of Cardioplegia: Assessment by Thermovision N. Shapira, M.D., G. M. Lemole, M.D., P. M. Spagna, M.D., F. J. Bonner, M.D.,

Transcription

1 Antegrade and Infusion of Cardioplegia: Assessment by Thermovision N. Shapira, M.D., G. M. Lemole, M.D., P. M. Spagna, M.D., F. J. Bonner, M.D., J. Fernandez, M.D., and D. Morse, M.D. ABSTRACT The efficacy of two routes of cardioplegia infusion was examined by assessing the hypothermia induced in patients with critically obstructed or occluded major coronary arteries. The antegrade (through the aorta) and the retrograde (through the coronary sinus) methods of cardioplegia infusion were compared using myocardial thermograms. Patients (N = 8) were matched according to angiographic similarities of their coronary artery disease. Adequate myocardial cooling distal to a critical obstruction could frequently not be obtained with antegrade infusion of cardioplegic solution. With retrograde infusion, the desired uniform cooling of the myocardium, as determined by thermographic analysis of the surface temperatures of the heart, was obtained. We conclude that retrograde infusion of cardioplegic solution induces more effective and homogeneous hypothermia in patients with critically obstructive multivessel coronary disease, and may provide improved myocardial protection. Adequate distribution of cardioplegic solution is essential for myocardial function and structural preservation during cardiac operations [l, 21. In the presence of severe multivessel coronary artery disease, protection by antegrade (aortic) infusion of cardioplegia may be inadequate because of underperfusion of myocardial territories supplied by occluded or severely obstructed arteries [3-51. It has been demonstrated in patients with normal coronary arteries that retrograde (coronary sinus) infusion may be an effective means to deliver cardioplegia [6]. The purpose of this study was to compare the results of antegrade and retrograde infusion in patients with severe coronary artery disease by assessment of the induced myocardial hypothermia. Myocardial temperatures were determined by infrared thermography. Material and Methods This study comprised 8 patients (Table). In 4 of them, cardioplegia was administered through the aorta. The first patient had aortic stenosis and normal coronary ar- From the Department of Surgery, Section of Cardiac Surgery, Washington Veterans Administration Hospital, and the Department of Surgery, Section of Cardiothoracic Surgery, Georgetown University, Washington, Dc. Accepted for publication Mar 3, Address reprint requests to Dr. Shapira, Department of Surgery, Section of Cardiac Surgery, Washington VA Hospital, 50 Irving St NW, Washington, DC teries; Patients 2 through 4 had severe and extensive coronary artery disease. In another 4 patients, cardioplegic solution was infused through the coronary sinus. Patient 5 had aortic stenosis and normal coronary arteries, and Patients 6 through 8 had coronary artery disease similar in severity and distribution to that in Patients 2 through 4. Patients with a previous myocardial infarction were not included. High-potassium (24 meql) cold (4" to 6 C) blood cardioplegia was used in all patients and was supplemented by systemic hypothermia (25" to 28 C). Topical cooling (iced normal saline lavage) was applied after the thermograms were obtained. When the aortic route was used, cardioplegic solution was infused through a 14F cannula for two minutes. Pressure at the proximal aorta was maintained at 100 to 120 mm Hg, and an average of 750 ml of cardioplegic solution was infused. When the retrograde method was employed, venous drainage was obtained by double caval cannulation through the right atrial appendage. The coronary sinus orifice was exposed through a short (2 to 3 cm) right atriotomy about 1 cm from and parallel to the atrioventricular groove. Cardioplegia was administered through a Foley catheter (22F), which was placed under direct vision. The Foley catheter balloon was inflated at the orifice of the coronary sinus just enough to occlude it. Cardioplegic infusion pressure was continuously monitored through a needle inserted into the Foley catheter lumen and connected to a Gould P23ID pressure transducer, and was maintained at 30 mm Hg. Infusion of 700 to 800 ml by this method required four to five minutes. Distal anastomoses were done first during a single aortic cross-clamp time. Aortic or coronary sinus cardioplegia infusion was repeated every twenty minutes, and was supplemented by infusion through each vein graft subsequent to completion of each distal anastomosis. In our investigation, we