CLINICAL AND HAEMODYNAMIC EFFECTS OF MILRINONE IN THE TREATMENT OF LOW CARDIAC OUTPUT AFTER CARDIAC SURGERY

Transcription

1 British Journal of Anaesthesia 1991; 67: CLINICAL AND HAEMODYNAMIC EFFECTS OF MILRINONE IN THE TREATMENT OF LOW CARDIAC OUTPUT AFTER CARDIAC SURGERY E. M. WRIGHT AND K. M. SHERRY SUMMARY We have studied the haemodynamic effects of i.v. milrinone. a newphosphodiesterase inhibitor, in patients with low cardiac output after cardiac surgery. Thirty-five patients with a cardiac index (Cl) < 2.5 litre min~' m~ 2 and a pulmonary capillary wedge pressure (PCWP) > 8 mm Hg were given a loading dose of milrinone 50 fig kg' 1 followed by an infusion at one of three rates: fig kg' 1 min' 1, 0.5 fig kg~' min~ 1 or 0.75 fig kg' 1 min' 1 for 12 h. After 1 h there were increases in Cl (35%) (P < 0.001), heart rate (13%) (P < 0.01) and stroke volume index (19%) (P < 0.005). There were decreases in mean arterial pressure (12%) (P < 0.01), systemic vascular resistance (35%) (P < 0.001) and PCWP (24%) (P < 0.05). Pulmonary vascular resistance was unchanged or reduced and left ventricular stroke work index was unchanged or increased. The haemodynamic improvements were sustained throughout the infusion period. Milrinone was tolerated well: three patients developed tachycardia > 125 beat min' 1, one patient developed atrial fibrillation and one patient had a short run of atrial bigemini. We conclude that milrinone is a useful agent in the treatment of patients with a reduced cardiac output after cardiac surgery. KEY WORDS Heart: low cardiac output. Pharmacology:phosphodiesterase inhibitor, milrinone. Surgery: cardiac. A reduced cardiac output is a well recognized postoperative complication of cardiac surgery. It may be caused by either preoperative myocardial damage or perioperative myocardial ischaemia or infarction. Management may involve use of positive inotropic agents, traditionally betaadrenergic agonists often in combination with direct acting peripheral vasodilators. The bipyridine derivative, milrinone, is a new agent combining both inotropic and vasodilator effects. It is unrelated structurally to the catecholamines, the digitalis glycosides or the imidazolone derivative enoximone. It is a derivative of amrinone and has similar pharmacological and haemodynamic effects. It acts mainly via phosphodiesterase 111 inhibition, resulting in an increase in intracellular cyclic adenosine monophosphate (camp) [1]. Milrinone is 10 to 20 times more potent than amrinone and appears to be devoid of the important adverse effects of thrombocytopenia and abnormalities of liver function observed with the older drug [2, 3]. Previous studies have shown milrinone to have beneficial haemodynamic and clinical effects in patients with congestive cardiac failure [4 6] without increasing myocardial oxygen consumption [7]. We therefore performed a dose finding study to investigate the clinical and haemodynamic effects of i.v. milrinone at different infusion rates in a different group of patients, notably those with decreased cardiac output after cardiac surgery. PATIENTS AND METHODS Local Ethics Committee approval was obtained, and written informed consent from all patients. We studied adults having either valvular or coronary artery bypass surgery. All female patients were either post-menopausal or infertile. Patients were excluded if they had hepatic impairment, a plasma creatinine concentration > E. M. WRIGHT*, B.M., B.CH., F.C.ANAES.; K. M. SHERRY, M.B., B.S., F.C.ANAES. ; Department of Anaesthesia, Northern General Hospital, Sheffield. Accepted for Publication: April 4, *Prcsent address, for correspondence: Department of Anaesthetics, Ysbyty Maelor, Wrecsam, Clwyd.

