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.

Background: The COVID-19 pandemic reached Germany in spring 2020. No proven treatment for SARS-CoV-2 was available at that time, especially for severe COVID-19-induced ARDS. We determined whether the infusion of mesenchymal stromal cells (MSCs) would help to improve pulmonary function and overall outcome in patients with severe COVID-19 ARDS. We offered MSC infusion as an extended indication to all critically ill COVID-19 patients with a Horovitz index <100. We treated 5 out of 23 patients with severe COVID-19 ARDS with an infusion of MSCs. One million MSCs/kg body weight was infused over 30 minutes, and the process was repeated in 3 patients twice and in 2 patients 3 times. Result: Four out of 5 MSC-treated patients compared to 50% of control patients (9 out of 18) received ECMO support (80%). The MSC group showed a higher Murray score on admission than control patients, reflecting more severe pulmonary compromise (3.5 ± 0.2 versus 2.8 ± 0.3). MSC infusion was safe and well tolerated. The MSC group had a significantly higher Horovitz score on discharge than the control group. Compared to controls, patients with MSC treatment showed a significantly lower Murray score upon discharge than controls. In the MSC group, 4 out of 5 patients (80%) survived to discharge and exhibited good pulmonary function, whereas only 8 out of 18 patients (45%) in the control group survived to discharge. Conclusion: MSC infusion is a safe treatment for COVID-19 ARDS that improves pulmonary function and overall outcome in this patient population.

In critically ill patients, hemodynamic disturbances are common and often lead to a detrimental outcome. Frequently, invasive hemodynamic monitoring is required for patients who are hemodynamically unstable. Although the pulmonary artery catheter enables a comprehensive assessment of the hemodynamic profile, this technique carries a substantial inherent risk of complications. Other less invasive techniques do not offer a full range of results to guide detailed hemodynamic therapies. An alternative with a lower risk profile is transthoracic echocardiography (TTE) or transesophageal echocardiography (TEE). After training, intensivists can obtain similar parameters on the hemodynamic profile using echocardiography, such as stroke volume and ejection fraction of the right and left ventricles, an estimate of the pulmonary artery wedge pressure, and cardiac output. Here, we will review individual echocardiography techniques that will help the intensivist obtain a comprehensive assessment of the hemodynamic profile using echocardiography.

Background: Percutaneous edge‐to‐edge mitral valve repair ( PMVR ) has become an established treatment option for mitral regurgitation in patients not eligible for surgical repair. Currently, most procedures are performed under general anesthesia ( GA ). An increasing number of centers, however, are performing the procedure under deep sedation ( DS ). Here, we compared patients undergoing PMVR with GA or DS . Methods and Results: A total of 271 consecutive patients underwent PMVR at our institution between May 2014 and December 2016. Seventy‐two procedures were performed under GA and 199 procedures under DS . We observed that in the DS group, doses of propofol (743±228 mg for GA versus 369±230 mg for DS , P <0.001) and norepinephrine (1.1±1.6 mg for GA versus 0.2±0.3 mg for DS , P <0.001) were significantly lower. Procedure time, fluoroscopy time, and dose area product were significantly higher in the GA group. There was no significant difference between GA and DS with respect to overall bleeding complications, postinterventional pneumonia (4% for GA versus 5% for DS ), or C‐reactive protein levels (361±351 nmol/L for GA versus 278±239 nmol/L for DS). Significantly fewer patients with DS needed a postinterventional stay in the intensive care unit (96% for GA versus 19% for DS , P <0.001). Importantly, there was no significant difference between DS and GA regarding intrahospital or 6‐month mortality. Conclusions: DS for PMVR is safe and feasible. No disadvantages with respect to procedural outcome or complications in comparison to GA were observed. Applying DS may simplify the PMVR procedure.