• John P Reilly, Fan Wang, Tiffanie K Jones, Jessica A Palakshappa, Brian J Anderson, ShashatyMichael G SMGSPulmonary, Allergy, and Critical Care Medicine Division, University of Pennsylvania Perelman School of Medicine, 3600 Spruce Street 5039 Gates Building, Philadelphia, PA, 19104, USA., Thomas G Dunn, Erik D Johansson, Thomas R Riley, Brian Lim, Jason Abbott, IttnerCaroline A GCAGPulmonary, Allergy, and Critical Care Medicine Division, University of Pennsylvania Perelman School of Medicine, 3600 Spruce Street 5039 Gates Building, Philadelphia, PA, 19104, USA., Edward Cantu, Xihong Lin, Carmen Mikacenic, Mark M Wurfel, David C Christiani, Carolyn S Calfee, Michael A Matthay, Jason D Christie, Rui Feng, and Nuala J Meyer.
• Pulmonary, Allergy, and Critical Care Medicine Division, University of Pennsylvania Perelman School of Medicine, 3600 Spruce Street 5039 Gates Building, Philadelphia, PA, 19104, USA.
• Intensive Care Med. 2018 Nov 1; 44 (11): 184918581849-1858.

PurposeA causal biomarker for acute respiratory distress syndrome (ARDS) could fuel precision therapy options. Plasma angiopoietin-2 (ANG2), a vascular permeability marker, is a strong candidate on the basis of experimental and observational evidence. We used genetic causal inference methods-Mendelian randomization and mediation-to infer potential effects of plasma ANG2.MethodsWe genotyped 703 septic subjects, measured ICU admission plasma ANG2, and performed a quantitative trait loci (QTL) analysis to determine variants in the ANGPT2 gene associated with plasma ANG2 (p < 0.005). We then used linear regression and post-estimation analysis to genetically predict plasma ANG2 and tested genetically predicted ANG2 for ARDS association using logistic regression. We estimated the proportion of the genetic effect explained by plasma ANG2 using mediation analysis.ResultsPlasma ANG2 was strongly associated with ARDS (OR 1.59 (95% CI 1.35, 1.88) per log). Five ANGPT2 variants were associated with ANG2 in European ancestry subjects (n = 404). Rs2442608C, the most extreme cis QTL (coefficient 0.22, 95% CI 0.09-0.36, p = 0.001), was associated with higher ARDS risk: adjusted OR 1.38 (95% CI 1.01, 1.87), p = 0.042. No significant QTL were identified in African ancestry subjects. Genetically predicted plasma ANG2 was associated with ARDS risk: adjusted OR 2.25 (95% CI 1.06-4.78), p = 0.035. Plasma ANG2 mediated 34% of the rs2442608C-related ARDS risk.ConclusionsIn septic European ancestry subjects, the strongest ANG2-determining ANGPT2 genetic variant is associated with higher ARDS risk. Plasma ANG2 may be a causal factor in ARDS development. Strategies to reduce plasma ANG2 warrant testing to prevent or treat sepsis-associated ARDS.

33 keywords...

hide…