Clinical Science (1990)78, 591-596
591
Cardiac output and oxygen uptake in patients with renal failure C. H. KONG, F. D. THOMPSON
AND
F. J. MMS*
Institute of Urology and St Peter’s Hospital, London, and *SherringtonSchool of Physiology, UMDS (St Thomas’ Campus),London
(Received 28 September 1989/7 February 1990; accepted 21 February 1990)
SUMMARY 1. Cardiac index, oxygen uptake and haemoglobin concentration have been measured in eight patients with end-stage renal failure, and the values compared with those in eight healthy control subjects. Assuming the arterial blood to be 97% saturated, the oxygen saturation of mixed venous blood has been calculated. 2. Cardiac index was lower in the patients (2.45 20.42 litres min-’ m-’) compared with the control subjects (3.74f0.17 litres min-’ m-,). Oxygen uptake was reduced from 147 k 16 ml STP min-’ m-* in the control subjects to 112 k 9 ml STP min-’ m-, in the patients. 3. Oxygen saturation of mixed venous blood was 5 3 2 8 % in the patients compared with 79+2% in the control subjects, suggesting some degree of tissue hypoxia. If oxygen uptake were not reduced in renal failure, then the mixed venous blood oxygen saturation would be as low as 40%. 4. The oxygen saturation of blood sampled from the superior vena cava of patients with renal failure was 57 f 6%. 5. The data on oxygen uptake and venous oxygen saturation suggest that in the patients with end-stage renal failure, the low values of cardiac index, previously measured by impedance cardiography, are accurate.
Key words: cardiac output, oxy8en uptake, renal failure. Abbreviations: Po,, partial pressure of oxygen; PR, peripheral resistance; Q, cardiac output; R, respiratory exchange ratio; Sao,, arterial oxygen saturation; SVo,, mixed venous oxygen saturation; Vo2,oxygen uptake.
INTRODUCTION In anaemic patients, the cardiac output (Q) at rest is usually increased [1-31 to maintain tissue oxygenation. In
Correspondence: Dr C. H. Kong, Institute of Urology and’St Peter’s Hospital, Sheffeld Street, London WC2A 2EX.
patients with anaemia associated with renal failure,,some previous studies have described a raised [4-81, whereas others have reported a low Q [7, 9-11] in such patients. In a previous study in this unit, Q was low in about half of 35 patients with end-stage renal failure [12]. The combination of anaemia and a low Q would suggest a greater than normal arteriovenous oxygen difference and a low partial pressure of oxygen (Po,) at the tissues. However, if oxygen uptake (vo,) was also reduced, then tissue Po, would be nearer to normal. Studies in rats with renal failure suggest that vo2 at rest may be lowered [ 13,141. In this study, vo, and Q in a group of uraemic patients have been compared with those in a comparable group of normal healthy subjects. The calculated values of mixed venous oxygen saturation ( SVo2) have been compared with those directly measured in a second group of patients with subclavian lines inserted for treatment purposes. To confirm that arterial oxygen saturation (Sao,) was unaffected in patients with renal failure, it was measured in a third group of patients with arteriovenous fistulae. METHODS Subjects
Group 1 consisted of eight patients, two females and six males with an age range of 24-60 years (mean 43 years), in end-stage renal failure, who were attending St Philip’s Hospital. Their plasma creatinine and haemoglobin concentrations were 891 2 183 mol/l and 7.8 f 1.0 g/100 ml (meanksD), respectively. Five of them had an arteriovenous fistula and only two were undergoing haemodialysis. Five were receiving B-adrenoceptor blocking agents. None had respiratory diseases or other pathology unrelated to renal failure. The controls were eight normal healthy subjects, two females and six males, recruited from the hospital staff to match the patients in group 1 for sex and age. One male was taking a p-adrenoceptor-p-blocking agent and diuretic. Group 2 comprised ten patients (eight females and two males) of mean age 42 f 17 years ( fSD) with a subclavian
592
C. H. Kong et al.
catheter already in situ for haemodialysis. None had intracardiac shunts. Their mean haemoglobin concentration was 8.1k1.0 g/100 ml. The duration of dialysis was between 2 and 10 weeks. The position on the tips of the catheters was determined by radiography to be in the superior vena cava in nine patients and in the right atrium in one. For the measurement of blood gases, two 5 ml samples were withdrawn from the catheter into heparinized tubes, taking care to avoid air bubbles. The samples were analysed within 2 min for pH, Po, and oxygen saturation using an ABL 3/30 acid-base analyser (Radiometer, Copenhagen, Denmark). The 95% confidence limits of estimation for the routine laboratory estimations were + 0.01 units for pH and + 0.3 Wa for Po2.pH and Po, were used in the computation of mixed venous oxygen saturation. Group 3 consisted of nine patients (five females and four males) of mean age 47 f 15 years, with an arteriovenous fistula and mean haemoglobin concentration of 8.1f1.0 g/100 ml. Sao, was determined by sampling blood from the arterial line before starting dialysis. This study was approved by the Ethics Committee of St Peter’s Hospital.
