Colloids versus crystalloids as volume substitutes: clinical relevance of the serum oncotic pressure

With some amplifications, Starling's concept of the serum colloid osmotic or oncotic pressure as the determinant of fluid partition between the intravascular and the interstitial compartment has been confirmed by modern physiological research. The relationship between serum oncotic pressure and interstitial edema is non-linear, i.e. edema becomes progressively greater per mm decrease of the oncotic pressure. The intravascular volume effect of crystalloids is inseparable from interstitial edema, because it depends on an expansion of the interstitium which increases the hydrostatic pressure in that compartment sufficiently to compensate for a lowered capillary oncotic pressure. With large crystalloid fluid loads, about 90% of the total edema accumulates in skeletal muscle, subcutaneous fat, and the skin. The skin is particularly susceptible to the development of edema associated with a hypoproteinemic fluid overload, presumably because its extracellular space is three times larger than the average whole-body value. The intestine also shows a marked reaction to a hypoproteinemic fluid overload. There is evidence for a pathogenetic significance of hypoproteinemic edema in the myocardium, the intestine, and the skin. To rid a patient of this type of edema, concentrated albumin is indispensable, and its effects are demonstrable in the above-mentioned tissues, whereas a diuretic alone is ineffective. The therapeutic implications of these mechanisms and findings are discussed. In the lung, fluid exchange and distribution between the intravascular and the interstitial compartment is influenced by additional factors, and opinions on the use of colloids versus crystalloids continue to differ, particularly with respect to those circumstances where capillary permeability is presumably or demonstrably abnormal. However, the weight of the evidence still favours the concept that in patients with a much greater than respiratory distress syndromes much greater than, the serum oncotic pressure should at any rate not be permitted to drop below a certain critical level. A condensed review of these complex and as yet incompletely clarified problems is presented.