2 586 BRITISH JOURNAL OF ANAESTHESIA TABLE I. Patient characteristics and operations (Milrinone (ig kg" 1 min" 1 ) (Milrinone 0.5 ug kg" 1 min- 1 ) (Milrinone 0.75 ug kg" 1 min" 1 ) Sex (M:F) Mean age (yr) Mean weight (kg) Operation (No.) Mitral valve replacement Coronary artery bypass graft Aortic valve replacement Mitral and aortic valve replacement 8: (46-76) 76.3 (56-87) :7 59.7(51-76) 67.8 (50-84) : (43-68) 70.9(51-92) umol litre" 1 or a platelet count < 100 x 10" litre" 1. They were excluded also if they had significant arrhythmias or had received inotropic therapy 1 h before the study. Long acting vasodilators were withheld for 12 h before the infusion of milrinone. A medical history, physical examination, ECG, chest x-ray and haematological analysis were made and serum concentrations of electrolytes measured 24 h before surgery. A suitably calibrated flowdirected, balloon-tipped pulmonary artery catheter was inserted before operation via the internal jugular vein and a radial artery cannulated for direct arterial pressure monitoring and blood-gas measurement. The anaesthetic method was not standardized. Extracorporeal circulation with non-pulsatile flow and moderate hypothermia were used during each operation. Myocardial protection was provided by cold cardioplegia solution and topical hypothermia. After the operation, platelet count, ECG and chest x-ray were performed. The ECG was monitored continuously throughout the study. Haemodynamic assessment of the patients was undertaken 2 h after operation. The haemodynamic measurements performed were: cardiac output (CO) by thermodilution (an average of three values was taken), mean pulmonary artery wedge pressure (PCWP), systolic, diastolic and mean systemic arterial pressure (SAP), mean pulmonary arterial pressure (MPAP), mean right atrial pressure (RAP) and heart rate. Derived haemodynamic values of cardiac index (CI), stroke volume index (SVI), systemic vascular resistance (SVR), pulmonary vascular resistance (PVR), left ventricular stroke work index (LVSWI), and right ventricular stroke work index (RVSWI) were calculated using standard formulae. The plasma potassium concentration was measured at 1-h intervals for 24 h after operation and maintained at mmol litre" 1. Patients were entered into the study if they had a stable CI < 2.5 litre min" 1 m~ 2 in the presence of a PCWP > 8 mm Hg. A minimum of two sets of readings were taken at least 10 min apart. If the lesser of the two readings for either CI or PCWP was not within 15% of the greater reading, an additional reading was taken. If this value was not within 15 % of the second reading, the patient was withdrawn from the study. Accepted patients were given a loading dose of milrinone 50 ug kg" 1 over 10 min, followed by an infusion of ug kg" 1 min" 1 (first 10 patients), 0.5 ig kg" 1 min" 1 (next 15 patients) or 0.75 ug kg" 1 min" 1 (last 10 patients). The dose regimens were non-randomized. The haemodynamic measurements were repeated at 15, 30 and 45 min, and 1, 3, 6 and 12 h after the start of milrinone. After 12 h the infusion was stopped and further haemodynamic measurements were taken after 2 and 4 h. Statistical atialysis Analysis of variance was used to assess withinand between-group changes and Student's paired t test was used to test the significance of observed differences. RESULTS We studied 22 males and 13 females aged yr. A summary of the patient data is given in table I. There was no significance difference

3 MILRINONE AND CARDIAC SURGERY 587 TABLE II. Mean (SD) haemodynamic data in groups 1 (milrinone fig kg 'mm' 1 ), 0.5 fig kg->min->) and 3 (0.75 fig kg-'min-'). *P < 0.05; ***P < (milrinone Time after infusion Baseline 15 min 1 h 2h 4h CI G'tre min" 1 ra" 1 ) SVR (dyn s an"') HR (beat miir 1 ) MAP (mm Hg) PCWP (mm Hg) LVSWI (g mm"') SVI (ml m" 1 ) 1.95 (0.36) 1.93 (0.29) 2.06 (0.24) 1623 (312) 1811 (515) 1537 (359) 79 (15.2) 84.8 (16.1) 83 (16.1) 79.4 (9.5) 80 (12.6) 77.7 (12) 10.7 (2.5) 10.6 (2) 11.2 (4.2) 23.3 (5.1) 21.8 (6) 23.2 (8.9) 24.8 (3.6) 23.1 (4.2) 25.4 (7.3) 2.99*** (0.6) 2.51*** (0.42) 2.52*** (0.47) 1000*** (242) 1281*** (417) 1160*** (286) 91.8*** (19.3) 97*** (20.2) 93.2* (18) 73.6 (6.3) 72.8* (15.5) 68.1* (10.3) 8.8 (2.7) 7.9*** (2.7) 7.1*** (3.8) 28.6* (4.8) 23.5 (7.8) 23.1 (8.8) 32.6* (4.5) 26.4* (5.5) 27.5 (8.4) 2.75*** (0.55) 2.59*** (0.38) 2.68*** (0.39) 