Ann Thorac Surg 2008;85:1061-1065. doi:10.1016/j.athoracsur.2007.09.037
© 2008 The Society of Thoracic Surgeons
New Technology
Beating Heart Cardioscopy: A Platform for Real-Time, Intracardiac Imaging
Tomislav Mihaljevic, MDa,*,
Yoshio Ootaki, MD, PhDa,b,
Jason O. Robertson, BSa,c,
Amir K. Durrani, BSa,c,
Keiji Kamohara, MDb,b,
Masatoshi Akiyama, MD, PhDb,
Faruk Cingoz, MDb,
Chiyo Ootaki, MDb,
Raymond Dessoffy, AAb,
Michael Kopcak, BAb,
Jenny Liu, BAb,
Kiyotaka Fukamachi, MD, PhDb
a Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, Ohio
b Department of Biomedical Engineering, Cleveland Clinic, Cleveland, Ohio
c Cleveland Clinic Lerner College of Medicine, Cleveland Clinic, Cleveland, Ohio
Accepted for publication September 18, 2007.
* Address correspondence Dr Mihaljevic, Cleveland Clinic, Department of Thoracic and Cardiovascular Surgery, 9500 Euclid Ave, Desk F24, Cleveland, OH 44195 (Email: mihaljt{at}ccf.org).
 |
Abstract
|
|---|
Purpose: The purpose of this study was to develop a method for real-time, fiberoptic, intracardiac imaging to serve as a platform for closed-chest, intracardiac surgery on the beating heart.
Description: Fiberoptic cardioscopy of the left and right heart was conducted in a porcine model. A cardiopulmonary bypass circuit maintained systemic organ perfusion and a separate circuit replaced intracardiac blood with oxygenated, modified Krebs-Henseleit perfusate.
Evaluation: Video images of structures in the left and right sides of an in vivo beating heart were obtained, including the inner surface of the left and right atria and ventricles, the mitral and aortic valves, the Thebesian veins, and the coronary sinus. Effective isolation of the heart from systemic and intracardiac blood flow and control of perfusion rates were important factors for successful image acquisition.
Conclusions: Fiberoptic cardioscopy is a novel approach that allows for visualization of the structures within a nonarrested heart on bypass. It lays the groundwork for a platform that could lead to more successful percutaneous valvular and intracardiac procedures in a stable hemodynamic environment.
|
Introduction
|
|---|
The successful development of a platform for beating heart cardioscopic surgery has been limited by difficulties of visualizing the inside of the heart through the blood. Other primary obstacles remain technological (ie, thin-walled cannulas and miniaturized endoscopic equipment with large flushing and instrument channels are important for a percutaneous platform [1]). Efforts to develop such a cardioscopic system have been underway since 1913 when the first cardioscopes were invented [2]. A majority, but not all, of these attempts were undertaken in arrested or explanted hearts and facilitated visualization of major right-sided, intracardiac structures through the displacement of blood [3–5].
|
Technology
|
|---|
This technique lays the groundwork for a platform for in vivo beating heart cardioscopy that could eventually permit safer, less invasive, closed-chest procedures in a hemodynamically stable environment (Figs 1A,
1B). A femoral-femoral bypass circuit would perfuse the systemic circulation, and a second cardiac circuit would perfuse the heart with clear solution. Such a platform would provide an isolated system in which intracardiac pressures, flow rates, and volumes are controllable.
View larger version (40K):
[in this window]
[in a new window]
|
Fig 1. Schema of the minimally invasive surgical approach. (A) The cardiopulmonary bypass (CPB) circuit maintains the systemic circulation, and the cardiac circuit flushes the heart with clear, oxygenated perfusate to enable visualization of the intracardiac structures during beating heart surgery. Multi-port catheters may be used for introduction of instruments into the chambers including endoscopes. (B) Percutaneous cannulation is depicted, including a femoral-femoral bypass and the cardiac circuit.
|
|
|
Technique
|
|---|
Animal Preparation
This study protocol and subsequent amendments were approved by the Institutional Animal Care and Use Committee at the Cleveland Clinic Foundation. All animals were treated in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No. 86-23, Revised 1985).
