| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Bicêtre Hospital, Le Kremlin Bicêtre, France
Correspondence to: Frédéric Michard, MD, PhD, Medical ICU, CHU de Bicêtre, 78 rue du Général Leclerc, 94275 Le Kremlin Bicêtre cedex, France; e-mail: michard.frederic{at}free.fr
To the Editor:
I read with interest the study by Martin et al1 (December 2002) describing correlations between the findings of portable chest radiographs and fluid balance in critically ill patients. In the study, the authors suggested that the vascular pedicle width (VPW) is an underutilized tool for the assessment of intravascular volume status. However, although the relationships reported in the study are statistically significant, those between the VPW and the wedge pressure or the fluid balance are weak (r2 range, 0.25 to 0.5), such that a given VPW value cannot be recommended to accurately predict intravascular volume status. Moreover, from a methodologic point of view, demonstrating that a parameter is sensitive to changes in volume status does not allow one to conclude that this parameter is useful in assessing intravascular volume. For example, central venous pressure goes up during fluid loading, goes down during fluid depletion, but cannot be used to assess intravascular blood volume,2 simply because intravascular volume is not the only determinant of central venous pressure. It must also be pointed out that chest radiographs were performed under the best conditions (prospective clinical study) and were interpreted by one experienced thoracic radiologist, which is usually not the case in real life.
The respiratory changes in arterial pressure were recognized as a clinical sign of hypovolemia > 35 years ago3 and have been shown to be correlated very closely with intravascular volume.4 In contrast to chest radiograph interpretation, the quantification of arterial pressure variation is not operator dependent and even now is automatically calculated by commercially available devices. Since most patients with acute lung injury/ARDS are sedated (at least transiently for the correct measurement of plateau inspiratory pressure and total positive end-expiratory pressure) and instrumented with an arterial line, there is no doubt that the arterial pressure variation is also an underutilized tool for assessing intravascular volume status. Finally, since the echocardiographic study of respiratory changes in aortic blood flow provides information that is similar to that from arterial pressure waveform analysis,5 I am afraid that the VPW cannot be considered as the "most sensitive noninvasive indicator of intravascular volume."
References
Emory University, Atlanta, GA Vanderbilt University, Nashville, TN
Correspondence to: Greg S. Martin, MD, FCCP, 69 Jesse Hill Jr. Dr SE, Room 2D-004, Atlanta, GA 30303; e-mail: Greg_Martin{at}emory.org
To the Editor:
We appreciate Dr. Michard highlighting the importance of monitoring intravascular volume in critically ill patients. Variations in arterial pressure are a valid and useful tool in estimating intravascular volume, particularly as it relates to cardiovascular responsiveness to fluid loading. Fluid responsiveness may be predicted based on variations in systolic pressure alone or, more accurately, on variations in pulse pressure.1 Variations in aortic blood velocity may provide even more information, but require transesophageal echocardiography,2 and may reflect changes in airway pressure and intrathoracic pressure more than changes in cardiovascular hemodynamics.3 As with variations in arterial pressure related to intrathoracic pressure and cardiac preload, variations in pulse oximetry tracings correlate with intrathoracic pressure changes in patients with acute exacerbations of obstructive airway disease.4 As Dr. Michard points out, new methods for quantifying systolic pressure variations in critically ill patients are available at selected institutions.5 The utility and accuracy of repeated measures over time for predicting changes in intravascular volume remains uncertain.
Despite the moderate correlation between radiographic vascular pedicle width (VPW) and fluid balance or hydrostatic pressure, practicing physicians use less reliable information on a daily basis. In the absence of invasive monitoring in the ICUs, we often base our estimates of intravascular volume on notoriously unreliable findings from the physical examination and vital signs.6 7 Even worse, we currently integrate findings from the readily available chest radiograph (such as the character of infiltrates, peribronchial cuffing, or effusions) in our daily patient evaluation without recognizing the superior nature of VPW in predicting volume status compared to classical roentgenographic findings.8 The consistent relationship between VPW and blood volume has been shown previously,9 making VPW in the context of this study the most sensitive radiographic indicator of intravascular volume. Furthermore, > 500 independent evaluations of VPW by multiple radiologists show intrareader and interreader reproducibility correlation to be excellent (r = 0.84 to 0.96). 8 10
In assessing intravascular volume or fluid responsiveness, true volumetric measures are better than pressure measures traditionally available from pulmonary artery catheterization.11 12 Despite growing availability and acceptance, these measurements are invasive. As with our data, VPW historically has been shown to correlate better with blood volume than with hydrostatic pressure,13 making the chest radiograph an essential tool for monitoring intravascular volume in critically ill patients. We as clinicians must integrate the emerging literature regarding noninvasive assessment of intravascular volume into our practice. Given the ready availability and familiarity of chest radiographs, the information available from this medium should be optimized and incorporated into a diagnostic algorithm for patient care.14 Only further study can determine if such an approach will improve outcomes for our most severely ill patients.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