1063*** (254) 1179*** (305) 1091*** (281) 87.3* (16.3) 98.7*** (19.3) 93.7* (19.3) 71.4* (7.8) 72*** (10) 67.7*** (11.3) 9.1* (2.7) 8.1* (3.1) 7.4*** (3.2) * 23.9 (7.1) 31.6* (4.6) 26.8*** (5.5) 28.8* (7.3) 2.84*** (0.44) 2.82*** (0.43) 2.87*** (0.58) 1030*** (253) 1113*** (277) 1023*** (285) 85.6* (11.2) 98.5* (20.5) 91 (21) 73.2 (11.8) 72.4* (8.4) 70.1* (7.6) 9.4 (3.5) 9.0* (2.9) * (5.4) 25.8* 26.3 (8.9) 33.1* (3.5) 29.8*** (4.7) 32.1* (9.3) 2.4* (0.4) 2.53* (0.58) 2.63* (0.88) 1393 (242) 1457 (544) 1426 (649) 82 (11.0) 89.7 (15.4) 90.6 (15) 84.8 (11) 83.7 (15.5) 80.3 (9.9) 12.5 (4.1) 12 (4.6) 11.1 (6.5) 28.1* (3.9) 27.3* (7.3) 27.3 (10.9) 28.8* (2.3) 28.2*** (5.1) 29.4 (11.1) between the baseline haemodynamic measurements in the three groups. Changes in haemodynamic variables are summarized in table II. There was a significant (P < 0.001) improvement in CI in all three dose groups, which was sustained throughout the period of infusion. The average increase was 34 % after 15 min and 35 % after 60 min. Between-groups analysis revealed a greater increase in CI in group 1 after 15 min (P <0.05) CI increased by 53%. This was the

4 588 BRITISH JOURNAL OF ANAESTHESIA only statistically significant difference between the groups. SVR decreased significantly (P < 0.001) in the three groups, by an average 32 % after 15 min and 35 % after 1 h. The vasodilating effect of milrinone was sustained throughout the infusion and there was no significant difference between groups. There was an initial increase in heart rate after the loading dose from an average of 82 beat min" 1 to an average of 94 beat min" 1 after 15 min and 93 beat min" 1 after 1 h (P < 0.01). In three patients the increase in heart rate was considered a clinically adverse event. These patients are discussed further below. Mean arterial pressure decreased and was significant (P < 0.01) in the mid and high dose groups after 15 min and in all groups after 1 h, when an average decrease of 9 mm Hg was seen. Although the decrease in arterial pressure was less in the low dose group, the difference between groups was not significant. PCWP decreased by an average 26% after 15 min and was greater in the mid and high dose groups: 25% (P < 0.001) and 37% (P < 0.001), respectively. The differences between the three groups was not significant. SVI increased by an average of 18% after 15 min and 19% after 60 min (P < 0.005). There was a significant decrease in MPAP in the mid and high dose groups after 15 min and a significant decrease in PVR in the low dose group after 15 min and 60 min. RAP was either unchanged or decreased and LVSWI was either unchanged or increased. There was no significant change in RVSWI. Following discontinuation of the infusion, CI decreased by 11% and SVR increased by 35%. Heart rate and mean arterial pressure returned towards baseline values. Haematological and biochemical measurements before and after milrinone revealed no changes in platelet count or liver function. Adverse events Two patients developed sinus tachycardia. The first was a patient in the mid dose group who had an aortic valve replacement for aortic stenosis and whose heart rate increased from 87 to 127 beat min" 1 after 6 h, settling to 86 beat min" 1 4 h after the infusion. The second was a patient in the high dose group who had coronary bypass surgery and showed a similar pattern of change in heart rate. A third patient who had a mitral valve replacement $23 > min 20 J PCWP (mm Hg) FIG. 1. Mean (SEM) LVSWI and PCWP of all patients at baseline (B) and 15 min after the start of the initial dose of milrinone. for mitral regurgitation was in atrial fibrillation before receiving milrinone; heart rate increased from 105 to 135 beat min" 1 within the first 1 h, but decreased after digoxin mg. One patient in the high dose group who had a mitral valve replacement for mitral regurgitation had episodes of atrial fibrillation after commencing milrinone, reverting to sinus rhythm at the end of the infusion. One patient in the high dose group had a 15-s period of atrial bigemini. DISCUSSION The aim of our study was to investigate the effects of milrinone in a group of patients different from those studied previously. Thus the study design was open and non-comparative. Patients were not allocated randomly to the three dose groups; instead the dose of milrinone was increased sequentially because the sensitivity of patients to milrinone after cardiac surgery was not known. The overall haemodynamic changes seen in our study confirmed the efficacy of milrinone as an inotrope and vasodilator. Although heart rate increased, there was also an increase in SVI, indicating that the improvement in CI was not caused solely by increase in heart rate. The improvement in left ventricular function after the initial loading dose of milrinone is shown in figure

