A central component of hemodynamic management in the ICU is fluid therapy. Our goal is to “fill the tank” and make sure that a patient has adequate blood volume. Too much fluid can be as bad as too little—pulmonary edema, bowel wall edema, and abdominal compartment syndrome are all adverse consequences of giving too much fluid. Therefore, we need some way of determining whether or not a patient needs fluids. If he’s septic, he needs either fluids or a vasopressor. If he has decompensated CHF, he may need inotropes but he still needs enough blood volume to maintain perfusion.
In ventilated patients, there are two commonly used tools to determine preload responsiveness. One is the use of bedside ultrasound. The other relies on changes in the arterial waveform and analyzes either pulse pressure variation or stroke volume variation.
Before we get started, please note that there is a difference between a static measure of preload (CVP, wedge pressure) and preload responsiveness. Measurements of preload are simply points in time, and they correlate very poorly with preload responsiveness, which is an increase in cardiac output of at least 15% with a 500 mL fluid bolus. Dynamic measurements like arterial variation and IVC variation are much more predictive of preload responsiveness. So, forget about the CVP and the wedge. They aren’t helpful.
The second point is that preload responsiveness doesn’t necessarily mean the patient needs more volume! You are most likely preload responsive right now, but you don’t need a fluid bolus. A patient who is off vasopressors, well-perfused, and making plenty of urine doesn’t need more volume, no matter what the ultrasound suggests. A patient who is on norepinephrine and oliguric, on the other hand, would definitely benefit from some sort of analysis of his preload-responsiveness.
When the ventilator cycles during positive pressure ventilation, it reduces right ventricular filling. As the blood goes through the pulmonary circulation and back to the left atrium, this translates to reduced left ventricular filling. You will see this on the arterial line as a reduction in the pulse pressure—the height of the arterial tracing. You can measure the maximum pulse pressure, the minimum pulse pressure, and then average the two. Pulse Pressure Variation (PPV) can then be calculated:
PPV = [PPmax – Ppmin] / PPmean, then multiplied by 100 to express it as a percentage
A PPV > 13% suggests preload responsiveness. A PPV ≤ 13% means that more fluid probably won’t help. This is only valid in patient receiving mechanical ventilation who is synchronous with the ventilator. Spontaneous breathing will increase blood return and will alter the measurements. If you want to be precise, increase the tidal volume transiently to 8-10 mL/kg—that’s more of a physiologic stress on the circulation, and all the published studies used that tidal volume. Turn it back down to 6 mL/kg when you’re done. You can do this by hand, but there are commercial devices (LidCO, PiCCO, FloTrac) that will calculate it continuously.
Another way, and possibly more accurate, to use the arterial changes is by calculating Stroke Volume Variation (SVV). The integral of the arterial tracing, or the area under the curve, represents the stroke volume. The stroke volume multiplied by the heart rate is the cardiac output. Mechanical ventilation has the same effect on the stroke volume as it does the pulse pressure, and devices like LidCO and FloTrac can calculate the stroke volume and its variation:
SVV = [SVmax – Svmin] / SVmean, then multiplied by 100 to express it as a percentage
A SVV > 10% suggests preload responsiveness. A SVV ≤ 10% means that more fluid probably won’t help. This is subject to the same conditions as with PPV—the patient needs to be on assist-control ventilation, synchronous with the vent, and the measurement is most accurate with a tidal volume of 8-10 mL/kg.
pulse pressure variation and
stroke volume variation