Open Access Highly Accessed Research article

Prediction of arterial pressure increase after fluid challenge

Giuseppe Natalini1*, Antonio Rosano1, Carmine Rocco Militano1, Antonella Di Maio1, Pierluigi Ferretti1, Michele Bertelli1, Federica de Giuli2 and Achille Bernardini1

Author Affiliations

1 Intensive Care Unit, Poliambulanza Hospital Foundation, Brescia, Italy

2 Cardiology Division, Salvatore Maugeri Foundation, Lumezzane, Italy

For all author emails, please log on.

BMC Anesthesiology 2012, 12:3  doi:10.1186/1471-2253-12-3

Published: 5 March 2012

Abstract

Background

Mean arterial pressure above 65 mmHg is recommended for critically ill hypotensive patients whereas they do not benefit from supranormal cardiac output values. In this study we investigated if the increase of mean arterial pressure after volume expansion could be predicted by cardiovascular and renal variables. This is a relevant topic because unnecessary positive fluid balance increases mortality, organ dysfunction and Intensive Care Unit length of stay.

Methods

Thirty-six hypotensive patients (mean arterial pressure < 65 mmH) received a fluid challenge with hydroxyethyl starch. Patients were excluded if they had active bleeding and/or required changes in vasoactive agents infusion rate in the previous 30 minutes. Responders were defined by the increase of mean arterial pressure value to over 65 mmHg or by more than 20% with respect to the value recorded before fluid challenge. Measurements were performed before and at one hour after the end of fluid challenge.

Results

Twenty-two patients (61%) increased arterial pressure after volume expansion. Baseline heart rate, arterial pressure, central venous pressure, central venous saturation, central venous to arterial PCO2 difference, lactate, urinary output, fractional excretion of sodium and urinary sodium/potassium ratio were similar between responder and non-responder. Only 7 out of 36 patients had valuable dynamic indices and then we excluded them from analysis. When the variables were tested as predictors of responders, they showed values of areas under the ROC curve ranging between 0.502 and 0.604. Logistic regression did not reveal any association between variables and responder definition.

Conclusions

Fluid challenge did not improve arterial pressure in about one third of hypotensive critically ill patients. Cardiovascular and renal variables did not enable us to predict the individual response to volume administration.

Trial registration

ClinicalTrials.gov: NCT00721604.