RTOG 0525: Exploratory Subset Analysis from a Randomized Phase III Trial Comparing Standard (std) Adjuvant Temozolomide (TMZ) with a Dose-dense (dd) Schedule for Glioblastoma (GBM)

Purpose/Objective(s): RTwith TMZ is the standard for GBM. dd TMZ causes prolongedMGMTdepletion in mononuclear cells and possibly in tumor. The RTOG 0525 trial (ASCO 2011) did not show an advantage from dd TMZ for survival or progression free survival. We conducted exploratory, hypothesis-generating subset analyses to detect possible benefit from dd TMZ.Materials/Methods: Patients were randomized to std (150-200 mg/m2 x 5 d) or dd TMZ (75-100 mg/m2 x 21 d) q 4 weeks for 6- 12 cycles. Eligibility included age.18, KPS$ 60, and. 1 cm2 tissue for prospective MGMTanalysis for stratification. Furtheranalyses were performed for all randomized patients (''intent-to-treat'', ITT), and for all patients starting protocol therapy (SPT). Subset analyses were performed by RPA class (III, IV, V), KPS (90-100, = 50,\50), resection (partial, total), gender (female, male), and neurologic dysfunction (nf = none, minor, moderate).Results: No significant difference was seen for median OS (16.6 vs. 14.9 months), or PFS (5.5 vs. 6.7 months, p = 0.06). MGMT methylation was linked to improved OS (21.2 vs. 14 months, p\0.0001), and PFS (8.7 vs. 5.7 months, p\0.0001). For the ITT (n = 833), there was no OS benefit from dd TMZ in any subset. Two subsets showed a PFS benefit for dd TMZ: RPA class III (6.2 vs. 12.6 months, HR 0.69, p = 0.03) and nf = minor (HR 0.77, p = 0.01). For RPA III, dd dramatically delayed progression, but post-progression dd patients died more quickly than std. A similar pattern for nf = minor was observed. For the SPT group (n = 714) there was neither PFS nor OS benefit for dd TMZ, overall. For RPA class III and nf = minor, there was a PFS benefit for dd TMZ (HR 0.73, p = 0.08; HR 0.77, p = 0.02). For nf = moderate subset, both ITT and SPT, the std arm showed superior OS (14.4 vs. 10.9 months) compared to dd, without improved PFS (HR 1.46, p = 0.03; and HR 1.74, p = 0.01. In terms of methylation status within this subset, there were more methylated patients in the std arm of the ITT subset (n = 159; 32 vs. 24%). For the SPT subset (n = 124), methylation status was similar between arms.Conclusions: This study did not demonstrate improved OS for dd TMZ for any subgroup, but for 2 highly functional subgroups, PFS was significantly increased. These data generate the testable hypothesis that intensive treatment may selectively improve disease control in those most likely able to tolerate dd therapy. Interpretation of this should be considered carefully due to small sample size, the process of multiple observations, and other confounders.Acknowledgment: This project was supported by RTOG grant U10 CA21661, and CCOP grant U10 CA37422 from the National Cancer Institute (NCI).

Multilayer perceptron with local constraint as an emerging method in spatial data analysis

The use of Geographic Information Systems has revolutionalized the handling and the visualization of geo-referenced data and has underlined the critic role of spatial analysis. The usual tools for such a purpose are geostatistics which are widely used in Earth science. Geostatistics are based upon several hypothesis which are not always verified in practice. On the other hand, Artificial Neural Network (ANN) a priori can be used without special assumptions and are known to be flexible. This paper proposes to discuss the application of ANN in the case of the interpolation of a geo-referenced variable.

Subclinical hypothyroidism and the risk of coronary heart disease and mortality: an individual participant data analysis from nine prospective cohort studies

Texte intégral: http://www.springerlink.com/content/3q68180337551r47/fulltext.pdf

Subclinical thyroid dysfunction and the risk of heart failure events: an individual participant data analysis from 6 prospective cohorts.

