Straining to Predict Fluid Responsiveness
Abstract & Commentary
By Andrew M. Luks, MD, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Associate Editor for Critical Care Alert.
Synopsis: This prospective clinical study demonstrates that arterial pressure changes during a 10-second Valsalva maneuver can be used to predict fluid responsiveness in spontaneously breathing, non-intubated patients.
Source: Monge García MI, et al. Arterial pressure changes during the Valsalva maneuver to predict fluid responsiveness in spontaneously breathing patients. Intensive Care Med 2008 Oct 2; Epub ahead of print.
Despite the prominent role that volume challenges play in the management of hypotensive patients, we lack simple, reliable methods to determine which patients will respond to such challenges. Monge Garcia and colleagues sought to address this problem by investigating whether a Valsalva maneuver could predict volume responsiveness in spontaneously breathing patients.
The authors studied patients in a multidisciplinary intensive care unit (ICU) at a single institution who had radial artery catheters in place and were deemed to require fluid administration for hypotension, tachycardia, or oliguria. Patients were excluded if they had arrhythmias, a history of syncope, or could not execute satisfactory Valsalva maneuvers. After obtaining baseline data, including measurement of the stroke volume index (SVI) using a FloTrac™ sensor connected to the arterial line, subjects were placed in the supine position and asked to perform a 10-second Valsalva maneuver during which they maintained a constant expiratory pressure of 30 cm H2O.
Central venous pressure, arterial pressure, and airway pressure were monitored continuously before, during, and after the maneuver. These data were used to calculate two primary variables including: 1) Valsalva pulse pressure variation (DVPP), the percent variation between the highest pulse pressure during Phase 1 of the Valsalva pressure response and the lowest pulse pressure during Phase 2, and 2) Valsalva systolic pressure variation (DVSP), the percent variation between the highest systolic pressure during Phase 1 and the lowest systolic pressure during Phase 2. Subjects were then administered a 500 mL bolus of 6% hydroxyethylstarch over 30 minutes, after which time the Valsalva maneuver and hemodynamic measurements were repeated. Patients were labeled as being volume responders if the SVI increased by > 15% following the fluid bolus. A variety of statistical analyses were then used to compare hemodynamic responses between responders and non-responders.
Of the 30 patients who participated in the study, there were 19 responders and 11 non-responders. It is important to note that prior to volume expansion, mean systolic blood pressures in the responders and non-responders were over 120 mm Hg while mean diastolic blood pressures were roughly 60 mm Hg. These average pressures did not change with volume expansion in either the responders or the non-responders. There were positive linear correlations between volume expansion-induced changes in the SVI and pre-infusion values of DVPP (r2 = 0.71; P < 0.0001) and DVSP (r2 = 0.60, P = 0001). There was no correlation between pre-infusion CVP and changes in SVI. A threshold DVPP of 52% predicted fluid responsiveness with a sensitivity of 91% and specificity of 95%. The positive and negative predictive values were 91% and 95%, respectively. A DVSP of 30% of greater predicted fluid responsiveness with a sensitivity of 73% and a specificity of 90%.
Commentary
Critical care clinicians repeatedly face the question of whether to administer fluids or vasopressors to hypotensive patients, but lack reliable means for determining which strategy is indicated for a particular patient. Static measures of volume status such as CVP and pulmonary capillary wedge pressure have proven to have little utility in this regard, while other proposed tactics such as passive leg raising are infeasible because they require echocardiographic measures of left ventricular outflow to determine whether the patient is in fact volume responsive.1 On the surface, the notion of using a Valsalva maneuver to predict volume responsiveness is very appealing. It is inexpensive, simple to perform, and the protocol described by Monge Garcia and colleagues does not require complicated measurement tools such as echocardiography as part of the formal assessment.
Unfortunately, the trial described above falls short in many important respects, and does not tell us whether this is a valid tool for determining volume responsiveness in our patients. First, even though hypotension is perhaps the most common reason for which we must decide whether to give fluids or vasopressors, the patients in this study were far from hypotensive, with mean blood pressures in the 120/60 mm Hg range prior to the volume infusion. As a result, we have no idea if this tactic has any predictive capability in the patients in whom we are most concerned about its utility. Second, to apply this tactic, patients must be cooperative and able to execute a 10-second Valsalva maneuver. Leaving aside the fact that most true Valsalva maneuvers require a longer duration expiratory pressure hold than 10 seconds, many if not most of the patients in the ICU in whom we want to assess volume responsiveness are not able to complete Valsalva maneuvers because they are either intubated or have pain, altered mental status, or other problems that limit patient cooperation.
Finally, there is an important item in the study protocol that limits the study's applicability. Rather than measuring hemodynamic responses to crystalloid administration, the study authors used 6% hydroxyethylstarch for volume resuscitation. Even before recent data suggested volume resuscitation with hydroxyethylstarch is associated with adverse side effects such as acute renal failure and increased need for renal replacement therapy,2 volume resuscitation with colloid solution was uncommon in many ICUs. Because of this protocol issue, we do not know whether the Valsalva maneuver predicts hemodynamic responses to the more commonly used crystalloid solutions. Clinicians who use such solutions for volume resuscitation should not rely on this maneuver to predict volume responsiveness until further studies address this limitation.
References
- Lamia B, et al. Echocardiographic prediction of volume responsiveness in critically ill patients with spontaneous breathing activity. Intensive Care Med 2007; 33:1125-1132.
- Brunkhorst FM, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008;358:125-139.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content