Ann Thorac Surg 1999;67:739-744
© 1999 The Society of Thoracic Surgeons
Original Articles
Oxygen transport in critically ill infants after congenital heart operations1
Anthony F. Rossi, MDa,b,
Howard S. Seiden, MDa,b,
Ronda P. Gross, MSNa,b,
Randall B. Griepp, MDa,b
a Department of Pediatrics, The Mount Sinai Medical Center, New York, New York USA
b Department of Cardiothoracic Surgery, The Mount Sinai Medical Center, New York, New York, USA
Accepted for publication July 22, 1998.
Address reprint requests to Dr Seiden, The Mount Sinai Medical Center, PCICU-Box 1201, One Gustave Levy Place, New York, NY 10029
e-mail: howard_seiden{at}smtplink.mssm.edu
 |
Abstract
|
|---|
Background. Oxygen transport variables reflect the balance of oxygen delivery and demand. Because oxygen transport in infants undergoing congenital cardiac operations is not well described, we examined oxygen transport in such patients. Differences in oxygen transport between survivors and nonsurvivors and variables that might be predictive of outcome were sought.
Methods. We reviewed hospital records of infants admitted to the pediatric cardiac intensive care unit in our institution from January 1996 through April 1997. Infants in whom simultaneous arterial blood gas and systemic venous oxygen saturation measurements were performed on admission and at 6 and 24 hours after admission were included. Analyses of arterial pH, base excess, arteriovenous oxygen saturation differences, and oxygen extraction ratio were performed, including comparisons of survivors and nonsurvivors and changes over time.
Results. Forty-nine infants were included in the study, with 39 survivors. There were no differences in any parameter between survivors and nonsurvivors on admission or at 24 hours. At 6 hours, differences between survivors and nonsurvivors were significant for arterial pH (7.48 versus 7.35, p < 0.001), base excess (2.9 versus -4.3 mmol/L, p < 0.01), arteriovenous oxygen saturation difference (34 versus 43, p < 0.05), and oxygen extraction ratio (0.28 versus 0.53, p < 0.001). The oxygen extraction ratio at 6 hours was at least 0.5 in 6 of 39 survivors and 7 of 10 nonsurvivors (p = 0.002).
Conclusions. Infants who die after cardiac operations have significant derangements of oxygen transport at 6 hours after admission to the intensive care unit. Infants with an oxygen extraction ratio greater than 0.5 at 6 hours are at highest risk.
|
Introduction
|
|---|
Pathologic imbalance of oxygen consumption and delivery has been suggested as a mechanism that leads to anaerobic metabolism, multisystem organ failure, and death [1–5]. During periods of diminished oxygen delivery, aerobic metabolism is maintained initially by increased oxygen extraction by the tissues. Through this mechanism, oxygen consumption remains independent of oxygen delivery until a critical point is reached. At that point (thought to correspond to an oxygen extraction ratio (OER) between 0.5 and 0.6 by most investigators), oxygen consumption becomes supply dependent and lactic acidosis develops [6–8]. There has been little investigation of oxygen transport in infants with congenital heart disease, especially in the period immediately after congenital heart operations. The objective of this study was to describe oxygen transport in this population, with particular attention to the OER. The OER might best reflect the balance of oxygen transport, particularly after congenital heart operations when oxygen delivery might be limited by hypoxemia. In addition, it can be calculated postoperatively from routine measurements. Differences in oxygen transport between survivors and nonsurvivors could lead to insights regarding the mechanism of death and the physiologic consequences of heart operations on the oxygen transport system. Furthermore, derangements in oxygen transport variables could identify infants at highest risk.
|
Material and methods
|
|---|
The pediatric cardiovascular database was searched for all infants less than 1 year of age who had a cardiac operation in our institution between January 1996 and April 1997. Patients with simultaneous measurements of arterial blood gas and systemic venous oxygen saturation (SvO2) on admission and at 6 and 24 hours after admission comprised the study population.