Measurement of Vo, and Q After 10 min rest in the supine position, the patients (group 1) and the control subjects breathed through a valved mouthpiece into a Douglas bag. They were allowed 5 min to familiarize themselves with the equipment and then two 5 min collections of expired air were obtained. Q was measured before and after the collection by impedance cardiography (IFM/Minnesota Cardiograph model 304A). The application of this technique in renal failure patients has been described before [ 151 and the coefficient of variation by this method is 4.1%. Excellent correlations between impedance cardiography and more standard methods have been obtained in normal subjects [16, 171. In patients with renal failure, Handt et al. [18] demonstrated statistical identity between impedance cardiography and the thermodilution method before, during and at the end of haemodialysis after the removal of 2.1 litres of fluid. A simultaneous electrocardiograph (incorporated in the impedance cardiograph) provided the heart rate. Blood pressure was also measured using the mercury sphygmomanometer, and peripheral resistance (PR) was calculated as mean arterial blood pressure/Q where mean arterial blood pressure = diastolic pressure plus one-third of pulse pressure. The fractional concentration of oxygen in expired air was measured immediately after collection using a Servomex oxygen analyser and the volume was measured with a dry gas-meter (Scientific and Research Instruments) and corrected to STP dry. Assuming the respiratory exchange ratio ( R ) to be 0.80, vo, was calculated as follows. If =volume of expired air (litres/min), FEo2= fractional
vE
concentration of oxygen in dry gas and F E ~=fractional 2 concentration of nitrogen in dry gas, then:
where: &N,
=
1- FEo2+ ( R x FEo2) [(Rx21)/79]+1
The arterial oxygen content ( Cao2) was calculated assuming an oxygen capacity of 1.34 ml/g of haemoglobin and 97% saturation. The mixed venous oxygen content (00,) was calculated using Fick’s principle: VO,
Cvo, = Cao, - -
Q
and Svo, as: Go2 SVO, = -
Cao,
x 100
Statistical analysis The results were analysed using Student’s t-test. Differences were regarded as statistically significant if they achieved a significance level of 5%. RESULTS The blood pressure and PR in the patients in group 1 were significantly higher than in the normal subjects (Tables 1 and 2, P
591
Cardiac output and oxygen uptake in patients with renal failure C. H. KONG, F. D. THOMPSON
AND
F. J. MMS*
Institute of Urology and St Peter’s Hospital, London, and *SherringtonSchool of Physiology, UMDS (St Thomas’ Campus),London
(Received 28 September 1989/7 February 1990; accepted 21 February 1990)
SUMMARY 1. Cardiac index, oxygen uptake and haemoglobin concentration have been measured in eight patients with end-stage renal failure, and the values compared with those in eight healthy control subjects. Assuming the arterial blood to be 97% saturated, the oxygen saturation of mixed venous blood has been calculated. 2. Cardiac index was lower in the patients (2.45 20.42 litres min-’ m-’) compared with the control subjects (3.74f0.17 litres min-’ m-,). Oxygen uptake was reduced from 147 k 16 ml STP min-’ m-* in the control subjects to 112 k 9 ml STP min-’ m-, in the patients. 3. Oxygen saturation of mixed venous blood was 5 3 2 8 % in the patients compared with 79+2% in the control subjects, suggesting some degree of tissue hypoxia. If oxygen uptake were not reduced in renal failure, then the mixed venous blood oxygen saturation would be as low as 40%. 4. The oxygen saturation of blood sampled from the superior vena cava of patients with renal failure was 57 f 6%. 5. The data on oxygen uptake and venous oxygen saturation suggest that in the patients with end-stage renal failure, the low values of cardiac index, previously measured by impedance cardiography, are accurate.
Key words: cardiac output, oxy8en uptake, renal failure. Abbreviations: Po,, partial pressure of oxygen; PR, peripheral resistance; Q, cardiac output; R, respiratory exchange ratio; Sao,, arterial oxygen saturation; SVo,, mixed venous oxygen saturation; Vo2,oxygen uptake.
