic cardiomyopathy
- NYHA Class III or IV in the past three calendar months prior to or at the time of enrollment
- Coronary artery bypass graft surgery or percutaneous coronary intervention (balloon and/or stent angioplasty) within the past three calendar months prior to enrollment
- Enzyme-positive myocardial infarction within the past three calendar months prior to enrollment
- Angiographic evidence of coronary disease who were candidates for coronary revascularization and are likely to undergo coronary artery bypass graft surgery or percutaneous coronary intervention in the foreseeable future
- Second or third degree heart block
- Irreversible brain damage from preexisting cerebral disease
- Pregnant or plan to become pregnant during the course of the study (Note: Women of childbearing potential must have had a negative pregnancy test within 7 days prior to enrollment)
- Reversible non-ischemic cardiomyopathy such as acute viral myocarditis or discontinuation of alcohol in alcohol-induced heart disease
- Chronic atrial fibrillation within one month prior to enrollment
- Presence of any disease, other than the patient's cardiac disease, associated with a reduced likelihood of survival for the duration of the study, e.g., cancer, uremia (BUN>70mg/dl or creatinine >3.0mg/dl), liver failure, etc
- Participating in any other clinical studies
- Unwilling or unable to cooperate with the protocol

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 9

15

{9}

- Live at such a distance from the clinic that travel for follow-up visits would be unusually difficult
- Did not anticipate being a resident of the area for the scheduled duration of the study
- Unwilling to sign the consent for participation

## 2. Follow-up Schedule

In the MADIT-CRT study 1820 patients were randomized to CRT-D or ICD and enrolled between December 22, 2004 and April 23, 2008. Table 1 provides a summary of the follow-up timeline and procedures performed.

Table 1: Follow-Up Schedule

|  Follow-Up | Description  |
| --- | --- |
|  Screening | Initial assessment of patient eligibility; taking of patient history  |
|  Pre-randomization confirmation testing | ECG and echocardiogram  |
|  Randomization | Randomization status (CRT-D or ICD) was assigned  |
|  Baseline Testing | Six-minute walk, Holter, QoL, BNP (US only)  |
|  Routine clinic follow-ups | 1, 3, and every 3 months  |
|  Special testing | Echocardiogram, 6-minute walk, Holter, BNP (US only), QoL (every 6 months)  |

As of December 31, 2009, the total patient follow-up months were 62,335 months: 24,683 in the ICD group and 37,653 in the CRT-D group. The mean follow up duration was 34.3±12.2 months; 33.8±12.9 months in the ICD arm and 34.6±11.7 months in the CRT-D arm. There was no statistically significant difference in the follow-up duration between treatment arms. The follow-up duration details are summarized below in Table 2.

Table 2: Follow-up Duration (all patients randomized, N=1820)

|  Measurement | Follow-up Duration (months) | ICD (N = 731) | CRT-D (N = 1089)  |
| --- | --- | --- | --- |
|  Mean ± SD | 34.3 ± 12.2 | 33.8 ± 12.9 | 34.6 ± 11.7  |
|  Range | 0.03 - 58.97 | 0.03 - 58.94 | 0.03 - 58.97  |
|  Total Patient Months | 62,335 | 24,683 | 37,653  |

### 3. Clinical Endpoints

#### Primary Study Objectives

Safety Endpoint: Determine if the CRT-D system-related complication-free rate observed was greater than 70% after three months of follow-up post-implant

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 10

16

{10}

**Primary Effectiveness Endpoint:** Determine whether CRT-D resulted in a statistically significant reduction in the combined endpoint of all-cause mortality or heart failure event, whichever came first, when compared to ICD.

All deaths were reviewed and adjudicated by the Mortality Event Committee. All heart failure events were reviewed and adjudicated by the Heart Failure Event Committee and the members were blinded to the randomized therapy.

# Additional Objectives

**Secondary:** Evaluate the effects of CRT-D, relative to ICD, on the patient-specific rates of recurrent heart failure events over the full study period.

**Tertiary:** Evaluate the effects of CRT-D on:

All-cause mortality
- Appropriate defibrillator therapy for ventricular tachycardia (VT) and/or ventricular fibrillation (VF)
Changes in echocardiographic structure and function at 12 months (echo-determined left ventricular internal volume at end-systole and diastole (LVESV and LVEDV) and changes in LVEF at 12 months)
Changes in NYHA functional class at 12 months
Changes in quality of life at 12 months and over the full study period
- Mitral regurgitation at 12 months
Functional capacity (six minute hall walk) at 12 months
The association between BNP and outcome with CRT-D
- Brain natriuretic peptide (BNP) levels at 12 months
- Holter-recorded electrocardiographic parameters and hemodynamic response

# Safety Endpoint

The MADIT-CRT study assessed the CRT-D safety by the system-related complication-free rate observed within the date of implant and three months of follow-up. For the purposes of the safety analysis, three month follow-up was defined as 91 days post-implant.

Hypothesis:

H₀: The system-related complication-free rate ≤ 70%

Hₐ: The system-related complication-free rate > 70%.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 11

17

{11}

# Primary Effectiveness Endpoint

The MADIT-CRT study assessed the effectiveness of CRT-D by the relative reduction of the risk of the combined endpoint of all-cause mortality or HF event, whichever occurred first, when compared to ICD.

The primary effectiveness analysis was based on comparing the Kaplan-Meier life-table event-free survival time graphs for the CRT-D and ICD groups. The stratified log-rank test (stratified by clinical center and ischemic status) was used to evaluate statistical significance, adjusting for the group-sequential stopping rule of the study.

# Hypothesis:

Ho: The event-free survival curves for the combined endpoint of HF event or all-cause mortality do not differ between the ICD and CRT-D groups.

Ha: The event-free survival curve for the CRT-D group is above that for the ICD group, with a hazard ratio less than unity.

# Secondary Endpoint

The MADIT-CRT study also evaluated the effects of CRT-D, relative to ICD, on the patient-specific rates of multiple HF events over the full study period.

# Hypothesis:

Ho: The HF event rates do not differ between the ICD and CRT-D groups.

Ha: The HF event rate for the CRT-D group is less than that for the ICD group, with an average-rate ratio less than unity.

# Tertiary Endpoints

The trial included ten tertiary, exploratory objectives as follows:

- Evaluate the effects of CRT-D on all-cause mortality.
- Evaluate the effects of CRT-D on appropriate ICD therapy for ventricular tachycardia (VT) and ventricular fibrillation (VF).
- Evaluate the effects of CRT-D, relative to ICD-only, on the changes from baseline to one year in ECHO-determined left ventricular internal volume at end systole (LVESV) and at end diastole (LVEDV).
- Evaluate the effects of CRT-D, relative to ICD-only, on the changes from baseline to one year in NYHA functional class. It is hypothesized that, at one year, the average NYHA class for the CRT-D group will be lower than that for the ICD-only group, after adjusting for any differences in baseline values.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 12

18

{12}

- Evaluate the effects of CRT-D, relative to ICD-only, on the accumulated changes in quality-of-life within the full study period. It is hypothesized that the assessed quality of life in the CRT-D group will, on average, exceed that in the ICD-only group.
- Evaluate by echocardiographic/Doppler technique at the 12 month follow-up whether CRT-D when compared to subjects receiving ICD-only reduces the degree of mitral regurgitation (echocardiographic subprotocol).
- Evaluate whether functional capacity (as measured by distance achieved during a 6 minute hall walk) at the 12 month follow-up is greater in subjects receiving CRT-D than in those receiving ICD-only.
- Evaluate the association between the level of brain natriuretic peptide (BNP) at baseline and outcome in subjects randomized to CRT-D.
- Evaluate whether the level of brain natriuretic peptide (BNP) at the 12-month follow-up visit is lower in the CRT-D group than the ICD-only group.
- Evaluate whether Holter-recorded non-invasive parameters can identify subjects with increased hemodynamic benefit in CRT responders and non-responders.

# B. Accountability of PMA Cohort

Two data sets were used in reporting MADIT-CRT results to FDA. The first was based on a data cutoff of June 22, 2009. At that time, Boston Scientific announced that the MADIT-CRT study met its primary effectiveness endpoint. Based on this data cutoff, a data set was delivered to the sponsor by the data coordinating center on September 30, 2009. A report was prepared and submitted to FDA on December 10, 2009. The results from this data set were also presented to the cardiovascular devices advisory panel on March 18, 2010.

