- Email: [email protected]
Inhibition of human internal mammary artery contractions
Inhibition of human internal mammary artery contractions An in vitro study of vasodilators The internal mammary artery is currently the preferred conduit for myocardial revascularization; however, perioperative vasospasm of the internal mammary artery may limit its use as a bypass graft. The ability of various vasodilators to inhibit internal mammary artery contraction was investigated with the use of discarded segments of human internal mammary artery not used in coronary artery bypass grafting. Ring segments of human internal mammary arteries were suspended on strain gauges in muscle baths containing 37° C Krebs solution for measurement of isometric tension in vitro. Arterial contraction was stimulated by elevating the extracellular potassium concentration to 70 mmol/L or by exposure to a 10 ~mol/L concentration of norepinephrine, and inhibition of contraction by vasodilators was measured. The order of potency to inhibit potassium-induced contraction was as follows: nifedipine > verapamil > nitroprusside> papaverine. At maximal effective doses, nifedipine, verapamil, and papaverine almost completely inhibited potassium-induced contraction, whereas nitroprusside inhibited contraction by only 55 %. When norepinephrine was used to contract the arteries, a biphasic relaxation curve was seen with nifedipine, but not with other vasodilator drugs. The order of potency to inhibit norepinephrine-induced contraction was as follows: nifedipine > nitroprusside> verapamil > papaverine. Maximal inhibition of norepinephrine contraction by these vasodi1ators ranged from 68 % to 95 %. Nitroglycerin, isoproterenol, and adenosine produced little or no inhibition of internal mammary artery contraction caused by potassium or norepinephrine. Although nifedipine was the most potent vasodilator, papaverine produced the greatest maximal inhibition of both potassium- and norepinephrine-induced contraction of human internal mammary artery. (J THORAC CARDIOVASC SURG 1992;104:977-82)
G. Kimble Jett, MD,a Robert A. Guyton, MD,a Charles R. Hatcher, Jr., MD,a and Peter W. Abel, PhD,b Atlanta, Ga.
h e proved long-term patency of the internal mammary artery (IMA) as a bypass graft has made it the preferred conduit for myocardial revascularization 1,2; however, vasospasm during isolation of this vessel with its pedicle is common.v" Treatment of vasospasm of the IMA not only may improve the immediate flow after From the Department of Cardiothoracic Surgery; The Carlyle Fraser Heart Center, Crawford W. Long Memorial Hospital of Emory University, and from the Department of Pharmacology," Emory University School of Medicine, Atlanta, Ga. Receivedfor publication June 21, 1990.
bypass grafting but also may influence the decision regarding the use of the artery as a graft. Various vasodilators, as well as mechanical dilation of the artery, have been used to prevent or reverse vasospasm'- 4 The risk of intimal injury with mechanical dilation renders pharmacologic manipulation as the preferred treatment of vasospasm.' This study was designed to determine the effectiveness of various vasodilator drugs in inhibiting contraction of human IMAs studied in vitro. This approach may provide information regarding the most effective drug treatment of vasospasm in patients.
Accepted for publication Aug. 27,1991.
Methods
Address for reprints: G. Kimble Jett, MD, Baylor University Medical Center, 3409 Worth Street, Suite 720, Dallas, TX 75246.
Tissue preparation and contraction measurements. Segments of 25 different human IMAs were obtained from the Emory University Hospital operating room. The extra length of
12/1/33401
977
978
The Journal of Thoracic and Cardiovascular Surgery
Jett et al.
artery not used for coronary artery grafting was used in these studies. The experimental protocol was submitted to the Human Investigations Committee of the Emory University School of Medicine, and permission to use these arteries was granted on the basis of their being discarded specimens. IMA segments were removed from patients, placed directly into Krebs solution (4° C), and transported to the laboratory. They were immediately placed in oxygenated Krebs solution, cleaned of loose connective tissue, and cut into 2.5 mm long rings. The outer diameter of the rings ranged from 2.4 to 3.8 mm. Two stainless steel wires were placed through the lumen of each ring. The rings were placed in separate muscle baths, and one wire was connected to an immovable support while the other wire was attached to a Grass FT 0.03 force transducer (Grass Instrument Co., Quincy, Mass.). The time from collection in the operating room to placement in the muscle bath was less than 15 minutes. Attention and care were given to maintaining and not disrupting the integrity of the endothelium of each segment of artery. The presence of intact endothelium with this method of tissue preparation has been demonstrated." Isometric contractions were recorded on a Beckman Dynograph (Beckman Instruments, Inc., Schiller Park, IIl.) or a Grass modelS polygraph. The rings were equilibrated in the muscle baths in Krebs solution containing (in millimoles per liter) NaCI, 120; KCI, 5.5; CaCh, 2.5; NaH zP04, 1.2; NaHC03, 20; dextrose, II; and CaNaz ethylenediaminetetraacetic acid, 0.029. They were gassed with 95% oxygen and 5% carbon dioxide and maintained at 37° C. Rings were maintained at 1.5 gresting tension, which was determined to be optimal for maximal agonist-induced tension development in preliminary experiments. Each ring was equilibrated in Krebs solution for approximately 30 to 40 minutes with continuous readjustment of resting tension to 1.5 g. Rings were then contracted either with Krebs solution having a potassium content of 70 mmol/L or with norepinephrine lO ~mol/L. The Krebs solution containing potassium, 70 mmol/L, was made by equimolar substitution of potassium for sodium chloride to bring the potassium concentration to 70 mmol/L, Arteries contracted with a high-potassium solution were incubated in Krebs solution containing phentolamine, I ~mol/L, to block contraction caused by potassium-induced release of norepinephrine. In preliminary experiments, it was determined that potassium, 70 mrnol/L, and norepinephrine, 10 ~mol/L, caused maximal contraction of the human IMA. The drugs used were obtained from the following sources: (-)-norepinephrine bitartate, (± )-verapamil HCI, sodium nitroprusside, adenosine, dipyridamole (Sigma Chemical Co., St. Louis, Mo.), phentolamine mesylate (Ciba Pharmaceutical, Summit, N.J.), nitroglycerin (American Critical Care, McGaw Park, 111.), and nifedipine (Pfizer Laboratories, New York, N.Y.). Stock solutions of drugs (usually I mrnol/L) were prepared daily in 0.9% saline, except for nifedipine, which was made in 0.9% saline containing I% acetone and protected from light. Stock solutions were diluted by lO-fold to lOoo-fold with Krebs solution. Experiments involving nifedipine were performed with reduced lighting to prevent drug breakdown. In separate experiments, it was determined that the nifedipine vehicle containing acetone and the nitroglycerin vehicle had no effect on norepinephrine- or potassium-induced contraction. Experimental protocol. The experiments were performed by simultaneously studying four- to six-ring segments from the same or different IMAs. After the equilibration period, rings
