Pre-eclampsia is the leading cause of morbidity of pregnant women. It is associated with high blood pressure, intrauterine growth retardation (IUGR), haemolysis, proteinuria and increased oedema, in part due to increased vascular permeability. To determine whether this increase in permeability is stimulated by a circulating macromolecule we have compared the effect of plasma from patients with severe and mild pre-eclampsia with plasma from patients with normal pregnancies on permeability of frog mesenteric microvessels.

With ethical committee and patient consent, plasma was taken from 6 patients with severe pre-eclampsia [3 with IUGR, 3 with haemolysis, elevated liver enzymes and low platelets (HELLP)], 10 patients with mild pre-eclampsia (proteinuria and hypertension but without IUGR or HELLP) and 45 normal pregnancy controls, matched for age and gestation. The samples were dialysed against frog Ringer using a 12 kDa molecular weight dialysis tubing. The colloid osmotic pressure of the dialysed plasma was measured in a modified Hansen oncometer. Frogs were anaesthetised by immersion in MS222 and the mesentery exposed as previously described. A microvessel was cannulated and perfused with a solution of human serum albumin that was matched for colloid osmotic pressure to the plasma to be perfused. A series of measurements of filtration rate were taken using a modification of the Landis Michel technique (Harper et al. 2002). The pipette was then refilled with dialysed plasma and filtration rate measured at two pressures (20 and 30cmH₂O) for up to ten min. Hydraulic conductivity (L_p) and oncotic reflection coefficient (ρ) were calculated from the slope and x intercept respectively of the filtration pressure relationship. The frogs were then humanely killed by destruction of the brain.

Neither control nor mild pre-eclamptic plasma caused a significant change in L_p or ρ within 10 min of perfusion. Perfusion of plasma from patients with severe pre-eclampsia, however, resulted in a rapid transient increase in L_p from mean ± S.E.M. 1.6 ± 0.26 to 11.5 ± 2.3 X 10^-7 cm.s^-1.cmH₂O^-1 (P < 0.02, paired t test) that returned to control within two min with all six plasma samples. ρ reduced from 0.98 ± 0.05 to 0.90 ± 0.05 (P = 0.1).

These results show that a macromolecule in pre-eclamptic plasma can increase microvascular permeability. The identity of the molecule is not known.