Hemodynamic Optimization By Non-Invasive Determination Of Cardiac Output In Critically Ill Patients

Patients with life-threatening hypotension are commonly cared for in intensive care units (ICUs). Pathophysiologic perturbations due to disease states such as sepsis, hypovolemia, and congestive heart failure may lead to tissue hypoxia, a critical development which precedes multi-organ failure and death. Not only is it difficult to rapidly identify patients heading down this path, but the execution of effective hemodynamic resuscitation to slow or reverse this process is challenging.

A cornerstone of hemodynamic resuscitation is volume administration. The goal of volume administration is to maximize cardiac output, and thereby systemic oxygen delivery, by optimizing cardiac preload. Bedside clinical assessment is inadequate for judging whether or not this goal has been achieved. Invasive techniques, such as central venous pressure (CVP) monitoring and pulmonary artery catheterization, carry risks, are costly and time-consuming, and may yield misleading data. Doppler ultrasound-based technology has been developed that allows for instantaneous, non-invasive monitoring of key hemodynamic parameters, such as cardiac output. This technology may facilitate determining the etiology of hemodynamic disturbances and assist the clinician in optimizing therapy based on changes in hemodynamic parameters. This randomized, controlled trial was designed to determine the impact of a volume resuscitation protocol, guided by non-invasive Doppler ultrasound technology, on outcomes in medical ICU patients with vasopressor-dependent hypotension.

Materials and methods:

Patient Population:

Study participants were recruited between January 19, 2010 and December 26, 2010 from two medical ICUs at Barnes-Jewish Hospital, a 1252-bed urban teaching hospital. The Washington University School of Medicine Human Research Protection Office approved the study, and informed consent was obtained from participants or their authorized representatives. Inclusion criteria were age ≥18 years; administration of vasopressors, defined as a continuous infusion of norepinephrine at a dose >5 mcg/min, dopamine ≥5 mcg/kg/min or any dose of another vasopressor; and passage of <18 hours since initiation of vasopressors at doses specified above. Exclusion criteria were hemorrhagic shock, need for immediate surgery, imminent risk of death in the next 48 hours (as judged by the attending ICU physician), level of care decision that precluded implementation of the study protocol, enrollment in any other clinical study, and pregnancy.

Study Protocol:

Subjects were assigned to treatment groups using blocked randomization to receive either volume resuscitation guided by the ultrasound cardiac output monitor (USCOM; USCOM Ltd., Sydney, Australia) or observation. The USCOM is a non-invasive device that uses continuous-wave Doppler ultrasound measurements of blood flow in the ascending aorta or pulmonary artery to estimate stroke volume (SV). Prior to the beginning of the study, one of the study investigators (LMD) underwent a supervised training period with the USCOM in 50 patients to insure reproducibility of the obtained measurements. Subjects randomized to the intervention group underwent a baseline USCOM measurement of SV by a single operator (LMD), followed by a pressurized infusion of 1 L of normal saline (NS) over approximately 15 minutes, after which the SV measurement was repeated. If the SV increased by ≥15%, the patient was deemed volume-responsive, and another 1 L NS bolus was administered. This process was repeated until the SV did not increase by ≥15%, two hours had elapsed since study enrollment, or 4 L of NS had been infused, whichever occurred first. With the exception of initial fluid management in the intervention group, as described above, use of all diagnostic and treatment modalities were at the discretion of the ICU physicians. Subjects in the control group underwent no intervention; ongoing care was carried out at the discretion of the ICU physicians. ICU physicians were unaware of subjects' group assignments and the USCOM data acquired in the intervention group.