The CardioMEMS Heart Failure Sensor: A Procedural Guide for Implanting Physicians

Abstract: The CardioMEMS heart failure system is a recent Food and Drug Administration approved device that can be implanted in patients with New York Heart Association class III heart failure and allows remote monitoring of pulmonary artery pressures. There is limited published information regarding the implantation procedure. Successful use of the CardioMEMS heart failure system requires an understanding of the technical issues surrounding the implantation procedure. The goal of the present review is to provide a summary of the implantation procedure, discuss the required imaging steps, review procedural supplies, and present a series of case studies to illustrate clinically relevant issues that may arise during sensor implantation.

J INVASIVE CARDIOL 2016;28(7):273-279

Key words: CardioMEMS heart failure sensor, heart failure, new devices, interventional devices, innovation

Device-based therapy for patients with congestive heart failure (CHF) continues to evolve.^1-4 The CardioMEMS heart failure (HF) system is a recent Food and Drug Administration (FDA)-approved device that is indicated for patients with New York Heart Association (NYHA) class III symptoms and a prior hospitalization for CHF within the last year, regardless of ejection fraction. The system includes a sensor that is percutaneously placed into a branch of the pulmonary artery and home and hospital electronics units that are used to interrogate the sensor. The system provides an 18 second measurement of the systolic, diastolic, and mean pulmonary artery pressures, thereby allowing adjustment of HF medications based upon pressure trends and specified pressure goals. In the randomized CHAMPION (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients) study, interventional, electrophysiology, heart failure specialists, and invasive cardiologists performed the implant procedure.⁵ The goal of the present review is to provide a summary of the implant procedure, discuss the required imaging steps, review procedural supplies, and present a series of case studies to illustrate clinically relevant issues that may arise during sensor implantation. 

Implantation Procedure

Components of the CardioMEMS HF system are shown in Figure 1. The sensor comes attached to an over-the-wire delivery catheter that accommodates a 0.018˝ exchange-length guidewire. Counter-clockwise rotation of the tethering cord releases the sensor and allows the two nitinol hoops to contact the adjacent pulmonary artery branch. Prior to the procedure, contraindications for sensor implantation should be reviewed (Table 1). The implantation procedure involves the steps outlined in Table 2. Variation of these steps may be required in the setting of difficult anatomy and/or patient-related issues. 

Venous access. Venous access is obtained percutaneously via the right common femoral vein with an 8 Fr sheath. Current instructions for use specify that femoral access should be used. Although right internal jugular venous access would be considered “off-label,” this approach may be useful in patients with morbid obesity and/or severe tricuspid regurgitation to aid in placement of the right heart catheter. The sheath is upsized to a standard, commercially available 12 cm or 15 cm long 12 Fr sheath. For patients receiving novel oral anticoagulants or warfarin, venous access is commonly done without interruption of these agents. Anticoagulation is not required during the procedure. 

Measurement of pressures and cardiac output. Accurate measurements of cardiac pressures and cardiac output are required as part of the implantation procedure in order assess the baseline hemodynamic status of the patient. A 7 Fr balloon wedge catheter (Arrow International) provides the ability to insert a 0.035˝ guidewire into the distal portwhen advancement through the right heart is challenging, allows improved torque control when advancing the catheter to the left pulmonary artery, and provides optimal imaging of the pulmonary artery (Figure 2). Although a standard 7 Fr Swan-Ganz catheter (Edwards Lifesciences) allows assessment of pressures and cardiac output, imaging via the distal pulmonary artery port is limited because of the smaller catheter lumen (0.021˝). The tricuspid valve should be crossed using balloon floatation (ie, with the balloon inflated) to avoid potential entanglement with tricuspid valve structures during sensor advancement (see below). 

Pulmonary artery access. Ideal sensor location is within an inferior and lateral branch of the left pulmonary artery measuring 7-15 mm in diameter. As opposed to right-sided pulmonary artery branches, left-sided branches tend to course in a posterior orientation such that interrogation of the sensor is more effective. Sensors have been placed within branches of the right pulmonary artery and although they allow measurement of pulmonary artery pressures, they may be limited by poor signal strength. Placement of the sensor in a right pulmonary artery branch should be considered when the size of left pulmonary artery branches is not adequate. Engagement of the left pulmonary artery can be achieved by selective rotation of the 7 Fr balloon wedge catheter, use of a 0.035˝ angled Glidewire (Terumo Medical Corporation), or use of any angled catheter (KMP, JR4) that can be selectively directed to the main left pulmonary artery. 