were contracted with potassium or norepinephrine, the contractile agent was washed out, and the tissues were equilibrated for an additional 20 to 30 minutes. This sequence of drug exposure and washout was repeated two to four times until stable and reproducible contractions were obtained. One ring from each segment of artery was not exposed to vasodilator and served as a control, whereas the other rings were incubated with various vasodilators for 20 to 30 minutes. The rings were then exposed to potassium or norepinephrine plus vasodilator and peak contraction was measured. The contractile agent was washed out, and a higher concentration of vasodilator was added and allowed to equilibrate for 20 to 30 minutes. This equilibration period allowed each segment to return to baseline tension before stimulation and therefore ensured drug washout. This sequence of contraction, washout, and vasodilator incubation was repeated for each concentration of vasodilator tested. Each arterial ring was exposed to only one vasodilator. Data analysis. The contractile force of arterial rings not exposed to vasodilator was used as the control (100%) response. The contraction after exposure to vasodilator was compared with the parallel control contraction to calculate the response as a percent of control. Contraction response curves were generated by plotting the percent of control contraction versus the log of concentration of vasodilator. EC so values (effective concentration of vasodilator required for 50% maximal inhibition of contraction) were calculated from the concentration-response curves by linear regression of all points between 20% and 80% of the maximal contraction. The biphasic concentration-response curve for nifedipine was analyzed by least squares nonlinear regression analysis of the relaxation dose-response curve. The mean ECso value for each agent was calculated and used for comparison of drug potency. EC so values and maximal relaxation were compared by analysis of variance followed by a Neuman-Keuls test (p < 0.05) for differences between responses for each drug. Results Contractile response to potassium and norepinephrine. Control segments not exposed to vasodilator drugs maintained relatively consistent maximal contractions with repeated potassium stimulation (Fig. I). The maximal contractile response elicited by a potassium concentration of 70 mmoljL produced a mean peak contraction of 2505 ± 300 mg. Higher potassium concentration did not increase the maximal contractile response. After washout of the potassium Krebs solution, arterial rings relaxed to the initial baseline tension within 15 to 20 minutes. Norepinephrine-induced contractions were less consistent than those caused by potassium and usually resulted in a progressive decline in contraction with repeated stimulation. The maximal decrease in norepinephrineinduced contraction of control arteries averaged 24% ± 6%. A 10 JlmoljL concentration of norepinephrine produced a mean peak contraction of 2835 ± 410 mg. Norepinephrine-contracted arteries usually relaxed to the initial baseline tension within 20 to 30 minutes after
Volume 104 Number 4 October 1992
c o o
-...
Inhibition of IMA contractions
100
c:
ttl
o :;: 100
c
...ttl
o
o
-
Nifedipine
Verapamil
•
o
Nitroprusside
Papaverine
o
c:
60
o
U (5
...
ttl
c:
o ?J?
•
o
979
20
•
Papa verine
o
Control
'E o
o_ 1--
, " - . . 1 . . - _ - - ' -_ _......1.-_ _
6
60
5
[Papaverine]
20
o ~
~
4
10
drug washout. Some arteries relaxed very slowly after drug washout and were not used in these studies. Vasodilator inhibition of potassium contraction. Concentration-response curves for vasodilator-induced inhibition of potassium contraction are shown in Fig. 2. The order of vasodilator potency to inhibit potassium-induced contraction was as follows: nifedipine > verapamil > nitroprusside> papaverine. Nitroglycerin, isoproterenol, and adenosine were not effective in inhibiting contraction and were tested only at a single high concentration (100 ~mol/L). In some experiments in which adenosine, was used, dipyridamole, 10 p,moljL, was added to ensure that adenosine uptake and breakdown were not responsible for the ineffectiveness of adenosine as a vasodilator. Dipyridamole had no effecton adenosine inhibition of potassium contraction. The maximal vasodilator inhibition of potassium contraction is listed in Table I. Papaverine, nifedipine, and verapamilwere the most effectiverelaxant drugs, almost completely inhibiting potassium contraction. Sodium nitroprusside, nitroglycerin, and isoproterenol caused less inhibition and adenosine had no effect. Vasodilator inhibition of norepinephrine contraction. Concentration-response curves for vasodilator inhi-
7
8
[Drug]
(-log M)
Fig. 1. Protocol for generating vasodilator concentration-response curves inhuman IMA ringsegments. Both control (open squares) andpapaverine-exposed (solid squares) arteries (n = 5 for each group) were stimulated with potassium every 30minutes. Aftertheresponse to potassium wasreproducible, thenext contraction was assigned the initial 100% response. In the absence of papaverine, potassium contractions were relatively constant, whereas therewas a dose-related decrease in contractile response in the presence of papaverine. In all subsequent experiments, the response in the presence of vasodilator was compared with thecontractile response ofa control preparation. Inallcases, control anddrug-exposed rings were takenfrom the same segment of the IMA.
9
6
4
5
(-log M)
Fig. 2. Mean concentration-response curves for vasodilator inhibition of potassium-induced contraction. Responses are plotted as a percent of the potassium response of control preparations not exposed to vasodilator drugs. Nifedipine was the most potent drug, and nifedipine, verapamil, and papaverine all caused nearly maximal inhibition of contraction. Each curve represents the mean of dose-response curves from five separate arteries. Table I. Relaxant effects of vasodilators on potassium-induced contraction of human IMAs
Nifedipine Verapamil Sodium nitroprusside Papaverine Nitroglycerin Isoproterenol Adenosine
EC50 (95% CI)
Maximum
(J.tmol/L)
(% inhibition)
0.004 (0.002-0.007)* 0.224 (0.148-0.339) 0.302 (0.240-0.380) 8.32 (3.89-17.0)*
92 ± I 91 55 96 30 21 6
± I
± ± ± ± ±
7 3 8 7 4
Tissues were incubated with various concentrations of vasodilator drugs and then contracted with potassium, 70 mmol/L, ECso values and maximal inhibition of contraction were calculated from inhibition concentration-response curves. For some vasodilators, only one concentration (100 "mol/L) was tested. Each value is the mean plus 95% confidence interval (95% CI) for EC so values, or the mean ± standard error of the mean for maximal relaxation, of responses from five arteries. •p
< 0.05 when compared
with the other vasodilators.
bition of norepinephrine-induced contractions are shown in Fig. 3. The concentration-responsecurve for nifedipine inhibition of norepinephrine contraction was biphasic, unlike nifedipine relaxation of potassium-contracted rings (Fig. 2). The order of potency to inhibit norepinephrine-induced contraction was as follows: first-phase nifedipine > nitroprusside> second-phase nifedipine > verapamil > papaverine. The least effectivevasodilators, nitroglycerin and isoproterenol, were tested only at a single concentration of I00 ~mol/L.
980
The Journal of Thoracic and Cardiovascular Surgery
Jett et al.
r::
•
Nitedipine
o
Verapamil
•
o
Table II. Relaxant effects of vasodilators on norepinephrine-induced contraction of human [MAs
Nitroprusside
Papaverine
~ 100
-... o
as r::
o
o
60
C o o _
20
...
