Prediction de l'année 1649. sur l'emprisonnement du prince du sang, surnommé la Cuirasse

[Mazarinade. 1650]

Utilité des biomarqueurs du sepsis en réanimation [Usefulness of biomarkers of sepsis in the ICU].

Despite some progress, the mortality of severe sepsis and septic shock remains high. Immunotherapy directed against inflammatory mediators failed, but new treatments more specifically tailored to individual situations are actively investigated. C-reactive protein (CRP) and procalcitonin (PCT) have not demonstrated to be useful for individual prognostic stratification. New biomarkers such as pancreatic stone protein (PSP) or growth arrest specific protein 6 (Gas6) could improve this prediction. Combined with the clinical course, "PCT" allows to tailor individually the duration of antibiotic therapy in ICU patients. This still contested innovative approach significantly reduces overall exposure to antibiotics.

Strategies for sequential prediction of stationary time series

We present simple procedures for the prediction of a real valued sequence. The algorithms are based on a combinationof several simple predictors. We show that if the sequence is a realization of a bounded stationary and ergodic random process then the average of squared errors converges, almost surely, to that of the optimum, given by the Bayes predictor. We offer an analog result for the prediction of stationary gaussian processes.

On prediction of individual sequences

Sequential randomized prediction of an arbitrary binary sequence isinvestigated. No assumption is made on the mechanism of generating the bit sequence. The goal of the predictor is to minimize its relative loss, i.e., to make (almost) as few mistakes as the best ``expert'' in a fixed, possibly infinite, set of experts. We point out a surprising connection between this prediction problem and empirical process theory. First, in the special case of static (memoryless) experts, we completely characterize the minimax relative loss in terms of the maximum of an associated Rademacher process. Then we show general upper and lower bounds on the minimaxrelative loss in terms of the geometry of the class of experts. As main examples, we determine the exact order of magnitude of the minimax relative loss for the class of autoregressive linear predictors and for the class of Markov experts.

Thyroid cancer mortality and incidence: a global overview.

In most areas of the world, thyroid cancer incidence has been appreciably increasing over the last few decades, whereas mortality has steadily declined. We updated global trends in thyroid cancer mortality and incidence using official mortality data from the World Health Organization (1970-2012) and incidence data from the Cancer Incidence in Five Continents (1960-2007). Male mortality declined in all the major countries considered, with annual percent changes around -2/-3% over the last decades. Only in the United States mortality declined up to the mid 1980s and increased thereafter. Similarly, in women mortality declined in most countries considered, with APCs around -2/-5% over the last decades, with the exception of the UK, the United States and Australia, where mortality has been declining up to the late 1980s/late 1990s to level off (or increase) thereafter. In 2008-2012, most countries had mortality rates (age-standardized, world population) between 0.20 and 0.40/100,000 men and 0.20 and 0.60/100,000 women, the highest rates being in Latvia, Hungary, the Republic of Moldova and Israel (over 0.40/100,000) for men and in Ecuador, Colombia and Israel (over 0.60/100,000) for women. In most countries, a steady increase in the incidence of thyroid cancer (mainly papillary carcinomas) was observed in both sexes. The declines in thyroid cancer mortality reflect both variations in risk factor exposure and changes in the diagnosis and treatment of the disease, while the increases in the incidence are likely due to the increase in the detection of this neoplasm over the last few decades.

Prediction of myocardial infarction size using the SYNTAX score in patients treated with primary percutaneous coronary intervention for acute ST- segment elevation myocardial infarction

Objectives The relevance of the SYNTAX score for the particular case of patients with acute ST- segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PPCI)  has previously only been studied in the setting of post hoc analysis of large prospective randomized clinical trials. A "real-life" population approach has never been explored before. The aim of this study was to evaluate the impact of the SYNTAX score for the prediction of the myocardial infarction size, estimated by the creatin-kinase (CK) peak value, using the SYNTAX score in patients treated with primary coronary intervention for acute ST-segment elevation myocardial infarction. Methods The primary endpoint of the study was myocardial infarction size as measured by the CK peak value. The SYNTAX score was calculated retrospectively in 253 consecutive patients with acute ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PPCI) in a large tertiary referral center in Switzerland, between January 2009 and June 2010. Linear regression analysis was performed to compare myocardial infarction size with the SYNTAX score. This same endpoint was then stratified according to SYNTAX score tertiles: low <22 (n=178), intermediate [22-32] (n=60), and high >=33 (n=15). Results There were no significant differences in terms of clinical characteristics between the three groups. When stratified according to the SYNTAX score tertiles, average CK peak values of 1985 (low<22), 3336 (intermediate [22-32]) and 3684 (high>=33) were obtained with a p-value <0.0001. Bartlett's test for equal variances between the three groups was 9.999 (p-value <0.0067). A moderate Pearson product-moment correlation coefficient (r=0.4074) with a high statistical significance level (p-value <0.0001) was found. The coefficient of determination (R^2=0.1660) showed that approximately 17% of the variation of CK peak value (myocardial infarction size) could be explained by the SYNTAX score, i.e. by the coronary disease complexity. Conclusion In an all-comers population, the SYNTAX score is an additional tool in predicting myocardial infarction size in patients treated with primary percutaneous coronary intervention (PPCI). The stratification of patients in different risk groups according to SYNTAX enables to identify a high-risk population that may warrant particular patient care.

