Support of mean arterial pressure during tepid cardiopulmonary bypass: Effects of phenylephrine and pump flow on systemic oxygen supply and demand

Support of Mean Arterial Pressure During Tepid Cardiopulmonary Bypass: Effects of Phenylephrine and Pump Flow on Systemic Oxygen Supply and Demand Walter PI6chl, MD, Thomas A. Orszulak, MD, David J. Cook, MD, Rajbir S. Sarpal, MD, and David L. Dickerman, MD Objective: To examine the effects of phenylephrine infusion and increases in pump flow on systemic oxygen supply and demand when they are used to support mean arterial pressure (MAP) during cardiopulmonary bypass (CPB). Design: Prospective, unblinded study. Setting: The animal cardiopulmonary laboratory at the Mayo Foundation (Rochester, MN). Participants: Twelve pigs. Interventions: Twelve pigs had systemic oxygen delivery (DO2) and consumption (VO2) measured before CPB and then underwent CPB at 35°C. During CPB, measurements of DO2 and 402 were obtained at an MAP of approximately 50 mmHg and a pump flow of 2.2 L/min/m 2, Thereafter, MAP was elevated to 70 mmHg either by increases in pump flow or by a phenylephrine infusion, and the balance between systemic oxygen supply and demand was reassessed, Measurements and Main Results: Before CPB, DO2 was 375 _+ 83 mL/min/m 2 and decreased with the onset of CPB

mainly because of the effects of hemodilution. During CPB, with a pump flow of 2.2 L/min/m 2 and an MAP of 53 mmHg, 1~O2was 218 +_ 40 mL/min/m z. Increasing perfusion pressure to an MAP of 72 mmHg with phenylephrine and maintaining pump flow constant (2.2 L/min/m 2) did not change DO2 (222 __ 37 mL/min/m2), and the oxygen extraction ratio (OER) was increased relative to pre-CPB levels. In contrast, increasing MAP to 71 mmHg by increasing pump flow to 3.2 L/min/m 2 resulted in a significantly greater DO2, and the OER normalized to the pre-CPB value. Conclusions: During CPB with conventional flow rates, I)O2 is decreased. Supporting MAP with increases in pump flow better maintains DO2 than the administration of an ~-agonist.

EMODILUTION is standard practice during cardiopulmonary bypass (CPB), and its beneficial effects are well recognized. CPB hemodilution reduces the demand for blood therapy and may reduce the incidence of transfusion-related complications. Depending on the priming volume of the extracorporeal circuit and patient body size, there may be a 20% to 40% reduction in hematocrit with the onset of CPB. This sudden decrease in hematocrit decreases arterial oxygen content and blood viscosity; therefore, decreases in systemic vascular resistance and mean arterial pressure (MAP) occur. In the intact circulation, MAP and systemic oxygen delivery (Do2) are supported during hemodilution by increases in cardiac output, such that 1902 is supported over a broad range of hematocrit values3 ,2 Pump flows (Q) on CPB do not approximate the cardiac outputs associated with hemodilution in the intact circulation; therefore, MAP may decrease substantially. 3 During CPB, if pump flow only approximates the pre-CPB cardiac index (which it often does not), DO2 will decrease secondary to a reduction in arterial oxygen content (CaO2) with hemodilution (I)O2 = Q . CaO2). Therefore, I)O2 during CPB is typically less than that under non-CPB conditions. If systemic oxygen consumption (~'O:) is stable, an increase in the oxygen extraction ratio (OER) is required to compensate for the reduced delivery. The combination of hemodilution and pump flows of 2.2 to 2.5 L/min/m 2 narrows the margin between oxygen supply and demand. Low perfusion pressures are particularly common at warmer CPB temperatures because hypothermia does not offset the decrease in systemic vascular resistance associated with hemodilution. Therefore, vasoconstrictors (typically phenylephrine) are used to elevate MAP during warm or tepid CPB.4,5 Whereas MAP is readily increased using an e~-agonist, this pharmacologic intervention is not physiologic. For a given M A E a better maintenance of DO2 is predictably obtained with an increase in pump flow than with peripheral vasoconstriction. Additionally, the use of e~-agonists during CPB may decrease splanchnic 6 or

