|
|
||||||||
Ann Thorac Surg 1998;66:471-476
© 1998 The Society of Thoracic Surgeons
a Department of Cardiology, Utrecht University Hospital, Utrecht, the Netherlands
Accepted for publication March 24, 1998.
Address reprint requests to Dr Borst, Department of Cardiology, Heart Lung Institute, Utrecht University Hospital, Rm G02.523, PO Box 85500, 3508 GA Utrecht, the Netherlands
e-mail: (exp.cardio{at}hli.azu.nl)
Presented at the Third Utrecht MICABG Workshop, September 27, 1997, Utrecht, the Netherlands.
| Abstract |
|---|
|
|
|---|
Methods. In 18 consecutive pigs, the left internal mammary artery was grafted to the left anterior descending coronary artery, and the arteriotomy was temporarily sealed luminally by a 200-µm-thick polyurethane seal. Endothelial denudation, medial necrosis, and intimal hyperplasia were measured quantitatively and compared with conventionally sutured anastomoses (n = 4 pigs).
Results. Insertion and retrieval of the seal required 28 ± 12 and 11 ± 6 seconds, respectively. Including the arteriotomy, coronary artery occlusion was limited to about 80 seconds. The seal provided a bloodless arteriotomy in all anastomoses with unimpeded coronary artery blood flow. Endothelial denudation was limited to two thirds of the circumference of the coronary artery. No medial necrosis was found. Intimal hyperplasia at the suture line was small, although more pronounced when compared with conventionally sutured anastomoses.
Conclusions. In off-pump, beating-heart coronary artery bypass grafting, the temporary luminal arteriotomy seal provided a bloodless arteriotomy with negligible obstruction to coronary artery blood flow, and with a minimum of arterial wall damage. It is conceivable that this seal may expand the indications for coronary surgical procedures without cardiopulmonary bypass.
| Introduction |
|---|
|
|
|---|
The aim of this study was to assess the feasibility of applying the TLAS during left internal mammary artery (LIMA) bypass grafting to the left anterior descending coronary artery (LAD) on the beating porcine heart, and to compare coronary artery wall injury and healing quantitatively with conventional suturing.
| Material and methods |
|---|
|
|
|---|
Anesthesia
The animals were sedated with azaperone (2.0 mg/kg) and ketamine (1.5 mg/kg), administered by intramuscular injection. An intravenous line was established and each animal received intravenous metomidate (2.0 mg/kg) and atropine (1.0 mg). The pigs were intubated and mechanically ventilated. Anesthesia was maintained by supplying a mixture of oxygen and air (1:1 vol/vol), with 1% halothane added, and an intravenous infusion of midazolam (0.06 mg/kg per hour). Analgesia was obtained by intravenous infusion of sufentanil citrate(0.6 µg/kg per hour). To reduce the mechanical irritability of the heart, each animal received intravenous propranolol (5 mg).
Surgical technique and study protocol
After median sternotomy, the LIMA was harvested in a skeletonized fashion. After heparinization (150 IU/kg), the LIMA was transected, severed distally, and flushed with papaverinesaline solution (0.5%) to prevent spasm. The pericardium was opened and its edges suspended. The middle one third of the LAD, distal to the first diagonal branch (internal diameter, 2.5 mm), was immobilized by two Octopus tentacles (Octopus Tissue Stabilizer, Medtronic; Inc, Minneapolis, MN) [2]. Under the operating microscope (Opmi-6; Zeiss, Oberkochen, Germany), the LAD was dissected free (1) proximally, to apply a microvascular Acland clamp (B-2V; S&T Marketing Ltd, Neuhausen, Switzerland) for temporary interruption of coronary blood flow, (2) at the anastomosis site, and (3) distally, to place a transit time flow probe (2.5S; Transonic, Ithaca, NY) for continuous monitoring of coronary blood flow.
In 18 animals, the LIMA was anastomosed to the LAD using the TLAS (Figs 1, 2). The 200-µm-thick seal was cut from the polyurethane balloon skin of a coronary sinus cannula for retrograde delivery of cardioplegic solution (RC-014; Research Medical, Inc, Midvale, UT) and measured 9 mm in length and 6 mm in width. The seal was slightly modified from the carotid artery seal reported previously [1], in particular with respect to the "shark fin" used for insertion, positioning, and retrieval of the seal (Fig 1). After proximal coronary artery occlusion, the seal was inserted through a 4- to 5-mm longitudinal arteriotomy. After positioning the seal correctly inside the artery, blood flow was restored. At the toe of the anastomosis, the continuous suture loops were loosely applied, creating a funnel-shaped exit in the axial direction. After coronary artery reocclusion, the TLAS was taken hold of at its fin and withdrawn from the artery past the suture, which subsequently was tightened to complete the anastomosis (see Figure 1 in reference 1). The pigs were evaluated at 2 (n = 8), 7 (n = 2), and 28 days (n = 8) after operation.
|
|
All arteries were connected with a continuous 8-0 polypropylene suture (Ethicon, Somerville, NJ). After restoration of LIMA graft flow, the LAD was occluded permanently by a vascular clip proximal to the anastomosis. All 22 anastomoses were performed by one investigator (R.H.H.).
