Search

Isolated mammalian hearts have been used to study cardiac physiology, pharmacology, and biomedical devices in order to separate myocardial characteristics from the milieu of the intact animal and to allow for increased control over experimental conditions. Considering these benefits and that MRI is the “gold” standard for measuring myocardial function, it was considered desirable to have a system which would allow simultaneous MR imaging of an isolated beating heart. Here we describe a unique portable system, which enables physiologic perfusion of an isolated heart during simultaneous MR imaging. A two unit system was designed to physiologically support a large mammalian isolated heart during MR imaging were a modified Krebs-Henseleit perfusate was used as a blood substitute. The first unit, which resides in an adjacent support room next to the scanner, contains all electronically powered equipment and components (with ferromagnetic materials) which cannot operate safely near the magnet, including (1) a thermal module and custom tube in tube heat exchanger warming the perfusate to 38°C; (2) a carbogen tank (95% O2 5% CO2) and hollow fiber oxygenator; and (3) two centrifugal blood pumps which circulates and pressurizes the left and right atrial filling chambers. The second unit, which resides next to the magnet and is free of ferromagnetic materials, receives warmed, oxygenated perfusate from the first unit via PVC tubing. The isolated hearts were connected to the second unit via four cannulae sutured to the great vessels. A support system placed inside the scanner on the patient bed secured the hearts and cannulae in the correct anatomical position. To date, this system was tested in a 1.5 T Siemens scanner using swine hearts (n=2). The hearts were arrested with St. Thomas cardioplegia and removed via a medial sternotomy. After cannulation of the great vessels, reperfusion, and defibrillation, four-chamber and tagged short-axis cine loops were acquired using standard ECG gating. Tagged short-axis images obtained at the base, mid-ventricle, and apex were used to measure the following functional parameters for one heart: LV end-diastolic volume=38.84 ml, LV end-systolic volume=23.23 ml, LV stroke volume=15.6 ml, LV ejection fraction=40.18%, and peak LV circumferential strain=16%. The feasibility of MR imaging an isolated, four-chamber working large mammalian heart was demonstrated using a custom designed and built portable MRI compatible perfusion system. This system will be useful in studying in vitro cardiac function (including human hearts) and developing MRI safe biomedical devices and MRI guided therapies in a controlled setting.

Using Visible Heart® methodologies we imaged coronary artery bypass grafts (CABGs) and coronary stents in isolated beating human hearts and perfusion fixed human hearts. Due to the varying cardiac health of the donor hearts it has been possible to see progressive levels of stent endothelialization and vascular calcification. The isolated heart model uses a clear Krebs–Henseleit buffer in place of blood, allowing for the unique opportunity to image the coronary vessels. In the isolated human heart a fiberscope was inserted into either the native coronary artery or the CABG with the heart in sinus rhythm. In order to verify cardiac function during the imaging process the following measurements were read at a sampling rate of 5 kHz: ECG, aortic flow, and ventricular pressures. Perfusion fixed hearts were fixed in an end diastolic state achieved by applying pressures comparable to physiological conditions. This process causes the coronary arteries to fix in a dilated state. CABGs of human hearts were then imaged using fluoroscopy (angiograms) and fiberscopic techniques. The stented native coronary arteries of human hearts were imaged via fluoroscopy and by dissection. Through a variety of imaging techniques and using Visible Heart® methodologies we have obtained a unique visualization of a CABG and a coronary artery stent in a beating human heart during sinus rhythm. Investigative studies in perfusion fixed human hearts have provided a more complete anatomical imaging study of stent endothelialization in the native coronary arteries and vascular calcification in bypass grafts.

This project aims to investigate the performance of edge-to-edge mitral valve repair (MVR) within reanimated swine hearts. Direct imaging and hemodynamic data of the mitral valve during normal cardiac function (Normal), after an induced prolapse (Prolapse), and post surgical repair (E2E) was obtained. Isolated swine hearts (n=6) were reanimated using a clear Krebs–Henseleit buffer. Mitral prolapse, and regurgitation, in the P2 region was induced by cutting chordae tendinae of the posterior leaflet. An edge-to-edge MVR procedure was performed, suturing the prolapsed P2 region to the A2 region of the anterior leaflet. The mitral valve was imaged using endoscopic cameras in the left atrium and ventricle allowing verification of stitch placement and leaflet coaptation. Analysis of the endoscopic images provided measures of annulus area, orifice area, and regurgitant area. Echocardiography, the standard clinical imaging modality, was used to determine the hemodynamic performance of the valve. Additionally, ECG and left chamber pressures were recorded at a sample rate of 5 kHz. Prolapse of the P2 region was consistently created, and edge-to-edge repair of the mitral leaflet showed full leaflet coaptation. The annulus area of the valve was tracked throughout the procedure and did not show significant variation. The orifice area, defined as the area of the annulus that does not contain leaflets, normalized to the corresponding annulus area for Normal, Prolapse and E2E were: 41±13%, 44±14% and 21±13%, p=0.02. The regurgitant area, normalized to the corresponding annulus area, increased from 2±2% for Normal to 8±3% for the Prolapse and then decreased to 1±1% for the E2E group. The regurgitant fraction, normalized against the maximum observed, for Normal, Prolapse and E2E was 10±6%, 57±26% and 13±13%, p