Using Deep Learning to Identify High-Risk Patients with Heart Failure with Reduced Ejection Fraction.

Background: Deep Learning (DL) has not been well-established as a method to identify high-risk patients among patients with heart failure (HF). Objectives: This study aimed to use DL models to predict hospitalizations, worsening HF events, and 30-day and 90-day readmissions in patients with heart failure with reduced ejection fraction (HFrEF). Methods: We analyzed the data of adult HFrEF patients from the IBM® MarketScan® Commercial and Medicare Supplement databases between January 1, 2015 and December 31, 2017. A sequential model architecture based on bi-directional long short-term memory (Bi-LSTM) layers was utilized. For DL models to predict HF hospitalizations and worsening HF events, we utilized two study designs: with and without a buffer window. For comparison, we also tested multiple traditional machine learning models including logistic regression, random forest, and eXtreme Gradient Boosting (XGBoost). Model performance was assessed by area under the curve (AUC) values, precision, and recall on an independent testing dataset. Results: A total of 47 498 HFrEF patients were included; 9427 with at least one HF hospitalization. The best AUCs of DL models without a buffer window in predicting HF hospitalizations and worsening HF events in the total patient cohort were 0.977 and 0.972; with a 7-day buffer window the best AUCs were 0.573 and 0.608, respectively. The best AUCs in predicting 30- and 90-day readmissions in all adult patients were 0.597 and 0.614, respectively. An AUC of 0.861 was attained for prediction of 90-day readmission in patients aged 18-64. For all outcomes assessed, the DL approach outperformed traditional machine learning models. Discussion: The DL approach can automate feature engineering during the model learning, which can increase the clinical applicability and lead to comparable or better model performance. However, the lack of granular clinical data, and sample size and imbalance issues may have limited the model's performance. Conclusions: A DL approach using Bi-LSTM was shown to be a feasible and useful tool to predict HF-related outcomes. This study can help inform the future development and deployment of predictive tools to identify high-risk HFrEF patients and ultimately facilitate targeted interventions in clinical practice.

Community-acquired pneumonia in the emergency department: an algorithm to facilitate diagnosis and guide chest CT scan indication.

OBJECTIVE: The aim was to create and validate a community-acquired pneumonia (CAP) diagnostic algorithm to facilitate diagnosis and guide chest computed tomography (CT) scan indication in patients with CAP suspicion in Emergency Departments (ED). METHODS: We performed an analysis of CAP suspected patients enrolled in the ESCAPED study who had undergone chest CT scan and detection of respiratory pathogens through nasopharyngeal PCRs. An adjudication committee assigned the final CAP probability (reference standard). Variables associated with confirmed CAP were used to create weighted CAP diagnostic scores. We estimated the score values for which CT scans helped correctly identify CAP, therefore creating a CAP diagnosis algorithm. Algorithms were externally validated in an independent cohort of 200 patients consecutively admitted in a Swiss hospital for CAP suspicion. RESULTS: Among the 319 patients included, 51% (163/319) were classified as confirmed CAP and 49% (156/319) as excluded CAP. Cough (weight = 1), chest pain (1), fever (1), positive PCR (except for rhinovirus) (1), C-reactive protein ≥50 mg/L (2) and chest X-ray parenchymal infiltrate (2) were associated with CAP. Patients with a score below 3 had a low probability of CAP (17%, 14/84), whereas those above 5 had a high probability (88%, 51/58). The algorithm (score calculation + CT scan in patients with score between 3 and 5) showed sensitivity 73% (95% CI 66-80), specificity 89% (95% CI 83-94), positive predictive value (PPV) 88% (95% CI 81-93), negative predictive value (NPV) 76% (95% CI 69-82) and area under the curve (AUC) 0.81 (95% CI 0.77-0.85). The algorithm displayed similar performance in the validation cohort (sensitivity 88% (95% CI 81-92), specificity 72% (95% CI 60-81), PPV 86% (95% CI 79-91), NPV 75% (95% CI 63-84) and AUC 0.80 (95% CI 0.73-0.87). CONCLUSION: Our CAP diagnostic algorithm may help reduce CAP misdiagnosis and optimize the use of chest CT scan.

