Hemoadsorption in the critically ill-Final results of the International CytoSorb Registry.

The aim of the current paper is to summarize the results of the International CytoSorb Registry. Data were collected on patients of the intensive care unit. The primary endpoint was actual in-hospital mortality compared to the mortality predicted by APACHE II score. The main secondary endpoints were SOFA scores, inflammatory biomarkers and overall evaluation of the general condition. 1434 patients were enrolled. Indications for hemoadsorption were sepsis/septic shock (N = 936); cardiac surgery perioperatively (N = 172); cardiac surgery postoperatively (N = 67) and "other" reasons (N = 259). APACHE-II-predicted mortality was 62.0±24.8%, whereas observed hospital mortality was 50.1%. Overall SOFA scores did not change but cardiovascular and pulmonary SOFA scores decreased by 0.4 [-0.5;-0.3] and -0.2 [-0.3;-0.2] points, respectively. Serum procalcitonin and C-reactive protein levels showed significant reduction: -15.4 [-19.6;-11.17] ng/mL; -17,52 [-70;44] mg/L, respectively. In the septic cohort PCT and IL-6 also showed significant reduction: -18.2 [-23.6;-12.8] ng/mL; -2.6 [-3.0;-2.2] pg/mL, respectively. Evaluation of the overall effect: minimal improvement (22%), much improvement (22%) and very much improvement (10%), no change observed (30%) and deterioration (4%). There was no significant difference in the primary outcome of mortality, but there were improvements in cardiovascular and pulmonary SOFA scores and a reduction in PCT, CRP and IL-6 levels. Trial registration: ClinicalTrials.gov Identifier: NCT02312024 (retrospectively registered).

Correlation of lung collapse and gas exchange - a computer tomographic study in sheep and pigs with atelectasis in otherwise normal lungs.

BACKGROUND: Atelectasis can provoke pulmonary and non-pulmonary complications after general anaesthesia. Unfortunately, there is no instrument to estimate atelectasis and prompt changes of mechanical ventilation during general anaesthesia. Although arterial partial pressure of oxygen (PaO2) and intrapulmonary shunt have both been suggested to correlate with atelectasis, studies yielded inconsistent results. Therefore, we investigated these correlations. METHODS: Shunt, PaO2 and atelectasis were measured in 11 sheep and 23 pigs with otherwise normal lungs. In pigs, contrasting measurements were available 12 hours after induction of acute respiratory distress syndrome (ARDS). Atelectasis was calculated by computed tomography relative to total lung mass (Mtotal). We logarithmically transformed PaO2 (lnPaO2) to linearize its relationships with shunt and atelectasis. Data are given as median (interquartile range). RESULTS: Mtotal was 768 (715-884) g in sheep and 543 (503-583) g in pigs. Atelectasis was 26 (16-47) % in sheep and 18 (13-23) % in pigs. PaO2 (FiO2 = 1.0) was 242 (106-414) mmHg in sheep and 480 (437-514) mmHg in pigs. Shunt was 39 (29-51) % in sheep and 15 (11-20) % in pigs. Atelectasis correlated closely with lnPaO2 (R2 = 0.78) and shunt (R2 = 0.79) in sheep (P-values<0.0001). The correlation of atelectasis with lnPaO2 (R2 = 0.63) and shunt (R2 = 0.34) was weaker in pigs, but R2 increased to 0.71 for lnPaO2 and 0.72 for shunt 12 hours after induction of ARDS. In both, sheep and pigs, changes in atelectasis correlated strongly with corresponding changes in lnPaO2 and shunt. DISCUSSION AND CONCLUSION: In lung-healthy sheep, atelectasis correlates closely with lnPaO2 and shunt, when blood gases are measured during ventilation with pure oxygen. In lung-healthy pigs, these correlations were significantly weaker, likely because pigs have stronger hypoxic pulmonary vasoconstriction (HPV) than sheep and humans. Nevertheless, correlations improved also in pigs after blunting of HPV during ARDS. In humans, the observed relationships may aid in assessing anaesthesia-related atelectasis.