- Email: [email protected]
Oxygen transport and oxygen debt and their relation to organ failure and mortality
3 Oxygen transport and oxygen debt and their relation to organ failure and mortality W I L L I A M C. S H O E M A K E R
The classic description of hypoxia was made by B arcroft in 1920. Low flow was classified as 'stagnant hypoxia', low arterial blood oxygen tension as 'hypoxic hypoxia', and low haemoglobin concentrations as 'anaemic hypoxia'. Later, histiocytic or tissue hypoxia was added to these concepts. Shock was subsequently defined by Blalock and associates (Levy and Blalock, 1938; Blalock, 1940) as a failure of tissue perfusion. They observed decreased total body, renal and splanchnic oxygen uptake in haemorrhaged dogs. Traditionally the overall rate of body metabolism has been evaluated by oxygen consumption (fVo2), which may be measured by direct or indirect calorimetry (Atwater and Rosa, 1899; Benedict and Higgins, 1912; DuB .ois, 1921). Cuthbertson (1932) reported that after orthopaedic injuries, Vo2 initially decreased and then increased. He regarded this ebb and flow of oxygen as the common pattern of body metabolism after trauma. Later Wiggers (1950) reviewed and evaluated cardiac output and other haemodynamic variables as the major features of shock and shock-related circulatory problems.
M E A S U R E M E N T OF B L O O D F L O W AND OXYGEN
METABOLISM Over a century ago cardiac output was estimated by the direct Fick method. In this method, 1702was measured directly by spirometry and this ~g02value was then divided by the arteriovenous oxygen content. C(a-9)02, expressed in comparable units, to calculate the cardiac output. Since the advent of the balloon-tipped flow-directed pulmonary artery catheter, cardiac output is conventionally measured by thermodilution and V02 is calculated as the product of the cardiac index and the C(a-'~)02. Oxygen delivery (/)o2) is calculated as the product of the cardiac index and the arterial oxygen content. (Ca02). The Do2 reflects the overall circulatory function, and increased Do2 values represent the capacity to compensate for inadequate tissue perfusion and tissue hypoxia. This is particularly relevant when there is increased body metabolism produced by trauma, stress, sepsis and other hypermetabolic clinical conditions. Bailli&e's Clinical Anaesthesiology--
Vol. 6, No. 1, March 1992 ISBN 0-7020-1616-0
39 Copyright 9 1992, by Bailli6re Tindall All rights of reproduction in any form reserved
40
W.C. SHOEMAKER
Low flow has been described after haemorrhage and cardiogenic shock produced experimentally in animals as well as in the comparable clinical conditions (Wiggers, 1950; Guyton and Crowell, 1961; Crowell and Smith, 1964). Cournand et al (1943) observed low flow and low Vo2 associated with hypovolaemia and hypotension in traumatized patients. However, increased cardiac output was reported after trauma, sepsis and cirrhosis (Shoemaker et al, 1973, 1988; Bland et al, 1985).
O X Y G E N T R A N S P O R T PATTERNS IN A C C I D E N T A L AND SURGICAL TRAUMA
Temporal haemodynamic and oxygen transport patterns have been described in survivors and non-survivors in high risk critically, ill surg!cal patients. Survivors had supranormal values for cardiac index, Do2 and Vo2 when their blood volume, central venous pressure and wedge pressures were normal or somewhat above normal. The haemodynamic and oxygen transport values of survivors were greater than those of the non-survivors. The arterial and mixed venous blood gases of both groups were usually within the normal range, while pulmonary artery pressure and the pulmonary vascular resistance index (PVRI) were higher in the non-survivors (Cournand et al, 1943; Shoemaker et al, 1973, 1988; Bland et al, 1985). In survivors of high risk surgery, the median values were: cardiac index 4.5 litres.min-l.m -2 oxygen delivery 600 ml.min-Lm -2 and Vo2 170 ml-min-l.m -2 (Shoemaker et al, 1988). In prospective controlled studies, high risk surgical patients were preoperatively randomized to a control group that were maintained at normal haemodynamic and oxygen transport values and a protocol group whose therapeutic goals were the empirically determined supranormal values. There were marked reductions in mortality, morbidity, organ failure, days spent in the intensive care unit, hospital days and costs in the protocol group (Shoemaker et al, 1988).