Implementation study of a clinical prediction score to rule out cardiogenic chest pain in community: Cluster randomized trial

1.1 Fundamentals Chest pain is a common complaint in primary care patients (1 to 3% of all consultations) (1) and its aetiology can be miscellaneous, from harmless to potentially life threatening conditions. In primary care practice, the most prevalent aetiologies are: chest wall syndrome (43%), coronary heart disease (12%) and anxiety (7%) (2). In up to 20% of cases, potentially serious conditions as cardiac, respiratory or neoplasic diseases underlie chest pain. In this context, a large number of laboratory tests are run (42%) and over 16% of patients are referred to a specialist or hospitalized (2).¦A cardiovascular origin to chest pain can threaten patient's life and investigations run to exclude a serious condition can be expensive and involve a large number of exams or referral to specialist -­‐ often without real clinical need. In emergency settings, up to 80% of chest pains in patients are due to cardiovascular events (3) and scoring methods have been developed to identify conditions such as coronary heart disease (HD) quickly and efficiently (4-­‐6). In primary care, a cardiovascular origin is present in only about 12% of patients with chest pain (2) and general practitioners (GPs) need to exclude as safely as possible a potential serious condition underlying chest pain. A simple clinical prediction rule (CPR) like those available in emergency settings may therefore help GPs and spare time and extra investigations in ruling out CHD in primary care patients. Such a tool may also help GPs reassure patients with more common origin to chest pain.

Myocardial infarct size and mortality depend on the time of day-a large multicenter study.

BACKGROUND: Different studies have shown circadian variation of ischemic burden among patients with ST-Elevation Myocardial Infarction (STEMI), but with controversial results. The aim of this study was to analyze circadian variation of myocardial infarction size and in-hospital mortality in a large multicenter registry. METHODS: This retrospective, registry-based study was based on data from AMIS Plus, a large multicenter Swiss registry of patients who suffered myocardial infarction between 1999 and 2013. Peak creatine kinase (CK) was used as a proxy measure for myocardial infarction size. Associations between peak CK, in-hospital mortality, and the time of day at symptom onset were modelled using polynomial-harmonic regression methods. RESULTS: 6,223 STEMI patients were admitted to 82 acute-care hospitals in Switzerland and treated with primary angioplasty within six hours of symptom onset. Only the 24-hour harmonic was significantly associated with peak CK (p = 0.0001). The maximum average peak CK value (2,315 U/L) was for patients with symptom onset at 23:00, whereas the minimum average (2,017 U/L) was for onset at 11:00. The amplitude of variation was 298 U/L. In addition, no correlation was observed between ischemic time and circadian peak CK variation. Of the 6,223 patients, 223 (3.58%) died during index hospitalization. Remarkably, only the 24-hour harmonic was significantly associated with in-hospital mortality. The risk of death from STEMI was highest for patients with symptom onset at 00:00 and lowest for those with onset at 12:00. DISCUSSION: As a part of this first large study of STEMI patients treated with primary angioplasty in Swiss hospitals, investigations confirmed a circadian pattern to both peak CK and in-hospital mortality which were independent of total ischemic time. Accordingly, this study proposes that symptom onset time be incorporated as a prognosis factor in patients with myocardial infarction.

Growth arrest-specific gene 6 » (Gas6) as an intra-hospital mortality predictor for patients in septic shock

Aim: Gas6 is known to be elevated in sepsis, correlating with the severity of infection and¦organ failure. We aimed to investigate the performance of Gas6 plasma levels at¦admission to predict the risk of mortality in a cohort of septic patients.¦Methods: We used prospectively collected data and plasma samples from the "Sepsis¦Cohorte Romande". Gas6 level was measured by ELISA at admission and expressed in¦percentage relative to its level in a pool of normal plasma.¦Results: Non-survivors (n=21) presented higher Gas6 levels than survivors (n=73) (median¦258% vs 164%, IQR 194 and 117 respectively) (p=0.0027). Gas6 correlated positively with¦different cytokines and was the best mortality predictor, as shown by the ROC curves area.¦In patients with septic shock (n=66), using 249% as a cut-off value, Gas6 measurement¦had a specificity of 67% and a sensitivity of 81% for predicting mortality. ROC curve area¦was 0.75. Positive and negative predictive values were 57% and 87%, respectively.¦Conclusion: Thus, Gas6 plasma level at admission might be a useful tool to predict¦mortality in patients with septic shock. Although Gas6 hold promise as an early sepsis¦marker, its precise implication in sepsis remains to be elucidated. Our observation should¦be further investigated in larger prospective clinical trials.