used the AGA Thermovision 782 System (AGEMA Infrared Systems, 550 County Ave, Secaucus, NJ 07094), which consists of a scanner, monitor, and videotape recorder (Fig 1). It provides a real-time pictorial reproduction of the infrared radiation emitted by the surface of the heart. The Thermovision 782 System converts this radiation through the scanner to a picture projected onto a color television screen and recorded on a magnetic videotape. Infrared energy from the object being viewed is focused via the front lens of the scanner through rotating prisms and collimating optics onto the infrared detector. The rotation of the prisms causes the detector to scan the focused image both vertically and horizontally and results in a video signal where the variation in video intensity is directly proportional to 92 Ann Thorac Surg 43:92-97, Jan 1987

2 93 Shapira, Lemole, Spagna, et al: Assessment of Cardioplegia Infusion by Thermovision Clinical Data Method of Patient Cardioplegia No. Infusion Coronary Artery Anatomy 1 Antegrade 2 Antegrade 3 Antegrade Antegrade Normal (aortic stenosis, no CAD) LAD, occlusion; circumflex, moderatekevere stenosis; RCA, occlusion Left main, moderate stenosis; LAD, moderatelsevere stenosis and occlusion of 1st diagonal; circumflex, severe stenosis of obtuse marginal and occlusion of 2nd marginal; RCA, occlusion LAD, occlusion (distal to 1st septal branch); circumflex, occlusion of 2nd marginal; RCA, occlusion Normal (aortic stenosis, no CAD) LAD, occlusion; circumflex, subtotal occlusion; RCA (dominant), severe stenosis at three locations LAD, moderate stenosis; circumflex, occlusion distal to occluded 1st marginal; RCA, occlusion LAD, moderatelsevere stenosis; circumflex, occlusion of 1st and 2nd marginal; RCA, occlusion CAD = coronary artery disease; LAD = left anterior descending coronary artery; RCA = right coronary artery. the temperature variation across the scene which is being scanned. The video signal is processed by the color monitor and is presented as a color map where the individual color contours (isotherms) are set to a predetermined temperature scale, that is, each color contour represents a fixed increment of temperature on the scanned object. In our investigations, the system was set to record temperatures ranging from 30" to 10 C. Ten different colors were projected, each corresponding to a discrete 2 C temperature range. The hottest areas (higher than 28 C) appear in white and the coldest areas (lower than 12"C), in black. A scale of the different temperature levels and the corresponding colors is presented in Figure 3. Thermograms were obtained both during and after cardioplegia infusion. Intramyocardial temperatures were also assessed in 4 patients by the use of Shiley myocardial needle thermistors. The needles were inserted in the distal interventricular septum and in myocardium covered by fat. These temperatures were compared with the corresponding surface temperatures displayed on the Thermovision screen. Fig 1. AGA Thermovision 782 System. It consists ofa scanner (far right), monitor, and videotape recorder. Results Eighteen selected frames from the thermograms obtained during and after infusion of cardioplegia in 8 patients are presented in Figures 2 and 3. These thermograms represent the temperature of the heart surfaces that were facing the scanner. The temperature of the nonfatty epicardial surface as displayed by the thermograms was slightly higher, always by 2" to 3"C, than the intramyocardial temperature measured by the needle thermistors. In contrast, areas covered by adipose tissue were always registered on the thermograms as patches of warm colors (red to yellow). The temperature differences between these areas and the myocardium beneath were wider (8" to 15 C) and variable. Adequate and even cooling of the heart was obtained by antegrade infusion in the absence of coronary artery disease (see Fig 2, frame la). Likewise, moderate or even more severely stenotic lesions (estimated as less than 90% narrowing) did not prevent adequate cooling of the corresponding myocardium, as demonstrated by the cooling of the left anterior descending coronary artery (LAD) territory in Patient 3 (see Fig 2, frame 3a). However, myocardium perfused by an occluded or subtotally occluded vessel was not adequately cooled when cardioplegic solution was infused through the aorta; this is shown in the territories of the LAD and posterior descending coronary artery in Patient 2 (see Fig 2, frames 2a, 2c) or the territories of the outer diagonal and circumflex marginal branches in Patient 3 (see Fig 2, frames 3b, 3d, 3e).

