Although the efficacy of treatment for all cardiovascular diseases has improved greatly in developed countries, it is expected that the incidence and prevalence of heart failure will increase due to the changing demographics of the population.1 The development of health management strategies that extend and improve the quality of life for heart failure patients at sustainable costs is of great clinical and societal importance. One strategy that may achieve both goals is the development of permanently implanted intracardiac hemodynamic monitors that permit real-time collection of intracardiac pressure data under ambulatory conditions.2,3 Two currently approved devices consist of a miniature pressure transducer connected by a cable to a signal conditioning unit, and a third device under consideration is a passive device energized by ultrasound. These devices have been implanted transvenously into the right heart or pulmonary artery (Chronicle, Medtronic; CardioMems, Inc.) or delivered by catheter to the left atrium via transseptal puncture (HeartPOD, St. Jude Medical). Recent clinical trials include the Chronicle Offers Management to Patients with Advanced Signs and Symptoms of Heart Failure (COMPASS-HF), Hemodynamically Guided Home Self-Therapy in Severe Heart Failure Patients (HOMEOSTASIS), and CardioMems Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients (CHAMPION). These studies suggest that monitoring devices may improve quality of life and reduce the frequency of hospitalizations for acute decompensation events by providing intracardiac pressure loading data to guide titration of medications.4–6 Moreover, as these devices and the criteria for appropriate application continue to evolve, survival times may be extended. In addition, the devices may be useful for the detection of other pathologies or as sensors for use with other implanted devices such as ventricular assist pumps.
This report focuses on pilot implantation studies of a high-fidelity wireless and battery-less blood pressure sensor [Integrated Sensing Systems, Inc. (ISSYS), Ypsilanti, MI], after surgical placement into the left ventricle of conditioned dogs. The primary goals of the study were to develop an implantation technique and assess safety and efficacy over 8 weeks. A secondary goal was to assess performance over the same period.
Methods and Materials
Blood Pressure Measurement System
The implantable device (Figure 1) is an “intelligent” cylindrical glass-encapsulated pressure sensor. It consists of a biocompatible glass tube (3.7 mm D × 17 mm L) that contains an ASIC (Application Specific Integrated Circuit). The flat 3.7 mm diameter end of the cylinder is the pressure sensing element made by a proprietary MEMS (Micro-Electro-Mechanical Systems) technology. This, in addition to the other electronic components that provide signal conditioning, power management, and radiofrequency transmission, is hermetically sealed within the glass capsule at a near-total vacuum to establish a stable “no drift” reference state and to prevent contamination from body fluids. After assembly and sealing, the calibration coefficients for offset and gain are determined for each sensor by the manufacturer and stored in a computer file for use by proprietary software that interprets the data sent by the extracorporeal components of the data capture system. Aside from a computer, the system consists of a proprietary transceiver unit and a handheld antenna (Figure 2, A and B). The transceiver unit powers the implanted sensor by electromagnetic induction and detects a radio frequency signal emitted from the implanted sensor that carries the encoded pressure data. The transceiver processes the signal and then passes it via a USB interface to a computer for real-time display and recording. The transceiver also passes ambient atmospheric pressure data from an internal reference barometer to the computer, so as to enable compensation for altitude and meteorological effects. It also provides digital to analog (D/A) voltage output for connection to other instruments.
All procedures were reviewed and approved by the William Beaumont Hospital Research Institute Animal Care Committee for compliance with regulatory requirements and ethical standards. Conditioned adult mongrel dogs (n = 4, weight 20–25 kg) were prepared as follows. After a 1 week quarantine period, the animals were anesthetized with thiopental (25–20 mg/kg), endotracheally intubated, and placed on ventilator that administered inhalational isoflurane (1%–3%). Antibiotic prophylaxis (500 mg cephalosporin IV) and preemptive analgesia (carprofen 2 mg/kg) were administered, and the left thorax was clipped and prepped for sterile surgery. Before incision, the site was infiltrated with 1% bupivacaine solution. The heart was approached through a fourth intercostal space thoracotomy, and the pericardium was opened broadly to expose the heart. The apex of the left ventricle was identified, and three to four 4-0 polypropylene sutures were placed in an approximately 1-cm circle. The apex was punctured within the perimeter of the sutures with a 14-gauge hypodermic needle, taking note of the depth of penetration that resulted in pulsatile blood flow, and the ventriculostomy was enlarged slightly with an instrument.
