- Email: [email protected]
Effects of magnesium sulphate on the cardiovascular system, coronary circulation and myocardial metabolism in anaesthetized dogs
British Journal of Anaesthesia 1997; 79: 363–368
Effects of magnesium sulphate on the cardiovascular system, coronary circulation and myocardial metabolism in anaesthetized dogs
Y. NAKAIGAWA, S. AKAZAWA, R. SHIMIZU, R. ISHII, S. IKENO, S. INOUE AND R. YAMATO
Summary We have studied the effects of i.v. bolus doses of magnesium sulphate (MgSO4) 60, 90 and 120 mg kg1 on haemodynamic state, the coronary circulation and myocardial metabolism in nine dogs anaesthetized with pentobarbitone and fentanyl. MgSO4 produced dose-dependent decreases in arterial pressure, heart rate, left ventricular dP/dtmax and left ventricular minute work index (LVMWI) and an increase in the time constant of left ventricular isovolumic relaxation. Stroke volume increased, systemic vascular resistance decreased and cardiac output did not change significantly. MgSO4 produced decreases in coronary perfusion pressure, coronary vascular resist• ance and myocardial oxygen consumption (MVO2) Coronary sinus blood flow, lactate extraction ratio • and the ratio of LVMWI to myocardial MVO2 , that is an index of cardiac efficiency, did not change significantly. This study indicated that the depressant effect of MgSO4 on cardiac function was offset by lowering of peripheral vascular resistance, so that cardiac pump function remained effective, and the almost constant coronary sinus blood flow resulted from the decrease in coronary vascular resistance even at higher doses. (Br. J. Anaesth. 1997; 79: 363–368). Key words Hypnotics barbiturate, pentobarbitone. system, effects. Heart, metabolism. magnesium sulphate. Dog.
Cardiovascular Pharmacology,
Magnesium sulphate has been used for decades in the management of eclampsia1 2 and pregnancyinduced hypertension.3 4 Recent reports on magnesium have emphasized its role in deficiency syndromes,5–7 including cardiac arrhythmias.8 9 10 The therapeutic importance and physiological roles of magnesium have been summarized by Iseri and French,11 and an extensive review on clinical experiences has been published by James.12 The evidence for cardiac depression by clinically useful concentrations of hypermagnesaemia is less clear, despite its wide clinical use. Some authors have demonstrated that there is no evidence of myocardial depression at any concentration of serum
magnesium,13 whereas others have reported that a single injection of magnesium sulphate leads to a decrease in cardiac performance.14 Although the cardiovascular haemodynamic effects of magnesium sulphate were assessed by some authors in baboons,13 sheep15 and humans,16 there have been few reports on magnesium-induced changes in left ventricular diastolic function. The effects of magnesium sulphate on the coronary circulation and myocardial metabolism have not been studied extensively. This study was designed to investigate the effects of magnesium sulphate on haemodynamic state, including left ventricular diastolic function, coronary circulation and myocardial metabolism, in open-chest dogs.
Materials and methods The study was approved by the management committee at Jichi Medical Laboratory of Experimental Medicine, based on the school’s Guide for Laboratory Animals, 1993. Nine mongrel dogs (mean weight 23.5 (range 18–30.5) kg) were anaesthetized with pentobarbitone 30 mg kg1. After tracheal intubation, the lungs were ventilated with a Harvard pump respirator ( FIO2 1.0). Arterial carbon dioxide tension was maintained at 4.8–5.3 kPa, as assessed by continuous measurement of end-tidal carbon dioxide concentration with a calibrated respiratory gas monitor (Model 5250, Ohmeda, Madison, WI.,USA) and repetitive analyses of arterial bloodgas tensions. Temperature was measured with an oesophageal thermistor and maintained at 36–37 C using an external heating pad. Anaesthesia was maintained during surgery with fentanyl 20 g kg1 h1 and 0.5–2.0 % sevoflurane in pure oxygen. All animals were paralysed with bolus doses of pancuronium 0.2 mg kg1 followed by small doses as needed. Heart rate (HR) was monitored by electrocardiography (ECG, lead II), and systolic, diastolic and YASUSHI NAKAIGAWA, MD, SATOSHI AKAZAWA, MD, REIJU SHIMIZU, MD, RYOUSUKE ISHII, MD, SHIGEO IKENO, MD, SOICHIRO INOUE, MD, REIKO YAMATO, MD, Department of Anaesthesiology, Jichi Medical School, Tochigi, 329–04, Japan. Accepted for publication: April 25, 1997. Correspondence to Y. N.
364
British Journal of Anaesthesia
mean arterial pressures (SAP, DAP and MAP, respectively) were measured with a micromanometer-tipped catheter (7F 45326, Toyoda, Ltd, Tokyo, Japan) placed via the left femoral artery into the abdominal aorta. A catheter was placed in the abdominal aorta via the right femoral artery for arterial blood sampling. Lactated Ringer’s solution was infused at 5–10 ml kg1 h1 into the right femoral vein throughout the experiment. After median sternotomy, an appropriately sized electromagnetic flow transducer (FB-140T, Nihon Kohden, Ltd, Tokyo, Japan) was placed around the root of the ascending aorta and connected to a flowmeter (MFW2100, Nihon Kohden, Ltd, Tokyo, Japan) for measurement of ascending aortic flow and cardiac output (CO). A micro-tip catheter (8F PC380, Millar, Inc.) was inserted through a stab wound in the apex of the left ventricle for measurements of left ventricular pressure (LVP) and left ventricular enddiastolic pressure (LVEDP). The rate of increase of left ventricular pressure (LVdP/dt) was obtained with an analogue differentiating circuit incorporated with an analogue-to-digital converter (Contractility Unit 1323, NEC San-ei, Ltd, Tokyo, Japan) and used as an index of left ventricular contractile function. Coronary sinus blood flow (CSBF) was measured by the retrograde thermodilution technique with a catheter (Thermoflow CCS-7U-90A, Webster Labs., Inc.) inserted into the coronary sinus via the right auricle.17 Cold normal saline as the indicator was infused at 20 ml min1 for 20 s with an infusion pump (SW-367, Sage Inst., Inc.) through the catheter for each measurement of CSBF. A pair of piezoelectric crystals (5 MHz, Sonotek Co., San Diego, CA, USA) were inserted into the subendocardium in the anterior and posterior walls of the left ventricle and connected to an amplifier (amplifier Unit 4105, NEC San-ei., Ltd, Tokyo, Japan) for measurement of left ventricular enddiastolic and end-systolic diameters (LVEDD and LVESD). Fractional shortening (FS) was obtained using the following formula and used as an index of left ventricular contractile function: FS (LVEDD– LVESD)/LVEDD100. All variables were recorded continuously on a polygraph (RECTI-HORIZ-8K23, NEC San-ei, Ltd, Tokyo, Japan).
Arterial and coronary sinus blood was sampled simultaneously to measure pH, PO2, PCO2, base excess, standard bicarbonate and haemoglobin concentrations (288 Blood Gas System, Ciba Corning), lactate, non-esterified fatty acid (NEFA) and magnesium concentrations. Lactate concentrations were measured using an enzymatic method. • Myocardial oxygen consumption (MV O2) , the ratio • of left ventricular minute work index to MV O2 • (LVMWI/MV O2) myocardial oxygen extraction ratio (MO2ext.), myocardial lactate extraction ratio (MLext.) and myocardial non-esterified fatty acid extraction ratio (MNEFAext.) were calculated. All derived variables and their abbreviations are shown in table 1. Left ventricular pressure (LVP), LVdP/dt, LVEDP and ECG signals were fed to a multichannel photographic oscillograph (Visigraph 5137, NEC San-ei Inst., Ltd, Tokyo, Japan) and recorded at a paper speed of 500 mm s1 to obtain the time constant of left ventricular isovolumic relaxation (T). T was calculated according to the method of Weiss, Frederiksen and Weisfeldt18 and used as an index of left ventricular relaxation. At the end of the surgical procedure, sevoflurane was discontinued and fentanyl was infused at a rate of 20 g kg1 h1. Approximately 30 min was allowed after withdrawal of sevoflurane for stabilization of haemodynamic variables and baseline values were obtained. All dogs were given a 12.3% solution of magnesium sulphate in a bolus dose of 60, 90 and 120 mg kg1 (0.49, 0.73 and 0.98 ml kg1, respectively) over 10 s at 20-min intervals consecutively. The ratio of osmolarity of magnesium sulphate to that of saline is 3.0 : 1.0. Haemodynamic measurements and blood sampling were performed 1 min before and after each administration of magnesium sulphate. All values are presented as mean (SEM). Data were analysed statistically using analysis of variance for repeated measurements (ANOVA). Where appropriate, this was followed by the Scheffe F test. A probability of chance occurrence less than 5% (P0.05) was considered statistically significant.
