Diuretic response to colloid and crystalloid fluid loading in critically ill patients

Aims In the critically ill patient, fluid loading is commonly done to stabilise hemodynamics and increase diuresis, whereas the absence of diuresis may predispose to harmful overloading. The goal of the current study was to evaluate the diuretic response and determinants thereof upon crystalloid and colloid fluid loading. Subjects and methods This is a substudy on 42 clinically hypovolemic, septic or non-septic patients without acute kidney injury, who were randomly assigned, after stratification for sepsis, to a 90-min fluid loading protocol with either 0.9 % saline or a colloid solution (gelatin, hydroxyethyl starch 200/ 0.5 or albumin). Hemodynamics, biochemical parameters and diuresis were recorded. A response was defined by an increase in diuresis of[ 10 % during fluid loading. Results Diuresis increased more during saline than colloid infusion, together with a decline in colloid osmotic pressure (COP) of plasma and less increase in plasma volume and global hemodynamics with saline, at similar fluid balance. Nine patients (82 %) receiving saline had a diuretic response, compared to 13 patients (42 %) receiving colloids (P = 0.04), and the response was not predicted by underlying condition, global hemodynamics, volume of fluid infused and COP. Conclusion In critically ill patients with clinical hypovolemia, diuresis increases more during saline than colloid fluid loading, only partly dependent of a fall in plasma COP.