AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 288: H1071-H1079, 2005. First published October 28, 2004; doi:10.1152/ajpheart.00884.2004
0363-6135/05 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
288/3/H1071    most recent
00884.2004v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Torres Filho, I. P.
Right arrow Articles by Ward, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Torres Filho, I. P.
Right arrow Articles by Ward, K. R.

Experimental analysis of critical oxygen delivery

Ivo P. Torres Filho,1,2,3 Bruce D. Spiess,1,2 Roland N. Pittman,2,3 R. Wayne Barbee,2,3 and Kevin R. Ward2,3

Departments of 1Anesthesiology, 2Emergency Medicine, and 3Physiology, Virginia Commonwealth University Reanimation Engineering Shock Center, Virginia Commonwealth University, Richmond, Virginia 23298-0695

Submitted 26 August 2004 ; accepted in final form 22 October 2004


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Systemic variables were evaluated with respect to O2 delivery to test the hypothesis that critical O2 delivery and critical Hb can be estimated by multiple variables collected simultaneously. Rats were subjected to transfusion with either fresh or stored blood and then subjected to stepwise isovolemic hemodilution. Critical levels were measured by the dual-regression method from plots of systemic variables against O2 delivery and Hb. Delivery was calculated from cardiac index and arterial O2 content. We found that 1) after hemodilution, O2 delivery changed in a nonlinear relationship with Hb; 2) critical delivery calculated using 30 different systemic variables was not statistically different from each other; 3) critical delivery and critical Hb were correlated but were not different between animals receiving fresh or stored blood; and 4) similar critical levels were found using a single variable from several animals and using several variables from the same subject. The best variables to estimate critical delivery were lactate, bicarbonate, base excess, O2 extraction ratio, expired CO2, pulse pressure, cardiac index, and systolic pressure. The data suggest that a multivariable analysis of critical delivery may help determine the physiological oxygenation boundary at the whole body level. This may assist in finding therapeutic triggers on an individual basis using systemic markers of the transition from aerobic to anaerobic metabolism.

hemodilution; transfusion; oxygen delivery and consumption

SYSTEMIC O2 DELIVERY (DO2) is defined as the product of cardiac output (CO) and arterial O2 content. At normal or high levels of DO2, O2 uptake is independent of DO2 because changes in DO2 result in reciprocal changes in tissue O2 extraction ratio. However, tissues exhibit a level of DO2, known as critical DO2 (DO2 crit), below which the extraction cannot increase enough to sustain O2 uptake, and O2 consumption (O2) becomes supply (or DO2) dependent (6, 26, 31, 33). Therefore, a O2-DO2 plot reveals that O2 is relatively constant (supply independent) at high levels of DO2 but falls off in a supply-dependent phase when DO2 is below DO2 crit.

Systemic DO2 crit has been considered the ultimate physiological threshold to the manifestation of tissue hypoxia and shock (32). Because DO2 dependency has been associated with organ damage and poor outcome in critically ill patients, some investigators recommended aggressive efforts to increase DO2 to replenish tissue O2 and prevent organ dysfunction (16, 34, 35). However, pathological DO2 dependency is not detectable in many patients (3), increased DO2 is not always beneficial (9, 16, 26), and pathological dependency remains a controversial topic (30).

The detection of the supply-dependent phase may depend on the ability to determine DO2 crit (5, 33), and most previous attempts to analyze DO2 crit focused on O2-DO2 relationships (5, 10, 26). DO2 crit does not seem to be affected by the method used to decrease DO2 (4) and is an important marker of the transition to anaerobic metabolism (26, 31, 32). Although different organs and tissues possess unique O2-DO2 relationships (5–7, 10, 29), the passage from aerobic to anaerobic metabolism changes multiple whole body physiological variables. Unfortunately, there are no studies presenting a detailed evaluation of systemic variables as a function of DO2. We hypothesized that, if changes in these variables would follow changes in DO2 in a predictable fashion, they could be used to detect the departure from aerobic metabolism and to estimate DO2 crit. This approach could be useful in settings with limited access to technologies for O2 monitoring (6).

The present investigation was designed to verify if changes in systemic variables would follow changes in DO2 in a predictable fashion and could be used to estimate DO2 crit. This was accomplished by decreasing DO2 and assessing the ensuing changes in systemic variables.

We tested the hypothesis that DO2 crit can be estimated by multiple variables. The DO2 crit estimation was tested by processing data in the following two ways: 1) data from all animals were averaged to provide a mean DO2 crit for each variable and 2) data from all variables were averaged to provide a mean DO2 crit for each animal. To further test the relationship between each variable and DO2, animals were subjected to transfusion with either fresh or stored blood. Because Hb concentration is easier to determine than DO2 and Hb levels are used to trigger more aggressive procedures, such as transfusion (1, 44), we also investigated the systemic changes as a function of Hb.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animal instrumentation and measurements. This study was approved in advance by the Institutional Animal Care and Use Committee of Virginia Commonwealth University Health System and conforms to the Public Health Service Policy on Human Care and Use of Laboratory Animals (August, 2002) and the American Physiological Society's "Guiding Principles in the Care and Use of Animals." Twenty-two male Sprague-Dawley rats (Harlan, Indianapolis, IN) weighing 430 ± 6 g (mean ± SE) were housed (12:12-h light-dark cycle, constant temperature and humidity) for a 1-wk adaptation period with free access to standard rat chow and water.

