- Email: [email protected]
Oxygen consumption-oxygen delivery relationship in children
Oxygen consumption-oxygen delivery relationship in children M i c h a e l Seear, FRCPC, David Wensley, FRCPC, a n d A n d r e w M a c N a b , FRCPC From the Department of Intensive Care, British Columbia's Children's Hospital, Vancouver, British Columbia, Canada We e x a m i n e d the relationship b e t w e e n o x y g e n consumption (Vo2) and o x y g e n delivery (Do2) over a range of m e t a b o l i c d e m a n d in two groups of children. We studied 15 children after c a r d i a c surgery (plasma lactate levels <2.2 mmol/L, Vo2 <6 ml/min per kilogram, o x y g e n extraction ratio <25%); 8 were given transfusions with erythrocytes, 10 to 15 ml/kg, and 7 received a d r e n a l i n e infusions (0.05 to 0.3/~g/kg per minute). Blood transfusions significantly increased [:)o2 (20.5 • 6.4 to 26.2 • 7.1 ml/min per kilogram; p <0.05) but did not alter Vo2. Adrenaline increased Do2 (19.9 _+ 5.0 to 25.9 _+ 6.1 ml/min per kilogram; p <0.05) and Vo2 (4.3 _+ 0.8 to 5.5 • 1.2 ml/min per kilogram; p <0.05), but the o x y g e n excretion ratio and the mixed venous o x y g e n saturation were u n c h a n g e d . We also measured Vo2 and Doppler-derived 002 in 25 normal children during exercise. The relationship during exercise is given by the following equation: Vo2 index (in milliliters per minute per kilogram) = 0.88 • Do2 index - 6.95. Adrenaline infusions, but not b l o o d transfusions, increased Vo2 and Do2 together. This effect m a y be due to increased d e m a n d , a n a l o g o u s to exercise, and p r o b a b l y does not represent improved pertusion. We also found significant measurement error in Do2 and spontaneous variation in Vo2. We b e l i e v e that the c o n c e p t of supplyd e p e n d e n t ~{o2 is based on a number of m e t h o d o l o g i c and measurement errors. It should not be used to justify potentially dangerous therapies in sick children. (J PEDIATR1993;123:208-14)
An imbalance between oxygen delivery and oxygen consumption is frequently observed in critically ill children, particularly those with sepsis. 1 Transported oxygen is not used efficiently; although DO2 may be elevated, ~/02 and the oxygen extraction ratio are often inappropriately low. Lactic acidosis provides additional evidence of poor tissue oxygenation in many patients. Because V02 is correlated with survival,2 treatments designed to increase V02 might hold therapeutic value and have been widely investigated. Until recently, the conventional view has been that ~-02
Presented in part at the annual meeting of the American Thoracic Society, Miami Beach, Fla., May 1992. Submitted for publication Jan. 7, 1993; accepted April 1, 1993. Reprint requests: Michael Seear, FRCPC, Intensive Care Unit, British Columbia's Children's Hospital, 4480 Oak St., Vancouver, British Columbia V6H 3V4, Canada. Copyright 9 1993 by Mosby-Year Book, Inc. 0022-3476/93/$1.00 + .10 9/20/47540
208
is solely a reflection of oxygen demand and remains independent of DO2 over a wide range. 3 As Do2 falls, V02 remains constant so that tissue needs are met by increasing the OER. Once this compensation is exceeded (critical ACSA Cad2 Cvo2 Do2 MAFV OER Q Sad2 Svo2 "Vo2
Aortic cross-sectional area Arterial oxygen content Mixed venous oxygen content Oxygendelivery Mean aortic flow velocity Oxygenextraction ratio Cardiac output Arterial oxygen saturation Mixed venous oxygen saturation O x y g e nconsumption
DO2) , ~-O2 also falls and enters a flow-dependent phase 4
characterized by anaerobic metabolism and lactic acidosis. This view has been challenged by many reports that have shown supply dependence of V02, above the critical Do2
The Journal of Pediatrics Volume 123, Number 2
Seear, Wensley, and MacNab
209
point,5, 6 even in some patients with normal plasma lactate levels.7 The clinical implication of this work is that although a critically ill child might appear well resuscitated (normal plasma lactate level, blood pressure, and urine output), there may still be benefit from driving Do2 to supranormal levels in an attempt to increase ~r02 and consequently improve prognosis. 8 The concept of flow dependency and the use of high doses of inotropes, which is based on this hypothesis, have been criticized by authors who claim that the proposed relationship between V02 and Do2 is artifactual and due either to spontaneous variation in oxygen demand 9 or to methodologic errors] ~ The use of high-dose inotropes is not without risk, so we decided to examine the hypothesis closely before introducing this practice into our pediatric intensive care unit. Unfortunately, pediatric work in this area is limited and contradictory. Lucking et al. 11 noted flow dependence in children with sepsis, whereas Mink and Pollack 12 and Dietrich et al. 13 did not. We theorized that the ~r02 response to a blood transfusion (where oxygen demand might be assumed to be relatively constant) would differ from the V02 response to an infusion of adrenaline, which has been shown to have a profound effect on oxygen demand. 14 Consequently, we studied the relationship between V02 and Do2 after heart surgery in children in our pediatric intensive care unit. One group received blood transfusions and the other inotropes. We also believed that measurement error would be an important factor in an interpretation of results, and we attempted to quantify it for V02 and Do> We found no information in the pediatric literature to support our hypothesis that oxygen supply dependence is a normal physiologic response to exercise (or, at least, to high levels of circulating catecholamines). To obtain normal values for comparison, we studied the relationship between V02 and Do2 during exercise in healthy children.