Meanwhile, a second data set was created that updated the results with events that occurred on or before December 31, 2009. Heart failure events were adjudicated and a data set delivered to the sponsor on February 1, 2010 by the data coordinating center. These additional data have been incorporated and the results from the December 10, 2009 report have been updated. A timeline illustrating the timing of data cutoffs and significant events in the regulatory history are shown below.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 13

19

{13}

![img-0.jpeg](img-0.jpeg)

# C. Study Population Demographics and Baseline Parameters

Key information related to enrollment, patient demographics, and the primary endpoints is represented below in Table 3. Table 3 includes a brief summary of the demographics for both the full MADIT-CRT population and the MADIT-CRT LBBB sub-population. Tables 4 through 9 provide more detailed information about the full demographics of the randomized CRT-D and ICD groups from the full MADIT-CRT study, in order to provide a more complete picture of the original supporting study and to demonstrate the similarities of the two (2) randomized groups of patients.

Table 3: Demographic Data (All MADIT-CRT Patients and LBBB Patients)

|  Data Item | Result (All Patients) | Result (LBBB Only)  |
| --- | --- | --- |
|  Number of enrolled patients | 1820 | 1281  |
|  Number of randomized patients (CRT-D/ICD) | 1089 / 731 | 761 / 520  |
|  Number of implanted patients (CRT-D/ICD) | 1078 / 712 | 757 / 507  |
|  Number of attempted patients (CRT-D/ICD) | 1 / 0 | 1 / 0  |
|  Number of intent patients (CRT-D/ICD) | 10 / 19 | 3 / 13  |
|  Mean follow-up time (± SD) | 34.3 ± 12.2 months | 34.8 ± 12.3 months  |
|  Implant phase | 01/05/05 - 05/05/08 | 01/05/05 - 05/05/08  |
|  Number of centers | 110 | 109†  |
|  Patient Demographics | Result | Result  |
|  Gender | 75 % Male | 69 % Male  |
|   |  25 % Female | 31 % Female  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 14

20

{14}

|  Data Item | Result (All Patients) | Result (LBBB Only)  |
| --- | --- | --- |
|  NYHA Classification | 15 % Class I Ischemic 40 % Class II Ischemic 45 % Class II Non-Ischemic | 11 % Class I Ischemic 33 % Class II Ischemic 56 % Class II Non-Ischemic  |
|  Mean Age (± SD) | 64 ± 11 years | 64 ± 11 years  |
|  Mean LVEF (± SD) | 24 ± 5 % | 24 ± 5 %  |
|  Mean QRS (± SD) | 158 ± 20 ms | 163 ± 19 ms  |
|  Endpoint Summary | Result | Result  |
|  Primary Safety System-Related Complication Free Rate (Lower One-Sided 95% Confidence Bound) | 84.8 (82.9) | 83.4 (81.0)  |
|  Primary Effectiveness Hazard Ratio (95% Confidence Interval) | 0.61 (0.50, 0.75); p<0.001* | 0.43 (0.33, 0.56); < 0.001*  |

† One (1) center enrolled no patients with LBBB morphology

* Hazard ratio adjusted for ischemic status and center

The general characteristics of the 1820 patients randomized in MADIT-CRT are presented in the following six tables (Tables 4 through 9) according to assigned randomization, including baseline demographics, cardiac history, cardiac risk factors, cardiac findings at enrollment, baseline echocardiographic volumes and cardiac medications. Patient characteristics were well-balanced between the therapy groups. While statistically significant differences were found between diastolic and systolic blood pressure, these differences are not considered clinically meaningful. The mean age of the patients was 64±11 years, 75% of the patients were male, and 90% of the patients reported their race as white. A majority of the patients had ischemic heart disease (55%), 85% of the patients were classified as NYHA Class II, 71% of the patients had a left bundle branch block, and 62% of the patients had never been hospitalized for heart failure prior to enrollment. MADIT-CRT patients had a mean LVEF of 24±5% and a mean QRS of 158±20 ms. The majority of patients were medicated on heart failure drugs: 96% of the patients were on an angiotensin converting enzyme (ACE) or angiotensin receptor blockers (ARB); 93% were on a beta blocker; 75% were on a diuretic; 32% were on an aldosterone antagonist; and 67% were on a statin.

Table 4: Baseline Demographics (all patients randomized, N=1820) - Full Patient Population

|  Characteristic | Measurement | ICD (N=731) | CRT-D (N=1089) | P-value  |
| --- | --- | --- | --- | --- |
|  Age at Implant (years) | N | 731 | 1089 |   |
|   | Mean ± SD | 64 ± 11 | 64 ± 11 | 0.74  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 15

21

{15}

|  Characteristic | Measurement | ICD (N=731) | CRT-D (N=1089) | P-value  |
| --- | --- | --- | --- | --- |
|   | Range | 32 - 88 | 25 - 90 |   |
|  Gender [N (%)] | Female | 178 (24.4) | 275 (25.3) | 0.66  |
|   | Male | 553 (75.6) | 814 (74.7) |   |
|  Race [N (%)] | White | 658 (90.8) | 980 (90.4) |   |
|   | Black/African American | 56 (7.7) | 87 (8.0) | 0.97  |
|   | Other* | 11 (1.5) | 17 (1.6) |   |

* Other races include Asian, American Indian/Alaska Native, Native Hawaiian or other Pacific Islander, and more than one race

Table 5: Cardiac History (all patients randomized, N=1820) - Full Patient Population

|  Characteristic | Measurement | ICD (N=731) | CRT-D (N=1089) | P-value  |
| --- | --- | --- | --- | --- |
|  NYHA Class/Ischemic [N (%)] | Class I Ischemic | 113 (15.5) | 152 (14.0) | 0.63  |
|   | Class II Ischemic | 288 (39.4) | 446 (41.0) |   |
|   | Class II Non-Ischemic | 330 (45.1) | 491 (45.1) |   |
|  Worst NYHA class > 3 mos prior to enrollment [N (%)]* | Class III/IV | 73 (10.4) | 109 (10.4) | 0.99  |
|  Number of CHF Hospitalizations Prior to Enrollment [N (%)] | None | 451 (63.3) | 656 (61.3) | 0.69  |
|   | 1 – 2 | 231 (32.4) | 365 (34.1) |   |
|   | 3 or more | 31 (4.3) | 50 (4.7) |   |

* Patients who were Class I or II at enrollment but had prior history of Class III or IV more than 3 months prior to enrollment.

Table 6: Cardiac Risk Factors (all patients randomized, N=1820) - Full Patient Population

|  Risk Factor | ICD (N=731) [N (%)] | CRT-D (N=1089) [N (%)] | P-value  |
| --- | --- | --- | --- |
|  Treatment for Hypertension | 461 (63.2) | 691 (63.7) | 0.82  |
|  Atrial fibrillation > 1 month before enrollment | 90 (12.3) | 118 (10.8) | 0.33  |
|  Diabetes Mellitus | 223 (30.6) | 329 (30.2) | 0.87  |
|  Cigarette Smoking | 92 (12.8) | 122 (11.4) | 0.37  |
|  Body-mass index >= 30 | 256 (35.4) | 378 (35.2) | 0.95  |
|  Coronary-bypass surgery | 208 (28.5) | 317 (29.1) | 0.77  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 16