3 94 The Annals of Thoracic Surgery Vol 43 No 1 January 1987 Fig 2. Selected frames from thermograms obtained during operation with antegrade infusion. The heart is in the orientation seen by the surgeon on the right side of the operating table; the apex of the heart is toward the right in each frame. Dotted lines indicate heart border; broken lines indicate the surgeon's fingers. ( l a ) (Patient I.) This patient had no coronary artery disease; there is adequate and normal distribution of cardioplegic solution. (2a) (Patient 2.) lnadequate cooling of myocardium perfused by an occluded LAD is indicated by warm colors. (2b) Same region during infusion of cardioplegia through a vein graft shows adequate cooling and cool colors. (2c) By cephulad rotation of the heart, the P D A territory is exposed. (2d) This territory remains warmer until completion of the vein graft anastomosis to the RCA and infusion of the cardioplegic solution through it. (3a) (Patient 3. ) Noncritical obstruction did not prevent myocardial cooling. (3b) With administration of cardioplegia through a vein graft anastornosed to the LAD, adequate cooling of the entire anterolateral wall was obtained, except for the myocardium corresponding to the occluded proximal diagonal branch. (3c) This region was cooled only after completion of a side-to-side anastomosis to the LAD vein graft. In the next threeframes, the posterohferal wall is exposed by the rotation of the heart to the right. (3d) Only partial cooling could be obtained because of severe disease in the circumflex system. After completion of an end-to-side anastomosis to the second marginal branch (3e) and a side-to-side anastomosis to the proximal marginal branch (3f), adequate cooling was obtained by cardioplegia administration through the vein grafts. (4) (Patient 4. ) Despite LAD occlusion, adequate cooling was achieved because of a rich network of collaterals. (LV = left ventricular; LAD = left anterior descending coronary artery; PDA = posterior descending coronary artery; RCA = right coronav artery.)

4 95 Shapira, Lemole, Spagna, et al: Assessment of Cardioplegia Infusion by Thermovision Fig 3. Selected frames from thennograms obtained during operation with retrograde infusion. The heart orientation and the dotted and broken lines are the same as in Figure 2. Adequate and homogeneous cooling of the right and left ventricles was accomplished by coronary sinus cardioplegia infusion in the absence (5a, 5b) or the presence ( 6, 7a-7c) of coronary artery disease. In frame Sb, the heart is elevated to expose the apex and in frames 7b and 7c, it is rotated to the left to expose the inferior wall. Temperature scale and the corresponding colors are demonstrated in the right lower corner. During antegrade infusion, the myocardial thermograms corresponded well with the coronary arterial anatomy, the coldest areas always being the areas of the coronary arteries. The areas in between were somewhat warmer. After coronary sinus cardioplegia infusion, thermograms demonstrated a very diffuse and homogeneous pattern without showing any distinct anatomy and involving the right and left ventricles equally. Myocardial cooling distal to these lesions could be obtained only after completion of the distal anastomosis to each of these vessels, by the administration of cardioplegia through the vein grafts (see Fig 2, frames 2b, 2d, 3c, 3e, 30. In Patient 4, excellent cooling of the anterior wall was achieved with aortic infusion, despite total occlusion of the LAD and right coronary artery (RCA) as well as the thud circumflex marginal branch (see Fig 2, frame 4). Inspection of the coronary angiogram of this patient (Fig 4) revealed that the occlusions of the LAD and RCA were located distal to the first septa1 perforator and conus branches, thereby enabling the development of a very rich network of collaterals. This angiogram contrasts with that of Patient 2 in whom the LAD and RCA were occluded proximal to these branches and to collaterals which were not well developed (Fig 5). With retrograde infusion of cardioplegia, adequate and homogeneous cooling of the heart was uniformly obtained, and was not