Nifedipine First phase Second phase Verapamil Sodium nitroprusside Papaverine Nitroglycerin Isoproterenol
(5
o
*-
6
7
[Drug]
6
5
4
3
(-log M)
Fig. 3. Mean concentration-response curves for selected vasodilators to inhibit norepinephrine-induced contractions of human IMA. Nifedipine produced a biphasic concentration-response curve and was the most potent vasodilator drug. Papaverinewasthe leastpotentdrug but produced the greatestmaximal inhibition of contraction. Each curveis the meanof five or sixindividual dose-response curves obtainedin separatetissues.
Maximal vasodilator inhibition of norepinephrine contraction is listed in Table II. The order of maximal relaxationwasasfollows:papaverine > verapamil > nitroprusside> nifedipine > nitroglycerin> isoproterenol. With the exception of papaverine, the vasodilators were less effective at inhibiting norepinephrine-induced contraction than potassium-induced contraction. Papaverine was equally effective for relaxation of norepinephrine- or potassium-stimulated arteries. Discussion The IMA is currently the preferred conduit for myocardial revascularization.I: 8 Vasospasm can occur during isolation of the IMA and its pedicle. The management of vasospasm not only may influence the decision regarding use of the artery as a graft, but also may improve the immediate flow through the IMA after the operation." Some studies have used mechanical dilation of the IMA to reverse vasospasm and also to judge the diameter of the artery- 9; however, mechanical dilation can cause intimal damage that may result in thrombosis or obstruct flow in the graft, or both. 5 , 10, II In addition, endothelial cell damage may enhance arterial spasm or constriction of the IMA by reducing the formation of endothelium-derived relaxing factors. 5, 6, 12, 13 Topical papaverine has been used to reverse IMA spasm.v 10, 14 but other vasodilators have received less attention. In this study, two methods were used to cause contraction of the IMA. Potassium contracts vascular smooth muscle by direct membrane depolarization, which results
EC50 (95% CI)
Maximum
(Jlmol/L)
(% inhibition)
0.019 3.29 4.68 0.447 4.07
(0.012-0.03 I)" ( 1.79-6.07) (3.12-7.01) (0.190-0.996) (2.26-7.33)
30 ± 3 38 ± 4 75 ± 5 71 ± 8 95 ± 5" 56 ± 5 35 ± 9
Tissues were contracted with norepinephrine, 10 I'mol/L, in the absence or presence of various concentrations of vasodilators. Concentration-response curves for inhibition of contraction by vasodilators were plotted, and EC,o and maximal responses were calculated. For nitroglycerin and isoproterenol, only one concentration (100 I'mol/L) was tested. Each value is the mean plus 95% confidence intervals (95% CO for EC,o values, or the mean ± standard error of the mean for maximal relaxation, of responses from five or six arteries. "p < 0.05 when compared with the other vasodilators.
in an influx of extracellular calcium across the cell membrane through voltage-operated calcium channels. 15, 16 Effects caused by potassium-induced release of endogenous norepinephrine from sympathetic nerve endings in the IMA were inhibited in our studies by the addition of phentolamine.l'' which blocks a-adrenergic receptors. Norepinephrine causes contraction of vascular muscle by stimulating a-adrenergic receptors, which cause a release of intracellular calcium'< 16 and influx of extracellular calcium through receptor-operated calcium channels. 16- 18 Because the mechanisms causing spasm of the human IMA are not known, two vasoconstrictors causing contraction by activating different calcium pools were used in these studies. The role of endothelium in arterial contraction has recently been ernphasized.P- 13, 19,20 Our study used a method of tissue preparation that preserves the integrity of the endothelium. Although we did not specifically test for the presence of endothelium, other studies in which an identical method of tissue preparation was used have demonstrated its presence.P: 13, 19, 21 Nifedipine and verapamil inhibit extracellular calcium influx through voltage-operated and receptor-operated calcium channels in vascular smooth muscle.F We found that the potency of nifedipine was greater for inhibition of potassium-induced contraction than norepinephrineinduced contraction. In addition, nifedipine caused a biphasic relaxation of norepinephrine-stimulated arteries. The low-affinity second phase of nifedipine relaxation may be caused by blockade of receptor-operated channels, whereas the high-affinity first phase may be due to blockade of voltage-dependent channels associated with norepinephrine stimulation. Verapamil was also potent
Volume 104
Inhibition of IMA contractions
Number 4 October 1992
and effective at inhibiting potassium-induced and norepinephrine-induced contraction. It did not show biphasic relaxation with norepinephrine, as did nifedipine. Nitroprusside and nitroglycerin can relax various smooth muscles including vascular, uterine, gastrointestinal, and bronchial muscles. Recent studies have found increased intracellular levels of cyclic guanosine monophosphate induced by nitroprusside and nitroglycerin23, 24; this increase in cyclic guanosine monophosphate may be responsible for relaxation of vascular smooth muscle. The mechanism of nitrate-induced relaxation is not voltage-dependent.P Our study found nitroprusside to be more potent and more effective than nitroglycerin. Although nitroglycerin has been shown to be potent at inhibiting contraction of human coronary arteries studied in vitro,26 it was not effective on the IMA in our studies. Nitroprusside had a similar potency to inhibit both potassium and norepinephrine contraction. Both nitroprusside and nitroglycerin were more effective in inhibiting norepinephrine-induced contraction than potassiuminduced contraction. The ;J-adrenergic agonist, isoproterenol, is also a potent vasodilator drug, which acts by increasing adenylate cyclase activity and causing an increase in intracellular cyclic adenosine monophosphate.I? Isoproterenol was neither potent nor effective in inhibiting contraction in our study. Isoproterenol did inhibit norepinephrine contraction slightly when used at high concentrations, possibly as a result of isoproterenol displacing norepinephrine from a-adrenergic receptors. In all cases, the arteries relaxed slowly after washout of isoproterenol, and relaxation to baseline often was incomplete. This was probably caused by persistent a-adrenergic receptor stimulation seen with high concentrations of isoproterenol.P The lack of relaxation with isoproterenol suggests that few ;J-adrenergic receptors are present in the human IMA. This is in contrast to the human coronary artery, which has more ;J-adrenergic receptors than a-adrenergic receptors.P: 29 A recent study also demonstrated weak ;J-adrenoceptormediated relaxation in the human IMA. 3o Adenosine is also a potent vasodilator, especially in the coronary circulation,25,31 which acts by causing a decrease in the membrane permeability to extracellular calcium ions" or by acting through an intracellular mechanism.V Adenosine was neither potent nor effective in inhibiting potassium-induced contraction in our experiments. Dipyridamole was added in some experiments to ensure that uptake and breakdown of adenosine were not responsible for its lack of effects. In all cases, dipyridamole had no effect on adenosine-induced relaxation. These results suggest that, unlike the coronary vascular bed, the IMA is not sensitive to adenosine relaxation.
