Pronostic du coma après lésion cérébrale aiguë chez l'homme

Contexte: L'ensemble des phénomènes aigus suivant un arrêt cardio-respiratoire (ACR) est décrit sous le nom de maladie de post-réanimation (MPR) (post-resuscitation disease). Celle- ci est la conséquence du syndrome de reperfusion et est caractérisée par une réponse inflammatoire systémique intense, d'allure septique. La procalcitonine (PCT) est un marqueur aigu de la réponse inflammatoire systémique, qui a été beaucoup étudiée aux soins intensifs (SI) dans le contexte du sepsis, et constitue un outil diagnostic et pronostique important. Toutefois la PCT n'est pas un marqueur spécifique pour le sepsis mais peut également augmenter lors de réponse inflammatoire systémique d'origine non infectieuse. Objectifs: 1) Evaluer s'il existe une corrélation entre la valeur plasmatique de PCT et la MPR ; 2) examiner la relation entre le taux au pic de PCT et le pronostic des patients avec coma post-ACR ; 3) comparer la valeur pronostique de la PCT à celle d'un marqueur pronostic connu du coma post-anoxique tel que la neuron specific enolase (NSE). Méthodologie: Analyse d'une base de données prospective comprenant des patients admis aux SI du centre hospitalier universitaire vaudoise (CHUV) entre décembre 2009 et juillet 2011 en raison d'un ACR et traités par hypothermie thérapeutique (33 - 34 °C pendant 24h), selon notre protocole standard de prise en charge. La concentration plasmatique de PCT est mesurée à 24-72h après ACR, la valeur maximale (PCTmax) étant incluse dans l'analyse. La durée de l'arrêt circulatoire et le score de SOFA (Sequential Organ Failure Assessment) sont utilisés pour quantifier la sévérité de la MPR. Le pronostic est composé de la mortalité hospitalière, ainsi que la mortalité et la récupération neurologique à trois mois, mesurée avec le score de « Cerebral Performance Categories » (CPC), dichotomisé en bonne récupération (CPC 1 = pas de handicap ; CPC 2 = handicap modéré) et mauvaise récupération (CPC 3 = handicap sévère ; CPC 4 = état végétatif ; CPC 5 = décès). Résultats: 68 patients consécutifs (âge médian 65 ans, durée médiane totale de l'arrêt circulatoire [time to ROSC] 20.5 min) ont été étudiés. La PCTmax corrélait avec la durée de l'arrêt circulatoire (p = 0.001) ainsi qu'avec les scores de SOFA à l'admission et aux jours 1 et 2 (p<0.001 pour les trois associations). Une association significative a été observée entre la PCTmax et la survie hospitalière (médiane 3.9 [écart interquartile (EI) 1.0 - 16.8] chez les non-survivants vs. 1.4 [EI 0.6 - 6.2] ng/ml chez les survivants, p=0.032) et à trois mois (médiane 3.8 ([EI 1.0 - 15.6] vs. 1.4 [EI 0.5 - 6.0] ng/ml, p=0.034). La PCTmax était aussi plus basse chez les patients avec bonne récupération neurologique à trois mois (p=0.064). En comparaison avec la NSEmax, la PCTmax avait une valeur prédictive supérieure pour la sévérité de la maladie de post-réanimation et inférieure pour le pronostic. Conclusions: La valeur plasmatique maximale de PCT corrèle avec la sévérité de la MPR et est associé à la mortalité et à l'état neurologique à trois mois après coma post-anoxique. Ces données suggèrent que la PCT peut être un marqueur utile dans la prise en charge des patients comateux après ACR et hypothermie thérapeutique. Des études à plus large échelle sont en cours pour confirmer ces résultats.

A fatal overdose of cocaine associated with coingestion of marijuana, buprenorphine, and fluoxetine. Body fluid and tissue distribution of cocaine and its metabolites determined by hydrophilic interaction chromatography-mass spectrometry(HILIC-MS).