renal blood flow. 7 Therefore, it is probably more rational to increase pump flow during periods of CPB hypotension than to use a vasoconstrictor. The purpose of this study was to examine the differential effects of the c~-agonist, phenylephrine, and increases in pump flow on I)O2 and X)O2 during tepid CPB.

H

Journal of Cardiothoracic and Vascular Anesthesia,

Copyright © 1999 by WB. Saunders Company KEY WORDS: cardiopulmonary bypass, pump flow, oxygen delivery, phenylephrine, oxygen demand

METHODS After review and approval by the Institutional Animal Care and Use Committee, 12 unmedicated and fasting juvenile pigs weighing 18 to 29 kg (mean, 22 _+ 3 kg) were studied. Pigs were premedicated with Telazol (Fort Dodge Labs, Fort Dodge, IA) (4 mg/kg) and xylazine (2 mg/kg) intramuscularly. General anesthesia was induced using halothane, 2%, by mask, and the trachea was intubated. Peripheral intravenous access was then secured and muscle relaxation was obtained with pancuronium, 0.1 mg/kg intravenously. Ventilation was controlled to maintain PaCO2 at 35 to 40 mmHg and PaO2 greater than 150 mmHg. Anesthesia was maintained with halothane and continuous infusion of fentanyl, 0.7 gg/kg/miu, and ketamine, 28 gg/kg/min. For muscle relaxation, pancuronium, 0.3 gg/kg/miu, was continuously administered. Cannulae were surgically inserted into a femoral artery for MAP measurements and blood sampling. A 7.5F introducer was placed in a femoral vein and a pediatric pulmonary artery catheter (4F Swan-Ganz; Baxter, Santa Ana, CA) was placed into the pulmonary artery. Before CPB, cardiac output measurements were obtained in triplicate, and

From the Department of Anesthesiology and Division of Cardiovascufar and Thoracic Surgery, and the Department of Surgery, Mayo Clinic, Rochester, MN. Supported in part by American Heart Association--Minnesota Affffiate and the Mayo Foundation. W.P. is the recipient of an ErwinSehrOdinger research fellowship from the Austrian Science Foundation, Vienna, Austria. Address reprint requests to David J. Cook, MD, Department of Anesthesiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. Copyright © 1999 by W.B. Saunders Company 1053-0770/99/1304-0012510.00/0

