On-line intraoperative quantitation of regional myocardial perfusion during coronary artery bypass graft operations with myocardial contrast two-dimensional echocardiography

On-line intraoperative quantitation of regional myocardial perfusion during coronary artery bypass graft operations with myocardial contrast two-dimensional echocardiography We hypothesized that the success of coronary artery bypass graft operations could be assessed by means of on-line quantitative myocardial contrast echocardiography. Accordingly, myocardial contrast echocardiography was performed at baseline and after each placement of venous graft in 21 patients undergoing coronary artery bypass graft operations. Time-intensity plots were generated on-line with the use of a dedicated computer system, and areas under the curve were assessed for each injection. Successful on-line quantitation of myocardial contrast echocardiography data was performed in 17 patients; this aUowed comparison before and after coronary artery bypass graft operations for 21 grafts, with agreement between expert visual analysis and quantitative data in 91 % of these cases. Three distinct perfusion patterns were noted on myocardial contrast echocardiography: (1) reduced contrast effect before coronary artery bypass graft operations with improvement after coronary artery bypass graft operations (n = 11); (2) adequate contrast effect before coronary artery bypass graft operations with no change after coronary artery bypass graft operations (n = 9) (for patients in group 2, the mean percentage of coronary stenosis was less than the mean for patients in group 1-67% ± 25 % vs, 88 % ± 20 %, p = 0.05); and (3) no contrast effect either before or after coronary artery bypass graft operations in one patient with previous infarction. One third of the time (34 of 95 injections), on-line quantitation was unsuccessful. Failure was related three times more often to problems associated with myocardial contrast echocardiography, such as attenuation and inadequate quality of bubbles, than to computer failure. Despite its limitations, on-line quantitative myocardial contrast echocardiography is feasible in patients undergoing coronary artery bypass graft operations and provides important objective information regarding the success of revascularization. (J THORAC CARDIOVASC SURG 1992;104:1524-31)

Flordeliza S. Villanueva, MD, William D. Spotnitz, MD, Ananda R. Jayaweera, PhD, John Dent, MD, Lawrence W. Gimple, MD, and Sanjiv Kaul, MD, Charlottesville, Va.

From the Division of Cardiology, Department of Internal Medicine, and Division of Thoracic and Cardiovascular Surgery, Department of Surgery, University of Virginia School of Medicine, Charlottesville, Va. Supported in part by grants-in-aid (Dr. Spotnitz and Dr. Kaul) from the Virginia Affiliate of the American Heart Association, Glen Allen, va, FIRST Awards (R29-HL38345-Dr. Kaul and R29HL43787-Dr. Spotnitz) from the National Institutes of Health, Bethesda, Md; and, an instrument grant from General Electric Medical Systems, Milwaukee, Wis. Dr. Villanueva is the recipient of a fellowship training grant from the Virginia affiliate of the American Heart Association. Dr. Kaul is the recipient of the

1524

Established Investigator Award from the National Center of the American Heart Association, Dallas, Tex. Presented in part at the Fortieth Scientific Session of the American College of Cardiology, Atlanta, Ga. Received for publication July 23, 1991. Accepted for publication Feb. 11, 1992. Address for reprints: Sanjiv Kaul, MD, Divisionof Cardiology, Box 158, University of Virginia, Charlottesville, VA 22908.

12/1/37029

Volume 104 Number 6 December 1992

On-line intraoperative quantitation of myocardial perfusion during CABG

Adequate myocardial preservation and successful rev ascularization are two major goals during coronary artery bypass graft (CABG) operations. Quantitative myocardial contrast echocardiography (MCE) has been shown to accurately assess the rate of delivery of cardioplegic solutions within the myocardium in a canine model of graded coronary artery stenosis. I The potential of this technique to assess the adequacy of revascularization has been demonstrated in experimentalv ' and clinical

