| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 1999;15:173-179
© 1999 Elsevier Science NL
a Department of Cardiothoracic Surgery, University of Cologne, Joseph-Stelzmann-Strasse 9, 50924 Cologne, Germany
Department of Experimental Medicine, University of Cologne, Cologne, Germany
Department of Pharmacology, University of Cologne, Cologne, Germany
Received 22 September 1998; received in revised form 10 November 1998; accepted 25 November 1998.
* Corresponding author. Tel.: +49-221-478-6043; fax: +49-221-478-5906; e-mail: otto.dapunt@medizin.uni-koeln.de
 |
Abstract |
|---|
 Top Abstract Introduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
Key Words: Minmally invasive direct coronary artery bypass grafting • Myocardial protection • Myocardial metabolism • Nucleotides • Heat-shock protein • Intracoronary shunts • Off-pump-revascularization
|
Introduction |
|---|
|
TopAbstractIntroduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
Our personal clinical experience has been in contrast to others who reported that ischemia does not seem to be an issue in off-pump coronary surgery because the occlusion of the coronary artery usually is located in the distal segments and does not last longer than 15 min [2]. We have observed both ECG alterations (ST segment elevations, arrhythmias) and myocardial contractility reduction documented by transesophageal echocardiography (TEE) in 40% of our MIDCAB patients even after very short periods of peripheral coronary artery occlusion. This is in agreement with another group reporting regional systolic dysfunction as measured by TEE in the anterior and apical left ventricular (LV) segments after a 5-min occlusion of the left anterior descending (LAD) coronary artery [3][4].
We have conducted an experimental study to determine the impact of regional myocardial ischemia/reperfusion on LV functional, structural, and metabolic parameters, and to evaluate possible benefits of intracoronary shunt tube insertion during simulated MIDCAB-operations in a porcine working heart model.
|
Materials and methods |
|---|
|
TopAbstractIntroduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
Electrocardiogram electrodes were placed to monitor the heart rate (HR). We placed fluid-filled catheters into the right common carotid artery and the right internal jugular vein for arterial blood pressure monitoring and fluid administration, respectively. Via the right external jugular vein we inserted a 7F Swan-Ganz thermodilution catheter into the pulmonary artery for determination of pulmonary artery (PAP) and central venous pressure (CVP) as well as cardiac output (CO). Following median sternotomy, we prepared the left internal thoracic artery (LITA) for later anastomosis with the left anterior descending (LAD) coronary artery. In ten pigs (`occlusion-group') we placed two occlusion sutures, about 2 cm apart, using 4.0 Prolene (Ethicon®, Germany) around the LAD just distal to the first diagonal branch. In the other ten pigs (`shunt-group') we placed occlusion sutures in the same fashion to create similar experimental conditions, however, did not pull them tight.
LITA-LAD anastomosis in the occlusion-group
To minimize cardiac motion during LITA-LAD anastomosis we used a stabilizer commercially available for minimally invasive coronary artery surgery (Cardio-Thoracic-Systems®, Cupertino, CA). We then snared the occlusion sutures around the LAD, incised the LAD, and performed the LITA-LAD end-to-side anastomosis with 7.0 Prolene® as usual. After 15 min, the distal occlusion suture was removed and blood flow to the distal LAD via the anastomosis was restored. Thereafter, the proximal occlusion suture and the stabilizer were removed.
LITA-LAD anastomosis in the shunt-group
In the shunt-group the same stabilizer was used to minimize LAD movement, however, the LAD occlusion sutures were not snared. The LAD was incised and an appropriately sized intra-coronary silicone shunt tube (Intravascular Arteriotomy Cannula, Medtronic®, USA; length 2 cm, available at different diameters of 1.5, 2.0 and 2.5 mm) was introduced into the LAD to provide coronary blood flow to the myocardium beyond the LAD incision. In all pigs of the shunt-group a 2 mm shunt was used. The LITA-LAD anastomosis was then performed with the shunt in place. After 15 min, prior to the last 2-3 stitches, the flexible shunt was carefully pulled out of the LAD, the anastomosis was completed, and the stabilizer was removed.
