← Product Code MOM · P100045

# CARDIOMEMS HF PRESSURE MEASUREMENT SYSTEM (P100045)

_ABBOTT MEDICAL · MOM · May 28, 2014 · Cardiovascular · APPR_

**Canonical URL:** https://fda.innolitics.com/device/P100045

## Device Facts

- **Applicant:** ABBOTT MEDICAL
- **Product Code:** MOM
- **Decision Date:** May 28, 2014
- **Decision:** APPR
- **Device Class:** Class 3
- **Review Panel:** Cardiovascular
- **Attributes:** Expedited Review

## Indications for Use

The CardioMEMS™ HF System is indicated for wirelessly measuring and monitoring pulmonary artery (PA) pressure and heart rate in New York Heart Association (NYHA) Class III heart failure patients who have been hospitalized for heart failure in the previous year. The hemodynamic data are used by physicians for heart failure management and with the goal of reducing heart failure hospitalizations.

## Device Story

The CardioMEMS HF System is a permanently implantable, battery-free, capacitive pressure sensor (PA Sensor) deployed in the distal pulmonary artery via a transvenous delivery catheter. The system includes external electronics (Hospital and Patient units) that use RF energy bursts to power the sensor and interrogate its resonant frequency, which is converted into systolic, diastolic, and mean PA pressure measurements. The patient uses the home unit to take daily measurements, which are automatically transmitted to a secure database. Physicians access this data via a website to monitor hemodynamic status and adjust heart failure therapy. The device is intended to reduce heart failure hospitalizations by enabling proactive, pressure-guided management. The system is used in both clinical and home settings; the sensor is implanted by a physician, while the patient performs daily measurements at home.

## Clinical Evidence

The pivotal CHAMPION study was a prospective, multi-center, randomized, single-blind clinical trial (N=550). The primary effectiveness endpoint was the rate of heart failure-related (HFR) hospitalizations at 6 months. The Treatment group showed a 28% reduction in HFR hospitalizations compared to the Control group (p=0.0002). Secondary endpoints, including change in PA mean pressure, proportion of patients hospitalized, days alive outside the hospital, and quality of life (MLHFQ), were also met. Safety was demonstrated by a 98.6% freedom from device/system-related complications rate at 6 months. Ancillary longitudinal analyses (Part 2) supported the effectiveness findings.

## Technological Characteristics

The PA Sensor is a capacitive pressure sensor (15 x 3.4 x 2 mm) made of fused silica encased in medical-grade silicone, featuring nitinol wire loops for anchoring and platinum/iridium marker bands. It operates via electromagnetic coupling (30-37.5 MHz) and is battery-free. The delivery system is a 12Fr transvenous catheter with a hydrophilic coating. The system is MR Conditional (1.5T/3T). Sterilization is via validated EtO process. Software is used for signal acquisition, processing, and database transmission.

## Reference Devices

- COMPASS-HF (J Am Coll Cardiol. 2008 Mar 18;51(11):1073-9)
- MIRACLE Study Group (N Engl J Med. 2002 Jun 13;346(24):1845-53)

## 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/).

SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED)

I. GENERAL INFORMATION

Device Generic Name: Heart Failure Monitoring System

Device Trade Name: CardioMEMS™ HF System

Device Procode: MOM

Applicant’s Name and Address: CardioMEMS, Inc.

Date(s) of Panel Recommendation: December 8, 2011 & October 9, 2013

Premarket Approval Application (PMA) Number: P100045

Date of FDA Notice of Approval: May, 28, 2014

Priority Review: N/A

II. INDICATIONS FOR USE

The CardioMEMS™ HF System is indicated for wirelessly measuring and monitoring pulmonary artery (PA) pressure and heart rate in New York Heart Association (NYHA) Class III heart failure patients who have been hospitalized for heart failure in the previous year. The hemodynamic data are used by physicians for heart failure management and with the goal of reducing heart failure hospitalizations.

III. CONTRAINDICATIONS

The CardioMEMS HF System is contraindicated for patients with an inability to take dual antiplatelet or anticoagulants for one month post implant.

IV. WARNINGS AND PRECAUTIONS

The warnings and precautions can be found in the CardioMEMS HF System labeling.

V. DEVICE DESCRIPTION

The CardioMEMS HF System is a permanently implantable pressure measurement system designed to provide daily PA pressure measurements including systolic, diastolic, and mean PA pressures. These measurements are used to guide treatment of congestive heart failure (CHF). The system consists of the following components:

- PA Sensor (CM3000) - The PA Sensor is a battery-free capacitive pressure sensor permanently implanted in the pulmonary artery.

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 1

- Delivery System (CM3000) – The Delivery System is a transvenous catheter designed to deploy the sensor within the distal PA. The catheter has a usable length of 120cm, has a hydrophilic coating on the distal end of the catheter, and is compatible with a 0.018" guidewire.
- CardioMEMS Hospital (CM2000) and Patient Electronics Systems (CM1000 (GSM) and CM1010 (Landline)) and Database (CardioMEMS HF Website) - The Electronics Systems acquire and process signals from the PA Sensor and transfers PA pressure measurements to a secure database. The Database receives data transmitted from the Electronics Systems, and presents the PA pressure data for review by medical professionals, who can make decisions regarding the status of the patient and initiate changes in medical therapy.

## PA Sensor

The PA Sensor consists of a three (3) dimensional coil and pressure sensitive capacitor encased between two (2) wafers of fused silica measuring 15 x 3.4 x 2 mm. The fused silica assembly is completely encased in medical grade silicone. The coil electromagnetically couples the pressure sensitive capacitor to the Electronics System, allowing the remote measurement of the resonant frequency of the circuit without the need for an on-board battery. This resonant frequency is then converted to a pressure measurement.

The sensor is implanted in a descending branch of the left or right PA using a transvenous catheter. Nitinol wire loops extend from the pressure sensor; they are larger than the sensor and keep the implant in a PA branch of substantially greater diameter than the sensor size. Two (2) platinum/iridium marker bands at each end of the sensor (total of four (4) marker bands) allow the device to be visualized under fluoroscopy during the implant procedure (and on imaging during follow-up visits) and indicate the position of the sensor. Tether wires connect the PA Sensor to the Delivery System until the physician determines that the sensor is properly positioned within the distal PA. Once the sensor is in position, the tether wires are withdrawn, releasing the sensor.

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 2

A photograph of the PA Sensor is provided in Figure 1 below.

![img-0.jpeg](img-0.jpeg)
Figure 1. PA Sensor

# Delivery System

The Delivery System is an over the wire transvenous catheter used to deploy the PA Sensor. The sensor is attached by tether wires to the Delivery System as shown in Figure 2 below.

The Delivery System includes a hydrophilic coating on the distal portion of the catheter shaft. The Delivery System (with the sensor) is introduced over a guidewire through a 12Fr sheath. The usable length is  $120\mathrm{cm}$  and it is compatible with a 0.018" guidewire.

![img-1.jpeg](img-1.jpeg)
Figure 2. Distal Section of the PA Sensor and Delivery System including Tether Wire and Nitinol Loops.

The Delivery System is used to maneuver the sensor into the PA over the guidewire. Once it is optimally positioned, the sensor is separated from the Delivery System by pulling the tether wires that are connected to the cap on the catheter hub (see Figure 3 below). The Delivery System is then removed. The sensor remains in the PA as a permanent implant.

PMA P100045: FDA Summary of Safety and Effectiveness Data

![img-2.jpeg](img-2.jpeg)
Figure 3. Proximal End of Delivery Catheter with Cap and tether wires.

CardioMEMS Hospital and Patient Electronics Systems and Database

The CardioMEMS HF Monitoring System consists of a Hospital Electronics System, Patient Electronics Systems, and the associated sterile PA Sensor and Delivery System. The Hospital Electronics System is used in the hospital or clinic and the Patient Electronics System is used for home patient monitoring. The hospital and patient systems are similar except for greater functionality in the hospital system including display and printing of the pressure data, which is not available on the patient version. The software for the hospital system allows pressure measurements to be visualized on the touch screen during sensor implant with systolic, diastolic, and mean PA pressure as well as a waveform. The software on the patient system prompts and guides the patient to make a PA pressure measurement and automatically uploads the information to the Database.

The physician accesses data for each of his/her patients via a secure CardioMEMS HF Website that allows the physician to utilize PA pressure measurements in the management of heart failure. When the patient is hospitalized or returns to the clinic/office setting, the Hospital Electronics System can be used to obtain PA pressure measurements and allows the physician to see not only the pressure data, but also the waveform. When the patient returns home the Patient Electronics System can be used to obtain and transmit PA pressure measurements to the Database for physician access.

There are two (2) main components in both units: the antenna and main unit.

**Antenna**

The antenna is used to interrogate the PA Sensor. There are two (2) versions of the antenna: a rigid plastic housing and a flat, flexible model. The rigid antenna is used during the implant procedure, while the home measurements will be made with the flat antenna which is designed to allow the patient to lie on it. During a reading, the antenna is placed in the vicinity of the passive sensor and the antenna powers it using bursts of RF energy. When the sensor is energized, it returns a signal with pressure information. This signal is received by the antenna and sent to the main unit for processing.

PMA P100045: FDA Summary of Safety and Effectiveness Data

# Main Unit

The main unit is the location of all the signal generation and processing for the Hospital and Patient Electronics Systems. The custom circuitry generates bursts of RF energy which powers the sensor, processes the return signal from the sensor, and transmits pressure information to the single board computer. This circuitry also contains barometric pressure sensors which provide information to compensate for changes in atmospheric pressure. The hospital and patient systems are similar except for greater functionality in the Hospital Electronics System including display and printing of the pressure data which is not available on the Patient Electronics System. The hospital system is illustrated in Figure 4 below.

![img-3.jpeg](img-3.jpeg)
Figure 4. CardioMEMS Hospital Electronics System

# Implant Procedure

Once the target vessel has been identified, the PA Sensor is deployed. A right heart catheter is then placed in the pulmonary artery to obtain pulmonary artery pressure readings for calibration of the sensor mean pressure values. Using the initial implant mode, the physician is able to collect simultaneous pressure readings with both the right heart catheter and sensor. Throughout the procedure, the physician is able to obtain any number of pressure readings.

Figure 5 below is a simulated waveform and data presentation available only for the physician during implant. The patient will see neither a waveform nor a numerical presentation of their pressure reading.

