American Journal of Emergency Medicine (2008) 26, 119 – 123

www.elsevier.com/locate/ajem

Original Contribution

A comparison of central venous and arterial base deficit

as a predictor of survival in acute traumaB
Thomas M. Schmelzer MDb, Andrew D. Perron MDa,
Michael H. Thomason MDb, Ronald F. Sing DOb,*
a
Department of Emergency Medicine, Maine Medical Center, Portland, ME 04102, USA
b
Department of Surgery, Section of Trauma, Carolinas Medical Center, Charlotte, NC 28203, USA

Received 8 December 2006; revised2 23 January 2007; accepted 24 January 2007

Abstract
Background: The arterial base deficit has been demonstrated to be a marker of shock and predictive of
survival in injured patients. The venous blood, however, may better reflect tissue perfusion. Its
usefulness in trauma is unknown. We compared central venous with arterial blood gas analysis to
determine which was a better predictor of survival in injured patients.
Methods: A prospective, nonrandomized series of acutely injured patients was investigated. Patients
who had an arterial blood gas analysis for acid-base determination had a simultaneous central venous
blood gas analysis and routine blood tests. Patient demographics, Injury Severity Score, and survival
past 24 hours were recorded. Arterial and venous blood samples were analyzed for pH, Pco2, Po2,
HCO3, hemoglobin-oxygen saturation, base deficit, and lactate.
Results: One hundred patients were enrolled. There were 76 survivors and 24 nonsurvivors. Wilcoxon
rank sum test and multivariate logistic regression were used for each recorded variable; only central
venous base deficit was predictive of survival past 24 hours ( P = .0081). Specifically, arterial base
deficit was not predictive of survival past 24 hours.
Conclusion: In a prospective series of acutely injured patients, central venous base deficit, not arterial
base deficit, was predictive of survival past 24 hours.
D 2008 Elsevier Inc. All rights reserved.

1. Introduction in diagnosis and treatment. Vital signs have traditionally

Rapidly available serologic markers of shock continue
to be investigated. Although circulatory collapse is an
obvious sign of shock, compensatory physiologic mech-
anisms can mask this impending collapse, especially in
young and otherwise healthy patients, leading to delays
B appropriate in the acutely injured patient.
This study was funded by Carolinas Medical Center, Department of
Surgery, Division of Trauma and Critical Care, Charlotte, NC.
* Corresponding author. Tel.: +1 704 355 3176; fax: +1 704 355 5619.
E-mail address: Ron.Sing@carolinashealthcare.org (R.F. Sing).
been the measurement of inadequate tissue perfusion;
however, until compensatory mechanisms are exhausted,
these can be misleading. Over the past 20 years,
additional knowledge concerning tissue perfusion and
oxygen transport has markedly increased our identifica-
tion and quantification of shock. However, much of this
involves invasive monitoring techniques, which are not
The arterial base deficit has been demonstrated to be
predictive of survival in injured patients [1]. Venous blood
is more easily obtained and may be more reflective of

0735-6757/$ – see front matter D 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.ajem.2007.01.024
120 T.M. Schmelzer et al.

