Potential Predictors for Deterioration of Renal Function After Transfusion.

BACKGROUND: Transfusion of packed red blood cells (pRBCs) is still associated with risks. This study aims to determine whether renal function deterioration in the context of individual transfusions in individual patients can be predicted using machine learning. Recipient and donor characteristics linked to increased risk are identified. METHODS: This study was registered at ClinicalTrials.gov (NCT05466370) and was conducted after local ethics committee approval. We evaluated 3366 transfusion episodes from a university hospital between October 31, 2016, and August 31, 2020. Random forest models were tuned and trained via Python auto-sklearn package to predict acute kidney injury (AKI). The models included recipients' and donors' demographic parameters and laboratory values, donor questionnaire results, and the age of the pRBCs. Bootstrapping on the test dataset was used to calculate the means and standard deviations of various performance metrics. RESULTS: AKI as defined by a modified Kidney Disease Improving Global Outcomes (KDIGO) criterion developed after 17.4% transfusion episodes (base rate). AKI could be predicted with an area under the curve of the receiver operating characteristic (AUC-ROC) of 0.73 ± 0.02. The negative (NPV) and positive (PPV) predictive values were 0.90 ± 0.02 and 0.32 ± 0.03, respectively. Feature importance and relative risk analyses revealed that donor features were far less important than recipient features for predicting posttransfusion AKI. CONCLUSIONS: Surprisingly, only the recipients' characteristics played a decisive role in AKI prediction. Based on this result, we speculate that the selection of a specific pRBC may have less influence than recipient characteristics.

A community effort in SARS-CoV-2 drug discovery.

The COVID-19 pandemic continues to pose a substantial threat to human lives and is likely to do so for years to come. Despite the availability of vaccines, searching for efficient small-molecule drugs that are widely available, including in low- and middle-income countries, is an ongoing challenge. In this work, we report the results of an open science community effort, the "Billion molecules against COVID-19 challenge", to identify small-molecule inhibitors against SARS-CoV-2 or relevant human receptors. Participating teams used a wide variety of computational methods to screen a minimum of 1 billion virtual molecules against 6 protein targets. Overall, 31 teams participated, and they suggested a total of 639,024 molecules, which were subsequently ranked to find 'consensus compounds'. The organizing team coordinated with various contract research organizations (CROs) and collaborating institutions to synthesize and test 878 compounds for biological activity against proteases (Nsp5, Nsp3, TMPRSS2), nucleocapsid N, RdRP (only the Nsp12 domain), and (alpha) spike protein S. Overall, 27 compounds with weak inhibition/binding were experimentally identified by binding-, cleavage-, and/or viral suppression assays and are presented here. Open science approaches such as the one presented here contribute to the knowledge base of future drug discovery efforts in finding better SARS-CoV-2 treatments.

Using emergency department triage for machine learning-based admission and mortality prediction.

AIMS: Patient admission is a decision relying on sparsely available data. This study aims to provide prediction models for discharge versus admission for ward observation or intensive care, and 30 day-mortality for patients triaged with the Manchester Triage System. METHODS: This is a single-centre, observational, retrospective cohort study from data within ten minutes of patient presentation at the interdisciplinary emergency department of the Kepler University Hospital, Linz, Austria. We trained machine learning models including Random Forests and Neural Networks individually to predict discharge versus ward observation or intensive care admission, and 30 day-mortality. For analysis of the features' relevance, we used permutation feature importance. RESULTS: A total of 58323 adult patients between 1 December 2015 and 31 August 2020 were included. Neural Networks and Random Forests predicted admission to ward observation with an AUC-ROC of 0.842 ±â€…0.00 with the most important features being age and chief complaint. For admission to intensive care, the models had an AUC-ROC of 0.819 ±â€…0.002 with the most important features being the Manchester Triage category and heart rate, and for the outcome 30 day-mortality an AUC-ROC of 0.925 ±â€…0.001. The most important features for the prediction of 30 day-mortality were age and general ward admission. CONCLUSION: Machine learning can provide prediction on discharge versus admission to general wards and intensive care and inform about risk on 30 day-mortality for patients in the emergency department.

Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks.

Finding synthesis routes for molecules of interest is essential in the discovery of new drugs and materials. To find such routes, computer-assisted synthesis planning (CASP) methods are employed, which rely on a single-step model of chemical reactivity. In this study, we introduce a template-based single-step retrosynthesis model based on Modern Hopfield Networks, which learn an encoding of both molecules and reaction templates in order to predict the relevance of templates for a given molecule. The template representation allows generalization across different reactions and significantly improves the performance of template relevance prediction, especially for templates with few or zero training examples. With inference speed up to orders of magnitude faster than baseline methods, we improve or match the state-of-the-art performance for top-k exact match accuracy for k ≥ 3 in the retrosynthesis benchmark USPTO-50k. Code to reproduce the results is available at github.com/ml-jku/mhn-react.

Known and new facts on basal cell carcinoma.