Five domestic pigs (weighing 48.4 to 60.6 kg) were anesthetized by intramuscular injection of ketamine (22 mg/kg) and 1 mg/kg to 2 mg/kg of xylazine followed by Pentothal (20 mg/kg, intravenously; Hospira Inc, Lake Forest, Ill); this was maintained by inhaled isoflurane with the animals intubated and mechanically ventilated.
Surgical Approach and Imaging
A median sternotomy was performed and the pericardium was opened. The great vessels were exposed and isolated. As part of our final model, cardiopulmonary bypass (CPB) was added and instituted after full heparinization (300 IU/kg) (Fig 2). A 20-French outflow cannula was inserted into the ascending aorta through pursestring sutures. Inflow cannulae (28-French) were inserted into the vena cavae. Both vena cavae were occluded to prevent blood return and isolate the right atrium. The proximal portion of the coronary sinus was snared to prevent venous backflow into the right atrium. The left azygous vein was also snared because it drains directly into the coronary sinus in pigs.
View larger version (41K):
[in this window]
[in a new window]
|
Fig 2. Schema of the actual surgical approach. Cardiopulmonary bypass (CPB) was conducted between the aorta and vena cavae. The endoscope (equipped with a sheath to enable infusion of perfusate) was inserted from the right atrial appendage, left atrial appendage, or left ventricular apex.
|
|
A 24-French fluid perfusion cannula was inserted into the right atrium through the appendage. A normothermic, oxygenated modified Krebs-Henseleit buffer [6] was infused through the cannula at a rate of 2.0 to 3.0 L/minutes using a separate roller-pump to provide a transparent field in the right heart chambers. This rate was adjusted to provide enough intra-chamber pressure to exceed venous pressure (restricting inflow from the Thebesian circulation) but was limited to avoid overdistention of the heart [1]. In the left heart, the pulmonary veins were a source of blood. The outflow from the right ventricle ran through cannula in the pulmonary artery and provided adequate drainage and an immediate assessment of effective isolation of the heart without the use of suction.
Visualization of structures on the right side of the heart was accomplished through insertion of an endoscope (Olympus GIF-140, 9.8 mm, 103 cm [Olympus, Center Valley, PA]), powered by a light source (Olympus CLV-U40 [Olympus]) and video processor (Olympus CV-140 [Olympus]), into the right atrial appendage using the tobacco-pouch suture technique (Fig 2, position 1). Left atrial structures were visualized by endoscopic insertion into the left atrium (Fig 2, position 2), and left ventricular structures were imaged through the left ventricular apex (Fig 2, position 3). All video images were recorded digitally at the time of the experiment (Sony DCR-TRV480 [Sony Corp, Tokyo, Japan]).
The first two pigs were used to determine the most efficient manner of right heart flushing and endoscopic navigation. In later pigs, the fiberoptic endoscope was placed inside a pliable, translucent, plastic outer sheath made from polyurethane, with an outside diameter slightly wider than the outer diameter of the endoscope (Fig 3). There was a communicating space between the outer diameter of the endoscope and the sheath that allowed for continuous infusion of additional modified Krebs-Henseleit buffer using a Bio-Pump (Medtronic, Inc, Minneapolis, MN) to clear the area in front of the lens.
View larger version (54K):
[in this window]
[in a new window]
|
Fig 3. Endoscope sheath. This provides a clear-viewing chamber and a port for infusion of perfusate to enhance visualization. The endoscope is advanced through the left end of the sheath and a Bio-Pump line (Medtronic, Inc, Minneapolis, MN) is attached to the spigot at the bottom.
|
|
Procedures lasted an average of 90 minutes and were terminated by administration of potassium chloride (240 mEq/kg) after image acquisition was complete.
|
Clinical Experience
|
|---|
Image Capture
All images are still captures of video acquired from the beating heart. Near perfect isolation of the heart from venous return and sufficient flushing (2.0 to 3.0 L/min) were essential to acquire clear images, particularly in the right chambers. The trabeculae of the right ventricle with blood emanating from the Thebesian vein outlets are seen in Figure 4A. The mitral valve and papillary muscles are displayed with clear visualization of both leaflets and the subvalvular apparatus in Figure 4B. Another papillary muscle is also depicted in Figure 4E. The ventricular surface of the aortic valve is represented in Figures 4C and 4D, whereas Figure 4C was acquired during diastole and Figure 4D was acquired during early systole.