Table 1 Derived variables. BWBody weight (kg); COcardiac output (litre min1); HRheart rate (beat min1); MAPmean arterial pressure (mm Hg); LVEDPleft ventricular end-diastolic pressure (mm Hg); CVPcentral venous pressure (mm Hg); LVEDDleft ventricular end-diastolic diameter (mm); LVESDleft ventricular endsystolic diameter (mm); DAPdiastolic arterial pressure (mm Hg); CSBFcoronary sinus blood flow (ml min1); (a–cs) DO2arterio-coronary sinus oxygen content difference (ml dl1); CaO2 arterial oxygen content (ml dl1); (a–cs) DLarterio–coronary sinus lactate concentration difference (mg dl1); Laarterial lactate concentration (mg dl1); (a–cs) DNEFAarterio- coronary sinus non-esterified fatty acid concentration difference (mEq litre1); NEFAaarterial non- esterified fatty acid concentration (mEq litre1) CI SV SVI LVMWI SVR FS CPP CVR MV!O2 MO2 ext. MLext. MNEFAext.
Cardiac indexCO/BW Stroke volume(CO/HR) 103 Stroke volume indexSV/BW Left ventricular minute work index(MAP–LVEDP) SVIHR0.0136 Systemic vascular resistance(MAP–CVP)/CO80 Fractional shortening(LVEDD–LVESD)/LVEDD100 Coronary perfusion pressureDAP–LVEDP Coronary vascular resistanceCPP/CSBF80 Myocardial oxygen consumption(a–cs) DO2CSBF102 Myocardial oxygen extraction ratio((a–cs) DO2/CaO2) 102 Myocardial lactate extraction ratio((a–cs) DL/La102 Myocardial non-esterified fatty acid extraction ratio((a–cs) DNEFA/NEFAa) 102
(litre min1 kg1) (ml) (ml kg1) (g m kg1 min1) (dyn s cm5) (%) (mmHg) (kdyn s cm5) (ml min1) (%) (%) (%)
Magnesium: haemodynamics and myocardial metabolism
365
Table 2 Effects of magnesium sulphate (MgSO4) on cardiovascular state (mean (SEM)). MgSerum concentration of magnesium; HRheart rate; SAPsystolic arterial pressure; DAPdiastolic arterial pressure; MAPmean arterial pressure; LVEDPleft ventricular end-diastolic pressure; COcardiac output; CIcardiac index; SVstroke volume; SVIstroke volume index; LVMWIleft ventricular minute work index; SVRsystemic vascular resistance; LVdP/dtmaxmaximum rate of increase of left ventricular pressure; LVEDDleft ventricular end-diastolic diameter; LVESDleft ventricular end-systolic diameter; FSfractional shortening; Ttime constant of fall in isovolumic left ventricular pressure. *P0.05, **P0.01 compared with baseline. †P0.05, ††P0.01 compared with MgSO4 60 mg kg1 MgSO4 (mg kg-1)
Mg (mg dl1) HR (beat min1) SAP (mm Hg) DAP (mm Hg) MAP (mm Hg) LVEDP (mm Hg) CO (litre min1) CI (litre min1 kg1) SV (ml) SVI (ml kg1) LVMWI (g m kg1 min1) SVR (dyn s cm-5) LVdP/dtmax (mm Hg s1) LVEDD (mm) LVESD (mm) FS (%) T (ms)
Baseline
60
90
120
1.3 (0.1) 109 (8) 145 (6) 99 (7) 115 (6) 3.9 (0.5) 2.2 (0.1) 0.10 (0.01) 21.3 (2.3) 0.98 (0.16) 146 (14) 4376 (506) 2468 (210) 36.0 (6.8) 42.8 (6.2) 19.1 (4.2) 48.0 (4.4)
6.3 (0.3)** 93 (6) 122 (6)** 72 (6)** 89 (6)** 4.1 (0.5) 2.4 (0.1) 0.11 (0.01) 26.6 (2.4) 1.21 (0.17)* 121 (12) 3076 (335)* 1974 (194)** 35.7 (6.6) 42.9 (6.2) 20.3 (3.6) 55.6 (5.1)*
9.7 (0.5)**†† 79 (5)**† 114 (6)** 65 (6)** 81 (6)** 4.7 (0.6) 2.3 (0.2) 0.10 (0.01) 29.3 (2.9)* 1.32 (0.19)** 104 (14)* 3058 (372)** 1721 (206)** 36.3 (6.6) 43.8 (6.0) 20.0 (4.0) 62.3 (5.8)**
14.4 (0.7)**†† 69 (4)**†† 97 (7)**†† 53 (6)**†† 69 (6)**†† 4.5 (0.6) 2.0 (0.1) 0.09 (0.01) 29.1 (3.0)* 1.29 (0.18)** 74 (9)**†† 2833 (303)** 1320 (188)**†† 36.6 (6.5) 43.3 (5.5) 20.0 (3.6) 67.5 (5.2)**††
Figure 1 Changes in cardiac output (CO) (!), heart rate (HR) (▲) and stroke volume (SV) (■) (mean, SEM) after administration of magnesium sulphate (MgSO4). *P0.05, **P0.01 compared with baseline (B).
Results Serum concentrations of magnesium 1 min after each bolus administration and the effects on haemodynamic state are shown in table 2. Haemodynamic variables decreased within the first minute and gradually returned to near baseline values after 20 min. Serum concentrations of magnesium before and after administration of magnesium sulphate of 60, 90 and 120 mg kg1 were 1.3 (0.1), 6.3 (0.3), 9.7 (0.5) and 14.4 (0.7) mg dl1, respectively. Magnesium sulphate produced dose-dependent
Figure 2 Changes in cardiac output (CO) (!), mean arterial pressure (MAP) (▲) and systemic vascular resistance (SVR) (■) (mean, SEM) after administration of magnesium sulphate (MgSO4). *P0.05, **P0.01 compared with baseline (B).