Table I. Descriptive features of the treated groups
METHODS
where Fx02 is the inspired or expired oxygen fraction, Px02 is the partial pressure of arterial or mixed venous oxygen PB is barometric pressure. Minute ventilation was calculated from a timed expired gas collection. Gas volume was measured with a hot wire anemometer. Most children had cuffed endotracheal tubes in place. Gas leaks around uncurled tubes were easily controlled by repositioning the head or by gentle external pressure. Cardiac output was calculated by the Fick equation:
The study protocols were approved by our hospital clinical screening committee for research involving human subjects. Informed consent was obtained from parents. Cardiac surgery group. We studied 15 children (aged 2 months to 8 years; 9 girls, 6 boys) after cardiac bypass surgery that included insertion of an oximetric pulmonary artery catheter. The decision to insert the pulmonary arterial line was a clinical judgment and was not part of the study. Strict criteria were used to select patients who were acceptably resuscitated (arterial lactate level <2.2 mmol/L, temperature >36.5 ~ C, pH >7.35, base excess < - 5 mmol/L) but whose cardiopulmonary values still fell to less than the prognostic levels defined by Pollack et al. 2 We chose a "Vo2 index <6 ml/min per kilogram and an OER <25% as val-
Treatmenf group Variable
Transfusion
Inotrope
No. 8 (5 F) 7 (4 F) Age range (mo) 7-96 2-55 Treatment* Packed erythrocytes Adrenaline 11.2 +_ 2.1 ml/kg 0.09 _+ 0.03 ~g/kg/min Deaths (No.) 0 1 *Values are mean _+ SD,
ues for selection based on median values for survivors of pediatric sepsis. 2 All the children were paralyzed and sedated and had undergone operation more than 12 hours previously. No alterations of ventilator settings or support drugs were made during the study. Eight children who were given a blood transfusion and seven who were given extra inotropes were studied further. Decisions to start inotrope therapy or to give blood were made on individual clinical grounds and were not part of the study. The transfusion group received packed erythrocytes, 10 to 15 ml/kg, for 2 hours. The inotrope group received an infusion of adrenaline, 0.05 to 0.30 ~g/kg per minute. Oxygen consumption was calculated from the following equation: r~O2 = gmin (FIO2 - FEO2)
(1)
where ~/min is minute ventilation, FI02 is inspired oxygen fraction, and FE02 is expired oxygen fraction. Partial pressures of oxygen in saturated inspired and expired gas samples were measured by oxygen electrode (model 178 analyzer, Corning, Medfield, Mass.) and converted to fractional concentration by using the following equation: FxO2
PxO2
(PB - 47)
(2)
Wo2
Q - (Cao2 -- Cv02)
(3)
where Ca02 and Cv02 are arterial and mixed venous oxygen contents respectively, calculated as follows: Cxo2 = I X 34 Hgb X SxO2+ 0.003 X Pxo2
(4)
2 10
Seear, Wensley, and MacNab
The Journal of Pediatrics August 1993
10
t"
E v
E
4
O4
.o>
2-
0
0
I
I
I
I
10
20
50
40
50
20
I 30
I 40
50
10-
.
6t.--
E E
4-
,P
04
o o>.
2
0 0
t 10 DO21
(ml/min/kg)
Fig. 4. Individual plot of ~/O2index versus Do2 index, before and after treatments designed to increase Do2 index. Top,
Eight children who received packed erythrocytes, 10 to 15 ml/kg; bottom, seven children who received an infusion of adrenaline, 0.05 to 0.30 ~g/kg per minute. T/off, Voz index; Doff, Do2 index.
where Hgb is hemoglobin level, SxO2 is arterial or mixed venous oxygen saturation, and PxO2 is the partial pressure of arterial or mixed venous oxygen. The Sao2 was measured by pulse oximeter and Svo2 by cooximeter. Values for Vo~ and Do2 were indexed to body weight in kilograms ('Vo2 and Do2 indexes). Exercise group. We measured "~o2 and Do2 in 26 children (aged 6 to 19 years; 14 girls, 12 boys) during a modified exercise protocol. An additional five children had technically inadequate measurements and were excluded from the study. The children were exercised on a supine cycle ergometer with electrical braking. Two to four levels of submaximal exercise were used according to the child's ability (range, 0 to 175 watts). A 2-minute equilibration period was allowed after changes in work load.
Cardiac output was measured noninvasively by using Doppler echocardiography. This technique has been validated for use during exercise in chiidren 15and adults.16 The method is described in detail elsewhere] 7 Briefly, Q can be calculated from the following equation: Q = ACSA x MAFV
(5)
where MAFV is mean aortic flow velocity and ACSA is aortic cross-sectional area calculated as follows: ACSA-
7rd2 4
(6)
where d is aortic diameter. The ACSA was calculated at rest by measuring aortic diameter with two-dimensional echocardiography. Mean velocity was then measured at each
The Journal of Pediatrics Volume 123, Number 2
Seear, Wensley, and MacNab
2 11
50
40
r-
30
O
E
E
%
20
o
0 ,.,>.
10-
0 0
I
I
I
I
1
10
20
30
40
50
D02I
(ml/min/kg)
60
Fig. 2. Regression plot of V02 index and Do2 index measured at various levels of submaximal exercise, with pooled data from 26 children. (/off, ~/02 index; Doff, Do2 index.
Table
II. Physiologic variables before and after t r e a t m e n t s Blood transfusion group Variable
Pretreatment
Vo2I (ml/min/kg) Do2I (ml/min/kg) OER (%) Heart rate (beats/rain) Temperature (~ Svo2 (%)
CI ( L / m i n / m 2) Hgb (gm/dl)
4.5 20.5 21.9 152 37.6 65 3.4 8.4
Posttreatment
_+ 0.8 _+ 6.4 _+ 3.3 _+ 18 _+ 1.9 _+ 3.2 + 0.9 _+ 1.4
4.4 26.2 18.8 156 37.8 67 3.8 9.9
_+ 0.8 _+ 7.1" _+ 3.9 _+ 18 + 1.8 _+ 4.4 _+ 1.3 + 1.7
Inotrope group Pretreatment
4.3 19.9 21.8 165 37.3 59 2.9 9.2
_+ 0.8 _+ 5.0 _+ 2.1 _+ 21 _+ 2.0 _+ 5.6 _+ 1.2 _+ 1.5
Posttreatment
5.5 25.9 21.9 177 37.8 61 3.6 9.2
_+ 1.2" _+ 6.1" + 3.0 _+ 19 _+ 1.8 _+ 5.1 _+ 1.4 _+ 1.5
All values are mean _+ SD. CI, Cardiac index; Doff, DO2 index; Hgb, hemoglobin; f/off, VO2 index. *Significant difference between pretreatment and posttreatment values (paired t test, p = 0.05).
