TY - JOUR
T1 - Etiology of Myocardial Injury in Critically Ill Patients with Sepsis
T2 - A Cohort Study
AU - Frencken, Jos F
AU - van Smeden, Maarten
AU - van de Groep, Kirsten
AU - Ong, David S Y
AU - Klein Klouwenberg, Peter M C
AU - Juffermans, Nicole
AU - Bonten, Marc J M
AU - van der Poll, Tom
AU - Cremer, Olaf L
N1 - Funding Information:
Supported by the Center for Translational Molecular Medicine (https://www.lygature.org/ctmm-portfolio) MARS (Molecular Diagnosis and Risk Stratification of Sepsis) project (grant 04I-201).
Publisher Copyright:
© 2022 by the American Thoracic Society.
PY - 2022/5
Y1 - 2022/5
N2 - Rationale: Myocardial injury occurs frequently during sepsis and is independently associated with mortality. However, its etiology remains largely unknown. Objectives: To assess the relative contributions of hyperinflammation, activated coagulation, and endothelial dysfunction to myocardial injury in critically ill patients with sepsis. Methods: We included consecutive patients with sepsis presenting to two tertiary intensive care units in the Netherlands between 2011 and 2013. High-sensitivity cardiac troponin I as well as a wide range of plasma biomarkers related to inflammation, coagulation, and endothelial function were measured. Structural equation modeling was used to construct latent variables representing each of these pathophysiological constructs and to subsequently study their associations with troponin elevation while adjusting for confounders. Results: We analyzed 908 (88%) of 1,037 eligible patients, 553 (61%) of whom had raised high-sensitivity cardiac troponin I levels upon intensive care unit admission. The latent variables included interleukin (IL)-6, IL-8, and IL-1b for inflammation; platelet count, prothrombin time, and protein C for coagulation; and soluble E-selectin, intercellular adhesion molecule-1, and angiopoietin-2 for endothelial function. After adjustment for age and cardiovascular comorbidities, structural equation modeling analysis showed that activated coagulation was independently associated with elevated troponin during sepsis (standardized regression coefficient, 0.551; 95% confidence interval [CI], 0.257-0.845; P,0.001) whereas hyperinflammation and endothelial dysfunction were not (standardized regression coefficient, 20.161; 95% CI, 20.418 to 0.096 and 20.054; 95% CI, 20.168 to 0.060, respectively). Conclusions: Our findings suggest that myocardial injury during sepsis is mediated by systemic activation of coagulation rather than by circulating inflammatory mediators or activation of the endothelium. These findings may guide evaluation of strategies to protect the myocardium during sepsis. Clinical trial registered with clinicaltrials.gov (NCT01905033).
AB - Rationale: Myocardial injury occurs frequently during sepsis and is independently associated with mortality. However, its etiology remains largely unknown. Objectives: To assess the relative contributions of hyperinflammation, activated coagulation, and endothelial dysfunction to myocardial injury in critically ill patients with sepsis. Methods: We included consecutive patients with sepsis presenting to two tertiary intensive care units in the Netherlands between 2011 and 2013. High-sensitivity cardiac troponin I as well as a wide range of plasma biomarkers related to inflammation, coagulation, and endothelial function were measured. Structural equation modeling was used to construct latent variables representing each of these pathophysiological constructs and to subsequently study their associations with troponin elevation while adjusting for confounders. Results: We analyzed 908 (88%) of 1,037 eligible patients, 553 (61%) of whom had raised high-sensitivity cardiac troponin I levels upon intensive care unit admission. The latent variables included interleukin (IL)-6, IL-8, and IL-1b for inflammation; platelet count, prothrombin time, and protein C for coagulation; and soluble E-selectin, intercellular adhesion molecule-1, and angiopoietin-2 for endothelial function. After adjustment for age and cardiovascular comorbidities, structural equation modeling analysis showed that activated coagulation was independently associated with elevated troponin during sepsis (standardized regression coefficient, 0.551; 95% confidence interval [CI], 0.257-0.845; P,0.001) whereas hyperinflammation and endothelial dysfunction were not (standardized regression coefficient, 20.161; 95% CI, 20.418 to 0.096 and 20.054; 95% CI, 20.168 to 0.060, respectively). Conclusions: Our findings suggest that myocardial injury during sepsis is mediated by systemic activation of coagulation rather than by circulating inflammatory mediators or activation of the endothelium. These findings may guide evaluation of strategies to protect the myocardium during sepsis. Clinical trial registered with clinicaltrials.gov (NCT01905033).
KW - Sepsis
KW - coagulation
KW - inflammation
KW - troponin
KW - vascular endothelium
UR - http://www.scopus.com/inward/record.url?scp=85129781856&partnerID=8YFLogxK
U2 - 10.1513/AnnalsATS.202106-689OC
DO - 10.1513/AnnalsATS.202106-689OC
M3 - Article
C2 - 34784496
SN - 2329-6933
VL - 19
SP - 773
EP - 780
JO - Annals of the American Thoracic Society
JF - Annals of the American Thoracic Society
IS - 5
ER -