Hemodynamic measurements. The animals were instrumented under anesthesia with a mixture of ketamine (70 mg/kg ip; Fort Dodge Animal Health, Fort Dodge, IA) and acepromazine (3 mg/kg ip; Vedco, St. Joseph, MO), followed by an intravenous infusion (0.24–0.36 mg·kg–1·min–1) of alfaxalone-alfadolone acetate (Saffan; Schering-Plough Animal Health, Welwyn Garden City, UK). The left femoral artery was connected to a disposable pressure transducer to continuously measure arterial blood pressure. The right jugular vein was cannulated with PE-90 tubing advanced to the entrance of the right atrium. This line was used to collect central venous blood samples and for recording central venous blood pressure. The right femoral artery and vein were connected to a syringe pump (model PHD2000; Harvard Apparatus, Holliston, MA). The right femoral artery was used for blood exchange and blood sampling. The core temperature was maintained at 36.5–37.0°C using a blanket (Harvard Apparatus).

Rats were ventilated by a pressure-controlled ventilator (Kent Scientific, Torrington, CT) using a Y-shaped tube and an inspired fraction of O2 (FIO2) of 0.95. This FIO2 is commonly used in surgical settings. To monitor minute ventilation (E) volume, the expiratory limb of the Y tube was connected to a pneumotach attached to a differential pressure amplifier (Biopac Systems, Goleta, CA). Calibrated pumps continuously sampled mixed expired gas, routing it to O2 and CO2 analyzers (O2100C and CO2100C modules; Biopac Systems) by means of water-permeable Nafion tubing. After placement in the ventilator and for the remainder of the experiment, the animals received a continuous intravenous infusion of pancuronium bromide and Hespan (for hydration). A median sternotomy was performed, and a transit-time ultrasonic flow probe (model 2.5SB; Transonic Systems, Ithaca, NY) was positioned around the ascending aorta. The probe was connected to a flowmeter (model T206; Transonic Systems) for continuous recording of aortic blood flow. This technique is stable and reproducible and recently has been used to estimate DO2 crit in rats subjected to prolonged hemorrhagic hypotension (39). The method has been fully validated under various experimental conditions and by comparison with other techniques (43).

Blood-gas, hematological, and biochemical measurements. Blood analyses were performed in arterial and venous samples (0.1 ml each) collected using heparinized glass capillaries. Blood gases and chemistry were measured with a blood gas analyzer (ABL 725; Radiometer, Copenhagen, Denmark). Total Hb concentration, methemoglobin, and Hb O2 saturation were measured with a multiwavelength CO oximeter adjusted for the rat's Hb absorption spectra (OSM3; Radiometer). Hematocrit was determined from an additional blood sample (0.06 ml) withdrawn in a heparinized microtube.

Solutions. The following solutions were used: 1) Hespan was used as hydration fluid and for isovolemic hemodilutions and 2) fresh and aged blood; blood was harvested from a separate group of Sprague-Dawley rats (400–500 g body wt) maintained under isoflurane anesthesia. Under sterile conditions, blood was withdrawn into 20-ml sterile syringes containing 2 ml citrate-phosphate-dextrose solution with adenine (CPD-A; Sigma Chemical, St. Louis, MO). Once collected, the blood was immediately centrifuged and stored as packed cells (60–70% hematocrit) in neonatal unit storage bags at 4°C. Blood was maintained for the appropriate length of time (<24 h for fresh blood and 10.5 days for stored blood) before reinfusion. Before blood exchange, packed cells were diluted with normal saline so that all animals received blood with the same Hb concentration (10.5 ± 0.5 g/dl). All injected solutions were warmed to room temperature (23 ± 2°C) and administered without further change in temperature. Just before injection, blood analyses were performed in each solution.

Protocol. Animals were heparinized, and baseline measurements were obtained. Rats were randomly allocated to experimental groups in which 50% isovolemic exchange transfusion was performed with fresh blood (n = 14) or stored blood (n = 8). At the end of the baseline period, animals were subjected to the exchange transfusion procedure with one of the test solutions. After the end of the isovolemic exchange (10 min), one set of measurements was performed. To decrease DO2 in a stepwise fashion, animals were subjected to 10–14 isovolemic hemodilutions at rates of 0.5–1 ml/min and dilution volumes of 3–8 ml/kg. Typically, higher rates and exchange volumes were used in the first five to six steps and slowly decreased as lower values of Hb were achieved. Blood was withdrawn from the right femoral artery, and Hespan was infused through the right femoral vein. After completion of each dilution (5 min), a complete set of the measures outlined above was collected. The average time interval between each data collection was 14 min. The dilution protocol was followed until a terminal stage was reached such that CO and arterial pressure were no longer constant from the beginning to the end of the data collection period. Typically, the animals died shortly after this terminal stage. A subgroup (n = 5) of the rats that received fresh blood was used as control animals for the hemodilution procedure. These rats were subjected to all experimental procedures except for the hemodilution. All animals requiring euthanasia received a pentobarbital sodium overdose of 100 mg/kg (iv).

Data acquisition and analysis. Signals from the pressure transducers, aortic flowmeter, pneumotach, and gas analyzers were digitized at a rate of 500 Hz (Acqknowledge 3.7.2 + MP150 hardware and software; Biopac Systems). Systolic, diastolic, pulse, and mean arterial pressures were calculated from the arterial pressure recording. Heart rate (HR) was calculated from the aortic flow signal. CO and mean E were estimated from the mean aortic flow and pneumotach signal, respectively. Mean stroke volume was calculated as CO/HR. Because surface area was not measured in each rat, relative stroke index and relative cardiac index (CI) were computed by dividing the appropriate variables by body mass. As an estimate of cardiac contractility, the first derivative of aortic flow with respect to time was calculated, and its maximum value (dF/dt) was obtained. All off-line calculations were based on 1-min segments of the digitized signals.

