T excellent conduit for coronary artery bypass grafting.

The Influence of Age and Sex on Human Internal Mammary Artery Size and Reactivity Rebecca J. Dignan, MD, Thomas Yeh, Jr, MD, Cornelius M. Dyke, MD, Harry A. Lutz 111, PhD, and Andrew S. Wechsler, MD Department of Surgery, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia Internal mammary arteries (IMAs) from women and the elderly have been postulated to be smaller and more reactive than IMAs from men and younger patients and, therefore, not as reliable for coronary artery bypass grafting in the short term. This study tests the physiologic basis for that hypothesis. Trimmed IMA segments were obtained from patients aged 5 to 76 years at coronary artery bypass grafting. Eighteen ring segments from 12 women and 35 ring segments from 17 men were mounted on a strain-gauge apparatus, and internal diameter at a transmural pressure of 1 mm Hg was determined from length-tension curves. Contractions to potassium chloride and a dose-response curve to norepinephrine or serotonin were obtained to simulate physiologic vasospasm. Sodium nitroprusside determined arterial relaxation. Linear regression was used to determine correlation of these parameters with age. Internal mammary arteries from women and men were of equal size. They had equal strength of contraction to potassium chloride and norepinephrine, but female IMAs had greater strength of contraction to serotonin. Female IMAs had weaker contraction to norepinephrine as a percent of maximum contraction to potassium chloride than IMAs from men. Internal mammary arteries from women had equal relaxation to sodium nitroprusside compared with IMAs from men. There was no correlation between age and arterial reactivity to vasoconstrictors, relaxation to sodium nitroprusside, or size. These data suggest that IMAs from women and the elderly are not more susceptible to reduction in flow due to smaller size. Postoperatively, it may be important that women be kept on platelet inhibitors because of their greater absolute contraction to serotonin and men on nitrovasodilators because of their greater relative contraction to norepinephrine. (Ann Thoruc Surg 1992;53:792-7) he internal mammary artery has been shown to be an T excellent conduit for coronary artery bypass grafting. The need for arterial revascularization has increased in women and the elderly as the frequency of coronary bypass grafting in these patients has increased. Internal mammary arteries from women and the elderly are thought to be smaller and more reactive than those from men and younger patients and, therefore, not as reliable for coronary artery bypass grafting in the short term. This becomes of particular importance when inotropic agents that possess a-adrenergic properties are required or during hypotension when neurogenic catecholamine release may occur. Frequency of vasospasm has not been tested in these subgroups. Comparison of arterial reactivity to vasoconstrictor substances such as circulating catecholamines and platelet products is important in understanding and preventing graft spasm. Canine and human internal mammary artery reactivity to various constrictor and dilator substances has been studied in the context of coronary artery bypass grafting [l-31, but little is known about differences between male and female arteries and those of older patients. This study tested the hypothesis that internal mammary arteries from women and elderly Presented at the Thirty-eighth Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 7-9, 1991. Address reprint requests to Dr Wechsler, PO Box 615, MCV Station, Richmond, VA 23298. patients are smaller and more reactive than internal mammary arteries from men and younger patients. A common vasoconstrictor circulating perioperatively, norepinephrine, and a product of platelet aggregation, serotonin, were chosen to initiate humoral, physiologic contraction. Potassium chloride was used to obtain maximum smooth muscle contraction. Vascular dilation was assessed using sodium nitroprusside because of its direct, smooth muscle action and its frequent perioperative use. Material and Methods Trimmed internal mammary artery segments (n = 53) were obtained from 12 women and 17 men between the ages of 5 and 76 years (men, 6. * 2.2 years; women, 66.1 * 2. years [not significant]) at coronary artery bypass operation. (This protocol was approved by the institutional committee on the conduct of human research and all patients gave informed consent.) The specimens were kept in iced, physiologic saline solution until studied. The solution had the following composition (in millimoles per liter): NaC1, 119; KC1, 4.6; CaCI,, 2.5; MgSO,, 1.17; NaHCO,, 25; KH,PO,, 1.17; EDTA,.27; and glucose, 11. (All chemicals were obtained from Sigma Chemical Company, St. Louis, MO.) A ring segment 3 mm in length was prepared by a transverse cut with a razor blade and mounted on two stainless steel wires in a 1992 by The Society of Thoracic Surgeons 3-4975/92/$5.

