[Implementation of the new collective agreement in 2021 in urological clinics : Expectations and reality]

The amendment to the collective agreement is intended to significantly improve the working conditions of physicians and includes longer-term duty scheduling, work on a maximum of two weekends per month, less overtime. Smaller hospitals often have problems implementing these requirements and have to make compromises. At least the overtime is now better paid-overall, an improvement in working conditions can only be achieved by increasing the number of staff, then better and more intensive training is also possible.

Using Automated Machine Learning to Predict the Mortality of Patients With COVID-19: Prediction Model Development Study

BACKGROUND

During a pandemic, it is important for clinicians to stratify patients and decide who receives limited medical resources. Machine learning models have been proposed to accurately predict COVID-19 disease severity. Previous studies have typically tested only one machine learning algorithm and limited performance evaluation to area under the curve analysis. To obtain the best results possible, it may be important to test different machine learning algorithms to find the best prediction model.

OBJECTIVE

In this study, we aimed to use automated machine learning (autoML) to train various machine learning algorithms. We selected the model that best predicted patients' chances of surviving a SARS-CoV-2 infection. In addition, we identified which variables (ie, vital signs, biomarkers, comorbidities, etc) were the most influential in generating an accurate model.

METHODS

Data were retrospectively collected from all patients who tested positive for COVID-19 at our institution between March 1 and July 3, 2020. We collected 48 variables from each patient within 36 hours before or after the index time (ie, real-time polymerase chain reaction positivity). Patients were followed for 30 days or until death. Patients' data were used to build 20 machine learning models with various algorithms via autoML. The performance of machine learning models was measured by analyzing the area under the precision-recall curve (AUPCR). Subsequently, we established model interpretability via Shapley additive explanation and partial dependence plots to identify and rank variables that drove model predictions. Afterward, we conducted dimensionality reduction to extract the 10 most influential variables. AutoML models were retrained by only using these 10 variables, and the output models were evaluated against the model that used 48 variables.

RESULTS

Data from 4313 patients were used to develop the models. The best model that was generated by using autoML and 48 variables was the stacked ensemble model (AUPRC=0.807). The two best independent models were the gradient boost machine and extreme gradient boost models, which had an AUPRC of 0.803 and 0.793, respectively. The deep learning model (AUPRC=0.73) was substantially inferior to the other models. The 10 most influential variables for generating high-performing models were systolic and diastolic blood pressure, age, pulse oximetry level, blood urea nitrogen level, lactate dehydrogenase level, D-dimer level, troponin level, respiratory rate, and Charlson comorbidity score. After the autoML models were retrained with these 10 variables, the stacked ensemble model still had the best performance (AUPRC=0.791).

CONCLUSIONS

We used autoML to develop high-performing models that predicted the survival of patients with COVID-19. In addition, we identified important variables that correlated with mortality. This is proof of concept that autoML is an efficient, effective, and informative method for generating machine learning-based clinical decision support tools.

AKI in Hospitalized Patients with and without COVID-19: A Comparison Study

BACKGROUND

Reports from centers treating patients with coronavirus disease 2019 (COVID-19) have noted that such patients frequently develop AKI. However, there have been no direct comparisons of AKI in hospitalized patients with and without COVID-19 that would reveal whether there are aspects of AKI risk, course, and outcomes unique to this infection.

METHODS

In a retrospective observational study, we evaluated AKI incidence, risk factors, and outcomes for 3345 adults with COVID-19 and 1265 without COVID-19 who were hospitalized in a large New York City health system and compared them with a historical cohort of 9859 individuals hospitalized a year earlier in the same health system. We also developed a model to identify predictors of stage 2 or 3 AKI in our COVID-19.

RESULTS

We found higher AKI incidence among patients with COVID-19 compared with the historical cohort (56.9% versus 25.1%, respectively). Patients with AKI and COVID-19 were more likely than those without COVID-19 to require RRT and were less likely to recover kidney function. Development of AKI was significantly associated with male sex, Black race, and older age (>50 years). Male sex and age >50 years associated with the composite outcome of RRT or mortality, regardless of COVID-19 status. Factors that were predictive of stage 2 or 3 AKI included initial respiratory rate, white blood cell count, neutrophil/lymphocyte ratio, and lactate dehydrogenase level.