Accuracy of comprehensive PCR analysis of nasopharyngeal and oropharyngeal swabs for CT-scan-confirmed pneumonia in elderly patients: a prospective cohort study.

OBJECTIVES: We aimed to assess the accuracy of PCR detection of viruses and bacteria on nasopharyngeal and oropharyngeal swabs (NPS) for the diagnosis of pneumonia in elderly individuals. METHODS: We included consecutive hospitalized elderly individuals suspected of having pneumonia. At inclusion, NPS were collected from all participants and tested by PCR for the presence of viral and bacterial respiratory pathogens (index test, defined as comprehensive molecular testing). Routine diagnostic tests (blood and sputum culture, urine antigen detection) were also performed. The reference standard was the presence of pneumonia on a low-dose CT scan as assessed by two independent expert radiologists. RESULTS: The diagnosis of pneumonia was confirmed in 127 of 199 (64%) included patients (mean age 83 years, community-acquired pneumonia in 105 (83%)). A pathogen was identified by comprehensive molecular testing in 114 patients (57%) and by routine methods in 22 (11%). Comprehensive molecular testing was positive for viruses in 62 patients (31%) and for bacteria in 73 (37%). The sensitivity and specificity were 61% (95% CI 53%-69%) and 50% (95% CI 39%-61%) for comprehensive molecular testing, and 14% (95% CI 82%-21%) and 94% (95% CI 86%-98%) for routine testing, respectively. Positive likelihood ratio was 2.55 for routine methods and 1.23 for comprehensive molecular testing. CONCLUSION: Comprehensive molecular testing of NPS increases the number of pathogens detected compared with routine methods, but results are poorly predictive of the presence of pneumonia. Hence, comprehensive molecular testing is unlikely to impact clinical decision-making (NCT02467192). CLINICAL TRIALS REGISTRATION: NCT02467192.

[Does treatment improve prognosis of heart failure with preserved ejection fraction?]. / Insuffisance cardiaque à fraction d'éjection préservée: le traitement améliore-t-il le pronostic?

Heart failure with preserved ejection fraction (HF-PEF) represents half of all heart failure. Morbi-mortality for HF-PEF is similar to that of reduced ejection fraction HF (HF-REF). Diagnosis of HF-REF is difficult because of the lack of highly specific criteria. It is based on the presence of signs and symptoms of heart failure, associated with a preserved or moderately decreased left ventricular function, the absence of left ventricular dilatation, and the presence of relevant structural disease such as left ventricular hypertrophy. Despite the use of prognosis modifying drugs commonly used for HF-REF, no therapeutic strategy has been shown to reduce morbi-mortality of HF-PEF. Evidence based guidelines are limited. Management of HF-PEF therefore resides in treatment of high blood pressure and cardiac rate, that of comorbidities, and the use of diuretics in case of congestion.

[Viscosity changes in thickened water due to the addition of highly prescribed drugs in geriatrics]. / Cambios en la viscosidad del agua con espesantes por la adición de fármacos altamente prescritos en geriatría.