Table 3 Effects of magnesium sulphate (MgSO4) on electrocardiogram (mean (SEM)). SCLSinus cycle length; QTcQT interval corrected for heart rate (Bazett’s formula). *P0.05, **P0.01 compared with baseline. †P0.05, ††P0.01 compared with MgSO4 60 mg kg1 MgSO4 (mg kg1)
SCL (ms) PQ (ms) QRS (ms) QT (ms) QTc (ms)
Baseline 60
90
120
571 (40) 124 (6) 79 (3) 429 (25) 570 (26)
783 (56)**†† 168 (8)**† 84 (2)** 513 (27)**†† 584 (29)
895 (53)**†† 184 (8)**†† 86 (2)** 563 (29)**†† 599 (29)*
664 (42)** 148 (6) 80 (2) 481 (26)**†† 592 (27)*
366
British Journal of Anaesthesia Table 4 Effects of magnesium sulphate (MgSO4) on coronary circulation and myocardial metabolism (mean (SEM)). CPPCoronary perfusion pressure; CSBFcoronary sinus blood flow; COcardiac output; CVRcoronary ˝ vascular resistance; MV O2 myocardial oxygen consumption; MO2ext. myocardial oxygen extraction ratio; MLext. myocardial lactate extraction ratio; MNEFAext. myocardial non-esterified fatty acid extraction ratio; LVMWIleft ventricular minute work index. *P0.05, **P0.01 compared with baseline. †P0.05, ††P0.01 compared with MgSO4 60 mg kg1 MgSO4 (mg kg1)
CPP (mm Hg) CSBF (ml min1) CSBF/CO (%) CVR (kdyn s cm5) ˝ MV O2 (ml min1) MO2ext. (%) MLext. (%) MNEFAext. (%) LVMWI/MV!O2 (g m kg1 litre1)
Baseline
60
90
120
90 (8) 91.7 (4.4) 4.4 (0.3) 81 (10) 5.4 (0.3) 44.3 (2.1) 42.1 (4.2) 20.4 (2.3) 24.0 (2.3)
64 (7)** 112.5 (12.1) 5.0 (0.6) 49 (6)** 4.8 (0.6) 36.6 (4.4) 37.1 (4.5) 24.1 (5.2) 24.0 (3.7)
57 (7)** 106.9 (12.9) 5.4 (0.8) 47 (7)** 3.9 (0.6) 32.7 (4.1)* 33.8 (4.6) 27.0 (7.1) 23.1 (2.4)
40 (4)**†† 89.2 (15.7) 4.7 (0.8) 42 (6)** 2.7 (0.3)**†† 31.6 (4.4)* 31.7 (7.9) 34.9 (9.5) 22.3 (3.7)
decreases in HR, SAP, DAP, MAP, LVMWI and LVdP/dt max (P0.01). Stroke volume increased significantly but cardiac output did not change significantly (fig. 1). The near constant CO may have been brought about by the decrease in systemic vascular resistance (SVR) (fig. 2). LVEDD, LVESD and FS did not change significantly. T increased in a dose-dependent manner, indicating a decrease in left ventricular diastolic function. The effects of magnesium sulphate on the electrocardiogram are shown in table 3. Magnesium sulphate increased sinus cycle length, PQ and QT intervals dose-dependently (P0.01). QRS duration and corrected QT interval (QTc) increased significantly, but not dose-dependently. P wave, ST segment and T wave configurations did not change significantly. Magnesium sulphate did not produce arrhythmia. The effects of magnesium sulphate on the coronary circulation and myocardial metabolism are shown in table 4. Magnesium sulphate produced dose-dependent decreases in coronary perfusion • pressure (CPP) and MV O2 (P0.01). Coronary vascular resistance (CVR) and MO2ext. ratio decreased, but not dose-dependently. Any significant changes in CSBF, the ratio of CSBF to CO, MLext. ratio, MNEFAext. ratio and • LVMWI/MV O2 were not observed at each dose of magnesium sulphate.
Discussion The major findings of this study were that magnesium sulphate, in a dose-dependent manner, lowered systemic vascular resistance, arterial pressure and heart rate, while cardiac output was well maintained. Cardiac work, derived from both stroke volume and heart rate, and myocardial oxygen consumption, decreased markedly in parallel with systemic vascular resistance. Magnesium sulphate maintained coronary perfusion and cardiac efficiency even at higher doses. With an unchanged cardiac output, the decrease in arterial pressures must result from a decrease in systemic vascular resistance, which may be caused mainly by the vasodilator effects of magnesium sulphate that have been demonstrated in vitro,19 20 in
intact animals13 21 and in humans.15 The cardiovascular effects of magnesium are as follows: (1) inhibition of the normal rhythm or vasomotion (spontaneous mechanical activity) of vascular smooth muscle, (2) depression of the contractile responses to endogenous neurohumoral substances that maintain vascular tone,22 (3) possibly blockade of sympathetic ganglia23 and (4) negative inotropism. In humans, Mroczek, Lee and Davidov15 reported that infusion of magnesium sulphate resulted in a rapid but transient decrease in arterial pressure in hypertensive patients, whereas normotensive subjects did not demonstrate any appreciable change in arterial pressure. McCubbin and co-workers24 reported cardiopulmonary arrest caused by acute maternal hypermagnesaemia (35.1 mg dl1). In animals, transient16 and marked13 25 hypotension has been reported. This transient hypotension may be explained partially by rapid clearance of magnesium from plasma through the kidney. James, Cork and Dennett13 demonstrated the difficulties of maintaining high serum magnesium concentrations in the presence of normal renal function. They administered in the baboon a loading dose of 60 mg kg1 over 1 min and then infused continuously 12.5 mg min1 for at least 5 min. The infusion rate was then increased to 25 mg min1 and by 25 mg min1 every 5 min up to a maximum of 200 mg min1. The maximum serum magnesium concentration attained was only 9.4 mmol litre1. Dandavino and colleagues16 observed that rapid administration of magnesium sulphate produced hypotension but a more gradual administration did not produce hypotension. In our study, magnesium sulphate at bolus doses of 60, 90 and 120 mg kg1 administered every 20 min produced a dose-dependent decrease in arterial pressure. This decrease in arterial pressure corresponded to increases in serum concentrations of magnesium. Magnesium sulphate is known to have a negative inotropic action and cause myocardial depression and cardiac arrest in massive doses.24 In this study, magnesium sulphate produced a dose-dependent decrease in LVdP/dtmax. Although the decrease in LVdP/dtmax can be explained partially by the decrease in heart rate,26 this indicates that magnesium sulphate exerted a dose-dependent
Magnesium: haemodynamics and myocardial metabolism negative inotropic action in the whole animal. Left ventricular pump function, however, was well compensated by the increase in stroke volume caused by the lowering of peripheral vascular resistance. In fact, magnesium sulphate 60 and 90 mg kg1 produced a significant increase in stroke volume despite a marked decrease in LVdP/dtmax. Cardiac output also was maintained well at a high dose of 120 mg kg1. The profound decrease in heart rate may be the main effect of magnesium sulphate, with the heart compensating by a Starling effect to increase stroke volume appreciably. This finding is consistent with previous reports.3 13 27 James, Cork and Dennett13 demonstrated that cardiac output was maintained at control values, at concentrations of serum magnesium greater than 5 mmol litre1. Cotton, Gonik and Dorman3 and Mroczek, Lee and Davidov15 also demonstrated that magnesium sulphate produced an increase in cardiac output despite significant myocardial depression. These findings and ours suggest that the direct myocardial depressant effect with magnesium may be compensated by decreased afterload, resulting in the maintenance of cardiac function as a pump. No significant change in fractional shortening as an index of left ventricular systolic function during ejection phase could be attributed to the decrease in afterload. Some authors16 25 have reported that tachycardia occurred during administration of magnesium sulphate, whereas others13 have reported that bradycardia was seen. In our study, magnesium sulphate produced a dose-dependent decrease in heart rate and prolongation of PQ and QTc intervals. The degree and rate of increase of blood magnesium concentration may partially explain the different findings observed among these studies. At a low serum concentration of magnesium, hypotension may produce tachycardia through the baroreflex pathway, whereas at a high concentration bradycardia may be brought about by both indirect and direct inhibitory effects of magnesium on the sinoatrial node. As for indirect inhibitory effects, Stanbury23 found that magnesium induces a short-lived blockade of the sympathetic ganglia and prevents the stimulating effects of potassium and acetylcholine on the superior cervical ganglion in anaesthetized cats. As for direct inhibitory effect, Opthof and colleagues28 demonstrated that magnesium itself decreased the rate of diastolic depolarization of the pacemaker cells in the sinus node. In this study, magnesium sulphate did not produce any arrhythmias or conduction disorders at serum concentrations of 6.3 (0.3) to 14.4 (0.7) mg dl1, although it produced a dose-dependent decrease in heart rate and prolongation of PQ, QRS and QTc intervals. This finding is consistent with our earlier study29 except for QTc interval. The difference of basal anaesthesia may be associated with the different results of QTc interval. Although the effects of magnesium sulphate on systemic haemodynamic state and myocardial contractility have been described in other studies,13 15 16 the effect on ventricular diastolic function in vivo or in vitro remains unexplored. In this study, the effect
367 of magnesium sulphate on left ventricular relaxing function was investigated by the time constant of left ventricular isovolumic relaxation (T). The shorter the T value, the better the relaxing function of the left ventricle, and vice versa.18 Magnesium sulphate produced a dose-dependent increase in T in this study. Therefore, magnesium may depress left ventricular relaxing function. Because T tends to increase with a decrease in heart rate and depressed systolic fibre shortening,18 the decrease in T observed here appeared to be related not only to decreased left ventricular relaxing function but also to decreased heart rate and depressed systolic fibre shortening. The effects of magnesium sulphate on the coronary circulation and myocardial metabolism have not been studied extensively. This study demonstrated that magnesium sulphate did not change coronary sinus blood flow remarkably despite a decrease in coronary perfusion pressure. This suggests that magnesium sulphate may have a coronary vasodilating effect. A dose-dependent • decrease in MV O2 appeared to parallel the decreased • external work of the heart, because LVMWI/MV O2 did not change significantly. The decrease in LVMWI is thought to be caused mainly by reduction in arterial pressure. Lactate production was not observed throughout this study. Therefore, magnesium sulphate may maintain an appropriate oxygen supply-and-demand relation in the myocardium and may exert no major adverse effects on cardiac function as a pump. The findings of this study could be related to basal anaesthesia because of an acute open-chest canine preparation under general anaesthesia. Priebe demonstrated an anaesthetic technique comprising pentobarbitone, with fentanyl maintaining a stable, global haemodynamic state over a prolonged period.30 Cox31 demonstrated that haemodynamic variables returned to control levels within 15 min after pentobarbitone 30 mg kg1 i.v. in dogs, although heart rate remained increased from the control value of 85 beat min1 to 132 beat min1 even after 60 min. In this study, however, a heart rate of 125 (5) beat min1 immediately after administration of pentobarbitone decreased to a baseline value of 109 (8) beat min1 immediately before initiation of measurements. The osmolarity of magnesium sulphate solutions used could not produce any transient cardiovascular effects because the volumes were small and osmolarity might be lowered by the circulating blood of the dogs. Magnesium sulphate competes with calcium at the cell membrane and is regarded in many clinical situations as a physiological calcium blocker. This concept may be supported by the facts that magnesium sulphate (1) exerted a mild consistent, dose-dependent hypotensive effect, (2) preserved cardiac pump and coronary sinus blood flow despite a decrease in coronary perfusion pressure, (3) produced negative inotropic effect, (4) decreased myocardial oxygen consumption and (5) did not produce arrhythmia. These effects are thought to be beneficial for patients with ischaemic heart disease.