Echocardiographic LV function determination
To determine LV function we placed a 5 MHz epimyocardial echocardiography probe (Vingmed®, Sonotron®, Horten, Norway) to provide a LV short-axis image at the midpapillary level. From the echocardiography images we derived the wall motion score index (WMSI) according to Kancz et al. and Prunier et al. [5][6]. Briefly, the midpapillary LV short axis image is divided into eight segments each of which is scored as recommended by the American Society of Echocardiography:
Global LV function (WMSIglobal) was derived as average of all eight LV segments, and regional LV anterior wall function (WMSIAW) was determined as average of the three segments representing the LV anterior wall (antero-septal, anterior, and antero-lateral segments) [5][6]. All echocardiography image analyses were performed by a cardiologist blinded to the procedure (occlusion vs. shunt).
Heat-shock protein-70
To quantitate structural myocardial alterations induced by either technique (occlusion vs. shunt) we used immuno-histochemical methods for measurement of inducible heat-shock protein-70 (HSP-70) expression in LV biopsies. These transmural biopsies were collected from the apex-near LV anterior wall using a 14G biopsy needle (Tru-Cut®, Baxter Healthcare, Deerfield, IL).
Staining
Following cryopreservation using liquid nitrogen, the biopsies were cut in 5 µm slices at -20°C, mounted on coverslips, air-dried, and immunostained as previously described [7]. Briefly, specimens were fixed in methanol at -20°C (30 min) followed by incubation with 0.1% Triton X-100 (20 min) and 0.01 mol/l phosphate-buffered saline (PBS; pH 7.4) +1 mg/ml bovine serum albumin (BSA) (20 min). Sections were then incubated with the primary antibody against the inducible HSP-70 (anti-HSP70i; dilution 1:200; monoclonal mouse antibody C92F3A-5, StressGen, Victoria, Canada) for 1 h. Thereafter, sections were rinsed with 0.01 mol/l PBS, followed by incubation in 0.01 mol/l PBS and 1 mg/ml BSA for 20 min. Subsequently, the secondary antibody (FITC-labeled goat anti-mouse IgG antibody; dilution 1:1000; Fc specific, Sigma, St. Louis, MO) was applied for 1 h and rinsed off using 0.01 mol/l PBS. Finally, the specimens were incubated in 0.01 mol/l PBS for 20 min [7].
Image analysis
Photographs were taken from each specimen at x1000 magnification, and processed using a computerized image analysis system (JAVA®, Jandel Scientific, Erkrath, Germany). After digitalization, the number of fluorescence spots indicating HSP-70 was counted per viewfield and expressed as HSP-70 spots per mm2 myocardial tissue.
Chemicals and solutions
PBS: 8 g NaCl, 0.2 g KCl, 0.2 g KH2PO4, 1.15 g Na2HPO4, adjusted to pH 7.4 with NaOH. BSA: fractionated by cold alcohol precipitation, fraction V 96–99% (gel electophoresis, Sigma, St. Louis, MO).
Experimental protocol
Following preparation we recorded baseline measurements of all hemodynamic parameters, including vascular pressures, cardiac output, epimyocardial echocardiography reading, and transmural LV biopsy collection as described above. Each pig received 5000 IU heparin for systemic anticoagulation. Thereafter, the LITA-LAD anastomosis was constructed during 15 min of either LAD occlusion (occlusion-group) or intracoronary shunt insertion (shunt-group) as described above. Following anastomosis completion and stabilizer removal we repeated measurements of all parameters and a second transmural LV biopsy was collected. At 30 min reperfusion, final measurements of all parameters were taken including a third LV biopsy. Thereafter, the pigs were sacrificed with anesthesia overdose and two myocardial tissue blocks at a size of about 1 cm3 were rapidly excised from the LV anterior wall (LAD perfusion region) and from the LV posterior wall (reference region), respectively. These tissue blocks were freeze-clamped in liquid nitrogen then freeze-dried at -40°C for later quantification of adenine nucleotides (ATP, ADP, AMP) and glycogen content using high performance liquid chromatography (HPLC) and enzymatic assays as previously described [8][9]. At autopsy, patency of the LITA-LAD anastomosis was confirmed in all pigs.