BACKGROUND: American College of Cardiology/American Heart Association guidelines for the diagnosis and management of heart failure recommend investigating exacerbating conditions such as thyroid dysfunction, but without specifying the impact of different thyroid-stimulation hormone (TSH) levels. Limited prospective data exist on the association between subclinical thyroid dysfunction and heart failure events. METHODS AND RESULTS: We performed a pooled analysis of individual participant data using all available prospective cohorts with thyroid function tests and subsequent follow-up of heart failure events. Individual data on 25 390 participants with 216 248 person-years of follow-up were supplied from 6 prospective cohorts in the United States and Europe. Euthyroidism was defined as TSH of 0.45 to 4.49 mIU/L, subclinical hypothyroidism as TSH of 4.5 to 19.9 mIU/L, and subclinical hyperthyroidism as TSH <0.45 mIU/L, the last two with normal free thyroxine levels. Among 25 390 participants, 2068 (8.1%) had subclinical hypothyroidism and 648 (2.6%) had subclinical hyperthyroidism. In age- and sex-adjusted analyses, risks of heart failure events were increased with both higher and lower TSH levels (P for quadratic pattern <0.01); the hazard ratio was 1.01 (95% confidence interval, 0.81-1.26) for TSH of 4.5 to 6.9 mIU/L, 1.65 (95% confidence interval, 0.84-3.23) for TSH of 7.0 to 9.9 mIU/L, 1.86 (95% confidence interval, 1.27-2.72) for TSH of 10.0 to 19.9 mIU/L (P for trend <0.01) and 1.31 (95% confidence interval, 0.88-1.95) for TSH of 0.10 to 0.44 mIU/L and 1.94 (95% confidence interval, 1.01-3.72) for TSH <0.10 mIU/L (P for trend=0.047). Risks remained similar after adjustment for cardiovascular risk factors. CONCLUSION: Risks of heart failure events were increased with both higher and lower TSH levels, particularly for TSH ≥10 and <0.10 mIU/L.

BIO-INSPIRED COMPUTATIONAL TECHNIQUES APPLIED TO THE CLUSTERING AND VISUALIZATION OF SPATIO-TEMPORAL GEOSPATIAL DATA

The coverage and volume of geo-referenced datasets are extensive and incessantly¦growing. The systematic capture of geo-referenced information generates large volumes¦of spatio-temporal data to be analyzed. Clustering and visualization play a key¦role in the exploratory data analysis and the extraction of knowledge embedded in¦these data. However, new challenges in visualization and clustering are posed when¦dealing with the special characteristics of this data. For instance, its complex structures,¦large quantity of samples, variables involved in a temporal context, high dimensionality¦and large variability in cluster shapes.¦The central aim of my thesis is to propose new algorithms and methodologies for¦clustering and visualization, in order to assist the knowledge extraction from spatiotemporal¦geo-referenced data, thus improving making decision processes.¦I present two original algorithms, one for clustering: the Fuzzy Growing Hierarchical¦Self-Organizing Networks (FGHSON), and the second for exploratory visual data analysis:¦the Tree-structured Self-organizing Maps Component Planes. In addition, I present¦methodologies that combined with FGHSON and the Tree-structured SOM Component¦Planes allow the integration of space and time seamlessly and simultaneously in¦order to extract knowledge embedded in a temporal context.¦The originality of the FGHSON lies in its capability to reflect the underlying structure¦of a dataset in a hierarchical fuzzy way. A hierarchical fuzzy representation of¦clusters is crucial when data include complex structures with large variability of cluster¦shapes, variances, densities and number of clusters. The most important characteristics¦of the FGHSON include: (1) It does not require an a-priori setup of the number¦of clusters. (2) The algorithm executes several self-organizing processes in parallel.¦Hence, when dealing with large datasets the processes can be distributed reducing the¦computational cost. (3) Only three parameters are necessary to set up the algorithm.¦In the case of the Tree-structured SOM Component Planes, the novelty of this algorithm¦lies in its ability to create a structure that allows the visual exploratory data analysis¦of large high-dimensional datasets. This algorithm creates a hierarchical structure¦of Self-Organizing Map Component Planes, arranging similar variables' projections in¦the same branches of the tree. Hence, similarities on variables' behavior can be easily¦detected (e.g. local correlations, maximal and minimal values and outliers).¦Both FGHSON and the Tree-structured SOM Component Planes were applied in¦several agroecological problems proving to be very efficient in the exploratory analysis¦and clustering of spatio-temporal datasets.¦In this thesis I also tested three soft competitive learning algorithms. Two of them¦well-known non supervised soft competitive algorithms, namely the Self-Organizing¦Maps (SOMs) and the Growing Hierarchical Self-Organizing Maps (GHSOMs); and the¦third was our original contribution, the FGHSON. Although the algorithms presented¦here have been used in several areas, to my knowledge there is not any work applying¦and comparing the performance of those techniques when dealing with spatiotemporal¦geospatial data, as it is presented in this thesis.¦I propose original methodologies to explore spatio-temporal geo-referenced datasets¦through time. Our approach uses time windows to capture temporal similarities and¦variations by using the FGHSON clustering algorithm. The developed methodologies¦are used in two case studies. In the first, the objective was to find similar agroecozones¦through time and in the second one it was to find similar environmental patterns¦shifted in time.¦Several results presented in this thesis have led to new contributions to agroecological¦knowledge, for instance, in sugar cane, and blackberry production.¦Finally, in the framework of this thesis we developed several software tools: (1)¦a Matlab toolbox that implements the FGHSON algorithm, and (2) a program called¦BIS (Bio-inspired Identification of Similar agroecozones) an interactive graphical user¦interface tool which integrates the FGHSON algorithm with Google Earth in order to¦show zones with similar agroecological characteristics.

