COVIDAL: A Machine Learning Classifier for Digital COVID-19 Diagnosis in German Hospitals

For the fight against the COVID-19 pandemic, it is particularly important to map the course of infection, in terms of patients who have currently tested SARS-CoV-2 positive, as accurately as possible. In hospitals, this is even more important because resources have become scarce. Although polymerase chain reaction (PCR) and point of care (POC) antigen testing capacities have been massively expanded, they are often very time-consuming and cost-intensive and, in some cases, lack appropriate performance. To meet these challenges, we propose the COVIDAL classifier for AI-based diagnosis of symptomatic COVID-19 subjects in hospitals based on laboratory parameters. We evaluate the algorithm's performance by unique multicenter data with approximately 4,000 patients and an extraordinary high ratio of SARS-CoV-2-positive patients. We analyze the influence of data preparation, flexibility in optimization targets, as well as the selection of the test set on the COVIDAL outcome. The algorithm is compared with standard AI, PCR, POC antigen testing and manual classifications of seven physicians by a decision theoretic scoring model including performance metrics, turnaround times and cost. Thereby, we define health care settings in which a certain classifier for COVID-19 diagnosis is to be applied. We find sensitivities, specificities, and accuracies of the COVIDAL algorithm of up to 90 percent. Our scoring model suggests using PCR testing for a focus on performance metrics. For turnaround times, POC antigen testing should be used. If balancing performance, turnaround times, and cost is of interest, as, for example, in the emergency department, COVIDAL is superior based on the scoring model. © 2023 Association for Computing Machinery.

A proof of concept study for the differentiation of SARS-CoV-2, hCoV-NL63, and IAV-H1N1 in vitro cultures using ion mobility spectrometry.

Rapid, high-throughput diagnostic tests are essential to decelerate the spread of the novel coronavirus disease 2019 (COVID-19) pandemic. While RT-PCR tests performed in centralized laboratories remain the gold standard, rapid point-of-care antigen tests might provide faster results. However, they are associated with markedly reduced sensitivity. Bedside breath gas analysis of volatile organic compounds detected by ion mobility spectrometry (IMS) may enable a quick and sensitive point-of-care testing alternative. In this proof-of-concept study, we investigated whether gas analysis by IMS can discriminate severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from other respiratory viruses in an experimental set-up. Repeated gas analyses of air samples collected from the headspace of virus-infected in vitro cultures were performed for 5 days. A three-step decision tree using the intensities of four spectrometry peaks correlating to unidentified volatile organic compounds allowed the correct classification of SARS-CoV-2, human coronavirus-NL63, and influenza A virus H1N1 without misassignment when the calculation was performed with data 3 days post infection. The forward selection assignment model allowed the identification of SARS-CoV-2 with high sensitivity and specificity, with only one of 231 measurements (0.43%) being misclassified. Thus, volatile organic compound analysis by IMS allows highly accurate differentiation of SARS-CoV-2 from other respiratory viruses in an experimental set-up, supporting further research and evaluation in clinical studies.

Geographical disparities in clinical outcomes of severe COVID-19 patients treated with remdesivir

Background: Remdesivir (RDV), a RNA polymerase inhibitor with potent in vitro activity against SARS-CoV-2, is the only treatment with demonstrated efficacy in shortening the duration of COVID-19. Here we report regional differences in clinical outcomes of severe COVID-19 patients treated with RDV, as part of an open-label, randomized phase-3 trial establishing RDV treatment duration. Methods: Hospitalized patients with oxygen saturation ≤94%, a positive SARS-CoV-2 PCR in the past 4 days and radiographic evidence of pneumonia were randomized 1:1 to receive 5d or 10d of intravenous RDV. We compared d14 clinical outcomes of patients from different geographical areas, as measured by mortality rates, change in clinical status from baseline (BL) on a 7-point ordinal scale and change in O2 requirements from BL. Based on previous analyses in compassionate use data showing region as an important predictor of outcome, Italy was examined separately from other regions. Results: 397 patients were treated with RDV, of which 229 (58%) were in the US, 77 (19%) Italy, 61 (15% in Spain), 12 (3%) Republic of Korea, 9 (2%) Singapore, 4 (1%) Germany, 4 (1%) Hong Kong and 1 (< 1%) Taiwan. BL clinical status was worse in Italy compared to other regions (72% vs 17% requiring high-flow oxygen delivery or higher), and Italian patients were more likely to be male than patients from other regions (69% vs 63%). Overall results showed 5d RDV was as effective as 10d. Mortality at d14 was higher in Italy (18%) compared to all other countries except Italy (7%). Similarly, clinical improvement at d14, measured as ≥2-point increase in the ordinal scale, was lower in Italian patients (39%) compared to all other countries combined (64%). (Fig.1). Conclusion: Overall, our results demonstrate significant geographical differences in the clinical course of severe COVID-19 patients treated with RDV. We observed worse outcomes, such as increased mortality and lower rate of clinical improvement, in patients from Italy compared to other regions. (Table Presented).

