| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Department of Anesthesiology and Critical Care, Faculty of Health Services, University of Stellenbosch, Tygerberg, South Africa
Address correspondence and reprint requests to Andrew I. Levin, Department of Anesthesiology and Critical Care, Faculty of Health Sciences, University of Stellenbosch, PO Box 19063, Tygerberg 7505, South Africa. Address e-mail to ail{at}sun.ac.za
Szegedi et al. (1) studied the effects of normovolemic acute hemodilution on arterial oxygenation in supine subjects during one-lung anesthesia (OLA). After hemodilution, patients who had chronic obstructive airways disease (COAD) exhibited decreases in PaO2, whereas there were no changes in subjects with normal lungs, nor in a control group with COAD who did not undergo hemodilution. The effects of hemodilution during OLA have not been previously described, and the reasons for their findings are not obvious. Intuitively, one may suspect that hemodilution increased pulmonary shunting in the COAD group. There are however, a number of factors influencing arterial oxygenation during OLA that need to be considered (2). These factors may be introduced by considering the shunt equation:
|
|
where Qs represents shunt flow, Qt cardiac output, Qs/Qt pulmonary shunt fraction CcO2 end-capillary oxygen content, CaO2 arterial oxygen content, CvO2 mixed venous oxygen content. Rearranging Equation 1 gives:
|
|
In addition to the magnitude of the shunt (Qs/Qt), Equation 2 indicates that the determinants of CaO2 also include CcO2 and CvO2. CcO2 is primarily determined by the hemoglobin concentration, provided lung parenchyma is relatively normal and inhaled oxygen partial pressures and alveolar ventilation are adequate. However, the determinants of CvO2 are more complex. These are given by the Fick relationship for oxygen consumption (
O2):
|
|
Therefore, the four variables that influence CaO2 during OLA are Qs/Qt, Hb concentration (predominant determinant of CcO2), and the two variables that determine CvO2: Qt and O2. It is important to note that it is the ratio O2/Qt that influences CvO2 and therefore arterial oxygenation in the presence of a shunt. Kelman et al. (3) combined the shunt and Fick equations into a single expression that describes these complex relationships:
|
|
A graph of Equation 4 (Figure 1) depicts the influence of cardiac output on CaO2. Plot A in Figure 1 portrays this relationship when Hb concentration is 15 g/dL and O2 150 mL/min in the presence of a Qs/Qt of 0.2. Inspection of plot A reveals that at low cardiac output values, CaO2 is reduced because high O2/Qt ratios increase the value of the second term in Equation 4. With increasing cardiac output, these ratios decrease, causing CaO2 to increase steeply until the O2/Qt ratios assume such small values that the second term in Equation 4 becomes negligible. The result is that CaO2 approaches CcO2 asymptotically at high cardiac outputs. Plot B illustrates how the relationship shifts downwards if the shunt fraction doubles to 0.4. Plot C shows how plot A moves upwards and to the left if O2 is halved. Plot D demonstrates that if Hb concentration is reduced from 15 to 10 g/dL, curve A is shifted downwards.
|
The first approach to improving arterial oxygenation is to increase cardiac output (Figure 1) and several publications have validated the usefulness of this approach (4,7–9). Two studies (7,10) have demonstrated increases in arterial oxygenation during OLA when increasing cardiac output by administering small doses of inotropic drugs (dobutamine infusion rates of 5 µg · kg–1 · min–1). Slinger and Scott (8) have also shown that an increased cardiac output during OLA was associated with better arterial oxygenation.
The benefits of increasing CaO2 by increasing cardiac output are limited, as is demonstrated by the flattening of the curve in Figure 1 that occurs when cardiac outputs approach high values. This has been confirmed in a recent study (21) using progressively increasing dobutamine infusion rates (3, 5, and 7 µg · kg–1 · min–1) to increase cardiac output. Furthermore, Equation 4 does not take into account the potential for increased cardiac output to increase PAP that may, in turn, overcome the weak forces that bring about HPV (11,12). In addition, inotropic drugs can inhibit HPV directly (10). Deleterious effects on arterial oxygenation during OLA in a supine animal model have been demonstrated when sufficient inotrope was administered to double cardiac output compared with baseline values (13).
Another approach, again evident from Equation 4, is that a decrease in O2 will lead to an increased CaO2 in the presence of a shunt. However, further inspection of Equation 4 reveals it is neither the absolute value of cardiac output or O2 but the O2/Qt ratio that is important in influencing arterial oxygenation in the presence of a shunt. Therefore, although administration of inhaled anesthetics decrease O2 (14), they may also decrease cardiac output and negate the beneficial effect. In this regard, it has come to our attention that balanced anesthesia (small-dose to moderate-dose opioid combined with small-dose volatile anesthetic) strikingly reduces O2 while maintaining cardiac output during thoracic surgery (21). This technique decreases the O2/Qt ratio, thereby benefiting arterial oxygenation during OLA.
Szegedi et al. (1) studied the effects of hemodilution on arterial oxygenation during OLA in the presence of an unchanged cardiac output. Hb concentration, a primary determinant of CcO2, is an important variable in Equation 4 and Figure 1 predicts that hemodilution, per se, should lead to decreased arterial oxygenation. Unfortunately, mixed venous blood was not sampled and therefore, it was not possible to calculate Qs/Qt or O2. Therefore, it is difficult to explain why, after acute hemodilution, only the COAD group experienced a decrease in PaO2. We can however speculate that shunt fraction was greater in the COAD group for the following reasons:
Another hypothesis is that the process of acute hemodilution (blood withdrawal and administration of room temperature fluids) may have decreased body temperatures and O2. Figure 1 illustrates that a decrease in O2 will shift the relationship between cardiac output and CaO2 upwards, thereby increasing CaO2. We speculate that in the presence of an unchanged cardiac output, decreases in O2 could have counteracted the tendency to decreased CaO2 after hemodilution in the normal group. However, in the COAD group, the influence of this decrease in O2 was possibly insufficient to prevent the decreases in arterial oxygenation that occurred because of that group’s larger shunt fractions.
The Szegedi et al. study investigated how Hb concentration influences arterial oxygenation in the presence of the shunt that occurs during OLA. It cannot, however, satisfactorily answer the question, as it is not possible to accurately determine certain variables that affect arterial oxygenation during OLA from the aforementioned study. Nonetheless, it generates hypotheses consistent with the background theory that acute decreases in Hb concentration will deleteriously affect arterial oxygenation during OLA if cardiac output, O2, and shunt fraction are kept constant.
The amount of blood shunted through the nonventilated lung is an important and widely quoted variable determining arterial oxygenation during OLA (6,12) Nonetheless, Equation 2 indicates that CvO2 is also an important variable affecting arterial oxygenation in the presence of a shunt. Investigations comparing the effect of various maneuvers on arterial oxygenation during OLA should report both shunt fraction and mixed venous oxygenation (including all the variables that influence it, namely the O2/Qt ratio and Hb concentration). Furthermore, anesthesiologists should consider manipulating factors affecting CvO2 to ensure adequate arterial oxygenation in the presence of right to left pulmonary shunting.
 |
References |
|---|
 Top References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