368 In future, magnesium will be used more widely in many medical situations, including anaesthetic practice as a physiological calcium blocker.
British Journal of Anaesthesia
16.
Acknowledgements This work was supported by a grant-in-aid for scientific research (B) from the Ministry of Education, Science and Culture, Japan. We thank K. Nozawa and N. Yoshida (NEC San-ei., Utsunomiya, Japan) for technical assistance.
References 1. Pritchard JA, Pritchard SA. Standardized treatment of 154 consecutive cases of eclampsia. American Journal of Obstetrics and Gynecology 1975; 123: 543–552. 2. Pritchard JA, Cunningham FG, Pritchard SA. The Parkland Memorial Hospital protocol for treatment of eclampsia: Evaluation of 245 cases. American Journal of Obstetrics and Gynecology 1984; 148: 951–963. 3. Cotton DB, Gonik B, Dorman KF. Cardiovascular alterations in severe pregnancy-induced hypertension: Acute effects of intravenous magnesium sulphate. American Journal of Obstetrics and Gynecology 1984; 148: 162–165. 4. Lee MI, Todd HM, Bowe A. The effects of magnesium sulphate infusion on blood pressure and vascular responsiveness during pregnancy. American Journal of Obstetrics and Gynecology 1984; 149: 705–708. 5. Turlapaty PDMV, Altura BM. Magnesium deficiency produces spasms of coronary arteries: Relationship to etiology of sudden death ischemic heart disease. Science 1980; 208: 198–200. 6. Vincent JL, Buset M, Dufaye P, Barre J, Degaute JP, Kahn RJ. Circulatory shock associated with magnesium depletion. Intensive Care Medicine 1982; 8: 149–152. 7. Iseri LT, Freed J, Bures AR. Magnesium deficiency and cardiac disorders. American Journal of Medicine 1975; 58: 837–846. 8. Cohen L, Kitzes R. Magnesium sulphate in the treatment of variant angina. Magnesium 1984; 3: 46–49. 9. Iseri LT, Fairshter RD, Hardemann JL, Brodsky MA. Magnesium and potassium therapy in multifocal arterial tachycardia. American Heart Journal 1985; 10: 789–794. 10. Solomon D, Bunegin L, Albin M. The effect of magnesium sulphate administration on cerebral and cardiac toxicity of bupivacaine in dogs. Anesthesiology 1990; 72: 341–346. 11. Iseri LT, French JH. Magnesium: Nature’s physiologic calcium blocker. American Heart Journal 1984; 108: 188–193. 12. James MFM. Clinical use of magnesium infusions in anesthesia. Anesthesia and Analgesia 1992; 74: 129–136. 13. James MFM, Cork RC, Dennett JE. Cardiovascular effects of magnesium sulphate in the baboon. Magnesium 1987; 6: 314–324. 14. Critelli G, Ferro G, Peschle C, Perticone FR, Rengo FR, Condorelli M. Myocardial contractility after injection or prolonged infusion of magnesium sulphate. Acta Cardiologica 1977; 32: 65–73. 15. Mroczek WJ, Lee WR, Davidov ME. Effect of magnesium
17.
18.
19. 20. 21. 22.
23. 24. 25. 26.
27. 28.
29.
30. 31.
sulphate on cardiovascular haemodynamics. Angiology 1970; 10: 720–724. Dandavino A, Woods JR, Murayama K, Brinkman CR, Assali NS. Circulatory effects of magnesium sulphate in normotensive and renal hypertensive pregnant sheep. American Journal of Obstetrics and Gynecology 1977; 127: 769–774. Ganz W, Tamura K, Marcus HS, Donoso R, Yoshida S, Swan HJC. Measurement of coronary sinus blood flow by continuous thermodilution in man. Circulation 1971; 44: 181–195. Weiss JL, Frederiksen JW, Weisfeldt ML. Haemodynamic determinant of the time course of fall in canine left ventricular pressure. Journal of Clinical Investigation 1976; 58: 751–760. Turlapaty PDMV, Altura BM. Extracellular magnesium ions control calcium exchange and content of vascular smooth muscle. European Journal of Pharmacology 1978; 52: 421–423. Altura BM, Alutra BT. Magnesium and contraction of arterial smooth muscle. Microvascular Research 1974; 7: 145–155. James MFM, Cork RC, Harlen GM, White JF. Interactions of adrenaline and magnesium on the cardiovascular system of the baboon. Magnesium 1988; 7: 37–43. Altura BM and Altura BT. General anesthetics and magnesium ions as calcium antagonists on vascular smooth muscle. In: Weiss GB, ed. New Perspectives on Calcium Antagonists. Washington: Maryland American Physiological Society, 1981; 131–145. Stanbury JB. The blocking action of magnesium ion on sympathetic ganglia. Journal of Pharmacology and Experimental Therapeutics 1948; 93: 52–62. McCubbin JH, Sibai BM, Abdella TN, Anderson GD. Cardiopulmonary arrest due to acute maternal hypermagnesaemia. Lancet 1981; 1: 1058. Maxwell GM, Elliott RB, Burnell RH. Effects of hypermagnesemia on general and coronary hemodynamics of the dog. American Journal of Physiology 1965; 208: 158–161. Msaon DT. Usefulness and limitations of the rate of rise of intraventricular pressure (dp/dt) in the evaluation of myocardial contractility in man. American Journal of Cardiology 1969; 23: 516–527. James MFM, Manson EDM. The use of magnesium sulphate infusions in the management of very severe tetanus. Intensive Care Medicine 1985; 11: 5–12. Opthof T, de Jonge B, Schade B, Jongsma HJ, Bouman LN. Cycle length dependence of the chronotropic effects of adrenaline, acetylcholine, Ca2+ and Mg2+ in the guinea-pig sinoatrial node. Journal of the Autonomic Nervous System 1984; 11: 349–366. Akazawa S, Shimizu R, Nakaigawa Y, Ishii R, Ikeno S, Yamato R. Effects of magnesium sulphate on atrioventricular conduction times and surface electrocardiogram in sevoflurane-anaesthetized dogs. British Journal of Anaesthesia 1997; 78: 75–80. Priebe HJ. Myocardial ischemia during vasodilator therapy in a canine model of pulmonary hypertension and coronary insufficiency. Anesthesiology 1992; 76: 781–791. Cox RH. Influence of pentobarbital anesthesia on cardiovascular function in trained dogs. American Journal of Physiology 1972; 223: 651–659.
