Diagnostic yield and cost analysis of electrocardiographic screening in Swiss paediatric athletes.

OBJECTIVES: Athletes performing sports on high level are at increased risk for sudden cardiac death. This includes paediatric athletes, even though data on screening strategies in this age group remain scarce. This study aimed to assess electrocardiogram interpretation criteria in paediatric athletes and to evaluate the cost of screening. METHODS: National, multicentre, retrospective, observational study on 891 athletes of paediatric age (<18 years) evaluated by history, physical examination and 12-lead electrocardiogram. The primary outcome measure was abnormal electrocardiogram findings according to the International Recommendations for Electrographic Interpretation in Athletes. The secondary outcome measure was cost of screening. RESULTS: 19 athletes (2.1%) presented abnormal electrocardiogram findings requiring further investigations, mainly abnormal T-wave inversion. These 19 athletes were predominantly males, performing endurance sports with a mean volume of 10 weekly hours for a mean duration of 6 years of training. Further investigations did not identify any relevant pathology. All athletes were cleared for competition with regular follow-up. Total costs of the screening were 108,860 USD (122 USD per athlete). CONCLUSIONS: Our study using the International Recommendations for Electrographic Interpretation in Athletes identified a low count of abnormal findings in paediatric athletes, yet raising substantially the cost of screening. Hence, the utility of electrocardiogram-inclusive screening of paediatric athletes remains to be elucidated by longitudinal data.

Probabilistic assessment of spatial heterogeneity of natural background concentrations in large-scale groundwater bodies through Functional Geostatistics.

We propose and exemplify a framework to assess Natural Background Levels (NBLs) of target chemical species in large-scale groundwater bodies based on the context of Object Oriented Spatial Statistics. The approach enables one to fully exploit the richness of the information content embedded in the probability density function (PDF) of the variables of interest, as estimated from historical records of chemical observations. As such, the population of the entire distribution functions of NBL concentrations monitored across a network of monitoring boreholes across a given aquifer is considered as the object of the spatial analysis. Our approach starkly differs from previous studies which are mainly focused on the estimation of NBLs on the basis of the median or selected quantiles of chemical concentrations, thus resulting in information loss and limitations related to the need to invoke parametric assumptions to obtain further summary statistics in addition to those considered for the spatial analysis. Our work enables one to (i) assess spatial dependencies among observed PDFs of natural background concentrations, (ii) provide spatially distributed kriging predictions of NBLs, as well as (iii) yield a robust quantification of the ensuing uncertainty and probability of exceeding given threshold concentration values via stochastic simulation. We illustrate the approach by considering the (probabilistic) characterization of spatially variable NBLs of ammonium and arsenic detected at a monitoring network across a large scale confined groundwater body in Northern Italy.