TISSUE O X Y G E N DEBT IN E X P E R I M E N T A L CONDITIONS
It is evident that l)oz measures the rate of oxygen actually consumed, and not necessarily the rate of Vo2 that is needed. In order to estimate the oxygen need in shock, Guyton and colleagues (Guyton and Crowell, 1961; Crowell and Smith, 1964) measured 1/o2 in dogs under controlled anaesthetized conditions before, during and after haemorrhage. They calculated the net cumulative oxygen debt as the time-integrated difference between the Vo2 after shock minus the 1)'o2measured under controlled conditions but extrapolated to the haemorrhagic and posthaemorrhagic periods. Dogs that accumulated oxygen debts of less than 100 ml.kg -1 all survived, while dogs with more than 140 ml.kg- 1 of oxygen debt all died; 50% mortality occurred with oxygen debts of 120 ml.kg-L
OXYGEN DEBT
41
METHOD FOR ESTIMATING THE TISSUE OXYGEN DEBT IN SURGICAL PATIENTS Correction terms were used to estimate the {7o2need under anaesthesia and at various temperatures. Lowe and Ernst (1981) reported that in healthy elective patients, the V02 under anaesthesia is 10 x kg~ This value was corrected for the effects of temperature and was used to estimate the V02 need during the time the patient underwent general anaesthesia. It was approximately equivalent to 100ml-min-l.m -2. The temperature correction term assumed that metabolic activity increased or decreased at the rate of 7% per degree Fahrenheit (Shoemaker et al, 1990). The I)02 need after recovery from anaesthesia in the postoperative period was estimated from the patient's own baseline r~02 measured pregperatively and corrected for the effects of temperature. Thu% the rate of V02 deficit was calculated from the measured Voz minus the V02 need estimated from the patient's own resting preoperative control values corrected for both temperature and anaesthesia.
HAEMODYNAMIC AND OXYGEN TRANSPORT PATTERNS IN RELATION TO OXYGEN DEBT IN CRITICALLY ILL POSTOPERATIVE PATIENTS Figure 1 shows the patterns of selected haemodynamic and oxygen transport variables in a series of 253 consecutively monitored high risk surgical patients in the preoperative, intraoperative and immediate postoperative period (Shoemaker et al, 1990, 1991). Cardiac index, oxygen delivery and Vo2 of the 158 survivors without organ failure were highest; these values were intermediate in 31 patients who survived with organ failure. The 64 patients who subsequently died had the lowest values which were within the normal range.
OXYGEN DEBT IN POSTOPERATIVE PATIENTS
The calculated oxygen debt of this series of consecutively monitored high risk patients was related to multiple organ failures, complications and outcome (Table 1). All 64 patients who died had organ failure; their cumulative ~'o2 deficit averaged 33.2 + 4 (SEM) litres.m -2 at its maximum, which occurred 17.8 + 2.2h postoperatively, and it took an average of 48.6+ 4.2h postoperatively before the oxygen debt was corrected. In the 31 survivors with organ failure, the cumulative Vo2 averaged 21.6 + 3.7 litres.m -2 at its maximum, which occurred at 10.1 + 2.7 h and lasted 29.2 + 4.7 h postoperatively (P < 0.05). In the 158 survivors without organ failure or major complications, Jr. 2 at the maximum cumulative (/o2 deficit averaged 9.2_l.31itres.m4.1 + 0.6 h and lasted 17.9 + 1.6 h postoperatively ( P < 0.05).
42
w.c.
Intraoperative 1 Preop I - ~ 2
4
8
" I I I
I
I
Hours Postoperative 12 24
I
I
36
SHOEMAKER
48
I
I
Cardiac Index
. "I
L/mJn.m 2
s"
O2Delivery
'.
i
j
/"
600
"~[ ml/min.m 2
',~ ,/ I
120
I
02 Consumption
.-'1 ......
,,o
500
!" *
t
400
t .......... t
"["........ ] ........... ""
..t ...... "-""
,oo?,o
~l
cess,
"'I"
LJm 2 2o
10 s
..... -
-
Survivors Without Organ Failure Survivors With Organ Failure Nonsurvivors
[
4o
Figure 1. Serial measurements of cardiac index, oxygen delivery, oxygen consumption and net cumulative 1/o2 deficit in survivors without organ failure, survivors with organ failure and non-survivors calculated for the intraoperative period and successive time periods postoperatively. Modified from Shoemaker et al (1992).