Effects of magnesium sulphate on the cardiovascular system, coronary circulation and myocardial metabolism in anaesthetized dogs
Y. NAKAIGAWA, S. AKAZAWA, R. SHIMIZU, R. ISHII, S. IKENO, S. INOUE AND R. YAMATO
Summary We have studied the effects of i.v. bolus doses of magnesium sulphate (MgSO4) 60, 90 and 120 mg kg1 on haemodynamic state, the coronary circulation and myocardial metabolism in nine dogs anaesthetized with pentobarbitone and fentanyl. MgSO4 produced dose-dependent decreases in arterial pressure, heart rate, left ventricular dP/dtmax and left ventricular minute work index (LVMWI) and an increase in the time constant of left ventricular isovolumic relaxation. Stroke volume increased, systemic vascular resistance decreased and cardiac output did not change significantly. MgSO4 produced decreases in coronary perfusion pressure, coronary vascular resist• ance and myocardial oxygen consumption (MVO2) Coronary sinus blood flow, lactate extraction ratio • and the ratio of LVMWI to myocardial MVO2 , that is an index of cardiac efficiency, did not change significantly. This study indicated that the depressant effect of MgSO4 on cardiac function was offset by lowering of peripheral vascular resistance, so that cardiac pump function remained effective, and the almost constant coronary sinus blood flow resulted from the decrease in coronary vascular resistance even at higher doses. (Br. J. Anaesth. 1997; 79: 363–368). Key words Hypnotics barbiturate, pentobarbitone. system, effects. Heart, metabolism. magnesium sulphate. Dog.
Cardiovascular Pharmacology,
Magnesium sulphate has been used for decades in the management of eclampsia1 2 and pregnancyinduced hypertension.3 4 Recent reports on magnesium have emphasized its role in deficiency syndromes,5–7 including cardiac arrhythmias.8 9 10 The therapeutic importance and physiological roles of magnesium have been summarized by Iseri and French,11 and an extensive review on clinical experiences has been published by James.12 The evidence for cardiac depression by clinically useful concentrations of hypermagnesaemia is less clear, despite its wide clinical use. Some authors have demonstrated that there is no evidence of myocardial depression at any concentration of serum
magnesium,13 whereas others have reported that a single injection of magnesium sulphate leads to a decrease in cardiac performance.14 Although the cardiovascular haemodynamic effects of magnesium sulphate were assessed by some authors in baboons,13 sheep15 and humans,16 there have been few reports on magnesium-induced changes in left ventricular diastolic function. The effects of magnesium sulphate on the coronary circulation and myocardial metabolism have not been studied extensively. This study was designed to investigate the effects of magnesium sulphate on haemodynamic state, including left ventricular diastolic function, coronary circulation and myocardial metabolism, in open-chest dogs.
Materials and methods The study was approved by the management committee at Jichi Medical Laboratory of Experimental Medicine, based on the school’s Guide for Laboratory Animals, 1993. Nine mongrel dogs (mean weight 23.5 (range 18–30.5) kg) were anaesthetized with pentobarbitone 30 mg kg1. After tracheal intubation, the lungs were ventilated with a Harvard pump respirator ( FIO2 1.0). Arterial carbon dioxide tension was maintained at 4.8–5.3 kPa, as assessed by continuous measurement of end-tidal carbon dioxide concentration with a calibrated respiratory gas monitor (Model 5250, Ohmeda, Madison, WI.,USA) and repetitive analyses of arterial bloodgas tensions. Temperature was measured with an oesophageal thermistor and maintained at 36–37 C using an external heating pad. Anaesthesia was maintained during surgery with fentanyl 20 g kg1 h1 and 0.5–2.0 % sevoflurane in pure oxygen. All animals were paralysed with bolus doses of pancuronium 0.2 mg kg1 followed by small doses as needed. Heart rate (HR) was monitored by electrocardiography (ECG, lead II), and systolic, diastolic and YASUSHI NAKAIGAWA, MD, SATOSHI AKAZAWA, MD, REIJU SHIMIZU, MD, RYOUSUKE ISHII, MD, SHIGEO IKENO, MD, SOICHIRO INOUE, MD, REIKO YAMATO, MD, Department of Anaesthesiology, Jichi Medical School, Tochigi, 329–04, Japan. Accepted for publication: April 25, 1997. Correspondence to Y. N.
364
British Journal of Anaesthesia
mean arterial pressures (SAP, DAP and MAP, respectively) were measured with a micromanometer-tipped catheter (7F 45326, Toyoda, Ltd, Tokyo, Japan) placed via the left femoral artery into the abdominal aorta. A catheter was placed in the abdominal aorta via the right femoral artery for arterial blood sampling. Lactated Ringer’s solution was infused at 5–10 ml kg1 h1 into the right femoral vein throughout the experiment. After median sternotomy, an appropriately sized electromagnetic flow transducer (FB-140T, Nihon Kohden, Ltd, Tokyo, Japan) was placed around the root of the ascending aorta and connected to a flowmeter (MFW2100, Nihon Kohden, Ltd, Tokyo, Japan) for measurement of ascending aortic flow and cardiac output (CO). A micro-tip catheter (8F PC380, Millar, Inc.) was inserted through a stab wound in the apex of the left ventricle for measurements of left ventricular pressure (LVP) and left ventricular enddiastolic pressure (LVEDP). The rate of increase of left ventricular pressure (LVdP/dt) was obtained with an analogue differentiating circuit incorporated with an analogue-to-digital converter (Contractility Unit 1323, NEC San-ei, Ltd, Tokyo, Japan) and used as an index of left ventricular contractile function. Coronary sinus blood flow (CSBF) was measured by the retrograde thermodilution technique with a catheter (Thermoflow CCS-7U-90A, Webster Labs., Inc.) inserted into the coronary sinus via the right auricle.17 Cold normal saline as the indicator was infused at 20 ml min1 for 20 s with an infusion pump (SW-367, Sage Inst., Inc.) through the catheter for each measurement of CSBF. A pair of piezoelectric crystals (5 MHz, Sonotek Co., San Diego, CA, USA) were inserted into the subendocardium in the anterior and posterior walls of the left ventricle and connected to an amplifier (amplifier Unit 4105, NEC San-ei., Ltd, Tokyo, Japan) for measurement of left ventricular enddiastolic and end-systolic diameters (LVEDD and LVESD). Fractional shortening (FS) was obtained using the following formula and used as an index of left ventricular contractile function: FS (LVEDD– LVESD)/LVEDD100. All variables were recorded continuously on a polygraph (RECTI-HORIZ-8K23, NEC San-ei, Ltd, Tokyo, Japan).
Arterial and coronary sinus blood was sampled simultaneously to measure pH, PO2, PCO2, base excess, standard bicarbonate and haemoglobin concentrations (288 Blood Gas System, Ciba Corning), lactate, non-esterified fatty acid (NEFA) and magnesium concentrations. Lactate concentrations were measured using an enzymatic method. • Myocardial oxygen consumption (MV O2) , the ratio • of left ventricular minute work index to MV O2 • (LVMWI/MV O2) myocardial oxygen extraction ratio (MO2ext.), myocardial lactate extraction ratio (MLext.) and myocardial non-esterified fatty acid extraction ratio (MNEFAext.) were calculated. All derived variables and their abbreviations are shown in table 1. Left ventricular pressure (LVP), LVdP/dt, LVEDP and ECG signals were fed to a multichannel photographic oscillograph (Visigraph 5137, NEC San-ei Inst., Ltd, Tokyo, Japan) and recorded at a paper speed of 500 mm s1 to obtain the time constant of left ventricular isovolumic relaxation (T). T was calculated according to the method of Weiss, Frederiksen and Weisfeldt18 and used as an index of left ventricular relaxation. At the end of the surgical procedure, sevoflurane was discontinued and fentanyl was infused at a rate of 20 g kg1 h1. Approximately 30 min was allowed after withdrawal of sevoflurane for stabilization of haemodynamic variables and baseline values were obtained. All dogs were given a 12.3% solution of magnesium sulphate in a bolus dose of 60, 90 and 120 mg kg1 (0.49, 0.73 and 0.98 ml kg1, respectively) over 10 s at 20-min intervals consecutively. The ratio of osmolarity of magnesium sulphate to that of saline is 3.0 : 1.0. Haemodynamic measurements and blood sampling were performed 1 min before and after each administration of magnesium sulphate. All values are presented as mean (SEM). Data were analysed statistically using analysis of variance for repeated measurements (ANOVA). Where appropriate, this was followed by the Scheffe F test. A probability of chance occurrence less than 5% (P0.05) was considered statistically significant.
