RESUMO
Background: The study aimed to investigate the association of bioelectrical impedance analysis (BIA) for predicting clinical outcomes in critically ill children. Methods: This single-center prospective observational study included patients admitted to a mixed Pediatric Intensive Care Unit (PICU). All patients underwent anthropometric measurement and BIA measurements in the first 24 h of admission. The patients were classified into different groups based on body mass index (BMI) for age. Electronic hospital medical records were reviewed to collect clinical data for each patient. All the obtained data were analyzed by the statistical methods. Results: There were 231 patients enrolled in our study, of which 31.6% were diagnosed with malnutrition. The phase angle (PhA) of 90-day survivors was significantly higher than that of the non-survivors (4.3° ± 1.1°vs. 3.1° ± 0.9°, P = 0.02). The age-adjusted Spearman partial correlation analysis showed a weak negative correlation between PhA and duration of medical ventilation (rs = -0.42, P < 0.05). Furthermore, length of stay in PICU has a very weak correlation with ECW/TBW (rs = 0.29, P < 0.05), and a negative correlation with protein (rs = -0.27, P < 0.05). Multivariate analysis found that PhA was a significant predictor associated with the 90-day mortality when it was adjusted for PRISM III score (adjusted OR = 1.51, CI: 1.10-2.07, p = 0.01). The area under the ROC (AUROC) of PhA for predicting 90-day mortality was 0.69 (95% CI: 0.53-0.85, p < 0.05), and the cutoff value of PhA was 3.0°, with a sensitivity and specificity of 83 and 53%, respectively. Conclusion: BIA-derived PhA was found to be an independent predictor of 90-day mortality among critically ill children. A low PhA was associated with a prolonged duration of medical ventilation.