Objectives: The aims of the study were to compare the consumption of blood products before and af... more Objectives: The aims of the study were to compare the consumption of blood products before and after the implementation of a bleeding management algorithm in patients undergoing liver transplantation and to determine the feasibility of a multicentre, randomized study. Background: Liver transplantation remains the only curative therapy for patients with end-stage liver disease, but it carries a high risk of surgical bleeding. Materials and Methods: Retrospective study of patients treated before (group 1) and after (group 2) implementation of a haemostatic algorithm guided by viscoelastic testing, including use of lyophilized coagulation factor concentrates (prothrombin complex and fibrinogen concentrates). Primary outcome was the number of units of blood products transfused in 24 h after surgery. Secondary outcomes included hospital stay, mortality, and cost. Results: Data from 30 consecutive patients was analysed; 14 in group 1 and 16 in group 2. Baseline data were similar between g...
Background: Differences in blood pressure can influence the risk of brain ischemia, perioperative... more Background: Differences in blood pressure can influence the risk of brain ischemia, perioperative complications, and postoperative neurocognitive function in patients undergoing carotid endarterectomy (CEA). Methods: In this single-center trial, patients scheduled for CEA under general anesthesia were randomized into an intervention group receiving near-infrared spectroscopy (NIRS)-guided blood pressure management during carotid cross-clamping and a control group receiving standard care. The primary endpoint was postoperative neurocognitive function assessed before surgery, on postoperative days 1 and 7, and eight weeks after surgery. Perioperative complications and cerebral autoregulatory capacity were secondary endpoints. Results: Systolic blood pressure (p < 0.001) and norepinephrine doses (89 (54–122) vs. 147 (116–242) µg; p < 0.001) during carotid cross-clamping were lower in the intervention group. No group differences in postoperative neurocognitive function were observ...
Many previous studies claim to have developed machine learning models that diagnose COVID-19 from... more Many previous studies claim to have developed machine learning models that diagnose COVID-19 from blood tests. However, we hypothesize that changes in the underlying distribution of the data, so called domain shifts, affect the predictive performance and reliability and are a reason for the failure of such machine learning models in clinical application. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (way of taking samples, laboratory procedures), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. We investigate whether domain shifts are present in COVID-19 datasets and how they affect machine learning methods. We further set out to estimate the mortality risk based on routinely acquired blood tests in a hospital setting throughout pandemics and under domain shifts. We reveal domain ...
For many years, physicians’ approach to the transfusion of allogeneic red blood cells (RBC) was n... more For many years, physicians’ approach to the transfusion of allogeneic red blood cells (RBC) was not individualized. It was accepted that a hemoglobin concentration (Hb) of less than 10 g/dL was a general transfusion threshold and the majority of patients were transfused immediately. In recent years, there has been increasing evidence that even significantly lower hemoglobin concentrations can be survived in the short term without sequelae. This somehow contradicts the observation that moderate or mild anemia is associated with relevant long-term morbidity and mortality. To resolve this apparent contradiction, it must be recognized that we have to avoid acute anemia or treat it by alternative methods. The aim of this article is to describe the physiological limits of acute anemia, match these considerations with clinical realities, and then present “patient blood management” (PBM) as the therapeutic concept that can prevent both anemia and unnecessary transfusion of RBC concentrates ...
Visually evoked potentials (VEPs) are widely used for diagnoses of different neurological disease... more Visually evoked potentials (VEPs) are widely used for diagnoses of different neurological diseases. Interestingly, there is limited research about the impact of the stimulus color onto the evoked response. Therefore, in our study we investigated the possibility of automatically classifying the stimulus color. The visual stimuli were selected to be red/black and green/black checkerboard patterns with equal light density. Both of these stimuli were presented in a random manner to nine subjects, while the electroencephalogram was recorded at the occipital lobe. After pre-processing and aligning the evoked potentials, an artificial neural network with one hidden layer was used to investigate the general possibility to automatically classify the stimulus color in three different settings. First, color classification with individually trained models, color classification with a common model, and color classification for each individual volunteer with a model trained on the data of the rem...
Summary Background In December 2019, the new virus infection coronavirus disease 2019 (COVID-19) ... more Summary Background In December 2019, the new virus infection coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged. Simple clinical risk scores may improve the management of COVID-19 patients. Therefore, the aim of this pilot study was to evaluate the quick Sequential Organ Failure Assessment (qSOFA) score, which is well established for other diseases, as an early risk assessment tool predicting a severe course of COVID-19. Methods We retrospectively analyzed data from adult COVID-19 patients hospitalized between March and July 2020. A critical disease progress was defined as admission to intensive care unit (ICU) or death. Results Of 64 COVID-19 patients, 33% (21/64) had a critical disease progression from which 13 patients had to be transferred to ICU. The COVID-19-associated mortality rate was 20%, increasing to 39% after ICU admission. All patients without a critical progress had a qSOFA score ≤ 1 at admission. Pat...
We investigate machine learning models that identify COVID-19 positive patients and estimate the ... more We investigate machine learning models that identify COVID-19 positive patients and estimate the mortality risk based on routinely acquired blood tests in a hospital setting. However, during pandemics or new outbreaks, disease and testing characteristics change, thus we face domain shifts. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (taking samples, laboratory), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. To countermand this effect, we propose methods that first identify domain shifts and then reverse their negative effects on the model performance. Frequent re-training and reassessment, as well as stronger weighting of more recent samples, keeps model performance and credibility at a high level over time. Our diagnosis models are constructed and tested on large-scale data ...
Background Uncertainty about the optimal respiratory support strategies in critically ill COVID-1... more Background Uncertainty about the optimal respiratory support strategies in critically ill COVID-19 patients is widespread. While the risks and benefits of noninvasive techniques versus early invasive mechanical ventilation (IMV) are intensely debated, actual evidence is lacking. We sought to assess the risks and benefits of different respiratory support strategies, employed in intensive care units during the first months of the COVID-19 pandemic on intubation and intensive care unit (ICU) mortality rates. Methods Subanalysis of a prospective, multinational registry of critically ill COVID-19 patients. Patients were subclassified into standard oxygen therapy ≥10 L/min (SOT), high-flow oxygen therapy (HFNC), noninvasive positive-pressure ventilation (NIV), and early IMV, according to the respiratory support strategy employed at the day of admission to ICU. Propensity score matching was performed to ensure comparability between groups. Results Initially, 1421 patients were assessed for...
Objective The diagnosis of COVID-19 is based on the detection of SARS-CoV-2 in respiratory secret... more Objective The diagnosis of COVID-19 is based on the detection of SARS-CoV-2 in respiratory secretions, blood, or stool. Currently, reverse transcription polymerase chain reaction (RT-PCR) is the most commonly used method to test for SARS-CoV-2. Methods In this retrospective cohort analysis, we evaluated whether machine learning could exclude SARS-CoV-2 infection using routinely available laboratory values. A Random Forests algorithm with 28 unique features was trained to predict the RT-PCR results. Results Out of 12,848 patients undergoing SARS-CoV-2 testing, routine blood tests were simultaneously performed in 1357 patients. The machine learning model could predict SARS-CoV-2 test results with an accuracy of 86% and an area under the receiver operating characteristic curve of 0.74. Conclusion Machine learning methods can reliably predict a negative SARS-CoV-2 RT-PCR test result using standard blood tests.
BackgroundThe ability to predict transfusions arising during hospital admission might enable econ... more BackgroundThe ability to predict transfusions arising during hospital admission might enable economized blood supply management and might furthermore increase patient safety by ensuring a sufficient stock of red blood cells (RBCs) for a specific patient. We therefore investigated the precision of four different machine learning–based prediction algorithms to predict transfusion, massive transfusion, and the number of transfusions in patients admitted to a hospital.Study Design and MethodsThis was a retrospective, observational study in three adult tertiary care hospitals in Western Australia between January 2008 and June 2017. Primary outcome measures for the classification tasks were the area under the curve for the receiver operating characteristics curve, the F1 score, and the average precision of the four machine learning algorithms used: neural networks (NNs), logistic regression (LR), random forests (RFs), and gradient boosting (GB) trees.ResultsUsing our four predictive model...
Background Acute normovolemic hemodilution reduces the circulating erythrocyte mass and, thus, th... more Background Acute normovolemic hemodilution reduces the circulating erythrocyte mass and, thus, the hemoglobin concentration. After extreme acute normovolemic hemodilution to the critical hemoglobin concentration (Hbcrit), oxygen demand of the tissues is no longer met by oxygen supply, and death occurs with increasing oxygen debt. The aim of the current study was to investigate whether ventilation with 100% oxygen (fraction of inspired oxygen [FiO2] = 1.0; hyperoxic ventilation) initiated at Hbcrit could restore adequate tissue oxygenation and prevent death. Methods Fourteen anesthetized pigs ventilated with room air (FiO2 = 0.21) were hemodiluted by exchange of whole blood for 6% hydroxyethyl starch (200,000:0.5) until the individual Hbcrit was reached. Hbcrit was defined as the onset of oxygen supply dependency of oxygen consumption and was identified with indirect calorimetry. For the next 6 h, animals were either ventilated with an FiO2 of 0.21 (n = 7) or an FiO2 of 1.0 (n = 7). ...
The first clinical impression of emergency patients conveys a myriad of information that has been... more The first clinical impression of emergency patients conveys a myriad of information that has been incompletely elucidated. In this prospective, observational study, the value of the first clinical impression, assessed by 18 observations, to predict the need for timely medical attention, the need for hospital admission, and in-hospital mortality in 1506 adult patients presenting to the triage desk of an emergency department was determined. Machine learning models were used for statistical analysis. The first clinical impression could predict the need for timely medical attention [area under the receiver operating characteristic curve (AUC ROC), 0.73; p = 0.01] and hospital admission (AUC ROC, 0.8; p = 0.004), but not in-hospital mortality (AUC ROC, 0.72; p = 0.13). The five most important features informing the prediction models were age, ability to walk, admission by emergency medical services, lying on a stretcher, breathing pattern, and bringing a suitcase. The inability to walk a...
BackgroundIt remains elusive how the characteristics, the course of disease, the clinical managem... more BackgroundIt remains elusive how the characteristics, the course of disease, the clinical management and the outcomes of critically ill COVID-19 patients admitted to intensive care units (ICU) worldwide have changed over the course of the pandemic.MethodsProspective, observational registry constituted by 90 ICUs across 22 countries worldwide including patients with a laboratory-confirmed, critical presentation of COVID-19 requiring advanced organ support. Hierarchical, generalized linear mixed-effect models accounting for hospital and country variability were employed to analyse the continuous evolution of the studied variables over the pandemic.ResultsFour thousand forty-one patients were included from March 2020 to September 2021. Over this period, the age of the admitted patients (62 [95% CI 60–63] years vs 64 [62–66] years,p < 0.001) and the severity of organ dysfunction at ICU admission decreased (Sequential Organ Failure Assessment 8.2 [7.6–9.0] vs 5.8 [5.3–6.4],p < 0.00...
Objectives: The aims of the study were to compare the consumption of blood products before and af... more Objectives: The aims of the study were to compare the consumption of blood products before and after the implementation of a bleeding management algorithm in patients undergoing liver transplantation and to determine the feasibility of a multicentre, randomized study. Background: Liver transplantation remains the only curative therapy for patients with end-stage liver disease, but it carries a high risk of surgical bleeding. Materials and Methods: Retrospective study of patients treated before (group 1) and after (group 2) implementation of a haemostatic algorithm guided by viscoelastic testing, including use of lyophilized coagulation factor concentrates (prothrombin complex and fibrinogen concentrates). Primary outcome was the number of units of blood products transfused in 24 h after surgery. Secondary outcomes included hospital stay, mortality, and cost. Results: Data from 30 consecutive patients was analysed; 14 in group 1 and 16 in group 2. Baseline data were similar between g...
Background: Differences in blood pressure can influence the risk of brain ischemia, perioperative... more Background: Differences in blood pressure can influence the risk of brain ischemia, perioperative complications, and postoperative neurocognitive function in patients undergoing carotid endarterectomy (CEA). Methods: In this single-center trial, patients scheduled for CEA under general anesthesia were randomized into an intervention group receiving near-infrared spectroscopy (NIRS)-guided blood pressure management during carotid cross-clamping and a control group receiving standard care. The primary endpoint was postoperative neurocognitive function assessed before surgery, on postoperative days 1 and 7, and eight weeks after surgery. Perioperative complications and cerebral autoregulatory capacity were secondary endpoints. Results: Systolic blood pressure (p < 0.001) and norepinephrine doses (89 (54–122) vs. 147 (116–242) µg; p < 0.001) during carotid cross-clamping were lower in the intervention group. No group differences in postoperative neurocognitive function were observ...
Many previous studies claim to have developed machine learning models that diagnose COVID-19 from... more Many previous studies claim to have developed machine learning models that diagnose COVID-19 from blood tests. However, we hypothesize that changes in the underlying distribution of the data, so called domain shifts, affect the predictive performance and reliability and are a reason for the failure of such machine learning models in clinical application. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (way of taking samples, laboratory procedures), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. We investigate whether domain shifts are present in COVID-19 datasets and how they affect machine learning methods. We further set out to estimate the mortality risk based on routinely acquired blood tests in a hospital setting throughout pandemics and under domain shifts. We reveal domain ...
For many years, physicians’ approach to the transfusion of allogeneic red blood cells (RBC) was n... more For many years, physicians’ approach to the transfusion of allogeneic red blood cells (RBC) was not individualized. It was accepted that a hemoglobin concentration (Hb) of less than 10 g/dL was a general transfusion threshold and the majority of patients were transfused immediately. In recent years, there has been increasing evidence that even significantly lower hemoglobin concentrations can be survived in the short term without sequelae. This somehow contradicts the observation that moderate or mild anemia is associated with relevant long-term morbidity and mortality. To resolve this apparent contradiction, it must be recognized that we have to avoid acute anemia or treat it by alternative methods. The aim of this article is to describe the physiological limits of acute anemia, match these considerations with clinical realities, and then present “patient blood management” (PBM) as the therapeutic concept that can prevent both anemia and unnecessary transfusion of RBC concentrates ...
Visually evoked potentials (VEPs) are widely used for diagnoses of different neurological disease... more Visually evoked potentials (VEPs) are widely used for diagnoses of different neurological diseases. Interestingly, there is limited research about the impact of the stimulus color onto the evoked response. Therefore, in our study we investigated the possibility of automatically classifying the stimulus color. The visual stimuli were selected to be red/black and green/black checkerboard patterns with equal light density. Both of these stimuli were presented in a random manner to nine subjects, while the electroencephalogram was recorded at the occipital lobe. After pre-processing and aligning the evoked potentials, an artificial neural network with one hidden layer was used to investigate the general possibility to automatically classify the stimulus color in three different settings. First, color classification with individually trained models, color classification with a common model, and color classification for each individual volunteer with a model trained on the data of the rem...
Summary Background In December 2019, the new virus infection coronavirus disease 2019 (COVID-19) ... more Summary Background In December 2019, the new virus infection coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged. Simple clinical risk scores may improve the management of COVID-19 patients. Therefore, the aim of this pilot study was to evaluate the quick Sequential Organ Failure Assessment (qSOFA) score, which is well established for other diseases, as an early risk assessment tool predicting a severe course of COVID-19. Methods We retrospectively analyzed data from adult COVID-19 patients hospitalized between March and July 2020. A critical disease progress was defined as admission to intensive care unit (ICU) or death. Results Of 64 COVID-19 patients, 33% (21/64) had a critical disease progression from which 13 patients had to be transferred to ICU. The COVID-19-associated mortality rate was 20%, increasing to 39% after ICU admission. All patients without a critical progress had a qSOFA score ≤ 1 at admission. Pat...
We investigate machine learning models that identify COVID-19 positive patients and estimate the ... more We investigate machine learning models that identify COVID-19 positive patients and estimate the mortality risk based on routinely acquired blood tests in a hospital setting. However, during pandemics or new outbreaks, disease and testing characteristics change, thus we face domain shifts. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (taking samples, laboratory), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. To countermand this effect, we propose methods that first identify domain shifts and then reverse their negative effects on the model performance. Frequent re-training and reassessment, as well as stronger weighting of more recent samples, keeps model performance and credibility at a high level over time. Our diagnosis models are constructed and tested on large-scale data ...
Background Uncertainty about the optimal respiratory support strategies in critically ill COVID-1... more Background Uncertainty about the optimal respiratory support strategies in critically ill COVID-19 patients is widespread. While the risks and benefits of noninvasive techniques versus early invasive mechanical ventilation (IMV) are intensely debated, actual evidence is lacking. We sought to assess the risks and benefits of different respiratory support strategies, employed in intensive care units during the first months of the COVID-19 pandemic on intubation and intensive care unit (ICU) mortality rates. Methods Subanalysis of a prospective, multinational registry of critically ill COVID-19 patients. Patients were subclassified into standard oxygen therapy ≥10 L/min (SOT), high-flow oxygen therapy (HFNC), noninvasive positive-pressure ventilation (NIV), and early IMV, according to the respiratory support strategy employed at the day of admission to ICU. Propensity score matching was performed to ensure comparability between groups. Results Initially, 1421 patients were assessed for...
Objective The diagnosis of COVID-19 is based on the detection of SARS-CoV-2 in respiratory secret... more Objective The diagnosis of COVID-19 is based on the detection of SARS-CoV-2 in respiratory secretions, blood, or stool. Currently, reverse transcription polymerase chain reaction (RT-PCR) is the most commonly used method to test for SARS-CoV-2. Methods In this retrospective cohort analysis, we evaluated whether machine learning could exclude SARS-CoV-2 infection using routinely available laboratory values. A Random Forests algorithm with 28 unique features was trained to predict the RT-PCR results. Results Out of 12,848 patients undergoing SARS-CoV-2 testing, routine blood tests were simultaneously performed in 1357 patients. The machine learning model could predict SARS-CoV-2 test results with an accuracy of 86% and an area under the receiver operating characteristic curve of 0.74. Conclusion Machine learning methods can reliably predict a negative SARS-CoV-2 RT-PCR test result using standard blood tests.
BackgroundThe ability to predict transfusions arising during hospital admission might enable econ... more BackgroundThe ability to predict transfusions arising during hospital admission might enable economized blood supply management and might furthermore increase patient safety by ensuring a sufficient stock of red blood cells (RBCs) for a specific patient. We therefore investigated the precision of four different machine learning–based prediction algorithms to predict transfusion, massive transfusion, and the number of transfusions in patients admitted to a hospital.Study Design and MethodsThis was a retrospective, observational study in three adult tertiary care hospitals in Western Australia between January 2008 and June 2017. Primary outcome measures for the classification tasks were the area under the curve for the receiver operating characteristics curve, the F1 score, and the average precision of the four machine learning algorithms used: neural networks (NNs), logistic regression (LR), random forests (RFs), and gradient boosting (GB) trees.ResultsUsing our four predictive model...
Background Acute normovolemic hemodilution reduces the circulating erythrocyte mass and, thus, th... more Background Acute normovolemic hemodilution reduces the circulating erythrocyte mass and, thus, the hemoglobin concentration. After extreme acute normovolemic hemodilution to the critical hemoglobin concentration (Hbcrit), oxygen demand of the tissues is no longer met by oxygen supply, and death occurs with increasing oxygen debt. The aim of the current study was to investigate whether ventilation with 100% oxygen (fraction of inspired oxygen [FiO2] = 1.0; hyperoxic ventilation) initiated at Hbcrit could restore adequate tissue oxygenation and prevent death. Methods Fourteen anesthetized pigs ventilated with room air (FiO2 = 0.21) were hemodiluted by exchange of whole blood for 6% hydroxyethyl starch (200,000:0.5) until the individual Hbcrit was reached. Hbcrit was defined as the onset of oxygen supply dependency of oxygen consumption and was identified with indirect calorimetry. For the next 6 h, animals were either ventilated with an FiO2 of 0.21 (n = 7) or an FiO2 of 1.0 (n = 7). ...
The first clinical impression of emergency patients conveys a myriad of information that has been... more The first clinical impression of emergency patients conveys a myriad of information that has been incompletely elucidated. In this prospective, observational study, the value of the first clinical impression, assessed by 18 observations, to predict the need for timely medical attention, the need for hospital admission, and in-hospital mortality in 1506 adult patients presenting to the triage desk of an emergency department was determined. Machine learning models were used for statistical analysis. The first clinical impression could predict the need for timely medical attention [area under the receiver operating characteristic curve (AUC ROC), 0.73; p = 0.01] and hospital admission (AUC ROC, 0.8; p = 0.004), but not in-hospital mortality (AUC ROC, 0.72; p = 0.13). The five most important features informing the prediction models were age, ability to walk, admission by emergency medical services, lying on a stretcher, breathing pattern, and bringing a suitcase. The inability to walk a...
BackgroundIt remains elusive how the characteristics, the course of disease, the clinical managem... more BackgroundIt remains elusive how the characteristics, the course of disease, the clinical management and the outcomes of critically ill COVID-19 patients admitted to intensive care units (ICU) worldwide have changed over the course of the pandemic.MethodsProspective, observational registry constituted by 90 ICUs across 22 countries worldwide including patients with a laboratory-confirmed, critical presentation of COVID-19 requiring advanced organ support. Hierarchical, generalized linear mixed-effect models accounting for hospital and country variability were employed to analyse the continuous evolution of the studied variables over the pandemic.ResultsFour thousand forty-one patients were included from March 2020 to September 2021. Over this period, the age of the admitted patients (62 [95% CI 60–63] years vs 64 [62–66] years,p < 0.001) and the severity of organ dysfunction at ICU admission decreased (Sequential Organ Failure Assessment 8.2 [7.6–9.0] vs 5.8 [5.3–6.4],p < 0.00...
Uploads