studies."? For MCE to have practical value, reliable information regarding the adequacy of cardioplegic solution delivery and the success of individual graft placement should be available to the surgeon at the time of operation. One approach for providing such information would be on-line quantitation of myocardial perfusion during delivery of cardioplegic solution at baseline and after each placement of a venous graft. In this study, we tested the hypothesis that intraoperative on-line quantitation of myocardial perfusion could be performed with the use of MCE. We had three aims in mind. The first was to determine whether the on-line quantitative information regarding perfusion before and after CABG was consistent with expert visual assessment. The second was to determine the patterns of perfusion before and after CABG so as to correlate these with preoperative findings. The third was to evaluate the feasibility of quantitative on-line MCE in the operating room. Methods Study population. We studied twenty-one patients (16 men and 5 women with a mean age of 61 ± 9 years) with multivessel coronary artery disease who were undergoing elective CABG. The study protocol was approved by the Human Investigations Committee at the University of Virginia, and all patients gave informed consent. The coronary angiograms were interpreted by a single blinded observer who used hand-held calipers to measure the percentage of luminal diameter narrowing. Myocardial contrast echocardiography. Epicardial imaging was performed in 14 patients with a hand-held 5 MHz transducer (ND-256, Biosound, or RT5000, GE Medical Systems, Waukesha, Wis.) placed in a sterile sheath (Civco Medical Instruments Co., Inc., Kalona, Iowa) containing sterile ultrasound gel. A saline-filled pericardial cradle acted as an acoustic interface between the heart and the transducer. Transesophageal imaging was performed in 7 patients with a 5 MHz transducer (RT 6800, GE Medical Systems) positioned in the stomach. All images were obtained in the short-axis view at the midpapillary muscle level and were recorded on 0.5-inch videotape (Panasonic model AG6200, Matsushita Electric Corp. of America, Secaucus, N.J.). Optimal gain settings were established initially and were held constant during the remainder of each operation.

I 525

Sonicated 5% human albumin (Miles, Inc., Cutter Biological, Covina, Calif.) was used as the contrast agent. It was prepared with a commercial system (model W-375, Heat Systems Ultrasonics Inc., Farmingdale, N.Y.) by a previously described method that produces microbubbles with a mean size of 4.5 ~m. Sonication was performed by the same person throughout the study period and was performed immediately before each injection. The same volume of contrast (0.5 to I ml) was injected into the patient during each stage. On-line quantitation. Our method for the on-line quantitation of myocardial perfusion, previously described, I! incorporates the use of a custom-designed dedicated system that uses an 80386-based microcomputer (system 310, Dell Computer Corp., Austin, Tex.). The digital conversion of analog images was performed by means of an internal frame grabber (PCVISIONPLUS, Imaging Technology, Inc., Woburn, Mass.). Up to six regions of interest, 9 X 9 pixels in size, can be positioned over the image by the operator with a hand-held mouse. Average videointensity was measured in real time (30 frames per second) for each region of interest within the frame grabber. Thedata and their corresponding times are written into a file, which is automatically transferred to a data analysis and graphics package (RS/I, Bolt, Beranek, and Newman, Cambridge, Mass.). After the injection is completed, the background-subtracted time-intensity plots are automatically depicted on the computer screen. Each data point on the plots represents average background-subtracted videointensity from three consecutive frames. Pre-CABG and post-CABG plots are depicted side-by-side for comparison. Quantitative image analysis. The time-intensity data generated in the operating room were fitted to a double exponential function!' I! that used RS/I on a minicomputer (VAX 8200, Digital Equipment Corp., Marlboro, Mass.). We have found this function to best describe time-intensity data for contrast injections in the cold, cardioplegia-arrested heart. I The initial area under the curve was calculated from the curve fit. For all series of injections, this measurement was made at the time of baseline injection, at which point maximal videointensity was observed. An improvement in regional flow was defined as an increase of 25% or more. We have shown that the area under the curve correlates well with radiolabeled microsphere flow measured in the cardioplegia-arrested heart.' This analysis was performed to determine whether the intraoperative on-line analysis had been both correct and meaningful. Intraoperative protocol. With the patient on cardiopulmonary bypass, the aorta was crossclamped and a cold, hyperkalemic crystalloid cardioplegic solution was delivered into the aortic root. The flow rate was adjusted to maintain cardioplegic solution line pressure at approximately 180 mm Hg. The corresponding mean aortic root pressure was measured concomitantly in three patients and was approximately two thirds of the value in the cardioplegic solution line. Shortly after cardiac arrest, during continuous delivery of cardioplegic solutions, the first injection of contrast medium was made through a port in the cardioplegic solution line. This injection was used to determine the adequacy of cardioplegic solution delivery to the myocardium before placement of any bypass graft. Before the injection of contrast medium the transducer was positioned to obtain an optimal short-axis view. Three to six regions of interest, representing different vascular beds, were

I 526

The Journal of Thoracic and Cardiovascular Surgery

Villanueva et al.