Vo113,No 4 (August), 1999:pp 441-445

441

442

PLOCHL ET AL

femoral and pulmonary arterial blood gases were obtained. Body temperature was measured with a nasopharyngeal thermocouple. For CPB, a left thoracotomy was performed. The CPB machine was primed with one unit of pig blood (approximately 250 mL) and 750 mL of 6% dextran 70. Venous drainage to the extracorporeal circuit was by a 36F cannula placed in the right atrium through the right atrial appendage. The blood was circulated by a centrifugal pump through a membrane oxygenator with integrated heat exchanger (Bentley Spiral Gold, Irvine, CA) and returned through a cannula (4.5-mm internal diameter) into the root of the aorta. A 40-grn filter was incorporated into the arterial inflow catheter. After completing the surgical preparation and obtaining pre-CPB measurements, CPB was initiated with a flow of 2.2 to 2.4 L/mirdm2, and body temperature was adjusted to 34.5°C to 35.5°C using the CPB heat exchanger. Hahithane was administered through an inline vaporizer on the gas inflow circuit. Arterial hemoglobin concentration and blood gas data were continuously monitored by inline detectors (CDI 100 and CD1400; Cardiovascular Devices, Inc, Tastin, CA). Alpha-slat management was used during CPB. Baseline CPB measurements were obtained with an MAP of 50 to 55 mmHg and a pump flow of approximately 2.2 L/minim 2. Thereafter, MAP was increased to 70 mmI-Ig by either increasing CPB pump flow or by fixing the flow at 2.2 L/min/m2 and adding a phenylephrine infusion. The phenylephrine infusion was titrated to the target MAP. Each animal underwent both peffusion conditions. Six animals received phenylephrine treatment first and flow adjustment second. The other six animals received flow adjustment first and phenylephrine treatment second. Under each of the three CPB conditions, physiologic measurements were obtained when steady-state conditions were reached; the time between measurements was approximately 20 minutes. Arterial blood was drawn from the femoral catheter and mixed venous blood was drawn from the venous return catheter just proximal to the CPB reservoir. Arterial and venous blood gases were measured by an IL 1306 pI-t/blood gas analyzer (Instrumentation Laboratory, Lexington, MA), hemoglobin concentration and oxygen saturation by an IL 482 COOximeter (Instrumentation Laboratory) using a coefficient for pig blood integrated into the software of the analyzer. In the blood gas analyzer, sample temperature was brought up to 37°C so reported blood gas values reflected normothermia. Arterial lactate level was determined by a YSI Model 23A analyzer (Yellow Springs Instrumentation, Yellow Springs, OH). I)O2 and ~-O2 were calculated according to standard formulae. Arterial or venous oxygen content:

Table 1. Biochemical Data (Arterial Blood) CPB 50

Hemoglobin (g/dL)

CPB 70 (MAP = 70)

Pre-CPB

(MAP = 50)

Phenylephrine

High 6_

10.5 _+ 0.5

7.3 ± 0.8*

7.2 ± 0.8*

7.2 ± 0.7"

PaO2 (mmHg) 536 ± 60 240 -+ 51" 237 _+ 76* 227 _+ 59* PaCO2(mmHg) 36±8 37_+2 37±3 37±4 pH 7.48 _+ 0.10 7.37 + 0.06" 7.38 ± 0.06* 7.38 ± 0.06* Lactate (IJmol/mL)

2.0 _+ 0.9

4.0 -+ 1.1"

4.3 _+ 1.2"

4.3 ± 1.1"

NOTE. Values expressed as mean ± SD (n = 12). * p < 0.05 versus pre-CPB, The three CPB periods did not differ f r o m each other for any physiologic variable.

assessed using a two-sample t-test. Repeated-measures analysis of variance followed by Bonferroni's correction was used to compare (~, 202, ~'O2, and other physiologic values between the pre-CPB period and three experimental CPB periods. A p less than 0.05 is considered significant. All data are reported as mean _+ SD.

RESULTS

Before CPB, M A P was 72 -+ 16 m m H g , cardiac index was 2.5 -+ 0.5 L / m i n / m 2, and h e m o g l o b i n level was 10.5 ± 0.5 g/dL. This resulted in a I302 o f 375 + 83 m L / m i r d m 2. "QO2 was 116 41 m L / m i n / m 2, O E R was 0.31, and m i x e d venous oxygen saturation was 72% _+ 10% (Tables 1 and 2). CPB was initiated and body temperature was allowed to drift to 35°C. In the initial CPB period, a flow o f 2.2 _+ 0.4 L/mirdm 2 was maintained; CPB hemodilution resulted in a h e m o g l o b i n level o f 7.3 g/dL (Table 1). U n d e r these conditions, the M A P was 53 - 5 m m H g and 2 0 2 and "QO2 were 218 ± 40 and 87 + 14 m L / m i n / m 2, respectively (Fig 1). U n d e r this condition, I)O2 was significantly reduced relative to the pre-CPB period (Table 2). At 35°C, the mean 9 0 2 was not statistically different from the pre-CPB value but approximated the predicted change based on the reduction in body temperature. With the reduction in 2 0 2 during CPB, the mixed venous oxygen saturation decreased to 58% ± 12% and the O E R increased to 0.42 ± 0.12. Both were significantly different from the pre-CPB values (Table 2).

