Ultrasound to Assess Volume Status--Why Bother?

Aaron B. Holley, MD


November 14, 2016

Volume Assessment by Bedside Debate

Indulge me and imagine that you are on intensive care unit rounds at an academic hospital. You are discussing a patient who is mechanically ventilated, hypotensive, and on vasopressors. Your team is sure that the patient is septic, but he has already had aggressive resuscitation with fluids. Your resident wants to bolus more fluid but the fellow thinks the patient is "volume-up." A lively debate over "volume status" ensues. Literature is cited and someone mentions the Starling curve. One team member, a medical student destined to be a cardiologist, has the audacity to claim that he could see the patient's jugular vein and that it was distended. Ultimately, a liter of normal saline is infused.

When it comes to volume status, the arguments are endless and everyone has a measurement preference. Central venous pressure (CVP), obtained using a pressure transducer attached to a central venous catheter, is probably the most popular. It is static and easy to interpret. Large, randomized controlled trials (RCTs) have added to CVP's prestige. The Rivers study[1] on early goal-directed therapy used a target of 12 cm H2O during resuscitation (18 cm H2O if the patient was mechanically ventilated). The ARDSnet investigators[2] gave credence to an ancient pulmonary proverb—"A dry lung is a happy lung"—by using a low CVP target (4 cm H2O) to reduce time on the ventilator. Its simplicity and the RCTs notwithstanding, CVP doesn't tell us what we want to know. It doesn't predict whether the patient's cardiac output will increase in response to a fluid bolus.[3,4]

What measure does predict response to a bolus? Pulse-pressure variation (PPV) on an arterial-line tracing is popular on the boards,[5] and the maneuver known as passive leg raise (PLR) has the advantage of increasing venous return to the heart without actually adding volume.[6] The PPV data were obtained from mechanically ventilated patients who were paralyzed, however, making it difficult to extrapolate to spontaneously breathing patients. And whatever the reason, I don't think I have ever seen a physician do a PLR.

Volume Assessment by Bedside Echo

These days, it seems that "everyone who is anyone" uses bedside ultrasound (US) to predict response to a fluid bolus. Inferior vena cava (IVC) diameter and respiratory variation are most often employed, with left ventricular ejection fraction thrown in for good measure.[7] How good are each of these US measures? Until now, individual US parameters have only been validated in small studies of patients with septic shock

A new study published in the American Journal of Respiratory and Critical Care Medicine [8] compared three different US measures with PPV and PLR in 540 patients. Respiratory variation of superior vena cava (SVC) diameter (measured with transesophageal echocardiography [TEE]) and IVC diameter (using trans-thoracic echocardiography [TTE]) were assessed, along with the maximal Doppler velocity in left ventricular outflow track (measured with both echocardiographic approaches). All patients had septic, cardiogenic, or hypovolemic circulatory failure.

The study was prospective, observational, and very well done. Given the large sample size, the investigators were able to examine different factors expected to affect the performance of each measure, such as the effect that intra-abdominal pressure has on IVC measurements. The investigators found that respiratory SVC variation measured by TEE has the best specificity and overall accuracy, whereas maximal Doppler velocity in the left ventricular outflow track has the best sensitivity for predicting fluid response. All measures studied were less accurate than previously reported.

Volume Assessment by Bed Manipulation

Where does this leave us? Nowhere, really. The measure that performed best—respiratory SVC variation—was obtained with TEE and required neuromuscular blockade. Even with that, the study authors concluded that accuracy isn't sufficient to override clinical judgement. For example, they recommend a bolus if the risk for harm is low and there are signs of hypoperfusion, even if US parameters or PPV predict lack of response.

Ironically, the authors used PLR as their gold standard. They decided that it wasn't ethical to administer fluids to all patients regardless of volume assessment, because some might be harmed. Instead, everyone got a PLR. Patients with an increase in the left ventricular outflow tract velocity-time integral after PLR were considered responders.

So, in short, this study compared several technically difficult US measures with a PLR that can be done for free, with no training and without equipment. It's a great study with useful information, but I still can't figure out why we don't do PLR more often.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.