Acta anaesth. scand. 1976, 20, 421-428
The Haemodynamic Effects of Nitrous Oxide Anaesthesia on Myocardial Blood Flow in Dogs 0. DOTTORI, E. HAGGENDAL, E. LINDER, G. NORDSTROM and T. SEEMAN Departments of Anaesthesiology and Surgery 11, University of Goteborg, Goteborg, Sweden The effects of ventilation with nitrous oxide in oxygen on myocardial blood flow and oxygen metabolism were investigated in 31 mongrel dogs. The results of this study showed that, compared with controls, hyperoxic nitrous oxide mixtures did not cause any great changes in myocardial haemodynamics, despite a decrease in cardiac output and an increase in systemic vascular resistance. Normoxic nitrous oxide mixtures produced an increase of the coronary blood flow due to decreased coronary vascular resistance. To what extent this coronary vasodilatation resulted from an increased myocardial metabolism or from a direct effect of nitrous oxide on the coronary vascular bed cannot be quantified from the present results.
Received I0 April, accepted f o r publication 15 December I975
Unsupplemented nitrous oxide was intro- release of vasodilatator metabolites (BERNE duced into clinical anaesthesia as the agent of 1964) or directly by decreased myocardial choice in “poor-risk’’ patients (HELLER et al. oxygen tension (GREGG& FISCHER 1963). 1956) and cardiac patients (GRAY& RIDING Moreover, previous studies of the coronary 1957, DOTTORI & EKESTROM 1964), mainly circulation have shown that the hyperoxia because it is presumed to have no deleterious results in a decreased coronary blood flow effects on the cardiovascular system. How- (SOBOL et al. 1962, LAMMERANT et al. 1969, ever, the effect of unsupplemented nitrous among others) and in a reduced myocardial oxide on the coronary blood flow is still contractility (DANIELL & BAGWELL 1968). unknown. The aim of the present investigation, A previous experimental study indicated therefore, was to study the effects of nitrous that nitrous oxide exerted different effects oxide on the coronary blood flow in normoxic on the systemic circulation depending on and hyperoxic gas-mixtures. the amount of oxygen in the gas-mixture. Hyperoxic nitrous oxide gas-mixtures seemed to produce vasoconstriction com- MATERIAL AND METHODS bined with decreased cardiac output, while The studies were carried out on a total of 31 mongrel normoxic nitrous oxide gas-mixtures had the dogs (16-39 kg) of either sex. The dogs were premedicated with diazepam (0.4-0.5 mg/kg body weight) opposite effect (DOTTORI et al. 1976). and atropine (0.01-0.02 mg/kg body weight). AnThe normal coronary vascular bed ex- aesthesia was induced by 80% nitrous oxide in oxygen hibits a considerable basal tone which can be using a semi-open system without rebreathing. adjusted so that the coronary-venous oxygen Muscular relaxation was obtained by suxamethonium difference of the coronary blood remains in a dose of 2-4 mg/kg body weight, given as a single largely constant under greatly varying work injection immediately after induction, in order to facilitate tracheal intubation with a cuffed Magill loads on the heart. The adjustment of this tube. The animals were ventilated with a n Engstrom vascular tone is linked to the actual oxygen respirator with nitrous oxide and oxygen (1 :1 or 4: 1) demands (BERGLUNDet al. 1958) either by in a non-rebreathing system. The body temperature
422
0 . DOTTORI, E. HAGGENDAL, E. LINDER, G . NORDSTROM AND T. SEEMAN
was measured in the rectum and was kept at 37-38°C. Muscle relaxation was maintained by intermittent administration of suxamethonium in a n average dose of 4 mg/kg b.w./h. The ventilation was adjusted throughout the experiment to give alveolar normoventilation, and the animals were hyperinsufflated at intervals in order to prevent the development of atelectasis. Four intercostal nerves on either side of the proposed thoracotomy incision were blocked by local injection of 2 ml bupivacaine 0.5% with adrenaline before the thoracotomy. A left thoracotomy was carried out and the pericardium was incised and sutured to the thoracic wall in order to minimize changes in the position of the heart due to ventilatory movements. Short sections of the anterio-descending coronary artery or one of its branches were dissected. A catheter for intra-arterial administration of indicator and for intra-arterial pressure registration was sutured directly through the wall of this main coronary artery. A polyethylene catheter with a side opening, heated and drawn to fit a n unclosed atraumatic