| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;76:805-809
© 2003 The Society of Thoracic Surgeons
a Experimental Cardiology Laboratory, Heart Lung Center Utrecht, University Medical Center Utrecht, Utrecht, The Netherlands
b Department of Design, Engineering and Production, Delft University of Technology, Delft, The Netherlands
Accepted for publication February 5, 2003.
* Address reprint requests to Dr Borst, University Medical Center Utrecht (Room G02.523), Heart Lung Center Utrecht, P.O. Box 85500, 3508 GA Utrecht, The Netherlands.
e-mail: c.borst{at}hli.azu.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: A stretch device that elongates a part of the artery by 80% was implanted in 8 pigs. Elongated coronary arteries (n = 14) and internal mammary arteries (n = 15) were examined histologically at either 2 days (4 pigs) or 5 weeks of follow-up (4 pigs).
RESULTS: No mural thrombus was observed at the elongated site. In the coronary artery at 2 days, few and only minor histologic changes were found. At 5 weeks, in two of seven coronary segments, a thin rim of intimal hyperplasia was found, in one case with a maximum thickness of 76 µm. The internal mammary artery hardly showed any changes.
CONCLUSIONS: Permanent longitudinal elongation by 80% caused little structural changes in the porcine coronary and internal mammary artery wall. Anastomotic connectors that impose relatively large deformations can be safely evaluated in the pig.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Sustained hypertension has been used to study the chronic effects of sustained circumferential stress in the vessel wall [2]. It leads to increased thickness of the medial smooth muscle cell layer due to smooth muscle cell proliferation. Little is known, however, about the wall injury caused by permanent, large mechanical deformation and the subsequent healing response of the vessel wall.
We recently reported the stress-strain relations in and beyond the range of physiologic deformations for the arteries used in the distal anastomotic connector, the coronary and internal mammary arteries [3]. The response of the arterial wall to permanent stretch, however, as might occur in an anastomotic connector, has never been studied in the coronary and internal mammary artery in vivo.
To date, arteries that have been subjected to semichronic longitudinal stretches have only been analyzed in vitro with respect to changes in the expression and activation of key molecules in vascular remodeling [4] and changes in vascular smooth muscle tone [5, 6]. The only in vivo study on longitudinal elongation was performed on rabbit carotid arteries [7].
To evaluate the safety of overstretching the arterial wall in testing coronary anastomotic connectors in the pig, we developed a stretch device that could apply a fixed longitudinal stretch on small arteries in vivo of up to 100%. In this study, we have analyzed the microscopic wall response to 80% longitudinal elongation of the porcine coronary and internal mammary arteries.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Anesthesia
Anesthesia was induced by ketamine (10 mg/kg) intramuscularly, thiopental sodium (4 mg/kg), atropine (1 mg), and the antibiotic amoxicillin (500 mg) intravenously. A mixture of oxygen and air (1:1 vol/vol) with 0.5% to 1% halothane, midazolam (0.3 mg/kg per hour) intravenously, and propranolol (5 mg) were administered. Analgesia was obtained by sufentanil citrate (1 µg/kg per hour) and muscle relaxation by pancuronium (0.1 mg/kg per hour).
Postoperatively, amoxicillin trihydrate (15 mg/kg) and buprenorphine (0.6 mg) as analgesic were administered intramuscularly for 2 or 3 days. Animals were sacrificed after 2 days or 5 weeks with pentobarbitalnatrium (200 mg/kg) intravenously.
