Interstitial washdown and vascular albumin refill during fluid infusion: novel kinetic analysis from three clinical trials

Abstract

Background and aims: Increased capillary filtration may paradoxically accelerate vascular refill of both fluid and albumin from the interstitial space, which is claimed to be edema-preventing. We characterized this proposed mechanism, called "interstitial washdown", by kinetic analyses of the hemodilution induced by intravenous infusion of crystalloid fluid during 3 distinct physiological states.

Methods: Greater plasma dilution of hemoglobin as compared to albumin during fluid therapy indicated recruitment of albumin, which was compared to the flow of interstitial fluid to the plasma as indicated by population volume kinetic analysis. Data for the comparison were derived from 24 infusions of crystalloid fluid in conscious volunteers, 30 in anesthetized patients, and 31 in patients with ketoacidosis from hyperglycemia.

Results: "Interstitial washdown" increased the plasma albumin concentration by between 0.3 and 1.0 g/L in the three series of infusions. The initial albumin concentration in the interstitial fluid returning to the plasma was estimated to between 22 g/L and 29 g/L, which decreased to an average of 50-75% lower during the subsequent 2-3 h. Kinetic simulations show that pronounced washdown was associated with increased capillary filtration (high k₁₂) and, in conscious subjects, with greater plasma and interstitial volume expansion and restricted urine flow. During anesthesia, the main effect was an increase in the non-exchangeable fluid volume ("third-spacing").

Conclusions: Crystalloid fluid accelerates lymphatic flow that moderately increases plasma albumin, but more clearly helps to maintain the intravascular volume. This "interstitial washdown" mechanism becomes exhausted after a few hours.

Keywords: Body water; Crystalloid solutions; Extracellular space; Isotonic; Pharmacokinetics; Physiology; Saline solution.

Fig. 1

Kinetic model. Schematic drawing of the kinetic model used to analyze the distribution and elimination of Ringer’s solution

Fig. 2

Albumin recruitment during crystalloid fluid therapy. The y-axis shows plasma albumin concentration that is due to interstitial washdown. Technically, each data point is the product of the Hb–albumin difference in plasma dilution and the plasma albumin concentration in A 20 volunteers receiving 1.7 L of Ringer’s acetate, B 30 patients given 1.7 L of Ringer’s acetate thyroid surgery, and C 31 infusions of 1 L of 0.9% saline in patients treated for diabetic ketoacidosis. Each infusion was given over 30 min

Fig. 3

Volume kinetic analyses based on the dilution of blood Hb. The subplots in the top row show the distribution of infused fluid in volunteers between the A central and the B peripheral fluid compartment, C the excess fluid in the central compartment when analyzing the volume kinetics based on Hb minus the volume expansion as obtained when albumin was used as the marker of plasma dilution. D Return flow of fluid from the peripheral to the central space (the plasma) when contrasting the influence of high-degree versus low-degree interstitial washdown (approximately 5–95% span). Subplots E–H show the same calculations when applied to patients under general anesthesia and Subplots I–L when applied to patients with ketoacidosis

Fig. 4

Influence of interstitial washdown on the distribution of crystalloid fluid. Volume kinetic analysis of the fluid distribution when 1.7 L of Ringer’s was infused in volunteers (top row; subplots A–D), 1.7 L to patients undergoing surgery (middle row; subplots E–H) and 1.0 L of 0.9% saline was given to patients with ketoacidosis (bottom row; subplots I–L). All infusions were given over 30 min. All volumes are shown depending on whether the interstitial washdown was in the low or high range (minimal to maximum washdown; approximately 5–95% span). For the volunteers, the range was between − 0.10 and + 0.10 (mean − 0.019), for the anesthesia patients − 0.01 to + 0.15 (mean, + 0.05), and for the patients with ketoacidosis the Hb–albumin difference in plasma dilution varied from − 0.10 to + 0.40 (mean, + 0.046)