Ann Thorac Surg 1992;53792-7 DlGNAN ET AL 793 Force Transducer Oxygen Port b Fig I. (a) Organ bath with strain-gauge transducer. (b) Arterial ring segment mounted on wire hooks. (Reprinted from Dignan Rl, Yeh T Ir, Dyke CM, et al. Reactivity of gastroepiploic and internal mammary arteries. Relevance to coronary artery bypass grafting. J Thorac Cardiovasc Surg 1992;13:116-23, by permission of Mosby-Year Book, Znc.) 6-mL water-jacketed glass organ bath (Fig 1). The upper wire was attached to a Grass FT3D force transducer (Grass Instrument Company, Quincy, MA). The lower wire was fixed to a micrometer (Mitutyo, Tokyo, Japan). Ring segments were allowed to relax in the organ bath in oxygenated (95% oxygen, 5% carbon dioxide), normothermic, buffered saline solution for 6 minutes. A wide range of diameters was observed in the vessels obtained. Therefore, a normalization procedure was performed to standardize baseline resting length at which subsequent tension measurements were to be obtained. This method has been previously described for small [4] and large arteries and large veins [I, 51. The normalization procedure predicted the circumference of each vessel at a pressure of 1 mm Hg (L,,,), allowing all arteries to be studied at comparable resting conditions. Passive length-tension curves were obtained for each vessel by sequentially separating the wires using a micrometer. Force generated by the vessel was recorded on a Grass Model 7 Polygraph and the micrometer setting noted at each step. Vessel tension (T) was calculated at each step from the equation T = ForceRg where g is the length of the segment. Internal circumference (L) was determined from the equation L = (7 + 2)d + 2f where d is the wire diameter and f is the distance between the wires. To ensure that the vessel was not overstretched, pressure was calculated from the tension and internal circumference after each step. When pressure was greater than 1 mm Hg, vessel stretch was released. Nonlinear regression was used to obtain the best exponential fit of the length-tension curve. A line was generated by solving the Laplace formula (P = 27fT/L) for tension, assuming a pressure equal to 1 mm Hg T = (1 mm Hg)L/27 and plotting the tensions on the isobaric line against the internal circumference. A personal computer-based program [6] modified in our laboratory was used to determine the intersection of the length-tension curve and the line by an iterative fitting technique (Fig 2). This intersection determined the internal circumference (L) the vessel was predicted to have at a transmural pressure of 1 mm Hg and is called L,,,. The internal circumference of every vessel was set at 9% of L,,, (L,,). The isometric force at this setting is the "passive" or "resting" force in the absence of constrictor tone. This method of determining resting force allows maximum tension development [7] while standardizing resting tension independent from vessel diameter. Passive resting force was therefore determined for each vessel from its length-tension curve and not by nonspecific methods that have frequently been used in similar studies using isolated vessel rings. Arterial segments were allowed to equilibrate for 15 to 2 minutes at L, before proceeding (Fig 3). They were contracted twice using 125 mmol/l potassium chloride, rinsed, and allowed to equilibrate before and after the second contraction. Concentration-response curves were obtained (one per ring segment) to norepinephrine ( lop9 25 2 Tension 15 (rnnewtonslrnm) 1 Internal Circumference (mm) Fig 2. Normalization procedure for one arterial segment. The curved line represents the experimentally determined length-tension relationship of one vessel. It was obtained by progressively stretching the ring segment and plotting tension measurements as a function of internal circumference. The best exponential fit was obtained using nonlinear regression. The straight line is the mathematical solution of the Laplace relation where pressure is equal to 1 mm Hg. The intersection of the curve and the line determined the internal circumference the vessel would have at a transmural pressure of 1 mm Hg (L,,,J. (P = pressure; T = tension.)

