Magnesium metabolism in open-heart surgery

Resuscitation,

Elsevier

13 (1986)

Scientific

MAGNESIUM

FRANC0 MONACO,

215-221

Publishers

215

Ireland

METABOLISM

Ltd.

IN OPEN-HEART

CAVALIERE, MARIO SCIARRA, SABRINA BONIFAZI and ROCCO

SURGERY

ROBERTO ZAMPARELLI, SCHIAVELLO

Universita Cattolica de1 S. Cuore, Istituto di Anestesiologia S-00168, Rome (Italy)

CAROLINA

e Rianimazione,

Largo A.

Gemelli

(Received (Accepted

November 5th, 1985) March 21st, 1986)

SUMMARY

The levels of magnesium in serum, urine and erythrocytes were studied in 22 patients undergoing cardiac surgery for valvular prothesis. Magnesium values were correlated with serum albumin and non-esterified fatty acids (NEFA). Data were collected before anesthesia, 10 min after sternotomy, heparinization and declamping of the aorta and in the 1st postoperative day. A slight decrease in magnesemia was observed before extracorporeal circulation (ECC) and was mainly due to haemodilution. The correlation of magnesium with NEFA was significant only after heparinization. The use of the St Thomas solution as cardioplegia fully corrected the hypomagnesemia previously reported during ECC as well as in the 1st postoperative day. A moderate hypermagnesemia was observed at the end of ECC, but no patient reached dangerous levels of serum magnesium. Urinary losses increased during and after ECC. Red blood cell magnesium showed a slight increase before ECC, followed by a significant reduction at the end of ECC. Key words: (NEFA)

Magnesium

-

Open heart

surgery - Non-esterified

fatty

acids

INTRODUCTION

The interest in magnesium changes in cardiopathic patients is justified by the cardiac anomalies that its deficiency or excess may cause. Magnesium abnormalities may increase the toxicity of other drugs, such as digitalis or antiarrhythmics. Symptomatic hypomagnesemia is characterized by ECG alterations, such as low voltage, ST-T abnormalities, the occurence of u waves, and an extended QT interval. Patients with hypomagnesemia may develop several cardiac arrhythmias, including Torsades de Pointes. Hyper0300-9572/86/$03.50 Printed and Published

0 1986 Elsevier in Ireland

Scientific

Publishers

Ireland

Ltd.

216

magnesemia causes a reduction in the atrial and-ventricular conductance, a retardation of the periodicity of the SA node, and a slowing of the AV transmission, bradycardia, prolongation of the PR interval and intraventricular conduction defects (Dyckner and Wester, 1982; Top01 and Lerman, 1983; Schipperheyn, 1984). Several authors have studied magnesium changes in patients undergoing cardiac surgery (Scheinman, Sullivan and Hyatt, 1969; Scheinman, Sullivan, Hutchinson and Hyatt, 1971; Turnier; Osborn, Gerbode and Popper, 1972; For&r, Stalder, Bloch and Suter, 1981; Bunton, 1983; Sciarra, Marana, Cavaliere, Loffreda and Schiavello, 1983). All have observed a post-operative decrease, similar to that observed after general surgery (Sawyer, Drew, Gesink, Sawyer and Sawyer, 1969). This effect is unrelated to perioperative myocardial infarction (Forster et al., 1981; Bunton, 1983). An initial deficit is often tied to diuretic therapy (Lim and Jacob, 1972), but it is also frequently observed in seriously ill patients (Ryzen, Wagers, Singer and Rude, 1985). Alternatively, haemodilution, urinary losses, exchange with the intracellular compartment, perioperative intake, blood transfusions and NEFA increases may influence the values of magnesemia (McLellan, Reid and Lane, 1984; Flink, Brick and Shane, 1981). The purpose of the present study was to elucidate the importance of some of these factors in influencing serum magnesium levels during openheart surgery. PATIENTS

