| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:1295-1298
© 2004 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Medical University of South Carolina, Charleston, South Carolina, USA
Accepted for publication January 14, 2004.
* Address reprint requests to Dr Kratz, Department of Surgery, Room 409, CSB, Medical University of South Carolina, 171 Ashley Ave, Charleston, SC 29425-2279, USA
kratzj{at}musc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Fifty-eight patients were randomized into three groups to compare papaverine's ability to prevent spasm and to treat established spasm. Group 1 was control and no treatment was employed. In group 2, papaverine was injected with a blunt needle through the endothoracic fascia parallel to the mammary artery before harvest to assess spasm prevention. In group 3, papaverine was delivered perivascular in an identical manner to group 2 but after the mammary artery was dissected from the chest wall. This group was an evaluation of spasm treatment. Drug dosage was the same for both groups and routine bypass grafting was performed. Before anastomosing the mammary artery to the left anterior descending artery, blood flow was recorded for 15 seconds and flow per minute calculated. Cardiopulmonary bypass pressures were maintained at 70 mm Hg during collection.
RESULTS: Mean blood flows were: group 1 = 86.2 mL/min, group 2 = 122.5 mL/min, and group 3 = 139.7 mL/min. Left internal mammary artery flow in group 3 was statistically different from control (p = 0.0457). Group 2 flow approached but did not reach statistical significance (p = 0.0874). Mammary artery dissection times for the three groups were not different.
CONCLUSIONS: Papaverine delivery to the left internal mammary artery after dissection treats spasm effectively, improves blood flow at the time of its anastamosis to the left anterior descending artery, and avoids any risk of intimal injury. Injection of papaverine before mammary artery harvest does not shorten dissection time, and flow is not statistically improved.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In our institution we routinely use papaverine to both prevent and treat LIMA spasm. Injection of papaverine beneath the endothoracic fascia before LIMA dissection may inhibit spasm and the additional hydrodissection may make harvest quicker and easier. This technique was evaluated and compared to papaverine's ability to treat spasm by delivering the drug after LIMA harvest.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Patients were randomized into 3 categories: group 1, no treatment (n = 20); group 2, papaverine injection of the LIMA pedicle before dissection (n = 18); and group 3, papaverine injection of the LIMA pedicle after harvest (n = 20). Routine sternotomy was done followed by harvest of the LIMA using a wide pedicle from the level of the subclavian vein to beyond the vessel's bifurcation. Dissection was performed by four cardiothoracic residents done with low power electrocautery and major branches were controlled with hemoclips. Heparin was then given systemically (400 units/kg); and after a 3-minute delay, the LIMA was transected beyond its bifurcation and occluded with hemoclips. In groups 2 and 3, 10 mg of room temperature papaverine (1 mg/mL 0.9% saline) was injected periarterially throughout the vessel using an olive-tipped needle (Number 31001; Medtronic, Minneapolis, MN) with care to avoid vascular wall infiltration. The only difference was that in the chest wall group (group 2), papaverine was administered parallel to the vessel before LIMA harvest, whereas in the pedicle group (group 3), papaverine was delivered in an identical manner but after the LIMA was dissected and occluded with hemoclips. The mammary artery was placed in the left hemithorax to await anastomosis to the LAD in all cases, this being a single anastomosis and the last coronary anastamosis performed. Routine CABG was done under mild hypothermia (32°C) with both antegrade and retrograde blood cardioplegia. No topical epicardial cooling is used at our institution.
Immediately before the LIMA-LAD anastomosis, the mammary artery was transected 2- to 3-cm proximal to its bifurcation and blood flow was collected over a 15-second period. Flow per minute was then calculated. Cardiopulmonary bypass pressures were maintained at 70 mm Hg during flow measurements. To ensure uniformity, all distal vein graft anastomoses and LIMA flow measurements were done before any mitral valve procedures were preformed. Aortic cross clamp time (XCT), cardiopulmonary bypass time (CPB), and LIMA dissection times differed slightly from patient to patient, but time differences between the groups were not statistically significant. The mean number of bypass grafts performed was 3.2, which was also not different for the three groups. Because papaverine was introduced to the mammary artery before dissection in group 2, LIMA exposure to the drug in that group was greater by 26 minutes than in group 3, the mean time required for group 2 mammary artery harvest.