Global O2 was calculated both by the Fick equation (O2b) and by direct measurement of air flow and inspired and expired gas concentrations (O2g). O2g was calculated as follows: O2g = E(FIO2 – FEO2), where FEO2 is the fraction of expired O2. O2b was also calculated by the product of CI and the difference between arterial (CaO2) and venous (CvO2) O2 contents:



where aHb O2 Sat and vHb O2 Sat are the arterial and venous Hb O2 saturations, respectively, and PaO2 and PvO2 are the arterial and venous O2 tensions, respectively. Whole animal DO2 was computed as the product of CaO2 and CI. Accordingly, the following two O2 extraction ratios (ER) were computed: O2ERg = O2g/DO2 and O2ERb = O2b/DO2.

Critical value estimation. DO2 crit was determined in each animal from a plot of O2 as a function of DO2 using a dual-line method previously described (27). For each rat, O2 was plotted against DO2, and a series of regression lines were fitted to the delivery-dependent and delivery-independent portions of the O2-DO2 curve using the least-squares method. The best pair of regressions was determined, based on the smallest sum of squared residuals. The DO2 at which these two regression lines intersected indicated the DO2 crit. This methodology was extended to determine DO2 crit from plots of all systemic variables as a function of DO2, and the data were processed in the following two ways: 1) data from all animals were averaged to provide a mean DO2 crit for each variable; and 2) data from all variables were averaged to provide a mean DO2 crit for each animal. Similar analysis was performed for all variables as a function of Hb to obtain estimates of critical Hb (Hbcrit) concentration. A computer program was developed to automate the processing of the {approx}30,000 regression lines required to compute all estimations in all animals. In 83% of the cases, the dual-regression method was "successful" in estimating critical values from DO2 plots: data could be adequately fitted by two crossing regression lines, and an inflection point was found in a given relationship. In 17% of the cases, an inflection point could not be found, and these cases were considered as "unsuccessful" critical determinations. For each variable, a success rate was computed by dividing the number of successful determinations by the total number of tested relationships.

Statistics. Values are reported as means ± SE. Differences between two groups (fresh and stored blood) for the mean DO2 crit of each variable were analyzed by using the Student's t-test. The coefficient of variation (CV, calculated as SD/mean x 100) was used as an index of variability. For correlation analysis, linear least-squares regressions were performed, and significance of the correlation coefficients was tested. The Student's t-tests and calculations of statistical significance and of the CV were performed using commercial computer software [Origin 7 (OriginLab) and Excel 2002 (Microsoft)]. All P values correspond to two-tailed tests with significance set at 0.05.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
The systemic variables from control animals remained relatively stable throughout the tested period of 5 h. Because the animals received Hespan to replace the volume of blood used during sampling, these animals showed a small decrease in blood Hb concentration over time, but O2 delivery and consumption were not statistically different from baseline measurements. The number of hemodilution steps and the total volume of Hespan used were similar for all hemodiluted animals, averaging 13 ± 1 and 63.2 ± 2.2 ml/kg, respectively. Whole body DO2 was relatively constant for Hb values >6 g/dl (Fig. 1). The isovolemic hemodilutions led to progressively lower Hb and correspondingly lower DO2 for all animals, following a nonlinear relationship. The steeper decrease in DO2 at low Hb levels indicates a change in CO at higher hemodilution levels. The decrease in DO2 led to changes in systemic variables (Fig. 2), and all animals responded in a similar fashion, with the main change occurring after DO2 reached 10 ml·min–1·kg–1. The variables were grouped into four categories (hemodynamic, cardiac, respiratory/oxygenation, and blood chemistry) to facilitate analysis and presentation.



View larger version (13K):
[in this window]
[in a new window]
  		Fig. 1. Whole body O2 delivery (DO2) as a function of Hb concentration during hemodilution, based on 200 measurements from animals that previously received fresh or stored blood. On average, each point represents 11 determinations of DO2 at a given range of Hb. Error bars represent SE.

 


View larger version (39K):
[in this window]
[in a new window]
  		Fig. 2. Data from a typical experiment depicting changes in hemodynamic (A), cardiac (B), respiratory/oxygenation (C), and blood biochemical (D) variables as a function of DO2 during isovolemic hemodilution. For clarity, not all studied variables are shown. Data are expressed as a percentage of the initial value except for arterial base excess (aBE) and lactate in D, which are presented as change from initial value (y-axis on far right). In D, the first y-axis (from left to right) refers to arterial pH (apH) and the second y-axis is used for arterial glucose (aGluc), potassium (aK+), bicarbonate (aHCO3–), and O2 tension (PaO2). aGluc, arterial glucose; aLactate, arterial lactate. Note that the curves for all variables change inflection at similar critical DO2 (DO2 crit), indicated by the vertical bars. dF/dt, maximum slope of aortic flow curve; CO2 max, peak expired CO2 concentration; O2b, O2ERb, O2g, and O2ERg, O2 consumption (O2) and O2 extraction ratio (ER) calculated from blood (b) and expired gas (g).

 
Some variables did not present any discernible relationship with DO2 and were not subjected to the dual-regression analysis. These variables included HR, central venous pressure, respiratory flow, and blood Na+, Cl–, and Ca2+ levels. The remaining 30 variables, listed on Table 1, were processed to estimate DO2 crit and Hbcrit. Figure 3 shows the application of the dual-regression method to estimate DO2 crit from systemic data in addition to the traditional use of the methodology in O2-DO2 relationships. Some data could not be fitted by two crossing regression lines, and the dual-regression method was not successful in estimating critical values. The success rate and the mean correlation coefficient (for the DO2-dependent portion) for each variable are presented in Table 1. The overall mean correlation coefficient was 0.946 ± 0.008. With the use of 26 variables, successful estimates of DO2 crit and Hbcrit were obtained in >50% of the cases (at least 9 of 17 animals).


