- Email: [email protected]
Topical Application of Amrinone (a Selective Phosphodiesterase III Inhibitor) for Relief of Vasospasm
Journal of Surgical Research 93, 149 –155 (2000) doi:10.1006/jsre.2000.5971, available online at http://www.idealibrary.com on
Topical Application of Amrinone (a Selective Phosphodiesterase III Inhibitor) for Relief of Vasospasm Shigeru Ichioka, M.D.,* ,1 Takashi Nakatsuka, M.D.,* Norihiko Ohura, M.D.,* Yuko Sato, M.D.,† and Kiyonori Harii, M.D.† *Department of Plastic and Reconstructive Surgery, Saitama Medical School, 38 Morohongo, Moroyama, Iruma-gun, Saitama, 350-0495, Japan; and †Department of Plastic and Reconstructive Surgery, Faculty of Medicine, University of Tokyo, Tokyo, Japan Submitted for publication January 31, 2000
Background. Amrinone, a selective phosphodiesterase III inhibitor, is an agent that possesses a combination of positive inotropic and vasodilating properties as a result of preventing the degradation of cAMP, and it has recently been licensed for the treatment of heart failure. The aim of this study was to investigate the potential therapeutic application of amrinone to resolve vasospasm, which is the major problem in reconstructive surgery. In this study its effect was compared with that of lidocaine, the most commonly used topical vasodilating agent clinically. Materials and methods. The probe of an ultrasonic transit-time volume flowmeter was applied to the femoral artery of rats to measure blood flow. After a baseline recording was obtained, 0.03 ml of epinephrine was applied topically to induce vasospasm. The vessels were then immersed in 1 ml of amrinone (5 mg/ml), 10% lidocaine hydrochloride, or normal physiological saline solution for 1 min in an attempt to resolve the spasm. In another group of animals, no solution was used following administration of epinephrine to allow observation of spontaneous resolution of the vasospasm over time. Results. The results showed an essentially immediate spasm-resolving effect in both the amrinone group and the lidocaine group. The amrinone group showed a significantly greater degree of maximum increase in blood flow than the lidocaine group. The effect of lidocaine decreased with time, whereas amrinone had a more lasting effect. Conclusions. The findings suggest that amrinone could be used as an effective topical vasodilating agent to resolve vasospasm in reconstructive surgery. © 2000 Academic Press
Key Words: amrinone; phosphodiesterase inhibitor; vasospasm; cAMP. 1
To whom correspondence should be addressed.
Amrinone is a selective phosphodiesterase III inhibitor that increases cAMP in both vascular smooth muscle and the myocardium by preventing its breakdown. Accumulation of cAMP induces an increase in cardiac muscle contraction and vascular smooth muscle relaxation. Because of its combined inotropic and vasodilating effects, amrinone has been developed as an agent for the treatment of heart failure. Based on its mechanism of action, we assumed that amrinone is expected to be useful not only for the treatment of heart failure but for the treatment of peripheral circulatory disorders, including vascular disease, and for ischemic flaps, by improving microcirculatory hemodynamics. To test this assumption, we conducted a study in which we employed an intravital microscopic technique and an experimental flap model, and demonstrated that amrinone significantly increased microcirculatory blood flow and ischemic flap survival [1]. In this earlier experiment, we observed such a remarkable vasodilating effect that we expected amrinone to have a substantial counteractive effect on vasospasm, the major problem in reconstructive surgery. Vasospasm is defined as the tight vasoconstriction occasionally encountered before or after vascular anastomosis that is refractory to mechanical dilation and significantly retards blood flow [2]. This condition remains a significant problem in reconstructive surgery, because it induces microvascular stasis and thrombosis and the resultant hypoperfusion contributes to ultimate failure of tissue transfers and replantations. Relief of this pathologic vasoconstriction must be rapid and efficient to allow adequate blood flow to be established and to maintain survival of the transplanted tissue. This study was designed to evaluate the efficacy of topical application of amrinone to relieve vasospasm in
149
0022-4804/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
150
JOURNAL OF SURGICAL RESEARCH: VOL. 93, NO. 1, SEPTEMBER 2000
FIG. 1. The femoral artery was carefully dissected from the vein and nerve under an operating microscope. The probe of an ultrasonic transit-time volume flowmeter was applied to the artery to measure blood flow. Note the loose fit of the probe, which allows changes in vessel diameter during measurement.
a rat experimental model. Its effect was compared with that of lidocaine, the most commonly used clinical topical vasodilating agent. MATERIALS AND METHODS
Effects on Experimental Vasospasm The femoral artery of 70 male Wistar rats weighing 400 – 425 g was used to estimate the effect of the topical agents. The animals were anesthetized with an intraperitoneal injection of urethane (1 g/kg) and placed on a warming blanket (Homoethermic Blanket Control, KN-474, Natume), the temperature of which was kept at 36°C. Room temperature was maintained at a constant 25°C. Surgical procedure. An incision was made in the right groin, and the femoral artery was dissected carefully from the vein and nerve under an operating microscope (Carl Zeiss, Inc.). The proximal limit of the dissection was the inferiomedial branch, which the femoral artery always gives off toward the inguinal ligament. The distal limit was the area of the bifurcation of the inferior epigastric artery. The periadventitial tissue was dissected free for this length of vessel and the probe of an ultrasonic transit-time volume flowmeter (Transonic Systems, Inc.) was applied to the artery to measure blood flow (Fig. 1). Experimental protocol. After a 60-min stabilization period, baseline blood flow was recorded before topically applying a drop of 1:1000 epinephrine (0.03 ml) to induce vasospasm. The time when the epinephrine was applied was recorded as 0 min, and blood flow at 0 min was regarded as baseline flow. At 2 min the vessels were immersed in 1 ml of solution for 1 min in an attempt to resolve the spasm; the solution was removed by gentle application of a gauze sponge. Flow was monitored at 1- or 2-min intervals, and a final measurement was made 21 min after the application of epinephrine. The solutions used were commercial preparations of amrinone (5 mg/ml), 10% lidocaine hydrochloride (100 mg/ml), and normal physiological saline. The temperature of the solutions was kept at 34°C with a constant-temperature water bath before application. The amrinone-, lidocaine-, and saline-treated groups consisted of 20, 18, and 17 animals, respectively. In addition to these three
groups, the arteries of another group of animals (n ⫽ 15) were not exposed to solution following administration of epinephrine to allow observation of the course of spontaneous resolution over time.