PMA P100045: FDA Summary of Safety and Effectiveness Data

![img-4.jpeg](img-4.jpeg)
Figure 5. User interface for physician - Measurement Mode

The Hospital Electronics System can be attached to an IV pole during sensor implant. Once the implant procedure is completed, a patient system with a flexible antenna is given to the patient to take home so that they may begin transmitting pressure readings.

After the implant procedure is completed, the software will provide audio and visual prompts for the patient to guide them through signal acquisition. Once the signal is acquired, the patient is notified of the successful reading and the data is automatically transmitted to a remote secure database where the data can be evaluated by the physician.

# VI. ALTERNATIVE PRACTICES AND PROCEDURES

The only alternative method for obtaining pulmonary artery pressure is currently through a right heart catheterization (RHC) procedure. This is a procedure during which a catheter is inserted through a large vein in the neck or groin and subsequently advanced into the pulmonary artery. In the hospital setting, the RHC is used to measure pulmonary artery pressure and tailor CHF therapy. However, use of this procedure to obtain pulmonary artery pressure frequently is impractical and associated with significant risks, including bruising and/or bleeding at the insertion site, trauma to the vein, trauma to the heart, and lung puncture. Other inherent risks include possible induction of cardiac arrhythmias, infection, and/or embolism.

Prior studies have concluded that changes in cardiac hemodynamics can be indicative of disease fluctuation or progression of the disease.

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

PMA P100045: FDA Summary of Safety and Effectiveness Data

VII. MARKETING HISTORY

The CardioMEMS HF System has not been marketed in the United States or any foreign country.

VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH

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

- Infection
- Upper respiratory infection
- Bronchitis
- Pneumonia
- Acute Bronchitis
- Groin abscess
- Methicilin-resistant staphylococcal aureus infection
- Pulmonary Infiltration
- Sepsis

- Arrhythmias
- Ventricular tachycardia
- Atrial fibrillation
- Ventricular arrhythmia
- Ventricular fibrillation
- Atrial fibrillation with rapid ventricular response
- Atrial flutter
- Cardiac dysrhythmias
- Tachycardia
- Wide complex tachycardia

- Bleeding
- Epistaxis
- Hemoptysis
- GI bleed
- Bleeding
- Blood in stool
- Catheter site bleeding
- Catheter site ecchymosis
- Hematuria
- Nose bleeds

- Hematoma
- Hematoma
- Catheter site hematoma
- Vessel puncture site hematoma

- Thrombus
- Arterial thrombosis (limbs)
- Blood clot

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 7

- Myocardial infarction
Stroke
- Transient Ischemic Attack
Death
Device embolization

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

# IX. SUMMARY OF PRECLINICAL STUDIES

# A. Laboratory Studies

Table 1. Summary of Testing- Sensor, Delivery System, and Electronics

|  Sensor Functional Testing  |   |   |   |
| --- | --- | --- | --- |
|  Test | Acceptance Criteria | Results | Analysis Type  |
|  Sensor Accuracy | Sensor must measure within ± 2 mmHg at baseline and ± 3% across the pressure range compared to a reference pressure measurement for a pressure range of 600-860 mm Hg (absolute). | Pass | Variable  |
|  Simulated Use Pressure Cycle Conditions | Sensor must continue to function throughout 10 years of simulated fatigue testing (400 million cycles). | Pass | Attribute  |
|   |  Sensor must maintain accuracy within +/- 10 mm Hg of a reference pressure measurement throughout 10 years of simulated fatigue testing (400 million cycles). | Pass | Variable  |
|  Sensor Detection Distance | Sensor must be detectable by the external measurement system at > 4” and > 50% signal strength in physiological saline. | Pass | Variable  |
|  Temperature Sensitivity | Sensor pressure measurement change per unit temperature change: +1 ± 1mm Hg / °C. | Pass | Variable  |
|  Over-Pressure Exposure | Sensor must meet functional requirements after exposure to 2.0 atm. | Pass | Attribute  |
|  Mechanical Shock | Sensor must meet functional requirements after shock and vibration testing, per ISTA-2A. | Pass | Attribute  |
|   |  Sensor must meet functional requirements after mechanical shock in testing per ISO 14708-1, part 23.1. | Pass | Attribute  |
|  Corrosion | Metallic components shall show no sign of corrosion in testing per ISO-10555-1, Annex A. | Pass | Attribute  |
|  Sensor Radiopacity | Catheter shaft is visible under fluoroscopy. | Pass | Attribute  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 9

|  Sensor Compatibility Testing  |   |   |   |
| --- | --- | --- | --- |
|  Test | Acceptance Criteria | Results | Analysis Type  |
|  MRI | Device must meet “MR Conditional” requirements for safe scanning immediately after placement under the following conditions:
- Static magnetic field of 1.5 or 3 Tesla
- Maximum spatial gradient magnetic field of 720-Gauss/cm (7200-mT/m) or less

In non-clinical testing with 1.5 and 3 Tesla systems, device must meet standard requirements for:
- Displacement Force, per ASTM F2052-06
- RF Heating, per ASTM F2182-11
- Torque, per ASTM F2213-06
- Image Artifact, per ASTM F2119-07 | Pass | Attribute -- Third Party Testing and Review  |
|   |  Sensor must meet functional and accuracy requirements after 3T MR exposure. | Pass | Attribute  |
|  Defibrillation | Must meet requirements for defibrillation shock testing, per ISO 14708-1, part 20.2. | Pass | Attribute  |
|  Ultrasound | Must meet requirements of Ultrasound compatibility testing, per ISO 14708-1, part 22. | Pass | Attribute  |
|  Pacemaker and ICD Compatibility | The normal operation of the system, pacemakers, and ICD’s must not be affected during simultaneous operation for potential modes of use. The in vitro test plan includes:
• Representative relative placement in a human torso anatomical model
• Multiple ICD and pacemaker models
• Variables addressed in the test plan:
○ Number of pacing chambers
○ Potential operating modes
○ Unipolar and bipolar lead configurations
○ Relative orientations between the external system and implanted devices. | Pass | Attribute -- Third Party Testing and Review  |

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 10

|  Delivery System with Sensor |   |  |   |
| --- | --- | --- | --- |
|  Test | Acceptance Criteria | Results | Analysis Type  |
|  Simulated Implant Procedure Testing | The PA Delivery System and Sensor must meet the following requirements in simulated use testing:
1) Removal of the catheter from the packaging,
2) Catheter preparation per IFU,
3) Advancement of the catheter over the guidewire,
4) Loading of the catheter through the venous sheath,
5) Tracking over the guidewire within a venous / right heart / pulmonary arterial anatomical model,
6) Positioning the sensor in the target implant location,
7) Retraction into the sheath,
8) Sensor deployment in target location,
9) Sensor visual inspection post-delivery,
10) Catheter integrity post-removal,
11) Wire loop integrity post-removal, and
12) Sensor meets functional and accuracy requirements after simulated implant procedure. | Pass | Attribute  |
|  Catheter Shaft Tensile | Catheter shaft and hub tensile forces must be ≥ 15N. Test performed per ISO 10555-1, Annex B. | Pass | Variable  |
|  Hydrophilic Coating Integrity and Uniformity | ≥ 90% coating coverage over coated length of shaft after durability and friction test, using Congo Red dye as an indicator | Pass | Attribute -- Third Party Testing and Review  |
|  Hydrophilic Coating Durability and Friction | ≥ 50% reduction in friction vs. uncoated shaft, after 15 passes through silicone pads under a 500 g load. | Pass | Variable - Third Party Testing and Review  |
|  Catheter Shaft Radiopacity | Catheter shaft is visible under fluoroscopy. | Pass | Attribute  |
|  Pouch Bubble Emission Test | No continuous streams of bubbles emanating from the pouch, per ASTM F2096-11. | Pass | Attribute  |
|  Pouch Seal Tensile Test | Pouch seal peak tensile load for 1 in. wide sample must be greater ≥ 1.0 lbf/in, per ASTM F88-09. | Pass | Variable  |
|  Shipping and Environmental Conditions | Product must meet specifications after exposure to shipping and environmental conditions, per ISTA-2A. | Pass | Attribute  |
|  Shelf-Life | Product must meet specifications after exposure to 2-year accelerated (per Q10 Theory, per ASTM F1980-07) and real-time aging. | Pass | Attribute  |

|  Sterilization  |   |   |   |
| --- | --- | --- | --- |
|  Test | Acceptance Criteria | Results | Analysis Type  |
|  Sterilization | A validated EtO sterilization process is used. It is considered an overkill sterilization cycle.The sterilization process must demonstrate a sterilization assurance level of ≤ 10-6using a “worst case” challenge configuration of the product in a sterilization process validation performed per ISO-11135 requirements. | Pass | Attribute -- Third Party Testing and Review  |
|  Sterilization Byproducts | EtO residuals must be within acceptable limits per ISO 10993-7. | Pass | Attribute -- Third Party Testing and Review  |
|  Electronics Unit  |   |   |   |
| --- | --- | --- | --- |
|  Test | Acceptance Criteria | Results | Analysis Type  |
|  Electrical Safety Testing | Electronics must meet safety requirements for medical electronic equipment defined by IEC 60601-1, AAMI ES60601-1 and CAN/CSA-C22.2 No. 60601-1. | Pass | Attribute -- Third Party Testing and Review  |
|  Emissions Testing (FCC and International) | Electronics and Sensor must meet national and international electromagnetic emissions requirements defined by IEC 60601-1-2, EN 55022, FCC Part 15 (Sensor authorization FCC identifier: R3PCSA-00051), FCC Part 18, and EN 302 510-1 & 2. | Pass | Attribute -- Third Party Testing and Review  |
|  Electromagnetic Compatibility Testing | Electronics must meet requirements for electromagnetic compatibility as defined by IEC 60601-1-2, ETSI EN 301 489-1, and ETSI EN 301 489-3. | Pass | Attribute -- Third Party Testing and Review  |
|  Design Testing | The Electronics must operate over a Frequency Range of 30 MHz to 37.5 MHz. | Pass | Attribute  |
|   |  The accuracy of the system must be +/- 2mmHg at baseline and +/- 3% over the remaining operating pressure range. | Pass | Variable  |
|   |  The resolution of a reading must be 1 mmHg. | Pass | Attribute  |
|   |  The electrical noise must be less than 20kHz peak. | Pass | Attribute  |
|   |  The sample rate shall be greater than or equal to 120 Hz. | Pass | Attribute  |
|  Thermal Assessment Test | Electronics must function accurately (within +/- 4mmHg of a reference standard) over the range of normal operating temperatures (5°C to 40°C) as defined in IEC 60601-1-11 and not exceed surface and internal temperatures as defined by IEC 60601-1. | Pass | Attribute -- Third Party Testing and Review  |
|  Flexible Antenna and Pad Set | Electronics must continue to function throughout 5 years of simulated use (1825 cycles). | Pass | Attribute  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

|  Electronics Unit  |   |   |   |
| --- | --- | --- | --- |
|  Test | Acceptance Criteria | Results | Analysis Type  |
|  Verification Testing |  |  |   |
|  Mechanical Testing | Electronics must function after rough handling, shock, and vibration as defined in IEC 60601-1 and IEC 60601-1-11. | Pass | Attribute -- Third Party Testing and Review  |
|  Label Durability Testing | The labeling on the Electronics must meet durability requirements defined in IEC 60601-1. | Pass | Attribute -- Third Party Testing and Review  |
|  Ship and Environmental Testing | The Electronics shall meet transport and storage conditions defined by IEC 60601-1-11 and meet transport testing defined by ISTA-3A. | Pass | Attribute -- Third Party Testing and Review  |