Analysis of variance was used to detect differences among the three groups. The Kruskal-Wallis one way analysis of variance on ranks was performed to analyze the number of CABG grafts done as this was not a continuous variable and the Bonferroni multiple comparison test was used for LIMA flow evaluation. All statistical analyses were performed with NCSS 2000 for Windows (Number Cruncher Statistical Systems, Kaysville, UT). All p values < 0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Whether or not intraluminal administration is the best method for treating LIMA spasm, it exposes the vessel to potential injury. Dregelid and associates [13, 14] sectioned and histologically analyzed discarded LIMA segments after intraluminal delivery of papaverine, and found 8 of 52 vessels (15%) had intimal or medial disruption. This occurred in the absence of hydrostatic dilation of the artery. Despite intimal injury, they could not intraoperatively identify which vessels were injured when blood flow was measured [14]. In another study, intimal and internal elastic lamina disruption occurred in 17% to 35% of arterial segments depending on where along the distal LIMA sections were obtained. Injury occurred despite keeping papaverine administration below intraluminal pressures of 50 mm Hg, perceived to be the threshold above which vessel trauma occurs. The effect of intraluminal delivery of papaverine on long-term LIMA patency is not known, but injury is speculated to contribute to intimal hyperplasia chronically [16]. Using a blunt needle, drug injection in the pedicle avoids the intima of the artery and the injury potential of intraluminal delivery to this layer. Also, pedicled delivery probably allows for better availability of papaverine to the LIMA over topical therapy. There is evidence that adventitial exposure prolongs the duration of the drug's vasodilatory effect. Chester and coworkers [17] found, in vitro, that inhibition of potassium induced contraction was extended at least 3 hours, from 2 to 5, if extraluminal papaverine exposure was added to intraluminal exposure.
Mulay and coworkers [18] described a method of delivering papaverine to the mammary artery bed before LIMA dissection, making harvest easier. For years, we have used this technique, but to our knowledge, no one has formally analyzed this method's ability to prevent LIMA spasm, whether it eases dissection, and how it ultimately influences vessel flow. LIMA spasm is complex and not fully understood, but surgical dissection certainly contributes to it. In administering papaverine before harvest, we were able to evaluate papaverine's ability to prevent LIMA spasm (group 2) and contrast this to papaverine's capacity to reverse already established spasm (group 3). In our study, LIMA flow in the chest wall group (group 2) was not statistically different from control. In vitro data suggests that papaverine is potentially a good agent at preventing LIMA spasm [8]. In addition, with the delivery of papaverine before dissection, drug exposure to the LIMA was increased by a mean of 26 minutes from that delivered to the pedicle after harvest. This intituitively could produce greater vasodilation over later drug delivery and hence greater LIMA flow, but our results do not support this. Even in the presence of longer drug exposure, flow in group 2 was no different from control suggesting papaverine may be more effective in the treatment of established spasm rather than in its prevention. We also found LIMA dissection times to be no shorter when papaverine was delivered before harvest. In fact, we discovered that both internal mammary veins were more prominent creating greater problems with hemostasis during harvest, blunting any theoretical advantage.
Without pharmacologic manipulation, will the LIMA spontaneously vasodilate after harvest if given sufficient time? Of the studies evaluating this, results are conflicting. In three reports, LIMA flow was measured before the onset of cardiopulmonary bypass (CPB), mean interval times between a first and second set of flow measurements ranging from 12 to 38 minutes [10, 11, 19]. Cooper and colleagues [10] and Bilgen and coworkers [19] demonstrated spontaneous relaxation does not occur between LIMA harvest and the start of cardiopulmonary bypass. Sasson and associates [11] discovered LIMA flow increased spontaneously before the start of CPB when there was at least a 30-minute interval between two sets of flow measurements. Vilandt and colleagues [20] recorded vessel flow later on bypass before anastomosing the LIMA to LAD, similar to our study, and found the LIMA spontaneously dilated. Mean interval between flow measurements reported by them was 75 minutes allowing a greater opportunity for natural relaxation [20]. Papaverine dilates smooth muscle slowly, slower than nitroglycerine, but more quickly than calcium-channel-blockers [7, 10, 12], although the precise time intervals are not fully appreciated. We recorded LIMA flow late in a CABG operation to factor spontaneous dilation (control) and to allow maximal exposure of the LIMA to papaverine in the experimental groups. Blood flow through the LIMA on CPB is not physiologic, but this is the moment when surgeons decide to use the LIMA in situ, as a free graft, or to discard it in favor of another conduit based in part on its flow quality.