Morphological Changes Associated with Drug Treatment To assess any morphological changes that might occur as a result of application of the vasodilatory drugs, both femoral arteries of 12 additional animals were exposed in the same manner as in the above experiment. Both arteries in each animal were exposed to either amrinone or lidocaine for 10, 30, or 60 min (n ⫽ 3 for each group). Following immersion, arteries were fixed with 10% buffered formaldehyde. Arteries from another group of rats (n ⫽ 3) were fixed 5 days later after immersion for 60 min. The specimens were resectioned in a routine manner for histologic examination and stained with hematoxylin and eosin. All experiments were conducted in accordance with the standards in The Guide for Animal Experimentation controlled by the Animal Research Committee, Saitama Medical School.
Statistical Analysis All data are summarized as means ⫾ SE. Changes in blood flow with time within each group were evaluated by comparing blood flow between individual points in time. Differences in blood flow among the four groups at the same point in time were also estimated. In both comparisons, analysis of variance was used to test for statistically significant differences and was followed by application of the Tukey method for multiple comparisons. Values of P ⬍ 0.05 were considered significant.
RESULTS
Effects on Experimental Vasospasm Topical epinephrine produced an immediate and reproducible decrease in blood flow within 1 min. Values
151
ICHIOKA ET AL.: AMRINONE FOR RELIEF OF VASOSPASM
FIG. 2. Changes in blood flow with time were evaluated in four groups. * Blood flow in the amrinone group was significantly greater than in the saline and no solution groups. ** Blood flow in the amrinone group was significantly greater than in the lidocaine, saline, and no solution groups. # Blood flow in the lidocaine group was significantly greater than in the no solution group.
at 2 min were the minimum during the time course in all groups (Fig. 2). Time course in each group. In the no solution group, blood flow slowly recovered from the minimum value at 2 min, but there was not a statistically significant return to the baseline flow level (before epinephrine application) at any time during the test period. In the saline group, blood flow gradually increased following immersion, and reached a level that was statistically identical to the baseline at 9 min. In the lidocaine and amrinone groups, blood flow rapidly increased after immersion, and blood flow at 4 min was statistically equivalent to the baseline flow levels in both groups. Comparison among groups. There were no significant differences among the four groups at 0 – 4 min. In the amrinone group, blood flow increased to a significantly greater level at 5 min than in the saline and no solution groups. At 7–19 min flow values were significantly greater than in the other three groups. Although flow at 21 min showed no significant difference compared with the lidocaine group, it was still significantly greater than in the saline and no solution groups. In the lidocaine group, flow values at 5–9 min were only significantly greater than in the no solution group. Flow in the lidocaine group tended to exceed flow in the
saline group at every time point, but the changes did not reach statistical significance. Table 1 demonstrates absolute blood flow values (means ⫾ SE) expressed in milliliters per minute in the amrinone and lidocaine groups to conveniently compare the effects in these two groups. P values derived from the multiple comparisons are also included in Table 1. At 7–19 min blood flow values in the amrinone group significantly exceeded those in the lidocaine
TABLE 1 Blood Flow in Amrinone and Lidocaine Groups Blood flow (ml/min) Time after epinephrine application (min)
Amrinone group
Lidocaine group
P
4 5 7 9 11 13 15 17 19 21
2.40 ⫾ 0.72 2.99 ⫾ 0.63 3.72 ⫾ 0.36 3.53 ⫾ 0.22 3.40 ⫾ 0.27 3.30 ⫾ 0.34 3.21 ⫾ 0.35 3.25 ⫾ 0.40 3.32 ⫾ 0.45 3.32 ⫾ 0.31
2.02 ⫾ 0.45 2.64 ⫾ 0.51 2.52 ⫾ 0.34 2.34 ⫾ 0.26 2.16 ⫾ 0.41 2.10 ⫾ 0.25 2.05 ⫾ 0.28 2.07 ⫾ 0.29 2.13 ⫾ 0.31 2.22 ⫾ 0.26
NS NS ⬍0.05 ⬍0.01 ⬍0.01 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05 NS
152
FIG. 3.
