| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2001;122:470-475
© 2001 The American Association for Thoracic Surgery
Cardiopulmonary Support and Physiology (CPS) |
From the Department of Cardiothoracic Surgery,a Royal Brompton and Harefield NHS Trust, and Department of Cardiac Medicine,b National Heart and Lung Institute at the Imperial College of Science, Technology and Medicine, London, United Kingdom.
Supported by the Kate Weeks Research Fellowship, Royal College of Surgeons, England, and the Clinical Research Committee Fellowship, Royal Brompton and Harefield NHS Trust.
Received for publication Sept 29, 2000. Revisions requested Dec 15, 2000; revisions received March 20, 2001. Accepted for publication March 22, 2001. Address for reprints: Peter Collins, MD, Department of Cardiac Medicine, National Heart and Lung Institute at Imperial College of Science, Technology and Medicine, Dovehouse St, London SW3 6LY (E-mail: peter.collins{at}ic.ac.uk).
Abstract
Objectives: Surgical preparation of coronary conduits for coronary artery bypass grafting may affect their early and long-term patency; one mechanism may involve endothelial damage. We investigated the effect of 3 commonly used solutions—Ringer's solution, normal saline solution, and heparinized whole blood—on in vitro endothelial and contractile functions of the human radial artery.
Methods: Radial artery segments were harvested, cut into 3-mm rings, and stored in unoxygenated Ringer's solution, normal saline solution, or heparinized whole blood for 45 minutes. Rings stored in Krebs solution were used as controls. The rings were then mounted and stretched to an optimal resting tension in oxygenated Krebs solution at 37°C. Contraction responses to potassium, norepinephrine, and serotonin and relaxation responses to acetylcholine, verapamil, and nitroprusside were evaluated.
Results: Fifty-six radial artery ring segments from 14 patients (n = 7 rings for each contaction-relaxation curve) were studied. Equilibrated resting tension was 9.6 ± 0.3 mN (5.9 ± 0.2 g), and resting internal circumference was 6.4 ± 0.2 mm. Absolute maximum contraction to potassium was significantly less in rings stored in normal saline solution than in rings stored in control solution (10.7 ± 0.6 g vs 14.5 ± 0.6 g, P < .01; 95% confidence intervals, 0.9-6.9). There was no difference in the contraction to norepinephrine (P = .11) and serotonin (P = .25) among the 3 solutions compared with the control solution. Rings stored in heparinized whole blood had significantly greater endothelium-dependent relaxation to acetylcholine (P < .007), whereas those stored in normal saline solution had reduced responses. Endothelium-independent relaxation to verapamil and nitroprusside were similar among the 3 solutions.
Conclusion: Heparinized whole blood is a better physiologic medium for preservation of radial artery endothelial and contractile functions during storage before grafting.
Since the reintroduction of the radial artery (RA) as an alternative arterial conduit in coronary artery bypass grafting (CABG) operations by Acar and colleagues,
1 it has been increasingly used. The improved results achieved by Acar and colleagues may have been due to a combination of careful harvesting with avoidance of mechanical trauma and the use of calcium channel blockers given after CABG to prevent graft vasospasm. Despite these changes in surgical practice, a 4% incidence of vasospasm is still reported in RA grafts, which therefore represents an important and unresolved issue. 2 Histologic studies have shown the RA to be a class III muscular artery with a greater propensity for vasospasm than any other arterial grafts that have been used previously. 3
Damage as a result of surgical harvesting and preparation of conduits for CABG can adversely affect endothelial function, which may result in graft vasoconstriction and myointimal hyperplasia, leading to early graft failure. The type of solution used to store saphenous vein conduits before grafting has been shown to affect smooth muscle contraction and relaxation. 4-7 Nevertheless, the type of solution used for storing RA grafts varies from center to center. The issue of the optimal solution for storing or rinsing RA grafts before anastomosis has not been investigated. We therefore investigated the 3 commonly used storage solutions—Ringer's solution, normal saline solution, and heparinized whole blood—and measured their effects on in vitro endothelial function and contraction of the human RA.
Materials and methods
Preparation of RA segments
Human RA segments were obtained from patients undergoing CABG operations with the RA as a conduit. These were carefully harvested, and side branches were ligated with Ligaclip devices (Ethicon Endo-Surgery, Inc, Somerville, NJ), as previously described by Reyes and colleagues. 8 The distal 1.5- to 2-cm segment of the artery that was not exposed to verapamil rinsing solution (2 mg/100 mL of Ringer's solution) immediately on removal from the forearm was collected for the in vitro vasoreactivity studies. Approval to use the discarded distal segments of the RA was approved by the Royal Brompton Hospital Ethics Committee.