PATIENTSI

121

/17(81%) PATIENTS

16 1

oJ,

INJECTIONS

39(64%) INJECTIONS SUCCESSFUL

I

/4(19%) PATIENTS

I

oJ,

I

1 2 INJECTIONS UNSUCCESSFUL

22(36%) INJECTIONS UNSUCCESSFUL

Fig. 1. Success and failure rates in acquisition of quantitative on-line data in 21 patients included in study.

Table I

Increase in area under the curve 2': 25% Increase in area under the curve <25%

Visual increase in contrast

No visual increase in contrast

II

o

2

8

placed over the myocardium. Data acquisition was then initiated, and contrast medium was injected. When contrast was no longer seen in the myocardium, data acquisition was terminated. Data were usually acquired for 30 to 60 seconds.These steps were repeated after placement of each venous graft. Data analysis. The time-intensity plots generated in the operating room were visually assessed with respect to the preCABG and post-CABG area under the curve.An assessmentof changes in the regional flow to the bypassed bed was made on the basis of these data. These evaluations were compared with expert visual analyses performed at a later date without knowledge of the quantitative results. Improvement in perfusion, assessedby visual analysis, was defined as a significantincrease in contrast effect in areas that previously showed little or no contrast. On-line quantitation was deemed successful when, after injection of contrast, quantifiable data were available for the bypassed vascular bed. The reasons for unsuccessful on-line quantitation were categorized as resulting from either MCE, such as acoustic attenuation, or the on-line method, such as system malfunction. All numerical data were compiled on a VAX 8200 minicomputer, and statistical analysis was performed with RSjl. The paired t test was used to compare pre-CABG and post-CABG

parameters. Differences in proportions were tested by X2 analysis. Statistical significancewasdefinedas p < 0.05 (two-sided). Results The 21 patients received a total of 38 saphenous vein bypass grafts: 12 to the left anterior descending coronary artery or its diagonal branch, 10 to an obtuse marginal branch of the left circumflex coronary artery, and 16 to the right coronary artery or its posterior descending branch. In all patients the proximal anastomosis was performed first. Bypasses incorporating the left internal mammary artery were excluded from the analysis. Reasons for this will be described later. Quantitative versus visual assessment. Quantitative M CE data were acquired in 17 (81 %) of the 21 patients (Fig. 1). These 17 patients received 61 pre-CABG and post-CABG injections for 21 vascular beds. Thirty-nine injections (64%) were successful. Expert visual and quantitative assessments of improvement in post-CABG perfusion in these 21 vascular beds were in agreement 91 % of the time (Table I). That is, when visual analysis suggested improvement in perfusion after CABG, there was at least a 25% increase in the area under the curve 85% of the time (p < 0.0002). Conversely, when there was no visual improvement in perfusion, in no case did the area under the curve increase by 25% or more. Perfusion patterns before and after CABG. Analysis of pre-CABG and post-CABG data demonstrated three distinct patterns of contrast enhancement based on analysis of areas under the curve: reduced contrast effect before and improved contrast effect after CABG; ade-

Volume 104 Number 6 December 1992

On-line intraoperative quantitation of myocardial perfusion during CABG

1527

c. 25

>

CABG

I

20

l-

ii)

z w

15

(5

10

,

toZ

w

PRE-RCA CABG

Q

s

5

I

0"----"---'--........--'-----'

o

8

16

24

32

40

TIME (sec)

Fig. 2. A, Prebypass image from patientwith rightcoronary artery stenosis showing poor contrasteffect in poste-

rior myocardium. B, Image from same patient after successful placement of bypass graft to right coronary artery showing adequate perfusion of bed. C, Quantitative data obtained on lineshowing improvement of perfusion after placement of graft.