98 I
Papaverine is a nonspecific, noncompetitive vasodilator. It inhibits oxidative phosphorylation'! but may cause vasodilation by inhibiting phosphodiesterase, which prevents breakdown of cyclic nucleotides that can lead to increased intracellular cyclic adenosine monophosphate or cyclic guanosine monophosphate concentrations.i?: 34 Papaverine was the least potent agent tested, but it was the most effective in maximally inhibiting both potassium-induced and norepinephrine-induced contraction. It was the only drug that completely reversed norepinephrine-induced contraction. Although papaverine is an effective drug, it has few therapeutic uses and may best be used by topical or intraluminal application to the IMA. Effective vasodilation with a resultant increase in IMA flow has been demonstrated when papaverine was injected intraluminally." Clinical implications. This study investigated the effectiveness of various vasodilators to inhibit potassiumand norepinephrine-induced contraction of the human IMA in vitro. Frequently, vascular spasm is the initial event that then necessitates treatment. Although we attempted to induce contraction and then relaxation of the artery with vasodilators, we found that the human IMA relaxes in a prolonged manner. We therefore accepted the limitations inherent in a pretreatment model to have more manageable data. Although papaverine is effective in preventing contraction of the IMA, it is not a clinically useful drug. It can best be used intraoperatively while the patient's chest is open. Nifedipine is a more clinically useful drug that can be used perioperatively with the chest closed. Nifedipine is potent and relatively effective in inhibiting both norepinephrine-induced and potassium-induced contraction. Nitroprusside and nitroglycerin are only moderately potent and effective. The human IMA appears to have few ;J-adrenergic receptors because it was only minimally responsive to isoproterenol. Our study therefore supports the intraoperative use of papaverine for treatment of vasospasm seen during isolation of the IMA and its pedicle. Nifedipine may be a better drug for treatment or prevention of perioperative IMA spasm. REFERENCES I. Lytle BW, Loop FD, Cosgrove DM, Ratliff NB, EasleyK, Taylor P'C, Long-term (5-12 years) serial studies of internal mammary artery and saphenousvein coronary bypass grafts. J THORAC CARDIOVASC SURG 1985;89:248-58. 2. Barner HB, Standeven JW, Reese J. Twelve-year experience with internal mammary artery for coronary artery bypass. J THORAC CARDIOVASC SURG 1985;90:668-75. 3. Green GE. Techniques of internal mammary--coronary artery anastomosis. J THORAC CARDIOVASC SURG 1979; 78:455-9.
982
Jett et al.
4. Mills NL, Bringaze WL. Preparation of the internal mammary artery graft. Which is the best method? J THORAC CARDIOVASC SURG 1989;98:73-9. 5. Johns RA, Peach MJ, Flanagan T, Kron IL. Probing ofthe canine mammary artery damages endothelium and impairs vasodilation resulting from prostacyclin and endotheliumderived relaxing factor. J THORAC CARDIOVASC SURG 1989;97:252-8. 6. Luscher TF, Diederich D, Siebenmann R, et al. Difference between endothelium-dependent relaxation in arterial and venous coronary bypass grafts. N Engl J Med 1988; 319:462-7. 7. Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal mammary artery graft on 10-year survival and other cardiac events. N Engl J Med 1986;314:1-6. 8. Tector AJ, Schmahl TM, Canino VR. Expanding the use of the internal mammary artery to improve patency in coronary artery bypass grafting. J THORAC CARDIOVASC SURG 1986;91:9-16. 9. Green GE. Coronary bypass highlights blood vessel biology. Ann Thorac Surg 1986;41:235-6. 10. Mills NL, Ochsner JL. Technique of internal mammaryto-coronary artery bypass. Ann Thorac Surg 1974;17:23746. 11. Barner HB: Double internal mammary-coronary artery bypass. Arch Surg 1974;109:627-30. 12. Tesfamarian B, Halpern W, Osol G. Effects of perfusion and endothelium on the reactivity of isolated resistance arteries. Blood Vessels 1986;22:301-5. 13. Yang Z, Diederich D, Schneider K, et al. Endothelium-derived relaxing factor and protection against contractions induced by histamine and serotonin in the human internal mammary artery and in the saphenous vein. Circulation 1989;80:1041-8. 14. Green GF. Internal mammary-coronary artery anastomosis for myocardial ischemia. In: Sabiston DC, Spencer FC, eds. Gibbon's surgery of the chest, Philadelphia: WB Saunders, 1983:1451-8. 15. Loutzenhiser R, Epstein M. Activation mechanisms of human renal artery: effects of KCI, norepinephrine, and nifedipine upon tension development and 45Ca influx. Eur J PharmacoI1984;106:47-52. 16. Hogestatt ED. Characterization of two different calcium entering pathways in small mesenteric arteries from rat. Acta Physiol Scand 1984;122:483-95. 17. Lorenz RR, Vanhoutte PM. Inhibition of adrenergic neurotransmission in isolated veins of dog by potassium ion. J Physiol 1975;246:479-500. 18. Haeusler G. Relationship between noradrenaline-induced depolarization and contraction in smooth muscle. Blood Vessels 1978;15:46-54.
The Journal of Thoracic and Cardiovascular Surgery
19. Furchgott RF, Zawadsk JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288:373-6. 20. Furchgott RF. Role of endothelium in responses of vascular smooth muscle. Circ Res 1983;53:557-73. 21. Oshiro MEM, Paiva AC, Paiva TB. Endothelium-dependent inhibition of the use of extracellular calcium for arterial response to vasoconstrictor agents. Gen Pharmacol 1985;16:567-72. 22. Sperelakis N, Mras S. Depression of rabbit aorta and guinea pig vena cava by mesudipine and other slow channel blockers. Blood Vessels 1983;20:172-83. 23. Bohme E, Graf H, Schultz G. Effects of sodium nitroprusside and other smooth muscle relaxants on cyclic GMP formation in smooth muscle and platelets. Adv Cyc Nucl Res 1978;9:131-43. 24. Rapaport RM, Murad F. Endothelium-dependent and nitrovasodilator-induced relaxation of vascular smooth muscle: role of cyclic GMP. J Cyc Nucl Prot Phosph Res 1983;9:281-96. 25. Schmaar RL, Sparks HV. Response of large and small arteries to nitroglycerin, NaNOz, and adenosine. Am J Physiol 1972;223:223-8. 26. Ginsburg R, Bristow MR, Harrison DC, Stinson EB. Studies with isolated human coronary arteries: some general observations, potential mediators of spasm, role of calcium antagonists. Chest 1980;78(suppl 1):180-6. 27. Triner L, Nahas GG, Overweg NIA, Verosky M, Habif DV, Ngai SH. Cyclic AMP and smooth muscle function. Ann NY Acad Sci 1971;185:458-76. 28. Dorevitch N. Effect of isoproterenol on adrenergic receptors in rabbit aorta. Arch Int Pharmacodyn Ther 1968; 174:98-107. 29. Godfraind T. Pharmacology of human coronary arteries. Acta Cardiol 1985;40:441-6. 30. He GW, Buxton B, Rosenfeldt FL, Wilson AC, Angus JA. Weak adrenoceptor-mediated relaxation in the human internal mammary artery. J THORAC CARDIOVASC SURG 1989;97:259-66. 31. Herlihy JT, Bockman EL, Berna RM, Rubio R. Adenosine relaxation of isolated vascular smooth muscle. Am J PhysioI1976;230:1239-43. 32. Verhueghe RH. Action of adenosine and adenine nucleotides on dog's isolated veins. Am J Physiol 1977; 233:HI14-21. 33. Santi R, Contessa AR, Ferrari M. Spasmolytic effect of the papaverine and inhibition of the oxidative phosphorylation. Biochem Biophys Res Commun 1963;11:156-9. 34. Triner L, Vulliemoz YU, Schwartz I, Nahas GG. Cyclic phosphodiesterase activity and the action of papaverine. Biochem Biophys Res Commun 1970;40:64-9.