CxO2 = 1.34. Hb(SxO2) + 0.003 (PxO2) where Hb = hemoglobin concentration; SxO 2 oxygen saturation; and PxO2 = partial pressure of oxygen; x = arterial or venous. Systemic oxygen delivery (I)Oz):

Table 2. Systemic Physiologic Values

=

1302 = 0 • CaO2 where (~ is pump flow. Systemic oxygen consumption ('QO2): "QO2 = Q. (Ca02 - Cv02) Oxygen extraction ratio (OER): OER = (CaO2 - CvO2)/CaOz Six animals received phenylephrine as the first treatment to increase MAP and increased pump flow as the second treatment. The other six animals underwent the reverse order. The effect of treatment order was

Body temperature(°C) MAP (mmHg) Flow (L/min/m 2)

Pre-CPB

CPB 50 (MAP = 50)

37.0 _+ 0.9 72 _+ 16 2.5 _+ 0.5

35.0-+ 0.3* 53 ± 5* 2.2 _+ 0,4

DO2 (mUmin/m 2) 375 _+ 83 218 402 (mL/min/m 2) 116 _+ 41 87 S~/O2 (%) 72 _+ 10 58 OER 0.31 _+ 0.09 0.42

CPB 70 (MAP = 70) Phenylephrine

34.9 + 0.3* 72 _+ 7 f 2.2 -- 0.3

-+ 40* 222 ± 14 99 -+ 12" 55 ± 0.12" 0.45

High

34.9 _+ 0.3* 71 _+ 71, 3.2 ± 0.5"1

-+ 37* 318 ± 14 91 _+ 9* 70 ± 0.08* 0.30

_+ 64"1, _+ 15 ± 81" _+ 0,071,

Values expressed as mean _+ SD (n = 12). Abbreviations: MAP, mean arterial pressure; (~, pump f l o w ; DO2, systemic oxygen delivery; ~/O2, systemic oxygen consumption; Sk/O2, mixed venous oxygen saturation; OER, oxygen extraction ratio. * p < 0.05 v pre-CPB. l - p < 0.05 vinitial CPB period (MAP = 50 mmHg).

MANAGEMENT OF 02 DELIVERY DURING BYPASS

400

443

T

300 -

0 2O0

100

0

Phenylephrine

Pre-CPB

CPB 50

High (~

CPB 70

Fig 1. Values for systemic oxygen consumption (VOz} and systemic oxygen delivery (I)Oz) are expressed as mean _+ SD in mL/ minim z (n = 12). CPB 50 (MAP = 53 - 5 mmHg, (~ = 2.2 _+ 0.4 L/mini ma); CPB 70; Phenylephrine (MAP, 72 -+ 7 mmHg by phenylephrine infusion, (~ = 2.2 -+ 0.3 L/minima); CPB 70, high (~ (MAP = 71 -+ 7 mmHg by increases in (~ to 3.2 -+ 0.5 L/min/mZ). II, VO2; [~, DOn.

