I’m putting this question out there so that everyone can reach into their bag of tricks and tell me “this is how I know how much to give.” Because I really do want to know.
To the best of my knowledge, that answer does not exist.
Of course, someone can say “we have to give 2 liters of crystalloids!” loudly and clearly, raising a fistful of papers and guidelines where it is printed, black on white.
Now lets think about exactly what that question is really asking, which is: how much fluid can I safely give my patient to improve the hemodynamic status without a significant cost in terms of complications, both short and long-term?
What are the implications of this statement?
a. My patient must be volume responsive.
b. I must not cross the threshold of excessive fluid administration.
c. I must KNOW that fluid loading is the most appropriate therapy for my patient.
Now when you start looking at those questions and really wondering whether you have a definite answer for all of them, if becomes quite difficult. It is much easier to say “well, such-and-such study says people do better if I give them x amount.” This is probably true, that most people do better with a certain amount of fluid. But not everyone, because there are a lot of variables within those three questions.
Lets go through these questions:
My patient must be volume responsive – this is worth an entire podcast/lecture/article and there are a number of technologies and techniques out there which have both physiological and practical merit. I’ll get into that soon, but the short answer is that a passive leg raise (PLR) coupled with good cardiac output monitoring would be the “gold medal,” bedside ultrasound of the IVC’s global dynamics would be the “silver medal” and most of the rest would be “bronze medal” material at best. I won’t even mention the traditional physical exam or history or labs, not only do they not make it onto the podium, they didn’t make it into the stadium. I have to thank Scott Weingart (www.emcrit.org) for the sports analogy which I think is really excellent, since making the podium is actually pretty good and the difference between the three medalists may not always be that great.
I must not cross the threshold of excessive fluid administration – this one is dicey, because that’s where you have to factor in the patient’s oncotic pressure, hydrostatic pressure, both of which can be somewhat estimated, but mostly the degree of leak or “capillaritis”, both systemic and local, and that, at initial presentation, is impossible. The best you can do is a very broad classification from not too leaky to really, really leaky. And that’s not science. Now my good and highly esteemed friend Daniel Lichtenstein (the true founder of bedside ultrasound) published the FALLS Protocol, which basically says you can fill someone up until the appearance of B lines (during resuscitation) by lung ultrasound is certainly right that you should stop at this point, but I would contend you probably need to stop a bit before the edema is actually happening, especially since, in all likelihood and especially with crystalloids, the extravasation will continue for a while. The SOAP and VASST studies clearly showed an association between positive fluid balance and mortality, all other factors adjusted. Hence, this question is almost impossible to answer with knowledge. An educated guess is the best we can hope for.
I must KNOW that fluid loading is the most appropriate therapy for my patient – now this is almost philosophical and anxiety-arousing, because it forces us to challenge the fundamentals of our therapy. In certain cases, it is quite clear. If someone has been hemorrhaging or overdiuresed into shock, it is fair to say that you know they need fluids. However, if we are dealing with a vasodilatory circulatory failure, why do we feel compelled to treat it with a therapy best reserved for hypovolemic circulatory failure… Hmmm. Food for thought. Perhaps we can start by looking back and seeing what the basis is for our almost unshakeable beliefs that fluids are benign and that vasoactive medications are evil… But that is for the near future.
Please, let me know what you think. I hope someone disagrees strongly.
Philippe Rola
thinking critical care 
“That answer does not exist”. I totally agree with you. Optimal volume status is always relative: if you say that “I must KNOW that fluid loading is the most appropriate therapy for my patient”, I would add “we don’t need to fill every fluid responsive patient”. Even when we are able to asses volume status with the most accurate techniques (gold or silver medal!), what is most important is interpreting those data. As Albert Einstein once said “Not everything that counts can be counted, and not everything that can be counted counts…”
Absolutely!