G. Kimble Jett, MD,a Robert A. Guyton, MD,a Charles R. Hatcher, Jr., MD,a and Peter W. Abel, PhD,b Atlanta, Ga.
h e proved long-term patency of the internal mammary artery (IMA) as a bypass graft has made it the preferred conduit for myocardial revascularization 1,2; however, vasospasm during isolation of this vessel with its pedicle is common.v" Treatment of vasospasm of the IMA not only may improve the immediate flow after From the Department of Cardiothoracic Surgery; The Carlyle Fraser Heart Center, Crawford W. Long Memorial Hospital of Emory University, and from the Department of Pharmacology," Emory University School of Medicine, Atlanta, Ga. Receivedfor publication June 21, 1990.
bypass grafting but also may influence the decision regarding the use of the artery as a graft. Various vasodilators, as well as mechanical dilation of the artery, have been used to prevent or reverse vasospasm'- 4 The risk of intimal injury with mechanical dilation renders pharmacologic manipulation as the preferred treatment of vasospasm.' This study was designed to determine the effectiveness of various vasodilator drugs in inhibiting contraction of human IMAs studied in vitro. This approach may provide information regarding the most effective drug treatment of vasospasm in patients.
Accepted for publication Aug. 27,1991.
Methods
Address for reprints: G. Kimble Jett, MD, Baylor University Medical Center, 3409 Worth Street, Suite 720, Dallas, TX 75246.
Tissue preparation and contraction measurements. Segments of 25 different human IMAs were obtained from the Emory University Hospital operating room. The extra length of
12/1/33401
977
978
The Journal of Thoracic and Cardiovascular Surgery
Jett et al.
artery not used for coronary artery grafting was used in these studies. The experimental protocol was submitted to the Human Investigations Committee of the Emory University School of Medicine, and permission to use these arteries was granted on the basis of their being discarded specimens. IMA segments were removed from patients, placed directly into Krebs solution (4° C), and transported to the laboratory. They were immediately placed in oxygenated Krebs solution, cleaned of loose connective tissue, and cut into 2.5 mm long rings. The outer diameter of the rings ranged from 2.4 to 3.8 mm. Two stainless steel wires were placed through the lumen of each ring. The rings were placed in separate muscle baths, and one wire was connected to an immovable support while the other wire was attached to a Grass FT 0.03 force transducer (Grass Instrument Co., Quincy, Mass.). The time from collection in the operating room to placement in the muscle bath was less than 15 minutes. Attention and care were given to maintaining and not disrupting the integrity of the endothelium of each segment of artery. The presence of intact endothelium with this method of tissue preparation has been demonstrated." Isometric contractions were recorded on a Beckman Dynograph (Beckman Instruments, Inc., Schiller Park, IIl.) or a Grass modelS polygraph. The rings were equilibrated in the muscle baths in Krebs solution containing (in millimoles per liter) NaCI, 120; KCI, 5.5; CaCh, 2.5; NaH zP04, 1.2; NaHC03, 20; dextrose, II; and CaNaz ethylenediaminetetraacetic acid, 0.029. They were gassed with 95% oxygen and 5% carbon dioxide and maintained at 37° C. Rings were maintained at 1.5 gresting tension, which was determined to be optimal for maximal agonist-induced tension development in preliminary experiments. Each ring was equilibrated in Krebs solution for approximately 30 to 40 minutes with continuous readjustment of resting tension to 1.5 g. Rings were then contracted either with Krebs solution having a potassium content of 70 mmol/L or with norepinephrine lO ~mol/L. The Krebs solution containing potassium, 70 mmol/L, was made by equimolar substitution of potassium for sodium chloride to bring the potassium concentration to 70 mmol/L, Arteries contracted with a high-potassium solution were incubated in Krebs solution containing phentolamine, I ~mol/L, to block contraction caused by potassium-induced release of norepinephrine. In preliminary experiments, it was determined that potassium, 70 mrnol/L, and norepinephrine, 10 ~mol/L, caused maximal contraction of the human IMA. The drugs used were obtained from the following sources: (-)-norepinephrine bitartate, (± )-verapamil HCI, sodium nitroprusside, adenosine, dipyridamole (Sigma Chemical Co., St. Louis, Mo.), phentolamine mesylate (Ciba Pharmaceutical, Summit, N.J.), nitroglycerin (American Critical Care, McGaw Park, 111.), and nifedipine (Pfizer Laboratories, New York, N.Y.). Stock solutions of drugs (usually I mrnol/L) were prepared daily in 0.9% saline, except for nifedipine, which was made in 0.9% saline containing I% acetone and protected from light. Stock solutions were diluted by lO-fold to lOoo-fold with Krebs solution. Experiments involving nifedipine were performed with reduced lighting to prevent drug breakdown. In separate experiments, it was determined that the nifedipine vehicle containing acetone and the nitroglycerin vehicle had no effect on norepinephrine- or potassium-induced contraction. Experimental protocol. The experiments were performed by simultaneously studying four- to six-ring segments from the same or different IMAs. After the equilibration period, rings
were contracted with potassium or norepinephrine, the contractile agent was washed out, and the tissues were equilibrated for an additional 20 to 30 minutes. This sequence of drug exposure and washout was repeated two to four times until stable and reproducible contractions were obtained. One ring from each segment of artery was not exposed to vasodilator and served as a control, whereas the other rings were incubated with various vasodilators for 20 to 30 minutes. The rings were then exposed to potassium or norepinephrine plus vasodilator and peak contraction was measured. The contractile agent was washed out, and a higher concentration of vasodilator was added and allowed to equilibrate for 20 to 30 minutes. This equilibration period allowed each segment to return to baseline tension before stimulation and therefore ensured drug washout. This sequence of contraction, washout, and vasodilator incubation was repeated for each concentration of vasodilator tested. Each arterial ring was exposed to only one vasodilator. Data analysis. The contractile force of arterial rings not exposed to vasodilator was used as the control (100%) response. The contraction after exposure to vasodilator was compared with the parallel control contraction to calculate the response as a percent of control. Contraction response curves were generated by plotting the percent of control contraction versus the log of concentration of vasodilator. EC so values (effective concentration of vasodilator required for 50% maximal inhibition of contraction) were calculated from the concentration-response curves by linear regression of all points between 20% and 80% of the maximal contraction. The biphasic concentration-response curve for nifedipine was analyzed by least squares nonlinear regression analysis of the relaxation dose-response curve. The mean ECso value for each agent was calculated and used for comparison of drug potency. EC so values and maximal relaxation were compared by analysis of variance followed by a Neuman-Keuls test (p < 0.05) for differences between responses for each drug. Results Contractile response to potassium and norepinephrine. Control segments not exposed to vasodilator drugs maintained relatively consistent maximal contractions with repeated potassium stimulation (Fig. I). The maximal contractile response elicited by a potassium concentration of 70 mmoljL produced a mean peak contraction of 2505 ± 300 mg. Higher potassium concentration did not increase the maximal contractile response. After washout of the potassium Krebs solution, arterial rings relaxed to the initial baseline tension within 15 to 20 minutes. Norepinephrine-induced contractions were less consistent than those caused by potassium and usually resulted in a progressive decline in contraction with repeated stimulation. The maximal decrease in norepinephrineinduced contraction of control arteries averaged 24% ± 6%. A 10 JlmoljL concentration of norepinephrine produced a mean peak contraction of 2835 ± 410 mg. Norepinephrine-contracted arteries usually relaxed to the initial baseline tension within 20 to 30 minutes after
Volume 104 Number 4 October 1992
c o o
-...