Remdesivir vs standard care in patients with moderate COVID-19

Background: Remdesivir (RDV) shortens time to recovery time in patients with severe COVID-19. Its effect in patients with moderate COVID-19 remains unclear. Methods: We conducted an open-label, phase 3 trial (NCT04252664) involving hospitalized patients with confirmed SARS-CoV-2 infection, evidence of pulmonary infiltrates, and oxygen saturation >94% on room air. Patients were randomly assigned 1:1:1 to receive up to 5d or 10d of RDV with standard of care (SoC), or SoC alone;patients could be discharged prior to completing per-protocol assigned treatment duration. RDV was dosed intravenously at 200 mg on d1, 100 mg daily thereafter. Patients were evaluated daily while hospitalized, and via telephone if discharged. The primary endpoint was clinical status on d11 assessed on a 7-point ordinal scale. Results regarding the primary endpoint are expected to be published before IDWeek 2020;we plan to present d28 results at the meeting. Results: In total, 584 patients underwent randomization and started their assigned treatment (191, 5d RDV;193, 10d RDV;200, SoC). By d11, 3 2 point improvement on the ordinal scale occurred in 70% of patients in the 5d arm, 65% in the 10d arm, and 61% in the SoC arm. Patients in the 5d RDV arm were significantly more likely to have an improvement in clinical status than those receiving SoC (odds ratio [OR], 1.65;95% confidence interval [CI], 1.09-2.48;P=0.017);OR of improvement for the 10d RDV arm compared to SoC was 1.31 (95% CI, 0.88-1.95];p=0.183). This improvement in the 5-day arm over the SOC arm was noted from d6 through d11. We observed a peak of discharges corresponding with the assigned treatment duration of RDV, with increased discharges at d6 in the 5-day arm and at d11 in the 10-day arm. A worsening of clinical status of ≥ 1 point in the ordinal scale was observed more commonly in the SoC am (n=19, 10%) versus the 5d RDV (n=7, 4%) and 10d RDV (n=9, 5%). Conclusion: RDV for up to 5 days was superior to SoC in improving the clinical status of patients with moderate COVID-19 by d11. We will report d28 outcomes at the meeting. (Table Presented).

[Role of university hospitals in Bavaria during the COVID-19 pandemic]. / Die Universitätsmedizin in Bayern im Rahmen der COVID-19-Pandemie.

BACKGROUND: In early 2020 the German healthcare system was put into a state of emergency due to the coronavirus disease 2019 (COVID-19) pandemic. Bavaria had to deal with more severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections than any other German state during the first wave and currently has over 270,000 cases, accounting for about one fifth of all COVID-19 cases in Germany. The Bavarian Ministry of Interior together with the Bavarian Sate Ministry of Health and Care issued a general ruling at the beginning of the first wave that ordered the centralised organisation of hospital capacity, a redesign of the information technology (IT) management system and introduced reporting obligations for SARS-CoV-2/COVID-19. The goal of this analysis was to investigate the role that university hospitals played in the inpatient treatment of COVID-19 patients. METHODS: A retrospective evaluation of all inpatient COVID-19 cases that were reported through the "IVENA Sonderlage" (Ivena eHEALTH, [IVENA, interdisziplinärer Versorgungsnachweis, mainis IT-Service GmbH, Offenbach am Main, Germany]), a special module for the "Interdisciplinary Medical Care Capacity Management System" designed for extraordinary events and circumstances, was conducted by analysing the number of reported treatment days of all Bavarian hospitals that participated in the treatment of COVID-19 patients. RESULTS: During the first wave university hospitals provided relevant scientific contributions and played an important role in advising physicians, hospitals and politicians on the pandemic. In all, 20% of intensive care unit (ICU) and intermediate care (IMC) treatment days were provided by the university hospitals in particular for treatment of complex courses of COVID-19.

[S2k Guideline - Recommendations for Inpatient Therapy of Patients with COVID-19]. / S2k-Leitlinie ­ Empfehlungen zur stationären Therapie von Patienten mit COVID-19.