We traditionally have used 10 mg of papaverine to prevent or treat LIMA spasm in a concentration of 1 mg/ml and all flow measurements were made after this dosage. Papaverine dosage in the literature has varied widely up to a maximum of 50 mg [15] and concentrations have ranged from 0.3 mg/mL to 1.5 mg/mL, making it somewhat difficult to compare study results. However, Dregelid and associates [13, 14], in two similar reports, found that increasing the concentration of topical papaverine from 0.8 mg/mL to 1.5 mg/mL did not increase LIMA flow, suggesting that concentrations in this range may have similar effects.
Several limitations are present in our study. First, the size of the study is small. Larger group sizes may have increased the statistical power of each group, in particular group 2 results. Accrual, with any study, was difficult as some patients were emotionally uncomfortable with randomization. We arbitrarily closed the study after 59 patients had been entered. A second limitation: our patient population was not uniformally similar. Six operations included were redo CABG procedures. Fibrotic bands around the LIMA can develop in response to primary operations, potentially compromising LIMA flow and our results. Subanalysis of the six redo cases showed vessel flow was statistically the same as their respective group. There were also three mitral valve procedures performed. Two were repairs done after revascularization to correct persistent insufficiency, these being done during a second cross clamp period after distal vein graft anastamosis and LIMA flow was recorded. The other mitral valve procedure was a patient who received a mechanical valve ten years ago and in whom one annular suture dehisced. Again, distal vein graft anastomoses were performed and LIMA flow measured before the prosthetic mitral valve was repaired establishing a uniform time period between papaverine administration and LIMA flow measurement. Subanalysis of flow rates in these three patients was statistically the same as their respective group.
We did not directly measure time intervals between papaverine administration and flow calculation in our study, but indirect measurements (XCT, CPB times) for these three valve patients were the same as their respective group suggesting that time delays were not present. Lastly, we used an ambient temperature solution of papaverine for injection. When papaverine is delivered to the LIMA before harvest, as reported by Mulay and colleagues [18], it is done with a normothermic solution. Normothermic papaverine has been revealed to be superior to room temperature solution when applied topically [18], the reasons have been speculated [7, 20]. Papaverine inhibits intracellular phosphodiesterase, increasing cyclic AMP and phosphodiesterase activity is greater at normothermia. Some have commented that the relative hypothermia of an ambient solution may be vasospastic, potentially blunting the full vasodilatory properties of papaverine. If this is the case, LIMA flow might have been improved toward statistical significance in group 2 and a corresponding increase in group 3 with normothermic papaverine, but this would require additional investigation.
Papaverine delivery to the pedicle after LIMA dissection treats spasm effectively, avoids any risk of intimal injury, and is quick and simple. Administering papaverine before LIMA harvest does not make dissection easier or faster, and vessel flow was not statistically improved.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Bahcivan, F. Kolbakir, H. Karamustafa, and H T. Keceligil Endothoracic Papaverine Application for Internal Thoracic Artery Harvest Asian Cardiovasc Thorac Ann, June 1, 2007; 15(3): 234 - 237. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Koramaz, M. Ozkan, G. Altun, K. Y. Guven, M. K. Duman, N. I. Kalyoncu, E. Yaris, and F. Ozcan Effects of papaverine and carbon dioxide alone or in combination on the blood flow of internal thoracic artery. J. Thorac. Cardiovasc. Surg., November 1, 2006; 132(5): 1126 - 1130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. P. Lima Canadas, A. C. Canas, J. L. Orradre Romeo, C. I. Rubio Martinez, L. F. Lopez Almodovar, and M. Calleja Hernandez Endothelium histological integrity after skeletonized dissection of the left internal mammary artery with ultrasonic scalpel Interactive CardioVascular and Thoracic Surgery, June 1, 2005; 4(3): 160 - 162. [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 |