level of work with a Doppler flow probe from a suprasternal position. Oxygen consumption was m e a s u r e d with a standard, commercially available exercise testing m a c h i n e ( J a e g e r Q-Plex; Quinton I n s t r u m e n t Co., Seattle, Wash.). Oxygen delivery was calculated from the following equation: Do2 = Q • Ca02
(7)
The Ca02 was calculated as follows: Cao2 = 1.34 x Hemoglobin • Sao2
(8)
No correction was m a d e for dissolved oxygen. W e believed t h a t the small error incurred was not worth the difficulties
involved in sampling arterial blood. Hemoglobin was measured from a capillary blood sample, and Sao2 was measured by continuous pulse oximetry. Error estimation and statistical methods. W e a t t e m p t e d to assess m e a s u r e m e n t error in "r and Do2 for the cardiac surgery group. T h e V02 was m e a s u r e d in a d u m m y ventilator circuit while varying oxygen flows were bled into the system by using an a c c u r a t e r o t a m e t e r (series 150 flowmeter; U n i o n Carbide, Somerset, N.J.). M e a s u r e d and actual oxygen flows were c o m p a r e d by using the method of Bland and Altman. 18 N o absolute s t a n d a r d for Q exists; we could measure only the individual error for each component of the Fick equation (V02, partial pressures of arterial and mixed
2 12
Seear, Wensley, and MacNab
Fig. 3. Schematic representation of relationship between Vo2 and Do2 in children at different levels of oxygen demand. Shaded area represents relationship during exercise(i.e., varying demand). Circle represents normal resting values. Solid lines represent the relationship at two levels of constant demand. True supply dependence occurs when a poorly perfused patient is resuscitated (a-b). Blood transfusion at point b produces no further increase in Vo2 index (b-c) because demand is constant. Adrenaline infusion at point b raises the Vo2 index plateau (exercise effect), giving a false impressionof supply dependence (b-e). iZo2I,~'o2 index; Do2I, Do2 index.
venous oxygen, hemoglobin level, Sao2, Svo2). For each variable, 5 to 10 measurements of a single sample were made and a coefficient of variation calculated: (SD/ mean) X 100. In the same way, we assessed the reproducibility of Doppler-derived Q during exercise by calculating a coefficient of variation for repeated measures of MAFV and aortic diameter. Spontaneous variability in 37o2 was assessed by making regular measurements of 3702 for at least 30 minutes before the treatment period began. The percentage increase between highest and lowest values was used as an indicator of variability, after the method of Weissman et al. 19 All data are expressed as mean +__SD. The regression line for pooled exercise data was calculated by the method of least squares. Comparisons between the 37o2 index and the Do2 index values before and after a treatment were made by paired t test; significance level was 5%. The error in measuring 3702 was expressed by the method of Bland and Altman, 18 where the difference between two simultaneous measurements is expressed as a percentage of their mean value. RESULTS Cardiac surgery groups. The descriptive features of the two treatment subgroups are given in Table I. The pre-
The Journal of Pediatrics August 1993
treatment and posttreatment hemodynamic measurements are shown in Table II. Blood transfusions increased the Do2 index significantly but did not alter any other measurements, including the V02 index (Fig. 1, top). Infusions of adrenaline increased both the 37o2 index and the Do2 index significantly (Fig. 1, bottom), but there were no significant increases in any other measurements, including Sv02 and OER (Table II). Exercise group. Successful measurements of Q were made in 26 exercising children by using Doppler echocardiography. The technique was dependent on the skills of the ultrasonographer; even under ideal conditions, we could not obtain adequate measurements of aortic velocity in five other children who were excluded from the study. The relationship between the 37o2 index and the Do2 index (with 37o2 and Do2 measured in milliliters per minute per kilogram of body weight) during submaximal exercise is shown in Fig. 2. It is described by the following equation: V02 index = Do2 index x 0,88 - 6.95
(9)
We could not find published values for the relationship between the 37o2 index and Do2 index during exercise in children. However, our results for Q (in liters per minute) and 37o2 measured during exercise compare well with the findings reported in the available literature. 2~The relationship is described by the following equation: Q = 5.5 x Vo2 + 2.3
(10)
Measurement errors. The average difference between measured and actual values of 3702 with the use of our technique was 4.7%. The coefficients of variation for repeated measures of the individual Fick components were as follows: Sao2, 2.2%; Svo2, unknown; partial pressures of arterial and mixed venous oxygen, 3.9%; hemoglobin level, 5.1%; and 37o2, 5.3%. Short-term 3702 was surprisingly variable (mean 12.3%, range 6.2% to 19.6%) even though care was taken not to disturb the patient. The coefficients of variation for additional measures of the Doppler components were as follows: MAFV, 9.3% and aortic diameter, 4.9%. DISCUSSION Although our study of the exercise group expands the information available from a standard pediatric exercise test, our main interest was in obtaining normal pediatric values for 37o2 and Do2 over a wide range of oxygen demand. Direct comparisons between the surgical and exercise groups should be made cautiously. During exercise, Do2 is dictated by oxygen demand; in the postsurgical group the reverse is true. In addition, measurement techniques varied between the groups. There was a striking difference between the two groups
The Journal of Pediatrics Volume 123, Number 2