View this table:
[in this window]
[in a new window]
  		Table 1. Systemic variables investigated as a function of DO2 and of Hb

 


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Each panel shows representative experimental data from one animal where DO2 crit was estimated from a systemic variable using the dual-regression method. Each point represents one measurement. The straight lines are the least-squares regression line for the DO2-dependent and DO2-independent portions of the data. In each case, at least 10 pairs of regressions were tested. The DO2 crit (arrows) was estimated from the intersection between the two regression lines, as denoted by dotted lines.

 
Values of DO2 crit estimated from various systemic variables were not statistically different (Figs. 4 and 5). In addition, independent of the variable used, values of DO2 crit for animals receiving fresh and stored blood were not statistically different. The DO2 crit calculated from cardiac variables (CI, stroke index, and dF/dt) showed the largest difference between animals receiving fresh and stored blood, but the difference did not reach statistical significance for any of the variables (0.05 < P < 0.10).



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 4. Estimated DO2 crit from blood biochemical variables. The number of DO2 crit determinations for a given variable is on the right of each bar. Error bars represent SE. Dotted vertical line indicates the overall mean DO2 crit for all displayed variables. Sat, saturation.

 


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5. Estimated DO2 crit from hemodynamic (A), cardiac (B), and respiratory/oxygenation (C) variables. The number of DO2 crit determinations for a given variable is on the right of each bar. Error bars represent SE. Dotted vertical line indicates the overall mean DO2 crit for all variables in A–C.

 
The success rate and the CV for each variable used to estimate DO2 crit are presented in Fig. 6. With the use of the criteria of both high success rate and low CV, the best variables to estimate DO2 crit were arterial lactate, arterial HCO3–, dF/dt, venous lactate, arterial base excess, O2ERg, peak expired CO2, pulse pressure, CI, and systolic pressure (Fig. 6, bottom right). PaCO2 possibly could be included among the best variables, since the combination of its relatively high success rate (88%) and low CV (15%) makes it comparable to the systolic pressure (94% success and 25% CV). The same rationale would apply to stroke index because of its high success rate (100%) but relatively high CV (29%).The worse variables were arterial Hb O2 saturation and PaO2 tension (Fig. 6, top left). The poor performance of these variables is likely because of the fact that arterial blood oxygenation was kept relatively constant by the high (0.95) FIO2. Although O2b showed the smallest variability (10%), this variable could only be used to estimate DO2 crit in 35% of cases.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 6. Success rate and coefficient of variation (CV) in the estimation of DO2 crit using 30 systemic variables, including hemodynamic ({triangledown}), cardiac ({square}), respiratory/oxygenation ({triangleup}), and blood biochemical ({circ}) variables. The positions of the dotted lines are arbitrary and are used only to denote regions of low and high rates and CVs. The prefixes a and v indicate arterial and venous, respectively. Lac, lactate; BE, base excess; Gluc, glucose; Hb O2Sat, Hb O2 saturation; MAP, mean arterial pressure; DP, diastolic pressure; SP, systolic pressure; PP, pulse pressure; CI, cardiac index; SI, stroke index.

 
In each animal, curves from at least 19 variables changed inflection at similar DO2 crit values. On average, DO2 crit could be estimated from 23 variables (range: 19–26) in each animal. The overall average DO2 crit, considering all 30 variables, was 10.7 ± 0.2 ml·min–1·kg–1. Additionally, the average DO2 crit calculated for each animal, using at least 19 different variables, was also 10.7 ± 0.4 ml·min–1·kg–1. The variability of DO2 crit among variables (9%) was smaller (P < 0.05) than the variability of DO2 crit among animals (14%).

The analysis of changes in systemic variables as a function of Hb yielded results very similar to the analysis performed with DO2. The correlation coefficient (for the DO2-dependent portion) for each variable was always high, and no difference was found between animals treated with fresh or stored blood. Similarly to DO2 crit, Hbcrit calculated from cardiac variables showed the largest difference between fresh (2.4 ± 0.1 g/dl) and stored (3.0 ± 0.3 g/dl) blood-treated rats, but the difference did not reach statistical significance for any of the three variables (0.05 < P < 0.10). The overall average Hbcrit, considering all 30 variables, was 2.8 ± 0.1 g/dl (CV = 13%). The average Hbcrit calculated for each animal, using at least 19 different variables, was also 2.8 ± 0.1 g/dl (CV = 14%). Despite the small range of DO2 crit values, a significant linear correlation was found between Hbcrit and DO2 crit calculated for all 30 variables (Fig. 7).



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 7. Estimated DO2 crit and critical Hb (Hbcrit) from all variables. The least-squares regression line is shown. r, Correlation coefficient; n, no. of points. Error bars represent SE.

 
The calculated "critical" values for each systemic variable at DO2 crit are presented in Table 2. For each variable, critical values found for fresh and stored blood-treated animals were not statistically different.


View this table:
[in this window]
[in a new window]
  		Table 2. Systemic variable values at critical DO2

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
To test the hypothesis that DO2 crit can be estimated by multiple variables, a systematic investigation of physiological variables was performed in animals subjected to stepwise hemodilutions until death. The main findings were that 1) DO2 crit calculated using 30 different systemic variables were not statistically different from each other; 2) DO2 crit and Hbcrit were linearly related, but they were not different between animals receiving fresh or stored blood; and 3) similar DO2 crit values were found using a single variable from several animals and using several variables from the same subject.