AND

METHODS

Twenty-two patients (aged 22-65; mean 51.6; 10 males), undergoing open-heart surgery for valvular prothesis, were studied. They were premedicated with morphine (0.1 mg/kg) and atropine or scopolamine (0.01 mg/kg). Anesthesia was induced with diazepam (0.3-0.4 mg/kg) and maintained with fentanyl (0.03-0.05 mg/kg, subdivided in smaller doses). Myprelaxation was obtained with pancuronium bromide (0.1 mg/kg). Patients were connected to an Engstrom 300 respirator and ventilated with an Oz/NzO mixture .(ratio 1: 1): The extracorporeal apparatus was a roller pump with a Temptrol Q bubble oxygenator. The circuit was primed with Ringer Lactate to give an haemodilution with an haematocrit of 25%. During the bypass the blood flow was maintained between 2000 and 2500 ml/m2 per min and the internal temperature was lowered to 25-28°C. At the beginning of the cardiopulmonary bypass, methylprednisolone (30 mg/kg), mannitol as 18% solution (0.5 g/kg), and diazepam (1 mg/kg per h) were administered to all the patients. Myocardial protection was obtained using cold St Thomas solution (Mg: 32 mEq/l) as cardioplegia: 500 ml/m2 b.s. were infused initially and repeated if necessary (total amount 700-3500 ml; mean 1300 ml). Patients received l-12 titrated blood units (mean 4.1)

217

a.nd, in 7 subjects, 2 heparinized self blood units. Periodically, blood samples were obtained for K and Ca and abnormalities were corrected. Five blood samples were collected from every patient: (1) before the induction of the anesthesia, (2) 10 min following sternotomy, (3) 10 min after administering heparin, (4) 10 min after declamping&he aorta, and (5) in the 1st postoperative day. Urine was collected during surgery, before, during and after ECC. Using a Varian DMS 90 spectrophotometer, samples were analyzed for magnesium in serum, urine and erythrocytes using Calmagite (Mg Kit, Biomerieux), for albumin (BCG test, Sclavo Diagnostic), and, in 11 patients, for NEFA (NEFAC kit, WAKO Pure Chemical Industries Ltd.). Ten healthy subjects of the medical staff were utilized as control group for magnesium in serum and in erythrocytes and for NEFA. Consecutive data were compared by paired t-test; the difference between controls and samples number 1 was tested by unpaired t-test. Four multiple correlations among the values of serum magnesium (y), albumin (xl) and NEFA (x2) were carried out utilizing the values of samples 1 and 2,1 and 3, 1 and 4, 1 and 5. Results were evaluated by the F of Fisher and, for the single coefficients of xl and x2, by T distribution (Armitage, 1971). RESULTS

The means and standard deviations of the collected data are reported in Table I. Before anesthesia, no significative difference was observed between patients and controls, except in NEFA values. Only one patient presented hypomagnesemia (1.5 mEq/l), but seven subjects showed levels of erythrocyte magnesium lower than 2.5 mEq/l. Samples 2 and 3, collected before ECC, showed a small decrease in serum magnesium (1-2 P < 0.05; 2-3 P < 0.05) and of serum albumin (1-2 P < 0.05; 2-3 P < 0.01); Magnesium increased slightly in red blood cells (l-2 P < 0.01)and NEFA rose (controls-l P < 0.01; 2-3 P < 0.01). The analysis of variance applied to the multiple correlations among the serum values of magnesium, albumin and NEFA (Table II) resulted significant differences only for 1-3 (P< O.Ol), like the T of the coefficient of NEFA (P < 0.05). The coefficient of albumin, however, was significant also for l--2 (1-2 P < 0.05; l-3 P < 0.05). Following ECC, there was a marked rise of magnesemium (P< 0.001)and a decrease of red blood cell magnesium (P < 0.05), albumin (P < O.OOl), and NEFA (P < 0.01). In the 1st postoperative day, values returned to normal. In spite of the marked increase of magnesemia observed after ECC, we did not find any value higher than 4 mEq/l, a level at which symptoms of hypermagnesemia frequently occur (Rude and Singer, 1981).