quate contrast enhancement at baseline and no change after CABG; and no contrast effect either before or after CABG. Improved contrast effect after CABG was noted in regions supplied by 11 grafts (area = 13 ± 13 before CABG versus 98 ± 76 after CABG, p = 0.006). Fig. 2, A illustrates a baseline image in a patient with right coronary artery disease and a pre-CABG relative reduction in contrast effect in the posterior wall. After a graft to the right coronary artery was constructed, contrast effect in this bed was significantly improved both visually (Fig. 2, B) and quantitatively (Fig. 2, C). In nine instances, adequate contrast enhancement was presentat baseline, with increase of 25%or lessin contrast effect after CABG (area = 36 ± 37 before CABG versus 22 ± 20 after CABG, p = 0.34). Patients in whom the contrast effect was not reduced at baseline and was not changed after CABG had less angiographic stenosis than patients who had less contrast effect before CABG and showed improvement in contrast effect after CABG (67% ± 25% versus 88% ± 20%, p = 0.05). Fig. 3, A illustrates adequate contrast effect (indicated by arrow) in the pre-CABG image of a patient undergoing bypass to a diagonal artery. The post-CABG image was identical.

Finally, the persistent absence of contrast enhancement was noted in one patient. In this patient with right coronary artery occlusion, contrast was not noted in the inferior interventricular septum before CABG and remained absent from this region despite seemingly successful CABG to the right coronary artery (Fig. 3, B). This patient had a remote inferior infarction associated with akinesia in that region. Reasons for unsuccessful on-line quantitation. Online quantitation was unsuccessful during 34 of the 95 injections (36%) (Fig. I). Twenty-six of these failures (three fourths) were related to MCE and 8 were related to the on-line computer system. Reasons for injection failure related to MCE included acoustic attenuation resulting from marked myocardial contrast enhancement in the near field (10 injections) or intense left ventricular cavity contrast resulting from aortic regurgitation (6 injections). Although quantitative data were acquired for these 16 injections, they were not meaningful because of the attenuation that occurred. In 10 injections, contrast medium was not detected in any myocardial segment despite the fact that the aortic root injection seemed to be successful (Table 11). Data loss related to the on-line system occurred with respect to 8 injections. Reasons for failure included maI-

15 28

The Journal of Thoracic and Cardiovascular Surgery

Villanueva et al.

Table II

A.

Reasons for unsuccessful quantitation

MCE-re1ated Attenuation caused by Anterior wall opacification Left ventricular cavity contrast from aortic regurgitation Nocontrast effect in the myocardium On-line method-related computer malfunction Area-of-interest misplacement

No. of injections

10 6 10

4 4

Discussion

B.

Fig. 3. A, Prebypass image from patient with stenosed diagonal branch of left anterior descending coronaryartery showing adequate contrast effect in anterior wall (arrow). This effect remained unchanged after successful graft placement to the diagonal vessel. B, Postbypass image from patient with severe right coronary artery stenosisand prior inferior wall infarction associated with regionaldyssynergy. No contrast is notedin the posteriorseptum and part of the posteriorwall (arrow), despite successful graft placementto the posteriordescending branch of the right coronary artery (the image was reversed during data acquisition). This contrast effect is unchanged from that in a prebypass image.

function of the computer (4 injections) and imprecise region-of-interest placement (4 injections, Table 11). These regions either overlapped the left ventricular cavity or the epicardium or subtended an irrelevant vascular bed. In all cases, the region of interest was located correctly before injection of contrast medium, but slight movement of the transducer during data acquisition caused the myocardium to shift out from under it.

Despite major advances in myocardial preservation and surgical techniques, myocardial infarction during or immediately after surgical procedures remains a significant problem in patients undergoing CABG. Depending on the sensitivity of the technique used to detect perioperative infarction, its frequency can vary from 1 in 20 to I in 5 patients. 12- 15 Potential reasons for myocardial infarction during CABG include inadequate delivery of cardioplegic solution to regions supplied by severe coronary stenosis'< 17 and inadequate revascularization. The need to develop a practical and feasible method for the intraoperative evaluation of delivery of cardioplegic solution to the myocardium and the success of CABG is, therefore, compelling. With such a method, regions of the myocardium receiving the least amount of cardioplegic solution could be bypassed first and cardioplegic solution could be delivered to these regions via the bypass grafts. Similarly, if a bypass did not result in adequate revascularization, it could be immediately revised. Although the potential contribution of MCE toward this end has appeared promising, 1-9 an objective means for the instantaneous intraoperative quantitation of MCE data is currently unavailable. The present study was intended to address the feasibility, logistics, and applicability of on-line quantitation of myocardial perfusion by means of MCE during delivery of cardioplegic solution. We found a high degree of agreement between expert visual assessment and on-line quantitative data. This concordance establishes the reliability of the data generated by the on-line method and paves the path for its use in settings in which the visual analysis of M CE images may be less familiar. Patterns of contrast effect. Three types of contrast pattern were identified on quantitative analysis. In most cases, reduced contrast effect was noted in regions of the myocardium that were to be bypassed; definite improvement in contrast enhancement in these regions after graft