View larger version (74K):
[in this window]
[in a new window]
|
Fig 4. Images acquired from the beating heart: (A) trabeculae of the right ventricle with arrows indicating the blood from the Thebesian circulation; (B) the mitral valve and papillary muscle; (C) the aortic valve during diastole; (D) the aortic valve during early systole; and (E) the right ventricular papillary muscle.
|
|
Systemic Effects of Perfusion
During initial imaging of the right heart, no attempts were made to remove any output from the circulation; these pigs demonstrated edema. This was subsequently avoided by collecting outflow at the pulmonary artery and placing a venting cannula in the left ventricular apex, such that infusions were matched by drainage; therefore, no edema was seen.
|
Comment
|
|---|
This is the first study to report successful visualization of the native intracardiac anatomy of the beating heart using a cardiopulmonary bypass circuit, transparent perfusate, and fiberoptic imaging equipment without external pacing [7]. Cardioscopy has been approached variably in the past. Images of intracardiac structures have been obtained through various methods. Method 1: displacement of blood with pressurized transparent fluid boluses [3, 4]; method 2: direct contact between the cardioscope and cardiac tissue [2]; method 3: use of a transparent balloon chamber to displace blood between the lens and the object [4, 5]; and method 4: the complete replacement of intracardiac blood with a transparent, nourishing perfusate [1, 8]. The application of these methods in experimental settings has resulted in hemodynamic instability and limited visualization, therefore limiting their clinical usefulness. In addition, the design of these approaches precludes percutaneous surgical applications because of their bulk and the fact that visualizing a structure implies that the structure is blocked by the instrument.
In contrast, method 4 (as previously described) provides a clear working environment in which the endoscope does not obstruct the working area, and it has been demonstrated to allow long-term inspection of more than 1 hour [8]. We successfully elected to use a modified Krebs-Henseleit perfusate designed with additives intended to support cardiac performance by chelating calcium and toxic metal ions, increasing glucose utilization, providing additional energy substrates, and increasing perfusate osmolarity to reduce cardiac edema [6]. Nevertheless, all of these methods, including method 4, are complicated by blood clouding the field of view when isolation of the heart is incomplete. Blood return from the Thebesian veins presented particular difficulty, as control of blood flow though these veins was limited. Novel endoscopic technology that allows visualization of structures through opaque blood is promising, but such technology is currently limited for percutaneous surgical applications by depth of field, angle of view, and lack of color imagery [9].
This study was able to capture images of intracardiac structures in an in vivo beating heart animal model. This advancement has potential for extrapolation to an entirely percutaneous method in a hemodynamically stable environment, allowing for percutaneous interventions such as valve repairs and replacements and septal defect repairs under direct vision. Presently, percutaneous stented valves [10] are placed in the beating heart using fluoroscopic and echocardiographic guidance in which the quality is often insufficient. Such valves may cause fatal hemodynamic instability if inappropriately placed. When used at the aortic position, inaccurate deployment has the potential to occlude the coronary arteries, and the potential exists to dislodge thrombi from calcified plaques. Cardioscopic visualization could improve the accuracy of deployment, and concomitant cardiopulmonary bypass would reduce the risk from the aforementioned complications. Such synergy between technologies, once fully developed, could provide an attractive alternative to conventional open heart surgery.
The primary goal of this study was proof of concept of our methodology; therefore, the approach was not yet percutaneous, no attempt was made to fully implement our proposed double bypass circuit (Figs 1A, 1B), and we did not collect data on the survival of the pigs or attempt to wean them from cardiopulmonary bypass. No study to date characterizes the basic challenges inherent to cardioscopy of the in vivo beating heart. Our study demonstrates both the technical possibility of beating heart cardioscopy, as well as the most pressing obstacles (ie, overcoming blood return from the Thebesian veins and the lungs). In contrast to accomplishments made in the arrested heart [1, 3], the use of beating heart cardioscopy presents the challenge of flushing a system that is more obstinate to emptying itself of blood and that provides a range of intracardiac pressures that sometimes dip below venous pressures. This latter fact permits the Thebesian veins to empty blood back into the right heart if flow and pressure are not carefully controlled [1]. In an open chest model, the pulmonary veins can be clamped to halt unwanted flow in the left heart, and if the problem persists (upon continued refinement) in a percutaneous system, the pulmonary veins can be occluded with a system of catheters, or the lungs may be flushed of blood.
Due to the absence of the double bypass circuit, hemodynamic stability was not a concern in this study. Finally, we acknowledge that our fiberoptic instrumentation is larger than would be ultimately desirable, as it was selected based on availability. Although improvements are necessary, technology already exists to advance our experiments in each of these areas as we proceed with the development of this platform into a reliable procedure for isolation and visualization of the intracardiac environment.
|
Disclosures and Freedom of Investigation
|
|---|
The authors independently developed this technology and had full control of the design of the study, methods used, outcome measurements, and production of the written report. This work was supported by a grant from the United States Army Telemedicine and Advanced Technologies Research Center.
|
Footnotes
|
|---|
Disclaimer The Society of Thoracic Surgeons, the Southern Thoracic Surgical Association, and The Annals of Thoracic Surgery neither endorse nor discourage use of the new technology described in this article.
|
References
|
|---|
- Legget ME, Shaw DP. Fiberoptic cardioscopy under cardiopulmonary bypass: potential for cardioscopic surgery? Ann Thorac Surg 1994;58:222-225.[Abstract]
- Cutler EC, Levine SA, Beck CS. The surgical treatment of mitral stenosis: experimental and clinical studies AMA Arch Surg 1924;9:689-821.[Abstract/Free Full Text]
- Dee P, Crosby I. Fibreoptic studies of the aortic valve in dogs Br Heart J 1977;39:459-461.[Abstract/Free Full Text]
- Sakakibara S, Iikawa T, Hattori J, Inomata K. Direct visual operation for aortic stenosis: cardioscopic studies J Int Coll Surg 1958;29:548-562.[Medline]
- Uchida Y, Oshima T, Fujimori Y, Hirose J, Mukai H, Kawashima M. Percutaneous fiberoptic angioscopy of the cardiac valves Am Heart J 1991;121:1791-1798.[Medline]
- Chinchoy E, Soule CL, Houlton AJ, et al. Isolated four-chamber working swine heart model Ann Thorac Surg 2000;70:1607-1614.[Abstract/Free Full Text]
- Ikeda K, Okazaki Y, Furukawa K, et al. Direct imaging of bileaflet mechanical valve behavior in the tricuspid position Eur J Cardiothorac Surg 2006;29:1014-1019.[Abstract/Free Full Text]
- Minato N, Itoh T. Direct imaging of the tricuspid valve annular motions by fiberoptic cardioscopy in dogs. I. Does De Vega’s annuloplasty preserve the annular motions?. J Thorac Cardiovasc Surg 1992;104:1545-1553.[Abstract]
- Nazarian S, Knight BP, Dickfeld TL, et al. Direct visualization of coronary sinus ostium and branches with a flexible steerable fiberoptic infrared endoscope Heart Rhythm 2005;2:844-848.[Medline]
- Babaliaros V, Block P. State of the art percutaneous intervention for the treatment of valvular heart disease: a review of the current technologies and ongoing research in the field of percutaneous valve replacement and repair Cardiology 2006;107:87-96.[Medline]
Related Article
-
Invited Commentary
- Georg Lutter
Ann. Thorac. Surg. 2008 85: 1065-1066.
[Extract]
[Full Text]
[PDF]
This article has been cited by other articles:
|
|
|
|
 
Y. Inoue
Beating Heart Cardioscopy: On-Pump or Off-Pump?
Ann. Thorac. Surg.,
October 1, 2008;
86(4):
1401 - 1401.
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. Mihaljevic
Reply.
Ann. Thorac. Surg.,
October 1, 2008;
86(4):
1401 - 1401.
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Lutter
Invited Commentary
Ann. Thorac. Surg.,
March 1, 2008;
85(3):
1065 - 1066.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