43
OXYGEN DEBT
1)'o2 debt and the time taken to reach the maximum Vo2 debt for non-survivors, survivors with organ failure and survivors without organ failure (values are mean + SEM). Table 1. Intraoperative
Intr.aoperative Maximum Time.to maximum Vo2 debt Vo2 debt Vo2 debt (litres.m -2) (litres.m 2) (h) Non-survivors (n = 64) Survivors with organ failure (n = 31) Survivors with no organ failure (n = 158)
12.0 + 1.3 11.8 _+1.6 5.7 + 0.9
33.2 + 4 21.6 + 3.7 9.2 + 1.3
17.8 + 2.2 10.1 + 2.7 4.1 _+0.6
T h u s , the m a g n i t u d e d u r a t i o n of the calculated 1)02 deficit was greatest with n o n - s u r v i v o r s , slightly less in survivors with o r g a n failure, a n d lowest in survivors w i t h o u t o r g a n failure.
O X Y G E N D E B T IN A R A N D O M I Z E D C L I N I C A L T R I A L T h e oxygen d e b t was calculated in a prospective r a n d o m i z e d study testing the effect of s u p r a n o r m a t values empirically o b s e r v e d in survivors in critically ill surgical p a t i e n t s ( S h o e m a k e r et al, 1990). Figure 2 shows smaller o x y g e n debts in p r o t o c o l p a t i e n t s who had s u p r a n o r m a l oxygen t r a n s p o r t values as t h e r a p e u t i c goals c o m p a r e d with the c o n t r o l group who had n o r m a l values as t h e r a p e u t i c goals.
Intra-
Hours post-op
Control 1 _ _ 2 _ _ 4
Cumulative excess.
8
12
24
36
48
or deficit (ml min4-m z) 20 OOO
10 000 0
\ ~
- 1 0 000
-f
•
"~---~-----~.~
T
T
CONTROL T
t
I//
-20 000 ----30
000
--CONTROL i
i
i
i
J-
I
~
4-
Figure 2. Serial measurements of the net cumulative oxygen deficit or excess for a prospectively randomized control group with normal values as therapeutic goals and a protocol group with supranormal oxygen transport values as goals of therapy. Modified from Shoemaker et al (1990).
44
W. C. SHOEMAKER
E +1+1+1+1+1 ii
t_., = o,I
9
"E
o=
9
"6
!
n~
kl
o
t.,
,.b, t~
'~ ~ . .
o~
~-~.= o~ .-, : on ,", "-~
=
o
0 (",1
"~
~
"~ gM
~
a~ u
";
o Z
9
A ~ A ~ I_~ =~x o ~A
I v =..u e ~ ,.. /~. = - =v'~
I~.=.~
~ ..,'-'
o-~
/ r~ o = . n ~ . n ..
| ~
~
~.;
.~, 9
~
E 0 ~Z
0
PI
OXYGEN DEBT
45
SPECIFIC O R G A N FAILURES AND THEIR TIMES OF ONSET
Table 2 details the number and type of organ failures in non-survivors and survivors. There were 1.72 organ failures per non-survivor, and 1.42 organ failures per survivor with organ failure or 0.28 organ failures per survivor with or without organ failure. The average time of appearance of each organ failure is listed in Table 2. Although there were wide ranges in the time of appearance of each organ failure, pulmonary failure often occurred first, followed by cardiac failure, renal failure, disseminated intravascular coagulation, sepsis and central nervous system failure. However, the order of appearance varied widely. The average organ failure occurred 5.5 + 4.3 days postoperatively; this was later than the oxygen debt, which began intraoperatively and reached maximum values in 10.1 h postoperatively for survivors and in 17.8 h for non-survivors. Organ failures usually became clinically apparent after the maximum cumulative oxygen debt was reached, but before the debt had been entirely corrected. SUMMARY AND CONCLUSIONS The data demonstrate a strong relationship between the magnitude and duration of the 17o2 deficit in the intraoperative and early postoperative period and the subsequent appearance of organ failure and death. There were greater tissue oxygen debts in surviving patients who subsequently developed multiple organ failure than in surviving patients without organ failure. The oxygen deficits were greater in magnitude and duration in non-survivors than in those who survived with organ failure. Moreover, the early appearance of oxygen debt suggest that reduced tissue oxygenation is the primary event, while organ failure and death are the result of this antecedent physiological event. This is supported by prospective clinical trials that demonstrated a reduced morbidity and mortality when the 1702 had been maintained at supranormal values. Thus, evidence suggests that reduced tissue oxygenation from maldistributed or inadequate tissue perfusion in the face of increased metabolic need is an early pathogenic mechanism that produces organ failure and subsequently death. The early underlying mechanism responsible for postoperative shock is the reduced or unevenly distributed blood flow and Do2 in the face of increased metabolic need, resulting in inadequate tissue perfusion and oxygenation. Insufficient tissue perfusion produces tissue hypoxia, that directly leads to organ failure and subsequently to death. The function of the circulation is to transport oxygen, oxidative substrates and other blood constituents. When deficiencies occur, the body compensates by increasing the cardiac output, oxygen delivery and oxygen extraction; these compensations attempt to maintain the Vo2 at satisfactory levels. The practical clinical implications for physicians at the bedside is to augment the body's natural compensations by further increasing oxygen delivery to achieve supranormal optimal values rapidly and expeditiously.
