Estimating the Starling equation parameters in living humans by using volume kinetic analysis.

Abstract

The transcapillary filtration of fluid is governed by the Starling forces. The purpose of the present study was to simultaneously estimate all Starling parameters in living humans by using likelihood mathematics. First, the capillary filtration was estimated by volume kinetic analysis at 3,015 time points during 193 intravenous infusion experiments using crystalloid and colloid fluid in adult volunteers and surgical patients placed in the supine position. Second, the obtained capillary filtration and the measured plasma colloid osmotic pressure were entered into a new kinetic analysis to estimate the five remaining parameters in the Starling equation. The result shows that the capillary hydrostatic pressure was 16 mmHg (P_c), the interstitial hydrostatic pressure - 2.2 mmHg (P_i), the interstitial colloid osmotic pressure 1.6 mmHg (π_i), the reflection coefficient 0.79, and the capillary filtration coefficient (K_f) 306 mL/min/mmHg for the whole body when crystalloid volume loading was performed. K_f was significantly lower when colloid fluid (albumin 5% and 20%) was infused, and π_i was lower in patients with subacute inflammation. Association plots showed that P_c was low in hypovolemia and inflammation but markedly higher when albumin was infused. Plots over time suggested that K_f was high early during an infusion, which was probably due to release of nitric oxide due to shear stress, but then decreased gradually. In conclusion, the parameters in the Starling equation could be estimated with acceptable confidence during fluid infusion experiments in humans. Changes of K_f appeared to be key parameter that modulated the capillary filtration rate during volume loading.