Explore the vast range of titles available.

Controversy surrounds the use of parenteral nutrition in critical illness. Previous overviews used composite scales to identify high-quality trials, which may mask important differences in true methodological quality. Using a component-based approach this meta-analysis investigated the effect of trial quality on overall conclusions reached when standard enteral nutrition is compared to standard parenteral nutrition in critically ill patients. An extensive literature search was undertaken to identify all eligible trials. We retrieved 465 publications, and 11 qualified for inclusion. Nine trials presented complete follow-up, allowing the conduct of an intention to treat analysis. Aggregation revealed a mortality benefit in favour of parenteral nutrition, with no heterogeneity. A priori specified subgroup analysis demonstrated the presence of a potentially important treatment-subgroup interaction between studies of parenteral vs. early enteral nutrition compared to parenteral vs. late enteral. Six trials with complete follow-up reported infectious complications. Infectious complications were increased with parenteral use. The I(2) measure of heterogeneity was 37.7%. Intention to treat trials demonstrated reduced mortality associated with parenteral nutrition use. A priori subgroup analysis attributed this reduction to trials comparing parenteral to delayed enteral nutrition. Despite an association with increased infectious complications, a grade B+ evidence-based recommendation (level II trials, no heterogeneity) can be generated for parenteral nutrition use in patients in whom enteral nutrition cannot be initiated within 24 h of ICU admission or injury.

Efficient literature searching and the application of formal rules of evidence in evaluating the clinical literature are the two key skills defining the practice of evidence-based medicine. Although clinicians embrace the concepts of evidence-based medicine, most identify limited personal time as the major barrier towards its implementation into daily practice. Busy clinicians who practice evidence-based medicine identify systematic reviews and evidence-based clinical practice guidelines as very useful resources. This review presents a simple, easy-to-follow, three-step searching strategy that emphasises the use of powerful new PubMed features that allow clinicians to retrieve high-quality systematic reviews, clinical practice guidelines and primary studies with a single mouse click. The overall effectiveness of the process is further improved by highlighting the major features of successful and unsuccessful literature searches. At the end of this tutorial the reader should be able to conduct efficient and effective literature searches that support clinical decision making in under 10 minutes.

Purpose. To compare the predictive performance of a series of logistic regression models (LMs) to a corresponding series of back-propagation artificial neural networks (ANNs). Location. A 30 bed adult general intensive care unit (ICU) that serves a 600-bed tertiary care teaching hospital. Patients. Consecutive patients with a duration of ICU stay greater than 72 hours. Outcome: ICU-based mortality. Methods. Data were collected on day one and day three of stay using a modified APACHE III methodology. A randomly generated 811 patient developmental database was used to build models using day one data (LM1 and ANN1), day three data (LM2 and ANN2) and a combination of day one and day three data (LMOT and ANNOT). Primary comparisons were based on area under the receiver operating curves (aROC) as measured on a 338 patient validation database. Outcome predictions were also obtained from experienced ICU clinicians on a subset of patients. Results. Of the 3,728 patients admitted to the ICU during the period from March 1, 1994 through February 28, 1996, 1,181 qualified for entry into the study. There was no significant difference between LM and ANN models developed using day one data. The ANN developed using day three data performed significantly better than the corresponding LM (aROC LM2 0.7158 vs. ANN2 0.7845, p = 0.0355). The time dependent ANN model also performed significantly better than the corresponding LM (aROC LM OT 0.7342 vs. ANNOT 0.8095, p = 0.0140). The predictions obtained from ICU consultants (aROC 0.8210) discriminated significantly better than LMOT (aROC 0.6814, p = 0.0015) but there was no difference between the consultants and ANNOT (aROC 0.8094, p = 0.7684). Conclusion. Although the 1,181 patients who became eligible for entry into this study represented only 32 percent of all ICU admissions, they accounted for 80 percent of the resources (costs) expended. ANNs demonstrated significantly better predictive performance in this clinically important group of patients. Four potential reasons are discussed: (1) ANNs are insensitive to problems associated with multicollinearity; (2) ANNs place importance on novel predictors; (3) ANNs automatically model nonlinear relationships and; (4) ANNs implicitly detect all possible interaction terms.

Objective To determine whether aspirin and antioxidant therapy, combined or alone, are more effective than placebo in reducing the development of cardiovascular events in patients with diabetes mellitus and asymptomatic peripheral arterial disease.Design Multicentre, randomised, double blind, 2×2 factorial, placebo controlled trial.Setting 16 hospital centres in Scotland, supported by 188 primary care groups.Participants 1276 adults aged 40 or more with type 1 or type 2 diabetes and an ankle brachial pressure index of 0.99 or less but no symptomatic cardiovascular disease.Interventions Daily, 100 mg aspirin tablet plus antioxidant capsule (n=320), aspirin tablet plus placebo capsule (n=318), placebo tablet plus antioxidant capsule (n=320), or placebo tablet plus placebo capsule (n=318).Main outcome measures Two hierarchical composite primary end points of death from coronary heart disease or stroke, non-fatal myocardial infarction or stroke, or amputation above the ankle for critical limb ischaemia; and death from coronary heart disease or stroke.Results No evidence was found of any interaction between aspirin and antioxidant. Overall, 116 of 638 primary events occurred in the aspirin groups compared with 117 of 638 in the no aspirin groups (18.2% v 18.3%): hazard ratio 0.98 (95% confidence interval 0.76 to 1.26). Forty three deaths from coronary heart disease or stroke occurred in the aspirin groups compared with 35 in the no aspirin groups (6.7% v 5.5%): 1.23 (0.79 to 1.93). Among the antioxidant groups 117 of 640 (18.3%) primary events occurred compared with 116 of 636 (18.2%) in the no antioxidant groups (1.03, 0.79 to 1.33). Forty two (6.6%) deaths from coronary heart disease or stroke occurred in the antioxidant groups compared with 36 (5.7%) in the no antioxidant groups (1.21, 0.78 to 1.89).Conclusion This trial does not provide evidence to support the use of aspirin or antioxidants in primary prevention of cardiovascular events and mortality in the population with diabetes studied.Trial registration Current Controlled Trials ISRCTN53295293.