Lactate levels at a flow of 2.2 L/min/m2 and an MAP of 53 mmHg were also elevated relative to before CPB (Table 1). After measurements were obtained with a mean flow of 2.2 L/min/m2 and an MAP of 53 mmHg, the MAP was increased to approximately 70 mmHg with either increases in pump flow or by phenylephrine infusion. The order of the interventionused to increase perfusion pressure (flow first or phenylephrine first) had no primary effect on 1)O2, "QO2, or any of the physiologic values. Therefore, the data of all 12 animals were pooled. Throughout the three CPB study periods, temperature, PaO2, PaCO2, pH, and hemoglobin concentrations were stable and did not differ (Table 1). With phenylephrine treatment, MAP increased from 53 +- 5 to 72 +- 7 mmHg. Bypass flow (2.2 L/mirdm2) was unchanged during this intervention (Tables 1 and 2). This increase in perfusion pressure was not associated with an increase in 1)O2 or "VO2relative to measurements obtained during CPB with an MAP of 53 mmHg. Relative to the pre-CPB period, 1)O2 remained reduced, as was mixed venous oxygen saturation. The OER also remained elevated relative to the pre-CPB period. Conversely, increasing MAP by increasing CPB flow normalized systemic oxygen balance. With a pump flow of 3.2 + 0.5 L/mirdm2, MAP increased to 71 +_ 7 mmHg: With this flow increase, "QO2remained unchanged, whereas DO2 increased to 318 -+ 64 mL/min/m2. Mixed venous oxygen saturation increased to 70% +- 8%, and the OER decreased to 0.30 + 0.07. Under the high-flow condition, each of these values differed from the other two CPB conditions (Table 2). Additionally, under this flow condition, MAP, mixed venous oxygen saturation, and OER were the same as before CPB. However, lactate levels remained elevated relative to pre-CPB (Table 1).

DISCUSSION During CPB, the determinants of whole-body oxyge.n balance are the same as those under nonbypass conditions. VO2 is

primarily determined by temperature, and D Q remains a function of cardiac output (pump flow) and arterial oxygen content. It is increasingly popular to allow body temperature to drift during CPB rather than actively cool. A body temperature of 33°C to 34°C is common in clinical practice, and this corresponds to the CPB temperature of 35°C in this study (normal pig temperature is approximately 2°C warmer than humans). Similarly, the pigs underwent a reduction in hematocrit of approximately 30% with CPB hemodilution, and MAP decreased to 53 mmHg with a pump flow of 2.2 L/min/m2 (80 mL/min/kg). This closely reflects what is seen during clinical CPB. However, the authors have shown that under this perfusion condition, 1)O2 is reduced relative to the non-CPB state and oxygen extraction must increase to maintain "VO2. This is true even with the reduction in temperature. I)O2 decreases because the reduction in CaO2 with CPB hemodilution is not compensated for by the increases in flow that would occur in the intact, non-CPB circulation. The reduction in hematoerit during CPB causes hypotension. Hematocrit is a fundamental determinant of blood viscosity; therefore with hemodilution, systemic vascular resistance decreases and MAP is reduced if pump flow is not increased. During hypothermic CPB, hypotension is less common because hypothermia offsets the reduction in systemic vascular resistance seen with hemodilution. However, with the trend to higher CPB temperatures, hypotension is frequent and the typical response is to support MAP with an c~-agonist, usually phenylephrine.4,5 Whereas other studies showed that use of an a-agonist to support MAP is acceptable for the central nervous system, cerebral perfusion may be preserved at the expense of other organ beds. The authors, 7,8 as well as others, 9-11have shown that cerebral perfusion is independent of CPB flow as long as perfusion pressure is maintained. Therefore, the use of phenylephrine to maintain cerebral perfusion is appropriate; however, if total CPB flow is reduced, maintenance of cerebral blood flow is achieved by the shunting of flow from other vascular beds. Renal blood flow7 and visceral perfusion more broadly6 may be compromised when phenylephrine infusion is used to maintain MAP during CPB. These peffusion conditions may be an important contributor to post-CPB dysfunction of splanchnic organs. In the pre-CPB period, the cardiac index of a typical anesthetized cardiac surgical patient is 2.3 to 2.7 L/min/m2, and their hemoglobin concentration is approximately 12 g/dE Under conditions of normoxia, this results in a 1)O2 of approximately 350 to 450 mL/mirdm2. During CPB, pump flows of 2.2 to 2.4 L/mirdmz are most common under conditions when the hemoglobin level is 7 to 8 g/dL. This results in a 1)O2 between 200 and 300 mL/mirdmL (These values are the same reported in the study animals.) Thus, for most patients, the 1)O2 during CPB is significantly less than their baseline non-CPB 1)O2. This results in greater systemic oxygen extraction and is reflected in a decreased mixed venous oxygen saturation. Under normal conditions, systemic oxygen delivery exceeds oxygen demand, and reductions in DO2 can be tolerated through increases in oxygen extraction. This is seen during CPB.