Volume 104 Number 6 December 1992

On-line intraoperative quantitation of myocardial perfusion during CABG

placement was found, suggesting improved flow to the bypassed bed. I. 2 In other instances, there was no evidence of reduced baseline contrast effect within regions supplied by coronary arteries deemed on angiography to be stenotic. These regions showed no change in contrast effect after placement of a bypass graft. The percentage of stenosis of the vessels supplying these regions, however, was less than that in vessels supplying regions that showed improved contrast effect after bypass. It is, therefore, possible that these stenoses were not flow-limiting under the hemodynamic conditions present during delivery of cardioplegic solution. In this situation it may be useful to elicit maximal blood flow reserve. IS, 19 Cold cardioplegia and myocardial hypoxia can result in coronary vasodilation, 20, 21 which, however, may not be enough to maximally dilate the coronary bed under the hemodynamic conditions present during aortic crossclamping. Pharmacologic agents'? or other means may be required to elicit maximal flow reserve, and the resulting flow heterogeneity could be measured with the use of MCE.22 In only one case was no contrast effect noted within a myocardial region both at baseline and after seemingly successful CABG. In the patient with this pattern, a previous infarction and dyssynergy were noted in that region. Mudra and colleagues? reported a similar patient in their intraoperative study with MCE. It is likely that, in such a situation, fibrous scar was present without any intramyocardial blood vessels. In the case reported by Mudra and colleagues, the flow in the graft itself was normal, and it is likely that it was being diverted to other regions of the myocardium via collateral vessels.P Comparison with previous studies using MCE. We believe that the introduction of contrast medium into the crossclamped aortic root has several advantages over direct injection of contrast medium into a graft. 6-9 First, there is less chance of injury to the graft. Second, the adequacy of delivery of cardioplegic solution is determined at baseline, which could determine the order of graft placement. Third, the operation does not have to be interrupted, since cardioplegic solution is routinely delivered into the crossclamped aorta at baseline and after placement of each graft. Fourth, if the flow of cardioplegic solution to the crossclamped aortic root and the amount of contrast medium injected are held constant at each stage, the input function remains essentially the same, allowing meaningful comparisons between preCABG and post-CABG data. 24 Finally, ifthe flow to the aortic root is known and the input function can be determined (by imaging of the crossclamped aortic root during contrast injection), then actual (rather than relative) flow to myocardial regions can be calculated"

15 2 9

Limitations of intraoperative MCE. Unsuccessful on-line quantitation occurred mostly during the early stages of our experience; it was difficult to optimize the injection dosage that yielded the least attenuation within the time available in the operating room. With time, unsuccessful on-line quantitation occurred less frequently. During some injections we did not see or measure any myocardial opacification; this might have been related to suboptimal quality of the bubbles. High pressures have been reported to cause reduction in bubble reflectivity.P which may be related to reduction in either their size or their number. Furthermore, although we consistently produce bubbles of approximately the same size, we are not as skilled in producing the same concentration of bubbles during each sonication. Commercial agents currently undergoing preclinical testing have standardized bubble sizes and concentrations.P: 27 These bubbles could eliminate the problems of attenuation and suboptimal image quality and would also make MCE more feasible. Transesophageal echocardiography, coupled with an automated method of contrast injection, would also be generally more acceptable in the operating room environment. For comparison of pre-CABG and post-CABG timeintensity curves, the input function should remain the same between injections.l" Thus it is difficult to compare data generated when distal anastomoses are performed first, where direct injection of contrast medium into the graft is done after CABG, with pre-CABG data, where the injection is made into the aortic root. In addition, variations in flow rate of cardioplegic solution, which can occur at any time in the series of injections, could make quantitative comparisons before and after CABG difficult to make, even when contrast medium is delivered directly into the aortic root. With the exception of one case, however, the variability in cardioplegic solution flow rates in our study was less than 15%. We believe, therefore, that the pre-CABG and post-CABG comparisons in this study were justifiable, since we chose a 25% difference in area under the curve as the minimum level for a significant change in perfusion. Our on-line system for quantitation of perfusion performed well, with incorrect region-of-interest placement being the major cause of unusable data related to the computer system. With placement of more regions of interest (up to six), we were able to sample more of the myocardium, which allowed us to examine data from alternative, neighboring regions in the event that placement in another region was incorrect. System failure itself rarely occurred. A major limitation of our technique of injecting