On the Schoenberg transformations in data analysis: theory and illustrations

The class of Schoenberg transformations, embedding Euclidean distances into higher dimensional Euclidean spaces, is presented, and derived from theorems on positive definite and conditionally negative definite matrices. Original results on the arc lengths, angles and curvature of the transformations are proposed, and visualized on artificial data sets by classical multidimensional scaling. A distance-based discriminant algorithm and a robust multidimensional centroid estimate illustrate the theory, closely connected to the Gaussian kernels of Machine Learning.

Advanced geostatistical and machine-learning models for spatial data analysis of radioactively contaminated regions

Radioactive soil-contamination mapping and risk assessment is a vital issue for decision makers. Traditional approaches for mapping the spatial concentration of radionuclides employ various regression-based models, which usually provide a single-value prediction realization accompanied (in some cases) by estimation error. Such approaches do not provide the capability for rigorous uncertainty quantification or probabilistic mapping. Machine learning is a recent and fast-developing approach based on learning patterns and information from data. Artificial neural networks for prediction mapping have been especially powerful in combination with spatial statistics. A data-driven approach provides the opportunity to integrate additional relevant information about spatial phenomena into a prediction model for more accurate spatial estimates and associated uncertainty. Machine-learning algorithms can also be used for a wider spectrum of problems than before: classification, probability density estimation, and so forth. Stochastic simulations are used to model spatial variability and uncertainty. Unlike regression models, they provide multiple realizations of a particular spatial pattern that allow uncertainty and risk quantification. This paper reviews the most recent methods of spatial data analysis, prediction, and risk mapping, based on machine learning and stochastic simulations in comparison with more traditional regression models. The radioactive fallout from the Chernobyl Nuclear Power Plant accident is used to illustrate the application of the models for prediction and classification problems. This fallout is a unique case study that provides the challenging task of analyzing huge amounts of data ('hard' direct measurements, as well as supplementary information and expert estimates) and solving particular decision-oriented problems.

How to interprete subclinical thyroid dysfunction in the prevention and management of heart failure events? Individual participant data analysis from six prospective cohorts

Background: Guidelines of the Diagnosis and Management of Heart Failure (HF) recommend investigating exacerbating conditions, such as thyroid dysfunction, but without specifying impact of different TSH levels. Limited prospective data exist regarding the association between subclinical thyroid dysfunction and HF events. Methods: We performed a pooled analysis of individual participant data using all available prospective cohorts with thyroid function tests and subsequent follow-up of HF events. Individual data on 25,390 participants with 216,247 person-years of follow-up were supplied from 6 prospective cohorts in the United States and Europe. Euthyroidism was defined as TSH 0.45-4.49 mIU/L, subclinical hypothyroidism as TSH 4.5-19.9 mIU/L and subclinical hyperthyroidism as TSH <0.45 mIU/L, both with normal free thyroxine levels. HF events were defined as acute HF events, hospitalization or death related to HF events. Results: Among 25,390 participants, 2068 had subclinical hypothyroidism (8.1%) and 648 subclinical hyperthyroidism (2.6%). In age- and gender-adjusted analyses, risks of HF events were increased with both higher and lower TSH levels (P for quadratic pattern<0.01): hazard ratio (HR) was 1.01 (95% confidence interval [CI] 0.81-1.26) for TSH 4.5-6.9 mIU/L, 1.65 (CI 0.84-3.23) for TSH 7.0-9.9 mIU/L, 1.86 (CI 1.27-2.72) for TSH 10.0-19.9 mIUL/L (P for trend <0.01), and was 1.31 (CI 0.88-1.95) for TSH 0.10-0.44 mIU/L and 1.94 (CI 1.01-3.72) for TSH <0.10 mIU/L (P for trend=0.047). Risks remained similar after adjustment for cardiovascular risk factors. Conclusion: Risks of HF events were increased with both higher and lower TSH levels, particularly for TSH ≥10 mIU/L and for TSH <0.10 mIU/L. Our findings might help to interpret TSH levels in the prevention and investigation of HF.

