@article{MaguniaLedererVerbuechelnetal.2021,
  author    = {Magunia, Harry and Lederer, Simone and Verbuecheln, Raphael and Gilot, Bryant Joseph and Koeppen, Michael and Haeberle, Helene A. and Mirakaj, Valbona and Hofmann, Pascal and Marx, Gernot and Bickenbach, Johannes and Nohe, Boris and Lay, Michael and Spies, Claudia and Edel, Andreas and Schiefenh{\"o}vel, Fridtjof and Rahmel, Tim and Putensen, Christian and Sellmann, Timur and Koch, Thea and Brandenburger, Timo and Kindgen-Milles, Detlef and Brenner, Thorsten and Berger, Marc and Zacharowski, Kai and Adam, Elisabeth and Posch, Matthias and Moerer, Onnen and Scheer, Christian S. and Sedding, Daniel and Weigand, Markus A. and Fichtner, Falk and Nau, Carla and Pr{\"a}tsch, Florian and Wiesmann, Thomas and Koch, Christian and Schneider, Gerhard and Lahmer, Tobias and Straub, Andreas and Meiser, Andreas and Weiss, Manfred and Jungwirth, Bettina and Wappler, Frank and Meybohm, Patrick and Herrmann, Johannes and Malek, Nisar and Kohlbacher, Oliver and Biergans, Stephanie and Rosenberger, Peter},
  title     = {Machine learning identifies ICU outcome predictors in a multicenter COVID-19 cohort},
  series = {Critical Care},
  volume    = {25},
  journal   = {Critical Care},
  doi       = {10.1186/s13054-021-03720-4},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-306766},
  year      = {2021},
  abstract  = {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.},
  language  = {en}
}
@article{HoppeKhanMeybohmetal.2023,
  author    = {Hoppe, K. and Khan, E. and Meybohm, P. and Riese, T.},
  title     = {Mechanical power of ventilation and driving pressure: two undervalued parameters for pre extracorporeal membrane oxygenation ventilation and during daily management?},
  series = {Critical Care},
  volume    = {27},
  journal   = {Critical Care},
  doi       = {10.1186/s13054-023-04375-z},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-357181},
  year      = {2023},
  abstract  = {The current ARDS guidelines highly recommend lung protective ventilation which include plateau pressure (Pplat < 30 cm H\(_2\)O), positive end expiratory pressure (PEEP > 5 cm H2O) and tidal volume (Vt of 6 ml/kg) of predicted body weight. In contrast, the ELSO guidelines suggest the evaluation of an indication of veno-venous extracorporeal membrane oxygenation (ECMO) due to hypoxemic or hypercapnic respiratory failure or as bridge to lung transplantation. Finally, these recommendations remain a wide range of scope of interpretation. However, particularly patients with moderate-severe to severe ARDS might benefit from strict adherence to lung protective ventilation strategies. Subsequently, we discuss whether extended physiological ventilation parameter analysis might be relevant for indication of ECMO support and can be implemented during the daily routine evaluation of ARDS patients. Particularly, this viewpoint focus on driving pressure and mechanical power.},
  language  = {en}
}
@article{SitterPecksRuedigeretal.2022,
  author    = {Sitter, Magdalena and Pecks, Ulrich and R{\"u}diger, Mario and Friedrich, Sabine and Fill Malfertheiner, Sara and Hein, Alexander and K{\"o}nigbauer, Josefine T. and Becke-Jakob, Karin and Z{\"o}llkau, Janine and Ramsauer, Babett and Rathberger, Katharina and Pontones, Constanza A. and Kraft, Katrina and Meybohm, Patrick and H{\"a}rtel, Christoph and Kranke, Peter},
  title     = {Pregnant and postpartum women requiring intensive care treatment for COVID-19 — first data from the CRONOS-registry},
  series = {Journal of Clinical Medicine},
  volume    = {11},
  journal   = {Journal of Clinical Medicine},
  number    = {3},
  issn      = {2077-0383},
  doi       = {10.3390/jcm11030701},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-255257},
  year      = {2022},
  abstract  = {(1) Background: Data on coronavirus 2 infection during pregnancy vary. We aimed to describe maternal characteristics and clinical presentation of SARS-CoV-2 positive women requiring intensive care treatment for COVID-19 during pregnancy and postpartum period based on data of a comprehensive German surveillance system in obstetric patients. (2) Methods: Data from COVID-19 Related Obstetric and Neonatal Outcome Study (CRONOS), a prospective multicenter registry for SARS-CoV-2 positive pregnant women, was analyzed with respect to ICU treatment. All women requiring intensive care treatment for COVID-19 were included and compared regarding maternal characteristics, course of disease, as well as maternal and neonatal outcomes. (3) Results: Of 2650 cases in CRONOS, 101 women (4\%) had a documented ICU stay. Median maternal age was 33 (IQR, 30-36) years. COVID-19 was diagnosed at a median gestational age of 33 (IQR, 28-35) weeks. As the most invasive form of COVID-19 treatment interventions, patients received either continuous monitoring of vital signs without further treatment requirement (n = 6), insufflation of oxygen (n = 30), non-invasive ventilation (n = 22), invasive ventilation (n = 28), or escalation to extracorporeal membrane oxygenation (n = 15). No significant clinical differences were identified between patients receiving different forms of ventilatory support for COVID-19. Prevalence of preterm delivery was significantly higher in women receiving invasive respiratory treatments. Four women died of COVID-19 and six fetuses were stillborn. (4) Conclusions: Our cohort shows that progression of COVID-19 is rare in pregnant and postpartum women treated in the ICU. Preterm birth rate is high and COVID-19 requiring respiratory support increases the risk of poor maternal and neonatal outcome.},
  language  = {en}
}