Chromatographic separation of highly polar basic drugs with ideal ionspray mass spectrometry volatile mobile phases is a difficult challenge. A new quantification procedure was developed using hydrophilic interaction chromatography-mass spectrometry with turbo-ionspray ionization in the positive mode. After addition of deuterated internal standards and simple clean-up liquid extraction, the dried extracts were reconstituted in 500 microL pure acetonitrile and 5 microL was directly injected onto a Waters Atlantis HILIC 150- x 2.1-mm, 3-microm column. Chromatographic separations of cocaine, seven metabolites, and anhydroecgonine were obtained by linear gradient-elution with decreasing high concentrations of acetonitrile (80-56% in 18 min). This high proportion of organic solvent makes it easier to be coupled with MS. The eluent was buffered with 2 mM ammonium acetate at pH 4.5. Except for m-hydroxy-benzoylecgonine, the within-day and between-day precisions at 20, 100, and 500 ng/mL were below 7 and 19.1%, respectively. Accuracy was also below +/- 13.5% at all tested concentrations. The limit of quantification was 5 ng/mL (%Diff < 16.1, %RSD < 4.3) and the limit of detection below 0.5 ng/mL. This method was successfully applied to a fatal overdose. In Switzerland, cocaine abuse has dramatically increased in the last few years. A 45-year-old man, a known HIV-positive drug user, was found dead at home. According to relatives, cocaine was self-injected about 10 times during the evening before death. A low amount of cocaine (0.45 mg) was detected in the bloody fluid taken from a syringe discovered near the corpse. Besides injection marks, no significant lesions were detected during the forensic autopsy. Toxicological investigations showed high cocaine concentrations in all body fluids and tissues. The peripheral blood concentrations of cocaine, benzoylecgonine, and methylecgonine were 5.0, 10.4, and 4.1 mg/L, respectively. The brain concentrations of cocaine, benzoylecgonine, and methylecgonine were 21.2, 3.8, and 3.3 mg/kg, respectively. The highest concentrations of norcocaine (about 1 mg/L) were measured in bile and urine. Very high levels of cocaine were determined in hair (160 ng/mg), indicating chronic cocaine use. A low concentration of anhydroecgonine methylester was also found in urine (0.65 mg/L) suggesting recent cocaine inhalation. Therapeutic blood concentrations of fluoxetine (0.15 mg/L) and buprenorphine (0.1 microg/L) were also discovered. A relatively high concentration of Delta(9)-THC was measured both in peripheral blood (8.2 microg/L) and brain cortex (13.5 microg/kg), suggesting that the victim was under the influence of cannabis at the time of death. In addition, fluoxetine might have enhanced the toxic effects of cocaine because of its weak pro-arrhythmogenic properties. Likewise, combination of cannabinoids and cocaine might have increase detrimental cardiovascular effects. Altogether, these results indicate a lethal cocaine overdose with a minor contribution of fluoxetine and cannabinoids.

A century of research: what have we learned about the interaction of Trypanosoma cruzi with host cells?

Since the discovery of Trypanosoma cruzi and the brilliant description of the then-referred to "new tripanosomiasis" by Carlos Chagas 100 years ago, a great deal of scientific effort and curiosity has been devoted to understanding how this parasite invades and colonises mammalian host cells. This is a key step in the survival of the parasite within the vertebrate host, and although much has been learned over this century, differences in strains or isolates used by different laboratories may have led to conclusions that are not as universal as originally interpreted. Molecular genotyping of the CL-Brener clone confirmed a genetic heterogeneity in the parasite that had been detected previously by other techniques, including zymodeme or schizodeme (kDNA) analysis. T. cruzi can be grouped into at least two major phylogenetic lineages: T. cruzi I, mostly associated with the sylvatic cycle and T. cruzi II, linked to human disease; however, a third lineage, T. cruziIII, has also been proposed. Hybrid isolates, such as the CL-Brener clone, which was chosen for sequencing the genome of the parasite (Elias et al. 2005, El Sayed et al. 2005a), have also been identified. The parasite must be able to invade cells in the mammalian host, and many studies have implicated the flagellated trypomastigotes as the main actor in this process. Several surface components of parasites and some of the host cell receptors with which they interact have been described. Herein, we have attempted to identify milestones in the history of understanding T. cruzi- host cell interactions. Different infective forms of T. cruzi have displayed unexpected requirements for the parasite to attach to the host cell, enter it, and translocate between the parasitophorous vacuole to its final cytoplasmic destination. It is noteworthy that some of the mechanisms originally proposed to be broad in function turned out not to be universal, and multiple interactions involving different repertoires of molecules seem to act in concert to give rise to a rather complex interplay of signalling cascades involving both parasite and cellular components.