444

PLOCHL ET AL

However, there is a limit to this physiologic response, and a critical oxygen delivery may be reached for the whole body or, more likely, for the individual organs at which tissue ischemia results. Because CPB is associated with man.y periods in which either v~rO2 is increasing (rewarming) or DO2 is decreasing (transient reductions in pump flow or decreasing hematocrit), it may be more rational to support perfusion pressure with increases in flow rather than a vasoconstrictor• This report clearly shows that under normal tepid CPB conditions, wholebody oxygen delivery is decreased and the margin between oxygen supply and demand is significantly reduced and is not improved by the use of vasoconstrictors. Conversely, greater pump flows better maintain DO2 and normalize the OER. This study might be criticized for several reasons. First, the results are predictable. Determination of I)O2 and ~rO 2 during tepid CPB does not produce unique physiologic insights; the values reported are what would be expected if these values were calculated during clinical CPB. However, these observations are of specific practical relevance with the trend to warmer CPB temperatures. Therefore, although the results are not surprising, they are worth paying attention to. Second, it might be suggested that the relevance of a pig model is limited, but the authors disagree. The determinants of whole-body oxygen delivery and consumption are no different in humans and pigs, and the CPB management in this study very closely mimics that used in clinical practice. Findings in a healthy pig are probably less dramatic than what might be expected in today's surgical population. The authors would expect young healthy animals without diabetes or vascular disease to be far more tolerant of these perfusion conditions than today's older surgical patients. Therefore, these observations in normal animals may be even more relevant to clinical practice• This study could also have been more sophisticated. It did not measure perfusion or oxygen balance in individual organ beds. Whereas flow and oxygenation to such critical organ beds as the brain are preserved at the expense of noncritical organs, it is not essential to make these measurements for this study to have relevance. These measurements have already been made for brain s~° and, to some extent, kidney and other visceral organs. 6,7,9 It has been shown that these organ beds may be compromised under CPB conditions that are conmaon. Regardless of how much individual organs are challenged, it is probably reasonable to maintain whole-body oxygen delivery as close to normal as possible. An argument by analogy can also be made. If an intact animal experiences an acute reduction in

hematocrit, whole-body oxygen delivery is supported by increases in cardiac output; increases in oxygen extraction occur secondarily, and only under extreme conditions will vasoconstriction occur and organ blood flow be compromised. During CPB, it should be the goal to mimic the intact circulation to the extent possible. There are also limitations in method that are faced by most studies of this type. As in this study, the determination of VO2 and 1~O2 may be linked mathematically as well as physiologically. This is of particular importance when correlations are performed, and this potential problem can be minimized by the determination of linked or coupled variables using independent techniques.22 This laboratory, like most others reporting similar studies, is unable to provide calorimetric studies or direct measurement of systemic oxygen consumption and so is compelled to rely on the Fick method. However, the authors realize the inherent limitation of the technique. An additional minor criticism is that increasing pump flow did not appear to improve lactate levels. The authors believe that the elevated lactate levels seen during high-flow CPB are a function of incomplete clearance of the lactate generated during the preceding lower flow CPB state. Given the normalization of DOe, mixed venous oxygen saturation, and OER, it would be predicted that, over time, the lactate levels would have returned to pre-CPB levels if the study had been extended. However, it should also be noted that lactate has been reported to increase during CPB despite maintained DO2 .13 Those investigators suggest this is a manifestation of a systemic inflammatory response to CPB and not inadequate perfusion. 13 Finally, it should be noted that to maintain adequate circuit volume, the use of high pump flows will be associated with greater fluid requirements and could increase the risk for hemolysis and platelet damage.14,15 In summary, this study documents systemic oxygen balance in pigs during tepid CPB. It was found that with conventional hemodilution and pump flow, 1302 is reduced and the margin between supply and demand is significantly narrowed. The authors found that compensating for hemodihition-associated hypotension with an c~-agonist improved perfusion pressure, but systemic oxygenation remained compromised. Conversely, supp.ort of perfusion pressure with increases in flow will normalize DO2 to prebypass conditions. This finding may lead to reconsideration of the routine use of vasoconstrictors during CPB. The authors urge frequent reevaluation of these physiologic variables during conditions of pressure, flow, hematocrit, and temperature change during CPB.