Arterial pH was measured directly for each sample. Arterial base excess was calculated automatically based on measured pH and pCO2 using the Sigaard-Anderson nomogram. Arterial oxygen saturation (SaO2) was also calculated automatically based on measured arterial partial pressure of oxygen and pH. The venous samples were from a right atrial catheter in 36 patients and a superior vena caval catheter in 13. The arteriovenous oxygen saturation (AVO2) difference was calculated by subtracting the venous oxygen saturation (SvO2), measured by cooximetry, from the SaO2. The OER was calculated using the following formula:
where VO2 = oxygen consumption; DO2 = oxygen delivery; CaO2 = oxygen content of arterial blood; CvO2 = oxygen content of systemic venous blood; SaO2 = oxygen saturation of arterial blood; SvO2 = oxygen saturation of systemic venous blood; C = constant; and Hb = hemoglobin concentration. Differences in pH, base excess, AVO2 differences, and OER between survivors and nonsurvivors, as well as changes over time, were analyzed. Survival was defined as survival to discharge from the hospital.
Patients
There were 118 infants less than 1 year of age admitted to the Pediatric Cardiac Intensive Care Unit after congenital heart operation during the study period. Forty-nine infants met inclusion criteria for the study. Their ages were 1 day to 12 months (median, 1 month). Open heart surgery was performed in 48 (closed, n = 1), including stage 1 palliation for hypoplastic left heart syndrome (n = 12), repair of tetralogy of Fallot (n = 12), repair of total anomalous pulmonary venous return (n = 5), arterial switch operation, closure of ventricular septal defect, interrupted aortic arch repair, bidirectional Glenn shunt (n = 3 each), and other (n = 7) (Table 1).
All infants were mechanically ventilated (volume control mode, tidal volume 12 to 16 mL/kg) at the time of the study. All were sedated with a continuous infusion of fentanyl (10 to 15 µg/kg per hour) and paralyzed with vecuronium (0.1 mg/kg every hour as needed). Patients were transfused with packed red blood cells to keep the hemoglobin level higher than 13 gm/dL, and stable core temperature was maintained with acetaminophen, cooling blankets, or both.
Statistics
Statistics were performed using SigmaStat (SPSS, Inc, Chicago, IL). Student’s t test and Mann-Whitney’s rank sum test were used to evaluate differences between survivors and nonsurvivors. One way analysis of variance was used to evaluate changes over time. The 
2 test was used to evaluate thresholds of OER for survival and mortality. A p value less than 0.05 was considered statistically significant.
|
Results
|
|---|
There were 39 survivors and 10 nonsurvivors. The age at operation of survivors and nonsurvivors differed significantly (3.2 months versus 0.44 months, p < 0.01).
The means and standard errors of the mean for pH, base excess, AVO2 difference, and OER at different times for survivors and nonsurvivors are presented in Figures 1 through 4. There were no statistically significant differences between survivors and nonsurvivors in any of the four variables at admission and at 24 hours. At 6 hours, arterial pH was higher in survivors than nonsurvivors (7.48 versus 7.34, respectively, p < 0.001) as was base excess (2.9 versus -4.3 mmol/L, respectively, p < 0.01). The mean AVO2 difference at 6 hours was lower in survivors than nonsurvivors (34 versus 43, respectively, p < 0.05), as was the OER calculated at 6 hours (0.28 versus 0.53, respectively, p < 0.001).
View larger version (17K):
[in this window]
[in a new window]
|
Fig 1. Changes in mean pH for survivors and nonsurvivors over time (circles = survivors, squares = nonsurvivors).
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig 2. Changes in mean base excess for survivors and nonsurvivors over time (circles = survivors, squares = nonsurvivors).
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig 3. Changes in mean arteriovenous oxygen saturation (AVO2) difference for survivors and nonsurvivors over time (circles = survivors, squares = nonsurvivors).
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig 4. Changes in mean oxygen extraction ratio (OER) for survivors and nonsurvivors over time (circles = survivors, squares = nonsurvivors).
|
|
The base excess in nonsurvivors increased from 6 to 24 hours (p < 0.01), with no significant change from admission to 24 hours. There was no change in base excess in survivors over time. Neither the changes in pH nor AVO2 difference over time was significant in survivors or nonsurvivors. There was an increase in the OER in nonsurvivors from admission to 6 hours from 0.36 to 0.53 (p = 0.01) and then a decrease to 0.34 at 24 hours (p = 0.05). There was no change in the OER over time in survivors. Mean arterial pH was greater than 7.25 in survivors and nonsurvivors at all times.