Table 1 Derived variables. BWBody weight (kg); COcardiac output (litre min1); HRheart rate (beat min1); MAPmean arterial pressure (mm Hg); LVEDPleft ventricular end-diastolic pressure (mm Hg); CVPcentral venous pressure (mm Hg); LVEDDleft ventricular end-diastolic diameter (mm); LVESDleft ventricular endsystolic diameter (mm); DAPdiastolic arterial pressure (mm Hg); CSBFcoronary sinus blood flow (ml min1); (a–cs) DO2arterio-coronary sinus oxygen content difference (ml dl1); CaO2 arterial oxygen content (ml dl1); (a–cs) DLarterio–coronary sinus lactate concentration difference (mg dl1); Laarterial lactate concentration (mg dl1); (a–cs) DNEFAarterio- coronary sinus non-esterified fatty acid concentration difference (mEq litre1); NEFAaarterial non- esterified fatty acid concentration (mEq litre1) CI SV SVI LVMWI SVR FS CPP CVR MV!O2 MO2 ext. MLext. MNEFAext.
Cardiac indexCO/BW Stroke volume(CO/HR) 103 Stroke volume indexSV/BW Left ventricular minute work index(MAP–LVEDP) SVIHR0.0136 Systemic vascular resistance(MAP–CVP)/CO80 Fractional shortening(LVEDD–LVESD)/LVEDD100 Coronary perfusion pressureDAP–LVEDP Coronary vascular resistanceCPP/CSBF80 Myocardial oxygen consumption(a–cs) DO2CSBF102 Myocardial oxygen extraction ratio((a–cs) DO2/CaO2) 102 Myocardial lactate extraction ratio((a–cs) DL/La102 Myocardial non-esterified fatty acid extraction ratio((a–cs) DNEFA/NEFAa) 102
(litre min1 kg1) (ml) (ml kg1) (g m kg1 min1) (dyn s cm5) (%) (mmHg) (kdyn s cm5) (ml min1) (%) (%) (%)
Magnesium: haemodynamics and myocardial metabolism
365
Table 2 Effects of magnesium sulphate (MgSO4) on cardiovascular state (mean (SEM)). MgSerum concentration of magnesium; HRheart rate; SAPsystolic arterial pressure; DAPdiastolic arterial pressure; MAPmean arterial pressure; LVEDPleft ventricular end-diastolic pressure; COcardiac output; CIcardiac index; SVstroke volume; SVIstroke volume index; LVMWIleft ventricular minute work index; SVRsystemic vascular resistance; LVdP/dtmaxmaximum rate of increase of left ventricular pressure; LVEDDleft ventricular end-diastolic diameter; LVESDleft ventricular end-systolic diameter; FSfractional shortening; Ttime constant of fall in isovolumic left ventricular pressure. *P0.05, **P0.01 compared with baseline. †P0.05, ††P0.01 compared with MgSO4 60 mg kg1 MgSO4 (mg kg-1)
Mg (mg dl1) HR (beat min1) SAP (mm Hg) DAP (mm Hg) MAP (mm Hg) LVEDP (mm Hg) CO (litre min1) CI (litre min1 kg1) SV (ml) SVI (ml kg1) LVMWI (g m kg1 min1) SVR (dyn s cm-5) LVdP/dtmax (mm Hg s1) LVEDD (mm) LVESD (mm) FS (%) T (ms)
Baseline
60
90
120
1.3 (0.1) 109 (8) 145 (6) 99 (7) 115 (6) 3.9 (0.5) 2.2 (0.1) 0.10 (0.01) 21.3 (2.3) 0.98 (0.16) 146 (14) 4376 (506) 2468 (210) 36.0 (6.8) 42.8 (6.2) 19.1 (4.2) 48.0 (4.4)
6.3 (0.3)** 93 (6) 122 (6)** 72 (6)** 89 (6)** 4.1 (0.5) 2.4 (0.1) 0.11 (0.01) 26.6 (2.4) 1.21 (0.17)* 121 (12) 3076 (335)* 1974 (194)** 35.7 (6.6) 42.9 (6.2) 20.3 (3.6) 55.6 (5.1)*
9.7 (0.5)**†† 79 (5)**† 114 (6)** 65 (6)** 81 (6)** 4.7 (0.6) 2.3 (0.2) 0.10 (0.01) 29.3 (2.9)* 1.32 (0.19)** 104 (14)* 3058 (372)** 1721 (206)** 36.3 (6.6) 43.8 (6.0) 20.0 (4.0) 62.3 (5.8)**
14.4 (0.7)**†† 69 (4)**†† 97 (7)**†† 53 (6)**†† 69 (6)**†† 4.5 (0.6) 2.0 (0.1) 0.09 (0.01) 29.1 (3.0)* 1.29 (0.18)** 74 (9)**†† 2833 (303)** 1320 (188)**†† 36.6 (6.5) 43.3 (5.5) 20.0 (3.6) 67.5 (5.2)**††
Figure 1 Changes in cardiac output (CO) (!), heart rate (HR) (▲) and stroke volume (SV) (■) (mean, SEM) after administration of magnesium sulphate (MgSO4). *P0.05, **P0.01 compared with baseline (B).
Results Serum concentrations of magnesium 1 min after each bolus administration and the effects on haemodynamic state are shown in table 2. Haemodynamic variables decreased within the first minute and gradually returned to near baseline values after 20 min. Serum concentrations of magnesium before and after administration of magnesium sulphate of 60, 90 and 120 mg kg1 were 1.3 (0.1), 6.3 (0.3), 9.7 (0.5) and 14.4 (0.7) mg dl1, respectively. Magnesium sulphate produced dose-dependent
Figure 2 Changes in cardiac output (CO) (!), mean arterial pressure (MAP) (▲) and systemic vascular resistance (SVR) (■) (mean, SEM) after administration of magnesium sulphate (MgSO4). *P0.05, **P0.01 compared with baseline (B).
