Refine
Document Type
- Article (2)
Language
- English (2) (remove)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- ARDS (1)
- COVID-19 (1)
- Critical care (1)
- Outcome (1)
- Prognostic models (1)
- cardiac surgery (1)
- humoral factors (1)
- molecular mechanisms (1)
- propofol anesthesia (1)
- remote ischemic preconditioning (1)
Institute
- Medizin (2)
Background: Intensive Care Resources are heavily utilized during the COVID-19 pandemic. However, risk stratification and prediction of SARS-CoV-2 patient clinical outcomes upon ICU admission remain inadequate. This study aimed to develop a machine learning model, based on retrospective & prospective clinical data, to stratify patient risk and predict ICU survival and outcomes. Methods: A Germany-wide electronic registry was established to pseudonymously collect admission, therapeutic and discharge information of SARS-CoV-2 ICU patients retrospectively and prospectively. Machine learning approaches were evaluated for the accuracy and interpretability of predictions. The Explainable Boosting Machine approach was selected as the most suitable method. Individual, non-linear shape functions for predictive parameters and parameter interactions are reported. Results: 1039 patients were included in the Explainable Boosting Machine model, 596 patients retrospectively collected, and 443 patients prospectively collected. The model for prediction of general ICU outcome was shown to be more reliable to predict “survival”. Age, inflammatory and thrombotic activity, and severity of ARDS at ICU admission were shown to be predictive of ICU survival. Patients’ age, pulmonary dysfunction and transfer from an external institution were predictors for ECMO therapy. The interaction of patient age with D-dimer levels on admission and creatinine levels with SOFA score without GCS were predictors for renal replacement therapy. Conclusions: Using Explainable Boosting Machine analysis, we confirmed and weighed previously reported and identified novel predictors for outcome in critically ill COVID-19 patients. Using this strategy, predictive modeling of COVID-19 ICU patient outcomes can be performed overcoming the limitations of linear regression models. Trial registration “ClinicalTrials” (clinicaltrials.gov) under NCT04455451.
In contrast to several smaller studies, which demonstrate that remote ischemic preconditioning (RIPC) reduces myocardial injury in patients that undergo cardiovascular surgery, the RIPHeart study failed to demonstrate beneficial effects of troponin release and clinical outcome in propofol-anesthetized cardiac surgery patients. Therefore, we addressed the potential biochemical mechanisms triggered by RIPC. This is a predefined prospective sub-analysis of the randomized and controlled RIPHeart study in cardiac surgery patients (n = 40) that was recently published. Blood samples were drawn from patients prior to surgery, after RIPC of four cycles of 5 min arm ischemia/5 min reperfusion (n = 19) and the sham (n = 21) procedure, after connection to cardiopulmonary bypass (CPB), at the end of surgery, 24 h postoperatively, and 48 h postoperatively for the measurement of troponin T, macrophage migration inhibitory factor (MIF), stromal cell-derived factor 1 (CXCL12), IL-6, CXCL8, and IL-10. After RIPC, right atrial tissue samples were taken for the measurement of extracellular-signal regulated kinase (ERK1/2), protein kinase B (AKT), Glycogen synthase kinase 3 (GSK-3β), protein kinase C (PKCε), and MIF content. RIPC did not significantly reduce the troponin release when compared with the sham procedure. MIF serum levels intraoperatively increased, peaking at intensive care unit (ICU) admission (with an increase of 48.04%, p = 0.164 in RIPC; and 69.64%, p = 0.023 over the baseline in the sham procedure), and decreased back to the baseline 24 h after surgery, with no differences between the groups. In the right atrial tissue, MIF content decreased after RIPC (1.040 ± 1.032 Arbitrary units [au] in RIPC vs. 2.028 ± 1.631 [au] in the sham procedure, p < 0.05). CXCL12 serum levels increased significantly over the baseline at the end of surgery, with no differences between the groups. ERK1/2, AKT, GSK-3β, and PKCɛ phosphorylation in the right atrial samples were no different between the groups. No difference was found in IL-6, CXCL8, and IL10 serum levels between the groups. In this cohort of cardiac surgery patients that received propofol anesthesia, we could not show a release of potential mediators of signaling, nor an effect on the inflammatory response, nor an activation of well-established protein kinases after RIPC. Based on these data, we cannot exclude that confounding factors, such as propofol, may have interfered with RIPC.