JOURNAL OF SURGICAL RESEARCH: VOL. 93, NO. 1, SEPTEMBER 2000
Photomicrograph of the artery soaked in amrinone for 10 min. The structure of the vessel wall has remained intact. Bar ⫽ 50 m.
group (P ⬍ 0.05). The difference was highly significant (P ⬍ 0.01) at 9 –11 min. There were no significant differences between the saline group and the no solution group at any time during the period of observation. Morphological Changes Associated with Drug Treatment Arteries examined 10 and 30 min after immersion in amrinone or lidocaine were essentially normal (Fig. 3). Exposure of arteries for 60 min caused patchy endothelial loss and occasional breaks and sparseness in the external layer. The elastic lamina was for the most part intact. There was no evidence of medial damage. The severity of the changes induced by amrinone and was not different from that for lidocaine (Fig. 4). Examination 5 days after 60 min of immersion showed a regenerated endothelial layer in both the amrinone group and the lidocaine group. Although there were inflammatory cells in the perivascular area, extensive damage was not evident (Fig. 5). DISCUSSION
Maintenance of sufficient blood flow at the anastomotic site of the vessel is essential in reconstructive microvascular surgery because retardation in blood flow may initiate microvascular stasis, which then results in thrombosis and ultimate failure of free-tissue transfers. Immediate and efficient relief of the vasospasm is needed to establish rapid flow and maintain it after anastomosis. Therefore topical agents applied to
the vessels for the purpose of acute resolution of the intraoperative spasm have been the subject of many experimental studies. Lidocaine, a potent local anesthetic, has been widely used topically to treat vasospasm in microvascular surgery because of its rapid action and availability. Its efficiency has been well demonstrated by different authors [3– 8] and it has been successfully used in most of the cases in our institution. However, we occasionally encounter severe vasospasm that is resistant to topical lidocaine application, and despite a variety of attempts to resolve spasm, more attractive agents are anticipated. This work was directed toward exploring the acute spasmolytic effect of amrinone as a candidate for a more effective agent to allow maximal blood flow and prevent vasospasm. We used amrinone solution (5 mg/ml), which is commercially available for treatment of heart failure. To compare the effect, blood flow was also measured in 10% lidocaine-treated animals. The effectiveness of lidocaine is known to be strongly related to the concentration used [3, 4, 6, 7]. Some investigators recommended 20% lidocaine as the optimal concentration [3, 8]. In contrast, it has been pointed out that toxicity may develop clinically if a high concentration of lidocaine is applied on a vessel [7]. Beekman et al. reported that 12% lidocaine is the best possible choice with the least potential toxicity and the maximum effect [5, 6]. In Japan 10% is the highest concentration of lidocaine that can be obtained as a commercial preparation. Our clinical experience using 10% lidocaine in approximately 2600 cases has indicated satisfactory relief of vasospasm in most instances. To get practical information to surgeons, we have evaluated the effects of com-
ICHIOKA ET AL.: AMRINONE FOR RELIEF OF VASOSPASM
153
FIG. 4. Photomicrographs of arteries immersed in (A) amrinone and (B) lidocaine for 60 min. Patchy endothelial loss and occasional breaks and sparseness in the adventitial layer are seen. The elastic laminae remain intact. There is no evidence of medial damage. The degree of the change is no different between the amrinone group and lidocaine group. Bar ⫽ 50 m.
mercial preparations of the drugs in this comparative study. A suitable procedure for inducing experimental spasm is required to investigate vasospasm. There have been experimental models where external force was directly applied to the vessels of animals [9]. In our pilot study, we initially tried to induce vasospasm by various types of surgical trauma to mimic clinical situations. However, as pointed out earlier [9], reproducibility was quite difficult by means of physical damage to the vessels. Thus we employed a vasoactive substance to induce constant vasospasm reproducibly. Although the mechanism of vasospasm is not completely understood, one of the most potent stimuli of periph-
eral vasoconstriction is circulating catecholamines [10]. Their blood levels are definitely increased by the stimulation of surgical trauma and postoperative pain. It has previously been confirmed that topical epinephrine produces sustained tight constriction of vessels much more consistently than other agents, such as trauma and cold [8]. The use of topical epinephrine to induce vasospasm in this study was convenient and provided a reproducible model as in previous reports [2, 8, 11]. Not until recently have technical advances made it possible to directly measure absolute blood flow in small vessels. Many of the previous experiments on vasospasm have measured vessel diameter as a means
154
JOURNAL OF SURGICAL RESEARCH: VOL. 93, NO. 1, SEPTEMBER 2000
FIG. 5. Photomicrographs taken at 5 days after 60 min of immersion in (A) amrinone and (B) lidocaine. Inflammatory cells are present in the outer layer of the vessels. Intact endothelial layers are observed. Extensive damage is not evident. Bar ⫽ 50 m.
of estimating flow in vessels [3, 8, 11, 12]. Some investigators have measured blood flow with an electromagnetic flowmeter [6, 13], but it has certain disadvantages. First, the electromagnetic flowmeter can measure relative changes in blood flow, but difficulties with zero offset and instability limit its accuracy. Second, electromagnetic probes are size specific. They must fit snugly around the artery, and the vessel is constricted when the probe is applied [13]. We therefore assumed that this type of probe was not suitable for experiments on vasospasm because vessel diameter varies during the period of observation. We employed an ultrasonic transit-time flowmeter which was recently developed. Offset and zero adjustment inaccuracies are not apparent with the ultrasonic
probe. The ultrasonic flowmeter measures blood flow from the difference in transit time of ultrasonic waves between an upstream and a downstream signal [14]. Because this method measures flow directly and is independent of vessel diameter, the probes are designed to fit loosely around artery (Fig. 1). The loose fit capability is an advantage in experiments in which vessel size changes during measurement. The results of this study showed essentially immediate effects in both the amrinone group and the lidocaine group. The amrinone group exhibited a significantly greater degree of maximum increase in blood flow than the lidocaine group. Lidocaine’s efficacy decreases with time, whereas amrinone seems to have a more lasting effect.