vascular needle, was used. Catheters were also inserted in the common carotid artery and, transmyocardially, into the left atrium and ventricle for continuous registration of vascular pressures during the whole experimental period. In some animals catheters were also inserted directly into the main pulmonary artery and into the coronary sinus via a jugular vein. All pressures were measured by means of a transducer of inductance type, and registered on a direct writer oscillograph (Mingograf-Elema). Mean blood pressure was obtained by electrical integration. For measurement of coronary blood flow, 0.1 ml ‘’krypton or 133xenon in saline was injected intracoronarily or directly at constant depth in the left et al. 1964, LINDER1966). anterior wall (JOHANSSON The clearance of the isotope was registered by an external scintillation detector, coupled to a ratemeter and recorded on a potentiometer writer. The coronary blood flow was calculated in ml/min x 100 g tissue from the monoexponential curve obtained by semilogarithmic plotting using the formula:
F = K X A X 100 where K is the rate constant of disappearance slope and A denotes the tissue-blood partition coefficient of the tracer divided by the specific weight of the tissue. The A-values used were 1.0 for krypton and 0.72 for xenon. The cardiac output was measured with a n electromagnetic flow meter (Nycotron, Oslo). T h e flow probe was chosen to fit the vessel snugly, and was placed on the aorta immediately above the aortic valves. Probe calibration was accomplished by perfusing the dog’s arteries with blood or 5 N saline solution. Zero flow for mean blood flow in vivo was obtained by shutting off the probe magnet instead
of inflow occlusion, as the latter was judged to interfere with the experimental condition. The mean flow in the ascending aorta was taken as a n indication of the cardiac output, and the stroke volumes were estimated from the peak flow curve. The diastolic “quiet interval”, when no flow except coronary flow occurred in the aorta, was used as a measure of zero for peak flow (SCHENK & DEDICHEN 1967). The peripheral vascular resistance was calculated as the quotient between mean arterial blood pressure in mmHg and cardiac output in l/min, thus ignoring the right atrial pressure. The coronary vascular resistance was calculated as the ratio between mean arterial blood pressure in mmHg and coronary blood flow in ml/min x 100 g tissue. Blood samples were taken at repeated intervals from the coronary artery and sinus for estimation of blood gas tensions. They were collected without exposure to air, and were stored at 0°C (ice water). The oxygen tensionin coronary arterial and venous blood was determined according to the method of GLEICHMANN & LUBBERS(1960). Carbon dioxide tension in blood was analysed according to the micromethod of SIGGAARD-ANDERSEN ( 1963). The correction factor for oxygen consumed during the interval between sampling and analysis and for temperature was & NU” (1966). I n calculated according to KELMAN the determination of oxygen content, the amount of physically dissolved oxygen was included in both arterial and sinus blood samples. The myocardial oxygen consumption was calculated per g tissue from determinations of arterio-sinus oxygen difference and coronary blood flow. The effect of nitrous oxide-oxygen anaesthesia was studied a t different oxygen-concentrations (1 : 1 and 4:I ) , and compared with the control periods where the animals were ventilated with nitrogen-oxygen at the same oxygen concentrations, and also with pure oxygen. In four dogs nitrous oxide-oxygen and nitrogenoxygen were compared in the proportion 1 : 1 ; and in another four dogs nitrous oxide-oxygen and nitrogenoxygen were compared in the proportion 4:1. In 18 dogs nitrous oxide-oxygen in the proportion 1 : 1 was compared with pure oxygen. Finally, in five dogs unsupplemented nitrous oxide-oxygen in the ratio 4:1, and nitrous oxide-oxygen-nitrogen in the ratio 2 :0.80: 1.20 were compared with nitrogen-oxygen supplemented by 5-1 0 mg diazepam a t the same oxygen concentrations (Table 1 ) . After any change in the ventilated gas mixture, no measurements were performed until a period of 7-10 min had passed, in order to obtain equilibrium. Each measurement period lasted 8-10 min. The values during nitrogen-oxygen or pure oxygen ventilation were used as control values, and those which were registered during nitrous oxide-oxygen ventilation are given in per cent of control values, each animal representing its own control.
