Impairment of small somatic and autonomic nerve fibres in intensive care unit patients with severe sepsis and critical illness polyneuropathy--a single center controlled observational study.

BACKGROUND: Axonal damage in large myelinated nerve fibres occurs in about 70% of patients with severe sepsis, known as critical illness polyneuropathy and contributes significantly to an increased short- and long-term morbidity and mortality in this population. Among other pathophysiological mechanisms, autonomic dysregulation, characterized by high concentrations of circulating catecholamines in the presence of impaired sympathetic modulation of heart and vessels have been discussed. We hypothesize that autonomic small fibre neuropathy play an important role in autonomic failure. METHODS/DESIGN: Single center, non-randomized, controlled, observational study. Skin biopsies of patients with severe sepsis and/or septic shock are compared with those of age-matched controls. In order to assess impairment of small nerve fibres, skin biopsies are taken at onset of severe sepsis, and two and 16 weeks later. Intraepidermal nerve fibre densities are histologically analyzed using anti protein gene product (PGP) 9.5 immunostaining. In addition, standardized clinical examinations, as Medical Research Council (MRC) scores of muscle strength, Rankin scores, and standardized nerve conduction studies of the right median nerve, the right tibial nerve, the left fibular nerve, and both sural nerves are performed, to identify critical illness polyneuropathy and to neurophysiologically quantify the damage of large nerve fibres. DISCUSSION: The study will allow to describe the frequency of small fibre neuropathy in patients with severe sepsis up to four months after onset of severe sepsis and to evaluate its relationship to critical illness polyneuropathy. TRIAL REGISTRATION: The trial has been registered to the German Clinical Trials Register. The trial registration number is DRKS-ID: DRKS00000642.

Muscle ultrasound for early assessment of critical illness neuromyopathy in severe sepsis.

INTRODUCTION: Muscle ultrasound is emerging as a promising tool in the diagnosis of neuromuscular diseases. The current observational study evaluates the usefulness of muscle ultrasound in patients with severe sepsis for assessment of critical illness polyneuropathy and myopathy (CINM) in the intensive care unit. METHODS: 28 patients with either septic shock or severe sepsis underwent clinical neurological examinations, muscle ultrasound, and nerve conduction studies on days 4 and 14 after onset of sepsis. 26 healthy controls of comparable age underwent clinical neurological evaluation and muscle ultrasound only. RESULTS: 26 of the 28 patients exhibited classic electrophysiological characteristics of CINM, and all showed typical clinical signs. Ultrasonic echogenicity of muscles was graded semiquantitatively and fasciculations were evaluated in muscles of proximal and distal arms and legs. 75% of patients showed a mean echotexture greater than 1.5, which was the maximal value found in the control group. A significant difference in mean muscle echotexture between patients and controls was found at day 4 and day 14 (both p < 0.001). In addition, from day 4 to day 14, the mean grades of muscle echotexture increased in the patient group, although the values did not reach significance levels (p = 0.085). Controls revealed the lowest number of fasciculations. In the patients group, fasciculations were detected in more muscular regions (lower and upper arm and leg) in comparison to controls (p = 0.08 at day 4 and p = 0.002 at day 14). CONCLUSIONS: Muscle ultrasound represents an easily applicable, non-invasive diagnostic tool which adds to neurophysiological testing information regarding morphological changes of muscles early in the course of sepsis. Muscle ultrasound could be useful for screening purposes prior to subjecting patients to more invasive techniques such as electromyography and/or muscle biopsy. TRIAL REGISTRATION: German Clinical Trials Register, DRKS-ID: DRKS00000642.

Graph based fusion of miRNA and mRNA expression data improves clinical outcome prediction in prostate cancer.

BACKGROUND: One of the main goals in cancer studies including high-throughput microRNA (miRNA) and mRNA data is to find and assess prognostic signatures capable of predicting clinical outcome. Both mRNA and miRNA expression changes in cancer diseases are described to reflect clinical characteristics like staging and prognosis. Furthermore, miRNA abundance can directly affect target transcripts and translation in tumor cells. Prediction models are trained to identify either mRNA or miRNA signatures for patient stratification. With the increasing number of microarray studies collecting mRNA and miRNA from the same patient cohort there is a need for statistical methods to integrate or fuse both kinds of data into one prediction model in order to find a combined signature that improves the prediction. RESULTS: Here, we propose a new method to fuse miRNA and mRNA data into one prediction model. Since miRNAs are known regulators of mRNAs we used the correlations between them as well as the target prediction information to build a bipartite graph representing the relations between miRNAs and mRNAs. This graph was used to guide the feature selection in order to improve the prediction. The method is illustrated on a prostate cancer data set comprising 98 patient samples with miRNA and mRNA expression data. The biochemical relapse was used as clinical endpoint. It could be shown that the bipartite graph in combination with both data sets could improve prediction performance as well as the stability of the feature selection. CONCLUSIONS: Fusion of mRNA and miRNA expression data into one prediction model improves clinical outcome prediction in terms of prediction error and stable feature selection. The R source code of the proposed method is available in the supplement.