CONCLUSIONS

Patients hospitalized with COVID-19 had a higher incidence of severe AKI compared with controls. Vital signs at admission and laboratory data may be useful for risk stratification to predict severe AKI. Although male sex, Black race, and older age associated with development of AKI, these associations were not unique to COVID-19.

[Regional Deprivation in Germany: Nation-wide Analysis of its Association with Mortality Using the German Index of Multiple Deprivation (GIMD)]

BACKGROUND

Deprivation indices are increasingly being used to assess the effects of contextual factors on health. In Germany, the recently developed 'German Index of Multiple Deprivation (GIMD)' integrates various dimensions of regional deprivation. We aim to assess the validity of the GIMD through a recalculation using more recent rural and urban district level data and by analysing its association with mortality at the national level.

METHODS

We calculated a new version of the GIMD based on data from 2007 to 2010 for all 412 rural and urban districts in Germany. Mortality was quantified using indirectly standardised mortality ratios (SMRs). Correlation analyses and Poisson regression analyses were used to assess the association between the GIMD scores and total mortality, as well as premature mortality (< 65 years).

RESULTS

Correlation analyses showed a positive association between the GIMD and both total mortality (p<0.001) and premature mortality (p<0.001). In the Poisson regression analyses, rural and urban districts in the quintile with the highest deprivation showed a significantly elevated risk of total mortality (RR: 1.29; 95% CI: 1.28-1.30) as well as premature mortality (RR: 1.50; 95% CI: 1.47-1.53), compared to the districts in the lowest quintile.

CONCLUSION

The association between regional deprivation and mortality has already been shown for the federal state of Bavaria. Using more recent data, this relationship could be confirmed here for Germany as a whole. The GIMD has been shown to be able to effectively assess regional deprivation. Concerning public health policy, the significant, positive and stable association between regional deprivation and mortality indicates an increased need for health care provision particularly in the most deprived districts. Further studies should examine, for example, whether and how the allocation of districts to quintiles of regional deprivation changes over time, and how this affects mortality.

Preservation of cocoa antioxidant activity, total polyphenols, flavan-3-ols, and procyanidin content in foods prepared with cocoa powder

Little is known about the effects of common cooking processes on cocoa flavanols. Antioxidant activity, total polyphenols (TP), flavanol monomers, and procyanidin oligomers were determined in chocolate frosting, a hot cocoa drink, chocolate cookies, and chocolate cake made with natural cocoa powder. Recoveries of antioxidant activity, TP, flavanol monomers, and procyanidins ranged from 86% to over 100% in the chocolate frosting, hot cocoa drink, and chocolate cookies. Losses were greatest in the chocolate cake with recoveries ranging from 5% for epicatechin to 54% for antioxidant activity. The causes of losses in baked chocolate cakes were investigated by exchanging baking soda with baking powder or combinations of the 2 leavening agents. Use of baking soda as a leavening agent was associated with increased pH and darkening color of cakes. Losses of antioxidant activity, TP, flavanol monomers, and procyanidins were associated with an increased extractable pH of the baked cakes. Chocolate cakes made with baking powder for leavening resulted in an average extractable pH of 6.2 with essentially complete retention of antioxidant activity and flavanol content, but with reduced cake heights and lighter cake color. Commercially available chocolate cake mixes had final pHs above 8.3 and contained no detectable monomeric flavanols after baking. These results suggest that baking soda causes an increase in pH and subsequent destruction of flavanol compounds and antioxidant activity. Use of an appropriate leavening agent to moderate the final cake pH to approximately 7.25 or less results in both good leavening and preservation of cocoa flavanols and procyanidins.

Mono- and di-nickellaazaphosphiranes of mono- and bis-(amido)cyclodiphosph(III)azanes

The syntheses and solid-state structures of the first three-membered nickel-phosphorus-nitrogen ring compounds, having anionic four-electron P=N moieties are reported.