This study presents the first quantitative evaluation of respiratory, cardiac, hemodynamic, and blood biochemical variables, measured simultaneously, in a significant range of DO2 levels, varying from fully aerobic to near-death anaerobic conditions. Our results are the first to demonstrate that DO2 crit and Hbcrit can be simultaneously estimated by a large number of systemic variables. We used a hemodilution model that was developed and standardized in our laboratory, providing direct and online measurements of multiple systemic variables and the determination of DO2 crit from a large number of consecutive DO2 measurements. In addition to providing quantitative estimations of DO2 crit, the present work presents a methodology to analyze systemic data as a function of DO2 (and/or Hb) to follow systemic markers of the transition from aerobic to anaerobic metabolism.

In an intact physiological system, several interrelated mechanisms act to provide adequate O2 supply to meet different tissue demands. Conversely, a major disturbance in the system, such as drastic decreases in DO2, should affect a number of the interconnected physiological systems, leading to changes in several physiological variables. Our study shows that it is possible to detect the transition to anaerobic metabolism at a whole body level using some traditional but nonconventional measures. Because perfusion and metabolic demand are heterogeneously distributed, a given variable may not reflect the adequacy of tissue oxygenation in individual organs. Mixed (central) venous blood biochemical changes can be relatively insensitive to organ imbalances, and changes in venous Hb O2Sat cannot always be used to identify tissue hypoxia in individual patients (32). Likewise, there may be causes of increased lactate levels that are not related to tissue hypoxia (18, 36). Therefore, a multivariable analysis of DO2 crit may be important, since a group of systemic variables may indicate the state of tissue oxygenation in a more reliable fashion than any single individual marker.

The tissue O2 extraction and the O2-DO2 relationship have been examined at whole body (3, 9, 12, 20, 26, 45), organ (10, 28), and even capillary (13) levels. Previous attempts to analyze DO2 crit have focused on O2 (3, 9, 12, 20, 26, 27, 32, 45), PCO2 (11, 15, 45), and CI (24) relationships with DO2 and lactate (42, 44) and pH (45) changes. We speculated whether other systemic variables could be used to simultaneously determine DO2 crit and found that similar DO2 crit levels could be obtained for nearly 30 systemic variables. The small variability of DO2 crit determinations among variables (9%) in comparison with DO2 crit values among animals (14%) suggests that the use of several variables to estimate DO2 crit is a relatively accurate procedure. Therefore, it may be possible to determine a critical point in whole body metabolism based on variables that are easier to measure experimentally and clinically. The mean DO2 crit in this study is similar to values reported previously for rats (39), rabbits (20), pigs (17), and dogs (45).

Estimation of critical values of systemic variables at DO2 crit has been so far mostly restricted to O2 (4, 5, 12, 26), CI (24), pH (3, 45), PCO2 (3, 45), and lactate (45). We extended these determinations and estimated critical values for a wide range of respiratory, biochemical, and hemodynamic variables (Table 2). Although average critical values for each variable may be used only as a reference for an animal subjected to decreases in DO2, the fact that these variables show progressive changes toward a critical point may be a very useful finding. If these concepts are confirmed in clinical settings, they will provide a valuable tool to determine the transfusion trigger on an individual basis (i.e., for each patient), without the need of the sophisticated methodologies such as those needed for CO and O2 measurement. The ability to predict this trigger level, by tracking progressive changes in these variables using unbiased, quantitative methods, may help in finding the appropriate time to start a therapeutic procedure. The possibility to determine a critical threshold level considering the time trend of several systemic variables (obtained before DO2 crit is reached) underscores the concept of an individual (Hb/DO2 crit) trigger obtained for each patient, as opposed to a single universal transfusion trigger.

We also demonstrated that changes in systemic variables as a function of Hb concentration can be used to estimate critical oxygenation levels. Despite the nonlinear relationship between DO2 and Hb, the changes in systemic variables as a function of Hb were very similar to those found as a function of DO2. Similar results were previously reported for rats (24) and dogs. The similar variability of Hbcrit and of DO2 crit suggests that Hb levels may be as accurate as DO2 in finding the critical decompensation point. This may be important, since, in most instances, levels of Hb are much easier to determine than levels of DO2 (22), and blood Hb concentration is an important variable directing transfusion therapy in patients suffering blood loss (1).

Even before DO2 crit was reached, some variables showed a systematic increase as DO2 was lowered. The continuous rise in O2ER was expected, since this is a mechanism that allows the maintenance of O2 consumption during reductions in DO2. Increased K+ levels and vasodilation (decreased total peripheral resistance) were also observed during hemodilution (Fig. 2). The decreased DO2 after hemodilution may evoke hypoxia-related mechanisms similar to those found during hemorrhagic hypotension, even though a number of differences exist between these two conditions. Vasodilation during the hemorrhagic shock has been associated with K+ channels (21), since shock is associated with derangements in the intracellular metabolic status and K+ channels can be opened by decreased intracellular ATP levels and by acidosis. In addition, hyperkalemia has been associated with death after hemorrhagic hypotension (19, 39). Ischemia-induced loss of hepatic K+ could account for a portion of the observed increase in extracellular K+ (19, 29), since the liver has been shown to experience the most severe reduction in O2 (2).

A common criticism in O2-DO2 experiments is that both variables depend on CO measurement, and some investigators pointed out that relationships determined between calculated variables in the presence of shared measurements could be erroneous (6, 30, 37). To obtain accurate results, we employed a continuous and reliable CO measurement method (39, 43) and also measured O2 by a CO-independent method. Although estimations using blood samples (O2b and O2ERb) and expired gas (O2g and O2ERg) gave similar DO2 crit values, the determination of DO2 crit was more often successful when measurements were based on expired gas.

