Endotheliopathy is associated with slower liberation from mechanical ventilation: a cohort study

Abstract

Background: Endotheliopathy is suggested as pivotal pathophysiology of sepsis and trauma-associated organ failure, but its role in acute respiratory failure is not yet determined. We investigated if endotheliopathy biomarkers at ICU admission are associated with illness severity and clinical outcomes in patients with acute respiratory failure requiring mechanical ventilation.

Methods: We conducted a prospective single-center cohort study including 459 mechanically ventilated adults at ICU admission. Plasma levels of three endotheliopathy biomarkers were measured at ICU admission: Syndecan-1, soluble Thrombomodulin (sTM), and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1). The primary outcome was the rate of liberation from mechanical ventilation, which is presented together with the rate of the competing risk of death while still on mechanical ventilation. Secondary outcomes were PaO₂/FiO₂-ratios on admission and on last measurement in patients dying within five days, and 30-day all-cause mortality. The primary outcome and 30-day all-cause mortality were analyzed using Cox regression, controlled for gender, age, chronic obstructive pulmonary disease, septic shock, heart failure, PaO₂/FiO₂-ratio at admission, respiratory infection, acute kidney injury, and bilirubin. PaO₂/FiO₂-ratios were analyzed using linear regression, controlled for age, chronic obstructive pulmonary disease, respiratory infection, and shock.

Results: Patients with high sTM were liberated from mechanical ventilation at a lower rate (adjusted hazard ratio (HR) 0.71, for an increase from the 25th to the 75th percentile, 95% confidence interval (CI) 0.54-0.93, p = 0.01). Patients with high PECAM-1 were liberated from mechanical ventilation at a lower rate, but only during the first 5 days (adjusted HR 0.72, for an increase from the 25th to the 75th percentile, 95% CI 0.58-0.9, p < 0.01). High levels of Syndecan-1 and PECAM-1 were associated with a higher rate of death while still on mechanical ventilation. sTM and PECAM-1 were negatively associated with PaO₂/FiO₂-ratio at ICU admission and no biomarker was associated with last measured PaO₂/FiO₂-ratio. High levels of all biomarkers were associated with higher 30-day all-cause mortality.

Conclusion: In acute respiratory failure, endotheliopathy biomarkers are associated with lower rates of liberation from mechanical ventilation, hypoxemia at ICU admission, and 30-day all-cause mortality.

Keywords: Endothelium; Observational study; Platelet Endothelial Cell Adhesion Molecule-1; Respiratory insufficiency/physiopathology; Syndecan-1; Thrombomodulin; Vascular.

Fig. 1

Flowchart of the inclusion/exclusion-process

Fig. 2

Cumulative incidences of liberation from mechanical ventilation and the competing risk of death on mechanical ventilation, divided on the grouped versions of Syndecan-1, soluble Thrombomodulin (sTM) and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1) respectively

Fig. 3

Adjusted hazard ratios (HRs) for liberation from mechanical ventilation and the competing risk of death on mechanical ventilation with point estimates and 95% confidence intervals against the levels of Syndecan-1, soluble Thrombomodulin (sTM) and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1) respectively. The HRs are adjusted for age, gender, chronic obstructive pulmonary disease, heart failure, bilirubin, PaO₂/FiO₂-ratio, septic shock and respiratory infection. Vertical lines inserted at the 25th and 75th percentiles of each biomarker for illustrative purposes

Fig. 4

Absolute risks of liberation from mechanical ventilation and of death on mechanical ventilation comparing patients with levels at the 25th percentile with patients with levels at the 75th percentile of Syndecan-1, Soluble Thrombomodulin (sTM) and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1) respectively. Solid lines are point estimates and dashed lines represent 95% confidence intervals. The absolute risks are calculated based on the cause specific hazard rates for the represented outcomes, obtained from multivariable competing risks Cox regressions introducing Syndecan-1, sTM or PECAM-1 as continuous variables. The absolute risks are calculated for a 71-year old female, with no chronic obstructive pulmonary disease, no chronic heart failure, no respiratory infection at baseline, no acute kidney injury at baseline, with a Bilirubin of 10.7 µmol/L and a PaO₂/FiO₂-ratio of 21.4 kPa

Fig. 5

Plotted associations of Syndecan-1, soluble Thrombomodulin (sTM) and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1) as categorical variables. 1st quartile is used as reference. The HRs are adjusted for age, gender, chronic obstructive pulmonary disease, heart failure, bilirubin, PaO₂/FiO₂-ratio, septic shock and respiratory infection. HR hazard ratio, 95% CI 95% confidence interval

Fig. 6

Adjusted hazard ratios (HRs) for 30-day all-cause mortality with point estimates and 95% confidence intervals against the levels of Syndecan-1, soluble Thrombomodulin (sTM) and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1) respectively. The HRs are adjusted for age, gender, chronic obstructive pulmonary disease, heart failure, bilirubin, PaO₂/FiO₂-ratio, septic shock and respiratory infection. Vertical lines inserted at the 25th and 75th percentiles of each biomarker for illustrative purposes