42 3
NITROUS OXIDE ANAESTHESIA AND MYOCARDIAL BLOOD FLOW
Table 1 Plan of investigation.
Compared mixtures
Concentrations
N20-02 vs N202
80 %-20 %
NZO-02 vs N202-diazepam
80 %-20
%
NZO-02 vs N202 50 %-50
7;
N20-02 vs
0, 50 %-50
% 50 %-20 %-30 %
vs
100 %
No. of dogs
Pressures & flow Myocardial O L consumption
No. of Pressures & flow observations Myocardial O2 consumption
.
N20-02-N2 vs N,O,-diazepam
vs 80 %-20
4
5*
4
18
5*
3
5
3
5
5
6
5
4
21
5
5
5
3
5
5
%
* same experimental dogs Statistical analysis of all data was performed using the formula:
In order to compare the results obtained in the four dogs ventilated with nitrous oxide at 1 :1 with those ventilated with nitrous oxide in the ratio 2:0.80: 1.20, the flows were reported in ml/kg body weight; this was because of the large differences of weight in the two groups. For this comparison the t-test for independent means was also used for statjstical analysis. The changes after the hyperoxic nitrous oxide ventilation were thus used as control values. The level of significance chosen was P < 0.05.
RESULTS Figure 1 (A, upper part) shows the effects of nitrous oxide-nitrogen-oxygen mixture (2:0.80: 1.20) on coronary blood flow and cardiac output, as compared with the effects of nitrous oxide in oxygen (1 :1). I n Figure I (B), a group ventilated with a gas mixture containing 50% nitrous oxide during normoxia is compared with a group ventilated with air supplemented by diazepam.
Finally, in Figure 1 (C), comparison is made between a group ventilated with nitrous oxide in oxygen (1 :1) and a group ventilated with pure oxygen. I n all groups (Fig. 1 A, B, C), blood pressure and heart rate were unchanged. The only significant change was an increase in cardiac index (from 86.5 & 16.5 to 105.2147.9 ml/min/kg body weight) in the group which was ventilated with a normoxic nitrous oxide mixture, as compared with the group ventilated with a hyperoxic nitrous oxide (Fig. 1A). Among the haemodynamic factors reported in the middle part of the figure, there was a significant decrease (from 1.21 kO.44 to 1.1OkO.48 arb. units) in the coronary vascular resistance in the nitrous oxide group compared with pure oxygen (Fig. 1 C). The increase of the arterial-sinus oxygen difference in the normoxic nitrous oxide group was marked as compared with the nitrous oxide-oxygen (1 :1) group (Fig. 1 A). A similarly marked decrease in the oxygen difference was found when the nitrous oxide group was compared with the group ventilated with air supplemented by diazepam (Fig. 1 B). Furthermore, when the effects of
424
0.DOTTORI, E. HAGGENDAL, E . LINDER, G . NORDSTROM AND T. SEEMAN
B
A
A
C
8
120
100
80 60
0 P4.05
PNS
Fig. 1. Changes in circulatory parameters during ventilation with normoxic nitrous oxide gas mixtures, as compared with measurements during ventilation with air and pure oxygen. Definition of abbreviations: CO = cardiac output, H R = heart rate, MBF = myocardial blood flow, CVR = coronary vascular resistance, LVW = left ventricular work, BP, = arterial systolic blood pressure, A S o 2 diff = arteriosinus-oxygen difference, To,cor = myocardial oxygen consumption.
0
P
The Haemodynamic Effects of Nitrous Oxide Anaesthesia on Myocardial Blood Flow in Dogs 0. DOTTORI, E. HAGGENDAL, E. LINDER, G. NORDSTROM and T. SEEMAN Departments of Anaesthesiology and Surgery 11, University of Goteborg, Goteborg, Sweden The effects of ventilation with nitrous oxide in oxygen on myocardial blood flow and oxygen metabolism were investigated in 31 mongrel dogs. The results of this study showed that, compared with controls, hyperoxic nitrous oxide mixtures did not cause any great changes in myocardial haemodynamics, despite a decrease in cardiac output and an increase in systemic vascular resistance. Normoxic nitrous oxide mixtures produced an increase of the coronary blood flow due to decreased coronary vascular resistance. To what extent this coronary vasodilatation resulted from an increased myocardial metabolism or from a direct effect of nitrous oxide on the coronary vascular bed cannot be quantified from the present results.
Received I0 April, accepted f o r publication 15 December I975
Unsupplemented nitrous oxide was intro- release of vasodilatator metabolites (BERNE duced into clinical anaesthesia as the agent of 1964) or directly by decreased myocardial choice in “poor-risk’’ patients (HELLER et al. oxygen tension (GREGG& FISCHER 1963). 1956) and cardiac patients (GRAY& RIDING Moreover, previous studies of the coronary 1957, DOTTORI & EKESTROM 1964), mainly circulation have shown that the hyperoxia because it is presumed to have no deleterious results in a decreased coronary blood flow effects on the cardiovascular system. How- (SOBOL et al. 1962, LAMMERANT et al. 1969, ever, the effect of unsupplemented nitrous among others) and in a reduced myocardial oxide on the coronary blood flow is still contractility (DANIELL & BAGWELL 1968). unknown. The aim of the present investigation, A previous experimental study indicated therefore, was to study the effects of nitrous that nitrous oxide exerted different effects oxide on the coronary blood flow in normoxic on the systemic circulation depending on and hyperoxic gas-mixtures. the amount of oxygen in the gas-mixture. Hyperoxic nitrous oxide gas-mixtures seemed to produce vasoconstriction com- MATERIAL AND METHODS bined with decreased cardiac output, while The studies were carried out on a total of 31 mongrel normoxic nitrous oxide gas-mixtures had the dogs (16-39 kg) of either sex. The dogs were premedicated with diazepam (0.4-0.5 mg/kg body weight) opposite effect (DOTTORI et al. 1976). and atropine (0.01-0.02 mg/kg body weight). AnThe normal coronary vascular bed ex- aesthesia was induced by 80% nitrous oxide in oxygen hibits a considerable basal tone which can be using a semi-open system without rebreathing. adjusted so that the coronary-venous oxygen Muscular relaxation was obtained by suxamethonium difference of the coronary blood remains in a dose of 2-4 mg/kg body weight, given as a single largely constant under greatly varying work injection immediately after induction, in order to facilitate tracheal intubation with a cuffed Magill loads on the heart. The adjustment of this tube. The animals were ventilated with a n Engstrom vascular tone is linked to the actual oxygen respirator with nitrous oxide and oxygen (1 :1 or 4: 1) demands (BERGLUNDet al. 1958) either by in a non-rebreathing system. The body temperature
422
0 . DOTTORI, E. HAGGENDAL, E. LINDER, G . NORDSTROM AND T. SEEMAN
was measured in the rectum and was kept at 37-38°C. Muscle relaxation was maintained by intermittent administration of suxamethonium in a n average dose of 4 mg/kg b.w./h. The ventilation was adjusted throughout the experiment to give alveolar normoventilation, and the animals were hyperinsufflated at intervals in order to prevent the development of atelectasis. Four intercostal nerves on either side of the proposed thoracotomy incision were blocked by local injection of 2 ml bupivacaine 0.5% with adrenaline before the thoracotomy. A left thoracotomy was carried out and the pericardium was incised and sutured to the thoracic wall in order to minimize changes in the position of the heart due to ventilatory movements. Short sections of the anterio-descending coronary artery or one of its branches were dissected. A catheter for intra-arterial administration of indicator and for intra-arterial pressure registration was sutured directly through the wall of this main coronary artery. A polyethylene catheter with a side opening, heated and drawn to fit a n unclosed atraumatic vascular needle, was