Flow measurements
Phasic and mean blood flow rate of the distal LAD and the LIMA graft during and after the procedure, respectively, were measured with a transit time flow probe connected to a flowmeter (T208) and recorded on an eight-channel recorder, together with the electrocardiogram and intraarterial blood pressure. After clamping the artery for 30 seconds, the coronary hyperemic response was determined as the ratio of the peak mean flow rate divided by the baseline mean flow rate. Baseline mean flow rate and the coronary hyperemic response were measured in duplicate at an interval of 10 minutes in the LAD, before and after insertion of the seal, and in the LIMA graft, postoperatively and at follow-up. Results are presented as the mean of two determinations. Before closure, LIMA graft flow was monitored continuously for 1 hour.
Angiography
Before sacrifice, in all animals the anastomosis was visualized by LIMA angiography using frontal and right anterior oblique projections (C-arm BV27; Philips, Eindhoven, the Netherlands).
Histologic analysis
To preserve anastomotic geometry, all arteries were pressure-fixed at 80 mm Hg as described before [4]. After overnight fixation in formalin 4%, the anastomotic segments were embedded in paraffin and subsequently sectioned in the transverse plane. Sections were stained with hematoxylin and eosin and with elastin van Gieson. Light microscopy was used to identify endothelial injury, medial necrosis, and intimal hyperplasia.
Endothelial injury
At 400x magnification, the absence of endothelial cell nuclei (endothelial denudation) along the circumference of the LAD was determined in hematoxylin and eosinstained cross-sections at 2-mm intervals throughout the anastomotic segment (Fig 3) [1]. The length of denuded areas was divided by the total length of the circumference, for each cross-section, to determine the percentage of the circumference not covered by an endothelial cell layer, irrespective of the diameter after histologic fixation.
|
Intimal hyperplasia
The area enclosed by the luminal border and the internal elastic lamina adjacent to the suture line in LIMA graft and LAD was defined as intimal hyperplasia. Three adjacent elastin-stained transverse sections in the mid segment of the arteriotomy were morphometrically analyzed. Mean values of intimal hyperplasia area (mm2) at the suture line were calculated for each anastomosis.
Statistical analysis
Data are presented as mean ± standard deviation or as median and range. The Students t test (two-tailed) was used to compare mean values. A p value less than 0.05 was regarded as statistically significant.
| Results |
|---|
|
|
|---|
After insertion, instantaneous complete sealing of the arteriotomy was obtained in all anastomoses. Neither suturing nor the sudden increase in blood flow during the coronary hyperemic response resulted in leakage, dislocation, expulsion, or inadvertent transfer downstream of the seal. As a result of the study protocol, the seal was inserted for 52 ± 8 minutes. No intraluminal thrombus formation was observed and no thrombus was found adhering to the seal.
Anastomotic suturing was successfully accomplished in 23 ± 5 minutes (range, 15 to 32 minutes) and required 13 ± 1 bites. An additional stitch to obtain complete hemostasis was required at two anastomosis sites.
Generally, no ventricular arrhythmias other than occasional ventricular premature beats were observed during the operative procedure. In 2 pigs, however, abrupt ventricular fibrillation occurred on opening the pericardium and after a mechanical stimulus during anastomotic suturing, which was successfully managed by electrical defibrillation (30 J).
Conventional anastomoses
After 5 minutes of stop-flow ischemia, the subsequent ischemic period required to perform the conventional anastomosis (n = 4) lasted 11 ± 1 minutes. Anastomotic suturing was accomplished in 10 ± 1 minutes (range, 8 to 11 minutes) and required 13 ± 1 bites.
In 2 of 4 pigs, the prolonged ischemic period during anastomotic suturing was accompanied by frequent ventricular premature beats, leading to ventricular fibrillation in 1 animal. On reperfusion, 3 of 4 pigs experienced a period of ventricular ectopic activity.
