Assessing SOFA score trajectories in sepsis using machine learning: A pragmatic approach to improve the accuracy of mortality prediction.

INTRODUCTION: An increasing amount of longitudinal health data is available on critically ill septic patients in the age of digital medicine, including daily sequential organ failure assessment (SOFA) score measurements. Thus, the assessment in sepsis focuses increasingly on the evaluation of the individual disease's trajectory. Machine learning (ML) algorithms may provide a promising approach here to improve the evaluation of daily SOFA score dynamics. We tested whether ML algorithms can outperform the conventional ΔSOFA score regarding the accuracy of 30-day mortality prediction. METHODS: We used the multicentric SepsisDataNet.NRW study cohort that prospectively enrolled 252 sepsis patients between 03/2018 and 09/2019 for training ML algorithms, i.e. support vector machine (SVM) with polynomial kernel and artificial neural network (aNN). We used the Amsterdam UMC database covering 1,790 sepsis patients for external and independent validation. RESULTS: Both SVM (AUC 0.84; 95% CI: 0.71-0.96) and aNN (AUC 0.82; 95% CI: 0.69-0.95) assessing the SOFA scores of the first seven days led to a more accurate prognosis of 30-day mortality compared to the ΔSOFA score between day 1 and 7 (AUC 0.73; 95% CI: 0.65-0.80; p = 0.02 and p = 0.05, respectively). These differences were even more prominent the shorter the time interval considered. Using the SOFA scores of day 1 to 3 SVM (AUC 0.82; 95% CI: 0.68 0.95) and aNN (AUC 0.80; 95% CI: 0.660.93) led to a more accurate prognosis of 30-day mortality compared to the ΔSOFA score (AUC 0.66; 95% CI: 0.58-0.74; p < 0.01 and p < 0.01, respectively). Strikingly, all these findings could be confirmed in the independent external validation cohort. CONCLUSIONS: The ML-based algorithms using daily SOFA scores markedly improved the accuracy of mortality compared to the conventional ΔSOFA score. Therefore, this approach could provide a promising and automated approach to assess the individual disease trajectory in sepsis. These findings reflect the potential of incorporating ML algorithms as robust and generalizable support tools on intensive care units.

Key characteristics impacting survival of COVID-19 extracorporeal membrane oxygenation.

BACKGROUND: Severe COVID-19 induced acute respiratory distress syndrome (ARDS) often requires extracorporeal membrane oxygenation (ECMO). Recent German health insurance data revealed low ICU survival rates. Patient characteristics and experience of the ECMO center may determine intensive care unit (ICU) survival. The current study aimed to identify factors affecting ICU survival of COVID-19 ECMO patients. METHODS: 673 COVID-19 ARDS ECMO patients treated in 26 centers between January 1st 2020 and March 22nd 2021 were included. Data on clinical characteristics, adjunct therapies, complications, and outcome were documented. Block wise logistic regression analysis was applied to identify variables associated with ICU-survival. RESULTS: Most patients were between 50 and 70 years of age. PaO2/FiO2 ratio prior to ECMO was 72 mmHg (IQR: 58-99). ICU survival was 31.4%. Survival was significantly lower during the 2nd wave of the COVID-19 pandemic. A subgroup of 284 (42%) patients fulfilling modified EOLIA criteria had a higher survival (38%) (p = 0.0014, OR 0.64 (CI 0.41-0.99)). Survival differed between low, intermediate, and high-volume centers with 20%, 30%, and 38%, respectively (p = 0.0024). Treatment in high volume centers resulted in an odds ratio of 0.55 (CI 0.28-1.02) compared to low volume centers. Additional factors associated with survival were younger age, shorter time between intubation and ECMO initiation, BMI > 35 (compared to < 25), absence of renal replacement therapy or major bleeding/thromboembolic events. CONCLUSIONS: Structural and patient-related factors, including age, comorbidities and ECMO case volume, determined the survival of COVID-19 ECMO. These factors combined with a more liberal ECMO indication during the 2nd wave may explain the reasonably overall low survival rate. Careful selection of patients and treatment in high volume ECMO centers was associated with higher odds of ICU survival. TRIAL REGISTRATION: Registered in the German Clinical Trials Register (study ID: DRKS00022964, retrospectively registered, September 7th 2020, https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00022964 .

Complement inhibition initiated recovery of a severe myasthenic crisis with COVID-19.

We report on a patient with refractory Myasthenia gravis with acetylcholine receptor antibodies with two prior myasthenic crises suffering from COVID-19 with rapid evolving weakness and respiratory failure. Respiratory failure developed and prolonged mechanical ventilation was necessary. After plasmapheresis, residual, severe generalized and bulbar weakness persisted. Complement inhibition with eculizumab was, therefore, introduced and lead to rapid recovery. In refractory myasthenic crisis individualised therapies could be successful.

Postoperative delirium - treatment and prevention.

PURPOSE OF REVIEW: Postoperative delirium (POD) is one of the most severe complications after surgery.The consequences are dramatic: longer hospitalization, a doubling of mortality and almost all cases develop permanent, yet subtle, cognitive deficits specific to everyday life. Actually, no global guideline with standardized concepts of management exists. Advances in prevention, diagnosis and treatment can improve recognition and risk stratification of delirium and its consequences. RECENT FINDINGS: Management of POD is a multiprofessional approach and consists of different parts: First, the detection of high-risk patients with a validated tool, preventive nonpharmacological concepts and an intraoperative anesthetic management plan that is individualized to the older patient (e.g. avoiding large swings in blood pressure, vigilance in maintaining normothermia, ensuring adequate analgesia and monitoring of anesthetic depth). In addition to preventive standards, treatment and diagnostic concepts must also be available, both pharmaceutical and nonpharmacological. SUMMARY: Not every POD can be prevented. It is important to detect patients with high risk for POD and have standardized concepts of management. The most important predisposing risk factors are a higher age, preexisting cognitive deficits, multimorbidity and an associated prodelirious polypharmacy. In view of demographic change, the implementation of multidisciplinary approaches to pharmacological and nonpharmacological POD management is highly recommended.

Inhaled carbon monoxide protects time-dependently from loss of hypoxic pulmonary vasoconstriction in endotoxemic mice.

BACKGROUND: Inhaled carbon monoxide (CO) appears to have beneficial effects on endotoxemia-induced impairment of hypoxic pulmonary vasoconstriction (HPV). This study aims to specify correct timing of CO application, it's biochemical mechanisms and effects on inflammatory reactions. METHODS: Mice (C57BL/6; n = 86) received lipopolysaccharide (LPS, 30 mg/kg) intraperitoneally and subsequently breathed 50 ppm CO continuously during defined intervals of 3, 6, 12 or 18 h. Two control groups received saline intraperitoneally and additionally either air or CO, and one control group received LPS but breathed air only. In an isolated lung perfusion model vasoconstrictor response to hypoxia (FiO2 = 0.01) was quantified by measurements of pulmonary artery pressure. Pulmonary capillary pressure was estimated by double occlusion technique. Further, inflammatory plasma cytokines and lung tissue mRNA of nitric-oxide-synthase-2 (NOS-2) and heme oxygenase-1 (HO-1) were measured. RESULTS: HPV was impaired after LPS-challenge (p < 0.01). CO exposure restored HPV-responsiveness if administered continuously for full 18 h, for the first 6 h and if given in the interval between the 3(rd) and 6(th) hour after LPS-challenge (p < 0.05). Preserved HPV was attributable to recovered arterial resistance and associated with significant reduction in NOS-2 mRNA when compared to controls (p < 0.05). We found no effects on inflammatory plasma cytokines. CONCLUSION: Low-dose CO prevented LPS-induced impairment of HPV in a time-dependent manner, associated with a decreased NOS-2 expression.