22

{16}

Table 7: Cardiac Findings at Enrollment (all patients randomized, N=1820) – Full Patient Population

|  Characteristic | Measurement | ICD (N=731) | CRT-D (N=1089) | P-value  |
| --- | --- | --- | --- | --- |
|  Conduction | LBBB [N (%)] | 520 (71.3) | 761 (69.9) | 0.51  |
|   | RBBB [N (%)] | 92 (12.6) | 136 (12.5) | 0.93  |
|   | IVCD [N (%)] | 111 (15.2) | 182 (16.7) | 0.40  |
|  Systolic Blood Pressure (mm Hg) | N | 719 | 1074 |   |
|   | Mean ± SD | 121 ± 18 | 124 ± 17 | 0.001*  |
|   | Range | 80 - 193 | 78 - 194 |   |
|  Diastolic Blood Pressure (mm Hg) | N | 719 | 1074 |   |
|   | Mean ± SD | 71 ± 10 | 72 ± 10 | 0.002*  |
|   | Range | 37 - 107 | 40 - 110 |   |
|  BUN (blood urea nitrogen) >= 26 mg/dL | N (%) | 173 (24.0) | 254 (23.5) | 0.79  |
|  Creatinine (mg/dl) | N | 725 | 1083 |   |
|   | Mean ± SD | 1.2 ± 0.4 | 1.2 ± 0.4 | 0.51  |
|   | Range | 0.5 - 7.2 | 0.4 - 5.6 |   |
|  QRS duration >= 150 ms | N (%) | 476 (65.1) | 699 (64.2) | 0.68  |
|  LVEF (%) | N | 731 | 1089 |   |
|   | Mean ± SD | 24 ± 5 | 24 ± 5 | 0.33  |
|   | Range | 6 - 32 | 7 - 35 |   |
|  BNP (pg/ml) | N | 473 | 724 |   |
|   | Mean ± SD | 114 ± 141 | 132 ± 173 | 0.06  |
|   | Range | 1 - 1209 | 1 - 1433 |   |
|  6 Minute Walk Distance (meters) | N | 696 | 1069 |   |
|   | Mean ± SD | 363 ± 108 | 358 ± 106 | 0.44  |
|   | Range | 31 - 896 | 0 - 744 |   |
|  Euro Qol Index | N | 726 | 1086 |   |
|   | Mean ± SD | 0.84 ± 0.13 | 0.84 ± 0.14 | 0.58  |
|   | Range | 0.27 - 1.00 | -0.04 - 1.00 |   |
|  KCCQ Overall Summary Score | N | 727 | 1087 |   |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 17

23

{17}

|  Characteristic | Measurement | ICD (N=731) | CRT-D (N=1089) | P-value  |
| --- | --- | --- | --- | --- |
|   | Mean ± SD | 75 ± 19 | 76 ± 18 | 0.43  |
|   | Range | 10 - 100 | 17 - 100 |   |
|  KCCQ Clinical Summary Score | N | 727 | 1087 |   |
|   | Mean ± SD | 80 ± 18 | 80 ± 17 | 0.66  |
|   | Range | 4 - 100 | 16 - 100 |   |
|  KCCQ Quality of Life Score | N | 727 | 1087 |   |
|   | Mean ± SD | 66 ± 25 | 67 ± 23 | 0.84  |
|   | Range | 0 - 100 | 0 - 100 |   |

* Statistically significant differences were found between systolic and diastolic blood pressure; these differences are not considered clinically meaningful.

Table 8: Baseline Echocardiographic Volumes (all patients randomized, N=1820) – Full Patient Population

|  Characteristic | Measurement | ICD (N=731) | CRT-D (N=1089) | P-value  |
| --- | --- | --- | --- | --- |
|  Left ventricular end-systolic volume (ml) | N | 724 | 1085 |   |
|   | Mean ± SD | 180 ± 52 | 176 ± 48 | 0.10  |
|   | Range | 94 - 465 | 83 - 434 |   |
|  Left ventricular end-diastolic volume (ml) | N | 724 | 1085 |   |
|   | Mean ± SD | 251 ± 65 | 246 ± 60 | 0.10  |
|   | Range | 134 - 601 | 134 - 564 |   |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 18

24

{18}

Table 9: Cardiac Medications (all patients randomized, N=1820) - Full Patient Population

|  Medication | ICD (N=731) [N (%)] | CRT-D (N=1089) [N (%)] | P-value  |
| --- | --- | --- | --- |
|  ACE (Angiotensin Converting Enzyme) / ARB (Angiotensin Receptor blockers)* | 699 (95.6) | 1039 (95.4) | 0.83  |
|  Aldosterone Antagonist** | 226 (30.9) | 352 (32.3) | 0.53  |
|  Amiodarone** | 51 (7.0) | 78 (7.2) | 0.88  |
|  Beta Blockers* | 681 (93.2) | 1016 (93.3) | 0.91  |
|  Class I antiarrhythmic agent** | 3 (0.4) | 12 (1.1) | 0.11  |
|  Digitalis** | 177 (24.2) | 291 (26.7) | 0.23  |
|  Diuretic* | 533 (72.9) | 824 (75.7) | 0.19  |
|  Statin* | 491 (67.2) | 735 (67.5) | 0.88  |

* Required Optimal Pharmacologic Therapy (OPT), unless contraindicated. Ischemic patients were required to have a statin prescribed.

** Adjunctive medications per medical discretion.

# D. Safety and Effectiveness Results

# 1. Results: Safety Endpoint

The MADIT-CRT study assessed the safety of the Cognis, Livian, and Contak Renewal 3 RF HE CRT-D's by the system-related complication-free rate observed within the date of implant and three (3) months of follow-up. For the purposes of the safety analysis, three (3) month follow-up was defined as 91 days post-implant.

Data Analysis: All CRT-D patients (n=1079) who underwent an implant procedure were included in the primary safety analysis and were analyzed according to randomization assignment. A system-related complication that occurred within 91 days post-implant was included as an event in this analysis.

The hypothesis for the system-related complication-free rate was evaluated using the lower one-sided 95% confidence bound from the Kaplan-Meier estimated system-related complication-free rate. The rate reported was based on Kaplan-Meier estimates and the lower confidence bound was based on a log cumulative hazard transformation.

Patients who did not have a system-related complication within 91 days post-implant were censored at their date of death, withdrawal (if the patient did not agree to phone contact to collect event data), or at 92 days post-implant.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 19

25

{19}

Results: A total of 1079 patients were randomized to CRT-D and underwent an implant procedure. Of these, 164 unique patients experienced 214 system-related complications (SRC)s within 91 days post-implant. The CRT-D Kaplan-Meier system-related complication-free rate was 84.8% with a lower one-sided 95% confidence bound of 82.9%. This rate was statistically significantly greater than 70% and therefore passed the pre-specified safety endpoint. The value of 70% was selected based on previous studies designed to evaluate CRT-D devices. The Kaplan-Meier system-related complication-free graph and supporting data are shown below in Figure 1 and Table 10.

Figure 1: System-Related Complication-Free Rate Within 91 days -Full Patient Population

![img-1.jpeg](img-1.jpeg)

Table 10: System-Related Complication-Free Summary Data - Full Patient Population

|  Statistic | Days from Implant  |   |   |
| --- | --- | --- | --- |
|   |  0-30 Days | 31-60 Days | 61-91 Days  |
|  Number at Risk at Start of Interval | 1079 | 933 | 917  |
|  Number of Patients in Interval | 144 | 14 | 6  |
|  Cumulative Number of Patients | 144 | 158 | 164  |
|  Number Censored in Interval | 2 | 2 | 0  |
|  Cumulative Number Censored | 2 | 4 | 4  |
|  Percent Free from Event | 86.6% | 85.3% | 84.8%  |
|  95% Lower Confidence Bound | 84.8% | 83.5% | 82.9%  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 20

26

{20}

There were 164 patients with system-related complications (214 total events). Of the 214 events, the cause of the event was related to the implant procedure in 84 events (39.3%), the LV lead in 62 events (29.0%), the RA lead in 35 events (16.4%), and the remaining 15.3% of events were related to the PG or RV lead. A summary of the CRT-D system-related complications that contribute to the safety endpoint is shown below in TableTable 11. The sum of patients across categories does not equal the total number of unique patients because some patients had more than one (1) System-Related Complication.