Pulmonary artery imaging. A complete pulmonary angiogram is not required for sensor implantation. Injection of 4-8 cc diluted contrast via the 7 Fr balloon wedge catheter provides adequate visualization of the required pulmonary vessels. Imaging should be done in both an anterior-posterior and steep left anterior oblique or lateral projections. The left anterior oblique or lateral projections ensure that the desired sensor location is within a pulmonary artery branch that is coursing in a posterior direction. A pulmonary artery branch measuring between 7 and 15 mm in diameter should be selected for sensor implantation. Vessel size can be measured using cardiac catheterization laboratory software (quantitative coronary angiography) or visually estimated using the size of the adjacent balloon wedge catheter shaft or the inflated distal balloon as a reference (Figure 3). If an optimal vessel location is not identified during initial imaging, the balloon wedge catheter should be withdrawn and another proximal pulmonary vessel should be engaged and imaged. Use of fluoroscopic landmarks such as surgical clips or ventricular leads can be useful when determining the location for sensor implantation (Figure 4). 

Device preparation and implantation. Once the desired sensor location is determined, a 0.018˝ wire is advanced to a distal portion of the pulmonary vessel. Commonly used 0.018˝ wires include the Platinum Plus (Boston Scientific), Hi-Torque Steelcore 18 (Abbott Vascular), and CardioMEMS guidewire (St. Jude Medical). Care should be taken to place the guidewire securely in a distal vessel location, without introducing a bend to the distal portion of the guidewire (Figure 5). A distal wire bend may dislodge the sensor during guidewire removal. The over-the-wire port of the sensor delivery catheter is flushed with heparinized saline and the sensor is agitated in heparinized saline for 10-15 seconds to activate the hydrophilic coating on the delivery catheter. When advancing the sensor through the 12 Fr venous sheath, the sensor should be held in place to prevent movement on the sensor relative to the delivery catheter. Under fluoroscopic guidance, the sensor is advanced through the right heart to the desired implantation location. The sensor should be clearly visualized as it traverses the tricuspid valve to avoid entanglement with any tricuspid valve structures (papillary muscles, chordae tendinae, and tricuspid valve leaflets). Resistance to sensor advancement during advancement through the tricuspid valve is likely related to entanglement with one of these structures and requires recrossing the tricuspid valve using balloon floatation. Once the sensor is advanced to the desired location, the tether release system is withdrawn in a slow, continuous manner under fluoroscopic imaging. Following release of the sensor, cinefluoroscopy is done to document sensor location (Figure 6). The delivery catheter is removed under fluoroscopy and the 0.018˝ guidewire is left in place. During removal of the delivery catheter, the sensor should be clearly visualized to assess for sensor movement. 

Device calibration. The balloon wedge or a Swan-Ganz catheter is reinserted to measure pulmonary artery pressures and calibrate the device. Care should be taken to avoid advancement of the balloon wedge or Swan-Ganz catheter into the same pulmonary artery branch as occupied by the sensor as this may result in sensor movement. The hospital electronics unit wand is placed behind the patient and signal strength is assessed (Figure 7). Three to four pressure readings are taken consecutively to calibrate the sensor.

Hemostasis. Multiple options exist for hemostasis, including manual hemostasis, suture based, or device based. Use of a modified figure-8 suture placed below and adjacent to the venous sheath allows immediate sheath removal, excellent hemostasis, and minimal additional cost (Figure 8). The suture can be removed 2-3 hours post procedure, prior to the patient ambulating. 

Post procedure. Prior to hospital discharge, patients are educated by nurse practitioners and/or trained hospital staff on the steps required to obtain pressure measurements using their home electronics unit. Measurements are taken and sent to the secure website prior to hospital discharge (Figure 9). Outpatients with stable hemodynamics can usually be discharged to home 4 hours after completion of the procedure, following successful transmission of pressures and ambulation.

A chest radiograph in anterior-posterior and lateral projections is obtained to document sensor location. For patients not receiving long-term warfarin or novel oral anticoagulants, clopidogrel 75 mg daily and aspirin 81 mg daily are prescribed for 30 days; after 30 days, aspirin is continued. 

Case Presentations

Sensor implantation requires approximately 2 minutes of fluoroscopy time and <20 minutes of overall procedural time. Difficult sensor implantation is uncommon, but likely related to severe tricuspid regurgitation, inability to advance the sensor, non-optimal vessel location for sensor placement, movement of the sensor following deployment, and inability to advance the 0.018˝ guidewire to the desired distal vessel location. Each of the following cases outlines issues that may arise during the implantation procedure. 