5 MILRINONE AND CARDIAC SURGERY 589 1, which demonstrates an upward and leftward shift of left ventricular function, indicating that the changes were probably caused by a combination of reduced SVR and increasing myocardial contractility. The vasodilating properties of milrinone were confirmed by the significant reduction in SVR seen in all three groups. This was accompanied by a decrease in MAP, but in no patients did this cause concern. The least value of MAP recorded was 52 mm Hg at 15 min in one patient in the middle dose group. The only difference between the groups which was statistically significant was a greater increase in CI in group 1 at 15 min. It is interesting to note that PCWP was maintained better in this group at 15 min, although the difference was not statistically significant. As the decrease in SVR was similar in all three groups, it is possible that this refleas the infusion of greater volumes of fluid to the low dose group. Our results otherwise showed no marked difference between the three dose groups. This probably reflects the fact that the initial loading dose of 50 ug kg" 1 was the same for all three groups. Previous pharmacokinetic studies in patients with congestive cardiac failure have shown an elimination half-life of between 1.5 and 1.7 h [8, 9] after an i.v. dose of ug kg- 1. Thus the initial blood concentration of milrinone would have been similar in all three groups. The haemodynamic improvement was sustained throughout the infusion period. Discontinuation of milrinone was followed by a decrease in cardiac index, although not to baseline values, and probably reflects an underlying recovery. However, there was a marked increase in SVR, thus confirming that the vasodilatation was primarily caused by milrinone and was not merely a reflection of patient warming after cardiac bypass. The incidence of adverse effects in this study was acceptable. We noted tachycardia as an important side effect in fewer than 10% of patients; however, this did not lead to clinical deterioration and it resolved on stopping the infusion. Anti-arrhythmic therapy was required for one patient who had chronic atrial fibrillation and suffered an increase in ventricular rate after milrinone was started; however, dose adjustment or drug discontinuation was not necessary. As arrhythmias are common in diese patients, it is difficult to evaluate the arrhythmic effects of milrinone. Electrophysiological studies have shown that i.v. milrinone enhances atrioventricular conduction, which may explain the increased ventricular rate in the patient with atrial fibrillation [10]. Long term trials of oral milrinone in chronic cardiac failure showed an increased mortality in patients assessed as New York Heart Association (NYHA) grade IV who received oral milrinone therapy, resulting in withdrawal of the oral preparation. However, trials of i.v. infusion of milrinone for 48 h to a similar group of patients with chronic cardiac failure (CCF) have shown a low incidence of adverse effects and no increased mortality [11]. An important question concerning the use of milrinone in postoperative cardiac surgical patients is its effect on myocardial oxygen consumption. The observed effects of milrinone produce opposing influences on the determinants of myocardial oxygen consumption: there is modest increase in heart rate and a modest decrease in systolic arterial pressure. In our study, the change in rate-pressure product was +7% after 60 min. There is evidence from previous studies that short-term administration of milrinone does not increase myocardial oxygen consumption [7, 12]. The combination of inotropic and vasodilator treatment in the management of patients with heart failure after cardiac surgery is often more beneficial than each therapy alone. Milrinone has been compared with dobutamine [5] and with sodium nitroprusside (SNP) [6] in