Table 1 Patients included in the trauma code 1 and 2

categories
Trauma code 1
Clinical evidence of shock (blood pressure of b90 mm Hg,
heart rate of b50 or N130 beats/min)
(GCS of V8) with significant mechanism of injury
GSW to head, neck, torso
Penetrating traumatic arrest with signs of life in the field
Inhalation injury
Trauma code 2
Pneumothorax, hemothorax, or wide mediastinum
Significant vascular injury
Other significant penetrating trauma to head, neck, torso
Adults/pediatrics with burns on N25% total body surface area,
with second- or third- degree burns
High voltage electrical injury
GCS of N8 or V12, with traumatic brain injury
Clinical evidence of spinal cord injury
Suspected abdominal injury
Proximal long bone, pelvic, or spinal fracture
Proximal amputations
GCS, Glasgow coma scale; GSW, gun shot wound.
overall tissue perfusion and therefore a better predictor of
severity of shock and survival [2]. The usefulness of central
venous blood gas analysis in the acutely injured patients has
not been studied. We examined arteriovenous blood gas
differences present during the initial evaluation of the
injured patient.
2. Methods
Our prospective, nonrandomized, observational clinical
investigation included acutely injured patients arriving at an
urban, American College of Surgeons– designated level I
trauma center. Institutional review board approval was
obtained before starting the study.
The variables recorded included patient demographics,
physiologic parameters (temperature, systolic blood pres-
sure, pulse rate, and Injury Severity Score), and survival
past 24 hours. In addition, each patient’s arterial and venous
blood gas results (pH, Pco2, Po2, HCO3, hemoglobin-
oxygen saturation, and base deficit) were recorded.
Patients arriving as a trauma code 1 or 2 (see Table 1) who
underwent arterial blood gas analysis at the discretion of the
trauma team leader (trauma attending, emergency medicine
attending, general surgery post graduate years (PGY)-4
resident, or emergency medicine PGY-3 resident) had a
concurrent phlebotomy for central venous blood. Patients
who had phlebotomy of arterial and venous blood greater
than 2 minutes apart were excluded from analysis. Our
institutional practice guidelines regarding the evaluation of
patients categorized as trauma codes 1 or 2 require 35 mL of
venous blood drawn for laboratory analysis and potential type
and cross matching. One half milliliter of venous blood was
obtained using a blood gas syringe via a 3-way stopcock at
the time of the phlebotomy. Only patients who had venous
blood gases analyzed from a central vein (subclavian
or femoral) and arterial blood gases were included in the
study. Only the initial blood gases obtained were used for
study analysis.
Descriptive statistics (mean and SD) were calculated for
all variables. Wilcoxon rank sum test was used on survivors
vs nonsurvivors, with a P value of less than .05 considered
as significant. Multivariate logistic regression was used to
determine which variables predicted survival past 24 hours.
3. Results
One hundred patients were enrolled over an 18-month
period. There were 76 survivors and 24 nonsurvivors. There
were 73 males and 27 females, and the mean age was 40.2
(F23.6) years. The mechanism of injury was blunt trauma
for 67 patients, with the remaining 33 having penetrating
traumatic injuries. The mean initial temperature, systolic
blood pressure, and pulse rate were 97.98F (F3.28F), 109.6
(F50.9) mm Hg, and 97.1 (F23.6) beats/min, respectively.
The mean Injury Severity Score was 14.92 (F8.46). There
were no statistically significant differences seen between the
survivors and nonsurvivors for any of the demographic
variables or initial physiologic parameters. Results for each
recorded variable from the blood gas analyses are shown in
Table 2. Using the Wilcoxon rank sum test to compare these
variables in survivors and nonsurvivors, we found that only