ICHIOKA ET AL.: AMRINONE FOR RELIEF OF VASOSPASM
Since we employed only one concentration of the solution for each drug, it is difficult to decide which of the two is better. To determine the difference in concentrations, we calculated their molarity and obtained 26.7 mmol/liter for amrinone and 426.7 mmol/liter for lidocaine. The results seem to suggest that the action of amrinone in resolving vasospasm is more potent than that of lidocaine. Amrinone may be superior to lidocaine because amrinone is more involved in the essential pathway of contractile mechanisms than lidocaine. However, the precise mechanisms of these agents’ activities remain to be elucidated. Local anesthetics, including lidocaine, act on cell membranes. They are known to influence ion influx in cell membranes, but the specific mechanism of their vasodilating effect is unknown [8]. Amrinone has been reported to cause vasodilation by both endothelium-dependent and -independent mechanisms. A recent experiment using ring segments of rat aorta demonstrated that amrinone enhances the release of endothelium-derived relaxing factor/nitric oxide, which is currently recognized as the most important factor in vasodilation [15]. On the other hand, other research using human small arteries indicated that the relaxant properties of amrinone are not dependent on nitric oxide release [16]. Clearly, further studies are required to determine the exact nature of the amrinone effect. Histologic examination of the arterial wall revealed limited endothelial loss and degeneration in the external layer when immersed in the drugs for up to 60 min. These acute minor changes were not evident in the specimen excised 5 days later. The degree of the changes was no different in the amrinone and lidocaine groups. The toxicity of these agents when applied topically to vessels does not seem to be a serious problem. The results of this experiment suggest that amrinone can be used effectively as a topical vasodilating agent in free or pedicled flap transfer. Further clinical studies are indicated to evaluate the significance of this drug.
hibitor, improves microcirculation and flap survival: A comparative study with prostaglandin E1. J. Surg. Res. 75: 42, 1998. 2.
Puckett, C. L., Winters, R. R., Geter, R. K., and Goebel, D. Studies of pathologic vasoconstriction (vasospasm) in microvascular surgery. J. Hand Surg. [Am.] 10: 343, 1985.
3.
Ohta, I., Kawai, H., Kawabata, H., Masada, K., and Ono, K. Topical use of xylocaine for relieving vasospasm: Effect of concentration. J. Reconstr. Microsurg. 7: 205, 1991.
4.
Chafin, J. B., Wax, M. K., Johnstone, R., and Bishop, D. The use of lidocaine in microvascular reconstruction. Otolaryngol. Head Neck Surg. 117: 93, 1997.
5.
Beekman, W. H., Sluimers, J. E., Kort, W. J., and van der Meulen, J. C. Resolution of experimental microvascular vasoconstriction in rats by topical application of lidocaine hydrochloride in varying concentrations. Ann. Plast. Surg. 21: 570, 1988.
6.
Beekman, W. H., Dammeijer, P. F., Sluimers, J. E., Kort, W. J., and van der Meulen, J. H. Improvement in blood flow and diameter of the postanastomotic rat tail artery by topical application of lidocaine in varying concentrations. Ann. Plast. Surg. 24: 248, 1990.
7.
Hou, S. M., Liu, T. K., and Yu, H. Y. Absorption of lidocaine following topical application in microvascular procedures on rabbits. J. Orthop. Res. 9: 545, 1991.
8.
Geter, R. K., Winters, R. R., and Puckett, C. L. Resolution of experimental microvascular spasm and improvement in anastomotic patency by direct topical agent application. Plast. Reconstr. Surg. 77: 105, 1986.
9.
Wadstrom, J. Studies on traumatic vasospasm in the central ear artery of the rabbit. Scand. J. Plast. Reconstr. Surg. Hand Surg. Suppl. 21: 1, 1990.
10.
Richards, R. R., Seaber, A. V., and Urbaniak, J. R. Chemically induced vasospasm: The effect of ischemia, vessel occlusion, and adrenergic blockade. Plast. Reconstr. Surg. 75: 238, 1985.
11.
Pederson, W. C., Singer, D. I., Angel, M. F., Mahoney, J. E., Mitchell, G. M., Morrison, W. A., and O’Brien, B. M. Hydrostatic distention and pharmacological treatment of epinephrine-induced microarterial spasm. Ann. Plast. Surg. 23: 17, 1989.
12.
Kerschner, J. E., and Futran, N. D. The effect of topical vasodilating agents on microvascular vessel diameter in the rat model. Laryngoscope 106: 1429, 1996.
13.
Seaber, A. V. Experimental vasospasm. Microsurgery 8: 234, 1987.
14.
Welch, W. J., Deng, X., Snellen, H., and Wilcox, C. S. Validation of miniature ultrasonic transit-time flow probes for measurement of renal blood flow in rats. Am. J. Physiol. 268: F175, 1995.
15.
Mori, K., Takeuchi, S., Moritoki, H., Tsuchiya, K., Nakaya, Y., Matsuoka, S., and Kuroda, Y. Endothelium-dependent relaxation of rat thoracic aorta by amrinone-induced nitric oxide release. Eur. Heart J. 17: 308, 1996.
16.
Vroom, M. B., Pfaffendorf, M., van Wezel, H. B., and van Zwieten, P. A. Effect of phosphodiesterase inhibitors on human arteries in vitro. Br. J. Anaesth. 76: 122, 1996.
ACKNOWLEDGMENT We thank Naomi Sekiya for her careful assistance with the experimental studies.
REFERENCES 1.