Table 11: CRT System-Related Complication-Free Summary Data - Full Patient Population

|  Complication | Number of Events | Number of Patients | Complication Free Rate (%) | Lower One-Sided 95% Confidence Bound (%)  |
| --- | --- | --- | --- | --- |
|  Procedure | 84 | 75 | 93.0 | 91.6  |
|  AV block | 6 | 6 | 99.4 | 98.9  |
|  Adverse reaction | 6 | 6 | 99.4 | 98.9  |
|  Hematoma – Pocket (<=30 days post-implant) | 14 | 14 | 98.7 | 98.0  |
|  Inadvertent VT/VF | 4 | 4 | 99.6 | 99.2  |
|  Other - Lead - Procedure | 5 | 5 | 99.5 | 99.0  |
|  Pericardial effusion | 4 | 3 | 99.7 | 99.3  |
|  Pneumothorax - Procedure | 15 | 15 | 98.6 | 97.9  |
|  Post-surgical infection (<= 30 days post-implant) | 5 | 5 | 99.5 | 99.0  |
|  Renal failure due to contrast media – Procedure | 4 | 4 | 99.6 | 99.2  |
|  Thromboembolic events | 8 | 8 | 99.3 | 98.7  |
|  Other Procedure * | 13 | 13 | 98.8 | 98.1  |
|  LV Lead | 62 | 57 | 94.7 | 93.5  |
|  Dislodgment | 51 | 46 | 95.7 | 94.6  |
|  Extracardiac stimulation - LV | 9 | 9 | 99.2 | 98.6  |
|  Other LV Lead ** | 2 | 2 | 99.8 | 99.4  |
|  PG ** | 16 | 16 | 98.5 | 97.8  |
|  RA Lead | 35 | 35 | 96.7 | 95.7  |
|  Dislodgment | 33 | 33 | 96.9 | 95.9  |
|  Other RA Lead ** | 2 | 2 | 99.8 | 99.4  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 21

27

{21}

|  Complication | Number of Events | Number of Patients | Complication Free Rate (%) | Lower One-Sided 95% Confidence Bound (%)  |
| --- | --- | --- | --- | --- |
|  RV Lead | 15 | 15 | 98.6 | 97.9  |
|  Dislodgment | 8 | 8 | 99.3 | 98.7  |
|  Elevated threshold - RV | 5 | 5 | 99.5 | 99.0  |
|  Other RV Lead ** | 2 | 2 | 99.8 | 99.4  |
|  Other *** | 2 | 2 | 99.8 | 99.4  |
|  Total System-Related Complications | 214 | 164 | 84.8 | 82.9  |

* Procedure related events that occurred three times or fewer: Arterial perforation - Procedure (1), Coronary venous perforation without tamponade (2), Inadvertent SVT (1), Myocardial perforation with tamponade (3), Other - PG system - Procedure (1), Pleural effusion - Procedure (1), Post-surgical pocket hemorrhage (2), Seroma - Pocket (<=30 days post-implant) (1), Venous occlusion (1).

** Device related events that occurred three times or fewer: Elevated threshold - LV (1), Insulation breach - LV (1), Early ERI - Random component failure (2), Elevated DFT - Defibrillation (3), Elevated threshold - RV (1), Extracardiac stimulation - LV (2), Inappropriate tachy therapy - Noise (1), Inappropriate tachy therapy - SVT (1), Infection (> 30 days post-implant) (3), Migration (1), Programmer / Software error code (1), Unable to convert - Defibrillation (1), Unable to capture - RA (2), Elevated DFT - Defibrillation lead (1), Unable to convert - Defibrillation lead (1).

*** Other events that occurred three times or fewer: Pulmonary edema - Heart failure (1), Systemic infection (1).

Results: Additional Safety Data

Cause of Death (All Patients)

Deaths for the full MADIT-CRT patient population are shown in Table 12 below. The most common causes of death in the study were pump failure (n=62, 38.8%) and non-cardiac causes (n=49, 30.6%), arrhythmic deaths (n=15, 9.4%), deaths from unknown causes (n=16, 10.0%).

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 22

28

{22}

Table 12: Cause of Death - Full Patient Population

|  Cause of Death Category | ICD (N=731) | CRT-D (N=1089) | Total (N=1820)  |
| --- | --- | --- | --- |
|  Cardiac: Pump failure | 30 (44.1%) | 32 (34.8%) | 62 (38.8%)  |
|  Non-Cardiac | 16 (23.5%) | 33 (35.9%) | 49 (30.6%)  |
|  Cardiac: Arrhythmic | 8 (11.8%) | 7 (7.6%) | 15 (9.4%)  |
|  Unknown | 7 (10.3%) | 9 (9.8%) | 16 (10%)  |
|  Cardiac: Ischemic | 2 (2.9%) | 6 (6.5%) | 8 (5%)  |
|  Cardiac: Other procedure | 2 (2.9%) | 0 (0%) | 2 (1.3%)  |
|  Cardiac: Other | 1 (1.5%) | 1 (1.1%) | 2 (1.3%)  |
|  Not yet classified | 2 (2.9%) | 4 (4.3%) | 6 (3.8%)  |
|  Total | 68 (100%) | 92 (100%) | 160 (100%)  |

# Cause of Death (Left Bundle Branch Block Patients Only)

The MADIT-CRT LBBB sub-population is the subset of patients enrolled in the MADIT-CRT study that have the same indications as those patients for which the new indication is being granted. The most common causes of death in the MADIT-CRT Left Bundle Branch Block (LBBB) sub-population were pump failure (n=41, 39.0%), non-cardiac causes (n=33, 31.4%), and arrhythmic deaths (n=10, 11.0%). Additionally, deaths from 12 patients (11.0%) were adjudicated and classified as unknown due to a lack of information from the investigational center. There were no statistically significant differences in the cause of death between treatment groups. The cause of death data are summarized below in Table 13.

Table 13: Cause of Death - LBBB Patient Sub-Population

|  Cause of Death Category | ICD (N=520) | CRT-D (N=761) | Total (N=1281)  |
| --- | --- | --- | --- |
|  Cardiac: Pump failure | 22 (43.1%) | 19 (35.2%) | 41 (39%)  |
|  Non-Cardiac | 16 (31.4%) | 17 (31.5%) | 33 (31.4%)  |
|  Cardiac: Arrhythmic | 6 (11.8%) | 4 (7.4%) | 10 (9.5%)  |
|  Unknown | 5 (9.8%) | 7 (13%) | 12 (11.4%)  |
|  Cardiac: Ischemic | 1 (2%) | 4 (7.4%) | 5 (4.8%)  |
|  Cardiac: Other | 1 (2%) | 1 (1.9%) | 2 (1.9%)  |
|  Not yet classified | 0 (0%) | 2 (3.7%) | 2 (1.9%)  |
|  Total | 51 (100%) | 54 (100%) | 105 (100%)  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 23

29

{23}

Of the 105 deaths in the LBBB patients, six (5.7%) were adjudicated as device-related (including procedure for device installation or maintenance) or possibly related to the device or procedure. Three occurred in the ICD group and three in the CRT-D group.

Of the 105 deaths in the LBBB patients, one occurred in a patient categorized as an intent (signed the consent and randomized, but never implanted) and five (4.8%) occurred prior to 91 days post implant. None of the deaths occurred within 24 hours of implant. Of the five patients that were implanted and then died within 91 days post-implant, one death occurred within 30 days of implant and four deaths occurred within 91 days post implant. The remaining 100 deaths occurred greater than 91 days post implant (95.2%).

# Adverse Event Definitions

Investigators were responsible for providing a description of each reported adverse event including the suspected cause, corrective actions, and clinical outcome. All adverse events reported by centers were reviewed and classified as described below.

Adverse events were defined as any untoward clinical event, including events that were not related to the implanted system. Adverse events were ranked by severity. Events that were life-threatening, required an invasive intervention, resulted in hospitalization, permanent loss of device therapy, permanent disability, or death were defined as complications. Those events that were transient and reversible and resolved non-invasively without hospitalization were defined as observations. In Europe (per ISO 14155) reportable adverse events were classified as serious (equivalent to complications) or non-serious (equivalent to observations).

# Patient-Related Adverse Events

In all patients, 5024 (81%) of the adverse events were related to patient condition rather than the implanted device, procedure or protocol. Of these, 2284 (45%) were non-cardiac related, 1711 (34%) were cardiac/non-heart failure related, and 1029 (20%) were cardiac/heart failure related.

In LBBB patients, 2161 (74%) of the adverse events were related to patient condition rather than the implanted device, procedure or protocol. Of these, 1091 (50%) were non-cardiac related, 762 (35%) were cardiac/non-heart failure related, and 308 (14%) were cardiac/heart failure related.