46
W. C. SHOEMAKER
Criteria to evaluate the adequacy of therapy in various clinical conditions may be determined empirically from survivor's values. These values are greater than normal, but may require further refinement and documentation for each diagnostic and physiological group.. However, from the practical operational viewpoint, the adequacy of the V02 may be tested by increasing the Do2 with fluids (colloids are more effective) or inotropic agents such as dobutamine (Shoemaker et al, 1991). If the V02 increases appreciably with increases in Do2, the V,02 may be considered to be supply dependent and further therapy may be indicated. If the V02 remains higher than normal but does not increase with further /)o2 increases, the Vo2 may be supply independent and, therefore, adequate for the moment. REFERENCES Atwater WD & Rosa EBA (1899) A New Respiration Colorimeter. Bulletin 63. Washington DC: US Department of Agriculture. Barcroft J (1920) On anoxaemia. Lancet ii: 485. Benedict FG & Higgins HL (1912) The influence of the respiratory exchange of varying amounts of carbohydrates in the diet. American Journal of Physiology 30: 217. Blalock A (1940) Principles of Surgical Care, Shock and Other Problems. St Louis: Mosby. Bland RD, Shoemaker WC, Abraham E et al (1985) Hemodynamic and oxygen transport patterns in surviving and nonsurviving postoperative patients. Critical Care Medicine 13: 85. Cournand A, Riley RL, Bradley SE et al (1943) Studies of the circulation in clinical shock. Surgery 13: 964. Crowell JW & Smith EE (1964) Oxygen deficit and irreversible hemorrhagic shock. American Journal of Physiology 106: 313. Cuthbertson DP (1932) Observations on disturbances of metabolism produced by injury to limbs. Quarterly Journal of Medicine 1: 233. DuBois EF (1921) Basal metabohsm in fever. Journal of the American Medical Association 77: 325. Guyton AC & Crowell JW (1961) Dynamics of the heart in shock. Federation Proceedings 20(supplement 9): 51. Levy SE & Blalock A (1938) Effects of unilateral nephrectomy on renal blood flow and oxygen consumption of unanesthetized dogs. American Journal of Physiology 122: 609. Lowe JR (1970) The anesthetic continum. In Aldarete JH, Lowe JH & Virtue RA (eds) Low Flow and Closed @stem Anesthesia. New York: Grune and Stratton. Lowe HJ & Ernst E A (1981) The Quantitative Practice of Anesthesia: Use of Closed Circuit, pp 146-147. Baltimore: Williams & Wilkins. Shoemaker WC, Montgomery ES, Kaplan E et al (1973) Physiologic patterns in surviving and nonsurviving shock patients. Archives of Surgery 106: 630. Shoemaker WC, Appel PL, Kram HB et al (1988) Prospective trial of supranormal values of survivors as therapeutic goals in high risk surgical patients. Chest 94: 1176. Shoemaker WC, Appel PL & Kram HB (1990) The efficacy of central venous and pulmonary artery catheters and therapy based upon them in reducing mortality and morbidity. Archives of Surgery 125: 1332. Shoemaker WC, Appel PL & Kram HB (1991) Oxygen transport measurements to evaluate tissue perfusion and titrate therapy. Critical Care Medicine 19: 672. Shoemaker WC, Appel PL & Kram HB (1992) Multiple vital organ failure in high risk postoperative patients. Chest (in press). Wiggers CJ (1950) The Physiology of Shock. New York: The Commonwealth Fund.