I 530

Villanueva et al.

contrast medium into the crossc1amped aortic root is that we cannot assess the success of internal mammary artery bypass because these vessels arise distal to the site of the aortic crossc1amp. It is for this reason that mammary artery grafts were not evaluated in this study. It is likely that mammary artery patency could be better assessed by injection of contrast material into the left atrium of the beating heart after the aortic crossc1amp has been removed. In such an instance, however, since one would be dealing with a blood-perfused beating heart, the parameters measured and the algorithms applied would be different. I, 11,24 Gating would also be required to obtain the same frame from each cardiac cycle.i" Injection of contrast medium into the left atrium would also permit quantitative pre-CABG and post-CABG comparison of regional perfusion when distal anastomoses are performed first. Such an approach is currently being evaluated in our laboratory.

Conclusions The results of our study indicate that quantitative MCE can be performed on line for the intraoperative assessment of myocardial perfusion in patients undergoing CABG. Quantitative data are in agreement with expert visual analysis. Three distinct myocardial contrast patterns are noted, with each providing unique pathophysiologic insight. Despite its current limitations, many of which can be overcome with experience and standardization of contrast agents, MCE provides an objective approach for the intraoperative assessment of myocardial perfusion. Further investigation is needed to determine the value of our on-line method in the enhancement of myocardial preservation, optimization of revascularization, and improvement in clinical outcome in patients undergoing CABG. REFERENCES 1. Keller MW, Spotnitz WD, Matthew TL, Glasheen WP, Watson DO, Kaul S. Intraoperative assessment of regional myocardial perfusion using quantitative myocardial contrast echocardiography: an experimental evaluation. J Am Coli CardioI1990;16:1267-79. 2. Spotnitz WD, Keller MW, Watson DO, Nolan SP, Kaul S. Success of internal mammary bypass grafting can be assessed intraoperatively using myocardial contrast echocardiography. J Am Coli Cardiol 1988;12:196-201. 3. Aronson S, Bender E, Feinstein SB, et al. Contrast echocardiography: a method to visualize changes in regional myocardial perfusion in the dog model for CABG operations. Anesthesiology 1990;72:295-30. 4. Goldman ME, Mindich BP. Intraoperative cardioplegic contrast echocardiography for assessing myocardial perfu-

The Journal of Thoracic and Cardiovascular Surgery

sion during open heart operations. J Am Coil Cardiol 1984;4: 1029-34. 5. Matthew TL, Keller MW, Kaul S, Spotnitz WD. Assessment of myocardial perfusion during coronary artery bypass graft operations in humans using myocardial contrast echocardiography. Surg Forum 1989;40:248-51. 6. Kabas JS, KissloJ, Flick CL, et al. Intraoperative perfusion contrast echocardiography: initial experience during coronary artery bypass grafting. J THORAC CARDIOVASC SURG 1990;99:536-42. 7. Mudra H, Zwehl W, Klauss V, et al. Intraoperative myocardial contrast echocardiography for assessmentof regional bypass perfusion. Am J Cardiol 1990;66:1077-81. 8. Smith J, Feinstein SB, Kapelanski DP, Karp RB, Roizen MF. Transesophageal echocardiographic determination of myocardial perfusion during cardiac operations [Abstract]. Circulation 1986;74(2 Pt 2):II475. 9. Aronson S, Lee BK, Wiencek JF, et al. Assessment of myocardial perfusion during CABG surgery with twodimensional transesophageal contrast echocardiography. Anesthesiology 1991;75:433-40. 10. Keller MW, Glasheen WP, Teja K, Gear A, Kaul S. Myocardial contrast echocardiography without significant hemodynamic effects or reactive hyperemia: a major advantage in the imaging of regional myocardial perfusion. J Am Coil CardioI1988;12:1039-47. II. Jayaweera AR, Matthew TL, Sklenar J, Spotnitz WD, Watson DO, Kaul S. Method for the quantitation of myocardial perfusion during myocardial contrast two-dimensional echocardiography. J Am Soc Echocardiogr 1990; 3:91-8. 12. Rose 0 M, Gelfish J, Jacobowitz IJ, et al. Analysis of morbidity and mortality in patients 70 years of age and over undergoing isolated coronary artery bypass surgery. Am Heart J 1985;110:341-6. 13. Force T, BloomfieldP, O'Boyle JE, et al. Quantitative twodimensional echocardiographic analysis of regional wall motion in patients with perioperative myocardial infarction. Circulation 1984;70:233-41. 14. Burns RJ, Gladstone PJ, Tremblay PC, et al. Myocardial infarction determined by technetium-99m pyrophosphate single-photon tomography complicating elective coronary artery bypass grafting for angina pectoris. Am J Cardiol 1989;63:1429-34. 15. Val PG, Pelletier LC, Hernandez MG, et al. Diagnostic criteria and prognosis of perioperative myocardial infarction following coronary bypass. J THORAC CARDIOVASC SURG 1983;86:878-86. 16. Hilton CJ, Teubl W, Acker M, et al. Inadequate cardioplegic protection with obstructed coronary arteries. Ann Thorae Surg 1979;28:323-34. 17. Grondin CM, Helias J, Vouhe PR, Robert P. Influence of a critical coronary artery stenosison myocardial protection through cold potassium cardioplegia. J THORAC CARDIOVASC SURG 1981;82:608-15. 18. Gould KL, Lipscomb K, Hamilton GW. Physiologicbasis