Statistical analysis
All data presented are mean±standard error of the mean (SEM). Data collected over the course of the experiment were analyzed using multivariate analysis of variance (MANOVA) for repeated measures followed by post hoc t-test at a significance level of P<0.05. Type of manipulation was a factor (two steps: shunt or occlusion), while HSP-70-expression was introduced as dependent variable with a repeat factor 3 (baseline, 15 min of manipulation, and 30 min reperfusion). Non-parametric data (i.e. WMSI) were analyzed using the Mann-Whitney U-test. Data on metabolic parameters only collected at the end of the experiment were analyzed using two-tailed t-test for independent samples. A value of P<0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
The quality of the left internal thoracic artery (LITA) graft anastomosis with the LAD performed on the beating heart has been debated especially regarding long-term patency results. To obtain a bloodless field during anastomosis it is usually necessary to occlude the coronary target vessel for a short period of time resulting in regional myocardial ischemia. The severity of ischemia is dependent on the site of occlusion as well as coronary artery morphology and extent of collateral blood flow. Occlusion sutures have to be placed in a `blind fashion' and bear the potential risk of septal coronary artery branch injury. These problems may be overcome by insertion of coronary artery shunt tubes which are assumed to sustain a certain amount of coronary blood flow during anastomosis suturing without blood in the operative field eliminating the need for occluding sutures around the coronary artery [10].
According to our clinical experience we do not agree with the observation that ischemia is never a problem in off-pump cardiac surgery [2]. We observed ECG changes indicative of local ischemia and corresponding hemodynamic compromise during LAD-occlusion in 42 of 100 MIDCAB patients operated during the last two years. In addition, the negative impact of temporary LAD occlusion on LV wall motion has recently been demonstrated in clinical studies [3][4]. Performing the operation under such conditions may have a negative influence on the quality of the anastomosis.
In our experimental model simulating MIDCAB operations in juvenile pigs absent of coronary artery disease and poor collateral coronary circulation we were able to show structural myocardial alteration, depletion of energy stores and functional deficits even after a short period of coronary occlusion and subsequent reperfusion. This `myocardial stunning' was demonstrated by depressed LV anterior wall motion as measured by epimyocardial echocardiography at 30 min of reperfusion following a clinically relevant LAD occlusion time of 15 min. These experimental data are in agreement with recent clinical studies demonstrating that MIDCAB patients exhibit regional LV dysfunction following temporary LAD occlusion as measured by TEE [3][4]. In contrast, our data determined in the shunt-group revealed that 15 min of intra-LAD shunt insertion during LITA-LAD anastomosis construction resulted in normal LV wall motion at 30 min of reperfusion. Thus, intracoronary shunt insertion could possibly prevent regional dysfunction in patients undergoing off-pump revascularization. We are currently investigating this modality.
The expression of the inducible heat-shock protein-70 (HSP-70) is known to be an indicator for the presence of various injury stimuli including stress, heat, and ischemia [11][12]. As both groups in our experiments were exposed to similar stress and experimental time frame, the higher HSP-70 expression in the occlusion group at 15 min of manipulation was probably due to the ischemic injury induced by 15 min of LAD occlusion. However, at 15 min of manipulation HSP-70 expression was also increased in the shunt-group, indicating that some myocardial injury due to ischemia induced by the shunt insertion procedure (about 20–30 s) still occurred in this group. Following reperfusion and thus cessation of the stimulus `ischemia', HSP-70 declined in the occlusion-group. At 30 min of reperfusion, HSP-70 values were similar between both groups. This is in agreement with studies demonstrating that HSP-70 is rapidly expressed (within about 15 min) following injury [13], and that increased HSP-70 levels decline with a half-life of about 2 h after cessation of ischemia [14] which is compatible with the 10–15% decline in HSP-70 we observed during the 30 min reperfusion period in the occlusion-group.
The benefits of intra-LAD shunt insertion vs. LAD occlusion is further demonstrated by measurement of myocardial energy stores (ATP, ADP, AMP and glycogen content) and derived parameters (TAN, ECP) [8][9]. Animals of the occlusion group showed significant depletion of myocardial energy stores at the end of the experiments. In contrast, intra-LAD shunt insertion resulted in significantly better preserved myocardial energy stores. These data corresponded well with our echocardiographic observations and suggest that intracoronary shunt insertion improves myocardial protection during off-pump revascularization.
Critique of the experimental design
A limitation of our study design is that our data were generated in juvenile pig hearts absent of coronary artery obstruction. In patients' hearts with either complete proximal occlusion or subtotal obstruction of a single coronary artery and present collateralization, temporary occlusion of the target vessel probably does not result in significant myocardial ischemia and cardiac dysfunction. However, in patients with moderate single vessel stenosis and poor collateralization, temporary coronary artery occlusion may result in cardiac dysfunction as has been recently demonstrated [4]. In addition, patients with multivessel disease and poor LV function might also not tolerate temporary coronary artery occlusion during off-pump revascularization well. Therefore, we believe that future studies are required to investigate the value of intracoronary shunt insertion in patients undergoing off-pump revascularization.