Machine learning for geospatial data : algorithms, software tools and case studies

Résumé Cette thèse est consacrée à l'analyse, la modélisation et la visualisation de données environnementales à référence spatiale à l'aide d'algorithmes d'apprentissage automatique (Machine Learning). L'apprentissage automatique peut être considéré au sens large comme une sous-catégorie de l'intelligence artificielle qui concerne particulièrement le développement de techniques et d'algorithmes permettant à une machine d'apprendre à partir de données. Dans cette thèse, les algorithmes d'apprentissage automatique sont adaptés pour être appliqués à des données environnementales et à la prédiction spatiale. Pourquoi l'apprentissage automatique ? Parce que la majorité des algorithmes d'apprentissage automatiques sont universels, adaptatifs, non-linéaires, robustes et efficaces pour la modélisation. Ils peuvent résoudre des problèmes de classification, de régression et de modélisation de densité de probabilités dans des espaces à haute dimension, composés de variables informatives spatialisées (« géo-features ») en plus des coordonnées géographiques. De plus, ils sont idéaux pour être implémentés en tant qu'outils d'aide à la décision pour des questions environnementales allant de la reconnaissance de pattern à la modélisation et la prédiction en passant par la cartographie automatique. Leur efficacité est comparable au modèles géostatistiques dans l'espace des coordonnées géographiques, mais ils sont indispensables pour des données à hautes dimensions incluant des géo-features. Les algorithmes d'apprentissage automatique les plus importants et les plus populaires sont présentés théoriquement et implémentés sous forme de logiciels pour les sciences environnementales. Les principaux algorithmes décrits sont le Perceptron multicouches (MultiLayer Perceptron, MLP) - l'algorithme le plus connu dans l'intelligence artificielle, le réseau de neurones de régression généralisée (General Regression Neural Networks, GRNN), le réseau de neurones probabiliste (Probabilistic Neural Networks, PNN), les cartes auto-organisées (SelfOrganized Maps, SOM), les modèles à mixture Gaussiennes (Gaussian Mixture Models, GMM), les réseaux à fonctions de base radiales (Radial Basis Functions Networks, RBF) et les réseaux à mixture de densité (Mixture Density Networks, MDN). Cette gamme d'algorithmes permet de couvrir des tâches variées telle que la classification, la régression ou l'estimation de densité de probabilité. L'analyse exploratoire des données (Exploratory Data Analysis, EDA) est le premier pas de toute analyse de données. Dans cette thèse les concepts d'analyse exploratoire de données spatiales (Exploratory Spatial Data Analysis, ESDA) sont traités selon l'approche traditionnelle de la géostatistique avec la variographie expérimentale et selon les principes de l'apprentissage automatique. La variographie expérimentale, qui étudie les relations entre pairs de points, est un outil de base pour l'analyse géostatistique de corrélations spatiales anisotropiques qui permet de détecter la présence de patterns spatiaux descriptible par une statistique. L'approche de l'apprentissage automatique pour l'ESDA est présentée à travers l'application de la méthode des k plus proches voisins qui est très simple et possède d'excellentes qualités d'interprétation et de visualisation. Une part importante de la thèse traite de sujets d'actualité comme la cartographie automatique de données spatiales. Le réseau de neurones de régression généralisée est proposé pour résoudre cette tâche efficacement. Les performances du GRNN sont démontrées par des données de Comparaison d'Interpolation Spatiale (SIC) de 2004 pour lesquelles le GRNN bat significativement toutes les autres méthodes, particulièrement lors de situations d'urgence. La thèse est composée de quatre chapitres : théorie, applications, outils logiciels et des exemples guidés. Une partie importante du travail consiste en une collection de logiciels : Machine Learning Office. Cette collection de logiciels a été développée durant les 15 dernières années et a été utilisée pour l'enseignement de nombreux cours, dont des workshops internationaux en Chine, France, Italie, Irlande et Suisse ainsi que dans des projets de recherche fondamentaux et appliqués. Les cas d'études considérés couvrent un vaste spectre de problèmes géoenvironnementaux réels à basse et haute dimensionnalité, tels que la pollution de l'air, du