Aortobronchial and aortoesophageal fistulae as risk factors in surgery of descending thoracic aortic aneurysms.

OBJECTIVE: Assess outcome of patients with descending thoracic aortic aneurysms complicated by aortobronchial and aortoesophageal fistulae in comparison to patients undergoing repair of aortic aneurysms without fistulae. METHODS: In a consecutive series of 145 patients (age 60 +/- 12 years) with repair of descending thoracic and thoracoabdominal aortic aneurysms, 11 patients (8%; age 63 +/- 9; NS) primarily presented for hematemesis and/or hemoptysis. In 8/11 patients (73%) an aortobronchial fistula was identified, and 3/11 patients (27%) suffered from an aortoesophageal fistula. Five of 11 patients (45%) had undergone previous aortic surgery in the same region. RESULTS: Extent of aortic segments (range 1-8) replaced was 3.1 +/- 1.4 for all versus 2.6 +/- 0.9 for fistulae (NS). Aortic cross clamp time was 38 +/- 22 min for all versus 45 +/- 15 min for fistulae (NS). Mortality at 30 days was 18/145 (12%) for all versus 16/134 (12%) without fistulae versus 2/11 (18%) with fistulae (NS). Paraparesis and or paraplegia was observed in 11/145 (8%) for all versus 10/134 (7%) without fistulae versus 1/11 (9%) for cases with fistulae (NS). Nine additional patients died after hospital discharge, seven without fistulae and two with fistulae (days 80, and 120) bringing the 1-year mortality up to 23/134 (17%) without fistulae versus 4/11 (36%) with fistulae (NS). Further analysis shows that the 1-year mortality accounts for 1/8 patients (13%) with aorto-bronchial fistulae versus to 3/3 patients (100%) with aorto-esophageal fistulae (esophageal versus bronchial fistula: P = 0.018; esophageal versus no fistula: P = 0.006). CONCLUSIONS: Outcome of patients suffering from descending thoracic aortic aneurysms complicated by aorto-bronchial fistulae can be similar to that without fistulae, whereas for cases complicated by aorto-esophageal fistulae the prognosis seems to remain poor even after successful hospital discharge.

Markers of atherosclerosis and inflammation for prediction of coronary heart disease in older adults.

Although both inflammatory and atherosclerosis markers have been associated with coronary heart disease (CHD) risk, data directly comparing their predictive value are limited. The authors compared the value of 2 atherosclerosis markers (ankle-arm index (AAI) and aortic pulse wave velocity (aPWV)) and 3 inflammatory markers (C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-alpha)) in predicting CHD events. Among 2,191 adults aged 70-79 years at baseline (1997-1998) from the Health, Aging, and Body Composition Study cohort, the authors examined adjudicated incident myocardial infarction or CHD death ("hard" events) and "hard" events plus hospitalization for angina or coronary revascularization (total CHD events). During 8 years of follow-up between 1997-1998 and June 2007, 351 participants developed total CHD events (197 "hard" events). IL-6 (highest quartile vs. lowest: hazard ratio = 1.82, 95% confidence interval: 1.33, 2.49; P-trend < 0.001) and AAI (AAI </= 0.9 vs. AAI 1.01-1.30: hazard ratio = 1.57, 95% confidence interval: 1.14, 2.18) predicted CHD events above traditional risk factors and modestly improved global measures of predictive accuracy. CRP, TNF-alpha, and aPWV had weaker associations. IL-6 and AAI accurately reclassified 6.6% and 3.3% of participants, respectively (P's </= 0.05). Results were similar for "hard" CHD, with higher reclassification rates for AAI. IL-6 and AAI are associated with future CHD events beyond traditional risk factors and modestly improve risk prediction in older adults.

Influence of the timing of cardiac surgery on the outcome of patients with infective endocarditis and stroke.