REFERENCES

1. Cain SM: Oxygen delivery and uptake in dogs during anemic and hypoxic hypoxia. J Appl Physio142:228-234, 1977 2. Messmer K: Hemodilution.Surg Clin NorthAm 55:659-678, 1975 3. Liam BL, Plochl W, Cook DJ, et al: Hemodilution and wholebody oxygen balance during normothermic cardiopulmonary bypass in dogs. J Thorac Cardiovasc Surg 115:1203-1208, 1998 4. ChristakisGT, Koch JP, Deemar KA, et al: A randomized study of the systemic effects of warm heart surgery.Ann Thorac Surg 54:449-459, 1992 5. Lehot JJ, Villard J, Piriz H, et al: Hemodynamic and hormonal responses to hypothermic and normothermic cardiopulmonary bypass. J Cardiothorac Vasc Anesth 6:132-139, 1992 6. O'Dwyer C, Woodson LC, Conroy BE et al: Regional perfusion

abnormalities with phenylephrine during normothermic bypass. Ann Thorac Surg 63:728-735, 1997 7. Cook DJ, Orszulak TA, Daly RC: The effects of pulsatile cardiopulrnonary bypass on cerebral and renal blood flow in dogs. J Cardiothorac Vasc Anesth 11:420-427, 1997 8. Cook DJ, Proper JA, Orszulak TA, et al: Effect of pump flow rate on cerebral blood flow during hypothermic cardiopnlmonary bypass in adults. J Cardiothorac Vasc Anesth 11:415-419, 1997 9. Fox LS, Blackstone EH, Kirklin JW, et al: Relationship of brain blood flow and oxygen consumption to perfusion flow rate during profoundly hypothermic cardiopulmonary bypass. An experimental study. J Thorac Cardiovasc Surg 87:658-664, 1984

MANAGEMENT OF 02 DELIVERY DURING BYPASS

10. Schwartz AE, Sandhu AA, Kaplon RJ, et al: Cerebral blood flow is determined by arterial pressure and not cardiopulmonary bypass flow rate. Ann Thorac Surg 60:165-169, 1995 11. Rogers AT, Prough DS, Roy RC, et al: Cerebrovascular and cerebral metabolic effects of alterations in perfusion flow rate during hypothennic cardiopulmonary bypass in man. J Thorac Cardiovasc Surg 103:363-368, 1992 12. Archie JPJ: Mathematic coupling of data. A common source of error. Ann Surg 193:296-303, 1981

445

13. Haisjackl M, Birnbanm J, Redlin M, et al: Splanchnic oxygen transport and lactate metabolism during normothermic cardiopulmonary bypass in humans. Anesth Analg 86:22-27, 1998 14. Nevaril CG, Lynch EC, Alfrey CR et al: Erythrocyte damage and destruction induced by shearing stress. J Lab Clin Med 71:784-790, 1968 15. Brown CH III, Lemuth RF, Hellums JD, et al: Response of human platelets to shear stress. Trans _Am Soc Artif Intern Organs 21:35-38, 1975