The classic description of hypoxia was made by B arcroft in 1920. Low flow was classified as 'stagnant hypoxia', low arterial blood oxygen tension as 'hypoxic hypoxia', and low haemoglobin concentrations as 'anaemic hypoxia'. Later, histiocytic or tissue hypoxia was added to these concepts. Shock was subsequently defined by Blalock and associates (Levy and Blalock, 1938; Blalock, 1940) as a failure of tissue perfusion. They observed decreased total body, renal and splanchnic oxygen uptake in haemorrhaged dogs. Traditionally the overall rate of body metabolism has been evaluated by oxygen consumption (fVo2), which may be measured by direct or indirect calorimetry (Atwater and Rosa, 1899; Benedict and Higgins, 1912; DuB .ois, 1921). Cuthbertson (1932) reported that after orthopaedic injuries, Vo2 initially decreased and then increased. He regarded this ebb and flow of oxygen as the common pattern of body metabolism after trauma. Later Wiggers (1950) reviewed and evaluated cardiac output and other haemodynamic variables as the major features of shock and shock-related circulatory problems.
M E A S U R E M E N T OF B L O O D F L O W AND OXYGEN
METABOLISM Over a century ago cardiac output was estimated by the direct Fick method. In this method, 1702was measured directly by spirometry and this ~g02value was then divided by the arteriovenous oxygen content. C(a-9)02, expressed in comparable units, to calculate the cardiac output. Since the advent of the balloon-tipped flow-directed pulmonary artery catheter, cardiac output is conventionally measured by thermodilution and V02 is calculated as the product of the cardiac index and the C(a-'~)02. Oxygen delivery (/)o2) is calculated as the product of the cardiac index and the arterial oxygen content. (Ca02). The Do2 reflects the overall circulatory function, and increased Do2 values represent the capacity to compensate for inadequate tissue perfusion and tissue hypoxia. This is particularly relevant when there is increased body metabolism produced by trauma, stress, sepsis and other hypermetabolic clinical conditions. Bailli&e's Clinical Anaesthesiology--
Vol. 6, No. 1, March 1992 ISBN 0-7020-1616-0
39 Copyright 9 1992, by Bailli6re Tindall All rights of reproduction in any form reserved
40
W.C. SHOEMAKER
Low flow has been described after haemorrhage and cardiogenic shock produced experimentally in animals as well as in the comparable clinical conditions (Wiggers, 1950; Guyton and Crowell, 1961; Crowell and Smith, 1964). Cournand et al (1943) observed low flow and low Vo2 associated with hypovolaemia and hypotension in traumatized patients. However, increased cardiac output was reported after trauma, sepsis and cirrhosis (Shoemaker et al, 1973, 1988; Bland et al, 1985).
O X Y G E N T R A N S P O R T PATTERNS IN A C C I D E N T A L AND SURGICAL TRAUMA
Temporal haemodynamic and oxygen transport patterns have been described in survivors and non-survivors in high risk critically, ill surg!cal patients. Survivors had supranormal values for cardiac index, Do2 and Vo2 when their blood volume, central venous pressure and wedge pressures were normal or somewhat above normal. The haemodynamic and oxygen transport values of survivors were greater than those of the non-survivors. The arterial and mixed venous blood gases of both groups were usually within the normal range, while pulmonary artery pressure and the pulmonary vascular resistance index (PVRI) were higher in the non-survivors (Cournand et al, 1943; Shoemaker et al, 1973, 1988; Bland et al, 1985). In survivors of high risk surgery, the median values were: cardiac index 4.5 litres.min-l.m -2 oxygen delivery 600 ml.min-Lm -2 and Vo2 170 ml-min-l.m -2 (Shoemaker et al, 1988). In prospective controlled studies, high risk surgical patients were preoperatively randomized to a control group that were maintained at normal haemodynamic and oxygen transport values and a protocol group whose therapeutic goals were the empirically determined supranormal values. There were marked reductions in mortality, morbidity, organ failure, days spent in the intensive care unit, hospital days and costs in the protocol group (Shoemaker et al, 1988).
