Endothelial Glycocalyx Degradation in Sepsis: Analysis of the Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis (CLOVERS) Trial, a Multicenter, Phase 3, Randomized Trial.

Abstract

Rationale: Prior evidence suggests that endothelial glycocalyx degradation contributes to sepsis pathogenesis and is potentially worsened by intravenous fluid resuscitation. Objectives: To assess 1) the association of endothelial glycocalyx degradation with sepsis mortality, 2) the impact of a randomly assigned liberal versus restrictive intravenous fluid resuscitation strategy on endothelial glycocalyx degradation, and 3) whether there is a differential treatment effect for mortality based on baseline endothelial glycocalyx degradation. Methods: We used an enriched sampling strategy to define a cohort of 574 patients enrolled in the CLOVERS (Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis) trial, which compared liberal versus restrictive intravenous fluid resuscitation strategies. We used mass spectrometry to quantify plasma heparan sulfate as the primary measure of endothelial glycocalyx degradation. Plasma syndecan-1, quantified by enzyme-linked immunoassay, served as a clinically feasible complementary index of endothelial glycocalyx degradation. The primary outcome was 90-day all-cause mortality. Results: There was an association between baseline heparan sulfate level and mortality, with increasing mortality by baseline heparan sulfate tertile: lower tertile, 9.9% (95% confidence interval, 7.0-12.7%); middle tertile, 20.4% (15.6-25.0%); and upper tertile, 44.2% (35.6-51.6%) (P < 0.001; log-rank test), with an adjusted hazard ratio for interquartile range change in heparan sulfate of 3.12 (95% confidence interval, 2.18-4.46). We observed no effect of assigned fluid resuscitation strategy on endothelial glycocalyx degradation 24 hours after randomization. We observed no evidence of differential treatment effect of fluid resuscitation strategy based on baseline plasma heparan sulfate for 90-day mortality. Similar findings were observed using plasma syndecan-1 as an index of endothelial glycocalyx degradation. Conclusions: Endothelial glycocalyx degradation is a strong predictor of mortality in sepsis. Fluid resuscitation strategy had no impact on endothelial glycocalyx degradation, and there was no evidence for a differential effect of resuscitation strategy by baseline levels of endothelial glycocalyx degradation. Clinical trial registered with www.clinicaltrials.gov (NCT03434028).