794 DIGNAN ET AL Ann Thorac Surg 1992;53:792-7 Fig 3. The conduct of the experiment is illustrated in this tracing of the force generated by a typical ring segment. The ring was contracted twice using 125 mmolll potassium chloride (KCL). Half-log molar concentrations of norepinephrine (NE) starting at lo-' molll and ending at lo-" molll were used to obtain a concentration-response curve. The ring was relaxed with sodium nitroprusside (SNP). FORCE (gms) RESIN6 FORCE REACTIVITY AT Lgo t t ttt t t tt t t ttt KCL KCL -9-8 -7-6 -5-4 -4 (LogMolar) 12SmM 125mY v NE SNP TIME 4 4 min. to mol/l) or serotonin (lo-' to lo-* mol/l). Drug concentrations used started below threshold and ended when maximum contraction was achieved. Maximum smooth muscle relaxation was measured using sodium nitroprusside (lop4 mom) (see Fig 3). Tension at L, was converted to pressure using the Laplace relationship (P = 27rT/L9J. All ring segments studied were included in the statistical analysis. Two measurements quantified arterial ring segment reactivity: maximum contraction and sensitivity (Fig 4). Maximum mean contraction was calculated for each constrictor and compared between and within groups of arteries. Norepinephrine and serotonin contractions were also compared as percent of maximum potassium contraction. Sensitivity was measured by effective drug concentration producing 5% of maximum contraction (EC,,), which was calculated from concentration-response curves generated by a symmetric sigmoidal curve model for every ring segment using personal computer-based software. Average EC,,s for each constrictor were compared between groups. Mean concentration-response curves to serotonin and norepinephrine for each group of vessels were generated using mean pressures at each concentration of serotonin and norepinephrine. Relaxation to sodium nitroprusside was expressed as a percent of contraction to the drug used. The comparisons were subject to analysis of variance, accepting a value of p less than.5 1 = EC.. 12 1 = Maximum Contraction Pressure 16 4 1 8o I: -1. -9. -8. -7. -6. -5. Vasoconstrictor (Log Molar) Fig 4. Two measurements quantified arterial ring segment reactivity. Maximum contraction was the highest pressure generated by each vasoconstrictor. Sensitivity was determined by efiective concentration producing 5% of maximum contraction (EC,d. Symmetric sigmoidal curves generated for two ring segments are pictured here. The EC,, for the hoo rings is very similar but the maximum contraction is significantly difierent. as significant. Regression analysis was used to test the correlation of age with seven variables: diameter; maximum contraction to potassium, norepinephrine, and serotonin; EC,, of norepinephrine and serotonin, and relaxation to sodium nitroprusside. Results Sex Diflerences SIZE. The diameters of the ring segments from men and women, as determined by L,,,, were not significantly different (1.76?.1 versus 1.6 f.6 mm, respectively). ARTERIAL CONTRACTIONS. Internal mammary arteries from women had equal strength of contraction to potassium (115 f 13 versus 13 f 6 mm Hg) and norepinephrine (96? 7 versus 112 2 4 mm Hg) but greater strength of contraction to serotonin (71 f 15 versus 26 f 3 mm Hg) compared with internal mammary arteries from men (Fig 5a). Both internal mammary arteries had similar contractions to potassium and norepinephrine. Male internal mammary arteries had significantly weaker contractions to serotonin than to norepinephrine or potassium (see Fig 5a). Female internal mammary arteries, however, had equal contractions to serotonin and norepinephrine. Female internal mammary arteries had weaker contraction to norepinephrine (82% f 9% versus 113% f 9%) as a percent of maximum potassium contraction than the internal mammary arteries from men (Fig 5b). ARTERIAL SENSITIVITY. Concentration-response curves to norepinephrine and serotonin (Figs 6, 7) were located in the same position on the horizontal axis for both internal mammary arteries. The calculated EC,,s of serotonin (1.2 4.4 x low7 versus 1.5 2.2 x lop7 mol/l) and norepinephrine (.63 2.2 x versus 1.3?.2 X lop7 mom) were equal in female and male internal mammary arteries, indicating equal sensitivity of the arteries to those drugs. There was a significant difference in the upper portion of the serotonin curve, however, because the strength of contraction was greater in the female internal mammary arteries (see Figs 6, 7). ARTERIAL RELAXATION. The vessels had no significant difference in relaxation to sodium nitroprusside (female, 132.6% 2 14.1%; male, 117.2% * 8.1%).