TISSUE O X Y G E N DEBT IN E X P E R I M E N T A L CONDITIONS
It is evident that l)oz measures the rate of oxygen actually consumed, and not necessarily the rate of Vo2 that is needed. In order to estimate the oxygen need in shock, Guyton and colleagues (Guyton and Crowell, 1961; Crowell and Smith, 1964) measured 1/o2 in dogs under controlled anaesthetized conditions before, during and after haemorrhage. They calculated the net cumulative oxygen debt as the time-integrated difference between the Vo2 after shock minus the 1)'o2measured under controlled conditions but extrapolated to the haemorrhagic and posthaemorrhagic periods. Dogs that accumulated oxygen debts of less than 100 ml.kg -1 all survived, while dogs with more than 140 ml.kg- 1 of oxygen debt all died; 50% mortality occurred with oxygen debts of 120 ml.kg-L
OXYGEN DEBT
41
METHOD FOR ESTIMATING THE TISSUE OXYGEN DEBT IN SURGICAL PATIENTS Correction terms were used to estimate the {7o2need under anaesthesia and at various temperatures. Lowe and Ernst (1981) reported that in healthy elective patients, the V02 under anaesthesia is 10 x kg~ This value was corrected for the effects of temperature and was used to estimate the V02 need during the time the patient underwent general anaesthesia. It was approximately equivalent to 100ml-min-l.m -2. The temperature correction term assumed that metabolic activity increased or decreased at the rate of 7% per degree Fahrenheit (Shoemaker et al, 1990). The I)02 need after recovery from anaesthesia in the postoperative period was estimated from the patient's own baseline r~02 measured pregperatively and corrected for the effects of temperature. Thu% the rate of V02 deficit was calculated from the measured Voz minus the V02 need estimated from the patient's own resting preoperative control values corrected for both temperature and anaesthesia.
HAEMODYNAMIC AND OXYGEN TRANSPORT PATTERNS IN RELATION TO OXYGEN DEBT IN CRITICALLY ILL POSTOPERATIVE PATIENTS Figure 1 shows the patterns of selected haemodynamic and oxygen transport variables in a series of 253 consecutively monitored high risk surgical patients in the preoperative, intraoperative and immediate postoperative period (Shoemaker et al, 1990, 1991). Cardiac index, oxygen delivery and Vo2 of the 158 survivors without organ failure were highest; these values were intermediate in 31 patients who survived with organ failure. The 64 patients who subsequently died had the lowest values which were within the normal range.
OXYGEN DEBT IN POSTOPERATIVE PATIENTS
The calculated oxygen debt of this series of consecutively monitored high risk patients was related to multiple organ failures, complications and outcome (Table 1). All 64 patients who died had organ failure; their cumulative ~'o2 deficit averaged 33.2 + 4 (SEM) litres.m -2 at its maximum, which occurred 17.8 + 2.2h postoperatively, and it took an average of 48.6+ 4.2h postoperatively before the oxygen debt was corrected. In the 31 survivors with organ failure, the cumulative Vo2 averaged 21.6 + 3.7 litres.m -2 at its maximum, which occurred at 10.1 + 2.7 h and lasted 29.2 + 4.7 h postoperatively (P < 0.05). In the 158 survivors without organ failure or major complications, Jr. 2 at the maximum cumulative (/o2 deficit averaged 9.2_l.31itres.m4.1 + 0.6 h and lasted 17.9 + 1.6 h postoperatively ( P < 0.05).
42
w.c.
Intraoperative 1 Preop I - ~ 2
4
8
" I I I
I
I
Hours Postoperative 12 24
I
I
36
SHOEMAKER
48
I
I
Cardiac Index
. "I
L/mJn.m 2
s"
O2Delivery
'.
i
j
/"
600
"~[ ml/min.m 2
',~ ,/ I
120
I
02 Consumption
.-'1 ......
,,o
500
!" *
t
400
t .......... t
"["........ ] ........... ""
..t ...... "-""
,oo?,o
~l
cess,
"'I"
LJm 2 2o
10 s
..... -
-
Survivors Without Organ Failure Survivors With Organ Failure Nonsurvivors
[
4o
Figure 1. Serial measurements of cardiac index, oxygen delivery, oxygen consumption and net cumulative 1/o2 deficit in survivors without organ failure, survivors with organ failure and non-survivors calculated for the intraoperative period and successive time periods postoperatively. Modified from Shoemaker et al (1992).
