Sorry for the delay gang, doing some batch blog posting today to catch up.
One of the richer discussions we had during today’s case was the general question of how to assess cardiac output in patients with shock, and when to decide whether someone has had an adequate fluid resuscitation trial. Dr. James Frank as usual enlightened the group with some approaches to this question using non-invasive (i.e. no PA or CVC monitoring, a-line okay) techniques.
- More recent data have called into question reliance on central venous pressure (CVP) as a marker of fluid responsiveness, and for the most part we have moved away from relying on it in isolation.
- Broadly defined, measures of fluid responsiveness can be classified as upstream, relating to the pressures in the vena cava, heart chambers, and aorta, or downstream, attempting to directly measure tissue perfusion. Think upstream as a PA cath, and downstream as a lactate.
- One option is pulse pressure variation from an A-line, which is calculated as a percent difference (PPmax-PPmin/PPmean) the mean taken over 3-4 respiratory cycles. Differences of greater than 15% predict fluid responsiveness, one systematic review found an AUC for PPvariation and subsequent fluid responsiveness of 0.94 compared to 0.55 for CVP
- Another James Frank pearl, don’t rely on MAP as your marker of fluid responsiveness, it’s not sensitive enough
- Good old passive leg raise is decent, 86% and 92% sens/spec respectively, with a greater than 10% change in MAP (poor proxy for CO but usually all we have without central access), remember, get a baseline of them in a semi recumbent position, then move to flat and raise the legs about 45 degrees
- Finally, the sexy stuff, electrical bioimpedance and bioreactance technologies, which detects changes in fluid levels in the thorax (ideally most of it being in the heart) and extrapolates to changes in systole and diastole, thus providing an estimate of CO (SVxHR). Studies are inconsistent as to whether these work all that well. An expansion of the technology called bioreactance, which adds a calculation for the pulsatile flow of blood in the heart to better isolate it from the “noise” of a pleural effusion, edema etc. that can also result in intrathoracic fluid. These seem to work better, the sense in our room was they may have some utility but no consensus could be reached on their use