Flow measurements
Baseline mean flow rate, 6 ± 2 mL/min, and the coronary hyperemic response, 5.1 ± 0.7, in the LAD was not changed after insertion of the seal, 6 ± 2 mL/min and 5.2 ± 0.9, respectively (Fig 4). Postoperatively, baseline mean flow rate and the coronary hyperemic response in the LAD and LIMA graft was 7 ± 3 mL/min and 4.3 ± 0.9, and 22 ± 9 mL/min and 3.8 ± 1.0, respectively. At follow-up, LIMA graft flow had increased to 36 ± 15 mL/min (p = 0.001). The coronary hyperemic response was unchanged, 3.5 ± 1.1.
|
Continuous observation of LIMA graft flow for 1 hour postoperatively did not reveal any abnormal flow patterns, ie, cyclic flow reductions, indicative of repetitive thrombus formation [5].
Angiography
At sacrifice, all anastomoses were patent, without evidence of luminal stenosis to be attributed to suboptimal surgical technique, intravascular thrombus formation, or excessive anastomotic intimal hyperplasia (Fig 5).
|
|
Intimal hyperplasia
In the 2 animals sacrificed at 7 days, suture line intimal hyperplasia measured 0.07 and 0.06 mm2 (Fig 6). At 28 days, intimal hyperplasia area at the suture line in the TLAS anastomoses, 0.23 ± 0.05 mm2, was small (Fig 7), although more pronounced when compared with conventionally sutured anastomoses, 0.08 and 0.08 mm2.
|
| Comment |
|---|
|
|
|---|
Potential complications of coronary occlusion for off-pump coronary artery bypass grafting
Suboptimal visualization of the distal coronary anastomosis site carries the risk of imperfect anastomotic suturing. To provide a bloodless arteriotomy during off-pump, beating-heart coronary artery bypass grafting, the recipient coronary artery is temporarily occluded proximally and distally [6, 7]. Perforating branches located in the isolated coronary segment, however, may produce retrograde blood flow into the arteriotomy. This blood may obscure the arteriotomy edges and jeopardize the quality of the anastomosis suturing. In addition, interruption of coronary blood flow results in regional myocardial ischemia. In the great majority of patients, the short ischemic period during construction of the graft is remarkably well tolerated without a rise in cardiac troponin levels [8, 9]. This may be attributed to well-developed collateral coronary circulation in patients with high-grade proximal stenosis or occlusion. In a few cases, however, regional myocardial ischemia led to serious ventricular arrhythmias or hemodynamic deterioration, which necessitated conversion to a conventional procedure with cardiopulmonary bypass [10, 11] or insertion of an intraluminal shunt [12, 13]. It is inferred that the preocclusion stenosis severity and the site of coronary artery occlusion play a crucial role in the magnitude and consequence of myocardial ischemia induced by interruption of coronary flow [14]. It is hard to predict, however, which patient will fail to tolerate the 10- to 20-minute interruption of coronary flow.
These two concerns regarding off-pump coronary artery bypass grafting, blood in arteriotomy and regional myocardial ischemia, also apply to coronary surgical procedures with normothermic cardiac arrest and continuous warm blood cardioplegia [15, 16].
Temporary luminal arteriotomy seal
In this study, we extended the experimental evaluation of the recently described briefly occlusive end-to-side anastomosis technique in the porcine carotid artery by means of a TLAS [1]. In contrast to the carotid artery, a longitudinal incision in the pressurized coronary artery widens to an almost round arteriotomy. Therefore, we adapted the seals oval shape (Fig 1). Furthermore, because the polypropylene 9-0 suture loop at the toe-side of the seal, which was umbilical cord originally, broke on retrieval in 2 of 32 cases [1], we replaced the suture loop by a polyurethane dorsal shark fin. A slit in the cannula segment facilitated insertion.
After only a short learning period, insertion and retrieval of the seal required about 20 and 10 seconds, respectively. Coronary artery occlusion time was limited to about 80 seconds. As a consequence, none of the animals experienced serious rate, rhythm, or conduction disturbances during the surgical procedure. In the control group (conventional anastomosis), in contrast, in spite of preconditioning 3 of 4 animals experienced severe ventricular arrhythmias throughout the ischemic period during anastomotic suturing or on reperfusion. Hence, the control group was limited to 4 animals.
In the present study, coronary artery blood flow was temporarily interrupted during insertion and retrieval of the seal by a microvascular clamp proximal to the anastomosis site. Retrograde blood flow, although minimal in the porcine heart, was used to prevent the entrance of air in the coronary artery. It is conceivable that in clinical cases, coronary artery blood flow is only interrupted during insertion of the seal by external pressure proximal to the arteriotomy, whereas on retrieval, both antegrade and retrograde coronary blood flow is used to remove air and flush debris from the anastomosis. As a consequence, dissection of the coronary artery might be required at the anastomosis site only.