in their ability to utilize transported oxygen. After surgery, patients had ~'o2 indexes barely higher than those of normal resting children, even though several patients could achieve Do2 indexes equal to moderate exercise. The primary problem appeared to be poor oxygen uptake, not flow limitation. The mean OER was less than 25%, considerably less than the 65% to 70% achieved by healthy exercising children. The newer theory of Vo2 supply dependence is compatible with the classic view when applied to patients with lactic acidosis and a large oxygen debt. If Do2 is at less than a critical level, the clinical manifestations are poor perfusion, low urine output, and elevated blood lactate concentration. The patient requires resuscitation, so any increase in Do2 will increase the volume of tissue participating in gas exchange; consequently, Vo2 will rise (Fig. 3). Both Haupt et al. 21 and Gilbert et al. 22 have shown in adults with sepsis that supply dependence after blood transfusions is found only in those patients who had lactic acidosis, although dependence has been described in patients with normal lactate levels. 23 The two theories differ once a patient is considered well resuscitated. This poorly defined condition usually rests on the attainment of predefined physiologic variables, including a normal blood lactate concentration,2 even though clinical assessment is relatively crude and measurements of pH and lactate levels vary widely for many reasons. In stable children with normal lactate levels, we found no significant rise in Vo2 when Do2 was increased by erythrocyte transfusion. Conversely, infusions of adrenaline appeared to uncover oxygen supply dependence. In the absence of a measure of cellular metabolism (such as cytochrome oxidase saturation),z4 it is difficult to decide whether this represents any advantage to the child. Because there was no change in OER (which correlates with microvascular surface area 25) or Svo2, we thought that the rise was not due to improved perfusion but was more likely the result of increased oxygen demand, analogous to exercise. The stimulation of Vo2 by catecholamine infusion is well described in human beings. 14 Villar et al. 9 demonstrated apparent supply dependence in adults even when no therapeutic interventionswere made and thought that it was due to spontaneous variations in ~ro2 and oxygen demand. This point is of more than academic interest because it is possible that the use of high doses of catecholamines is harmful in critically ill children with normal blood lactate values. The combination of local vasoconstriction and increased metabolic demand might lead to areas of unnecessary anaerobic metabolism. Fellows et al. 14 found that infusions of adrenaline, 0.05/zg/kg per minute, caused a 24% increase in metabolic rate with a significant reduction
Seear, Wensley, and MacNab
2 13
in hand blood flow; blood lactate values increased significantly in healthy volunteers. Wilson et al. 26 reported a similar finding when using adrenaline infusions in adults with septic shock; blood lactate levels rose significantly despite an increase in Do2 of 25%. From our results, we believe that in stable children after cardiac surgery with normal blood lactate values, V02 is independent of increases in Do2 (Fig. 3). Any therapy that increases demand (such as high doses of inotropes) simply raises the overall metabolism of perfused cells. Because demand cannot be quantified, the increase in "r can be misinterpreted as implying supply dependency when, in fact, the change represents normal exercise physiology and gives no perfusion advantage to the child. The use of the V02 and Do2 relationship as an indicator of tissue perfusion has other weaknesses, apart from the nonquantifiableeffect of varying oxygen demand. The slope of the relationship during supply dependence is commonly reported in the range of 10% to 15%.27 Unfortunately, measurement error is at least as big as this proposed slope. We found that spontaneous variability in ~/o 2 averaged 12.3% with a range up to 19.6%; although we could assess error only of the individual components of Do2, its total error was probably at least this large. Even if measurements could be made accurately, the clinical value would still be limited because whole body values for "~o2 and Do2 can never provide information about perfusion of individual organs such as kidney and gut. In summary, we could not support the concept of oxygen supply dependence or the clinical use of ~'o2 and Do2 measurements on which it is based. The measurement errors and methodologic problems that we report are not specific to children after cardiac surgery and may apply to other critically ill patients. REFERENCES
1. Pollack MM, Fields AI, Ruttimann UE. Sequential cardiopulmonary variables of infants and children in septic shock. Crit Care Med 1984;12:554-9. 2. Pollack MM, Fields AI, Ruttimann UE. Distribution of eardiopulmonaryvariables in pediatric survivorsand nonsurvivors of septic shock. Crit Care Med 1985;13:454-9. 3. Shepard AP, Granger AJ, Smith EE, Guyton AC. Local control of tissue oxygen delivery and its contribution to the regulation of cardiac output. Am J Physiol 1973;225:747-55. 4. Cain SM. Peripheral oxygen uptake and deliveryin health and disease. Clin Chest Med 1983;4:139-48. 5. Kaufman BS, Rackow EC, Falk JL. The relationship between oxygen delivery and consumption during fluid resuscitation of hypovolemic and septic shock. Chest 1984;85:336-40. 6. Shoemaker WC, Appel PL, Kram HB. Hemodynamicand oxygen transport effects of dobutamine in critically ill general surgical patients. Crit Care Med 1986;14:1032-6. 7. Mohsenifar Z~ Goldbach P, Tashkin DP, Campisi DJ. Rela-
2 14
8.
9.
10.
11.
12.
13.
14.
15.
16.
Seear, Wensley, and MacNab
tionship between 02 delivery and 0 2 consumption in the adult respiratory distress syndrome. Chest 1983;84:267-71. Clarke C, Denis Edwards J, Nightingale P, Mortimer A J, Morris J. Persistence of supply dependency of oxygen uptake at high levels of delivery in adult respiratory distress syndrome. Crit Care Med 1991;19:497-502. Villar J, Slutsky AS, Hew E, Aberman A. Oxygen transport and oxygen consumption in critically ill patients. Chest 1990; 98:687-92. Ronco J J, Phang PT, Walley KR, Wiggs B, Fenwick JC, Russell JA. Oxygen consumption is independent of changes in oxygen delivery in severe adult respiratory distress syndrome. Am Rev Respir Dis 1991;143:1267-73. Lucking SE, Williams TM, Chaten FC, Metz RI, Mickell JJ. Dependence of oxygen consumption on oxygen delivery in children with hyperdynamic septic shock and low oxygen extraction. Crit Care Med 1990;18:1316-9. Mink RB, Pollack MM. Effect of blood transfusion on oxygen consumption in pediatric septic shock. Crit Care Med 1990; 18:1087-91. Dietrich KA, Conrad SA, Herbert CA, Levy GF, Romero MD. Cardiovascular and metabolic response to red blood cell transfusion in critically ill volume-resuscitated nonsurgical patients. Crit Care Med 1990;18:940-4. Fellows IW, Bennet T, MacDonald IA. The effect of adrenaline upon the cardiovascular and metabolic functions in man. Clin Sci 1985;69:215-22. Marx GR, Hicks RW, Allen HD, Kinger SM. Measurement of cardiac output and exercise factor by pulsed Doppler echocardiography during supine bicycle ergometry in normal young adolescent boys. J Am Coll Cardiol 1987;10:430-4. Christie J, Sheldahl CM, Tristani FE, Sagar KB, Ptacin M J, Wann S. Determination of stroke volume and cardiac output during exercise: comparison of two-dimensional and Doppler echocardiography, Fick oximetry, and thermodilution. Circulation 1987;76:539-47.