Case #1. Straightforward implantation. A 65-year-old female with CHF with preserved ejection (HFpEF), prior aortic valve replacement, severe left ventricular hypertrophy, and three prior episodes of HF requiring hospitalization over the last year was referred for CardioMEMS implantation. Procedural imaging showed a left-sided pulmonary artery branch measuring approximately 10 mm in diameter (Figure 10). A 0.018˝ wire was placed into a distal branch and provided excellent support. The sensor was implanted without difficulty. 

Case #2. Severe tricuspid regurgitation. A 52-year-old male with morbid obesity (body mass index, 36 kg/m²; chest circumference, 64 inches) and HF with reduced ejection fraction (HFrEF), ventricular tachycardia, and three prior episodes of HF requiring hospitalization over the last year was referred for CardioMEMS implantation. A recent transthoracic echocardiogram showed an ejection fraction of <25%, severe tricuspid regurgitation, and moderate pulmonary hypertension. Given the presence of severe tricuspid regurgitation and morbid obesity, venous access was obtained using ultrasound guidance of the right internal jugular vein (Figure 11). A selective pulmonary angiogram showed a left-side pulmonary artery branch measuring approximately 10 mm in diameter. Sensor advancement across the tricuspid valve was straightforward and the sensor was implanted without difficulty. 

Case #3. Inability to advance sensor. A 72-year-old male with coronary artery disease, prior myocardial infarction, prior coronary artery bypass surgery, and multiple prior coronary stent procedures with HFrEF and three prior episodes of HF requiring hospitalization over the last year was referred for CardioMEMS implantation. A 7 Fr balloon wedge catheter was advanced from the right common femoral vein to the left pulmonary artery with the balloon inflated. Cardiac pressures and cardiac output were measured. A selective pulmonary angiogram was obtained. A 0.018˝ Platinum Plus wire was placed into the distal portion of an inferior and lateral branch of the left pulmonary artery. The sensor could not be advanced across the tricuspid valve despite multiple attempts, including rotation of the delivery catheter. The sensor and guidewire were removed as a unit. Care was taken to hold the sensor as it was withdrawn through the 12 Fr venous sheath to avoid movement of the sensor relative to the delivery catheter. The 7 Fr balloon wedge catheter was readvanced across the tricuspid valve using balloon floatation. A selective pulmonary angiogram was repeated and the 0.018˝ Platinum Plus wire was reinserted. The sensor was easily advanced across the tricuspid valve, suggesting that the initial delivery catheter location was entrapped within one of the tricuspid valve structures. The sensor was released without difficulty. 

Case #4. Non-optimal vessel location for sensor location. A 68-year-old male with coronary artery disease, prior myocardial infarction, prior coronary artery bypass surgery, and multiple prior coronary stent procedures with HFrEF and three prior episodes of HF requiring hospitalization over the last year was referred for CardioMEMS implantation. A 7 Fr balloon wedge catheter was advanced from the right common femoral vein to the left pulmonary artery and a selective pulmonary angiogram was obtained. Significant tapering of the distal pulmonary artery branch vessels was noted (Figure 12). The optimal sensor location was thought to be slightly distal, to allow apposition of the nitinol loops to the vessel wall and to avoid having the sensor “float” in the proximal portion of the branch. A 0.018˝ CardioMEMS guidewire was placed into the distal portion of the pulmonary artery branch and the sensor was released without difficulty. There was no movement of the sensor post deployment. 

Case #5. Movement of sensor following deployment. A 72-year-old female with atrial fibrillation, three prior unsuccessful attempts at atrial fibrillation ablation, HFpEF, and two prior episodes of HF requiring hospitalization over the last year was referred for CardioMEMS implantation. Selective pulmonary angiography showed two possible locations for sensor implantation. A 0.018˝ Platinum Plus guidewire was advanced to the distal vessel and the sensor was advanced to the desired implantation location. The sensor was released and the delivery catheter was withdrawn. When the distal portion of the delivery catheter was within the right atrium, the sensor was noted to move proximally (Figure 13). The 0.018˝ wire was left in place and the 7 Fr balloon wedge catheter was rapidly advanced to the proximal left pulmonary artery. The balloon was inflated and used to advance the sensor. Once the balloon moved past the proximal portion of the sensor, the balloon was deflated and this process was repeated several times until the sensor was advanced to the desired location. A chest radiograph the following day confirmed excellent position of the sensor. 