RA segments were immediately transferred to the laboratory in cold (4°C) oxygenated Krebs solution: Na+, 144 mmol/L; K+, 5.9 mmol/L; Ca2+, 2.5 mmol/L; Mg2+, 1.2 mmol/L; Cl–, 128.7 mmol/L; HCO3–, 25 mmol/L; SO42–, 1.2 mmol/L; H2PO4–, 1.2 mmol/L; and glucose, 11 mmol/L. The RA segments were dissected free of surrounding satellite veins and connective tissue and cut into 3-mm rings. A set of 4 rings was obtained from each artery, with the resection margins discarded.
Protocol
Each RA ring was stored in 10 mL of either unoxygenated Ringer's solution, normal saline solution, or heparinized whole blood (individual patient's blood containing 100 IU of heparin per kilogram of patient weight obtained from the arterial circuit of the cardiopulmonary bypass pump) at constant room temperature (20°C-22°C) for 45 minutes. Rings stored in Krebs solution were used as controls for comparison.
Organ bath procedure
According to storage protocol, rings were mounted on specially designed wire hooks and attached to a strain gauge in 10-mL jacketed organ baths containing oxygenated Krebs solution at 37°C. The RA rings were equilibrated to resting tension by the length-tension curves, as described previously. 9 In brief, RA rings were stretched in progressive steps of increasing tension to determine the associated internal circumference changes. These parameters were plotted to determine the length-tension curve for each ring (Microsoft Excel; Microsoft Corporation, Redmond, Wash). The stretch procedure was stopped once the exponential length-tension curve intersected the linear length-tension curve, a point at which the internal circumference of the ring equates to a transmural pressure of 100 mm Hg. Each ring was then set at the resting tension equivalent to that required to stretch the ring by 90% of its internal circumference when distended with a transmural pressure of 100 mm Hg.
After this normalization procedure, the RA rings were equilibrated for 1 hour. Each ring was precontracted with 100 mmol of potassium before the vasoreactivity study. The organ bath Krebs solution was changed every 20 minutes. Between each contraction or relaxation study, the rings were rinsed regularly at 20-minute intervals and allowed to equilibrate to baseline resting tension. Each arterial ring was subjected to cumulative doses of 1 of the 3 contractile and relaxation agents in random order.
Contraction
Cumulative contraction curves to potassium (10-80 mmol/L), norepinephrine (10–8-10–4.5 mol/L), and serotonin (10–8-10–4.5 mol/L) were performed to assess voltage-mediated and receptor-mediated contraction. Median effective concentrations (concentration at 50% of maximum contraction [EC50]) were calculated.
Relaxation
Endothelium-dependent relaxation to acetylcholine (10–8-10–4.5 mol/L) and to the calcium channel blocker verapamil (10–8-10–4.5 mol/L) and endothelium-independent relaxation to nitroprusside (10–10-10–6.5 mol/L) were measured. Relaxation was expressed as the percentage of relaxation of the precontraction with potassium (EC50). Cumulative relaxation curves were performed, and median effective concentration EC50 values were also calculated. All drugs were obtained from Sigma-Aldrich Company Ltd, Poole, Dorset, United Kingdom.
Data analysis
Results were expressed as means ± standard error of the mean. Absolute maximum contraction and cumulative relaxation achieved by rings stored in the 3 types of storage solutions and those stored in the control solution were analyzed by 1-way analysis of variance (Primer of Biostatistics 4.0, McGraw-Hill). Post hoc analysis was performed by the Student-Newman-Keuls test for multiple comparisons.
Results
Fifty-six RA ring segments from 14 patients were studied (n = 7 for each experiment). Mean equilibrated resting tension was 9.6 ± 0.3 mN (5.9 ± 0.2 g), and mean resting internal circumference was 6.4 ± 0.2 mm.