Mortality was 70% (7 of 10) in patients with an OER at least 0.5 at 6 hours. Of the 5 patients with an OER at least 0.5 on admission, all survived. Survival was 91% in patients in which the OER was less than 0.5 at all times. The OER was at least 0.5 at any time in 8 of 39 survivors and 7 of 10 nonsurvivors (p = 0.005). Of the infants with an OER at least 0.5 at any time, 3 of 8 survivors and 7 of 7 nonsurvivors had their maximum OER at 6 hours.
|
Comment
|
|---|
Infants who have congenital heart operations are at significant risk of dying, with mortality for certain operations over 20% in many centers. Predictors of outcome can improve risk stratification, which is especially important for high-risk patients. Furthermore, early signs of circulatory insufficiency can be instrumental in decision making regarding escalating medical therapy.
Recently, investigators have used blood lactate levels as a marker of diminished systemic perfusion and as a predictor of outcome in infants and children after heart operations [9–11]. Interventions can be made based on serial blood lactate levels. Unfortunately, blood lactate levels become higher only after significant circulatory dysfunction, below the point when oxygen consumption becomes dependent on oxygen delivery [7], which might be too late to prevent end organ damage. Furthermore, increased blood lactate might not be related to current circulatory well-being but might rather reflect prior hemodynamic instability (preoperative or intraoperative) that led to end organ dysfunction and an inability to metabolize circulating lactate. Although hyperlactatemia is potentially useful as a predictor of outcome, it is a nonspecific condition related to current circulatory conditions, complex biochemical interactions, and end organ function. In critically ill infants interventions to correct suspected tissue dysoxia based on elevated lactate values could be detrimental rather than helpful [12, 13]. Earlier, more specific warning signs could potentially allow appropriate medical intervention before significant systemic tissue dysoxia occurs.
Despite the considerable amount of investigation of cardiac output determination in patients after heart operations, few data are available describing oxygen transport in infants after congenital heart operations [14–16]. Cardiac output measurements alone might be misleading in the postoperative period. Normal cardiac output might be inadequate at times of increased oxygen demand, and a lower cardiac output might be sufficient during times of lower oxygen demand. Cardiac output measurements made without knowledge of oxygen demand might lead the clinician to erroneous conclusions regarding the well-being of the postoperative patient. The OER reflects the current relationship between oxygen delivery and oxygen demand and is directly related to circulatory performance in patients with cardiogenic shock.
This retrospective study of 49 infants after congenital heart operations helps to better define oxygen transport in these patients. On admission and at 24 hours postoperatively, there were no differences in measured or derived oxygen transport data between survivors and nonsurvivors. However, 6 hours after admission to the cardiac intensive care unit, significant differences in pH, base excess, AVO2 difference, and OER were present between these subgroups. Nonsurvivors had significant derangements of oxygen transport at 6 hours, closer to the critical OER of at least 0.5, above which there is much less cardiovascular reserve. In nonsurvivors there was also evidence of significant tissue dysoxia, with a diminished base excess. Reacting to derangements in base excess, clinicians attempted to normalize blood gas levels. The improvement in base excess and pH in nonsurvivors from 6 to 24 hours might reflect a combination of the generous use of buffering agents in the sickest patients or a true improvement in oxygen delivery. The decrease in the OER suggests a significant improvement in oxygen delivery. Despite an OER in nonsurvivors that is indistinguishable from survivors at 24 hours, patients with elevated OER at 6 hours still were more likely to die.
In this study, an OER of at least 0.5 was predictive of death, with 70% of nonsurvivors and 25% of survivors achieving this value. In all nonsurvivors the greatest OER was recorded at 6 hours. Seventy percent of infants in whom the OER was at least 0.5 at 6 hours died. An OER of at least 0.5 on admission did not have negative prognostic implications, which suggests that if corrected expediently, severe derangements in oxygen transport are not necessarily lethal. Oxygen transport data on admission reflect only the short time from cessation of cardiopulmonary bypass to transfer to the intensive care unit. The data collected at 6 hours after admission could reflect a longer period of poor hemodynamics.
