Clinical Article

Can Venous Blood Gas Samples Replace Arterial Blood Gas Samples for Measurement of Base Excess in Severely Injured Trauma Patients?

by

Tools



Tools

Abstract

The role of the arterial blood gas (ABG) base excess value has been well established as an important clinical indicator in the acutely injured trauma patient. 1  In this pilot study, thirty adult trauma patients at a level 1 trauma center were studied to correlate the base excess value of their ABG with their venous blood gas (VBG) base excess value during acute resuscitation.  We found a strong positive correlation between the arterial and venous samples with a correlation coefficient of 0.922, indicating that the two samples are clinically equivalent.  However, using a Bland Altman Analysis, we observed that as the base excess became more negative this correlation was less reliable.  A VBG can be more conveniently and safely sampled and is associated with less pain for the patient.  We conclude that the VBG may be able to replace the ABG for purposes of base excess measurement in an acutely injured trauma patient; however, further study will be needed.

Background

Many studies in the trauma and critical care literature have established the utility of the ABG base excess in the initial evaluation of the severely injured adult trauma patient. 1   It is an indicator of shock 2 and the efficacy of resuscitation, 3 blood product requirements, 4 a screen for the presence of intrabdominal injury, 5 as well as a predictor of overall mortality after trauma. 6  The American College of Surgeons “Advanced Trauma Life Support for Physicians” guidelines identify the ABG base excess as a useful tool in estimating the severity of a trauma patient’s perfusion deficit. 1  The ABG, however, is a painful procedure for the patient and poses a risk, albeit small, for vascular complications and infection.  The ABG is more time-consuming and the added arterial puncture poses a potential source for needle stick accidents.  Other studies have illustrated that the pH value of ABG and VBG samples are strongly correlated in a nontrauma setting. 7  We studied the arterial and venous base excess values in adult trauma patients to see if the VBG could replace the ABG for base excess interpretation in trauma patients.

Patients and Methods

This prospective study occurred at our Level 1 trauma center.  One hundred adult trauma patients presenting to our facility had an arterial and venous blood gas drawn upon presentation as part of the initial workup between May 2005 and May 2006.  The study protocol specified that the blood gases were to be drawn concurrently.  All laboratory samples for trauma patients were taken immediately to the laboratory.  The time of blood sample laboratory registration was recorded as the time of the blood draw.  However, simultaneous draws were not always possible.  From the original treatment group, 67 patients whose blood gases were drawn more than 60 seconds apart from each other were excluded, and 3 patients whose samples were both venous were also excluded. Thus, 30 patients were included in the final analysis.

Data analysis was done using simple linear regression and construction of a Bland Altman plot using Microsoft EXCEL.  The Bland-Altman plot is a graphical method that permits the comparison of two measurement techniques that is usually used to assess changes in instrumentation.

Results

We analyzed 30 adult trauma patients using a simple regression analysis.  The correlation between predictor (VBG-Base Excess) and criterion (ABG-Base Excess) is 0.922; the shared variance (Adjusted R Square) is .845, statistically significant at p&ln;.001 (F=159.48). The adjusted R square (the coefficient of determination) estimates the magnitude of the effect, or shared measurement characteristics, of the predictor (VBG) on the criterion (ABG).  A large R square indicates that VBG provides essentially the same information as ABG, that is, one can predict VBG from ABG with an accuracy of about 85%.

We then generated a Bland Altman plot, as shown on Figure 1, which illustrates that, as the base excess became more negative, the relationship between ABG and VBG becomes less reliable. The regression of Mean Base Excess on Deviation yielded a very small R Square, indicating a poor relationship, but inspection reveals that the low R Square is due to the high dispersion at highly negative values of the Mean Base Excess.

Figure 1
Figure 1

Discussion

The VBG is easier, quicker, and safer to obtain and is associated with significantly less pain for the patient.  It would be convenient for physician and patient to be able to replace the ABG with the VBG for analysis of base excess in trauma patients.  Regression analysis showed that VBG base excess value is a strong predictor of the ABG base excess value.  In our study however, as illustrated with the Bland Altman plot, as patients’ base excess became more negative, the relationship became less reliable.  ABG and VBG base excess seems not correlate well in our sickest patients.  Based on our results, VBG may be able to replace ABG in the initial evaluation of trauma patients who are not severely critically injured.

The failure of the correlation in the sickest patients is difficult to explain.  Ideally in our study, ABG and VBG were to be drawn concurrently.  This proved to be a difficult task, as seen by the fact that greater than two thirds of our initial study group had been excluded.  Additionally, we were unable to record the exact time of the blood draw and instead used the time the sample arrived at the lab by convention as the time of the draw.  This introduced several opportunities for error.  Since the sickest patients are the more hemodynamically labile, the second to second resuscitation may involve relatively dramatic changes in acid-base balance.  As such, very small time differences in VBG and ABG blood draws may result in very large base excess differences.  We postulate that this may be the most important reason for our unexplained finding.

These errors could have been obviated in two ways.  First, the time of the blood draw should be recorded in real time.  Second, a more stringent study will be needed to repeat the work in severely injured trauma patients with strictly enforced guideline for truly concurrent blood draws.

In conclusion, this study illustrates the potential of the VBG to replace the ABG for analysis of base excess in trauma patients.  The two values are strongly correlated.  A more stringent study will be needed in the future to validate this approach in more seriously injured patients.

References

  1. Committee on Trauma. Advanced Trauma Life Support for Physicians, Sixth Edition.Chicago, Illinois: American College of Surgeons, 1997.

  2. Davis JW, Shackford SR, Mackersie RC, Hoyt DB. Base deficit as a guide to volume resuscitation. Journal of Trauma-Injury Infection & Critical Care. 1988;28:1464-7.

  3. Davis JW, Kaups KL, Parks SN. Base deficit is superior to pH in evaluating the clearance of acidosis after traumatic shock. Journal of Trauma-Injury Infection & Critical Care. 1998;44:114-8.

  4. Davis JW JW. Parks SN. Kaups KL. Gladen HE. O’Donnell-Nicol S. Admission base deficit predicts transfusion requirements and risk of complications. Journal of Trauma-Injury Infection & Critical Care. 1996;41:769-74.

  5. Davis JW. Mackersie RC. Holbrook TL. Hoyt DB. Base deficit is an indicator of significant abdominal injury. Annals of Emergency Medicine. 1991;20:842-4.

  6. Rutherford EJ. Morris JA Jr. Reed GW. Hall KS. Base deficit stratifies mortality and determines therapy. Journal of Trauma-Injury Infection & Critical Care.1992;33:417-23.

  7. Kelly AM, McAlpine R, Kyle E. Venous pH can safely replace arterial pH in the initial evaluation of patients in the emergency department. Emergency Medicine Journal.2001;18:340-2.