Inhibition of IMA contractions
100
c:
ttl
o :;: 100
c
...ttl
o
o
-
Nifedipine
Verapamil
•
o
Nitroprusside
Papaverine
o
c:
60
o
U (5
...
ttl
c:
o ?J?
•
o
979
20
•
Papa verine
o
Control
'E o
o_ 1--
, " - . . 1 . . - _ - - ' -_ _......1.-_ _
6
60
5
[Papaverine]
20
o ~
~
4
10
drug washout. Some arteries relaxed very slowly after drug washout and were not used in these studies. Vasodilator inhibition of potassium contraction. Concentration-response curves for vasodilator-induced inhibition of potassium contraction are shown in Fig. 2. The order of vasodilator potency to inhibit potassium-induced contraction was as follows: nifedipine > verapamil > nitroprusside> papaverine. Nitroglycerin, isoproterenol, and adenosine were not effective in inhibiting contraction and were tested only at a single high concentration (100 ~mol/L). In some experiments in which adenosine, was used, dipyridamole, 10 p,moljL, was added to ensure that adenosine uptake and breakdown were not responsible for the ineffectiveness of adenosine as a vasodilator. Dipyridamole had no effecton adenosine inhibition of potassium contraction. The maximal vasodilator inhibition of potassium contraction is listed in Table I. Papaverine, nifedipine, and verapamilwere the most effectiverelaxant drugs, almost completely inhibiting potassium contraction. Sodium nitroprusside, nitroglycerin, and isoproterenol caused less inhibition and adenosine had no effect. Vasodilator inhibition of norepinephrine contraction. Concentration-response curves for vasodilator inhi-
7
8
[Drug]
(-log M)
Fig. 1. Protocol for generating vasodilator concentration-response curves inhuman IMA ringsegments. Both control (open squares) andpapaverine-exposed (solid squares) arteries (n = 5 for each group) were stimulated with potassium every 30minutes. Aftertheresponse to potassium wasreproducible, thenext contraction was assigned the initial 100% response. In the absence of papaverine, potassium contractions were relatively constant, whereas therewas a dose-related decrease in contractile response in the presence of papaverine. In all subsequent experiments, the response in the presence of vasodilator was compared with thecontractile response ofa control preparation. Inallcases, control anddrug-exposed rings were takenfrom the same segment of the IMA.
9
6
4
5
(-log M)
Fig. 2. Mean concentration-response curves for vasodilator inhibition of potassium-induced contraction. Responses are plotted as a percent of the potassium response of control preparations not exposed to vasodilator drugs. Nifedipine was the most potent drug, and nifedipine, verapamil, and papaverine all caused nearly maximal inhibition of contraction. Each curve represents the mean of dose-response curves from five separate arteries. Table I. Relaxant effects of vasodilators on potassium-induced contraction of human IMAs
Nifedipine Verapamil Sodium nitroprusside Papaverine Nitroglycerin Isoproterenol Adenosine
EC50 (95% CI)
Maximum
(J.tmol/L)
(% inhibition)
0.004 (0.002-0.007)* 0.224 (0.148-0.339) 0.302 (0.240-0.380) 8.32 (3.89-17.0)*
92 ± I 91 55 96 30 21 6
± I
± ± ± ± ±
7 3 8 7 4
Tissues were incubated with various concentrations of vasodilator drugs and then contracted with potassium, 70 mmol/L, ECso values and maximal inhibition of contraction were calculated from inhibition concentration-response curves. For some vasodilators, only one concentration (100 "mol/L) was tested. Each value is the mean plus 95% confidence interval (95% CI) for EC so values, or the mean ± standard error of the mean for maximal relaxation, of responses from five arteries. •p
< 0.05 when compared
with the other vasodilators.
bition of norepinephrine-induced contractions are shown in Fig. 3. The concentration-responsecurve for nifedipine inhibition of norepinephrine contraction was biphasic, unlike nifedipine relaxation of potassium-contracted rings (Fig. 2). The order of potency to inhibit norepinephrine-induced contraction was as follows: first-phase nifedipine > nitroprusside> second-phase nifedipine > verapamil > papaverine. The least effectivevasodilators, nitroglycerin and isoproterenol, were tested only at a single concentration of I00 ~mol/L.
980
The Journal of Thoracic and Cardiovascular Surgery
Jett et al.
r::
•
Nitedipine
o
Verapamil
•
o
Table II. Relaxant effects of vasodilators on norepinephrine-induced contraction of human [MAs
Nitroprusside
Papaverine
~ 100
-... o
as r::
o
o
60
C o o _
20
...
Nifedipine First phase Second phase Verapamil Sodium nitroprusside Papaverine Nitroglycerin Isoproterenol
(5
o
*-
6
7
[Drug]
6
5
4
3
(-log M)
Fig. 3. Mean concentration-response curves for selected vasodilators to inhibit norepinephrine-induced contractions of human IMA. Nifedipine produced a biphasic concentration-response curve and was the most potent vasodilator drug. Papaverinewasthe leastpotentdrug but produced the greatestmaximal inhibition of contraction. Each curveis the meanof five or sixindividual dose-response curves obtainedin separatetissues.