INTRODUCTION: Dysphagia is a swallowing disorder with a high incidence in the geriatric patient related with an increased risk for undernutrition and pneumonia due to bronchial aspiration. In this condition, it is usual to add commercial thickeners in liquids, as well as the addition of drugs in this mixture to improve their administration. However, there are no studies regarding the possible change in viscosity produced by their addition. OBJECTIVES: To assess the change in viscosity of water thickened with commercial products by adding the drugs frequently used in elderly patients. METHODS: Samples of water mixed with the commercial thickener Resource (modified corn starch) or Nutilis (modified corn starch, maltodextrin, and gums: tara, xhantan, and guar) to achieve an intermediate consistence as "honey". The viscosity of these samples was measured as well as for similar samples to which one of the following drugs was added: galantamine, rivastigmin, ciprofloxacin, cholecalciferol, memantine, fosfomycin, calcium, and amoxicillin/clavulanic acid. RESULTS: In the samples with Resource thickener we observed decreased viscosity by adding galantamine, memantine, fosfomycin or calcium, and increased viscosity with amoxicillin/clavulanic acid. The viscosity of the samples with Nutilis® decreased with galantamine, rivastigmine, amoxicillin/clavulanic acid, fosfomycin and calcium. CONCLUSION: The viscosity of water with commercial thickeners may be affected by some drugs or their preservatives, which may influence the swallowing capability. It is recommended to perform further in vitro and in vivo studies in order to adjust these formulations if necessary.

[Q fever: a cause of fever of unknown origin in Switzerland]. / La fièvre Q: une cause d'état fébrile sans foyer en Suisse.

We describe two cases of Q fever in previously healthy women presenting with fever of unknown origin. The diagnosis was made after several days of investigations. Symptoms and signs of acute or chronic Coxiella burnetii infection are protean and non-specific. Q fever should be included in the differential diagnosis of fever of unknown origin and appropriate serologic studies should be done. We review the clinical presentation of Q fever. Use of serology for the diagnosis and the follow-up is discussed.

Cambios en la viscosidad del agua con espesantes por la adición de fármacos altamente prescritos en geriatría / Viscosity changes in thickened water due to the addition of highly prescribed drugs in geriatrics

Introducción: La disfagia es una alteración de la deglución con una elevada incidencia en el paciente geriátrico relacionada con un aumento del riesgo de desnutrición y neumonía por broncoaspiración. La adición de espesantes comerciales en líquidos es frecuente en esta situación, así como la adición de fármacos en esta mezcla para facilitar su administración. Sin embargo, no existen estudios referentes al posible cambio de viscosidad producido por la adición de los mismos. Objetivos: Evaluar la variación ejercida sobre la viscosidad del agua espesada con preparados comerciales al añadir fármacos frecuentemente utilizados en pacientes de edad avanzada. Métodos: Se prepararon muestras de agua con espesante comercial Resource® (almidón de maíz modificado) o Nutilis® (almidón modificado de maíz, maltodextrina y gomas: tara, xantana y guar) para conseguir consistencia intermedia tipo "miel". Se midió la viscosidad de estas muestras y para muestras similares a las que se había añadido alguno de los siguientes fármacos: galantamina, rivastigmina, ciprofloxacino, colecalciferol, memantina, fosfomicina, calcio y amoxilina/clavulánico. Resultados: En las muestras con espesante Resource® se observó una disminución de la viscosidad al añadir galantamina, memantina, fosfomicina o calcio, y un aumento de la viscosidad con amoxicilina/clavulánico. La viscosidad de la muestras con Nutilis® disminuyó con galantamina, rivastigmina, amoxicilina/clavulánico, fosfomicina y calcio. Conclusión: La viscosidad del agua con espesantes comerciales puede verse afectada por algunos fármacos o sus excipientes, lo que puede incidir en la capacidad de deglución de los mismos. Es aconsejable realizar más estudios in vitro e in vivo para valorar ajustar dichas pautas en caso necesario (AU)