affected by the presence of severe and extensive coronary artery disease, as demonstrated by the thermograms of Patients 6 through 8 (see Fig 3, frames 6, 7a-7c). Myocardial cooling in these patients was similar to that achieved by a similar technique in a patient with normal coronary arteries (Patient 5) (see Fig 3, frames 5a, 5b). Thermographic methods have been previously used to study the peripheral circulation [71, to delineate myocardial ischemic areas [8], and to study myocardial temperature during infusion of cardioplegia in patients with coronary disease [3, 41. The concept of thermographic evaluation of retrograde cardioplegia was also described by Moravcsik and co-authors [9].In our study, the antegrade and the retrograde cardioplegia infusion techniques were compared in patients with severe and extensive coronary artery disease. With the Thermovision method, a real-time pictorial projection of the cardioplegia-induced myocardial hypothermia with each method was obtained, and could be analyzed in relation to the anatomy of the coronary artery diesease. Global intraoperative mapping of the distribution of the cold cardioplegic solution by the Thermovision method was found to be superior to needle thermistors, which provide only local temperature information. There are, however, some limitations to the Thermovision method. The thermal picture can reflect only the surface temperature, not the intramyocardial temperature, and it projects only what is "seen" by and faces the scanner. On the other hand, we have learned from experience that unless the epicardium is covered by fat, the epicardial temperature is only slightly (2"to 3 C) higher Comment

5 96 The Annals of Thoracic Surgery Vol 43 No 1 January 1987 Fig 4. Coronary angiogram of Patient 4. The occlusions of the left anterior descending coronary artery (LAD) and the right coronary artery (RCA) were located distal to the first septa1 perforator and the conus branches, thereby allowing the development ofa rich network of collaterals. Fig 5. Coronary angiogram of Patient 2. There was a poor network of collaterals related to the proximal locations of the occlusions of the left anterior descending coronary artery (LAD) and the right coronary artery.

6 97 Shapira, Lemole, Spagna, et al: Assessment of Cardioplegia Infusion by Thermovision than the intramyocardial temperature. Because this difference is consistent, we think that thermal pictures of the surface of the heart can serve as reliable indicators of myocardial cooling. Furthermore, by tilting the heart, thermal pictures of other aspects of the heart can be obtained. The thermograms in this study demonstrate that with antegrade infusion of cardioplegia, regions of myocardium that are perfused by severely diseased coronary arteries receive an inadequate amount of cardioplegic solution and remain warm. In some patients, as in Patient 4 in the present study, even distribution of cardioplegia was achieved with the antegrade method despite severe triple-vessel coronary artery disease. This may be attributed to the more distal location of the occlusive lesions, thereby sparing the proximal segment of the vessels, including the takeoff of the first septa1 perforator, conus, and obtuse marginal branches, and allowing the development of a rich network of collaterals. It has been demonstrated that in most patients, however, the collateral vessels are unable to provide adequate blood flow to adjacent areas perfused by stenosed or occluded vessels [lo]. This finding is supported by the work of Ekroth and associates [3] and by the thermograms of our Patients 2 and 3. Only when a coronary bypass operation is executed intelligently by determining the sequence of the bypass grafts using myocardial temperature mapping as proposed by Daggett [ll], Vander Salm [12], and their associates is early infusion of cardioplegic solution to these "warm" regions possible. It requires completion of one or more distal anastomoses, and thereby creates a delay that may be critical, prior to the delivery of cardioplegia to myocardial regions that are at the highest risk for myocardial injury. The adequacy of myocardial