An overview of techniques for linking high-dimensional molecular data to time-to-event endpoints by risk prediction models.

Analysis of molecular data promises identification of biomarkers for improving prognostic models, thus potentially enabling better patient management. For identifying such biomarkers, risk prediction models can be employed that link high-dimensional molecular covariate data to a clinical endpoint. In low-dimensional settings, a multitude of statistical techniques already exists for building such models, e.g. allowing for variable selection or for quantifying the added value of a new biomarker. We provide an overview of techniques for regularized estimation that transfer this toward high-dimensional settings, with a focus on models for time-to-event endpoints. Techniques for incorporating specific covariate structure are discussed, as well as techniques for dealing with more complex endpoints. Employing gene expression data from patients with diffuse large B-cell lymphoma, some typical modeling issues from low-dimensional settings are illustrated in a high-dimensional application. First, the performance of classical stepwise regression is compared to stage-wise regression, as implemented by a component-wise likelihood-based boosting approach. A second issues arises, when artificially transforming the response into a binary variable. The effects of the resulting loss of efficiency and potential bias in a high-dimensional setting are illustrated, and a link to competing risks models is provided. Finally, we discuss conditions for adequately quantifying the added value of high-dimensional gene expression measurements, both at the stage of model fitting and when performing evaluation.

Leveraging external knowledge on molecular interactions in classification methods for risk prediction of patients.

Classification of patients based on molecular markers, for example into different risk groups, is a modern field in medical research. The aim of this classification is often a better diagnosis or individualized therapy. The search for molecular markers often utilizes extremely high-dimensional data sets (e.g. gene-expression microarrays). However, in situations where the number of measured markers (genes) is intrinsically higher than the number of available patients, standard methods from statistical learning fail to deal correctly with this so-called "curse of dimensionality". Also feature or dimension reduction techniques based on statistical models promise only limited success. Several recent methods explore ideas of how to quantify and incorporate biological prior knowledge of molecular interactions and known cellular processes into the feature selection process. This article aims to give an overview of such current methods as well as the databases, where this external knowledge can be obtained from. For illustration, two recent methods are compared in detail, a feature selection approach for support vector machines as well as a boosting approach for regression models. As a practical example, data on patients with acute lymphoblastic leukemia are considered, where the binary endpoint "relapse within first year" should be predicted.

Parallelized prediction error estimation for evaluation of high-dimensional models.

UNLABELLED: There is a multitude of new techniques that promise to extract predictive information in bioinformatics applications. It has been recognized that a first step for validation of the resulting model fits should rely on proper use of resampling techniques. However, this advice is frequently not followed, potential reasons being difficulty of correct implementation and computational demand. This is addressed by the R package peperr, which is designed for reliable prediction error estimation through resampling, potentially accelerated by parallel execution on a compute cluster. Its interface allows easy connection to newly developed model fitting routines. Performance evaluation of the latter is furthermore guided by diagnostic plots, which helps to detect specific problems due to high-dimensional data structures. AVAILABILITY: http://cran.r-project.org, http://www.imbi.uni-freiburg.de/parallel. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

A general, prediction error-based criterion for selecting model complexity for high-dimensional survival models.

When fitting predictive survival models to high-dimensional data, an adequate criterion for selecting model complexity is needed to avoid overfitting. The complexity parameter is typically selected by the predictive partial log-likelihood (PLL) estimated via cross-validation. As an alternative criterion, we propose a relative version of the integrated prediction error curve (IPEC), which can be stably estimated via bootstrap resampling. The IPEC has the advantage of being applicable for models and fitting techniques where the PLL is not available. To investigate the performance of this new criterion, a simulation study is carried out, mimicking microarray survival data. Additionally, model selection by predictive PLL, estimated via bootstrap resampling instead of cross-validation, is examined. It is seen that this mostly results in similar prediction performance of the selected models, compared to estimates based on cross-validation. Model selection by bootstrap estimates of the IPEC performs about as well as selection by cross-validation estimates of the PLL. Therefore, it is expected to be a reasonable alternative in cases where there is no PLL. Similar results are seen in the analysis of a microarray survival data set from patients with diffuse large-B-cell lymphoma.

Microarray-based classification and clinical predictors: on combined classifiers and additional predictive value.

MOTIVATION: In the context of clinical bioinformatics methods are needed for assessing the additional predictive value of microarray data compared to simple clinical parameters alone. Such methods should also provide an optimal prediction rule making use of all potentialities of both types of data: they should ideally be able to catch subtypes which are not identified by clinical parameters alone. Moreover, they should address the question of the additional predictive value of microarray data in a fair framework. RESULTS: We propose a novel but simple two-step approach based on random forests and partial least squares (PLS) dimension reduction embedding the idea of pre-validation suggested by Tibshirani and colleagues, which is based on an internal cross-validation for avoiding overfitting. Our approach is fast, flexible and can be used both for assessing the overall additional significance of the microarray data and for building optimal hybrid classification rules. Its efficiency is demonstrated through simulations and an application to breast cancer and colorectal cancer data. AVAILABILITY: Our method is implemented in the freely available R package 'MAclinical' which can be downloaded from http://www.stat.uni-muenchen.de/~socher/MAclinical

Hemoglobin levels above anemia thresholds are maximally predictive for long-term survival in COPD with chronic respiratory failure.