Maximal vasodilator inhibition of norepinephrine contraction is listed in Table II. The order of maximal relaxationwasasfollows:papaverine > verapamil > nitroprusside> nifedipine > nitroglycerin> isoproterenol. With the exception of papaverine, the vasodilators were less effective at inhibiting norepinephrine-induced contraction than potassium-induced contraction. Papaverine was equally effective for relaxation of norepinephrine- or potassium-stimulated arteries. Discussion The IMA is currently the preferred conduit for myocardial revascularization.I: 8 Vasospasm can occur during isolation of the IMA and its pedicle. The management of vasospasm not only may influence the decision regarding use of the artery as a graft, but also may improve the immediate flow through the IMA after the operation." Some studies have used mechanical dilation of the IMA to reverse vasospasm and also to judge the diameter of the artery- 9; however, mechanical dilation can cause intimal damage that may result in thrombosis or obstruct flow in the graft, or both. 5 , 10, II In addition, endothelial cell damage may enhance arterial spasm or constriction of the IMA by reducing the formation of endothelium-derived relaxing factors. 5, 6, 12, 13 Topical papaverine has been used to reverse IMA spasm.v 10, 14 but other vasodilators have received less attention. In this study, two methods were used to cause contraction of the IMA. Potassium contracts vascular smooth muscle by direct membrane depolarization, which results
EC50 (95% CI)
Maximum
(Jlmol/L)
(% inhibition)
0.019 3.29 4.68 0.447 4.07
(0.012-0.03 I)" ( 1.79-6.07) (3.12-7.01) (0.190-0.996) (2.26-7.33)
30 ± 3 38 ± 4 75 ± 5 71 ± 8 95 ± 5" 56 ± 5 35 ± 9
Tissues were contracted with norepinephrine, 10 I'mol/L, in the absence or presence of various concentrations of vasodilators. Concentration-response curves for inhibition of contraction by vasodilators were plotted, and EC,o and maximal responses were calculated. For nitroglycerin and isoproterenol, only one concentration (100 I'mol/L) was tested. Each value is the mean plus 95% confidence intervals (95% CO for EC,o values, or the mean ± standard error of the mean for maximal relaxation, of responses from five or six arteries. "p < 0.05 when compared with the other vasodilators.
in an influx of extracellular calcium across the cell membrane through voltage-operated calcium channels. 15, 16 Effects caused by potassium-induced release of endogenous norepinephrine from sympathetic nerve endings in the IMA were inhibited in our studies by the addition of phentolamine.l'' which blocks a-adrenergic receptors. Norepinephrine causes contraction of vascular muscle by stimulating a-adrenergic receptors, which cause a release of intracellular calcium'< 16 and influx of extracellular calcium through receptor-operated calcium channels. 16- 18 Because the mechanisms causing spasm of the human IMA are not known, two vasoconstrictors causing contraction by activating different calcium pools were used in these studies. The role of endothelium in arterial contraction has recently been ernphasized.P- 13, 19,20 Our study used a method of tissue preparation that preserves the integrity of the endothelium. Although we did not specifically test for the presence of endothelium, other studies in which an identical method of tissue preparation was used have demonstrated its presence.P: 13, 19, 21 Nifedipine and verapamil inhibit extracellular calcium influx through voltage-operated and receptor-operated calcium channels in vascular smooth muscle.F We found that the potency of nifedipine was greater for inhibition of potassium-induced contraction than norepinephrineinduced contraction. In addition, nifedipine caused a biphasic relaxation of norepinephrine-stimulated arteries. The low-affinity second phase of nifedipine relaxation may be caused by blockade of receptor-operated channels, whereas the high-affinity first phase may be due to blockade of voltage-dependent channels associated with norepinephrine stimulation. Verapamil was also potent
Volume 104
Inhibition of IMA contractions
Number 4 October 1992
and effective at inhibiting potassium-induced and norepinephrine-induced contraction. It did not show biphasic relaxation with norepinephrine, as did nifedipine. Nitroprusside and nitroglycerin can relax various smooth muscles including vascular, uterine, gastrointestinal, and bronchial muscles. Recent studies have found increased intracellular levels of cyclic guanosine monophosphate induced by nitroprusside and nitroglycerin23, 24; this increase in cyclic guanosine monophosphate may be responsible for relaxation of vascular smooth muscle. The mechanism of nitrate-induced relaxation is not voltage-dependent.P Our study found nitroprusside to be more potent and more effective than nitroglycerin. Although nitroglycerin has been shown to be potent at inhibiting contraction of human coronary arteries studied in vitro,26 it was not effective on the IMA in our studies. Nitroprusside had a similar potency to inhibit both potassium and norepinephrine contraction. Both nitroprusside and nitroglycerin were more effective in inhibiting norepinephrine-induced contraction than potassiuminduced contraction. The ;J-adrenergic agonist, isoproterenol, is also a potent vasodilator drug, which acts by increasing adenylate cyclase activity and causing an increase in intracellular cyclic adenosine monophosphate.I? Isoproterenol was neither potent nor effective in inhibiting contraction in our study. Isoproterenol did inhibit norepinephrine contraction slightly when used at high concentrations, possibly as a result of isoproterenol displacing norepinephrine from a-adrenergic receptors. In all cases, the arteries relaxed slowly after washout of isoproterenol, and relaxation to baseline often was incomplete. This was probably caused by persistent a-adrenergic receptor stimulation seen with high concentrations of isoproterenol.P The lack of relaxation with isoproterenol suggests that few ;J-adrenergic receptors are present in the human IMA. This is in contrast to the human coronary artery, which has more ;J-adrenergic receptors than a-adrenergic receptors.P: 29 A recent study also demonstrated weak ;J-adrenoceptormediated relaxation in the human IMA. 3o Adenosine is also a potent vasodilator, especially in the coronary circulation,25,31 which acts by causing a decrease in the membrane permeability to extracellular calcium ions" or by acting through an intracellular mechanism.V Adenosine was neither potent nor effective in inhibiting potassium-induced contraction in our experiments. Dipyridamole was added in some experiments to ensure that uptake and breakdown of adenosine were not responsible for its lack of effects. In all cases, dipyridamole had no effect on adenosine-induced relaxation. These results suggest that, unlike the coronary vascular bed, the IMA is not sensitive to adenosine relaxation.