sol et de l'eau par des produits radioactifs et des métaux lourds, la classification de types de sols et d'unités hydrogéologiques, la cartographie des incertitudes pour l'aide à la décision et l'estimation de risques naturels (glissements de terrain, avalanches). Des outils complémentaires pour l'analyse exploratoire des données et la visualisation ont également été développés en prenant soin de créer une interface conviviale et facile à l'utilisation. Machine Learning for geospatial data: algorithms, software tools and case studies Abstract The thesis is devoted to the analysis, modeling and visualisation of spatial environmental data using machine learning algorithms. In a broad sense machine learning can be considered as a subfield of artificial intelligence. It mainly concerns with the development of techniques and algorithms that allow computers to learn from data. In this thesis machine learning algorithms are adapted to learn from spatial environmental data and to make spatial predictions. Why machine learning? In few words most of machine learning algorithms are universal, adaptive, nonlinear, robust and efficient modeling tools. They can find solutions for the classification, regression, and probability density modeling problems in high-dimensional geo-feature spaces, composed of geographical space and additional relevant spatially referenced features. They are well-suited to be implemented as predictive engines in decision support systems, for the purposes of environmental data mining including pattern recognition, modeling and predictions as well as automatic data mapping. They have competitive efficiency to the geostatistical models in low dimensional geographical spaces but are indispensable in high-dimensional geo-feature spaces. The most important and popular machine learning algorithms and models interesting for geo- and environmental sciences are presented in details: from theoretical description of the concepts to the software implementation. The main algorithms and models considered are the following: multi-layer perceptron (a workhorse of machine learning), general regression neural networks, probabilistic neural networks, self-organising (Kohonen) maps, Gaussian mixture models, radial basis functions networks, mixture density networks. This set of models covers machine learning tasks such as classification, regression, and density estimation. Exploratory data analysis (EDA) is initial and very important part of data analysis. In this thesis the concepts of exploratory spatial data analysis (ESDA) is considered using both traditional geostatistical approach such as_experimental variography and machine learning. Experimental variography is a basic tool for geostatistical analysis of anisotropic spatial correlations which helps to understand the presence of spatial patterns, at least described by two-point statistics. A machine learning approach for ESDA is presented by applying the k-nearest neighbors (k-NN) method which is simple and has very good interpretation and visualization properties. Important part of the thesis deals with a hot topic of nowadays, namely, an automatic mapping of geospatial data. General regression neural networks (GRNN) is proposed as efficient model to solve this task. Performance of the GRNN model is demonstrated on Spatial Interpolation Comparison (SIC) 2004 data where GRNN model significantly outperformed all other approaches, especially in case of emergency conditions. The thesis consists of four chapters and has the following structure: theory, applications, software tools, and how-to-do-it examples. An important part of the work is a collection of software tools - Machine Learning Office. Machine Learning Office tools were developed during last 15 years and was used both for many teaching courses, including international workshops in China, France, Italy, Ireland, Switzerland and for realizing fundamental and applied research projects. Case studies considered cover wide spectrum of the real-life low and high-dimensional geo- and environmental problems, such as air, soil and water pollution by radionuclides and heavy metals, soil types and hydro-geological units classification, decision-oriented mapping with uncertainties, natural hazards (landslides, avalanches) assessments and susceptibility mapping. Complementary tools useful for the exploratory data analysis and visualisation were developed as well. The software is user friendly and easy to use.