Ann Thorac Surg 1992;53:792-7 DIGNAN ET AL 795 Pressure 1 5 R Y 15 1 Contraction 5 - - j ifl In li; (11.35) (n.8) (nz6) (n-9) KCL NE STN NE In= 18) 'p<.5 (Female vs. Male) "p<.5 (NE vs. STN).- -. STN Fig 5. (A) Contractile function to potassium chloride (KCL), norepinephrine (NE), and serotonin (STN) is demonstrated for female and male internal mamma y arteries. Strength of contraction to KCL and NE were equal, but contraction to STN was significantly greater in the female internal mammay arteries. Female internal mamma y arteries had similar contrac-.tions to KCL, NE, and STN. Male internal mammay arteries had significantly weaker contractions to STN than to NE or KCL (indicated with double asterisks). (B) Norepinephrine and STN contractions were normalized to the KCL contractions by calculating them as percent of KCL contraction, as is frequently done by other authors when diameter and, therefore, pressure are unknown. There is no significant difference in contraction to STN. Male internal mamma y arteries, however, had significantly weaker contractions to NE as a percent of KCL contraction. Age Differences There was no linear correlation, as determined by the? coefficient, between age and measurements of sensitivity, maximum contraction to potassium, norepinephrine, and serotonin, or relaxation to nitroprusside (Table 1). Plots of age versus maximum contraction to potassium, serotonin, or norepinephrine or EC5,, of norepinephrine or serotonin were scattered, as were those of age versus relaxation to sodium nitroprusside (Fig 8). The Llo was not significantly different as age increased. There was also no correlation between any of the variables and age when female or male subgroups were analyzed. Comment This study shows equal size, equal sensitivity to norepinephrine and serotonin, equal maximum contraction to potassium and norepinephrine, and equal relative contraction to serotonin of internal mammary arteries from men and women. The maximum contraction to serotonin and relative contraction to norepinephrine, however, are different. Serotonin is thought to cause coronary artery vasospasm [8]. It is also known to cause contraction in the human internal mammary artery [9, 11 and is likely to cause vasospasm in the internal mammary artery as well. Although the difference in reactivity of female and male internal mammary arteries to serotonin shown in this study has not been previously studied, the literature suggests interaction between estrogen and serotonin receptors. Serotonin interacts with specific serotonin receptors, probably the S, subtype, on vascular smooth muscle [ll, 121. Estradiol increases serotonin receptor binding in rats [13] and also induces uterine sensitivity to serotonin (S,-receptor mediated) [ 141. This estrogen and S,-receptor A Pressure B Percent Contraction 15 1 c. Female (n=8) 1 5 15 1 5 1-1. -9. -8. -7. -6. -5. -4. l++hdc5-1. -9. -8. -7. -6. -5. p=ns -4. Norepinephrine (Log Molar) Fig 6. Concentration-response curves to norepinephrine for female and male internal mammay arteries. (A) The curves are located at the same position on the horizontal axis, demonstrating that the effective concentration producing 5% contraction and, therefore, sensitivity to norepinephrine was equivalent in both groups. (B) When mean contractions at specific concentrations are normalized to potassium chloride contractions, there is no difference in strength of contraction to norepinephrine. (NS = not significant.)