Application stretch device
A stretch device was developed to apply a fixed longitudinal stretch on the coronary and mammary arterial walls in vivo (Fig 1). After it appeared in pilot experiments that in vivo circumferential stretch could not be applied without creating a hemodynamically significant stenosis, or without direct contact between the arterial wall and the device, longitudinal stretch was evaluated only. The stretch device consisted of two separate bows, which each have four needles (diameter, 0.1 mm) pointing in the longitudinal direction of the artery. First, the needles of one bow (Fig 1A) were inserted into the arterial wall intramurally. Second, the needles of the other bow (Fig 1B) were inserted in the arterial wall with the needles pointing in the opposite direction. The tips of the needles were first inserted at an angle, approximately 3 to 5 mm away from the first bow. Subsequently, through a combination of tilting, turning the needles to the horizontal plane, and sliding them in the wall, the second bow was fixed. Figure 2
shows the position of the needles in the arterial wall after insertion of the bows. The exact distance between the bows was measured, and the new distance, necessary for the desired elongation, was calculated. Pilot experiments showed that 40% and 60% elongation of the arterial wall did not result in any microscopic damage to the wall. On the other hand, 100% elongation caused leakage from the needle insertion sites where tearing of the vessel wall created holes. Therefore, we chose to apply an elongation of 80%.
|
|
All animals received two stretch devices on the left anterior descending coronary artery (LAD) and two on the left internal mammary artery (LIMA). Four pigs had the devices implanted for 5 weeks and the other 4 had them implanted for 2 days. The device was implanted on the near side of the arterial wall. The opposite side, which was not stretched, served as control. In this way, both the stretched and the control parts of the arterial wall could be analyzed in one histologic cross-section. The stretched near wall was recognized in the histologic cross-sections by the damaged adventitia and by the repair tissue that surrounded the screw (Fig 3A).
|
At 2-day follow-up, the intima was inspected for intraluminal and mural thrombus formation. When the endothelium was focally absent, endothelium was scored as absent. Furthermore, leukocyte adhesion and fractures in the internal elastic lamina were noted as present or absent. At 5 weeks, the presence or absence of intimal hyperplasia was noted and its maximal thickness was measured as the maximal distance between the luminal border and the internal elastic lamina.
The media was checked for the presence or absence of pyknotic nuclei, cell necrosis (areas without smooth muscle cell nuclei), and infiltration of inflammatory cells. When fields of smooth muscle cell nuclei had conspicuously changed in their orientation compared with the control side, this was considered to be a positive observation of smooth muscle cell nuclei orientation change. The alignment of elastin layers was studied. Alteration with respect to the control side in waviness or number of elastin layers as well as the distance between the layers was scored as "changed elastin structure." The presence or absence of a disrupted or dissected external elastic lamina was also scored.
Overall, in the hematoxilin-eosin–stained cross-sections, the histologic changes in the arterial wall due to stretch were considered to be minor and impossible to quantify other than by dichotomous score: present or absent. Two observers evaluated each histologic section simultaneously under the (teaching) microscope until consensus opinion was reached about the presence or absence of the abnormalities delineated above and listed in Table 1.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
The changes in the intima and media at 2 days and 5 weeks of the elongated near wall compared with the unstretched far wall are listed in Table 1. No mural thrombus was observed in any of the elongated segments. The differences between the stretched and control segment of the arterial wall proved to be minor; therefore, changes compared with control were scored as either present or absent. At 2-day follow-up, limited changes in the intima and adventitia were found in the stretched wall of the LAD. At 5-week follow-up, only few and insignificant alterations were observed. The maximum intimal hyperplasia found (76 µm thick, ie, 2.6% of the arterial diameter based on the internal elastic lamina) is illustrated in Figure 3. An intimal hyperplasia rim of this thickness (Fig 3A) is unlikely to have hemodynamic consequences of importance. The stretched IMA wall showed virtually no microscopic changes due to 80% elongation, both at 2-day and 5-week follow-up.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
In the porcine model, a 5-week follow-up period is considered by Fischel and Virmani [8] to be equivalent to approximately 15 to 30 weeks of wall healing in the human coronary artery. A previous study in our laboratory [9] showed completed reendothelialization 5 weeks after coronary connector implantation and streamlining repair tissue. Several weeks later, no further changes in healing response were observed (unpublished results). In the pig model, a 5-week follow-up period provides solid experimental evidence of the (intermediate) healing response.