Environmental data mining and modeling based on machine learning algorithms and geostatistics

The paper presents some contemporary approaches to spatial environmental data analysis. The main topics are concentrated on the decision-oriented problems of environmental spatial data mining and modeling: valorization and representativity of data with the help of exploratory data analysis, spatial predictions, probabilistic and risk mapping, development and application of conditional stochastic simulation models. The innovative part of the paper presents integrated/hybrid model-machine learning (ML) residuals sequential simulations-MLRSS. The models are based on multilayer perceptron and support vector regression ML algorithms used for modeling long-range spatial trends and sequential simulations of the residuals. NIL algorithms deliver non-linear solution for the spatial non-stationary problems, which are difficult for geostatistical approach. Geostatistical tools (variography) are used to characterize performance of ML algorithms, by analyzing quality and quantity of the spatially structured information extracted from data with ML algorithms. Sequential simulations provide efficient assessment of uncertainty and spatial variability. Case study from the Chernobyl fallouts illustrates the performance of the proposed model. It is shown that probability mapping, provided by the combination of ML data driven and geostatistical model based approaches, can be efficiently used in decision-making process. (C) 2003 Elsevier Ltd. All rights reserved.

Thyroid antibody status, subclinical hypothyroidism, and the risk of coronary heart disease: an individual participant data analysis.

CONTEXT: Subclinical hypothyroidism has been associated with increased risk of coronary heart disease (CHD), particularly with thyrotropin levels of 10.0 mIU/L or greater. The measurement of thyroid antibodies helps predict the progression to overt hypothyroidism, but it is unclear whether thyroid autoimmunity independently affects CHD risk. OBJECTIVE: The objective of the study was to compare the CHD risk of subclinical hypothyroidism with and without thyroid peroxidase antibodies (TPOAbs). DATA SOURCES AND STUDY SELECTION: A MEDLINE and EMBASE search from 1950 to 2011 was conducted for prospective cohorts, reporting baseline thyroid function, antibodies, and CHD outcomes. DATA EXTRACTION: Individual data of 38 274 participants from six cohorts for CHD mortality followed up for 460 333 person-years and 33 394 participants from four cohorts for CHD events. DATA SYNTHESIS: Among 38 274 adults (median age 55 y, 63% women), 1691 (4.4%) had subclinical hypothyroidism, of whom 775 (45.8%) had positive TPOAbs. During follow-up, 1436 participants died of CHD and 3285 had CHD events. Compared with euthyroid individuals, age- and gender-adjusted risks of CHD mortality in subclinical hypothyroidism were similar among individuals with and without TPOAbs [hazard ratio (HR) 1.15, 95% confidence interval (CI) 0.87-1.53 vs HR 1.26, CI 1.01-1.58, P for interaction = .62], as were risks of CHD events (HR 1.16, CI 0.87-1.56 vs HR 1.26, CI 1.02-1.56, P for interaction = .65). Risks of CHD mortality and events increased with higher thyrotropin, but within each stratum, risks did not differ by TPOAb status. CONCLUSIONS: CHD risk associated with subclinical hypothyroidism did not differ by TPOAb status, suggesting that biomarkers of thyroid autoimmunity do not add independent prognostic information for CHD outcomes.

Subclinical Hypothyroidism and the Risk of Stroke Events and Fatal Stroke: An Individual Participant Data Analysis.

OBJECTIVE: The objective was to determine the risk of stroke associated with subclinical hypothyroidism. DATA SOURCES AND STUDY SELECTION: Published prospective cohort studies were identified through a systematic search through November 2013 without restrictions in several databases. Unpublished studies were identified through the Thyroid Studies Collaboration. We collected individual participant data on thyroid function and stroke outcome. Euthyroidism was defined as TSH levels of 0.45-4.49 mIU/L, and subclinical hypothyroidism was defined as TSH levels of 4.5-19.9 mIU/L with normal T4 levels. DATA EXTRACTION AND SYNTHESIS: We collected individual participant data on 47 573 adults (3451 subclinical hypothyroidism) from 17 cohorts and followed up from 1972-2014 (489 192 person-years). Age- and sex-adjusted pooled hazard ratios (HRs) for participants with subclinical hypothyroidism compared to euthyroidism were 1.05 (95% confidence interval [CI], 0.91-1.21) for stroke events (combined fatal and nonfatal stroke) and 1.07 (95% CI, 0.80-1.42) for fatal stroke. Stratified by age, the HR for stroke events was 3.32 (95% CI, 1.25-8.80) for individuals aged 18-49 years. There was an increased risk of fatal stroke in the age groups 18-49 and 50-64 years, with a HR of 4.22 (95% CI, 1.08-16.55) and 2.86 (95% CI, 1.31-6.26), respectively (p trend 0.04). We found no increased risk for those 65-79 years old (HR, 1.00; 95% CI, 0.86-1.18) or ≥ 80 years old (HR, 1.31; 95% CI, 0.79-2.18). There was a pattern of increased risk of fatal stroke with higher TSH concentrations. CONCLUSIONS: Although no overall effect of subclinical hypothyroidism on stroke could be demonstrated, an increased risk in subjects younger than 65 years and those with higher TSH concentrations was observed.