Ichioka, S., Nakatsuka, T., Sato, Y., Shibata, M., Kamiya, A., and Harii, K. Amrinone, a selective phosphodiesterase III in-
155
Topical Application of Amrinone (a Selective Phosphodiesterase III Inhibitor) for Relief of Vasospasm Shigeru Ichioka, M.D.,* ,1 Takashi Nakatsuka, M.D.,* Norihiko Ohura, M.D.,* Yuko Sato, M.D.,† and Kiyonori Harii, M.D.† *Department of Plastic and Reconstructive Surgery, Saitama Medical School, 38 Morohongo, Moroyama, Iruma-gun, Saitama, 350-0495, Japan; and †Department of Plastic and Reconstructive Surgery, Faculty of Medicine, University of Tokyo, Tokyo, Japan Submitted for publication January 31, 2000
Background. Amrinone, a selective phosphodiesterase III inhibitor, is an agent that possesses a combination of positive inotropic and vasodilating properties as a result of preventing the degradation of cAMP, and it has recently been licensed for the treatment of heart failure. The aim of this study was to investigate the potential therapeutic application of amrinone to resolve vasospasm, which is the major problem in reconstructive surgery. In this study its effect was compared with that of lidocaine, the most commonly used topical vasodilating agent clinically. Materials and methods. The probe of an ultrasonic transit-time volume flowmeter was applied to the femoral artery of rats to measure blood flow. After a baseline recording was obtained, 0.03 ml of epinephrine was applied topically to induce vasospasm. The vessels were then immersed in 1 ml of amrinone (5 mg/ml), 10% lidocaine hydrochloride, or normal physiological saline solution for 1 min in an attempt to resolve the spasm. In another group of animals, no solution was used following administration of epinephrine to allow observation of spontaneous resolution of the vasospasm over time. Results. The results showed an essentially immediate spasm-resolving effect in both the amrinone group and the lidocaine group. The amrinone group showed a significantly greater degree of maximum increase in blood flow than the lidocaine group. The effect of lidocaine decreased with time, whereas amrinone had a more lasting effect. Conclusions. The findings suggest that amrinone could be used as an effective topical vasodilating agent to resolve vasospasm in reconstructive surgery. © 2000 Academic Press
Key Words: amrinone; phosphodiesterase inhibitor; vasospasm; cAMP. 1
To whom correspondence should be addressed.
Amrinone is a selective phosphodiesterase III inhibitor that increases cAMP in both vascular smooth muscle and the myocardium by preventing its breakdown. Accumulation of cAMP induces an increase in cardiac muscle contraction and vascular smooth muscle relaxation. Because of its combined inotropic and vasodilating effects, amrinone has been developed as an agent for the treatment of heart failure. Based on its mechanism of action, we assumed that amrinone is expected to be useful not only for the treatment of heart failure but for the treatment of peripheral circulatory disorders, including vascular disease, and for ischemic flaps, by improving microcirculatory hemodynamics. To test this assumption, we conducted a study in which we employed an intravital microscopic technique and an experimental flap model, and demonstrated that amrinone significantly increased microcirculatory blood flow and ischemic flap survival [1]. In this earlier experiment, we observed such a remarkable vasodilating effect that we expected amrinone to have a substantial counteractive effect on vasospasm, the major problem in reconstructive surgery. Vasospasm is defined as the tight vasoconstriction occasionally encountered before or after vascular anastomosis that is refractory to mechanical dilation and significantly retards blood flow [2]. This condition remains a significant problem in reconstructive surgery, because it induces microvascular stasis and thrombosis and the resultant hypoperfusion contributes to ultimate failure of tissue transfers and replantations. Relief of this pathologic vasoconstriction must be rapid and efficient to allow adequate blood flow to be established and to maintain survival of the transplanted tissue. This study was designed to evaluate the efficacy of topical application of amrinone to relieve vasospasm in
149
0022-4804/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
150
JOURNAL OF SURGICAL RESEARCH: VOL. 93, NO. 1, SEPTEMBER 2000
FIG. 1. The femoral artery was carefully dissected from the vein and nerve under an operating microscope. The probe of an ultrasonic transit-time volume flowmeter was applied to the artery to measure blood flow. Note the loose fit of the probe, which allows changes in vessel diameter during measurement.
a rat experimental model. Its effect was compared with that of lidocaine, the most commonly used clinical topical vasodilating agent. MATERIALS AND METHODS
Effects on Experimental Vasospasm The femoral artery of 70 male Wistar rats weighing 400 – 425 g was used to estimate the effect of the topical agents. The animals were anesthetized with an intraperitoneal injection of urethane (1 g/kg) and placed on a warming blanket (Homoethermic Blanket Control, KN-474, Natume), the temperature of which was kept at 36°C. Room temperature was maintained at a constant 25°C. Surgical procedure. An incision was made in the right groin, and the femoral artery was dissected carefully from the vein and nerve under an operating microscope (Carl Zeiss, Inc.). The proximal limit of the dissection was the inferiomedial branch, which the femoral artery always gives off toward the inguinal ligament. The distal limit was the area of the bifurcation of the inferior epigastric artery. The periadventitial tissue was dissected free for this length of vessel and the probe of an ultrasonic transit-time volume flowmeter (Transonic Systems, Inc.) was applied to the artery to measure blood flow (Fig. 1). Experimental protocol. After a 60-min stabilization period, baseline blood flow was recorded before topically applying a drop of 1:1000 epinephrine (0.03 ml) to induce vasospasm. The time when the epinephrine was applied was recorded as 0 min, and blood flow at 0 min was regarded as baseline flow. At 2 min the vessels were immersed in 1 ml of solution for 1 min in an attempt to resolve the spasm; the solution was removed by gentle application of a gauze sponge. Flow was monitored at 1- or 2-min intervals, and a final measurement was made 21 min after the application of epinephrine. The solutions used were commercial preparations of amrinone (5 mg/ml), 10% lidocaine hydrochloride (100 mg/ml), and normal physiological saline. The temperature of the solutions was kept at 34°C with a constant-temperature water bath before application. The amrinone-, lidocaine-, and saline-treated groups consisted of 20, 18, and 17 animals, respectively. In addition to these three
groups, the arteries of another group of animals (n ⫽ 15) were not exposed to solution following administration of epinephrine to allow observation of the course of spontaneous resolution over time.
Morphological Changes Associated with Drug Treatment To assess any morphological changes that might occur as a result of application of the vasodilatory drugs, both femoral arteries of 12 additional animals were exposed in the same manner as in the above experiment. Both arteries in each animal were exposed to either amrinone or lidocaine for 10, 30, or 60 min (n ⫽ 3 for each group). Following immersion, arteries were fixed with 10% buffered formaldehyde. Arteries from another group of rats (n ⫽ 3) were fixed 5 days later after immersion for 60 min. The specimens were resectioned in a routine manner for histologic examination and stained with hematoxylin and eosin. All experiments were conducted in accordance with the standards in The Guide for Animal Experimentation controlled by the Animal Research Committee, Saitama Medical School.
Statistical Analysis All data are summarized as means ⫾ SE. Changes in blood flow with time within each group were evaluated by comparing blood flow between individual points in time. Differences in blood flow among the four groups at the same point in time were also estimated. In both comparisons, analysis of variance was used to test for statistically significant differences and was followed by application of the Tukey method for multiple comparisons. Values of P ⬍ 0.05 were considered significant.
RESULTS
Effects on Experimental Vasospasm Topical epinephrine produced an immediate and reproducible decrease in blood flow within 1 min. Values
151
ICHIOKA ET AL.: AMRINONE FOR RELIEF OF VASOSPASM
FIG. 2. Changes in blood flow with time were evaluated in four groups. * Blood flow in the amrinone group was significantly greater than in the saline and no solution groups. ** Blood flow in the amrinone group was significantly greater than in the lidocaine, saline, and no solution groups. # Blood flow in the lidocaine group was significantly greater than in the no solution group.
at 2 min were the minimum during the time course in all groups (Fig. 2). Time course in each group. In the no solution group, blood flow slowly recovered from the minimum value at 2 min, but there was not a statistically significant return to the baseline flow level (before epinephrine application) at any time during the test period. In the saline group, blood flow gradually increased following immersion, and reached a level that was statistically identical to the baseline at 9 min. In the lidocaine and amrinone groups, blood flow rapidly increased after immersion, and blood flow at 4 min was statistically equivalent to the baseline flow levels in both groups. Comparison among groups. There were no significant differences among the four groups at 0 – 4 min. In the amrinone group, blood flow increased to a significantly greater level at 5 min than in the saline and no solution groups. At 7–19 min flow values were significantly greater than in the other three groups. Although flow at 21 min showed no significant difference compared with the lidocaine group, it was still significantly greater than in the saline and no solution groups. In the lidocaine group, flow values at 5–9 min were only significantly greater than in the no solution group. Flow in the lidocaine group tended to exceed flow in the
saline group at every time point, but the changes did not reach statistical significance. Table 1 demonstrates absolute blood flow values (means ⫾ SE) expressed in milliliters per minute in the amrinone and lidocaine groups to conveniently compare the effects in these two groups. P values derived from the multiple comparisons are also included in Table 1. At 7–19 min blood flow values in the amrinone group significantly exceeded those in the lidocaine
TABLE 1 Blood Flow in Amrinone and Lidocaine Groups Blood flow (ml/min) Time after epinephrine application (min)
Amrinone group
Lidocaine group
P
4 5 7 9 11 13 15 17 19 21
2.40 ⫾ 0.72 2.99 ⫾ 0.63 3.72 ⫾ 0.36 3.53 ⫾ 0.22 3.40 ⫾ 0.27 3.30 ⫾ 0.34 3.21 ⫾ 0.35 3.25 ⫾ 0.40 3.32 ⫾ 0.45 3.32 ⫾ 0.31
2.02 ⫾ 0.45 2.64 ⫾ 0.51 2.52 ⫾ 0.34 2.34 ⫾ 0.26 2.16 ⫾ 0.41 2.10 ⫾ 0.25 2.05 ⫾ 0.28 2.07 ⫾ 0.29 2.13 ⫾ 0.31 2.22 ⫾ 0.26
NS NS ⬍0.05 ⬍0.01 ⬍0.01 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05 NS
152
FIG. 3.
JOURNAL OF SURGICAL RESEARCH: VOL. 93, NO. 1, SEPTEMBER 2000
Photomicrograph of the artery soaked in amrinone for 10 min. The structure of the vessel wall has remained intact. Bar ⫽ 50 m.
group (P ⬍ 0.05). The difference was highly significant (P ⬍ 0.01) at 9 –11 min. There were no significant differences between the saline group and the no solution group at any time during the period of observation. Morphological Changes Associated with Drug Treatment Arteries examined 10 and 30 min after immersion in amrinone or lidocaine were essentially normal (Fig. 3). Exposure of arteries for 60 min caused patchy endothelial loss and occasional breaks and sparseness in the external layer. The elastic lamina was for the most part intact. There was no evidence of medial damage. The severity of the changes induced by amrinone and was not different from that for lidocaine (Fig. 4). Examination 5 days after 60 min of immersion showed a regenerated endothelial layer in both the amrinone group and the lidocaine group. Although there were inflammatory cells in the perivascular area, extensive damage was not evident (Fig. 5). DISCUSSION
Maintenance of sufficient blood flow at the anastomotic site of the vessel is essential in reconstructive microvascular surgery because retardation in blood flow may initiate microvascular stasis, which then results in thrombosis and ultimate failure of free-tissue transfers. Immediate and efficient relief of the vasospasm is needed to establish rapid flow and maintain it after anastomosis. Therefore topical agents applied to
the vessels for the purpose of acute resolution of the intraoperative spasm have been the subject of many experimental studies. Lidocaine, a potent local anesthetic, has been widely used topically to treat vasospasm in microvascular surgery because of its rapid action and availability. Its efficiency has been well demonstrated by different authors [3– 8] and it has been successfully used in most of the cases in our institution. However, we occasionally encounter severe vasospasm that is resistant to topical lidocaine application, and despite a variety of attempts to resolve spasm, more attractive agents are anticipated. This work was directed toward exploring the acute spasmolytic effect of amrinone as a candidate for a more effective agent to allow maximal blood flow and prevent vasospasm. We used amrinone solution (5 mg/ml), which is commercially available for treatment of heart failure. To compare the effect, blood flow was also measured in 10% lidocaine-treated animals. The effectiveness of lidocaine is known to be strongly related to the concentration used [3, 4, 6, 7]. Some investigators recommended 20% lidocaine as the optimal concentration [3, 8]. In contrast, it has been pointed out that toxicity may develop clinically if a high concentration of lidocaine is applied on a vessel [7]. Beekman et al. reported that 12% lidocaine is the best possible choice with the least potential toxicity and the maximum effect [5, 6]. In Japan 10% is the highest concentration of lidocaine that can be obtained as a commercial preparation. Our clinical experience using 10% lidocaine in approximately 2600 cases has indicated satisfactory relief of vasospasm in most instances. To get practical information to surgeons, we have evaluated the effects of com-
ICHIOKA ET AL.: AMRINONE FOR RELIEF OF VASOSPASM
153
FIG. 4. Photomicrographs of arteries immersed in (A) amrinone and (B) lidocaine for 60 min. Patchy endothelial loss and occasional breaks and sparseness in the adventitial layer are seen. The elastic laminae remain intact. There is no evidence of medial damage. The degree of the change is no different between the amrinone group and lidocaine group. Bar ⫽ 50 m.
mercial preparations of the drugs in this comparative study. A suitable procedure for inducing experimental spasm is required to investigate vasospasm. There have been experimental models where external force was directly applied to the vessels of animals [9]. In our pilot study, we initially tried to induce vasospasm by various types of surgical trauma to mimic clinical situations. However, as pointed out earlier [9], reproducibility was quite difficult by means of physical damage to the vessels. Thus we employed a vasoactive substance to induce constant vasospasm reproducibly. Although the mechanism of vasospasm is not completely understood, one of the most potent stimuli of periph-
eral vasoconstriction is circulating catecholamines [10]. Their blood levels are definitely increased by the stimulation of surgical trauma and postoperative pain. It has previously been confirmed that topical epinephrine produces sustained tight constriction of vessels much more consistently than other agents, such as trauma and cold [8]. The use of topical epinephrine to induce vasospasm in this study was convenient and provided a reproducible model as in previous reports [2, 8, 11]. Not until recently have technical advances made it possible to directly measure absolute blood flow in small vessels. Many of the previous experiments on vasospasm have measured vessel diameter as a means
154
JOURNAL OF SURGICAL RESEARCH: VOL. 93, NO. 1, SEPTEMBER 2000
FIG. 5. Photomicrographs taken at 5 days after 60 min of immersion in (A) amrinone and (B) lidocaine. Inflammatory cells are present in the outer layer of the vessels. Intact endothelial layers are observed. Extensive damage is not evident. Bar ⫽ 50 m.
of estimating flow in vessels [3, 8, 11, 12]. Some investigators have measured blood flow with an electromagnetic flowmeter [6, 13], but it has certain disadvantages. First, the electromagnetic flowmeter can measure relative changes in blood flow, but difficulties with zero offset and instability limit its accuracy. Second, electromagnetic probes are size specific. They must fit snugly around the artery, and the vessel is constricted when the probe is applied [13]. We therefore assumed that this type of probe was not suitable for experiments on vasospasm because vessel diameter varies during the period of observation. We employed an ultrasonic transit-time flowmeter which was recently developed. Offset and zero adjustment inaccuracies are not apparent with the ultrasonic
probe. The ultrasonic flowmeter measures blood flow from the difference in transit time of ultrasonic waves between an upstream and a downstream signal [14]. Because this method measures flow directly and is independent of vessel diameter, the probes are designed to fit loosely around artery (Fig. 1). The loose fit capability is an advantage in experiments in which vessel size changes during measurement. The results of this study showed essentially immediate effects in both the amrinone group and the lidocaine group. The amrinone group exhibited a significantly greater degree of maximum increase in blood flow than the lidocaine group. Lidocaine’s efficacy decreases with time, whereas amrinone seems to have a more lasting effect.