Volume 104

Number 6 December 1992

19.

20.

21. 22.

23.

On-line intraoperative quantitation of myocardial perfusion during CABG

for assessing critical coronary stenosis: instantaneous flow response and regional distribution during coronary hyperemia as measures of coronary flow reserve. Am J Cardiol 1974;33:87-94. Wilson RF, White CW. Intracoronary papaverine: an ideal coronary vasodilator for studies of the coronary circulation in conscious humans. Circulation 1986;73:444-51. McDonagh PF, Laks H. Use of cold blood cardioplegia to protect against coronary microcirculatory injury due to ischemia and reperfusion. J THORAC CARDIOVASC SURG 1982;84:609-18. Berne RM, Blackman JR, Gardner TH. Hypoxemia and coronary flow. J Clin Invest 1957;36:1101-6. Keller MW, Glasheen W, Smucker ML, Burwell LR, Watson DD, Kaul S. Myocardial contrast echocardiography in humans. II. Assessment of coronary blood flow reserve. J Am Coli Cardiol 1988;12:925-34. Kaul S. Quantitation of myocardial perfusion with contrast echocardiography. Am J Card Imag 1991;5:200-216.

I53I

24. Shapiro JR, Reisner SA, Lichtenberg GS, Meltzer RS. Intravenous contrast echocardiography with use of sonicated albumin in humans: systolic disappearance of left ventricular contrast after transpulmonary transmission. J Am Coli Cardiol 1990;16:1603-07. 25. Keller MW, Glasheen WP, Kaul S. Albunex: a safe and effectivecommercially produced agent for myocardial contrast echocardiography. J Am Soc Echocardiogr 1989;2: 48-52. 26. SchliefR, Stacks T, Mahler M, et al. Enhanced color Doppler echocardiography of the left heart after intravenous injection of a new saccharide based agent in humans [Abstract]. Circulation I990;(Pt 2):III28. 27. Kaul S, Kelly P, Oliner JD, Glasheen WP, Keller MW, Watson DD. Assessment of regional myocardial blood flow with myocardial contrast two-dimensional echocardiography. J Am Coli Cardiol 1989;13:468-82.

Bound volumes available to subscribers Bound volumes ofTHE JOURNAL OF THORACIC AND CARDIOVASCULAR SURGERY areavailable tosubscribers (only) forthe 1992 issues from thePublisher, at a costof $71.00 fordomestic, $93.97 forCanadian, and $89.00 forinternational subscribers forVol. 103 (JanuaryJune) and Vol. 104 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising isremoved. Copies are shipped within 60 daysafterpublication of thelastissue of thevolume. Thebinding isdurable buckram with theJOURNAL name, volume number, andyearstamped in gold onthe spine. Payment must accompany all orders. Contact MosbyYear Book, Inc., Subscription Services, 11830 Westline Industrial Drive, St. Louis, Missouri 63146-3318, USA;phone (800) 325-4177, ext. 4351, or (314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.