The Journal of Pediatrics August 1993
17. Huntsman LL, Stewart DK, Barnes SR, Franklin SB, Colocousis JS, Hessef EA. Non-invasive Doppler determination of cardiac output in man: clinical validation. Circulation 1983; 67:593-7. 18. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. 19. Weissman C, Kemper M. Damask MC, Askanazi J, Hyman AL, Kinney JM. Effect of routine intensive care interactions on metabolic rate. Chest 1984;86:815-8. 20. Godfrey S. Exercise testing in children. Philadelphia: WB Saunders, 1974:97-101. 21. Haupt MT, Gilbert EM, Carlson RW. Fluid loading increases oxygen consumption in septic patients with lactic acidosis. Am Rev Respir Dis 1985;131:912-6. 22. Gilbert EM, Haupt MT, Mandanas RY, Huaringa A J, Carlson RW. The effect of fluid loading, blood transfusion, and catechotamine infusion on oxygen delivery and consumption in patients with sepsis. Am Rev Respir Dis 1986;134:873-8. 23. Weg J. Oxygen transport in adult respiratory distress syndrome and other acute circulatory problems: relationship of oxygen delivery and oxygen consumption. Crit Care Med 1991;19:650-7. 24. Brazy J. Near-infrared spectroscopy. Clin Perinatol 1991; 18:519-34. 25. Beer G, Yonce LR. Blood flow, oxygen uptake, and capillary filtration in resting skeletal muscle. Am J Physiol 1972; 223:492-8. 26. Wilson W, Lipman J, Scribante J, et al. Septic shock: does adrenaline have a role as a first line inotropic agent? Anaesth Intensive Care 1992;20:470-4. 27. Dantzker DR, Foresman B, Gutierrez G. Oxygen supply and utilization relationships: a re-evaluation. Am Rev Respir Dis 199l;143:675-9.
An imbalance between oxygen delivery and oxygen consumption is frequently observed in critically ill children, particularly those with sepsis. 1 Transported oxygen is not used efficiently; although DO2 may be elevated, ~/02 and the oxygen extraction ratio are often inappropriately low. Lactic acidosis provides additional evidence of poor tissue oxygenation in many patients. Because V02 is correlated with survival,2 treatments designed to increase V02 might hold therapeutic value and have been widely investigated. Until recently, the conventional view has been that ~-02
Presented in part at the annual meeting of the American Thoracic Society, Miami Beach, Fla., May 1992. Submitted for publication Jan. 7, 1993; accepted April 1, 1993. Reprint requests: Michael Seear, FRCPC, Intensive Care Unit, British Columbia's Children's Hospital, 4480 Oak St., Vancouver, British Columbia V6H 3V4, Canada. Copyright 9 1993 by Mosby-Year Book, Inc. 0022-3476/93/$1.00 + .10 9/20/47540
208
is solely a reflection of oxygen demand and remains independent of DO2 over a wide range. 3 As Do2 falls, V02 remains constant so that tissue needs are met by increasing the OER. Once this compensation is exceeded (critical ACSA Cad2 Cvo2 Do2 MAFV OER Q Sad2 Svo2 "Vo2
Aortic cross-sectional area Arterial oxygen content Mixed venous oxygen content Oxygendelivery Mean aortic flow velocity Oxygenextraction ratio Cardiac output Arterial oxygen saturation Mixed venous oxygen saturation O x y g e nconsumption
DO2) , ~-O2 also falls and enters a flow-dependent phase 4
characterized by anaerobic metabolism and lactic acidosis. This view has been challenged by many reports that have shown supply dependence of V02, above the critical Do2
The Journal of Pediatrics Volume 123, Number 2
Seear, Wensley, and MacNab
209
point,5, 6 even in some patients with normal plasma lactate levels.7 The clinical implication of this work is that although a critically ill child might appear well resuscitated (normal plasma lactate level, blood pressure, and urine output), there may still be benefit from driving Do2 to supranormal levels in an attempt to increase ~r02 and consequently improve prognosis. 8 The concept of flow dependency and the use of high doses of inotropes, which is based on this hypothesis, have been criticized by authors who claim that the proposed relationship between V02 and Do2 is artifactual and due either to spontaneous variation in oxygen demand 9 or to methodologic errors] ~ The use of high-dose inotropes is not without risk, so we decided to examine the hypothesis closely before introducing this practice into our pediatric intensive care unit. Unfortunately, pediatric work in this area is limited and contradictory. Lucking et al. 11 noted flow dependence in children with sepsis, whereas Mink and Pollack 12 and Dietrich et al. 13 did not. We theorized that the ~r02 response to a blood transfusion (where oxygen demand might be assumed to be relatively constant) would differ from the V02 response to an infusion of adrenaline, which has been shown to have a profound effect on oxygen demand. 14 Consequently, we studied the relationship between V02 and Do2 after heart surgery in children in our pediatric intensive care unit. One group received blood transfusions and the other inotropes. We also believed that measurement error would be an important factor in an interpretation of results, and we attempted to quantify it for V02 and Do> We found no information in the pediatric literature to support our hypothesis that oxygen supply dependence is a normal physiologic response to exercise (or, at least, to high levels of circulating catecholamines). To obtain normal values for comparison, we studied the relationship between V02 and Do2 during exercise in healthy children.
Table I. Descriptive features of the treated groups
METHODS
where Fx02 is the inspired or expired oxygen fraction, Px02 is the partial pressure of arterial or mixed venous oxygen PB is barometric pressure. Minute ventilation was calculated from a timed expired gas collection. Gas volume was measured with a hot wire anemometer. Most children had cuffed endotracheal tubes in place. Gas leaks around uncurled tubes were easily controlled by repositioning the head or by gentle external pressure. Cardiac output was calculated by the Fick equation:
The study protocols were approved by our hospital clinical screening committee for research involving human subjects. Informed consent was obtained from parents. Cardiac surgery group. We studied 15 children (aged 2 months to 8 years; 9 girls, 6 boys) after cardiac bypass surgery that included insertion of an oximetric pulmonary artery catheter. The decision to insert the pulmonary arterial line was a clinical judgment and was not part of the study. Strict criteria were used to select patients who were acceptably resuscitated (arterial lactate level <2.2 mmol/L, temperature >36.5 ~ C, pH >7.35, base excess < - 5 mmol/L) but whose cardiopulmonary values still fell to less than the prognostic levels defined by Pollack et al. 2 We chose a "Vo2 index <6 ml/min per kilogram and an OER <25% as val-
Treatmenf group Variable
Transfusion
Inotrope
No. 8 (5 F) 7 (4 F) Age range (mo) 7-96 2-55 Treatment* Packed erythrocytes Adrenaline 11.2 +_ 2.1 ml/kg 0.09 _+ 0.03 ~g/kg/min Deaths (No.) 0 1 *Values are mean _+ SD,
ues for selection based on median values for survivors of pediatric sepsis. 2 All the children were paralyzed and sedated and had undergone operation more than 12 hours previously. No alterations of ventilator settings or support drugs were made during the study. Eight children who were given a blood transfusion and seven who were given extra inotropes were studied further. Decisions to start inotrope therapy or to give blood were made on individual clinical grounds and were not part of the study. The transfusion group received packed erythrocytes, 10 to 15 ml/kg, for 2 hours. The inotrope group received an infusion of adrenaline, 0.05 to 0.30 ~g/kg per minute. Oxygen consumption was calculated from the following equation: r~O2 = gmin (FIO2 - FEO2)
(1)
where ~/min is minute ventilation, FI02 is inspired oxygen fraction, and FE02 is expired oxygen fraction. Partial pressures of oxygen in saturated inspired and expired gas samples were measured by oxygen electrode (model 178 analyzer, Corning, Medfield, Mass.) and converted to fractional concentration by using the following equation: FxO2
PxO2
(PB - 47)
(2)
Wo2
Q - (Cao2 -- Cv02)
(3)
where Ca02 and Cv02 are arterial and mixed venous oxygen contents respectively, calculated as follows: Cxo2 = I X 34 Hgb X SxO2+ 0.003 X Pxo2
(4)
2 10
Seear, Wensley, and MacNab
The Journal of Pediatrics August 1993
10
t"
E v
E
4
O4
.o>
2-
0
0
I
I
I
I
10
20
50
40
50
20
I 30
I 40
50
10-
.