Table 3 Effects of magnesium sulphate (MgSO4) on electrocardiogram (mean (SEM)). SCLSinus cycle length; QTcQT interval corrected for heart rate (Bazett’s formula). *P0.05, **P0.01 compared with baseline. †P0.05, ††P0.01 compared with MgSO4 60 mg kg1 MgSO4 (mg kg1)
SCL (ms) PQ (ms) QRS (ms) QT (ms) QTc (ms)
Baseline 60
90
120
571 (40) 124 (6) 79 (3) 429 (25) 570 (26)
783 (56)**†† 168 (8)**† 84 (2)** 513 (27)**†† 584 (29)
895 (53)**†† 184 (8)**†† 86 (2)** 563 (29)**†† 599 (29)*
664 (42)** 148 (6) 80 (2) 481 (26)**†† 592 (27)*
366
British Journal of Anaesthesia Table 4 Effects of magnesium sulphate (MgSO4) on coronary circulation and myocardial metabolism (mean (SEM)). CPPCoronary perfusion pressure; CSBFcoronary sinus blood flow; COcardiac output; CVRcoronary ˝ vascular resistance; MV O2 myocardial oxygen consumption; MO2ext. myocardial oxygen extraction ratio; MLext. myocardial lactate extraction ratio; MNEFAext. myocardial non-esterified fatty acid extraction ratio; LVMWIleft ventricular minute work index. *P0.05, **P0.01 compared with baseline. †P0.05, ††P0.01 compared with MgSO4 60 mg kg1 MgSO4 (mg kg1)
CPP (mm Hg) CSBF (ml min1) CSBF/CO (%) CVR (kdyn s cm5) ˝ MV O2 (ml min1) MO2ext. (%) MLext. (%) MNEFAext. (%) LVMWI/MV!O2 (g m kg1 litre1)
Baseline
60
90
120
90 (8) 91.7 (4.4) 4.4 (0.3) 81 (10) 5.4 (0.3) 44.3 (2.1) 42.1 (4.2) 20.4 (2.3) 24.0 (2.3)
64 (7)** 112.5 (12.1) 5.0 (0.6) 49 (6)** 4.8 (0.6) 36.6 (4.4) 37.1 (4.5) 24.1 (5.2) 24.0 (3.7)
57 (7)** 106.9 (12.9) 5.4 (0.8) 47 (7)** 3.9 (0.6) 32.7 (4.1)* 33.8 (4.6) 27.0 (7.1) 23.1 (2.4)
40 (4)**†† 89.2 (15.7) 4.7 (0.8) 42 (6)** 2.7 (0.3)**†† 31.6 (4.4)* 31.7 (7.9) 34.9 (9.5) 22.3 (3.7)
decreases in HR, SAP, DAP, MAP, LVMWI and LVdP/dt max (P0.01). Stroke volume increased significantly but cardiac output did not change significantly (fig. 1). The near constant CO may have been brought about by the decrease in systemic vascular resistance (SVR) (fig. 2). LVEDD, LVESD and FS did not change significantly. T increased in a dose-dependent manner, indicating a decrease in left ventricular diastolic function. The effects of magnesium sulphate on the electrocardiogram are shown in table 3. Magnesium sulphate increased sinus cycle length, PQ and QT intervals dose-dependently (P0.01). QRS duration and corrected QT interval (QTc) increased significantly, but not dose-dependently. P wave, ST segment and T wave configurations did not change significantly. Magnesium sulphate did not produce arrhythmia. The effects of magnesium sulphate on the coronary circulation and myocardial metabolism are shown in table 4. Magnesium sulphate produced dose-dependent decreases in coronary perfusion • pressure (CPP) and MV O2 (P0.01). Coronary vascular resistance (CVR) and MO2ext. ratio decreased, but not dose-dependently. Any significant changes in CSBF, the ratio of CSBF to CO, MLext. ratio, MNEFAext. ratio and • LVMWI/MV O2 were not observed at each dose of magnesium sulphate.
Discussion The major findings of this study were that magnesium sulphate, in a dose-dependent manner, lowered systemic vascular resistance, arterial pressure and heart rate, while cardiac output was well maintained. Cardiac work, derived from both stroke volume and heart rate, and myocardial oxygen consumption, decreased markedly in parallel with systemic vascular resistance. Magnesium sulphate maintained coronary perfusion and cardiac efficiency even at higher doses. With an unchanged cardiac output, the decrease in arterial pressures must result from a decrease in systemic vascular resistance, which may be caused mainly by the vasodilator effects of magnesium sulphate that have been demonstrated in vitro,19 20 in
intact animals13 21 and in humans.15 The cardiovascular effects of magnesium are as follows: (1) inhibition of the normal rhythm or vasomotion (spontaneous mechanical activity) of vascular smooth muscle, (2) depression of the contractile responses to endogenous neurohumoral substances that maintain vascular tone,22 (3) possibly blockade of sympathetic ganglia23 and (4) negative inotropism. In humans, Mroczek, Lee and Davidov15 reported that infusion of magnesium sulphate resulted in a rapid but transient decrease in arterial pressure in hypertensive patients, whereas normotensive subjects did not demonstrate any appreciable change in arterial pressure. McCubbin and co-workers24 reported cardiopulmonary arrest caused by acute maternal hypermagnesaemia (35.1 mg dl1). In animals, transient16 and marked13 25 hypotension has been reported. This transient hypotension may be explained partially by rapid clearance of magnesium from plasma through the kidney. James, Cork and Dennett13 demonstrated the difficulties of maintaining high serum magnesium concentrations in the presence of normal renal function. They administered in the baboon a loading dose of 60 mg kg1 over 1 min and then infused continuously 12.5 mg min1 for at least 5 min. The infusion rate was then increased to 25 mg min1 and by 25 mg min1 every 5 min up to a maximum of 200 mg min1. The maximum serum magnesium concentration attained was only 9.4 mmol litre1. Dandavino and colleagues16 observed that rapid administration of magnesium sulphate produced hypotension but a more gradual administration did not produce hypotension. In our study, magnesium sulphate at bolus doses of 60, 90 and 120 mg kg1 administered every 20 min produced a dose-dependent decrease in arterial pressure. This decrease in arterial pressure corresponded to increases in serum concentrations of magnesium. Magnesium sulphate is known to have a negative inotropic action and cause myocardial depression and cardiac arrest in massive doses.24 In this study, magnesium sulphate produced a dose-dependent decrease in LVdP/dtmax. Although the decrease in LVdP/dtmax can be explained partially by the decrease in heart rate,26 this indicates that magnesium sulphate exerted a dose-dependent
Magnesium: haemodynamics and myocardial metabolism negative inotropic action in the whole animal. Left ventricular pump function, however, was well compensated by the increase in stroke volume caused by the lowering of peripheral vascular resistance. In fact, magnesium sulphate 60 and 90 mg kg1 produced a significant increase in stroke volume despite a marked decrease in LVdP/dtmax. Cardiac output also was maintained well at a high dose of 120 mg kg1. The profound decrease in heart rate may be the main effect of magnesium sulphate, with the heart compensating by a Starling effect to increase stroke volume appreciably. This finding is consistent with previous reports.3 13 27 James, Cork and Dennett13 demonstrated that cardiac output was maintained at control values, at concentrations of serum magnesium greater than 5 mmol litre1. Cotton, Gonik and Dorman3 and Mroczek, Lee and Davidov15 also demonstrated that magnesium sulphate produced an increase in cardiac output despite significant myocardial depression. These findings and ours suggest that the direct myocardial depressant effect with magnesium may be compensated by decreased afterload, resulting in the maintenance of cardiac function as a pump. No significant change in fractional shortening as an index of left ventricular systolic function during ejection phase could be attributed to the decrease in afterload. Some authors16 25 have reported that tachycardia occurred during administration of magnesium sulphate, whereas others13 have reported that bradycardia was seen. In our study, magnesium sulphate produced a dose-dependent decrease in heart rate and prolongation of PQ and QTc intervals. The degree and rate of increase of blood magnesium concentration may partially explain the different findings observed among these studies. At a low serum concentration of magnesium, hypotension may produce tachycardia through the baroreflex pathway, whereas at a high concentration bradycardia may be brought about by both indirect and direct inhibitory effects of magnesium on the sinoatrial node. As for indirect inhibitory effects, Stanbury23 found that magnesium induces a short-lived blockade of the sympathetic ganglia and prevents the stimulating effects of potassium and acetylcholine on the superior cervical ganglion in anaesthetized cats. As for direct inhibitory effect, Opthof and colleagues28 demonstrated that magnesium itself decreased the rate of diastolic depolarization of the pacemaker cells in the sinus node. In this study, magnesium sulphate did not produce any arrhythmias or conduction disorders at serum concentrations of 6.3 (0.3) to 14.4 (0.7) mg dl1, although it produced a dose-dependent decrease in heart rate and prolongation of PQ, QRS and QTc intervals. This finding is consistent with our earlier study29 except for QTc interval. The difference of basal anaesthesia may be associated with the different results of QTc interval. Although the effects of magnesium sulphate on systemic haemodynamic state and myocardial contractility have been described in other studies,13 15 16 the effect on ventricular diastolic function in vivo or in vitro remains unexplored. In this study, the effect
367 of magnesium sulphate on left ventricular relaxing function was investigated by the time constant of left ventricular isovolumic relaxation (T). The shorter the T value, the better the relaxing function of the left ventricle, and vice versa.18 Magnesium sulphate produced a dose-dependent increase in T in this study. Therefore, magnesium may depress left ventricular relaxing function. Because T tends to increase with a decrease in heart rate and depressed systolic fibre shortening,18 the decrease in T observed here appeared to be related not only to decreased left ventricular relaxing function but also to decreased heart rate and depressed systolic fibre shortening. The effects of magnesium sulphate on the coronary circulation and myocardial metabolism have not been studied extensively. This study demonstrated that magnesium sulphate did not change coronary sinus blood flow remarkably despite a decrease in coronary perfusion pressure. This suggests that magnesium sulphate may have a coronary vasodilating effect. A dose-dependent • decrease in MV O2 appeared to parallel the decreased • external work of the heart, because LVMWI/MV O2 did not change significantly. The decrease in LVMWI is thought to be caused mainly by reduction in arterial pressure. Lactate production was not observed throughout this study. Therefore, magnesium sulphate may maintain an appropriate oxygen supply-and-demand relation in the myocardium and may exert no major adverse effects on cardiac function as a pump. The findings of this study could be related to basal anaesthesia because of an acute open-chest canine preparation under general anaesthesia. Priebe demonstrated an anaesthetic technique comprising pentobarbitone, with fentanyl maintaining a stable, global haemodynamic state over a prolonged period.30 Cox31 demonstrated that haemodynamic variables returned to control levels within 15 min after pentobarbitone 30 mg kg1 i.v. in dogs, although heart rate remained increased from the control value of 85 beat min1 to 132 beat min1 even after 60 min. In this study, however, a heart rate of 125 (5) beat min1 immediately after administration of pentobarbitone decreased to a baseline value of 109 (8) beat min1 immediately before initiation of measurements. The osmolarity of magnesium sulphate solutions used could not produce any transient cardiovascular effects because the volumes were small and osmolarity might be lowered by the circulating blood of the dogs. Magnesium sulphate competes with calcium at the cell membrane and is regarded in many clinical situations as a physiological calcium blocker. This concept may be supported by the facts that magnesium sulphate (1) exerted a mild consistent, dose-dependent hypotensive effect, (2) preserved cardiac pump and coronary sinus blood flow despite a decrease in coronary perfusion pressure, (3) produced negative inotropic effect, (4) decreased myocardial oxygen consumption and (5) did not produce arrhythmia. These effects are thought to be beneficial for patients with ischaemic heart disease.