Voltage-dependent contraction to KCl
Data on voltage-dependent contraction to KCl are shown inFigure 1, A. There was a significant difference in the absolute maximum contraction measured in the groups (P = .002), with rings stored in normal saline solution achieving significantly lower absolute maximum contraction when compared with control rings and rings in Ringer's solution (10.7 ± 0.6 g vs 14.5 ± 0.6 g and 15.7 ± 1.1 g, respectively; P < .05). There was no difference between rings stored in Ringer's solution and heparinized whole blood and rings stored in control solution (15.7 ± 1.1 g vs 13.0 ± 1.0 g vs 14.5 ± 0.6 g, respectively; P > .05). There was also no difference in the potassium EC50 values between the tested storage solutions and control solution(Table 1).
|
|
, B; P = .11) or serotonin(Figure 1, C; P = .25). There was also no difference in the EC50 values between the tested storage solutions and the control solution(Table 1).
Endothelium-dependent relaxation
Data on endothelium-dependent relaxation are shown inFigure 2, A. A functional endothelium was confirmed by relaxation to acetylcholine in all rings, ranging from 31% (normal saline solution) to 82% (heparinized whole blood). RA rings stored in heparinized whole blood had significantly greater relaxation to acetylcholine than did those stored in control and normal saline solution at a concentration of 10–7 to 10–4.5 mol/L (P < .01) and Ringer's solution at 10–6 mol/L. Rings stored in normal saline solution consistently achieved the lowest percentage of endothelium-dependent relaxation to acetylcholine. Sensitivities, as indicated by the EC50 values for acetylcholine, were not significantly different between the solutions investigated(Table 1).
|
, B. Maximum relaxation to verapamil ranged from 69% (normal saline solution) to 80% (Ringer's solution), with no significant difference among the groups (P > .2). EC50 values for verapamil were similar among the tested storage solutions(Table 1).
Endothelium-independent relaxation
Data on endothelium-independent relaxation are shown inFigure 2, C. Maximum endothelium-independent relaxation achieved with nitroprusside ranged from 83% (control) to 98% (heparinized whole blood). There was no significant difference among the groups (P > .2). However, relaxation was significantly greater than that achieved with acetylcholine. There was no difference in EC50 values to nitroprusside among the tested storage solutions(Table 1).
Discussion
There is a degree of variability in the types of solution used for storing conduits before bypass grafting. The potential grafts are commonly stored for 45 minutes or more. This can disrupt the vascular endothelium and the integrity of the media in saphenous vein grafts. 4-7,10 Our study has demonstrated that a storage duration of 45 minutes in the 3 tested storage solutions does not significantly affect receptor-mediated contraction or their sensitivity, as indicated by EC50 to norepinephrine or serotonin when compared with control rings in Krebs solution. However, storing in normal saline solution does significantly attenuate contraction to potassium.
Tatoulis and colleagues 11 recently reported a significant increase in sensitivity and magnitude of contraction of RA stored in heparinized whole blood compared with RA stored in Krebs solution to various agonists. This is different from our findings, which could be due to the fact that the above investigators compared freshly harvested distal segments of RA with surgically prepared proximal segments that were stored in heparinized whole blood containing papaverine until grafting. Chester and colleagues 12 demonstrated a significant difference in contractile properties between the proximal and distal segments of the RA, with the former exhibiting significantly greater contractility than the distal segments. Thus, the increase in contractility found by Tatoulis and colleagues 11 could be attributed to the greater contractile properties of the proximal segment rather than the effect of storage in heparinized whole blood.
RA rings stored in Ringer's solution and heparinized whole blood contracted to an equivalent degree to control rings. This may indicate a preservation of the contractile properties of the RA rings stored in the above solutions, unlike those stored in normal saline solution, with which there is an attenuation of contraction. The success of the pedicled left internal thoracic artery has been attributed to its ability to adapt to changes in myocardial demand by contracting or dilating in response to local stimuli, indicating the preservation of its vasomotor properties. 13 Thus, the preservation of RA contractile properties by Ringer's solution and heparinized whole blood during storage may be important in determining the early and long-term function and patency of this conduit.
All 3 storage solutions tested for 45 minutes preserved endothelium-dependent and endothelium-independent relaxation to a certain extent. However, storage in heparinized whole blood significantly enhanced endothelium-dependent relaxation to acetylcholine and has been shown to cause minimal morphologic changes. 14 He 15 reported a marked inhibition of relaxation to acetylcholine in the RA stored in normal saline solution with 0.1% papaverine. This is important because an indication of early endothelial dysfunction can be expressed as reduction or loss of endothelium-dependent relaxation to acetylcholine. 16 He 15 hypothesized that the acidic nature of the solution, secondary to the papaverine, could have resulted in damage to the endothelium. We have also shown similar attenuation of endothelium-dependent relaxation to acetylcholine in rings stored in normal saline solution alone, which has a neutral pH. Storage of saphenous vein in crystalloid solution, particularly normal saline solution, has been associated with loss of endothelial coverage of more than 80%, with minor injury to media and adventitia. 6,17 Therefore, the effect reported by He and associates 15 could be partly attributed to storage in normal saline solution rather than papaverine alone.
Media like the University of Wisconsin solution contain antioxidants and adenosine, which are included to help the storage and preservation of the transplant organ. 18 Adenosine may delay the degradation and also stimulate the synthesis of adenosine nucleotides. Blood is a good source of elements, such as nitric oxide, HCO3–, antioxidants, adenosine, and other phosphate energy sources, which are protective of endothelial function. They also have better oxygen-carrying capacity than the nonsanguineous solutions, such as normal saline solution. This may explain the better results observed in rings stored in heparinized whole blood. The prolonged storage of RA in normal saline solution may result in a redistribution of sodium ions from the breakdown of the sodium-potassium pump as a result of the lack of supply of energy sources, causing endothelial cell edema, separation, and desquamation of cells. 14 This cytotoxicity of crystalloid solutions with prolonged contact has also been shown in endothelial cell cultures. 19 However, unlike normal saline solution, Ringer's solution showed better preservation of contractile and endothelial function. This could be due to the more physiologic composition of Ringer's solution. Krebs solution was used as the control in our experiment, and this may not be ideal because RA rings stored in Krebs solution achieved significantly less endothelium-dependent relaxation to acetylcholine than heparinized whole blood.
The 45-minute storage of RA in the 3 types of storage solutions may not be sufficiently long to produce permanent damage to the media because both calcium channel receptor blockade–induced relaxation caused by verapamil and endothelium-independent relaxation caused by nitroprusside were preserved, indicating a normally functioning smooth muscle media.
The addition of vasodilators, such as papaverine, verapamil, or nitrates, to the storage solution, which is a common practice in all cardiac surgical units, may alter the effect of the storage solution on endothelial function. This was not the primary aim of the experiments. For this reason, we ensured that study segments were not exposed to verapamil or other vasodilators during the harvesting or surgical preparation techniques.
Conclusions
The type of media used in surgical preparation and storage of graft conduits has received little attention, despite the fact that the initial insult resulting from the media may have sealed the fate of both the short- and long-term patency of the grafts. Thus, it is important that the types of storage media used should maximize endothelial preservation in addition to contributing to the benefits of better harvesting techniques. Autologous blood, which has bathed the endothelium of the RA in its native anatomic position before removal, should be used for storing RA conduits to preserve both endothelial and contractile function. Normal saline solution should be avoided because we have demonstrated that it results in a disruption of endothelial integrity and impairs function.
References
This article has been cited by other articles:
|
|
 |
|
  K. L. Sloots, J. Vinten-Johansen, and G. P. Dobson Warm nondepolarizing adenosine and lidocaine cardioplegia: Continuous versus intermittent delivery J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1171 - 1178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Y. Oo, A. R. Conant, M. R. Chester, W. C. Dihmis, and A. W.M. Simpson Temperature Changes Stimulate Contraction in the Human Radial Artery and Affect Response to Vasoconstrictors Ann. Thorac. Surg., January 1, 2007; 83(1): 126 - 132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. F. Chong, P. J. L. Ong, N. Moat, and P. Collins Effects of hydrostatic distention on in vitro vasoreactivity of radial artery conduits J. Thorac. Cardiovasc. Surg., October 1, 2004; 128(4): 609 - 614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Emir, M. K. Gol, K. Ozisik, V. Bakuy, M. F. Sargon, S. Yavas, K. Cagli, K. Kilinc, and E. Sener Harvesting Techniques Affect the Integrity of the Radial Artery: An Electron Microscopic Evaluation Ann. Thorac. Surg., October 1, 2004; 78(4): 1319 - 1325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mayranpaa, J. Simpanen, M. W. Hess, K. Werkkala, and P. T. Kovanen Arterial endothelial denudation by intraluminal use of papaverine-NaCl solution in coronary bypass surgery Eur. J. Cardiothorac. Surg., April 1, 2004; 25(4): 560 - 566. [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 |