protection achieved by Menasche and colleagues [6] using coronary sinus cardioplegia infusion in patients with normal coronary arteries, and the experimental demonstration of modifying myocardial ischemia due to coronary artery occlusion by retroperfusion of the coronary sinus [ 13-15] stimulated us to apply the retrograde method in patients with coronary artery disease. The patients selected for this technique had coronary artery disease comparable in severity and distribution to the patients studied by the antegrade technique. Yet adequate and homogeneous cooling of the heart was achieved in all 4. By delivering the cardioplegic solution with the retrograde method, which, unlike the antegrade method, is not dependent on the patency of the coronary arterial tree, even distribution to all myocardial areas regardless of the extent of coronary artery disease, in the left as well as the right ventricles, could be accomplished without untoward delay. In conclusion, in patients with severe and extensive coronary artery disease, homogeneous cooling of the heart may not be accomplished by aortic infusion of cardioplegia and some territories may remain inadequately protected. In these patients, the coronary sinus appears to offer a superior delivery route for cardioplegia. We appreciate the help of AGEMA Infrared Systems in providing us with the Thermovision 782 System for this study. Our special thanks go to Mr. Steven DeFilipo for his expertise and technical assistance and to Mrs. Carole Eiting for the preparation of the manuscript. References Macmanus Q, Grunkemeier G, Lambert L, et al: Aortic valve replacement and aorta-coronary bypass surgery: results with perfusion of proximal and distal coronary ar- teries. J Thorac Cardiovasc Surg 75:865, Becker H, Vinten-Johansen J, Buckberg GD, et al: Critical importance of ensuring cardioplegic delivery with coronary stenoses. J Thorac Cardiovasc Surg 81:507, Ekroth R, Berggren H, Sudow G, et al: Thermographic demonstration of uneven myocardial cooling in patients with coronary lesions. Ann Thorac Surg 2931, Gagliardi C, Porntales D, Del-Naja C, et al: Thermographic evaluation of cardioplegia diffusion in coronary patients. In Coronary Artery Surgery. New York, Springer-Verlag, 1984, pp Chiu RCJ, Blundell PE, Scott HJ, Cain S: The importance of monitoring intramyocardial temperature during hypothermic myocardial protection. J Thorac Cardiovasc Surg 28:317, Menasche P, Kural S, Fauchet M, et al: coronary sinus perfusion: a safe alternative for ensuring cardioplegic delivery in aortic valve surgery. Ann Thorac Surg 34547, Winsor T: Vascular aspects of thermography. J Cardiovasc Surg (Torino) 12:379, Robicsek F, Masters TN, Svenson RH, et al: The application of thermography in the study of coronary blood flow. Surgery 84:858, Moravcsik E, Papp L, Lengyel I, Szabo Z Thermographic evaluation of retrograde cardioplegia: experimental and clinical studies. In Mohl W, Wolner D, Gloger D (eds): The Coronary Sinus: First International Symposium on Myocardial Protection via the Coronary Sinus. New York, Springer-Verlag, 1984, pp Oldham HN, Jones RH, Harris CC, et al: Intraoperative relationships between regional myocardial distribution of bypass graft flow and the coronary collateral circulation. J Thorac Cardiovasc Surg 77:32, Daggett WM, Jacocks MA, Coleman WS, et al: Myocardial temperature mapping: improved intraoperative myocardial preservation. J Thorac Cardiovasc Surg 82:883, Vander Salm TJ, Okike ON, Cutler BS, et al: Improved myocardial preservation by improved distribution of cardioplegic solutions. J Thorac Cardiovasc Surg 83:767, Gundry SR Modification of myocardial ischemia in normal and hypertrophied hearts utilizing diastolic retroperfusion of the coronary veins. J Thorac Cardiovasc Surg 83:659, Meerbaum S, Haendchen RV, Corday E, et al: Hypothermic coronary venous phased retroperfusion: a closed chest treatment of acute regional myocardial ischemia. Circulation 65:1435, 1982 Bolling SF, Flaherty JT, Bulkley BH, et al: Improved myo- cardial preservation during global ischemia by continuous retrograde coronary sinus perfusion. J Thorac Cardiovasc Surg 86:659, 1983