368 In future, magnesium will be used more widely in many medical situations, including anaesthetic practice as a physiological calcium blocker.
British Journal of Anaesthesia
16.
Acknowledgements This work was supported by a grant-in-aid for scientific research (B) from the Ministry of Education, Science and Culture, Japan. We thank K. Nozawa and N. Yoshida (NEC San-ei., Utsunomiya, Japan) for technical assistance.
References 1. Pritchard JA, Pritchard SA. Standardized treatment of 154 consecutive cases of eclampsia. American Journal of Obstetrics and Gynecology 1975; 123: 543–552. 2. Pritchard JA, Cunningham FG, Pritchard SA. The Parkland Memorial Hospital protocol for treatment of eclampsia: Evaluation of 245 cases. American Journal of Obstetrics and Gynecology 1984; 148: 951–963. 3. Cotton DB, Gonik B, Dorman KF. Cardiovascular alterations in severe pregnancy-induced hypertension: Acute effects of intravenous magnesium sulphate. American Journal of Obstetrics and Gynecology 1984; 148: 162–165. 4. Lee MI, Todd HM, Bowe A. The effects of magnesium sulphate infusion on blood pressure and vascular responsiveness during pregnancy. American Journal of Obstetrics and Gynecology 1984; 149: 705–708. 5. Turlapaty PDMV, Altura BM. Magnesium deficiency produces spasms of coronary arteries: Relationship to etiology of sudden death ischemic heart disease. Science 1980; 208: 198–200. 6. Vincent JL, Buset M, Dufaye P, Barre J, Degaute JP, Kahn RJ. Circulatory shock associated with magnesium depletion. Intensive Care Medicine 1982; 8: 149–152. 7. Iseri LT, Freed J, Bures AR. Magnesium deficiency and cardiac disorders. American Journal of Medicine 1975; 58: 837–846. 8. Cohen L, Kitzes R. Magnesium sulphate in the treatment of variant angina. Magnesium 1984; 3: 46–49. 9. Iseri LT, Fairshter RD, Hardemann JL, Brodsky MA. Magnesium and potassium therapy in multifocal arterial tachycardia. American Heart Journal 1985; 10: 789–794. 10. Solomon D, Bunegin L, Albin M. The effect of magnesium sulphate administration on cerebral and cardiac toxicity of bupivacaine in dogs. Anesthesiology 1990; 72: 341–346. 11. Iseri LT, French JH. Magnesium: Nature’s physiologic calcium blocker. American Heart Journal 1984; 108: 188–193. 12. James MFM. Clinical use of magnesium infusions in anesthesia. Anesthesia and Analgesia 1992; 74: 129–136. 13. James MFM, Cork RC, Dennett JE. Cardiovascular effects of magnesium sulphate in the baboon. Magnesium 1987; 6: 314–324. 14. Critelli G, Ferro G, Peschle C, Perticone FR, Rengo FR, Condorelli M. Myocardial contractility after injection or prolonged infusion of magnesium sulphate. Acta Cardiologica 1977; 32: 65–73. 15. Mroczek WJ, Lee WR, Davidov ME. Effect of magnesium
17.
18.
19. 20. 21. 22.
23. 24. 25. 26.
27. 28.
29.
30. 31.
sulphate on cardiovascular haemodynamics. Angiology 1970; 10: 720–724. Dandavino A, Woods JR, Murayama K, Brinkman CR, Assali NS. Circulatory effects of magnesium sulphate in normotensive and renal hypertensive pregnant sheep. American Journal of Obstetrics and Gynecology 1977; 127: 769–774. Ganz W, Tamura K, Marcus HS, Donoso R, Yoshida S, Swan HJC. Measurement of coronary sinus blood flow by continuous thermodilution in man. Circulation 1971; 44: 181–195. Weiss JL, Frederiksen JW, Weisfeldt ML. Haemodynamic determinant of the time course of fall in canine left ventricular pressure. Journal of Clinical Investigation 1976; 58: 751–760. Turlapaty PDMV, Altura BM. Extracellular magnesium ions control calcium exchange and content of vascular smooth muscle. European Journal of Pharmacology 1978; 52: 421–423. Altura BM, Alutra BT. Magnesium and contraction of arterial smooth muscle. Microvascular Research 1974; 7: 145–155. James MFM, Cork RC, Harlen GM, White JF. Interactions of adrenaline and magnesium on the cardiovascular system of the baboon. Magnesium 1988; 7: 37–43. Altura BM and Altura BT. General anesthetics and magnesium ions as calcium antagonists on vascular smooth muscle. In: Weiss GB, ed. New Perspectives on Calcium Antagonists. Washington: Maryland American Physiological Society, 1981; 131–145. Stanbury JB. The blocking action of magnesium ion on sympathetic ganglia. Journal of Pharmacology and Experimental Therapeutics 1948; 93: 52–62. McCubbin JH, Sibai BM, Abdella TN, Anderson GD. Cardiopulmonary arrest due to acute maternal hypermagnesaemia. Lancet 1981; 1: 1058. Maxwell GM, Elliott RB, Burnell RH. Effects of hypermagnesemia on general and coronary hemodynamics of the dog. American Journal of Physiology 1965; 208: 158–161. Msaon DT. Usefulness and limitations of the rate of rise of intraventricular pressure (dp/dt) in the evaluation of myocardial contractility in man. American Journal of Cardiology 1969; 23: 516–527. James MFM, Manson EDM. The use of magnesium sulphate infusions in the management of very severe tetanus. Intensive Care Medicine 1985; 11: 5–12. Opthof T, de Jonge B, Schade B, Jongsma HJ, Bouman LN. Cycle length dependence of the chronotropic effects of adrenaline, acetylcholine, Ca2+ and Mg2+ in the guinea-pig sinoatrial node. Journal of the Autonomic Nervous System 1984; 11: 349–366. Akazawa S, Shimizu R, Nakaigawa Y, Ishii R, Ikeno S, Yamato R. Effects of magnesium sulphate on atrioventricular conduction times and surface electrocardiogram in sevoflurane-anaesthetized dogs. British Journal of Anaesthesia 1997; 78: 75–80. Priebe HJ. Myocardial ischemia during vasodilator therapy in a canine model of pulmonary hypertension and coronary insufficiency. Anesthesiology 1992; 76: 781–791. Cox RH. Influence of pentobarbital anesthesia on cardiovascular function in trained dogs. American Journal of Physiology 1972; 223: 651–659.