The engineering study results for the Sensor demonstrated the following conclusions:

- Remains functional after 10 years of simulated use;
- Temperature, over-pressurization and mechanical shock have a negligible effect on Sensor function;
- Meets its specifications for accuracy during the hermeticity and calibration testing;
- Meets RF signal detection requirements for distance between the antenna and the implanted sensor in a simulation;
- Remains securely attached to the Delivery System until release;
- Is resistant to corrosion; and
- Is compatible with MRI, defibrillators, ultrasound, pacemakers and ICDs

The engineering study results for the Delivery System demonstrated the following conclusions:

- May be removed from the packaging, flushed with saline, advanced over an 0.018" guidewire and loaded into a venous sheath;
- Positions the Sensor in the target vessel, retracts into the sheath and releases the HF Sensor at the appropriate time;
- Does not damage the catheter or HF Sensor during delivery and has sufficient tensile strength to maintain its integrity during use;
- Is corrosion resistant and is sufficiently radiopaque; and
- Hydrophilic coating is durable and maintains its integrity during use.

Table 2. Biocompatibility

|  Test | Acceptance Criteria | Results | Analysis Type  |
| --- | --- | --- | --- |
|  Cytotoxicity | Meet requirements in an ISO Elution Method study (1xMEM Extract), per ISO 10993-5. | Pass | Attribute -- Third Party Testing  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

|  Test | Acceptance Criteria | Results | Analysis Type  |
| --- | --- | --- | --- |
|  Sensitization | Meet requirements in an ISO Maximization Sensitization Study (Extract), per ISO 10993-10. | Pass | Attribute -- Third Party Testing  |
|  Intracutaneous Reactivity | Meet requirements in an ISO Intracutaneous Study (Extract), per ISO 10993-10. | Pass | Attribute -- Third Party Testing  |
|  Acute and Subchronic Systemic Toxicity | Meet requirements in an ISO Systemic Toxicity Study (Extract), per ISO 10993-11. | Pass | Attribute -- Third Party Testing  |
|  Hemolysis | Meet requirements in an In Vitro Hemolysis Study (ASTM-Extraction Method), per ISO 10993-4. | Pass | Attribute -- Third Party Testing  |
|  C3a Compliment Activation | Meet requirements in a C3a Complement Activation Assay, per ISO 10993-4. | Pass | Attribute -- Third Party Testing  |
|  SC5b-9 Compliment Activation | Meet requirements in a SC5b-9 Complement Activation Assay, per ISO 10993-4. | Pass | Attribute -- Third Party Testing  |
|  USP Pryogen Study | Meet requirements in a USP Pyrogen Study (Material Mediated), per EN ISO 10993-11. | Pass | Attribute -- Third Party Testing  |
|  Chromosomal Aberration | Meet requirements in a Mouse Bone Marrow Micronucleus Study, per ISO 10993-3. | Pass | Attribute -- Third Party Testing  |
|  Bacterial Reverse Mutation | Meet requirements in a Bacterial Reverse Mutation Study (Saline Extract and DMSO Extract), per ISO 10993-3. | Pass | Attribute -- Third Party Testing  |
|  Mouse Micronucleus | Meet requirements in a Mouse Bone Marrow Micronucleus Study, per ISO-10993-3. | Pass | Attribute -- Third Party Testing  |
|  Muscle Implantation | Meet requirements in a ISO Muscle Implantation Study, per ISO-10993-6. | Pass | Attribute -- Third Party Testing  |
|  Carcinogenicity | Scientific rationale provided for no risk of carcinogenesis associated with clinical use. | Pass | Review  |
|  Chronic Toxicity, Thromboresistance, and Histopathology | Must demonstrate acceptable long term tissue response, thromboresistance, and no chronic toxicity after 12 months implantation in a porcine animal model. | Pass | Attribute -- Third Party Testing  |
|  Particulate Testing | Must meet particulate requirements per USP-788 after simulated use. | Pass | Attribute -- Third Party Testing  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 3. US and International Standards

|  ISO 11607 | Packaging for terminally sterilized medical Devices  |
| --- | --- |
|  ISO 14708-1, EN 45502 | Active Implantable Medical Devices, General Requirements  |
|  ISTA-2A / ISTA-3A | International Safe Transit Association Medical Packaging Testing  |
|  ISO 11135 | Sterilization of Health Care Products by Ethylene oxide  |
|  ISO 10993 | Biological Evaluation of Medical Devices  |
|  ISO 10555-1 | Sterile, Single-Use Intravascular Catheters General requirements  |
|  ASTM F2096-11 | Detecting Gross Leaks in Packaging by Internal Pressurization  |
|  ASTM F88-09 | Seal Strength of Flexible Barrier Materials  |
|  ASTM F1980-07 | Standard Guide for Accelerated Aging of Sterile Barrier Systems for Medical Devices  |
|  ASTM F2052-06 | Measurement of Magnetically Induced Displacement Force on Medical Devices in the MR Environment  |
|  ASTM F2182-11 | Measurement of RF Induced Heating on or near Passive Implants during MR Imaging  |
|  ASTM F2213-06 | Measurement of Magnetically Induced Torque on Medical Devices in the MR Environment  |
|  ASTM F2119-07 | Evaluation of MR Image Artifacts from Passive Implants  |
|  ISO 11607 | Packaging for terminally sterilized medical Devices  |
|  EN 20594-1, EN 1707 | Conical fittings with a 6% (Luer) taper  |
|  EN 980 | Symbols for Medical Devices  |
|  IEC 60601-1, CAN/CSA-C22.2, ES60601-1 | Medical Electrical Equipment  |
|  EN 302 510, IEC 60601-1-2, EN 301 489 | Electromagnetic Compatibility  |
|  EN 55022, EN 302 510, FCC Part 18, FCC Part 15 | Radiated Emissions  |
|  IEC 60601-1-11 | Medical Electrical Systems in the Home Healthcare Environment  |
|  EN 62304 | Life Cycle Requirements for Medical Device Software  |

B. Animal Studies

Chronic studies were performed in nine (9) pigs with two (2) sensors in each animal with follow-up periods of 3, 6, and 12 months. The sensors were fully endothelialized and well tolerated. Readings were obtained from all sensors throughout the studies.

PMA P100045: FDA Summary of Safety and Effectiveness Data

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 15

# X. SUMMARY OF PRIMARY CLINICAL STUDIES

CardioMEMS, Inc. conducted a randomized, controlled pivotal study, CHAMPION, of the device under Investigational Device Exemption (IDE) application G060187. The purpose of this study was to establish a reasonable assurance of safety and effectiveness of the reduction of heart failure hospitalizations with the CardioMEMS HF System for wirelessly measuring and monitoring pulmonary artery (PA) pressure and heart rate in New York Heart Association (NYHA) Class III heart failure patients.

All subjects enrolled had the device implanted. Subjects randomized to the Treatment group were managed by their physicians using the PA pressure data. Subjects randomized to the control group had the device implanted, but data from the device was not made available to the physicians for making treatment decisions. Following completion of follow-up necessary for analysis of the primary endpoint (mean 17.6 months) (referred to as Part 1 or the Randomized Access Period), PA data was made available to physicians for all subjects, including those originally randomized to the control group. This second phase was referred to as Part 2 or the Open Access Period. CardioMEMS, Inc. continued to follow subjects enrolled in the Randomized Access Period of the Study and continued to collect data during the Open Access Period of the Study.

The Part 1 data revealed that trial conduct included subject-specific treatment recommendations sent by nurses employed by the CardioMEMS to the treating physicians. These subject-specific recommendations were limited to subjects in the treatment arm of the study. The possible impact of nurse communications was determined to severely limit the interpretability of the data in terms of effectiveness. Additionally, the post-hoc gender analysis noted a statistically significant treatment by gender interaction. Therefore, the evidence about device effectiveness in females is unclear.

To address these concerns, the CardioMEMS commissioned an independent third party audit to identify and characterize the nature of all communications between CardioMEMS and the investigative sites. CardioMEMS also continued to follow patients implanted with the device in Part 2 (Open Access Period). The results of the audit were found to be acceptable. Notably, the audit results provided assurance that the nurse communications were limited to Part 1 (Randomized Access Period) of the study. CardioMEMS provided a Clinical Analysis that included a clinical evaluation of the nurse communications to assess the clinical impact on Heart Failure Related (HFR) hospitalizations. Finally, CardioMEMS conducted multiple analyses of the Part 2 data in order to demonstrate that the observed effect could be attributed to the device and not only to the nurse communications with investigational sites.

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 16

# A. Study Design

## Randomized Access (Part 1)

For the initial Randomized Access Period of the Study (Part 1), patients were treated between September 6, 2007 and August 12, 2010. The database for this PMA reflected data collected through April 30, 2012 and included 550 patients. There were 64 investigational sites.

The study was a prospective, multi-center, randomized, single-blind clinical trial conducted in the United States (US). All subjects who met the eligibility criteria at the Screening Visit and provided informed consent form were eligible to participate within the study. Following the Screening Visit, all subjects were implanted in conjunction with a right heart catheterization (RHC) procedure. Following the RHC and after sensor implant, but prior to hospital discharge, subjects were randomized to one of two (2) groups:

- Treatment group: standard of care HF management plus HF management based upon hemodynamic information obtained from the HF System
- Control group: standard of care HF management

Following the Sensor implant, subjects were hospitalized overnight for observation and evaluation. Prior to hospital discharge, subjects were trained in the use of the equipment, including how to take their daily HF pressure measurements and how to initiate the transfer of their pressure reading to a secure database.

For the Treatment group, the investigator provided standard of care HF management plus HF management based upon hemodynamic information obtained from the HF System. If the PA pressures were outside the prescribed limits, the investigator used the data in their evaluation of the medical condition of the patient and initiated treatment options per recommendations specified within the clinical protocol.

For the Control group, the investigator provided standard of care HF management and did not have access to the home pressure measurements.

All subjects were blinded to the randomization assignment and did not have access to their pulmonary artery pressures.

During the study, patient contact by the investigative sites by phone was scripted for both Treatment and Control groups. The script for both groups was identical except for the medication adjustment in the Treatment group. The contact was balanced to assure that when a Treatment patient was contacted by phone for a PA pressure based intervention, a matching phone contact was made to a randomly selected control patient.

# 1. Clinical Inclusion and Exclusion Criteria

Enrollment in the CHAMPION study was limited to patients who met the following inclusion criteria:

- Written informed consent obtained from subject or legal representative.
- Male or female, at least 18 years of age
- Diagnosis of HF ≥ 3 months, with either preserved or reduced Left Ventricular Ejection Fraction (LVEF)
- Diagnosis of NYHA Class III HF (historical assessment documented at screening visit)
- At least one HF-related hospitalization within 12 months of Screening Visit
- Subjects with reduced LVEF must be receiving a beta blocker for three (3) months and an Angiotensin-Converting Enzyme Inhibitor (ACE-I) or Angiotensin Receptor Blocker (ARB) for one (1) month unless in the investigator's opinion, the subject is intolerant to beta blockers, ACE-I, or ARB. Beta blockers and ACE-I (or ARB) doses should be stable for one (1) month prior to study entry.
- Subjects with a BMI ≤ 35 or chest circumference ≤ 52 inches. In subjects with BMI >35 and chest circumference >52 inches, the distance from the subject's back to the pulmonary artery must be < 10 cm on lateral angiography during the RHC. Patients with chest circumference >65 inches were excluded.
- Subjects with implant pulmonary artery branch diameter between 7mm and 15mm.
- Female subjects of childbearing age with a negative urine or serum pregnancy test (at Screening Visit), and who have agreed to use a reliable mechanical or hormonal form of contraception during the study will be allowed to enter the study. Note: A female is considered of child-bearing potential unless she is postmenopausal for two (2) years, has had a total hysterectomy, or has had a bilateral tubal ligation.
- Subjects willing and able to comply with the follow up requirements of the study.

Patients were not permitted to enroll in the CHAMPION study if they met any of the following exclusion criteria:

- Subjects with an active infection.
- Subjects with history of recurrent (>1) pulmonary embolism or deep vein thrombosis.
- Subjects, in the investigator's opinion, unable to tolerate a right heart catheterization.
- Subjects who have had a major cardiovascular event (e.g., myocardial infarction, stroke) within two (2) months of Screening Visit.
- Subjects with Cardiac Resynchronization Device (CRT) implanted ≤ 3 months prior to enrollment.

PMA P100045: FDA Summary of Safety and Effectiveness Data

- Subjects with a Glomerular Filtration Rate (GFR)  $< 25 \, \text{ml/min}$  who are non-responsive to diuretic therapy or who are on chronic renal dialysis.
- Subjects likely to undergo heart transplantation within six (6) months of Screening Visit.
- Subjects with congenital heart disease or mechanical right heart valve(s).
- Subjects with known coagulation disorders.
- Subjects with a hypersensitivity or allergy to aspirin, and/or clopidogrel.
- Subjects enrolled in concurrent studies that may confound the results of this study.
- Subjects whose clinical condition, in the investigator's opinion, would not allow them to complete the study.

# 2. Follow-up Schedule

All patients were scheduled to return for follow-up examinations at 1, 3, and 6 months and every 6 months thereafter. Table 4 identifies preoperative and postoperative evaluations and timeframes for study subjects. Adverse events and complications were recorded at all visits

Table 4. Schedule of Events

|   | Screening | Baseline | Month 1 | Month 3 | Month 6 | Every 6 Months or Study Termination  |
| --- | --- | --- | --- | --- | --- | --- |
|  Procedures | Visit 1 | Visit 2 Sensor Implant (≤2 weeks of Visit 1) | Visit 3 (30 ±7 days) | Visit 4 (90 ±14 days) | Visit 5 (180 ±14 days) | Visits 6-10 or until marketing approval (±30 day window)  |
|  Informed Consent | X |  |  |  |  |   |
|  Serum or Urine Pregnancy Test | X[1] |  |  |  |  |   |
|  Demographics | X |  |  |  |  |   |
|  Past Medical & Surgical History | X |  |  |  |  |   |
|  Blood Chemistry (Creatinine) | X |  |  |  | X |   |
|  Inclusion/Exclusion Criteria Review | X | X[2] |  |  |  |   |
|  GFR | X |  |  |  |  |   |
|  INR (if indicated) |  | X |  |  |  |   |
|  Swan-Ganz measurement |  | X |  |  |  |   |
|  Physical Examination (including weight) | X[3] | X (Abbreviated PE)[3] | X[3] | X[3] | X[3] | X[3]  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

|   | Screening | Baseline | Month 1 | Month 3 | Month 6 | Every 6 Months or Study Termination  |
| --- | --- | --- | --- | --- | --- | --- |
|  NYHA HF Classification | X | X | X | X | X | X  |
|  QOL questionnaire (Minnesota) | X |  | X | X | X | X (12 month only)  |
|  SF-12 Survey | X |  | X | X | X | X (12 month only)  |
|  EQ-5D Instrument | X |  | X | X | X | X (12 month only)  |
|  Randomization |  | X |  |  |  |   |
|  Pulmonary Artery Angiography |  | X[4] |  |  |  |   |
|  Sensor Implant |  | X |  |  |  |   |
|  Sensor Measurements |  | X[5] |  |  |  |   |
|  HF Sensor Support Questionnaire | X |  |  |  |  |   |
|  Adverse Events Assessment |  | X | X | X | X | X  |
|  Medication | X | X | X | X | X | X  |
|  Phone Contact |  |  | As needed[6] | As needed[6] | As needed[6] | As needed[6]  |

[1]Females of child bearing potential
[2]Review of clinical laboratory findings against clinical laboratory inclusion/exclusion criteria for eligibility verification
[3]Include weight, height and vital signs (temperature, blood pressure, pulse, respirations). BASELINE: Abbreviated Physical Exam (i.e., weight, vital signs and significant changes since Screening)
[4]Subjects with BMI  $>35$  and chest circumference between 52" and 65", need to have appropriately located pulmonary artery branch (defined as  $<10$  cm from the pulmonary artery branch to the skin of the back) prior to implant procedure as measured by angiography at the Baseline Visit will receive the Sensor implant.
[5]Sensor measurements will be performed for both groups of subjects, however the control group's measurements will be blinded to the physician to better reflect standard of care.
[6]Refer to section 6.1.10 of the clinical protocol

# 3. Clinical Endpoints

With regards to safety, the primary safety endpoints were tested hierarchically, in order to control for multiplicity. Employing the O'Brien Fleming analysis methodology for one (1) interim analysis, the primary safety analysis nominal significance level was set at 0.048 for the final analysis. First, the freedom from device/system-related complication (DSRC) rate was tested. If the result was statistically significant (i.e.,  $p \leq 0.048$ ), then the freedom from pressure sensor failure rate was also tested for significance (i.e.,  $p \leq 0.048$ ). The study was

PMA P100045: FDA Summary of Safety and Effectiveness Data

judged to have provided positive safety results if both tests of the primary safety analysis endpoints were statistically significant (i.e., $p \leq 0.048$).

Analysis of DSRC was based on the following objective performance criteria: the lower limit of the two-sided $95.2\%$ confidence interval on the freedom from DSRC rate for the combined patient groups at six (6) months was at least $80\%$. The statistical hypotheses were:

$\mathrm{H_0}$: $\pi$ (Freedom from device $I$ system-related complications at six months) $\leq 80\%$

$\mathrm{H}_{1}$: $\pi$ (Freedom from device $I$ system-related complications at six months) $>80\%$

Analysis of sensor failures was based on the following objective performance criteria: the lower limit of the two-sided $95.2\%$ confidence interval on the freedom from pressure sensor failure rate for the combined patient groups at six (6) months was at least $90\%$. The statistical hypotheses were:

$\mathrm{H_0}$: $\pi$ (Freedom from pressure sensor failure at six months) $\leq 90\%$

$\mathrm{H}_{1}$: $\pi$ (Freedom from pressure sensor failure at six months) $>90\%$

With regards to effectiveness, employing the O'Brien Fleming analysis methodology for one (1) interim analysis, the primary efficacy analysis nominal significance level was set at 0.048 for the final analysis. The study was judged to have provided positive efficacy results if the final efficacy result was statistically significant (i.e., $p \leq 0.048$) using the negative binomial regression procedure. The primary alternative hypothesis of interest was that the Treatment group (standard of care HF management plus HF management based upon hemodynamic information obtained from the CardioMEMS HF System) will have a lower rate of HF hospitalizations at 6 months than the control group (standard of care HF management only). The statistical hypotheses were:

$\mathrm{H_0}$: $\mu$ (Treatment Group) $= \mu$ (Control Group)

$\mathrm{H}_{\mathrm{a}}$: $\mu$ (Treatment Group) $\neq \mu$ (Control Group)

where, $\mu$ is the rate of heart failure-related hospitalizations through six (6) months.

Additionally, there were four (4) secondary effectiveness endpoints analyzed at the six (6) month visit. The statistical analysis tested the secondary effectiveness endpoints according to a hierarchical strategy in order to preserve an overall Type I error rate of $5\%$. These secondary effectiveness endpoints included:

- Change from baseline in PA mean pressures;
- Proportion of patients hospitalized for heart failure;
- Days alive outside of the hospital; and
- Quality of Life – Minnesota Living with Heart Failure Questionnaire (MLHFQ).

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 20

An interim analysis was conducted by the Data Safety Monitoring Board (DSMB) after 50% of the subjects completed at least six (6) months on study (or prematurely discontinued).

Supplementary Analyses were performed on the full duration of follow-up data (12 months) (Part 1) and included:

- analyzing the primary safety endpoints and effectiveness endpoints over the whole study duration;
- analyzing the effectiveness endpoints under the per-protocol population where some subjects were excluded; and
- performing survival analyses to compare the survival curves and HFR hospitalization free survival curves between the treatment group and the control group.

## Open Access (Part 2)

For the Open Access Period, patients were treated between August 12, 2010 and April 30, 2012. The database for this PMA reflected data collected through April 30, 2012 and included 347 patients. There were 64 investigational sites.

Following the completion of the period of Randomized Access (Part 1), the investigators continued to receive PA pressure data for Treatment group subjects, and began to receive PA pressure data for Control group subjects. In other words, subjects in Part 1 transitioned to a period of Open Access, defined as the Part 2 of the study. During Part 2, investigators received automated alerts and had access to subject PA pressure measurements for all subjects (both Treatment and Control groups) but received no CardioMEMS nurse subject-specific treatment recommendations as established by an independent third party audit.

A series of ancillary analyses were used to evaluate outcomes when all subjects' investigators received access to PA pressure information during Part 2 of the study. Part 2 study results were compared to Part 1. Specifically, the longitudinal analyses (Open Access Part 2), discussed below, focused on the changes in the HFR hospitalizations as the subjects transitioned from Part 1 to Part 2. Table 5 below outlines the differences between the study periods, randomized groups, and study components.

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 21

Table 5. Distinctions for Part 1 (Randomized Access) and Part 2 (Open Access).

|  Study Period | Randomized Group | Study Component  |   |   |
| --- | --- | --- | --- | --- |
|   |   |  Standard of Care Heart Failure Management | Physician Knowledge of PA Pressures | Nurse Communications to Enhance Protocol Compliance  |
|  Randomized Access (Part 1) | Treatment | Yes | Yes | Yes  |
|   |  Control | Yes | No | No  |
|  Open Access (Part 2) | Former Control | Yes | Yes | No  |
|   |  Former Treatment | Yes | Yes | No  |

1. Clinical Inclusion and Exclusion Criteria
The Inclusion and Exclusion Criteria did not change from Part 1.

2. Follow-up Schedule
All patients continued to be followed per the protocol. Adverse events and complications were recorded at all visits.

3. Clinical Endpoints
The longitudinal analyses were based on the Randomized Access Period of the Study (Part 1) and Open Access Period of the Study (Part 2). The Randomized Access (Part 1) focused on the differences in the rate of HFR hospitalization between Treatment and Control groups. The longitudinal analyses compared the rate of HFR hospitalization in subjects followed during the Open Access (Part 2). These longitudinal analyses were designed to further evaluate whether knowledge of PA pressures, un-confounded by nurse communications, reduced the rate of HFR hospitalizations.

CardioMEMS analyzed an intent-to-treat (ITT) population, which consisted of all subjects who were randomized into the study, regardless of study completion status. Subjects who were lost-to-follow-up, underwent VAD implantation or heart transplantation, withdrew consent or died were censored at the time of occurrence of these events. Censoring these subjects excluded the subject's subsequent events after the censoring from the analyses. CardioMEMS proposed the Anderson-Gill multiplicative hazards model to accommodate variable follow-up times as well as recurrent heart failure events using the combined Part 1 and Part 2 longitudinal data. CardioMEMS used an Anderson-Gill model with Frailty, which allows for random effects, to address the correlated data.

To assess the effect of using PA pressure measurements to guide medical therapy to prevent HFR hospitalizations, and to establish device effectiveness among females, four (4) analyses were performed: Longitudinal, Gender, Propensity, and Clinical. P-values should be interpreted with caution because the analyses including Part 2 data were not specified before the onset of the study and there are

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 22

various sources of confounding effects which cannot be separated from the treatment effect. Each analysis is briefly described below.

a. Longitudinal Analyses

i. Comparison of Former Control (Part 2) to Control (Part 1): To determine whether the HFR hospitalization rate was lower in the Former Control group than the Control group, when physicians of Former Control patients received access to PA pressures (neither had nurse communications).

ii. Comparison of Former Treatment (Part 2) to Treatment (Part 1): To evaluate whether HFR hospitalization rates remain the same in subjects whose physician's access to PA pressures remained unchanged, but no longer received nurse communications.

iii. Comparison of Former Control (Part 2) to Former Treatment (Part 2): To demonstrate that the rates of HFR hospitalizations were similar during Part 2 when both groups were managed in an identical fashion (access to PA pressure and no nurse communications).

iv. Change in HFR Hospitalization Rates in the Control group (Part 2 vs. Part 1) compared to the Change in HFR Hospitalization Rates in the Treatment Group (Part 2 vs. Part 1): To demonstrate that the magnitude of change in HFR hospitalization rates after the transition from Control to Former Control (Part 1 vs. Part 2, initiation of physician access to PA pressures in Part 2) was greater than the magnitude of change in HFR hospitalization rates after the transition from Treatment to Former Treatment (Part 1 vs. Part 2, no change in physician access to PA pressure).

CardioMEMS proposed the Anderson-Gill multiplicative hazards model to accommodate variable follow-up times as well as recurrent events using the combined Part 1 and Part 2 longitudinal data. An additional random variable, $w_{i}$, was added to the model to account for the level of frailty, where the log-frailty random variable has a normal distribution with mean zero and unknown variance $\sigma^2$. The model of the hazard rate for the $i^{\text{th}}$ subject, $i = 1,\ldots,n$, is structured as follows:

$$
\lambda_{i}(t) = \lambda_{0}(t) \exp\left\{\beta_{1} X_{1i} + \beta_{2} X_{2i} + \beta_{3} X_{3i} + \gamma w_{i}\right\}
$$

$$
X_{1} = \begin{cases} 1 & \text{Treatment Group} \\ 0 & \text{Control Group} \end{cases}
$$

$$
X_{2} = \begin{cases} 1 & \text{if } t \in \text{Part 2} \\ 0 & \text{if } t \in \text{Part 1} \end{cases}
$$

$$
X_{3} = X_{1} \cdot X_{2} = \begin{cases} 1 & \text{if } X_{1} = 1 \text{ (Treatment Group) and } X_{2} = 1 \text{ (Part 2)} \\ 0 & \text{otherwise} \end{cases}
$$

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 23

CardioMEMS performed multiple supporting analyses to evaluate the assumptions and robustness including:

- Proportional Hazards and Independence of the Recurrent Hospitalization used in the A-G model;
- Robustness of the A-G model including GEE models, non-parametric methods and non-parametric bootstrapping procedures;
- Longitudinal analyses using individual data from Part 1 and Part 2;
- Competing risk analysis to assess the impact of death when it is considered an event;
- Covariate adjusted analysis; and
- Analysis to evaluate missing data.

b. Gender Analysis

An ancillary subgroup analysis of 6 month HFR hospitalization based on gender was performed in an effort to address the issues raised regarding the Randomized Access Period of the Study (Part 1) gender analysis. The initial gender analysis was a post hoc analysis, which compared HFR hospitalization rates between males and females in the Treatment and Control groups. The initial gender analysis noted a statistically significant treatment by gender interaction. In order to examine whether the treatment-by-gender interaction was driven by early deaths in the Control group females, the composite endpoint of "death or first HFR hospitalization" was analyzed using a Cox proportional hazard model over Part 1 and the full duration of Part 1 plus Part 2. The concern was that death had created a significant competing risk problem in the Control group women and therefore led to lower HFR hospitalization rates, since early death precludes the possibility of further HFR hospitalizations. In addition, the endpoints of time to first HFR hospitalization over Part 1 and over full Duration Part 1 plus Part 2 were assessed in the Cox proportional hazard model.

To demonstrate the robustness of the findings, CardioMEMS performed the composite endpoints of recurrent HFR hospitalization or death (death is treated as a HFR hospitalization) over Part 1 and over full Duration Part 1 + Part 2 using Andersen-Gill model with robust sandwich estimates, Anderson-Gill model with Frailty and using the Negative Binomial Regression.

c. Propensity Analysis

Part 1 results were analyzed after excluding all Treatment group subjects whose treating investigators received a patient-specific CardioMEMS nurse recommendation. In order to have an adequate Control group for this analysis, a Propensity Score model was developed prior to any final data analysis. The Treatment group (N=270) was divided based on whether the study subjects were the topic of a nurse communication. Those patients in the Treatment group who

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 24

were never the topic of a nurse communication were placed in the Treatment No Nurse Communications (TNNC) group (N=99). The propensity modeling, using one-to-one nearest neighbor approach, matched the cohort of TNNC group with a comparable group of subjects in the Control group (N=99). An independent statistician evaluated and included the baseline variables in the final propensity model, using backward elimination, with the threshold for retaining a variable in the model at p<0.3. A Propensity Model with all covariates forced into the model was also considered. The propensity score was calculated using logistic regression with a treatment indicator as the outcome variable. The treatment indicator identified whether a patient belonged to the TNNC group or the Control group.

Based on the two (2) propensity score models, 30 sets of matched data were generated. These sets were generated to explore the robustness of the matching due to the dependency of the matching procedure on the sorting order. Each matched data set has a different random sorting of the 99 TNNC participants prior to the matching and 99 matched participants in the Control group based on their estimated propensity scores. For each matched data set, the independent statistician identified and quantified the potential imbalance that existed between the two (2) groups prior to performing the propensity score modeling. The approaches that evaluated the potential imbalances included Wilcoxon rank sum tests, variance ratios, standardized differences in performance, quantile-quantile (Q-Q) plots, and distribution plots for continuous variables; Fisher's Exact test, and observed proportions were used for categorical variables.