Surprisingly, we did not find differences between rats treated with fresh and stored blood. In our experiments, blood was stored for 10.5 days, which may be equivalent to 42 days storage in humans, since one study (8) has shown that red blood cells (RBCs) from rats age four times faster than human RBCs. However, only a limited number of variables was previously considered to evaluate the aging process of rat RBCs (8), and these cells may require longer storage times to develop all the changes that will make them comparable to aged human cells. For instance, rat blood stored for 28 days failed to improve oxygenation in models of sepsis (14) and hemorrhage (40). In addition, our experiments lasted a few hours from the time stored blood was infused until DO2 crit was reached. It is possible that the reinfused blood could have recovered part of its storage dysfunction during that time, since it is known that stored RBCs start regenerating immediately after infusion.

Limitations. Limitations to interpretation of the current study should be considered. Other systemic variables, not investigated in our study, may follow similar changes with DO2 crit and may also be used to estimate DO2 crit. Additionally, calculations based on some variables (e.g., arteriovenous differences in PCO2 and pH) may yield similar results (45). Although a number of indicators of anaerobic metabolism could be used, our study focused on those already clinically available and that can be easily monitored in a clinical setting.

Another limitation is the use of the dual-regression method. We noted that the changes in the variable as a function of DO2 were sometimes more complex than the program could resolve by trying to fit the data with two straight lines. However, a change in the relationship (at the same DO2 crit estimated by other variables) could be easily identified by visual inspection of the tracing. Therefore, it is possible that a more sophisticated (nonlinear) fitting of the data could yield higher success rates in estimating DO2 crit. For example, O2-DO2 relationships have been successfully fitted by exponentials (25). On the other hand, the dual-line method was used in the current analysis because it provided an unbiased means to estimate DO2 crit, and it is the most frequently employed method (12, 17, 20, 39). It also bears the advantage of ease of implementation and a small sensitivity to experimental error (27).

Anesthesia has been shown to affect the O2-DO2 relationship (25, 41), whereas ketamine and halogenated anesthetics may limit the tolerance to severe hemodilution (25, 41, 42). The anesthetic Saffan was chosen for this study because it preserves cardiovascular reflex activity (23) and because studies on cardiorespiratory functions in the rat used this anesthetic (38), including determinations of DO2 and O2 (39), as performed in our studies. In addition, we found that DO2 decreased sharply with a reduction in Hb once critical levels were reached (Fig. 1), indicating a strong CO response to hemodilution during Saffan anesthesia. These responses contrast with reports that ketamine prevented the expected increase in CO during hemodilution (42) and further support the use of Saffan.

The present experiments were designed to simulate some of the conditions of a surgical setting where blood transfusion, general anesthesia, mechanical ventilation, and changes in DO2 may occur. In addition, our study was performed under relatively controlled conditions using a homogenous group of animals. In clinical scenarios, Hbcrit (and DO2 crit) will depend on several factors, such as time course and intensity of hemorrhage/hemodilution, age, and comorbidity (1). In view of these limitations, further studies are necessary before our results can be applied to clinical situations.

The results of the present investigations may have implications for further understanding of prior controversial topics, e.g., instances where O2-DO2 plots fail to show a clear DO2-dependent segment (3, 9). In our study, O2 was not among the best variables to estimate DO2 crit. However, all animals that did not show supply dependency using O2 showed a clearly defined DO2 crit, as determined by at least 17 different variables. These results suggest that, although supply dependency of O2 may not have been detected in some previous studies, a critical level of DO2 could have been reached. Therefore, in some circumstances, a O2-DO2 plot may not be sufficient to determine DO2 crit.

In summary, the present studies in hemodiluted rats document similar patterns of systemic variables as a function of either DO2 or Hb. The data suggest that a multivariable analysis of critical O2 levels may help in reliably determining the physiological oxygenation boundary at the whole body level. This may assist in finding therapeutic triggers on an individual basis using variables that are readily available experimentally and clinically.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported, in part, by a grant from Hemosol, Toronto, Ontario, Canada.


   ACKNOWLEDGMENTS
 
We are grateful to Dr. Jiepei Zhu and Brian Berger for expert assistance in the methodology and to Dr. Penny Reynolds, Ben Pierce, and Leonardo Somera III, who assisted in data analysis.


