Objectives: Meningococcal septic shock is a devastating illness associated with an increase in vascular permeability leading to hypovolemia and accumulation of plasma proteins and fluid in tissues. The capillary leak syndrome is often associated with widespread thrombosis in the skin, limbs, and digits. We postulated that the increase in vascular permeability and the intravascular thrombosis might be caused by an inflammation-induced loss of endothelial and basement membrane glycosaminoglycans (GAGs), which play a role in the permeability and thromboresistant properties of the microvasculature.
Design: Prospective, single-center observational study.
Setting: University-affiliated meningococcal research unit and pediatric intensive care unit.
Patients: Eighteen children requiring intensive care for meningococcal sepsis, 18 children with steroid-responsive nephrotic syndrome, and 18 healthy control children.
Interventions: None.
Measurements and main results: Serum concentrations and urine excretion of glycosaminoglycans were measured and related to changes in glomerular permeability to plasma proteins. The size-distribution and nature of glycosaminoglycans were defined by Polyacrylamide Gel Electrophoresis and specific enzyme digestion. Urinary excretion of heparan sulfate, chondroitin-4-sulfate, and chondroitin-6-sulfate were significantly increased in meningococcal disease (MD) relative to healthy controls and children with steroid-responsive nephrotic syndrome. The urinary GAGs in MD were of similar size to those in controls when analyzed after pronase digestion. However, analysis of proteoglycan size before proteolytic digestion showed the urinary GAGs in MD were of lower molecular weight and unattached to proteins. The fractional excretion of albumin and immunoglobulin G in MD increased with severity of disease. Patients with severe or fatal MD had albumin clearances overlapping those seen in steroid-responsive nephrotic syndrome. There was a significant correlation between proteinuria in MD and urinary excretion of heparan sulfate (r2 = 0.611, p < .0001), chondroitin-4-sulfate (r2 = 0.721, p < .0001), and chondroitin-6-sulfate (r2 = 0.395, p < .0001).
Conclusions: The capillary leak in meningococcal disease is associated with increased plasma and urine concentrations of GAGs, which may be proteolytically cleaved from endothelial and basement membrane sites. The correlation between the severity of protein leakage and the urine excretion of GAGs suggests that loss of GAGs may be causally related to the increase in permeability to proteins.