Introduction: Dysphagia is a swallowing disorder with a high incidence in the geriatric patient related with an increased risk for undernutrition and pneumonia due to bronchial aspiration. In this condition, it is usual to add commercial thickeners in liquids, as well as the addition of drugs in this mixture to improve their administration. However, there are no studies regarding the possible change in viscosity produced by their addition. Objectives: To assess the change in viscosity of water thickened with commercial products by adding the drugs frequently used in elderly patients. Methods: Samples of water mixed with the commercial thickener Resource® (modified corn starch) or Nutilis® (modified corn starch, maltodextrin, and gums: tara, xhantan, and guar) to achieve an intermediate consistence as "honey". The viscosity of these samples was measured as well as for similar samples to which one of the following drugs was added: galantamine, rivastigmin, ciprofloxacin, cholecalciferol, memantine, fosfomycin, calcium, and amoxicillin/clavulanic acid. Results: In the samples with Resource® thickener we observed decreased viscosity by adding galantamine, memantine, fosfomycin or calcium, and increased viscosity with amoxicillin/clavulanic acid. The viscosity of the samples with Nutilis® decreased with galantamine, rivastigmine, amoxicillin/clavulanic acid, fosfomycin and calcium. Conclusion: The viscosity of water with commercial thickeners may be affected by some drugs or their preservatives, which may influence the swallowing capability. It is recommended to perform further in vitro and in vivo studies in order to adjust these formulations if necessary (AU)

[Clinical itinerary for heart failure: a program designed by primary care practitioners in Geneva]. / Itinéraire clinique pour l'insuffisance cardiaque: un programme conçu par des médecins de premier recours genevois.

Episodes of heart failure impact on patients' quality of life as well as their morbidity and mortality. This article describes a series of interventions designed by a group of primary care practitioners in Geneva. Some interventions aim to improve patients' autonomy in identifying the first signs of heart failure to act immediately. Others focus on patients' motivation to adopt appropriate behaviours (physical activity, etc.). And finally others have the objective to improve coordination between ambulatory and hospital care, as well as the transmission of clinical information. The implementation of these interventions highlights the need for individualised objectives of care in complex cases where patients have several co-morbidities and/or complicated social situations. In these situations an interdisciplinary approach is also essential.

[Opioid rotation: from theory to practice. Interdisciplinary recommendations from the Geneva University Hospitals pain network]. / Rotation des opioïdes: de ta théorie à la pratique. Recommandations interdisciptinaires du réseau douleur des HUG.

Opioids are widely used to treat moderate to severe pain of cancer or non cancer origin. Although opioids provide an adequate analgesia in many patients, their use can be limited by inefficacy and/or intolerable side effects. Opioid rotation is one of the strategies that have been proposed to overcome these therapeutic difficulties. This article revisits the concept of opioid rotation, from pharmacological rational to clinical application.

[Critical pathway for community-acquired pneumonia]. / Itinéraire clinique et pneumonie acquise à domicile.

A clinical pathway is a methodological tool for standardizing medical practice, improving the quality and efficiency of care delivery, and enhancing the diffusion of evidence-based medicine. Despite the fact that a majority of trials have shown that the use of clinical pathways improves certain specific outcomes such as length of stay or complications, the overall impact of these pathways in the clinical setting has yet to be documented. In the setting of community-acquired pneumonia, a few observational and one large randomized trial have shown positive effects on various outcomes. We describe in this article the clinical pathway for community-acquired pneumonia developed at our institution.

[Update in hospital medicine 2008]. / Médecine interne hospitalière.

This paper summarizes several important studies published during the previous year that have an impact on the practice of inpatient internal medicine, because they either modify or reinforce current practices. The selected domains include cardiovascular disease, for example the management of hypertension in very old patients, the effects of blockade of the renin-angiotensin-aldosterone system, and the use of biomarkers in cardiology; neurovascular pathology, specifically the prognosis of transient ischemic attacks and some aspects of cardioembolic stroke due to atrial fibrillation. Other topics include pneumonia prognosis, the management of ascitis fluid or of septic shock, and methodology.

[Empiric therapy of community-acquired pneumonia: a transatlantic controversy]. / Antibiothérapie des pneumonies acquises à domicile--une controverse transatlantique.