~~ ~ ~~ 796 DIGNAN ET AL Ann Thorac Surg 199253792-7 Fig, 7. Concentration-response < curves ; to serotonin for internal mamma y arteries. (A) The curues are located at the same place on the horizontal axis, showing that the effective concentration producing 5% contraction and therefore sensitivity to serotonin was equal in female and male internal mammary arteries. (B) Mean maximum contraction and contraction as percent of potassium contraction for the highest three concentrations of serotonin for female internal mammay arteries is greater than for male internal mammay arteries. ; A loo 1 -Female (n.6). * Pressure B!! Percent Contraction ; 25-9. -8. -7. -6. -5. p<.5, 25-9. -8. -7. -6. -5. Serotonin (Log Molar) p<.5 interaction may be the mechanism for greater contraction to serotonin in internal mammary arteries from women. Further studies are required to verify this. Many patients undergoing coronary bypass are given platelet inhibitors. In this study, 1 women (55%) and 13 men (37%) were taking aspirin at the time of operation. Platelet inhibitors such as aspirin are known to inhibit the release of products of arachidonic acid metabolism. In one study, however, pretreatment of arteries with aspirin did not affect the contractile responses of rabbit coronary arteries and aorta to platelets [15]. The relation between internal mammary artery reactivity and aspirin ingestion was analyzed by comparing reactivity and relaxation of internal mammary arteries from patients receiving aspirin and those not receiving aspirin. There was no difference in any of the variables of any of the vasoconstrictors (potassium, norepinephrine, or serotonin). There was no difference in contraction to potassium in either men given aspirin or women given aspirin compared with those not given aspirin. Numbers were inadequate to compare reactivity to norepinephrine or serotonin of female and male internal mammary subgroups receiving aspirin. Studies on aortic and large artery stiffness [16] show a progressive stiffening due to structural changes in the seventh decade. There is a lower degree of stiffness in women until menopause, after which they show an accelerated stiffening equal to men in the seventh decade. Compliance factors were not analyzed in this study but may contribute to differences in reactivity in internal mammary arteries from men and women at a younger age. This study showed no correlation between age and reactivity. Studies in rats have suggested that serotonin response increases with age in coronary resistance arteries [17]. This effect was not demonstrated in this study, possibly due to low numbers or because the ring segments were taken from large arteries rather than resistance vessels. A well-known age-dependent change in vascular reac- tivity is the disappearance of preceptor-mediated relaxation of rabbit and rat aorta [18]. Increasing contraction to norepinephrine with age, if demonstrated in this study, would have indicated possible decrease in number of preceptors with age. Human internal mammary artery, however, has been shown to have only a few preceptors. Adenosine-mediated vasodilatation diminishes with age also, indicating a general loss of vasodilator potential in the aorta with age [19]. Although vasoconstrictor responses to potassium, norepinephrine, and serotonin increase in small animal aorta in the early part of life [2], in the later part, they decrease [19]. This does not appear to be true in human internal mammary artery, where there appears to be no change in contraction or relaxation with age. In conclusion, internal mammary arteries from women and the elderly are not more susceptible to reduction in flow due to size. The use of platelet inhibitors may be more important in women because of greater contraction to serotonin and nitrovasodilators in men because of greater relative contraction to norepinephrine in the perioperative interval when the internal mammary artery is Table 1. Influence of Age on Internal Mammary Artery Size and Reactivitv Variable Total Group Female Male I1.9.1.1 KCI contraction.17.18.14 NE contraction.9.25.9 STN contraction.5.3.28 SNP relaxation.18.34.3 a Regression analysis: numbers represent? values for correlation between age and variables. (p = not significant.) KCI = potassium chloride; I, = internal circumference at 1 mm Hg; NE = norepinephrine; SNP = sodium nitroprusside; STN = serotonin.