In pilot experiments, we learned that stretch by less than or equal to 60% did not induce any histologic changes in the arterial wall. Stretch by 100% resulted in leakage from the needle insertion points. Therefore, we focused on 80% stretch, a value that does have a bearing on clinically applied connectors. For example, the GraftConnector [10] requires the IMA to be everted over a Nitinol ring of 2 to 2.5 mm diameter. The stretch at the everted inner wall can be estimated as follows:
 |
When this is applied to human IMAs of persons over 40 years of age where the wall-thickness has increased [11], the stretch at the inner wall may be as large as 75%.
Another example is the St. Jude Medical stainless steel connector [12], where the coronary arteriotomy is created by inserting a small needle in the arterial wall. The arteriotomy is then dilated to a diameter of 2 mm. The dilated segment maybe stretched by up to 100%.
Longitudinal stretching of coronary arteries has been reported by Frobert and associates [5, 6], but their experiments were performed in vitro. The applied stretch was small, a maximum of 10% on top of the in situ length. They found that varying the degree of longitudinal stretch plays a significant role for coronary artery constriction at extreme values of transmural pressure: at low pressures, stretch increases constriction, whereas at high pressures, stretch decreases constriction. Furthermore, they reported a decreased sensitivity of the response to nifedipine of the excised porcine coronary artery. Clearly, axial stretching affects the vascular smooth muscle tone. They did not report on structural changes, but, given the results in our study, these small stretches probably did not cause any structural damages.
The one earlier in vivo study on arterial stretching was reported by Jackson and associates [7]. They stretched rabbit carotid arteries in vivo by surgically removing a part of the artery. This increased the longitudinal stretch from a normal 62% ± 2% to 97% ± 2% compared with the ex vivo length. Remodeling of the arteries, however, caused the axial strain to be normalized within 1 week. Endothelial and medial cell replication rate, elastin and collagen content, vessel wall thickness, and circumference were all significantly increased in the stretched artery. Also, increased rates of apoptosis were found. In our experiments, the reduction of the tension applied by the stretch device observed at sacrifice is also likely to be due to compensatory remodeling of the vessel wall. The lengthening of the artery, however, was more controlled in our device because the stretched part was small and fixed in the device and therefore not influenced by the proximal or distal part of the artery.
Limitations of the study
To discriminate between an inflammatory reaction to an implanted foreign body and the response to wall stretch, it would have been standard procedure to include in the study implantation of the device without imposing the 80% stretch. There were four reasons, however, for not including the control implantation of the stretch device. First, the stretch device had been designed to avoid any contact between the screw body and the stretched artery segment (distance, 0.5 to 0.8 mm).
Second, the investigated arterial cross-section was taken from the middle of the stretched segment, at least 1.8 mm removed from the inserted needles that served as retracting claws. As a result, we anticipated that in this cross-section, a foreign body (claws and screw) inflammatory response would be minor. Third, connector-induced overstretching is likely to involve a foreign body reaction to the connector material too.
Fourth, in four pilot experiments with 40% and 60% stretch, no effects on the arterial wall were observed at all at 4 weeks. In particular, no intimal hyperplasia was found. The damaged adventitia at 5 weeks is attributed to the implantation procedure, not to stretch or an inflammatory response to a foreign body. Before implantation of the stretch device, we removed all periadventitial loose connective tissue, similar to a conventional coronary bypass procedure or a coronary connector procedure. We attribute the intimal hyperplasia at 5 weeks to the 80% stretch, not to an inflammatory response, because it was absent in the four pilot experiments with 40% and 60% stretch, and because it was also absent in five of seven LADs and seven of seven IMAs stretched by 80%.
The response of normal arteries to stretch is likely to be different from the response of atherosclerotic arteries. The composition of the atherosclerotic wall and its elasticity are different. It is conceivable that in similar experiments in diet pigs, more pronounced wall injury would be elicited.
In conclusion, 80% elongation of the healthy porcine coronary and internal mammary arteries in vivo caused little structural changes in the arterial wall. Anastomotic connectors that impose this amount of stretch to the porcine arterial wall are not expected to give rise to any mural thrombus or to intimal hyperplasia of any importance. The response of the human artery to such connectors remains to be determined.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