The final propensity model and the matched data were provided to a separate and independent 3rd party data analysis center for the outcome analysis, i.e. 6 month HFR hospitalization rates (the same as the pre-specified primary effectiveness endpoint).

d. Clinical Analysis

The clinical impact of the nurse communications with the goal of identifying and discussing the potential influence on the rate of HFR hospitalization was assessed. Two (2) cardiologists, acting independently of each other, with expertise in HF and clinical trials and who had not been involved in the design, recruitment, execution, or initial analysis of the CHAMPION study, performed the clinical analyses. These two (2) cardiologists identified and reviewed every email and logged phone communication between CardioMEMS and the investigators. A nurse communication was defined as potentially providing a treatment recommendation if the text of the communication referred to the potential desirability of, or the need for, a change in a specific type of medication or treatment, regardless of whether the text referred to a class of drug, a specific agent by name, or a specific dose or specific route of administration. A treatment recommendation and a medication change were considered 'concordant' if the medication change took place within a specified period of having received the

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 25

nurse communication. The analysis was conducted using four (4) different concordance time periods (0-1, 0-2, 0-3, and 0-7 days). Furthermore, the medication change was labeled as 'consistent' or 'not consistent' with the study protocol. The cardiologists classified the nurse communications as either:

- Concordant use of drugs consistent with study protocol and hypothesis.
- Concordant use of drugs not consistent with study protocol and hypotheses.
- No concordant medication change.

After classification, the cardiologists estimated the percentage of the treatment effect that may have been related to nurse communications.

Importantly, this analysis also served to establish the appropriate time point after which the data should not be considered potentially biased by nurse communications.

## B. Accountability of PMA Cohort

At the time of database lock, a total of 550 subjects were implanted with the device and then randomized 1:1 to either the Treatment group (n=270 subjects) or to the Control group (n=280 subjects). A total of 347 subjects (177 in the Treatment group and 170 in the Control group) completed the full period of Randomized Access (Part 1). During the course of Part 1, 93 subjects in the Treatment group and 110 subjects in the Control group exited for reasons described in Figure 6 below. The average duration of follow-up for Part 1 was 533.5 days in the Treatment group and 524.7 days in the Control group. For Part 2, the average duration of follow-up was 372.7 days in the Former Treatment group and 405.4 days in the Former Control group.

Subject demographics and medical history were reasonably matched between the Treatment and Control groups in Part 1 and between the Former Treatment and Former Control in Part 2 in regard to their original baseline characteristics which were measured prior to the onset of Part 1.

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 26

Figure 6. Patient Disposition
![img-5.jpeg](img-5.jpeg)
Additionally, Table 6 shows an assessment of the deaths that occurred in Part 1 and Part 2 of the study. There was a relative reduction in the death rate of  $29\%$  ((17.5% - 12.4%) ÷ 17.5%) comparing the Former Control group to the Former Treatment group.

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 6. Deaths Occurring in Part 1 and Part 2

|   | Part 1 | Part 2  |
| --- | --- | --- |
|  Deaths in Treatment Arm | 50/270 (18.5%) | 31/177 (17.5%)  |
|  Cardiac | 40/270 (14.8%) | 25/177 (14.1%)  |
|  Non-Cardiac | 10/270 (3.7%) | 6/177 (3.4%)  |
|  Deaths in Control Arm | 64/280 (22.9%) | 21/170 (12.4%)  |
|  Cardiac | 49/280 (17.5%) | 17/170 (10.0%)  |
|  Non-Cardiac | 15/280 (5.3%) | 4/170 (2.4%)  |

# C. Study Population Demographics and Baseline Parameters

The demographics of the study population are typical for a heart failure study performed in the US.

Patient Demographics for Part 1 and Part 2

Table 7 includes the patient demographics at the time of enrollment for the Randomized Access (Part 1). Table 8 includes the patient demographics as patients transferred into the Open Access (Part 2).

Table 7. Part 1 Population Demographics and Baseline Parameters

|  Variables | Randomized Group |   | p-value[1]  |
| --- | --- | --- | --- |
|   |  Treatment (N=270) | Control (N=280)  |   |
|  Age (years) | 61.3 ± 12.98 (270) | 61.8 ± 12.73 (280) | 0.5927  |
|  Male | 194/270 (71.9%) | 205/280 (73.2%) | 0.7745  |
|  Race (White) | 196/270 (72.6%) | 205/280 (73.2%) | 0.9236  |
|  Systolic BP (mmHg) | 121.2 ± 22.52 (270) | 123.2 ± 21.01 (280) | 0.1286  |
|  Heart Rate (bpm) | 72.4 ± 12.91 (269) | 73.0 ± 12.14 (280) | 0.4873  |
|  BMI | 30.5 ± 6.50 (270) | 30.9 ± 7.35 (280) | 0.6228  |
|  BUN (mg/dL) | 29.6 ± 17.99 (248) | 28.1 ± 16.17 (267) | 0.6325  |
|  Creatinine (mg/dL) | 1.4 ± 0.47 (270) | 1.4 ± 0.42 (280) | 0.5560  |
|  GFR (mL/min/1.73m²) | 60.4 ± 22.50 (270) | 61.8 ± 23.20 (280) | 0.5638  |
|  Ejection Fraction (EF>=40%) | 62/270 (23.0%) | 57/279 (20.4%) | 0.5343  |
|  Cardiac Output (L/min) | 4.5 ± 1.41 (270) | 4.6 ± 1.54 (278) | 0.5499  |
|  Cardiac Index (L/min/m²) | 2.1 ± 0.59 (270) | 2.2 ± 0.64 (278) | 0.4405  |
|  PVR | 2.9 ± 2.02 (270) | 2.7 ± 1.82 (278) | 0.4609  |
|  PA Wedge Pressure (mmHg) | 17.5 ± 7.97 (270) | 19.0 ± 8.12 (280) | 0.0276  |
|  PA Mean Pressure (mmHg) | 28.9 ± 9.92 (270) | 29.9 ± 10.05 (280) | 0.3021  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 8. Part 2 Population Demographics and Baseline Parameters

|  Variables | Group |   | p-value[1]  |
| --- | --- | --- | --- |
|   |  Former Treatment (N=177) | Former Control (N=170)  |   |
|  Age (years) | 60.5 ± 12.14 (177) | 60.0 ± 12.78 (170) | 0.8506  |
|  Male | 123/177 (69.5%) | 119/170 (70.0%) | 1.0000  |
|  Race (White) | 123/177 (69.5%) | 119/170 (70.0%) | 1.0000  |
|  Systolic BP (mmHg) | 121.8 ± 22.85 (177) | 123.4 ± 19.98 (170) | 0.2815  |
|  Heart Rate (bpm) | 71.2 ± 11.55 (177) | 71.6 ± 11.48 (170) | 0.7648  |
|  BMI | 31.1 ± 6.41 (177) | 31.6 ± 7.56 (170) | 0.6047  |
|  BUN (mg/dL) | 27.8 ± 16.85 (162) | 26.2 ± 13.70 (159) | 0.7724  |
|  Creatinine (mg/dL) | 1.4 ± 0.44 (177) | 1.3 ± 0.41 (170) | 0.7778  |
|  GFR (mL/min/1.73m2) | 61.8 ± 22.33 (177) | 63.2 ± 23.63 (170) | 0.6676  |
|  Ejection Fraction (EF>=40%) | 44/177 (24.9%) | 39/169 (23.1%) | 0.7076  |
|  Cardiac Output (L/min) | 4.6 ± 1.34 (177) | 4.8 ± 1.48 (168) | 0.2931  |
|  Cardiac Index (L/min/m2) | 2.2 ± 0.58 (177) | 2.3 ± 0.60 (168) | 0.1324  |
|  PVR | 2.6 ± 1.73 (177) | 2.4 ± 1.66 (168) | 0.2451  |
|  PMR | 2.6 ± 1.73 (177) | 2.4 ± 1.66 (168) | 0.2451  |

[1] Wilcoxon Rank-Sum Test for continuous measures and Fisher's exact test for categorical measures.

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 9 summarizes the duration of patient participation during the trial. For the Randomized Access Study (Part 1), the average duration of follow-up was 533.5 days in the Treatment group and 524.7 days in the Control group. For the Open Access Study (Part 2), the average duration of follow-up was 372.7 days in the Former Treatment group and 405.4 days in the Former Control group. The total number of patient years was 797 for Part 1 and 1,166 for Part 1 + Part 2 combined.

Table 9. Patient Follow Up Duration (Days) in Study: Part 1 & Full Study Duration

|   | Treatment (270) | Control (280) | All Patients (550)  |
| --- | --- | --- | --- |
|  Part 1 Follow-up (days)  |   |   |   |
|  Mean±StdDev (N) | 533.5±236.9 (270) | 524.7±231.8 (280) | 529.0±234.1 (550)  |
|  Median | 521.5 | 524.5 | 523.0  |
|  (Min, Max) | (4, 1,036) | (1, 1,010) | (1, 1,036)  |
|  Total Patient Days | 144,054 | 146,910 | 290,964  |
|  Part 1 + Part 2 Follow-up (days)  |   |   |   |
|  Mean±StdDev (N) | 777.9±353.3 (270) | 770.8±353.3 (280) | 774.3±353.0 (550)  |
|  Median | 951.0 | 946.0 | 949.5  |
|  (Min, Max) | (4, 1,258) | (1, 1,308) | (1, 1,308)  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

Patient Duration (Days) in Study: Part 2

|   | Former Treatment (177) | Former Control (170) | All Patients (347)  |
| --- | --- | --- | --- |
|  Part 2 Follow-up (days)  |   |   |   |
|  Mean±StdDev (N) | 372.7±182.6 (177) | 405.4±187.8 (170) | 388.7±185.6 (347)  |
|  Median | 345.0 | 397.0 | 372.0  |
|  (Min, Max) | (31, 662) | (25, 662) | (25, 662)  |
|  Total Patient Days | 65,968 | 68,911 | 134,879  |

A pre-specified analysis on the background Heart Failure Medical Therapy was performed, as shown in Tables 10 and 11. The p-values in the table were not adjusted for multiplicity.

Table 10. Baseline HF Drug Therapy

|  HF Medication | Treatment (270) | Control (280) | All Patients (550) | p-value[1]  |
| --- | --- | --- | --- | --- |
|  ACE/ARB | 205 (75.9%) | 222 (79.3%) | 427 (77.6%) | 0.3584  |
|  Beta Blocker | 243 (90.0%) | 256 (91.4%) | 499 (90.7%) | 0.6595  |
|  Aldosterone Antagonist | 117 (43.3%) | 114 (40.7%) | 231 (42.0%) | 0.5463  |
|  Nitrate | 64 (23.7%) | 56 (20.0%) | 120 (21.8%) | 0.3035  |
|  Hydralazine | 36 (13.3%) | 33 (11.8%) | 69 (12.5%) | 0.6084  |
|  Diuretic-Loop | 248 (91.9%) | 258 (92.1%) | 506 (92.0%) | >0.9999  |
|  Diuretic-Thiazide-Standing | 30 (11.1%) | 35 (12.5%) | 65 (11.8%) | 0.6922  |
|  Diuretic-Thiazide-PRN | 20 (7.4%) | 18 (6.4%) | 38 (6.9%) | 0.7374  |

[1]p-value testing Treatment vs. Control obtained from Fisher's Exact Test.