BACKGROUND: The timing of cardiac surgery after stroke in infective endocarditis (IE) remains controversial. We examined the relationship between the timing of surgery after stroke and the incidence of in-hospital and 1-year mortalities. METHODS: Data were obtained from the International Collaboration on Endocarditis-Prospective Cohort Study of 4794 patients with definite IE who were admitted to 64 centers from June 2000 through December 2006. Multivariate logistic regression and Cox regression analyses were performed to estimate the impact of early surgery on hospital and 1-year mortality after adjustments for other significant covariates. RESULTS: Of the 857 patients with IE complicated by ischemic stroke syndromes, 198 who underwent valve replacement surgery poststroke were available for analysis. Overall, 58 (29.3%) patients underwent early surgical treatment vs 140 (70.7%) patients who underwent late surgical treatment. After adjustment for other risk factors, early surgery was not significantly associated with increased in-hospital mortality rates (odds ratio, 2.308; 95% confidence interval [CI], .942-5.652). Overall, probability of death after 1-year follow-up did not differ between 2 treatment groups (27.1% in early surgery and 19.2% in late surgery group, P = .328; adjusted hazard ratio, 1.138; 95% CI, .802-1.650). CONCLUSIONS: There is no apparent survival benefit in delaying surgery when indicated in IE patients after ischemic stroke. Further observational analyses that include detailed pre- and postoperative clinical neurologic findings and advanced imaging data (eg, ischemic stroke size), may allow for more refined recommendations on the optimal timing of valvular surgery in patients with IE and recent stroke syndromes.

Robust response transformations based on optimal prediction

Nonlinear regression problems can often be reduced to linearity by transforming the response variable (e.g., using the Box-Cox family of transformations). The classic estimates of the parameter defining the transformation as well as of the regression coefficients are based on the maximum likelihood criterion, assuming homoscedastic normal errors for the transformed response. These estimates are nonrobust in the presence of outliers and can be inconsistent when the errors are nonnormal or heteroscedastic. This article proposes new robust estimates that are consistent and asymptotically normal for any unimodal and homoscedastic error distribution. For this purpose, a robust version of conditional expectation is introduced for which the prediction mean squared error is replaced with an M scale. This concept is then used to develop a nonparametric criterion to estimate the transformation parameter as well as the regression coefficients. A finite sample estimate of this criterion based on a robust version of smearing is also proposed. Monte Carlo experiments show that the new estimates compare favorably with respect to the available competitors.

Estimating the probability of leaving unemployment using uncompleted spells from repeated cross-section data

We propose a new econometric estimation method for analyzing the probabilityof leaving unemployment using uncompleted spells from repeated cross-sectiondata, which can be especially useful when panel data are not available. Theproposed method-of-moments-based estimator has two important features:(1) it estimates the exit probability at the individual level and(2) it does not rely on the stationarity assumption of the inflowcomposition. We illustrate and gauge the performance of the proposedestimator using the Spanish Labor Force Survey data, and analyze the changesin distribution of unemployment between the 1980s and 1990s during a periodof labor market reform. We find that the relative probability of leavingunemployment of the short-term unemployed versus the long-term unemployedbecomes significantly higher in the 1990s.

Influence of noninjecting and injecting drug use on mortality, retention in the cohort, and antiretroviral therapy, in participants in the Swiss HIV Cohort Study.

OBJECTIVES: We studied the influence of noninjecting and injecting drug use on mortality, dropout rate, and the course of antiretroviral therapy (ART), in the Swiss HIV Cohort Study (SHCS). METHODS: Cohort participants, registered prior to April 2007 and with at least one drug use questionnaire completed until May 2013, were categorized according to their self-reported drug use behaviour. The probabilities of death and dropout were separately analysed using multivariable competing risks proportional hazards regression models with mutual correction for the other endpoint. Furthermore, we describe the influence of drug use on the course of ART. RESULTS: A total of 6529 participants (including 31% women) were followed during 31 215 person-years; 5.1% participants died; 10.5% were lost to follow-up. Among persons with homosexual or heterosexual HIV transmission, noninjecting drug use was associated with higher all-cause mortality [subhazard rate (SHR) 1.73; 95% confidence interval (CI) 1.07-2.83], compared with no drug use. Also, mortality was increased among former injecting drug users (IDUs) who reported noninjecting drug use (SHR 2.34; 95% CI 1.49-3.69). Noninjecting drug use was associated with higher dropout rates. The mean proportion of time with suppressed viral replication was 82.2% in all participants, irrespective of ART status, and 91.2% in those on ART. Drug use lowered adherence, and increased rates of ART change and ART interruptions. Virological failure on ART was more frequent in participants who reported concomitant drug injections while on opiate substitution, and in current IDUs, but not among noninjecting drug users. CONCLUSIONS: Noninjecting drug use and injecting drug use are modifiable risks for death, and they lower retention in a cohort and complicate ART.