A sensor of appropriate length (myocardial thickness plus ∼3-5 mm) was selected to ensure that the sensing surface was not embedded within the myocardium, where it would transmit intramuscular pressure instead of the blood pressure within the ventricle. The sensor was inserted through the ventriculostomy and secured to the epicardium using the plastic anchor formed from polyether ether ketone (PEEK). Then a 5-Fr micromanometer-tipped catheter (MPC-500, SP-350, Millar Instruments, Houston, TX) was placed via a small ventriculostomy for simultaneous collection of comparative left ventricular (LV) pressure measurements (Figure 3). A sterilized miniature antenna was placed over the sensor to energize it and confirm transmission of the LV pressure waveform. After data collection, the Millar catheter was removed, the pericardium was loosely reapproximated, and the wound was closed in layers following standard operative techniques. Analgesia was provided by bolus parenteral administration of acepromazine maleate (0.05 mg/kg) and buprenophrine (0.03–0.05 mg/kg) until a transdermal fentanyl patch (150 mg) provided postoperative pain relief from 12 to 72 hours. Carprofen was administered thereafter as needed.
At 4 weeks, the animals were anesthetized as described above and the neck was prepared in a sterile manner for LV catheterization, and the left common carotid artery was circumferentially exposed in the neck for cannulation. After catheterization, the artery was repaired and the wound closed by standard techniques. At 8 weeks, another catheterization study was performed in a terminal procedure and the sensor was retrieved en bloc, and the kidneys were harvested for examination. The explanted sensors were returned to the manufacturer for evaluation of their mechanical and electrical properties.
Data Collection and Processing
After implantation, an external antenna was used to confirm sensor power-up and signal transmission across the chest wall. Pressure data from the sensor and the micromanometer catheter were collected simultaneously at the implantation procedure and during the catheterizations at 4 and 8 weeks. The micromanometer catheter was connected to a Gould 4600 series signal conditioner, and its analog output was recorded on a computerized D/A data acquisition system at 200 samples per second (WinDaq, DATAQ Instruments, Akron, OH). Data from the MEMS sensor were also collected with the WinDaq system via the D/A output port of the ISSYS transceiver unit. Postexperimental processing of the data was performed to correct zero offsets that were attributable to external conditions or components of the equipment, such as ambient barometric pressure, body temperature, or interconnection of the recording equipment. Apparent zero offsets or other changes that were attributable to the effects of implantation were not adjusted, nor was the 40–75 milliseconds delay that was due to processing by the hardware and software of the ISSYS transceiver system. After baseline correction, the signals were filtered with an 8-10 point moving average filter to attenuate high-frequency artifacts greater than approximately 20 Hz and then differentiated with respect to time to obtain LV dP/dt.
All animals were ambulatory, demonstrated good appetite on the first postoperative day, and were capable of vigorous exercise within 1–2 weeks. There were no behavioral abnormalities or clinical signs of heart failure, neurologic damage, or distal thromboembolism at any time during the 8 weeks of implantation. At autopsy (Figure 4, A and B) the epicardial anchors of the implanted sensors were encapsulated in connective tissue. Figure 5A is a representative composite photomicrograph of the apical implantation site of dog 1 stained with hematoxylin and eosin. Figure 5B shows a high-power magnification (400×) of the thin connective tissue film [mean = 0.66 ± 0.44 mm (SD)] that covered the lumenal pressure-transducing surfaces of the implanted sensors. In dog 4, it was discovered upon explantation that the sensor was partially embedded at the base of a papillary muscle. Intracardiac thrombus was not observed in any animal, nor was there evidence of distal thromboembolism in the kidney as evidenced by gross infarcts. Examination of sensors that were recovered without iatrogenic damage revealed that the seals were intact and that their gain and frequency response characteristics were not changed by the implantation, as determined by testing upon the manufacturer’s calibration apparatus.
Left ventricular pressure waveforms and their first derivatives are displayed for all four animals (Figure 6) at the time of implantation and at 4 and 8 weeks. In all animals, the morphology of the waveform transmitted by the implanted sensor was similar to that of the Millar catheter at every time point (note that the data from the experimental sensor always follows the catheter-based data due to signal processing delay). Small hydrostatic pressure discrepancies occurred that were attributed to the fact that we did not attempt to insert the catheter exactly adjacent to the sensor so as to avoid damage to the implanted sensors’ diaphragms by inadvertent contact. In two dogs (Figure 6; dogs 2 and 4), there were minor differences in pressure at 4 and 8 weeks. These differences did not exceed the recording range of the system, and the calculated LV dP/dT demonstrated good correlation between the sensor and the catheter across the time points. Data from both sensors showed slight baseline offsets over time, as well as an additional variance that was linked to the respiratory cycle.
This study demonstrated that permanent implantation of miniature wireless, battery-less high-fidelity pressure sensors into the healthy and dynamic left ventricle in mongrel dogs was technically successful, safe, and functional for 8 weeks after implantation. Importantly, the sensors provided robust intracardiac pressure data to asses LV function over time. Eight weeks was chosen so that the wound healing process was largely complete and sufficient time passed for potential evidence of thromboembolism to be observed. All sensors functioned as expected throughout the implant period and no cardiac, neurologic, or distal thromboembolic complications were observed. We believe that these sensors will function for the intended design life of 10 years and likely longer, as they are hermetically sealed in a glass capsule.