BACKGROUND: In patients with COPD, chronic anemia is known as an unfavorable prognostic factor. Whether the association between hemoglobin (Hb) levels and long-term survival is restricted to anemia or extends to higher Hb levels has not yet been systematically assessed. METHODS: We determined Hb levels in 309 subjects with COPD and chronic respiratory failure prior to initiation of noninvasive ventilation, accounting for confounders that might affect Hb. Subjects were categorized as anemic (Hb < 12 g/dL in females, Hb < 13 g/dL in males), polycythemic (Hb ≥ 15 g/dL in females, Hb ≥ 17 g/dL in males), or normocythemic. In addition, percentiles of Hb values were analyzed with regard to mortality from any cause. RESULTS: Two-hundred seven subjects (67.0%) showed normal Hb levels, 46 (14.9%) had anemia, and 56 (18.1%) had polycythemia. Polycythemic subjects showed a higher survival rate than anemic (P = .01) and normocythemic subjects (P = .043). In a univariate Cox hazards model, Hb was associated with long-term survival (hazard ratio 0.855; 95% CI 0.783-0.934, P < .001). The 58th percentiles of Hb (14.3 g/dL in females, 15.1 g/dL in males) yielded the highest discriminative value for predicting survival (hazard ratio 0.463, 95% CI 0.324-0.660, P < .001). In the multivariate analysis this cutoff was an independent predictor for survival (hazard ratio 0.627, 95% CI 0.414-0.949, P = .03), in addition to age and body mass index. CONCLUSIONS: In subjects with COPD and chronic respiratory failure undergoing treatment with noninvasive ventilation and LTOT, high Hb levels are associated with better long-term survival. The optimal cutoff level for prediction was above the established threshold defining anemia. Thus, predicting survival only on the basis of anemia does not fully utilize the prognostic potential of Hb values in COPD.

Microarray-based genotyping and clinical outcomes of Staphylococcus aureus bloodstream infection: an exploratory study.

The clinical course of Staphylococcus aureus bacteremia varies extensively. We sought to determine the relationship between genetic characteristics of the infecting pathogen and clinical outcomes in an exploratory study. In two study centers, 317 blood culture isolates were analyzed by DNA microarray and spa genotyping. By uni- and multivariate regression analyses associations of genotype data with 30-day all-cause mortality, severe sepsis/septic shock, disseminated disease, endocarditis, and osteoarticular infection were investigated. Univariate analysis showed significant association between S. aureus genes/gene-clusters or clonal complexes and clinical endpoints. For example CC15 was associated with 30-day mortality and CC22 with osteoarticular infection. In multivariate analysis methicillin resistance (mecA, OR 4.8 [1.43-16.06]) and the beta-lactamase-gene (bla, OR 3.12 [1.17-8.30]) remained independently associated with 30-day mortality. The presence of genes for enterotoxins (sed/sej/ser) was associated with endocarditis (OR 5.11 [1.14-18.62]). Host factors such as McCabe classification (OR 4.52 [2.09-9.79] for mortality), age (OR 1.06 [1.03-1.10] per year), and community-acquisition (OR 3.40 [1.31-8.81]) had a major influence on disease severity, dissemination and mortality. Individual genotypes and clonal complexes of S. aureus can only partially explain clinical features and outcomes of S. aureus bacteremia. Genotype-phenotype association studies need to include adjustments for host factors like age, comorbidity and community-acquisition.

Morphometric long-term evaluation and comparison of brow position and shape after endoscopic forehead lift and transpalpebral browpexy.

BACKGROUND: Although a multitude of proposed forehead rejuvenation procedures have been described, long-term and systematic morphometric evaluation is rare. There are no studies comparing endoscopic forehead lifts and transpalpebral browpexy techniques according to their efficacy in raising and reshaping the brow. METHODS: In a retrospective study, standardized photographic documentations of patients undergoing an endoscopic forehead lift or a transpalpebral browpexy were morphometrically analyzed. Five measurements were taken to evaluate brow height, and two measurements were taken to describe the change in brow shape. RESULTS: Fifty-six and 29 patients, respectively, in the two groups were analyzed up to 5 years postoperatively. Morphometric evaluation proved a significant elevation of the total brow and an improvement in brow shape for the endoforehead group, whereas the brow position after the transpalpebral browpexy significantly descends, despite an improvement in subjective aesthetic outcome. An additional blepharoplasty after the endoscopic forehead lift does not lower the brow significantly. CONCLUSIONS: The efficacy of the endoforehead lift is supported by extensive, systematic, and long-term data, and its superiority over transpalpebral approaches has been proven. The descent of the brow after transpalpebral browpexy is most likely caused by a decrease of frontalis hyperactivity after the simultaneously performed blepharoplasty. Further studies need to evaluate its effect compared with blepharoplasty alone. CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, III.