North American and European guidelines for empiric antimicrobial therapy of community-acquired pneumonia differ. This stems from a different mechanism of pneumococcal resistance to macrolides, and a different perception of the need to cover atypical germs. For mild cases, two recent meta-analyses conclude that this coverage is not associated with any benefit. For more severe pneumonias, data are scarce and experts generally recommend wide coverage. Except for particular situations, the new fluoroquinolones should not represent the first-line therapy because of rapidly growing resistance to these antibiotics.

The development of inhibitory synaptic specializations in the mouse deep cerebellar nuclei.

Using confocal laser scanning microscopy and immunohistochemistry, this study shows the complete morphological development of GABAergic synaptic contacts between Purkinje cells and neurons of the deep cerebellar nuclei of the mouse. Firstly, presynaptic varicosities visualized with antibodies against synaptophysin, synapsin or glutamic acid decarboxylase, were detected when the postsynaptic GABA(A) receptors were not yet aggregated in the membrane but had a diffuse cytoplasmic distribution, which indicated a lead in maturation of presynaptic terminals over target cells. Secondly, receptor aggregates developed suddenly after an initial week of diffuse expression and these clusters matured into more numerous and larger synaptic aggregates. During this postsynaptic maturation, the presynaptic varicosities develop into numerous and well-defined spots. As soon as both pre- and postsynaptic clusters were detectable, these sites are always colocalized. We therefore consider the aggregation of postsynaptic receptor during development as a landmark of synapse formation. Our observations are consistent with a developmental model in which the presynaptic neuron differentiates its axon before the target neuron expresses the mature form of its receptors on the membrane. The presynaptic neuron can therefore instruct the target neuron about the distribution and aggregation of the postsynaptic receptors at the synapse.

Implication of the extracellular disulfide bond on myelin protein zero expression.

Mutations in myelin protein zero (P0) are responsible for several peripheral neuropathies. We studied transport and membrane integration of the truncated P0 mutants using transfected oligodendroglial cell line (Oln93). Starting with rat cDNA, we produced two P0 deletions. The first, called P0-Tyr contains a 66 amino acid deletion in the extracellular domain and a tyrosine at the new position 32. In the second, called P0-Cys, the tyrosine 32 is replaced by a cysteine. This replacement restores a disulfide bond in the extracellular domain. Our results show that P0 proteins, truncated or not, were expressed in the plasma membrane of the transfected cells. Transcription rates of both mutants were normal. However, P0-Tyr was detected in only 3-5% of the cells compared to the P0-Cys and the wild type. Thus, the disulfide bond in the extracellular domain is important for stability and correct addressing of the P0 protein.

A large outdoor radial maze for comparative studies in birds and mammals.

For a comparative neurobiological analysis of spatial learning and memory, a large outdoor eight-arm radial maze was constructed which permits behavioral assessment of many avian and mammalian species both from the laboratory or the wild, using the same metric space and session schedules. It consists of a central part of 250cm diameter, and has arms of 650cm length, 170cm height and 80cm width. In order to determine appropriate training schedules for comparison of different species, we tested four mammalian and two avian species during 9-15 sessions: 18 albino rats (Rattus norvegicus), nine outdoors and nine in a conventional small indoor maze; six guinea pigs (Cavia porcellus); six rabbits (Oryctolagus cuniculus); five hedgehogs (Erinaceus europaeus); seven hooded crows (Corvus corone cornix) and six chickens (Gallus domesticus). Rats learned fast in both mazes yet significantly better in the large one. Good-to-excellent learning was also observed in juvenile rabbits and wild-caught crows, although the latter tended to avoid arms in the vicinity of the observer. Hedgehogs and chickens did not show significant learning as a group, but some individuals appeared to learn the task. Guinea pigs remained continuously passive and could not be trained. Thus, in spite of species-specific demands for reward, adaptation and pre-training, this type of radial maze permits to directly compare a wide variety of species. Such comparability is essential for an analysis of underlying neurobiological mechanisms.