98 I
Papaverine is a nonspecific, noncompetitive vasodilator. It inhibits oxidative phosphorylation'! but may cause vasodilation by inhibiting phosphodiesterase, which prevents breakdown of cyclic nucleotides that can lead to increased intracellular cyclic adenosine monophosphate or cyclic guanosine monophosphate concentrations.i?: 34 Papaverine was the least potent agent tested, but it was the most effective in maximally inhibiting both potassium-induced and norepinephrine-induced contraction. It was the only drug that completely reversed norepinephrine-induced contraction. Although papaverine is an effective drug, it has few therapeutic uses and may best be used by topical or intraluminal application to the IMA. Effective vasodilation with a resultant increase in IMA flow has been demonstrated when papaverine was injected intraluminally." Clinical implications. This study investigated the effectiveness of various vasodilators to inhibit potassiumand norepinephrine-induced contraction of the human IMA in vitro. Frequently, vascular spasm is the initial event that then necessitates treatment. Although we attempted to induce contraction and then relaxation of the artery with vasodilators, we found that the human IMA relaxes in a prolonged manner. We therefore accepted the limitations inherent in a pretreatment model to have more manageable data. Although papaverine is effective in preventing contraction of the IMA, it is not a clinically useful drug. It can best be used intraoperatively while the patient's chest is open. Nifedipine is a more clinically useful drug that can be used perioperatively with the chest closed. Nifedipine is potent and relatively effective in inhibiting both norepinephrine-induced and potassium-induced contraction. Nitroprusside and nitroglycerin are only moderately potent and effective. The human IMA appears to have few ;J-adrenergic receptors because it was only minimally responsive to isoproterenol. Our study therefore supports the intraoperative use of papaverine for treatment of vasospasm seen during isolation of the IMA and its pedicle. Nifedipine may be a better drug for treatment or prevention of perioperative IMA spasm. REFERENCES I. Lytle BW, Loop FD, Cosgrove DM, Ratliff NB, EasleyK, Taylor P'C, Long-term (5-12 years) serial studies of internal mammary artery and saphenousvein coronary bypass grafts. J THORAC CARDIOVASC SURG 1985;89:248-58. 2. Barner HB, Standeven JW, Reese J. Twelve-year experience with internal mammary artery for coronary artery bypass. J THORAC CARDIOVASC SURG 1985;90:668-75. 3. Green GE. Techniques of internal mammary--coronary artery anastomosis. J THORAC CARDIOVASC SURG 1979; 78:455-9.
982
Jett et al.
4. Mills NL, Bringaze WL. Preparation of the internal mammary artery graft. Which is the best method? J THORAC CARDIOVASC SURG 1989;98:73-9. 5. Johns RA, Peach MJ, Flanagan T, Kron IL. Probing ofthe canine mammary artery damages endothelium and impairs vasodilation resulting from prostacyclin and endotheliumderived relaxing factor. J THORAC CARDIOVASC SURG 1989;97:252-8. 6. Luscher TF, Diederich D, Siebenmann R, et al. Difference between endothelium-dependent relaxation in arterial and venous coronary bypass grafts. N Engl J Med 1988; 319:462-7. 7. Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal mammary artery graft on 10-year survival and other cardiac events. N Engl J Med 1986;314:1-6. 8. Tector AJ, Schmahl TM, Canino VR. Expanding the use of the internal mammary artery to improve patency in coronary artery bypass grafting. J THORAC CARDIOVASC SURG 1986;91:9-16. 9. Green GE. Coronary bypass highlights blood vessel biology. Ann Thorac Surg 1986;41:235-6. 10. Mills NL, Ochsner JL. Technique of internal mammaryto-coronary artery bypass. Ann Thorac Surg 1974;17:23746. 11. Barner HB: Double internal mammary-coronary artery bypass. Arch Surg 1974;109:627-30. 12. Tesfamarian B, Halpern W, Osol G. Effects of perfusion and endothelium on the reactivity of isolated resistance arteries. Blood Vessels 1986;22:301-5. 13. Yang Z, Diederich D, Schneider K, et al. Endothelium-derived relaxing factor and protection against contractions induced by histamine and serotonin in the human internal mammary artery and in the saphenous vein. Circulation 1989;80:1041-8. 14. Green GF. Internal mammary-coronary artery anastomosis for myocardial ischemia. In: Sabiston DC, Spencer FC, eds. Gibbon's surgery of the chest, Philadelphia: WB Saunders, 1983:1451-8. 15. Loutzenhiser R, Epstein M. Activation mechanisms of human renal artery: effects of KCI, norepinephrine, and nifedipine upon tension development and 45Ca influx. Eur J PharmacoI1984;106:47-52. 16. Hogestatt ED. Characterization of two different calcium entering pathways in small mesenteric arteries from rat. Acta Physiol Scand 1984;122:483-95. 17. Lorenz RR, Vanhoutte PM. Inhibition of adrenergic neurotransmission in isolated veins of dog by potassium ion. J Physiol 1975;246:479-500. 18. Haeusler G. Relationship between noradrenaline-induced depolarization and contraction in smooth muscle. Blood Vessels 1978;15:46-54.
The Journal of Thoracic and Cardiovascular Surgery
19. Furchgott RF, Zawadsk JV. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288:373-6. 20. Furchgott RF. Role of endothelium in responses of vascular smooth muscle. Circ Res 1983;53:557-73. 21. Oshiro MEM, Paiva AC, Paiva TB. Endothelium-dependent inhibition of the use of extracellular calcium for arterial response to vasoconstrictor agents. Gen Pharmacol 1985;16:567-72. 22. Sperelakis N, Mras S. Depression of rabbit aorta and guinea pig vena cava by mesudipine and other slow channel blockers. Blood Vessels 1983;20:172-83. 23. Bohme E, Graf H, Schultz G. Effects of sodium nitroprusside and other smooth muscle relaxants on cyclic GMP formation in smooth muscle and platelets. Adv Cyc Nucl Res 1978;9:131-43. 24. Rapaport RM, Murad F. Endothelium-dependent and nitrovasodilator-induced relaxation of vascular smooth muscle: role of cyclic GMP. J Cyc Nucl Prot Phosph Res 1983;9:281-96. 25. Schmaar RL, Sparks HV. Response of large and small arteries to nitroglycerin, NaNOz, and adenosine. Am J Physiol 1972;223:223-8. 26. Ginsburg R, Bristow MR, Harrison DC, Stinson EB. Studies with isolated human coronary arteries: some general observations, potential mediators of spasm, role of calcium antagonists. Chest 1980;78(suppl 1):180-6. 27. Triner L, Nahas GG, Overweg NIA, Verosky M, Habif DV, Ngai SH. Cyclic AMP and smooth muscle function. Ann NY Acad Sci 1971;185:458-76. 28. Dorevitch N. Effect of isoproterenol on adrenergic receptors in rabbit aorta. Arch Int Pharmacodyn Ther 1968; 174:98-107. 29. Godfraind T. Pharmacology of human coronary arteries. Acta Cardiol 1985;40:441-6. 30. He GW, Buxton B, Rosenfeldt FL, Wilson AC, Angus JA. Weak adrenoceptor-mediated relaxation in the human internal mammary artery. J THORAC CARDIOVASC SURG 1989;97:259-66. 31. Herlihy JT, Bockman EL, Berna RM, Rubio R. Adenosine relaxation of isolated vascular smooth muscle. Am J PhysioI1976;230:1239-43. 32. Verhueghe RH. Action of adenosine and adenine nucleotides on dog's isolated veins. Am J Physiol 1977; 233:HI14-21. 33. Santi R, Contessa AR, Ferrari M. Spasmolytic effect of the papaverine and inhibition of the oxidative phosphorylation. Biochem Biophys Res Commun 1963;11:156-9. 34. Triner L, Vulliemoz YU, Schwartz I, Nahas GG. Cyclic phosphodiesterase activity and the action of papaverine. Biochem Biophys Res Commun 1970;40:64-9.