INTRODUCTION In anaemic patients, the cardiac output (Q) at rest is usually increased [1-31 to maintain tissue oxygenation. In
Correspondence: Dr C. H. Kong, Institute of Urology and’St Peter’s Hospital, Sheffeld Street, London WC2A 2EX.
patients with anaemia associated with renal failure,,some previous studies have described a raised [4-81, whereas others have reported a low Q [7, 9-11] in such patients. In a previous study in this unit, Q was low in about half of 35 patients with end-stage renal failure [12]. The combination of anaemia and a low Q would suggest a greater than normal arteriovenous oxygen difference and a low partial pressure of oxygen (Po,) at the tissues. However, if oxygen uptake (vo,) was also reduced, then tissue Po, would be nearer to normal. Studies in rats with renal failure suggest that vo2 at rest may be lowered [ 13,141. In this study, vo, and Q in a group of uraemic patients have been compared with those in a comparable group of normal healthy subjects. The calculated values of mixed venous oxygen saturation ( SVo2) have been compared with those directly measured in a second group of patients with subclavian lines inserted for treatment purposes. To confirm that arterial oxygen saturation (Sao,) was unaffected in patients with renal failure, it was measured in a third group of patients with arteriovenous fistulae. METHODS Subjects
Group 1 consisted of eight patients, two females and six males with an age range of 24-60 years (mean 43 years), in end-stage renal failure, who were attending St Philip’s Hospital. Their plasma creatinine and haemoglobin concentrations were 891 2 183 mol/l and 7.8 f 1.0 g/100 ml (meanksD), respectively. Five of them had an arteriovenous fistula and only two were undergoing haemodialysis. Five were receiving B-adrenoceptor blocking agents. None had respiratory diseases or other pathology unrelated to renal failure. The controls were eight normal healthy subjects, two females and six males, recruited from the hospital staff to match the patients in group 1 for sex and age. One male was taking a p-adrenoceptor-p-blocking agent and diuretic. Group 2 comprised ten patients (eight females and two males) of mean age 42 f 17 years ( fSD) with a subclavian
592
C. H. Kong et al.
catheter already in situ for haemodialysis. None had intracardiac shunts. Their mean haemoglobin concentration was 8.1k1.0 g/100 ml. The duration of dialysis was between 2 and 10 weeks. The position on the tips of the catheters was determined by radiography to be in the superior vena cava in nine patients and in the right atrium in one. For the measurement of blood gases, two 5 ml samples were withdrawn from the catheter into heparinized tubes, taking care to avoid air bubbles. The samples were analysed within 2 min for pH, Po, and oxygen saturation using an ABL 3/30 acid-base analyser (Radiometer, Copenhagen, Denmark). The 95% confidence limits of estimation for the routine laboratory estimations were + 0.01 units for pH and + 0.3 Wa for Po2.pH and Po, were used in the computation of mixed venous oxygen saturation. Group 3 consisted of nine patients (five females and four males) of mean age 47 f 15 years, with an arteriovenous fistula and mean haemoglobin concentration of 8.1f1.0 g/100 ml. Sao, was determined by sampling blood from the arterial line before starting dialysis. This study was approved by the Ethics Committee of St Peter’s Hospital.
Measurement of Vo, and Q After 10 min rest in the supine position, the patients (group 1) and the control subjects breathed through a valved mouthpiece into a Douglas bag. They were allowed 5 min to familiarize themselves with the equipment and then two 5 min collections of expired air were obtained. Q was measured before and after the collection by impedance cardiography (IFM/Minnesota Cardiograph model 304A). The application of this technique in renal failure patients has been described before [ 151 and the coefficient of variation by this method is 4.1%. Excellent correlations between impedance cardiography and more standard methods have been obtained in normal subjects [16, 171. In patients with renal failure, Handt et al. [18] demonstrated statistical identity between impedance cardiography and the thermodilution method before, during and at the end of haemodialysis after the removal of 2.1 litres of fluid. A simultaneous electrocardiograph (incorporated in the impedance cardiograph) provided the heart rate. Blood pressure was also measured using the mercury sphygmomanometer, and peripheral resistance (PR) was calculated as mean arterial blood pressure/Q where mean arterial blood pressure = diastolic pressure plus one-third of pulse pressure. The fractional concentration of oxygen in expired air was measured immediately after collection using a Servomex oxygen analyser and the volume was measured with a dry gas-meter (Scientific and Research Instruments) and corrected to STP dry. Assuming the respiratory exchange ratio ( R ) to be 0.80, vo, was calculated as follows. If =volume of expired air (litres/min), FEo2= fractional
vE
concentration of oxygen in dry gas and F E ~=fractional 2 concentration of nitrogen in dry gas, then:
where: &N,
=
1- FEo2+ ( R x FEo2) [(Rx21)/79]+1
The arterial oxygen content ( Cao2) was calculated assuming an oxygen capacity of 1.34 ml/g of haemoglobin and 97% saturation. The mixed venous oxygen content (00,) was calculated using Fick’s principle: VO,
Cvo, = Cao, - -
Q
and Svo, as: Go2 SVO, = -
Cao,
x 100
Statistical analysis The results were analysed using Student’s t-test. Differences were regarded as statistically significant if they achieved a significance level of 5%. RESULTS The blood pressure and PR in the patients in group 1 were significantly higher than in the normal subjects (Tables 1 and 2, P