Previous investigators have described the role of monitoring systemic venous partial pressure of oxygen, SvO2 and AVO2 difference to assess cardiac output and tissue perfusion in patients after congenital heart operation [14, 17–19]. Because these factors are affected by anemia, sepsis, and oxygen consumption, there is controversy regarding their use. They are also dependent on systemic arterial hypoxemia. In cyanotic infants after heart operation, the difference might not accurately reflect the balance of oxygen delivery and consumption. The OER describes this balance even in the presence of arterial desaturation. Although the OER has not been shown to more accurately estimate cardiac output, it might better reflect overall cardiovascular well-being than cardiac output alone. The normal ratio of 4 to 5 times the amount of oxygen delivery to oxygen consumption suggests considerable reserve, and decreasing this factor (or increasing the OER) reflects less cardiovascular reserve. It has been suggested that at an OER between 0.5 and 0.6, oxygen consumption becomes pathologically dependent on oxygen delivery. At this critical OER level, tissue hypoxia occurs and anaerobic metabolism begins, leading to lactic acid production. This situation can lead to end organ damage, multisystem organ dysfunction, and ultimately, death.
In the current patient population anemia was controlled with transfusions of packed red blood cells. Attempts were also made to maintain stable oxygen consumption with aggressive temperature control, sedation, and mechanical ventilation. Therefore, any increase in the OER can be attributed primarily to a decrease in cardiac output.
A true SvO2 should reflect complete mixing of all the systemic venous blood return, making the pulmonary artery the most accurate sampling site in most patients. After congenital heart operation, pulmonary artery oxygen saturation is often a poor estimate of SvO2. For example, any left-to-right shunt will cause the pulmonary artery saturation to be higher, falsely elevating the estimated SvO2. Previous investigators have validated the use of both right atrial and superior vena caval saturations as an estimate of SvO2 [20–22].
The OER calculations did not take into consideration dissolved oxygen content, i.e., that which is not bound to hemoglobin. For all venous samples in all patients and for all arterial samples in patients that remained cyanotic postoperatively (14 of 49 patients), the amount of dissolved oxygen was trivial compared with that bound to hemoglobin and was disregarded. Because all patients, except those who had stage I palliation for hypoplastic left heart syndrome, were treated initially with a 100% fraction of inspired oxygen, there could be an error in calculation for the initial OER in patients who had a biventricular repair. However, the fraction of inspired oxygen is quickly weaned in all patients as long as the arterial partial pressure of oxygen is greater than 100 mm Hg, where the amount of dissolved oxygen is trivial compared with that bound to hemoglobin. There should, therefore, be no such error in the 6- and 24-hour calculations.
The number of patients in the study is just less than half of all patients eligible by the age criterion. Most excluded patients were hemodynamically stable enough such that the continued measurement of arterial or venous blood gases up to 6 hours postoperatively was not considered necessary. The remainder had no reliable source of venous blood samples. For this reason, the conclusions might be applied only to the most critical cases.
These findings suggest that severe derangements in oxygen transport (as indicated by an elevated OER) occur frequently in nonsurvivors of infant heart operations. Furthermore, these derangements are most common 6 hours after admission to the intensive care unit and are accompanied by tissue dysoxia. The OER can be used as a predictor of outcome and as a real-time estimate of oxygen transport. Infants with an OER of at least 0.5 at 6 hours after admission to the intensive care unit are at significant risk of dying.
|
Footnotes
|
|---|
1 This article has been selected for the open discussion forum on the STS Web Site: http://www.sts.org/section/atsdiscussion/ 
|
References
|
|---|
-
Shoemaker W.C. Oxygen transport and oxygen metabolism in shock and critical illness. Invasive and noninvasive monitoring of circulatory dysfunction and shock. Crit Care Clin 1996;12:939-969.[Medline]
-
Baigorri F., Russel J.A. Oxygen delivery in critical illness. Crit Care Clin 1996;12:971-994.[Medline]