Basal cell carcinoma (BCC) is the most common malignant tumor in light-skinned people and amounts to about 75 % of all cases of skin cancer. Increasing incidence rates have been reported for decades all over the world. The main risk factors include UV radiation, male sex, light skin type, advanced age, long-term immunosuppression, a positive individual or family history, and certain genodermatoses. BCC metastasizes only rarely, and its mortality is low, but it is associated with significant morbidity. Genetic mutations especially in the hedgehog pathway play an important role in BCC pathogenesis. Non-invasive procedures such as optical coherence tomography or confocal laser scan microscopy are increasingly utilized for diagnostics in addition to visual inspection and dermatoscopy, but only in exceptional cases can histological confirmation of the diagnosis be dispensed with. Various clinical and histological subtypes have been defined. Differentiating between BCC with high and low risk of recurrence has a significant influence on the choice of treatment. Most BCC can be treated effectively and safely with standard surgery, or in selected cases with topical treatment. Locally advanced and metastasized BCC must be treated with radiation or systemic therapy. Radiation is also an option for older patients with contraindications for surgery. The hedgehog inhibitors vismodegib and sonidegib are currently approved for systemic therapy of BCC in Europe. Approval for the PD1 inhibitor cemiplimab as second-line therapy is expected in the near future.

Laser-assisted epicutaneous immunization to target human skin dendritic cells.

Dendritic cells (DC) are promising targets for immunotherapy of cancer. Clinically, immunization against cancer antigens by means of the most potent antigen-presenting cells, that is DC, remains an important treatment option in combination with the modern immune checkpoint approaches. Instead of adoptively transferring in vitro monocyte-derived DC, they can also be loaded in situ by antibody-mediated targeting of antigen. Conventionally, these vaccines are delivered by classical intradermal injections. Here, we tested an alternative approach, namely laser-assisted epicutaneous immunization. With an infrared laser ("Precise Laser Epidermal System"/P.L.E.A.S.E.® Laser System), we created micropores in human skin and applied monoclonal antibodies (mAbs) against C-type lectins, for example DEC-205/CD205 and Langerin/CD207. Optimal parameters for formation of pores in epidermis and dermis were determined. We could induce pores of defined depths without enhanced apoptosis around them. Antibodies applied epicutaneously to the laser-porated skin could be detected both in Langerhans cells (LC) in situ in the epidermis and in migratory skin DC subsets from short term human skin explant culture, demonstrating uptake and transport of Langerin and DEC-205 mAbs. Efficacy of targeting was similar between the different laser treatments and pore depths. Thus, laser-assisted epicutaneous immunization may be a valuable alternative to intradermal injection, yet the loading efficacy of DC needs to be further improved.

Pulmonary vein isolation using second-generation single-shot devices: not all the same?

INTRODUCTION: Single-shot devices have been developed to simplify pulmonary vein isolation (PVI). Randomized studies of the second-generation cryoballoon (CB 2nd) demonstrated excellent results. There are limited data comparing results of circular pulmonary vein ablation catheter (PVAC) with conventional RF ablation or CB for PVI. OBJECTIVE: Using a sequential registry cohort and a prospective randomized study, we aimed to compare the acute and long-term results of CB 2nd and PVAC Gold. METHODS: In the registry, consecutive patients with paroxysmal atrial fibrillation (AF) undergoing their first PVI were included. The preferred method used was PVAC Gold in 2014 and CB 2nd in 2015. Subsequently, a randomized study (PVAC vs. CB 2nd) was performed. Ablation success was measured as freedom of AF or atrial tachycardias (AT) off antiarrhythmic drugs. RESULTS: In the registry cohort, PVAC Gold was used in 60 patients and CB 2nd in 56 patients (age 66 ± 11 years, 52% male, LAD 43 ± 6). In the randomized study, 20 patients were treated with PVAC Gold and 22 with CB 2nd (age 67 ± 9; 43% men, LAD 40 ± 7 mm). During a mean follow up of 13.2 ± 3.6 months, success was 54% in PVAC Gold patients and 81% in CB 2nd cases (p = 0.001). In the randomized study 12 months success was 50% versus 86%, p < 0.05. Complications occurred rare in both groups. CONCLUSIONS: Our registry data and the randomized study both suggest superiority of PVI using CB 2nd as compared with PVI using PVAC Gold.

Spektrum der Ichthyosen in einer österreichischen Ichthyosekohorte von 2004-2007.