I

Urine Mg (mEq)

(mEq/i) Eryth. Mg (mEq/i) Albumin (g/di) NEFA (mEq/i)

Serum mg

0.47

-

3.2

1.7

C

between

Pre-ECC

? 0.14

? 0.3

k 0.2 ? 1.0 ? 0.6 It 0.37**

3.8

0.84

+ 0.4

3.3

1.6

1

(5)

3.1

2 0.6*

ECC 37.6

+ 71.9

2.92

3.4

1.2

3

**P < 0.01;

+ 0.75**

+ O.B**

% 0.7

k 0.5*

*P < 0.05;

+ 0.9**

? 0.3*

by the f-test.

1.02 + 0.34

3.3

3.5

1.4

2

values were evaluated

4.9 + 7.2

consecutive

DAY

+_0.6***

2.1

Post-ECC

1.02 % 0.39**

+ 0.5*

? 0.7***

< 0.001.

3.1

2.2

4

***P

41.9

+ 49

0.46

3.6

3.3

1.5

5

(4),

t 0.21**

? 0.4***

+ 0.6*

+ 0.5***

S.D. OF THE VALUES OBSERVED IN THE CONTROL GROUP (C), BEFORE ANESTHESIA (l), 10 MIN THE STERNOTOMY (2), 10 MIN AFTER GIVING HEPARIN (3) AND AFTER DECLAMPING THE AORTA

IN THE 1 st POSTOPERATIVE

The differences

AND

MEANS AND FOLLOWING

TABLE

219 TABLE

II

MULTIPLE ALBUMIN

CORRELATIONS (x2)

The degrees of freedom xl and x2 are reported.

AMONG

l-4 l-5

MAGNESIUM

(y),

NEFA

(xl)

AND

(d.f.), the F of Fisher (F) and the T values for the coefficients *P < 0.05; **P < 0.01.

of

d.f.

-. l-2 l-3

SERUM

y = 0.88 t0.03xl y = 1.22 -0.17~1 y = 2.20 +0.68x1 y = 0.58 t 0.15x1

+0.27x2 +0.24x2

2.19 2.19

2.70 8.47**

0.12 2.11*

-0.15x2 +0.32x2

2.19 2.19

1.72 2.30

1.45 0.62

Urinary losses were low before patients during ECC.

ECC, and increased

in the majority

2.26* 2.31* 0.99 1.88

of the

DISCUSSION

In a former study, we observed significant changes in magnesemia during open-heart surgery. The decrease noted during ECC was explained by haemodilution with priming solution poor in magnesium (Sciarra et al., 1983). We pointed out that hypomagnesemia may cause important disturbances of cardiac activity at the end of ECC, a very delicate phase of cardiac surgery. In our present study we utilized the St Thomas solution as cardioplegia. This solution is very rich in magnesium and fully corrected the hypomagnesemia previously reported during ECC. However, no patient presented dangerous levels of serum magnesium and the usual postoperative hypomagnesemia (Scheinman et al., 1969; Scheinman et al., 1971; Turnier et al., 1983) was not observed. Before ECC, we found a slight decrease in serum magnesium probably due to haemodilution, or the increase of NEFA. To evaluate the degree of haemodilution, we studied the changes in serum albumin, a protein which binds about 30% of the serum magnesium (Edel and Gunther, 1980). We observed a progressive decrease of albumin until the end of ECC, followed by a return to the initial values. Magnesium and NEFA levels are inversely proportional in patients affected by myocardial infarction and Flink et al. hypothesized that NEFA may bind this ion (Flink et al., 1981). Our patients developed high values of NEFA, probably caused by the operative stress. An important rise was also observed after giving heparin, because of the clarifying action of this drug, The multiple correlations among magnesium, albumin and NEFA in serum indicate that the decrease in magnesium before ECC was mainly caused by