6t.--
E E
4-
,P
04
o o>.
2
0 0
t 10 DO21
(ml/min/kg)
Fig. 4. Individual plot of ~/O2index versus Do2 index, before and after treatments designed to increase Do2 index. Top,
Eight children who received packed erythrocytes, 10 to 15 ml/kg; bottom, seven children who received an infusion of adrenaline, 0.05 to 0.30 ~g/kg per minute. T/off, Voz index; Doff, Do2 index.
where Hgb is hemoglobin level, SxO2 is arterial or mixed venous oxygen saturation, and PxO2 is the partial pressure of arterial or mixed venous oxygen. The Sao2 was measured by pulse oximeter and Svo2 by cooximeter. Values for Vo~ and Do2 were indexed to body weight in kilograms ('Vo2 and Do2 indexes). Exercise group. We measured "~o2 and Do2 in 26 children (aged 6 to 19 years; 14 girls, 12 boys) during a modified exercise protocol. An additional five children had technically inadequate measurements and were excluded from the study. The children were exercised on a supine cycle ergometer with electrical braking. Two to four levels of submaximal exercise were used according to the child's ability (range, 0 to 175 watts). A 2-minute equilibration period was allowed after changes in work load.
Cardiac output was measured noninvasively by using Doppler echocardiography. This technique has been validated for use during exercise in chiidren 15and adults.16 The method is described in detail elsewhere] 7 Briefly, Q can be calculated from the following equation: Q = ACSA x MAFV
(5)
where MAFV is mean aortic flow velocity and ACSA is aortic cross-sectional area calculated as follows: ACSA-
7rd2 4
(6)
where d is aortic diameter. The ACSA was calculated at rest by measuring aortic diameter with two-dimensional echocardiography. Mean velocity was then measured at each
The Journal of Pediatrics Volume 123, Number 2
Seear, Wensley, and MacNab
2 11
50
40
r-
30
O
E
E
%
20
o
0 ,.,>.
10-
0 0
I
I
I
I
1
10
20
30
40
50
D02I
(ml/min/kg)
60
Fig. 2. Regression plot of V02 index and Do2 index measured at various levels of submaximal exercise, with pooled data from 26 children. (/off, ~/02 index; Doff, Do2 index.
Table
II. Physiologic variables before and after t r e a t m e n t s Blood transfusion group Variable
Pretreatment
Vo2I (ml/min/kg) Do2I (ml/min/kg) OER (%) Heart rate (beats/rain) Temperature (~ Svo2 (%)
CI ( L / m i n / m 2) Hgb (gm/dl)
4.5 20.5 21.9 152 37.6 65 3.4 8.4
Posttreatment
_+ 0.8 _+ 6.4 _+ 3.3 _+ 18 _+ 1.9 _+ 3.2 + 0.9 _+ 1.4
4.4 26.2 18.8 156 37.8 67 3.8 9.9
_+ 0.8 _+ 7.1" _+ 3.9 _+ 18 + 1.8 _+ 4.4 _+ 1.3 + 1.7
Inotrope group Pretreatment
4.3 19.9 21.8 165 37.3 59 2.9 9.2
_+ 0.8 _+ 5.0 _+ 2.1 _+ 21 _+ 2.0 _+ 5.6 _+ 1.2 _+ 1.5
Posttreatment
5.5 25.9 21.9 177 37.8 61 3.6 9.2
_+ 1.2" _+ 6.1" + 3.0 _+ 19 _+ 1.8 _+ 5.1 _+ 1.4 _+ 1.5
All values are mean _+ SD. CI, Cardiac index; Doff, DO2 index; Hgb, hemoglobin; f/off, VO2 index. *Significant difference between pretreatment and posttreatment values (paired t test, p = 0.05).
level of work with a Doppler flow probe from a suprasternal position. Oxygen consumption was m e a s u r e d with a standard, commercially available exercise testing m a c h i n e ( J a e g e r Q-Plex; Quinton I n s t r u m e n t Co., Seattle, Wash.). Oxygen delivery was calculated from the following equation: Do2 = Q • Ca02
(7)
The Ca02 was calculated as follows: Cao2 = 1.34 x Hemoglobin • Sao2
(8)
No correction was m a d e for dissolved oxygen. W e believed t h a t the small error incurred was not worth the difficulties
involved in sampling arterial blood. Hemoglobin was measured from a capillary blood sample, and Sao2 was measured by continuous pulse oximetry. Error estimation and statistical methods. W e a t t e m p t e d to assess m e a s u r e m e n t error in "r and Do2 for the cardiac surgery group. T h e V02 was m e a s u r e d in a d u m m y ventilator circuit while varying oxygen flows were bled into the system by using an a c c u r a t e r o t a m e t e r (series 150 flowmeter; U n i o n Carbide, Somerset, N.J.). M e a s u r e d and actual oxygen flows were c o m p a r e d by using the method of Bland and Altman. 18 N o absolute s t a n d a r d for Q exists; we could measure only the individual error for each component of the Fick equation (V02, partial pressures of arterial and mixed
2 12
Seear, Wensley, and MacNab
Fig. 3. Schematic representation of relationship between Vo2 and Do2 in children at different levels of oxygen demand. Shaded area represents relationship during exercise(i.e., varying demand). Circle represents normal resting values. Solid lines represent the relationship at two levels of constant demand. True supply dependence occurs when a poorly perfused patient is resuscitated (a-b). Blood transfusion at point b produces no further increase in Vo2 index (b-c) because demand is constant. Adrenaline infusion at point b raises the Vo2 index plateau (exercise effect), giving a false impressionof supply dependence (b-e). iZo2I,~'o2 index; Do2I, Do2 index.