Since December 2019, the novel coronavirus SARS-CoV-2 (Severe Acute Respiratory Syndrome - Corona Virus-2) has been spreading rapidly in the sense of a global pandemic. This poses significant challenges for clinicians and hospitals and is placing unprecedented strain on the healthcare systems of many countries. The majority of patients with Coronavirus Disease 2019 (COVID-19) present with only mild symptoms such as cough and fever. However, about 6 % require hospitalization. Early clarification of whether inpatient and, if necessary, intensive care treatment is medically appropriate and desired by the patient is of particular importance in the pandemic. Acute hypoxemic respiratory insufficiency with dyspnea and high respiratory rate (> 30/min) usually leads to admission to the intensive care unit. Often, bilateral pulmonary infiltrates/consolidations or even pulmonary emboli are already found on imaging. As the disease progresses, some of these patients develop acute respiratory distress syndrome (ARDS). Mortality reduction of available drug therapy in severe COVID-19 disease has only been demonstrated for dexamethasone in randomized controlled trials. The main goal of supportive therapy is to ensure adequate oxygenation. In this regard, invasive ventilation and repeated prone positioning are important elements in the treatment of severely hypoxemic COVID-19 patients. Strict adherence to basic hygiene, including hand hygiene, and the correct wearing of adequate personal protective equipment are essential when handling patients. Medically necessary actions on patients that could result in aerosol formation should be performed with extreme care and preparation.

Impact of baseline alanine aminotransferase levels on the safety and efficacy of remdesivir in severe COVID-19 patients

Background: Remdesivir (RDV), a nucleotide analogue prodrug that inhibits viral RNA polymerases, has demonstrated potent in vitro and in vivo activity against SAR-CoV-2 and favorable clinical efficacy and tolerability in patients with moderate and severe COVID-19 Elevated transaminase levels are commonly seen in patients with severe COVID-19 prior to treatment Here we report safety and clinical outcomes after RDV treatment in patients with normal versus elevated baseline alanine aminotransferase (ALT) levels Methods: We conducted a randomized, open-label, phase 3 trial, involving hospitalized patients with confirmed COVID-19 pneumonia with Sat<94% Patients with screening ALT or AST> 5x the upper limit of normal (ULN) were excluded from the study Patients were randomized 1:1 to receive either 5 or 10 days of intravenous RDV once daily We compared patients with baseline ALT below and above the ULN based on AASLD criteria (ALT 35 U/L for males and 25 U/L for females) Covariates for adjustment included age, sex, race and baseline oxygen support Clinical recovery and all-cause mortality were evaluated using logistic regression Clinical outcomes and adverse events (AEs) were assessed through day 28 Results: Of 397 patients treated with RDV, 215 (54%) had elevated baseline ALT Median ALT was 53 U/L (IQR: 40 - 78 U/L) in the high ALT group Patients with high ALT at time of RDV initiation were younger (median 58 vs 65 years, p<0 001), required less oxygen (p=0 02), and had longer symptom duration (median 10 vs 8 days, p<0.001) prior to first dose of RDV. Incidence of serious AEs, grade ≥3 AEs, and AE leading to discontinuation were similar between groups (Table1). Grade ≥3 hepatobiliary adverse events, particularly transaminase elevations, were not common but numerically higher in the high ALT group (8 8% vs 3 3%, p=0 068) Time to clinical recovery, 2-point clinical improvement, 1-point clinical improvement, room air, and death were similar between groups Conclusion: In severe COVID-19 patients, adverse events and clinical outcomes after RDV initiation were similar among patients with baseline normal ALT and those with elevated ALT (1-5x ULN)(Table Presented).

[Numbers of emergency room patients during the COVID-19 pandemic]. / Patientenzahlen im Rahmen der COVID-19-Pandemie in einer zentralen Notaufnahme.

BACKGROUND: Since end of March, the health care system in Germany has been placed into a state of emergency in order to gain resources for the spreading coronavirus disease 2019 (COVID-19) pandemic. The overall goal of this study is to evaluate the number of emergency room patients at the time of the pandemic in order to draw conclusions about the influence of the COVID 19 pandemic on the number of patients in an emergency department. MATERIALS AND METHODS: With this descriptive epidemiologic study we collected and analyzed anonymized patient-related data of 19,357 cases presenting to the emergency department of the Klinikum rechts der Isar (Munich) from 01 February 2019 to 30 April 2019 and from 01 February 2020 to 30 April 2020. RESULTS: Despite an increase in the number of patients from 2019 to 2020, there was a significant drop in the number of emergencies from February to March 2020 and proceeding in April to a level below that of 2019. This was particularly observed in the field of trauma surgery, with a 40% decrease in the number of patients. With regard to the individual complaint patterns in March 2020, it was found that an increased incidence of malaise (+47%) and breathing problems (+36%) was recorded, whereas back pain (-41%), wounds (-29%), thoracic (-24%) and abdominal pain (-23%) were significantly less common than in the previous year. In terms of the severity of the complaints, the decline was mainly due to complaints with a low degree of urgency. CONCLUSION: In the course of the COVID-19 pandemic we observed a significant decline in the number of patients in one of the largest emergency rooms in Munich. This has to be avoided with existing hospital capacities, in order to prevent potential damage to health caused by postponed or missing emergency presentations.