Ann Thorac Surg 1992;53:792-7 DIGNAN ET AL 797 Pressure 225 1 75. 8 : I I I i Fig 8. Plots of age versus maximum contraction to potassium of the female internal mammay artey were scattered, as were those of age versus every variable in the combined female and male group and in subgroups. used for coronary artery bypass operation. Obviously, further pharmacologic studies of receptor differences are required. Better understanding of physiologic peculiarities of the internal mammary artery may influence shortterm perioperative management. Supported by grant 5 R1 HL 2632-11 from the National Institutes of Health. References 1. 2. 3. 4. 5. He GW, Angus JA, Rosenfeldt FL. Reactivity of the canine isolated internal mammary artery, saphenous vein, and coronary artery to constrictor and dilator substances: relevance to coronary bypass graft surgery. J Cardiovasc Pharmacol 1988;12:12-22. He GW, Buxton 8, Rosenfeldt FL, Wilson AC, Angus JA. Weak padrenoceptor-mediated relaxation in the human internal mammary artery. J Thorac Cardiovasc Surg 1989;97 259-66. He GW, Rosenfeldt FL, Buxton BF, Angus JA. Reactivity of human isolated internal mammary artery to constrictor and dilator agents: implications for treatment of internal mammary artery spasm. Circulation 1989;8O(Suppl 1):141-5. Mulvany MF, Halpern W. Mechanical properties of vascular smooth muscle in situ. Nature 1976;26:617-9. Angus JA, Cocks TM, Satoh K. a,-adrenoceptors and endo- thelium-dependent relaxation in canine large arteries. Br J Pharmacol 1986;88:767-77. 6. Mulvany MJ. Procedures for investigation of small vessels using small vessel myograph. Aurhus: Department of Pharmacology, Aurhus University, Denmark, 1988. 7. Mulvany MJ, Halpern W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res 1977;41:19-26. 8. Yasue H, Omote S, Takizawa A, Nagao M. Coronary arterial spasm in ischemic heart disease and its pathogenesis. Circ Res 1983;52(Suppl 1):147-52. 9. Dignan RJ, Yeh T Jr, Dyke CM, et al. Reactivity of gastroepiploic and internal mammary arteries. Relevance to coronary artery bypass grafting. J Thorac Cardiovasc Surg 1992;13: 116-23. 1. Koike R, Suma H, Kondo D, et al. Pharmacological response of internal mammary artery and gastroepiploic artery. Ann Thorac Surg 199;5384-6. 11. Awouters F. The pharmacology of ketanserin, the first selective serotonin S,-antagonist. Drug Development Res 1985;6: 263-3. 12. Braxenor RM, Angus JA. Action of serotonin antagonists on dog coronary artery. Eur J Pharmacol 1982;81:569-76. 13. Williams J, Unhouse L. Serotonin binding sites during proestrus and following estradiol treatment. Pharmacol Biochem Behav 1989;33:615-2. 14. Campos-Lara G, Caracheo F, Valenda-Sanchez A, Ponce- Monter H. The sensitivity of rat uterus to serotonin in vitro is a late estrogenic response. Arch Invest Med 199;21:71-5. 15. Awano K, Yokoyama M, Fukuzaki H. Role of serotonin, histamine, and thromboxane A2 in platelet-induced contractions of coronary arteries and aortae from rabbits. J Cardiovasc Pharmacol 1989;13:781-92. 16. Hickler RB. Aortic and large artery stiffness: current methodology and clinical correlations. Clin Cardiol 199;13:317-22. 17. Nyborg NCB, Mikkelsen EO. Serotonin response increases with age in rat coronary resistance arteries. Cardiovasc Res 1988;22:131-7. 18. Fleich JH, Maling MH, Brodie BB. Beta-receptor activity in aorta. Circ Res 197;26:151-62. 19. Hayashi S, Toda N. Age related changes in the response of rabbit isolated aortae to vasoactive agents. Br J Pharmacol 1978;64:229-37. 2. Duckles SP, Banner W. Changes in vascular smooth muscle reactivity during development. Annu Rev Pharmacol 1984; 246.583. DISCUSSION DR TIMOTHY J. GARDNER (Baltimore, MD): That was a nice presentation. I think that it emphasizes a point that we made based on a clinical series a couple of years ago that the internal mammary artery is quite suitable for use in elderly patients and also in women. In fact, not only is the internal mammary artery probably no different anatomically or physiologically, but when compared with available vein conduits in elderly women, it may in fact be superior. One clinical point that we have noticed in elderly patients more often than in younger patients and in females more often than in males is that the distal internal mammary artery, after having been infused with papaverine, appears to be more friable and susceptible to dissection of the vessel wall. This may be a point of some technical importance, because if you have a partially dissected or friable mammary tip and your sutures include only the outer wall of the artery, the inner wall can telescope backward and result in an obstruction. DR DIGNAN: Thank you for that comment. We do not have data to support or disprove your experience. In elderly patients one would expect more atherosclerosis, which might contribute to dissection. Internal mammary arteries, however, have been shown to develop atherosclerosis infrequently. DR FREDERICK L. GROVER (Denver, CO): Did you correlate reactivity with diabetes? DR DIGNAN: No, we did not correlate reactivity with diabetes. We have information as to which patients were receiving insulin, however. A study comparing internal mammary artery reactivity of patients receiving insulin with that of those not receiving insulin would be interesting.