Can ultrasound predict volume responsiveness better than CVP?

For decades we have relied on static measurements of volume status such as central venous pressure, pulmonary artery occlusion pressure and right ventricular end diastolic volume.  CVP has been shown to have little better than a coin toss chance of predicting volume responsiveness¹ and the other measurements require an invasive pulmonary artery catheter.  More recently dynamic measurements including variation in right atrial pressure, pulse pressure and peak aortic flow velocity have been shown to more closely predict changes in cardiac index.²  They rely on changes in blood flow that occur in responses to normal changes in intrathoracic pressure with the respiratory cycle.  Unfortunately, these measures can be challenging to obtain.  For example PPV and right atrial pressure changes are most accurate when the patient has a regular heart rhythm and controlled breathing (sedated on a ventilator) and PAFV is best measured with transesophageal echocardiography.

Point of care ultrasound has also gained popularity as a method to assess volume responsiveness.  Measurements of IVC diameter and respiratory variation are easy to learn, can be performed quickly at the bedside and initial studies showed promise.    Bariber, et al. found the IVC distensibility index to be % sensitive for a 15% rise in cardiac index.³ A recent metaanalysis showed an IVC diameter < 2 cm to accurately predict a low CVP.⁴  While IVC diameter and distensibility has been shown to predict CVP, it is subject to the some of the same challenges as CVP in predicting volume responsiveness, including inaccuracies with right heart failure, tricuspid regurgitation and increased abdominal pressure.

Recently, investigators have evaluated carotid flow velocities as a measure of volume responsiveness.  Like aortic flow velocity, carotid flow velocity varies according to volume status and when a patient is volume depleted, improves with fluid challenge.  Unlike aortic flow velocity, the carotid artery is easily accessible using a linear probe in the anterior cervical triangle of the neck and can be performed with limited ultrasound skills.  In an elegant blood donor study, Mackenzie et al. measured carotid flow velocity before and after blood donation in 68 healthy volunteers.⁵  These investigators were able to demonstrate a statistically significant reduction in carotid flow velocity after blood donation.  Further, they were able to demonstrate a significant increase in carotid flow velocity with passive leg raise performed after, but not before blood donation.  Passive leg raise rapidly mobilizes 300 to 500 ml of whole blood and simulates a fluid bolus.

Carotid flow velocity holds promise in predicting volume responsiveness.  It’s accuracy rivals that of more invasive measurements, but it can be performed noninvasively at the bedside by clinicians with minimal training.⁶  More studies are needed to evaluate carotid flow velocities across a variety of clinical contexts.

References

1. Marik, et al. Does Central Venous Pressure Predict Fluid Responsiveness, CHEST 2008; 134:172–178
2. Marik, et al.  Noninvasive Cardiac Output Monitors: A State of the Art Review, Journal of Cardiothoracic and Vascular Anesthesia, 2012:1-14
3. Bariber et al.  Intensive Care Medicine.  Sept 2004:1740-1746
4. Frekker et al.  Critical Care Medicine. March 2013
5. Mackenzie et al.  Poster Presentation, 24^th International Symposium on Intensive Care, 2014, 18(suppl 1): P131
6. Evans, et al.  Volume Responsiveness in Critically Ill Patients:  Use of sonography to guide management.  J UTZ Medicine, 2014; 33:3-7