In conclusion, our data show that in a porcine MIDCAB model 15 min LAD occlusion and 30 min reperfusion causes significant myocardial ischemia associated with structural, metabolic, and functional alterations. In contrast, maintenance of LAD perfusion using intracoronary shunt insertion minimizes ischemia/reperfusion injury and prevents regional LV dysfunction. As our experiments were conducted in healthy pig hearts absent of coronary artery disease, similar results may probably only be expected in humans with moderate coronary artery stenosis and poor collateralization. However, this has to be evaluated in clinical studies. In addition, intracoronary shunt placement does not pose risks to septal coronary artery branches as no occlusion sutures are required, and thus, intracoronary shunts could be a useful tool for myocardial protection during `off-pump revascularization'. Further research is required to investigate the impact of shunt insertion on coronary artery endothelial structure and function.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstractIntroduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
Dr Mehlhorn: I think prior to a broad clinical application, we need to look at the endothelium, what happens at the site of shunt insertion to the coronary artery endothelium, because it might well be that there is some damage to the endothelium, just like with the occlusion sutures, so there might be possible or potential coronary artery stenosis. However, that's our current investigation, so I can't present data on that.
Dr D. Steinbruechel (Aarhus, Denmark): How proximally did you occlude the LAD? And how do you prevent fibrillation of the heart if you are proximally from the first diagonal?
Dr Mehlhorn: To answer the first question, we occluded or we put the shunts versus the occlusion distal to the first diagonal branch. And second, interestingly, we never experienced fibrillation when we occluded the coronary artery or incised it. However, after removal of the CTS stabilizer, we saw ventricular fibrillation in some of the pigs, yes, but that was easily terminated by intrathoracic defibrillation.
Dr K. Leitz (Bremen, Germany): Do you have an idea how much is flowing through your shunt, how much flow?
Dr Mehlhorn: That's what we are currently investigating, depending on different sizes and shunt diameters. However, I think, even if there is only little flow, that's better than no flow. But we will determine numbers and present them.
Dr D. Loisance (Creteil, France): The point is to be very clearly well taken, especially by those who claim that we can occlude the arteries without any problem. I think that you should repeat this presentation at meetings on MIDCAB surgery, and it would give probably a word of caution.
Dr Mehlhorn: Actually, we heard yesterday about shunt insertion in patients and that was beneficial, too. And there is a paper by Malkowski, recently published in JACC, and they showed that if you occlude, you get dysfunction. So it is significant.
Dr A. Haverich (Hannover, Germany): If you repeat these experiments, and also look for endothelial damage maybe, why don't you consider preconditioning of those hearts? Because this has not been included in these experiments, but something we do routinely on a clinical basis, number one. Also, in the clinical setting, we do believe that all the patients are preconditioned because they have ischemia of the anterior wall, so part of your explanation and the results of your experiments may look different if you would have included routine protocol of preconditioning.
Dr Mehlhorn: Yes, but ischemic preconditioning is controversial as well. And in this cited study by Malkowski, they showed exactly that, actually, that if you use a 5-minute ischemic preconditioning protocol prior to the occlusion, you see the same kind of dysfunction at the time of anastomosis construction as you see during the ischemic preconditioning. So I don't know if there is a real benefit in functional terms.
Dr Haverich: Could I have the final question about the heart rate during your occlusion? Because in the clinical setting we try to keep the heart rate real low and also the oxygen consumption. What was the heart rate during your experiments?
Dr Mehlhorn: 100 a minute or so, 100 or 120, but we use the CTS stabilizer there, so heart rate is no problem.
|
References |
|---|
|
TopAbstractIntroduction
Materials and methodsResults
Discussion
Appendix A. Conference...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
) demonstrated significant global LV dysfunction at 30 min reperfusion as indicated by increased WMSIglobal. In contrast, pigs of the Shunt-Group (•) did not show global LV function changes versus baseline. REP, reperfusion. #P<0.05 vs. baseline; *P<0.05 between groups.
) as compared with the Occlusion-Group (