The particular advantages offered by transmural placement of this compact miniature sensor are that it 1) presents minimal blood contact surface area thus minimizing the risk of thrombus formation, 2) does not tether the heart or suffer from excessive motion artifact, and 3) is less likely to interfere with critical structures or 4) escape to form an embolus. The device can be placed during open chest revascularization procedures or other corrective surgeries for arrhythmias or structural defects (e.g., valvular repairs). Future studies will explore the possibilities of delivery by minimally invasive surgical or percutaneous techniques as well as implantation in other sites such as the right ventricle or left atrium.
In this pilot study, we chose to implant the device transmurally into the left ventricular apex to directly measure LV diastolic, systolic, and mean pressures. In addition, it is possible to derive other indices of both systolic and diastolic function, including rates of pressure development and relaxation, the tension-time period, and the temporal ratio of systolic to diastolic periods (Tei Index).7 Moreover, if LV pressure data can be simultaneously combined with volumetric data as might be obtained by four-dimensional imaging techniques, then it would be possible to assess the ventricular function of a conscious patient by the “gold standard” pressure-volume relationship, whether in the office, emergency room, or radiology suite.
In examination of the data in this study, insignificant baseline offsets occurred in two devices. We speculate that this related in part to the deposition of collagen and smooth muscle cells on the sensors’ surfaces as observed in the postmortem examinations (Figure 4, A and B). However, it is important to note that a baseline shift does not effect a change of the first derivative of pressure or alter the measurements of temporal relationships of the cardiac cycle that may be used to calculate the Tei Index. In Figure 6, the LV dP/dt data derived from the permanently implanted sensors are in good agreement with the simultaneous measurements derived from the temporarily placed Millar catheters.
In one animal, the sensor was noted to be embedded in a papillary muscle at necropsy, but it is notable that there were no signs of behavioral distress, dyspnea, or exercise intolerance. A pressure trace from this animal showed a significant baseline shift that varied with the respiratory cycle. With respect to changes in the sensors’ gain, the tissue overgrowth at the sensor’s surface did not appear to affect the precision of the pressure waveforms obtained. One option is to recalibrate the sensor by a noninvasive technique developed by McClean et al.,8 and also collect data at a defined time point in the cardiorespiratory cycle. Furthermore, it may be possible to attenuate the encapsulation by coating the sensor with substances that inhibit growth of connective tissue. In future studies, preoperative echocardiographic imaging may aid in selection of an optimal implantation site.
In summary, our study demonstrated the feasibility, safety, and efficacy of a wireless, battery-less high-fidelity pressure transducer implanted permanently into the left ventricle of ambulatory dogs. The transducer performed well over the 8-week study period without evidence of device-related complications. This technology has broad implications for individualized treatment of heart failure at a sustainable cost if timely data processing can be achieved. In the future, modifications of this technology may have application in other organ systems (e.g., neural, ocular, and urinary systems) where pressure measurements may be useful to guide treatment.
Supported by a grant from the Michigan Economic Development Council (MTTC 125-05). The authors thank Diane Studzinski, Rose Callahan, and the Research Services of William Beaumont Hospital Research Institute.
1. Kuznetsova T, Herbots L, López B, et al. Prevalence of left ventricular diastolic dysfunction in a general population. Circ Heart Fail.;2:105–112 2009
2. Merchant FM, Dec GW, Singh JP. Implantable sensors for heart failure. Circ Arrhythm Electrophysiol.;3:657–667 2010
3. Conraads VM, Vrints CJ. Managing heart failure patients: when good-old-fashioned clinical care is not enough. Heart.;95:1036–1037 2009
4. Bourge RC, Abraham WT, Adamson PB, et al.COMPASS-HF Study Group. Randomized controlled trial of an implantable continuous hemodynamic monitor in patients with advanced heart failure: the COMPASS-HF study. J Am Coll Cardiol.;51:1073–1079 2008
5. Ritzema J, Troughton R, Melton I, et al.Hemodynamically Guided Home Self-Therapy in Severe Heart Failure Patients (HOMEOSTASIS) Study Group. Physician-directed patient self-management of left atrial pressure in advanced chronic heart failure. Circulation.;121:1086–1095 2010
6. Adamson PB, Abraham WT, Aaron M, et al. CHAMPION trial rationale and design: The long-term safety and clinical efficacy of a wireless pulmonary artery pressure monitoring system. J Card Fail. 2011;17:3–10
7. Tei C, Nishimura RA, Seward JB, Tajik AJ. Noninvasive Doppler-derived myocardial performance index: correlation with simultaneous measurements of cardiac catheterization measurements. J Am Soc Echocardiogr.;10:169–178 1997
8. McClean D, Aragon J, Jamali A, et al. Noninvasive calibration of cardiac pressure transducers in patients with heart failure: an aid to implantable hemodynamic monitoring and therapeutic guidance. J Card Fail.;12:568–576 2006