In concordance with our previous findings in the porcine carotid artery [1], the intraluminal seal did not impede baseline coronary artery blood flow. In addition, the coronary hyperemic response was unaffected by the seal. No difference was observed between the two types of anastomoses with respect to baseline mean flow rate and the coronary hyperemic response in the LIMA graft, postoperatively and at follow-up. The data were indicative of an adequate anastomotic orifice, which was confirmed by angiography and subsequent histologic analysis. In all animals, LIMA graft flow at follow-up was significantly higher than in the immediate postoperative period. This may be attributed to the administration of propranolol (ß-blockade) before the operative procedure only.
The noted discrepancy in baseline mean flow rate of the distal LAD, 7 ± 3 mL/min, and the LIMA graft, 22 ± 9 mL/min postoperatively, illustrates the large number of perforating side branches located in the short coronary artery segment between the proximal clip and the flow probe distally. Isolating a segment of the recipient coronary artery by temporary occlusion proximal and distal to the anastomosis site [6, 7], or by insertion of an intraluminal tubular shunt [12, 13, 17], blocks antegrade residual or retrograde collateral flow in many of these perforating side branches, which may result in regional myocardial ischemia and its sequelae. The present device, however, contacts only half of the circumference of the coronary artery without blocking the entrance to side branches. In addition, the present device perfectly sealed the arteriotomy, whereas intraluminal tubular shunts [17] may still allow blood to well up retrogradely from perforating branches.
The present study in the porcine coronary artery supports our previous findings in the carotid artery concerning arterial wall damage and subsequent healing [1]. Endothelial damage was limited to the area covered by the seal, and it was less pronounced downstream. Disruption of the internal elastic lamina was never observed. In 1 of 2 animals, endothelial-like cells were already observed at the suture line at 7 days postoperatively. The flimsy seal caused no medial necrosis, and at 28 days intimal hyperplasia at the suture line was small (occupying about 6% of the cross-sectional luminal area of LIMA graft and LAD in the mid segment of the anastomosis), although more pronounced when compared with conventionally sutured anastomoses (approximately 2%). The observed difference in intimal hyperplasia between the two types of anastomoses may be attributed to the study protocol. The LAD in the TLAS anastomosis was exposed to air for about 60 minutes, in contrast to 15 minutes in the conventional anastomosis.
Arteriosclerotic human coronary artery
In the arteriosclerotic human coronary artery, decreased transmural pressure or arteriosclerotic luminal wall surface irregularities may frustrate sealing. However, preliminary observations in variably pressure-perfused postmortem human arteriosclerotic coronary arteries, (3 hearts, 12 coronary arteries, 15 anastomoses) showed that the TLAS sealed adequately.
Conclusion
In off-pump, beating-heart coronary artery bypass grafting, the TLAS enabled bloodless anastomosis suturing without obstruction to coronary artery blood flow and with minimal damage to the arterial wall. It is conceivable that this seal may expand the indications for coronary surgical procedures without cardiopulmonary bypass.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. S. Scheltes, C. J. van Andel, P. V. Pistecky, and C. Borst Coronary anastomotic devices: blood-exposed non-intimal surface and coronary wall stress J. Thorac. Cardiovasc. Surg., July 1, 2003; 126(1): 191 - 199. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Rama, S. Mohammadi, P. Leprince, and I. Gandjbakhch A simple method for heart stabilization during off-pump multi-vessel coronary artery bypass grafting: surgical technique and short term results Eur. J. Cardiothorac. Surg., January 1, 2001; 19(1): 105 - 107. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Bastiaanse, C. Borst, Y. J.M. van der Helm, K. H.H. Loo, and P. F. Grundeman Arteriotomy closure by glued patch in the porcine carotid artery Ann. Thorac. Surg., October 1, 2000; 70(4): 1384 - 1388. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Borst and P. F. Grundeman Minimally Invasive Coronary Artery Bypass Grafting : An Experimental Perspective Circulation, March 23, 1999; 99(11): 1400 - 1403. [Full Text] [PDF] |
||||
![]() |
R. H. Heijmen, C. Borst, C. M. Moues, Y. J.M. van der Helm, P. F. Grundeman, and G. Pasterkamp Temporary luminal arteriotomy seal: III. Postmortem arteriosclerotic human coronary artery Ann. Thorac. Surg., January 1, 1999; 67(1): 120 - 123. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. H. Heijmen, P. Hinchliffe, C. Borst, C. W. J. Verlaan, C. M. Moues, Y. J. M. v. d. Helm, S. Manzo, E. W.L. Jansen, and P. F. Grundeman A NOVEL ONE-SHOT ANASTOMOTIC STAPLER PROTOTYPE FOR CORONARY BYPASS GRAFTING ON THE BEATING HEART: FEASIBILITY IN THE PIG J. Thorac. Cardiovasc. Surg., January 1, 1999; 117(1): 117 - 125. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |