Anesth Analg 2004;98:1618-1622
© 2004 International Anesthesia Research Society
doi: 10.1213/01.ANE.0000115781.69209.59
CARDIOVASCULAR ANESTHESIA
Enhanced Vasodilatory Responses to Milrinone in Catecholamine-Precontracted Small Pulmonary Arteries
Rajiv Jhaveri, MD,
Soonyul Kim,
A. Ron White, PhD,
Sean Burke, BS,
Dan E. Berkowitz, MD, and
Daniel Nyhan, MD
Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland
Address correspondence and reprint requests to Dan Berkowitz, MD, Anesthesia, Tower 711, The Johns Hopkins Hospital, 600 N. Wolfe St., Baltimore, MD 21287-8711. Address e-mail to dberkowi @bme.jhu.edu.
 |
Abstract
|
|---|
ß-Adrenergic agonists (e.g., epinephrine [E] and norepinephrine [NE]) and phosphodiesterase-III inhibitors (e.g., milrinone) are often used in combination to augment ventricular function in the perioperative period. In the myocardium, milrinone acts synergistically with ß-adrenergic agonists to increase contractility. However, the potential interaction between catecholamines with combined 
- and ß-adrenergic activity and milrinone in the pulmonary circulation has not been determined. We evaluated the vasodilatory effects of milrinone and nitroglycerine on large elastic and small muscular porcine pulmonary vascular rings precontracted with catecholamines with ß-adrenergic agonist activity (E and NE), the -adrenergic agonist phenylephrine, and a nonadrenergic agonist, the thromboxane analog U46619. In small pulmonary arteries, the vasorelaxation with milrinone was significantly enhanced in rings precontracted with E or NE compared with those precontracted with phenylephrine or U46619. However, in large pulmonary arteries, the vasorelaxation with milrinone was similar in all vessel rings and was not influenced by the agonist used to induce precontraction. In marked contrast, the vasorelaxant responses to nitroglycerine were not altered by the specific agonist used for precontraction in either small or large pulmonary vascular rings. Thus, the pulmonary vascular effects of milrinone are enhanced when combined with drugs with ß-adrenoreceptor agonist activity. The vasodilatory interactions exhibited by phosphodiesterase-III inhibitors and the catecholamines NE and E suggest that their combined use might be beneficial in circumstances in which ventricular dysfunction and increased pulmonary vascular resistance occur.
IMPLICATIONS: This study demonstrated that milrinone had enhanced vasodilator effects when combined with drugs with ß-adrenoreceptor agonist activity in small pulmonary artery segments removed from pigs.
|
Introduction
|
|---|
Inotropes with ß-agonist activity are the first line of treatment for perioperative myocardial dysfunction. Phosphodiesterase-III (PDE3) inhibitors such as milrinone are also frequently used because of their combined vasodilator and inotropic effects. In the myocardium, clear synergistic effects have been demonstrated between PDE3 inhibitors and ß-adrenoreceptor agonists (1–3). In clinical practice, myocardial dysfunction requiring inotropic support is often associated with increased pulmonary vascular resistance (PVR) due to passive increases in pulmonary venous pressures, reactive increases in PVR, and structural remodeling of the pulmonary arterial wall (4). The inotropes used in this setting can exacerbate disturbances in PVR because of their 1-adrenoreceptor activity.
The drugs often used to treat increased PVR include nitric oxide donors (e.g., nitroglycerine [NTG] and inhaled nitric oxide) and PDE3 inhibitors (e.g., milrinone). NTG and milrinone have different mechanisms of action, at least proximally, on intracellular signaling pathways. NTG activates guanylate cyclase, leading to increased production of cyclic guanylate monophosphate (cGMP) (5–7). The cGMP induces relaxation in vascular smooth muscle, including pulmonary vascular smooth muscle, by activation of cGMP-dependent protein kinases and modulation of calcium homeostasis. In the heart, milrinone inhibits the enzyme PDE3 and, thus, cyclic adenosine monophosphate (cAMP) breakdown, with modulation of downstream pathways that results in positive inotropic and lusitropic effects (8–10). In vascular smooth muscle, cAMP activation of protein kinase A has multiple effects, including 1) uncoupling myosin light chain phosphorylation from increases in calcium; 2) modulating sarcolemmal channels, membrane potential, and voltage-dependent calcium channels; 3) inhibiting phospholipase C; and 4) facilitating calcium reuptake by the sarcoplasmic reticulum (11).
Although NTG and milrinone have different proximal mechanisms of action, the activated pathways converge distally to cause similar physiologic responses. Although NTG is a potent pulmonary vasodilator, it has little direct effect on myocardial contractility. In contrast, milrinone is both an inotrope and a pulmonary vasodilator and thus can be a suitable drug for the treatment of patients with both ventricular dysfunction and disturbances in PVR. However, when used independently, milrinone can cause excessive decreases in systemic vascular resistance and diastolic blood pressure, and compromised coronary perfusion.
Epinephrine (E), norepinephrine (NE) (combined - and ß-adrenoreceptor agonists), and phenylephrine (PE) are often used to treat milrinone-induced hypotension. However, there is no information detailing the effects of this combination therapy on the pulmonary vasculature. Moreover, it is entirely possible that the pulmonary vascular effects of combination therapy may differ in small and large pulmonary vessels. Similarly, the potential differential modulating effects of catecholamines on the pulmonary vascular response to NTG in both large and small pulmonary vessels have not been characterized. Under physiologic conditions, pulmonary vascular pressures, PVR, and intrinsic tone are low. Thus, demonstration of the potential pulmonary vasorelaxant effects of any intervention requires precontraction and increase of pulmonary vascular tone.
We investigated the pulmonary vascular effects of milrinone and NTG in both large and small pulmonary vessels ex vivo after precontraction with mechanistically different agonists. We hypothesized that the pulmonary vasodilator response to milrinone would be enhanced after precontraction with adrenergic agonists that cause ß-adrenoreceptor activation.
|
Methods
|
|---|
Porcine lungs were obtained fresh from the slaughterhouse. Intralobar segments of pulmonary arteries were dissected for use as bioassay tissues. Arteries were cleaned of loose connective tissue and cut into rings 1–2 mm long (small) and 4–5 mm long (large) and were then immersed in cold modified Krebs-Ringer solution. To avoid the confounding variable of endothelial modulation of underlying smooth muscle on the interpretation of the results, the rings were deendothelialized by gentle rubbing of the luminal surface. The rings were then suspended horizontally between 2 stainless-steel stirrups in organ chambers filled with 24 mL of Krebs-Ringer solution (95% oxygen, 5% CO2, balance N2; 37°C; pH 7.4). Rings were stretched to 900 mg (small) and 2500 mg (large) of tension in 200- and 500-mg increments, respectively, at 10-min intervals. Tension in the pulmonary rings was continuously recorded by using the MacLab® system (AD Instruments, Grand Junction, CO).
After passive stretching of the rings, stretch-induced myogenic tone was determined by adding sodium nitroprusside (SNP) (10–6 M). The SNP was then washed out, and the viability of the vascular rings was tested after depolarization with 60 mM KCl solution. After washout of the KCl, the vessels were precontracted with PE (10–6 M), and responses to acetylcholine (10–6 M) were tested to confirm that the rings had been deendothelialized. Separate groups of rings were then precontracted with the following agonists: E 10–7 M; NE 10–7 M; PE 10–6 M; or a thromboxane analog, U46619 (10–8 M). The vessels were precontracted to approximately 50%–60% of the KCl maximal response.
After precontraction, vasodilator dose-response relationships were generated to NTG (10–9 to 10–5 M). The NTG was then washed out. Dose-response relationships were then generated to the PDE3 inhibitor milrinone (10–8 to 10–5 M). Because of the pharmacokinetics of milrinone (avid tissue binding, enzyme inhibition, and slow washout), NTG dose-response relationships were always generated first. In a subset of experiments, dose responses to milrinone were performed without prior administration of NTG.
Data were collected online and analyzed with the software Dose Response® (AD Instruments, Grand Rapids, CO). The 50% effective concentration (EC50) and maximum effective concentration (EMAX) were derived by using logistic regression with the software Prizm (GraphPad, San Diego, CA). All statistical comparisons were made with two-way analysis of variance with post hoc Student’s t-tests. Differences were considered significant when P was <0.05.
|
Results
|
|---|
The diameter of the large vessels isolated was 3.7 ± 0.11 mm (n = 24 rings). The diameter of the small pulmonary artery rings was 1.1 ± 0.007 mm (n = 10 rings).
Vessels were preconstricted with the specific agonists before vasodilatory responses to NTG and milrinone were examined. We attempted to obtain equal preconstriction with all agonists. In large vessels, there was no significant difference in the preconstricted tension induced by any of the pressors (Fig. 1a). In the small vessels, tension induced by U46619 was significantly more than that induced by the other agonists. However, there was no significant difference in the preconstricted tension among the other contractile agonists (Fig. 1b).
View larger version (28K):
[in this window]
[in a new window]
|
Figure 1. Levels of preconstriction tension (mg) induced by phenylephrine (P4), epinephrine (E), norepinephrine (NE), and U46619 (U4) in large-diameter (A) and small-diameter (B) vessels (* P < 0.05).
|
|
Stretch-induced tone was assessed by measuring the extent of relaxation after the addition of SNP to the stretch rings. Relaxation (and thus stretch-induced tone) is expressed as the percentage inhibition (reversal) of tone (milligrams) present at the end of stretching. SNP caused 9.0% ± 1.0% and 9.83% ± 6.17% inhibition of tension in small (n = 3) and large (n = 3) rings, respectively.
In small pulmonary arteries, the pulmonary vasodilator responses to NTG were similar in all protocols and were independent of the specific agonist used to induce precontraction (Fig. 2a). In marked contrast, the pulmonary vasorelaxant responses to milrinone were significantly more in small pulmonary artery rings precontracted with either NE or E compared with those precontracted with either PE or U46619 (Fig. 2b). This was reflected in a significant leftward shift in the dose responses in both NE- and E-preconstricted as compared with U46619- and PE-preconstricted rings and in a significant reduction in the EC50 values (Fig. 2c).
View larger version (21K):
[in this window]
[in a new window]
|
Figure 2. Dose response to nitroglycerine (NTG) and milrinone in small vessels after precontraction with U46619 (U4), phenylephrine (PE), norepinephrine (NE), and epinephrine (E). A, Nitroglycerine induced dose-dependent relaxation in pulmonary vessels that was independent of the agonist used for precontraction. B, Milrinone also induced dose-dependent relaxation. However, in marked contrast to nitroglycerine, the specific agonist used for precontraction significantly altered the vasodilator response to milrinone. The responses to milrinone at each dose were markedly enhanced in vessels precontracted with E and NE (*P < 0.05, E and NE versus U4 and PE by two-way analysis of variance). C, Milrinone EC50 values (concentration of milrinone required to induce 50% relaxation) in the presence of agonist-induced preconstriction. The EC50 values were significantly decreased in the presence of preconstriction induced by either E or NE compared with P or U (*P < 0.05; **P < 0.01).
|
|
In the large pulmonary vessels, NTG again produced similar vasorelaxant responses in all protocols, indicating that the response to NTG was independent of the agonist used for precontraction (Fig. 3a). In contrast to small pulmonary vessel rings, milrinone also produced similar vasorelaxant responses when the rings were precontracted with NE, E, or PE. Interestingly, although this was not statistically significant, there was a trend toward milrinone manifesting enhanced vasodilatory responses in rings precontracted with the thromboxane analog U46619 (Fig. 3b).
View larger version (21K):
[in this window]
[in a new window]
|
Figure 3. Dose response to nitroglycerine (NTG) and milrinone in large vessels after precontraction with U46619 (U4), phenylephrine (PE), norepinephrine (NE), and epinephrine (E). NTG induced dose-dependent relaxation in pulmonary vessels. There were no significant differences in the vasorelaxant responses in vessels precontracted with U46619, PE, NE, or E. The responses were independent of the agonist used for precontraction. Milrinone also induced dose-dependent relaxation in the large vessels.
|
|
|
Discussion
|
|---|
The seminal and unique finding of this study is that the pulmonary vasodilator response to milrinone is significantly enhanced in small porcine pulmonary vessels that are precontracted by using agonists that have ß-adrenergic activity. In marked contrast, the responses to NTG were not influenced by the specific agonist used for precontraction. The differential response to milrinone was not observed in large conduit pulmonary arteries. This supports the concept that there is an interaction between vasoactive drugs with ß-adrenoreceptor activity and milrinone in small-diameter pulmonary vessels.
In the myocardium there is a synergistic interaction between ß-adrenoreceptor agonists and PDE3 inhibitors (3,12,13). This synergy arises as a result of these drugs’ effects on cAMP production and degradation, respectively. Wagner et al. (14) demonstrated similar interactions between ß-adrenergic agonists and PDE inhibitors in pulmonary vessels from normal rat lungs and in chronically hypoxic rat lungs. In their study, PDE inhibition with milrinone or rolipram significantly enhanced the magnitude of vasorelaxation produced by either isoproterenol or forskolin. Our results are consistent with those of Wagner et al. and suggest that concomitant pulmonary hypertension and myocardial pump failure may be preferentially treated by combining milrinone and an inotrope with inherent ß-adrenoreceptor agonist activity.
Can these in vitro data obtained in isolated pulmonary vascular rings be extrapolated to the in vivo clinical setting? Specifically, are the concentrations of drugs that produce pulmonary vasorelaxant responses comparable to those likely to be seen in cardiac surgical patients? Bailey et al. (15) investigated the pharmacokinetics and pharmacodynamics of milrinone in patients undergoing cardiac surgery. They found that the relationship between the plasma concentration of milrinone and the percentage increase in cardiac index could be described by a sigmoidal dose-response curve whereby a plasma concentration of 167 ng/mL was associated with a 50% increase in cardiac index (effective EC50). Given a molecular weight of 211 D for milrinone, this translates to an EC50 of 
10–6 M. We derived the EC50 for milrinone from the data in this pulmonary vascular study by using logistic regression and found that the EC50 in small pulmonary vessels preconstricted with NE and E was between 10–6 and 10–7 M. Thus, the in vitro concentrations are very similar to those observed for the cardiac effects in vivo. In contrast, a significantly larger concentration of milrinone (10–6 M) is required to cause pulmonary vasorelaxation in vessels precontracted with PE.
Our study also demonstrated a differential response to pulmonary vasodilators depending on the vessel size studied. This is consistent with previous studies which demonstrated heterogeneity across the pulmonary vasculature with respect to the vascular response to vasoactive activation. This heterogeneity in the pulmonary vasculature extends beyond the modulating influence of vessel size to encompass also the specific agonist studied (16). Specifically, Boels et al. (16) demonstrated a significant difference in endothelial-dependent responses to acetylcholine depending on vessel size. In contrast, the endothelial-dependent vasodilator bradykinin was not dependent on vessel size. Thus, in our study, it would actually be surprising if vasodilator responses to NTG and milrinone had been identical in different vessel sizes.
The pulmonary vascular rings used in this study were obtained from the lungs of normal pigs. The pulmonary vasculature undergoes major structural changes in chronic pulmonary hypertension and may not be responsive to pulmonary vasodilators, including milrinone and NTG. Thus, the results from this study may not be directly applicable to patients who have long-standing pulmonary hypertension. However, vessels that have chronic pulmonary vascular disease are likely to have a fixed PVR that may not be further exacerbated by E or NE. Vessels that exhibit reactive pulmonary vasoconstriction after catecholamine administration are likely to respond to milrinone, leading to amelioration of the pulmonary hypertension. Therefore, although catecholamines are used to treat right ventricular dysfunction in the presence of pulmonary hypertension, they should be used judiciously and, on the basis of our results, might be preferentially combined with milrinone.
ß-Adrenergic agonists continue to be widely used as the first-line treatment for augmenting myocardial contractility. This study confirms that both E and NE increase pulmonary vascular tone. Milrinone effectively reverses the pulmonary vasoconstriction induced by E or NE. -Adrenoreceptor agonists, for example, PE, are frequently used to support systemic blood pressure. However, these agonists may also cause a significant increase in PVR. Our results are consistent with the premise that milrinone is less effective in attenuating the increases in PVR induced by pure -adrenoreceptor (or nonadrenoreceptor) agonists compared with increases in PVR induced by agonists with combined - and ß-adrenoreceptor activity.
|
Acknowledgments
|
|---|
Supported, in part, by the National Space Biomedical Research Institute and the National Institute of Health (NIH-R01 AG021523-01A1).
|
References
|
|---|
- Siostrzonek P, Koreny M, Delle-Karth G, et al. Milrinone therapy in catecholamine-dependent critically ill patients with heart failure. Acta Anaesthesiol Scand 2000; 44: 403–9.[Web of Science][Medline]
- Ono K, Tareen FM, Yoshida A, Noma A. Synergistic action of cyclic GMP on catecholamine-induced chloride current in guinea-pig ventricular cells. J Physiol 1992; 453: 647–61.[Abstract/Free Full Text]
- Bohm M, Diet F, Kemkes B, Erdmann E. Enhancement of the effectiveness of milrinone to increase force of contraction by stimulation of cardiac beta-adrenoceptors in the failing human heart. Klin Wochenschr 1988; 66: 957–62.[Medline]
- Moraes DL, Colucci WS, Givertz MD. Secondary pulmonary hypertension in chronic heart failure: the role of the endothelium in pathophysiology and management. Circulation 2000; 102: 1718–23.[Abstract/Free Full Text]
- Feelisch M. Biotransformation to nitric oxide of organic nitrates in comparison to other nitrovasodilators. Eur Heart J 1993; 14: 123–32.[Web of Science][Medline]
- Ignarro LJ, Ross G, Tillisch J. Pharmacology of endothelium-derived nitric oxide and nitrovasodilators. West J Med 1991; 154: 51–62.[Web of Science][Medline]
- Fung HL. Clinical pharmacology of organic nitrates. Am J Cardiol 1993; 72: 9C–13;discussion 14C–5.
- Schmitz W, von der Leyen H, Meyer W, et al. Phosphodiesterase inhibition and positive inotropic effects. J Cardiovasc Pharmacol 1989; 14: S11–4.[Medline]
- Honerjager P, Nawrath H. Pharmacology of bipyridine phosphodiesterase III inhibitors. Eur J Anaesthesiol 1992; (Suppl 5): 7–14.
- Honerjager P. Pharmacology of bipyridine phosphodiesterase III inhibitors. Am Heart J 1991; 121: 1939–44.[Web of Science][Medline]
- Hopkins PM, O’Beirne HA. Smooth muscle. In: Hemmings HC, Hopkins PM, eds. Foundations of anesthesia: basic and clinical sciences. London: Mosby, 2000: 353–60.
- Erdmann E. The effect of positive inotropes on the failing human myocardium. Cardiology 1997; 88: 7–11.[Medline]
- Orime Y, Shiono M, Hata H, et al. Effects of concomitant usage of milrinone and catecholamine for weaning from cardiopulmonary bypass. Jpn J Thorac Cardiovasc Surg 1998; 46: 803–9.[Medline]
- Wagner RS, Smith CJ, Taylor AM, Rhoades RA. Phosphodiesterase inhibition improves agonist-induced relaxation of hypertensive pulmonary arteries. J Pharmacol Exp Ther 1997; 282: 1650–7.[Abstract/Free Full Text]
- Bailey JM, Levy JH, Kikura M, et al. Pharmacokinetics of intravenous milrinone in patients undergoing cardiac surgery. Anesthesiology 1994; 81: 616–22.[Web of Science][Medline]
- Boels P, Deutsch J, Gao B, Haworth SG. Maturation of the response to bradykinin in resistance and conduit pulmonary arteries. Cardiovasc Res 1999; 44: 416–28.[Abstract/Free Full Text]
Accepted for publication December 10, 2003.
This article has been cited by other articles:
|
|
|
|
 
M. Onomoto, I. Tsuneyoshi, A. Yonetani, S. Suehiro, K. Matsumoto, R. Sakata, and Y. Kanmura
Differential Pharmacologic Sensitivities of Phosphodiesterase-3 Inhibitors Among Human Isolated Gastroepiploic, Internal Mammary, and Radial Arteries
Anesth. Analg.,
October 1, 2005;
101(4):
950 - 956.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