Simultaneous estimation of indirect and interaction effects using structural equation models

Interaction effects are usually modeled by means of moderated regression analysis. Structural equation models with non-linear constraints make it possible to estimate interaction effects while correcting formeasurement error. From the various specifications, Jöreskog and Yang's(1996, 1998), likely the most parsimonious, has been chosen and further simplified. Up to now, only direct effects have been specified, thus wasting much of the capability of the structural equation approach. This paper presents and discusses an extension of Jöreskog and Yang's specification that can handle direct, indirect and interaction effects simultaneously. The model is illustrated by a study of the effects of an interactive style of use of budgets on both company innovation and performance

Correcting Measurement Error Bias in Interaction Models with Small Samples

Several methods have been suggested to estimate non-linear models with interaction terms in the presence of measurement error. Structural equation models eliminate measurement error bias, but require large samples. Ordinary least squares regression on summated scales, regression on factor scores and partial least squares are appropriate for small samples but do not correct measurement error bias. Two stage least squares regression does correct measurement error bias but the results strongly depend on the instrumental variable choice. This article discusses the old disattenuated regression method as an alternative for correcting measurement error in small samples. The method is extended to the case of interaction terms and is illustrated on a model that examines the interaction effect of innovation and style of use of budgets on business performance. Alternative reliability estimates that can be used to disattenuate the estimates are discussed. A comparison is made with the alternative methods. Methods that do not correct for measurement error bias perform very similarly and considerably worse than disattenuated regression

Differential epipolar constraint in mobile robot egomotion estimation

The estimation of camera egomotion is a well established problem in computer vision. Many approaches have been proposed based on both the discrete and the differential epipolar constraint. The discrete case is mainly used in self-calibrated stereoscopic systems, whereas the differential case deals with a unique moving camera. The article surveys several methods for mobile robot egomotion estimation covering more than 0.5 million samples using synthetic data. Results from real data are also given

Efficient learning of reactive robot behaviors with a Neural-Q_learning approach

The purpose of this paper is to propose a Neural-Q_learning approach designed for online learning of simple and reactive robot behaviors. In this approach, the Q_function is generalized by a multi-layer neural network allowing the use of continuous states and actions. The algorithm uses a database of the most recent learning samples to accelerate and guarantee the convergence. Each Neural-Q_learning function represents an independent, reactive and adaptive behavior which maps sensorial states to robot control actions. A group of these behaviors constitutes a reactive control scheme designed to fulfill simple missions. The paper centers on the description of the Neural-Q_learning based behaviors showing their performance with an underwater robot in a target following task. Real experiments demonstrate the convergence and stability of the learning system, pointing out its suitability for online robot learning. Advantages and limitations are discussed

Semi-online neural-Q_leaming for real-time robot learning

Reinforcement learning (RL) is a very suitable technique for robot learning, as it can learn in unknown environments and in real-time computation. The main difficulties in adapting classic RL algorithms to robotic systems are the generalization problem and the correct observation of the Markovian state. This paper attempts to solve the generalization problem by proposing the semi-online neural-Q_learning algorithm (SONQL). The algorithm uses the classic Q_learning technique with two modifications. First, a neural network (NN) approximates the Q_function allowing the use of continuous states and actions. Second, a database of the most representative learning samples accelerates and stabilizes the convergence. The term semi-online is referred to the fact that the algorithm uses the current but also past learning samples. However, the algorithm is able to learn in real-time while the robot is interacting with the environment. The paper shows simulated results with the "mountain-car" benchmark and, also, real results with an underwater robot in a target following behavior

Vision-based localization of an underwater robot in a structured environment

This paper presents a vision-based localization approach for an underwater robot in a structured environment. The system is based on a coded pattern placed on the bottom of a water tank and an onboard down looking camera. Main features are, absolute and map-based localization, landmark detection and tracking, and real-time computation (12.5 Hz). The proposed system provides three-dimensional position and orientation of the vehicle along with its velocity. Accuracy of the drift-free estimates is very high, allowing them to be used as feedback measures of a velocity-based low-level controller. The paper details the localization algorithm, by showing some graphical results, and the accuracy of the system

Towards Direct Policy Search Reinforcement Learning for Robot Control

This paper proposes a high-level reinforcement learning (RL) control system for solving the action selection problem of an autonomous robot. Although the dominant approach, when using RL, has been to apply value function based algorithms, the system here detailed is characterized by the use of direct policy search methods. Rather than approximating a value function, these methodologies approximate a policy using an independent function approximator with its own parameters, trying to maximize the future expected reward. The policy based algorithm presented in this paper is used for learning the internal state/action mapping of a behavior. In this preliminary work, we demonstrate its feasibility with simulated experiments using the underwater robot GARBI in a target reaching task