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 11. HF Drug Therapy at 6 months

|  HF Medication | Treatment (270) | Control (280) | All Patients (550) | p-value[1]  |
| --- | --- | --- | --- | --- |
|  ACE/ARB | 203 (75.2%) | 213 (76.1%) | 416 (75.6%) | 0.8428  |
|  Beta Blocker | 236 (87.4%) | 246 (87.9%) | 482 (87.6%) | 0.8975  |
|  Aldosterone Antagonist | 130 (48.1%) | 124 (44.3%) | 254 (46.2%) | 0.3926  |
|  Nitrate | 113 (41.9%) | 65 (23.2%) | 178 (32.4%) | <0.0001  |
|  Hydralazine | 61 (22.6%) | 42 (15.0%) | 103 (18.7%) | 0.0285  |
|  Diuretic-Loop | 239 (88.5%) | 251 (89.6%) | 490 (89.1%) | 0.6840  |
|  Diuretic-Thiazide-Standing | 53 (19.6%) | 41 (14.6%) | 94 (17.1%) | 0.1407  |
|  Diuretic-Thiazide-PRN | 33 (12.2%) | 30 (10.7%) | 63 (11.5%) | 0.5948  |

$^{[1]}$ p-value testing Treatment vs. Control obtained from Fisher's Exact Test.

At 6 months, the Treatment group had a significantly greater proportion of patients on nitrates (41.9%) compared to Control (23.2%). There was also a greater proportion of patients on hydralazine in the Treatment group (22.6%) compared to Control (15.0%). The proportion of subjects taking ACE-I/ARB's, B-blockers, aldosterone antagonists, and other diuretics was similar between the two (2) groups.

# D. Safety and Effectiveness Results

# 1. Safety Results

The analysis of safety was based on the heart failure cohort of 575 patients who underwent right heart catheterization (RHC). Of these 575 patients, 25  $(4.3\%)$  underwent a RHC, but did not receive an implant primarily because of anatomical/physiological conditions identified during the catheterization. The key safety outcomes for this study were evaluated at 6 months and are presented below in Table 12. Adverse effects are reported in Table 13.

# a. Primary (Randomized Access (Part 1)) Safety Endpoint #1

This endpoint captured freedom from a DSRC through 6 months. It was tested against a pre-specified performance goal of  $80\%$ . The performance goal is similar to performance goals FDA has accepted for other permanent implants for heart failure devices. The analysis population included all subjects consented who had a right heart catheterization attempted.

There were 567 patients out of 575 patients that were free from DSRC. The freedom from DSRC rate was  $98.6\%$ , with a  $95.2\%$  lower confidence bound (LCB) of  $97.3\%$ . The pre-specified performance goal was  $80\%$ . The endpoint was met. These results are listed in Table 12 below.

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 12. Primary Safety Endpoint #1- DSRC at 6 Months

|  Acute Safety Results | Sample Size (N=575)  |
| --- | --- |
|  Number of patients free from a DSRC | 567 (98.6%)  |
|  95.2% Lower Confidence Boundry[1] | 97.3%  |
|  p-value of H0: Rate ≥80% | <0.0001[2]  |

[1]Exact  $95.2\%$  Clopper-Pearson lower confidence limit
[2]p-value from exact test of binomial proportions compared to  $80\%$  for all patients.

The adverse events included in the acute primary safety endpoint analysis are summarized in Table 13 below.

Table 13. Adverse Events in the Primary Safety Endpoint

|  Description | Number of Subjects with Device or System related complication (%) (N = 575)  |
| --- | --- |
|  Hemoptysis | 1 (0.2%)  |
|  Sensor did not deploy | 1 (0.2%)  |
|  Transient Ischemic Attack (TIA) | 1 (0.2%)  |
|  Atypical chest pain | 1 (0.2%)  |
|  Sepsis → death | 1 (0.2%)  |
|  Atrial arrhythmia → death | 1 (0.2%)  |
|  Arterial embolism (upper extremity) | 1 (0.2%)  |
|  Pulmonary artery (in-situ) thrombus | 1 (0.2%)  |
|  Total Subjects Experiencing a DSRC | 8 (1.4% [1], 95.2% LCB 97.3%)  |

[1]DSRCs (8 total) by group: Consented by not randomized (2), Treatment (3), Control (3)

b. Primary (Randomized Access (Part 1)) Safety Endpoint #2

This endpoint captured the freedom from pressure sensor failure rate through 6 months. It was tested against a pre-specified performance goal of  $90\%$ . The analysis cohort included all subjects that had an investigational sensor implanted.

There were zero (0) pressure sensor failures out of 550 implanted devices. The freedom from pressure sensor failure rate was  $100\%$  with a  $95.2\%$  LCB of  $99.3\%$ . The pre-specified performance goal was  $90\%$ . This endpoint was met.

PMA P100045: FDA Summary of Safety and Effectiveness Data

c. Open Access (Part 2) Safety Results

There were no Unanticipated Serious Adverse Device Events, Serious Adverse Device Events, Non-Serious Adverse Device Events, or Device-System Related Complications. In addition, there were no sensor failures over the entire study duration (mean follow-up of 26 months, range: 1 day – 44 months).

d. Adverse Events

Tables 14 through 17 identify the adverse events observed during both Part 1 and Part 2.

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 34

Table 14. Non-serious Adverse Events Not Related to the Device Over Part 1 and Part 2

|   | Part 1 |   |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
|   | Treatment (270) |   | Control (280) |   | All Patients (550) |   | All Patients (347)  |   |
|   | Subjects | Events | Subjects | Events | Subjects | Events | Subjects | Events  |
|  All Patients with an Event | 216 (80.0%) | 1229 | 223 (79.6%) | 1135 | 439 (79.8%) | 2364 | 219 (63.1%) | 787  |
|  Blood and lymphatic system disorders | 27 (10.0%) | 37 | 22 (7.9%) | 28 | 49 (8.9%) | 65 | 13 (3.7%) | 16  |
|  Cardiac disorders | 81 (30.0%) | 140 | 69 (24.6%) | 117 | 150 (27.3%) | 257 | 49 (14.1%) | 71  |
|  Congenital, familial and genetic disorders | 0 (0.0%) | 0 | 0 (0.0%) | 0 | 0 (0.0%) | 0 | 3 (0.9%) | 3  |
|  Ear and labyrinth disorders | 6 (2.2%) | 6 | 2 (0.7%) | 2 | 8 (1.5%) | 8 | 2 (0.6%) | 2  |
|  Endocrine disorders | 4 (1.5%) | 4 | 9 (3.2%) | 10 | 13 (2.4%) | 14 | 7 (2.0%) | 7  |
|  Eye disorders | 12 (4.4%) | 12 | 14 (5.0%) | 16 | 26 (4.7%) | 28 | 7 (2.0%) | 8  |
|  Gastrointestinal disorders | 64 (23.7%) | 104 | 60 (21.4%) | 96 | 124 (22.5%) | 200 | 48 (13.8%) | 70  |
|  General disorders and administration site conditions | 64 (23.7%) | 102 | 45 (16.1%) | 80 | 109 (19.8%) | 182 | 50 (14.4%) | 62  |
|  Hepatobiliary disorders | 1 (0.4%) | 1 | 7 (2.5%) | 10 | 8 (1.5%) | 11 | 3 (0.9%) | 3  |
|  Immune system disorders | 4 (1.5%) | 4 | 4 (1.4%) | 4 | 8 (1.5%) | 8 | 4 (1.2%) | 4  |
|  Infections and infestations | 76 (28.1%) | 129 | 91 (32.5%) | 150 | 167 (30.4%) | 279 | 65 (18.7%) | 99  |
|  Injury, poisoning and procedural complications | 32 (11.9%) | 44 | 32 (11.4%) | 37 | 64 (11.6%) | 81 | 32 (9.2%) | 43  |
|  Investigations | 32 (11.9%) | 51 | 26 (9.3%) | 40 | 58 (10.5%) | 91 | 22 (6.3%) | 25  |
|  Metabolism and nutrition disorders | 66 (24.4%) | 116 | 52 (18.6%) | 88 | 118 (21.5%) | 204 | 37 (10.7%) | 53  |
|  Musculoskeletal and connective tissue disorders | 49 (18.1%) | 75 | 58 (20.7%) | 73 | 107 (19.5%) | 148 | 56 (16.1%) | 70  |
|  Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 6 (2.2%) | 8 | 9 (3.2%) | 9 | 15 (2.7%) | 17 | 6 (1.7%) | 7  |

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 35

|   | Part 1 |   |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
|   | Treatment (270) |   | Control (280) |   | All Patients (550) |   | All Patients (347)  |   |
|   | Subjects | Events | Subjects | Events | Subjects | Events | Subjects | Events  |
|  Nervous system disorders | 61 (22.6%) | 86 | 50 (17.9%) | 67 | 111 (20.2%) | 153 | 47 (13.5%) | 56  |
|  Psychiatric disorders | 34 (12.6%) | 46 | 29 (10.4%) | 36 | 63 (11.5%) | 82 | 25 (7.2%) | 31  |
|  Renal and urinary disorders | 33 (12.2%) | 55 | 35 (12.5%) | 45 | 68 (12.4%) | 100 | 21 (6.1%) | 21  |
|  Reproductive system and breast disorders | 7 (2.6%) | 8 | 16 (5.7%) | 16 | 23 (4.2%) | 24 | 11 (3.2%) | 13  |
|  Respiratory, thoracic and mediastinal disorders | 68 (25.2%) | 97 | 70 (25.0%) | 117 | 138 (25.1%) | 214 | 47 (13.5%) | 66  |
|  Skin and subcutaneous tissue disorders | 23 (8.5%) | 26 | 24 (8.6%) | 28 | 47 (8.5%) | 54 | 9 (2.6%) | 9  |
|  Surgical and medical procedures | 17 (6.3%) | 21 | 16 (5.7%) | 20 | 33 (6.0%) | 41 | 16 (4.6%) | 19  |
|  Vascular disorders | 41 (15.2%) | 57 | 39 (13.9%) | 46 | 80 (14.5%) | 103 | 27 (7.8%) | 29  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