Table 2 Measured variables for survivors vs nonsurvivors

Variable Survivors Nonsurvivors Wilcoxon rank
(n= 76 [mean F SD]) (n= 24 [mean F SD]) sum test P
Venous base deficit 0.3 F 4.3 2.9 F 4.9 .0040*
Arterial base deficit 2.2 F 4.4 4.1 F 5.4 .1195
Venous pH 7.3 F 0.1 7.3 F 0.1 .1947
Arterial pH 7.4 F 0.1 7.4 F 0.1 .2226
Venous serum lactate 3.4 F 2.6 4.00 F 2.9 .1999
Arterial serum lactate 3.4 F 2.9 4.2 F 2.9 .0656
* P b .05.
Arterial vs. venous base deficit in trauma
central venous base deficit achieved statistical significance
( P = .004); the arterial base deficit did not achieve statistical
significance ( P = .1195). When multivariate logistic
regression was used for each variable, only the central
venous base deficit was predictive of survival past 24 hours
( P = .0081).
4. Discussion
In our prospective, observational, clinical investigation,
we demonstrated that central venous base deficit is an accurate
predictor of 24-hour survival in acutely injured patients.
Several indicators of shock have been reported includ-
ing central venous oxygen saturation, serum lactate, and
base deficit. Scalea and colleagues [3], using a hemorrhage
model in dogs, reported the central venous oxygen
saturation to be a useful measure of intravascular volume
status during hemorrhage. Scalea et al [4] later followed
this report with a clinical study of the utility of central
venous oxygen saturation in injured patients. In this
consecutive series of injured patients, the central venous
oxygen saturation identified patients with increased blood
loss and transfusion requirements despite normal vital
signs. The use of central venous oxygen saturation to
monitor resuscitation of critically ill patients has been
shown to improve survival; however, no data show initial
values to be predictive of survival outcomes [5,6]. Also,
patients must have central venous catheters placed for
serial measurements of central venous oxygen, which is
invasive and may not be necessary or feasible in the
acutely injured patient.
Lactate has consistently been shown to be a marker for
oxygen debt and a predictor of mortality in trauma patients
[7-9]. Siegel and colleagues [10] showed that the median
lethal dose in patients with blunt multitrauma was associated
with an admission lactate concentration of 8 mmol/L. The
measurement of lactate, however, is not readily available in
many acute trauma settings, and elevated levels of lactate can
be associated with other pathologic states (eg, alcoholism and
diabetic ketoacidosis) [11]. Furthermore, correction of tissue
oxygen debt does not always correlate with changes in serum
lactate because it is also produced via other mechanisms in
the body (eg, red blood cell production, physiologic shunts,
impairment of pyruvate dehydrogenase), and only a small
fraction of elevated lactate levels is directly related to tissue
hypoperfusion [7].
Much attention has also been given to the measurement
of the base deficit. The arterial base deficit is an indicator of
shock as well as a guide to resuscitation. This measurement
can be obtained with a standard arterial blood gas analysis.
Laboratory and clinical research has shown the base deficit
to be a sensitive indicator of oxygen debt and a qualitative
indicator of mortality and severity of hemorrhagic shock
[12-15]. Davis and coworkers [16], using a porcine
hemorrhagic shock model, studied the relationship of the
121
base deficit to the invasively measured dynamic oxygen
transport variable and showed the base deficit to accurately
reflect hemodynamic and tissue perfusion changes during
hemorrhagic shock. Davis [17] later demonstrated a
correlation between base deficit and serum lactate levels.
Dunham et al [18], using a canine hemorrhage model,
demonstrated the base deficit to be the single variable most
predictive of mortality. Davis and colleagues [19] applied
the use of the base deficit in a clinical study of more than
200 patients and found direct correlation between base
deficit and volume requirements. An increase of the base
deficit in the setting of fluid resuscitation was a signal for
ongoing hemorrhage. Rutherford and colleagues [20]
substantiated the work of Dunham et al in a retrospective
review of 3791 trauma patients. In this study, a base deficit
was a significant marker for shock and a predictor of
mortality (a base deficit of b15 mmol/L in patients
b55 years was associated with a 25% mortality, and in
patients N55 years, a base deficit of b8 mmol/L was
associated with a 25% mortality). This measurement also
determined the need for ongoing resuscitative efforts in
injured patients. Trauma patients with a base deficit of