-
Edwards J.D. Oxygen transport in cardiogenic and septic shock. Crit Care Med 1991;19:658-663.[Medline]
-
Dunham C.M., Siegel J.H., Weireter L., et al. Oxygen debt and metabolic acidemia as quantitative predictors of mortality and the severity of the ischemic insult in hemorragic shock. Crit Care Med 1991;19:231-243.[Medline]
-
Lorente J.A., Renes E., Gomez-Aquinaga M.A., Landin L., de la Morena J., Liste D. Oxygen delivery-dependent oxygen consumption in acute respiratory failure. Crit Care Med 1991;19:770-775.[Medline]
-
Trouwborst A., Tenbrinck R., van Woerkens EC Blood gas analysis of mixed venous blood during normoxic acute isovolemic hemodilution in pigs. Anesth Analg 1990;70:523-529.[Abstract/Free Full Text]
-
Ronco J.J., Fenwick J.C., Tweeddale M.G., et al. Identification of the critical oxygen delivery for anaerobic metabolism in critically ill septic and nonseptic humans. JAMA 1993;270:1724-1730.[Abstract/Free Full Text]
-
Lugo G., Arizpe D., Dominguez G., Ramirez M., Tamariz O. Relationship between oxygen consumption and oxygen delivery during anesthesia in high risk surgical patients. Crit Care Med 1993;21:64-69.[Medline]
-
Siegel L.B., Dalton H.J., Hertzog J.H., Hopkins R.A., Hannan R.L., Hauser G.J. Initial postoperative serum lactate levels predict survival in children after open heart surgery. Intensive Care Med 1996;22:1418-1423.[Medline]
-
Cheifetz I.M., Kern F.H., Schulman S.R., Greeley W.J., Ungerleider R.M., Meliones J.N. Serum lactates correlate with mortality after operations for complex congenital heart disease. Ann Thorac Surg 1997;64:735-738.[Abstract/Free Full Text]
-
Toffaletti J., Hansell D. Interpretation of blood lactate measurements in paediatric open-heart surgery and in extracorporeal membrane oxygenation. Scand J Clin Lab Invest 1995;55:301-307.[Medline]
-
Bakker J., Gris P., Coffernils M., Kahn R.J., Vincent J.L. Serial blood lactate levels can predict the development of multiple organ failure following septic shock. Am J Surg 1996;171:221-226.[Medline]
-
Gutierrez G., Wulf M.E. Lactic acidosis in sepsis: a commentary. Intensive Care Med 1996;22:6-16.[Medline]
-
Buheitel G., Scharf J., Hofbeck M., Singer H. Estimation of cardiac index by means of the arterial and the mixed venous oxygen content and pulmonary oxygen uptake determination in the early post-operative period following surgery of congenital heart disease. Intensive Care Med 1994;20:500-503.[Medline]
-
Shenaq S.A., Casar G., Chelly J.E., Ott H., Crawford E.S. Continuous monitoring of mixed venous oxygen saturation during aortic surgery. Chest 1987;92:796-799.[Abstract/Free Full Text]
-
Baele P., Goenen M., Kestens-Servaye Y. Clinical use of continuous monitoring of mixed venous oxygen saturation during and after cardiovascular surgery. Acta Anaesthesiol Belg 1985;36:5-13.[Medline]
-
Parr G.V., Blackstone E.H., Kirklin J.W. Cardiac performance and mortality after intracardiac surgery in infants and young children. Circulation 1975;51:867-874.[Abstract/Free Full Text]
-
De la Rocha A.G., Edmonds J.F., Williams W.G., Poirier C., Trusler R.N. Importance of mixed venous oxygen saturation in the care of critically ill patients. Can J Surg 1978;21:227-229.[Medline]
-
Rossi A.R., Sommer R.J., Lotvin A., et al. Usefulness of intermittent monitoring of mixed venous oxygen saturation after stage I palliation for hypoplastic left heart syndrome. Am J Cardiol 1994;73:1118-1123.[Medline]
-
Freed M.D., Miettinen O.S., Nadas A.S. Oximetric detection of intracardiac left-to-right shunts. Br Heart J 1979;42:690-694.[Free Full Text]
-
Thayssen P., Klarholt E. Relation between caval and pulmonary artery oxygen saturation in children. Br Heart J 1980;43:574-578.[Abstract/Free Full Text]
-
Tahvanainen J., Meretoja O., Nikki P. Can central venous blood replace mixed venous blood samples?. Crit Care Med 1982;10:758-761.[Medline]
This article has been cited by other articles:
|
|
|
|
 
T. Juvonen, F. Biancari, J. Rimpilainen, V. Anttila, M. Pokela, V. Vainionpaa, P. Romsi, and K. Kiviluoma
Determinants of mortality after hypothermic circulatory arrest in a chronic porcine model
Eur. J. Cardiothorac. Surg.,
October 1, 2001;
20(4):
803 - 810.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Taeed, S. M. Schwartz, J. M. Pearl, J. L. Raake, R. H. Beekman III, P. B. Manning, and D. P. Nelson
Unrecognized Pulmonary Venous Desaturation Early After Norwood Palliation Confounds Gp:Gs Assessment and Compromises Oxygen Delivery
Circulation,
June 5, 2001;
103(22):
2699 - 2704.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