Table 15. Serious Adverse Events Over Part 1 and Part 2

|   | Part 1 |   |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
|   |  Treatment (270)Subjects Events |   | Control (280)Subjects Events |   | All Patients (550)Subjects Events |   | All Patients (347)Subjects Events  |   |
|  All Patients with an Event | 198(73.3%) | 797 | 217(77.5%) | 956 | 415(75.5%) | 1753 | 201(57.9%) | 647  |
|  Cardiac disorders | 138(51.1%) | 333 | 151(53.9%) | 443 | 289(52.5%) | 776 | 119(34.3%) | 238  |
|  Congestive heart failure | 99 | 204 | 121 | 274 | 220 | 478 | 80 | 140  |
|  Heart failure | 10 | 15 | 13 | 22 | 23 | 37 | 10 | 12  |
|  Ventricular tachycardia* | 10 | 14 | 12 | 16 | 22 | 30 | 8 | 9  |
|  Myocardial infarction* | 7 | 7 | 14 | 14 | 21 | 21 | 9 | 9  |
|  Cardiac pain | 13 | 19 | 7 | 22 | 20 | 41 | 0 | 0  |
|  Atrial fibrillation* | 3 | 5 | 10 | 11 | 13 | 16 | 4 | 4  |
|  Cardiomyopathy | 5 | 6 | 8 | 11 | 13 | 17 | 6 | 7  |
|  Cardiopulmonary arrest | 3 | 3 | 7 | 7 | 10 | 10 | 3 | 3  |
|  Unstable angina | 4 | 4 | 5 | 5 | 9 | 9 | 4 | 6  |
|  Coronary artery disease | 5 | 5 | 3 | 3 | 8 | 8 | 4 | 4  |
|  Ventricular arrhythmia* | 3 | 3 | 5 | 7 | 8 | 10 | 0 | 0  |
|  Ventricular fibrillation* | 5 | 6 | 2 | 2 | 7 | 8 | 2 | 2  |
|  Anginal discomfort | 1 | 1 | 5 | 8 | 6 | 9 | 2 | 4  |
|  Cardiac arrest | 2 | 2 | 4 | 4 | 6 | 6 | 6 | 6  |
|  Ischemic cardiomyopathy | 3 | 4 | 3 | 3 | 6 | 7 | 6 | 7  |
|  Atrial flutter* | 2 | 2 | 3 | 3 | 5 | 5 | 3 | 3  |
|  Cardiogenic shock | 2 | 2 | 3 | 3 | 5 | 5 | 3 | 3  |
|  Acute decompensated heart failure | 2 | 2 | 1 | 1 | 3 | 3 | 0 | 0  |
|  ADHF | 2 | 2 | 0 | 0 | 2 | 2 | 0 | 0  |
|  Acute coronary syndrome | 1 | 2 | 1 | 1 | 2 | 3 | 1 | 1  |
|  Arrhythmia* | 1 | 1 | 1 | 1 | 2 | 2 | 1 | 1  |
|  Atrial arrhythmia* | 1 | 1 | 1 | 1 | 2 | 2 | 0 | 0  |
|  Cardiac failure | 0 | 0 | 2 | 2 | 2 | 2 | 0 | 0  |
|  Heart disease, unspecified | 1 | 1 | 1 | 1 | 2 | 2 | 1 | 1  |

PMA P100045: FDA Summary of Safety and Effectiveness Data
Page 37

|   | Part 1 |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- |
|   |  Treatment (270)Subjects Events |   | Control (280)Subjects Events |   | All Patients (550)Subjects Events | All Patients (347)Subjects Events  |   |
|   |  Subjects | Events | Subjects | Events | Subjects | Events | Subjects  |
|  Non-ischemic cardiomyopathy | 1 | 1 | 1 | 1 | 2 | 2 | 0  |
|  Sick sinus syndrome | 1 | 1 | 1 | 1 | 2 | 2 | 0  |
|  Angina unstable | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Arrhythmia ventricular* | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Arrhythmia ventricular (NOS) * | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Atrial tachycardia* | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Bradycardia* | 0 | 0 | 1 | 1 | 1 | 1 | 2  |
|  Bradycardia-tachycardia syndrome | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Cardiac arrhythmia* | 1 | 1 | 0 | 0 | 1 | 1 | 1  |
|  Cardiomegaly | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Cardiorenal syndrome | 1 | 1 | 0 | 0 | 1 | 1 | 1  |
|  Chronic heart failure | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Congestive cardiac failure aggravated | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Coronary artery disease progression | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Coronary atherosclerosis | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Coronary spasm | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Decompensated heart failure | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  End stage cardiac failure | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Heart failure, congestive | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Heart valve incompetence | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Intermediate coronary syndrome | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Junctional tachycardia* | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Left ventricular dysfunction | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Mitral valve incompetence | 1 | 1 | 0 | 0 | 1 | 1 | 1  |
|  Multi-valvular regurgitation | 0 | 0 | 1 | 1 | 1 | 1 | 0  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

|   | Part 1 |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- |
|   |  Treatment (270)Subjects Events |   | Control (280)Subjects Events |   | All Patients (550)Subjects Events | All Patients (347)Subjects Events  |   |
|   |  Subjects | Events | Subjects | Events | Subjects | Events | Subjects  |
|  Non ST segment elevation myocardial infarction* | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Non-sustained ventricular tachycardia* | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Pacemaker mediated tachycardia | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Paroxysmal supraventricular tachycardia* | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Pericardial disease | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Pericardial effusion | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Pericarditis | 0 | 0 | 1 | 1 | 1 | 1 | 1  |
|  Premature ventricular contractions* | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Supraventricular tachycardia* | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Sustained ventricular tachycardia* | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Tachycardia* | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Tricuspid insufficiency | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Ventricular ectopic beats | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Ventricular rhythm* | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Wide complex tachycardia | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Wide complex ventricular tachycardia* | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Asystole | 0 | 0 | 0 | 0 | 0 | 2 | 2  |
|  Congestive cardiomyopathy | 0 | 0 | 0 | 0 | 0 | 1 | 1  |
|  End stage heart disease | 0 | 0 | 0 | 0 | 0 | 1 | 1  |
|  Hemopericardium | 0 | 0 | 0 | 0 | 0 | 1 | 1  |
|  Palpitation | 0 | 0 | 0 | 0 | 0 | 1 | 1  |
|  Paroxysmal atrial fibrillation | 0 | 0 | 0 | 0 | 0 | 1 | 1  |
|  Polymorphic ventricular tachycardia | 0 | 0 | 0 | 0 | 0 | 1 | 1  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

|   | Part 1 |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- |
|   |  Treatment (270) |   | Control (280) |   | All Patients (550) | All Patients (347)  |   |
|   |  Subjects | Events | Subjects | Events | Subjects | Events | Subjects  |
|  Tachycardia supraventricular | 0 | 0 | 0 | 0 | 0 | 1 | 1  |
|  Infections and infestations | 45 (16.7%) | 62 | 61 (21.8%) | 90 | 106 (19.3%) | 52 (15.0%) | 76  |
|  Pneumonia | 11 | 11 | 15 | 16 | 26 | 27 | 17  |
|  Urinary tract infection | 5 | 7 | 5 | 6 | 10 | 13 | 5  |
|  Bronchitis | 3 | 3 | 5 | 6 | 8 | 9 | 3  |
|  Cellulitis | 1 | 1 | 6 | 7 | 7 | 8 | 1  |
|  Sepsis | 3 | 4 | 4 | 4 | 7 | 8 | 7  |
|  Acute bronchitis | 1 | 1 | 4 | 4 | 5 | 5 | 2  |
|  Bacteremia | 1 | 1 | 3 | 5 | 4 | 6 | 2  |
|  Upper respiratory infection | 2 | 2 | 2 | 2 | 4 | 4 | 1  |
|  Influenza | 3 | 3 | 0 | 0 | 3 | 3 | 0  |
|  Cellulitis of leg | 0 | 0 | 2 | 2 | 2 | 2 | 0  |
|  Cellulitis of legs | 0 | 0 | 2 | 2 | 2 | 2 | 0  |
|  Central line infection | 0 | 0 | 2 | 2 | 2 | 2 | 2  |
|  Endocarditis | 0 | 0 | 2 | 2 | 2 | 2 | 0  |
|  Foot infection | 2 | 3 | 0 | 0 | 2 | 3 | 0  |
|  Gastroenteritis | 2 | 3 | 0 | 0 | 2 | 3 | 3  |
|  Incision site infection | 1 | 3 | 1 | 4 | 2 | 7 | 1  |
|  Infection | 0 | 0 | 2 | 2 | 2 | 2 | 1  |
|  Osteomyelitis | 1 | 1 | 1 | 1 | 2 | 2 | 1  |
|  Pyelonephritis | 1 | 1 | 1 | 1 | 2 | 2 | 0  |
|  Respiratory infection | 1 | 1 | 1 | 1 | 2 | 2 | 0  |
|  Viral gastroenteritis | 1 | 1 | 1 | 1 | 2 | 2 | 0  |
|  Abscess | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Acute diverticulitis | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  Acute pyelonephritis | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Bacterial endocarditis | 0 | 0 | 1 | 1 | 1 | 1 | 0  |
|  Bacterial infection | 1 | 1 | 0 | 0 | 1 | 1 | 0  |
|  C.difficile colitis | 1 | 1 | 0 | 0 | 1 | 1 | 0  |

PMA P100045: FDA Summary of Safety and Effectiveness Data

|   | Part 1 |   |   |   |   | Part 2  |   |
| --- | --- | --- | --- | --- | --- | --- | --- |
|   |  Treatment (270)Subjects Events |   | Control (280)Subjects Events |   | All Patients (550)Subjects Events | All Patients (347)Subjects Events  |   |
|   |  |   |   |   |   |   |   |
|  Catheter site infection | 0 | 0 | 1 | 1 | 1 | 0 | 0  |
|  Cellulitis of arm | 0 | 0 | 1 | 1 | 1 | 0 | 0  |
|  Cellulitis of hand | 1 | 1 | 0 | 0 | 1 | 0 | 0…

---

**Source:** [https://fda.innolitics.com/device/P100045](https://fda.innolitics.com/device/P100045)

**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