greater than 6 mmol/L have an increased risk of developing
early adult respiratory distress syndrome and severe
multiple organ dysfunction syndrome [21,22].
Investigations of patients undergoing cardiac arrest and
cardiopulmonary resuscitation (CPR) have demonstrated
significant differences in the arterial and venous blood gases
obtained during CPR. Adrogue and coworkers [23] dem-
onstrated an increased arteriovenous pH difference and an
arteriovenous Pco2 difference associated with increased
oxygen dependence. Other studies have demonstrated
selective acidosis in venous blood (ie, increased venous
hypercarbia) during CPR [24,25]. Of note in these studies,
the significant respiratory acidosis present in venous blood
was not present in the arterial blood. In addition, the
arteriovenous difference was more pronounced in patients
with severe hemodynamic instability [26].
Another useful initial physiologic index that has been
used to predict mortality in trauma patients is end-tidal
carbon dioxide. Levine and colleagues [27] demonstrated
that in trauma patients with cardiac arrest associated with
electrical activity but no pulse, an initial end-tidal carbon
dioxide of 10 mm Hg or less measured 20 minutes after the
initiation of advanced cardiac life support accurately
predicted death. Unfortunately, we do not have this
equipment readily available in our department, so we were
unable to evaluate this index in our study.
Tissue hypoxia leads to hypercarbia of 2 mechanisms: 1)
Hypoxia increases adenosine triphosphate decomposition
and hydrogen ion production, which stimulates anaerobic
metabolism and the production of organic acids, such as
lactate, therefore increasing carbon dioxide production. 2)
Reduced cardiac output from decreased preload in the
hemorrhage state also slows the removal and transport of
carbon dioxide to the lungs; therefore, carbon dioxide
122
accumulates in the venous blood. Because of the rapid
ability of the lungs to excrete carbon dioxide and to readily
diffuse oxygen into the alveoli, carbon dioxide can be
quickly ventilated off; therefore, arterial blood gases may
not demonstrate the hypercarbia that exists on the venous
side. Because of this, venous blood gases appear to better
reflect tissue perfusion, and arterial blood gases better
reflect the adequacy of ventilation and oxygenation [28].
Several recent studies examined venous hypercarbia in a
hemorrhage model. Benjamin et al [29] studied acid-base
correction using sodium bicarbonate in several canine
hemorrhagic shock models. Although their conclusion was
that arterial pH correction did not improve the outcome in
this animal model, one of their observations was a
significant difference in arterial carbon dioxide compared
with venous carbon dioxide related to the severity of shock.
Oropello and coworkers [2] looked at continuous fiber-optic
arterial and venous blood gas monitoring during hemor-
rhagic shock in another animal model and again found
significant arteriovenous pH, Pco2, and Po2 changes. Most
notably, the changes in the central venous blood gasses were
greater and occurred significantly earlier than in the arterial
blood gasses.
Another benefit of using venous base deficit compared
with arterial base deficit is the lower cost and complications.
Arterial sticks require specialized equipment and procedures
for collection and special sample storage and analysis. In
addition, multiple arterial sticks or an indwelling arterial
catheter is needed, which causes increased pain and
complications (ie, hematoma, infection, pseudoaneurysm,
distal embolization, and in rare cases, hand or limb ischemia
or loss) compared with standard venipuncture [30]. Arterial
blood gases also add to the cost of patient care [31].
The results of our study demonstrated that central venous
base deficit was a more accurate predictor of survival than
arterial base deficit. We believe this is due to the venous
blood reflecting systemic effects of shock more directly and
the arterial base deficit reflecting ventilation and oxygena-
tion deficits. In this prospective study, central venous base
deficit, not arterial base deficit, predicted survival past 24
hours in acutely injured patients.
References
[1] Kincaid EH, et al. Elevated arterial base deficit in trauma patients: a
marker of impaired oxygen utilization. J Am Coll Surg 1998;
187(4):384 - 92.
[2] Oropello JM, et al. Continuous fiberoptic arterial and venous
blood gas monitoring in hemorrhagic shock. Chest 1996;109(4):
1049 - 55.
[3] Scalea TM, et al. Central venous blood oxygen saturation: an early,
accurate measurement of volume during hemorrhage. J Trauma