# Device-Related Adverse Events

In all patients, 1276 (29.5%) of the adverse events were related to the device, implant procedure or protocol related testing. A summary of the device-related adverse events for the CRT-D treatment group is provided below in Table 14 for all patients and in Table 15 for LBBB patients only.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 24

30

{24}

Table 14: Device-Related Adverse Events by System Component (Full Patient Population)

|  Component | Number Of Events (% of events) | Number Of Patients (% of patients with events)*  |
| --- | --- | --- |
|  Pulse Generator Related | 568 (44.5%) | 409 (57%)  |
|  Procedure Related | 353 (27.7%) | 283 (39.4%)  |
|  Left Ventricular Lead Related | 174 (13.6%) | 149 (20.8%)  |
|  Right Atrial Lead Related | 94 (7.4%) | 81 (11.3%)  |
|  Right Ventricular Lead Related | 81 (6.3%) | 73 (10.2%)  |
|  Protocol-mandated Testing Related | 6 (0.5%) | 6 (0.8%)  |
|  Total Device/Procedure Adverse Events | 1276 (100%) | 718 (100%)  |

*Patients can be in more than one category so percent of patients with events does not equal 100%

Table 15: Device-Related Adverse Events by System Component (LBBB Patient Sub-Population)

|  Component | Number Of Events (% of events) | Number Of Patients (% of patients with events)*  |
| --- | --- | --- |
|  Pulse Generator Related | 390 (43.0%) | 286 (57.5%)  |
|  Procedure Related | 262 (28.9%) | 208 (41.9%)  |
|  Left Ventricular Lead Related | 129 (14.2%) | 106 (21.3%)  |
|  Right Atrial Lead Related | 66 (7.3%) | 57 (11.5%)  |
|  Right Ventricular Lead Related | 54 (6.0%) | 49 (9.9%)  |
|  Protocol-mandated Testing Related | 5 (0.6%) | 5 (1.0%)  |
|  Total Device/Procedure Adverse Events | 906 (100%) | 497 (100%)  |

* Patients can be in more than one category so percent of patients with events does not equal 100%.

# Summary of All Adverse Events

A summary of the observations and complications for the CRT-D treatment group is provided below in Table 16 for all patients. In order to provide a broad yet concise summary of the adverse events, a summary of all observations and complications for the LBBB patients is not provided here. Infrequent events (defined as observations occurring in fewer than four patients) are summarized at the end of each section within the table.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 25

31

{25}

Table 16: CRT-D Clinical Observation and Complication Summary (All Patients)

|   |   | Complications |   | Observations  |   |
| --- | --- | --- | --- | --- | --- |
|  Adverse Event | Total Number Of Events (Number of Patients) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events)  |
|  Total Adverse Events | 4323 (942) | 65.2 (704) | 5.39 (2014) | 71.6 (773) | 6.18 (2309)  |
|  Pulse Generator (PG) Related Events  |   |   |   |   |   |
|  Early elective replacement indicator | 44 (43) | 3.9 (42) | 0.12 (43) | 0.1 (1) | 0.00 (1)  |
|  Elevated Defibrillation Thresholds | 4 (4) | 0.3 (3) | 0.01 (3) | 0.1 (1) | 0.00 (1)  |
|  Erosion | 6 (6) | 0.6 (6) | 0.02 (6) | 0.0 (0) | 0.00 (0)  |
|  Extracardiac stimulation – Left Ventricular Lead | 127 (104) | 0.2 (2) | 0.01 (2) | 9.5 (102) | 0.33 (125)  |
|  Inappropriate tachyarrhythmia therapy | 33 (27) | 0.6 (7) | 0.02 (7) | 1.9 (21) | 0.07 (26)  |
|  Infection (> 30 days post-implant) | 12 (10) | 0.6 (7) | 0.02 (9) | 0.3 (3) | 0.01 (3)  |
|  Migration | 4 (4) | 0.1 (1) | 0.00 (1) | 0.3 (3) | 0.01 (3)  |
|  Other – PG System - Patient | 3 (3) | 0.1 (1) | 0.00 (1) | 0.2 (2) | 0.01 (2)  |
|  Oversensing – Right Atrial Lead | 13 (11) | 0.0 (0) | 0.00 (0) | 1.0 (11) | 0.03 (13)  |
|  Oversensing – Right Ventricular Lead | 5 (4) | 0.1 (1) | 0.00 (1) | 0.3 (3) | 0.01 (4)  |
|  Pacemaker-mediated tachycardia (PMT) | 154 (116) | 0.3 (3) | 0.01 (3) | 10.5 (113) | 0.40 (151)  |
|  Programmer / Software error code | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Threshold elevated/unable to capture – Left Ventricular Lead | 32 (26) | 0.2 (2) | 0.01 (2) | 2.3 (25) | 0.08 (30)  |
|  Threshold elevated/unable to capture – Right Atrial Lead | 4 (4) | 0.0 (0) | 0.00 (0) | 0.4 (4) | 0.01 (4)  |
|  Threshold elevated/unable to capture – Right Ventricular Lead | 6 (6) | 0.2 (2) | 0.01 (2) | 0.4 (4) | 0.01 (4)  |
|  Unable to convert - Defibrillation | 4 (4) | 0.3 (3) | 0.01 (3) | 0.1 (1) | 0.00 (1)  |
|  Undersensing - Defibrillation | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Infrequent events include: Accelerated to AF (1), Cannot measure left ventricular lead impedance (1), Inappropriate AV delay (3), pulse generator system diagnosis - other (3), Psychological effect due to device therapy (1), Seroma - Pocket (> 30 days post-implant) (3), Undersensing – right atrial (1)  |   |   |   |   |   |
|  Subtotal Pulse Generator Related Events | 466 (326) | 7.5 (81) | 0.23 (85) | 25.4 (274) | 1.02 (381)  |
|  Right Atrial (RA) Lead Related Events  |   |   |   |   |   |
|  Conductor coil fracture – RA | 2 (2) | 0.1 (1) | 0.00 (1) | 0.1 (1) | 0.00 (1)  |
|  Impedance > 2000 ohms - RA | 3 (3) | 0.2 (2) | 0.01 (2) | 0.1 (1) | 0.00 (1)  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 26

32

{26}

|   |   | Complications |   | Observations  |   |
| --- | --- | --- | --- | --- | --- |
|  Adverse Event | Total Number Of Events (Number of Patients) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events)  |
|  Insulation breach - RA | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Lead dislodgment - RA | 43 (40) | 3.6 (39) | 0.11 (42) | 0.1 (1) | 0.00 (1)  |
|  Oversensing - RA | 9 (7) | 0.1 (1) | 0.00 (1) | 0.6 (6) | 0.02 (8)  |
|  Threshold elevated/unable to capture - RA | 14 (14) | 0.4 (4) | 0.01 (4) | 0.9 (10) | 0.03 (10)  |
|  Subtotal RA Lead Related Events | 72 (62) | 4.4 (47) | 0.14 (51) | 1.8 (19) | 0.06 (21)  |
|  Right Ventricular (RV) Lead Related Events  |   |   |   |   |   |
|  Conductor coil fracture - RV | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Impedance < 300 ohms - RV | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Impedance > 2000 ohms - RV | 1 (1) | 0.0 (0) | 0.00 (0) | 0.1 (1) | 0.00 (1)  |
|  Insulation breach - RV | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Lead dislodgment - RV | 10 (9) | 0.8 (9) | 0.03 (10) | 0.0 (0) | 0.00 (0)  |
|  Oversensing - RV | 6 (6) | 0.2 (2) | 0.01 (2) | 0.4 (4) | 0.01 (4)  |
|  RV lead dislodgment | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Threshold elevated/unable to capture - RV | 10 (10) | 0.4 (4) | 0.01 (4) | 0.6 (6) | 0.02 (6)  |
|  Undersensing - RV | 2 (2) | 0.1 (1) | 0.00 (1) | 0.1 (1) | 0.00 (1)  |
|  Subtotal RV Lead Related Events | 33 (32) | 1.9 (20) | 0.06 (21) | 1.1 (12) | 0.03 (12)  |
|  Left Ventricular (LV) Lead Related Events  |   |   |   |   |   |
|  Conductor coil fracture - LV | 2 (2) | 0.2 (2) | 0.01 (2) | 0.0 (0) | 0.00 (0)  |
|  Extracardiac stimulation - LV | 43 (40) | 1.4 (15) | 0.05 (17) | 2.4 (26) | 0.07 (26)  |
|  Impedance > 2000 ohms - LV | 6 (6) | 0.3 (3) | 0.01 (3) | 0.3 (3) | 0.01 (3)  |
|  Insulation breach - LV | 4 (4) | 0.4 (4) | 0.01 (4) | 0.0 (0) | 0.00 (0)  |
|  Lead dislodgment - LV | 80 (68) | 5.8 (63) | 0.20 (73) | 0.6 (7) | 0.02 (7)  |
|  Threshold elevated/unable to capture - LV | 24 (23) | 0.6 (6) | 0.02 (6) | 1.6 (17) | 0.05 (18)  |
|  Infrequent events include: Impedance < 300 ohms - LV (2), Oversensing - LV (1)  |   |   |   |   |   |
|  Subtotal LV Lead Related Events | 162 (137) | 8.2 (88) | 0.28 (105) | 5.0 (54) | 0.15 (57)  |
|  Defibrillator Lead Related Events  |   |   |   |   |   |
|  Elevated DFT/ unable to convert | 5 (5) | 0.4 (4) | 0.01 (4) | 0.1 (1) | 0.00 (1)  |
|  Inappropriate tachy therapy | 5 (5) | 0.2 (2) | 0.01 (2) | 0.3 (3) | 0.01 (3)  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 27