ICHIOKA ET AL.: AMRINONE FOR RELIEF OF VASOSPASM
Since we employed only one concentration of the solution for each drug, it is difficult to decide which of the two is better. To determine the difference in concentrations, we calculated their molarity and obtained 26.7 mmol/liter for amrinone and 426.7 mmol/liter for lidocaine. The results seem to suggest that the action of amrinone in resolving vasospasm is more potent than that of lidocaine. Amrinone may be superior to lidocaine because amrinone is more involved in the essential pathway of contractile mechanisms than lidocaine. However, the precise mechanisms of these agents’ activities remain to be elucidated. Local anesthetics, including lidocaine, act on cell membranes. They are known to influence ion influx in cell membranes, but the specific mechanism of their vasodilating effect is unknown [8]. Amrinone has been reported to cause vasodilation by both endothelium-dependent and -independent mechanisms. A recent experiment using ring segments of rat aorta demonstrated that amrinone enhances the release of endothelium-derived relaxing factor/nitric oxide, which is currently recognized as the most important factor in vasodilation [15]. On the other hand, other research using human small arteries indicated that the relaxant properties of amrinone are not dependent on nitric oxide release [16]. Clearly, further studies are required to determine the exact nature of the amrinone effect. Histologic examination of the arterial wall revealed limited endothelial loss and degeneration in the external layer when immersed in the drugs for up to 60 min. These acute minor changes were not evident in the specimen excised 5 days later. The degree of the changes was no different in the amrinone and lidocaine groups. The toxicity of these agents when applied topically to vessels does not seem to be a serious problem. The results of this experiment suggest that amrinone can be used effectively as a topical vasodilating agent in free or pedicled flap transfer. Further clinical studies are indicated to evaluate the significance of this drug.
hibitor, improves microcirculation and flap survival: A comparative study with prostaglandin E1. J. Surg. Res. 75: 42, 1998. 2.
Puckett, C. L., Winters, R. R., Geter, R. K., and Goebel, D. Studies of pathologic vasoconstriction (vasospasm) in microvascular surgery. J. Hand Surg. [Am.] 10: 343, 1985.
3.
Ohta, I., Kawai, H., Kawabata, H., Masada, K., and Ono, K. Topical use of xylocaine for relieving vasospasm: Effect of concentration. J. Reconstr. Microsurg. 7: 205, 1991.
4.
Chafin, J. B., Wax, M. K., Johnstone, R., and Bishop, D. The use of lidocaine in microvascular reconstruction. Otolaryngol. Head Neck Surg. 117: 93, 1997.
5.
Beekman, W. H., Sluimers, J. E., Kort, W. J., and van der Meulen, J. C. Resolution of experimental microvascular vasoconstriction in rats by topical application of lidocaine hydrochloride in varying concentrations. Ann. Plast. Surg. 21: 570, 1988.
6.
Beekman, W. H., Dammeijer, P. F., Sluimers, J. E., Kort, W. J., and van der Meulen, J. H. Improvement in blood flow and diameter of the postanastomotic rat tail artery by topical application of lidocaine in varying concentrations. Ann. Plast. Surg. 24: 248, 1990.
7.
Hou, S. M., Liu, T. K., and Yu, H. Y. Absorption of lidocaine following topical application in microvascular procedures on rabbits. J. Orthop. Res. 9: 545, 1991.
8.
Geter, R. K., Winters, R. R., and Puckett, C. L. Resolution of experimental microvascular spasm and improvement in anastomotic patency by direct topical agent application. Plast. Reconstr. Surg. 77: 105, 1986.
9.
Wadstrom, J. Studies on traumatic vasospasm in the central ear artery of the rabbit. Scand. J. Plast. Reconstr. Surg. Hand Surg. Suppl. 21: 1, 1990.
10.
Richards, R. R., Seaber, A. V., and Urbaniak, J. R. Chemically induced vasospasm: The effect of ischemia, vessel occlusion, and adrenergic blockade. Plast. Reconstr. Surg. 75: 238, 1985.
11.
Pederson, W. C., Singer, D. I., Angel, M. F., Mahoney, J. E., Mitchell, G. M., Morrison, W. A., and O’Brien, B. M. Hydrostatic distention and pharmacological treatment of epinephrine-induced microarterial spasm. Ann. Plast. Surg. 23: 17, 1989.
12.
Kerschner, J. E., and Futran, N. D. The effect of topical vasodilating agents on microvascular vessel diameter in the rat model. Laryngoscope 106: 1429, 1996.
13.
Seaber, A. V. Experimental vasospasm. Microsurgery 8: 234, 1987.
14.
Welch, W. J., Deng, X., Snellen, H., and Wilcox, C. S. Validation of miniature ultrasonic transit-time flow probes for measurement of renal blood flow in rats. Am. J. Physiol. 268: F175, 1995.
15.
Mori, K., Takeuchi, S., Moritoki, H., Tsuchiya, K., Nakaya, Y., Matsuoka, S., and Kuroda, Y. Endothelium-dependent relaxation of rat thoracic aorta by amrinone-induced nitric oxide release. Eur. Heart J. 17: 308, 1996.
16.
Vroom, M. B., Pfaffendorf, M., van Wezel, H. B., and van Zwieten, P. A. Effect of phosphodiesterase inhibitors on human arteries in vitro. Br. J. Anaesth. 76: 122, 1996.
ACKNOWLEDGMENT We thank Naomi Sekiya for her careful assistance with the experimental studies.
REFERENCES 1.
Ichioka, S., Nakatsuka, T., Sato, Y., Shibata, M., Kamiya, A., and Harii, K. Amrinone, a selective phosphodiesterase III in-
155