used. Catheters were also inserted in the common carotid artery and, transmyocardially, into the left atrium and ventricle for continuous registration of vascular pressures during the whole experimental period. In some animals catheters were also inserted directly into the main pulmonary artery and into the coronary sinus via a jugular vein. All pressures were measured by means of a transducer of inductance type, and registered on a direct writer oscillograph (Mingograf-Elema). Mean blood pressure was obtained by electrical integration. For measurement of coronary blood flow, 0.1 ml ‘’krypton or 133xenon in saline was injected intracoronarily or directly at constant depth in the left et al. 1964, LINDER1966). anterior wall (JOHANSSON The clearance of the isotope was registered by an external scintillation detector, coupled to a ratemeter and recorded on a potentiometer writer. The coronary blood flow was calculated in ml/min x 100 g tissue from the monoexponential curve obtained by semilogarithmic plotting using the formula:
F = K X A X 100 where K is the rate constant of disappearance slope and A denotes the tissue-blood partition coefficient of the tracer divided by the specific weight of the tissue. The A-values used were 1.0 for krypton and 0.72 for xenon. The cardiac output was measured with a n electromagnetic flow meter (Nycotron, Oslo). T h e flow probe was chosen to fit the vessel snugly, and was placed on the aorta immediately above the aortic valves. Probe calibration was accomplished by perfusing the dog’s arteries with blood or 5 N saline solution. Zero flow for mean blood flow in vivo was obtained by shutting off the probe magnet instead
of inflow occlusion, as the latter was judged to interfere with the experimental condition. The mean flow in the ascending aorta was taken as a n indication of the cardiac output, and the stroke volumes were estimated from the peak flow curve. The diastolic “quiet interval”, when no flow except coronary flow occurred in the aorta, was used as a measure of zero for peak flow (SCHENK & DEDICHEN 1967). The peripheral vascular resistance was calculated as the quotient between mean arterial blood pressure in mmHg and cardiac output in l/min, thus ignoring the right atrial pressure. The coronary vascular resistance was calculated as the ratio between mean arterial blood pressure in mmHg and coronary blood flow in ml/min x 100 g tissue. Blood samples were taken at repeated intervals from the coronary artery and sinus for estimation of blood gas tensions. They were collected without exposure to air, and were stored at 0°C (ice water). The oxygen tensionin coronary arterial and venous blood was determined according to the method of GLEICHMANN & LUBBERS(1960). Carbon dioxide tension in blood was analysed according to the micromethod of SIGGAARD-ANDERSEN ( 1963). The correction factor for oxygen consumed during the interval between sampling and analysis and for temperature was & NU” (1966). I n calculated according to KELMAN the determination of oxygen content, the amount of physically dissolved oxygen was included in both arterial and sinus blood samples. The myocardial oxygen consumption was calculated per g tissue from determinations of arterio-sinus oxygen difference and coronary blood flow. The effect of nitrous oxide-oxygen anaesthesia was studied a t different oxygen-concentrations (1 : 1 and 4:I ) , and compared with the control periods where the animals were ventilated with nitrogen-oxygen at the same oxygen concentrations, and also with pure oxygen. In four dogs nitrous oxide-oxygen and nitrogenoxygen were compared in the proportion 1 : 1 ; and in another four dogs nitrous oxide-oxygen and nitrogenoxygen were compared in the proportion 4:1. In 18 dogs nitrous oxide-oxygen in the proportion 1 : 1 was compared with pure oxygen. Finally, in five dogs unsupplemented nitrous oxide-oxygen in the ratio 4:1, and nitrous oxide-oxygen-nitrogen in the ratio 2 :0.80: 1.20 were compared with nitrogen-oxygen supplemented by 5-1 0 mg diazepam a t the same oxygen concentrations (Table 1 ) . After any change in the ventilated gas mixture, no measurements were performed until a period of 7-10 min had passed, in order to obtain equilibrium. Each measurement period lasted 8-10 min. The values during nitrogen-oxygen or pure oxygen ventilation were used as control values, and those which were registered during nitrous oxide-oxygen ventilation are given in per cent of control values, each animal representing its own control.
