Critical Care

official impact factor 4.60

Open Access Research

Angiopoietin-2 is increased in sepsis and inversely associated with nitric oxide-dependent microvascular reactivity

Joshua S Davis1,2, Tsin W Yeo1, Kim A Piera1, Tonia Woodberry1, David S Celermajer3, Dianne P Stephens4 and Nicholas M Anstey1,2*

Author Affiliations

1 International Health Division, Menzies School of Health Research and Charles Darwin University, Ellengowan Drive, Casuarina, Darwin, NT 0810, Australia

2 Department of Infectious Diseases, Royal Darwin Hospital, Rocklands Drive, Tiwi, Darwin, NT 0810, Australia

3 Department of Medicine, University of Sydney and Department of Cardiology, Royal Prince Alfred Hospital, Missenden Road, Camperdown, Sydney, NSW 2006, Australia

4 Intensive Care Unit, Royal Darwin Hospital, Rocklands Drive, Tiwi, Darwin, NT, 0810, Australia

For all author emails, please log on.

Critical Care 2010, 14:R89 doi:10.1186/cc9020


See related commentary by van Nieuw Amerongen and Groeneveld, http://ccforum.com/content/14/3/166

Published: 18 May 2010

Abstract

Introduction

Angiopoietin-2 (ang-2), an angiogenic peptide released by endothelial cell Weibel-Palade bodies (WPBs), increases endothelial activation and vascular permeability. Ang-2 is raised in severe sepsis but the mechanisms underlying this are not known. Nitric oxide (NO) inhibits WPB exocytosis, and bioavailability of endothelial NO is decreased in sepsis. We hypothesized that endothelial NO bioavailability would be inversely correlated with ang-2 concentrations in sepsis.

Methods

Plasma ang-2, vascular endothelial growth factor (VEGF) and endothelial-active cytokines were assessed in 83 patients with early sepsis and 41 hospital controls, and related to reactive hyperaemia-peripheral arterial tonometry, RH-PAT, a measure of endothelial NO bioavailability.

Results

Plasma Ang-2 was elevated in sepsis (median [interquartile range (IQR)], ng/ml: severe sepsis 12.4 [8.5-33.4], sepsis without organ failure 6.1 [5.0-10.4], controls 2.7 [2.2-3.6], P < 0.0001). It correlated inversely with RH-PAT (r = -0.38, P < 0.0001) and positively with IL-6 (r = 0.57, P < 0.0001) and degree of organ failure (sequential organ function assessment score) (r = 0.58, P < 0.0001). The correlation of ang-2 with RH-PAT persisted after controlling for sepsis severity. In a longitudinal mixed-effects model, recovery of RH-PAT over time was associated with decline in ang-2.

Conclusions

Ang-2 is elevated in proportion to sepsis severity, and inversely correlated with NO-dependent microvascular reactivity. Impaired endothelial NO bioavailability may contribute to increased endothelial cell release of ang-2, endothelial activation and capillary leak. Agents that increase endothelial NO bioavailability or inhibit WPB exocytosis and/or Ang-2 activity may have therapeutic potential in sepsis.