1988;28(6):725 - 32.
[4] Scalea TM, et al. Central venous oxygen saturation: a useful clinical
tool in trauma patients. J Trauma 1990;30(12):1539 - 43.
[5] Reinhart K, Bloos F. The value of venous oximetry. Curr Opin Crit
Care 2005;11(3):259 - 63.
T.M. Schmelzer et al.
[6] Powell CC, Schultz SC, Malcolm DS. Diaspirin crosslinked
hemoglobin (DCLHb): more effective than lactated Ringer’s solution
in restoring central venous oxygen saturation after hemorrhagic
shock in rats. Artif Cells Blood Substit Immobil Biotechnol 1996;
24(3):197 - 200.
[7] Moomey Jr CB, et al. Prognostic value of blood lactate, base deficit,
and oxygen-derived variables in an LD50 model of penetrating trauma.
Crit Care Med 1999;27(1):154 - 61.
[8] Husain FA, et al. Serum lactate and base deficit as predictors of
mortality and morbidity. Am J Surg 2003;185(5):485 - 91.
[9] Aslar AK, et al. Admission lactate level and the APACHE II score are
the most useful predictors of prognosis following torso trauma. Injury
2004;35(8):746 - 52.
[10] Siegel JH, et al. Early physiologic predictors of injury severity and
death in blunt multiple trauma. Arch Surg 1990;125(4):498 - 508.
[11] Dunne JR, et al. Lactate and base deficit in trauma: does alcohol
or drug use impair their predictive accuracy? J Trauma 2005;
58(5):959 - 66.
[12] Rixen D, Siegel JH. Bench-to-bedside review: oxygen debt and its
metabolic correlates as quantifiers of the severity of hemorrhagic and
post-traumatic shock. Crit Care 2005;9(5):441 - 53.
[13] Martin M, et al. Diagnosis of acid-base derangements and mortality
prediction in the trauma intensive care unit: the physiochemical
approach. J Trauma 2005;58(2):238 - 43.
[14] Tremblay LN, Feliciano DV, Rozycki GS. Assessment of initial base
deficit as a predictor of outcome: mechanism of injury does make a
difference. Am Surg 2002;68(8):689 - 93 [discussion 693-4].
[15] Kaplan LJ, Kellum JA. Initial pH, base deficit, lactate, anion gap,
strong ion difference, and strong ion gap predict outcome from major
vascular injury. Crit Care Med 2004;32(5):1120 - 4.
[16] Davis JW, et al. Base deficit as an indicator of significant abdominal
injury. Ann Emerg Med 1991;20(8):842 - 4.
[17] Davis JW. The relationship of base deficit to lactate in porcine
hemorrhagic shock and resuscitation. J Trauma 1994;36(2):168 - 72.
[18] Dunham CM, et al. Oxygen debt and metabolic acidemia as
quantitative predictors of mortality and the severity of the
ischemic insult in hemorrhagic shock. Crit Care Med 1991;19(2):
231 - 43.
[19] Davis JW, Kaups KL, Parks SN. Base deficit is superior to pH in
evaluating clearance of acidosis after traumatic shock. J Trauma
1998;44(1):114 - 8.
[20] Rutherford EJ, et al. Base deficit stratifies mortality and determines
therapy. J Trauma 1992;33(3):417 - 23.
[21] Rixen D, Siegel JH. Metabolic correlates of oxygen debt predict
posttrauma early acute respiratory distress syndrome and the related
cytokine response. J Trauma 2000;49(3):392 - 403.
[22] Cartotto R, et al. A prospective study on the implications of a base
deficit during fluid resuscitation. J Burn Care Rehabil 2003;
24(2):75 - 84.
[23] Adrogue HJ, et al. Assessing acid-base status in circulatory failure.
Differences between arterial and central venous blood. N Engl J Med
1989;320(20):1312 - 6.
[24] Grundler W, Weil MH, Rackow EC. Arteriovenous carbon dioxide
and pH gradients during cardiac arrest. Circulation 1986;74(5):
1071 - 4.
[25] Weil MH, et al. Difference in acid-base state between venous and
arterial blood during cardiopulmonary resuscitation. N Engl J Med
1986;315(3):153 - 6.
[26] Middleton P, et al. Agreement between arterial and central venous
values for pH, bicarbonate, base excess, and lactate. Emerg Med J
2006;23(8):622 - 4.
[27] Levine RL, Wayne MA, Miller CC. End-tidal carbon dioxide and
outcome of out-of-hospital cardiac arrest. N Engl J Med
1997;337(5):301 - 6.
[28] Mathias DW, Clifford PS, Klopfenstein HS. Mixed venous blood
gases are superior to arterial blood gases in assessing acid-base status
Arterial vs. venous base deficit in trauma
and oxygenation during acute cardiac tamponade in dogs. J Clin
Invest 1988;82(3):833 - 8.
[29] Benjamin E, et al. Effects of acid-base correction on hemodynamics,
oxygen dynamics, and resuscitability in severe canine hemorrhagic
shock. Crit Care Med 1994;22(10):1616 - 23.
123
[30] FitzSullivan E, et al. Serum bicarbonate may replace the arterial base
deficit in the trauma intensive care unit. Am J Surg 2005;190(6):941 - 6.
[31] Jacobs IA, et al. Cost savings associated with changes in routine
laboratory tests ordered for victims of trauma. Am Surg
2000;66(6):579 - 84.