220

haemodilution. However, the rise of NEFA after giving heparin may also be important. After ECC no significative correlation was observed, probably because of the external intake of magnesium. The evaluation of the intracellular content of magnesium is difficult because it is often poorly correlated to the serum levels. We determined the red blood cell content despite its poor correlation with that of muscles (Alfrey, Miller and Butkus, 1974). The initial data were not significantly different from those of the controls, but in seven subjects values were very low (2.5 mEq/l), in spite of a normal serum magnesium. During surgery, magnesium showed two significant changes in erythrocytes: an increase before ECC and then a reduction. This trend was very different from that observed in serum and its cause is not clear. Hypothermia may cause the decrease found at the end of ECC, by means of a reduction of active cellular transport. In conclusion, our data show that the decrease of the magnesemia during cardiac surgery is mainly caused by haemodilution. The use of the St Thomas solution as cardioplegia fully corrects this hypomagnesemia. Hypermagnesemia may occur at the end of ECC but it does not reach dangerous levels. On the contrary, the intake of magnesium may be advantageous in two ways: (1) to correct, in part, a possible preexisting magnesium deficit, and (2) to reduce the output from the cells if the reduction of red blood cell magnesium will be confirmed. Finally, it must be remarked that the use of solutions with a high content of magnesium is dangerous in some clinical conditions, such as renal failure and preexisting disturbances of intracardiac conduction. REFERENCES Alfrey, A.C., Miller, N.L. and Butkus, D. (1974) Evaluation of body magnesium stores. J. Lab. Clin. Med., 84, 153-162. Armitage, P. (1971) Statistical methods in medical research, Blackwell Scientific Publ., Oxford, p. 310. Bunton, R.W. (1983) Value of serum magnesium estimation in diagnosing myocardial infarction and predicting dysrhythmias after coronary artery bypass grafting. Thorax, 38, 946-950. Dychner, T. and Wester, P.O. (1982) Magnesium in cardiology. Acta Med. Stand. (Suppl.), 661, 27-31. Edel, H. and Gunther, T. (1980) Magnesium metabolism: a review. J. Clin. Chem. Clin. B&hem., 18, 257-270. Flink, E.B., Brick, J.E. and Shane, S.R. (1981) Alterations of long-chain free fatty acid and magnesium concentrations in acute myocardial infarction. Arch. Int. Med., 141, 441-443. Forster, A., Stalder, R., Bloch, A. and Suter, P. (1981) The value of serum magnesium estimations in the diagnosis of acute perioperative myocardial infarction aftercoronary artery surgery. J. Cardiovasc. Surg., 22, 163-165. Lim, P. and Jacob, E. (1972) Magnesium deficiency in patients on long term diuretic therapy for heart failure. Br. Med. J., 3, 620-624. MC Lellan, B.A., Reid, S.R. and Lane, P.L. (1984) Massive blood transfusion causing hypomagnesemia. Crit. Care Med., 12, 146-147.

221 Rude, R.K. and Singer, F.R. (1981) Magnesium deficiency and excess. Ann. Rev. Med., 32, 245-259. Ryzen, E., Wagers, P.W., Singer, F.R. and Rude, R.K. (1985) Magnesium deficiency in a medical ICU population. Crit. Care Med., 13, 19-21. Sawyer, R.B., Drew, H.A., Gesink, H.H., Sawyer, K.C., Jr and Sawyer, K.C. (1969) Postoperative magnesium metabolism. Arch. Surg., 100, 343-348. Scheinman, M.M., Sullivan, R.W., Hutchinson, J.C. and Hyatt, K.H. (1971) Clinical significance of changes in serum magnesium in patients undergoing cardiopulmonary bypass. J. Cardiovasc. Surg., 61, 135-140. Scheinman, M.M., Sullivan, R.W. and Hyatt, K.H. (1969) Magnesium metabolism in patients undergoing cardiopulmonary bypass. Circulation. XXXIX and XL, Suppl. 1. Schipperheyn, J.J. (1984) The pathophysiology of potassium and magnesium disturbances. A cardiac perspective. Drugs, 28 (Suppl. l), 112-119. Sciarra, M., Marana, E., Cavaliere, F., Loffreda, C. and Schiavello R. (1983) Changes of magnesium ions in open-heart surgery. Resuscitation, 10, 253-257. Topol, E.J. and Lerman, B.B. (1983) Hypomagnesemic Torsades De Pointes. Am. J. Cardiol., 52, 1367-1368.