HINTERGRUND: Ichthyosen sind eine heterogene Gruppe von Krankheiten, deren klinische Klassifizierung schwierig ist. Hier wird die Ichthyosekohorte eines Expertisezentrums für Genodermatosen im Detail beschrieben. PATIENTEN UND METHODIK: Eingeschlossen wurden Patienten mit klinisch oder genetisch bestätigter Ichthyose, die zwischen 2004 und 2017 untersucht und in einer Datenbank aufgenommen wurden. Krankheitsbeginn, Phänotyp, Histologie, Komorbiditäten und Familienanamnese wurden detailliert beschrieben. Bei den genetisch getesteten Patienten wurden Jahr und Methode der genetischen Testung protokolliert und die Prävalenz der unterschiedlichen Autosomal-rezessive-kongenitale Ichthyose (ARCI)-Gene und -Phänotypen, die Prävalenz der syndromalen Ichthyosen und die Genotyp-Phänotyp-Korrelationen analysiert. ERGEBNISSE UND METHODIK: Von den insgesamt 198 eingeschlossenen Patienten wurden 151 genetisch getestet. 81 Patienten hatten eine Ichthyosis vulgaris (IV), 43 eine X-chromosomale Ichthyose (XLI), 38 eine ARCI, 9 eine keratinopathische Ichthyose (KPI) und ein Patient eine Exfoliative Ichthyose. 26 Patienten litten an einer syndromalen Ichthyose. Im Vergleich zu den syndromalen Ichthyosen wurde bei den häufigen Ichthyosen (IV, XLI) und KPI eine gute Phänotyp-Genotyp-Korrelation beobachtet. In 91 % der ARCI-Patienten konnte die exakte Diagnose durch genetische Testung gestellt werden. Lediglich bei 33 % der Patienten mit syndromaler Ichthyose bestand vor der genetischen Testung ein Verdacht auf die tatsächliche Diagnose. In 86 % der Fälle wurde eine kausale Mutation nachgewiesen. SCHLUSSFOLGERUNGEN: Die Arbeit beschreibt das Spektrum der Ichthyosen an einem Expertisezentrum und zeigt, dass für diese Gruppe die genetische Testung von Genodermatosen ein diagnostischer Standard werden sollte.

Spectrum of ichthyoses in an Austrian ichthyosis cohort from 2004 to 2017.

BACKGROUND: Ichthyoses are a heterogeneous disease group, which makes clinical classification challenging. An ichthyosis cohort at a center for genodermatoses is presented in detail. PATIENTS AND METHODS: Patients with clinically and/or genetically confirmed ichthyosis seen from 2004 to 2017 and listed in a database were included. Disease onset, phenotype, histology, comorbidities and family history were described in detail. In genetically tested patients, the prevalence of various ARCI genes, ARCI phenotypes and syndromic ichthyoses, as well as genotype-phenotype correlation and year/method of genetic testing was assessed. RESULTS: Of all 198 patients who were included in the cohort, 151 were genetically tested. 81 had ichthyosis vulgaris, 43 X-linked ichthyosis, 38 autosomal recessive congenital ichthyosis (ARCI), 9 keratinopathic ichthyosis (KPI) and one exfoliative ichthyosis. 26 individuals suffered from syndromic ichthyoses. A good genotype-phenotype correlation was observed for common ichthyoses and KPI; the correlation was less good in syndromic ichthyoses. In 91 % of ARCI patients an accurate diagnosis was obtained by genetic testing. In only 33 % of syndromic ichthyoses was the definitive diagnosis suspected before genetic testing, which revealed a causative mutation in 86 % of cases. CONCLUSION: This study describes the spectrum of ichthyoses in a center of expertise and shows that genetic testing should become a diagnostic standard for this disease group.

Videolaryngoscopy with glidescope reduces cervical spine movement in patients with unsecured cervical spine.

BACKGROUND: Unconscious patients with severe trauma often require urgent endotracheal intubation. In trauma victims with possible cervical spine injury, any movement of the head and neck should be avoided. STUDY OBJECTIVES: We investigated the effect of GlideScope videolaryngoscopy on cervical spine movement compared with conventional laryngoscopy in anesthetized patients with unsecured cervical spines. METHODS: Sixty patients scheduled for elective surgery with general anesthesia and without anticipated airway problems were enrolled in the study after ethics committee approval and written informed consent. Intubation was performed with videolaryngoscopy (GlideScope(®), Verathon Inc., Bothell, WA) or conventional laryngoscopy (MacIntosh). Using video motion analysis with a lateral view, the maximum extension angle α was measured with reference to anatomical points (baseline and line drawn from processus mastoideus to os frontale [glabella]). Values were analyzed using Mann Whitney U-tests. RESULTS: The deviation of α was a median 11.8° in the videolaryngoscope group and 14.3° in the conventional group (p = 0.045), with a maximum of 19.2° (videolaryngoscopy) vs. 29.3° (conventional). Intubation by physicians with some experience in videolaryngoscopy was associated with a reduced angle deviation (α = 10.3°) compared to inexperienced physicians (12.8°, p = 0.019). Intubation time was a median 24 s (min/max 12/75 s) in the MacIntosh group and 53 s (min/max 28/210 s) in the GlideScope group. In 3 patients randomized to the conventional group in whom conventional intubation failed, intubation could be successfully performed using videolaryngoscopy. CONCLUSION: GlideScope videolaryngoscopy reduces movements of the cervical spine in patients with unsecured cervical spines and therefore might reduce the risk of secondary damage during emergency intubation of patients with cervical spine trauma.