33

{27}

|   |   | Complications |   | Observations  |   |
| --- | --- | --- | --- | --- | --- |
|  Adverse Event | Total Number Of Events (Number of Patients) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events)  |
|  Subtotal Defib Lead Related Events | 10 (10) | 0.6 (6) | 0.02 (6) | 0.4 (4) | 0.01 (4)  |
|  Procedure Related Events  |   |   |   |   |   |
|  AV block | 11 (11) | 0.6 (7) | 0.02 (7) | 0.4 (4) | 0.01 (4)  |
|  Adverse reaction | 23 (22) | 0.6 (7) | 0.02 (7) | 1.4 (15) | 0.04 (16)  |
|  Arterial perforation - Procedure | 2 (2) | 0.1 (1) | 0.00 (1) | 0.1 (1) | 0.00 (1)  |
|  Coronary venous dissection | 5 (5) | 0.0 (0) | 0.00 (0) | 0.5 (5) | 0.01 (5)  |
|  Coronary venous perforation without tamponade | 5 (5) | 0.2 (2) | 0.01 (2) | 0.3 (3) | 0.01 (3)  |
|  Hematoma - Pocket (<=30 days post-implant) | 39 (39) | 1.4 (15) | 0.04 (15) | 2.2 (24) | 0.06 (24)  |
|  Inadvertent Supraventricular Tachycardia | 2 (2) | 0.1 (1) | 0.00 (1) | 0.1 (1) | 0.00 (1)  |
|  Inadvertent VT/VF | 5 (5) | 0.5 (5) | 0.01 (5) | 0.0 (0) | 0.00 (0)  |
|  Myocardial perforation with tamponade | 3 (3) | 0.3 (3) | 0.01 (3) | 0.0 (0) | 0.00 (0)  |
|  Other - Lead - Procedure | 8 (8) | 0.7 (8) | 0.02 (8) | 0.0 (0) | 0.00 (0)  |
|  Other - PG system - Procedure | 18 (17) | 0.6 (6) | 0.02 (6) | 1.1 (12) | 0.03 (12)  |
|  Pericardial effusion | 9 (7) | 0.4 (4) | 0.01 (5) | 0.3 (3) | 0.01 (4)  |
|  Pleural effusion - Procedure | 3 (3) | 0.1 (1) | 0.00 (1) | 0.2 (2) | 0.01 (2)  |
|  Pneumothorax - Procedure | 21 (21) | 1.5 (16) | 0.04 (16) | 0.5 (5) | 0.01 (5)  |
|  Post-surgical infection (<= 30 days post-implant) | 20 (17) | 0.7 (8) | 0.03 (11) | 0.8 (9) | 0.02 (9)  |
|  Post-surgical pocket hemorrhage | 4 (4) | 0.2 (2) | 0.01 (2) | 0.2 (2) | 0.01 (2)  |
|  Post-surgical wound discomfort | 29 (28) | 0.0 (0) | 0.00 (0) | 2.6 (28) | 0.08 (29)  |
|  Renal failure due to contrast media - Procedure | 4 (4) | 0.4 (4) | 0.01 (4) | 0.0 (0) | 0.00 (0)  |
|  Seroma - Pocket (<=30 days post-implant) | 4 (4) | 0.1 (1) | 0.00 (1) | 0.3 (3) | 0.01 (3)  |
|  Thromboembolic events | 15 (15) | 0.7 (8) | 0.02 (8) | 0.6 (7) | 0.02 (7)  |
|  Venous occlusion | 3 (3) | 0.1 (1) | 0.00 (1) | 0.2 (2) | 0.01 (2)  |
|  Infrequent events include: Chest pain (1), Hemorrhage (2), Vasovagal (2)  |   |   |   |   |   |
|  Subtotal Procedure Related Events | 238 (187) | 8.3 (90) | 0.28 (104) | 10.5 (113) | 0.36 (134)  |
|  Protocol Testing Related Events  |   |   |   |   |   |
|  Chest pain | 3 (3) | 0.1 (1) | 0.00 (1) | 0.2 (2) | 0.01 (2)  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 28

34

{28}

|   |   | Complications |   | Observations  |   |
| --- | --- | --- | --- | --- | --- |
|  Adverse Event | Total Number Of Events (Number of Patients) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events) | % of Patients (N Patients) | N Events/ 100 Device Months (N Events)  |
|  Fatigue | 1 (1) | 0.0 (0) | 0.00 (0) | 0.1 (1) | 0.00 (1)  |
|  Integumentary | 2 (2) | 0.0 (0) | 0.00 (0) | 0.2 (2) | 0.01 (2)  |
|  Subtotal Protocol Testing Related Events | 6 (6) | 0.1 (1) | 0.00 (1) | 0.5 (5) | 0.01 (5)  |
|  Cardiovascular - Heart Failure (HF) Related Events  |   |   |   |   |   |
|  Worsening heart failure | 533 (287) | 16.9 (184) | 0.87 (325) | 13.7 (149) | 0.56 (208)  |
|  Subtotal Cardiovascular - HF Related Events | 533 (287) | 16.9 (184) | 0.87 (325) | 13.7 (149) | 0.56 (208)  |
|  Cardiovascular - Non-Heart Failure (Non-HF) Related Events  |   |   |   |   |   |
|  Bleeding | 12 (12) | 0.8 (9) | 0.02 (9) | 0.3 (3) | 0.01 (3)  |
|  Bradyarrhythmias | 24 (24) | 0.5 (5) | 0.01 (5) | 1.7 (19) | 0.05 (19)  |
|  Cardiogenic shock | 1 (1) | 0.1 (1) | 0.00 (1) | 0.0 (0) | 0.00 (0)  |
|  Chest pain | 233 (162) | 8.2 (89) | 0.31 (116) | 8.4 (91) | 0.31 (117)  |
|  Dizziness | 67 (64) | 1.2 (13) | 0.03 (13) | 4.9 (53) | 0.14 (54)  |
|  Dyspnea | 49 (45) | 1.2 (13) | 0.03 (13) | 3.2 (35) | 0.10 (36)  |
|  Fatigue | 25 (25) | 0.1 (1) | 0.00 (1) | 2.2 (24) | 0.06 (24)  |
|  Hypo/hypertension | 105 (88) | 2.1 (23) | 0.07 (27) | 6.2 (67) | 0.21 (78)  |
|  Myocardial infarction | 37 (31) | 2.8 (31) | 0.10 (37) | 0.0 (0) | 0.00 (0)  |
|  Other - Patient condition - Cardiovascular | 92 (85) | 4.0 (44) | 0.12 (46) | 3.9 (42) | 0.12 (46)  |
|  Palpitations | 37 (28) | 0.2 (2) | 0.01 (2) | 2.4 (26) | 0.09 (35)  |
|  Related to Vasculature | 2 (2) | 0.2 (2) | 0.01 (2) | 0.0 (0) | 0.00 (0)  |
|  Supraventricular tachyarrhythmias | 238 (173) | 6.9 (75) | 0.26 (96) | 10.8 (118) | 0.38 (142)  |
|  Syncope | 43 (36) | 1.6 (17) | 0.05 (19) | 1.8 (20) | 0.06 (24)  |
|  Vascular | 102 (82) | 6.1 (66) | 0.22 (82) | 1.7 (19) | 0.05 (20)  |
|  Ventricular tachyarrhythmias | 126 (91) | 5.6 (61) | 0.22 (81) | 3.7 (40) | 0.12 (45)  |
|  Subtotal Cardiovascular - Non-HF Related Events | 1193 (580) | 30.6 (333) | 1.47 (550) | 36.4 (396) | 1.72 (643)  |
|  Subtotal Non-cardiovascular Related Events | 1610 (635) | 38.0 (414) | 2.05 (766) | 39.2 (427) | 2.26 (844)  |

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 29

35

{29}

## 2. Results: Primary Effectiveness Endpoint

The MADIT-CRT study assessed the effectiveness of CRT-D by the relative reduction of the risk of the combined endpoint of all-cause mortality or heart failure event, whichever occurred first, when compared to ICD.