venous oxygen, hemoglobin level, Sao2, Svo2). For each variable, 5 to 10 measurements of a single sample were made and a coefficient of variation calculated: (SD/ mean) X 100. In the same way, we assessed the reproducibility of Doppler-derived Q during exercise by calculating a coefficient of variation for repeated measures of MAFV and aortic diameter. Spontaneous variability in 37o2 was assessed by making regular measurements of 3702 for at least 30 minutes before the treatment period began. The percentage increase between highest and lowest values was used as an indicator of variability, after the method of Weissman et al. 19 All data are expressed as mean +__SD. The regression line for pooled exercise data was calculated by the method of least squares. Comparisons between the 37o2 index and the Do2 index values before and after a treatment were made by paired t test; significance level was 5%. The error in measuring 3702 was expressed by the method of Bland and Altman, 18 where the difference between two simultaneous measurements is expressed as a percentage of their mean value. RESULTS Cardiac surgery groups. The descriptive features of the two treatment subgroups are given in Table I. The pre-
The Journal of Pediatrics August 1993
treatment and posttreatment hemodynamic measurements are shown in Table II. Blood transfusions increased the Do2 index significantly but did not alter any other measurements, including the V02 index (Fig. 1, top). Infusions of adrenaline increased both the 37o2 index and the Do2 index significantly (Fig. 1, bottom), but there were no significant increases in any other measurements, including Sv02 and OER (Table II). Exercise group. Successful measurements of Q were made in 26 exercising children by using Doppler echocardiography. The technique was dependent on the skills of the ultrasonographer; even under ideal conditions, we could not obtain adequate measurements of aortic velocity in five other children who were excluded from the study. The relationship between the 37o2 index and the Do2 index (with 37o2 and Do2 measured in milliliters per minute per kilogram of body weight) during submaximal exercise is shown in Fig. 2. It is described by the following equation: V02 index = Do2 index x 0,88 - 6.95
(9)
We could not find published values for the relationship between the 37o2 index and Do2 index during exercise in children. However, our results for Q (in liters per minute) and 37o2 measured during exercise compare well with the findings reported in the available literature. 2~The relationship is described by the following equation: Q = 5.5 x Vo2 + 2.3
(10)
Measurement errors. The average difference between measured and actual values of 3702 with the use of our technique was 4.7%. The coefficients of variation for repeated measures of the individual Fick components were as follows: Sao2, 2.2%; Svo2, unknown; partial pressures of arterial and mixed venous oxygen, 3.9%; hemoglobin level, 5.1%; and 37o2, 5.3%. Short-term 3702 was surprisingly variable (mean 12.3%, range 6.2% to 19.6%) even though care was taken not to disturb the patient. The coefficients of variation for additional measures of the Doppler components were as follows: MAFV, 9.3% and aortic diameter, 4.9%. DISCUSSION Although our study of the exercise group expands the information available from a standard pediatric exercise test, our main interest was in obtaining normal pediatric values for 37o2 and Do2 over a wide range of oxygen demand. Direct comparisons between the surgical and exercise groups should be made cautiously. During exercise, Do2 is dictated by oxygen demand; in the postsurgical group the reverse is true. In addition, measurement techniques varied between the groups. There was a striking difference between the two groups
The Journal of Pediatrics Volume 123, Number 2
in their ability to utilize transported oxygen. After surgery, patients had ~'o2 indexes barely higher than those of normal resting children, even though several patients could achieve Do2 indexes equal to moderate exercise. The primary problem appeared to be poor oxygen uptake, not flow limitation. The mean OER was less than 25%, considerably less than the 65% to 70% achieved by healthy exercising children. The newer theory of Vo2 supply dependence is compatible with the classic view when applied to patients with lactic acidosis and a large oxygen debt. If Do2 is at less than a critical level, the clinical manifestations are poor perfusion, low urine output, and elevated blood lactate concentration. The patient requires resuscitation, so any increase in Do2 will increase the volume of tissue participating in gas exchange; consequently, Vo2 will rise (Fig. 3). Both Haupt et al. 21 and Gilbert et al. 22 have shown in adults with sepsis that supply dependence after blood transfusions is found only in those patients who had lactic acidosis, although dependence has been described in patients with normal lactate levels. 23 The two theories differ once a patient is considered well resuscitated. This poorly defined condition usually rests on the attainment of predefined physiologic variables, including a normal blood lactate concentration,2 even though clinical assessment is relatively crude and measurements of pH and lactate levels vary widely for many reasons. In stable children with normal lactate levels, we found no significant rise in Vo2 when Do2 was increased by erythrocyte transfusion. Conversely, infusions of adrenaline appeared to uncover oxygen supply dependence. In the absence of a measure of cellular metabolism (such as cytochrome oxidase saturation),z4 it is difficult to decide whether this represents any advantage to the child. Because there was no change in OER (which correlates with microvascular surface area 25) or Svo2, we thought that the rise was not due to improved perfusion but was more likely the result of increased oxygen demand, analogous to exercise. The stimulation of Vo2 by catecholamine infusion is well described in human beings. 14 Villar et al. 9 demonstrated apparent supply dependence in adults even when no therapeutic interventionswere made and thought that it was due to spontaneous variations in ~ro2 and oxygen demand. This point is of more than academic interest because it is possible that the use of high doses of catecholamines is harmful in critically ill children with normal blood lactate values. The combination of local vasoconstriction and increased metabolic demand might lead to areas of unnecessary anaerobic metabolism. Fellows et al. 14 found that infusions of adrenaline, 0.05/zg/kg per minute, caused a 24% increase in metabolic rate with a significant reduction