43
OXYGEN DEBT
1)'o2 debt and the time taken to reach the maximum Vo2 debt for non-survivors, survivors with organ failure and survivors without organ failure (values are mean + SEM). Table 1. Intraoperative
Intr.aoperative Maximum Time.to maximum Vo2 debt Vo2 debt Vo2 debt (litres.m -2) (litres.m 2) (h) Non-survivors (n = 64) Survivors with organ failure (n = 31) Survivors with no organ failure (n = 158)
12.0 + 1.3 11.8 _+1.6 5.7 + 0.9
33.2 + 4 21.6 + 3.7 9.2 + 1.3
17.8 + 2.2 10.1 + 2.7 4.1 _+0.6
T h u s , the m a g n i t u d e d u r a t i o n of the calculated 1)02 deficit was greatest with n o n - s u r v i v o r s , slightly less in survivors with o r g a n failure, a n d lowest in survivors w i t h o u t o r g a n failure.
O X Y G E N D E B T IN A R A N D O M I Z E D C L I N I C A L T R I A L T h e oxygen d e b t was calculated in a prospective r a n d o m i z e d study testing the effect of s u p r a n o r m a t values empirically o b s e r v e d in survivors in critically ill surgical p a t i e n t s ( S h o e m a k e r et al, 1990). Figure 2 shows smaller o x y g e n debts in p r o t o c o l p a t i e n t s who had s u p r a n o r m a l oxygen t r a n s p o r t values as t h e r a p e u t i c goals c o m p a r e d with the c o n t r o l group who had n o r m a l values as t h e r a p e u t i c goals.
Intra-
Hours post-op
Control 1 _ _ 2 _ _ 4
Cumulative excess.
8
12
24
36
48
or deficit (ml min4-m z) 20 OOO
10 000 0
\ ~
- 1 0 000
-f
•
"~---~-----~.~
T
T
CONTROL T
t
I//
-20 000 ----30
000
--CONTROL i
i
i
i
J-
I
~
4-
Figure 2. Serial measurements of the net cumulative oxygen deficit or excess for a prospectively randomized control group with normal values as therapeutic goals and a protocol group with supranormal oxygen transport values as goals of therapy. Modified from Shoemaker et al (1990).
44
W. C. SHOEMAKER
E +1+1+1+1+1 ii
t_., = o,I
9
"E
o=
9
"6
!
n~
kl
o
t.,
,.b, t~
'~ ~ . .
o~
~-~.= o~ .-, : on ,", "-~
=
o
0 (",1
"~
~
"~ gM
~
a~ u
";
o Z
9
A ~ A ~ I_~ =~x o ~A
I v =..u e ~ ,.. /~. = - =v'~
I~.=.~
~ ..,'-'
o-~
/ r~ o = . n ~ . n ..
| ~
~
~.;
.~, 9
~
E 0 ~Z
0
PI
OXYGEN DEBT
45
SPECIFIC O R G A N FAILURES AND THEIR TIMES OF ONSET
Table 2 details the number and type of organ failures in non-survivors and survivors. There were 1.72 organ failures per non-survivor, and 1.42 organ failures per survivor with organ failure or 0.28 organ failures per survivor with or without organ failure. The average time of appearance of each organ failure is listed in Table 2. Although there were wide ranges in the time of appearance of each organ failure, pulmonary failure often occurred first, followed by cardiac failure, renal failure, disseminated intravascular coagulation, sepsis and central nervous system failure. However, the order of appearance varied widely. The average organ failure occurred 5.5 + 4.3 days postoperatively; this was later than the oxygen debt, which began intraoperatively and reached maximum values in 10.1 h postoperatively for survivors and in 17.8 h for non-survivors. Organ failures usually became clinically apparent after the maximum cumulative oxygen debt was reached, but before the debt had been entirely corrected. SUMMARY AND CONCLUSIONS The data demonstrate a strong relationship between the magnitude and duration of the 17o2 deficit in the intraoperative and early postoperative period and the subsequent appearance of organ failure and death. There were greater tissue oxygen debts in surviving patients who subsequently developed multiple organ failure than in surviving patients without organ failure. The oxygen deficits were greater in magnitude and duration in non-survivors than in those who survived with organ failure. Moreover, the early appearance of oxygen debt suggest that reduced tissue oxygenation is the primary event, while organ failure and death are the result of this antecedent physiological event. This is supported by prospective clinical trials that demonstrated a reduced morbidity and mortality when the 1702 had been maintained at supranormal values. Thus, evidence suggests that reduced tissue oxygenation from maldistributed or inadequate tissue perfusion in the face of increased metabolic need is an early pathogenic mechanism that produces organ failure and subsequently death. The early underlying mechanism responsible for postoperative shock is the reduced or unevenly distributed blood flow and Do2 in the face of increased metabolic need, resulting in inadequate tissue perfusion and oxygenation. Insufficient tissue perfusion produces tissue hypoxia, that directly leads to organ failure and subsequently to death. The function of the circulation is to transport oxygen, oxidative substrates and other blood constituents. When deficiencies occur, the body compensates by increasing the cardiac output, oxygen delivery and oxygen extraction; these compensations attempt to maintain the Vo2 at satisfactory levels. The practical clinical implications for physicians at the bedside is to augment the body's natural compensations by further increasing oxygen delivery to achieve supranormal optimal values rapidly and expeditiously.
