A note on the errors of using venous congestion in intact rats for determinations of microvascular permeability.

The established ideas of transcapillary exchange have recently been challenged based on studies in intact rats. In vivo measurements of net fluid flux and albumin clearance in muscle (and skin) have given estimates of the reflection coefficient (sigma) for albumin of 0.98-0.99 compared to the sigma value of 0.90 obtained by most other techniques. This discrepancy has vast consequences for the understanding of the transcapillary passage of macromolecules. A sigma for albumin near unity implies that there is virtually no coupling between protein and fluid transfer as induced by, for example, increases in vascular hydrostatic pressures. However, there are several assumptions inherent in the seemingly straight-forward experiments on intact rats, and in the present study we tested the hypothesis that occlusion of the femoral vein by ligation induces only moderate and transient increments of venous pressure (PV). During control conditions PV was 6.3 mmHg and pressure increased to 12.8 mmHg immediately following venous occlusion. However, PV declined with time and after 30 minutes of occlusion the capillary hydrostatic pressure was only increased by 3.0 mmHg. Calculations of the capillary filtration coefficient gave completely unrealistic values, close to those of maximally vasodilated skeletal muscle. The findings suggest that data obtained from intact rats, albeit important and interesting, should be evaluated with great care due to possible experimental errors in the in vivo approaches. In particular, the technique of estimating sigma for albumin in intact rats must be subjected to modifications before allowing any reliable conclusions.(ABSTRACT TRUNCATED AT 250 WORDS)