   FOOTNOTES
 

Address for reprint requests and other correspondence: I. P. Torres Filho, Dept. of Anesthesiology, MCV-VCU, 1101 East Marshall St., Rm. B1-012, PO Box 980695, Richmond, VA 23298-0695 (E-mail: itorres{at}vcu.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. American Society of Anesthesiologists Task Force on Blood Component Therapy. Practice guidelines for blood component therapy. Anesthesiology 84: 732–747, 1996.
  2. Ba ZF, Wang P, Koo DJ, Cioffi WG, Bland KI, and Chaudry IH. Alterations in tissue oxygen consumption and extraction after trauma and hemorrhagic shock. Crit Care Med 28: 2837–2842, 2000.
  3. Baigorri F and Russell JA. Oxygen delivery in critical illness. Crit Care Clin 12: 971–994, 1996.
  4. Cain SM. Oxygen delivery and uptake in dogs during anemic and hypoxic hypoxia. J Appl Physiol 42: 228–234, 1977.
  5. Cain SM and Curtis SE. Experimental models of pathologic oxygen supply dependency. Crit Care Med 19: 603–612, 1991.
  6. Chittock D, Ronco J, and Russell J. Monitoring of oxygen transport and oxygen consumption. In: Principles and Practice of Intensive Care Monitoring (1st ed.), edited by Tobin M. New York: McGraw-Hill, 1998, chapt. 20, p. 317–343.
  7. Curtis SE and Cain SM. Systemic and regional O2 delivery and uptake in bled dogs given hypertonic saline, whole blood, or dextran. Am J Physiol Heart Circ Physiol 262: H778–H786, 1992.
  8. D'Almeida MS, Jagger J, Duggan M, White M, Ellis C, and Chin-Yee IH. A comparison of biochemical and functional alterations of rat and human erythrocytes stored in CPDA-1 for 29 days: implications for animal models of transfusion. Transfus Med 10: 291–303, 2000.
  9. Dantzker DR, Foresman B, and Gutierrez G. Oxygen supply and utilization relationships. A reevaluation. Am Rev Respir Dis 143: 675–679, 1991.
  10. De Blasi RA, Ferrari M, Antonelli M, Conti G, Almenrader N, and Gasparetto A. O2 consumption-O2 delivery relationship and arteriolar resistance in the forearm of critically ill patients measured by near infrared spectroscopy. Shock 6: 319–325, 1996.
  11. Dubin A, Murias G, Estenssoro E, Canales H, Sottile P, Badie J, Baran M, Rossi S, Laporte M, Palizas F, Giampieri J, Mediavilla D, Vacca E, and Botta D. End-tidal CO2 pressure determinants during hemorrhagic shock. Intensive Care Med 26: 1619–1623, 2000.
  12. Eichelbronner O, D'Almeida M, Sielenkamper A, Sibbald WJ, and Chin-Yee IH. Increasing P50 does not improve DO2 CRIT or systemic O2 in severe anemia. Am J Physiol Heart Circ Physiol 283: H92–H101, 2002.
  13. Ellis CG, Bateman RM, Sharpe MD, Sibbald WJ, and Gill R. Effect of a maldistribution of microvascular blood flow on capillary O2 extraction in sepsis. Am J Physiol Heart Circ Physiol 282: H156–H164, 2002.
  14. Fitzgerald RD, Martin CM, Dietz GE, Doig GS, Potter RF, and Sibbald WJ. Transfusing red blood cells stored in citrate phosphate dextrose adenine-1 for 28 days fails to improve tissue oxygenation in rats. Crit Care Med 25: 726–732, 1997.
  15. Guzman JA, Lacoma FJ, Najar A, and Kruse JA. End-tidal partial pressure of carbon dioxide as a noninvasive indicator of systemic oxygen supply dependency during hemorrhagic shock and resuscitation. Shock 8: 427–431, 1997.
  16. Hayes MA, Timmins AC, Yau EH, Palazzo M, Hinds CJ, and Watson D. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 330: 1717–1722, 1994.
  17. Heard SO, Baum TD, Wang HL, Rothschild HR, and Fink MP. Systemic and mesenteric O2 metabolism in endotoxic pigs: effect of graded hemorrhage. Circ Shock 35: 44–52, 1991.
  18. James JH, Luchette FA, McCarter FD, and Fischer JE. Lactate is an unreliable indicator of tissue hypoxia in injury or sepsis. Lancet 354: 505–508, 1999.
  19. Johnson KB, Wiesmann WP, and Pearce FJ. The effect of hypothermia on potassium and glucose changes in isobaric hemorrhagic shock in the rat. Shock 6: 223–229, 1996.
  20. Kuwagata Y, Oda J, Matsuyama S, Nakamori Y, Fujimi S, Ogura H, Nishino M, and Sugimoto H. Dopamine does not correct oxygen consumption/oxygen delivery relation abnormality during vasomotor shock induced by interleukin-1beta. Shock 18: 536–541, 2002.
  21. Landry DW and Oliver JA. The pathogenesis of vasodilatory shock. N Engl J Med 345: 588–595, 2001.
  22. Lurie F, Jahr JS, and Driessen B. The novel HemoCu plasma/low hemoglobin system accurately measures small concentrations of three different hemoglobin-based oxygen carriers in plasma: hemoglobin glutamer-200 (bovine) (Oxyglobin), hemoglobin glutamer-250 (bovine) (Hemopure), and hemoglobin-Raffimer (Hemolink). Anesth Analg 95: 870–873, 2002.
  23. Mackway-Jones K, Foex BA, Kirkman E, and Little RA. Modification of the cardiovascular response to hemorrhage by somatic afferent nerve stimulation with special reference to gut and skeletal muscle blood flow. J Trauma 47: 481–485, 1999.
  24. Morita Y, Chin-Yee I, Yu P, Sibbald WJ, and Martin CM. Critical oxygen delivery in conscious septic rats under stagnant or anemic hypoxia. Am J Respir Crit Care Med 167: 868–872, 2003.
  25. Rock P, Beattie C, Kimball AW, Nyhan DP, Chen BB, Fehr DM, Derrer SA, Parker SD, and Murray PA. Halothane alters the oxygen consumption-oxygen delivery relationship compared with conscious state. Anesthesiology 73: 1186–1197, 1990.
  26. Ronco JJ, Fenwick JC, Tweeddale MG, Wiggs BR, Phang PT, Cooper DJ, Cunningham KF, Russell JA, and Walley KR. Identification of the critical oxygen delivery for anaerobic metabolism in critically ill septic and nonseptic humans. JAMA 270: 1724–1730, 1993.
  27. Samsel RW and Schumacker PT. Determination of the critical O2 delivery from experimental data: sensitivity to error. J Appl Physiol 64: 2074–2082, 1988.
  28. Schlichtig R, Klions HA, Kramer DJ, and Nemoto EM. Hepatic dysoxia commences during O2 supply dependence. J Appl Physiol 72: 1499–1505, 1992.
  29. Schlichtig R, Kramer DJ, and Pinsky MR. Flow redistribution during progressive hemorrhage is a determinant of critical O2 delivery. J Appl Physiol 70: 169–178, 1991.
  30. Schumacker PT. Oxygen supply dependency in critical illness: an evolving understanding. Intensive Care Med 24: 97–99, 1998.
  31. Schumacker PT and Cain SM. The concept of a critical oxygen delivery. Intensive Care Med 13: 223–229, 1987.
  32. Schumacker PT and Samsel RW. Oxygen delivery and uptake by peripheral tissues: physiology and pathophysiology. Crit Care Clin 5: 255–269, 1989.
  33. Shoemaker WC. Oxygen transport and oxygen metabolism in shock and critical illness. Invasive and noninvasive monitoring of circulatory dysfunction and shock. Crit Care Clin 12: 939–969, 1996.
  34. Shoemaker WC, Appel PL, and Kram HB. Tissue oxygen debt as a determinant of lethal and nonlethal postoperative organ failure. Crit Care Med 16: 1117–1120, 1988.
  35. Shoemaker WC, Appel PL, Kram HB, Waxman K, and Lee TS. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest 94: 1176–1186, 1988.
  36. Sibbald WJ, Messmer K, and Fink MP. Roundtable conference on tissue oxygenation in acute medicine, Brussels, Belgium, 14–16 March 1998. Intensive Care Med 26: 780–791, 2000.
  37. Stratton HH, Feustel PJ, and Newell JC. Regression of calculated variables in the presence of shared measurement error. J Appl Physiol 62: 2083–2093, 1987.
  38. Thomas T and Marshall JM. Interdependence of respiratory and cardiovascular changes induced by systemic hypoxia in the rat: the roles of adenosine. J Physiol 480: 627–636, 1994.
  39. Torres LN, Torres Filho IP, Barbee RW, Tiba MH, Ward KR, and Pittman RN. Systemic responses to prolonged hemorrhagic hypotension. Am J Physiol Heart Circ Physiol 286: H1811–H1820, 2004.
  40. van Bommel J, de Korte D, Lind A, Siegemund M, Trouwborst A, Verhoeven AJ, Ince C, and Henny CP. The effect of the transfusion of stored RBCs on intestinal microvascular oxygenation in the rat. Transfusion 41: 1515–1523, 2001.
  41. Van der Linden P, Gilbart E, Engelman E, Schmartz D, and Vincent JL. Effects of anesthetic agents on systemic critical O2 delivery. J Appl Physiol 71: 83–93, 1991.
  42. Van der Linden P, Schmartz D, De Groote F, Mathieu N, Willaert P, Rausin I, and Vincent JL. Critical haemoglobin concentration in anaesthetized dogs: comparison of two plasma substitutes. Br J Anaesth 81: 556–562, 1998.
  43. Veal N, Moal F, Wang J, Vuillemin E, Oberti F, Roy E, Kaassis M, Trouve R, Saumet JL, and Cales P. New method of cardiac output measurement using ultrasound velocity dilution in rats. J Appl Physiol 91: 1274–1282, 2001.
  44. Weiskopf RB. Do we know when to transfuse red cells to treat acute anemia? Transfusion 38: 517–521, 1998.
  45. Zhang H and Vincent JL. Arteriovenous differences in PCO2 and pH are good indicators of critical hypoperfusion. Am Rev Respir Dis 148: 867–871, 1993.



