Ultrasound-Derived Muscle Biomarkers as a Gateway toward Quantitative Pediatric Frailty in Palliative Care.

Abstract

Adult frailty scales rely on performance tests that are rarely feasible in children. Quantitative muscle ultrasound (US) could provide an objective bridge, but its pediatric utility is untested. To explore whether US-derived biomarkers Z-scores can cluster into recognizable patterns that serve as preliminary building blocks toward a quantitative definition of pediatric frailty. Eight-month, single-center exploratory cross-sectional observational study in a pediatric palliative care unit. Twenty-four consecutively enrolled children (1-18 years) with advanced but clinically stable, life-limiting conditions (83.5% neurometabolic, 16.5% oncologic), recruited with parental informed consent. Rectus femoris and vastus intermedius were scanned with standardized presets (64 dB gain; 4 cm depth). Age- or weight-adjusted Z-scores for muscle thickness, echo-intensity (EI), and subcutaneous-fat were derived from a published pediatric dataset. Each child was plotted in 3-D "health cubes" (x = muscle-thickness z, y = EI z, z = subcutaneous-fat z). All patients lay outside the ± 2 SD "normality cube" for at least one reference set.Median (IQR) muscle-thickness z = -4.9 (-5.6 to -4.1); EI z = +3.7 (+2.1 to +5.8).Thickness and EI were inversely correlated (ρ = -0.62, <i>p</i> = 0.002); lower thickness modestly predicted shorter time-to-death (ρ = -0.55, <i>p</i> = 0.01). In this modest, single-center cohort, pediatric palliative patients consistently deviated from normative muscle US ranges, forming a distinct cluster outside the "healthy" 3-D zone. While not diagnostic, these patterns support further, larger, longitudinal studies to develop composite US indices and formal cut-points for pediatric frailty. This visual framework may accelerate that process by highlighting multidimensional outliers at the bedside.