Data Analysis: Statistical tests of the difference in the primary effectiveness endpoint (rate of combined all-cause mortality or heart failure event, whichever occurred first) between the randomized CRT-D and ICD groups were performed. The primary effectiveness analysis was based on comparing the Kaplan-Meier life-table event-free survival time graphs for the CRT-D and ICD groups. The stratified log-rank test (stratified by clinical center and ischemic status) was used to evaluate statistical significance, adjusting for the group-sequential stopping rule of the study. The hazard ratio for CRT-D relative to ICD, based on proportional hazards modeling, was also estimated, along with the corresponding 95% confidence limits.

All analyses were carried out according to the intention-to-treat principle and include data occurring on or before December 31, 2009. Patients who did not have a primary endpoint by December 31, 2009, were censored at either their date of withdrawal (if the patient did not agree to phone contact to collect event data) or last visit.

Results: In the full patient population, CRT-D was associated with a statistically significant reduction in the combined endpoint of all-cause mortality or heart failure event, whichever occurred first, when compared to ICD (adjusted log-rank p<0.001. However, post-hoc subgroup analyses revealed that there was no evidence of benefit in the non-LBBB patient sub-population and that the results were driven by the LBBB patient sub-population, which comprised 70% of the total cohort. The adjusted hazard ratio was 0.61, with 95% confidence interval (0.50 to 0.75), and p<0.001.

In the LBBB patient sub-population, CRT-D was associated with a reduction in the relative risk of death or heart failure event by 57% as compared to ICD. The Kaplan-Meier curves demonstrate separation in the early months and continue to separate throughout the subsequent follow-up period as shown below in Figure 2. A table summarizing the data supporting the Kaplan-Meier curves is shown below in Table 17. The components of the primary effectiveness endpoint are shown below in Table 18.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 30

36

{30}

Figure 2: K-M Curves of Time to All-Cause Mortality or HF Event (LBBB Patients Only)
All LBBB patients, N=1281

![img-2.jpeg](img-2.jpeg)

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 31

37

{31}

Table 17: Summary of K-M All-Cause Mortality or HF Events (LBBB Patients Only)
All LBBB patients, N=1281

|   | ICD |   |   |   |   | CRT-D  |   |   |   |   |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|  Months | 0-12 | 12-24 | 24-36 | 36-48 | 48-60 | 0-12 | 12-24 | 24-36 | 36-48 | 48-60  |
|  N Patients at Start of Interval | 520 | 436 | 332 | 181 | 60 | 761 | 700 | 596 | 324 | 93  |
|  N Patients with Endpoint Events | 63 | 58 | 26 | 11 | 4 | 47 | 37 | 22 | 11 | 3  |
|  N Patients with Endpoint Events (cumulative) | 63 | 121 | 147 | 158 | 162 | 47 | 84 | 106 | 117 | 120  |
|  N Patients Censored | 21 | 46 | 125 | 110 | 56 | 14 | 67 | 250 | 220 | 90  |
|  N Patients Censored (cumulative) | 21 | 67 | 192 | 302 | 358 | 14 | 81 | 331 | 551 | 641  |
|  Percent Free from Event | 87.6% | 75.8% | 68.2% | 61.8% | 55.8% | 93.8% | 88.7% | 84.5% | 79.4% | 74.7%  |
|  95% Lower Confidence Bound | 84.4% | 71.8% | 63.6% | 56.0% | 47.9% | 91.8% | 86.2% | 81.5% | 74.9% | 67.4%  |

Table 18: Primary Effectiveness Endpoint Components (LBBB Patients Only)
All LBBB patients, N=1281

|  Item | Number of Patients (% of All Patients in Treatment Group)  |   |   |   |
| --- | --- | --- | --- | --- |
|   |  ICD (N=520) | CRT-D (N=761) | Hazard Ratio (95% Confidence Interval) | P-value  |
|  Patients with Primary Endpoint Event | 162 (31%) | 120 (16%) | 0.43 (0.33, 0.56) | <.001  |
|  Patients with All-Cause Mortality at Any Time* | 51 (10%) | 54 (7%) | 0.65 (0.42, 1.00) | 0.044  |
|  Patients with HF Event | 144 (28%) | 89 (12%) | 0.37 (0.28, 0.50) | <.001  |
|  Inpatient HF Event | 116 (22%) | 76 (10%) | --- | ---  |
|  Outpatient HF Event | 28 (5%) | 13 (2%) | --- | ---  |

* This category includes all deaths, including those that occurred after the first heart-failure event.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 32

38

{32}

Kaplan-Meier curves for the primary endpoint and its components are presented in Figures 3 through 5 below.

Figure 3: Kaplan-Meier Curves for Primary Endpoint

![img-3.jpeg](img-3.jpeg)

Figure 4: Kaplan-Meier Curves for First Heart Failure Event

![img-4.jpeg](img-4.jpeg)

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 33

39

{33}

Figure 5: Kaplan-Meier Curves for All-Cause Mortality

![img-5.jpeg](img-5.jpeg)

# Results: Secondary Endpoint

The MADIT-CRT study also evaluated the effects of CRT-D, relative to ICD, on the recurrence of heart failure events over the full study period.

Data Analysis: The analysis was based on the intention-to-treat principle. The number of heart failure events occurring within the period of active follow-up of each patient was analyzed to determine whether the average rate within the CRT-D group differed from that in the ICD group. An Andersen-Gill regression analysis was performed to assess the benefit of CRT-D on recurrent heart failure events. In the Andersen-Gill regression analysis, patients were censored at their date of death, withdrawal (if patient did not agree to phone contact to collect event data) or last visit.

Results: In the full patient population, CRT-D was associated with a statistically significant reduction in the risk of recurrent heart failure events when compared to ICD. However, there was no evidence of benefit in the non-LBBB patient sub-population, and the results were driven by the LBBB patient sub-population.

In the LBBB patient sub-population, CRT-D was associated with a reduction in the risk of recurrent heart failure events by 43% when compared to ICD. A summary of the number of all heart failure events experienced for LBBB patients according to treatment group is presented below in Table 19.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 34

40

{34}

Table 19: Number of Heart Failure Events by Treatment Arm (LBBB Patients Only)
All LBBB patients, N=1281

|  Number of HF Events | Number of Patients (% of All Patients in Treatment Group)  |   |
| --- | --- | --- |
|   |  ICD (N=520) | CRT-D (N=761)  |
|  0 | 376 (72.3%) | 672 (88.3%)  |
|  1 | 89 (17.1%) | 51 (6.7%)  |
|  2+ | 55 (10.6%) | 38 (5.0%)  |

Rates of heart failure events can be presented two ways, separately for each treatment group: first as the count of heart failure events per 100 patients and second as the count of heart failure events per 100 patient-years of follow-up. Patients randomized to ICD experienced 53 HF events for every 100 patients and 18.5 HF events for every 100 patient-years of follow-up, whereas patients randomized to CRT-D experienced 22 HF events for every 100 patients and 7.5 HF events for every 100 patient-years of follow-up.