Syntheses and structures of heterobicyclic bis(tert-butylamido)cyclodiphosph(III)azane compounds having phosphorus(III) and arsenic(III) centers

Syntheses and single-crystal X-ray diffraction studies of heterobicyclic cyclodiphosphazanes having central phosphorus(III) and arsenic(III) atoms are described. Interaction of PCl3 or AsCl3 with cis-[(tBuNP)2(tBuNLixTHF)2] produced the isomorphous ([(tBuNP)2(tBuN)2]ECl); E = P(1), As(4), respectively. These Cs-symmetric molecules crystallize with two molecules in the monoclinic space group P2(1)/m. Unit cell dimensions of 1 are (293 K) a = 9.777(1) A, b = 11.745(1) A, c = 9.986(2) A, and beta = 97.44(1) degrees; those of 4 are (213 K) a = 9.688(3) A, b = 11.873(3) A, c = 9.975(3) A, and beta = 97.80(3) degrees. When ([(tBuNP)2(tBuN)2]PCl) was treated with NaN3 or LiN(SiMe3)2, ([(tBuNP)2(tBuN)2]PN3)(2) and ([(tBuNP)2(tBuN)2]PN(SiMe3)2)(3), respectively, were obtained. Compound 2 crystallizes in the monoclinic space group P2(1)/m and has until-cell dimensions (213 K) of a = 9496(7) A, b = 12455(7) A, c = 10043(6) A, and beta = 9723(4) degrees, Z = 2.

Infections with nonthoracotomy implantable cardioverter defibrillators: can these be prevented? Endotak Lead Clinical Investigators

Nonthoracotomy ICDs are believed to be the best therapeutic modality for treatment of life-threatening ventricular arrhythmias. Little is known about the risk of infection with initial implantation of these devices. We studied the incidence, clinical characteristics, and risk factors associated with infections in 1,831 patients with nonthoracotomy ICD from the Endotak-C nonthoracotomy lead registry of Cardiac Pacemakers, Inc. A transvenous lead was implanted in 950 patients (51.9%) and a combination transvenous plus subcutaneous patch was used in 881 patients (48.1%). Nine preselected data variables were studied, and all investigators identified as having patients with infections were personally contacted. Infections occurred in 22 (1.2%) of 1,831 patients receiving this nonthoracotomy ICD system. The mean time to infection was 5.7 +/- 6.5 months (range 1-25 months). Staphylococci were isolated in 58% of patients with reported infection. The presence of a subcutaneous defibrillator patch system was associated with the development of infection. Six of 950 patients (0.63%) with a totally transvenous lead system developed infection versus 16 of 838 (1.9%) patients with a transvenous lead plus subcutaneous patch system configuration (P = 0.015, Chi-square test), with an unadjusted estimated odds ratio of 3.06 (CI 1.19-7.86). The risk of infection encountered with the nonthoracotomy ICD is low, estimated from our data to be 1.2%. Placement of a subcutaneous defibrillator patch appears to be an independent risk factor for development of infection.

Preservation of coronary flow reserve in stunned myocardium

Microvascular obstruction and persistent focal ischemia have been suggested as a possible cause of myocardial dysfunction (stunning) after brief coronary occlusion. Microvascular occlusion should result in a reduction in maximal coronary flow reserve, although resting transmural coronary flow may be maintained by release of local vasodilators, such as adenosine. To test the microvascular occlusion hypothesis, coronary flow reserve was measured in 14 anesthetized dogs, before and after myocardial stunning produced by 10 min of ischemia. Intracoronary adenosine infusion (5,900 microM/min) increased coronary flow to the same degree in normal [195 +/- 20 (SE) ml/min] and stunned (212 +/- 23 ml/min) myocardium. Peak hyperemic flow after 100 s of coronary occlusion was also similar in normal (205 +/- 25 ml/min) and stunned (218 +/- 23 ml/min) myocardium. The adenosine antagonist 8-phenyltheophylline (5 mg/kg) reduced the flow response to exogenous adenosine, but neither resting coronary flow nor peak hyperemic flow in stunned myocardium was altered. In stunned myocardium, myocardial shortening at rest (0.2 +/- 2.0%) increased during reactive hyperemia (to 13.8 +/- 2.5%, P less than 0.01), but shortening promptly returned to basal levels after each hyperemia. These findings indicate that fixed microvascular occlusion is unlikely to be an important factor in the pathogenesis of stunned myocardium and that local adenosine release does not appear to have a compensatory role in coronary vasoregulation in stunned myocardium.

Lymph node dissection in malignant melanoma

Regional node dissection was performed in 120 patients with malignant melanoma. Patients with clinically negative nodes had a 90% survival, whereas patients with enlarged nodes had a 15% survival. In 45 inguinal dissections, usually associated with a high morbidity, there was only a 4.5% incidence of infection and a 6.5% incidence of skin edge necrosis. For all types of node dissection, the overall incidence of wound infection was 5.8%, and that of necrosis of skin edges was 5%.