An Analysis of Reliability and Accuracy of Muscle Thickness Ultrasonography in Critically Ill Children and Adults.

Abstract

Muscle wasting starts already within the first week in critically patients and is strongly related to poor outcome. Nevertheless, the early detection of muscle wasting is difficult. Therefore, we investigated the reliability and accuracy of ultrasonography to evaluate skeletal muscle wasting in critically ill children and adults. This prospective observational study enrolled 30 sedated critically ill children and 14 critically ill adults. Two independent investigators made 210 ultrasonographical assessments of muscle thigh thickness. Inter- and intraobserver reliability and cutoff levels were calculated as a function of muscle thickness and the expected reduction in muscle size (predefined at 20% and 30%). Mean ± SD muscle thickness was 1.67 ± 0.55 cm in the pediatric and 2.10 ± 0.85 cm in the adult population. The median absolute interobserver variability was 0.07 cm (interquartile range [IQR], 0.04-0.20 cm) in the pediatric population and 0.05 cm (IQR, 0.03-0.09 cm) in the adult population. However, the absolute intraobserver accuracy had a 95% confidence interval of 0.43 cm in children and 0.22 cm in adults. Only a 30% decrease (0.50 cm) in muscle thickness can be detected in critically ill children. Although the interobserver variability is acceptable in the pediatric population, the intraobserver variability is too large with respect to the expected reduction in muscle thickness. In adults, ultrasonography may be a reliable tool for early detection of muscle mass wasting.