### 3. Results: Subgroup Analyses

A subgroup analysis based on pre-specified clinical baseline characteristics revealed statistically significant interactions with sex and QRS width such that female patients and those patients with QRS ≥ 150 ms derived the greatest benefit from CRT-D. During a post hoc investigation of benefit in female patients, a disparity in the prevalence of left bundle branch block (LBBB) morphology between women and men was discovered in which women were more likely than men to have LBBB. The study results were subsequently examined by bundle branch morphology in a post-hoc analysis. When compared to the non-LBBB cohort, LBBB was associated with substantially greater improvement across the primary endpoint and its components, the secondary endpoint, and some tertiary endpoints as shown in Figure 6 below.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 35

41

{35}

Figure 6: Selected Outcomes Stratified by Bundle Branch Block Morphology

![img-6.jpeg](img-6.jpeg)

VT=ventricular tachycardia, VF=ventricular fibrillation

An exploratory analysis restricted to the LBBB patient population revisited the primary endpoint of the prespecified covariates. As shown in Figure 7, each patient population demonstrated consistent improvement with CRT-D.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 36

42

{36}

Figure 7: Primary Endpoint for Prespecified Subgroups (LBBB Patient Sub-Population)

![img-7.jpeg](img-7.jpeg)

# Multivariate Analysis (LBBB Patients Only)

Multivariate analyses were performed in order to determine which baseline clinical variables in addition to treatment were significantly associated with outcomes. The outcomes examined included the primary endpoint and its components and the secondary endpoint. For the primary endpoint and its components, a multivariate Cox proportional hazards model was constructed. For the secondary endpoint, an Andersen-Gill model was used.

CRT-D as a treatment was consistently associated with significantly improved outcomes. In general, covariates associated with worsened baseline condition (e.g. six minute walk distance, NYHA Class and impaired renal function) were also associated with the greater risk of an event. Heart failure medications that improve outcomes, such as angiotensin converting enzymes (ACE), angiotension receptor blockers (ARB), and beta blockers at the target dose were also associated with better prognosis. However, the use of loop diuretics

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 37

43

{37}

and amiodarone were associated with poorer outcomes, possibly linked to their use to treat volume overload and atrial fibrillation, respectively. For the secondary endpoint, the most powerful predictor of recurrent heart failure events included prior events.

Two (2) variables were associated with significant interactions: sex and systolic blood pressure. Women were significantly more likely to derive benefit from CRT-D than men while CRT-D conferred greater benefit to patients with lower systolic blood pressure.

# Study Conclusion of Safety and Effectiveness

In the MADIT-CRT study, the safety and primary effectiveness endpoints were met for the full population studied. However, there was no evidence of benefit in the non-LBBB patient sub-population as the results were predominately driven by the LBBB patient sub-population. A retrospective analysis by bundle branch morphology revealed that CRT-D conferred the greatest benefit in patients with left bundle branch block.

In patients with LBBB, the CRT-D system-related complication-free rate between implant and three months of follow-up was 83.4%; this result was greater than the pre-specified boundary of 70%. CRT-D, when compared to ICD, also reduced the relative risk of the following:

- Combined endpoint of all-cause mortality or heart failure event by \(57\%\)
- Heart failure events alone by \(63\%\)
- All-cause mortality by \(35\%\)
- Recurrent heart failure events by \(43\%\)

Therefore, the LBBB sub-population from the MADIT-CRT study demonstrated the safety and effectiveness of Boston Scientific CRT-D devices in patients that have LBBB with QRS ≥ 130 ms, EF ≤ 30%, and mild (NYHA Class II) ischemic or nonischemic heart failure or asymptomatic (NYHA Class I) ischemic heart failure.

# Tertiary Endpoints (LBBB Patients Only)

There were ten (10) pre-specified tertiary endpoints for this study. For each endpoint, the objective, analysis methods, results and conclusions are provided. Core labs were utilized to reduce variability as well as evaluate and classify tertiary endpoint data specifically related to electrogram analysis and device interrogation, echocardiogram, quality of life (QOL), brain natriuretic peptide (BNP), and Holter. These analyses are exploratory, and the results should be considered suggestive and not definitive. Future studies would be needed to confirm these results.

# All-Cause Mortality

Objective: Evaluate the effects of CRT-D on all-cause mortality.

Endpoint Results: Death occurred in 7.1% (n=54) of the patients in the CRT-D group and 9.8% (n=51) of the patients in the ICD group. The hazard ratio for all-cause mortality was

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 38

44

{38}

0.65. These results indicated an association with CRT-D and a reduction in all-cause mortality rate.

**All-Cause Mortality Conclusion:** Within the MADIT-CRT LBBB sub-population, CRT-D was associated with a reduction of 35% in the risk of all-cause mortality.

# *Appropriate Defibrillator Therapy*

**Objective:** Evaluate the effects of CRT-D on appropriate defibrillator therapy for ventricular tachycardia (VT) and ventricular fibrillation (VF).

**Endpoint Results:** Rates of tachyarrhythmias are calculated by the total number of events (VT, VF and VT or VF) in the randomized group divided by the total follow-up years in the randomized group. Patients randomized to ICD experienced 49.6 VT events, 11.6 VF events, and 61.2 VT or VF events per 100 patient-years of follow-up; whereas patients randomized to CRT-D experienced 31.6 VT events, 2.0 VF events, and 33.7 VT or VF events per 100 patient-years of follow-up respectively. From the time-to-first event model, CRT-D was associated with reductions in VT, VF, or combination of VT and VF of 33%, 57%, and 34% respectively. The results from the recurrent event model were consistent in magnitude with the first event model.

**Appropriate Therapy Conclusion:** When compared to ICD, CRT-D with the MADIT-CRT LBBB sub-population was associated with a reduction in the risk of ventricular tachyarrhythmias using a time to first event model. When using a recurrent events model, CRT-D was associated with a reduction in VT, VF, and combined VT and VF.

# *Echocardiographic Structure and Function*

**Objective:** Evaluate the effects of CRT-D, relative to ICD, on the changes from baseline to one (1) year in echo-determined left ventricular internal volume at end-systole (LVESV) and at end-diastole (LVEDV). The changes from baseline to one (1) year in left ventricular ejection fraction (LVEF) were also evaluated.

**Endpoint Results:** At 12 months, the mean change in LVESV in the CRT-D group from baseline was a reduction of 62 ml, as compared to 19 ml in the ICD group. Similarly, the mean change in LVEDV at 12 months in the CRT-D group was a reduction of 57 ml, as compared to 15 ml in the ICD group. Additionally, the mean change at 12 months in LVEF in the CRT-D group was an improvement of 12%, as compared to 3% in the ICD group.

**Echocardiographic Structure and Function Conclusion:** CRT-D was associated with a reduction in left ventricular volumes and an improvement in left ventricular ejection fraction as compared to ICD. Note, however, that these results should be interpreted with extreme caution, as CRT was not turned off but remained on during the echocardiographic measurements and might have influenced the results.

PMA P010012/S230: FDA Summary of Safety and Effectiveness Data

page 39

45

{39}

New York Heart Association Class

Objective: Evaluate the effects of CRT-D, relative to ICD, on the changes from baseline to one (1) year in NYHA functional class. It was hypothesized that, at one (1) year, the proportion of patients with symptomatic improvement would be greater with CRT-D when compared to ICD, after adjusting for any differences in baseline values.

Endpoint Results: Overall, there were associations between treatment group and change in NYHA based on the five (5) degrees of freedom test. For NYHA class I patients at baseline, there was not an association between NYHA functional class and treatment group. For NYHA class II patients at baseline, the results demonstrated changes for both the ischemic and non-ischemic subgroups. Combining the same and improved groups together also demonstrated an association between treatment group and change in NYHA class at 12 months, confirming analysis of all three (3) change groups (same, worsened, improved from the baseline functional class).

New York Heart Association Class Conclusion: There was no evidence of an association between treatment group and improvement in NYHA at 12 months for the subgroup of patients who were NYHA I at baseline. However, there were associations between treatment group and improvement in NYHA class at 12 months in NYHA II patients, both overall and within the ischemic/non-ischemic patients. Thus, in the MADIT-CRT LBBB sub-population, the Cognis, Livian, and Contak Renewal 3 RF HE CRT-D's lowered NYHA func…

---

**Source:** [https://fda-staging.innolitics.com/device/P010012S230](https://fda-staging.innolitics.com/device/P010012S230)

**Published by [Innolitics](https://innolitics.com)** — a medical-device software consultancy. We help companies design, build, and clear FDA-regulated software and AI/ML devices. If you're preparing [a PMA](https://innolitics.com/services/regulatory/), [a 510(k)](https://innolitics.com/services/510ks/), [a SaMD](https://innolitics.com/services/end-to-end-samd/), [an AI/ML medical device](https://innolitics.com/services/medical-imaging-ai-development/), or [an FDA regulatory strategy](https://innolitics.com/services/regulatory/), [get in touch](https://innolitics.com/contact).

**Cite:** Innolitics at https://innolitics.com