studies of patients with CCF. When i.v. administration of milrinone and SNP were compared for a matched reduction in MAP or SVR, milrinone caused a significantly greater increase in LVSWI at the same or lesser left ventricular end-diastolic pressure (LVEDP). When compared with dobutamine, both milrinone and dobutamine improved cardiac index to a similar degree, but milrinone caused a significantly greater reduction in RAP, PCWP and LVEDP. Milrinone has a mechanism of action which differs from that of the catecholamines. The latter increase camp by stimulation of myocardial betaadrenergic receptors which are linked to the camp synthesizing enzyme adenylate cyclase. Prolonged use of catecholamines is known to lead to a loss of responsiveness caused first by down regulation of receptors and second because the prolonged increase in camp concentration leads

6 590 BRITISH JOURNAL OF ANAESTHESIA to increases in phosphodiesterase activity and enhances tolerance to beta-agonists. Large doses of catecholamines may also cause tachycardia and arrhythmias. Patients who require beta-adrenergic inotropic therapy for several days are therefore at risk of tachyphylaxis and the effects of prolonged catecholamine stimulation. Down regulation of beta receptors may occur also in patients with long-standing left ventricular failure. As phosphodiesterase inhibitors have a mechanism of action different from that of catecholamines, their efficacy is not reduced by down regulation of betareceptors and thus they may be of benefit in patients requiring prolonged inotropic therapy and for patients with chronic left ventricular failure [13]. ACKNOWLEDGEMENT The study was supported by Sterling Research Group Europe. REFERENCES 1. Earl CQ, Linden JM, Weglicki WB. Inhibition of phosphodiesterase by amrinone and milrinone in rat heart. Federation Proceedings 1984; 43: Massie B, Bourassa M, DiBianco R. Long term oral administration of amrinone for congestive heart failure: lack of efficacy in a multi-centre controlled trial. Circulation 1985; 71: Fitzpatrick PG, Cinquegrani MP, Vakiener AR, Baggs JG, Biddle TL, Liang CS. Hemodynamic and regional blood flow response to milrinone in patients with severe congestive heart failure: a dose-ranging study. American Heart Journal 1987; 114: Bairn DS, McDowell AV, Cherniles J, Monrad ES, Parker JA, Edelson J, Braunwald E, Grossman W. Evaluation of a new bipyridine inotropic agent milrinone in patients with severe congestive heart failure. New England Journal of Medicine 1983; 309: Colucci WS, Wright RF, Jaski BE, Fifer MA, Braunwald E. Milrinone and dobutamine in severe heart failure: differing hemodynamic effects and individual patient responsiveness. Circulation 1986; 73 (Suppl. 3): Jasli BE, Fifer MA, Wright RF, Braunwald E, Colucci WS. Positive inotropic and vasodilator actions of milrinone in patients with severe congestive heart failure: Doseresponse relationships and comparison to nitroprusside. Journal of Clinical Investigation 1985; 75: Monrad ES, Bairn DS, Smith HS, Lanoue A, Braunwald E, Grossman W. Effects of milrinone on coronary hemodynamics and myocardial energetics in patients with congestive heart failure. Circulation 1985; 71: Wilson H, Larijani GE, Likoff M, Weber K, Rocci ML. The pharmacokinetics of milrinone in patients with chronic cardiac failure. Clinical Pharmacology and Therapeutics 1984; 35: 238 (Abstract No. A27). 9. Lesko LJ, McCue JE, Brady P, Alpert JS. Pharmacokinctics and pharmacodynamics of intravenous milrinone. Circulation 1984; 70 (Suppl. 2): 168 (Abstract No. 670). 10. Goldstein RA, Geraci SA, Gray EL, Rinkenbcrger RL, Hamilton Dougherty A, Naccarcelli GV. Electrophysiological effects of milrinone in patients with congestive heart failure. American Journal of Cardiology 1986; 57: Anderson JL, Bairn DS, Fein SA, Goldstein RA, Lejemtel TH, Likoff MJ. Efficacy and safety of sustained (48 hour) intravenous infusions of milrinone in patients with severe congestive heart failure: a multicenter study. Journal of the American College of Cardiology 1987; 9: Lejemtel TH, Gumbardo D, Chadwick B, Rutman HI, Sonnenblick EH. Milrinone for long term therapy of heart failure: clinical experience with special reference to maximal exercise tolerance. Circulation 1986; 73 (Suppl. 3): Watson DM, Sherry KM, Weston GA. Milrinone a bridge to heart transplant. Anaesthesia 1991; 46: