Lukas Martin

Proenkephalin A 119–159 (PENK) is a promising functional kidney biomarker, evaluated in various clinical settings. In critical care medicine, early diagnosis of acute kidney injury (AKI) is crucial; however, to date, the diagnosis and the assessment of kidney function is still based on serum creatinine (sCr) and urine output, both associated with several limitations. Between November 2020 and March 2022, we implemented PENK in our daily practice on our intensive care units (ICU). PENK, sCr, AKI stage, and the start and duration of renal replacement therapy (RRT) were documented. Almost 18,000 PENK measurements from 4169 patients were analyzed, and the glomerular filtration rate (GFR) was estimated with the new PENK-GFR formula. PENK outperformed sCR in the kidney function assessment and sCR trajectory over time. Moreover, PENK predicted the use of RRT and thus showed its usefulness in critical care daily practice.

Background: The medical teams in intensive care units (ICUs) spend increasing amounts of time at computer systems for data processing, input, and interpretation purposes. As each patient creates about 1000 data points per hour, the available information is abundant, making the interpretation difficult and time-consuming. This data flood leads to a decrease in time for evidence-based, patient-centered care. Information systems, such as patient data management systems (PDMSs), are increasingly used at ICUs. However, they often create new challenges arising from the increasing documentation burden. Objective: New concepts, such as artificial intelligence (AI)–based assistant systems, are hence introduced to the workflow to cope with these challenges. However, there is a lack of standardized, published metrics in order to compare the various data input and management systems in the ICU setting. The objective of this study is to compare established documentation and retrieval processes with newer methods, such as PDMSs and voice information and documentation systems (VIDSs). Methods: In this crossover study, we compare traditional, paper-based documentation systems with PDMSs and newer AI-based VIDSs in terms of performance (required time), accuracy, mental workload, and user experience in an intensive care setting. Performance is assessed on a set of 6 standardized, typical ICU tasks, ranging from documentation to medical interpretation. Results: A total of 60 ICU-experienced medical professionals participated in the study. The VIDS showed a statistically significant advantage compared to the other 2 systems. The tasks were completed significantly faster with the VIDS than with the PDMS (1-tailed t59=12.48; Cohen d=1.61; P<.001) or paper documentation (t59=20.41; Cohen d=2.63; P<.001). Significantly fewer errors were made with VIDS than with the PDMS (t59=3.45; Cohen d=0.45; P=.03) and paper-based documentation (t59=11.2; Cohen d=1.45; P<.001). The analysis of the mental workload of VIDS and PDMS showed no statistically significant difference (P=.06). However, the analysis of subjective user perception showed a statistically significant perceived benefit of the VIDS compared to the PDMS (P<.001) and paper documentation (P<.001). Conclusions: The results of this study show that the VIDS reduced error rate, documentation time, and mental workload regarding the set of 6 standardized typical ICU tasks. In conclusion, this indicates that AI-based systems such as the VIDS tested in this study have the potential to reduce this workload and improve evidence-based and safe patient care.

The COVID-19 pandemic caused by the new SARS-CoV-2 coronavirus is the most recent and well-known outbreak of a coronavirus. RNase 1 is a small endogenous antimicrobial polypeptide that possesses antiviral activity against viral diseases. In this study, we investigated a potential association between ribonuclease 1 and the outcome in COVID-19 patients and the impact of increased and decreased RNase 1 levels serum during the course of the disease. Therefore, two patient populations, Cohort A (n = 35) and B (n = 80), were subclassified into two groups, in which the RNase 1 concentration increased or decreased from time point one to time point two. We show that the RNase 1 serum levels significantly increased in the increasing group of both cohorts (p = 0.0171; p < 0.0001). We detect that patients in the increasing group who died had significantly higher RNase 1 serum levels at both time points in Cohort A (p = 0.0170; p = 0.0393) and Cohort B (p = 0.0253; p = 0.0034) than patients who survived. Additionally, we measured a significant correlation of RNase 1 serum levels with serum creatinine as well as creatinine clearance in the increasing and decreasing group at both time points of Cohort A. Based on these results, there is now good evidence that RNase 1 may play a role in renal dysfunction associated with ICU COVID-19 patients and that increasing RNase 1 serum level may be a potential biomarker to predict outcome in COVID-19 patients.

The discipline of process mining has a solid track record of successful applications to the healthcare domain. Within such research space, we conducted a case study related to the Intensive Care Unit (ICU) ward of the Uniklinik Aachen hospital in Germany. The aim of this work is twofold: developing a normative model representing the clinical guidelines for the treatment of COVID-19 patients, and analyzing the adherence of the observed behavior (recorded in the information system of the hospital) to such guidelines. We show that, through conformance checking techniques, it is possible to analyze the care process for COVID-19 patients, highlighting the main deviations from the clinical guidelines. The results provide physicians with useful indications for improving the process and ensuring service quality and patient satisfaction. We share the resulting model as an open-source BPMN file.

Treatment of acute pulmonary embolism (PE) varies based upon risk stratification and ranges from outpatient oral anticoagulation to emergency surgical embolectomy. Patients with high-risk PE can be considered for systemic thrombolytic (ST) based upon guideline recommendations, but intermediate-risk PE does not currently have strong evidence to guide primary reperfusion strategies via thrombolytic administration. Ultrasound-assisted catheter-directed thrombolysis (USAT) is an alternative reperfusion option to ST but is not currently recommended as first line in any key guidelines due to limited available evidence. This retrospective, multicenter, observational study compares 210 patients treated with USAT (n = 105) or ST (n = 105) for acute high- or intermediate-risk PE in three hospitals. Baseline characteristics were significant in that severity of illness was higher in those that received ST, which limited comparisons of outcomes. The primary outcome of major bleeding in patients receiving USAT was 15.2% and 22.9% in those that received ST. Efficacy of reperfusion strategy was observed to be 86.7% of patients in USAT group and 65.7% in ST group. Reperfusion strategies had no difference in in-hospital death, intensive care length of stay, or hospital length of stay. Predefined subgroup analysis found that high-risk PE had higher mortality (14.7%) than intermediate-risk PE (0%) regardless of reperfusion strategy. Upon multivariate analysis, high-risk PE was the only independent risk factor for major bleeding while USAT therapy and intermediate-risk PE were independent predictors of efficacy. Due to the difference in baseline severity of illness, direct comparisons in primary outcomes to each group was not performed. We have described real world usage of both USAT and ST and which patients were likely to receive each therapy at these institutions.

Cardiac dysfunction is a life-threatening complication in sepsis. Upon infection and cardiac stress, the cardiac macrophage population expands. Recruited macrophages exhibit a predominantly proinflammatory phenotype and release danger-associated molecular patterns (DAMPs) that contribute to cardiac dysfunction. However, the underlying pathomechanisms are highly complex and not fully understood. Here, we utilized an indirect macrophage–cardiomyocyte co-culture model to study the effects of proinflammatory macrophages on the activation of different cardiac receptors (TLR3, TLR4, and TNFR) and their role in cardiac inflammation and caspase-3/7 activation. The stimulation of cardiomyocytes with conditioned medium of LPS-stimulated macrophages resulted in elevated IL-6 protein concentrations and relative IL-6 and TNFα mRNA levels. Conditioned medium from LPS-stimulated macrophages also induced NFκB translocation and increased caspase-3/7 activation in cardiomyocytes. Analyzing the role of different cardiac receptors, we found that TLR4 and TNFR inhibition reduces cardiac inflammation and that the inhibition of TNFR prevents NFκB translocation into the nuclei of cardiomyocytes, induced by exposure to conditioned medium of proinflammatory macrophages. Moreover, we demonstrated that TLR3 inhibition reduces macrophage-mediated caspase-3/7 activation. Our results suggest that the immune response of macrophages under inflammatory conditions leads to the release of DAMPs, such as eRNA and cytokines, which in turn induce cardiomyocyte dysfunction. Thus, the data obtained in this study contribute to a better understanding of the pathophysiological mechanisms of cardiac dysfunction.

BACKGROUND The medical teams in intensive care units (ICUs) spend increasing amounts of time at computer systems for data processing, input, and interpretation purposes. As each patient creates about 1000 data points per hour, the available information is abundant, making the interpretation difficult and time-consuming. This data flood leads to a decrease in time for evidence-based, patient-centered care. Information systems, such as patient data management systems (PDMSs), are increasingly used at ICUs. However, they often create new challenges arising from the increasing documentation burden. OBJECTIVE New concepts, such as artificial intelligence (AI)–based assistant systems, are hence introduced to the workflow to cope with these challenges. However, there is a lack of standardized, published metrics in order to compare the various data input and management systems in the ICU setting. The objective of this study is to compare established documentation and retrieval processes with newer methods, such as PDMSs and voice information and documentation systems (VIDSs). METHODS In this crossover study, we compare traditional, paper-based documentation systems with PDMSs and newer AI-based VIDSs in terms of performance (required time), accuracy, mental workload, and user experience in an intensive care setting. Performance is assessed on a set of 6 standardized, typical ICU tasks, ranging from documentation to medical interpretation. RESULTS A total of 60 ICU-experienced medical professionals participated in the study. The VIDS showed a statistically significant advantage compared to the other 2 systems. The tasks were completed significantly faster with the VIDS than with the PDMS (1-tailed t59=12.48; Cohen d=1.61; P<.001) or paper documentation (t59=20.41; Cohen d=2.63; P<.001). Significantly fewer errors were made with VIDS than with the PDMS (t59=3.45; Cohen d=0.45; P=.03) and paper-based documentation (t59=11.2; Cohen d=1.45; P<.001). The analysis of the mental workload of VIDS and PDMS showed no statistically significant difference (P=.06). However, the analysis of subjective user perception showed a statistically significant perceived benefit of the VIDS compared to the PDMS (P<.001) and paper documentation (P<.001). CONCLUSIONS The results of this study show that the VIDS reduced error rate, documentation time, and mental workload regarding the set of 6 standardized typical ICU tasks. In conclusion, this indicates that AI-based systems such as the VIDS tested in this study have the potential to reduce this workload and improve evidence-based and safe patient care. CLINICALTRIAL

The discipline of process mining has a solid track record of successful applications to the healthcare domain. Within such research space, we conducted a case study related to the Intensive Care Unit (ICU) ward of the Uniklinik Aachen hospital in Germany. The aim of this work is twofold: developing a normative model representing the clinical guidelines for the treatment of COVID-19 patients, and analyzing the adherence of the observed behavior (recorded in the information system of the hospital) to such guidelines. We show that, through conformance checking techniques, it is possible to analyze the care process for COVID-19 patients, highlighting the main deviations from the clinical guidelines. The results provide physicians with useful indications for improving the process and ensuring service quality and patient satisfaction. We share the resulting model as an open-source BPMN file.

SARS-CoV-2 was first detected in 2019 in Wuhan, China. It has been found to be the most pathogenic virus among coronaviruses and is associated with endothelial damage resulting in respiratory failure. Determine whether heparanase and heparan sulfate fragments, biomarkers of endothelial function, can assist in the risk stratification and clinical management of critically ill COVID-19 patients admitted to the intensive care unit. We investigated 53 critically ill patients with severe COVID-19 admitted between March and April 2020 to the University Hospital RWTH Aachen. Heparanase activity and serum levels of both heparanase and heparan sulfate were measured on day one (day of diagnosis) and day three in patients with COVID-19. The patients were classified into four groups according to the severity of ARDS. When compared to baseline data (day one), heparanase activity increased and the heparan sulfate serum levels decreased with increasing severity of ARDS. The heparanase activity significantly correlated with the lactate concentration on day one (r = 0.34, p = 0.024) and on day three (r = 0.43, p = 0.006). Heparanase activity and heparan sulfate levels correlate with COVID-19 disease severity and outcome. Both biomarkers might be helpful in predicting clinical course and outcomes in COVID-19 patients.

Background: In recent years, the volume of medical knowledge and health data has increased rapidly. For example, the increased availability of electronic health records (EHRs) provides accurate, up-to-date, and complete information about patients at the point of care and enables medical staff to have quick access to patient records for more coordinated and efficient care. With this increase in knowledge, the complexity of accurate, evidence-based medicine tends to grow all the time. Health care workers must deal with an increasing amount of data and documentation. Meanwhile, relevant patient data are frequently overshadowed by a layer of less relevant data, causing medical staff to often miss important values or abnormal trends and their importance to the progression of the patient’s case. Objective: The goal of this work is to analyze the current laboratory results for patients in the intensive care unit (ICU) and classify which of these lab values could be abnormal the next time the test is done. Detecting near-future abnormalities can be useful to support clinicians in their decision-making process in the ICU by drawing their attention to the important values and focus on future lab testing, saving them both time and money. Additionally, it will give doctors more time to spend with patients, rather than skimming through a long list of lab values. Methods: We used Structured Query Language to extract 25 lab values for mechanically ventilated patients in the ICU from the MIMIC-III and eICU data sets. Additionally, we applied time-windowed sampling and holding, and a support vector machine to fill in the missing values in the sparse time series, as well as the Tukey range to detect and delete anomalies. Then, we used the data to train 4 deep learning models for time series classification, as well as a gradient boosting–based algorithm and compared their performance on both data sets. Results: The models tested in this work (deep neural networks and gradient boosting), combined with the preprocessing pipeline, achieved an accuracy of at least 80% on the multilabel classification task. Moreover, the model based on the multiple convolutional neural network outperformed the other algorithms on both data sets, with the accuracy exceeding 89%. Conclusions: In this work, we show that using machine learning and deep neural networks to predict near-future abnormalities in lab values can achieve satisfactory results. Our system was trained, validated, and tested on 2 well-known data sets to ensure that our system bridged the reality gap as much as possible. Finally, the model can be used in combination with our preprocessing pipeline on real-life EHRs to improve patients’ diagnosis and treatment.