LAUSR

BACKGROUND Nutrition markers may be useful for diagnosis and monitoring and, also, as additional indicators of estimating death risk. We tested the association of body composition indicators (mid-upper arm circumference and phase angle) with pediatric intensive care unit (PICU) length of stay and mortality in critically ill pediatric patients. METHODS Data from children aged 2 months-18 years were collected, and bioelectrical impedance was performed to obtain phase angle. Severity was evaluated by scoring the Pediatric Index of Mortality. Descriptive statistics were reported for nominal variables. Receiver operating characteristic curve was used to analyze the association of phase angle with 30-day mortality and to find the best cutoff. Survival probabilities and PICU length of stay were estimated using the Kaplan-Meier method. RESULTS We evaluated 247 children with a median age of 4.8 years whose main cause of admission was sepsis. Survival curves showed higher survival in patients with phase angle >2.8° compared with patients with phase angle ≤2.8° (P < .0001). Kaplan-Meier time-to-event analysis showed that children with lower phase angle values were more likely to remain in the PICU (hazard ratio, 1.84; P = .003). Lower survival was also observed in patients who presented mid-upper arm circumference values ≤5th percentile (P < .03). CONCLUSIONS Mid-upper arm circumference and phase angle were associated with mortality and morbidity in critically ill children, suggesting that these parameters may be useful not only for nutrition diagnosis and monitoring, but also as an additional indicator in estimating prognosis.