This article has been cited by other articles:


Home page
 CirculationHome page
G. J. Murphy, B. C. Reeves, C. A. Rogers, S. I.A. Rizvi, L. Culliford, and G. D. Angelini
Increased Mortality, Postoperative Morbidity, and Cost After Red Blood Cell Transfusion in Patients Having Cardiac Surgery
Circulation, November 27, 2007; 116(22): 2544 - 2552.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
The Society of Thoracic Surgeons Blood Conservatio, V. A. Ferraris, S. P. Ferraris, S. P. Saha, E. A. Hessel II, C. K. Haan, B. D. Royston, C. R. Bridges, R. S.D. Higgins, G. Despotis, et al.
Perioperative Blood Transfusion and Blood Conservation in Cardiac Surgery: The Society of Thoracic Surgeons and The Society of Cardiovascular Anesthesiologists Clinical Practice Guideline
Ann. Thorac. Surg., May 1, 2007; 83(5_Supplement): S27 - S86.
[Abstract] [Full Text] [PDF]

Home page
 Br J AnaesthHome page
D. A. Otsuki, D. T. Fantoni, C. B. Margarido, C. K. Marumo, T. Intelizano, C. A. Pasqualucci, and J. O. Costa Auler Jr
Hydroxyethyl starch is superior to lactated Ringer as a replacement fluid in a pig model of acute normovolaemic haemodilution
Br. J. Anaesth., January 1, 2007; 98(1): 29 - 37.
[Abstract] [Full Text] [PDF]

Home page
 Ann. Thorac. Surg.Home page
G. J. Murphy and G. D. Angelini
Indications for Blood Transfusion in Cardiac Surgery
Ann. Thorac. Surg., December 1, 2006; 82(6): 2323 - 2334.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
288/3/H1071    most recent
00884.2004v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Torres Filho, I. P.
Right arrow Articles by Ward, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Torres Filho, I. P.
Right arrow Articles by Ward, K. R.

HOME HELP FEEDBACK