Case #6. Inability to advance the guidewire to the desired distal vessel location. In rare cases, it can be difficult to advance the 0.018˝ wire into the desired branch of the pulmonary artery. In this setting, use of a co-pilot device (Abbott Vascular) in combination with a torqueing device allows for imaging of the desired location during wire manipulation and advancement (Figure 14). 

Patient Selection

Patients with NYHA class III symptoms and one prior hospitalization for HF should be considered for CardioMEMS implantation. In clinical practice, patients who appear to achieve the most benefit from pressure-guided HF management are those on optimal medical therapy with difficult to control volume status and frequent hospitalizations for decompensated HF. Patients who are compliant with medical therapy and with communications with their health-care providers, and those with a solid knowledge base regarding medical therapy and concepts regarding diuretic medication titration also appear to achieve benefit from the CardioMEMS device. 

Discussion

The CardioMEMS HF system has been available for clinical use in the United States since FDA approval in October 2014. Although the implantation procedure is straightforward and commonly performed in <20 minutes, several procedural issues may arise. Difficult sensor implantation may occur in the setting of severe tricuspid regurgitation, the inability to advance the sensor across the tricuspid valve, rapidly tapering pulmonary vessels with a non-optimal vessel location for sensor implantation, and movement of the sensor following deployment. The most challenging issue that arises during sensor implantation is proximal movement of the sensor post deployment. This can be addressed using an inflated balloon wedge catheter with advancement of the balloon to advance the sensor to the desired location.

Implantation of the CardioMEMS sensor is a minor component of successful implementation of this technology. To effectively use the CardioMEMS system to reduce future hospitalizations for HF, an efficient HF management program must be in place, protocols for pressure-guided HF management must be established and rigorously followed, compliance with frequent pressure transmission must be encouraged, and appropriate patients likely to benefit from remote monitoring must be identified. 

The ongoing post-FDA approval study is evaluating use of the CardioMEMS HF system in 1200 patients with NYHA class III HF symptoms and one prior hospitalization for HF within the last year.⁶ As physicians gain additional experience with sensor implantation and pressure-guided HF management, device-based therapy for HF will continue to improve. 

References

1.    Bourge RC, Abraham WT, Adamson PB, et al; the COMPASS-HF group. Randomized controlled trial of an implantable continuous hemodynamic monitor in patients with advanced heart failure: the COMPASS-HF study. J Am Coll Cardiol. 2008;51:1073-1079.

2.    Smith SA, Abraham WT. Device therapy in advanced heart failure: what to put in and what to turn off: remote telemonitoring and implantable hemodynamic devices for advanced heart failure monitoring in the ambulatory setting and the evolving role of cardiac resynchronization therapy. Congest Heart Fail. 2011;17:220-226.

3.    Mabote T, Wong K, Cleland JG. The utility of novel non-invasive technologies for remote hemodynamic monitoring in chronic heart failure. Expert Rev Cardiovasc Ther. 2014;12:923-928.

4.    Ritzema J, Troughton R, Melton I, et al; the Hemodynamically Guided Home Self-Therapy in Severe Heart Failure Patients Study Group. Physician-directed patient self-management of left atrial pressure in advanced chronic heart failure. Circulation. 2010;121:1086-1095.

5.    Abraham WT, Adamson PB, Bourge RC, et al; CHAMPION Trial Study Group. Wireless pulmonary artery haemodynamic monitoring in chronic heart failure: a randomised controlled trial. Lancet. 2011;377:658-666.

6.     CardioMEMS HF system post approval study. Clinical trials.gov/Identifier NCT02279888.

From the ¹Division of Cardiovascular Medicine, University of Southern California, Los Angeles, California; and ²Division of Heart Failure, Department of Cardiology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Shavelle has served as a paid consultant  for St. Jude Medical (peer to peer physician course) and reports research support from St. Jude Medical, Abbott Vascular, and the NIH. Dr Jermyn reports research support from St. Jude Medical. 

Manuscript submitted May 6, 2016, provisional acceptance given May 17, 2016, final version accepted May 23, 2016.

Address for correspondence: David M. Shavelle, MD, FACC, FSCAI, Division of Cardiovascular Medicine, Keck School of Medicine, 1510 San Pablo Street, Suite 322, Los Angeles, CA 90033. Email: shavelle@usc.edu