46
W. C. SHOEMAKER
Criteria to evaluate the adequacy of therapy in various clinical conditions may be determined empirically from survivor's values. These values are greater than normal, but may require further refinement and documentation for each diagnostic and physiological group.. However, from the practical operational viewpoint, the adequacy of the V02 may be tested by increasing the Do2 with fluids (colloids are more effective) or inotropic agents such as dobutamine (Shoemaker et al, 1991). If the V02 increases appreciably with increases in Do2, the V,02 may be considered to be supply dependent and further therapy may be indicated. If the V02 remains higher than normal but does not increase with further /)o2 increases, the Vo2 may be supply independent and, therefore, adequate for the moment. REFERENCES Atwater WD & Rosa EBA (1899) A New Respiration Colorimeter. Bulletin 63. Washington DC: US Department of Agriculture. Barcroft J (1920) On anoxaemia. Lancet ii: 485. Benedict FG & Higgins HL (1912) The influence of the respiratory exchange of varying amounts of carbohydrates in the diet. American Journal of Physiology 30: 217. Blalock A (1940) Principles of Surgical Care, Shock and Other Problems. St Louis: Mosby. Bland RD, Shoemaker WC, Abraham E et al (1985) Hemodynamic and oxygen transport patterns in surviving and nonsurviving postoperative patients. Critical Care Medicine 13: 85. Cournand A, Riley RL, Bradley SE et al (1943) Studies of the circulation in clinical shock. Surgery 13: 964. Crowell JW & Smith EE (1964) Oxygen deficit and irreversible hemorrhagic shock. American Journal of Physiology 106: 313. Cuthbertson DP (1932) Observations on disturbances of metabolism produced by injury to limbs. Quarterly Journal of Medicine 1: 233. DuBois EF (1921) Basal metabohsm in fever. Journal of the American Medical Association 77: 325. Guyton AC & Crowell JW (1961) Dynamics of the heart in shock. Federation Proceedings 20(supplement 9): 51. Levy SE & Blalock A (1938) Effects of unilateral nephrectomy on renal blood flow and oxygen consumption of unanesthetized dogs. American Journal of Physiology 122: 609. Lowe JR (1970) The anesthetic continum. In Aldarete JH, Lowe JH & Virtue RA (eds) Low Flow and Closed @stem Anesthesia. New York: Grune and Stratton. Lowe HJ & Ernst E A (1981) The Quantitative Practice of Anesthesia: Use of Closed Circuit, pp 146-147. Baltimore: Williams & Wilkins. Shoemaker WC, Montgomery ES, Kaplan E et al (1973) Physiologic patterns in surviving and nonsurviving shock patients. Archives of Surgery 106: 630. Shoemaker WC, Appel PL, Kram HB et al (1988) Prospective trial of supranormal values of survivors as therapeutic goals in high risk surgical patients. Chest 94: 1176. Shoemaker WC, Appel PL & Kram HB (1990) The efficacy of central venous and pulmonary artery catheters and therapy based upon them in reducing mortality and morbidity. Archives of Surgery 125: 1332. Shoemaker WC, Appel PL & Kram HB (1991) Oxygen transport measurements to evaluate tissue perfusion and titrate therapy. Critical Care Medicine 19: 672. Shoemaker WC, Appel PL & Kram HB (1992) Multiple vital organ failure in high risk postoperative patients. Chest (in press). Wiggers CJ (1950) The Physiology of Shock. New York: The Commonwealth Fund.