42 3
NITROUS OXIDE ANAESTHESIA AND MYOCARDIAL BLOOD FLOW
Table 1 Plan of investigation.
Compared mixtures
Concentrations
N20-02 vs N202
80 %-20 %
NZO-02 vs N202-diazepam
80 %-20
%
NZO-02 vs N202 50 %-50
7;
N20-02 vs
0, 50 %-50
% 50 %-20 %-30 %
vs
100 %
No. of dogs
Pressures & flow Myocardial O L consumption
No. of Pressures & flow observations Myocardial O2 consumption
.
N20-02-N2 vs N,O,-diazepam
vs 80 %-20
4
5*
4
18
5*
3
5
3
5
5
6
5
4
21
5
5
5
3
5
5
%
* same experimental dogs Statistical analysis of all data was performed using the formula:
In order to compare the results obtained in the four dogs ventilated with nitrous oxide at 1 :1 with those ventilated with nitrous oxide in the ratio 2:0.80: 1.20, the flows were reported in ml/kg body weight; this was because of the large differences of weight in the two groups. For this comparison the t-test for independent means was also used for statjstical analysis. The changes after the hyperoxic nitrous oxide ventilation were thus used as control values. The level of significance chosen was P < 0.05.
RESULTS Figure 1 (A, upper part) shows the effects of nitrous oxide-nitrogen-oxygen mixture (2:0.80: 1.20) on coronary blood flow and cardiac output, as compared with the effects of nitrous oxide in oxygen (1 :1). I n Figure I (B), a group ventilated with a gas mixture containing 50% nitrous oxide during normoxia is compared with a group ventilated with air supplemented by diazepam.
Finally, in Figure 1 (C), comparison is made between a group ventilated with nitrous oxide in oxygen (1 :1) and a group ventilated with pure oxygen. I n all groups (Fig. 1 A, B, C), blood pressure and heart rate were unchanged. The only significant change was an increase in cardiac index (from 86.5 & 16.5 to 105.2147.9 ml/min/kg body weight) in the group which was ventilated with a normoxic nitrous oxide mixture, as compared with the group ventilated with a hyperoxic nitrous oxide (Fig. 1A). Among the haemodynamic factors reported in the middle part of the figure, there was a significant decrease (from 1.21 kO.44 to 1.1OkO.48 arb. units) in the coronary vascular resistance in the nitrous oxide group compared with pure oxygen (Fig. 1 C). The increase of the arterial-sinus oxygen difference in the normoxic nitrous oxide group was marked as compared with the nitrous oxide-oxygen (1 :1) group (Fig. 1 A). A similarly marked decrease in the oxygen difference was found when the nitrous oxide group was compared with the group ventilated with air supplemented by diazepam (Fig. 1 B). Furthermore, when the effects of
424
0.DOTTORI, E. HAGGENDAL, E . LINDER, G . NORDSTROM AND T. SEEMAN
B
A
A
C
8
120
100
80 60
0 P4.05
PNS
Fig. 1. Changes in circulatory parameters during ventilation with normoxic nitrous oxide gas mixtures, as compared with measurements during ventilation with air and pure oxygen. Definition of abbreviations: CO = cardiac output, H R = heart rate, MBF = myocardial blood flow, CVR = coronary vascular resistance, LVW = left ventricular work, BP, = arterial systolic blood pressure, A S o 2 diff = arteriosinus-oxygen difference, To,cor = myocardial oxygen consumption.
0
P