Seear, Wensley, and MacNab
2 13
in hand blood flow; blood lactate values increased significantly in healthy volunteers. Wilson et al. 26 reported a similar finding when using adrenaline infusions in adults with septic shock; blood lactate levels rose significantly despite an increase in Do2 of 25%. From our results, we believe that in stable children after cardiac surgery with normal blood lactate values, V02 is independent of increases in Do2 (Fig. 3). Any therapy that increases demand (such as high doses of inotropes) simply raises the overall metabolism of perfused cells. Because demand cannot be quantified, the increase in "r can be misinterpreted as implying supply dependency when, in fact, the change represents normal exercise physiology and gives no perfusion advantage to the child. The use of the V02 and Do2 relationship as an indicator of tissue perfusion has other weaknesses, apart from the nonquantifiableeffect of varying oxygen demand. The slope of the relationship during supply dependence is commonly reported in the range of 10% to 15%.27 Unfortunately, measurement error is at least as big as this proposed slope. We found that spontaneous variability in ~/o 2 averaged 12.3% with a range up to 19.6%; although we could assess error only of the individual components of Do2, its total error was probably at least this large. Even if measurements could be made accurately, the clinical value would still be limited because whole body values for "~o2 and Do2 can never provide information about perfusion of individual organs such as kidney and gut. In summary, we could not support the concept of oxygen supply dependence or the clinical use of ~'o2 and Do2 measurements on which it is based. The measurement errors and methodologic problems that we report are not specific to children after cardiac surgery and may apply to other critically ill patients. REFERENCES
1. Pollack MM, Fields AI, Ruttimann UE. Sequential cardiopulmonary variables of infants and children in septic shock. Crit Care Med 1984;12:554-9. 2. Pollack MM, Fields AI, Ruttimann UE. Distribution of eardiopulmonaryvariables in pediatric survivorsand nonsurvivors of septic shock. Crit Care Med 1985;13:454-9. 3. Shepard AP, Granger AJ, Smith EE, Guyton AC. Local control of tissue oxygen delivery and its contribution to the regulation of cardiac output. Am J Physiol 1973;225:747-55. 4. Cain SM. Peripheral oxygen uptake and deliveryin health and disease. Clin Chest Med 1983;4:139-48. 5. Kaufman BS, Rackow EC, Falk JL. The relationship between oxygen delivery and consumption during fluid resuscitation of hypovolemic and septic shock. Chest 1984;85:336-40. 6. Shoemaker WC, Appel PL, Kram HB. Hemodynamicand oxygen transport effects of dobutamine in critically ill general surgical patients. Crit Care Med 1986;14:1032-6. 7. Mohsenifar Z~ Goldbach P, Tashkin DP, Campisi DJ. Rela-
2 14
8.
9.
10.
11.
12.
13.
14.
15.
16.
Seear, Wensley, and MacNab
tionship between 02 delivery and 0 2 consumption in the adult respiratory distress syndrome. Chest 1983;84:267-71. Clarke C, Denis Edwards J, Nightingale P, Mortimer A J, Morris J. Persistence of supply dependency of oxygen uptake at high levels of delivery in adult respiratory distress syndrome. Crit Care Med 1991;19:497-502. Villar J, Slutsky AS, Hew E, Aberman A. Oxygen transport and oxygen consumption in critically ill patients. Chest 1990; 98:687-92. Ronco J J, Phang PT, Walley KR, Wiggs B, Fenwick JC, Russell JA. Oxygen consumption is independent of changes in oxygen delivery in severe adult respiratory distress syndrome. Am Rev Respir Dis 1991;143:1267-73. Lucking SE, Williams TM, Chaten FC, Metz RI, Mickell JJ. Dependence of oxygen consumption on oxygen delivery in children with hyperdynamic septic shock and low oxygen extraction. Crit Care Med 1990;18:1316-9. Mink RB, Pollack MM. Effect of blood transfusion on oxygen consumption in pediatric septic shock. Crit Care Med 1990; 18:1087-91. Dietrich KA, Conrad SA, Herbert CA, Levy GF, Romero MD. Cardiovascular and metabolic response to red blood cell transfusion in critically ill volume-resuscitated nonsurgical patients. Crit Care Med 1990;18:940-4. Fellows IW, Bennet T, MacDonald IA. The effect of adrenaline upon the cardiovascular and metabolic functions in man. Clin Sci 1985;69:215-22. Marx GR, Hicks RW, Allen HD, Kinger SM. Measurement of cardiac output and exercise factor by pulsed Doppler echocardiography during supine bicycle ergometry in normal young adolescent boys. J Am Coll Cardiol 1987;10:430-4. Christie J, Sheldahl CM, Tristani FE, Sagar KB, Ptacin M J, Wann S. Determination of stroke volume and cardiac output during exercise: comparison of two-dimensional and Doppler echocardiography, Fick oximetry, and thermodilution. Circulation 1987;76:539-47.
The Journal of Pediatrics August 1993
17. Huntsman LL, Stewart DK, Barnes SR, Franklin SB, Colocousis JS, Hessef EA. Non-invasive Doppler determination of cardiac output in man: clinical validation. Circulation 1983; 67:593-7. 18. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. 19. Weissman C, Kemper M. Damask MC, Askanazi J, Hyman AL, Kinney JM. Effect of routine intensive care interactions on metabolic rate. Chest 1984;86:815-8. 20. Godfrey S. Exercise testing in children. Philadelphia: WB Saunders, 1974:97-101. 21. Haupt MT, Gilbert EM, Carlson RW. Fluid loading increases oxygen consumption in septic patients with lactic acidosis. Am Rev Respir Dis 1985;131:912-6. 22. Gilbert EM, Haupt MT, Mandanas RY, Huaringa A J, Carlson RW. The effect of fluid loading, blood transfusion, and catechotamine infusion on oxygen delivery and consumption in patients with sepsis. Am Rev Respir Dis 1986;134:873-8. 23. Weg J. Oxygen transport in adult respiratory distress syndrome and other acute circulatory problems: relationship of oxygen delivery and oxygen consumption. Crit Care Med 1991;19:650-7. 24. Brazy J. Near-infrared spectroscopy. Clin Perinatol 1991; 18:519-34. 25. Beer G, Yonce LR. Blood flow, oxygen uptake, and capillary filtration in resting skeletal muscle. Am J Physiol 1972; 223:492-8. 26. Wilson W, Lipman J, Scribante J, et al. Septic shock: does adrenaline have a role as a first line inotropic agent? Anaesth Intensive Care 1992;20:470-4. 27. Dantzker DR, Foresman B, Gutierrez G. Oxygen supply and utilization relationships: a re-evaluation. Am Rev Respir Dis 199l;143:675-9.